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([231, 450]) >>> trigger_label_ffn.layers.1.bias: torch.Size([231]) >>> trigger_crf.transition: torch.Size([233, 233]) n_trainable_params: 560510152, 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 08:25:40.673547: step: 4/533, loss: 13.145635604858398 2023-01-21 08:25:41.849386: step: 8/533, loss: 19.22907829284668 2023-01-21 08:25:43.023544: step: 12/533, loss: 12.959906578063965 2023-01-21 08:25:44.178041: step: 16/533, loss: 4.805675506591797 2023-01-21 08:25:45.341920: step: 20/533, loss: 14.847404479980469 2023-01-21 08:25:46.493162: step: 24/533, loss: 25.32388687133789 2023-01-21 08:25:47.671244: step: 28/533, loss: 12.053035736083984 2023-01-21 08:25:48.854218: step: 32/533, loss: 17.16490364074707 2023-01-21 08:25:50.018403: step: 36/533, loss: 14.677508354187012 2023-01-21 08:25:51.201492: step: 40/533, loss: 3.136167049407959 2023-01-21 08:25:52.363362: step: 44/533, loss: 3.092291831970215 2023-01-21 08:25:53.550950: step: 48/533, loss: 5.116755485534668 2023-01-21 08:25:54.727752: step: 52/533, loss: 8.502379417419434 2023-01-21 08:25:55.912554: step: 56/533, loss: 16.00510025024414 2023-01-21 08:25:57.059256: step: 60/533, loss: 2.2202422618865967 2023-01-21 08:25:58.213947: step: 64/533, loss: 13.906517028808594 2023-01-21 08:25:59.416582: step: 68/533, loss: 4.964485168457031 2023-01-21 08:26:00.599046: step: 72/533, loss: 3.6854076385498047 2023-01-21 08:26:01.774933: step: 76/533, loss: 4.585338592529297 2023-01-21 08:26:02.957058: step: 80/533, loss: 20.101791381835938 2023-01-21 08:26:04.132342: step: 84/533, loss: 6.803764343261719 2023-01-21 08:26:05.321146: step: 88/533, loss: 4.213061332702637 2023-01-21 08:26:06.450180: step: 92/533, loss: 9.807540893554688 2023-01-21 08:26:07.591485: step: 96/533, loss: 3.6892173290252686 2023-01-21 08:26:08.741244: step: 100/533, loss: 14.63908576965332 2023-01-21 08:26:09.902868: step: 104/533, loss: 3.5595526695251465 2023-01-21 08:26:11.063114: step: 108/533, loss: 2.4618730545043945 2023-01-21 08:26:12.265998: step: 112/533, loss: 10.619168281555176 2023-01-21 08:26:13.464491: step: 116/533, loss: 14.758976936340332 2023-01-21 08:26:14.622805: step: 120/533, loss: 22.126323699951172 2023-01-21 08:26:15.836654: step: 124/533, loss: 14.038152694702148 2023-01-21 08:26:16.976665: step: 128/533, loss: 1.9014348983764648 2023-01-21 08:26:18.149515: step: 132/533, loss: 10.615972518920898 2023-01-21 08:26:19.283306: step: 136/533, loss: 7.768258094787598 2023-01-21 08:26:20.436145: step: 140/533, loss: 2.7860686779022217 2023-01-21 08:26:21.592898: step: 144/533, loss: 18.730642318725586 2023-01-21 08:26:22.839720: step: 148/533, loss: 3.895756483078003 2023-01-21 08:26:23.978113: step: 152/533, loss: 17.74464225769043 2023-01-21 08:26:25.151271: step: 156/533, loss: 10.170475006103516 2023-01-21 08:26:26.351826: step: 160/533, loss: 5.223288536071777 2023-01-21 08:26:27.565971: step: 164/533, loss: 2.46091890335083 2023-01-21 08:26:28.744114: step: 168/533, loss: 4.119781494140625 2023-01-21 08:26:29.942400: step: 172/533, loss: 11.421395301818848 2023-01-21 08:26:31.085017: step: 176/533, loss: 12.13332748413086 2023-01-21 08:26:32.246302: step: 180/533, loss: 13.479781150817871 2023-01-21 08:26:33.429962: step: 184/533, loss: 2.916821002960205 2023-01-21 08:26:34.612753: step: 188/533, loss: 14.945432662963867 2023-01-21 08:26:35.816796: step: 192/533, loss: 3.467480182647705 2023-01-21 08:26:37.009709: step: 196/533, loss: 19.44829750061035 2023-01-21 08:26:38.188353: step: 200/533, loss: 3.2405571937561035 2023-01-21 08:26:39.398179: step: 204/533, loss: 18.749374389648438 2023-01-21 08:26:40.564129: step: 208/533, loss: 10.947992324829102 2023-01-21 08:26:41.711592: step: 212/533, loss: 12.88017749786377 2023-01-21 08:26:42.907089: step: 216/533, loss: 11.977092742919922 2023-01-21 08:26:44.078450: step: 220/533, loss: 12.171873092651367 2023-01-21 08:26:45.298798: step: 224/533, loss: 22.430875778198242 2023-01-21 08:26:46.412168: step: 228/533, loss: 8.519892692565918 2023-01-21 08:26:47.589340: step: 232/533, loss: 10.795780181884766 2023-01-21 08:26:48.740659: step: 236/533, loss: 3.2259273529052734 2023-01-21 08:26:49.901226: step: 240/533, loss: 11.365700721740723 2023-01-21 08:26:51.106881: step: 244/533, loss: 3.0791263580322266 2023-01-21 08:26:52.277046: step: 248/533, loss: 1.945581078529358 2023-01-21 08:26:53.463401: step: 252/533, loss: 32.50908660888672 2023-01-21 08:26:54.649780: step: 256/533, loss: 19.508695602416992 2023-01-21 08:26:55.845959: step: 260/533, loss: 9.221532821655273 2023-01-21 08:26:57.035557: step: 264/533, loss: 2.031940460205078 2023-01-21 08:26:58.233115: step: 268/533, loss: 2.9496140480041504 2023-01-21 08:26:59.395984: step: 272/533, loss: 5.251761436462402 2023-01-21 08:27:00.610533: step: 276/533, loss: 2.8276400566101074 2023-01-21 08:27:01.777750: step: 280/533, loss: 5.040543556213379 2023-01-21 08:27:02.966827: step: 284/533, loss: 14.219460487365723 2023-01-21 08:27:04.178065: step: 288/533, loss: 2.935567855834961 2023-01-21 08:27:05.354939: step: 292/533, loss: 3.5496878623962402 2023-01-21 08:27:06.573664: step: 296/533, loss: 16.49407196044922 2023-01-21 08:27:07.786725: step: 300/533, loss: 19.036121368408203 2023-01-21 08:27:08.973204: step: 304/533, loss: 2.9939088821411133 2023-01-21 08:27:10.229720: step: 308/533, loss: 4.264180660247803 2023-01-21 08:27:11.420754: step: 312/533, loss: 22.403202056884766 2023-01-21 08:27:12.581164: step: 316/533, loss: 2.887939691543579 2023-01-21 08:27:13.755429: step: 320/533, loss: 4.152337074279785 2023-01-21 08:27:14.929757: step: 324/533, loss: 3.6799814701080322 2023-01-21 08:27:16.114465: step: 328/533, loss: 37.86791229248047 2023-01-21 08:27:17.307456: step: 332/533, loss: 2.171797037124634 2023-01-21 08:27:18.516545: step: 336/533, loss: 16.571880340576172 2023-01-21 08:27:19.710824: step: 340/533, loss: 25.26451873779297 2023-01-21 08:27:20.920484: step: 344/533, loss: 20.807138442993164 2023-01-21 08:27:22.097445: step: 348/533, loss: 2.651780843734741 2023-01-21 08:27:23.287256: step: 352/533, loss: 19.05449676513672 2023-01-21 08:27:24.468212: step: 356/533, loss: 19.279260635375977 2023-01-21 08:27:25.620165: step: 360/533, loss: 4.750474452972412 2023-01-21 08:27:26.797621: step: 364/533, loss: 6.298741340637207 2023-01-21 08:27:28.009339: step: 368/533, loss: 7.545099258422852 2023-01-21 08:27:29.157288: step: 372/533, loss: 2.0527303218841553 2023-01-21 08:27:30.317740: step: 376/533, loss: 7.285812854766846 2023-01-21 08:27:31.475502: step: 380/533, loss: 6.0223188400268555 2023-01-21 08:27:32.610964: step: 384/533, loss: 5.087244510650635 2023-01-21 08:27:33.807429: step: 388/533, loss: 3.4977898597717285 2023-01-21 08:27:34.992483: step: 392/533, loss: 3.954305648803711 2023-01-21 08:27:36.167490: step: 396/533, loss: 6.2261738777160645 2023-01-21 08:27:37.332768: step: 400/533, loss: 2.483565330505371 2023-01-21 08:27:38.530861: step: 404/533, loss: 3.413839817047119 2023-01-21 08:27:39.730072: step: 408/533, loss: 3.6666359901428223 2023-01-21 08:27:40.949269: step: 412/533, loss: 4.7104339599609375 2023-01-21 08:27:42.179780: step: 416/533, loss: 3.085981845855713 2023-01-21 08:27:43.367373: step: 420/533, loss: 2.365067720413208 2023-01-21 08:27:44.559702: step: 424/533, loss: 3.2868666648864746 2023-01-21 08:27:45.733445: step: 428/533, loss: 4.227965831756592 2023-01-21 08:27:46.890622: step: 432/533, loss: 8.1939115524292 2023-01-21 08:27:48.084169: step: 436/533, loss: 2.514676094055176 2023-01-21 08:27:49.250509: step: 440/533, loss: 18.518136978149414 2023-01-21 08:27:50.465675: step: 444/533, loss: 13.300690650939941 2023-01-21 08:27:51.667846: step: 448/533, loss: 15.62561321258545 2023-01-21 08:27:52.859278: step: 452/533, loss: 5.538820743560791 2023-01-21 08:27:54.016700: step: 456/533, loss: 6.011515140533447 2023-01-21 08:27:55.208926: step: 460/533, loss: 13.365700721740723 2023-01-21 08:27:56.397923: step: 464/533, loss: 16.098880767822266 2023-01-21 08:27:57.592998: step: 468/533, loss: 8.444330215454102 2023-01-21 08:27:58.809065: step: 472/533, loss: 17.922931671142578 2023-01-21 08:27:59.986524: step: 476/533, loss: 2.046785593032837 2023-01-21 08:28:01.206978: step: 480/533, loss: 13.46902084350586 2023-01-21 08:28:02.392994: step: 484/533, loss: 15.22579288482666 2023-01-21 08:28:03.552887: step: 488/533, loss: 1.474422574043274 2023-01-21 08:28:04.739693: step: 492/533, loss: 2.030651807785034 2023-01-21 08:28:05.899302: step: 496/533, loss: 7.230257987976074 2023-01-21 08:28:07.058250: step: 500/533, loss: 9.662696838378906 2023-01-21 08:28:08.231448: step: 504/533, loss: 12.307378768920898 2023-01-21 08:28:09.436195: step: 508/533, loss: 6.826079368591309 2023-01-21 08:28:10.596416: step: 512/533, loss: 2.6584668159484863 2023-01-21 08:28:11.788141: step: 516/533, loss: 6.260522842407227 2023-01-21 08:28:12.997396: step: 520/533, loss: 6.591512203216553 2023-01-21 08:28:14.153902: step: 524/533, loss: 0.7286615967750549 2023-01-21 08:28:15.335028: step: 528/533, loss: 2.6164960861206055 2023-01-21 08:28:16.524136: step: 532/533, loss: 1.4323185682296753 2023-01-21 08:28:17.649105: step: 536/533, loss: 1.4544968605041504 2023-01-21 08:28:18.879631: step: 540/533, loss: 5.8127617835998535 2023-01-21 08:28:20.073907: step: 544/533, loss: 1.2871625423431396 2023-01-21 08:28:21.240633: step: 548/533, loss: 1.926548719406128 2023-01-21 08:28:22.408212: step: 552/533, loss: 1.577078938484192 2023-01-21 08:28:23.646343: step: 556/533, loss: 4.6855998039245605 2023-01-21 08:28:24.841128: step: 560/533, loss: 5.157561779022217 2023-01-21 08:28:26.052012: step: 564/533, loss: 1.6243854761123657 2023-01-21 08:28:27.238825: step: 568/533, loss: 1.0705506801605225 2023-01-21 08:28:28.418241: step: 572/533, loss: 4.837060451507568 2023-01-21 08:28:29.587217: step: 576/533, loss: 3.3355207443237305 2023-01-21 08:28:30.764263: step: 580/533, loss: 1.5827611684799194 2023-01-21 08:28:31.923945: step: 584/533, loss: 1.5407795906066895 2023-01-21 08:28:33.120829: step: 588/533, loss: 1.5949573516845703 2023-01-21 08:28:34.278793: step: 592/533, loss: 8.876623153686523 2023-01-21 08:28:35.516828: step: 596/533, loss: 0.9607012867927551 2023-01-21 08:28:36.721944: step: 600/533, loss: 0.6043447256088257 2023-01-21 08:28:37.876156: step: 604/533, loss: 1.9733870029449463 2023-01-21 08:28:39.074613: step: 608/533, loss: 0.6224600076675415 2023-01-21 08:28:40.265838: step: 612/533, loss: 2.3000543117523193 2023-01-21 08:28:41.477972: step: 616/533, loss: 2.936113119125366 2023-01-21 08:28:42.631168: step: 620/533, loss: 0.7199267148971558 2023-01-21 08:28:43.785164: step: 624/533, loss: 1.7188756465911865 2023-01-21 08:28:44.967033: step: 628/533, loss: 3.489527702331543 2023-01-21 08:28:46.111606: step: 632/533, loss: 2.524557113647461 2023-01-21 08:28:47.276845: step: 636/533, loss: 1.7253302335739136 2023-01-21 08:28:48.430631: step: 640/533, loss: 0.2792018949985504 2023-01-21 08:28:49.592431: step: 644/533, loss: 8.905596733093262 2023-01-21 08:28:50.756997: step: 648/533, loss: 0.6786611080169678 2023-01-21 08:28:51.945482: step: 652/533, loss: 2.0452775955200195 2023-01-21 08:28:53.130167: step: 656/533, loss: 0.5380074977874756 2023-01-21 08:28:54.307374: step: 660/533, loss: 2.289801836013794 2023-01-21 08:28:55.449043: step: 664/533, loss: 0.36594659090042114 2023-01-21 08:28:56.612886: step: 668/533, loss: 2.646902084350586 2023-01-21 08:28:57.765073: step: 672/533, loss: 0.4852398931980133 2023-01-21 08:28:58.960320: step: 676/533, loss: 4.140330791473389 2023-01-21 08:29:00.126649: step: 680/533, loss: 0.5002164840698242 2023-01-21 08:29:01.285982: step: 684/533, loss: 2.7602343559265137 2023-01-21 08:29:02.451419: step: 688/533, loss: 4.075169563293457 2023-01-21 08:29:03.654366: step: 692/533, loss: 0.9605245590209961 2023-01-21 08:29:04.862054: step: 696/533, loss: 3.2553060054779053 2023-01-21 08:29:06.065516: step: 700/533, loss: 2.454166889190674 2023-01-21 08:29:07.242371: step: 704/533, loss: 0.664567768573761 2023-01-21 08:29:08.446652: step: 708/533, loss: 1.123242735862732 2023-01-21 08:29:09.576693: step: 712/533, loss: 0.6399648785591125 2023-01-21 08:29:10.750659: step: 716/533, loss: 5.524782657623291 2023-01-21 08:29:11.927724: step: 720/533, loss: 1.5474045276641846 2023-01-21 08:29:13.111559: step: 724/533, loss: 1.0461201667785645 2023-01-21 08:29:14.275718: step: 728/533, loss: 0.6861740350723267 2023-01-21 08:29:15.432707: step: 732/533, loss: 1.4719107151031494 2023-01-21 08:29:16.598447: step: 736/533, loss: 2.194241523742676 2023-01-21 08:29:17.798810: step: 740/533, loss: 0.322665274143219 2023-01-21 08:29:18.952882: step: 744/533, loss: 1.455559492111206 2023-01-21 08:29:20.110361: step: 748/533, loss: 1.8877043724060059 2023-01-21 08:29:21.286899: step: 752/533, loss: 1.974555492401123 2023-01-21 08:29:22.476319: step: 756/533, loss: 0.6265078783035278 2023-01-21 08:29:23.659054: step: 760/533, loss: 0.35264188051223755 2023-01-21 08:29:24.839753: step: 764/533, loss: 0.7506839036941528 2023-01-21 08:29:26.027767: step: 768/533, loss: 2.4636168479919434 2023-01-21 08:29:27.193738: step: 772/533, loss: 1.1075924634933472 2023-01-21 08:29:28.381976: step: 776/533, loss: 0.8517844676971436 2023-01-21 08:29:29.546109: step: 780/533, loss: 1.1073848009109497 2023-01-21 08:29:30.729377: step: 784/533, loss: 2.039799690246582 2023-01-21 08:29:31.897333: step: 788/533, loss: 0.1933211386203766 2023-01-21 08:29:33.052944: step: 792/533, loss: 6.374067783355713 2023-01-21 08:29:34.213241: step: 796/533, loss: 0.2266598343849182 2023-01-21 08:29:35.415341: step: 800/533, loss: 4.059456825256348 2023-01-21 08:29:36.607387: step: 804/533, loss: 1.0205544233322144 2023-01-21 08:29:37.772745: step: 808/533, loss: 2.309664726257324 2023-01-21 08:29:38.940814: step: 812/533, loss: 3.5434622764587402 2023-01-21 08:29:40.094883: step: 816/533, loss: 0.23684760928153992 2023-01-21 08:29:41.240454: step: 820/533, loss: 1.341080904006958 2023-01-21 08:29:42.433329: step: 824/533, loss: 0.9833893179893494 2023-01-21 08:29:43.600153: step: 828/533, loss: 1.579512596130371 2023-01-21 08:29:44.780109: step: 832/533, loss: 0.3956812024116516 2023-01-21 08:29:45.970818: step: 836/533, loss: 1.6221565008163452 2023-01-21 08:29:47.143633: step: 840/533, loss: 1.349205732345581 2023-01-21 08:29:48.391865: step: 844/533, loss: 1.8290324211120605 2023-01-21 08:29:49.519478: step: 848/533, loss: 0.8911094069480896 2023-01-21 08:29:50.687682: step: 852/533, loss: 0.6542081832885742 2023-01-21 08:29:51.866228: step: 856/533, loss: 1.627138614654541 2023-01-21 08:29:53.066318: step: 860/533, loss: 0.5866850018501282 2023-01-21 08:29:54.214015: step: 864/533, loss: 1.0946797132492065 2023-01-21 08:29:55.383960: step: 868/533, loss: 1.0102338790893555 2023-01-21 08:29:56.564215: step: 872/533, loss: 1.191148281097412 2023-01-21 08:29:57.728738: step: 876/533, loss: 0.5396867990493774 2023-01-21 08:29:58.936845: step: 880/533, loss: 2.4249701499938965 2023-01-21 08:30:00.120181: step: 884/533, loss: 7.279786109924316 2023-01-21 08:30:01.307335: step: 888/533, loss: 2.475615978240967 2023-01-21 08:30:02.444988: step: 892/533, loss: 0.5411392450332642 2023-01-21 08:30:03.657042: step: 896/533, loss: 0.5999356508255005 2023-01-21 08:30:04.849924: step: 900/533, loss: 2.413778066635132 2023-01-21 08:30:06.033919: step: 904/533, loss: 1.3172987699508667 2023-01-21 08:30:07.222691: step: 908/533, loss: 0.8468945026397705 2023-01-21 08:30:08.421257: step: 912/533, loss: 4.539299011230469 2023-01-21 08:30:09.585384: step: 916/533, loss: 1.444950819015503 2023-01-21 08:30:10.750128: step: 920/533, loss: 1.354472279548645 2023-01-21 08:30:11.943197: step: 924/533, loss: 0.6375842690467834 2023-01-21 08:30:13.128902: step: 928/533, loss: 2.4856770038604736 2023-01-21 08:30:14.324218: step: 932/533, loss: 0.8675698637962341 2023-01-21 08:30:15.538962: step: 936/533, loss: 1.526894211769104 2023-01-21 08:30:16.705261: step: 940/533, loss: 2.504338502883911 2023-01-21 08:30:17.918238: step: 944/533, loss: 0.5813045501708984 2023-01-21 08:30:19.105840: step: 948/533, loss: 0.29295188188552856 2023-01-21 08:30:20.301393: step: 952/533, loss: 7.454026222229004 2023-01-21 08:30:21.437304: step: 956/533, loss: 0.6717919111251831 2023-01-21 08:30:22.611654: step: 960/533, loss: 0.5095861554145813 2023-01-21 08:30:23.803080: step: 964/533, loss: 3.4565818309783936 2023-01-21 08:30:24.986731: step: 968/533, loss: 1.2004553079605103 2023-01-21 08:30:26.151417: step: 972/533, loss: 1.4490216970443726 2023-01-21 08:30:27.320994: step: 976/533, loss: 1.7319591045379639 2023-01-21 08:30:28.491184: step: 980/533, loss: 0.34661394357681274 2023-01-21 08:30:29.667820: step: 984/533, loss: 0.7132916450500488 2023-01-21 08:30:30.878894: step: 988/533, loss: 1.4547441005706787 2023-01-21 08:30:32.030565: step: 992/533, loss: 1.3637043237686157 2023-01-21 08:30:33.188761: step: 996/533, loss: 1.6559770107269287 2023-01-21 08:30:34.349491: step: 1000/533, loss: 1.0394364595413208 2023-01-21 08:30:35.502402: step: 1004/533, loss: 0.6348270177841187 2023-01-21 08:30:36.698866: step: 1008/533, loss: 0.40477514266967773 2023-01-21 08:30:37.929435: step: 1012/533, loss: 0.8926259875297546 2023-01-21 08:30:39.082370: step: 1016/533, loss: 0.346503883600235 2023-01-21 08:30:40.269928: step: 1020/533, loss: 0.2392118275165558 2023-01-21 08:30:41.436869: step: 1024/533, loss: 4.181270122528076 2023-01-21 08:30:42.587653: step: 1028/533, loss: 0.36705470085144043 2023-01-21 08:30:43.750095: step: 1032/533, loss: 2.478487014770508 2023-01-21 08:30:44.869492: step: 1036/533, loss: 3.8078560829162598 2023-01-21 08:30:46.045756: step: 1040/533, loss: 1.1824283599853516 2023-01-21 08:30:47.247327: step: 1044/533, loss: 1.399385690689087 2023-01-21 08:30:48.425162: step: 1048/533, loss: 2.3647425174713135 2023-01-21 08:30:49.616049: step: 1052/533, loss: 1.3252263069152832 2023-01-21 08:30:50.801911: step: 1056/533, loss: 0.5123209953308105 2023-01-21 08:30:51.962031: step: 1060/533, loss: 0.6275076270103455 2023-01-21 08:30:53.132062: step: 1064/533, loss: 1.1519665718078613 2023-01-21 08:30:54.301149: step: 1068/533, loss: 2.201920509338379 2023-01-21 08:30:55.432932: step: 1072/533, loss: 0.4908309578895569 2023-01-21 08:30:56.589468: step: 1076/533, loss: 2.691150665283203 2023-01-21 08:30:57.725159: step: 1080/533, loss: 0.522924542427063 2023-01-21 08:30:58.932005: step: 1084/533, loss: 0.7473450303077698 2023-01-21 08:31:00.096776: step: 1088/533, loss: 1.4121553897857666 2023-01-21 08:31:01.263628: step: 1092/533, loss: 1.576360821723938 2023-01-21 08:31:02.485147: step: 1096/533, loss: 0.37594738602638245 2023-01-21 08:31:03.667945: step: 1100/533, loss: 3.142935276031494 2023-01-21 08:31:04.837094: step: 1104/533, loss: 0.5358239412307739 2023-01-21 08:31:06.046302: step: 1108/533, loss: 0.6444010138511658 2023-01-21 08:31:07.244840: step: 1112/533, loss: 0.528475284576416 2023-01-21 08:31:08.403426: step: 1116/533, loss: 0.28249630331993103 2023-01-21 08:31:09.571701: step: 1120/533, loss: 0.6195205450057983 2023-01-21 08:31:10.753133: step: 1124/533, loss: 0.550328254699707 2023-01-21 08:31:11.896827: step: 1128/533, loss: 0.569743275642395 2023-01-21 08:31:13.075965: step: 1132/533, loss: 3.132331609725952 2023-01-21 08:31:14.282605: step: 1136/533, loss: 0.27634888887405396 2023-01-21 08:31:15.525449: step: 1140/533, loss: 3.176569938659668 2023-01-21 08:31:16.725886: step: 1144/533, loss: 0.6582551002502441 2023-01-21 08:31:17.869992: step: 1148/533, loss: 0.46856653690338135 2023-01-21 08:31:19.130566: step: 1152/533, loss: 1.2697393894195557 2023-01-21 08:31:20.331175: step: 1156/533, loss: 1.2549058198928833 2023-01-21 08:31:21.587961: step: 1160/533, loss: 1.1552642583847046 2023-01-21 08:31:22.742349: step: 1164/533, loss: 2.0883054733276367 2023-01-21 08:31:23.926437: step: 1168/533, loss: 1.2584812641143799 2023-01-21 08:31:25.106545: step: 1172/533, loss: 0.6239482760429382 2023-01-21 08:31:26.285941: step: 1176/533, loss: 0.8790833353996277 2023-01-21 08:31:27.456120: step: 1180/533, loss: 2.6410627365112305 2023-01-21 08:31:28.620961: step: 1184/533, loss: 1.0360885858535767 2023-01-21 08:31:29.765010: step: 1188/533, loss: 0.12752699851989746 2023-01-21 08:31:30.946209: step: 1192/533, loss: 0.23638108372688293 2023-01-21 08:31:32.179422: step: 1196/533, loss: 2.8699378967285156 2023-01-21 08:31:33.373365: step: 1200/533, loss: 0.7668875455856323 2023-01-21 08:31:34.571523: step: 1204/533, loss: 0.5818021893501282 2023-01-21 08:31:35.803683: step: 1208/533, loss: 0.13527949154376984 2023-01-21 08:31:36.969953: step: 1212/533, loss: 2.060091972351074 2023-01-21 08:31:38.109937: step: 1216/533, loss: 0.3165915608406067 2023-01-21 08:31:39.304885: step: 1220/533, loss: 2.0273361206054688 2023-01-21 08:31:40.487305: step: 1224/533, loss: 1.0736255645751953 2023-01-21 08:31:41.665908: step: 1228/533, loss: 1.0269742012023926 2023-01-21 08:31:42.843820: step: 1232/533, loss: 1.7775332927703857 2023-01-21 08:31:44.065505: step: 1236/533, loss: 1.2202086448669434 2023-01-21 08:31:45.264311: step: 1240/533, loss: 2.08382511138916 2023-01-21 08:31:46.458352: step: 1244/533, loss: 2.4972329139709473 2023-01-21 08:31:47.642158: step: 1248/533, loss: 0.4306202530860901 2023-01-21 08:31:48.830325: step: 1252/533, loss: 4.437713146209717 2023-01-21 08:31:49.962526: step: 1256/533, loss: 1.4629840850830078 2023-01-21 08:31:51.160994: step: 1260/533, loss: 2.1052424907684326 2023-01-21 08:31:52.365854: step: 1264/533, loss: 1.056288480758667 2023-01-21 08:31:53.558545: step: 1268/533, loss: 3.5432915687561035 2023-01-21 08:31:54.728510: step: 1272/533, loss: 1.5018134117126465 2023-01-21 08:31:55.951593: step: 1276/533, loss: 3.1959073543548584 2023-01-21 08:31:57.124043: step: 1280/533, loss: 3.9127073287963867 2023-01-21 08:31:58.307727: step: 1284/533, loss: 0.6725099086761475 2023-01-21 08:31:59.485713: step: 1288/533, loss: 0.5000278353691101 2023-01-21 08:32:00.684345: step: 1292/533, loss: 1.617366909980774 2023-01-21 08:32:01.882704: step: 1296/533, loss: 0.34270036220550537 2023-01-21 08:32:03.002251: step: 1300/533, loss: 0.2997363209724426 2023-01-21 08:32:04.176760: step: 1304/533, loss: 1.3547724485397339 2023-01-21 08:32:05.377268: step: 1308/533, loss: 0.3017137050628662 2023-01-21 08:32:06.535998: step: 1312/533, loss: 0.2045232355594635 2023-01-21 08:32:07.693323: step: 1316/533, loss: 1.3004283905029297 2023-01-21 08:32:08.899036: step: 1320/533, loss: 3.418627977371216 2023-01-21 08:32:10.112265: step: 1324/533, loss: 0.5456693172454834 2023-01-21 08:32:11.283082: step: 1328/533, loss: 0.9518420696258545 2023-01-21 08:32:12.509118: step: 1332/533, loss: 3.071730613708496 2023-01-21 08:32:13.684569: step: 1336/533, loss: 0.6783522367477417 2023-01-21 08:32:14.869212: step: 1340/533, loss: 0.475480318069458 2023-01-21 08:32:16.070864: step: 1344/533, loss: 0.5549815893173218 2023-01-21 08:32:17.214235: step: 1348/533, loss: 0.425025075674057 2023-01-21 08:32:18.402619: step: 1352/533, loss: 2.243117332458496 2023-01-21 08:32:19.587785: step: 1356/533, loss: 1.3087692260742188 2023-01-21 08:32:20.800709: step: 1360/533, loss: 1.550156593322754 2023-01-21 08:32:21.946352: step: 1364/533, loss: 1.8158907890319824 2023-01-21 08:32:23.144821: step: 1368/533, loss: 1.799782156944275 2023-01-21 08:32:24.326770: step: 1372/533, loss: 0.27897748351097107 2023-01-21 08:32:25.535305: step: 1376/533, loss: 0.5272464752197266 2023-01-21 08:32:26.701347: step: 1380/533, loss: 0.848552405834198 2023-01-21 08:32:27.913020: step: 1384/533, loss: 1.9821054935455322 2023-01-21 08:32:29.087281: step: 1388/533, loss: 2.0987937450408936 2023-01-21 08:32:30.264217: step: 1392/533, loss: 0.6874128580093384 2023-01-21 08:32:31.434138: step: 1396/533, loss: 7.185245037078857 2023-01-21 08:32:32.679961: step: 1400/533, loss: 0.46852028369903564 2023-01-21 08:32:33.847431: step: 1404/533, loss: 0.7313648462295532 2023-01-21 08:32:34.996934: step: 1408/533, loss: 0.3206093907356262 2023-01-21 08:32:36.165548: step: 1412/533, loss: 0.7183647155761719 2023-01-21 08:32:37.329926: step: 1416/533, loss: 1.5642496347427368 2023-01-21 08:32:38.531748: step: 1420/533, loss: 0.3313596844673157 2023-01-21 08:32:39.739049: step: 1424/533, loss: 0.8808186054229736 2023-01-21 08:32:40.920740: step: 1428/533, loss: 0.5782290697097778 2023-01-21 08:32:42.105823: step: 1432/533, loss: 0.38926661014556885 2023-01-21 08:32:43.302760: step: 1436/533, loss: 0.5657832026481628 2023-01-21 08:32:44.464098: step: 1440/533, loss: 1.0370789766311646 2023-01-21 08:32:45.634157: step: 1444/533, loss: 1.1857184171676636 2023-01-21 08:32:46.804657: step: 1448/533, loss: 0.3721585273742676 2023-01-21 08:32:48.035458: step: 1452/533, loss: 1.4785434007644653 2023-01-21 08:32:49.305815: step: 1456/533, loss: 1.6584553718566895 2023-01-21 08:32:50.497847: step: 1460/533, loss: 0.3198855519294739 2023-01-21 08:32:51.660530: step: 1464/533, loss: 0.4314543604850769 2023-01-21 08:32:52.814162: step: 1468/533, loss: 0.7182348370552063 2023-01-21 08:32:53.998426: step: 1472/533, loss: 0.4239721894264221 2023-01-21 08:32:55.151380: step: 1476/533, loss: 0.26039448380470276 2023-01-21 08:32:56.336543: step: 1480/533, loss: 0.7978843450546265 2023-01-21 08:32:57.526698: step: 1484/533, loss: 0.7496509552001953 2023-01-21 08:32:58.691771: step: 1488/533, loss: 0.46709907054901123 2023-01-21 08:32:59.863400: step: 1492/533, loss: 0.4759085774421692 2023-01-21 08:33:01.058759: step: 1496/533, loss: 0.3654235899448395 2023-01-21 08:33:02.207790: step: 1500/533, loss: 0.4483720064163208 2023-01-21 08:33:03.430798: step: 1504/533, loss: 1.3782504796981812 2023-01-21 08:33:04.654777: step: 1508/533, loss: 0.40954285860061646 2023-01-21 08:33:05.819593: step: 1512/533, loss: 1.1658687591552734 2023-01-21 08:33:07.038654: step: 1516/533, loss: 2.2199409008026123 2023-01-21 08:33:08.251786: step: 1520/533, loss: 1.2348964214324951 2023-01-21 08:33:09.450858: step: 1524/533, loss: 0.43148165941238403 2023-01-21 08:33:10.619701: step: 1528/533, loss: 0.45100802183151245 2023-01-21 08:33:11.795692: step: 1532/533, loss: 0.6193769574165344 2023-01-21 08:33:13.019654: step: 1536/533, loss: 1.130348801612854 2023-01-21 08:33:14.169056: step: 1540/533, loss: 1.2036304473876953 2023-01-21 08:33:15.375923: step: 1544/533, loss: 2.1202661991119385 2023-01-21 08:33:16.592128: step: 1548/533, loss: 2.759725332260132 2023-01-21 08:33:17.791036: step: 1552/533, loss: 0.36631566286087036 2023-01-21 08:33:18.951979: step: 1556/533, loss: 1.2751449346542358 2023-01-21 08:33:20.121114: step: 1560/533, loss: 0.5678756237030029 2023-01-21 08:33:21.292576: step: 1564/533, loss: 0.6512054204940796 2023-01-21 08:33:22.473281: step: 1568/533, loss: 0.5060820579528809 2023-01-21 08:33:23.635375: step: 1572/533, loss: 0.15566979348659515 2023-01-21 08:33:24.807494: step: 1576/533, loss: 1.3091468811035156 2023-01-21 08:33:26.050346: step: 1580/533, loss: 2.03983211517334 2023-01-21 08:33:27.242949: step: 1584/533, loss: 0.13604021072387695 2023-01-21 08:33:28.432241: step: 1588/533, loss: 0.2660253643989563 2023-01-21 08:33:29.593398: step: 1592/533, loss: 1.4109845161437988 2023-01-21 08:33:30.777239: step: 1596/533, loss: 0.35731926560401917 2023-01-21 08:33:31.959549: step: 1600/533, loss: 0.5896186828613281 2023-01-21 08:33:33.140674: step: 1604/533, loss: 0.9611473083496094 2023-01-21 08:33:34.373302: step: 1608/533, loss: 0.33070236444473267 2023-01-21 08:33:35.562862: step: 1612/533, loss: 0.9193414449691772 2023-01-21 08:33:36.692516: step: 1616/533, loss: 0.7878321409225464 2023-01-21 08:33:37.857674: step: 1620/533, loss: 3.335897445678711 2023-01-21 08:33:39.043230: step: 1624/533, loss: 1.4354864358901978 2023-01-21 08:33:40.279608: step: 1628/533, loss: 2.3773746490478516 2023-01-21 08:33:41.442810: step: 1632/533, loss: 1.2009724378585815 2023-01-21 08:33:42.663430: step: 1636/533, loss: 3.2265028953552246 2023-01-21 08:33:43.871327: step: 1640/533, loss: 1.3407931327819824 2023-01-21 08:33:45.049883: step: 1644/533, loss: 1.351318359375 2023-01-21 08:33:46.219237: step: 1648/533, loss: 0.5832690000534058 2023-01-21 08:33:47.406603: step: 1652/533, loss: 1.2264254093170166 2023-01-21 08:33:48.551683: step: 1656/533, loss: 0.9430838227272034 2023-01-21 08:33:49.728217: step: 1660/533, loss: 0.515834629535675 2023-01-21 08:33:50.914156: step: 1664/533, loss: 0.40780550241470337 2023-01-21 08:33:52.111302: step: 1668/533, loss: 7.985874652862549 2023-01-21 08:33:53.293058: step: 1672/533, loss: 0.5383089780807495 2023-01-21 08:33:54.474295: step: 1676/533, loss: 1.4420864582061768 2023-01-21 08:33:55.636710: step: 1680/533, loss: 3.4140679836273193 2023-01-21 08:33:56.795995: step: 1684/533, loss: 0.41258716583251953 2023-01-21 08:33:57.989134: step: 1688/533, loss: 0.3350537419319153 2023-01-21 08:33:59.216398: step: 1692/533, loss: 2.8005568981170654 2023-01-21 08:34:00.361075: step: 1696/533, loss: 0.838647186756134 2023-01-21 08:34:01.524827: step: 1700/533, loss: 1.7298648357391357 2023-01-21 08:34:02.686614: step: 1704/533, loss: 0.2837178707122803 2023-01-21 08:34:03.864574: step: 1708/533, loss: 0.8221243619918823 2023-01-21 08:34:05.031530: step: 1712/533, loss: 0.20631763339042664 2023-01-21 08:34:06.209538: step: 1716/533, loss: 0.4717278480529785 2023-01-21 08:34:07.403621: step: 1720/533, loss: 0.4660990834236145 2023-01-21 08:34:08.579728: step: 1724/533, loss: 1.2459135055541992 2023-01-21 08:34:09.725187: step: 1728/533, loss: 0.8349850177764893 2023-01-21 08:34:10.913084: step: 1732/533, loss: 2.735856771469116 2023-01-21 08:34:12.137072: step: 1736/533, loss: 1.6801704168319702 2023-01-21 08:34:13.304187: step: 1740/533, loss: 0.4961080551147461 2023-01-21 08:34:14.523195: step: 1744/533, loss: 2.493788719177246 2023-01-21 08:34:15.701171: step: 1748/533, loss: 0.5395064353942871 2023-01-21 08:34:16.915198: step: 1752/533, loss: 0.2596380114555359 2023-01-21 08:34:18.100182: step: 1756/533, loss: 0.8152353763580322 2023-01-21 08:34:19.259057: step: 1760/533, loss: 0.2847483158111572 2023-01-21 08:34:20.438640: step: 1764/533, loss: 0.19927635788917542 2023-01-21 08:34:21.616854: step: 1768/533, loss: 0.3931877613067627 2023-01-21 08:34:22.777668: step: 1772/533, loss: 4.3392486572265625 2023-01-21 08:34:23.984430: step: 1776/533, loss: 0.7024887800216675 2023-01-21 08:34:25.170037: step: 1780/533, loss: 0.36346569657325745 2023-01-21 08:34:26.383368: step: 1784/533, loss: 1.6709012985229492 2023-01-21 08:34:27.597896: step: 1788/533, loss: 0.3318673372268677 2023-01-21 08:34:28.790301: step: 1792/533, loss: 2.3311517238616943 2023-01-21 08:34:29.964031: step: 1796/533, loss: 0.8531425595283508 2023-01-21 08:34:31.138915: step: 1800/533, loss: 1.2707364559173584 2023-01-21 08:34:32.309186: step: 1804/533, loss: 0.40326642990112305 2023-01-21 08:34:33.498528: step: 1808/533, loss: 1.4189283847808838 2023-01-21 08:34:34.668096: step: 1812/533, loss: 0.4235231578350067 2023-01-21 08:34:35.868818: step: 1816/533, loss: 1.348684549331665 2023-01-21 08:34:37.103684: step: 1820/533, loss: 1.0400131940841675 2023-01-21 08:34:38.338507: step: 1824/533, loss: 4.387889385223389 2023-01-21 08:34:39.546440: step: 1828/533, loss: 1.3777626752853394 2023-01-21 08:34:40.751143: step: 1832/533, loss: 2.2717339992523193 2023-01-21 08:34:41.909068: step: 1836/533, loss: 0.43283140659332275 2023-01-21 08:34:43.075247: step: 1840/533, loss: 0.73629230260849 2023-01-21 08:34:44.271761: step: 1844/533, loss: 1.5241594314575195 2023-01-21 08:34:45.451162: step: 1848/533, loss: 0.41628575325012207 2023-01-21 08:34:46.673103: step: 1852/533, loss: 0.3854314684867859 2023-01-21 08:34:47.854130: step: 1856/533, loss: 0.9270620942115784 2023-01-21 08:34:49.027290: step: 1860/533, loss: 0.8662399053573608 2023-01-21 08:34:50.239682: step: 1864/533, loss: 0.8201812505722046 2023-01-21 08:34:51.423878: step: 1868/533, loss: 1.4643592834472656 2023-01-21 08:34:52.632219: step: 1872/533, loss: 0.1555122435092926 2023-01-21 08:34:53.798722: step: 1876/533, loss: 0.21470078825950623 2023-01-21 08:34:54.992624: step: 1880/533, loss: 1.4232566356658936 2023-01-21 08:34:56.189921: step: 1884/533, loss: 1.3975878953933716 2023-01-21 08:34:57.363239: step: 1888/533, loss: 0.2776089310646057 2023-01-21 08:34:58.582245: step: 1892/533, loss: 0.5662600994110107 2023-01-21 08:34:59.790504: step: 1896/533, loss: 0.4286738634109497 2023-01-21 08:35:00.967952: step: 1900/533, loss: 0.38017719984054565 2023-01-21 08:35:02.152723: step: 1904/533, loss: 0.6671310663223267 2023-01-21 08:35:03.354953: step: 1908/533, loss: 2.746434450149536 2023-01-21 08:35:04.561058: step: 1912/533, loss: 0.38533443212509155 2023-01-21 08:35:05.774822: step: 1916/533, loss: 1.6261848211288452 2023-01-21 08:35:06.928162: step: 1920/533, loss: 2.455322265625 2023-01-21 08:35:08.051621: step: 1924/533, loss: 0.8101444244384766 2023-01-21 08:35:09.268218: step: 1928/533, loss: 1.4234592914581299 2023-01-21 08:35:10.475156: step: 1932/533, loss: 0.3791810870170593 2023-01-21 08:35:11.664013: step: 1936/533, loss: 1.9108490943908691 2023-01-21 08:35:12.819508: step: 1940/533, loss: 0.273362398147583 2023-01-21 08:35:14.045862: step: 1944/533, loss: 0.7735607624053955 2023-01-21 08:35:15.242397: step: 1948/533, loss: 2.2022252082824707 2023-01-21 08:35:16.435708: step: 1952/533, loss: 0.6361154317855835 2023-01-21 08:35:17.588348: step: 1956/533, loss: 0.5247913599014282 2023-01-21 08:35:18.755401: step: 1960/533, loss: 0.39000555872917175 2023-01-21 08:35:19.954394: step: 1964/533, loss: 1.4923536777496338 2023-01-21 08:35:21.101677: step: 1968/533, loss: 0.4410415589809418 2023-01-21 08:35:22.280050: step: 1972/533, loss: 0.2872112989425659 2023-01-21 08:35:23.435216: step: 1976/533, loss: 1.3799985647201538 2023-01-21 08:35:24.594966: step: 1980/533, loss: 0.9958469271659851 2023-01-21 08:35:25.780644: step: 1984/533, loss: 0.27800047397613525 2023-01-21 08:35:26.963759: step: 1988/533, loss: 0.6145942211151123 2023-01-21 08:35:28.149266: step: 1992/533, loss: 1.2298780679702759 2023-01-21 08:35:29.344466: step: 1996/533, loss: 0.7616209983825684 2023-01-21 08:35:30.499484: step: 2000/533, loss: 0.4867037832736969 2023-01-21 08:35:31.701593: step: 2004/533, loss: 0.22924967110157013 2023-01-21 08:35:32.899490: step: 2008/533, loss: 0.18077389895915985 2023-01-21 08:35:34.072424: step: 2012/533, loss: 0.39582863450050354 2023-01-21 08:35:35.276713: step: 2016/533, loss: 0.3094213604927063 2023-01-21 08:35:36.444332: step: 2020/533, loss: 1.1123430728912354 2023-01-21 08:35:37.629947: step: 2024/533, loss: 2.7838902473449707 2023-01-21 08:35:38.782181: step: 2028/533, loss: 1.039722204208374 2023-01-21 08:35:39.947660: step: 2032/533, loss: 0.9073008894920349 2023-01-21 08:35:41.087370: step: 2036/533, loss: 0.20761491358280182 2023-01-21 08:35:42.251162: step: 2040/533, loss: 0.991794764995575 2023-01-21 08:35:43.400142: step: 2044/533, loss: 0.19815826416015625 2023-01-21 08:35:44.593299: step: 2048/533, loss: 1.2739765644073486 2023-01-21 08:35:45.780356: step: 2052/533, loss: 1.3405394554138184 2023-01-21 08:35:46.898336: step: 2056/533, loss: 1.065006136894226 2023-01-21 08:35:48.060588: step: 2060/533, loss: 1.6213529109954834 2023-01-21 08:35:49.243995: step: 2064/533, loss: 0.235340416431427 2023-01-21 08:35:50.388168: step: 2068/533, loss: 0.31616178154945374 2023-01-21 08:35:51.547642: step: 2072/533, loss: 0.8285520672798157 2023-01-21 08:35:52.736740: step: 2076/533, loss: 1.5899884700775146 2023-01-21 08:35:53.934519: step: 2080/533, loss: 1.8968029022216797 2023-01-21 08:35:55.127453: step: 2084/533, loss: 0.5193651914596558 2023-01-21 08:35:56.295633: step: 2088/533, loss: 1.701128363609314 2023-01-21 08:35:57.453073: step: 2092/533, loss: 0.5211665034294128 2023-01-21 08:35:58.601034: step: 2096/533, loss: 1.4466166496276855 2023-01-21 08:35:59.783810: step: 2100/533, loss: 0.7893010377883911 2023-01-21 08:36:00.949819: step: 2104/533, loss: 0.1537456065416336 2023-01-21 08:36:02.119460: step: 2108/533, loss: 0.9257146716117859 2023-01-21 08:36:03.299906: step: 2112/533, loss: 0.2685169279575348 2023-01-21 08:36:04.454926: step: 2116/533, loss: 0.31897249817848206 2023-01-21 08:36:05.648264: step: 2120/533, loss: 0.20759698748588562 2023-01-21 08:36:06.814813: step: 2124/533, loss: 1.727454662322998 2023-01-21 08:36:08.004821: step: 2128/533, loss: 0.9758910536766052 2023-01-21 08:36:09.203561: step: 2132/533, loss: 0.6938608288764954 ================================================== Loss: 3.352 -------------------- Dev: {'event': {'p': 0.47613636363636364, 'r': 0.5579227696404794, 'f1': 0.5137952176578787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Test: {'event': {'p': 0.443666899930021, 'r': 0.4053708439897698, 'f1': 0.4236551954560641}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Chinese: {'event': {'p': 0.5384615384615384, 'r': 0.3888888888888889, 'f1': 0.45161290322580644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Korean: {'event': {'p': 0.72, 'r': 0.2857142857142857, 'f1': 0.409090909090909}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Russian: {'event': {'p': 0.42105263157894735, 'r': 0.2222222222222222, 'f1': 0.2909090909090909}, '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': 0.47613636363636364, 'r': 0.5579227696404794, 'f1': 0.5137952176578787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Chinese: {'event': {'p': 0.443666899930021, 'r': 0.4053708439897698, 'f1': 0.4236551954560641}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Chinese: {'event': {'p': 0.5384615384615384, 'r': 0.3888888888888889, 'f1': 0.45161290322580644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Korean: {'event': {'p': 0.47613636363636364, 'r': 0.5579227696404794, 'f1': 0.5137952176578787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Korean: {'event': {'p': 0.443666899930021, 'r': 0.4053708439897698, 'f1': 0.4236551954560641}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Korean: {'event': {'p': 0.72, 'r': 0.2857142857142857, 'f1': 0.409090909090909}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Russian: {'event': {'p': 0.47613636363636364, 'r': 0.5579227696404794, 'f1': 0.5137952176578787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Russian: {'event': {'p': 0.443666899930021, 'r': 0.4053708439897698, 'f1': 0.4236551954560641}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Russian: {'event': {'p': 0.42105263157894735, 'r': 0.2222222222222222, 'f1': 0.2909090909090909}, '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 08:37:08.423402: step: 4/533, loss: 1.0354243516921997 2023-01-21 08:37:09.585988: step: 8/533, loss: 0.6560386419296265 2023-01-21 08:37:10.749750: step: 12/533, loss: 0.39212942123413086 2023-01-21 08:37:11.992643: step: 16/533, loss: 2.5404789447784424 2023-01-21 08:37:13.184493: step: 20/533, loss: 0.4714636206626892 2023-01-21 08:37:14.361212: step: 24/533, loss: 0.5660895109176636 2023-01-21 08:37:15.536463: step: 28/533, loss: 0.5721099972724915 2023-01-21 08:37:16.741054: step: 32/533, loss: 2.7766735553741455 2023-01-21 08:37:17.958781: step: 36/533, loss: 0.4042859971523285 2023-01-21 08:37:19.177691: step: 40/533, loss: 0.6899076104164124 2023-01-21 08:37:20.398465: step: 44/533, loss: 0.7919952273368835 2023-01-21 08:37:21.588059: step: 48/533, loss: 6.525989532470703 2023-01-21 08:37:22.741709: step: 52/533, loss: 0.21311740577220917 2023-01-21 08:37:23.907774: step: 56/533, loss: 1.2778058052062988 2023-01-21 08:37:25.119309: step: 60/533, loss: 0.9194133877754211 2023-01-21 08:37:26.282619: step: 64/533, loss: 0.288774311542511 2023-01-21 08:37:27.462060: step: 68/533, loss: 0.13113990426063538 2023-01-21 08:37:28.667215: step: 72/533, loss: 1.1873517036437988 2023-01-21 08:37:29.845841: step: 76/533, loss: 2.0232064723968506 2023-01-21 08:37:31.044021: step: 80/533, loss: 3.1798782348632812 2023-01-21 08:37:32.184207: step: 84/533, loss: 0.3318081498146057 2023-01-21 08:37:33.361699: step: 88/533, loss: 1.6192948818206787 2023-01-21 08:37:34.553514: step: 92/533, loss: 3.523315191268921 2023-01-21 08:37:35.757381: step: 96/533, loss: 1.138803243637085 2023-01-21 08:37:36.936715: step: 100/533, loss: 1.2268428802490234 2023-01-21 08:37:38.099085: step: 104/533, loss: 0.37400999665260315 2023-01-21 08:37:39.293766: step: 108/533, loss: 0.6838157773017883 2023-01-21 08:37:40.467991: step: 112/533, loss: 0.41508930921554565 2023-01-21 08:37:41.629077: step: 116/533, loss: 0.8689453601837158 2023-01-21 08:37:42.811486: step: 120/533, loss: 0.25466614961624146 2023-01-21 08:37:44.007212: step: 124/533, loss: 0.7410454154014587 2023-01-21 08:37:45.149019: step: 128/533, loss: 0.2254638671875 2023-01-21 08:37:46.326517: step: 132/533, loss: 3.513627052307129 2023-01-21 08:37:47.489080: step: 136/533, loss: 0.8366312980651855 2023-01-21 08:37:48.713263: step: 140/533, loss: 0.20423099398612976 2023-01-21 08:37:49.912443: step: 144/533, loss: 0.4385802745819092 2023-01-21 08:37:51.146773: step: 148/533, loss: 1.250146508216858 2023-01-21 08:37:52.301092: step: 152/533, loss: 1.2558205127716064 2023-01-21 08:37:53.486701: step: 156/533, loss: 0.703910231590271 2023-01-21 08:37:54.675540: step: 160/533, loss: 0.6373254656791687 2023-01-21 08:37:55.854681: step: 164/533, loss: 0.928286075592041 2023-01-21 08:37:57.041584: step: 168/533, loss: 0.5824216604232788 2023-01-21 08:37:58.198911: step: 172/533, loss: 0.19328445196151733 2023-01-21 08:37:59.394685: step: 176/533, loss: 0.9951791167259216 2023-01-21 08:38:00.542614: step: 180/533, loss: 0.23292537033557892 2023-01-21 08:38:01.767976: step: 184/533, loss: 0.3331875205039978 2023-01-21 08:38:02.930476: step: 188/533, loss: 0.16349191963672638 2023-01-21 08:38:04.116046: step: 192/533, loss: 2.5099377632141113 2023-01-21 08:38:05.298923: step: 196/533, loss: 0.9419357180595398 2023-01-21 08:38:06.440561: step: 200/533, loss: 0.15835857391357422 2023-01-21 08:38:07.604285: step: 204/533, loss: 0.8913013935089111 2023-01-21 08:38:08.779165: step: 208/533, loss: 0.23131799697875977 2023-01-21 08:38:09.925358: step: 212/533, loss: 0.19361896812915802 2023-01-21 08:38:11.069235: step: 216/533, loss: 2.761167526245117 2023-01-21 08:38:12.242927: step: 220/533, loss: 0.34818974137306213 2023-01-21 08:38:13.408324: step: 224/533, loss: 0.1616200953722 2023-01-21 08:38:14.572601: step: 228/533, loss: 0.3424082398414612 2023-01-21 08:38:15.753169: step: 232/533, loss: 1.104100227355957 2023-01-21 08:38:16.958993: step: 236/533, loss: 0.1488325148820877 2023-01-21 08:38:18.183858: step: 240/533, loss: 0.3915451169013977 2023-01-21 08:38:19.374018: step: 244/533, loss: 0.6880553960800171 2023-01-21 08:38:20.586749: step: 248/533, loss: 0.27069178223609924 2023-01-21 08:38:21.795296: step: 252/533, loss: 0.5047810077667236 2023-01-21 08:38:22.951807: step: 256/533, loss: 2.100074052810669 2023-01-21 08:38:24.118523: step: 260/533, loss: 0.1853216141462326 2023-01-21 08:38:25.305688: step: 264/533, loss: 0.35941487550735474 2023-01-21 08:38:26.505716: step: 268/533, loss: 2.569343090057373 2023-01-21 08:38:27.675227: step: 272/533, loss: 0.566318154335022 2023-01-21 08:38:28.847211: step: 276/533, loss: 0.2727154791355133 2023-01-21 08:38:30.088604: step: 280/533, loss: 2.3410072326660156 2023-01-21 08:38:31.261444: step: 284/533, loss: 0.9303352236747742 2023-01-21 08:38:32.406415: step: 288/533, loss: 0.11808319389820099 2023-01-21 08:38:33.631948: step: 292/533, loss: 1.683610439300537 2023-01-21 08:38:34.825733: step: 296/533, loss: 1.1566574573516846 2023-01-21 08:38:36.041653: step: 300/533, loss: 0.12567663192749023 2023-01-21 08:38:37.231836: step: 304/533, loss: 0.8735305666923523 2023-01-21 08:38:38.387992: step: 308/533, loss: 0.3775220513343811 2023-01-21 08:38:39.558913: step: 312/533, loss: 0.6042418479919434 2023-01-21 08:38:40.726148: step: 316/533, loss: 0.2863849699497223 2023-01-21 08:38:41.885586: step: 320/533, loss: 0.31863218545913696 2023-01-21 08:38:43.080721: step: 324/533, loss: 0.3744821548461914 2023-01-21 08:38:44.256058: step: 328/533, loss: 0.29896843433380127 2023-01-21 08:38:45.461203: step: 332/533, loss: 0.28597211837768555 2023-01-21 08:38:46.640544: step: 336/533, loss: 1.6122771501541138 2023-01-21 08:38:47.809847: step: 340/533, loss: 1.3248423337936401 2023-01-21 08:38:48.971796: step: 344/533, loss: 0.3494013547897339 2023-01-21 08:38:50.184700: step: 348/533, loss: 0.19187764823436737 2023-01-21 08:38:51.358421: step: 352/533, loss: 0.3479959964752197 2023-01-21 08:38:52.528596: step: 356/533, loss: 0.7737563848495483 2023-01-21 08:38:53.711006: step: 360/533, loss: 1.4335170984268188 2023-01-21 08:38:54.861570: step: 364/533, loss: 1.905852198600769 2023-01-21 08:38:56.069500: step: 368/533, loss: 0.3818017244338989 2023-01-21 08:38:57.230645: step: 372/533, loss: 0.49197322130203247 2023-01-21 08:38:58.406505: step: 376/533, loss: 0.14738327264785767 2023-01-21 08:38:59.585164: step: 380/533, loss: 1.6405372619628906 2023-01-21 08:39:00.816470: step: 384/533, loss: 0.3434966206550598 2023-01-21 08:39:01.970581: step: 388/533, loss: 0.7020226716995239 2023-01-21 08:39:03.173435: step: 392/533, loss: 0.42894333600997925 2023-01-21 08:39:04.363728: step: 396/533, loss: 0.984560489654541 2023-01-21 08:39:05.545157: step: 400/533, loss: 0.1585921347141266 2023-01-21 08:39:06.720188: step: 404/533, loss: 1.4114590883255005 2023-01-21 08:39:07.915429: step: 408/533, loss: 1.7944252490997314 2023-01-21 08:39:09.096366: step: 412/533, loss: 0.2702641487121582 2023-01-21 08:39:10.226395: step: 416/533, loss: 0.8641048669815063 2023-01-21 08:39:11.454644: step: 420/533, loss: 0.27141672372817993 2023-01-21 08:39:12.655656: step: 424/533, loss: 0.8849024176597595 2023-01-21 08:39:13.829955: step: 428/533, loss: 1.5761537551879883 2023-01-21 08:39:14.994804: step: 432/533, loss: 0.24044305086135864 2023-01-21 08:39:16.173603: step: 436/533, loss: 1.4541585445404053 2023-01-21 08:39:17.358476: step: 440/533, loss: 1.3601478338241577 2023-01-21 08:39:18.556109: step: 444/533, loss: 1.650113582611084 2023-01-21 08:39:19.726989: step: 448/533, loss: 0.24510188400745392 2023-01-21 08:39:20.909898: step: 452/533, loss: 0.4233432710170746 2023-01-21 08:39:22.072592: step: 456/533, loss: 0.3117285966873169 2023-01-21 08:39:23.280241: step: 460/533, loss: 0.31069380044937134 2023-01-21 08:39:24.466255: step: 464/533, loss: 0.42756518721580505 2023-01-21 08:39:25.639882: step: 468/533, loss: 0.284812331199646 2023-01-21 08:39:26.842507: step: 472/533, loss: 0.1648479402065277 2023-01-21 08:39:28.049241: step: 476/533, loss: 0.3525228500366211 2023-01-21 08:39:29.240086: step: 480/533, loss: 0.37308454513549805 2023-01-21 08:39:30.454170: step: 484/533, loss: 0.16926679015159607 2023-01-21 08:39:31.644430: step: 488/533, loss: 0.31133148074150085 2023-01-21 08:39:32.822090: step: 492/533, loss: 0.30838558077812195 2023-01-21 08:39:34.044866: step: 496/533, loss: 1.9954447746276855 2023-01-21 08:39:35.206066: step: 500/533, loss: 1.1287389993667603 2023-01-21 08:39:36.442649: step: 504/533, loss: 4.503689289093018 2023-01-21 08:39:37.553349: step: 508/533, loss: 0.5619094371795654 2023-01-21 08:39:38.716191: step: 512/533, loss: 0.8835994005203247 2023-01-21 08:39:39.898079: step: 516/533, loss: 0.41967689990997314 2023-01-21 08:39:41.058771: step: 520/533, loss: 0.5207710266113281 2023-01-21 08:39:42.229684: step: 524/533, loss: 0.34112748503685 2023-01-21 08:39:43.420220: step: 528/533, loss: 0.30970126390457153 2023-01-21 08:39:44.571297: step: 532/533, loss: 0.21265669167041779 2023-01-21 08:39:45.778772: step: 536/533, loss: 0.636131763458252 2023-01-21 08:39:46.964775: step: 540/533, loss: 2.11444091796875 2023-01-21 08:39:48.180673: step: 544/533, loss: 2.981522560119629 2023-01-21 08:39:49.341920: step: 548/533, loss: 1.6722217798233032 2023-01-21 08:39:50.488643: step: 552/533, loss: 0.22657814621925354 2023-01-21 08:39:51.648846: step: 556/533, loss: 0.47271353006362915 2023-01-21 08:39:52.817206: step: 560/533, loss: 1.6469544172286987 2023-01-21 08:39:53.989609: step: 564/533, loss: 0.5778610706329346 2023-01-21 08:39:55.198065: step: 568/533, loss: 0.33917656540870667 2023-01-21 08:39:56.359296: step: 572/533, loss: 0.6833656430244446 2023-01-21 08:39:57.531806: step: 576/533, loss: 0.4177490472793579 2023-01-21 08:39:58.733334: step: 580/533, loss: 1.7253884077072144 2023-01-21 08:39:59.936649: step: 584/533, loss: 0.21751871705055237 2023-01-21 08:40:01.109212: step: 588/533, loss: 0.3941943049430847 2023-01-21 08:40:02.295677: step: 592/533, loss: 0.45982152223587036 2023-01-21 08:40:03.502763: step: 596/533, loss: 0.3533408045768738 2023-01-21 08:40:04.674227: step: 600/533, loss: 0.6673813462257385 2023-01-21 08:40:05.854426: step: 604/533, loss: 0.699525773525238 2023-01-21 08:40:07.049014: step: 608/533, loss: 0.1774233877658844 2023-01-21 08:40:08.220128: step: 612/533, loss: 7.702902317047119 2023-01-21 08:40:09.446648: step: 616/533, loss: 0.1863287091255188 2023-01-21 08:40:10.618855: step: 620/533, loss: 0.46469393372535706 2023-01-21 08:40:11.816978: step: 624/533, loss: 1.0272042751312256 2023-01-21 08:40:13.031742: step: 628/533, loss: 1.2684119939804077 2023-01-21 08:40:14.187312: step: 632/533, loss: 8.235013961791992 2023-01-21 08:40:15.400657: step: 636/533, loss: 1.1135437488555908 2023-01-21 08:40:16.555364: step: 640/533, loss: 0.1753322184085846 2023-01-21 08:40:17.716270: step: 644/533, loss: 0.7635591626167297 2023-01-21 08:40:18.882948: step: 648/533, loss: 1.4237725734710693 2023-01-21 08:40:20.051233: step: 652/533, loss: 1.0969231128692627 2023-01-21 08:40:21.227817: step: 656/533, loss: 0.5957016944885254 2023-01-21 08:40:22.390799: step: 660/533, loss: 0.5933224558830261 2023-01-21 08:40:23.581279: step: 664/533, loss: 0.19568271934986115 2023-01-21 08:40:24.721655: step: 668/533, loss: 0.13619375228881836 2023-01-21 08:40:25.914204: step: 672/533, loss: 0.2319856584072113 2023-01-21 08:40:27.090058: step: 676/533, loss: 0.10898929089307785 2023-01-21 08:40:28.292023: step: 680/533, loss: 0.3891049921512604 2023-01-21 08:40:29.516906: step: 684/533, loss: 1.5643413066864014 2023-01-21 08:40:30.737379: step: 688/533, loss: 0.31160032749176025 2023-01-21 08:40:31.873778: step: 692/533, loss: 0.43434616923332214 2023-01-21 08:40:33.075959: step: 696/533, loss: 0.11282258480787277 2023-01-21 08:40:34.274954: step: 700/533, loss: 0.34091776609420776 2023-01-21 08:40:35.479639: step: 704/533, loss: 1.8270622491836548 2023-01-21 08:40:36.677359: step: 708/533, loss: 0.3535654842853546 2023-01-21 08:40:37.854583: step: 712/533, loss: 3.5323214530944824 2023-01-21 08:40:39.070046: step: 716/533, loss: 0.023487448692321777 2023-01-21 08:40:40.247067: step: 720/533, loss: 0.9441156983375549 2023-01-21 08:40:41.427744: step: 724/533, loss: 0.5851117372512817 2023-01-21 08:40:42.598243: step: 728/533, loss: 0.6507365107536316 2023-01-21 08:40:43.812258: step: 732/533, loss: 0.35127145051956177 2023-01-21 08:40:44.951381: step: 736/533, loss: 0.732417643070221 2023-01-21 08:40:46.171935: step: 740/533, loss: 0.8329192996025085 2023-01-21 08:40:47.349919: step: 744/533, loss: 0.6095622181892395 2023-01-21 08:40:48.578356: step: 748/533, loss: 2.958179235458374 2023-01-21 08:40:49.734887: step: 752/533, loss: 0.21832752227783203 2023-01-21 08:40:50.914671: step: 756/533, loss: 0.212521031498909 2023-01-21 08:40:52.084346: step: 760/533, loss: 0.26764345169067383 2023-01-21 08:40:53.252410: step: 764/533, loss: 0.22837097942829132 2023-01-21 08:40:54.489106: step: 768/533, loss: 0.19268932938575745 2023-01-21 08:40:55.672154: step: 772/533, loss: 0.13291996717453003 2023-01-21 08:40:56.851394: step: 776/533, loss: 0.15635986626148224 2023-01-21 08:40:58.033663: step: 780/533, loss: 0.23405605554580688 2023-01-21 08:40:59.255541: step: 784/533, loss: 0.2388055920600891 2023-01-21 08:41:00.443110: step: 788/533, loss: 0.1785575896501541 2023-01-21 08:41:01.618319: step: 792/533, loss: 0.6539970636367798 2023-01-21 08:41:02.787114: step: 796/533, loss: 1.5934582948684692 2023-01-21 08:41:03.967952: step: 800/533, loss: 0.994404137134552 2023-01-21 08:41:05.157374: step: 804/533, loss: 0.9540376663208008 2023-01-21 08:41:06.333438: step: 808/533, loss: 0.09399070590734482 2023-01-21 08:41:07.515304: step: 812/533, loss: 1.6626752614974976 2023-01-21 08:41:08.692386: step: 816/533, loss: 2.802852153778076 2023-01-21 08:41:09.940507: step: 820/533, loss: 0.9512723684310913 2023-01-21 08:41:11.123965: step: 824/533, loss: 0.8457459211349487 2023-01-21 08:41:12.281638: step: 828/533, loss: 0.5066772699356079 2023-01-21 08:41:13.553398: step: 832/533, loss: 0.5556012988090515 2023-01-21 08:41:14.722230: step: 836/533, loss: 0.2553671896457672 2023-01-21 08:41:15.872635: step: 840/533, loss: 1.140232801437378 2023-01-21 08:41:17.061075: step: 844/533, loss: 1.0770596265792847 2023-01-21 08:41:18.253366: step: 848/533, loss: 0.1377793848514557 2023-01-21 08:41:19.405136: step: 852/533, loss: 0.21938037872314453 2023-01-21 08:41:20.597896: step: 856/533, loss: 1.3110233545303345 2023-01-21 08:41:21.764937: step: 860/533, loss: 0.9007653594017029 2023-01-21 08:41:22.964943: step: 864/533, loss: 0.2547755241394043 2023-01-21 08:41:24.139712: step: 868/533, loss: 7.441074371337891 2023-01-21 08:41:25.326184: step: 872/533, loss: 0.4394218325614929 2023-01-21 08:41:26.505132: step: 876/533, loss: 0.22106260061264038 2023-01-21 08:41:27.655469: step: 880/533, loss: 0.12388768792152405 2023-01-21 08:41:28.840144: step: 884/533, loss: 0.17823491990566254 2023-01-21 08:41:29.989149: step: 888/533, loss: 0.5554996728897095 2023-01-21 08:41:31.139274: step: 892/533, loss: 0.16692335903644562 2023-01-21 08:41:32.283468: step: 896/533, loss: 0.21521052718162537 2023-01-21 08:41:33.455696: step: 900/533, loss: 0.7005617022514343 2023-01-21 08:41:34.656758: step: 904/533, loss: 0.18899306654930115 2023-01-21 08:41:35.886648: step: 908/533, loss: 4.294531345367432 2023-01-21 08:41:37.053769: step: 912/533, loss: 0.3245202302932739 2023-01-21 08:41:38.207147: step: 916/533, loss: 0.3368349075317383 2023-01-21 08:41:39.379641: step: 920/533, loss: 0.09404177963733673 2023-01-21 08:41:40.537836: step: 924/533, loss: 0.4819769859313965 2023-01-21 08:41:41.669670: step: 928/533, loss: 3.002568483352661 2023-01-21 08:41:42.844072: step: 932/533, loss: 1.454717993736267 2023-01-21 08:41:44.079696: step: 936/533, loss: 0.328425794839859 2023-01-21 08:41:45.298747: step: 940/533, loss: 1.8779464960098267 2023-01-21 08:41:46.458258: step: 944/533, loss: 0.2418876737356186 2023-01-21 08:41:47.642656: step: 948/533, loss: 1.7883442640304565 2023-01-21 08:41:48.839434: step: 952/533, loss: 0.24025803804397583 2023-01-21 08:41:50.001673: step: 956/533, loss: 0.2860255241394043 2023-01-21 08:41:51.168966: step: 960/533, loss: 0.14447283744812012 2023-01-21 08:41:52.362998: step: 964/533, loss: 0.13047447800636292 2023-01-21 08:41:53.589127: step: 968/533, loss: 2.645369052886963 2023-01-21 08:41:54.793798: step: 972/533, loss: 0.7749943733215332 2023-01-21 08:41:55.960247: step: 976/533, loss: 0.20647449791431427 2023-01-21 08:41:57.155708: step: 980/533, loss: 0.49702727794647217 2023-01-21 08:41:58.339123: step: 984/533, loss: 0.9022245407104492 2023-01-21 08:41:59.501017: step: 988/533, loss: 0.33355823159217834 2023-01-21 08:42:00.690785: step: 992/533, loss: 0.6512640118598938 2023-01-21 08:42:01.858815: step: 996/533, loss: 1.454237937927246 2023-01-21 08:42:03.036810: step: 1000/533, loss: 0.13608050346374512 2023-01-21 08:42:04.222922: step: 1004/533, loss: 0.28959551453590393 2023-01-21 08:42:05.432673: step: 1008/533, loss: 0.6602293252944946 2023-01-21 08:42:06.601965: step: 1012/533, loss: 0.35750120878219604 2023-01-21 08:42:07.765435: step: 1016/533, loss: 1.0920332670211792 2023-01-21 08:42:08.997146: step: 1020/533, loss: 0.9741709232330322 2023-01-21 08:42:10.194142: step: 1024/533, loss: 0.29904240369796753 2023-01-21 08:42:11.409030: step: 1028/533, loss: 0.8901816010475159 2023-01-21 08:42:12.553786: step: 1032/533, loss: 0.16749007999897003 2023-01-21 08:42:13.774991: step: 1036/533, loss: 0.15685425698757172 2023-01-21 08:42:14.954230: step: 1040/533, loss: 0.47795915603637695 2023-01-21 08:42:16.144529: step: 1044/533, loss: 0.37385550141334534 2023-01-21 08:42:17.303982: step: 1048/533, loss: 0.2544828951358795 2023-01-21 08:42:18.471304: step: 1052/533, loss: 0.34292542934417725 2023-01-21 08:42:19.679597: step: 1056/533, loss: 0.17216157913208008 2023-01-21 08:42:20.845588: step: 1060/533, loss: 0.20723217725753784 2023-01-21 08:42:22.014492: step: 1064/533, loss: 0.40144920349121094 2023-01-21 08:42:23.171929: step: 1068/533, loss: 0.12985210120677948 2023-01-21 08:42:24.395700: step: 1072/533, loss: 1.9000437259674072 2023-01-21 08:42:25.600553: step: 1076/533, loss: 0.8238351345062256 2023-01-21 08:42:26.732602: step: 1080/533, loss: 0.3972499966621399 2023-01-21 08:42:27.933431: step: 1084/533, loss: 0.8402547836303711 2023-01-21 08:42:29.123066: step: 1088/533, loss: 2.5658740997314453 2023-01-21 08:42:30.324297: step: 1092/533, loss: 0.20307406783103943 2023-01-21 08:42:31.503530: step: 1096/533, loss: 0.7247675061225891 2023-01-21 08:42:32.671359: step: 1100/533, loss: 0.15931224822998047 2023-01-21 08:42:33.864987: step: 1104/533, loss: 1.7298507690429688 2023-01-21 08:42:35.051662: step: 1108/533, loss: 0.23708420991897583 2023-01-21 08:42:36.255820: step: 1112/533, loss: 1.1382371187210083 2023-01-21 08:42:37.455346: step: 1116/533, loss: 0.36733686923980713 2023-01-21 08:42:38.663895: step: 1120/533, loss: 0.12278571724891663 2023-01-21 08:42:39.845445: step: 1124/533, loss: 1.1160228252410889 2023-01-21 08:42:41.036116: step: 1128/533, loss: 0.4858914315700531 2023-01-21 08:42:42.232326: step: 1132/533, loss: 2.625290632247925 2023-01-21 08:42:43.399354: step: 1136/533, loss: 0.41810664534568787 2023-01-21 08:42:44.547885: step: 1140/533, loss: 1.226264476776123 2023-01-21 08:42:45.739148: step: 1144/533, loss: 0.3684443235397339 2023-01-21 08:42:46.926118: step: 1148/533, loss: 0.17214256525039673 2023-01-21 08:42:48.121162: step: 1152/533, loss: 0.24763727188110352 2023-01-21 08:42:49.309968: step: 1156/533, loss: 0.3016742765903473 2023-01-21 08:42:50.496310: step: 1160/533, loss: 0.3701697885990143 2023-01-21 08:42:51.664362: step: 1164/533, loss: 0.11853313446044922 2023-01-21 08:42:52.830353: step: 1168/533, loss: 0.31025782227516174 2023-01-21 08:42:53.993036: step: 1172/533, loss: 0.530258059501648 2023-01-21 08:42:55.146544: step: 1176/533, loss: 1.364376187324524 2023-01-21 08:42:56.327658: step: 1180/533, loss: 0.8632562160491943 2023-01-21 08:42:57.522319: step: 1184/533, loss: 1.4107623100280762 2023-01-21 08:42:58.687599: step: 1188/533, loss: 0.14337699115276337 2023-01-21 08:42:59.877511: step: 1192/533, loss: 2.232888698577881 2023-01-21 08:43:01.040049: step: 1196/533, loss: 0.3871815800666809 2023-01-21 08:43:02.220377: step: 1200/533, loss: 0.8312267065048218 2023-01-21 08:43:03.400511: step: 1204/533, loss: 0.2820187211036682 2023-01-21 08:43:04.585815: step: 1208/533, loss: 0.21394038200378418 2023-01-21 08:43:05.760909: step: 1212/533, loss: 0.16216059029102325 2023-01-21 08:43:06.935944: step: 1216/533, loss: 0.8990379571914673 2023-01-21 08:43:08.071142: step: 1220/533, loss: 0.13687849044799805 2023-01-21 08:43:09.256418: step: 1224/533, loss: 0.7648783326148987 2023-01-21 08:43:10.446424: step: 1228/533, loss: 0.6988527774810791 2023-01-21 08:43:11.622886: step: 1232/533, loss: 0.8141835927963257 2023-01-21 08:43:12.801037: step: 1236/533, loss: 0.06783771514892578 2023-01-21 08:43:14.007436: step: 1240/533, loss: 1.1130763292312622 2023-01-21 08:43:15.240981: step: 1244/533, loss: 0.2957499623298645 2023-01-21 08:43:16.449984: step: 1248/533, loss: 0.4456924498081207 2023-01-21 08:43:17.603429: step: 1252/533, loss: 0.47412949800491333 2023-01-21 08:43:18.756967: step: 1256/533, loss: 1.2063266038894653 2023-01-21 08:43:19.940729: step: 1260/533, loss: 1.640655517578125 2023-01-21 08:43:21.104383: step: 1264/533, loss: 0.16568291187286377 2023-01-21 08:43:22.298498: step: 1268/533, loss: 0.24979759752750397 2023-01-21 08:43:23.444310: step: 1272/533, loss: 0.13588576018810272 2023-01-21 08:43:24.678134: step: 1276/533, loss: 0.24857358634471893 2023-01-21 08:43:25.898645: step: 1280/533, loss: 0.29324567317962646 2023-01-21 08:43:27.056535: step: 1284/533, loss: 0.13492503762245178 2023-01-21 08:43:28.268634: step: 1288/533, loss: 0.383827805519104 2023-01-21 08:43:29.437240: step: 1292/533, loss: 0.3037358522415161 2023-01-21 08:43:30.624666: step: 1296/533, loss: 0.23564472794532776 2023-01-21 08:43:31.782101: step: 1300/533, loss: 0.15794897079467773 2023-01-21 08:43:32.977427: step: 1304/533, loss: 0.1066683828830719 2023-01-21 08:43:34.142909: step: 1308/533, loss: 0.4907299280166626 2023-01-21 08:43:35.328748: step: 1312/533, loss: 0.3331003189086914 2023-01-21 08:43:36.526749: step: 1316/533, loss: 0.1393260657787323 2023-01-21 08:43:37.709221: step: 1320/533, loss: 0.5686827898025513 2023-01-21 08:43:38.887311: step: 1324/533, loss: 0.7912129759788513 2023-01-21 08:43:40.048355: step: 1328/533, loss: 0.4062058925628662 2023-01-21 08:43:41.259563: step: 1332/533, loss: 0.3385017514228821 2023-01-21 08:43:42.428674: step: 1336/533, loss: 0.2562173008918762 2023-01-21 08:43:43.648333: step: 1340/533, loss: 0.20449256896972656 2023-01-21 08:43:44.819870: step: 1344/533, loss: 0.8572694063186646 2023-01-21 08:43:46.101125: step: 1348/533, loss: 0.605523407459259 2023-01-21 08:43:47.268632: step: 1352/533, loss: 0.7122437357902527 2023-01-21 08:43:48.460395: step: 1356/533, loss: 0.5115454792976379 2023-01-21 08:43:49.644598: step: 1360/533, loss: 0.3410715162754059 2023-01-21 08:43:50.868821: step: 1364/533, loss: 0.28134116530418396 2023-01-21 08:43:52.035226: step: 1368/533, loss: 0.24053707718849182 2023-01-21 08:43:53.230635: step: 1372/533, loss: 0.13209930062294006 2023-01-21 08:43:54.368373: step: 1376/533, loss: 0.6349261403083801 2023-01-21 08:43:55.554641: step: 1380/533, loss: 0.20161685347557068 2023-01-21 08:43:56.758681: step: 1384/533, loss: 1.0845767259597778 2023-01-21 08:43:57.962506: step: 1388/533, loss: 0.20318767428398132 2023-01-21 08:43:59.176811: step: 1392/533, loss: 1.6933789253234863 2023-01-21 08:44:00.341075: step: 1396/533, loss: 1.3705992698669434 2023-01-21 08:44:01.495298: step: 1400/533, loss: 0.3725167214870453 2023-01-21 08:44:02.686833: step: 1404/533, loss: 0.3230894207954407 2023-01-21 08:44:03.895897: step: 1408/533, loss: 0.19021745026111603 2023-01-21 08:44:05.061752: step: 1412/533, loss: 1.5556437969207764 2023-01-21 08:44:06.248313: step: 1416/533, loss: 1.4127694368362427 2023-01-21 08:44:07.457758: step: 1420/533, loss: 0.8863551020622253 2023-01-21 08:44:08.650109: step: 1424/533, loss: 0.11446066200733185 2023-01-21 08:44:09.819742: step: 1428/533, loss: 0.4817333221435547 2023-01-21 08:44:10.972307: step: 1432/533, loss: 0.41757750511169434 2023-01-21 08:44:12.112058: step: 1436/533, loss: 0.8214594721794128 2023-01-21 08:44:13.306530: step: 1440/533, loss: 0.3887789845466614 2023-01-21 08:44:14.489528: step: 1444/533, loss: 0.5875848531723022 2023-01-21 08:44:15.648215: step: 1448/533, loss: 7.613314628601074 2023-01-21 08:44:16.767360: step: 1452/533, loss: 0.34646815061569214 2023-01-21 08:44:17.918594: step: 1456/533, loss: 0.3636614680290222 2023-01-21 08:44:19.100649: step: 1460/533, loss: 0.5438183546066284 2023-01-21 08:44:20.264899: step: 1464/533, loss: 1.1973438262939453 2023-01-21 08:44:21.443886: step: 1468/533, loss: 0.18392372131347656 2023-01-21 08:44:22.627416: step: 1472/533, loss: 0.4506836533546448 2023-01-21 08:44:23.814927: step: 1476/533, loss: 0.1776662915945053 2023-01-21 08:44:24.960480: step: 1480/533, loss: 0.3532768189907074 2023-01-21 08:44:26.114371: step: 1484/533, loss: 0.15288801491260529 2023-01-21 08:44:27.275856: step: 1488/533, loss: 0.4506000578403473 2023-01-21 08:44:28.438712: step: 1492/533, loss: 0.51244056224823 2023-01-21 08:44:29.585215: step: 1496/533, loss: 0.2831088602542877 2023-01-21 08:44:30.763466: step: 1500/533, loss: 0.10649213939905167 2023-01-21 08:44:32.045482: step: 1504/533, loss: 0.09102578461170197 2023-01-21 08:44:33.192525: step: 1508/533, loss: 0.9720889925956726 2023-01-21 08:44:34.380868: step: 1512/533, loss: 0.5511001944541931 2023-01-21 08:44:35.621996: step: 1516/533, loss: 0.21419945359230042 2023-01-21 08:44:36.741625: step: 1520/533, loss: 0.9294561147689819 2023-01-21 08:44:37.903797: step: 1524/533, loss: 0.2315596640110016 2023-01-21 08:44:39.109588: step: 1528/533, loss: 0.1455668956041336 2023-01-21 08:44:40.312134: step: 1532/533, loss: 0.5210424661636353 2023-01-21 08:44:41.528960: step: 1536/533, loss: 0.1889040619134903 2023-01-21 08:44:42.737262: step: 1540/533, loss: 0.9127049446105957 2023-01-21 08:44:43.915989: step: 1544/533, loss: 0.13219304382801056 2023-01-21 08:44:45.083446: step: 1548/533, loss: 1.001331090927124 2023-01-21 08:44:46.253655: step: 1552/533, loss: 1.5616778135299683 2023-01-21 08:44:47.494196: step: 1556/533, loss: 1.5641981363296509 2023-01-21 08:44:48.662729: step: 1560/533, loss: 1.0800390243530273 2023-01-21 08:44:49.901555: step: 1564/533, loss: 0.19683510065078735 2023-01-21 08:44:51.110242: step: 1568/533, loss: 0.32040977478027344 2023-01-21 08:44:52.303366: step: 1572/533, loss: 0.7974711060523987 2023-01-21 08:44:53.475362: step: 1576/533, loss: 0.30020105838775635 2023-01-21 08:44:54.656140: step: 1580/533, loss: 0.18065744638442993 2023-01-21 08:44:55.815531: step: 1584/533, loss: 0.20464010536670685 2023-01-21 08:44:57.085229: step: 1588/533, loss: 0.9827674627304077 2023-01-21 08:44:58.275430: step: 1592/533, loss: 2.80283522605896 2023-01-21 08:44:59.422498: step: 1596/533, loss: 1.3079277276992798 2023-01-21 08:45:00.611417: step: 1600/533, loss: 1.040922999382019 2023-01-21 08:45:01.777643: step: 1604/533, loss: 6.412429332733154 2023-01-21 08:45:02.929306: step: 1608/533, loss: 0.1268180012702942 2023-01-21 08:45:04.113987: step: 1612/533, loss: 1.9018254280090332 2023-01-21 08:45:05.276132: step: 1616/533, loss: 0.20163805782794952 2023-01-21 08:45:06.464640: step: 1620/533, loss: 0.9594154953956604 2023-01-21 08:45:07.671687: step: 1624/533, loss: 0.5759618282318115 2023-01-21 08:45:08.841043: step: 1628/533, loss: 0.8610288500785828 2023-01-21 08:45:10.019845: step: 1632/533, loss: 0.1434132605791092 2023-01-21 08:45:11.232738: step: 1636/533, loss: 0.20625987648963928 2023-01-21 08:45:12.416724: step: 1640/533, loss: 2.2320892810821533 2023-01-21 08:45:13.585939: step: 1644/533, loss: 0.5052288174629211 2023-01-21 08:45:14.787313: step: 1648/533, loss: 0.6492370367050171 2023-01-21 08:45:15.968500: step: 1652/533, loss: 0.788342297077179 2023-01-21 08:45:17.130768: step: 1656/533, loss: 0.18628616631031036 2023-01-21 08:45:18.293137: step: 1660/533, loss: 0.18082067370414734 2023-01-21 08:45:19.466503: step: 1664/533, loss: 0.1602247655391693 2023-01-21 08:45:20.644633: step: 1668/533, loss: 0.17520609498023987 2023-01-21 08:45:21.811389: step: 1672/533, loss: 0.6110613346099854 2023-01-21 08:45:22.963696: step: 1676/533, loss: 0.6510634422302246 2023-01-21 08:45:24.142034: step: 1680/533, loss: 0.7063851356506348 2023-01-21 08:45:25.268760: step: 1684/533, loss: 0.6892544031143188 2023-01-21 08:45:26.457678: step: 1688/533, loss: 0.49946165084838867 2023-01-21 08:45:27.633631: step: 1692/533, loss: 0.7132205963134766 2023-01-21 08:45:28.856427: step: 1696/533, loss: 0.1858876347541809 2023-01-21 08:45:30.056103: step: 1700/533, loss: 1.2561460733413696 2023-01-21 08:45:31.279998: step: 1704/533, loss: 0.9905737042427063 2023-01-21 08:45:32.463011: step: 1708/533, loss: 0.33987924456596375 2023-01-21 08:45:33.625969: step: 1712/533, loss: 0.7124394774436951 2023-01-21 08:45:34.827565: step: 1716/533, loss: 0.15509900450706482 2023-01-21 08:45:36.048593: step: 1720/533, loss: 0.4252755045890808 2023-01-21 08:45:37.272500: step: 1724/533, loss: 0.21397143602371216 2023-01-21 08:45:38.449095: step: 1728/533, loss: 0.2233814150094986 2023-01-21 08:45:39.618719: step: 1732/533, loss: 0.14459353685379028 2023-01-21 08:45:40.825780: step: 1736/533, loss: 0.43625637888908386 2023-01-21 08:45:41.988419: step: 1740/533, loss: 0.18639174103736877 2023-01-21 08:45:43.159429: step: 1744/533, loss: 0.1797471046447754 2023-01-21 08:45:44.327414: step: 1748/533, loss: 0.3187692165374756 2023-01-21 08:45:45.498499: step: 1752/533, loss: 0.22615423798561096 2023-01-21 08:45:46.646823: step: 1756/533, loss: 0.8571557998657227 2023-01-21 08:45:47.845569: step: 1760/533, loss: 6.347506046295166 2023-01-21 08:45:49.057978: step: 1764/533, loss: 1.7205748558044434 2023-01-21 08:45:50.266403: step: 1768/533, loss: 1.0612514019012451 2023-01-21 08:45:51.447812: step: 1772/533, loss: 0.2512054443359375 2023-01-21 08:45:52.623602: step: 1776/533, loss: 0.8201772570610046 2023-01-21 08:45:53.838410: step: 1780/533, loss: 0.17684555053710938 2023-01-21 08:45:55.037578: step: 1784/533, loss: 0.7339959144592285 2023-01-21 08:45:56.273816: step: 1788/533, loss: 7.235861301422119 2023-01-21 08:45:57.463986: step: 1792/533, loss: 0.14361485838890076 2023-01-21 08:45:58.616377: step: 1796/533, loss: 0.20178404450416565 2023-01-21 08:45:59.798336: step: 1800/533, loss: 0.447252094745636 2023-01-21 08:46:00.959382: step: 1804/533, loss: 0.33057937026023865 2023-01-21 08:46:02.165260: step: 1808/533, loss: 0.8879483938217163 2023-01-21 08:46:03.386015: step: 1812/533, loss: 1.0193923711776733 2023-01-21 08:46:04.551376: step: 1816/533, loss: 0.553665041923523 2023-01-21 08:46:05.726505: step: 1820/533, loss: 0.8712189793586731 2023-01-21 08:46:06.925658: step: 1824/533, loss: 0.15548363327980042 2023-01-21 08:46:08.135580: step: 1828/533, loss: 0.5508195757865906 2023-01-21 08:46:09.331007: step: 1832/533, loss: 0.11083962768316269 2023-01-21 08:46:10.532711: step: 1836/533, loss: 0.3949201703071594 2023-01-21 08:46:11.681991: step: 1840/533, loss: 0.3378902077674866 2023-01-21 08:46:12.863450: step: 1844/533, loss: 0.21268385648727417 2023-01-21 08:46:14.098239: step: 1848/533, loss: 0.14636117219924927 2023-01-21 08:46:15.313007: step: 1852/533, loss: 1.2633007764816284 2023-01-21 08:46:16.530649: step: 1856/533, loss: 0.3618050813674927 2023-01-21 08:46:17.730775: step: 1860/533, loss: 0.7293722033500671 2023-01-21 08:46:18.913298: step: 1864/533, loss: 0.18918466567993164 2023-01-21 08:46:20.119320: step: 1868/533, loss: 2.40671968460083 2023-01-21 08:46:21.297008: step: 1872/533, loss: 0.11944542825222015 2023-01-21 08:46:22.459523: step: 1876/533, loss: 0.2083059400320053 2023-01-21 08:46:23.628501: step: 1880/533, loss: 0.1916859745979309 2023-01-21 08:46:24.811567: step: 1884/533, loss: 0.35312873125076294 2023-01-21 08:46:26.000164: step: 1888/533, loss: 0.5087341070175171 2023-01-21 08:46:27.144114: step: 1892/533, loss: 0.44750314950942993 2023-01-21 08:46:28.325830: step: 1896/533, loss: 0.7753688097000122 2023-01-21 08:46:29.546115: step: 1900/533, loss: 0.36229783296585083 2023-01-21 08:46:30.718999: step: 1904/533, loss: 0.6711670756340027 2023-01-21 08:46:31.883844: step: 1908/533, loss: 0.7190624475479126 2023-01-21 08:46:33.074158: step: 1912/533, loss: 1.3056414127349854 2023-01-21 08:46:34.232704: step: 1916/533, loss: 0.17223167419433594 2023-01-21 08:46:35.401445: step: 1920/533, loss: 0.07773604989051819 2023-01-21 08:46:36.603462: step: 1924/533, loss: 0.0669272243976593 2023-01-21 08:46:37.781861: step: 1928/533, loss: 0.08887290954589844 2023-01-21 08:46:38.946329: step: 1932/533, loss: 0.853316068649292 2023-01-21 08:46:40.166801: step: 1936/533, loss: 1.3761448860168457 2023-01-21 08:46:41.370078: step: 1940/533, loss: 0.2315848469734192 2023-01-21 08:46:42.528182: step: 1944/533, loss: 0.2900312542915344 2023-01-21 08:46:43.705450: step: 1948/533, loss: 0.26392853260040283 2023-01-21 08:46:44.916084: step: 1952/533, loss: 0.1180468499660492 2023-01-21 08:46:46.147231: step: 1956/533, loss: 0.49521002173423767 2023-01-21 08:46:47.401488: step: 1960/533, loss: 0.07162399590015411 2023-01-21 08:46:48.592067: step: 1964/533, loss: 0.1672719419002533 2023-01-21 08:46:49.780095: step: 1968/533, loss: 0.08329484611749649 2023-01-21 08:46:50.935303: step: 1972/533, loss: 0.05844435840845108 2023-01-21 08:46:52.116722: step: 1976/533, loss: 0.2748605012893677 2023-01-21 08:46:53.335237: step: 1980/533, loss: 0.18915005028247833 2023-01-21 08:46:54.534486: step: 1984/533, loss: 0.2961672246456146 2023-01-21 08:46:55.773566: step: 1988/533, loss: 2.322105884552002 2023-01-21 08:46:56.973381: step: 1992/533, loss: 0.9499852657318115 2023-01-21 08:46:58.146057: step: 1996/533, loss: 0.2910190224647522 2023-01-21 08:46:59.344225: step: 2000/533, loss: 0.6286906003952026 2023-01-21 08:47:00.575470: step: 2004/533, loss: 0.20375236868858337 2023-01-21 08:47:01.742830: step: 2008/533, loss: 0.16577622294425964 2023-01-21 08:47:02.991679: step: 2012/533, loss: 0.33581504225730896 2023-01-21 08:47:04.209822: step: 2016/533, loss: 1.2614092826843262 2023-01-21 08:47:05.439177: step: 2020/533, loss: 0.26006898283958435 2023-01-21 08:47:06.626978: step: 2024/533, loss: 0.21648313105106354 2023-01-21 08:47:07.765460: step: 2028/533, loss: 0.08915567398071289 2023-01-21 08:47:08.958033: step: 2032/533, loss: 0.2993401885032654 2023-01-21 08:47:10.175521: step: 2036/533, loss: 0.27792254090309143 2023-01-21 08:47:11.336471: step: 2040/533, loss: 0.5655492544174194 2023-01-21 08:47:12.530526: step: 2044/533, loss: 0.7827044725418091 2023-01-21 08:47:13.710363: step: 2048/533, loss: 0.7342962622642517 2023-01-21 08:47:14.879978: step: 2052/533, loss: 0.17950135469436646 2023-01-21 08:47:16.080277: step: 2056/533, loss: 0.570751428604126 2023-01-21 08:47:17.286233: step: 2060/533, loss: 1.1367592811584473 2023-01-21 08:47:18.457159: step: 2064/533, loss: 0.31864356994628906 2023-01-21 08:47:19.656816: step: 2068/533, loss: 0.1877949833869934 2023-01-21 08:47:20.863454: step: 2072/533, loss: 1.6193100214004517 2023-01-21 08:47:22.024753: step: 2076/533, loss: 0.4787091314792633 2023-01-21 08:47:23.204076: step: 2080/533, loss: 0.1201920285820961 2023-01-21 08:47:24.415734: step: 2084/533, loss: 1.5718034505844116 2023-01-21 08:47:25.593766: step: 2088/533, loss: 0.5335476994514465 2023-01-21 08:47:26.801726: step: 2092/533, loss: 1.9365079402923584 2023-01-21 08:47:27.984857: step: 2096/533, loss: 0.15172234177589417 2023-01-21 08:47:29.153766: step: 2100/533, loss: 0.3086782395839691 2023-01-21 08:47:30.346409: step: 2104/533, loss: 1.6743295192718506 2023-01-21 08:47:31.515183: step: 2108/533, loss: 0.2194240391254425 2023-01-21 08:47:32.690183: step: 2112/533, loss: 0.04186415672302246 2023-01-21 08:47:33.906320: step: 2116/533, loss: 2.0739517211914062 2023-01-21 08:47:35.147308: step: 2120/533, loss: 0.9264987707138062 2023-01-21 08:47:36.329437: step: 2124/533, loss: 0.15454182028770447 2023-01-21 08:47:37.525328: step: 2128/533, loss: 0.5753645896911621 2023-01-21 08:47:38.691787: step: 2132/533, loss: 0.22760553658008575 ================================================== Loss: 0.780 -------------------- Dev: {'event': {'p': 0.49306197964847365, 'r': 0.7097203728362184, 'f1': 0.5818777292576419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Test: {'event': {'p': 0.5165719696969697, 'r': 0.6975703324808185, 'f1': 0.5935799782372144}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Chinese: {'event': {'p': 0.449438202247191, 'r': 0.7407407407407407, 'f1': 0.5594405594405595}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Korean: {'event': {'p': 0.5535714285714286, 'r': 0.49206349206349204, 'f1': 0.5210084033613446}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} 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': 1} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.49306197964847365, 'r': 0.7097203728362184, 'f1': 0.5818777292576419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Chinese: {'event': {'p': 0.5165719696969697, 'r': 0.6975703324808185, 'f1': 0.5935799782372144}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Chinese: {'event': {'p': 0.449438202247191, 'r': 0.7407407407407407, 'f1': 0.5594405594405595}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Korean: {'event': {'p': 0.49306197964847365, 'r': 0.7097203728362184, 'f1': 0.5818777292576419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5165719696969697, 'r': 0.6975703324808185, 'f1': 0.5935799782372144}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.5535714285714286, 'r': 0.49206349206349204, 'f1': 0.5210084033613446}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.49306197964847365, 'r': 0.7097203728362184, 'f1': 0.5818777292576419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5165719696969697, 'r': 0.6975703324808185, 'f1': 0.5935799782372144}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample 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': 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 08:48:35.433383: step: 4/533, loss: 0.2273930162191391 2023-01-21 08:48:36.595863: step: 8/533, loss: 0.11745396256446838 2023-01-21 08:48:37.797762: step: 12/533, loss: 0.8195672631263733 2023-01-21 08:48:39.010178: step: 16/533, loss: 0.9431135058403015 2023-01-21 08:48:40.221327: step: 20/533, loss: 1.7726690769195557 2023-01-21 08:48:41.378646: step: 24/533, loss: 0.18240299820899963 2023-01-21 08:48:42.528601: step: 28/533, loss: 0.5393819808959961 2023-01-21 08:48:43.695907: step: 32/533, loss: 0.35707321763038635 2023-01-21 08:48:44.861194: step: 36/533, loss: 0.1625724732875824 2023-01-21 08:48:46.100473: step: 40/533, loss: 0.444752961397171 2023-01-21 08:48:47.279710: step: 44/533, loss: 0.6659727096557617 2023-01-21 08:48:48.490184: step: 48/533, loss: 0.41809549927711487 2023-01-21 08:48:49.696196: step: 52/533, loss: 0.14960438013076782 2023-01-21 08:48:50.887395: step: 56/533, loss: 0.4455803334712982 2023-01-21 08:48:52.036436: step: 60/533, loss: 0.11054739356040955 2023-01-21 08:48:53.212091: step: 64/533, loss: 0.15095539391040802 2023-01-21 08:48:54.420885: step: 68/533, loss: 0.20855793356895447 2023-01-21 08:48:55.613813: step: 72/533, loss: 0.09738487750291824 2023-01-21 08:48:56.794966: step: 76/533, loss: 0.1486688256263733 2023-01-21 08:48:58.015149: step: 80/533, loss: 0.2405291646718979 2023-01-21 08:48:59.192556: step: 84/533, loss: 0.18646469712257385 2023-01-21 08:49:00.422571: step: 88/533, loss: 0.8988907337188721 2023-01-21 08:49:01.597448: step: 92/533, loss: 0.5040372014045715 2023-01-21 08:49:02.818444: step: 96/533, loss: 1.945794701576233 2023-01-21 08:49:04.003792: step: 100/533, loss: 0.07698164135217667 2023-01-21 08:49:05.189567: step: 104/533, loss: 0.28211650252342224 2023-01-21 08:49:06.402673: step: 108/533, loss: 1.2997928857803345 2023-01-21 08:49:07.602713: step: 112/533, loss: 0.8248919248580933 2023-01-21 08:49:08.782530: step: 116/533, loss: 0.1631982922554016 2023-01-21 08:49:09.928942: step: 120/533, loss: 0.3366301655769348 2023-01-21 08:49:11.116189: step: 124/533, loss: 0.22430419921875 2023-01-21 08:49:12.302197: step: 128/533, loss: 1.0289384126663208 2023-01-21 08:49:13.467534: step: 132/533, loss: 0.3962440490722656 2023-01-21 08:49:14.635546: step: 136/533, loss: 0.24397139251232147 2023-01-21 08:49:15.825804: step: 140/533, loss: 0.9660804271697998 2023-01-21 08:49:17.029567: step: 144/533, loss: 0.35625410079956055 2023-01-21 08:49:18.259567: step: 148/533, loss: 1.0034176111221313 2023-01-21 08:49:19.457630: step: 152/533, loss: 0.060857679694890976 2023-01-21 08:49:20.673591: step: 156/533, loss: 0.2796187400817871 2023-01-21 08:49:21.829102: step: 160/533, loss: 0.06856928020715714 2023-01-21 08:49:23.015950: step: 164/533, loss: 0.42069217562675476 2023-01-21 08:49:24.221238: step: 168/533, loss: 1.717408299446106 2023-01-21 08:49:25.404110: step: 172/533, loss: 0.53558349609375 2023-01-21 08:49:26.552041: step: 176/533, loss: 0.12073850631713867 2023-01-21 08:49:27.725658: step: 180/533, loss: 0.13690312206745148 2023-01-21 08:49:28.960790: step: 184/533, loss: 1.05412757396698 2023-01-21 08:49:30.180695: step: 188/533, loss: 6.908061981201172 2023-01-21 08:49:31.325641: step: 192/533, loss: 0.5959917902946472 2023-01-21 08:49:32.473043: step: 196/533, loss: 0.07897396385669708 2023-01-21 08:49:33.668721: step: 200/533, loss: 3.012596368789673 2023-01-21 08:49:34.885024: step: 204/533, loss: 0.21523308753967285 2023-01-21 08:49:36.060794: step: 208/533, loss: 0.12647505104541779 2023-01-21 08:49:37.234817: step: 212/533, loss: 0.6816475987434387 2023-01-21 08:49:38.417448: step: 216/533, loss: 0.11413614451885223 2023-01-21 08:49:39.617930: step: 220/533, loss: 0.7274017930030823 2023-01-21 08:49:40.800502: step: 224/533, loss: 0.43675634264945984 2023-01-21 08:49:41.964823: step: 228/533, loss: 0.7292206883430481 2023-01-21 08:49:43.111604: step: 232/533, loss: 0.16568660736083984 2023-01-21 08:49:44.316876: step: 236/533, loss: 0.2401791661977768 2023-01-21 08:49:45.522374: step: 240/533, loss: 1.20443856716156 2023-01-21 08:49:46.707855: step: 244/533, loss: 0.08509984612464905 2023-01-21 08:49:47.931103: step: 248/533, loss: 1.643349528312683 2023-01-21 08:49:49.139055: step: 252/533, loss: 0.22615928947925568 2023-01-21 08:49:50.338698: step: 256/533, loss: 0.05160102993249893 2023-01-21 08:49:51.505436: step: 260/533, loss: 0.390639990568161 2023-01-21 08:49:52.703133: step: 264/533, loss: 0.14797353744506836 2023-01-21 08:49:53.917114: step: 268/533, loss: 0.9673904776573181 2023-01-21 08:49:55.136067: step: 272/533, loss: 0.3699526786804199 2023-01-21 08:49:56.316543: step: 276/533, loss: 0.636673092842102 2023-01-21 08:49:57.487370: step: 280/533, loss: 1.3742446899414062 2023-01-21 08:49:58.659009: step: 284/533, loss: 1.0109410285949707 2023-01-21 08:49:59.850958: step: 288/533, loss: 0.38311767578125 2023-01-21 08:50:01.064644: step: 292/533, loss: 0.47721540927886963 2023-01-21 08:50:02.268674: step: 296/533, loss: 1.3345869779586792 2023-01-21 08:50:03.504753: step: 300/533, loss: 0.8420354723930359 2023-01-21 08:50:04.727957: step: 304/533, loss: 0.24124260246753693 2023-01-21 08:50:05.933406: step: 308/533, loss: 0.23484164476394653 2023-01-21 08:50:07.116954: step: 312/533, loss: 0.303290456533432 2023-01-21 08:50:08.319546: step: 316/533, loss: 0.8358780145645142 2023-01-21 08:50:09.510430: step: 320/533, loss: 0.13641862571239471 2023-01-21 08:50:10.755055: step: 324/533, loss: 1.2870972156524658 2023-01-21 08:50:11.963013: step: 328/533, loss: 0.7284207940101624 2023-01-21 08:50:13.133091: step: 332/533, loss: 0.5092397928237915 2023-01-21 08:50:14.334457: step: 336/533, loss: 0.892011284828186 2023-01-21 08:50:15.556140: step: 340/533, loss: 0.8644672632217407 2023-01-21 08:50:16.775492: step: 344/533, loss: 0.04526777192950249 2023-01-21 08:50:17.966133: step: 348/533, loss: 0.3578372895717621 2023-01-21 08:50:19.164886: step: 352/533, loss: 1.10097336769104 2023-01-21 08:50:20.367021: step: 356/533, loss: 0.7220709323883057 2023-01-21 08:50:21.522042: step: 360/533, loss: 0.1650390625 2023-01-21 08:50:22.716772: step: 364/533, loss: 0.19318580627441406 2023-01-21 08:50:23.874808: step: 368/533, loss: 1.388374924659729 2023-01-21 08:50:25.010763: step: 372/533, loss: 0.20563922822475433 2023-01-21 08:50:26.231198: step: 376/533, loss: 1.1659611463546753 2023-01-21 08:50:27.416923: step: 380/533, loss: 0.49511605501174927 2023-01-21 08:50:28.576404: step: 384/533, loss: 0.08295345306396484 2023-01-21 08:50:29.763695: step: 388/533, loss: 0.7571626305580139 2023-01-21 08:50:30.966177: step: 392/533, loss: 0.6921822428703308 2023-01-21 08:50:32.156447: step: 396/533, loss: 0.17010803520679474 2023-01-21 08:50:33.344469: step: 400/533, loss: 0.20108766853809357 2023-01-21 08:50:34.579092: step: 404/533, loss: 1.3238558769226074 2023-01-21 08:50:35.784530: step: 408/533, loss: 1.0351576805114746 2023-01-21 08:50:36.965254: step: 412/533, loss: 0.37799471616744995 2023-01-21 08:50:38.121806: step: 416/533, loss: 0.7027173042297363 2023-01-21 08:50:39.296959: step: 420/533, loss: 0.11830691993236542 2023-01-21 08:50:40.490170: step: 424/533, loss: 0.2692562937736511 2023-01-21 08:50:41.683033: step: 428/533, loss: 0.0980944111943245 2023-01-21 08:50:42.882234: step: 432/533, loss: 0.372312068939209 2023-01-21 08:50:44.044830: step: 436/533, loss: 0.39765799045562744 2023-01-21 08:50:45.235805: step: 440/533, loss: 0.48756155371665955 2023-01-21 08:50:46.409956: step: 444/533, loss: 0.08249840885400772 2023-01-21 08:50:47.605380: step: 448/533, loss: 1.5067517757415771 2023-01-21 08:50:48.798111: step: 452/533, loss: 0.2680312991142273 2023-01-21 08:50:49.996342: step: 456/533, loss: 1.0247547626495361 2023-01-21 08:50:51.191266: step: 460/533, loss: 0.25795361399650574 2023-01-21 08:50:52.362719: step: 464/533, loss: 0.6210354566574097 2023-01-21 08:50:53.553529: step: 468/533, loss: 0.06505861133337021 2023-01-21 08:50:54.758647: step: 472/533, loss: 0.9583412408828735 2023-01-21 08:50:55.898636: step: 476/533, loss: 0.1402682363986969 2023-01-21 08:50:57.074919: step: 480/533, loss: 0.7231633067131042 2023-01-21 08:50:58.263636: step: 484/533, loss: 0.6232940554618835 2023-01-21 08:50:59.449602: step: 488/533, loss: 0.28297579288482666 2023-01-21 08:51:00.621365: step: 492/533, loss: 0.1705310344696045 2023-01-21 08:51:01.812333: step: 496/533, loss: 0.23556914925575256 2023-01-21 08:51:02.996084: step: 500/533, loss: 0.35297250747680664 2023-01-21 08:51:04.124618: step: 504/533, loss: 0.06654729694128036 2023-01-21 08:51:05.320382: step: 508/533, loss: 0.29731130599975586 2023-01-21 08:51:06.525204: step: 512/533, loss: 0.7469516396522522 2023-01-21 08:51:07.676923: step: 516/533, loss: 0.47540247440338135 2023-01-21 08:51:08.881143: step: 520/533, loss: 0.27656564116477966 2023-01-21 08:51:10.028519: step: 524/533, loss: 0.7852625250816345 2023-01-21 08:51:11.234011: step: 528/533, loss: 0.48299741744995117 2023-01-21 08:51:12.424648: step: 532/533, loss: 0.49583935737609863 2023-01-21 08:51:13.612308: step: 536/533, loss: 0.238701730966568 2023-01-21 08:51:14.806381: step: 540/533, loss: 0.1043207198381424 2023-01-21 08:51:16.007616: step: 544/533, loss: 1.0323508977890015 2023-01-21 08:51:17.165276: step: 548/533, loss: 0.24594631791114807 2023-01-21 08:51:18.337880: step: 552/533, loss: 0.7298108339309692 2023-01-21 08:51:19.531759: step: 556/533, loss: 0.5059745907783508 2023-01-21 08:51:20.719480: step: 560/533, loss: 0.1618526428937912 2023-01-21 08:51:21.862629: step: 564/533, loss: 0.09630686044692993 2023-01-21 08:51:23.078081: step: 568/533, loss: 0.6778080463409424 2023-01-21 08:51:24.209367: step: 572/533, loss: 0.07121472805738449 2023-01-21 08:51:25.404076: step: 576/533, loss: 0.5281625986099243 2023-01-21 08:51:26.606692: step: 580/533, loss: 0.06455035507678986 2023-01-21 08:51:27.811997: step: 584/533, loss: 7.478963851928711 2023-01-21 08:51:29.012568: step: 588/533, loss: 0.7464495897293091 2023-01-21 08:51:30.171500: step: 592/533, loss: 0.18666720390319824 2023-01-21 08:51:31.330631: step: 596/533, loss: 0.7304089665412903 2023-01-21 08:51:32.507562: step: 600/533, loss: 0.7658006548881531 2023-01-21 08:51:33.731793: step: 604/533, loss: 0.09449243545532227 2023-01-21 08:51:34.915269: step: 608/533, loss: 0.3147008419036865 2023-01-21 08:51:36.108393: step: 612/533, loss: 0.17011529207229614 2023-01-21 08:51:37.290843: step: 616/533, loss: 0.22066155076026917 2023-01-21 08:51:38.468213: step: 620/533, loss: 0.7816414833068848 2023-01-21 08:51:39.664586: step: 624/533, loss: 1.3505761623382568 2023-01-21 08:51:40.821307: step: 628/533, loss: 1.430994987487793 2023-01-21 08:51:42.020982: step: 632/533, loss: 0.2059134542942047 2023-01-21 08:51:43.186209: step: 636/533, loss: 0.6495293974876404 2023-01-21 08:51:44.390796: step: 640/533, loss: 0.7417030334472656 2023-01-21 08:51:45.564611: step: 644/533, loss: 0.4340957701206207 2023-01-21 08:51:46.704104: step: 648/533, loss: 0.3562898635864258 2023-01-21 08:51:47.861313: step: 652/533, loss: 1.0801200866699219 2023-01-21 08:51:49.028927: step: 656/533, loss: 0.21649666130542755 2023-01-21 08:51:50.193581: step: 660/533, loss: 0.1450033187866211 2023-01-21 08:51:51.360409: step: 664/533, loss: 0.8730990290641785 2023-01-21 08:51:52.538743: step: 668/533, loss: 0.17216749489307404 2023-01-21 08:51:53.716481: step: 672/533, loss: 0.16926175355911255 2023-01-21 08:51:54.898966: step: 676/533, loss: 0.6977423429489136 2023-01-21 08:51:56.087577: step: 680/533, loss: 0.4301180839538574 2023-01-21 08:51:57.284668: step: 684/533, loss: 0.199737548828125 2023-01-21 08:51:58.518455: step: 688/533, loss: 1.4486771821975708 2023-01-21 08:51:59.663008: step: 692/533, loss: 0.5181573033332825 2023-01-21 08:52:00.843956: step: 696/533, loss: 0.07946591824293137 2023-01-21 08:52:02.007989: step: 700/533, loss: 0.4264151155948639 2023-01-21 08:52:03.204049: step: 704/533, loss: 0.21281233429908752 2023-01-21 08:52:04.411883: step: 708/533, loss: 0.9650977849960327 2023-01-21 08:52:05.635155: step: 712/533, loss: 0.33989372849464417 2023-01-21 08:52:06.810006: step: 716/533, loss: 0.12522058188915253 2023-01-21 08:52:08.006965: step: 720/533, loss: 0.1704375296831131 2023-01-21 08:52:09.204870: step: 724/533, loss: 0.21190881729125977 2023-01-21 08:52:10.378435: step: 728/533, loss: 0.24946996569633484 2023-01-21 08:52:11.553889: step: 732/533, loss: 0.7441123723983765 2023-01-21 08:52:12.736091: step: 736/533, loss: 0.05891933664679527 2023-01-21 08:52:13.937815: step: 740/533, loss: 0.20319104194641113 2023-01-21 08:52:15.161837: step: 744/533, loss: 1.7787359952926636 2023-01-21 08:52:16.348613: step: 748/533, loss: 1.0156267881393433 2023-01-21 08:52:17.554064: step: 752/533, loss: 0.16084283590316772 2023-01-21 08:52:18.773130: step: 756/533, loss: 0.3467342257499695 2023-01-21 08:52:19.940169: step: 760/533, loss: 1.0924389362335205 2023-01-21 08:52:21.105719: step: 764/533, loss: 0.09603042900562286 2023-01-21 08:52:22.318181: step: 768/533, loss: 0.3103495240211487 2023-01-21 08:52:23.482337: step: 772/533, loss: 0.150017648935318 2023-01-21 08:52:24.674186: step: 776/533, loss: 0.1918269693851471 2023-01-21 08:52:25.847947: step: 780/533, loss: 0.11927299201488495 2023-01-21 08:52:27.021235: step: 784/533, loss: 0.3759794235229492 2023-01-21 08:52:28.193207: step: 788/533, loss: 0.8447084426879883 2023-01-21 08:52:29.370161: step: 792/533, loss: 0.13505783677101135 2023-01-21 08:52:30.552298: step: 796/533, loss: 0.14870567619800568 2023-01-21 08:52:31.754220: step: 800/533, loss: 0.25327959656715393 2023-01-21 08:52:32.908776: step: 804/533, loss: 0.05911578983068466 2023-01-21 08:52:34.123608: step: 808/533, loss: 1.1324101686477661 2023-01-21 08:52:35.284133: step: 812/533, loss: 0.7645180821418762 2023-01-21 08:52:36.453391: step: 816/533, loss: 0.11914453655481339 2023-01-21 08:52:37.650746: step: 820/533, loss: 0.6584722399711609 2023-01-21 08:52:38.816289: step: 824/533, loss: 0.11634072661399841 2023-01-21 08:52:40.051331: step: 828/533, loss: 0.296133816242218 2023-01-21 08:52:41.277811: step: 832/533, loss: 2.2104575634002686 2023-01-21 08:52:42.469120: step: 836/533, loss: 0.26594123244285583 2023-01-21 08:52:43.707458: step: 840/533, loss: 0.29918622970581055 2023-01-21 08:52:44.916465: step: 844/533, loss: 1.1086997985839844 2023-01-21 08:52:46.090869: step: 848/533, loss: 0.13089317083358765 2023-01-21 08:52:47.279836: step: 852/533, loss: 0.08148898929357529 2023-01-21 08:52:48.440197: step: 856/533, loss: 0.07526750862598419 2023-01-21 08:52:49.608594: step: 860/533, loss: 0.05287761986255646 2023-01-21 08:52:50.777569: step: 864/533, loss: 0.24665486812591553 2023-01-21 08:52:51.969510: step: 868/533, loss: 0.9336276650428772 2023-01-21 08:52:53.140203: step: 872/533, loss: 0.24038414657115936 2023-01-21 08:52:54.338747: step: 876/533, loss: 0.3202973008155823 2023-01-21 08:52:55.525183: step: 880/533, loss: 0.47325411438941956 2023-01-21 08:52:56.703924: step: 884/533, loss: 0.718765914440155 2023-01-21 08:52:57.866850: step: 888/533, loss: 0.05490746721625328 2023-01-21 08:52:59.023245: step: 892/533, loss: 0.4450978636741638 2023-01-21 08:53:00.234478: step: 896/533, loss: 0.13848859071731567 2023-01-21 08:53:01.405516: step: 900/533, loss: 0.4307509958744049 2023-01-21 08:53:02.553579: step: 904/533, loss: 6.461243152618408 2023-01-21 08:53:03.718517: step: 908/533, loss: 1.3066799640655518 2023-01-21 08:53:04.895961: step: 912/533, loss: 0.7175735831260681 2023-01-21 08:53:06.067785: step: 916/533, loss: 1.0465493202209473 2023-01-21 08:53:07.342971: step: 920/533, loss: 0.42484796047210693 2023-01-21 08:53:08.521379: step: 924/533, loss: 0.28290843963623047 2023-01-21 08:53:09.698493: step: 928/533, loss: 0.09106452763080597 2023-01-21 08:53:10.855904: step: 932/533, loss: 0.16259175539016724 2023-01-21 08:53:12.073169: step: 936/533, loss: 0.15368108451366425 2023-01-21 08:53:13.303978: step: 940/533, loss: 0.18690046668052673 2023-01-21 08:53:14.450049: step: 944/533, loss: 1.2590035200119019 2023-01-21 08:53:15.607319: step: 948/533, loss: 0.16667777299880981 2023-01-21 08:53:16.822785: step: 952/533, loss: 0.14748039841651917 2023-01-21 08:53:18.019303: step: 956/533, loss: 0.17940634489059448 2023-01-21 08:53:19.196635: step: 960/533, loss: 0.5820997953414917 2023-01-21 08:53:20.377833: step: 964/533, loss: 1.5327001810073853 2023-01-21 08:53:21.559556: step: 968/533, loss: 0.6402400732040405 2023-01-21 08:53:22.723444: step: 972/533, loss: 0.14987602829933167 2023-01-21 08:53:23.937631: step: 976/533, loss: 0.7734742164611816 2023-01-21 08:53:25.122333: step: 980/533, loss: 1.6735639572143555 2023-01-21 08:53:26.297719: step: 984/533, loss: 0.6829612255096436 2023-01-21 08:53:27.456670: step: 988/533, loss: 8.01281452178955 2023-01-21 08:53:28.657589: step: 992/533, loss: 0.45811301469802856 2023-01-21 08:53:29.841249: step: 996/533, loss: 0.10504613071680069 2023-01-21 08:53:31.050399: step: 1000/533, loss: 0.1916123628616333 2023-01-21 08:53:32.215699: step: 1004/533, loss: 0.23422221839427948 2023-01-21 08:53:33.437687: step: 1008/533, loss: 0.35816287994384766 2023-01-21 08:53:34.622155: step: 1012/533, loss: 0.21505865454673767 2023-01-21 08:53:35.781159: step: 1016/533, loss: 0.3315998315811157 2023-01-21 08:53:36.967659: step: 1020/533, loss: 0.08506689220666885 2023-01-21 08:53:38.139158: step: 1024/533, loss: 6.508427619934082 2023-01-21 08:53:39.307151: step: 1028/533, loss: 0.16155119240283966 2023-01-21 08:53:40.447340: step: 1032/533, loss: 0.21006269752979279 2023-01-21 08:53:41.621634: step: 1036/533, loss: 0.1542145311832428 2023-01-21 08:53:42.844379: step: 1040/533, loss: 0.24551811814308167 2023-01-21 08:53:44.022604: step: 1044/533, loss: 0.3328459858894348 2023-01-21 08:53:45.192883: step: 1048/533, loss: 0.6866517663002014 2023-01-21 08:53:46.349533: step: 1052/533, loss: 0.21656756103038788 2023-01-21 08:53:47.540808: step: 1056/533, loss: 0.16592836380004883 2023-01-21 08:53:48.739920: step: 1060/533, loss: 0.3690887689590454 2023-01-21 08:53:49.941915: step: 1064/533, loss: 0.23401856422424316 2023-01-21 08:53:51.120003: step: 1068/533, loss: 0.5108942985534668 2023-01-21 08:53:52.328464: step: 1072/533, loss: 0.09856872260570526 2023-01-21 08:53:53.478565: step: 1076/533, loss: 0.5681627988815308 2023-01-21 08:53:54.679909: step: 1080/533, loss: 0.6977648735046387 2023-01-21 08:53:55.869495: step: 1084/533, loss: 7.22731351852417 2023-01-21 08:53:57.040340: step: 1088/533, loss: 0.2014133632183075 2023-01-21 08:53:58.245603: step: 1092/533, loss: 0.28654471039772034 2023-01-21 08:53:59.430665: step: 1096/533, loss: 0.12103691697120667 2023-01-21 08:54:00.595282: step: 1100/533, loss: 0.9488382339477539 2023-01-21 08:54:01.827948: step: 1104/533, loss: 0.07610297203063965 2023-01-21 08:54:03.018526: step: 1108/533, loss: 0.16119880974292755 2023-01-21 08:54:04.223943: step: 1112/533, loss: 0.13445310294628143 2023-01-21 08:54:05.399883: step: 1116/533, loss: 0.11755600571632385 2023-01-21 08:54:06.564459: step: 1120/533, loss: 0.29532432556152344 2023-01-21 08:54:07.763725: step: 1124/533, loss: 0.8633362054824829 2023-01-21 08:54:08.941573: step: 1128/533, loss: 6.540116310119629 2023-01-21 08:54:10.099466: step: 1132/533, loss: 0.2515488266944885 2023-01-21 08:54:11.233050: step: 1136/533, loss: 0.8181860446929932 2023-01-21 08:54:12.421154: step: 1140/533, loss: 0.2142030894756317 2023-01-21 08:54:13.644695: step: 1144/533, loss: 0.3745500445365906 2023-01-21 08:54:14.833856: step: 1148/533, loss: 0.8272171020507812 2023-01-21 08:54:15.992858: step: 1152/533, loss: 0.06278391182422638 2023-01-21 08:54:17.173611: step: 1156/533, loss: 0.1989230215549469 2023-01-21 08:54:18.386573: step: 1160/533, loss: 0.27896958589553833 2023-01-21 08:54:19.572690: step: 1164/533, loss: 0.20488262176513672 2023-01-21 08:54:20.779996: step: 1168/533, loss: 0.38272780179977417 2023-01-21 08:54:21.950110: step: 1172/533, loss: 0.06994666904211044 2023-01-21 08:54:23.097775: step: 1176/533, loss: 0.18634013831615448 2023-01-21 08:54:24.243366: step: 1180/533, loss: 0.743073582649231 2023-01-21 08:54:25.423352: step: 1184/533, loss: 0.0941368043422699 2023-01-21 08:54:26.613294: step: 1188/533, loss: 0.8243561387062073 2023-01-21 08:54:27.836423: step: 1192/533, loss: 0.19166569411754608 2023-01-21 08:54:29.081626: step: 1196/533, loss: 6.266689777374268 2023-01-21 08:54:30.271677: step: 1200/533, loss: 0.06868753582239151 2023-01-21 08:54:31.487736: step: 1204/533, loss: 0.19443902373313904 2023-01-21 08:54:32.643080: step: 1208/533, loss: 0.13472004234790802 2023-01-21 08:54:33.860484: step: 1212/533, loss: 0.9054728746414185 2023-01-21 08:54:35.063442: step: 1216/533, loss: 0.12195806950330734 2023-01-21 08:54:36.246419: step: 1220/533, loss: 0.17970800399780273 2023-01-21 08:54:37.418057: step: 1224/533, loss: 0.12070713937282562 2023-01-21 08:54:38.646894: step: 1228/533, loss: 1.4752812385559082 2023-01-21 08:54:39.843186: step: 1232/533, loss: 0.7454493641853333 2023-01-21 08:54:41.010535: step: 1236/533, loss: 0.06914544105529785 2023-01-21 08:54:42.200554: step: 1240/533, loss: 1.2100070714950562 2023-01-21 08:54:43.377023: step: 1244/533, loss: 0.10302410274744034 2023-01-21 08:54:44.550610: step: 1248/533, loss: 0.21132364869117737 2023-01-21 08:54:45.717322: step: 1252/533, loss: 0.2659677565097809 2023-01-21 08:54:46.898326: step: 1256/533, loss: 0.19050368666648865 2023-01-21 08:54:48.082899: step: 1260/533, loss: 0.2024466097354889 2023-01-21 08:54:49.256624: step: 1264/533, loss: 0.5930944681167603 2023-01-21 08:54:50.462610: step: 1268/533, loss: 0.44186297059059143 2023-01-21 08:54:51.695773: step: 1272/533, loss: 0.3731943368911743 2023-01-21 08:54:52.930942: step: 1276/533, loss: 0.1335306167602539 2023-01-21 08:54:54.103232: step: 1280/533, loss: 1.0604174137115479 2023-01-21 08:54:55.294200: step: 1284/533, loss: 0.13002681732177734 2023-01-21 08:54:56.476681: step: 1288/533, loss: 0.9454756379127502 2023-01-21 08:54:57.676907: step: 1292/533, loss: 0.18780165910720825 2023-01-21 08:54:58.847947: step: 1296/533, loss: 0.20250433683395386 2023-01-21 08:55:00.049941: step: 1300/533, loss: 0.6443038582801819 2023-01-21 08:55:01.218490: step: 1304/533, loss: 0.8326841592788696 2023-01-21 08:55:02.394939: step: 1308/533, loss: 0.2797355055809021 2023-01-21 08:55:03.555173: step: 1312/533, loss: 0.09797344356775284 2023-01-21 08:55:04.756250: step: 1316/533, loss: 0.7025970220565796 2023-01-21 08:55:05.934892: step: 1320/533, loss: 0.08682501316070557 2023-01-21 08:55:07.141454: step: 1324/533, loss: 0.7677488923072815 2023-01-21 08:55:08.305909: step: 1328/533, loss: 1.3780176639556885 2023-01-21 08:55:09.476975: step: 1332/533, loss: 1.6198054552078247 2023-01-21 08:55:10.696826: step: 1336/533, loss: 0.3698953688144684 2023-01-21 08:55:11.867750: step: 1340/533, loss: 0.21812386810779572 2023-01-21 08:55:13.031296: step: 1344/533, loss: 1.2627613544464111 2023-01-21 08:55:14.208349: step: 1348/533, loss: 0.28452783823013306 2023-01-21 08:55:15.383918: step: 1352/533, loss: 0.879878580570221 2023-01-21 08:55:16.590981: step: 1356/533, loss: 0.3955039978027344 2023-01-21 08:55:17.764158: step: 1360/533, loss: 0.06800947338342667 2023-01-21 08:55:18.933994: step: 1364/533, loss: 0.04477305710315704 2023-01-21 08:55:20.098458: step: 1368/533, loss: 0.1684991866350174 2023-01-21 08:55:21.285897: step: 1372/533, loss: 0.35392212867736816 2023-01-21 08:55:22.460067: step: 1376/533, loss: 0.13616938889026642 2023-01-21 08:55:23.640863: step: 1380/533, loss: 7.286977767944336 2023-01-21 08:55:24.818905: step: 1384/533, loss: 0.15678483247756958 2023-01-21 08:55:26.040392: step: 1388/533, loss: 0.8577432036399841 2023-01-21 08:55:27.200106: step: 1392/533, loss: 0.11835689097642899 2023-01-21 08:55:28.398246: step: 1396/533, loss: 0.10267496109008789 2023-01-21 08:55:29.585214: step: 1400/533, loss: 0.6827402114868164 2023-01-21 08:55:30.790811: step: 1404/533, loss: 0.3727491497993469 2023-01-21 08:55:31.904506: step: 1408/533, loss: 0.19902926683425903 2023-01-21 08:55:33.056583: step: 1412/533, loss: 0.05669289082288742 2023-01-21 08:55:34.237754: step: 1416/533, loss: 0.4112869203090668 2023-01-21 08:55:35.433182: step: 1420/533, loss: 0.11731824278831482 2023-01-21 08:55:36.605544: step: 1424/533, loss: 0.2478063702583313 2023-01-21 08:55:37.785877: step: 1428/533, loss: 0.6963310241699219 2023-01-21 08:55:38.963046: step: 1432/533, loss: 1.1325640678405762 2023-01-21 08:55:40.198898: step: 1436/533, loss: 0.2118019163608551 2023-01-21 08:55:41.354299: step: 1440/533, loss: 0.31069833040237427 2023-01-21 08:55:42.548346: step: 1444/533, loss: 0.301180362701416 2023-01-21 08:55:43.719656: step: 1448/533, loss: 0.11280126869678497 2023-01-21 08:55:44.879484: step: 1452/533, loss: 6.658930778503418 2023-01-21 08:55:46.083023: step: 1456/533, loss: 0.20187149941921234 2023-01-21 08:55:47.314594: step: 1460/533, loss: 0.46133309602737427 2023-01-21 08:55:48.502473: step: 1464/533, loss: 0.1584792584180832 2023-01-21 08:55:49.660527: step: 1468/533, loss: 0.5543267130851746 2023-01-21 08:55:50.827227: step: 1472/533, loss: 0.8263395428657532 2023-01-21 08:55:52.040977: step: 1476/533, loss: 0.48075762391090393 2023-01-21 08:55:53.207482: step: 1480/533, loss: 0.09838028252124786 2023-01-21 08:55:54.404736: step: 1484/533, loss: 0.15344391763210297 2023-01-21 08:55:55.617812: step: 1488/533, loss: 0.737304151058197 2023-01-21 08:55:56.829120: step: 1492/533, loss: 0.8947690725326538 2023-01-21 08:55:58.040042: step: 1496/533, loss: 0.511588454246521 2023-01-21 08:55:59.201361: step: 1500/533, loss: 0.1525021642446518 2023-01-21 08:56:00.399770: step: 1504/533, loss: 0.11327085644006729 2023-01-21 08:56:01.581475: step: 1508/533, loss: 1.546418309211731 2023-01-21 08:56:02.748962: step: 1512/533, loss: 0.2898986041545868 2023-01-21 08:56:03.902542: step: 1516/533, loss: 0.7356467247009277 2023-01-21 08:56:05.058291: step: 1520/533, loss: 7.215470790863037 2023-01-21 08:56:06.247269: step: 1524/533, loss: 1.7791252136230469 2023-01-21 08:56:07.424012: step: 1528/533, loss: 0.27869662642478943 2023-01-21 08:56:08.585025: step: 1532/533, loss: 0.9292815327644348 2023-01-21 08:56:09.801498: step: 1536/533, loss: 0.2675338685512543 2023-01-21 08:56:10.982181: step: 1540/533, loss: 0.4899996817111969 2023-01-21 08:56:12.130703: step: 1544/533, loss: 0.10808859020471573 2023-01-21 08:56:13.305517: step: 1548/533, loss: 0.3228895664215088 2023-01-21 08:56:14.481081: step: 1552/533, loss: 0.06798505783081055 2023-01-21 08:56:15.694072: step: 1556/533, loss: 0.3320944905281067 2023-01-21 08:56:16.870336: step: 1560/533, loss: 1.806983232498169 2023-01-21 08:56:18.072922: step: 1564/533, loss: 1.048262119293213 2023-01-21 08:56:19.216844: step: 1568/533, loss: 0.13699588179588318 2023-01-21 08:56:20.360705: step: 1572/533, loss: 0.2673904001712799 2023-01-21 08:56:21.544751: step: 1576/533, loss: 0.8985964059829712 2023-01-21 08:56:22.754444: step: 1580/533, loss: 1.1982507705688477 2023-01-21 08:56:23.933605: step: 1584/533, loss: 0.22475290298461914 2023-01-21 08:56:25.169989: step: 1588/533, loss: 1.7302381992340088 2023-01-21 08:56:26.385633: step: 1592/533, loss: 0.8524916768074036 2023-01-21 08:56:27.596254: step: 1596/533, loss: 0.2432946264743805 2023-01-21 08:56:28.729457: step: 1600/533, loss: 13.707395553588867 2023-01-21 08:56:29.936791: step: 1604/533, loss: 0.15675364434719086 2023-01-21 08:56:31.114344: step: 1608/533, loss: 0.0466889850795269 2023-01-21 08:56:32.289180: step: 1612/533, loss: 0.4817764163017273 2023-01-21 08:56:33.505486: step: 1616/533, loss: 0.2190229445695877 2023-01-21 08:56:34.672749: step: 1620/533, loss: 0.4787437319755554 2023-01-21 08:56:35.839374: step: 1624/533, loss: 0.2182387113571167 2023-01-21 08:56:37.055706: step: 1628/533, loss: 1.2036696672439575 2023-01-21 08:56:38.239557: step: 1632/533, loss: 1.0908328294754028 2023-01-21 08:56:39.405304: step: 1636/533, loss: 0.12211751937866211 2023-01-21 08:56:40.594537: step: 1640/533, loss: 0.4149482548236847 2023-01-21 08:56:41.777491: step: 1644/533, loss: 0.34169453382492065 2023-01-21 08:56:42.989705: step: 1648/533, loss: 0.9192399978637695 2023-01-21 08:56:44.197361: step: 1652/533, loss: 0.0711483508348465 2023-01-21 08:56:45.370396: step: 1656/533, loss: 0.12819892168045044 2023-01-21 08:56:46.564991: step: 1660/533, loss: 0.052781008183956146 2023-01-21 08:56:47.724508: step: 1664/533, loss: 0.5178384780883789 2023-01-21 08:56:48.894680: step: 1668/533, loss: 0.20390301942825317 2023-01-21 08:56:50.047068: step: 1672/533, loss: 0.1226007267832756 2023-01-21 08:56:51.220483: step: 1676/533, loss: 0.16699305176734924 2023-01-21 08:56:52.413498: step: 1680/533, loss: 0.36003148555755615 2023-01-21 08:56:53.585603: step: 1684/533, loss: 0.14527592062950134 2023-01-21 08:56:54.764819: step: 1688/533, loss: 0.09590520709753036 2023-01-21 08:56:55.981076: step: 1692/533, loss: 0.11949887871742249 2023-01-21 08:56:57.169620: step: 1696/533, loss: 0.21050110459327698 2023-01-21 08:56:58.389139: step: 1700/533, loss: 0.17268845438957214 2023-01-21 08:56:59.555185: step: 1704/533, loss: 0.24717727303504944 2023-01-21 08:57:00.717693: step: 1708/533, loss: 0.38135844469070435 2023-01-21 08:57:01.898866: step: 1712/533, loss: 0.06689977645874023 2023-01-21 08:57:03.103860: step: 1716/533, loss: 0.12919697165489197 2023-01-21 08:57:04.276385: step: 1720/533, loss: 0.17385978996753693 2023-01-21 08:57:05.472538: step: 1724/533, loss: 0.4880429208278656 2023-01-21 08:57:06.703550: step: 1728/533, loss: 0.14656944572925568 2023-01-21 08:57:07.890955: step: 1732/533, loss: 0.16757328808307648 2023-01-21 08:57:09.128903: step: 1736/533, loss: 0.58705735206604 2023-01-21 08:57:10.283439: step: 1740/533, loss: 0.46039533615112305 2023-01-21 08:57:11.430690: step: 1744/533, loss: 0.08018722385168076 2023-01-21 08:57:12.625515: step: 1748/533, loss: 0.6703826785087585 2023-01-21 08:57:13.771073: step: 1752/533, loss: 0.08001342415809631 2023-01-21 08:57:14.935488: step: 1756/533, loss: 0.5173451900482178 2023-01-21 08:57:16.171063: step: 1760/533, loss: 1.1369085311889648 2023-01-21 08:57:17.376018: step: 1764/533, loss: 0.47411566972732544 2023-01-21 08:57:18.567225: step: 1768/533, loss: 0.17189660668373108 2023-01-21 08:57:19.712773: step: 1772/533, loss: 0.16097822785377502 2023-01-21 08:57:20.892100: step: 1776/533, loss: 1.8662508726119995 2023-01-21 08:57:22.069753: step: 1780/533, loss: 0.49722155928611755 2023-01-21 08:57:23.270715: step: 1784/533, loss: 0.21301144361495972 2023-01-21 08:57:24.511403: step: 1788/533, loss: 0.15982326865196228 2023-01-21 08:57:25.766415: step: 1792/533, loss: 0.17234042286872864 2023-01-21 08:57:26.909730: step: 1796/533, loss: 0.14246368408203125 2023-01-21 08:57:28.159140: step: 1800/533, loss: 0.2712755799293518 2023-01-21 08:57:29.355271: step: 1804/533, loss: 0.10894688963890076 2023-01-21 08:57:30.528687: step: 1808/533, loss: 0.8479633927345276 2023-01-21 08:57:31.699016: step: 1812/533, loss: 1.3127999305725098 2023-01-21 08:57:32.914437: step: 1816/533, loss: 0.17744627594947815 2023-01-21 08:57:34.088131: step: 1820/533, loss: 0.07554864883422852 2023-01-21 08:57:35.256862: step: 1824/533, loss: 0.34288546442985535 2023-01-21 08:57:36.449517: step: 1828/533, loss: 0.08664403110742569 2023-01-21 08:57:37.699120: step: 1832/533, loss: 0.13270169496536255 2023-01-21 08:57:38.850797: step: 1836/533, loss: 0.3446725010871887 2023-01-21 08:57:40.013772: step: 1840/533, loss: 1.8795182704925537 2023-01-21 08:57:41.214516: step: 1844/533, loss: 0.18244333565235138 2023-01-21 08:57:42.400710: step: 1848/533, loss: 0.10259667038917542 2023-01-21 08:57:43.615202: step: 1852/533, loss: 0.03538132086396217 2023-01-21 08:57:44.797656: step: 1856/533, loss: 0.45798835158348083 2023-01-21 08:57:45.994596: step: 1860/533, loss: 0.27287137508392334 2023-01-21 08:57:47.286937: step: 1864/533, loss: 0.3578135371208191 2023-01-21 08:57:48.471093: step: 1868/533, loss: 0.3007594347000122 2023-01-21 08:57:49.660193: step: 1872/533, loss: 1.4543057680130005 2023-01-21 08:57:50.875035: step: 1876/533, loss: 1.3446812629699707 2023-01-21 08:57:52.074228: step: 1880/533, loss: 0.25744399428367615 2023-01-21 08:57:53.261206: step: 1884/533, loss: 0.5162826776504517 2023-01-21 08:57:54.451585: step: 1888/533, loss: 0.42663878202438354 2023-01-21 08:57:55.647020: step: 1892/533, loss: 0.16847020387649536 2023-01-21 08:57:56.852513: step: 1896/533, loss: 0.6822004318237305 2023-01-21 08:57:57.990903: step: 1900/533, loss: 0.10992636531591415 2023-01-21 08:57:59.198004: step: 1904/533, loss: 0.14849665760993958 2023-01-21 08:58:00.368233: step: 1908/533, loss: 1.0728156566619873 2023-01-21 08:58:01.520652: step: 1912/533, loss: 0.14673951268196106 2023-01-21 08:58:02.729965: step: 1916/533, loss: 0.6810401082038879 2023-01-21 08:58:03.924251: step: 1920/533, loss: 0.07864876091480255 2023-01-21 08:58:05.104951: step: 1924/533, loss: 0.2458069920539856 2023-01-21 08:58:06.243372: step: 1928/533, loss: 0.1809266209602356 2023-01-21 08:58:07.482379: step: 1932/533, loss: 0.5229305028915405 2023-01-21 08:58:08.661084: step: 1936/533, loss: 0.21158117055892944 2023-01-21 08:58:09.815710: step: 1940/533, loss: 0.0708746463060379 2023-01-21 08:58:11.021393: step: 1944/533, loss: 0.3087455630302429 2023-01-21 08:58:12.189407: step: 1948/533, loss: 0.42598843574523926 2023-01-21 08:58:13.359824: step: 1952/533, loss: 0.907239556312561 2023-01-21 08:58:14.575753: step: 1956/533, loss: 0.10535035282373428 2023-01-21 08:58:15.734373: step: 1960/533, loss: 0.24729867279529572 2023-01-21 08:58:16.946109: step: 1964/533, loss: 0.9914241433143616 2023-01-21 08:58:18.127014: step: 1968/533, loss: 0.10492689907550812 2023-01-21 08:58:19.305523: step: 1972/533, loss: 0.8063637018203735 2023-01-21 08:58:20.507984: step: 1976/533, loss: 0.2916424870491028 2023-01-21 08:58:21.737800: step: 1980/533, loss: 0.237802654504776 2023-01-21 08:58:22.939473: step: 1984/533, loss: 0.08947930485010147 2023-01-21 08:58:24.151858: step: 1988/533, loss: 0.18570509552955627 2023-01-21 08:58:25.336235: step: 1992/533, loss: 0.5579148530960083 2023-01-21 08:58:26.484673: step: 1996/533, loss: 0.19813914597034454 2023-01-21 08:58:27.711180: step: 2000/533, loss: 1.77040696144104 2023-01-21 08:58:28.880541: step: 2004/533, loss: 0.2940627336502075 2023-01-21 08:58:30.127070: step: 2008/533, loss: 0.1400291472673416 2023-01-21 08:58:31.304689: step: 2012/533, loss: 0.2583116590976715 2023-01-21 08:58:32.449429: step: 2016/533, loss: 0.19395039975643158 2023-01-21 08:58:33.678596: step: 2020/533, loss: 1.147090196609497 2023-01-21 08:58:34.861153: step: 2024/533, loss: 0.7672001123428345 2023-01-21 08:58:36.046507: step: 2028/533, loss: 0.13498002290725708 2023-01-21 08:58:37.227401: step: 2032/533, loss: 0.22591599822044373 2023-01-21 08:58:38.449697: step: 2036/533, loss: 1.6391170024871826 2023-01-21 08:58:39.654733: step: 2040/533, loss: 0.1355711966753006 2023-01-21 08:58:40.853220: step: 2044/533, loss: 0.21271300315856934 2023-01-21 08:58:42.022454: step: 2048/533, loss: 0.11225919425487518 2023-01-21 08:58:43.201722: step: 2052/533, loss: 0.1445845067501068 2023-01-21 08:58:44.347595: step: 2056/533, loss: 0.3594485819339752 2023-01-21 08:58:45.536746: step: 2060/533, loss: 0.4189876616001129 2023-01-21 08:58:46.773902: step: 2064/533, loss: 0.13552206754684448 2023-01-21 08:58:47.923509: step: 2068/533, loss: 0.2117624282836914 2023-01-21 08:58:49.148504: step: 2072/533, loss: 0.6755533218383789 2023-01-21 08:58:50.326657: step: 2076/533, loss: 0.19395361840724945 2023-01-21 08:58:51.519615: step: 2080/533, loss: 0.7784324884414673 2023-01-21 08:58:52.732239: step: 2084/533, loss: 1.1049057245254517 2023-01-21 08:58:53.943652: step: 2088/533, loss: 0.09831076115369797 2023-01-21 08:58:55.114356: step: 2092/533, loss: 0.19836750626564026 2023-01-21 08:58:56.301798: step: 2096/533, loss: 0.18707247078418732 2023-01-21 08:58:57.527478: step: 2100/533, loss: 0.08321471512317657 2023-01-21 08:58:58.769081: step: 2104/533, loss: 0.358722448348999 2023-01-21 08:58:59.965694: step: 2108/533, loss: 0.911118745803833 2023-01-21 08:59:01.144715: step: 2112/533, loss: 0.7241929173469543 2023-01-21 08:59:02.320615: step: 2116/533, loss: 0.7691901326179504 2023-01-21 08:59:03.506168: step: 2120/533, loss: 0.2270602285861969 2023-01-21 08:59:04.689614: step: 2124/533, loss: 0.2018546164035797 2023-01-21 08:59:05.894061: step: 2128/533, loss: 0.8860490918159485 2023-01-21 08:59:07.103234: step: 2132/533, loss: 0.5212861895561218 ================================================== Loss: 0.626 -------------------- Dev: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Test: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Korean: {'event': {'p': 0.8, 'r': 0.31746031746031744, 'f1': 0.45454545454545453}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Russian: {'event': {'p': 0.45454545454545453, 'r': 0.2777777777777778, 'f1': 0.3448275862068966}, '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': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.8, 'r': 0.31746031746031744, 'f1': 0.45454545454545453}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.49306197964847365, 'r': 0.7097203728362184, 'f1': 0.5818777292576419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5165719696969697, 'r': 0.6975703324808185, 'f1': 0.5935799782372144}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample 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': 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 09:00:04.788731: step: 4/533, loss: 0.03451957553625107 2023-01-21 09:00:05.960188: step: 8/533, loss: 0.19599905610084534 2023-01-21 09:00:07.093871: step: 12/533, loss: 0.26731598377227783 2023-01-21 09:00:08.283623: step: 16/533, loss: 0.18239393830299377 2023-01-21 09:00:09.441902: step: 20/533, loss: 0.6576451659202576 2023-01-21 09:00:10.600474: step: 24/533, loss: 0.2004776895046234 2023-01-21 09:00:11.811383: step: 28/533, loss: 0.18090343475341797 2023-01-21 09:00:12.964356: step: 32/533, loss: 0.49251270294189453 2023-01-21 09:00:14.133411: step: 36/533, loss: 0.13193264603614807 2023-01-21 09:00:15.319085: step: 40/533, loss: 0.43752098083496094 2023-01-21 09:00:16.489968: step: 44/533, loss: 0.559822678565979 2023-01-21 09:00:17.663937: step: 48/533, loss: 0.15219825506210327 2023-01-21 09:00:18.851801: step: 52/533, loss: 0.9357835650444031 2023-01-21 09:00:20.035998: step: 56/533, loss: 0.6957262754440308 2023-01-21 09:00:21.242019: step: 60/533, loss: 6.301454067230225 2023-01-21 09:00:22.426735: step: 64/533, loss: 1.0453317165374756 2023-01-21 09:00:23.630629: step: 68/533, loss: 0.1851538121700287 2023-01-21 09:00:24.847627: step: 72/533, loss: 0.11183881759643555 2023-01-21 09:00:26.052942: step: 76/533, loss: 0.057894326746463776 2023-01-21 09:00:27.227380: step: 80/533, loss: 0.20177365839481354 2023-01-21 09:00:28.395502: step: 84/533, loss: 0.7120344638824463 2023-01-21 09:00:29.587673: step: 88/533, loss: 0.07235679775476456 2023-01-21 09:00:30.797276: step: 92/533, loss: 0.3799402415752411 2023-01-21 09:00:32.007720: step: 96/533, loss: 0.028688574209809303 2023-01-21 09:00:33.243344: step: 100/533, loss: 0.2685028910636902 2023-01-21 09:00:34.394611: step: 104/533, loss: 0.07654419541358948 2023-01-21 09:00:35.644625: step: 108/533, loss: 0.07747369259595871 2023-01-21 09:00:36.799047: step: 112/533, loss: 0.13297343254089355 2023-01-21 09:00:38.009136: step: 116/533, loss: 0.7147554755210876 2023-01-21 09:00:39.170777: step: 120/533, loss: 0.2567634582519531 2023-01-21 09:00:40.321192: step: 124/533, loss: 0.13873139023780823 2023-01-21 09:00:41.515522: step: 128/533, loss: 0.9747810363769531 2023-01-21 09:00:42.688280: step: 132/533, loss: 0.37831661105155945 2023-01-21 09:00:43.831704: step: 136/533, loss: 0.6275434494018555 2023-01-21 09:00:45.039180: step: 140/533, loss: 0.19439983367919922 2023-01-21 09:00:46.232097: step: 144/533, loss: 0.08895187824964523 2023-01-21 09:00:47.377955: step: 148/533, loss: 0.28726911544799805 2023-01-21 09:00:48.538309: step: 152/533, loss: 0.10586166381835938 2023-01-21 09:00:49.728366: step: 156/533, loss: 0.5252687335014343 2023-01-21 09:00:50.958901: step: 160/533, loss: 0.7363172769546509 2023-01-21 09:00:52.132511: step: 164/533, loss: 1.0096620321273804 2023-01-21 09:00:53.290107: step: 168/533, loss: 0.09869274497032166 2023-01-21 09:00:54.459767: step: 172/533, loss: 0.2539038360118866 2023-01-21 09:00:55.619263: step: 176/533, loss: 0.25012826919555664 2023-01-21 09:00:56.793958: step: 180/533, loss: 0.20629939436912537 2023-01-21 09:00:57.998116: step: 184/533, loss: 0.24908486008644104 2023-01-21 09:00:59.183149: step: 188/533, loss: 0.08796806633472443 2023-01-21 09:01:00.379136: step: 192/533, loss: 0.028368141502141953 2023-01-21 09:01:01.573177: step: 196/533, loss: 0.12687864899635315 2023-01-21 09:01:02.766952: step: 200/533, loss: 0.29567041993141174 2023-01-21 09:01:03.934321: step: 204/533, loss: 0.21308287978172302 2023-01-21 09:01:05.169881: step: 208/533, loss: 0.3525891900062561 2023-01-21 09:01:06.307834: step: 212/533, loss: 0.3552277684211731 2023-01-21 09:01:07.474015: step: 216/533, loss: 0.21406956017017365 2023-01-21 09:01:08.636616: step: 220/533, loss: 0.31315916776657104 2023-01-21 09:01:09.825103: step: 224/533, loss: 0.1957150250673294 2023-01-21 09:01:11.012605: step: 228/533, loss: 0.1896463930606842 2023-01-21 09:01:12.161954: step: 232/533, loss: 0.22514191269874573 2023-01-21 09:01:13.349423: step: 236/533, loss: 0.7129614949226379 2023-01-21 09:01:14.481342: step: 240/533, loss: 0.011224795132875443 2023-01-21 09:01:15.711858: step: 244/533, loss: 0.17230349779129028 2023-01-21 09:01:16.913052: step: 248/533, loss: 0.15486077964305878 2023-01-21 09:01:18.080936: step: 252/533, loss: 0.163020521402359 2023-01-21 09:01:19.287377: step: 256/533, loss: 0.11416491866111755 2023-01-21 09:01:20.472508: step: 260/533, loss: 0.21514463424682617 2023-01-21 09:01:21.715393: step: 264/533, loss: 0.6189179420471191 2023-01-21 09:01:22.887852: step: 268/533, loss: 0.0948396772146225 2023-01-21 09:01:24.067667: step: 272/533, loss: 0.49893808364868164 2023-01-21 09:01:25.216135: step: 276/533, loss: 0.34505051374435425 2023-01-21 09:01:26.387510: step: 280/533, loss: 1.1180148124694824 2023-01-21 09:01:27.573248: step: 284/533, loss: 0.526872456073761 2023-01-21 09:01:28.755797: step: 288/533, loss: 0.16351699829101562 2023-01-21 09:01:29.945038: step: 292/533, loss: 1.0271607637405396 2023-01-21 09:01:31.161345: step: 296/533, loss: 0.9073029160499573 2023-01-21 09:01:32.328546: step: 300/533, loss: 0.40234214067459106 2023-01-21 09:01:33.512391: step: 304/533, loss: 0.14024820923805237 2023-01-21 09:01:34.668387: step: 308/533, loss: 0.15397652983665466 2023-01-21 09:01:35.810455: step: 312/533, loss: 0.1487516462802887 2023-01-21 09:01:36.976718: step: 316/533, loss: 0.5063775777816772 2023-01-21 09:01:38.164009: step: 320/533, loss: 0.228526309132576 2023-01-21 09:01:39.443496: step: 324/533, loss: 0.9932713508605957 2023-01-21 09:01:40.674093: step: 328/533, loss: 0.29136142134666443 2023-01-21 09:01:41.868419: step: 332/533, loss: 0.46229496598243713 2023-01-21 09:01:43.047262: step: 336/533, loss: 0.14601492881774902 2023-01-21 09:01:44.226141: step: 340/533, loss: 0.0966564193367958 2023-01-21 09:01:45.435401: step: 344/533, loss: 0.31992846727371216 2023-01-21 09:01:46.617908: step: 348/533, loss: 0.09903021156787872 2023-01-21 09:01:47.799516: step: 352/533, loss: 0.047904398292303085 2023-01-21 09:01:48.949476: step: 356/533, loss: 0.0857873409986496 2023-01-21 09:01:50.103832: step: 360/533, loss: 0.7139033079147339 2023-01-21 09:01:51.327312: step: 364/533, loss: 0.1520092934370041 2023-01-21 09:01:52.508443: step: 368/533, loss: 1.0046950578689575 2023-01-21 09:01:53.656471: step: 372/533, loss: 0.11109619587659836 2023-01-21 09:01:54.831507: step: 376/533, loss: 0.7636460661888123 2023-01-21 09:01:56.034234: step: 380/533, loss: 0.22293290495872498 2023-01-21 09:01:57.285022: step: 384/533, loss: 0.3910292088985443 2023-01-21 09:01:58.494900: step: 388/533, loss: 0.8029780983924866 2023-01-21 09:01:59.690160: step: 392/533, loss: 0.08425979316234589 2023-01-21 09:02:00.869068: step: 396/533, loss: 0.4124286472797394 2023-01-21 09:02:02.046536: step: 400/533, loss: 0.807341992855072 2023-01-21 09:02:03.243630: step: 404/533, loss: 0.19572040438652039 2023-01-21 09:02:04.401987: step: 408/533, loss: 0.6098631024360657 2023-01-21 09:02:05.599907: step: 412/533, loss: 0.09129834175109863 2023-01-21 09:02:06.741817: step: 416/533, loss: 0.05229043960571289 2023-01-21 09:02:07.903482: step: 420/533, loss: 0.06809864193201065 2023-01-21 09:02:09.087908: step: 424/533, loss: 0.43989038467407227 2023-01-21 09:02:10.257863: step: 428/533, loss: 0.18254362046718597 2023-01-21 09:02:11.444641: step: 432/533, loss: 0.047667600214481354 2023-01-21 09:02:12.662151: step: 436/533, loss: 0.10156774520874023 2023-01-21 09:02:13.808212: step: 440/533, loss: 0.04419431835412979 2023-01-21 09:02:15.005509: step: 444/533, loss: 0.1719702184200287 2023-01-21 09:02:16.158839: step: 448/533, loss: 0.10167141258716583 2023-01-21 09:02:17.363940: step: 452/533, loss: 0.4255094528198242 2023-01-21 09:02:18.542905: step: 456/533, loss: 1.0728917121887207 2023-01-21 09:02:19.725382: step: 460/533, loss: 0.656667947769165 2023-01-21 09:02:20.896450: step: 464/533, loss: 0.31740856170654297 2023-01-21 09:02:22.075516: step: 468/533, loss: 0.8777336478233337 2023-01-21 09:02:23.287447: step: 472/533, loss: 0.21998901665210724 2023-01-21 09:02:24.478405: step: 476/533, loss: 0.21157197654247284 2023-01-21 09:02:25.686738: step: 480/533, loss: 0.08625831454992294 2023-01-21 09:02:26.841940: step: 484/533, loss: 0.05295391008257866 2023-01-21 09:02:28.006883: step: 488/533, loss: 1.3534992933273315 2023-01-21 09:02:29.124758: step: 492/533, loss: 0.05635356903076172 2023-01-21 09:02:30.281760: step: 496/533, loss: 0.1512811779975891 2023-01-21 09:02:31.474506: step: 500/533, loss: 1.4139525890350342 2023-01-21 09:02:32.657466: step: 504/533, loss: 0.48238927125930786 2023-01-21 09:02:33.822328: step: 508/533, loss: 1.2085622549057007 2023-01-21 09:02:34.989778: step: 512/533, loss: 0.85533607006073 2023-01-21 09:02:36.197209: step: 516/533, loss: 0.37146711349487305 2023-01-21 09:02:37.394882: step: 520/533, loss: 0.11755583435297012 2023-01-21 09:02:38.561267: step: 524/533, loss: 0.6158665418624878 2023-01-21 09:02:39.741147: step: 528/533, loss: 0.04983863979578018 2023-01-21 09:02:40.915184: step: 532/533, loss: 0.5918863415718079 2023-01-21 09:02:42.075615: step: 536/533, loss: 0.3632376492023468 2023-01-21 09:02:43.235894: step: 540/533, loss: 0.47278451919555664 2023-01-21 09:02:44.433927: step: 544/533, loss: 0.7704907655715942 2023-01-21 09:02:45.619048: step: 548/533, loss: 0.5547401309013367 2023-01-21 09:02:46.793304: step: 552/533, loss: 0.06354232132434845 2023-01-21 09:02:47.977025: step: 556/533, loss: 0.11705227196216583 2023-01-21 09:02:49.220885: step: 560/533, loss: 1.9675414562225342 2023-01-21 09:02:50.403377: step: 564/533, loss: 0.6091313362121582 2023-01-21 09:02:51.581002: step: 568/533, loss: 0.06256427615880966 2023-01-21 09:02:52.729544: step: 572/533, loss: 1.065568447113037 2023-01-21 09:02:53.942205: step: 576/533, loss: 6.238678455352783 2023-01-21 09:02:55.193713: step: 580/533, loss: 0.20755037665367126 2023-01-21 09:02:56.438609: step: 584/533, loss: 1.1084903478622437 2023-01-21 09:02:57.642154: step: 588/533, loss: 0.4266526401042938 2023-01-21 09:02:58.825268: step: 592/533, loss: 1.9573957920074463 2023-01-21 09:03:00.053512: step: 596/533, loss: 0.14478835463523865 2023-01-21 09:03:01.245386: step: 600/533, loss: 0.11911492794752121 2023-01-21 09:03:02.409606: step: 604/533, loss: 0.7593408823013306 2023-01-21 09:03:03.592099: step: 608/533, loss: 0.7655779123306274 2023-01-21 09:03:04.769757: step: 612/533, loss: 0.23260065913200378 2023-01-21 09:03:05.967842: step: 616/533, loss: 0.07270164787769318 2023-01-21 09:03:07.179204: step: 620/533, loss: 0.09500313550233841 2023-01-21 09:03:08.392190: step: 624/533, loss: 0.2175094187259674 2023-01-21 09:03:09.542775: step: 628/533, loss: 0.1838330626487732 2023-01-21 09:03:10.766920: step: 632/533, loss: 0.04129800945520401 2023-01-21 09:03:11.972091: step: 636/533, loss: 0.17840024828910828 2023-01-21 09:03:13.161182: step: 640/533, loss: 1.0283324718475342 2023-01-21 09:03:14.364708: step: 644/533, loss: 6.115992069244385 2023-01-21 09:03:15.528787: step: 648/533, loss: 0.22435492277145386 2023-01-21 09:03:16.722391: step: 652/533, loss: 0.13172969222068787 2023-01-21 09:03:17.883345: step: 656/533, loss: 0.2503325045108795 2023-01-21 09:03:19.056263: step: 660/533, loss: 0.24290724098682404 2023-01-21 09:03:20.226281: step: 664/533, loss: 0.047755979001522064 2023-01-21 09:03:21.467579: step: 668/533, loss: 0.6896345615386963 2023-01-21 09:03:22.689080: step: 672/533, loss: 0.17214851081371307 2023-01-21 09:03:23.848907: step: 676/533, loss: 0.16003593802452087 2023-01-21 09:03:25.064914: step: 680/533, loss: 0.45051613450050354 2023-01-21 09:03:26.238162: step: 684/533, loss: 0.41030359268188477 2023-01-21 09:03:27.404897: step: 688/533, loss: 0.23761333525180817 2023-01-21 09:03:28.574736: step: 692/533, loss: 0.3367490768432617 2023-01-21 09:03:29.736445: step: 696/533, loss: 0.8366636633872986 2023-01-21 09:03:30.934559: step: 700/533, loss: 0.21287240087985992 2023-01-21 09:03:32.143837: step: 704/533, loss: 0.10609011352062225 2023-01-21 09:03:33.322860: step: 708/533, loss: 0.325504869222641 2023-01-21 09:03:34.504584: step: 712/533, loss: 5.976997375488281 2023-01-21 09:03:35.683511: step: 716/533, loss: 0.07249946892261505 2023-01-21 09:03:36.867802: step: 720/533, loss: 0.5006676316261292 2023-01-21 09:03:38.047833: step: 724/533, loss: 0.09480872750282288 2023-01-21 09:03:39.225893: step: 728/533, loss: 0.08126058429479599 2023-01-21 09:03:40.404796: step: 732/533, loss: 0.18725129961967468 2023-01-21 09:03:41.563883: step: 736/533, loss: 0.6739752292633057 2023-01-21 09:03:42.811791: step: 740/533, loss: 0.10861635208129883 2023-01-21 09:03:43.955461: step: 744/533, loss: 0.4173053503036499 2023-01-21 09:03:45.147368: step: 748/533, loss: 0.5523170232772827 2023-01-21 09:03:46.313832: step: 752/533, loss: 0.1912127584218979 2023-01-21 09:03:47.520918: step: 756/533, loss: 0.1156759262084961 2023-01-21 09:03:48.721367: step: 760/533, loss: 0.06699047237634659 2023-01-21 09:03:49.889443: step: 764/533, loss: 0.08671140670776367 2023-01-21 09:03:51.041476: step: 768/533, loss: 0.13751859962940216 2023-01-21 09:03:52.218121: step: 772/533, loss: 0.514961838722229 2023-01-21 09:03:53.426911: step: 776/533, loss: 0.39676913619041443 2023-01-21 09:03:54.572575: step: 780/533, loss: 0.576019823551178 2023-01-21 09:03:55.726251: step: 784/533, loss: 0.09464402496814728 2023-01-21 09:03:56.874143: step: 788/533, loss: 0.3749733865261078 2023-01-21 09:03:58.041747: step: 792/533, loss: 0.07085695117712021 2023-01-21 09:03:59.283690: step: 796/533, loss: 0.08774380385875702 2023-01-21 09:04:00.503525: step: 800/533, loss: 0.11243218928575516 2023-01-21 09:04:01.689412: step: 804/533, loss: 0.13983449339866638 2023-01-21 09:04:02.903940: step: 808/533, loss: 0.12128515541553497 2023-01-21 09:04:04.070598: step: 812/533, loss: 0.0763828307390213 2023-01-21 09:04:05.268673: step: 816/533, loss: 0.18989324569702148 2023-01-21 09:04:06.413358: step: 820/533, loss: 0.4272615313529968 2023-01-21 09:04:07.617772: step: 824/533, loss: 1.0054242610931396 2023-01-21 09:04:08.829067: step: 828/533, loss: 0.42077115178108215 2023-01-21 09:04:10.035191: step: 832/533, loss: 0.7691338062286377 2023-01-21 09:04:11.237707: step: 836/533, loss: 0.33931148052215576 2023-01-21 09:04:12.421104: step: 840/533, loss: 0.7439698576927185 2023-01-21 09:04:13.632375: step: 844/533, loss: 0.8667982816696167 2023-01-21 09:04:14.820175: step: 848/533, loss: 0.21416302025318146 2023-01-21 09:04:15.986079: step: 852/533, loss: 0.4676376283168793 2023-01-21 09:04:17.190569: step: 856/533, loss: 0.13931503891944885 2023-01-21 09:04:18.378941: step: 860/533, loss: 0.1378774642944336 2023-01-21 09:04:19.543157: step: 864/533, loss: 0.14342756569385529 2023-01-21 09:04:20.699175: step: 868/533, loss: 0.14753732085227966 2023-01-21 09:04:21.869073: step: 872/533, loss: 0.07717466354370117 2023-01-21 09:04:23.048545: step: 876/533, loss: 0.5566136837005615 2023-01-21 09:04:24.238093: step: 880/533, loss: 0.09720521420240402 2023-01-21 09:04:25.443843: step: 884/533, loss: 0.6886968612670898 2023-01-21 09:04:26.644063: step: 888/533, loss: 6.780792236328125 2023-01-21 09:04:27.835716: step: 892/533, loss: 0.8442405462265015 2023-01-21 09:04:29.044105: step: 896/533, loss: 0.15167587995529175 2023-01-21 09:04:30.222050: step: 900/533, loss: 0.27124518156051636 2023-01-21 09:04:31.380899: step: 904/533, loss: 0.1327369660139084 2023-01-21 09:04:32.522783: step: 908/533, loss: 0.13409757614135742 2023-01-21 09:04:33.724076: step: 912/533, loss: 0.21246260404586792 2023-01-21 09:04:34.959185: step: 916/533, loss: 0.13756155967712402 2023-01-21 09:04:36.138679: step: 920/533, loss: 0.12368479371070862 2023-01-21 09:04:37.333863: step: 924/533, loss: 0.21089425683021545 2023-01-21 09:04:38.536040: step: 928/533, loss: 0.178009033203125 2023-01-21 09:04:39.712097: step: 932/533, loss: 0.18765611946582794 2023-01-21 09:04:40.883265: step: 936/533, loss: 0.7292583584785461 2023-01-21 09:04:42.056505: step: 940/533, loss: 0.41813531517982483 2023-01-21 09:04:43.232228: step: 944/533, loss: 0.4608857035636902 2023-01-21 09:04:44.441493: step: 948/533, loss: 0.07561693340539932 2023-01-21 09:04:45.617562: step: 952/533, loss: 0.19039887189865112 2023-01-21 09:04:46.790477: step: 956/533, loss: 0.34387293457984924 2023-01-21 09:04:47.981442: step: 960/533, loss: 0.26746273040771484 2023-01-21 09:04:49.159774: step: 964/533, loss: 0.7302387952804565 2023-01-21 09:04:50.338719: step: 968/533, loss: 1.9227488040924072 2023-01-21 09:04:51.503951: step: 972/533, loss: 0.5365772247314453 2023-01-21 09:04:52.765713: step: 976/533, loss: 0.439666748046875 2023-01-21 09:04:53.950282: step: 980/533, loss: 0.11347942054271698 2023-01-21 09:04:55.155494: step: 984/533, loss: 0.14102010428905487 2023-01-21 09:04:56.330247: step: 988/533, loss: 0.10964584350585938 2023-01-21 09:04:57.554756: step: 992/533, loss: 0.2236180305480957 2023-01-21 09:04:58.723369: step: 996/533, loss: 0.7999707460403442 2023-01-21 09:04:59.932818: step: 1000/533, loss: 0.06100459024310112 2023-01-21 09:05:01.105911: step: 1004/533, loss: 0.06685248017311096 2023-01-21 09:05:02.318657: step: 1008/533, loss: 0.06022777408361435 2023-01-21 09:05:03.505163: step: 1012/533, loss: 0.16972537338733673 2023-01-21 09:05:04.731236: step: 1016/533, loss: 0.30189791321754456 2023-01-21 09:05:05.938899: step: 1020/533, loss: 0.5730999708175659 2023-01-21 09:05:07.098165: step: 1024/533, loss: 0.6522205471992493 2023-01-21 09:05:08.257774: step: 1028/533, loss: 0.3931344151496887 2023-01-21 09:05:09.446377: step: 1032/533, loss: 0.5140711069107056 2023-01-21 09:05:10.615204: step: 1036/533, loss: 0.22558140754699707 2023-01-21 09:05:11.818344: step: 1040/533, loss: 0.1763581931591034 2023-01-21 09:05:13.039509: step: 1044/533, loss: 1.507238507270813 2023-01-21 09:05:14.155661: step: 1048/533, loss: 0.1031225174665451 2023-01-21 09:05:15.320614: step: 1052/533, loss: 0.19427213072776794 2023-01-21 09:05:16.510562: step: 1056/533, loss: 0.26278573274612427 2023-01-21 09:05:17.718739: step: 1060/533, loss: 0.24707774817943573 2023-01-21 09:05:18.916501: step: 1064/533, loss: 0.3065307140350342 2023-01-21 09:05:20.157756: step: 1068/533, loss: 0.5000457763671875 2023-01-21 09:05:21.335803: step: 1072/533, loss: 1.0486862659454346 2023-01-21 09:05:22.519615: step: 1076/533, loss: 0.683367133140564 2023-01-21 09:05:23.697808: step: 1080/533, loss: 0.1699684113264084 2023-01-21 09:05:24.884677: step: 1084/533, loss: 0.21722298860549927 2023-01-21 09:05:26.088127: step: 1088/533, loss: 0.13793249428272247 2023-01-21 09:05:27.286178: step: 1092/533, loss: 1.2792967557907104 2023-01-21 09:05:28.445769: step: 1096/533, loss: 0.27423784136772156 2023-01-21 09:05:29.655213: step: 1100/533, loss: 0.15517958998680115 2023-01-21 09:05:30.886702: step: 1104/533, loss: 0.9255038499832153 2023-01-21 09:05:32.050846: step: 1108/533, loss: 0.264323353767395 2023-01-21 09:05:33.206329: step: 1112/533, loss: 0.7559067010879517 2023-01-21 09:05:34.466108: step: 1116/533, loss: 1.1000087261199951 2023-01-21 09:05:35.654188: step: 1120/533, loss: 0.22105997800827026 2023-01-21 09:05:36.837678: step: 1124/533, loss: 0.8161914348602295 2023-01-21 09:05:38.017251: step: 1128/533, loss: 0.6911132335662842 2023-01-21 09:05:39.210388: step: 1132/533, loss: 0.6746658682823181 2023-01-21 09:05:40.391931: step: 1136/533, loss: 0.2609933614730835 2023-01-21 09:05:41.573812: step: 1140/533, loss: 0.35272732377052307 2023-01-21 09:05:42.771365: step: 1144/533, loss: 0.2991991937160492 2023-01-21 09:05:43.966832: step: 1148/533, loss: 0.19277286529541016 2023-01-21 09:05:45.175929: step: 1152/533, loss: 0.8033561706542969 2023-01-21 09:05:46.342595: step: 1156/533, loss: 0.051830388605594635 2023-01-21 09:05:47.580086: step: 1160/533, loss: 6.071428298950195 2023-01-21 09:05:48.798232: step: 1164/533, loss: 0.4722738265991211 2023-01-21 09:05:49.988967: step: 1168/533, loss: 0.20875605940818787 2023-01-21 09:05:51.191864: step: 1172/533, loss: 0.2725340723991394 2023-01-21 09:05:52.398083: step: 1176/533, loss: 0.7279965877532959 2023-01-21 09:05:53.558021: step: 1180/533, loss: 0.6772966384887695 2023-01-21 09:05:54.707908: step: 1184/533, loss: 0.15030308067798615 2023-01-21 09:05:55.866964: step: 1188/533, loss: 0.09388285130262375 2023-01-21 09:05:57.081460: step: 1192/533, loss: 0.29090338945388794 2023-01-21 09:05:58.244594: step: 1196/533, loss: 0.14142310619354248 2023-01-21 09:05:59.465427: step: 1200/533, loss: 0.17211371660232544 2023-01-21 09:06:00.644193: step: 1204/533, loss: 0.1851998269557953 2023-01-21 09:06:01.780333: step: 1208/533, loss: 0.4577353596687317 2023-01-21 09:06:02.942474: step: 1212/533, loss: 1.147730827331543 2023-01-21 09:06:04.148423: step: 1216/533, loss: 6.743643283843994 2023-01-21 09:06:05.340967: step: 1220/533, loss: 0.1123572364449501 2023-01-21 09:06:06.503291: step: 1224/533, loss: 0.13448172807693481 2023-01-21 09:06:07.693565: step: 1228/533, loss: 0.8281000256538391 2023-01-21 09:06:08.857908: step: 1232/533, loss: 0.26371994614601135 2023-01-21 09:06:10.017301: step: 1236/533, loss: 0.5202805399894714 2023-01-21 09:06:11.186597: step: 1240/533, loss: 0.10507631301879883 2023-01-21 09:06:12.419455: step: 1244/533, loss: 0.7237327694892883 2023-01-21 09:06:13.599804: step: 1248/533, loss: 0.11498404294252396 2023-01-21 09:06:14.766546: step: 1252/533, loss: 0.05696067959070206 2023-01-21 09:06:15.952272: step: 1256/533, loss: 0.10937843471765518 2023-01-21 09:06:17.121340: step: 1260/533, loss: 0.11326217651367188 2023-01-21 09:06:18.300549: step: 1264/533, loss: 0.09295278042554855 2023-01-21 09:06:19.524166: step: 1268/533, loss: 0.3635866641998291 2023-01-21 09:06:20.691127: step: 1272/533, loss: 0.15917611122131348 2023-01-21 09:06:21.896521: step: 1276/533, loss: 0.8526588678359985 2023-01-21 09:06:23.088144: step: 1280/533, loss: 0.13858576118946075 2023-01-21 09:06:24.251556: step: 1284/533, loss: 0.0664517879486084 2023-01-21 09:06:25.438866: step: 1288/533, loss: 0.8690705299377441 2023-01-21 09:06:26.605292: step: 1292/533, loss: 0.7336040735244751 2023-01-21 09:06:27.806651: step: 1296/533, loss: 0.26735228300094604 2023-01-21 09:06:28.966019: step: 1300/533, loss: 0.12874527275562286 2023-01-21 09:06:30.146936: step: 1304/533, loss: 0.8387083411216736 2023-01-21 09:06:31.335428: step: 1308/533, loss: 0.29012781381607056 2023-01-21 09:06:32.502617: step: 1312/533, loss: 0.29116612672805786 2023-01-21 09:06:33.653348: step: 1316/533, loss: 0.06747221946716309 2023-01-21 09:06:34.828415: step: 1320/533, loss: 0.14437732100486755 2023-01-21 09:06:36.010370: step: 1324/533, loss: 0.950222373008728 2023-01-21 09:06:37.158331: step: 1328/533, loss: 1.8799209594726562 2023-01-21 09:06:38.363903: step: 1332/533, loss: 0.10964985191822052 2023-01-21 09:06:39.537041: step: 1336/533, loss: 0.11989335715770721 2023-01-21 09:06:40.691166: step: 1340/533, loss: 0.14323726296424866 2023-01-21 09:06:41.913401: step: 1344/533, loss: 0.21820011734962463 2023-01-21 09:06:43.126251: step: 1348/533, loss: 0.9188598394393921 2023-01-21 09:06:44.341595: step: 1352/533, loss: 0.06784620881080627 2023-01-21 09:06:45.539103: step: 1356/533, loss: 0.07366757094860077 2023-01-21 09:06:46.744173: step: 1360/533, loss: 1.4543697834014893 2023-01-21 09:06:47.932643: step: 1364/533, loss: 0.2703947424888611 2023-01-21 09:06:49.111543: step: 1368/533, loss: 0.2699599266052246 2023-01-21 09:06:50.340958: step: 1372/533, loss: 0.06850510090589523 2023-01-21 09:06:51.528100: step: 1376/533, loss: 0.4204274117946625 2023-01-21 09:06:52.664484: step: 1380/533, loss: 0.1322219967842102 2023-01-21 09:06:53.869375: step: 1384/533, loss: 0.1301138997077942 2023-01-21 09:06:55.076617: step: 1388/533, loss: 0.1648695021867752 2023-01-21 09:06:56.268120: step: 1392/533, loss: 0.3226913511753082 2023-01-21 09:06:57.439765: step: 1396/533, loss: 0.027720261365175247 2023-01-21 09:06:58.610960: step: 1400/533, loss: 0.1715250462293625 2023-01-21 09:06:59.764486: step: 1404/533, loss: 0.545713484287262 2023-01-21 09:07:00.924838: step: 1408/533, loss: 0.09135523438453674 2023-01-21 09:07:02.112198: step: 1412/533, loss: 0.1428672969341278 2023-01-21 09:07:03.339391: step: 1416/533, loss: 0.04816432297229767 2023-01-21 09:07:04.521828: step: 1420/533, loss: 0.04769277945160866 2023-01-21 09:07:05.682823: step: 1424/533, loss: 0.6040971875190735 2023-01-21 09:07:06.874812: step: 1428/533, loss: 1.1631361246109009 2023-01-21 09:07:08.093630: step: 1432/533, loss: 0.6827511787414551 2023-01-21 09:07:09.256800: step: 1436/533, loss: 0.3993563652038574 2023-01-21 09:07:10.436786: step: 1440/533, loss: 0.8167518377304077 2023-01-21 09:07:11.623195: step: 1444/533, loss: 0.823337733745575 2023-01-21 09:07:12.852445: step: 1448/533, loss: 0.17267131805419922 2023-01-21 09:07:14.017356: step: 1452/533, loss: 0.060135677456855774 2023-01-21 09:07:15.155754: step: 1456/533, loss: 1.0252174139022827 2023-01-21 09:07:16.357350: step: 1460/533, loss: 0.34034109115600586 2023-01-21 09:07:17.586119: step: 1464/533, loss: 0.6563622355461121 2023-01-21 09:07:18.783101: step: 1468/533, loss: 0.32413730025291443 2023-01-21 09:07:19.984206: step: 1472/533, loss: 0.036279771476984024 2023-01-21 09:07:21.155125: step: 1476/533, loss: 0.09050703048706055 2023-01-21 09:07:22.360727: step: 1480/533, loss: 0.8064340949058533 2023-01-21 09:07:23.523569: step: 1484/533, loss: 0.7008271217346191 2023-01-21 09:07:24.703943: step: 1488/533, loss: 0.3770490884780884 2023-01-21 09:07:25.895930: step: 1492/533, loss: 0.19439055025577545 2023-01-21 09:07:27.119446: step: 1496/533, loss: 0.5236929655075073 2023-01-21 09:07:28.309236: step: 1500/533, loss: 1.528775691986084 2023-01-21 09:07:29.531021: step: 1504/533, loss: 0.1137852668762207 2023-01-21 09:07:30.760822: step: 1508/533, loss: 0.43502336740493774 2023-01-21 09:07:31.946361: step: 1512/533, loss: 0.038903046399354935 2023-01-21 09:07:33.148442: step: 1516/533, loss: 0.3854188024997711 2023-01-21 09:07:34.368745: step: 1520/533, loss: 0.251146137714386 2023-01-21 09:07:35.555906: step: 1524/533, loss: 0.12915687263011932 2023-01-21 09:07:36.704341: step: 1528/533, loss: 0.673798143863678 2023-01-21 09:07:37.894031: step: 1532/533, loss: 0.12621363997459412 2023-01-21 09:07:39.069819: step: 1536/533, loss: 0.6278102993965149 2023-01-21 09:07:40.296245: step: 1540/533, loss: 0.06361560523509979 2023-01-21 09:07:41.469334: step: 1544/533, loss: 0.13660703599452972 2023-01-21 09:07:42.675047: step: 1548/533, loss: 0.15256424248218536 2023-01-21 09:07:43.884951: step: 1552/533, loss: 0.18162879347801208 2023-01-21 09:07:45.092408: step: 1556/533, loss: 0.3840428292751312 2023-01-21 09:07:46.263517: step: 1560/533, loss: 0.8691293001174927 2023-01-21 09:07:47.454981: step: 1564/533, loss: 0.0928887352347374 2023-01-21 09:07:48.612338: step: 1568/533, loss: 0.09873300045728683 2023-01-21 09:07:49.784440: step: 1572/533, loss: 0.3925350308418274 2023-01-21 09:07:50.957746: step: 1576/533, loss: 0.0813567191362381 2023-01-21 09:07:52.119507: step: 1580/533, loss: 0.1296248435974121 2023-01-21 09:07:53.286646: step: 1584/533, loss: 0.4144035577774048 2023-01-21 09:07:54.467656: step: 1588/533, loss: 0.7897875905036926 2023-01-21 09:07:55.663884: step: 1592/533, loss: 0.21992692351341248 2023-01-21 09:07:56.843985: step: 1596/533, loss: 0.17542344331741333 2023-01-21 09:07:57.990216: step: 1600/533, loss: 0.21425633132457733 2023-01-21 09:07:59.159975: step: 1604/533, loss: 0.22274942696094513 2023-01-21 09:08:00.359136: step: 1608/533, loss: 0.6761957406997681 2023-01-21 09:08:01.547528: step: 1612/533, loss: 0.26131343841552734 2023-01-21 09:08:02.681803: step: 1616/533, loss: 0.45399951934814453 2023-01-21 09:08:03.894090: step: 1620/533, loss: 0.7500147819519043 2023-01-21 09:08:05.086311: step: 1624/533, loss: 1.0922762155532837 2023-01-21 09:08:06.246086: step: 1628/533, loss: 0.7369341254234314 2023-01-21 09:08:07.417682: step: 1632/533, loss: 0.4089612066745758 2023-01-21 09:08:08.597303: step: 1636/533, loss: 0.897318959236145 2023-01-21 09:08:09.748734: step: 1640/533, loss: 0.6893734931945801 2023-01-21 09:08:10.973063: step: 1644/533, loss: 0.1404230147600174 2023-01-21 09:08:12.242105: step: 1648/533, loss: 0.21221551299095154 2023-01-21 09:08:13.417938: step: 1652/533, loss: 0.8159072995185852 2023-01-21 09:08:14.602357: step: 1656/533, loss: 0.19141311943531036 2023-01-21 09:08:15.801972: step: 1660/533, loss: 0.08337202668190002 2023-01-21 09:08:16.962883: step: 1664/533, loss: 1.0861512422561646 2023-01-21 09:08:18.126144: step: 1668/533, loss: 1.4600284099578857 2023-01-21 09:08:19.277678: step: 1672/533, loss: 0.06867833435535431 2023-01-21 09:08:20.450217: step: 1676/533, loss: 0.09389019012451172 2023-01-21 09:08:21.649211: step: 1680/533, loss: 0.03448162227869034 2023-01-21 09:08:22.874912: step: 1684/533, loss: 0.3975580632686615 2023-01-21 09:08:24.059520: step: 1688/533, loss: 0.16914740204811096 2023-01-21 09:08:25.280910: step: 1692/533, loss: 0.20642197132110596 2023-01-21 09:08:26.456776: step: 1696/533, loss: 0.11696415394544601 2023-01-21 09:08:27.649490: step: 1700/533, loss: 0.12218408286571503 2023-01-21 09:08:28.833396: step: 1704/533, loss: 0.05383472889661789 2023-01-21 09:08:29.996423: step: 1708/533, loss: 0.12333174049854279 2023-01-21 09:08:31.190069: step: 1712/533, loss: 0.9496538639068604 2023-01-21 09:08:32.349818: step: 1716/533, loss: 0.4803905189037323 2023-01-21 09:08:33.515933: step: 1720/533, loss: 0.22776460647583008 2023-01-21 09:08:34.702616: step: 1724/533, loss: 0.26100996136665344 2023-01-21 09:08:35.934027: step: 1728/533, loss: 0.7702292799949646 2023-01-21 09:08:37.115975: step: 1732/533, loss: 0.867817759513855 2023-01-21 09:08:38.314387: step: 1736/533, loss: 0.3525756895542145 2023-01-21 09:08:39.514405: step: 1740/533, loss: 0.2566913664340973 2023-01-21 09:08:40.691085: step: 1744/533, loss: 0.6176469922065735 2023-01-21 09:08:41.893327: step: 1748/533, loss: 0.14198055863380432 2023-01-21 09:08:43.101272: step: 1752/533, loss: 0.13684920966625214 2023-01-21 09:08:44.320540: step: 1756/533, loss: 0.28108978271484375 2023-01-21 09:08:45.536056: step: 1760/533, loss: 0.18111354112625122 2023-01-21 09:08:46.736265: step: 1764/533, loss: 0.4277893006801605 2023-01-21 09:08:47.919570: step: 1768/533, loss: 0.10900545865297318 2023-01-21 09:08:49.093325: step: 1772/533, loss: 0.11225280910730362 2023-01-21 09:08:50.292178: step: 1776/533, loss: 0.1711202710866928 2023-01-21 09:08:51.472377: step: 1780/533, loss: 0.09322099387645721 2023-01-21 09:08:52.638646: step: 1784/533, loss: 0.16490697860717773 2023-01-21 09:08:53.849621: step: 1788/533, loss: 0.5194489359855652 2023-01-21 09:08:55.008275: step: 1792/533, loss: 0.1701022982597351 2023-01-21 09:08:56.204340: step: 1796/533, loss: 0.3623226284980774 2023-01-21 09:08:57.370978: step: 1800/533, loss: 0.8791519403457642 2023-01-21 09:08:58.547673: step: 1804/533, loss: 1.0270390510559082 2023-01-21 09:08:59.731661: step: 1808/533, loss: 0.02732677385210991 2023-01-21 09:09:00.899764: step: 1812/533, loss: 0.24584579467773438 2023-01-21 09:09:02.044929: step: 1816/533, loss: 0.7096970677375793 2023-01-21 09:09:03.220080: step: 1820/533, loss: 0.12158603966236115 2023-01-21 09:09:04.403417: step: 1824/533, loss: 0.23516026139259338 2023-01-21 09:09:05.619496: step: 1828/533, loss: 0.14820708334445953 2023-01-21 09:09:06.772743: step: 1832/533, loss: 0.3918459415435791 2023-01-21 09:09:07.969839: step: 1836/533, loss: 0.13952083885669708 2023-01-21 09:09:09.134157: step: 1840/533, loss: 0.13386079668998718 2023-01-21 09:09:10.334044: step: 1844/533, loss: 0.1306193321943283 2023-01-21 09:09:11.548797: step: 1848/533, loss: 0.14490899443626404 2023-01-21 09:09:12.764382: step: 1852/533, loss: 0.1578243225812912 2023-01-21 09:09:13.959943: step: 1856/533, loss: 0.5925222635269165 2023-01-21 09:09:15.158627: step: 1860/533, loss: 0.44138574600219727 2023-01-21 09:09:16.348000: step: 1864/533, loss: 0.12281332910060883 2023-01-21 09:09:17.548647: step: 1868/533, loss: 0.4627809524536133 2023-01-21 09:09:18.710708: step: 1872/533, loss: 1.246553659439087 2023-01-21 09:09:19.905666: step: 1876/533, loss: 0.1605004519224167 2023-01-21 09:09:21.132091: step: 1880/533, loss: 0.28873175382614136 2023-01-21 09:09:22.299891: step: 1884/533, loss: 0.15015211701393127 2023-01-21 09:09:23.473419: step: 1888/533, loss: 0.15393801033496857 2023-01-21 09:09:24.679013: step: 1892/533, loss: 1.2088112831115723 2023-01-21 09:09:25.865464: step: 1896/533, loss: 0.06461191177368164 2023-01-21 09:09:26.999387: step: 1900/533, loss: 0.6443314552307129 2023-01-21 09:09:28.195412: step: 1904/533, loss: 0.19854307174682617 2023-01-21 09:09:29.372504: step: 1908/533, loss: 0.09643955528736115 2023-01-21 09:09:30.588130: step: 1912/533, loss: 1.6777030229568481 2023-01-21 09:09:31.792223: step: 1916/533, loss: 0.0324217826128006 2023-01-21 09:09:32.966156: step: 1920/533, loss: 0.05962047725915909 2023-01-21 09:09:34.188705: step: 1924/533, loss: 0.36219388246536255 2023-01-21 09:09:35.391868: step: 1928/533, loss: 0.23248833417892456 2023-01-21 09:09:36.588750: step: 1932/533, loss: 0.24731898307800293 2023-01-21 09:09:37.902692: step: 1936/533, loss: 1.1649845838546753 2023-01-21 09:09:39.073130: step: 1940/533, loss: 0.14077988266944885 2023-01-21 09:09:40.237413: step: 1944/533, loss: 0.8158080577850342 2023-01-21 09:09:41.385371: step: 1948/533, loss: 0.12570805847644806 2023-01-21 09:09:42.596864: step: 1952/533, loss: 0.2574155926704407 2023-01-21 09:09:43.733922: step: 1956/533, loss: 0.09502439200878143 2023-01-21 09:09:44.886691: step: 1960/533, loss: 0.726071834564209 2023-01-21 09:09:46.029656: step: 1964/533, loss: 0.7745004296302795 2023-01-21 09:09:47.176888: step: 1968/533, loss: 0.4669771194458008 2023-01-21 09:09:48.359800: step: 1972/533, loss: 0.4462575912475586 2023-01-21 09:09:49.468808: step: 1976/533, loss: 0.08509598672389984 2023-01-21 09:09:50.686371: step: 1980/533, loss: 0.5676898956298828 2023-01-21 09:09:51.893635: step: 1984/533, loss: 0.4625686705112457 2023-01-21 09:09:53.097209: step: 1988/533, loss: 7.225518703460693 2023-01-21 09:09:54.260102: step: 1992/533, loss: 0.11657486110925674 2023-01-21 09:09:55.440757: step: 1996/533, loss: 0.18175660073757172 2023-01-21 09:09:56.602833: step: 2000/533, loss: 1.0071289539337158 2023-01-21 09:09:57.747861: step: 2004/533, loss: 0.24380053579807281 2023-01-21 09:09:58.946097: step: 2008/533, loss: 0.23419040441513062 2023-01-21 09:10:00.096394: step: 2012/533, loss: 0.3166806697845459 2023-01-21 09:10:01.277501: step: 2016/533, loss: 0.20011290907859802 2023-01-21 09:10:02.454033: step: 2020/533, loss: 0.2254011034965515 2023-01-21 09:10:03.635504: step: 2024/533, loss: 0.21801738440990448 2023-01-21 09:10:04.788846: step: 2028/533, loss: 0.40872716903686523 2023-01-21 09:10:05.971218: step: 2032/533, loss: 0.507516086101532 2023-01-21 09:10:07.179978: step: 2036/533, loss: 0.07941646873950958 2023-01-21 09:10:08.363365: step: 2040/533, loss: 0.11726884543895721 2023-01-21 09:10:09.545417: step: 2044/533, loss: 0.3451942503452301 2023-01-21 09:10:10.767982: step: 2048/533, loss: 0.3090072572231293 2023-01-21 09:10:11.922023: step: 2052/533, loss: 0.10676155239343643 2023-01-21 09:10:13.093910: step: 2056/533, loss: 0.02714524418115616 2023-01-21 09:10:14.235897: step: 2060/533, loss: 0.20507307350635529 2023-01-21 09:10:15.419737: step: 2064/533, loss: 0.08268632739782333 2023-01-21 09:10:16.591704: step: 2068/533, loss: 0.10210895538330078 2023-01-21 09:10:17.792513: step: 2072/533, loss: 0.14143720269203186 2023-01-21 09:10:19.013108: step: 2076/533, loss: 0.05337691307067871 2023-01-21 09:10:20.171071: step: 2080/533, loss: 0.35757824778556824 2023-01-21 09:10:21.346161: step: 2084/533, loss: 7.25294303894043 2023-01-21 09:10:22.571419: step: 2088/533, loss: 1.0735610723495483 2023-01-21 09:10:23.746948: step: 2092/533, loss: 0.27614539861679077 2023-01-21 09:10:24.951110: step: 2096/533, loss: 1.4364241361618042 2023-01-21 09:10:26.092063: step: 2100/533, loss: 0.09752216935157776 2023-01-21 09:10:27.357273: step: 2104/533, loss: 0.2030380219221115 2023-01-21 09:10:28.552067: step: 2108/533, loss: 0.40994346141815186 2023-01-21 09:10:29.715788: step: 2112/533, loss: 0.07958989590406418 2023-01-21 09:10:30.861185: step: 2116/533, loss: 0.07022424042224884 2023-01-21 09:10:32.007134: step: 2120/533, loss: 0.20016518235206604 2023-01-21 09:10:33.155817: step: 2124/533, loss: 0.10159797966480255 2023-01-21 09:10:34.347855: step: 2128/533, loss: 0.20515227317810059 2023-01-21 09:10:35.509248: step: 2132/533, loss: 0.0617949478328228 ================================================== Loss: 0.478 -------------------- Dev: {'event': {'p': 0.6363636363636364, 'r': 0.7456724367509987, 'f1': 0.6866952789699571}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Test: {'event': {'p': 0.6506849315068494, 'r': 0.7289002557544757, 'f1': 0.6875753920386006}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Chinese: {'event': {'p': 0.5394736842105263, 'r': 0.7592592592592593, 'f1': 0.6307692307692309}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Korean: {'event': {'p': 0.6086956521739131, 'r': 0.4444444444444444, 'f1': 0.5137614678899082}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Russian: {'event': {'p': 0.45714285714285713, 'r': 0.4444444444444444, 'f1': 0.4507042253521127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6363636363636364, 'r': 0.7456724367509987, 'f1': 0.6866952789699571}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Korean: {'event': {'p': 0.6506849315068494, 'r': 0.7289002557544757, 'f1': 0.6875753920386006}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Korean: {'event': {'p': 0.6086956521739131, 'r': 0.4444444444444444, 'f1': 0.5137614678899082}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Russian: {'event': {'p': 0.6363636363636364, 'r': 0.7456724367509987, 'f1': 0.6866952789699571}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Russian: {'event': {'p': 0.6506849315068494, 'r': 0.7289002557544757, 'f1': 0.6875753920386006}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Russian: {'event': {'p': 0.45714285714285713, 'r': 0.4444444444444444, 'f1': 0.4507042253521127}, '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 09:11:33.037749: step: 4/533, loss: 0.16080769896507263 2023-01-21 09:11:34.224417: step: 8/533, loss: 0.2467871606349945 2023-01-21 09:11:35.389690: step: 12/533, loss: 0.12963686883449554 2023-01-21 09:11:36.553307: step: 16/533, loss: 0.06429843604564667 2023-01-21 09:11:37.718818: step: 20/533, loss: 1.1561965942382812 2023-01-21 09:11:38.854275: step: 24/533, loss: 0.2859885096549988 2023-01-21 09:11:40.026426: step: 28/533, loss: 0.13757991790771484 2023-01-21 09:11:41.216185: step: 32/533, loss: 0.7600545883178711 2023-01-21 09:11:42.405800: step: 36/533, loss: 0.06282529979944229 2023-01-21 09:11:43.590695: step: 40/533, loss: 0.03997869789600372 2023-01-21 09:11:44.752912: step: 44/533, loss: 0.17397208511829376 2023-01-21 09:11:45.918137: step: 48/533, loss: 0.11370354145765305 2023-01-21 09:11:47.071362: step: 52/533, loss: 0.046379707753658295 2023-01-21 09:11:48.215199: step: 56/533, loss: 0.7432664036750793 2023-01-21 09:11:49.375922: step: 60/533, loss: 0.16724443435668945 2023-01-21 09:11:50.530755: step: 64/533, loss: 0.0855528861284256 2023-01-21 09:11:51.695816: step: 68/533, loss: 0.31670600175857544 2023-01-21 09:11:52.830933: step: 72/533, loss: 0.09054632484912872 2023-01-21 09:11:53.977126: step: 76/533, loss: 0.04326023906469345 2023-01-21 09:11:55.186541: step: 80/533, loss: 0.12371959537267685 2023-01-21 09:11:56.452522: step: 84/533, loss: 0.16149845719337463 2023-01-21 09:11:57.619497: step: 88/533, loss: 0.2936019003391266 2023-01-21 09:11:58.805632: step: 92/533, loss: 0.2979351282119751 2023-01-21 09:11:59.980471: step: 96/533, loss: 0.14254418015480042 2023-01-21 09:12:01.188962: step: 100/533, loss: 0.7453964352607727 2023-01-21 09:12:02.372043: step: 104/533, loss: 0.3402903974056244 2023-01-21 09:12:03.558456: step: 108/533, loss: 0.1641683578491211 2023-01-21 09:12:04.714784: step: 112/533, loss: 0.2477407604455948 2023-01-21 09:12:05.926970: step: 116/533, loss: 0.07361679524183273 2023-01-21 09:12:07.100677: step: 120/533, loss: 0.043189048767089844 2023-01-21 09:12:08.270950: step: 124/533, loss: 0.38951969146728516 2023-01-21 09:12:09.425980: step: 128/533, loss: 1.3054147958755493 2023-01-21 09:12:10.591409: step: 132/533, loss: 0.4066828787326813 2023-01-21 09:12:11.727716: step: 136/533, loss: 0.05499096214771271 2023-01-21 09:12:12.899810: step: 140/533, loss: 0.03694338724017143 2023-01-21 09:12:14.068254: step: 144/533, loss: 0.2237229347229004 2023-01-21 09:12:15.234515: step: 148/533, loss: 0.17048291862010956 2023-01-21 09:12:16.435654: step: 152/533, loss: 0.1532120257616043 2023-01-21 09:12:17.657314: step: 156/533, loss: 0.1119285598397255 2023-01-21 09:12:18.817590: step: 160/533, loss: 0.16961650550365448 2023-01-21 09:12:20.010121: step: 164/533, loss: 0.08274183422327042 2023-01-21 09:12:21.156812: step: 168/533, loss: 0.9294984936714172 2023-01-21 09:12:22.354655: step: 172/533, loss: 0.5300347208976746 2023-01-21 09:12:23.501369: step: 176/533, loss: 0.0946936160326004 2023-01-21 09:12:24.726571: step: 180/533, loss: 0.1105434000492096 2023-01-21 09:12:25.925038: step: 184/533, loss: 0.9350584745407104 2023-01-21 09:12:27.102129: step: 188/533, loss: 0.35646992921829224 2023-01-21 09:12:28.305633: step: 192/533, loss: 0.20065680146217346 2023-01-21 09:12:29.483695: step: 196/533, loss: 0.10344738513231277 2023-01-21 09:12:30.633537: step: 200/533, loss: 0.2196740210056305 2023-01-21 09:12:31.785902: step: 204/533, loss: 7.556061267852783 2023-01-21 09:12:32.957524: step: 208/533, loss: 0.3177177309989929 2023-01-21 09:12:34.144898: step: 212/533, loss: 0.24030885100364685 2023-01-21 09:12:35.317406: step: 216/533, loss: 6.5061445236206055 2023-01-21 09:12:36.465012: step: 220/533, loss: 0.18813514709472656 2023-01-21 09:12:37.665560: step: 224/533, loss: 0.08334064483642578 2023-01-21 09:12:38.837843: step: 228/533, loss: 0.6746306419372559 2023-01-21 09:12:40.042102: step: 232/533, loss: 0.8121949434280396 2023-01-21 09:12:41.229972: step: 236/533, loss: 0.01866598054766655 2023-01-21 09:12:42.432317: step: 240/533, loss: 0.8045177459716797 2023-01-21 09:12:43.590160: step: 244/533, loss: 0.1645314246416092 2023-01-21 09:12:44.748905: step: 248/533, loss: 0.1700821816921234 2023-01-21 09:12:45.936866: step: 252/533, loss: 0.1886609047651291 2023-01-21 09:12:47.124377: step: 256/533, loss: 0.47371262311935425 2023-01-21 09:12:48.335588: step: 260/533, loss: 0.10804824531078339 2023-01-21 09:12:49.519922: step: 264/533, loss: 0.16475573182106018 2023-01-21 09:12:50.725466: step: 268/533, loss: 0.13311119377613068 2023-01-21 09:12:51.936361: step: 272/533, loss: 0.4139326214790344 2023-01-21 09:12:53.099222: step: 276/533, loss: 0.05772047117352486 2023-01-21 09:12:54.268481: step: 280/533, loss: 0.2530384063720703 2023-01-21 09:12:55.491091: step: 284/533, loss: 0.12808284163475037 2023-01-21 09:12:56.672709: step: 288/533, loss: 0.14574074745178223 2023-01-21 09:12:57.842844: step: 292/533, loss: 0.8609334826469421 2023-01-21 09:12:58.990385: step: 296/533, loss: 0.12245547771453857 2023-01-21 09:13:00.193219: step: 300/533, loss: 0.10447988659143448 2023-01-21 09:13:01.355102: step: 304/533, loss: 0.39735668897628784 2023-01-21 09:13:02.549476: step: 308/533, loss: 1.2928032875061035 2023-01-21 09:13:03.758143: step: 312/533, loss: 0.4389309883117676 2023-01-21 09:13:04.948574: step: 316/533, loss: 0.16427643597126007 2023-01-21 09:13:06.129245: step: 320/533, loss: 0.09371490776538849 2023-01-21 09:13:07.271425: step: 324/533, loss: 0.22092023491859436 2023-01-21 09:13:08.405888: step: 328/533, loss: 0.29821139574050903 2023-01-21 09:13:09.588946: step: 332/533, loss: 0.7922015190124512 2023-01-21 09:13:10.799440: step: 336/533, loss: 0.18724536895751953 2023-01-21 09:13:12.081590: step: 340/533, loss: 0.0818777084350586 2023-01-21 09:13:13.295417: step: 344/533, loss: 0.18011704087257385 2023-01-21 09:13:14.490046: step: 348/533, loss: 0.08503279834985733 2023-01-21 09:13:15.621328: step: 352/533, loss: 0.17050866782665253 2023-01-21 09:13:16.806333: step: 356/533, loss: 0.23868036270141602 2023-01-21 09:13:17.993049: step: 360/533, loss: 0.11245355755090714 2023-01-21 09:13:19.181370: step: 364/533, loss: 0.11277976632118225 2023-01-21 09:13:20.389519: step: 368/533, loss: 0.411241739988327 2023-01-21 09:13:21.563475: step: 372/533, loss: 0.7163174152374268 2023-01-21 09:13:22.741069: step: 376/533, loss: 0.09073544293642044 2023-01-21 09:13:23.899748: step: 380/533, loss: 0.7070478200912476 2023-01-21 09:13:25.071492: step: 384/533, loss: 0.46448269486427307 2023-01-21 09:13:26.258528: step: 388/533, loss: 0.1333259642124176 2023-01-21 09:13:27.430899: step: 392/533, loss: 0.36320582032203674 2023-01-21 09:13:28.624733: step: 396/533, loss: 0.38122254610061646 2023-01-21 09:13:29.828570: step: 400/533, loss: 0.3049568235874176 2023-01-21 09:13:31.031598: step: 404/533, loss: 0.23573628067970276 2023-01-21 09:13:32.203766: step: 408/533, loss: 0.6420763731002808 2023-01-21 09:13:33.369035: step: 412/533, loss: 1.008461356163025 2023-01-21 09:13:34.581642: step: 416/533, loss: 0.22769832611083984 2023-01-21 09:13:35.715848: step: 420/533, loss: 0.7806648015975952 2023-01-21 09:13:36.885916: step: 424/533, loss: 1.0949163436889648 2023-01-21 09:13:38.071791: step: 428/533, loss: 0.10100089013576508 2023-01-21 09:13:39.253863: step: 432/533, loss: 0.08356934040784836 2023-01-21 09:13:40.405570: step: 436/533, loss: 0.12097835540771484 2023-01-21 09:13:41.606715: step: 440/533, loss: 0.412949800491333 2023-01-21 09:13:42.758099: step: 444/533, loss: 0.4608314633369446 2023-01-21 09:13:44.007402: step: 448/533, loss: 0.1743953824043274 2023-01-21 09:13:45.169068: step: 452/533, loss: 0.6128075122833252 2023-01-21 09:13:46.366067: step: 456/533, loss: 0.19651412963867188 2023-01-21 09:13:47.595834: step: 460/533, loss: 0.6721065044403076 2023-01-21 09:13:48.771747: step: 464/533, loss: 0.15967969596385956 2023-01-21 09:13:49.940007: step: 468/533, loss: 0.20060434937477112 2023-01-21 09:13:51.110603: step: 472/533, loss: 0.19437763094902039 2023-01-21 09:13:52.343376: step: 476/533, loss: 0.11729685962200165 2023-01-21 09:13:53.556217: step: 480/533, loss: 0.43636640906333923 2023-01-21 09:13:54.744275: step: 484/533, loss: 0.38820579648017883 2023-01-21 09:13:55.952019: step: 488/533, loss: 0.036032892763614655 2023-01-21 09:13:57.122627: step: 492/533, loss: 0.20904521644115448 2023-01-21 09:13:58.311456: step: 496/533, loss: 0.14324331283569336 2023-01-21 09:13:59.459033: step: 500/533, loss: 0.8581582307815552 2023-01-21 09:14:00.590668: step: 504/533, loss: 0.009099816903471947 2023-01-21 09:14:01.763099: step: 508/533, loss: 6.096617698669434 2023-01-21 09:14:02.916783: step: 512/533, loss: 0.07436814159154892 2023-01-21 09:14:04.095050: step: 516/533, loss: 0.8195649981498718 2023-01-21 09:14:05.249113: step: 520/533, loss: 0.7428781390190125 2023-01-21 09:14:06.433954: step: 524/533, loss: 0.3563980460166931 2023-01-21 09:14:07.616198: step: 528/533, loss: 0.2819378972053528 2023-01-21 09:14:08.803099: step: 532/533, loss: 0.07229113578796387 2023-01-21 09:14:09.993111: step: 536/533, loss: 0.10850057750940323 2023-01-21 09:14:11.146976: step: 540/533, loss: 0.7728476524353027 2023-01-21 09:14:12.354055: step: 544/533, loss: 0.27718108892440796 2023-01-21 09:14:13.514731: step: 548/533, loss: 0.20332804322242737 2023-01-21 09:14:14.724772: step: 552/533, loss: 0.1661197692155838 2023-01-21 09:14:15.930827: step: 556/533, loss: 0.34860095381736755 2023-01-21 09:14:17.117055: step: 560/533, loss: 0.013684368692338467 2023-01-21 09:14:18.273126: step: 564/533, loss: 0.15758386254310608 2023-01-21 09:14:19.469171: step: 568/533, loss: 0.2174517810344696 2023-01-21 09:14:20.657431: step: 572/533, loss: 0.3024938702583313 2023-01-21 09:14:21.859427: step: 576/533, loss: 0.1184653490781784 2023-01-21 09:14:23.053174: step: 580/533, loss: 1.110622525215149 2023-01-21 09:14:24.226976: step: 584/533, loss: 0.16563330590724945 2023-01-21 09:14:25.403684: step: 588/533, loss: 0.6151212453842163 2023-01-21 09:14:26.597823: step: 592/533, loss: 0.11995840072631836 2023-01-21 09:14:27.815534: step: 596/533, loss: 0.07827548682689667 2023-01-21 09:14:28.963665: step: 600/533, loss: 0.2466479241847992 2023-01-21 09:14:30.166116: step: 604/533, loss: 0.8222202658653259 2023-01-21 09:14:31.346387: step: 608/533, loss: 0.10267267376184464 2023-01-21 09:14:32.460761: step: 612/533, loss: 0.17090922594070435 2023-01-21 09:14:33.648317: step: 616/533, loss: 0.338049054145813 2023-01-21 09:14:34.810466: step: 620/533, loss: 0.1157594695687294 2023-01-21 09:14:35.995541: step: 624/533, loss: 1.3425092697143555 2023-01-21 09:14:37.155398: step: 628/533, loss: 0.20160922408103943 2023-01-21 09:14:38.350410: step: 632/533, loss: 0.2886773943901062 2023-01-21 09:14:39.585452: step: 636/533, loss: 0.059935905039310455 2023-01-21 09:14:40.753046: step: 640/533, loss: 0.10588303208351135 2023-01-21 09:14:41.918475: step: 644/533, loss: 0.5502110719680786 2023-01-21 09:14:43.097927: step: 648/533, loss: 0.298610121011734 2023-01-21 09:14:44.301224: step: 652/533, loss: 0.2920886278152466 2023-01-21 09:14:45.477923: step: 656/533, loss: 0.10092958807945251 2023-01-21 09:14:46.596187: step: 660/533, loss: 0.33596089482307434 2023-01-21 09:14:47.769475: step: 664/533, loss: 0.6211967468261719 2023-01-21 09:14:48.968388: step: 668/533, loss: 0.1802423596382141 2023-01-21 09:14:50.125164: step: 672/533, loss: 0.09360642731189728 2023-01-21 09:14:51.348730: step: 676/533, loss: 0.38748520612716675 2023-01-21 09:14:52.543567: step: 680/533, loss: 0.6602237820625305 2023-01-21 09:14:53.702744: step: 684/533, loss: 0.20110264420509338 2023-01-21 09:14:54.874810: step: 688/533, loss: 6.789928436279297 2023-01-21 09:14:56.066532: step: 692/533, loss: 0.7050453424453735 2023-01-21 09:14:57.237672: step: 696/533, loss: 6.557606220245361 2023-01-21 09:14:58.402848: step: 700/533, loss: 0.2180258333683014 2023-01-21 09:14:59.556306: step: 704/533, loss: 1.156533122062683 2023-01-21 09:15:00.750465: step: 708/533, loss: 0.18626396358013153 2023-01-21 09:15:01.947573: step: 712/533, loss: 0.16130739450454712 2023-01-21 09:15:03.131473: step: 716/533, loss: 0.11807730793952942 2023-01-21 09:15:04.303592: step: 720/533, loss: 0.1919388771057129 2023-01-21 09:15:05.486776: step: 724/533, loss: 0.23629441857337952 2023-01-21 09:15:06.662866: step: 728/533, loss: 0.14452704787254333 2023-01-21 09:15:07.850803: step: 732/533, loss: 0.1256023347377777 2023-01-21 09:15:09.001224: step: 736/533, loss: 5.827259063720703 2023-01-21 09:15:10.161145: step: 740/533, loss: 0.12067222595214844 2023-01-21 09:15:11.363607: step: 744/533, loss: 0.0689108818769455 2023-01-21 09:15:12.553126: step: 748/533, loss: 0.028519347310066223 2023-01-21 09:15:13.712373: step: 752/533, loss: 0.6715900301933289 2023-01-21 09:15:14.856427: step: 756/533, loss: 0.11868033558130264 2023-01-21 09:15:16.044214: step: 760/533, loss: 0.20304298400878906 2023-01-21 09:15:17.230116: step: 764/533, loss: 0.4125555157661438 2023-01-21 09:15:18.420310: step: 768/533, loss: 0.6204870343208313 2023-01-21 09:15:19.656855: step: 772/533, loss: 0.7362143993377686 2023-01-21 09:15:20.833846: step: 776/533, loss: 0.11665868759155273 2023-01-21 09:15:22.014005: step: 780/533, loss: 0.7963329553604126 2023-01-21 09:15:23.187488: step: 784/533, loss: 0.0415625125169754 2023-01-21 09:15:24.355811: step: 788/533, loss: 0.12673553824424744 2023-01-21 09:15:25.537415: step: 792/533, loss: 1.5688321590423584 2023-01-21 09:15:26.765107: step: 796/533, loss: 0.20103083550930023 2023-01-21 09:15:27.988911: step: 800/533, loss: 0.33872976899147034 2023-01-21 09:15:29.208062: step: 804/533, loss: 0.4234919548034668 2023-01-21 09:15:30.353740: step: 808/533, loss: 0.6318029761314392 2023-01-21 09:15:31.550627: step: 812/533, loss: 0.1803821623325348 2023-01-21 09:15:32.696640: step: 816/533, loss: 0.1712852418422699 2023-01-21 09:15:33.815414: step: 820/533, loss: 0.09087991714477539 2023-01-21 09:15:34.969359: step: 824/533, loss: 0.2951028048992157 2023-01-21 09:15:36.170909: step: 828/533, loss: 0.3447265326976776 2023-01-21 09:15:37.386151: step: 832/533, loss: 0.2435164451599121 2023-01-21 09:15:38.619388: step: 836/533, loss: 0.5090039968490601 2023-01-21 09:15:39.767097: step: 840/533, loss: 0.6650123000144958 2023-01-21 09:15:40.930709: step: 844/533, loss: 0.6050577759742737 2023-01-21 09:15:42.118418: step: 848/533, loss: 0.11153145134449005 2023-01-21 09:15:43.260202: step: 852/533, loss: 0.13925257325172424 2023-01-21 09:15:44.453181: step: 856/533, loss: 0.7219201326370239 2023-01-21 09:15:45.616249: step: 860/533, loss: 0.2796954810619354 2023-01-21 09:15:46.772176: step: 864/533, loss: 0.09976521134376526 2023-01-21 09:15:47.924068: step: 868/533, loss: 0.1735800802707672 2023-01-21 09:15:49.129138: step: 872/533, loss: 0.6152204275131226 2023-01-21 09:15:50.314372: step: 876/533, loss: 0.1786670684814453 2023-01-21 09:15:51.466259: step: 880/533, loss: 0.556827962398529 2023-01-21 09:15:52.697279: step: 884/533, loss: 0.725416362285614 2023-01-21 09:15:53.874732: step: 888/533, loss: 0.9925830960273743 2023-01-21 09:15:55.029164: step: 892/533, loss: 0.0687982589006424 2023-01-21 09:15:56.199817: step: 896/533, loss: 0.04963874816894531 2023-01-21 09:15:57.350689: step: 900/533, loss: 0.09570202976465225 2023-01-21 09:15:58.525423: step: 904/533, loss: 1.0855121612548828 2023-01-21 09:15:59.706180: step: 908/533, loss: 0.08802032470703125 2023-01-21 09:16:00.898349: step: 912/533, loss: 1.807556390762329 2023-01-21 09:16:02.082860: step: 916/533, loss: 1.6945182085037231 2023-01-21 09:16:03.244646: step: 920/533, loss: 0.09521179646253586 2023-01-21 09:16:04.410880: step: 924/533, loss: 0.07061167061328888 2023-01-21 09:16:05.575450: step: 928/533, loss: 0.2166360318660736 2023-01-21 09:16:06.723794: step: 932/533, loss: 0.09166759997606277 2023-01-21 09:16:07.925180: step: 936/533, loss: 0.07578954845666885 2023-01-21 09:16:09.096656: step: 940/533, loss: 0.8715420961380005 2023-01-21 09:16:10.256156: step: 944/533, loss: 0.0767279639840126 2023-01-21 09:16:11.391647: step: 948/533, loss: 0.19810661673545837 2023-01-21 09:16:12.585435: step: 952/533, loss: 0.08481831848621368 2023-01-21 09:16:13.769508: step: 956/533, loss: 0.43105068802833557 2023-01-21 09:16:14.911517: step: 960/533, loss: 0.28990793228149414 2023-01-21 09:16:16.098423: step: 964/533, loss: 0.08686523139476776 2023-01-21 09:16:17.290889: step: 968/533, loss: 0.718145489692688 2023-01-21 09:16:18.496791: step: 972/533, loss: 0.33432263135910034 2023-01-21 09:16:19.663614: step: 976/533, loss: 0.8757674694061279 2023-01-21 09:16:20.826743: step: 980/533, loss: 0.05391073226928711 2023-01-21 09:16:22.006380: step: 984/533, loss: 0.045043185353279114 2023-01-21 09:16:23.172445: step: 988/533, loss: 0.11546371132135391 2023-01-21 09:16:24.369373: step: 992/533, loss: 0.7150843143463135 2023-01-21 09:16:25.514289: step: 996/533, loss: 0.24153746664524078 2023-01-21 09:16:26.706622: step: 1000/533, loss: 0.7552597522735596 2023-01-21 09:16:27.891712: step: 1004/533, loss: 0.6755239963531494 2023-01-21 09:16:29.032640: step: 1008/533, loss: 0.42555132508277893 2023-01-21 09:16:30.250085: step: 1012/533, loss: 0.17450609803199768 2023-01-21 09:16:31.400278: step: 1016/533, loss: 0.07074947655200958 2023-01-21 09:16:32.581120: step: 1020/533, loss: 0.25498390197753906 2023-01-21 09:16:33.785554: step: 1024/533, loss: 0.0510433204472065 2023-01-21 09:16:34.996416: step: 1028/533, loss: 0.13398166000843048 2023-01-21 09:16:36.162743: step: 1032/533, loss: 0.1743348240852356 2023-01-21 09:16:37.311999: step: 1036/533, loss: 0.2929289937019348 2023-01-21 09:16:38.467794: step: 1040/533, loss: 0.17708474397659302 2023-01-21 09:16:39.677563: step: 1044/533, loss: 0.04255685955286026 2023-01-21 09:16:40.818926: step: 1048/533, loss: 0.09301743656396866 2023-01-21 09:16:41.987144: step: 1052/533, loss: 0.35009899735450745 2023-01-21 09:16:43.196034: step: 1056/533, loss: 0.42279911041259766 2023-01-21 09:16:44.365839: step: 1060/533, loss: 0.04605751112103462 2023-01-21 09:16:45.539889: step: 1064/533, loss: 0.040482521057128906 2023-01-21 09:16:46.701536: step: 1068/533, loss: 0.7885123491287231 2023-01-21 09:16:47.855757: step: 1072/533, loss: 5.510984420776367 2023-01-21 09:16:48.997132: step: 1076/533, loss: 0.7045764923095703 2023-01-21 09:16:50.174323: step: 1080/533, loss: 0.11668120324611664 2023-01-21 09:16:51.381119: step: 1084/533, loss: 0.15884742140769958 2023-01-21 09:16:52.537292: step: 1088/533, loss: 0.14098411798477173 2023-01-21 09:16:53.752256: step: 1092/533, loss: 0.19659024477005005 2023-01-21 09:16:54.935970: step: 1096/533, loss: 1.9845985174179077 2023-01-21 09:16:56.152949: step: 1100/533, loss: 0.45900583267211914 2023-01-21 09:16:57.304915: step: 1104/533, loss: 0.7520493865013123 2023-01-21 09:16:58.483683: step: 1108/533, loss: 6.177628993988037 2023-01-21 09:16:59.686519: step: 1112/533, loss: 0.3970717489719391 2023-01-21 09:17:00.878408: step: 1116/533, loss: 0.08741851150989532 2023-01-21 09:17:02.054693: step: 1120/533, loss: 0.01384887658059597 2023-01-21 09:17:03.210528: step: 1124/533, loss: 0.0630807876586914 2023-01-21 09:17:04.434101: step: 1128/533, loss: 0.1139598861336708 2023-01-21 09:17:05.581254: step: 1132/533, loss: 0.09872536361217499 2023-01-21 09:17:06.728416: step: 1136/533, loss: 1.0366337299346924 2023-01-21 09:17:07.960009: step: 1140/533, loss: 0.6433200836181641 2023-01-21 09:17:09.156663: step: 1144/533, loss: 0.07232493907213211 2023-01-21 09:17:10.331164: step: 1148/533, loss: 0.0791909247636795 2023-01-21 09:17:11.501416: step: 1152/533, loss: 0.17272216081619263 2023-01-21 09:17:12.677577: step: 1156/533, loss: 0.3607533574104309 2023-01-21 09:17:13.847658: step: 1160/533, loss: 0.2840801477432251 2023-01-21 09:17:15.016848: step: 1164/533, loss: 0.09294863045215607 2023-01-21 09:17:16.178417: step: 1168/533, loss: 0.1951891928911209 2023-01-21 09:17:17.352765: step: 1172/533, loss: 0.23752471804618835 2023-01-21 09:17:18.505670: step: 1176/533, loss: 0.08013544231653214 2023-01-21 09:17:19.677428: step: 1180/533, loss: 0.385166734457016 2023-01-21 09:17:20.907079: step: 1184/533, loss: 0.28750577569007874 2023-01-21 09:17:22.109582: step: 1188/533, loss: 0.20149211585521698 2023-01-21 09:17:23.272285: step: 1192/533, loss: 0.06264295428991318 2023-01-21 09:17:24.465869: step: 1196/533, loss: 0.5978087186813354 2023-01-21 09:17:25.619308: step: 1200/533, loss: 0.11043882369995117 2023-01-21 09:17:26.795570: step: 1204/533, loss: 0.6323989629745483 2023-01-21 09:17:27.968192: step: 1208/533, loss: 0.541048526763916 2023-01-21 09:17:29.132220: step: 1212/533, loss: 0.14680881798267365 2023-01-21 09:17:30.299598: step: 1216/533, loss: 0.19817982614040375 2023-01-21 09:17:31.528294: step: 1220/533, loss: 0.10579681396484375 2023-01-21 09:17:32.716014: step: 1224/533, loss: 1.2712929248809814 2023-01-21 09:17:33.891299: step: 1228/533, loss: 0.5510305762290955 2023-01-21 09:17:35.080845: step: 1232/533, loss: 0.08197517693042755 2023-01-21 09:17:36.264855: step: 1236/533, loss: 0.14559394121170044 2023-01-21 09:17:37.448779: step: 1240/533, loss: 0.21898965537548065 2023-01-21 09:17:38.634207: step: 1244/533, loss: 0.11944732815027237 2023-01-21 09:17:39.880661: step: 1248/533, loss: 0.711075484752655 2023-01-21 09:17:41.141306: step: 1252/533, loss: 0.24308347702026367 2023-01-21 09:17:42.347533: step: 1256/533, loss: 0.7453237771987915 2023-01-21 09:17:43.495997: step: 1260/533, loss: 0.7348448038101196 2023-01-21 09:17:44.713199: step: 1264/533, loss: 0.5355383157730103 2023-01-21 09:17:45.861813: step: 1268/533, loss: 0.17442288994789124 2023-01-21 09:17:47.019296: step: 1272/533, loss: 0.081713005900383 2023-01-21 09:17:48.190030: step: 1276/533, loss: 5.200689315795898 2023-01-21 09:17:49.384589: step: 1280/533, loss: 0.38226261734962463 2023-01-21 09:17:50.604558: step: 1284/533, loss: 0.31229841709136963 2023-01-21 09:17:51.777077: step: 1288/533, loss: 0.0915495902299881 2023-01-21 09:17:52.956485: step: 1292/533, loss: 0.11697855591773987 2023-01-21 09:17:54.111608: step: 1296/533, loss: 0.1356651484966278 2023-01-21 09:17:55.267176: step: 1300/533, loss: 0.3107897639274597 2023-01-21 09:17:56.442438: step: 1304/533, loss: 0.32370811700820923 2023-01-21 09:17:57.612393: step: 1308/533, loss: 0.4293191432952881 2023-01-21 09:17:58.810747: step: 1312/533, loss: 0.13465996086597443 2023-01-21 09:18:00.030442: step: 1316/533, loss: 0.18126791715621948 2023-01-21 09:18:01.202179: step: 1320/533, loss: 0.16612711548805237 2023-01-21 09:18:02.352502: step: 1324/533, loss: 0.4484199583530426 2023-01-21 09:18:03.543473: step: 1328/533, loss: 0.3617344796657562 2023-01-21 09:18:04.693807: step: 1332/533, loss: 0.040073588490486145 2023-01-21 09:18:05.836438: step: 1336/533, loss: 0.25802773237228394 2023-01-21 09:18:06.984598: step: 1340/533, loss: 0.16844701766967773 2023-01-21 09:18:08.138372: step: 1344/533, loss: 0.1379452347755432 2023-01-21 09:18:09.311656: step: 1348/533, loss: 0.8102596998214722 2023-01-21 09:18:10.488480: step: 1352/533, loss: 0.029878616333007812 2023-01-21 09:18:11.666134: step: 1356/533, loss: 0.21733397245407104 2023-01-21 09:18:12.878980: step: 1360/533, loss: 0.1776810735464096 2023-01-21 09:18:14.075292: step: 1364/533, loss: 0.3859017491340637 2023-01-21 09:18:15.260476: step: 1368/533, loss: 0.2634678781032562 2023-01-21 09:18:16.433529: step: 1372/533, loss: 0.29899561405181885 2023-01-21 09:18:17.614529: step: 1376/533, loss: 0.38006791472435 2023-01-21 09:18:18.743194: step: 1380/533, loss: 0.04742402955889702 2023-01-21 09:18:19.928964: step: 1384/533, loss: 0.18840648233890533 2023-01-21 09:18:21.090639: step: 1388/533, loss: 0.03719749301671982 2023-01-21 09:18:22.245820: step: 1392/533, loss: 0.6020131707191467 2023-01-21 09:18:23.465097: step: 1396/533, loss: 0.3525051176548004 2023-01-21 09:18:24.648166: step: 1400/533, loss: 0.42236506938934326 2023-01-21 09:18:25.834670: step: 1404/533, loss: 1.0156099796295166 2023-01-21 09:18:26.972845: step: 1408/533, loss: 0.31141966581344604 2023-01-21 09:18:28.128280: step: 1412/533, loss: 0.06403665244579315 2023-01-21 09:18:29.311099: step: 1416/533, loss: 0.16745629906654358 2023-01-21 09:18:30.469473: step: 1420/533, loss: 0.20169982314109802 2023-01-21 09:18:31.698533: step: 1424/533, loss: 0.871826171875 2023-01-21 09:18:32.901629: step: 1428/533, loss: 0.8681539297103882 2023-01-21 09:18:34.081032: step: 1432/533, loss: 0.09811706840991974 2023-01-21 09:18:35.293791: step: 1436/533, loss: 0.14755316078662872 2023-01-21 09:18:36.455439: step: 1440/533, loss: 0.23911935091018677 2023-01-21 09:18:37.657534: step: 1444/533, loss: 0.3370138108730316 2023-01-21 09:18:38.844290: step: 1448/533, loss: 0.6119120717048645 2023-01-21 09:18:40.028323: step: 1452/533, loss: 0.36522483825683594 2023-01-21 09:18:41.197197: step: 1456/533, loss: 0.06301660090684891 2023-01-21 09:18:42.377581: step: 1460/533, loss: 0.10820493847131729 2023-01-21 09:18:43.528840: step: 1464/533, loss: 0.28848162293434143 2023-01-21 09:18:44.676741: step: 1468/533, loss: 0.079480841755867 2023-01-21 09:18:45.893465: step: 1472/533, loss: 0.10893673449754715 2023-01-21 09:18:47.083940: step: 1476/533, loss: 0.9639902114868164 2023-01-21 09:18:48.263938: step: 1480/533, loss: 0.6345762610435486 2023-01-21 09:18:49.397285: step: 1484/533, loss: 0.08421941101551056 2023-01-21 09:18:50.558276: step: 1488/533, loss: 0.0578124038875103 2023-01-21 09:18:51.750179: step: 1492/533, loss: 0.31327182054519653 2023-01-21 09:18:52.936379: step: 1496/533, loss: 0.08017121255397797 2023-01-21 09:18:54.137147: step: 1500/533, loss: 0.22312337160110474 2023-01-21 09:18:55.306449: step: 1504/533, loss: 0.637442409992218 2023-01-21 09:18:56.530748: step: 1508/533, loss: 1.16358482837677 2023-01-21 09:18:57.756967: step: 1512/533, loss: 0.08317212760448456 2023-01-21 09:18:58.907416: step: 1516/533, loss: 0.2635626792907715 2023-01-21 09:19:00.093881: step: 1520/533, loss: 0.11569662392139435 2023-01-21 09:19:01.304999: step: 1524/533, loss: 0.12475547939538956 2023-01-21 09:19:02.489622: step: 1528/533, loss: 0.7947450280189514 2023-01-21 09:19:03.668463: step: 1532/533, loss: 0.5174509882926941 2023-01-21 09:19:04.890903: step: 1536/533, loss: 0.14672188460826874 2023-01-21 09:19:06.052959: step: 1540/533, loss: 0.01967933215200901 2023-01-21 09:19:07.199880: step: 1544/533, loss: 0.47687575221061707 2023-01-21 09:19:08.366078: step: 1548/533, loss: 0.06152964010834694 2023-01-21 09:19:09.513136: step: 1552/533, loss: 0.09312210977077484 2023-01-21 09:19:10.686975: step: 1556/533, loss: 0.089032843708992 2023-01-21 09:19:11.890394: step: 1560/533, loss: 0.4784398078918457 2023-01-21 09:19:13.091686: step: 1564/533, loss: 0.14908543229103088 2023-01-21 09:19:14.280484: step: 1568/533, loss: 0.49519291520118713 2023-01-21 09:19:15.439106: step: 1572/533, loss: 0.20832185447216034 2023-01-21 09:19:16.578446: step: 1576/533, loss: 0.23972997069358826 2023-01-21 09:19:17.768129: step: 1580/533, loss: 0.2782464027404785 2023-01-21 09:19:18.927966: step: 1584/533, loss: 0.04326915740966797 2023-01-21 09:19:20.147358: step: 1588/533, loss: 0.08292331546545029 2023-01-21 09:19:21.361277: step: 1592/533, loss: 0.7723281383514404 2023-01-21 09:19:22.551556: step: 1596/533, loss: 0.49383050203323364 2023-01-21 09:19:23.716316: step: 1600/533, loss: 0.3659517765045166 2023-01-21 09:19:24.906261: step: 1604/533, loss: 0.06499271094799042 2023-01-21 09:19:26.083881: step: 1608/533, loss: 0.25885510444641113 2023-01-21 09:19:27.263655: step: 1612/533, loss: 0.05401439964771271 2023-01-21 09:19:28.504384: step: 1616/533, loss: 0.267822265625 2023-01-21 09:19:29.664209: step: 1620/533, loss: 0.6613141298294067 2023-01-21 09:19:30.868218: step: 1624/533, loss: 0.3519867956638336 2023-01-21 09:19:32.057819: step: 1628/533, loss: 0.22230637073516846 2023-01-21 09:19:33.208165: step: 1632/533, loss: 0.03190779685974121 2023-01-21 09:19:34.385293: step: 1636/533, loss: 1.0905355215072632 2023-01-21 09:19:35.568876: step: 1640/533, loss: 0.05632734298706055 2023-01-21 09:19:36.728506: step: 1644/533, loss: 0.18079432845115662 2023-01-21 09:19:37.927246: step: 1648/533, loss: 0.14281006157398224 2023-01-21 09:19:39.086185: step: 1652/533, loss: 1.0845324993133545 2023-01-21 09:19:40.262029: step: 1656/533, loss: 0.10688714683055878 2023-01-21 09:19:41.408576: step: 1660/533, loss: 0.08639836311340332 2023-01-21 09:19:42.642571: step: 1664/533, loss: 0.5937854647636414 2023-01-21 09:19:43.819588: step: 1668/533, loss: 0.05844841152429581 2023-01-21 09:19:44.982966: step: 1672/533, loss: 0.07084226608276367 2023-01-21 09:19:46.130462: step: 1676/533, loss: 0.10019215941429138 2023-01-21 09:19:47.324945: step: 1680/533, loss: 0.052054405212402344 2023-01-21 09:19:48.494221: step: 1684/533, loss: 0.5961360931396484 2023-01-21 09:19:49.650701: step: 1688/533, loss: 0.6336082816123962 2023-01-21 09:19:50.851047: step: 1692/533, loss: 0.17626100778579712 2023-01-21 09:19:52.042099: step: 1696/533, loss: 0.2126356065273285 2023-01-21 09:19:53.246866: step: 1700/533, loss: 0.6530532240867615 2023-01-21 09:19:54.430072: step: 1704/533, loss: 0.1026669517159462 2023-01-21 09:19:55.622587: step: 1708/533, loss: 0.06398515403270721 2023-01-21 09:19:56.800904: step: 1712/533, loss: 0.019005395472049713 2023-01-21 09:19:57.982060: step: 1716/533, loss: 0.14805489778518677 2023-01-21 09:19:59.185037: step: 1720/533, loss: 0.807794451713562 2023-01-21 09:20:00.367302: step: 1724/533, loss: 0.11912989616394043 2023-01-21 09:20:01.597739: step: 1728/533, loss: 0.19164490699768066 2023-01-21 09:20:02.738312: step: 1732/533, loss: 0.4997274875640869 2023-01-21 09:20:03.902974: step: 1736/533, loss: 0.08028631657361984 2023-01-21 09:20:05.102706: step: 1740/533, loss: 0.38860541582107544 2023-01-21 09:20:06.294288: step: 1744/533, loss: 0.7654044032096863 2023-01-21 09:20:07.439654: step: 1748/533, loss: 0.23120225965976715 2023-01-21 09:20:08.610356: step: 1752/533, loss: 0.05283026397228241 2023-01-21 09:20:09.766223: step: 1756/533, loss: 0.2377423346042633 2023-01-21 09:20:10.977410: step: 1760/533, loss: 0.09864912182092667 2023-01-21 09:20:12.150281: step: 1764/533, loss: 0.6943032741546631 2023-01-21 09:20:13.338334: step: 1768/533, loss: 0.4640032649040222 2023-01-21 09:20:14.513520: step: 1772/533, loss: 0.6499159336090088 2023-01-21 09:20:15.681493: step: 1776/533, loss: 0.136498361825943 2023-01-21 09:20:16.860293: step: 1780/533, loss: 0.7912119626998901 2023-01-21 09:20:18.025760: step: 1784/533, loss: 0.17415256798267365 2023-01-21 09:20:19.166803: step: 1788/533, loss: 0.39642849564552307 2023-01-21 09:20:20.342862: step: 1792/533, loss: 0.25583165884017944 2023-01-21 09:20:21.506504: step: 1796/533, loss: 1.0334510803222656 2023-01-21 09:20:22.671292: step: 1800/533, loss: 0.030234910547733307 2023-01-21 09:20:23.861870: step: 1804/533, loss: 0.8843833208084106 2023-01-21 09:20:25.049471: step: 1808/533, loss: 0.11465645581483841 2023-01-21 09:20:26.239083: step: 1812/533, loss: 0.29103779792785645 2023-01-21 09:20:27.420768: step: 1816/533, loss: 0.17670278251171112 2023-01-21 09:20:28.601677: step: 1820/533, loss: 0.07809782028198242 2023-01-21 09:20:29.810545: step: 1824/533, loss: 0.15545521676540375 2023-01-21 09:20:30.966421: step: 1828/533, loss: 0.21935686469078064 2023-01-21 09:20:32.151071: step: 1832/533, loss: 0.1724845916032791 2023-01-21 09:20:33.294113: step: 1836/533, loss: 0.07348985970020294 2023-01-21 09:20:34.475931: step: 1840/533, loss: 0.10673341155052185 2023-01-21 09:20:35.642269: step: 1844/533, loss: 1.6208428144454956 2023-01-21 09:20:36.774971: step: 1848/533, loss: 0.09668588638305664 2023-01-21 09:20:38.004849: step: 1852/533, loss: 0.7784967422485352 2023-01-21 09:20:39.170254: step: 1856/533, loss: 0.36850759387016296 2023-01-21 09:20:40.361843: step: 1860/533, loss: 0.2463562935590744 2023-01-21 09:20:41.559980: step: 1864/533, loss: 1.013809084892273 2023-01-21 09:20:42.750477: step: 1868/533, loss: 0.16174575686454773 2023-01-21 09:20:43.947100: step: 1872/533, loss: 0.1303505003452301 2023-01-21 09:20:45.138294: step: 1876/533, loss: 0.2305046170949936 2023-01-21 09:20:46.327639: step: 1880/533, loss: 0.44880956411361694 2023-01-21 09:20:47.522086: step: 1884/533, loss: 0.10925905406475067 2023-01-21 09:20:48.671673: step: 1888/533, loss: 0.25126829743385315 2023-01-21 09:20:49.865686: step: 1892/533, loss: 0.27894097566604614 2023-01-21 09:20:51.065041: step: 1896/533, loss: 0.5337204933166504 2023-01-21 09:20:52.258414: step: 1900/533, loss: 0.10193166136741638 2023-01-21 09:20:53.445362: step: 1904/533, loss: 0.0545649528503418 2023-01-21 09:20:54.631855: step: 1908/533, loss: 0.09411277621984482 2023-01-21 09:20:55.792735: step: 1912/533, loss: 0.12842321395874023 2023-01-21 09:20:56.979654: step: 1916/533, loss: 0.07013235241174698 2023-01-21 09:20:58.192914: step: 1920/533, loss: 0.10923346877098083 2023-01-21 09:20:59.407839: step: 1924/533, loss: 0.5884883999824524 2023-01-21 09:21:00.569676: step: 1928/533, loss: 0.2668037414550781 2023-01-21 09:21:01.736990: step: 1932/533, loss: 0.045733071863651276 2023-01-21 09:21:02.898830: step: 1936/533, loss: 0.07686290889978409 2023-01-21 09:21:04.066284: step: 1940/533, loss: 0.1623860001564026 2023-01-21 09:21:05.230800: step: 1944/533, loss: 4.371690273284912 2023-01-21 09:21:06.397847: step: 1948/533, loss: 0.049350835382938385 2023-01-21 09:21:07.561185: step: 1952/533, loss: 0.30196380615234375 2023-01-21 09:21:08.780707: step: 1956/533, loss: 0.6507329344749451 2023-01-21 09:21:09.927769: step: 1960/533, loss: 0.09905920177698135 2023-01-21 09:21:11.098540: step: 1964/533, loss: 0.047605134546756744 2023-01-21 09:21:12.262427: step: 1968/533, loss: 0.18005454540252686 2023-01-21 09:21:13.401271: step: 1972/533, loss: 0.32386618852615356 2023-01-21 09:21:14.607534: step: 1976/533, loss: 2.035412311553955 2023-01-21 09:21:15.812898: step: 1980/533, loss: 0.23445968329906464 2023-01-21 09:21:16.984682: step: 1984/533, loss: 0.06588487327098846 2023-01-21 09:21:18.145149: step: 1988/533, loss: 0.29961585998535156 2023-01-21 09:21:19.311908: step: 1992/533, loss: 0.41132137179374695 2023-01-21 09:21:20.508485: step: 1996/533, loss: 0.15920743346214294 2023-01-21 09:21:21.667426: step: 2000/533, loss: 1.2604306936264038 2023-01-21 09:21:22.859278: step: 2004/533, loss: 0.7499514222145081 2023-01-21 09:21:24.048253: step: 2008/533, loss: 0.16321706771850586 2023-01-21 09:21:25.196454: step: 2012/533, loss: 0.39645081758499146 2023-01-21 09:21:26.396492: step: 2016/533, loss: 0.1803477257490158 2023-01-21 09:21:27.547509: step: 2020/533, loss: 0.13040466606616974 2023-01-21 09:21:28.722032: step: 2024/533, loss: 0.14299148321151733 2023-01-21 09:21:29.898212: step: 2028/533, loss: 0.2886943817138672 2023-01-21 09:21:31.098002: step: 2032/533, loss: 0.05794897302985191 2023-01-21 09:21:32.271221: step: 2036/533, loss: 0.06020651012659073 2023-01-21 09:21:33.458126: step: 2040/533, loss: 0.06842012703418732 2023-01-21 09:21:34.637713: step: 2044/533, loss: 0.05030565336346626 2023-01-21 09:21:35.870490: step: 2048/533, loss: 0.28099098801612854 2023-01-21 09:21:37.038441: step: 2052/533, loss: 0.051462698727846146 2023-01-21 09:21:38.201337: step: 2056/533, loss: 0.44959595799446106 2023-01-21 09:21:39.392673: step: 2060/533, loss: 0.2987767457962036 2023-01-21 09:21:40.563040: step: 2064/533, loss: 0.4543399214744568 2023-01-21 09:21:41.726171: step: 2068/533, loss: 0.5146217346191406 2023-01-21 09:21:42.916682: step: 2072/533, loss: 5.33129358291626 2023-01-21 09:21:44.065457: step: 2076/533, loss: 0.06337728351354599 2023-01-21 09:21:45.264043: step: 2080/533, loss: 0.1781458854675293 2023-01-21 09:21:46.432710: step: 2084/533, loss: 0.15403088927268982 2023-01-21 09:21:47.610292: step: 2088/533, loss: 0.10929388552904129 2023-01-21 09:21:48.838056: step: 2092/533, loss: 0.13719424605369568 2023-01-21 09:21:49.991678: step: 2096/533, loss: 0.03629489243030548 2023-01-21 09:21:51.153056: step: 2100/533, loss: 0.3804849684238434 2023-01-21 09:21:52.297822: step: 2104/533, loss: 0.4966081380844116 2023-01-21 09:21:53.444638: step: 2108/533, loss: 0.04897712171077728 2023-01-21 09:21:54.654848: step: 2112/533, loss: 0.4192419946193695 2023-01-21 09:21:55.882506: step: 2116/533, loss: 0.11249351501464844 2023-01-21 09:21:57.072562: step: 2120/533, loss: 1.5381256341934204 2023-01-21 09:21:58.239160: step: 2124/533, loss: 0.3666750490665436 2023-01-21 09:21:59.443276: step: 2128/533, loss: 0.17426948249340057 2023-01-21 09:22:00.619397: step: 2132/533, loss: 0.09110651165246964 ================================================== Loss: 0.443 -------------------- Dev: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Test: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Chinese: {'event': {'p': 0.5517241379310345, 'r': 0.8888888888888888, 'f1': 0.6808510638297872}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample 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': 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 09:22:56.987298: step: 4/533, loss: 0.2078016698360443 2023-01-21 09:22:58.151571: step: 8/533, loss: 0.31518229842185974 2023-01-21 09:22:59.313312: step: 12/533, loss: 0.026435114443302155 2023-01-21 09:23:00.500980: step: 16/533, loss: 0.17656183242797852 2023-01-21 09:23:01.673942: step: 20/533, loss: 0.11464977264404297 2023-01-21 09:23:02.878400: step: 24/533, loss: 0.0932437926530838 2023-01-21 09:23:04.073489: step: 28/533, loss: 0.16593581438064575 2023-01-21 09:23:05.245680: step: 32/533, loss: 0.3303230404853821 2023-01-21 09:23:06.434633: step: 36/533, loss: 0.09351110458374023 2023-01-21 09:23:07.613014: step: 40/533, loss: 0.4381864070892334 2023-01-21 09:23:08.802622: step: 44/533, loss: 0.5056236386299133 2023-01-21 09:23:09.940780: step: 48/533, loss: 0.053463414311409 2023-01-21 09:23:11.109176: step: 52/533, loss: 0.4643387794494629 2023-01-21 09:23:12.258845: step: 56/533, loss: 0.05888638645410538 2023-01-21 09:23:13.432706: step: 60/533, loss: 0.09870719909667969 2023-01-21 09:23:14.585418: step: 64/533, loss: 0.050325583666563034 2023-01-21 09:23:15.711744: step: 68/533, loss: 0.4720114469528198 2023-01-21 09:23:16.904828: step: 72/533, loss: 0.11935101449489594 2023-01-21 09:23:18.076691: step: 76/533, loss: 0.1484021246433258 2023-01-21 09:23:19.256095: step: 80/533, loss: 0.6812185049057007 2023-01-21 09:23:20.443564: step: 84/533, loss: 0.11881309002637863 2023-01-21 09:23:21.636281: step: 88/533, loss: 0.13198252022266388 2023-01-21 09:23:22.820888: step: 92/533, loss: 0.11519059538841248 2023-01-21 09:23:23.972387: step: 96/533, loss: 0.13022871315479279 2023-01-21 09:23:25.136399: step: 100/533, loss: 0.075426384806633 2023-01-21 09:23:26.311470: step: 104/533, loss: 0.6256099939346313 2023-01-21 09:23:27.519481: step: 108/533, loss: 0.5357329845428467 2023-01-21 09:23:28.699151: step: 112/533, loss: 0.25639086961746216 2023-01-21 09:23:29.860740: step: 116/533, loss: 0.2864781320095062 2023-01-21 09:23:31.121655: step: 120/533, loss: 0.11975307762622833 2023-01-21 09:23:32.297032: step: 124/533, loss: 0.13876304030418396 2023-01-21 09:23:33.487478: step: 128/533, loss: 0.0528528206050396 2023-01-21 09:23:34.665006: step: 132/533, loss: 0.07103414833545685 2023-01-21 09:23:35.844458: step: 136/533, loss: 0.1213715523481369 2023-01-21 09:23:36.972042: step: 140/533, loss: 0.2966233193874359 2023-01-21 09:23:38.154213: step: 144/533, loss: 0.33138924837112427 2023-01-21 09:23:39.301878: step: 148/533, loss: 0.8489260673522949 2023-01-21 09:23:40.494311: step: 152/533, loss: 0.12019577622413635 2023-01-21 09:23:41.650149: step: 156/533, loss: 0.14811936020851135 2023-01-21 09:23:42.798484: step: 160/533, loss: 0.03802354261279106 2023-01-21 09:23:43.981678: step: 164/533, loss: 0.07348743081092834 2023-01-21 09:23:45.157311: step: 168/533, loss: 0.7071317434310913 2023-01-21 09:23:46.347113: step: 172/533, loss: 0.5400540828704834 2023-01-21 09:23:47.509933: step: 176/533, loss: 0.329581618309021 2023-01-21 09:23:48.682867: step: 180/533, loss: 0.5030377507209778 2023-01-21 09:23:49.847314: step: 184/533, loss: 0.2839488387107849 2023-01-21 09:23:51.025423: step: 188/533, loss: 0.15923671424388885 2023-01-21 09:23:52.239065: step: 192/533, loss: 0.219902902841568 2023-01-21 09:23:53.423644: step: 196/533, loss: 2.0716328620910645 2023-01-21 09:23:54.641789: step: 200/533, loss: 0.14099450409412384 2023-01-21 09:23:55.813615: step: 204/533, loss: 0.11872296780347824 2023-01-21 09:23:56.966012: step: 208/533, loss: 0.02985839918255806 2023-01-21 09:23:58.145650: step: 212/533, loss: 0.19700823724269867 2023-01-21 09:23:59.336091: step: 216/533, loss: 0.10918435454368591 2023-01-21 09:24:00.482603: step: 220/533, loss: 0.22313785552978516 2023-01-21 09:24:01.658726: step: 224/533, loss: 0.09669509530067444 2023-01-21 09:24:02.814792: step: 228/533, loss: 0.17720898985862732 2023-01-21 09:24:03.988535: step: 232/533, loss: 1.2134418487548828 2023-01-21 09:24:05.158405: step: 236/533, loss: 0.3068639636039734 2023-01-21 09:24:06.368026: step: 240/533, loss: 0.3462057411670685 2023-01-21 09:24:07.577885: step: 244/533, loss: 0.1848122626543045 2023-01-21 09:24:08.784645: step: 248/533, loss: 1.5631258487701416 2023-01-21 09:24:09.927323: step: 252/533, loss: 0.13114280998706818 2023-01-21 09:24:11.098325: step: 256/533, loss: 0.11863727867603302 2023-01-21 09:24:12.296191: step: 260/533, loss: 0.07429871708154678 2023-01-21 09:24:13.435092: step: 264/533, loss: 0.18646402657032013 2023-01-21 09:24:14.647984: step: 268/533, loss: 0.27133816480636597 2023-01-21 09:24:15.844164: step: 272/533, loss: 0.1807132363319397 2023-01-21 09:24:16.989862: step: 276/533, loss: 0.5122114419937134 2023-01-21 09:24:18.178960: step: 280/533, loss: 0.07589464634656906 2023-01-21 09:24:19.374994: step: 284/533, loss: 0.1204366683959961 2023-01-21 09:24:20.558167: step: 288/533, loss: 0.2831389605998993 2023-01-21 09:24:21.774041: step: 292/533, loss: 0.5479009747505188 2023-01-21 09:24:22.969007: step: 296/533, loss: 0.1523316204547882 2023-01-21 09:24:24.133391: step: 300/533, loss: 0.08619347214698792 2023-01-21 09:24:25.288920: step: 304/533, loss: 0.22536487877368927 2023-01-21 09:24:26.500648: step: 308/533, loss: 0.38911542296409607 2023-01-21 09:24:27.691557: step: 312/533, loss: 0.04329654946923256 2023-01-21 09:24:28.887680: step: 316/533, loss: 0.45962390303611755 2023-01-21 09:24:30.058120: step: 320/533, loss: 0.04925408214330673 2023-01-21 09:24:31.209804: step: 324/533, loss: 0.24926291406154633 2023-01-21 09:24:32.385740: step: 328/533, loss: 0.4987621009349823 2023-01-21 09:24:33.573909: step: 332/533, loss: 0.006787109188735485 2023-01-21 09:24:34.784845: step: 336/533, loss: 0.24444876611232758 2023-01-21 09:24:35.995543: step: 340/533, loss: 0.29878556728363037 2023-01-21 09:24:37.224115: step: 344/533, loss: 0.04171190410852432 2023-01-21 09:24:38.429247: step: 348/533, loss: 0.8884342908859253 2023-01-21 09:24:39.667554: step: 352/533, loss: 0.15009284019470215 2023-01-21 09:24:40.817403: step: 356/533, loss: 0.022879624739289284 2023-01-21 09:24:42.039811: step: 360/533, loss: 0.4236254394054413 2023-01-21 09:24:43.226967: step: 364/533, loss: 0.710014283657074 2023-01-21 09:24:44.407494: step: 368/533, loss: 0.5040040612220764 2023-01-21 09:24:45.619379: step: 372/533, loss: 0.1399248093366623 2023-01-21 09:24:46.819990: step: 376/533, loss: 0.07454462349414825 2023-01-21 09:24:47.988896: step: 380/533, loss: 0.10650920867919922 2023-01-21 09:24:49.179733: step: 384/533, loss: 0.12517786026000977 2023-01-21 09:24:50.335922: step: 388/533, loss: 0.1347772628068924 2023-01-21 09:24:51.541438: step: 392/533, loss: 0.3723783493041992 2023-01-21 09:24:52.747305: step: 396/533, loss: 0.2512499988079071 2023-01-21 09:24:53.876747: step: 400/533, loss: 0.15802955627441406 2023-01-21 09:24:55.111546: step: 404/533, loss: 0.24489697813987732 2023-01-21 09:24:56.262223: step: 408/533, loss: 0.18426179885864258 2023-01-21 09:24:57.481429: step: 412/533, loss: 0.33918437361717224 2023-01-21 09:24:58.667917: step: 416/533, loss: 0.03830733522772789 2023-01-21 09:24:59.873279: step: 420/533, loss: 0.07256555557250977 2023-01-21 09:25:01.034438: step: 424/533, loss: 0.23671989142894745 2023-01-21 09:25:02.234472: step: 428/533, loss: 0.17922210693359375 2023-01-21 09:25:03.440661: step: 432/533, loss: 0.19453468918800354 2023-01-21 09:25:04.584616: step: 436/533, loss: 0.02808399125933647 2023-01-21 09:25:05.801820: step: 440/533, loss: 0.158246710896492 2023-01-21 09:25:06.978675: step: 444/533, loss: 0.7060526013374329 2023-01-21 09:25:08.106696: step: 448/533, loss: 0.07113365828990936 2023-01-21 09:25:09.277396: step: 452/533, loss: 0.13560420274734497 2023-01-21 09:25:10.445463: step: 456/533, loss: 0.39642783999443054 2023-01-21 09:25:11.598546: step: 460/533, loss: 0.21954207122325897 2023-01-21 09:25:12.787842: step: 464/533, loss: 0.48731398582458496 2023-01-21 09:25:13.971071: step: 468/533, loss: 0.4911355972290039 2023-01-21 09:25:15.115059: step: 472/533, loss: 0.1939440220594406 2023-01-21 09:25:16.309949: step: 476/533, loss: 0.1469784677028656 2023-01-21 09:25:17.454468: step: 480/533, loss: 0.12989595532417297 2023-01-21 09:25:18.619546: step: 484/533, loss: 0.8503895998001099 2023-01-21 09:25:19.802035: step: 488/533, loss: 0.5811647176742554 2023-01-21 09:25:20.977587: step: 492/533, loss: 0.16675075888633728 2023-01-21 09:25:22.201881: step: 496/533, loss: 0.11213760823011398 2023-01-21 09:25:23.356788: step: 500/533, loss: 0.07026652991771698 2023-01-21 09:25:24.560373: step: 504/533, loss: 0.2696276605129242 2023-01-21 09:25:25.707864: step: 508/533, loss: 1.6246525049209595 2023-01-21 09:25:26.859801: step: 512/533, loss: 0.19958114624023438 2023-01-21 09:25:28.031431: step: 516/533, loss: 0.2736547589302063 2023-01-21 09:25:29.286014: step: 520/533, loss: 0.4516318440437317 2023-01-21 09:25:30.473124: step: 524/533, loss: 0.09942984580993652 2023-01-21 09:25:31.609256: step: 528/533, loss: 0.1929490566253662 2023-01-21 09:25:32.858606: step: 532/533, loss: 0.12010325491428375 2023-01-21 09:25:34.009685: step: 536/533, loss: 0.8290392756462097 2023-01-21 09:25:35.184602: step: 540/533, loss: 0.03588051721453667 2023-01-21 09:25:36.333972: step: 544/533, loss: 0.6883276104927063 2023-01-21 09:25:37.538827: step: 548/533, loss: 1.1545380353927612 2023-01-21 09:25:38.711031: step: 552/533, loss: 0.06733036041259766 2023-01-21 09:25:39.833806: step: 556/533, loss: 0.07664375007152557 2023-01-21 09:25:41.034030: step: 560/533, loss: 0.16379842162132263 2023-01-21 09:25:42.236796: step: 564/533, loss: 0.24413833022117615 2023-01-21 09:25:43.406583: step: 568/533, loss: 0.13898450136184692 2023-01-21 09:25:44.577032: step: 572/533, loss: 0.30169177055358887 2023-01-21 09:25:45.759088: step: 576/533, loss: 0.44153380393981934 2023-01-21 09:25:46.984131: step: 580/533, loss: 0.18375855684280396 2023-01-21 09:25:48.176961: step: 584/533, loss: 1.6990687847137451 2023-01-21 09:25:49.330756: step: 588/533, loss: 3.2265725135803223 2023-01-21 09:25:50.537034: step: 592/533, loss: 0.10718021541833878 2023-01-21 09:25:51.745654: step: 596/533, loss: 0.2942487597465515 2023-01-21 09:25:52.926388: step: 600/533, loss: 0.07089582085609436 2023-01-21 09:25:54.052934: step: 604/533, loss: 0.29861927032470703 2023-01-21 09:25:55.231762: step: 608/533, loss: 0.5086458325386047 2023-01-21 09:25:56.395258: step: 612/533, loss: 0.17752285301685333 2023-01-21 09:25:57.565812: step: 616/533, loss: 0.2619171142578125 2023-01-21 09:25:58.764881: step: 620/533, loss: 0.26793354749679565 2023-01-21 09:25:59.916393: step: 624/533, loss: 0.15851421654224396 2023-01-21 09:26:01.100725: step: 628/533, loss: 0.14425459504127502 2023-01-21 09:26:02.313908: step: 632/533, loss: 0.5166572332382202 2023-01-21 09:26:03.558944: step: 636/533, loss: 0.05711860582232475 2023-01-21 09:26:04.703052: step: 640/533, loss: 0.9694008231163025 2023-01-21 09:26:05.841717: step: 644/533, loss: 0.10738392174243927 2023-01-21 09:26:07.005930: step: 648/533, loss: 0.027312662452459335 2023-01-21 09:26:08.209578: step: 652/533, loss: 0.1380002498626709 2023-01-21 09:26:09.397270: step: 656/533, loss: 0.0690101608633995 2023-01-21 09:26:10.605648: step: 660/533, loss: 0.04692526161670685 2023-01-21 09:26:11.775352: step: 664/533, loss: 0.36581698060035706 2023-01-21 09:26:12.910884: step: 668/533, loss: 0.05739751085639 2023-01-21 09:26:14.080457: step: 672/533, loss: 0.16228342056274414 2023-01-21 09:26:15.243699: step: 676/533, loss: 0.12853717803955078 2023-01-21 09:26:16.429608: step: 680/533, loss: 0.35547828674316406 2023-01-21 09:26:17.638754: step: 684/533, loss: 0.370439350605011 2023-01-21 09:26:18.829000: step: 688/533, loss: 0.04332399368286133 2023-01-21 09:26:19.974177: step: 692/533, loss: 0.17293567955493927 2023-01-21 09:26:21.146243: step: 696/533, loss: 0.349107563495636 2023-01-21 09:26:22.307016: step: 700/533, loss: 0.3088824450969696 2023-01-21 09:26:23.536728: step: 704/533, loss: 0.06935205310583115 2023-01-21 09:26:24.700912: step: 708/533, loss: 0.0868774950504303 2023-01-21 09:26:25.883220: step: 712/533, loss: 0.5752468705177307 2023-01-21 09:26:27.051567: step: 716/533, loss: 0.10252895951271057 2023-01-21 09:26:28.246139: step: 720/533, loss: 0.966886043548584 2023-01-21 09:26:29.440261: step: 724/533, loss: 0.14027100801467896 2023-01-21 09:26:30.608077: step: 728/533, loss: 0.0520419105887413 2023-01-21 09:26:31.783012: step: 732/533, loss: 0.07258539646863937 2023-01-21 09:26:32.945296: step: 736/533, loss: 0.37988173961639404 2023-01-21 09:26:34.118753: step: 740/533, loss: 0.03662538528442383 2023-01-21 09:26:35.312477: step: 744/533, loss: 0.04324188083410263 2023-01-21 09:26:36.506559: step: 748/533, loss: 0.3613608777523041 2023-01-21 09:26:37.657953: step: 752/533, loss: 0.12816905975341797 2023-01-21 09:26:38.864513: step: 756/533, loss: 0.0955532118678093 2023-01-21 09:26:40.103584: step: 760/533, loss: 0.06491170078516006 2023-01-21 09:26:41.295946: step: 764/533, loss: 0.2155565321445465 2023-01-21 09:26:42.528650: step: 768/533, loss: 0.17227935791015625 2023-01-21 09:26:43.692223: step: 772/533, loss: 0.17179450392723083 2023-01-21 09:26:44.870220: step: 776/533, loss: 0.27961111068725586 2023-01-21 09:26:46.056324: step: 780/533, loss: 0.16273269057273865 2023-01-21 09:26:47.221372: step: 784/533, loss: 0.11638174206018448 2023-01-21 09:26:48.430139: step: 788/533, loss: 0.0422457717359066 2023-01-21 09:26:49.620751: step: 792/533, loss: 0.2984282374382019 2023-01-21 09:26:50.819406: step: 796/533, loss: 1.4745967388153076 2023-01-21 09:26:51.993568: step: 800/533, loss: 0.22456084191799164 2023-01-21 09:26:53.174172: step: 804/533, loss: 0.07785816490650177 2023-01-21 09:26:54.390804: step: 808/533, loss: 0.14456290006637573 2023-01-21 09:26:55.569016: step: 812/533, loss: 0.10354909300804138 2023-01-21 09:26:56.717529: step: 816/533, loss: 0.04806175455451012 2023-01-21 09:26:57.879313: step: 820/533, loss: 0.1270865499973297 2023-01-21 09:26:59.054741: step: 824/533, loss: 0.1418808102607727 2023-01-21 09:27:00.213582: step: 828/533, loss: 0.08975763618946075 2023-01-21 09:27:01.356856: step: 832/533, loss: 0.0381811149418354 2023-01-21 09:27:02.533924: step: 836/533, loss: 0.08536271750926971 2023-01-21 09:27:03.701020: step: 840/533, loss: 1.3843010663986206 2023-01-21 09:27:04.917884: step: 844/533, loss: 0.163020521402359 2023-01-21 09:27:06.087530: step: 848/533, loss: 0.04792146757245064 2023-01-21 09:27:07.248598: step: 852/533, loss: 0.08335190266370773 2023-01-21 09:27:08.424793: step: 856/533, loss: 0.14321881532669067 2023-01-21 09:27:09.578985: step: 860/533, loss: 0.11247396469116211 2023-01-21 09:27:10.794058: step: 864/533, loss: 0.08266621083021164 2023-01-21 09:27:11.962520: step: 868/533, loss: 0.19085007905960083 2023-01-21 09:27:13.107303: step: 872/533, loss: 0.18547268211841583 2023-01-21 09:27:14.292940: step: 876/533, loss: 0.2471088171005249 2023-01-21 09:27:15.461427: step: 880/533, loss: 0.24844466149806976 2023-01-21 09:27:16.681293: step: 884/533, loss: 0.6363113522529602 2023-01-21 09:27:17.844161: step: 888/533, loss: 0.21207237243652344 2023-01-21 09:27:19.046488: step: 892/533, loss: 0.50953608751297 2023-01-21 09:27:20.233108: step: 896/533, loss: 0.08926382660865784 2023-01-21 09:27:21.434165: step: 900/533, loss: 0.038840580731630325 2023-01-21 09:27:22.606901: step: 904/533, loss: 0.3272953927516937 2023-01-21 09:27:23.758479: step: 908/533, loss: 0.09617023915052414 2023-01-21 09:27:24.893426: step: 912/533, loss: 0.2607877254486084 2023-01-21 09:27:26.043961: step: 916/533, loss: 0.0365811362862587 2023-01-21 09:27:27.210061: step: 920/533, loss: 0.09311385452747345 2023-01-21 09:27:28.435196: step: 924/533, loss: 0.24119792878627777 2023-01-21 09:27:29.635193: step: 928/533, loss: 0.3606806993484497 2023-01-21 09:27:30.796256: step: 932/533, loss: 0.656897246837616 2023-01-21 09:27:31.983744: step: 936/533, loss: 0.0672733336687088 2023-01-21 09:27:33.136958: step: 940/533, loss: 0.05885028839111328 2023-01-21 09:27:34.311985: step: 944/533, loss: 0.03974022716283798 2023-01-21 09:27:35.489387: step: 948/533, loss: 0.24166402220726013 2023-01-21 09:27:36.655323: step: 952/533, loss: 0.36412110924720764 2023-01-21 09:27:37.892906: step: 956/533, loss: 0.2218734323978424 2023-01-21 09:27:39.055625: step: 960/533, loss: 0.019949007779359818 2023-01-21 09:27:40.241025: step: 964/533, loss: 0.2599373757839203 2023-01-21 09:27:41.429965: step: 968/533, loss: 0.21649064123630524 2023-01-21 09:27:42.613002: step: 972/533, loss: 0.6959112882614136 2023-01-21 09:27:43.860285: step: 976/533, loss: 1.9110101461410522 2023-01-21 09:27:45.028851: step: 980/533, loss: 0.060066986829042435 2023-01-21 09:27:46.247978: step: 984/533, loss: 0.10456404834985733 2023-01-21 09:27:47.440587: step: 988/533, loss: 0.09126629680395126 2023-01-21 09:27:48.621088: step: 992/533, loss: 0.8815129399299622 2023-01-21 09:27:49.794809: step: 996/533, loss: 0.18886065483093262 2023-01-21 09:27:51.009691: step: 1000/533, loss: 0.30123645067214966 2023-01-21 09:27:52.189391: step: 1004/533, loss: 0.0677306205034256 2023-01-21 09:27:53.374816: step: 1008/533, loss: 0.24569815397262573 2023-01-21 09:27:54.555961: step: 1012/533, loss: 0.04990663379430771 2023-01-21 09:27:55.741041: step: 1016/533, loss: 0.06525164097547531 2023-01-21 09:27:56.929257: step: 1020/533, loss: 0.525295615196228 2023-01-21 09:27:58.064832: step: 1024/533, loss: 0.02988424338400364 2023-01-21 09:27:59.278656: step: 1028/533, loss: 0.021146394312381744 2023-01-21 09:28:00.441345: step: 1032/533, loss: 0.23003879189491272 2023-01-21 09:28:01.648657: step: 1036/533, loss: 0.2026832103729248 2023-01-21 09:28:02.822028: step: 1040/533, loss: 0.0364932045340538 2023-01-21 09:28:04.005836: step: 1044/533, loss: 0.30065423250198364 2023-01-21 09:28:05.157508: step: 1048/533, loss: 0.035550735890865326 2023-01-21 09:28:06.327732: step: 1052/533, loss: 0.06179413944482803 2023-01-21 09:28:07.522225: step: 1056/533, loss: 0.04478273540735245 2023-01-21 09:28:08.715656: step: 1060/533, loss: 0.5006459951400757 2023-01-21 09:28:09.893645: step: 1064/533, loss: 0.2556353211402893 2023-01-21 09:28:11.096299: step: 1068/533, loss: 0.22370100021362305 2023-01-21 09:28:12.294589: step: 1072/533, loss: 0.10955343395471573 2023-01-21 09:28:13.467895: step: 1076/533, loss: 0.2563779950141907 2023-01-21 09:28:14.657108: step: 1080/533, loss: 0.16635161638259888 2023-01-21 09:28:15.870575: step: 1084/533, loss: 0.08671098202466965 2023-01-21 09:28:17.049216: step: 1088/533, loss: 0.06804871559143066 2023-01-21 09:28:18.214799: step: 1092/533, loss: 0.5283897519111633 2023-01-21 09:28:19.398486: step: 1096/533, loss: 0.04945874214172363 2023-01-21 09:28:20.570376: step: 1100/533, loss: 0.10594506561756134 2023-01-21 09:28:21.819104: step: 1104/533, loss: 0.057581618428230286 2023-01-21 09:28:22.972986: step: 1108/533, loss: 0.3890157639980316 2023-01-21 09:28:24.176071: step: 1112/533, loss: 0.4649909436702728 2023-01-21 09:28:25.351771: step: 1116/533, loss: 0.43836021423339844 2023-01-21 09:28:26.544531: step: 1120/533, loss: 0.07939185947179794 2023-01-21 09:28:27.746887: step: 1124/533, loss: 0.11801280826330185 2023-01-21 09:28:28.954417: step: 1128/533, loss: 0.04644632339477539 2023-01-21 09:28:30.128095: step: 1132/533, loss: 0.27586880326271057 2023-01-21 09:28:31.321377: step: 1136/533, loss: 0.5773674845695496 2023-01-21 09:28:32.481341: step: 1140/533, loss: 0.09778346866369247 2023-01-21 09:28:33.682016: step: 1144/533, loss: 0.1590556651353836 2023-01-21 09:28:34.843055: step: 1148/533, loss: 0.19908609986305237 2023-01-21 09:28:36.040702: step: 1152/533, loss: 0.04899187386035919 2023-01-21 09:28:37.235546: step: 1156/533, loss: 0.41234758496284485 2023-01-21 09:28:38.391890: step: 1160/533, loss: 0.1556684970855713 2023-01-21 09:28:39.537647: step: 1164/533, loss: 1.1117031574249268 2023-01-21 09:28:40.713143: step: 1168/533, loss: 0.19716626405715942 2023-01-21 09:28:41.942591: step: 1172/533, loss: 0.0710611343383789 2023-01-21 09:28:43.136228: step: 1176/533, loss: 0.08309850841760635 2023-01-21 09:28:44.337071: step: 1180/533, loss: 0.32324203848838806 2023-01-21 09:28:45.546054: step: 1184/533, loss: 0.8226678967475891 2023-01-21 09:28:46.773019: step: 1188/533, loss: 0.4102906286716461 2023-01-21 09:28:47.946949: step: 1192/533, loss: 1.0576732158660889 2023-01-21 09:28:49.091389: step: 1196/533, loss: 0.14293083548545837 2023-01-21 09:28:50.275893: step: 1200/533, loss: 0.37836790084838867 2023-01-21 09:28:51.451136: step: 1204/533, loss: 0.1536066085100174 2023-01-21 09:28:52.593468: step: 1208/533, loss: 0.028438378125429153 2023-01-21 09:28:53.768559: step: 1212/533, loss: 0.331953763961792 2023-01-21 09:28:54.940465: step: 1216/533, loss: 0.14810489118099213 2023-01-21 09:28:56.131794: step: 1220/533, loss: 0.5138829946517944 2023-01-21 09:28:57.310067: step: 1224/533, loss: 0.1982186883687973 2023-01-21 09:28:58.475412: step: 1228/533, loss: 0.09420661628246307 2023-01-21 09:28:59.627928: step: 1232/533, loss: 0.24847650527954102 2023-01-21 09:29:00.799170: step: 1236/533, loss: 0.12884683907032013 2023-01-21 09:29:01.995421: step: 1240/533, loss: 0.18810588121414185 2023-01-21 09:29:03.172941: step: 1244/533, loss: 0.3753485679626465 2023-01-21 09:29:04.323002: step: 1248/533, loss: 0.9338828921318054 2023-01-21 09:29:05.497129: step: 1252/533, loss: 0.22380857169628143 2023-01-21 09:29:06.687144: step: 1256/533, loss: 0.6245174407958984 2023-01-21 09:29:07.875671: step: 1260/533, loss: 0.14597196877002716 2023-01-21 09:29:09.086601: step: 1264/533, loss: 0.36217156052589417 2023-01-21 09:29:10.258942: step: 1268/533, loss: 0.0613522082567215 2023-01-21 09:29:11.450976: step: 1272/533, loss: 0.09672181308269501 2023-01-21 09:29:12.638492: step: 1276/533, loss: 0.13940119743347168 2023-01-21 09:29:13.815593: step: 1280/533, loss: 0.09196672588586807 2023-01-21 09:29:14.990576: step: 1284/533, loss: 0.16214370727539062 2023-01-21 09:29:16.171322: step: 1288/533, loss: 0.18966837227344513 2023-01-21 09:29:17.376652: step: 1292/533, loss: 0.12006721645593643 2023-01-21 09:29:18.583109: step: 1296/533, loss: 0.0970979705452919 2023-01-21 09:29:19.762743: step: 1300/533, loss: 0.1135433167219162 2023-01-21 09:29:20.916784: step: 1304/533, loss: 0.029100514948368073 2023-01-21 09:29:22.112413: step: 1308/533, loss: 0.14718112349510193 2023-01-21 09:29:23.303399: step: 1312/533, loss: 1.097081184387207 2023-01-21 09:29:24.500124: step: 1316/533, loss: 0.40221983194351196 2023-01-21 09:29:25.704094: step: 1320/533, loss: 0.04940800741314888 2023-01-21 09:29:26.907443: step: 1324/533, loss: 0.6038603782653809 2023-01-21 09:29:28.058794: step: 1328/533, loss: 0.7928571701049805 2023-01-21 09:29:29.216139: step: 1332/533, loss: 0.10265494138002396 2023-01-21 09:29:30.368236: step: 1336/533, loss: 0.21025848388671875 2023-01-21 09:29:31.542773: step: 1340/533, loss: 0.41907915472984314 2023-01-21 09:29:32.725231: step: 1344/533, loss: 0.12884274125099182 2023-01-21 09:29:33.873048: step: 1348/533, loss: 0.17052879929542542 2023-01-21 09:29:35.092258: step: 1352/533, loss: 0.0990772470831871 2023-01-21 09:29:36.276604: step: 1356/533, loss: 0.06579726934432983 2023-01-21 09:29:37.462974: step: 1360/533, loss: 0.03973455727100372 2023-01-21 09:29:38.656215: step: 1364/533, loss: 0.09333725273609161 2023-01-21 09:29:39.829213: step: 1368/533, loss: 0.17254649102687836 2023-01-21 09:29:41.001315: step: 1372/533, loss: 0.2807408571243286 2023-01-21 09:29:42.149570: step: 1376/533, loss: 0.19102182984352112 2023-01-21 09:29:43.311141: step: 1380/533, loss: 0.9477138519287109 2023-01-21 09:29:44.547235: step: 1384/533, loss: 0.16673335433006287 2023-01-21 09:29:45.699561: step: 1388/533, loss: 0.5129029750823975 2023-01-21 09:29:46.903323: step: 1392/533, loss: 0.4011189043521881 2023-01-21 09:29:48.047908: step: 1396/533, loss: 0.05289926752448082 2023-01-21 09:29:49.282030: step: 1400/533, loss: 0.25938501954078674 2023-01-21 09:29:50.455955: step: 1404/533, loss: 0.13770073652267456 2023-01-21 09:29:51.663674: step: 1408/533, loss: 0.06882219016551971 2023-01-21 09:29:52.858004: step: 1412/533, loss: 0.38377562165260315 2023-01-21 09:29:54.031286: step: 1416/533, loss: 0.06996621936559677 2023-01-21 09:29:55.242273: step: 1420/533, loss: 0.14846988022327423 2023-01-21 09:29:56.415754: step: 1424/533, loss: 0.08452146500349045 2023-01-21 09:29:57.575761: step: 1428/533, loss: 0.6743655204772949 2023-01-21 09:29:58.739627: step: 1432/533, loss: 0.06838846206665039 2023-01-21 09:29:59.905673: step: 1436/533, loss: 0.691069483757019 2023-01-21 09:30:01.112543: step: 1440/533, loss: 0.09301143139600754 2023-01-21 09:30:02.317796: step: 1444/533, loss: 0.1438518613576889 2023-01-21 09:30:03.491022: step: 1448/533, loss: 0.04686949402093887 2023-01-21 09:30:04.699973: step: 1452/533, loss: 0.05652890354394913 2023-01-21 09:30:05.907857: step: 1456/533, loss: 0.10818997025489807 2023-01-21 09:30:07.074358: step: 1460/533, loss: 0.014242172241210938 2023-01-21 09:30:08.246057: step: 1464/533, loss: 0.043192245066165924 2023-01-21 09:30:09.409737: step: 1468/533, loss: 0.5309242010116577 2023-01-21 09:30:10.592096: step: 1472/533, loss: 0.457075297832489 2023-01-21 09:30:11.770477: step: 1476/533, loss: 0.1744905561208725 2023-01-21 09:30:12.956948: step: 1480/533, loss: 0.6160749793052673 2023-01-21 09:30:14.121378: step: 1484/533, loss: 0.24872151017189026 2023-01-21 09:30:15.312126: step: 1488/533, loss: 0.024968530982732773 2023-01-21 09:30:16.486499: step: 1492/533, loss: 0.14192934334278107 2023-01-21 09:30:17.710034: step: 1496/533, loss: 0.13072550296783447 2023-01-21 09:30:18.882984: step: 1500/533, loss: 1.138831615447998 2023-01-21 09:30:20.054286: step: 1504/533, loss: 0.2735590636730194 2023-01-21 09:30:21.203352: step: 1508/533, loss: 3.957540512084961 2023-01-21 09:30:22.354845: step: 1512/533, loss: 0.5816661715507507 2023-01-21 09:30:23.572068: step: 1516/533, loss: 0.023665238171815872 2023-01-21 09:30:24.732017: step: 1520/533, loss: 0.17918796837329865 2023-01-21 09:30:25.921874: step: 1524/533, loss: 0.3125152587890625 2023-01-21 09:30:27.170605: step: 1528/533, loss: 0.1004018783569336 2023-01-21 09:30:28.368742: step: 1532/533, loss: 0.3702074885368347 2023-01-21 09:30:29.555195: step: 1536/533, loss: 0.33720093965530396 2023-01-21 09:30:30.728195: step: 1540/533, loss: 0.05317385122179985 2023-01-21 09:30:31.887696: step: 1544/533, loss: 0.0034165619872510433 2023-01-21 09:30:33.039521: step: 1548/533, loss: 0.2887263298034668 2023-01-21 09:30:34.247542: step: 1552/533, loss: 0.057515766471624374 2023-01-21 09:30:35.433728: step: 1556/533, loss: 0.08571477234363556 2023-01-21 09:30:36.612805: step: 1560/533, loss: 0.2643674910068512 2023-01-21 09:30:37.824636: step: 1564/533, loss: 0.17403718829154968 2023-01-21 09:30:38.978442: step: 1568/533, loss: 0.1666909158229828 2023-01-21 09:30:40.137844: step: 1572/533, loss: 0.032784365117549896 2023-01-21 09:30:41.336241: step: 1576/533, loss: 0.2608538269996643 2023-01-21 09:30:42.510853: step: 1580/533, loss: 0.12190522998571396 2023-01-21 09:30:43.701269: step: 1584/533, loss: 0.029227543622255325 2023-01-21 09:30:44.872469: step: 1588/533, loss: 1.949853539466858 2023-01-21 09:30:46.030027: step: 1592/533, loss: 0.7524545788764954 2023-01-21 09:30:47.230524: step: 1596/533, loss: 0.25709474086761475 2023-01-21 09:30:48.365027: step: 1600/533, loss: 0.03538703918457031 2023-01-21 09:30:49.558091: step: 1604/533, loss: 1.2393126487731934 2023-01-21 09:30:50.706130: step: 1608/533, loss: 0.061524104326963425 2023-01-21 09:30:51.872833: step: 1612/533, loss: 0.05912356823682785 2023-01-21 09:30:53.036647: step: 1616/533, loss: 0.525540828704834 2023-01-21 09:30:54.212567: step: 1620/533, loss: 0.3413012623786926 2023-01-21 09:30:55.431728: step: 1624/533, loss: 0.14334268867969513 2023-01-21 09:30:56.660293: step: 1628/533, loss: 0.18554410338401794 2023-01-21 09:30:57.873077: step: 1632/533, loss: 0.08620801568031311 2023-01-21 09:30:59.043201: step: 1636/533, loss: 0.2511632442474365 2023-01-21 09:31:00.230659: step: 1640/533, loss: 0.21081668138504028 2023-01-21 09:31:01.423759: step: 1644/533, loss: 0.4084090292453766 2023-01-21 09:31:02.607827: step: 1648/533, loss: 0.04076967388391495 2023-01-21 09:31:03.791118: step: 1652/533, loss: 0.057720281183719635 2023-01-21 09:31:04.975710: step: 1656/533, loss: 0.07515545189380646 2023-01-21 09:31:06.163885: step: 1660/533, loss: 0.7464092373847961 2023-01-21 09:31:07.351159: step: 1664/533, loss: 0.06995511054992676 2023-01-21 09:31:08.585129: step: 1668/533, loss: 0.8005859851837158 2023-01-21 09:31:09.777330: step: 1672/533, loss: 0.05395703762769699 2023-01-21 09:31:10.963605: step: 1676/533, loss: 0.13379746675491333 2023-01-21 09:31:12.164507: step: 1680/533, loss: 0.205827996134758 2023-01-21 09:31:13.341519: step: 1684/533, loss: 0.05636405944824219 2023-01-21 09:31:14.482402: step: 1688/533, loss: 0.2690616846084595 2023-01-21 09:31:15.662377: step: 1692/533, loss: 0.4825805723667145 2023-01-21 09:31:16.846299: step: 1696/533, loss: 0.1953682005405426 2023-01-21 09:31:18.006427: step: 1700/533, loss: 0.029216576367616653 2023-01-21 09:31:19.232501: step: 1704/533, loss: 0.15461483597755432 2023-01-21 09:31:20.443049: step: 1708/533, loss: 0.7828990817070007 2023-01-21 09:31:21.627389: step: 1712/533, loss: 0.10142626613378525 2023-01-21 09:31:22.849950: step: 1716/533, loss: 0.05300397798418999 2023-01-21 09:31:23.992518: step: 1720/533, loss: 0.6581972241401672 2023-01-21 09:31:25.201710: step: 1724/533, loss: 0.09413953125476837 2023-01-21 09:31:26.370683: step: 1728/533, loss: 0.07101168483495712 2023-01-21 09:31:27.557889: step: 1732/533, loss: 0.2001507729291916 2023-01-21 09:31:28.741245: step: 1736/533, loss: 0.24617183208465576 2023-01-21 09:31:29.914201: step: 1740/533, loss: 0.09033975750207901 2023-01-21 09:31:31.048814: step: 1744/533, loss: 0.20538422465324402 2023-01-21 09:31:32.243006: step: 1748/533, loss: 0.1850377917289734 2023-01-21 09:31:33.506792: step: 1752/533, loss: 0.25849050283432007 2023-01-21 09:31:34.670767: step: 1756/533, loss: 0.22616156935691833 2023-01-21 09:31:35.828007: step: 1760/533, loss: 0.11476631462574005 2023-01-21 09:31:37.052903: step: 1764/533, loss: 0.1577913761138916 2023-01-21 09:31:38.227021: step: 1768/533, loss: 0.14762765169143677 2023-01-21 09:31:39.415986: step: 1772/533, loss: 0.040448904037475586 2023-01-21 09:31:40.637655: step: 1776/533, loss: 0.12067694962024689 2023-01-21 09:31:41.809876: step: 1780/533, loss: 0.0686771422624588 2023-01-21 09:31:43.019247: step: 1784/533, loss: 0.11080141365528107 2023-01-21 09:31:44.257498: step: 1788/533, loss: 0.05756950378417969 2023-01-21 09:31:45.468208: step: 1792/533, loss: 0.4624618887901306 2023-01-21 09:31:46.690440: step: 1796/533, loss: 0.12290406227111816 2023-01-21 09:31:47.883999: step: 1800/533, loss: 0.16624422371387482 2023-01-21 09:31:49.039736: step: 1804/533, loss: 0.2789624333381653 2023-01-21 09:31:50.199937: step: 1808/533, loss: 0.0747896283864975 2023-01-21 09:31:51.409880: step: 1812/533, loss: 0.3055858016014099 2023-01-21 09:31:52.612618: step: 1816/533, loss: 0.26524078845977783 2023-01-21 09:31:53.762404: step: 1820/533, loss: 0.06191863864660263 2023-01-21 09:31:54.960706: step: 1824/533, loss: 0.12996242940425873 2023-01-21 09:31:56.130361: step: 1828/533, loss: 0.34519368410110474 2023-01-21 09:31:57.318290: step: 1832/533, loss: 0.42146873474121094 2023-01-21 09:31:58.488764: step: 1836/533, loss: 0.03204717859625816 2023-01-21 09:31:59.704396: step: 1840/533, loss: 0.06813183426856995 2023-01-21 09:32:00.904375: step: 1844/533, loss: 0.9512029886245728 2023-01-21 09:32:02.094306: step: 1848/533, loss: 0.05611982196569443 2023-01-21 09:32:03.295349: step: 1852/533, loss: 0.14940910041332245 2023-01-21 09:32:04.491780: step: 1856/533, loss: 0.08887653052806854 2023-01-21 09:32:05.698123: step: 1860/533, loss: 0.01161956787109375 2023-01-21 09:32:06.913448: step: 1864/533, loss: 0.1938934326171875 2023-01-21 09:32:08.078290: step: 1868/533, loss: 0.1492149382829666 2023-01-21 09:32:09.260038: step: 1872/533, loss: 0.26267844438552856 2023-01-21 09:32:10.477898: step: 1876/533, loss: 0.12772627174854279 2023-01-21 09:32:11.685223: step: 1880/533, loss: 0.09009389579296112 2023-01-21 09:32:12.881169: step: 1884/533, loss: 0.34288281202316284 2023-01-21 09:32:14.074378: step: 1888/533, loss: 0.10200528800487518 2023-01-21 09:32:15.277318: step: 1892/533, loss: 0.49737465381622314 2023-01-21 09:32:16.419479: step: 1896/533, loss: 0.1520950347185135 2023-01-21 09:32:17.611806: step: 1900/533, loss: 0.21713094413280487 2023-01-21 09:32:18.762353: step: 1904/533, loss: 0.48465415835380554 2023-01-21 09:32:19.938784: step: 1908/533, loss: 0.13304853439331055 2023-01-21 09:32:21.110380: step: 1912/533, loss: 2.005528211593628 2023-01-21 09:32:22.299098: step: 1916/533, loss: 0.5247257351875305 2023-01-21 09:32:23.459876: step: 1920/533, loss: 0.1867447793483734 2023-01-21 09:32:24.641516: step: 1924/533, loss: 0.10536666214466095 2023-01-21 09:32:25.825089: step: 1928/533, loss: 0.6730639338493347 2023-01-21 09:32:26.995921: step: 1932/533, loss: 0.9028729200363159 2023-01-21 09:32:28.173055: step: 1936/533, loss: 2.4237864017486572 2023-01-21 09:32:29.373270: step: 1940/533, loss: 0.2181251496076584 2023-01-21 09:32:30.515522: step: 1944/533, loss: 0.03633255884051323 2023-01-21 09:32:31.676769: step: 1948/533, loss: 0.4216315448284149 2023-01-21 09:32:32.864404: step: 1952/533, loss: 0.026334190741181374 2023-01-21 09:32:34.069312: step: 1956/533, loss: 0.07945594936609268 2023-01-21 09:32:35.241050: step: 1960/533, loss: 0.12200164794921875 2023-01-21 09:32:36.425529: step: 1964/533, loss: 0.03914046287536621 2023-01-21 09:32:37.657866: step: 1968/533, loss: 0.053531549870967865 2023-01-21 09:32:38.854317: step: 1972/533, loss: 0.1782490313053131 2023-01-21 09:32:40.021243: step: 1976/533, loss: 0.0967465415596962 2023-01-21 09:32:41.231730: step: 1980/533, loss: 0.0998849868774414 2023-01-21 09:32:42.433806: step: 1984/533, loss: 0.1548910140991211 2023-01-21 09:32:43.621686: step: 1988/533, loss: 0.11932344734668732 2023-01-21 09:32:44.756989: step: 1992/533, loss: 0.2317657470703125 2023-01-21 09:32:45.964935: step: 1996/533, loss: 0.13084612786769867 2023-01-21 09:32:47.132097: step: 2000/533, loss: 0.2056548297405243 2023-01-21 09:32:48.284354: step: 2004/533, loss: 0.2298460453748703 2023-01-21 09:32:49.487801: step: 2008/533, loss: 0.12498773634433746 2023-01-21 09:32:50.672272: step: 2012/533, loss: 0.17964085936546326 2023-01-21 09:32:51.827786: step: 2016/533, loss: 0.2789931297302246 2023-01-21 09:32:52.994842: step: 2020/533, loss: 0.08965711295604706 2023-01-21 09:32:54.185909: step: 2024/533, loss: 0.17179150879383087 2023-01-21 09:32:55.374076: step: 2028/533, loss: 0.09418530762195587 2023-01-21 09:32:56.569038: step: 2032/533, loss: 0.16949692368507385 2023-01-21 09:32:57.779201: step: 2036/533, loss: 0.08645877987146378 2023-01-21 09:32:58.955547: step: 2040/533, loss: 0.0520998015999794 2023-01-21 09:33:00.133376: step: 2044/533, loss: 0.23317575454711914 2023-01-21 09:33:01.397811: step: 2048/533, loss: 0.14852294325828552 2023-01-21 09:33:02.579281: step: 2052/533, loss: 0.13098320364952087 2023-01-21 09:33:03.718875: step: 2056/533, loss: 0.03686361387372017 2023-01-21 09:33:04.942926: step: 2060/533, loss: 0.1774548590183258 2023-01-21 09:33:06.074896: step: 2064/533, loss: 0.12040863186120987 2023-01-21 09:33:07.356645: step: 2068/533, loss: 0.023991085588932037 2023-01-21 09:33:08.540272: step: 2072/533, loss: 0.26984015107154846 2023-01-21 09:33:09.709263: step: 2076/533, loss: 1.1108540296554565 2023-01-21 09:33:10.869244: step: 2080/533, loss: 0.05016741901636124 2023-01-21 09:33:12.080420: step: 2084/533, loss: 0.304918110370636 2023-01-21 09:33:13.286380: step: 2088/533, loss: 0.21580210328102112 2023-01-21 09:33:14.486113: step: 2092/533, loss: 0.28309839963912964 2023-01-21 09:33:15.666322: step: 2096/533, loss: 0.09128522872924805 2023-01-21 09:33:16.867332: step: 2100/533, loss: 0.09254761040210724 2023-01-21 09:33:18.012296: step: 2104/533, loss: 0.1728985756635666 2023-01-21 09:33:19.183773: step: 2108/533, loss: 0.028179071843624115 2023-01-21 09:33:20.367407: step: 2112/533, loss: 0.0651882141828537 2023-01-21 09:33:21.566204: step: 2116/533, loss: 0.5313538312911987 2023-01-21 09:33:22.821034: step: 2120/533, loss: 0.6243495941162109 2023-01-21 09:33:23.984231: step: 2124/533, loss: 0.23429462313652039 2023-01-21 09:33:25.137350: step: 2128/533, loss: 0.12990951538085938 2023-01-21 09:33:26.279140: step: 2132/533, loss: 0.15905046463012695 ================================================== Loss: 0.279 -------------------- Dev: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Test: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} 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': 5} Korean: {'event': {'p': 0.6444444444444445, 'r': 0.4603174603174603, 'f1': 0.537037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 09:34:13.635944: step: 4/533, loss: 0.1402624249458313 2023-01-21 09:34:14.816550: step: 8/533, loss: 0.04164600372314453 2023-01-21 09:34:15.990813: step: 12/533, loss: 0.1231774389743805 2023-01-21 09:34:17.188076: step: 16/533, loss: 0.022655677050352097 2023-01-21 09:34:18.375711: step: 20/533, loss: 0.11645087599754333 2023-01-21 09:34:19.547804: step: 24/533, loss: 0.23251761496067047 2023-01-21 09:34:20.716929: step: 28/533, loss: 0.12609200179576874 2023-01-21 09:34:21.888394: step: 32/533, loss: 0.036342814564704895 2023-01-21 09:34:23.048252: step: 36/533, loss: 0.07379398494958878 2023-01-21 09:34:24.228782: step: 40/533, loss: 0.16314354538917542 2023-01-21 09:34:25.431925: step: 44/533, loss: 0.09132575988769531 2023-01-21 09:34:26.617133: step: 48/533, loss: 0.14595623314380646 2023-01-21 09:34:27.807377: step: 52/533, loss: 0.10838431864976883 2023-01-21 09:34:29.023080: step: 56/533, loss: 0.05974111706018448 2023-01-21 09:34:30.203600: step: 60/533, loss: 0.096981480717659 2023-01-21 09:34:31.369455: step: 64/533, loss: 0.09481563419103622 2023-01-21 09:34:32.557523: step: 68/533, loss: 0.14430314302444458 2023-01-21 09:34:33.761360: step: 72/533, loss: 0.10095834732055664 2023-01-21 09:34:34.973967: step: 76/533, loss: 0.11170845478773117 2023-01-21 09:34:36.155615: step: 80/533, loss: 0.23635263741016388 2023-01-21 09:34:37.339721: step: 84/533, loss: 0.0709206610918045 2023-01-21 09:34:38.522550: step: 88/533, loss: 0.40613335371017456 2023-01-21 09:34:39.729607: step: 92/533, loss: 0.0773090347647667 2023-01-21 09:34:40.892316: step: 96/533, loss: 0.044652268290519714 2023-01-21 09:34:42.066916: step: 100/533, loss: 0.13224896788597107 2023-01-21 09:34:43.250154: step: 104/533, loss: 0.16517484188079834 2023-01-21 09:34:44.450474: step: 108/533, loss: 0.17807331681251526 2023-01-21 09:34:45.643425: step: 112/533, loss: 0.10436401516199112 2023-01-21 09:34:46.850235: step: 116/533, loss: 0.18117032945156097 2023-01-21 09:34:48.012528: step: 120/533, loss: 0.145610511302948 2023-01-21 09:34:49.210358: step: 124/533, loss: 0.2857432961463928 2023-01-21 09:34:50.381357: step: 128/533, loss: 0.23638640344142914 2023-01-21 09:34:51.538166: step: 132/533, loss: 0.05786857753992081 2023-01-21 09:34:52.707196: step: 136/533, loss: 0.08663368225097656 2023-01-21 09:34:53.860732: step: 140/533, loss: 0.03582906723022461 2023-01-21 09:34:55.027886: step: 144/533, loss: 0.11095075309276581 2023-01-21 09:34:56.195687: step: 148/533, loss: 0.3409159481525421 2023-01-21 09:34:57.367232: step: 152/533, loss: 0.07475566864013672 2023-01-21 09:34:58.592503: step: 156/533, loss: 0.6491065621376038 2023-01-21 09:34:59.808733: step: 160/533, loss: 0.13936004042625427 2023-01-21 09:35:01.043260: step: 164/533, loss: 0.16715583205223083 2023-01-21 09:35:02.279073: step: 168/533, loss: 0.04724021255970001 2023-01-21 09:35:03.481977: step: 172/533, loss: 0.04715633764863014 2023-01-21 09:35:04.669417: step: 176/533, loss: 0.11720866709947586 2023-01-21 09:35:05.833401: step: 180/533, loss: 0.098785400390625 2023-01-21 09:35:06.993331: step: 184/533, loss: 0.13359108567237854 2023-01-21 09:35:08.185004: step: 188/533, loss: 0.34552955627441406 2023-01-21 09:35:09.394215: step: 192/533, loss: 0.4464438259601593 2023-01-21 09:35:10.627167: step: 196/533, loss: 0.08923645317554474 2023-01-21 09:35:11.794678: step: 200/533, loss: 0.31227704882621765 2023-01-21 09:35:12.997800: step: 204/533, loss: 0.26824918389320374 2023-01-21 09:35:14.206765: step: 208/533, loss: 0.020005034282803535 2023-01-21 09:35:15.370104: step: 212/533, loss: 0.1812088042497635 2023-01-21 09:35:16.548750: step: 216/533, loss: 0.039708614349365234 2023-01-21 09:35:17.744587: step: 220/533, loss: 0.4655044376850128 2023-01-21 09:35:18.927573: step: 224/533, loss: 1.5419625043869019 2023-01-21 09:35:20.143047: step: 228/533, loss: 0.15453405678272247 2023-01-21 09:35:21.419707: step: 232/533, loss: 0.2798457145690918 2023-01-21 09:35:22.591859: step: 236/533, loss: 0.015505028888583183 2023-01-21 09:35:23.867260: step: 240/533, loss: 0.11501885205507278 2023-01-21 09:35:25.028171: step: 244/533, loss: 0.2427239716053009 2023-01-21 09:35:26.214345: step: 248/533, loss: 0.20234784483909607 2023-01-21 09:35:27.405563: step: 252/533, loss: 0.2615823745727539 2023-01-21 09:35:28.572154: step: 256/533, loss: 0.09938888251781464 2023-01-21 09:35:29.733188: step: 260/533, loss: 0.13757282495498657 2023-01-21 09:35:30.919127: step: 264/533, loss: 0.16167640686035156 2023-01-21 09:35:32.156479: step: 268/533, loss: 0.2587317228317261 2023-01-21 09:35:33.332221: step: 272/533, loss: 0.24378043413162231 2023-01-21 09:35:34.522298: step: 276/533, loss: 0.05969972535967827 2023-01-21 09:35:35.693507: step: 280/533, loss: 0.15841493010520935 2023-01-21 09:35:36.877523: step: 284/533, loss: 0.312185138463974 2023-01-21 09:35:38.064798: step: 288/533, loss: 0.4866515100002289 2023-01-21 09:35:39.300543: step: 292/533, loss: 0.01793675497174263 2023-01-21 09:35:40.478387: step: 296/533, loss: 0.12395727634429932 2023-01-21 09:35:41.625031: step: 300/533, loss: 0.024558259174227715 2023-01-21 09:35:42.822974: step: 304/533, loss: 0.11003799736499786 2023-01-21 09:35:43.990146: step: 308/533, loss: 0.05248909071087837 2023-01-21 09:35:45.178714: step: 312/533, loss: 0.19564762711524963 2023-01-21 09:35:46.384264: step: 316/533, loss: 0.03859877586364746 2023-01-21 09:35:47.586595: step: 320/533, loss: 0.032102394849061966 2023-01-21 09:35:48.732116: step: 324/533, loss: 0.030251741409301758 2023-01-21 09:35:49.952273: step: 328/533, loss: 1.7634994983673096 2023-01-21 09:35:51.164026: step: 332/533, loss: 0.06786461174488068 2023-01-21 09:35:52.367296: step: 336/533, loss: 0.13286533951759338 2023-01-21 09:35:53.570152: step: 340/533, loss: 1.558632493019104 2023-01-21 09:35:54.745109: step: 344/533, loss: 0.10671892017126083 2023-01-21 09:35:55.925750: step: 348/533, loss: 0.12205901741981506 2023-01-21 09:35:57.109940: step: 352/533, loss: 0.0779503807425499 2023-01-21 09:35:58.297901: step: 356/533, loss: 0.02380542829632759 2023-01-21 09:35:59.503719: step: 360/533, loss: 0.1602693498134613 2023-01-21 09:36:00.638987: step: 364/533, loss: 0.9128307104110718 2023-01-21 09:36:01.813406: step: 368/533, loss: 0.1600487232208252 2023-01-21 09:36:02.980406: step: 372/533, loss: 0.39343318343162537 2023-01-21 09:36:04.177819: step: 376/533, loss: 0.1305026113986969 2023-01-21 09:36:05.404598: step: 380/533, loss: 0.0362975150346756 2023-01-21 09:36:06.610384: step: 384/533, loss: 0.7242580056190491 2023-01-21 09:36:07.777182: step: 388/533, loss: 0.04973917454481125 2023-01-21 09:36:08.959626: step: 392/533, loss: 0.20595207810401917 2023-01-21 09:36:10.150792: step: 396/533, loss: 0.048276614397764206 2023-01-21 09:36:11.334913: step: 400/533, loss: 0.8103103637695312 2023-01-21 09:36:12.523548: step: 404/533, loss: 0.8164023160934448 2023-01-21 09:36:13.713655: step: 408/533, loss: 0.4811538755893707 2023-01-21 09:36:14.901322: step: 412/533, loss: 0.19216632843017578 2023-01-21 09:36:16.068409: step: 416/533, loss: 0.19023312628269196 2023-01-21 09:36:17.232800: step: 420/533, loss: 0.9657771587371826 2023-01-21 09:36:18.404660: step: 424/533, loss: 0.09076862037181854 2023-01-21 09:36:19.595301: step: 428/533, loss: 0.1797008067369461 2023-01-21 09:36:20.756482: step: 432/533, loss: 0.0892309695482254 2023-01-21 09:36:21.925168: step: 436/533, loss: 0.06246676295995712 2023-01-21 09:36:23.099938: step: 440/533, loss: 0.13895034790039062 2023-01-21 09:36:24.259046: step: 444/533, loss: 0.19411660730838776 2023-01-21 09:36:25.463339: step: 448/533, loss: 0.10427618026733398 2023-01-21 09:36:26.638018: step: 452/533, loss: 0.15523472428321838 2023-01-21 09:36:27.824091: step: 456/533, loss: 0.07152634114027023 2023-01-21 09:36:28.999465: step: 460/533, loss: 0.18934230506420135 2023-01-21 09:36:30.163126: step: 464/533, loss: 0.3714737892150879 2023-01-21 09:36:31.359807: step: 468/533, loss: 0.08996625244617462 2023-01-21 09:36:32.507276: step: 472/533, loss: 0.062832310795784 2023-01-21 09:36:33.713979: step: 476/533, loss: 0.32677918672561646 2023-01-21 09:36:34.925269: step: 480/533, loss: 0.1360362470149994 2023-01-21 09:36:36.083285: step: 484/533, loss: 0.04621868580579758 2023-01-21 09:36:37.252293: step: 488/533, loss: 0.38293105363845825 2023-01-21 09:36:38.439296: step: 492/533, loss: 0.1301508992910385 2023-01-21 09:36:39.631165: step: 496/533, loss: 0.20964142680168152 2023-01-21 09:36:40.792321: step: 500/533, loss: 0.06040621176362038 2023-01-21 09:36:41.974202: step: 504/533, loss: 0.3040216565132141 2023-01-21 09:36:43.178799: step: 508/533, loss: 0.11116790771484375 2023-01-21 09:36:44.360512: step: 512/533, loss: 0.13114652037620544 2023-01-21 09:36:45.632104: step: 516/533, loss: 0.2801080346107483 2023-01-21 09:36:46.824485: step: 520/533, loss: 0.05532646179199219 2023-01-21 09:36:48.065590: step: 524/533, loss: 0.3578099012374878 2023-01-21 09:36:49.270110: step: 528/533, loss: 0.2572367787361145 2023-01-21 09:36:50.407581: step: 532/533, loss: 3.4118878841400146 2023-01-21 09:36:51.595080: step: 536/533, loss: 0.26610809564590454 2023-01-21 09:36:52.739733: step: 540/533, loss: 0.06443443894386292 2023-01-21 09:36:53.906463: step: 544/533, loss: 0.022541046142578125 2023-01-21 09:36:55.062123: step: 548/533, loss: 0.17633609473705292 2023-01-21 09:36:56.233815: step: 552/533, loss: 0.11661338806152344 2023-01-21 09:36:57.425604: step: 556/533, loss: 0.1085977554321289 2023-01-21 09:36:58.610611: step: 560/533, loss: 0.09120988845825195 2023-01-21 09:36:59.800557: step: 564/533, loss: 0.08338051289319992 2023-01-21 09:37:00.959062: step: 568/533, loss: 0.3780990540981293 2023-01-21 09:37:02.125133: step: 572/533, loss: 0.16336117684841156 2023-01-21 09:37:03.335812: step: 576/533, loss: 0.2157839834690094 2023-01-21 09:37:04.514098: step: 580/533, loss: 0.12675128877162933 2023-01-21 09:37:05.665068: step: 584/533, loss: 0.17588502168655396 2023-01-21 09:37:06.858500: step: 588/533, loss: 0.1416037678718567 2023-01-21 09:37:08.068979: step: 592/533, loss: 0.5350645780563354 2023-01-21 09:37:09.258393: step: 596/533, loss: 0.0289019588381052 2023-01-21 09:37:10.428084: step: 600/533, loss: 0.03147625923156738 2023-01-21 09:37:11.584727: step: 604/533, loss: 0.011325454339385033 2023-01-21 09:37:12.807993: step: 608/533, loss: 0.42858293652534485 2023-01-21 09:37:13.986681: step: 612/533, loss: 0.17268818616867065 2023-01-21 09:37:15.175326: step: 616/533, loss: 0.2072030007839203 2023-01-21 09:37:16.336770: step: 620/533, loss: 0.02739272080361843 2023-01-21 09:37:17.556999: step: 624/533, loss: 0.239209845662117 2023-01-21 09:37:18.741533: step: 628/533, loss: 0.157927468419075 2023-01-21 09:37:19.898632: step: 632/533, loss: 0.05565395578742027 2023-01-21 09:37:21.064476: step: 636/533, loss: 0.0764245018362999 2023-01-21 09:37:22.234691: step: 640/533, loss: 0.39503756165504456 2023-01-21 09:37:23.464466: step: 644/533, loss: 0.723455548286438 2023-01-21 09:37:24.643013: step: 648/533, loss: 0.05753688886761665 2023-01-21 09:37:25.818726: step: 652/533, loss: 0.03455867990851402 2023-01-21 09:37:27.032724: step: 656/533, loss: 0.20306864380836487 2023-01-21 09:37:28.238099: step: 660/533, loss: 0.12418485432863235 2023-01-21 09:37:29.410859: step: 664/533, loss: 0.15200595557689667 2023-01-21 09:37:30.588017: step: 668/533, loss: 0.05314607545733452 2023-01-21 09:37:31.782474: step: 672/533, loss: 0.16093826293945312 2023-01-21 09:37:32.958850: step: 676/533, loss: 0.012057686224579811 2023-01-21 09:37:34.138186: step: 680/533, loss: 0.061133477836847305 2023-01-21 09:37:35.334452: step: 684/533, loss: 0.08942899852991104 2023-01-21 09:37:36.521291: step: 688/533, loss: 0.05464329570531845 2023-01-21 09:37:37.666640: step: 692/533, loss: 0.2680133283138275 2023-01-21 09:37:38.846656: step: 696/533, loss: 0.0571599006652832 2023-01-21 09:37:40.001977: step: 700/533, loss: 0.9248478412628174 2023-01-21 09:37:41.185995: step: 704/533, loss: 0.03651714324951172 2023-01-21 09:37:42.385624: step: 708/533, loss: 0.11802182346582413 2023-01-21 09:37:43.537288: step: 712/533, loss: 0.30104243755340576 2023-01-21 09:37:44.745627: step: 716/533, loss: 0.03824158012866974 2023-01-21 09:37:45.894906: step: 720/533, loss: 0.0996273010969162 2023-01-21 09:37:47.080719: step: 724/533, loss: 0.5776216983795166 2023-01-21 09:37:48.223668: step: 728/533, loss: 0.08605937659740448 2023-01-21 09:37:49.410973: step: 732/533, loss: 0.08463068306446075 2023-01-21 09:37:50.606951: step: 736/533, loss: 0.04151606559753418 2023-01-21 09:37:51.795397: step: 740/533, loss: 0.1516055166721344 2023-01-21 09:37:53.011914: step: 744/533, loss: 0.114264115691185 2023-01-21 09:37:54.179219: step: 748/533, loss: 0.08029527962207794 2023-01-21 09:37:55.360371: step: 752/533, loss: 0.047203924506902695 2023-01-21 09:37:56.522358: step: 756/533, loss: 0.45689982175827026 2023-01-21 09:37:57.688102: step: 760/533, loss: 0.49262797832489014 2023-01-21 09:37:58.849109: step: 764/533, loss: 0.10497431457042694 2023-01-21 09:38:00.001913: step: 768/533, loss: 0.1463664174079895 2023-01-21 09:38:01.176845: step: 772/533, loss: 0.02423887327313423 2023-01-21 09:38:02.377318: step: 776/533, loss: 0.10864267498254776 2023-01-21 09:38:03.561030: step: 780/533, loss: 0.06768804043531418 2023-01-21 09:38:04.734249: step: 784/533, loss: 0.18495555222034454 2023-01-21 09:38:05.952766: step: 788/533, loss: 1.8760281801223755 2023-01-21 09:38:07.156035: step: 792/533, loss: 0.12130908668041229 2023-01-21 09:38:08.347493: step: 796/533, loss: 0.21055154502391815 2023-01-21 09:38:09.598171: step: 800/533, loss: 0.05789361149072647 2023-01-21 09:38:10.795932: step: 804/533, loss: 0.19672508537769318 2023-01-21 09:38:12.004959: step: 808/533, loss: 0.42760956287384033 2023-01-21 09:38:13.173319: step: 812/533, loss: 0.05177664756774902 2023-01-21 09:38:14.325602: step: 816/533, loss: 0.41149386763572693 2023-01-21 09:38:15.514824: step: 820/533, loss: 0.05115308612585068 2023-01-21 09:38:16.746578: step: 824/533, loss: 0.10420207679271698 2023-01-21 09:38:17.930146: step: 828/533, loss: 0.08954200893640518 2023-01-21 09:38:19.106076: step: 832/533, loss: 0.18194933235645294 2023-01-21 09:38:20.275095: step: 836/533, loss: 0.12253104150295258 2023-01-21 09:38:21.451655: step: 840/533, loss: 0.07624082267284393 2023-01-21 09:38:22.605059: step: 844/533, loss: 0.164313405752182 2023-01-21 09:38:23.753471: step: 848/533, loss: 0.023150037974119186 2023-01-21 09:38:24.929184: step: 852/533, loss: 0.11395011097192764 2023-01-21 09:38:26.098348: step: 856/533, loss: 0.23876214027404785 2023-01-21 09:38:27.236348: step: 860/533, loss: 0.05722036957740784 2023-01-21 09:38:28.410113: step: 864/533, loss: 0.1533019095659256 2023-01-21 09:38:29.573328: step: 868/533, loss: 0.04178767278790474 2023-01-21 09:38:30.729496: step: 872/533, loss: 0.4499811828136444 2023-01-21 09:38:31.940771: step: 876/533, loss: 0.13845939934253693 2023-01-21 09:38:33.132712: step: 880/533, loss: 0.09106773883104324 2023-01-21 09:38:34.273773: step: 884/533, loss: 0.07382631301879883 2023-01-21 09:38:35.463690: step: 888/533, loss: 0.16635771095752716 2023-01-21 09:38:36.628146: step: 892/533, loss: 0.1024908721446991 2023-01-21 09:38:37.803470: step: 896/533, loss: 0.29661786556243896 2023-01-21 09:38:39.010992: step: 900/533, loss: 0.1438174843788147 2023-01-21 09:38:40.240170: step: 904/533, loss: 0.16420002281665802 2023-01-21 09:38:41.414485: step: 908/533, loss: 0.11474857479333878 2023-01-21 09:38:42.587365: step: 912/533, loss: 0.04106488078832626 2023-01-21 09:38:43.821547: step: 916/533, loss: 0.23226386308670044 2023-01-21 09:38:44.956935: step: 920/533, loss: 0.0999455451965332 2023-01-21 09:38:46.163052: step: 924/533, loss: 0.06648831069469452 2023-01-21 09:38:47.335866: step: 928/533, loss: 0.08065791428089142 2023-01-21 09:38:48.548384: step: 932/533, loss: 0.2053905427455902 2023-01-21 09:38:49.741073: step: 936/533, loss: 0.1347431242465973 2023-01-21 09:38:50.972673: step: 940/533, loss: 0.09002552181482315 2023-01-21 09:38:52.250320: step: 944/533, loss: 0.06162893772125244 2023-01-21 09:38:53.421088: step: 948/533, loss: 0.17965379357337952 2023-01-21 09:38:54.603734: step: 952/533, loss: 0.16462893784046173 2023-01-21 09:38:55.777246: step: 956/533, loss: 0.0900600403547287 2023-01-21 09:38:57.007555: step: 960/533, loss: 0.12690864503383636 2023-01-21 09:38:58.192814: step: 964/533, loss: 0.015422534197568893 2023-01-21 09:38:59.345075: step: 968/533, loss: 0.4357540011405945 2023-01-21 09:39:00.489472: step: 972/533, loss: 0.10814743489027023 2023-01-21 09:39:01.615305: step: 976/533, loss: 0.06888599693775177 2023-01-21 09:39:02.802992: step: 980/533, loss: 0.033872321248054504 2023-01-21 09:39:03.964399: step: 984/533, loss: 0.1013982743024826 2023-01-21 09:39:05.139816: step: 988/533, loss: 0.036885831505060196 2023-01-21 09:39:06.373205: step: 992/533, loss: 0.320893257856369 2023-01-21 09:39:07.585740: step: 996/533, loss: 0.3332177400588989 2023-01-21 09:39:08.708281: step: 1000/533, loss: 0.3373338282108307 2023-01-21 09:39:09.917790: step: 1004/533, loss: 0.6432033777236938 2023-01-21 09:39:11.121053: step: 1008/533, loss: 0.28733140230178833 2023-01-21 09:39:12.339656: step: 1012/533, loss: 0.2184928059577942 2023-01-21 09:39:13.541381: step: 1016/533, loss: 0.08183488249778748 2023-01-21 09:39:14.718446: step: 1020/533, loss: 0.14864301681518555 2023-01-21 09:39:15.891434: step: 1024/533, loss: 0.16043797135353088 2023-01-21 09:39:17.059327: step: 1028/533, loss: 0.6593327522277832 2023-01-21 09:39:18.221505: step: 1032/533, loss: 0.13295136392116547 2023-01-21 09:39:19.382322: step: 1036/533, loss: 0.18243809044361115 2023-01-21 09:39:20.571409: step: 1040/533, loss: 0.03352494165301323 2023-01-21 09:39:21.741416: step: 1044/533, loss: 0.5756750106811523 2023-01-21 09:39:22.957379: step: 1048/533, loss: 0.5511318445205688 2023-01-21 09:39:24.161478: step: 1052/533, loss: 0.20697203278541565 2023-01-21 09:39:25.350136: step: 1056/533, loss: 0.05725250393152237 2023-01-21 09:39:26.522536: step: 1060/533, loss: 3.5092246532440186 2023-01-21 09:39:27.655401: step: 1064/533, loss: 0.2597506642341614 2023-01-21 09:39:28.882242: step: 1068/533, loss: 0.22108283638954163 2023-01-21 09:39:30.072649: step: 1072/533, loss: 0.310195654630661 2023-01-21 09:39:31.249899: step: 1076/533, loss: 0.07069483399391174 2023-01-21 09:39:32.428396: step: 1080/533, loss: 0.22458162903785706 2023-01-21 09:39:33.572367: step: 1084/533, loss: 0.02254791371524334 2023-01-21 09:39:34.764252: step: 1088/533, loss: 0.18851758539676666 2023-01-21 09:39:35.945912: step: 1092/533, loss: 0.2145695984363556 2023-01-21 09:39:37.101427: step: 1096/533, loss: 0.34991785883903503 2023-01-21 09:39:38.298819: step: 1100/533, loss: 0.12251511216163635 2023-01-21 09:39:39.536012: step: 1104/533, loss: 0.05593429133296013 2023-01-21 09:39:40.689965: step: 1108/533, loss: 0.15074720978736877 2023-01-21 09:39:41.934797: step: 1112/533, loss: 0.3482500910758972 2023-01-21 09:39:43.138416: step: 1116/533, loss: 0.11167445778846741 2023-01-21 09:39:44.299648: step: 1120/533, loss: 0.05750274658203125 2023-01-21 09:39:45.487672: step: 1124/533, loss: 0.513885498046875 2023-01-21 09:39:46.686804: step: 1128/533, loss: 0.07086972892284393 2023-01-21 09:39:47.888442: step: 1132/533, loss: 0.2156183272600174 2023-01-21 09:39:49.045339: step: 1136/533, loss: 0.2882498800754547 2023-01-21 09:39:50.212385: step: 1140/533, loss: 0.08378219604492188 2023-01-21 09:39:51.444996: step: 1144/533, loss: 0.021956156939268112 2023-01-21 09:39:52.656917: step: 1148/533, loss: 0.1018834114074707 2023-01-21 09:39:53.825978: step: 1152/533, loss: 0.5925976634025574 2023-01-21 09:39:55.008951: step: 1156/533, loss: 0.025112057104706764 2023-01-21 09:39:56.231961: step: 1160/533, loss: 0.11240149289369583 2023-01-21 09:39:57.412253: step: 1164/533, loss: 0.2043600082397461 2023-01-21 09:39:58.569860: step: 1168/533, loss: 0.3127053380012512 2023-01-21 09:39:59.749568: step: 1172/533, loss: 0.11704130470752716 2023-01-21 09:40:00.962757: step: 1176/533, loss: 0.07777661830186844 2023-01-21 09:40:02.148032: step: 1180/533, loss: 0.1534065306186676 2023-01-21 09:40:03.346761: step: 1184/533, loss: 0.09513416886329651 2023-01-21 09:40:04.550416: step: 1188/533, loss: 0.06841164082288742 2023-01-21 09:40:05.703980: step: 1192/533, loss: 0.12703752517700195 2023-01-21 09:40:06.879099: step: 1196/533, loss: 0.10455399006605148 2023-01-21 09:40:08.099515: step: 1200/533, loss: 0.07830943912267685 2023-01-21 09:40:09.291770: step: 1204/533, loss: 0.43790698051452637 2023-01-21 09:40:10.497971: step: 1208/533, loss: 0.3851598799228668 2023-01-21 09:40:11.653489: step: 1212/533, loss: 0.1309981346130371 2023-01-21 09:40:12.899696: step: 1216/533, loss: 0.13694210350513458 2023-01-21 09:40:14.116090: step: 1220/533, loss: 0.11538295447826385 2023-01-21 09:40:15.320802: step: 1224/533, loss: 0.2820868492126465 2023-01-21 09:40:16.542594: step: 1228/533, loss: 0.495748907327652 2023-01-21 09:40:17.744391: step: 1232/533, loss: 0.04419393837451935 2023-01-21 09:40:18.911331: step: 1236/533, loss: 0.11558160930871964 2023-01-21 09:40:20.076591: step: 1240/533, loss: 0.08236837387084961 2023-01-21 09:40:21.221498: step: 1244/533, loss: 0.06681094318628311 2023-01-21 09:40:22.437276: step: 1248/533, loss: 0.03396787494421005 2023-01-21 09:40:23.656318: step: 1252/533, loss: 0.07566051930189133 2023-01-21 09:40:24.836838: step: 1256/533, loss: 0.17411479353904724 2023-01-21 09:40:25.999367: step: 1260/533, loss: 0.13344541192054749 2023-01-21 09:40:27.166912: step: 1264/533, loss: 0.037888478487730026 2023-01-21 09:40:28.383959: step: 1268/533, loss: 0.2684181332588196 2023-01-21 09:40:29.602875: step: 1272/533, loss: 0.07422362267971039 2023-01-21 09:40:30.765060: step: 1276/533, loss: 0.419964075088501 2023-01-21 09:40:31.932539: step: 1280/533, loss: 0.5850145220756531 2023-01-21 09:40:33.098355: step: 1284/533, loss: 0.038173869252204895 2023-01-21 09:40:34.250927: step: 1288/533, loss: 0.1673487275838852 2023-01-21 09:40:35.427667: step: 1292/533, loss: 0.11415676772594452 2023-01-21 09:40:36.608106: step: 1296/533, loss: 0.14867329597473145 2023-01-21 09:40:37.843456: step: 1300/533, loss: 0.011673260480165482 2023-01-21 09:40:39.025981: step: 1304/533, loss: 1.3752771615982056 2023-01-21 09:40:40.201805: step: 1308/533, loss: 0.15026354789733887 2023-01-21 09:40:41.383288: step: 1312/533, loss: 0.08691949397325516 2023-01-21 09:40:42.578144: step: 1316/533, loss: 0.44630491733551025 2023-01-21 09:40:43.753672: step: 1320/533, loss: 0.04766340181231499 2023-01-21 09:40:44.980782: step: 1324/533, loss: 0.051876164972782135 2023-01-21 09:40:46.164945: step: 1328/533, loss: 0.9295927286148071 2023-01-21 09:40:47.391614: step: 1332/533, loss: 0.09872036427259445 2023-01-21 09:40:48.621592: step: 1336/533, loss: 0.16506114602088928 2023-01-21 09:40:49.830198: step: 1340/533, loss: 0.14645300805568695 2023-01-21 09:40:51.004458: step: 1344/533, loss: 1.0066637992858887 2023-01-21 09:40:52.216990: step: 1348/533, loss: 0.19292087852954865 2023-01-21 09:40:53.359138: step: 1352/533, loss: 0.11848153918981552 2023-01-21 09:40:54.578061: step: 1356/533, loss: 0.221858412027359 2023-01-21 09:40:55.863614: step: 1360/533, loss: 0.2674716114997864 2023-01-21 09:40:57.053618: step: 1364/533, loss: 0.059876635670661926 2023-01-21 09:40:58.204824: step: 1368/533, loss: 0.1604660004377365 2023-01-21 09:40:59.369133: step: 1372/533, loss: 0.12207386642694473 2023-01-21 09:41:00.537813: step: 1376/533, loss: 0.12905368208885193 2023-01-21 09:41:01.752519: step: 1380/533, loss: 0.2698141932487488 2023-01-21 09:41:02.951223: step: 1384/533, loss: 0.2435825914144516 2023-01-21 09:41:04.098784: step: 1388/533, loss: 0.06786127388477325 2023-01-21 09:41:05.270051: step: 1392/533, loss: 0.38354578614234924 2023-01-21 09:41:06.451348: step: 1396/533, loss: 0.12366710603237152 2023-01-21 09:41:07.665002: step: 1400/533, loss: 0.5397354364395142 2023-01-21 09:41:08.841961: step: 1404/533, loss: 0.028628351166844368 2023-01-21 09:41:10.092228: step: 1408/533, loss: 0.09934577345848083 2023-01-21 09:41:11.267578: step: 1412/533, loss: 0.7715206146240234 2023-01-21 09:41:12.418031: step: 1416/533, loss: 1.4465574026107788 2023-01-21 09:41:13.630140: step: 1420/533, loss: 0.1617651879787445 2023-01-21 09:41:14.821742: step: 1424/533, loss: 0.0394839309155941 2023-01-21 09:41:15.989403: step: 1428/533, loss: 0.05683021619915962 2023-01-21 09:41:17.172993: step: 1432/533, loss: 0.10786976665258408 2023-01-21 09:41:18.363976: step: 1436/533, loss: 0.027826596051454544 2023-01-21 09:41:19.576784: step: 1440/533, loss: 1.2564316987991333 2023-01-21 09:41:20.737616: step: 1444/533, loss: 0.12281899899244308 2023-01-21 09:41:21.938808: step: 1448/533, loss: 1.341007947921753 2023-01-21 09:41:23.131902: step: 1452/533, loss: 0.2766227722167969 2023-01-21 09:41:24.363003: step: 1456/533, loss: 0.17998160421848297 2023-01-21 09:41:25.512764: step: 1460/533, loss: 0.11274652928113937 2023-01-21 09:41:26.699791: step: 1464/533, loss: 0.06542710959911346 2023-01-21 09:41:27.911864: step: 1468/533, loss: 0.6500186920166016 2023-01-21 09:41:29.119738: step: 1472/533, loss: 0.11893444508314133 2023-01-21 09:41:30.301634: step: 1476/533, loss: 0.06078615039587021 2023-01-21 09:41:31.458347: step: 1480/533, loss: 0.13525590300559998 2023-01-21 09:41:32.610084: step: 1484/533, loss: 0.0734526664018631 2023-01-21 09:41:33.818350: step: 1488/533, loss: 0.402272492647171 2023-01-21 09:41:35.023953: step: 1492/533, loss: 0.10789927840232849 2023-01-21 09:41:36.206633: step: 1496/533, loss: 0.08972835540771484 2023-01-21 09:41:37.374796: step: 1500/533, loss: 0.1338546723127365 2023-01-21 09:41:38.518954: step: 1504/533, loss: 0.6546284556388855 2023-01-21 09:41:39.700887: step: 1508/533, loss: 0.09639425575733185 2023-01-21 09:41:40.884632: step: 1512/533, loss: 1.2416266202926636 2023-01-21 09:41:42.124915: step: 1516/533, loss: 0.7368118166923523 2023-01-21 09:41:43.302847: step: 1520/533, loss: 0.09771113842725754 2023-01-21 09:41:44.500540: step: 1524/533, loss: 0.163905531167984 2023-01-21 09:41:45.764786: step: 1528/533, loss: 0.13659754395484924 2023-01-21 09:41:46.941558: step: 1532/533, loss: 0.1884136199951172 2023-01-21 09:41:48.102966: step: 1536/533, loss: 0.0712137222290039 2023-01-21 09:41:49.267792: step: 1540/533, loss: 0.09784231334924698 2023-01-21 09:41:50.442621: step: 1544/533, loss: 0.39156195521354675 2023-01-21 09:41:51.677979: step: 1548/533, loss: 0.22279366850852966 2023-01-21 09:41:52.861380: step: 1552/533, loss: 0.0961771011352539 2023-01-21 09:41:54.072648: step: 1556/533, loss: 0.3711252212524414 2023-01-21 09:41:55.256230: step: 1560/533, loss: 0.11000586301088333 2023-01-21 09:41:56.446537: step: 1564/533, loss: 0.03015165403485298 2023-01-21 09:41:57.651289: step: 1568/533, loss: 0.13989749550819397 2023-01-21 09:41:58.837976: step: 1572/533, loss: 0.1575462371110916 2023-01-21 09:42:00.020427: step: 1576/533, loss: 0.1334587037563324 2023-01-21 09:42:01.222098: step: 1580/533, loss: 0.09696273505687714 2023-01-21 09:42:02.398580: step: 1584/533, loss: 0.05098571628332138 2023-01-21 09:42:03.596228: step: 1588/533, loss: 0.0862436294555664 2023-01-21 09:42:04.774816: step: 1592/533, loss: 0.04604396969079971 2023-01-21 09:42:05.946418: step: 1596/533, loss: 0.060735225677490234 2023-01-21 09:42:07.084068: step: 1600/533, loss: 0.02472858503460884 2023-01-21 09:42:08.254653: step: 1604/533, loss: 0.16002511978149414 2023-01-21 09:42:09.435990: step: 1608/533, loss: 0.5418399572372437 2023-01-21 09:42:10.625762: step: 1612/533, loss: 0.07610159367322922 2023-01-21 09:42:11.796071: step: 1616/533, loss: 0.1806151270866394 2023-01-21 09:42:12.963860: step: 1620/533, loss: 0.08388242870569229 2023-01-21 09:42:14.188406: step: 1624/533, loss: 0.2008608877658844 2023-01-21 09:42:15.385962: step: 1628/533, loss: 0.13060970604419708 2023-01-21 09:42:16.584823: step: 1632/533, loss: 0.052858926355838776 2023-01-21 09:42:17.781960: step: 1636/533, loss: 0.09687748551368713 2023-01-21 09:42:18.996336: step: 1640/533, loss: 0.615983784198761 2023-01-21 09:42:20.182990: step: 1644/533, loss: 0.2445390820503235 2023-01-21 09:42:21.340241: step: 1648/533, loss: 0.03076920472085476 2023-01-21 09:42:22.565703: step: 1652/533, loss: 0.05043325573205948 2023-01-21 09:42:23.763618: step: 1656/533, loss: 0.10877757519483566 2023-01-21 09:42:24.914876: step: 1660/533, loss: 0.17157940566539764 2023-01-21 09:42:26.104565: step: 1664/533, loss: 0.2575721740722656 2023-01-21 09:42:27.270898: step: 1668/533, loss: 0.05763950198888779 2023-01-21 09:42:28.462749: step: 1672/533, loss: 0.2720033824443817 2023-01-21 09:42:29.655741: step: 1676/533, loss: 0.22390387952327728 2023-01-21 09:42:30.848481: step: 1680/533, loss: 0.3240469992160797 2023-01-21 09:42:32.012506: step: 1684/533, loss: 0.17517486214637756 2023-01-21 09:42:33.202278: step: 1688/533, loss: 0.13015690445899963 2023-01-21 09:42:34.373099: step: 1692/533, loss: 0.45919427275657654 2023-01-21 09:42:35.534962: step: 1696/533, loss: 0.072052001953125 2023-01-21 09:42:36.738393: step: 1700/533, loss: 0.13439303636550903 2023-01-21 09:42:37.944655: step: 1704/533, loss: 0.05600857734680176 2023-01-21 09:42:39.134669: step: 1708/533, loss: 0.05688543617725372 2023-01-21 09:42:40.322069: step: 1712/533, loss: 0.14664840698242188 2023-01-21 09:42:41.488728: step: 1716/533, loss: 0.384204626083374 2023-01-21 09:42:42.672955: step: 1720/533, loss: 1.031225562095642 2023-01-21 09:42:43.888995: step: 1724/533, loss: 0.26040613651275635 2023-01-21 09:42:45.061534: step: 1728/533, loss: 0.037535667419433594 2023-01-21 09:42:46.237597: step: 1732/533, loss: 0.12251291424036026 2023-01-21 09:42:47.444096: step: 1736/533, loss: 0.07688362151384354 2023-01-21 09:42:48.636960: step: 1740/533, loss: 0.128939688205719 2023-01-21 09:42:49.837244: step: 1744/533, loss: 0.0218703281134367 2023-01-21 09:42:51.063346: step: 1748/533, loss: 0.07857814431190491 2023-01-21 09:42:52.244706: step: 1752/533, loss: 0.07582254707813263 2023-01-21 09:42:53.438086: step: 1756/533, loss: 0.067365363240242 2023-01-21 09:42:54.607844: step: 1760/533, loss: 0.10041113197803497 2023-01-21 09:42:55.794953: step: 1764/533, loss: 0.15190574526786804 2023-01-21 09:42:56.940307: step: 1768/533, loss: 0.13488759100437164 2023-01-21 09:42:58.128083: step: 1772/533, loss: 0.2523534893989563 2023-01-21 09:42:59.290496: step: 1776/533, loss: 0.15841913223266602 2023-01-21 09:43:00.452893: step: 1780/533, loss: 0.0786147192120552 2023-01-21 09:43:01.618563: step: 1784/533, loss: 0.30462560057640076 2023-01-21 09:43:02.806748: step: 1788/533, loss: 0.07425765693187714 2023-01-21 09:43:03.966053: step: 1792/533, loss: 0.020304633304476738 2023-01-21 09:43:05.163990: step: 1796/533, loss: 0.5724536180496216 2023-01-21 09:43:06.365972: step: 1800/533, loss: 0.14271003007888794 2023-01-21 09:43:07.568748: step: 1804/533, loss: 0.06013636663556099 2023-01-21 09:43:08.785044: step: 1808/533, loss: 0.14017534255981445 2023-01-21 09:43:09.984427: step: 1812/533, loss: 0.05206337571144104 2023-01-21 09:43:11.221575: step: 1816/533, loss: 0.37365150451660156 2023-01-21 09:43:12.421348: step: 1820/533, loss: 0.01740436628460884 2023-01-21 09:43:13.642131: step: 1824/533, loss: 0.27691972255706787 2023-01-21 09:43:14.820888: step: 1828/533, loss: 0.24644887447357178 2023-01-21 09:43:15.971413: step: 1832/533, loss: 0.09711380302906036 2023-01-21 09:43:17.144051: step: 1836/533, loss: 0.2083951085805893 2023-01-21 09:43:18.346573: step: 1840/533, loss: 0.0981113463640213 2023-01-21 09:43:19.545066: step: 1844/533, loss: 0.20288285613059998 2023-01-21 09:43:20.733466: step: 1848/533, loss: 0.4039088487625122 2023-01-21 09:43:21.895885: step: 1852/533, loss: 0.027602385729551315 2023-01-21 09:43:23.081564: step: 1856/533, loss: 0.14193540811538696 2023-01-21 09:43:24.230378: step: 1860/533, loss: 0.10248861461877823 2023-01-21 09:43:25.404734: step: 1864/533, loss: 0.2411825954914093 2023-01-21 09:43:26.573014: step: 1868/533, loss: 0.06684742122888565 2023-01-21 09:43:27.777060: step: 1872/533, loss: 0.009373664855957031 2023-01-21 09:43:28.930599: step: 1876/533, loss: 0.13146862387657166 2023-01-21 09:43:30.106679: step: 1880/533, loss: 0.19396571815013885 2023-01-21 09:43:31.289860: step: 1884/533, loss: 0.12971439957618713 2023-01-21 09:43:32.451883: step: 1888/533, loss: 0.08772548288106918 2023-01-21 09:43:33.647763: step: 1892/533, loss: 0.0742061585187912 2023-01-21 09:43:34.831611: step: 1896/533, loss: 0.15593385696411133 2023-01-21 09:43:36.006161: step: 1900/533, loss: 0.14757271111011505 2023-01-21 09:43:37.177978: step: 1904/533, loss: 0.27709275484085083 2023-01-21 09:43:38.375585: step: 1908/533, loss: 0.12831354141235352 2023-01-21 09:43:39.556526: step: 1912/533, loss: 0.1672619879245758 2023-01-21 09:43:40.766526: step: 1916/533, loss: 0.3313051462173462 2023-01-21 09:43:41.932859: step: 1920/533, loss: 0.19065017998218536 2023-01-21 09:43:43.146983: step: 1924/533, loss: 0.06419233977794647 2023-01-21 09:43:44.346048: step: 1928/533, loss: 0.12727728486061096 2023-01-21 09:43:45.505023: step: 1932/533, loss: 0.419650137424469 2023-01-21 09:43:46.687606: step: 1936/533, loss: 0.065036341547966 2023-01-21 09:43:47.845791: step: 1940/533, loss: 0.07918167114257812 2023-01-21 09:43:49.050246: step: 1944/533, loss: 0.07085132598876953 2023-01-21 09:43:50.218520: step: 1948/533, loss: 0.0782196968793869 2023-01-21 09:43:51.425449: step: 1952/533, loss: 0.6852022409439087 2023-01-21 09:43:52.625600: step: 1956/533, loss: 0.9026672840118408 2023-01-21 09:43:53.810291: step: 1960/533, loss: 0.11578083038330078 2023-01-21 09:43:54.992204: step: 1964/533, loss: 0.1018071174621582 2023-01-21 09:43:56.181999: step: 1968/533, loss: 0.1267421692609787 2023-01-21 09:43:57.310608: step: 1972/533, loss: 0.10802315175533295 2023-01-21 09:43:58.533370: step: 1976/533, loss: 0.0272674560546875 2023-01-21 09:43:59.689195: step: 1980/533, loss: 0.05474843084812164 2023-01-21 09:44:00.885424: step: 1984/533, loss: 0.7037945985794067 2023-01-21 09:44:02.100610: step: 1988/533, loss: 0.03250541538000107 2023-01-21 09:44:03.334886: step: 1992/533, loss: 0.019031710922718048 2023-01-21 09:44:04.543055: step: 1996/533, loss: 0.15862607955932617 2023-01-21 09:44:05.750794: step: 2000/533, loss: 0.11570720374584198 2023-01-21 09:44:06.924188: step: 2004/533, loss: 0.029933834448456764 2023-01-21 09:44:08.105164: step: 2008/533, loss: 0.18323147296905518 2023-01-21 09:44:09.291154: step: 2012/533, loss: 0.07221489399671555 2023-01-21 09:44:10.492025: step: 2016/533, loss: 0.14682435989379883 2023-01-21 09:44:11.688555: step: 2020/533, loss: 0.5569009184837341 2023-01-21 09:44:12.882608: step: 2024/533, loss: 0.01458828430622816 2023-01-21 09:44:14.063622: step: 2028/533, loss: 0.05969729647040367 2023-01-21 09:44:15.259260: step: 2032/533, loss: 0.13389864563941956 2023-01-21 09:44:16.414709: step: 2036/533, loss: 0.13498875498771667 2023-01-21 09:44:17.592425: step: 2040/533, loss: 0.09449644386768341 2023-01-21 09:44:18.772785: step: 2044/533, loss: 1.1591969728469849 2023-01-21 09:44:19.979862: step: 2048/533, loss: 0.7760132551193237 2023-01-21 09:44:21.166335: step: 2052/533, loss: 0.10990545898675919 2023-01-21 09:44:22.363370: step: 2056/533, loss: 0.05424699932336807 2023-01-21 09:44:23.576205: step: 2060/533, loss: 0.12772150337696075 2023-01-21 09:44:24.781933: step: 2064/533, loss: 0.08707280457019806 2023-01-21 09:44:25.959851: step: 2068/533, loss: 0.09374499320983887 2023-01-21 09:44:27.163058: step: 2072/533, loss: 0.06009012460708618 2023-01-21 09:44:28.337836: step: 2076/533, loss: 0.08118324726819992 2023-01-21 09:44:29.487049: step: 2080/533, loss: 1.3016350269317627 2023-01-21 09:44:30.661869: step: 2084/533, loss: 0.23098011314868927 2023-01-21 09:44:31.862092: step: 2088/533, loss: 0.01660327985882759 2023-01-21 09:44:33.041334: step: 2092/533, loss: 0.1556406021118164 2023-01-21 09:44:34.234530: step: 2096/533, loss: 0.13567021489143372 2023-01-21 09:44:35.394011: step: 2100/533, loss: 0.10422177612781525 2023-01-21 09:44:36.585910: step: 2104/533, loss: 0.16072861850261688 2023-01-21 09:44:37.763642: step: 2108/533, loss: 0.005278587341308594 2023-01-21 09:44:38.969870: step: 2112/533, loss: 0.08607663959264755 2023-01-21 09:44:40.224199: step: 2116/533, loss: 0.0637931376695633 2023-01-21 09:44:41.377975: step: 2120/533, loss: 0.0776335746049881 2023-01-21 09:44:42.593766: step: 2124/533, loss: 0.20560836791992188 2023-01-21 09:44:43.755875: step: 2128/533, loss: 0.36538657546043396 2023-01-21 09:44:44.925831: step: 2132/533, loss: 0.16393613815307617 ================================================== Loss: 0.219 -------------------- Dev: {'event': {'p': 0.5718475073313783, 'r': 0.7789613848202397, 'f1': 0.6595264937993235}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Test: {'event': {'p': 0.6085657370517928, 'r': 0.7813299232736572, 'f1': 0.6842105263157895}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Chinese: {'event': {'p': 0.6233766233766234, 'r': 0.8888888888888888, 'f1': 0.732824427480916}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Korean: {'event': {'p': 0.6538461538461539, 'r': 0.5396825396825397, 'f1': 0.5913043478260869}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Russian: {'event': {'p': 0.40384615384615385, 'r': 0.5833333333333334, 'f1': 0.4772727272727273}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 09:45:26.755420: step: 4/533, loss: 0.02713499218225479 2023-01-21 09:45:27.996145: step: 8/533, loss: 0.09234371781349182 2023-01-21 09:45:29.211753: step: 12/533, loss: 0.0528598316013813 2023-01-21 09:45:30.418288: step: 16/533, loss: 0.16958141326904297 2023-01-21 09:45:31.633932: step: 20/533, loss: 0.1115323081612587 2023-01-21 09:45:32.812252: step: 24/533, loss: 0.5008176565170288 2023-01-21 09:45:33.953668: step: 28/533, loss: 0.5069162845611572 2023-01-21 09:45:35.101156: step: 32/533, loss: 0.07732424885034561 2023-01-21 09:45:36.249047: step: 36/533, loss: 0.10948900878429413 2023-01-21 09:45:37.421989: step: 40/533, loss: 0.08771447837352753 2023-01-21 09:45:38.540814: step: 44/533, loss: 0.013901353813707829 2023-01-21 09:45:39.745261: step: 48/533, loss: 0.14765433967113495 2023-01-21 09:45:40.907229: step: 52/533, loss: 0.10826416313648224 2023-01-21 09:45:42.070599: step: 56/533, loss: 0.04156780242919922 2023-01-21 09:45:43.226392: step: 60/533, loss: 0.0886441245675087 2023-01-21 09:45:44.448689: step: 64/533, loss: 0.419236958026886 2023-01-21 09:45:45.588383: step: 68/533, loss: 0.051343392580747604 2023-01-21 09:45:46.795515: step: 72/533, loss: 0.3384891450405121 2023-01-21 09:45:47.949868: step: 76/533, loss: 0.15001440048217773 2023-01-21 09:45:49.145521: step: 80/533, loss: 0.08281965553760529 2023-01-21 09:45:50.306784: step: 84/533, loss: 0.027667714282870293 2023-01-21 09:45:51.481179: step: 88/533, loss: 0.0781317725777626 2023-01-21 09:45:52.700965: step: 92/533, loss: 0.6364617347717285 2023-01-21 09:45:53.868969: step: 96/533, loss: 0.007638502400368452 2023-01-21 09:45:55.017952: step: 100/533, loss: 0.04420776292681694 2023-01-21 09:45:56.188601: step: 104/533, loss: 0.08147278428077698 2023-01-21 09:45:57.345038: step: 108/533, loss: 0.022054623812437057 2023-01-21 09:45:58.549630: step: 112/533, loss: 0.17314930260181427 2023-01-21 09:45:59.754357: step: 116/533, loss: 0.283458411693573 2023-01-21 09:46:00.954286: step: 120/533, loss: 0.046570874750614166 2023-01-21 09:46:02.167633: step: 124/533, loss: 0.027858782559633255 2023-01-21 09:46:03.310167: step: 128/533, loss: 0.01189116295427084 2023-01-21 09:46:04.485331: step: 132/533, loss: 0.5573993921279907 2023-01-21 09:46:05.670801: step: 136/533, loss: 0.07571068406105042 2023-01-21 09:46:06.906452: step: 140/533, loss: 0.13586430251598358 2023-01-21 09:46:08.092877: step: 144/533, loss: 0.018926765769720078 2023-01-21 09:46:09.252901: step: 148/533, loss: 0.0792180523276329 2023-01-21 09:46:10.434638: step: 152/533, loss: 0.062178902328014374 2023-01-21 09:46:11.604600: step: 156/533, loss: 0.1420845091342926 2023-01-21 09:46:12.865002: step: 160/533, loss: 0.1116710677742958 2023-01-21 09:46:14.030796: step: 164/533, loss: 0.11904354393482208 2023-01-21 09:46:15.214084: step: 168/533, loss: 0.7697548866271973 2023-01-21 09:46:16.413033: step: 172/533, loss: 0.33512943983078003 2023-01-21 09:46:17.614136: step: 176/533, loss: 0.6424466371536255 2023-01-21 09:46:18.793819: step: 180/533, loss: 0.04398174583911896 2023-01-21 09:46:20.007309: step: 184/533, loss: 0.11134495586156845 2023-01-21 09:46:21.190470: step: 188/533, loss: 0.11911516636610031 2023-01-21 09:46:22.362202: step: 192/533, loss: 0.1014779582619667 2023-01-21 09:46:23.580156: step: 196/533, loss: 0.21229958534240723 2023-01-21 09:46:24.774076: step: 200/533, loss: 0.1393236368894577 2023-01-21 09:46:25.977018: step: 204/533, loss: 0.09390535205602646 2023-01-21 09:46:27.139500: step: 208/533, loss: 0.09418316185474396 2023-01-21 09:46:28.359380: step: 212/533, loss: 0.1424139142036438 2023-01-21 09:46:29.597838: step: 216/533, loss: 0.07017631828784943 2023-01-21 09:46:30.785261: step: 220/533, loss: 0.27648013830184937 2023-01-21 09:46:31.970580: step: 224/533, loss: 0.21756906807422638 2023-01-21 09:46:33.165966: step: 228/533, loss: 0.10032311081886292 2023-01-21 09:46:34.325776: step: 232/533, loss: 0.8077481389045715 2023-01-21 09:46:35.501861: step: 236/533, loss: 0.0897817611694336 2023-01-21 09:46:36.717669: step: 240/533, loss: 0.11320037394762039 2023-01-21 09:46:37.892321: step: 244/533, loss: 0.5346193909645081 2023-01-21 09:46:39.073642: step: 248/533, loss: 0.26862525939941406 2023-01-21 09:46:40.239421: step: 252/533, loss: 0.096012644469738 2023-01-21 09:46:41.428626: step: 256/533, loss: 0.30831077694892883 2023-01-21 09:46:42.637062: step: 260/533, loss: 0.15489159524440765 2023-01-21 09:46:43.834323: step: 264/533, loss: 0.031189823523163795 2023-01-21 09:46:44.988564: step: 268/533, loss: 0.06381230056285858 2023-01-21 09:46:46.186462: step: 272/533, loss: 0.20449256896972656 2023-01-21 09:46:47.327245: step: 276/533, loss: 0.11977416276931763 2023-01-21 09:46:48.569200: step: 280/533, loss: 0.048119544982910156 2023-01-21 09:46:49.747291: step: 284/533, loss: 0.0661199539899826 2023-01-21 09:46:50.942762: step: 288/533, loss: 0.15234041213989258 2023-01-21 09:46:52.112157: step: 292/533, loss: 0.352247416973114 2023-01-21 09:46:53.309886: step: 296/533, loss: 0.1439352035522461 2023-01-21 09:46:54.500805: step: 300/533, loss: 0.09108009934425354 2023-01-21 09:46:55.727661: step: 304/533, loss: 0.10483226925134659 2023-01-21 09:46:56.931424: step: 308/533, loss: 0.18390808999538422 2023-01-21 09:46:58.108440: step: 312/533, loss: 0.08191729336977005 2023-01-21 09:46:59.254498: step: 316/533, loss: 0.08554497361183167 2023-01-21 09:47:00.464131: step: 320/533, loss: 0.1712614744901657 2023-01-21 09:47:01.614867: step: 324/533, loss: 0.24714070558547974 2023-01-21 09:47:02.800941: step: 328/533, loss: 0.06959524005651474 2023-01-21 09:47:03.990574: step: 332/533, loss: 0.04762458801269531 2023-01-21 09:47:05.183890: step: 336/533, loss: 0.36471718549728394 2023-01-21 09:47:06.353408: step: 340/533, loss: 0.055223919451236725 2023-01-21 09:47:07.546195: step: 344/533, loss: 0.021492719650268555 2023-01-21 09:47:08.723489: step: 348/533, loss: 0.4865197241306305 2023-01-21 09:47:09.917087: step: 352/533, loss: 0.40627679228782654 2023-01-21 09:47:11.064245: step: 356/533, loss: 0.5815869569778442 2023-01-21 09:47:12.261997: step: 360/533, loss: 0.13450995087623596 2023-01-21 09:47:13.439935: step: 364/533, loss: 0.13007330894470215 2023-01-21 09:47:14.664809: step: 368/533, loss: 0.03940916061401367 2023-01-21 09:47:15.843241: step: 372/533, loss: 0.13993777334690094 2023-01-21 09:47:17.040709: step: 376/533, loss: 0.8265101313591003 2023-01-21 09:47:18.218578: step: 380/533, loss: 0.34481391310691833 2023-01-21 09:47:19.413855: step: 384/533, loss: 0.19091807305812836 2023-01-21 09:47:20.588810: step: 388/533, loss: 0.4733350872993469 2023-01-21 09:47:21.791509: step: 392/533, loss: 0.11199293285608292 2023-01-21 09:47:22.956666: step: 396/533, loss: 0.06990843266248703 2023-01-21 09:47:24.130599: step: 400/533, loss: 0.1847863495349884 2023-01-21 09:47:25.318241: step: 404/533, loss: 0.11941175907850266 2023-01-21 09:47:26.488139: step: 408/533, loss: 0.06626243889331818 2023-01-21 09:47:27.732251: step: 412/533, loss: 0.10838775336742401 2023-01-21 09:47:28.917831: step: 416/533, loss: 0.013866320252418518 2023-01-21 09:47:30.115997: step: 420/533, loss: 0.10283956676721573 2023-01-21 09:47:31.335873: step: 424/533, loss: 0.09228868782520294 2023-01-21 09:47:32.510388: step: 428/533, loss: 0.5761860013008118 2023-01-21 09:47:33.719419: step: 432/533, loss: 0.06501880288124084 2023-01-21 09:47:34.885746: step: 436/533, loss: 0.10980406403541565 2023-01-21 09:47:36.022933: step: 440/533, loss: 0.06611061096191406 2023-01-21 09:47:37.215545: step: 444/533, loss: 0.030341625213623047 2023-01-21 09:47:38.382387: step: 448/533, loss: 0.050723735243082047 2023-01-21 09:47:39.601500: step: 452/533, loss: 0.2815723419189453 2023-01-21 09:47:40.806341: step: 456/533, loss: 0.12159442901611328 2023-01-21 09:47:41.994937: step: 460/533, loss: 0.13366861641407013 2023-01-21 09:47:43.170513: step: 464/533, loss: 0.13840627670288086 2023-01-21 09:47:44.380182: step: 468/533, loss: 0.2861112356185913 2023-01-21 09:47:45.592200: step: 472/533, loss: 0.1198093444108963 2023-01-21 09:47:46.788151: step: 476/533, loss: 0.15332946181297302 2023-01-21 09:47:47.971799: step: 480/533, loss: 0.18241333961486816 2023-01-21 09:47:49.169084: step: 484/533, loss: 0.0022195817437022924 2023-01-21 09:47:50.315388: step: 488/533, loss: 0.08952093124389648 2023-01-21 09:47:51.485793: step: 492/533, loss: 0.09961710125207901 2023-01-21 09:47:52.646996: step: 496/533, loss: 0.20637111365795135 2023-01-21 09:47:53.831864: step: 500/533, loss: 0.16375738382339478 2023-01-21 09:47:55.017647: step: 504/533, loss: 0.34664592146873474 2023-01-21 09:47:56.195636: step: 508/533, loss: 0.1622917205095291 2023-01-21 09:47:57.434375: step: 512/533, loss: 0.10765047371387482 2023-01-21 09:47:58.605297: step: 516/533, loss: 0.153484046459198 2023-01-21 09:47:59.779615: step: 520/533, loss: 0.03779182583093643 2023-01-21 09:48:01.000508: step: 524/533, loss: 2.8432631492614746 2023-01-21 09:48:02.188415: step: 528/533, loss: 0.04412193223834038 2023-01-21 09:48:03.404655: step: 532/533, loss: 0.12501993775367737 2023-01-21 09:48:04.570971: step: 536/533, loss: 0.08911094814538956 2023-01-21 09:48:05.730848: step: 540/533, loss: 0.04807615280151367 2023-01-21 09:48:06.915596: step: 544/533, loss: 0.13882999122142792 2023-01-21 09:48:08.097927: step: 548/533, loss: 0.005447817035019398 2023-01-21 09:48:09.282449: step: 552/533, loss: 0.12212295830249786 2023-01-21 09:48:10.453028: step: 556/533, loss: 0.060848332941532135 2023-01-21 09:48:11.636311: step: 560/533, loss: 0.04181380197405815 2023-01-21 09:48:12.800887: step: 564/533, loss: 0.14441928267478943 2023-01-21 09:48:13.996142: step: 568/533, loss: 0.08375763893127441 2023-01-21 09:48:15.151158: step: 572/533, loss: 0.1010388433933258 2023-01-21 09:48:16.342191: step: 576/533, loss: 0.12010860443115234 2023-01-21 09:48:17.520850: step: 580/533, loss: 0.14491453766822815 2023-01-21 09:48:18.709763: step: 584/533, loss: 0.3436940312385559 2023-01-21 09:48:19.898181: step: 588/533, loss: 0.2553713917732239 2023-01-21 09:48:21.047264: step: 592/533, loss: 0.6016233563423157 2023-01-21 09:48:22.242531: step: 596/533, loss: 0.02080359496176243 2023-01-21 09:48:23.420680: step: 600/533, loss: 0.4463450014591217 2023-01-21 09:48:24.585435: step: 604/533, loss: 0.7715851664543152 2023-01-21 09:48:25.774230: step: 608/533, loss: 0.02622537687420845 2023-01-21 09:48:26.971192: step: 612/533, loss: 0.4134342074394226 2023-01-21 09:48:28.200358: step: 616/533, loss: 0.3035464286804199 2023-01-21 09:48:29.393378: step: 620/533, loss: 0.04574203863739967 2023-01-21 09:48:30.625668: step: 624/533, loss: 0.0404236800968647 2023-01-21 09:48:31.823241: step: 628/533, loss: 0.11040210723876953 2023-01-21 09:48:32.974817: step: 632/533, loss: 0.062007904052734375 2023-01-21 09:48:34.109494: step: 636/533, loss: 0.07534551620483398 2023-01-21 09:48:35.251864: step: 640/533, loss: 0.08489511162042618 2023-01-21 09:48:36.380839: step: 644/533, loss: 0.00962753314524889 2023-01-21 09:48:37.564550: step: 648/533, loss: 0.09941711276769638 2023-01-21 09:48:38.741847: step: 652/533, loss: 0.11434593051671982 2023-01-21 09:48:39.944003: step: 656/533, loss: 0.8069518208503723 2023-01-21 09:48:41.159179: step: 660/533, loss: 0.15600843727588654 2023-01-21 09:48:42.314492: step: 664/533, loss: 0.07689161598682404 2023-01-21 09:48:43.493576: step: 668/533, loss: 0.09479685872793198 2023-01-21 09:48:44.685476: step: 672/533, loss: 0.10828132182359695 2023-01-21 09:48:45.897885: step: 676/533, loss: 0.05332536622881889 2023-01-21 09:48:47.042898: step: 680/533, loss: 0.26875317096710205 2023-01-21 09:48:48.199110: step: 684/533, loss: 0.05133800581097603 2023-01-21 09:48:49.392196: step: 688/533, loss: 0.035854484885931015 2023-01-21 09:48:50.558936: step: 692/533, loss: 0.09907674789428711 2023-01-21 09:48:51.745776: step: 696/533, loss: 0.14481498301029205 2023-01-21 09:48:52.920292: step: 700/533, loss: 0.07191634178161621 2023-01-21 09:48:54.072768: step: 704/533, loss: 0.14179524779319763 2023-01-21 09:48:55.205591: step: 708/533, loss: 0.05259838327765465 2023-01-21 09:48:56.364197: step: 712/533, loss: 0.006909561343491077 2023-01-21 09:48:57.551849: step: 716/533, loss: 0.06766758114099503 2023-01-21 09:48:58.742578: step: 720/533, loss: 0.095367431640625 2023-01-21 09:48:59.954794: step: 724/533, loss: 0.3103080689907074 2023-01-21 09:49:01.186362: step: 728/533, loss: 0.06444225460290909 2023-01-21 09:49:02.411203: step: 732/533, loss: 0.24143868684768677 2023-01-21 09:49:03.578263: step: 736/533, loss: 0.07462029904127121 2023-01-21 09:49:04.802523: step: 740/533, loss: 0.17867432534694672 2023-01-21 09:49:05.961230: step: 744/533, loss: 0.0874989926815033 2023-01-21 09:49:07.179824: step: 748/533, loss: 0.04517421871423721 2023-01-21 09:49:08.353962: step: 752/533, loss: 0.3197963535785675 2023-01-21 09:49:09.538676: step: 756/533, loss: 0.062395237386226654 2023-01-21 09:49:10.710395: step: 760/533, loss: 0.007513904944062233 2023-01-21 09:49:11.880662: step: 764/533, loss: 0.06632880866527557 2023-01-21 09:49:13.090238: step: 768/533, loss: 0.014541245065629482 2023-01-21 09:49:14.243910: step: 772/533, loss: 0.20517253875732422 2023-01-21 09:49:15.476015: step: 776/533, loss: 0.04048910364508629 2023-01-21 09:49:16.664559: step: 780/533, loss: 0.1029973104596138 2023-01-21 09:49:17.850715: step: 784/533, loss: 0.41770344972610474 2023-01-21 09:49:19.017695: step: 788/533, loss: 0.13063812255859375 2023-01-21 09:49:20.176805: step: 792/533, loss: 0.058751873672008514 2023-01-21 09:49:21.382385: step: 796/533, loss: 0.1576654464006424 2023-01-21 09:49:22.570735: step: 800/533, loss: 0.020013142377138138 2023-01-21 09:49:23.754702: step: 804/533, loss: 0.17426013946533203 2023-01-21 09:49:24.950865: step: 808/533, loss: 0.10956916958093643 2023-01-21 09:49:26.117266: step: 812/533, loss: 0.031053420156240463 2023-01-21 09:49:27.328413: step: 816/533, loss: 0.040668584406375885 2023-01-21 09:49:28.486218: step: 820/533, loss: 0.04122123867273331 2023-01-21 09:49:29.694040: step: 824/533, loss: 0.0990133285522461 2023-01-21 09:49:30.828754: step: 828/533, loss: 0.14162521064281464 2023-01-21 09:49:32.007300: step: 832/533, loss: 0.11816645413637161 2023-01-21 09:49:33.179097: step: 836/533, loss: 0.05524645000696182 2023-01-21 09:49:34.328491: step: 840/533, loss: 0.0916467159986496 2023-01-21 09:49:35.505483: step: 844/533, loss: 0.1974247395992279 2023-01-21 09:49:36.686995: step: 848/533, loss: 0.7947461605072021 2023-01-21 09:49:37.860282: step: 852/533, loss: 0.15344282984733582 2023-01-21 09:49:39.085662: step: 856/533, loss: 0.3938547968864441 2023-01-21 09:49:40.279480: step: 860/533, loss: 0.06485573947429657 2023-01-21 09:49:41.450158: step: 864/533, loss: 0.05940370634198189 2023-01-21 09:49:42.643725: step: 868/533, loss: 0.022136593237519264 2023-01-21 09:49:43.854500: step: 872/533, loss: 0.12865644693374634 2023-01-21 09:49:45.019378: step: 876/533, loss: 0.3195739686489105 2023-01-21 09:49:46.214066: step: 880/533, loss: 0.39717569947242737 2023-01-21 09:49:47.409365: step: 884/533, loss: 0.05437278747558594 2023-01-21 09:49:48.554251: step: 888/533, loss: 0.3693615794181824 2023-01-21 09:49:49.747651: step: 892/533, loss: 0.4145874083042145 2023-01-21 09:49:50.953850: step: 896/533, loss: 0.10774698108434677 2023-01-21 09:49:52.120865: step: 900/533, loss: 0.01544121466577053 2023-01-21 09:49:53.278613: step: 904/533, loss: 0.35072413086891174 2023-01-21 09:49:54.454127: step: 908/533, loss: 0.399954229593277 2023-01-21 09:49:55.636882: step: 912/533, loss: 0.049668122082948685 2023-01-21 09:49:56.821949: step: 916/533, loss: 0.17268256843090057 2023-01-21 09:49:58.019641: step: 920/533, loss: 0.614274263381958 2023-01-21 09:49:59.199461: step: 924/533, loss: 0.17506122589111328 2023-01-21 09:50:00.396199: step: 928/533, loss: 0.7478906512260437 2023-01-21 09:50:01.567160: step: 932/533, loss: 0.12009774148464203 2023-01-21 09:50:02.775727: step: 936/533, loss: 0.22066831588745117 2023-01-21 09:50:03.939726: step: 940/533, loss: 0.31020450592041016 2023-01-21 09:50:05.139888: step: 944/533, loss: 0.04259634017944336 2023-01-21 09:50:06.376238: step: 948/533, loss: 0.03404679149389267 2023-01-21 09:50:07.608205: step: 952/533, loss: 0.11127567291259766 2023-01-21 09:50:08.803909: step: 956/533, loss: 0.049219321459531784 2023-01-21 09:50:10.009403: step: 960/533, loss: 0.14533177018165588 2023-01-21 09:50:11.187125: step: 964/533, loss: 0.059020426124334335 2023-01-21 09:50:12.409848: step: 968/533, loss: 0.12678509950637817 2023-01-21 09:50:13.617648: step: 972/533, loss: 0.055660344660282135 2023-01-21 09:50:14.832590: step: 976/533, loss: 0.09729135036468506 2023-01-21 09:50:16.048343: step: 980/533, loss: 0.14687786996364594 2023-01-21 09:50:17.261187: step: 984/533, loss: 0.06933341175317764 2023-01-21 09:50:18.445832: step: 988/533, loss: 0.15269014239311218 2023-01-21 09:50:19.607084: step: 992/533, loss: 0.03150606155395508 2023-01-21 09:50:20.809605: step: 996/533, loss: 0.2112255096435547 2023-01-21 09:50:21.975501: step: 1000/533, loss: 0.07571356743574142 2023-01-21 09:50:23.164206: step: 1004/533, loss: 0.031698036938905716 2023-01-21 09:50:24.365758: step: 1008/533, loss: 0.025605155155062675 2023-01-21 09:50:25.596634: step: 1012/533, loss: 0.021311093121767044 2023-01-21 09:50:26.766405: step: 1016/533, loss: 0.26800423860549927 2023-01-21 09:50:27.988608: step: 1020/533, loss: 0.18607187271118164 2023-01-21 09:50:29.179843: step: 1024/533, loss: 0.09499606490135193 2023-01-21 09:50:30.364141: step: 1028/533, loss: 0.04922132194042206 2023-01-21 09:50:31.552289: step: 1032/533, loss: 0.060540199279785156 2023-01-21 09:50:32.692365: step: 1036/533, loss: 0.013698672875761986 2023-01-21 09:50:33.916640: step: 1040/533, loss: 0.08650960773229599 2023-01-21 09:50:35.096085: step: 1044/533, loss: 0.08960485458374023 2023-01-21 09:50:36.278260: step: 1048/533, loss: 0.16634587943553925 2023-01-21 09:50:37.463484: step: 1052/533, loss: 0.14437341690063477 2023-01-21 09:50:38.657411: step: 1056/533, loss: 0.07751629501581192 2023-01-21 09:50:39.865087: step: 1060/533, loss: 0.08586493134498596 2023-01-21 09:50:41.012437: step: 1064/533, loss: 0.1762179434299469 2023-01-21 09:50:42.165495: step: 1068/533, loss: 0.03575735166668892 2023-01-21 09:50:43.421277: step: 1072/533, loss: 0.11446776241064072 2023-01-21 09:50:44.627848: step: 1076/533, loss: 0.15126128494739532 2023-01-21 09:50:45.773081: step: 1080/533, loss: 0.14959660172462463 2023-01-21 09:50:46.940517: step: 1084/533, loss: 0.07745137065649033 2023-01-21 09:50:48.127704: step: 1088/533, loss: 0.005724430549889803 2023-01-21 09:50:49.307247: step: 1092/533, loss: 0.5174845457077026 2023-01-21 09:50:50.489852: step: 1096/533, loss: 0.1122356504201889 2023-01-21 09:50:51.658736: step: 1100/533, loss: 0.10003185272216797 2023-01-21 09:50:52.790150: step: 1104/533, loss: 0.005584716796875 2023-01-21 09:50:54.009678: step: 1108/533, loss: 0.7720202207565308 2023-01-21 09:50:55.203508: step: 1112/533, loss: 0.5403192043304443 2023-01-21 09:50:56.384564: step: 1116/533, loss: 0.35966265201568604 2023-01-21 09:50:57.591262: step: 1120/533, loss: 0.12716464698314667 2023-01-21 09:50:58.791854: step: 1124/533, loss: 0.17202362418174744 2023-01-21 09:50:59.960250: step: 1128/533, loss: 0.016559697687625885 2023-01-21 09:51:01.119077: step: 1132/533, loss: 0.15614385902881622 2023-01-21 09:51:02.344951: step: 1136/533, loss: 0.9566184878349304 2023-01-21 09:51:03.593946: step: 1140/533, loss: 0.1273021250963211 2023-01-21 09:51:04.766503: step: 1144/533, loss: 0.0437101349234581 2023-01-21 09:51:05.953618: step: 1148/533, loss: 0.03266763687133789 2023-01-21 09:51:07.149675: step: 1152/533, loss: 0.06007976830005646 2023-01-21 09:51:08.347053: step: 1156/533, loss: 0.05297403410077095 2023-01-21 09:51:09.513759: step: 1160/533, loss: 0.06830110400915146 2023-01-21 09:51:10.692198: step: 1164/533, loss: 0.07320580631494522 2023-01-21 09:51:11.915796: step: 1168/533, loss: 0.40680330991744995 2023-01-21 09:51:13.091815: step: 1172/533, loss: 0.13231736421585083 2023-01-21 09:51:14.240125: step: 1176/533, loss: 0.018236160278320312 2023-01-21 09:51:15.458808: step: 1180/533, loss: 0.10143108665943146 2023-01-21 09:51:16.612199: step: 1184/533, loss: 0.26360803842544556 2023-01-21 09:51:17.776648: step: 1188/533, loss: 0.1318511962890625 2023-01-21 09:51:18.909851: step: 1192/533, loss: 0.05794224888086319 2023-01-21 09:51:20.106023: step: 1196/533, loss: 0.2904471457004547 2023-01-21 09:51:21.320404: step: 1200/533, loss: 0.08380518108606339 2023-01-21 09:51:22.539049: step: 1204/533, loss: 0.19169996678829193 2023-01-21 09:51:23.689043: step: 1208/533, loss: 0.134568989276886 2023-01-21 09:51:24.879833: step: 1212/533, loss: 0.08094777911901474 2023-01-21 09:51:26.076396: step: 1216/533, loss: 0.2527502179145813 2023-01-21 09:51:27.244834: step: 1220/533, loss: 0.09208475798368454 2023-01-21 09:51:28.440957: step: 1224/533, loss: 0.25244027376174927 2023-01-21 09:51:29.656597: step: 1228/533, loss: 0.06860509514808655 2023-01-21 09:51:30.818982: step: 1232/533, loss: 0.1862872987985611 2023-01-21 09:51:32.028447: step: 1236/533, loss: 0.7058597207069397 2023-01-21 09:51:33.196844: step: 1240/533, loss: 0.02703228034079075 2023-01-21 09:51:34.409671: step: 1244/533, loss: 0.689183235168457 2023-01-21 09:51:35.597131: step: 1248/533, loss: 0.10156326740980148 2023-01-21 09:51:36.761013: step: 1252/533, loss: 0.11410484462976456 2023-01-21 09:51:37.993819: step: 1256/533, loss: 0.14767113327980042 2023-01-21 09:51:39.180565: step: 1260/533, loss: 0.0669654831290245 2023-01-21 09:51:40.359459: step: 1264/533, loss: 0.17472442984580994 2023-01-21 09:51:41.569405: step: 1268/533, loss: 0.07428856194019318 2023-01-21 09:51:42.748244: step: 1272/533, loss: 0.0847872719168663 2023-01-21 09:51:43.927584: step: 1276/533, loss: 0.019122647121548653 2023-01-21 09:51:45.106402: step: 1280/533, loss: 0.07606898248195648 2023-01-21 09:51:46.325291: step: 1284/533, loss: 0.6748707294464111 2023-01-21 09:51:47.537524: step: 1288/533, loss: 0.015431642532348633 2023-01-21 09:51:48.751203: step: 1292/533, loss: 0.19803038239479065 2023-01-21 09:51:49.911705: step: 1296/533, loss: 0.011018085293471813 2023-01-21 09:51:51.142444: step: 1300/533, loss: 0.2448538839817047 2023-01-21 09:51:52.290857: step: 1304/533, loss: 0.04956841468811035 2023-01-21 09:51:53.527698: step: 1308/533, loss: 0.039197683334350586 2023-01-21 09:51:54.731348: step: 1312/533, loss: 0.07702169567346573 2023-01-21 09:51:55.897423: step: 1316/533, loss: 0.2669532895088196 2023-01-21 09:51:57.064079: step: 1320/533, loss: 0.7712525725364685 2023-01-21 09:51:58.242039: step: 1324/533, loss: 0.2272128164768219 2023-01-21 09:51:59.438235: step: 1328/533, loss: 0.05593233183026314 2023-01-21 09:52:00.656047: step: 1332/533, loss: 0.04806847870349884 2023-01-21 09:52:01.837610: step: 1336/533, loss: 0.03831439092755318 2023-01-21 09:52:03.049722: step: 1340/533, loss: 0.09763441234827042 2023-01-21 09:52:04.263956: step: 1344/533, loss: 0.19350768625736237 2023-01-21 09:52:05.451648: step: 1348/533, loss: 0.22850441932678223 2023-01-21 09:52:06.746895: step: 1352/533, loss: 0.20074662566184998 2023-01-21 09:52:07.926188: step: 1356/533, loss: 0.2425546646118164 2023-01-21 09:52:09.102290: step: 1360/533, loss: 0.017459489405155182 2023-01-21 09:52:10.265284: step: 1364/533, loss: 0.12814387679100037 2023-01-21 09:52:11.479161: step: 1368/533, loss: 0.13456392288208008 2023-01-21 09:52:12.655307: step: 1372/533, loss: 0.02603388950228691 2023-01-21 09:52:13.821081: step: 1376/533, loss: 0.023082830011844635 2023-01-21 09:52:15.039097: step: 1380/533, loss: 0.85345458984375 2023-01-21 09:52:16.221399: step: 1384/533, loss: 0.05691461265087128 2023-01-21 09:52:17.426812: step: 1388/533, loss: 0.08820381760597229 2023-01-21 09:52:18.635918: step: 1392/533, loss: 0.6441293954849243 2023-01-21 09:52:19.830317: step: 1396/533, loss: 0.1557263433933258 2023-01-21 09:52:20.982518: step: 1400/533, loss: 2.3298323154449463 2023-01-21 09:52:22.133863: step: 1404/533, loss: 0.05295753479003906 2023-01-21 09:52:23.311694: step: 1408/533, loss: 0.09047737717628479 2023-01-21 09:52:24.573465: step: 1412/533, loss: 0.07927742600440979 2023-01-21 09:52:25.751232: step: 1416/533, loss: 0.01557533722370863 2023-01-21 09:52:26.938873: step: 1420/533, loss: 0.17565536499023438 2023-01-21 09:52:28.177103: step: 1424/533, loss: 0.12403789162635803 2023-01-21 09:52:29.369917: step: 1428/533, loss: 0.06530608981847763 2023-01-21 09:52:30.499863: step: 1432/533, loss: 0.11824546009302139 2023-01-21 09:52:31.674084: step: 1436/533, loss: 0.061544135212898254 2023-01-21 09:52:32.856907: step: 1440/533, loss: 0.135945126414299 2023-01-21 09:52:34.073905: step: 1444/533, loss: 0.33694860339164734 2023-01-21 09:52:35.254052: step: 1448/533, loss: 0.16456662118434906 2023-01-21 09:52:36.413446: step: 1452/533, loss: 0.0867166519165039 2023-01-21 09:52:37.590480: step: 1456/533, loss: 0.13213959336280823 2023-01-21 09:52:38.821215: step: 1460/533, loss: 0.07199926674365997 2023-01-21 09:52:39.978575: step: 1464/533, loss: 0.29781877994537354 2023-01-21 09:52:41.193252: step: 1468/533, loss: 0.056651026010513306 2023-01-21 09:52:42.372366: step: 1472/533, loss: 0.3756037652492523 2023-01-21 09:52:43.569333: step: 1476/533, loss: 0.14920340478420258 2023-01-21 09:52:44.762720: step: 1480/533, loss: 0.10907058417797089 2023-01-21 09:52:45.942689: step: 1484/533, loss: 0.3637380301952362 2023-01-21 09:52:47.140798: step: 1488/533, loss: 0.922110915184021 2023-01-21 09:52:48.330153: step: 1492/533, loss: 0.22944754362106323 2023-01-21 09:52:49.525221: step: 1496/533, loss: 0.01664004474878311 2023-01-21 09:52:50.711942: step: 1500/533, loss: 0.13072019815444946 2023-01-21 09:52:51.959194: step: 1504/533, loss: 0.14062805473804474 2023-01-21 09:52:53.146410: step: 1508/533, loss: 0.09868965297937393 2023-01-21 09:52:54.361975: step: 1512/533, loss: 0.2841409742832184 2023-01-21 09:52:55.563154: step: 1516/533, loss: 0.07551197707653046 2023-01-21 09:52:56.837319: step: 1520/533, loss: 0.26093220710754395 2023-01-21 09:52:58.041760: step: 1524/533, loss: 0.0073053836822509766 2023-01-21 09:52:59.183091: step: 1528/533, loss: 0.41684991121292114 2023-01-21 09:53:00.370864: step: 1532/533, loss: 0.077867791056633 2023-01-21 09:53:01.559611: step: 1536/533, loss: 0.2675896883010864 2023-01-21 09:53:02.772157: step: 1540/533, loss: 0.1291884481906891 2023-01-21 09:53:03.945736: step: 1544/533, loss: 0.06301793456077576 2023-01-21 09:53:05.169176: step: 1548/533, loss: 0.1364070028066635 2023-01-21 09:53:06.322083: step: 1552/533, loss: 0.10266055911779404 2023-01-21 09:53:07.517960: step: 1556/533, loss: 0.008716249838471413 2023-01-21 09:53:08.726365: step: 1560/533, loss: 0.3959083557128906 2023-01-21 09:53:09.872154: step: 1564/533, loss: 0.03552446514368057 2023-01-21 09:53:11.047629: step: 1568/533, loss: 0.061426304280757904 2023-01-21 09:53:12.257897: step: 1572/533, loss: 0.006771659944206476 2023-01-21 09:53:13.412328: step: 1576/533, loss: 0.09923876821994781 2023-01-21 09:53:14.576862: step: 1580/533, loss: 0.09676344692707062 2023-01-21 09:53:15.739590: step: 1584/533, loss: 0.0769491195678711 2023-01-21 09:53:16.929351: step: 1588/533, loss: 0.14762984216213226 2023-01-21 09:53:18.112749: step: 1592/533, loss: 0.0663335770368576 2023-01-21 09:53:19.343977: step: 1596/533, loss: 0.6944124698638916 2023-01-21 09:53:20.591243: step: 1600/533, loss: 0.08068704605102539 2023-01-21 09:53:21.758817: step: 1604/533, loss: 0.07248697429895401 2023-01-21 09:53:22.931891: step: 1608/533, loss: 0.12503671646118164 2023-01-21 09:53:24.086475: step: 1612/533, loss: 0.0302626620978117 2023-01-21 09:53:25.242756: step: 1616/533, loss: 0.06951122730970383 2023-01-21 09:53:26.439637: step: 1620/533, loss: 0.43427911400794983 2023-01-21 09:53:27.606676: step: 1624/533, loss: 0.09474726021289825 2023-01-21 09:53:28.786426: step: 1628/533, loss: 0.026811406016349792 2023-01-21 09:53:29.949240: step: 1632/533, loss: 0.11235599219799042 2023-01-21 09:53:31.152199: step: 1636/533, loss: 0.10576677322387695 2023-01-21 09:53:32.335116: step: 1640/533, loss: 0.14707250893115997 2023-01-21 09:53:33.528202: step: 1644/533, loss: 0.023218631744384766 2023-01-21 09:53:34.718932: step: 1648/533, loss: 0.04017763212323189 2023-01-21 09:53:35.913416: step: 1652/533, loss: 0.2986489236354828 2023-01-21 09:53:37.065074: step: 1656/533, loss: 0.06629820168018341 2023-01-21 09:53:38.242980: step: 1660/533, loss: 0.25984689593315125 2023-01-21 09:53:39.416961: step: 1664/533, loss: 0.10866174846887589 2023-01-21 09:53:40.579363: step: 1668/533, loss: 0.46603700518608093 2023-01-21 09:53:41.754829: step: 1672/533, loss: 0.05942163243889809 2023-01-21 09:53:42.912599: step: 1676/533, loss: 0.11981496959924698 2023-01-21 09:53:44.048702: step: 1680/533, loss: 0.17088022828102112 2023-01-21 09:53:45.259870: step: 1684/533, loss: 0.05519905686378479 2023-01-21 09:53:46.433274: step: 1688/533, loss: 0.06767778098583221 2023-01-21 09:53:47.622986: step: 1692/533, loss: 0.03444407135248184 2023-01-21 09:53:48.806829: step: 1696/533, loss: 0.0202044490724802 2023-01-21 09:53:49.993205: step: 1700/533, loss: 0.016474580392241478 2023-01-21 09:53:51.195570: step: 1704/533, loss: 0.09185848385095596 2023-01-21 09:53:52.344029: step: 1708/533, loss: 0.23659612238407135 2023-01-21 09:53:53.534476: step: 1712/533, loss: 0.5814775228500366 2023-01-21 09:53:54.717667: step: 1716/533, loss: 0.07386360317468643 2023-01-21 09:53:55.920368: step: 1720/533, loss: 1.4356608390808105 2023-01-21 09:53:57.115238: step: 1724/533, loss: 0.08315916359424591 2023-01-21 09:53:58.295761: step: 1728/533, loss: 0.34634724259376526 2023-01-21 09:53:59.432634: step: 1732/533, loss: 0.060264021158218384 2023-01-21 09:54:00.615165: step: 1736/533, loss: 0.0956021100282669 2023-01-21 09:54:01.779776: step: 1740/533, loss: 0.10569701343774796 2023-01-21 09:54:03.011100: step: 1744/533, loss: 0.14919549226760864 2023-01-21 09:54:04.192588: step: 1748/533, loss: 0.13188667595386505 2023-01-21 09:54:05.414100: step: 1752/533, loss: 0.13377352058887482 2023-01-21 09:54:06.623259: step: 1756/533, loss: 0.09226135909557343 2023-01-21 09:54:07.815226: step: 1760/533, loss: 0.12719126045703888 2023-01-21 09:54:09.009911: step: 1764/533, loss: 0.13516291975975037 2023-01-21 09:54:10.205838: step: 1768/533, loss: 0.14136266708374023 2023-01-21 09:54:11.382318: step: 1772/533, loss: 0.08780984580516815 2023-01-21 09:54:12.569128: step: 1776/533, loss: 0.09126882255077362 2023-01-21 09:54:13.784281: step: 1780/533, loss: 0.19873599708080292 2023-01-21 09:54:14.975238: step: 1784/533, loss: 0.16518214344978333 2023-01-21 09:54:16.186037: step: 1788/533, loss: 0.08995547890663147 2023-01-21 09:54:17.424525: step: 1792/533, loss: 0.1451520472764969 2023-01-21 09:54:18.615115: step: 1796/533, loss: 0.5091859698295593 2023-01-21 09:54:19.813919: step: 1800/533, loss: 0.3400367796421051 2023-01-21 09:54:20.995988: step: 1804/533, loss: 0.5751129388809204 2023-01-21 09:54:22.158036: step: 1808/533, loss: 0.13906565308570862 2023-01-21 09:54:23.355088: step: 1812/533, loss: 0.04004368931055069 2023-01-21 09:54:24.549599: step: 1816/533, loss: 0.5197814702987671 2023-01-21 09:54:25.739309: step: 1820/533, loss: 0.058991577476263046 2023-01-21 09:54:26.893562: step: 1824/533, loss: 0.19924607872962952 2023-01-21 09:54:28.133227: step: 1828/533, loss: 0.4027549922466278 2023-01-21 09:54:29.359440: step: 1832/533, loss: 0.25962895154953003 2023-01-21 09:54:30.535739: step: 1836/533, loss: 0.022650720551609993 2023-01-21 09:54:31.697620: step: 1840/533, loss: 0.10708286613225937 2023-01-21 09:54:32.877673: step: 1844/533, loss: 0.10166168212890625 2023-01-21 09:54:34.036013: step: 1848/533, loss: 0.02556152269244194 2023-01-21 09:54:35.233033: step: 1852/533, loss: 0.4256401062011719 2023-01-21 09:54:36.433439: step: 1856/533, loss: 0.06053295359015465 2023-01-21 09:54:37.627702: step: 1860/533, loss: 0.1242123693227768 2023-01-21 09:54:38.825653: step: 1864/533, loss: 0.2137494683265686 2023-01-21 09:54:40.015594: step: 1868/533, loss: 0.10713911056518555 2023-01-21 09:54:41.189338: step: 1872/533, loss: 0.13193221390247345 2023-01-21 09:54:42.377181: step: 1876/533, loss: 0.400832861661911 2023-01-21 09:54:43.530528: step: 1880/533, loss: 0.06214485317468643 2023-01-21 09:54:44.722093: step: 1884/533, loss: 0.13905023038387299 2023-01-21 09:54:45.881862: step: 1888/533, loss: 0.04362001270055771 2023-01-21 09:54:47.080918: step: 1892/533, loss: 0.3122302293777466 2023-01-21 09:54:48.271800: step: 1896/533, loss: 0.03895587846636772 2023-01-21 09:54:49.430693: step: 1900/533, loss: 0.15119047462940216 2023-01-21 09:54:50.640062: step: 1904/533, loss: 0.6767929196357727 2023-01-21 09:54:51.818153: step: 1908/533, loss: 0.25180110335350037 2023-01-21 09:54:53.005245: step: 1912/533, loss: 0.1086040511727333 2023-01-21 09:54:54.216459: step: 1916/533, loss: 0.19573555886745453 2023-01-21 09:54:55.356582: step: 1920/533, loss: 0.16889649629592896 2023-01-21 09:54:56.525830: step: 1924/533, loss: 0.3492767810821533 2023-01-21 09:54:57.676126: step: 1928/533, loss: 0.060495663434267044 2023-01-21 09:54:58.864432: step: 1932/533, loss: 0.26421308517456055 2023-01-21 09:55:00.066251: step: 1936/533, loss: 0.2601969838142395 2023-01-21 09:55:01.223755: step: 1940/533, loss: 0.09201221913099289 2023-01-21 09:55:02.424234: step: 1944/533, loss: 0.15824276208877563 2023-01-21 09:55:03.612594: step: 1948/533, loss: 0.10685815662145615 2023-01-21 09:55:04.800683: step: 1952/533, loss: 0.030006123706698418 2023-01-21 09:55:05.996757: step: 1956/533, loss: 0.3846948742866516 2023-01-21 09:55:07.196752: step: 1960/533, loss: 0.07690100371837616 2023-01-21 09:55:08.347162: step: 1964/533, loss: 0.44955766201019287 2023-01-21 09:55:09.540325: step: 1968/533, loss: 0.13609905540943146 2023-01-21 09:55:10.725287: step: 1972/533, loss: 0.11102294921875 2023-01-21 09:55:11.950763: step: 1976/533, loss: 0.12062263488769531 2023-01-21 09:55:13.103203: step: 1980/533, loss: 0.32266101241111755 2023-01-21 09:55:14.296827: step: 1984/533, loss: 0.09616956859827042 2023-01-21 09:55:15.504705: step: 1988/533, loss: 0.1433626264333725 2023-01-21 09:55:16.676374: step: 1992/533, loss: 0.02000696025788784 2023-01-21 09:55:17.798789: step: 1996/533, loss: 0.021545404568314552 2023-01-21 09:55:19.021104: step: 2000/533, loss: 0.0699489563703537 2023-01-21 09:55:20.180527: step: 2004/533, loss: 0.12893332540988922 2023-01-21 09:55:21.368359: step: 2008/533, loss: 0.05516872555017471 2023-01-21 09:55:22.530512: step: 2012/533, loss: 0.07231850922107697 2023-01-21 09:55:23.714104: step: 2016/533, loss: 0.11212797462940216 2023-01-21 09:55:24.919730: step: 2020/533, loss: 0.04125042259693146 2023-01-21 09:55:26.131543: step: 2024/533, loss: 0.13496848940849304 2023-01-21 09:55:27.327737: step: 2028/533, loss: 0.04353313520550728 2023-01-21 09:55:28.471979: step: 2032/533, loss: 0.04877294600009918 2023-01-21 09:55:29.654524: step: 2036/533, loss: 0.08072672039270401 2023-01-21 09:55:30.816424: step: 2040/533, loss: 0.4876982569694519 2023-01-21 09:55:31.966239: step: 2044/533, loss: 0.024735068902373314 2023-01-21 09:55:33.122211: step: 2048/533, loss: 0.030648136511445045 2023-01-21 09:55:34.321969: step: 2052/533, loss: 0.6002691984176636 2023-01-21 09:55:35.516065: step: 2056/533, loss: 0.04818249121308327 2023-01-21 09:55:36.740080: step: 2060/533, loss: 0.04724721983075142 2023-01-21 09:55:37.902489: step: 2064/533, loss: 0.02690756320953369 2023-01-21 09:55:39.059322: step: 2068/533, loss: 0.08322863280773163 2023-01-21 09:55:40.250872: step: 2072/533, loss: 0.1908152550458908 2023-01-21 09:55:41.471816: step: 2076/533, loss: 0.21371030807495117 2023-01-21 09:55:42.677093: step: 2080/533, loss: 0.07740859687328339 2023-01-21 09:55:43.879722: step: 2084/533, loss: 0.056576963514089584 2023-01-21 09:55:45.101984: step: 2088/533, loss: 0.09218082576990128 2023-01-21 09:55:46.323518: step: 2092/533, loss: 0.8551380038261414 2023-01-21 09:55:47.489197: step: 2096/533, loss: 0.1529284417629242 2023-01-21 09:55:48.641484: step: 2100/533, loss: 0.031125929206609726 2023-01-21 09:55:49.839929: step: 2104/533, loss: 0.041991040110588074 2023-01-21 09:55:51.063250: step: 2108/533, loss: 0.022683238610625267 2023-01-21 09:55:52.243859: step: 2112/533, loss: 0.12014572322368622 2023-01-21 09:55:53.428691: step: 2116/533, loss: 0.22935214638710022 2023-01-21 09:55:54.656099: step: 2120/533, loss: 0.07466840744018555 2023-01-21 09:55:55.831746: step: 2124/533, loss: 0.1327281892299652 2023-01-21 09:55:57.004733: step: 2128/533, loss: 0.3902902603149414 2023-01-21 09:55:58.173161: step: 2132/533, loss: 0.17620038986206055 ================================================== Loss: 0.179 -------------------- Dev: {'event': {'p': 0.5535714285714286, 'r': 0.7842876165113183, 'f1': 0.6490358126721764}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Test: {'event': {'p': 0.6090335114133074, 'r': 0.8017902813299232, 'f1': 0.6922439966878277}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Chinese: {'event': {'p': 0.5268817204301075, 'r': 0.9074074074074074, 'f1': 0.6666666666666666}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Korean: {'event': {'p': 0.5714285714285714, 'r': 0.5714285714285714, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} 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': 7} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 09:56:40.020302: step: 4/533, loss: 0.6840684413909912 2023-01-21 09:56:41.205931: step: 8/533, loss: 0.19611205160617828 2023-01-21 09:56:42.347666: step: 12/533, loss: 0.04957152530550957 2023-01-21 09:56:43.498136: step: 16/533, loss: 0.07229804992675781 2023-01-21 09:56:44.676739: step: 20/533, loss: 0.06724605709314346 2023-01-21 09:56:45.884942: step: 24/533, loss: 0.03101954609155655 2023-01-21 09:56:47.061540: step: 28/533, loss: 0.03050823137164116 2023-01-21 09:56:48.229604: step: 32/533, loss: 0.23211756348609924 2023-01-21 09:56:49.414493: step: 36/533, loss: 0.15388011932373047 2023-01-21 09:56:50.594041: step: 40/533, loss: 0.05956144258379936 2023-01-21 09:56:51.822886: step: 44/533, loss: 0.09288463741540909 2023-01-21 09:56:53.031806: step: 48/533, loss: 0.02809920348227024 2023-01-21 09:56:54.242618: step: 52/533, loss: 0.1705632209777832 2023-01-21 09:56:55.411007: step: 56/533, loss: 0.036454200744628906 2023-01-21 09:56:56.578165: step: 60/533, loss: 0.06403279304504395 2023-01-21 09:56:57.752337: step: 64/533, loss: 0.07089615613222122 2023-01-21 09:56:58.897333: step: 68/533, loss: 0.0332825630903244 2023-01-21 09:57:00.062887: step: 72/533, loss: 0.05311160162091255 2023-01-21 09:57:01.224613: step: 76/533, loss: 0.076184943318367 2023-01-21 09:57:02.394211: step: 80/533, loss: 0.12787123024463654 2023-01-21 09:57:03.584601: step: 84/533, loss: 0.15155789256095886 2023-01-21 09:57:04.774314: step: 88/533, loss: 0.08287163823843002 2023-01-21 09:57:05.961815: step: 92/533, loss: 0.21548710763454437 2023-01-21 09:57:07.166420: step: 96/533, loss: 0.07911403477191925 2023-01-21 09:57:08.339510: step: 100/533, loss: 0.10535149276256561 2023-01-21 09:57:09.520846: step: 104/533, loss: 0.11290760338306427 2023-01-21 09:57:10.689647: step: 108/533, loss: 0.04956026002764702 2023-01-21 09:57:11.871998: step: 112/533, loss: 0.07664909213781357 2023-01-21 09:57:13.049990: step: 116/533, loss: 0.18215522170066833 2023-01-21 09:57:14.225168: step: 120/533, loss: 0.0694248229265213 2023-01-21 09:57:15.395662: step: 124/533, loss: 0.01795663870871067 2023-01-21 09:57:16.565443: step: 128/533, loss: 0.0804353728890419 2023-01-21 09:57:17.792717: step: 132/533, loss: 0.3450789749622345 2023-01-21 09:57:18.954101: step: 136/533, loss: 0.16633854806423187 2023-01-21 09:57:20.122052: step: 140/533, loss: 0.07037067413330078 2023-01-21 09:57:21.370383: step: 144/533, loss: 0.4730842709541321 2023-01-21 09:57:22.534282: step: 148/533, loss: 0.07848282158374786 2023-01-21 09:57:23.761124: step: 152/533, loss: 0.0747932642698288 2023-01-21 09:57:24.932544: step: 156/533, loss: 0.045522309839725494 2023-01-21 09:57:26.109658: step: 160/533, loss: 0.06635656207799911 2023-01-21 09:57:27.285128: step: 164/533, loss: 0.03924846649169922 2023-01-21 09:57:28.465497: step: 168/533, loss: 0.0030699074268341064 2023-01-21 09:57:29.614688: step: 172/533, loss: 0.011506748385727406 2023-01-21 09:57:30.812948: step: 176/533, loss: 0.05214982107281685 2023-01-21 09:57:31.974375: step: 180/533, loss: 0.012051582336425781 2023-01-21 09:57:33.151785: step: 184/533, loss: 0.08306588977575302 2023-01-21 09:57:34.334835: step: 188/533, loss: 0.03041238896548748 2023-01-21 09:57:35.541879: step: 192/533, loss: 0.21867509186267853 2023-01-21 09:57:36.711999: step: 196/533, loss: 0.053096964955329895 2023-01-21 09:57:37.895867: step: 200/533, loss: 0.016430282965302467 2023-01-21 09:57:39.092057: step: 204/533, loss: 0.08459319919347763 2023-01-21 09:57:40.274810: step: 208/533, loss: 0.07347631454467773 2023-01-21 09:57:41.435931: step: 212/533, loss: 0.04259805753827095 2023-01-21 09:57:42.626365: step: 216/533, loss: 0.05538778752088547 2023-01-21 09:57:43.840127: step: 220/533, loss: 0.055837154388427734 2023-01-21 09:57:45.024725: step: 224/533, loss: 0.009785771369934082 2023-01-21 09:57:46.175068: step: 228/533, loss: 0.31007400155067444 2023-01-21 09:57:47.356663: step: 232/533, loss: 0.04368572682142258 2023-01-21 09:57:48.530002: step: 236/533, loss: 0.013491630554199219 2023-01-21 09:57:49.715222: step: 240/533, loss: 0.11121578514575958 2023-01-21 09:57:50.927467: step: 244/533, loss: 0.040283918380737305 2023-01-21 09:57:52.101373: step: 248/533, loss: 0.062049105763435364 2023-01-21 09:57:53.281575: step: 252/533, loss: 0.1786242425441742 2023-01-21 09:57:54.490384: step: 256/533, loss: 0.02454845979809761 2023-01-21 09:57:55.713175: step: 260/533, loss: 0.0715000182390213 2023-01-21 09:57:56.926195: step: 264/533, loss: 0.059846021234989166 2023-01-21 09:57:58.095250: step: 268/533, loss: 0.08256607502698898 2023-01-21 09:57:59.297581: step: 272/533, loss: 0.13933268189430237 2023-01-21 09:58:00.482599: step: 276/533, loss: 0.05463256686925888 2023-01-21 09:58:01.645938: step: 280/533, loss: 0.04744548723101616 2023-01-21 09:58:02.842735: step: 284/533, loss: 0.03658599779009819 2023-01-21 09:58:03.983718: step: 288/533, loss: 0.06907138973474503 2023-01-21 09:58:05.128393: step: 292/533, loss: 0.00803365744650364 2023-01-21 09:58:06.337798: step: 296/533, loss: 0.10439768433570862 2023-01-21 09:58:07.543473: step: 300/533, loss: 0.2719841003417969 2023-01-21 09:58:08.691744: step: 304/533, loss: 0.15916454792022705 2023-01-21 09:58:09.841342: step: 308/533, loss: 0.1406663954257965 2023-01-21 09:58:11.043622: step: 312/533, loss: 0.07233677059412003 2023-01-21 09:58:12.212370: step: 316/533, loss: 0.059840667992830276 2023-01-21 09:58:13.422841: step: 320/533, loss: 0.005750083830207586 2023-01-21 09:58:14.586855: step: 324/533, loss: 0.19576987624168396 2023-01-21 09:58:15.765959: step: 328/533, loss: 0.04022081196308136 2023-01-21 09:58:16.936775: step: 332/533, loss: 0.07618796825408936 2023-01-21 09:58:18.131741: step: 336/533, loss: 0.04661369323730469 2023-01-21 09:58:19.367063: step: 340/533, loss: 0.15816710889339447 2023-01-21 09:58:20.576147: step: 344/533, loss: 0.062172770500183105 2023-01-21 09:58:21.789850: step: 348/533, loss: 0.11814107745885849 2023-01-21 09:58:22.920800: step: 352/533, loss: 0.005133724305778742 2023-01-21 09:58:24.076669: step: 356/533, loss: 0.1578054428100586 2023-01-21 09:58:25.250800: step: 360/533, loss: 0.04465850442647934 2023-01-21 09:58:26.442015: step: 364/533, loss: 0.04959249496459961 2023-01-21 09:58:27.588741: step: 368/533, loss: 0.07189063727855682 2023-01-21 09:58:28.743686: step: 372/533, loss: 0.1457217037677765 2023-01-21 09:58:29.905729: step: 376/533, loss: 0.027528857812285423 2023-01-21 09:58:31.086012: step: 380/533, loss: 0.01366500835865736 2023-01-21 09:58:32.272485: step: 384/533, loss: 0.002990341279655695 2023-01-21 09:58:33.437355: step: 388/533, loss: 1.1891587972640991 2023-01-21 09:58:34.631754: step: 392/533, loss: 0.07135549187660217 2023-01-21 09:58:35.823790: step: 396/533, loss: 0.13265901803970337 2023-01-21 09:58:37.036307: step: 400/533, loss: 0.09585314244031906 2023-01-21 09:58:38.222481: step: 404/533, loss: 0.018650341778993607 2023-01-21 09:58:39.426234: step: 408/533, loss: 0.41071557998657227 2023-01-21 09:58:40.570150: step: 412/533, loss: 0.11216192692518234 2023-01-21 09:58:41.779619: step: 416/533, loss: 0.19438429176807404 2023-01-21 09:58:42.955184: step: 420/533, loss: 0.036748696118593216 2023-01-21 09:58:44.174852: step: 424/533, loss: 0.06768226623535156 2023-01-21 09:58:45.345596: step: 428/533, loss: 0.09394869953393936 2023-01-21 09:58:46.528777: step: 432/533, loss: 0.15862369537353516 2023-01-21 09:58:47.735788: step: 436/533, loss: 0.13674621284008026 2023-01-21 09:58:48.900894: step: 440/533, loss: 0.11397209763526917 2023-01-21 09:58:50.070974: step: 444/533, loss: 0.08727769553661346 2023-01-21 09:58:51.277725: step: 448/533, loss: 0.12621507048606873 2023-01-21 09:58:52.434017: step: 452/533, loss: 0.054107144474983215 2023-01-21 09:58:53.662851: step: 456/533, loss: 0.15346698462963104 2023-01-21 09:58:54.814005: step: 460/533, loss: 0.3318294584751129 2023-01-21 09:58:55.995094: step: 464/533, loss: 0.21256455779075623 2023-01-21 09:58:57.159471: step: 468/533, loss: 0.004202795214951038 2023-01-21 09:58:58.361971: step: 472/533, loss: 0.17262953519821167 2023-01-21 09:58:59.514649: step: 476/533, loss: 0.03719501569867134 2023-01-21 09:59:00.685822: step: 480/533, loss: 0.09452610462903976 2023-01-21 09:59:01.860681: step: 484/533, loss: 0.17246972024440765 2023-01-21 09:59:03.103387: step: 488/533, loss: 0.03460388258099556 2023-01-21 09:59:04.275091: step: 492/533, loss: 0.9785146117210388 2023-01-21 09:59:05.433018: step: 496/533, loss: 0.11171851307153702 2023-01-21 09:59:06.586505: step: 500/533, loss: 0.02204873599112034 2023-01-21 09:59:07.777188: step: 504/533, loss: 0.04498577117919922 2023-01-21 09:59:08.997302: step: 508/533, loss: 0.041539859026670456 2023-01-21 09:59:10.203103: step: 512/533, loss: 0.031057119369506836 2023-01-21 09:59:11.378715: step: 516/533, loss: 0.059025704860687256 2023-01-21 09:59:12.607060: step: 520/533, loss: 0.09988422691822052 2023-01-21 09:59:13.824946: step: 524/533, loss: 0.08969402313232422 2023-01-21 09:59:15.034132: step: 528/533, loss: 0.09846248477697372 2023-01-21 09:59:16.179130: step: 532/533, loss: 0.02994859218597412 2023-01-21 09:59:17.331354: step: 536/533, loss: 0.031388092786073685 2023-01-21 09:59:18.475367: step: 540/533, loss: 0.12991762161254883 2023-01-21 09:59:19.668196: step: 544/533, loss: 0.17067213356494904 2023-01-21 09:59:20.837937: step: 548/533, loss: 0.22777089476585388 2023-01-21 09:59:22.043020: step: 552/533, loss: 0.09216437488794327 2023-01-21 09:59:23.230053: step: 556/533, loss: 0.12074422836303711 2023-01-21 09:59:24.373239: step: 560/533, loss: 0.12821780145168304 2023-01-21 09:59:25.512948: step: 564/533, loss: 0.3531808853149414 2023-01-21 09:59:26.715731: step: 568/533, loss: 1.0519897937774658 2023-01-21 09:59:27.872328: step: 572/533, loss: 0.0063069346360862255 2023-01-21 09:59:29.043923: step: 576/533, loss: 0.04196124151349068 2023-01-21 09:59:30.248753: step: 580/533, loss: 0.17628784477710724 2023-01-21 09:59:31.419372: step: 584/533, loss: 0.12298412621021271 2023-01-21 09:59:32.581370: step: 588/533, loss: 0.1853950023651123 2023-01-21 09:59:33.784067: step: 592/533, loss: 0.11240754276514053 2023-01-21 09:59:34.941207: step: 596/533, loss: 0.16599902510643005 2023-01-21 09:59:36.157500: step: 600/533, loss: 0.1002839133143425 2023-01-21 09:59:37.324887: step: 604/533, loss: 0.03844108805060387 2023-01-21 09:59:38.482970: step: 608/533, loss: 0.0167099479585886 2023-01-21 09:59:39.743152: step: 612/533, loss: 0.03913535922765732 2023-01-21 09:59:40.895594: step: 616/533, loss: 0.08450829982757568 2023-01-21 09:59:42.116891: step: 620/533, loss: 0.012268638238310814 2023-01-21 09:59:43.288087: step: 624/533, loss: 0.056078244000673294 2023-01-21 09:59:44.473924: step: 628/533, loss: 0.48550042510032654 2023-01-21 09:59:45.651645: step: 632/533, loss: 1.5149446725845337 2023-01-21 09:59:46.839634: step: 636/533, loss: 0.28021401166915894 2023-01-21 09:59:48.031137: step: 640/533, loss: 0.07205267250537872 2023-01-21 09:59:49.219256: step: 644/533, loss: 0.14142942428588867 2023-01-21 09:59:50.423796: step: 648/533, loss: 0.3264802098274231 2023-01-21 09:59:51.646433: step: 652/533, loss: 0.17765235900878906 2023-01-21 09:59:52.807130: step: 656/533, loss: 0.005532455630600452 2023-01-21 09:59:54.009338: step: 660/533, loss: 0.03502330929040909 2023-01-21 09:59:55.209720: step: 664/533, loss: 0.7579852342605591 2023-01-21 09:59:56.402724: step: 668/533, loss: 0.1398542821407318 2023-01-21 09:59:57.571855: step: 672/533, loss: 0.05399642139673233 2023-01-21 09:59:58.765620: step: 676/533, loss: 0.19067582488059998 2023-01-21 09:59:59.947194: step: 680/533, loss: 0.24744835495948792 2023-01-21 10:00:01.120479: step: 684/533, loss: 0.11159972846508026 2023-01-21 10:00:02.315684: step: 688/533, loss: 0.02066049724817276 2023-01-21 10:00:03.500395: step: 692/533, loss: 0.06876301765441895 2023-01-21 10:00:04.663988: step: 696/533, loss: 0.03165634721517563 2023-01-21 10:00:05.856770: step: 700/533, loss: 0.07921639084815979 2023-01-21 10:00:07.028291: step: 704/533, loss: 0.010765934363007545 2023-01-21 10:00:08.237543: step: 708/533, loss: 0.21481665968894958 2023-01-21 10:00:09.392584: step: 712/533, loss: 0.24281997978687286 2023-01-21 10:00:10.588226: step: 716/533, loss: 0.11822627484798431 2023-01-21 10:00:11.793394: step: 720/533, loss: 0.6449282169342041 2023-01-21 10:00:12.974911: step: 724/533, loss: 0.35939139127731323 2023-01-21 10:00:14.140175: step: 728/533, loss: 0.24447423219680786 2023-01-21 10:00:15.328521: step: 732/533, loss: 0.016836263239383698 2023-01-21 10:00:16.519730: step: 736/533, loss: 0.044773247092962265 2023-01-21 10:00:17.720049: step: 740/533, loss: 0.018569564446806908 2023-01-21 10:00:18.836436: step: 744/533, loss: 0.0799143984913826 2023-01-21 10:00:20.023560: step: 748/533, loss: 0.09197226166725159 2023-01-21 10:00:21.204277: step: 752/533, loss: 0.1904076635837555 2023-01-21 10:00:22.396970: step: 756/533, loss: 0.3678344786167145 2023-01-21 10:00:23.571264: step: 760/533, loss: 0.015276337042450905 2023-01-21 10:00:24.787830: step: 764/533, loss: 0.09168615192174911 2023-01-21 10:00:25.984606: step: 768/533, loss: 0.08190512657165527 2023-01-21 10:00:27.189138: step: 772/533, loss: 0.175513356924057 2023-01-21 10:00:28.366567: step: 776/533, loss: 0.06053218990564346 2023-01-21 10:00:29.600426: step: 780/533, loss: 0.03240928798913956 2023-01-21 10:00:30.810646: step: 784/533, loss: 0.05183429643511772 2023-01-21 10:00:31.965831: step: 788/533, loss: 0.005655050277709961 2023-01-21 10:00:33.109377: step: 792/533, loss: 0.08929681777954102 2023-01-21 10:00:34.280604: step: 796/533, loss: 0.09898243099451065 2023-01-21 10:00:35.517782: step: 800/533, loss: 0.03399248048663139 2023-01-21 10:00:36.757947: step: 804/533, loss: 0.03037128411233425 2023-01-21 10:00:37.935443: step: 808/533, loss: 0.03187217563390732 2023-01-21 10:00:39.117272: step: 812/533, loss: 0.10390167683362961 2023-01-21 10:00:40.313634: step: 816/533, loss: 0.06381683051586151 2023-01-21 10:00:41.498930: step: 820/533, loss: 0.06803464889526367 2023-01-21 10:00:42.704963: step: 824/533, loss: 0.097596175968647 2023-01-21 10:00:43.905763: step: 828/533, loss: 0.16625767946243286 2023-01-21 10:00:45.039005: step: 832/533, loss: 0.03342418745160103 2023-01-21 10:00:46.219399: step: 836/533, loss: 0.01952815055847168 2023-01-21 10:00:47.401483: step: 840/533, loss: 0.03822450712323189 2023-01-21 10:00:48.586752: step: 844/533, loss: 0.06044311448931694 2023-01-21 10:00:49.815394: step: 848/533, loss: 0.11998863518238068 2023-01-21 10:00:50.998971: step: 852/533, loss: 0.038886167109012604 2023-01-21 10:00:52.227137: step: 856/533, loss: 0.19750748574733734 2023-01-21 10:00:53.501804: step: 860/533, loss: 0.42549440264701843 2023-01-21 10:00:54.690059: step: 864/533, loss: 0.04626961052417755 2023-01-21 10:00:55.911783: step: 868/533, loss: 0.63702392578125 2023-01-21 10:00:57.075351: step: 872/533, loss: 0.018661784008145332 2023-01-21 10:00:58.254128: step: 876/533, loss: 0.01962861977517605 2023-01-21 10:00:59.491085: step: 880/533, loss: 0.022882748395204544 2023-01-21 10:01:00.710046: step: 884/533, loss: 0.04386615753173828 2023-01-21 10:01:01.875521: step: 888/533, loss: 0.034467220306396484 2023-01-21 10:01:03.045289: step: 892/533, loss: 0.20613127946853638 2023-01-21 10:01:04.222920: step: 896/533, loss: 0.23075255751609802 2023-01-21 10:01:05.438092: step: 900/533, loss: 0.160674050450325 2023-01-21 10:01:06.629672: step: 904/533, loss: 0.2709992229938507 2023-01-21 10:01:07.811026: step: 908/533, loss: 0.02138206921517849 2023-01-21 10:01:08.969188: step: 912/533, loss: 0.9458889365196228 2023-01-21 10:01:10.129818: step: 916/533, loss: 0.038884639739990234 2023-01-21 10:01:11.299711: step: 920/533, loss: 0.08989983052015305 2023-01-21 10:01:12.472258: step: 924/533, loss: 0.0558956116437912 2023-01-21 10:01:13.649073: step: 928/533, loss: 0.09872579574584961 2023-01-21 10:01:14.829986: step: 932/533, loss: 0.039171792566776276 2023-01-21 10:01:15.997386: step: 936/533, loss: 0.15904836356639862 2023-01-21 10:01:17.161534: step: 940/533, loss: 0.019061803817749023 2023-01-21 10:01:18.369729: step: 944/533, loss: 0.0789029598236084 2023-01-21 10:01:19.528778: step: 948/533, loss: 0.05996885523200035 2023-01-21 10:01:20.745989: step: 952/533, loss: 0.16926145553588867 2023-01-21 10:01:21.918052: step: 956/533, loss: 0.035991862416267395 2023-01-21 10:01:23.082315: step: 960/533, loss: 0.03003845363855362 2023-01-21 10:01:24.291407: step: 964/533, loss: 0.2123720347881317 2023-01-21 10:01:25.463971: step: 968/533, loss: 0.7410622239112854 2023-01-21 10:01:26.640603: step: 972/533, loss: 0.017244840040802956 2023-01-21 10:01:27.807969: step: 976/533, loss: 0.05120887607336044 2023-01-21 10:01:29.012523: step: 980/533, loss: 0.04309334605932236 2023-01-21 10:01:30.197036: step: 984/533, loss: 0.3598502278327942 2023-01-21 10:01:31.402307: step: 988/533, loss: 0.4101582467556 2023-01-21 10:01:32.593910: step: 992/533, loss: 0.03446369245648384 2023-01-21 10:01:33.780988: step: 996/533, loss: 0.01908884011209011 2023-01-21 10:01:34.945110: step: 1000/533, loss: 0.029134750366210938 2023-01-21 10:01:36.167386: step: 1004/533, loss: 0.049874309450387955 2023-01-21 10:01:37.307271: step: 1008/533, loss: 0.17342795431613922 2023-01-21 10:01:38.508637: step: 1012/533, loss: 0.07612524926662445 2023-01-21 10:01:39.682383: step: 1016/533, loss: 0.047574520111083984 2023-01-21 10:01:40.870295: step: 1020/533, loss: 0.14100809395313263 2023-01-21 10:01:42.033915: step: 1024/533, loss: 0.07453164458274841 2023-01-21 10:01:43.215973: step: 1028/533, loss: 0.4932495057582855 2023-01-21 10:01:44.400602: step: 1032/533, loss: 0.017429828643798828 2023-01-21 10:01:45.569690: step: 1036/533, loss: 0.3841678500175476 2023-01-21 10:01:46.755450: step: 1040/533, loss: 0.2241520881652832 2023-01-21 10:01:47.971764: step: 1044/533, loss: 0.336454302072525 2023-01-21 10:01:49.115307: step: 1048/533, loss: 0.11091289669275284 2023-01-21 10:01:50.288666: step: 1052/533, loss: 0.04796476662158966 2023-01-21 10:01:51.460549: step: 1056/533, loss: 0.1912672072649002 2023-01-21 10:01:52.608292: step: 1060/533, loss: 0.12299275398254395 2023-01-21 10:01:53.808512: step: 1064/533, loss: 0.06454887241125107 2023-01-21 10:01:55.069163: step: 1068/533, loss: 0.09541969001293182 2023-01-21 10:01:56.272464: step: 1072/533, loss: 0.02104644849896431 2023-01-21 10:01:57.470268: step: 1076/533, loss: 0.035081129521131516 2023-01-21 10:01:58.633974: step: 1080/533, loss: 0.1482558250427246 2023-01-21 10:01:59.802265: step: 1084/533, loss: 0.05635084956884384 2023-01-21 10:02:01.010610: step: 1088/533, loss: 0.13622379302978516 2023-01-21 10:02:02.171080: step: 1092/533, loss: 0.10148197412490845 2023-01-21 10:02:03.321323: step: 1096/533, loss: 0.040871430188417435 2023-01-21 10:02:04.580961: step: 1100/533, loss: 0.1324821412563324 2023-01-21 10:02:05.759047: step: 1104/533, loss: 0.1031554713845253 2023-01-21 10:02:06.932043: step: 1108/533, loss: 0.15277177095413208 2023-01-21 10:02:08.097965: step: 1112/533, loss: 0.013832282274961472 2023-01-21 10:02:09.299260: step: 1116/533, loss: 0.151885986328125 2023-01-21 10:02:10.503035: step: 1120/533, loss: 0.06178426742553711 2023-01-21 10:02:11.660264: step: 1124/533, loss: 0.0672360435128212 2023-01-21 10:02:12.822873: step: 1128/533, loss: 0.061821937561035156 2023-01-21 10:02:14.010015: step: 1132/533, loss: 0.17634135484695435 2023-01-21 10:02:15.209731: step: 1136/533, loss: 0.2373599112033844 2023-01-21 10:02:16.386685: step: 1140/533, loss: 0.39508095383644104 2023-01-21 10:02:17.590417: step: 1144/533, loss: 0.08182048797607422 2023-01-21 10:02:18.719240: step: 1148/533, loss: 0.1069403737783432 2023-01-21 10:02:19.880439: step: 1152/533, loss: 0.01784048229455948 2023-01-21 10:02:21.116615: step: 1156/533, loss: 0.09308870136737823 2023-01-21 10:02:22.297405: step: 1160/533, loss: 0.17533540725708008 2023-01-21 10:02:23.509415: step: 1164/533, loss: 0.08245744556188583 2023-01-21 10:02:24.708003: step: 1168/533, loss: 0.1243671402335167 2023-01-21 10:02:25.906249: step: 1172/533, loss: 0.6975938677787781 2023-01-21 10:02:27.093035: step: 1176/533, loss: 0.1372303068637848 2023-01-21 10:02:28.281275: step: 1180/533, loss: 0.06776552647352219 2023-01-21 10:02:29.459278: step: 1184/533, loss: 0.18441256880760193 2023-01-21 10:02:30.614175: step: 1188/533, loss: 0.11154685169458389 2023-01-21 10:02:31.822175: step: 1192/533, loss: 0.06885939091444016 2023-01-21 10:02:32.984588: step: 1196/533, loss: 0.16055002808570862 2023-01-21 10:02:34.186647: step: 1200/533, loss: 0.14407922327518463 2023-01-21 10:02:35.365318: step: 1204/533, loss: 0.03295435756444931 2023-01-21 10:02:36.517007: step: 1208/533, loss: 0.04848499596118927 2023-01-21 10:02:37.679123: step: 1212/533, loss: 0.007522201631218195 2023-01-21 10:02:38.865880: step: 1216/533, loss: 0.30566588044166565 2023-01-21 10:02:40.059120: step: 1220/533, loss: 0.09707650542259216 2023-01-21 10:02:41.226132: step: 1224/533, loss: 0.05437551811337471 2023-01-21 10:02:42.420730: step: 1228/533, loss: 0.0364563949406147 2023-01-21 10:02:43.629877: step: 1232/533, loss: 0.051595401018857956 2023-01-21 10:02:44.800799: step: 1236/533, loss: 0.02667818032205105 2023-01-21 10:02:45.988069: step: 1240/533, loss: 0.12089768052101135 2023-01-21 10:02:47.143184: step: 1244/533, loss: 0.21270255744457245 2023-01-21 10:02:48.315881: step: 1248/533, loss: 0.0508122444152832 2023-01-21 10:02:49.486991: step: 1252/533, loss: 0.29473793506622314 2023-01-21 10:02:50.639425: step: 1256/533, loss: 0.10584226250648499 2023-01-21 10:02:51.843902: step: 1260/533, loss: 0.12005731463432312 2023-01-21 10:02:53.006609: step: 1264/533, loss: 0.4417663514614105 2023-01-21 10:02:54.154854: step: 1268/533, loss: 0.1455962210893631 2023-01-21 10:02:55.286556: step: 1272/533, loss: 0.07612204551696777 2023-01-21 10:02:56.454688: step: 1276/533, loss: 0.0481874942779541 2023-01-21 10:02:57.650295: step: 1280/533, loss: 0.1576584428548813 2023-01-21 10:02:58.855934: step: 1284/533, loss: 0.049852754920721054 2023-01-21 10:03:00.006208: step: 1288/533, loss: 0.2352529615163803 2023-01-21 10:03:01.169068: step: 1292/533, loss: 0.007737541571259499 2023-01-21 10:03:02.323931: step: 1296/533, loss: 0.0856243371963501 2023-01-21 10:03:03.504609: step: 1300/533, loss: 0.048775482922792435 2023-01-21 10:03:04.709106: step: 1304/533, loss: 0.02902846410870552 2023-01-21 10:03:05.894149: step: 1308/533, loss: 0.07591553032398224 2023-01-21 10:03:07.046575: step: 1312/533, loss: 0.06961230933666229 2023-01-21 10:03:08.213733: step: 1316/533, loss: 0.05663866922259331 2023-01-21 10:03:09.410000: step: 1320/533, loss: 0.011522769927978516 2023-01-21 10:03:10.578121: step: 1324/533, loss: 0.08625419437885284 2023-01-21 10:03:11.740509: step: 1328/533, loss: 0.1429588794708252 2023-01-21 10:03:12.911784: step: 1332/533, loss: 0.03054199367761612 2023-01-21 10:03:14.071631: step: 1336/533, loss: 0.07631373405456543 2023-01-21 10:03:15.238284: step: 1340/533, loss: 0.037290096282958984 2023-01-21 10:03:16.402685: step: 1344/533, loss: 0.038230106234550476 2023-01-21 10:03:17.603761: step: 1348/533, loss: 0.06629786640405655 2023-01-21 10:03:18.816531: step: 1352/533, loss: 0.024962520226836205 2023-01-21 10:03:19.991734: step: 1356/533, loss: 0.10138006508350372 2023-01-21 10:03:21.124374: step: 1360/533, loss: 0.11009187996387482 2023-01-21 10:03:22.284876: step: 1364/533, loss: 0.39571821689605713 2023-01-21 10:03:23.496985: step: 1368/533, loss: 0.04840097576379776 2023-01-21 10:03:24.671330: step: 1372/533, loss: 0.2133781909942627 2023-01-21 10:03:25.809603: step: 1376/533, loss: 0.23290252685546875 2023-01-21 10:03:26.969205: step: 1380/533, loss: 0.07012281566858292 2023-01-21 10:03:28.167888: step: 1384/533, loss: 0.15407808125019073 2023-01-21 10:03:29.402203: step: 1388/533, loss: 0.05259094387292862 2023-01-21 10:03:30.573792: step: 1392/533, loss: 0.11284966766834259 2023-01-21 10:03:31.709721: step: 1396/533, loss: 0.45853570103645325 2023-01-21 10:03:32.861759: step: 1400/533, loss: 0.10231009125709534 2023-01-21 10:03:34.037465: step: 1404/533, loss: 0.05192689970135689 2023-01-21 10:03:35.173547: step: 1408/533, loss: 0.013852620497345924 2023-01-21 10:03:36.352276: step: 1412/533, loss: 0.029465390369296074 2023-01-21 10:03:37.563828: step: 1416/533, loss: 0.3853006958961487 2023-01-21 10:03:38.745417: step: 1420/533, loss: 0.016729354858398438 2023-01-21 10:03:39.880130: step: 1424/533, loss: 0.17632311582565308 2023-01-21 10:03:41.052146: step: 1428/533, loss: 0.08743259310722351 2023-01-21 10:03:42.255869: step: 1432/533, loss: 0.11501336097717285 2023-01-21 10:03:43.426925: step: 1436/533, loss: 0.05981488525867462 2023-01-21 10:03:44.569674: step: 1440/533, loss: 0.07624292373657227 2023-01-21 10:03:45.756233: step: 1444/533, loss: 0.032958388328552246 2023-01-21 10:03:46.920329: step: 1448/533, loss: 0.013405894860625267 2023-01-21 10:03:48.110311: step: 1452/533, loss: 0.09428711235523224 2023-01-21 10:03:49.259520: step: 1456/533, loss: 0.07770691066980362 2023-01-21 10:03:50.454771: step: 1460/533, loss: 0.2388835847377777 2023-01-21 10:03:51.734737: step: 1464/533, loss: 0.16202107071876526 2023-01-21 10:03:52.897872: step: 1468/533, loss: 0.03350029140710831 2023-01-21 10:03:54.044795: step: 1472/533, loss: 0.04074613004922867 2023-01-21 10:03:55.209373: step: 1476/533, loss: 0.41753920912742615 2023-01-21 10:03:56.368251: step: 1480/533, loss: 0.04913058131933212 2023-01-21 10:03:57.559279: step: 1484/533, loss: 0.08075688779354095 2023-01-21 10:03:58.741512: step: 1488/533, loss: 0.1480119824409485 2023-01-21 10:03:59.931125: step: 1492/533, loss: 0.04343461990356445 2023-01-21 10:04:01.117008: step: 1496/533, loss: 0.05710973963141441 2023-01-21 10:04:02.276331: step: 1500/533, loss: 0.6117690205574036 2023-01-21 10:04:03.500584: step: 1504/533, loss: 0.5654932856559753 2023-01-21 10:04:04.662151: step: 1508/533, loss: 0.25297728180885315 2023-01-21 10:04:05.819996: step: 1512/533, loss: 0.004511928651481867 2023-01-21 10:04:07.055716: step: 1516/533, loss: 0.10465659946203232 2023-01-21 10:04:08.251972: step: 1520/533, loss: 0.33815255761146545 2023-01-21 10:04:09.399599: step: 1524/533, loss: 0.077989861369133 2023-01-21 10:04:10.574053: step: 1528/533, loss: 0.024226760491728783 2023-01-21 10:04:11.741888: step: 1532/533, loss: 0.22960643470287323 2023-01-21 10:04:12.969391: step: 1536/533, loss: 0.19098205864429474 2023-01-21 10:04:14.122784: step: 1540/533, loss: 0.1172613650560379 2023-01-21 10:04:15.325084: step: 1544/533, loss: 0.16278590261936188 2023-01-21 10:04:16.483303: step: 1548/533, loss: 0.12067241221666336 2023-01-21 10:04:17.708893: step: 1552/533, loss: 0.08397092670202255 2023-01-21 10:04:18.875944: step: 1556/533, loss: 0.08799795806407928 2023-01-21 10:04:20.005305: step: 1560/533, loss: 0.007019233889877796 2023-01-21 10:04:21.221808: step: 1564/533, loss: 0.051380157470703125 2023-01-21 10:04:22.431617: step: 1568/533, loss: 0.05458813160657883 2023-01-21 10:04:23.654555: step: 1572/533, loss: 0.05836434289813042 2023-01-21 10:04:24.863219: step: 1576/533, loss: 0.07878632843494415 2023-01-21 10:04:26.089598: step: 1580/533, loss: 0.15848851203918457 2023-01-21 10:04:27.264434: step: 1584/533, loss: 0.09809398651123047 2023-01-21 10:04:28.469059: step: 1588/533, loss: 0.10737352818250656 2023-01-21 10:04:29.650060: step: 1592/533, loss: 0.10923920571804047 2023-01-21 10:04:30.859199: step: 1596/533, loss: 0.06701450794935226 2023-01-21 10:04:32.015709: step: 1600/533, loss: 0.10820188373327255 2023-01-21 10:04:33.209311: step: 1604/533, loss: 0.5899614691734314 2023-01-21 10:04:34.363201: step: 1608/533, loss: 0.08746099472045898 2023-01-21 10:04:35.561763: step: 1612/533, loss: 0.23179256916046143 2023-01-21 10:04:36.726083: step: 1616/533, loss: 0.17375750839710236 2023-01-21 10:04:37.874753: step: 1620/533, loss: 0.8971181511878967 2023-01-21 10:04:39.032624: step: 1624/533, loss: 0.054903555661439896 2023-01-21 10:04:40.188860: step: 1628/533, loss: 0.061953071504831314 2023-01-21 10:04:41.341364: step: 1632/533, loss: 0.057358644902706146 2023-01-21 10:04:42.506123: step: 1636/533, loss: 0.09211969375610352 2023-01-21 10:04:43.735828: step: 1640/533, loss: 0.039336059242486954 2023-01-21 10:04:44.912301: step: 1644/533, loss: 0.07075109332799911 2023-01-21 10:04:46.076990: step: 1648/533, loss: 0.1728501319885254 2023-01-21 10:04:47.236067: step: 1652/533, loss: 0.11759848892688751 2023-01-21 10:04:48.433357: step: 1656/533, loss: 0.13034382462501526 2023-01-21 10:04:49.632073: step: 1660/533, loss: 0.046736910939216614 2023-01-21 10:04:50.820924: step: 1664/533, loss: 0.11750298738479614 2023-01-21 10:04:51.988427: step: 1668/533, loss: 0.177720844745636 2023-01-21 10:04:53.129695: step: 1672/533, loss: 0.1376950740814209 2023-01-21 10:04:54.312132: step: 1676/533, loss: 0.08894872665405273 2023-01-21 10:04:55.497428: step: 1680/533, loss: 0.05386357381939888 2023-01-21 10:04:56.687732: step: 1684/533, loss: 0.0661015510559082 2023-01-21 10:04:57.857581: step: 1688/533, loss: 0.07546535134315491 2023-01-21 10:04:58.996771: step: 1692/533, loss: 0.008820438757538795 2023-01-21 10:05:00.171591: step: 1696/533, loss: 0.029964257031679153 2023-01-21 10:05:01.364468: step: 1700/533, loss: 0.19186973571777344 2023-01-21 10:05:02.565127: step: 1704/533, loss: 0.5331511497497559 2023-01-21 10:05:03.726637: step: 1708/533, loss: 0.13748803734779358 2023-01-21 10:05:04.955639: step: 1712/533, loss: 0.056466102600097656 2023-01-21 10:05:06.100763: step: 1716/533, loss: 0.0015454532112926245 2023-01-21 10:05:07.293933: step: 1720/533, loss: 0.013463020324707031 2023-01-21 10:05:08.503738: step: 1724/533, loss: 0.04285430908203125 2023-01-21 10:05:09.644176: step: 1728/533, loss: 0.033825017511844635 2023-01-21 10:05:10.820853: step: 1732/533, loss: 0.08803386986255646 2023-01-21 10:05:11.994841: step: 1736/533, loss: 0.04089755937457085 2023-01-21 10:05:13.177518: step: 1740/533, loss: 0.092307448387146 2023-01-21 10:05:14.378043: step: 1744/533, loss: 0.37547388672828674 2023-01-21 10:05:15.547944: step: 1748/533, loss: 0.4492069482803345 2023-01-21 10:05:16.733260: step: 1752/533, loss: 0.03988780826330185 2023-01-21 10:05:17.937017: step: 1756/533, loss: 0.15885190665721893 2023-01-21 10:05:19.074421: step: 1760/533, loss: 0.3061714172363281 2023-01-21 10:05:20.233085: step: 1764/533, loss: 0.3835746645927429 2023-01-21 10:05:21.378229: step: 1768/533, loss: 0.27670979499816895 2023-01-21 10:05:22.608261: step: 1772/533, loss: 0.12942752242088318 2023-01-21 10:05:23.759448: step: 1776/533, loss: 0.12261597812175751 2023-01-21 10:05:24.985852: step: 1780/533, loss: 0.09932427853345871 2023-01-21 10:05:26.186782: step: 1784/533, loss: 0.118072509765625 2023-01-21 10:05:27.385106: step: 1788/533, loss: 1.369811773300171 2023-01-21 10:05:28.659864: step: 1792/533, loss: 0.06129207834601402 2023-01-21 10:05:29.799470: step: 1796/533, loss: 0.0732000395655632 2023-01-21 10:05:30.962562: step: 1800/533, loss: 0.08286122977733612 2023-01-21 10:05:32.168899: step: 1804/533, loss: 0.08141598850488663 2023-01-21 10:05:33.325489: step: 1808/533, loss: 0.12271225452423096 2023-01-21 10:05:34.477578: step: 1812/533, loss: 0.0843229815363884 2023-01-21 10:05:35.665840: step: 1816/533, loss: 0.07953443378210068 2023-01-21 10:05:36.839433: step: 1820/533, loss: 0.01404027920216322 2023-01-21 10:05:38.046340: step: 1824/533, loss: 0.14238977432250977 2023-01-21 10:05:39.204758: step: 1828/533, loss: 0.11053504794836044 2023-01-21 10:05:40.383861: step: 1832/533, loss: 0.055158376693725586 2023-01-21 10:05:41.560058: step: 1836/533, loss: 1.2640025615692139 2023-01-21 10:05:42.728280: step: 1840/533, loss: 0.01196146011352539 2023-01-21 10:05:43.874887: step: 1844/533, loss: 0.02628335915505886 2023-01-21 10:05:45.047194: step: 1848/533, loss: 0.026090241968631744 2023-01-21 10:05:46.204052: step: 1852/533, loss: 0.1779797524213791 2023-01-21 10:05:47.414588: step: 1856/533, loss: 0.09468249976634979 2023-01-21 10:05:48.574454: step: 1860/533, loss: 0.029116058722138405 2023-01-21 10:05:49.757442: step: 1864/533, loss: 0.10984411835670471 2023-01-21 10:05:50.916786: step: 1868/533, loss: 0.1984615921974182 2023-01-21 10:05:52.078794: step: 1872/533, loss: 0.009050846099853516 2023-01-21 10:05:53.245483: step: 1876/533, loss: 0.016206074506044388 2023-01-21 10:05:54.412393: step: 1880/533, loss: 0.03429281711578369 2023-01-21 10:05:55.579631: step: 1884/533, loss: 0.11189261078834534 2023-01-21 10:05:56.752842: step: 1888/533, loss: 0.13980932533740997 2023-01-21 10:05:57.930435: step: 1892/533, loss: 0.296229749917984 2023-01-21 10:05:59.122938: step: 1896/533, loss: 0.11454787105321884 2023-01-21 10:06:00.292748: step: 1900/533, loss: 0.1393490880727768 2023-01-21 10:06:01.491078: step: 1904/533, loss: 0.03891587257385254 2023-01-21 10:06:02.645126: step: 1908/533, loss: 0.0973021537065506 2023-01-21 10:06:03.842539: step: 1912/533, loss: 0.11823282390832901 2023-01-21 10:06:04.973275: step: 1916/533, loss: 0.5313553214073181 2023-01-21 10:06:06.159963: step: 1920/533, loss: 0.12328267842531204 2023-01-21 10:06:07.335698: step: 1924/533, loss: 0.07467503845691681 2023-01-21 10:06:08.524892: step: 1928/533, loss: 0.1543058454990387 2023-01-21 10:06:09.732334: step: 1932/533, loss: 0.06829843670129776 2023-01-21 10:06:10.905115: step: 1936/533, loss: 0.04492764547467232 2023-01-21 10:06:12.076615: step: 1940/533, loss: 0.07383785396814346 2023-01-21 10:06:13.216294: step: 1944/533, loss: 0.05159907415509224 2023-01-21 10:06:14.348815: step: 1948/533, loss: 0.020679570734500885 2023-01-21 10:06:15.550425: step: 1952/533, loss: 0.06067943572998047 2023-01-21 10:06:16.704213: step: 1956/533, loss: 0.10070066899061203 2023-01-21 10:06:17.897611: step: 1960/533, loss: 0.10214195400476456 2023-01-21 10:06:19.047785: step: 1964/533, loss: 0.0035020112991333008 2023-01-21 10:06:20.219446: step: 1968/533, loss: 0.3237728178501129 2023-01-21 10:06:21.415267: step: 1972/533, loss: 0.6307231187820435 2023-01-21 10:06:22.577334: step: 1976/533, loss: 0.11528892815113068 2023-01-21 10:06:23.822708: step: 1980/533, loss: 0.10718250274658203 2023-01-21 10:06:24.997540: step: 1984/533, loss: 0.3092205822467804 2023-01-21 10:06:26.205009: step: 1988/533, loss: 0.06763891875743866 2023-01-21 10:06:27.488232: step: 1992/533, loss: 0.06109762191772461 2023-01-21 10:06:28.720322: step: 1996/533, loss: 0.5236159563064575 2023-01-21 10:06:29.890986: step: 2000/533, loss: 0.036244723945856094 2023-01-21 10:06:31.120431: step: 2004/533, loss: 0.1054566353559494 2023-01-21 10:06:32.277332: step: 2008/533, loss: 0.2676524221897125 2023-01-21 10:06:33.466883: step: 2012/533, loss: 0.030649995431303978 2023-01-21 10:06:34.614332: step: 2016/533, loss: 0.43426164984703064 2023-01-21 10:06:35.799281: step: 2020/533, loss: 0.01166553609073162 2023-01-21 10:06:36.978730: step: 2024/533, loss: 0.18203334510326385 2023-01-21 10:06:38.160936: step: 2028/533, loss: 0.038648199290037155 2023-01-21 10:06:39.359969: step: 2032/533, loss: 0.08750849217176437 2023-01-21 10:06:40.543335: step: 2036/533, loss: 0.7998626232147217 2023-01-21 10:06:41.714153: step: 2040/533, loss: 0.1191432848572731 2023-01-21 10:06:42.904240: step: 2044/533, loss: 0.05177684128284454 2023-01-21 10:06:44.089699: step: 2048/533, loss: 0.04636268690228462 2023-01-21 10:06:45.260338: step: 2052/533, loss: 0.09539518505334854 2023-01-21 10:06:46.484536: step: 2056/533, loss: 0.017030049115419388 2023-01-21 10:06:47.676287: step: 2060/533, loss: 0.040022946894168854 2023-01-21 10:06:48.909635: step: 2064/533, loss: 0.3287423849105835 2023-01-21 10:06:50.077329: step: 2068/533, loss: 0.21219034492969513 2023-01-21 10:06:51.251340: step: 2072/533, loss: 0.033795833587646484 2023-01-21 10:06:52.385071: step: 2076/533, loss: 0.08236217498779297 2023-01-21 10:06:53.554441: step: 2080/533, loss: 0.07740440964698792 2023-01-21 10:06:54.738787: step: 2084/533, loss: 0.2799074649810791 2023-01-21 10:06:55.917043: step: 2088/533, loss: 0.04357290267944336 2023-01-21 10:06:57.074919: step: 2092/533, loss: 0.0749397799372673 2023-01-21 10:06:58.198587: step: 2096/533, loss: 0.25775814056396484 2023-01-21 10:06:59.367463: step: 2100/533, loss: 0.08026123046875 2023-01-21 10:07:00.571511: step: 2104/533, loss: 0.13429725170135498 2023-01-21 10:07:01.761175: step: 2108/533, loss: 0.1385565847158432 2023-01-21 10:07:02.968860: step: 2112/533, loss: 0.049806784838438034 2023-01-21 10:07:04.112826: step: 2116/533, loss: 0.03262138366699219 2023-01-21 10:07:05.270612: step: 2120/533, loss: 0.15529866516590118 2023-01-21 10:07:06.419607: step: 2124/533, loss: 0.08780555427074432 2023-01-21 10:07:07.641906: step: 2128/533, loss: 0.08337926864624023 2023-01-21 10:07:08.857634: step: 2132/533, loss: 0.10202445834875107 ================================================== Loss: 0.138 -------------------- Dev: {'event': {'p': 0.5778648383937316, 'r': 0.7856191744340879, 'f1': 0.6659142212189617}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Test: {'event': {'p': 0.6300893743793445, 'r': 0.8113810741687979, 'f1': 0.7093348239239798}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Chinese: {'event': {'p': 0.5666666666666667, 'r': 0.9444444444444444, 'f1': 0.7083333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Korean: {'event': {'p': 0.6379310344827587, 'r': 0.5873015873015873, 'f1': 0.6115702479338844}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} 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': 8} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6832460732984293, 'r': 0.6950732356857523, 'f1': 0.689108910891089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6548406494287432, 'r': 0.6962915601023018, 'f1': 0.6749302757979547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.7037037037037037, 'r': 0.7037037037037037, 'f1': 0.7037037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5864197530864198, 'r': 0.758988015978695, 'f1': 0.6616366802089378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6193447737909517, 'r': 0.7615089514066496, 'f1': 0.6831086894178378}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.723404255319149, 'r': 0.5396825396825397, 'f1': 0.6181818181818182}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 10:07:50.076642: step: 4/533, loss: 0.09926968812942505 2023-01-21 10:07:51.222077: step: 8/533, loss: 0.026259899139404297 2023-01-21 10:07:52.380721: step: 12/533, loss: 0.04331560432910919 2023-01-21 10:07:53.594859: step: 16/533, loss: 0.10419845581054688 2023-01-21 10:07:54.771707: step: 20/533, loss: 0.02791421487927437 2023-01-21 10:07:55.954509: step: 24/533, loss: 0.048871614038944244 2023-01-21 10:07:57.125882: step: 28/533, loss: 0.05272989720106125 2023-01-21 10:07:58.330190: step: 32/533, loss: 0.0406985804438591 2023-01-21 10:07:59.497060: step: 36/533, loss: 0.04748249053955078 2023-01-21 10:08:00.654330: step: 40/533, loss: 0.06372527778148651 2023-01-21 10:08:01.850107: step: 44/533, loss: 0.1344389021396637 2023-01-21 10:08:03.017568: step: 48/533, loss: 0.08515739440917969 2023-01-21 10:08:04.222761: step: 52/533, loss: 0.07617368549108505 2023-01-21 10:08:05.404711: step: 56/533, loss: 0.02903785929083824 2023-01-21 10:08:06.560686: step: 60/533, loss: 0.03367490693926811 2023-01-21 10:08:07.704910: step: 64/533, loss: 0.11510520428419113 2023-01-21 10:08:08.899440: step: 68/533, loss: 0.010165119543671608 2023-01-21 10:08:10.050623: step: 72/533, loss: 0.052768900990486145 2023-01-21 10:08:11.173216: step: 76/533, loss: 0.1772814393043518 2023-01-21 10:08:12.361522: step: 80/533, loss: 0.0358455665409565 2023-01-21 10:08:13.521159: step: 84/533, loss: 0.006482315249741077 2023-01-21 10:08:14.737562: step: 88/533, loss: 0.24554085731506348 2023-01-21 10:08:15.922130: step: 92/533, loss: 0.6262931823730469 2023-01-21 10:08:17.102634: step: 96/533, loss: 0.05124311521649361 2023-01-21 10:08:18.271301: step: 100/533, loss: 0.06116313859820366 2023-01-21 10:08:19.427680: step: 104/533, loss: 0.7051877379417419 2023-01-21 10:08:20.600937: step: 108/533, loss: 0.05737247318029404 2023-01-21 10:08:21.807145: step: 112/533, loss: 0.04718508571386337 2023-01-21 10:08:22.957803: step: 116/533, loss: 0.07925932854413986 2023-01-21 10:08:24.113915: step: 120/533, loss: 0.17257767915725708 2023-01-21 10:08:25.284335: step: 124/533, loss: 0.025682760402560234 2023-01-21 10:08:26.449630: step: 128/533, loss: 0.014387511648237705 2023-01-21 10:08:27.563888: step: 132/533, loss: 0.010199499316513538 2023-01-21 10:08:28.744585: step: 136/533, loss: 0.08140130341053009 2023-01-21 10:08:29.881337: step: 140/533, loss: 0.12280035018920898 2023-01-21 10:08:31.063057: step: 144/533, loss: 0.013095760717988014 2023-01-21 10:08:32.225372: step: 148/533, loss: 0.01890258863568306 2023-01-21 10:08:33.361518: step: 152/533, loss: 0.028209973126649857 2023-01-21 10:08:34.551798: step: 156/533, loss: 0.06222410500049591 2023-01-21 10:08:35.710038: step: 160/533, loss: 0.043053437024354935 2023-01-21 10:08:36.920017: step: 164/533, loss: 0.01863894611597061 2023-01-21 10:08:38.098178: step: 168/533, loss: 0.1328442543745041 2023-01-21 10:08:39.278893: step: 172/533, loss: 0.014139366336166859 2023-01-21 10:08:40.489356: step: 176/533, loss: 0.17969512939453125 2023-01-21 10:08:41.703057: step: 180/533, loss: 0.1496437042951584 2023-01-21 10:08:42.897425: step: 184/533, loss: 0.07733212411403656 2023-01-21 10:08:44.084566: step: 188/533, loss: 0.04834775999188423 2023-01-21 10:08:45.271851: step: 192/533, loss: 0.05633850023150444 2023-01-21 10:08:46.472377: step: 196/533, loss: 0.07619714736938477 2023-01-21 10:08:47.651818: step: 200/533, loss: 0.13846178352832794 2023-01-21 10:08:48.814272: step: 204/533, loss: 0.26538050174713135 2023-01-21 10:08:49.998439: step: 208/533, loss: 0.09557390213012695 2023-01-21 10:08:51.212091: step: 212/533, loss: 0.17834702134132385 2023-01-21 10:08:52.360380: step: 216/533, loss: 0.1351640820503235 2023-01-21 10:08:53.524765: step: 220/533, loss: 0.007816505618393421 2023-01-21 10:08:54.717112: step: 224/533, loss: 0.12863358855247498 2023-01-21 10:08:55.916184: step: 228/533, loss: 0.025182437151670456 2023-01-21 10:08:57.104391: step: 232/533, loss: 0.0069138528779149055 2023-01-21 10:08:58.255772: step: 236/533, loss: 0.02809600904583931 2023-01-21 10:08:59.438571: step: 240/533, loss: 0.03894472494721413 2023-01-21 10:09:00.631451: step: 244/533, loss: 0.14274054765701294 2023-01-21 10:09:01.821197: step: 248/533, loss: 0.08537349104881287 2023-01-21 10:09:03.024808: step: 252/533, loss: 0.11126642674207687 2023-01-21 10:09:04.155444: step: 256/533, loss: 0.03346996381878853 2023-01-21 10:09:05.326297: step: 260/533, loss: 0.29826951026916504 2023-01-21 10:09:06.500378: step: 264/533, loss: 0.20949554443359375 2023-01-21 10:09:07.663738: step: 268/533, loss: 0.032239485532045364 2023-01-21 10:09:08.820436: step: 272/533, loss: 0.013752412050962448 2023-01-21 10:09:10.021398: step: 276/533, loss: 0.03142566606402397 2023-01-21 10:09:11.240997: step: 280/533, loss: 0.10265292972326279 2023-01-21 10:09:12.458977: step: 284/533, loss: 0.16017714142799377 2023-01-21 10:09:13.645390: step: 288/533, loss: 0.04518106207251549 2023-01-21 10:09:14.803530: step: 292/533, loss: 0.05074930191040039 2023-01-21 10:09:15.955334: step: 296/533, loss: 0.04854927957057953 2023-01-21 10:09:17.218336: step: 300/533, loss: 0.020022202283143997 2023-01-21 10:09:18.360626: step: 304/533, loss: 0.03747405856847763 2023-01-21 10:09:19.560000: step: 308/533, loss: 0.036989498883485794 2023-01-21 10:09:20.758179: step: 312/533, loss: 0.06270818412303925 2023-01-21 10:09:21.911772: step: 316/533, loss: 0.02303600311279297 2023-01-21 10:09:23.143494: step: 320/533, loss: 0.12723135948181152 2023-01-21 10:09:24.411699: step: 324/533, loss: 0.10956735908985138 2023-01-21 10:09:25.579534: step: 328/533, loss: 0.08900395035743713 2023-01-21 10:09:26.746891: step: 332/533, loss: 0.10268016159534454 2023-01-21 10:09:27.926794: step: 336/533, loss: 0.05550365522503853 2023-01-21 10:09:29.150042: step: 340/533, loss: 0.057116128504276276 2023-01-21 10:09:30.330568: step: 344/533, loss: 0.04482236132025719 2023-01-21 10:09:31.536469: step: 348/533, loss: 0.034926414489746094 2023-01-21 10:09:32.769916: step: 352/533, loss: 0.14513015747070312 2023-01-21 10:09:33.961712: step: 356/533, loss: 0.03851809352636337 2023-01-21 10:09:35.096797: step: 360/533, loss: 0.006290304474532604 2023-01-21 10:09:36.257451: step: 364/533, loss: 0.018521975725889206 2023-01-21 10:09:37.429026: step: 368/533, loss: 0.010516547597944736 2023-01-21 10:09:38.621486: step: 372/533, loss: 0.057587720453739166 2023-01-21 10:09:39.811934: step: 376/533, loss: 0.0536588691174984 2023-01-21 10:09:40.983000: step: 380/533, loss: 0.05293913185596466 2023-01-21 10:09:42.140260: step: 384/533, loss: 0.04788952320814133 2023-01-21 10:09:43.364275: step: 388/533, loss: 0.032726287841796875 2023-01-21 10:09:44.576116: step: 392/533, loss: 0.011357784271240234 2023-01-21 10:09:45.769472: step: 396/533, loss: 0.018758393824100494 2023-01-21 10:09:46.950355: step: 400/533, loss: 0.04614190384745598 2023-01-21 10:09:48.111344: step: 404/533, loss: 0.09147845208644867 2023-01-21 10:09:49.279401: step: 408/533, loss: 0.005666446406394243 2023-01-21 10:09:50.429062: step: 412/533, loss: 0.014667415991425514 2023-01-21 10:09:51.605789: step: 416/533, loss: 0.022772978991270065 2023-01-21 10:09:52.776526: step: 420/533, loss: 0.09483551234006882 2023-01-21 10:09:54.043038: step: 424/533, loss: 0.01939101330935955 2023-01-21 10:09:55.235335: step: 428/533, loss: 0.018538666889071465 2023-01-21 10:09:56.387350: step: 432/533, loss: 0.024607086554169655 2023-01-21 10:09:57.581646: step: 436/533, loss: 0.2652323246002197 2023-01-21 10:09:58.754773: step: 440/533, loss: 0.10097656399011612 2023-01-21 10:09:59.948365: step: 444/533, loss: 0.27409666776657104 2023-01-21 10:10:01.118336: step: 448/533, loss: 0.08909034729003906 2023-01-21 10:10:02.303467: step: 452/533, loss: 0.3133987486362457 2023-01-21 10:10:03.506178: step: 456/533, loss: 0.1810930371284485 2023-01-21 10:10:04.653082: step: 460/533, loss: 0.040762949734926224 2023-01-21 10:10:05.828564: step: 464/533, loss: 0.061676979064941406 2023-01-21 10:10:07.061138: step: 468/533, loss: 0.4010699987411499 2023-01-21 10:10:08.264242: step: 472/533, loss: 0.07260093837976456 2023-01-21 10:10:09.456038: step: 476/533, loss: 0.28557929396629333 2023-01-21 10:10:10.677346: step: 480/533, loss: 0.14278583228588104 2023-01-21 10:10:11.830944: step: 484/533, loss: 0.041107941418886185 2023-01-21 10:10:12.992909: step: 488/533, loss: 0.058317460119724274 2023-01-21 10:10:14.144491: step: 492/533, loss: 0.017256595194339752 2023-01-21 10:10:15.340014: step: 496/533, loss: 0.08629989624023438 2023-01-21 10:10:16.527125: step: 500/533, loss: 0.7166303992271423 2023-01-21 10:10:17.650902: step: 504/533, loss: 0.03475242108106613 2023-01-21 10:10:18.817917: step: 508/533, loss: 0.5990545749664307 2023-01-21 10:10:20.031101: step: 512/533, loss: 0.0847008228302002 2023-01-21 10:10:21.194762: step: 516/533, loss: 0.3145454525947571 2023-01-21 10:10:22.361020: step: 520/533, loss: 0.04445791244506836 2023-01-21 10:10:23.545767: step: 524/533, loss: 0.12211110442876816 2023-01-21 10:10:24.727778: step: 528/533, loss: 0.03650302812457085 2023-01-21 10:10:25.926096: step: 532/533, loss: 0.022868730127811432 2023-01-21 10:10:27.130435: step: 536/533, loss: 0.4974576234817505 2023-01-21 10:10:28.341529: step: 540/533, loss: 0.05934934690594673 2023-01-21 10:10:29.542364: step: 544/533, loss: 0.6201225519180298 2023-01-21 10:10:30.755330: step: 548/533, loss: 0.0677005797624588 2023-01-21 10:10:31.934116: step: 552/533, loss: 0.056618500500917435 2023-01-21 10:10:33.133673: step: 556/533, loss: 0.05579443275928497 2023-01-21 10:10:34.326493: step: 560/533, loss: 0.07386646419763565 2023-01-21 10:10:35.490335: step: 564/533, loss: 0.08603496849536896 2023-01-21 10:10:36.685879: step: 568/533, loss: 0.09671372920274734 2023-01-21 10:10:37.853899: step: 572/533, loss: 0.07626190781593323 2023-01-21 10:10:39.014046: step: 576/533, loss: 0.12589764595031738 2023-01-21 10:10:40.198736: step: 580/533, loss: 0.07626771926879883 2023-01-21 10:10:41.387039: step: 584/533, loss: 0.1767188161611557 2023-01-21 10:10:42.580914: step: 588/533, loss: 0.0031238081865012646 2023-01-21 10:10:43.770392: step: 592/533, loss: 0.05519752576947212 2023-01-21 10:10:44.981042: step: 596/533, loss: 0.042662620544433594 2023-01-21 10:10:46.154448: step: 600/533, loss: 0.16373272240161896 2023-01-21 10:10:47.298704: step: 604/533, loss: 0.012190532870590687 2023-01-21 10:10:48.460032: step: 608/533, loss: 0.20656099915504456 2023-01-21 10:10:49.600699: step: 612/533, loss: 0.025418471544981003 2023-01-21 10:10:50.759050: step: 616/533, loss: 0.005350637715309858 2023-01-21 10:10:51.928726: step: 620/533, loss: 0.12743797898292542 2023-01-21 10:10:53.117470: step: 624/533, loss: 0.1162261962890625 2023-01-21 10:10:54.291338: step: 628/533, loss: 0.03257916122674942 2023-01-21 10:10:55.421578: step: 632/533, loss: 0.004297447390854359 2023-01-21 10:10:56.605330: step: 636/533, loss: 0.15054330229759216 2023-01-21 10:10:57.743739: step: 640/533, loss: 0.058246467262506485 2023-01-21 10:10:58.915494: step: 644/533, loss: 0.047281697392463684 2023-01-21 10:11:00.095151: step: 648/533, loss: 0.013740921393036842 2023-01-21 10:11:01.306036: step: 652/533, loss: 1.265947937965393 2023-01-21 10:11:02.504992: step: 656/533, loss: 0.053545381873846054 2023-01-21 10:11:03.672743: step: 660/533, loss: 0.08232641965150833 2023-01-21 10:11:04.875955: step: 664/533, loss: 0.01431188639253378 2023-01-21 10:11:06.073831: step: 668/533, loss: 0.009477544575929642 2023-01-21 10:11:07.260802: step: 672/533, loss: 0.2790432572364807 2023-01-21 10:11:08.435704: step: 676/533, loss: 0.05599184334278107 2023-01-21 10:11:09.600000: step: 680/533, loss: 0.10838194191455841 2023-01-21 10:11:10.756674: step: 684/533, loss: 0.04290275648236275 2023-01-21 10:11:11.892276: step: 688/533, loss: 0.09902272373437881 2023-01-21 10:11:13.085457: step: 692/533, loss: 0.013250255957245827 2023-01-21 10:11:14.269056: step: 696/533, loss: 0.06524792313575745 2023-01-21 10:11:15.447965: step: 700/533, loss: 0.019201230257749557 2023-01-21 10:11:16.627044: step: 704/533, loss: 0.01433572731912136 2023-01-21 10:11:17.812330: step: 708/533, loss: 0.01153421401977539 2023-01-21 10:11:19.002157: step: 712/533, loss: 0.08900967240333557 2023-01-21 10:11:20.181735: step: 716/533, loss: 0.1447269320487976 2023-01-21 10:11:21.402010: step: 720/533, loss: 0.11871615052223206 2023-01-21 10:11:22.548046: step: 724/533, loss: 0.02577047422528267 2023-01-21 10:11:23.724255: step: 728/533, loss: 0.08094902336597443 2023-01-21 10:11:24.954627: step: 732/533, loss: 0.06724414229393005 2023-01-21 10:11:26.115638: step: 736/533, loss: 0.05045163631439209 2023-01-21 10:11:27.292928: step: 740/533, loss: 0.017099715769290924 2023-01-21 10:11:28.520313: step: 744/533, loss: 0.2363205850124359 2023-01-21 10:11:29.718921: step: 748/533, loss: 0.0421605110168457 2023-01-21 10:11:30.965928: step: 752/533, loss: 0.09516634792089462 2023-01-21 10:11:32.170672: step: 756/533, loss: 0.07830887287855148 2023-01-21 10:11:33.343536: step: 760/533, loss: 0.0657321959733963 2023-01-21 10:11:34.526165: step: 764/533, loss: 0.038173675537109375 2023-01-21 10:11:35.753388: step: 768/533, loss: 0.0770900696516037 2023-01-21 10:11:36.918086: step: 772/533, loss: 0.011811303906142712 2023-01-21 10:11:38.076324: step: 776/533, loss: 0.04569902643561363 2023-01-21 10:11:39.271442: step: 780/533, loss: 0.10124779492616653 2023-01-21 10:11:40.479203: step: 784/533, loss: 0.0378355048596859 2023-01-21 10:11:41.641952: step: 788/533, loss: 0.10238370299339294 2023-01-21 10:11:42.862045: step: 792/533, loss: 0.0881500244140625 2023-01-21 10:11:44.054373: step: 796/533, loss: 0.020652318373322487 2023-01-21 10:11:45.264131: step: 800/533, loss: 0.06561565399169922 2023-01-21 10:11:46.478852: step: 804/533, loss: 0.5448048114776611 2023-01-21 10:11:47.679766: step: 808/533, loss: 0.039614297449588776 2023-01-21 10:11:48.829912: step: 812/533, loss: 0.019077610224485397 2023-01-21 10:11:50.022919: step: 816/533, loss: 0.10220970958471298 2023-01-21 10:11:51.205920: step: 820/533, loss: 0.03624997287988663 2023-01-21 10:11:52.376387: step: 824/533, loss: 0.008788347244262695 2023-01-21 10:11:53.588836: step: 828/533, loss: 0.06470656394958496 2023-01-21 10:11:54.741615: step: 832/533, loss: 0.06646556407213211 2023-01-21 10:11:55.891916: step: 836/533, loss: 0.060130029916763306 2023-01-21 10:11:57.055215: step: 840/533, loss: 0.038254644721746445 2023-01-21 10:11:58.261578: step: 844/533, loss: 0.0641595870256424 2023-01-21 10:11:59.460629: step: 848/533, loss: 0.007167530246078968 2023-01-21 10:12:00.636223: step: 852/533, loss: 0.05634784698486328 2023-01-21 10:12:01.845452: step: 856/533, loss: 0.11353941261768341 2023-01-21 10:12:03.046245: step: 860/533, loss: 0.05974188446998596 2023-01-21 10:12:04.204374: step: 864/533, loss: 0.17184743285179138 2023-01-21 10:12:05.371644: step: 868/533, loss: 0.015978015959262848 2023-01-21 10:12:06.508506: step: 872/533, loss: 0.4191598892211914 2023-01-21 10:12:07.695458: step: 876/533, loss: 0.13737726211547852 2023-01-21 10:12:08.891512: step: 880/533, loss: 0.019935226067900658 2023-01-21 10:12:10.061465: step: 884/533, loss: 0.8298965692520142 2023-01-21 10:12:11.267138: step: 888/533, loss: 0.12010689079761505 2023-01-21 10:12:12.409232: step: 892/533, loss: 0.12362580001354218 2023-01-21 10:12:13.622900: step: 896/533, loss: 0.10992546379566193 2023-01-21 10:12:14.820979: step: 900/533, loss: 0.06212148815393448 2023-01-21 10:12:16.020558: step: 904/533, loss: 0.034334421157836914 2023-01-21 10:12:17.170258: step: 908/533, loss: 0.04552020877599716 2023-01-21 10:12:18.384514: step: 912/533, loss: 0.014476490207016468 2023-01-21 10:12:19.602774: step: 916/533, loss: 0.05216169357299805 2023-01-21 10:12:20.792058: step: 920/533, loss: 0.054259397089481354 2023-01-21 10:12:21.976439: step: 924/533, loss: 0.05218372121453285 2023-01-21 10:12:23.178621: step: 928/533, loss: 0.05224800109863281 2023-01-21 10:12:24.354727: step: 932/533, loss: 0.02460627630352974 2023-01-21 10:12:25.521858: step: 936/533, loss: 0.12872420251369476 2023-01-21 10:12:26.705849: step: 940/533, loss: 0.37724569439888 2023-01-21 10:12:27.875404: step: 944/533, loss: 0.015972351655364037 2023-01-21 10:12:29.027016: step: 948/533, loss: 0.017873143777251244 2023-01-21 10:12:30.204857: step: 952/533, loss: 0.017248056828975677 2023-01-21 10:12:31.364512: step: 956/533, loss: 0.02011718787252903 2023-01-21 10:12:32.484611: step: 960/533, loss: 0.11006336659193039 2023-01-21 10:12:33.664093: step: 964/533, loss: 0.02671186998486519 2023-01-21 10:12:34.861446: step: 968/533, loss: 0.04826552793383598 2023-01-21 10:12:36.037669: step: 972/533, loss: 0.02652893029153347 2023-01-21 10:12:37.220147: step: 976/533, loss: 0.11934099346399307 2023-01-21 10:12:38.419642: step: 980/533, loss: 0.008259296417236328 2023-01-21 10:12:39.617445: step: 984/533, loss: 0.08100166916847229 2023-01-21 10:12:40.759834: step: 988/533, loss: 0.031232452020049095 2023-01-21 10:12:41.911151: step: 992/533, loss: 0.06072964891791344 2023-01-21 10:12:43.044680: step: 996/533, loss: 0.01717977598309517 2023-01-21 10:12:44.251501: step: 1000/533, loss: 0.017211150377988815 2023-01-21 10:12:45.465164: step: 1004/533, loss: 0.034348390996456146 2023-01-21 10:12:46.627298: step: 1008/533, loss: 0.034008074551820755 2023-01-21 10:12:47.802355: step: 1012/533, loss: 0.04258880764245987 2023-01-21 10:12:48.944244: step: 1016/533, loss: 0.09461527317762375 2023-01-21 10:12:50.113986: step: 1020/533, loss: 0.1257382333278656 2023-01-21 10:12:51.270241: step: 1024/533, loss: 0.10754475742578506 2023-01-21 10:12:52.413810: step: 1028/533, loss: 0.09646854549646378 2023-01-21 10:12:53.577209: step: 1032/533, loss: 0.0010723829036578536 2023-01-21 10:12:54.767882: step: 1036/533, loss: 0.07256393879652023 2023-01-21 10:12:55.927813: step: 1040/533, loss: 0.07233639061450958 2023-01-21 10:12:57.146462: step: 1044/533, loss: 0.06914357841014862 2023-01-21 10:12:58.378369: step: 1048/533, loss: 0.03955082967877388 2023-01-21 10:12:59.556778: step: 1052/533, loss: 0.3587758243083954 2023-01-21 10:13:00.721375: step: 1056/533, loss: 0.06304188072681427 2023-01-21 10:13:01.901792: step: 1060/533, loss: 0.06420469284057617 2023-01-21 10:13:03.074329: step: 1064/533, loss: 0.057447340339422226 2023-01-21 10:13:04.240569: step: 1068/533, loss: 0.027955342084169388 2023-01-21 10:13:05.404517: step: 1072/533, loss: 0.1225438117980957 2023-01-21 10:13:06.578278: step: 1076/533, loss: 0.11520595848560333 2023-01-21 10:13:07.746285: step: 1080/533, loss: 0.16201798617839813 2023-01-21 10:13:08.938313: step: 1084/533, loss: 0.12088815867900848 2023-01-21 10:13:10.110214: step: 1088/533, loss: 0.04511423408985138 2023-01-21 10:13:11.296088: step: 1092/533, loss: 0.0017524241702631116 2023-01-21 10:13:12.447524: step: 1096/533, loss: 0.10824938118457794 2023-01-21 10:13:13.656366: step: 1100/533, loss: 0.07720164954662323 2023-01-21 10:13:14.860856: step: 1104/533, loss: 0.1040860116481781 2023-01-21 10:13:16.029588: step: 1108/533, loss: 0.034945059567689896 2023-01-21 10:13:17.169610: step: 1112/533, loss: 0.04620242491364479 2023-01-21 10:13:18.322285: step: 1116/533, loss: 0.07007589191198349 2023-01-21 10:13:19.496874: step: 1120/533, loss: 0.035433389246463776 2023-01-21 10:13:20.692575: step: 1124/533, loss: 0.008829689584672451 2023-01-21 10:13:21.931285: step: 1128/533, loss: 0.09057541191577911 2023-01-21 10:13:23.176234: step: 1132/533, loss: 0.08630113303661346 2023-01-21 10:13:24.351920: step: 1136/533, loss: 0.4293048679828644 2023-01-21 10:13:25.530531: step: 1140/533, loss: 0.029358576983213425 2023-01-21 10:13:26.708055: step: 1144/533, loss: 0.19015425443649292 2023-01-21 10:13:27.856655: step: 1148/533, loss: 0.04289841651916504 2023-01-21 10:13:29.037602: step: 1152/533, loss: 0.12921875715255737 2023-01-21 10:13:30.219297: step: 1156/533, loss: 0.08202410489320755 2023-01-21 10:13:31.394196: step: 1160/533, loss: 0.06566992402076721 2023-01-21 10:13:32.541847: step: 1164/533, loss: 0.272839218378067 2023-01-21 10:13:33.717203: step: 1168/533, loss: 0.07983341068029404 2023-01-21 10:13:34.852634: step: 1172/533, loss: 0.0463099479675293 2023-01-21 10:13:35.991215: step: 1176/533, loss: 0.5903288125991821 2023-01-21 10:13:37.127309: step: 1180/533, loss: 0.11411400139331818 2023-01-21 10:13:38.294586: step: 1184/533, loss: 0.2203051596879959 2023-01-21 10:13:39.474919: step: 1188/533, loss: 0.2530983090400696 2023-01-21 10:13:40.643410: step: 1192/533, loss: 0.06348276138305664 2023-01-21 10:13:41.809089: step: 1196/533, loss: 0.09556732326745987 2023-01-21 10:13:42.969130: step: 1200/533, loss: 0.06467476487159729 2023-01-21 10:13:44.134487: step: 1204/533, loss: 0.07003694027662277 2023-01-21 10:13:45.310488: step: 1208/533, loss: 0.019532132893800735 2023-01-21 10:13:46.477815: step: 1212/533, loss: 0.06544151157140732 2023-01-21 10:13:47.672454: step: 1216/533, loss: 0.10628766566514969 2023-01-21 10:13:48.852950: step: 1220/533, loss: 0.025108743458986282 2023-01-21 10:13:50.033388: step: 1224/533, loss: 0.046537257730960846 2023-01-21 10:13:51.190990: step: 1228/533, loss: 0.04871067777276039 2023-01-21 10:13:52.385170: step: 1232/533, loss: 0.08525543659925461 2023-01-21 10:13:53.541392: step: 1236/533, loss: 0.020444059744477272 2023-01-21 10:13:54.675221: step: 1240/533, loss: 0.026122761890292168 2023-01-21 10:13:55.895668: step: 1244/533, loss: 0.09374479949474335 2023-01-21 10:13:57.066943: step: 1248/533, loss: 0.07252321392297745 2023-01-21 10:13:58.215885: step: 1252/533, loss: 0.1519780158996582 2023-01-21 10:13:59.405030: step: 1256/533, loss: 0.22880668938159943 2023-01-21 10:14:00.608229: step: 1260/533, loss: 0.07265176624059677 2023-01-21 10:14:01.804017: step: 1264/533, loss: 0.09801264107227325 2023-01-21 10:14:02.978938: step: 1268/533, loss: 0.005519008729606867 2023-01-21 10:14:04.181750: step: 1272/533, loss: 0.01893758773803711 2023-01-21 10:14:05.365253: step: 1276/533, loss: 0.006653213407844305 2023-01-21 10:14:06.530130: step: 1280/533, loss: 0.04139719158411026 2023-01-21 10:14:07.695178: step: 1284/533, loss: 0.07049236446619034 2023-01-21 10:14:08.848949: step: 1288/533, loss: 0.08337441086769104 2023-01-21 10:14:10.021041: step: 1292/533, loss: 0.030786896124482155 2023-01-21 10:14:11.216762: step: 1296/533, loss: 0.02945718728005886 2023-01-21 10:14:12.407904: step: 1300/533, loss: 0.16059261560440063 2023-01-21 10:14:13.596604: step: 1304/533, loss: 0.11569375544786453 2023-01-21 10:14:14.807489: step: 1308/533, loss: 0.06544365733861923 2023-01-21 10:14:15.977392: step: 1312/533, loss: 0.015305234119296074 2023-01-21 10:14:17.153126: step: 1316/533, loss: 0.11670169979333878 2023-01-21 10:14:18.304661: step: 1320/533, loss: 0.028336383402347565 2023-01-21 10:14:19.486715: step: 1324/533, loss: 0.005647182464599609 2023-01-21 10:14:20.686675: step: 1328/533, loss: 0.722735583782196 2023-01-21 10:14:21.823432: step: 1332/533, loss: 0.08525466918945312 2023-01-21 10:14:23.012723: step: 1336/533, loss: 0.1007390096783638 2023-01-21 10:14:24.197536: step: 1340/533, loss: 0.04582986980676651 2023-01-21 10:14:25.352194: step: 1344/533, loss: 0.23077526688575745 2023-01-21 10:14:26.512108: step: 1348/533, loss: 0.15750399231910706 2023-01-21 10:14:27.707009: step: 1352/533, loss: 0.4472717344760895 2023-01-21 10:14:28.892925: step: 1356/533, loss: 0.05208740383386612 2023-01-21 10:14:30.088749: step: 1360/533, loss: 0.0662989616394043 2023-01-21 10:14:31.258131: step: 1364/533, loss: 0.04457426071166992 2023-01-21 10:14:32.450060: step: 1368/533, loss: 0.08263588696718216 2023-01-21 10:14:33.613830: step: 1372/533, loss: 0.0627906322479248 2023-01-21 10:14:34.761796: step: 1376/533, loss: 0.1397140920162201 2023-01-21 10:14:35.926815: step: 1380/533, loss: 0.043340303003787994 2023-01-21 10:14:37.101611: step: 1384/533, loss: 0.38710179924964905 2023-01-21 10:14:38.259093: step: 1388/533, loss: 0.01959371566772461 2023-01-21 10:14:39.413579: step: 1392/533, loss: 0.06831777095794678 2023-01-21 10:14:40.609449: step: 1396/533, loss: 0.036202672868967056 2023-01-21 10:14:41.779035: step: 1400/533, loss: 0.12237277626991272 2023-01-21 10:14:42.990186: step: 1404/533, loss: 0.06173763424158096 2023-01-21 10:14:44.170645: step: 1408/533, loss: 0.05011396482586861 2023-01-21 10:14:45.362902: step: 1412/533, loss: 0.23885507881641388 2023-01-21 10:14:46.504485: step: 1416/533, loss: 0.09139814227819443 2023-01-21 10:14:47.660843: step: 1420/533, loss: 0.42762821912765503 2023-01-21 10:14:48.832442: step: 1424/533, loss: 0.11530523747205734 2023-01-21 10:14:49.990628: step: 1428/533, loss: 0.11973657459020615 2023-01-21 10:14:51.167885: step: 1432/533, loss: 0.21258427202701569 2023-01-21 10:14:52.355414: step: 1436/533, loss: 0.04773378372192383 2023-01-21 10:14:53.575325: step: 1440/533, loss: 0.0718592181801796 2023-01-21 10:14:54.784641: step: 1444/533, loss: 0.006921005435287952 2023-01-21 10:14:55.992010: step: 1448/533, loss: 0.01573786698281765 2023-01-21 10:14:57.243835: step: 1452/533, loss: 0.33907264471054077 2023-01-21 10:14:58.395726: step: 1456/533, loss: 0.08991508185863495 2023-01-21 10:14:59.525677: step: 1460/533, loss: 0.09660325199365616 2023-01-21 10:15:00.702990: step: 1464/533, loss: 0.06763105094432831 2023-01-21 10:15:01.893099: step: 1468/533, loss: 0.2460470199584961 2023-01-21 10:15:03.053151: step: 1472/533, loss: 0.01610422134399414 2023-01-21 10:15:04.216467: step: 1476/533, loss: 0.025304174050688744 2023-01-21 10:15:05.356947: step: 1480/533, loss: 0.06752438098192215 2023-01-21 10:15:06.558505: step: 1484/533, loss: 0.09731343388557434 2023-01-21 10:15:07.768806: step: 1488/533, loss: 0.04540739208459854 2023-01-21 10:15:08.937271: step: 1492/533, loss: 0.08259592205286026 2023-01-21 10:15:10.084658: step: 1496/533, loss: 0.03957099840044975 2023-01-21 10:15:11.235472: step: 1500/533, loss: 0.06335163116455078 2023-01-21 10:15:12.421461: step: 1504/533, loss: 0.0810893103480339 2023-01-21 10:15:13.583171: step: 1508/533, loss: 0.013915968127548695 2023-01-21 10:15:14.742639: step: 1512/533, loss: 0.1194813996553421 2023-01-21 10:15:15.955169: step: 1516/533, loss: 0.0778312236070633 2023-01-21 10:15:17.116402: step: 1520/533, loss: 0.06056080013513565 2023-01-21 10:15:18.282933: step: 1524/533, loss: 0.07550602406263351 2023-01-21 10:15:19.414250: step: 1528/533, loss: 0.0341365821659565 2023-01-21 10:15:20.603725: step: 1532/533, loss: 0.12445640563964844 2023-01-21 10:15:21.781388: step: 1536/533, loss: 0.3284502923488617 2023-01-21 10:15:22.981208: step: 1540/533, loss: 0.12210474163293839 2023-01-21 10:15:24.210134: step: 1544/533, loss: 0.08749590069055557 2023-01-21 10:15:25.392616: step: 1548/533, loss: 0.13944703340530396 2023-01-21 10:15:26.580117: step: 1552/533, loss: 0.028626251965761185 2023-01-21 10:15:27.751143: step: 1556/533, loss: 0.2883397042751312 2023-01-21 10:15:28.894613: step: 1560/533, loss: 0.19004663825035095 2023-01-21 10:15:30.059427: step: 1564/533, loss: 0.17749014496803284 2023-01-21 10:15:31.235695: step: 1568/533, loss: 0.12720413506031036 2023-01-21 10:15:32.433368: step: 1572/533, loss: 0.5603672862052917 2023-01-21 10:15:33.625192: step: 1576/533, loss: 0.018171025440096855 2023-01-21 10:15:34.815657: step: 1580/533, loss: 0.40120047330856323 2023-01-21 10:15:36.051702: step: 1584/533, loss: 0.015099716372787952 2023-01-21 10:15:37.239950: step: 1588/533, loss: 0.07071518898010254 2023-01-21 10:15:38.375854: step: 1592/533, loss: 0.008141708560287952 2023-01-21 10:15:39.555019: step: 1596/533, loss: 0.09460707008838654 2023-01-21 10:15:40.744563: step: 1600/533, loss: 0.17722895741462708 2023-01-21 10:15:41.907121: step: 1604/533, loss: 0.09161920845508575 2023-01-21 10:15:43.080387: step: 1608/533, loss: 0.04561030492186546 2023-01-21 10:15:44.306041: step: 1612/533, loss: 0.15457019209861755 2023-01-21 10:15:45.514787: step: 1616/533, loss: 0.15258640050888062 2023-01-21 10:15:46.689008: step: 1620/533, loss: 0.017450714483857155 2023-01-21 10:15:47.881618: step: 1624/533, loss: 0.07702770829200745 2023-01-21 10:15:49.077821: step: 1628/533, loss: 0.08610725402832031 2023-01-21 10:15:50.241500: step: 1632/533, loss: 0.02127513848245144 2023-01-21 10:15:51.492423: step: 1636/533, loss: 0.07588864117860794 2023-01-21 10:15:52.655680: step: 1640/533, loss: 0.040285490453243256 2023-01-21 10:15:53.862296: step: 1644/533, loss: 0.06690329313278198 2023-01-21 10:15:55.028311: step: 1648/533, loss: 0.027987100183963776 2023-01-21 10:15:56.228817: step: 1652/533, loss: 0.021415090188384056 2023-01-21 10:15:57.392511: step: 1656/533, loss: 0.1005639135837555 2023-01-21 10:15:58.615917: step: 1660/533, loss: 0.6033145785331726 2023-01-21 10:15:59.809913: step: 1664/533, loss: 0.057854942977428436 2023-01-21 10:16:00.971632: step: 1668/533, loss: 0.13870087265968323 2023-01-21 10:16:02.137761: step: 1672/533, loss: 0.06631994247436523 2023-01-21 10:16:03.380386: step: 1676/533, loss: 0.04903984069824219 2023-01-21 10:16:04.594361: step: 1680/533, loss: 0.05277226120233536 2023-01-21 10:16:05.756862: step: 1684/533, loss: 0.12280307710170746 2023-01-21 10:16:06.962925: step: 1688/533, loss: 0.12222793698310852 2023-01-21 10:16:08.141916: step: 1692/533, loss: 0.0692349448800087 2023-01-21 10:16:09.350778: step: 1696/533, loss: 0.02035648748278618 2023-01-21 10:16:10.568547: step: 1700/533, loss: 0.055666401982307434 2023-01-21 10:16:11.749981: step: 1704/533, loss: 0.05515652149915695 2023-01-21 10:16:12.899666: step: 1708/533, loss: 0.09315033257007599 2023-01-21 10:16:14.084675: step: 1712/533, loss: 0.00919046439230442 2023-01-21 10:16:15.246281: step: 1716/533, loss: 0.056665562093257904 2023-01-21 10:16:16.420430: step: 1720/533, loss: 0.10833263397216797 2023-01-21 10:16:17.592490: step: 1724/533, loss: 0.08019638061523438 2023-01-21 10:16:18.768281: step: 1728/533, loss: 0.03905963897705078 2023-01-21 10:16:19.970217: step: 1732/533, loss: 0.04802331700921059 2023-01-21 10:16:21.199854: step: 1736/533, loss: 0.1047694981098175 2023-01-21 10:16:22.396710: step: 1740/533, loss: 0.039585210382938385 2023-01-21 10:16:23.569421: step: 1744/533, loss: 0.005497646518051624 2023-01-21 10:16:24.761055: step: 1748/533, loss: 0.3444347083568573 2023-01-21 10:16:26.019055: step: 1752/533, loss: 0.08689117431640625 2023-01-21 10:16:27.197389: step: 1756/533, loss: 0.016423894092440605 2023-01-21 10:16:28.400205: step: 1760/533, loss: 0.13254165649414062 2023-01-21 10:16:29.619916: step: 1764/533, loss: 0.05878373235464096 2023-01-21 10:16:30.804977: step: 1768/533, loss: 0.03589468076825142 2023-01-21 10:16:31.968525: step: 1772/533, loss: 0.6198861598968506 2023-01-21 10:16:33.119172: step: 1776/533, loss: 0.11236973106861115 2023-01-21 10:16:34.293782: step: 1780/533, loss: 0.12118339538574219 2023-01-21 10:16:35.487263: step: 1784/533, loss: 0.03707880899310112 2023-01-21 10:16:36.651758: step: 1788/533, loss: 0.004126453772187233 2023-01-21 10:16:37.810315: step: 1792/533, loss: 0.01949899084866047 2023-01-21 10:16:38.995399: step: 1796/533, loss: 0.2387535274028778 2023-01-21 10:16:40.144562: step: 1800/533, loss: 0.02178182639181614 2023-01-21 10:16:41.360294: step: 1804/533, loss: 0.1231098398566246 2023-01-21 10:16:42.524495: step: 1808/533, loss: 0.05806431919336319 2023-01-21 10:16:43.718203: step: 1812/533, loss: 0.015900611877441406 2023-01-21 10:16:44.941258: step: 1816/533, loss: 0.036623287945985794 2023-01-21 10:16:46.139716: step: 1820/533, loss: 0.044013749808073044 2023-01-21 10:16:47.326661: step: 1824/533, loss: 0.03783436119556427 2023-01-21 10:16:48.485203: step: 1828/533, loss: 0.027425026521086693 2023-01-21 10:16:49.668064: step: 1832/533, loss: 0.06180458143353462 2023-01-21 10:16:50.901146: step: 1836/533, loss: 0.025116827338933945 2023-01-21 10:16:52.122855: step: 1840/533, loss: 0.04083309322595596 2023-01-21 10:16:53.338692: step: 1844/533, loss: 0.058376505970954895 2023-01-21 10:16:54.543659: step: 1848/533, loss: 0.24160251021385193 2023-01-21 10:16:55.713053: step: 1852/533, loss: 0.013720654882490635 2023-01-21 10:16:56.902337: step: 1856/533, loss: 0.0057541849091649055 2023-01-21 10:16:58.061687: step: 1860/533, loss: 0.001074123429134488 2023-01-21 10:16:59.186782: step: 1864/533, loss: 0.03571796417236328 2023-01-21 10:17:00.349176: step: 1868/533, loss: 0.041402243077754974 2023-01-21 10:17:01.519116: step: 1872/533, loss: 0.02385539934039116 2023-01-21 10:17:02.698031: step: 1876/533, loss: 0.048597145825624466 2023-01-21 10:17:03.859895: step: 1880/533, loss: 0.08617869019508362 2023-01-21 10:17:05.053100: step: 1884/533, loss: 0.058133698999881744 2023-01-21 10:17:06.257716: step: 1888/533, loss: 0.043524548411369324 2023-01-21 10:17:07.429471: step: 1892/533, loss: 0.11809349060058594 2023-01-21 10:17:08.636164: step: 1896/533, loss: 0.12552142143249512 2023-01-21 10:17:09.807984: step: 1900/533, loss: 0.08240270614624023 2023-01-21 10:17:10.979232: step: 1904/533, loss: 0.025552939623594284 2023-01-21 10:17:12.148081: step: 1908/533, loss: 0.02073340304195881 2023-01-21 10:17:13.332686: step: 1912/533, loss: 0.12882208824157715 2023-01-21 10:17:14.529089: step: 1916/533, loss: 0.1296786367893219 2023-01-21 10:17:15.709695: step: 1920/533, loss: 0.09653263539075851 2023-01-21 10:17:16.890157: step: 1924/533, loss: 0.07321052253246307 2023-01-21 10:17:18.046375: step: 1928/533, loss: 0.12343092262744904 2023-01-21 10:17:19.241477: step: 1932/533, loss: 0.06080026552081108 2023-01-21 10:17:20.433533: step: 1936/533, loss: 0.03781747817993164 2023-01-21 10:17:21.612128: step: 1940/533, loss: 0.025562716647982597 2023-01-21 10:17:22.790945: step: 1944/533, loss: 0.11140938103199005 2023-01-21 10:17:23.983371: step: 1948/533, loss: 0.030676936730742455 2023-01-21 10:17:25.129658: step: 1952/533, loss: 0.06337670981884003 2023-01-21 10:17:26.312482: step: 1956/533, loss: 0.025703812018036842 2023-01-21 10:17:27.490622: step: 1960/533, loss: 0.11980142444372177 2023-01-21 10:17:28.646421: step: 1964/533, loss: 0.08251719921827316 2023-01-21 10:17:29.814957: step: 1968/533, loss: 0.21277998387813568 2023-01-21 10:17:31.011444: step: 1972/533, loss: 0.05088827386498451 2023-01-21 10:17:32.175650: step: 1976/533, loss: 0.3476164638996124 2023-01-21 10:17:33.338081: step: 1980/533, loss: 0.07201957702636719 2023-01-21 10:17:34.517360: step: 1984/533, loss: 0.02053804323077202 2023-01-21 10:17:35.670798: step: 1988/533, loss: 0.3541708290576935 2023-01-21 10:17:36.854020: step: 1992/533, loss: 0.04240398481488228 2023-01-21 10:17:38.055020: step: 1996/533, loss: 0.13211165368556976 2023-01-21 10:17:39.227566: step: 2000/533, loss: 0.10249615460634232 2023-01-21 10:17:40.428001: step: 2004/533, loss: 0.1062891036272049 2023-01-21 10:17:41.638672: step: 2008/533, loss: 0.16611194610595703 2023-01-21 10:17:42.816965: step: 2012/533, loss: 0.053447723388671875 2023-01-21 10:17:43.962545: step: 2016/533, loss: 1.185105562210083 2023-01-21 10:17:45.144358: step: 2020/533, loss: 0.11249179393053055 2023-01-21 10:17:46.355002: step: 2024/533, loss: 0.18461018800735474 2023-01-21 10:17:47.513786: step: 2028/533, loss: 0.07553768157958984 2023-01-21 10:17:48.682264: step: 2032/533, loss: 0.1200229674577713 2023-01-21 10:17:49.836467: step: 2036/533, loss: 0.1078365296125412 2023-01-21 10:17:50.984248: step: 2040/533, loss: 0.09825363755226135 2023-01-21 10:17:52.138057: step: 2044/533, loss: 0.023118305951356888 2023-01-21 10:17:53.329206: step: 2048/533, loss: 0.03482646867632866 2023-01-21 10:17:54.517114: step: 2052/533, loss: 0.019292794167995453 2023-01-21 10:17:55.759767: step: 2056/533, loss: 0.033036209642887115 2023-01-21 10:17:56.909433: step: 2060/533, loss: 0.03395429998636246 2023-01-21 10:17:58.099802: step: 2064/533, loss: 0.04866061359643936 2023-01-21 10:17:59.258301: step: 2068/533, loss: 0.1779816597700119 2023-01-21 10:18:00.430364: step: 2072/533, loss: 0.5045151114463806 2023-01-21 10:18:01.613342: step: 2076/533, loss: 1.1006358861923218 2023-01-21 10:18:02.783511: step: 2080/533, loss: 0.1859375685453415 2023-01-21 10:18:03.957858: step: 2084/533, loss: 0.150996595621109 2023-01-21 10:18:05.194731: step: 2088/533, loss: 0.024134159088134766 2023-01-21 10:18:06.363053: step: 2092/533, loss: 0.07079525291919708 2023-01-21 10:18:07.572822: step: 2096/533, loss: 0.10572786629199982 2023-01-21 10:18:08.719544: step: 2100/533, loss: 0.07872743904590607 2023-01-21 10:18:09.861238: step: 2104/533, loss: 0.12590566277503967 2023-01-21 10:18:11.015847: step: 2108/533, loss: 0.0893617644906044 2023-01-21 10:18:12.196821: step: 2112/533, loss: 0.054971419274806976 2023-01-21 10:18:13.347008: step: 2116/533, loss: 0.022372521460056305 2023-01-21 10:18:14.529037: step: 2120/533, loss: 0.04605302959680557 2023-01-21 10:18:15.734656: step: 2124/533, loss: 0.014611768536269665 2023-01-21 10:18:16.901059: step: 2128/533, loss: 0.1182146966457367 2023-01-21 10:18:18.040745: step: 2132/533, loss: 0.0826452299952507 ================================================== Loss: 0.104 -------------------- Dev: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Test: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Russian: {'event': {'p': 0.38636363636363635, 'r': 0.4722222222222222, 'f1': 0.425}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} New best chinese model... New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 10:19:09.737882: step: 4/533, loss: 0.045775607228279114 2023-01-21 10:19:10.922492: step: 8/533, loss: 0.041579537093639374 2023-01-21 10:19:12.126647: step: 12/533, loss: 0.08029906451702118 2023-01-21 10:19:13.288850: step: 16/533, loss: 0.03169608116149902 2023-01-21 10:19:14.472307: step: 20/533, loss: 0.028263472020626068 2023-01-21 10:19:15.662932: step: 24/533, loss: 0.30895429849624634 2023-01-21 10:19:16.855231: step: 28/533, loss: 0.10805211961269379 2023-01-21 10:19:18.043802: step: 32/533, loss: 0.07137098908424377 2023-01-21 10:19:19.223318: step: 36/533, loss: 0.02403836138546467 2023-01-21 10:19:20.438690: step: 40/533, loss: 0.06653881072998047 2023-01-21 10:19:21.620502: step: 44/533, loss: 0.020636750385165215 2023-01-21 10:19:22.814549: step: 48/533, loss: 0.010600566864013672 2023-01-21 10:19:24.033829: step: 52/533, loss: 0.04670543596148491 2023-01-21 10:19:25.223260: step: 56/533, loss: 0.27407702803611755 2023-01-21 10:19:26.430192: step: 60/533, loss: 0.08262339234352112 2023-01-21 10:19:27.620757: step: 64/533, loss: 0.05509508401155472 2023-01-21 10:19:28.836223: step: 68/533, loss: 0.13145074248313904 2023-01-21 10:19:30.044494: step: 72/533, loss: 0.06372890621423721 2023-01-21 10:19:31.199526: step: 76/533, loss: 0.15012378990650177 2023-01-21 10:19:32.393413: step: 80/533, loss: 0.001420688582584262 2023-01-21 10:19:33.561298: step: 84/533, loss: 0.08213672786951065 2023-01-21 10:19:34.731342: step: 88/533, loss: 0.26611536741256714 2023-01-21 10:19:35.948392: step: 92/533, loss: 0.05539955943822861 2023-01-21 10:19:37.101509: step: 96/533, loss: 0.42285194993019104 2023-01-21 10:19:38.285804: step: 100/533, loss: 0.08684978634119034 2023-01-21 10:19:39.462147: step: 104/533, loss: 0.010871458798646927 2023-01-21 10:19:40.636765: step: 108/533, loss: 0.2614063322544098 2023-01-21 10:19:41.856813: step: 112/533, loss: 0.18899306654930115 2023-01-21 10:19:43.047266: step: 116/533, loss: 0.009564495645463467 2023-01-21 10:19:44.182711: step: 120/533, loss: 0.014503668993711472 2023-01-21 10:19:45.413935: step: 124/533, loss: 0.05736379697918892 2023-01-21 10:19:46.589943: step: 128/533, loss: 0.2594629228115082 2023-01-21 10:19:47.778563: step: 132/533, loss: 0.12998512387275696 2023-01-21 10:19:48.951059: step: 136/533, loss: 0.0793793722987175 2023-01-21 10:19:50.134960: step: 140/533, loss: 0.002503824420273304 2023-01-21 10:19:51.275157: step: 144/533, loss: 0.07496003806591034 2023-01-21 10:19:52.418080: step: 148/533, loss: 0.023552369326353073 2023-01-21 10:19:53.600843: step: 152/533, loss: 0.05401415750384331 2023-01-21 10:19:54.754233: step: 156/533, loss: 0.00713500939309597 2023-01-21 10:19:55.958027: step: 160/533, loss: 0.06813764572143555 2023-01-21 10:19:57.107190: step: 164/533, loss: 0.023251915350556374 2023-01-21 10:19:58.318223: step: 168/533, loss: 0.032544899731874466 2023-01-21 10:19:59.487244: step: 172/533, loss: 0.11059360206127167 2023-01-21 10:20:00.642252: step: 176/533, loss: 0.09497089684009552 2023-01-21 10:20:01.808228: step: 180/533, loss: 0.02761373668909073 2023-01-21 10:20:02.987996: step: 184/533, loss: 0.002765703247860074 2023-01-21 10:20:04.121400: step: 188/533, loss: 0.0524420291185379 2023-01-21 10:20:05.303960: step: 192/533, loss: 0.2073441445827484 2023-01-21 10:20:06.464478: step: 196/533, loss: 0.02808818779885769 2023-01-21 10:20:07.658625: step: 200/533, loss: 0.072445347905159 2023-01-21 10:20:08.844349: step: 204/533, loss: 0.13380718231201172 2023-01-21 10:20:10.027254: step: 208/533, loss: 0.03206147998571396 2023-01-21 10:20:11.193808: step: 212/533, loss: 0.003026771591976285 2023-01-21 10:20:12.351089: step: 216/533, loss: 0.1567460149526596 2023-01-21 10:20:13.536822: step: 220/533, loss: 0.03535861894488335 2023-01-21 10:20:14.706480: step: 224/533, loss: 0.09771156311035156 2023-01-21 10:20:15.869838: step: 228/533, loss: 0.17733033001422882 2023-01-21 10:20:17.031561: step: 232/533, loss: 0.1689428836107254 2023-01-21 10:20:18.196718: step: 236/533, loss: 0.1843942105770111 2023-01-21 10:20:19.410368: step: 240/533, loss: 0.08619699627161026 2023-01-21 10:20:20.592041: step: 244/533, loss: 0.08284597098827362 2023-01-21 10:20:21.825399: step: 248/533, loss: 0.06134033203125 2023-01-21 10:20:23.016293: step: 252/533, loss: 0.09902391582727432 2023-01-21 10:20:24.157540: step: 256/533, loss: 0.06408510357141495 2023-01-21 10:20:25.325053: step: 260/533, loss: 0.2765384912490845 2023-01-21 10:20:26.501142: step: 264/533, loss: 0.061306096613407135 2023-01-21 10:20:27.683034: step: 268/533, loss: 0.02085442654788494 2023-01-21 10:20:28.819647: step: 272/533, loss: 0.04109468683600426 2023-01-21 10:20:30.009545: step: 276/533, loss: 0.02451331540942192 2023-01-21 10:20:31.200869: step: 280/533, loss: 0.005088329315185547 2023-01-21 10:20:32.424970: step: 284/533, loss: 0.01680002361536026 2023-01-21 10:20:33.637008: step: 288/533, loss: 0.18105259537696838 2023-01-21 10:20:34.805889: step: 292/533, loss: 0.02563619613647461 2023-01-21 10:20:35.938786: step: 296/533, loss: 0.07982616126537323 2023-01-21 10:20:37.101339: step: 300/533, loss: 0.09824800491333008 2023-01-21 10:20:38.258737: step: 304/533, loss: 0.11986671388149261 2023-01-21 10:20:39.451638: step: 308/533, loss: 0.03739624097943306 2023-01-21 10:20:40.617196: step: 312/533, loss: 0.06771669536828995 2023-01-21 10:20:41.785963: step: 316/533, loss: 0.0616641640663147 2023-01-21 10:20:42.941769: step: 320/533, loss: 0.02994365803897381 2023-01-21 10:20:44.096035: step: 324/533, loss: 0.017020177096128464 2023-01-21 10:20:45.276861: step: 328/533, loss: 0.06182847172021866 2023-01-21 10:20:46.434817: step: 332/533, loss: 0.021282721310853958 2023-01-21 10:20:47.569613: step: 336/533, loss: 0.01936183124780655 2023-01-21 10:20:48.778658: step: 340/533, loss: 0.08292493969202042 2023-01-21 10:20:49.956882: step: 344/533, loss: 0.02923121489584446 2023-01-21 10:20:51.133390: step: 348/533, loss: 0.01003417931497097 2023-01-21 10:20:52.286202: step: 352/533, loss: 0.04278698191046715 2023-01-21 10:20:53.500757: step: 356/533, loss: 0.0335271842777729 2023-01-21 10:20:54.724549: step: 360/533, loss: 0.06862793117761612 2023-01-21 10:20:55.915051: step: 364/533, loss: 0.016425227746367455 2023-01-21 10:20:57.063764: step: 368/533, loss: 0.038833893835544586 2023-01-21 10:20:58.259096: step: 372/533, loss: 0.16099020838737488 2023-01-21 10:20:59.455646: step: 376/533, loss: 0.08399849385023117 2023-01-21 10:21:00.637290: step: 380/533, loss: 0.06945972889661789 2023-01-21 10:21:01.790188: step: 384/533, loss: 0.08206329494714737 2023-01-21 10:21:02.975397: step: 388/533, loss: 0.001986599061638117 2023-01-21 10:21:04.143962: step: 392/533, loss: 0.09394264221191406 2023-01-21 10:21:05.303041: step: 396/533, loss: 0.05575890466570854 2023-01-21 10:21:06.520368: step: 400/533, loss: 0.0068223001435399055 2023-01-21 10:21:07.680920: step: 404/533, loss: 0.04055070877075195 2023-01-21 10:21:08.831992: step: 408/533, loss: 0.02395925484597683 2023-01-21 10:21:09.970441: step: 412/533, loss: 0.008512115105986595 2023-01-21 10:21:11.113830: step: 416/533, loss: 0.03124713897705078 2023-01-21 10:21:12.309794: step: 420/533, loss: 0.11245842278003693 2023-01-21 10:21:13.433791: step: 424/533, loss: 0.09764566272497177 2023-01-21 10:21:14.642821: step: 428/533, loss: 0.07032451778650284 2023-01-21 10:21:15.846892: step: 432/533, loss: 0.060978200286626816 2023-01-21 10:21:17.007994: step: 436/533, loss: 0.04816289246082306 2023-01-21 10:21:18.163034: step: 440/533, loss: 0.18946895003318787 2023-01-21 10:21:19.303739: step: 444/533, loss: 0.06754856556653976 2023-01-21 10:21:20.516292: step: 448/533, loss: 0.007838916964828968 2023-01-21 10:21:21.706272: step: 452/533, loss: 0.0027926205657422543 2023-01-21 10:21:22.855818: step: 456/533, loss: 0.038857340812683105 2023-01-21 10:21:23.998135: step: 460/533, loss: 0.0311871524900198 2023-01-21 10:21:25.117396: step: 464/533, loss: 0.12137894332408905 2023-01-21 10:21:26.300768: step: 468/533, loss: 0.003990650177001953 2023-01-21 10:21:27.521408: step: 472/533, loss: 0.0762060135602951 2023-01-21 10:21:28.688506: step: 476/533, loss: 0.039781905710697174 2023-01-21 10:21:29.862599: step: 480/533, loss: 0.09409356117248535 2023-01-21 10:21:30.982058: step: 484/533, loss: 0.21389159560203552 2023-01-21 10:21:32.145775: step: 488/533, loss: 0.041011907160282135 2023-01-21 10:21:33.305276: step: 492/533, loss: 0.12473776936531067 2023-01-21 10:21:34.509689: step: 496/533, loss: 0.16382838785648346 2023-01-21 10:21:35.708956: step: 500/533, loss: 0.04423399269580841 2023-01-21 10:21:36.866621: step: 504/533, loss: 0.08094783127307892 2023-01-21 10:21:38.074834: step: 508/533, loss: 0.05893287807703018 2023-01-21 10:21:39.297534: step: 512/533, loss: 0.2006063461303711 2023-01-21 10:21:40.517378: step: 516/533, loss: 0.0035518647637218237 2023-01-21 10:21:41.705428: step: 520/533, loss: 0.012249970808625221 2023-01-21 10:21:42.910247: step: 524/533, loss: 0.03733997419476509 2023-01-21 10:21:44.160408: step: 528/533, loss: 0.19792062044143677 2023-01-21 10:21:45.365089: step: 532/533, loss: 0.022332191467285156 2023-01-21 10:21:46.597545: step: 536/533, loss: 0.08081942051649094 2023-01-21 10:21:47.773631: step: 540/533, loss: 0.1203208938241005 2023-01-21 10:21:48.953458: step: 544/533, loss: 0.01227712631225586 2023-01-21 10:21:50.171372: step: 548/533, loss: 0.05710621178150177 2023-01-21 10:21:51.381797: step: 552/533, loss: 0.09651708602905273 2023-01-21 10:21:52.583975: step: 556/533, loss: 0.059904247522354126 2023-01-21 10:21:53.743931: step: 560/533, loss: 0.026563310995697975 2023-01-21 10:21:54.927851: step: 564/533, loss: 0.02484150044620037 2023-01-21 10:21:56.123128: step: 568/533, loss: 0.12736962735652924 2023-01-21 10:21:57.315005: step: 572/533, loss: 0.017676640301942825 2023-01-21 10:21:58.517800: step: 576/533, loss: 0.028828907757997513 2023-01-21 10:21:59.744449: step: 580/533, loss: 0.06970787048339844 2023-01-21 10:22:00.943008: step: 584/533, loss: 0.017116164788603783 2023-01-21 10:22:02.151385: step: 588/533, loss: 0.06243934482336044 2023-01-21 10:22:03.304257: step: 592/533, loss: 0.017549894750118256 2023-01-21 10:22:04.504974: step: 596/533, loss: 0.0860927551984787 2023-01-21 10:22:05.689511: step: 600/533, loss: 0.030481241643428802 2023-01-21 10:22:06.846231: step: 604/533, loss: 0.031983088701963425 2023-01-21 10:22:08.037512: step: 608/533, loss: 0.05323949083685875 2023-01-21 10:22:09.254698: step: 612/533, loss: 0.05301237106323242 2023-01-21 10:22:10.414918: step: 616/533, loss: 0.08764810860157013 2023-01-21 10:22:11.598250: step: 620/533, loss: 0.008958530612289906 2023-01-21 10:22:12.791050: step: 624/533, loss: 0.08596744388341904 2023-01-21 10:22:13.961480: step: 628/533, loss: 0.017841290682554245 2023-01-21 10:22:15.123721: step: 632/533, loss: 0.07892937958240509 2023-01-21 10:22:16.292935: step: 636/533, loss: 0.09280043095350266 2023-01-21 10:22:17.456865: step: 640/533, loss: 0.002223205752670765 2023-01-21 10:22:18.696559: step: 644/533, loss: 0.028485393151640892 2023-01-21 10:22:19.901396: step: 648/533, loss: 0.0022100924979895353 2023-01-21 10:22:21.079291: step: 652/533, loss: 0.005769098177552223 2023-01-21 10:22:22.272919: step: 656/533, loss: 0.04408903419971466 2023-01-21 10:22:23.452325: step: 660/533, loss: 0.06102009117603302 2023-01-21 10:22:24.660049: step: 664/533, loss: 0.10442186892032623 2023-01-21 10:22:25.809689: step: 668/533, loss: 0.17542466521263123 2023-01-21 10:22:26.969243: step: 672/533, loss: 0.04020867496728897 2023-01-21 10:22:28.139936: step: 676/533, loss: 0.14757074415683746 2023-01-21 10:22:29.312842: step: 680/533, loss: 0.017836570739746094 2023-01-21 10:22:30.496108: step: 684/533, loss: 0.01849689520895481 2023-01-21 10:22:31.667151: step: 688/533, loss: 0.0216786228120327 2023-01-21 10:22:32.915457: step: 692/533, loss: 0.05392627790570259 2023-01-21 10:22:34.133269: step: 696/533, loss: 0.21014156937599182 2023-01-21 10:22:35.294421: step: 700/533, loss: 0.021427344530820847 2023-01-21 10:22:36.464287: step: 704/533, loss: 0.04213247075676918 2023-01-21 10:22:37.620247: step: 708/533, loss: 0.08668547123670578 2023-01-21 10:22:38.839289: step: 712/533, loss: 0.3106168806552887 2023-01-21 10:22:40.042319: step: 716/533, loss: 0.017412377521395683 2023-01-21 10:22:41.277678: step: 720/533, loss: 0.059807490557432175 2023-01-21 10:22:42.442955: step: 724/533, loss: 0.028153324499726295 2023-01-21 10:22:43.611499: step: 728/533, loss: 0.003921255469322205 2023-01-21 10:22:44.801501: step: 732/533, loss: 0.02863330766558647 2023-01-21 10:22:46.006650: step: 736/533, loss: 0.005204916000366211 2023-01-21 10:22:47.182455: step: 740/533, loss: 0.029224969446659088 2023-01-21 10:22:48.350405: step: 744/533, loss: 0.02072594314813614 2023-01-21 10:22:49.543025: step: 748/533, loss: 0.11536255478858948 2023-01-21 10:22:50.716283: step: 752/533, loss: 0.1391172856092453 2023-01-21 10:22:51.913034: step: 756/533, loss: 0.0448765754699707 2023-01-21 10:22:53.052344: step: 760/533, loss: 0.026940584182739258 2023-01-21 10:22:54.212331: step: 764/533, loss: 0.01531610544770956 2023-01-21 10:22:55.398968: step: 768/533, loss: 0.0840090811252594 2023-01-21 10:22:56.563097: step: 772/533, loss: 0.21093778312206268 2023-01-21 10:22:57.719284: step: 776/533, loss: 0.15608978271484375 2023-01-21 10:22:58.892222: step: 780/533, loss: 0.007524467073380947 2023-01-21 10:23:00.076235: step: 784/533, loss: 0.2596149444580078 2023-01-21 10:23:01.267434: step: 788/533, loss: 0.03057042509317398 2023-01-21 10:23:02.456508: step: 792/533, loss: 0.034690190106630325 2023-01-21 10:23:03.655185: step: 796/533, loss: 0.09901542961597443 2023-01-21 10:23:04.874035: step: 800/533, loss: 0.013355541042983532 2023-01-21 10:23:06.048672: step: 804/533, loss: 0.013167286291718483 2023-01-21 10:23:07.239300: step: 808/533, loss: 0.0480194091796875 2023-01-21 10:23:08.390572: step: 812/533, loss: 0.02352001704275608 2023-01-21 10:23:09.567514: step: 816/533, loss: 0.037485502660274506 2023-01-21 10:23:10.829851: step: 820/533, loss: 0.06921150535345078 2023-01-21 10:23:12.024005: step: 824/533, loss: 0.40658819675445557 2023-01-21 10:23:13.179908: step: 828/533, loss: 0.039452601224184036 2023-01-21 10:23:14.345664: step: 832/533, loss: 0.017307210713624954 2023-01-21 10:23:15.512027: step: 836/533, loss: 0.1110198050737381 2023-01-21 10:23:16.768433: step: 840/533, loss: 0.04405651241540909 2023-01-21 10:23:17.953805: step: 844/533, loss: 0.28958702087402344 2023-01-21 10:23:19.122700: step: 848/533, loss: 0.0905223861336708 2023-01-21 10:23:20.310655: step: 852/533, loss: 0.03428377956151962 2023-01-21 10:23:21.526090: step: 856/533, loss: 0.2751867175102234 2023-01-21 10:23:22.753649: step: 860/533, loss: 0.056435585021972656 2023-01-21 10:23:23.932636: step: 864/533, loss: 0.004986953921616077 2023-01-21 10:23:25.108348: step: 868/533, loss: 0.05244483798742294 2023-01-21 10:23:26.323622: step: 872/533, loss: 0.018985938280820847 2023-01-21 10:23:27.529792: step: 876/533, loss: 0.1250377595424652 2023-01-21 10:23:28.713299: step: 880/533, loss: 0.07025227695703506 2023-01-21 10:23:29.909834: step: 884/533, loss: 0.0242125503718853 2023-01-21 10:23:31.113693: step: 888/533, loss: 0.05815563350915909 2023-01-21 10:23:32.264938: step: 892/533, loss: 0.04133930429816246 2023-01-21 10:23:33.535298: step: 896/533, loss: 0.0975547730922699 2023-01-21 10:23:34.701275: step: 900/533, loss: 0.08729038387537003 2023-01-21 10:23:35.886336: step: 904/533, loss: 0.09723357856273651 2023-01-21 10:23:37.020856: step: 908/533, loss: 0.08272366225719452 2023-01-21 10:23:38.209331: step: 912/533, loss: 0.27123039960861206 2023-01-21 10:23:39.374147: step: 916/533, loss: 0.07418441772460938 2023-01-21 10:23:40.561447: step: 920/533, loss: 0.02607116848230362 2023-01-21 10:23:41.755689: step: 924/533, loss: 0.08553676307201385 2023-01-21 10:23:42.944877: step: 928/533, loss: 0.08645935356616974 2023-01-21 10:23:44.112329: step: 932/533, loss: 0.23334969580173492 2023-01-21 10:23:45.309010: step: 936/533, loss: 0.07576780766248703 2023-01-21 10:23:46.472589: step: 940/533, loss: 0.12365283817052841 2023-01-21 10:23:47.633196: step: 944/533, loss: 0.0713711753487587 2023-01-21 10:23:48.795009: step: 948/533, loss: 0.05747837945818901 2023-01-21 10:23:49.990807: step: 952/533, loss: 0.009887218475341797 2023-01-21 10:23:51.181674: step: 956/533, loss: 0.03767995908856392 2023-01-21 10:23:52.364436: step: 960/533, loss: 0.019109057262539864 2023-01-21 10:23:53.567619: step: 964/533, loss: 0.08429031074047089 2023-01-21 10:23:54.722754: step: 968/533, loss: 0.6642215847969055 2023-01-21 10:23:55.868864: step: 972/533, loss: 0.05595884472131729 2023-01-21 10:23:57.067709: step: 976/533, loss: 0.11274567246437073 2023-01-21 10:23:58.260719: step: 980/533, loss: 0.027388382703065872 2023-01-21 10:23:59.447638: step: 984/533, loss: 0.012281752191483974 2023-01-21 10:24:00.662659: step: 988/533, loss: 0.06368140876293182 2023-01-21 10:24:01.796600: step: 992/533, loss: 0.07388248294591904 2023-01-21 10:24:03.014521: step: 996/533, loss: 0.016553591936826706 2023-01-21 10:24:04.176911: step: 1000/533, loss: 0.10744866728782654 2023-01-21 10:24:05.347191: step: 1004/533, loss: 0.08306851238012314 2023-01-21 10:24:06.561522: step: 1008/533, loss: 0.032976724207401276 2023-01-21 10:24:07.737136: step: 1012/533, loss: 0.05050821602344513 2023-01-21 10:24:08.962278: step: 1016/533, loss: 0.049886129796504974 2023-01-21 10:24:10.236908: step: 1020/533, loss: 0.0882379561662674 2023-01-21 10:24:11.384562: step: 1024/533, loss: 0.059326935559511185 2023-01-21 10:24:12.536715: step: 1028/533, loss: 0.045450758188962936 2023-01-21 10:24:13.753764: step: 1032/533, loss: 0.026250028982758522 2023-01-21 10:24:14.977953: step: 1036/533, loss: 0.6288866996765137 2023-01-21 10:24:16.165869: step: 1040/533, loss: 0.0344233512878418 2023-01-21 10:24:17.353391: step: 1044/533, loss: 0.7731708288192749 2023-01-21 10:24:18.553425: step: 1048/533, loss: 0.04690227657556534 2023-01-21 10:24:19.721319: step: 1052/533, loss: 0.2962002754211426 2023-01-21 10:24:20.872712: step: 1056/533, loss: 0.008244156837463379 2023-01-21 10:24:22.031441: step: 1060/533, loss: 0.11076736450195312 2023-01-21 10:24:23.209621: step: 1064/533, loss: 0.004121589940041304 2023-01-21 10:24:24.405797: step: 1068/533, loss: 0.0828530341386795 2023-01-21 10:24:25.610593: step: 1072/533, loss: 0.11859636753797531 2023-01-21 10:24:26.788013: step: 1076/533, loss: 0.06641598045825958 2023-01-21 10:24:27.962153: step: 1080/533, loss: 0.09994659572839737 2023-01-21 10:24:29.146957: step: 1084/533, loss: 0.21724766492843628 2023-01-21 10:24:30.320066: step: 1088/533, loss: 0.13935308158397675 2023-01-21 10:24:31.487387: step: 1092/533, loss: 0.03557634353637695 2023-01-21 10:24:32.699309: step: 1096/533, loss: 0.1536761373281479 2023-01-21 10:24:33.902048: step: 1100/533, loss: 0.1869945526123047 2023-01-21 10:24:35.037674: step: 1104/533, loss: 0.22792643308639526 2023-01-21 10:24:36.220891: step: 1108/533, loss: 0.9061795473098755 2023-01-21 10:24:37.433568: step: 1112/533, loss: 0.01720714569091797 2023-01-21 10:24:38.615286: step: 1116/533, loss: 0.1820284128189087 2023-01-21 10:24:39.770997: step: 1120/533, loss: 0.054145097732543945 2023-01-21 10:24:40.953689: step: 1124/533, loss: 0.6930950880050659 2023-01-21 10:24:42.120294: step: 1128/533, loss: 0.08696766197681427 2023-01-21 10:24:43.277298: step: 1132/533, loss: 0.09825573861598969 2023-01-21 10:24:44.417354: step: 1136/533, loss: 0.09787445515394211 2023-01-21 10:24:45.588587: step: 1140/533, loss: 0.038423679769039154 2023-01-21 10:24:46.761943: step: 1144/533, loss: 0.11393798887729645 2023-01-21 10:24:47.962694: step: 1148/533, loss: 0.11200561374425888 2023-01-21 10:24:49.138375: step: 1152/533, loss: 0.012944317422807217 2023-01-21 10:24:50.301461: step: 1156/533, loss: 0.03660225868225098 2023-01-21 10:24:51.452947: step: 1160/533, loss: 0.10471335053443909 2023-01-21 10:24:52.597327: step: 1164/533, loss: 0.07390861958265305 2023-01-21 10:24:53.816951: step: 1168/533, loss: 0.1529351770877838 2023-01-21 10:24:54.962753: step: 1172/533, loss: 0.004880237393081188 2023-01-21 10:24:56.156519: step: 1176/533, loss: 0.040322497487068176 2023-01-21 10:24:57.364379: step: 1180/533, loss: 0.11745148152112961 2023-01-21 10:24:58.535287: step: 1184/533, loss: 0.08561897277832031 2023-01-21 10:24:59.737448: step: 1188/533, loss: 0.034182358533144 2023-01-21 10:25:00.938883: step: 1192/533, loss: 0.4548143446445465 2023-01-21 10:25:02.127322: step: 1196/533, loss: 0.17241840064525604 2023-01-21 10:25:03.348759: step: 1200/533, loss: 0.08510598540306091 2023-01-21 10:25:04.524757: step: 1204/533, loss: 0.15125522017478943 2023-01-21 10:25:05.670054: step: 1208/533, loss: 0.0975104346871376 2023-01-21 10:25:06.855079: step: 1212/533, loss: 0.3133293092250824 2023-01-21 10:25:08.041508: step: 1216/533, loss: 0.054933834820985794 2023-01-21 10:25:09.210394: step: 1220/533, loss: 0.046628620475530624 2023-01-21 10:25:10.378894: step: 1224/533, loss: 0.024113750085234642 2023-01-21 10:25:11.550825: step: 1228/533, loss: 0.012184619903564453 2023-01-21 10:25:12.755856: step: 1232/533, loss: 0.05818767845630646 2023-01-21 10:25:13.950905: step: 1236/533, loss: 0.07443217933177948 2023-01-21 10:25:15.152708: step: 1240/533, loss: 0.2787153720855713 2023-01-21 10:25:16.341350: step: 1244/533, loss: 0.010120391845703125 2023-01-21 10:25:17.526731: step: 1248/533, loss: 0.07431783527135849 2023-01-21 10:25:18.713570: step: 1252/533, loss: 0.046751346439123154 2023-01-21 10:25:19.899738: step: 1256/533, loss: 0.11606941372156143 2023-01-21 10:25:21.075898: step: 1260/533, loss: 0.08818984031677246 2023-01-21 10:25:22.244081: step: 1264/533, loss: 0.1536249816417694 2023-01-21 10:25:23.399228: step: 1268/533, loss: 0.07218757271766663 2023-01-21 10:25:24.553366: step: 1272/533, loss: 0.10483994334936142 2023-01-21 10:25:25.693753: step: 1276/533, loss: 0.05454890802502632 2023-01-21 10:25:26.858372: step: 1280/533, loss: 0.24248504638671875 2023-01-21 10:25:28.070297: step: 1284/533, loss: 0.0508207343518734 2023-01-21 10:25:29.254646: step: 1288/533, loss: 0.12971706688404083 2023-01-21 10:25:30.418365: step: 1292/533, loss: 0.07170099765062332 2023-01-21 10:25:31.584697: step: 1296/533, loss: 0.02202749252319336 2023-01-21 10:25:32.760352: step: 1300/533, loss: 0.08108548820018768 2023-01-21 10:25:33.970452: step: 1304/533, loss: 0.011467933654785156 2023-01-21 10:25:35.157030: step: 1308/533, loss: 0.07238147407770157 2023-01-21 10:25:36.330908: step: 1312/533, loss: 0.08613242954015732 2023-01-21 10:25:37.503161: step: 1316/533, loss: 0.2801237106323242 2023-01-21 10:25:38.685315: step: 1320/533, loss: 0.013946629129350185 2023-01-21 10:25:39.901847: step: 1324/533, loss: 0.08583274483680725 2023-01-21 10:25:41.074896: step: 1328/533, loss: 0.06180715560913086 2023-01-21 10:25:42.280032: step: 1332/533, loss: 0.10338515788316727 2023-01-21 10:25:43.487717: step: 1336/533, loss: 1.0009410381317139 2023-01-21 10:25:44.675482: step: 1340/533, loss: 0.17573462426662445 2023-01-21 10:25:45.848264: step: 1344/533, loss: 0.05526895448565483 2023-01-21 10:25:47.007823: step: 1348/533, loss: 0.0029705048073083162 2023-01-21 10:25:48.206740: step: 1352/533, loss: 0.18709523975849152 2023-01-21 10:25:49.374228: step: 1356/533, loss: 0.5884739756584167 2023-01-21 10:25:50.565120: step: 1360/533, loss: 0.036551713943481445 2023-01-21 10:25:51.727354: step: 1364/533, loss: 0.036616209894418716 2023-01-21 10:25:52.899572: step: 1368/533, loss: 0.38821572065353394 2023-01-21 10:25:54.056892: step: 1372/533, loss: 0.062314461916685104 2023-01-21 10:25:55.293751: step: 1376/533, loss: 0.24531981348991394 2023-01-21 10:25:56.458399: step: 1380/533, loss: 0.026241686195135117 2023-01-21 10:25:57.647890: step: 1384/533, loss: 0.027741573750972748 2023-01-21 10:25:58.802661: step: 1388/533, loss: 0.03669700771570206 2023-01-21 10:25:59.982841: step: 1392/533, loss: 0.07896633446216583 2023-01-21 10:26:01.153944: step: 1396/533, loss: 0.027110911905765533 2023-01-21 10:26:02.335697: step: 1400/533, loss: 0.10955075919628143 2023-01-21 10:26:03.525187: step: 1404/533, loss: 0.5084964036941528 2023-01-21 10:26:04.725452: step: 1408/533, loss: 0.06081056594848633 2023-01-21 10:26:05.921501: step: 1412/533, loss: 0.043141938745975494 2023-01-21 10:26:07.071784: step: 1416/533, loss: 0.16898289322853088 2023-01-21 10:26:08.337899: step: 1420/533, loss: 0.12807735800743103 2023-01-21 10:26:09.513291: step: 1424/533, loss: 0.028343772515654564 2023-01-21 10:26:10.677954: step: 1428/533, loss: 0.010363107547163963 2023-01-21 10:26:11.838430: step: 1432/533, loss: 0.06041097640991211 2023-01-21 10:26:13.040788: step: 1436/533, loss: 0.04304962605237961 2023-01-21 10:26:14.200817: step: 1440/533, loss: 0.005700159352272749 2023-01-21 10:26:15.375495: step: 1444/533, loss: 0.03150491788983345 2023-01-21 10:26:16.672079: step: 1448/533, loss: 0.04747028648853302 2023-01-21 10:26:17.835178: step: 1452/533, loss: 0.05173196643590927 2023-01-21 10:26:19.041784: step: 1456/533, loss: 0.05852065235376358 2023-01-21 10:26:20.251738: step: 1460/533, loss: 0.12316127121448517 2023-01-21 10:26:21.415948: step: 1464/533, loss: 0.12001600116491318 2023-01-21 10:26:22.583422: step: 1468/533, loss: 0.05540933459997177 2023-01-21 10:26:23.765801: step: 1472/533, loss: 0.007329070474952459 2023-01-21 10:26:24.950510: step: 1476/533, loss: 0.051676083356142044 2023-01-21 10:26:26.127147: step: 1480/533, loss: 0.026768827810883522 2023-01-21 10:26:27.365823: step: 1484/533, loss: 0.10381804406642914 2023-01-21 10:26:28.543697: step: 1488/533, loss: 0.05727798864245415 2023-01-21 10:26:29.689537: step: 1492/533, loss: 0.5386823415756226 2023-01-21 10:26:30.852342: step: 1496/533, loss: 0.027991319075226784 2023-01-21 10:26:32.000074: step: 1500/533, loss: 0.003662109375 2023-01-21 10:26:33.145189: step: 1504/533, loss: 0.49916714429855347 2023-01-21 10:26:34.339503: step: 1508/533, loss: 0.08819127082824707 2023-01-21 10:26:35.514156: step: 1512/533, loss: 0.12923279404640198 2023-01-21 10:26:36.733473: step: 1516/533, loss: 0.010399436578154564 2023-01-21 10:26:37.930391: step: 1520/533, loss: 0.23699027299880981 2023-01-21 10:26:39.091184: step: 1524/533, loss: 0.008065986447036266 2023-01-21 10:26:40.276875: step: 1528/533, loss: 0.02567291259765625 2023-01-21 10:26:41.491128: step: 1532/533, loss: 0.04861469566822052 2023-01-21 10:26:42.663777: step: 1536/533, loss: 0.044959452003240585 2023-01-21 10:26:43.837689: step: 1540/533, loss: 0.025795841589570045 2023-01-21 10:26:45.071285: step: 1544/533, loss: 0.373870313167572 2023-01-21 10:26:46.238400: step: 1548/533, loss: 0.025064660236239433 2023-01-21 10:26:47.407323: step: 1552/533, loss: 0.0037708759773522615 2023-01-21 10:26:48.622295: step: 1556/533, loss: 0.02398090437054634 2023-01-21 10:26:49.768754: step: 1560/533, loss: 0.04906916990876198 2023-01-21 10:26:50.929065: step: 1564/533, loss: 0.1152244359254837 2023-01-21 10:26:52.055178: step: 1568/533, loss: 0.12641267478466034 2023-01-21 10:26:53.195707: step: 1572/533, loss: 0.024326419457793236 2023-01-21 10:26:54.404037: step: 1576/533, loss: 0.02207660675048828 2023-01-21 10:26:55.617690: step: 1580/533, loss: 0.0548248291015625 2023-01-21 10:26:56.820338: step: 1584/533, loss: 0.2030935287475586 2023-01-21 10:26:57.972923: step: 1588/533, loss: 0.03352651745080948 2023-01-21 10:26:59.142142: step: 1592/533, loss: 0.03094806708395481 2023-01-21 10:27:00.304960: step: 1596/533, loss: 0.06598720699548721 2023-01-21 10:27:01.493705: step: 1600/533, loss: 0.007987022399902344 2023-01-21 10:27:02.679051: step: 1604/533, loss: 0.015272426418960094 2023-01-21 10:27:03.872017: step: 1608/533, loss: 0.2675851881504059 2023-01-21 10:27:05.070593: step: 1612/533, loss: 0.10724973678588867 2023-01-21 10:27:06.282370: step: 1616/533, loss: 0.05555405840277672 2023-01-21 10:27:07.408693: step: 1620/533, loss: 0.09273891896009445 2023-01-21 10:27:08.579124: step: 1624/533, loss: 0.016599655151367188 2023-01-21 10:27:09.758597: step: 1628/533, loss: 0.07123704254627228 2023-01-21 10:27:10.935966: step: 1632/533, loss: 0.059188082814216614 2023-01-21 10:27:12.110325: step: 1636/533, loss: 0.008829879574477673 2023-01-21 10:27:13.279782: step: 1640/533, loss: 0.03722744062542915 2023-01-21 10:27:14.492791: step: 1644/533, loss: 0.16563081741333008 2023-01-21 10:27:15.673254: step: 1648/533, loss: 0.12208674103021622 2023-01-21 10:27:16.876216: step: 1652/533, loss: 0.06706981360912323 2023-01-21 10:27:18.045806: step: 1656/533, loss: 0.0668870061635971 2023-01-21 10:27:19.210413: step: 1660/533, loss: 0.015508986078202724 2023-01-21 10:27:20.388022: step: 1664/533, loss: 0.05443458631634712 2023-01-21 10:27:21.589835: step: 1668/533, loss: 0.04190950468182564 2023-01-21 10:27:22.749535: step: 1672/533, loss: 0.05709293484687805 2023-01-21 10:27:23.962686: step: 1676/533, loss: 0.06490955501794815 2023-01-21 10:27:25.121935: step: 1680/533, loss: 0.0038893460296094418 2023-01-21 10:27:26.274675: step: 1684/533, loss: 0.009616660885512829 2023-01-21 10:27:27.448260: step: 1688/533, loss: 0.07100296020507812 2023-01-21 10:27:28.605980: step: 1692/533, loss: 0.0056012156419456005 2023-01-21 10:27:29.805682: step: 1696/533, loss: 0.05840704217553139 2023-01-21 10:27:30.990750: step: 1700/533, loss: 0.12703028321266174 2023-01-21 10:27:32.175522: step: 1704/533, loss: 0.057176969945430756 2023-01-21 10:27:33.335157: step: 1708/533, loss: 0.23405419290065765 2023-01-21 10:27:34.526547: step: 1712/533, loss: 0.3543122410774231 2023-01-21 10:27:35.704787: step: 1716/533, loss: 0.021964453160762787 2023-01-21 10:27:36.888499: step: 1720/533, loss: 0.02976525016129017 2023-01-21 10:27:38.064121: step: 1724/533, loss: 0.08387728035449982 2023-01-21 10:27:39.228722: step: 1728/533, loss: 0.018842125311493874 2023-01-21 10:27:40.429031: step: 1732/533, loss: 0.07530169934034348 2023-01-21 10:27:41.598759: step: 1736/533, loss: 0.06779966503381729 2023-01-21 10:27:42.809482: step: 1740/533, loss: 0.026753904297947884 2023-01-21 10:27:43.977527: step: 1744/533, loss: 0.04340362548828125 2023-01-21 10:27:45.157986: step: 1748/533, loss: 0.12866048514842987 2023-01-21 10:27:46.324078: step: 1752/533, loss: 0.02243213728070259 2023-01-21 10:27:47.556390: step: 1756/533, loss: 0.15219421684741974 2023-01-21 10:27:48.741830: step: 1760/533, loss: 0.1093904972076416 2023-01-21 10:27:49.907229: step: 1764/533, loss: 0.0665835440158844 2023-01-21 10:27:51.146504: step: 1768/533, loss: 0.10058832168579102 2023-01-21 10:27:52.325029: step: 1772/533, loss: 0.017956828698515892 2023-01-21 10:27:53.499971: step: 1776/533, loss: 0.03499488905072212 2023-01-21 10:27:54.677515: step: 1780/533, loss: 0.014957617968320847 2023-01-21 10:27:55.834787: step: 1784/533, loss: 0.08211478590965271 2023-01-21 10:27:56.975958: step: 1788/533, loss: 0.024079417809844017 2023-01-21 10:27:58.153218: step: 1792/533, loss: 0.05085492134094238 2023-01-21 10:27:59.315525: step: 1796/533, loss: 0.02582411840558052 2023-01-21 10:28:00.499531: step: 1800/533, loss: 0.026110172271728516 2023-01-21 10:28:01.736478: step: 1804/533, loss: 0.07222280651330948 2023-01-21 10:28:02.915710: step: 1808/533, loss: 0.7108290791511536 2023-01-21 10:28:04.098729: step: 1812/533, loss: 0.005486679263412952 2023-01-21 10:28:05.271443: step: 1816/533, loss: 0.05412931740283966 2023-01-21 10:28:06.461683: step: 1820/533, loss: 0.057809069752693176 2023-01-21 10:28:07.654663: step: 1824/533, loss: 0.07038374990224838 2023-01-21 10:28:08.810608: step: 1828/533, loss: 0.05788679048418999 2023-01-21 10:28:10.008113: step: 1832/533, loss: 0.09710827469825745 2023-01-21 10:28:11.165840: step: 1836/533, loss: 0.0164810661226511 2023-01-21 10:28:12.363756: step: 1840/533, loss: 0.01407403964549303 2023-01-21 10:28:13.629348: step: 1844/533, loss: 0.20030422508716583 2023-01-21 10:28:14.813927: step: 1848/533, loss: 0.03567218780517578 2023-01-21 10:28:15.996629: step: 1852/533, loss: 0.11867332458496094 2023-01-21 10:28:17.144057: step: 1856/533, loss: 0.11448794603347778 2023-01-21 10:28:18.347812: step: 1860/533, loss: 0.08703909069299698 2023-01-21 10:28:19.512982: step: 1864/533, loss: 0.17075638473033905 2023-01-21 10:28:20.678043: step: 1868/533, loss: 0.009350347332656384 2023-01-21 10:28:21.859455: step: 1872/533, loss: 0.02880687639117241 2023-01-21 10:28:23.055235: step: 1876/533, loss: 0.06279440224170685 2023-01-21 10:28:24.215464: step: 1880/533, loss: 0.0051780701614916325 2023-01-21 10:28:25.411953: step: 1884/533, loss: 0.06368546187877655 2023-01-21 10:28:26.614285: step: 1888/533, loss: 0.01476383302360773 2023-01-21 10:28:27.783653: step: 1892/533, loss: 0.12096557766199112 2023-01-21 10:28:28.944979: step: 1896/533, loss: 0.09713239222764969 2023-01-21 10:28:30.121286: step: 1900/533, loss: 0.03229885175824165 2023-01-21 10:28:31.311316: step: 1904/533, loss: 0.1363920271396637 2023-01-21 10:28:32.497158: step: 1908/533, loss: 0.040161460638046265 2023-01-21 10:28:33.652169: step: 1912/533, loss: 0.03166034445166588 2023-01-21 10:28:34.844165: step: 1916/533, loss: 0.49534720182418823 2023-01-21 10:28:36.043304: step: 1920/533, loss: 0.05235690996050835 2023-01-21 10:28:37.307313: step: 1924/533, loss: 0.016606617718935013 2023-01-21 10:28:38.463196: step: 1928/533, loss: 0.006987619213759899 2023-01-21 10:28:39.661436: step: 1932/533, loss: 0.0011625289916992188 2023-01-21 10:28:40.825318: step: 1936/533, loss: 0.011188840493559837 2023-01-21 10:28:41.992749: step: 1940/533, loss: 0.17567414045333862 2023-01-21 10:28:43.235620: step: 1944/533, loss: 0.09800954163074493 2023-01-21 10:28:44.435358: step: 1948/533, loss: 0.05898590385913849 2023-01-21 10:28:45.616103: step: 1952/533, loss: 0.01071014441549778 2023-01-21 10:28:46.765588: step: 1956/533, loss: 0.36879587173461914 2023-01-21 10:28:47.937897: step: 1960/533, loss: 0.5534471273422241 2023-01-21 10:28:49.091670: step: 1964/533, loss: 0.016251325607299805 2023-01-21 10:28:50.241668: step: 1968/533, loss: 0.04796028137207031 2023-01-21 10:28:51.407037: step: 1972/533, loss: 0.16088706254959106 2023-01-21 10:28:52.627143: step: 1976/533, loss: 0.153499037027359 2023-01-21 10:28:53.818512: step: 1980/533, loss: 0.019681455567479134 2023-01-21 10:28:55.042392: step: 1984/533, loss: 0.11831459403038025 2023-01-21 10:28:56.214573: step: 1988/533, loss: 0.006252622697502375 2023-01-21 10:28:57.360686: step: 1992/533, loss: 0.10022459179162979 2023-01-21 10:28:58.535028: step: 1996/533, loss: 0.31878718733787537 2023-01-21 10:28:59.682455: step: 2000/533, loss: 9.713172767078504e-05 2023-01-21 10:29:00.917020: step: 2004/533, loss: 0.026334000751376152 2023-01-21 10:29:02.110458: step: 2008/533, loss: 0.04110260307788849 2023-01-21 10:29:03.319273: step: 2012/533, loss: 0.0028736114036291838 2023-01-21 10:29:04.495865: step: 2016/533, loss: 0.09804189205169678 2023-01-21 10:29:05.658740: step: 2020/533, loss: 0.05880336835980415 2023-01-21 10:29:06.854227: step: 2024/533, loss: 0.14047737419605255 2023-01-21 10:29:08.036891: step: 2028/533, loss: 0.01078042946755886 2023-01-21 10:29:09.220962: step: 2032/533, loss: 0.11109104007482529 2023-01-21 10:29:10.455356: step: 2036/533, loss: 0.013492370024323463 2023-01-21 10:29:11.607638: step: 2040/533, loss: 0.05885725095868111 2023-01-21 10:29:12.757318: step: 2044/533, loss: 0.029235459864139557 2023-01-21 10:29:13.957250: step: 2048/533, loss: 0.13680404424667358 2023-01-21 10:29:15.127948: step: 2052/533, loss: 0.28125572204589844 2023-01-21 10:29:16.294960: step: 2056/533, loss: 0.24199867248535156 2023-01-21 10:29:17.465908: step: 2060/533, loss: 0.04656543582677841 2023-01-21 10:29:18.627522: step: 2064/533, loss: 0.022696923464536667 2023-01-21 10:29:19.826458: step: 2068/533, loss: 0.0015400409465655684 2023-01-21 10:29:21.055530: step: 2072/533, loss: 0.029370786622166634 2023-01-21 10:29:22.225015: step: 2076/533, loss: 0.07193336635828018 2023-01-21 10:29:23.394116: step: 2080/533, loss: 0.025753021240234375 2023-01-21 10:29:24.628414: step: 2084/533, loss: 0.020557023584842682 2023-01-21 10:29:25.855446: step: 2088/533, loss: 0.033214569091796875 2023-01-21 10:29:27.076202: step: 2092/533, loss: 0.13563615083694458 2023-01-21 10:29:28.244346: step: 2096/533, loss: 0.017001725733280182 2023-01-21 10:29:29.463938: step: 2100/533, loss: 0.1857103407382965 2023-01-21 10:29:30.653576: step: 2104/533, loss: 0.10610218346118927 2023-01-21 10:29:31.829720: step: 2108/533, loss: 0.04030952602624893 2023-01-21 10:29:33.018972: step: 2112/533, loss: 0.04674091190099716 2023-01-21 10:29:34.214315: step: 2116/533, loss: 0.1484365016222 2023-01-21 10:29:35.367070: step: 2120/533, loss: 0.07701349258422852 2023-01-21 10:29:36.536075: step: 2124/533, loss: 0.08461757749319077 2023-01-21 10:29:37.722116: step: 2128/533, loss: 0.26792973279953003 2023-01-21 10:29:38.871157: step: 2132/533, loss: 0.06239013746380806 ================================================== Loss: 0.092 -------------------- Dev: {'event': {'p': 0.5639810426540285, 'r': 0.7922769640479361, 'f1': 0.6589147286821706}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Test: {'event': {'p': 0.6056812710640347, 'r': 0.8043478260869565, 'f1': 0.6910189508376821}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Chinese: {'event': {'p': 0.5777777777777777, 'r': 0.9629629629629629, 'f1': 0.7222222222222221}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Korean: {'event': {'p': 0.6101694915254238, 'r': 0.5714285714285714, 'f1': 0.5901639344262296}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Russian: {'event': {'p': 0.37209302325581395, 'r': 0.4444444444444444, 'f1': 0.40506329113924044}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 10:30:21.873528: step: 4/533, loss: 0.11098746955394745 2023-01-21 10:30:23.033778: step: 8/533, loss: 0.11326369643211365 2023-01-21 10:30:24.239866: step: 12/533, loss: 0.015026998706161976 2023-01-21 10:30:25.444996: step: 16/533, loss: 0.04696798324584961 2023-01-21 10:30:26.595531: step: 20/533, loss: 0.006007767282426357 2023-01-21 10:30:27.736865: step: 24/533, loss: 0.036834146827459335 2023-01-21 10:30:28.945400: step: 28/533, loss: 0.05240399390459061 2023-01-21 10:30:30.106814: step: 32/533, loss: 0.005521345417946577 2023-01-21 10:30:31.314744: step: 36/533, loss: 0.08041973412036896 2023-01-21 10:30:32.543378: step: 40/533, loss: 0.010283757001161575 2023-01-21 10:30:33.735472: step: 44/533, loss: 0.02172107622027397 2023-01-21 10:30:34.901907: step: 48/533, loss: 0.02050170861184597 2023-01-21 10:30:36.087322: step: 52/533, loss: 0.2266945242881775 2023-01-21 10:30:37.251988: step: 56/533, loss: 0.005156993865966797 2023-01-21 10:30:38.477073: step: 60/533, loss: 0.049368664622306824 2023-01-21 10:30:39.675979: step: 64/533, loss: 0.06065859645605087 2023-01-21 10:30:40.852203: step: 68/533, loss: 0.02526865154504776 2023-01-21 10:30:42.089974: step: 72/533, loss: 0.02322525903582573 2023-01-21 10:30:43.318891: step: 76/533, loss: 0.03267580270767212 2023-01-21 10:30:44.476013: step: 80/533, loss: 0.04438326507806778 2023-01-21 10:30:45.668479: step: 84/533, loss: 0.009611892513930798 2023-01-21 10:30:46.846601: step: 88/533, loss: 0.02946758270263672 2023-01-21 10:30:48.019335: step: 92/533, loss: 0.019115973263978958 2023-01-21 10:30:49.172839: step: 96/533, loss: 0.22646526992321014 2023-01-21 10:30:50.399509: step: 100/533, loss: 0.061250876635313034 2023-01-21 10:30:51.608771: step: 104/533, loss: 0.00987234152853489 2023-01-21 10:30:52.802383: step: 108/533, loss: 0.02112426795065403 2023-01-21 10:30:53.978243: step: 112/533, loss: 0.0675802230834961 2023-01-21 10:30:55.151814: step: 116/533, loss: 0.017433859407901764 2023-01-21 10:30:56.288880: step: 120/533, loss: 0.06914959102869034 2023-01-21 10:30:57.460332: step: 124/533, loss: 0.04367981106042862 2023-01-21 10:30:58.618653: step: 128/533, loss: 0.006170368287712336 2023-01-21 10:30:59.816993: step: 132/533, loss: 0.02725696563720703 2023-01-21 10:31:01.003116: step: 136/533, loss: 0.11591873317956924 2023-01-21 10:31:02.206925: step: 140/533, loss: 0.0342339389026165 2023-01-21 10:31:03.406963: step: 144/533, loss: 0.03218948841094971 2023-01-21 10:31:04.560164: step: 148/533, loss: 0.020868491381406784 2023-01-21 10:31:05.739258: step: 152/533, loss: 0.07970891147851944 2023-01-21 10:31:06.933893: step: 156/533, loss: 0.07470874488353729 2023-01-21 10:31:08.103546: step: 160/533, loss: 0.06186676025390625 2023-01-21 10:31:09.287133: step: 164/533, loss: 0.026323892176151276 2023-01-21 10:31:10.441727: step: 168/533, loss: 0.011422348208725452 2023-01-21 10:31:11.652035: step: 172/533, loss: 0.023266172036528587 2023-01-21 10:31:12.841362: step: 176/533, loss: 0.006433963775634766 2023-01-21 10:31:14.003878: step: 180/533, loss: 0.05391855537891388 2023-01-21 10:31:15.163138: step: 184/533, loss: 0.004941749386489391 2023-01-21 10:31:16.339829: step: 188/533, loss: 0.04730267822742462 2023-01-21 10:31:17.540930: step: 192/533, loss: 0.06706647574901581 2023-01-21 10:31:18.717225: step: 196/533, loss: 0.010280609130859375 2023-01-21 10:31:19.909704: step: 200/533, loss: 0.08075451850891113 2023-01-21 10:31:21.105114: step: 204/533, loss: 0.06409773975610733 2023-01-21 10:31:22.306648: step: 208/533, loss: 0.024731159210205078 2023-01-21 10:31:23.485253: step: 212/533, loss: 0.09119682013988495 2023-01-21 10:31:24.647420: step: 216/533, loss: 0.032472554594278336 2023-01-21 10:31:25.859690: step: 220/533, loss: 0.16315937042236328 2023-01-21 10:31:27.000325: step: 224/533, loss: 0.046383023262023926 2023-01-21 10:31:28.214580: step: 228/533, loss: 0.020339488983154297 2023-01-21 10:31:29.371823: step: 232/533, loss: 0.043627165257930756 2023-01-21 10:31:30.568272: step: 236/533, loss: 0.011780072003602982 2023-01-21 10:31:31.707436: step: 240/533, loss: 0.03435640037059784 2023-01-21 10:31:32.886378: step: 244/533, loss: 0.12655138969421387 2023-01-21 10:31:34.070540: step: 248/533, loss: 0.03272189944982529 2023-01-21 10:31:35.226740: step: 252/533, loss: 0.012595271691679955 2023-01-21 10:31:36.416727: step: 256/533, loss: 0.15964969992637634 2023-01-21 10:31:37.595477: step: 260/533, loss: 0.03986222296953201 2023-01-21 10:31:38.760650: step: 264/533, loss: 0.07054682075977325 2023-01-21 10:31:39.896015: step: 268/533, loss: 0.40937739610671997 2023-01-21 10:31:41.072235: step: 272/533, loss: 0.15408268570899963 2023-01-21 10:31:42.258746: step: 276/533, loss: 0.023018455132842064 2023-01-21 10:31:43.423100: step: 280/533, loss: 0.06060600280761719 2023-01-21 10:31:44.568179: step: 284/533, loss: 0.03645515441894531 2023-01-21 10:31:45.738292: step: 288/533, loss: 0.06190476566553116 2023-01-21 10:31:46.912081: step: 292/533, loss: 0.016421889886260033 2023-01-21 10:31:48.057671: step: 296/533, loss: 0.08073997497558594 2023-01-21 10:31:49.235276: step: 300/533, loss: 0.04174823686480522 2023-01-21 10:31:50.390336: step: 304/533, loss: 0.03557872772216797 2023-01-21 10:31:51.534446: step: 308/533, loss: 0.03222084045410156 2023-01-21 10:31:52.740248: step: 312/533, loss: 0.05733766779303551 2023-01-21 10:31:53.923945: step: 316/533, loss: 0.00252113351598382 2023-01-21 10:31:55.153488: step: 320/533, loss: 0.06051836162805557 2023-01-21 10:31:56.348479: step: 324/533, loss: 0.16255348920822144 2023-01-21 10:31:57.533713: step: 328/533, loss: 0.0007349014631472528 2023-01-21 10:31:58.700634: step: 332/533, loss: 0.014552832581102848 2023-01-21 10:31:59.885352: step: 336/533, loss: 0.22178708016872406 2023-01-21 10:32:01.055346: step: 340/533, loss: 0.07792253792285919 2023-01-21 10:32:02.254294: step: 344/533, loss: 0.02086925506591797 2023-01-21 10:32:03.454103: step: 348/533, loss: 0.03763217851519585 2023-01-21 10:32:04.640983: step: 352/533, loss: 0.01214828435331583 2023-01-21 10:32:05.813438: step: 356/533, loss: 0.0013957977062091231 2023-01-21 10:32:07.021010: step: 360/533, loss: 0.018419837579131126 2023-01-21 10:32:08.223307: step: 364/533, loss: 0.025835610926151276 2023-01-21 10:32:09.380140: step: 368/533, loss: 0.17872829735279083 2023-01-21 10:32:10.562510: step: 372/533, loss: 0.004904317669570446 2023-01-21 10:32:11.771280: step: 376/533, loss: 0.08587322384119034 2023-01-21 10:32:12.933816: step: 380/533, loss: 0.1208728551864624 2023-01-21 10:32:14.105719: step: 384/533, loss: 0.05300531163811684 2023-01-21 10:32:15.278420: step: 388/533, loss: 0.006340742111206055 2023-01-21 10:32:16.451144: step: 392/533, loss: 0.013455485925078392 2023-01-21 10:32:17.580499: step: 396/533, loss: 0.05456986650824547 2023-01-21 10:32:18.782300: step: 400/533, loss: 0.0044717793352901936 2023-01-21 10:32:19.988442: step: 404/533, loss: 0.04871797561645508 2023-01-21 10:32:21.164932: step: 408/533, loss: 0.022593878209590912 2023-01-21 10:32:22.303588: step: 412/533, loss: 0.09443550556898117 2023-01-21 10:32:23.459495: step: 416/533, loss: 0.06721778213977814 2023-01-21 10:32:24.639460: step: 420/533, loss: 0.12110957503318787 2023-01-21 10:32:25.820459: step: 424/533, loss: 0.011982965283095837 2023-01-21 10:32:27.017830: step: 428/533, loss: 0.02914142608642578 2023-01-21 10:32:28.189380: step: 432/533, loss: 0.05497932434082031 2023-01-21 10:32:29.345908: step: 436/533, loss: 0.007151222787797451 2023-01-21 10:32:30.568917: step: 440/533, loss: 0.17595058679580688 2023-01-21 10:32:31.759528: step: 444/533, loss: 0.027852630242705345 2023-01-21 10:32:32.912909: step: 448/533, loss: 0.055014923214912415 2023-01-21 10:32:34.144544: step: 452/533, loss: 0.044091418385505676 2023-01-21 10:32:35.317729: step: 456/533, loss: 0.05246572569012642 2023-01-21 10:32:36.501773: step: 460/533, loss: 0.06380920112133026 2023-01-21 10:32:37.674788: step: 464/533, loss: 0.008252429775893688 2023-01-21 10:32:38.860354: step: 468/533, loss: 0.1192958801984787 2023-01-21 10:32:40.056729: step: 472/533, loss: 0.036467358469963074 2023-01-21 10:32:41.213045: step: 476/533, loss: 0.0562971867620945 2023-01-21 10:32:42.388946: step: 480/533, loss: 0.02591428905725479 2023-01-21 10:32:43.562979: step: 484/533, loss: 0.07569951564073563 2023-01-21 10:32:44.776683: step: 488/533, loss: 0.05282320827245712 2023-01-21 10:32:45.963263: step: 492/533, loss: 0.04266824945807457 2023-01-21 10:32:47.107045: step: 496/533, loss: 0.13862422108650208 2023-01-21 10:32:48.289945: step: 500/533, loss: 0.15836314857006073 2023-01-21 10:32:49.443331: step: 504/533, loss: 0.055207252502441406 2023-01-21 10:32:50.646690: step: 508/533, loss: 0.015988362953066826 2023-01-21 10:32:51.809073: step: 512/533, loss: 0.053771305829286575 2023-01-21 10:32:53.009922: step: 516/533, loss: 0.08956480026245117 2023-01-21 10:32:54.232633: step: 520/533, loss: 0.13842526078224182 2023-01-21 10:32:55.406758: step: 524/533, loss: 0.02561025694012642 2023-01-21 10:32:56.600164: step: 528/533, loss: 0.04951143264770508 2023-01-21 10:32:57.764977: step: 532/533, loss: 0.00173273088876158 2023-01-21 10:32:58.910890: step: 536/533, loss: 0.06415396183729172 2023-01-21 10:33:00.073984: step: 540/533, loss: 0.12729597091674805 2023-01-21 10:33:01.267851: step: 544/533, loss: 0.025625038892030716 2023-01-21 10:33:02.472040: step: 548/533, loss: 0.03579330816864967 2023-01-21 10:33:03.730725: step: 552/533, loss: 0.011406230740249157 2023-01-21 10:33:04.911253: step: 556/533, loss: 0.24583172798156738 2023-01-21 10:33:06.077922: step: 560/533, loss: 0.05475788190960884 2023-01-21 10:33:07.229717: step: 564/533, loss: 0.014833450317382812 2023-01-21 10:33:08.400152: step: 568/533, loss: 0.03657722473144531 2023-01-21 10:33:09.600510: step: 572/533, loss: 0.03354253992438316 2023-01-21 10:33:10.774020: step: 576/533, loss: 0.027268314734101295 2023-01-21 10:33:11.980001: step: 580/533, loss: 0.2000339776277542 2023-01-21 10:33:13.154141: step: 584/533, loss: 0.00023698806762695312 2023-01-21 10:33:14.411301: step: 588/533, loss: 0.02266082912683487 2023-01-21 10:33:15.605021: step: 592/533, loss: 0.09441972523927689 2023-01-21 10:33:16.737703: step: 596/533, loss: 0.18726615607738495 2023-01-21 10:33:17.901977: step: 600/533, loss: 0.005542755126953125 2023-01-21 10:33:19.090560: step: 604/533, loss: 0.0045563699677586555 2023-01-21 10:33:20.290680: step: 608/533, loss: 0.015193367376923561 2023-01-21 10:33:21.456346: step: 612/533, loss: 0.005956745240837336 2023-01-21 10:33:22.643493: step: 616/533, loss: 0.017676448449492455 2023-01-21 10:33:23.856513: step: 620/533, loss: 0.02517547644674778 2023-01-21 10:33:25.016143: step: 624/533, loss: 0.08792870491743088 2023-01-21 10:33:26.215414: step: 628/533, loss: 0.34034356474876404 2023-01-21 10:33:27.413711: step: 632/533, loss: 0.09360618889331818 2023-01-21 10:33:28.598366: step: 636/533, loss: 0.31324413418769836 2023-01-21 10:33:29.779394: step: 640/533, loss: 0.03742113336920738 2023-01-21 10:33:30.953699: step: 644/533, loss: 0.02005005069077015 2023-01-21 10:33:32.084869: step: 648/533, loss: 0.007426857948303223 2023-01-21 10:33:33.250811: step: 652/533, loss: 0.035787202417850494 2023-01-21 10:33:34.458303: step: 656/533, loss: 0.040491294115781784 2023-01-21 10:33:35.659679: step: 660/533, loss: 0.08698378503322601 2023-01-21 10:33:36.874513: step: 664/533, loss: 0.32028070092201233 2023-01-21 10:33:38.090776: step: 668/533, loss: 0.12318526208400726 2023-01-21 10:33:39.302780: step: 672/533, loss: 0.09167566150426865 2023-01-21 10:33:40.477332: step: 676/533, loss: 0.03479957580566406 2023-01-21 10:33:41.686568: step: 680/533, loss: 0.06010895222425461 2023-01-21 10:33:42.870734: step: 684/533, loss: 0.04062480852007866 2023-01-21 10:33:44.088790: step: 688/533, loss: 0.05848865583539009 2023-01-21 10:33:45.287595: step: 692/533, loss: 0.03724632412195206 2023-01-21 10:33:46.457944: step: 696/533, loss: 0.07625627517700195 2023-01-21 10:33:47.621304: step: 700/533, loss: 0.008632754907011986 2023-01-21 10:33:48.816898: step: 704/533, loss: 0.015421295538544655 2023-01-21 10:33:50.005018: step: 708/533, loss: 0.041405774652957916 2023-01-21 10:33:51.186782: step: 712/533, loss: 0.02901287004351616 2023-01-21 10:33:52.403411: step: 716/533, loss: 0.12156916409730911 2023-01-21 10:33:53.590133: step: 720/533, loss: 0.10615520924329758 2023-01-21 10:33:54.783284: step: 724/533, loss: 0.0732932984828949 2023-01-21 10:33:55.980494: step: 728/533, loss: 0.12164249271154404 2023-01-21 10:33:57.194683: step: 732/533, loss: 0.014490867033600807 2023-01-21 10:33:58.333010: step: 736/533, loss: 0.07560920715332031 2023-01-21 10:33:59.499126: step: 740/533, loss: 0.008276844397187233 2023-01-21 10:34:00.675535: step: 744/533, loss: 0.048221781849861145 2023-01-21 10:34:01.875927: step: 748/533, loss: 0.06289515644311905 2023-01-21 10:34:03.071385: step: 752/533, loss: 0.008064167574048042 2023-01-21 10:34:04.262947: step: 756/533, loss: 0.0025943757500499487 2023-01-21 10:34:05.487653: step: 760/533, loss: 0.05871171876788139 2023-01-21 10:34:06.638370: step: 764/533, loss: 0.022032355889678 2023-01-21 10:34:07.801827: step: 768/533, loss: 0.016852283850312233 2023-01-21 10:34:08.971444: step: 772/533, loss: 0.000576019287109375 2023-01-21 10:34:10.167573: step: 776/533, loss: 0.1648513376712799 2023-01-21 10:34:11.391972: step: 780/533, loss: 0.09577111899852753 2023-01-21 10:34:12.599906: step: 784/533, loss: 0.0040313247591257095 2023-01-21 10:34:13.753678: step: 788/533, loss: 0.03049830161035061 2023-01-21 10:34:14.908726: step: 792/533, loss: 0.0767643004655838 2023-01-21 10:34:16.101542: step: 796/533, loss: 0.09438591450452805 2023-01-21 10:34:17.252459: step: 800/533, loss: 0.10875463485717773 2023-01-21 10:34:18.454633: step: 804/533, loss: 0.11499723792076111 2023-01-21 10:34:19.650649: step: 808/533, loss: 0.06844482570886612 2023-01-21 10:34:20.836160: step: 812/533, loss: 0.020340634509921074 2023-01-21 10:34:22.047945: step: 816/533, loss: 0.100629523396492 2023-01-21 10:34:23.234171: step: 820/533, loss: 0.028775835409760475 2023-01-21 10:34:24.425117: step: 824/533, loss: 0.037395671010017395 2023-01-21 10:34:25.611090: step: 828/533, loss: 0.4053604304790497 2023-01-21 10:34:26.785280: step: 832/533, loss: 0.06030865013599396 2023-01-21 10:34:27.962538: step: 836/533, loss: 0.008334731683135033 2023-01-21 10:34:29.151751: step: 840/533, loss: 0.010800456628203392 2023-01-21 10:34:30.335510: step: 844/533, loss: 0.028617478907108307 2023-01-21 10:34:31.547629: step: 848/533, loss: 0.6874879598617554 2023-01-21 10:34:32.727555: step: 852/533, loss: 0.29638758301734924 2023-01-21 10:34:33.929773: step: 856/533, loss: 0.11476555466651917 2023-01-21 10:34:35.085691: step: 860/533, loss: 0.09508753567934036 2023-01-21 10:34:36.247511: step: 864/533, loss: 0.01524662971496582 2023-01-21 10:34:37.399592: step: 868/533, loss: 0.06694383919239044 2023-01-21 10:34:38.566375: step: 872/533, loss: 0.0042227269150316715 2023-01-21 10:34:39.746016: step: 876/533, loss: 0.16895513236522675 2023-01-21 10:34:40.878430: step: 880/533, loss: 0.06287021934986115 2023-01-21 10:34:42.070658: step: 884/533, loss: 0.03757200017571449 2023-01-21 10:34:43.261143: step: 888/533, loss: 0.06285929679870605 2023-01-21 10:34:44.431420: step: 892/533, loss: 0.023138046264648438 2023-01-21 10:34:45.634748: step: 896/533, loss: 0.017914963886141777 2023-01-21 10:34:46.812248: step: 900/533, loss: 0.002557039260864258 2023-01-21 10:34:47.991419: step: 904/533, loss: 0.0672997385263443 2023-01-21 10:34:49.175831: step: 908/533, loss: 0.03160591050982475 2023-01-21 10:34:50.361034: step: 912/533, loss: 0.2144254595041275 2023-01-21 10:34:51.505127: step: 916/533, loss: 0.047751907259225845 2023-01-21 10:34:52.659214: step: 920/533, loss: 0.08782310783863068 2023-01-21 10:34:53.835272: step: 924/533, loss: 0.0461915023624897 2023-01-21 10:34:54.994037: step: 928/533, loss: 0.013929082080721855 2023-01-21 10:34:56.190419: step: 932/533, loss: 0.11307516694068909 2023-01-21 10:34:57.334911: step: 936/533, loss: 0.04707388952374458 2023-01-21 10:34:58.543184: step: 940/533, loss: 0.09293480217456818 2023-01-21 10:34:59.732374: step: 944/533, loss: 0.0036401748657226562 2023-01-21 10:35:00.917777: step: 948/533, loss: 0.08520784974098206 2023-01-21 10:35:02.103842: step: 952/533, loss: 0.017494583502411842 2023-01-21 10:35:03.288028: step: 956/533, loss: 0.011858273297548294 2023-01-21 10:35:04.444785: step: 960/533, loss: 0.0019500732887536287 2023-01-21 10:35:05.601713: step: 964/533, loss: 0.06429986655712128 2023-01-21 10:35:06.800294: step: 968/533, loss: 0.12511055171489716 2023-01-21 10:35:08.010335: step: 972/533, loss: 0.02482757531106472 2023-01-21 10:35:09.198333: step: 976/533, loss: 0.01854996755719185 2023-01-21 10:35:10.367116: step: 980/533, loss: 0.033623307943344116 2023-01-21 10:35:11.566113: step: 984/533, loss: 0.06684722751379013 2023-01-21 10:35:12.763503: step: 988/533, loss: 0.02318502776324749 2023-01-21 10:35:13.944660: step: 992/533, loss: 0.02847576141357422 2023-01-21 10:35:15.123048: step: 996/533, loss: 0.08658294379711151 2023-01-21 10:35:16.307802: step: 1000/533, loss: 0.02443409152328968 2023-01-21 10:35:17.513483: step: 1004/533, loss: 0.007188225165009499 2023-01-21 10:35:18.684519: step: 1008/533, loss: 0.019654560834169388 2023-01-21 10:35:19.820674: step: 1012/533, loss: 0.053139831870794296 2023-01-21 10:35:21.031276: step: 1016/533, loss: 0.05817057937383652 2023-01-21 10:35:22.217278: step: 1020/533, loss: 0.13626976311206818 2023-01-21 10:35:23.395113: step: 1024/533, loss: 0.05506887286901474 2023-01-21 10:35:24.563679: step: 1028/533, loss: 0.027257252484560013 2023-01-21 10:35:25.779992: step: 1032/533, loss: 0.037662312388420105 2023-01-21 10:35:26.939206: step: 1036/533, loss: 0.051063500344753265 2023-01-21 10:35:28.108397: step: 1040/533, loss: 0.042043305933475494 2023-01-21 10:35:29.303121: step: 1044/533, loss: 0.05184240639209747 2023-01-21 10:35:30.473884: step: 1048/533, loss: 0.018011093139648438 2023-01-21 10:35:31.615445: step: 1052/533, loss: 0.03362541273236275 2023-01-21 10:35:32.793064: step: 1056/533, loss: 0.0038625719025731087 2023-01-21 10:35:33.955538: step: 1060/533, loss: 0.011679555289447308 2023-01-21 10:35:35.161366: step: 1064/533, loss: 0.01837022230029106 2023-01-21 10:35:36.380530: step: 1068/533, loss: 0.04417762905359268 2023-01-21 10:35:37.549009: step: 1072/533, loss: 0.010522127151489258 2023-01-21 10:35:38.746467: step: 1076/533, loss: 0.061770249158144 2023-01-21 10:35:39.892703: step: 1080/533, loss: 0.07209939509630203 2023-01-21 10:35:41.057967: step: 1084/533, loss: 0.6111618280410767 2023-01-21 10:35:42.271462: step: 1088/533, loss: 0.02486734464764595 2023-01-21 10:35:43.430824: step: 1092/533, loss: 0.010312545113265514 2023-01-21 10:35:44.641839: step: 1096/533, loss: 0.0774257630109787 2023-01-21 10:35:45.830494: step: 1100/533, loss: 0.007196473889052868 2023-01-21 10:35:46.974204: step: 1104/533, loss: 0.046445272862911224 2023-01-21 10:35:48.155511: step: 1108/533, loss: 0.03284912183880806 2023-01-21 10:35:49.360545: step: 1112/533, loss: 0.03728685528039932 2023-01-21 10:35:50.478351: step: 1116/533, loss: 0.062476254999637604 2023-01-21 10:35:51.663242: step: 1120/533, loss: 0.0112457275390625 2023-01-21 10:35:52.834627: step: 1124/533, loss: 0.0990542471408844 2023-01-21 10:35:54.045048: step: 1128/533, loss: 0.06675414741039276 2023-01-21 10:35:55.217110: step: 1132/533, loss: 0.013537216931581497 2023-01-21 10:35:56.361423: step: 1136/533, loss: 0.13270586729049683 2023-01-21 10:35:57.511547: step: 1140/533, loss: 0.07421837002038956 2023-01-21 10:35:58.670865: step: 1144/533, loss: 0.0027669547125697136 2023-01-21 10:35:59.860630: step: 1148/533, loss: 0.11421117931604385 2023-01-21 10:36:01.062742: step: 1152/533, loss: 0.054032422602176666 2023-01-21 10:36:02.262497: step: 1156/533, loss: 0.025281094014644623 2023-01-21 10:36:03.470614: step: 1160/533, loss: 0.17904329299926758 2023-01-21 10:36:04.665699: step: 1164/533, loss: 0.04856109619140625 2023-01-21 10:36:05.814788: step: 1168/533, loss: 0.058324433863162994 2023-01-21 10:36:07.018722: step: 1172/533, loss: 0.06331968307495117 2023-01-21 10:36:08.188816: step: 1176/533, loss: 0.2400626242160797 2023-01-21 10:36:09.340800: step: 1180/533, loss: 0.03560075908899307 2023-01-21 10:36:10.534023: step: 1184/533, loss: 0.009624456986784935 2023-01-21 10:36:11.729906: step: 1188/533, loss: 0.04755058512091637 2023-01-21 10:36:12.895400: step: 1192/533, loss: 0.06912145763635635 2023-01-21 10:36:14.067420: step: 1196/533, loss: 0.0572635643184185 2023-01-21 10:36:15.257920: step: 1200/533, loss: 0.0376192107796669 2023-01-21 10:36:16.469884: step: 1204/533, loss: 0.057989977300167084 2023-01-21 10:36:17.649829: step: 1208/533, loss: 0.03752708435058594 2023-01-21 10:36:18.810123: step: 1212/533, loss: 0.08052120357751846 2023-01-21 10:36:19.964910: step: 1216/533, loss: 0.003288400126621127 2023-01-21 10:36:21.136856: step: 1220/533, loss: 0.022491026669740677 2023-01-21 10:36:22.329472: step: 1224/533, loss: 0.0004567146534100175 2023-01-21 10:36:23.523729: step: 1228/533, loss: 0.15962447226047516 2023-01-21 10:36:24.686434: step: 1232/533, loss: 0.06012604385614395 2023-01-21 10:36:25.843785: step: 1236/533, loss: 0.06118650361895561 2023-01-21 10:36:27.010394: step: 1240/533, loss: 0.04592762142419815 2023-01-21 10:36:28.165366: step: 1244/533, loss: 0.01642293855547905 2023-01-21 10:36:29.353873: step: 1248/533, loss: 0.0709286704659462 2023-01-21 10:36:30.573998: step: 1252/533, loss: 0.6426746249198914 2023-01-21 10:36:31.733123: step: 1256/533, loss: 0.024199390783905983 2023-01-21 10:36:32.903535: step: 1260/533, loss: 0.029232501983642578 2023-01-21 10:36:34.068314: step: 1264/533, loss: 0.020049571990966797 2023-01-21 10:36:35.257849: step: 1268/533, loss: 0.05558958277106285 2023-01-21 10:36:36.409952: step: 1272/533, loss: 0.06585562974214554 2023-01-21 10:36:37.576042: step: 1276/533, loss: 0.0010228157043457031 2023-01-21 10:36:38.784186: step: 1280/533, loss: 0.2014915496110916 2023-01-21 10:36:39.894217: step: 1284/533, loss: 0.046613503247499466 2023-01-21 10:36:41.086965: step: 1288/533, loss: 0.044942282140254974 2023-01-21 10:36:42.248023: step: 1292/533, loss: 0.010610103607177734 2023-01-21 10:36:43.454066: step: 1296/533, loss: 0.029195215553045273 2023-01-21 10:36:44.680132: step: 1300/533, loss: 0.43042439222335815 2023-01-21 10:36:45.859294: step: 1304/533, loss: 0.018293000757694244 2023-01-21 10:36:47.103927: step: 1308/533, loss: 0.04098348692059517 2023-01-21 10:36:48.270643: step: 1312/533, loss: 0.11825218796730042 2023-01-21 10:36:49.430124: step: 1316/533, loss: 0.06375002861022949 2023-01-21 10:36:50.641267: step: 1320/533, loss: 0.08055982738733292 2023-01-21 10:36:51.831828: step: 1324/533, loss: 0.039058782160282135 2023-01-21 10:36:53.035027: step: 1328/533, loss: 0.06780214607715607 2023-01-21 10:36:54.216501: step: 1332/533, loss: 0.05446338653564453 2023-01-21 10:36:55.408242: step: 1336/533, loss: 0.05097074434161186 2023-01-21 10:36:56.592545: step: 1340/533, loss: 1.152879238128662 2023-01-21 10:36:57.759629: step: 1344/533, loss: 0.001506805419921875 2023-01-21 10:36:58.953374: step: 1348/533, loss: 0.06723959743976593 2023-01-21 10:37:00.138616: step: 1352/533, loss: 0.01587352715432644 2023-01-21 10:37:01.303445: step: 1356/533, loss: 0.035622358322143555 2023-01-21 10:37:02.499276: step: 1360/533, loss: 0.34401464462280273 2023-01-21 10:37:03.699939: step: 1364/533, loss: 0.10021839290857315 2023-01-21 10:37:04.866446: step: 1368/533, loss: 0.037081338465213776 2023-01-21 10:37:06.044593: step: 1372/533, loss: 0.03386535495519638 2023-01-21 10:37:07.224339: step: 1376/533, loss: 0.011300945654511452 2023-01-21 10:37:08.385385: step: 1380/533, loss: 0.13766269385814667 2023-01-21 10:37:09.583128: step: 1384/533, loss: 0.018306540325284004 2023-01-21 10:37:10.797808: step: 1388/533, loss: 0.00264053326100111 2023-01-21 10:37:11.965299: step: 1392/533, loss: 0.0549396276473999 2023-01-21 10:37:13.165435: step: 1396/533, loss: 0.08497276902198792 2023-01-21 10:37:14.353471: step: 1400/533, loss: 0.030950356274843216 2023-01-21 10:37:15.521122: step: 1404/533, loss: 0.049021054059267044 2023-01-21 10:37:16.668243: step: 1408/533, loss: 0.024622201919555664 2023-01-21 10:37:17.961405: step: 1412/533, loss: 0.039582788944244385 2023-01-21 10:37:19.129779: step: 1416/533, loss: 0.04685864597558975 2023-01-21 10:37:20.293766: step: 1420/533, loss: 0.014054680243134499 2023-01-21 10:37:21.487439: step: 1424/533, loss: 0.1489175409078598 2023-01-21 10:37:22.667177: step: 1428/533, loss: 0.10843181610107422 2023-01-21 10:37:23.826147: step: 1432/533, loss: 0.11393924057483673 2023-01-21 10:37:25.083043: step: 1436/533, loss: 0.025747396051883698 2023-01-21 10:37:26.319112: step: 1440/533, loss: 0.41678953170776367 2023-01-21 10:37:27.517809: step: 1444/533, loss: 0.039380647242069244 2023-01-21 10:37:28.706844: step: 1448/533, loss: 0.051689907908439636 2023-01-21 10:37:29.937024: step: 1452/533, loss: 0.535626232624054 2023-01-21 10:37:31.146169: step: 1456/533, loss: 0.1808479279279709 2023-01-21 10:37:32.306880: step: 1460/533, loss: 0.04927721247076988 2023-01-21 10:37:33.455306: step: 1464/533, loss: 0.03549213334918022 2023-01-21 10:37:34.644805: step: 1468/533, loss: 0.025571823120117188 2023-01-21 10:37:35.811493: step: 1472/533, loss: 0.01535797119140625 2023-01-21 10:37:36.977308: step: 1476/533, loss: 0.06740112602710724 2023-01-21 10:37:38.189222: step: 1480/533, loss: 0.3737823963165283 2023-01-21 10:37:39.382588: step: 1484/533, loss: 0.023802900686860085 2023-01-21 10:37:40.515726: step: 1488/533, loss: 0.015548801980912685 2023-01-21 10:37:41.665965: step: 1492/533, loss: 0.04977254942059517 2023-01-21 10:37:42.892548: step: 1496/533, loss: 0.2738347351551056 2023-01-21 10:37:44.065336: step: 1500/533, loss: 0.027628321200609207 2023-01-21 10:37:45.229478: step: 1504/533, loss: 0.011148596182465553 2023-01-21 10:37:46.477812: step: 1508/533, loss: 0.003384876297786832 2023-01-21 10:37:47.683372: step: 1512/533, loss: 0.22552672028541565 2023-01-21 10:37:48.872834: step: 1516/533, loss: 0.02197723463177681 2023-01-21 10:37:50.069428: step: 1520/533, loss: 0.3720621168613434 2023-01-21 10:37:51.266385: step: 1524/533, loss: 0.009566117078065872 2023-01-21 10:37:52.481669: step: 1528/533, loss: 0.060518693178892136 2023-01-21 10:37:53.665801: step: 1532/533, loss: 0.007303190417587757 2023-01-21 10:37:54.877223: step: 1536/533, loss: 0.07793235778808594 2023-01-21 10:37:56.046455: step: 1540/533, loss: 0.003921413328498602 2023-01-21 10:37:57.219696: step: 1544/533, loss: 0.04175548627972603 2023-01-21 10:37:58.416717: step: 1548/533, loss: 0.05214815214276314 2023-01-21 10:37:59.610202: step: 1552/533, loss: 0.013699413277208805 2023-01-21 10:38:00.872625: step: 1556/533, loss: 0.05700492858886719 2023-01-21 10:38:02.089542: step: 1560/533, loss: 0.02967863157391548 2023-01-21 10:38:03.308228: step: 1564/533, loss: 0.014600848779082298 2023-01-21 10:38:04.482118: step: 1568/533, loss: 0.10859213024377823 2023-01-21 10:38:05.730961: step: 1572/533, loss: 0.006503486540168524 2023-01-21 10:38:06.921881: step: 1576/533, loss: 0.017542744055390358 2023-01-21 10:38:08.118526: step: 1580/533, loss: 0.05604248121380806 2023-01-21 10:38:09.299364: step: 1584/533, loss: 0.08737903833389282 2023-01-21 10:38:10.498419: step: 1588/533, loss: 0.04217586666345596 2023-01-21 10:38:11.669179: step: 1592/533, loss: 0.026972675696015358 2023-01-21 10:38:12.837836: step: 1596/533, loss: 0.0016846180660650134 2023-01-21 10:38:14.054977: step: 1600/533, loss: 0.060553815215826035 2023-01-21 10:38:15.249497: step: 1604/533, loss: 0.1027032881975174 2023-01-21 10:38:16.415266: step: 1608/533, loss: 0.07762566208839417 2023-01-21 10:38:17.563102: step: 1612/533, loss: 0.005103397648781538 2023-01-21 10:38:18.754033: step: 1616/533, loss: 0.022433947771787643 2023-01-21 10:38:19.921995: step: 1620/533, loss: 0.04688744619488716 2023-01-21 10:38:21.118046: step: 1624/533, loss: 0.07281551510095596 2023-01-21 10:38:22.328223: step: 1628/533, loss: 0.131850004196167 2023-01-21 10:38:23.518068: step: 1632/533, loss: 0.013134479522705078 2023-01-21 10:38:24.721414: step: 1636/533, loss: 0.08298487961292267 2023-01-21 10:38:25.898231: step: 1640/533, loss: 0.011401604861021042 2023-01-21 10:38:27.054858: step: 1644/533, loss: 0.15960845351219177 2023-01-21 10:38:28.244569: step: 1648/533, loss: 0.05998096615076065 2023-01-21 10:38:29.416800: step: 1652/533, loss: 0.042473673820495605 2023-01-21 10:38:30.596011: step: 1656/533, loss: 0.06607169657945633 2023-01-21 10:38:31.740480: step: 1660/533, loss: 0.006520366761833429 2023-01-21 10:38:32.878391: step: 1664/533, loss: 0.02886677160859108 2023-01-21 10:38:34.044276: step: 1668/533, loss: 0.015095710754394531 2023-01-21 10:38:35.222919: step: 1672/533, loss: 0.08982840180397034 2023-01-21 10:38:36.445817: step: 1676/533, loss: 0.06776418536901474 2023-01-21 10:38:37.613185: step: 1680/533, loss: 0.061571888625621796 2023-01-21 10:38:38.864682: step: 1684/533, loss: 0.05083566904067993 2023-01-21 10:38:40.068392: step: 1688/533, loss: 0.06410541385412216 2023-01-21 10:38:41.243149: step: 1692/533, loss: 0.13023224472999573 2023-01-21 10:38:42.391802: step: 1696/533, loss: 0.03539867326617241 2023-01-21 10:38:43.559584: step: 1700/533, loss: 0.014899969100952148 2023-01-21 10:38:44.694689: step: 1704/533, loss: 0.158883735537529 2023-01-21 10:38:45.869364: step: 1708/533, loss: 0.009968948550522327 2023-01-21 10:38:47.049993: step: 1712/533, loss: 0.05930042266845703 2023-01-21 10:38:48.218923: step: 1716/533, loss: 0.019159698858857155 2023-01-21 10:38:49.373265: step: 1720/533, loss: 0.031523894518613815 2023-01-21 10:38:50.588714: step: 1724/533, loss: 0.031030654907226562 2023-01-21 10:38:51.750920: step: 1728/533, loss: 0.017488574609160423 2023-01-21 10:38:52.968966: step: 1732/533, loss: 0.09975738823413849 2023-01-21 10:38:54.153361: step: 1736/533, loss: 0.014637804590165615 2023-01-21 10:38:55.329817: step: 1740/533, loss: 0.02560257911682129 2023-01-21 10:38:56.507514: step: 1744/533, loss: 0.011232377029955387 2023-01-21 10:38:57.653054: step: 1748/533, loss: 0.035165030509233475 2023-01-21 10:38:58.842169: step: 1752/533, loss: 0.03615207597613335 2023-01-21 10:39:00.031949: step: 1756/533, loss: 0.05477473884820938 2023-01-21 10:39:01.223891: step: 1760/533, loss: 0.07021446526050568 2023-01-21 10:39:02.409898: step: 1764/533, loss: 0.029623890295624733 2023-01-21 10:39:03.588863: step: 1768/533, loss: 0.029053211212158203 2023-01-21 10:39:04.780873: step: 1772/533, loss: 0.097753144800663 2023-01-21 10:39:05.977110: step: 1776/533, loss: 0.03115224838256836 2023-01-21 10:39:07.137536: step: 1780/533, loss: 0.012836361303925514 2023-01-21 10:39:08.283886: step: 1784/533, loss: 0.8643342852592468 2023-01-21 10:39:09.460534: step: 1788/533, loss: 0.03354082256555557 2023-01-21 10:39:10.634289: step: 1792/533, loss: 0.03175721317529678 2023-01-21 10:39:11.838399: step: 1796/533, loss: 0.09133930504322052 2023-01-21 10:39:13.041522: step: 1800/533, loss: 0.12361745536327362 2023-01-21 10:39:14.199170: step: 1804/533, loss: 0.04615011066198349 2023-01-21 10:39:15.432475: step: 1808/533, loss: 0.11400061100721359 2023-01-21 10:39:16.588173: step: 1812/533, loss: 0.017228269949555397 2023-01-21 10:39:17.784871: step: 1816/533, loss: 0.08150921016931534 2023-01-21 10:39:19.022025: step: 1820/533, loss: 0.02705707587301731 2023-01-21 10:39:20.214138: step: 1824/533, loss: 0.07153358310461044 2023-01-21 10:39:21.415254: step: 1828/533, loss: 0.028616715222597122 2023-01-21 10:39:22.616981: step: 1832/533, loss: 0.015455532819032669 2023-01-21 10:39:23.868727: step: 1836/533, loss: 0.12773366272449493 2023-01-21 10:39:25.031888: step: 1840/533, loss: 0.1605411171913147 2023-01-21 10:39:26.181297: step: 1844/533, loss: 0.01254820916801691 2023-01-21 10:39:27.373884: step: 1848/533, loss: 0.04598388820886612 2023-01-21 10:39:28.584899: step: 1852/533, loss: 0.11140614002943039 2023-01-21 10:39:29.758656: step: 1856/533, loss: 0.01506586093455553 2023-01-21 10:39:30.943343: step: 1860/533, loss: 0.019997311756014824 2023-01-21 10:39:32.093006: step: 1864/533, loss: 0.03732867166399956 2023-01-21 10:39:33.220046: step: 1868/533, loss: 0.006541824899613857 2023-01-21 10:39:34.408270: step: 1872/533, loss: 0.04486050829291344 2023-01-21 10:39:35.617635: step: 1876/533, loss: 0.08218993991613388 2023-01-21 10:39:36.854865: step: 1880/533, loss: 0.015301275067031384 2023-01-21 10:39:38.015328: step: 1884/533, loss: 0.03984098136425018 2023-01-21 10:39:39.195327: step: 1888/533, loss: 0.050959013402462006 2023-01-21 10:39:40.389862: step: 1892/533, loss: 0.1539476364850998 2023-01-21 10:39:41.568678: step: 1896/533, loss: 0.03992071375250816 2023-01-21 10:39:42.728633: step: 1900/533, loss: 0.09901275485754013 2023-01-21 10:39:43.906566: step: 1904/533, loss: 0.10107841342687607 2023-01-21 10:39:45.079146: step: 1908/533, loss: 0.020864101126790047 2023-01-21 10:39:46.283279: step: 1912/533, loss: 0.07817935943603516 2023-01-21 10:39:47.487585: step: 1916/533, loss: 0.09975185245275497 2023-01-21 10:39:48.664539: step: 1920/533, loss: 0.06512641906738281 2023-01-21 10:39:49.869009: step: 1924/533, loss: 0.057581521570682526 2023-01-21 10:39:51.110243: step: 1928/533, loss: 0.03314018249511719 2023-01-21 10:39:52.264754: step: 1932/533, loss: 0.013709831982851028 2023-01-21 10:39:53.475076: step: 1936/533, loss: 0.06103997677564621 2023-01-21 10:39:54.684100: step: 1940/533, loss: 0.047936033457517624 2023-01-21 10:39:55.857974: step: 1944/533, loss: 0.040651895105838776 2023-01-21 10:39:57.049992: step: 1948/533, loss: 0.008717156015336514 2023-01-21 10:39:58.238664: step: 1952/533, loss: 0.025026893243193626 2023-01-21 10:39:59.408470: step: 1956/533, loss: 0.019580364227294922 2023-01-21 10:40:00.588711: step: 1960/533, loss: 0.014453220181167126 2023-01-21 10:40:01.787818: step: 1964/533, loss: 0.03233394771814346 2023-01-21 10:40:02.979790: step: 1968/533, loss: 0.6248613595962524 2023-01-21 10:40:04.194972: step: 1972/533, loss: 0.12208108603954315 2023-01-21 10:40:05.387613: step: 1976/533, loss: 0.04608125612139702 2023-01-21 10:40:06.588009: step: 1980/533, loss: 0.09121675789356232 2023-01-21 10:40:07.794131: step: 1984/533, loss: 0.08527243137359619 2023-01-21 10:40:08.974394: step: 1988/533, loss: 0.07880210876464844 2023-01-21 10:40:10.153082: step: 1992/533, loss: 0.06456947326660156 2023-01-21 10:40:11.360200: step: 1996/533, loss: 0.026406098157167435 2023-01-21 10:40:12.573133: step: 2000/533, loss: 0.05671081691980362 2023-01-21 10:40:13.771633: step: 2004/533, loss: 0.0009198188781738281 2023-01-21 10:40:14.929821: step: 2008/533, loss: 0.04607219621539116 2023-01-21 10:40:16.145745: step: 2012/533, loss: 0.05803294479846954 2023-01-21 10:40:17.316334: step: 2016/533, loss: 0.02836422808468342 2023-01-21 10:40:18.498016: step: 2020/533, loss: 0.06971683353185654 2023-01-21 10:40:19.715884: step: 2024/533, loss: 0.0008287429809570312 2023-01-21 10:40:20.937814: step: 2028/533, loss: 0.051264382898807526 2023-01-21 10:40:22.111068: step: 2032/533, loss: 0.0007672309875488281 2023-01-21 10:40:23.326988: step: 2036/533, loss: 0.029651736840605736 2023-01-21 10:40:24.541513: step: 2040/533, loss: 0.11418915539979935 2023-01-21 10:40:25.762486: step: 2044/533, loss: 0.033976174890995026 2023-01-21 10:40:26.952431: step: 2048/533, loss: 0.057324983179569244 2023-01-21 10:40:28.102785: step: 2052/533, loss: 0.06552743911743164 2023-01-21 10:40:29.339719: step: 2056/533, loss: 0.05742397531867027 2023-01-21 10:40:30.504062: step: 2060/533, loss: 0.034281156957149506 2023-01-21 10:40:31.688513: step: 2064/533, loss: 0.004154014401137829 2023-01-21 10:40:32.915734: step: 2068/533, loss: 0.04763736575841904 2023-01-21 10:40:34.072987: step: 2072/533, loss: 0.062172889709472656 2023-01-21 10:40:35.262727: step: 2076/533, loss: 0.03974580764770508 2023-01-21 10:40:36.450565: step: 2080/533, loss: 0.08486881107091904 2023-01-21 10:40:37.631660: step: 2084/533, loss: 0.06954784691333771 2023-01-21 10:40:38.809865: step: 2088/533, loss: 0.058583930134773254 2023-01-21 10:40:40.026471: step: 2092/533, loss: 0.04322100058197975 2023-01-21 10:40:41.196744: step: 2096/533, loss: 0.04362912476062775 2023-01-21 10:40:42.372655: step: 2100/533, loss: 0.05357861518859863 2023-01-21 10:40:43.536658: step: 2104/533, loss: 0.047242093831300735 2023-01-21 10:40:44.690756: step: 2108/533, loss: 0.04727339744567871 2023-01-21 10:40:45.890656: step: 2112/533, loss: 0.1299700289964676 2023-01-21 10:40:47.092817: step: 2116/533, loss: 0.024875832721590996 2023-01-21 10:40:48.292746: step: 2120/533, loss: 0.020792867988348007 2023-01-21 10:40:49.526379: step: 2124/533, loss: 0.08784227073192596 2023-01-21 10:40:50.698545: step: 2128/533, loss: 0.010356160812079906 2023-01-21 10:40:51.859639: step: 2132/533, loss: 0.08221583813428879 ================================================== Loss: 0.069 -------------------- Dev: {'event': {'p': 0.6244541484716157, 'r': 0.7616511318242344, 'f1': 0.6862627474505099}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Test: {'event': {'p': 0.6416624429802331, 'r': 0.809462915601023, 'f1': 0.7158608990670059}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} 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': 11} Korean: {'event': {'p': 0.5769230769230769, 'r': 0.47619047619047616, 'f1': 0.5217391304347826}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} 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': 11} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 10:41:33.524179: step: 4/533, loss: 0.041904352605342865 2023-01-21 10:41:34.700559: step: 8/533, loss: 0.0374697670340538 2023-01-21 10:41:35.913424: step: 12/533, loss: 0.06795825809240341 2023-01-21 10:41:37.101383: step: 16/533, loss: 0.07436218857765198 2023-01-21 10:41:38.291625: step: 20/533, loss: 0.02218928374350071 2023-01-21 10:41:39.481045: step: 24/533, loss: 0.028714274987578392 2023-01-21 10:41:40.658097: step: 28/533, loss: 0.01710367389023304 2023-01-21 10:41:41.864107: step: 32/533, loss: 0.02336578257381916 2023-01-21 10:41:43.014027: step: 36/533, loss: 0.017100047320127487 2023-01-21 10:41:44.177493: step: 40/533, loss: 0.1393013894557953 2023-01-21 10:41:45.372462: step: 44/533, loss: 0.03557715564966202 2023-01-21 10:41:46.556452: step: 48/533, loss: 0.03423051908612251 2023-01-21 10:41:47.744533: step: 52/533, loss: 0.039509013295173645 2023-01-21 10:41:48.941637: step: 56/533, loss: 0.007215403951704502 2023-01-21 10:41:50.148625: step: 60/533, loss: 0.062174417078495026 2023-01-21 10:41:51.382512: step: 64/533, loss: 0.052642252296209335 2023-01-21 10:41:52.556821: step: 68/533, loss: 0.06284160912036896 2023-01-21 10:41:53.743137: step: 72/533, loss: 0.03643188625574112 2023-01-21 10:41:54.922249: step: 76/533, loss: 0.003572177840396762 2023-01-21 10:41:56.130294: step: 80/533, loss: 0.09779243916273117 2023-01-21 10:41:57.288165: step: 84/533, loss: 0.17891749739646912 2023-01-21 10:41:58.470219: step: 88/533, loss: 0.02024994045495987 2023-01-21 10:41:59.683401: step: 92/533, loss: 0.18497686088085175 2023-01-21 10:42:00.843121: step: 96/533, loss: 0.07873706519603729 2023-01-21 10:42:02.022423: step: 100/533, loss: 0.035846613347530365 2023-01-21 10:42:03.201901: step: 104/533, loss: 0.019582651555538177 2023-01-21 10:42:04.364667: step: 108/533, loss: 0.019284486770629883 2023-01-21 10:42:05.562717: step: 112/533, loss: 0.027866506949067116 2023-01-21 10:42:06.780263: step: 116/533, loss: 0.1550152748823166 2023-01-21 10:42:08.012127: step: 120/533, loss: 0.12320385128259659 2023-01-21 10:42:09.181965: step: 124/533, loss: 0.03342805057764053 2023-01-21 10:42:10.368335: step: 128/533, loss: 0.0074793933890759945 2023-01-21 10:42:11.531675: step: 132/533, loss: 0.012518501840531826 2023-01-21 10:42:12.752676: step: 136/533, loss: 0.0877528190612793 2023-01-21 10:42:13.959988: step: 140/533, loss: 0.0618189200758934 2023-01-21 10:42:15.145105: step: 144/533, loss: 0.01933436468243599 2023-01-21 10:42:16.318556: step: 148/533, loss: 0.03780355304479599 2023-01-21 10:42:17.497025: step: 152/533, loss: 0.02359323389828205 2023-01-21 10:42:18.689445: step: 156/533, loss: 0.040283203125 2023-01-21 10:42:19.842951: step: 160/533, loss: 0.06320181488990784 2023-01-21 10:42:21.052213: step: 164/533, loss: 0.02956409379839897 2023-01-21 10:42:22.219020: step: 168/533, loss: 0.002128219697624445 2023-01-21 10:42:23.412784: step: 172/533, loss: 0.027791213244199753 2023-01-21 10:42:24.543389: step: 176/533, loss: 0.06364583969116211 2023-01-21 10:42:25.697422: step: 180/533, loss: 0.027123406529426575 2023-01-21 10:42:26.880500: step: 184/533, loss: 0.03582916408777237 2023-01-21 10:42:28.087125: step: 188/533, loss: 0.004851293750107288 2023-01-21 10:42:29.269226: step: 192/533, loss: 0.1057400330901146 2023-01-21 10:42:30.419747: step: 196/533, loss: 0.013115882873535156 2023-01-21 10:42:31.607669: step: 200/533, loss: 0.04010210186243057 2023-01-21 10:42:32.853720: step: 204/533, loss: 0.029671097174286842 2023-01-21 10:42:34.043837: step: 208/533, loss: 0.00581812858581543 2023-01-21 10:42:35.231236: step: 212/533, loss: 0.008448028936982155 2023-01-21 10:42:36.454778: step: 216/533, loss: 0.013164937496185303 2023-01-21 10:42:37.646333: step: 220/533, loss: 0.005957365036010742 2023-01-21 10:42:38.790882: step: 224/533, loss: 0.01655890978872776 2023-01-21 10:42:39.934947: step: 228/533, loss: 0.1682807058095932 2023-01-21 10:42:41.099906: step: 232/533, loss: 0.005534934811294079 2023-01-21 10:42:42.255849: step: 236/533, loss: 0.016397953033447266 2023-01-21 10:42:43.467077: step: 240/533, loss: 0.201178640127182 2023-01-21 10:42:44.621661: step: 244/533, loss: 0.00960841216146946 2023-01-21 10:42:45.804046: step: 248/533, loss: 0.057245831936597824 2023-01-21 10:42:46.994429: step: 252/533, loss: 0.020941734313964844 2023-01-21 10:42:48.179757: step: 256/533, loss: 0.051650047302246094 2023-01-21 10:42:49.354830: step: 260/533, loss: 0.005865669343620539 2023-01-21 10:42:50.522921: step: 264/533, loss: 0.1919335275888443 2023-01-21 10:42:51.687724: step: 268/533, loss: 0.16353721916675568 2023-01-21 10:42:52.831853: step: 272/533, loss: 0.0005916595691815019 2023-01-21 10:42:54.016337: step: 276/533, loss: 0.044652655720710754 2023-01-21 10:42:55.182436: step: 280/533, loss: 0.01705150678753853 2023-01-21 10:42:56.360944: step: 284/533, loss: 0.004418182652443647 2023-01-21 10:42:57.580028: step: 288/533, loss: 0.0002839088556356728 2023-01-21 10:42:58.795800: step: 292/533, loss: 0.05885763093829155 2023-01-21 10:42:59.960114: step: 296/533, loss: 0.004337406251579523 2023-01-21 10:43:01.148429: step: 300/533, loss: 0.06694946438074112 2023-01-21 10:43:02.389776: step: 304/533, loss: 0.038599491119384766 2023-01-21 10:43:03.543224: step: 308/533, loss: 0.05937004089355469 2023-01-21 10:43:04.754949: step: 312/533, loss: 0.00040993691072799265 2023-01-21 10:43:05.914126: step: 316/533, loss: 0.011014366522431374 2023-01-21 10:43:07.102048: step: 320/533, loss: 0.048088837414979935 2023-01-21 10:43:08.269061: step: 324/533, loss: 0.01701197773218155 2023-01-21 10:43:09.433028: step: 328/533, loss: 0.025472545996308327 2023-01-21 10:43:10.633873: step: 332/533, loss: 0.011154365725815296 2023-01-21 10:43:11.846199: step: 336/533, loss: 0.024979686364531517 2023-01-21 10:43:13.006112: step: 340/533, loss: 0.02335391193628311 2023-01-21 10:43:14.181364: step: 344/533, loss: 0.10278654098510742 2023-01-21 10:43:15.386436: step: 348/533, loss: 0.052660562098026276 2023-01-21 10:43:16.610262: step: 352/533, loss: 0.05932150036096573 2023-01-21 10:43:17.741752: step: 356/533, loss: 0.020996378734707832 2023-01-21 10:43:18.927652: step: 360/533, loss: 0.04142170026898384 2023-01-21 10:43:20.167680: step: 364/533, loss: 0.11402206122875214 2023-01-21 10:43:21.377470: step: 368/533, loss: 0.029016535729169846 2023-01-21 10:43:22.552259: step: 372/533, loss: 0.05256099998950958 2023-01-21 10:43:23.719755: step: 376/533, loss: 0.07133827358484268 2023-01-21 10:43:24.904387: step: 380/533, loss: 0.02216320112347603 2023-01-21 10:43:26.098224: step: 384/533, loss: 0.0813487097620964 2023-01-21 10:43:27.254135: step: 388/533, loss: 0.021233130246400833 2023-01-21 10:43:28.499450: step: 392/533, loss: 0.02132740058004856 2023-01-21 10:43:29.746693: step: 396/533, loss: 0.027507878839969635 2023-01-21 10:43:30.947792: step: 400/533, loss: 0.008081817999482155 2023-01-21 10:43:32.130871: step: 404/533, loss: 0.019214915111660957 2023-01-21 10:43:33.336403: step: 408/533, loss: 0.01882343180477619 2023-01-21 10:43:34.541485: step: 412/533, loss: 0.08627300709486008 2023-01-21 10:43:35.763432: step: 416/533, loss: 0.03166494518518448 2023-01-21 10:43:36.940781: step: 420/533, loss: 0.007539940066635609 2023-01-21 10:43:38.134811: step: 424/533, loss: 0.01225423812866211 2023-01-21 10:43:39.317240: step: 428/533, loss: 0.48720666766166687 2023-01-21 10:43:40.501124: step: 432/533, loss: 0.013778448104858398 2023-01-21 10:43:41.719445: step: 436/533, loss: 0.023970890790224075 2023-01-21 10:43:42.914635: step: 440/533, loss: 0.0032465935219079256 2023-01-21 10:43:44.108712: step: 444/533, loss: 0.06096343696117401 2023-01-21 10:43:45.322964: step: 448/533, loss: 0.02353210374712944 2023-01-21 10:43:46.488727: step: 452/533, loss: 0.03164968639612198 2023-01-21 10:43:47.682128: step: 456/533, loss: 0.21471329033374786 2023-01-21 10:43:48.863462: step: 460/533, loss: 0.04640321806073189 2023-01-21 10:43:50.038504: step: 464/533, loss: 0.02782764658331871 2023-01-21 10:43:51.230752: step: 468/533, loss: 0.0024102211464196444 2023-01-21 10:43:52.445381: step: 472/533, loss: 0.04049396514892578 2023-01-21 10:43:53.670033: step: 476/533, loss: 0.019942332059144974 2023-01-21 10:43:54.836834: step: 480/533, loss: 0.012668239884078503 2023-01-21 10:43:56.005875: step: 484/533, loss: 0.04583416134119034 2023-01-21 10:43:57.173700: step: 488/533, loss: 0.002342796418815851 2023-01-21 10:43:58.386314: step: 492/533, loss: 0.04883222654461861 2023-01-21 10:43:59.529762: step: 496/533, loss: 0.08726353943347931 2023-01-21 10:44:00.725246: step: 500/533, loss: 0.0261735450476408 2023-01-21 10:44:01.936769: step: 504/533, loss: 0.0060156588442623615 2023-01-21 10:44:03.133403: step: 508/533, loss: 0.004028129391372204 2023-01-21 10:44:04.294238: step: 512/533, loss: 0.00636897049844265 2023-01-21 10:44:05.462500: step: 516/533, loss: 0.019833946600556374 2023-01-21 10:44:06.629669: step: 520/533, loss: 0.01276550255715847 2023-01-21 10:44:07.833869: step: 524/533, loss: 0.026750564575195312 2023-01-21 10:44:09.023888: step: 528/533, loss: 0.12476720660924911 2023-01-21 10:44:10.203720: step: 532/533, loss: 0.0007033348083496094 2023-01-21 10:44:11.403638: step: 536/533, loss: 0.06892824172973633 2023-01-21 10:44:12.573721: step: 540/533, loss: 0.05283260717988014 2023-01-21 10:44:13.745699: step: 544/533, loss: 0.09020958095788956 2023-01-21 10:44:14.912135: step: 548/533, loss: 0.018908215686678886 2023-01-21 10:44:16.068996: step: 552/533, loss: 0.012073302641510963 2023-01-21 10:44:17.267831: step: 556/533, loss: 0.23959484696388245 2023-01-21 10:44:18.449899: step: 560/533, loss: 0.04039249196648598 2023-01-21 10:44:19.636727: step: 564/533, loss: 0.013921642675995827 2023-01-21 10:44:20.828440: step: 568/533, loss: 0.004255819600075483 2023-01-21 10:44:21.987303: step: 572/533, loss: 0.015288257971405983 2023-01-21 10:44:23.150570: step: 576/533, loss: 0.03288526460528374 2023-01-21 10:44:24.332217: step: 580/533, loss: 0.06351232528686523 2023-01-21 10:44:25.495317: step: 584/533, loss: 0.004592800047248602 2023-01-21 10:44:26.695331: step: 588/533, loss: 0.017434120178222656 2023-01-21 10:44:27.882243: step: 592/533, loss: 0.12036561965942383 2023-01-21 10:44:29.057022: step: 596/533, loss: 0.04384660720825195 2023-01-21 10:44:30.272336: step: 600/533, loss: 0.034758470952510834 2023-01-21 10:44:31.470047: step: 604/533, loss: 0.02042102813720703 2023-01-21 10:44:32.668237: step: 608/533, loss: 0.021448900923132896 2023-01-21 10:44:33.875787: step: 612/533, loss: 0.060921575874090195 2023-01-21 10:44:35.050051: step: 616/533, loss: 0.18937024474143982 2023-01-21 10:44:36.227186: step: 620/533, loss: 0.013827895745635033 2023-01-21 10:44:37.394151: step: 624/533, loss: 0.0035732269752770662 2023-01-21 10:44:38.580840: step: 628/533, loss: 0.014711190015077591 2023-01-21 10:44:39.734018: step: 632/533, loss: 0.03498706966638565 2023-01-21 10:44:40.928107: step: 636/533, loss: 0.0016311646904796362 2023-01-21 10:44:42.134100: step: 640/533, loss: 0.0909392386674881 2023-01-21 10:44:43.289581: step: 644/533, loss: 0.024828530848026276 2023-01-21 10:44:44.469246: step: 648/533, loss: 0.03652148321270943 2023-01-21 10:44:45.646956: step: 652/533, loss: 0.021696995943784714 2023-01-21 10:44:46.844869: step: 656/533, loss: 1.1339837312698364 2023-01-21 10:44:48.022706: step: 660/533, loss: 0.06812019646167755 2023-01-21 10:44:49.175862: step: 664/533, loss: 0.03695373609662056 2023-01-21 10:44:50.341895: step: 668/533, loss: 0.01607351377606392 2023-01-21 10:44:51.572726: step: 672/533, loss: 0.3977532684803009 2023-01-21 10:44:52.788640: step: 676/533, loss: 0.019930077716708183 2023-01-21 10:44:54.019134: step: 680/533, loss: 0.0021008490584790707 2023-01-21 10:44:55.216989: step: 684/533, loss: 0.0179348886013031 2023-01-21 10:44:56.377202: step: 688/533, loss: 0.030675314366817474 2023-01-21 10:44:57.556244: step: 692/533, loss: 0.0865689218044281 2023-01-21 10:44:58.756935: step: 696/533, loss: 0.037986092269420624 2023-01-21 10:44:59.951277: step: 700/533, loss: 0.04570160061120987 2023-01-21 10:45:01.111425: step: 704/533, loss: 0.05926094204187393 2023-01-21 10:45:02.294395: step: 708/533, loss: 0.01253213919699192 2023-01-21 10:45:03.528659: step: 712/533, loss: 0.017019128426909447 2023-01-21 10:45:04.690548: step: 716/533, loss: 0.0007432937854900956 2023-01-21 10:45:05.895932: step: 720/533, loss: 0.12261772155761719 2023-01-21 10:45:07.055361: step: 724/533, loss: 0.20159253478050232 2023-01-21 10:45:08.263412: step: 728/533, loss: 0.08737373352050781 2023-01-21 10:45:09.444673: step: 732/533, loss: 0.0029486059211194515 2023-01-21 10:45:10.595717: step: 736/533, loss: 0.04386758804321289 2023-01-21 10:45:11.732775: step: 740/533, loss: 0.08595972508192062 2023-01-21 10:45:12.893436: step: 744/533, loss: 0.10736212879419327 2023-01-21 10:45:14.077538: step: 748/533, loss: 0.04567518085241318 2023-01-21 10:45:15.266726: step: 752/533, loss: 0.011414623819291592 2023-01-21 10:45:16.450635: step: 756/533, loss: 0.004131627269089222 2023-01-21 10:45:17.693655: step: 760/533, loss: 0.02365722693502903 2023-01-21 10:45:18.932767: step: 764/533, loss: 0.032416343688964844 2023-01-21 10:45:20.115485: step: 768/533, loss: 0.04998932033777237 2023-01-21 10:45:21.357602: step: 772/533, loss: 0.09170132130384445 2023-01-21 10:45:22.547474: step: 776/533, loss: 0.01733417622745037 2023-01-21 10:45:23.745092: step: 780/533, loss: 0.02901754342019558 2023-01-21 10:45:24.948685: step: 784/533, loss: 0.007569694425910711 2023-01-21 10:45:26.124502: step: 788/533, loss: 0.008176279254257679 2023-01-21 10:45:27.315672: step: 792/533, loss: 0.00899195671081543 2023-01-21 10:45:28.480732: step: 796/533, loss: 0.06712212413549423 2023-01-21 10:45:29.670167: step: 800/533, loss: 0.14444255828857422 2023-01-21 10:45:30.850796: step: 804/533, loss: 0.04913497343659401 2023-01-21 10:45:32.059213: step: 808/533, loss: 0.05984792858362198 2023-01-21 10:45:33.253365: step: 812/533, loss: 0.043523404747247696 2023-01-21 10:45:34.417611: step: 816/533, loss: 0.1107281744480133 2023-01-21 10:45:35.595289: step: 820/533, loss: 0.03707904741168022 2023-01-21 10:45:36.784896: step: 824/533, loss: 0.012945270165801048 2023-01-21 10:45:37.973942: step: 828/533, loss: 0.01947174221277237 2023-01-21 10:45:39.161923: step: 832/533, loss: 0.02723522298038006 2023-01-21 10:45:40.328181: step: 836/533, loss: 0.020889664068818092 2023-01-21 10:45:41.501383: step: 840/533, loss: 0.017010213807225227 2023-01-21 10:45:42.691810: step: 844/533, loss: 0.017997123301029205 2023-01-21 10:45:43.839747: step: 848/533, loss: 0.005139636807143688 2023-01-21 10:45:45.006617: step: 852/533, loss: 0.03821602091193199 2023-01-21 10:45:46.181374: step: 856/533, loss: 0.028556011617183685 2023-01-21 10:45:47.378798: step: 860/533, loss: 0.043375395238399506 2023-01-21 10:45:48.575942: step: 864/533, loss: 0.03889598697423935 2023-01-21 10:45:49.731016: step: 868/533, loss: 0.07777624577283859 2023-01-21 10:45:50.934969: step: 872/533, loss: 0.007169913966208696 2023-01-21 10:45:52.103346: step: 876/533, loss: 0.031205464154481888 2023-01-21 10:45:53.312260: step: 880/533, loss: 0.0543147549033165 2023-01-21 10:45:54.507643: step: 884/533, loss: 0.023758603259921074 2023-01-21 10:45:55.708096: step: 888/533, loss: 0.25472497940063477 2023-01-21 10:45:56.861883: step: 892/533, loss: 0.14260028302669525 2023-01-21 10:45:58.036131: step: 896/533, loss: 0.019707489758729935 2023-01-21 10:45:59.194418: step: 900/533, loss: 0.002470600651577115 2023-01-21 10:46:00.343569: step: 904/533, loss: 0.05762834846973419 2023-01-21 10:46:01.526460: step: 908/533, loss: 0.17950496077537537 2023-01-21 10:46:02.717858: step: 912/533, loss: 0.03777436912059784 2023-01-21 10:46:03.915424: step: 916/533, loss: 0.05789833515882492 2023-01-21 10:46:05.091528: step: 920/533, loss: 0.033000778406858444 2023-01-21 10:46:06.290772: step: 924/533, loss: 0.016828252002596855 2023-01-21 10:46:07.503947: step: 928/533, loss: 0.0033605576027184725 2023-01-21 10:46:08.679067: step: 932/533, loss: 0.06819181144237518 2023-01-21 10:46:09.845938: step: 936/533, loss: 0.023807812482118607 2023-01-21 10:46:11.059966: step: 940/533, loss: 0.015148640610277653 2023-01-21 10:46:12.213103: step: 944/533, loss: 0.0016361236339434981 2023-01-21 10:46:13.422976: step: 948/533, loss: 0.035016633570194244 2023-01-21 10:46:14.604189: step: 952/533, loss: 0.01717844046652317 2023-01-21 10:46:15.818212: step: 956/533, loss: 0.10387001931667328 2023-01-21 10:46:16.983500: step: 960/533, loss: 0.014607143588364124 2023-01-21 10:46:18.134172: step: 964/533, loss: 0.02431488037109375 2023-01-21 10:46:19.339160: step: 968/533, loss: 0.017378902062773705 2023-01-21 10:46:20.515707: step: 972/533, loss: 0.027135562151670456 2023-01-21 10:46:21.701368: step: 976/533, loss: 0.009358692914247513 2023-01-21 10:46:22.870646: step: 980/533, loss: 0.06468673050403595 2023-01-21 10:46:24.043046: step: 984/533, loss: 0.06808729469776154 2023-01-21 10:46:25.236104: step: 988/533, loss: 0.006745166145265102 2023-01-21 10:46:26.404660: step: 992/533, loss: 0.004541206173598766 2023-01-21 10:46:27.572815: step: 996/533, loss: 0.008391380310058594 2023-01-21 10:46:28.735349: step: 1000/533, loss: 0.03845653682947159 2023-01-21 10:46:29.931131: step: 1004/533, loss: 0.08253125846385956 2023-01-21 10:46:31.105247: step: 1008/533, loss: 0.00016336439875885844 2023-01-21 10:46:32.335194: step: 1012/533, loss: 0.2629735767841339 2023-01-21 10:46:33.484353: step: 1016/533, loss: 0.6583120226860046 2023-01-21 10:46:34.671082: step: 1020/533, loss: 0.005272984504699707 2023-01-21 10:46:35.848337: step: 1024/533, loss: 0.024669364094734192 2023-01-21 10:46:37.096367: step: 1028/533, loss: 0.20830345153808594 2023-01-21 10:46:38.271348: step: 1032/533, loss: 0.009858131408691406 2023-01-21 10:46:39.434006: step: 1036/533, loss: 0.02657509036362171 2023-01-21 10:46:40.614157: step: 1040/533, loss: 0.018193816766142845 2023-01-21 10:46:41.763691: step: 1044/533, loss: 0.010870719328522682 2023-01-21 10:46:42.945011: step: 1048/533, loss: 0.0379091277718544 2023-01-21 10:46:44.086189: step: 1052/533, loss: 0.013233566656708717 2023-01-21 10:46:45.258085: step: 1056/533, loss: 0.04811582341790199 2023-01-21 10:46:46.430630: step: 1060/533, loss: 0.021058082580566406 2023-01-21 10:46:47.660063: step: 1064/533, loss: 0.0037311555352061987 2023-01-21 10:46:48.808522: step: 1068/533, loss: 0.03063049539923668 2023-01-21 10:46:49.967971: step: 1072/533, loss: 0.016056345775723457 2023-01-21 10:46:51.156240: step: 1076/533, loss: 0.01564941555261612 2023-01-21 10:46:52.393224: step: 1080/533, loss: 0.10672007501125336 2023-01-21 10:46:53.572956: step: 1084/533, loss: 0.03578929975628853 2023-01-21 10:46:54.768209: step: 1088/533, loss: 0.0438106544315815 2023-01-21 10:46:55.955913: step: 1092/533, loss: 0.021296977996826172 2023-01-21 10:46:57.196204: step: 1096/533, loss: 0.008884811773896217 2023-01-21 10:46:58.376655: step: 1100/533, loss: 0.07901906967163086 2023-01-21 10:46:59.540992: step: 1104/533, loss: 0.1819925308227539 2023-01-21 10:47:00.703804: step: 1108/533, loss: 0.029536008834838867 2023-01-21 10:47:01.906519: step: 1112/533, loss: 0.016701603308320045 2023-01-21 10:47:03.094875: step: 1116/533, loss: 0.0659646987915039 2023-01-21 10:47:04.264737: step: 1120/533, loss: 0.02088022232055664 2023-01-21 10:47:05.469744: step: 1124/533, loss: 0.013384056277573109 2023-01-21 10:47:06.622691: step: 1128/533, loss: 0.017646025866270065 2023-01-21 10:47:07.778776: step: 1132/533, loss: 0.046300508081912994 2023-01-21 10:47:08.907035: step: 1136/533, loss: 0.05372676998376846 2023-01-21 10:47:10.105996: step: 1140/533, loss: 0.02361116372048855 2023-01-21 10:47:11.287300: step: 1144/533, loss: 0.01811380498111248 2023-01-21 10:47:12.454223: step: 1148/533, loss: 0.0357694998383522 2023-01-21 10:47:13.622334: step: 1152/533, loss: 0.026819705963134766 2023-01-21 10:47:14.814689: step: 1156/533, loss: 0.07977437973022461 2023-01-21 10:47:15.996237: step: 1160/533, loss: 0.02685832977294922 2023-01-21 10:47:17.169269: step: 1164/533, loss: 0.004259300418198109 2023-01-21 10:47:18.343989: step: 1168/533, loss: 0.014326381497085094 2023-01-21 10:47:19.504597: step: 1172/533, loss: 0.06263842433691025 2023-01-21 10:47:20.682293: step: 1176/533, loss: 0.0031723796855658293 2023-01-21 10:47:21.841312: step: 1180/533, loss: 0.04707822948694229 2023-01-21 10:47:22.994759: step: 1184/533, loss: 0.007716464810073376 2023-01-21 10:47:24.168918: step: 1188/533, loss: 0.018033696338534355 2023-01-21 10:47:25.336511: step: 1192/533, loss: 0.15719851851463318 2023-01-21 10:47:26.509233: step: 1196/533, loss: 0.0605626106262207 2023-01-21 10:47:27.685385: step: 1200/533, loss: 0.007368660066276789 2023-01-21 10:47:28.870215: step: 1204/533, loss: 0.06865191459655762 2023-01-21 10:47:30.036733: step: 1208/533, loss: 0.035880088806152344 2023-01-21 10:47:31.238871: step: 1212/533, loss: 0.003882694523781538 2023-01-21 10:47:32.428346: step: 1216/533, loss: 0.049204446375370026 2023-01-21 10:47:33.609003: step: 1220/533, loss: 0.014983702450990677 2023-01-21 10:47:34.796251: step: 1224/533, loss: 0.06709900498390198 2023-01-21 10:47:36.013869: step: 1228/533, loss: 0.11667509377002716 2023-01-21 10:47:37.179962: step: 1232/533, loss: 0.028460979461669922 2023-01-21 10:47:38.375235: step: 1236/533, loss: 0.02120800130069256 2023-01-21 10:47:39.580838: step: 1240/533, loss: 0.034026432782411575 2023-01-21 10:47:40.752355: step: 1244/533, loss: 0.06484775245189667 2023-01-21 10:47:41.981967: step: 1248/533, loss: 0.06524619460105896 2023-01-21 10:47:43.141598: step: 1252/533, loss: 0.05959830433130264 2023-01-21 10:47:44.321625: step: 1256/533, loss: 0.04763012006878853 2023-01-21 10:47:45.469224: step: 1260/533, loss: 0.10088081657886505 2023-01-21 10:47:46.637951: step: 1264/533, loss: 0.03941183164715767 2023-01-21 10:47:47.792607: step: 1268/533, loss: 0.019593000411987305 2023-01-21 10:47:49.001019: step: 1272/533, loss: 0.165852352976799 2023-01-21 10:47:50.189804: step: 1276/533, loss: 0.13181647658348083 2023-01-21 10:47:51.406429: step: 1280/533, loss: 0.06385259330272675 2023-01-21 10:47:52.621994: step: 1284/533, loss: 0.14597482979297638 2023-01-21 10:47:53.812948: step: 1288/533, loss: 0.07788078486919403 2023-01-21 10:47:54.984751: step: 1292/533, loss: 0.05867443233728409 2023-01-21 10:47:56.144649: step: 1296/533, loss: 0.017969608306884766 2023-01-21 10:47:57.347891: step: 1300/533, loss: 0.09136438369750977 2023-01-21 10:47:58.548046: step: 1304/533, loss: 0.0328548438847065 2023-01-21 10:47:59.681413: step: 1308/533, loss: 0.04424486309289932 2023-01-21 10:48:00.841944: step: 1312/533, loss: 0.0019918440375477076 2023-01-21 10:48:02.015625: step: 1316/533, loss: 0.007418251130729914 2023-01-21 10:48:03.213760: step: 1320/533, loss: 0.03332176059484482 2023-01-21 10:48:04.381937: step: 1324/533, loss: 0.009627675637602806 2023-01-21 10:48:05.622394: step: 1328/533, loss: 0.08733449131250381 2023-01-21 10:48:06.803753: step: 1332/533, loss: 0.04447555914521217 2023-01-21 10:48:07.967944: step: 1336/533, loss: 0.06751923263072968 2023-01-21 10:48:09.169013: step: 1340/533, loss: 0.032973192632198334 2023-01-21 10:48:10.356476: step: 1344/533, loss: 0.005038452334702015 2023-01-21 10:48:11.522024: step: 1348/533, loss: 0.004536533262580633 2023-01-21 10:48:12.684982: step: 1352/533, loss: 0.058122068643569946 2023-01-21 10:48:13.898621: step: 1356/533, loss: 0.012687254697084427 2023-01-21 10:48:15.075900: step: 1360/533, loss: 0.00492329616099596 2023-01-21 10:48:16.233293: step: 1364/533, loss: 0.09077616035938263 2023-01-21 10:48:17.439804: step: 1368/533, loss: 0.005219841375946999 2023-01-21 10:48:18.604849: step: 1372/533, loss: 0.028905438259243965 2023-01-21 10:48:19.788656: step: 1376/533, loss: 0.009671163745224476 2023-01-21 10:48:20.977374: step: 1380/533, loss: 0.002448535058647394 2023-01-21 10:48:22.147687: step: 1384/533, loss: 0.13827520608901978 2023-01-21 10:48:23.360446: step: 1388/533, loss: 0.6082713007926941 2023-01-21 10:48:24.559457: step: 1392/533, loss: 0.0005320549244061112 2023-01-21 10:48:25.761494: step: 1396/533, loss: 0.032445717602968216 2023-01-21 10:48:26.934650: step: 1400/533, loss: 0.019359493628144264 2023-01-21 10:48:28.110532: step: 1404/533, loss: 0.05150227248668671 2023-01-21 10:48:29.302693: step: 1408/533, loss: 0.3610328733921051 2023-01-21 10:48:30.520018: step: 1412/533, loss: 0.17799559235572815 2023-01-21 10:48:31.731565: step: 1416/533, loss: 0.017168521881103516 2023-01-21 10:48:32.940722: step: 1420/533, loss: 0.201358363032341 2023-01-21 10:48:34.143861: step: 1424/533, loss: 0.06886868178844452 2023-01-21 10:48:35.321853: step: 1428/533, loss: 0.16228222846984863 2023-01-21 10:48:36.542393: step: 1432/533, loss: 0.021047307178378105 2023-01-21 10:48:37.726908: step: 1436/533, loss: 0.10263442993164062 2023-01-21 10:48:38.915476: step: 1440/533, loss: 0.2091541290283203 2023-01-21 10:48:40.071583: step: 1444/533, loss: 0.09670257568359375 2023-01-21 10:48:41.234972: step: 1448/533, loss: 0.052000902593135834 2023-01-21 10:48:42.394877: step: 1452/533, loss: 0.06897849589586258 2023-01-21 10:48:43.576972: step: 1456/533, loss: 0.00041856765164993703 2023-01-21 10:48:44.722985: step: 1460/533, loss: 0.03280916064977646 2023-01-21 10:48:45.889087: step: 1464/533, loss: 0.013363933190703392 2023-01-21 10:48:47.081450: step: 1468/533, loss: 0.09364490956068039 2023-01-21 10:48:48.268214: step: 1472/533, loss: 0.01431436650454998 2023-01-21 10:48:49.477399: step: 1476/533, loss: 0.005146122071892023 2023-01-21 10:48:50.676229: step: 1480/533, loss: 0.060553837567567825 2023-01-21 10:48:51.965642: step: 1484/533, loss: 0.30208471417427063 2023-01-21 10:48:53.126481: step: 1488/533, loss: 0.10384144634008408 2023-01-21 10:48:54.307662: step: 1492/533, loss: 0.03745314106345177 2023-01-21 10:48:55.484300: step: 1496/533, loss: 0.02348005771636963 2023-01-21 10:48:56.659419: step: 1500/533, loss: 0.008070326410233974 2023-01-21 10:48:57.812755: step: 1504/533, loss: 0.045976921916007996 2023-01-21 10:48:58.944486: step: 1508/533, loss: 0.06866731494665146 2023-01-21 10:49:00.164318: step: 1512/533, loss: 0.060430336743593216 2023-01-21 10:49:01.355605: step: 1516/533, loss: 0.04514675214886665 2023-01-21 10:49:02.593814: step: 1520/533, loss: 0.040711406618356705 2023-01-21 10:49:03.765482: step: 1524/533, loss: 0.01841134950518608 2023-01-21 10:49:04.949533: step: 1528/533, loss: 0.018446827307343483 2023-01-21 10:49:06.118162: step: 1532/533, loss: 0.01768198050558567 2023-01-21 10:49:07.320507: step: 1536/533, loss: 0.14561490714550018 2023-01-21 10:49:08.515228: step: 1540/533, loss: 0.01342697162181139 2023-01-21 10:49:09.698970: step: 1544/533, loss: 0.033490944653749466 2023-01-21 10:49:10.902134: step: 1548/533, loss: 0.02055206336081028 2023-01-21 10:49:12.041829: step: 1552/533, loss: 0.034474849700927734 2023-01-21 10:49:13.226829: step: 1556/533, loss: 0.07722511142492294 2023-01-21 10:49:14.416309: step: 1560/533, loss: 0.0552278533577919 2023-01-21 10:49:15.610638: step: 1564/533, loss: 0.06279754638671875 2023-01-21 10:49:16.771872: step: 1568/533, loss: 0.0029401779174804688 2023-01-21 10:49:17.944686: step: 1572/533, loss: 0.026673031970858574 2023-01-21 10:49:19.148675: step: 1576/533, loss: 0.05111350864171982 2023-01-21 10:49:20.362300: step: 1580/533, loss: 0.04513407498598099 2023-01-21 10:49:21.559058: step: 1584/533, loss: 0.018364334478974342 2023-01-21 10:49:22.768360: step: 1588/533, loss: 0.007852268405258656 2023-01-21 10:49:23.978250: step: 1592/533, loss: 0.1326128989458084 2023-01-21 10:49:25.161000: step: 1596/533, loss: 0.11506939679384232 2023-01-21 10:49:26.312250: step: 1600/533, loss: 0.07705860584974289 2023-01-21 10:49:27.482846: step: 1604/533, loss: 0.45621684193611145 2023-01-21 10:49:28.644391: step: 1608/533, loss: 0.05713510513305664 2023-01-21 10:49:29.842570: step: 1612/533, loss: 0.08268871158361435 2023-01-21 10:49:31.007582: step: 1616/533, loss: 0.010688400827348232 2023-01-21 10:49:32.172805: step: 1620/533, loss: 0.05480222776532173 2023-01-21 10:49:33.362851: step: 1624/533, loss: 0.006007385440170765 2023-01-21 10:49:34.529047: step: 1628/533, loss: 0.014450264163315296 2023-01-21 10:49:35.747993: step: 1632/533, loss: 0.08145341277122498 2023-01-21 10:49:36.935329: step: 1636/533, loss: 0.05900127440690994 2023-01-21 10:49:38.083770: step: 1640/533, loss: 0.0033389092423021793 2023-01-21 10:49:39.255232: step: 1644/533, loss: 0.043526653200387955 2023-01-21 10:49:40.422598: step: 1648/533, loss: 0.017824554815888405 2023-01-21 10:49:41.577821: step: 1652/533, loss: 0.04410190507769585 2023-01-21 10:49:42.718925: step: 1656/533, loss: 0.039089299738407135 2023-01-21 10:49:43.933451: step: 1660/533, loss: 0.7259637117385864 2023-01-21 10:49:45.130482: step: 1664/533, loss: 0.2234129011631012 2023-01-21 10:49:46.287639: step: 1668/533, loss: 0.057437799870967865 2023-01-21 10:49:47.481518: step: 1672/533, loss: 0.07039479911327362 2023-01-21 10:49:48.656604: step: 1676/533, loss: 0.00327472691424191 2023-01-21 10:49:49.909262: step: 1680/533, loss: 0.05289306864142418 2023-01-21 10:49:51.072230: step: 1684/533, loss: 0.019138097763061523 2023-01-21 10:49:52.264752: step: 1688/533, loss: 0.02441711537539959 2023-01-21 10:49:53.443732: step: 1692/533, loss: 0.028813552111387253 2023-01-21 10:49:54.597195: step: 1696/533, loss: 0.024274826049804688 2023-01-21 10:49:55.794169: step: 1700/533, loss: 0.03939366713166237 2023-01-21 10:49:56.953382: step: 1704/533, loss: 0.013880109414458275 2023-01-21 10:49:58.124223: step: 1708/533, loss: 0.018095040693879128 2023-01-21 10:49:59.290459: step: 1712/533, loss: 0.034016892313957214 2023-01-21 10:50:00.464292: step: 1716/533, loss: 0.027688028290867805 2023-01-21 10:50:01.638149: step: 1720/533, loss: 0.03873515501618385 2023-01-21 10:50:02.769891: step: 1724/533, loss: 0.049902867525815964 2023-01-21 10:50:03.994207: step: 1728/533, loss: 0.06859683990478516 2023-01-21 10:50:05.179200: step: 1732/533, loss: 0.029404640197753906 2023-01-21 10:50:06.342798: step: 1736/533, loss: 0.04838988929986954 2023-01-21 10:50:07.548734: step: 1740/533, loss: 0.06674300134181976 2023-01-21 10:50:08.710749: step: 1744/533, loss: 0.025876617059111595 2023-01-21 10:50:09.891017: step: 1748/533, loss: 0.0449770912528038 2023-01-21 10:50:11.062031: step: 1752/533, loss: 0.015346718020737171 2023-01-21 10:50:12.247215: step: 1756/533, loss: 0.10980740189552307 2023-01-21 10:50:13.411666: step: 1760/533, loss: 0.05595569685101509 2023-01-21 10:50:14.589258: step: 1764/533, loss: 0.05331411212682724 2023-01-21 10:50:15.772679: step: 1768/533, loss: 0.08244156837463379 2023-01-21 10:50:16.962758: step: 1772/533, loss: 0.06902127712965012 2023-01-21 10:50:18.109571: step: 1776/533, loss: 0.043268442153930664 2023-01-21 10:50:19.318787: step: 1780/533, loss: 0.07462511211633682 2023-01-21 10:50:20.508370: step: 1784/533, loss: 0.009383583441376686 2023-01-21 10:50:21.697480: step: 1788/533, loss: 0.11410693824291229 2023-01-21 10:50:22.921023: step: 1792/533, loss: 0.02843337133526802 2023-01-21 10:50:24.090600: step: 1796/533, loss: 0.03494281694293022 2023-01-21 10:50:25.317362: step: 1800/533, loss: 0.002321624895557761 2023-01-21 10:50:26.485773: step: 1804/533, loss: 0.03250131756067276 2023-01-21 10:50:27.649656: step: 1808/533, loss: 0.020795442163944244 2023-01-21 10:50:28.911247: step: 1812/533, loss: 0.02325439453125 2023-01-21 10:50:30.116296: step: 1816/533, loss: 0.006741237826645374 2023-01-21 10:50:31.304254: step: 1820/533, loss: 0.1205529272556305 2023-01-21 10:50:32.517827: step: 1824/533, loss: 0.11564426124095917 2023-01-21 10:50:33.708016: step: 1828/533, loss: 0.30382683873176575 2023-01-21 10:50:34.900227: step: 1832/533, loss: 0.056415751576423645 2023-01-21 10:50:36.058352: step: 1836/533, loss: 0.10739608108997345 2023-01-21 10:50:37.258189: step: 1840/533, loss: 0.0007989883306436241 2023-01-21 10:50:38.430378: step: 1844/533, loss: 0.02347707748413086 2023-01-21 10:50:39.604246: step: 1848/533, loss: 0.01653442531824112 2023-01-21 10:50:40.776640: step: 1852/533, loss: 0.07408218830823898 2023-01-21 10:50:41.893479: step: 1856/533, loss: 0.027668381109833717 2023-01-21 10:50:43.093069: step: 1860/533, loss: 0.15898758172988892 2023-01-21 10:50:44.244937: step: 1864/533, loss: 0.4568532109260559 2023-01-21 10:50:45.459558: step: 1868/533, loss: 0.00966711062937975 2023-01-21 10:50:46.636137: step: 1872/533, loss: 0.0014947891468182206 2023-01-21 10:50:47.792700: step: 1876/533, loss: 0.004918289370834827 2023-01-21 10:50:48.956967: step: 1880/533, loss: 0.04282064363360405 2023-01-21 10:50:50.121822: step: 1884/533, loss: 0.1184360459446907 2023-01-21 10:50:51.289801: step: 1888/533, loss: 0.3993797302246094 2023-01-21 10:50:52.447088: step: 1892/533, loss: 1.2934364080429077 2023-01-21 10:50:53.660166: step: 1896/533, loss: 0.09910546243190765 2023-01-21 10:50:54.824778: step: 1900/533, loss: 0.009246539324522018 2023-01-21 10:50:56.031374: step: 1904/533, loss: 0.08051595836877823 2023-01-21 10:50:57.196063: step: 1908/533, loss: 0.004052114672958851 2023-01-21 10:50:58.376282: step: 1912/533, loss: 0.006758785340934992 2023-01-21 10:50:59.557829: step: 1916/533, loss: 0.03931179270148277 2023-01-21 10:51:00.716584: step: 1920/533, loss: 0.07021927833557129 2023-01-21 10:51:01.882990: step: 1924/533, loss: 0.05391426384449005 2023-01-21 10:51:03.045216: step: 1928/533, loss: 0.020174002274870872 2023-01-21 10:51:04.222930: step: 1932/533, loss: 0.21702241897583008 2023-01-21 10:51:05.425457: step: 1936/533, loss: 0.0025521754287183285 2023-01-21 10:51:06.623104: step: 1940/533, loss: 0.04597644880414009 2023-01-21 10:51:07.824570: step: 1944/533, loss: 0.028826426714658737 2023-01-21 10:51:08.984613: step: 1948/533, loss: 0.022482728585600853 2023-01-21 10:51:10.162361: step: 1952/533, loss: 0.03686704859137535 2023-01-21 10:51:11.386022: step: 1956/533, loss: 0.05378265678882599 2023-01-21 10:51:12.529277: step: 1960/533, loss: 0.003544426057487726 2023-01-21 10:51:13.753651: step: 1964/533, loss: 0.10948581993579865 2023-01-21 10:51:14.976721: step: 1968/533, loss: 0.037540435791015625 2023-01-21 10:51:16.145431: step: 1972/533, loss: 0.0022695541847497225 2023-01-21 10:51:17.323322: step: 1976/533, loss: 0.056593991816043854 2023-01-21 10:51:18.508186: step: 1980/533, loss: 0.050704099237918854 2023-01-21 10:51:19.707088: step: 1984/533, loss: 0.033888839185237885 2023-01-21 10:51:20.895662: step: 1988/533, loss: 0.003427839372307062 2023-01-21 10:51:22.096066: step: 1992/533, loss: 0.035056211054325104 2023-01-21 10:51:23.260500: step: 1996/533, loss: 0.025683116167783737 2023-01-21 10:51:24.434779: step: 2000/533, loss: 0.049903109669685364 2023-01-21 10:51:25.630575: step: 2004/533, loss: 0.024570655077695847 2023-01-21 10:51:26.897030: step: 2008/533, loss: 0.05321057140827179 2023-01-21 10:51:28.086790: step: 2012/533, loss: 0.05623655766248703 2023-01-21 10:51:29.294581: step: 2016/533, loss: 0.03146815299987793 2023-01-21 10:51:30.474239: step: 2020/533, loss: 0.07843560725450516 2023-01-21 10:51:31.677140: step: 2024/533, loss: 0.0027111531235277653 2023-01-21 10:51:32.897860: step: 2028/533, loss: 0.004670906346291304 2023-01-21 10:51:34.058427: step: 2032/533, loss: 0.09960909187793732 2023-01-21 10:51:35.227067: step: 2036/533, loss: 0.4701419472694397 2023-01-21 10:51:36.471706: step: 2040/533, loss: 0.04217939451336861 2023-01-21 10:51:37.677251: step: 2044/533, loss: 0.08248773217201233 2023-01-21 10:51:38.902413: step: 2048/533, loss: 0.05019540712237358 2023-01-21 10:51:40.058084: step: 2052/533, loss: 0.030640697106719017 2023-01-21 10:51:41.232012: step: 2056/533, loss: 0.02095479890704155 2023-01-21 10:51:42.406210: step: 2060/533, loss: 0.02147965505719185 2023-01-21 10:51:43.625172: step: 2064/533, loss: 0.01578063890337944 2023-01-21 10:51:44.832873: step: 2068/533, loss: 0.04016117751598358 2023-01-21 10:51:46.005971: step: 2072/533, loss: 0.08902301639318466 2023-01-21 10:51:47.177078: step: 2076/533, loss: 0.033548880368471146 2023-01-21 10:51:48.345022: step: 2080/533, loss: 0.08504059165716171 2023-01-21 10:51:49.545453: step: 2084/533, loss: 0.06880378723144531 2023-01-21 10:51:50.745460: step: 2088/533, loss: 0.09760390222072601 2023-01-21 10:51:51.935728: step: 2092/533, loss: 0.019560767337679863 2023-01-21 10:51:53.085026: step: 2096/533, loss: 1.6983377933502197 2023-01-21 10:51:54.267446: step: 2100/533, loss: 0.046353720128536224 2023-01-21 10:51:55.420818: step: 2104/533, loss: 0.03412804752588272 2023-01-21 10:51:56.572928: step: 2108/533, loss: 0.0052394866943359375 2023-01-21 10:51:57.735517: step: 2112/533, loss: 0.05005950853228569 2023-01-21 10:51:58.933877: step: 2116/533, loss: 0.023495342582464218 2023-01-21 10:52:00.095770: step: 2120/533, loss: 0.015263080596923828 2023-01-21 10:52:01.291074: step: 2124/533, loss: 0.03577251732349396 2023-01-21 10:52:02.449073: step: 2128/533, loss: 0.28675755858421326 2023-01-21 10:52:03.611764: step: 2132/533, loss: 0.028914406895637512 ================================================== Loss: 0.064 -------------------- Dev: {'event': {'p': 0.5838206627680312, 'r': 0.7976031957390146, 'f1': 0.6741699493528419}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Test: {'event': {'p': 0.6255506607929515, 'r': 0.8171355498721228, 'f1': 0.7086221236484613}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} 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': 12} Korean: {'event': {'p': 0.5901639344262295, 'r': 0.5714285714285714, 'f1': 0.5806451612903226}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Russian: {'event': {'p': 0.4090909090909091, 'r': 0.5, 'f1': 0.45}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 10:52:45.644049: step: 4/533, loss: 0.02113480493426323 2023-01-21 10:52:46.816128: step: 8/533, loss: 0.2180517166852951 2023-01-21 10:52:47.989743: step: 12/533, loss: 0.04204254224896431 2023-01-21 10:52:49.186996: step: 16/533, loss: 0.05729112774133682 2023-01-21 10:52:50.349606: step: 20/533, loss: 0.011809063144028187 2023-01-21 10:52:51.513839: step: 24/533, loss: 0.09803657233715057 2023-01-21 10:52:52.738398: step: 28/533, loss: 0.005874681286513805 2023-01-21 10:52:53.877594: step: 32/533, loss: 0.03490648418664932 2023-01-21 10:52:55.107484: step: 36/533, loss: 0.022076226770877838 2023-01-21 10:52:56.281100: step: 40/533, loss: 0.04399728775024414 2023-01-21 10:52:57.426277: step: 44/533, loss: 0.0050290110521018505 2023-01-21 10:52:58.641558: step: 48/533, loss: 0.13323011994361877 2023-01-21 10:52:59.818154: step: 52/533, loss: 0.010277272202074528 2023-01-21 10:53:01.015987: step: 56/533, loss: 0.03768301010131836 2023-01-21 10:53:02.175205: step: 60/533, loss: 0.04535641893744469 2023-01-21 10:53:03.386206: step: 64/533, loss: 0.009997940622270107 2023-01-21 10:53:04.600640: step: 68/533, loss: 0.015745162963867188 2023-01-21 10:53:05.800665: step: 72/533, loss: 0.08374081552028656 2023-01-21 10:53:06.976124: step: 76/533, loss: 0.07598571479320526 2023-01-21 10:53:08.158949: step: 80/533, loss: 0.002165889833122492 2023-01-21 10:53:09.334962: step: 84/533, loss: 0.34626901149749756 2023-01-21 10:53:10.517052: step: 88/533, loss: 0.01954049989581108 2023-01-21 10:53:11.690859: step: 92/533, loss: 0.0006309509044513106 2023-01-21 10:53:12.843643: step: 96/533, loss: 0.011398744769394398 2023-01-21 10:53:14.057419: step: 100/533, loss: 0.053518008440732956 2023-01-21 10:53:15.261315: step: 104/533, loss: 0.0004321098676882684 2023-01-21 10:53:16.446888: step: 108/533, loss: 0.0034241201356053352 2023-01-21 10:53:17.663523: step: 112/533, loss: 0.014015293680131435 2023-01-21 10:53:18.806682: step: 116/533, loss: 0.03946514055132866 2023-01-21 10:53:19.974277: step: 120/533, loss: 0.0601312629878521 2023-01-21 10:53:21.164734: step: 124/533, loss: 0.012287044897675514 2023-01-21 10:53:22.339947: step: 128/533, loss: 0.10448139160871506 2023-01-21 10:53:23.505758: step: 132/533, loss: 0.05132708698511124 2023-01-21 10:53:24.663418: step: 136/533, loss: 0.027512740343809128 2023-01-21 10:53:25.935564: step: 140/533, loss: 0.041411686688661575 2023-01-21 10:53:27.087186: step: 144/533, loss: 0.00450897216796875 2023-01-21 10:53:28.306406: step: 148/533, loss: 0.018312646076083183 2023-01-21 10:53:29.480233: step: 152/533, loss: 0.08904209733009338 2023-01-21 10:53:30.639323: step: 156/533, loss: 0.00920324306935072 2023-01-21 10:53:31.846049: step: 160/533, loss: 0.01207885704934597 2023-01-21 10:53:33.038867: step: 164/533, loss: 0.029880523681640625 2023-01-21 10:53:34.254284: step: 168/533, loss: 0.011577606201171875 2023-01-21 10:53:35.471374: step: 172/533, loss: 0.12033109366893768 2023-01-21 10:53:36.663821: step: 176/533, loss: 0.049056150019168854 2023-01-21 10:53:37.845380: step: 180/533, loss: 0.09324140846729279 2023-01-21 10:53:39.030718: step: 184/533, loss: 0.04524211958050728 2023-01-21 10:53:40.242932: step: 188/533, loss: 0.006444644648581743 2023-01-21 10:53:41.460157: step: 192/533, loss: 0.04369932785630226 2023-01-21 10:53:42.642075: step: 196/533, loss: 0.021640397608280182 2023-01-21 10:53:43.839782: step: 200/533, loss: 0.08095663785934448 2023-01-21 10:53:45.057232: step: 204/533, loss: 0.0007752418168820441 2023-01-21 10:53:46.221082: step: 208/533, loss: 0.019136715680360794 2023-01-21 10:53:47.441288: step: 212/533, loss: 0.06638136506080627 2023-01-21 10:53:48.645422: step: 216/533, loss: 0.041887760162353516 2023-01-21 10:53:49.812951: step: 220/533, loss: 0.0011137962574139237 2023-01-21 10:53:50.973814: step: 224/533, loss: 0.05277423933148384 2023-01-21 10:53:52.176151: step: 228/533, loss: 0.014643478207290173 2023-01-21 10:53:53.344973: step: 232/533, loss: 0.00011196137347724289 2023-01-21 10:53:54.536757: step: 236/533, loss: 0.02535724639892578 2023-01-21 10:53:55.728663: step: 240/533, loss: 0.014371919445693493 2023-01-21 10:53:56.900828: step: 244/533, loss: 0.01745319366455078 2023-01-21 10:53:58.090785: step: 248/533, loss: 0.04930629953742027 2023-01-21 10:53:59.296428: step: 252/533, loss: 0.04411773756146431 2023-01-21 10:54:00.463104: step: 256/533, loss: 0.05849814414978027 2023-01-21 10:54:01.668776: step: 260/533, loss: 0.063349150121212 2023-01-21 10:54:02.847824: step: 264/533, loss: 0.02587909810245037 2023-01-21 10:54:04.046550: step: 268/533, loss: 0.000789642333984375 2023-01-21 10:54:05.216498: step: 272/533, loss: 0.0025743008591234684 2023-01-21 10:54:06.386381: step: 276/533, loss: 0.04830064997076988 2023-01-21 10:54:07.609152: step: 280/533, loss: 0.13837222754955292 2023-01-21 10:54:08.783959: step: 284/533, loss: 0.027942851185798645 2023-01-21 10:54:09.967335: step: 288/533, loss: 0.0027311325538903475 2023-01-21 10:54:11.160787: step: 292/533, loss: 0.03370227664709091 2023-01-21 10:54:12.350593: step: 296/533, loss: 0.09649143368005753 2023-01-21 10:54:13.497653: step: 300/533, loss: 0.03746652230620384 2023-01-21 10:54:14.685973: step: 304/533, loss: 0.0020686150528490543 2023-01-21 10:54:15.872896: step: 308/533, loss: 0.03869514539837837 2023-01-21 10:54:16.989795: step: 312/533, loss: 0.04442758858203888 2023-01-21 10:54:18.235991: step: 316/533, loss: 0.011024285107851028 2023-01-21 10:54:19.371144: step: 320/533, loss: 0.04921398311853409 2023-01-21 10:54:20.513165: step: 324/533, loss: 0.007674884516745806 2023-01-21 10:54:21.728339: step: 328/533, loss: 0.05484924465417862 2023-01-21 10:54:22.888600: step: 332/533, loss: 0.005054187960922718 2023-01-21 10:54:24.014493: step: 336/533, loss: 0.0066925049759447575 2023-01-21 10:54:25.233780: step: 340/533, loss: 0.30472761392593384 2023-01-21 10:54:26.417053: step: 344/533, loss: 0.06115546450018883 2023-01-21 10:54:27.557854: step: 348/533, loss: 0.0355132594704628 2023-01-21 10:54:28.724732: step: 352/533, loss: 0.0890776664018631 2023-01-21 10:54:29.913905: step: 356/533, loss: 0.04704098775982857 2023-01-21 10:54:31.128155: step: 360/533, loss: 0.04420604556798935 2023-01-21 10:54:32.319556: step: 364/533, loss: 0.019759131595492363 2023-01-21 10:54:33.498624: step: 368/533, loss: 0.021140575408935547 2023-01-21 10:54:34.641200: step: 372/533, loss: 0.006780624855309725 2023-01-21 10:54:35.827395: step: 376/533, loss: 0.003692054655402899 2023-01-21 10:54:36.993248: step: 380/533, loss: 0.00440139789134264 2023-01-21 10:54:38.174472: step: 384/533, loss: 0.07288307696580887 2023-01-21 10:54:39.330289: step: 388/533, loss: 0.010738945566117764 2023-01-21 10:54:40.492784: step: 392/533, loss: 0.027880383655428886 2023-01-21 10:54:41.679740: step: 396/533, loss: 0.05134258419275284 2023-01-21 10:54:42.874253: step: 400/533, loss: 0.0022987842094153166 2023-01-21 10:54:44.024276: step: 404/533, loss: 0.005730914883315563 2023-01-21 10:54:45.249386: step: 408/533, loss: 0.07759633660316467 2023-01-21 10:54:46.467113: step: 412/533, loss: 0.005066728685051203 2023-01-21 10:54:47.663771: step: 416/533, loss: 0.020241450518369675 2023-01-21 10:54:48.892910: step: 420/533, loss: 0.02721252292394638 2023-01-21 10:54:50.129099: step: 424/533, loss: 0.020453788340091705 2023-01-21 10:54:51.276232: step: 428/533, loss: 0.029588891193270683 2023-01-21 10:54:52.449352: step: 432/533, loss: 0.0313752181828022 2023-01-21 10:54:53.633179: step: 436/533, loss: 0.025255393236875534 2023-01-21 10:54:54.891331: step: 440/533, loss: 0.22620391845703125 2023-01-21 10:54:56.035415: step: 444/533, loss: 0.001817131182178855 2023-01-21 10:54:57.195310: step: 448/533, loss: 0.033921051770448685 2023-01-21 10:54:58.349155: step: 452/533, loss: 0.700085461139679 2023-01-21 10:54:59.506904: step: 456/533, loss: 0.02322845533490181 2023-01-21 10:55:00.692437: step: 460/533, loss: 0.05259857326745987 2023-01-21 10:55:01.880411: step: 464/533, loss: 0.08695525676012039 2023-01-21 10:55:03.072030: step: 468/533, loss: 0.0037970542907714844 2023-01-21 10:55:04.265969: step: 472/533, loss: 0.018614768981933594 2023-01-21 10:55:05.453921: step: 476/533, loss: 0.08401613682508469 2023-01-21 10:55:06.648420: step: 480/533, loss: 0.00024237632169388235 2023-01-21 10:55:07.830995: step: 484/533, loss: 0.07679080963134766 2023-01-21 10:55:09.012829: step: 488/533, loss: 0.05887727811932564 2023-01-21 10:55:10.222057: step: 492/533, loss: 0.01378641091287136 2023-01-21 10:55:11.392074: step: 496/533, loss: 0.03404855728149414 2023-01-21 10:55:12.535561: step: 500/533, loss: 0.008977937512099743 2023-01-21 10:55:13.696366: step: 504/533, loss: 0.054403115063905716 2023-01-21 10:55:14.867328: step: 508/533, loss: 0.049040794372558594 2023-01-21 10:55:16.080485: step: 512/533, loss: 0.12543697655200958 2023-01-21 10:55:17.289681: step: 516/533, loss: 0.0028532028663903475 2023-01-21 10:55:18.419457: step: 520/533, loss: 0.04877519607543945 2023-01-21 10:55:19.598686: step: 524/533, loss: 0.020182037726044655 2023-01-21 10:55:20.737385: step: 528/533, loss: 0.038889121264219284 2023-01-21 10:55:21.949276: step: 532/533, loss: 0.007474326994270086 2023-01-21 10:55:23.086411: step: 536/533, loss: 0.0029264481272548437 2023-01-21 10:55:24.291170: step: 540/533, loss: 0.013007258996367455 2023-01-21 10:55:25.485957: step: 544/533, loss: 0.03889331966638565 2023-01-21 10:55:26.649050: step: 548/533, loss: 0.010523987002670765 2023-01-21 10:55:27.844044: step: 552/533, loss: 0.045534897595644 2023-01-21 10:55:29.029723: step: 556/533, loss: 0.0723661407828331 2023-01-21 10:55:30.237655: step: 560/533, loss: 0.07396040111780167 2023-01-21 10:55:31.390358: step: 564/533, loss: 0.10192445665597916 2023-01-21 10:55:32.542028: step: 568/533, loss: 0.07721743732690811 2023-01-21 10:55:33.750914: step: 572/533, loss: 0.03134002909064293 2023-01-21 10:55:34.927105: step: 576/533, loss: 0.004663276951760054 2023-01-21 10:55:36.136349: step: 580/533, loss: 0.07088317722082138 2023-01-21 10:55:37.363985: step: 584/533, loss: 0.03112011030316353 2023-01-21 10:55:38.571927: step: 588/533, loss: 0.1224648505449295 2023-01-21 10:55:39.782298: step: 592/533, loss: 0.14820709824562073 2023-01-21 10:55:40.942596: step: 596/533, loss: 0.0059379576705396175 2023-01-21 10:55:42.115111: step: 600/533, loss: 0.01222152728587389 2023-01-21 10:55:43.285983: step: 604/533, loss: 0.033492375165224075 2023-01-21 10:55:44.483741: step: 608/533, loss: 0.020554065704345703 2023-01-21 10:55:45.676304: step: 612/533, loss: 0.004041099455207586 2023-01-21 10:55:46.885965: step: 616/533, loss: 0.025343704968690872 2023-01-21 10:55:48.072698: step: 620/533, loss: 0.11975374817848206 2023-01-21 10:55:49.231005: step: 624/533, loss: 0.06233959645032883 2023-01-21 10:55:50.421728: step: 628/533, loss: 0.019565487280488014 2023-01-21 10:55:51.598720: step: 632/533, loss: 0.015001583844423294 2023-01-21 10:55:52.805246: step: 636/533, loss: 0.1642434149980545 2023-01-21 10:55:53.994410: step: 640/533, loss: 0.028392791748046875 2023-01-21 10:55:55.174577: step: 644/533, loss: 0.04311332851648331 2023-01-21 10:55:56.357986: step: 648/533, loss: 0.006820689886808395 2023-01-21 10:55:57.573976: step: 652/533, loss: 0.021359825506806374 2023-01-21 10:55:58.744683: step: 656/533, loss: 0.039544109255075455 2023-01-21 10:55:59.919317: step: 660/533, loss: 0.00016994477482512593 2023-01-21 10:56:01.104496: step: 664/533, loss: 0.0352630615234375 2023-01-21 10:56:02.301258: step: 668/533, loss: 0.035451509058475494 2023-01-21 10:56:03.506459: step: 672/533, loss: 0.08077006042003632 2023-01-21 10:56:04.713178: step: 676/533, loss: 0.06459636241197586 2023-01-21 10:56:05.885144: step: 680/533, loss: 0.013711739331483841 2023-01-21 10:56:07.054601: step: 684/533, loss: 0.03627767786383629 2023-01-21 10:56:08.209036: step: 688/533, loss: 0.03703594207763672 2023-01-21 10:56:09.367251: step: 692/533, loss: 0.0052512651309370995 2023-01-21 10:56:10.557136: step: 696/533, loss: 0.03257384151220322 2023-01-21 10:56:11.764298: step: 700/533, loss: 0.035390663892030716 2023-01-21 10:56:12.945645: step: 704/533, loss: 0.005215453915297985 2023-01-21 10:56:14.149772: step: 708/533, loss: 0.00872497633099556 2023-01-21 10:56:15.377362: step: 712/533, loss: 0.023395873606204987 2023-01-21 10:56:16.537353: step: 716/533, loss: 0.0072959898971021175 2023-01-21 10:56:17.742658: step: 720/533, loss: 0.029417896643280983 2023-01-21 10:56:18.919667: step: 724/533, loss: 0.03926267474889755 2023-01-21 10:56:20.101445: step: 728/533, loss: 0.01951618120074272 2023-01-21 10:56:21.283691: step: 732/533, loss: 0.00012855530076194555 2023-01-21 10:56:22.422766: step: 736/533, loss: 0.018889330327510834 2023-01-21 10:56:23.656407: step: 740/533, loss: 0.03725109249353409 2023-01-21 10:56:24.795067: step: 744/533, loss: 0.043996550142765045 2023-01-21 10:56:25.992299: step: 748/533, loss: 0.10337982326745987 2023-01-21 10:56:27.145901: step: 752/533, loss: 0.12036962807178497 2023-01-21 10:56:28.343392: step: 756/533, loss: 0.04981708526611328 2023-01-21 10:56:29.537833: step: 760/533, loss: 0.00131816859357059 2023-01-21 10:56:30.776267: step: 764/533, loss: 0.07454271614551544 2023-01-21 10:56:31.993159: step: 768/533, loss: 0.0735284835100174 2023-01-21 10:56:33.168622: step: 772/533, loss: 0.07411356270313263 2023-01-21 10:56:34.363972: step: 776/533, loss: 0.02219703048467636 2023-01-21 10:56:35.541164: step: 780/533, loss: 0.057050369679927826 2023-01-21 10:56:36.717432: step: 784/533, loss: 0.059995558112859726 2023-01-21 10:56:37.916864: step: 788/533, loss: 0.02738466113805771 2023-01-21 10:56:39.115905: step: 792/533, loss: 0.004936409182846546 2023-01-21 10:56:40.316509: step: 796/533, loss: 0.037609003484249115 2023-01-21 10:56:41.483621: step: 800/533, loss: 0.05502338707447052 2023-01-21 10:56:42.665238: step: 804/533, loss: 0.056107472628355026 2023-01-21 10:56:43.873584: step: 808/533, loss: 0.0258181095123291 2023-01-21 10:56:45.053373: step: 812/533, loss: 0.010819625109434128 2023-01-21 10:56:46.227277: step: 816/533, loss: 0.03242502361536026 2023-01-21 10:56:47.403531: step: 820/533, loss: 0.08075661957263947 2023-01-21 10:56:48.561324: step: 824/533, loss: 0.05600786209106445 2023-01-21 10:56:49.761043: step: 828/533, loss: 0.02236041985452175 2023-01-21 10:56:50.917146: step: 832/533, loss: 0.0878625437617302 2023-01-21 10:56:52.110312: step: 836/533, loss: 0.07311764359474182 2023-01-21 10:56:53.340095: step: 840/533, loss: 0.04112996906042099 2023-01-21 10:56:54.518270: step: 844/533, loss: 0.02455897442996502 2023-01-21 10:56:55.699637: step: 848/533, loss: 0.041886892169713974 2023-01-21 10:56:56.874981: step: 852/533, loss: 0.01795930787920952 2023-01-21 10:56:58.050774: step: 856/533, loss: 0.027780722826719284 2023-01-21 10:56:59.239129: step: 860/533, loss: 0.05400047451257706 2023-01-21 10:57:00.456532: step: 864/533, loss: 0.0428224578499794 2023-01-21 10:57:01.639772: step: 868/533, loss: 0.019554710015654564 2023-01-21 10:57:02.861759: step: 872/533, loss: 0.011345005594193935 2023-01-21 10:57:04.049054: step: 876/533, loss: 0.34164875745773315 2023-01-21 10:57:05.225809: step: 880/533, loss: 0.004983139224350452 2023-01-21 10:57:06.409812: step: 884/533, loss: 0.03833484649658203 2023-01-21 10:57:07.572613: step: 888/533, loss: 0.006659126374870539 2023-01-21 10:57:08.749220: step: 892/533, loss: 0.020348072052001953 2023-01-21 10:57:09.954434: step: 896/533, loss: 0.0044612884521484375 2023-01-21 10:57:11.119281: step: 900/533, loss: 0.04254169389605522 2023-01-21 10:57:12.307831: step: 904/533, loss: 0.00836181640625 2023-01-21 10:57:13.513868: step: 908/533, loss: 0.03066062740981579 2023-01-21 10:57:14.656377: step: 912/533, loss: 0.026505853980779648 2023-01-21 10:57:15.841090: step: 916/533, loss: 0.005989265628159046 2023-01-21 10:57:17.049566: step: 920/533, loss: 0.027514267712831497 2023-01-21 10:57:18.232972: step: 924/533, loss: 0.08614940941333771 2023-01-21 10:57:19.438876: step: 928/533, loss: 0.007572269532829523 2023-01-21 10:57:20.603686: step: 932/533, loss: 0.03605690225958824 2023-01-21 10:57:21.742314: step: 936/533, loss: 0.026875877752900124 2023-01-21 10:57:22.913493: step: 940/533, loss: 0.02230405993759632 2023-01-21 10:57:24.071519: step: 944/533, loss: 0.02019815519452095 2023-01-21 10:57:25.255262: step: 948/533, loss: 0.02164745330810547 2023-01-21 10:57:26.413795: step: 952/533, loss: 0.012897110544145107 2023-01-21 10:57:27.604008: step: 956/533, loss: 0.09667067974805832 2023-01-21 10:57:28.766200: step: 960/533, loss: 0.013701343908905983 2023-01-21 10:57:29.947939: step: 964/533, loss: 0.007405811455100775 2023-01-21 10:57:31.121608: step: 968/533, loss: 0.012143898755311966 2023-01-21 10:57:32.331883: step: 972/533, loss: 0.0043468475341796875 2023-01-21 10:57:33.503709: step: 976/533, loss: 0.002060031984001398 2023-01-21 10:57:34.693343: step: 980/533, loss: 0.0745355635881424 2023-01-21 10:57:35.886991: step: 984/533, loss: 0.0032449723221361637 2023-01-21 10:57:37.090021: step: 988/533, loss: 0.040094755589962006 2023-01-21 10:57:38.257796: step: 992/533, loss: 0.27566835284233093 2023-01-21 10:57:39.416661: step: 996/533, loss: 0.08316078037023544 2023-01-21 10:57:40.589677: step: 1000/533, loss: 0.052486710250377655 2023-01-21 10:57:41.740149: step: 1004/533, loss: 0.011968517675995827 2023-01-21 10:57:42.916326: step: 1008/533, loss: 0.08449907600879669 2023-01-21 10:57:44.066245: step: 1012/533, loss: 0.07400409132242203 2023-01-21 10:57:45.271100: step: 1016/533, loss: 0.07389393448829651 2023-01-21 10:57:46.473670: step: 1020/533, loss: 0.020061874762177467 2023-01-21 10:57:47.655455: step: 1024/533, loss: 0.045684244483709335 2023-01-21 10:57:48.798908: step: 1028/533, loss: 0.007948494516313076 2023-01-21 10:57:49.955927: step: 1032/533, loss: 0.0016446113586425781 2023-01-21 10:57:51.192202: step: 1036/533, loss: 0.02089214324951172 2023-01-21 10:57:52.350421: step: 1040/533, loss: 0.05728769302368164 2023-01-21 10:57:53.516854: step: 1044/533, loss: 0.004139709752053022 2023-01-21 10:57:54.723411: step: 1048/533, loss: 0.6455539464950562 2023-01-21 10:57:55.903207: step: 1052/533, loss: 0.12276172637939453 2023-01-21 10:57:57.092410: step: 1056/533, loss: 0.43716782331466675 2023-01-21 10:57:58.270708: step: 1060/533, loss: 0.06069774925708771 2023-01-21 10:57:59.436490: step: 1064/533, loss: 0.026159286499023438 2023-01-21 10:58:00.581312: step: 1068/533, loss: 4.129410081077367e-05 2023-01-21 10:58:01.762160: step: 1072/533, loss: 0.054787732660770416 2023-01-21 10:58:02.946393: step: 1076/533, loss: 0.04379386827349663 2023-01-21 10:58:04.087838: step: 1080/533, loss: 0.03880128636956215 2023-01-21 10:58:05.245908: step: 1084/533, loss: 0.04727020487189293 2023-01-21 10:58:06.415794: step: 1088/533, loss: 0.06523828953504562 2023-01-21 10:58:07.586985: step: 1092/533, loss: 0.03663768991827965 2023-01-21 10:58:08.757794: step: 1096/533, loss: 0.0030756474006921053 2023-01-21 10:58:09.908608: step: 1100/533, loss: 0.022153282538056374 2023-01-21 10:58:11.139492: step: 1104/533, loss: 0.026297569274902344 2023-01-21 10:58:12.295953: step: 1108/533, loss: 0.06271553039550781 2023-01-21 10:58:13.484136: step: 1112/533, loss: 0.03436391428112984 2023-01-21 10:58:14.639203: step: 1116/533, loss: 0.07579460740089417 2023-01-21 10:58:15.814301: step: 1120/533, loss: 0.06359796226024628 2023-01-21 10:58:16.993709: step: 1124/533, loss: 0.040679167956113815 2023-01-21 10:58:18.187703: step: 1128/533, loss: 0.029019545763731003 2023-01-21 10:58:19.382751: step: 1132/533, loss: 0.036148551851511 2023-01-21 10:58:20.607041: step: 1136/533, loss: 0.005886936094611883 2023-01-21 10:58:21.776045: step: 1140/533, loss: 0.006011772435158491 2023-01-21 10:58:22.954857: step: 1144/533, loss: 0.007932091131806374 2023-01-21 10:58:24.119642: step: 1148/533, loss: 0.003106242511421442 2023-01-21 10:58:25.341625: step: 1152/533, loss: 0.05227470397949219 2023-01-21 10:58:26.521768: step: 1156/533, loss: 0.018625259399414062 2023-01-21 10:58:27.724383: step: 1160/533, loss: 0.04417724907398224 2023-01-21 10:58:28.949190: step: 1164/533, loss: 0.09038429707288742 2023-01-21 10:58:30.157358: step: 1168/533, loss: 0.011069584637880325 2023-01-21 10:58:31.325796: step: 1172/533, loss: 0.012411308474838734 2023-01-21 10:58:32.489889: step: 1176/533, loss: 0.03125328943133354 2023-01-21 10:58:33.695548: step: 1180/533, loss: 0.007350158877670765 2023-01-21 10:58:34.901367: step: 1184/533, loss: 0.0676090270280838 2023-01-21 10:58:36.067946: step: 1188/533, loss: 0.04961109161376953 2023-01-21 10:58:37.240412: step: 1192/533, loss: 0.011720085516571999 2023-01-21 10:58:38.398189: step: 1196/533, loss: 0.02475709840655327 2023-01-21 10:58:39.545783: step: 1200/533, loss: 0.03511079400777817 2023-01-21 10:58:40.741823: step: 1204/533, loss: 0.006617927458137274 2023-01-21 10:58:41.938586: step: 1208/533, loss: 0.006246662233024836 2023-01-21 10:58:43.126591: step: 1212/533, loss: 0.0549495704472065 2023-01-21 10:58:44.308182: step: 1216/533, loss: 0.019649125635623932 2023-01-21 10:58:45.502368: step: 1220/533, loss: 0.007434082217514515 2023-01-21 10:58:46.740994: step: 1224/533, loss: 0.026411820203065872 2023-01-21 10:58:47.913374: step: 1228/533, loss: 0.06571540981531143 2023-01-21 10:58:49.155115: step: 1232/533, loss: 0.05336766317486763 2023-01-21 10:58:50.366515: step: 1236/533, loss: 0.0635339766740799 2023-01-21 10:58:51.554151: step: 1240/533, loss: 0.025406312197446823 2023-01-21 10:58:52.733008: step: 1244/533, loss: 0.0009971618419513106 2023-01-21 10:58:53.921028: step: 1248/533, loss: 0.055283788591623306 2023-01-21 10:58:55.096545: step: 1252/533, loss: 0.014269554056227207 2023-01-21 10:58:56.232188: step: 1256/533, loss: 0.0013463019859045744 2023-01-21 10:58:57.433133: step: 1260/533, loss: 0.008423805236816406 2023-01-21 10:58:58.604257: step: 1264/533, loss: 0.004395103547722101 2023-01-21 10:58:59.785659: step: 1268/533, loss: 0.08881063759326935 2023-01-21 10:59:00.943358: step: 1272/533, loss: 0.032858848571777344 2023-01-21 10:59:02.138148: step: 1276/533, loss: 0.045183755457401276 2023-01-21 10:59:03.324866: step: 1280/533, loss: 0.10130295902490616 2023-01-21 10:59:04.522139: step: 1284/533, loss: 0.0013909817207604647 2023-01-21 10:59:05.798479: step: 1288/533, loss: 0.010302829556167126 2023-01-21 10:59:06.967347: step: 1292/533, loss: 0.05099201574921608 2023-01-21 10:59:08.187333: step: 1296/533, loss: 0.1515444815158844 2023-01-21 10:59:09.355268: step: 1300/533, loss: 0.013141632080078125 2023-01-21 10:59:10.549604: step: 1304/533, loss: 0.0005883693811483681 2023-01-21 10:59:11.703065: step: 1308/533, loss: 0.01961502991616726 2023-01-21 10:59:12.886851: step: 1312/533, loss: 1.2897058725357056 2023-01-21 10:59:14.084983: step: 1316/533, loss: 0.008539343252778053 2023-01-21 10:59:15.257797: step: 1320/533, loss: 0.08233537524938583 2023-01-21 10:59:16.434331: step: 1324/533, loss: 0.0383848212659359 2023-01-21 10:59:17.618937: step: 1328/533, loss: 0.10562209784984589 2023-01-21 10:59:18.845022: step: 1332/533, loss: 0.017515946179628372 2023-01-21 10:59:20.006995: step: 1336/533, loss: 0.007807398214936256 2023-01-21 10:59:21.204932: step: 1340/533, loss: 0.040183451026678085 2023-01-21 10:59:22.382305: step: 1344/533, loss: 0.02508239820599556 2023-01-21 10:59:23.578282: step: 1348/533, loss: 0.02649688720703125 2023-01-21 10:59:24.756843: step: 1352/533, loss: 0.00017156600370071828 2023-01-21 10:59:25.961543: step: 1356/533, loss: 0.12630033493041992 2023-01-21 10:59:27.123832: step: 1360/533, loss: 0.05860919877886772 2023-01-21 10:59:28.338492: step: 1364/533, loss: 1.308081865310669 2023-01-21 10:59:29.568489: step: 1368/533, loss: 0.02837085723876953 2023-01-21 10:59:30.742230: step: 1372/533, loss: 0.053876303136348724 2023-01-21 10:59:31.891140: step: 1376/533, loss: 0.022728348150849342 2023-01-21 10:59:33.064559: step: 1380/533, loss: 0.024346735328435898 2023-01-21 10:59:34.231482: step: 1384/533, loss: 0.0023559569381177425 2023-01-21 10:59:35.377837: step: 1388/533, loss: 0.20595036447048187 2023-01-21 10:59:36.606808: step: 1392/533, loss: 0.0008267402881756425 2023-01-21 10:59:37.807526: step: 1396/533, loss: 0.04612855985760689 2023-01-21 10:59:39.008450: step: 1400/533, loss: 0.008933735080063343 2023-01-21 10:59:40.203519: step: 1404/533, loss: 0.0012525557540357113 2023-01-21 10:59:41.399470: step: 1408/533, loss: 0.015462398529052734 2023-01-21 10:59:42.596096: step: 1412/533, loss: 0.003404903458431363 2023-01-21 10:59:43.785641: step: 1416/533, loss: 0.050734709948301315 2023-01-21 10:59:44.986515: step: 1420/533, loss: 0.05393047258257866 2023-01-21 10:59:46.185961: step: 1424/533, loss: 0.0025018840096890926 2023-01-21 10:59:47.346538: step: 1428/533, loss: 0.053116071969270706 2023-01-21 10:59:48.560085: step: 1432/533, loss: 0.04021759331226349 2023-01-21 10:59:49.746521: step: 1436/533, loss: 0.008732987567782402 2023-01-21 10:59:50.967115: step: 1440/533, loss: 0.09969702363014221 2023-01-21 10:59:52.154052: step: 1444/533, loss: 0.042491450905799866 2023-01-21 10:59:53.346938: step: 1448/533, loss: 0.09365804493427277 2023-01-21 10:59:54.545887: step: 1452/533, loss: 0.03684242069721222 2023-01-21 10:59:55.744835: step: 1456/533, loss: 0.04148521274328232 2023-01-21 10:59:56.963731: step: 1460/533, loss: 0.3614819645881653 2023-01-21 10:59:58.151487: step: 1464/533, loss: 0.04030151292681694 2023-01-21 10:59:59.336938: step: 1468/533, loss: 0.040897928178310394 2023-01-21 11:00:00.568612: step: 1472/533, loss: 0.07197470963001251 2023-01-21 11:00:01.786104: step: 1476/533, loss: 0.016492558643221855 2023-01-21 11:00:02.990987: step: 1480/533, loss: 0.03463897854089737 2023-01-21 11:00:04.121588: step: 1484/533, loss: 0.0009126663208007812 2023-01-21 11:00:05.353540: step: 1488/533, loss: 0.09991320967674255 2023-01-21 11:00:06.550276: step: 1492/533, loss: 0.04668637365102768 2023-01-21 11:00:07.726443: step: 1496/533, loss: 0.008671903051435947 2023-01-21 11:00:08.965011: step: 1500/533, loss: 0.8305399417877197 2023-01-21 11:00:10.149086: step: 1504/533, loss: 0.005974197760224342 2023-01-21 11:00:11.339547: step: 1508/533, loss: 0.09068632125854492 2023-01-21 11:00:12.507355: step: 1512/533, loss: 0.034741759300231934 2023-01-21 11:00:13.667179: step: 1516/533, loss: 0.040307048708200455 2023-01-21 11:00:14.837691: step: 1520/533, loss: 0.07411651313304901 2023-01-21 11:00:16.054766: step: 1524/533, loss: 0.02290353737771511 2023-01-21 11:00:17.267083: step: 1528/533, loss: 0.014486503787338734 2023-01-21 11:00:18.434005: step: 1532/533, loss: 0.009680259972810745 2023-01-21 11:00:19.607040: step: 1536/533, loss: 0.12689457833766937 2023-01-21 11:00:20.841501: step: 1540/533, loss: 0.033223867416381836 2023-01-21 11:00:22.041823: step: 1544/533, loss: 0.002190208528190851 2023-01-21 11:00:23.213793: step: 1548/533, loss: 0.015666579827666283 2023-01-21 11:00:24.381371: step: 1552/533, loss: 0.16559754312038422 2023-01-21 11:00:25.576288: step: 1556/533, loss: 0.05890865623950958 2023-01-21 11:00:26.733970: step: 1560/533, loss: 0.015046787448227406 2023-01-21 11:00:27.899491: step: 1564/533, loss: 0.011388206854462624 2023-01-21 11:00:29.162380: step: 1568/533, loss: 0.0432308204472065 2023-01-21 11:00:30.330602: step: 1572/533, loss: 0.7456789016723633 2023-01-21 11:00:31.511560: step: 1576/533, loss: 0.0708545669913292 2023-01-21 11:00:32.694894: step: 1580/533, loss: 0.16015291213989258 2023-01-21 11:00:33.886566: step: 1584/533, loss: 0.10407057404518127 2023-01-21 11:00:35.049977: step: 1588/533, loss: 0.049579620361328125 2023-01-21 11:00:36.243826: step: 1592/533, loss: 0.002498817630112171 2023-01-21 11:00:37.411032: step: 1596/533, loss: 0.005932044703513384 2023-01-21 11:00:38.552578: step: 1600/533, loss: 0.013750839047133923 2023-01-21 11:00:39.764689: step: 1604/533, loss: 0.0308137908577919 2023-01-21 11:00:40.916340: step: 1608/533, loss: 0.01919250376522541 2023-01-21 11:00:42.118811: step: 1612/533, loss: 0.0014053345657885075 2023-01-21 11:00:43.309798: step: 1616/533, loss: 0.04550447687506676 2023-01-21 11:00:44.491656: step: 1620/533, loss: 0.004902282729744911 2023-01-21 11:00:45.670704: step: 1624/533, loss: 0.05277309566736221 2023-01-21 11:00:46.897752: step: 1628/533, loss: 0.037142276763916016 2023-01-21 11:00:48.056895: step: 1632/533, loss: 0.03200273588299751 2023-01-21 11:00:49.225563: step: 1636/533, loss: 0.07220450043678284 2023-01-21 11:00:50.424826: step: 1640/533, loss: 0.022613907232880592 2023-01-21 11:00:51.617638: step: 1644/533, loss: 0.001593732857145369 2023-01-21 11:00:52.779606: step: 1648/533, loss: 0.009949112311005592 2023-01-21 11:00:53.963930: step: 1652/533, loss: 0.0013437271118164062 2023-01-21 11:00:55.198940: step: 1656/533, loss: 0.04954414814710617 2023-01-21 11:00:56.365442: step: 1660/533, loss: 0.0308440700173378 2023-01-21 11:00:57.529346: step: 1664/533, loss: 0.05828666687011719 2023-01-21 11:00:58.728476: step: 1668/533, loss: 0.07057533413171768 2023-01-21 11:00:59.910391: step: 1672/533, loss: 0.04824848473072052 2023-01-21 11:01:01.107854: step: 1676/533, loss: 0.3716858923435211 2023-01-21 11:01:02.305785: step: 1680/533, loss: 0.004019451327621937 2023-01-21 11:01:03.475404: step: 1684/533, loss: 0.04919147491455078 2023-01-21 11:01:04.627053: step: 1688/533, loss: 0.05374279245734215 2023-01-21 11:01:05.790168: step: 1692/533, loss: 0.01891307905316353 2023-01-21 11:01:06.956444: step: 1696/533, loss: 0.021967411041259766 2023-01-21 11:01:08.124923: step: 1700/533, loss: 0.011027527041733265 2023-01-21 11:01:09.327019: step: 1704/533, loss: 0.0049224854446947575 2023-01-21 11:01:10.484490: step: 1708/533, loss: 0.04515886679291725 2023-01-21 11:01:11.673071: step: 1712/533, loss: 0.04739999771118164 2023-01-21 11:01:12.831645: step: 1716/533, loss: 0.05396304279565811 2023-01-21 11:01:13.999740: step: 1720/533, loss: 0.009928131476044655 2023-01-21 11:01:15.157007: step: 1724/533, loss: 0.056420229375362396 2023-01-21 11:01:16.364237: step: 1728/533, loss: 0.007881450466811657 2023-01-21 11:01:17.532308: step: 1732/533, loss: 0.012360621243715286 2023-01-21 11:01:18.725827: step: 1736/533, loss: 0.0024271011352539062 2023-01-21 11:01:19.877943: step: 1740/533, loss: 0.06875310093164444 2023-01-21 11:01:21.074408: step: 1744/533, loss: 0.010922432877123356 2023-01-21 11:01:22.249689: step: 1748/533, loss: 0.006643486209213734 2023-01-21 11:01:23.406088: step: 1752/533, loss: 0.04122591018676758 2023-01-21 11:01:24.560455: step: 1756/533, loss: 0.08097944408655167 2023-01-21 11:01:25.746599: step: 1760/533, loss: 0.03387737274169922 2023-01-21 11:01:26.928311: step: 1764/533, loss: 0.07992801815271378 2023-01-21 11:01:28.109915: step: 1768/533, loss: 5.626814365386963 2023-01-21 11:01:29.304924: step: 1772/533, loss: 0.023672878742218018 2023-01-21 11:01:30.477973: step: 1776/533, loss: 0.11575527489185333 2023-01-21 11:01:31.646360: step: 1780/533, loss: 0.033416748046875 2023-01-21 11:01:32.848189: step: 1784/533, loss: 0.005836295895278454 2023-01-21 11:01:34.054210: step: 1788/533, loss: 0.027249742299318314 2023-01-21 11:01:35.238171: step: 1792/533, loss: 0.1162288710474968 2023-01-21 11:01:36.435885: step: 1796/533, loss: 0.031856488436460495 2023-01-21 11:01:37.604311: step: 1800/533, loss: 0.0013091088039800525 2023-01-21 11:01:38.785290: step: 1804/533, loss: 0.03713731840252876 2023-01-21 11:01:39.984420: step: 1808/533, loss: 0.02197857014834881 2023-01-21 11:01:41.161617: step: 1812/533, loss: 0.002892112825065851 2023-01-21 11:01:42.382130: step: 1816/533, loss: 0.02422657050192356 2023-01-21 11:01:43.562087: step: 1820/533, loss: 0.0020391466096043587 2023-01-21 11:01:44.756452: step: 1824/533, loss: 0.09324569255113602 2023-01-21 11:01:46.002691: step: 1828/533, loss: 0.040514566004276276 2023-01-21 11:01:47.161873: step: 1832/533, loss: 0.015990257263183594 2023-01-21 11:01:48.325335: step: 1836/533, loss: 0.019208334386348724 2023-01-21 11:01:49.485867: step: 1840/533, loss: 0.01714625395834446 2023-01-21 11:01:50.721921: step: 1844/533, loss: 0.38581982254981995 2023-01-21 11:01:51.893413: step: 1848/533, loss: 0.039667606353759766 2023-01-21 11:01:53.066809: step: 1852/533, loss: 0.01530914381146431 2023-01-21 11:01:54.194126: step: 1856/533, loss: 0.05333404242992401 2023-01-21 11:01:55.377085: step: 1860/533, loss: 0.02271137200295925 2023-01-21 11:01:56.627749: step: 1864/533, loss: 0.03758373111486435 2023-01-21 11:01:57.852535: step: 1868/533, loss: 0.019816016778349876 2023-01-21 11:01:59.034218: step: 1872/533, loss: 0.01566324196755886 2023-01-21 11:02:00.216778: step: 1876/533, loss: 0.3339584469795227 2023-01-21 11:02:01.395024: step: 1880/533, loss: 0.06104555353522301 2023-01-21 11:02:02.541758: step: 1884/533, loss: 0.02511494979262352 2023-01-21 11:02:03.702308: step: 1888/533, loss: 0.03393096849322319 2023-01-21 11:02:04.862195: step: 1892/533, loss: 0.1022665947675705 2023-01-21 11:02:06.039258: step: 1896/533, loss: 0.022139834240078926 2023-01-21 11:02:07.199646: step: 1900/533, loss: 0.009954738430678844 2023-01-21 11:02:08.372352: step: 1904/533, loss: 0.012344075366854668 2023-01-21 11:02:09.511106: step: 1908/533, loss: 0.07537984848022461 2023-01-21 11:02:10.700061: step: 1912/533, loss: 0.016133736819028854 2023-01-21 11:02:11.888808: step: 1916/533, loss: 0.020180368795990944 2023-01-21 11:02:13.061499: step: 1920/533, loss: 0.04780330881476402 2023-01-21 11:02:14.228828: step: 1924/533, loss: 0.1405627727508545 2023-01-21 11:02:15.431810: step: 1928/533, loss: 0.060349512845277786 2023-01-21 11:02:16.654052: step: 1932/533, loss: 0.02381448820233345 2023-01-21 11:02:17.880161: step: 1936/533, loss: 0.06789477914571762 2023-01-21 11:02:19.041041: step: 1940/533, loss: 0.009193516336381435 2023-01-21 11:02:20.211265: step: 1944/533, loss: 0.019691942259669304 2023-01-21 11:02:21.414433: step: 1948/533, loss: 0.33529210090637207 2023-01-21 11:02:22.551261: step: 1952/533, loss: 0.04544544592499733 2023-01-21 11:02:23.752864: step: 1956/533, loss: 0.026929188519716263 2023-01-21 11:02:24.924687: step: 1960/533, loss: 0.059825848788022995 2023-01-21 11:02:26.124133: step: 1964/533, loss: 0.03160800784826279 2023-01-21 11:02:27.324899: step: 1968/533, loss: 0.05579938739538193 2023-01-21 11:02:28.527332: step: 1972/533, loss: 0.0343564972281456 2023-01-21 11:02:29.700640: step: 1976/533, loss: 0.3329122066497803 2023-01-21 11:02:30.901156: step: 1980/533, loss: 0.0194975845515728 2023-01-21 11:02:32.089361: step: 1984/533, loss: 0.02982030063867569 2023-01-21 11:02:33.285667: step: 1988/533, loss: 0.027073288336396217 2023-01-21 11:02:34.455865: step: 1992/533, loss: 0.0241559986025095 2023-01-21 11:02:35.657561: step: 1996/533, loss: 0.03220415115356445 2023-01-21 11:02:36.842451: step: 2000/533, loss: 0.16389961540699005 2023-01-21 11:02:38.017441: step: 2004/533, loss: 0.1568075716495514 2023-01-21 11:02:39.238926: step: 2008/533, loss: 0.008940506726503372 2023-01-21 11:02:40.422640: step: 2012/533, loss: 0.0014833450550213456 2023-01-21 11:02:41.590103: step: 2016/533, loss: 0.006083965301513672 2023-01-21 11:02:42.760022: step: 2020/533, loss: 0.009312057867646217 2023-01-21 11:02:43.942820: step: 2024/533, loss: 0.0312558189034462 2023-01-21 11:02:45.167962: step: 2028/533, loss: 0.03570146486163139 2023-01-21 11:02:46.396794: step: 2032/533, loss: 0.0021544217597693205 2023-01-21 11:02:47.574921: step: 2036/533, loss: 0.06766524165868759 2023-01-21 11:02:48.784061: step: 2040/533, loss: 0.05466528236865997 2023-01-21 11:02:49.993137: step: 2044/533, loss: 0.015527725219726562 2023-01-21 11:02:51.206464: step: 2048/533, loss: 0.044196225702762604 2023-01-21 11:02:52.383298: step: 2052/533, loss: 0.05046520382165909 2023-01-21 11:02:53.546413: step: 2056/533, loss: 0.1475900411605835 2023-01-21 11:02:54.766594: step: 2060/533, loss: 0.002491188235580921 2023-01-21 11:02:56.016912: step: 2064/533, loss: 0.006779193878173828 2023-01-21 11:02:57.136231: step: 2068/533, loss: 0.0022489072289317846 2023-01-21 11:02:58.295659: step: 2072/533, loss: 0.007942772470414639 2023-01-21 11:02:59.464096: step: 2076/533, loss: 0.04199938848614693 2023-01-21 11:03:00.628880: step: 2080/533, loss: 0.012682009488344193 2023-01-21 11:03:01.818374: step: 2084/533, loss: 0.09639053046703339 2023-01-21 11:03:02.985301: step: 2088/533, loss: 0.13996896147727966 2023-01-21 11:03:04.173581: step: 2092/533, loss: 0.006974983029067516 2023-01-21 11:03:05.349417: step: 2096/533, loss: 0.029192544519901276 2023-01-21 11:03:06.530152: step: 2100/533, loss: 0.015396308153867722 2023-01-21 11:03:07.723169: step: 2104/533, loss: 0.11576671153306961 2023-01-21 11:03:08.929917: step: 2108/533, loss: 0.12584161758422852 2023-01-21 11:03:10.133653: step: 2112/533, loss: 0.012666946277022362 2023-01-21 11:03:11.366141: step: 2116/533, loss: 0.002259921981021762 2023-01-21 11:03:12.566829: step: 2120/533, loss: 0.025191783905029297 2023-01-21 11:03:13.751451: step: 2124/533, loss: 0.02025766484439373 2023-01-21 11:03:14.965543: step: 2128/533, loss: 0.0872829481959343 2023-01-21 11:03:16.136305: step: 2132/533, loss: 0.004870033357292414 ================================================== Loss: 0.065 -------------------- Dev: {'event': {'p': 0.5871369294605809, 'r': 0.7536617842876165, 'f1': 0.660058309037901}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Test: {'event': {'p': 0.6406406406406406, 'r': 0.8184143222506394, 'f1': 0.7186973610331273}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Chinese: {'event': {'p': 0.5384615384615384, 'r': 0.9074074074074074, 'f1': 0.6758620689655174}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Korean: {'event': {'p': 0.5862068965517241, 'r': 0.5396825396825397, 'f1': 0.5619834710743802}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} 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': 13} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 11:03:58.162080: step: 4/533, loss: 0.0518094077706337 2023-01-21 11:03:59.364582: step: 8/533, loss: 0.05276336893439293 2023-01-21 11:04:00.536934: step: 12/533, loss: 0.011742830276489258 2023-01-21 11:04:01.718640: step: 16/533, loss: 0.009232045151293278 2023-01-21 11:04:02.924618: step: 20/533, loss: 0.0018863677978515625 2023-01-21 11:04:04.111206: step: 24/533, loss: 0.03952836990356445 2023-01-21 11:04:05.287057: step: 28/533, loss: 1.133405327796936 2023-01-21 11:04:06.472187: step: 32/533, loss: 0.03339672088623047 2023-01-21 11:04:07.680935: step: 36/533, loss: 0.003053999040275812 2023-01-21 11:04:08.850646: step: 40/533, loss: 0.001923418021760881 2023-01-21 11:04:10.051645: step: 44/533, loss: 0.01966419257223606 2023-01-21 11:04:11.294865: step: 48/533, loss: 0.022713089361786842 2023-01-21 11:04:12.484674: step: 52/533, loss: 0.016245078295469284 2023-01-21 11:04:13.679064: step: 56/533, loss: 0.002317714737728238 2023-01-21 11:04:14.840326: step: 60/533, loss: 0.037776585668325424 2023-01-21 11:04:16.037031: step: 64/533, loss: 0.031238652765750885 2023-01-21 11:04:17.190252: step: 68/533, loss: 0.026325320824980736 2023-01-21 11:04:18.373621: step: 72/533, loss: 0.003839445300400257 2023-01-21 11:04:19.537078: step: 76/533, loss: 0.12901124358177185 2023-01-21 11:04:20.721350: step: 80/533, loss: 0.021156905218958855 2023-01-21 11:04:21.906555: step: 84/533, loss: 0.020279789343476295 2023-01-21 11:04:23.159054: step: 88/533, loss: 0.01670360565185547 2023-01-21 11:04:24.360484: step: 92/533, loss: 0.0005053520435467362 2023-01-21 11:04:25.562977: step: 96/533, loss: 0.000706577324308455 2023-01-21 11:04:26.758147: step: 100/533, loss: 0.013101291842758656 2023-01-21 11:04:27.957036: step: 104/533, loss: 0.004114056006073952 2023-01-21 11:04:29.120361: step: 108/533, loss: 0.02924833446741104 2023-01-21 11:04:30.287700: step: 112/533, loss: 0.00824208278208971 2023-01-21 11:04:31.473879: step: 116/533, loss: 0.06478758156299591 2023-01-21 11:04:32.700546: step: 120/533, loss: 0.004162550438195467 2023-01-21 11:04:33.879245: step: 124/533, loss: 0.04760627821087837 2023-01-21 11:04:35.103847: step: 128/533, loss: 0.002372550778090954 2023-01-21 11:04:36.286930: step: 132/533, loss: 0.011991691775619984 2023-01-21 11:04:37.473025: step: 136/533, loss: 0.022435570135712624 2023-01-21 11:04:38.614155: step: 140/533, loss: 0.023341942578554153 2023-01-21 11:04:39.796840: step: 144/533, loss: 0.011906379833817482 2023-01-21 11:04:40.991256: step: 148/533, loss: 0.005397891625761986 2023-01-21 11:04:42.189271: step: 152/533, loss: 0.026522351428866386 2023-01-21 11:04:43.393678: step: 156/533, loss: 0.00797882117331028 2023-01-21 11:04:44.580007: step: 160/533, loss: 0.00012578963651321828 2023-01-21 11:04:45.769141: step: 164/533, loss: 0.017604637891054153 2023-01-21 11:04:46.989405: step: 168/533, loss: 0.010859060101211071 2023-01-21 11:04:48.220226: step: 172/533, loss: 0.021741580218076706 2023-01-21 11:04:49.428234: step: 176/533, loss: 0.013530444353818893 2023-01-21 11:04:50.649504: step: 180/533, loss: 0.003749370574951172 2023-01-21 11:04:51.829245: step: 184/533, loss: 0.023764800280332565 2023-01-21 11:04:53.014178: step: 188/533, loss: 0.003438282059505582 2023-01-21 11:04:54.190383: step: 192/533, loss: 0.007263851352035999 2023-01-21 11:04:55.369135: step: 196/533, loss: 0.049765635281801224 2023-01-21 11:04:56.581789: step: 200/533, loss: 0.08769015967845917 2023-01-21 11:04:57.759513: step: 204/533, loss: 0.18049907684326172 2023-01-21 11:04:58.963237: step: 208/533, loss: 0.014322662726044655 2023-01-21 11:05:00.159752: step: 212/533, loss: 0.05369434505701065 2023-01-21 11:05:01.359171: step: 216/533, loss: 0.06156301498413086 2023-01-21 11:05:02.529274: step: 220/533, loss: 0.033228494226932526 2023-01-21 11:05:03.688299: step: 224/533, loss: 0.007882023230195045 2023-01-21 11:05:04.852574: step: 228/533, loss: 0.001933765597641468 2023-01-21 11:05:06.033200: step: 232/533, loss: 0.05712109059095383 2023-01-21 11:05:07.177089: step: 236/533, loss: 0.011195087805390358 2023-01-21 11:05:08.335034: step: 240/533, loss: 0.032080747187137604 2023-01-21 11:05:09.568581: step: 244/533, loss: 0.027440454810857773 2023-01-21 11:05:10.732367: step: 248/533, loss: 0.0005819320795126259 2023-01-21 11:05:11.892684: step: 252/533, loss: 0.036676786839962006 2023-01-21 11:05:13.108481: step: 256/533, loss: 0.05094585195183754 2023-01-21 11:05:14.291147: step: 260/533, loss: 0.0014437198406085372 2023-01-21 11:05:15.451571: step: 264/533, loss: 0.012420654296875 2023-01-21 11:05:16.606025: step: 268/533, loss: 0.002605485962703824 2023-01-21 11:05:17.781013: step: 272/533, loss: 0.00310173025354743 2023-01-21 11:05:18.912772: step: 276/533, loss: 0.002975702518597245 2023-01-21 11:05:20.089373: step: 280/533, loss: 0.04784102365374565 2023-01-21 11:05:21.271670: step: 284/533, loss: 0.0175428856164217 2023-01-21 11:05:22.416899: step: 288/533, loss: 0.004173708148300648 2023-01-21 11:05:23.601866: step: 292/533, loss: 0.2548210024833679 2023-01-21 11:05:24.781646: step: 296/533, loss: 0.0013281822903081775 2023-01-21 11:05:25.970177: step: 300/533, loss: 0.02344522438943386 2023-01-21 11:05:27.205888: step: 304/533, loss: 0.05363183096051216 2023-01-21 11:05:28.387465: step: 308/533, loss: 0.013500213623046875 2023-01-21 11:05:29.572466: step: 312/533, loss: 0.06136922910809517 2023-01-21 11:05:30.780560: step: 316/533, loss: 0.0007150173187255859 2023-01-21 11:05:31.988436: step: 320/533, loss: 0.002105522435158491 2023-01-21 11:05:33.183185: step: 324/533, loss: 0.5246507525444031 2023-01-21 11:05:34.381911: step: 328/533, loss: 0.03873424604535103 2023-01-21 11:05:35.555731: step: 332/533, loss: 0.0029366493690758944 2023-01-21 11:05:36.754003: step: 336/533, loss: 0.01386947650462389 2023-01-21 11:05:37.912570: step: 340/533, loss: 0.013595200143754482 2023-01-21 11:05:39.112682: step: 344/533, loss: 0.03603091463446617 2023-01-21 11:05:40.299627: step: 348/533, loss: 0.0012850761413574219 2023-01-21 11:05:41.519631: step: 352/533, loss: 0.004223919007927179 2023-01-21 11:05:42.734115: step: 356/533, loss: 0.03832054138183594 2023-01-21 11:05:43.919560: step: 360/533, loss: 0.04459753260016441 2023-01-21 11:05:45.118818: step: 364/533, loss: 0.0398649238049984 2023-01-21 11:05:46.287919: step: 368/533, loss: 0.03520069271326065 2023-01-21 11:05:47.474014: step: 372/533, loss: 0.03638019412755966 2023-01-21 11:05:48.614830: step: 376/533, loss: 0.05231475830078125 2023-01-21 11:05:49.796162: step: 380/533, loss: 0.4421539306640625 2023-01-21 11:05:51.031927: step: 384/533, loss: 0.20982475578784943 2023-01-21 11:05:52.156159: step: 388/533, loss: 0.07425327599048615 2023-01-21 11:05:53.354058: step: 392/533, loss: 0.0016329764621332288 2023-01-21 11:05:54.539962: step: 396/533, loss: 0.05830249935388565 2023-01-21 11:05:55.733481: step: 400/533, loss: 0.008459853939712048 2023-01-21 11:05:56.918497: step: 404/533, loss: 0.004722786135971546 2023-01-21 11:05:58.122228: step: 408/533, loss: 0.0774863213300705 2023-01-21 11:05:59.263941: step: 412/533, loss: 0.0028838159050792456 2023-01-21 11:06:00.489768: step: 416/533, loss: 0.10506601631641388 2023-01-21 11:06:01.647968: step: 420/533, loss: 0.27414941787719727 2023-01-21 11:06:02.870859: step: 424/533, loss: 0.010890627279877663 2023-01-21 11:06:04.164528: step: 428/533, loss: 0.01600799523293972 2023-01-21 11:06:05.346189: step: 432/533, loss: 0.03158445283770561 2023-01-21 11:06:06.597595: step: 436/533, loss: 0.03226461261510849 2023-01-21 11:06:07.746246: step: 440/533, loss: 0.02806882932782173 2023-01-21 11:06:08.940089: step: 444/533, loss: 0.6819657683372498 2023-01-21 11:06:10.160141: step: 448/533, loss: 0.012016487307846546 2023-01-21 11:06:11.335923: step: 452/533, loss: 0.007284260354936123 2023-01-21 11:06:12.511386: step: 456/533, loss: 0.009635353460907936 2023-01-21 11:06:13.714446: step: 460/533, loss: 0.010797214694321156 2023-01-21 11:06:14.903916: step: 464/533, loss: 0.0497320182621479 2023-01-21 11:06:16.084252: step: 468/533, loss: 0.0053924559615552425 2023-01-21 11:06:17.271813: step: 472/533, loss: 0.03964591026306152 2023-01-21 11:06:18.428176: step: 476/533, loss: 0.0021451949141919613 2023-01-21 11:06:19.596218: step: 480/533, loss: 0.02539977990090847 2023-01-21 11:06:20.808155: step: 484/533, loss: 0.0027429580222815275 2023-01-21 11:06:22.014054: step: 488/533, loss: 0.09211654216051102 2023-01-21 11:06:23.206134: step: 492/533, loss: 0.00011256933066761121 2023-01-21 11:06:24.390526: step: 496/533, loss: 0.023706912994384766 2023-01-21 11:06:25.555056: step: 500/533, loss: 0.00475311279296875 2023-01-21 11:06:26.706954: step: 504/533, loss: 0.010573959909379482 2023-01-21 11:06:27.927413: step: 508/533, loss: 0.00876550655812025 2023-01-21 11:06:29.077205: step: 512/533, loss: 0.0004525184631347656 2023-01-21 11:06:30.291879: step: 516/533, loss: 0.0011173248058184981 2023-01-21 11:06:31.470281: step: 520/533, loss: 0.009097481146454811 2023-01-21 11:06:32.631137: step: 524/533, loss: 0.2122957855463028 2023-01-21 11:06:33.826565: step: 528/533, loss: 0.37422817945480347 2023-01-21 11:06:34.991815: step: 532/533, loss: 0.03604459762573242 2023-01-21 11:06:36.187652: step: 536/533, loss: 0.023160744458436966 2023-01-21 11:06:37.374396: step: 540/533, loss: 0.003912448883056641 2023-01-21 11:06:38.545525: step: 544/533, loss: 0.009313011541962624 2023-01-21 11:06:39.678681: step: 548/533, loss: 0.006199741270393133 2023-01-21 11:06:40.873325: step: 552/533, loss: 0.006425977218896151 2023-01-21 11:06:42.120984: step: 556/533, loss: 0.01684131659567356 2023-01-21 11:06:43.295614: step: 560/533, loss: 0.00794363021850586 2023-01-21 11:06:44.464166: step: 564/533, loss: 0.04762868955731392 2023-01-21 11:06:45.649945: step: 568/533, loss: 0.0741356909275055 2023-01-21 11:06:46.854042: step: 572/533, loss: 0.0008775173919275403 2023-01-21 11:06:48.022682: step: 576/533, loss: 0.04950456693768501 2023-01-21 11:06:49.229975: step: 580/533, loss: 0.021826554089784622 2023-01-21 11:06:50.415520: step: 584/533, loss: 0.0016706467140465975 2023-01-21 11:06:51.602407: step: 588/533, loss: 0.03202934190630913 2023-01-21 11:06:52.782276: step: 592/533, loss: 0.0002475738583598286 2023-01-21 11:06:53.931445: step: 596/533, loss: 1.538374662399292 2023-01-21 11:06:55.089656: step: 600/533, loss: 0.0041249278001487255 2023-01-21 11:06:56.293459: step: 604/533, loss: 0.04962005838751793 2023-01-21 11:06:57.486838: step: 608/533, loss: 0.04288854822516441 2023-01-21 11:06:58.676439: step: 612/533, loss: 0.05195961147546768 2023-01-21 11:06:59.869452: step: 616/533, loss: 0.04928302764892578 2023-01-21 11:07:01.042755: step: 620/533, loss: 0.0015430449275299907 2023-01-21 11:07:02.190195: step: 624/533, loss: 0.005393601022660732 2023-01-21 11:07:03.395408: step: 628/533, loss: 0.01617260091006756 2023-01-21 11:07:04.542523: step: 632/533, loss: 0.023660849779844284 2023-01-21 11:07:05.679072: step: 636/533, loss: 0.00185480120126158 2023-01-21 11:07:06.854736: step: 640/533, loss: 0.04813957214355469 2023-01-21 11:07:08.052262: step: 644/533, loss: 0.0995282232761383 2023-01-21 11:07:09.197379: step: 648/533, loss: 0.073021799325943 2023-01-21 11:07:10.375316: step: 652/533, loss: 0.003402900882065296 2023-01-21 11:07:11.549019: step: 656/533, loss: 0.0019078254699707031 2023-01-21 11:07:12.716521: step: 660/533, loss: 0.004020499996840954 2023-01-21 11:07:13.922312: step: 664/533, loss: 0.06028938665986061 2023-01-21 11:07:15.114550: step: 668/533, loss: 0.09243932366371155 2023-01-21 11:07:16.322128: step: 672/533, loss: 0.009998321533203125 2023-01-21 11:07:17.497676: step: 676/533, loss: 0.020939921960234642 2023-01-21 11:07:18.654223: step: 680/533, loss: 0.004695129580795765 2023-01-21 11:07:19.805704: step: 684/533, loss: 0.010105801746249199 2023-01-21 11:07:21.028442: step: 688/533, loss: 0.007549285888671875 2023-01-21 11:07:22.227976: step: 692/533, loss: 0.0031150816939771175 2023-01-21 11:07:23.355636: step: 696/533, loss: 0.04190712049603462 2023-01-21 11:07:24.546002: step: 700/533, loss: 0.02550687827169895 2023-01-21 11:07:25.748783: step: 704/533, loss: 0.002892398973926902 2023-01-21 11:07:26.936035: step: 708/533, loss: 0.34170034527778625 2023-01-21 11:07:28.124037: step: 712/533, loss: 0.0010183334816247225 2023-01-21 11:07:29.315053: step: 716/533, loss: 0.04951591417193413 2023-01-21 11:07:30.489886: step: 720/533, loss: 0.09577446430921555 2023-01-21 11:07:31.689092: step: 724/533, loss: 0.03147487714886665 2023-01-21 11:07:32.884303: step: 728/533, loss: 0.00046219825162552297 2023-01-21 11:07:34.110137: step: 732/533, loss: 0.07083601504564285 2023-01-21 11:07:35.278310: step: 736/533, loss: 0.08560695499181747 2023-01-21 11:07:36.453205: step: 740/533, loss: 0.04132995754480362 2023-01-21 11:07:37.622387: step: 744/533, loss: 0.017970561981201172 2023-01-21 11:07:38.845750: step: 748/533, loss: 0.018801499158143997 2023-01-21 11:07:40.003511: step: 752/533, loss: 0.06247539818286896 2023-01-21 11:07:41.191057: step: 756/533, loss: 0.019234562292695045 2023-01-21 11:07:42.373968: step: 760/533, loss: 0.02002892643213272 2023-01-21 11:07:43.564500: step: 764/533, loss: 0.015103436075150967 2023-01-21 11:07:44.732869: step: 768/533, loss: 0.008491325192153454 2023-01-21 11:07:45.920061: step: 772/533, loss: 0.0006873130914755166 2023-01-21 11:07:47.123813: step: 776/533, loss: 0.0072487834841012955 2023-01-21 11:07:48.376769: step: 780/533, loss: 0.045395851135253906 2023-01-21 11:07:49.540724: step: 784/533, loss: 0.5315955877304077 2023-01-21 11:07:50.742020: step: 788/533, loss: 0.020901966840028763 2023-01-21 11:07:51.912709: step: 792/533, loss: 0.0020432472229003906 2023-01-21 11:07:53.096269: step: 796/533, loss: 0.0017360687488690019 2023-01-21 11:07:54.276148: step: 800/533, loss: 0.013344001956284046 2023-01-21 11:07:55.486589: step: 804/533, loss: 0.02614021487534046 2023-01-21 11:07:56.657594: step: 808/533, loss: 0.0005293846479617059 2023-01-21 11:07:57.830803: step: 812/533, loss: 0.0006503105396404862 2023-01-21 11:07:59.020057: step: 816/533, loss: 0.03363943099975586 2023-01-21 11:08:00.215739: step: 820/533, loss: 0.05826082453131676 2023-01-21 11:08:01.368015: step: 824/533, loss: 0.007171630859375 2023-01-21 11:08:02.565641: step: 828/533, loss: 0.00195732107385993 2023-01-21 11:08:03.713639: step: 832/533, loss: 0.012225913815200329 2023-01-21 11:08:04.907230: step: 836/533, loss: 0.025861550122499466 2023-01-21 11:08:06.122903: step: 840/533, loss: 0.009180068969726562 2023-01-21 11:08:07.310296: step: 844/533, loss: 0.0018133163684979081 2023-01-21 11:08:08.466024: step: 848/533, loss: 0.02747049368917942 2023-01-21 11:08:09.638948: step: 852/533, loss: 0.00331459054723382 2023-01-21 11:08:10.822185: step: 856/533, loss: 0.009013843722641468 2023-01-21 11:08:11.992109: step: 860/533, loss: 0.02509593963623047 2023-01-21 11:08:13.193932: step: 864/533, loss: 0.032434847205877304 2023-01-21 11:08:14.405168: step: 868/533, loss: 0.16066695749759674 2023-01-21 11:08:15.584184: step: 872/533, loss: 0.049582671374082565 2023-01-21 11:08:16.747467: step: 876/533, loss: 0.0167510025203228 2023-01-21 11:08:17.870095: step: 880/533, loss: 0.001733493758365512 2023-01-21 11:08:19.035625: step: 884/533, loss: 0.023692702874541283 2023-01-21 11:08:20.195701: step: 888/533, loss: 1.039563775062561 2023-01-21 11:08:21.361960: step: 892/533, loss: 1.1634827387752011e-05 2023-01-21 11:08:22.569005: step: 896/533, loss: 0.043808698654174805 2023-01-21 11:08:23.737687: step: 900/533, loss: 0.019817162305116653 2023-01-21 11:08:24.938641: step: 904/533, loss: 0.011703675612807274 2023-01-21 11:08:26.160820: step: 908/533, loss: 0.00259494804777205 2023-01-21 11:08:27.340576: step: 912/533, loss: 0.029726792126893997 2023-01-21 11:08:28.540354: step: 916/533, loss: 0.061312485486269 2023-01-21 11:08:29.711956: step: 920/533, loss: 0.010663175955414772 2023-01-21 11:08:30.884849: step: 924/533, loss: 0.004093170631676912 2023-01-21 11:08:32.070286: step: 928/533, loss: 0.16270485520362854 2023-01-21 11:08:33.227611: step: 932/533, loss: 0.002553558209910989 2023-01-21 11:08:34.434007: step: 936/533, loss: 0.20938794314861298 2023-01-21 11:08:35.594356: step: 940/533, loss: 0.044144246727228165 2023-01-21 11:08:36.766930: step: 944/533, loss: 0.0159467700868845 2023-01-21 11:08:37.983111: step: 948/533, loss: 0.12271471321582794 2023-01-21 11:08:39.158212: step: 952/533, loss: 0.005106735043227673 2023-01-21 11:08:40.383611: step: 956/533, loss: 0.08501043170690536 2023-01-21 11:08:41.550377: step: 960/533, loss: 0.082502081990242 2023-01-21 11:08:42.722773: step: 964/533, loss: 0.02762739732861519 2023-01-21 11:08:43.921668: step: 968/533, loss: 0.02583923563361168 2023-01-21 11:08:45.141476: step: 972/533, loss: 0.33078011870384216 2023-01-21 11:08:46.343095: step: 976/533, loss: 0.008112717419862747 2023-01-21 11:08:47.521225: step: 980/533, loss: 0.07246056199073792 2023-01-21 11:08:48.679413: step: 984/533, loss: 0.011728477664291859 2023-01-21 11:08:49.881252: step: 988/533, loss: 0.0017777920002117753 2023-01-21 11:08:51.110697: step: 992/533, loss: 0.07626251876354218 2023-01-21 11:08:52.300074: step: 996/533, loss: 0.01080942153930664 2023-01-21 11:08:53.451049: step: 1000/533, loss: 0.10545816272497177 2023-01-21 11:08:54.642123: step: 1004/533, loss: 0.005780220031738281 2023-01-21 11:08:55.843327: step: 1008/533, loss: 0.03603062778711319 2023-01-21 11:08:57.037491: step: 1012/533, loss: 0.07141857594251633 2023-01-21 11:08:58.220312: step: 1016/533, loss: 0.031074905768036842 2023-01-21 11:08:59.417502: step: 1020/533, loss: 0.0625300407409668 2023-01-21 11:09:00.631329: step: 1024/533, loss: 0.0076808929443359375 2023-01-21 11:09:01.823984: step: 1028/533, loss: 0.04914788901805878 2023-01-21 11:09:02.986208: step: 1032/533, loss: 0.0006085395580157638 2023-01-21 11:09:04.225468: step: 1036/533, loss: 0.0002342224179301411 2023-01-21 11:09:05.408639: step: 1040/533, loss: 0.08393802493810654 2023-01-21 11:09:06.560917: step: 1044/533, loss: 0.03736257553100586 2023-01-21 11:09:07.763129: step: 1048/533, loss: 0.013930320739746094 2023-01-21 11:09:08.903836: step: 1052/533, loss: 0.04577846825122833 2023-01-21 11:09:10.062961: step: 1056/533, loss: 0.00041623116703704 2023-01-21 11:09:11.237237: step: 1060/533, loss: 0.06527204811573029 2023-01-21 11:09:12.411284: step: 1064/533, loss: 0.005023288540542126 2023-01-21 11:09:13.571425: step: 1068/533, loss: 0.0074495794251561165 2023-01-21 11:09:14.788297: step: 1072/533, loss: 0.05359682813286781 2023-01-21 11:09:15.968016: step: 1076/533, loss: 0.03528108820319176 2023-01-21 11:09:17.147557: step: 1080/533, loss: 0.0033839941024780273 2023-01-21 11:09:18.340724: step: 1084/533, loss: 0.03252124786376953 2023-01-21 11:09:19.557530: step: 1088/533, loss: 0.030231047421693802 2023-01-21 11:09:20.742846: step: 1092/533, loss: 0.09924106299877167 2023-01-21 11:09:21.951748: step: 1096/533, loss: 0.005743694491684437 2023-01-21 11:09:23.151753: step: 1100/533, loss: 0.0025201798416674137 2023-01-21 11:09:24.319073: step: 1104/533, loss: 0.015850067138671875 2023-01-21 11:09:25.482067: step: 1108/533, loss: 0.021021222695708275 2023-01-21 11:09:26.662279: step: 1112/533, loss: 0.08223210275173187 2023-01-21 11:09:27.862520: step: 1116/533, loss: 0.0052163125947117805 2023-01-21 11:09:29.020555: step: 1120/533, loss: 0.014155101962387562 2023-01-21 11:09:30.192679: step: 1124/533, loss: 0.020010948181152344 2023-01-21 11:09:31.355830: step: 1128/533, loss: 0.04718637838959694 2023-01-21 11:09:32.510686: step: 1132/533, loss: 0.012091445736587048 2023-01-21 11:09:33.690030: step: 1136/533, loss: 0.06243286281824112 2023-01-21 11:09:34.846180: step: 1140/533, loss: 0.03770342096686363 2023-01-21 11:09:36.015464: step: 1144/533, loss: 0.0015277862548828125 2023-01-21 11:09:37.241020: step: 1148/533, loss: 0.07014961540699005 2023-01-21 11:09:38.422855: step: 1152/533, loss: 0.032598257064819336 2023-01-21 11:09:39.575601: step: 1156/533, loss: 0.11392717063426971 2023-01-21 11:09:40.736969: step: 1160/533, loss: 0.002681923098862171 2023-01-21 11:09:41.920726: step: 1164/533, loss: 0.0587984099984169 2023-01-21 11:09:43.122502: step: 1168/533, loss: 0.019454097375273705 2023-01-21 11:09:44.294359: step: 1172/533, loss: 0.004249954130500555 2023-01-21 11:09:45.449854: step: 1176/533, loss: 0.08586225658655167 2023-01-21 11:09:46.652544: step: 1180/533, loss: 0.04163970798254013 2023-01-21 11:09:47.838431: step: 1184/533, loss: 0.013718700036406517 2023-01-21 11:09:49.032471: step: 1188/533, loss: 0.035169318318367004 2023-01-21 11:09:50.197055: step: 1192/533, loss: 0.018397999927401543 2023-01-21 11:09:51.406428: step: 1196/533, loss: 0.09522581100463867 2023-01-21 11:09:52.614291: step: 1200/533, loss: 0.003457868006080389 2023-01-21 11:09:53.786169: step: 1204/533, loss: 0.02933807298541069 2023-01-21 11:09:54.983464: step: 1208/533, loss: 0.06172552332282066 2023-01-21 11:09:56.182856: step: 1212/533, loss: 0.026673413813114166 2023-01-21 11:09:57.344316: step: 1216/533, loss: 0.003959846682846546 2023-01-21 11:09:58.520354: step: 1220/533, loss: 0.0015020370483398438 2023-01-21 11:09:59.693222: step: 1224/533, loss: 0.0026490213349461555 2023-01-21 11:10:00.885283: step: 1228/533, loss: 0.0009955406421795487 2023-01-21 11:10:02.048656: step: 1232/533, loss: 0.0056493766605854034 2023-01-21 11:10:03.239263: step: 1236/533, loss: 0.008498001843690872 2023-01-21 11:10:04.414385: step: 1240/533, loss: 0.028267955407500267 2023-01-21 11:10:05.651490: step: 1244/533, loss: 0.24183063209056854 2023-01-21 11:10:06.831849: step: 1248/533, loss: 0.07068081200122833 2023-01-21 11:10:08.043663: step: 1252/533, loss: 0.0030951499938964844 2023-01-21 11:10:09.259275: step: 1256/533, loss: 0.08301430195569992 2023-01-21 11:10:10.418780: step: 1260/533, loss: 0.025484465062618256 2023-01-21 11:10:11.613296: step: 1264/533, loss: 0.05164022371172905 2023-01-21 11:10:12.803606: step: 1268/533, loss: 0.12735433876514435 2023-01-21 11:10:13.998669: step: 1272/533, loss: 0.042942240834236145 2023-01-21 11:10:15.225285: step: 1276/533, loss: 0.043050576001405716 2023-01-21 11:10:16.431581: step: 1280/533, loss: 0.02764721028506756 2023-01-21 11:10:17.634298: step: 1284/533, loss: 0.08372178673744202 2023-01-21 11:10:18.797674: step: 1288/533, loss: 0.22312016785144806 2023-01-21 11:10:19.978417: step: 1292/533, loss: 0.0036039352416992188 2023-01-21 11:10:21.189730: step: 1296/533, loss: 0.0007527351845055819 2023-01-21 11:10:22.410019: step: 1300/533, loss: 0.04929714277386665 2023-01-21 11:10:23.565737: step: 1304/533, loss: 0.005889320746064186 2023-01-21 11:10:24.756749: step: 1308/533, loss: 0.014123249799013138 2023-01-21 11:10:25.895379: step: 1312/533, loss: 0.00188446044921875 2023-01-21 11:10:27.070251: step: 1316/533, loss: 0.0700952559709549 2023-01-21 11:10:28.234948: step: 1320/533, loss: 0.04431352764368057 2023-01-21 11:10:29.470019: step: 1324/533, loss: 0.003768348600715399 2023-01-21 11:10:30.652572: step: 1328/533, loss: 0.18981055915355682 2023-01-21 11:10:31.920892: step: 1332/533, loss: 0.09900808334350586 2023-01-21 11:10:33.122054: step: 1336/533, loss: 0.0039764405228197575 2023-01-21 11:10:34.317406: step: 1340/533, loss: 0.022089386358857155 2023-01-21 11:10:35.566407: step: 1344/533, loss: 0.0007393836858682334 2023-01-21 11:10:36.770865: step: 1348/533, loss: 0.1096283420920372 2023-01-21 11:10:37.949192: step: 1352/533, loss: 0.03020324744284153 2023-01-21 11:10:39.126092: step: 1356/533, loss: 0.07181215286254883 2023-01-21 11:10:40.319457: step: 1360/533, loss: 0.03059682995080948 2023-01-21 11:10:41.489495: step: 1364/533, loss: 0.00811699591577053 2023-01-21 11:10:42.673682: step: 1368/533, loss: 0.0016086578834801912 2023-01-21 11:10:43.877314: step: 1372/533, loss: 0.16164740920066833 2023-01-21 11:10:45.047697: step: 1376/533, loss: 0.0023539543617516756 2023-01-21 11:10:46.211894: step: 1380/533, loss: 0.06281928718090057 2023-01-21 11:10:47.390674: step: 1384/533, loss: 0.019055986776947975 2023-01-21 11:10:48.623561: step: 1388/533, loss: 0.0008141517755575478 2023-01-21 11:10:49.794362: step: 1392/533, loss: 0.024837160483002663 2023-01-21 11:10:50.938826: step: 1396/533, loss: 0.031973838806152344 2023-01-21 11:10:52.083808: step: 1400/533, loss: 0.020502157509326935 2023-01-21 11:10:53.260870: step: 1404/533, loss: 0.01667661778628826 2023-01-21 11:10:54.438343: step: 1408/533, loss: 0.0005800724611617625 2023-01-21 11:10:55.629630: step: 1412/533, loss: 0.1979566514492035 2023-01-21 11:10:56.778232: step: 1416/533, loss: 0.03205595165491104 2023-01-21 11:10:57.951848: step: 1420/533, loss: 0.00010177493095397949 2023-01-21 11:10:59.196216: step: 1424/533, loss: 0.03745689615607262 2023-01-21 11:11:00.392354: step: 1428/533, loss: 0.0004849433898925781 2023-01-21 11:11:01.595057: step: 1432/533, loss: 0.012656641192734241 2023-01-21 11:11:02.747523: step: 1436/533, loss: 0.4612763524055481 2023-01-21 11:11:03.923835: step: 1440/533, loss: 0.0041783335618674755 2023-01-21 11:11:05.092068: step: 1444/533, loss: 0.0037330626510083675 2023-01-21 11:11:06.265289: step: 1448/533, loss: 0.07242584228515625 2023-01-21 11:11:07.484948: step: 1452/533, loss: 0.07093124091625214 2023-01-21 11:11:08.638402: step: 1456/533, loss: 0.040244292467832565 2023-01-21 11:11:09.814532: step: 1460/533, loss: 0.024210548028349876 2023-01-21 11:11:10.976605: step: 1464/533, loss: 0.0807284340262413 2023-01-21 11:11:12.176954: step: 1468/533, loss: 0.010180473327636719 2023-01-21 11:11:13.347734: step: 1472/533, loss: 0.04467420652508736 2023-01-21 11:11:14.524061: step: 1476/533, loss: 0.05566120147705078 2023-01-21 11:11:15.686121: step: 1480/533, loss: 0.03325071558356285 2023-01-21 11:11:16.868565: step: 1484/533, loss: 0.04644813388586044 2023-01-21 11:11:18.111011: step: 1488/533, loss: 0.006442260928452015 2023-01-21 11:11:19.341382: step: 1492/533, loss: 0.04137783125042915 2023-01-21 11:11:20.526255: step: 1496/533, loss: 0.002364254090934992 2023-01-21 11:11:21.691595: step: 1500/533, loss: 0.012958860956132412 2023-01-21 11:11:22.865333: step: 1504/533, loss: 0.032246969640254974 2023-01-21 11:11:24.014835: step: 1508/533, loss: 0.012474918738007545 2023-01-21 11:11:25.192788: step: 1512/533, loss: 0.005384350195527077 2023-01-21 11:11:26.404622: step: 1516/533, loss: 0.021234415471553802 2023-01-21 11:11:27.536459: step: 1520/533, loss: 0.0019318581325933337 2023-01-21 11:11:28.715690: step: 1524/533, loss: 0.10267606377601624 2023-01-21 11:11:29.905624: step: 1528/533, loss: 0.03939533233642578 2023-01-21 11:11:31.118076: step: 1532/533, loss: 0.012689972296357155 2023-01-21 11:11:32.344887: step: 1536/533, loss: 0.0326385498046875 2023-01-21 11:11:33.540793: step: 1540/533, loss: 0.009597301483154297 2023-01-21 11:11:34.701793: step: 1544/533, loss: 0.0153534896671772 2023-01-21 11:11:35.841692: step: 1548/533, loss: 0.09442967921495438 2023-01-21 11:11:37.004381: step: 1552/533, loss: 0.02636394463479519 2023-01-21 11:11:38.154972: step: 1556/533, loss: 0.008846051059663296 2023-01-21 11:11:39.323799: step: 1560/533, loss: 0.018706750124692917 2023-01-21 11:11:40.484893: step: 1564/533, loss: 0.04613618552684784 2023-01-21 11:11:41.651769: step: 1568/533, loss: 0.012685585767030716 2023-01-21 11:11:42.828233: step: 1572/533, loss: 0.02436218224465847 2023-01-21 11:11:43.987573: step: 1576/533, loss: 0.03539569675922394 2023-01-21 11:11:45.181099: step: 1580/533, loss: 0.0009920119773596525 2023-01-21 11:11:46.349939: step: 1584/533, loss: 0.009822189807891846 2023-01-21 11:11:47.505231: step: 1588/533, loss: 0.0009979248279705644 2023-01-21 11:11:48.689318: step: 1592/533, loss: 0.045466091483831406 2023-01-21 11:11:49.903182: step: 1596/533, loss: 0.0031791210640221834 2023-01-21 11:11:51.081739: step: 1600/533, loss: 0.14055366814136505 2023-01-21 11:11:52.254357: step: 1604/533, loss: 0.034258175641298294 2023-01-21 11:11:53.400568: step: 1608/533, loss: 0.04239988327026367 2023-01-21 11:11:54.569468: step: 1612/533, loss: 0.016264747828245163 2023-01-21 11:11:55.735634: step: 1616/533, loss: 0.262167751789093 2023-01-21 11:11:56.907996: step: 1620/533, loss: 0.0024843215942382812 2023-01-21 11:11:58.085646: step: 1624/533, loss: 0.025490522384643555 2023-01-21 11:11:59.251511: step: 1628/533, loss: 0.060387518256902695 2023-01-21 11:12:00.403089: step: 1632/533, loss: 0.004184436984360218 2023-01-21 11:12:01.561985: step: 1636/533, loss: 0.004380142781883478 2023-01-21 11:12:02.777558: step: 1640/533, loss: 0.00661888113245368 2023-01-21 11:12:03.935273: step: 1644/533, loss: 0.0020897863432765007 2023-01-21 11:12:05.140039: step: 1648/533, loss: 0.12932910025119781 2023-01-21 11:12:06.287676: step: 1652/533, loss: 0.0013526440598070621 2023-01-21 11:12:07.462639: step: 1656/533, loss: 0.005809688940644264 2023-01-21 11:12:08.627829: step: 1660/533, loss: 0.0036900995764881372 2023-01-21 11:12:09.811963: step: 1664/533, loss: 0.03878364711999893 2023-01-21 11:12:10.991370: step: 1668/533, loss: 0.0022054819855839014 2023-01-21 11:12:12.161035: step: 1672/533, loss: 0.03097858466207981 2023-01-21 11:12:13.390013: step: 1676/533, loss: 0.04583339765667915 2023-01-21 11:12:14.628239: step: 1680/533, loss: 0.04856610298156738 2023-01-21 11:12:15.778362: step: 1684/533, loss: 0.1337287873029709 2023-01-21 11:12:16.955135: step: 1688/533, loss: 0.015915585681796074 2023-01-21 11:12:18.173907: step: 1692/533, loss: 0.01108570210635662 2023-01-21 11:12:19.340616: step: 1696/533, loss: 0.003035640809684992 2023-01-21 11:12:20.489634: step: 1700/533, loss: 0.04025306925177574 2023-01-21 11:12:21.649901: step: 1704/533, loss: 0.0033143043983727694 2023-01-21 11:12:22.823199: step: 1708/533, loss: 0.012143706902861595 2023-01-21 11:12:24.000982: step: 1712/533, loss: 0.0008431434980593622 2023-01-21 11:12:25.136463: step: 1716/533, loss: 0.00021071435185149312 2023-01-21 11:12:26.298304: step: 1720/533, loss: 0.04438972845673561 2023-01-21 11:12:27.505076: step: 1724/533, loss: 0.10260956734418869 2023-01-21 11:12:28.674812: step: 1728/533, loss: 0.003399467561393976 2023-01-21 11:12:29.857737: step: 1732/533, loss: 0.03218822553753853 2023-01-21 11:12:31.036137: step: 1736/533, loss: 0.013613224029541016 2023-01-21 11:12:32.243783: step: 1740/533, loss: 0.023714255541563034 2023-01-21 11:12:33.407573: step: 1744/533, loss: 0.013700771145522594 2023-01-21 11:12:34.610843: step: 1748/533, loss: 0.026090431958436966 2023-01-21 11:12:35.770874: step: 1752/533, loss: 0.004889106843620539 2023-01-21 11:12:36.946264: step: 1756/533, loss: 0.01535043679177761 2023-01-21 11:12:38.101983: step: 1760/533, loss: 0.010417843237519264 2023-01-21 11:12:39.290796: step: 1764/533, loss: 0.07286892831325531 2023-01-21 11:12:40.483516: step: 1768/533, loss: 0.05481967702507973 2023-01-21 11:12:41.642269: step: 1772/533, loss: 0.05034580081701279 2023-01-21 11:12:42.859955: step: 1776/533, loss: 0.012636279687285423 2023-01-21 11:12:44.070596: step: 1780/533, loss: 0.021080780774354935 2023-01-21 11:12:45.246023: step: 1784/533, loss: 0.009203148074448109 2023-01-21 11:12:46.412126: step: 1788/533, loss: 0.02357616275548935 2023-01-21 11:12:47.566906: step: 1792/533, loss: 0.020754529163241386 2023-01-21 11:12:48.772181: step: 1796/533, loss: 0.012844180688261986 2023-01-21 11:12:50.067909: step: 1800/533, loss: 0.0021793365012854338 2023-01-21 11:12:51.222970: step: 1804/533, loss: 0.013502311892807484 2023-01-21 11:12:52.475440: step: 1808/533, loss: 0.02083413116633892 2023-01-21 11:12:53.663782: step: 1812/533, loss: 0.006758308503776789 2023-01-21 11:12:54.893597: step: 1816/533, loss: 0.08245468139648438 2023-01-21 11:12:56.078408: step: 1820/533, loss: 0.029354475438594818 2023-01-21 11:12:57.302041: step: 1824/533, loss: 0.01232290267944336 2023-01-21 11:12:58.507432: step: 1828/533, loss: 0.08789655566215515 2023-01-21 11:12:59.717579: step: 1832/533, loss: 0.05191917344927788 2023-01-21 11:13:00.917798: step: 1836/533, loss: 0.015374278649687767 2023-01-21 11:13:02.082517: step: 1840/533, loss: 0.031821396201848984 2023-01-21 11:13:03.284142: step: 1844/533, loss: 0.005908298306167126 2023-01-21 11:13:04.469756: step: 1848/533, loss: 0.05773887783288956 2023-01-21 11:13:05.644320: step: 1852/533, loss: 0.004272079560905695 2023-01-21 11:13:06.843483: step: 1856/533, loss: 0.23922786116600037 2023-01-21 11:13:08.067684: step: 1860/533, loss: 0.05212083086371422 2023-01-21 11:13:09.231934: step: 1864/533, loss: 0.06520719826221466 2023-01-21 11:13:10.471603: step: 1868/533, loss: 0.07926884293556213 2023-01-21 11:13:11.692562: step: 1872/533, loss: 0.014132505282759666 2023-01-21 11:13:12.926842: step: 1876/533, loss: 0.015330505557358265 2023-01-21 11:13:14.129294: step: 1880/533, loss: 0.023017311468720436 2023-01-21 11:13:15.316130: step: 1884/533, loss: 0.018812984228134155 2023-01-21 11:13:16.477302: step: 1888/533, loss: 0.014664220623672009 2023-01-21 11:13:17.717039: step: 1892/533, loss: 0.0025262832641601562 2023-01-21 11:13:18.953743: step: 1896/533, loss: 0.0036348344292491674 2023-01-21 11:13:20.149252: step: 1900/533, loss: 0.058869458734989166 2023-01-21 11:13:21.368916: step: 1904/533, loss: 0.05035886913537979 2023-01-21 11:13:22.582331: step: 1908/533, loss: 0.01423492468893528 2023-01-21 11:13:23.737977: step: 1912/533, loss: 0.00346794119104743 2023-01-21 11:13:24.932379: step: 1916/533, loss: 0.0018219948979094625 2023-01-21 11:13:26.197121: step: 1920/533, loss: 0.0005479812971316278 2023-01-21 11:13:27.396998: step: 1924/533, loss: 0.07553939521312714 2023-01-21 11:13:28.591796: step: 1928/533, loss: 0.017013169825077057 2023-01-21 11:13:29.769281: step: 1932/533, loss: 0.011327933520078659 2023-01-21 11:13:31.016428: step: 1936/533, loss: 0.07743445038795471 2023-01-21 11:13:32.243842: step: 1940/533, loss: 0.12663136422634125 2023-01-21 11:13:33.442823: step: 1944/533, loss: 0.009558677673339844 2023-01-21 11:13:34.686957: step: 1948/533, loss: 0.03538475185632706 2023-01-21 11:13:35.840202: step: 1952/533, loss: 0.0012446403270587325 2023-01-21 11:13:37.068645: step: 1956/533, loss: 0.025933455675840378 2023-01-21 11:13:38.256254: step: 1960/533, loss: 0.0037683488335460424 2023-01-21 11:13:39.423995: step: 1964/533, loss: 0.023306941613554955 2023-01-21 11:13:40.542182: step: 1968/533, loss: 0.010752391070127487 2023-01-21 11:13:41.743002: step: 1972/533, loss: 0.004403495695441961 2023-01-21 11:13:42.939371: step: 1976/533, loss: 0.021851349622011185 2023-01-21 11:13:44.138398: step: 1980/533, loss: 0.022672273218631744 2023-01-21 11:13:45.365651: step: 1984/533, loss: 0.0005634308326989412 2023-01-21 11:13:46.535737: step: 1988/533, loss: 0.02189350128173828 2023-01-21 11:13:47.755102: step: 1992/533, loss: 0.00907678622752428 2023-01-21 11:13:48.979418: step: 1996/533, loss: 0.06172523647546768 2023-01-21 11:13:50.164027: step: 2000/533, loss: 0.04457101970911026 2023-01-21 11:13:51.397261: step: 2004/533, loss: 0.058938220143318176 2023-01-21 11:13:52.573427: step: 2008/533, loss: 0.0027115345001220703 2023-01-21 11:13:53.783543: step: 2012/533, loss: 0.08951206505298615 2023-01-21 11:13:54.935278: step: 2016/533, loss: 0.005914401728659868 2023-01-21 11:13:56.134583: step: 2020/533, loss: 0.024036791175603867 2023-01-21 11:13:57.340922: step: 2024/533, loss: 0.07169260829687119 2023-01-21 11:13:58.490513: step: 2028/533, loss: 0.02427082136273384 2023-01-21 11:13:59.641285: step: 2032/533, loss: 0.09004479646682739 2023-01-21 11:14:00.854349: step: 2036/533, loss: 0.011097240261733532 2023-01-21 11:14:02.080199: step: 2040/533, loss: 0.015466880984604359 2023-01-21 11:14:03.249563: step: 2044/533, loss: 0.0009699821239337325 2023-01-21 11:14:04.463907: step: 2048/533, loss: 0.022615719586610794 2023-01-21 11:14:05.672682: step: 2052/533, loss: 0.02037983015179634 2023-01-21 11:14:06.858938: step: 2056/533, loss: 0.025789640843868256 2023-01-21 11:14:08.052673: step: 2060/533, loss: 0.03392486646771431 2023-01-21 11:14:09.223004: step: 2064/533, loss: 0.08136949688196182 2023-01-21 11:14:10.395830: step: 2068/533, loss: 0.024516774341464043 2023-01-21 11:14:11.563869: step: 2072/533, loss: 0.007665443699806929 2023-01-21 11:14:12.752931: step: 2076/533, loss: 0.02284068986773491 2023-01-21 11:14:13.925176: step: 2080/533, loss: 0.007357692811638117 2023-01-21 11:14:15.087448: step: 2084/533, loss: 0.00414619455114007 2023-01-21 11:14:16.253380: step: 2088/533, loss: 0.017296601086854935 2023-01-21 11:14:17.438514: step: 2092/533, loss: 0.008288383483886719 2023-01-21 11:14:18.613915: step: 2096/533, loss: 0.024480629712343216 2023-01-21 11:14:19.786527: step: 2100/533, loss: 0.062108803540468216 2023-01-21 11:14:21.001664: step: 2104/533, loss: 0.05815162882208824 2023-01-21 11:14:22.184632: step: 2108/533, loss: 0.023702431470155716 2023-01-21 11:14:23.326749: step: 2112/533, loss: 0.01562356948852539 2023-01-21 11:14:24.507158: step: 2116/533, loss: 0.0021365166176110506 2023-01-21 11:14:25.670588: step: 2120/533, loss: 0.06192522495985031 2023-01-21 11:14:26.831310: step: 2124/533, loss: 0.042099762707948685 2023-01-21 11:14:28.030670: step: 2128/533, loss: 0.014585400000214577 2023-01-21 11:14:29.217277: step: 2132/533, loss: 0.03098588064312935 ================================================== Loss: 0.046 -------------------- Dev: {'event': {'p': 0.6091954022988506, 'r': 0.7762982689747004, 'f1': 0.6826697892271664}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Test: {'event': {'p': 0.6437185929648241, 'r': 0.8190537084398977, 'f1': 0.7208778840742824}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} 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': 14} Korean: {'event': {'p': 0.6, 'r': 0.5238095238095238, 'f1': 0.559322033898305}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Russian: {'event': {'p': 0.5, 'r': 0.6111111111111112, 'f1': 0.55}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 11:15:11.112034: step: 4/533, loss: 0.032935142517089844 2023-01-21 11:15:12.323397: step: 8/533, loss: 0.20816020667552948 2023-01-21 11:15:13.486643: step: 12/533, loss: 0.010228538885712624 2023-01-21 11:15:14.656916: step: 16/533, loss: 0.00011982917931163684 2023-01-21 11:15:15.836800: step: 20/533, loss: 0.05382366105914116 2023-01-21 11:15:17.025725: step: 24/533, loss: 0.024899769574403763 2023-01-21 11:15:18.216991: step: 28/533, loss: 0.012189102359116077 2023-01-21 11:15:19.426665: step: 32/533, loss: 0.014241504482924938 2023-01-21 11:15:20.648248: step: 36/533, loss: 0.037456318736076355 2023-01-21 11:15:21.842766: step: 40/533, loss: 0.00794839859008789 2023-01-21 11:15:23.041752: step: 44/533, loss: 0.012491417117416859 2023-01-21 11:15:24.206653: step: 48/533, loss: 0.012137413024902344 2023-01-21 11:15:25.383994: step: 52/533, loss: 0.006480109877884388 2023-01-21 11:15:26.595146: step: 56/533, loss: 0.0018396377563476562 2023-01-21 11:15:27.777129: step: 60/533, loss: 0.03527842089533806 2023-01-21 11:15:28.931033: step: 64/533, loss: 0.028742078691720963 2023-01-21 11:15:30.139280: step: 68/533, loss: 0.0031180381774902344 2023-01-21 11:15:31.321468: step: 72/533, loss: 0.007898425683379173 2023-01-21 11:15:32.497962: step: 76/533, loss: 0.01277923583984375 2023-01-21 11:15:33.646647: step: 80/533, loss: 0.013762188144028187 2023-01-21 11:15:34.808688: step: 84/533, loss: 0.00080957415048033 2023-01-21 11:15:35.988423: step: 88/533, loss: 0.004513836000114679 2023-01-21 11:15:37.191156: step: 92/533, loss: 0.006750058848410845 2023-01-21 11:15:38.360692: step: 96/533, loss: 0.0025457381270825863 2023-01-21 11:15:39.511574: step: 100/533, loss: 0.01690845564007759 2023-01-21 11:15:40.684580: step: 104/533, loss: 0.00952682550996542 2023-01-21 11:15:41.922099: step: 108/533, loss: 0.5799604654312134 2023-01-21 11:15:43.139042: step: 112/533, loss: 0.027771949768066406 2023-01-21 11:15:44.295917: step: 116/533, loss: 0.01718606986105442 2023-01-21 11:15:45.473981: step: 120/533, loss: 0.014300156384706497 2023-01-21 11:15:46.631656: step: 124/533, loss: 0.029419898986816406 2023-01-21 11:15:47.809919: step: 128/533, loss: 0.00897455308586359 2023-01-21 11:15:48.989409: step: 132/533, loss: 0.025732040405273438 2023-01-21 11:15:50.206318: step: 136/533, loss: 0.021947670727968216 2023-01-21 11:15:51.377875: step: 140/533, loss: 0.08651962131261826 2023-01-21 11:15:52.626534: step: 144/533, loss: 0.11216086894273758 2023-01-21 11:15:53.814454: step: 148/533, loss: 0.03270130231976509 2023-01-21 11:15:54.952475: step: 152/533, loss: 0.006653499323874712 2023-01-21 11:15:56.131987: step: 156/533, loss: 0.04458275064826012 2023-01-21 11:15:57.273531: step: 160/533, loss: 0.0029197693802416325 2023-01-21 11:15:58.506954: step: 164/533, loss: 0.00010404587374068797 2023-01-21 11:15:59.675118: step: 168/533, loss: 0.006921195890754461 2023-01-21 11:16:00.872963: step: 172/533, loss: 0.015159130096435547 2023-01-21 11:16:02.087895: step: 176/533, loss: 0.053437136113643646 2023-01-21 11:16:03.288834: step: 180/533, loss: 0.010609244927763939 2023-01-21 11:16:04.440145: step: 184/533, loss: 0.00041809084359556437 2023-01-21 11:16:05.639222: step: 188/533, loss: 0.06620364636182785 2023-01-21 11:16:06.855547: step: 192/533, loss: 0.03919067233800888 2023-01-21 11:16:08.060713: step: 196/533, loss: 0.008757306262850761 2023-01-21 11:16:09.216221: step: 200/533, loss: 0.0003612518194131553 2023-01-21 11:16:10.392925: step: 204/533, loss: 0.05581951141357422 2023-01-21 11:16:11.580284: step: 208/533, loss: 0.0005365372053347528 2023-01-21 11:16:12.793261: step: 212/533, loss: 0.0021191597916185856 2023-01-21 11:16:13.976549: step: 216/533, loss: 0.0073677063919603825 2023-01-21 11:16:15.154289: step: 220/533, loss: 0.023686980828642845 2023-01-21 11:16:16.320070: step: 224/533, loss: 0.04401359707117081 2023-01-21 11:16:17.531006: step: 228/533, loss: 0.005532217212021351 2023-01-21 11:16:18.702171: step: 232/533, loss: 0.024250458925962448 2023-01-21 11:16:19.868902: step: 236/533, loss: 0.032639503479003906 2023-01-21 11:16:21.085433: step: 240/533, loss: 0.04319038614630699 2023-01-21 11:16:22.221601: step: 244/533, loss: 0.004134106449782848 2023-01-21 11:16:23.404731: step: 248/533, loss: 0.0012115002609789371 2023-01-21 11:16:24.619465: step: 252/533, loss: 0.08519001305103302 2023-01-21 11:16:25.810877: step: 256/533, loss: 0.022072412073612213 2023-01-21 11:16:27.044674: step: 260/533, loss: 0.024069691076874733 2023-01-21 11:16:28.218932: step: 264/533, loss: 0.02736072614789009 2023-01-21 11:16:29.459651: step: 268/533, loss: 0.03606872260570526 2023-01-21 11:16:30.663121: step: 272/533, loss: 0.04927101358771324 2023-01-21 11:16:31.828210: step: 276/533, loss: 0.0025703429710119963 2023-01-21 11:16:32.995880: step: 280/533, loss: 0.12955741584300995 2023-01-21 11:16:34.191828: step: 284/533, loss: 0.01866474188864231 2023-01-21 11:16:35.362421: step: 288/533, loss: 0.009340954013168812 2023-01-21 11:16:36.555304: step: 292/533, loss: 0.01603412628173828 2023-01-21 11:16:37.760846: step: 296/533, loss: 0.022847937420010567 2023-01-21 11:16:38.911768: step: 300/533, loss: 0.037826765328645706 2023-01-21 11:16:40.103620: step: 304/533, loss: 0.0024423152208328247 2023-01-21 11:16:41.299406: step: 308/533, loss: 0.04450559616088867 2023-01-21 11:16:42.489698: step: 312/533, loss: 0.05001239851117134 2023-01-21 11:16:43.664236: step: 316/533, loss: 0.0037585259415209293 2023-01-21 11:16:44.845533: step: 320/533, loss: 0.013702106662094593 2023-01-21 11:16:46.037433: step: 324/533, loss: 0.004167270381003618 2023-01-21 11:16:47.218851: step: 328/533, loss: 0.026240825653076172 2023-01-21 11:16:48.424809: step: 332/533, loss: 0.006727981381118298 2023-01-21 11:16:49.623315: step: 336/533, loss: 0.04866666719317436 2023-01-21 11:16:50.808235: step: 340/533, loss: 0.04574894905090332 2023-01-21 11:16:51.977962: step: 344/533, loss: 0.05682888254523277 2023-01-21 11:16:53.156800: step: 348/533, loss: 0.04966564103960991 2023-01-21 11:16:54.355254: step: 352/533, loss: 0.012169087305665016 2023-01-21 11:16:55.504903: step: 356/533, loss: 0.05115833133459091 2023-01-21 11:16:56.632104: step: 360/533, loss: 0.003943538758903742 2023-01-21 11:16:57.810224: step: 364/533, loss: 0.010458231903612614 2023-01-21 11:16:58.962796: step: 368/533, loss: 0.029062747955322266 2023-01-21 11:17:00.149903: step: 372/533, loss: 0.08163433521986008 2023-01-21 11:17:01.332080: step: 376/533, loss: 0.003917217254638672 2023-01-21 11:17:02.543630: step: 380/533, loss: 0.043286897242069244 2023-01-21 11:17:03.786625: step: 384/533, loss: 0.0018850326305255294 2023-01-21 11:17:04.996007: step: 388/533, loss: 0.00558547955006361 2023-01-21 11:17:06.158188: step: 392/533, loss: 0.011851501651108265 2023-01-21 11:17:07.354259: step: 396/533, loss: 0.006362342741340399 2023-01-21 11:17:08.484088: step: 400/533, loss: 0.13735751807689667 2023-01-21 11:17:09.699685: step: 404/533, loss: 0.05071859061717987 2023-01-21 11:17:10.855409: step: 408/533, loss: 0.01619955338537693 2023-01-21 11:17:12.035451: step: 412/533, loss: 0.019878387451171875 2023-01-21 11:17:13.247219: step: 416/533, loss: 0.052910711616277695 2023-01-21 11:17:14.399399: step: 420/533, loss: 0.00669331569224596 2023-01-21 11:17:15.582422: step: 424/533, loss: 0.026935769245028496 2023-01-21 11:17:16.748230: step: 428/533, loss: 0.0018197059398517013 2023-01-21 11:17:17.952012: step: 432/533, loss: 0.5512388348579407 2023-01-21 11:17:19.124087: step: 436/533, loss: 0.0006567001692019403 2023-01-21 11:17:20.290906: step: 440/533, loss: 0.022025395184755325 2023-01-21 11:17:21.482611: step: 444/533, loss: 0.004497576039284468 2023-01-21 11:17:22.665048: step: 448/533, loss: 0.006063556764274836 2023-01-21 11:17:23.849660: step: 452/533, loss: 0.23575058579444885 2023-01-21 11:17:25.026328: step: 456/533, loss: 0.020454170182347298 2023-01-21 11:17:26.214704: step: 460/533, loss: 0.011699771508574486 2023-01-21 11:17:27.381411: step: 464/533, loss: 0.01693134382367134 2023-01-21 11:17:28.555941: step: 468/533, loss: 0.01447305642068386 2023-01-21 11:17:29.701382: step: 472/533, loss: 0.00260753626935184 2023-01-21 11:17:30.886900: step: 476/533, loss: 0.006367969326674938 2023-01-21 11:17:32.153493: step: 480/533, loss: 0.0010057449107989669 2023-01-21 11:17:33.400976: step: 484/533, loss: 0.001689910888671875 2023-01-21 11:17:34.586889: step: 488/533, loss: 0.0008909225580282509 2023-01-21 11:17:35.750326: step: 492/533, loss: 0.03721008449792862 2023-01-21 11:17:36.979197: step: 496/533, loss: 0.0230318084359169 2023-01-21 11:17:38.189052: step: 500/533, loss: 0.02500610426068306 2023-01-21 11:17:39.421246: step: 504/533, loss: 0.011840629391372204 2023-01-21 11:17:40.609291: step: 508/533, loss: 0.0052495962008833885 2023-01-21 11:17:41.773237: step: 512/533, loss: 0.0015333176124840975 2023-01-21 11:17:42.963932: step: 516/533, loss: 0.0017180442810058594 2023-01-21 11:17:44.164719: step: 520/533, loss: 0.00868835486471653 2023-01-21 11:17:45.349837: step: 524/533, loss: 0.09515104442834854 2023-01-21 11:17:46.540263: step: 528/533, loss: 0.051025390625 2023-01-21 11:17:47.690296: step: 532/533, loss: 0.005434084217995405 2023-01-21 11:17:48.885162: step: 536/533, loss: 0.026868535205721855 2023-01-21 11:17:50.072776: step: 540/533, loss: 0.035859107971191406 2023-01-21 11:17:51.292764: step: 544/533, loss: 0.06793337315320969 2023-01-21 11:17:52.472297: step: 548/533, loss: 0.021899700164794922 2023-01-21 11:17:53.660665: step: 552/533, loss: 0.5063167810440063 2023-01-21 11:17:54.853887: step: 556/533, loss: 0.015806769952178 2023-01-21 11:17:55.991362: step: 560/533, loss: 0.008835411630570889 2023-01-21 11:17:57.144715: step: 564/533, loss: 0.01644115522503853 2023-01-21 11:17:58.304445: step: 568/533, loss: 0.0074520111083984375 2023-01-21 11:17:59.483716: step: 572/533, loss: 0.011236190795898438 2023-01-21 11:18:00.684049: step: 576/533, loss: 0.00153017055708915 2023-01-21 11:18:01.863046: step: 580/533, loss: 0.009859466925263405 2023-01-21 11:18:03.051673: step: 584/533, loss: 0.027907848358154297 2023-01-21 11:18:04.209633: step: 588/533, loss: 0.029274415224790573 2023-01-21 11:18:05.398050: step: 592/533, loss: 0.051645852625370026 2023-01-21 11:18:06.568350: step: 596/533, loss: 0.01381602231413126 2023-01-21 11:18:07.732611: step: 600/533, loss: 0.009810996241867542 2023-01-21 11:18:08.901940: step: 604/533, loss: 1.0538564920425415 2023-01-21 11:18:10.040900: step: 608/533, loss: 0.03410739824175835 2023-01-21 11:18:11.211985: step: 612/533, loss: 0.020374584943056107 2023-01-21 11:18:12.459493: step: 616/533, loss: 0.010302925482392311 2023-01-21 11:18:13.639842: step: 620/533, loss: 0.10747719556093216 2023-01-21 11:18:14.815109: step: 624/533, loss: 0.0033604621421545744 2023-01-21 11:18:15.956985: step: 628/533, loss: 0.005664443597197533 2023-01-21 11:18:17.149049: step: 632/533, loss: 0.010132789611816406 2023-01-21 11:18:18.334434: step: 636/533, loss: 0.007479286286979914 2023-01-21 11:18:19.537495: step: 640/533, loss: 0.00010354519326938316 2023-01-21 11:18:20.726771: step: 644/533, loss: 3.647804260253906e-05 2023-01-21 11:18:21.919045: step: 648/533, loss: 0.0018508911598473787 2023-01-21 11:18:23.110033: step: 652/533, loss: 0.025847626850008965 2023-01-21 11:18:24.267490: step: 656/533, loss: 0.001186108565889299 2023-01-21 11:18:25.517799: step: 660/533, loss: 0.009152794256806374 2023-01-21 11:18:26.692959: step: 664/533, loss: 0.0001338958682026714 2023-01-21 11:18:27.880683: step: 668/533, loss: 0.029302407056093216 2023-01-21 11:18:29.060166: step: 672/533, loss: 0.0017469406593590975 2023-01-21 11:18:30.239428: step: 676/533, loss: 0.05545921251177788 2023-01-21 11:18:31.398229: step: 680/533, loss: 0.012060356326401234 2023-01-21 11:18:32.572870: step: 684/533, loss: 0.03998775780200958 2023-01-21 11:18:33.729206: step: 688/533, loss: 0.01690850406885147 2023-01-21 11:18:34.919989: step: 692/533, loss: 0.04131384193897247 2023-01-21 11:18:36.107351: step: 696/533, loss: 0.05395827442407608 2023-01-21 11:18:37.307242: step: 700/533, loss: 0.0005521774291992188 2023-01-21 11:18:38.506642: step: 704/533, loss: 0.018268680199980736 2023-01-21 11:18:39.719465: step: 708/533, loss: 0.0668523758649826 2023-01-21 11:18:40.894653: step: 712/533, loss: 0.001416969345882535 2023-01-21 11:18:42.108138: step: 716/533, loss: 0.019535064697265625 2023-01-21 11:18:43.304983: step: 720/533, loss: 0.08260726928710938 2023-01-21 11:18:44.476005: step: 724/533, loss: 0.0035907744895666838 2023-01-21 11:18:45.629119: step: 728/533, loss: 0.028116608038544655 2023-01-21 11:18:46.747016: step: 732/533, loss: 0.052947234362363815 2023-01-21 11:18:47.983613: step: 736/533, loss: 0.007778358645737171 2023-01-21 11:18:49.155417: step: 740/533, loss: 0.05410642549395561 2023-01-21 11:18:50.333106: step: 744/533, loss: 0.003478527069091797 2023-01-21 11:18:51.516570: step: 748/533, loss: 0.07205858081579208 2023-01-21 11:18:52.720417: step: 752/533, loss: 0.011920833960175514 2023-01-21 11:18:53.878752: step: 756/533, loss: 0.0007012844434939325 2023-01-21 11:18:55.047959: step: 760/533, loss: 0.01123061217367649 2023-01-21 11:18:56.241129: step: 764/533, loss: 0.08117532730102539 2023-01-21 11:18:57.435487: step: 768/533, loss: 0.005683326628059149 2023-01-21 11:18:58.604257: step: 772/533, loss: 0.021115398034453392 2023-01-21 11:18:59.749848: step: 776/533, loss: 0.01758279651403427 2023-01-21 11:19:01.013580: step: 780/533, loss: 0.001708984375 2023-01-21 11:19:02.216700: step: 784/533, loss: 0.02628192864358425 2023-01-21 11:19:03.388754: step: 788/533, loss: 0.00523422984406352 2023-01-21 11:19:04.552144: step: 792/533, loss: 0.005763435736298561 2023-01-21 11:19:05.725502: step: 796/533, loss: 0.012892818078398705 2023-01-21 11:19:06.903899: step: 800/533, loss: 0.0067414757795631886 2023-01-21 11:19:08.081606: step: 804/533, loss: 0.04166240990161896 2023-01-21 11:19:09.272414: step: 808/533, loss: 0.042984962463378906 2023-01-21 11:19:10.444819: step: 812/533, loss: 0.02524395100772381 2023-01-21 11:19:11.621421: step: 816/533, loss: 0.06830673664808273 2023-01-21 11:19:12.859039: step: 820/533, loss: 0.018540572375059128 2023-01-21 11:19:14.052968: step: 824/533, loss: 0.012559700757265091 2023-01-21 11:19:15.251931: step: 828/533, loss: 0.06579142063856125 2023-01-21 11:19:16.453723: step: 832/533, loss: 0.030162906274199486 2023-01-21 11:19:17.657113: step: 836/533, loss: 0.031333159655332565 2023-01-21 11:19:18.816889: step: 840/533, loss: 0.01032400131225586 2023-01-21 11:19:19.976313: step: 844/533, loss: 0.005405331030488014 2023-01-21 11:19:21.158907: step: 848/533, loss: 0.004728603642433882 2023-01-21 11:19:22.365581: step: 852/533, loss: 0.03109598159790039 2023-01-21 11:19:23.590449: step: 856/533, loss: 0.003120040986686945 2023-01-21 11:19:24.824476: step: 860/533, loss: 0.010536766611039639 2023-01-21 11:19:25.985829: step: 864/533, loss: 0.00366802210919559 2023-01-21 11:19:27.154084: step: 868/533, loss: 0.0921415314078331 2023-01-21 11:19:28.359934: step: 872/533, loss: 0.011440182104706764 2023-01-21 11:19:29.541295: step: 876/533, loss: 0.0031838417053222656 2023-01-21 11:19:30.714423: step: 880/533, loss: 0.011308335699141026 2023-01-21 11:19:31.999412: step: 884/533, loss: 0.01524534821510315 2023-01-21 11:19:33.206565: step: 888/533, loss: 0.029624175280332565 2023-01-21 11:19:34.383558: step: 892/533, loss: 0.021938515827059746 2023-01-21 11:19:35.573767: step: 896/533, loss: 0.03240537643432617 2023-01-21 11:19:36.774231: step: 900/533, loss: 0.055516816675662994 2023-01-21 11:19:37.985181: step: 904/533, loss: 0.03161277621984482 2023-01-21 11:19:39.156184: step: 908/533, loss: 0.022112274542450905 2023-01-21 11:19:40.354088: step: 912/533, loss: 9.52720656641759e-05 2023-01-21 11:19:41.554448: step: 916/533, loss: 0.0011038780212402344 2023-01-21 11:19:42.727300: step: 920/533, loss: 0.023032188415527344 2023-01-21 11:19:43.904740: step: 924/533, loss: 0.005356502719223499 2023-01-21 11:19:45.119654: step: 928/533, loss: 0.03163165971636772 2023-01-21 11:19:46.342569: step: 932/533, loss: 0.007514477241784334 2023-01-21 11:19:47.529898: step: 936/533, loss: 0.015540122985839844 2023-01-21 11:19:48.697864: step: 940/533, loss: 0.11558723449707031 2023-01-21 11:19:49.858317: step: 944/533, loss: 0.002057838486507535 2023-01-21 11:19:51.017264: step: 948/533, loss: 0.01934661902487278 2023-01-21 11:19:52.277117: step: 952/533, loss: 0.11768823117017746 2023-01-21 11:19:53.469396: step: 956/533, loss: 0.020032405853271484 2023-01-21 11:19:54.675511: step: 960/533, loss: 0.25579720735549927 2023-01-21 11:19:55.878633: step: 964/533, loss: 0.010859108529984951 2023-01-21 11:19:57.079977: step: 968/533, loss: 0.06880789250135422 2023-01-21 11:19:58.265682: step: 972/533, loss: 0.018734503537416458 2023-01-21 11:19:59.439645: step: 976/533, loss: 0.2389226108789444 2023-01-21 11:20:00.628212: step: 980/533, loss: 0.05467681959271431 2023-01-21 11:20:01.818463: step: 984/533, loss: 0.09200669080018997 2023-01-21 11:20:02.998149: step: 988/533, loss: 0.03554544597864151 2023-01-21 11:20:04.161749: step: 992/533, loss: 0.017479050904512405 2023-01-21 11:20:05.329540: step: 996/533, loss: 0.016999052837491035 2023-01-21 11:20:06.513143: step: 1000/533, loss: 0.006008339114487171 2023-01-21 11:20:07.683251: step: 1004/533, loss: 0.025869179517030716 2023-01-21 11:20:08.872891: step: 1008/533, loss: 0.02622394822537899 2023-01-21 11:20:10.091019: step: 1012/533, loss: 0.0006391525384970009 2023-01-21 11:20:11.260710: step: 1016/533, loss: 0.014249706640839577 2023-01-21 11:20:12.440221: step: 1020/533, loss: 0.04812910407781601 2023-01-21 11:20:13.609140: step: 1024/533, loss: 0.03662566840648651 2023-01-21 11:20:14.786671: step: 1028/533, loss: 0.010272979736328125 2023-01-21 11:20:15.973234: step: 1032/533, loss: 0.0028142931405454874 2023-01-21 11:20:17.156907: step: 1036/533, loss: 0.05822286382317543 2023-01-21 11:20:18.324063: step: 1040/533, loss: 0.030423451215028763 2023-01-21 11:20:19.516420: step: 1044/533, loss: 0.023008251562714577 2023-01-21 11:20:20.707660: step: 1048/533, loss: 0.00150384905282408 2023-01-21 11:20:21.913608: step: 1052/533, loss: 0.10438661277294159 2023-01-21 11:20:23.100638: step: 1056/533, loss: 0.02017350308597088 2023-01-21 11:20:24.316725: step: 1060/533, loss: 0.01994161680340767 2023-01-21 11:20:25.461951: step: 1064/533, loss: 0.04570809006690979 2023-01-21 11:20:26.635087: step: 1068/533, loss: 0.01996765099465847 2023-01-21 11:20:27.786010: step: 1072/533, loss: 0.0008236885187216103 2023-01-21 11:20:29.005802: step: 1076/533, loss: 0.01607356034219265 2023-01-21 11:20:30.180290: step: 1080/533, loss: 0.06569354236125946 2023-01-21 11:20:31.366876: step: 1084/533, loss: 0.5582121014595032 2023-01-21 11:20:32.504881: step: 1088/533, loss: 0.006081867031753063 2023-01-21 11:20:33.702782: step: 1092/533, loss: 0.0364801399409771 2023-01-21 11:20:34.872292: step: 1096/533, loss: 0.002303695771843195 2023-01-21 11:20:36.067565: step: 1100/533, loss: 0.05634870380163193 2023-01-21 11:20:37.252923: step: 1104/533, loss: 0.07834739983081818 2023-01-21 11:20:38.432676: step: 1108/533, loss: 0.015307998284697533 2023-01-21 11:20:39.611926: step: 1112/533, loss: 0.02333064004778862 2023-01-21 11:20:40.806561: step: 1116/533, loss: 0.0076385498978197575 2023-01-21 11:20:42.003382: step: 1120/533, loss: 0.042043305933475494 2023-01-21 11:20:43.193538: step: 1124/533, loss: 0.0008339881896972656 2023-01-21 11:20:44.311778: step: 1128/533, loss: 0.00040121079655364156 2023-01-21 11:20:45.539548: step: 1132/533, loss: 0.019464731216430664 2023-01-21 11:20:46.720411: step: 1136/533, loss: 0.002195644425228238 2023-01-21 11:20:47.944441: step: 1140/533, loss: 0.034500885754823685 2023-01-21 11:20:49.130511: step: 1144/533, loss: 0.03498554229736328 2023-01-21 11:20:50.287112: step: 1148/533, loss: 0.0002364158717682585 2023-01-21 11:20:51.533726: step: 1152/533, loss: 0.004755592439323664 2023-01-21 11:20:52.736841: step: 1156/533, loss: 0.014267158694565296 2023-01-21 11:20:53.908439: step: 1160/533, loss: 0.08406372368335724 2023-01-21 11:20:55.081641: step: 1164/533, loss: 0.001859378768131137 2023-01-21 11:20:56.278774: step: 1168/533, loss: 0.007318973541259766 2023-01-21 11:20:57.461223: step: 1172/533, loss: 0.008948421105742455 2023-01-21 11:20:58.633404: step: 1176/533, loss: 0.025196265429258347 2023-01-21 11:20:59.831310: step: 1180/533, loss: 0.00012750625319313258 2023-01-21 11:21:00.963104: step: 1184/533, loss: 0.0464697889983654 2023-01-21 11:21:02.129068: step: 1188/533, loss: 0.00582809466868639 2023-01-21 11:21:03.328665: step: 1192/533, loss: 0.016617584973573685 2023-01-21 11:21:04.522816: step: 1196/533, loss: 0.009745216928422451 2023-01-21 11:21:05.720704: step: 1200/533, loss: 0.0041717528365552425 2023-01-21 11:21:06.896744: step: 1204/533, loss: 0.0827278196811676 2023-01-21 11:21:08.081077: step: 1208/533, loss: 0.005058670416474342 2023-01-21 11:21:09.258190: step: 1212/533, loss: 0.04741477966308594 2023-01-21 11:21:10.457855: step: 1216/533, loss: 0.04038963466882706 2023-01-21 11:21:11.587688: step: 1220/533, loss: 0.007138824090361595 2023-01-21 11:21:12.781532: step: 1224/533, loss: 0.00027523041353560984 2023-01-21 11:21:13.951247: step: 1228/533, loss: 0.0023010254371911287 2023-01-21 11:21:15.132880: step: 1232/533, loss: 0.06627855449914932 2023-01-21 11:21:16.326757: step: 1236/533, loss: 0.021790886297822 2023-01-21 11:21:17.552476: step: 1240/533, loss: 0.0030642985366284847 2023-01-21 11:21:18.758908: step: 1244/533, loss: 0.08760223537683487 2023-01-21 11:21:19.990926: step: 1248/533, loss: 0.01812133751809597 2023-01-21 11:21:21.198103: step: 1252/533, loss: 0.008239174261689186 2023-01-21 11:21:22.370767: step: 1256/533, loss: 0.005572319030761719 2023-01-21 11:21:23.574537: step: 1260/533, loss: 0.0014656067360192537 2023-01-21 11:21:24.762635: step: 1264/533, loss: 0.019694900140166283 2023-01-21 11:21:25.945430: step: 1268/533, loss: 0.008084679022431374 2023-01-21 11:21:27.161462: step: 1272/533, loss: 0.003276062197983265 2023-01-21 11:21:28.360769: step: 1276/533, loss: 0.05052213370800018 2023-01-21 11:21:29.520316: step: 1280/533, loss: 0.03057537041604519 2023-01-21 11:21:30.746781: step: 1284/533, loss: 0.021741202101111412 2023-01-21 11:21:31.930450: step: 1288/533, loss: 0.00624427804723382 2023-01-21 11:21:33.148779: step: 1292/533, loss: 0.03257598727941513 2023-01-21 11:21:34.320609: step: 1296/533, loss: 0.008803748525679111 2023-01-21 11:21:35.475712: step: 1300/533, loss: 0.00012302398681640625 2023-01-21 11:21:36.664162: step: 1304/533, loss: 0.028060341253876686 2023-01-21 11:21:37.845313: step: 1308/533, loss: 0.051899246871471405 2023-01-21 11:21:39.037916: step: 1312/533, loss: 0.06038370355963707 2023-01-21 11:21:40.242065: step: 1316/533, loss: 0.0061799525283277035 2023-01-21 11:21:41.414728: step: 1320/533, loss: 0.018686868250370026 2023-01-21 11:21:42.575811: step: 1324/533, loss: 0.0014800071949139237 2023-01-21 11:21:43.769138: step: 1328/533, loss: 0.01181702595204115 2023-01-21 11:21:44.962658: step: 1332/533, loss: 0.026486776769161224 2023-01-21 11:21:46.122433: step: 1336/533, loss: 0.015202522277832031 2023-01-21 11:21:47.311168: step: 1340/533, loss: 0.03638286888599396 2023-01-21 11:21:48.484589: step: 1344/533, loss: 0.051151469349861145 2023-01-21 11:21:49.697528: step: 1348/533, loss: 0.012768936343491077 2023-01-21 11:21:50.929354: step: 1352/533, loss: 3.23295607813634e-05 2023-01-21 11:21:52.107395: step: 1356/533, loss: 0.02454528957605362 2023-01-21 11:21:53.234102: step: 1360/533, loss: 0.022385787218809128 2023-01-21 11:21:54.418382: step: 1364/533, loss: 0.09749583899974823 2023-01-21 11:21:55.563886: step: 1368/533, loss: 0.013975143432617188 2023-01-21 11:21:56.743430: step: 1372/533, loss: 0.046065568923950195 2023-01-21 11:21:57.920062: step: 1376/533, loss: 0.0007600784301757812 2023-01-21 11:21:59.089751: step: 1380/533, loss: 0.002143955323845148 2023-01-21 11:22:00.249306: step: 1384/533, loss: 0.011771773919463158 2023-01-21 11:22:01.427286: step: 1388/533, loss: 0.008600425906479359 2023-01-21 11:22:02.597956: step: 1392/533, loss: 0.02682333067059517 2023-01-21 11:22:03.768771: step: 1396/533, loss: 0.004392004106193781 2023-01-21 11:22:04.967971: step: 1400/533, loss: 0.02995305135846138 2023-01-21 11:22:06.157486: step: 1404/533, loss: 0.024753950536251068 2023-01-21 11:22:07.340014: step: 1408/533, loss: 0.6111812591552734 2023-01-21 11:22:08.508628: step: 1412/533, loss: 0.009926986880600452 2023-01-21 11:22:09.711659: step: 1416/533, loss: 0.012932014651596546 2023-01-21 11:22:10.887320: step: 1420/533, loss: 0.40571194887161255 2023-01-21 11:22:12.062394: step: 1424/533, loss: 0.060767367482185364 2023-01-21 11:22:13.208740: step: 1428/533, loss: 0.005069160833954811 2023-01-21 11:22:14.351151: step: 1432/533, loss: 0.0010920524364337325 2023-01-21 11:22:15.523052: step: 1436/533, loss: 0.02393188513815403 2023-01-21 11:22:16.727130: step: 1440/533, loss: 0.001622200128622353 2023-01-21 11:22:17.934079: step: 1444/533, loss: 0.1077723503112793 2023-01-21 11:22:19.124973: step: 1448/533, loss: 0.0002967834589071572 2023-01-21 11:22:20.315408: step: 1452/533, loss: 0.004766655154526234 2023-01-21 11:22:21.571592: step: 1456/533, loss: 0.0057089803740382195 2023-01-21 11:22:22.776923: step: 1460/533, loss: 0.011112308129668236 2023-01-21 11:22:23.965237: step: 1464/533, loss: 0.8716068267822266 2023-01-21 11:22:25.150354: step: 1468/533, loss: 0.0208574291318655 2023-01-21 11:22:26.335159: step: 1472/533, loss: 0.0012716293567791581 2023-01-21 11:22:27.512513: step: 1476/533, loss: 0.0011721611954271793 2023-01-21 11:22:28.704920: step: 1480/533, loss: 0.041300203651189804 2023-01-21 11:22:29.871390: step: 1484/533, loss: 0.023172473534941673 2023-01-21 11:22:31.096363: step: 1488/533, loss: 0.19410498440265656 2023-01-21 11:22:32.267725: step: 1492/533, loss: 0.0006723880651406944 2023-01-21 11:22:33.427560: step: 1496/533, loss: 0.07472310215234756 2023-01-21 11:22:34.611608: step: 1500/533, loss: 0.0003501892206259072 2023-01-21 11:22:35.838390: step: 1504/533, loss: 0.005684948060661554 2023-01-21 11:22:37.023139: step: 1508/533, loss: 0.04314479976892471 2023-01-21 11:22:38.200897: step: 1512/533, loss: 0.039985354989767075 2023-01-21 11:22:39.360959: step: 1516/533, loss: 0.000335502641974017 2023-01-21 11:22:40.564191: step: 1520/533, loss: 0.04457368701696396 2023-01-21 11:22:41.747954: step: 1524/533, loss: 0.05116824805736542 2023-01-21 11:22:43.023661: step: 1528/533, loss: 0.007270717993378639 2023-01-21 11:22:44.239979: step: 1532/533, loss: 0.009715080261230469 2023-01-21 11:22:45.433427: step: 1536/533, loss: 0.07018737494945526 2023-01-21 11:22:46.612776: step: 1540/533, loss: 0.10267038643360138 2023-01-21 11:22:47.811404: step: 1544/533, loss: 0.004651117138564587 2023-01-21 11:22:48.948834: step: 1548/533, loss: 0.0390382744371891 2023-01-21 11:22:50.152950: step: 1552/533, loss: 0.05172806233167648 2023-01-21 11:22:51.323134: step: 1556/533, loss: 0.6601153612136841 2023-01-21 11:22:52.546958: step: 1560/533, loss: 0.6721023321151733 2023-01-21 11:22:53.715092: step: 1564/533, loss: 0.02232213132083416 2023-01-21 11:22:54.899054: step: 1568/533, loss: 0.05605268478393555 2023-01-21 11:22:56.074626: step: 1572/533, loss: 0.08810672909021378 2023-01-21 11:22:57.247647: step: 1576/533, loss: 0.008544350042939186 2023-01-21 11:22:58.439400: step: 1580/533, loss: 0.001749324845150113 2023-01-21 11:22:59.601799: step: 1584/533, loss: 0.039250753819942474 2023-01-21 11:23:00.808696: step: 1588/533, loss: 0.03836536407470703 2023-01-21 11:23:01.975119: step: 1592/533, loss: 0.012895394116640091 2023-01-21 11:23:03.161152: step: 1596/533, loss: 0.036333274096250534 2023-01-21 11:23:04.344173: step: 1600/533, loss: 0.09430017322301865 2023-01-21 11:23:05.524678: step: 1604/533, loss: 0.02545313909649849 2023-01-21 11:23:06.710608: step: 1608/533, loss: 0.014666581526398659 2023-01-21 11:23:07.921917: step: 1612/533, loss: 0.01267700269818306 2023-01-21 11:23:09.103423: step: 1616/533, loss: 0.002629518508911133 2023-01-21 11:23:10.256577: step: 1620/533, loss: 0.0003755807993002236 2023-01-21 11:23:11.445420: step: 1624/533, loss: 0.015447616577148438 2023-01-21 11:23:12.622341: step: 1628/533, loss: 0.06693296879529953 2023-01-21 11:23:13.779364: step: 1632/533, loss: 0.026526546105742455 2023-01-21 11:23:15.002727: step: 1636/533, loss: 0.0023128509055823088 2023-01-21 11:23:16.212402: step: 1640/533, loss: 0.030338764190673828 2023-01-21 11:23:17.405029: step: 1644/533, loss: 0.0010441303020343184 2023-01-21 11:23:18.601427: step: 1648/533, loss: 0.002232170198112726 2023-01-21 11:23:19.773532: step: 1652/533, loss: 0.05335855484008789 2023-01-21 11:23:20.954285: step: 1656/533, loss: 0.048605918884277344 2023-01-21 11:23:22.125000: step: 1660/533, loss: 0.023540209978818893 2023-01-21 11:23:23.312092: step: 1664/533, loss: 0.027623845264315605 2023-01-21 11:23:24.472291: step: 1668/533, loss: 0.05308714136481285 2023-01-21 11:23:25.690479: step: 1672/533, loss: 0.0011747360695153475 2023-01-21 11:23:26.832901: step: 1676/533, loss: 0.03023710288107395 2023-01-21 11:23:27.970062: step: 1680/533, loss: 0.018737316131591797 2023-01-21 11:23:29.144956: step: 1684/533, loss: 0.006157779600471258 2023-01-21 11:23:30.319390: step: 1688/533, loss: 0.032965995371341705 2023-01-21 11:23:31.472240: step: 1692/533, loss: 0.03528919070959091 2023-01-21 11:23:32.632996: step: 1696/533, loss: 0.024686051532626152 2023-01-21 11:23:33.819001: step: 1700/533, loss: 0.02487964555621147 2023-01-21 11:23:35.012277: step: 1704/533, loss: 0.026212597265839577 2023-01-21 11:23:36.179415: step: 1708/533, loss: 0.04367341846227646 2023-01-21 11:23:37.319383: step: 1712/533, loss: 0.002299404004588723 2023-01-21 11:23:38.526862: step: 1716/533, loss: 0.018983840942382812 2023-01-21 11:23:39.668895: step: 1720/533, loss: 0.019336320459842682 2023-01-21 11:23:40.839518: step: 1724/533, loss: 0.06380148231983185 2023-01-21 11:23:42.043645: step: 1728/533, loss: 0.026281356811523438 2023-01-21 11:23:43.200996: step: 1732/533, loss: 0.04699697345495224 2023-01-21 11:23:44.450708: step: 1736/533, loss: 0.03758583217859268 2023-01-21 11:23:45.655326: step: 1740/533, loss: 0.00016431808762717992 2023-01-21 11:23:46.856147: step: 1744/533, loss: 0.011332893744111061 2023-01-21 11:23:48.034000: step: 1748/533, loss: 0.0074713705107569695 2023-01-21 11:23:49.209008: step: 1752/533, loss: 0.04449405521154404 2023-01-21 11:23:50.362440: step: 1756/533, loss: 0.020090175792574883 2023-01-21 11:23:51.543932: step: 1760/533, loss: 0.0004707336483988911 2023-01-21 11:23:52.768926: step: 1764/533, loss: 0.008679008111357689 2023-01-21 11:23:53.960008: step: 1768/533, loss: 0.002693748567253351 2023-01-21 11:23:55.128794: step: 1772/533, loss: 0.00797424279153347 2023-01-21 11:23:56.331179: step: 1776/533, loss: 0.055773161351680756 2023-01-21 11:23:57.506679: step: 1780/533, loss: 0.011396408081054688 2023-01-21 11:23:58.676311: step: 1784/533, loss: 0.0009837150573730469 2023-01-21 11:23:59.829506: step: 1788/533, loss: 0.001769256661646068 2023-01-21 11:24:01.007286: step: 1792/533, loss: 0.20864161849021912 2023-01-21 11:24:02.194095: step: 1796/533, loss: 0.0011324883671477437 2023-01-21 11:24:03.375100: step: 1800/533, loss: 0.048406124114990234 2023-01-21 11:24:04.558891: step: 1804/533, loss: 0.04061460494995117 2023-01-21 11:24:05.754860: step: 1808/533, loss: 0.06837115436792374 2023-01-21 11:24:06.939757: step: 1812/533, loss: 0.019831180572509766 2023-01-21 11:24:08.126701: step: 1816/533, loss: 0.0007309913635253906 2023-01-21 11:24:09.275512: step: 1820/533, loss: 0.03796567767858505 2023-01-21 11:24:10.422309: step: 1824/533, loss: 0.003537082578986883 2023-01-21 11:24:11.599067: step: 1828/533, loss: 0.05063915252685547 2023-01-21 11:24:12.768850: step: 1832/533, loss: 0.06044836342334747 2023-01-21 11:24:13.963397: step: 1836/533, loss: 0.053933046758174896 2023-01-21 11:24:15.150798: step: 1840/533, loss: 0.07410397380590439 2023-01-21 11:24:16.333691: step: 1844/533, loss: 0.04561617225408554 2023-01-21 11:24:17.495061: step: 1848/533, loss: 0.001814508461393416 2023-01-21 11:24:18.697196: step: 1852/533, loss: 0.013540947809815407 2023-01-21 11:24:19.837379: step: 1856/533, loss: 0.04883613437414169 2023-01-21 11:24:20.995760: step: 1860/533, loss: 0.006354331970214844 2023-01-21 11:24:22.161778: step: 1864/533, loss: 0.03805875778198242 2023-01-21 11:24:23.359775: step: 1868/533, loss: 0.0682777389883995 2023-01-21 11:24:24.568893: step: 1872/533, loss: 0.04222317039966583 2023-01-21 11:24:25.749005: step: 1876/533, loss: 0.044712066650390625 2023-01-21 11:24:26.969962: step: 1880/533, loss: 0.008061408996582031 2023-01-21 11:24:28.147257: step: 1884/533, loss: 0.026677321642637253 2023-01-21 11:24:29.275837: step: 1888/533, loss: 0.008876991458237171 2023-01-21 11:24:30.463777: step: 1892/533, loss: 0.0035416604951024055 2023-01-21 11:24:31.678398: step: 1896/533, loss: 0.15971823036670685 2023-01-21 11:24:32.857146: step: 1900/533, loss: 0.015170670114457607 2023-01-21 11:24:34.060368: step: 1904/533, loss: 0.016233444213867188 2023-01-21 11:24:35.239197: step: 1908/533, loss: 0.0054069519974291325 2023-01-21 11:24:36.429696: step: 1912/533, loss: 0.002009582705795765 2023-01-21 11:24:37.623670: step: 1916/533, loss: 0.0009664535755291581 2023-01-21 11:24:38.799666: step: 1920/533, loss: 0.0011308669345453382 2023-01-21 11:24:39.950529: step: 1924/533, loss: 0.008218956179916859 2023-01-21 11:24:41.145483: step: 1928/533, loss: 0.07188782840967178 2023-01-21 11:24:42.311715: step: 1932/533, loss: 0.038410499691963196 2023-01-21 11:24:43.473430: step: 1936/533, loss: 0.005018759053200483 2023-01-21 11:24:44.639650: step: 1940/533, loss: 0.17368750274181366 2023-01-21 11:24:45.763926: step: 1944/533, loss: 0.004603862762451172 2023-01-21 11:24:46.983174: step: 1948/533, loss: 0.006214046385139227 2023-01-21 11:24:48.184708: step: 1952/533, loss: 0.010112953372299671 2023-01-21 11:24:49.380932: step: 1956/533, loss: 0.04880419000983238 2023-01-21 11:24:50.529013: step: 1960/533, loss: 0.004610681440681219 2023-01-21 11:24:51.755527: step: 1964/533, loss: 0.004082346335053444 2023-01-21 11:24:52.937445: step: 1968/533, loss: 0.004262733273208141 2023-01-21 11:24:54.131574: step: 1972/533, loss: 0.03580818697810173 2023-01-21 11:24:55.316890: step: 1976/533, loss: 0.003840160323306918 2023-01-21 11:24:56.459787: step: 1980/533, loss: 0.0005839825025759637 2023-01-21 11:24:57.658602: step: 1984/533, loss: 0.09142742305994034 2023-01-21 11:24:58.828783: step: 1988/533, loss: 0.000720596348401159 2023-01-21 11:24:59.982608: step: 1992/533, loss: 0.00953598041087389 2023-01-21 11:25:01.158354: step: 1996/533, loss: 0.05358543246984482 2023-01-21 11:25:02.320261: step: 2000/533, loss: 0.05969047546386719 2023-01-21 11:25:03.501620: step: 2004/533, loss: 0.012916183099150658 2023-01-21 11:25:04.724147: step: 2008/533, loss: 0.01000747736543417 2023-01-21 11:25:05.916404: step: 2012/533, loss: 0.0004625797155313194 2023-01-21 11:25:07.069420: step: 2016/533, loss: 0.008774567395448685 2023-01-21 11:25:08.242084: step: 2020/533, loss: 0.013058471493422985 2023-01-21 11:25:09.412512: step: 2024/533, loss: 0.02618112601339817 2023-01-21 11:25:10.627036: step: 2028/533, loss: 0.06427936255931854 2023-01-21 11:25:11.862667: step: 2032/533, loss: 0.032117556780576706 2023-01-21 11:25:13.078645: step: 2036/533, loss: 0.08823671191930771 2023-01-21 11:25:14.239663: step: 2040/533, loss: 0.03191203996539116 2023-01-21 11:25:15.419613: step: 2044/533, loss: 0.03502722084522247 2023-01-21 11:25:16.598360: step: 2048/533, loss: 0.0031945230439305305 2023-01-21 11:25:17.801380: step: 2052/533, loss: 0.11333017796278 2023-01-21 11:25:19.010879: step: 2056/533, loss: 0.0027659893967211246 2023-01-21 11:25:20.190690: step: 2060/533, loss: 1.087188684323337e-05 2023-01-21 11:25:21.391928: step: 2064/533, loss: 0.02297971211373806 2023-01-21 11:25:22.644428: step: 2068/533, loss: 0.033625029027462006 2023-01-21 11:25:23.857536: step: 2072/533, loss: 0.00562896765768528 2023-01-21 11:25:25.021990: step: 2076/533, loss: 0.02445383183658123 2023-01-21 11:25:26.195267: step: 2080/533, loss: 0.004440879914909601 2023-01-21 11:25:27.360702: step: 2084/533, loss: 0.03657379373908043 2023-01-21 11:25:28.546615: step: 2088/533, loss: 0.013322449289262295 2023-01-21 11:25:29.732354: step: 2092/533, loss: 0.024021148681640625 2023-01-21 11:25:30.932836: step: 2096/533, loss: 0.001678562257438898 2023-01-21 11:25:32.115379: step: 2100/533, loss: 0.0024990083184093237 2023-01-21 11:25:33.282799: step: 2104/533, loss: 0.0009291649330407381 2023-01-21 11:25:34.472966: step: 2108/533, loss: 0.04841213300824165 2023-01-21 11:25:35.680386: step: 2112/533, loss: 0.05562658607959747 2023-01-21 11:25:36.878072: step: 2116/533, loss: 0.004183387849479914 2023-01-21 11:25:38.112392: step: 2120/533, loss: 0.02273120917379856 2023-01-21 11:25:39.283132: step: 2124/533, loss: 0.022293664515018463 2023-01-21 11:25:40.476746: step: 2128/533, loss: 0.06163139268755913 2023-01-21 11:25:41.684878: step: 2132/533, loss: 0.035263922065496445 ================================================== Loss: 0.039 -------------------- Dev: {'event': {'p': 0.6156387665198237, 'r': 0.7443408788282291, 'f1': 0.6738999397227246}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Test: {'event': {'p': 0.6574556830031283, 'r': 0.8062659846547314, 'f1': 0.7242963813900059}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} 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': 15} 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': 15} Russian: {'event': {'p': 0.34285714285714286, 'r': 0.3333333333333333, 'f1': 0.3380281690140845}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 11:26:23.316776: step: 4/533, loss: 0.008470917120575905 2023-01-21 11:26:24.481770: step: 8/533, loss: 0.0465204231441021 2023-01-21 11:26:25.649386: step: 12/533, loss: 0.006655311677604914 2023-01-21 11:26:26.861152: step: 16/533, loss: 0.0010618210071697831 2023-01-21 11:26:28.055393: step: 20/533, loss: 0.15059979259967804 2023-01-21 11:26:29.220193: step: 24/533, loss: 0.23148241639137268 2023-01-21 11:26:30.355050: step: 28/533, loss: 0.021610165014863014 2023-01-21 11:26:31.529644: step: 32/533, loss: 0.030868148431181908 2023-01-21 11:26:32.711587: step: 36/533, loss: 0.03508281707763672 2023-01-21 11:26:33.862119: step: 40/533, loss: 0.001137542654760182 2023-01-21 11:26:35.087733: step: 44/533, loss: 0.020200539380311966 2023-01-21 11:26:36.246211: step: 48/533, loss: 0.011140108108520508 2023-01-21 11:26:37.422466: step: 52/533, loss: 0.028125381097197533 2023-01-21 11:26:38.588670: step: 56/533, loss: 0.003948402591049671 2023-01-21 11:26:39.773245: step: 60/533, loss: 0.0028722763527184725 2023-01-21 11:26:40.932817: step: 64/533, loss: 0.02617664448916912 2023-01-21 11:26:42.161781: step: 68/533, loss: 0.009011936374008656 2023-01-21 11:26:43.324627: step: 72/533, loss: 0.003072547959163785 2023-01-21 11:26:44.494704: step: 76/533, loss: 0.003905010409653187 2023-01-21 11:26:45.643418: step: 80/533, loss: 0.31844061613082886 2023-01-21 11:26:46.809353: step: 84/533, loss: 0.0043312073685228825 2023-01-21 11:26:47.956381: step: 88/533, loss: 0.0011698722373694181 2023-01-21 11:26:49.157894: step: 92/533, loss: 0.002591466996818781 2023-01-21 11:26:50.361415: step: 96/533, loss: 0.04778652265667915 2023-01-21 11:26:51.582102: step: 100/533, loss: 0.08303088694810867 2023-01-21 11:26:52.723028: step: 104/533, loss: 0.0005508899921551347 2023-01-21 11:26:53.926935: step: 108/533, loss: 0.09066527336835861 2023-01-21 11:26:55.091762: step: 112/533, loss: 0.027950001880526543 2023-01-21 11:26:56.264372: step: 116/533, loss: 0.07192688435316086 2023-01-21 11:26:57.499290: step: 120/533, loss: 0.011118126101791859 2023-01-21 11:26:58.667622: step: 124/533, loss: 0.01728353463113308 2023-01-21 11:26:59.805689: step: 128/533, loss: 0.011571216396987438 2023-01-21 11:27:01.055748: step: 132/533, loss: 0.027283286675810814 2023-01-21 11:27:02.283386: step: 136/533, loss: 0.018111610785126686 2023-01-21 11:27:03.445808: step: 140/533, loss: 0.04141692817211151 2023-01-21 11:27:04.640224: step: 144/533, loss: 0.0010954856406897306 2023-01-21 11:27:05.835550: step: 148/533, loss: 0.00011777877807617188 2023-01-21 11:27:07.025148: step: 152/533, loss: 0.019156837835907936 2023-01-21 11:27:08.203346: step: 156/533, loss: 0.0023279192391783 2023-01-21 11:27:09.335678: step: 160/533, loss: 0.0026392461732029915 2023-01-21 11:27:10.490792: step: 164/533, loss: 0.01635732688009739 2023-01-21 11:27:11.660450: step: 168/533, loss: 0.0048995972611010075 2023-01-21 11:27:12.839203: step: 172/533, loss: 0.01196365337818861 2023-01-21 11:27:13.997420: step: 176/533, loss: 0.0156586654484272 2023-01-21 11:27:15.187616: step: 180/533, loss: 0.0022554397583007812 2023-01-21 11:27:16.357823: step: 184/533, loss: 0.0028951645363122225 2023-01-21 11:27:17.527239: step: 188/533, loss: 0.00784082431346178 2023-01-21 11:27:18.732713: step: 192/533, loss: 0.01173181552439928 2023-01-21 11:27:19.876547: step: 196/533, loss: 0.0005043029668740928 2023-01-21 11:27:21.089565: step: 200/533, loss: 0.05573310703039169 2023-01-21 11:27:22.220062: step: 204/533, loss: 0.0011247635120525956 2023-01-21 11:27:23.414137: step: 208/533, loss: 0.035033416002988815 2023-01-21 11:27:24.575985: step: 212/533, loss: 0.009194565936923027 2023-01-21 11:27:25.778300: step: 216/533, loss: 0.04939556121826172 2023-01-21 11:27:26.958234: step: 220/533, loss: 0.008978605270385742 2023-01-21 11:27:28.118959: step: 224/533, loss: 0.026622582226991653 2023-01-21 11:27:29.357064: step: 228/533, loss: 0.07035829871892929 2023-01-21 11:27:30.508315: step: 232/533, loss: 0.023929214105010033 2023-01-21 11:27:31.693045: step: 236/533, loss: 0.0026329518295824528 2023-01-21 11:27:32.869931: step: 240/533, loss: 0.03444099426269531 2023-01-21 11:27:34.068471: step: 244/533, loss: 0.03175296634435654 2023-01-21 11:27:35.267558: step: 248/533, loss: 0.033638764172792435 2023-01-21 11:27:36.416559: step: 252/533, loss: 0.011321449652314186 2023-01-21 11:27:37.598785: step: 256/533, loss: 0.001935768174007535 2023-01-21 11:27:38.793675: step: 260/533, loss: 0.026566125452518463 2023-01-21 11:27:39.969152: step: 264/533, loss: 0.0031630517914891243 2023-01-21 11:27:41.162804: step: 268/533, loss: 0.00252532958984375 2023-01-21 11:27:42.379863: step: 272/533, loss: 0.031397342681884766 2023-01-21 11:27:43.543528: step: 276/533, loss: 0.00027408599271439016 2023-01-21 11:27:44.726470: step: 280/533, loss: 0.029116153717041016 2023-01-21 11:27:45.948524: step: 284/533, loss: 0.013803101144731045 2023-01-21 11:27:47.085991: step: 288/533, loss: 0.0073143006302416325 2023-01-21 11:27:48.255626: step: 292/533, loss: 0.013009835034608841 2023-01-21 11:27:49.424778: step: 296/533, loss: 0.003967761993408203 2023-01-21 11:27:50.621052: step: 300/533, loss: 0.0006698608631268144 2023-01-21 11:27:51.808960: step: 304/533, loss: 0.1048440933227539 2023-01-21 11:27:52.993309: step: 308/533, loss: 0.01865367963910103 2023-01-21 11:27:54.142382: step: 312/533, loss: 0.0009140968322753906 2023-01-21 11:27:55.307368: step: 316/533, loss: 0.022235143929719925 2023-01-21 11:27:56.499371: step: 320/533, loss: 0.036298468708992004 2023-01-21 11:27:57.721291: step: 324/533, loss: 0.019890213385224342 2023-01-21 11:27:58.926273: step: 328/533, loss: 0.010746574029326439 2023-01-21 11:28:00.114217: step: 332/533, loss: 0.13982105255126953 2023-01-21 11:28:01.331504: step: 336/533, loss: 0.010132885538041592 2023-01-21 11:28:02.503643: step: 340/533, loss: 1.6117095583467744e-05 2023-01-21 11:28:03.669505: step: 344/533, loss: 0.007548713590949774 2023-01-21 11:28:04.839439: step: 348/533, loss: 0.02690095826983452 2023-01-21 11:28:05.980893: step: 352/533, loss: 0.0007046699756756425 2023-01-21 11:28:07.157092: step: 356/533, loss: 0.020682239904999733 2023-01-21 11:28:08.343762: step: 360/533, loss: 0.0009099007584154606 2023-01-21 11:28:09.531334: step: 364/533, loss: 0.0006016731495037675 2023-01-21 11:28:10.727869: step: 368/533, loss: 0.018793772906064987 2023-01-21 11:28:11.920623: step: 372/533, loss: 0.08054371178150177 2023-01-21 11:28:13.078485: step: 376/533, loss: 0.0019276619423180819 2023-01-21 11:28:14.212236: step: 380/533, loss: 0.02097911946475506 2023-01-21 11:28:15.374950: step: 384/533, loss: 0.02664213255047798 2023-01-21 11:28:16.565758: step: 388/533, loss: 0.002897262806072831 2023-01-21 11:28:17.769415: step: 392/533, loss: 0.032498933374881744 2023-01-21 11:28:18.923439: step: 396/533, loss: 0.003563452046364546 2023-01-21 11:28:20.062059: step: 400/533, loss: 0.008996676653623581 2023-01-21 11:28:21.220993: step: 404/533, loss: 0.004610633943229914 2023-01-21 11:28:22.353446: step: 408/533, loss: 0.02196197584271431 2023-01-21 11:28:23.568214: step: 412/533, loss: 0.00047779083251953125 2023-01-21 11:28:24.746244: step: 416/533, loss: 0.0006655693287029862 2023-01-21 11:28:25.930850: step: 420/533, loss: 0.005791568662971258 2023-01-21 11:28:27.129444: step: 424/533, loss: 0.04312610626220703 2023-01-21 11:28:28.343787: step: 428/533, loss: 0.01964435540139675 2023-01-21 11:28:29.533360: step: 432/533, loss: 0.02618255652487278 2023-01-21 11:28:30.716736: step: 436/533, loss: 0.0003886222839355469 2023-01-21 11:28:31.928731: step: 440/533, loss: 0.018010282889008522 2023-01-21 11:28:33.075725: step: 444/533, loss: 0.00016660690016578883 2023-01-21 11:28:34.218826: step: 448/533, loss: 0.030162811279296875 2023-01-21 11:28:35.422401: step: 452/533, loss: 0.033753298223018646 2023-01-21 11:28:36.574216: step: 456/533, loss: 0.005419731140136719 2023-01-21 11:28:37.729620: step: 460/533, loss: 0.09558341652154922 2023-01-21 11:28:38.944104: step: 464/533, loss: 0.049311257898807526 2023-01-21 11:28:40.115284: step: 468/533, loss: 0.008684826083481312 2023-01-21 11:28:41.305872: step: 472/533, loss: 0.004789256956428289 2023-01-21 11:28:42.466798: step: 476/533, loss: 0.045457080006599426 2023-01-21 11:28:43.617785: step: 480/533, loss: 0.16615954041481018 2023-01-21 11:28:44.786078: step: 484/533, loss: 0.011584663763642311 2023-01-21 11:28:45.935236: step: 488/533, loss: 0.09564094245433807 2023-01-21 11:28:47.079073: step: 492/533, loss: 0.003418445587158203 2023-01-21 11:28:48.266627: step: 496/533, loss: 0.006530619226396084 2023-01-21 11:28:49.425309: step: 500/533, loss: 0.042191032320261 2023-01-21 11:28:50.604712: step: 504/533, loss: 0.018095016479492188 2023-01-21 11:28:51.806951: step: 508/533, loss: 0.028162287548184395 2023-01-21 11:28:53.044450: step: 512/533, loss: 0.0015274047618731856 2023-01-21 11:28:54.229183: step: 516/533, loss: 0.01014175359159708 2023-01-21 11:28:55.411405: step: 520/533, loss: 0.002201271243393421 2023-01-21 11:28:56.605481: step: 524/533, loss: 0.009698057547211647 2023-01-21 11:28:57.807793: step: 528/533, loss: 0.000864005065523088 2023-01-21 11:28:58.963507: step: 532/533, loss: 0.07200102508068085 2023-01-21 11:29:00.171070: step: 536/533, loss: 1.6215065717697144 2023-01-21 11:29:01.317167: step: 540/533, loss: 0.0024261474609375 2023-01-21 11:29:02.488998: step: 544/533, loss: 0.00011811256990768015 2023-01-21 11:29:03.683428: step: 548/533, loss: 0.00041313175461255014 2023-01-21 11:29:04.906399: step: 552/533, loss: 0.0007096290355548263 2023-01-21 11:29:06.133926: step: 556/533, loss: 0.023937225341796875 2023-01-21 11:29:07.308979: step: 560/533, loss: 0.00608139019459486 2023-01-21 11:29:08.489016: step: 564/533, loss: 0.01667614094913006 2023-01-21 11:29:09.660954: step: 568/533, loss: 0.04230823740363121 2023-01-21 11:29:10.841997: step: 572/533, loss: 0.005451011937111616 2023-01-21 11:29:12.030298: step: 576/533, loss: 0.0013081550132483244 2023-01-21 11:29:13.215561: step: 580/533, loss: 0.008057022467255592 2023-01-21 11:29:14.394244: step: 584/533, loss: 0.012000751681625843 2023-01-21 11:29:15.550456: step: 588/533, loss: 0.011235046200454235 2023-01-21 11:29:16.767731: step: 592/533, loss: 0.003505706787109375 2023-01-21 11:29:17.963481: step: 596/533, loss: 0.14706459641456604 2023-01-21 11:29:19.112011: step: 600/533, loss: 0.008831977844238281 2023-01-21 11:29:20.264532: step: 604/533, loss: 0.004325867164880037 2023-01-21 11:29:21.463219: step: 608/533, loss: 0.011186790652573109 2023-01-21 11:29:22.633510: step: 612/533, loss: 0.04059867933392525 2023-01-21 11:29:23.791401: step: 616/533, loss: 0.0018993377452716231 2023-01-21 11:29:24.913982: step: 620/533, loss: 0.026645947247743607 2023-01-21 11:29:26.092378: step: 624/533, loss: 0.03933725506067276 2023-01-21 11:29:27.267592: step: 628/533, loss: 0.08598098158836365 2023-01-21 11:29:28.448371: step: 632/533, loss: 0.02301044389605522 2023-01-21 11:29:29.654839: step: 636/533, loss: 0.02397918701171875 2023-01-21 11:29:30.835641: step: 640/533, loss: 0.0008863449329510331 2023-01-21 11:29:32.024195: step: 644/533, loss: 0.004969883244484663 2023-01-21 11:29:33.231672: step: 648/533, loss: 0.02785663679242134 2023-01-21 11:29:34.446030: step: 652/533, loss: 0.03374824672937393 2023-01-21 11:29:35.614794: step: 656/533, loss: 0.03947162628173828 2023-01-21 11:29:36.811327: step: 660/533, loss: 0.006059455685317516 2023-01-21 11:29:37.987147: step: 664/533, loss: 0.00020484924607444555 2023-01-21 11:29:39.148536: step: 668/533, loss: 0.010035991668701172 2023-01-21 11:29:40.309113: step: 672/533, loss: 0.0008369446150027215 2023-01-21 11:29:41.457582: step: 676/533, loss: 0.10975103080272675 2023-01-21 11:29:42.648764: step: 680/533, loss: 0.008195066824555397 2023-01-21 11:29:43.806676: step: 684/533, loss: 0.04231100529432297 2023-01-21 11:29:44.977384: step: 688/533, loss: 0.005776023957878351 2023-01-21 11:29:46.161206: step: 692/533, loss: 0.08877818286418915 2023-01-21 11:29:47.351534: step: 696/533, loss: 0.03502826765179634 2023-01-21 11:29:48.557732: step: 700/533, loss: 0.015086794272065163 2023-01-21 11:29:49.750999: step: 704/533, loss: 0.10844888538122177 2023-01-21 11:29:50.914204: step: 708/533, loss: 0.007921409793198109 2023-01-21 11:29:52.088618: step: 712/533, loss: 0.0001663207949604839 2023-01-21 11:29:53.269087: step: 716/533, loss: 0.0853106677532196 2023-01-21 11:29:54.450409: step: 720/533, loss: 0.0296204574406147 2023-01-21 11:29:55.668073: step: 724/533, loss: 0.0007078170892782509 2023-01-21 11:29:56.825604: step: 728/533, loss: 0.018746759742498398 2023-01-21 11:29:58.007792: step: 732/533, loss: 0.0012365340953692794 2023-01-21 11:29:59.192038: step: 736/533, loss: 0.008796310983598232 2023-01-21 11:30:00.365625: step: 740/533, loss: 0.06804495304822922 2023-01-21 11:30:01.552148: step: 744/533, loss: 0.09098707139492035 2023-01-21 11:30:02.739867: step: 748/533, loss: 0.0015560149913653731 2023-01-21 11:30:03.899385: step: 752/533, loss: 0.0029108047019690275 2023-01-21 11:30:05.048645: step: 756/533, loss: 0.021446991711854935 2023-01-21 11:30:06.273025: step: 760/533, loss: 0.00452499371021986 2023-01-21 11:30:07.462639: step: 764/533, loss: 0.015717601403594017 2023-01-21 11:30:08.649726: step: 768/533, loss: 0.06093960255384445 2023-01-21 11:30:09.826970: step: 772/533, loss: 0.022891521453857422 2023-01-21 11:30:11.011698: step: 776/533, loss: 0.010380077175796032 2023-01-21 11:30:12.177615: step: 780/533, loss: 0.00443954486399889 2023-01-21 11:30:13.335481: step: 784/533, loss: 0.011400127783417702 2023-01-21 11:30:14.498076: step: 788/533, loss: 0.05018024891614914 2023-01-21 11:30:15.701174: step: 792/533, loss: 0.002664852188900113 2023-01-21 11:30:16.888657: step: 796/533, loss: 9.74655122263357e-05 2023-01-21 11:30:18.057034: step: 800/533, loss: 0.003055382054299116 2023-01-21 11:30:19.224834: step: 804/533, loss: 0.061838772147893906 2023-01-21 11:30:20.423859: step: 808/533, loss: 0.003964710049331188 2023-01-21 11:30:21.577917: step: 812/533, loss: 0.0037606239784508944 2023-01-21 11:30:22.823940: step: 816/533, loss: 0.03313946723937988 2023-01-21 11:30:24.018892: step: 820/533, loss: 0.019649790599942207 2023-01-21 11:30:25.205369: step: 824/533, loss: 0.03895380347967148 2023-01-21 11:30:26.384778: step: 828/533, loss: 0.026616811752319336 2023-01-21 11:30:27.546581: step: 832/533, loss: 0.010266399942338467 2023-01-21 11:30:28.737444: step: 836/533, loss: 0.011691570281982422 2023-01-21 11:30:29.924458: step: 840/533, loss: 0.07766079902648926 2023-01-21 11:30:31.107975: step: 844/533, loss: 0.00442085275426507 2023-01-21 11:30:32.257376: step: 848/533, loss: 0.01811370998620987 2023-01-21 11:30:33.442222: step: 852/533, loss: 0.05416903644800186 2023-01-21 11:30:34.655087: step: 856/533, loss: 0.03901004418730736 2023-01-21 11:30:35.787286: step: 860/533, loss: 6.341934204101562e-05 2023-01-21 11:30:36.961191: step: 864/533, loss: 0.011152410879731178 2023-01-21 11:30:38.155590: step: 868/533, loss: 0.09143200516700745 2023-01-21 11:30:39.363011: step: 872/533, loss: 0.016480159014463425 2023-01-21 11:30:40.610368: step: 876/533, loss: 0.018391134217381477 2023-01-21 11:30:41.795787: step: 880/533, loss: 0.020070480182766914 2023-01-21 11:30:42.939653: step: 884/533, loss: 0.0003344535653013736 2023-01-21 11:30:44.102040: step: 888/533, loss: 0.027069903910160065 2023-01-21 11:30:45.283675: step: 892/533, loss: 0.03187832981348038 2023-01-21 11:30:46.429911: step: 896/533, loss: 0.00434532156214118 2023-01-21 11:30:47.590899: step: 900/533, loss: 0.004500770475715399 2023-01-21 11:30:48.778393: step: 904/533, loss: 0.0017803192604333162 2023-01-21 11:30:49.971623: step: 908/533, loss: 0.0022199153900146484 2023-01-21 11:30:51.220274: step: 912/533, loss: 0.010034753009676933 2023-01-21 11:30:52.429431: step: 916/533, loss: 0.010270786471664906 2023-01-21 11:30:53.644064: step: 920/533, loss: 0.004616737365722656 2023-01-21 11:30:54.831096: step: 924/533, loss: 0.010496329516172409 2023-01-21 11:30:55.965724: step: 928/533, loss: 0.0013918876647949219 2023-01-21 11:30:57.154406: step: 932/533, loss: 0.03287268057465553 2023-01-21 11:30:58.325563: step: 936/533, loss: 0.01556539535522461 2023-01-21 11:30:59.476626: step: 940/533, loss: 0.05309810861945152 2023-01-21 11:31:00.651645: step: 944/533, loss: 0.0014232636895030737 2023-01-21 11:31:01.800218: step: 948/533, loss: 0.018986130133271217 2023-01-21 11:31:02.976800: step: 952/533, loss: 0.015268420800566673 2023-01-21 11:31:04.148461: step: 956/533, loss: 0.14075346291065216 2023-01-21 11:31:05.320913: step: 960/533, loss: 0.05768918991088867 2023-01-21 11:31:06.442584: step: 964/533, loss: 0.0023357390891760588 2023-01-21 11:31:07.629520: step: 968/533, loss: 0.0012866973411291838 2023-01-21 11:31:08.812024: step: 972/533, loss: 0.007906151004135609 2023-01-21 11:31:09.984440: step: 976/533, loss: 0.08674231171607971 2023-01-21 11:31:11.197158: step: 980/533, loss: 0.039389897137880325 2023-01-21 11:31:12.358533: step: 984/533, loss: 0.009799767285585403 2023-01-21 11:31:13.516500: step: 988/533, loss: 0.025008518248796463 2023-01-21 11:31:14.707286: step: 992/533, loss: 0.01917734183371067 2023-01-21 11:31:15.884108: step: 996/533, loss: 0.01778383180499077 2023-01-21 11:31:17.055223: step: 1000/533, loss: 0.006135463714599609 2023-01-21 11:31:18.233298: step: 1004/533, loss: 0.008789444342255592 2023-01-21 11:31:19.393753: step: 1008/533, loss: 0.026110364124178886 2023-01-21 11:31:20.568548: step: 1012/533, loss: 0.044086359441280365 2023-01-21 11:31:21.741605: step: 1016/533, loss: 0.004754066467285156 2023-01-21 11:31:22.961598: step: 1020/533, loss: 0.03907470777630806 2023-01-21 11:31:24.122948: step: 1024/533, loss: 0.0020124674774706364 2023-01-21 11:31:25.346638: step: 1028/533, loss: 0.04017849266529083 2023-01-21 11:31:26.521114: step: 1032/533, loss: 0.012788104824721813 2023-01-21 11:31:27.677196: step: 1036/533, loss: 0.005555343814194202 2023-01-21 11:31:28.929178: step: 1040/533, loss: 0.004494285676628351 2023-01-21 11:31:30.089049: step: 1044/533, loss: 0.0002636909484863281 2023-01-21 11:31:31.294739: step: 1048/533, loss: 0.005420684814453125 2023-01-21 11:31:32.495209: step: 1052/533, loss: 0.004572534468024969 2023-01-21 11:31:33.646902: step: 1056/533, loss: 0.07770762592554092 2023-01-21 11:31:34.799458: step: 1060/533, loss: 0.014161204919219017 2023-01-21 11:31:35.970329: step: 1064/533, loss: 0.026057055220007896 2023-01-21 11:31:37.150929: step: 1068/533, loss: 7.514953904319555e-05 2023-01-21 11:31:38.384678: step: 1072/533, loss: 0.004003399517387152 2023-01-21 11:31:39.591268: step: 1076/533, loss: 0.01516580581665039 2023-01-21 11:31:40.798559: step: 1080/533, loss: 0.0058991434052586555 2023-01-21 11:31:42.003005: step: 1084/533, loss: 0.02819213829934597 2023-01-21 11:31:43.237603: step: 1088/533, loss: 0.0011773109436035156 2023-01-21 11:31:44.419072: step: 1092/533, loss: 0.020191382616758347 2023-01-21 11:31:45.589250: step: 1096/533, loss: 0.04162616655230522 2023-01-21 11:31:46.777135: step: 1100/533, loss: 0.02677922323346138 2023-01-21 11:31:47.977010: step: 1104/533, loss: 0.0011736869346350431 2023-01-21 11:31:49.168404: step: 1108/533, loss: 0.03926978260278702 2023-01-21 11:31:50.343370: step: 1112/533, loss: 0.02059917524456978 2023-01-21 11:31:51.555948: step: 1116/533, loss: 0.04444694519042969 2023-01-21 11:31:52.766208: step: 1120/533, loss: 0.1583656370639801 2023-01-21 11:31:53.962774: step: 1124/533, loss: 0.0014865398406982422 2023-01-21 11:31:55.132050: step: 1128/533, loss: 0.036760617047548294 2023-01-21 11:31:56.311575: step: 1132/533, loss: 0.017153168097138405 2023-01-21 11:31:57.482189: step: 1136/533, loss: 0.000324058550177142 2023-01-21 11:31:58.622838: step: 1140/533, loss: 0.0009197235340252519 2023-01-21 11:31:59.771147: step: 1144/533, loss: 0.016248513013124466 2023-01-21 11:32:00.961860: step: 1148/533, loss: 0.0032559395767748356 2023-01-21 11:32:02.115988: step: 1152/533, loss: 0.0001277923583984375 2023-01-21 11:32:03.267121: step: 1156/533, loss: 0.18440285325050354 2023-01-21 11:32:04.450802: step: 1160/533, loss: 0.04303378984332085 2023-01-21 11:32:05.678322: step: 1164/533, loss: 0.0405062660574913 2023-01-21 11:32:06.865624: step: 1168/533, loss: 0.01967163011431694 2023-01-21 11:32:08.041229: step: 1172/533, loss: 0.01926717907190323 2023-01-21 11:32:09.221928: step: 1176/533, loss: 0.0012675285106524825 2023-01-21 11:32:10.395454: step: 1180/533, loss: 0.03308048099279404 2023-01-21 11:32:11.580498: step: 1184/533, loss: 0.03112502209842205 2023-01-21 11:32:12.736655: step: 1188/533, loss: 0.0011125088203698397 2023-01-21 11:32:13.888083: step: 1192/533, loss: 0.0049453736282885075 2023-01-21 11:32:15.028063: step: 1196/533, loss: 0.00047240257845260203 2023-01-21 11:32:16.195603: step: 1200/533, loss: 9.841918654274195e-05 2023-01-21 11:32:17.322687: step: 1204/533, loss: 0.00141487130895257 2023-01-21 11:32:18.480028: step: 1208/533, loss: 0.009282112121582031 2023-01-21 11:32:19.679381: step: 1212/533, loss: 0.015368938446044922 2023-01-21 11:32:20.841973: step: 1216/533, loss: 0.0258820541203022 2023-01-21 11:32:22.052489: step: 1220/533, loss: 0.0005972862127237022 2023-01-21 11:32:23.213504: step: 1224/533, loss: 0.002756786299869418 2023-01-21 11:32:24.376477: step: 1228/533, loss: 0.0035709382500499487 2023-01-21 11:32:25.580532: step: 1232/533, loss: 0.005856037139892578 2023-01-21 11:32:26.776227: step: 1236/533, loss: 0.008848332799971104 2023-01-21 11:32:27.967646: step: 1240/533, loss: 0.0033036230597645044 2023-01-21 11:32:29.139602: step: 1244/533, loss: 0.0023911476600915194 2023-01-21 11:32:30.380911: step: 1248/533, loss: 0.012393951416015625 2023-01-21 11:32:31.511501: step: 1252/533, loss: 0.0021102428436279297 2023-01-21 11:32:32.705517: step: 1256/533, loss: 0.008630752563476562 2023-01-21 11:32:33.872773: step: 1260/533, loss: 0.03329286351799965 2023-01-21 11:32:35.051795: step: 1264/533, loss: 0.00894699152559042 2023-01-21 11:32:36.205418: step: 1268/533, loss: 0.07978720963001251 2023-01-21 11:32:37.432125: step: 1272/533, loss: 0.004378318786621094 2023-01-21 11:32:38.607470: step: 1276/533, loss: 0.0025140761863440275 2023-01-21 11:32:39.846456: step: 1280/533, loss: 0.04782400280237198 2023-01-21 11:32:41.045750: step: 1284/533, loss: 0.0032793045975267887 2023-01-21 11:32:42.240146: step: 1288/533, loss: 0.0032743930350989103 2023-01-21 11:32:43.424271: step: 1292/533, loss: 0.03894253075122833 2023-01-21 11:32:44.625664: step: 1296/533, loss: 0.001995944883674383 2023-01-21 11:32:45.801134: step: 1300/533, loss: 0.016109563410282135 2023-01-21 11:32:46.966517: step: 1304/533, loss: 0.008039665408432484 2023-01-21 11:32:48.146588: step: 1308/533, loss: 0.023818302899599075 2023-01-21 11:32:49.319872: step: 1312/533, loss: 0.06377563625574112 2023-01-21 11:32:50.508348: step: 1316/533, loss: 0.0023560526315122843 2023-01-21 11:32:51.700753: step: 1320/533, loss: 0.0037340165581554174 2023-01-21 11:32:52.845830: step: 1324/533, loss: 0.008930588141083717 2023-01-21 11:32:54.025129: step: 1328/533, loss: 0.008122109808027744 2023-01-21 11:32:55.255338: step: 1332/533, loss: 0.0007085800170898438 2023-01-21 11:32:56.458452: step: 1336/533, loss: 0.00172767648473382 2023-01-21 11:32:57.666336: step: 1340/533, loss: 0.008748245425522327 2023-01-21 11:32:58.808327: step: 1344/533, loss: 0.016813278198242188 2023-01-21 11:32:59.958385: step: 1348/533, loss: 0.019668007269501686 2023-01-21 11:33:01.157664: step: 1352/533, loss: 0.00862894020974636 2023-01-21 11:33:02.339384: step: 1356/533, loss: 0.03913841396570206 2023-01-21 11:33:03.552126: step: 1360/533, loss: 0.005063533782958984 2023-01-21 11:33:04.715429: step: 1364/533, loss: 0.004771709442138672 2023-01-21 11:33:05.851593: step: 1368/533, loss: 0.010756874457001686 2023-01-21 11:33:07.020624: step: 1372/533, loss: 0.004741382785141468 2023-01-21 11:33:08.198259: step: 1376/533, loss: 0.0027503492310643196 2023-01-21 11:33:09.384621: step: 1380/533, loss: 0.011645698919892311 2023-01-21 11:33:10.572228: step: 1384/533, loss: 0.016314219683408737 2023-01-21 11:33:11.740425: step: 1388/533, loss: 0.0008418560028076172 2023-01-21 11:33:12.900517: step: 1392/533, loss: 0.03683042526245117 2023-01-21 11:33:14.081893: step: 1396/533, loss: 0.011615562252700329 2023-01-21 11:33:15.328138: step: 1400/533, loss: 0.006627941038459539 2023-01-21 11:33:16.509106: step: 1404/533, loss: 0.0002004623383982107 2023-01-21 11:33:17.683231: step: 1408/533, loss: 0.06742353737354279 2023-01-21 11:33:18.880938: step: 1412/533, loss: 0.019146442413330078 2023-01-21 11:33:20.001482: step: 1416/533, loss: 0.006477737333625555 2023-01-21 11:33:21.145737: step: 1420/533, loss: 0.0006521224859170616 2023-01-21 11:33:22.333920: step: 1424/533, loss: 0.04995422437787056 2023-01-21 11:33:23.514478: step: 1428/533, loss: 0.010791946202516556 2023-01-21 11:33:24.687911: step: 1432/533, loss: 0.0004720747529063374 2023-01-21 11:33:25.911572: step: 1436/533, loss: 0.053525160998106 2023-01-21 11:33:27.083037: step: 1440/533, loss: 0.012418174184858799 2023-01-21 11:33:28.255773: step: 1444/533, loss: 0.0002233505219919607 2023-01-21 11:33:29.427452: step: 1448/533, loss: 0.01474857423454523 2023-01-21 11:33:30.626357: step: 1452/533, loss: 0.03698024898767471 2023-01-21 11:33:31.828729: step: 1456/533, loss: 0.018057728186249733 2023-01-21 11:33:32.986783: step: 1460/533, loss: 0.003914451692253351 2023-01-21 11:33:34.129822: step: 1464/533, loss: 0.002291393233463168 2023-01-21 11:33:35.320262: step: 1468/533, loss: 0.000687933003064245 2023-01-21 11:33:36.480027: step: 1472/533, loss: 0.006572341546416283 2023-01-21 11:33:37.659497: step: 1476/533, loss: 0.017885446548461914 2023-01-21 11:33:38.834430: step: 1480/533, loss: 0.070431187748909 2023-01-21 11:33:40.061422: step: 1484/533, loss: 0.07761158794164658 2023-01-21 11:33:41.225699: step: 1488/533, loss: 0.018985940143465996 2023-01-21 11:33:42.448607: step: 1492/533, loss: 0.0036679268814623356 2023-01-21 11:33:43.652170: step: 1496/533, loss: 0.001142311142757535 2023-01-21 11:33:44.856088: step: 1500/533, loss: 0.05302829667925835 2023-01-21 11:33:46.024909: step: 1504/533, loss: 0.008706331253051758 2023-01-21 11:33:47.237120: step: 1508/533, loss: 0.014579296112060547 2023-01-21 11:33:48.449479: step: 1512/533, loss: 0.03141956403851509 2023-01-21 11:33:49.625563: step: 1516/533, loss: 0.04047403484582901 2023-01-21 11:33:50.850412: step: 1520/533, loss: 0.0005851745372638106 2023-01-21 11:33:52.009627: step: 1524/533, loss: 0.06057176738977432 2023-01-21 11:33:53.210736: step: 1528/533, loss: 0.00021476745314430445 2023-01-21 11:33:54.368354: step: 1532/533, loss: 0.01809978298842907 2023-01-21 11:33:55.483385: step: 1536/533, loss: 0.02373666875064373 2023-01-21 11:33:56.657892: step: 1540/533, loss: 0.001889038016088307 2023-01-21 11:33:57.854037: step: 1544/533, loss: 0.007563877385109663 2023-01-21 11:33:59.028957: step: 1548/533, loss: 0.011583710089325905 2023-01-21 11:34:00.232219: step: 1552/533, loss: 0.006509304046630859 2023-01-21 11:34:01.422809: step: 1556/533, loss: 0.00034122465876862407 2023-01-21 11:34:02.589110: step: 1560/533, loss: 0.003436517668887973 2023-01-21 11:34:03.759965: step: 1564/533, loss: 0.023935507982969284 2023-01-21 11:34:04.937120: step: 1568/533, loss: 0.05130348354578018 2023-01-21 11:34:06.109932: step: 1572/533, loss: 0.0046652317978441715 2023-01-21 11:34:07.297370: step: 1576/533, loss: 0.006455517373979092 2023-01-21 11:34:08.507133: step: 1580/533, loss: 0.0006789207691326737 2023-01-21 11:34:09.666994: step: 1584/533, loss: 0.0489516481757164 2023-01-21 11:34:10.866667: step: 1588/533, loss: 0.004643535707145929 2023-01-21 11:34:12.021451: step: 1592/533, loss: 0.0033933641389012337 2023-01-21 11:34:13.174999: step: 1596/533, loss: 0.03144850954413414 2023-01-21 11:34:14.345023: step: 1600/533, loss: 0.0007343292818404734 2023-01-21 11:34:15.507680: step: 1604/533, loss: 0.0011828423012048006 2023-01-21 11:34:16.706849: step: 1608/533, loss: 0.024991797283291817 2023-01-21 11:34:17.887544: step: 1612/533, loss: 0.02314753644168377 2023-01-21 11:34:19.126928: step: 1616/533, loss: 0.008723830804228783 2023-01-21 11:34:20.248791: step: 1620/533, loss: 0.0005520820850506425 2023-01-21 11:34:21.399574: step: 1624/533, loss: 0.010486459359526634 2023-01-21 11:34:22.588632: step: 1628/533, loss: 0.007105732336640358 2023-01-21 11:34:23.772092: step: 1632/533, loss: 0.1719016134738922 2023-01-21 11:34:24.924167: step: 1636/533, loss: 0.00025384427863173187 2023-01-21 11:34:26.100344: step: 1640/533, loss: 7.600784010719508e-05 2023-01-21 11:34:27.305387: step: 1644/533, loss: 0.0025642395485192537 2023-01-21 11:34:28.507726: step: 1648/533, loss: 0.03659496456384659 2023-01-21 11:34:29.670062: step: 1652/533, loss: 0.004700374789535999 2023-01-21 11:34:30.846820: step: 1656/533, loss: 0.011045646853744984 2023-01-21 11:34:32.010786: step: 1660/533, loss: 0.08599185943603516 2023-01-21 11:34:33.199225: step: 1664/533, loss: 0.008962058462202549 2023-01-21 11:34:34.395295: step: 1668/533, loss: 0.03727264702320099 2023-01-21 11:34:35.610068: step: 1672/533, loss: 0.005661964416503906 2023-01-21 11:34:36.812371: step: 1676/533, loss: 5.511342525482178 2023-01-21 11:34:37.999918: step: 1680/533, loss: 0.12205095589160919 2023-01-21 11:34:39.183908: step: 1684/533, loss: 0.043882180005311966 2023-01-21 11:34:40.366332: step: 1688/533, loss: 0.007207632064819336 2023-01-21 11:34:41.568820: step: 1692/533, loss: 0.054799843579530716 2023-01-21 11:34:42.762386: step: 1696/533, loss: 0.024207402020692825 2023-01-21 11:34:43.908809: step: 1700/533, loss: 0.009629536420106888 2023-01-21 11:34:45.078923: step: 1704/533, loss: 0.029888534918427467 2023-01-21 11:34:46.250171: step: 1708/533, loss: 0.03759927675127983 2023-01-21 11:34:47.406658: step: 1712/533, loss: 0.009524344466626644 2023-01-21 11:34:48.600907: step: 1716/533, loss: 0.5328686237335205 2023-01-21 11:34:49.814911: step: 1720/533, loss: 0.0009124755742959678 2023-01-21 11:34:50.993917: step: 1724/533, loss: 0.02554798126220703 2023-01-21 11:34:52.163727: step: 1728/533, loss: 0.005244636442512274 2023-01-21 11:34:53.358146: step: 1732/533, loss: 0.01390447560697794 2023-01-21 11:34:54.556991: step: 1736/533, loss: 0.0038839338812977076 2023-01-21 11:34:55.748445: step: 1740/533, loss: 0.008599664084613323 2023-01-21 11:34:56.926974: step: 1744/533, loss: 0.026338864117860794 2023-01-21 11:34:58.163947: step: 1748/533, loss: 0.001549529959447682 2023-01-21 11:34:59.342858: step: 1752/533, loss: 0.0003589630068745464 2023-01-21 11:35:00.562280: step: 1756/533, loss: 0.0014263152843341231 2023-01-21 11:35:01.772802: step: 1760/533, loss: 0.057195089757442474 2023-01-21 11:35:02.968013: step: 1764/533, loss: 0.025406265631318092 2023-01-21 11:35:04.136720: step: 1768/533, loss: 0.0036479949485510588 2023-01-21 11:35:05.356263: step: 1772/533, loss: 0.13120490312576294 2023-01-21 11:35:06.503892: step: 1776/533, loss: 0.001035499619320035 2023-01-21 11:35:07.664544: step: 1780/533, loss: 0.001547098159790039 2023-01-21 11:35:08.874217: step: 1784/533, loss: 0.0009371757623739541 2023-01-21 11:35:10.097322: step: 1788/533, loss: 0.01241159439086914 2023-01-21 11:35:11.280690: step: 1792/533, loss: 0.16809825599193573 2023-01-21 11:35:12.479188: step: 1796/533, loss: 0.007153129670768976 2023-01-21 11:35:13.636577: step: 1800/533, loss: 0.0023974417708814144 2023-01-21 11:35:14.838511: step: 1804/533, loss: 0.04204883426427841 2023-01-21 11:35:15.995780: step: 1808/533, loss: 0.013253641314804554 2023-01-21 11:35:17.179434: step: 1812/533, loss: 0.03452930226922035 2023-01-21 11:35:18.431826: step: 1816/533, loss: 0.022757722064852715 2023-01-21 11:35:19.615271: step: 1820/533, loss: 0.03851603344082832 2023-01-21 11:35:20.784649: step: 1824/533, loss: 0.0040035247802734375 2023-01-21 11:35:21.963541: step: 1828/533, loss: 0.05450177192687988 2023-01-21 11:35:23.114723: step: 1832/533, loss: 0.007073306944221258 2023-01-21 11:35:24.275381: step: 1836/533, loss: 0.08658181130886078 2023-01-21 11:35:25.498209: step: 1840/533, loss: 0.0004989624139852822 2023-01-21 11:35:26.708031: step: 1844/533, loss: 0.012637806124985218 2023-01-21 11:35:27.980885: step: 1848/533, loss: 0.017045021057128906 2023-01-21 11:35:29.148826: step: 1852/533, loss: 0.021886637434363365 2023-01-21 11:35:30.340042: step: 1856/533, loss: 0.04786987230181694 2023-01-21 11:35:31.535203: step: 1860/533, loss: 0.04563789442181587 2023-01-21 11:35:32.705851: step: 1864/533, loss: 0.022282185032963753 2023-01-21 11:35:33.872471: step: 1868/533, loss: 0.03771495819091797 2023-01-21 11:35:35.047361: step: 1872/533, loss: 0.027369119226932526 2023-01-21 11:35:36.286580: step: 1876/533, loss: 0.6058931350708008 2023-01-21 11:35:37.447387: step: 1880/533, loss: 0.6500256657600403 2023-01-21 11:35:38.637437: step: 1884/533, loss: 0.04814014583826065 2023-01-21 11:35:39.797867: step: 1888/533, loss: 0.029430102556943893 2023-01-21 11:35:41.017747: step: 1892/533, loss: 0.0046901702880859375 2023-01-21 11:35:42.215583: step: 1896/533, loss: 0.05810112878680229 2023-01-21 11:35:43.403458: step: 1900/533, loss: 0.010535335168242455 2023-01-21 11:35:44.615784: step: 1904/533, loss: 0.0058329105377197266 2023-01-21 11:35:45.769119: step: 1908/533, loss: 0.007765769958496094 2023-01-21 11:35:46.943864: step: 1912/533, loss: 0.004822921939194202 2023-01-21 11:35:48.158789: step: 1916/533, loss: 0.000716781651135534 2023-01-21 11:35:49.370507: step: 1920/533, loss: 0.008598709478974342 2023-01-21 11:35:50.538376: step: 1924/533, loss: 0.05542631447315216 2023-01-21 11:35:51.767159: step: 1928/533, loss: 0.0026082992553710938 2023-01-21 11:35:52.944144: step: 1932/533, loss: 0.041727352887392044 2023-01-21 11:35:54.160138: step: 1936/533, loss: 0.001020717667415738 2023-01-21 11:35:55.320676: step: 1940/533, loss: 0.003590202424675226 2023-01-21 11:35:56.487526: step: 1944/533, loss: 0.009755706414580345 2023-01-21 11:35:57.706983: step: 1948/533, loss: 0.00464973459020257 2023-01-21 11:35:58.903880: step: 1952/533, loss: 0.04869265481829643 2023-01-21 11:36:00.103873: step: 1956/533, loss: 0.03931822627782822 2023-01-21 11:36:01.238249: step: 1960/533, loss: 0.03744201734662056 2023-01-21 11:36:02.402712: step: 1964/533, loss: 0.10401459038257599 2023-01-21 11:36:03.586191: step: 1968/533, loss: 0.0023281096946448088 2023-01-21 11:36:04.778386: step: 1972/533, loss: 0.0008807182312011719 2023-01-21 11:36:06.003380: step: 1976/533, loss: 0.020523929968476295 2023-01-21 11:36:07.176194: step: 1980/533, loss: 0.0007357597351074219 2023-01-21 11:36:08.366965: step: 1984/533, loss: 0.04567699506878853 2023-01-21 11:36:09.559352: step: 1988/533, loss: 0.0010856628650799394 2023-01-21 11:36:10.735678: step: 1992/533, loss: 0.003289890242740512 2023-01-21 11:36:11.902052: step: 1996/533, loss: 0.034958742558956146 2023-01-21 11:36:13.074597: step: 2000/533, loss: 0.0006689071888104081 2023-01-21 11:36:14.248659: step: 2004/533, loss: 0.05834522470831871 2023-01-21 11:36:15.441124: step: 2008/533, loss: 0.00016880035400390625 2023-01-21 11:36:16.592652: step: 2012/533, loss: 0.08300953358411789 2023-01-21 11:36:17.752347: step: 2016/533, loss: 0.0039192200638353825 2023-01-21 11:36:18.960161: step: 2020/533, loss: 0.002334976103156805 2023-01-21 11:36:20.119505: step: 2024/533, loss: 0.06384678184986115 2023-01-21 11:36:21.304216: step: 2028/533, loss: 0.011546612717211246 2023-01-21 11:36:22.496585: step: 2032/533, loss: 0.005102920811623335 2023-01-21 11:36:23.643312: step: 2036/533, loss: 0.457741916179657 2023-01-21 11:36:24.844801: step: 2040/533, loss: 0.027019884437322617 2023-01-21 11:36:26.014138: step: 2044/533, loss: 0.07772817462682724 2023-01-21 11:36:27.237875: step: 2048/533, loss: 9.927750215865672e-05 2023-01-21 11:36:28.460490: step: 2052/533, loss: 0.024082567542791367 2023-01-21 11:36:29.651319: step: 2056/533, loss: 0.02911539003252983 2023-01-21 11:36:30.842266: step: 2060/533, loss: 0.053293369710445404 2023-01-21 11:36:32.027872: step: 2064/533, loss: 0.003379488131031394 2023-01-21 11:36:33.212399: step: 2068/533, loss: 0.007106208708137274 2023-01-21 11:36:34.409701: step: 2072/533, loss: 0.03746318817138672 2023-01-21 11:36:35.601880: step: 2076/533, loss: 0.041051484644412994 2023-01-21 11:36:36.806954: step: 2080/533, loss: 0.0014678002335131168 2023-01-21 11:36:37.970209: step: 2084/533, loss: 0.023686887696385384 2023-01-21 11:36:39.144034: step: 2088/533, loss: 0.02539215050637722 2023-01-21 11:36:40.326943: step: 2092/533, loss: 0.002061557723209262 2023-01-21 11:36:41.581572: step: 2096/533, loss: 0.04375810921192169 2023-01-21 11:36:42.754953: step: 2100/533, loss: 0.019135857000947 2023-01-21 11:36:43.936697: step: 2104/533, loss: 0.0030751230660825968 2023-01-21 11:36:45.143327: step: 2108/533, loss: 0.08152560889720917 2023-01-21 11:36:46.322202: step: 2112/533, loss: 0.026894284412264824 2023-01-21 11:36:47.494524: step: 2116/533, loss: 0.04390363767743111 2023-01-21 11:36:48.676945: step: 2120/533, loss: 0.005885505583137274 2023-01-21 11:36:49.855912: step: 2124/533, loss: 0.011088562197983265 2023-01-21 11:36:51.087665: step: 2128/533, loss: 0.04149675369262695 2023-01-21 11:36:52.341279: step: 2132/533, loss: 0.061532117426395416 ================================================== Loss: 0.041 -------------------- Dev: {'event': {'p': 0.6240681576144835, 'r': 0.7802929427430093, 'f1': 0.6934911242603551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Test: {'event': {'p': 0.6589229144667371, 'r': 0.7979539641943734, 'f1': 0.7218045112781956}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} 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': 16} Korean: {'event': {'p': 0.6415094339622641, 'r': 0.5396825396825397, 'f1': 0.5862068965517241}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Russian: {'event': {'p': 0.4878048780487805, 'r': 0.5555555555555556, 'f1': 0.5194805194805195}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 11:37:33.835946: step: 4/533, loss: 0.05608072504401207 2023-01-21 11:37:35.000170: step: 8/533, loss: 0.0068302154541015625 2023-01-21 11:37:36.156414: step: 12/533, loss: 0.052759744226932526 2023-01-21 11:37:37.391679: step: 16/533, loss: 0.008755017071962357 2023-01-21 11:37:38.582164: step: 20/533, loss: 0.006179142277687788 2023-01-21 11:37:39.745750: step: 24/533, loss: 0.049042705446481705 2023-01-21 11:37:40.911816: step: 28/533, loss: 0.003762722248211503 2023-01-21 11:37:42.139280: step: 32/533, loss: 0.04843645170331001 2023-01-21 11:37:43.319807: step: 36/533, loss: 0.1207214817404747 2023-01-21 11:37:44.481692: step: 40/533, loss: 0.027240967378020287 2023-01-21 11:37:45.647043: step: 44/533, loss: 0.06505122780799866 2023-01-21 11:37:46.826314: step: 48/533, loss: 0.001573848770931363 2023-01-21 11:37:48.027442: step: 52/533, loss: 0.006667900364845991 2023-01-21 11:37:49.235958: step: 56/533, loss: 0.005202674772590399 2023-01-21 11:37:50.464177: step: 60/533, loss: 0.0001806259242584929 2023-01-21 11:37:51.643689: step: 64/533, loss: 0.008336687460541725 2023-01-21 11:37:52.863070: step: 68/533, loss: 0.0003641128714662045 2023-01-21 11:37:54.003062: step: 72/533, loss: 0.0024078369606286287 2023-01-21 11:37:55.174237: step: 76/533, loss: 0.013912821188569069 2023-01-21 11:37:56.319241: step: 80/533, loss: 0.0032484056428074837 2023-01-21 11:37:57.483111: step: 84/533, loss: 0.002562236739322543 2023-01-21 11:37:58.651954: step: 88/533, loss: 0.0006509781233035028 2023-01-21 11:37:59.820864: step: 92/533, loss: 0.011055183596909046 2023-01-21 11:38:00.985005: step: 96/533, loss: 0.00506095914170146 2023-01-21 11:38:02.197872: step: 100/533, loss: 0.015534305945038795 2023-01-21 11:38:03.414249: step: 104/533, loss: 0.028289128094911575 2023-01-21 11:38:04.595815: step: 108/533, loss: 0.02078886143863201 2023-01-21 11:38:05.778743: step: 112/533, loss: 0.004430580418556929 2023-01-21 11:38:06.965871: step: 116/533, loss: 0.0007699966663494706 2023-01-21 11:38:08.175493: step: 120/533, loss: 0.0001239776611328125 2023-01-21 11:38:09.334592: step: 124/533, loss: 2.9805119037628174 2023-01-21 11:38:10.491443: step: 128/533, loss: 0.009613228030502796 2023-01-21 11:38:11.650121: step: 132/533, loss: 0.004459572024643421 2023-01-21 11:38:12.811201: step: 136/533, loss: 0.004940084647387266 2023-01-21 11:38:13.959837: step: 140/533, loss: 0.25174829363822937 2023-01-21 11:38:15.129701: step: 144/533, loss: 0.03089718706905842 2023-01-21 11:38:16.333068: step: 148/533, loss: 0.023326683789491653 2023-01-21 11:38:17.490350: step: 152/533, loss: 0.023358916863799095 2023-01-21 11:38:18.690424: step: 156/533, loss: 0.005521631333976984 2023-01-21 11:38:19.852766: step: 160/533, loss: 0.0020106316078454256 2023-01-21 11:38:21.063298: step: 164/533, loss: 0.04149749502539635 2023-01-21 11:38:22.217653: step: 168/533, loss: 0.003925132565200329 2023-01-21 11:38:23.405384: step: 172/533, loss: 0.012125111185014248 2023-01-21 11:38:24.600160: step: 176/533, loss: 0.009181595407426357 2023-01-21 11:38:25.760417: step: 180/533, loss: 0.06652164459228516 2023-01-21 11:38:26.959110: step: 184/533, loss: 0.00015954971604514867 2023-01-21 11:38:28.134281: step: 188/533, loss: 0.001555633614771068 2023-01-21 11:38:29.337999: step: 192/533, loss: 0.003946781158447266 2023-01-21 11:38:30.515162: step: 196/533, loss: 0.00089263916015625 2023-01-21 11:38:31.645842: step: 200/533, loss: 0.10044444352388382 2023-01-21 11:38:32.785477: step: 204/533, loss: 0.0017809867858886719 2023-01-21 11:38:34.003388: step: 208/533, loss: 0.0005335807800292969 2023-01-21 11:38:35.178305: step: 212/533, loss: 0.0393100269138813 2023-01-21 11:38:36.356962: step: 216/533, loss: 0.023301126435399055 2023-01-21 11:38:37.548144: step: 220/533, loss: 0.004351234529167414 2023-01-21 11:38:38.723263: step: 224/533, loss: 0.01137557066977024 2023-01-21 11:38:39.890514: step: 228/533, loss: 0.0106497285887599 2023-01-21 11:38:41.071166: step: 232/533, loss: 0.0031280517578125 2023-01-21 11:38:42.297964: step: 236/533, loss: 0.004107809159904718 2023-01-21 11:38:43.473035: step: 240/533, loss: 0.008036613464355469 2023-01-21 11:38:44.645611: step: 244/533, loss: 0.0030488967895507812 2023-01-21 11:38:45.763956: step: 248/533, loss: 0.004109001252800226 2023-01-21 11:38:46.918156: step: 252/533, loss: 0.04989318922162056 2023-01-21 11:38:48.099034: step: 256/533, loss: 0.018237780779600143 2023-01-21 11:38:49.253745: step: 260/533, loss: 0.005792808718979359 2023-01-21 11:38:50.412815: step: 264/533, loss: 0.001018607639707625 2023-01-21 11:38:51.566520: step: 268/533, loss: 0.024995993822813034 2023-01-21 11:38:52.725347: step: 272/533, loss: 0.0034825324546545744 2023-01-21 11:38:53.880339: step: 276/533, loss: 0.18073216080665588 2023-01-21 11:38:55.109904: step: 280/533, loss: 0.009951591491699219 2023-01-21 11:38:56.271947: step: 284/533, loss: 0.0006700515514239669 2023-01-21 11:38:57.459209: step: 288/533, loss: 0.000870227871928364 2023-01-21 11:38:58.609766: step: 292/533, loss: 0.011655330657958984 2023-01-21 11:38:59.790094: step: 296/533, loss: 0.05896587669849396 2023-01-21 11:39:00.984110: step: 300/533, loss: 0.09211063385009766 2023-01-21 11:39:02.142318: step: 304/533, loss: 0.0387604720890522 2023-01-21 11:39:03.315316: step: 308/533, loss: 0.08025378733873367 2023-01-21 11:39:04.501646: step: 312/533, loss: 0.037885189056396484 2023-01-21 11:39:05.672069: step: 316/533, loss: 0.21114139258861542 2023-01-21 11:39:06.860133: step: 320/533, loss: 0.002173805143684149 2023-01-21 11:39:08.025341: step: 324/533, loss: -4.196166628389619e-06 2023-01-21 11:39:09.191719: step: 328/533, loss: 0.0019367218483239412 2023-01-21 11:39:10.359978: step: 332/533, loss: 0.05145730823278427 2023-01-21 11:39:11.536194: step: 336/533, loss: 0.006091403774917126 2023-01-21 11:39:12.682895: step: 340/533, loss: 0.006446457002311945 2023-01-21 11:39:13.860847: step: 344/533, loss: 0.00121221540030092 2023-01-21 11:39:15.052017: step: 348/533, loss: 0.018650149926543236 2023-01-21 11:39:16.199052: step: 352/533, loss: 0.0018839837284758687 2023-01-21 11:39:17.370249: step: 356/533, loss: 0.0008254051208496094 2023-01-21 11:39:18.596297: step: 360/533, loss: 0.01699669286608696 2023-01-21 11:39:19.854785: step: 364/533, loss: 0.02749962918460369 2023-01-21 11:39:21.088773: step: 368/533, loss: 0.013296891003847122 2023-01-21 11:39:22.272374: step: 372/533, loss: 0.009439850226044655 2023-01-21 11:39:23.461108: step: 376/533, loss: 0.02125081978738308 2023-01-21 11:39:24.677768: step: 380/533, loss: 0.034911155700683594 2023-01-21 11:39:25.875010: step: 384/533, loss: 0.0003604888916015625 2023-01-21 11:39:27.053339: step: 388/533, loss: 0.0019594430923461914 2023-01-21 11:39:28.253362: step: 392/533, loss: 0.08834762871265411 2023-01-21 11:39:29.415435: step: 396/533, loss: 0.014138413593173027 2023-01-21 11:39:30.613798: step: 400/533, loss: 0.003722858615219593 2023-01-21 11:39:31.780589: step: 404/533, loss: 0.01670980453491211 2023-01-21 11:39:32.943803: step: 408/533, loss: 0.013614273630082607 2023-01-21 11:39:34.138413: step: 412/533, loss: 0.10627670586109161 2023-01-21 11:39:35.339310: step: 416/533, loss: 0.03528871759772301 2023-01-21 11:39:36.516548: step: 420/533, loss: 0.00450520496815443 2023-01-21 11:39:37.701097: step: 424/533, loss: 0.0008152961963787675 2023-01-21 11:39:38.925833: step: 428/533, loss: 0.0012554168934002519 2023-01-21 11:39:40.056168: step: 432/533, loss: 0.04270394146442413 2023-01-21 11:39:41.215491: step: 436/533, loss: 0.0003074169217143208 2023-01-21 11:39:42.379818: step: 440/533, loss: 0.042651571333408356 2023-01-21 11:39:43.537639: step: 444/533, loss: 0.07190332561731339 2023-01-21 11:39:44.721498: step: 448/533, loss: 0.02822551690042019 2023-01-21 11:39:45.905791: step: 452/533, loss: 0.03028698079288006 2023-01-21 11:39:47.117883: step: 456/533, loss: 0.012123984284698963 2023-01-21 11:39:48.340950: step: 460/533, loss: 0.006522560026496649 2023-01-21 11:39:49.548790: step: 464/533, loss: 0.01727910153567791 2023-01-21 11:39:50.729786: step: 468/533, loss: 0.05306796729564667 2023-01-21 11:39:51.888601: step: 472/533, loss: 0.004248524084687233 2023-01-21 11:39:53.124457: step: 476/533, loss: 0.002651214599609375 2023-01-21 11:39:54.288760: step: 480/533, loss: 0.002184104872867465 2023-01-21 11:39:55.455379: step: 484/533, loss: 0.02514057233929634 2023-01-21 11:39:56.627444: step: 488/533, loss: 0.00786514300853014 2023-01-21 11:39:57.799510: step: 492/533, loss: 0.018297530710697174 2023-01-21 11:39:58.977941: step: 496/533, loss: 0.003752422519028187 2023-01-21 11:40:00.187268: step: 500/533, loss: 0.01578083075582981 2023-01-21 11:40:01.352198: step: 504/533, loss: 0.0007875442388467491 2023-01-21 11:40:02.539020: step: 508/533, loss: 0.00044369700481183827 2023-01-21 11:40:03.742245: step: 512/533, loss: 0.0015524388290941715 2023-01-21 11:40:04.928682: step: 516/533, loss: 0.009407234378159046 2023-01-21 11:40:06.084251: step: 520/533, loss: 0.006123637780547142 2023-01-21 11:40:07.254650: step: 524/533, loss: 0.007503509521484375 2023-01-21 11:40:08.428519: step: 528/533, loss: 0.010577201843261719 2023-01-21 11:40:09.591828: step: 532/533, loss: 0.026291703805327415 2023-01-21 11:40:10.796371: step: 536/533, loss: 0.0040225982666015625 2023-01-21 11:40:11.961325: step: 540/533, loss: 0.0010875702137127519 2023-01-21 11:40:13.167637: step: 544/533, loss: 0.07167740166187286 2023-01-21 11:40:14.333982: step: 548/533, loss: 0.00013570785813499242 2023-01-21 11:40:15.515224: step: 552/533, loss: 0.6052299737930298 2023-01-21 11:40:16.686100: step: 556/533, loss: 0.00020322800264693797 2023-01-21 11:40:17.838995: step: 560/533, loss: 0.4142410159111023 2023-01-21 11:40:19.021147: step: 564/533, loss: 0.0003871917724609375 2023-01-21 11:40:20.204930: step: 568/533, loss: 0.0008869171142578125 2023-01-21 11:40:21.417307: step: 572/533, loss: 0.012749576941132545 2023-01-21 11:40:22.594217: step: 576/533, loss: 0.02688579633831978 2023-01-21 11:40:23.770407: step: 580/533, loss: 0.06965198367834091 2023-01-21 11:40:24.938467: step: 584/533, loss: 0.012274456210434437 2023-01-21 11:40:26.117729: step: 588/533, loss: 0.0014977456303313375 2023-01-21 11:40:27.315871: step: 592/533, loss: 0.08007049560546875 2023-01-21 11:40:28.500711: step: 596/533, loss: 0.0019989013671875 2023-01-21 11:40:29.653023: step: 600/533, loss: 0.00740737933665514 2023-01-21 11:40:30.848699: step: 604/533, loss: 0.059052374213933945 2023-01-21 11:40:32.040130: step: 608/533, loss: 8.37326078908518e-05 2023-01-21 11:40:33.212876: step: 612/533, loss: 0.00994024332612753 2023-01-21 11:40:34.352449: step: 616/533, loss: 0.02215109020471573 2023-01-21 11:40:35.534179: step: 620/533, loss: 0.005905819125473499 2023-01-21 11:40:36.722540: step: 624/533, loss: 0.0372898131608963 2023-01-21 11:40:37.924924: step: 628/533, loss: 0.04229144752025604 2023-01-21 11:40:39.093414: step: 632/533, loss: 0.004037666134536266 2023-01-21 11:40:40.242261: step: 636/533, loss: 0.00018367767916060984 2023-01-21 11:40:41.434023: step: 640/533, loss: 0.03332529217004776 2023-01-21 11:40:42.628152: step: 644/533, loss: 0.018091201782226562 2023-01-21 11:40:43.801100: step: 648/533, loss: 0.016399670392274857 2023-01-21 11:40:44.987049: step: 652/533, loss: 0.0014136313693597913 2023-01-21 11:40:46.213125: step: 656/533, loss: 0.0541507825255394 2023-01-21 11:40:47.394279: step: 660/533, loss: 0.0584072582423687 2023-01-21 11:40:48.607780: step: 664/533, loss: 0.05647401884198189 2023-01-21 11:40:49.804998: step: 668/533, loss: 1.0480600595474243 2023-01-21 11:40:51.008927: step: 672/533, loss: 0.022043418139219284 2023-01-21 11:40:52.210018: step: 676/533, loss: 0.0027786255814135075 2023-01-21 11:40:53.422569: step: 680/533, loss: 0.010172463022172451 2023-01-21 11:40:54.644413: step: 684/533, loss: 0.0022869110107421875 2023-01-21 11:40:55.816270: step: 688/533, loss: 0.0011724472278729081 2023-01-21 11:40:56.996801: step: 692/533, loss: 0.005409527570009232 2023-01-21 11:40:58.159267: step: 696/533, loss: 0.009091068059206009 2023-01-21 11:40:59.314340: step: 700/533, loss: 0.03131723403930664 2023-01-21 11:41:00.518474: step: 704/533, loss: 0.017778396606445312 2023-01-21 11:41:01.696265: step: 708/533, loss: 0.002792692044749856 2023-01-21 11:41:02.860961: step: 712/533, loss: 0.021972371265292168 2023-01-21 11:41:04.039078: step: 716/533, loss: 0.030235623940825462 2023-01-21 11:41:05.223200: step: 720/533, loss: 0.05753831937909126 2023-01-21 11:41:06.383043: step: 724/533, loss: 0.028432749211788177 2023-01-21 11:41:07.580770: step: 728/533, loss: 8.76426711329259e-05 2023-01-21 11:41:08.785983: step: 732/533, loss: 0.04440498352050781 2023-01-21 11:41:09.910140: step: 736/533, loss: 7.534026735811494e-06 2023-01-21 11:41:11.084799: step: 740/533, loss: 0.0031850815284997225 2023-01-21 11:41:12.260893: step: 744/533, loss: 0.0006715774652548134 2023-01-21 11:41:13.429589: step: 748/533, loss: 0.010947704315185547 2023-01-21 11:41:14.655473: step: 752/533, loss: 0.0019267082680016756 2023-01-21 11:41:15.850300: step: 756/533, loss: 0.05553417652845383 2023-01-21 11:41:17.025759: step: 760/533, loss: 0.06920681148767471 2023-01-21 11:41:18.229812: step: 764/533, loss: 0.029787994921207428 2023-01-21 11:41:19.438582: step: 768/533, loss: 0.021555567160248756 2023-01-21 11:41:20.621492: step: 772/533, loss: 0.017038488760590553 2023-01-21 11:41:21.818726: step: 776/533, loss: 1.717919111251831 2023-01-21 11:41:22.988176: step: 780/533, loss: 0.005867290776222944 2023-01-21 11:41:24.123537: step: 784/533, loss: 0.09713178873062134 2023-01-21 11:41:25.288316: step: 788/533, loss: 0.036733344197273254 2023-01-21 11:41:26.500835: step: 792/533, loss: 0.019828511402010918 2023-01-21 11:41:27.680029: step: 796/533, loss: 0.0029081343673169613 2023-01-21 11:41:28.908575: step: 800/533, loss: 0.017275989055633545 2023-01-21 11:41:30.084815: step: 804/533, loss: 0.09756974875926971 2023-01-21 11:41:31.271688: step: 808/533, loss: 0.005640506744384766 2023-01-21 11:41:32.478175: step: 812/533, loss: 0.0070188045501708984 2023-01-21 11:41:33.651819: step: 816/533, loss: 0.007050991524010897 2023-01-21 11:41:34.874450: step: 820/533, loss: 0.00012218952178955078 2023-01-21 11:41:36.046362: step: 824/533, loss: 0.0050903321243822575 2023-01-21 11:41:37.246688: step: 828/533, loss: 0.00038170814514160156 2023-01-21 11:41:38.403394: step: 832/533, loss: 6.055832272977568e-06 2023-01-21 11:41:39.581073: step: 836/533, loss: 0.11572265625 2023-01-21 11:41:40.763972: step: 840/533, loss: 0.0002931595081463456 2023-01-21 11:41:41.938950: step: 844/533, loss: 0.03839264065027237 2023-01-21 11:41:43.168567: step: 848/533, loss: 0.0022394179832190275 2023-01-21 11:41:44.355622: step: 852/533, loss: 0.006460762117058039 2023-01-21 11:41:45.549921: step: 856/533, loss: 0.005555248819291592 2023-01-21 11:41:46.748312: step: 860/533, loss: 0.006723976228386164 2023-01-21 11:41:47.924249: step: 864/533, loss: 0.08497696369886398 2023-01-21 11:41:49.105796: step: 868/533, loss: 0.016167260706424713 2023-01-21 11:41:50.296982: step: 872/533, loss: 0.0004943847889080644 2023-01-21 11:41:51.521923: step: 876/533, loss: 0.05033750459551811 2023-01-21 11:41:52.685928: step: 880/533, loss: 0.004552364349365234 2023-01-21 11:41:53.868882: step: 884/533, loss: 0.00016300679999403656 2023-01-21 11:41:55.038488: step: 888/533, loss: 0.16002637147903442 2023-01-21 11:41:56.200020: step: 892/533, loss: 0.0014913559425622225 2023-01-21 11:41:57.357730: step: 896/533, loss: 0.005386924836784601 2023-01-21 11:41:58.539971: step: 900/533, loss: 0.015250921249389648 2023-01-21 11:41:59.730086: step: 904/533, loss: 0.01175675354897976 2023-01-21 11:42:00.934780: step: 908/533, loss: 0.04188823699951172 2023-01-21 11:42:02.097782: step: 912/533, loss: 0.023948539048433304 2023-01-21 11:42:03.289979: step: 916/533, loss: 0.011065912432968616 2023-01-21 11:42:04.482965: step: 920/533, loss: 0.0011293889256194234 2023-01-21 11:42:05.729945: step: 924/533, loss: 0.02864704094827175 2023-01-21 11:42:06.931512: step: 928/533, loss: 0.00133857736364007 2023-01-21 11:42:08.055045: step: 932/533, loss: 0.008938217535614967 2023-01-21 11:42:09.233321: step: 936/533, loss: 0.02033844031393528 2023-01-21 11:42:10.481740: step: 940/533, loss: 0.003681755159050226 2023-01-21 11:42:11.637113: step: 944/533, loss: 0.017348384484648705 2023-01-21 11:42:12.845062: step: 948/533, loss: 0.18990212678909302 2023-01-21 11:42:14.001199: step: 952/533, loss: 0.0006432533264160156 2023-01-21 11:42:15.204953: step: 956/533, loss: 0.0032904627732932568 2023-01-21 11:42:16.385249: step: 960/533, loss: 0.0005182266468182206 2023-01-21 11:42:17.554829: step: 964/533, loss: 0.003928565885871649 2023-01-21 11:42:18.748272: step: 968/533, loss: 0.029495621100068092 2023-01-21 11:42:19.935736: step: 972/533, loss: 0.00980234146118164 2023-01-21 11:42:21.129616: step: 976/533, loss: 0.01341257058084011 2023-01-21 11:42:22.298032: step: 980/533, loss: 0.16279855370521545 2023-01-21 11:42:23.504211: step: 984/533, loss: 0.1024242416024208 2023-01-21 11:42:24.626902: step: 988/533, loss: 0.0001791000395314768 2023-01-21 11:42:25.795344: step: 992/533, loss: 0.021137427538633347 2023-01-21 11:42:26.987202: step: 996/533, loss: 0.008636856451630592 2023-01-21 11:42:28.193095: step: 1000/533, loss: 0.0004884600639343262 2023-01-21 11:42:29.388545: step: 1004/533, loss: 0.08552370220422745 2023-01-21 11:42:30.607575: step: 1008/533, loss: 0.026104547083377838 2023-01-21 11:42:31.778986: step: 1012/533, loss: 0.00365715054795146 2023-01-21 11:42:32.932177: step: 1016/533, loss: 0.00258979806676507 2023-01-21 11:42:34.111737: step: 1020/533, loss: 0.01114511489868164 2023-01-21 11:42:35.253421: step: 1024/533, loss: 0.06952658295631409 2023-01-21 11:42:36.407881: step: 1028/533, loss: 0.041036225855350494 2023-01-21 11:42:37.580758: step: 1032/533, loss: 0.003975486848503351 2023-01-21 11:42:38.759484: step: 1036/533, loss: 0.009595775976777077 2023-01-21 11:42:39.970351: step: 1040/533, loss: 0.013378525152802467 2023-01-21 11:42:41.094698: step: 1044/533, loss: 0.019137000665068626 2023-01-21 11:42:42.257901: step: 1048/533, loss: 4.7969817387638614e-05 2023-01-21 11:42:43.427739: step: 1052/533, loss: 0.0026799202896654606 2023-01-21 11:42:44.589939: step: 1056/533, loss: 0.0016494751907885075 2023-01-21 11:42:45.788441: step: 1060/533, loss: 0.039968397468328476 2023-01-21 11:42:46.957878: step: 1064/533, loss: 0.0075446609407663345 2023-01-21 11:42:48.149157: step: 1068/533, loss: 0.007601452060043812 2023-01-21 11:42:49.363718: step: 1072/533, loss: 0.023512650281190872 2023-01-21 11:42:50.564988: step: 1076/533, loss: 0.06672406196594238 2023-01-21 11:42:51.733502: step: 1080/533, loss: 0.0727386474609375 2023-01-21 11:42:52.905555: step: 1084/533, loss: 0.006413841620087624 2023-01-21 11:42:54.045164: step: 1088/533, loss: 0.022216796875 2023-01-21 11:42:55.234527: step: 1092/533, loss: 0.016120338812470436 2023-01-21 11:42:56.407817: step: 1096/533, loss: 0.030898286029696465 2023-01-21 11:42:57.602055: step: 1100/533, loss: 0.011331653222441673 2023-01-21 11:42:58.843335: step: 1104/533, loss: 0.11912579834461212 2023-01-21 11:42:59.997911: step: 1108/533, loss: 0.00015935898409225047 2023-01-21 11:43:01.186526: step: 1112/533, loss: 0.00235404958948493 2023-01-21 11:43:02.357486: step: 1116/533, loss: 0.030215073376893997 2023-01-21 11:43:03.546583: step: 1120/533, loss: 0.0014148711925372481 2023-01-21 11:43:04.714661: step: 1124/533, loss: 0.0008066296577453613 2023-01-21 11:43:05.906523: step: 1128/533, loss: 0.0031019686721265316 2023-01-21 11:43:07.109914: step: 1132/533, loss: 0.0009886742336675525 2023-01-21 11:43:08.269617: step: 1136/533, loss: 0.0015953541733324528 2023-01-21 11:43:09.444825: step: 1140/533, loss: 0.02370443381369114 2023-01-21 11:43:10.591408: step: 1144/533, loss: 0.08602094650268555 2023-01-21 11:43:11.754003: step: 1148/533, loss: 0.008855437859892845 2023-01-21 11:43:12.962120: step: 1152/533, loss: 0.025487661361694336 2023-01-21 11:43:14.122714: step: 1156/533, loss: 0.0011738777393475175 2023-01-21 11:43:15.279763: step: 1160/533, loss: 0.04270558804273605 2023-01-21 11:43:16.467952: step: 1164/533, loss: 0.004864215850830078 2023-01-21 11:43:17.666319: step: 1168/533, loss: 0.020808840170502663 2023-01-21 11:43:18.887056: step: 1172/533, loss: 0.024790192022919655 2023-01-21 11:43:20.075448: step: 1176/533, loss: 0.002162837889045477 2023-01-21 11:43:21.243179: step: 1180/533, loss: 0.023140335455536842 2023-01-21 11:43:22.424280: step: 1184/533, loss: 0.04875240474939346 2023-01-21 11:43:23.613597: step: 1188/533, loss: 0.0011785507667809725 2023-01-21 11:43:24.800771: step: 1192/533, loss: 0.0038541792891919613 2023-01-21 11:43:25.984139: step: 1196/533, loss: 0.027614861726760864 2023-01-21 11:43:27.129509: step: 1200/533, loss: 0.006805134005844593 2023-01-21 11:43:28.324517: step: 1204/533, loss: 0.03253033384680748 2023-01-21 11:43:29.521335: step: 1208/533, loss: 5.6552886235294864e-05 2023-01-21 11:43:30.727734: step: 1212/533, loss: 0.03117389790713787 2023-01-21 11:43:31.918478: step: 1216/533, loss: 0.021603966131806374 2023-01-21 11:43:33.082923: step: 1220/533, loss: 0.018318558111786842 2023-01-21 11:43:34.218409: step: 1224/533, loss: 0.013022899627685547 2023-01-21 11:43:35.439259: step: 1228/533, loss: 0.009419441223144531 2023-01-21 11:43:36.605215: step: 1232/533, loss: 1.3256072634248994e-05 2023-01-21 11:43:37.804212: step: 1236/533, loss: 0.007796573918312788 2023-01-21 11:43:38.940926: step: 1240/533, loss: 0.0009675026522018015 2023-01-21 11:43:40.103079: step: 1244/533, loss: 0.0860045924782753 2023-01-21 11:43:41.313911: step: 1248/533, loss: 0.029773332178592682 2023-01-21 11:43:42.470602: step: 1252/533, loss: 0.015056420117616653 2023-01-21 11:43:43.646671: step: 1256/533, loss: 0.06025705486536026 2023-01-21 11:43:44.825307: step: 1260/533, loss: 0.01129903830587864 2023-01-21 11:43:45.997081: step: 1264/533, loss: 0.02307424508035183 2023-01-21 11:43:47.161777: step: 1268/533, loss: 0.04053802415728569 2023-01-21 11:43:48.354819: step: 1272/533, loss: 0.004135513212531805 2023-01-21 11:43:49.568310: step: 1276/533, loss: 0.035432055592536926 2023-01-21 11:43:50.750769: step: 1280/533, loss: 0.013380718417465687 2023-01-21 11:43:51.936507: step: 1284/533, loss: 0.0014565468300133944 2023-01-21 11:43:53.170054: step: 1288/533, loss: 0.009642865508794785 2023-01-21 11:43:54.400513: step: 1292/533, loss: 9.403229341842234e-05 2023-01-21 11:43:55.575641: step: 1296/533, loss: 0.038231659680604935 2023-01-21 11:43:56.777827: step: 1300/533, loss: 0.01034850999712944 2023-01-21 11:43:57.929021: step: 1304/533, loss: 0.024612808600068092 2023-01-21 11:43:59.108516: step: 1308/533, loss: 0.0072463033720850945 2023-01-21 11:44:00.310367: step: 1312/533, loss: 0.01629180833697319 2023-01-21 11:44:01.489263: step: 1316/533, loss: 0.0032887414563447237 2023-01-21 11:44:02.696537: step: 1320/533, loss: 0.0037805556785315275 2023-01-21 11:44:03.910616: step: 1324/533, loss: 0.004913234617561102 2023-01-21 11:44:05.069911: step: 1328/533, loss: 0.010395479388535023 2023-01-21 11:44:06.313565: step: 1332/533, loss: 0.7065010070800781 2023-01-21 11:44:07.514451: step: 1336/533, loss: 0.01079416275024414 2023-01-21 11:44:08.704406: step: 1340/533, loss: 0.046254731714725494 2023-01-21 11:44:09.871879: step: 1344/533, loss: 0.024059485644102097 2023-01-21 11:44:11.025891: step: 1348/533, loss: 0.08166928589344025 2023-01-21 11:44:12.172150: step: 1352/533, loss: 0.002643489744514227 2023-01-21 11:44:13.326645: step: 1356/533, loss: 0.03202419355511665 2023-01-21 11:44:14.509914: step: 1360/533, loss: 0.0095405587926507 2023-01-21 11:44:15.694868: step: 1364/533, loss: 0.0013853549025952816 2023-01-21 11:44:16.851646: step: 1368/533, loss: 0.008166885003447533 2023-01-21 11:44:17.980656: step: 1372/533, loss: 0.0072609903290867805 2023-01-21 11:44:19.191595: step: 1376/533, loss: 0.0057697300799191 2023-01-21 11:44:20.395175: step: 1380/533, loss: 0.05821714550256729 2023-01-21 11:44:21.604052: step: 1384/533, loss: 0.048148348927497864 2023-01-21 11:44:22.800750: step: 1388/533, loss: 0.0068649291060864925 2023-01-21 11:44:23.930825: step: 1392/533, loss: 0.019165897741913795 2023-01-21 11:44:25.099554: step: 1396/533, loss: 0.020553112030029297 2023-01-21 11:44:26.266532: step: 1400/533, loss: 0.04664144665002823 2023-01-21 11:44:27.495055: step: 1404/533, loss: 6.64711042190902e-05 2023-01-21 11:44:28.693942: step: 1408/533, loss: 0.010936547070741653 2023-01-21 11:44:29.846767: step: 1412/533, loss: 0.021195411682128906 2023-01-21 11:44:30.989545: step: 1416/533, loss: 0.027533099055290222 2023-01-21 11:44:32.169267: step: 1420/533, loss: 0.0031245232094079256 2023-01-21 11:44:33.346115: step: 1424/533, loss: 0.002873325487598777 2023-01-21 11:44:34.499741: step: 1428/533, loss: 0.05027656629681587 2023-01-21 11:44:35.673442: step: 1432/533, loss: 0.011453866958618164 2023-01-21 11:44:36.858922: step: 1436/533, loss: 0.003962039947509766 2023-01-21 11:44:38.017668: step: 1440/533, loss: 0.005975246429443359 2023-01-21 11:44:39.206948: step: 1444/533, loss: 0.010951423086225986 2023-01-21 11:44:40.407728: step: 1448/533, loss: 0.0022405623458325863 2023-01-21 11:44:41.590528: step: 1452/533, loss: 0.01731410063803196 2023-01-21 11:44:42.753614: step: 1456/533, loss: 0.04520225524902344 2023-01-21 11:44:43.953846: step: 1460/533, loss: 0.005767345428466797 2023-01-21 11:44:45.122596: step: 1464/533, loss: 0.014719819650053978 2023-01-21 11:44:46.303843: step: 1468/533, loss: 0.002204799558967352 2023-01-21 11:44:47.484412: step: 1472/533, loss: 0.04959557205438614 2023-01-21 11:44:48.657302: step: 1476/533, loss: 0.032160572707653046 2023-01-21 11:44:49.844399: step: 1480/533, loss: 0.028227951377630234 2023-01-21 11:44:51.039427: step: 1484/533, loss: 0.0010269165504723787 2023-01-21 11:44:52.222670: step: 1488/533, loss: 0.008346510119736195 2023-01-21 11:44:53.415982: step: 1492/533, loss: 0.004586410243064165 2023-01-21 11:44:54.580536: step: 1496/533, loss: 0.001663637231104076 2023-01-21 11:44:55.732930: step: 1500/533, loss: 0.8587007522583008 2023-01-21 11:44:56.954549: step: 1504/533, loss: 0.0015378951793536544 2023-01-21 11:44:58.162674: step: 1508/533, loss: 0.0017135620582848787 2023-01-21 11:44:59.328388: step: 1512/533, loss: 0.0026695250999182463 2023-01-21 11:45:00.523142: step: 1516/533, loss: 0.0011548995971679688 2023-01-21 11:45:01.686916: step: 1520/533, loss: 0.004550743382424116 2023-01-21 11:45:02.846298: step: 1524/533, loss: 0.04035091772675514 2023-01-21 11:45:04.015489: step: 1528/533, loss: 0.0012311935424804688 2023-01-21 11:45:05.159584: step: 1532/533, loss: 0.026397133246064186 2023-01-21 11:45:06.334043: step: 1536/533, loss: 0.002847003983333707 2023-01-21 11:45:07.519856: step: 1540/533, loss: 0.001199436141178012 2023-01-21 11:45:08.704028: step: 1544/533, loss: 0.010825681500136852 2023-01-21 11:45:09.901342: step: 1548/533, loss: 0.024338722229003906 2023-01-21 11:45:11.140590: step: 1552/533, loss: 1.5991666316986084 2023-01-21 11:45:12.324349: step: 1556/533, loss: 0.029135845601558685 2023-01-21 11:45:13.484566: step: 1560/533, loss: 0.012155341915786266 2023-01-21 11:45:14.669774: step: 1564/533, loss: 0.022369956597685814 2023-01-21 11:45:15.833752: step: 1568/533, loss: 0.045952893793582916 2023-01-21 11:45:17.012760: step: 1572/533, loss: 0.002908039139583707 2023-01-21 11:45:18.220460: step: 1576/533, loss: 0.053562358021736145 2023-01-21 11:45:19.411481: step: 1580/533, loss: 0.0067806243896484375 2023-01-21 11:45:20.626442: step: 1584/533, loss: 0.06362471729516983 2023-01-21 11:45:21.821165: step: 1588/533, loss: 0.01238708570599556 2023-01-21 11:45:22.992479: step: 1592/533, loss: 0.01600179634988308 2023-01-21 11:45:24.164969: step: 1596/533, loss: 0.01671447791159153 2023-01-21 11:45:25.342906: step: 1600/533, loss: 0.5335648059844971 2023-01-21 11:45:26.495937: step: 1604/533, loss: 0.009768676944077015 2023-01-21 11:45:27.695590: step: 1608/533, loss: 0.026198577135801315 2023-01-21 11:45:28.876783: step: 1612/533, loss: 0.10621795803308487 2023-01-21 11:45:30.032441: step: 1616/533, loss: 0.027897994965314865 2023-01-21 11:45:31.190869: step: 1620/533, loss: 0.004445361904799938 2023-01-21 11:45:32.370816: step: 1624/533, loss: 0.03170432895421982 2023-01-21 11:45:33.544110: step: 1628/533, loss: 0.03311891481280327 2023-01-21 11:45:34.714806: step: 1632/533, loss: 0.022612379863858223 2023-01-21 11:45:35.880534: step: 1636/533, loss: 0.038114309310913086 2023-01-21 11:45:37.078988: step: 1640/533, loss: 0.0055901529267430305 2023-01-21 11:45:38.272278: step: 1644/533, loss: 0.0008411050075665116 2023-01-21 11:45:39.511825: step: 1648/533, loss: 0.05695762485265732 2023-01-21 11:45:40.675146: step: 1652/533, loss: 0.06915159523487091 2023-01-21 11:45:41.848400: step: 1656/533, loss: 0.0007195472717285156 2023-01-21 11:45:43.017792: step: 1660/533, loss: 0.002749919891357422 2023-01-21 11:45:44.187997: step: 1664/533, loss: 0.018449973315000534 2023-01-21 11:45:45.383466: step: 1668/533, loss: 0.2560577392578125 2023-01-21 11:45:46.627432: step: 1672/533, loss: 5.111694190418348e-05 2023-01-21 11:45:47.838765: step: 1676/533, loss: 0.018149472773075104 2023-01-21 11:45:49.036400: step: 1680/533, loss: 0.004533195402473211 2023-01-21 11:45:50.224430: step: 1684/533, loss: 0.013851833529770374 2023-01-21 11:45:51.397481: step: 1688/533, loss: 0.002399635501205921 2023-01-21 11:45:52.558401: step: 1692/533, loss: 0.021000387147068977 2023-01-21 11:45:53.714938: step: 1696/533, loss: 0.03911762312054634 2023-01-21 11:45:54.933224: step: 1700/533, loss: 0.004843330476433039 2023-01-21 11:45:56.179022: step: 1704/533, loss: 0.06415177136659622 2023-01-21 11:45:57.394925: step: 1708/533, loss: 0.031184006482362747 2023-01-21 11:45:58.590345: step: 1712/533, loss: 0.013955879025161266 2023-01-21 11:45:59.814701: step: 1716/533, loss: 0.04959449917078018 2023-01-21 11:46:00.970041: step: 1720/533, loss: 0.0045303343795239925 2023-01-21 11:46:02.159448: step: 1724/533, loss: 0.010566044598817825 2023-01-21 11:46:03.335643: step: 1728/533, loss: 0.033652305603027344 2023-01-21 11:46:04.526333: step: 1732/533, loss: 0.023145103827118874 2023-01-21 11:46:05.715012: step: 1736/533, loss: 0.003997230436652899 2023-01-21 11:46:06.868790: step: 1740/533, loss: 9.732246689964086e-05 2023-01-21 11:46:08.032067: step: 1744/533, loss: 0.0053689004853367805 2023-01-21 11:46:09.274588: step: 1748/533, loss: 0.04672088474035263 2023-01-21 11:46:10.481027: step: 1752/533, loss: 0.01916523091495037 2023-01-21 11:46:11.645187: step: 1756/533, loss: 0.017249250784516335 2023-01-21 11:46:12.849293: step: 1760/533, loss: 0.013771915808320045 2023-01-21 11:46:14.049222: step: 1764/533, loss: 0.0003450870281085372 2023-01-21 11:46:15.234122: step: 1768/533, loss: 0.0036611079704016447 2023-01-21 11:46:16.416185: step: 1772/533, loss: 0.00032405852107331157 2023-01-21 11:46:17.578826: step: 1776/533, loss: 0.0033330442383885384 2023-01-21 11:46:18.823050: step: 1780/533, loss: 0.04648284614086151 2023-01-21 11:46:20.061176: step: 1784/533, loss: 0.0020838738419115543 2023-01-21 11:46:21.265071: step: 1788/533, loss: 0.00030288693960756063 2023-01-21 11:46:22.436146: step: 1792/533, loss: 0.08451161533594131 2023-01-21 11:46:23.651469: step: 1796/533, loss: 0.03381318971514702 2023-01-21 11:46:24.841795: step: 1800/533, loss: 0.09086885303258896 2023-01-21 11:46:26.044772: step: 1804/533, loss: 0.04130687564611435 2023-01-21 11:46:27.237326: step: 1808/533, loss: 0.0025913240388035774 2023-01-21 11:46:28.414967: step: 1812/533, loss: 0.07474642246961594 2023-01-21 11:46:29.605377: step: 1816/533, loss: 0.019781876355409622 2023-01-21 11:46:30.787533: step: 1820/533, loss: 0.000301361083984375 2023-01-21 11:46:31.969325: step: 1824/533, loss: 0.0002887725713662803 2023-01-21 11:46:33.148289: step: 1828/533, loss: 0.012577342800796032 2023-01-21 11:46:34.316505: step: 1832/533, loss: 9.441375732421875e-05 2023-01-21 11:46:35.487500: step: 1836/533, loss: 0.007608985994011164 2023-01-21 11:46:36.677980: step: 1840/533, loss: 0.0354861281812191 2023-01-21 11:46:37.894383: step: 1844/533, loss: 0.0065691471099853516 2023-01-21 11:46:39.088692: step: 1848/533, loss: 0.02790858969092369 2023-01-21 11:46:40.296816: step: 1852/533, loss: 0.022379687055945396 2023-01-21 11:46:41.511010: step: 1856/533, loss: 0.0019664764404296875 2023-01-21 11:46:42.685460: step: 1860/533, loss: 0.009407997131347656 2023-01-21 11:46:43.879928: step: 1864/533, loss: 0.0008975982782430947 2023-01-21 11:46:45.056978: step: 1868/533, loss: 0.0029234886169433594 2023-01-21 11:46:46.227972: step: 1872/533, loss: 0.05127887800335884 2023-01-21 11:46:47.472919: step: 1876/533, loss: 0.005353259854018688 2023-01-21 11:46:48.656370: step: 1880/533, loss: 0.010486721992492676 2023-01-21 11:46:49.870719: step: 1884/533, loss: 0.0009550094837322831 2023-01-21 11:46:51.057785: step: 1888/533, loss: 1.3399124327406753e-05 2023-01-21 11:46:52.189734: step: 1892/533, loss: 0.04675102233886719 2023-01-21 11:46:53.418160: step: 1896/533, loss: 0.03129234537482262 2023-01-21 11:46:54.612022: step: 1900/533, loss: 0.023981522768735886 2023-01-21 11:46:55.777229: step: 1904/533, loss: 0.0007944107055664062 2023-01-21 11:46:56.989700: step: 1908/533, loss: 0.019039299339056015 2023-01-21 11:46:58.167029: step: 1912/533, loss: 0.0014998435508459806 2023-01-21 11:46:59.374638: step: 1916/533, loss: 0.04495801776647568 2023-01-21 11:47:00.577729: step: 1920/533, loss: 0.002214336534962058 2023-01-21 11:47:01.723443: step: 1924/533, loss: 0.026740361005067825 2023-01-21 11:47:02.885009: step: 1928/533, loss: 0.09192061424255371 2023-01-21 11:47:04.077175: step: 1932/533, loss: 0.04042892903089523 2023-01-21 11:47:05.253257: step: 1936/533, loss: 0.032320547848939896 2023-01-21 11:47:06.375182: step: 1940/533, loss: 0.0020843506790697575 2023-01-21 11:47:07.556175: step: 1944/533, loss: 0.05872688442468643 2023-01-21 11:47:08.722232: step: 1948/533, loss: 0.023344803601503372 2023-01-21 11:47:09.875710: step: 1952/533, loss: 0.001634883927181363 2023-01-21 11:47:11.054625: step: 1956/533, loss: 0.07500836998224258 2023-01-21 11:47:12.272896: step: 1960/533, loss: 0.011383152566850185 2023-01-21 11:47:13.458978: step: 1964/533, loss: 0.04769134521484375 2023-01-21 11:47:14.644197: step: 1968/533, loss: 0.07409077137708664 2023-01-21 11:47:15.881037: step: 1972/533, loss: 0.0036670684348791838 2023-01-21 11:47:17.072951: step: 1976/533, loss: 0.0016620636451989412 2023-01-21 11:47:18.244074: step: 1980/533, loss: 0.013271236792206764 2023-01-21 11:47:19.433330: step: 1984/533, loss: 0.002618217607960105 2023-01-21 11:47:20.584727: step: 1988/533, loss: 0.0463135726749897 2023-01-21 11:47:21.787881: step: 1992/533, loss: 0.01051025465130806 2023-01-21 11:47:22.982127: step: 1996/533, loss: 0.025898220017552376 2023-01-21 11:47:24.147629: step: 2000/533, loss: 0.10018301010131836 2023-01-21 11:47:25.352136: step: 2004/533, loss: 0.011593151837587357 2023-01-21 11:47:26.528055: step: 2008/533, loss: 0.012183571234345436 2023-01-21 11:47:27.702921: step: 2012/533, loss: 0.0626705214381218 2023-01-21 11:47:28.865958: step: 2016/533, loss: 0.004366493318229914 2023-01-21 11:47:30.032544: step: 2020/533, loss: 0.017737198621034622 2023-01-21 11:47:31.175775: step: 2024/533, loss: 0.014255713671445847 2023-01-21 11:47:32.344915: step: 2028/533, loss: 0.025807952508330345 2023-01-21 11:47:33.486729: step: 2032/533, loss: 0.011851645074784756 2023-01-21 11:47:34.647394: step: 2036/533, loss: 0.0013482094509527087 2023-01-21 11:47:35.853657: step: 2040/533, loss: 0.007368850987404585 2023-01-21 11:47:37.020224: step: 2044/533, loss: 0.008652460761368275 2023-01-21 11:47:38.177464: step: 2048/533, loss: 0.017990780994296074 2023-01-21 11:47:39.347773: step: 2052/533, loss: 0.009862900711596012 2023-01-21 11:47:40.510371: step: 2056/533, loss: 0.10551033169031143 2023-01-21 11:47:41.696411: step: 2060/533, loss: 0.00060358049813658 2023-01-21 11:47:42.891219: step: 2064/533, loss: 0.0018243789672851562 2023-01-21 11:47:44.153175: step: 2068/533, loss: 0.07365989685058594 2023-01-21 11:47:45.423675: step: 2072/533, loss: 0.0021131515968590975 2023-01-21 11:47:46.581548: step: 2076/533, loss: 0.6265827417373657 2023-01-21 11:47:47.773832: step: 2080/533, loss: 0.010940074920654297 2023-01-21 11:47:48.957104: step: 2084/533, loss: 0.021912382915616035 2023-01-21 11:47:50.151880: step: 2088/533, loss: 0.0029693127144128084 2023-01-21 11:47:51.337429: step: 2092/533, loss: 8.726119995117188e-05 2023-01-21 11:47:52.571486: step: 2096/533, loss: 0.08318972587585449 2023-01-21 11:47:53.745114: step: 2100/533, loss: 0.02552352100610733 2023-01-21 11:47:54.940958: step: 2104/533, loss: 0.009144211187958717 2023-01-21 11:47:56.178077: step: 2108/533, loss: 0.03104572184383869 2023-01-21 11:47:57.333474: step: 2112/533, loss: 0.04920797422528267 2023-01-21 11:47:58.522721: step: 2116/533, loss: 0.019005203619599342 2023-01-21 11:47:59.661850: step: 2120/533, loss: 0.016808509826660156 2023-01-21 11:48:00.844307: step: 2124/533, loss: 0.11634770035743713 2023-01-21 11:48:01.994032: step: 2128/533, loss: 0.014425563625991344 2023-01-21 11:48:03.135775: step: 2132/533, loss: 0.030226517468690872 ================================================== Loss: 0.044 -------------------- Dev: {'event': {'p': 0.5883534136546185, 'r': 0.7802929427430093, 'f1': 0.6708643388666284}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Test: {'event': {'p': 0.6426029486527707, 'r': 0.8081841432225064, 'f1': 0.7159444916454262}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Chinese: {'event': {'p': 0.5185185185185185, 'r': 0.7777777777777778, 'f1': 0.6222222222222222}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Korean: {'event': {'p': 0.660377358490566, 'r': 0.5555555555555556, 'f1': 0.6034482758620691}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Russian: {'event': {'p': 0.46153846153846156, 'r': 0.5, 'f1': 0.48000000000000004}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977130977130977, 'r': 0.7656458055925432, 'f1': 0.6713368359603036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.6403940886699507, 'r': 0.7480818414322251, 'f1': 0.690061928634621}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.55, 'r': 0.6111111111111112, 'f1': 0.5789473684210527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** 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 11:48:45.249442: step: 4/533, loss: 0.003078651614487171 2023-01-21 11:48:46.455023: step: 8/533, loss: 0.020116426050662994 2023-01-21 11:48:47.645649: step: 12/533, loss: 0.001172828720882535 2023-01-21 11:48:48.826096: step: 16/533, loss: 0.01778268814086914 2023-01-21 11:48:49.970772: step: 20/533, loss: 0.012190199457108974 2023-01-21 11:48:51.187797: step: 24/533, loss: 0.008250046521425247 2023-01-21 11:48:52.347140: step: 28/533, loss: 0.0290509220212698 2023-01-21 11:48:53.549960: step: 32/533, loss: 0.00017261505126953125 2023-01-21 11:48:54.807545: step: 36/533, loss: 0.006304740905761719 2023-01-21 11:48:55.972272: step: 40/533, loss: 7.514953904319555e-05 2023-01-21 11:48:57.134783: step: 44/533, loss: 0.009279251098632812 2023-01-21 11:48:58.398058: step: 48/533, loss: 0.044998932629823685 2023-01-21 11:48:59.553667: step: 52/533, loss: 0.04044933617115021 2023-01-21 11:49:00.743527: step: 56/533, loss: 0.0008599281427450478 2023-01-21 11:49:01.992495: step: 60/533, loss: 0.002803802490234375 2023-01-21 11:49:03.192889: step: 64/533, loss: 0.03216443210840225 2023-01-21 11:49:04.371147: step: 68/533, loss: 0.00028076174203306437 2023-01-21 11:49:05.584770: step: 72/533, loss: 0.0486719124019146 2023-01-21 11:49:06.762921: step: 76/533, loss: 0.27453774213790894 2023-01-21 11:49:07.921187: step: 80/533, loss: 0.026564789935946465 2023-01-21 11:49:09.125730: step: 84/533, loss: 0.024771692231297493 2023-01-21 11:49:10.337329: step: 88/533, loss: 0.008466911502182484 2023-01-21 11:49:11.483291: step: 92/533, loss: 0.0011946677695959806 2023-01-21 11:49:12.645107: step: 96/533, loss: 0.0038028715644031763 2023-01-21 11:49:13.864259: step: 100/533, loss: 0.0006295203929767013 2023-01-21 11:49:15.094480: step: 104/533, loss: 0.000762558018323034 2023-01-21 11:49:16.273940: step: 108/533, loss: 0.002228641649708152 2023-01-21 11:49:17.473815: step: 112/533, loss: 0.027641868218779564 2023-01-21 11:49:18.631034: step: 116/533, loss: 0.05996112525463104 2023-01-21 11:49:19.807814: step: 120/533, loss: 0.002659320831298828 2023-01-21 11:49:20.985808: step: 124/533, loss: 0.012964916415512562 2023-01-21 11:49:22.176863: step: 128/533, loss: 0.003347206162288785 2023-01-21 11:49:23.388341: step: 132/533, loss: 0.00575675955042243 2023-01-21 11:49:24.572185: step: 136/533, loss: 0.038973618298769 2023-01-21 11:49:25.717710: step: 140/533, loss: 0.00017900466627907008 2023-01-21 11:49:26.862652: step: 144/533, loss: 0.0008783340454101562 2023-01-21 11:49:28.010017: step: 148/533, loss: 0.0035134314093738794 2023-01-21 11:49:29.158449: step: 152/533, loss: 0.002569693373516202 2023-01-21 11:49:30.318792: step: 156/533, loss: 0.0291748046875 2023-01-21 11:49:31.573990: step: 160/533, loss: 0.00010786056373035535 2023-01-21 11:49:32.757631: step: 164/533, loss: 0.018441010266542435 2023-01-21 11:49:33.916314: step: 168/533, loss: 0.0010353088146075606 2023-01-21 11:49:35.102091: step: 172/533, loss: 0.03807525709271431 2023-01-21 11:49:36.305945: step: 176/533, loss: 0.013902425765991211 2023-01-21 11:49:37.437197: step: 180/533, loss: 0.02832059934735298 2023-01-21 11:49:38.651769: step: 184/533, loss: 3.06129441014491e-05 2023-01-21 11:49:39.881672: step: 188/533, loss: 0.0019176483619958162 2023-01-21 11:49:41.080868: step: 192/533, loss: 0.012591552920639515 2023-01-21 11:49:42.287781: step: 196/533, loss: 0.004140091128647327 2023-01-21 11:49:43.463991: step: 200/533, loss: 0.005182075314223766 2023-01-21 11:49:44.636695: step: 204/533, loss: 0.00023326874361373484 2023-01-21 11:49:45.789502: step: 208/533, loss: 0.2693680226802826 2023-01-21 11:49:46.933281: step: 212/533, loss: 0.007162714377045631 2023-01-21 11:49:48.118819: step: 216/533, loss: 0.08151502907276154 2023-01-21 11:49:49.336637: step: 220/533, loss: 0.016516782343387604 2023-01-21 11:49:50.534979: step: 224/533, loss: 0.011872291564941406 2023-01-21 11:49:51.742759: step: 228/533, loss: 0.020183373242616653 2023-01-21 11:49:52.920247: step: 232/533, loss: 0.0018589019309729338 2023-01-21 11:49:54.129312: step: 236/533, loss: 6.923676119185984e-05 2023-01-21 11:49:55.295990: step: 240/533, loss: 0.005663013551384211 2023-01-21 11:49:56.435495: step: 244/533, loss: 0.004565620329231024 2023-01-21 11:49:57.621993: step: 248/533, loss: 0.04617328569293022 2023-01-21 11:49:58.800217: step: 252/533, loss: 0.0018202782375738025 2023-01-21 11:50:00.007842: step: 256/533, loss: 0.024161625653505325 2023-01-21 11:50:01.145050: step: 260/533, loss: 0.0044615743681788445 2023-01-21 11:50:02.303752: step: 264/533, loss: 0.0030312538146972656 2023-01-21 11:50:03.506297: step: 268/533, loss: 0.012202071957290173 2023-01-21 11:50:04.681512: step: 272/533, loss: -1.4877317880745977e-05 2023-01-21 11:50:05.794075: step: 276/533, loss: 0.02927684783935547 2023-01-21 11:50:06.957382: step: 280/533, loss: 0.003538465593010187 2023-01-21 11:50:08.119912: step: 284/533, loss: 0.0011259079910814762 2023-01-21 11:50:09.312982: step: 288/533, loss: 0.027346229180693626 2023-01-21 11:50:10.502803: step: 292/533, loss: 0.003432083409279585 2023-01-21 11:50:11.713717: step: 296/533, loss: 0.03890342637896538 2023-01-21 11:50:12.870709: step: 300/533, loss: 0.04705004766583443 2023-01-21 11:50:14.039349: step: 304/533, loss: 0.004475212190300226 2023-01-21 11:50:15.226185: step: 308/533, loss: 0.00018997192091774195 2023-01-21 11:50:16.365011: step: 312/533, loss: 0.007609653752297163 2023-01-21 11:50:17.505058: step: 316/533, loss: 2.0360947019071318e-05 2023-01-21 11:50:18.689274: step: 320/533, loss: 0.0027098655700683594 2023-01-21 11:50:19.953981: step: 324/533, loss: 0.02936725690960884 2023-01-21 11:50:21.134824: step: 328/533, loss: 0.00278053293004632 2023-01-21 11:50:22.285924: step: 332/533, loss: 0.013946057297289371 2023-01-21 11:50:23.452814: step: 336/533, loss: 0.09047951549291611 2023-01-21 11:50:24.642561: step: 340/533, loss: 0.06602182239294052 2023-01-21 11:50:25.826049: step: 344/533, loss: 0.005893421359360218 2023-01-21 11:50:26.983374: step: 348/533, loss: 0.003125905990600586 2023-01-21 11:50:28.213943: step: 352/533, loss: 0.009057235904037952 2023-01-21 11:50:29.360624: step: 356/533, loss: 0.0009259224170818925 2023-01-21 11:50:30.529859: step: 360/533, loss: 0.011692619882524014 2023-01-21 11:50:31.732926: step: 364/533, loss: 0.026745129376649857 2023-01-21 11:50:32.901261: step: 368/533, loss: 0.0017299175960943103 2023-01-21 11:50:34.071206: step: 372/533, loss: 0.014490318484604359 2023-01-21 11:50:35.258756: step: 376/533, loss: 0.0031873704865574837 2023-01-21 11:50:36.423553: step: 380/533, loss: 0.0009585380321368575 2023-01-21 11:50:37.595240: step: 384/533, loss: 0.021773338317871094 2023-01-21 11:50:38.795154: step: 388/533, loss: 0.02523784711956978 2023-01-21 11:50:39.973918: step: 392/533, loss: 0.03062143363058567 2023-01-21 11:50:41.137419: step: 396/533, loss: 0.014317465014755726 2023-01-21 11:50:42.305945: step: 400/533, loss: 0.009530831128358841 2023-01-21 11:50:43.475535: step: 404/533, loss: 0.00107488629873842 2023-01-21 11:50:44.629221: step: 408/533, loss: 0.0005081176641397178 2023-01-21 11:50:45.809272: step: 412/533, loss: 0.04646573215723038 2023-01-21 11:50:46.998584: step: 416/533, loss: 7.286071922862902e-05 2023-01-21 11:50:48.165554: step: 420/533, loss: 0.0032161714043468237 2023-01-21 11:50:49.336530: step: 424/533, loss: 0.0007998943910934031 2023-01-21 11:50:50.513369: step: 428/533, loss: 0.02635193057358265 2023-01-21 11:50:51.693379: step: 432/533, loss: 0.06146125867962837 2023-01-21 11:50:52.881276: step: 436/533, loss: 0.01206274051219225 2023-01-21 11:50:54.038926: step: 440/533, loss: 0.015166091732680798 2023-01-21 11:50:55.231263: step: 444/533, loss: 0.034293174743652344 2023-01-21 11:50:56.391120: step: 448/533, loss: 0.008524084463715553 2023-01-21 11:50:57.558243: step: 452/533, loss: 0.005695343483239412 2023-01-21 11:50:58.739555: step: 456/533, loss: 0.0012744903797283769 2023-01-21 11:50:59.956012: step: 460/533, loss: 0.00030517578125 2023-01-21 11:51:01.143557: step: 464/533, loss: 0.0374692939221859 2023-01-21 11:51:02.335440: step: 468/533, loss: 0.008836555294692516 2023-01-21 11:51:03.545858: step: 472/533, loss: 3.8528440200025216e-05 2023-01-21 11:51:04.740209: step: 476/533, loss: 0.0033369064331054688 2023-01-21 11:51:05.928614: step: 480/533, loss: 0.01641864888370037 2023-01-21 11:51:07.097281: step: 484/533, loss: 0.04925060272216797 2023-01-21 11:51:08.268581: step: 488/533, loss: 0.008362484164536 2023-01-21 11:51:09.439993: step: 492/533, loss: 0.39628347754478455 2023-01-21 11:51:10.589311: step: 496/533, loss: 0.06452713161706924 2023-01-21 11:51:11.762772: step: 500/533, loss: 0.00809793546795845 2023-01-21 11:51:12.914721: step: 504/533, loss: 0.00017147065955214202 2023-01-21 11:51:14.091393: step: 508/533, loss: 0.035984039306640625 2023-01-21 11:51:15.304662: step: 512/533, loss: 0.08561354130506516 2023-01-21 11:51:16.481590: step: 516/533, loss: 0.0022647858131676912 2023-01-21 11:51:17.633124: step: 520/533, loss: 0.00040712355985306203 2023-01-21 11:51:18.790418: step: 524/533, loss: 0.02590971067547798 2023-01-21 11:51:19.965310: step: 528/533, loss: 0.0012377442326396704 2023-01-21 11:51:21.162355: step: 532/533, loss: 0.0007405758369714022 2023-01-21 11:51:22.326508: step: 536/533, loss: 0.0511658675968647 2023-01-21 11:51:23.519617: step: 540/533, loss: 0.0012251854641363025 2023-01-21 11:51:24.765600: step: 544/533, loss: 0.04508533328771591 2023-01-21 11:51:25.948752: step: 548/533, loss: 0.004385852720588446 2023-01-21 11:51:27.154815: step: 552/533, loss: 0.007276725955307484 2023-01-21 11:51:28.385476: step: 556/533, loss: 0.5972205400466919 2023-01-21 11:51:29.540746: step: 560/533, loss: 0.10096587985754013 2023-01-21 11:51:30.717336: step: 564/533, loss: 0.0010144234402105212 2023-01-21 11:51:31.855135: step: 568/533, loss: 0.0004378319135867059 2023-01-21 11:51:33.094325: step: 572/533, loss: 0.013380623422563076 2023-01-21 11:51:34.272748: step: 576/533, loss: 0.004177570808678865 2023-01-21 11:51:35.429461: step: 580/533, loss: 0.007358217146247625 2023-01-21 11:51:36.623950: step: 584/533, loss: 0.00415306119248271 2023-01-21 11:51:37.762403: step: 588/533, loss: 0.011286735534667969 2023-01-21 11:51:38.964009: step: 592/533, loss: 0.0008275985601358116 2023-01-21 11:51:40.124557: step: 596/533, loss: 0.0003675461048260331 2023-01-21 11:51:41.280491: step: 600/533, loss: 0.0019462585914880037 2023-01-21 11:51:42.449149: step: 604/533, loss: 0.04704933241009712 2023-01-21 11:51:43.626061: step: 608/533, loss: 0.021963119506835938 2023-01-21 11:51:44.802375: step: 612/533, loss: 0.0009529591188766062 2023-01-21 11:51:45.971190: step: 616/533, loss: 0.0005486488225869834 2023-01-21 11:51:47.180775: step: 620/533, loss: 0.0003541946643963456 2023-01-21 11:51:48.349576: step: 624/533, loss: 0.0007778167491778731 2023-01-21 11:51:49.542254: step: 628/533, loss: 0.0508481040596962 2023-01-21 11:51:50.740169: step: 632/533, loss: 0.032167911529541016 2023-01-21 11:51:51.898577: step: 636/533, loss: 0.0075897215865552425 2023-01-21 11:51:53.068519: step: 640/533, loss: 0.002895164769142866 2023-01-21 11:51:54.249956: step: 644/533, loss: 0.009769916534423828 2023-01-21 11:51:55.445303: step: 648/533, loss: 0.022771263495087624 2023-01-21 11:51:56.591047: step: 652/533, loss: 0.007764053530991077 2023-01-21 11:51:57.804470: step: 656/533, loss: 0.017029428854584694 2023-01-21 11:51:58.995136: step: 660/533, loss: 0.0016609192825853825 2023-01-21 11:52:00.185131: step: 664/533, loss: 0.001756381941959262 2023-01-21 11:52:01.398127: step: 668/533, loss: 0.021846963092684746 2023-01-21 11:52:02.586680: step: 672/533, loss: 0.05139732360839844 2023-01-21 11:52:03.759810: step: 676/533, loss: 0.10620041191577911 2023-01-21 11:52:04.924584: step: 680/533, loss: 0.0011187554337084293 2023-01-21 11:52:06.097084: step: 684/533, loss: 0.03479604795575142 2023-01-21 11:52:07.268094: step: 688/533, loss: 0.02920999750494957 2023-01-21 11:52:08.489573: step: 692/533, loss: 0.004007434472441673 2023-01-21 11:52:09.669085: step: 696/533, loss: 0.011062145233154297 2023-01-21 11:52:10.799349: step: 700/533, loss: 6.904602196300402e-05 2023-01-21 11:52:11.978008: step: 704/533, loss: 0.012273025698959827 2023-01-21 11:52:13.145507: step: 708/533, loss: 0.032195378094911575 2023-01-21 11:52:14.334932: step: 712/533, loss: 0.03704576566815376 2023-01-21 11:52:15.557476: step: 716/533, loss: 0.00537452707067132 2023-01-21 11:52:16.702100: step: 720/533, loss: 0.0506470687687397 2023-01-21 11:52:17.895909: step: 724/533, loss: -3.8146970382513246e-06 2023-01-21 11:52:19.094308: step: 728/533, loss: 0.00796289462596178 2023-01-21 11:52:20.291218: step: 732/533, loss: 0.06104631349444389 2023-01-21 11:52:21.493423: step: 736/533, loss: 0.02778659015893936 2023-01-21 11:52:22.673838: step: 740/533, loss: 0.031070711091160774 2023-01-21 11:52:23.870229: step: 744/533, loss: 0.0019535066094249487 2023-01-21 11:52:25.051302: step: 748/533, loss: 0.0007670879131183028 2023-01-21 11:52:26.224303: step: 752/533, loss: 0.012935805134475231 2023-01-21 11:52:27.432694: step: 756/533, loss: 0.006496238987892866 2023-01-21 11:52:28.628178: step: 760/533, loss: 0.05417289957404137 2023-01-21 11:52:29.820648: step: 764/533, loss: 0.016036653891205788 2023-01-21 11:52:31.010748: step: 768/533, loss: 0.0013278961414471269 2023-01-21 11:52:32.181321: step: 772/533, loss: 0.0029485702980309725 2023-01-21 11:52:33.386684: step: 776/533, loss: 0.0016141892410814762 2023-01-21 11:52:34.575339: step: 780/533, loss: 0.026542093604803085 2023-01-21 11:52:35.732670: step: 784/533, loss: 0.0016596794594079256 2023-01-21 11:52:36.945725: step: 788/533, loss: 0.00856161117553711 2023-01-21 11:52:38.126290: step: 792/533, loss: 0.010188340209424496 2023-01-21 11:52:39.355159: step: 796/533, loss: 0.03049163892865181 2023-01-21 11:52:40.505576: step: 800/533, loss: 0.016820145770907402 2023-01-21 11:52:41.690721: step: 804/533, loss: 5.7220458984375e-06 2023-01-21 11:52:42.844864: step: 808/533, loss: 0.0003688812430482358 2023-01-21 11:52:44.025264: step: 812/533, loss: 0.13359051942825317 2023-01-21 11:52:45.177601: step: 816/533, loss: 0.013157271780073643 2023-01-21 11:52:46.378831: step: 820/533, loss: 0.007832241244614124 2023-01-21 11:52:47.558847: step: 824/533, loss: 0.00125293736346066 2023-01-21 11:52:48.731474: step: 828/533, loss: 0.0002062320854747668 2023-01-21 11:52:49.887328: step: 832/533, loss: 0.028976034373044968 2023-01-21 11:52:51.044985: step: 836/533, loss: 0.00889368075877428 2023-01-21 11:52:52.192386: step: 840/533, loss: 0.02768421173095703 2023-01-21 11:52:53.328035: step: 844/533, loss: 8.010864803509321e-06 2023-01-21 11:52:54.486763: step: 848/533, loss: 0.33837267756462097 2023-01-21 11:52:55.644957: step: 852/533, loss: 0.006473541725426912 2023-01-21 11:52:56.827826: step: 856/533, loss: 0.005878258030861616 2023-01-21 11:52:57.974243: step: 860/533, loss: 0.011524581350386143 2023-01-21 11:52:59.177484: step: 864/533, loss: 0.004676878452301025 2023-01-21 11:53:00.356924: step: 868/533, loss: 0.006819725502282381 2023-01-21 11:53:01.556508: step: 872/533, loss: 0.0007096290937624872 2023-01-21 11:53:02.714250: step: 876/533, loss: 0.012262153439223766 2023-01-21 11:53:03.924093: step: 880/533, loss: 0.0032433033920824528 2023-01-21 11:53:05.080170: step: 884/533, loss: 0.06317378580570221 2023-01-21 11:53:06.288032: step: 888/533, loss: 0.022116471081972122 2023-01-21 11:53:07.445885: step: 892/533, loss: 0.001499176025390625 2023-01-21 11:53:08.592051: step: 896/533, loss: 0.02615683153271675 2023-01-21 11:53:09.749717: step: 900/533, loss: 6.780624971725047e-05 2023-01-21 11:53:10.916220: step: 904/533, loss: 0.05661144107580185 2023-01-21 11:53:12.094901: step: 908/533, loss: 0.006216239649802446 2023-01-21 11:53:13.300108: step: 912/533, loss: 0.010548114776611328 2023-01-21 11:53:14.478313: step: 916/533, loss: 0.0012091159587725997 2023-01-21 11:53:15.683775: step: 920/533, loss: 0.06478162109851837 2023-01-21 11:53:16.896203: step: 924/533, loss: 0.046471498906612396 2023-01-21 11:53:18.101537: step: 928/533, loss: 0.024771392345428467 2023-01-21 11:53:19.280522: step: 932/533, loss: 0.0794704407453537 2023-01-21 11:53:20.461673: step: 936/533, loss: 5.226135181146674e-05 2023-01-21 11:53:21.620219: step: 940/533, loss: 0.004506682977080345 2023-01-21 11:53:22.835770: step: 944/533, loss: 0.01990632526576519 2023-01-21 11:53:24.030556: step: 948/533, loss: 0.02489328570663929 2023-01-21 11:53:25.275941: step: 952/533, loss: 0.013418388552963734 2023-01-21 11:53:26.438104: step: 956/533, loss: 0.02723369561135769 2023-01-21 11:53:27.642716: step: 960/533, loss: 0.008179282769560814 2023-01-21 11:53:28.785840: step: 964/533, loss: 0.02629275433719158 2023-01-21 11:53:29.930058: step: 968/533, loss: 0.008550834842026234 2023-01-21 11:53:31.171274: step: 972/533, loss: 0.045240309089422226 2023-01-21 11:53:32.337908: step: 976/533, loss: 0.00171070103533566 2023-01-21 11:53:33.538536: step: 980/533, loss: 0.08156146854162216 2023-01-21 11:53:34.746738: step: 984/533, loss: 0.0013447761302813888 2023-01-21 11:53:35.913581: step: 988/533, loss: 0.004789638798683882 2023-01-21 11:53:37.111445: step: 992/533, loss: 0.0020843506790697575 2023-01-21 11:53:38.306234: step: 996/533, loss: 0.019885826855897903 2023-01-21 11:53:39.470727: step: 1000/533, loss: 0.007515716832131147 2023-01-21 11:53:40.673875: step: 1004/533, loss: 0.007029056549072266 2023-01-21 11:53:41.873165: step: 1008/533, loss: 0.0129852294921875 2023-01-21 11:53:43.036946: step: 1012/533, loss: 0.018865205347537994 2023-01-21 11:53:44.223541: step: 1016/533, loss: 0.00030233862344175577 2023-01-21 11:53:45.391969: step: 1020/533, loss: 0.09663896262645721 2023-01-21 11:53:46.581374: step: 1024/533, loss: 0.06732263416051865 2023-01-21 11:53:47.784498: step: 1028/533, loss: 0.01478595845401287 2023-01-21 11:53:48.922508: step: 1032/533, loss: 0.0420168861746788 2023-01-21 11:53:50.084963: step: 1036/533, loss: 0.07473564147949219 2023-01-21 11:53:51.227728: step: 1040/533, loss: 0.019064616411924362 2023-01-21 11:53:52.410735: step: 1044/533, loss: 0.13124370574951172 2023-01-21 11:53:53.555996: step: 1048/533, loss: 0.005586433224380016 2023-01-21 11:53:54.726150: step: 1052/533, loss: 0.023783493787050247 2023-01-21 11:53:55.907365: step: 1056/533, loss: 0.004500770941376686 2023-01-21 11:53:57.062843: step: 1060/533, loss: 0.06214962154626846 2023-01-21 11:53:58.224443: step: 1064/533, loss: 0.009563160128891468 2023-01-21 11:53:59.438171: step: 1068/533, loss: 0.06312179565429688 2023-01-21 11:54:00.623640: step: 1072/533, loss: 0.03809795901179314 2023-01-21 11:54:01.757709: step: 1076/533, loss: 0.004991292953491211 2023-01-21 11:54:02.892294: step: 1080/533, loss: 0.016753865405917168 2023-01-21 11:54:04.051370: step: 1084/533, loss: 0.022898579016327858 2023-01-21 11:54:05.302763: step: 1088/533, loss: 0.012741565704345703 2023-01-21 11:54:06.485440: step: 1092/533, loss: 0.15026415884494781 2023-01-21 11:54:07.656385: step: 1096/533, loss: 0.003010272979736328 2023-01-21 11:54:08.836929: step: 1100/533, loss: 0.016104698181152344 2023-01-21 11:54:10.039458: step: 1104/533, loss: 0.005273390095680952 2023-01-21 11:54:11.220334: step: 1108/533, loss: 0.0022059441544115543 2023-01-21 11:54:12.393803: step: 1112/533, loss: 0.011533929035067558 2023-01-21 11:54:13.585843: step: 1116/533, loss: 0.00527610769495368 2023-01-21 11:54:14.795749: step: 1120/533, loss: 0.008863449096679688 2023-01-21 11:54:15.974927: step: 1124/533, loss: 0.13792219758033752 2023-01-21 11:54:17.112848: step: 1128/533, loss: 0.0013053894508630037 2023-01-21 11:54:18.306644: step: 1132/533, loss: 0.004211998078972101 2023-01-21 11:54:19.518121: step: 1136/533, loss: 0.03299827501177788 2023-01-21 11:54:20.674923: step: 1140/533, loss: 0.03654613718390465 2023-01-21 11:54:21.832586: step: 1144/533, loss: 0.017261218279600143 2023-01-21 11:54:23.024649: step: 1148/533, loss: 0.024605702608823776 2023-01-21 11:54:24.202021: step: 1152/533, loss: 0.005193901248276234 2023-01-21 11:54:25.348775: step: 1156/533, loss: 0.0039022446144372225 2023-01-21 11:54:26.543981: step: 1160/533, loss: 0.016078472137451172 2023-01-21 11:54:27.750001: step: 1164/533, loss: 0.034882355481386185 2023-01-21 11:54:28.920598: step: 1168/533, loss: 0.0022861480247229338 2023-01-21 11:54:30.093992: step: 1172/533, loss: 6.87599167576991e-05 2023-01-21 11:54:31.260119: step: 1176/533, loss: 0.047203633934259415 2023-01-21 11:54:32.402473: step: 1180/533, loss: 0.012865830212831497 2023-01-21 11:54:33.586226: step: 1184/533, loss: 0.008940315805375576 2023-01-21 11:54:34.766707: step: 1188/533, loss: 0.03618317097425461 2023-01-21 11:54:35.934445: step: 1192/533, loss: 0.0059108734130859375 2023-01-21 11:54:37.105675: step: 1196/533, loss: 0.0028356551192700863 2023-01-21 11:54:38.269360: step: 1200/533, loss: 0.0007039070478640497 2023-01-21 11:54:39.437713: step: 1204/533, loss: 0.0004631996271200478 2023-01-21 11:54:40.621073: step: 1208/533, loss: 0.0008699417230673134 2023-01-21 11:54:41.820768: step: 1212/533, loss: 0.018605614081025124 2023-01-21 11:54:43.000605: step: 1216/533, loss: 0.013643073849380016 2023-01-21 11:54:44.171877: step: 1220/533, loss: 0.03987293317914009 2023-01-21 11:54:45.300995: step: 1224/533, loss: 0.005715799517929554 2023-01-21 11:54:46.505148: step: 1228/533, loss: 0.01930828206241131 2023-01-21 11:54:47.684583: step: 1232/533, loss: 0.0009300232632085681 2023-01-21 11:54:48.858317: step: 1236/533, loss: 0.03018360212445259 2023-01-21 11:54:50.043910: step: 1240/533, loss: 0.037119388580322266 2023-01-21 11:54:51.248441: step: 1244/533, loss: 0.03416137769818306 2023-01-21 11:54:52.407003: step: 1248/533, loss: 0.00020008085994049907 2023-01-21 11:54:53.552933: step: 1252/533, loss: 0.00022554397583007812 2023-01-21 11:54:54.749840: step: 1256/533, loss: 0.010517645627260208 2023-01-21 11:54:55.917064: step: 1260/533, loss: 0.01484232023358345 2023-01-21 11:54:57.116155: step: 1264/533, loss: 0.013745211996138096 2023-01-21 11:54:58.280031: step: 1268/533, loss: 5.016326758777723e-05 2023-01-21 11:54:59.454921: step: 1272/533, loss: 0.0025022507179528475 2023-01-21 11:55:00.662750: step: 1276/533, loss: 0.0003076553111895919 2023-01-21 11:55:01.854222: step: 1280/533, loss: 0.00577468890696764 2023-01-21 11:55:03.035743: step: 1284/533, loss: 0.0010597228538244963 2023-01-21 11:55:04.239224: step: 1288/533, loss: 0.005928707309067249 2023-01-21 11:55:05.427356: step: 1292/533, loss: 0.022014712914824486 2023-01-21 11:55:06.597372: step: 1296/533, loss: 0.05559501424431801 2023-01-21 11:55:07.768510: step: 1300/533, loss: 0.011143875308334827 2023-01-21 11:55:08.938066: step: 1304/533, loss: 0.006691360380500555 2023-01-21 11:55:10.106822: step: 1308/533, loss: 0.004321432206779718 2023-01-21 11:55:11.256246: step: 1312/533, loss: 0.009820937179028988 2023-01-21 11:55:12.405540: step: 1316/533, loss: 0.0040412903763353825 2023-01-21 11:55:13.625341: step: 1320/533, loss: 0.04025459289550781 2023-01-21 11:55:14.789272: step: 1324/533, loss: 0.012469768524169922 2023-01-21 11:55:15.997705: step: 1328/533, loss: 0.049068763852119446 2023-01-21 11:55:17.216098: step: 1332/533, loss: 0.016312599182128906 2023-01-21 11:55:18.411346: step: 1336/533, loss: 0.0009770393371582031 2023-01-21 11:55:19.583958: step: 1340/533, loss: 0.04604502022266388 2023-01-21 11:55:20.806034: step: 1344/533, loss: 0.03741512447595596 2023-01-21 11:55:21.992984: step: 1348/533, loss: 0.5269002914428711 2023-01-21 11:55:23.126883: step: 1352/533, loss: 0.0010107994312420487 2023-01-21 11:55:24.315966: step: 1356/533, loss: 0.0008288383833132684 2023-01-21 11:55:25.457678: step: 1360/533, loss: 0.0003665924014057964 2023-01-21 11:55:26.646919: step: 1364/533, loss: 0.00013344288163352758 2023-01-21 11:55:27.826403: step: 1368/533, loss: 0.04731407389044762 2023-01-21 11:55:29.004197: step: 1372/533, loss: 0.004719066899269819 2023-01-21 11:55:30.156766: step: 1376/533, loss: 0.002026286907494068 2023-01-21 11:55:31.290988: step: 1380/533, loss: 0.006644344422966242 2023-01-21 11:55:32.464832: step: 1384/533, loss: 0.0011403083335608244 2023-01-21 11:55:33.671792: step: 1388/533, loss: 0.03219585865736008 2023-01-21 11:55:34.842470: step: 1392/533, loss: 0.011390590108931065 2023-01-21 11:55:36.012458: step: 1396/533, loss: 0.059751417487859726 2023-01-21 11:55:37.169466: step: 1400/533, loss: 0.0038070681039243937 2023-01-21 11:55:38.353071: step: 1404/533, loss: 0.004620933905243874 2023-01-21 11:55:39.493831: step: 1408/533, loss: 0.005663537885993719 2023-01-21 11:55:40.704733: step: 1412/533, loss: 0.001876068185083568 2023-01-21 11:55:41.904884: step: 1416/533, loss: 0.07971706241369247 2023-01-21 11:55:43.110214: step: 1420/533, loss: 0.01218566857278347 2023-01-21 11:55:44.335180: step: 1424/533, loss: 0.01216878928244114 2023-01-21 11:55:45.490418: step: 1428/533, loss: 0.05049515143036842 2023-01-21 11:55:46.684643: step: 1432/533, loss: 0.048594094812870026 2023-01-21 11:55:47.856137: step: 1436/533, loss: 0.029944991692900658 2023-01-21 11:55:49.036435: step: 1440/533, loss: 0.005917740054428577 2023-01-21 11:55:50.228475: step: 1444/533, loss: 0.0016349792713299394 2023-01-21 11:55:51.419132: step: 1448/533, loss: 0.024352073669433594 2023-01-21 11:55:52.572129: step: 1452/533, loss: 0.0016530037391930819 2023-01-21 11:55:53.802625: step: 1456/533, loss: 0.002508545061573386 2023-01-21 11:55:54.967438: step: 1460/533, loss: 0.0540466345846653 2023-01-21 11:55:56.119724: step: 1464/533, loss: 0.007477569859474897 2023-01-21 11:55:57.294851: step: 1468/533, loss: 0.03422980010509491 2023-01-21 11:55:58.429011: step: 1472/533, loss: 0.0006747245788574219 2023-01-21 11:55:59.586247: step: 1476/533, loss: 0.0018912197556346655 2023-01-21 11:56:00.775644: step: 1480/533, loss: 0.0799618735909462 2023-01-21 11:56:01.894548: step: 1484/533, loss: 0.0011687278747558594 2023-01-21 11:56:03.085104: step: 1488/533, loss: 0.00216922746039927 2023-01-21 11:56:04.246691: step: 1492/533, loss: 0.04210357740521431 2023-01-21 11:56:05.412629: step: 1496/533, loss: 0.01060781441628933 2023-01-21 11:56:06.578088: step: 1500/533, loss: 0.028268814086914062 2023-01-21 11:56:07.739645: step: 1504/533, loss: 0.0071805003099143505 2023-01-21 11:56:08.913432: step: 1508/533, loss: 0.03542815148830414 2023-01-21 11:56:10.103775: step: 1512/533, loss: 0.018245410174131393 2023-01-21 11:56:11.262624: step: 1516/533, loss: 0.007300567347556353 2023-01-21 11:56:12.436847: step: 1520/533, loss: 0.004657936282455921 2023-01-21 11:56:13.626240: step: 1524/533, loss: 0.0015552521217614412 2023-01-21 11:56:14.856518: step: 1528/533, loss: 0.013571548275649548 2023-01-21 11:56:16.034304: step: 1532/533, loss: 0.00352821359410882 2023-01-21 11:56:17.218656: step: 1536/533, loss: 0.0027129172813147306 2023-01-21 11:56:18.426330: step: 1540/533, loss: 0.036319371312856674 2023-01-21 11:56:19.598457: step: 1544/533, loss: 0.03829554468393326 2023-01-21 11:56:20.800382: step: 1548/533, loss: 0.008475875481963158 2023-01-21 11:56:22.034133: step: 1552/533, loss: 0.0070514678955078125 2023-01-21 11:56:23.220438: step: 1556/533, loss: 0.0005015373462811112 2023-01-21 11:56:24.397628: step: 1560/533, loss: 0.04219703748822212 2023-01-21 11:56:25.597041: step: 1564/533, loss: 0.02217083051800728 2023-01-21 11:56:26.799482: step: 1568/533, loss: 0.004139900207519531 2023-01-21 11:56:27.966639: step: 1572/533, loss: 7.07626313669607e-05 2023-01-21 11:56:29.106781: step: 1576/533, loss: 0.027157379314303398 2023-01-21 11:56:30.266296: step: 1580/533, loss: 0.14897410571575165 2023-01-21 11:56:31.431253: step: 1584/533, loss: 0.005349826533347368 2023-01-21 11:56:32.622006: step: 1588/533, loss: 0.010769559070467949 2023-01-21 11:56:33.813661: step: 1592/533, loss: 0.02519550547003746 2023-01-21 11:56:35.024030: step: 1596/533, loss: 0.04316768795251846 2023-01-21 11:56:36.193302: step: 1600/533, loss: 0.0012823105789721012 2023-01-21 11:56:37.354032: step: 1604/533, loss: 0.015904640778899193 2023-01-21 11:56:38.530087: step: 1608/533, loss: 0.0019223212730139494 2023-01-21 11:56:39.725763: step: 1612/533, loss: 0.005879020784050226 2023-01-21 11:56:40.886761: step: 1616/533, loss: 0.0035712718963623047 2023-01-21 11:56:42.072403: step: 1620/533, loss: 0.023318292573094368 2023-01-21 11:56:43.247448: step: 1624/533, loss: 0.009914744645357132 2023-01-21 11:56:44.408451: step: 1628/533, loss: 0.0034047127701342106 2023-01-21 11:56:45.596506: step: 1632/533, loss: 0.014000892639160156 2023-01-21 11:56:46.776835: step: 1636/533, loss: 0.0201581958681345 2023-01-21 11:56:47.987218: step: 1640/533, loss: 0.0027790546882897615 2023-01-21 11:56:49.180988: step: 1644/533, loss: 0.0003223419189453125 2023-01-21 11:56:50.359923: step: 1648/533, loss: 0.008738135918974876 2023-01-21 11:56:51.512867: step: 1652/533, loss: 0.06151562184095383 2023-01-21 11:56:52.715190: step: 1656/533, loss: 0.0006672859308309853 2023-01-21 11:56:53.824640: step: 1660/533, loss: 0.006988143548369408 2023-01-21 11:56:55.017109: step: 1664/533, loss: 0.0005323886871337891 2023-01-21 11:56:56.206944: step: 1668/533, loss: 0.0012575149303302169 2023-01-21 11:56:57.375535: step: 1672/533, loss: 0.0011177064152434468 2023-01-21 11:56:58.537820: step: 1676/533, loss: 0.05244407430291176 2023-01-21 11:56:59.708801: step: 1680/533, loss: 0.006606960203498602 2023-01-21 11:57:00.933966: step: 1684/533, loss: 0.00037298203096725047 2023-01-21 11:57:02.087288: step: 1688/533, loss: 0.0564691536128521 2023-01-21 11:57:03.283061: step: 1692/533, loss: 0.015074063092470169 2023-01-21 11:57:04.476974: step: 1696/533, loss: 0.014576150104403496 2023-01-21 11:57:05.653855: step: 1700/533, loss: 0.0021986961364746094 2023-01-21 11:57:06.818359: step: 1704/533, loss: 0.003827739041298628 2023-01-21 11:57:07.995566: step: 1708/533, loss: 0.001714515732601285 2023-01-21 11:57:09.168190: step: 1712/533, loss: 0.022550201043486595 2023-01-21 11:57:10.333161: step: 1716/533, loss: 0.17543230950832367 2023-01-21 11:57:11.483442: step: 1720/533, loss: 0.014978217892348766 2023-01-21 11:57:12.645040: step: 1724/533, loss: 0.008336162194609642 2023-01-21 11:57:13.833612: step: 1728/533, loss: 0.0009703636169433594 2023-01-21 11:57:15.004515: step: 1732/533, loss: 0.0013994217151775956 2023-01-21 11:57:16.231015: step: 1736/533, loss: 0.032283976674079895 2023-01-21 11:57:17.366791: step: 1740/533, loss: 0.00791158713400364 2023-01-21 11:57:18.527813: step: 1744/533, loss: 0.0012325287098065019 2023-01-21 11:57:19.695419: step: 1748/533, loss: 0.1075240969657898 2023-01-21 11:57:20.899302: step: 1752/533, loss: 0.019856836646795273 2023-01-21 11:57:22.059602: step: 1756/533, loss: 0.005785751156508923 2023-01-21 11:57:23.199013: step: 1760/533, loss: 0.0017606735927984118 2023-01-21 11:57:24.417476: step: 1764/533, loss: 0.040729716420173645 2023-01-21 11:57:25.604855: step: 1768/533, loss: 0.0003330230829305947 2023-01-21 11:57:26.765326: step: 1772/533, loss: 0.003000545548275113 2023-01-21 11:57:27.913143: step: 1776/533, loss: 0.12473531067371368 2023-01-21 11:57:29.087634: step: 1780/533, loss: 0.011386299505829811 2023-01-21 11:57:30.250385: step: 1784/533, loss: 0.009250640869140625 2023-01-21 11:57:31.435127: step: 1788/533, loss: 0.035471536219120026 2023-01-21 11:57:32.613763: step: 1792/533, loss: 0.054877471178770065 2023-01-21 11:57:33.782235: step: 1796/533, loss: 0.009039211086928844 2023-01-21 11:57:34.986195: step: 1800/533, loss: 0.03161327913403511 2023-01-21 11:57:36.124299: step: 1804/533, loss: 0.0035272599197924137 2023-01-21 11:57:37.274054: step: 1808/533, loss: 0.012872028164565563 2023-01-21 11:57:38.461314: step: 1812/533, loss: 0.00036458970862440765 2023-01-21 11:57:39.665453: step: 1816/533, loss: 0.010746288113296032 2023-01-21 11:57:40.847853: step: 1820/533, loss: 0.04239597171545029 2023-01-21 11:57:41.988587: step: 1824/533, loss: 0.005365753546357155 2023-01-21 11:57:43.205259: step: 1828/533, loss: 0.005748176481574774 2023-01-21 11:57:44.382100: step: 1832/533, loss: 0.13444003462791443 2023-01-21 11:57:45.544221: step: 1836/533, loss: 0.038729477673769 2023-01-21 11:57:46.686777: step: 1840/533, loss: 0.004156065173447132 2023-01-21 11:57:47.875294: step: 1844/533, loss: 0.011990547180175781 2023-01-21 11:57:49.075805: step: 1848/533, loss: 0.5351228713989258 2023-01-21 11:57:50.251425: step: 1852/533, loss: 0.00042018890962935984 2023-01-21 11:57:51.402023: step: 1856/533, loss: 0.0038950920570641756 2023-01-21 11:57:52.578787: step: 1860/533, loss: 0.0020399093627929688 2023-01-21 11:57:53.748873: step: 1864/533, loss: 0.016020776703953743 2023-01-21 11:57:54.886328: step: 1868/533, loss: 0.0003823757288046181 2023-01-21 11:57:56.050794: step: 1872/533, loss: 0.160267174243927 2023-01-21 11:57:57.248563: step: 1876/533, loss: 0.055043794214725494 2023-01-21 11:57:58.386475: step: 1880/533, loss: 0.004176521208137274 2023-01-21 11:57:59.546159: step: 1884/533, loss: 0.0003499031299725175 2023-01-21 11:58:00.747013: step: 1888/533, loss: 0.03513612598180771 2023-01-21 11:58:01.912833: step: 1892/533, loss: 0.010804463177919388 2023-01-21 11:58:03.071117: step: 1896/533, loss: 0.004789483733475208 2023-01-21 11:58:04.228786: step: 1900/533, loss: 0.0002534866507630795 2023-01-21 11:58:05.394473: step: 1904/533, loss: 0.002976846881210804 2023-01-21 11:58:06.569995: step: 1908/533, loss: 0.022188186645507812 2023-01-21 11:58:07.757683: step: 1912/533, loss: 0.042273327708244324 2023-01-21 11:58:08.981239: step: 1916/533, loss: 0.0001011848435155116 2023-01-21 11:58:10.181381: step: 1920/533, loss: 0.006221294403076172 2023-01-21 11:58:11.348667: step: 1924/533, loss: 0.04015479236841202 2023-01-21 11:58:12.518358: step: 1928/533, loss: 0.02574901655316353 2023-01-21 11:58:13.677438: step: 1932/533, loss: 0.03933687135577202 2023-01-21 11:58:14.834972: step: 1936/533, loss: 0.00240325927734375 2023-01-21 11:58:16.052476: step: 1940/533, loss: 0.0028482438065111637 2023-01-21 11:58:17.238324: step: 1944/533, loss: 0.013387490063905716 2023-01-21 11:58:18.405787: step: 1948/533, loss: 0.002329874085262418 2023-01-21 11:58:19.596517: step: 1952/533, loss: 0.01641235314309597 2023-01-21 11:58:20.797362: step: 1956/533, loss: 0.017055273056030273 2023-01-21 11:58:21.967685: step: 1960/533, loss: 0.0067986492067575455 2023-01-21 11:58:23.152190: step: 1964/533, loss: 0.03813019022345543 2023-01-21 11:58:24.359092: step: 1968/533, loss: 0.11629810184240341 2023-01-21 11:58:25.550670: step: 1972/533, loss: 0.002849387936294079 2023-01-21 11:58:26.727253: step: 1976/533, loss: 0.0020013332832604647 2023-01-21 11:58:27.900406: step: 1980/533, loss: 0.00494384765625 2023-01-21 11:58:29.081473: step: 1984/533, loss: 0.001073646591976285 2023-01-21 11:58:30.266348: step: 1988/533, loss: 0.04678606986999512 2023-01-21 11:58:31.436692: step: 1992/533, loss: 0.022136688232421875 2023-01-21 11:58:32.620104: step: 1996/533, loss: 0.09453306347131729 2023-01-21 11:58:33.772320: step: 2000/533, loss: 0.008246994577348232 2023-01-21 11:58:34.928191: step: 2004/533, loss: 0.002620697021484375 2023-01-21 11:58:36.090795: step: 2008/533, loss: 0.08653044700622559 2023-01-21 11:58:37.237223: step: 2012/533, loss: 0.0023775578010827303 2023-01-21 11:58:38.389892: step: 2016/533, loss: 0.023912716656923294 2023-01-21 11:58:39.548487: step: 2020/533, loss: 0.003780317259952426 2023-01-21 11:58:40.739391: step: 2024/533, loss: 0.023583412170410156 2023-01-21 11:58:41.937527: step: 2028/533, loss: 0.00495491037145257 2023-01-21 11:58:43.096009: step: 2032/533, loss: 0.03590545803308487 2023-01-21 11:58:44.297567: step: 2036/533, loss: 0.004186439327895641 2023-01-21 11:58:45.438207: step: 2040/533, loss: 0.00045447348384186625 2023-01-21 11:58:46.635446: step: 2044/533, loss: 0.015240669250488281 2023-01-21 11:58:47.838347: step: 2048/533, loss: 0.008154154755175114 2023-01-21 11:58:49.021697: step: 2052/533, loss: 6.160736666060984e-05 2023-01-21 11:58:50.223603: step: 2056/533, loss: 0.008100509643554688 2023-01-21 11:58:51.402841: step: 2060/533, loss: 0.02331714704632759 2023-01-21 11:58:52.550959: step: 2064/533, loss: 4.8344292640686035 2023-01-21 11:58:53.727750: step: 2068/533, loss: 0.0019422532059252262 2023-01-21 11:58:54.883519: step: 2072/533, loss: 0.0006623268709518015 2023-01-21 11:58:56.084952: step: 2076/533, loss: 8.454322960460559e-05 2023-01-21 11:58:57.246590: step: 2080/533, loss: 0.023875214159488678 2023-01-21 11:58:58.472459: step: 2084/533, loss: 0.0006901741144247353 2023-01-21 11:58:59.637402: step: 2088/533, loss: 0.026457881554961205 2023-01-21 11:59:00.798339: step: 2092/533, loss: 0.013144207186996937 2023-01-21 11:59:01.998827: step: 2096/533, loss: 0.011030007153749466 2023-01-21 11:59:03.196223: step: 2100/533, loss: 0.03676357492804527 2023-01-21 11:59:04.348470: step: 2104/533, loss: 0.00734214810654521 2023-01-21 11:59:05.509741: step: 2108/533, loss: 0.01333236787468195 2023-01-21 11:59:06.650768: step: 2112/533, loss: 0.0020069123711436987 2023-01-21 11:59:07.823584: step: 2116/533, loss: 0.03376322239637375 2023-01-21 11:59:09.028695: step: 2120/533, loss: 0.01103897113353014 2023-01-21 11:59:10.180801: step: 2124/533, loss: 0.0008355140453204513 2023-01-21 11:59:11.375890: step: 2128/533, loss: 0.005996895022690296 2023-01-21 11:59:12.530903: step: 2132/533, loss: 0.008705330081284046 ================================================== Loss: 0.033 -------------------- Dev: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Test: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Chinese: {'event': {'p': 0.5287356321839081, 'r': 0.8518518518518519, 'f1': 0.6524822695035462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Korean: {'event': {'p': 0.7115384615384616, 'r': 0.5873015873015873, 'f1': 0.6434782608695652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:00:02.007302: step: 4/533, loss: 0.0005886077997274697 2023-01-21 12:00:03.184256: step: 8/533, loss: 0.0020846365951001644 2023-01-21 12:00:04.351942: step: 12/533, loss: 0.06853103637695312 2023-01-21 12:00:05.495961: step: 16/533, loss: 0.0003220558282919228 2023-01-21 12:00:06.647578: step: 20/533, loss: 0.013419532217085361 2023-01-21 12:00:07.817909: step: 24/533, loss: 0.004877281375229359 2023-01-21 12:00:08.993322: step: 28/533, loss: 0.04770174250006676 2023-01-21 12:00:10.232643: step: 32/533, loss: 0.002239227294921875 2023-01-21 12:00:11.408182: step: 36/533, loss: 0.00337486295029521 2023-01-21 12:00:12.613649: step: 40/533, loss: 0.005089664366096258 2023-01-21 12:00:13.782241: step: 44/533, loss: 0.0008498192764818668 2023-01-21 12:00:14.973055: step: 48/533, loss: 0.012123679742217064 2023-01-21 12:00:16.127969: step: 52/533, loss: 0.046187594532966614 2023-01-21 12:00:17.312362: step: 56/533, loss: 0.018197156488895416 2023-01-21 12:00:18.499466: step: 60/533, loss: 0.015445232391357422 2023-01-21 12:00:19.718663: step: 64/533, loss: 0.03211021423339844 2023-01-21 12:00:20.912959: step: 68/533, loss: 0.00043125153752043843 2023-01-21 12:00:22.115033: step: 72/533, loss: 5.359649730962701e-05 2023-01-21 12:00:23.281284: step: 76/533, loss: 0.0009490966913290322 2023-01-21 12:00:24.435797: step: 80/533, loss: 0.037011146545410156 2023-01-21 12:00:25.602921: step: 84/533, loss: 0.00044002532376907766 2023-01-21 12:00:26.786007: step: 88/533, loss: 0.17801494896411896 2023-01-21 12:00:27.915886: step: 92/533, loss: 0.0012642861111089587 2023-01-21 12:00:29.063313: step: 96/533, loss: 0.025786591693758965 2023-01-21 12:00:30.200655: step: 100/533, loss: 0.0003948211669921875 2023-01-21 12:00:31.379910: step: 104/533, loss: 0.0006008148193359375 2023-01-21 12:00:32.576381: step: 108/533, loss: 5.309581683832221e-05 2023-01-21 12:00:33.769358: step: 112/533, loss: 0.006973231211304665 2023-01-21 12:00:34.901718: step: 116/533, loss: 7.009506134636467e-06 2023-01-21 12:00:36.046391: step: 120/533, loss: 0.016931915655732155 2023-01-21 12:00:37.257765: step: 124/533, loss: 0.01997542567551136 2023-01-21 12:00:38.427129: step: 128/533, loss: 0.0645233616232872 2023-01-21 12:00:39.562260: step: 132/533, loss: 0.010686302557587624 2023-01-21 12:00:40.736889: step: 136/533, loss: 0.042328834533691406 2023-01-21 12:00:41.891432: step: 140/533, loss: 0.013307953253388405 2023-01-21 12:00:43.041282: step: 144/533, loss: 0.0037047863006591797 2023-01-21 12:00:44.201466: step: 148/533, loss: 0.002474021865054965 2023-01-21 12:00:45.341344: step: 152/533, loss: 0.034833334386348724 2023-01-21 12:00:46.511510: step: 156/533, loss: 0.00022029876708984375 2023-01-21 12:00:47.698655: step: 160/533, loss: 0.039119720458984375 2023-01-21 12:00:48.829927: step: 164/533, loss: 0.00017089844914153218 2023-01-21 12:00:50.028863: step: 168/533, loss: 0.002909374190494418 2023-01-21 12:00:51.201171: step: 172/533, loss: 0.002466106554493308 2023-01-21 12:00:52.354844: step: 176/533, loss: 0.0013028144603595138 2023-01-21 12:00:53.555739: step: 180/533, loss: 0.03087158128619194 2023-01-21 12:00:54.731543: step: 184/533, loss: 0.08638477325439453 2023-01-21 12:00:55.867739: step: 188/533, loss: 0.04150676727294922 2023-01-21 12:00:57.063430: step: 192/533, loss: 0.05215311422944069 2023-01-21 12:00:58.251124: step: 196/533, loss: 0.008701801300048828 2023-01-21 12:00:59.403159: step: 200/533, loss: 0.00032444001408293843 2023-01-21 12:01:00.573573: step: 204/533, loss: 0.030159424990415573 2023-01-21 12:01:01.753333: step: 208/533, loss: 0.04551716148853302 2023-01-21 12:01:02.918221: step: 212/533, loss: 0.0031419754959642887 2023-01-21 12:01:04.101182: step: 216/533, loss: 0.00016069412231445312 2023-01-21 12:01:05.289416: step: 220/533, loss: 0.0012417794205248356 2023-01-21 12:01:06.466546: step: 224/533, loss: 0.003115367842838168 2023-01-21 12:01:07.609264: step: 228/533, loss: 0.005041694734245539 2023-01-21 12:01:08.779071: step: 232/533, loss: 0.018837736919522285 2023-01-21 12:01:09.922928: step: 236/533, loss: 0.0005652428371831775 2023-01-21 12:01:11.116558: step: 240/533, loss: 0.004841423127800226 2023-01-21 12:01:12.314769: step: 244/533, loss: 0.024548912420868874 2023-01-21 12:01:13.480477: step: 248/533, loss: 0.0027175426948815584 2023-01-21 12:01:14.644281: step: 252/533, loss: 0.0006651878356933594 2023-01-21 12:01:15.833088: step: 256/533, loss: 0.051253221929073334 2023-01-21 12:01:17.003117: step: 260/533, loss: 0.0347231887280941 2023-01-21 12:01:18.177625: step: 264/533, loss: 8.420944504905492e-05 2023-01-21 12:01:19.390864: step: 268/533, loss: 0.005065918434411287 2023-01-21 12:01:20.578091: step: 272/533, loss: 0.01858234405517578 2023-01-21 12:01:21.858765: step: 276/533, loss: 0.3724966049194336 2023-01-21 12:01:23.023006: step: 280/533, loss: 0.0010100125800818205 2023-01-21 12:01:24.193610: step: 284/533, loss: 0.012977314181625843 2023-01-21 12:01:25.404747: step: 288/533, loss: 0.0003398895205464214 2023-01-21 12:01:26.558952: step: 292/533, loss: 0.04905519634485245 2023-01-21 12:01:27.718850: step: 296/533, loss: 0.11014318466186523 2023-01-21 12:01:28.956753: step: 300/533, loss: 0.0010835647117346525 2023-01-21 12:01:30.090313: step: 304/533, loss: 6.0749054682673886e-05 2023-01-21 12:01:31.255303: step: 308/533, loss: 0.00024385451979469508 2023-01-21 12:01:32.405102: step: 312/533, loss: 0.0003857612609863281 2023-01-21 12:01:33.583304: step: 316/533, loss: 0.06520500034093857 2023-01-21 12:01:34.770658: step: 320/533, loss: 0.0005734444130212069 2023-01-21 12:01:35.957567: step: 324/533, loss: 0.02917947992682457 2023-01-21 12:01:37.079636: step: 328/533, loss: 0.017774105072021484 2023-01-21 12:01:38.223813: step: 332/533, loss: 0.00014009475125931203 2023-01-21 12:01:39.410780: step: 336/533, loss: 6.723403930664062e-05 2023-01-21 12:01:40.607691: step: 340/533, loss: 0.0009504318004474044 2023-01-21 12:01:41.791150: step: 344/533, loss: 0.0014159202110022306 2023-01-21 12:01:42.974503: step: 348/533, loss: 0.006699276156723499 2023-01-21 12:01:44.133448: step: 352/533, loss: 0.02167348936200142 2023-01-21 12:01:45.305792: step: 356/533, loss: 0.00017671585374046117 2023-01-21 12:01:46.469679: step: 360/533, loss: 0.007201528642326593 2023-01-21 12:01:47.656714: step: 364/533, loss: 0.0010522842640057206 2023-01-21 12:01:48.819890: step: 368/533, loss: 0.014066696166992188 2023-01-21 12:01:50.037847: step: 372/533, loss: 0.0008108139154501259 2023-01-21 12:01:51.212288: step: 376/533, loss: 0.005213260650634766 2023-01-21 12:01:52.371738: step: 380/533, loss: 0.08440346270799637 2023-01-21 12:01:53.505916: step: 384/533, loss: 0.0059181214310228825 2023-01-21 12:01:54.720204: step: 388/533, loss: 0.0453365333378315 2023-01-21 12:01:55.850482: step: 392/533, loss: 0.02545952796936035 2023-01-21 12:01:57.036531: step: 396/533, loss: 0.0031192777678370476 2023-01-21 12:01:58.224204: step: 400/533, loss: 0.004544734954833984 2023-01-21 12:01:59.395159: step: 404/533, loss: 0.0013781547313556075 2023-01-21 12:02:00.595922: step: 408/533, loss: 0.027086447924375534 2023-01-21 12:02:01.761179: step: 412/533, loss: 0.0352291576564312 2023-01-21 12:02:02.911715: step: 416/533, loss: 0.6356148719787598 2023-01-21 12:02:04.079442: step: 420/533, loss: 0.10403952747583389 2023-01-21 12:02:05.250778: step: 424/533, loss: 0.008613109588623047 2023-01-21 12:02:06.407748: step: 428/533, loss: 0.005357981193810701 2023-01-21 12:02:07.579724: step: 432/533, loss: 0.0004409789980854839 2023-01-21 12:02:08.767609: step: 436/533, loss: 5.15937790623866e-05 2023-01-21 12:02:09.911338: step: 440/533, loss: 0.02735147438943386 2023-01-21 12:02:11.057984: step: 444/533, loss: 0.0007070541614666581 2023-01-21 12:02:12.219072: step: 448/533, loss: 0.009868335910141468 2023-01-21 12:02:13.431216: step: 452/533, loss: 0.0013532638549804688 2023-01-21 12:02:14.570207: step: 456/533, loss: 0.00036716461181640625 2023-01-21 12:02:15.734506: step: 460/533, loss: 0.021535206586122513 2023-01-21 12:02:16.909397: step: 464/533, loss: 0.00033512114896439016 2023-01-21 12:02:18.046810: step: 468/533, loss: 0.0006453990936279297 2023-01-21 12:02:19.250017: step: 472/533, loss: 0.03422722965478897 2023-01-21 12:02:20.421640: step: 476/533, loss: 0.00012903213792014867 2023-01-21 12:02:21.575661: step: 480/533, loss: 0.014339828863739967 2023-01-21 12:02:22.748105: step: 484/533, loss: 0.01044159010052681 2023-01-21 12:02:23.909132: step: 488/533, loss: 0.05192422866821289 2023-01-21 12:02:25.085979: step: 492/533, loss: 0.0005170822259970009 2023-01-21 12:02:26.259776: step: 496/533, loss: 0.0045454977080225945 2023-01-21 12:02:27.422160: step: 500/533, loss: 0.00013699532428290695 2023-01-21 12:02:28.594392: step: 504/533, loss: 0.02984147146344185 2023-01-21 12:02:29.780385: step: 508/533, loss: 0.0005741119384765625 2023-01-21 12:02:30.926758: step: 512/533, loss: 0.004507732577621937 2023-01-21 12:02:32.105455: step: 516/533, loss: 0.002103519393131137 2023-01-21 12:02:33.286341: step: 520/533, loss: 2.9277802241267636e-05 2023-01-21 12:02:34.467241: step: 524/533, loss: 0.008983231149613857 2023-01-21 12:02:35.646773: step: 528/533, loss: 0.0468318946659565 2023-01-21 12:02:36.786616: step: 532/533, loss: 0.0013917923206463456 2023-01-21 12:02:37.939013: step: 536/533, loss: 0.0016618728404864669 2023-01-21 12:02:39.093665: step: 540/533, loss: 0.0043538096360862255 2023-01-21 12:02:40.278770: step: 544/533, loss: 0.007143926806747913 2023-01-21 12:02:41.411416: step: 548/533, loss: 0.004426193423569202 2023-01-21 12:02:42.539708: step: 552/533, loss: 0.00013427734666038305 2023-01-21 12:02:43.701005: step: 556/533, loss: 0.00029239655123092234 2023-01-21 12:02:44.858844: step: 560/533, loss: 0.0009088218212127686 2023-01-21 12:02:46.041017: step: 564/533, loss: 0.0022357942070811987 2023-01-21 12:02:47.241157: step: 568/533, loss: 0.0011319637997075915 2023-01-21 12:02:48.451270: step: 572/533, loss: 0.0021406172309070826 2023-01-21 12:02:49.596413: step: 576/533, loss: 0.0006431579240597785 2023-01-21 12:02:50.766076: step: 580/533, loss: 0.00021286011906340718 2023-01-21 12:02:51.935656: step: 584/533, loss: 0.01475515402853489 2023-01-21 12:02:53.060264: step: 588/533, loss: 0.00048542022705078125 2023-01-21 12:02:54.243824: step: 592/533, loss: 0.004576161503791809 2023-01-21 12:02:55.453435: step: 596/533, loss: 0.0001426696835551411 2023-01-21 12:02:56.689035: step: 600/533, loss: 0.014153290539979935 2023-01-21 12:02:57.897970: step: 604/533, loss: 0.0007802963373251259 2023-01-21 12:02:59.054961: step: 608/533, loss: 0.00010232925706077367 2023-01-21 12:03:00.263076: step: 612/533, loss: 0.01934070698916912 2023-01-21 12:03:01.401940: step: 616/533, loss: 0.010976982302963734 2023-01-21 12:03:02.592119: step: 620/533, loss: 0.00015578270540572703 2023-01-21 12:03:03.790119: step: 624/533, loss: 0.0016590118175372481 2023-01-21 12:03:04.956007: step: 628/533, loss: 0.007845116779208183 2023-01-21 12:03:06.151339: step: 632/533, loss: 0.007744312286376953 2023-01-21 12:03:07.326640: step: 636/533, loss: 0.0002020835963776335 2023-01-21 12:03:08.501035: step: 640/533, loss: 0.001143646310083568 2023-01-21 12:03:09.676651: step: 644/533, loss: 0.0006561279296875 2023-01-21 12:03:10.844831: step: 648/533, loss: 0.004530525300651789 2023-01-21 12:03:12.023240: step: 652/533, loss: 0.001343584037385881 2023-01-21 12:03:13.202824: step: 656/533, loss: 0.00165472028311342 2023-01-21 12:03:14.357726: step: 660/533, loss: 0.0007586479769088328 2023-01-21 12:03:15.559548: step: 664/533, loss: 0.01009063795208931 2023-01-21 12:03:16.730301: step: 668/533, loss: 0.0008470535976812243 2023-01-21 12:03:17.894639: step: 672/533, loss: 0.0005537032848224044 2023-01-21 12:03:19.089287: step: 676/533, loss: 0.0009410798666067421 2023-01-21 12:03:20.250889: step: 680/533, loss: 0.30586767196655273 2023-01-21 12:03:21.452276: step: 684/533, loss: 0.0010246277088299394 2023-01-21 12:03:22.615824: step: 688/533, loss: 0.0006807327154092491 2023-01-21 12:03:23.806263: step: 692/533, loss: 0.002361869905143976 2023-01-21 12:03:24.935729: step: 696/533, loss: 0.0003974914725404233 2023-01-21 12:03:26.116781: step: 700/533, loss: 0.002469873521476984 2023-01-21 12:03:27.265016: step: 704/533, loss: 0.003798961639404297 2023-01-21 12:03:28.400077: step: 708/533, loss: 0.0011365890968590975 2023-01-21 12:03:29.542177: step: 712/533, loss: 0.07625456154346466 2023-01-21 12:03:30.681303: step: 716/533, loss: 0.00031499864417128265 2023-01-21 12:03:31.858938: step: 720/533, loss: 0.00072393415030092 2023-01-21 12:03:33.008734: step: 724/533, loss: 0.04470658302307129 2023-01-21 12:03:34.187183: step: 728/533, loss: 0.0015951157547533512 2023-01-21 12:03:35.371227: step: 732/533, loss: 0.009563303552567959 2023-01-21 12:03:36.557414: step: 736/533, loss: 0.003547382541000843 2023-01-21 12:03:37.706987: step: 740/533, loss: 0.0012043475871905684 2023-01-21 12:03:38.915543: step: 744/533, loss: 0.1874251365661621 2023-01-21 12:03:40.116055: step: 748/533, loss: 0.037401676177978516 2023-01-21 12:03:41.298349: step: 752/533, loss: 0.08503856509923935 2023-01-21 12:03:42.426085: step: 756/533, loss: 0.10025987029075623 2023-01-21 12:03:43.608055: step: 760/533, loss: 0.005613732151687145 2023-01-21 12:03:44.845181: step: 764/533, loss: 0.007354640867561102 2023-01-21 12:03:46.027459: step: 768/533, loss: 0.026511192321777344 2023-01-21 12:03:47.251120: step: 772/533, loss: 0.00554161099717021 2023-01-21 12:03:48.405912: step: 776/533, loss: 0.0007149696466512978 2023-01-21 12:03:49.567893: step: 780/533, loss: 0.004193687345832586 2023-01-21 12:03:50.770359: step: 784/533, loss: 0.04234008863568306 2023-01-21 12:03:51.951592: step: 788/533, loss: 0.0004261017020326108 2023-01-21 12:03:53.131611: step: 792/533, loss: 0.0012734413612633944 2023-01-21 12:03:54.301864: step: 796/533, loss: 0.01250085886567831 2023-01-21 12:03:55.476249: step: 800/533, loss: 0.007552337367087603 2023-01-21 12:03:56.659819: step: 804/533, loss: 0.02440948598086834 2023-01-21 12:03:57.847533: step: 808/533, loss: 0.004042148590087891 2023-01-21 12:03:59.003587: step: 812/533, loss: 4.9209596909349784e-05 2023-01-21 12:04:00.234902: step: 816/533, loss: 0.1505233347415924 2023-01-21 12:04:01.382375: step: 820/533, loss: 7.152557373046875e-05 2023-01-21 12:04:02.533113: step: 824/533, loss: 0.004483032505959272 2023-01-21 12:04:03.735963: step: 828/533, loss: 0.0030532032251358032 2023-01-21 12:04:04.886281: step: 832/533, loss: 0.00026721955509856343 2023-01-21 12:04:06.100403: step: 836/533, loss: 0.00019340515427757055 2023-01-21 12:04:07.287280: step: 840/533, loss: 0.048323437571525574 2023-01-21 12:04:08.485952: step: 844/533, loss: 0.004909897223114967 2023-01-21 12:04:09.662799: step: 848/533, loss: 0.09964561462402344 2023-01-21 12:04:10.845872: step: 852/533, loss: 0.0630183219909668 2023-01-21 12:04:12.027220: step: 856/533, loss: 0.20232120156288147 2023-01-21 12:04:13.201171: step: 860/533, loss: 0.0067619322799146175 2023-01-21 12:04:14.383921: step: 864/533, loss: 0.0007730484358035028 2023-01-21 12:04:15.590314: step: 868/533, loss: 0.026348590850830078 2023-01-21 12:04:16.761956: step: 872/533, loss: 0.022862529382109642 2023-01-21 12:04:17.920210: step: 876/533, loss: 0.00386981968767941 2023-01-21 12:04:19.072648: step: 880/533, loss: 0.0009071350214071572 2023-01-21 12:04:20.288413: step: 884/533, loss: 0.042486000806093216 2023-01-21 12:04:21.483412: step: 888/533, loss: 0.11995582282543182 2023-01-21 12:04:22.678788: step: 892/533, loss: 0.00020332336134742945 2023-01-21 12:04:23.873977: step: 896/533, loss: 0.055998802185058594 2023-01-21 12:04:25.039864: step: 900/533, loss: 0.0024772644974291325 2023-01-21 12:04:26.240755: step: 904/533, loss: 0.00737838726490736 2023-01-21 12:04:27.386487: step: 908/533, loss: 0.000621795654296875 2023-01-21 12:04:28.572123: step: 912/533, loss: 0.06251354515552521 2023-01-21 12:04:29.742923: step: 916/533, loss: 0.004185867495834827 2023-01-21 12:04:30.929129: step: 920/533, loss: 0.006374740973114967 2023-01-21 12:04:32.107261: step: 924/533, loss: 0.0002525329473428428 2023-01-21 12:04:33.278210: step: 928/533, loss: 0.02062520943582058 2023-01-21 12:04:34.443849: step: 932/533, loss: 0.002858448075130582 2023-01-21 12:04:35.596699: step: 936/533, loss: 0.005259895231574774 2023-01-21 12:04:36.764182: step: 940/533, loss: 0.0004816532018594444 2023-01-21 12:04:37.945944: step: 944/533, loss: 0.011879158206284046 2023-01-21 12:04:39.122007: step: 948/533, loss: 0.00376129150390625 2023-01-21 12:04:40.322964: step: 952/533, loss: 0.03909454494714737 2023-01-21 12:04:41.471086: step: 956/533, loss: 0.0024597167503088713 2023-01-21 12:04:42.659178: step: 960/533, loss: 0.04390716552734375 2023-01-21 12:04:43.823887: step: 964/533, loss: 0.00010299683344783261 2023-01-21 12:04:44.984138: step: 968/533, loss: 0.0005018234369345009 2023-01-21 12:04:46.168022: step: 972/533, loss: 0.03115081787109375 2023-01-21 12:04:47.317715: step: 976/533, loss: 0.0972156673669815 2023-01-21 12:04:48.493069: step: 980/533, loss: 0.009401989169418812 2023-01-21 12:04:49.669240: step: 984/533, loss: 0.004040527157485485 2023-01-21 12:04:50.865002: step: 988/533, loss: 0.049971677362918854 2023-01-21 12:04:52.025778: step: 992/533, loss: 0.0004322052118368447 2023-01-21 12:04:53.212604: step: 996/533, loss: 0.0016117095947265625 2023-01-21 12:04:54.411648: step: 1000/533, loss: 0.00027065275935456157 2023-01-21 12:04:55.576774: step: 1004/533, loss: 0.4066218435764313 2023-01-21 12:04:56.749698: step: 1008/533, loss: 0.12249297648668289 2023-01-21 12:04:57.939605: step: 1012/533, loss: 0.011709118261933327 2023-01-21 12:04:59.099694: step: 1016/533, loss: 0.007724660448729992 2023-01-21 12:05:00.289896: step: 1020/533, loss: 0.02372436597943306 2023-01-21 12:05:01.440427: step: 1024/533, loss: 1.316070574830519e-05 2023-01-21 12:05:02.616220: step: 1028/533, loss: 0.014858436770737171 2023-01-21 12:05:03.764097: step: 1032/533, loss: 0.0002283096400788054 2023-01-21 12:05:04.931011: step: 1036/533, loss: 0.09309688210487366 2023-01-21 12:05:06.079008: step: 1040/533, loss: 0.02438983879983425 2023-01-21 12:05:07.218992: step: 1044/533, loss: 0.11804670840501785 2023-01-21 12:05:08.405431: step: 1048/533, loss: 0.01589689403772354 2023-01-21 12:05:09.582330: step: 1052/533, loss: 0.011453819461166859 2023-01-21 12:05:10.741027: step: 1056/533, loss: 0.054758742451667786 2023-01-21 12:05:11.880410: step: 1060/533, loss: 0.00033435822115279734 2023-01-21 12:05:13.089126: step: 1064/533, loss: 0.0005811691517010331 2023-01-21 12:05:14.250183: step: 1068/533, loss: 0.015230655670166016 2023-01-21 12:05:15.424978: step: 1072/533, loss: 0.003242635866627097 2023-01-21 12:05:16.608955: step: 1076/533, loss: 0.007020187564194202 2023-01-21 12:05:17.783333: step: 1080/533, loss: 0.00015792845806572586 2023-01-21 12:05:18.958308: step: 1084/533, loss: 0.008802413940429688 2023-01-21 12:05:20.112912: step: 1088/533, loss: 0.018341543152928352 2023-01-21 12:05:21.344098: step: 1092/533, loss: 0.03094339370727539 2023-01-21 12:05:22.464059: step: 1096/533, loss: 0.0026141167618334293 2023-01-21 12:05:23.631149: step: 1100/533, loss: 0.08951663970947266 2023-01-21 12:05:24.863950: step: 1104/533, loss: 0.0002616286219563335 2023-01-21 12:05:26.068515: step: 1108/533, loss: 0.0001354217529296875 2023-01-21 12:05:27.245687: step: 1112/533, loss: 0.006215572357177734 2023-01-21 12:05:28.445969: step: 1116/533, loss: 0.008065510541200638 2023-01-21 12:05:29.632859: step: 1120/533, loss: 0.006081008818000555 2023-01-21 12:05:30.814375: step: 1124/533, loss: 0.0047782897017896175 2023-01-21 12:05:31.952016: step: 1128/533, loss: 0.00404701242223382 2023-01-21 12:05:33.169103: step: 1132/533, loss: 0.004487133119255304 2023-01-21 12:05:34.325569: step: 1136/533, loss: 4.0531158447265625e-06 2023-01-21 12:05:35.539074: step: 1140/533, loss: 0.05892696604132652 2023-01-21 12:05:36.733897: step: 1144/533, loss: 0.00040349960909225047 2023-01-21 12:05:37.897698: step: 1148/533, loss: 0.05106792598962784 2023-01-21 12:05:39.050766: step: 1152/533, loss: 0.001617473317310214 2023-01-21 12:05:40.225797: step: 1156/533, loss: 0.08778515458106995 2023-01-21 12:05:41.392753: step: 1160/533, loss: 0.002426719758659601 2023-01-21 12:05:42.563933: step: 1164/533, loss: 0.0054111480712890625 2023-01-21 12:05:43.794173: step: 1168/533, loss: 0.004352569580078125 2023-01-21 12:05:44.980351: step: 1172/533, loss: 0.016757775098085403 2023-01-21 12:05:46.186908: step: 1176/533, loss: 5.122110366821289 2023-01-21 12:05:47.348176: step: 1180/533, loss: 0.02698221430182457 2023-01-21 12:05:48.527972: step: 1184/533, loss: 0.004998448304831982 2023-01-21 12:05:49.701164: step: 1188/533, loss: 0.0003503799671307206 2023-01-21 12:05:50.928234: step: 1192/533, loss: 0.020367145538330078 2023-01-21 12:05:52.117751: step: 1196/533, loss: 0.2772860527038574 2023-01-21 12:05:53.341192: step: 1200/533, loss: 0.0034521103370934725 2023-01-21 12:05:54.505409: step: 1204/533, loss: 1.487731969973538e-05 2023-01-21 12:05:55.693429: step: 1208/533, loss: 0.00666203536093235 2023-01-21 12:05:56.822580: step: 1212/533, loss: 0.0018039703136309981 2023-01-21 12:05:57.989415: step: 1216/533, loss: 0.0131378173828125 2023-01-21 12:05:59.152035: step: 1220/533, loss: 0.0062576294876635075 2023-01-21 12:06:00.331031: step: 1224/533, loss: 0.0024105070624500513 2023-01-21 12:06:01.502614: step: 1228/533, loss: 0.0026803971268236637 2023-01-21 12:06:02.696904: step: 1232/533, loss: 0.005998420529067516 2023-01-21 12:06:03.869723: step: 1236/533, loss: 1.024240255355835 2023-01-21 12:06:05.044290: step: 1240/533, loss: 0.031406402587890625 2023-01-21 12:06:06.216809: step: 1244/533, loss: 0.025767900049686432 2023-01-21 12:06:07.417089: step: 1248/533, loss: 0.007350135128945112 2023-01-21 12:06:08.587255: step: 1252/533, loss: 0.01758279837667942 2023-01-21 12:06:09.732134: step: 1256/533, loss: 0.007219123654067516 2023-01-21 12:06:10.915020: step: 1260/533, loss: 0.005278205964714289 2023-01-21 12:06:12.090478: step: 1264/533, loss: 0.013664436526596546 2023-01-21 12:06:13.252183: step: 1268/533, loss: 0.19037647545337677 2023-01-21 12:06:14.414594: step: 1272/533, loss: 0.009997367858886719 2023-01-21 12:06:15.581486: step: 1276/533, loss: 0.021406935527920723 2023-01-21 12:06:16.718069: step: 1280/533, loss: 0.07100877910852432 2023-01-21 12:06:17.865474: step: 1284/533, loss: 0.0025356293190270662 2023-01-21 12:06:19.018806: step: 1288/533, loss: 0.08441966027021408 2023-01-21 12:06:20.156305: step: 1292/533, loss: 0.0005296230083331466 2023-01-21 12:06:21.342793: step: 1296/533, loss: 0.007633018773049116 2023-01-21 12:06:22.521934: step: 1300/533, loss: 0.02295827865600586 2023-01-21 12:06:23.759489: step: 1304/533, loss: 0.013332939706742764 2023-01-21 12:06:24.973946: step: 1308/533, loss: 0.033670809119939804 2023-01-21 12:06:26.144200: step: 1312/533, loss: 0.04257016256451607 2023-01-21 12:06:27.327764: step: 1316/533, loss: 0.0009433746454305947 2023-01-21 12:06:28.541058: step: 1320/533, loss: 0.008856773376464844 2023-01-21 12:06:29.710375: step: 1324/533, loss: 0.0036297799088060856 2023-01-21 12:06:30.869112: step: 1328/533, loss: 0.0019344330066815019 2023-01-21 12:06:32.030630: step: 1332/533, loss: 0.011247063055634499 2023-01-21 12:06:33.175735: step: 1336/533, loss: 0.00048666002112440765 2023-01-21 12:06:34.383292: step: 1340/533, loss: 0.003752136370167136 2023-01-21 12:06:35.597721: step: 1344/533, loss: 0.0425909049808979 2023-01-21 12:06:36.774743: step: 1348/533, loss: 0.006059551611542702 2023-01-21 12:06:37.941830: step: 1352/533, loss: 0.005645942874252796 2023-01-21 12:06:39.126613: step: 1356/533, loss: 0.03469143062829971 2023-01-21 12:06:40.325239: step: 1360/533, loss: 0.0069252969697117805 2023-01-21 12:06:41.575528: step: 1364/533, loss: 0.5461196899414062 2023-01-21 12:06:42.743403: step: 1368/533, loss: 0.07063484191894531 2023-01-21 12:06:43.937911: step: 1372/533, loss: 0.0021238327026367188 2023-01-21 12:06:45.097137: step: 1376/533, loss: 0.025496339425444603 2023-01-21 12:06:46.289102: step: 1380/533, loss: 0.2572908401489258 2023-01-21 12:06:47.448910: step: 1384/533, loss: 0.0003055572451557964 2023-01-21 12:06:48.626593: step: 1388/533, loss: 0.025017881765961647 2023-01-21 12:06:49.781422: step: 1392/533, loss: 0.0010271072387695312 2023-01-21 12:06:50.931884: step: 1396/533, loss: 0.0025419234298169613 2023-01-21 12:06:52.129268: step: 1400/533, loss: 0.022492026910185814 2023-01-21 12:06:53.327236: step: 1404/533, loss: 0.03459892421960831 2023-01-21 12:06:54.491616: step: 1408/533, loss: 0.019272994250059128 2023-01-21 12:06:55.627247: step: 1412/533, loss: 0.011844253167510033 2023-01-21 12:06:56.806250: step: 1416/533, loss: 0.02653675153851509 2023-01-21 12:06:57.966003: step: 1420/533, loss: 3.218650817871094e-05 2023-01-21 12:06:59.149777: step: 1424/533, loss: 0.00127410888671875 2023-01-21 12:07:00.342006: step: 1428/533, loss: 0.001157474471256137 2023-01-21 12:07:01.520658: step: 1432/533, loss: 0.03273334354162216 2023-01-21 12:07:02.694775: step: 1436/533, loss: 0.001287269638851285 2023-01-21 12:07:03.899291: step: 1440/533, loss: 0.02900104597210884 2023-01-21 12:07:05.085398: step: 1444/533, loss: 4.7206878662109375e-05 2023-01-21 12:07:06.248787: step: 1448/533, loss: 0.27194613218307495 2023-01-21 12:07:07.416317: step: 1452/533, loss: 0.0010843276977539062 2023-01-21 12:07:08.617565: step: 1456/533, loss: 0.0332002155482769 2023-01-21 12:07:09.859139: step: 1460/533, loss: 0.03449106216430664 2023-01-21 12:07:11.067975: step: 1464/533, loss: 0.0015352248447015882 2023-01-21 12:07:12.250617: step: 1468/533, loss: 0.0028256417717784643 2023-01-21 12:07:13.425272: step: 1472/533, loss: 0.1281047761440277 2023-01-21 12:07:14.607518: step: 1476/533, loss: 0.001175737357698381 2023-01-21 12:07:15.854117: step: 1480/533, loss: 0.006404686253517866 2023-01-21 12:07:17.019043: step: 1484/533, loss: 0.013490390963852406 2023-01-21 12:07:18.176810: step: 1488/533, loss: 0.04128146171569824 2023-01-21 12:07:19.343189: step: 1492/533, loss: 0.002332592150196433 2023-01-21 12:07:20.511191: step: 1496/533, loss: 0.017383957281708717 2023-01-21 12:07:21.651756: step: 1500/533, loss: 0.0044089797884225845 2023-01-21 12:07:22.788346: step: 1504/533, loss: 0.04332747310400009 2023-01-21 12:07:23.927627: step: 1508/533, loss: 0.021468639373779297 2023-01-21 12:07:25.100872: step: 1512/533, loss: 0.00043354035005904734 2023-01-21 12:07:26.288721: step: 1516/533, loss: 0.04013385996222496 2023-01-21 12:07:27.495568: step: 1520/533, loss: 0.012973022647202015 2023-01-21 12:07:28.659011: step: 1524/533, loss: 4.1198727558366954e-05 2023-01-21 12:07:29.809501: step: 1528/533, loss: 0.0010884524090215564 2023-01-21 12:07:31.048084: step: 1532/533, loss: 0.02466287650167942 2023-01-21 12:07:32.226451: step: 1536/533, loss: 0.03653149679303169 2023-01-21 12:07:33.411417: step: 1540/533, loss: 0.02111187018454075 2023-01-21 12:07:34.548706: step: 1544/533, loss: 0.0004723549063783139 2023-01-21 12:07:35.759313: step: 1548/533, loss: 0.0007199287647381425 2023-01-21 12:07:36.921422: step: 1552/533, loss: 0.0566747672855854 2023-01-21 12:07:38.086417: step: 1556/533, loss: 0.001899623777717352 2023-01-21 12:07:39.246036: step: 1560/533, loss: 0.5502438545227051 2023-01-21 12:07:40.392749: step: 1564/533, loss: 0.004647445864975452 2023-01-21 12:07:41.533478: step: 1568/533, loss: 0.00039348602876998484 2023-01-21 12:07:42.725677: step: 1572/533, loss: 0.0006938934093341231 2023-01-21 12:07:43.937017: step: 1576/533, loss: 0.15553809702396393 2023-01-21 12:07:45.093709: step: 1580/533, loss: 0.0007397651206701994 2023-01-21 12:07:46.275488: step: 1584/533, loss: 0.02375955693423748 2023-01-21 12:07:47.450586: step: 1588/533, loss: 0.00013420582399703562 2023-01-21 12:07:48.645698: step: 1592/533, loss: 0.01761808432638645 2023-01-21 12:07:49.830958: step: 1596/533, loss: 0.05400505289435387 2023-01-21 12:07:51.011150: step: 1600/533, loss: 0.005527639761567116 2023-01-21 12:07:52.184027: step: 1604/533, loss: 0.06259031593799591 2023-01-21 12:07:53.389330: step: 1608/533, loss: 0.07448416203260422 2023-01-21 12:07:54.545440: step: 1612/533, loss: 0.10164089500904083 2023-01-21 12:07:55.738078: step: 1616/533, loss: 0.003637647721916437 2023-01-21 12:07:56.977171: step: 1620/533, loss: 0.014238452538847923 2023-01-21 12:07:58.167777: step: 1624/533, loss: 0.06112994998693466 2023-01-21 12:07:59.366617: step: 1628/533, loss: 0.0048576355911791325 2023-01-21 12:08:00.507275: step: 1632/533, loss: 0.004140663426369429 2023-01-21 12:08:01.670826: step: 1636/533, loss: 0.0073407175950706005 2023-01-21 12:08:02.880499: step: 1640/533, loss: 0.004893111996352673 2023-01-21 12:08:04.056813: step: 1644/533, loss: 0.00447921734303236 2023-01-21 12:08:05.236811: step: 1648/533, loss: 0.030223656445741653 2023-01-21 12:08:06.467165: step: 1652/533, loss: 0.06376419216394424 2023-01-21 12:08:07.610478: step: 1656/533, loss: 0.037380170077085495 2023-01-21 12:08:08.765977: step: 1660/533, loss: 0.0037206648848950863 2023-01-21 12:08:09.946910: step: 1664/533, loss: 0.0024383545387536287 2023-01-21 12:08:11.110213: step: 1668/533, loss: 0.040135763585567474 2023-01-21 12:08:12.270229: step: 1672/533, loss: 0.00013532637967728078 2023-01-21 12:08:13.469469: step: 1676/533, loss: 0.0004868507385253906 2023-01-21 12:08:14.625621: step: 1680/533, loss: 4.9591064453125e-05 2023-01-21 12:08:15.798295: step: 1684/533, loss: 2.9373170036706142e-05 2023-01-21 12:08:16.986257: step: 1688/533, loss: 0.00013394355482887477 2023-01-21 12:08:18.154026: step: 1692/533, loss: 4.024505687993951e-05 2023-01-21 12:08:19.299580: step: 1696/533, loss: 0.017134379595518112 2023-01-21 12:08:20.468646: step: 1700/533, loss: 0.006768226623535156 2023-01-21 12:08:21.636548: step: 1704/533, loss: 0.0004404068167787045 2023-01-21 12:08:22.824973: step: 1708/533, loss: 0.008815002627670765 2023-01-21 12:08:23.972713: step: 1712/533, loss: 0.08487816154956818 2023-01-21 12:08:25.104274: step: 1716/533, loss: 0.0005656242137774825 2023-01-21 12:08:26.284061: step: 1720/533, loss: 0.0015456199180334806 2023-01-21 12:08:27.493722: step: 1724/533, loss: 0.018176937475800514 2023-01-21 12:08:28.694739: step: 1728/533, loss: 0.12450714409351349 2023-01-21 12:08:29.874742: step: 1732/533, loss: 0.006127930246293545 2023-01-21 12:08:31.070306: step: 1736/533, loss: 0.014917850494384766 2023-01-21 12:08:32.233803: step: 1740/533, loss: 0.0043621063232421875 2023-01-21 12:08:33.412356: step: 1744/533, loss: 0.004134416580200195 2023-01-21 12:08:34.609677: step: 1748/533, loss: 0.022272301837801933 2023-01-21 12:08:35.785082: step: 1752/533, loss: 0.002297639846801758 2023-01-21 12:08:36.940987: step: 1756/533, loss: 0.0017546177841722965 2023-01-21 12:08:38.096577: step: 1760/533, loss: 0.5152382850646973 2023-01-21 12:08:39.264139: step: 1764/533, loss: 0.0025721550919115543 2023-01-21 12:08:40.444198: step: 1768/533, loss: 0.0020834924653172493 2023-01-21 12:08:41.622900: step: 1772/533, loss: 0.03519783169031143 2023-01-21 12:08:42.858207: step: 1776/533, loss: 0.05718240886926651 2023-01-21 12:08:44.046107: step: 1780/533, loss: 0.029549717903137207 2023-01-21 12:08:45.220873: step: 1784/533, loss: 0.04425792768597603 2023-01-21 12:08:46.407822: step: 1788/533, loss: 0.004256058018654585 2023-01-21 12:08:47.581968: step: 1792/533, loss: 0.0013744353782385588 2023-01-21 12:08:48.730527: step: 1796/533, loss: 0.005530834197998047 2023-01-21 12:08:49.921133: step: 1800/533, loss: 0.010906601324677467 2023-01-21 12:08:51.123708: step: 1804/533, loss: 0.02026844024658203 2023-01-21 12:08:52.270290: step: 1808/533, loss: 0.02513866499066353 2023-01-21 12:08:53.391863: step: 1812/533, loss: 0.0009048461797647178 2023-01-21 12:08:54.572660: step: 1816/533, loss: 0.00675168028101325 2023-01-21 12:08:55.755071: step: 1820/533, loss: 0.0035154344514012337 2023-01-21 12:08:56.955427: step: 1824/533, loss: 0.00260505685582757 2023-01-21 12:08:58.181928: step: 1828/533, loss: 0.041312407702207565 2023-01-21 12:08:59.340026: step: 1832/533, loss: 0.0010979652870446444 2023-01-21 12:09:00.523652: step: 1836/533, loss: 0.008643600158393383 2023-01-21 12:09:01.703211: step: 1840/533, loss: 0.004321003332734108 2023-01-21 12:09:02.893114: step: 1844/533, loss: 0.003247928572818637 2023-01-21 12:09:04.066738: step: 1848/533, loss: 0.0011290073161944747 2023-01-21 12:09:05.277299: step: 1852/533, loss: 0.00388507847674191 2023-01-21 12:09:06.445210: step: 1856/533, loss: 0.00414276123046875 2023-01-21 12:09:07.624226: step: 1860/533, loss: 0.21076726913452148 2023-01-21 12:09:08.810762: step: 1864/533, loss: 0.054128263145685196 2023-01-21 12:09:10.013842: step: 1868/533, loss: 0.00983820017427206 2023-01-21 12:09:11.198854: step: 1872/533, loss: 0.04686005041003227 2023-01-21 12:09:12.340487: step: 1876/533, loss: 0.0007840156904421747 2023-01-21 12:09:13.526933: step: 1880/533, loss: 0.00798187218606472 2023-01-21 12:09:14.704883: step: 1884/533, loss: 0.005003738682717085 2023-01-21 12:09:15.886139: step: 1888/533, loss: 0.00497207697480917 2023-01-21 12:09:17.021926: step: 1892/533, loss: 0.0006148338434286416 2023-01-21 12:09:18.181249: step: 1896/533, loss: 0.03353137895464897 2023-01-21 12:09:19.375738: step: 1900/533, loss: 0.05414552986621857 2023-01-21 12:09:20.524093: step: 1904/533, loss: 0.0008921623229980469 2023-01-21 12:09:21.710586: step: 1908/533, loss: 0.0067070964723825455 2023-01-21 12:09:22.886968: step: 1912/533, loss: 0.014531135559082031 2023-01-21 12:09:24.049558: step: 1916/533, loss: 0.0013049126137048006 2023-01-21 12:09:25.194803: step: 1920/533, loss: 0.00024919508723542094 2023-01-21 12:09:26.377758: step: 1924/533, loss: 0.0003031730593647808 2023-01-21 12:09:27.570239: step: 1928/533, loss: 0.0020389556884765625 2023-01-21 12:09:28.737425: step: 1932/533, loss: 0.05325908586382866 2023-01-21 12:09:29.982145: step: 1936/533, loss: 0.09697379916906357 2023-01-21 12:09:31.160128: step: 1940/533, loss: 0.0003757476806640625 2023-01-21 12:09:32.305696: step: 1944/533, loss: 6.27517729299143e-05 2023-01-21 12:09:33.442939: step: 1948/533, loss: 4.57763690064894e-06 2023-01-21 12:09:34.629395: step: 1952/533, loss: 0.024541093036532402 2023-01-21 12:09:35.841253: step: 1956/533, loss: 0.028744984418153763 2023-01-21 12:09:36.986097: step: 1960/533, loss: 0.000919818936381489 2023-01-21 12:09:38.154390: step: 1964/533, loss: 0.0005827904096804559 2023-01-21 12:09:39.314701: step: 1968/533, loss: 0.02389349788427353 2023-01-21 12:09:40.534186: step: 1972/533, loss: 0.1714106649160385 2023-01-21 12:09:41.669599: step: 1976/533, loss: 0.0001039028211380355 2023-01-21 12:09:42.844513: step: 1980/533, loss: 0.053618624806404114 2023-01-21 12:09:44.042517: step: 1984/533, loss: 0.00476760882884264 2023-01-21 12:09:45.214919: step: 1988/533, loss: 0.0037973406724631786 2023-01-21 12:09:46.377106: step: 1992/533, loss: 0.05294475704431534 2023-01-21 12:09:47.545973: step: 1996/533, loss: 0.00338916783221066 2023-01-21 12:09:48.746628: step: 2000/533, loss: 0.0022109029814600945 2023-01-21 12:09:49.918572: step: 2004/533, loss: 0.045327283442020416 2023-01-21 12:09:51.111261: step: 2008/533, loss: 0.0035783289931714535 2023-01-21 12:09:52.294754: step: 2012/533, loss: 0.012261963449418545 2023-01-21 12:09:53.448402: step: 2016/533, loss: 0.04805422201752663 2023-01-21 12:09:54.628377: step: 2020/533, loss: 0.009329653345048428 2023-01-21 12:09:55.771464: step: 2024/533, loss: 0.0013481139903888106 2023-01-21 12:09:56.937699: step: 2028/533, loss: 0.028441905975341797 2023-01-21 12:09:58.086862: step: 2032/533, loss: 0.003007602645084262 2023-01-21 12:09:59.300717: step: 2036/533, loss: 0.010295892134308815 2023-01-21 12:10:00.482378: step: 2040/533, loss: 0.00782165490090847 2023-01-21 12:10:01.664912: step: 2044/533, loss: 0.00035581589327193797 2023-01-21 12:10:02.851016: step: 2048/533, loss: 0.008481930941343307 2023-01-21 12:10:04.005147: step: 2052/533, loss: 9.479522850597277e-05 2023-01-21 12:10:05.191536: step: 2056/533, loss: 0.02820129320025444 2023-01-21 12:10:06.349667: step: 2060/533, loss: 0.008162785321474075 2023-01-21 12:10:07.533745: step: 2064/533, loss: 0.05465507507324219 2023-01-21 12:10:08.744893: step: 2068/533, loss: 0.0004142761172261089 2023-01-21 12:10:09.920141: step: 2072/533, loss: 0.05378603935241699 2023-01-21 12:10:11.078586: step: 2076/533, loss: 0.0031736374367028475 2023-01-21 12:10:12.281238: step: 2080/533, loss: 0.02724475972354412 2023-01-21 12:10:13.399335: step: 2084/533, loss: 0.001712703611701727 2023-01-21 12:10:14.561442: step: 2088/533, loss: 0.05514903366565704 2023-01-21 12:10:15.775696: step: 2092/533, loss: 0.014711570926010609 2023-01-21 12:10:16.961776: step: 2096/533, loss: 0.007003688719123602 2023-01-21 12:10:18.155161: step: 2100/533, loss: 0.0008621216402389109 2023-01-21 12:10:19.346080: step: 2104/533, loss: 0.0004925727844238281 2023-01-21 12:10:20.481352: step: 2108/533, loss: 0.0013575553894042969 2023-01-21 12:10:21.636693: step: 2112/533, loss: 0.037487030029296875 2023-01-21 12:10:22.825955: step: 2116/533, loss: 0.009822273626923561 2023-01-21 12:10:24.002267: step: 2120/533, loss: 0.009489918127655983 2023-01-21 12:10:25.211564: step: 2124/533, loss: 0.0016937257023528218 2023-01-21 12:10:26.384560: step: 2128/533, loss: 0.01073532085865736 2023-01-21 12:10:27.596803: step: 2132/533, loss: 0.011197661980986595 ================================================== Loss: 0.038 -------------------- Dev: {'event': {'p': 0.6279594137542277, 'r': 0.7416777629826897, 'f1': 0.68009768009768}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Test: {'event': {'p': 0.6633663366336634, 'r': 0.8139386189258312, 'f1': 0.7309790410565605}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Chinese: {'event': {'p': 0.5822784810126582, 'r': 0.8518518518518519, 'f1': 0.6917293233082706}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Korean: {'event': {'p': 0.5961538461538461, 'r': 0.49206349206349204, 'f1': 0.5391304347826087}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Russian: {'event': {'p': 0.5, 'r': 0.5, 'f1': 0.5}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:11:08.361552: step: 4/533, loss: 0.2114221602678299 2023-01-21 12:11:09.512696: step: 8/533, loss: 0.0007483959197998047 2023-01-21 12:11:10.712292: step: 12/533, loss: 0.054314516484737396 2023-01-21 12:11:11.880382: step: 16/533, loss: 0.0029611587524414062 2023-01-21 12:11:13.060126: step: 20/533, loss: 0.008453655056655407 2023-01-21 12:11:14.242063: step: 24/533, loss: 0.012305354699492455 2023-01-21 12:11:15.422420: step: 28/533, loss: 0.0008993148803710938 2023-01-21 12:11:16.552489: step: 32/533, loss: 0.0005504608270712197 2023-01-21 12:11:17.713191: step: 36/533, loss: 0.00023708344087935984 2023-01-21 12:11:18.951486: step: 40/533, loss: 0.4376079738140106 2023-01-21 12:11:20.099967: step: 44/533, loss: 0.003211879637092352 2023-01-21 12:11:21.267419: step: 48/533, loss: 0.0015024185413494706 2023-01-21 12:11:22.419662: step: 52/533, loss: 0.00014915467181708664 2023-01-21 12:11:23.622013: step: 56/533, loss: 0.3636995255947113 2023-01-21 12:11:24.781963: step: 60/533, loss: 0.010753822512924671 2023-01-21 12:11:25.976644: step: 64/533, loss: 0.003022480057552457 2023-01-21 12:11:27.135142: step: 68/533, loss: 0.0028556822799146175 2023-01-21 12:11:28.312283: step: 72/533, loss: 0.018155287951231003 2023-01-21 12:11:29.498039: step: 76/533, loss: 0.0011999130947515368 2023-01-21 12:11:30.702876: step: 80/533, loss: 0.07744219154119492 2023-01-21 12:11:31.869803: step: 84/533, loss: 0.025572966784238815 2023-01-21 12:11:33.027836: step: 88/533, loss: 0.00023365022207144648 2023-01-21 12:11:34.201682: step: 92/533, loss: 0.0009333133930340409 2023-01-21 12:11:35.390997: step: 96/533, loss: 0.024586105719208717 2023-01-21 12:11:36.563392: step: 100/533, loss: 0.0007363319746218622 2023-01-21 12:11:37.760284: step: 104/533, loss: 0.0004058837948832661 2023-01-21 12:11:38.929656: step: 108/533, loss: 0.006507396697998047 2023-01-21 12:11:40.134438: step: 112/533, loss: 3.337860107421875e-06 2023-01-21 12:11:41.319341: step: 116/533, loss: 0.09514904022216797 2023-01-21 12:11:42.438034: step: 120/533, loss: 0.010058355517685413 2023-01-21 12:11:43.627576: step: 124/533, loss: 0.12287445366382599 2023-01-21 12:11:44.814904: step: 128/533, loss: 0.04804229736328125 2023-01-21 12:11:45.925684: step: 132/533, loss: 0.010461139492690563 2023-01-21 12:11:47.130938: step: 136/533, loss: 5.998611595714465e-05 2023-01-21 12:11:48.281786: step: 140/533, loss: 0.031055260449647903 2023-01-21 12:11:49.447872: step: 144/533, loss: 0.0007986068958416581 2023-01-21 12:11:50.594803: step: 148/533, loss: 0.0013193131890147924 2023-01-21 12:11:51.747845: step: 152/533, loss: 0.0019273757934570312 2023-01-21 12:11:52.923082: step: 156/533, loss: 0.053510092198848724 2023-01-21 12:11:54.116539: step: 160/533, loss: 0.00019378661818336695 2023-01-21 12:11:55.268797: step: 164/533, loss: 0.002864933107048273 2023-01-21 12:11:56.461270: step: 168/533, loss: 0.004442405886948109 2023-01-21 12:11:57.631530: step: 172/533, loss: 0.059233762323856354 2023-01-21 12:11:58.800796: step: 176/533, loss: 0.00818567257374525 2023-01-21 12:12:00.036366: step: 180/533, loss: 0.020816516131162643 2023-01-21 12:12:01.191269: step: 184/533, loss: 0.044381141662597656 2023-01-21 12:12:02.338274: step: 188/533, loss: 0.006442356389015913 2023-01-21 12:12:03.522286: step: 192/533, loss: 0.0042479513213038445 2023-01-21 12:12:04.695036: step: 196/533, loss: 0.00030841826810501516 2023-01-21 12:12:05.899517: step: 200/533, loss: 0.001988649368286133 2023-01-21 12:12:07.042857: step: 204/533, loss: 0.0006410598871298134 2023-01-21 12:12:08.150475: step: 208/533, loss: 0.06265778839588165 2023-01-21 12:12:09.321063: step: 212/533, loss: 0.029312897473573685 2023-01-21 12:12:10.484019: step: 216/533, loss: 0.00338916783221066 2023-01-21 12:12:11.639838: step: 220/533, loss: 3.06129441014491e-05 2023-01-21 12:12:12.857728: step: 224/533, loss: 0.006083297543227673 2023-01-21 12:12:14.012836: step: 228/533, loss: 0.004646826069802046 2023-01-21 12:12:15.157646: step: 232/533, loss: 0.006245899014174938 2023-01-21 12:12:16.338532: step: 236/533, loss: 0.016956519335508347 2023-01-21 12:12:17.529566: step: 240/533, loss: 0.008573818020522594 2023-01-21 12:12:18.746202: step: 244/533, loss: 0.00300350203178823 2023-01-21 12:12:19.911790: step: 248/533, loss: 0.00211334228515625 2023-01-21 12:12:21.111303: step: 252/533, loss: 0.0003257751523051411 2023-01-21 12:12:22.283073: step: 256/533, loss: 0.1889963150024414 2023-01-21 12:12:23.494701: step: 260/533, loss: 0.00031490327091887593 2023-01-21 12:12:24.717530: step: 264/533, loss: 9.660721116233617e-05 2023-01-21 12:12:25.869147: step: 268/533, loss: 0.028276730328798294 2023-01-21 12:12:27.016246: step: 272/533, loss: 0.0002155303955078125 2023-01-21 12:12:28.211602: step: 276/533, loss: 0.008719349279999733 2023-01-21 12:12:29.366350: step: 280/533, loss: 0.0025572776794433594 2023-01-21 12:12:30.531649: step: 284/533, loss: 0.025408077985048294 2023-01-21 12:12:31.692952: step: 288/533, loss: 0.0012783050769940019 2023-01-21 12:12:32.884370: step: 292/533, loss: 0.017495252192020416 2023-01-21 12:12:34.053583: step: 296/533, loss: 4.3678286601789296e-05 2023-01-21 12:12:35.267182: step: 300/533, loss: 0.031350038945674896 2023-01-21 12:12:36.454787: step: 304/533, loss: 0.01266541425138712 2023-01-21 12:12:37.651414: step: 308/533, loss: 0.03453044965863228 2023-01-21 12:12:38.803166: step: 312/533, loss: 0.03584041818976402 2023-01-21 12:12:40.021051: step: 316/533, loss: 0.031690120697021484 2023-01-21 12:12:41.218005: step: 320/533, loss: 0.008761787787079811 2023-01-21 12:12:42.352308: step: 324/533, loss: 0.007081317715346813 2023-01-21 12:12:43.530571: step: 328/533, loss: 0.00044803618220612407 2023-01-21 12:12:44.719318: step: 332/533, loss: 0.00107240688521415 2023-01-21 12:12:45.927089: step: 336/533, loss: 0.0007678032270632684 2023-01-21 12:12:47.072861: step: 340/533, loss: 0.07104358822107315 2023-01-21 12:12:48.211203: step: 344/533, loss: 2.307891918462701e-05 2023-01-21 12:12:49.393047: step: 348/533, loss: 0.01395959872752428 2023-01-21 12:12:50.638986: step: 352/533, loss: 0.00013370515080168843 2023-01-21 12:12:51.775261: step: 356/533, loss: 0.0005534171941690147 2023-01-21 12:12:52.971149: step: 360/533, loss: 0.05240488424897194 2023-01-21 12:12:54.094864: step: 364/533, loss: 3.604888843256049e-05 2023-01-21 12:12:55.248109: step: 368/533, loss: 0.0011074065696448088 2023-01-21 12:12:56.441103: step: 372/533, loss: 0.008075714111328125 2023-01-21 12:12:57.578630: step: 376/533, loss: 0.0010364532936364412 2023-01-21 12:12:58.741848: step: 380/533, loss: 0.002929878421127796 2023-01-21 12:12:59.901051: step: 384/533, loss: 1.583099401614163e-05 2023-01-21 12:13:01.097111: step: 388/533, loss: 6.427765038097277e-05 2023-01-21 12:13:02.282149: step: 392/533, loss: 0.00099945068359375 2023-01-21 12:13:03.438417: step: 396/533, loss: 1.831054760259576e-05 2023-01-21 12:13:04.565632: step: 400/533, loss: 5.664825584972277e-05 2023-01-21 12:13:05.699459: step: 404/533, loss: 0.0048836711794137955 2023-01-21 12:13:06.842884: step: 408/533, loss: 0.011579811573028564 2023-01-21 12:13:08.034625: step: 412/533, loss: 0.05727224424481392 2023-01-21 12:13:09.230334: step: 416/533, loss: 0.011657333932816982 2023-01-21 12:13:10.411336: step: 420/533, loss: 0.07051172107458115 2023-01-21 12:13:11.592559: step: 424/533, loss: 0.02228241041302681 2023-01-21 12:13:12.757879: step: 428/533, loss: 0.01579265668988228 2023-01-21 12:13:13.941290: step: 432/533, loss: 0.012528609484434128 2023-01-21 12:13:15.118806: step: 436/533, loss: 0.018306922167539597 2023-01-21 12:13:16.275291: step: 440/533, loss: 0.01898346096277237 2023-01-21 12:13:17.482202: step: 444/533, loss: 0.00022840499877929688 2023-01-21 12:13:18.657128: step: 448/533, loss: 0.014329385943710804 2023-01-21 12:13:19.800416: step: 452/533, loss: 0.019863223657011986 2023-01-21 12:13:21.002902: step: 456/533, loss: 0.005616378504782915 2023-01-21 12:13:22.196092: step: 460/533, loss: -1.0156631105928682e-05 2023-01-21 12:13:23.367909: step: 464/533, loss: 0.0011076927185058594 2023-01-21 12:13:24.574528: step: 468/533, loss: 0.04111747443675995 2023-01-21 12:13:25.733543: step: 472/533, loss: 0.0002289771946379915 2023-01-21 12:13:26.902563: step: 476/533, loss: 0.0017118453979492188 2023-01-21 12:13:28.085441: step: 480/533, loss: 0.00015544891357421875 2023-01-21 12:13:29.297400: step: 484/533, loss: 0.026804352179169655 2023-01-21 12:13:30.494872: step: 488/533, loss: 0.0007459640619345009 2023-01-21 12:13:31.704627: step: 492/533, loss: 0.027368545532226562 2023-01-21 12:13:32.851180: step: 496/533, loss: 1.926422191900201e-05 2023-01-21 12:13:34.009883: step: 500/533, loss: 0.028551483526825905 2023-01-21 12:13:35.181248: step: 504/533, loss: 0.10369911044836044 2023-01-21 12:13:36.349086: step: 508/533, loss: 0.09845905005931854 2023-01-21 12:13:37.518036: step: 512/533, loss: 7.123946852516383e-05 2023-01-21 12:13:38.673525: step: 516/533, loss: 0.051432374864816666 2023-01-21 12:13:39.817860: step: 520/533, loss: 4.7063826059456915e-05 2023-01-21 12:13:41.008610: step: 524/533, loss: 0.007857322692871094 2023-01-21 12:13:42.149814: step: 528/533, loss: 0.05198555067181587 2023-01-21 12:13:43.318998: step: 532/533, loss: 0.001773929689079523 2023-01-21 12:13:44.450984: step: 536/533, loss: 6.031990051269531e-05 2023-01-21 12:13:45.610720: step: 540/533, loss: 0.010657978244125843 2023-01-21 12:13:46.792532: step: 544/533, loss: 0.020488500595092773 2023-01-21 12:13:48.001993: step: 548/533, loss: 0.0020929814781993628 2023-01-21 12:13:49.162073: step: 552/533, loss: 0.023758888244628906 2023-01-21 12:13:50.311400: step: 556/533, loss: 0.0066515919752418995 2023-01-21 12:13:51.509466: step: 560/533, loss: 0.0038924694526940584 2023-01-21 12:13:52.704504: step: 564/533, loss: 0.013571548275649548 2023-01-21 12:13:53.860635: step: 568/533, loss: 3.089904930675402e-05 2023-01-21 12:13:55.033442: step: 572/533, loss: 0.0016660690307617188 2023-01-21 12:13:56.224863: step: 576/533, loss: 0.11463909596204758 2023-01-21 12:13:57.427384: step: 580/533, loss: 0.13136129081249237 2023-01-21 12:13:58.631907: step: 584/533, loss: 0.050052642822265625 2023-01-21 12:13:59.779624: step: 588/533, loss: 0.0009054660913534462 2023-01-21 12:14:00.914119: step: 592/533, loss: 0.0008127689943648875 2023-01-21 12:14:02.083548: step: 596/533, loss: 0.0009230375289916992 2023-01-21 12:14:03.271977: step: 600/533, loss: 0.1051357239484787 2023-01-21 12:14:04.452876: step: 604/533, loss: 0.00024251938157249242 2023-01-21 12:14:05.655275: step: 608/533, loss: 0.041836928576231 2023-01-21 12:14:06.829543: step: 612/533, loss: 0.024132154881954193 2023-01-21 12:14:08.001683: step: 616/533, loss: 0.007435894571244717 2023-01-21 12:14:09.146009: step: 620/533, loss: 0.005153465550392866 2023-01-21 12:14:10.290471: step: 624/533, loss: 0.0006205082172527909 2023-01-21 12:14:11.408988: step: 628/533, loss: 0.001353359199129045 2023-01-21 12:14:12.596195: step: 632/533, loss: 0.009461212903261185 2023-01-21 12:14:13.798699: step: 636/533, loss: 0.018471335992217064 2023-01-21 12:14:14.965778: step: 640/533, loss: 0.0191816333681345 2023-01-21 12:14:16.154267: step: 644/533, loss: 0.032549094408750534 2023-01-21 12:14:17.344822: step: 648/533, loss: 0.026899147778749466 2023-01-21 12:14:18.501383: step: 652/533, loss: 0.033650971949100494 2023-01-21 12:14:19.658673: step: 656/533, loss: 0.017066096886992455 2023-01-21 12:14:20.822614: step: 660/533, loss: 0.0011143684387207031 2023-01-21 12:14:21.987306: step: 664/533, loss: 0.01226501539349556 2023-01-21 12:14:23.136202: step: 668/533, loss: 0.00025739672128111124 2023-01-21 12:14:24.322347: step: 672/533, loss: 0.0008543014409951866 2023-01-21 12:14:25.475039: step: 676/533, loss: 0.0025899887550622225 2023-01-21 12:14:26.656147: step: 680/533, loss: 0.08083362877368927 2023-01-21 12:14:27.831037: step: 684/533, loss: 0.010121727362275124 2023-01-21 12:14:29.069817: step: 688/533, loss: 0.00011482238915050402 2023-01-21 12:14:30.228215: step: 692/533, loss: 0.0028248669113963842 2023-01-21 12:14:31.402153: step: 696/533, loss: 0.04645667225122452 2023-01-21 12:14:32.574178: step: 700/533, loss: 0.022718334570527077 2023-01-21 12:14:33.708991: step: 704/533, loss: 0.002287614392116666 2023-01-21 12:14:34.897170: step: 708/533, loss: 0.031145477667450905 2023-01-21 12:14:36.110356: step: 712/533, loss: 0.008761787787079811 2023-01-21 12:14:37.308421: step: 716/533, loss: 0.00011377334885764867 2023-01-21 12:14:38.562554: step: 720/533, loss: 0.0005422592512331903 2023-01-21 12:14:39.729164: step: 724/533, loss: 4.682540748035535e-05 2023-01-21 12:14:40.887323: step: 728/533, loss: 0.00013380050950217992 2023-01-21 12:14:42.023980: step: 732/533, loss: 0.04857831075787544 2023-01-21 12:14:43.208371: step: 736/533, loss: 0.0002054929791484028 2023-01-21 12:14:44.383884: step: 740/533, loss: 0.02416095696389675 2023-01-21 12:14:45.541128: step: 744/533, loss: 0.001089763711206615 2023-01-21 12:14:46.713504: step: 748/533, loss: 0.005036068148910999 2023-01-21 12:14:47.875088: step: 752/533, loss: 9.722709364723414e-05 2023-01-21 12:14:49.052111: step: 756/533, loss: 0.002954196883365512 2023-01-21 12:14:50.196337: step: 760/533, loss: 0.0008681297767907381 2023-01-21 12:14:51.344579: step: 764/533, loss: 0.0043032169342041016 2023-01-21 12:14:52.598826: step: 768/533, loss: 0.04064664617180824 2023-01-21 12:14:53.778661: step: 772/533, loss: 0.08902730792760849 2023-01-21 12:14:54.977051: step: 776/533, loss: 0.0171966552734375 2023-01-21 12:14:56.141569: step: 780/533, loss: 0.009490680880844593 2023-01-21 12:14:57.309937: step: 784/533, loss: 0.012571334838867188 2023-01-21 12:14:58.485839: step: 788/533, loss: 0.029239654541015625 2023-01-21 12:14:59.636688: step: 792/533, loss: 0.006228351499885321 2023-01-21 12:15:00.801010: step: 796/533, loss: 0.0014430045848712325 2023-01-21 12:15:01.939747: step: 800/533, loss: 0.04431457817554474 2023-01-21 12:15:03.110578: step: 804/533, loss: 0.000459051167126745 2023-01-21 12:15:04.292818: step: 808/533, loss: 0.009197045117616653 2023-01-21 12:15:05.448396: step: 812/533, loss: 0.03645050898194313 2023-01-21 12:15:06.646822: step: 816/533, loss: 0.003945732023566961 2023-01-21 12:15:07.792648: step: 820/533, loss: 0.03385419771075249 2023-01-21 12:15:08.939341: step: 824/533, loss: 1.2302399227337446e-05 2023-01-21 12:15:10.064427: step: 828/533, loss: 0.006670760922133923 2023-01-21 12:15:11.255363: step: 832/533, loss: 0.0007033348665572703 2023-01-21 12:15:12.395325: step: 836/533, loss: 0.039798736572265625 2023-01-21 12:15:13.598000: step: 840/533, loss: 0.004310226533561945 2023-01-21 12:15:14.766782: step: 844/533, loss: 0.0002499580441508442 2023-01-21 12:15:15.943871: step: 848/533, loss: 0.0006010055658407509 2023-01-21 12:15:17.135427: step: 852/533, loss: 0.010681104846298695 2023-01-21 12:15:18.296370: step: 856/533, loss: 0.011629200540482998 2023-01-21 12:15:19.503180: step: 860/533, loss: 0.0016603947151452303 2023-01-21 12:15:20.694736: step: 864/533, loss: 0.00121650705114007 2023-01-21 12:15:21.889070: step: 868/533, loss: 0.0017477036453783512 2023-01-21 12:15:23.071835: step: 872/533, loss: 0.006692838855087757 2023-01-21 12:15:24.265734: step: 876/533, loss: 0.006122398190200329 2023-01-21 12:15:25.486649: step: 880/533, loss: 0.0006157875177450478 2023-01-21 12:15:26.638777: step: 884/533, loss: 0.020538795739412308 2023-01-21 12:15:27.819221: step: 888/533, loss: 0.02670116536319256 2023-01-21 12:15:29.049635: step: 892/533, loss: 0.01110620517283678 2023-01-21 12:15:30.206478: step: 896/533, loss: 0.2819734215736389 2023-01-21 12:15:31.382271: step: 900/533, loss: 0.0001598358212504536 2023-01-21 12:15:32.560098: step: 904/533, loss: 0.0011553765507414937 2023-01-21 12:15:33.800683: step: 908/533, loss: 0.006258392706513405 2023-01-21 12:15:34.982133: step: 912/533, loss: 0.0023877143394201994 2023-01-21 12:15:36.152322: step: 916/533, loss: 0.0025774003006517887 2023-01-21 12:15:37.295881: step: 920/533, loss: 2.2602082026423886e-05 2023-01-21 12:15:38.458379: step: 924/533, loss: 0.005808830261230469 2023-01-21 12:15:39.656640: step: 928/533, loss: 0.018293190747499466 2023-01-21 12:15:40.812198: step: 932/533, loss: 0.03265724331140518 2023-01-21 12:15:41.970640: step: 936/533, loss: 0.020064163953065872 2023-01-21 12:15:43.111735: step: 940/533, loss: 0.003011989640071988 2023-01-21 12:15:44.277685: step: 944/533, loss: 0.0037488937377929688 2023-01-21 12:15:45.442338: step: 948/533, loss: 0.033568382263183594 2023-01-21 12:15:46.633861: step: 952/533, loss: 1.316070574830519e-05 2023-01-21 12:15:47.780980: step: 956/533, loss: 0.012466716580092907 2023-01-21 12:15:49.010812: step: 960/533, loss: 0.009816360659897327 2023-01-21 12:15:50.164945: step: 964/533, loss: 8.535385131835938e-05 2023-01-21 12:15:51.370775: step: 968/533, loss: 2.0027162008773303e-06 2023-01-21 12:15:52.563987: step: 972/533, loss: 0.0032600401900708675 2023-01-21 12:15:53.712562: step: 976/533, loss: 0.02121276967227459 2023-01-21 12:15:54.877278: step: 980/533, loss: 0.0603116974234581 2023-01-21 12:15:56.030017: step: 984/533, loss: 0.0004171371692791581 2023-01-21 12:15:57.182779: step: 988/533, loss: 5.722047262679553e-07 2023-01-21 12:15:58.352651: step: 992/533, loss: 0.0035126686561852694 2023-01-21 12:15:59.500760: step: 996/533, loss: 0.0013684272998943925 2023-01-21 12:16:00.664402: step: 1000/533, loss: 0.0013780593872070312 2023-01-21 12:16:01.805083: step: 1004/533, loss: 1.4209747860149946e-05 2023-01-21 12:16:02.949537: step: 1008/533, loss: 0.006527316756546497 2023-01-21 12:16:04.127041: step: 1012/533, loss: 0.02345886267721653 2023-01-21 12:16:05.283837: step: 1016/533, loss: 0.02124919928610325 2023-01-21 12:16:06.451043: step: 1020/533, loss: 0.010608386248350143 2023-01-21 12:16:07.724281: step: 1024/533, loss: 0.16931067407131195 2023-01-21 12:16:08.932666: step: 1028/533, loss: 0.0012843131553381681 2023-01-21 12:16:10.144397: step: 1032/533, loss: 0.0008766175014898181 2023-01-21 12:16:11.328124: step: 1036/533, loss: 0.007630157750099897 2023-01-21 12:16:12.460160: step: 1040/533, loss: 0.00045099260751158 2023-01-21 12:16:13.582927: step: 1044/533, loss: 0.0005715370643883944 2023-01-21 12:16:14.732464: step: 1048/533, loss: 0.024192430078983307 2023-01-21 12:16:15.871760: step: 1052/533, loss: 0.000790500664152205 2023-01-21 12:16:17.059369: step: 1056/533, loss: 0.0063911438919603825 2023-01-21 12:16:18.227901: step: 1060/533, loss: 0.00025405886117368937 2023-01-21 12:16:19.395697: step: 1064/533, loss: 0.06196136772632599 2023-01-21 12:16:20.592946: step: 1068/533, loss: 0.000476837158203125 2023-01-21 12:16:21.788511: step: 1072/533, loss: 0.013082504272460938 2023-01-21 12:16:22.997993: step: 1076/533, loss: 0.0014033317565917969 2023-01-21 12:16:24.172286: step: 1080/533, loss: 0.000850677490234375 2023-01-21 12:16:25.311604: step: 1084/533, loss: 0.0023277283180505037 2023-01-21 12:16:26.499082: step: 1088/533, loss: 0.004790782928466797 2023-01-21 12:16:27.677502: step: 1092/533, loss: 0.0006177902687340975 2023-01-21 12:16:28.901662: step: 1096/533, loss: 0.08150079846382141 2023-01-21 12:16:30.095837: step: 1100/533, loss: 0.006209183018654585 2023-01-21 12:16:31.246850: step: 1104/533, loss: 0.005821609403938055 2023-01-21 12:16:32.408442: step: 1108/533, loss: 0.00037744047585874796 2023-01-21 12:16:33.601521: step: 1112/533, loss: 0.021162033081054688 2023-01-21 12:16:34.745731: step: 1116/533, loss: 0.03668374940752983 2023-01-21 12:16:35.918962: step: 1120/533, loss: 0.018657110631465912 2023-01-21 12:16:37.066745: step: 1124/533, loss: 0.05146846920251846 2023-01-21 12:16:38.298119: step: 1128/533, loss: 0.0019295692909508944 2023-01-21 12:16:39.498996: step: 1132/533, loss: 0.026489736512303352 2023-01-21 12:16:40.709605: step: 1136/533, loss: 0.018149565905332565 2023-01-21 12:16:41.882820: step: 1140/533, loss: 0.007332992274314165 2023-01-21 12:16:43.097806: step: 1144/533, loss: 0.09552960097789764 2023-01-21 12:16:44.258604: step: 1148/533, loss: 0.002246379852294922 2023-01-21 12:16:45.520995: step: 1152/533, loss: 0.021952057257294655 2023-01-21 12:16:46.677762: step: 1156/533, loss: 0.0018913268577307463 2023-01-21 12:16:47.849258: step: 1160/533, loss: 0.0004182815901003778 2023-01-21 12:16:49.043245: step: 1164/533, loss: 0.006509971804916859 2023-01-21 12:16:50.216608: step: 1168/533, loss: 0.04781994968652725 2023-01-21 12:16:51.421469: step: 1172/533, loss: 0.00017130374908447266 2023-01-21 12:16:52.655525: step: 1176/533, loss: 0.18276672065258026 2023-01-21 12:16:53.815445: step: 1180/533, loss: 0.03608705848455429 2023-01-21 12:16:55.061657: step: 1184/533, loss: 0.0004178047238383442 2023-01-21 12:16:56.247936: step: 1188/533, loss: 0.001820898149162531 2023-01-21 12:16:57.397135: step: 1192/533, loss: 0.00043807030306197703 2023-01-21 12:16:58.585718: step: 1196/533, loss: 0.0058074952103197575 2023-01-21 12:16:59.802262: step: 1200/533, loss: 6.723403657815652e-06 2023-01-21 12:17:00.967991: step: 1204/533, loss: 0.004756308160722256 2023-01-21 12:17:02.149779: step: 1208/533, loss: 0.034429457038640976 2023-01-21 12:17:03.348234: step: 1212/533, loss: 0.0032428978011012077 2023-01-21 12:17:04.539938: step: 1216/533, loss: 0.014985943213105202 2023-01-21 12:17:05.703259: step: 1220/533, loss: 0.0006476640701293945 2023-01-21 12:17:06.872181: step: 1224/533, loss: 0.00029773713322356343 2023-01-21 12:17:08.050240: step: 1228/533, loss: 0.030493546277284622 2023-01-21 12:17:09.270328: step: 1232/533, loss: 0.007457160856574774 2023-01-21 12:17:10.451625: step: 1236/533, loss: 0.012656117789447308 2023-01-21 12:17:11.638966: step: 1240/533, loss: 0.014214038848876953 2023-01-21 12:17:12.858805: step: 1244/533, loss: 0.016066743060946465 2023-01-21 12:17:14.020380: step: 1248/533, loss: 0.007273959927260876 2023-01-21 12:17:15.180133: step: 1252/533, loss: 9.613037400413305e-05 2023-01-21 12:17:16.362407: step: 1256/533, loss: 4.9591064453125e-05 2023-01-21 12:17:17.538899: step: 1260/533, loss: 2.841949390131049e-05 2023-01-21 12:17:18.727852: step: 1264/533, loss: 0.15430641174316406 2023-01-21 12:17:19.876842: step: 1268/533, loss: 0.0002168655482819304 2023-01-21 12:17:21.095783: step: 1272/533, loss: 0.00010948181443382055 2023-01-21 12:17:22.275783: step: 1276/533, loss: 2.4700166250113398e-05 2023-01-21 12:17:23.478798: step: 1280/533, loss: 0.20417319238185883 2023-01-21 12:17:24.666955: step: 1284/533, loss: 0.013249779120087624 2023-01-21 12:17:25.838649: step: 1288/533, loss: 0.007322215940803289 2023-01-21 12:17:27.034404: step: 1292/533, loss: 0.00030851367046125233 2023-01-21 12:17:28.224994: step: 1296/533, loss: 0.027322864159941673 2023-01-21 12:17:29.446440: step: 1300/533, loss: 0.0010341644519940019 2023-01-21 12:17:30.617330: step: 1304/533, loss: 0.00280342111364007 2023-01-21 12:17:31.785705: step: 1308/533, loss: 0.0010660648113116622 2023-01-21 12:17:32.957585: step: 1312/533, loss: 0.00029616354731842875 2023-01-21 12:17:34.158232: step: 1316/533, loss: 0.02420072630047798 2023-01-21 12:17:35.361435: step: 1320/533, loss: 0.22925511002540588 2023-01-21 12:17:36.566019: step: 1324/533, loss: 0.003690481185913086 2023-01-21 12:17:37.838259: step: 1328/533, loss: 0.004816913977265358 2023-01-21 12:17:38.995587: step: 1332/533, loss: 0.0010329962242394686 2023-01-21 12:17:40.174827: step: 1336/533, loss: 0.02021799236536026 2023-01-21 12:17:41.391163: step: 1340/533, loss: 0.01763634756207466 2023-01-21 12:17:42.510899: step: 1344/533, loss: 7.019043550826609e-05 2023-01-21 12:17:43.646376: step: 1348/533, loss: 0.0015884400345385075 2023-01-21 12:17:44.836066: step: 1352/533, loss: 0.031409457325935364 2023-01-21 12:17:46.039808: step: 1356/533, loss: 0.23956279456615448 2023-01-21 12:17:47.204122: step: 1360/533, loss: 0.007388305850327015 2023-01-21 12:17:48.328159: step: 1364/533, loss: 0.01363153476268053 2023-01-21 12:17:49.479772: step: 1368/533, loss: 0.06320323795080185 2023-01-21 12:17:50.650272: step: 1372/533, loss: 0.011235857382416725 2023-01-21 12:17:51.823086: step: 1376/533, loss: 0.0015499591827392578 2023-01-21 12:17:52.997149: step: 1380/533, loss: 0.039724256843328476 2023-01-21 12:17:54.189147: step: 1384/533, loss: 0.021163273602724075 2023-01-21 12:17:55.378434: step: 1388/533, loss: 0.0013381005264818668 2023-01-21 12:17:56.532949: step: 1392/533, loss: 0.05538537725806236 2023-01-21 12:17:57.701792: step: 1396/533, loss: 0.00010185241262661293 2023-01-21 12:17:58.858245: step: 1400/533, loss: 0.00037603380042128265 2023-01-21 12:18:00.038204: step: 1404/533, loss: -4.482269105210435e-06 2023-01-21 12:18:01.186712: step: 1408/533, loss: 0.0010151863098144531 2023-01-21 12:18:02.345954: step: 1412/533, loss: 0.0003846168692689389 2023-01-21 12:18:03.510454: step: 1416/533, loss: 0.00011072755296481773 2023-01-21 12:18:04.680695: step: 1420/533, loss: 0.0009510040399618447 2023-01-21 12:18:05.899156: step: 1424/533, loss: 0.01688366010785103 2023-01-21 12:18:07.045063: step: 1428/533, loss: 0.03987493738532066 2023-01-21 12:18:08.196321: step: 1432/533, loss: 0.009581565856933594 2023-01-21 12:18:09.416561: step: 1436/533, loss: 0.013789081946015358 2023-01-21 12:18:10.608234: step: 1440/533, loss: 0.021574975922703743 2023-01-21 12:18:11.768102: step: 1444/533, loss: 6.942749314475805e-05 2023-01-21 12:18:12.931627: step: 1448/533, loss: 0.0017833709716796875 2023-01-21 12:18:14.111624: step: 1452/533, loss: 7.25746140233241e-05 2023-01-21 12:18:15.296040: step: 1456/533, loss: 3.62396240234375e-05 2023-01-21 12:18:16.485128: step: 1460/533, loss: 0.012083339504897594 2023-01-21 12:18:17.651549: step: 1464/533, loss: 0.014619636349380016 2023-01-21 12:18:18.850899: step: 1468/533, loss: 0.0023204803001135588 2023-01-21 12:18:20.008313: step: 1472/533, loss: 0.005540370941162109 2023-01-21 12:18:21.212570: step: 1476/533, loss: 0.040253639221191406 2023-01-21 12:18:22.350326: step: 1480/533, loss: 0.08708486706018448 2023-01-21 12:18:23.544119: step: 1484/533, loss: 0.02092895470559597 2023-01-21 12:18:24.705198: step: 1488/533, loss: 8.125304884742945e-05 2023-01-21 12:18:25.890812: step: 1492/533, loss: 0.00382823939435184 2023-01-21 12:18:27.077002: step: 1496/533, loss: 0.03235873952507973 2023-01-21 12:18:28.291456: step: 1500/533, loss: 0.03460102155804634 2023-01-21 12:18:29.447233: step: 1504/533, loss: 0.01672344282269478 2023-01-21 12:18:30.624623: step: 1508/533, loss: 0.0007333874818868935 2023-01-21 12:18:31.816784: step: 1512/533, loss: 0.14904403686523438 2023-01-21 12:18:33.002698: step: 1516/533, loss: 0.0004961013910360634 2023-01-21 12:18:34.186718: step: 1520/533, loss: 0.06422577053308487 2023-01-21 12:18:35.356926: step: 1524/533, loss: 0.003205871442332864 2023-01-21 12:18:36.603181: step: 1528/533, loss: 0.0173205379396677 2023-01-21 12:18:37.768514: step: 1532/533, loss: 0.01211938913911581 2023-01-21 12:18:38.968283: step: 1536/533, loss: 0.0001068115234375 2023-01-21 12:18:40.151136: step: 1540/533, loss: 0.004449176602065563 2023-01-21 12:18:41.334562: step: 1544/533, loss: 0.011485290713608265 2023-01-21 12:18:42.473855: step: 1548/533, loss: 0.007732868194580078 2023-01-21 12:18:43.640485: step: 1552/533, loss: 5.795508695882745e-05 2023-01-21 12:18:44.799238: step: 1556/533, loss: 0.0039581297896802425 2023-01-21 12:18:45.961281: step: 1560/533, loss: 0.0801682397723198 2023-01-21 12:18:47.142871: step: 1564/533, loss: 0.0008066177251748741 2023-01-21 12:18:48.286732: step: 1568/533, loss: 0.00022730827913619578 2023-01-21 12:18:49.443738: step: 1572/533, loss: 0.009067249484360218 2023-01-21 12:18:50.620150: step: 1576/533, loss: 0.014584064483642578 2023-01-21 12:18:51.810161: step: 1580/533, loss: 0.21547985076904297 2023-01-21 12:18:52.965145: step: 1584/533, loss: 0.007395362947136164 2023-01-21 12:18:54.109272: step: 1588/533, loss: 0.003168773604556918 2023-01-21 12:18:55.248806: step: 1592/533, loss: 0.005688357166945934 2023-01-21 12:18:56.421918: step: 1596/533, loss: 0.0075018880888819695 2023-01-21 12:18:57.604307: step: 1600/533, loss: 0.002256202744320035 2023-01-21 12:18:58.764189: step: 1604/533, loss: 0.005157852079719305 2023-01-21 12:18:59.936524: step: 1608/533, loss: 0.03178606182336807 2023-01-21 12:19:01.075748: step: 1612/533, loss: 0.06984749436378479 2023-01-21 12:19:02.244600: step: 1616/533, loss: 0.006915188394486904 2023-01-21 12:19:03.432259: step: 1620/533, loss: 0.006665086839348078 2023-01-21 12:19:04.570399: step: 1624/533, loss: 0.004380130674690008 2023-01-21 12:19:05.768180: step: 1628/533, loss: 0.0045452117919921875 2023-01-21 12:19:06.912427: step: 1632/533, loss: 8.869171142578125e-05 2023-01-21 12:19:08.065780: step: 1636/533, loss: 0.020156098529696465 2023-01-21 12:19:09.264503: step: 1640/533, loss: 7.214546349132434e-05 2023-01-21 12:19:10.391827: step: 1644/533, loss: 0.0008726119995117188 2023-01-21 12:19:11.552964: step: 1648/533, loss: 0.0004152298206463456 2023-01-21 12:19:12.696670: step: 1652/533, loss: 0.0005063056596554816 2023-01-21 12:19:13.862157: step: 1656/533, loss: 0.0020912170875817537 2023-01-21 12:19:15.035025: step: 1660/533, loss: 0.005460834596306086 2023-01-21 12:19:16.258157: step: 1664/533, loss: 0.000476837158203125 2023-01-21 12:19:17.395867: step: 1668/533, loss: 0.006083107087761164 2023-01-21 12:19:18.597328: step: 1672/533, loss: 0.0004906654357910156 2023-01-21 12:19:19.777548: step: 1676/533, loss: 0.054097943007946014 2023-01-21 12:19:20.952007: step: 1680/533, loss: 0.0013564825057983398 2023-01-21 12:19:22.101513: step: 1684/533, loss: 0.033176470547914505 2023-01-21 12:19:23.231903: step: 1688/533, loss: 0.0036859512329101562 2023-01-21 12:19:24.381655: step: 1692/533, loss: 0.0007610321044921875 2023-01-21 12:19:25.579566: step: 1696/533, loss: 0.14615498483181 2023-01-21 12:19:26.777391: step: 1700/533, loss: 0.007613563910126686 2023-01-21 12:19:27.921757: step: 1704/533, loss: 0.00026841164799407125 2023-01-21 12:19:29.072101: step: 1708/533, loss: 0.002239227294921875 2023-01-21 12:19:30.280507: step: 1712/533, loss: 0.055242352187633514 2023-01-21 12:19:31.502190: step: 1716/533, loss: 0.02281207963824272 2023-01-21 12:19:32.690636: step: 1720/533, loss: 3.857910633087158e-05 2023-01-21 12:19:33.866535: step: 1724/533, loss: 0.004992103669792414 2023-01-21 12:19:35.042614: step: 1728/533, loss: 5.826949927723035e-05 2023-01-21 12:19:36.190776: step: 1732/533, loss: 0.0025222108233720064 2023-01-21 12:19:37.326455: step: 1736/533, loss: 0.0007436752784997225 2023-01-21 12:19:38.490905: step: 1740/533, loss: 0.001551818917505443 2023-01-21 12:19:39.686516: step: 1744/533, loss: 0.0027424811851233244 2023-01-21 12:19:40.850394: step: 1748/533, loss: 0.01200180035084486 2023-01-21 12:19:42.007783: step: 1752/533, loss: 0.00654869107529521 2023-01-21 12:19:43.195534: step: 1756/533, loss: 0.12989655137062073 2023-01-21 12:19:44.336366: step: 1760/533, loss: 0.021578121930360794 2023-01-21 12:19:45.550258: step: 1764/533, loss: 0.005482363980263472 2023-01-21 12:19:46.748382: step: 1768/533, loss: 0.14501523971557617 2023-01-21 12:19:47.924879: step: 1772/533, loss: 0.006252766121178865 2023-01-21 12:19:49.090728: step: 1776/533, loss: 0.0041370391845703125 2023-01-21 12:19:50.262345: step: 1780/533, loss: 0.0011947632301598787 2023-01-21 12:19:51.423872: step: 1784/533, loss: 0.001958179520443082 2023-01-21 12:19:52.574686: step: 1788/533, loss: 0.0005060196272097528 2023-01-21 12:19:53.766368: step: 1792/533, loss: 0.027555085718631744 2023-01-21 12:19:54.932917: step: 1796/533, loss: 0.018183518201112747 2023-01-21 12:19:56.121460: step: 1800/533, loss: 0.004184532444924116 2023-01-21 12:19:57.278794: step: 1804/533, loss: 0.025298310443758965 2023-01-21 12:19:58.436015: step: 1808/533, loss: 0.01170425396412611 2023-01-21 12:19:59.580744: step: 1812/533, loss: 0.016721343621611595 2023-01-21 12:20:00.751815: step: 1816/533, loss: 0.0005083084106445312 2023-01-21 12:20:01.958250: step: 1820/533, loss: 0.031094741076231003 2023-01-21 12:20:03.112941: step: 1824/533, loss: 0.004650688264518976 2023-01-21 12:20:04.293413: step: 1828/533, loss: 0.0014319419860839844 2023-01-21 12:20:05.460958: step: 1832/533, loss: 0.0025077343452721834 2023-01-21 12:20:06.624161: step: 1836/533, loss: 0.0005243301275186241 2023-01-21 12:20:07.807211: step: 1840/533, loss: 0.0027610778342932463 2023-01-21 12:20:08.991171: step: 1844/533, loss: 0.00160560617223382 2023-01-21 12:20:10.167379: step: 1848/533, loss: 0.0021846771705895662 2023-01-21 12:20:11.328289: step: 1852/533, loss: 0.002509212587028742 2023-01-21 12:20:12.485344: step: 1856/533, loss: 0.0032723904587328434 2023-01-21 12:20:13.691429: step: 1860/533, loss: 0.00010719299461925402 2023-01-21 12:20:14.868666: step: 1864/533, loss: 0.001210975693538785 2023-01-21 12:20:16.001202: step: 1868/533, loss: 0.0691111609339714 2023-01-21 12:20:17.178977: step: 1872/533, loss: 0.0005935669178143144 2023-01-21 12:20:18.329327: step: 1876/533, loss: 0.011607361026108265 2023-01-21 12:20:19.524618: step: 1880/533, loss: 0.004014777950942516 2023-01-21 12:20:20.715873: step: 1884/533, loss: 0.0020259858574718237 2023-01-21 12:20:21.952457: step: 1888/533, loss: 3.33309180859942e-05 2023-01-21 12:20:23.123888: step: 1892/533, loss: 0.0012325287098065019 2023-01-21 12:20:24.306998: step: 1896/533, loss: 0.0015484809409826994 2023-01-21 12:20:25.463181: step: 1900/533, loss: 0.0038679123390465975 2023-01-21 12:20:26.624827: step: 1904/533, loss: 0.004544353578239679 2023-01-21 12:20:27.753394: step: 1908/533, loss: 0.0004002094210591167 2023-01-21 12:20:28.931934: step: 1912/533, loss: 0.013720226474106312 2023-01-21 12:20:30.127048: step: 1916/533, loss: 4.836684703826904 2023-01-21 12:20:31.332641: step: 1920/533, loss: 0.009340668097138405 2023-01-21 12:20:32.472622: step: 1924/533, loss: 0.0009104252094402909 2023-01-21 12:20:33.659873: step: 1928/533, loss: 0.00782699603587389 2023-01-21 12:20:34.836208: step: 1932/533, loss: 0.016136646270751953 2023-01-21 12:20:36.046800: step: 1936/533, loss: 0.01700305938720703 2023-01-21 12:20:37.257006: step: 1940/533, loss: 0.019713401794433594 2023-01-21 12:20:38.370648: step: 1944/533, loss: 0.010913277044892311 2023-01-21 12:20:39.538332: step: 1948/533, loss: 0.01380233746021986 2023-01-21 12:20:40.744004: step: 1952/533, loss: 0.015938568860292435 2023-01-21 12:20:41.925847: step: 1956/533, loss: 0.002467680023983121 2023-01-21 12:20:43.097299: step: 1960/533, loss: 0.00034103397047147155 2023-01-21 12:20:44.264603: step: 1964/533, loss: 0.0011853218311443925 2023-01-21 12:20:45.401084: step: 1968/533, loss: 4.148483185417717e-06 2023-01-21 12:20:46.536692: step: 1972/533, loss: 0.0076265339739620686 2023-01-21 12:20:47.678096: step: 1976/533, loss: 0.0030502318404614925 2023-01-21 12:20:48.912344: step: 1980/533, loss: 0.007844544015824795 2023-01-21 12:20:50.059325: step: 1984/533, loss: 0.007668781094253063 2023-01-21 12:20:51.224816: step: 1988/533, loss: 0.017430590465664864 2023-01-21 12:20:52.371264: step: 1992/533, loss: 0.0008695602300576866 2023-01-21 12:20:53.557207: step: 1996/533, loss: 0.0018198012840002775 2023-01-21 12:20:54.720812: step: 2000/533, loss: 0.031736090779304504 2023-01-21 12:20:55.860583: step: 2004/533, loss: 0.022118758410215378 2023-01-21 12:20:57.029974: step: 2008/533, loss: 0.0017561912536621094 2023-01-21 12:20:58.185902: step: 2012/533, loss: 0.0020606995094567537 2023-01-21 12:20:59.330132: step: 2016/533, loss: 0.0823429599404335 2023-01-21 12:21:00.474779: step: 2020/533, loss: 0.00012249946303199977 2023-01-21 12:21:01.666192: step: 2024/533, loss: 0.0686667189002037 2023-01-21 12:21:02.886020: step: 2028/533, loss: 0.00014419556828215718 2023-01-21 12:21:04.056173: step: 2032/533, loss: 0.005441809073090553 2023-01-21 12:21:05.213030: step: 2036/533, loss: 0.0034411430824548006 2023-01-21 12:21:06.383906: step: 2040/533, loss: 0.0027456283569335938 2023-01-21 12:21:07.567098: step: 2044/533, loss: 0.008579921908676624 2023-01-21 12:21:08.751857: step: 2048/533, loss: 0.0001771926908986643 2023-01-21 12:21:09.995009: step: 2052/533, loss: 0.11936378479003906 2023-01-21 12:21:11.170952: step: 2056/533, loss: 0.022103404626250267 2023-01-21 12:21:12.304213: step: 2060/533, loss: 0.04020385816693306 2023-01-21 12:21:13.464826: step: 2064/533, loss: 0.016656970605254173 2023-01-21 12:21:14.640188: step: 2068/533, loss: 0.03689517825841904 2023-01-21 12:21:15.828885: step: 2072/533, loss: 0.7788367867469788 2023-01-21 12:21:17.067185: step: 2076/533, loss: 0.010155201889574528 2023-01-21 12:21:18.255614: step: 2080/533, loss: 0.011760711669921875 2023-01-21 12:21:19.421205: step: 2084/533, loss: 0.011496353894472122 2023-01-21 12:21:20.624538: step: 2088/533, loss: 0.021622180938720703 2023-01-21 12:21:21.773522: step: 2092/533, loss: 0.0004516601620707661 2023-01-21 12:21:22.911861: step: 2096/533, loss: 0.0009643555385991931 2023-01-21 12:21:24.090839: step: 2100/533, loss: 0.5990684032440186 2023-01-21 12:21:25.282001: step: 2104/533, loss: 0.0037865638732910156 2023-01-21 12:21:26.434993: step: 2108/533, loss: 0.01347818411886692 2023-01-21 12:21:27.622296: step: 2112/533, loss: 0.0003016471746377647 2023-01-21 12:21:28.784925: step: 2116/533, loss: 0.1178341880440712 2023-01-21 12:21:29.955660: step: 2120/533, loss: 0.07783952355384827 2023-01-21 12:21:31.127208: step: 2124/533, loss: 0.001119422959163785 2023-01-21 12:21:32.292365: step: 2128/533, loss: 0.059322454035282135 2023-01-21 12:21:33.422878: step: 2132/533, loss: 0.005815315525978804 ================================================== Loss: 0.032 -------------------- Dev: {'event': {'p': 0.5891238670694864, 'r': 0.7789613848202397, 'f1': 0.6708715596330275}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Test: {'event': {'p': 0.6351626016260162, 'r': 0.7992327365728901, 'f1': 0.707814269535674}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Chinese: {'event': {'p': 0.6, 'r': 0.8888888888888888, 'f1': 0.7164179104477612}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Korean: {'event': {'p': 0.55, 'r': 0.5238095238095238, 'f1': 0.5365853658536585}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Russian: {'event': {'p': 0.3953488372093023, 'r': 0.4722222222222222, 'f1': 0.43037974683544306}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:22:14.670691: step: 4/533, loss: 0.0010752677917480469 2023-01-21 12:22:15.866163: step: 8/533, loss: 0.0005086899036541581 2023-01-21 12:22:17.032740: step: 12/533, loss: 0.05015525966882706 2023-01-21 12:22:18.240740: step: 16/533, loss: 0.0031757354736328125 2023-01-21 12:22:19.444509: step: 20/533, loss: 0.0002834320184774697 2023-01-21 12:22:20.640067: step: 24/533, loss: 0.009250640869140625 2023-01-21 12:22:21.823734: step: 28/533, loss: 0.00024132727412506938 2023-01-21 12:22:22.991687: step: 32/533, loss: 0.00344772357493639 2023-01-21 12:22:24.158539: step: 36/533, loss: 0.010338879190385342 2023-01-21 12:22:25.371244: step: 40/533, loss: 0.032149411737918854 2023-01-21 12:22:26.531036: step: 44/533, loss: 0.03542356938123703 2023-01-21 12:22:27.688895: step: 48/533, loss: 0.0007952690357342362 2023-01-21 12:22:28.872915: step: 52/533, loss: 0.0009147644159384072 2023-01-21 12:22:30.102893: step: 56/533, loss: 0.11987175792455673 2023-01-21 12:22:31.236527: step: 60/533, loss: 0.00032634736271575093 2023-01-21 12:22:32.393974: step: 64/533, loss: 0.0013103962410241365 2023-01-21 12:22:33.553447: step: 68/533, loss: 0.0482419990003109 2023-01-21 12:22:34.676060: step: 72/533, loss: 0.06710374355316162 2023-01-21 12:22:35.844255: step: 76/533, loss: 0.018765831366181374 2023-01-21 12:22:37.017734: step: 80/533, loss: 0.02208137698471546 2023-01-21 12:22:38.191151: step: 84/533, loss: 0.012534905225038528 2023-01-21 12:22:39.363631: step: 88/533, loss: 0.006365632638335228 2023-01-21 12:22:40.545627: step: 92/533, loss: 0.010386228561401367 2023-01-21 12:22:41.692521: step: 96/533, loss: 0.0027452469803392887 2023-01-21 12:22:42.856496: step: 100/533, loss: 0.08927460014820099 2023-01-21 12:22:44.040061: step: 104/533, loss: 0.015231896191835403 2023-01-21 12:22:45.222656: step: 108/533, loss: 0.011481285095214844 2023-01-21 12:22:46.398950: step: 112/533, loss: 0.0051631927490234375 2023-01-21 12:22:47.594166: step: 116/533, loss: 0.00141057965811342 2023-01-21 12:22:48.752671: step: 120/533, loss: 4.482269287109375e-05 2023-01-21 12:22:49.940797: step: 124/533, loss: 0.0001451492280466482 2023-01-21 12:22:51.123361: step: 128/533, loss: 0.0004528045828919858 2023-01-21 12:22:52.308103: step: 132/533, loss: 0.0026713372208178043 2023-01-21 12:22:53.475078: step: 136/533, loss: 0.006187820341438055 2023-01-21 12:22:54.672879: step: 140/533, loss: 9.72747802734375e-05 2023-01-21 12:22:55.830154: step: 144/533, loss: 0.03494882956147194 2023-01-21 12:22:57.004312: step: 148/533, loss: 0.010295105166733265 2023-01-21 12:22:58.161449: step: 152/533, loss: 0.00013155936903785914 2023-01-21 12:22:59.327475: step: 156/533, loss: 0.01892986334860325 2023-01-21 12:23:00.542930: step: 160/533, loss: 0.00103759765625 2023-01-21 12:23:01.720784: step: 164/533, loss: 0.003035259200260043 2023-01-21 12:23:02.886726: step: 168/533, loss: 0.005446529481559992 2023-01-21 12:23:04.063701: step: 172/533, loss: 0.0014709472889080644 2023-01-21 12:23:05.270175: step: 176/533, loss: 0.007541466038674116 2023-01-21 12:23:06.425796: step: 180/533, loss: 0.0003601074276957661 2023-01-21 12:23:07.571441: step: 184/533, loss: 0.0014446259010583162 2023-01-21 12:23:08.730240: step: 188/533, loss: 0.002903175540268421 2023-01-21 12:23:09.911686: step: 192/533, loss: 0.006575584877282381 2023-01-21 12:23:11.103458: step: 196/533, loss: 0.0003159523184876889 2023-01-21 12:23:12.281263: step: 200/533, loss: 0.0037467004731297493 2023-01-21 12:23:13.424527: step: 204/533, loss: 0.025455093011260033 2023-01-21 12:23:14.571639: step: 208/533, loss: 0.019953157752752304 2023-01-21 12:23:15.730201: step: 212/533, loss: 0.003736400743946433 2023-01-21 12:23:16.929942: step: 216/533, loss: 0.016729259863495827 2023-01-21 12:23:18.061590: step: 220/533, loss: 8.20159948489163e-06 2023-01-21 12:23:19.212319: step: 224/533, loss: 0.04910917207598686 2023-01-21 12:23:20.427949: step: 228/533, loss: 0.012179184705018997 2023-01-21 12:23:21.618094: step: 232/533, loss: 0.0005213737604208291 2023-01-21 12:23:22.778656: step: 236/533, loss: 0.01016860082745552 2023-01-21 12:23:23.942512: step: 240/533, loss: 0.0001675605890341103 2023-01-21 12:23:25.102678: step: 244/533, loss: 0.0002992630179505795 2023-01-21 12:23:26.280449: step: 248/533, loss: 0.011136901564896107 2023-01-21 12:23:27.444202: step: 252/533, loss: 1.9073486612342094e-07 2023-01-21 12:23:28.671049: step: 256/533, loss: 0.009944343939423561 2023-01-21 12:23:29.838485: step: 260/533, loss: 0.005520582664757967 2023-01-21 12:23:30.983652: step: 264/533, loss: 0.029515648260712624 2023-01-21 12:23:32.133949: step: 268/533, loss: 0.00013809204392600805 2023-01-21 12:23:33.316432: step: 272/533, loss: 0.007865333929657936 2023-01-21 12:23:34.509231: step: 276/533, loss: 0.0011631011730059981 2023-01-21 12:23:35.682663: step: 280/533, loss: 0.029093172401189804 2023-01-21 12:23:36.864181: step: 284/533, loss: 0.3226785659790039 2023-01-21 12:23:38.027804: step: 288/533, loss: 0.0004228592151775956 2023-01-21 12:23:39.250676: step: 292/533, loss: 0.019213486462831497 2023-01-21 12:23:40.448890: step: 296/533, loss: 0.014779663644731045 2023-01-21 12:23:41.627236: step: 300/533, loss: 0.008392905816435814 2023-01-21 12:23:42.811160: step: 304/533, loss: 0.015485906973481178 2023-01-21 12:23:43.995138: step: 308/533, loss: 0.00043864251347258687 2023-01-21 12:23:45.150240: step: 312/533, loss: 0.008238411508500576 2023-01-21 12:23:46.316252: step: 316/533, loss: 0.002128553343936801 2023-01-21 12:23:47.486595: step: 320/533, loss: 0.006801700685173273 2023-01-21 12:23:48.705846: step: 324/533, loss: 0.003962373826652765 2023-01-21 12:23:49.886411: step: 328/533, loss: 0.07114367187023163 2023-01-21 12:23:51.060467: step: 332/533, loss: 0.0028912543784826994 2023-01-21 12:23:52.237351: step: 336/533, loss: 0.000156402587890625 2023-01-21 12:23:53.407077: step: 340/533, loss: 0.05251359939575195 2023-01-21 12:23:54.601539: step: 344/533, loss: 0.0012472153175622225 2023-01-21 12:23:55.765623: step: 348/533, loss: 0.0014495849609375 2023-01-21 12:23:56.917685: step: 352/533, loss: 0.006665611173957586 2023-01-21 12:23:58.235884: step: 356/533, loss: 0.13494834303855896 2023-01-21 12:23:59.405385: step: 360/533, loss: 0.0005622864118777215 2023-01-21 12:24:00.562342: step: 364/533, loss: 0.05162405967712402 2023-01-21 12:24:01.706203: step: 368/533, loss: 0.01754598692059517 2023-01-21 12:24:02.901792: step: 372/533, loss: 0.003567266510799527 2023-01-21 12:24:04.102300: step: 376/533, loss: 2.250671423098538e-05 2023-01-21 12:24:05.322390: step: 380/533, loss: 0.0237839687615633 2023-01-21 12:24:06.522608: step: 384/533, loss: 0.009996986947953701 2023-01-21 12:24:07.665248: step: 388/533, loss: 0.0007330895168706775 2023-01-21 12:24:08.825735: step: 392/533, loss: 0.00145807268563658 2023-01-21 12:24:09.957381: step: 396/533, loss: 0.006991052534431219 2023-01-21 12:24:11.136288: step: 400/533, loss: 0.01845083199441433 2023-01-21 12:24:12.296493: step: 404/533, loss: 0.0005084990989416838 2023-01-21 12:24:13.445792: step: 408/533, loss: 0.054672908037900925 2023-01-21 12:24:14.600511: step: 412/533, loss: 0.001952362130396068 2023-01-21 12:24:15.770250: step: 416/533, loss: 0.00040774344233796 2023-01-21 12:24:16.938141: step: 420/533, loss: 0.004485798068344593 2023-01-21 12:24:18.119392: step: 424/533, loss: 0.0005653381813317537 2023-01-21 12:24:19.265035: step: 428/533, loss: 0.01516647357493639 2023-01-21 12:24:20.411837: step: 432/533, loss: 0.17142286896705627 2023-01-21 12:24:21.573699: step: 436/533, loss: 0.010178184136748314 2023-01-21 12:24:22.800239: step: 440/533, loss: 0.002794647356495261 2023-01-21 12:24:23.969384: step: 444/533, loss: 0.0011768341064453125 2023-01-21 12:24:25.156259: step: 448/533, loss: 6.28471389063634e-05 2023-01-21 12:24:26.310515: step: 452/533, loss: 0.0034820795990526676 2023-01-21 12:24:27.490275: step: 456/533, loss: 0.0008982181898318231 2023-01-21 12:24:28.657950: step: 460/533, loss: 0.00012836456880904734 2023-01-21 12:24:29.890758: step: 464/533, loss: 0.0581602118909359 2023-01-21 12:24:31.037852: step: 468/533, loss: 0.0005725860828533769 2023-01-21 12:24:32.202018: step: 472/533, loss: 0.0020710944663733244 2023-01-21 12:24:33.365585: step: 476/533, loss: 0.0005970954662188888 2023-01-21 12:24:34.511851: step: 480/533, loss: 0.0041342019103467464 2023-01-21 12:24:35.680254: step: 484/533, loss: 0.008154869079589844 2023-01-21 12:24:36.838195: step: 488/533, loss: 0.015286731533706188 2023-01-21 12:24:37.996344: step: 492/533, loss: 0.05025453865528107 2023-01-21 12:24:39.149436: step: 496/533, loss: 0.005192804615944624 2023-01-21 12:24:40.348561: step: 500/533, loss: 0.11076297610998154 2023-01-21 12:24:41.503326: step: 504/533, loss: 0.0039459229446947575 2023-01-21 12:24:42.714682: step: 508/533, loss: 0.02257680892944336 2023-01-21 12:24:43.912735: step: 512/533, loss: 0.07266369462013245 2023-01-21 12:24:45.079381: step: 516/533, loss: 0.02223796956241131 2023-01-21 12:24:46.262012: step: 520/533, loss: 0.014933967962861061 2023-01-21 12:24:47.488811: step: 524/533, loss: 0.0014959335094317794 2023-01-21 12:24:48.662865: step: 528/533, loss: 0.015108847990632057 2023-01-21 12:24:49.872198: step: 532/533, loss: 0.0011769295670092106 2023-01-21 12:24:51.039437: step: 536/533, loss: 0.00025663376436568797 2023-01-21 12:24:52.156734: step: 540/533, loss: 0.04293036460876465 2023-01-21 12:24:53.332416: step: 544/533, loss: 0.005969238467514515 2023-01-21 12:24:54.471784: step: 548/533, loss: 0.0011371612781658769 2023-01-21 12:24:55.607261: step: 552/533, loss: 0.0007307052728720009 2023-01-21 12:24:56.804292: step: 556/533, loss: 0.0010875702137127519 2023-01-21 12:24:57.933674: step: 560/533, loss: 1.0681153071345761e-05 2023-01-21 12:24:59.124323: step: 564/533, loss: 0.000659942626953125 2023-01-21 12:25:00.281440: step: 568/533, loss: 0.0009937286376953125 2023-01-21 12:25:01.453276: step: 572/533, loss: 0.10395746678113937 2023-01-21 12:25:02.604573: step: 576/533, loss: 0.015338420867919922 2023-01-21 12:25:03.758733: step: 580/533, loss: 0.0017696857685223222 2023-01-21 12:25:04.913344: step: 584/533, loss: 0.02059359662234783 2023-01-21 12:25:06.085614: step: 588/533, loss: 0.018854428082704544 2023-01-21 12:25:07.211582: step: 592/533, loss: 3.6716461181640625e-05 2023-01-21 12:25:08.439304: step: 596/533, loss: 0.003392982529476285 2023-01-21 12:25:09.629933: step: 600/533, loss: 9.851455979514867e-05 2023-01-21 12:25:10.819400: step: 604/533, loss: 0.026817036792635918 2023-01-21 12:25:12.013988: step: 608/533, loss: 0.028690434992313385 2023-01-21 12:25:13.203131: step: 612/533, loss: 0.01290206890553236 2023-01-21 12:25:14.370453: step: 616/533, loss: 0.004898834507912397 2023-01-21 12:25:15.532921: step: 620/533, loss: 0.012697410769760609 2023-01-21 12:25:16.699260: step: 624/533, loss: 0.03265371173620224 2023-01-21 12:25:17.854035: step: 628/533, loss: 0.24329720437526703 2023-01-21 12:25:19.036045: step: 632/533, loss: 0.007614231202751398 2023-01-21 12:25:20.173318: step: 636/533, loss: 0.005146980285644531 2023-01-21 12:25:21.339334: step: 640/533, loss: 0.00010375976853538305 2023-01-21 12:25:22.488940: step: 644/533, loss: 0.00281867990270257 2023-01-21 12:25:23.683418: step: 648/533, loss: 0.00011215210543014109 2023-01-21 12:25:24.835111: step: 652/533, loss: 0.02332477457821369 2023-01-21 12:25:26.046502: step: 656/533, loss: 0.008139610290527344 2023-01-21 12:25:27.264214: step: 660/533, loss: 0.04134407266974449 2023-01-21 12:25:28.478022: step: 664/533, loss: 0.0021888732444494963 2023-01-21 12:25:29.681497: step: 668/533, loss: 0.012498951517045498 2023-01-21 12:25:30.894386: step: 672/533, loss: 0.002067661378532648 2023-01-21 12:25:32.057896: step: 676/533, loss: 0.003961086273193359 2023-01-21 12:25:33.238537: step: 680/533, loss: 0.02997622638940811 2023-01-21 12:25:34.412457: step: 684/533, loss: -1.010894811770413e-05 2023-01-21 12:25:35.591795: step: 688/533, loss: 0.004114627838134766 2023-01-21 12:25:36.749792: step: 692/533, loss: 0.006018257234245539 2023-01-21 12:25:37.952045: step: 696/533, loss: 0.0013607025612145662 2023-01-21 12:25:39.133960: step: 700/533, loss: 0.0024507520720362663 2023-01-21 12:25:40.281146: step: 704/533, loss: 0.03011307679116726 2023-01-21 12:25:41.495356: step: 708/533, loss: 0.06468620151281357 2023-01-21 12:25:42.644234: step: 712/533, loss: 0.02297673374414444 2023-01-21 12:25:43.796605: step: 716/533, loss: 0.002057647565379739 2023-01-21 12:25:44.953630: step: 720/533, loss: 0.0001374244602629915 2023-01-21 12:25:46.143689: step: 724/533, loss: 0.0007652282947674394 2023-01-21 12:25:47.305398: step: 728/533, loss: 0.0020050047896802425 2023-01-21 12:25:48.491703: step: 732/533, loss: 0.08466528356075287 2023-01-21 12:25:49.662398: step: 736/533, loss: 0.0009654999012127519 2023-01-21 12:25:50.821763: step: 740/533, loss: 7.085800461936742e-05 2023-01-21 12:25:51.962809: step: 744/533, loss: 0.0028558732010424137 2023-01-21 12:25:53.113578: step: 748/533, loss: 0.0003326416190247983 2023-01-21 12:25:54.287125: step: 752/533, loss: 0.09139051288366318 2023-01-21 12:25:55.461273: step: 756/533, loss: 0.0011870384914800525 2023-01-21 12:25:56.670403: step: 760/533, loss: 0.0959661528468132 2023-01-21 12:25:57.835974: step: 764/533, loss: 0.0029108047019690275 2023-01-21 12:25:58.998575: step: 768/533, loss: 0.0015790463658049703 2023-01-21 12:26:00.208433: step: 772/533, loss: 0.012923812493681908 2023-01-21 12:26:01.358087: step: 776/533, loss: 0.022732067853212357 2023-01-21 12:26:02.526945: step: 780/533, loss: 0.0006853103986941278 2023-01-21 12:26:03.666110: step: 784/533, loss: 0.01432724017649889 2023-01-21 12:26:04.853056: step: 788/533, loss: 0.0016889573307707906 2023-01-21 12:26:06.083086: step: 792/533, loss: 0.00040435791015625 2023-01-21 12:26:07.244026: step: 796/533, loss: 0.010360385291278362 2023-01-21 12:26:08.388234: step: 800/533, loss: 4.091262962901965e-05 2023-01-21 12:26:09.617497: step: 804/533, loss: 0.005774307530373335 2023-01-21 12:26:10.806541: step: 808/533, loss: 0.005314636044204235 2023-01-21 12:26:12.052337: step: 812/533, loss: 0.022045612335205078 2023-01-21 12:26:13.237792: step: 816/533, loss: 0.0003689766163006425 2023-01-21 12:26:14.411719: step: 820/533, loss: 9.107589721679688e-05 2023-01-21 12:26:15.597903: step: 824/533, loss: 2.536773718020413e-05 2023-01-21 12:26:16.760583: step: 828/533, loss: 0.0548795685172081 2023-01-21 12:26:17.910035: step: 832/533, loss: 0.0016007423400878906 2023-01-21 12:26:19.079062: step: 836/533, loss: 0.005694914143532515 2023-01-21 12:26:20.313654: step: 840/533, loss: 0.04848938062787056 2023-01-21 12:26:21.505922: step: 844/533, loss: 0.00029430389986373484 2023-01-21 12:26:22.675524: step: 848/533, loss: 0.006320238579064608 2023-01-21 12:26:23.825354: step: 852/533, loss: 0.0019893646240234375 2023-01-21 12:26:25.044801: step: 856/533, loss: 0.059297751635313034 2023-01-21 12:26:26.180950: step: 860/533, loss: 0.0003581047058105469 2023-01-21 12:26:27.344101: step: 864/533, loss: 0.008666038513183594 2023-01-21 12:26:28.563849: step: 868/533, loss: 0.0025396347045898438 2023-01-21 12:26:29.735824: step: 872/533, loss: 0.000274658203125 2023-01-21 12:26:30.882876: step: 876/533, loss: 0.0016363144386559725 2023-01-21 12:26:32.063408: step: 880/533, loss: 0.03193187713623047 2023-01-21 12:26:33.226981: step: 884/533, loss: 0.1317039579153061 2023-01-21 12:26:34.383780: step: 888/533, loss: 0.05788488686084747 2023-01-21 12:26:35.585273: step: 892/533, loss: 0.019421959295868874 2023-01-21 12:26:36.779952: step: 896/533, loss: 0.0003930091916117817 2023-01-21 12:26:37.935904: step: 900/533, loss: 0.0020294189453125 2023-01-21 12:26:39.108745: step: 904/533, loss: 0.00029382703360170126 2023-01-21 12:26:40.309054: step: 908/533, loss: 0.026293182745575905 2023-01-21 12:26:41.467935: step: 912/533, loss: 0.007247066590934992 2023-01-21 12:26:42.678113: step: 916/533, loss: 0.028051234781742096 2023-01-21 12:26:43.848800: step: 920/533, loss: 0.00021848679170943797 2023-01-21 12:26:45.019657: step: 924/533, loss: 0.0006963253254070878 2023-01-21 12:26:46.211322: step: 928/533, loss: 0.002893638564273715 2023-01-21 12:26:47.366051: step: 932/533, loss: 0.0013071061111986637 2023-01-21 12:26:48.553139: step: 936/533, loss: 0.013422345742583275 2023-01-21 12:26:49.724270: step: 940/533, loss: 0.04858889803290367 2023-01-21 12:26:50.900051: step: 944/533, loss: 0.0016326905461028218 2023-01-21 12:26:52.042045: step: 948/533, loss: 3.24249276673072e-06 2023-01-21 12:26:53.296736: step: 952/533, loss: 0.0061667440459132195 2023-01-21 12:26:54.432633: step: 956/533, loss: 0.0009001732105389237 2023-01-21 12:26:55.596483: step: 960/533, loss: 0.014184570871293545 2023-01-21 12:26:56.775946: step: 964/533, loss: 0.001047372817993164 2023-01-21 12:26:57.942765: step: 968/533, loss: 0.01500549353659153 2023-01-21 12:26:59.106693: step: 972/533, loss: 0.005068588070571423 2023-01-21 12:27:00.297461: step: 976/533, loss: 0.11187916249036789 2023-01-21 12:27:01.488298: step: 980/533, loss: 0.00046386721078306437 2023-01-21 12:27:02.675250: step: 984/533, loss: 0.013736248016357422 2023-01-21 12:27:03.820843: step: 988/533, loss: 0.023029232397675514 2023-01-21 12:27:04.989880: step: 992/533, loss: 0.0005396843189373612 2023-01-21 12:27:06.138421: step: 996/533, loss: 0.0003070831298828125 2023-01-21 12:27:07.316442: step: 1000/533, loss: 0.014395141042768955 2023-01-21 12:27:08.522296: step: 1004/533, loss: 0.03348350524902344 2023-01-21 12:27:09.686022: step: 1008/533, loss: 3.32355521095451e-05 2023-01-21 12:27:10.864397: step: 1012/533, loss: 0.022552872076630592 2023-01-21 12:27:12.007743: step: 1016/533, loss: 0.0120690343901515 2023-01-21 12:27:13.149444: step: 1020/533, loss: 0.0007764816400595009 2023-01-21 12:27:14.361643: step: 1024/533, loss: 0.008112048730254173 2023-01-21 12:27:15.518073: step: 1028/533, loss: 0.009698200039565563 2023-01-21 12:27:16.682028: step: 1032/533, loss: 0.0020278929732739925 2023-01-21 12:27:17.826284: step: 1036/533, loss: 3.185272362316027e-05 2023-01-21 12:27:19.006875: step: 1040/533, loss: 0.00225658412091434 2023-01-21 12:27:20.225415: step: 1044/533, loss: 0.0024166107177734375 2023-01-21 12:27:21.392861: step: 1048/533, loss: 0.005865669809281826 2023-01-21 12:27:22.518372: step: 1052/533, loss: 0.00010905266390182078 2023-01-21 12:27:23.694967: step: 1056/533, loss: 0.0002785682736430317 2023-01-21 12:27:24.846434: step: 1060/533, loss: 0.005165576934814453 2023-01-21 12:27:26.045640: step: 1064/533, loss: 0.0017800331115722656 2023-01-21 12:27:27.212905: step: 1068/533, loss: 2.7179718017578125e-05 2023-01-21 12:27:28.402558: step: 1072/533, loss: 6.513595872092992e-05 2023-01-21 12:27:29.571424: step: 1076/533, loss: 0.0008892774349078536 2023-01-21 12:27:30.766619: step: 1080/533, loss: 0.013199138455092907 2023-01-21 12:27:31.964285: step: 1084/533, loss: 0.0036388395819813013 2023-01-21 12:27:33.107310: step: 1088/533, loss: 0.00034961701021529734 2023-01-21 12:27:34.269462: step: 1092/533, loss: 0.00227012624964118 2023-01-21 12:27:35.456511: step: 1096/533, loss: 0.004424476530402899 2023-01-21 12:27:36.706395: step: 1100/533, loss: 0.004406643100082874 2023-01-21 12:27:37.872933: step: 1104/533, loss: 0.009217835031449795 2023-01-21 12:27:39.055016: step: 1108/533, loss: 0.008301163092255592 2023-01-21 12:27:40.253341: step: 1112/533, loss: 2.9277802241267636e-05 2023-01-21 12:27:41.452543: step: 1116/533, loss: 0.020472336560487747 2023-01-21 12:27:42.632219: step: 1120/533, loss: 0.06597833335399628 2023-01-21 12:27:43.812309: step: 1124/533, loss: 0.05019645765423775 2023-01-21 12:27:45.006662: step: 1128/533, loss: 0.0011815071338787675 2023-01-21 12:27:46.179671: step: 1132/533, loss: 0.0002745628298725933 2023-01-21 12:27:47.330505: step: 1136/533, loss: 3.662109520519152e-05 2023-01-21 12:27:48.470972: step: 1140/533, loss: 0.000997686292976141 2023-01-21 12:27:49.681394: step: 1144/533, loss: 0.02737903781235218 2023-01-21 12:27:50.936349: step: 1148/533, loss: 1.640319896978326e-05 2023-01-21 12:27:52.147096: step: 1152/533, loss: 0.000651931797619909 2023-01-21 12:27:53.283620: step: 1156/533, loss: 0.000788784003816545 2023-01-21 12:27:54.529470: step: 1160/533, loss: 0.005009269807487726 2023-01-21 12:27:55.714653: step: 1164/533, loss: 0.010225200094282627 2023-01-21 12:27:56.954843: step: 1168/533, loss: 0.001161861466243863 2023-01-21 12:27:58.103359: step: 1172/533, loss: 0.00019083023653365672 2023-01-21 12:27:59.305442: step: 1176/533, loss: 0.03815264627337456 2023-01-21 12:28:00.523252: step: 1180/533, loss: 0.001218584249727428 2023-01-21 12:28:01.654218: step: 1184/533, loss: 0.00033731458825059235 2023-01-21 12:28:02.818016: step: 1188/533, loss: 0.0032871246803551912 2023-01-21 12:28:03.991894: step: 1192/533, loss: 0.002836036728695035 2023-01-21 12:28:05.164802: step: 1196/533, loss: 0.011802864260971546 2023-01-21 12:28:06.322359: step: 1200/533, loss: 0.0024247169494628906 2023-01-21 12:28:07.491700: step: 1204/533, loss: 0.02355041541159153 2023-01-21 12:28:08.662913: step: 1208/533, loss: 0.005244255065917969 2023-01-21 12:28:09.854851: step: 1212/533, loss: 0.04604225233197212 2023-01-21 12:28:11.014646: step: 1216/533, loss: 0.005480453372001648 2023-01-21 12:28:12.161258: step: 1220/533, loss: 0.054716113954782486 2023-01-21 12:28:13.313994: step: 1224/533, loss: 0.02812042087316513 2023-01-21 12:28:14.519321: step: 1228/533, loss: 0.03990650177001953 2023-01-21 12:28:15.676124: step: 1232/533, loss: 0.01297688577324152 2023-01-21 12:28:16.895610: step: 1236/533, loss: 0.0258147232234478 2023-01-21 12:28:18.073133: step: 1240/533, loss: 0.028432894498109818 2023-01-21 12:28:19.260066: step: 1244/533, loss: 0.010734654031693935 2023-01-21 12:28:20.415912: step: 1248/533, loss: 0.025060415267944336 2023-01-21 12:28:21.622728: step: 1252/533, loss: 0.044008638709783554 2023-01-21 12:28:22.790705: step: 1256/533, loss: 0.0012048721546307206 2023-01-21 12:28:24.009453: step: 1260/533, loss: 0.0008306503295898438 2023-01-21 12:28:25.202139: step: 1264/533, loss: 0.0081024169921875 2023-01-21 12:28:26.376332: step: 1268/533, loss: 0.058712102472782135 2023-01-21 12:28:27.586039: step: 1272/533, loss: 0.013739204034209251 2023-01-21 12:28:28.783742: step: 1276/533, loss: 0.0015888214111328125 2023-01-21 12:28:29.929211: step: 1280/533, loss: 0.13101644814014435 2023-01-21 12:28:31.130435: step: 1284/533, loss: 0.0021768570877611637 2023-01-21 12:28:32.294616: step: 1288/533, loss: 0.0010646821465343237 2023-01-21 12:28:33.455023: step: 1292/533, loss: 0.02703394927084446 2023-01-21 12:28:34.645511: step: 1296/533, loss: 0.005271339789032936 2023-01-21 12:28:35.844178: step: 1300/533, loss: 0.002501964569091797 2023-01-21 12:28:37.023314: step: 1304/533, loss: 0.013520240783691406 2023-01-21 12:28:38.200069: step: 1308/533, loss: 0.0021947859786450863 2023-01-21 12:28:39.370165: step: 1312/533, loss: 0.036550141870975494 2023-01-21 12:28:40.598984: step: 1316/533, loss: 0.0007497787592001259 2023-01-21 12:28:41.771726: step: 1320/533, loss: 0.0007326126797124743 2023-01-21 12:28:42.976911: step: 1324/533, loss: 0.06887874752283096 2023-01-21 12:28:44.139190: step: 1328/533, loss: 2.86102294921875e-05 2023-01-21 12:28:45.315822: step: 1332/533, loss: 0.0006643295055255294 2023-01-21 12:28:46.448937: step: 1336/533, loss: 0.00013718605623580515 2023-01-21 12:28:47.621937: step: 1340/533, loss: 0.04394559934735298 2023-01-21 12:28:48.765189: step: 1344/533, loss: 0.031722258776426315 2023-01-21 12:28:49.954461: step: 1348/533, loss: 0.0006387710454873741 2023-01-21 12:28:51.115329: step: 1352/533, loss: 0.005569267552345991 2023-01-21 12:28:52.323793: step: 1356/533, loss: 0.012485313229262829 2023-01-21 12:28:53.499296: step: 1360/533, loss: 0.0012495993869379163 2023-01-21 12:28:54.650772: step: 1364/533, loss: 0.0001277923583984375 2023-01-21 12:28:55.819160: step: 1368/533, loss: 0.003922081086784601 2023-01-21 12:28:57.012211: step: 1372/533, loss: 0.04140782356262207 2023-01-21 12:28:58.221202: step: 1376/533, loss: 6.046295311534777e-05 2023-01-21 12:28:59.404240: step: 1380/533, loss: 0.004292297642678022 2023-01-21 12:29:00.570987: step: 1384/533, loss: 5.826950291520916e-05 2023-01-21 12:29:01.793048: step: 1388/533, loss: 0.017092324793338776 2023-01-21 12:29:02.949251: step: 1392/533, loss: 0.002660989761352539 2023-01-21 12:29:04.083374: step: 1396/533, loss: 0.00088334089377895 2023-01-21 12:29:05.247917: step: 1400/533, loss: 1.2016297660011332e-05 2023-01-21 12:29:06.434493: step: 1404/533, loss: 0.0024902820587158203 2023-01-21 12:29:07.589875: step: 1408/533, loss: 0.003922462463378906 2023-01-21 12:29:08.776394: step: 1412/533, loss: 0.005764675326645374 2023-01-21 12:29:09.922307: step: 1416/533, loss: 0.004608058836311102 2023-01-21 12:29:11.147297: step: 1420/533, loss: 0.001043939497321844 2023-01-21 12:29:12.325631: step: 1424/533, loss: 0.028919031843543053 2023-01-21 12:29:13.524430: step: 1428/533, loss: 0.000705718994140625 2023-01-21 12:29:14.702153: step: 1432/533, loss: 0.011714077554643154 2023-01-21 12:29:15.909689: step: 1436/533, loss: 0.00036087038461118937 2023-01-21 12:29:17.073308: step: 1440/533, loss: -8.01086389401462e-06 2023-01-21 12:29:18.275112: step: 1444/533, loss: 1.5211106074275449e-05 2023-01-21 12:29:19.454756: step: 1448/533, loss: 0.00012025833711959422 2023-01-21 12:29:20.641106: step: 1452/533, loss: 0.00018367767916060984 2023-01-21 12:29:21.828196: step: 1456/533, loss: 0.00063581473659724 2023-01-21 12:29:22.997153: step: 1460/533, loss: 0.041513923555612564 2023-01-21 12:29:24.204151: step: 1464/533, loss: 0.007718944456428289 2023-01-21 12:29:25.405046: step: 1468/533, loss: 0.011109257116913795 2023-01-21 12:29:26.590685: step: 1472/533, loss: 0.037636902183294296 2023-01-21 12:29:27.783436: step: 1476/533, loss: 0.02476225048303604 2023-01-21 12:29:28.993767: step: 1480/533, loss: 0.00011863708641612902 2023-01-21 12:29:30.234748: step: 1484/533, loss: 0.00042591095552779734 2023-01-21 12:29:31.367275: step: 1488/533, loss: 0.011008739471435547 2023-01-21 12:29:32.551693: step: 1492/533, loss: 0.01116724032908678 2023-01-21 12:29:33.730343: step: 1496/533, loss: 2.593994213384576e-05 2023-01-21 12:29:34.871637: step: 1500/533, loss: 9.430646605324e-05 2023-01-21 12:29:36.098643: step: 1504/533, loss: 0.09664344787597656 2023-01-21 12:29:37.247329: step: 1508/533, loss: 0.033812522888183594 2023-01-21 12:29:38.456341: step: 1512/533, loss: 0.0002402305690338835 2023-01-21 12:29:39.616438: step: 1516/533, loss: 0.0005529880872927606 2023-01-21 12:29:40.745425: step: 1520/533, loss: 1.4829635802016128e-05 2023-01-21 12:29:41.899965: step: 1524/533, loss: 0.04460430145263672 2023-01-21 12:29:43.055924: step: 1528/533, loss: 0.003004360245540738 2023-01-21 12:29:44.231524: step: 1532/533, loss: 0.01606011390686035 2023-01-21 12:29:45.398377: step: 1536/533, loss: 0.012944984249770641 2023-01-21 12:29:46.547612: step: 1540/533, loss: 0.02269144169986248 2023-01-21 12:29:47.667062: step: 1544/533, loss: 0.03440279886126518 2023-01-21 12:29:48.847097: step: 1548/533, loss: 0.007594776339828968 2023-01-21 12:29:50.026748: step: 1552/533, loss: 0.013205528259277344 2023-01-21 12:29:51.174137: step: 1556/533, loss: 0.0015297889476642013 2023-01-21 12:29:52.327808: step: 1560/533, loss: 0.005613517481833696 2023-01-21 12:29:53.507980: step: 1564/533, loss: 0.00030956268892623484 2023-01-21 12:29:54.701010: step: 1568/533, loss: 0.004345702938735485 2023-01-21 12:29:55.905598: step: 1572/533, loss: 0.0006660461658611894 2023-01-21 12:29:57.090379: step: 1576/533, loss: 0.00014314652071334422 2023-01-21 12:29:58.244018: step: 1580/533, loss: 0.016733551397919655 2023-01-21 12:29:59.399474: step: 1584/533, loss: 0.00011148452904308215 2023-01-21 12:30:00.573860: step: 1588/533, loss: 8.0108642578125e-05 2023-01-21 12:30:01.769544: step: 1592/533, loss: 0.000789642333984375 2023-01-21 12:30:02.940164: step: 1596/533, loss: 0.0005065918085165322 2023-01-21 12:30:04.125651: step: 1600/533, loss: 0.004064703360199928 2023-01-21 12:30:05.279496: step: 1604/533, loss: 0.001863193465396762 2023-01-21 12:30:06.448188: step: 1608/533, loss: 0.0022901534102857113 2023-01-21 12:30:07.624850: step: 1612/533, loss: 0.0025522231590002775 2023-01-21 12:30:08.855245: step: 1616/533, loss: 0.005135727114975452 2023-01-21 12:30:09.999108: step: 1620/533, loss: 0.025900553911924362 2023-01-21 12:30:11.196174: step: 1624/533, loss: 0.0015659332275390625 2023-01-21 12:30:12.356415: step: 1628/533, loss: 0.00013856888108421117 2023-01-21 12:30:13.517886: step: 1632/533, loss: 0.008434295654296875 2023-01-21 12:30:14.683575: step: 1636/533, loss: 0.002241802401840687 2023-01-21 12:30:15.849458: step: 1640/533, loss: 4.00543194700731e-06 2023-01-21 12:30:17.005533: step: 1644/533, loss: 0.00035753249539993703 2023-01-21 12:30:18.171560: step: 1648/533, loss: 0.0006530284881591797 2023-01-21 12:30:19.358270: step: 1652/533, loss: 0.07139892876148224 2023-01-21 12:30:20.577586: step: 1656/533, loss: 2.021789623540826e-05 2023-01-21 12:30:21.766281: step: 1660/533, loss: 0.0004839897155761719 2023-01-21 12:30:22.886809: step: 1664/533, loss: 0.00011167526099598035 2023-01-21 12:30:24.054628: step: 1668/533, loss: 0.019932890310883522 2023-01-21 12:30:25.279840: step: 1672/533, loss: 0.0004402160702738911 2023-01-21 12:30:26.412934: step: 1676/533, loss: 0.0004188537714071572 2023-01-21 12:30:27.604562: step: 1680/533, loss: 1.106262243411038e-05 2023-01-21 12:30:28.743087: step: 1684/533, loss: 0.0006122589111328125 2023-01-21 12:30:29.926341: step: 1688/533, loss: 0.003991031553596258 2023-01-21 12:30:31.082248: step: 1692/533, loss: 0.002306175185367465 2023-01-21 12:30:32.230410: step: 1696/533, loss: 4.7874447773210704e-05 2023-01-21 12:30:33.372230: step: 1700/533, loss: 0.0022580146323889494 2023-01-21 12:30:34.555053: step: 1704/533, loss: 0.01635761186480522 2023-01-21 12:30:35.715976: step: 1708/533, loss: 2.622604552016128e-05 2023-01-21 12:30:36.867627: step: 1712/533, loss: 0.0005837440839968622 2023-01-21 12:30:38.063580: step: 1716/533, loss: -1.640319896978326e-05 2023-01-21 12:30:39.250753: step: 1720/533, loss: 0.00027065275935456157 2023-01-21 12:30:40.427055: step: 1724/533, loss: 0.00124950404278934 2023-01-21 12:30:41.601887: step: 1728/533, loss: 0.006613636389374733 2023-01-21 12:30:42.742822: step: 1732/533, loss: 0.0005489349132403731 2023-01-21 12:30:43.918083: step: 1736/533, loss: 0.00011410714068915695 2023-01-21 12:30:45.096960: step: 1740/533, loss: 0.02530527114868164 2023-01-21 12:30:46.252367: step: 1744/533, loss: 0.39516574144363403 2023-01-21 12:30:47.478680: step: 1748/533, loss: 0.03044586069881916 2023-01-21 12:30:48.638760: step: 1752/533, loss: 0.012116623111069202 2023-01-21 12:30:49.827774: step: 1756/533, loss: 0.026382064446806908 2023-01-21 12:30:50.992002: step: 1760/533, loss: 0.00031147003755904734 2023-01-21 12:30:52.153544: step: 1764/533, loss: 0.012786484323441982 2023-01-21 12:30:53.342834: step: 1768/533, loss: 0.00014591217041015625 2023-01-21 12:30:54.482812: step: 1772/533, loss: 0.002908134600147605 2023-01-21 12:30:55.635482: step: 1776/533, loss: 0.00021276474581100047 2023-01-21 12:30:56.777273: step: 1780/533, loss: 6.122589547885582e-05 2023-01-21 12:30:57.904742: step: 1784/533, loss: 5.354881432140246e-05 2023-01-21 12:30:59.124864: step: 1788/533, loss: 0.00426330603659153 2023-01-21 12:31:00.306140: step: 1792/533, loss: 0.003810405731201172 2023-01-21 12:31:01.460029: step: 1796/533, loss: 0.000842237495817244 2023-01-21 12:31:02.639843: step: 1800/533, loss: 0.0013736725086346269 2023-01-21 12:31:03.815181: step: 1804/533, loss: 0.05387058109045029 2023-01-21 12:31:04.980464: step: 1808/533, loss: 0.0006219864008016884 2023-01-21 12:31:06.137473: step: 1812/533, loss: 0.0009620666969567537 2023-01-21 12:31:07.305051: step: 1816/533, loss: 0.00445480365306139 2023-01-21 12:31:08.432053: step: 1820/533, loss: 8.230209641624242e-05 2023-01-21 12:31:09.600696: step: 1824/533, loss: 0.0006955146673135459 2023-01-21 12:31:10.757865: step: 1828/533, loss: 4.425048973644152e-05 2023-01-21 12:31:11.905182: step: 1832/533, loss: 5.1975250244140625e-05 2023-01-21 12:31:13.088505: step: 1836/533, loss: 0.0005017280345782638 2023-01-21 12:31:14.292918: step: 1840/533, loss: 0.00013079644122626632 2023-01-21 12:31:15.492603: step: 1844/533, loss: 0.007572496309876442 2023-01-21 12:31:16.661086: step: 1848/533, loss: 0.016297675669193268 2023-01-21 12:31:17.809153: step: 1852/533, loss: 0.004268741700798273 2023-01-21 12:31:18.953171: step: 1856/533, loss: 0.006599808111786842 2023-01-21 12:31:20.111744: step: 1860/533, loss: 0.16262473165988922 2023-01-21 12:31:21.270601: step: 1864/533, loss: 0.0015596390003338456 2023-01-21 12:31:22.440463: step: 1868/533, loss: 0.0005552291986532509 2023-01-21 12:31:23.640721: step: 1872/533, loss: 0.11578655987977982 2023-01-21 12:31:24.823340: step: 1876/533, loss: 0.029906749725341797 2023-01-21 12:31:25.977078: step: 1880/533, loss: 0.0011843681568279862 2023-01-21 12:31:27.132079: step: 1884/533, loss: 0.003837108612060547 2023-01-21 12:31:28.315653: step: 1888/533, loss: 0.0002840042288880795 2023-01-21 12:31:29.526547: step: 1892/533, loss: -4.959105353918858e-06 2023-01-21 12:31:30.684679: step: 1896/533, loss: 0.03408050537109375 2023-01-21 12:31:31.875867: step: 1900/533, loss: 0.007250118535012007 2023-01-21 12:31:33.052929: step: 1904/533, loss: 8.144378080032766e-05 2023-01-21 12:31:34.271144: step: 1908/533, loss: 0.01952528953552246 2023-01-21 12:31:35.463517: step: 1912/533, loss: 0.000389528286177665 2023-01-21 12:31:36.588114: step: 1916/533, loss: 6.880760338390246e-05 2023-01-21 12:31:37.752989: step: 1920/533, loss: 0.012068557552993298 2023-01-21 12:31:38.949136: step: 1924/533, loss: 0.00035943984403274953 2023-01-21 12:31:40.122380: step: 1928/533, loss: 0.0052436827681958675 2023-01-21 12:31:41.361107: step: 1932/533, loss: 0.007184791844338179 2023-01-21 12:31:42.556146: step: 1936/533, loss: 0.0014678954612463713 2023-01-21 12:31:43.695127: step: 1940/533, loss: 0.0016061782371252775 2023-01-21 12:31:44.881660: step: 1944/533, loss: 0.1350231170654297 2023-01-21 12:31:46.049502: step: 1948/533, loss: 0.02249317243695259 2023-01-21 12:31:47.237676: step: 1952/533, loss: 0.004653358366340399 2023-01-21 12:31:48.423261: step: 1956/533, loss: 0.016799213364720345 2023-01-21 12:31:49.606107: step: 1960/533, loss: 0.016709424555301666 2023-01-21 12:31:50.753046: step: 1964/533, loss: 0.07454991340637207 2023-01-21 12:31:51.912927: step: 1968/533, loss: 0.01202240027487278 2023-01-21 12:31:53.083711: step: 1972/533, loss: 0.0009447097545489669 2023-01-21 12:31:54.214270: step: 1976/533, loss: 0.0615813285112381 2023-01-21 12:31:55.400412: step: 1980/533, loss: 0.0004860878107137978 2023-01-21 12:31:56.578331: step: 1984/533, loss: 0.5839420557022095 2023-01-21 12:31:57.766751: step: 1988/533, loss: 0.0011266708606854081 2023-01-21 12:31:58.914357: step: 1992/533, loss: 0.05644693225622177 2023-01-21 12:32:00.067263: step: 1996/533, loss: 0.0016967772971838713 2023-01-21 12:32:01.288116: step: 2000/533, loss: 3.452301098150201e-05 2023-01-21 12:32:02.442083: step: 2004/533, loss: 0.009470843710005283 2023-01-21 12:32:03.579086: step: 2008/533, loss: 0.0005991458892822266 2023-01-21 12:32:04.739782: step: 2012/533, loss: 5.321502612787299e-05 2023-01-21 12:32:05.921290: step: 2016/533, loss: 5.3882604333921336e-06 2023-01-21 12:32:07.062848: step: 2020/533, loss: 0.001840782118961215 2023-01-21 12:32:08.217487: step: 2024/533, loss: 0.004614782519638538 2023-01-21 12:32:09.371800: step: 2028/533, loss: 2.7942656743107364e-05 2023-01-21 12:32:10.547143: step: 2032/533, loss: 0.004661274142563343 2023-01-21 12:32:11.736393: step: 2036/533, loss: 0.009046172723174095 2023-01-21 12:32:12.940876: step: 2040/533, loss: 0.0019807815551757812 2023-01-21 12:32:14.158756: step: 2044/533, loss: 0.037517644464969635 2023-01-21 12:32:15.359452: step: 2048/533, loss: 0.010144973173737526 2023-01-21 12:32:16.525997: step: 2052/533, loss: 0.3174971640110016 2023-01-21 12:32:17.695537: step: 2056/533, loss: 0.042220547795295715 2023-01-21 12:32:18.850801: step: 2060/533, loss: 6.141662743175402e-05 2023-01-21 12:32:20.012208: step: 2064/533, loss: 0.018467331305146217 2023-01-21 12:32:21.164325: step: 2068/533, loss: 0.0009114264976233244 2023-01-21 12:32:22.341168: step: 2072/533, loss: 0.02417278289794922 2023-01-21 12:32:23.539303: step: 2076/533, loss: 0.00592803955078125 2023-01-21 12:32:24.720133: step: 2080/533, loss: 0.003192997071892023 2023-01-21 12:32:25.897360: step: 2084/533, loss: 0.0006876945262774825 2023-01-21 12:32:27.045593: step: 2088/533, loss: 0.005936718080192804 2023-01-21 12:32:28.208743: step: 2092/533, loss: 0.010973739437758923 2023-01-21 12:32:29.387437: step: 2096/533, loss: 0.0002742767392192036 2023-01-21 12:32:30.549747: step: 2100/533, loss: 0.0765453353524208 2023-01-21 12:32:31.713587: step: 2104/533, loss: 0.08053645491600037 2023-01-21 12:32:32.881178: step: 2108/533, loss: 0.0009613991132937372 2023-01-21 12:32:34.067618: step: 2112/533, loss: 0.002557468367740512 2023-01-21 12:32:35.260200: step: 2116/533, loss: 0.00816116388887167 2023-01-21 12:32:36.408975: step: 2120/533, loss: 0.0049690245650708675 2023-01-21 12:32:37.572909: step: 2124/533, loss: 0.0006447792402468622 2023-01-21 12:32:38.748098: step: 2128/533, loss: 5.7220458984375e-06 2023-01-21 12:32:39.915240: step: 2132/533, loss: 4.725456528831273e-05 ================================================== Loss: 0.017 -------------------- Dev: {'event': {'p': 0.6135416666666667, 'r': 0.7842876165113183, 'f1': 0.6884862653419054}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Test: {'event': {'p': 0.6512583461736005, 'r': 0.8107416879795396, 'f1': 0.7223013386499573}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Chinese: {'event': {'p': 0.5542168674698795, 'r': 0.8518518518518519, 'f1': 0.6715328467153285}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Korean: {'event': {'p': 0.6808510638297872, 'r': 0.5079365079365079, 'f1': 0.5818181818181817}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Russian: {'event': {'p': 0.4864864864864865, 'r': 0.5, 'f1': 0.4931506849315069}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:33:20.674612: step: 4/533, loss: 4.6348573960131034e-05 2023-01-21 12:33:21.884969: step: 8/533, loss: 0.01979808881878853 2023-01-21 12:33:23.055627: step: 12/533, loss: 0.0006522178300656378 2023-01-21 12:33:24.207942: step: 16/533, loss: 0.0006101608159951866 2023-01-21 12:33:25.374500: step: 20/533, loss: 0.0013151168823242188 2023-01-21 12:33:26.545214: step: 24/533, loss: 1.3637542906508315e-05 2023-01-21 12:33:27.685895: step: 28/533, loss: 0.002569293836131692 2023-01-21 12:33:28.837165: step: 32/533, loss: 0.0043472289107739925 2023-01-21 12:33:30.052710: step: 36/533, loss: 0.0025729178451001644 2023-01-21 12:33:31.223292: step: 40/533, loss: 4.043579247081652e-05 2023-01-21 12:33:32.413295: step: 44/533, loss: 0.004203224554657936 2023-01-21 12:33:33.560832: step: 48/533, loss: 0.00024824144202284515 2023-01-21 12:33:34.698013: step: 52/533, loss: 3.814697265625e-05 2023-01-21 12:33:35.908566: step: 56/533, loss: 0.014941597357392311 2023-01-21 12:33:37.099468: step: 60/533, loss: 7.143020775401965e-05 2023-01-21 12:33:38.279218: step: 64/533, loss: 0.9156786203384399 2023-01-21 12:33:39.468191: step: 68/533, loss: 0.0006795883527956903 2023-01-21 12:33:40.655089: step: 72/533, loss: 0.003953456878662109 2023-01-21 12:33:41.833100: step: 76/533, loss: 0.01984882354736328 2023-01-21 12:33:42.973276: step: 80/533, loss: 2.47955313170678e-06 2023-01-21 12:33:44.201289: step: 84/533, loss: 0.0030334473121911287 2023-01-21 12:33:45.383358: step: 88/533, loss: 0.005707931704819202 2023-01-21 12:33:46.601610: step: 92/533, loss: 0.00021300316439010203 2023-01-21 12:33:47.728389: step: 96/533, loss: 6.270408630371094e-05 2023-01-21 12:33:48.847570: step: 100/533, loss: 0.05145569145679474 2023-01-21 12:33:50.040910: step: 104/533, loss: 0.007511043455451727 2023-01-21 12:33:51.193576: step: 108/533, loss: -4.19616708313697e-06 2023-01-21 12:33:52.365609: step: 112/533, loss: 3.62396240234375e-05 2023-01-21 12:33:53.498647: step: 116/533, loss: 0.0008966922760009766 2023-01-21 12:33:54.656662: step: 120/533, loss: 2.8610231765924254e-06 2023-01-21 12:33:55.863011: step: 124/533, loss: 0.02390899695456028 2023-01-21 12:33:57.047568: step: 128/533, loss: 0.07706312835216522 2023-01-21 12:33:58.193737: step: 132/533, loss: 1.7929078239831142e-05 2023-01-21 12:33:59.399523: step: 136/533, loss: 0.058489225804805756 2023-01-21 12:34:00.592667: step: 140/533, loss: 2.86102294921875e-06 2023-01-21 12:34:01.744362: step: 144/533, loss: 0.01895446702837944 2023-01-21 12:34:02.942811: step: 148/533, loss: 0.05796413496136665 2023-01-21 12:34:04.088120: step: 152/533, loss: 0.001422119210474193 2023-01-21 12:34:05.251522: step: 156/533, loss: 0.007993698120117188 2023-01-21 12:34:06.408462: step: 160/533, loss: 0.00015840530977584422 2023-01-21 12:34:07.549613: step: 164/533, loss: 0.0163726806640625 2023-01-21 12:34:08.709940: step: 168/533, loss: 0.004786300472915173 2023-01-21 12:34:09.857087: step: 172/533, loss: 0.012909699231386185 2023-01-21 12:34:11.069227: step: 176/533, loss: 0.0011219978332519531 2023-01-21 12:34:12.241212: step: 180/533, loss: 0.0015806198352947831 2023-01-21 12:34:13.427526: step: 184/533, loss: 4.3201445805607364e-05 2023-01-21 12:34:14.640805: step: 188/533, loss: 0.0008132934453897178 2023-01-21 12:34:15.784600: step: 192/533, loss: 5.53131121705519e-06 2023-01-21 12:34:16.997640: step: 196/533, loss: 0.00020055771165061742 2023-01-21 12:34:18.152699: step: 200/533, loss: 0.009841728955507278 2023-01-21 12:34:19.341230: step: 204/533, loss: 0.006096077151596546 2023-01-21 12:34:20.506319: step: 208/533, loss: 0.000798344612121582 2023-01-21 12:34:21.667878: step: 212/533, loss: 0.014163780957460403 2023-01-21 12:34:22.856863: step: 216/533, loss: 0.01322474516928196 2023-01-21 12:34:24.049523: step: 220/533, loss: 0.001019382500089705 2023-01-21 12:34:25.239629: step: 224/533, loss: 0.01369171217083931 2023-01-21 12:34:26.417353: step: 228/533, loss: 0.0028247833251953125 2023-01-21 12:34:27.616191: step: 232/533, loss: 0.0005532264476642013 2023-01-21 12:34:28.790479: step: 236/533, loss: 0.0021217346657067537 2023-01-21 12:34:29.929622: step: 240/533, loss: 0.0004964828258380294 2023-01-21 12:34:31.108748: step: 244/533, loss: 1.3065338862361386e-05 2023-01-21 12:34:32.264228: step: 248/533, loss: 0.01525197084993124 2023-01-21 12:34:33.423352: step: 252/533, loss: 0.0016655444633215666 2023-01-21 12:34:34.598035: step: 256/533, loss: 0.008540058508515358 2023-01-21 12:34:35.785988: step: 260/533, loss: 0.005357933230698109 2023-01-21 12:34:36.954486: step: 264/533, loss: 0.0001466751127736643 2023-01-21 12:34:38.131409: step: 268/533, loss: 0.040259361267089844 2023-01-21 12:34:39.287806: step: 272/533, loss: 0.0012897491687908769 2023-01-21 12:34:40.452947: step: 276/533, loss: 0.0006779670948162675 2023-01-21 12:34:41.613179: step: 280/533, loss: 0.0011564254527911544 2023-01-21 12:34:42.804094: step: 284/533, loss: 0.0009406090248376131 2023-01-21 12:34:43.957054: step: 288/533, loss: 6.198883056640625e-05 2023-01-21 12:34:45.214925: step: 292/533, loss: 0.001065158867277205 2023-01-21 12:34:46.402810: step: 296/533, loss: 6.64711042190902e-05 2023-01-21 12:34:47.528761: step: 300/533, loss: 0.003765964647755027 2023-01-21 12:34:48.716108: step: 304/533, loss: 0.03945455700159073 2023-01-21 12:34:49.912852: step: 308/533, loss: 0.0054145813919603825 2023-01-21 12:34:51.120869: step: 312/533, loss: 0.0028831481467932463 2023-01-21 12:34:52.297493: step: 316/533, loss: 0.00039892198401503265 2023-01-21 12:34:53.502652: step: 320/533, loss: 0.0007991790771484375 2023-01-21 12:34:54.691675: step: 324/533, loss: 0.035404156893491745 2023-01-21 12:34:55.876536: step: 328/533, loss: 0.0020177841652184725 2023-01-21 12:34:57.054304: step: 332/533, loss: 0.003953456878662109 2023-01-21 12:34:58.222327: step: 336/533, loss: 0.0003470420779194683 2023-01-21 12:34:59.408838: step: 340/533, loss: 0.00016918181790970266 2023-01-21 12:35:00.566162: step: 344/533, loss: 0.02539677731692791 2023-01-21 12:35:01.694215: step: 348/533, loss: 0.000522613525390625 2023-01-21 12:35:02.876194: step: 352/533, loss: 2.117157055181451e-05 2023-01-21 12:35:04.055160: step: 356/533, loss: 0.00034656524076126516 2023-01-21 12:35:05.222607: step: 360/533, loss: 0.04780769348144531 2023-01-21 12:35:06.405838: step: 364/533, loss: 9.5367431640625e-06 2023-01-21 12:35:07.545908: step: 368/533, loss: 0.0003541946643963456 2023-01-21 12:35:08.738934: step: 372/533, loss: 0.0038636208046227694 2023-01-21 12:35:09.928531: step: 376/533, loss: 0.010775757022202015 2023-01-21 12:35:11.090819: step: 380/533, loss: 0.007929230108857155 2023-01-21 12:35:12.262847: step: 384/533, loss: 0.0018610954284667969 2023-01-21 12:35:13.416715: step: 388/533, loss: -6.67572021484375e-06 2023-01-21 12:35:14.582492: step: 392/533, loss: 0.0009245872497558594 2023-01-21 12:35:15.762644: step: 396/533, loss: 0.0067008971236646175 2023-01-21 12:35:16.915395: step: 400/533, loss: 1.4781951449549524e-06 2023-01-21 12:35:18.126502: step: 404/533, loss: 0.0010811806423589587 2023-01-21 12:35:19.316474: step: 408/533, loss: 0.00144023890607059 2023-01-21 12:35:20.556584: step: 412/533, loss: 0.04804077371954918 2023-01-21 12:35:21.729843: step: 416/533, loss: 0.003083610674366355 2023-01-21 12:35:22.892187: step: 420/533, loss: 0.14540882408618927 2023-01-21 12:35:24.099444: step: 424/533, loss: 0.07040577381849289 2023-01-21 12:35:25.317998: step: 428/533, loss: 0.10770759731531143 2023-01-21 12:35:26.510124: step: 432/533, loss: 0.001545047853142023 2023-01-21 12:35:27.693354: step: 436/533, loss: 0.00029778480529785156 2023-01-21 12:35:28.860527: step: 440/533, loss: 0.0012248039711266756 2023-01-21 12:35:30.085277: step: 444/533, loss: 0.0018113135593011975 2023-01-21 12:35:31.259918: step: 448/533, loss: 0.0013878822792321444 2023-01-21 12:35:32.476404: step: 452/533, loss: 0.015762805938720703 2023-01-21 12:35:33.654231: step: 456/533, loss: 0.006797600071877241 2023-01-21 12:35:34.850718: step: 460/533, loss: 0.030245399102568626 2023-01-21 12:35:36.009946: step: 464/533, loss: 0.00077228550799191 2023-01-21 12:35:37.259315: step: 468/533, loss: 0.0056113312020897865 2023-01-21 12:35:38.398810: step: 472/533, loss: 0.06577315926551819 2023-01-21 12:35:39.545820: step: 476/533, loss: 7.305145118152723e-05 2023-01-21 12:35:40.686198: step: 480/533, loss: 0.043172646313905716 2023-01-21 12:35:41.871062: step: 484/533, loss: 0.003369951155036688 2023-01-21 12:35:43.065251: step: 488/533, loss: 1.889200210571289 2023-01-21 12:35:44.229528: step: 492/533, loss: 0.00013494491577148438 2023-01-21 12:35:45.441513: step: 496/533, loss: 0.03617754206061363 2023-01-21 12:35:46.643219: step: 500/533, loss: 0.002237987471744418 2023-01-21 12:35:47.840176: step: 504/533, loss: 0.009354686364531517 2023-01-21 12:35:48.975948: step: 508/533, loss: 0.04181298986077309 2023-01-21 12:35:50.148978: step: 512/533, loss: 0.09755564481019974 2023-01-21 12:35:51.363979: step: 516/533, loss: 0.016693497076630592 2023-01-21 12:35:52.537962: step: 520/533, loss: 0.006962776184082031 2023-01-21 12:35:53.733777: step: 524/533, loss: 0.0037490844260901213 2023-01-21 12:35:54.938837: step: 528/533, loss: 0.0029821395874023438 2023-01-21 12:35:56.125087: step: 532/533, loss: 0.0003068924124818295 2023-01-21 12:35:57.322103: step: 536/533, loss: 0.004728508181869984 2023-01-21 12:35:58.502775: step: 540/533, loss: 2.746581958490424e-05 2023-01-21 12:35:59.649394: step: 544/533, loss: 0.00024232864961959422 2023-01-21 12:36:00.801291: step: 548/533, loss: 0.00011367797560524195 2023-01-21 12:36:01.942575: step: 552/533, loss: 0.00011091232590842992 2023-01-21 12:36:03.141709: step: 556/533, loss: 0.006826496217399836 2023-01-21 12:36:04.316342: step: 560/533, loss: 0.034980013966560364 2023-01-21 12:36:05.513217: step: 564/533, loss: 0.022213172167539597 2023-01-21 12:36:06.689223: step: 568/533, loss: 0.02323455922305584 2023-01-21 12:36:07.890721: step: 572/533, loss: 2.1293745040893555 2023-01-21 12:36:09.043612: step: 576/533, loss: 0.005896091461181641 2023-01-21 12:36:10.204899: step: 580/533, loss: 0.002432918408885598 2023-01-21 12:36:11.358063: step: 584/533, loss: 0.0011341571807861328 2023-01-21 12:36:12.508280: step: 588/533, loss: 0.011813854798674583 2023-01-21 12:36:13.711596: step: 592/533, loss: 1.61647803906817e-05 2023-01-21 12:36:14.877719: step: 596/533, loss: 0.023606205359101295 2023-01-21 12:36:16.093970: step: 600/533, loss: 0.003218269208446145 2023-01-21 12:36:17.251220: step: 604/533, loss: 0.0007703781593590975 2023-01-21 12:36:18.465792: step: 608/533, loss: 0.036215901374816895 2023-01-21 12:36:19.628001: step: 612/533, loss: 0.9816697835922241 2023-01-21 12:36:20.787174: step: 616/533, loss: 0.0006397724500857294 2023-01-21 12:36:21.963004: step: 620/533, loss: 0.006338882725685835 2023-01-21 12:36:23.145302: step: 624/533, loss: 0.08328447490930557 2023-01-21 12:36:24.333350: step: 628/533, loss: 0.007692146580666304 2023-01-21 12:36:25.483533: step: 632/533, loss: 0.0005227089277468622 2023-01-21 12:36:26.645807: step: 636/533, loss: 6.50405854685232e-05 2023-01-21 12:36:27.825565: step: 640/533, loss: 0.00810155924409628 2023-01-21 12:36:29.022762: step: 644/533, loss: 0.04750204086303711 2023-01-21 12:36:30.178123: step: 648/533, loss: 0.0011665343772619963 2023-01-21 12:36:31.361803: step: 652/533, loss: 0.07217235863208771 2023-01-21 12:36:32.520968: step: 656/533, loss: 0.0003842354053631425 2023-01-21 12:36:33.699993: step: 660/533, loss: 0.0027728083077818155 2023-01-21 12:36:34.872409: step: 664/533, loss: 9.498596045887098e-05 2023-01-21 12:36:36.060669: step: 668/533, loss: 0.007309627719223499 2023-01-21 12:36:37.179479: step: 672/533, loss: 0.02184305153787136 2023-01-21 12:36:38.365953: step: 676/533, loss: 0.06633082032203674 2023-01-21 12:36:39.501651: step: 680/533, loss: 0.0002643585321493447 2023-01-21 12:36:40.692328: step: 684/533, loss: 0.0016119956271722913 2023-01-21 12:36:41.862035: step: 688/533, loss: 0.00028591154841706157 2023-01-21 12:36:43.036986: step: 692/533, loss: 0.021193992346525192 2023-01-21 12:36:44.179314: step: 696/533, loss: 0.0004324913024902344 2023-01-21 12:36:45.362308: step: 700/533, loss: 0.0001886367826955393 2023-01-21 12:36:46.569270: step: 704/533, loss: 0.015512657351791859 2023-01-21 12:36:47.767668: step: 708/533, loss: 0.5523599982261658 2023-01-21 12:36:48.920558: step: 712/533, loss: 0.0019481659401208162 2023-01-21 12:36:50.071723: step: 716/533, loss: 0.014204693026840687 2023-01-21 12:36:51.256650: step: 720/533, loss: 0.001001644181087613 2023-01-21 12:36:52.430864: step: 724/533, loss: 0.00131902692373842 2023-01-21 12:36:53.593772: step: 728/533, loss: 0.012248230166733265 2023-01-21 12:36:54.756094: step: 732/533, loss: 0.022366046905517578 2023-01-21 12:36:55.921766: step: 736/533, loss: 0.0021962167229503393 2023-01-21 12:36:57.065184: step: 740/533, loss: 2.346038854739163e-05 2023-01-21 12:36:58.235227: step: 744/533, loss: 0.0032279968727380037 2023-01-21 12:36:59.360318: step: 748/533, loss: 6.189346458995715e-05 2023-01-21 12:37:00.557347: step: 752/533, loss: 0.0012270926963537931 2023-01-21 12:37:01.757758: step: 756/533, loss: 0.040390968322753906 2023-01-21 12:37:02.949287: step: 760/533, loss: 0.006791782099753618 2023-01-21 12:37:04.085116: step: 764/533, loss: -3.24249276673072e-06 2023-01-21 12:37:05.280995: step: 768/533, loss: 0.04647274315357208 2023-01-21 12:37:06.453245: step: 772/533, loss: 0.014141177758574486 2023-01-21 12:37:07.633524: step: 776/533, loss: 6.67572021484375e-05 2023-01-21 12:37:08.814547: step: 780/533, loss: 0.03485088422894478 2023-01-21 12:37:09.978511: step: 784/533, loss: 0.0005096435779705644 2023-01-21 12:37:11.164528: step: 788/533, loss: 0.027565479278564453 2023-01-21 12:37:12.367537: step: 792/533, loss: 0.08724422752857208 2023-01-21 12:37:13.512280: step: 796/533, loss: 0.006330347154289484 2023-01-21 12:37:14.674411: step: 800/533, loss: 0.05926513671875 2023-01-21 12:37:15.887066: step: 804/533, loss: 0.0013380051823332906 2023-01-21 12:37:17.077565: step: 808/533, loss: 0.029644252732396126 2023-01-21 12:37:18.241086: step: 812/533, loss: 0.013859033584594727 2023-01-21 12:37:19.432279: step: 816/533, loss: 0.0990932434797287 2023-01-21 12:37:20.589841: step: 820/533, loss: 0.0036014558281749487 2023-01-21 12:37:21.784871: step: 824/533, loss: 0.021534014493227005 2023-01-21 12:37:22.941223: step: 828/533, loss: 0.09225676208734512 2023-01-21 12:37:24.060171: step: 832/533, loss: 0.00712165841832757 2023-01-21 12:37:25.238800: step: 836/533, loss: 2.059936559817288e-05 2023-01-21 12:37:26.394393: step: 840/533, loss: 0.004351520445197821 2023-01-21 12:37:27.538783: step: 844/533, loss: 0.011018180288374424 2023-01-21 12:37:28.741730: step: 848/533, loss: 0.025036906823515892 2023-01-21 12:37:29.907244: step: 852/533, loss: 0.0002452850458212197 2023-01-21 12:37:31.113156: step: 856/533, loss: 0.002333450596779585 2023-01-21 12:37:32.245331: step: 860/533, loss: 0.0005685806390829384 2023-01-21 12:37:33.473915: step: 864/533, loss: 0.00010480881610419601 2023-01-21 12:37:34.647203: step: 868/533, loss: 0.023266220465302467 2023-01-21 12:37:35.852835: step: 872/533, loss: 0.07908955216407776 2023-01-21 12:37:37.051238: step: 876/533, loss: 0.003699874971061945 2023-01-21 12:37:38.254623: step: 880/533, loss: 0.030781077221035957 2023-01-21 12:37:39.457936: step: 884/533, loss: 0.049390316009521484 2023-01-21 12:37:40.642592: step: 888/533, loss: 0.008008194155991077 2023-01-21 12:37:41.835799: step: 892/533, loss: 0.0022375108674168587 2023-01-21 12:37:43.000341: step: 896/533, loss: 0.0018934727413579822 2023-01-21 12:37:44.216537: step: 900/533, loss: 0.0015453338855877519 2023-01-21 12:37:45.376034: step: 904/533, loss: 0.10147485136985779 2023-01-21 12:37:46.577184: step: 908/533, loss: 0.010884379968047142 2023-01-21 12:37:47.732175: step: 912/533, loss: 0.0781635269522667 2023-01-21 12:37:48.860649: step: 916/533, loss: 4.100799560546875e-05 2023-01-21 12:37:50.030055: step: 920/533, loss: 0.00021457672119140625 2023-01-21 12:37:51.215902: step: 924/533, loss: 0.030443573370575905 2023-01-21 12:37:52.392925: step: 928/533, loss: 0.026337813585996628 2023-01-21 12:37:53.598541: step: 932/533, loss: 0.005729484371840954 2023-01-21 12:37:54.773319: step: 936/533, loss: 0.00021305083646439016 2023-01-21 12:37:55.978079: step: 940/533, loss: 0.026098251342773438 2023-01-21 12:37:57.136819: step: 944/533, loss: 0.045636557042598724 2023-01-21 12:37:58.307972: step: 948/533, loss: 0.043384552001953125 2023-01-21 12:37:59.491073: step: 952/533, loss: 0.009241295047104359 2023-01-21 12:38:00.684750: step: 956/533, loss: 0.000935936055611819 2023-01-21 12:38:01.818115: step: 960/533, loss: 0.05775656923651695 2023-01-21 12:38:02.980661: step: 964/533, loss: 0.11051978915929794 2023-01-21 12:38:04.143863: step: 968/533, loss: 0.05667925253510475 2023-01-21 12:38:05.352116: step: 972/533, loss: 0.006842327304184437 2023-01-21 12:38:06.562439: step: 976/533, loss: 0.04330272972583771 2023-01-21 12:38:07.740902: step: 980/533, loss: 0.001110172364860773 2023-01-21 12:38:08.976399: step: 984/533, loss: 0.07031986117362976 2023-01-21 12:38:10.185296: step: 988/533, loss: 0.0008275032741948962 2023-01-21 12:38:11.376167: step: 992/533, loss: 0.0011075020302087069 2023-01-21 12:38:12.562112: step: 996/533, loss: 0.06190691143274307 2023-01-21 12:38:13.751656: step: 1000/533, loss: 0.007774639409035444 2023-01-21 12:38:14.934681: step: 1004/533, loss: 0.0054145813919603825 2023-01-21 12:38:16.082369: step: 1008/533, loss: 0.0041713714599609375 2023-01-21 12:38:17.210952: step: 1012/533, loss: 0.00029468536376953125 2023-01-21 12:38:18.368327: step: 1016/533, loss: 0.0006761550903320312 2023-01-21 12:38:19.536263: step: 1020/533, loss: 0.03555736690759659 2023-01-21 12:38:20.728575: step: 1024/533, loss: 0.048803139477968216 2023-01-21 12:38:21.949279: step: 1028/533, loss: 1.983642505365424e-05 2023-01-21 12:38:23.137666: step: 1032/533, loss: 0.00019598007202148438 2023-01-21 12:38:24.357866: step: 1036/533, loss: 1.1171705722808838 2023-01-21 12:38:25.637254: step: 1040/533, loss: 0.001381874200887978 2023-01-21 12:38:26.842257: step: 1044/533, loss: 9.155274165095761e-05 2023-01-21 12:38:28.001480: step: 1048/533, loss: 8.106231689453125e-05 2023-01-21 12:38:29.169739: step: 1052/533, loss: 0.00841598492115736 2023-01-21 12:38:30.342335: step: 1056/533, loss: 0.00014925003051757812 2023-01-21 12:38:31.493262: step: 1060/533, loss: 0.006827068515121937 2023-01-21 12:38:32.716231: step: 1064/533, loss: 0.05668392404913902 2023-01-21 12:38:33.900941: step: 1068/533, loss: 0.004292106721550226 2023-01-21 12:38:35.105801: step: 1072/533, loss: 0.0001983642578125 2023-01-21 12:38:36.323031: step: 1076/533, loss: 0.006275367923080921 2023-01-21 12:38:37.492877: step: 1080/533, loss: 4.110336521989666e-05 2023-01-21 12:38:38.697097: step: 1084/533, loss: -1.3971328371553682e-05 2023-01-21 12:38:39.840111: step: 1088/533, loss: 0.03936729207634926 2023-01-21 12:38:41.011951: step: 1092/533, loss: 2.4700164431123994e-05 2023-01-21 12:38:42.218751: step: 1096/533, loss: 0.0028278352692723274 2023-01-21 12:38:43.370507: step: 1100/533, loss: 0.007265186402946711 2023-01-21 12:38:44.538614: step: 1104/533, loss: 0.0366390235722065 2023-01-21 12:38:45.709248: step: 1108/533, loss: 0.0009763717534951866 2023-01-21 12:38:46.875483: step: 1112/533, loss: 0.0024343968834728003 2023-01-21 12:38:48.040343: step: 1116/533, loss: 0.01678180694580078 2023-01-21 12:38:49.183362: step: 1120/533, loss: 0.0451175719499588 2023-01-21 12:38:50.328873: step: 1124/533, loss: 0.0002361297665629536 2023-01-21 12:38:51.514531: step: 1128/533, loss: 0.008493232540786266 2023-01-21 12:38:52.725123: step: 1132/533, loss: 0.03020477294921875 2023-01-21 12:38:53.928481: step: 1136/533, loss: 0.003135872073471546 2023-01-21 12:38:55.115305: step: 1140/533, loss: 0.020868968218564987 2023-01-21 12:38:56.311193: step: 1144/533, loss: 0.0019865036010742188 2023-01-21 12:38:57.476188: step: 1148/533, loss: 0.000226879128604196 2023-01-21 12:38:58.622799: step: 1152/533, loss: 0.0009277344215661287 2023-01-21 12:38:59.787970: step: 1156/533, loss: 0.0006402969593182206 2023-01-21 12:39:00.999190: step: 1160/533, loss: 0.00225658412091434 2023-01-21 12:39:02.181794: step: 1164/533, loss: 0.004957771394401789 2023-01-21 12:39:03.376669: step: 1168/533, loss: 2.727508581301663e-05 2023-01-21 12:39:04.571671: step: 1172/533, loss: 0.0015538216102868319 2023-01-21 12:39:05.717302: step: 1176/533, loss: 0.001525783445686102 2023-01-21 12:39:06.863815: step: 1180/533, loss: 0.012783623300492764 2023-01-21 12:39:08.047737: step: 1184/533, loss: 0.0001906395045807585 2023-01-21 12:39:09.194196: step: 1188/533, loss: 0.010441686026751995 2023-01-21 12:39:10.415155: step: 1192/533, loss: 0.01654043234884739 2023-01-21 12:39:11.603135: step: 1196/533, loss: 0.0035977838560938835 2023-01-21 12:39:12.761938: step: 1200/533, loss: 0.0007568359724245965 2023-01-21 12:39:13.919033: step: 1204/533, loss: 0.00042791367741301656 2023-01-21 12:39:15.096974: step: 1208/533, loss: 0.10214042663574219 2023-01-21 12:39:16.275675: step: 1212/533, loss: 1.544952465337701e-05 2023-01-21 12:39:17.418413: step: 1216/533, loss: 0.0001062512383214198 2023-01-21 12:39:18.586947: step: 1220/533, loss: 0.00014152527728583664 2023-01-21 12:39:19.775360: step: 1224/533, loss: 0.00012798310490325093 2023-01-21 12:39:20.943207: step: 1228/533, loss: 0.06654767692089081 2023-01-21 12:39:22.098498: step: 1232/533, loss: 0.001120758126489818 2023-01-21 12:39:23.328039: step: 1236/533, loss: 0.004656028933823109 2023-01-21 12:39:24.520033: step: 1240/533, loss: 0.06841182708740234 2023-01-21 12:39:25.694630: step: 1244/533, loss: 0.0008581161964684725 2023-01-21 12:39:26.837425: step: 1248/533, loss: 0.01417932566255331 2023-01-21 12:39:27.999630: step: 1252/533, loss: 0.11068897694349289 2023-01-21 12:39:29.201135: step: 1256/533, loss: 0.004182625096291304 2023-01-21 12:39:30.341898: step: 1260/533, loss: 0.00025472641573287547 2023-01-21 12:39:31.540856: step: 1264/533, loss: 0.008537864312529564 2023-01-21 12:39:32.690444: step: 1268/533, loss: 0.0010258675320073962 2023-01-21 12:39:33.855907: step: 1272/533, loss: 0.013204192742705345 2023-01-21 12:39:35.034361: step: 1276/533, loss: 0.000745582627132535 2023-01-21 12:39:36.221420: step: 1280/533, loss: 0.0014459609519690275 2023-01-21 12:39:37.384562: step: 1284/533, loss: 0.0023626326583325863 2023-01-21 12:39:38.539289: step: 1288/533, loss: 0.001220750855281949 2023-01-21 12:39:39.734050: step: 1292/533, loss: 7.810592069290578e-05 2023-01-21 12:39:40.921898: step: 1296/533, loss: 0.011261940002441406 2023-01-21 12:39:42.093024: step: 1300/533, loss: 0.012563610449433327 2023-01-21 12:39:43.230489: step: 1304/533, loss: 0.0019123078091070056 2023-01-21 12:39:44.417613: step: 1308/533, loss: 0.008049964904785156 2023-01-21 12:39:45.611795: step: 1312/533, loss: 0.0001605987490620464 2023-01-21 12:39:46.792111: step: 1316/533, loss: 0.003508329391479492 2023-01-21 12:39:47.968430: step: 1320/533, loss: 0.06928787380456924 2023-01-21 12:39:49.167874: step: 1324/533, loss: 0.0006918907165527344 2023-01-21 12:39:50.354974: step: 1328/533, loss: 0.03987693786621094 2023-01-21 12:39:51.500149: step: 1332/533, loss: 0.009356594644486904 2023-01-21 12:39:52.702373: step: 1336/533, loss: 0.002071189694106579 2023-01-21 12:39:53.867804: step: 1340/533, loss: 0.00019121171499136835 2023-01-21 12:39:55.069003: step: 1344/533, loss: 0.0008920669206418097 2023-01-21 12:39:56.232757: step: 1348/533, loss: 0.0001051902727340348 2023-01-21 12:39:57.403541: step: 1352/533, loss: 0.00044116974459029734 2023-01-21 12:39:58.573013: step: 1356/533, loss: 0.010896396823227406 2023-01-21 12:39:59.745332: step: 1360/533, loss: 0.0002187728969147429 2023-01-21 12:40:00.938035: step: 1364/533, loss: 0.014450455084443092 2023-01-21 12:40:02.141337: step: 1368/533, loss: 2.403259350103326e-05 2023-01-21 12:40:03.341670: step: 1372/533, loss: 0.015905380249023438 2023-01-21 12:40:04.510073: step: 1376/533, loss: 0.00020828247943427414 2023-01-21 12:40:05.660020: step: 1380/533, loss: 0.0001461505889892578 2023-01-21 12:40:06.828431: step: 1384/533, loss: 0.008422470651566982 2023-01-21 12:40:07.975016: step: 1388/533, loss: 0.00789489783346653 2023-01-21 12:40:09.134824: step: 1392/533, loss: 0.018627356737852097 2023-01-21 12:40:10.269881: step: 1396/533, loss: 0.00071291922358796 2023-01-21 12:40:11.446731: step: 1400/533, loss: 2.765655517578125e-05 2023-01-21 12:40:12.620269: step: 1404/533, loss: 0.06094570457935333 2023-01-21 12:40:13.755330: step: 1408/533, loss: 7.05718994140625e-05 2023-01-21 12:40:14.911399: step: 1412/533, loss: 3.62396240234375e-05 2023-01-21 12:40:16.076588: step: 1416/533, loss: 0.0292219165712595 2023-01-21 12:40:17.222180: step: 1420/533, loss: 0.00011034011549782008 2023-01-21 12:40:18.387248: step: 1424/533, loss: 0.011471081525087357 2023-01-21 12:40:19.562202: step: 1428/533, loss: 0.006596899125725031 2023-01-21 12:40:20.836156: step: 1432/533, loss: 0.023598099127411842 2023-01-21 12:40:22.007121: step: 1436/533, loss: 0.01854524575173855 2023-01-21 12:40:23.149712: step: 1440/533, loss: 8.583069188716763e-07 2023-01-21 12:40:24.321067: step: 1444/533, loss: 0.0002729415718931705 2023-01-21 12:40:25.512418: step: 1448/533, loss: 0.0002134323149221018 2023-01-21 12:40:26.652261: step: 1452/533, loss: 0.018523788079619408 2023-01-21 12:40:27.839843: step: 1456/533, loss: 0.0010927199618890882 2023-01-21 12:40:29.032913: step: 1460/533, loss: 0.010115623474121094 2023-01-21 12:40:30.177545: step: 1464/533, loss: 0.0007582664838992059 2023-01-21 12:40:31.336120: step: 1468/533, loss: 0.000745582627132535 2023-01-21 12:40:32.531617: step: 1472/533, loss: 0.019166089594364166 2023-01-21 12:40:33.675726: step: 1476/533, loss: 0.020336437970399857 2023-01-21 12:40:34.846996: step: 1480/533, loss: 0.0014175415271893144 2023-01-21 12:40:35.998913: step: 1484/533, loss: 0.004273796454071999 2023-01-21 12:40:37.173696: step: 1488/533, loss: 0.025166893377900124 2023-01-21 12:40:38.327696: step: 1492/533, loss: 0.0003503799671307206 2023-01-21 12:40:39.463148: step: 1496/533, loss: 0.0001297950657317415 2023-01-21 12:40:40.611722: step: 1500/533, loss: 0.0010652542114257812 2023-01-21 12:40:41.841792: step: 1504/533, loss: 0.05767173692584038 2023-01-21 12:40:43.021259: step: 1508/533, loss: 0.0036220550537109375 2023-01-21 12:40:44.191369: step: 1512/533, loss: 0.00531616248190403 2023-01-21 12:40:45.338962: step: 1516/533, loss: 0.007601452060043812 2023-01-21 12:40:46.492502: step: 1520/533, loss: 0.00654182443395257 2023-01-21 12:40:47.689992: step: 1524/533, loss: 0.0023254393599927425 2023-01-21 12:40:48.900512: step: 1528/533, loss: 0.010938835330307484 2023-01-21 12:40:50.075225: step: 1532/533, loss: 0.007700061891227961 2023-01-21 12:40:51.286919: step: 1536/533, loss: 0.0013273239601403475 2023-01-21 12:40:52.442532: step: 1540/533, loss: 3.700256274896674e-05 2023-01-21 12:40:53.649878: step: 1544/533, loss: 2.384185791015625e-06 2023-01-21 12:40:54.841612: step: 1548/533, loss: 0.00159454345703125 2023-01-21 12:40:56.005813: step: 1552/533, loss: 0.0012510298984125257 2023-01-21 12:40:57.203183: step: 1556/533, loss: 0.014195061288774014 2023-01-21 12:40:58.360622: step: 1560/533, loss: 0.0009794235229492188 2023-01-21 12:40:59.497800: step: 1564/533, loss: 0.0018888532649725676 2023-01-21 12:41:00.690400: step: 1568/533, loss: 0.00152759556658566 2023-01-21 12:41:01.850312: step: 1572/533, loss: 0.0032791136763989925 2023-01-21 12:41:02.998052: step: 1576/533, loss: 6.828307959949598e-05 2023-01-21 12:41:04.159487: step: 1580/533, loss: 0.0016378402942791581 2023-01-21 12:41:05.361687: step: 1584/533, loss: 0.008200645446777344 2023-01-21 12:41:06.531797: step: 1588/533, loss: -5.817413239128655e-06 2023-01-21 12:41:07.696088: step: 1592/533, loss: 0.009557152166962624 2023-01-21 12:41:08.885638: step: 1596/533, loss: 0.011639404110610485 2023-01-21 12:41:10.068187: step: 1600/533, loss: 0.0003681182861328125 2023-01-21 12:41:11.222872: step: 1604/533, loss: 6.4849853515625e-05 2023-01-21 12:41:12.419338: step: 1608/533, loss: 0.000152587890625 2023-01-21 12:41:13.614849: step: 1612/533, loss: 0.0019096374744549394 2023-01-21 12:41:14.826983: step: 1616/533, loss: 4.596710277837701e-05 2023-01-21 12:41:15.989563: step: 1620/533, loss: 5.588531348621473e-05 2023-01-21 12:41:17.163567: step: 1624/533, loss: 0.04605436325073242 2023-01-21 12:41:18.358103: step: 1628/533, loss: 0.00020551681518554688 2023-01-21 12:41:19.533078: step: 1632/533, loss: 9.098053124034777e-05 2023-01-21 12:41:20.662945: step: 1636/533, loss: 0.00812215730547905 2023-01-21 12:41:21.807732: step: 1640/533, loss: 5.435943603515625e-05 2023-01-21 12:41:22.991594: step: 1644/533, loss: 0.023671245202422142 2023-01-21 12:41:24.160017: step: 1648/533, loss: 0.0004153251647949219 2023-01-21 12:41:25.332190: step: 1652/533, loss: 7.114410254871473e-05 2023-01-21 12:41:26.452162: step: 1656/533, loss: 0.0008428573491983116 2023-01-21 12:41:27.682729: step: 1660/533, loss: 0.00035705568734556437 2023-01-21 12:41:28.869230: step: 1664/533, loss: 0.004678058438003063 2023-01-21 12:41:30.008561: step: 1668/533, loss: 0.0011800766224041581 2023-01-21 12:41:31.238997: step: 1672/533, loss: 0.02404804155230522 2023-01-21 12:41:32.404305: step: 1676/533, loss: 0.0011755942832678556 2023-01-21 12:41:33.603364: step: 1680/533, loss: 0.0001852989080362022 2023-01-21 12:41:34.790320: step: 1684/533, loss: 0.0021029473282396793 2023-01-21 12:41:35.970224: step: 1688/533, loss: 0.01467285118997097 2023-01-21 12:41:37.147634: step: 1692/533, loss: 2.918243444582913e-05 2023-01-21 12:41:38.341758: step: 1696/533, loss: 6.723404112563003e-06 2023-01-21 12:41:39.530508: step: 1700/533, loss: 0.019610309973359108 2023-01-21 12:41:40.683106: step: 1704/533, loss: 0.04709291458129883 2023-01-21 12:41:41.841492: step: 1708/533, loss: 0.02316885255277157 2023-01-21 12:41:43.025170: step: 1712/533, loss: 4.019737389171496e-05 2023-01-21 12:41:44.184652: step: 1716/533, loss: 8.20159948489163e-06 2023-01-21 12:41:45.356596: step: 1720/533, loss: 0.0009641647338867188 2023-01-21 12:41:46.569146: step: 1724/533, loss: 0.07786465436220169 2023-01-21 12:41:47.742328: step: 1728/533, loss: 0.0015408992767333984 2023-01-21 12:41:48.931108: step: 1732/533, loss: 0.013975907117128372 2023-01-21 12:41:50.079150: step: 1736/533, loss: 0.001447486924007535 2023-01-21 12:41:51.256765: step: 1740/533, loss: 2.7084352041129023e-05 2023-01-21 12:41:52.400239: step: 1744/533, loss: -4.57763690064894e-06 2023-01-21 12:41:53.557750: step: 1748/533, loss: -4.482269105210435e-06 2023-01-21 12:41:54.719131: step: 1752/533, loss: 0.011700249277055264 2023-01-21 12:41:55.890306: step: 1756/533, loss: 0.002574157901108265 2023-01-21 12:41:57.074011: step: 1760/533, loss: 0.024260330945253372 2023-01-21 12:41:58.277092: step: 1764/533, loss: 0.0008701324695721269 2023-01-21 12:41:59.467555: step: 1768/533, loss: 0.007428646087646484 2023-01-21 12:42:00.618121: step: 1772/533, loss: 0.055863380432128906 2023-01-21 12:42:01.787478: step: 1776/533, loss: 0.02773122861981392 2023-01-21 12:42:02.911934: step: 1780/533, loss: 0.0001354217529296875 2023-01-21 12:42:04.097438: step: 1784/533, loss: 0.05065784603357315 2023-01-21 12:42:05.303110: step: 1788/533, loss: -7.4386593951203395e-06 2023-01-21 12:42:06.485902: step: 1792/533, loss: 0.0011419296497479081 2023-01-21 12:42:07.681220: step: 1796/533, loss: 0.0008434295887127519 2023-01-21 12:42:08.865969: step: 1800/533, loss: 0.014854812994599342 2023-01-21 12:42:10.013626: step: 1804/533, loss: 0.024970054626464844 2023-01-21 12:42:11.213111: step: 1808/533, loss: 0.0001202583298436366 2023-01-21 12:42:12.353263: step: 1812/533, loss: -4.00543194700731e-06 2023-01-21 12:42:13.579262: step: 1816/533, loss: 8.320808410644531e-05 2023-01-21 12:42:14.751418: step: 1820/533, loss: 0.007039832882583141 2023-01-21 12:42:15.928820: step: 1824/533, loss: 1.8358230590820312e-05 2023-01-21 12:42:17.092531: step: 1828/533, loss: 5.340576535672881e-06 2023-01-21 12:42:18.320604: step: 1832/533, loss: 0.00851221103221178 2023-01-21 12:42:19.535881: step: 1836/533, loss: 0.013712120242416859 2023-01-21 12:42:20.726742: step: 1840/533, loss: 5.664825584972277e-05 2023-01-21 12:42:21.904449: step: 1844/533, loss: 0.024763107299804688 2023-01-21 12:42:23.076617: step: 1848/533, loss: 0.0003149986150674522 2023-01-21 12:42:24.269823: step: 1852/533, loss: 2.0980836779926904e-05 2023-01-21 12:42:25.481108: step: 1856/533, loss: 0.00046830179053358734 2023-01-21 12:42:26.678910: step: 1860/533, loss: 1.6927719116210938e-05 2023-01-21 12:42:27.860035: step: 1864/533, loss: 0.017177963629364967 2023-01-21 12:42:29.073081: step: 1868/533, loss: 0.0006961823091842234 2023-01-21 12:42:30.324237: step: 1872/533, loss: 0.0007681846618652344 2023-01-21 12:42:31.510060: step: 1876/533, loss: 0.0012486458290368319 2023-01-21 12:42:32.742941: step: 1880/533, loss: 0.0022342682350426912 2023-01-21 12:42:33.937835: step: 1884/533, loss: 0.029091835021972656 2023-01-21 12:42:35.102395: step: 1888/533, loss: 0.0012881278526037931 2023-01-21 12:42:36.236464: step: 1892/533, loss: 0.020928192883729935 2023-01-21 12:42:37.449573: step: 1896/533, loss: 0.007745838258415461 2023-01-21 12:42:38.644727: step: 1900/533, loss: 0.0026636123657226562 2023-01-21 12:42:39.843765: step: 1904/533, loss: 0.02485194243490696 2023-01-21 12:42:41.053208: step: 1908/533, loss: -4.00543194700731e-06 2023-01-21 12:42:42.204990: step: 1912/533, loss: 0.003954124636948109 2023-01-21 12:42:43.377474: step: 1916/533, loss: 5.607604907709174e-05 2023-01-21 12:42:44.629337: step: 1920/533, loss: 1.0204315003647935e-05 2023-01-21 12:42:45.798293: step: 1924/533, loss: 0.0020041943062096834 2023-01-21 12:42:46.983455: step: 1928/533, loss: 0.00025773048400878906 2023-01-21 12:42:48.131521: step: 1932/533, loss: 0.02681426890194416 2023-01-21 12:42:49.308788: step: 1936/533, loss: 0.016598891466856003 2023-01-21 12:42:50.462958: step: 1940/533, loss: 0.01302194595336914 2023-01-21 12:42:51.641385: step: 1944/533, loss: 0.025763513520359993 2023-01-21 12:42:52.795776: step: 1948/533, loss: 0.02256469801068306 2023-01-21 12:42:53.972204: step: 1952/533, loss: 0.004913234617561102 2023-01-21 12:42:55.144927: step: 1956/533, loss: 0.0017461776733398438 2023-01-21 12:42:56.306391: step: 1960/533, loss: 0.039456557482481 2023-01-21 12:42:57.509107: step: 1964/533, loss: 0.009808349423110485 2023-01-21 12:42:58.648900: step: 1968/533, loss: 0.004604053683578968 2023-01-21 12:42:59.823537: step: 1972/533, loss: 0.019207384437322617 2023-01-21 12:43:01.008626: step: 1976/533, loss: 0.00277194962836802 2023-01-21 12:43:02.181170: step: 1980/533, loss: -2.0885467165498994e-05 2023-01-21 12:43:03.357434: step: 1984/533, loss: 1.4019012269272935e-05 2023-01-21 12:43:04.557532: step: 1988/533, loss: 6.46591215627268e-05 2023-01-21 12:43:05.723620: step: 1992/533, loss: 0.006814527790993452 2023-01-21 12:43:06.911750: step: 1996/533, loss: 0.021334027871489525 2023-01-21 12:43:08.076456: step: 2000/533, loss: 0.009670543484389782 2023-01-21 12:43:09.237084: step: 2004/533, loss: 0.002830171724781394 2023-01-21 12:43:10.449241: step: 2008/533, loss: 0.01386880874633789 2023-01-21 12:43:11.619134: step: 2012/533, loss: 0.0037967683747410774 2023-01-21 12:43:12.815494: step: 2016/533, loss: 0.006990909576416016 2023-01-21 12:43:14.001976: step: 2020/533, loss: 0.0005186080816201866 2023-01-21 12:43:15.187338: step: 2024/533, loss: 0.00025320128770545125 2023-01-21 12:43:16.395306: step: 2028/533, loss: 0.00026721955509856343 2023-01-21 12:43:17.587024: step: 2032/533, loss: 0.00011672973778331652 2023-01-21 12:43:18.752497: step: 2036/533, loss: 0.006080150604248047 2023-01-21 12:43:19.932588: step: 2040/533, loss: 0.7187438011169434 2023-01-21 12:43:21.123012: step: 2044/533, loss: 0.00107240688521415 2023-01-21 12:43:22.276221: step: 2048/533, loss: 0.0003261566162109375 2023-01-21 12:43:23.448827: step: 2052/533, loss: 0.0005474090576171875 2023-01-21 12:43:24.643419: step: 2056/533, loss: 7.476806786144152e-05 2023-01-21 12:43:25.856782: step: 2060/533, loss: 0.016642285510897636 2023-01-21 12:43:27.036268: step: 2064/533, loss: 0.0021890641655772924 2023-01-21 12:43:28.244361: step: 2068/533, loss: 0.0004442215140443295 2023-01-21 12:43:29.383668: step: 2072/533, loss: 0.00383586878888309 2023-01-21 12:43:30.589268: step: 2076/533, loss: 0.003705692244693637 2023-01-21 12:43:31.771144: step: 2080/533, loss: 0.0005316734313964844 2023-01-21 12:43:32.980470: step: 2084/533, loss: 0.002096223644912243 2023-01-21 12:43:34.201896: step: 2088/533, loss: 0.0032352448906749487 2023-01-21 12:43:35.368374: step: 2092/533, loss: 0.008084392175078392 2023-01-21 12:43:36.575462: step: 2096/533, loss: 0.0405704490840435 2023-01-21 12:43:37.774951: step: 2100/533, loss: 0.029516983777284622 2023-01-21 12:43:38.954801: step: 2104/533, loss: 0.019709492102265358 2023-01-21 12:43:40.120666: step: 2108/533, loss: 0.00034705401048995554 2023-01-21 12:43:41.290024: step: 2112/533, loss: 0.02143106609582901 2023-01-21 12:43:42.519904: step: 2116/533, loss: 0.08330383151769638 2023-01-21 12:43:43.742204: step: 2120/533, loss: 0.00034933091956190765 2023-01-21 12:43:44.948933: step: 2124/533, loss: 6.752014451194555e-05 2023-01-21 12:43:46.138014: step: 2128/533, loss: 0.0017266273498535156 2023-01-21 12:43:47.309489: step: 2132/533, loss: 0.0057968138717114925 ================================================== Loss: 0.028 -------------------- Dev: {'event': {'p': 0.6149789029535865, 'r': 0.7762982689747004, 'f1': 0.6862860506180106}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Test: {'event': {'p': 0.6622760800842993, 'r': 0.8037084398976982, 'f1': 0.7261698440207972}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Chinese: {'event': {'p': 0.5662650602409639, 'r': 0.8703703703703703, 'f1': 0.6861313868613138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Korean: {'event': {'p': 0.6851851851851852, 'r': 0.5873015873015873, 'f1': 0.6324786324786325}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} 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': 22} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:44:28.390258: step: 4/533, loss: 4.367828296381049e-05 2023-01-21 12:44:29.541706: step: 8/533, loss: 0.004833698738366365 2023-01-21 12:44:30.735523: step: 12/533, loss: -1.5258788153005298e-06 2023-01-21 12:44:31.913914: step: 16/533, loss: 0.0010787963401526213 2023-01-21 12:44:33.099047: step: 20/533, loss: 0.01408369466662407 2023-01-21 12:44:34.303748: step: 24/533, loss: 0.5773662328720093 2023-01-21 12:44:35.521995: step: 28/533, loss: 0.0076806070283055305 2023-01-21 12:44:36.722884: step: 32/533, loss: 0.00199718470685184 2023-01-21 12:44:37.900545: step: 36/533, loss: 0.010397911071777344 2023-01-21 12:44:39.074803: step: 40/533, loss: 0.00027551650418899953 2023-01-21 12:44:40.279219: step: 44/533, loss: 0.011960124596953392 2023-01-21 12:44:41.467656: step: 48/533, loss: 0.0021364688873291016 2023-01-21 12:44:42.635532: step: 52/533, loss: 0.0010891437996178865 2023-01-21 12:44:43.824883: step: 56/533, loss: 0.00436820974573493 2023-01-21 12:44:45.013846: step: 60/533, loss: 0.0032970430329442024 2023-01-21 12:44:46.153808: step: 64/533, loss: 0.001438140869140625 2023-01-21 12:44:47.298405: step: 68/533, loss: 1.3685226804227568e-05 2023-01-21 12:44:48.449907: step: 72/533, loss: -4.00543194700731e-06 2023-01-21 12:44:49.642364: step: 76/533, loss: 0.018964387476444244 2023-01-21 12:44:50.798075: step: 80/533, loss: 0.013823652639985085 2023-01-21 12:44:51.950983: step: 84/533, loss: 0.000173473366885446 2023-01-21 12:44:53.118122: step: 88/533, loss: 0.11178937554359436 2023-01-21 12:44:54.265747: step: 92/533, loss: 0.002576160477474332 2023-01-21 12:44:55.489874: step: 96/533, loss: 0.0018013000953942537 2023-01-21 12:44:56.647271: step: 100/533, loss: -9.5367431640625e-07 2023-01-21 12:44:57.801301: step: 104/533, loss: 2.1743775505456142e-05 2023-01-21 12:44:59.000302: step: 108/533, loss: 0.0002487182500772178 2023-01-21 12:45:00.233084: step: 112/533, loss: 0.003396320389583707 2023-01-21 12:45:01.430045: step: 116/533, loss: 0.0013130188453942537 2023-01-21 12:45:02.638818: step: 120/533, loss: 0.002260589739307761 2023-01-21 12:45:03.770669: step: 124/533, loss: 0.00011034011549782008 2023-01-21 12:45:04.966358: step: 128/533, loss: 0.00017328262038063258 2023-01-21 12:45:06.124047: step: 132/533, loss: 0.022417353466153145 2023-01-21 12:45:07.263738: step: 136/533, loss: 0.00011119843111373484 2023-01-21 12:45:08.483243: step: 140/533, loss: 0.0045659542083740234 2023-01-21 12:45:09.628324: step: 144/533, loss: 0.012973308563232422 2023-01-21 12:45:10.805042: step: 148/533, loss: -1.697540210443549e-05 2023-01-21 12:45:11.978624: step: 152/533, loss: 0.0008994102827273309 2023-01-21 12:45:13.119485: step: 156/533, loss: 0.0024919509887695312 2023-01-21 12:45:14.261978: step: 160/533, loss: 0.0009292602771893144 2023-01-21 12:45:15.426796: step: 164/533, loss: 0.00080108642578125 2023-01-21 12:45:16.603641: step: 168/533, loss: 0.003117084503173828 2023-01-21 12:45:17.759888: step: 172/533, loss: 0.015945052728056908 2023-01-21 12:45:18.974281: step: 176/533, loss: 2.250671423098538e-05 2023-01-21 12:45:20.136436: step: 180/533, loss: 9.150505502475426e-05 2023-01-21 12:45:21.347558: step: 184/533, loss: 7.295608611457283e-06 2023-01-21 12:45:22.541750: step: 188/533, loss: 0.049378011375665665 2023-01-21 12:45:23.698281: step: 192/533, loss: 0.04559612274169922 2023-01-21 12:45:24.868625: step: 196/533, loss: 0.0034296035300940275 2023-01-21 12:45:26.040493: step: 200/533, loss: 0.0013187408912926912 2023-01-21 12:45:27.212449: step: 204/533, loss: 0.0005696296575479209 2023-01-21 12:45:28.368856: step: 208/533, loss: 0.01294555701315403 2023-01-21 12:45:29.607514: step: 212/533, loss: 0.0030471801292151213 2023-01-21 12:45:30.804106: step: 216/533, loss: 0.015167427249252796 2023-01-21 12:45:31.973784: step: 220/533, loss: 0.0001539230433991179 2023-01-21 12:45:33.133189: step: 224/533, loss: 0.0017995834350585938 2023-01-21 12:45:34.312195: step: 228/533, loss: 0.026882745325565338 2023-01-21 12:45:35.452061: step: 232/533, loss: 0.0005903244018554688 2023-01-21 12:45:36.593944: step: 236/533, loss: 5.33103957423009e-05 2023-01-21 12:45:37.760942: step: 240/533, loss: 0.0048084259033203125 2023-01-21 12:45:39.022981: step: 244/533, loss: 0.006752967834472656 2023-01-21 12:45:40.161197: step: 248/533, loss: 0.01136703509837389 2023-01-21 12:45:41.320956: step: 252/533, loss: 0.0007987976423464715 2023-01-21 12:45:42.482442: step: 256/533, loss: 0.0003628730773925781 2023-01-21 12:45:43.646584: step: 260/533, loss: 0.0075005535036325455 2023-01-21 12:45:44.792367: step: 264/533, loss: 0.01676158793270588 2023-01-21 12:45:45.902605: step: 268/533, loss: 0.5639263987541199 2023-01-21 12:45:47.053664: step: 272/533, loss: 0.007390069775283337 2023-01-21 12:45:48.245925: step: 276/533, loss: 0.0006742477999068797 2023-01-21 12:45:49.470170: step: 280/533, loss: 3.223419116693549e-05 2023-01-21 12:45:50.675152: step: 284/533, loss: 0.0007882118807174265 2023-01-21 12:45:51.810369: step: 288/533, loss: 0.02160205878317356 2023-01-21 12:45:53.031630: step: 292/533, loss: 0.0036558150313794613 2023-01-21 12:45:54.241827: step: 296/533, loss: 0.0003621101495809853 2023-01-21 12:45:55.389729: step: 300/533, loss: 0.017501259222626686 2023-01-21 12:45:56.547408: step: 304/533, loss: 0.008803939446806908 2023-01-21 12:45:57.710594: step: 308/533, loss: 0.0008853912586346269 2023-01-21 12:45:58.890610: step: 312/533, loss: 3.204345557605848e-05 2023-01-21 12:46:00.122750: step: 316/533, loss: 0.0013242721324786544 2023-01-21 12:46:01.261830: step: 320/533, loss: 0.01422338467091322 2023-01-21 12:46:02.462374: step: 324/533, loss: 0.007374000269919634 2023-01-21 12:46:03.677072: step: 328/533, loss: 0.02530651167035103 2023-01-21 12:46:04.846177: step: 332/533, loss: 0.0005330085987225175 2023-01-21 12:46:05.993640: step: 336/533, loss: 0.005850601475685835 2023-01-21 12:46:07.149795: step: 340/533, loss: 0.006619072053581476 2023-01-21 12:46:08.383683: step: 344/533, loss: 0.0011375427711755037 2023-01-21 12:46:09.567016: step: 348/533, loss: 0.004350757692009211 2023-01-21 12:46:10.737225: step: 352/533, loss: 0.013763475231826305 2023-01-21 12:46:11.866370: step: 356/533, loss: 3.1757354008732364e-05 2023-01-21 12:46:13.044829: step: 360/533, loss: 0.01386342104524374 2023-01-21 12:46:14.222914: step: 364/533, loss: 0.010579300113022327 2023-01-21 12:46:15.406295: step: 368/533, loss: 0.0001661300630075857 2023-01-21 12:46:16.547080: step: 372/533, loss: 0.0073909759521484375 2023-01-21 12:46:17.702690: step: 376/533, loss: 0.014061165042221546 2023-01-21 12:46:18.846037: step: 380/533, loss: 3.452301098150201e-05 2023-01-21 12:46:20.019932: step: 384/533, loss: 4.95910626341356e-06 2023-01-21 12:46:21.269021: step: 388/533, loss: 0.006706476677209139 2023-01-21 12:46:22.431257: step: 392/533, loss: 0.00013456345186568797 2023-01-21 12:46:23.584887: step: 396/533, loss: 0.0027136802673339844 2023-01-21 12:46:24.717770: step: 400/533, loss: 0.0012274743057787418 2023-01-21 12:46:25.877517: step: 404/533, loss: 0.0027151107788085938 2023-01-21 12:46:27.085159: step: 408/533, loss: 0.0020917891524732113 2023-01-21 12:46:28.285572: step: 412/533, loss: 0.0005847931024618447 2023-01-21 12:46:29.484059: step: 416/533, loss: 0.0001483917294535786 2023-01-21 12:46:30.640229: step: 420/533, loss: 0.0032260895241051912 2023-01-21 12:46:31.829791: step: 424/533, loss: 0.008239937014877796 2023-01-21 12:46:33.032429: step: 428/533, loss: 0.0019099235069006681 2023-01-21 12:46:34.195409: step: 432/533, loss: 8.049011375987902e-05 2023-01-21 12:46:35.370581: step: 436/533, loss: 0.00030984877957962453 2023-01-21 12:46:36.502991: step: 440/533, loss: 0.0001843452628236264 2023-01-21 12:46:37.664813: step: 444/533, loss: 2.2602082026423886e-05 2023-01-21 12:46:38.838362: step: 448/533, loss: 5.917549424339086e-05 2023-01-21 12:46:40.040043: step: 452/533, loss: 0.0009412765502929688 2023-01-21 12:46:41.223801: step: 456/533, loss: 0.0033445358276367188 2023-01-21 12:46:42.432918: step: 460/533, loss: 0.020433617755770683 2023-01-21 12:46:43.632811: step: 464/533, loss: 0.004730605985969305 2023-01-21 12:46:44.811339: step: 468/533, loss: 0.0006296157953329384 2023-01-21 12:46:45.954603: step: 472/533, loss: 0.00010242462303722277 2023-01-21 12:46:47.104397: step: 476/533, loss: 0.030965041369199753 2023-01-21 12:46:48.276999: step: 480/533, loss: 0.0010493517620489001 2023-01-21 12:46:49.484155: step: 484/533, loss: 0.0013270378112792969 2023-01-21 12:46:50.644514: step: 488/533, loss: 0.00020847321138717234 2023-01-21 12:46:51.813741: step: 492/533, loss: 0.0003181457577738911 2023-01-21 12:46:53.020285: step: 496/533, loss: 6.771087555534905e-06 2023-01-21 12:46:54.172031: step: 500/533, loss: 0.0013645172584801912 2023-01-21 12:46:55.347456: step: 504/533, loss: 0.02997570112347603 2023-01-21 12:46:56.519223: step: 508/533, loss: 0.028525162488222122 2023-01-21 12:46:57.696049: step: 512/533, loss: 3.237724376958795e-05 2023-01-21 12:46:58.872974: step: 516/533, loss: 7.62939453125e-06 2023-01-21 12:47:00.078493: step: 520/533, loss: 0.7232208251953125 2023-01-21 12:47:01.281802: step: 524/533, loss: 0.14858703315258026 2023-01-21 12:47:02.440456: step: 528/533, loss: 0.03774090111255646 2023-01-21 12:47:03.638402: step: 532/533, loss: 0.0007862091297283769 2023-01-21 12:47:04.815906: step: 536/533, loss: 0.05137762799859047 2023-01-21 12:47:05.959045: step: 540/533, loss: 0.0007997513166628778 2023-01-21 12:47:07.154998: step: 544/533, loss: 0.026436423882842064 2023-01-21 12:47:08.328939: step: 548/533, loss: 0.0015860558487474918 2023-01-21 12:47:09.531296: step: 552/533, loss: 0.038579367101192474 2023-01-21 12:47:10.688854: step: 556/533, loss: 0.00045642853365279734 2023-01-21 12:47:11.870276: step: 560/533, loss: 0.003823947859928012 2023-01-21 12:47:13.060635: step: 564/533, loss: 0.00037822723970748484 2023-01-21 12:47:14.216782: step: 568/533, loss: 0.1228710189461708 2023-01-21 12:47:15.377836: step: 572/533, loss: 0.061507608741521835 2023-01-21 12:47:16.534046: step: 576/533, loss: 0.14891338348388672 2023-01-21 12:47:17.687432: step: 580/533, loss: 8.37326078908518e-05 2023-01-21 12:47:18.901858: step: 584/533, loss: 0.016631508246064186 2023-01-21 12:47:20.076716: step: 588/533, loss: 0.01926266774535179 2023-01-21 12:47:21.237567: step: 592/533, loss: 0.004851913545280695 2023-01-21 12:47:22.432963: step: 596/533, loss: 0.06506597995758057 2023-01-21 12:47:23.611424: step: 600/533, loss: 0.0013857842423021793 2023-01-21 12:47:24.766401: step: 604/533, loss: 0.051195334643125534 2023-01-21 12:47:25.958856: step: 608/533, loss: 0.00739550543949008 2023-01-21 12:47:27.160528: step: 612/533, loss: 0.13472633063793182 2023-01-21 12:47:28.357291: step: 616/533, loss: 0.014884376898407936 2023-01-21 12:47:29.508311: step: 620/533, loss: 0.027770044282078743 2023-01-21 12:47:30.674253: step: 624/533, loss: 0.008984469808638096 2023-01-21 12:47:31.837851: step: 628/533, loss: 0.0031900405883789062 2023-01-21 12:47:33.001317: step: 632/533, loss: 0.0038488388527184725 2023-01-21 12:47:34.220301: step: 636/533, loss: 0.06429142504930496 2023-01-21 12:47:35.418694: step: 640/533, loss: 0.024149131029844284 2023-01-21 12:47:36.581536: step: 644/533, loss: 5.14984139954322e-06 2023-01-21 12:47:37.769691: step: 648/533, loss: 0.010907459072768688 2023-01-21 12:47:38.971132: step: 652/533, loss: 0.00024356841458939016 2023-01-21 12:47:40.130191: step: 656/533, loss: 0.00762863177806139 2023-01-21 12:47:41.310775: step: 660/533, loss: 4.272460864740424e-05 2023-01-21 12:47:42.494534: step: 664/533, loss: 0.0015058518620207906 2023-01-21 12:47:43.738364: step: 668/533, loss: 1.3703391551971436 2023-01-21 12:47:44.910539: step: 672/533, loss: 0.002596283098682761 2023-01-21 12:47:46.092077: step: 676/533, loss: 0.0028486251831054688 2023-01-21 12:47:47.284211: step: 680/533, loss: 0.03347458690404892 2023-01-21 12:47:48.467014: step: 684/533, loss: 0.006866455078125 2023-01-21 12:47:49.629378: step: 688/533, loss: 0.012720465660095215 2023-01-21 12:47:50.817316: step: 692/533, loss: 0.008841896429657936 2023-01-21 12:47:52.039949: step: 696/533, loss: 0.001014661742374301 2023-01-21 12:47:53.205199: step: 700/533, loss: 0.0015799523098394275 2023-01-21 12:47:54.394769: step: 704/533, loss: 0.00843434315174818 2023-01-21 12:47:55.540697: step: 708/533, loss: 0.02506399154663086 2023-01-21 12:47:56.737516: step: 712/533, loss: 0.00012054442777298391 2023-01-21 12:47:57.908529: step: 716/533, loss: -1.7166130419354886e-06 2023-01-21 12:47:59.072332: step: 720/533, loss: 0.0081055648624897 2023-01-21 12:48:00.190847: step: 724/533, loss: 0.004156970884650946 2023-01-21 12:48:01.375495: step: 728/533, loss: 0.01972188986837864 2023-01-21 12:48:02.559164: step: 732/533, loss: 0.025399019941687584 2023-01-21 12:48:03.722514: step: 736/533, loss: 0.01169891282916069 2023-01-21 12:48:04.913891: step: 740/533, loss: 0.042191121727228165 2023-01-21 12:48:06.056097: step: 744/533, loss: 0.0007309913635253906 2023-01-21 12:48:07.265353: step: 748/533, loss: 0.0024110793601721525 2023-01-21 12:48:08.461440: step: 752/533, loss: 0.015259361825883389 2023-01-21 12:48:09.604772: step: 756/533, loss: 0.002070522168651223 2023-01-21 12:48:10.750765: step: 760/533, loss: 0.008665180765092373 2023-01-21 12:48:11.939842: step: 764/533, loss: 0.0001693725644145161 2023-01-21 12:48:13.098096: step: 768/533, loss: 0.02346820756793022 2023-01-21 12:48:14.263520: step: 772/533, loss: 0.0014692307449877262 2023-01-21 12:48:15.434866: step: 776/533, loss: 0.0006662369123660028 2023-01-21 12:48:16.589347: step: 780/533, loss: 0.0003328800085000694 2023-01-21 12:48:17.729074: step: 784/533, loss: 0.0017113685607910156 2023-01-21 12:48:18.907003: step: 788/533, loss: 0.009048175998032093 2023-01-21 12:48:20.073349: step: 792/533, loss: 0.00018091201491188258 2023-01-21 12:48:21.265543: step: 796/533, loss: 0.0008546829340048134 2023-01-21 12:48:22.412658: step: 800/533, loss: 0.005116653628647327 2023-01-21 12:48:23.607565: step: 804/533, loss: 0.011718940921127796 2023-01-21 12:48:24.803934: step: 808/533, loss: 0.0022395611740648746 2023-01-21 12:48:26.006038: step: 812/533, loss: 0.00012645722017623484 2023-01-21 12:48:27.200740: step: 816/533, loss: 0.0007635116344317794 2023-01-21 12:48:28.402434: step: 820/533, loss: 0.011646557599306107 2023-01-21 12:48:29.583990: step: 824/533, loss: 0.00012969970703125 2023-01-21 12:48:30.783452: step: 828/533, loss: 0.032987214624881744 2023-01-21 12:48:31.982804: step: 832/533, loss: 0.0005928516620770097 2023-01-21 12:48:33.189606: step: 836/533, loss: 0.021815728396177292 2023-01-21 12:48:34.337802: step: 840/533, loss: 0.00014200211444403976 2023-01-21 12:48:35.485725: step: 844/533, loss: 0.0036428452003747225 2023-01-21 12:48:36.656496: step: 848/533, loss: 0.0002111435023834929 2023-01-21 12:48:37.891845: step: 852/533, loss: 0.03106403537094593 2023-01-21 12:48:39.080254: step: 856/533, loss: 0.0004092007875442505 2023-01-21 12:48:40.257475: step: 860/533, loss: 0.015527820214629173 2023-01-21 12:48:41.438580: step: 864/533, loss: 0.021834183484315872 2023-01-21 12:48:42.573833: step: 868/533, loss: 0.005451297853142023 2023-01-21 12:48:43.726564: step: 872/533, loss: 0.0022825240157544613 2023-01-21 12:48:44.892806: step: 876/533, loss: 0.0010254860389977694 2023-01-21 12:48:46.106984: step: 880/533, loss: 0.004307175055146217 2023-01-21 12:48:47.355960: step: 884/533, loss: 0.030779076740145683 2023-01-21 12:48:48.549719: step: 888/533, loss: 5.7888031733455136e-05 2023-01-21 12:48:49.707959: step: 892/533, loss: 0.00014953613572288305 2023-01-21 12:48:50.926911: step: 896/533, loss: 0.0017480850219726562 2023-01-21 12:48:52.112457: step: 900/533, loss: 0.016500188037753105 2023-01-21 12:48:53.280299: step: 904/533, loss: 0.0003707885625772178 2023-01-21 12:48:54.468857: step: 908/533, loss: 0.025084495544433594 2023-01-21 12:48:55.651543: step: 912/533, loss: 0.00115375523455441 2023-01-21 12:48:56.791843: step: 916/533, loss: 0.03572063520550728 2023-01-21 12:48:57.935366: step: 920/533, loss: 4.997253563487902e-05 2023-01-21 12:48:59.091322: step: 924/533, loss: 0.0014259337913244963 2023-01-21 12:49:00.266595: step: 928/533, loss: 0.0014012337196618319 2023-01-21 12:49:01.419565: step: 932/533, loss: 0.05708455666899681 2023-01-21 12:49:02.657223: step: 936/533, loss: 0.020565509796142578 2023-01-21 12:49:03.861037: step: 940/533, loss: 0.24678516387939453 2023-01-21 12:49:05.051107: step: 944/533, loss: 0.0027359009254723787 2023-01-21 12:49:06.205966: step: 948/533, loss: 0.0003315925714559853 2023-01-21 12:49:07.372303: step: 952/533, loss: 0.00045986176701262593 2023-01-21 12:49:08.533878: step: 956/533, loss: 0.0006501436000689864 2023-01-21 12:49:09.730209: step: 960/533, loss: 2.441406286379788e-05 2023-01-21 12:49:10.934650: step: 964/533, loss: 0.0015298009384423494 2023-01-21 12:49:12.060776: step: 968/533, loss: 3.24249276673072e-06 2023-01-21 12:49:13.232762: step: 972/533, loss: 0.21640262007713318 2023-01-21 12:49:14.431261: step: 976/533, loss: 0.036772727966308594 2023-01-21 12:49:15.621362: step: 980/533, loss: 0.06259270012378693 2023-01-21 12:49:16.795711: step: 984/533, loss: 6.637573096668348e-05 2023-01-21 12:49:17.977388: step: 988/533, loss: 0.002273750491440296 2023-01-21 12:49:19.149085: step: 992/533, loss: 0.0649871826171875 2023-01-21 12:49:20.300247: step: 996/533, loss: 0.0004590034659486264 2023-01-21 12:49:21.540746: step: 1000/533, loss: 0.007580948062241077 2023-01-21 12:49:22.672808: step: 1004/533, loss: 0.00927047710865736 2023-01-21 12:49:23.842408: step: 1008/533, loss: 0.05645904690027237 2023-01-21 12:49:25.041167: step: 1012/533, loss: 0.01929454877972603 2023-01-21 12:49:26.236602: step: 1016/533, loss: 2.7132035029353574e-05 2023-01-21 12:49:27.402753: step: 1020/533, loss: 0.056095123291015625 2023-01-21 12:49:28.574435: step: 1024/533, loss: 0.0004031181160826236 2023-01-21 12:49:29.754652: step: 1028/533, loss: 0.0003883361932821572 2023-01-21 12:49:30.909452: step: 1032/533, loss: 0.2314375787973404 2023-01-21 12:49:32.099420: step: 1036/533, loss: 0.021697998046875 2023-01-21 12:49:33.312036: step: 1040/533, loss: 0.008711433969438076 2023-01-21 12:49:34.508824: step: 1044/533, loss: 0.0008431434980593622 2023-01-21 12:49:35.658727: step: 1048/533, loss: 0.00032129290048033 2023-01-21 12:49:36.823957: step: 1052/533, loss: 0.0010913849109783769 2023-01-21 12:49:37.967966: step: 1056/533, loss: 0.001083970069885254 2023-01-21 12:49:39.131107: step: 1060/533, loss: 0.2573177218437195 2023-01-21 12:49:40.275352: step: 1064/533, loss: 0.09269677102565765 2023-01-21 12:49:41.457721: step: 1068/533, loss: 0.05773629993200302 2023-01-21 12:49:42.569106: step: 1072/533, loss: 0.0183271411806345 2023-01-21 12:49:43.748622: step: 1076/533, loss: 0.006753826513886452 2023-01-21 12:49:44.935105: step: 1080/533, loss: 0.0029914856422692537 2023-01-21 12:49:46.139171: step: 1084/533, loss: 0.001609754515811801 2023-01-21 12:49:47.296843: step: 1088/533, loss: 0.004450416658073664 2023-01-21 12:49:48.471299: step: 1092/533, loss: 0.243892103433609 2023-01-21 12:49:49.615050: step: 1096/533, loss: 0.0043305871076881886 2023-01-21 12:49:50.754933: step: 1100/533, loss: 2.47955313170678e-06 2023-01-21 12:49:51.922487: step: 1104/533, loss: 0.010229015722870827 2023-01-21 12:49:53.092787: step: 1108/533, loss: 0.010856819339096546 2023-01-21 12:49:54.300827: step: 1112/533, loss: 0.05640153959393501 2023-01-21 12:49:55.446509: step: 1116/533, loss: 0.0008594513055868447 2023-01-21 12:49:56.600874: step: 1120/533, loss: 8.697509474586695e-05 2023-01-21 12:49:57.791164: step: 1124/533, loss: 0.00024261475482489914 2023-01-21 12:49:59.010594: step: 1128/533, loss: 0.0005855560302734375 2023-01-21 12:50:00.130024: step: 1132/533, loss: 0.4725748896598816 2023-01-21 12:50:01.338824: step: 1136/533, loss: 0.0004037380567751825 2023-01-21 12:50:02.541482: step: 1140/533, loss: 0.003464603563770652 2023-01-21 12:50:03.701168: step: 1144/533, loss: 0.00033664703369140625 2023-01-21 12:50:04.905405: step: 1148/533, loss: 0.00016498565673828125 2023-01-21 12:50:06.108168: step: 1152/533, loss: 0.002050781389698386 2023-01-21 12:50:07.268176: step: 1156/533, loss: 0.002957296557724476 2023-01-21 12:50:08.464275: step: 1160/533, loss: 0.0021316527854651213 2023-01-21 12:50:09.644174: step: 1164/533, loss: 0.0026208877097815275 2023-01-21 12:50:10.792885: step: 1168/533, loss: 0.0025114058516919613 2023-01-21 12:50:11.974474: step: 1172/533, loss: 0.00016860962205100805 2023-01-21 12:50:13.117283: step: 1176/533, loss: 0.0015295027988031507 2023-01-21 12:50:14.306397: step: 1180/533, loss: 0.010804367251694202 2023-01-21 12:50:15.510797: step: 1184/533, loss: 0.005790138151496649 2023-01-21 12:50:16.651359: step: 1188/533, loss: 0.05527706444263458 2023-01-21 12:50:17.846601: step: 1192/533, loss: 0.0040986062958836555 2023-01-21 12:50:19.034759: step: 1196/533, loss: 0.4861486554145813 2023-01-21 12:50:20.236054: step: 1200/533, loss: 0.0005045891157351434 2023-01-21 12:50:21.461870: step: 1204/533, loss: 0.00016527176194358617 2023-01-21 12:50:22.641016: step: 1208/533, loss: 0.00028266909066587687 2023-01-21 12:50:23.816927: step: 1212/533, loss: 0.0005316734313964844 2023-01-21 12:50:25.010246: step: 1216/533, loss: 0.003757762722671032 2023-01-21 12:50:26.157908: step: 1220/533, loss: 0.00011310577247058973 2023-01-21 12:50:27.365733: step: 1224/533, loss: 0.496656596660614 2023-01-21 12:50:28.557957: step: 1228/533, loss: 0.02617950551211834 2023-01-21 12:50:29.727928: step: 1232/533, loss: 0.00017681121244095266 2023-01-21 12:50:30.898329: step: 1236/533, loss: 0.0004814147832803428 2023-01-21 12:50:32.119792: step: 1240/533, loss: 0.02547617070376873 2023-01-21 12:50:33.272397: step: 1244/533, loss: 0.009776783175766468 2023-01-21 12:50:34.395777: step: 1248/533, loss: 6.217956251930445e-05 2023-01-21 12:50:35.661509: step: 1252/533, loss: 4.028190612792969 2023-01-21 12:50:36.842976: step: 1256/533, loss: 0.07298431545495987 2023-01-21 12:50:38.051606: step: 1260/533, loss: 0.0029594420921057463 2023-01-21 12:50:39.249009: step: 1264/533, loss: 0.000111627574369777 2023-01-21 12:50:40.426306: step: 1268/533, loss: 0.014233589172363281 2023-01-21 12:50:41.594779: step: 1272/533, loss: 3.4168362617492676e-05 2023-01-21 12:50:42.759061: step: 1276/533, loss: 0.00037889485247433186 2023-01-21 12:50:43.947376: step: 1280/533, loss: 0.057022858411073685 2023-01-21 12:50:45.125448: step: 1284/533, loss: 0.02221984975039959 2023-01-21 12:50:46.291161: step: 1288/533, loss: 0.0013111113803461194 2023-01-21 12:50:47.474394: step: 1292/533, loss: 0.07339143753051758 2023-01-21 12:50:48.624852: step: 1296/533, loss: 0.00025682448176667094 2023-01-21 12:50:49.773310: step: 1300/533, loss: 0.0005340576171875 2023-01-21 12:50:50.952131: step: 1304/533, loss: 0.008294487372040749 2023-01-21 12:50:52.121261: step: 1308/533, loss: 0.00045070648775435984 2023-01-21 12:50:53.297900: step: 1312/533, loss: 9.918212890625e-05 2023-01-21 12:50:54.495522: step: 1316/533, loss: 0.004061460494995117 2023-01-21 12:50:55.694749: step: 1320/533, loss: 0.002850818680599332 2023-01-21 12:50:56.864321: step: 1324/533, loss: 0.021628951653838158 2023-01-21 12:50:58.034393: step: 1328/533, loss: 0.04279213026165962 2023-01-21 12:50:59.169809: step: 1332/533, loss: 0.0003364562871865928 2023-01-21 12:51:00.347098: step: 1336/533, loss: 0.08291897922754288 2023-01-21 12:51:01.503878: step: 1340/533, loss: 3.271103196311742e-05 2023-01-21 12:51:02.683920: step: 1344/533, loss: 4.444122168933973e-05 2023-01-21 12:51:03.810243: step: 1348/533, loss: 0.0006788730970583856 2023-01-21 12:51:04.988385: step: 1352/533, loss: 0.013336753472685814 2023-01-21 12:51:06.157453: step: 1356/533, loss: 0.0570068359375 2023-01-21 12:51:07.295214: step: 1360/533, loss: 3.561973790056072e-05 2023-01-21 12:51:08.529810: step: 1364/533, loss: 0.013704109936952591 2023-01-21 12:51:09.734852: step: 1368/533, loss: 0.0008209228981286287 2023-01-21 12:51:10.912215: step: 1372/533, loss: 0.08327322453260422 2023-01-21 12:51:12.100629: step: 1376/533, loss: 0.0008445740095339715 2023-01-21 12:51:13.264623: step: 1380/533, loss: 0.05320386588573456 2023-01-21 12:51:14.439123: step: 1384/533, loss: 0.002938842633739114 2023-01-21 12:51:15.614393: step: 1388/533, loss: 6.84738188283518e-05 2023-01-21 12:51:16.761341: step: 1392/533, loss: 0.00256004324182868 2023-01-21 12:51:17.927741: step: 1396/533, loss: 3.552436828613281e-05 2023-01-21 12:51:19.087883: step: 1400/533, loss: 0.00030241013155318797 2023-01-21 12:51:20.232928: step: 1404/533, loss: 0.06723956763744354 2023-01-21 12:51:21.392582: step: 1408/533, loss: 0.0002961158752441406 2023-01-21 12:51:22.592463: step: 1412/533, loss: 0.027453135699033737 2023-01-21 12:51:23.759490: step: 1416/533, loss: 0.0010979652870446444 2023-01-21 12:51:24.916607: step: 1420/533, loss: 0.00952377263456583 2023-01-21 12:51:26.077336: step: 1424/533, loss: 0.0015462875599041581 2023-01-21 12:51:27.248061: step: 1428/533, loss: 0.008257769979536533 2023-01-21 12:51:28.429563: step: 1432/533, loss: 9.765625145519152e-05 2023-01-21 12:51:29.631959: step: 1436/533, loss: 8.239746239269152e-05 2023-01-21 12:51:30.764505: step: 1440/533, loss: 0.00243780598975718 2023-01-21 12:51:31.933927: step: 1444/533, loss: 0.022835826501250267 2023-01-21 12:51:33.109235: step: 1448/533, loss: 0.00011386872211005539 2023-01-21 12:51:34.278077: step: 1452/533, loss: 0.0006555557483807206 2023-01-21 12:51:35.413038: step: 1456/533, loss: 0.004465865902602673 2023-01-21 12:51:36.576715: step: 1460/533, loss: 0.001409387681633234 2023-01-21 12:51:37.753055: step: 1464/533, loss: 0.0009399414411745965 2023-01-21 12:51:38.924931: step: 1468/533, loss: 0.0008329391130246222 2023-01-21 12:51:40.080222: step: 1472/533, loss: 0.0006933696568012238 2023-01-21 12:51:41.249683: step: 1476/533, loss: 0.005269050598144531 2023-01-21 12:51:42.411741: step: 1480/533, loss: 0.014586448669433594 2023-01-21 12:51:43.555519: step: 1484/533, loss: 0.0021942139137536287 2023-01-21 12:51:44.715330: step: 1488/533, loss: 1.583099401614163e-05 2023-01-21 12:51:45.884023: step: 1492/533, loss: 0.00037889479426667094 2023-01-21 12:51:47.044916: step: 1496/533, loss: 0.01611328125 2023-01-21 12:51:48.234159: step: 1500/533, loss: 0.00028998852940276265 2023-01-21 12:51:49.407426: step: 1504/533, loss: 0.003301429795101285 2023-01-21 12:51:50.592295: step: 1508/533, loss: 0.0011800765059888363 2023-01-21 12:51:51.733087: step: 1512/533, loss: 0.0052337646484375 2023-01-21 12:51:52.938784: step: 1516/533, loss: 0.004521846771240234 2023-01-21 12:51:54.122766: step: 1520/533, loss: 0.0034716606605798006 2023-01-21 12:51:55.281337: step: 1524/533, loss: 7.21931501175277e-05 2023-01-21 12:51:56.488291: step: 1528/533, loss: 0.0005996703985147178 2023-01-21 12:51:57.670788: step: 1532/533, loss: 0.004365730099380016 2023-01-21 12:51:58.829256: step: 1536/533, loss: 0.0011579514248296618 2023-01-21 12:51:59.989280: step: 1540/533, loss: 4.2629242670955136e-05 2023-01-21 12:52:01.190262: step: 1544/533, loss: 0.006268214900046587 2023-01-21 12:52:02.333602: step: 1548/533, loss: 1.5926361811580136e-05 2023-01-21 12:52:03.485319: step: 1552/533, loss: 0.0029651641380041838 2023-01-21 12:52:04.689597: step: 1556/533, loss: 0.0060555459931492805 2023-01-21 12:52:05.854385: step: 1560/533, loss: 0.012828826904296875 2023-01-21 12:52:07.002972: step: 1564/533, loss: 0.017372703179717064 2023-01-21 12:52:08.188639: step: 1568/533, loss: 0.0002750396670307964 2023-01-21 12:52:09.343513: step: 1572/533, loss: 6.370543997036293e-05 2023-01-21 12:52:10.499799: step: 1576/533, loss: 0.004369258880615234 2023-01-21 12:52:11.674488: step: 1580/533, loss: 0.025567438453435898 2023-01-21 12:52:12.850595: step: 1584/533, loss: 0.009633255191147327 2023-01-21 12:52:14.064538: step: 1588/533, loss: 0.0020591735374182463 2023-01-21 12:52:15.236532: step: 1592/533, loss: 0.011599158868193626 2023-01-21 12:52:16.443601: step: 1596/533, loss: 0.0035119056701660156 2023-01-21 12:52:17.565174: step: 1600/533, loss: 0.00023536682419944555 2023-01-21 12:52:18.809139: step: 1604/533, loss: 0.04906482622027397 2023-01-21 12:52:19.987757: step: 1608/533, loss: 0.00012674331082962453 2023-01-21 12:52:21.203110: step: 1612/533, loss: 0.0005786895635537803 2023-01-21 12:52:22.360867: step: 1616/533, loss: 0.008405876345932484 2023-01-21 12:52:23.541321: step: 1620/533, loss: 0.0001960754452738911 2023-01-21 12:52:24.704677: step: 1624/533, loss: 3.0040740966796875e-05 2023-01-21 12:52:25.893351: step: 1628/533, loss: 0.0006561279878951609 2023-01-21 12:52:27.059029: step: 1632/533, loss: 0.0015196801396086812 2023-01-21 12:52:28.229965: step: 1636/533, loss: 0.04878272861242294 2023-01-21 12:52:29.375773: step: 1640/533, loss: 0.002242469694465399 2023-01-21 12:52:30.554766: step: 1644/533, loss: 1.5163421267061494e-05 2023-01-21 12:52:31.748390: step: 1648/533, loss: 0.05905933305621147 2023-01-21 12:52:32.956153: step: 1652/533, loss: 0.009740972891449928 2023-01-21 12:52:34.135293: step: 1656/533, loss: 0.06339530646800995 2023-01-21 12:52:35.263049: step: 1660/533, loss: 0.023625755682587624 2023-01-21 12:52:36.476415: step: 1664/533, loss: 0.0003610610729083419 2023-01-21 12:52:37.681391: step: 1668/533, loss: 0.004610252566635609 2023-01-21 12:52:38.834771: step: 1672/533, loss: 0.02607612870633602 2023-01-21 12:52:39.984432: step: 1676/533, loss: 0.00010704994929255918 2023-01-21 12:52:41.122369: step: 1680/533, loss: 0.0007863044738769531 2023-01-21 12:52:42.286819: step: 1684/533, loss: 0.0424046516418457 2023-01-21 12:52:43.463738: step: 1688/533, loss: 0.0006410598871298134 2023-01-21 12:52:44.649057: step: 1692/533, loss: 0.001499176025390625 2023-01-21 12:52:45.845110: step: 1696/533, loss: 0.0027582645416259766 2023-01-21 12:52:46.986053: step: 1700/533, loss: 1.7452239262638614e-05 2023-01-21 12:52:48.178173: step: 1704/533, loss: 0.03904452174901962 2023-01-21 12:52:49.327348: step: 1708/533, loss: 0.00035881996154785156 2023-01-21 12:52:50.495137: step: 1712/533, loss: 0.0007788657676428556 2023-01-21 12:52:51.626410: step: 1716/533, loss: 0.014426421374082565 2023-01-21 12:52:52.748876: step: 1720/533, loss: 1.7452239262638614e-05 2023-01-21 12:52:53.910593: step: 1724/533, loss: 0.00018796921358443797 2023-01-21 12:52:55.033945: step: 1728/533, loss: 0.0021246911492198706 2023-01-21 12:52:56.186187: step: 1732/533, loss: 2.722740282479208e-05 2023-01-21 12:52:57.341311: step: 1736/533, loss: 0.0004248619079589844 2023-01-21 12:52:58.509603: step: 1740/533, loss: 0.00027742385282181203 2023-01-21 12:52:59.636140: step: 1744/533, loss: 0.013812637887895107 2023-01-21 12:53:00.777276: step: 1748/533, loss: 0.000461387651739642 2023-01-21 12:53:01.996708: step: 1752/533, loss: 0.050911761820316315 2023-01-21 12:53:03.190188: step: 1756/533, loss: 0.0004821777401957661 2023-01-21 12:53:04.354965: step: 1760/533, loss: 0.14102792739868164 2023-01-21 12:53:05.487428: step: 1764/533, loss: 0.001533126924186945 2023-01-21 12:53:06.639927: step: 1768/533, loss: 0.028049850836396217 2023-01-21 12:53:07.825654: step: 1772/533, loss: 0.030653953552246094 2023-01-21 12:53:08.965930: step: 1776/533, loss: 3.8719179428881034e-05 2023-01-21 12:53:10.149350: step: 1780/533, loss: 0.005081558134406805 2023-01-21 12:53:11.267524: step: 1784/533, loss: 5.602836608886719e-05 2023-01-21 12:53:12.437049: step: 1788/533, loss: 0.0060096741653978825 2023-01-21 12:53:13.602956: step: 1792/533, loss: 3.247260974603705e-05 2023-01-21 12:53:14.804273: step: 1796/533, loss: 0.0004333496035542339 2023-01-21 12:53:15.972757: step: 1800/533, loss: 0.00029277801513671875 2023-01-21 12:53:17.159832: step: 1804/533, loss: 0.00944442767649889 2023-01-21 12:53:18.359060: step: 1808/533, loss: 0.13763079047203064 2023-01-21 12:53:19.513303: step: 1812/533, loss: 0.028946973383426666 2023-01-21 12:53:20.648526: step: 1816/533, loss: 0.002588272327557206 2023-01-21 12:53:21.829000: step: 1820/533, loss: 0.0024524687323719263 2023-01-21 12:53:22.997452: step: 1824/533, loss: 0.02628307417035103 2023-01-21 12:53:24.145993: step: 1828/533, loss: 0.0023279190063476562 2023-01-21 12:53:25.305381: step: 1832/533, loss: 0.012312699109315872 2023-01-21 12:53:26.463428: step: 1836/533, loss: 0.04829578474164009 2023-01-21 12:53:27.646463: step: 1840/533, loss: 0.09863433241844177 2023-01-21 12:53:28.816850: step: 1844/533, loss: 0.031348228454589844 2023-01-21 12:53:29.977703: step: 1848/533, loss: 0.00034146307734772563 2023-01-21 12:53:31.130285: step: 1852/533, loss: 0.01343150157481432 2023-01-21 12:53:32.262749: step: 1856/533, loss: 3.7813188100699335e-05 2023-01-21 12:53:33.427541: step: 1860/533, loss: 0.0021717071067541838 2023-01-21 12:53:34.600823: step: 1864/533, loss: 0.01147542055696249 2023-01-21 12:53:35.797637: step: 1868/533, loss: 1.0278065204620361 2023-01-21 12:53:36.948727: step: 1872/533, loss: 0.0061668395064771175 2023-01-21 12:53:38.137009: step: 1876/533, loss: 0.044216301292181015 2023-01-21 12:53:39.291438: step: 1880/533, loss: 0.016011809930205345 2023-01-21 12:53:40.484119: step: 1884/533, loss: 0.0029096603393554688 2023-01-21 12:53:41.644342: step: 1888/533, loss: 0.02132740058004856 2023-01-21 12:53:42.802940: step: 1892/533, loss: 0.0008987426408566535 2023-01-21 12:53:43.968549: step: 1896/533, loss: 0.004156399052590132 2023-01-21 12:53:45.140688: step: 1900/533, loss: 0.00032887462293729186 2023-01-21 12:53:46.255003: step: 1904/533, loss: 1.792907642084174e-05 2023-01-21 12:53:47.455375: step: 1908/533, loss: 0.012526893988251686 2023-01-21 12:53:48.661619: step: 1912/533, loss: 0.031160498037934303 2023-01-21 12:53:49.840468: step: 1916/533, loss: 0.014987754635512829 2023-01-21 12:53:51.088906: step: 1920/533, loss: 0.11016568541526794 2023-01-21 12:53:52.260002: step: 1924/533, loss: 0.004780960269272327 2023-01-21 12:53:53.417896: step: 1928/533, loss: 0.008588695898652077 2023-01-21 12:53:54.613421: step: 1932/533, loss: 0.002572488971054554 2023-01-21 12:53:55.780176: step: 1936/533, loss: 0.008165550418198109 2023-01-21 12:53:56.994717: step: 1940/533, loss: 0.026792287826538086 2023-01-21 12:53:58.161078: step: 1944/533, loss: 0.00024280548677779734 2023-01-21 12:53:59.326995: step: 1948/533, loss: 0.000796413398347795 2023-01-21 12:54:00.539242: step: 1952/533, loss: 0.0007739067077636719 2023-01-21 12:54:01.701806: step: 1956/533, loss: 0.01584186591207981 2023-01-21 12:54:02.934979: step: 1960/533, loss: 0.01906299591064453 2023-01-21 12:54:04.098510: step: 1964/533, loss: 0.001792526338249445 2023-01-21 12:54:05.264457: step: 1968/533, loss: 0.0002571106015238911 2023-01-21 12:54:06.453920: step: 1972/533, loss: 0.005412483587861061 2023-01-21 12:54:07.639665: step: 1976/533, loss: 0.027904702350497246 2023-01-21 12:54:08.847157: step: 1980/533, loss: 0.0051780701614916325 2023-01-21 12:54:10.082687: step: 1984/533, loss: 0.0005664110649377108 2023-01-21 12:54:11.265911: step: 1988/533, loss: 0.027359582483768463 2023-01-21 12:54:12.452928: step: 1992/533, loss: 0.06540117412805557 2023-01-21 12:54:13.652811: step: 1996/533, loss: 7.162094698287547e-05 2023-01-21 12:54:14.819593: step: 2000/533, loss: 4.243850980856223e-06 2023-01-21 12:54:16.047210: step: 2004/533, loss: 0.5318679213523865 2023-01-21 12:54:17.247452: step: 2008/533, loss: 0.45101621747016907 2023-01-21 12:54:18.409329: step: 2012/533, loss: 0.00593195017427206 2023-01-21 12:54:19.559011: step: 2016/533, loss: 2.746581958490424e-05 2023-01-21 12:54:20.697579: step: 2020/533, loss: 0.01992168463766575 2023-01-21 12:54:21.888330: step: 2024/533, loss: 0.005261612124741077 2023-01-21 12:54:23.058049: step: 2028/533, loss: 0.0010038375621661544 2023-01-21 12:54:24.274525: step: 2032/533, loss: 0.006363296881318092 2023-01-21 12:54:25.449843: step: 2036/533, loss: 0.009067535400390625 2023-01-21 12:54:26.656856: step: 2040/533, loss: 0.015846826136112213 2023-01-21 12:54:27.835150: step: 2044/533, loss: 4.711151268566027e-05 2023-01-21 12:54:29.027909: step: 2048/533, loss: 0.005828666500747204 2023-01-21 12:54:30.209489: step: 2052/533, loss: 0.04320545122027397 2023-01-21 12:54:31.407197: step: 2056/533, loss: -4.38690176451928e-06 2023-01-21 12:54:32.586178: step: 2060/533, loss: 4.243851071805693e-05 2023-01-21 12:54:33.742919: step: 2064/533, loss: 0.044243812561035156 2023-01-21 12:54:34.895830: step: 2068/533, loss: 0.03879880905151367 2023-01-21 12:54:36.069468: step: 2072/533, loss: 0.7287365794181824 2023-01-21 12:54:37.217334: step: 2076/533, loss: 0.013099098578095436 2023-01-21 12:54:38.383336: step: 2080/533, loss: 0.00027809146558865905 2023-01-21 12:54:39.559207: step: 2084/533, loss: 6.961822691664565e-06 2023-01-21 12:54:40.703576: step: 2088/533, loss: 0.008771514520049095 2023-01-21 12:54:41.869801: step: 2092/533, loss: 0.0011700630420818925 2023-01-21 12:54:43.055935: step: 2096/533, loss: 0.006569195073097944 2023-01-21 12:54:44.206659: step: 2100/533, loss: 0.0061782835982739925 2023-01-21 12:54:45.443088: step: 2104/533, loss: 0.007445717230439186 2023-01-21 12:54:46.580076: step: 2108/533, loss: 0.024918079376220703 2023-01-21 12:54:47.773348: step: 2112/533, loss: 0.0028689862228929996 2023-01-21 12:54:48.921067: step: 2116/533, loss: 0.0009232998127117753 2023-01-21 12:54:50.082719: step: 2120/533, loss: 0.0002359390346100554 2023-01-21 12:54:51.257264: step: 2124/533, loss: 0.00056629185564816 2023-01-21 12:54:52.453820: step: 2128/533, loss: 0.00014734268188476562 2023-01-21 12:54:53.587695: step: 2132/533, loss: 0.0004146099090576172 ================================================== Loss: 0.035 -------------------- Dev: {'event': {'p': 0.6143162393162394, 'r': 0.7656458055925432, 'f1': 0.6816834617664493}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Test: {'event': {'p': 0.6556870818322182, 'r': 0.8145780051150895, 'f1': 0.7265469061876247}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Chinese: {'event': {'p': 0.5411764705882353, 'r': 0.8518518518518519, 'f1': 0.6618705035971222}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} 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': 23} Russian: {'event': {'p': 0.43243243243243246, 'r': 0.4444444444444444, 'f1': 0.4383561643835616}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 12:55:34.810653: step: 4/533, loss: 0.0014982223510742188 2023-01-21 12:55:35.953651: step: 8/533, loss: 2.117157055181451e-05 2023-01-21 12:55:37.116031: step: 12/533, loss: 0.004729080013930798 2023-01-21 12:55:38.339616: step: 16/533, loss: 0.003379392670467496 2023-01-21 12:55:39.557483: step: 20/533, loss: 0.0043249609880149364 2023-01-21 12:55:40.714064: step: 24/533, loss: 0.010345363058149815 2023-01-21 12:55:41.831284: step: 28/533, loss: -5.7220458984375e-06 2023-01-21 12:55:42.995029: step: 32/533, loss: 0.018908215686678886 2023-01-21 12:55:44.170460: step: 36/533, loss: 0.006651115603744984 2023-01-21 12:55:45.340661: step: 40/533, loss: 5.321502612787299e-05 2023-01-21 12:55:46.531007: step: 44/533, loss: 0.0044120075181126595 2023-01-21 12:55:47.691880: step: 48/533, loss: 0.0025789260398596525 2023-01-21 12:55:48.880303: step: 52/533, loss: 0.0016567230923101306 2023-01-21 12:55:50.049478: step: 56/533, loss: 0.005040633957833052 2023-01-21 12:55:51.281214: step: 60/533, loss: 0.00017986298189498484 2023-01-21 12:55:52.489303: step: 64/533, loss: 0.021048355847597122 2023-01-21 12:55:53.681742: step: 68/533, loss: 3.9577484130859375e-05 2023-01-21 12:55:54.828465: step: 72/533, loss: 0.0019787789788097143 2023-01-21 12:55:55.970275: step: 76/533, loss: 0.0010746002662926912 2023-01-21 12:55:57.145982: step: 80/533, loss: 9.460449655307457e-05 2023-01-21 12:55:58.329246: step: 84/533, loss: 0.004900455474853516 2023-01-21 12:55:59.520119: step: 88/533, loss: 0.017937565222382545 2023-01-21 12:56:00.691758: step: 92/533, loss: 0.0011552810901775956 2023-01-21 12:56:01.976964: step: 96/533, loss: 3.490447852527723e-05 2023-01-21 12:56:03.126789: step: 100/533, loss: 0.00010614395432639867 2023-01-21 12:56:04.269830: step: 104/533, loss: 0.0002875328063964844 2023-01-21 12:56:05.422327: step: 108/533, loss: 0.004686927888542414 2023-01-21 12:56:06.607825: step: 112/533, loss: 5.145073009771295e-05 2023-01-21 12:56:07.764745: step: 116/533, loss: 0.0008823394891805947 2023-01-21 12:56:08.937005: step: 120/533, loss: 0.015026092529296875 2023-01-21 12:56:10.089995: step: 124/533, loss: 7.152557373046875e-06 2023-01-21 12:56:11.236303: step: 128/533, loss: 5.836486889165826e-05 2023-01-21 12:56:12.421363: step: 132/533, loss: 0.043589115142822266 2023-01-21 12:56:13.612571: step: 136/533, loss: 0.00158519740216434 2023-01-21 12:56:14.788592: step: 140/533, loss: 0.001020240830257535 2023-01-21 12:56:15.992678: step: 144/533, loss: 0.012314415536820889 2023-01-21 12:56:17.142109: step: 148/533, loss: 0.49116066098213196 2023-01-21 12:56:18.334947: step: 152/533, loss: 0.026417160406708717 2023-01-21 12:56:19.494243: step: 156/533, loss: 0.00011157989501953125 2023-01-21 12:56:20.695767: step: 160/533, loss: 0.009055137634277344 2023-01-21 12:56:21.855103: step: 164/533, loss: 0.0026679993607103825 2023-01-21 12:56:23.012884: step: 168/533, loss: 0.6604849100112915 2023-01-21 12:56:24.146106: step: 172/533, loss: 0.0007463455549441278 2023-01-21 12:56:25.309843: step: 176/533, loss: 0.18963924050331116 2023-01-21 12:56:26.464817: step: 180/533, loss: 0.0006026267656125128 2023-01-21 12:56:27.645729: step: 184/533, loss: 0.0003443718305788934 2023-01-21 12:56:28.795546: step: 188/533, loss: 0.002713107969611883 2023-01-21 12:56:29.956546: step: 192/533, loss: 0.07361268997192383 2023-01-21 12:56:31.114934: step: 196/533, loss: 0.2587224841117859 2023-01-21 12:56:32.267450: step: 200/533, loss: 0.010776519775390625 2023-01-21 12:56:33.417487: step: 204/533, loss: 0.0046325684525072575 2023-01-21 12:56:34.586065: step: 208/533, loss: -4.2915345943583816e-07 2023-01-21 12:56:35.768170: step: 212/533, loss: 0.0023481370881199837 2023-01-21 12:56:36.967333: step: 216/533, loss: 7.743835885776207e-05 2023-01-21 12:56:38.201078: step: 220/533, loss: 7.596016075694934e-05 2023-01-21 12:56:39.301413: step: 224/533, loss: 0.00010766983177745715 2023-01-21 12:56:40.500519: step: 228/533, loss: 0.001570892403833568 2023-01-21 12:56:41.675797: step: 232/533, loss: 0.00017271042452193797 2023-01-21 12:56:42.845214: step: 236/533, loss: 0.00013027191744185984 2023-01-21 12:56:44.004689: step: 240/533, loss: 0.0488218329846859 2023-01-21 12:56:45.207204: step: 244/533, loss: 0.17562231421470642 2023-01-21 12:56:46.405941: step: 248/533, loss: 0.0055484771728515625 2023-01-21 12:56:47.661769: step: 252/533, loss: 0.003929710481315851 2023-01-21 12:56:48.898630: step: 256/533, loss: 0.0005680083995684981 2023-01-21 12:56:50.049220: step: 260/533, loss: 0.00023746490478515625 2023-01-21 12:56:51.262228: step: 264/533, loss: 0.005959606263786554 2023-01-21 12:56:52.416064: step: 268/533, loss: 0.00901808775961399 2023-01-21 12:56:53.622047: step: 272/533, loss: 0.037245895713567734 2023-01-21 12:56:54.805935: step: 276/533, loss: 1.3065338862361386e-05 2023-01-21 12:56:55.961329: step: 280/533, loss: 0.00537528982385993 2023-01-21 12:56:57.169081: step: 284/533, loss: 0.0012947083450853825 2023-01-21 12:56:58.345958: step: 288/533, loss: 0.010526848025619984 2023-01-21 12:56:59.558810: step: 292/533, loss: 0.00020656586275435984 2023-01-21 12:57:00.775853: step: 296/533, loss: 0.0017289519309997559 2023-01-21 12:57:01.930870: step: 300/533, loss: 3.080368333030492e-05 2023-01-21 12:57:03.119795: step: 304/533, loss: 0.02760915830731392 2023-01-21 12:57:04.322166: step: 308/533, loss: 0.00186500558629632 2023-01-21 12:57:05.458556: step: 312/533, loss: 0.0007704735035076737 2023-01-21 12:57:06.678921: step: 316/533, loss: 0.00027847292949445546 2023-01-21 12:57:07.820742: step: 320/533, loss: 0.03551159054040909 2023-01-21 12:57:09.034865: step: 324/533, loss: 0.07000233978033066 2023-01-21 12:57:10.176593: step: 328/533, loss: 0.0018533708062022924 2023-01-21 12:57:11.366908: step: 332/533, loss: 5.9890749980695546e-05 2023-01-21 12:57:12.550431: step: 336/533, loss: 4.38690185546875e-05 2023-01-21 12:57:13.708124: step: 340/533, loss: 5.016326758777723e-05 2023-01-21 12:57:14.829690: step: 344/533, loss: 0.016464997082948685 2023-01-21 12:57:16.046628: step: 348/533, loss: 0.015217496082186699 2023-01-21 12:57:17.224823: step: 352/533, loss: 0.0009018898126669228 2023-01-21 12:57:18.425778: step: 356/533, loss: 0.00095028878422454 2023-01-21 12:57:19.580118: step: 360/533, loss: 0.004959774203598499 2023-01-21 12:57:20.797681: step: 364/533, loss: 0.0031316757667809725 2023-01-21 12:57:21.927780: step: 368/533, loss: 9.012222108140122e-06 2023-01-21 12:57:23.047101: step: 372/533, loss: 0.0005079269758425653 2023-01-21 12:57:24.234828: step: 376/533, loss: 0.027384473010897636 2023-01-21 12:57:25.409243: step: 380/533, loss: 0.0008356571197509766 2023-01-21 12:57:26.569066: step: 384/533, loss: 5.684867574018426e-05 2023-01-21 12:57:27.737625: step: 388/533, loss: 2.193450927734375e-05 2023-01-21 12:57:28.895662: step: 392/533, loss: 0.0004699706914834678 2023-01-21 12:57:30.054166: step: 396/533, loss: 0.040029145777225494 2023-01-21 12:57:31.227941: step: 400/533, loss: 4.5490265620173886e-05 2023-01-21 12:57:32.373834: step: 404/533, loss: 0.017969371750950813 2023-01-21 12:57:33.543794: step: 408/533, loss: 0.0003772735653910786 2023-01-21 12:57:34.716269: step: 412/533, loss: 0.005244255065917969 2023-01-21 12:57:35.897815: step: 416/533, loss: 0.00014572143845725805 2023-01-21 12:57:37.086412: step: 420/533, loss: 0.0002780914364848286 2023-01-21 12:57:38.260334: step: 424/533, loss: 0.016827965155243874 2023-01-21 12:57:39.461811: step: 428/533, loss: 0.011587334796786308 2023-01-21 12:57:40.688727: step: 432/533, loss: 0.0019344331230968237 2023-01-21 12:57:41.892772: step: 436/533, loss: 0.17042502760887146 2023-01-21 12:57:43.088567: step: 440/533, loss: 0.00025768281193450093 2023-01-21 12:57:44.249567: step: 444/533, loss: 0.03531613573431969 2023-01-21 12:57:45.404825: step: 448/533, loss: 0.21429137885570526 2023-01-21 12:57:46.590125: step: 452/533, loss: 0.0023428918793797493 2023-01-21 12:57:47.762377: step: 456/533, loss: 0.003290414810180664 2023-01-21 12:57:48.946311: step: 460/533, loss: 0.037320517003536224 2023-01-21 12:57:50.179759: step: 464/533, loss: 0.044169191271066666 2023-01-21 12:57:51.378470: step: 468/533, loss: 0.002331733936443925 2023-01-21 12:57:52.583163: step: 472/533, loss: 0.03938446193933487 2023-01-21 12:57:53.796309: step: 476/533, loss: 0.002490520477294922 2023-01-21 12:57:54.966388: step: 480/533, loss: 0.005198383703827858 2023-01-21 12:57:56.144138: step: 484/533, loss: 0.0333745963871479 2023-01-21 12:57:57.272421: step: 488/533, loss: 3.290176664449973e-06 2023-01-21 12:57:58.430823: step: 492/533, loss: 0.0028717040549963713 2023-01-21 12:57:59.574437: step: 496/533, loss: 0.0167559627443552 2023-01-21 12:58:00.764183: step: 500/533, loss: 0.0025938034523278475 2023-01-21 12:58:01.965728: step: 504/533, loss: 0.05666179582476616 2023-01-21 12:58:03.169833: step: 508/533, loss: 0.006939411163330078 2023-01-21 12:58:04.339278: step: 512/533, loss: 0.00011105537851108238 2023-01-21 12:58:05.521469: step: 516/533, loss: 0.005019569303840399 2023-01-21 12:58:06.679919: step: 520/533, loss: 0.26031914353370667 2023-01-21 12:58:07.894871: step: 524/533, loss: -1.9073468138230965e-07 2023-01-21 12:58:09.084985: step: 528/533, loss: 1.5258789289873675e-06 2023-01-21 12:58:10.233907: step: 532/533, loss: 0.0020549774635583162 2023-01-21 12:58:11.396312: step: 536/533, loss: 7.629395258845761e-05 2023-01-21 12:58:12.538812: step: 540/533, loss: 0.002950048539787531 2023-01-21 12:58:13.690175: step: 544/533, loss: 0.08048410713672638 2023-01-21 12:58:14.877882: step: 548/533, loss: 0.00038318632869049907 2023-01-21 12:58:16.085938: step: 552/533, loss: 0.001358986017294228 2023-01-21 12:58:17.258010: step: 556/533, loss: -8.583068620282575e-07 2023-01-21 12:58:18.421495: step: 560/533, loss: 0.03986091539263725 2023-01-21 12:58:19.564255: step: 564/533, loss: 2.937316821771674e-05 2023-01-21 12:58:20.738360: step: 568/533, loss: 0.016434669494628906 2023-01-21 12:58:21.908324: step: 572/533, loss: 0.004740905947983265 2023-01-21 12:58:23.083785: step: 576/533, loss: 8.01086389401462e-06 2023-01-21 12:58:24.242869: step: 580/533, loss: 0.0002947807079181075 2023-01-21 12:58:25.436341: step: 584/533, loss: 0.01135711744427681 2023-01-21 12:58:26.613650: step: 588/533, loss: 2.448723554611206 2023-01-21 12:58:27.769805: step: 592/533, loss: 0.011686992831528187 2023-01-21 12:58:28.918685: step: 596/533, loss: 0.020264243707060814 2023-01-21 12:58:30.079224: step: 600/533, loss: 0.002237558364868164 2023-01-21 12:58:31.255771: step: 604/533, loss: 0.00012111663818359375 2023-01-21 12:58:32.473888: step: 608/533, loss: 0.002114773029461503 2023-01-21 12:58:33.700526: step: 612/533, loss: 4.38690185546875e-05 2023-01-21 12:58:34.874025: step: 616/533, loss: 0.0010427475208416581 2023-01-21 12:58:36.032651: step: 620/533, loss: 8.96453857421875e-05 2023-01-21 12:58:37.204025: step: 624/533, loss: 0.01702127419412136 2023-01-21 12:58:38.363812: step: 628/533, loss: 0.02596092037856579 2023-01-21 12:58:39.544778: step: 632/533, loss: 0.00015163421630859375 2023-01-21 12:58:40.709616: step: 636/533, loss: 0.0008037567604333162 2023-01-21 12:58:41.889062: step: 640/533, loss: 0.0015946388011798263 2023-01-21 12:58:43.037071: step: 644/533, loss: 7.033348083496094e-05 2023-01-21 12:58:44.241876: step: 648/533, loss: 0.0001253843365702778 2023-01-21 12:58:45.399606: step: 652/533, loss: 0.0597134605050087 2023-01-21 12:58:46.576812: step: 656/533, loss: 0.006297588814049959 2023-01-21 12:58:47.723384: step: 660/533, loss: 0.0007574081537313759 2023-01-21 12:58:48.883768: step: 664/533, loss: 0.033930208534002304 2023-01-21 12:58:50.068957: step: 668/533, loss: 7.286071922862902e-05 2023-01-21 12:58:51.245877: step: 672/533, loss: 0.026606369763612747 2023-01-21 12:58:52.421795: step: 676/533, loss: 7.286071922862902e-05 2023-01-21 12:58:53.623216: step: 680/533, loss: 0.0002827644348144531 2023-01-21 12:58:54.836025: step: 684/533, loss: 0.03730640560388565 2023-01-21 12:58:56.016481: step: 688/533, loss: 0.00074176792986691 2023-01-21 12:58:57.191796: step: 692/533, loss: 3.23295607813634e-05 2023-01-21 12:58:58.471394: step: 696/533, loss: 0.004149341490119696 2023-01-21 12:58:59.626270: step: 700/533, loss: 0.00017681121244095266 2023-01-21 12:59:00.810387: step: 704/533, loss: 0.0032795907463878393 2023-01-21 12:59:01.967382: step: 708/533, loss: 5.645752025884576e-05 2023-01-21 12:59:03.190994: step: 712/533, loss: 0.003503084182739258 2023-01-21 12:59:04.329592: step: 716/533, loss: 0.01647644303739071 2023-01-21 12:59:05.509175: step: 720/533, loss: 0.00027370452880859375 2023-01-21 12:59:06.656390: step: 724/533, loss: 0.00021629335242323577 2023-01-21 12:59:07.817060: step: 728/533, loss: 0.0017112254863604903 2023-01-21 12:59:09.009018: step: 732/533, loss: 0.0016474724980071187 2023-01-21 12:59:10.260924: step: 736/533, loss: 0.00041790009709075093 2023-01-21 12:59:11.497560: step: 740/533, loss: 0.006649160757660866 2023-01-21 12:59:12.654186: step: 744/533, loss: 4.353523036115803e-05 2023-01-21 12:59:13.856530: step: 748/533, loss: 0.021708868443965912 2023-01-21 12:59:15.040279: step: 752/533, loss: 0.005659389775246382 2023-01-21 12:59:16.209644: step: 756/533, loss: 0.0002841949462890625 2023-01-21 12:59:17.373327: step: 760/533, loss: 9.021758887683973e-05 2023-01-21 12:59:18.544821: step: 764/533, loss: 0.0033752440940588713 2023-01-21 12:59:19.708484: step: 768/533, loss: 0.008475112728774548 2023-01-21 12:59:20.946155: step: 772/533, loss: 0.05258503183722496 2023-01-21 12:59:22.126335: step: 776/533, loss: 0.018725205212831497 2023-01-21 12:59:23.274341: step: 780/533, loss: 0.0021941661834716797 2023-01-21 12:59:24.438451: step: 784/533, loss: 0.006798935122787952 2023-01-21 12:59:25.613360: step: 788/533, loss: 0.0001029014601954259 2023-01-21 12:59:26.884680: step: 792/533, loss: 0.14094963669776917 2023-01-21 12:59:28.048503: step: 796/533, loss: 0.00047369004460051656 2023-01-21 12:59:29.231118: step: 800/533, loss: 0.0003575325245037675 2023-01-21 12:59:30.417914: step: 804/533, loss: 0.0017885209526866674 2023-01-21 12:59:31.538158: step: 808/533, loss: 1.2683868590102065e-05 2023-01-21 12:59:32.752174: step: 812/533, loss: 1.4868286848068237 2023-01-21 12:59:33.890456: step: 816/533, loss: 0.009005260653793812 2023-01-21 12:59:35.083930: step: 820/533, loss: 9.269714064430445e-05 2023-01-21 12:59:36.233085: step: 824/533, loss: 0.01559295691549778 2023-01-21 12:59:37.402801: step: 828/533, loss: 0.0008003234979696572 2023-01-21 12:59:38.587926: step: 832/533, loss: 0.0001672744838288054 2023-01-21 12:59:39.771828: step: 836/533, loss: 0.01175780314952135 2023-01-21 12:59:40.904198: step: 840/533, loss: 0.016665076836943626 2023-01-21 12:59:42.096425: step: 844/533, loss: 8.711814734851941e-05 2023-01-21 12:59:43.263469: step: 848/533, loss: 0.03712005913257599 2023-01-21 12:59:44.435730: step: 852/533, loss: 0.0038500784430652857 2023-01-21 12:59:45.564006: step: 856/533, loss: 0.0008716583251953125 2023-01-21 12:59:46.722373: step: 860/533, loss: 6.67572021484375e-05 2023-01-21 12:59:47.917593: step: 864/533, loss: 0.039530184119939804 2023-01-21 12:59:49.099943: step: 868/533, loss: 0.006504202261567116 2023-01-21 12:59:50.270631: step: 872/533, loss: 1.0967254638671875e-05 2023-01-21 12:59:51.509933: step: 876/533, loss: 0.0012950897216796875 2023-01-21 12:59:52.674832: step: 880/533, loss: 0.0003065109485760331 2023-01-21 12:59:53.823856: step: 884/533, loss: 0.0003217697376385331 2023-01-21 12:59:55.018095: step: 888/533, loss: 0.03380412980914116 2023-01-21 12:59:56.176434: step: 892/533, loss: 0.004138564690947533 2023-01-21 12:59:57.383380: step: 896/533, loss: 0.013583374209702015 2023-01-21 12:59:58.559754: step: 900/533, loss: 0.0009149551624432206 2023-01-21 12:59:59.744574: step: 904/533, loss: 0.0003045797348022461 2023-01-21 13:00:00.934683: step: 908/533, loss: 8.759499178268015e-05 2023-01-21 13:00:02.105335: step: 912/533, loss: 0.0005922317504882812 2023-01-21 13:00:03.283368: step: 916/533, loss: 0.011572790332138538 2023-01-21 13:00:04.499857: step: 920/533, loss: 0.036786459386348724 2023-01-21 13:00:05.696288: step: 924/533, loss: 0.0003307342412881553 2023-01-21 13:00:06.908124: step: 928/533, loss: 0.0009876250987872481 2023-01-21 13:00:08.045278: step: 932/533, loss: 0.11505375057458878 2023-01-21 13:00:09.218380: step: 936/533, loss: 0.0005016326904296875 2023-01-21 13:00:10.364465: step: 940/533, loss: 0.027603626251220703 2023-01-21 13:00:11.533803: step: 944/533, loss: 0.0115814208984375 2023-01-21 13:00:12.710438: step: 948/533, loss: 0.0003444671747274697 2023-01-21 13:00:13.889095: step: 952/533, loss: 0.002324962755665183 2023-01-21 13:00:15.104124: step: 956/533, loss: 0.014894676394760609 2023-01-21 13:00:16.291650: step: 960/533, loss: 0.05403175204992294 2023-01-21 13:00:17.464104: step: 964/533, loss: 0.0507262721657753 2023-01-21 13:00:18.625277: step: 968/533, loss: 0.0005002021789550781 2023-01-21 13:00:19.838721: step: 972/533, loss: 0.001088905381038785 2023-01-21 13:00:21.018853: step: 976/533, loss: 0.001658630440942943 2023-01-21 13:00:22.184261: step: 980/533, loss: 0.002433204557746649 2023-01-21 13:00:23.354001: step: 984/533, loss: 0.0034229280427098274 2023-01-21 13:00:24.521615: step: 988/533, loss: 0.002062988467514515 2023-01-21 13:00:25.694607: step: 992/533, loss: 0.04250679165124893 2023-01-21 13:00:26.847350: step: 996/533, loss: 0.00473316153511405 2023-01-21 13:00:28.014446: step: 1000/533, loss: 0.001293182373046875 2023-01-21 13:00:29.216960: step: 1004/533, loss: 0.00026597976102493703 2023-01-21 13:00:30.351173: step: 1008/533, loss: 0.010640907101333141 2023-01-21 13:00:31.537968: step: 1012/533, loss: 0.014592457562685013 2023-01-21 13:00:32.722833: step: 1016/533, loss: 0.0036890029441565275 2023-01-21 13:00:33.855847: step: 1020/533, loss: 0.00022878646268509328 2023-01-21 13:00:35.010470: step: 1024/533, loss: 0.20881108939647675 2023-01-21 13:00:36.180514: step: 1028/533, loss: 0.0181534755975008 2023-01-21 13:00:37.333665: step: 1032/533, loss: 0.39380112290382385 2023-01-21 13:00:38.517200: step: 1036/533, loss: 0.000217437744140625 2023-01-21 13:00:39.694457: step: 1040/533, loss: 0.00019626619177870452 2023-01-21 13:00:40.870148: step: 1044/533, loss: 0.0017273903358727694 2023-01-21 13:00:42.029251: step: 1048/533, loss: 0.0006781101692467928 2023-01-21 13:00:43.181614: step: 1052/533, loss: 0.0786321684718132 2023-01-21 13:00:44.310498: step: 1056/533, loss: 3.204345557605848e-05 2023-01-21 13:00:45.477949: step: 1060/533, loss: 0.06606197357177734 2023-01-21 13:00:46.634988: step: 1064/533, loss: 0.014790629968047142 2023-01-21 13:00:47.805090: step: 1068/533, loss: 0.0026713847182691097 2023-01-21 13:00:48.982859: step: 1072/533, loss: 0.000154399880557321 2023-01-21 13:00:50.188194: step: 1076/533, loss: 0.0007165909046307206 2023-01-21 13:00:51.363353: step: 1080/533, loss: 0.009022235870361328 2023-01-21 13:00:52.516648: step: 1084/533, loss: 0.0007832527044229209 2023-01-21 13:00:53.721807: step: 1088/533, loss: 0.002002716064453125 2023-01-21 13:00:54.855911: step: 1092/533, loss: 0.002108240034431219 2023-01-21 13:00:56.034281: step: 1096/533, loss: 0.0008775711175985634 2023-01-21 13:00:57.265574: step: 1100/533, loss: 0.01584758795797825 2023-01-21 13:00:58.444734: step: 1104/533, loss: 0.01760435290634632 2023-01-21 13:00:59.669303: step: 1108/533, loss: 0.0007223129505291581 2023-01-21 13:01:00.811289: step: 1112/533, loss: 0.001470375107601285 2023-01-21 13:01:01.992957: step: 1116/533, loss: 7.939338684082031e-05 2023-01-21 13:01:03.220549: step: 1120/533, loss: 0.02177581936120987 2023-01-21 13:01:04.343884: step: 1124/533, loss: 0.0007461547502316535 2023-01-21 13:01:05.487322: step: 1128/533, loss: 5.054474058852065e-06 2023-01-21 13:01:06.636484: step: 1132/533, loss: 4.940032886224799e-05 2023-01-21 13:01:07.841435: step: 1136/533, loss: 0.00809392984956503 2023-01-21 13:01:09.014841: step: 1140/533, loss: 6.341934204101562e-05 2023-01-21 13:01:10.148645: step: 1144/533, loss: 0.007476234808564186 2023-01-21 13:01:11.318695: step: 1148/533, loss: 0.0005296707386150956 2023-01-21 13:01:12.496685: step: 1152/533, loss: 0.004399109166115522 2023-01-21 13:01:13.677356: step: 1156/533, loss: 0.014605332165956497 2023-01-21 13:01:14.853488: step: 1160/533, loss: 0.012473869137465954 2023-01-21 13:01:15.992866: step: 1164/533, loss: 0.009981537237763405 2023-01-21 13:01:17.126482: step: 1168/533, loss: 0.0016018867027014494 2023-01-21 13:01:18.331460: step: 1172/533, loss: 0.0006052017561160028 2023-01-21 13:01:19.470633: step: 1176/533, loss: 0.0015629768604412675 2023-01-21 13:01:20.630478: step: 1180/533, loss: 0.021303748711943626 2023-01-21 13:01:21.814342: step: 1184/533, loss: 0.004734134767204523 2023-01-21 13:01:22.979411: step: 1188/533, loss: 0.0010828971862792969 2023-01-21 13:01:24.177420: step: 1192/533, loss: 0.013994026929140091 2023-01-21 13:01:25.350595: step: 1196/533, loss: 0.00028057099552825093 2023-01-21 13:01:26.526184: step: 1200/533, loss: 0.002509593963623047 2023-01-21 13:01:27.710165: step: 1204/533, loss: 0.0016264915466308594 2023-01-21 13:01:28.868187: step: 1208/533, loss: 0.0005393981700763106 2023-01-21 13:01:30.108202: step: 1212/533, loss: 0.01260986365377903 2023-01-21 13:01:31.276917: step: 1216/533, loss: 0.05681347846984863 2023-01-21 13:01:32.438466: step: 1220/533, loss: 0.00026760101900435984 2023-01-21 13:01:33.656896: step: 1224/533, loss: 0.0006829261546954513 2023-01-21 13:01:34.819105: step: 1228/533, loss: 0.0022098543122410774 2023-01-21 13:01:36.000470: step: 1232/533, loss: 0.04069128260016441 2023-01-21 13:01:37.197717: step: 1236/533, loss: 0.0013536453479900956 2023-01-21 13:01:38.349918: step: 1240/533, loss: 0.0004873752477578819 2023-01-21 13:01:39.534739: step: 1244/533, loss: 0.029383039101958275 2023-01-21 13:01:40.675081: step: 1248/533, loss: 0.0016596914501860738 2023-01-21 13:01:41.898453: step: 1252/533, loss: 7.82012921263231e-06 2023-01-21 13:01:43.091883: step: 1256/533, loss: 8.239746239269152e-05 2023-01-21 13:01:44.256177: step: 1260/533, loss: 0.00013425351062323898 2023-01-21 13:01:45.393850: step: 1264/533, loss: 0.0007048130501061678 2023-01-21 13:01:46.585507: step: 1268/533, loss: 0.008801555261015892 2023-01-21 13:01:47.711803: step: 1272/533, loss: 0.00040836335392668843 2023-01-21 13:01:48.855591: step: 1276/533, loss: 0.0011699676979333162 2023-01-21 13:01:50.030037: step: 1280/533, loss: 0.007870912551879883 2023-01-21 13:01:51.170994: step: 1284/533, loss: 0.027782440185546875 2023-01-21 13:01:52.319675: step: 1288/533, loss: 3.571510751498863e-05 2023-01-21 13:01:53.502184: step: 1292/533, loss: 0.0010124206310138106 2023-01-21 13:01:54.681561: step: 1296/533, loss: 6.799697439419106e-05 2023-01-21 13:01:55.872466: step: 1300/533, loss: 0.00496673583984375 2023-01-21 13:01:57.030333: step: 1304/533, loss: 0.0011420012451708317 2023-01-21 13:01:58.217322: step: 1308/533, loss: 0.2273712158203125 2023-01-21 13:01:59.353627: step: 1312/533, loss: 0.03544168919324875 2023-01-21 13:02:00.490589: step: 1316/533, loss: 0.0009374141809530556 2023-01-21 13:02:01.651224: step: 1320/533, loss: 0.0007168293814174831 2023-01-21 13:02:02.805691: step: 1324/533, loss: 0.026342440396547318 2023-01-21 13:02:03.980414: step: 1328/533, loss: 0.006759834475815296 2023-01-21 13:02:05.165556: step: 1332/533, loss: 5.0258637202205136e-05 2023-01-21 13:02:06.368895: step: 1336/533, loss: 8.144378807628527e-05 2023-01-21 13:02:07.523641: step: 1340/533, loss: 0.0002568245108705014 2023-01-21 13:02:08.702156: step: 1344/533, loss: 0.0033945085015147924 2023-01-21 13:02:09.843555: step: 1348/533, loss: 0.0196043960750103 2023-01-21 13:02:11.002632: step: 1352/533, loss: 0.08306007832288742 2023-01-21 13:02:12.165994: step: 1356/533, loss: 0.009069060906767845 2023-01-21 13:02:13.323954: step: 1360/533, loss: 0.00041980744572356343 2023-01-21 13:02:14.506955: step: 1364/533, loss: 0.0005861282697878778 2023-01-21 13:02:15.716113: step: 1368/533, loss: 0.0033693313598632812 2023-01-21 13:02:16.874041: step: 1372/533, loss: 0.020530126988887787 2023-01-21 13:02:18.021492: step: 1376/533, loss: 0.0007372855907306075 2023-01-21 13:02:19.189578: step: 1380/533, loss: 0.002046871231868863 2023-01-21 13:02:20.333660: step: 1384/533, loss: 9.15527380129788e-06 2023-01-21 13:02:21.507339: step: 1388/533, loss: 1.0681153071345761e-05 2023-01-21 13:02:22.685206: step: 1392/533, loss: 1.2922286259708926e-05 2023-01-21 13:02:23.891388: step: 1396/533, loss: 0.0009120941394940019 2023-01-21 13:02:25.068867: step: 1400/533, loss: 0.0014757155440747738 2023-01-21 13:02:26.246366: step: 1404/533, loss: 0.03947634994983673 2023-01-21 13:02:27.400397: step: 1408/533, loss: 5.8650975915952586e-06 2023-01-21 13:02:28.605303: step: 1412/533, loss: 0.002025031950324774 2023-01-21 13:02:29.773322: step: 1416/533, loss: 0.03777818754315376 2023-01-21 13:02:30.971919: step: 1420/533, loss: 0.004207038786262274 2023-01-21 13:02:32.170810: step: 1424/533, loss: 0.0033987045753747225 2023-01-21 13:02:33.327951: step: 1428/533, loss: 1.3508878946304321 2023-01-21 13:02:34.480034: step: 1432/533, loss: 6.237030174816027e-05 2023-01-21 13:02:35.607430: step: 1436/533, loss: 0.030152320861816406 2023-01-21 13:02:36.744614: step: 1440/533, loss: 6.999969627941027e-05 2023-01-21 13:02:37.925182: step: 1444/533, loss: 0.09783058613538742 2023-01-21 13:02:39.079084: step: 1448/533, loss: 6.27517729299143e-05 2023-01-21 13:02:40.232195: step: 1452/533, loss: 0.0020839690696448088 2023-01-21 13:02:41.415033: step: 1456/533, loss: 0.0014142037834972143 2023-01-21 13:02:42.588182: step: 1460/533, loss: 0.002233791397884488 2023-01-21 13:02:43.772185: step: 1464/533, loss: 0.00030918122502043843 2023-01-21 13:02:44.893992: step: 1468/533, loss: 0.012582874856889248 2023-01-21 13:02:46.092544: step: 1472/533, loss: 0.010568427853286266 2023-01-21 13:02:47.265799: step: 1476/533, loss: 0.00021409988403320312 2023-01-21 13:02:48.461449: step: 1480/533, loss: 0.00041370390681549907 2023-01-21 13:02:49.621563: step: 1484/533, loss: 0.02518758736550808 2023-01-21 13:02:50.786513: step: 1488/533, loss: 0.00594520615413785 2023-01-21 13:02:51.963257: step: 1492/533, loss: 0.0007223129505291581 2023-01-21 13:02:53.114281: step: 1496/533, loss: 0.00371971121057868 2023-01-21 13:02:54.296415: step: 1500/533, loss: 2.9850005375919864e-05 2023-01-21 13:02:55.452921: step: 1504/533, loss: 7.996558997547254e-05 2023-01-21 13:02:56.624395: step: 1508/533, loss: 0.007046890445053577 2023-01-21 13:02:57.771071: step: 1512/533, loss: 0.01240387000143528 2023-01-21 13:02:58.928770: step: 1516/533, loss: 0.004190349485725164 2023-01-21 13:03:00.131405: step: 1520/533, loss: 0.0005035400390625 2023-01-21 13:03:01.308477: step: 1524/533, loss: 0.005941963288933039 2023-01-21 13:03:02.468427: step: 1528/533, loss: 0.011968541890382767 2023-01-21 13:03:03.653419: step: 1532/533, loss: 0.0006523132906295359 2023-01-21 13:03:04.816114: step: 1536/533, loss: 0.003118753433227539 2023-01-21 13:03:05.993435: step: 1540/533, loss: 0.013037538155913353 2023-01-21 13:03:07.168872: step: 1544/533, loss: 0.015993595123291016 2023-01-21 13:03:08.349317: step: 1548/533, loss: 0.0017304421635344625 2023-01-21 13:03:09.544396: step: 1552/533, loss: 0.000209808349609375 2023-01-21 13:03:10.707434: step: 1556/533, loss: 0.0011954070068895817 2023-01-21 13:03:11.865009: step: 1560/533, loss: -2.4795533590804553e-06 2023-01-21 13:03:13.027458: step: 1564/533, loss: 0.021761227399110794 2023-01-21 13:03:14.169829: step: 1568/533, loss: 0.0016402244800701737 2023-01-21 13:03:15.335437: step: 1572/533, loss: 0.027640342712402344 2023-01-21 13:03:16.513397: step: 1576/533, loss: 0.05317964777350426 2023-01-21 13:03:17.734046: step: 1580/533, loss: 0.015453719533979893 2023-01-21 13:03:18.904224: step: 1584/533, loss: 0.004646873567253351 2023-01-21 13:03:20.064076: step: 1588/533, loss: 0.0002182483731303364 2023-01-21 13:03:21.256378: step: 1592/533, loss: 0.00022077560424804688 2023-01-21 13:03:22.417531: step: 1596/533, loss: 0.03993396833539009 2023-01-21 13:03:23.648375: step: 1600/533, loss: 0.03127565234899521 2023-01-21 13:03:24.778463: step: 1604/533, loss: 0.00031414031400345266 2023-01-21 13:03:25.959442: step: 1608/533, loss: 0.014753628522157669 2023-01-21 13:03:27.150767: step: 1612/533, loss: 0.004345322027802467 2023-01-21 13:03:28.284694: step: 1616/533, loss: 0.00010051728168036789 2023-01-21 13:03:29.457371: step: 1620/533, loss: 0.03536186367273331 2023-01-21 13:03:30.644661: step: 1624/533, loss: 0.015340805053710938 2023-01-21 13:03:31.828715: step: 1628/533, loss: 0.004257774446159601 2023-01-21 13:03:33.007798: step: 1632/533, loss: 0.0011619568103924394 2023-01-21 13:03:34.216027: step: 1636/533, loss: 0.0029239654541015625 2023-01-21 13:03:35.402635: step: 1640/533, loss: 0.004975604824721813 2023-01-21 13:03:36.570960: step: 1644/533, loss: 0.005943679716438055 2023-01-21 13:03:37.734095: step: 1648/533, loss: 0.006720066070556641 2023-01-21 13:03:38.934280: step: 1652/533, loss: 0.00027964115724898875 2023-01-21 13:03:40.064696: step: 1656/533, loss: 8.869170414982364e-06 2023-01-21 13:03:41.259697: step: 1660/533, loss: 5.7220458984375e-06 2023-01-21 13:03:42.430600: step: 1664/533, loss: 0.06058015674352646 2023-01-21 13:03:43.547284: step: 1668/533, loss: 0.00027742385282181203 2023-01-21 13:03:44.760744: step: 1672/533, loss: 0.0005420685047283769 2023-01-21 13:03:45.908828: step: 1676/533, loss: 0.00010056496103061363 2023-01-21 13:03:47.050953: step: 1680/533, loss: 0.00025577546330168843 2023-01-21 13:03:48.211544: step: 1684/533, loss: 0.0041335588321089745 2023-01-21 13:03:49.346408: step: 1688/533, loss: 0.003565979190170765 2023-01-21 13:03:50.533302: step: 1692/533, loss: 0.06697248667478561 2023-01-21 13:03:51.772052: step: 1696/533, loss: 0.0018962860340252519 2023-01-21 13:03:52.949838: step: 1700/533, loss: 0.0030312540475279093 2023-01-21 13:03:54.149945: step: 1704/533, loss: 0.00044269562931731343 2023-01-21 13:03:55.298960: step: 1708/533, loss: 0.027163410559296608 2023-01-21 13:03:56.468390: step: 1712/533, loss: 0.00010833740816451609 2023-01-21 13:03:57.640372: step: 1716/533, loss: 0.03129911422729492 2023-01-21 13:03:58.810442: step: 1720/533, loss: 1.9073488033427566e-07 2023-01-21 13:03:59.978863: step: 1724/533, loss: 0.0001616478111827746 2023-01-21 13:04:01.175021: step: 1728/533, loss: 0.008506583981215954 2023-01-21 13:04:02.357276: step: 1732/533, loss: 0.00022010803513694555 2023-01-21 13:04:03.556682: step: 1736/533, loss: 0.0564180389046669 2023-01-21 13:04:04.738286: step: 1740/533, loss: 0.00010404586646473035 2023-01-21 13:04:05.885221: step: 1744/533, loss: 0.0002197742578573525 2023-01-21 13:04:07.061528: step: 1748/533, loss: 0.08783798664808273 2023-01-21 13:04:08.192992: step: 1752/533, loss: 0.004085349850356579 2023-01-21 13:04:09.366684: step: 1756/533, loss: 7.324219041038305e-05 2023-01-21 13:04:10.550410: step: 1760/533, loss: 0.014804650098085403 2023-01-21 13:04:11.723088: step: 1764/533, loss: 0.04349031671881676 2023-01-21 13:04:12.923783: step: 1768/533, loss: 1.1920930774067529e-05 2023-01-21 13:04:14.115711: step: 1772/533, loss: 9.15527380129788e-06 2023-01-21 13:04:15.298991: step: 1776/533, loss: 0.0006799697875976562 2023-01-21 13:04:16.461087: step: 1780/533, loss: 0.00044841767521575093 2023-01-21 13:04:17.641038: step: 1784/533, loss: 4.310607982915826e-05 2023-01-21 13:04:18.818593: step: 1788/533, loss: 0.04329051822423935 2023-01-21 13:04:20.026142: step: 1792/533, loss: 0.010746193118393421 2023-01-21 13:04:21.246473: step: 1796/533, loss: 0.00022792816162109375 2023-01-21 13:04:22.406097: step: 1800/533, loss: 0.003046226454898715 2023-01-21 13:04:23.555778: step: 1804/533, loss: 0.00012149810208939016 2023-01-21 13:04:24.748193: step: 1808/533, loss: 0.01930561102926731 2023-01-21 13:04:25.953550: step: 1812/533, loss: 4.234314110362902e-05 2023-01-21 13:04:27.129069: step: 1816/533, loss: 5.445480564958416e-05 2023-01-21 13:04:28.306292: step: 1820/533, loss: 0.003831768175587058 2023-01-21 13:04:29.484877: step: 1824/533, loss: 0.0299745574593544 2023-01-21 13:04:30.664802: step: 1828/533, loss: 9.717942157294601e-05 2023-01-21 13:04:31.855989: step: 1832/533, loss: 0.045876119285821915 2023-01-21 13:04:33.048779: step: 1836/533, loss: 0.00013175010099075735 2023-01-21 13:04:34.223615: step: 1840/533, loss: 0.00037260056706145406 2023-01-21 13:04:35.401671: step: 1844/533, loss: 0.030855130404233932 2023-01-21 13:04:36.560233: step: 1848/533, loss: 0.0003955840948037803 2023-01-21 13:04:37.715000: step: 1852/533, loss: 0.0023347854148596525 2023-01-21 13:04:38.913255: step: 1856/533, loss: 0.0037843703757971525 2023-01-21 13:04:40.050014: step: 1860/533, loss: -1.7166134966828395e-06 2023-01-21 13:04:41.250545: step: 1864/533, loss: 0.0440642386674881 2023-01-21 13:04:42.403321: step: 1868/533, loss: 0.05874033272266388 2023-01-21 13:04:43.613219: step: 1872/533, loss: 2.880096508306451e-05 2023-01-21 13:04:44.772484: step: 1876/533, loss: 0.003457927843555808 2023-01-21 13:04:45.962571: step: 1880/533, loss: 0.008210564032196999 2023-01-21 13:04:47.167563: step: 1884/533, loss: 0.007722091861069202 2023-01-21 13:04:48.310947: step: 1888/533, loss: 0.0012625694507732987 2023-01-21 13:04:49.519729: step: 1892/533, loss: 0.0007431030389852822 2023-01-21 13:04:50.684078: step: 1896/533, loss: 0.009793472476303577 2023-01-21 13:04:51.874539: step: 1900/533, loss: 8.611679368186742e-05 2023-01-21 13:04:53.085136: step: 1904/533, loss: 0.0005155563703738153 2023-01-21 13:04:54.284760: step: 1908/533, loss: 0.009315108880400658 2023-01-21 13:04:55.496125: step: 1912/533, loss: 0.0002994537353515625 2023-01-21 13:04:56.687316: step: 1916/533, loss: 0.016551781445741653 2023-01-21 13:04:57.853821: step: 1920/533, loss: 0.00035457609919831157 2023-01-21 13:04:59.004003: step: 1924/533, loss: 0.00037097930908203125 2023-01-21 13:05:00.173173: step: 1928/533, loss: 0.0001262664736714214 2023-01-21 13:05:01.358622: step: 1932/533, loss: 0.001853180001489818 2023-01-21 13:05:02.528806: step: 1936/533, loss: 0.001208400703035295 2023-01-21 13:05:03.676248: step: 1940/533, loss: 8.592977974331006e-05 2023-01-21 13:05:04.851371: step: 1944/533, loss: 0.05228424072265625 2023-01-21 13:05:06.013243: step: 1948/533, loss: 0.03904237970709801 2023-01-21 13:05:07.190138: step: 1952/533, loss: 3.519058373058215e-05 2023-01-21 13:05:08.302143: step: 1956/533, loss: 0.008115005679428577 2023-01-21 13:05:09.489333: step: 1960/533, loss: 4.010200791526586e-05 2023-01-21 13:05:10.661379: step: 1964/533, loss: 3.471374657237902e-05 2023-01-21 13:05:11.846052: step: 1968/533, loss: 0.016849135980010033 2023-01-21 13:05:13.022829: step: 1972/533, loss: 0.005553150083869696 2023-01-21 13:05:14.234225: step: 1976/533, loss: 0.013669204898178577 2023-01-21 13:05:15.379940: step: 1980/533, loss: 0.0012292861938476562 2023-01-21 13:05:16.581077: step: 1984/533, loss: 0.0021957396529614925 2023-01-21 13:05:17.800906: step: 1988/533, loss: 0.010261917486786842 2023-01-21 13:05:18.974967: step: 1992/533, loss: 0.0032817842438817024 2023-01-21 13:05:20.143854: step: 1996/533, loss: 0.003951930906623602 2023-01-21 13:05:21.305869: step: 2000/533, loss: 0.0030815124046057463 2023-01-21 13:05:22.534265: step: 2004/533, loss: 3.232955714338459e-05 2023-01-21 13:05:23.715924: step: 2008/533, loss: 0.0003964424249716103 2023-01-21 13:05:24.890088: step: 2012/533, loss: 0.0013454437721520662 2023-01-21 13:05:26.087559: step: 2016/533, loss: 0.0006504058837890625 2023-01-21 13:05:27.259049: step: 2020/533, loss: 0.005679011344909668 2023-01-21 13:05:28.445239: step: 2024/533, loss: 0.0008004665724001825 2023-01-21 13:05:29.642138: step: 2028/533, loss: 0.006409263703972101 2023-01-21 13:05:30.861755: step: 2032/533, loss: 0.0013885498046875 2023-01-21 13:05:32.017919: step: 2036/533, loss: 0.001469946000725031 2023-01-21 13:05:33.153106: step: 2040/533, loss: 9.17434663278982e-05 2023-01-21 13:05:34.342129: step: 2044/533, loss: 6.294250852079131e-06 2023-01-21 13:05:35.565646: step: 2048/533, loss: 0.00026197434635832906 2023-01-21 13:05:36.769868: step: 2052/533, loss: 0.0003758430539164692 2023-01-21 13:05:37.983068: step: 2056/533, loss: 2.6226043701171875e-06 2023-01-21 13:05:39.147265: step: 2060/533, loss: 5.53131121705519e-06 2023-01-21 13:05:40.350588: step: 2064/533, loss: 0.00010986327833961695 2023-01-21 13:05:41.515027: step: 2068/533, loss: 0.0020662308670580387 2023-01-21 13:05:42.710671: step: 2072/533, loss: -1.316070574830519e-05 2023-01-21 13:05:43.885785: step: 2076/533, loss: 5.378723290050402e-05 2023-01-21 13:05:45.053695: step: 2080/533, loss: 0.010884500108659267 2023-01-21 13:05:46.233568: step: 2084/533, loss: 0.005922507960349321 2023-01-21 13:05:47.422124: step: 2088/533, loss: 0.0003299713134765625 2023-01-21 13:05:48.597441: step: 2092/533, loss: 0.0008209228981286287 2023-01-21 13:05:49.760161: step: 2096/533, loss: 0.0015687941340729594 2023-01-21 13:05:50.927868: step: 2100/533, loss: 0.02914581447839737 2023-01-21 13:05:52.129714: step: 2104/533, loss: 0.006865215487778187 2023-01-21 13:05:53.361982: step: 2108/533, loss: 0.0027782439719885588 2023-01-21 13:05:54.558919: step: 2112/533, loss: 0.010496318340301514 2023-01-21 13:05:55.786517: step: 2116/533, loss: 0.013097573071718216 2023-01-21 13:05:56.951743: step: 2120/533, loss: 0.014444352127611637 2023-01-21 13:05:58.139170: step: 2124/533, loss: 0.03334693983197212 2023-01-21 13:05:59.396293: step: 2128/533, loss: 8.020401583053172e-05 2023-01-21 13:06:00.554735: step: 2132/533, loss: 4.673004241340095e-06 ================================================== Loss: 0.025 -------------------- Dev: {'event': {'p': 0.6048472075869337, 'r': 0.7643142476697736, 'f1': 0.6752941176470589}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Test: {'event': {'p': 0.6530612244897959, 'r': 0.7979539641943734, 'f1': 0.718273381294964}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Chinese: {'event': {'p': 0.5647058823529412, 'r': 0.8888888888888888, 'f1': 0.6906474820143885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} 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': 24} Russian: {'event': {'p': 0.3953488372093023, 'r': 0.4722222222222222, 'f1': 0.43037974683544306}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 13:06:41.571974: step: 4/533, loss: 0.11187629401683807 2023-01-21 13:06:42.781609: step: 8/533, loss: 0.0007357597351074219 2023-01-21 13:06:43.975014: step: 12/533, loss: 0.0008819579961709678 2023-01-21 13:06:45.124797: step: 16/533, loss: 0.0005161285516805947 2023-01-21 13:06:46.261556: step: 20/533, loss: 0.0001636505185160786 2023-01-21 13:06:47.437978: step: 24/533, loss: 0.007214903831481934 2023-01-21 13:06:48.595512: step: 28/533, loss: 0.00075616838876158 2023-01-21 13:06:49.794428: step: 32/533, loss: 0.02224902994930744 2023-01-21 13:06:50.960835: step: 36/533, loss: 0.00033230779808945954 2023-01-21 13:06:52.142074: step: 40/533, loss: 0.0005222320323809981 2023-01-21 13:06:53.307850: step: 44/533, loss: 0.0003184318484272808 2023-01-21 13:06:54.475610: step: 48/533, loss: 0.0045608519576489925 2023-01-21 13:06:55.633934: step: 52/533, loss: 0.00021038056001998484 2023-01-21 13:06:56.797363: step: 56/533, loss: 9.298324584960938e-05 2023-01-21 13:06:57.954281: step: 60/533, loss: 6.115436553955078e-05 2023-01-21 13:06:59.125091: step: 64/533, loss: 7.133484177757055e-05 2023-01-21 13:07:00.338455: step: 68/533, loss: 0.057862281799316406 2023-01-21 13:07:01.497218: step: 72/533, loss: 0.035753536969423294 2023-01-21 13:07:02.698488: step: 76/533, loss: 0.00014028548321221024 2023-01-21 13:07:03.868801: step: 80/533, loss: 0.0010984421242028475 2023-01-21 13:07:05.034630: step: 84/533, loss: 0.03076777420938015 2023-01-21 13:07:06.172774: step: 88/533, loss: 0.0024822235573083162 2023-01-21 13:07:07.328612: step: 92/533, loss: 0.009707736782729626 2023-01-21 13:07:08.481803: step: 96/533, loss: 0.032755278050899506 2023-01-21 13:07:09.667685: step: 100/533, loss: 0.014388466253876686 2023-01-21 13:07:10.854404: step: 104/533, loss: 0.0012490273220464587 2023-01-21 13:07:12.088989: step: 108/533, loss: 0.01947803422808647 2023-01-21 13:07:13.269795: step: 112/533, loss: 0.00010051727440441027 2023-01-21 13:07:14.413361: step: 116/533, loss: 0.0011391640873625875 2023-01-21 13:07:15.561721: step: 120/533, loss: 0.0001617431698832661 2023-01-21 13:07:16.752200: step: 124/533, loss: 0.013813018798828125 2023-01-21 13:07:17.911210: step: 128/533, loss: 0.001083803130313754 2023-01-21 13:07:19.101548: step: 132/533, loss: 4.27722952736076e-05 2023-01-21 13:07:20.264573: step: 136/533, loss: 6.814003427280113e-05 2023-01-21 13:07:21.463132: step: 140/533, loss: 0.0005793571472167969 2023-01-21 13:07:22.669185: step: 144/533, loss: 0.015749169513583183 2023-01-21 13:07:23.887123: step: 148/533, loss: 0.019277572631835938 2023-01-21 13:07:25.070110: step: 152/533, loss: 0.00017080307588912547 2023-01-21 13:07:26.208747: step: 156/533, loss: 0.0004277229309082031 2023-01-21 13:07:27.408521: step: 160/533, loss: 0.0004178047238383442 2023-01-21 13:07:28.572865: step: 164/533, loss: 0.027348613366484642 2023-01-21 13:07:29.726160: step: 168/533, loss: 0.0006777763483114541 2023-01-21 13:07:30.894402: step: 172/533, loss: 0.0006651878356933594 2023-01-21 13:07:32.104590: step: 176/533, loss: 0.0005069732433184981 2023-01-21 13:07:33.353566: step: 180/533, loss: 0.008580387569963932 2023-01-21 13:07:34.542964: step: 184/533, loss: 0.00039348602876998484 2023-01-21 13:07:35.714489: step: 188/533, loss: 1.0626280307769775 2023-01-21 13:07:36.904267: step: 192/533, loss: 0.0007314682006835938 2023-01-21 13:07:38.149613: step: 196/533, loss: 7.41958647267893e-05 2023-01-21 13:07:39.356630: step: 200/533, loss: 0.01301670167595148 2023-01-21 13:07:40.536952: step: 204/533, loss: 0.0005212783580645919 2023-01-21 13:07:41.749408: step: 208/533, loss: 0.003574275877326727 2023-01-21 13:07:42.911998: step: 212/533, loss: 0.0015644074883311987 2023-01-21 13:07:44.144696: step: 216/533, loss: 0.02648754231631756 2023-01-21 13:07:45.333795: step: 220/533, loss: 0.0005815983167849481 2023-01-21 13:07:46.594542: step: 224/533, loss: 0.04251975938677788 2023-01-21 13:07:47.826685: step: 228/533, loss: 0.005886840634047985 2023-01-21 13:07:48.989107: step: 232/533, loss: 0.00017662048048805445 2023-01-21 13:07:50.154897: step: 236/533, loss: 0.010887622833251953 2023-01-21 13:07:51.329809: step: 240/533, loss: 0.004258537199348211 2023-01-21 13:07:52.551545: step: 244/533, loss: 0.0005887985462322831 2023-01-21 13:07:53.714636: step: 248/533, loss: 0.05128965526819229 2023-01-21 13:07:54.882462: step: 252/533, loss: 0.0010391235118731856 2023-01-21 13:07:56.044602: step: 256/533, loss: 0.0002510070917196572 2023-01-21 13:07:57.236412: step: 260/533, loss: 3.71456153516192e-05 2023-01-21 13:07:58.423145: step: 264/533, loss: 0.0056063649244606495 2023-01-21 13:07:59.571402: step: 268/533, loss: 0.0002210617094533518 2023-01-21 13:08:00.722453: step: 272/533, loss: 1.5211105164780747e-05 2023-01-21 13:08:01.892668: step: 276/533, loss: 0.0035533905029296875 2023-01-21 13:08:03.093097: step: 280/533, loss: 4.0817263652570546e-05 2023-01-21 13:08:04.262448: step: 284/533, loss: 0.003212356474250555 2023-01-21 13:08:05.429184: step: 288/533, loss: 0.0012172699207440019 2023-01-21 13:08:06.599639: step: 292/533, loss: 1.62601463671308e-05 2023-01-21 13:08:07.800454: step: 296/533, loss: 0.05152120813727379 2023-01-21 13:08:08.984408: step: 300/533, loss: 0.00019426345534157008 2023-01-21 13:08:10.170252: step: 304/533, loss: 0.011230086907744408 2023-01-21 13:08:11.363154: step: 308/533, loss: 0.04899730533361435 2023-01-21 13:08:12.515853: step: 312/533, loss: 0.008487128652632236 2023-01-21 13:08:13.662083: step: 316/533, loss: 0.0018607140518724918 2023-01-21 13:08:14.897929: step: 320/533, loss: 0.002101325895637274 2023-01-21 13:08:16.077449: step: 324/533, loss: 0.006189728155732155 2023-01-21 13:08:17.274311: step: 328/533, loss: 3.863573510898277e-05 2023-01-21 13:08:18.457796: step: 332/533, loss: 0.00797185953706503 2023-01-21 13:08:19.634733: step: 336/533, loss: 8.39233416627394e-06 2023-01-21 13:08:20.826792: step: 340/533, loss: 0.0007979392539709806 2023-01-21 13:08:22.064903: step: 344/533, loss: 0.03519124910235405 2023-01-21 13:08:23.204081: step: 348/533, loss: 0.0005229950184002519 2023-01-21 13:08:24.364618: step: 352/533, loss: 0.0005412101745605469 2023-01-21 13:08:25.505648: step: 356/533, loss: 0.0002808570861816406 2023-01-21 13:08:26.631945: step: 360/533, loss: 0.0005090713384561241 2023-01-21 13:08:27.783108: step: 364/533, loss: 8.964539119915571e-06 2023-01-21 13:08:28.950407: step: 368/533, loss: 5.3787229262525216e-05 2023-01-21 13:08:30.125353: step: 372/533, loss: 0.0006996155134402215 2023-01-21 13:08:31.337176: step: 376/533, loss: 7.724762326688506e-06 2023-01-21 13:08:32.508667: step: 380/533, loss: 0.06823787838220596 2023-01-21 13:08:33.650400: step: 384/533, loss: 0.001269864966161549 2023-01-21 13:08:34.856624: step: 388/533, loss: 3.147125244140625e-05 2023-01-21 13:08:36.047474: step: 392/533, loss: 3.767013913602568e-05 2023-01-21 13:08:37.244700: step: 396/533, loss: 0.00026063917903229594 2023-01-21 13:08:38.408772: step: 400/533, loss: -5.34057608092553e-06 2023-01-21 13:08:39.584766: step: 404/533, loss: 8.311271812999621e-05 2023-01-21 13:08:40.749225: step: 408/533, loss: 0.011280441656708717 2023-01-21 13:08:41.904433: step: 412/533, loss: 0.0004051208670716733 2023-01-21 13:08:43.079719: step: 416/533, loss: 0.0001504898100392893 2023-01-21 13:08:44.267589: step: 420/533, loss: 3.6239625842426904e-06 2023-01-21 13:08:45.423313: step: 424/533, loss: 0.010908221825957298 2023-01-21 13:08:46.581039: step: 428/533, loss: 0.03046722337603569 2023-01-21 13:08:47.737447: step: 432/533, loss: 0.004014777950942516 2023-01-21 13:08:48.906782: step: 436/533, loss: 5.912781489314511e-06 2023-01-21 13:08:50.067223: step: 440/533, loss: 0.044097900390625 2023-01-21 13:08:51.248982: step: 444/533, loss: 0.00017986298189498484 2023-01-21 13:08:52.407252: step: 448/533, loss: 0.0013586044078692794 2023-01-21 13:08:53.599466: step: 452/533, loss: 0.0005167007329873741 2023-01-21 13:08:54.807388: step: 456/533, loss: 0.000335693359375 2023-01-21 13:08:55.998771: step: 460/533, loss: 0.00014629363431595266 2023-01-21 13:08:57.184106: step: 464/533, loss: 0.0005573272937908769 2023-01-21 13:08:58.429573: step: 468/533, loss: 0.04576678201556206 2023-01-21 13:08:59.608109: step: 472/533, loss: 0.04335803911089897 2023-01-21 13:09:00.771211: step: 476/533, loss: -4.386902219266631e-06 2023-01-21 13:09:01.980911: step: 480/533, loss: 0.00110206613317132 2023-01-21 13:09:03.134826: step: 484/533, loss: 0.00012626648822333664 2023-01-21 13:09:04.330369: step: 488/533, loss: 0.005187797360122204 2023-01-21 13:09:05.532762: step: 492/533, loss: 0.0012491225497797132 2023-01-21 13:09:06.756991: step: 496/533, loss: 0.048845864832401276 2023-01-21 13:09:07.953761: step: 500/533, loss: 0.0017173767555505037 2023-01-21 13:09:09.127964: step: 504/533, loss: 0.00175132742151618 2023-01-21 13:09:10.317924: step: 508/533, loss: 0.03961363062262535 2023-01-21 13:09:11.507858: step: 512/533, loss: 0.005207347683608532 2023-01-21 13:09:12.708196: step: 516/533, loss: 0.010910320095717907 2023-01-21 13:09:13.891233: step: 520/533, loss: 0.0005066872108727694 2023-01-21 13:09:15.052298: step: 524/533, loss: 8.554458327125758e-05 2023-01-21 13:09:16.226889: step: 528/533, loss: 0.04548225551843643 2023-01-21 13:09:17.406065: step: 532/533, loss: 3.185272362316027e-05 2023-01-21 13:09:18.550706: step: 536/533, loss: 6.809234764659777e-05 2023-01-21 13:09:19.737717: step: 540/533, loss: 0.0005782127263955772 2023-01-21 13:09:20.914161: step: 544/533, loss: 0.0005617141723632812 2023-01-21 13:09:22.111215: step: 548/533, loss: 0.005755377002060413 2023-01-21 13:09:23.311901: step: 552/533, loss: 0.00954656582325697 2023-01-21 13:09:24.538121: step: 556/533, loss: 0.026959801092743874 2023-01-21 13:09:25.724421: step: 560/533, loss: 0.07968740910291672 2023-01-21 13:09:26.915248: step: 564/533, loss: 0.024430274963378906 2023-01-21 13:09:28.121233: step: 568/533, loss: -6.198882601893274e-06 2023-01-21 13:09:29.255332: step: 572/533, loss: 0.05496664345264435 2023-01-21 13:09:30.446405: step: 576/533, loss: 0.033574867993593216 2023-01-21 13:09:31.627873: step: 580/533, loss: 0.006825065705925226 2023-01-21 13:09:32.805309: step: 584/533, loss: 2.3555756342830136e-05 2023-01-21 13:09:33.998843: step: 588/533, loss: 0.09605465084314346 2023-01-21 13:09:35.130735: step: 592/533, loss: 0.019091034308075905 2023-01-21 13:09:36.323380: step: 596/533, loss: 7.658005051780492e-05 2023-01-21 13:09:37.517856: step: 600/533, loss: 0.002284526824951172 2023-01-21 13:09:38.703876: step: 604/533, loss: 0.009167099371552467 2023-01-21 13:09:39.933622: step: 608/533, loss: 2.765655517578125e-05 2023-01-21 13:09:41.113943: step: 612/533, loss: 0.0008384704706259072 2023-01-21 13:09:42.296187: step: 616/533, loss: 0.044295214116573334 2023-01-21 13:09:43.480375: step: 620/533, loss: 0.0038341046310961246 2023-01-21 13:09:44.647669: step: 624/533, loss: 0.0005313873407430947 2023-01-21 13:09:45.811676: step: 628/533, loss: 0.0007600784301757812 2023-01-21 13:09:46.982707: step: 632/533, loss: 0.01061263494193554 2023-01-21 13:09:48.212265: step: 636/533, loss: 0.009187126532196999 2023-01-21 13:09:49.426843: step: 640/533, loss: 8.845329284667969e-05 2023-01-21 13:09:50.637098: step: 644/533, loss: 0.010899257846176624 2023-01-21 13:09:51.857276: step: 648/533, loss: 0.0005669593228958547 2023-01-21 13:09:53.040148: step: 652/533, loss: 0.04291486740112305 2023-01-21 13:09:54.257962: step: 656/533, loss: 0.0007602691766805947 2023-01-21 13:09:55.430247: step: 660/533, loss: 2.079009937006049e-05 2023-01-21 13:09:56.619424: step: 664/533, loss: 0.0001142501860158518 2023-01-21 13:09:57.770625: step: 668/533, loss: 4.95910626341356e-06 2023-01-21 13:09:58.981763: step: 672/533, loss: 0.0002685069921426475 2023-01-21 13:10:00.150715: step: 676/533, loss: 0.024470187723636627 2023-01-21 13:10:01.349685: step: 680/533, loss: 0.004728889558464289 2023-01-21 13:10:02.470805: step: 684/533, loss: 4.76837158203125e-07 2023-01-21 13:10:03.661030: step: 688/533, loss: 0.00017051695613190532 2023-01-21 13:10:04.842778: step: 692/533, loss: 0.006273365113884211 2023-01-21 13:10:05.997278: step: 696/533, loss: 9.565353684592992e-05 2023-01-21 13:10:07.202722: step: 700/533, loss: 0.001807403634302318 2023-01-21 13:10:08.349497: step: 704/533, loss: 2.7084352041129023e-05 2023-01-21 13:10:09.547490: step: 708/533, loss: 0.044736552983522415 2023-01-21 13:10:10.729643: step: 712/533, loss: 0.1034703254699707 2023-01-21 13:10:11.923802: step: 716/533, loss: 0.003424358321353793 2023-01-21 13:10:13.109320: step: 720/533, loss: 0.25962716341018677 2023-01-21 13:10:14.299040: step: 724/533, loss: 0.0012144566280767322 2023-01-21 13:10:15.467996: step: 728/533, loss: 0.001938820001669228 2023-01-21 13:10:16.663348: step: 732/533, loss: 0.015377044677734375 2023-01-21 13:10:17.788237: step: 736/533, loss: 0.0173155777156353 2023-01-21 13:10:18.997719: step: 740/533, loss: 0.001611268613487482 2023-01-21 13:10:20.160761: step: 744/533, loss: 0.005706692114472389 2023-01-21 13:10:21.334873: step: 748/533, loss: 0.027206992730498314 2023-01-21 13:10:22.527748: step: 752/533, loss: 5.6648252211743966e-05 2023-01-21 13:10:23.695980: step: 756/533, loss: 0.0010866641532629728 2023-01-21 13:10:24.846216: step: 760/533, loss: 0.0010593414772301912 2023-01-21 13:10:26.048131: step: 764/533, loss: 0.009982443414628506 2023-01-21 13:10:27.286137: step: 768/533, loss: 0.035612963140010834 2023-01-21 13:10:28.447279: step: 772/533, loss: 0.003970527555793524 2023-01-21 13:10:29.626872: step: 776/533, loss: 0.011561202816665173 2023-01-21 13:10:30.800303: step: 780/533, loss: 0.003782367566600442 2023-01-21 13:10:31.980322: step: 784/533, loss: 0.0054168701171875 2023-01-21 13:10:33.172663: step: 788/533, loss: 0.01558761578053236 2023-01-21 13:10:34.368749: step: 792/533, loss: 0.004012966062873602 2023-01-21 13:10:35.552559: step: 796/533, loss: 0.07053347676992416 2023-01-21 13:10:36.724217: step: 800/533, loss: 0.03857855871319771 2023-01-21 13:10:37.890064: step: 804/533, loss: 0.002723312471061945 2023-01-21 13:10:39.077308: step: 808/533, loss: 0.0044186594896018505 2023-01-21 13:10:40.238238: step: 812/533, loss: 1.0013580322265625e-05 2023-01-21 13:10:41.385913: step: 816/533, loss: 8.325576345669106e-05 2023-01-21 13:10:42.586974: step: 820/533, loss: 0.0002503395080566406 2023-01-21 13:10:43.758498: step: 824/533, loss: 0.0004323005850892514 2023-01-21 13:10:44.935905: step: 828/533, loss: 7.82012939453125e-05 2023-01-21 13:10:46.116338: step: 832/533, loss: 0.0005410194862633944 2023-01-21 13:10:47.299289: step: 836/533, loss: 0.005740070249885321 2023-01-21 13:10:48.470787: step: 840/533, loss: 1.9073486328125e-05 2023-01-21 13:10:49.640514: step: 844/533, loss: 0.00019674302893690765 2023-01-21 13:10:50.845471: step: 848/533, loss: 0.020435143262147903 2023-01-21 13:10:52.027148: step: 852/533, loss: 0.057543568313121796 2023-01-21 13:10:53.186950: step: 856/533, loss: 0.00160560617223382 2023-01-21 13:10:54.429238: step: 860/533, loss: 1.7499922250863165e-05 2023-01-21 13:10:55.575685: step: 864/533, loss: 0.00021610260591842234 2023-01-21 13:10:56.710488: step: 868/533, loss: 0.00015373229689430445 2023-01-21 13:10:57.920058: step: 872/533, loss: 0.00016098022751975805 2023-01-21 13:10:59.055082: step: 876/533, loss: 0.00025768281193450093 2023-01-21 13:11:00.239753: step: 880/533, loss: 0.00939950905740261 2023-01-21 13:11:01.402649: step: 884/533, loss: 0.14021149277687073 2023-01-21 13:11:02.554543: step: 888/533, loss: 3.23295607813634e-05 2023-01-21 13:11:03.724933: step: 892/533, loss: 0.09389925003051758 2023-01-21 13:11:04.918099: step: 896/533, loss: 0.0006057739374227822 2023-01-21 13:11:06.099205: step: 900/533, loss: 1.144409225162235e-06 2023-01-21 13:11:07.286454: step: 904/533, loss: 0.004372787661850452 2023-01-21 13:11:08.455643: step: 908/533, loss: 3.185272362316027e-05 2023-01-21 13:11:09.626196: step: 912/533, loss: 0.0025501251220703125 2023-01-21 13:11:10.744961: step: 916/533, loss: 0.0006044865003786981 2023-01-21 13:11:11.908120: step: 920/533, loss: 0.019122600555419922 2023-01-21 13:11:13.070256: step: 924/533, loss: 0.005641079042106867 2023-01-21 13:11:14.221706: step: 928/533, loss: 0.00014085769362282008 2023-01-21 13:11:15.378076: step: 932/533, loss: 1.4781951904296875e-05 2023-01-21 13:11:16.574213: step: 936/533, loss: 0.0013290406204760075 2023-01-21 13:11:17.792082: step: 940/533, loss: 0.003409004071727395 2023-01-21 13:11:18.975538: step: 944/533, loss: 0.007462692447006702 2023-01-21 13:11:20.156276: step: 948/533, loss: 0.005602407269179821 2023-01-21 13:11:21.380144: step: 952/533, loss: 0.004732513800263405 2023-01-21 13:11:22.597057: step: 956/533, loss: 0.0025073052383959293 2023-01-21 13:11:23.776617: step: 960/533, loss: 0.003074741456657648 2023-01-21 13:11:24.927068: step: 964/533, loss: 3.4999848139705136e-05 2023-01-21 13:11:26.082880: step: 968/533, loss: 0.00028791429940611124 2023-01-21 13:11:27.236120: step: 972/533, loss: 1.1157990229548886e-05 2023-01-21 13:11:28.405207: step: 976/533, loss: 0.00041961669921875 2023-01-21 13:11:29.613589: step: 980/533, loss: 6.246566772460938e-05 2023-01-21 13:11:30.813812: step: 984/533, loss: 0.021274566650390625 2023-01-21 13:11:32.005307: step: 988/533, loss: 0.0003974914725404233 2023-01-21 13:11:33.194257: step: 992/533, loss: 0.002517604734748602 2023-01-21 13:11:34.390481: step: 996/533, loss: 0.007695198059082031 2023-01-21 13:11:35.562724: step: 1000/533, loss: 1.4114380974206142e-05 2023-01-21 13:11:36.730492: step: 1004/533, loss: 0.0017191886436194181 2023-01-21 13:11:37.861038: step: 1008/533, loss: 0.006597613915801048 2023-01-21 13:11:39.018444: step: 1012/533, loss: 8.134842209983617e-05 2023-01-21 13:11:40.188162: step: 1016/533, loss: 0.045996859669685364 2023-01-21 13:11:41.381738: step: 1020/533, loss: 1.5640260244254023e-05 2023-01-21 13:11:42.549829: step: 1024/533, loss: 0.000274658203125 2023-01-21 13:11:43.756718: step: 1028/533, loss: 0.0004460334894247353 2023-01-21 13:11:44.933018: step: 1032/533, loss: 0.005132771097123623 2023-01-21 13:11:46.127392: step: 1036/533, loss: 0.00021157263836357743 2023-01-21 13:11:47.335616: step: 1040/533, loss: 0.013942909426987171 2023-01-21 13:11:48.535747: step: 1044/533, loss: 0.0032739639282226562 2023-01-21 13:11:49.753283: step: 1048/533, loss: 0.001295375870540738 2023-01-21 13:11:50.953911: step: 1052/533, loss: 0.0023937225341796875 2023-01-21 13:11:52.204676: step: 1056/533, loss: 0.0003565788210835308 2023-01-21 13:11:53.342417: step: 1060/533, loss: 0.08198156952857971 2023-01-21 13:11:54.573655: step: 1064/533, loss: 0.023420333862304688 2023-01-21 13:11:55.780223: step: 1068/533, loss: 0.0005703926435671747 2023-01-21 13:11:56.985465: step: 1072/533, loss: 0.005052519030869007 2023-01-21 13:11:58.150177: step: 1076/533, loss: 0.004059124272316694 2023-01-21 13:11:59.320051: step: 1080/533, loss: 0.022756576538085938 2023-01-21 13:12:00.505878: step: 1084/533, loss: 0.0025980949867516756 2023-01-21 13:12:01.658423: step: 1088/533, loss: 0.00041370390681549907 2023-01-21 13:12:02.876322: step: 1092/533, loss: 0.010689354501664639 2023-01-21 13:12:04.067850: step: 1096/533, loss: 0.012452316470444202 2023-01-21 13:12:05.211602: step: 1100/533, loss: 0.0003386497846804559 2023-01-21 13:12:06.416974: step: 1104/533, loss: 0.010661983862519264 2023-01-21 13:12:07.574982: step: 1108/533, loss: 0.0008134842501021922 2023-01-21 13:12:08.732624: step: 1112/533, loss: 0.0011558354599401355 2023-01-21 13:12:09.864703: step: 1116/533, loss: 4.444122168933973e-05 2023-01-21 13:12:11.081360: step: 1120/533, loss: 0.00040092470590025187 2023-01-21 13:12:12.305415: step: 1124/533, loss: 0.009002303704619408 2023-01-21 13:12:13.445869: step: 1128/533, loss: 0.013695811852812767 2023-01-21 13:12:14.661420: step: 1132/533, loss: 0.0006999969482421875 2023-01-21 13:12:15.795149: step: 1136/533, loss: 0.0010070801945403218 2023-01-21 13:12:16.991547: step: 1140/533, loss: 0.029296113178133965 2023-01-21 13:12:18.197724: step: 1144/533, loss: 6.0272213886491954e-05 2023-01-21 13:12:19.339806: step: 1148/533, loss: 0.0015497207641601562 2023-01-21 13:12:20.504342: step: 1152/533, loss: 0.004676247015595436 2023-01-21 13:12:21.666244: step: 1156/533, loss: 0.011130047030746937 2023-01-21 13:12:22.865509: step: 1160/533, loss: 8.988380432128906e-05 2023-01-21 13:12:24.044245: step: 1164/533, loss: 1.76429750808893e-06 2023-01-21 13:12:25.224755: step: 1168/533, loss: 1.1634827387752011e-05 2023-01-21 13:12:26.413068: step: 1172/533, loss: 0.030812358483672142 2023-01-21 13:12:27.613455: step: 1176/533, loss: 4.310607982915826e-05 2023-01-21 13:12:28.786072: step: 1180/533, loss: 0.002107906388118863 2023-01-21 13:12:29.942232: step: 1184/533, loss: 0.018909836187958717 2023-01-21 13:12:31.114089: step: 1188/533, loss: 0.0052969930693507195 2023-01-21 13:12:32.260892: step: 1192/533, loss: 0.00013313292583916336 2023-01-21 13:12:33.470053: step: 1196/533, loss: 0.046775247901678085 2023-01-21 13:12:34.670948: step: 1200/533, loss: 0.00278987898491323 2023-01-21 13:12:35.858972: step: 1204/533, loss: 2.384185791015625e-05 2023-01-21 13:12:37.060990: step: 1208/533, loss: 0.003305053571239114 2023-01-21 13:12:38.221068: step: 1212/533, loss: 0.013186216354370117 2023-01-21 13:12:39.405275: step: 1216/533, loss: 0.0031128881964832544 2023-01-21 13:12:40.595917: step: 1220/533, loss: 0.0005787849659100175 2023-01-21 13:12:41.762863: step: 1224/533, loss: 0.00012273788161110133 2023-01-21 13:12:42.957083: step: 1228/533, loss: 0.0012176514137536287 2023-01-21 13:12:44.135524: step: 1232/533, loss: 7.400513277389109e-05 2023-01-21 13:12:45.317670: step: 1236/533, loss: 0.0033599853049963713 2023-01-21 13:12:46.464584: step: 1240/533, loss: 1.7166138377433526e-06 2023-01-21 13:12:47.629049: step: 1244/533, loss: 6.332397606456652e-05 2023-01-21 13:12:48.776652: step: 1248/533, loss: 0.05436830595135689 2023-01-21 13:12:49.969914: step: 1252/533, loss: 0.0068357475101947784 2023-01-21 13:12:51.158634: step: 1256/533, loss: -6.48498553346144e-06 2023-01-21 13:12:52.306505: step: 1260/533, loss: 0.04023933410644531 2023-01-21 13:12:53.470242: step: 1264/533, loss: 0.02508525922894478 2023-01-21 13:12:54.634458: step: 1268/533, loss: 0.029063226655125618 2023-01-21 13:12:55.847850: step: 1272/533, loss: 0.0005532264476642013 2023-01-21 13:12:57.000529: step: 1276/533, loss: 0.0001367569057038054 2023-01-21 13:12:58.184905: step: 1280/533, loss: 0.0017366409301757812 2023-01-21 13:12:59.404149: step: 1284/533, loss: 0.022514915093779564 2023-01-21 13:13:00.578697: step: 1288/533, loss: 0.0881490707397461 2023-01-21 13:13:01.847353: step: 1292/533, loss: 6.237030174816027e-05 2023-01-21 13:13:03.041415: step: 1296/533, loss: 0.010213851928710938 2023-01-21 13:13:04.250386: step: 1300/533, loss: 0.03126201778650284 2023-01-21 13:13:05.372429: step: 1304/533, loss: 0.004970932379364967 2023-01-21 13:13:06.552806: step: 1308/533, loss: 5.7220458984375e-06 2023-01-21 13:13:07.768479: step: 1312/533, loss: 0.023522090166807175 2023-01-21 13:13:08.942254: step: 1316/533, loss: 0.00018234254093840718 2023-01-21 13:13:10.149952: step: 1320/533, loss: 0.00069088937016204 2023-01-21 13:13:11.297015: step: 1324/533, loss: 0.015287209302186966 2023-01-21 13:13:12.476432: step: 1328/533, loss: 0.0001544952392578125 2023-01-21 13:13:13.627038: step: 1332/533, loss: 9.34600848268019e-06 2023-01-21 13:13:14.790804: step: 1336/533, loss: 0.044062234461307526 2023-01-21 13:13:15.950839: step: 1340/533, loss: 0.007546902168542147 2023-01-21 13:13:17.131444: step: 1344/533, loss: 4.2343137465650216e-05 2023-01-21 13:13:18.292324: step: 1348/533, loss: 3.681182715808973e-05 2023-01-21 13:13:19.503754: step: 1352/533, loss: 3.070831371587701e-05 2023-01-21 13:13:20.711120: step: 1356/533, loss: 0.000644683837890625 2023-01-21 13:13:21.865413: step: 1360/533, loss: 8.39233416627394e-06 2023-01-21 13:13:23.049902: step: 1364/533, loss: 0.00973806343972683 2023-01-21 13:13:24.234303: step: 1368/533, loss: 0.00025281906710006297 2023-01-21 13:13:25.385783: step: 1372/533, loss: 0.0015600204933434725 2023-01-21 13:13:26.544743: step: 1376/533, loss: 0.009256219491362572 2023-01-21 13:13:27.698100: step: 1380/533, loss: 0.009183120913803577 2023-01-21 13:13:28.880499: step: 1384/533, loss: 5.893707202631049e-05 2023-01-21 13:13:30.037439: step: 1388/533, loss: 0.00030236245947889984 2023-01-21 13:13:31.207565: step: 1392/533, loss: 0.004322052001953125 2023-01-21 13:13:32.384393: step: 1396/533, loss: 0.0019245147705078125 2023-01-21 13:13:33.551925: step: 1400/533, loss: 0.0005702972412109375 2023-01-21 13:13:34.787958: step: 1404/533, loss: 0.0007865906227380037 2023-01-21 13:13:35.999580: step: 1408/533, loss: 0.003495597979053855 2023-01-21 13:13:37.223605: step: 1412/533, loss: 0.00019149782019667327 2023-01-21 13:13:38.450371: step: 1416/533, loss: 1.62601463671308e-05 2023-01-21 13:13:39.677287: step: 1420/533, loss: 0.016955852508544922 2023-01-21 13:13:40.869222: step: 1424/533, loss: 0.0009185791132040322 2023-01-21 13:13:42.018837: step: 1428/533, loss: 0.015378189273178577 2023-01-21 13:13:43.203133: step: 1432/533, loss: 0.005722379311919212 2023-01-21 13:13:44.402388: step: 1436/533, loss: 6.999969627941027e-05 2023-01-21 13:13:45.559836: step: 1440/533, loss: 7.43865966796875e-05 2023-01-21 13:13:46.745541: step: 1444/533, loss: 0.0013107300037518144 2023-01-21 13:13:47.940966: step: 1448/533, loss: 1.144409225162235e-06 2023-01-21 13:13:49.140351: step: 1452/533, loss: -7.05719003235572e-06 2023-01-21 13:13:50.315036: step: 1456/533, loss: 0.012697791680693626 2023-01-21 13:13:51.519819: step: 1460/533, loss: 3.757476952159777e-05 2023-01-21 13:13:52.677062: step: 1464/533, loss: 0.0001029014601954259 2023-01-21 13:13:53.888712: step: 1468/533, loss: 0.0032018660567700863 2023-01-21 13:13:55.060666: step: 1472/533, loss: 0.03953571245074272 2023-01-21 13:13:56.210949: step: 1476/533, loss: 9.059906005859375e-05 2023-01-21 13:13:57.385045: step: 1480/533, loss: 0.0004643440479412675 2023-01-21 13:13:58.552234: step: 1484/533, loss: 0.1530340164899826 2023-01-21 13:13:59.704816: step: 1488/533, loss: 0.0010137557983398438 2023-01-21 13:14:00.874194: step: 1492/533, loss: 6.337165541481227e-05 2023-01-21 13:14:02.028308: step: 1496/533, loss: 0.002912807511165738 2023-01-21 13:14:03.183302: step: 1500/533, loss: 0.0008403778192587197 2023-01-21 13:14:04.366048: step: 1504/533, loss: 0.0038170814514160156 2023-01-21 13:14:05.537167: step: 1508/533, loss: 0.008150959387421608 2023-01-21 13:14:06.677799: step: 1512/533, loss: 0.00011172294762218371 2023-01-21 13:14:07.872865: step: 1516/533, loss: 0.022918129339814186 2023-01-21 13:14:09.052536: step: 1520/533, loss: 0.011520194821059704 2023-01-21 13:14:10.269807: step: 1524/533, loss: 0.0005056381342001259 2023-01-21 13:14:11.447199: step: 1528/533, loss: 0.0002267837553517893 2023-01-21 13:14:12.660539: step: 1532/533, loss: 0.027029801160097122 2023-01-21 13:14:13.830429: step: 1536/533, loss: 0.004332542419433594 2023-01-21 13:14:15.054415: step: 1540/533, loss: 0.0001625060976948589 2023-01-21 13:14:16.235843: step: 1544/533, loss: 0.4024145007133484 2023-01-21 13:14:17.428856: step: 1548/533, loss: 0.0007131577003747225 2023-01-21 13:14:18.593661: step: 1552/533, loss: 0.048481278121471405 2023-01-21 13:14:19.738531: step: 1556/533, loss: 0.027990151196718216 2023-01-21 13:14:20.925486: step: 1560/533, loss: 0.0007340431329794228 2023-01-21 13:14:22.109716: step: 1564/533, loss: 0.014251136220991611 2023-01-21 13:14:23.289895: step: 1568/533, loss: 0.048250962048769 2023-01-21 13:14:24.475813: step: 1572/533, loss: 0.0034626247361302376 2023-01-21 13:14:25.662802: step: 1576/533, loss: 6.999969627941027e-05 2023-01-21 13:14:26.849143: step: 1580/533, loss: 0.01412420254200697 2023-01-21 13:14:28.008640: step: 1584/533, loss: 9.212493750965223e-05 2023-01-21 13:14:29.173843: step: 1588/533, loss: 0.0006238937494345009 2023-01-21 13:14:30.370590: step: 1592/533, loss: 0.013140678405761719 2023-01-21 13:14:31.597002: step: 1596/533, loss: 0.002686786698177457 2023-01-21 13:14:32.857691: step: 1600/533, loss: 0.005118751898407936 2023-01-21 13:14:34.062825: step: 1604/533, loss: 0.008208608254790306 2023-01-21 13:14:35.240609: step: 1608/533, loss: 0.00014591217041015625 2023-01-21 13:14:36.426651: step: 1612/533, loss: 0.006329535972326994 2023-01-21 13:14:37.638740: step: 1616/533, loss: 0.012324143201112747 2023-01-21 13:14:38.848521: step: 1620/533, loss: 0.07143593579530716 2023-01-21 13:14:40.038518: step: 1624/533, loss: 3.852844383800402e-05 2023-01-21 13:14:41.188930: step: 1628/533, loss: 0.0004291534423828125 2023-01-21 13:14:42.362682: step: 1632/533, loss: 0.03128309175372124 2023-01-21 13:14:43.515298: step: 1636/533, loss: 0.05012550204992294 2023-01-21 13:14:44.699961: step: 1640/533, loss: 0.0017783165676519275 2023-01-21 13:14:45.891064: step: 1644/533, loss: 0.004736137576401234 2023-01-21 13:14:47.111707: step: 1648/533, loss: 0.0016430855030193925 2023-01-21 13:14:48.291576: step: 1652/533, loss: 0.0004444122314453125 2023-01-21 13:14:49.480999: step: 1656/533, loss: 0.006460190284997225 2023-01-21 13:14:50.679684: step: 1660/533, loss: 2.9468537832144648e-05 2023-01-21 13:14:51.871633: step: 1664/533, loss: 0.0019041537307202816 2023-01-21 13:14:53.011589: step: 1668/533, loss: 0.002951717469841242 2023-01-21 13:14:54.196524: step: 1672/533, loss: 8.411407907260582e-05 2023-01-21 13:14:55.410476: step: 1676/533, loss: 0.013868285343050957 2023-01-21 13:14:56.554171: step: 1680/533, loss: 0.0009562492487020791 2023-01-21 13:14:57.732999: step: 1684/533, loss: 0.0038997649680823088 2023-01-21 13:14:58.907846: step: 1688/533, loss: 4.501342846197076e-05 2023-01-21 13:15:00.102165: step: 1692/533, loss: 0.0030788423027843237 2023-01-21 13:15:01.263727: step: 1696/533, loss: 0.00024003982252907008 2023-01-21 13:15:02.452163: step: 1700/533, loss: 0.00046606065006926656 2023-01-21 13:15:03.706198: step: 1704/533, loss: 0.001483058906160295 2023-01-21 13:15:04.893101: step: 1708/533, loss: 0.0007295608520507812 2023-01-21 13:15:06.087082: step: 1712/533, loss: 0.026917647570371628 2023-01-21 13:15:07.278395: step: 1716/533, loss: 0.0028310774359852076 2023-01-21 13:15:08.445130: step: 1720/533, loss: 0.00023860932560637593 2023-01-21 13:15:09.614587: step: 1724/533, loss: 0.0017264842754229903 2023-01-21 13:15:10.821515: step: 1728/533, loss: 0.0003843307786155492 2023-01-21 13:15:12.029932: step: 1732/533, loss: 0.007222652435302734 2023-01-21 13:15:13.176165: step: 1736/533, loss: 3.395378735149279e-05 2023-01-21 13:15:14.327198: step: 1740/533, loss: 4.034042649436742e-05 2023-01-21 13:15:15.539594: step: 1744/533, loss: 0.0003303527773823589 2023-01-21 13:15:16.731998: step: 1748/533, loss: 3.986358569818549e-05 2023-01-21 13:15:17.899984: step: 1752/533, loss: 9.975433204090223e-05 2023-01-21 13:15:19.049460: step: 1756/533, loss: 0.015137768350541592 2023-01-21 13:15:20.216684: step: 1760/533, loss: 0.0007786750793457031 2023-01-21 13:15:21.366361: step: 1764/533, loss: 0.005710697267204523 2023-01-21 13:15:22.517013: step: 1768/533, loss: 1.6498564946232364e-05 2023-01-21 13:15:23.669280: step: 1772/533, loss: 0.0006136417505331337 2023-01-21 13:15:24.844742: step: 1776/533, loss: 2.059936559817288e-05 2023-01-21 13:15:25.998804: step: 1780/533, loss: 0.0007009506225585938 2023-01-21 13:15:27.194127: step: 1784/533, loss: 0.006754684261977673 2023-01-21 13:15:28.322442: step: 1788/533, loss: 0.0013906479580327868 2023-01-21 13:15:29.532864: step: 1792/533, loss: 0.0007092476007528603 2023-01-21 13:15:30.708191: step: 1796/533, loss: 0.02512378618121147 2023-01-21 13:15:31.880456: step: 1800/533, loss: 0.004793929867446423 2023-01-21 13:15:33.006212: step: 1804/533, loss: 2.9373170036706142e-05 2023-01-21 13:15:34.215203: step: 1808/533, loss: 0.005444717593491077 2023-01-21 13:15:35.391956: step: 1812/533, loss: 8.544922457076609e-05 2023-01-21 13:15:36.566390: step: 1816/533, loss: 0.000518798828125 2023-01-21 13:15:37.695102: step: 1820/533, loss: 0.013496875762939453 2023-01-21 13:15:38.878458: step: 1824/533, loss: 0.009257126599550247 2023-01-21 13:15:40.014028: step: 1828/533, loss: 0.013383245095610619 2023-01-21 13:15:41.229998: step: 1832/533, loss: 0.0001522064267192036 2023-01-21 13:15:42.415360: step: 1836/533, loss: 1.7642974853515625e-05 2023-01-21 13:15:43.580226: step: 1840/533, loss: 0.003491020295768976 2023-01-21 13:15:44.758685: step: 1844/533, loss: 0.005664658732712269 2023-01-21 13:15:45.929276: step: 1848/533, loss: 0.05427103117108345 2023-01-21 13:15:47.112192: step: 1852/533, loss: 0.03789882734417915 2023-01-21 13:15:48.330125: step: 1856/533, loss: 0.021619606763124466 2023-01-21 13:15:49.481639: step: 1860/533, loss: 0.015170765109360218 2023-01-21 13:15:50.662687: step: 1864/533, loss: 0.00214805593714118 2023-01-21 13:15:51.919275: step: 1868/533, loss: 0.0010038375621661544 2023-01-21 13:15:53.104137: step: 1872/533, loss: 0.0052665709517896175 2023-01-21 13:15:54.288186: step: 1876/533, loss: 0.017368890345096588 2023-01-21 13:15:55.476401: step: 1880/533, loss: 0.003692722413688898 2023-01-21 13:15:56.681521: step: 1884/533, loss: 0.06499912589788437 2023-01-21 13:15:57.880971: step: 1888/533, loss: 0.0008035659557208419 2023-01-21 13:15:59.060808: step: 1892/533, loss: 0.0019177437061443925 2023-01-21 13:16:00.254301: step: 1896/533, loss: 7.41958647267893e-05 2023-01-21 13:16:01.432291: step: 1900/533, loss: 0.001054143882356584 2023-01-21 13:16:02.603081: step: 1904/533, loss: 0.003231239505112171 2023-01-21 13:16:03.746457: step: 1908/533, loss: 0.0005390167352743447 2023-01-21 13:16:04.961296: step: 1912/533, loss: 0.000940561352763325 2023-01-21 13:16:06.161426: step: 1916/533, loss: 0.007178878877311945 2023-01-21 13:16:07.310556: step: 1920/533, loss: 0.0060096741653978825 2023-01-21 13:16:08.473282: step: 1924/533, loss: 0.009455871768295765 2023-01-21 13:16:09.647424: step: 1928/533, loss: 0.03778943791985512 2023-01-21 13:16:10.828570: step: 1932/533, loss: 0.0014142990112304688 2023-01-21 13:16:12.013937: step: 1936/533, loss: 0.002552604768425226 2023-01-21 13:16:13.185243: step: 1940/533, loss: 0.0017974853981286287 2023-01-21 13:16:14.346333: step: 1944/533, loss: 0.07040510326623917 2023-01-21 13:16:15.511211: step: 1948/533, loss: 0.0019392013782635331 2023-01-21 13:16:16.666153: step: 1952/533, loss: 0.00138511648401618 2023-01-21 13:16:17.873562: step: 1956/533, loss: 0.0006892204401083291 2023-01-21 13:16:19.073514: step: 1960/533, loss: 0.0014643669128417969 2023-01-21 13:16:20.238325: step: 1964/533, loss: 0.0018978120060637593 2023-01-21 13:16:21.387933: step: 1968/533, loss: 0.00013651848712470382 2023-01-21 13:16:22.558250: step: 1972/533, loss: 0.008381461724638939 2023-01-21 13:16:23.744174: step: 1976/533, loss: 0.0005361631629057229 2023-01-21 13:16:24.905632: step: 1980/533, loss: 0.002209186553955078 2023-01-21 13:16:26.102036: step: 1984/533, loss: 1.1405349141568877e-05 2023-01-21 13:16:27.253468: step: 1988/533, loss: 0.005408096592873335 2023-01-21 13:16:28.443132: step: 1992/533, loss: 0.0002956390380859375 2023-01-21 13:16:29.595728: step: 1996/533, loss: 0.0031279087997972965 2023-01-21 13:16:30.763164: step: 2000/533, loss: 0.0012835502857342362 2023-01-21 13:16:31.962825: step: 2004/533, loss: 0.0001204729123855941 2023-01-21 13:16:33.165155: step: 2008/533, loss: 0.0006654739263467491 2023-01-21 13:16:34.424135: step: 2012/533, loss: 0.013453292660415173 2023-01-21 13:16:35.611607: step: 2016/533, loss: 0.0010333069367334247 2023-01-21 13:16:36.820906: step: 2020/533, loss: 6.122589547885582e-05 2023-01-21 13:16:38.017282: step: 2024/533, loss: 0.0063915252685546875 2023-01-21 13:16:39.210368: step: 2028/533, loss: 0.0021713257301598787 2023-01-21 13:16:40.457425: step: 2032/533, loss: 0.09097137302160263 2023-01-21 13:16:41.602087: step: 2036/533, loss: 0.00018024444580078125 2023-01-21 13:16:42.765254: step: 2040/533, loss: 0.0010258674155920744 2023-01-21 13:16:43.927361: step: 2044/533, loss: 0.002700996585190296 2023-01-21 13:16:45.136669: step: 2048/533, loss: 2.117157055181451e-05 2023-01-21 13:16:46.307557: step: 2052/533, loss: 0.07578067481517792 2023-01-21 13:16:47.512136: step: 2056/533, loss: 3.581047349143773e-05 2023-01-21 13:16:48.689617: step: 2060/533, loss: 0.004752730950713158 2023-01-21 13:16:49.910468: step: 2064/533, loss: 0.020041847601532936 2023-01-21 13:16:51.129485: step: 2068/533, loss: 0.00010447502427268773 2023-01-21 13:16:52.317147: step: 2072/533, loss: 0.00010900497727561742 2023-01-21 13:16:53.503867: step: 2076/533, loss: 0.012491464614868164 2023-01-21 13:16:54.695254: step: 2080/533, loss: 0.0030254365410655737 2023-01-21 13:16:55.908710: step: 2084/533, loss: 0.01447000540792942 2023-01-21 13:16:57.048341: step: 2088/533, loss: 0.0015224457019940019 2023-01-21 13:16:58.188895: step: 2092/533, loss: 0.0006699622026644647 2023-01-21 13:16:59.383290: step: 2096/533, loss: 0.00600776681676507 2023-01-21 13:17:00.557414: step: 2100/533, loss: 0.018784618005156517 2023-01-21 13:17:01.772682: step: 2104/533, loss: 0.04771747440099716 2023-01-21 13:17:02.947812: step: 2108/533, loss: 0.008324814029037952 2023-01-21 13:17:04.162952: step: 2112/533, loss: 0.00017213821411132812 2023-01-21 13:17:05.317189: step: 2116/533, loss: 0.013968085870146751 2023-01-21 13:17:06.543582: step: 2120/533, loss: 0.0004980087396688759 2023-01-21 13:17:07.722942: step: 2124/533, loss: 3.42369094141759e-05 2023-01-21 13:17:08.904184: step: 2128/533, loss: 0.00022010803513694555 2023-01-21 13:17:10.038099: step: 2132/533, loss: 0.0018367290031164885 ================================================== Loss: 0.013 -------------------- Dev: {'event': {'p': 0.6161934805467929, 'r': 0.7802929427430093, 'f1': 0.6886016451233843}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Test: {'event': {'p': 0.643223819301848, 'r': 0.801150895140665, 'f1': 0.7135535307517085}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} 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': 25} Korean: {'event': {'p': 0.6140350877192983, 'r': 0.5555555555555556, 'f1': 0.5833333333333335}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Russian: {'event': {'p': 0.45714285714285713, 'r': 0.4444444444444444, 'f1': 0.4507042253521127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 13:17:51.260620: step: 4/533, loss: 0.017394160851836205 2023-01-21 13:17:52.402165: step: 8/533, loss: 0.0012245177058503032 2023-01-21 13:17:53.568281: step: 12/533, loss: 0.0001485824614064768 2023-01-21 13:17:54.749032: step: 16/533, loss: 2.899169885495212e-05 2023-01-21 13:17:55.900897: step: 20/533, loss: 0.0005673408741131425 2023-01-21 13:17:57.047041: step: 24/533, loss: 0.008204269222915173 2023-01-21 13:17:58.211572: step: 28/533, loss: 0.09262824058532715 2023-01-21 13:17:59.368236: step: 32/533, loss: 0.04372682422399521 2023-01-21 13:18:00.524626: step: 36/533, loss: 0.0021409988403320312 2023-01-21 13:18:01.679251: step: 40/533, loss: 0.00638427771627903 2023-01-21 13:18:02.869310: step: 44/533, loss: 0.014252089895308018 2023-01-21 13:18:04.044575: step: 48/533, loss: 2.384185791015625e-05 2023-01-21 13:18:05.191524: step: 52/533, loss: 0.0006363391294144094 2023-01-21 13:18:06.380963: step: 56/533, loss: 0.006514930631965399 2023-01-21 13:18:07.548979: step: 60/533, loss: 0.0009164810180664062 2023-01-21 13:18:08.736601: step: 64/533, loss: 0.00033988954965025187 2023-01-21 13:18:09.901720: step: 68/533, loss: 0.0011954307556152344 2023-01-21 13:18:11.088316: step: 72/533, loss: 0.0009373784414492548 2023-01-21 13:18:12.295881: step: 76/533, loss: 0.002245187759399414 2023-01-21 13:18:13.458796: step: 80/533, loss: 0.02531757391989231 2023-01-21 13:18:14.628120: step: 84/533, loss: 0.0006640434148721397 2023-01-21 13:18:15.784373: step: 88/533, loss: 0.009225464425981045 2023-01-21 13:18:16.979109: step: 92/533, loss: 0.0005470275646075606 2023-01-21 13:18:18.182426: step: 96/533, loss: 0.00023269654775504023 2023-01-21 13:18:19.389274: step: 100/533, loss: 0.0004094124014955014 2023-01-21 13:18:20.599034: step: 104/533, loss: 0.0009828568436205387 2023-01-21 13:18:21.781106: step: 108/533, loss: 0.018595505505800247 2023-01-21 13:18:22.958132: step: 112/533, loss: 6.0558319091796875e-05 2023-01-21 13:18:24.146988: step: 116/533, loss: -6.10351571594947e-06 2023-01-21 13:18:25.297347: step: 120/533, loss: 0.00015726088895462453 2023-01-21 13:18:26.441652: step: 124/533, loss: 0.0009961128234863281 2023-01-21 13:18:27.623533: step: 128/533, loss: 0.0007328033680096269 2023-01-21 13:18:28.795831: step: 132/533, loss: 8.497238013660535e-05 2023-01-21 13:18:29.986540: step: 136/533, loss: 0.004687499720603228 2023-01-21 13:18:31.181375: step: 140/533, loss: 0.023482512682676315 2023-01-21 13:18:32.314581: step: 144/533, loss: 0.03089323081076145 2023-01-21 13:18:33.496474: step: 148/533, loss: 3.0231476557673886e-05 2023-01-21 13:18:34.682933: step: 152/533, loss: 0.00011148453631903976 2023-01-21 13:18:35.858541: step: 156/533, loss: 0.010329199023544788 2023-01-21 13:18:37.012701: step: 160/533, loss: 0.012707972899079323 2023-01-21 13:18:38.181827: step: 164/533, loss: 0.0033180236350744963 2023-01-21 13:18:39.366134: step: 168/533, loss: 2.6798248654813506e-05 2023-01-21 13:18:40.579089: step: 172/533, loss: 0.001982307294383645 2023-01-21 13:18:41.764964: step: 176/533, loss: 5.187988426769152e-05 2023-01-21 13:18:42.990713: step: 180/533, loss: 0.0009992599952965975 2023-01-21 13:18:44.155133: step: 184/533, loss: 0.002347278641536832 2023-01-21 13:18:45.342394: step: 188/533, loss: 6.256103370105848e-05 2023-01-21 13:18:46.497325: step: 192/533, loss: 0.0021039010025560856 2023-01-21 13:18:47.712452: step: 196/533, loss: 0.007149219512939453 2023-01-21 13:18:48.883917: step: 200/533, loss: 0.0005437851068563759 2023-01-21 13:18:50.111640: step: 204/533, loss: 0.00038356782170012593 2023-01-21 13:18:51.254868: step: 208/533, loss: 7.152557373046875e-06 2023-01-21 13:18:52.500962: step: 212/533, loss: 0.00013747216144111007 2023-01-21 13:18:53.678759: step: 216/533, loss: 0.00045843125553801656 2023-01-21 13:18:54.861481: step: 220/533, loss: 0.0004973411560058594 2023-01-21 13:18:56.051358: step: 224/533, loss: 0.00015544892812613398 2023-01-21 13:18:57.192411: step: 228/533, loss: 4.6253204345703125e-05 2023-01-21 13:18:58.388646: step: 232/533, loss: 0.005884838290512562 2023-01-21 13:18:59.598265: step: 236/533, loss: 0.0001049041748046875 2023-01-21 13:19:00.754464: step: 240/533, loss: -5.91278057981981e-06 2023-01-21 13:19:01.949894: step: 244/533, loss: 0.017470359802246094 2023-01-21 13:19:03.085769: step: 248/533, loss: 1.1825562069134321e-05 2023-01-21 13:19:04.240457: step: 252/533, loss: 1.9073486612342094e-07 2023-01-21 13:19:05.440544: step: 256/533, loss: 2.689361645025201e-05 2023-01-21 13:19:06.605206: step: 260/533, loss: 0.0010174751514568925 2023-01-21 13:19:07.741945: step: 264/533, loss: 0.0001680374116403982 2023-01-21 13:19:08.889784: step: 268/533, loss: 1.0681153071345761e-05 2023-01-21 13:19:10.031692: step: 272/533, loss: 0.014096545986831188 2023-01-21 13:19:11.186387: step: 276/533, loss: 0.002058601239696145 2023-01-21 13:19:12.370119: step: 280/533, loss: 9.91821252682712e-06 2023-01-21 13:19:13.592562: step: 284/533, loss: 0.026860618963837624 2023-01-21 13:19:14.751908: step: 288/533, loss: 0.0037556171882897615 2023-01-21 13:19:15.932137: step: 292/533, loss: 0.004272937774658203 2023-01-21 13:19:17.154876: step: 296/533, loss: 9.727478754939511e-06 2023-01-21 13:19:18.445770: step: 300/533, loss: 3.395080420887098e-05 2023-01-21 13:19:19.637012: step: 304/533, loss: 0.0026064871344715357 2023-01-21 13:19:20.788900: step: 308/533, loss: 0.0009754180791787803 2023-01-21 13:19:21.962637: step: 312/533, loss: 1.678466833254788e-05 2023-01-21 13:19:23.110773: step: 316/533, loss: 0.001332950545474887 2023-01-21 13:19:24.276434: step: 320/533, loss: 0.00213794712908566 2023-01-21 13:19:25.451029: step: 324/533, loss: 2.5749204723979346e-06 2023-01-21 13:19:26.643879: step: 328/533, loss: 0.009528731927275658 2023-01-21 13:19:27.776490: step: 332/533, loss: 0.00036535263643600047 2023-01-21 13:19:28.907504: step: 336/533, loss: 0.022148799151182175 2023-01-21 13:19:30.089912: step: 340/533, loss: 0.07264023274183273 2023-01-21 13:19:31.280037: step: 344/533, loss: 0.00027370452880859375 2023-01-21 13:19:32.446997: step: 348/533, loss: 0.0017305373912677169 2023-01-21 13:19:33.637118: step: 352/533, loss: 0.0002513885556254536 2023-01-21 13:19:34.840705: step: 356/533, loss: 0.00596542377024889 2023-01-21 13:19:36.039306: step: 360/533, loss: 0.0019187451107427478 2023-01-21 13:19:37.232564: step: 364/533, loss: 0.0021162035409361124 2023-01-21 13:19:38.404995: step: 368/533, loss: 0.011855793185532093 2023-01-21 13:19:39.585662: step: 372/533, loss: 0.021173572167754173 2023-01-21 13:19:40.739185: step: 376/533, loss: -4.38690176451928e-06 2023-01-21 13:19:41.922851: step: 380/533, loss: 0.0404851920902729 2023-01-21 13:19:43.162271: step: 384/533, loss: 0.04347210004925728 2023-01-21 13:19:44.330541: step: 388/533, loss: 0.005990028381347656 2023-01-21 13:19:45.545328: step: 392/533, loss: 2.059936559817288e-05 2023-01-21 13:19:46.704469: step: 396/533, loss: 0.010973549447953701 2023-01-21 13:19:47.895030: step: 400/533, loss: 5.455017162603326e-05 2023-01-21 13:19:49.063729: step: 404/533, loss: 7.45773286325857e-05 2023-01-21 13:19:50.240259: step: 408/533, loss: 0.00027141571626998484 2023-01-21 13:19:51.411089: step: 412/533, loss: 0.006593286991119385 2023-01-21 13:19:52.648063: step: 416/533, loss: 0.00036237240419723094 2023-01-21 13:19:53.771708: step: 420/533, loss: 1.7833710444392636e-05 2023-01-21 13:19:54.953388: step: 424/533, loss: 0.011798190884292126 2023-01-21 13:19:56.147751: step: 428/533, loss: 0.0002721786731854081 2023-01-21 13:19:57.312463: step: 432/533, loss: 0.4555656909942627 2023-01-21 13:19:58.462217: step: 436/533, loss: -1.5616417385899695e-06 2023-01-21 13:19:59.686745: step: 440/533, loss: 0.0016931057907640934 2023-01-21 13:20:00.889033: step: 444/533, loss: 0.01959824748337269 2023-01-21 13:20:02.104207: step: 448/533, loss: 0.0029212951194494963 2023-01-21 13:20:03.305369: step: 452/533, loss: 0.012137794867157936 2023-01-21 13:20:04.536110: step: 456/533, loss: 0.019290734082460403 2023-01-21 13:20:05.693711: step: 460/533, loss: 0.0007053375011309981 2023-01-21 13:20:06.831496: step: 464/533, loss: 0.00020515918731689453 2023-01-21 13:20:08.003213: step: 468/533, loss: 0.00018568038649391383 2023-01-21 13:20:09.184123: step: 472/533, loss: 0.7206413745880127 2023-01-21 13:20:10.379499: step: 476/533, loss: 8.468627493130043e-05 2023-01-21 13:20:11.561853: step: 480/533, loss: 0.004154682159423828 2023-01-21 13:20:12.793722: step: 484/533, loss: 0.03941326215863228 2023-01-21 13:20:13.966489: step: 488/533, loss: 0.004136324394494295 2023-01-21 13:20:15.134103: step: 492/533, loss: 0.0002960205019917339 2023-01-21 13:20:16.303044: step: 496/533, loss: 0.0004982947721146047 2023-01-21 13:20:17.488200: step: 500/533, loss: 0.076807402074337 2023-01-21 13:20:18.655783: step: 504/533, loss: 0.00012149810936534777 2023-01-21 13:20:19.838641: step: 508/533, loss: 0.004178046714514494 2023-01-21 13:20:21.004083: step: 512/533, loss: 0.0025717734824866056 2023-01-21 13:20:22.170891: step: 516/533, loss: 0.03859663009643555 2023-01-21 13:20:23.315704: step: 520/533, loss: 0.0017368317348882556 2023-01-21 13:20:24.609246: step: 524/533, loss: 2.593994213384576e-05 2023-01-21 13:20:25.836775: step: 528/533, loss: 0.00013647080049850047 2023-01-21 13:20:27.038901: step: 532/533, loss: 0.01704278215765953 2023-01-21 13:20:28.229175: step: 536/533, loss: 0.0002662658807821572 2023-01-21 13:20:29.379023: step: 540/533, loss: 0.00027885439340025187 2023-01-21 13:20:30.580915: step: 544/533, loss: 0.0009891510708257556 2023-01-21 13:20:31.770411: step: 548/533, loss: 2.746581958490424e-05 2023-01-21 13:20:32.946289: step: 552/533, loss: 0.00013837814913131297 2023-01-21 13:20:34.134692: step: 556/533, loss: 0.02695159986615181 2023-01-21 13:20:35.298789: step: 560/533, loss: 0.000798988388851285 2023-01-21 13:20:36.486734: step: 564/533, loss: 0.01008386630564928 2023-01-21 13:20:37.678289: step: 568/533, loss: 0.0001522064267192036 2023-01-21 13:20:38.843957: step: 572/533, loss: 0.010600757785141468 2023-01-21 13:20:39.969675: step: 576/533, loss: 0.00015230178541969508 2023-01-21 13:20:41.139965: step: 580/533, loss: 0.0007265090825967491 2023-01-21 13:20:42.333796: step: 584/533, loss: 0.5062990188598633 2023-01-21 13:20:43.492104: step: 588/533, loss: 0.06403694301843643 2023-01-21 13:20:44.691487: step: 592/533, loss: 0.0001540184166515246 2023-01-21 13:20:45.866454: step: 596/533, loss: 0.0007529258728027344 2023-01-21 13:20:47.025103: step: 600/533, loss: 0.03568553924560547 2023-01-21 13:20:48.198125: step: 604/533, loss: 0.02764911763370037 2023-01-21 13:20:49.363980: step: 608/533, loss: 0.0014818668132647872 2023-01-21 13:20:50.586678: step: 612/533, loss: 0.0014579773414880037 2023-01-21 13:20:51.750873: step: 616/533, loss: 4.117488788324408e-05 2023-01-21 13:20:52.978736: step: 620/533, loss: 0.017935562878847122 2023-01-21 13:20:54.153225: step: 624/533, loss: 0.0016592026222497225 2023-01-21 13:20:55.307682: step: 628/533, loss: 0.0003465652698650956 2023-01-21 13:20:56.471326: step: 632/533, loss: 0.03862953558564186 2023-01-21 13:20:57.668381: step: 636/533, loss: 0.012893295846879482 2023-01-21 13:20:58.850120: step: 640/533, loss: 0.006782150361686945 2023-01-21 13:21:00.012549: step: 644/533, loss: 9.250640869140625e-05 2023-01-21 13:21:01.161943: step: 648/533, loss: 0.0032339096069335938 2023-01-21 13:21:02.352439: step: 652/533, loss: 0.08789139240980148 2023-01-21 13:21:03.540088: step: 656/533, loss: 0.0018465996254235506 2023-01-21 13:21:04.736143: step: 660/533, loss: 0.01641969569027424 2023-01-21 13:21:05.916368: step: 664/533, loss: 1.4019013178767636e-05 2023-01-21 13:21:07.112531: step: 668/533, loss: 0.006941699888557196 2023-01-21 13:21:08.282616: step: 672/533, loss: 0.00023593902005814016 2023-01-21 13:21:09.489916: step: 676/533, loss: 0.009453010745346546 2023-01-21 13:21:10.705793: step: 680/533, loss: 0.0004684448358602822 2023-01-21 13:21:11.888180: step: 684/533, loss: 6.790161569369957e-05 2023-01-21 13:21:13.067189: step: 688/533, loss: 0.013985968194901943 2023-01-21 13:21:14.272659: step: 692/533, loss: 0.0026304246857762337 2023-01-21 13:21:15.469926: step: 696/533, loss: 0.00792837142944336 2023-01-21 13:21:16.706798: step: 700/533, loss: 0.015484190545976162 2023-01-21 13:21:17.877773: step: 704/533, loss: 0.00025806427584029734 2023-01-21 13:21:19.087317: step: 708/533, loss: 2.245903124276083e-05 2023-01-21 13:21:20.281018: step: 712/533, loss: 0.03350868448615074 2023-01-21 13:21:21.469321: step: 716/533, loss: 4.6253204345703125e-05 2023-01-21 13:21:22.601201: step: 720/533, loss: 0.02397127076983452 2023-01-21 13:21:23.748507: step: 724/533, loss: 0.0016825676430016756 2023-01-21 13:21:24.949972: step: 728/533, loss: 9.088516526389867e-05 2023-01-21 13:21:26.163014: step: 732/533, loss: 0.013457870110869408 2023-01-21 13:21:27.329391: step: 736/533, loss: 0.0016349792713299394 2023-01-21 13:21:28.493935: step: 740/533, loss: 0.001959133194759488 2023-01-21 13:21:29.705687: step: 744/533, loss: 0.0008723259088583291 2023-01-21 13:21:30.901718: step: 748/533, loss: 0.0002521514834370464 2023-01-21 13:21:32.090370: step: 752/533, loss: 0.0005171776283532381 2023-01-21 13:21:33.241784: step: 756/533, loss: 0.024413015693426132 2023-01-21 13:21:34.376312: step: 760/533, loss: 2.098083541568485e-06 2023-01-21 13:21:35.541806: step: 764/533, loss: 0.00033254624577239156 2023-01-21 13:21:36.728350: step: 768/533, loss: 7.4386593951203395e-06 2023-01-21 13:21:37.952200: step: 772/533, loss: 0.0003940582391805947 2023-01-21 13:21:39.133669: step: 776/533, loss: 0.0003329276805743575 2023-01-21 13:21:40.329792: step: 780/533, loss: 0.0011396408081054688 2023-01-21 13:21:41.506016: step: 784/533, loss: 0.08072586357593536 2023-01-21 13:21:42.709898: step: 788/533, loss: 0.0035265921615064144 2023-01-21 13:21:43.925144: step: 792/533, loss: 0.0005336761241778731 2023-01-21 13:21:45.075913: step: 796/533, loss: 0.009946012869477272 2023-01-21 13:21:46.250332: step: 800/533, loss: 0.0023246766068041325 2023-01-21 13:21:47.425941: step: 804/533, loss: 0.004134082701057196 2023-01-21 13:21:48.622376: step: 808/533, loss: 0.00019073486328125 2023-01-21 13:21:49.857307: step: 812/533, loss: 0.03328218311071396 2023-01-21 13:21:51.052940: step: 816/533, loss: 0.01727275922894478 2023-01-21 13:21:52.265265: step: 820/533, loss: 0.12867556512355804 2023-01-21 13:21:53.415230: step: 824/533, loss: 0.0018402099376544356 2023-01-21 13:21:54.638475: step: 828/533, loss: 0.05869045481085777 2023-01-21 13:21:55.835846: step: 832/533, loss: 0.038787271827459335 2023-01-21 13:21:57.024878: step: 836/533, loss: 0.010917281731963158 2023-01-21 13:21:58.133526: step: 840/533, loss: 0.0007252693176269531 2023-01-21 13:21:59.315767: step: 844/533, loss: 0.9349040985107422 2023-01-21 13:22:00.488507: step: 848/533, loss: 8.03947405074723e-05 2023-01-21 13:22:01.679286: step: 852/533, loss: 0.0007444858201779425 2023-01-21 13:22:02.880443: step: 856/533, loss: 0.015094757080078125 2023-01-21 13:22:04.027913: step: 860/533, loss: 0.0014700889587402344 2023-01-21 13:22:05.194919: step: 864/533, loss: 2.403259350103326e-05 2023-01-21 13:22:06.333138: step: 868/533, loss: 0.0002386093110544607 2023-01-21 13:22:07.541499: step: 872/533, loss: 1.3208388736529741e-05 2023-01-21 13:22:08.764972: step: 876/533, loss: 0.0007001876947470009 2023-01-21 13:22:09.949336: step: 880/533, loss: 0.0047089578583836555 2023-01-21 13:22:11.128536: step: 884/533, loss: 3.814697265625e-06 2023-01-21 13:22:12.297339: step: 888/533, loss: 7.781983003951609e-05 2023-01-21 13:22:13.438272: step: 892/533, loss: 0.020638607442378998 2023-01-21 13:22:14.591661: step: 896/533, loss: 0.0002821922244038433 2023-01-21 13:22:15.717546: step: 900/533, loss: 0.0010000229813158512 2023-01-21 13:22:16.900110: step: 904/533, loss: 0.013739585876464844 2023-01-21 13:22:18.058541: step: 908/533, loss: 0.0005388259887695312 2023-01-21 13:22:19.284982: step: 912/533, loss: 0.0003123283386230469 2023-01-21 13:22:20.467762: step: 916/533, loss: 0.0003717422659974545 2023-01-21 13:22:21.741696: step: 920/533, loss: 0.0005903244018554688 2023-01-21 13:22:22.876624: step: 924/533, loss: 0.01337890699505806 2023-01-21 13:22:24.037426: step: 928/533, loss: 0.00024890899658203125 2023-01-21 13:22:25.175426: step: 932/533, loss: 5.340576535672881e-06 2023-01-21 13:22:26.366486: step: 936/533, loss: 0.0008636951679363847 2023-01-21 13:22:27.557079: step: 940/533, loss: 0.001029872801154852 2023-01-21 13:22:28.699420: step: 944/533, loss: 0.00025329593336209655 2023-01-21 13:22:29.854838: step: 948/533, loss: 1.0681153071345761e-05 2023-01-21 13:22:31.035191: step: 952/533, loss: 1.0013579867518274e-06 2023-01-21 13:22:32.226925: step: 956/533, loss: 1.0871887752728071e-05 2023-01-21 13:22:33.394059: step: 960/533, loss: 0.599378764629364 2023-01-21 13:22:34.591474: step: 964/533, loss: 0.045965004712343216 2023-01-21 13:22:35.771180: step: 968/533, loss: 0.006877326872199774 2023-01-21 13:22:36.963188: step: 972/533, loss: 0.0017168044578284025 2023-01-21 13:22:38.145153: step: 976/533, loss: 0.015292549505829811 2023-01-21 13:22:39.315935: step: 980/533, loss: 0.010364246554672718 2023-01-21 13:22:40.478418: step: 984/533, loss: 0.0005341529613360763 2023-01-21 13:22:41.615668: step: 988/533, loss: 0.0003478527069091797 2023-01-21 13:22:42.753129: step: 992/533, loss: 0.0008901596302166581 2023-01-21 13:22:43.896164: step: 996/533, loss: 0.051703836768865585 2023-01-21 13:22:45.082565: step: 1000/533, loss: 2.6226043701171875e-05 2023-01-21 13:22:46.324368: step: 1004/533, loss: 0.0009669304126873612 2023-01-21 13:22:47.476371: step: 1008/533, loss: 0.002134704729542136 2023-01-21 13:22:48.666288: step: 1012/533, loss: 0.005488729570060968 2023-01-21 13:22:49.886982: step: 1016/533, loss: 0.0002879142703022808 2023-01-21 13:22:51.103586: step: 1020/533, loss: 0.0007899284246377647 2023-01-21 13:22:52.247751: step: 1024/533, loss: 0.00017375947209075093 2023-01-21 13:22:53.432678: step: 1028/533, loss: 0.0008118629921227694 2023-01-21 13:22:54.637401: step: 1032/533, loss: 0.01477889996021986 2023-01-21 13:22:55.840498: step: 1036/533, loss: 0.0010653972858563066 2023-01-21 13:22:57.018365: step: 1040/533, loss: 0.004500294104218483 2023-01-21 13:22:58.181684: step: 1044/533, loss: 0.0009019851568154991 2023-01-21 13:22:59.335754: step: 1048/533, loss: 0.034046366810798645 2023-01-21 13:23:00.533605: step: 1052/533, loss: 0.0005244255298748612 2023-01-21 13:23:01.757763: step: 1056/533, loss: 0.0019186020363122225 2023-01-21 13:23:02.964308: step: 1060/533, loss: 1.8930435544461943e-05 2023-01-21 13:23:04.160787: step: 1064/533, loss: 0.0006726741557940841 2023-01-21 13:23:05.322610: step: 1068/533, loss: 0.0031662462279200554 2023-01-21 13:23:06.539862: step: 1072/533, loss: 0.0001829147367971018 2023-01-21 13:23:07.737568: step: 1076/533, loss: 0.02013077773153782 2023-01-21 13:23:08.892835: step: 1080/533, loss: 7.314682443393394e-05 2023-01-21 13:23:10.073020: step: 1084/533, loss: 0.006153106689453125 2023-01-21 13:23:11.232875: step: 1088/533, loss: 0.00039653776912018657 2023-01-21 13:23:12.408899: step: 1092/533, loss: 0.0010359525913372636 2023-01-21 13:23:13.557729: step: 1096/533, loss: 1.1587142580538057e-05 2023-01-21 13:23:14.678379: step: 1100/533, loss: 0.011453533545136452 2023-01-21 13:23:15.863522: step: 1104/533, loss: 0.004266834352165461 2023-01-21 13:23:17.049848: step: 1108/533, loss: 0.6015817523002625 2023-01-21 13:23:18.246789: step: 1112/533, loss: 0.00026664737379178405 2023-01-21 13:23:19.436986: step: 1116/533, loss: 5.607604907709174e-05 2023-01-21 13:23:20.651802: step: 1120/533, loss: 0.03093128278851509 2023-01-21 13:23:21.850700: step: 1124/533, loss: 0.46698227524757385 2023-01-21 13:23:23.073107: step: 1128/533, loss: 0.4058647155761719 2023-01-21 13:23:24.258968: step: 1132/533, loss: 0.0016777992714196444 2023-01-21 13:23:25.411651: step: 1136/533, loss: 0.00014381408982444555 2023-01-21 13:23:26.608168: step: 1140/533, loss: 0.013336563482880592 2023-01-21 13:23:27.806585: step: 1144/533, loss: 0.0003948211669921875 2023-01-21 13:23:29.011900: step: 1148/533, loss: 8.20159948489163e-06 2023-01-21 13:23:30.174110: step: 1152/533, loss: 7.953644671943039e-05 2023-01-21 13:23:31.404594: step: 1156/533, loss: 6.70433000777848e-05 2023-01-21 13:23:32.559753: step: 1160/533, loss: 0.20153702795505524 2023-01-21 13:23:33.745744: step: 1164/533, loss: 0.02801218070089817 2023-01-21 13:23:34.879941: step: 1168/533, loss: 0.051758669316768646 2023-01-21 13:23:36.043575: step: 1172/533, loss: 0.007121944800019264 2023-01-21 13:23:37.199408: step: 1176/533, loss: 4.5442582631949335e-05 2023-01-21 13:23:38.420458: step: 1180/533, loss: 0.005846786312758923 2023-01-21 13:23:39.584425: step: 1184/533, loss: 0.0016781806480139494 2023-01-21 13:23:40.782739: step: 1188/533, loss: 0.00016689300537109375 2023-01-21 13:23:41.973775: step: 1192/533, loss: 0.012545204721391201 2023-01-21 13:23:43.149093: step: 1196/533, loss: 0.0010945319663733244 2023-01-21 13:23:44.358574: step: 1200/533, loss: 0.03044452704489231 2023-01-21 13:23:45.593325: step: 1204/533, loss: 0.005610752385109663 2023-01-21 13:23:46.763179: step: 1208/533, loss: 4.5347213017521426e-05 2023-01-21 13:23:47.968389: step: 1212/533, loss: 0.0028774261008948088 2023-01-21 13:23:49.154732: step: 1216/533, loss: 0.029937315732240677 2023-01-21 13:23:50.331785: step: 1220/533, loss: 0.00047178269596770406 2023-01-21 13:23:51.518016: step: 1224/533, loss: 2.727508581301663e-05 2023-01-21 13:23:52.699151: step: 1228/533, loss: 1.9836426872643642e-05 2023-01-21 13:23:53.901117: step: 1232/533, loss: 0.00942077673971653 2023-01-21 13:23:55.054588: step: 1236/533, loss: 0.0005599975702352822 2023-01-21 13:23:56.249005: step: 1240/533, loss: 0.03793029487133026 2023-01-21 13:23:57.417827: step: 1244/533, loss: 0.001160144922323525 2023-01-21 13:23:58.620661: step: 1248/533, loss: 0.0005642891046591103 2023-01-21 13:23:59.818541: step: 1252/533, loss: 0.12300319969654083 2023-01-21 13:24:01.017851: step: 1256/533, loss: 0.0025081634521484375 2023-01-21 13:24:02.168624: step: 1260/533, loss: 0.00015316010103560984 2023-01-21 13:24:03.362714: step: 1264/533, loss: 0.0003952980041503906 2023-01-21 13:24:04.547695: step: 1268/533, loss: 0.04303998872637749 2023-01-21 13:24:05.714064: step: 1272/533, loss: 0.009730529971420765 2023-01-21 13:24:06.898645: step: 1276/533, loss: 0.011086845770478249 2023-01-21 13:24:08.090175: step: 1280/533, loss: 0.000935459160245955 2023-01-21 13:24:09.268653: step: 1284/533, loss: 8.659363084007055e-05 2023-01-21 13:24:10.430053: step: 1288/533, loss: 0.040098000317811966 2023-01-21 13:24:11.621771: step: 1292/533, loss: 0.026552582159638405 2023-01-21 13:24:12.786550: step: 1296/533, loss: 1.6331673577951733e-06 2023-01-21 13:24:13.957120: step: 1300/533, loss: 0.00018215179443359375 2023-01-21 13:24:15.131693: step: 1304/533, loss: 1.010894811770413e-05 2023-01-21 13:24:16.324776: step: 1308/533, loss: 0.0017570494674146175 2023-01-21 13:24:17.450820: step: 1312/533, loss: 0.0001998901425395161 2023-01-21 13:24:18.616546: step: 1316/533, loss: 0.00016136169142555445 2023-01-21 13:24:19.834175: step: 1320/533, loss: 0.013141632080078125 2023-01-21 13:24:20.988605: step: 1324/533, loss: 0.0237717404961586 2023-01-21 13:24:22.119919: step: 1328/533, loss: 0.0001028060942189768 2023-01-21 13:24:23.325983: step: 1332/533, loss: 1.3065337952866685e-05 2023-01-21 13:24:24.479207: step: 1336/533, loss: 0.0016981124645099044 2023-01-21 13:24:25.626640: step: 1340/533, loss: 5.855560448253527e-05 2023-01-21 13:24:26.821502: step: 1344/533, loss: 0.001315641449764371 2023-01-21 13:24:27.992312: step: 1348/533, loss: 0.015293121337890625 2023-01-21 13:24:29.169759: step: 1352/533, loss: 8.449554297840223e-05 2023-01-21 13:24:30.365412: step: 1356/533, loss: 0.01225423812866211 2023-01-21 13:24:31.572816: step: 1360/533, loss: 0.0001655578671488911 2023-01-21 13:24:32.749288: step: 1364/533, loss: 4.024505687993951e-05 2023-01-21 13:24:33.901225: step: 1368/533, loss: 0.0024289607536047697 2023-01-21 13:24:35.121457: step: 1372/533, loss: 5.9700014389818534e-05 2023-01-21 13:24:36.307669: step: 1376/533, loss: 0.001974773360416293 2023-01-21 13:24:37.514228: step: 1380/533, loss: 3.1471254260395654e-06 2023-01-21 13:24:38.706929: step: 1384/533, loss: 0.0027975081466138363 2023-01-21 13:24:39.914608: step: 1388/533, loss: 1.1682510375976562e-05 2023-01-21 13:24:41.058451: step: 1392/533, loss: 0.0029973983764648438 2023-01-21 13:24:42.203987: step: 1396/533, loss: 0.0010484710801392794 2023-01-21 13:24:43.400154: step: 1400/533, loss: 0.032292939722537994 2023-01-21 13:24:44.566644: step: 1404/533, loss: 0.05029275640845299 2023-01-21 13:24:45.762077: step: 1408/533, loss: 0.0003791809140238911 2023-01-21 13:24:46.938055: step: 1412/533, loss: 1.6117097402457148e-05 2023-01-21 13:24:48.078018: step: 1416/533, loss: 0.0002233505219919607 2023-01-21 13:24:49.269947: step: 1420/533, loss: 2.6798248654813506e-05 2023-01-21 13:24:50.453993: step: 1424/533, loss: 0.07446031272411346 2023-01-21 13:24:51.633404: step: 1428/533, loss: 0.11768770962953568 2023-01-21 13:24:52.814641: step: 1432/533, loss: 0.02891254611313343 2023-01-21 13:24:54.021674: step: 1436/533, loss: 1.869201696536038e-05 2023-01-21 13:24:55.190085: step: 1440/533, loss: 8.401871309615672e-05 2023-01-21 13:24:56.393196: step: 1444/533, loss: 0.038555338978767395 2023-01-21 13:24:57.523547: step: 1448/533, loss: 0.0002514839288778603 2023-01-21 13:24:58.707305: step: 1452/533, loss: 4.138946678722277e-05 2023-01-21 13:24:59.875205: step: 1456/533, loss: 0.01672077178955078 2023-01-21 13:25:01.062564: step: 1460/533, loss: 0.010339641943573952 2023-01-21 13:25:02.262779: step: 1464/533, loss: 0.0001712799130473286 2023-01-21 13:25:03.488020: step: 1468/533, loss: 0.06185588985681534 2023-01-21 13:25:04.625227: step: 1472/533, loss: 1.0013580322265625e-05 2023-01-21 13:25:05.782339: step: 1476/533, loss: 0.008479214273393154 2023-01-21 13:25:06.949572: step: 1480/533, loss: 0.0011293411953374743 2023-01-21 13:25:08.111690: step: 1484/533, loss: 0.02370147779583931 2023-01-21 13:25:09.285330: step: 1488/533, loss: 0.0016771316295489669 2023-01-21 13:25:10.455249: step: 1492/533, loss: 0.0001272201625397429 2023-01-21 13:25:11.629017: step: 1496/533, loss: 0.009363556280732155 2023-01-21 13:25:12.794953: step: 1500/533, loss: 0.00040836335392668843 2023-01-21 13:25:13.914694: step: 1504/533, loss: 0.00010375976853538305 2023-01-21 13:25:15.141286: step: 1508/533, loss: 0.0006948471418581903 2023-01-21 13:25:16.327120: step: 1512/533, loss: 0.00017180443683173507 2023-01-21 13:25:17.502412: step: 1516/533, loss: 0.0001964569091796875 2023-01-21 13:25:18.633937: step: 1520/533, loss: 0.0019647120498120785 2023-01-21 13:25:19.856708: step: 1524/533, loss: 4.310607982915826e-05 2023-01-21 13:25:21.060740: step: 1528/533, loss: 3.490447852527723e-05 2023-01-21 13:25:22.263156: step: 1532/533, loss: 0.00022830962552689016 2023-01-21 13:25:23.385799: step: 1536/533, loss: 0.0016679763793945312 2023-01-21 13:25:24.541902: step: 1540/533, loss: 0.002149868058040738 2023-01-21 13:25:25.756584: step: 1544/533, loss: 0.00040092470590025187 2023-01-21 13:25:26.920175: step: 1548/533, loss: 0.006554603576660156 2023-01-21 13:25:28.100521: step: 1552/533, loss: 5.722046125811175e-07 2023-01-21 13:25:29.231388: step: 1556/533, loss: 0.0026386261451989412 2023-01-21 13:25:30.381528: step: 1560/533, loss: 0.002605438232421875 2023-01-21 13:25:31.564082: step: 1564/533, loss: 0.007409798912703991 2023-01-21 13:25:32.765377: step: 1568/533, loss: 2.117157055181451e-05 2023-01-21 13:25:33.955151: step: 1572/533, loss: 0.0017765044467523694 2023-01-21 13:25:35.122241: step: 1576/533, loss: 0.014275645837187767 2023-01-21 13:25:36.255821: step: 1580/533, loss: 7.872581045376137e-05 2023-01-21 13:25:37.411436: step: 1584/533, loss: 2.498626781743951e-05 2023-01-21 13:25:38.588811: step: 1588/533, loss: 0.00115795130841434 2023-01-21 13:25:39.763210: step: 1592/533, loss: 5.145073009771295e-05 2023-01-21 13:25:40.997176: step: 1596/533, loss: 0.0002783775271382183 2023-01-21 13:25:42.153948: step: 1600/533, loss: 0.0013859749305993319 2023-01-21 13:25:43.340589: step: 1604/533, loss: 0.09344358742237091 2023-01-21 13:25:44.500361: step: 1608/533, loss: 0.0003075599670410156 2023-01-21 13:25:45.705336: step: 1612/533, loss: 0.0008344650268554688 2023-01-21 13:25:46.899155: step: 1616/533, loss: 0.001014804933220148 2023-01-21 13:25:48.081455: step: 1620/533, loss: 0.0007132529863156378 2023-01-21 13:25:49.254804: step: 1624/533, loss: 9.689331636764109e-05 2023-01-21 13:25:50.472296: step: 1628/533, loss: 0.5569049715995789 2023-01-21 13:25:51.652653: step: 1632/533, loss: 0.04131278768181801 2023-01-21 13:25:52.823184: step: 1636/533, loss: 0.0006598949548788369 2023-01-21 13:25:54.006919: step: 1640/533, loss: 0.8051006197929382 2023-01-21 13:25:55.159180: step: 1644/533, loss: 0.0003419399436097592 2023-01-21 13:25:56.323137: step: 1648/533, loss: 0.00048828125 2023-01-21 13:25:57.513677: step: 1652/533, loss: 0.015282916836440563 2023-01-21 13:25:58.715358: step: 1656/533, loss: 0.0034267425071448088 2023-01-21 13:25:59.854777: step: 1660/533, loss: 0.00016379357839468867 2023-01-21 13:26:01.025914: step: 1664/533, loss: 0.00534820556640625 2023-01-21 13:26:02.186371: step: 1668/533, loss: 0.00021829605975653976 2023-01-21 13:26:03.353828: step: 1672/533, loss: 0.004005813971161842 2023-01-21 13:26:04.536584: step: 1676/533, loss: -1.316070574830519e-05 2023-01-21 13:26:05.675579: step: 1680/533, loss: 4.348754737293348e-05 2023-01-21 13:26:06.899352: step: 1684/533, loss: 5.378723290050402e-05 2023-01-21 13:26:08.054269: step: 1688/533, loss: 0.004181433003395796 2023-01-21 13:26:09.221656: step: 1692/533, loss: 5.292892637953628e-06 2023-01-21 13:26:10.394713: step: 1696/533, loss: 0.0015766143333166838 2023-01-21 13:26:11.571506: step: 1700/533, loss: 0.0006336212391033769 2023-01-21 13:26:12.737970: step: 1704/533, loss: 0.00013790131197310984 2023-01-21 13:26:13.882268: step: 1708/533, loss: 0.00092315673828125 2023-01-21 13:26:15.053574: step: 1712/533, loss: 0.009691429324448109 2023-01-21 13:26:16.193665: step: 1716/533, loss: 0.00014019013906363398 2023-01-21 13:26:17.341865: step: 1720/533, loss: 0.0008672714466229081 2023-01-21 13:26:18.479603: step: 1724/533, loss: 2.9087066195643274e-06 2023-01-21 13:26:19.653646: step: 1728/533, loss: 3.4141543437726796e-05 2023-01-21 13:26:20.878378: step: 1732/533, loss: 7.724762690486386e-05 2023-01-21 13:26:22.077288: step: 1736/533, loss: 0.0018504143226891756 2023-01-21 13:26:23.226366: step: 1740/533, loss: 0.0002808570861816406 2023-01-21 13:26:24.370854: step: 1744/533, loss: 0.03470306470990181 2023-01-21 13:26:25.567914: step: 1748/533, loss: 0.0001253128139069304 2023-01-21 13:26:26.777511: step: 1752/533, loss: 0.0007068634149618447 2023-01-21 13:26:27.996415: step: 1756/533, loss: 3.61442580469884e-05 2023-01-21 13:26:29.131947: step: 1760/533, loss: 3.25202927342616e-05 2023-01-21 13:26:30.335918: step: 1764/533, loss: 9.241104271495715e-05 2023-01-21 13:26:31.520464: step: 1768/533, loss: 0.0008534431108273566 2023-01-21 13:26:32.702855: step: 1772/533, loss: 0.004237615969032049 2023-01-21 13:26:33.903587: step: 1776/533, loss: 0.005154228303581476 2023-01-21 13:26:35.051512: step: 1780/533, loss: 0.30229464173316956 2023-01-21 13:26:36.217372: step: 1784/533, loss: 0.0005659103626385331 2023-01-21 13:26:37.363714: step: 1788/533, loss: 0.0009385109296999872 2023-01-21 13:26:38.496569: step: 1792/533, loss: 2.6941299438476562e-05 2023-01-21 13:26:39.695700: step: 1796/533, loss: 0.0019096375908702612 2023-01-21 13:26:40.850318: step: 1800/533, loss: 0.00014696121797896922 2023-01-21 13:26:42.037397: step: 1804/533, loss: 0.00023689270892646164 2023-01-21 13:26:43.208307: step: 1808/533, loss: 8.39233416627394e-06 2023-01-21 13:26:44.351012: step: 1812/533, loss: 0.030894756317138672 2023-01-21 13:26:45.522918: step: 1816/533, loss: 0.0019169808365404606 2023-01-21 13:26:46.691675: step: 1820/533, loss: -1.106262243411038e-05 2023-01-21 13:26:47.923360: step: 1824/533, loss: 0.0003183365042787045 2023-01-21 13:26:49.176380: step: 1828/533, loss: 6.50405854685232e-05 2023-01-21 13:26:50.349445: step: 1832/533, loss: 0.0013456344604492188 2023-01-21 13:26:51.510372: step: 1836/533, loss: 6.961822509765625e-05 2023-01-21 13:26:52.683147: step: 1840/533, loss: 0.017479753121733665 2023-01-21 13:26:53.835491: step: 1844/533, loss: 5.970001075183973e-05 2023-01-21 13:26:54.977732: step: 1848/533, loss: -4.95910626341356e-06 2023-01-21 13:26:56.198818: step: 1852/533, loss: 1.5354156857938506e-05 2023-01-21 13:26:57.388177: step: 1856/533, loss: 0.00021991730318404734 2023-01-21 13:26:58.594717: step: 1860/533, loss: 3.409385681152344e-05 2023-01-21 13:26:59.789065: step: 1864/533, loss: 0.04563913494348526 2023-01-21 13:27:01.022108: step: 1868/533, loss: 0.0006259918445721269 2023-01-21 13:27:02.171901: step: 1872/533, loss: 0.0003208160342182964 2023-01-21 13:27:03.348447: step: 1876/533, loss: 4.38690176451928e-06 2023-01-21 13:27:04.534147: step: 1880/533, loss: 0.016524506732821465 2023-01-21 13:27:05.699799: step: 1884/533, loss: 0.0010344504844397306 2023-01-21 13:27:06.810743: step: 1888/533, loss: 9.727478754939511e-06 2023-01-21 13:27:07.975786: step: 1892/533, loss: 0.0054416656494140625 2023-01-21 13:27:09.128456: step: 1896/533, loss: 0.0007518768543377519 2023-01-21 13:27:10.342942: step: 1900/533, loss: 0.00045452118501998484 2023-01-21 13:27:11.515361: step: 1904/533, loss: 3.814697265625e-06 2023-01-21 13:27:12.680181: step: 1908/533, loss: 3.5762786865234375e-06 2023-01-21 13:27:13.857360: step: 1912/533, loss: 0.013080311007797718 2023-01-21 13:27:15.048365: step: 1916/533, loss: 5.8650970458984375e-05 2023-01-21 13:27:16.185020: step: 1920/533, loss: 0.006737709045410156 2023-01-21 13:27:17.338077: step: 1924/533, loss: 0.00010623932757880539 2023-01-21 13:27:18.486873: step: 1928/533, loss: 0.006479215808212757 2023-01-21 13:27:19.656643: step: 1932/533, loss: 0.0008396149496547878 2023-01-21 13:27:20.849531: step: 1936/533, loss: 2.384185791015625e-05 2023-01-21 13:27:22.020798: step: 1940/533, loss: 0.0038040163926780224 2023-01-21 13:27:23.209824: step: 1944/533, loss: 2.7275087632006034e-05 2023-01-21 13:27:24.370647: step: 1948/533, loss: 0.00019674302893690765 2023-01-21 13:27:25.542783: step: 1952/533, loss: 4.276037361705676e-05 2023-01-21 13:27:26.679431: step: 1956/533, loss: 0.002428436418995261 2023-01-21 13:27:27.866061: step: 1960/533, loss: 0.0007152557955123484 2023-01-21 13:27:29.044673: step: 1964/533, loss: 6.198883056640625e-05 2023-01-21 13:27:30.207787: step: 1968/533, loss: 7.057190487103071e-06 2023-01-21 13:27:31.375535: step: 1972/533, loss: 0.0029434680473059416 2023-01-21 13:27:32.565986: step: 1976/533, loss: 0.0020652771927416325 2023-01-21 13:27:33.792250: step: 1980/533, loss: 0.02981853485107422 2023-01-21 13:27:34.980902: step: 1984/533, loss: 0.00014734269643668085 2023-01-21 13:27:36.138823: step: 1988/533, loss: 0.00014128685870673507 2023-01-21 13:27:37.333632: step: 1992/533, loss: 1.296997106692288e-05 2023-01-21 13:27:38.508677: step: 1996/533, loss: 0.00013809204392600805 2023-01-21 13:27:39.653237: step: 2000/533, loss: 0.008411980234086514 2023-01-21 13:27:40.855042: step: 2004/533, loss: 0.06284943222999573 2023-01-21 13:27:41.980791: step: 2008/533, loss: 0.0012033461825922132 2023-01-21 13:27:43.169254: step: 2012/533, loss: 2.5320052372990176e-05 2023-01-21 13:27:44.358057: step: 2016/533, loss: 0.006291294004768133 2023-01-21 13:27:45.529686: step: 2020/533, loss: 0.0004938125493936241 2023-01-21 13:27:46.708919: step: 2024/533, loss: 0.0012397767277434468 2023-01-21 13:27:47.903403: step: 2028/533, loss: 0.011632729321718216 2023-01-21 13:27:49.108662: step: 2032/533, loss: 0.005409622564911842 2023-01-21 13:27:50.280278: step: 2036/533, loss: 7.095337059581652e-05 2023-01-21 13:27:51.486329: step: 2040/533, loss: 0.0026950836181640625 2023-01-21 13:27:52.699823: step: 2044/533, loss: 0.0006472587701864541 2023-01-21 13:27:53.862108: step: 2048/533, loss: 1.4686585018353071e-05 2023-01-21 13:27:55.071676: step: 2052/533, loss: 0.02822704240679741 2023-01-21 13:27:56.260607: step: 2056/533, loss: 4.487037949729711e-05 2023-01-21 13:27:57.403957: step: 2060/533, loss: 0.0025330542121082544 2023-01-21 13:27:58.571497: step: 2064/533, loss: 4.954338146490045e-05 2023-01-21 13:27:59.738723: step: 2068/533, loss: 0.02870206907391548 2023-01-21 13:28:00.845794: step: 2072/533, loss: 0.00018787382578011602 2023-01-21 13:28:02.048590: step: 2076/533, loss: 4.38690176451928e-06 2023-01-21 13:28:03.250154: step: 2080/533, loss: 0.012172508984804153 2023-01-21 13:28:04.432629: step: 2084/533, loss: 0.09059123694896698 2023-01-21 13:28:05.567919: step: 2088/533, loss: 0.0024335861671715975 2023-01-21 13:28:06.708734: step: 2092/533, loss: 4.0435788832837716e-05 2023-01-21 13:28:07.905374: step: 2096/533, loss: 0.0021583556663244963 2023-01-21 13:28:09.078849: step: 2100/533, loss: 0.0218061450868845 2023-01-21 13:28:10.254438: step: 2104/533, loss: 0.01072168443351984 2023-01-21 13:28:11.413990: step: 2108/533, loss: -4.19616708313697e-06 2023-01-21 13:28:12.615374: step: 2112/533, loss: 0.00042304996168240905 2023-01-21 13:28:13.794319: step: 2116/533, loss: 0.0009634971502237022 2023-01-21 13:28:14.959796: step: 2120/533, loss: 0.00010538101196289062 2023-01-21 13:28:16.094630: step: 2124/533, loss: 0.012665748596191406 2023-01-21 13:28:17.283825: step: 2128/533, loss: 0.048429109156131744 2023-01-21 13:28:18.427519: step: 2132/533, loss: 1.5354156857938506e-05 ================================================== Loss: 0.020 -------------------- Dev: {'event': {'p': 0.6211453744493393, 'r': 0.7509986684420772, 'f1': 0.6799276672694394}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Test: {'event': {'p': 0.6510660426417056, 'r': 0.8005115089514067, 'f1': 0.7180957843418411}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Chinese: {'event': {'p': 0.5357142857142857, 'r': 0.8333333333333334, 'f1': 0.6521739130434783}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Korean: {'event': {'p': 0.6, 'r': 0.5714285714285714, 'f1': 0.5853658536585366}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Russian: {'event': {'p': 0.48484848484848486, 'r': 0.4444444444444444, 'f1': 0.463768115942029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Chinese: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 13:28:59.656417: step: 4/533, loss: 3.807544999290258e-05 2023-01-21 13:29:00.808307: step: 8/533, loss: 0.00021266937255859375 2023-01-21 13:29:01.962270: step: 12/533, loss: 5.5837630497990176e-05 2023-01-21 13:29:03.152640: step: 16/533, loss: 0.007767486851662397 2023-01-21 13:29:04.353157: step: 20/533, loss: 0.9967159032821655 2023-01-21 13:29:05.541525: step: 24/533, loss: 0.0017751693958416581 2023-01-21 13:29:06.685706: step: 28/533, loss: 0.00023698806762695312 2023-01-21 13:29:07.876391: step: 32/533, loss: 0.022069884464144707 2023-01-21 13:29:09.055926: step: 36/533, loss: 0.0028072355780750513 2023-01-21 13:29:10.223130: step: 40/533, loss: 0.0034592628944665194 2023-01-21 13:29:11.391045: step: 44/533, loss: 0.006538105197250843 2023-01-21 13:29:12.604679: step: 48/533, loss: 0.51051926612854 2023-01-21 13:29:13.822994: step: 52/533, loss: 0.0002382278471486643 2023-01-21 13:29:14.950267: step: 56/533, loss: -1.9073485191256623e-07 2023-01-21 13:29:16.171945: step: 60/533, loss: 0.00017662049503996968 2023-01-21 13:29:17.372337: step: 64/533, loss: 7.700920104980469e-05 2023-01-21 13:29:18.563124: step: 68/533, loss: 0.009666157886385918 2023-01-21 13:29:19.733890: step: 72/533, loss: 0.014373588375747204 2023-01-21 13:29:20.920259: step: 76/533, loss: 0.001781463623046875 2023-01-21 13:29:22.064086: step: 80/533, loss: 0.0014844894176349044 2023-01-21 13:29:23.220847: step: 84/533, loss: 0.0009613037109375 2023-01-21 13:29:24.391974: step: 88/533, loss: 0.015164375305175781 2023-01-21 13:29:25.561631: step: 92/533, loss: 0.0005056381342001259 2023-01-21 13:29:26.752595: step: 96/533, loss: 0.05765227973461151 2023-01-21 13:29:27.882394: step: 100/533, loss: 0.00011482238915050402 2023-01-21 13:29:29.077166: step: 104/533, loss: 0.00018968583026435226 2023-01-21 13:29:30.254456: step: 108/533, loss: 0.002274131868034601 2023-01-21 13:29:31.393629: step: 112/533, loss: 0.026640035212039948 2023-01-21 13:29:32.590499: step: 116/533, loss: 0.012826728634536266 2023-01-21 13:29:33.788126: step: 120/533, loss: 0.03502178564667702 2023-01-21 13:29:34.969520: step: 124/533, loss: 0.00028486252995207906 2023-01-21 13:29:36.159303: step: 128/533, loss: 0.010875320062041283 2023-01-21 13:29:37.331810: step: 132/533, loss: 9.794234938453883e-05 2023-01-21 13:29:38.504620: step: 136/533, loss: 0.0014381409855559468 2023-01-21 13:29:39.664866: step: 140/533, loss: 4.692077709478326e-05 2023-01-21 13:29:40.823457: step: 144/533, loss: 2.3937225705594756e-05 2023-01-21 13:29:42.028931: step: 148/533, loss: 0.07252006232738495 2023-01-21 13:29:43.195920: step: 152/533, loss: 0.0008319854969158769 2023-01-21 13:29:44.353550: step: 156/533, loss: 0.41022226214408875 2023-01-21 13:29:45.498038: step: 160/533, loss: 4.806518700206652e-05 2023-01-21 13:29:46.688814: step: 164/533, loss: 0.00023412704467773438 2023-01-21 13:29:47.865822: step: 168/533, loss: 0.0007537842029705644 2023-01-21 13:29:48.994162: step: 172/533, loss: -5.14984139954322e-06 2023-01-21 13:29:50.151328: step: 176/533, loss: 0.0002980232238769531 2023-01-21 13:29:51.384537: step: 180/533, loss: 0.001044464181177318 2023-01-21 13:29:52.564759: step: 184/533, loss: 0.01426782552152872 2023-01-21 13:29:53.691735: step: 188/533, loss: 3.566742088878527e-05 2023-01-21 13:29:54.893252: step: 192/533, loss: 0.0010480880737304688 2023-01-21 13:29:56.083933: step: 196/533, loss: 0.03789710998535156 2023-01-21 13:29:57.326123: step: 200/533, loss: 0.0044303894974291325 2023-01-21 13:29:58.526567: step: 204/533, loss: 7.295608611457283e-06 2023-01-21 13:29:59.737271: step: 208/533, loss: 2.0027162463520654e-05 2023-01-21 13:30:00.906452: step: 212/533, loss: 0.00035085680428892374 2023-01-21 13:30:02.109423: step: 216/533, loss: 0.00884027499705553 2023-01-21 13:30:03.351423: step: 220/533, loss: 0.00045452118501998484 2023-01-21 13:30:04.543315: step: 224/533, loss: 0.001056671142578125 2023-01-21 13:30:05.706197: step: 228/533, loss: 0.000720977783203125 2023-01-21 13:30:06.859981: step: 232/533, loss: 0.0004276275576557964 2023-01-21 13:30:08.079282: step: 236/533, loss: 0.02678699605166912 2023-01-21 13:30:09.310549: step: 240/533, loss: 0.0009007453918457031 2023-01-21 13:30:10.525160: step: 244/533, loss: 0.0023681640159338713 2023-01-21 13:30:11.701366: step: 248/533, loss: 0.007801055442541838 2023-01-21 13:30:12.873259: step: 252/533, loss: 0.0017336845630779862 2023-01-21 13:30:14.059051: step: 256/533, loss: 0.06660271435976028 2023-01-21 13:30:15.238941: step: 260/533, loss: 0.00033373833866789937 2023-01-21 13:30:16.414935: step: 264/533, loss: 1.0824203855008818e-05 2023-01-21 13:30:17.610418: step: 268/533, loss: 0.00015802383131813258 2023-01-21 13:30:18.770625: step: 272/533, loss: 0.020971108227968216 2023-01-21 13:30:19.932965: step: 276/533, loss: 3.8814545405330136e-05 2023-01-21 13:30:21.109641: step: 280/533, loss: 0.000720596348401159 2023-01-21 13:30:22.279081: step: 284/533, loss: 0.002125263214111328 2023-01-21 13:30:23.491388: step: 288/533, loss: 0.00043773651123046875 2023-01-21 13:30:24.648009: step: 292/533, loss: 1.106262243411038e-05 2023-01-21 13:30:25.796046: step: 296/533, loss: 0.0004241943533997983 2023-01-21 13:30:26.924174: step: 300/533, loss: 3.147125607938506e-05 2023-01-21 13:30:28.056532: step: 304/533, loss: 0.000173473366885446 2023-01-21 13:30:29.247444: step: 308/533, loss: 0.024636555463075638 2023-01-21 13:30:30.430687: step: 312/533, loss: 0.01586446911096573 2023-01-21 13:30:31.609366: step: 316/533, loss: 0.0009374618530273438 2023-01-21 13:30:32.832051: step: 320/533, loss: 0.013634777627885342 2023-01-21 13:30:34.001990: step: 324/533, loss: 0.00011720657494151965 2023-01-21 13:30:35.137035: step: 328/533, loss: 2.813339506246848e-06 2023-01-21 13:30:36.315390: step: 332/533, loss: 4.482269287109375e-05 2023-01-21 13:30:37.514314: step: 336/533, loss: 6.3896181927702855e-06 2023-01-21 13:30:38.693972: step: 340/533, loss: 0.0015603393549099565 2023-01-21 13:30:39.872604: step: 344/533, loss: 0.00388164515607059 2023-01-21 13:30:41.041700: step: 348/533, loss: 0.01164708100259304 2023-01-21 13:30:42.185914: step: 352/533, loss: 0.01867714151740074 2023-01-21 13:30:43.374684: step: 356/533, loss: 0.00010871887934627011 2023-01-21 13:30:44.538493: step: 360/533, loss: 1.792907642084174e-05 2023-01-21 13:30:45.708626: step: 364/533, loss: 4.806518700206652e-05 2023-01-21 13:30:46.876652: step: 368/533, loss: 0.007267188746482134 2023-01-21 13:30:48.057586: step: 372/533, loss: 0.0607847198843956 2023-01-21 13:30:49.217131: step: 376/533, loss: -8.583069188716763e-07 2023-01-21 13:30:50.406613: step: 380/533, loss: 0.011996460147202015 2023-01-21 13:30:51.644903: step: 384/533, loss: 0.03175077587366104 2023-01-21 13:30:52.801498: step: 388/533, loss: 0.003375244326889515 2023-01-21 13:30:53.940561: step: 392/533, loss: 4.215240551275201e-05 2023-01-21 13:30:55.100746: step: 396/533, loss: 9.918212890625e-05 2023-01-21 13:30:56.251981: step: 400/533, loss: 5.8937075664289296e-05 2023-01-21 13:30:57.406096: step: 404/533, loss: -5.91278057981981e-06 2023-01-21 13:30:58.590190: step: 408/533, loss: 0.005866623017936945 2023-01-21 13:30:59.749527: step: 412/533, loss: -2.0503998712229077e-06 2023-01-21 13:31:00.956775: step: 416/533, loss: 0.0063298228196799755 2023-01-21 13:31:02.127222: step: 420/533, loss: 0.000274658203125 2023-01-21 13:31:03.336059: step: 424/533, loss: 0.005877113435417414 2023-01-21 13:31:04.507297: step: 428/533, loss: 0.0067727090790867805 2023-01-21 13:31:05.676515: step: 432/533, loss: 7.772445678710938e-05 2023-01-21 13:31:06.869279: step: 436/533, loss: 3.910065061063506e-05 2023-01-21 13:31:08.039342: step: 440/533, loss: 0.004544067196547985 2023-01-21 13:31:09.221919: step: 444/533, loss: 1.2493133908719756e-05 2023-01-21 13:31:10.363238: step: 448/533, loss: 6.046295311534777e-05 2023-01-21 13:31:11.544129: step: 452/533, loss: 0.0002361297665629536 2023-01-21 13:31:12.720653: step: 456/533, loss: 0.00035190582275390625 2023-01-21 13:31:13.878494: step: 460/533, loss: 0.0003218650817871094 2023-01-21 13:31:15.031048: step: 464/533, loss: 0.010560035705566406 2023-01-21 13:31:16.243221: step: 468/533, loss: 0.006488990969955921 2023-01-21 13:31:17.408355: step: 472/533, loss: 0.005353736691176891 2023-01-21 13:31:18.554535: step: 476/533, loss: 0.0019417285220697522 2023-01-21 13:31:19.726803: step: 480/533, loss: 6.294249942584429e-06 2023-01-21 13:31:20.924427: step: 484/533, loss: 0.0026119232643395662 2023-01-21 13:31:22.079151: step: 488/533, loss: 1.0490417707842425e-06 2023-01-21 13:31:23.298506: step: 492/533, loss: 0.013564681634306908 2023-01-21 13:31:24.501287: step: 496/533, loss: 2.9563905172835803e-06 2023-01-21 13:31:25.685794: step: 500/533, loss: 0.023888587951660156 2023-01-21 13:31:26.864058: step: 504/533, loss: 0.04116563871502876 2023-01-21 13:31:28.027174: step: 508/533, loss: 0.00047068597632460296 2023-01-21 13:31:29.201899: step: 512/533, loss: 1.1205673217773438e-05 2023-01-21 13:31:30.438865: step: 516/533, loss: 0.0050483704544603825 2023-01-21 13:31:31.597655: step: 520/533, loss: 0.0015597344608977437 2023-01-21 13:31:32.756610: step: 524/533, loss: 0.0035635950043797493 2023-01-21 13:31:33.926996: step: 528/533, loss: 2.28881845032447e-06 2023-01-21 13:31:35.063237: step: 532/533, loss: 0.0015805244911462069 2023-01-21 13:31:36.241395: step: 536/533, loss: 0.00013208390737418085 2023-01-21 13:31:37.393990: step: 540/533, loss: 0.000285911577520892 2023-01-21 13:31:38.541409: step: 544/533, loss: 9.51766996877268e-05 2023-01-21 13:31:39.750971: step: 548/533, loss: 0.006950855255126953 2023-01-21 13:31:40.941776: step: 552/533, loss: 3.6811830796068534e-05 2023-01-21 13:31:42.131897: step: 556/533, loss: 6.446838233387098e-05 2023-01-21 13:31:43.366289: step: 560/533, loss: 4.920959327137098e-05 2023-01-21 13:31:44.586653: step: 564/533, loss: 0.001375484513118863 2023-01-21 13:31:45.750354: step: 568/533, loss: 0.03983908146619797 2023-01-21 13:31:46.904579: step: 572/533, loss: 4.043579247081652e-05 2023-01-21 13:31:48.101686: step: 576/533, loss: 0.006997585296630859 2023-01-21 13:31:49.260189: step: 580/533, loss: 0.0030935765244066715 2023-01-21 13:31:50.395678: step: 584/533, loss: 5.7220458984375e-05 2023-01-21 13:31:51.587350: step: 588/533, loss: 0.0029630661010742188 2023-01-21 13:31:52.772567: step: 592/533, loss: 0.03827390819787979 2023-01-21 13:31:53.962528: step: 596/533, loss: 0.06122303009033203 2023-01-21 13:31:55.116818: step: 600/533, loss: 0.00024874211521819234 2023-01-21 13:31:56.270126: step: 604/533, loss: 0.00014429092698264867 2023-01-21 13:31:57.483542: step: 608/533, loss: 0.005464363377541304 2023-01-21 13:31:58.671484: step: 612/533, loss: 1.5258789289873675e-06 2023-01-21 13:31:59.906041: step: 616/533, loss: 0.0035102844703942537 2023-01-21 13:32:01.060842: step: 620/533, loss: 0.00526008615270257 2023-01-21 13:32:02.234035: step: 624/533, loss: 0.0005893707275390625 2023-01-21 13:32:03.462997: step: 628/533, loss: 0.0026865960098803043 2023-01-21 13:32:04.655867: step: 632/533, loss: 0.0006772995693609118 2023-01-21 13:32:05.859772: step: 636/533, loss: 0.00025959016056731343 2023-01-21 13:32:07.094694: step: 640/533, loss: 0.0017457961803302169 2023-01-21 13:32:08.219033: step: 644/533, loss: 0.006958675570785999 2023-01-21 13:32:09.368348: step: 648/533, loss: 2.0313264030846767e-05 2023-01-21 13:32:10.536344: step: 652/533, loss: 0.00019569396681617945 2023-01-21 13:32:11.701100: step: 656/533, loss: 0.0028858184814453125 2023-01-21 13:32:12.868051: step: 660/533, loss: 0.0032699585426598787 2023-01-21 13:32:14.095294: step: 664/533, loss: 0.0007886886596679688 2023-01-21 13:32:15.289350: step: 668/533, loss: 0.013704395852982998 2023-01-21 13:32:16.431413: step: 672/533, loss: 0.0003761291445698589 2023-01-21 13:32:17.595756: step: 676/533, loss: 0.0061893463134765625 2023-01-21 13:32:18.803881: step: 680/533, loss: 0.0010457038879394531 2023-01-21 13:32:19.984591: step: 684/533, loss: 4.177093433099799e-05 2023-01-21 13:32:21.160328: step: 688/533, loss: 2.975463939947076e-05 2023-01-21 13:32:22.338392: step: 692/533, loss: 0.0013466834789142013 2023-01-21 13:32:23.492499: step: 696/533, loss: 0.00013155341730453074 2023-01-21 13:32:24.650927: step: 700/533, loss: 3.6525725590763614e-05 2023-01-21 13:32:25.794702: step: 704/533, loss: 0.0008505344158038497 2023-01-21 13:32:26.986130: step: 708/533, loss: 0.0019144058460369706 2023-01-21 13:32:28.173644: step: 712/533, loss: 0.0015656471950933337 2023-01-21 13:32:29.344642: step: 716/533, loss: 1.544952465337701e-05 2023-01-21 13:32:30.501346: step: 720/533, loss: 8.106231689453125e-06 2023-01-21 13:32:31.662185: step: 724/533, loss: 2.6702880859375e-05 2023-01-21 13:32:32.865716: step: 728/533, loss: 0.04627342149615288 2023-01-21 13:32:34.054117: step: 732/533, loss: 0.0053611756302416325 2023-01-21 13:32:35.218184: step: 736/533, loss: 0.0007402420742437243 2023-01-21 13:32:36.342583: step: 740/533, loss: 0.0005681038019247353 2023-01-21 13:32:37.518341: step: 744/533, loss: 0.009149361401796341 2023-01-21 13:32:38.707082: step: 748/533, loss: 0.0006128310924395919 2023-01-21 13:32:39.890321: step: 752/533, loss: 6.88552827341482e-05 2023-01-21 13:32:41.068304: step: 756/533, loss: 0.013961220160126686 2023-01-21 13:32:42.283246: step: 760/533, loss: 0.007540511898696423 2023-01-21 13:32:43.488448: step: 764/533, loss: 0.0015172958374023438 2023-01-21 13:32:44.661947: step: 768/533, loss: 2.117157055181451e-05 2023-01-21 13:32:45.827765: step: 772/533, loss: 0.00021724702673964202 2023-01-21 13:32:46.987309: step: 776/533, loss: 0.0002311706484761089 2023-01-21 13:32:48.220332: step: 780/533, loss: 0.05750885233283043 2023-01-21 13:32:49.370947: step: 784/533, loss: 0.0005660057067871094 2023-01-21 13:32:50.535387: step: 788/533, loss: 2.0647050405386835e-05 2023-01-21 13:32:51.724397: step: 792/533, loss: 0.00044736862764693797 2023-01-21 13:32:52.874349: step: 796/533, loss: 0.0006258010980673134 2023-01-21 13:32:54.031163: step: 800/533, loss: 0.0001451492280466482 2023-01-21 13:32:55.221588: step: 804/533, loss: 0.00026073455228470266 2023-01-21 13:32:56.409723: step: 808/533, loss: 0.02436218410730362 2023-01-21 13:32:57.562309: step: 812/533, loss: 0.0007188796880654991 2023-01-21 13:32:58.743327: step: 816/533, loss: 0.0661131888628006 2023-01-21 13:32:59.948519: step: 820/533, loss: 0.0003314971982035786 2023-01-21 13:33:01.124740: step: 824/533, loss: 7.686614844715223e-05 2023-01-21 13:33:02.290966: step: 828/533, loss: 0.0005083084106445312 2023-01-21 13:33:03.432214: step: 832/533, loss: 0.037106703966856 2023-01-21 13:33:04.645109: step: 836/533, loss: 0.02768269181251526 2023-01-21 13:33:05.813318: step: 840/533, loss: 0.0011528492905199528 2023-01-21 13:33:07.031251: step: 844/533, loss: 1.1634827387752011e-05 2023-01-21 13:33:08.259069: step: 848/533, loss: 0.0006323814741335809 2023-01-21 13:33:09.452246: step: 852/533, loss: 0.0011382102966308594 2023-01-21 13:33:10.597797: step: 856/533, loss: 1.8119812921213452e-06 2023-01-21 13:33:11.754094: step: 860/533, loss: 0.0001122474714065902 2023-01-21 13:33:12.905225: step: 864/533, loss: 0.030983783304691315 2023-01-21 13:33:14.072299: step: 868/533, loss: 0.01650695875287056 2023-01-21 13:33:15.254558: step: 872/533, loss: 0.01674938201904297 2023-01-21 13:33:16.416495: step: 876/533, loss: 8.220673043979332e-05 2023-01-21 13:33:17.581905: step: 880/533, loss: 7.400512549793348e-05 2023-01-21 13:33:18.736285: step: 884/533, loss: 5.817413239128655e-06 2023-01-21 13:33:19.897985: step: 888/533, loss: 0.014087105169892311 2023-01-21 13:33:21.097178: step: 892/533, loss: 0.030108071863651276 2023-01-21 13:33:22.284382: step: 896/533, loss: 4.091262962901965e-05 2023-01-21 13:33:23.546040: step: 900/533, loss: 1.296997106692288e-05 2023-01-21 13:33:24.683260: step: 904/533, loss: 0.0024122237227857113 2023-01-21 13:33:25.838407: step: 908/533, loss: 0.0006353378412313759 2023-01-21 13:33:27.033711: step: 912/533, loss: 0.018374014645814896 2023-01-21 13:33:28.204908: step: 916/533, loss: 0.012214088812470436 2023-01-21 13:33:29.373872: step: 920/533, loss: 0.03250427171587944 2023-01-21 13:33:30.574706: step: 924/533, loss: 2.0503997802734375e-05 2023-01-21 13:33:31.753645: step: 928/533, loss: 0.0008580207941122353 2023-01-21 13:33:32.951769: step: 932/533, loss: 0.0991085022687912 2023-01-21 13:33:34.094648: step: 936/533, loss: 0.11838188022375107 2023-01-21 13:33:35.269495: step: 940/533, loss: 0.03463344648480415 2023-01-21 13:33:36.434250: step: 944/533, loss: 3.4809112548828125e-05 2023-01-21 13:33:37.576049: step: 948/533, loss: 0.0001669883931754157 2023-01-21 13:33:38.778279: step: 952/533, loss: 0.0011203765170648694 2023-01-21 13:33:39.928215: step: 956/533, loss: 2.0790101189049892e-05 2023-01-21 13:33:41.113634: step: 960/533, loss: 0.0002492904895916581 2023-01-21 13:33:42.252947: step: 964/533, loss: 0.0057006836868822575 2023-01-21 13:33:43.431351: step: 968/533, loss: 0.0070161824114620686 2023-01-21 13:33:44.584114: step: 972/533, loss: 0.0001336097629973665 2023-01-21 13:33:45.769245: step: 976/533, loss: 0.05814504623413086 2023-01-21 13:33:46.918646: step: 980/533, loss: 7.877349707996473e-05 2023-01-21 13:33:48.092317: step: 984/533, loss: 5.4168704082258046e-05 2023-01-21 13:33:49.258624: step: 988/533, loss: 0.0009250640869140625 2023-01-21 13:33:50.407291: step: 992/533, loss: 0.0037965774536132812 2023-01-21 13:33:51.588519: step: 996/533, loss: 0.0011080740951001644 2023-01-21 13:33:52.755126: step: 1000/533, loss: 0.0014207839267328382 2023-01-21 13:33:53.926196: step: 1004/533, loss: 0.016039753332734108 2023-01-21 13:33:55.084904: step: 1008/533, loss: 0.0005547046894207597 2023-01-21 13:33:56.247140: step: 1012/533, loss: 2.6702880859375e-05 2023-01-21 13:33:57.454470: step: 1016/533, loss: 0.05676741898059845 2023-01-21 13:33:58.592818: step: 1020/533, loss: 0.017014503479003906 2023-01-21 13:33:59.742987: step: 1024/533, loss: 5.569458153331652e-05 2023-01-21 13:34:00.939771: step: 1028/533, loss: 7.42912307032384e-05 2023-01-21 13:34:02.118842: step: 1032/533, loss: 0.0021759034134447575 2023-01-21 13:34:03.283797: step: 1036/533, loss: 0.0002665519714355469 2023-01-21 13:34:04.450189: step: 1040/533, loss: 0.011576080694794655 2023-01-21 13:34:05.625318: step: 1044/533, loss: 0.004502486903220415 2023-01-21 13:34:06.788385: step: 1048/533, loss: 8.354187593795359e-05 2023-01-21 13:34:07.968726: step: 1052/533, loss: 0.001936578773893416 2023-01-21 13:34:09.119343: step: 1056/533, loss: 0.0008191585657186806 2023-01-21 13:34:10.281459: step: 1060/533, loss: 0.0017066956497728825 2023-01-21 13:34:11.447204: step: 1064/533, loss: 0.00045566557673737407 2023-01-21 13:34:12.594836: step: 1068/533, loss: 5.3787229262525216e-05 2023-01-21 13:34:13.769185: step: 1072/533, loss: 0.006452894303947687 2023-01-21 13:34:14.982457: step: 1076/533, loss: 0.0017498015658929944 2023-01-21 13:34:16.180451: step: 1080/533, loss: 0.09159460663795471 2023-01-21 13:34:17.317060: step: 1084/533, loss: 3.72409813280683e-05 2023-01-21 13:34:18.525845: step: 1088/533, loss: -9.15527380129788e-06 2023-01-21 13:34:19.687002: step: 1092/533, loss: 0.0010288238991051912 2023-01-21 13:34:20.820541: step: 1096/533, loss: 3.5285947888041846e-06 2023-01-21 13:34:21.963931: step: 1100/533, loss: 0.0006485939375124872 2023-01-21 13:34:23.152688: step: 1104/533, loss: 1.0442733582749497e-05 2023-01-21 13:34:24.324942: step: 1108/533, loss: 0.132389634847641 2023-01-21 13:34:25.483987: step: 1112/533, loss: 0.0003182411310262978 2023-01-21 13:34:26.621601: step: 1116/533, loss: 0.0048274993896484375 2023-01-21 13:34:27.834365: step: 1120/533, loss: 0.006980514619499445 2023-01-21 13:34:29.059640: step: 1124/533, loss: 0.001069354941137135 2023-01-21 13:34:30.230520: step: 1128/533, loss: 0.0010107040870934725 2023-01-21 13:34:31.408349: step: 1132/533, loss: 0.0009754180791787803 2023-01-21 13:34:32.617016: step: 1136/533, loss: 1.9073486328125e-06 2023-01-21 13:34:33.790618: step: 1140/533, loss: 0.00041484832763671875 2023-01-21 13:34:35.004960: step: 1144/533, loss: 0.0007080078357830644 2023-01-21 13:34:36.201481: step: 1148/533, loss: 5.34057608092553e-06 2023-01-21 13:34:37.361650: step: 1152/533, loss: 0.0040458678267896175 2023-01-21 13:34:38.537232: step: 1156/533, loss: 0.04434919357299805 2023-01-21 13:34:39.731566: step: 1160/533, loss: 3.8623809814453125e-05 2023-01-21 13:34:40.870909: step: 1164/533, loss: 0.0002884864807128906 2023-01-21 13:34:42.054537: step: 1168/533, loss: 0.0005552291986532509 2023-01-21 13:34:43.258185: step: 1172/533, loss: 0.06897010654211044 2023-01-21 13:34:44.437656: step: 1176/533, loss: 0.00013504027447197586 2023-01-21 13:34:45.599750: step: 1180/533, loss: 0.06146087870001793 2023-01-21 13:34:46.751540: step: 1184/533, loss: 5.53131121705519e-06 2023-01-21 13:34:47.918408: step: 1188/533, loss: 0.009692477993667126 2023-01-21 13:34:49.074470: step: 1192/533, loss: 4.85897071484942e-05 2023-01-21 13:34:50.231285: step: 1196/533, loss: 0.06887397915124893 2023-01-21 13:34:51.472987: step: 1200/533, loss: 3.638267662608996e-05 2023-01-21 13:34:52.634064: step: 1204/533, loss: 0.0002434253692626953 2023-01-21 13:34:53.815536: step: 1208/533, loss: 1.52587890625e-05 2023-01-21 13:34:54.981540: step: 1212/533, loss: 0.043121337890625 2023-01-21 13:34:56.161340: step: 1216/533, loss: 0.02290048636496067 2023-01-21 13:34:57.313688: step: 1220/533, loss: 0.00018668771372176707 2023-01-21 13:34:58.500332: step: 1224/533, loss: 0.0028312683571130037 2023-01-21 13:34:59.642221: step: 1228/533, loss: 0.00012245179095771164 2023-01-21 13:35:00.799893: step: 1232/533, loss: 0.36537235975265503 2023-01-21 13:35:01.947764: step: 1236/533, loss: 2.775192479020916e-05 2023-01-21 13:35:03.099950: step: 1240/533, loss: 0.00021872521028853953 2023-01-21 13:35:04.263014: step: 1244/533, loss: 0.002084016799926758 2023-01-21 13:35:05.466247: step: 1248/533, loss: 0.00045843125553801656 2023-01-21 13:35:06.623026: step: 1252/533, loss: 0.0015971660614013672 2023-01-21 13:35:07.776698: step: 1256/533, loss: 0.04697055742144585 2023-01-21 13:35:08.991868: step: 1260/533, loss: 0.017285728827118874 2023-01-21 13:35:10.163763: step: 1264/533, loss: 4.29630272265058e-05 2023-01-21 13:35:11.350716: step: 1268/533, loss: 0.006230545230209827 2023-01-21 13:35:12.536806: step: 1272/533, loss: -1.010894811770413e-05 2023-01-21 13:35:13.743947: step: 1276/533, loss: 0.041637323796749115 2023-01-21 13:35:14.917153: step: 1280/533, loss: 0.0001161575346486643 2023-01-21 13:35:16.063954: step: 1284/533, loss: 0.00017299651517532766 2023-01-21 13:35:17.276760: step: 1288/533, loss: 0.001446628593839705 2023-01-21 13:35:18.423517: step: 1292/533, loss: 0.00021438598923850805 2023-01-21 13:35:19.584390: step: 1296/533, loss: 0.0003723144473042339 2023-01-21 13:35:20.731678: step: 1300/533, loss: 0.0004852295096497983 2023-01-21 13:35:21.872453: step: 1304/533, loss: 0.0005516052478924394 2023-01-21 13:35:23.072018: step: 1308/533, loss: 0.004182147793471813 2023-01-21 13:35:24.251905: step: 1312/533, loss: 1.9073486328125e-05 2023-01-21 13:35:25.433717: step: 1316/533, loss: 0.012036323547363281 2023-01-21 13:35:26.548706: step: 1320/533, loss: 0.0009317874792031944 2023-01-21 13:35:27.739372: step: 1324/533, loss: 7.2479248046875e-05 2023-01-21 13:35:28.922416: step: 1328/533, loss: 0.002017498016357422 2023-01-21 13:35:30.112898: step: 1332/533, loss: 0.005322456359863281 2023-01-21 13:35:31.317803: step: 1336/533, loss: 0.002285194583237171 2023-01-21 13:35:32.480452: step: 1340/533, loss: 0.006087875925004482 2023-01-21 13:35:33.673971: step: 1344/533, loss: 0.05332136154174805 2023-01-21 13:35:34.852802: step: 1348/533, loss: 0.0036059380508959293 2023-01-21 13:35:36.027770: step: 1352/533, loss: 0.009685516357421875 2023-01-21 13:35:37.202683: step: 1356/533, loss: 0.0020960806868970394 2023-01-21 13:35:38.408338: step: 1360/533, loss: 0.0010589599842205644 2023-01-21 13:35:39.527828: step: 1364/533, loss: 0.001273727510124445 2023-01-21 13:35:40.688692: step: 1368/533, loss: 0.0007979393121786416 2023-01-21 13:35:41.872657: step: 1372/533, loss: 0.0006975174183025956 2023-01-21 13:35:43.000048: step: 1376/533, loss: 0.01880960538983345 2023-01-21 13:35:44.187091: step: 1380/533, loss: 0.030681611970067024 2023-01-21 13:35:45.390476: step: 1384/533, loss: 8.640289888717234e-05 2023-01-21 13:35:46.587057: step: 1388/533, loss: 0.0021041869185864925 2023-01-21 13:35:47.749285: step: 1392/533, loss: 3.5858156479662284e-05 2023-01-21 13:35:48.934586: step: 1396/533, loss: -1.430511474609375e-06 2023-01-21 13:35:50.087162: step: 1400/533, loss: 0.00374088273383677 2023-01-21 13:35:51.272710: step: 1404/533, loss: 0.013428688049316406 2023-01-21 13:35:52.446140: step: 1408/533, loss: 0.008450508117675781 2023-01-21 13:35:53.614275: step: 1412/533, loss: 7.82012939453125e-05 2023-01-21 13:35:54.777465: step: 1416/533, loss: 0.0005204200861044228 2023-01-21 13:35:55.960186: step: 1420/533, loss: 0.01613311842083931 2023-01-21 13:35:57.126136: step: 1424/533, loss: 1.4400482541532256e-05 2023-01-21 13:35:58.274720: step: 1428/533, loss: 0.00037846562918275595 2023-01-21 13:35:59.434796: step: 1432/533, loss: 0.028551960363984108 2023-01-21 13:36:00.650538: step: 1436/533, loss: 0.0005429268348962069 2023-01-21 13:36:01.861741: step: 1440/533, loss: 0.013313674367964268 2023-01-21 13:36:03.043364: step: 1444/533, loss: 0.0022737502586096525 2023-01-21 13:36:04.219605: step: 1448/533, loss: 0.0027341842651367188 2023-01-21 13:36:05.416673: step: 1452/533, loss: 0.00194549560546875 2023-01-21 13:36:06.590393: step: 1456/533, loss: 0.006162834353744984 2023-01-21 13:36:07.762557: step: 1460/533, loss: 3.204345557605848e-05 2023-01-21 13:36:08.983956: step: 1464/533, loss: 0.007847595028579235 2023-01-21 13:36:10.149762: step: 1468/533, loss: 0.0015055655967444181 2023-01-21 13:36:11.321453: step: 1472/533, loss: 2.059936559817288e-05 2023-01-21 13:36:12.514709: step: 1476/533, loss: 0.016317415982484818 2023-01-21 13:36:13.650403: step: 1480/533, loss: 2.9373170036706142e-05 2023-01-21 13:36:14.826278: step: 1484/533, loss: 1.2683868590102065e-05 2023-01-21 13:36:16.033777: step: 1488/533, loss: 0.003624534700065851 2023-01-21 13:36:17.264939: step: 1492/533, loss: 4.100799742445815e-06 2023-01-21 13:36:18.424184: step: 1496/533, loss: 2.632141149661038e-05 2023-01-21 13:36:19.609013: step: 1500/533, loss: 0.052933502942323685 2023-01-21 13:36:20.783312: step: 1504/533, loss: 0.0001771926908986643 2023-01-21 13:36:21.930163: step: 1508/533, loss: 0.003331470536068082 2023-01-21 13:36:23.136615: step: 1512/533, loss: 0.00020513535127975047 2023-01-21 13:36:24.363172: step: 1516/533, loss: 0.00304241175763309 2023-01-21 13:36:25.560181: step: 1520/533, loss: 0.0025769234634935856 2023-01-21 13:36:26.721567: step: 1524/533, loss: 0.00026569367037154734 2023-01-21 13:36:27.943947: step: 1528/533, loss: 6.10351571594947e-06 2023-01-21 13:36:29.108843: step: 1532/533, loss: 0.002432107925415039 2023-01-21 13:36:30.293531: step: 1536/533, loss: 0.0015659332275390625 2023-01-21 13:36:31.466749: step: 1540/533, loss: 0.0002457142109051347 2023-01-21 13:36:32.640112: step: 1544/533, loss: 0.0016598702641204 2023-01-21 13:36:33.794475: step: 1548/533, loss: 0.00038299558218568563 2023-01-21 13:36:35.001859: step: 1552/533, loss: 7.534026735811494e-06 2023-01-21 13:36:36.170312: step: 1556/533, loss: 0.010682192631065845 2023-01-21 13:36:37.396297: step: 1560/533, loss: 0.006688976194709539 2023-01-21 13:36:38.573914: step: 1564/533, loss: 2.2113323211669922e-05 2023-01-21 13:36:39.740927: step: 1568/533, loss: 5.655288987327367e-05 2023-01-21 13:36:40.935809: step: 1572/533, loss: 0.025667190551757812 2023-01-21 13:36:42.124156: step: 1576/533, loss: 0.02109069749712944 2023-01-21 13:36:43.302251: step: 1580/533, loss: 0.015968991443514824 2023-01-21 13:36:44.486359: step: 1584/533, loss: 0.0007537603378295898 2023-01-21 13:36:45.665671: step: 1588/533, loss: 0.00017671585374046117 2023-01-21 13:36:46.842561: step: 1592/533, loss: 0.006217861548066139 2023-01-21 13:36:48.032289: step: 1596/533, loss: 0.0044998168013989925 2023-01-21 13:36:49.180496: step: 1600/533, loss: 0.0004170417960267514 2023-01-21 13:36:50.333806: step: 1604/533, loss: 0.020654775202274323 2023-01-21 13:36:51.494337: step: 1608/533, loss: 0.001601791474968195 2023-01-21 13:36:52.689111: step: 1612/533, loss: 0.0013985633850097656 2023-01-21 13:36:53.842140: step: 1616/533, loss: 7.209778414107859e-05 2023-01-21 13:36:55.034044: step: 1620/533, loss: 0.005980396643280983 2023-01-21 13:36:56.256554: step: 1624/533, loss: 3.1280520488508046e-05 2023-01-21 13:36:57.411825: step: 1628/533, loss: 0.005026912782341242 2023-01-21 13:36:58.566016: step: 1632/533, loss: 4.96864304295741e-05 2023-01-21 13:36:59.764926: step: 1636/533, loss: 0.0001162529006251134 2023-01-21 13:37:00.921006: step: 1640/533, loss: 2.86102294921875e-05 2023-01-21 13:37:02.109264: step: 1644/533, loss: 0.0011852264869958162 2023-01-21 13:37:03.326956: step: 1648/533, loss: 0.03663644567131996 2023-01-21 13:37:04.519313: step: 1652/533, loss: 0.00025539397029206157 2023-01-21 13:37:05.732108: step: 1656/533, loss: 0.013941383920609951 2023-01-21 13:37:06.927649: step: 1660/533, loss: 0.0005298614269122481 2023-01-21 13:37:08.085409: step: 1664/533, loss: 0.0001453399599995464 2023-01-21 13:37:09.282944: step: 1668/533, loss: 0.021195530891418457 2023-01-21 13:37:10.426103: step: 1672/533, loss: 0.010399436578154564 2023-01-21 13:37:11.582041: step: 1676/533, loss: 0.002645206404849887 2023-01-21 13:37:12.756775: step: 1680/533, loss: 0.058966636657714844 2023-01-21 13:37:13.921434: step: 1684/533, loss: 0.01644315756857395 2023-01-21 13:37:15.120741: step: 1688/533, loss: 0.00021677017502952367 2023-01-21 13:37:16.284026: step: 1692/533, loss: 0.2044660598039627 2023-01-21 13:37:17.453298: step: 1696/533, loss: 0.0032224655151367188 2023-01-21 13:37:18.639674: step: 1700/533, loss: 0.0013557435013353825 2023-01-21 13:37:19.790808: step: 1704/533, loss: 0.00016217232041526586 2023-01-21 13:37:20.981428: step: 1708/533, loss: 0.00020866394334007055 2023-01-21 13:37:22.163372: step: 1712/533, loss: 0.11762447655200958 2023-01-21 13:37:23.297613: step: 1716/533, loss: -7.629396350239404e-07 2023-01-21 13:37:24.553621: step: 1720/533, loss: 0.0019397735595703125 2023-01-21 13:37:25.737926: step: 1724/533, loss: 0.010009956546127796 2023-01-21 13:37:26.930379: step: 1728/533, loss: 0.06170034781098366 2023-01-21 13:37:28.113609: step: 1732/533, loss: 0.0002636909775901586 2023-01-21 13:37:29.238448: step: 1736/533, loss: 0.0002613067626953125 2023-01-21 13:37:30.418464: step: 1740/533, loss: -1.468658410885837e-05 2023-01-21 13:37:31.691629: step: 1744/533, loss: 0.0012242316734045744 2023-01-21 13:37:32.889621: step: 1748/533, loss: 0.0003984451468568295 2023-01-21 13:37:34.128714: step: 1752/533, loss: 0.0017774105072021484 2023-01-21 13:37:35.362399: step: 1756/533, loss: 2.294331898156088e-05 2023-01-21 13:37:36.554166: step: 1760/533, loss: 0.010045815259218216 2023-01-21 13:37:37.761088: step: 1764/533, loss: 0.042183876037597656 2023-01-21 13:37:38.917365: step: 1768/533, loss: 0.0514240525662899 2023-01-21 13:37:40.099906: step: 1772/533, loss: 0.008601188659667969 2023-01-21 13:37:41.328247: step: 1776/533, loss: 0.014454650692641735 2023-01-21 13:37:42.494928: step: 1780/533, loss: 0.017203712835907936 2023-01-21 13:37:43.678770: step: 1784/533, loss: 0.0014197349082678556 2023-01-21 13:37:44.841891: step: 1788/533, loss: 0.00025691985501907766 2023-01-21 13:37:45.980256: step: 1792/533, loss: 0.0017223358154296875 2023-01-21 13:37:47.144575: step: 1796/533, loss: 5.204826811677776e-05 2023-01-21 13:37:48.351888: step: 1800/533, loss: 1.0728836059570312e-05 2023-01-21 13:37:49.528299: step: 1804/533, loss: 1.0681153071345761e-05 2023-01-21 13:37:50.709547: step: 1808/533, loss: 0.0005338668706826866 2023-01-21 13:37:51.864666: step: 1812/533, loss: 0.0004192352353129536 2023-01-21 13:37:53.071326: step: 1816/533, loss: 0.0018618584144860506 2023-01-21 13:37:54.243532: step: 1820/533, loss: 0.0008495807996951044 2023-01-21 13:37:55.451055: step: 1824/533, loss: 0.042612459510564804 2023-01-21 13:37:56.632105: step: 1828/533, loss: 0.00043506623478606343 2023-01-21 13:37:57.834855: step: 1832/533, loss: 0.0023750304244458675 2023-01-21 13:37:59.024503: step: 1836/533, loss: 9.965896606445312e-05 2023-01-21 13:38:00.193784: step: 1840/533, loss: 0.0002407073916401714 2023-01-21 13:38:01.363734: step: 1844/533, loss: 0.04739675298333168 2023-01-21 13:38:02.518898: step: 1848/533, loss: 8.087158494163305e-05 2023-01-21 13:38:03.706609: step: 1852/533, loss: 8.869171142578125e-05 2023-01-21 13:38:04.915600: step: 1856/533, loss: 0.004744147881865501 2023-01-21 13:38:06.107795: step: 1860/533, loss: 1.3732911611441523e-05 2023-01-21 13:38:07.279254: step: 1864/533, loss: 0.023932980373501778 2023-01-21 13:38:08.468953: step: 1868/533, loss: 0.004844284150749445 2023-01-21 13:38:09.672873: step: 1872/533, loss: 0.02382946014404297 2023-01-21 13:38:10.839028: step: 1876/533, loss: 0.00027885439340025187 2023-01-21 13:38:12.020052: step: 1880/533, loss: 0.0016535759204998612 2023-01-21 13:38:13.221899: step: 1884/533, loss: 0.026770973578095436 2023-01-21 13:38:14.396365: step: 1888/533, loss: 0.045336056500673294 2023-01-21 13:38:15.540030: step: 1892/533, loss: 0.0038851737044751644 2023-01-21 13:38:16.666878: step: 1896/533, loss: 0.0002075195370707661 2023-01-21 13:38:17.851063: step: 1900/533, loss: 0.01721658743917942 2023-01-21 13:38:19.079009: step: 1904/533, loss: 0.03827933967113495 2023-01-21 13:38:20.275875: step: 1908/533, loss: 0.0021682740189135075 2023-01-21 13:38:21.486819: step: 1912/533, loss: 0.0011571884388104081 2023-01-21 13:38:22.618807: step: 1916/533, loss: 0.028677750378847122 2023-01-21 13:38:23.802499: step: 1920/533, loss: 1.2397766795402276e-06 2023-01-21 13:38:25.002603: step: 1924/533, loss: 0.00036411284236237407 2023-01-21 13:38:26.170965: step: 1928/533, loss: 0.00029792787972837687 2023-01-21 13:38:27.366625: step: 1932/533, loss: 0.007142257876694202 2023-01-21 13:38:28.576783: step: 1936/533, loss: 0.0006432533264160156 2023-01-21 13:38:29.756165: step: 1940/533, loss: -8.77380352903856e-06 2023-01-21 13:38:30.929999: step: 1944/533, loss: 0.016586685553193092 2023-01-21 13:38:32.168045: step: 1948/533, loss: 0.0001255035458598286 2023-01-21 13:38:33.318708: step: 1952/533, loss: 9.679793947725557e-06 2023-01-21 13:38:34.472028: step: 1956/533, loss: 0.039911746978759766 2023-01-21 13:38:35.639861: step: 1960/533, loss: 0.051844026893377304 2023-01-21 13:38:36.807631: step: 1964/533, loss: 0.001210975693538785 2023-01-21 13:38:37.989986: step: 1968/533, loss: 0.00010075569298351184 2023-01-21 13:38:39.220848: step: 1972/533, loss: 3.8528440200025216e-05 2023-01-21 13:38:40.389481: step: 1976/533, loss: 0.00030765534029342234 2023-01-21 13:38:41.614328: step: 1980/533, loss: 0.06222648546099663 2023-01-21 13:38:42.783577: step: 1984/533, loss: 0.023264123126864433 2023-01-21 13:38:43.976128: step: 1988/533, loss: 0.006372451316565275 2023-01-21 13:38:45.187446: step: 1992/533, loss: 3.862380708596902e-06 2023-01-21 13:38:46.338984: step: 1996/533, loss: 0.0002157211274607107 2023-01-21 13:38:47.516529: step: 2000/533, loss: 0.03579854965209961 2023-01-21 13:38:48.720823: step: 2004/533, loss: 0.000868940376676619 2023-01-21 13:38:49.894757: step: 2008/533, loss: 0.002105140592902899 2023-01-21 13:38:51.097078: step: 2012/533, loss: 9.250640687241685e-06 2023-01-21 13:38:52.262424: step: 2016/533, loss: 0.0036342618986964226 2023-01-21 13:38:53.462683: step: 2020/533, loss: 0.00029354094294831157 2023-01-21 13:38:54.627800: step: 2024/533, loss: -1.4591217222914565e-05 2023-01-21 13:38:55.774335: step: 2028/533, loss: 0.005563068203628063 2023-01-21 13:38:56.910169: step: 2032/533, loss: 0.0003765106375794858 2023-01-21 13:38:58.140571: step: 2036/533, loss: 4.367828296381049e-05 2023-01-21 13:38:59.341930: step: 2040/533, loss: 0.003053188556805253 2023-01-21 13:39:00.541109: step: 2044/533, loss: 0.007087517064064741 2023-01-21 13:39:01.726231: step: 2048/533, loss: 0.011091899126768112 2023-01-21 13:39:02.922165: step: 2052/533, loss: 3.061294773942791e-05 2023-01-21 13:39:04.131983: step: 2056/533, loss: 0.006381988525390625 2023-01-21 13:39:05.285599: step: 2060/533, loss: 0.001312255859375 2023-01-21 13:39:06.476683: step: 2064/533, loss: 0.0008981704595498741 2023-01-21 13:39:07.632672: step: 2068/533, loss: 3.356933666509576e-05 2023-01-21 13:39:08.813646: step: 2072/533, loss: -1.5258774510584772e-06 2023-01-21 13:39:10.044876: step: 2076/533, loss: 0.0002628326474223286 2023-01-21 13:39:11.189899: step: 2080/533, loss: 0.0009059906587935984 2023-01-21 13:39:12.393021: step: 2084/533, loss: 0.0008560181013308465 2023-01-21 13:39:13.541433: step: 2088/533, loss: 0.00017938614473678172 2023-01-21 13:39:14.706229: step: 2092/533, loss: 1.773834264895413e-05 2023-01-21 13:39:15.871894: step: 2096/533, loss: 2.9945374990347773e-05 2023-01-21 13:39:17.083108: step: 2100/533, loss: 1.8290832042694092 2023-01-21 13:39:18.256063: step: 2104/533, loss: 0.005245590582489967 2023-01-21 13:39:19.391869: step: 2108/533, loss: 9.34600848268019e-06 2023-01-21 13:39:20.535436: step: 2112/533, loss: 7.05719003235572e-06 2023-01-21 13:39:21.696409: step: 2116/533, loss: 0.00044631960918195546 2023-01-21 13:39:22.866257: step: 2120/533, loss: 0.0003170013369526714 2023-01-21 13:39:24.031838: step: 2124/533, loss: 0.003554105758666992 2023-01-21 13:39:25.212031: step: 2128/533, loss: 1.392364538332913e-05 2023-01-21 13:39:26.385283: step: 2132/533, loss: 0.002403307007625699 ================================================== Loss: 0.016 -------------------- Dev: {'event': {'p': 0.6183206106870229, 'r': 0.7549933422103862, 'f1': 0.6798561151079138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Test: {'event': {'p': 0.6558164354322306, 'r': 0.7858056265984654, 'f1': 0.7149505526468877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} Korean: {'event': {'p': 0.6428571428571429, 'r': 0.5714285714285714, 'f1': 0.6050420168067226}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6183206106870229, 'r': 0.7549933422103862, 'f1': 0.6798561151079138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Chinese: {'event': {'p': 0.6558164354322306, 'r': 0.7858056265984654, 'f1': 0.7149505526468877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 13:40:11.976050: step: 4/533, loss: 7.534028554800898e-06 2023-01-21 13:40:13.138832: step: 8/533, loss: 7.896423630882055e-05 2023-01-21 13:40:14.304115: step: 12/533, loss: 0.0013439179165288806 2023-01-21 13:40:15.474948: step: 16/533, loss: 0.013418961316347122 2023-01-21 13:40:16.617895: step: 20/533, loss: 0.004796028137207031 2023-01-21 13:40:17.846629: step: 24/533, loss: 0.0005672455299645662 2023-01-21 13:40:19.045759: step: 28/533, loss: 0.019650841131806374 2023-01-21 13:40:20.271819: step: 32/533, loss: 0.006049156654626131 2023-01-21 13:40:21.513532: step: 36/533, loss: 3.628731064964086e-05 2023-01-21 13:40:22.661314: step: 40/533, loss: 0.004432487301528454 2023-01-21 13:40:23.825103: step: 44/533, loss: 9.078979928744957e-05 2023-01-21 13:40:25.007759: step: 48/533, loss: 0.0009373664506711066 2023-01-21 13:40:26.161755: step: 52/533, loss: 0.0012390136253088713 2023-01-21 13:40:27.312383: step: 56/533, loss: 0.00015735626220703125 2023-01-21 13:40:28.509308: step: 60/533, loss: 7.781982276355848e-05 2023-01-21 13:40:29.694685: step: 64/533, loss: 0.03666190803050995 2023-01-21 13:40:30.851979: step: 68/533, loss: 0.0004473209264688194 2023-01-21 13:40:32.022578: step: 72/533, loss: 0.001277923583984375 2023-01-21 13:40:33.151766: step: 76/533, loss: -1.1920928955078125e-06 2023-01-21 13:40:34.332681: step: 80/533, loss: 0.02877168543636799 2023-01-21 13:40:35.571670: step: 84/533, loss: 0.0004143714904785156 2023-01-21 13:40:36.779644: step: 88/533, loss: 0.001654815743677318 2023-01-21 13:40:37.983954: step: 92/533, loss: 0.008838463574647903 2023-01-21 13:40:39.187726: step: 96/533, loss: 3.166198803228326e-05 2023-01-21 13:40:40.403147: step: 100/533, loss: 0.06480560451745987 2023-01-21 13:40:41.540284: step: 104/533, loss: 7.247925168485381e-06 2023-01-21 13:40:42.736129: step: 108/533, loss: 0.0008224487537518144 2023-01-21 13:40:43.961929: step: 112/533, loss: 0.00101642613299191 2023-01-21 13:40:45.127063: step: 116/533, loss: 2.880096508306451e-05 2023-01-21 13:40:46.305859: step: 120/533, loss: 0.0006071090465411544 2023-01-21 13:40:47.461471: step: 124/533, loss: 5.092620995128527e-05 2023-01-21 13:40:48.650822: step: 128/533, loss: -8.678436643094756e-06 2023-01-21 13:40:49.788097: step: 132/533, loss: 0.002089214278385043 2023-01-21 13:40:50.982536: step: 136/533, loss: 0.11586780846118927 2023-01-21 13:40:52.184076: step: 140/533, loss: 0.029987145215272903 2023-01-21 13:40:53.379958: step: 144/533, loss: 0.0008866310818120837 2023-01-21 13:40:54.588685: step: 148/533, loss: 0.0030837536323815584 2023-01-21 13:40:55.827605: step: 152/533, loss: 0.5049497485160828 2023-01-21 13:40:57.029683: step: 156/533, loss: 0.0014310836559161544 2023-01-21 13:40:58.202113: step: 160/533, loss: 1.0681153071345761e-05 2023-01-21 13:40:59.369594: step: 164/533, loss: 1.2111663636460435e-05 2023-01-21 13:41:00.578399: step: 168/533, loss: 3.056526111322455e-05 2023-01-21 13:41:01.741342: step: 172/533, loss: 0.0006353378412313759 2023-01-21 13:41:02.982744: step: 176/533, loss: 0.0001543045073049143 2023-01-21 13:41:04.192380: step: 180/533, loss: 9.493828110862523e-05 2023-01-21 13:41:05.378217: step: 184/533, loss: 0.00387668632902205 2023-01-21 13:41:06.599302: step: 188/533, loss: 5.0258637202205136e-05 2023-01-21 13:41:07.756194: step: 192/533, loss: 2.8419495720299892e-05 2023-01-21 13:41:08.938624: step: 196/533, loss: 0.02336139604449272 2023-01-21 13:41:10.129237: step: 200/533, loss: 0.00423088064417243 2023-01-21 13:41:11.331647: step: 204/533, loss: 0.00907211285084486 2023-01-21 13:41:12.551114: step: 208/533, loss: 0.0005422591930255294 2023-01-21 13:41:13.716389: step: 212/533, loss: 0.00043468474177643657 2023-01-21 13:41:14.915548: step: 216/533, loss: -1.1444091796875e-05 2023-01-21 13:41:16.062238: step: 220/533, loss: 0.03506450727581978 2023-01-21 13:41:17.245822: step: 224/533, loss: -3.337860050578456e-07 2023-01-21 13:41:18.443532: step: 228/533, loss: 0.0006381034618243575 2023-01-21 13:41:19.620771: step: 232/533, loss: 2.47955322265625e-05 2023-01-21 13:41:20.793668: step: 236/533, loss: 0.10172824561595917 2023-01-21 13:41:21.986708: step: 240/533, loss: 0.0008962184656411409 2023-01-21 13:41:23.164158: step: 244/533, loss: 0.0015045166946947575 2023-01-21 13:41:24.405182: step: 248/533, loss: 0.0007680415874347091 2023-01-21 13:41:25.579720: step: 252/533, loss: 0.03281812742352486 2023-01-21 13:41:26.746688: step: 256/533, loss: 0.005105876829475164 2023-01-21 13:41:27.925548: step: 260/533, loss: 0.036783602088689804 2023-01-21 13:41:29.089399: step: 264/533, loss: 0.0008500099065713584 2023-01-21 13:41:30.251557: step: 268/533, loss: 1.568794141348917e-05 2023-01-21 13:41:31.440746: step: 272/533, loss: 0.03252124786376953 2023-01-21 13:41:32.625914: step: 276/533, loss: 0.005089378450065851 2023-01-21 13:41:33.819388: step: 280/533, loss: 0.00024280548677779734 2023-01-21 13:41:35.033921: step: 284/533, loss: 0.00330772390589118 2023-01-21 13:41:36.199544: step: 288/533, loss: 0.004314422607421875 2023-01-21 13:41:37.406143: step: 292/533, loss: 0.006544876378029585 2023-01-21 13:41:38.642285: step: 296/533, loss: 0.002794456435367465 2023-01-21 13:41:39.829168: step: 300/533, loss: 0.038854122161865234 2023-01-21 13:41:40.990071: step: 304/533, loss: 0.059088896960020065 2023-01-21 13:41:42.176239: step: 308/533, loss: 3.8623809814453125e-05 2023-01-21 13:41:43.343183: step: 312/533, loss: 6.86645489622606e-06 2023-01-21 13:41:44.541254: step: 316/533, loss: 0.005772209260612726 2023-01-21 13:41:45.701983: step: 320/533, loss: 0.0005887032020837069 2023-01-21 13:41:46.873703: step: 324/533, loss: 5.378723290050402e-05 2023-01-21 13:41:48.078574: step: 328/533, loss: 0.0012348175514489412 2023-01-21 13:41:49.265293: step: 332/533, loss: 0.0017560959095135331 2023-01-21 13:41:50.495260: step: 336/533, loss: 0.009603691287338734 2023-01-21 13:41:51.673671: step: 340/533, loss: 0.025124521926045418 2023-01-21 13:41:52.862131: step: 344/533, loss: 0.0001396179141011089 2023-01-21 13:41:54.105423: step: 348/533, loss: 0.0002620696905069053 2023-01-21 13:41:55.291817: step: 352/533, loss: 0.011796570383012295 2023-01-21 13:41:56.461905: step: 356/533, loss: 4.56810012110509e-05 2023-01-21 13:41:57.659809: step: 360/533, loss: 0.0041792867705225945 2023-01-21 13:41:58.821147: step: 364/533, loss: 1.411438006471144e-05 2023-01-21 13:42:00.070259: step: 368/533, loss: 1.9502640498103574e-05 2023-01-21 13:42:01.268287: step: 372/533, loss: 4.901886131847277e-05 2023-01-21 13:42:02.480368: step: 376/533, loss: 3.24249276673072e-06 2023-01-21 13:42:03.663995: step: 380/533, loss: 0.003913307096809149 2023-01-21 13:42:04.849601: step: 384/533, loss: 0.0003589630068745464 2023-01-21 13:42:06.020048: step: 388/533, loss: 1.907348234908568e-07 2023-01-21 13:42:07.185879: step: 392/533, loss: 0.05034380033612251 2023-01-21 13:42:08.375815: step: 396/533, loss: 0.0002077102690236643 2023-01-21 13:42:09.569257: step: 400/533, loss: 0.0006966590299271047 2023-01-21 13:42:10.754557: step: 404/533, loss: 2.19345088225964e-06 2023-01-21 13:42:11.961510: step: 408/533, loss: 0.0029265405610203743 2023-01-21 13:42:13.211004: step: 412/533, loss: 0.04236021265387535 2023-01-21 13:42:14.408950: step: 416/533, loss: 1.5544892448815517e-05 2023-01-21 13:42:15.607817: step: 420/533, loss: 1.1825562069134321e-05 2023-01-21 13:42:16.787520: step: 424/533, loss: 6.8664553509734105e-06 2023-01-21 13:42:17.998150: step: 428/533, loss: 0.00042285918607376516 2023-01-21 13:42:19.183728: step: 432/533, loss: 0.0018085479969158769 2023-01-21 13:42:20.421188: step: 436/533, loss: 0.0022339820861816406 2023-01-21 13:42:21.613061: step: 440/533, loss: 0.0012959956657141447 2023-01-21 13:42:22.765280: step: 444/533, loss: 0.00014696121797896922 2023-01-21 13:42:23.980857: step: 448/533, loss: 0.022188376635313034 2023-01-21 13:42:25.163570: step: 452/533, loss: 0.0009449005010537803 2023-01-21 13:42:26.366658: step: 456/533, loss: 0.00046324729919433594 2023-01-21 13:42:27.567677: step: 460/533, loss: 0.016710758209228516 2023-01-21 13:42:28.761177: step: 464/533, loss: 0.0005097389221191406 2023-01-21 13:42:29.928206: step: 468/533, loss: 0.0002099990815622732 2023-01-21 13:42:31.099061: step: 472/533, loss: 0.008814096450805664 2023-01-21 13:42:32.282273: step: 476/533, loss: 9.012222290039062e-05 2023-01-21 13:42:33.486123: step: 480/533, loss: 0.00017547608877066523 2023-01-21 13:42:34.671841: step: 484/533, loss: 0.002627849578857422 2023-01-21 13:42:35.902670: step: 488/533, loss: 0.0024357796646654606 2023-01-21 13:42:37.100320: step: 492/533, loss: 0.00024127960205078125 2023-01-21 13:42:38.340617: step: 496/533, loss: 0.015493392944335938 2023-01-21 13:42:39.530752: step: 500/533, loss: 4.7206878662109375e-05 2023-01-21 13:42:40.725601: step: 504/533, loss: 7.009506589383818e-06 2023-01-21 13:42:41.916837: step: 508/533, loss: 3.662109520519152e-05 2023-01-21 13:42:43.085794: step: 512/533, loss: 0.004395675845444202 2023-01-21 13:42:44.348994: step: 516/533, loss: 0.060263827443122864 2023-01-21 13:42:45.525463: step: 520/533, loss: 0.00018377303786110133 2023-01-21 13:42:46.795961: step: 524/533, loss: 0.002733421279117465 2023-01-21 13:42:48.002454: step: 528/533, loss: 0.009518241509795189 2023-01-21 13:42:49.231277: step: 532/533, loss: 0.0005384445539675653 2023-01-21 13:42:50.409882: step: 536/533, loss: 0.0003921031893696636 2023-01-21 13:42:51.616709: step: 540/533, loss: 3.643035961431451e-05 2023-01-21 13:42:52.772373: step: 544/533, loss: 0.0008705139625817537 2023-01-21 13:42:54.017418: step: 548/533, loss: 0.04143171012401581 2023-01-21 13:42:55.217737: step: 552/533, loss: 0.14688262343406677 2023-01-21 13:42:56.397038: step: 556/533, loss: 0.00016708375187590718 2023-01-21 13:42:57.570093: step: 560/533, loss: 0.0003758430539164692 2023-01-21 13:42:58.705479: step: 564/533, loss: 0.024804307147860527 2023-01-21 13:42:59.950776: step: 568/533, loss: 3.7193301523075206e-06 2023-01-21 13:43:01.132841: step: 572/533, loss: 0.0008938789833337069 2023-01-21 13:43:02.319326: step: 576/533, loss: 0.0005808830610476434 2023-01-21 13:43:03.532449: step: 580/533, loss: 0.0007548332214355469 2023-01-21 13:43:04.730664: step: 584/533, loss: 0.037284087389707565 2023-01-21 13:43:05.901283: step: 588/533, loss: 0.00010795592970680445 2023-01-21 13:43:07.082671: step: 592/533, loss: 0.0010578156216070056 2023-01-21 13:43:08.299924: step: 596/533, loss: 0.0027248861733824015 2023-01-21 13:43:09.478587: step: 600/533, loss: 0.005249595735222101 2023-01-21 13:43:10.634311: step: 604/533, loss: 8.02993745310232e-05 2023-01-21 13:43:11.842060: step: 608/533, loss: 0.0167097095400095 2023-01-21 13:43:13.031674: step: 612/533, loss: 0.02184000052511692 2023-01-21 13:43:14.243652: step: 616/533, loss: 0.0019045353401452303 2023-01-21 13:43:15.390272: step: 620/533, loss: 0.00011730194091796875 2023-01-21 13:43:16.572574: step: 624/533, loss: 0.00767097482457757 2023-01-21 13:43:17.749790: step: 628/533, loss: 0.00014634133549407125 2023-01-21 13:43:18.920988: step: 632/533, loss: 0.00572891253978014 2023-01-21 13:43:20.117106: step: 636/533, loss: 5.111694190418348e-05 2023-01-21 13:43:21.314831: step: 640/533, loss: 0.0001430511474609375 2023-01-21 13:43:22.498688: step: 644/533, loss: 0.00017509459576103836 2023-01-21 13:43:23.699635: step: 648/533, loss: 0.0009133339044637978 2023-01-21 13:43:24.866382: step: 652/533, loss: 5.6934357417048886e-05 2023-01-21 13:43:26.053907: step: 656/533, loss: 0.0014568328624591231 2023-01-21 13:43:27.242412: step: 660/533, loss: 0.0003809928894042969 2023-01-21 13:43:28.421947: step: 664/533, loss: 1.3256073543743696e-05 2023-01-21 13:43:29.607396: step: 668/533, loss: 0.0012298584915697575 2023-01-21 13:43:30.787412: step: 672/533, loss: 0.0037435530684888363 2023-01-21 13:43:31.961201: step: 676/533, loss: 0.0036605834029614925 2023-01-21 13:43:33.156872: step: 680/533, loss: 0.0015289306174963713 2023-01-21 13:43:34.363188: step: 684/533, loss: 0.0008705139625817537 2023-01-21 13:43:35.572305: step: 688/533, loss: 0.0019101144280284643 2023-01-21 13:43:36.730252: step: 692/533, loss: 3.43322744811303e-06 2023-01-21 13:43:37.937549: step: 696/533, loss: 0.0003459453582763672 2023-01-21 13:43:39.135381: step: 700/533, loss: 0.00263042445294559 2023-01-21 13:43:40.306646: step: 704/533, loss: 0.0017639159923419356 2023-01-21 13:43:41.474823: step: 708/533, loss: 3.356933666509576e-05 2023-01-21 13:43:42.647062: step: 712/533, loss: 3.490447852527723e-05 2023-01-21 13:43:43.874477: step: 716/533, loss: 2.574920654296875e-05 2023-01-21 13:43:45.064650: step: 720/533, loss: 0.00217266078107059 2023-01-21 13:43:46.277239: step: 724/533, loss: 0.001885795732960105 2023-01-21 13:43:47.470348: step: 728/533, loss: 0.0061120036989450455 2023-01-21 13:43:48.666649: step: 732/533, loss: 0.00022430419630836695 2023-01-21 13:43:49.854396: step: 736/533, loss: 0.0016601084498688579 2023-01-21 13:43:51.052509: step: 740/533, loss: 0.004161739256232977 2023-01-21 13:43:52.229113: step: 744/533, loss: 0.0015872955555096269 2023-01-21 13:43:53.423541: step: 748/533, loss: 3.538131932145916e-05 2023-01-21 13:43:54.625644: step: 752/533, loss: 0.0001951217564055696 2023-01-21 13:43:55.799940: step: 756/533, loss: 1.907349087559851e-07 2023-01-21 13:43:56.989966: step: 760/533, loss: 0.0009496688726358116 2023-01-21 13:43:58.140006: step: 764/533, loss: 0.0004805564822163433 2023-01-21 13:43:59.336273: step: 768/533, loss: 0.002004146808758378 2023-01-21 13:44:00.530077: step: 772/533, loss: 0.0023417354095727205 2023-01-21 13:44:01.715460: step: 776/533, loss: 0.013773537240922451 2023-01-21 13:44:02.902458: step: 780/533, loss: 0.00018539429584052414 2023-01-21 13:44:04.093088: step: 784/533, loss: 0.47250357270240784 2023-01-21 13:44:05.270221: step: 788/533, loss: -9.536743306171047e-08 2023-01-21 13:44:06.424018: step: 792/533, loss: 6.198883056640625e-06 2023-01-21 13:44:07.597255: step: 796/533, loss: 0.003001594450324774 2023-01-21 13:44:08.740666: step: 800/533, loss: 1.106262243411038e-05 2023-01-21 13:44:09.891877: step: 804/533, loss: 2.95162208203692e-05 2023-01-21 13:44:11.052473: step: 808/533, loss: 0.013832474127411842 2023-01-21 13:44:12.231613: step: 812/533, loss: 0.0277315154671669 2023-01-21 13:44:13.405305: step: 816/533, loss: 0.05567779764533043 2023-01-21 13:44:14.559404: step: 820/533, loss: 8.893013728084043e-05 2023-01-21 13:44:15.720866: step: 824/533, loss: 6.351471529342234e-05 2023-01-21 13:44:16.917070: step: 828/533, loss: 6.46591215627268e-05 2023-01-21 13:44:18.089264: step: 832/533, loss: 0.0013504029484465718 2023-01-21 13:44:19.291423: step: 836/533, loss: 0.00016469955153297633 2023-01-21 13:44:20.438587: step: 840/533, loss: 1.773834264895413e-05 2023-01-21 13:44:21.627364: step: 844/533, loss: 0.004299211781471968 2023-01-21 13:44:22.813184: step: 848/533, loss: 0.007451486308127642 2023-01-21 13:44:23.978786: step: 852/533, loss: 0.000179290771484375 2023-01-21 13:44:25.179777: step: 856/533, loss: 0.0008804321405477822 2023-01-21 13:44:26.353351: step: 860/533, loss: 0.0007562637329101562 2023-01-21 13:44:27.555691: step: 864/533, loss: 0.4774315655231476 2023-01-21 13:44:28.779310: step: 868/533, loss: 0.007146644406020641 2023-01-21 13:44:29.974249: step: 872/533, loss: 0.0006508827209472656 2023-01-21 13:44:31.158896: step: 876/533, loss: 0.00016288757615257055 2023-01-21 13:44:32.337159: step: 880/533, loss: 0.008348274044692516 2023-01-21 13:44:33.510450: step: 884/533, loss: 4.00543212890625e-05 2023-01-21 13:44:34.672463: step: 888/533, loss: 0.0050824168138206005 2023-01-21 13:44:35.860785: step: 892/533, loss: 0.0004955768818035722 2023-01-21 13:44:37.051289: step: 896/533, loss: 0.03819713369011879 2023-01-21 13:44:38.282611: step: 900/533, loss: 0.0019178390502929688 2023-01-21 13:44:39.438292: step: 904/533, loss: 0.0057343486696481705 2023-01-21 13:44:40.626456: step: 908/533, loss: 1.9931792849092744e-05 2023-01-21 13:44:41.773232: step: 912/533, loss: -5.245208853921213e-07 2023-01-21 13:44:42.923997: step: 916/533, loss: 0.00013856888108421117 2023-01-21 13:44:44.130838: step: 920/533, loss: 0.002774286549538374 2023-01-21 13:44:45.305537: step: 924/533, loss: 1.697540210443549e-05 2023-01-21 13:44:46.458102: step: 928/533, loss: 0.008076668716967106 2023-01-21 13:44:47.648481: step: 932/533, loss: 2.86102294921875e-06 2023-01-21 13:44:48.823341: step: 936/533, loss: 2.8419495720299892e-05 2023-01-21 13:44:49.949243: step: 940/533, loss: 0.0008134842501021922 2023-01-21 13:44:51.117114: step: 944/533, loss: 0.0012616157764568925 2023-01-21 13:44:52.309008: step: 948/533, loss: 9.040832810569555e-05 2023-01-21 13:44:53.467486: step: 952/533, loss: 0.0010336876148357987 2023-01-21 13:44:54.665528: step: 956/533, loss: 0.0003501892206259072 2023-01-21 13:44:55.838743: step: 960/533, loss: 0.0036535265389829874 2023-01-21 13:44:57.044900: step: 964/533, loss: 0.00019607543072197586 2023-01-21 13:44:58.218446: step: 968/533, loss: 2.1123885744600557e-05 2023-01-21 13:44:59.440179: step: 972/533, loss: 0.00491986284032464 2023-01-21 13:45:00.618191: step: 976/533, loss: 0.007938195019960403 2023-01-21 13:45:01.774352: step: 980/533, loss: 4.653930591302924e-05 2023-01-21 13:45:02.976779: step: 984/533, loss: 0.00041484832763671875 2023-01-21 13:45:04.144096: step: 988/533, loss: 0.048197317868471146 2023-01-21 13:45:05.265381: step: 992/533, loss: 0.025508642196655273 2023-01-21 13:45:06.420373: step: 996/533, loss: 0.021366501227021217 2023-01-21 13:45:07.610425: step: 1000/533, loss: 4.95910626341356e-06 2023-01-21 13:45:08.796323: step: 1004/533, loss: 0.000618457852397114 2023-01-21 13:45:09.992679: step: 1008/533, loss: 0.07714968174695969 2023-01-21 13:45:11.169337: step: 1012/533, loss: 5.14984130859375e-05 2023-01-21 13:45:12.362776: step: 1016/533, loss: 1.6021729607018642e-05 2023-01-21 13:45:13.553618: step: 1020/533, loss: 2.346038854739163e-05 2023-01-21 13:45:14.757561: step: 1024/533, loss: 0.0005279540782794356 2023-01-21 13:45:15.925229: step: 1028/533, loss: 1.697540210443549e-05 2023-01-21 13:45:17.099162: step: 1032/533, loss: 1.3303756531968247e-05 2023-01-21 13:45:18.278493: step: 1036/533, loss: 9.078979201149195e-05 2023-01-21 13:45:19.458757: step: 1040/533, loss: 0.00485992431640625 2023-01-21 13:45:20.644773: step: 1044/533, loss: 0.004650878719985485 2023-01-21 13:45:21.824645: step: 1048/533, loss: 0.0032660483848303556 2023-01-21 13:45:23.002009: step: 1052/533, loss: 0.03774967044591904 2023-01-21 13:45:24.158467: step: 1056/533, loss: 0.023192785680294037 2023-01-21 13:45:25.352653: step: 1060/533, loss: 0.007928658276796341 2023-01-21 13:45:26.510130: step: 1064/533, loss: 0.00019159316434524953 2023-01-21 13:45:27.709499: step: 1068/533, loss: 0.028194429352879524 2023-01-21 13:45:28.884283: step: 1072/533, loss: 0.00010051727440441027 2023-01-21 13:45:30.043461: step: 1076/533, loss: 3.376007225597277e-05 2023-01-21 13:45:31.202058: step: 1080/533, loss: 2.207756187999621e-05 2023-01-21 13:45:32.365622: step: 1084/533, loss: 0.0008795738103799522 2023-01-21 13:45:33.522793: step: 1088/533, loss: 9.784698340808973e-05 2023-01-21 13:45:34.713475: step: 1092/533, loss: 0.012301064096391201 2023-01-21 13:45:35.874698: step: 1096/533, loss: 1.3446808225126006e-05 2023-01-21 13:45:37.055094: step: 1100/533, loss: 1.5258789289873675e-06 2023-01-21 13:45:38.245739: step: 1104/533, loss: 0.014218902215361595 2023-01-21 13:45:39.423077: step: 1108/533, loss: 0.0026243210304528475 2023-01-21 13:45:40.579638: step: 1112/533, loss: 7.152557373046875e-06 2023-01-21 13:45:41.797605: step: 1116/533, loss: 0.0006278991932049394 2023-01-21 13:45:42.981004: step: 1120/533, loss: 0.00031414031400345266 2023-01-21 13:45:44.134501: step: 1124/533, loss: 8.087158494163305e-05 2023-01-21 13:45:45.327334: step: 1128/533, loss: 5.397796485340223e-05 2023-01-21 13:45:46.509395: step: 1132/533, loss: 0.010439587756991386 2023-01-21 13:45:47.692725: step: 1136/533, loss: 2.47955322265625e-05 2023-01-21 13:45:48.937498: step: 1140/533, loss: 0.007459640968590975 2023-01-21 13:45:50.141401: step: 1144/533, loss: 0.0021137238945811987 2023-01-21 13:45:51.383751: step: 1148/533, loss: 4.129410081077367e-05 2023-01-21 13:45:52.575194: step: 1152/533, loss: 1.5163421267061494e-05 2023-01-21 13:45:53.809867: step: 1156/533, loss: 1.7166138377433526e-06 2023-01-21 13:45:55.009880: step: 1160/533, loss: 0.03963002935051918 2023-01-21 13:45:56.196533: step: 1164/533, loss: 0.08506827056407928 2023-01-21 13:45:57.387639: step: 1168/533, loss: 0.0009277343633584678 2023-01-21 13:45:58.521351: step: 1172/533, loss: 0.011252212338149548 2023-01-21 13:45:59.656955: step: 1176/533, loss: 0.0007089615100994706 2023-01-21 13:46:00.833878: step: 1180/533, loss: 0.0010353088146075606 2023-01-21 13:46:02.010584: step: 1184/533, loss: 0.0006683349492959678 2023-01-21 13:46:03.220402: step: 1188/533, loss: 0.592664897441864 2023-01-21 13:46:04.398026: step: 1192/533, loss: 0.0042938231490552425 2023-01-21 13:46:05.562879: step: 1196/533, loss: 0.012715148739516735 2023-01-21 13:46:06.774920: step: 1200/533, loss: 0.1252616047859192 2023-01-21 13:46:07.964827: step: 1204/533, loss: 0.0007080078357830644 2023-01-21 13:46:09.147088: step: 1208/533, loss: 0.006268978118896484 2023-01-21 13:46:10.325981: step: 1212/533, loss: 7.534027099609375e-05 2023-01-21 13:46:11.550370: step: 1216/533, loss: 0.02620716206729412 2023-01-21 13:46:12.759910: step: 1220/533, loss: 0.033933356404304504 2023-01-21 13:46:13.997482: step: 1224/533, loss: 5.617141869151965e-05 2023-01-21 13:46:15.163538: step: 1228/533, loss: 0.0019325256580486894 2023-01-21 13:46:16.369065: step: 1232/533, loss: 2.3555756342830136e-05 2023-01-21 13:46:17.538003: step: 1236/533, loss: 0.0031822205055505037 2023-01-21 13:46:18.709634: step: 1240/533, loss: 1.1157988410559483e-05 2023-01-21 13:46:19.904373: step: 1244/533, loss: 0.0002727508544921875 2023-01-21 13:46:21.052841: step: 1248/533, loss: 0.00011658668518066406 2023-01-21 13:46:22.250425: step: 1252/533, loss: 0.054642681032419205 2023-01-21 13:46:23.385279: step: 1256/533, loss: 0.00229721050709486 2023-01-21 13:46:24.559744: step: 1260/533, loss: 0.002063751220703125 2023-01-21 13:46:25.729603: step: 1264/533, loss: 0.06561774760484695 2023-01-21 13:46:26.942464: step: 1268/533, loss: 0.00047388079110533 2023-01-21 13:46:28.133342: step: 1272/533, loss: 0.00015816689119674265 2023-01-21 13:46:29.316240: step: 1276/533, loss: 2.593994213384576e-05 2023-01-21 13:46:30.515462: step: 1280/533, loss: 0.01809406280517578 2023-01-21 13:46:31.695806: step: 1284/533, loss: 0.00015687942504882812 2023-01-21 13:46:32.861422: step: 1288/533, loss: 0.0005552291986532509 2023-01-21 13:46:34.016639: step: 1292/533, loss: 0.008483218960464 2023-01-21 13:46:35.135582: step: 1296/533, loss: 6.084441702114418e-05 2023-01-21 13:46:36.345960: step: 1300/533, loss: -3.6239616747479886e-06 2023-01-21 13:46:37.508264: step: 1304/533, loss: 8.687973604537547e-05 2023-01-21 13:46:38.675611: step: 1308/533, loss: 4.0626528061693534e-05 2023-01-21 13:46:39.878861: step: 1312/533, loss: 0.008015441708266735 2023-01-21 13:46:41.040847: step: 1316/533, loss: 0.0031057356391102076 2023-01-21 13:46:42.261541: step: 1320/533, loss: 4.634857032215223e-05 2023-01-21 13:46:43.410260: step: 1324/533, loss: 0.0013029099209234118 2023-01-21 13:46:44.604167: step: 1328/533, loss: 1.4686585018353071e-05 2023-01-21 13:46:45.789616: step: 1332/533, loss: 8.454322960460559e-05 2023-01-21 13:46:46.973125: step: 1336/533, loss: 0.001157474471256137 2023-01-21 13:46:48.167483: step: 1340/533, loss: 0.06334400177001953 2023-01-21 13:46:49.353361: step: 1344/533, loss: 0.004132700152695179 2023-01-21 13:46:50.483686: step: 1348/533, loss: 8.525847806595266e-05 2023-01-21 13:46:51.734528: step: 1352/533, loss: 0.0012422562576830387 2023-01-21 13:46:52.928011: step: 1356/533, loss: 0.00020294189744163305 2023-01-21 13:46:54.113309: step: 1360/533, loss: 0.01821422576904297 2023-01-21 13:46:55.287824: step: 1364/533, loss: 0.00543289165943861 2023-01-21 13:46:56.475793: step: 1368/533, loss: 0.0052276612259447575 2023-01-21 13:46:57.648174: step: 1372/533, loss: 0.0002896309015341103 2023-01-21 13:46:58.769150: step: 1376/533, loss: 3.337860107421875e-05 2023-01-21 13:46:59.952503: step: 1380/533, loss: 0.006535553839057684 2023-01-21 13:47:01.108327: step: 1384/533, loss: 6.198883511387976e-06 2023-01-21 13:47:02.264566: step: 1388/533, loss: 5.340576171875e-05 2023-01-21 13:47:03.468803: step: 1392/533, loss: 0.004298400599509478 2023-01-21 13:47:04.620853: step: 1396/533, loss: 5.664825584972277e-05 2023-01-21 13:47:05.751429: step: 1400/533, loss: 0.00010757446580100805 2023-01-21 13:47:06.941578: step: 1404/533, loss: 0.018986891955137253 2023-01-21 13:47:08.093427: step: 1408/533, loss: 0.00041837693424895406 2023-01-21 13:47:09.237091: step: 1412/533, loss: 0.023760223761200905 2023-01-21 13:47:10.401735: step: 1416/533, loss: 0.0023074152413755655 2023-01-21 13:47:11.589599: step: 1420/533, loss: 0.0027145384810864925 2023-01-21 13:47:12.796247: step: 1424/533, loss: 0.0033185959327965975 2023-01-21 13:47:13.974189: step: 1428/533, loss: 0.002271270612254739 2023-01-21 13:47:15.151609: step: 1432/533, loss: 7.371902756858617e-05 2023-01-21 13:47:16.300653: step: 1436/533, loss: 0.0003232002491131425 2023-01-21 13:47:17.458892: step: 1440/533, loss: 0.00011377334885764867 2023-01-21 13:47:18.615259: step: 1444/533, loss: 0.0015453338855877519 2023-01-21 13:47:19.754628: step: 1448/533, loss: 0.0002846241113729775 2023-01-21 13:47:21.020983: step: 1452/533, loss: 0.0038608552422374487 2023-01-21 13:47:22.194903: step: 1456/533, loss: 0.0005678177112713456 2023-01-21 13:47:23.407485: step: 1460/533, loss: 0.01827402226626873 2023-01-21 13:47:24.564771: step: 1464/533, loss: 0.04267864301800728 2023-01-21 13:47:25.696593: step: 1468/533, loss: 2.002716064453125e-05 2023-01-21 13:47:26.883954: step: 1472/533, loss: 0.000370216352166608 2023-01-21 13:47:28.061183: step: 1476/533, loss: 0.0009328365558758378 2023-01-21 13:47:29.276690: step: 1480/533, loss: -9.536742595628311e-08 2023-01-21 13:47:30.455133: step: 1484/533, loss: 0.017430400475859642 2023-01-21 13:47:31.696572: step: 1488/533, loss: 0.00017061232938431203 2023-01-21 13:47:32.933566: step: 1492/533, loss: 0.000292873359285295 2023-01-21 13:47:34.103968: step: 1496/533, loss: 0.00029964448185637593 2023-01-21 13:47:35.289570: step: 1500/533, loss: 0.00011558533151401207 2023-01-21 13:47:36.481986: step: 1504/533, loss: 0.0001277923583984375 2023-01-21 13:47:37.728535: step: 1508/533, loss: 0.00012931822857353836 2023-01-21 13:47:38.950647: step: 1512/533, loss: 0.009451198391616344 2023-01-21 13:47:40.145724: step: 1516/533, loss: 0.029564427211880684 2023-01-21 13:47:41.307404: step: 1520/533, loss: 0.0017645478947088122 2023-01-21 13:47:42.449906: step: 1524/533, loss: 1.9073486328125e-05 2023-01-21 13:47:43.672611: step: 1528/533, loss: 5.769729796156753e-06 2023-01-21 13:47:44.845033: step: 1532/533, loss: -9.5367431640625e-06 2023-01-21 13:47:46.029634: step: 1536/533, loss: 0.0004646778106689453 2023-01-21 13:47:47.214786: step: 1540/533, loss: 1.9359587895451114e-05 2023-01-21 13:47:48.400245: step: 1544/533, loss: 7.400513277389109e-05 2023-01-21 13:47:49.610925: step: 1548/533, loss: 0.05515308305621147 2023-01-21 13:47:50.799091: step: 1552/533, loss: 0.00010905266390182078 2023-01-21 13:47:51.959074: step: 1556/533, loss: 4.38690176451928e-06 2023-01-21 13:47:53.134069: step: 1560/533, loss: 3.5905839467886835e-05 2023-01-21 13:47:54.301780: step: 1564/533, loss: 2.613067590573337e-05 2023-01-21 13:47:55.509821: step: 1568/533, loss: 2.1457681214087643e-06 2023-01-21 13:47:56.705486: step: 1572/533, loss: 0.0020317076705396175 2023-01-21 13:47:57.947978: step: 1576/533, loss: 0.0002750396670307964 2023-01-21 13:47:59.122866: step: 1580/533, loss: 0.0011255263816565275 2023-01-21 13:48:00.324656: step: 1584/533, loss: 1.049041748046875e-05 2023-01-21 13:48:01.501155: step: 1588/533, loss: 3.6716461181640625e-05 2023-01-21 13:48:02.701378: step: 1592/533, loss: 0.00021114348783157766 2023-01-21 13:48:03.915536: step: 1596/533, loss: 0.10731296986341476 2023-01-21 13:48:05.119227: step: 1600/533, loss: 1.621246337890625e-05 2023-01-21 13:48:06.352018: step: 1604/533, loss: -4.19616708313697e-06 2023-01-21 13:48:07.521784: step: 1608/533, loss: 2.3174285161076114e-05 2023-01-21 13:48:08.693592: step: 1612/533, loss: 1.8787384760798886e-05 2023-01-21 13:48:09.903971: step: 1616/533, loss: 0.0007943630334921181 2023-01-21 13:48:11.096763: step: 1620/533, loss: 4.615783836925402e-05 2023-01-21 13:48:12.322743: step: 1624/533, loss: 0.0007746696355752647 2023-01-21 13:48:13.528304: step: 1628/533, loss: 0.0026132583152502775 2023-01-21 13:48:14.711605: step: 1632/533, loss: 0.17014513909816742 2023-01-21 13:48:15.916438: step: 1636/533, loss: -1.029968279908644e-05 2023-01-21 13:48:17.079306: step: 1640/533, loss: 0.0020588873885571957 2023-01-21 13:48:18.310521: step: 1644/533, loss: -9.346007573185489e-06 2023-01-21 13:48:19.494909: step: 1648/533, loss: 1.392364538332913e-05 2023-01-21 13:48:20.688798: step: 1652/533, loss: 0.0018345832359045744 2023-01-21 13:48:21.860055: step: 1656/533, loss: 1.449584942747606e-05 2023-01-21 13:48:23.015672: step: 1660/533, loss: -1.8119811784345075e-06 2023-01-21 13:48:24.186588: step: 1664/533, loss: 0.00015630721463821828 2023-01-21 13:48:25.408943: step: 1668/533, loss: 2.4223329091910273e-05 2023-01-21 13:48:26.614213: step: 1672/533, loss: 0.0002478599490132183 2023-01-21 13:48:27.804414: step: 1676/533, loss: 0.0008592605590820312 2023-01-21 13:48:28.982679: step: 1680/533, loss: 0.00012741087994072586 2023-01-21 13:48:30.189144: step: 1684/533, loss: 0.017216874286532402 2023-01-21 13:48:31.346333: step: 1688/533, loss: -1.1348724001436494e-05 2023-01-21 13:48:32.507930: step: 1692/533, loss: 4.472732689464465e-05 2023-01-21 13:48:33.693431: step: 1696/533, loss: 1.7929078239831142e-05 2023-01-21 13:48:34.857111: step: 1700/533, loss: 0.0003194809250999242 2023-01-21 13:48:36.052394: step: 1704/533, loss: 0.01260919589549303 2023-01-21 13:48:37.233214: step: 1708/533, loss: 0.06233177334070206 2023-01-21 13:48:38.450287: step: 1712/533, loss: 0.0012255193432793021 2023-01-21 13:48:39.621652: step: 1716/533, loss: 0.00013828277587890625 2023-01-21 13:48:40.824132: step: 1720/533, loss: 4.024505687993951e-05 2023-01-21 13:48:42.030042: step: 1724/533, loss: 0.01676464080810547 2023-01-21 13:48:43.223091: step: 1728/533, loss: 6.45637555862777e-05 2023-01-21 13:48:44.400329: step: 1732/533, loss: 0.023438740521669388 2023-01-21 13:48:45.563674: step: 1736/533, loss: 0.011902046389877796 2023-01-21 13:48:46.759442: step: 1740/533, loss: 0.10964813083410263 2023-01-21 13:48:47.975624: step: 1744/533, loss: 4.806518700206652e-05 2023-01-21 13:48:49.187254: step: 1748/533, loss: 0.0003323554992675781 2023-01-21 13:48:50.366539: step: 1752/533, loss: 0.006012439727783203 2023-01-21 13:48:51.543736: step: 1756/533, loss: 0.00010862351336982101 2023-01-21 13:48:52.744766: step: 1760/533, loss: 4.148483185417717e-06 2023-01-21 13:48:53.934242: step: 1764/533, loss: 0.011141777038574219 2023-01-21 13:48:55.113267: step: 1768/533, loss: 0.0028526782989501953 2023-01-21 13:48:56.293983: step: 1772/533, loss: 0.007021522149443626 2023-01-21 13:48:57.484037: step: 1776/533, loss: 0.03824453428387642 2023-01-21 13:48:58.703747: step: 1780/533, loss: 0.0009559631580486894 2023-01-21 13:48:59.905234: step: 1784/533, loss: 0.0009809493785724044 2023-01-21 13:49:01.068951: step: 1788/533, loss: 0.0008879661909304559 2023-01-21 13:49:02.313236: step: 1792/533, loss: 0.00031299592228606343 2023-01-21 13:49:03.512293: step: 1796/533, loss: 2.3412703740177676e-05 2023-01-21 13:49:04.713687: step: 1800/533, loss: 0.023003196343779564 2023-01-21 13:49:05.911132: step: 1804/533, loss: 0.059653475880622864 2023-01-21 13:49:07.151351: step: 1808/533, loss: 0.00021667480177711695 2023-01-21 13:49:08.364585: step: 1812/533, loss: 0.02505521848797798 2023-01-21 13:49:09.615840: step: 1816/533, loss: 3.776550147449598e-05 2023-01-21 13:49:10.770569: step: 1820/533, loss: 5.826950291520916e-05 2023-01-21 13:49:12.020737: step: 1824/533, loss: 0.03996095806360245 2023-01-21 13:49:13.203736: step: 1828/533, loss: 0.04258919134736061 2023-01-21 13:49:14.393293: step: 1832/533, loss: 0.00017242431931663305 2023-01-21 13:49:15.646326: step: 1836/533, loss: 0.00022563934908248484 2023-01-21 13:49:16.855352: step: 1840/533, loss: 2.6893614631262608e-05 2023-01-21 13:49:18.045715: step: 1844/533, loss: 0.02109680138528347 2023-01-21 13:49:19.238188: step: 1848/533, loss: 5.168914867681451e-05 2023-01-21 13:49:20.428597: step: 1852/533, loss: 0.0014940262772142887 2023-01-21 13:49:21.723913: step: 1856/533, loss: 0.0011319160694256425 2023-01-21 13:49:22.927317: step: 1860/533, loss: 0.002082443330436945 2023-01-21 13:49:24.089988: step: 1864/533, loss: 9.536745437799254e-08 2023-01-21 13:49:25.320762: step: 1868/533, loss: 1.564025842526462e-05 2023-01-21 13:49:26.495146: step: 1872/533, loss: 0.03885927051305771 2023-01-21 13:49:27.683630: step: 1876/533, loss: 0.00048704148503020406 2023-01-21 13:49:28.876730: step: 1880/533, loss: 9.813308861339465e-05 2023-01-21 13:49:30.025045: step: 1884/533, loss: 6.446838960982859e-05 2023-01-21 13:49:31.215372: step: 1888/533, loss: 0.004540538880974054 2023-01-21 13:49:32.440391: step: 1892/533, loss: 2.040863000729587e-05 2023-01-21 13:49:33.615739: step: 1896/533, loss: 0.0002704620419535786 2023-01-21 13:49:34.794911: step: 1900/533, loss: 1.9073468138230965e-07 2023-01-21 13:49:35.973160: step: 1904/533, loss: 0.0018741608364507556 2023-01-21 13:49:37.163449: step: 1908/533, loss: 5.455017526401207e-05 2023-01-21 13:49:38.329382: step: 1912/533, loss: 0.001864624093286693 2023-01-21 13:49:39.530379: step: 1916/533, loss: 5.1164628530386835e-05 2023-01-21 13:49:40.704786: step: 1920/533, loss: 3.2138825190486386e-05 2023-01-21 13:49:41.904987: step: 1924/533, loss: 0.007644557394087315 2023-01-21 13:49:43.111308: step: 1928/533, loss: 0.012665558606386185 2023-01-21 13:49:44.297323: step: 1932/533, loss: 1.4972687495173886e-05 2023-01-21 13:49:45.465349: step: 1936/533, loss: 0.0006610869895666838 2023-01-21 13:49:46.605014: step: 1940/533, loss: 3.299712989246473e-05 2023-01-21 13:49:47.752005: step: 1944/533, loss: -1.573562599332945e-06 2023-01-21 13:49:48.900481: step: 1948/533, loss: 3.9601327443961054e-05 2023-01-21 13:49:50.087200: step: 1952/533, loss: 0.0003427505725994706 2023-01-21 13:49:51.277497: step: 1956/533, loss: 0.014409447088837624 2023-01-21 13:49:52.466831: step: 1960/533, loss: 0.026736069470643997 2023-01-21 13:49:53.671287: step: 1964/533, loss: 0.0008687973604537547 2023-01-21 13:49:54.887961: step: 1968/533, loss: 0.0003067016659770161 2023-01-21 13:49:56.053085: step: 1972/533, loss: 2.6702880859375e-05 2023-01-21 13:49:57.235845: step: 1976/533, loss: 0.0012264251708984375 2023-01-21 13:49:58.401585: step: 1980/533, loss: 2.9659271604032256e-05 2023-01-21 13:49:59.629796: step: 1984/533, loss: 0.0006897926214151084 2023-01-21 13:50:00.807776: step: 1988/533, loss: 0.01601553149521351 2023-01-21 13:50:02.045828: step: 1992/533, loss: 0.03540649637579918 2023-01-21 13:50:03.222994: step: 1996/533, loss: 0.0002895355282817036 2023-01-21 13:50:04.415611: step: 2000/533, loss: 0.00019130707369185984 2023-01-21 13:50:05.643142: step: 2004/533, loss: 0.0019168853759765625 2023-01-21 13:50:06.801120: step: 2008/533, loss: 1.964569128176663e-05 2023-01-21 13:50:07.994881: step: 2012/533, loss: 0.00013847350783180445 2023-01-21 13:50:09.192140: step: 2016/533, loss: 0.0010629654861986637 2023-01-21 13:50:10.383856: step: 2020/533, loss: 0.005642128176987171 2023-01-21 13:50:11.546031: step: 2024/533, loss: 0.004029464907944202 2023-01-21 13:50:12.726236: step: 2028/533, loss: 0.0009443283197470009 2023-01-21 13:50:13.927055: step: 2032/533, loss: 0.00011205673217773438 2023-01-21 13:50:15.089749: step: 2036/533, loss: 0.00021800995455123484 2023-01-21 13:50:16.283419: step: 2040/533, loss: 6.294250852079131e-06 2023-01-21 13:50:17.441457: step: 2044/533, loss: 0.005199051462113857 2023-01-21 13:50:18.695398: step: 2048/533, loss: 0.01759166643023491 2023-01-21 13:50:19.883280: step: 2052/533, loss: 0.013919640332460403 2023-01-21 13:50:21.074643: step: 2056/533, loss: 0.39671382308006287 2023-01-21 13:50:22.216757: step: 2060/533, loss: 1.430511474609375e-06 2023-01-21 13:50:23.406727: step: 2064/533, loss: 5.297661118675023e-05 2023-01-21 13:50:24.600219: step: 2068/533, loss: 4.95910626341356e-06 2023-01-21 13:50:25.799557: step: 2072/533, loss: 7.543563697254285e-05 2023-01-21 13:50:26.969214: step: 2076/533, loss: 0.04627237096428871 2023-01-21 13:50:28.144378: step: 2080/533, loss: 0.003780079074203968 2023-01-21 13:50:29.402798: step: 2084/533, loss: 6.014108657836914e-05 2023-01-21 13:50:30.575071: step: 2088/533, loss: 0.0005304336664266884 2023-01-21 13:50:31.778016: step: 2092/533, loss: 0.0005899906391277909 2023-01-21 13:50:32.993697: step: 2096/533, loss: 0.0004552841419354081 2023-01-21 13:50:34.193713: step: 2100/533, loss: 0.00018424988957121968 2023-01-21 13:50:35.361682: step: 2104/533, loss: 0.028882216662168503 2023-01-21 13:50:36.548376: step: 2108/533, loss: 6.151199340820312e-05 2023-01-21 13:50:37.719068: step: 2112/533, loss: 0.002463913056999445 2023-01-21 13:50:38.915313: step: 2116/533, loss: 0.012499427422881126 2023-01-21 13:50:40.049536: step: 2120/533, loss: 1.3494492122845259e-05 2023-01-21 13:50:41.253990: step: 2124/533, loss: 0.026134587824344635 2023-01-21 13:50:42.398418: step: 2128/533, loss: -8.416175660386216e-06 2023-01-21 13:50:43.551238: step: 2132/533, loss: 2.059936559817288e-05 ================================================== Loss: 0.012 -------------------- Dev: {'event': {'p': 0.5916496945010183, 'r': 0.7736351531291611, 'f1': 0.6705135603000577}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Test: {'event': {'p': 0.6411167512690356, 'r': 0.8075447570332481, 'f1': 0.7147707979626485}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Chinese: {'event': {'p': 0.5833333333333334, 'r': 0.9074074074074074, 'f1': 0.7101449275362318}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Korean: {'event': {'p': 0.5238095238095238, 'r': 0.5238095238095238, 'f1': 0.5238095238095238}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Russian: {'event': {'p': 0.3684210526315789, 'r': 0.3888888888888889, 'f1': 0.37837837837837834}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6183206106870229, 'r': 0.7549933422103862, 'f1': 0.6798561151079138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Chinese: {'event': {'p': 0.6558164354322306, 'r': 0.7858056265984654, 'f1': 0.7149505526468877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** 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 13:51:25.463578: step: 4/533, loss: 7.963180541992188e-05 2023-01-21 13:51:26.674833: step: 8/533, loss: 0.0004528045537881553 2023-01-21 13:51:27.851303: step: 12/533, loss: 2.6798248654813506e-05 2023-01-21 13:51:29.025117: step: 16/533, loss: 0.0009815215598791838 2023-01-21 13:51:30.253597: step: 20/533, loss: 0.021444130688905716 2023-01-21 13:51:31.444303: step: 24/533, loss: 0.003556156065315008 2023-01-21 13:51:32.638599: step: 28/533, loss: -4.00543194700731e-06 2023-01-21 13:51:33.869438: step: 32/533, loss: 0.004832458682358265 2023-01-21 13:51:35.036506: step: 36/533, loss: 7.362365431617945e-05 2023-01-21 13:51:36.221672: step: 40/533, loss: 1.487731969973538e-05 2023-01-21 13:51:37.410415: step: 44/533, loss: 5.798339770990424e-05 2023-01-21 13:51:38.574809: step: 48/533, loss: 0.001138973282650113 2023-01-21 13:51:39.774842: step: 52/533, loss: 0.001449632691219449 2023-01-21 13:51:40.967394: step: 56/533, loss: 0.014890670776367188 2023-01-21 13:51:42.122654: step: 60/533, loss: 0.005891228094696999 2023-01-21 13:51:43.311592: step: 64/533, loss: 0.014198875986039639 2023-01-21 13:51:44.522367: step: 68/533, loss: 5.855560448253527e-05 2023-01-21 13:51:45.693329: step: 72/533, loss: 0.003890609834343195 2023-01-21 13:51:46.909173: step: 76/533, loss: 3.128051685052924e-05 2023-01-21 13:51:48.123248: step: 80/533, loss: 0.008204556070268154 2023-01-21 13:51:49.303366: step: 84/533, loss: 0.0004915714380331337 2023-01-21 13:51:50.446312: step: 88/533, loss: 0.006618499755859375 2023-01-21 13:51:51.636674: step: 92/533, loss: 8.125305612338707e-05 2023-01-21 13:51:52.846445: step: 96/533, loss: 0.00039987562922760844 2023-01-21 13:51:53.999408: step: 100/533, loss: 0.02913045883178711 2023-01-21 13:51:55.183136: step: 104/533, loss: 2.670288040462765e-06 2023-01-21 13:51:56.365820: step: 108/533, loss: 0.00022010803513694555 2023-01-21 13:51:57.487498: step: 112/533, loss: 5.42163870704826e-05 2023-01-21 13:51:58.674041: step: 116/533, loss: 1.9264220100012608e-05 2023-01-21 13:51:59.846313: step: 120/533, loss: 0.0003942489856854081 2023-01-21 13:52:01.021620: step: 124/533, loss: 1.1730194273695815e-05 2023-01-21 13:52:02.210884: step: 128/533, loss: -1.0490418844710803e-06 2023-01-21 13:52:03.422549: step: 132/533, loss: 0.003448104951530695 2023-01-21 13:52:04.608512: step: 136/533, loss: 2.346038854739163e-05 2023-01-21 13:52:05.788436: step: 140/533, loss: 1.220703143189894e-05 2023-01-21 13:52:06.918460: step: 144/533, loss: 0.0005262374761514366 2023-01-21 13:52:08.104245: step: 148/533, loss: 0.005027580540627241 2023-01-21 13:52:09.351649: step: 152/533, loss: 0.00013599396334029734 2023-01-21 13:52:10.557020: step: 156/533, loss: 0.00015163421630859375 2023-01-21 13:52:11.730301: step: 160/533, loss: 3.0040741876291577e-06 2023-01-21 13:52:12.902440: step: 164/533, loss: 2.474784923833795e-05 2023-01-21 13:52:14.089513: step: 168/533, loss: 7.672310312045738e-05 2023-01-21 13:52:15.265179: step: 172/533, loss: 2.8038026357535273e-05 2023-01-21 13:52:16.463148: step: 176/533, loss: -2.86102294921875e-06 2023-01-21 13:52:17.625791: step: 180/533, loss: 6.84738188283518e-05 2023-01-21 13:52:18.830260: step: 184/533, loss: 0.37682199478149414 2023-01-21 13:52:20.004648: step: 188/533, loss: 1.7642974853515625e-05 2023-01-21 13:52:21.218775: step: 192/533, loss: 1.0442733582749497e-05 2023-01-21 13:52:22.401729: step: 196/533, loss: 5.073547436040826e-05 2023-01-21 13:52:23.581583: step: 200/533, loss: 0.0001874923618743196 2023-01-21 13:52:24.779445: step: 204/533, loss: 2.6512147087487392e-05 2023-01-21 13:52:25.939953: step: 208/533, loss: 0.0018810273613780737 2023-01-21 13:52:27.098085: step: 212/533, loss: 0.00013179780216887593 2023-01-21 13:52:28.290684: step: 216/533, loss: 0.0004856109735555947 2023-01-21 13:52:29.467902: step: 220/533, loss: 0.06237411871552467 2023-01-21 13:52:30.684092: step: 224/533, loss: 0.011865615844726562 2023-01-21 13:52:31.866295: step: 228/533, loss: 0.005190658383071423 2023-01-21 13:52:33.017885: step: 232/533, loss: 5.664825584972277e-05 2023-01-21 13:52:34.206079: step: 236/533, loss: 1.0967255548166577e-05 2023-01-21 13:52:35.396672: step: 240/533, loss: 1.7642974853515625e-05 2023-01-21 13:52:36.559839: step: 244/533, loss: 6.67572021484375e-06 2023-01-21 13:52:37.761957: step: 248/533, loss: 0.0003154754813294858 2023-01-21 13:52:38.953432: step: 252/533, loss: 0.00016288757615257055 2023-01-21 13:52:40.136720: step: 256/533, loss: 2.021789623540826e-05 2023-01-21 13:52:41.313621: step: 260/533, loss: 5.1116945542162284e-05 2023-01-21 13:52:42.464703: step: 264/533, loss: 7.724762326688506e-06 2023-01-21 13:52:43.612185: step: 268/533, loss: 0.0011365890968590975 2023-01-21 13:52:44.746491: step: 272/533, loss: 0.001000261283479631 2023-01-21 13:52:45.896570: step: 276/533, loss: 0.012322806753218174 2023-01-21 13:52:47.081418: step: 280/533, loss: 0.004737759009003639 2023-01-21 13:52:48.249169: step: 284/533, loss: -5.7220458984375e-06 2023-01-21 13:52:49.457385: step: 288/533, loss: -3.814697265625e-06 2023-01-21 13:52:50.620575: step: 292/533, loss: 2.3269654775504023e-05 2023-01-21 13:52:51.815400: step: 296/533, loss: -4.2343137465650216e-05 2023-01-21 13:52:53.039527: step: 300/533, loss: 0.01856079138815403 2023-01-21 13:52:54.246048: step: 304/533, loss: 0.0003593444998841733 2023-01-21 13:52:55.462888: step: 308/533, loss: 0.009319210425019264 2023-01-21 13:52:56.679746: step: 312/533, loss: 0.0005790710565634072 2023-01-21 13:52:57.876036: step: 316/533, loss: 0.011573279276490211 2023-01-21 13:52:59.068056: step: 320/533, loss: 0.0005815505865029991 2023-01-21 13:53:00.273700: step: 324/533, loss: 4.653930591302924e-05 2023-01-21 13:53:01.483619: step: 328/533, loss: 0.0021497728303074837 2023-01-21 13:53:02.690676: step: 332/533, loss: 3.1805036996956915e-05 2023-01-21 13:53:03.862332: step: 336/533, loss: 0.0003741264226846397 2023-01-21 13:53:05.056358: step: 340/533, loss: -1.201629675051663e-05 2023-01-21 13:53:06.242850: step: 344/533, loss: 0.0018428802723065019 2023-01-21 13:53:07.420647: step: 348/533, loss: 3.261566234868951e-05 2023-01-21 13:53:08.582235: step: 352/533, loss: 0.026845265179872513 2023-01-21 13:53:09.722599: step: 356/533, loss: 0.00016660690016578883 2023-01-21 13:53:10.927093: step: 360/533, loss: 0.010023879818618298 2023-01-21 13:53:12.186276: step: 364/533, loss: 0.00011177062697242945 2023-01-21 13:53:13.370407: step: 368/533, loss: 0.03643150255084038 2023-01-21 13:53:14.536291: step: 372/533, loss: 0.008087730966508389 2023-01-21 13:53:15.707221: step: 376/533, loss: 3.2424925393570447e-06 2023-01-21 13:53:16.885894: step: 380/533, loss: 7.286071922862902e-05 2023-01-21 13:53:18.110378: step: 384/533, loss: 8.678436643094756e-06 2023-01-21 13:53:19.307400: step: 388/533, loss: 0.0009160518529824913 2023-01-21 13:53:20.486167: step: 392/533, loss: 0.0005702972412109375 2023-01-21 13:53:21.665863: step: 396/533, loss: 0.04440345615148544 2023-01-21 13:53:22.845479: step: 400/533, loss: 4.653930591302924e-05 2023-01-21 13:53:24.052329: step: 404/533, loss: 0.0009916304843500257 2023-01-21 13:53:25.279009: step: 408/533, loss: 0.0004911422729492188 2023-01-21 13:53:26.456717: step: 412/533, loss: 0.01899242401123047 2023-01-21 13:53:27.630273: step: 416/533, loss: 0.00021104811457917094 2023-01-21 13:53:28.807576: step: 420/533, loss: 0.00021486282639671117 2023-01-21 13:53:29.980779: step: 424/533, loss: 0.002027368638664484 2023-01-21 13:53:31.127708: step: 428/533, loss: 0.0003986835654359311 2023-01-21 13:53:32.415168: step: 432/533, loss: 0.00330352783203125 2023-01-21 13:53:33.583222: step: 436/533, loss: 0.013103485107421875 2023-01-21 13:53:34.752360: step: 440/533, loss: 8.220673043979332e-05 2023-01-21 13:53:35.945829: step: 444/533, loss: 0.006476211827248335 2023-01-21 13:53:37.123035: step: 448/533, loss: -4.57763690064894e-06 2023-01-21 13:53:38.300849: step: 452/533, loss: -0.0 2023-01-21 13:53:39.449890: step: 456/533, loss: 3.7765505112474784e-05 2023-01-21 13:53:40.634485: step: 460/533, loss: 3.395080420887098e-05 2023-01-21 13:53:41.801821: step: 464/533, loss: 4.310607982915826e-05 2023-01-21 13:53:42.995517: step: 468/533, loss: 0.004195690155029297 2023-01-21 13:53:44.153335: step: 472/533, loss: 0.00789022445678711 2023-01-21 13:53:45.315179: step: 476/533, loss: 5.91278057981981e-06 2023-01-21 13:53:46.486290: step: 480/533, loss: 1.3351440202313825e-06 2023-01-21 13:53:47.664341: step: 484/533, loss: 0.0003985404910054058 2023-01-21 13:53:48.864023: step: 488/533, loss: 0.06752987205982208 2023-01-21 13:53:50.008415: step: 492/533, loss: 0.008328819647431374 2023-01-21 13:53:51.168043: step: 496/533, loss: 0.015819359570741653 2023-01-21 13:53:52.434034: step: 500/533, loss: 0.019759178161621094 2023-01-21 13:53:53.606754: step: 504/533, loss: 0.00322723388671875 2023-01-21 13:53:54.781615: step: 508/533, loss: 0.00013881921768188477 2023-01-21 13:53:55.991083: step: 512/533, loss: 0.00067815778311342 2023-01-21 13:53:57.153813: step: 516/533, loss: -8.96453821042087e-06 2023-01-21 13:53:58.340184: step: 520/533, loss: 7.696151442360133e-05 2023-01-21 13:53:59.504350: step: 524/533, loss: 0.0009431838989257812 2023-01-21 13:54:00.689607: step: 528/533, loss: 7.686614844715223e-05 2023-01-21 13:54:01.836675: step: 532/533, loss: 1.106262243411038e-05 2023-01-21 13:54:02.963880: step: 536/533, loss: 0.004604148678481579 2023-01-21 13:54:04.154689: step: 540/533, loss: 0.3151291012763977 2023-01-21 13:54:05.319685: step: 544/533, loss: -7.62939453125e-06 2023-01-21 13:54:06.536955: step: 548/533, loss: 0.006260585971176624 2023-01-21 13:54:07.721461: step: 552/533, loss: 0.00654838141053915 2023-01-21 13:54:08.894630: step: 556/533, loss: 1.9073495423072018e-06 2023-01-21 13:54:10.104454: step: 560/533, loss: 0.0005549431079998612 2023-01-21 13:54:11.278748: step: 564/533, loss: 0.0011314392322674394 2023-01-21 13:54:12.448571: step: 568/533, loss: 0.00021576881408691406 2023-01-21 13:54:13.634606: step: 572/533, loss: 0.005048561375588179 2023-01-21 13:54:14.839806: step: 576/533, loss: 4.95910626341356e-06 2023-01-21 13:54:16.002908: step: 580/533, loss: 0.00028133392333984375 2023-01-21 13:54:17.190497: step: 584/533, loss: 6.0844424297101796e-05 2023-01-21 13:54:18.382541: step: 588/533, loss: 0.0031113624572753906 2023-01-21 13:54:19.534834: step: 592/533, loss: 0.011923504061996937 2023-01-21 13:54:20.702258: step: 596/533, loss: 0.013973427005112171 2023-01-21 13:54:21.898786: step: 600/533, loss: 0.048525430262088776 2023-01-21 13:54:23.113736: step: 604/533, loss: 0.0027065277099609375 2023-01-21 13:54:24.275066: step: 608/533, loss: 8.75473051564768e-05 2023-01-21 13:54:25.453451: step: 612/533, loss: 0.0033147812355309725 2023-01-21 13:54:26.629835: step: 616/533, loss: 0.03016185760498047 2023-01-21 13:54:27.836604: step: 620/533, loss: 0.00013847352238371968 2023-01-21 13:54:29.013086: step: 624/533, loss: 0.0020377635955810547 2023-01-21 13:54:30.225160: step: 628/533, loss: 0.00010318756540073082 2023-01-21 13:54:31.353812: step: 632/533, loss: 0.00015163421630859375 2023-01-21 13:54:32.534664: step: 636/533, loss: 0.05630455166101456 2023-01-21 13:54:33.725908: step: 640/533, loss: 0.007293033413589001 2023-01-21 13:54:34.876180: step: 644/533, loss: -6.86645489622606e-06 2023-01-21 13:54:36.029983: step: 648/533, loss: 3.3664702641544864e-05 2023-01-21 13:54:37.181839: step: 652/533, loss: 6.771088010282256e-06 2023-01-21 13:54:38.359822: step: 656/533, loss: 0.003407812211662531 2023-01-21 13:54:39.516073: step: 660/533, loss: -1.382828031637473e-06 2023-01-21 13:54:40.667175: step: 664/533, loss: 0.021144604310393333 2023-01-21 13:54:41.859234: step: 668/533, loss: 6.580352419405244e-06 2023-01-21 13:54:43.058095: step: 672/533, loss: 3.013610876223538e-05 2023-01-21 13:54:44.249493: step: 676/533, loss: 0.0015642166836187243 2023-01-21 13:54:45.403192: step: 680/533, loss: 3.967285010730848e-05 2023-01-21 13:54:46.576339: step: 684/533, loss: -1.907349087559851e-06 2023-01-21 13:54:47.735628: step: 688/533, loss: 0.01734914816915989 2023-01-21 13:54:48.911428: step: 692/533, loss: 0.00022182465181685984 2023-01-21 13:54:50.111397: step: 696/533, loss: 0.00023326874361373484 2023-01-21 13:54:51.293939: step: 700/533, loss: 0.00026979445829056203 2023-01-21 13:54:52.432336: step: 704/533, loss: 2.012253025895916e-05 2023-01-21 13:54:53.573267: step: 708/533, loss: 1.1110305422334932e-05 2023-01-21 13:54:54.754138: step: 712/533, loss: 0.04693937301635742 2023-01-21 13:54:55.870445: step: 716/533, loss: 2.4795533590804553e-06 2023-01-21 13:54:57.048031: step: 720/533, loss: 0.0017507553566247225 2023-01-21 13:54:58.206547: step: 724/533, loss: -1.4591217222914565e-05 2023-01-21 13:54:59.435583: step: 728/533, loss: 2.346038854739163e-05 2023-01-21 13:55:00.598989: step: 732/533, loss: 0.000988006591796875 2023-01-21 13:55:01.794062: step: 736/533, loss: 0.08970393985509872 2023-01-21 13:55:02.992094: step: 740/533, loss: 0.025391673669219017 2023-01-21 13:55:04.168418: step: 744/533, loss: 0.0323491096496582 2023-01-21 13:55:05.439134: step: 748/533, loss: 0.017469406127929688 2023-01-21 13:55:06.629869: step: 752/533, loss: 4.130596160888672 2023-01-21 13:55:07.794036: step: 756/533, loss: 0.030504608526825905 2023-01-21 13:55:08.946371: step: 760/533, loss: 0.0010730744106695056 2023-01-21 13:55:10.108453: step: 764/533, loss: 9.346007573185489e-06 2023-01-21 13:55:11.350611: step: 768/533, loss: 1.316070574830519e-05 2023-01-21 13:55:12.501775: step: 772/533, loss: 7.64846772653982e-05 2023-01-21 13:55:13.638970: step: 776/533, loss: 0.000743865966796875 2023-01-21 13:55:14.833530: step: 780/533, loss: 0.0004986286512576044 2023-01-21 13:55:16.037768: step: 784/533, loss: 0.003444671630859375 2023-01-21 13:55:17.288719: step: 788/533, loss: 0.05000896379351616 2023-01-21 13:55:18.467655: step: 792/533, loss: 0.000462436699308455 2023-01-21 13:55:19.644928: step: 796/533, loss: 3.743171691894531e-05 2023-01-21 13:55:20.845406: step: 800/533, loss: 0.01708526536822319 2023-01-21 13:55:22.052175: step: 804/533, loss: 0.0003654480096884072 2023-01-21 13:55:23.221626: step: 808/533, loss: 0.0001984596165129915 2023-01-21 13:55:24.406525: step: 812/533, loss: 0.0005933761713095009 2023-01-21 13:55:25.581329: step: 816/533, loss: 9.193421283271164e-05 2023-01-21 13:55:26.757020: step: 820/533, loss: 0.0007369995582848787 2023-01-21 13:55:27.925774: step: 824/533, loss: 0.0017513275379315019 2023-01-21 13:55:29.080095: step: 828/533, loss: 4.76837158203125e-06 2023-01-21 13:55:30.223454: step: 832/533, loss: 2.7465821403893642e-05 2023-01-21 13:55:31.455379: step: 836/533, loss: 0.0003027915954589844 2023-01-21 13:55:32.623274: step: 840/533, loss: 9.202957153320312e-05 2023-01-21 13:55:33.790082: step: 844/533, loss: 0.0002960205019917339 2023-01-21 13:55:34.961844: step: 848/533, loss: 4.959106809110381e-05 2023-01-21 13:55:36.181198: step: 852/533, loss: 0.0026914598420262337 2023-01-21 13:55:37.342510: step: 856/533, loss: 7.328986976062879e-05 2023-01-21 13:55:38.505317: step: 860/533, loss: 0.002540302462875843 2023-01-21 13:55:39.676693: step: 864/533, loss: 5.097389657748863e-05 2023-01-21 13:55:40.820587: step: 868/533, loss: 0.00167675013653934 2023-01-21 13:55:42.043324: step: 872/533, loss: 0.0019192696781829 2023-01-21 13:55:43.193195: step: 876/533, loss: 0.00300426478497684 2023-01-21 13:55:44.415687: step: 880/533, loss: 1.1348724001436494e-05 2023-01-21 13:55:45.596427: step: 884/533, loss: 0.00228805560618639 2023-01-21 13:55:46.743957: step: 888/533, loss: 8.583068620282575e-07 2023-01-21 13:55:47.929353: step: 892/533, loss: 0.0011803627712652087 2023-01-21 13:55:49.141079: step: 896/533, loss: 0.0001911163271870464 2023-01-21 13:55:50.310862: step: 900/533, loss: 0.005280113313347101 2023-01-21 13:55:51.478308: step: 904/533, loss: 2.403259350103326e-05 2023-01-21 13:55:52.650766: step: 908/533, loss: 0.02379913441836834 2023-01-21 13:55:53.843266: step: 912/533, loss: 1.201629675051663e-05 2023-01-21 13:55:55.023151: step: 916/533, loss: 0.00927658099681139 2023-01-21 13:55:56.233738: step: 920/533, loss: 8.96453821042087e-06 2023-01-21 13:55:57.368158: step: 924/533, loss: 1.144409225162235e-06 2023-01-21 13:55:58.555325: step: 928/533, loss: -2.7179717108083423e-06 2023-01-21 13:55:59.745441: step: 932/533, loss: 4.711151268566027e-05 2023-01-21 13:56:00.929060: step: 936/533, loss: 0.00026340485783293843 2023-01-21 13:56:02.112703: step: 940/533, loss: 0.0008029937744140625 2023-01-21 13:56:03.328202: step: 944/533, loss: 0.006027603521943092 2023-01-21 13:56:04.504185: step: 948/533, loss: 0.009347820654511452 2023-01-21 13:56:05.718220: step: 952/533, loss: 9.641647193348035e-05 2023-01-21 13:56:06.926647: step: 956/533, loss: 1.602172778802924e-05 2023-01-21 13:56:08.111408: step: 960/533, loss: 0.00015525818162132055 2023-01-21 13:56:09.290836: step: 964/533, loss: -3.6865474157821154e-06 2023-01-21 13:56:10.466983: step: 968/533, loss: 0.0006953001138754189 2023-01-21 13:56:11.690969: step: 972/533, loss: 4.234314110362902e-05 2023-01-21 13:56:12.847810: step: 976/533, loss: 0.03715667873620987 2023-01-21 13:56:14.032321: step: 980/533, loss: 0.09836140275001526 2023-01-21 13:56:15.218341: step: 984/533, loss: 0.04682312160730362 2023-01-21 13:56:16.381447: step: 988/533, loss: 0.0005405425908975303 2023-01-21 13:56:17.591927: step: 992/533, loss: 0.0016632080078125 2023-01-21 13:56:18.743651: step: 996/533, loss: 0.0006349563482217491 2023-01-21 13:56:19.936885: step: 1000/533, loss: 0.00038509367732331157 2023-01-21 13:56:21.110036: step: 1004/533, loss: 0.055867768824100494 2023-01-21 13:56:22.300469: step: 1008/533, loss: 1.2111663636460435e-05 2023-01-21 13:56:23.481434: step: 1012/533, loss: 0.0002483367861714214 2023-01-21 13:56:24.641254: step: 1016/533, loss: 0.0001218795805471018 2023-01-21 13:56:25.802247: step: 1020/533, loss: 0.00724868755787611 2023-01-21 13:56:27.020355: step: 1024/533, loss: 0.00033988954965025187 2023-01-21 13:56:28.189545: step: 1028/533, loss: 0.0335543118417263 2023-01-21 13:56:29.392136: step: 1032/533, loss: 3.63349899998866e-05 2023-01-21 13:56:30.605732: step: 1036/533, loss: 0.006031036376953125 2023-01-21 13:56:31.790870: step: 1040/533, loss: 8.58306884765625e-06 2023-01-21 13:56:32.960890: step: 1044/533, loss: 1.8119808373739943e-06 2023-01-21 13:56:34.156703: step: 1048/533, loss: 0.019656943157315254 2023-01-21 13:56:35.337013: step: 1052/533, loss: -1.6689295989635866e-06 2023-01-21 13:56:36.531051: step: 1056/533, loss: 2.250671423098538e-05 2023-01-21 13:56:37.714760: step: 1060/533, loss: 0.9803696870803833 2023-01-21 13:56:38.892195: step: 1064/533, loss: 0.0001221656712004915 2023-01-21 13:56:40.080753: step: 1068/533, loss: 6.618499901378527e-05 2023-01-21 13:56:41.203763: step: 1072/533, loss: 4.9591067181609105e-06 2023-01-21 13:56:42.365866: step: 1076/533, loss: 0.2125784009695053 2023-01-21 13:56:43.605090: step: 1080/533, loss: 0.00014238357834983617 2023-01-21 13:56:44.786094: step: 1084/533, loss: 3.9672853745287284e-05 2023-01-21 13:56:45.988082: step: 1088/533, loss: 0.0005665779463015497 2023-01-21 13:56:47.149030: step: 1092/533, loss: 0.009566307067871094 2023-01-21 13:56:48.343379: step: 1096/533, loss: 0.00021522045426536351 2023-01-21 13:56:49.507475: step: 1100/533, loss: 0.00147333147469908 2023-01-21 13:56:50.704152: step: 1104/533, loss: 0.014426995068788528 2023-01-21 13:56:51.910570: step: 1108/533, loss: 0.0001589298335602507 2023-01-21 13:56:53.085337: step: 1112/533, loss: 4.57763690064894e-06 2023-01-21 13:56:54.284793: step: 1116/533, loss: 0.00017995835514739156 2023-01-21 13:56:55.432779: step: 1120/533, loss: 0.0002035141078522429 2023-01-21 13:56:56.639352: step: 1124/533, loss: 0.7335087060928345 2023-01-21 13:56:57.835803: step: 1128/533, loss: 0.004726409446448088 2023-01-21 13:56:59.012805: step: 1132/533, loss: 1.9359587895451114e-05 2023-01-21 13:57:00.198806: step: 1136/533, loss: 0.06349802017211914 2023-01-21 13:57:01.357090: step: 1140/533, loss: 0.016556929796934128 2023-01-21 13:57:02.511153: step: 1144/533, loss: 1.0776519957289565e-05 2023-01-21 13:57:03.662510: step: 1148/533, loss: 1.6498564946232364e-05 2023-01-21 13:57:04.817383: step: 1152/533, loss: 0.0007374763954430819 2023-01-21 13:57:05.982785: step: 1156/533, loss: 8.77380352903856e-06 2023-01-21 13:57:07.166264: step: 1160/533, loss: 0.018012046813964844 2023-01-21 13:57:08.350547: step: 1164/533, loss: 0.00022640227689407766 2023-01-21 13:57:09.503930: step: 1168/533, loss: 0.0001028060942189768 2023-01-21 13:57:10.687749: step: 1172/533, loss: 0.0008440017700195312 2023-01-21 13:57:11.847052: step: 1176/533, loss: 0.0001661300630075857 2023-01-21 13:57:13.007378: step: 1180/533, loss: 0.00020494461932685226 2023-01-21 13:57:14.190194: step: 1184/533, loss: 0.00011949539475608617 2023-01-21 13:57:15.360902: step: 1188/533, loss: 0.00013704299635719508 2023-01-21 13:57:16.531457: step: 1192/533, loss: 0.04097404703497887 2023-01-21 13:57:17.700549: step: 1196/533, loss: 0.0008673667325638235 2023-01-21 13:57:18.884618: step: 1200/533, loss: 0.0002759933704510331 2023-01-21 13:57:20.117035: step: 1204/533, loss: 4.653930591302924e-05 2023-01-21 13:57:21.274403: step: 1208/533, loss: 0.002628421876579523 2023-01-21 13:57:22.465405: step: 1212/533, loss: 0.00014371873112395406 2023-01-21 13:57:23.672582: step: 1216/533, loss: 0.028267668560147285 2023-01-21 13:57:24.861679: step: 1220/533, loss: 0.007448578253388405 2023-01-21 13:57:26.012882: step: 1224/533, loss: 0.010971451178193092 2023-01-21 13:57:27.157535: step: 1228/533, loss: 0.00033292770967818797 2023-01-21 13:57:28.366185: step: 1232/533, loss: 0.001904296805150807 2023-01-21 13:57:29.553901: step: 1236/533, loss: 8.668899681651965e-05 2023-01-21 13:57:30.754579: step: 1240/533, loss: 0.009830474853515625 2023-01-21 13:57:31.961334: step: 1244/533, loss: 0.00023908616276457906 2023-01-21 13:57:33.152057: step: 1248/533, loss: 7.43865984986769e-06 2023-01-21 13:57:34.340228: step: 1252/533, loss: 1.3351441339182202e-06 2023-01-21 13:57:35.512699: step: 1256/533, loss: 0.0018934250110760331 2023-01-21 13:57:36.692832: step: 1260/533, loss: 0.01131963636726141 2023-01-21 13:57:37.879374: step: 1264/533, loss: 3.99589553126134e-05 2023-01-21 13:57:39.050370: step: 1268/533, loss: 0.0003822326543740928 2023-01-21 13:57:40.244604: step: 1272/533, loss: 0.02717895433306694 2023-01-21 13:57:41.417906: step: 1276/533, loss: 0.031896304339170456 2023-01-21 13:57:42.601957: step: 1280/533, loss: 0.08291064202785492 2023-01-21 13:57:43.785695: step: 1284/533, loss: 0.01981964148581028 2023-01-21 13:57:44.954608: step: 1288/533, loss: 0.005047607235610485 2023-01-21 13:57:46.075583: step: 1292/533, loss: 0.0006946563953533769 2023-01-21 13:57:47.249900: step: 1296/533, loss: 0.009035111404955387 2023-01-21 13:57:48.450600: step: 1300/533, loss: 0.0017980575794354081 2023-01-21 13:57:49.604130: step: 1304/533, loss: 0.008378982543945312 2023-01-21 13:57:50.794919: step: 1308/533, loss: 0.014710140414536 2023-01-21 13:57:51.979564: step: 1312/533, loss: 8.487701052217744e-06 2023-01-21 13:57:53.167006: step: 1316/533, loss: 1.8787384760798886e-05 2023-01-21 13:57:54.312776: step: 1320/533, loss: 0.002651214599609375 2023-01-21 13:57:55.512937: step: 1324/533, loss: 8.850097947288305e-05 2023-01-21 13:57:56.678134: step: 1328/533, loss: 0.005335903260856867 2023-01-21 13:57:57.855863: step: 1332/533, loss: 0.0066134450025856495 2023-01-21 13:57:59.060044: step: 1336/533, loss: 2.7227401005802676e-05 2023-01-21 13:58:00.216389: step: 1340/533, loss: 0.08215832710266113 2023-01-21 13:58:01.382333: step: 1344/533, loss: 6.29425039733178e-06 2023-01-21 13:58:02.553399: step: 1348/533, loss: 0.3367481231689453 2023-01-21 13:58:03.749270: step: 1352/533, loss: 0.05454311519861221 2023-01-21 13:58:04.951195: step: 1356/533, loss: 0.17127694189548492 2023-01-21 13:58:06.096832: step: 1360/533, loss: 0.011753655038774014 2023-01-21 13:58:07.262216: step: 1364/533, loss: 1.5735626220703125e-05 2023-01-21 13:58:08.405996: step: 1368/533, loss: 0.001644134521484375 2023-01-21 13:58:09.597826: step: 1372/533, loss: 0.02727527730166912 2023-01-21 13:58:10.765287: step: 1376/533, loss: 0.0006443978054448962 2023-01-21 13:58:11.912617: step: 1380/533, loss: 0.002175712725147605 2023-01-21 13:58:13.085995: step: 1384/533, loss: 0.0008421898237429559 2023-01-21 13:58:14.259786: step: 1388/533, loss: 0.03786773607134819 2023-01-21 13:58:15.425584: step: 1392/533, loss: 0.004493999294936657 2023-01-21 13:58:16.634325: step: 1396/533, loss: 0.0005971908685751259 2023-01-21 13:58:17.809810: step: 1400/533, loss: -3.871917579090223e-05 2023-01-21 13:58:18.997197: step: 1404/533, loss: 0.004050064366310835 2023-01-21 13:58:20.212318: step: 1408/533, loss: 5.416870044427924e-05 2023-01-21 13:58:21.492374: step: 1412/533, loss: 7.419585745083168e-05 2023-01-21 13:58:22.700221: step: 1416/533, loss: 0.00016250611224677414 2023-01-21 13:58:23.894426: step: 1420/533, loss: 0.0002401351957814768 2023-01-21 13:58:25.095206: step: 1424/533, loss: 0.0506587028503418 2023-01-21 13:58:26.266930: step: 1428/533, loss: 0.0009137153392657638 2023-01-21 13:58:27.447908: step: 1432/533, loss: 0.00253467564471066 2023-01-21 13:58:28.591213: step: 1436/533, loss: 0.00018959045701194555 2023-01-21 13:58:29.736536: step: 1440/533, loss: 0.2060968279838562 2023-01-21 13:58:30.961983: step: 1444/533, loss: 0.0008440017700195312 2023-01-21 13:58:32.139177: step: 1448/533, loss: 0.0007823944324627519 2023-01-21 13:58:33.281114: step: 1452/533, loss: 0.004110145848244429 2023-01-21 13:58:34.456639: step: 1456/533, loss: 0.000278472900390625 2023-01-21 13:58:35.619730: step: 1460/533, loss: 3.5929682780988514e-05 2023-01-21 13:58:36.788310: step: 1464/533, loss: 0.014162182807922363 2023-01-21 13:58:37.942794: step: 1468/533, loss: 0.0003181457577738911 2023-01-21 13:58:39.096376: step: 1472/533, loss: 4.95910626341356e-06 2023-01-21 13:58:40.323454: step: 1476/533, loss: 0.0009329795720987022 2023-01-21 13:58:41.498488: step: 1480/533, loss: 6.008148375258315e-06 2023-01-21 13:58:42.688796: step: 1484/533, loss: 0.10419555008411407 2023-01-21 13:58:43.872642: step: 1488/533, loss: 1.7547608877066523e-05 2023-01-21 13:58:45.039132: step: 1492/533, loss: 0.0022815705742686987 2023-01-21 13:58:46.212239: step: 1496/533, loss: -1.4209746950655244e-05 2023-01-21 13:58:47.451994: step: 1500/533, loss: 3.3664702641544864e-05 2023-01-21 13:58:48.602333: step: 1504/533, loss: 0.00010995864431606606 2023-01-21 13:58:49.750181: step: 1508/533, loss: 0.004834878724068403 2023-01-21 13:58:50.942378: step: 1512/533, loss: 0.00010757446580100805 2023-01-21 13:58:52.111499: step: 1516/533, loss: 0.0011323929065838456 2023-01-21 13:58:53.296732: step: 1520/533, loss: 0.001860809396021068 2023-01-21 13:58:54.549917: step: 1524/533, loss: 0.011013126000761986 2023-01-21 13:58:55.695119: step: 1528/533, loss: 0.00016269684419967234 2023-01-21 13:58:56.863365: step: 1532/533, loss: 0.0019326210021972656 2023-01-21 13:58:58.078142: step: 1536/533, loss: 0.11945199966430664 2023-01-21 13:58:59.261153: step: 1540/533, loss: 0.0001066207914846018 2023-01-21 13:59:00.483377: step: 1544/533, loss: 0.00934305228292942 2023-01-21 13:59:01.647585: step: 1548/533, loss: 0.0007795334095135331 2023-01-21 13:59:02.817874: step: 1552/533, loss: 0.04730243980884552 2023-01-21 13:59:03.982726: step: 1556/533, loss: 9.5367431640625e-07 2023-01-21 13:59:05.178844: step: 1560/533, loss: 0.00014905929856467992 2023-01-21 13:59:06.340472: step: 1564/533, loss: 0.02530956268310547 2023-01-21 13:59:07.592866: step: 1568/533, loss: 0.00016403199697379023 2023-01-21 13:59:08.778258: step: 1572/533, loss: 0.00035371779813431203 2023-01-21 13:59:09.977787: step: 1576/533, loss: 0.0015872002113610506 2023-01-21 13:59:11.138930: step: 1580/533, loss: 3.4046173823298886e-05 2023-01-21 13:59:12.303587: step: 1584/533, loss: 0.004333305638283491 2023-01-21 13:59:13.470462: step: 1588/533, loss: 0.0031821727752685547 2023-01-21 13:59:14.676574: step: 1592/533, loss: 0.005445766728371382 2023-01-21 13:59:15.825409: step: 1596/533, loss: 2.117157055181451e-05 2023-01-21 13:59:17.012200: step: 1600/533, loss: -7.343292054429185e-06 2023-01-21 13:59:18.179705: step: 1604/533, loss: 0.003073024796321988 2023-01-21 13:59:19.308346: step: 1608/533, loss: 0.014204598031938076 2023-01-21 13:59:20.461943: step: 1612/533, loss: 0.0127601632848382 2023-01-21 13:59:21.586811: step: 1616/533, loss: 0.0013735414249822497 2023-01-21 13:59:22.757845: step: 1620/533, loss: 0.003961372654885054 2023-01-21 13:59:23.943186: step: 1624/533, loss: 0.004844475071877241 2023-01-21 13:59:25.120076: step: 1628/533, loss: 0.022129535675048828 2023-01-21 13:59:26.292296: step: 1632/533, loss: 0.08166561275720596 2023-01-21 13:59:27.469507: step: 1636/533, loss: 0.00010881424532271922 2023-01-21 13:59:28.658950: step: 1640/533, loss: 2.574920654296875e-05 2023-01-21 13:59:29.882743: step: 1644/533, loss: 0.004196357913315296 2023-01-21 13:59:31.074281: step: 1648/533, loss: 4.9924849008675665e-05 2023-01-21 13:59:32.256137: step: 1652/533, loss: 0.025603104382753372 2023-01-21 13:59:33.492536: step: 1656/533, loss: 0.00566940288990736 2023-01-21 13:59:34.676194: step: 1660/533, loss: 0.004317379090934992 2023-01-21 13:59:35.865334: step: 1664/533, loss: 0.002075386233627796 2023-01-21 13:59:37.055648: step: 1668/533, loss: 0.00019884110952261835 2023-01-21 13:59:38.260192: step: 1672/533, loss: 0.11667118221521378 2023-01-21 13:59:39.420450: step: 1676/533, loss: 0.0006956100114621222 2023-01-21 13:59:40.596668: step: 1680/533, loss: 0.0003963470517192036 2023-01-21 13:59:41.754180: step: 1684/533, loss: 1.2493133908719756e-05 2023-01-21 13:59:42.924141: step: 1688/533, loss: 0.2241135537624359 2023-01-21 13:59:44.108688: step: 1692/533, loss: 0.0001409530668752268 2023-01-21 13:59:45.272124: step: 1696/533, loss: 2.3651124138268642e-05 2023-01-21 13:59:46.434727: step: 1700/533, loss: 0.0007869720575399697 2023-01-21 13:59:47.654656: step: 1704/533, loss: 0.0007658004760742188 2023-01-21 13:59:48.825946: step: 1708/533, loss: 0.11649684607982635 2023-01-21 13:59:49.977761: step: 1712/533, loss: 0.0012390136253088713 2023-01-21 13:59:51.122141: step: 1716/533, loss: 3.070831371587701e-05 2023-01-21 13:59:52.341686: step: 1720/533, loss: 0.008800124749541283 2023-01-21 13:59:53.486763: step: 1724/533, loss: 0.006130409426987171 2023-01-21 13:59:54.709886: step: 1728/533, loss: 0.00027709006099030375 2023-01-21 13:59:55.878618: step: 1732/533, loss: 0.012064743787050247 2023-01-21 13:59:57.068071: step: 1736/533, loss: 0.00026836394681595266 2023-01-21 13:59:58.297767: step: 1740/533, loss: 0.012546730227768421 2023-01-21 13:59:59.440747: step: 1744/533, loss: 0.00032796862069517374 2023-01-21 14:00:00.595231: step: 1748/533, loss: 0.8723475933074951 2023-01-21 14:00:01.754078: step: 1752/533, loss: 0.0007363319164142013 2023-01-21 14:00:02.934892: step: 1756/533, loss: 2.0980833141948096e-06 2023-01-21 14:00:04.086033: step: 1760/533, loss: 0.0038959502708166838 2023-01-21 14:00:05.216640: step: 1764/533, loss: -2.28881845032447e-06 2023-01-21 14:00:06.380294: step: 1768/533, loss: -2.86102294921875e-06 2023-01-21 14:00:07.574804: step: 1772/533, loss: 0.0002876758517231792 2023-01-21 14:00:08.760093: step: 1776/533, loss: 3.318786548334174e-05 2023-01-21 14:00:09.961464: step: 1780/533, loss: 8.583069188716763e-07 2023-01-21 14:00:11.157254: step: 1784/533, loss: 0.0015190124977380037 2023-01-21 14:00:12.315075: step: 1788/533, loss: 2.1743777324445546e-05 2023-01-21 14:00:13.451712: step: 1792/533, loss: 3.2901764370762976e-06 2023-01-21 14:00:14.617030: step: 1796/533, loss: -9.34600848268019e-06 2023-01-21 14:00:15.755074: step: 1800/533, loss: 0.00046949388342909515 2023-01-21 14:00:16.947987: step: 1804/533, loss: 0.00037751198397018015 2023-01-21 14:00:18.146842: step: 1808/533, loss: 0.05875739827752113 2023-01-21 14:00:19.331404: step: 1812/533, loss: 0.000251579302130267 2023-01-21 14:00:20.507454: step: 1816/533, loss: 9.794235666049644e-05 2023-01-21 14:00:21.667686: step: 1820/533, loss: 0.0001445770321879536 2023-01-21 14:00:22.807191: step: 1824/533, loss: 0.0004661559942178428 2023-01-21 14:00:23.980446: step: 1828/533, loss: 4.081726001459174e-05 2023-01-21 14:00:25.195630: step: 1832/533, loss: 0.004021263215690851 2023-01-21 14:00:26.376067: step: 1836/533, loss: 0.18091773986816406 2023-01-21 14:00:27.537070: step: 1840/533, loss: 0.008635115809738636 2023-01-21 14:00:28.707075: step: 1844/533, loss: 0.0007866382948122919 2023-01-21 14:00:29.876893: step: 1848/533, loss: 0.000278282182989642 2023-01-21 14:00:31.015782: step: 1852/533, loss: 5.7029727031476796e-05 2023-01-21 14:00:32.198185: step: 1856/533, loss: 0.00012121201143600047 2023-01-21 14:00:33.392055: step: 1860/533, loss: 0.012169933877885342 2023-01-21 14:00:34.582440: step: 1864/533, loss: 7.629394644936838e-07 2023-01-21 14:00:35.775898: step: 1868/533, loss: 2.1457672119140625e-06 2023-01-21 14:00:36.932003: step: 1872/533, loss: 0.0015893937088549137 2023-01-21 14:00:38.082523: step: 1876/533, loss: 2.384185791015625e-05 2023-01-21 14:00:39.267650: step: 1880/533, loss: 0.05921308323740959 2023-01-21 14:00:40.442779: step: 1884/533, loss: 0.0787748321890831 2023-01-21 14:00:41.635989: step: 1888/533, loss: 0.017388343811035156 2023-01-21 14:00:42.781038: step: 1892/533, loss: 0.0003908157523255795 2023-01-21 14:00:43.995739: step: 1896/533, loss: 1.0013580322265625e-05 2023-01-21 14:00:45.197301: step: 1900/533, loss: 0.0004707336483988911 2023-01-21 14:00:46.397015: step: 1904/533, loss: -8.583068620282575e-07 2023-01-21 14:00:47.552436: step: 1908/533, loss: 0.0005311965942382812 2023-01-21 14:00:48.763666: step: 1912/533, loss: 8.859634544933215e-05 2023-01-21 14:00:49.938838: step: 1916/533, loss: 0.017566299065947533 2023-01-21 14:00:51.148695: step: 1920/533, loss: 5.7029727031476796e-05 2023-01-21 14:00:52.352462: step: 1924/533, loss: 0.00013599396334029734 2023-01-21 14:00:53.528511: step: 1928/533, loss: 8.745193190407008e-05 2023-01-21 14:00:54.694480: step: 1932/533, loss: 0.0009029388311319053 2023-01-21 14:00:55.888101: step: 1936/533, loss: 0.00646400498226285 2023-01-21 14:00:57.053204: step: 1940/533, loss: 7.324219041038305e-05 2023-01-21 14:00:58.200149: step: 1944/533, loss: 0.0006534576532430947 2023-01-21 14:00:59.390956: step: 1948/533, loss: -3.814697265625e-06 2023-01-21 14:01:00.555591: step: 1952/533, loss: 5.7220458984375e-06 2023-01-21 14:01:01.706658: step: 1956/533, loss: 0.005449104588478804 2023-01-21 14:01:02.834763: step: 1960/533, loss: 0.00021533967810682952 2023-01-21 14:01:04.017540: step: 1964/533, loss: 4.510879807639867e-05 2023-01-21 14:01:05.203947: step: 1968/533, loss: 0.0032398223411291838 2023-01-21 14:01:06.358240: step: 1972/533, loss: 0.003849601838737726 2023-01-21 14:01:07.519287: step: 1976/533, loss: 0.028018666431307793 2023-01-21 14:01:08.715694: step: 1980/533, loss: 0.0008300781482830644 2023-01-21 14:01:09.920292: step: 1984/533, loss: 0.0006818771362304688 2023-01-21 14:01:11.085080: step: 1988/533, loss: 2.3365020751953125e-05 2023-01-21 14:01:12.269074: step: 1992/533, loss: 0.013470840640366077 2023-01-21 14:01:13.427871: step: 1996/533, loss: 0.004361343570053577 2023-01-21 14:01:14.619950: step: 2000/533, loss: 2.3365021206700476e-06 2023-01-21 14:01:15.823687: step: 2004/533, loss: 0.006718588061630726 2023-01-21 14:01:17.009554: step: 2008/533, loss: 0.0021800994873046875 2023-01-21 14:01:18.194790: step: 2012/533, loss: 0.0013469696277752519 2023-01-21 14:01:19.364324: step: 2016/533, loss: 2.708435022213962e-05 2023-01-21 14:01:20.588515: step: 2020/533, loss: 0.0006650924915447831 2023-01-21 14:01:21.802314: step: 2024/533, loss: 0.009285212494432926 2023-01-21 14:01:22.990920: step: 2028/533, loss: 2.7358531951904297e-05 2023-01-21 14:01:24.158079: step: 2032/533, loss: 0.006835746578872204 2023-01-21 14:01:25.334226: step: 2036/533, loss: 2.7751922971219756e-05 2023-01-21 14:01:26.535345: step: 2040/533, loss: 0.0006422043079510331 2023-01-21 14:01:27.699890: step: 2044/533, loss: 0.0018190385308116674 2023-01-21 14:01:28.933193: step: 2048/533, loss: 0.0017075538635253906 2023-01-21 14:01:30.110192: step: 2052/533, loss: 0.008793830871582031 2023-01-21 14:01:31.293614: step: 2056/533, loss: 0.0747842788696289 2023-01-21 14:01:32.477072: step: 2060/533, loss: 0.0003153800789732486 2023-01-21 14:01:33.668816: step: 2064/533, loss: 0.000965118408203125 2023-01-21 14:01:34.816041: step: 2068/533, loss: 0.0011789321433752775 2023-01-21 14:01:36.029968: step: 2072/533, loss: 0.0024553299881517887 2023-01-21 14:01:37.243117: step: 2076/533, loss: 0.002536964602768421 2023-01-21 14:01:38.444161: step: 2080/533, loss: 0.010658025741577148 2023-01-21 14:01:39.619029: step: 2084/533, loss: 0.0002533912775106728 2023-01-21 14:01:40.799614: step: 2088/533, loss: 3.0326846172101796e-05 2023-01-21 14:01:41.988143: step: 2092/533, loss: 0.022591495886445045 2023-01-21 14:01:43.142638: step: 2096/533, loss: 0.004225540440529585 2023-01-21 14:01:44.322898: step: 2100/533, loss: 3.538131568348035e-05 2023-01-21 14:01:45.516003: step: 2104/533, loss: 3.328323509776965e-05 2023-01-21 14:01:46.693991: step: 2108/533, loss: 0.0013039589393883944 2023-01-21 14:01:47.898814: step: 2112/533, loss: 0.07391653209924698 2023-01-21 14:01:49.062856: step: 2116/533, loss: 5.865097409696318e-05 2023-01-21 14:01:50.243234: step: 2120/533, loss: 0.00950164720416069 2023-01-21 14:01:51.395051: step: 2124/533, loss: 0.04941196367144585 2023-01-21 14:01:52.576323: step: 2128/533, loss: 0.010000801645219326 2023-01-21 14:01:53.780979: step: 2132/533, loss: 6.399155245162547e-05 ================================================== Loss: 0.023 -------------------- Dev: {'event': {'p': 0.5993914807302231, 'r': 0.7869507323568575, 'f1': 0.6804835924006909}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Test: {'event': {'p': 0.6507447354904982, 'r': 0.8101023017902813, 'f1': 0.7217317003702648}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Chinese: {'event': {'p': 0.5402298850574713, 'r': 0.8703703703703703, 'f1': 0.6666666666666666}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Korean: {'event': {'p': 0.6, 'r': 0.5238095238095238, 'f1': 0.559322033898305}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Russian: {'event': {'p': 0.4444444444444444, 'r': 0.5555555555555556, 'f1': 0.49382716049382713}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6183206106870229, 'r': 0.7549933422103862, 'f1': 0.6798561151079138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Chinese: {'event': {'p': 0.6558164354322306, 'r': 0.7858056265984654, 'f1': 0.7149505526468877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} -------------------- Eng Dev for Korean: {'event': {'p': 0.6301969365426696, 'r': 0.7669773635153129, 'f1': 0.6918918918918918}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Eng Test for Korean: {'event': {'p': 0.6444444444444445, 'r': 0.7787723785166241, 'f1': 0.7052692530399536}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Sample Korean: {'event': {'p': 0.7058823529411765, 'r': 0.5714285714285714, 'f1': 0.6315789473684211}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} -------------------- Eng Dev for Russian: {'event': {'p': 0.5875984251968503, 'r': 0.7949400798934754, 'f1': 0.6757215619694397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.643248629795715, 'r': 0.8254475703324808, 'f1': 0.7230467656118734}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}