Command that produces this log: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4
----------------------------------------------------------------------------------------------------
> trainable params:
>>> xlmr.embeddings.word_embeddings.weight: torch.Size([250002, 1024])
>>> xlmr.embeddings.position_embeddings.weight: torch.Size([514, 1024])
>>> xlmr.embeddings.token_type_embeddings.weight: torch.Size([1, 1024])
>>> xlmr.embeddings.LayerNorm.weight: torch.Size([1024])
>>> xlmr.embeddings.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.0.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.0.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.0.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.0.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.0.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.0.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.0.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.0.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.0.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.0.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.1.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.1.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.1.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.1.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.1.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.1.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.1.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.1.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.1.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.1.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.2.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.2.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.2.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.2.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.2.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.2.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.2.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.2.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.2.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.2.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.3.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.3.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.3.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.3.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.3.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.3.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.3.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.3.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.3.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.3.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.4.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.4.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.4.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.4.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.4.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.4.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.4.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.4.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.4.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.4.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.5.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.5.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.5.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.5.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.5.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.5.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.5.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.5.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.5.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.5.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.6.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.6.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.6.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.6.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.6.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.6.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.6.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.6.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.6.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.6.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.7.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.7.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.7.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.7.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.7.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.7.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.7.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.7.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.7.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.7.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.8.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.8.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.8.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.8.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.8.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.8.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.8.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.8.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.8.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.8.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.9.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.9.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.9.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.9.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.9.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.9.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.9.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.9.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.9.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.9.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.10.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.10.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.10.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.10.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.10.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.10.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.10.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.10.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.10.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.10.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.11.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.11.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.11.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.11.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.11.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.11.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.11.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.11.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.11.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.11.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.12.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.12.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.12.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.12.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.12.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.12.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.12.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.12.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.12.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.12.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.13.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.13.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.13.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.13.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.13.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.13.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.13.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.13.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.13.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.13.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.14.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.14.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.14.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.14.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.14.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.14.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.14.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.14.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.14.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.14.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.15.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.15.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.15.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.15.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.15.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.15.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.15.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.15.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.15.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.15.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.16.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.16.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.16.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.16.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.16.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.16.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.16.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.16.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.16.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.16.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.17.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.17.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.17.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.17.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.17.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.17.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.17.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.17.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.17.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.17.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.18.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.18.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.18.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.18.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.18.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.18.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.18.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.18.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.18.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.18.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.19.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.19.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.19.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.19.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.19.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.19.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.19.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.19.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.19.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.19.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.20.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.20.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.20.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.20.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.20.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.20.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.20.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.20.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.20.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.20.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.21.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.21.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.21.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.21.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.21.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.21.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.21.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.21.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.21.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.21.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.22.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.22.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.22.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.22.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.22.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.22.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.22.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.22.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.22.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.22.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.self.query.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.23.attention.self.query.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.self.key.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.23.attention.self.key.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.self.value.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.23.attention.self.value.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.output.dense.weight: torch.Size([1024, 1024])
>>> xlmr.encoder.layer.23.attention.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.23.attention.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.intermediate.dense.weight: torch.Size([4096, 1024])
>>> xlmr.encoder.layer.23.intermediate.dense.bias: torch.Size([4096])
>>> xlmr.encoder.layer.23.output.dense.weight: torch.Size([1024, 4096])
>>> xlmr.encoder.layer.23.output.dense.bias: torch.Size([1024])
>>> xlmr.encoder.layer.23.output.LayerNorm.weight: torch.Size([1024])
>>> xlmr.encoder.layer.23.output.LayerNorm.bias: torch.Size([1024])
>>> xlmr.pooler.dense.weight: torch.Size([1024, 1024])
>>> xlmr.pooler.dense.bias: torch.Size([1024])
>>> trigger_label_ffn.layers.0.weight: torch.Size([450, 1024])
>>> trigger_label_ffn.layers.0.bias: torch.Size([450])
>>> trigger_label_ffn.layers.1.weight: torch.Size([233, 450])
>>> trigger_label_ffn.layers.1.bias: torch.Size([233])
>>> trigger_crf.transition: torch.Size([235, 235])
n_trainable_params: 560511990, n_nontrainable_params: 0
----------------------------------------------------------------------------------------------------
******************************
Epoch: 0
command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4
2023-01-21 08:25:39.664587: step: 4/527, loss: 23.298704147338867
2023-01-21 08:25:40.762314: step: 8/527, loss: 11.17903995513916
2023-01-21 08:25:41.865987: step: 12/527, loss: 26.089847564697266
2023-01-21 08:25:42.950699: step: 16/527, loss: 3.1199731826782227
2023-01-21 08:25:44.047583: step: 20/527, loss: 3.0875887870788574
2023-01-21 08:25:45.125387: step: 24/527, loss: 16.458778381347656
2023-01-21 08:25:46.229449: step: 28/527, loss: 22.77322769165039
2023-01-21 08:25:47.350051: step: 32/527, loss: 14.06472110748291
2023-01-21 08:25:48.499881: step: 36/527, loss: 11.991277694702148
2023-01-21 08:25:49.622611: step: 40/527, loss: 9.615082740783691
2023-01-21 08:25:50.768212: step: 44/527, loss: 13.315023422241211
2023-01-21 08:25:51.877439: step: 48/527, loss: 11.216556549072266
2023-01-21 08:25:53.007980: step: 52/527, loss: 10.861495018005371
2023-01-21 08:25:54.106508: step: 56/527, loss: 19.935651779174805
2023-01-21 08:25:55.196575: step: 60/527, loss: 2.597614288330078
2023-01-21 08:25:56.320166: step: 64/527, loss: 16.626096725463867
2023-01-21 08:25:57.423870: step: 68/527, loss: 12.195844650268555
2023-01-21 08:25:58.531098: step: 72/527, loss: 3.4090874195098877
2023-01-21 08:25:59.648636: step: 76/527, loss: 14.483343124389648
2023-01-21 08:26:00.816042: step: 80/527, loss: 24.72446632385254
2023-01-21 08:26:01.913102: step: 84/527, loss: 5.178360939025879
2023-01-21 08:26:03.023399: step: 88/527, loss: 2.646289348602295
2023-01-21 08:26:04.143491: step: 92/527, loss: 11.701351165771484
2023-01-21 08:26:05.263411: step: 96/527, loss: 10.688973426818848
2023-01-21 08:26:06.362845: step: 100/527, loss: 12.138690948486328
2023-01-21 08:26:07.473506: step: 104/527, loss: 2.0844249725341797
2023-01-21 08:26:08.584665: step: 108/527, loss: 3.36696195602417
2023-01-21 08:26:09.693276: step: 112/527, loss: 29.702808380126953
2023-01-21 08:26:10.818901: step: 116/527, loss: 8.852483749389648
2023-01-21 08:26:11.917630: step: 120/527, loss: 3.291964292526245
2023-01-21 08:26:13.037412: step: 124/527, loss: 16.746883392333984
2023-01-21 08:26:14.192478: step: 128/527, loss: 7.291083812713623
2023-01-21 08:26:15.321778: step: 132/527, loss: 18.865188598632812
2023-01-21 08:26:16.445162: step: 136/527, loss: 3.309422492980957
2023-01-21 08:26:17.551513: step: 140/527, loss: 4.419771194458008
2023-01-21 08:26:18.666388: step: 144/527, loss: 10.253215789794922
2023-01-21 08:26:19.763034: step: 148/527, loss: 19.14675521850586
2023-01-21 08:26:20.854486: step: 152/527, loss: 2.9764788150787354
2023-01-21 08:26:21.960747: step: 156/527, loss: 9.114055633544922
2023-01-21 08:26:23.086863: step: 160/527, loss: 4.554574966430664
2023-01-21 08:26:24.250092: step: 164/527, loss: 4.2900590896606445
2023-01-21 08:26:25.395518: step: 168/527, loss: 4.220833778381348
2023-01-21 08:26:26.501489: step: 172/527, loss: 3.238072395324707
2023-01-21 08:26:27.634676: step: 176/527, loss: 17.893260955810547
2023-01-21 08:26:28.741445: step: 180/527, loss: 4.618285179138184
2023-01-21 08:26:29.858750: step: 184/527, loss: 4.224296569824219
2023-01-21 08:26:30.979025: step: 188/527, loss: 13.275412559509277
2023-01-21 08:26:32.069924: step: 192/527, loss: 4.548651695251465
2023-01-21 08:26:33.188231: step: 196/527, loss: 4.975196838378906
2023-01-21 08:26:34.285052: step: 200/527, loss: 2.763195514678955
2023-01-21 08:26:35.383592: step: 204/527, loss: 9.431214332580566
2023-01-21 08:26:36.502250: step: 208/527, loss: 3.18403959274292
2023-01-21 08:26:37.614203: step: 212/527, loss: 8.645074844360352
2023-01-21 08:26:38.738409: step: 216/527, loss: 13.355669021606445
2023-01-21 08:26:39.844075: step: 220/527, loss: 3.613096237182617
2023-01-21 08:26:40.944503: step: 224/527, loss: 13.06393051147461
2023-01-21 08:26:42.060818: step: 228/527, loss: 13.4402437210083
2023-01-21 08:26:43.170639: step: 232/527, loss: 3.1571521759033203
2023-01-21 08:26:44.288742: step: 236/527, loss: 15.730257034301758
2023-01-21 08:26:45.406318: step: 240/527, loss: 14.734528541564941
2023-01-21 08:26:46.523171: step: 244/527, loss: 29.4234676361084
2023-01-21 08:26:47.638688: step: 248/527, loss: 13.96733570098877
2023-01-21 08:26:48.740842: step: 252/527, loss: 11.352165222167969
2023-01-21 08:26:49.838707: step: 256/527, loss: 13.285268783569336
2023-01-21 08:26:50.971060: step: 260/527, loss: 32.28551483154297
2023-01-21 08:26:52.100495: step: 264/527, loss: 17.00605583190918
2023-01-21 08:26:53.224943: step: 268/527, loss: 2.4271886348724365
2023-01-21 08:26:54.333728: step: 272/527, loss: 3.566790819168091
2023-01-21 08:26:55.502952: step: 276/527, loss: 3.7898523807525635
2023-01-21 08:26:56.586536: step: 280/527, loss: 2.877229928970337
2023-01-21 08:26:57.712497: step: 284/527, loss: 3.207869529724121
2023-01-21 08:26:58.831506: step: 288/527, loss: 3.6515989303588867
2023-01-21 08:26:59.979036: step: 292/527, loss: 16.385099411010742
2023-01-21 08:27:01.102264: step: 296/527, loss: 14.307119369506836
2023-01-21 08:27:02.207160: step: 300/527, loss: 11.828729629516602
2023-01-21 08:27:03.321270: step: 304/527, loss: 3.2087841033935547
2023-01-21 08:27:04.445139: step: 308/527, loss: 8.430937767028809
2023-01-21 08:27:05.588544: step: 312/527, loss: 4.066773414611816
2023-01-21 08:27:06.713626: step: 316/527, loss: 7.552035808563232
2023-01-21 08:27:07.832733: step: 320/527, loss: 14.486663818359375
2023-01-21 08:27:08.924035: step: 324/527, loss: 2.7562456130981445
2023-01-21 08:27:10.051891: step: 328/527, loss: 20.86489486694336
2023-01-21 08:27:11.167750: step: 332/527, loss: 9.661310195922852
2023-01-21 08:27:12.271930: step: 336/527, loss: 2.8045647144317627
2023-01-21 08:27:13.390979: step: 340/527, loss: 15.216832160949707
2023-01-21 08:27:14.509616: step: 344/527, loss: 3.346928119659424
2023-01-21 08:27:15.667879: step: 348/527, loss: 15.470113754272461
2023-01-21 08:27:16.767177: step: 352/527, loss: 19.84587287902832
2023-01-21 08:27:17.876963: step: 356/527, loss: 15.20010757446289
2023-01-21 08:27:18.979830: step: 360/527, loss: 1.4692747592926025
2023-01-21 08:27:20.057153: step: 364/527, loss: 2.4089255332946777
2023-01-21 08:27:21.197366: step: 368/527, loss: 5.787039756774902
2023-01-21 08:27:22.301542: step: 372/527, loss: 11.902318000793457
2023-01-21 08:27:23.415718: step: 376/527, loss: 15.947282791137695
2023-01-21 08:27:24.526775: step: 380/527, loss: 9.210901260375977
2023-01-21 08:27:25.631558: step: 384/527, loss: 11.299958229064941
2023-01-21 08:27:26.734893: step: 388/527, loss: 2.2086942195892334
2023-01-21 08:27:27.859013: step: 392/527, loss: 8.672484397888184
2023-01-21 08:27:28.998872: step: 396/527, loss: 13.195979118347168
2023-01-21 08:27:30.143174: step: 400/527, loss: 11.198301315307617
2023-01-21 08:27:31.254026: step: 404/527, loss: 10.764354705810547
2023-01-21 08:27:32.337196: step: 408/527, loss: 1.770259141921997
2023-01-21 08:27:33.467723: step: 412/527, loss: 6.66694450378418
2023-01-21 08:27:34.555549: step: 416/527, loss: 11.2218017578125
2023-01-21 08:27:35.675493: step: 420/527, loss: 11.64199161529541
2023-01-21 08:27:36.772366: step: 424/527, loss: 5.863048553466797
2023-01-21 08:27:37.890642: step: 428/527, loss: 1.362597942352295
2023-01-21 08:27:39.018116: step: 432/527, loss: 25.305295944213867
2023-01-21 08:27:40.158168: step: 436/527, loss: 19.804800033569336
2023-01-21 08:27:41.269430: step: 440/527, loss: 8.378047943115234
2023-01-21 08:27:42.388869: step: 444/527, loss: 2.0445876121520996
2023-01-21 08:27:43.514503: step: 448/527, loss: 10.909842491149902
2023-01-21 08:27:44.640344: step: 452/527, loss: 2.1260976791381836
2023-01-21 08:27:45.746264: step: 456/527, loss: 9.138895034790039
2023-01-21 08:27:46.878129: step: 460/527, loss: 1.4602832794189453
2023-01-21 08:27:47.964112: step: 464/527, loss: 1.927833080291748
2023-01-21 08:27:49.084167: step: 468/527, loss: 2.6051862239837646
2023-01-21 08:27:50.184373: step: 472/527, loss: 10.984026908874512
2023-01-21 08:27:51.296118: step: 476/527, loss: 5.466901779174805
2023-01-21 08:27:52.390854: step: 480/527, loss: 1.729323387145996
2023-01-21 08:27:53.541114: step: 484/527, loss: 3.0027432441711426
2023-01-21 08:27:54.673495: step: 488/527, loss: 5.657947540283203
2023-01-21 08:27:55.793459: step: 492/527, loss: 2.444866180419922
2023-01-21 08:27:56.909488: step: 496/527, loss: 1.8976857662200928
2023-01-21 08:27:58.036054: step: 500/527, loss: 1.5381652116775513
2023-01-21 08:27:59.167677: step: 504/527, loss: 4.749081134796143
2023-01-21 08:28:00.276620: step: 508/527, loss: 1.6343910694122314
2023-01-21 08:28:01.376367: step: 512/527, loss: 1.5177936553955078
2023-01-21 08:28:02.480202: step: 516/527, loss: 6.4152984619140625
2023-01-21 08:28:03.599594: step: 520/527, loss: 1.229395866394043
2023-01-21 08:28:04.695260: step: 524/527, loss: 7.334926128387451
2023-01-21 08:28:05.796857: step: 528/527, loss: 5.558004379272461
2023-01-21 08:28:06.913880: step: 532/527, loss: 4.138823986053467
2023-01-21 08:28:08.014405: step: 536/527, loss: 1.235190749168396
2023-01-21 08:28:09.147032: step: 540/527, loss: 2.811471462249756
2023-01-21 08:28:10.273625: step: 544/527, loss: 0.8452047109603882
2023-01-21 08:28:11.374889: step: 548/527, loss: 1.1159751415252686
2023-01-21 08:28:12.475783: step: 552/527, loss: 3.3278636932373047
2023-01-21 08:28:13.616745: step: 556/527, loss: 1.3206007480621338
2023-01-21 08:28:14.724442: step: 560/527, loss: 2.820141553878784
2023-01-21 08:28:15.839950: step: 564/527, loss: 3.703164577484131
2023-01-21 08:28:16.970262: step: 568/527, loss: 2.5737457275390625
2023-01-21 08:28:18.122734: step: 572/527, loss: 0.7740475535392761
2023-01-21 08:28:19.224447: step: 576/527, loss: 0.551067054271698
2023-01-21 08:28:20.364880: step: 580/527, loss: 0.9118250012397766
2023-01-21 08:28:21.485726: step: 584/527, loss: 2.628262758255005
2023-01-21 08:28:22.605346: step: 588/527, loss: 1.581235408782959
2023-01-21 08:28:23.709088: step: 592/527, loss: 0.5293468236923218
2023-01-21 08:28:24.814700: step: 596/527, loss: 3.2675068378448486
2023-01-21 08:28:25.928979: step: 600/527, loss: 0.9840301871299744
2023-01-21 08:28:27.060664: step: 604/527, loss: 0.8499903678894043
2023-01-21 08:28:28.168947: step: 608/527, loss: 0.7708979845046997
2023-01-21 08:28:29.293542: step: 612/527, loss: 0.5774157047271729
2023-01-21 08:28:30.438458: step: 616/527, loss: 0.7160772085189819
2023-01-21 08:28:31.550533: step: 620/527, loss: 0.78236323595047
2023-01-21 08:28:32.640759: step: 624/527, loss: 4.907248020172119
2023-01-21 08:28:33.753369: step: 628/527, loss: 1.2145988941192627
2023-01-21 08:28:34.879938: step: 632/527, loss: 2.2446177005767822
2023-01-21 08:28:35.973903: step: 636/527, loss: 1.3998390436172485
2023-01-21 08:28:37.099800: step: 640/527, loss: 6.518995761871338
2023-01-21 08:28:38.184025: step: 644/527, loss: 1.2100446224212646
2023-01-21 08:28:39.303234: step: 648/527, loss: 7.992973327636719
2023-01-21 08:28:40.423274: step: 652/527, loss: 2.0180697441101074
2023-01-21 08:28:41.515821: step: 656/527, loss: 1.200575590133667
2023-01-21 08:28:42.620061: step: 660/527, loss: 1.2557106018066406
2023-01-21 08:28:43.713527: step: 664/527, loss: 0.6658799648284912
2023-01-21 08:28:44.873131: step: 668/527, loss: 2.404552698135376
2023-01-21 08:28:46.010364: step: 672/527, loss: 2.729940176010132
2023-01-21 08:28:47.097355: step: 676/527, loss: 2.278646469116211
2023-01-21 08:28:48.210797: step: 680/527, loss: 0.37135210633277893
2023-01-21 08:28:49.317633: step: 684/527, loss: 0.28588616847991943
2023-01-21 08:28:50.444392: step: 688/527, loss: 2.0488429069519043
2023-01-21 08:28:51.553377: step: 692/527, loss: 0.6801798939704895
2023-01-21 08:28:52.680865: step: 696/527, loss: 1.479017734527588
2023-01-21 08:28:53.770403: step: 700/527, loss: 0.4229392409324646
2023-01-21 08:28:54.913122: step: 704/527, loss: 1.5955603122711182
2023-01-21 08:28:56.050859: step: 708/527, loss: 1.8345609903335571
2023-01-21 08:28:57.173018: step: 712/527, loss: 0.5905696153640747
2023-01-21 08:28:58.259683: step: 716/527, loss: 0.48341989517211914
2023-01-21 08:28:59.356176: step: 720/527, loss: 1.385216236114502
2023-01-21 08:29:00.470830: step: 724/527, loss: 5.307806491851807
2023-01-21 08:29:01.579597: step: 728/527, loss: 0.6275202631950378
2023-01-21 08:29:02.692510: step: 732/527, loss: 0.4168476462364197
2023-01-21 08:29:03.802552: step: 736/527, loss: 1.195639967918396
2023-01-21 08:29:04.939218: step: 740/527, loss: 0.5715931057929993
2023-01-21 08:29:06.052629: step: 744/527, loss: 0.7279413938522339
2023-01-21 08:29:07.175696: step: 748/527, loss: 0.7316750884056091
2023-01-21 08:29:08.323095: step: 752/527, loss: 1.8104383945465088
2023-01-21 08:29:09.456300: step: 756/527, loss: 0.35374611616134644
2023-01-21 08:29:10.592313: step: 760/527, loss: 1.613642930984497
2023-01-21 08:29:11.676514: step: 764/527, loss: 3.2861971855163574
2023-01-21 08:29:12.754991: step: 768/527, loss: 0.9485872983932495
2023-01-21 08:29:13.838481: step: 772/527, loss: 0.6092826128005981
2023-01-21 08:29:14.934269: step: 776/527, loss: 2.5584630966186523
2023-01-21 08:29:16.051606: step: 780/527, loss: 0.3839254379272461
2023-01-21 08:29:17.201281: step: 784/527, loss: 0.7904701232910156
2023-01-21 08:29:18.297152: step: 788/527, loss: 1.8957479000091553
2023-01-21 08:29:19.395280: step: 792/527, loss: 0.2699732184410095
2023-01-21 08:29:20.518684: step: 796/527, loss: 1.9948726892471313
2023-01-21 08:29:21.630748: step: 800/527, loss: 1.1398056745529175
2023-01-21 08:29:22.760104: step: 804/527, loss: 3.541609525680542
2023-01-21 08:29:23.883384: step: 808/527, loss: 0.8285200595855713
2023-01-21 08:29:24.995550: step: 812/527, loss: 0.18635237216949463
2023-01-21 08:29:26.131740: step: 816/527, loss: 0.7432762384414673
2023-01-21 08:29:27.213839: step: 820/527, loss: 1.7212629318237305
2023-01-21 08:29:28.327021: step: 824/527, loss: 0.6409740447998047
2023-01-21 08:29:29.436870: step: 828/527, loss: 1.890131950378418
2023-01-21 08:29:30.553189: step: 832/527, loss: 2.3066790103912354
2023-01-21 08:29:31.659317: step: 836/527, loss: 0.32247740030288696
2023-01-21 08:29:32.787178: step: 840/527, loss: 1.4502936601638794
2023-01-21 08:29:33.916463: step: 844/527, loss: 1.6325743198394775
2023-01-21 08:29:35.034036: step: 848/527, loss: 0.6449156403541565
2023-01-21 08:29:36.170995: step: 852/527, loss: 3.262441396713257
2023-01-21 08:29:37.292032: step: 856/527, loss: 0.27129870653152466
2023-01-21 08:29:38.413421: step: 860/527, loss: 0.4194567799568176
2023-01-21 08:29:39.528749: step: 864/527, loss: 0.4426535665988922
2023-01-21 08:29:40.630552: step: 868/527, loss: 1.0641406774520874
2023-01-21 08:29:41.739745: step: 872/527, loss: 0.5724228024482727
2023-01-21 08:29:42.845742: step: 876/527, loss: 2.0590264797210693
2023-01-21 08:29:43.972523: step: 880/527, loss: 0.6681488752365112
2023-01-21 08:29:45.065217: step: 884/527, loss: 2.3275246620178223
2023-01-21 08:29:46.166724: step: 888/527, loss: 1.1875947713851929
2023-01-21 08:29:47.232692: step: 892/527, loss: 2.179564952850342
2023-01-21 08:29:48.364777: step: 896/527, loss: 2.9457297325134277
2023-01-21 08:29:49.467271: step: 900/527, loss: 4.995299339294434
2023-01-21 08:29:50.597217: step: 904/527, loss: 0.3346892297267914
2023-01-21 08:29:51.714987: step: 908/527, loss: 0.2718140184879303
2023-01-21 08:29:52.868997: step: 912/527, loss: 0.5803699493408203
2023-01-21 08:29:53.958550: step: 916/527, loss: 0.294408917427063
2023-01-21 08:29:55.054211: step: 920/527, loss: 1.2341583967208862
2023-01-21 08:29:56.162971: step: 924/527, loss: 0.4182208180427551
2023-01-21 08:29:57.273040: step: 928/527, loss: 0.4166804254055023
2023-01-21 08:29:58.376282: step: 932/527, loss: 1.071181058883667
2023-01-21 08:29:59.490458: step: 936/527, loss: 7.8886399269104
2023-01-21 08:30:00.619650: step: 940/527, loss: 3.144742488861084
2023-01-21 08:30:01.730162: step: 944/527, loss: 1.3239195346832275
2023-01-21 08:30:02.834578: step: 948/527, loss: 0.9189945459365845
2023-01-21 08:30:03.943370: step: 952/527, loss: 1.1120198965072632
2023-01-21 08:30:05.060969: step: 956/527, loss: 1.6622108221054077
2023-01-21 08:30:06.148457: step: 960/527, loss: 0.6418648362159729
2023-01-21 08:30:07.275717: step: 964/527, loss: 2.4725942611694336
2023-01-21 08:30:08.390938: step: 968/527, loss: 2.957956314086914
2023-01-21 08:30:09.482149: step: 972/527, loss: 0.12524865567684174
2023-01-21 08:30:10.613952: step: 976/527, loss: 4.013078212738037
2023-01-21 08:30:11.736357: step: 980/527, loss: 0.19023752212524414
2023-01-21 08:30:12.871943: step: 984/527, loss: 1.1993707418441772
2023-01-21 08:30:13.953993: step: 988/527, loss: 0.4049256443977356
2023-01-21 08:30:15.055982: step: 992/527, loss: 0.6819015741348267
2023-01-21 08:30:16.218907: step: 996/527, loss: 0.4340020716190338
2023-01-21 08:30:17.327915: step: 1000/527, loss: 0.7013236880302429
2023-01-21 08:30:18.418006: step: 1004/527, loss: 2.070523977279663
2023-01-21 08:30:19.533130: step: 1008/527, loss: 3.735775947570801
2023-01-21 08:30:20.677377: step: 1012/527, loss: 0.4360175132751465
2023-01-21 08:30:21.804549: step: 1016/527, loss: 0.5247083902359009
2023-01-21 08:30:22.964979: step: 1020/527, loss: 3.2997775077819824
2023-01-21 08:30:24.112393: step: 1024/527, loss: 2.4600210189819336
2023-01-21 08:30:25.217727: step: 1028/527, loss: 0.5743644833564758
2023-01-21 08:30:26.321445: step: 1032/527, loss: 1.6084976196289062
2023-01-21 08:30:27.436643: step: 1036/527, loss: 1.3321189880371094
2023-01-21 08:30:28.565982: step: 1040/527, loss: 0.19133087992668152
2023-01-21 08:30:29.673525: step: 1044/527, loss: 0.4211674928665161
2023-01-21 08:30:30.796516: step: 1048/527, loss: 0.6423962712287903
2023-01-21 08:30:31.939770: step: 1052/527, loss: 1.8461812734603882
2023-01-21 08:30:33.067776: step: 1056/527, loss: 2.1372907161712646
2023-01-21 08:30:34.166134: step: 1060/527, loss: 1.204171895980835
2023-01-21 08:30:35.265228: step: 1064/527, loss: 0.3970535397529602
2023-01-21 08:30:36.372686: step: 1068/527, loss: 0.43122321367263794
2023-01-21 08:30:37.490985: step: 1072/527, loss: 2.170353889465332
2023-01-21 08:30:38.593797: step: 1076/527, loss: 1.472381830215454
2023-01-21 08:30:39.692983: step: 1080/527, loss: 0.9341005086898804
2023-01-21 08:30:40.789584: step: 1084/527, loss: 1.4656577110290527
2023-01-21 08:30:41.921230: step: 1088/527, loss: 0.5932026505470276
2023-01-21 08:30:43.026031: step: 1092/527, loss: 3.2769429683685303
2023-01-21 08:30:44.138940: step: 1096/527, loss: 0.6678022146224976
2023-01-21 08:30:45.234892: step: 1100/527, loss: 1.446798324584961
2023-01-21 08:30:46.344034: step: 1104/527, loss: 3.290282964706421
2023-01-21 08:30:47.444805: step: 1108/527, loss: 1.2525103092193604
2023-01-21 08:30:48.549080: step: 1112/527, loss: 1.8739287853240967
2023-01-21 08:30:49.655232: step: 1116/527, loss: 1.6859760284423828
2023-01-21 08:30:50.822014: step: 1120/527, loss: 0.8996938467025757
2023-01-21 08:30:51.896869: step: 1124/527, loss: 1.9738459587097168
2023-01-21 08:30:53.001976: step: 1128/527, loss: 3.137787103652954
2023-01-21 08:30:54.131659: step: 1132/527, loss: 0.5248814821243286
2023-01-21 08:30:55.231950: step: 1136/527, loss: 2.0746726989746094
2023-01-21 08:30:56.325531: step: 1140/527, loss: 1.0960570573806763
2023-01-21 08:30:57.426650: step: 1144/527, loss: 0.31056922674179077
2023-01-21 08:30:58.541610: step: 1148/527, loss: 2.9502995014190674
2023-01-21 08:30:59.657437: step: 1152/527, loss: 0.5570597052574158
2023-01-21 08:31:00.764453: step: 1156/527, loss: 0.8082716464996338
2023-01-21 08:31:01.874955: step: 1160/527, loss: 0.723256528377533
2023-01-21 08:31:02.970704: step: 1164/527, loss: 0.4372190833091736
2023-01-21 08:31:04.046192: step: 1168/527, loss: 0.2185339480638504
2023-01-21 08:31:05.145690: step: 1172/527, loss: 0.6127884984016418
2023-01-21 08:31:06.255174: step: 1176/527, loss: 0.5128602981567383
2023-01-21 08:31:07.416513: step: 1180/527, loss: 0.7026681900024414
2023-01-21 08:31:08.527835: step: 1184/527, loss: 0.5505791306495667
2023-01-21 08:31:09.640480: step: 1188/527, loss: 0.39161816239356995
2023-01-21 08:31:10.738686: step: 1192/527, loss: 0.4343474507331848
2023-01-21 08:31:11.861511: step: 1196/527, loss: 0.5897807478904724
2023-01-21 08:31:12.971330: step: 1200/527, loss: 0.3465515971183777
2023-01-21 08:31:14.120016: step: 1204/527, loss: 0.594846785068512
2023-01-21 08:31:15.229743: step: 1208/527, loss: 2.2052197456359863
2023-01-21 08:31:16.321779: step: 1212/527, loss: 0.4247792959213257
2023-01-21 08:31:17.431242: step: 1216/527, loss: 0.24631667137145996
2023-01-21 08:31:18.557590: step: 1220/527, loss: 2.401458978652954
2023-01-21 08:31:19.664792: step: 1224/527, loss: 2.119553327560425
2023-01-21 08:31:20.748081: step: 1228/527, loss: 0.43200409412384033
2023-01-21 08:31:21.851902: step: 1232/527, loss: 1.5831866264343262
2023-01-21 08:31:22.981917: step: 1236/527, loss: 0.5252863168716431
2023-01-21 08:31:24.081692: step: 1240/527, loss: 0.8064610362052917
2023-01-21 08:31:25.199865: step: 1244/527, loss: 2.0962066650390625
2023-01-21 08:31:26.330693: step: 1248/527, loss: 0.515377402305603
2023-01-21 08:31:27.425553: step: 1252/527, loss: 0.27525532245635986
2023-01-21 08:31:28.559522: step: 1256/527, loss: 0.5191051959991455
2023-01-21 08:31:29.650077: step: 1260/527, loss: 0.2484794557094574
2023-01-21 08:31:30.759873: step: 1264/527, loss: 1.324589729309082
2023-01-21 08:31:31.865743: step: 1268/527, loss: 0.7637085318565369
2023-01-21 08:31:32.964972: step: 1272/527, loss: 0.6067211627960205
2023-01-21 08:31:34.100032: step: 1276/527, loss: 0.7518082857131958
2023-01-21 08:31:35.247387: step: 1280/527, loss: 0.5272313356399536
2023-01-21 08:31:36.331734: step: 1284/527, loss: 3.1354522705078125
2023-01-21 08:31:37.450898: step: 1288/527, loss: 0.8052797317504883
2023-01-21 08:31:38.588229: step: 1292/527, loss: 1.3554528951644897
2023-01-21 08:31:39.709615: step: 1296/527, loss: 0.8909964561462402
2023-01-21 08:31:40.835636: step: 1300/527, loss: 0.5566272735595703
2023-01-21 08:31:41.956269: step: 1304/527, loss: 0.4766858220100403
2023-01-21 08:31:43.067151: step: 1308/527, loss: 1.9354594945907593
2023-01-21 08:31:44.195716: step: 1312/527, loss: 0.29067105054855347
2023-01-21 08:31:45.303938: step: 1316/527, loss: 0.27712929248809814
2023-01-21 08:31:46.425840: step: 1320/527, loss: 0.4867759943008423
2023-01-21 08:31:47.555467: step: 1324/527, loss: 0.4455300569534302
2023-01-21 08:31:48.652631: step: 1328/527, loss: 0.2211351990699768
2023-01-21 08:31:49.751084: step: 1332/527, loss: 0.28282785415649414
2023-01-21 08:31:50.848704: step: 1336/527, loss: 2.0443952083587646
2023-01-21 08:31:51.954061: step: 1340/527, loss: 0.44707414507865906
2023-01-21 08:31:53.086462: step: 1344/527, loss: 1.613107681274414
2023-01-21 08:31:54.199144: step: 1348/527, loss: 1.103171944618225
2023-01-21 08:31:55.298585: step: 1352/527, loss: 1.4725905656814575
2023-01-21 08:31:56.405164: step: 1356/527, loss: 1.538714051246643
2023-01-21 08:31:57.523902: step: 1360/527, loss: 2.90914249420166
2023-01-21 08:31:58.613749: step: 1364/527, loss: 0.20819544792175293
2023-01-21 08:31:59.726308: step: 1368/527, loss: 4.111978530883789
2023-01-21 08:32:00.856260: step: 1372/527, loss: 2.559821844100952
2023-01-21 08:32:01.952840: step: 1376/527, loss: 0.5388041734695435
2023-01-21 08:32:03.066555: step: 1380/527, loss: 0.3268166780471802
2023-01-21 08:32:04.173907: step: 1384/527, loss: 0.28429141640663147
2023-01-21 08:32:05.288832: step: 1388/527, loss: 0.26779043674468994
2023-01-21 08:32:06.398197: step: 1392/527, loss: 0.1634858250617981
2023-01-21 08:32:07.498879: step: 1396/527, loss: 0.23047037422657013
2023-01-21 08:32:08.614031: step: 1400/527, loss: 1.6838372945785522
2023-01-21 08:32:09.737175: step: 1404/527, loss: 0.8953489661216736
2023-01-21 08:32:10.834774: step: 1408/527, loss: 0.424577534198761
2023-01-21 08:32:11.939268: step: 1412/527, loss: 1.6298184394836426
2023-01-21 08:32:13.064273: step: 1416/527, loss: 0.4104093909263611
2023-01-21 08:32:14.163410: step: 1420/527, loss: 0.95198655128479
2023-01-21 08:32:15.308517: step: 1424/527, loss: 0.5748883485794067
2023-01-21 08:32:16.405846: step: 1428/527, loss: 0.3222489356994629
2023-01-21 08:32:17.544884: step: 1432/527, loss: 2.584757089614868
2023-01-21 08:32:18.654915: step: 1436/527, loss: 0.9204427003860474
2023-01-21 08:32:19.778146: step: 1440/527, loss: 0.5373786687850952
2023-01-21 08:32:20.877132: step: 1444/527, loss: 0.37053626775741577
2023-01-21 08:32:21.987316: step: 1448/527, loss: 0.9937222003936768
2023-01-21 08:32:23.115781: step: 1452/527, loss: 0.3383520245552063
2023-01-21 08:32:24.229932: step: 1456/527, loss: 0.6849720478057861
2023-01-21 08:32:25.317421: step: 1460/527, loss: 0.2560383081436157
2023-01-21 08:32:26.426221: step: 1464/527, loss: 1.824838638305664
2023-01-21 08:32:27.523773: step: 1468/527, loss: 0.3990805745124817
2023-01-21 08:32:28.637951: step: 1472/527, loss: 1.3486689329147339
2023-01-21 08:32:29.759579: step: 1476/527, loss: 1.3528306484222412
2023-01-21 08:32:30.899481: step: 1480/527, loss: 1.2125530242919922
2023-01-21 08:32:31.998629: step: 1484/527, loss: 0.8566564321517944
2023-01-21 08:32:33.126320: step: 1488/527, loss: 0.4572031497955322
2023-01-21 08:32:34.227822: step: 1492/527, loss: 0.18106165528297424
2023-01-21 08:32:35.358285: step: 1496/527, loss: 1.1860369443893433
2023-01-21 08:32:36.442386: step: 1500/527, loss: 0.6649714708328247
2023-01-21 08:32:37.544050: step: 1504/527, loss: 0.19908590614795685
2023-01-21 08:32:38.684596: step: 1508/527, loss: 1.3781462907791138
2023-01-21 08:32:39.799398: step: 1512/527, loss: 0.8104506731033325
2023-01-21 08:32:40.899530: step: 1516/527, loss: 0.824360728263855
2023-01-21 08:32:42.031202: step: 1520/527, loss: 0.29937633872032166
2023-01-21 08:32:43.162724: step: 1524/527, loss: 4.915238380432129
2023-01-21 08:32:44.302220: step: 1528/527, loss: 2.220987319946289
2023-01-21 08:32:45.393919: step: 1532/527, loss: 0.7077765464782715
2023-01-21 08:32:46.491766: step: 1536/527, loss: 0.37568509578704834
2023-01-21 08:32:47.606527: step: 1540/527, loss: 2.2647054195404053
2023-01-21 08:32:48.683071: step: 1544/527, loss: 1.9965778589248657
2023-01-21 08:32:49.816173: step: 1548/527, loss: 1.3086352348327637
2023-01-21 08:32:50.922958: step: 1552/527, loss: 0.6749702095985413
2023-01-21 08:32:52.052120: step: 1556/527, loss: 0.4000672399997711
2023-01-21 08:32:53.163889: step: 1560/527, loss: 1.5065810680389404
2023-01-21 08:32:54.295674: step: 1564/527, loss: 2.4180479049682617
2023-01-21 08:32:55.401447: step: 1568/527, loss: 0.3390617370605469
2023-01-21 08:32:56.506915: step: 1572/527, loss: 1.172560214996338
2023-01-21 08:32:57.639081: step: 1576/527, loss: 0.6292140483856201
2023-01-21 08:32:58.768424: step: 1580/527, loss: 0.8719101548194885
2023-01-21 08:32:59.917020: step: 1584/527, loss: 1.8703562021255493
2023-01-21 08:33:01.046076: step: 1588/527, loss: 0.7215129733085632
2023-01-21 08:33:02.162756: step: 1592/527, loss: 3.0191850662231445
2023-01-21 08:33:03.269570: step: 1596/527, loss: 1.450263500213623
2023-01-21 08:33:04.402487: step: 1600/527, loss: 2.0740275382995605
2023-01-21 08:33:05.500665: step: 1604/527, loss: 2.142699718475342
2023-01-21 08:33:06.605484: step: 1608/527, loss: 0.16845369338989258
2023-01-21 08:33:07.719856: step: 1612/527, loss: 0.32914966344833374
2023-01-21 08:33:08.853295: step: 1616/527, loss: 0.7662199139595032
2023-01-21 08:33:09.979366: step: 1620/527, loss: 1.9744610786437988
2023-01-21 08:33:11.074864: step: 1624/527, loss: 0.4635851979255676
2023-01-21 08:33:12.203507: step: 1628/527, loss: 0.6281766891479492
2023-01-21 08:33:13.302145: step: 1632/527, loss: 1.1585177183151245
2023-01-21 08:33:14.450311: step: 1636/527, loss: 2.6801319122314453
2023-01-21 08:33:15.556574: step: 1640/527, loss: 0.7020303010940552
2023-01-21 08:33:16.666645: step: 1644/527, loss: 0.3365822434425354
2023-01-21 08:33:17.782947: step: 1648/527, loss: 1.0705496072769165
2023-01-21 08:33:18.901037: step: 1652/527, loss: 1.5442209243774414
2023-01-21 08:33:19.991597: step: 1656/527, loss: 4.564550876617432
2023-01-21 08:33:21.127578: step: 1660/527, loss: 0.31720733642578125
2023-01-21 08:33:22.221218: step: 1664/527, loss: 1.8574330806732178
2023-01-21 08:33:23.332829: step: 1668/527, loss: 0.9798682928085327
2023-01-21 08:33:24.448157: step: 1672/527, loss: 0.19750958681106567
2023-01-21 08:33:25.552819: step: 1676/527, loss: 0.18295665085315704
2023-01-21 08:33:26.703524: step: 1680/527, loss: 0.5710514783859253
2023-01-21 08:33:27.813656: step: 1684/527, loss: 0.40340834856033325
2023-01-21 08:33:28.902365: step: 1688/527, loss: 0.45031118392944336
2023-01-21 08:33:30.045190: step: 1692/527, loss: 0.49437159299850464
2023-01-21 08:33:31.175536: step: 1696/527, loss: 0.44882121682167053
2023-01-21 08:33:32.301848: step: 1700/527, loss: 2.6795060634613037
2023-01-21 08:33:33.413201: step: 1704/527, loss: 0.688677191734314
2023-01-21 08:33:34.521864: step: 1708/527, loss: 1.096032977104187
2023-01-21 08:33:35.601305: step: 1712/527, loss: 0.5273263454437256
2023-01-21 08:33:36.733532: step: 1716/527, loss: 1.9204206466674805
2023-01-21 08:33:37.834445: step: 1720/527, loss: 0.2613970637321472
2023-01-21 08:33:38.957774: step: 1724/527, loss: 2.618271827697754
2023-01-21 08:33:40.086997: step: 1728/527, loss: 0.5051380395889282
2023-01-21 08:33:41.187692: step: 1732/527, loss: 1.3047374486923218
2023-01-21 08:33:42.291469: step: 1736/527, loss: 0.473741739988327
2023-01-21 08:33:43.403619: step: 1740/527, loss: 1.3489004373550415
2023-01-21 08:33:44.509956: step: 1744/527, loss: 0.42201292514801025
2023-01-21 08:33:45.637541: step: 1748/527, loss: 2.2122833728790283
2023-01-21 08:33:46.777446: step: 1752/527, loss: 1.8622092008590698
2023-01-21 08:33:47.875209: step: 1756/527, loss: 1.1798655986785889
2023-01-21 08:33:48.981466: step: 1760/527, loss: 0.4417538642883301
2023-01-21 08:33:50.093823: step: 1764/527, loss: 1.4459447860717773
2023-01-21 08:33:51.205525: step: 1768/527, loss: 0.43311917781829834
2023-01-21 08:33:52.294438: step: 1772/527, loss: 0.28755927085876465
2023-01-21 08:33:53.373997: step: 1776/527, loss: 4.705626487731934
2023-01-21 08:33:54.494220: step: 1780/527, loss: 6.888480186462402
2023-01-21 08:33:55.586120: step: 1784/527, loss: 3.0172150135040283
2023-01-21 08:33:56.743736: step: 1788/527, loss: 0.27122220396995544
2023-01-21 08:33:57.853599: step: 1792/527, loss: 1.080742597579956
2023-01-21 08:33:58.961836: step: 1796/527, loss: 0.5888214111328125
2023-01-21 08:34:00.074480: step: 1800/527, loss: 0.33384573459625244
2023-01-21 08:34:01.214349: step: 1804/527, loss: 0.5154205560684204
2023-01-21 08:34:02.299983: step: 1808/527, loss: 2.7377469539642334
2023-01-21 08:34:03.386529: step: 1812/527, loss: 0.7977047562599182
2023-01-21 08:34:04.508900: step: 1816/527, loss: 0.5075576305389404
2023-01-21 08:34:05.630896: step: 1820/527, loss: 0.4341048002243042
2023-01-21 08:34:06.746835: step: 1824/527, loss: 0.9525284767150879
2023-01-21 08:34:07.890882: step: 1828/527, loss: 0.7073334455490112
2023-01-21 08:34:09.003898: step: 1832/527, loss: 2.572289228439331
2023-01-21 08:34:10.116535: step: 1836/527, loss: 0.3693413734436035
2023-01-21 08:34:11.223446: step: 1840/527, loss: 2.424553871154785
2023-01-21 08:34:12.339731: step: 1844/527, loss: 1.1136678457260132
2023-01-21 08:34:13.463289: step: 1848/527, loss: 1.049425721168518
2023-01-21 08:34:14.581255: step: 1852/527, loss: 8.466513633728027
2023-01-21 08:34:15.684178: step: 1856/527, loss: 0.2021808624267578
2023-01-21 08:34:16.769760: step: 1860/527, loss: 0.2533864974975586
2023-01-21 08:34:17.871244: step: 1864/527, loss: 0.267726331949234
2023-01-21 08:34:18.971863: step: 1868/527, loss: 0.19612160325050354
2023-01-21 08:34:20.076979: step: 1872/527, loss: 1.1699615716934204
2023-01-21 08:34:21.189296: step: 1876/527, loss: 3.9304897785186768
2023-01-21 08:34:22.340593: step: 1880/527, loss: 0.5027409791946411
2023-01-21 08:34:23.447788: step: 1884/527, loss: 2.0073020458221436
2023-01-21 08:34:24.582162: step: 1888/527, loss: 0.40575096011161804
2023-01-21 08:34:25.704408: step: 1892/527, loss: 1.5256905555725098
2023-01-21 08:34:26.830042: step: 1896/527, loss: 4.968748092651367
2023-01-21 08:34:27.909415: step: 1900/527, loss: 0.299780011177063
2023-01-21 08:34:29.017643: step: 1904/527, loss: 0.5224254131317139
2023-01-21 08:34:30.112310: step: 1908/527, loss: 0.21581010520458221
2023-01-21 08:34:31.185977: step: 1912/527, loss: 1.128339171409607
2023-01-21 08:34:32.322078: step: 1916/527, loss: 2.557401418685913
2023-01-21 08:34:33.423262: step: 1920/527, loss: 1.1100330352783203
2023-01-21 08:34:34.524816: step: 1924/527, loss: 0.498715341091156
2023-01-21 08:34:35.630803: step: 1928/527, loss: 7.218281269073486
2023-01-21 08:34:36.750259: step: 1932/527, loss: 0.997623085975647
2023-01-21 08:34:37.821584: step: 1936/527, loss: 0.27027639746665955
2023-01-21 08:34:38.924067: step: 1940/527, loss: 0.5404546856880188
2023-01-21 08:34:40.035260: step: 1944/527, loss: 0.8608279228210449
2023-01-21 08:34:41.145939: step: 1948/527, loss: 0.4378185272216797
2023-01-21 08:34:42.269888: step: 1952/527, loss: 1.277090311050415
2023-01-21 08:34:43.365557: step: 1956/527, loss: 0.1199650838971138
2023-01-21 08:34:44.463283: step: 1960/527, loss: 0.8804100155830383
2023-01-21 08:34:45.620838: step: 1964/527, loss: 0.2628374397754669
2023-01-21 08:34:46.724268: step: 1968/527, loss: 7.704134941101074
2023-01-21 08:34:47.818435: step: 1972/527, loss: 0.7970272898674011
2023-01-21 08:34:48.932755: step: 1976/527, loss: 0.7692806124687195
2023-01-21 08:34:50.058293: step: 1980/527, loss: 2.452371120452881
2023-01-21 08:34:51.211570: step: 1984/527, loss: 0.6181381344795227
2023-01-21 08:34:52.337364: step: 1988/527, loss: 0.2897105813026428
2023-01-21 08:34:53.476899: step: 1992/527, loss: 3.9894003868103027
2023-01-21 08:34:54.589544: step: 1996/527, loss: 0.5279205441474915
2023-01-21 08:34:55.682925: step: 2000/527, loss: 0.34472066164016724
2023-01-21 08:34:56.774177: step: 2004/527, loss: 0.19982147216796875
2023-01-21 08:34:57.927179: step: 2008/527, loss: 0.5787714719772339
2023-01-21 08:34:59.054548: step: 2012/527, loss: 0.1962549388408661
2023-01-21 08:35:00.178798: step: 2016/527, loss: 1.4819895029067993
2023-01-21 08:35:01.277601: step: 2020/527, loss: 0.90080326795578
2023-01-21 08:35:02.398051: step: 2024/527, loss: 0.816665530204773
2023-01-21 08:35:03.497487: step: 2028/527, loss: 0.7738326787948608
2023-01-21 08:35:04.632655: step: 2032/527, loss: 2.7024121284484863
2023-01-21 08:35:05.760594: step: 2036/527, loss: 0.4749559462070465
2023-01-21 08:35:06.984959: step: 2040/527, loss: 0.30739325284957886
2023-01-21 08:35:08.080151: step: 2044/527, loss: 0.36025017499923706
2023-01-21 08:35:09.176944: step: 2048/527, loss: 0.5957506895065308
2023-01-21 08:35:10.280109: step: 2052/527, loss: 0.33476340770721436
2023-01-21 08:35:11.363048: step: 2056/527, loss: 1.6934397220611572
2023-01-21 08:35:12.494195: step: 2060/527, loss: 0.4859869182109833
2023-01-21 08:35:13.633324: step: 2064/527, loss: 1.5259751081466675
2023-01-21 08:35:14.759190: step: 2068/527, loss: 0.7743636965751648
2023-01-21 08:35:15.889730: step: 2072/527, loss: 0.3628057539463043
2023-01-21 08:35:16.993333: step: 2076/527, loss: 1.2008018493652344
2023-01-21 08:35:18.109808: step: 2080/527, loss: 0.3144242465496063
2023-01-21 08:35:19.222136: step: 2084/527, loss: 0.2734830379486084
2023-01-21 08:35:20.319704: step: 2088/527, loss: 0.40227755904197693
2023-01-21 08:35:21.450142: step: 2092/527, loss: 0.571160614490509
2023-01-21 08:35:22.572543: step: 2096/527, loss: 1.4052577018737793
2023-01-21 08:35:23.693819: step: 2100/527, loss: 3.8038036823272705
2023-01-21 08:35:24.806100: step: 2104/527, loss: 0.20042724907398224
2023-01-21 08:35:25.926190: step: 2108/527, loss: 0.31341809034347534
==================================================
Loss: 3.319
--------------------
Dev: {'event': {'p': 0.627208480565371, 'r': 0.47270306258322237, 'f1': 0.5391040242976463}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Test: {'event': {'p': 0.5309396485867074, 'r': 0.39714285714285713, 'f1': 0.4543968617195162}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Chinese: {'event': {'p': 0.5769230769230769, 'r': 0.2777777777777778, 'f1': 0.375}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Korean: {'event': {'p': 0.7647058823529411, 'r': 0.20634920634920634, 'f1': 0.32499999999999996}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Russian: {'event': {'p': 0.5, 'r': 0.08333333333333333, 'f1': 0.14285714285714285}, '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': 0.627208480565371, 'r': 0.47270306258322237, 'f1': 0.5391040242976463}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Eng Test for Chinese: {'event': {'p': 0.5309396485867074, 'r': 0.39714285714285713, 'f1': 0.4543968617195162}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Sample Chinese: {'event': {'p': 0.5769230769230769, 'r': 0.2777777777777778, 'f1': 0.375}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
--------------------
Eng Dev for Korean: {'event': {'p': 0.627208480565371, 'r': 0.47270306258322237, 'f1': 0.5391040242976463}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Eng Test for Korean: {'event': {'p': 0.5309396485867074, 'r': 0.39714285714285713, 'f1': 0.4543968617195162}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Sample Korean: {'event': {'p': 0.7647058823529411, 'r': 0.20634920634920634, 'f1': 0.32499999999999996}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
--------------------
Eng Dev for Russian: {'event': {'p': 0.627208480565371, 'r': 0.47270306258322237, 'f1': 0.5391040242976463}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Eng Test for Russian: {'event': {'p': 0.5309396485867074, 'r': 0.39714285714285713, 'f1': 0.4543968617195162}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0}
Sample Russian: {'event': {'p': 0.5, 'r': 0.08333333333333333, 'f1': 0.14285714285714285}, '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 08:36:25.532047: step: 4/527, loss: 0.48955804109573364
2023-01-21 08:36:26.670566: step: 8/527, loss: 1.466174840927124
2023-01-21 08:36:27.805766: step: 12/527, loss: 0.35630694031715393
2023-01-21 08:36:28.926809: step: 16/527, loss: 0.16813354194164276
2023-01-21 08:36:30.046388: step: 20/527, loss: 0.5058125853538513
2023-01-21 08:36:31.170960: step: 24/527, loss: 0.996319055557251
2023-01-21 08:36:32.290198: step: 28/527, loss: 1.1384345293045044
2023-01-21 08:36:33.386942: step: 32/527, loss: 1.8335384130477905
2023-01-21 08:36:34.507554: step: 36/527, loss: 0.3281337022781372
2023-01-21 08:36:35.616252: step: 40/527, loss: 0.4102742373943329
2023-01-21 08:36:36.743334: step: 44/527, loss: 0.9760766625404358
2023-01-21 08:36:37.842058: step: 48/527, loss: 0.4607202112674713
2023-01-21 08:36:38.957914: step: 52/527, loss: 0.21661405265331268
2023-01-21 08:36:40.044655: step: 56/527, loss: 0.7160504460334778
2023-01-21 08:36:41.141774: step: 60/527, loss: 0.27713119983673096
2023-01-21 08:36:42.237489: step: 64/527, loss: 0.9000151753425598
2023-01-21 08:36:43.369284: step: 68/527, loss: 1.534056544303894
2023-01-21 08:36:44.501057: step: 72/527, loss: 0.6578210592269897
2023-01-21 08:36:45.575950: step: 76/527, loss: 1.5781265497207642
2023-01-21 08:36:46.701933: step: 80/527, loss: 0.3551687002182007
2023-01-21 08:36:47.813647: step: 84/527, loss: 6.695432662963867
2023-01-21 08:36:48.908365: step: 88/527, loss: 0.21213141083717346
2023-01-21 08:36:50.032918: step: 92/527, loss: 0.30454015731811523
2023-01-21 08:36:51.173402: step: 96/527, loss: 0.21898995339870453
2023-01-21 08:36:52.273822: step: 100/527, loss: 0.6064965724945068
2023-01-21 08:36:53.375220: step: 104/527, loss: 0.32294169068336487
2023-01-21 08:36:54.516112: step: 108/527, loss: 7.444394111633301
2023-01-21 08:36:55.648674: step: 112/527, loss: 1.1095454692840576
2023-01-21 08:36:56.760707: step: 116/527, loss: 0.29709187150001526
2023-01-21 08:36:57.892627: step: 120/527, loss: 0.13045310974121094
2023-01-21 08:36:58.996704: step: 124/527, loss: 0.16961747407913208
2023-01-21 08:37:00.080457: step: 128/527, loss: 0.17125794291496277
2023-01-21 08:37:01.198532: step: 132/527, loss: 5.192537784576416
2023-01-21 08:37:02.292032: step: 136/527, loss: 1.659091591835022
2023-01-21 08:37:03.380921: step: 140/527, loss: 0.2833220660686493
2023-01-21 08:37:04.486350: step: 144/527, loss: 0.6889618039131165
2023-01-21 08:37:05.587116: step: 148/527, loss: 0.7711418867111206
2023-01-21 08:37:06.741827: step: 152/527, loss: 2.9161343574523926
2023-01-21 08:37:07.883672: step: 156/527, loss: 0.4966874122619629
2023-01-21 08:37:09.004390: step: 160/527, loss: 0.3882254958152771
2023-01-21 08:37:10.091621: step: 164/527, loss: 0.8941822052001953
2023-01-21 08:37:11.222310: step: 168/527, loss: 0.547783613204956
2023-01-21 08:37:12.344727: step: 172/527, loss: 0.16548863053321838
2023-01-21 08:37:13.431042: step: 176/527, loss: 0.9105170965194702
2023-01-21 08:37:14.547132: step: 180/527, loss: 1.0770020484924316
2023-01-21 08:37:15.650242: step: 184/527, loss: 0.17274294793605804
2023-01-21 08:37:16.750181: step: 188/527, loss: 0.5551362633705139
2023-01-21 08:37:17.864272: step: 192/527, loss: 2.2001235485076904
2023-01-21 08:37:19.003489: step: 196/527, loss: 2.36144757270813
2023-01-21 08:37:20.131754: step: 200/527, loss: 1.6640422344207764
2023-01-21 08:37:21.251951: step: 204/527, loss: 0.40534576773643494
2023-01-21 08:37:22.352789: step: 208/527, loss: 0.8676781058311462
2023-01-21 08:37:23.459079: step: 212/527, loss: 0.8855735659599304
2023-01-21 08:37:24.565189: step: 216/527, loss: 0.34090396761894226
2023-01-21 08:37:25.667562: step: 220/527, loss: 0.10328011959791183
2023-01-21 08:37:26.783116: step: 224/527, loss: 0.34760522842407227
2023-01-21 08:37:27.891101: step: 228/527, loss: 0.2622261941432953
2023-01-21 08:37:28.991834: step: 232/527, loss: 0.7437978982925415
2023-01-21 08:37:30.114745: step: 236/527, loss: 1.2518001794815063
2023-01-21 08:37:31.205886: step: 240/527, loss: 0.5707436800003052
2023-01-21 08:37:32.320200: step: 244/527, loss: 0.7825153470039368
2023-01-21 08:37:33.477424: step: 248/527, loss: 1.836161732673645
2023-01-21 08:37:34.589979: step: 252/527, loss: 0.3499748110771179
2023-01-21 08:37:35.738422: step: 256/527, loss: 2.0855844020843506
2023-01-21 08:37:36.841984: step: 260/527, loss: 0.386616051197052
2023-01-21 08:37:37.992386: step: 264/527, loss: 1.3517508506774902
2023-01-21 08:37:39.080981: step: 268/527, loss: 0.2443222999572754
2023-01-21 08:37:40.194925: step: 272/527, loss: 1.2600563764572144
2023-01-21 08:37:41.321710: step: 276/527, loss: 0.7719128131866455
2023-01-21 08:37:42.434403: step: 280/527, loss: 3.9093616008758545
2023-01-21 08:37:43.542104: step: 284/527, loss: 0.8474264740943909
2023-01-21 08:37:44.662081: step: 288/527, loss: 0.904846727848053
2023-01-21 08:37:45.752195: step: 292/527, loss: 0.543824315071106
2023-01-21 08:37:46.866569: step: 296/527, loss: 1.0935778617858887
2023-01-21 08:37:47.983837: step: 300/527, loss: 0.9467129707336426
2023-01-21 08:37:49.079522: step: 304/527, loss: 0.3917180895805359
2023-01-21 08:37:50.184745: step: 308/527, loss: 0.32930245995521545
2023-01-21 08:37:51.297777: step: 312/527, loss: 0.18159084022045135
2023-01-21 08:37:52.409899: step: 316/527, loss: 0.9713813066482544
2023-01-21 08:37:53.508873: step: 320/527, loss: 0.2717186212539673
2023-01-21 08:37:54.639057: step: 324/527, loss: 0.5812457799911499
2023-01-21 08:37:55.770114: step: 328/527, loss: 1.818238377571106
2023-01-21 08:37:56.876525: step: 332/527, loss: 0.4060227572917938
2023-01-21 08:37:57.959285: step: 336/527, loss: 6.988563537597656
2023-01-21 08:37:59.088058: step: 340/527, loss: 1.8308956623077393
2023-01-21 08:38:00.171453: step: 344/527, loss: 0.9451266527175903
2023-01-21 08:38:01.327226: step: 348/527, loss: 0.4262138307094574
2023-01-21 08:38:02.440894: step: 352/527, loss: 0.1653548777103424
2023-01-21 08:38:03.569921: step: 356/527, loss: 0.6653217077255249
2023-01-21 08:38:04.667077: step: 360/527, loss: 0.24937179684638977
2023-01-21 08:38:05.788552: step: 364/527, loss: 1.1947906017303467
2023-01-21 08:38:06.872655: step: 368/527, loss: 0.7476380467414856
2023-01-21 08:38:07.971898: step: 372/527, loss: 0.7044621706008911
2023-01-21 08:38:09.100071: step: 376/527, loss: 1.4395205974578857
2023-01-21 08:38:10.249074: step: 380/527, loss: 0.46668338775634766
2023-01-21 08:38:11.381269: step: 384/527, loss: 0.6788533926010132
2023-01-21 08:38:12.481107: step: 388/527, loss: 6.986363410949707
2023-01-21 08:38:13.571496: step: 392/527, loss: 0.22851315140724182
2023-01-21 08:38:14.670282: step: 396/527, loss: 2.8738744258880615
2023-01-21 08:38:15.775512: step: 400/527, loss: 0.20713792741298676
2023-01-21 08:38:16.903477: step: 404/527, loss: 1.1463865041732788
2023-01-21 08:38:18.002854: step: 408/527, loss: 7.201968669891357
2023-01-21 08:38:19.115021: step: 412/527, loss: 0.15522675216197968
2023-01-21 08:38:20.235979: step: 416/527, loss: 0.13088180124759674
2023-01-21 08:38:21.362869: step: 420/527, loss: 0.33661580085754395
2023-01-21 08:38:22.492279: step: 424/527, loss: 0.4904075562953949
2023-01-21 08:38:23.625608: step: 428/527, loss: 0.3506813049316406
2023-01-21 08:38:24.739703: step: 432/527, loss: 2.774151563644409
2023-01-21 08:38:25.873237: step: 436/527, loss: 0.7703391313552856
2023-01-21 08:38:26.961296: step: 440/527, loss: 0.500515341758728
2023-01-21 08:38:28.075646: step: 444/527, loss: 0.2721587121486664
2023-01-21 08:38:29.207241: step: 448/527, loss: 0.3008522689342499
2023-01-21 08:38:30.314113: step: 452/527, loss: 1.2819223403930664
2023-01-21 08:38:31.411017: step: 456/527, loss: 0.49417054653167725
2023-01-21 08:38:32.553302: step: 460/527, loss: 0.08008956909179688
2023-01-21 08:38:33.642165: step: 464/527, loss: 0.36218443512916565
2023-01-21 08:38:34.744321: step: 468/527, loss: 0.31525689363479614
2023-01-21 08:38:35.864184: step: 472/527, loss: 0.9883606433868408
2023-01-21 08:38:36.987613: step: 476/527, loss: 0.23636884987354279
2023-01-21 08:38:38.106178: step: 480/527, loss: 1.193852186203003
2023-01-21 08:38:39.199337: step: 484/527, loss: 1.4867358207702637
2023-01-21 08:38:40.324299: step: 488/527, loss: 0.5293081998825073
2023-01-21 08:38:41.417263: step: 492/527, loss: 0.2099650800228119
2023-01-21 08:38:42.539767: step: 496/527, loss: 0.6615556478500366
2023-01-21 08:38:43.642726: step: 500/527, loss: 0.23160554468631744
2023-01-21 08:38:44.729164: step: 504/527, loss: 0.366739958524704
2023-01-21 08:38:45.832265: step: 508/527, loss: 1.613642930984497
2023-01-21 08:38:46.933972: step: 512/527, loss: 0.23827491700649261
2023-01-21 08:38:48.035883: step: 516/527, loss: 0.2559822201728821
2023-01-21 08:38:49.147156: step: 520/527, loss: 0.10874795913696289
2023-01-21 08:38:50.296489: step: 524/527, loss: 0.2618190050125122
2023-01-21 08:38:51.405601: step: 528/527, loss: 1.1079684495925903
2023-01-21 08:38:52.532777: step: 532/527, loss: 1.0028618574142456
2023-01-21 08:38:53.636706: step: 536/527, loss: 0.12403498589992523
2023-01-21 08:38:54.762153: step: 540/527, loss: 0.10908882319927216
2023-01-21 08:38:55.863830: step: 544/527, loss: 0.20107778906822205
2023-01-21 08:38:56.985130: step: 548/527, loss: 0.3245050609111786
2023-01-21 08:38:58.105208: step: 552/527, loss: 1.3465570211410522
2023-01-21 08:38:59.192256: step: 556/527, loss: 0.47787612676620483
2023-01-21 08:39:00.322196: step: 560/527, loss: 0.10955943912267685
2023-01-21 08:39:01.458269: step: 564/527, loss: 0.30516186356544495
2023-01-21 08:39:02.573413: step: 568/527, loss: 0.356654554605484
2023-01-21 08:39:03.696600: step: 572/527, loss: 0.4576318860054016
2023-01-21 08:39:04.809047: step: 576/527, loss: 1.3618965148925781
2023-01-21 08:39:05.914215: step: 580/527, loss: 0.8448148965835571
2023-01-21 08:39:07.008172: step: 584/527, loss: 0.19629907608032227
2023-01-21 08:39:08.117703: step: 588/527, loss: 0.2771604657173157
2023-01-21 08:39:09.238109: step: 592/527, loss: 0.9000728130340576
2023-01-21 08:39:10.353104: step: 596/527, loss: 1.2041209936141968
2023-01-21 08:39:11.478049: step: 600/527, loss: 0.525337815284729
2023-01-21 08:39:12.586438: step: 604/527, loss: 1.5502173900604248
2023-01-21 08:39:13.695391: step: 608/527, loss: 0.8513447642326355
2023-01-21 08:39:14.824327: step: 612/527, loss: 3.1387481689453125
2023-01-21 08:39:15.941160: step: 616/527, loss: 0.32541051506996155
2023-01-21 08:39:17.049760: step: 620/527, loss: 0.18478699028491974
2023-01-21 08:39:18.163115: step: 624/527, loss: 0.32146158814430237
2023-01-21 08:39:19.263828: step: 628/527, loss: 0.2090333104133606
2023-01-21 08:39:20.384708: step: 632/527, loss: 0.2918381690979004
2023-01-21 08:39:21.481627: step: 636/527, loss: 1.7401584386825562
2023-01-21 08:39:22.580511: step: 640/527, loss: 0.4173397123813629
2023-01-21 08:39:23.706206: step: 644/527, loss: 1.3212857246398926
2023-01-21 08:39:24.822875: step: 648/527, loss: 0.3152189254760742
2023-01-21 08:39:25.958251: step: 652/527, loss: 0.07118840515613556
2023-01-21 08:39:27.060183: step: 656/527, loss: 0.8042302131652832
2023-01-21 08:39:28.170358: step: 660/527, loss: 0.7121111154556274
2023-01-21 08:39:29.303502: step: 664/527, loss: 0.4650789499282837
2023-01-21 08:39:30.417313: step: 668/527, loss: 0.1480245590209961
2023-01-21 08:39:31.509528: step: 672/527, loss: 0.28826895356178284
2023-01-21 08:39:32.612826: step: 676/527, loss: 0.772638201713562
2023-01-21 08:39:33.707034: step: 680/527, loss: 0.13559189438819885
2023-01-21 08:39:34.827431: step: 684/527, loss: 1.9147491455078125
2023-01-21 08:39:35.950305: step: 688/527, loss: 0.3616272807121277
2023-01-21 08:39:37.095194: step: 692/527, loss: 0.7064182758331299
2023-01-21 08:39:38.214142: step: 696/527, loss: 2.249508857727051
2023-01-21 08:39:39.325746: step: 700/527, loss: 0.8944169282913208
2023-01-21 08:39:40.442859: step: 704/527, loss: 0.8163971304893494
2023-01-21 08:39:41.541360: step: 708/527, loss: 0.18458214402198792
2023-01-21 08:39:42.632546: step: 712/527, loss: 0.15241804718971252
2023-01-21 08:39:43.743888: step: 716/527, loss: 0.19965657591819763
2023-01-21 08:39:44.886161: step: 720/527, loss: 1.0484287738800049
2023-01-21 08:39:46.001476: step: 724/527, loss: 1.5262222290039062
2023-01-21 08:39:47.093994: step: 728/527, loss: 0.10742178559303284
2023-01-21 08:39:48.199393: step: 732/527, loss: 1.6643099784851074
2023-01-21 08:39:49.306435: step: 736/527, loss: 0.2840823233127594
2023-01-21 08:39:50.431056: step: 740/527, loss: 0.30755615234375
2023-01-21 08:39:51.571506: step: 744/527, loss: 0.18293671309947968
2023-01-21 08:39:52.677805: step: 748/527, loss: 1.617738127708435
2023-01-21 08:39:53.821884: step: 752/527, loss: 0.2676420211791992
2023-01-21 08:39:54.924104: step: 756/527, loss: 0.35020875930786133
2023-01-21 08:39:56.044612: step: 760/527, loss: 0.13954877853393555
2023-01-21 08:39:57.194924: step: 764/527, loss: 0.5549153089523315
2023-01-21 08:39:58.355368: step: 768/527, loss: 1.6945971250534058
2023-01-21 08:39:59.473828: step: 772/527, loss: 0.21222396194934845
2023-01-21 08:40:00.603851: step: 776/527, loss: 0.35026460886001587
2023-01-21 08:40:01.732244: step: 780/527, loss: 1.1629325151443481
2023-01-21 08:40:02.853126: step: 784/527, loss: 0.18467026948928833
2023-01-21 08:40:03.963543: step: 788/527, loss: 0.2201048880815506
2023-01-21 08:40:05.106851: step: 792/527, loss: 0.7045050859451294
2023-01-21 08:40:06.220503: step: 796/527, loss: 0.17034488916397095
2023-01-21 08:40:07.325418: step: 800/527, loss: 0.61002117395401
2023-01-21 08:40:08.420327: step: 804/527, loss: 0.57471764087677
2023-01-21 08:40:09.540278: step: 808/527, loss: 0.3278518617153168
2023-01-21 08:40:10.665006: step: 812/527, loss: 0.17721137404441833
2023-01-21 08:40:11.753464: step: 816/527, loss: 1.3893016576766968
2023-01-21 08:40:12.863576: step: 820/527, loss: 0.16541129350662231
2023-01-21 08:40:13.974897: step: 824/527, loss: 1.5428470373153687
2023-01-21 08:40:15.100224: step: 828/527, loss: 0.255013108253479
2023-01-21 08:40:16.212269: step: 832/527, loss: 0.6745571494102478
2023-01-21 08:40:17.306105: step: 836/527, loss: 0.5526704788208008
2023-01-21 08:40:18.409482: step: 840/527, loss: 0.34180185198783875
2023-01-21 08:40:19.514131: step: 844/527, loss: 1.331214189529419
2023-01-21 08:40:20.657412: step: 848/527, loss: 0.17377133667469025
2023-01-21 08:40:21.755374: step: 852/527, loss: 1.3320116996765137
2023-01-21 08:40:22.854935: step: 856/527, loss: 1.1521515846252441
2023-01-21 08:40:23.968721: step: 860/527, loss: 0.8331040143966675
2023-01-21 08:40:25.066850: step: 864/527, loss: 0.290000855922699
2023-01-21 08:40:26.130880: step: 868/527, loss: 0.22493115067481995
2023-01-21 08:40:27.278979: step: 872/527, loss: 0.8376018404960632
2023-01-21 08:40:28.380888: step: 876/527, loss: 0.061017513275146484
2023-01-21 08:40:29.521175: step: 880/527, loss: 0.27890872955322266
2023-01-21 08:40:30.658598: step: 884/527, loss: 0.6691693663597107
2023-01-21 08:40:31.745843: step: 888/527, loss: 0.5690058469772339
2023-01-21 08:40:32.905747: step: 892/527, loss: 0.3967205286026001
2023-01-21 08:40:34.024898: step: 896/527, loss: 0.19949685037136078
2023-01-21 08:40:35.127772: step: 900/527, loss: 0.7724200487136841
2023-01-21 08:40:36.227146: step: 904/527, loss: 0.27631789445877075
2023-01-21 08:40:37.358907: step: 908/527, loss: 0.3881289064884186
2023-01-21 08:40:38.485339: step: 912/527, loss: 0.8545264005661011
2023-01-21 08:40:39.589567: step: 916/527, loss: 0.9950926303863525
2023-01-21 08:40:40.694266: step: 920/527, loss: 0.9637256860733032
2023-01-21 08:40:41.819630: step: 924/527, loss: 0.204999178647995
2023-01-21 08:40:42.951388: step: 928/527, loss: 0.80538409948349
2023-01-21 08:40:44.086314: step: 932/527, loss: 0.16891475021839142
2023-01-21 08:40:45.218660: step: 936/527, loss: 0.49927815794944763
2023-01-21 08:40:46.325338: step: 940/527, loss: 0.5177841186523438
2023-01-21 08:40:47.450417: step: 944/527, loss: 0.10658760368824005
2023-01-21 08:40:48.593570: step: 948/527, loss: 0.8161695599555969
2023-01-21 08:40:49.718249: step: 952/527, loss: 1.055686354637146
2023-01-21 08:40:50.831048: step: 956/527, loss: 0.31629544496536255
2023-01-21 08:40:51.944472: step: 960/527, loss: 1.5012067556381226
2023-01-21 08:40:53.057579: step: 964/527, loss: 0.21044759452342987
2023-01-21 08:40:54.196055: step: 968/527, loss: 2.1814136505126953
2023-01-21 08:40:55.281977: step: 972/527, loss: 0.4474642276763916
2023-01-21 08:40:56.404361: step: 976/527, loss: 0.19904372096061707
2023-01-21 08:40:57.507186: step: 980/527, loss: 1.011764645576477
2023-01-21 08:40:58.603770: step: 984/527, loss: 0.2791154384613037
2023-01-21 08:40:59.697478: step: 988/527, loss: 0.5204005241394043
2023-01-21 08:41:00.846415: step: 992/527, loss: 0.264761745929718
2023-01-21 08:41:01.939579: step: 996/527, loss: 0.23842459917068481
2023-01-21 08:41:03.020723: step: 1000/527, loss: 0.09732875972986221
2023-01-21 08:41:04.124463: step: 1004/527, loss: 0.5582643151283264
2023-01-21 08:41:05.273731: step: 1008/527, loss: 0.24666127562522888
2023-01-21 08:41:06.377903: step: 1012/527, loss: 0.3967497944831848
2023-01-21 08:41:07.541897: step: 1016/527, loss: 0.3107507526874542
2023-01-21 08:41:08.682023: step: 1020/527, loss: 0.2532840669155121
2023-01-21 08:41:09.824330: step: 1024/527, loss: 0.2240493893623352
2023-01-21 08:41:10.929263: step: 1028/527, loss: 2.051966428756714
2023-01-21 08:41:12.048957: step: 1032/527, loss: 0.33020439743995667
2023-01-21 08:41:13.158253: step: 1036/527, loss: 0.287565141916275
2023-01-21 08:41:14.259147: step: 1040/527, loss: 1.6228686571121216
2023-01-21 08:41:15.363536: step: 1044/527, loss: 0.4113144874572754
2023-01-21 08:41:16.492035: step: 1048/527, loss: 0.3564572334289551
2023-01-21 08:41:17.595087: step: 1052/527, loss: 0.2971695065498352
2023-01-21 08:41:18.713100: step: 1056/527, loss: 0.258919894695282
2023-01-21 08:41:19.823322: step: 1060/527, loss: 0.7856267690658569
2023-01-21 08:41:20.954479: step: 1064/527, loss: 0.7656046748161316
2023-01-21 08:41:22.074086: step: 1068/527, loss: 1.8671091794967651
2023-01-21 08:41:23.185272: step: 1072/527, loss: 0.38578784465789795
2023-01-21 08:41:24.325585: step: 1076/527, loss: 0.4394359290599823
2023-01-21 08:41:25.439996: step: 1080/527, loss: 0.18547268211841583
2023-01-21 08:41:26.563541: step: 1084/527, loss: 0.2080131620168686
2023-01-21 08:41:27.666172: step: 1088/527, loss: 1.4777467250823975
2023-01-21 08:41:28.801413: step: 1092/527, loss: 0.21778731048107147
2023-01-21 08:41:29.918376: step: 1096/527, loss: 7.022336006164551
2023-01-21 08:41:31.015714: step: 1100/527, loss: 0.20288430154323578
2023-01-21 08:41:32.119016: step: 1104/527, loss: 0.48537150025367737
2023-01-21 08:41:33.266233: step: 1108/527, loss: 0.196641206741333
2023-01-21 08:41:34.376921: step: 1112/527, loss: 0.24698826670646667
2023-01-21 08:41:35.504541: step: 1116/527, loss: 0.48211997747421265
2023-01-21 08:41:36.635268: step: 1120/527, loss: 0.1801077425479889
2023-01-21 08:41:37.732049: step: 1124/527, loss: 0.10775332152843475
2023-01-21 08:41:38.811816: step: 1128/527, loss: 0.381283700466156
2023-01-21 08:41:39.912424: step: 1132/527, loss: 0.09606008231639862
2023-01-21 08:41:41.022131: step: 1136/527, loss: 0.1600854992866516
2023-01-21 08:41:42.133642: step: 1140/527, loss: 0.3644355833530426
2023-01-21 08:41:43.260359: step: 1144/527, loss: 0.9671630859375
2023-01-21 08:41:44.370924: step: 1148/527, loss: 1.2280585765838623
2023-01-21 08:41:45.478841: step: 1152/527, loss: 1.2624545097351074
2023-01-21 08:41:46.591945: step: 1156/527, loss: 1.665624737739563
2023-01-21 08:41:47.768446: step: 1160/527, loss: 0.26979005336761475
2023-01-21 08:41:48.907524: step: 1164/527, loss: 0.29928797483444214
2023-01-21 08:41:50.040647: step: 1168/527, loss: 0.3420730531215668
2023-01-21 08:41:51.149868: step: 1172/527, loss: 2.4461779594421387
2023-01-21 08:41:52.324266: step: 1176/527, loss: 0.31473052501678467
2023-01-21 08:41:53.423556: step: 1180/527, loss: 0.8288987874984741
2023-01-21 08:41:54.572004: step: 1184/527, loss: 1.4570404291152954
2023-01-21 08:41:55.699700: step: 1188/527, loss: 0.6560747027397156
2023-01-21 08:41:56.792326: step: 1192/527, loss: 0.1135459914803505
2023-01-21 08:41:57.895214: step: 1196/527, loss: 0.6497849822044373
2023-01-21 08:41:59.019347: step: 1200/527, loss: 0.4643517732620239
2023-01-21 08:42:00.132217: step: 1204/527, loss: 0.23417240381240845
2023-01-21 08:42:01.241920: step: 1208/527, loss: 0.11558208614587784
2023-01-21 08:42:02.344408: step: 1212/527, loss: 0.23825012147426605
2023-01-21 08:42:03.441055: step: 1216/527, loss: 0.41295814514160156
2023-01-21 08:42:04.530500: step: 1220/527, loss: 0.20897871255874634
2023-01-21 08:42:05.637033: step: 1224/527, loss: 0.9823281168937683
2023-01-21 08:42:06.739204: step: 1228/527, loss: 0.094927117228508
2023-01-21 08:42:07.876416: step: 1232/527, loss: 1.93697988986969
2023-01-21 08:42:09.046834: step: 1236/527, loss: 0.4876123070716858
2023-01-21 08:42:10.159810: step: 1240/527, loss: 0.15807271003723145
2023-01-21 08:42:11.261218: step: 1244/527, loss: 0.3072170317173004
2023-01-21 08:42:12.347802: step: 1248/527, loss: 0.22928762435913086
2023-01-21 08:42:13.478104: step: 1252/527, loss: 0.24584990739822388
2023-01-21 08:42:14.599140: step: 1256/527, loss: 0.41490334272384644
2023-01-21 08:42:15.714221: step: 1260/527, loss: 0.2963305711746216
2023-01-21 08:42:16.797840: step: 1264/527, loss: 0.10053520649671555
2023-01-21 08:42:17.931238: step: 1268/527, loss: 0.7227115631103516
2023-01-21 08:42:19.028825: step: 1272/527, loss: 1.0870440006256104
2023-01-21 08:42:20.169124: step: 1276/527, loss: 0.9287055730819702
2023-01-21 08:42:21.279374: step: 1280/527, loss: 0.39011627435684204
2023-01-21 08:42:22.406865: step: 1284/527, loss: 1.5694022178649902
2023-01-21 08:42:23.535102: step: 1288/527, loss: 0.2660086750984192
2023-01-21 08:42:24.632383: step: 1292/527, loss: 0.20604261755943298
2023-01-21 08:42:25.741902: step: 1296/527, loss: 1.3050010204315186
2023-01-21 08:42:26.845306: step: 1300/527, loss: 1.0639495849609375
2023-01-21 08:42:27.971775: step: 1304/527, loss: 0.9227062463760376
2023-01-21 08:42:29.114715: step: 1308/527, loss: 0.43316563963890076
2023-01-21 08:42:30.231213: step: 1312/527, loss: 0.47301241755485535
2023-01-21 08:42:31.354151: step: 1316/527, loss: 1.5292580127716064
2023-01-21 08:42:32.473950: step: 1320/527, loss: 1.5884895324707031
2023-01-21 08:42:33.583253: step: 1324/527, loss: 0.6982278227806091
2023-01-21 08:42:34.683306: step: 1328/527, loss: 0.6975236535072327
2023-01-21 08:42:35.800566: step: 1332/527, loss: 0.1938372701406479
2023-01-21 08:42:36.906394: step: 1336/527, loss: 0.06430401653051376
2023-01-21 08:42:38.058807: step: 1340/527, loss: 0.2321162223815918
2023-01-21 08:42:39.177666: step: 1344/527, loss: 2.779973030090332
2023-01-21 08:42:40.335657: step: 1348/527, loss: 0.5846744775772095
2023-01-21 08:42:41.460769: step: 1352/527, loss: 0.14413009583950043
2023-01-21 08:42:42.575119: step: 1356/527, loss: 0.3913077116012573
2023-01-21 08:42:43.682227: step: 1360/527, loss: 0.3192369341850281
2023-01-21 08:42:44.795422: step: 1364/527, loss: 0.6409803628921509
2023-01-21 08:42:45.912714: step: 1368/527, loss: 0.04847393184900284
2023-01-21 08:42:47.037657: step: 1372/527, loss: 0.08990946412086487
2023-01-21 08:42:48.161865: step: 1376/527, loss: 1.3807991743087769
2023-01-21 08:42:49.303981: step: 1380/527, loss: 1.8782823085784912
2023-01-21 08:42:50.427290: step: 1384/527, loss: 0.07490682601928711
2023-01-21 08:42:51.536802: step: 1388/527, loss: 0.18190395832061768
2023-01-21 08:42:52.628755: step: 1392/527, loss: 0.7041284441947937
2023-01-21 08:42:53.734708: step: 1396/527, loss: 0.17479224503040314
2023-01-21 08:42:54.863890: step: 1400/527, loss: 0.9497172832489014
2023-01-21 08:42:55.963573: step: 1404/527, loss: 0.4697246551513672
2023-01-21 08:42:57.049304: step: 1408/527, loss: 0.6437335014343262
2023-01-21 08:42:58.171750: step: 1412/527, loss: 0.5565390586853027
2023-01-21 08:42:59.290938: step: 1416/527, loss: 0.4132024645805359
2023-01-21 08:43:00.441822: step: 1420/527, loss: 0.4425075650215149
2023-01-21 08:43:01.553669: step: 1424/527, loss: 0.20741325616836548
2023-01-21 08:43:02.701362: step: 1428/527, loss: 0.40257158875465393
2023-01-21 08:43:03.815000: step: 1432/527, loss: 0.30314359068870544
2023-01-21 08:43:04.911996: step: 1436/527, loss: 0.2702926993370056
2023-01-21 08:43:06.015478: step: 1440/527, loss: 0.17117223143577576
2023-01-21 08:43:07.138392: step: 1444/527, loss: 1.8015871047973633
2023-01-21 08:43:08.251775: step: 1448/527, loss: 1.0656838417053223
2023-01-21 08:43:09.376465: step: 1452/527, loss: 0.5567184686660767
2023-01-21 08:43:10.519085: step: 1456/527, loss: 0.2849445044994354
2023-01-21 08:43:11.635150: step: 1460/527, loss: 0.26066744327545166
2023-01-21 08:43:12.761695: step: 1464/527, loss: 0.4099099040031433
2023-01-21 08:43:13.872234: step: 1468/527, loss: 0.21477475762367249
2023-01-21 08:43:15.023387: step: 1472/527, loss: 1.5466792583465576
2023-01-21 08:43:16.140998: step: 1476/527, loss: 0.2308189421892166
2023-01-21 08:43:17.259326: step: 1480/527, loss: 0.3078714907169342
2023-01-21 08:43:18.386871: step: 1484/527, loss: 0.2572745680809021
2023-01-21 08:43:19.493835: step: 1488/527, loss: 1.0283666849136353
2023-01-21 08:43:20.592177: step: 1492/527, loss: 0.2598365843296051
2023-01-21 08:43:21.726343: step: 1496/527, loss: 1.9921785593032837
2023-01-21 08:43:22.832449: step: 1500/527, loss: 0.22850003838539124
2023-01-21 08:43:23.943757: step: 1504/527, loss: 0.9154081344604492
2023-01-21 08:43:25.047551: step: 1508/527, loss: 0.2039547860622406
2023-01-21 08:43:26.180893: step: 1512/527, loss: 1.2271795272827148
2023-01-21 08:43:27.277914: step: 1516/527, loss: 0.13336606323719025
2023-01-21 08:43:28.403963: step: 1520/527, loss: 0.4729459285736084
2023-01-21 08:43:29.496821: step: 1524/527, loss: 0.1932951956987381
2023-01-21 08:43:30.636992: step: 1528/527, loss: 1.945611834526062
2023-01-21 08:43:31.745522: step: 1532/527, loss: 1.7316627502441406
2023-01-21 08:43:32.877931: step: 1536/527, loss: 2.6612277030944824
2023-01-21 08:43:33.981676: step: 1540/527, loss: 0.18422779440879822
2023-01-21 08:43:35.079897: step: 1544/527, loss: 0.09578094631433487
2023-01-21 08:43:36.169700: step: 1548/527, loss: 0.46580085158348083
2023-01-21 08:43:37.275819: step: 1552/527, loss: 0.787284255027771
2023-01-21 08:43:38.370027: step: 1556/527, loss: 0.12551012635231018
2023-01-21 08:43:39.498020: step: 1560/527, loss: 1.0120360851287842
2023-01-21 08:43:40.605946: step: 1564/527, loss: 0.13845491409301758
2023-01-21 08:43:41.740062: step: 1568/527, loss: 0.10843883454799652
2023-01-21 08:43:42.850251: step: 1572/527, loss: 3.7761125564575195
2023-01-21 08:43:43.952786: step: 1576/527, loss: 0.20947308838367462
2023-01-21 08:43:45.075056: step: 1580/527, loss: 0.12904532253742218
2023-01-21 08:43:46.236466: step: 1584/527, loss: 2.431868314743042
2023-01-21 08:43:47.347933: step: 1588/527, loss: 0.2086331844329834
2023-01-21 08:43:48.460195: step: 1592/527, loss: 0.3782004415988922
2023-01-21 08:43:49.554274: step: 1596/527, loss: 0.9793174862861633
2023-01-21 08:43:50.668331: step: 1600/527, loss: 0.07106056809425354
2023-01-21 08:43:51.759330: step: 1604/527, loss: 0.6578100919723511
2023-01-21 08:43:52.854967: step: 1608/527, loss: 0.2724035382270813
2023-01-21 08:43:53.960352: step: 1612/527, loss: 0.7021909356117249
2023-01-21 08:43:55.095735: step: 1616/527, loss: 0.3221290707588196
2023-01-21 08:43:56.230387: step: 1620/527, loss: 0.22217103838920593
2023-01-21 08:43:57.314548: step: 1624/527, loss: 6.8734965324401855
2023-01-21 08:43:58.441842: step: 1628/527, loss: 1.6036103963851929
2023-01-21 08:43:59.567144: step: 1632/527, loss: 0.7212901711463928
2023-01-21 08:44:00.691039: step: 1636/527, loss: 0.36735811829566956
2023-01-21 08:44:01.803951: step: 1640/527, loss: 0.33795303106307983
2023-01-21 08:44:02.902123: step: 1644/527, loss: 0.3518761396408081
2023-01-21 08:44:04.016175: step: 1648/527, loss: 1.5078743696212769
2023-01-21 08:44:05.107672: step: 1652/527, loss: 0.1540747582912445
2023-01-21 08:44:06.196087: step: 1656/527, loss: 0.16613636910915375
2023-01-21 08:44:07.277158: step: 1660/527, loss: 0.07519713044166565
2023-01-21 08:44:08.380272: step: 1664/527, loss: 0.7167521715164185
2023-01-21 08:44:09.507861: step: 1668/527, loss: 0.18565607070922852
2023-01-21 08:44:10.621316: step: 1672/527, loss: 0.3964042663574219
2023-01-21 08:44:11.728147: step: 1676/527, loss: 0.05642149597406387
2023-01-21 08:44:12.840972: step: 1680/527, loss: 0.15735659003257751
2023-01-21 08:44:13.936872: step: 1684/527, loss: 1.0823335647583008
2023-01-21 08:44:15.064591: step: 1688/527, loss: 0.9088883399963379
2023-01-21 08:44:16.162247: step: 1692/527, loss: 0.40829670429229736
2023-01-21 08:44:17.270666: step: 1696/527, loss: 0.1751539409160614
2023-01-21 08:44:18.378610: step: 1700/527, loss: 0.26682233810424805
2023-01-21 08:44:19.489662: step: 1704/527, loss: 0.24486637115478516
2023-01-21 08:44:20.593086: step: 1708/527, loss: 0.23768892884254456
2023-01-21 08:44:21.719067: step: 1712/527, loss: 0.21505403518676758
2023-01-21 08:44:22.833390: step: 1716/527, loss: 0.21891427040100098
2023-01-21 08:44:23.972240: step: 1720/527, loss: 0.14747437834739685
2023-01-21 08:44:25.082908: step: 1724/527, loss: 0.20381461083889008
2023-01-21 08:44:26.175387: step: 1728/527, loss: 9.230738639831543
2023-01-21 08:44:27.285269: step: 1732/527, loss: 1.4502363204956055
2023-01-21 08:44:28.407576: step: 1736/527, loss: 1.0933117866516113
2023-01-21 08:44:29.541882: step: 1740/527, loss: 0.761847972869873
2023-01-21 08:44:30.670313: step: 1744/527, loss: 0.5283024311065674
2023-01-21 08:44:31.800115: step: 1748/527, loss: 0.9965077042579651
2023-01-21 08:44:32.927042: step: 1752/527, loss: 1.0281659364700317
2023-01-21 08:44:34.036641: step: 1756/527, loss: 0.1795811653137207
2023-01-21 08:44:35.161068: step: 1760/527, loss: 0.2241329699754715
2023-01-21 08:44:36.294946: step: 1764/527, loss: 0.9095956683158875
2023-01-21 08:44:37.406239: step: 1768/527, loss: 0.28363296389579773
2023-01-21 08:44:38.514495: step: 1772/527, loss: 0.3083708882331848
2023-01-21 08:44:39.636678: step: 1776/527, loss: 0.65910404920578
2023-01-21 08:44:40.750537: step: 1780/527, loss: 0.09091377258300781
2023-01-21 08:44:41.870818: step: 1784/527, loss: 1.3716351985931396
2023-01-21 08:44:42.992413: step: 1788/527, loss: 0.1879730224609375
2023-01-21 08:44:44.094098: step: 1792/527, loss: 1.139765977859497
2023-01-21 08:44:45.184927: step: 1796/527, loss: 2.39186954498291
2023-01-21 08:44:46.294766: step: 1800/527, loss: 1.9225071668624878
2023-01-21 08:44:47.404903: step: 1804/527, loss: 0.6676848530769348
2023-01-21 08:44:48.521857: step: 1808/527, loss: 1.0369346141815186
2023-01-21 08:44:49.675282: step: 1812/527, loss: 0.203252911567688
2023-01-21 08:44:50.785002: step: 1816/527, loss: 0.30394524335861206
2023-01-21 08:44:51.911714: step: 1820/527, loss: 0.748797595500946
2023-01-21 08:44:53.027970: step: 1824/527, loss: 1.1922361850738525
2023-01-21 08:44:54.148392: step: 1828/527, loss: 0.7607652544975281
2023-01-21 08:44:55.239499: step: 1832/527, loss: 0.14432699978351593
2023-01-21 08:44:56.353037: step: 1836/527, loss: 0.2402600347995758
2023-01-21 08:44:57.499034: step: 1840/527, loss: 0.6234188079833984
2023-01-21 08:44:58.619493: step: 1844/527, loss: 1.4823126792907715
2023-01-21 08:44:59.749371: step: 1848/527, loss: 0.20493902266025543
2023-01-21 08:45:00.853537: step: 1852/527, loss: 0.44853726029396057
2023-01-21 08:45:01.993596: step: 1856/527, loss: 2.113954544067383
2023-01-21 08:45:03.103994: step: 1860/527, loss: 0.29280200600624084
2023-01-21 08:45:04.209395: step: 1864/527, loss: 0.4451705813407898
2023-01-21 08:45:05.350392: step: 1868/527, loss: 1.202926516532898
2023-01-21 08:45:06.474123: step: 1872/527, loss: 0.15281638503074646
2023-01-21 08:45:07.577426: step: 1876/527, loss: 0.5236794352531433
2023-01-21 08:45:08.684161: step: 1880/527, loss: 0.7365829348564148
2023-01-21 08:45:09.784288: step: 1884/527, loss: 0.5049744844436646
2023-01-21 08:45:10.908886: step: 1888/527, loss: 0.3946446478366852
2023-01-21 08:45:12.011987: step: 1892/527, loss: 0.1823606938123703
2023-01-21 08:45:13.117731: step: 1896/527, loss: 0.2072470635175705
2023-01-21 08:45:14.259853: step: 1900/527, loss: 0.4567932188510895
2023-01-21 08:45:15.400181: step: 1904/527, loss: 0.6553575396537781
2023-01-21 08:45:16.504416: step: 1908/527, loss: 0.2890511751174927
2023-01-21 08:45:17.597488: step: 1912/527, loss: 0.11705876141786575
2023-01-21 08:45:18.693058: step: 1916/527, loss: 2.7663867473602295
2023-01-21 08:45:19.843639: step: 1920/527, loss: 0.18343916535377502
2023-01-21 08:45:20.964384: step: 1924/527, loss: 0.17275866866111755
2023-01-21 08:45:22.091224: step: 1928/527, loss: 0.16366301476955414
2023-01-21 08:45:23.184607: step: 1932/527, loss: 0.03639573976397514
2023-01-21 08:45:24.287534: step: 1936/527, loss: 0.16920800507068634
2023-01-21 08:45:25.403798: step: 1940/527, loss: 0.317962646484375
2023-01-21 08:45:26.512647: step: 1944/527, loss: 0.7147844433784485
2023-01-21 08:45:27.639006: step: 1948/527, loss: 0.9657028913497925
2023-01-21 08:45:28.762238: step: 1952/527, loss: 1.3206175565719604
2023-01-21 08:45:29.896172: step: 1956/527, loss: 0.8681854009628296
2023-01-21 08:45:31.017214: step: 1960/527, loss: 0.5280121564865112
2023-01-21 08:45:32.140609: step: 1964/527, loss: 0.18877273797988892
2023-01-21 08:45:33.285021: step: 1968/527, loss: 0.2010248750448227
2023-01-21 08:45:34.446887: step: 1972/527, loss: 0.033550359308719635
2023-01-21 08:45:35.524782: step: 1976/527, loss: 0.4398748278617859
2023-01-21 08:45:36.627820: step: 1980/527, loss: 0.18755817413330078
2023-01-21 08:45:37.755444: step: 1984/527, loss: 1.0861334800720215
2023-01-21 08:45:38.840188: step: 1988/527, loss: 0.5326269268989563
2023-01-21 08:45:39.939310: step: 1992/527, loss: 0.2384411245584488
2023-01-21 08:45:41.045788: step: 1996/527, loss: 0.1842707097530365
2023-01-21 08:45:42.155823: step: 2000/527, loss: 0.2859558165073395
2023-01-21 08:45:43.272680: step: 2004/527, loss: 0.043778322637081146
2023-01-21 08:45:44.389668: step: 2008/527, loss: 1.393704891204834
2023-01-21 08:45:45.498774: step: 2012/527, loss: 0.7679749131202698
2023-01-21 08:45:46.601399: step: 2016/527, loss: 1.2648415565490723
2023-01-21 08:45:47.727744: step: 2020/527, loss: 0.46009406447410583
2023-01-21 08:45:48.834665: step: 2024/527, loss: 0.26336565613746643
2023-01-21 08:45:49.939978: step: 2028/527, loss: 0.08755011856555939
2023-01-21 08:45:51.047166: step: 2032/527, loss: 0.22198128700256348
2023-01-21 08:45:52.145502: step: 2036/527, loss: 0.9999496340751648
2023-01-21 08:45:53.247287: step: 2040/527, loss: 0.9257692098617554
2023-01-21 08:45:54.389138: step: 2044/527, loss: 0.2057945281267166
2023-01-21 08:45:55.468959: step: 2048/527, loss: 0.38532111048698425
2023-01-21 08:45:56.575080: step: 2052/527, loss: 0.2405363917350769
2023-01-21 08:45:57.711413: step: 2056/527, loss: 1.3372652530670166
2023-01-21 08:45:58.809614: step: 2060/527, loss: 0.4042108356952667
2023-01-21 08:45:59.913925: step: 2064/527, loss: 6.940031051635742
2023-01-21 08:46:01.043095: step: 2068/527, loss: 1.7936713695526123
2023-01-21 08:46:02.148596: step: 2072/527, loss: 0.18203188478946686
2023-01-21 08:46:03.250056: step: 2076/527, loss: 0.10747461766004562
2023-01-21 08:46:04.360141: step: 2080/527, loss: 0.3761855363845825
2023-01-21 08:46:05.548487: step: 2084/527, loss: 0.17500925064086914
2023-01-21 08:46:06.630689: step: 2088/527, loss: 0.04080705717206001
2023-01-21 08:46:07.762149: step: 2092/527, loss: 0.8781456351280212
2023-01-21 08:46:08.861723: step: 2096/527, loss: 0.2540857195854187
2023-01-21 08:46:09.987205: step: 2100/527, loss: 0.2488357573747635
2023-01-21 08:46:11.129324: step: 2104/527, loss: 0.09587278962135315
2023-01-21 08:46:12.238617: step: 2108/527, loss: 0.09662125259637833
==================================================
Loss: 0.774
--------------------
Dev: {'event': {'p': 0.5560344827586207, 'r': 0.6870838881491345, 'f1': 0.6146515783204287}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Test: {'event': {'p': 0.5831313620940378, 'r': 0.6874285714285714, 'f1': 0.6309992132179386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Chinese: {'event': {'p': 0.559322033898305, 'r': 0.6111111111111112, 'f1': 0.5840707964601771}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Korean: {'event': {'p': 0.631578947368421, 'r': 0.38095238095238093, 'f1': 0.4752475247524752}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Russian: {'event': {'p': 0.5526315789473685, 'r': 0.5833333333333334, 'f1': 0.5675675675675677}, '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': 0.5560344827586207, 'r': 0.6870838881491345, 'f1': 0.6146515783204287}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Eng Test for Chinese: {'event': {'p': 0.5831313620940378, 'r': 0.6874285714285714, 'f1': 0.6309992132179386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Sample Chinese: {'event': {'p': 0.559322033898305, 'r': 0.6111111111111112, 'f1': 0.5840707964601771}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
--------------------
Eng Dev for Korean: {'event': {'p': 0.5560344827586207, 'r': 0.6870838881491345, 'f1': 0.6146515783204287}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Eng Test for Korean: {'event': {'p': 0.5831313620940378, 'r': 0.6874285714285714, 'f1': 0.6309992132179386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Sample Korean: {'event': {'p': 0.631578947368421, 'r': 0.38095238095238093, 'f1': 0.4752475247524752}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
--------------------
Eng Dev for Russian: {'event': {'p': 0.5560344827586207, 'r': 0.6870838881491345, 'f1': 0.6146515783204287}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Eng Test for Russian: {'event': {'p': 0.5831313620940378, 'r': 0.6874285714285714, 'f1': 0.6309992132179386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Sample Russian: {'event': {'p': 0.5526315789473685, 'r': 0.5833333333333334, 'f1': 0.5675675675675677}, '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 08:47:10.297517: step: 4/527, loss: 0.508809506893158
2023-01-21 08:47:11.466842: step: 8/527, loss: 0.08051471412181854
2023-01-21 08:47:12.592007: step: 12/527, loss: 1.1197643280029297
2023-01-21 08:47:13.690799: step: 16/527, loss: 0.7132484912872314
2023-01-21 08:47:14.790285: step: 20/527, loss: 0.6464167237281799
2023-01-21 08:47:15.872675: step: 24/527, loss: 0.09567365050315857
2023-01-21 08:47:16.985714: step: 28/527, loss: 0.3732467591762543
2023-01-21 08:47:18.098928: step: 32/527, loss: 1.211378812789917
2023-01-21 08:47:19.207365: step: 36/527, loss: 0.6955430507659912
2023-01-21 08:47:20.339589: step: 40/527, loss: 0.6174247860908508
2023-01-21 08:47:21.450836: step: 44/527, loss: 1.0429612398147583
2023-01-21 08:47:22.552464: step: 48/527, loss: 0.38752707839012146
2023-01-21 08:47:23.669310: step: 52/527, loss: 0.8731134533882141
2023-01-21 08:47:24.746907: step: 56/527, loss: 7.186557292938232
2023-01-21 08:47:25.818174: step: 60/527, loss: 0.11953286826610565
2023-01-21 08:47:26.943743: step: 64/527, loss: 0.9726343154907227
2023-01-21 08:47:28.068233: step: 68/527, loss: 0.17540550231933594
2023-01-21 08:47:29.186165: step: 72/527, loss: 0.4749521017074585
2023-01-21 08:47:30.282127: step: 76/527, loss: 0.16490773856639862
2023-01-21 08:47:31.415873: step: 80/527, loss: 0.49111253023147583
2023-01-21 08:47:32.477282: step: 84/527, loss: 0.7758349180221558
2023-01-21 08:47:33.608603: step: 88/527, loss: 0.11393478512763977
2023-01-21 08:47:34.731354: step: 92/527, loss: 0.8853338360786438
2023-01-21 08:47:35.819008: step: 96/527, loss: 0.3115384578704834
2023-01-21 08:47:36.935410: step: 100/527, loss: 0.209774449467659
2023-01-21 08:47:38.083757: step: 104/527, loss: 0.3178957998752594
2023-01-21 08:47:39.216351: step: 108/527, loss: 0.5851931571960449
2023-01-21 08:47:40.327199: step: 112/527, loss: 0.2537763714790344
2023-01-21 08:47:41.467660: step: 116/527, loss: 2.578620672225952
2023-01-21 08:47:42.569218: step: 120/527, loss: 0.49000898003578186
2023-01-21 08:47:43.664121: step: 124/527, loss: 2.6648151874542236
2023-01-21 08:47:44.786772: step: 128/527, loss: 0.1741112768650055
2023-01-21 08:47:45.880122: step: 132/527, loss: 0.22406668961048126
2023-01-21 08:47:46.998025: step: 136/527, loss: 0.23551559448242188
2023-01-21 08:47:48.122911: step: 140/527, loss: 0.8233947157859802
2023-01-21 08:47:49.270906: step: 144/527, loss: 0.17566613852977753
2023-01-21 08:47:50.413045: step: 148/527, loss: 0.29378029704093933
2023-01-21 08:47:51.518295: step: 152/527, loss: 0.48547208309173584
2023-01-21 08:47:52.609749: step: 156/527, loss: 0.24952569603919983
2023-01-21 08:47:53.723478: step: 160/527, loss: 0.20701494812965393
2023-01-21 08:47:54.832957: step: 164/527, loss: 0.7263386845588684
2023-01-21 08:47:55.943194: step: 168/527, loss: 1.0169198513031006
2023-01-21 08:47:57.067577: step: 172/527, loss: 0.10271382331848145
2023-01-21 08:47:58.170277: step: 176/527, loss: 0.1205935925245285
2023-01-21 08:47:59.300105: step: 180/527, loss: 0.24087849259376526
2023-01-21 08:48:00.409441: step: 184/527, loss: 0.06403245776891708
2023-01-21 08:48:01.535654: step: 188/527, loss: 2.921477794647217
2023-01-21 08:48:02.648504: step: 192/527, loss: 1.823606252670288
2023-01-21 08:48:03.756221: step: 196/527, loss: 0.04609017074108124
2023-01-21 08:48:04.881309: step: 200/527, loss: 0.4537319540977478
2023-01-21 08:48:05.972942: step: 204/527, loss: 0.6913636922836304
2023-01-21 08:48:07.101893: step: 208/527, loss: 0.19006015360355377
2023-01-21 08:48:08.275870: step: 212/527, loss: 1.1852829456329346
2023-01-21 08:48:09.387025: step: 216/527, loss: 0.43352508544921875
2023-01-21 08:48:10.489119: step: 220/527, loss: 0.20796769857406616
2023-01-21 08:48:11.582488: step: 224/527, loss: 0.6497514247894287
2023-01-21 08:48:12.681868: step: 228/527, loss: 0.285788357257843
2023-01-21 08:48:13.800071: step: 232/527, loss: 0.06697969883680344
2023-01-21 08:48:14.931424: step: 236/527, loss: 0.6509712934494019
2023-01-21 08:48:16.043456: step: 240/527, loss: 0.2568542957305908
2023-01-21 08:48:17.143424: step: 244/527, loss: 0.16602811217308044
2023-01-21 08:48:18.258221: step: 248/527, loss: 0.4349152445793152
2023-01-21 08:48:19.337439: step: 252/527, loss: 0.048705484718084335
2023-01-21 08:48:20.454247: step: 256/527, loss: 0.17351874709129333
2023-01-21 08:48:21.535318: step: 260/527, loss: 0.15449538826942444
2023-01-21 08:48:22.667068: step: 264/527, loss: 0.7872732281684875
2023-01-21 08:48:23.773790: step: 268/527, loss: 0.08169607818126678
2023-01-21 08:48:24.948596: step: 272/527, loss: 0.18699485063552856
2023-01-21 08:48:26.102802: step: 276/527, loss: 0.3095025420188904
2023-01-21 08:48:27.226994: step: 280/527, loss: 2.523277521133423
2023-01-21 08:48:28.353054: step: 284/527, loss: 0.8526454567909241
2023-01-21 08:48:29.484537: step: 288/527, loss: 0.18119679391384125
2023-01-21 08:48:30.568130: step: 292/527, loss: 0.1618306189775467
2023-01-21 08:48:31.710724: step: 296/527, loss: 0.20381546020507812
2023-01-21 08:48:32.798822: step: 300/527, loss: 0.13776426017284393
2023-01-21 08:48:33.930031: step: 304/527, loss: 0.4403475821018219
2023-01-21 08:48:35.089797: step: 308/527, loss: 1.9599671363830566
2023-01-21 08:48:36.218721: step: 312/527, loss: 0.8410625457763672
2023-01-21 08:48:37.350133: step: 316/527, loss: 0.11824560165405273
2023-01-21 08:48:38.480851: step: 320/527, loss: 0.10421428829431534
2023-01-21 08:48:39.607666: step: 324/527, loss: 1.369499921798706
2023-01-21 08:48:40.706598: step: 328/527, loss: 0.7528577446937561
2023-01-21 08:48:41.853945: step: 332/527, loss: 0.7158708572387695
2023-01-21 08:48:42.967679: step: 336/527, loss: 0.6861649751663208
2023-01-21 08:48:44.075086: step: 340/527, loss: 0.2016763687133789
2023-01-21 08:48:45.200871: step: 344/527, loss: 1.667033314704895
2023-01-21 08:48:46.309189: step: 348/527, loss: 0.31558865308761597
2023-01-21 08:48:47.418654: step: 352/527, loss: 2.111006736755371
2023-01-21 08:48:48.534737: step: 356/527, loss: 0.19294653832912445
2023-01-21 08:48:49.615030: step: 360/527, loss: 0.5113725662231445
2023-01-21 08:48:50.756611: step: 364/527, loss: 0.12709054350852966
2023-01-21 08:48:51.883803: step: 368/527, loss: 0.4004729390144348
2023-01-21 08:48:52.983410: step: 372/527, loss: 0.3806811273097992
2023-01-21 08:48:54.096429: step: 376/527, loss: 0.05758824571967125
2023-01-21 08:48:55.233109: step: 380/527, loss: 0.19645214080810547
2023-01-21 08:48:56.371391: step: 384/527, loss: 0.2338651716709137
2023-01-21 08:48:57.512755: step: 388/527, loss: 0.14209969341754913
2023-01-21 08:48:58.635364: step: 392/527, loss: 0.9109208583831787
2023-01-21 08:48:59.734253: step: 396/527, loss: 0.11826296150684357
2023-01-21 08:49:00.848536: step: 400/527, loss: 0.3750064969062805
2023-01-21 08:49:01.959636: step: 404/527, loss: 1.0944753885269165
2023-01-21 08:49:03.051569: step: 408/527, loss: 0.13722772896289825
2023-01-21 08:49:04.159207: step: 412/527, loss: 0.0429803840816021
2023-01-21 08:49:05.296951: step: 416/527, loss: 1.160137414932251
2023-01-21 08:49:06.419558: step: 420/527, loss: 0.11524347960948944
2023-01-21 08:49:07.548683: step: 424/527, loss: 0.3231050670146942
2023-01-21 08:49:08.666072: step: 428/527, loss: 0.9715282320976257
2023-01-21 08:49:09.760596: step: 432/527, loss: 0.449584424495697
2023-01-21 08:49:10.857954: step: 436/527, loss: 0.7992695569992065
2023-01-21 08:49:11.941288: step: 440/527, loss: 0.40562957525253296
2023-01-21 08:49:13.068962: step: 444/527, loss: 0.2608977258205414
2023-01-21 08:49:14.182502: step: 448/527, loss: 0.271353155374527
2023-01-21 08:49:15.313305: step: 452/527, loss: 0.1906244456768036
2023-01-21 08:49:16.444204: step: 456/527, loss: 0.1685314178466797
2023-01-21 08:49:17.562726: step: 460/527, loss: 0.6777358055114746
2023-01-21 08:49:18.678103: step: 464/527, loss: 0.12882956862449646
2023-01-21 08:49:19.808063: step: 468/527, loss: 0.1659383773803711
2023-01-21 08:49:20.923451: step: 472/527, loss: 0.49204179644584656
2023-01-21 08:49:22.033556: step: 476/527, loss: 0.08797939121723175
2023-01-21 08:49:23.148033: step: 480/527, loss: 0.22639484703540802
2023-01-21 08:49:24.258938: step: 484/527, loss: 1.5357418060302734
2023-01-21 08:49:25.372913: step: 488/527, loss: 0.3083256781101227
2023-01-21 08:49:26.466273: step: 492/527, loss: 0.11426444351673126
2023-01-21 08:49:27.569509: step: 496/527, loss: 0.20675823092460632
2023-01-21 08:49:28.656586: step: 500/527, loss: 0.25360843539237976
2023-01-21 08:49:29.764327: step: 504/527, loss: 0.1589408814907074
2023-01-21 08:49:30.862707: step: 508/527, loss: 0.07644256949424744
2023-01-21 08:49:31.978578: step: 512/527, loss: 0.46519842743873596
2023-01-21 08:49:33.133524: step: 516/527, loss: 0.7580075263977051
2023-01-21 08:49:34.250883: step: 520/527, loss: 0.7077133059501648
2023-01-21 08:49:35.377863: step: 524/527, loss: 0.599079966545105
2023-01-21 08:49:36.485966: step: 528/527, loss: 0.5845640897750854
2023-01-21 08:49:37.592796: step: 532/527, loss: 0.4947095811367035
2023-01-21 08:49:38.717116: step: 536/527, loss: 0.41240742802619934
2023-01-21 08:49:39.806701: step: 540/527, loss: 0.149766743183136
2023-01-21 08:49:40.915150: step: 544/527, loss: 0.7612636685371399
2023-01-21 08:49:42.045233: step: 548/527, loss: 1.1650491952896118
2023-01-21 08:49:43.164823: step: 552/527, loss: 0.19652405381202698
2023-01-21 08:49:44.272225: step: 556/527, loss: 0.8278763294219971
2023-01-21 08:49:45.431546: step: 560/527, loss: 0.1411007046699524
2023-01-21 08:49:46.518880: step: 564/527, loss: 0.17499347031116486
2023-01-21 08:49:47.627570: step: 568/527, loss: 0.1611528843641281
2023-01-21 08:49:48.714942: step: 572/527, loss: 0.1049351692199707
2023-01-21 08:49:49.865258: step: 576/527, loss: 0.6969813704490662
2023-01-21 08:49:51.020751: step: 580/527, loss: 0.8075006604194641
2023-01-21 08:49:52.098724: step: 584/527, loss: 0.1691059172153473
2023-01-21 08:49:53.241620: step: 588/527, loss: 0.5877780914306641
2023-01-21 08:49:54.359155: step: 592/527, loss: 1.3590736389160156
2023-01-21 08:49:55.503517: step: 596/527, loss: 0.04550343006849289
2023-01-21 08:49:56.615448: step: 600/527, loss: 0.0836394801735878
2023-01-21 08:49:57.705564: step: 604/527, loss: 0.0873233824968338
2023-01-21 08:49:58.819995: step: 608/527, loss: 0.2814594507217407
2023-01-21 08:49:59.949689: step: 612/527, loss: 0.15087956190109253
2023-01-21 08:50:01.050674: step: 616/527, loss: 0.7622390389442444
2023-01-21 08:50:02.147311: step: 620/527, loss: 0.700497567653656
2023-01-21 08:50:03.239666: step: 624/527, loss: 0.3470827043056488
2023-01-21 08:50:04.381933: step: 628/527, loss: 0.08805885165929794
2023-01-21 08:50:05.467736: step: 632/527, loss: 0.582568347454071
2023-01-21 08:50:06.592884: step: 636/527, loss: 0.20462360978126526
2023-01-21 08:50:07.700511: step: 640/527, loss: 0.07517413794994354
2023-01-21 08:50:08.779777: step: 644/527, loss: 0.13510771095752716
2023-01-21 08:50:09.893261: step: 648/527, loss: 0.14372387528419495
2023-01-21 08:50:11.004817: step: 652/527, loss: 0.08327741920948029
2023-01-21 08:50:12.154718: step: 656/527, loss: 0.7339638471603394
2023-01-21 08:50:13.286831: step: 660/527, loss: 0.22133898735046387
2023-01-21 08:50:14.383783: step: 664/527, loss: 0.04356861114501953
2023-01-21 08:50:15.501171: step: 668/527, loss: 0.11088243126869202
2023-01-21 08:50:16.635212: step: 672/527, loss: 0.43668168783187866
2023-01-21 08:50:17.747835: step: 676/527, loss: 0.6186572909355164
2023-01-21 08:50:18.880127: step: 680/527, loss: 0.12462206184864044
2023-01-21 08:50:20.001827: step: 684/527, loss: 0.7752035856246948
2023-01-21 08:50:21.098532: step: 688/527, loss: 0.13928189873695374
2023-01-21 08:50:22.196157: step: 692/527, loss: 0.913519561290741
2023-01-21 08:50:23.325695: step: 696/527, loss: 0.869016706943512
2023-01-21 08:50:24.411120: step: 700/527, loss: 0.4711214005947113
2023-01-21 08:50:25.521257: step: 704/527, loss: 0.3347194194793701
2023-01-21 08:50:26.638134: step: 708/527, loss: 0.1446743905544281
2023-01-21 08:50:27.750877: step: 712/527, loss: 0.057020239531993866
2023-01-21 08:50:28.853223: step: 716/527, loss: 0.19327889382839203
2023-01-21 08:50:29.974268: step: 720/527, loss: 0.14151154458522797
2023-01-21 08:50:31.110145: step: 724/527, loss: 1.6256033182144165
2023-01-21 08:50:32.233760: step: 728/527, loss: 0.6168744564056396
2023-01-21 08:50:33.383787: step: 732/527, loss: 0.7165476083755493
2023-01-21 08:50:34.479542: step: 736/527, loss: 0.7513967752456665
2023-01-21 08:50:35.569254: step: 740/527, loss: 0.09074673801660538
2023-01-21 08:50:36.682178: step: 744/527, loss: 0.3354890048503876
2023-01-21 08:50:37.798447: step: 748/527, loss: 1.184076189994812
2023-01-21 08:50:38.914606: step: 752/527, loss: 0.028474999591708183
2023-01-21 08:50:40.031586: step: 756/527, loss: 0.21302320063114166
2023-01-21 08:50:41.169020: step: 760/527, loss: 0.05036201700568199
2023-01-21 08:50:42.303704: step: 764/527, loss: 0.18559235334396362
2023-01-21 08:50:43.415395: step: 768/527, loss: 0.32539892196655273
2023-01-21 08:50:44.534199: step: 772/527, loss: 0.7735402584075928
2023-01-21 08:50:45.621207: step: 776/527, loss: 0.05064937844872475
2023-01-21 08:50:46.719684: step: 780/527, loss: 0.40441039204597473
2023-01-21 08:50:47.817991: step: 784/527, loss: 0.17137542366981506
2023-01-21 08:50:48.946362: step: 788/527, loss: 0.3164135813713074
2023-01-21 08:50:50.057284: step: 792/527, loss: 0.7469159960746765
2023-01-21 08:50:51.205759: step: 796/527, loss: 0.2222343385219574
2023-01-21 08:50:52.307523: step: 800/527, loss: 0.13113151490688324
2023-01-21 08:50:53.423410: step: 804/527, loss: 0.2250564694404602
2023-01-21 08:50:54.536195: step: 808/527, loss: 0.19738531112670898
2023-01-21 08:50:55.654555: step: 812/527, loss: 0.5884044170379639
2023-01-21 08:50:56.763439: step: 816/527, loss: 0.15230293571949005
2023-01-21 08:50:57.881353: step: 820/527, loss: 3.5776891708374023
2023-01-21 08:50:59.024002: step: 824/527, loss: 0.09833803027868271
2023-01-21 08:51:00.129496: step: 828/527, loss: 0.14841929078102112
2023-01-21 08:51:01.230989: step: 832/527, loss: 8.15155029296875
2023-01-21 08:51:02.338986: step: 836/527, loss: 0.18123993277549744
2023-01-21 08:51:03.461870: step: 840/527, loss: 0.08970227092504501
2023-01-21 08:51:04.562493: step: 844/527, loss: 0.2677004337310791
2023-01-21 08:51:05.669738: step: 848/527, loss: 0.16472646594047546
2023-01-21 08:51:06.800534: step: 852/527, loss: 0.36883941292762756
2023-01-21 08:51:07.916660: step: 856/527, loss: 0.924812376499176
2023-01-21 08:51:09.042795: step: 860/527, loss: 0.21590128540992737
2023-01-21 08:51:10.166839: step: 864/527, loss: 0.18724651634693146
2023-01-21 08:51:11.279825: step: 868/527, loss: 0.12695884704589844
2023-01-21 08:51:12.416131: step: 872/527, loss: 0.8031549453735352
2023-01-21 08:51:13.530450: step: 876/527, loss: 0.13297969102859497
2023-01-21 08:51:14.637834: step: 880/527, loss: 1.044066309928894
2023-01-21 08:51:15.740543: step: 884/527, loss: 0.18827533721923828
2023-01-21 08:51:16.853260: step: 888/527, loss: 1.5006786584854126
2023-01-21 08:51:17.979960: step: 892/527, loss: 0.287637323141098
2023-01-21 08:51:19.107209: step: 896/527, loss: 0.080182746052742
2023-01-21 08:51:20.235852: step: 900/527, loss: 0.43495798110961914
2023-01-21 08:51:21.395925: step: 904/527, loss: 0.38494405150413513
2023-01-21 08:51:22.477185: step: 908/527, loss: 0.2986149191856384
2023-01-21 08:51:23.579756: step: 912/527, loss: 0.09951372444629669
2023-01-21 08:51:24.682368: step: 916/527, loss: 0.07324619591236115
2023-01-21 08:51:25.839924: step: 920/527, loss: 0.15875226259231567
2023-01-21 08:51:26.968652: step: 924/527, loss: 0.4924079179763794
2023-01-21 08:51:28.070651: step: 928/527, loss: 0.1666729897260666
2023-01-21 08:51:29.176546: step: 932/527, loss: 0.1510227918624878
2023-01-21 08:51:30.312294: step: 936/527, loss: 0.1508079469203949
2023-01-21 08:51:31.423334: step: 940/527, loss: 0.0981116071343422
2023-01-21 08:51:32.534218: step: 944/527, loss: 0.473650723695755
2023-01-21 08:51:33.647919: step: 948/527, loss: 0.4679538607597351
2023-01-21 08:51:34.788288: step: 952/527, loss: 1.9771320819854736
2023-01-21 08:51:35.878063: step: 956/527, loss: 0.5962545275688171
2023-01-21 08:51:36.961765: step: 960/527, loss: 0.08491936326026917
2023-01-21 08:51:38.092103: step: 964/527, loss: 0.4256337285041809
2023-01-21 08:51:39.198288: step: 968/527, loss: 0.974960207939148
2023-01-21 08:51:40.279365: step: 972/527, loss: 0.14189091324806213
2023-01-21 08:51:41.402476: step: 976/527, loss: 0.21337108314037323
2023-01-21 08:51:42.494154: step: 980/527, loss: 0.21850091218948364
2023-01-21 08:51:43.601021: step: 984/527, loss: 0.021103763952851295
2023-01-21 08:51:44.723393: step: 988/527, loss: 0.06029815599322319
2023-01-21 08:51:45.856511: step: 992/527, loss: 0.38073769211769104
2023-01-21 08:51:46.989283: step: 996/527, loss: 0.09785423427820206
2023-01-21 08:51:48.092534: step: 1000/527, loss: 0.1756284236907959
2023-01-21 08:51:49.297672: step: 1004/527, loss: 0.13993793725967407
2023-01-21 08:51:50.407567: step: 1008/527, loss: 0.11952514946460724
2023-01-21 08:51:51.508831: step: 1012/527, loss: 0.2876017689704895
2023-01-21 08:51:52.603969: step: 1016/527, loss: 0.1903502494096756
2023-01-21 08:51:53.725910: step: 1020/527, loss: 0.7590410709381104
2023-01-21 08:51:54.840396: step: 1024/527, loss: 0.21371503174304962
2023-01-21 08:51:55.977031: step: 1028/527, loss: 0.13532547652721405
2023-01-21 08:51:57.087201: step: 1032/527, loss: 0.17921629548072815
2023-01-21 08:51:58.211401: step: 1036/527, loss: 0.10391278564929962
2023-01-21 08:51:59.299352: step: 1040/527, loss: 0.5931336283683777
2023-01-21 08:52:00.415087: step: 1044/527, loss: 0.44791144132614136
2023-01-21 08:52:01.530561: step: 1048/527, loss: 0.2544996738433838
2023-01-21 08:52:02.624107: step: 1052/527, loss: 0.6879607439041138
2023-01-21 08:52:03.726731: step: 1056/527, loss: 1.048452615737915
2023-01-21 08:52:04.859654: step: 1060/527, loss: 0.1454302817583084
2023-01-21 08:52:05.989192: step: 1064/527, loss: 0.16948223114013672
2023-01-21 08:52:07.116854: step: 1068/527, loss: 6.81207799911499
2023-01-21 08:52:08.240017: step: 1072/527, loss: 0.09495621174573898
2023-01-21 08:52:09.335101: step: 1076/527, loss: 0.36988967657089233
2023-01-21 08:52:10.454131: step: 1080/527, loss: 1.2426155805587769
2023-01-21 08:52:11.572343: step: 1084/527, loss: 0.7794598937034607
2023-01-21 08:52:12.726284: step: 1088/527, loss: 0.11607380211353302
2023-01-21 08:52:13.834814: step: 1092/527, loss: 0.5163612365722656
2023-01-21 08:52:14.930515: step: 1096/527, loss: 0.7128528952598572
2023-01-21 08:52:16.070146: step: 1100/527, loss: 0.2510445713996887
2023-01-21 08:52:17.181138: step: 1104/527, loss: 6.766900062561035
2023-01-21 08:52:18.294537: step: 1108/527, loss: 2.0661354064941406
2023-01-21 08:52:19.419852: step: 1112/527, loss: 0.20085038244724274
2023-01-21 08:52:20.541818: step: 1116/527, loss: 0.15568295121192932
2023-01-21 08:52:21.670940: step: 1120/527, loss: 0.5789952278137207
2023-01-21 08:52:22.791500: step: 1124/527, loss: 0.1073971763253212
2023-01-21 08:52:23.891541: step: 1128/527, loss: 0.139032781124115
2023-01-21 08:52:25.050310: step: 1132/527, loss: 0.5800192356109619
2023-01-21 08:52:26.164128: step: 1136/527, loss: 0.07965927571058273
2023-01-21 08:52:27.252255: step: 1140/527, loss: 0.2582598626613617
2023-01-21 08:52:28.359995: step: 1144/527, loss: 0.06700535118579865
2023-01-21 08:52:29.478347: step: 1148/527, loss: 0.041037701070308685
2023-01-21 08:52:30.577066: step: 1152/527, loss: 0.04176654666662216
2023-01-21 08:52:31.686274: step: 1156/527, loss: 0.08579878509044647
2023-01-21 08:52:32.763138: step: 1160/527, loss: 0.14223642647266388
2023-01-21 08:52:33.876534: step: 1164/527, loss: 0.3107718825340271
2023-01-21 08:52:35.011764: step: 1168/527, loss: 1.2067625522613525
2023-01-21 08:52:36.142139: step: 1172/527, loss: 0.264103502035141
2023-01-21 08:52:37.253610: step: 1176/527, loss: 0.12725920975208282
2023-01-21 08:52:38.350606: step: 1180/527, loss: 0.5885448455810547
2023-01-21 08:52:39.484233: step: 1184/527, loss: 0.21629352867603302
2023-01-21 08:52:40.571540: step: 1188/527, loss: 0.08324048668146133
2023-01-21 08:52:41.689336: step: 1192/527, loss: 0.4931715726852417
2023-01-21 08:52:42.776692: step: 1196/527, loss: 0.14759770035743713
2023-01-21 08:52:43.871247: step: 1200/527, loss: 0.40107476711273193
2023-01-21 08:52:44.963538: step: 1204/527, loss: 0.13036805391311646
2023-01-21 08:52:46.084356: step: 1208/527, loss: 2.3990983963012695
2023-01-21 08:52:47.172263: step: 1212/527, loss: 0.20966807007789612
2023-01-21 08:52:48.293642: step: 1216/527, loss: 0.03937487676739693
2023-01-21 08:52:49.393609: step: 1220/527, loss: 0.11510948836803436
2023-01-21 08:52:50.502208: step: 1224/527, loss: 0.08761286735534668
2023-01-21 08:52:51.605034: step: 1228/527, loss: 0.19578705728054047
2023-01-21 08:52:52.727607: step: 1232/527, loss: 0.7736690640449524
2023-01-21 08:52:53.855653: step: 1236/527, loss: 0.4297598898410797
2023-01-21 08:52:54.978013: step: 1240/527, loss: 0.5477831959724426
2023-01-21 08:52:56.065227: step: 1244/527, loss: 0.7490943074226379
2023-01-21 08:52:57.161687: step: 1248/527, loss: 0.11891823261976242
2023-01-21 08:52:58.283327: step: 1252/527, loss: 0.06427083164453506
2023-01-21 08:52:59.417919: step: 1256/527, loss: 0.14949139952659607
2023-01-21 08:53:00.523471: step: 1260/527, loss: 0.7002460360527039
2023-01-21 08:53:01.625492: step: 1264/527, loss: 0.0852302610874176
2023-01-21 08:53:02.732900: step: 1268/527, loss: 2.069143295288086
2023-01-21 08:53:03.830643: step: 1272/527, loss: 0.1096169501543045
2023-01-21 08:53:04.928307: step: 1276/527, loss: 1.391528606414795
2023-01-21 08:53:06.018535: step: 1280/527, loss: 0.09534025192260742
2023-01-21 08:53:07.107556: step: 1284/527, loss: 0.2504711151123047
2023-01-21 08:53:08.214613: step: 1288/527, loss: 2.4685471057891846
2023-01-21 08:53:09.321800: step: 1292/527, loss: 0.7661879658699036
2023-01-21 08:53:10.431338: step: 1296/527, loss: 0.034363701939582825
2023-01-21 08:53:11.546338: step: 1300/527, loss: 0.6315479278564453
2023-01-21 08:53:12.652462: step: 1304/527, loss: 0.3477337062358856
2023-01-21 08:53:13.766790: step: 1308/527, loss: 0.19355988502502441
2023-01-21 08:53:14.839996: step: 1312/527, loss: 2.4004874229431152
2023-01-21 08:53:15.972710: step: 1316/527, loss: 0.25543269515037537
2023-01-21 08:53:17.041187: step: 1320/527, loss: 0.1204003393650055
2023-01-21 08:53:18.163008: step: 1324/527, loss: 0.2172868251800537
2023-01-21 08:53:19.294870: step: 1328/527, loss: 1.6003550291061401
2023-01-21 08:53:20.418380: step: 1332/527, loss: 1.1160590648651123
2023-01-21 08:53:21.530249: step: 1336/527, loss: 0.143450066447258
2023-01-21 08:53:22.659603: step: 1340/527, loss: 0.22737398743629456
2023-01-21 08:53:23.830796: step: 1344/527, loss: 0.09410958737134933
2023-01-21 08:53:24.973844: step: 1348/527, loss: 0.4932255744934082
2023-01-21 08:53:26.065287: step: 1352/527, loss: 0.10659746825695038
2023-01-21 08:53:27.200012: step: 1356/527, loss: 6.355819225311279
2023-01-21 08:53:28.318680: step: 1360/527, loss: 0.15165939927101135
2023-01-21 08:53:29.435881: step: 1364/527, loss: 0.3142017424106598
2023-01-21 08:53:30.566488: step: 1368/527, loss: 0.9376929998397827
2023-01-21 08:53:31.695961: step: 1372/527, loss: 0.995358943939209
2023-01-21 08:53:32.804168: step: 1376/527, loss: 1.19259512424469
2023-01-21 08:53:33.897419: step: 1380/527, loss: 0.08177123218774796
2023-01-21 08:53:35.043711: step: 1384/527, loss: 0.11256589740514755
2023-01-21 08:53:36.135559: step: 1388/527, loss: 0.08342233300209045
2023-01-21 08:53:37.264024: step: 1392/527, loss: 0.12773257493972778
2023-01-21 08:53:38.379011: step: 1396/527, loss: 0.12763291597366333
2023-01-21 08:53:39.488381: step: 1400/527, loss: 0.10100296139717102
2023-01-21 08:53:40.634628: step: 1404/527, loss: 6.4847636222839355
2023-01-21 08:53:41.755857: step: 1408/527, loss: 0.30751386284828186
2023-01-21 08:53:42.850099: step: 1412/527, loss: 0.1453719139099121
2023-01-21 08:53:44.002408: step: 1416/527, loss: 0.7211806178092957
2023-01-21 08:53:45.110977: step: 1420/527, loss: 0.1758289933204651
2023-01-21 08:53:46.204852: step: 1424/527, loss: 0.24334917962551117
2023-01-21 08:53:47.346335: step: 1428/527, loss: 0.11822004616260529
2023-01-21 08:53:48.432432: step: 1432/527, loss: 0.6254950165748596
2023-01-21 08:53:49.538065: step: 1436/527, loss: 0.7044002413749695
2023-01-21 08:53:50.675616: step: 1440/527, loss: 0.26617228984832764
2023-01-21 08:53:51.828955: step: 1444/527, loss: 0.11087627708911896
2023-01-21 08:53:52.957147: step: 1448/527, loss: 0.2744206488132477
2023-01-21 08:53:54.079525: step: 1452/527, loss: 1.1198296546936035
2023-01-21 08:53:55.207303: step: 1456/527, loss: 0.549225926399231
2023-01-21 08:53:56.328006: step: 1460/527, loss: 0.11290311813354492
2023-01-21 08:53:57.444830: step: 1464/527, loss: 0.2970294952392578
2023-01-21 08:53:58.524468: step: 1468/527, loss: 0.18934154510498047
2023-01-21 08:53:59.657303: step: 1472/527, loss: 0.05210290104150772
2023-01-21 08:54:00.770197: step: 1476/527, loss: 0.10870710015296936
2023-01-21 08:54:01.898850: step: 1480/527, loss: 0.30841749906539917
2023-01-21 08:54:02.999495: step: 1484/527, loss: 0.1739308387041092
2023-01-21 08:54:04.104734: step: 1488/527, loss: 0.8741557598114014
2023-01-21 08:54:05.219192: step: 1492/527, loss: 0.4375793933868408
2023-01-21 08:54:06.335538: step: 1496/527, loss: 0.2901790142059326
2023-01-21 08:54:07.445205: step: 1500/527, loss: 1.1248713731765747
2023-01-21 08:54:08.540802: step: 1504/527, loss: 1.3819787502288818
2023-01-21 08:54:09.632912: step: 1508/527, loss: 0.281766802072525
2023-01-21 08:54:10.749350: step: 1512/527, loss: 0.06937137246131897
2023-01-21 08:54:11.910608: step: 1516/527, loss: 0.3909940719604492
2023-01-21 08:54:12.996772: step: 1520/527, loss: 0.2660723924636841
2023-01-21 08:54:14.107834: step: 1524/527, loss: 1.8916209936141968
2023-01-21 08:54:15.225788: step: 1528/527, loss: 0.7295657396316528
2023-01-21 08:54:16.334183: step: 1532/527, loss: 1.5496559143066406
2023-01-21 08:54:17.452519: step: 1536/527, loss: 0.0745358020067215
2023-01-21 08:54:18.557866: step: 1540/527, loss: 0.28694668412208557
2023-01-21 08:54:19.671012: step: 1544/527, loss: 0.7614359855651855
2023-01-21 08:54:20.747974: step: 1548/527, loss: 0.17527265846729279
2023-01-21 08:54:21.879530: step: 1552/527, loss: 0.2575010359287262
2023-01-21 08:54:22.978019: step: 1556/527, loss: 0.08287763595581055
2023-01-21 08:54:24.101611: step: 1560/527, loss: 0.861185610294342
2023-01-21 08:54:25.206347: step: 1564/527, loss: 1.2788594961166382
2023-01-21 08:54:26.329288: step: 1568/527, loss: 0.7603168487548828
2023-01-21 08:54:27.454137: step: 1572/527, loss: 0.5276376008987427
2023-01-21 08:54:28.553871: step: 1576/527, loss: 0.17102785408496857
2023-01-21 08:54:29.702125: step: 1580/527, loss: 0.3335081934928894
2023-01-21 08:54:30.841331: step: 1584/527, loss: 0.09093207865953445
2023-01-21 08:54:31.979463: step: 1588/527, loss: 0.2155836969614029
2023-01-21 08:54:33.090275: step: 1592/527, loss: 0.5206926465034485
2023-01-21 08:54:34.182934: step: 1596/527, loss: 1.7206834554672241
2023-01-21 08:54:35.312144: step: 1600/527, loss: 0.7576001286506653
2023-01-21 08:54:36.427243: step: 1604/527, loss: 0.18618564307689667
2023-01-21 08:54:37.526213: step: 1608/527, loss: 3.081526279449463
2023-01-21 08:54:38.653859: step: 1612/527, loss: 0.19366303086280823
2023-01-21 08:54:39.756981: step: 1616/527, loss: 0.4724538028240204
2023-01-21 08:54:40.874764: step: 1620/527, loss: 0.3779865503311157
2023-01-21 08:54:41.975433: step: 1624/527, loss: 1.2595628499984741
2023-01-21 08:54:43.087691: step: 1628/527, loss: 0.07382285594940186
2023-01-21 08:54:44.201157: step: 1632/527, loss: 0.12430582195520401
2023-01-21 08:54:45.327242: step: 1636/527, loss: 0.24232524633407593
2023-01-21 08:54:46.414105: step: 1640/527, loss: 0.3426475524902344
2023-01-21 08:54:47.550654: step: 1644/527, loss: 0.2910544276237488
2023-01-21 08:54:48.657984: step: 1648/527, loss: 0.3327888548374176
2023-01-21 08:54:49.778320: step: 1652/527, loss: 0.11315460503101349
2023-01-21 08:54:50.908465: step: 1656/527, loss: 0.24544401466846466
2023-01-21 08:54:52.023017: step: 1660/527, loss: 0.10720377415418625
2023-01-21 08:54:53.130901: step: 1664/527, loss: 0.953731894493103
2023-01-21 08:54:54.251273: step: 1668/527, loss: 1.661987066268921
2023-01-21 08:54:55.353464: step: 1672/527, loss: 0.3664335310459137
2023-01-21 08:54:56.506299: step: 1676/527, loss: 0.47709712386131287
2023-01-21 08:54:57.678182: step: 1680/527, loss: 0.01964864693582058
2023-01-21 08:54:58.784076: step: 1684/527, loss: 0.09250341355800629
2023-01-21 08:54:59.915140: step: 1688/527, loss: 0.04162721708416939
2023-01-21 08:55:01.018486: step: 1692/527, loss: 0.21190519630908966
2023-01-21 08:55:02.123348: step: 1696/527, loss: 1.5235652923583984
2023-01-21 08:55:03.251806: step: 1700/527, loss: 0.0755079984664917
2023-01-21 08:55:04.372309: step: 1704/527, loss: 0.39317160844802856
2023-01-21 08:55:05.498083: step: 1708/527, loss: 0.18369001150131226
2023-01-21 08:55:06.602498: step: 1712/527, loss: 0.17624130845069885
2023-01-21 08:55:07.728348: step: 1716/527, loss: 0.24089355766773224
2023-01-21 08:55:08.877875: step: 1720/527, loss: 0.1675625890493393
2023-01-21 08:55:09.991209: step: 1724/527, loss: 1.108909010887146
2023-01-21 08:55:11.090085: step: 1728/527, loss: 0.08311252295970917
2023-01-21 08:55:12.218588: step: 1732/527, loss: 0.8410710096359253
2023-01-21 08:55:13.315195: step: 1736/527, loss: 0.45169803500175476
2023-01-21 08:55:14.466376: step: 1740/527, loss: 0.3264934718608856
2023-01-21 08:55:15.568057: step: 1744/527, loss: 0.05677900090813637
2023-01-21 08:55:16.665486: step: 1748/527, loss: 0.3532448410987854
2023-01-21 08:55:17.780139: step: 1752/527, loss: 0.13659420609474182
2023-01-21 08:55:18.921892: step: 1756/527, loss: 0.13202890753746033
2023-01-21 08:55:20.065183: step: 1760/527, loss: 0.2342444658279419
2023-01-21 08:55:21.202601: step: 1764/527, loss: 0.11645559966564178
2023-01-21 08:55:22.312420: step: 1768/527, loss: 0.0801548957824707
2023-01-21 08:55:23.436330: step: 1772/527, loss: 0.16870741546154022
2023-01-21 08:55:24.556270: step: 1776/527, loss: 0.3093426823616028
2023-01-21 08:55:25.654087: step: 1780/527, loss: 0.4707120954990387
2023-01-21 08:55:26.764879: step: 1784/527, loss: 0.1456715613603592
2023-01-21 08:55:27.906540: step: 1788/527, loss: 0.7960838079452515
2023-01-21 08:55:29.045737: step: 1792/527, loss: 0.6862504482269287
2023-01-21 08:55:30.142353: step: 1796/527, loss: 0.7375777959823608
2023-01-21 08:55:31.219972: step: 1800/527, loss: 0.03491075336933136
2023-01-21 08:55:32.328372: step: 1804/527, loss: 0.03256731107831001
2023-01-21 08:55:33.449195: step: 1808/527, loss: 0.6108474135398865
2023-01-21 08:55:34.538229: step: 1812/527, loss: 0.11911864578723907
2023-01-21 08:55:35.633168: step: 1816/527, loss: 0.049486566334962845
2023-01-21 08:55:36.748450: step: 1820/527, loss: 0.1568562537431717
2023-01-21 08:55:37.858706: step: 1824/527, loss: 0.19569067656993866
2023-01-21 08:55:38.980002: step: 1828/527, loss: 0.13616882264614105
2023-01-21 08:55:40.116918: step: 1832/527, loss: 0.337595134973526
2023-01-21 08:55:41.220035: step: 1836/527, loss: 0.4788685739040375
2023-01-21 08:55:42.338122: step: 1840/527, loss: 0.48046931624412537
2023-01-21 08:55:43.426349: step: 1844/527, loss: 0.0921226516366005
2023-01-21 08:55:44.535824: step: 1848/527, loss: 0.2914409637451172
2023-01-21 08:55:45.647644: step: 1852/527, loss: 0.38010454177856445
2023-01-21 08:55:46.753946: step: 1856/527, loss: 0.6869344115257263
2023-01-21 08:55:47.894786: step: 1860/527, loss: 0.2785852551460266
2023-01-21 08:55:49.009200: step: 1864/527, loss: 0.10527582466602325
2023-01-21 08:55:50.077824: step: 1868/527, loss: 0.5615676045417786
2023-01-21 08:55:51.188498: step: 1872/527, loss: 0.22778892517089844
2023-01-21 08:55:52.339171: step: 1876/527, loss: 0.7978864908218384
2023-01-21 08:55:53.461553: step: 1880/527, loss: 0.2650766372680664
2023-01-21 08:55:54.572055: step: 1884/527, loss: 0.37344110012054443
2023-01-21 08:55:55.659592: step: 1888/527, loss: 0.07013054937124252
2023-01-21 08:55:56.757754: step: 1892/527, loss: 0.35323771834373474
2023-01-21 08:55:57.841854: step: 1896/527, loss: 0.07066917419433594
2023-01-21 08:55:58.962790: step: 1900/527, loss: 0.3653683066368103
2023-01-21 08:56:00.079022: step: 1904/527, loss: 0.6925190091133118
2023-01-21 08:56:01.200283: step: 1908/527, loss: 0.49712473154067993
2023-01-21 08:56:02.281306: step: 1912/527, loss: 0.22238358855247498
2023-01-21 08:56:03.386184: step: 1916/527, loss: 0.225530207157135
2023-01-21 08:56:04.486973: step: 1920/527, loss: 0.046700358390808105
2023-01-21 08:56:05.618888: step: 1924/527, loss: 0.1559397280216217
2023-01-21 08:56:06.752943: step: 1928/527, loss: 0.53546541929245
2023-01-21 08:56:07.867889: step: 1932/527, loss: 1.5677920579910278
2023-01-21 08:56:08.979970: step: 1936/527, loss: 0.20982904732227325
2023-01-21 08:56:10.114985: step: 1940/527, loss: 0.18935346603393555
2023-01-21 08:56:11.241852: step: 1944/527, loss: 0.9220942854881287
2023-01-21 08:56:12.383404: step: 1948/527, loss: 0.1754719614982605
2023-01-21 08:56:13.504028: step: 1952/527, loss: 0.16006864607334137
2023-01-21 08:56:14.633938: step: 1956/527, loss: 0.16700410842895508
2023-01-21 08:56:15.769659: step: 1960/527, loss: 1.0917448997497559
2023-01-21 08:56:16.905961: step: 1964/527, loss: 0.0658954605460167
2023-01-21 08:56:18.045195: step: 1968/527, loss: 1.2004772424697876
2023-01-21 08:56:19.184679: step: 1972/527, loss: 1.4372750520706177
2023-01-21 08:56:20.292891: step: 1976/527, loss: 0.9758621454238892
2023-01-21 08:56:21.380037: step: 1980/527, loss: 0.03387107700109482
2023-01-21 08:56:22.504269: step: 1984/527, loss: 3.595634937286377
2023-01-21 08:56:23.628902: step: 1988/527, loss: 0.9712764620780945
2023-01-21 08:56:24.736911: step: 1992/527, loss: 0.42004719376564026
2023-01-21 08:56:25.842434: step: 1996/527, loss: 0.11016550660133362
2023-01-21 08:56:26.980302: step: 2000/527, loss: 0.23884287476539612
2023-01-21 08:56:28.100947: step: 2004/527, loss: 1.3455641269683838
2023-01-21 08:56:29.233115: step: 2008/527, loss: 0.2710852026939392
2023-01-21 08:56:30.362542: step: 2012/527, loss: 6.032918930053711
2023-01-21 08:56:31.470603: step: 2016/527, loss: 0.17980480194091797
2023-01-21 08:56:32.602290: step: 2020/527, loss: 0.7440292239189148
2023-01-21 08:56:33.705893: step: 2024/527, loss: 0.14048504829406738
2023-01-21 08:56:34.823657: step: 2028/527, loss: 1.5821094512939453
2023-01-21 08:56:35.914126: step: 2032/527, loss: 1.537947416305542
2023-01-21 08:56:37.017987: step: 2036/527, loss: 0.11527901142835617
2023-01-21 08:56:38.132085: step: 2040/527, loss: 1.6896952390670776
2023-01-21 08:56:39.246619: step: 2044/527, loss: 0.2969578802585602
2023-01-21 08:56:40.365623: step: 2048/527, loss: 0.41979628801345825
2023-01-21 08:56:41.464065: step: 2052/527, loss: 0.8157370686531067
2023-01-21 08:56:42.587144: step: 2056/527, loss: 0.7642280459403992
2023-01-21 08:56:43.682282: step: 2060/527, loss: 0.4924098551273346
2023-01-21 08:56:44.800864: step: 2064/527, loss: 0.2628178298473358
2023-01-21 08:56:45.916600: step: 2068/527, loss: 0.12557315826416016
2023-01-21 08:56:47.008975: step: 2072/527, loss: 0.1145901009440422
2023-01-21 08:56:48.141715: step: 2076/527, loss: 0.3862895965576172
2023-01-21 08:56:49.263306: step: 2080/527, loss: 2.2748115062713623
2023-01-21 08:56:50.395489: step: 2084/527, loss: 0.8024913668632507
2023-01-21 08:56:51.497145: step: 2088/527, loss: 0.4189472198486328
2023-01-21 08:56:52.601751: step: 2092/527, loss: 0.2964984178543091
2023-01-21 08:56:53.744796: step: 2096/527, loss: 0.9785525798797607
2023-01-21 08:56:54.829493: step: 2100/527, loss: 0.6473172307014465
2023-01-21 08:56:55.923543: step: 2104/527, loss: 0.2812657952308655
2023-01-21 08:56:57.042382: step: 2108/527, loss: 0.06238508224487305
==================================================
Loss: 0.562
--------------------
Dev: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Test: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Chinese: {'event': {'p': 0.6, 'r': 0.8333333333333334, 'f1': 0.6976744186046512}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Russian: {'event': {'p': 0.45161290322580644, 'r': 0.3888888888888889, 'f1': 0.417910447761194}, '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': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Chinese: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Chinese: {'event': {'p': 0.6, 'r': 0.8333333333333334, 'f1': 0.6976744186046512}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Korean: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Korean: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Russian: {'event': {'p': 0.5560344827586207, 'r': 0.6870838881491345, 'f1': 0.6146515783204287}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Eng Test for Russian: {'event': {'p': 0.5831313620940378, 'r': 0.6874285714285714, 'f1': 0.6309992132179386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1}
Sample Russian: {'event': {'p': 0.5526315789473685, 'r': 0.5833333333333334, 'f1': 0.5675675675675677}, '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 08:57:53.228338: step: 4/527, loss: 0.29523342847824097
2023-01-21 08:57:54.323980: step: 8/527, loss: 0.24055567383766174
2023-01-21 08:57:55.444859: step: 12/527, loss: 0.30832982063293457
2023-01-21 08:57:56.586621: step: 16/527, loss: 0.311708927154541
2023-01-21 08:57:57.698849: step: 20/527, loss: 0.44299525022506714
2023-01-21 08:57:58.791845: step: 24/527, loss: 0.08590765297412872
2023-01-21 08:57:59.916831: step: 28/527, loss: 0.23331400752067566
2023-01-21 08:58:01.036169: step: 32/527, loss: 0.5599204301834106
2023-01-21 08:58:02.124742: step: 36/527, loss: 0.06011553108692169
2023-01-21 08:58:03.279940: step: 40/527, loss: 0.28026342391967773
2023-01-21 08:58:04.426451: step: 44/527, loss: 0.09607162326574326
2023-01-21 08:58:05.542942: step: 48/527, loss: 0.06594524532556534
2023-01-21 08:58:06.642796: step: 52/527, loss: 0.19180460274219513
2023-01-21 08:58:07.749307: step: 56/527, loss: 0.7360360622406006
2023-01-21 08:58:08.865837: step: 60/527, loss: 0.09769344329833984
2023-01-21 08:58:09.980966: step: 64/527, loss: 0.1952771246433258
2023-01-21 08:58:11.099011: step: 68/527, loss: 0.28659144043922424
2023-01-21 08:58:12.234961: step: 72/527, loss: 0.22840900719165802
2023-01-21 08:58:13.344528: step: 76/527, loss: 0.037955570966005325
2023-01-21 08:58:14.460681: step: 80/527, loss: 0.7833908796310425
2023-01-21 08:58:15.595653: step: 84/527, loss: 0.12808018922805786
2023-01-21 08:58:16.704164: step: 88/527, loss: 0.12343281507492065
2023-01-21 08:58:17.823211: step: 92/527, loss: 0.7955486178398132
2023-01-21 08:58:18.944596: step: 96/527, loss: 0.12088718265295029
2023-01-21 08:58:20.049460: step: 100/527, loss: 0.31135714054107666
2023-01-21 08:58:21.210392: step: 104/527, loss: 0.21968041360378265
2023-01-21 08:58:22.316146: step: 108/527, loss: 0.16879454255104065
2023-01-21 08:58:23.420604: step: 112/527, loss: 0.7538687586784363
2023-01-21 08:58:24.548585: step: 116/527, loss: 0.24970674514770508
2023-01-21 08:58:25.659776: step: 120/527, loss: 0.12112836539745331
2023-01-21 08:58:26.777051: step: 124/527, loss: 0.16490879654884338
2023-01-21 08:58:27.879045: step: 128/527, loss: 0.23180007934570312
2023-01-21 08:58:28.979472: step: 132/527, loss: 0.21754293143749237
2023-01-21 08:58:30.097108: step: 136/527, loss: 0.1884976327419281
2023-01-21 08:58:31.218987: step: 140/527, loss: 0.3693264126777649
2023-01-21 08:58:32.331264: step: 144/527, loss: 0.2415207326412201
2023-01-21 08:58:33.438611: step: 148/527, loss: 0.7395491003990173
2023-01-21 08:58:34.576804: step: 152/527, loss: 0.20872434973716736
2023-01-21 08:58:35.662872: step: 156/527, loss: 0.03007795847952366
2023-01-21 08:58:36.766108: step: 160/527, loss: 0.072475865483284
2023-01-21 08:58:37.896457: step: 164/527, loss: 0.11533994972705841
2023-01-21 08:58:38.981689: step: 168/527, loss: 0.11754465103149414
2023-01-21 08:58:40.119200: step: 172/527, loss: 0.09689471870660782
2023-01-21 08:58:41.207183: step: 176/527, loss: 0.21636077761650085
2023-01-21 08:58:42.357478: step: 180/527, loss: 0.1440087854862213
2023-01-21 08:58:43.472897: step: 184/527, loss: 0.3535681366920471
2023-01-21 08:58:44.635826: step: 188/527, loss: 0.1323660910129547
2023-01-21 08:58:45.741097: step: 192/527, loss: 0.13328666985034943
2023-01-21 08:58:46.860933: step: 196/527, loss: 0.4558882713317871
2023-01-21 08:58:47.985468: step: 200/527, loss: 0.24560528993606567
2023-01-21 08:58:49.106180: step: 204/527, loss: 0.3231223225593567
2023-01-21 08:58:50.219789: step: 208/527, loss: 0.9360366463661194
2023-01-21 08:58:51.343486: step: 212/527, loss: 0.7182988524436951
2023-01-21 08:58:52.496718: step: 216/527, loss: 1.230064868927002
2023-01-21 08:58:53.635112: step: 220/527, loss: 0.14730653166770935
2023-01-21 08:58:54.708508: step: 224/527, loss: 0.2925889194011688
2023-01-21 08:58:55.815262: step: 228/527, loss: 0.3287316560745239
2023-01-21 08:58:56.949794: step: 232/527, loss: 0.16178306937217712
2023-01-21 08:58:58.069970: step: 236/527, loss: 0.35347530245780945
2023-01-21 08:58:59.216417: step: 240/527, loss: 0.05575313791632652
2023-01-21 08:59:00.309792: step: 244/527, loss: 0.23297587037086487
2023-01-21 08:59:01.432979: step: 248/527, loss: 0.09823159873485565
2023-01-21 08:59:02.558609: step: 252/527, loss: 0.8506431579589844
2023-01-21 08:59:03.655510: step: 256/527, loss: 0.8213114738464355
2023-01-21 08:59:04.773698: step: 260/527, loss: 0.2058243751525879
2023-01-21 08:59:05.875947: step: 264/527, loss: 0.055585384368896484
2023-01-21 08:59:06.973012: step: 268/527, loss: 2.41886043548584
2023-01-21 08:59:08.097315: step: 272/527, loss: 0.16041022539138794
2023-01-21 08:59:09.223831: step: 276/527, loss: 0.18471652269363403
2023-01-21 08:59:10.321796: step: 280/527, loss: 0.5527294278144836
2023-01-21 08:59:11.455125: step: 284/527, loss: 0.09649810940027237
2023-01-21 08:59:12.567258: step: 288/527, loss: 0.20128989219665527
2023-01-21 08:59:13.689477: step: 292/527, loss: 1.1198081970214844
2023-01-21 08:59:14.803961: step: 296/527, loss: 1.2830760478973389
2023-01-21 08:59:15.889533: step: 300/527, loss: 0.3204280138015747
2023-01-21 08:59:17.013773: step: 304/527, loss: 0.16644175350666046
2023-01-21 08:59:18.129633: step: 308/527, loss: 0.8513381481170654
2023-01-21 08:59:19.231517: step: 312/527, loss: 0.30492687225341797
2023-01-21 08:59:20.324281: step: 316/527, loss: 0.1365855187177658
2023-01-21 08:59:21.435377: step: 320/527, loss: 0.47999104857444763
2023-01-21 08:59:22.562098: step: 324/527, loss: 0.09173393249511719
2023-01-21 08:59:23.719521: step: 328/527, loss: 0.07055292278528214
2023-01-21 08:59:24.830199: step: 332/527, loss: 0.20234471559524536
2023-01-21 08:59:25.979175: step: 336/527, loss: 0.17705698311328888
2023-01-21 08:59:27.077075: step: 340/527, loss: 0.09585509449243546
2023-01-21 08:59:28.161508: step: 344/527, loss: 0.4019201397895813
2023-01-21 08:59:29.258582: step: 348/527, loss: 0.038372136652469635
2023-01-21 08:59:30.354575: step: 352/527, loss: 0.04337067902088165
2023-01-21 08:59:31.475067: step: 356/527, loss: 0.9830796718597412
2023-01-21 08:59:32.601608: step: 360/527, loss: 0.6228225827217102
2023-01-21 08:59:33.745611: step: 364/527, loss: 0.16078686714172363
2023-01-21 08:59:34.852053: step: 368/527, loss: 0.19050121307373047
2023-01-21 08:59:35.933810: step: 372/527, loss: 0.08886180073022842
2023-01-21 08:59:37.047324: step: 376/527, loss: 6.747163772583008
2023-01-21 08:59:38.190761: step: 380/527, loss: 0.7045419216156006
2023-01-21 08:59:39.272473: step: 384/527, loss: 0.10610118508338928
2023-01-21 08:59:40.362518: step: 388/527, loss: 0.23508206009864807
2023-01-21 08:59:41.474339: step: 392/527, loss: 0.2466500848531723
2023-01-21 08:59:42.598091: step: 396/527, loss: 0.05903768539428711
2023-01-21 08:59:43.712778: step: 400/527, loss: 0.12119865417480469
2023-01-21 08:59:44.800510: step: 404/527, loss: 0.39661216735839844
2023-01-21 08:59:45.913546: step: 408/527, loss: 0.29129326343536377
2023-01-21 08:59:47.035368: step: 412/527, loss: 1.1994248628616333
2023-01-21 08:59:48.149488: step: 416/527, loss: 0.08095411956310272
2023-01-21 08:59:49.292197: step: 420/527, loss: 0.531620979309082
2023-01-21 08:59:50.436010: step: 424/527, loss: 0.0606144443154335
2023-01-21 08:59:51.538692: step: 428/527, loss: 0.14632532000541687
2023-01-21 08:59:52.649737: step: 432/527, loss: 0.15788884460926056
2023-01-21 08:59:53.791098: step: 436/527, loss: 0.17314834892749786
2023-01-21 08:59:54.955513: step: 440/527, loss: 0.2165517807006836
2023-01-21 08:59:56.076663: step: 444/527, loss: 0.3165510296821594
2023-01-21 08:59:57.217045: step: 448/527, loss: 0.23098278045654297
2023-01-21 08:59:58.323143: step: 452/527, loss: 0.800719141960144
2023-01-21 08:59:59.456022: step: 456/527, loss: 0.5590640306472778
2023-01-21 09:00:00.603344: step: 460/527, loss: 0.13624043762683868
2023-01-21 09:00:01.733484: step: 464/527, loss: 0.25043773651123047
2023-01-21 09:00:02.847982: step: 468/527, loss: 0.23645392060279846
2023-01-21 09:00:03.965143: step: 472/527, loss: 0.4604577124118805
2023-01-21 09:00:05.091474: step: 476/527, loss: 0.083251953125
2023-01-21 09:00:06.208602: step: 480/527, loss: 0.7196712493896484
2023-01-21 09:00:07.309154: step: 484/527, loss: 0.07926731556653976
2023-01-21 09:00:08.404841: step: 488/527, loss: 0.15770551562309265
2023-01-21 09:00:09.556845: step: 492/527, loss: 0.4499916136264801
2023-01-21 09:00:10.671764: step: 496/527, loss: 6.143533706665039
2023-01-21 09:00:11.776032: step: 500/527, loss: 0.14929285645484924
2023-01-21 09:00:12.918811: step: 504/527, loss: 0.3751159906387329
2023-01-21 09:00:14.029045: step: 508/527, loss: 0.19858865439891815
2023-01-21 09:00:15.117791: step: 512/527, loss: 0.13074883818626404
2023-01-21 09:00:16.221487: step: 516/527, loss: 0.18720608949661255
2023-01-21 09:00:17.345948: step: 520/527, loss: 0.08892136067152023
2023-01-21 09:00:18.444393: step: 524/527, loss: 0.09759245067834854
2023-01-21 09:00:19.552763: step: 528/527, loss: 0.1223384365439415
2023-01-21 09:00:20.672930: step: 532/527, loss: 0.7035112977027893
2023-01-21 09:00:21.839330: step: 536/527, loss: 0.8711899518966675
2023-01-21 09:00:22.989559: step: 540/527, loss: 0.2516135275363922
2023-01-21 09:00:24.133719: step: 544/527, loss: 0.06325416266918182
2023-01-21 09:00:25.265389: step: 548/527, loss: 0.26714763045310974
2023-01-21 09:00:26.390644: step: 552/527, loss: 0.5036737322807312
2023-01-21 09:00:27.504501: step: 556/527, loss: 0.5743014216423035
2023-01-21 09:00:28.629860: step: 560/527, loss: 0.3543386459350586
2023-01-21 09:00:29.727855: step: 564/527, loss: 0.10639238357543945
2023-01-21 09:00:30.819405: step: 568/527, loss: 0.7924562692642212
2023-01-21 09:00:31.965559: step: 572/527, loss: 0.3038768768310547
2023-01-21 09:00:33.103662: step: 576/527, loss: 0.034146498888731
2023-01-21 09:00:34.202944: step: 580/527, loss: 1.4056048393249512
2023-01-21 09:00:35.318368: step: 584/527, loss: 1.034292459487915
2023-01-21 09:00:36.460783: step: 588/527, loss: 0.15713483095169067
2023-01-21 09:00:37.560056: step: 592/527, loss: 0.085680291056633
2023-01-21 09:00:38.670786: step: 596/527, loss: 0.6299510598182678
2023-01-21 09:00:39.801642: step: 600/527, loss: 0.1197303831577301
2023-01-21 09:00:40.894322: step: 604/527, loss: 0.6524761319160461
2023-01-21 09:00:41.992925: step: 608/527, loss: 0.4268460273742676
2023-01-21 09:00:43.100283: step: 612/527, loss: 0.2464122772216797
2023-01-21 09:00:44.209328: step: 616/527, loss: 0.11744027584791183
2023-01-21 09:00:45.295130: step: 620/527, loss: 0.9237141609191895
2023-01-21 09:00:46.409820: step: 624/527, loss: 0.08192434906959534
2023-01-21 09:00:47.521620: step: 628/527, loss: 0.14642827212810516
2023-01-21 09:00:48.656758: step: 632/527, loss: 0.13101109862327576
2023-01-21 09:00:49.781591: step: 636/527, loss: 0.3748794496059418
2023-01-21 09:00:50.922205: step: 640/527, loss: 1.3404457569122314
2023-01-21 09:00:52.031978: step: 644/527, loss: 0.20574846863746643
2023-01-21 09:00:53.160483: step: 648/527, loss: 0.5100795030593872
2023-01-21 09:00:54.264283: step: 652/527, loss: 1.888468861579895
2023-01-21 09:00:55.352571: step: 656/527, loss: 0.12523789703845978
2023-01-21 09:00:56.461784: step: 660/527, loss: 0.1742839813232422
2023-01-21 09:00:57.562379: step: 664/527, loss: 0.08021669834852219
2023-01-21 09:00:58.676915: step: 668/527, loss: 0.12213466316461563
2023-01-21 09:00:59.780849: step: 672/527, loss: 0.009029579348862171
2023-01-21 09:01:00.920178: step: 676/527, loss: 0.9232625365257263
2023-01-21 09:01:02.044010: step: 680/527, loss: 0.6095352172851562
2023-01-21 09:01:03.150313: step: 684/527, loss: 0.10016965866088867
2023-01-21 09:01:04.252052: step: 688/527, loss: 0.4289236068725586
2023-01-21 09:01:05.367372: step: 692/527, loss: 0.7149953246116638
2023-01-21 09:01:06.481246: step: 696/527, loss: 0.05515170097351074
2023-01-21 09:01:07.602313: step: 700/527, loss: 0.18643493950366974
2023-01-21 09:01:08.694053: step: 704/527, loss: 0.6594498157501221
2023-01-21 09:01:09.767685: step: 708/527, loss: 0.0828583687543869
2023-01-21 09:01:10.918365: step: 712/527, loss: 0.09805183857679367
2023-01-21 09:01:12.028918: step: 716/527, loss: 0.6854934692382812
2023-01-21 09:01:13.142553: step: 720/527, loss: 0.07641954720020294
2023-01-21 09:01:14.226684: step: 724/527, loss: 0.10716669261455536
2023-01-21 09:01:15.340574: step: 728/527, loss: 1.2215745449066162
2023-01-21 09:01:16.446318: step: 732/527, loss: 0.4225223660469055
2023-01-21 09:01:17.574709: step: 736/527, loss: 0.23105916380882263
2023-01-21 09:01:18.689513: step: 740/527, loss: 0.1052466481924057
2023-01-21 09:01:19.816553: step: 744/527, loss: 0.3296758532524109
2023-01-21 09:01:20.957342: step: 748/527, loss: 0.7393607497215271
2023-01-21 09:01:22.050708: step: 752/527, loss: 0.10656967759132385
2023-01-21 09:01:23.157364: step: 756/527, loss: 0.1621202826499939
2023-01-21 09:01:24.289029: step: 760/527, loss: 0.13115721940994263
2023-01-21 09:01:25.403820: step: 764/527, loss: 0.34071341156959534
2023-01-21 09:01:26.510563: step: 768/527, loss: 0.12973089516162872
2023-01-21 09:01:27.622411: step: 772/527, loss: 0.16026464104652405
2023-01-21 09:01:28.718126: step: 776/527, loss: 0.7662369608879089
2023-01-21 09:01:29.880214: step: 780/527, loss: 0.12174063175916672
2023-01-21 09:01:31.001275: step: 784/527, loss: 0.07869014889001846
2023-01-21 09:01:32.109326: step: 788/527, loss: 0.7192503213882446
2023-01-21 09:01:33.239442: step: 792/527, loss: 0.2659946382045746
2023-01-21 09:01:34.373058: step: 796/527, loss: 0.16486701369285583
2023-01-21 09:01:35.494706: step: 800/527, loss: 0.33068400621414185
2023-01-21 09:01:36.641412: step: 804/527, loss: 0.15061970055103302
2023-01-21 09:01:37.764308: step: 808/527, loss: 0.2260732799768448
2023-01-21 09:01:38.868930: step: 812/527, loss: 0.1231963187456131
2023-01-21 09:01:39.971798: step: 816/527, loss: 0.056687548756599426
2023-01-21 09:01:41.086614: step: 820/527, loss: 0.3718331456184387
2023-01-21 09:01:42.205022: step: 824/527, loss: 0.6463532447814941
2023-01-21 09:01:43.315807: step: 828/527, loss: 0.11277799308300018
2023-01-21 09:01:44.419350: step: 832/527, loss: 0.25255244970321655
2023-01-21 09:01:45.510323: step: 836/527, loss: 0.2428019940853119
2023-01-21 09:01:46.631724: step: 840/527, loss: 0.6127815246582031
2023-01-21 09:01:47.760899: step: 844/527, loss: 0.805828332901001
2023-01-21 09:01:48.890445: step: 848/527, loss: 0.10094957798719406
2023-01-21 09:01:50.039797: step: 852/527, loss: 0.13343198597431183
2023-01-21 09:01:51.200152: step: 856/527, loss: 1.540419578552246
2023-01-21 09:01:52.322598: step: 860/527, loss: 0.07517042756080627
2023-01-21 09:01:53.495327: step: 864/527, loss: 0.26802197098731995
2023-01-21 09:01:54.582242: step: 868/527, loss: 0.12183710932731628
2023-01-21 09:01:55.676831: step: 872/527, loss: 0.09167900681495667
2023-01-21 09:01:56.811043: step: 876/527, loss: 0.26292967796325684
2023-01-21 09:01:57.920261: step: 880/527, loss: 0.08565893024206161
2023-01-21 09:01:59.038389: step: 884/527, loss: 0.191510871052742
2023-01-21 09:02:00.140290: step: 888/527, loss: 0.21745187044143677
2023-01-21 09:02:01.248866: step: 892/527, loss: 0.578854501247406
2023-01-21 09:02:02.373636: step: 896/527, loss: 0.11720170825719833
2023-01-21 09:02:03.520069: step: 900/527, loss: 0.7298274636268616
2023-01-21 09:02:04.614419: step: 904/527, loss: 0.5172414183616638
2023-01-21 09:02:05.732536: step: 908/527, loss: 0.17615079879760742
2023-01-21 09:02:06.842942: step: 912/527, loss: 0.6806604862213135
2023-01-21 09:02:07.945773: step: 916/527, loss: 0.28286781907081604
2023-01-21 09:02:09.048891: step: 920/527, loss: 0.7868010401725769
2023-01-21 09:02:10.134556: step: 924/527, loss: 0.3200637102127075
2023-01-21 09:02:11.263365: step: 928/527, loss: 0.2528351843357086
2023-01-21 09:02:12.403826: step: 932/527, loss: 0.09375253319740295
2023-01-21 09:02:13.547620: step: 936/527, loss: 1.6057020425796509
2023-01-21 09:02:14.633394: step: 940/527, loss: 0.1982731819152832
2023-01-21 09:02:15.725934: step: 944/527, loss: 0.34476107358932495
2023-01-21 09:02:16.872254: step: 948/527, loss: 0.30681857466697693
2023-01-21 09:02:17.989913: step: 952/527, loss: 0.395455539226532
2023-01-21 09:02:19.119662: step: 956/527, loss: 0.13369092345237732
2023-01-21 09:02:20.219424: step: 960/527, loss: 0.059206388890743256
2023-01-21 09:02:21.323694: step: 964/527, loss: 0.7263615727424622
2023-01-21 09:02:22.446901: step: 968/527, loss: 0.05437064170837402
2023-01-21 09:02:23.552964: step: 972/527, loss: 0.5808128118515015
2023-01-21 09:02:24.648536: step: 976/527, loss: 0.2736876606941223
2023-01-21 09:02:25.750533: step: 980/527, loss: 0.09005871415138245
2023-01-21 09:02:26.857635: step: 984/527, loss: 0.8632826209068298
2023-01-21 09:02:27.985661: step: 988/527, loss: 1.0252221822738647
2023-01-21 09:02:29.087331: step: 992/527, loss: 0.17467424273490906
2023-01-21 09:02:30.197361: step: 996/527, loss: 0.05576139688491821
2023-01-21 09:02:31.311641: step: 1000/527, loss: 0.2604485750198364
2023-01-21 09:02:32.428428: step: 1004/527, loss: 1.2448192834854126
2023-01-21 09:02:33.534589: step: 1008/527, loss: 0.7224701642990112
2023-01-21 09:02:34.689010: step: 1012/527, loss: 0.05256319046020508
2023-01-21 09:02:35.778999: step: 1016/527, loss: 0.40065836906433105
2023-01-21 09:02:36.913731: step: 1020/527, loss: 0.8348200917243958
2023-01-21 09:02:38.034858: step: 1024/527, loss: 0.3438807725906372
2023-01-21 09:02:39.116441: step: 1028/527, loss: 0.7351655960083008
2023-01-21 09:02:40.265423: step: 1032/527, loss: 6.24078369140625
2023-01-21 09:02:41.375898: step: 1036/527, loss: 1.2371599674224854
2023-01-21 09:02:42.463785: step: 1040/527, loss: 0.20953096449375153
2023-01-21 09:02:43.564626: step: 1044/527, loss: 0.03357229381799698
2023-01-21 09:02:44.704670: step: 1048/527, loss: 0.2951011657714844
2023-01-21 09:02:45.843243: step: 1052/527, loss: 0.12666818499565125
2023-01-21 09:02:46.952239: step: 1056/527, loss: 0.7795705199241638
2023-01-21 09:02:48.068882: step: 1060/527, loss: 0.26206228137016296
2023-01-21 09:02:49.184543: step: 1064/527, loss: 0.34461164474487305
2023-01-21 09:02:50.322394: step: 1068/527, loss: 0.13036489486694336
2023-01-21 09:02:51.428740: step: 1072/527, loss: 0.07648658752441406
2023-01-21 09:02:52.523662: step: 1076/527, loss: 0.08582830429077148
2023-01-21 09:02:53.685711: step: 1080/527, loss: 0.35444217920303345
2023-01-21 09:02:54.847099: step: 1084/527, loss: 0.14558488130569458
2023-01-21 09:02:55.972332: step: 1088/527, loss: 0.8985475301742554
2023-01-21 09:02:57.095483: step: 1092/527, loss: 0.23319482803344727
2023-01-21 09:02:58.253732: step: 1096/527, loss: 0.09455452114343643
2023-01-21 09:02:59.340630: step: 1100/527, loss: 0.7094722986221313
2023-01-21 09:03:00.435013: step: 1104/527, loss: 0.17710772156715393
2023-01-21 09:03:01.573098: step: 1108/527, loss: 0.13092699646949768
2023-01-21 09:03:02.684262: step: 1112/527, loss: 0.5701183676719666
2023-01-21 09:03:03.791323: step: 1116/527, loss: 0.2542566955089569
2023-01-21 09:03:04.913439: step: 1120/527, loss: 6.269968509674072
2023-01-21 09:03:06.017155: step: 1124/527, loss: 0.10636921226978302
2023-01-21 09:03:07.122434: step: 1128/527, loss: 0.6528869271278381
2023-01-21 09:03:08.249735: step: 1132/527, loss: 0.23219867050647736
2023-01-21 09:03:09.392713: step: 1136/527, loss: 0.19824066758155823
2023-01-21 09:03:10.508049: step: 1140/527, loss: 0.24031782150268555
2023-01-21 09:03:11.602589: step: 1144/527, loss: 0.06825733184814453
2023-01-21 09:03:12.711280: step: 1148/527, loss: 0.6413553357124329
2023-01-21 09:03:13.839135: step: 1152/527, loss: 0.07785339653491974
2023-01-21 09:03:14.939337: step: 1156/527, loss: 0.3656767010688782
2023-01-21 09:03:16.058290: step: 1160/527, loss: 0.0957103744149208
2023-01-21 09:03:17.182700: step: 1164/527, loss: 0.04970159754157066
2023-01-21 09:03:18.311094: step: 1168/527, loss: 0.20569229125976562
2023-01-21 09:03:19.411015: step: 1172/527, loss: 0.08003134280443192
2023-01-21 09:03:20.526471: step: 1176/527, loss: 0.07006935775279999
2023-01-21 09:03:21.655134: step: 1180/527, loss: 0.20141926407814026
2023-01-21 09:03:22.770899: step: 1184/527, loss: 0.1382962167263031
2023-01-21 09:03:23.868162: step: 1188/527, loss: 0.023036815226078033
2023-01-21 09:03:24.964395: step: 1192/527, loss: 0.12034359574317932
2023-01-21 09:03:26.073834: step: 1196/527, loss: 0.7818715572357178
2023-01-21 09:03:27.184732: step: 1200/527, loss: 0.10831747204065323
2023-01-21 09:03:28.325469: step: 1204/527, loss: 0.11144642531871796
2023-01-21 09:03:29.463048: step: 1208/527, loss: 0.14342650771141052
2023-01-21 09:03:30.579857: step: 1212/527, loss: 6.843223571777344
2023-01-21 09:03:31.697926: step: 1216/527, loss: 0.058576010167598724
2023-01-21 09:03:32.815799: step: 1220/527, loss: 0.4922071695327759
2023-01-21 09:03:33.901580: step: 1224/527, loss: 0.11455860733985901
2023-01-21 09:03:35.003191: step: 1228/527, loss: 0.4115406274795532
2023-01-21 09:03:36.135677: step: 1232/527, loss: 2.1886959075927734
2023-01-21 09:03:37.238129: step: 1236/527, loss: 1.1516563892364502
2023-01-21 09:03:38.320352: step: 1240/527, loss: 0.22236990928649902
2023-01-21 09:03:39.427169: step: 1244/527, loss: 0.18879419565200806
2023-01-21 09:03:40.547416: step: 1248/527, loss: 0.6179380416870117
2023-01-21 09:03:41.663665: step: 1252/527, loss: 0.7509297728538513
2023-01-21 09:03:42.796669: step: 1256/527, loss: 0.08869829773902893
2023-01-21 09:03:43.931309: step: 1260/527, loss: 6.988436698913574
2023-01-21 09:03:45.030326: step: 1264/527, loss: 0.06967001408338547
2023-01-21 09:03:46.151930: step: 1268/527, loss: 0.19424334168434143
2023-01-21 09:03:47.255790: step: 1272/527, loss: 0.15633173286914825
2023-01-21 09:03:48.337475: step: 1276/527, loss: 0.07303605228662491
2023-01-21 09:03:49.427494: step: 1280/527, loss: 0.17554417252540588
2023-01-21 09:03:50.552584: step: 1284/527, loss: 0.44688090682029724
2023-01-21 09:03:51.662834: step: 1288/527, loss: 0.09542637318372726
2023-01-21 09:03:52.777915: step: 1292/527, loss: 0.1234808936715126
2023-01-21 09:03:53.866469: step: 1296/527, loss: 0.20519113540649414
2023-01-21 09:03:54.993028: step: 1300/527, loss: 0.16131410002708435
2023-01-21 09:03:56.093843: step: 1304/527, loss: 0.16981634497642517
2023-01-21 09:03:57.249515: step: 1308/527, loss: 0.27565404772758484
2023-01-21 09:03:58.404026: step: 1312/527, loss: 0.15511161088943481
2023-01-21 09:03:59.511003: step: 1316/527, loss: 0.3203316926956177
2023-01-21 09:04:00.606799: step: 1320/527, loss: 0.3474600613117218
2023-01-21 09:04:01.701329: step: 1324/527, loss: 0.20824985206127167
2023-01-21 09:04:02.834727: step: 1328/527, loss: 0.021573161706328392
2023-01-21 09:04:03.919092: step: 1332/527, loss: 0.14113540947437286
2023-01-21 09:04:05.001012: step: 1336/527, loss: 0.15560971200466156
2023-01-21 09:04:06.114532: step: 1340/527, loss: 0.7649518251419067
2023-01-21 09:04:07.219810: step: 1344/527, loss: 0.05924525111913681
2023-01-21 09:04:08.346931: step: 1348/527, loss: 0.2367693930864334
2023-01-21 09:04:09.456680: step: 1352/527, loss: 0.16110849380493164
2023-01-21 09:04:10.559795: step: 1356/527, loss: 0.7183638215065002
2023-01-21 09:04:11.672789: step: 1360/527, loss: 1.31695556640625
2023-01-21 09:04:12.762900: step: 1364/527, loss: 0.3987087309360504
2023-01-21 09:04:13.874255: step: 1368/527, loss: 0.16208620369434357
2023-01-21 09:04:15.001530: step: 1372/527, loss: 0.8116208910942078
2023-01-21 09:04:16.145599: step: 1376/527, loss: 0.17888890206813812
2023-01-21 09:04:17.258470: step: 1380/527, loss: 0.3769391179084778
2023-01-21 09:04:18.363240: step: 1384/527, loss: 0.0617159865796566
2023-01-21 09:04:19.478739: step: 1388/527, loss: 0.7561659812927246
2023-01-21 09:04:20.594857: step: 1392/527, loss: 1.257905125617981
2023-01-21 09:04:21.716434: step: 1396/527, loss: 0.6649717092514038
2023-01-21 09:04:22.806603: step: 1400/527, loss: 0.09816465526819229
2023-01-21 09:04:23.962651: step: 1404/527, loss: 0.21050962805747986
2023-01-21 09:04:25.081824: step: 1408/527, loss: 0.22895203530788422
2023-01-21 09:04:26.197788: step: 1412/527, loss: 0.16873522102832794
2023-01-21 09:04:27.310377: step: 1416/527, loss: 0.1545039266347885
2023-01-21 09:04:28.453590: step: 1420/527, loss: 0.3784274458885193
2023-01-21 09:04:29.561206: step: 1424/527, loss: 0.1585555374622345
2023-01-21 09:04:30.662573: step: 1428/527, loss: 0.12912093102931976
2023-01-21 09:04:31.776043: step: 1432/527, loss: 0.1897331327199936
2023-01-21 09:04:32.920612: step: 1436/527, loss: 0.1998147964477539
2023-01-21 09:04:34.059515: step: 1440/527, loss: 1.7084550857543945
2023-01-21 09:04:35.166519: step: 1444/527, loss: 1.9001847505569458
2023-01-21 09:04:36.259756: step: 1448/527, loss: 0.5831559300422668
2023-01-21 09:04:37.364120: step: 1452/527, loss: 0.16930732131004333
2023-01-21 09:04:38.507706: step: 1456/527, loss: 0.11557817459106445
2023-01-21 09:04:39.635258: step: 1460/527, loss: 0.036350131034851074
2023-01-21 09:04:40.747492: step: 1464/527, loss: 0.06388416886329651
2023-01-21 09:04:41.851734: step: 1468/527, loss: 0.3397369086742401
2023-01-21 09:04:42.975381: step: 1472/527, loss: 1.7740471363067627
2023-01-21 09:04:44.083794: step: 1476/527, loss: 0.15655669569969177
2023-01-21 09:04:45.193784: step: 1480/527, loss: 0.4824514389038086
2023-01-21 09:04:46.364445: step: 1484/527, loss: 0.05881619453430176
2023-01-21 09:04:47.506917: step: 1488/527, loss: 0.22328053414821625
2023-01-21 09:04:48.591180: step: 1492/527, loss: 0.10837321728467941
2023-01-21 09:04:49.687476: step: 1496/527, loss: 0.13701844215393066
2023-01-21 09:04:50.758141: step: 1500/527, loss: 0.13396620750427246
2023-01-21 09:04:51.873255: step: 1504/527, loss: 0.4210335612297058
2023-01-21 09:04:53.008078: step: 1508/527, loss: 0.2519668638706207
2023-01-21 09:04:54.142709: step: 1512/527, loss: 0.37600135803222656
2023-01-21 09:04:55.255206: step: 1516/527, loss: 0.048534706234931946
2023-01-21 09:04:56.360778: step: 1520/527, loss: 0.0802985206246376
2023-01-21 09:04:57.446330: step: 1524/527, loss: 0.3425601124763489
2023-01-21 09:04:58.559691: step: 1528/527, loss: 0.2651395797729492
2023-01-21 09:04:59.665217: step: 1532/527, loss: 0.3466445505619049
2023-01-21 09:05:00.761137: step: 1536/527, loss: 0.7408485412597656
2023-01-21 09:05:01.887544: step: 1540/527, loss: 0.15022669732570648
2023-01-21 09:05:02.990798: step: 1544/527, loss: 0.20141810178756714
2023-01-21 09:05:04.096946: step: 1548/527, loss: 0.6393409967422485
2023-01-21 09:05:05.203776: step: 1552/527, loss: 0.1797538697719574
2023-01-21 09:05:06.341402: step: 1556/527, loss: 0.2448950856924057
2023-01-21 09:05:07.428950: step: 1560/527, loss: 0.6922838091850281
2023-01-21 09:05:08.541410: step: 1564/527, loss: 1.196817398071289
2023-01-21 09:05:09.639778: step: 1568/527, loss: 1.151154637336731
2023-01-21 09:05:10.752556: step: 1572/527, loss: 0.20969842374324799
2023-01-21 09:05:11.841214: step: 1576/527, loss: 0.13624997437000275
2023-01-21 09:05:12.961047: step: 1580/527, loss: 0.26747751235961914
2023-01-21 09:05:14.090395: step: 1584/527, loss: 0.4422232210636139
2023-01-21 09:05:15.171715: step: 1588/527, loss: 0.21091842651367188
2023-01-21 09:05:16.302945: step: 1592/527, loss: 0.11456441879272461
2023-01-21 09:05:17.412746: step: 1596/527, loss: 0.23135709762573242
2023-01-21 09:05:18.523004: step: 1600/527, loss: 0.07480859756469727
2023-01-21 09:05:19.647121: step: 1604/527, loss: 0.5939818024635315
2023-01-21 09:05:20.777213: step: 1608/527, loss: 0.08332765102386475
2023-01-21 09:05:21.876563: step: 1612/527, loss: 0.729077935218811
2023-01-21 09:05:22.985012: step: 1616/527, loss: 0.3042232394218445
2023-01-21 09:05:24.091137: step: 1620/527, loss: 0.17822542786598206
2023-01-21 09:05:25.182999: step: 1624/527, loss: 0.07442331314086914
2023-01-21 09:05:26.303284: step: 1628/527, loss: 0.2780519127845764
2023-01-21 09:05:27.418055: step: 1632/527, loss: 0.1747821867465973
2023-01-21 09:05:28.550187: step: 1636/527, loss: 0.08693154156208038
2023-01-21 09:05:29.698003: step: 1640/527, loss: 0.1609300673007965
2023-01-21 09:05:30.798093: step: 1644/527, loss: 0.15866021811962128
2023-01-21 09:05:31.922159: step: 1648/527, loss: 0.13303261995315552
2023-01-21 09:05:33.013801: step: 1652/527, loss: 0.1694045066833496
2023-01-21 09:05:34.150886: step: 1656/527, loss: 0.7706464529037476
2023-01-21 09:05:35.270253: step: 1660/527, loss: 0.7723090052604675
2023-01-21 09:05:36.382348: step: 1664/527, loss: 0.14559917151927948
2023-01-21 09:05:37.494720: step: 1668/527, loss: 1.1145273447036743
2023-01-21 09:05:38.630871: step: 1672/527, loss: 0.8525142669677734
2023-01-21 09:05:39.748741: step: 1676/527, loss: 0.15494604408740997
2023-01-21 09:05:40.865983: step: 1680/527, loss: 0.20186564326286316
2023-01-21 09:05:41.969719: step: 1684/527, loss: 0.034897495061159134
2023-01-21 09:05:43.085234: step: 1688/527, loss: 0.7521612644195557
2023-01-21 09:05:44.180255: step: 1692/527, loss: 1.5351735353469849
2023-01-21 09:05:45.279028: step: 1696/527, loss: 0.1274941861629486
2023-01-21 09:05:46.415064: step: 1700/527, loss: 0.3413775861263275
2023-01-21 09:05:47.515856: step: 1704/527, loss: 0.24382968246936798
2023-01-21 09:05:48.644979: step: 1708/527, loss: 0.7066400051116943
2023-01-21 09:05:49.792915: step: 1712/527, loss: 0.1125233918428421
2023-01-21 09:05:50.920439: step: 1716/527, loss: 0.6647725701332092
2023-01-21 09:05:52.040358: step: 1720/527, loss: 0.35052013397216797
2023-01-21 09:05:53.166638: step: 1724/527, loss: 0.9961753487586975
2023-01-21 09:05:54.263697: step: 1728/527, loss: 0.639833390712738
2023-01-21 09:05:55.368734: step: 1732/527, loss: 0.7160775661468506
2023-01-21 09:05:56.501422: step: 1736/527, loss: 0.4022991359233856
2023-01-21 09:05:57.610752: step: 1740/527, loss: 0.07399425655603409
2023-01-21 09:05:58.733865: step: 1744/527, loss: 0.15703420341014862
2023-01-21 09:05:59.853474: step: 1748/527, loss: 0.23415395617485046
2023-01-21 09:06:00.957133: step: 1752/527, loss: 0.7301901578903198
2023-01-21 09:06:02.087969: step: 1756/527, loss: 0.09709759056568146
2023-01-21 09:06:03.206622: step: 1760/527, loss: 1.3265905380249023
2023-01-21 09:06:04.322973: step: 1764/527, loss: 0.12964686751365662
2023-01-21 09:06:05.454656: step: 1768/527, loss: 0.056077100336551666
2023-01-21 09:06:06.553360: step: 1772/527, loss: 0.05269885063171387
2023-01-21 09:06:07.674425: step: 1776/527, loss: 0.05644979700446129
2023-01-21 09:06:08.788631: step: 1780/527, loss: 1.9284570217132568
2023-01-21 09:06:09.914034: step: 1784/527, loss: 0.45022791624069214
2023-01-21 09:06:11.009424: step: 1788/527, loss: 0.41081351041793823
2023-01-21 09:06:12.132937: step: 1792/527, loss: 0.6521373987197876
2023-01-21 09:06:13.263234: step: 1796/527, loss: 1.4138745069503784
2023-01-21 09:06:14.377000: step: 1800/527, loss: 0.2815832197666168
2023-01-21 09:06:15.526938: step: 1804/527, loss: 0.18337178230285645
2023-01-21 09:06:16.652919: step: 1808/527, loss: 0.19891025125980377
2023-01-21 09:06:17.779845: step: 1812/527, loss: 1.1391477584838867
2023-01-21 09:06:18.922187: step: 1816/527, loss: 0.0474453940987587
2023-01-21 09:06:20.032449: step: 1820/527, loss: 0.6373782753944397
2023-01-21 09:06:21.139422: step: 1824/527, loss: 0.7021394371986389
2023-01-21 09:06:22.223993: step: 1828/527, loss: 0.03942599147558212
2023-01-21 09:06:23.305550: step: 1832/527, loss: 0.9061006307601929
2023-01-21 09:06:24.420483: step: 1836/527, loss: 0.6624301075935364
2023-01-21 09:06:25.532850: step: 1840/527, loss: 0.22355647385120392
2023-01-21 09:06:26.644299: step: 1844/527, loss: 0.05921673774719238
2023-01-21 09:06:27.757897: step: 1848/527, loss: 0.8650643229484558
2023-01-21 09:06:28.870990: step: 1852/527, loss: 0.8127278089523315
2023-01-21 09:06:29.963670: step: 1856/527, loss: 0.3343259394168854
2023-01-21 09:06:31.098199: step: 1860/527, loss: 0.08522005379199982
2023-01-21 09:06:32.201124: step: 1864/527, loss: 0.15020093321800232
2023-01-21 09:06:33.305554: step: 1868/527, loss: 0.09190855175256729
2023-01-21 09:06:34.444431: step: 1872/527, loss: 0.3572346270084381
2023-01-21 09:06:35.577321: step: 1876/527, loss: 0.12306909263134003
2023-01-21 09:06:36.717650: step: 1880/527, loss: 0.7557978630065918
2023-01-21 09:06:37.847339: step: 1884/527, loss: 0.14303618669509888
2023-01-21 09:06:38.969760: step: 1888/527, loss: 0.21837492287158966
2023-01-21 09:06:40.097904: step: 1892/527, loss: 0.6044619083404541
2023-01-21 09:06:41.185132: step: 1896/527, loss: 0.0841360092163086
2023-01-21 09:06:42.306487: step: 1900/527, loss: 0.16436046361923218
2023-01-21 09:06:43.425257: step: 1904/527, loss: 0.08870497345924377
2023-01-21 09:06:44.550932: step: 1908/527, loss: 0.13764634728431702
2023-01-21 09:06:45.659811: step: 1912/527, loss: 0.09183311462402344
2023-01-21 09:06:46.780567: step: 1916/527, loss: 0.23198318481445312
2023-01-21 09:06:47.887515: step: 1920/527, loss: 0.06905551254749298
2023-01-21 09:06:49.006857: step: 1924/527, loss: 0.5190895199775696
2023-01-21 09:06:50.170214: step: 1928/527, loss: 0.14674490690231323
2023-01-21 09:06:51.274649: step: 1932/527, loss: 0.2626269459724426
2023-01-21 09:06:52.388108: step: 1936/527, loss: 0.23745432496070862
2023-01-21 09:06:53.506778: step: 1940/527, loss: 0.7011728882789612
2023-01-21 09:06:54.624964: step: 1944/527, loss: 0.2089247703552246
2023-01-21 09:06:55.735104: step: 1948/527, loss: 0.05228424072265625
2023-01-21 09:06:56.837227: step: 1952/527, loss: 0.22551411390304565
2023-01-21 09:06:57.949500: step: 1956/527, loss: 0.1240018904209137
2023-01-21 09:06:59.076149: step: 1960/527, loss: 0.08612652122974396
2023-01-21 09:07:00.194330: step: 1964/527, loss: 2.1947362422943115
2023-01-21 09:07:01.334448: step: 1968/527, loss: 0.4895205497741699
2023-01-21 09:07:02.468250: step: 1972/527, loss: 0.2622237205505371
2023-01-21 09:07:03.564704: step: 1976/527, loss: 0.14788514375686646
2023-01-21 09:07:04.661276: step: 1980/527, loss: 1.4838156700134277
2023-01-21 09:07:05.773759: step: 1984/527, loss: 1.0007070302963257
2023-01-21 09:07:06.898312: step: 1988/527, loss: 3.2435779571533203
2023-01-21 09:07:08.005082: step: 1992/527, loss: 0.8127609491348267
2023-01-21 09:07:09.154231: step: 1996/527, loss: 0.12645292282104492
2023-01-21 09:07:10.274926: step: 2000/527, loss: 0.821395218372345
2023-01-21 09:07:11.416685: step: 2004/527, loss: 0.23000317811965942
2023-01-21 09:07:12.558512: step: 2008/527, loss: 1.283876657485962
2023-01-21 09:07:13.660195: step: 2012/527, loss: 0.11036530137062073
2023-01-21 09:07:14.803316: step: 2016/527, loss: 0.24143162369728088
2023-01-21 09:07:15.898748: step: 2020/527, loss: 0.08629532158374786
2023-01-21 09:07:17.013022: step: 2024/527, loss: 0.2083907574415207
2023-01-21 09:07:18.131967: step: 2028/527, loss: 0.8790580034255981
2023-01-21 09:07:19.216969: step: 2032/527, loss: 0.06619081646203995
2023-01-21 09:07:20.298426: step: 2036/527, loss: 0.37637922167778015
2023-01-21 09:07:21.384823: step: 2040/527, loss: 0.2209973782300949
2023-01-21 09:07:22.495407: step: 2044/527, loss: 0.5480155944824219
2023-01-21 09:07:23.624288: step: 2048/527, loss: 0.1440381109714508
2023-01-21 09:07:24.712569: step: 2052/527, loss: 0.1772206425666809
2023-01-21 09:07:25.836791: step: 2056/527, loss: 0.1490059792995453
2023-01-21 09:07:26.953409: step: 2060/527, loss: 2.06956148147583
2023-01-21 09:07:28.084574: step: 2064/527, loss: 0.1219981238245964
2023-01-21 09:07:29.195651: step: 2068/527, loss: 0.5840121507644653
2023-01-21 09:07:30.313129: step: 2072/527, loss: 0.7571640014648438
2023-01-21 09:07:31.400830: step: 2076/527, loss: 0.6306487321853638
2023-01-21 09:07:32.494253: step: 2080/527, loss: 0.13957777619361877
2023-01-21 09:07:33.615378: step: 2084/527, loss: 0.08821021020412445
2023-01-21 09:07:34.729981: step: 2088/527, loss: 0.1441861242055893
2023-01-21 09:07:35.841215: step: 2092/527, loss: 0.18021021783351898
2023-01-21 09:07:36.947125: step: 2096/527, loss: 0.09722509235143661
2023-01-21 09:07:38.071314: step: 2100/527, loss: 0.07278265804052353
2023-01-21 09:07:39.180326: step: 2104/527, loss: 1.0719960927963257
2023-01-21 09:07:40.317503: step: 2108/527, loss: 0.19550323486328125
==================================================
Loss: 0.449
--------------------
Dev: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Test: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Chinese: {'event': {'p': 0.5897435897435898, 'r': 0.8518518518518519, 'f1': 0.6969696969696971}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Korean: {'event': {'p': 0.7297297297297297, 'r': 0.42857142857142855, 'f1': 0.54}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
New best chinese model...
New best russian model...
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Chinese: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Chinese: {'event': {'p': 0.5897435897435898, 'r': 0.8518518518518519, 'f1': 0.6969696969696971}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
--------------------
Eng Dev for Korean: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Korean: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, '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 09:08:36.534377: step: 4/527, loss: 0.10690689086914062
2023-01-21 09:08:37.687353: step: 8/527, loss: 0.26229238510131836
2023-01-21 09:08:38.805885: step: 12/527, loss: 0.18392285704612732
2023-01-21 09:08:39.938244: step: 16/527, loss: 0.1851082742214203
2023-01-21 09:08:41.057959: step: 20/527, loss: 0.9686121940612793
2023-01-21 09:08:42.180436: step: 24/527, loss: 0.21167385578155518
2023-01-21 09:08:43.294566: step: 28/527, loss: 0.6079787611961365
2023-01-21 09:08:44.405658: step: 32/527, loss: 0.1102290153503418
2023-01-21 09:08:45.519710: step: 36/527, loss: 0.10351839661598206
2023-01-21 09:08:46.627062: step: 40/527, loss: 0.07550635933876038
2023-01-21 09:08:47.727986: step: 44/527, loss: 0.40787068009376526
2023-01-21 09:08:48.822541: step: 48/527, loss: 0.01969430409371853
2023-01-21 09:08:49.982305: step: 52/527, loss: 0.10179881751537323
2023-01-21 09:08:51.115426: step: 56/527, loss: 1.273101568222046
2023-01-21 09:08:52.225913: step: 60/527, loss: 0.14134541153907776
2023-01-21 09:08:53.332734: step: 64/527, loss: 0.33622679114341736
2023-01-21 09:08:54.472998: step: 68/527, loss: 0.48188328742980957
2023-01-21 09:08:55.609419: step: 72/527, loss: 1.5705112218856812
2023-01-21 09:08:56.694526: step: 76/527, loss: 0.11758680641651154
2023-01-21 09:08:57.820282: step: 80/527, loss: 0.09144806861877441
2023-01-21 09:08:58.928215: step: 84/527, loss: 0.4159158766269684
2023-01-21 09:09:00.023900: step: 88/527, loss: 0.1799517273902893
2023-01-21 09:09:01.132977: step: 92/527, loss: 0.30521997809410095
2023-01-21 09:09:02.242475: step: 96/527, loss: 0.0161711685359478
2023-01-21 09:09:03.366619: step: 100/527, loss: 0.21460795402526855
2023-01-21 09:09:04.465582: step: 104/527, loss: 0.6638267040252686
2023-01-21 09:09:05.566975: step: 108/527, loss: 0.612761378288269
2023-01-21 09:09:06.676396: step: 112/527, loss: 0.18181133270263672
2023-01-21 09:09:07.763207: step: 116/527, loss: 0.2613412141799927
2023-01-21 09:09:08.885976: step: 120/527, loss: 0.2105141133069992
2023-01-21 09:09:10.012499: step: 124/527, loss: 0.13384270668029785
2023-01-21 09:09:11.128624: step: 128/527, loss: 0.22524939477443695
2023-01-21 09:09:12.269907: step: 132/527, loss: 0.2948891818523407
2023-01-21 09:09:13.361831: step: 136/527, loss: 0.27858057618141174
2023-01-21 09:09:14.444545: step: 140/527, loss: 0.30813902616500854
2023-01-21 09:09:15.553893: step: 144/527, loss: 0.9296542406082153
2023-01-21 09:09:16.701216: step: 148/527, loss: 0.1598132699728012
2023-01-21 09:09:17.834483: step: 152/527, loss: 0.08925390243530273
2023-01-21 09:09:18.917553: step: 156/527, loss: 0.11261416226625443
2023-01-21 09:09:20.043004: step: 160/527, loss: 0.21356457471847534
2023-01-21 09:09:21.179686: step: 164/527, loss: 0.10367031395435333
2023-01-21 09:09:22.288559: step: 168/527, loss: 0.1113584041595459
2023-01-21 09:09:23.386911: step: 172/527, loss: 5.876473426818848
2023-01-21 09:09:24.485887: step: 176/527, loss: 0.14457111060619354
2023-01-21 09:09:25.594792: step: 180/527, loss: 0.2923074960708618
2023-01-21 09:09:26.753535: step: 184/527, loss: 0.12595386803150177
2023-01-21 09:09:27.859889: step: 188/527, loss: 0.432263046503067
2023-01-21 09:09:28.982279: step: 192/527, loss: 0.04789486154913902
2023-01-21 09:09:30.093311: step: 196/527, loss: 0.27775058150291443
2023-01-21 09:09:31.190076: step: 200/527, loss: 0.1511821299791336
2023-01-21 09:09:32.286775: step: 204/527, loss: 0.07255931198596954
2023-01-21 09:09:33.414124: step: 208/527, loss: 0.5041448473930359
2023-01-21 09:09:34.542292: step: 212/527, loss: 0.6822780966758728
2023-01-21 09:09:35.639834: step: 216/527, loss: 0.23523148894309998
2023-01-21 09:09:36.763620: step: 220/527, loss: 6.788954257965088
2023-01-21 09:09:37.897033: step: 224/527, loss: 0.18758010864257812
2023-01-21 09:09:39.053361: step: 228/527, loss: 0.21922169625759125
2023-01-21 09:09:40.179051: step: 232/527, loss: 0.15649166703224182
2023-01-21 09:09:41.277124: step: 236/527, loss: 0.09689555317163467
2023-01-21 09:09:42.400559: step: 240/527, loss: 0.12026543915271759
2023-01-21 09:09:43.505888: step: 244/527, loss: 0.12343807518482208
2023-01-21 09:09:44.605965: step: 248/527, loss: 0.41481637954711914
2023-01-21 09:09:45.737267: step: 252/527, loss: 0.22074469923973083
2023-01-21 09:09:46.902851: step: 256/527, loss: 0.3606138229370117
2023-01-21 09:09:48.054884: step: 260/527, loss: 0.3048554062843323
2023-01-21 09:09:49.143806: step: 264/527, loss: 0.3297635018825531
2023-01-21 09:09:50.563023: step: 268/527, loss: 0.07887134701013565
2023-01-21 09:09:51.688393: step: 272/527, loss: 0.1862729787826538
2023-01-21 09:09:52.792015: step: 276/527, loss: 0.07785310596227646
2023-01-21 09:09:53.883564: step: 280/527, loss: 0.16687121987342834
2023-01-21 09:09:55.011467: step: 284/527, loss: 0.7596178650856018
2023-01-21 09:09:56.130614: step: 288/527, loss: 0.12910166382789612
2023-01-21 09:09:57.252748: step: 292/527, loss: 0.21187114715576172
2023-01-21 09:09:58.389733: step: 296/527, loss: 0.5705634355545044
2023-01-21 09:09:59.502350: step: 300/527, loss: 0.7281942367553711
2023-01-21 09:10:00.611453: step: 304/527, loss: 0.20015564560890198
2023-01-21 09:10:01.735289: step: 308/527, loss: 6.149100303649902
2023-01-21 09:10:02.852257: step: 312/527, loss: 0.4192318022251129
2023-01-21 09:10:03.985538: step: 316/527, loss: 0.18486247956752777
2023-01-21 09:10:05.099169: step: 320/527, loss: 0.1175541952252388
2023-01-21 09:10:06.206009: step: 324/527, loss: 0.16281580924987793
2023-01-21 09:10:07.318260: step: 328/527, loss: 0.3077685832977295
2023-01-21 09:10:08.427901: step: 332/527, loss: 0.0775519385933876
2023-01-21 09:10:09.532583: step: 336/527, loss: 0.17245282232761383
2023-01-21 09:10:10.633577: step: 340/527, loss: 0.08445768058300018
2023-01-21 09:10:11.769928: step: 344/527, loss: 0.10412216186523438
2023-01-21 09:10:12.871061: step: 348/527, loss: 0.1898246705532074
2023-01-21 09:10:13.985676: step: 352/527, loss: 0.14269724488258362
2023-01-21 09:10:15.134483: step: 356/527, loss: 0.2525274455547333
2023-01-21 09:10:16.253277: step: 360/527, loss: 0.16955356299877167
2023-01-21 09:10:17.351240: step: 364/527, loss: 0.0355035774409771
2023-01-21 09:10:18.500206: step: 368/527, loss: 0.08516263961791992
2023-01-21 09:10:19.599532: step: 372/527, loss: 0.10598832368850708
2023-01-21 09:10:20.702858: step: 376/527, loss: 0.2322016805410385
2023-01-21 09:10:21.823914: step: 380/527, loss: 1.0321524143218994
2023-01-21 09:10:22.928031: step: 384/527, loss: 1.0343544483184814
2023-01-21 09:10:24.044965: step: 388/527, loss: 0.11659832298755646
2023-01-21 09:10:25.178137: step: 392/527, loss: 0.7954214811325073
2023-01-21 09:10:26.286424: step: 396/527, loss: 0.3000028133392334
2023-01-21 09:10:27.416880: step: 400/527, loss: 0.4770169258117676
2023-01-21 09:10:28.516308: step: 404/527, loss: 0.05647563934326172
2023-01-21 09:10:29.636651: step: 408/527, loss: 0.18130803108215332
2023-01-21 09:10:30.726533: step: 412/527, loss: 0.08880696445703506
2023-01-21 09:10:31.860617: step: 416/527, loss: 0.05320234224200249
2023-01-21 09:10:32.959189: step: 420/527, loss: 0.11156439781188965
2023-01-21 09:10:34.073252: step: 424/527, loss: 1.1430052518844604
2023-01-21 09:10:35.186748: step: 428/527, loss: 0.8291699290275574
2023-01-21 09:10:36.299170: step: 432/527, loss: 0.3138306140899658
2023-01-21 09:10:37.368670: step: 436/527, loss: 0.11644463241100311
2023-01-21 09:10:38.492476: step: 440/527, loss: 0.012163449078798294
2023-01-21 09:10:39.611135: step: 444/527, loss: 0.09956265240907669
2023-01-21 09:10:40.728006: step: 448/527, loss: 0.482940673828125
2023-01-21 09:10:41.834024: step: 452/527, loss: 0.020507145673036575
2023-01-21 09:10:42.998065: step: 456/527, loss: 0.23200541734695435
2023-01-21 09:10:44.121931: step: 460/527, loss: 0.6038606762886047
2023-01-21 09:10:45.230006: step: 464/527, loss: 0.06216559186577797
2023-01-21 09:10:46.373315: step: 468/527, loss: 0.2516061067581177
2023-01-21 09:10:47.465680: step: 472/527, loss: 0.7467668056488037
2023-01-21 09:10:48.570436: step: 476/527, loss: 0.04760603979229927
2023-01-21 09:10:49.687325: step: 480/527, loss: 0.2646981477737427
2023-01-21 09:10:50.846520: step: 484/527, loss: 0.1797521561384201
2023-01-21 09:10:51.983866: step: 488/527, loss: 0.04645397514104843
2023-01-21 09:10:53.069841: step: 492/527, loss: 0.060700275003910065
2023-01-21 09:10:54.180598: step: 496/527, loss: 0.22435550391674042
2023-01-21 09:10:55.284756: step: 500/527, loss: 0.06895676255226135
2023-01-21 09:10:56.411328: step: 504/527, loss: 0.2748515009880066
2023-01-21 09:10:57.501022: step: 508/527, loss: 0.21761950850486755
2023-01-21 09:10:58.603517: step: 512/527, loss: 0.2205769568681717
2023-01-21 09:10:59.701371: step: 516/527, loss: 0.14005526900291443
2023-01-21 09:11:00.777684: step: 520/527, loss: 0.2663953900337219
2023-01-21 09:11:01.911237: step: 524/527, loss: 0.6959317922592163
2023-01-21 09:11:03.022328: step: 528/527, loss: 0.07471341639757156
2023-01-21 09:11:04.146964: step: 532/527, loss: 0.4083843231201172
2023-01-21 09:11:05.251845: step: 536/527, loss: 0.9699186086654663
2023-01-21 09:11:06.363189: step: 540/527, loss: 0.2076442837715149
2023-01-21 09:11:07.482181: step: 544/527, loss: 0.7029988765716553
2023-01-21 09:11:08.578131: step: 548/527, loss: 0.08358597755432129
2023-01-21 09:11:09.711438: step: 552/527, loss: 0.42553555965423584
2023-01-21 09:11:10.819757: step: 556/527, loss: 0.22079545259475708
2023-01-21 09:11:11.925503: step: 560/527, loss: 0.06804303824901581
2023-01-21 09:11:13.077217: step: 564/527, loss: 0.24682527780532837
2023-01-21 09:11:14.211752: step: 568/527, loss: 0.16522203385829926
2023-01-21 09:11:15.341073: step: 572/527, loss: 0.35852736234664917
2023-01-21 09:11:16.498804: step: 576/527, loss: 1.2968554496765137
2023-01-21 09:11:17.609041: step: 580/527, loss: 0.7931557297706604
2023-01-21 09:11:18.692661: step: 584/527, loss: 0.16539031267166138
2023-01-21 09:11:19.854571: step: 588/527, loss: 0.4904360771179199
2023-01-21 09:11:21.015894: step: 592/527, loss: 0.03752746805548668
2023-01-21 09:11:22.146548: step: 596/527, loss: 0.6924483180046082
2023-01-21 09:11:23.279005: step: 600/527, loss: 0.6220345497131348
2023-01-21 09:11:24.377954: step: 604/527, loss: 0.8555251955986023
2023-01-21 09:11:25.494915: step: 608/527, loss: 0.4294985830783844
2023-01-21 09:11:26.601373: step: 612/527, loss: 0.016052579507231712
2023-01-21 09:11:27.708488: step: 616/527, loss: 0.1309412121772766
2023-01-21 09:11:28.832762: step: 620/527, loss: 0.045116521418094635
2023-01-21 09:11:29.937812: step: 624/527, loss: 0.2054397612810135
2023-01-21 09:11:31.062320: step: 628/527, loss: 0.15325704216957092
2023-01-21 09:11:32.196476: step: 632/527, loss: 0.20916244387626648
2023-01-21 09:11:33.325681: step: 636/527, loss: 0.023050928488373756
2023-01-21 09:11:34.437863: step: 640/527, loss: 0.5019675493240356
2023-01-21 09:11:35.550135: step: 644/527, loss: 0.07485733181238174
2023-01-21 09:11:36.640715: step: 648/527, loss: 0.2559559941291809
2023-01-21 09:11:37.738371: step: 652/527, loss: 0.6436958909034729
2023-01-21 09:11:38.849590: step: 656/527, loss: 0.6133899688720703
2023-01-21 09:11:39.933211: step: 660/527, loss: 0.903583824634552
2023-01-21 09:11:41.061658: step: 664/527, loss: 0.06538905948400497
2023-01-21 09:11:42.155901: step: 668/527, loss: 0.0531894713640213
2023-01-21 09:11:43.281512: step: 672/527, loss: 0.15315064787864685
2023-01-21 09:11:44.417993: step: 676/527, loss: 0.4527393579483032
2023-01-21 09:11:45.498171: step: 680/527, loss: 0.06289754062891006
2023-01-21 09:11:46.622222: step: 684/527, loss: 0.2575264573097229
2023-01-21 09:11:47.735350: step: 688/527, loss: 0.013828087598085403
2023-01-21 09:11:48.831689: step: 692/527, loss: 0.17731815576553345
2023-01-21 09:11:49.926453: step: 696/527, loss: 0.4480947256088257
2023-01-21 09:11:51.040975: step: 700/527, loss: 1.1558541059494019
2023-01-21 09:11:52.175722: step: 704/527, loss: 0.03804426267743111
2023-01-21 09:11:53.315759: step: 708/527, loss: 0.16213403642177582
2023-01-21 09:11:54.467560: step: 712/527, loss: 1.238487958908081
2023-01-21 09:11:55.619632: step: 716/527, loss: 0.18230657279491425
2023-01-21 09:11:56.760974: step: 720/527, loss: 0.5329722166061401
2023-01-21 09:11:57.829997: step: 724/527, loss: 0.14703664183616638
2023-01-21 09:11:58.935772: step: 728/527, loss: 0.09924321621656418
2023-01-21 09:12:00.038902: step: 732/527, loss: 0.20120000839233398
2023-01-21 09:12:01.119078: step: 736/527, loss: 0.3734327554702759
2023-01-21 09:12:02.235454: step: 740/527, loss: 0.21190989017486572
2023-01-21 09:12:03.317511: step: 744/527, loss: 0.5763669013977051
2023-01-21 09:12:04.422948: step: 748/527, loss: 0.1756018102169037
2023-01-21 09:12:05.554537: step: 752/527, loss: 0.8739427328109741
2023-01-21 09:12:06.672228: step: 756/527, loss: 0.6947381496429443
2023-01-21 09:12:07.762760: step: 760/527, loss: 0.13500681519508362
2023-01-21 09:12:08.870441: step: 764/527, loss: 0.03397350385785103
2023-01-21 09:12:09.957181: step: 768/527, loss: 0.050150156021118164
2023-01-21 09:12:11.090935: step: 772/527, loss: 0.07563390582799911
2023-01-21 09:12:12.191287: step: 776/527, loss: 0.243687242269516
2023-01-21 09:12:13.293188: step: 780/527, loss: 0.05087833106517792
2023-01-21 09:12:14.412990: step: 784/527, loss: 0.07746048271656036
2023-01-21 09:12:15.526205: step: 788/527, loss: 0.1341322511434555
2023-01-21 09:12:16.639252: step: 792/527, loss: 0.06158266216516495
2023-01-21 09:12:17.734922: step: 796/527, loss: 0.06087656319141388
2023-01-21 09:12:18.855280: step: 800/527, loss: 0.2464689314365387
2023-01-21 09:12:19.976093: step: 804/527, loss: 0.25684911012649536
2023-01-21 09:12:21.065442: step: 808/527, loss: 0.13031105697155
2023-01-21 09:12:22.179559: step: 812/527, loss: 0.17095699906349182
2023-01-21 09:12:23.283459: step: 816/527, loss: 0.12209143489599228
2023-01-21 09:12:24.389456: step: 820/527, loss: 0.2230725884437561
2023-01-21 09:12:25.548865: step: 824/527, loss: 0.5473877787590027
2023-01-21 09:12:26.665676: step: 828/527, loss: 0.11638985574245453
2023-01-21 09:12:27.778157: step: 832/527, loss: 0.6785234808921814
2023-01-21 09:12:28.880825: step: 836/527, loss: 0.2687937021255493
2023-01-21 09:12:29.997047: step: 840/527, loss: 0.29614678025245667
2023-01-21 09:12:31.093256: step: 844/527, loss: 0.2697829306125641
2023-01-21 09:12:32.213000: step: 848/527, loss: 0.03577737882733345
2023-01-21 09:12:33.330527: step: 852/527, loss: 0.6415659189224243
2023-01-21 09:12:34.417110: step: 856/527, loss: 0.6541442275047302
2023-01-21 09:12:35.527053: step: 860/527, loss: 0.19686970114707947
2023-01-21 09:12:36.601663: step: 864/527, loss: 0.6205440163612366
2023-01-21 09:12:37.716229: step: 868/527, loss: 0.724902868270874
2023-01-21 09:12:38.839003: step: 872/527, loss: 0.43067342042922974
2023-01-21 09:12:39.929708: step: 876/527, loss: 0.0367342010140419
2023-01-21 09:12:41.058864: step: 880/527, loss: 0.2426941990852356
2023-01-21 09:12:42.181511: step: 884/527, loss: 0.44762009382247925
2023-01-21 09:12:43.307845: step: 888/527, loss: 0.1117599755525589
2023-01-21 09:12:44.429566: step: 892/527, loss: 0.043029118329286575
2023-01-21 09:12:45.550840: step: 896/527, loss: 0.09100999683141708
2023-01-21 09:12:46.645714: step: 900/527, loss: 0.105923131108284
2023-01-21 09:12:47.747201: step: 904/527, loss: 0.12159473448991776
2023-01-21 09:12:48.844349: step: 908/527, loss: 0.06646624207496643
2023-01-21 09:12:49.932197: step: 912/527, loss: 0.15469437837600708
2023-01-21 09:12:51.065655: step: 916/527, loss: 0.3962244391441345
2023-01-21 09:12:52.217268: step: 920/527, loss: 0.15693531930446625
2023-01-21 09:12:53.304673: step: 924/527, loss: 0.3318410813808441
2023-01-21 09:12:54.425938: step: 928/527, loss: 0.1302165985107422
2023-01-21 09:12:55.519129: step: 932/527, loss: 0.09787073731422424
2023-01-21 09:12:56.629764: step: 936/527, loss: 5.116474628448486
2023-01-21 09:12:57.770419: step: 940/527, loss: 0.10111570358276367
2023-01-21 09:12:58.885927: step: 944/527, loss: 0.7220234870910645
2023-01-21 09:13:00.002468: step: 948/527, loss: 0.3562307357788086
2023-01-21 09:13:01.117112: step: 952/527, loss: 0.441702663898468
2023-01-21 09:13:02.216721: step: 956/527, loss: 0.09431767463684082
2023-01-21 09:13:03.333456: step: 960/527, loss: 0.01877765730023384
2023-01-21 09:13:04.429232: step: 964/527, loss: 0.24384915828704834
2023-01-21 09:13:05.552305: step: 968/527, loss: 0.11317439377307892
2023-01-21 09:13:06.690178: step: 972/527, loss: 0.1555279642343521
2023-01-21 09:13:07.792307: step: 976/527, loss: 0.1521296501159668
2023-01-21 09:13:08.922224: step: 980/527, loss: 0.26489123702049255
2023-01-21 09:13:10.027079: step: 984/527, loss: 0.02657175064086914
2023-01-21 09:13:11.150527: step: 988/527, loss: 0.9635842442512512
2023-01-21 09:13:12.224072: step: 992/527, loss: 0.05721640586853027
2023-01-21 09:13:13.329390: step: 996/527, loss: 0.5187878608703613
2023-01-21 09:13:14.422627: step: 1000/527, loss: 0.17131757736206055
2023-01-21 09:13:15.571038: step: 1004/527, loss: 1.59785795211792
2023-01-21 09:13:16.647124: step: 1008/527, loss: 0.030515337362885475
2023-01-21 09:13:17.803103: step: 1012/527, loss: 0.6989672183990479
2023-01-21 09:13:18.931390: step: 1016/527, loss: 0.09899745136499405
2023-01-21 09:13:20.023695: step: 1020/527, loss: 0.601532518863678
2023-01-21 09:13:21.187548: step: 1024/527, loss: 1.5280327796936035
2023-01-21 09:13:22.279891: step: 1028/527, loss: 0.25149011611938477
2023-01-21 09:13:23.384052: step: 1032/527, loss: 0.03057580068707466
2023-01-21 09:13:24.506269: step: 1036/527, loss: 0.18764066696166992
2023-01-21 09:13:25.632251: step: 1040/527, loss: 0.14220742881298065
2023-01-21 09:13:26.737052: step: 1044/527, loss: 0.4160740375518799
2023-01-21 09:13:27.858490: step: 1048/527, loss: 0.14959710836410522
2023-01-21 09:13:28.956734: step: 1052/527, loss: 0.08783617615699768
2023-01-21 09:13:30.073655: step: 1056/527, loss: 0.20530100166797638
2023-01-21 09:13:31.190951: step: 1060/527, loss: 1.0951528549194336
2023-01-21 09:13:32.293226: step: 1064/527, loss: 0.6230490803718567
2023-01-21 09:13:33.415423: step: 1068/527, loss: 0.047351837158203125
2023-01-21 09:13:34.521885: step: 1072/527, loss: 0.30140554904937744
2023-01-21 09:13:35.663647: step: 1076/527, loss: 0.17081165313720703
2023-01-21 09:13:36.773879: step: 1080/527, loss: 0.11976084858179092
2023-01-21 09:13:37.941139: step: 1084/527, loss: 0.015096187591552734
2023-01-21 09:13:39.058956: step: 1088/527, loss: 0.25514861941337585
2023-01-21 09:13:40.172110: step: 1092/527, loss: 0.08299532532691956
2023-01-21 09:13:41.275136: step: 1096/527, loss: 0.05821748077869415
2023-01-21 09:13:42.371202: step: 1100/527, loss: 0.07984709739685059
2023-01-21 09:13:43.499123: step: 1104/527, loss: 0.03862800449132919
2023-01-21 09:13:44.608991: step: 1108/527, loss: 0.17441534996032715
2023-01-21 09:13:45.684238: step: 1112/527, loss: 0.04745130613446236
2023-01-21 09:13:46.800034: step: 1116/527, loss: 0.5034268498420715
2023-01-21 09:13:47.915613: step: 1120/527, loss: 0.025683045387268066
2023-01-21 09:13:49.025787: step: 1124/527, loss: 0.6742849349975586
2023-01-21 09:13:50.165247: step: 1128/527, loss: 0.06588998436927795
2023-01-21 09:13:51.276402: step: 1132/527, loss: 0.1916070133447647
2023-01-21 09:13:52.379380: step: 1136/527, loss: 0.7642965316772461
2023-01-21 09:13:53.494551: step: 1140/527, loss: 0.12646256387233734
2023-01-21 09:13:54.629977: step: 1144/527, loss: 0.12535008788108826
2023-01-21 09:13:55.748445: step: 1148/527, loss: 0.6683266758918762
2023-01-21 09:13:56.833722: step: 1152/527, loss: 0.19703570008277893
2023-01-21 09:13:57.998990: step: 1156/527, loss: 0.20910978317260742
2023-01-21 09:13:59.096344: step: 1160/527, loss: 0.21245956420898438
2023-01-21 09:14:00.200781: step: 1164/527, loss: 0.06803891807794571
2023-01-21 09:14:01.315781: step: 1168/527, loss: 0.6156831979751587
2023-01-21 09:14:02.424858: step: 1172/527, loss: 0.02805934101343155
2023-01-21 09:14:03.533088: step: 1176/527, loss: 0.4477941393852234
2023-01-21 09:14:04.652158: step: 1180/527, loss: 0.20378351211547852
2023-01-21 09:14:05.781658: step: 1184/527, loss: 1.4286248683929443
2023-01-21 09:14:06.896025: step: 1188/527, loss: 0.1440032571554184
2023-01-21 09:14:08.015063: step: 1192/527, loss: 0.06988263130187988
2023-01-21 09:14:09.120195: step: 1196/527, loss: 0.10316705703735352
2023-01-21 09:14:10.251230: step: 1200/527, loss: 0.39985477924346924
2023-01-21 09:14:11.345435: step: 1204/527, loss: 0.19708210229873657
2023-01-21 09:14:12.479640: step: 1208/527, loss: 0.07931695878505707
2023-01-21 09:14:13.606915: step: 1212/527, loss: 0.19265621900558472
2023-01-21 09:14:14.726270: step: 1216/527, loss: 0.19477620720863342
2023-01-21 09:14:15.836560: step: 1220/527, loss: 0.14170321822166443
2023-01-21 09:14:16.972620: step: 1224/527, loss: 0.23198643326759338
2023-01-21 09:14:18.055772: step: 1228/527, loss: 0.08336324989795685
2023-01-21 09:14:19.163442: step: 1232/527, loss: 0.6265767216682434
2023-01-21 09:14:20.269892: step: 1236/527, loss: 0.06663751602172852
2023-01-21 09:14:21.414307: step: 1240/527, loss: 0.1593008041381836
2023-01-21 09:14:22.541615: step: 1244/527, loss: 0.5316864848136902
2023-01-21 09:14:23.724518: step: 1248/527, loss: 0.5278263092041016
2023-01-21 09:14:24.839600: step: 1252/527, loss: 0.04160184785723686
2023-01-21 09:14:25.966309: step: 1256/527, loss: 0.036322880536317825
2023-01-21 09:14:27.128902: step: 1260/527, loss: 0.6000651121139526
2023-01-21 09:14:28.234875: step: 1264/527, loss: 0.31250160932540894
2023-01-21 09:14:29.335495: step: 1268/527, loss: 0.17543601989746094
2023-01-21 09:14:30.434984: step: 1272/527, loss: 0.34024086594581604
2023-01-21 09:14:31.557881: step: 1276/527, loss: 0.2732410430908203
2023-01-21 09:14:32.699297: step: 1280/527, loss: 0.7535012364387512
2023-01-21 09:14:33.810931: step: 1284/527, loss: 0.13631367683410645
2023-01-21 09:14:34.917973: step: 1288/527, loss: 0.6097455620765686
2023-01-21 09:14:36.063398: step: 1292/527, loss: 0.6293331980705261
2023-01-21 09:14:37.183841: step: 1296/527, loss: 0.7128167152404785
2023-01-21 09:14:38.292891: step: 1300/527, loss: 0.13039374351501465
2023-01-21 09:14:39.417559: step: 1304/527, loss: 0.171274334192276
2023-01-21 09:14:40.556316: step: 1308/527, loss: 0.6098413467407227
2023-01-21 09:14:41.667783: step: 1312/527, loss: 0.6271671652793884
2023-01-21 09:14:42.788444: step: 1316/527, loss: 0.17098703980445862
2023-01-21 09:14:43.884944: step: 1320/527, loss: 0.04796471819281578
2023-01-21 09:14:44.994256: step: 1324/527, loss: 0.7401866912841797
2023-01-21 09:14:46.102940: step: 1328/527, loss: 0.3096674680709839
2023-01-21 09:14:47.195205: step: 1332/527, loss: 0.11526908725500107
2023-01-21 09:14:48.312688: step: 1336/527, loss: 5.885909080505371
2023-01-21 09:14:49.428523: step: 1340/527, loss: 0.10175705701112747
2023-01-21 09:14:50.539194: step: 1344/527, loss: 0.14329329133033752
2023-01-21 09:14:51.647641: step: 1348/527, loss: 0.12524500489234924
2023-01-21 09:14:52.749803: step: 1352/527, loss: 0.5256365537643433
2023-01-21 09:14:53.853110: step: 1356/527, loss: 0.14834603667259216
2023-01-21 09:14:54.979926: step: 1360/527, loss: 0.3145448565483093
2023-01-21 09:14:56.094735: step: 1364/527, loss: 0.3525451719760895
2023-01-21 09:14:57.194227: step: 1368/527, loss: 0.10944939404726028
2023-01-21 09:14:58.327441: step: 1372/527, loss: 0.019724082201719284
2023-01-21 09:14:59.463336: step: 1376/527, loss: 0.16537638008594513
2023-01-21 09:15:00.586977: step: 1380/527, loss: 0.5396057963371277
2023-01-21 09:15:01.668275: step: 1384/527, loss: 0.09598135948181152
2023-01-21 09:15:02.779253: step: 1388/527, loss: 0.9520591497421265
2023-01-21 09:15:03.868678: step: 1392/527, loss: 0.07424316555261612
2023-01-21 09:15:04.987740: step: 1396/527, loss: 0.07592492550611496
2023-01-21 09:15:06.080236: step: 1400/527, loss: 0.15157172083854675
2023-01-21 09:15:07.205591: step: 1404/527, loss: 0.09792271256446838
2023-01-21 09:15:08.320858: step: 1408/527, loss: 0.0961262658238411
2023-01-21 09:15:09.470370: step: 1412/527, loss: 1.803153395652771
2023-01-21 09:15:10.577142: step: 1416/527, loss: 0.1287805587053299
2023-01-21 09:15:11.707676: step: 1420/527, loss: 0.1581663191318512
2023-01-21 09:15:12.846608: step: 1424/527, loss: 0.21145859360694885
2023-01-21 09:15:13.948117: step: 1428/527, loss: 0.6952801942825317
2023-01-21 09:15:15.051532: step: 1432/527, loss: 0.08553238213062286
2023-01-21 09:15:16.192166: step: 1436/527, loss: 0.11029711365699768
2023-01-21 09:15:17.304407: step: 1440/527, loss: 0.14753007888793945
2023-01-21 09:15:18.400168: step: 1444/527, loss: 0.21220636367797852
2023-01-21 09:15:19.531156: step: 1448/527, loss: 0.14068298041820526
2023-01-21 09:15:20.672510: step: 1452/527, loss: 0.6115983128547668
2023-01-21 09:15:21.798426: step: 1456/527, loss: 0.4948551058769226
2023-01-21 09:15:22.921628: step: 1460/527, loss: 0.04553709179162979
2023-01-21 09:15:24.037185: step: 1464/527, loss: 0.04666939005255699
2023-01-21 09:15:25.151790: step: 1468/527, loss: 0.2741253972053528
2023-01-21 09:15:26.259803: step: 1472/527, loss: 0.09618870913982391
2023-01-21 09:15:27.366868: step: 1476/527, loss: 0.5667693018913269
2023-01-21 09:15:28.492950: step: 1480/527, loss: 0.38979265093803406
2023-01-21 09:15:29.605492: step: 1484/527, loss: 0.1593582183122635
2023-01-21 09:15:30.720927: step: 1488/527, loss: 0.11751909554004669
2023-01-21 09:15:31.844410: step: 1492/527, loss: 0.1928863227367401
2023-01-21 09:15:32.969242: step: 1496/527, loss: 0.8210095763206482
2023-01-21 09:15:34.071824: step: 1500/527, loss: 0.29117992520332336
2023-01-21 09:15:35.187415: step: 1504/527, loss: 0.08087539672851562
2023-01-21 09:15:36.277268: step: 1508/527, loss: 0.19064950942993164
2023-01-21 09:15:37.379772: step: 1512/527, loss: 0.15087157487869263
2023-01-21 09:15:38.511956: step: 1516/527, loss: 0.321188747882843
2023-01-21 09:15:39.629461: step: 1520/527, loss: 0.10306596755981445
2023-01-21 09:15:40.726916: step: 1524/527, loss: 0.20539608597755432
2023-01-21 09:15:41.833225: step: 1528/527, loss: 0.5874325633049011
2023-01-21 09:15:42.936993: step: 1532/527, loss: 0.08216361701488495
2023-01-21 09:15:44.047025: step: 1536/527, loss: 0.809807538986206
2023-01-21 09:15:45.172859: step: 1540/527, loss: 0.4458211064338684
2023-01-21 09:15:46.293080: step: 1544/527, loss: 0.17472851276397705
2023-01-21 09:15:47.395971: step: 1548/527, loss: 0.5652610659599304
2023-01-21 09:15:48.512959: step: 1552/527, loss: 0.027675582095980644
2023-01-21 09:15:49.669330: step: 1556/527, loss: 0.11164265125989914
2023-01-21 09:15:50.780066: step: 1560/527, loss: 0.3571251332759857
2023-01-21 09:15:51.909440: step: 1564/527, loss: 0.45429757237434387
2023-01-21 09:15:53.004785: step: 1568/527, loss: 1.0791453123092651
2023-01-21 09:15:54.121063: step: 1572/527, loss: 0.20986071228981018
2023-01-21 09:15:55.229779: step: 1576/527, loss: 0.33028706908226013
2023-01-21 09:15:56.331252: step: 1580/527, loss: 0.6444014310836792
2023-01-21 09:15:57.437752: step: 1584/527, loss: 0.05755796283483505
2023-01-21 09:15:58.571415: step: 1588/527, loss: 0.106346994638443
2023-01-21 09:15:59.681160: step: 1592/527, loss: 0.7262479662895203
2023-01-21 09:16:00.828361: step: 1596/527, loss: 0.5133785009384155
2023-01-21 09:16:01.939200: step: 1600/527, loss: 0.02716188319027424
2023-01-21 09:16:03.043326: step: 1604/527, loss: 1.0088305473327637
2023-01-21 09:16:04.136604: step: 1608/527, loss: 0.07211685180664062
2023-01-21 09:16:05.241280: step: 1612/527, loss: 0.689362645149231
2023-01-21 09:16:06.369970: step: 1616/527, loss: 0.7506176233291626
2023-01-21 09:16:07.505204: step: 1620/527, loss: 0.062085725367069244
2023-01-21 09:16:08.614335: step: 1624/527, loss: 0.4119747281074524
2023-01-21 09:16:09.729012: step: 1628/527, loss: 0.5336955189704895
2023-01-21 09:16:10.826948: step: 1632/527, loss: 0.06760063022375107
2023-01-21 09:16:11.922469: step: 1636/527, loss: 0.09099073708057404
2023-01-21 09:16:13.054846: step: 1640/527, loss: 0.5419539213180542
2023-01-21 09:16:14.174908: step: 1644/527, loss: 1.2363033294677734
2023-01-21 09:16:15.284406: step: 1648/527, loss: 0.13272853195667267
2023-01-21 09:16:16.402706: step: 1652/527, loss: 0.14073458313941956
2023-01-21 09:16:17.529459: step: 1656/527, loss: 0.1024385392665863
2023-01-21 09:16:18.671857: step: 1660/527, loss: 0.19618120789527893
2023-01-21 09:16:19.776288: step: 1664/527, loss: 0.3330685496330261
2023-01-21 09:16:20.915855: step: 1668/527, loss: 0.061608217656612396
2023-01-21 09:16:22.041392: step: 1672/527, loss: 0.5016913414001465
2023-01-21 09:16:23.155561: step: 1676/527, loss: 0.2912317216396332
2023-01-21 09:16:24.253365: step: 1680/527, loss: 0.054673001170158386
2023-01-21 09:16:25.347193: step: 1684/527, loss: 0.22520524263381958
2023-01-21 09:16:26.444052: step: 1688/527, loss: 0.24312329292297363
2023-01-21 09:16:27.556885: step: 1692/527, loss: 0.2241046130657196
2023-01-21 09:16:28.715993: step: 1696/527, loss: 0.2024269998073578
2023-01-21 09:16:29.848668: step: 1700/527, loss: 0.507845401763916
2023-01-21 09:16:30.948301: step: 1704/527, loss: 0.1972704976797104
2023-01-21 09:16:32.040480: step: 1708/527, loss: 0.9078596830368042
2023-01-21 09:16:33.153580: step: 1712/527, loss: 0.10781069099903107
2023-01-21 09:16:34.279530: step: 1716/527, loss: 0.3044893741607666
2023-01-21 09:16:35.395389: step: 1720/527, loss: 0.24299149215221405
2023-01-21 09:16:36.545657: step: 1724/527, loss: 1.3196039199829102
2023-01-21 09:16:37.659413: step: 1728/527, loss: 0.1933065503835678
2023-01-21 09:16:38.806957: step: 1732/527, loss: 0.2124234139919281
2023-01-21 09:16:39.900112: step: 1736/527, loss: 0.0949990302324295
2023-01-21 09:16:40.984428: step: 1740/527, loss: 1.154528260231018
2023-01-21 09:16:42.120138: step: 1744/527, loss: 3.1624863147735596
2023-01-21 09:16:43.281011: step: 1748/527, loss: 0.21432171761989594
2023-01-21 09:16:44.392177: step: 1752/527, loss: 1.3081138134002686
2023-01-21 09:16:45.520383: step: 1756/527, loss: 0.2940703332424164
2023-01-21 09:16:46.678483: step: 1760/527, loss: 0.513171374797821
2023-01-21 09:16:47.773913: step: 1764/527, loss: 0.029730796813964844
2023-01-21 09:16:48.894877: step: 1768/527, loss: 0.17151489853858948
2023-01-21 09:16:50.046468: step: 1772/527, loss: 0.1870718002319336
2023-01-21 09:16:51.167549: step: 1776/527, loss: 1.1802641153335571
2023-01-21 09:16:52.290500: step: 1780/527, loss: 0.036284636706113815
2023-01-21 09:16:53.406416: step: 1784/527, loss: 0.1877608299255371
2023-01-21 09:16:54.496950: step: 1788/527, loss: 0.16261368989944458
2023-01-21 09:16:55.583204: step: 1792/527, loss: 0.22908459603786469
2023-01-21 09:16:56.694933: step: 1796/527, loss: 0.23925809562206268
2023-01-21 09:16:57.823313: step: 1800/527, loss: 0.8778618574142456
2023-01-21 09:16:58.924629: step: 1804/527, loss: 0.5116405487060547
2023-01-21 09:17:00.058011: step: 1808/527, loss: 1.1821987628936768
2023-01-21 09:17:01.174120: step: 1812/527, loss: 0.6773964166641235
2023-01-21 09:17:02.267575: step: 1816/527, loss: 0.19796791672706604
2023-01-21 09:17:03.391093: step: 1820/527, loss: 0.5183804035186768
2023-01-21 09:17:04.497784: step: 1824/527, loss: 0.07474000006914139
2023-01-21 09:17:05.604023: step: 1828/527, loss: 0.06196761131286621
2023-01-21 09:17:06.708218: step: 1832/527, loss: 0.1263589859008789
2023-01-21 09:17:07.830569: step: 1836/527, loss: 0.1731864959001541
2023-01-21 09:17:08.916500: step: 1840/527, loss: 0.10313358157873154
2023-01-21 09:17:10.034683: step: 1844/527, loss: 0.6231164336204529
2023-01-21 09:17:11.129801: step: 1848/527, loss: 0.15423031151294708
2023-01-21 09:17:12.276054: step: 1852/527, loss: 0.07791309058666229
2023-01-21 09:17:13.387714: step: 1856/527, loss: 0.0723499283194542
2023-01-21 09:17:14.505460: step: 1860/527, loss: 0.010405349545180798
2023-01-21 09:17:15.612540: step: 1864/527, loss: 0.5426110029220581
2023-01-21 09:17:16.750516: step: 1868/527, loss: 2.9066708087921143
2023-01-21 09:17:17.897905: step: 1872/527, loss: 0.08476868271827698
2023-01-21 09:17:19.027186: step: 1876/527, loss: 0.16761358082294464
2023-01-21 09:17:20.116081: step: 1880/527, loss: 0.02445092238485813
2023-01-21 09:17:21.232923: step: 1884/527, loss: 0.07886552810668945
2023-01-21 09:17:22.344343: step: 1888/527, loss: 0.6830229759216309
2023-01-21 09:17:23.443558: step: 1892/527, loss: 0.21814775466918945
2023-01-21 09:17:24.540800: step: 1896/527, loss: 0.073696568608284
2023-01-21 09:17:25.660283: step: 1900/527, loss: 0.05319714546203613
2023-01-21 09:17:26.810744: step: 1904/527, loss: 0.04640341177582741
2023-01-21 09:17:27.917009: step: 1908/527, loss: 0.16652211546897888
2023-01-21 09:17:29.018192: step: 1912/527, loss: 0.1321442574262619
2023-01-21 09:17:30.106648: step: 1916/527, loss: 0.10971913486719131
2023-01-21 09:17:31.245313: step: 1920/527, loss: 0.14570856094360352
2023-01-21 09:17:32.352425: step: 1924/527, loss: 0.3235597610473633
2023-01-21 09:17:33.477554: step: 1928/527, loss: 0.17856568098068237
2023-01-21 09:17:34.612092: step: 1932/527, loss: 0.30888310074806213
2023-01-21 09:17:35.698585: step: 1936/527, loss: 0.7898846864700317
2023-01-21 09:17:36.813665: step: 1940/527, loss: 1.4510753154754639
2023-01-21 09:17:37.892122: step: 1944/527, loss: 0.09784574061632156
2023-01-21 09:17:39.010340: step: 1948/527, loss: 0.33908191323280334
2023-01-21 09:17:40.138437: step: 1952/527, loss: 0.14806756377220154
2023-01-21 09:17:41.273574: step: 1956/527, loss: 0.160518079996109
2023-01-21 09:17:42.389337: step: 1960/527, loss: 0.15122422575950623
2023-01-21 09:17:43.510618: step: 1964/527, loss: 0.25120019912719727
2023-01-21 09:17:44.649907: step: 1968/527, loss: 4.559635639190674
2023-01-21 09:17:45.761573: step: 1972/527, loss: 0.1597447395324707
2023-01-21 09:17:46.914652: step: 1976/527, loss: 0.5630172491073608
2023-01-21 09:17:48.041058: step: 1980/527, loss: 0.0700603500008583
2023-01-21 09:17:49.183687: step: 1984/527, loss: 0.23006907105445862
2023-01-21 09:17:50.287290: step: 1988/527, loss: 0.36249151825904846
2023-01-21 09:17:51.381782: step: 1992/527, loss: 0.05795612558722496
2023-01-21 09:17:52.494886: step: 1996/527, loss: 0.05022997781634331
2023-01-21 09:17:53.629730: step: 2000/527, loss: 0.1965283900499344
2023-01-21 09:17:54.731811: step: 2004/527, loss: 0.08582153916358948
2023-01-21 09:17:55.831544: step: 2008/527, loss: 0.08638457953929901
2023-01-21 09:17:56.940295: step: 2012/527, loss: 0.5372883677482605
2023-01-21 09:17:58.051787: step: 2016/527, loss: 0.07991671562194824
2023-01-21 09:17:59.145812: step: 2020/527, loss: 0.09569688141345978
2023-01-21 09:18:00.260773: step: 2024/527, loss: 0.3317447006702423
2023-01-21 09:18:01.356587: step: 2028/527, loss: 0.13375358283519745
2023-01-21 09:18:02.493309: step: 2032/527, loss: 0.3296220004558563
2023-01-21 09:18:03.586535: step: 2036/527, loss: 0.14134007692337036
2023-01-21 09:18:04.693039: step: 2040/527, loss: 0.8119940757751465
2023-01-21 09:18:05.831688: step: 2044/527, loss: 0.19644784927368164
2023-01-21 09:18:06.943284: step: 2048/527, loss: 0.09602288901805878
2023-01-21 09:18:08.041792: step: 2052/527, loss: 0.3183956742286682
2023-01-21 09:18:09.140812: step: 2056/527, loss: 0.202362060546875
2023-01-21 09:18:10.274764: step: 2060/527, loss: 0.24712371826171875
2023-01-21 09:18:11.377384: step: 2064/527, loss: 0.051843930035829544
2023-01-21 09:18:12.503519: step: 2068/527, loss: 0.106096550822258
2023-01-21 09:18:13.635742: step: 2072/527, loss: 0.4491764307022095
2023-01-21 09:18:14.740670: step: 2076/527, loss: 0.18740062415599823
2023-01-21 09:18:15.854732: step: 2080/527, loss: 0.5244362354278564
2023-01-21 09:18:16.971615: step: 2084/527, loss: 0.20706963539123535
2023-01-21 09:18:18.079182: step: 2088/527, loss: 0.09140148013830185
2023-01-21 09:18:19.181196: step: 2092/527, loss: 0.1178104430437088
2023-01-21 09:18:20.259710: step: 2096/527, loss: 0.13795657455921173
2023-01-21 09:18:21.378412: step: 2100/527, loss: 0.3109671473503113
2023-01-21 09:18:22.499354: step: 2104/527, loss: 0.10626889020204544
2023-01-21 09:18:23.612900: step: 2108/527, loss: 0.2091844081878662
==================================================
Loss: 0.377
--------------------
Dev: {'event': {'p': 0.5508637236084453, 'r': 0.7643142476697736, 'f1': 0.6402677077523703}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}
Test: {'event': {'p': 0.6072245084590764, 'r': 0.7588571428571429, 'f1': 0.6746253492506985}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}
Chinese: {'event': {'p': 0.5783132530120482, 'r': 0.8888888888888888, 'f1': 0.7007299270072992}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}
Korean: {'event': {'p': 0.6078431372549019, 'r': 0.49206349206349204, 'f1': 0.543859649122807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}
Russian: {'event': {'p': 0.45, 'r': 0.5, 'f1': 0.4736842105263158}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Chinese: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Chinese: {'event': {'p': 0.5897435897435898, 'r': 0.8518518518518519, 'f1': 0.6969696969696971}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
--------------------
Eng Dev for Korean: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Korean: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 09:19:05.493680: step: 4/527, loss: 0.022249173372983932
2023-01-21 09:19:06.631960: step: 8/527, loss: 0.052853774279356
2023-01-21 09:19:07.780988: step: 12/527, loss: 0.7081000804901123
2023-01-21 09:19:08.903133: step: 16/527, loss: 0.74346524477005
2023-01-21 09:19:10.019583: step: 20/527, loss: 0.14211177825927734
2023-01-21 09:19:11.137345: step: 24/527, loss: 0.2531282305717468
2023-01-21 09:19:12.270643: step: 28/527, loss: 0.30989572405815125
2023-01-21 09:19:13.375809: step: 32/527, loss: 0.5147432088851929
2023-01-21 09:19:14.492207: step: 36/527, loss: 0.3979882299900055
2023-01-21 09:19:15.626446: step: 40/527, loss: 0.4900001287460327
2023-01-21 09:19:16.774741: step: 44/527, loss: 0.3972986340522766
2023-01-21 09:19:17.918060: step: 48/527, loss: 0.18098661303520203
2023-01-21 09:19:19.026052: step: 52/527, loss: 0.2125433385372162
2023-01-21 09:19:20.157201: step: 56/527, loss: 0.1582840532064438
2023-01-21 09:19:21.315495: step: 60/527, loss: 0.08227815479040146
2023-01-21 09:19:22.437175: step: 64/527, loss: 0.2074366956949234
2023-01-21 09:19:23.531864: step: 68/527, loss: 0.8821884393692017
2023-01-21 09:19:24.671904: step: 72/527, loss: 0.4126943051815033
2023-01-21 09:19:25.792080: step: 76/527, loss: 0.1910543441772461
2023-01-21 09:19:26.920775: step: 80/527, loss: 0.07960166782140732
2023-01-21 09:19:28.055905: step: 84/527, loss: 1.8199048042297363
2023-01-21 09:19:29.164352: step: 88/527, loss: 0.19938337802886963
2023-01-21 09:19:30.284262: step: 92/527, loss: 0.27888283133506775
2023-01-21 09:19:31.395213: step: 96/527, loss: 0.024596024304628372
2023-01-21 09:19:32.507305: step: 100/527, loss: 0.6486958265304565
2023-01-21 09:19:33.613626: step: 104/527, loss: 0.5802132487297058
2023-01-21 09:19:34.722862: step: 108/527, loss: 0.06379161030054092
2023-01-21 09:19:35.816938: step: 112/527, loss: 0.12056417763233185
2023-01-21 09:19:36.952829: step: 116/527, loss: 0.03934536129236221
2023-01-21 09:19:38.047685: step: 120/527, loss: 0.12554463744163513
2023-01-21 09:19:39.164306: step: 124/527, loss: 0.36919790506362915
2023-01-21 09:19:40.276710: step: 128/527, loss: 0.2116708904504776
2023-01-21 09:19:41.391187: step: 132/527, loss: 0.07913322746753693
2023-01-21 09:19:42.489455: step: 136/527, loss: 0.03686084970831871
2023-01-21 09:19:43.616436: step: 140/527, loss: 0.06286859512329102
2023-01-21 09:19:44.771677: step: 144/527, loss: 0.16066904366016388
2023-01-21 09:19:45.860673: step: 148/527, loss: 0.013411570340394974
2023-01-21 09:19:46.973695: step: 152/527, loss: 0.05042882263660431
2023-01-21 09:19:48.108411: step: 156/527, loss: 0.12395496666431427
2023-01-21 09:19:49.217323: step: 160/527, loss: 0.04236145317554474
2023-01-21 09:19:50.348212: step: 164/527, loss: 0.2898007929325104
2023-01-21 09:19:51.496364: step: 168/527, loss: 0.12594422698020935
2023-01-21 09:19:52.590103: step: 172/527, loss: 0.06793558597564697
2023-01-21 09:19:53.701725: step: 176/527, loss: 0.15639400482177734
2023-01-21 09:19:54.797681: step: 180/527, loss: 0.0663963332772255
2023-01-21 09:19:55.915204: step: 184/527, loss: 0.13980217278003693
2023-01-21 09:19:57.040974: step: 188/527, loss: 0.07688732445240021
2023-01-21 09:19:58.150837: step: 192/527, loss: 0.33318862318992615
2023-01-21 09:19:59.285294: step: 196/527, loss: 0.5603718161582947
2023-01-21 09:20:00.406328: step: 200/527, loss: 0.11913757771253586
2023-01-21 09:20:01.521854: step: 204/527, loss: 0.22479209303855896
2023-01-21 09:20:02.641015: step: 208/527, loss: 0.04867896810173988
2023-01-21 09:20:03.765434: step: 212/527, loss: 0.2195838838815689
2023-01-21 09:20:04.891388: step: 216/527, loss: 0.29979240894317627
2023-01-21 09:20:06.008670: step: 220/527, loss: 0.07239756733179092
2023-01-21 09:20:07.133758: step: 224/527, loss: 0.13301487267017365
2023-01-21 09:20:08.270020: step: 228/527, loss: 0.05738077312707901
2023-01-21 09:20:09.385296: step: 232/527, loss: 0.17286130785942078
2023-01-21 09:20:10.522728: step: 236/527, loss: 0.0887751579284668
2023-01-21 09:20:11.634393: step: 240/527, loss: 0.15568368136882782
2023-01-21 09:20:12.795346: step: 244/527, loss: 0.13179701566696167
2023-01-21 09:20:13.900915: step: 248/527, loss: 0.253068208694458
2023-01-21 09:20:15.009199: step: 252/527, loss: 0.01938309706747532
2023-01-21 09:20:16.103310: step: 256/527, loss: 0.3003976047039032
2023-01-21 09:20:17.229798: step: 260/527, loss: 0.07840452343225479
2023-01-21 09:20:18.335222: step: 264/527, loss: 0.22693189978599548
2023-01-21 09:20:19.459434: step: 268/527, loss: 0.13255755603313446
2023-01-21 09:20:20.569477: step: 272/527, loss: 0.08591623604297638
2023-01-21 09:20:21.704606: step: 276/527, loss: 0.1224118247628212
2023-01-21 09:20:22.800206: step: 280/527, loss: 0.2676970362663269
2023-01-21 09:20:23.924087: step: 284/527, loss: 0.8498929142951965
2023-01-21 09:20:25.028599: step: 288/527, loss: 0.17844171822071075
2023-01-21 09:20:26.138823: step: 292/527, loss: 0.17654944956302643
2023-01-21 09:20:27.288329: step: 296/527, loss: 0.27493420243263245
2023-01-21 09:20:28.443851: step: 300/527, loss: 0.5626609921455383
2023-01-21 09:20:29.546724: step: 304/527, loss: 0.10025587677955627
2023-01-21 09:20:30.681521: step: 308/527, loss: 0.2725190222263336
2023-01-21 09:20:31.800722: step: 312/527, loss: 0.13540850579738617
2023-01-21 09:20:32.942445: step: 316/527, loss: 0.3991667926311493
2023-01-21 09:20:34.030044: step: 320/527, loss: 0.2279730886220932
2023-01-21 09:20:35.151063: step: 324/527, loss: 0.10485020279884338
2023-01-21 09:20:36.270245: step: 328/527, loss: 0.2993044853210449
2023-01-21 09:20:37.383830: step: 332/527, loss: 0.44559991359710693
2023-01-21 09:20:38.469156: step: 336/527, loss: 0.8305598497390747
2023-01-21 09:20:39.588644: step: 340/527, loss: 0.024629592895507812
2023-01-21 09:20:40.680050: step: 344/527, loss: 0.09807997196912766
2023-01-21 09:20:41.798755: step: 348/527, loss: 0.10104046016931534
2023-01-21 09:20:42.904112: step: 352/527, loss: 0.47202205657958984
2023-01-21 09:20:44.027188: step: 356/527, loss: 0.08598623424768448
2023-01-21 09:20:45.169379: step: 360/527, loss: 0.05252885818481445
2023-01-21 09:20:46.311856: step: 364/527, loss: 0.14439114928245544
2023-01-21 09:20:47.444546: step: 368/527, loss: 0.23421978950500488
2023-01-21 09:20:48.586211: step: 372/527, loss: 0.02374706231057644
2023-01-21 09:20:49.703998: step: 376/527, loss: 0.38174647092819214
2023-01-21 09:20:50.811008: step: 380/527, loss: 0.12252788245677948
2023-01-21 09:20:51.928892: step: 384/527, loss: 0.0832308754324913
2023-01-21 09:20:53.038942: step: 388/527, loss: 0.4216980040073395
2023-01-21 09:20:54.169214: step: 392/527, loss: 0.10022832453250885
2023-01-21 09:20:55.265979: step: 396/527, loss: 0.4777892231941223
2023-01-21 09:20:56.373975: step: 400/527, loss: 0.14322252571582794
2023-01-21 09:20:57.492890: step: 404/527, loss: 0.36082419753074646
2023-01-21 09:20:58.613844: step: 408/527, loss: 0.11528797447681427
2023-01-21 09:20:59.711927: step: 412/527, loss: 0.09323558956384659
2023-01-21 09:21:00.827446: step: 416/527, loss: 0.2644132673740387
2023-01-21 09:21:01.936545: step: 420/527, loss: 0.14598503708839417
2023-01-21 09:21:03.025837: step: 424/527, loss: 0.05398502200841904
2023-01-21 09:21:04.114441: step: 428/527, loss: 0.02928008884191513
2023-01-21 09:21:05.226703: step: 432/527, loss: 0.1828937530517578
2023-01-21 09:21:06.360077: step: 436/527, loss: 0.06944485008716583
2023-01-21 09:21:07.487466: step: 440/527, loss: 0.06689062714576721
2023-01-21 09:21:08.602143: step: 444/527, loss: 0.3215312957763672
2023-01-21 09:21:09.724167: step: 448/527, loss: 1.0646305084228516
2023-01-21 09:21:10.836661: step: 452/527, loss: 0.09793760627508163
2023-01-21 09:21:11.942882: step: 456/527, loss: 0.037320468574762344
2023-01-21 09:21:13.066947: step: 460/527, loss: 0.03612460941076279
2023-01-21 09:21:14.197808: step: 464/527, loss: 0.4685177803039551
2023-01-21 09:21:15.328276: step: 468/527, loss: 0.21176129579544067
2023-01-21 09:21:16.440322: step: 472/527, loss: 0.15110769867897034
2023-01-21 09:21:17.538636: step: 476/527, loss: 0.03734846040606499
2023-01-21 09:21:18.657848: step: 480/527, loss: 0.11501837521791458
2023-01-21 09:21:19.744621: step: 484/527, loss: 0.12085433304309845
2023-01-21 09:21:20.858180: step: 488/527, loss: 0.184031680226326
2023-01-21 09:21:21.954515: step: 492/527, loss: 0.22017021477222443
2023-01-21 09:21:23.070303: step: 496/527, loss: 0.0701756477355957
2023-01-21 09:21:24.180854: step: 500/527, loss: 0.34339240193367004
2023-01-21 09:21:25.301476: step: 504/527, loss: 0.12254400551319122
2023-01-21 09:21:26.429608: step: 508/527, loss: 0.1262405514717102
2023-01-21 09:21:27.548510: step: 512/527, loss: 0.2515473961830139
2023-01-21 09:21:28.676095: step: 516/527, loss: 0.15226033329963684
2023-01-21 09:21:29.797465: step: 520/527, loss: 0.29322385787963867
2023-01-21 09:21:30.903975: step: 524/527, loss: 0.04788990318775177
2023-01-21 09:21:32.041313: step: 528/527, loss: 0.11253318190574646
2023-01-21 09:21:33.157785: step: 532/527, loss: 0.3688744306564331
2023-01-21 09:21:34.274950: step: 536/527, loss: 0.007495689671486616
2023-01-21 09:21:35.378753: step: 540/527, loss: 1.9881023168563843
2023-01-21 09:21:36.524034: step: 544/527, loss: 0.25205785036087036
2023-01-21 09:21:37.641619: step: 548/527, loss: 0.5413957834243774
2023-01-21 09:21:38.740236: step: 552/527, loss: 0.03450331836938858
2023-01-21 09:21:39.869196: step: 556/527, loss: 0.05203437805175781
2023-01-21 09:21:41.005506: step: 560/527, loss: 0.15648341178894043
2023-01-21 09:21:42.099581: step: 564/527, loss: 0.2538606822490692
2023-01-21 09:21:43.189297: step: 568/527, loss: 0.5688015222549438
2023-01-21 09:21:44.295677: step: 572/527, loss: 0.05319194868206978
2023-01-21 09:21:45.416308: step: 576/527, loss: 0.08528414368629456
2023-01-21 09:21:46.533431: step: 580/527, loss: 0.018718242645263672
2023-01-21 09:21:47.643072: step: 584/527, loss: 0.18431530892848969
2023-01-21 09:21:48.741705: step: 588/527, loss: 0.23473891615867615
2023-01-21 09:21:49.849339: step: 592/527, loss: 0.31978341937065125
2023-01-21 09:21:50.983447: step: 596/527, loss: 0.23661017417907715
2023-01-21 09:21:52.117343: step: 600/527, loss: 0.06694173812866211
2023-01-21 09:21:53.255533: step: 604/527, loss: 0.09180283546447754
2023-01-21 09:21:54.357036: step: 608/527, loss: 0.27530938386917114
2023-01-21 09:21:55.468981: step: 612/527, loss: 0.084027960896492
2023-01-21 09:21:56.566790: step: 616/527, loss: 0.49331170320510864
2023-01-21 09:21:57.684468: step: 620/527, loss: 0.0631282776594162
2023-01-21 09:21:58.803091: step: 624/527, loss: 0.33109167218208313
2023-01-21 09:21:59.962663: step: 628/527, loss: 0.09209757298231125
2023-01-21 09:22:01.073506: step: 632/527, loss: 0.06318364292383194
2023-01-21 09:22:02.190933: step: 636/527, loss: 0.07507705688476562
2023-01-21 09:22:03.295042: step: 640/527, loss: 0.07749853283166885
2023-01-21 09:22:04.395225: step: 644/527, loss: 0.26230868697166443
2023-01-21 09:22:05.503041: step: 648/527, loss: 0.016105175018310547
2023-01-21 09:22:06.608994: step: 652/527, loss: 0.15562590956687927
2023-01-21 09:22:07.710484: step: 656/527, loss: 0.11723342537879944
2023-01-21 09:22:08.868892: step: 660/527, loss: 0.09571543335914612
2023-01-21 09:22:09.983990: step: 664/527, loss: 0.39289337396621704
2023-01-21 09:22:11.124371: step: 668/527, loss: 0.21759510040283203
2023-01-21 09:22:12.260699: step: 672/527, loss: 0.32942602038383484
2023-01-21 09:22:13.379467: step: 676/527, loss: 0.3686848282814026
2023-01-21 09:22:14.493902: step: 680/527, loss: 0.6266034245491028
2023-01-21 09:22:15.609367: step: 684/527, loss: 0.5773298144340515
2023-01-21 09:22:16.723191: step: 688/527, loss: 0.17440158128738403
2023-01-21 09:22:17.837244: step: 692/527, loss: 0.4018300771713257
2023-01-21 09:22:18.961161: step: 696/527, loss: 0.7968790531158447
2023-01-21 09:22:20.118504: step: 700/527, loss: 1.1410709619522095
2023-01-21 09:22:21.248870: step: 704/527, loss: 0.7502236366271973
2023-01-21 09:22:22.366833: step: 708/527, loss: 0.4125533699989319
2023-01-21 09:22:23.483666: step: 712/527, loss: 0.19168511033058167
2023-01-21 09:22:24.614477: step: 716/527, loss: 0.3345426619052887
2023-01-21 09:22:25.721366: step: 720/527, loss: 0.4023723006248474
2023-01-21 09:22:26.849998: step: 724/527, loss: 0.4025490880012512
2023-01-21 09:22:27.967280: step: 728/527, loss: 0.07412385940551758
2023-01-21 09:22:29.077380: step: 732/527, loss: 1.1686017513275146
2023-01-21 09:22:30.198473: step: 736/527, loss: 0.10230188816785812
2023-01-21 09:22:31.294526: step: 740/527, loss: 0.11895523220300674
2023-01-21 09:22:32.390530: step: 744/527, loss: 0.11851787567138672
2023-01-21 09:22:33.518721: step: 748/527, loss: 0.09862575680017471
2023-01-21 09:22:34.624988: step: 752/527, loss: 0.14316025376319885
2023-01-21 09:22:35.743805: step: 756/527, loss: 0.026618385687470436
2023-01-21 09:22:36.833337: step: 760/527, loss: 0.10191135108470917
2023-01-21 09:22:37.927120: step: 764/527, loss: 0.06429853290319443
2023-01-21 09:22:39.018026: step: 768/527, loss: 0.10434617847204208
2023-01-21 09:22:40.131767: step: 772/527, loss: 0.4852098524570465
2023-01-21 09:22:41.268297: step: 776/527, loss: 0.12042541801929474
2023-01-21 09:22:42.366729: step: 780/527, loss: 0.3627198338508606
2023-01-21 09:22:43.468810: step: 784/527, loss: 0.06677771359682083
2023-01-21 09:22:44.589704: step: 788/527, loss: 0.07106852531433105
2023-01-21 09:22:45.677546: step: 792/527, loss: 0.24752560257911682
2023-01-21 09:22:46.827465: step: 796/527, loss: 0.31318798661231995
2023-01-21 09:22:47.965335: step: 800/527, loss: 0.8733689785003662
2023-01-21 09:22:49.099821: step: 804/527, loss: 0.0827760249376297
2023-01-21 09:22:50.236219: step: 808/527, loss: 0.19814014434814453
2023-01-21 09:22:51.382588: step: 812/527, loss: 0.7050227522850037
2023-01-21 09:22:52.495111: step: 816/527, loss: 0.0867125540971756
2023-01-21 09:22:53.610881: step: 820/527, loss: 0.06167206913232803
2023-01-21 09:22:54.735202: step: 824/527, loss: 0.10035638511180878
2023-01-21 09:22:55.848933: step: 828/527, loss: 0.3124014139175415
2023-01-21 09:22:56.950500: step: 832/527, loss: 0.09035825729370117
2023-01-21 09:22:58.054669: step: 836/527, loss: 0.19395585358142853
2023-01-21 09:22:59.191471: step: 840/527, loss: 0.26939234137535095
2023-01-21 09:23:00.288860: step: 844/527, loss: 0.22080013155937195
2023-01-21 09:23:01.387159: step: 848/527, loss: 0.1174074113368988
2023-01-21 09:23:02.515273: step: 852/527, loss: 0.20611357688903809
2023-01-21 09:23:03.628685: step: 856/527, loss: 0.2556767463684082
2023-01-21 09:23:04.745870: step: 860/527, loss: 0.2072371393442154
2023-01-21 09:23:05.837156: step: 864/527, loss: 0.14181672036647797
2023-01-21 09:23:06.952577: step: 868/527, loss: 0.01876373216509819
2023-01-21 09:23:08.058215: step: 872/527, loss: 0.06636762619018555
2023-01-21 09:23:09.170865: step: 876/527, loss: 0.3427914083003998
2023-01-21 09:23:10.296833: step: 880/527, loss: 0.3641493618488312
2023-01-21 09:23:11.397451: step: 884/527, loss: 0.10763339698314667
2023-01-21 09:23:12.488108: step: 888/527, loss: 0.024660492315888405
2023-01-21 09:23:13.602755: step: 892/527, loss: 0.0815097838640213
2023-01-21 09:23:14.726910: step: 896/527, loss: 0.9242773056030273
2023-01-21 09:23:15.813655: step: 900/527, loss: 0.17459973692893982
2023-01-21 09:23:16.929601: step: 904/527, loss: 0.15603943169116974
2023-01-21 09:23:18.022646: step: 908/527, loss: 0.10006008297204971
2023-01-21 09:23:19.107114: step: 912/527, loss: 0.30490848422050476
2023-01-21 09:23:20.252313: step: 916/527, loss: 0.6718155145645142
2023-01-21 09:23:21.387122: step: 920/527, loss: 0.2529800832271576
2023-01-21 09:23:22.495544: step: 924/527, loss: 0.14264421164989471
2023-01-21 09:23:23.594680: step: 928/527, loss: 0.14165306091308594
2023-01-21 09:23:24.723593: step: 932/527, loss: 0.9555657505989075
2023-01-21 09:23:25.808888: step: 936/527, loss: 0.02001338079571724
2023-01-21 09:23:26.938417: step: 940/527, loss: 0.09285005927085876
2023-01-21 09:23:28.065801: step: 944/527, loss: 0.22490891814231873
2023-01-21 09:23:29.183820: step: 948/527, loss: 0.25561201572418213
2023-01-21 09:23:30.285923: step: 952/527, loss: 0.1803903579711914
2023-01-21 09:23:31.387466: step: 956/527, loss: 0.05386696010828018
2023-01-21 09:23:32.514680: step: 960/527, loss: 0.09731750190258026
2023-01-21 09:23:33.623435: step: 964/527, loss: 0.9887893199920654
2023-01-21 09:23:34.738313: step: 968/527, loss: 0.09517116844654083
2023-01-21 09:23:35.855344: step: 972/527, loss: 1.279137134552002
2023-01-21 09:23:36.985317: step: 976/527, loss: 0.05282296985387802
2023-01-21 09:23:38.074334: step: 980/527, loss: 0.09685230255126953
2023-01-21 09:23:39.171863: step: 984/527, loss: 0.5726767778396606
2023-01-21 09:23:40.318499: step: 988/527, loss: 0.052752211689949036
2023-01-21 09:23:41.431646: step: 992/527, loss: 0.21002750098705292
2023-01-21 09:23:42.557847: step: 996/527, loss: 0.18055109679698944
2023-01-21 09:23:43.704286: step: 1000/527, loss: 0.02756948582828045
2023-01-21 09:23:44.835064: step: 1004/527, loss: 0.11835814267396927
2023-01-21 09:23:45.960566: step: 1008/527, loss: 0.019714046269655228
2023-01-21 09:23:47.044077: step: 1012/527, loss: 0.17755194008350372
2023-01-21 09:23:48.149506: step: 1016/527, loss: 1.000170350074768
2023-01-21 09:23:49.292865: step: 1020/527, loss: 0.2504728436470032
2023-01-21 09:23:50.427745: step: 1024/527, loss: 1.7355129718780518
2023-01-21 09:23:51.536160: step: 1028/527, loss: 0.21115180850028992
2023-01-21 09:23:52.669805: step: 1032/527, loss: 0.10982055217027664
2023-01-21 09:23:53.858152: step: 1036/527, loss: 0.2593066990375519
2023-01-21 09:23:54.961045: step: 1040/527, loss: 0.08927793800830841
2023-01-21 09:23:56.075553: step: 1044/527, loss: 0.19734343886375427
2023-01-21 09:23:57.208328: step: 1048/527, loss: 0.04375801235437393
2023-01-21 09:23:58.351181: step: 1052/527, loss: 0.15594534575939178
2023-01-21 09:23:59.456131: step: 1056/527, loss: 0.2277686595916748
2023-01-21 09:24:00.535422: step: 1060/527, loss: 0.12292556464672089
2023-01-21 09:24:01.658865: step: 1064/527, loss: 0.08499407768249512
2023-01-21 09:24:02.747210: step: 1068/527, loss: 0.07709364593029022
2023-01-21 09:24:03.894174: step: 1072/527, loss: 1.4358896017074585
2023-01-21 09:24:04.987822: step: 1076/527, loss: 0.26507455110549927
2023-01-21 09:24:06.111067: step: 1080/527, loss: 0.15279458463191986
2023-01-21 09:24:07.233828: step: 1084/527, loss: 0.4682307243347168
2023-01-21 09:24:08.341423: step: 1088/527, loss: 0.09165802597999573
2023-01-21 09:24:09.439242: step: 1092/527, loss: 0.03277845308184624
2023-01-21 09:24:10.581008: step: 1096/527, loss: 0.04624834284186363
2023-01-21 09:24:11.680799: step: 1100/527, loss: 0.8685977458953857
2023-01-21 09:24:12.835956: step: 1104/527, loss: 0.22323516011238098
2023-01-21 09:24:13.945216: step: 1108/527, loss: 0.048952389508485794
2023-01-21 09:24:15.103409: step: 1112/527, loss: 0.26619797945022583
2023-01-21 09:24:16.235704: step: 1116/527, loss: 0.08635836094617844
2023-01-21 09:24:17.358136: step: 1120/527, loss: 0.07696514576673508
2023-01-21 09:24:18.475717: step: 1124/527, loss: 0.4653652012348175
2023-01-21 09:24:19.572656: step: 1128/527, loss: 0.4134241044521332
2023-01-21 09:24:20.672765: step: 1132/527, loss: 0.04846487194299698
2023-01-21 09:24:21.790337: step: 1136/527, loss: 0.09378290176391602
2023-01-21 09:24:22.904958: step: 1140/527, loss: 0.22416406869888306
2023-01-21 09:24:23.990961: step: 1144/527, loss: 0.0863683745265007
2023-01-21 09:24:25.101772: step: 1148/527, loss: 0.28976762294769287
2023-01-21 09:24:26.211639: step: 1152/527, loss: 0.10418453067541122
2023-01-21 09:24:27.350044: step: 1156/527, loss: 0.1785154789686203
2023-01-21 09:24:28.458829: step: 1160/527, loss: 0.5722765922546387
2023-01-21 09:24:29.553986: step: 1164/527, loss: 0.03684301674365997
2023-01-21 09:24:30.668097: step: 1168/527, loss: 0.039452649652957916
2023-01-21 09:24:31.762015: step: 1172/527, loss: 0.18124571442604065
2023-01-21 09:24:32.892420: step: 1176/527, loss: 0.07443609833717346
2023-01-21 09:24:34.012230: step: 1180/527, loss: 0.05713377147912979
2023-01-21 09:24:35.136503: step: 1184/527, loss: 0.1508459597826004
2023-01-21 09:24:36.268675: step: 1188/527, loss: 0.17405909299850464
2023-01-21 09:24:37.382434: step: 1192/527, loss: 0.9470130801200867
2023-01-21 09:24:38.497029: step: 1196/527, loss: 0.09310570359230042
2023-01-21 09:24:39.622550: step: 1200/527, loss: 0.1418277770280838
2023-01-21 09:24:40.715254: step: 1204/527, loss: 0.05359811708331108
2023-01-21 09:24:41.830607: step: 1208/527, loss: 0.14250335097312927
2023-01-21 09:24:42.932495: step: 1212/527, loss: 0.03633112832903862
2023-01-21 09:24:44.025638: step: 1216/527, loss: 0.24199900031089783
2023-01-21 09:24:45.133718: step: 1220/527, loss: 0.1411367505788803
2023-01-21 09:24:46.238374: step: 1224/527, loss: 0.21746310591697693
2023-01-21 09:24:47.347309: step: 1228/527, loss: 0.0654749870300293
2023-01-21 09:24:48.449201: step: 1232/527, loss: 0.1332019865512848
2023-01-21 09:24:49.561004: step: 1236/527, loss: 0.5273067355155945
2023-01-21 09:24:50.688700: step: 1240/527, loss: 0.09529409557580948
2023-01-21 09:24:51.809202: step: 1244/527, loss: 0.1311378926038742
2023-01-21 09:24:52.934888: step: 1248/527, loss: 0.9126175045967102
2023-01-21 09:24:54.046870: step: 1252/527, loss: 0.1864130049943924
2023-01-21 09:24:55.138968: step: 1256/527, loss: 0.5510692000389099
2023-01-21 09:24:56.266449: step: 1260/527, loss: 0.17051827907562256
2023-01-21 09:24:57.385550: step: 1264/527, loss: 0.5297226905822754
2023-01-21 09:24:58.486636: step: 1268/527, loss: 0.3993288278579712
2023-01-21 09:24:59.587479: step: 1272/527, loss: 0.20910167694091797
2023-01-21 09:25:00.704791: step: 1276/527, loss: 0.1272641271352768
2023-01-21 09:25:01.805315: step: 1280/527, loss: 0.1559308022260666
2023-01-21 09:25:02.944631: step: 1284/527, loss: 0.16905517876148224
2023-01-21 09:25:04.045279: step: 1288/527, loss: 0.3641207218170166
2023-01-21 09:25:05.200770: step: 1292/527, loss: 0.08693476021289825
2023-01-21 09:25:06.334901: step: 1296/527, loss: 0.3631511628627777
2023-01-21 09:25:07.458698: step: 1300/527, loss: 0.19771808385849
2023-01-21 09:25:08.563746: step: 1304/527, loss: 0.6395736932754517
2023-01-21 09:25:09.682008: step: 1308/527, loss: 0.16837921738624573
2023-01-21 09:25:10.816827: step: 1312/527, loss: 0.24905472993850708
2023-01-21 09:25:11.942688: step: 1316/527, loss: 0.2387489676475525
2023-01-21 09:25:13.078880: step: 1320/527, loss: 0.04274997487664223
2023-01-21 09:25:14.189611: step: 1324/527, loss: 0.09521861374378204
2023-01-21 09:25:15.296545: step: 1328/527, loss: 0.3333342671394348
2023-01-21 09:25:16.416327: step: 1332/527, loss: 0.1949302703142166
2023-01-21 09:25:17.552592: step: 1336/527, loss: 0.12058629840612411
2023-01-21 09:25:18.669594: step: 1340/527, loss: 0.0865846648812294
2023-01-21 09:25:19.794452: step: 1344/527, loss: 0.11176042258739471
2023-01-21 09:25:20.941214: step: 1348/527, loss: 0.3597325086593628
2023-01-21 09:25:22.049774: step: 1352/527, loss: 0.09371452778577805
2023-01-21 09:25:23.173781: step: 1356/527, loss: 0.14018574357032776
2023-01-21 09:25:24.277534: step: 1360/527, loss: 0.13045254349708557
2023-01-21 09:25:25.395951: step: 1364/527, loss: 0.22600993514060974
2023-01-21 09:25:26.518870: step: 1368/527, loss: 0.13538646697998047
2023-01-21 09:25:27.665137: step: 1372/527, loss: 0.13528437912464142
2023-01-21 09:25:28.778664: step: 1376/527, loss: 0.1861620396375656
2023-01-21 09:25:29.889002: step: 1380/527, loss: 0.24410080909729004
2023-01-21 09:25:30.982181: step: 1384/527, loss: 0.11294479668140411
2023-01-21 09:25:32.123223: step: 1388/527, loss: 0.3299015760421753
2023-01-21 09:25:33.242665: step: 1392/527, loss: 0.7020249962806702
2023-01-21 09:25:34.344881: step: 1396/527, loss: 0.1549573540687561
2023-01-21 09:25:35.442805: step: 1400/527, loss: 0.10581937432289124
2023-01-21 09:25:36.562497: step: 1404/527, loss: 0.18154793977737427
2023-01-21 09:25:37.696983: step: 1408/527, loss: 0.10994873195886612
2023-01-21 09:25:38.801337: step: 1412/527, loss: 0.10219564288854599
2023-01-21 09:25:39.914988: step: 1416/527, loss: 0.18357139825820923
2023-01-21 09:25:41.009197: step: 1420/527, loss: 0.1429995596408844
2023-01-21 09:25:42.138659: step: 1424/527, loss: 0.20690345764160156
2023-01-21 09:25:43.249847: step: 1428/527, loss: 0.03540918976068497
2023-01-21 09:25:44.373515: step: 1432/527, loss: 0.6291588544845581
2023-01-21 09:25:45.455765: step: 1436/527, loss: 0.023952938616275787
2023-01-21 09:25:46.583066: step: 1440/527, loss: 0.24245738983154297
2023-01-21 09:25:47.695485: step: 1444/527, loss: 0.3907793164253235
2023-01-21 09:25:48.798840: step: 1448/527, loss: 0.37806135416030884
2023-01-21 09:25:49.908130: step: 1452/527, loss: 0.08191061019897461
2023-01-21 09:25:50.994628: step: 1456/527, loss: 0.027959156781435013
2023-01-21 09:25:52.063136: step: 1460/527, loss: 1.3866853713989258
2023-01-21 09:25:53.171439: step: 1464/527, loss: 0.023647405207157135
2023-01-21 09:25:54.295623: step: 1468/527, loss: 0.4275180697441101
2023-01-21 09:25:55.424476: step: 1472/527, loss: 0.2591736614704132
2023-01-21 09:25:56.554067: step: 1476/527, loss: 0.20774689316749573
2023-01-21 09:25:57.667057: step: 1480/527, loss: 0.11025485396385193
2023-01-21 09:25:58.763295: step: 1484/527, loss: 0.14052283763885498
2023-01-21 09:25:59.873975: step: 1488/527, loss: 0.24376507103443146
2023-01-21 09:26:01.002414: step: 1492/527, loss: 0.15917068719863892
2023-01-21 09:26:02.093951: step: 1496/527, loss: 0.2980186343193054
2023-01-21 09:26:03.190201: step: 1500/527, loss: 0.05484504997730255
2023-01-21 09:26:04.328082: step: 1504/527, loss: 0.20234127342700958
2023-01-21 09:26:05.422264: step: 1508/527, loss: 0.1779327392578125
2023-01-21 09:26:06.558717: step: 1512/527, loss: 1.5668984651565552
2023-01-21 09:26:07.687100: step: 1516/527, loss: 0.20066681504249573
2023-01-21 09:26:08.834685: step: 1520/527, loss: 0.5626883506774902
2023-01-21 09:26:09.945435: step: 1524/527, loss: 0.10938291251659393
2023-01-21 09:26:11.039597: step: 1528/527, loss: 0.05707111209630966
2023-01-21 09:26:12.143497: step: 1532/527, loss: 0.0390351302921772
2023-01-21 09:26:13.261326: step: 1536/527, loss: 0.36508435010910034
2023-01-21 09:26:14.393329: step: 1540/527, loss: 2.013167142868042
2023-01-21 09:26:15.506775: step: 1544/527, loss: 0.5956695079803467
2023-01-21 09:26:16.585300: step: 1548/527, loss: 0.033182717859745026
2023-01-21 09:26:17.688414: step: 1552/527, loss: 0.24339599907398224
2023-01-21 09:26:18.826050: step: 1556/527, loss: 0.18161793053150177
2023-01-21 09:26:19.914166: step: 1560/527, loss: 0.0639905035495758
2023-01-21 09:26:21.017995: step: 1564/527, loss: 0.17906102538108826
2023-01-21 09:26:22.125170: step: 1568/527, loss: 0.1424793303012848
2023-01-21 09:26:23.218198: step: 1572/527, loss: 0.06548518687486649
2023-01-21 09:26:24.360061: step: 1576/527, loss: 0.47618475556373596
2023-01-21 09:26:25.475864: step: 1580/527, loss: 0.27528414130210876
2023-01-21 09:26:26.605111: step: 1584/527, loss: 0.10735063999891281
2023-01-21 09:26:27.709067: step: 1588/527, loss: 0.08886084705591202
2023-01-21 09:26:28.833855: step: 1592/527, loss: 0.19100405275821686
2023-01-21 09:26:29.948131: step: 1596/527, loss: 0.05273895710706711
2023-01-21 09:26:31.067903: step: 1600/527, loss: 0.10789547115564346
2023-01-21 09:26:32.189053: step: 1604/527, loss: 0.2897655963897705
2023-01-21 09:26:33.296092: step: 1608/527, loss: 0.12071742862462997
2023-01-21 09:26:34.415469: step: 1612/527, loss: 0.0781133696436882
2023-01-21 09:26:35.515144: step: 1616/527, loss: 0.29869768023490906
2023-01-21 09:26:36.639228: step: 1620/527, loss: 0.21792864799499512
2023-01-21 09:26:37.764681: step: 1624/527, loss: 0.1269351989030838
2023-01-21 09:26:38.876962: step: 1628/527, loss: 0.07096557319164276
2023-01-21 09:26:39.976532: step: 1632/527, loss: 0.017712164670228958
2023-01-21 09:26:41.086980: step: 1636/527, loss: 0.2003255933523178
2023-01-21 09:26:42.181954: step: 1640/527, loss: 0.5588818788528442
2023-01-21 09:26:43.307791: step: 1644/527, loss: 0.11722178757190704
2023-01-21 09:26:44.429614: step: 1648/527, loss: 0.1632780134677887
2023-01-21 09:26:45.592676: step: 1652/527, loss: 0.15093430876731873
2023-01-21 09:26:46.707790: step: 1656/527, loss: 1.1809275150299072
2023-01-21 09:26:47.835368: step: 1660/527, loss: 0.12432112544775009
2023-01-21 09:26:48.957106: step: 1664/527, loss: 0.02001657523214817
2023-01-21 09:26:50.102652: step: 1668/527, loss: 0.21676737070083618
2023-01-21 09:26:51.213134: step: 1672/527, loss: 0.1052466407418251
2023-01-21 09:26:52.313481: step: 1676/527, loss: 0.26912203431129456
2023-01-21 09:26:53.422682: step: 1680/527, loss: 0.042536139488220215
2023-01-21 09:26:54.529958: step: 1684/527, loss: 0.23775464296340942
2023-01-21 09:26:55.635127: step: 1688/527, loss: 0.07776384800672531
2023-01-21 09:26:56.842781: step: 1692/527, loss: 0.6749762296676636
2023-01-21 09:26:57.948161: step: 1696/527, loss: 0.03224661573767662
2023-01-21 09:26:59.055207: step: 1700/527, loss: 0.017928026616573334
2023-01-21 09:27:00.168753: step: 1704/527, loss: 0.06147823482751846
2023-01-21 09:27:01.275056: step: 1708/527, loss: 0.05404110252857208
2023-01-21 09:27:02.415668: step: 1712/527, loss: 0.17946463823318481
2023-01-21 09:27:03.544228: step: 1716/527, loss: 0.11328163743019104
2023-01-21 09:27:04.637036: step: 1720/527, loss: 0.11372726410627365
2023-01-21 09:27:05.795130: step: 1724/527, loss: 0.08207159489393234
2023-01-21 09:27:06.897894: step: 1728/527, loss: 0.13389606773853302
2023-01-21 09:27:08.025572: step: 1732/527, loss: 0.11606454849243164
2023-01-21 09:27:09.162802: step: 1736/527, loss: 0.07518234848976135
2023-01-21 09:27:10.261775: step: 1740/527, loss: 0.5552347898483276
2023-01-21 09:27:11.352360: step: 1744/527, loss: 0.07792038470506668
2023-01-21 09:27:12.474066: step: 1748/527, loss: 0.454611212015152
2023-01-21 09:27:13.576446: step: 1752/527, loss: 3.4578685760498047
2023-01-21 09:27:14.660890: step: 1756/527, loss: 0.10688943415880203
2023-01-21 09:27:15.769722: step: 1760/527, loss: 0.2683517336845398
2023-01-21 09:27:16.876896: step: 1764/527, loss: 0.11338291317224503
2023-01-21 09:27:17.998421: step: 1768/527, loss: 0.582638144493103
2023-01-21 09:27:19.105081: step: 1772/527, loss: 0.2119365781545639
2023-01-21 09:27:20.236886: step: 1776/527, loss: 0.013252854347229004
2023-01-21 09:27:21.326567: step: 1780/527, loss: 0.14856967329978943
2023-01-21 09:27:22.433563: step: 1784/527, loss: 0.06847252696752548
2023-01-21 09:27:23.538495: step: 1788/527, loss: 0.15705688297748566
2023-01-21 09:27:24.642277: step: 1792/527, loss: 0.5870369672775269
2023-01-21 09:27:25.772351: step: 1796/527, loss: 0.09677582234144211
2023-01-21 09:27:26.875653: step: 1800/527, loss: 0.03032945841550827
2023-01-21 09:27:28.025696: step: 1804/527, loss: 0.10984992980957031
2023-01-21 09:27:29.140659: step: 1808/527, loss: 0.09253750741481781
2023-01-21 09:27:30.232631: step: 1812/527, loss: 0.045008473098278046
2023-01-21 09:27:31.321026: step: 1816/527, loss: 0.07884550094604492
2023-01-21 09:27:32.438420: step: 1820/527, loss: 0.15091238915920258
2023-01-21 09:27:33.552326: step: 1824/527, loss: 0.9885495901107788
2023-01-21 09:27:34.648074: step: 1828/527, loss: 0.5015361905097961
2023-01-21 09:27:35.758665: step: 1832/527, loss: 0.027435969561338425
2023-01-21 09:27:36.873809: step: 1836/527, loss: 0.33904364705085754
2023-01-21 09:27:38.023625: step: 1840/527, loss: 0.17369243502616882
2023-01-21 09:27:39.154083: step: 1844/527, loss: 0.3427468240261078
2023-01-21 09:27:40.303926: step: 1848/527, loss: 0.076171875
2023-01-21 09:27:41.423940: step: 1852/527, loss: 1.4057501554489136
2023-01-21 09:27:42.536828: step: 1856/527, loss: 0.345808207988739
2023-01-21 09:27:43.673579: step: 1860/527, loss: 0.32316648960113525
2023-01-21 09:27:44.786872: step: 1864/527, loss: 0.15814723074436188
2023-01-21 09:27:45.912084: step: 1868/527, loss: 0.1599147915840149
2023-01-21 09:27:47.028378: step: 1872/527, loss: 0.1452324390411377
2023-01-21 09:27:48.151040: step: 1876/527, loss: 0.04798183590173721
2023-01-21 09:27:49.225041: step: 1880/527, loss: 0.7147918939590454
2023-01-21 09:27:50.338899: step: 1884/527, loss: 0.14694365859031677
2023-01-21 09:27:51.458066: step: 1888/527, loss: 0.11013995110988617
2023-01-21 09:27:52.569183: step: 1892/527, loss: 0.32118573784828186
2023-01-21 09:27:53.692073: step: 1896/527, loss: 0.14987125992774963
2023-01-21 09:27:54.809877: step: 1900/527, loss: 0.37460213899612427
2023-01-21 09:27:55.963189: step: 1904/527, loss: 0.47931337356567383
2023-01-21 09:27:57.104914: step: 1908/527, loss: 0.16490508615970612
2023-01-21 09:27:58.226747: step: 1912/527, loss: 0.3935590982437134
2023-01-21 09:27:59.364849: step: 1916/527, loss: 0.14565859735012054
2023-01-21 09:28:00.471982: step: 1920/527, loss: 0.02524557150900364
2023-01-21 09:28:01.564045: step: 1924/527, loss: 0.041150666773319244
2023-01-21 09:28:02.655558: step: 1928/527, loss: 0.08393120765686035
2023-01-21 09:28:03.774255: step: 1932/527, loss: 0.40841445326805115
2023-01-21 09:28:04.944677: step: 1936/527, loss: 0.04874372482299805
2023-01-21 09:28:06.054736: step: 1940/527, loss: 0.7660877108573914
2023-01-21 09:28:07.172769: step: 1944/527, loss: 0.04994244873523712
2023-01-21 09:28:08.232312: step: 1948/527, loss: 0.039541102945804596
2023-01-21 09:28:09.322350: step: 1952/527, loss: 0.8475232124328613
2023-01-21 09:28:10.467746: step: 1956/527, loss: 0.3495897352695465
2023-01-21 09:28:11.557893: step: 1960/527, loss: 0.23900824785232544
2023-01-21 09:28:12.646681: step: 1964/527, loss: 0.2032717615365982
2023-01-21 09:28:13.811641: step: 1968/527, loss: 0.13190627098083496
2023-01-21 09:28:14.938960: step: 1972/527, loss: 0.8238687515258789
2023-01-21 09:28:16.060892: step: 1976/527, loss: 0.27192696928977966
2023-01-21 09:28:17.184592: step: 1980/527, loss: 0.2585316300392151
2023-01-21 09:28:18.298055: step: 1984/527, loss: 0.14649590849876404
2023-01-21 09:28:19.411655: step: 1988/527, loss: 0.08172807097434998
2023-01-21 09:28:20.546722: step: 1992/527, loss: 5.662846565246582
2023-01-21 09:28:21.657389: step: 1996/527, loss: 0.1646881401538849
2023-01-21 09:28:22.765569: step: 2000/527, loss: 0.1943332850933075
2023-01-21 09:28:23.914644: step: 2004/527, loss: 0.08647593855857849
2023-01-21 09:28:25.064038: step: 2008/527, loss: 0.11360268294811249
2023-01-21 09:28:26.211414: step: 2012/527, loss: 0.6866371035575867
2023-01-21 09:28:27.326590: step: 2016/527, loss: 0.01580023765563965
2023-01-21 09:28:28.426626: step: 2020/527, loss: 0.3055339753627777
2023-01-21 09:28:29.508455: step: 2024/527, loss: 0.3335486352443695
2023-01-21 09:28:30.615140: step: 2028/527, loss: 0.14706259965896606
2023-01-21 09:28:31.741310: step: 2032/527, loss: 0.31563299894332886
2023-01-21 09:28:32.850862: step: 2036/527, loss: 0.07901842892169952
2023-01-21 09:28:33.955027: step: 2040/527, loss: 0.48686444759368896
2023-01-21 09:28:35.068781: step: 2044/527, loss: 0.06225104629993439
2023-01-21 09:28:36.170473: step: 2048/527, loss: 0.15736566483974457
2023-01-21 09:28:37.333964: step: 2052/527, loss: 0.1980557143688202
2023-01-21 09:28:38.432317: step: 2056/527, loss: 0.07029370963573456
2023-01-21 09:28:39.560143: step: 2060/527, loss: 0.19861502945423126
2023-01-21 09:28:40.678583: step: 2064/527, loss: 0.18253564834594727
2023-01-21 09:28:41.792021: step: 2068/527, loss: 0.919429361820221
2023-01-21 09:28:42.896377: step: 2072/527, loss: 0.1285136193037033
2023-01-21 09:28:44.012918: step: 2076/527, loss: 0.14362964034080505
2023-01-21 09:28:45.130576: step: 2080/527, loss: 0.045040179044008255
2023-01-21 09:28:46.248694: step: 2084/527, loss: 0.15095248818397522
2023-01-21 09:28:47.362037: step: 2088/527, loss: 0.034040164202451706
2023-01-21 09:28:48.457064: step: 2092/527, loss: 0.1280737817287445
2023-01-21 09:28:49.581780: step: 2096/527, loss: 0.0901232659816742
2023-01-21 09:28:50.721715: step: 2100/527, loss: 0.12633295357227325
2023-01-21 09:28:51.839215: step: 2104/527, loss: 0.040778160095214844
2023-01-21 09:28:52.950426: step: 2108/527, loss: 0.1386338174343109
==================================================
Loss: 0.265
--------------------
Dev: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Test: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, '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': 0.5135135135135135, 'r': 0.5277777777777778, 'f1': 0.5205479452054794}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
New best chinese model...
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Korean: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 09:29:41.190406: step: 4/527, loss: 0.09809327125549316
2023-01-21 09:29:42.285917: step: 8/527, loss: 0.05085396766662598
2023-01-21 09:29:43.401056: step: 12/527, loss: 0.019361257553100586
2023-01-21 09:29:44.506587: step: 16/527, loss: 0.16312438249588013
2023-01-21 09:29:45.659630: step: 20/527, loss: 0.19972950220108032
2023-01-21 09:29:46.772180: step: 24/527, loss: 0.227146714925766
2023-01-21 09:29:47.902796: step: 28/527, loss: 0.13906002044677734
2023-01-21 09:29:48.997997: step: 32/527, loss: 0.00816967524588108
2023-01-21 09:29:50.113195: step: 36/527, loss: 0.08307237923145294
2023-01-21 09:29:51.274520: step: 40/527, loss: 0.2395917922258377
2023-01-21 09:29:52.412893: step: 44/527, loss: 0.34595829248428345
2023-01-21 09:29:53.515821: step: 48/527, loss: 0.09413719177246094
2023-01-21 09:29:54.632249: step: 52/527, loss: 0.13661280274391174
2023-01-21 09:29:55.758286: step: 56/527, loss: 0.07389058917760849
2023-01-21 09:29:56.923806: step: 60/527, loss: 0.318773478269577
2023-01-21 09:29:58.026685: step: 64/527, loss: 0.04285860061645508
2023-01-21 09:29:59.146088: step: 68/527, loss: 0.09297715127468109
2023-01-21 09:30:00.274498: step: 72/527, loss: 0.3081987202167511
2023-01-21 09:30:01.385252: step: 76/527, loss: 0.0583585761487484
2023-01-21 09:30:02.513177: step: 80/527, loss: 0.17066267132759094
2023-01-21 09:30:03.618503: step: 84/527, loss: 0.0739145278930664
2023-01-21 09:30:04.736547: step: 88/527, loss: 0.04386329650878906
2023-01-21 09:30:05.896162: step: 92/527, loss: 0.3526807725429535
2023-01-21 09:30:07.021843: step: 96/527, loss: 0.10786505043506622
2023-01-21 09:30:08.150878: step: 100/527, loss: 0.13941210508346558
2023-01-21 09:30:09.258954: step: 104/527, loss: 0.0392460823059082
2023-01-21 09:30:10.388354: step: 108/527, loss: 0.14055700600147247
2023-01-21 09:30:11.503345: step: 112/527, loss: 0.034890368580818176
2023-01-21 09:30:12.632838: step: 116/527, loss: 0.03204517439007759
2023-01-21 09:30:13.737536: step: 120/527, loss: 0.49269258975982666
2023-01-21 09:30:14.835645: step: 124/527, loss: 0.02144632488489151
2023-01-21 09:30:15.921619: step: 128/527, loss: 0.4735269546508789
2023-01-21 09:30:17.043656: step: 132/527, loss: 0.20490741729736328
2023-01-21 09:30:18.175620: step: 136/527, loss: 0.21948319673538208
2023-01-21 09:30:19.269872: step: 140/527, loss: 0.09926557540893555
2023-01-21 09:30:20.414050: step: 144/527, loss: 0.12586374580860138
2023-01-21 09:30:21.522273: step: 148/527, loss: 0.05742521584033966
2023-01-21 09:30:22.649351: step: 152/527, loss: 0.0542752742767334
2023-01-21 09:30:23.766681: step: 156/527, loss: 0.21279259026050568
2023-01-21 09:30:24.862301: step: 160/527, loss: 0.10694971680641174
2023-01-21 09:30:25.979413: step: 164/527, loss: 0.21865972876548767
2023-01-21 09:30:27.098373: step: 168/527, loss: 0.1016080379486084
2023-01-21 09:30:28.207486: step: 172/527, loss: 0.25411853194236755
2023-01-21 09:30:29.289994: step: 176/527, loss: 0.07370634377002716
2023-01-21 09:30:30.431570: step: 180/527, loss: 0.06605801731348038
2023-01-21 09:30:31.543441: step: 184/527, loss: 0.7231829166412354
2023-01-21 09:30:32.676300: step: 188/527, loss: 0.23981569707393646
2023-01-21 09:30:33.790127: step: 192/527, loss: 0.07329946011304855
2023-01-21 09:30:34.910509: step: 196/527, loss: 0.1070750504732132
2023-01-21 09:30:36.008529: step: 200/527, loss: 0.5285487174987793
2023-01-21 09:30:37.147772: step: 204/527, loss: 7.36193323135376
2023-01-21 09:30:38.259999: step: 208/527, loss: 1.5692079067230225
2023-01-21 09:30:39.387890: step: 212/527, loss: 0.2993311882019043
2023-01-21 09:30:40.487587: step: 216/527, loss: 0.24765712022781372
2023-01-21 09:30:41.587288: step: 220/527, loss: 0.06100483238697052
2023-01-21 09:30:42.678868: step: 224/527, loss: 0.28924334049224854
2023-01-21 09:30:43.805309: step: 228/527, loss: 0.13619975745677948
2023-01-21 09:30:44.912941: step: 232/527, loss: 0.023258520290255547
2023-01-21 09:30:46.043846: step: 236/527, loss: 0.18993701040744781
2023-01-21 09:30:47.140670: step: 240/527, loss: 0.04937797412276268
2023-01-21 09:30:48.255660: step: 244/527, loss: 0.38007059693336487
2023-01-21 09:30:49.372837: step: 248/527, loss: 0.07458534836769104
2023-01-21 09:30:50.490671: step: 252/527, loss: 0.09627914428710938
2023-01-21 09:30:51.606857: step: 256/527, loss: 0.1149645745754242
2023-01-21 09:30:52.725379: step: 260/527, loss: 0.22831077873706818
2023-01-21 09:30:53.839118: step: 264/527, loss: 0.16829939186573029
2023-01-21 09:30:54.947282: step: 268/527, loss: 0.026418304070830345
2023-01-21 09:30:56.074096: step: 272/527, loss: 0.08385644108057022
2023-01-21 09:30:57.178544: step: 276/527, loss: 0.19703388214111328
2023-01-21 09:30:58.302772: step: 280/527, loss: 0.5657750368118286
2023-01-21 09:30:59.435848: step: 284/527, loss: 0.443074107170105
2023-01-21 09:31:00.537208: step: 288/527, loss: 0.04611387103796005
2023-01-21 09:31:01.649920: step: 292/527, loss: 0.18945404887199402
2023-01-21 09:31:02.774163: step: 296/527, loss: 0.14467281103134155
2023-01-21 09:31:03.874105: step: 300/527, loss: 0.05260620266199112
2023-01-21 09:31:05.008382: step: 304/527, loss: 0.05623479187488556
2023-01-21 09:31:06.152617: step: 308/527, loss: 0.8390406370162964
2023-01-21 09:31:07.258334: step: 312/527, loss: 0.1849168837070465
2023-01-21 09:31:08.403183: step: 316/527, loss: 0.09425649791955948
2023-01-21 09:31:09.511796: step: 320/527, loss: 0.02252521552145481
2023-01-21 09:31:10.607392: step: 324/527, loss: 0.03591623529791832
2023-01-21 09:31:11.745002: step: 328/527, loss: 1.6285374164581299
2023-01-21 09:31:12.833451: step: 332/527, loss: 0.2274969518184662
2023-01-21 09:31:13.984159: step: 336/527, loss: 0.06355658173561096
2023-01-21 09:31:15.094283: step: 340/527, loss: 0.26239073276519775
2023-01-21 09:31:16.213929: step: 344/527, loss: 0.06323843449354172
2023-01-21 09:31:17.395148: step: 348/527, loss: 0.12584057450294495
2023-01-21 09:31:18.514128: step: 352/527, loss: 0.11452827602624893
2023-01-21 09:31:19.601354: step: 356/527, loss: 0.15861892700195312
2023-01-21 09:31:20.728520: step: 360/527, loss: 0.13074392080307007
2023-01-21 09:31:21.844785: step: 364/527, loss: 0.4625195562839508
2023-01-21 09:31:22.964479: step: 368/527, loss: 0.030454207211732864
2023-01-21 09:31:24.104838: step: 372/527, loss: 0.06957979500293732
2023-01-21 09:31:25.208848: step: 376/527, loss: 0.05140667036175728
2023-01-21 09:31:26.347111: step: 380/527, loss: 0.4253025949001312
2023-01-21 09:31:27.451348: step: 384/527, loss: 0.09943151473999023
2023-01-21 09:31:28.593054: step: 388/527, loss: 0.05313815921545029
2023-01-21 09:31:29.701973: step: 392/527, loss: 0.09376392513513565
2023-01-21 09:31:30.801961: step: 396/527, loss: 0.18427881598472595
2023-01-21 09:31:31.916828: step: 400/527, loss: 0.4059511423110962
2023-01-21 09:31:33.031942: step: 404/527, loss: 1.1235990524291992
2023-01-21 09:31:34.135961: step: 408/527, loss: 0.08466877788305283
2023-01-21 09:31:35.278160: step: 412/527, loss: 0.1253274828195572
2023-01-21 09:31:36.383071: step: 416/527, loss: 0.10244733095169067
2023-01-21 09:31:37.486175: step: 420/527, loss: 0.037759922444820404
2023-01-21 09:31:38.606467: step: 424/527, loss: 0.0929255485534668
2023-01-21 09:31:39.770789: step: 428/527, loss: 0.09901618957519531
2023-01-21 09:31:40.868417: step: 432/527, loss: 0.07042674720287323
2023-01-21 09:31:41.962887: step: 436/527, loss: 0.112521693110466
2023-01-21 09:31:43.130287: step: 440/527, loss: 0.06343488395214081
2023-01-21 09:31:44.238207: step: 444/527, loss: 0.2544664740562439
2023-01-21 09:31:45.350230: step: 448/527, loss: 0.10858345031738281
2023-01-21 09:31:46.489268: step: 452/527, loss: 0.1982189118862152
2023-01-21 09:31:47.584021: step: 456/527, loss: 0.462616503238678
2023-01-21 09:31:48.713825: step: 460/527, loss: 0.01927213743329048
2023-01-21 09:31:49.849094: step: 464/527, loss: 0.06015825644135475
2023-01-21 09:31:50.958442: step: 468/527, loss: 0.021100426092743874
2023-01-21 09:31:52.067300: step: 472/527, loss: 0.5226963758468628
2023-01-21 09:31:53.187324: step: 476/527, loss: 0.18448218703269958
2023-01-21 09:31:54.287194: step: 480/527, loss: 0.4267534017562866
2023-01-21 09:31:55.392100: step: 484/527, loss: 0.0962924063205719
2023-01-21 09:31:56.501636: step: 488/527, loss: 0.3233264088630676
2023-01-21 09:31:57.634796: step: 492/527, loss: 0.2946937382221222
2023-01-21 09:31:58.778650: step: 496/527, loss: 0.22921191155910492
2023-01-21 09:31:59.878962: step: 500/527, loss: 0.22058506309986115
2023-01-21 09:32:01.003573: step: 504/527, loss: 0.3670656085014343
2023-01-21 09:32:02.110809: step: 508/527, loss: 0.027770616114139557
2023-01-21 09:32:03.224641: step: 512/527, loss: 0.40341219305992126
2023-01-21 09:32:04.348395: step: 516/527, loss: 0.16455334424972534
2023-01-21 09:32:05.449600: step: 520/527, loss: 0.06385669857263565
2023-01-21 09:32:06.598618: step: 524/527, loss: 0.14708839356899261
2023-01-21 09:32:07.705854: step: 528/527, loss: 0.07627630233764648
2023-01-21 09:32:08.836067: step: 532/527, loss: 0.14719390869140625
2023-01-21 09:32:09.944627: step: 536/527, loss: 0.08033581078052521
2023-01-21 09:32:11.028983: step: 540/527, loss: 0.23325125873088837
2023-01-21 09:32:12.136523: step: 544/527, loss: 2.487367630004883
2023-01-21 09:32:13.237776: step: 548/527, loss: 0.06263256072998047
2023-01-21 09:32:14.361537: step: 552/527, loss: 0.06555996090173721
2023-01-21 09:32:15.458551: step: 556/527, loss: 0.6510850191116333
2023-01-21 09:32:16.610459: step: 560/527, loss: 0.051460932940244675
2023-01-21 09:32:17.765320: step: 564/527, loss: 0.2417098879814148
2023-01-21 09:32:18.862643: step: 568/527, loss: 0.15583720803260803
2023-01-21 09:32:19.947447: step: 572/527, loss: 0.09012976288795471
2023-01-21 09:32:21.088223: step: 576/527, loss: 0.09545796364545822
2023-01-21 09:32:22.201344: step: 580/527, loss: 0.1428963840007782
2023-01-21 09:32:23.297673: step: 584/527, loss: 0.02486112155020237
2023-01-21 09:32:24.420445: step: 588/527, loss: 0.6363394856452942
2023-01-21 09:32:25.518800: step: 592/527, loss: 0.15953369438648224
2023-01-21 09:32:26.604237: step: 596/527, loss: 0.09888716042041779
2023-01-21 09:32:27.733478: step: 600/527, loss: 0.17457886040210724
2023-01-21 09:32:28.872185: step: 604/527, loss: 0.12611256539821625
2023-01-21 09:32:29.976644: step: 608/527, loss: 0.1384473741054535
2023-01-21 09:32:31.098671: step: 612/527, loss: 0.5675891637802124
2023-01-21 09:32:32.200919: step: 616/527, loss: 0.04823513329029083
2023-01-21 09:32:33.328848: step: 620/527, loss: 0.1206444725394249
2023-01-21 09:32:34.434987: step: 624/527, loss: 0.13086660206317902
2023-01-21 09:32:35.538170: step: 628/527, loss: 0.03149242326617241
2023-01-21 09:32:36.653244: step: 632/527, loss: 1.693342924118042
2023-01-21 09:32:37.765220: step: 636/527, loss: 0.06635136902332306
2023-01-21 09:32:38.886650: step: 640/527, loss: 0.1403800994157791
2023-01-21 09:32:40.001063: step: 644/527, loss: 0.16130180656909943
2023-01-21 09:32:41.164177: step: 648/527, loss: 0.21009722352027893
2023-01-21 09:32:42.254281: step: 652/527, loss: 0.117999367415905
2023-01-21 09:32:43.337492: step: 656/527, loss: 0.06665630638599396
2023-01-21 09:32:44.434788: step: 660/527, loss: 0.09412749111652374
2023-01-21 09:32:45.552042: step: 664/527, loss: 0.0505555160343647
2023-01-21 09:32:46.655334: step: 668/527, loss: 0.10793104022741318
2023-01-21 09:32:47.771314: step: 672/527, loss: 0.09758541733026505
2023-01-21 09:32:48.870945: step: 676/527, loss: 0.028674745932221413
2023-01-21 09:32:49.979060: step: 680/527, loss: 0.17769642174243927
2023-01-21 09:32:51.103154: step: 684/527, loss: 0.05619053915143013
2023-01-21 09:32:52.232753: step: 688/527, loss: 0.1361272782087326
2023-01-21 09:32:53.377909: step: 692/527, loss: 0.047132205218076706
2023-01-21 09:32:54.475976: step: 696/527, loss: 0.02039952203631401
2023-01-21 09:32:55.583589: step: 700/527, loss: 0.2235385924577713
2023-01-21 09:32:56.695961: step: 704/527, loss: 0.5704684257507324
2023-01-21 09:32:57.813325: step: 708/527, loss: 0.06690311431884766
2023-01-21 09:32:58.906613: step: 712/527, loss: 0.04849109798669815
2023-01-21 09:33:00.011973: step: 716/527, loss: 0.04433765262365341
2023-01-21 09:33:01.144220: step: 720/527, loss: 0.1945153772830963
2023-01-21 09:33:02.285014: step: 724/527, loss: 1.1448918581008911
2023-01-21 09:33:03.418892: step: 728/527, loss: 0.12088766694068909
2023-01-21 09:33:04.507356: step: 732/527, loss: 0.15637846291065216
2023-01-21 09:33:05.622629: step: 736/527, loss: 0.16790294647216797
2023-01-21 09:33:06.755563: step: 740/527, loss: 0.22962361574172974
2023-01-21 09:33:07.880973: step: 744/527, loss: 0.2084241807460785
2023-01-21 09:33:08.982826: step: 748/527, loss: 0.1070745512843132
2023-01-21 09:33:10.100460: step: 752/527, loss: 0.044196177273988724
2023-01-21 09:33:11.206497: step: 756/527, loss: 0.08995571732521057
2023-01-21 09:33:12.329523: step: 760/527, loss: 0.13057290017604828
2023-01-21 09:33:13.448114: step: 764/527, loss: 0.042356349527835846
2023-01-21 09:33:14.541355: step: 768/527, loss: 0.06146450340747833
2023-01-21 09:33:15.645400: step: 772/527, loss: 0.8324622511863708
2023-01-21 09:33:16.776123: step: 776/527, loss: 0.10629577934741974
2023-01-21 09:33:17.897611: step: 780/527, loss: 0.1776294708251953
2023-01-21 09:33:19.024289: step: 784/527, loss: 0.09487247467041016
2023-01-21 09:33:20.183057: step: 788/527, loss: 0.17201289534568787
2023-01-21 09:33:21.281253: step: 792/527, loss: 0.09751377254724503
2023-01-21 09:33:22.371473: step: 796/527, loss: 0.06043205410242081
2023-01-21 09:33:23.476023: step: 800/527, loss: 0.6203104257583618
2023-01-21 09:33:24.602806: step: 804/527, loss: 0.1668541133403778
2023-01-21 09:33:25.707598: step: 808/527, loss: 0.019964074715971947
2023-01-21 09:33:26.846175: step: 812/527, loss: 0.10484543442726135
2023-01-21 09:33:27.945665: step: 816/527, loss: 0.06051645427942276
2023-01-21 09:33:29.047075: step: 820/527, loss: 0.12209854274988174
2023-01-21 09:33:30.169761: step: 824/527, loss: 0.04891569912433624
2023-01-21 09:33:31.268422: step: 828/527, loss: 0.15463551878929138
2023-01-21 09:33:32.397870: step: 832/527, loss: 0.04897718504071236
2023-01-21 09:33:33.573087: step: 836/527, loss: 0.07085323333740234
2023-01-21 09:33:34.728661: step: 840/527, loss: 0.22726650536060333
2023-01-21 09:33:35.854576: step: 844/527, loss: 0.1680097132921219
2023-01-21 09:33:36.998315: step: 848/527, loss: 0.05432019382715225
2023-01-21 09:33:38.086416: step: 852/527, loss: 0.20855771005153656
2023-01-21 09:33:39.199210: step: 856/527, loss: 0.09163542091846466
2023-01-21 09:33:40.316182: step: 860/527, loss: 0.6569075584411621
2023-01-21 09:33:41.452032: step: 864/527, loss: 0.20458078384399414
2023-01-21 09:33:42.573728: step: 868/527, loss: 0.03446998447179794
2023-01-21 09:33:43.681307: step: 872/527, loss: 0.09217791259288788
2023-01-21 09:33:44.824501: step: 876/527, loss: 0.12784966826438904
2023-01-21 09:33:45.962941: step: 880/527, loss: 0.3509131371974945
2023-01-21 09:33:47.097214: step: 884/527, loss: 0.09518971294164658
2023-01-21 09:33:48.230082: step: 888/527, loss: 0.12601538002490997
2023-01-21 09:33:49.337318: step: 892/527, loss: 0.37552404403686523
2023-01-21 09:33:50.439012: step: 896/527, loss: 0.027489997446537018
2023-01-21 09:33:51.538581: step: 900/527, loss: 0.5072728395462036
2023-01-21 09:33:52.652973: step: 904/527, loss: 0.15026246011257172
2023-01-21 09:33:53.745554: step: 908/527, loss: 0.03210487961769104
2023-01-21 09:33:54.837187: step: 912/527, loss: 0.33800965547561646
2023-01-21 09:33:55.948727: step: 916/527, loss: 0.1711035966873169
2023-01-21 09:33:57.066898: step: 920/527, loss: 0.27540695667266846
2023-01-21 09:33:58.199576: step: 924/527, loss: 0.14751377701759338
2023-01-21 09:33:59.334901: step: 928/527, loss: 0.1953502595424652
2023-01-21 09:34:00.489031: step: 932/527, loss: 0.11080723255872726
2023-01-21 09:34:01.617841: step: 936/527, loss: 0.27181875705718994
2023-01-21 09:34:02.753493: step: 940/527, loss: 0.2269311547279358
2023-01-21 09:34:03.865743: step: 944/527, loss: 0.1031261682510376
2023-01-21 09:34:04.999356: step: 948/527, loss: 0.22477851808071136
2023-01-21 09:34:06.124239: step: 952/527, loss: 0.2573592960834503
2023-01-21 09:34:07.250155: step: 956/527, loss: 0.01927013322710991
2023-01-21 09:34:08.350838: step: 960/527, loss: 0.10109458118677139
2023-01-21 09:34:09.455071: step: 964/527, loss: 0.26626792550086975
2023-01-21 09:34:10.554859: step: 968/527, loss: 0.06411395221948624
2023-01-21 09:34:11.648658: step: 972/527, loss: 0.07802343368530273
2023-01-21 09:34:12.761233: step: 976/527, loss: 0.05781755596399307
2023-01-21 09:34:13.864263: step: 980/527, loss: 0.19363024830818176
2023-01-21 09:34:14.959025: step: 984/527, loss: 0.04104762151837349
2023-01-21 09:34:16.114862: step: 988/527, loss: 0.10942211747169495
2023-01-21 09:34:17.205718: step: 992/527, loss: 0.24775008857250214
2023-01-21 09:34:18.305220: step: 996/527, loss: 0.059580616652965546
2023-01-21 09:34:19.422616: step: 1000/527, loss: 0.19865284860134125
2023-01-21 09:34:20.552078: step: 1004/527, loss: 0.07532624900341034
2023-01-21 09:34:21.697340: step: 1008/527, loss: 0.0459408275783062
2023-01-21 09:34:22.782113: step: 1012/527, loss: 0.19169330596923828
2023-01-21 09:34:23.899997: step: 1016/527, loss: 0.16757772862911224
2023-01-21 09:34:25.031451: step: 1020/527, loss: 0.18535690009593964
2023-01-21 09:34:26.140522: step: 1024/527, loss: 0.10012922435998917
2023-01-21 09:34:27.275946: step: 1028/527, loss: 0.33195796608924866
2023-01-21 09:34:28.379571: step: 1032/527, loss: 0.08743729442358017
2023-01-21 09:34:29.488323: step: 1036/527, loss: 0.11253470927476883
2023-01-21 09:34:30.612357: step: 1040/527, loss: 0.0761350616812706
2023-01-21 09:34:31.712625: step: 1044/527, loss: 0.4874473512172699
2023-01-21 09:34:32.860266: step: 1048/527, loss: 0.10132065415382385
2023-01-21 09:34:33.975419: step: 1052/527, loss: 0.06235170364379883
2023-01-21 09:34:35.069725: step: 1056/527, loss: 0.12807168066501617
2023-01-21 09:34:36.174679: step: 1060/527, loss: 0.07180386036634445
2023-01-21 09:34:37.276452: step: 1064/527, loss: 0.04184355586767197
2023-01-21 09:34:38.386873: step: 1068/527, loss: 0.17963910102844238
2023-01-21 09:34:39.496306: step: 1072/527, loss: 0.1311967819929123
2023-01-21 09:34:40.616606: step: 1076/527, loss: 0.15443992614746094
2023-01-21 09:34:41.746147: step: 1080/527, loss: 0.05564575642347336
2023-01-21 09:34:42.864476: step: 1084/527, loss: 0.10488510131835938
2023-01-21 09:34:43.998218: step: 1088/527, loss: 0.07519850879907608
2023-01-21 09:34:45.089236: step: 1092/527, loss: 0.03843050077557564
2023-01-21 09:34:46.183809: step: 1096/527, loss: 0.06683111190795898
2023-01-21 09:34:47.327704: step: 1100/527, loss: 0.11720244586467743
2023-01-21 09:34:48.425564: step: 1104/527, loss: 0.173865407705307
2023-01-21 09:34:49.528443: step: 1108/527, loss: 0.10403481125831604
2023-01-21 09:34:50.645942: step: 1112/527, loss: 0.14471574127674103
2023-01-21 09:34:51.770603: step: 1116/527, loss: 0.12222138047218323
2023-01-21 09:34:52.864599: step: 1120/527, loss: 0.4596376419067383
2023-01-21 09:34:53.973647: step: 1124/527, loss: 0.04904680326581001
2023-01-21 09:34:55.083912: step: 1128/527, loss: 0.05549602583050728
2023-01-21 09:34:56.167318: step: 1132/527, loss: 0.06192345544695854
2023-01-21 09:34:57.281788: step: 1136/527, loss: 0.21808868646621704
2023-01-21 09:34:58.379393: step: 1140/527, loss: 0.016543006524443626
2023-01-21 09:34:59.472004: step: 1144/527, loss: 0.03120427206158638
2023-01-21 09:35:00.588056: step: 1148/527, loss: 0.09503450989723206
2023-01-21 09:35:01.718601: step: 1152/527, loss: 0.13844004273414612
2023-01-21 09:35:02.829166: step: 1156/527, loss: 0.03258562088012695
2023-01-21 09:35:03.926247: step: 1160/527, loss: 0.16277696192264557
2023-01-21 09:35:05.055230: step: 1164/527, loss: 0.104385145008564
2023-01-21 09:35:06.148735: step: 1168/527, loss: 0.3542178273200989
2023-01-21 09:35:07.246998: step: 1172/527, loss: 0.03200025483965874
2023-01-21 09:35:08.371669: step: 1176/527, loss: 1.0159810781478882
2023-01-21 09:35:09.464939: step: 1180/527, loss: 0.02443542517721653
2023-01-21 09:35:10.571778: step: 1184/527, loss: 0.07146044075489044
2023-01-21 09:35:11.705577: step: 1188/527, loss: 0.07184791564941406
2023-01-21 09:35:12.816582: step: 1192/527, loss: 0.20940819382667542
2023-01-21 09:35:13.962697: step: 1196/527, loss: 0.04494314268231392
2023-01-21 09:35:15.083175: step: 1200/527, loss: 0.042475320398807526
2023-01-21 09:35:16.213208: step: 1204/527, loss: 0.20813989639282227
2023-01-21 09:35:17.299877: step: 1208/527, loss: 0.08634022623300552
2023-01-21 09:35:18.438863: step: 1212/527, loss: 0.10069496929645538
2023-01-21 09:35:19.559168: step: 1216/527, loss: 0.06731309741735458
2023-01-21 09:35:20.675278: step: 1220/527, loss: 0.042203620076179504
2023-01-21 09:35:21.794565: step: 1224/527, loss: 0.066911980509758
2023-01-21 09:35:22.938195: step: 1228/527, loss: 0.1049363911151886
2023-01-21 09:35:24.047962: step: 1232/527, loss: 0.13648033142089844
2023-01-21 09:35:25.133893: step: 1236/527, loss: 0.057218264788389206
2023-01-21 09:35:26.256146: step: 1240/527, loss: 0.09241737425327301
2023-01-21 09:35:27.379383: step: 1244/527, loss: 0.08478298783302307
2023-01-21 09:35:28.544022: step: 1248/527, loss: 0.04098280519247055
2023-01-21 09:35:29.667835: step: 1252/527, loss: 0.5447441935539246
2023-01-21 09:35:30.755013: step: 1256/527, loss: 0.10794176906347275
2023-01-21 09:35:31.875796: step: 1260/527, loss: 0.09547443687915802
2023-01-21 09:35:32.994718: step: 1264/527, loss: 0.6401312947273254
2023-01-21 09:35:34.123570: step: 1268/527, loss: 0.08853588253259659
2023-01-21 09:35:35.253286: step: 1272/527, loss: 0.06858411431312561
2023-01-21 09:35:36.347589: step: 1276/527, loss: 0.047343719750642776
2023-01-21 09:35:37.457896: step: 1280/527, loss: 0.07362031936645508
2023-01-21 09:35:38.585858: step: 1284/527, loss: 0.030834747478365898
2023-01-21 09:35:39.673829: step: 1288/527, loss: 0.09466381371021271
2023-01-21 09:35:40.772270: step: 1292/527, loss: 0.033593546599149704
2023-01-21 09:35:41.907493: step: 1296/527, loss: 0.7086694240570068
2023-01-21 09:35:43.010880: step: 1300/527, loss: 0.8558475375175476
2023-01-21 09:35:44.113185: step: 1304/527, loss: 0.2469218373298645
2023-01-21 09:35:45.208292: step: 1308/527, loss: 0.06691642105579376
2023-01-21 09:35:46.342275: step: 1312/527, loss: 0.2022683173418045
2023-01-21 09:35:47.468779: step: 1316/527, loss: 0.07356643676757812
2023-01-21 09:35:48.599984: step: 1320/527, loss: 0.3981837034225464
2023-01-21 09:35:49.721432: step: 1324/527, loss: 0.5606789588928223
2023-01-21 09:35:50.844025: step: 1328/527, loss: 0.04904594272375107
2023-01-21 09:35:51.969525: step: 1332/527, loss: 0.1244468241930008
2023-01-21 09:35:53.064599: step: 1336/527, loss: 0.07025432586669922
2023-01-21 09:35:54.174308: step: 1340/527, loss: 0.16344553232192993
2023-01-21 09:35:55.331656: step: 1344/527, loss: 0.32741180062294006
2023-01-21 09:35:56.472423: step: 1348/527, loss: 0.07296828925609589
2023-01-21 09:35:57.572623: step: 1352/527, loss: 0.27629411220550537
2023-01-21 09:35:58.676870: step: 1356/527, loss: 0.12563781440258026
2023-01-21 09:35:59.788040: step: 1360/527, loss: 0.027243472635746002
2023-01-21 09:36:00.908613: step: 1364/527, loss: 0.05697689205408096
2023-01-21 09:36:02.032528: step: 1368/527, loss: 0.1365257352590561
2023-01-21 09:36:03.130117: step: 1372/527, loss: 0.0675632506608963
2023-01-21 09:36:04.272059: step: 1376/527, loss: 0.104736328125
2023-01-21 09:36:05.402535: step: 1380/527, loss: 0.11787891387939453
2023-01-21 09:36:06.531812: step: 1384/527, loss: 0.08608703315258026
2023-01-21 09:36:07.679917: step: 1388/527, loss: 0.11276207119226456
2023-01-21 09:36:08.810266: step: 1392/527, loss: 0.05546441301703453
2023-01-21 09:36:09.933765: step: 1396/527, loss: 0.09754600375890732
2023-01-21 09:36:11.060143: step: 1400/527, loss: 0.11838255077600479
2023-01-21 09:36:12.208229: step: 1404/527, loss: 0.29293060302734375
2023-01-21 09:36:13.322093: step: 1408/527, loss: 0.2585332989692688
2023-01-21 09:36:14.504341: step: 1412/527, loss: 0.0649944320321083
2023-01-21 09:36:15.629651: step: 1416/527, loss: 0.4411774277687073
2023-01-21 09:36:16.770822: step: 1420/527, loss: 0.3748853802680969
2023-01-21 09:36:17.895870: step: 1424/527, loss: 0.14224694669246674
2023-01-21 09:36:18.992803: step: 1428/527, loss: 0.12376852333545685
2023-01-21 09:36:20.110559: step: 1432/527, loss: 0.0445614829659462
2023-01-21 09:36:21.229261: step: 1436/527, loss: 0.10244321823120117
2023-01-21 09:36:22.341863: step: 1440/527, loss: 0.049125123769044876
2023-01-21 09:36:23.453258: step: 1444/527, loss: 0.0648014098405838
2023-01-21 09:36:24.553419: step: 1448/527, loss: 0.043448470532894135
2023-01-21 09:36:25.656980: step: 1452/527, loss: 2.2965469360351562
2023-01-21 09:36:26.776347: step: 1456/527, loss: 0.08939695358276367
2023-01-21 09:36:27.870270: step: 1460/527, loss: 0.49545416235923767
2023-01-21 09:36:28.965646: step: 1464/527, loss: 0.2712860107421875
2023-01-21 09:36:30.129217: step: 1468/527, loss: 0.12156401574611664
2023-01-21 09:36:31.215519: step: 1472/527, loss: 0.06898212432861328
2023-01-21 09:36:32.331259: step: 1476/527, loss: 0.7155297994613647
2023-01-21 09:36:33.464429: step: 1480/527, loss: 0.5859881639480591
2023-01-21 09:36:34.590811: step: 1484/527, loss: 0.2621147036552429
2023-01-21 09:36:35.702646: step: 1488/527, loss: 0.2753903269767761
2023-01-21 09:36:36.824252: step: 1492/527, loss: 0.07720744609832764
2023-01-21 09:36:37.960746: step: 1496/527, loss: 0.3880332112312317
2023-01-21 09:36:39.074083: step: 1500/527, loss: 0.029722118750214577
2023-01-21 09:36:40.172769: step: 1504/527, loss: 0.12754212319850922
2023-01-21 09:36:41.326867: step: 1508/527, loss: 0.03767247498035431
2023-01-21 09:36:42.449272: step: 1512/527, loss: 0.1910894364118576
2023-01-21 09:36:43.544545: step: 1516/527, loss: 0.20364342629909515
2023-01-21 09:36:44.684249: step: 1520/527, loss: 0.10594988614320755
2023-01-21 09:36:45.791095: step: 1524/527, loss: 0.39296627044677734
2023-01-21 09:36:46.898602: step: 1528/527, loss: 0.20036441087722778
2023-01-21 09:36:47.999110: step: 1532/527, loss: 0.22469262778759003
2023-01-21 09:36:49.090820: step: 1536/527, loss: 0.7947580218315125
2023-01-21 09:36:50.185676: step: 1540/527, loss: 0.14896979928016663
2023-01-21 09:36:51.309171: step: 1544/527, loss: 0.07383136451244354
2023-01-21 09:36:52.413554: step: 1548/527, loss: 0.3677327036857605
2023-01-21 09:36:53.523851: step: 1552/527, loss: 0.2320484071969986
2023-01-21 09:36:54.598312: step: 1556/527, loss: 0.045197200030088425
2023-01-21 09:36:55.725663: step: 1560/527, loss: 0.04570980370044708
2023-01-21 09:36:56.827536: step: 1564/527, loss: 0.1397724151611328
2023-01-21 09:36:57.918821: step: 1568/527, loss: 0.050203919410705566
2023-01-21 09:36:59.031588: step: 1572/527, loss: 0.10314774513244629
2023-01-21 09:37:00.143260: step: 1576/527, loss: 0.14121532440185547
2023-01-21 09:37:01.253159: step: 1580/527, loss: 0.1390089988708496
2023-01-21 09:37:02.371086: step: 1584/527, loss: 0.16678285598754883
2023-01-21 09:37:03.498161: step: 1588/527, loss: 0.06695995479822159
2023-01-21 09:37:04.632664: step: 1592/527, loss: 0.1356671303510666
2023-01-21 09:37:05.773415: step: 1596/527, loss: 0.12512041628360748
2023-01-21 09:37:06.885171: step: 1600/527, loss: 0.15033353865146637
2023-01-21 09:37:07.986967: step: 1604/527, loss: 0.08384790271520615
2023-01-21 09:37:09.083078: step: 1608/527, loss: 0.12716065347194672
2023-01-21 09:37:10.194307: step: 1612/527, loss: 1.3559350967407227
2023-01-21 09:37:11.317545: step: 1616/527, loss: 0.2778850495815277
2023-01-21 09:37:12.413534: step: 1620/527, loss: 0.21592837572097778
2023-01-21 09:37:13.540968: step: 1624/527, loss: 1.6223134994506836
2023-01-21 09:37:14.661317: step: 1628/527, loss: 0.11719150841236115
2023-01-21 09:37:15.787855: step: 1632/527, loss: 0.02186565473675728
2023-01-21 09:37:16.914863: step: 1636/527, loss: 0.06387953460216522
2023-01-21 09:37:18.039788: step: 1640/527, loss: 0.28336697816848755
2023-01-21 09:37:19.137719: step: 1644/527, loss: 0.059557151049375534
2023-01-21 09:37:20.255403: step: 1648/527, loss: 0.10761547088623047
2023-01-21 09:37:21.364020: step: 1652/527, loss: 0.0354095958173275
2023-01-21 09:37:22.462420: step: 1656/527, loss: 0.023825978860259056
2023-01-21 09:37:23.614476: step: 1660/527, loss: 0.6483904123306274
2023-01-21 09:37:24.735209: step: 1664/527, loss: 0.13121700286865234
2023-01-21 09:37:25.875597: step: 1668/527, loss: 0.08512725681066513
2023-01-21 09:37:26.993057: step: 1672/527, loss: 0.24677523970603943
2023-01-21 09:37:28.154057: step: 1676/527, loss: 0.1847589612007141
2023-01-21 09:37:29.253635: step: 1680/527, loss: 0.05692744627594948
2023-01-21 09:37:30.354385: step: 1684/527, loss: 0.03351978957653046
2023-01-21 09:37:31.470123: step: 1688/527, loss: 0.20769162476062775
2023-01-21 09:37:32.583657: step: 1692/527, loss: 0.1787862777709961
2023-01-21 09:37:33.688811: step: 1696/527, loss: 0.11295967549085617
2023-01-21 09:37:34.832751: step: 1700/527, loss: 0.2951013445854187
2023-01-21 09:37:35.970537: step: 1704/527, loss: 0.7350033521652222
2023-01-21 09:37:37.077038: step: 1708/527, loss: 0.1782231330871582
2023-01-21 09:37:38.177747: step: 1712/527, loss: 0.06197957694530487
2023-01-21 09:37:39.325221: step: 1716/527, loss: 0.137616828083992
2023-01-21 09:37:40.448936: step: 1720/527, loss: 0.12330102920532227
2023-01-21 09:37:41.556237: step: 1724/527, loss: 0.07322268187999725
2023-01-21 09:37:42.690673: step: 1728/527, loss: 0.10931578278541565
2023-01-21 09:37:43.792335: step: 1732/527, loss: 0.21992158889770508
2023-01-21 09:37:44.904782: step: 1736/527, loss: 0.03928103670477867
2023-01-21 09:37:46.021075: step: 1740/527, loss: 0.1692342758178711
2023-01-21 09:37:47.148393: step: 1744/527, loss: 0.8762350678443909
2023-01-21 09:37:48.282832: step: 1748/527, loss: 0.08703046292066574
2023-01-21 09:37:49.392762: step: 1752/527, loss: 0.14437074959278107
2023-01-21 09:37:50.507521: step: 1756/527, loss: 0.6991404294967651
2023-01-21 09:37:51.642159: step: 1760/527, loss: 1.4307262897491455
2023-01-21 09:37:52.728720: step: 1764/527, loss: 0.11755216121673584
2023-01-21 09:37:53.846597: step: 1768/527, loss: 0.044787146151065826
2023-01-21 09:37:54.977756: step: 1772/527, loss: 0.3315788507461548
2023-01-21 09:37:56.111605: step: 1776/527, loss: 0.0509905107319355
2023-01-21 09:37:57.232031: step: 1780/527, loss: 0.10356564819812775
2023-01-21 09:37:58.360488: step: 1784/527, loss: 0.11475391685962677
2023-01-21 09:37:59.446049: step: 1788/527, loss: 0.07465533912181854
2023-01-21 09:38:00.540511: step: 1792/527, loss: 0.623273491859436
2023-01-21 09:38:01.636951: step: 1796/527, loss: 1.0531699657440186
2023-01-21 09:38:02.741195: step: 1800/527, loss: 0.16995516419410706
2023-01-21 09:38:03.854956: step: 1804/527, loss: 0.3234786093235016
2023-01-21 09:38:04.979190: step: 1808/527, loss: 0.07282867282629013
2023-01-21 09:38:06.089579: step: 1812/527, loss: 0.08937902003526688
2023-01-21 09:38:07.228500: step: 1816/527, loss: 0.08460541069507599
2023-01-21 09:38:08.331278: step: 1820/527, loss: 0.25552287697792053
2023-01-21 09:38:09.436697: step: 1824/527, loss: 0.511398434638977
2023-01-21 09:38:10.586547: step: 1828/527, loss: 0.2710249423980713
2023-01-21 09:38:11.685277: step: 1832/527, loss: 0.11441951245069504
2023-01-21 09:38:12.833287: step: 1836/527, loss: 0.08043336868286133
2023-01-21 09:38:13.950220: step: 1840/527, loss: 0.09224581718444824
2023-01-21 09:38:15.075175: step: 1844/527, loss: 0.05091162025928497
2023-01-21 09:38:16.184224: step: 1848/527, loss: 0.033593371510505676
2023-01-21 09:38:17.285900: step: 1852/527, loss: 0.10790939629077911
2023-01-21 09:38:18.374528: step: 1856/527, loss: 0.12572938203811646
2023-01-21 09:38:19.509225: step: 1860/527, loss: 0.26958292722702026
2023-01-21 09:38:20.645805: step: 1864/527, loss: 0.23192210495471954
2023-01-21 09:38:21.747380: step: 1868/527, loss: 0.25937768816947937
2023-01-21 09:38:22.883883: step: 1872/527, loss: 0.11124172806739807
2023-01-21 09:38:23.994469: step: 1876/527, loss: 0.14579525589942932
2023-01-21 09:38:25.110171: step: 1880/527, loss: 0.21267780661582947
2023-01-21 09:38:26.221303: step: 1884/527, loss: 0.13840064406394958
2023-01-21 09:38:27.339929: step: 1888/527, loss: 0.24192224442958832
2023-01-21 09:38:28.481817: step: 1892/527, loss: 0.21090145409107208
2023-01-21 09:38:29.629911: step: 1896/527, loss: 0.3962695002555847
2023-01-21 09:38:30.727861: step: 1900/527, loss: 0.45139873027801514
2023-01-21 09:38:31.820349: step: 1904/527, loss: 0.11555171757936478
2023-01-21 09:38:32.920131: step: 1908/527, loss: 0.07022295147180557
2023-01-21 09:38:34.030938: step: 1912/527, loss: 0.14107860624790192
2023-01-21 09:38:35.124238: step: 1916/527, loss: 0.06297574192285538
2023-01-21 09:38:36.221981: step: 1920/527, loss: 0.17644786834716797
2023-01-21 09:38:37.334551: step: 1924/527, loss: 0.8610503077507019
2023-01-21 09:38:38.445555: step: 1928/527, loss: 0.10697450488805771
2023-01-21 09:38:39.558186: step: 1932/527, loss: 0.16575948894023895
2023-01-21 09:38:40.671543: step: 1936/527, loss: 0.055447958409786224
2023-01-21 09:38:41.773983: step: 1940/527, loss: 0.04218924418091774
2023-01-21 09:38:42.862649: step: 1944/527, loss: 0.1842663288116455
2023-01-21 09:38:43.964972: step: 1948/527, loss: 0.03675966337323189
2023-01-21 09:38:45.083308: step: 1952/527, loss: 0.0435916893184185
2023-01-21 09:38:46.204162: step: 1956/527, loss: 0.283983439207077
2023-01-21 09:38:47.319261: step: 1960/527, loss: 0.1147836223244667
2023-01-21 09:38:48.432740: step: 1964/527, loss: 0.04272041469812393
2023-01-21 09:38:49.508203: step: 1968/527, loss: 0.04331010952591896
2023-01-21 09:38:50.654099: step: 1972/527, loss: 0.4950031340122223
2023-01-21 09:38:51.793691: step: 1976/527, loss: 0.0833769366145134
2023-01-21 09:38:52.883516: step: 1980/527, loss: 0.1930265873670578
2023-01-21 09:38:53.991318: step: 1984/527, loss: 0.3233078420162201
2023-01-21 09:38:55.136705: step: 1988/527, loss: 0.11017389595508575
2023-01-21 09:38:56.262566: step: 1992/527, loss: 0.2645139694213867
2023-01-21 09:38:57.364735: step: 1996/527, loss: 0.16826733946800232
2023-01-21 09:38:58.454524: step: 2000/527, loss: 0.21377840638160706
2023-01-21 09:38:59.559973: step: 2004/527, loss: 0.027663325890898705
2023-01-21 09:39:00.680052: step: 2008/527, loss: 1.346925139427185
2023-01-21 09:39:01.792641: step: 2012/527, loss: 0.12795507907867432
2023-01-21 09:39:02.917411: step: 2016/527, loss: 0.14580735564231873
2023-01-21 09:39:04.031064: step: 2020/527, loss: 0.06216411292552948
2023-01-21 09:39:05.138663: step: 2024/527, loss: 0.2667986750602722
2023-01-21 09:39:06.252947: step: 2028/527, loss: 0.2472350150346756
2023-01-21 09:39:07.362151: step: 2032/527, loss: 0.3867262005805969
2023-01-21 09:39:08.482355: step: 2036/527, loss: 0.1298755705356598
2023-01-21 09:39:09.584936: step: 2040/527, loss: 0.10681380331516266
2023-01-21 09:39:10.666787: step: 2044/527, loss: 0.07795391231775284
2023-01-21 09:39:11.803135: step: 2048/527, loss: 0.1490764617919922
2023-01-21 09:39:12.929069: step: 2052/527, loss: 0.045057106763124466
2023-01-21 09:39:14.033441: step: 2056/527, loss: 0.014367294497787952
2023-01-21 09:39:15.175573: step: 2060/527, loss: 0.11018409579992294
2023-01-21 09:39:16.275129: step: 2064/527, loss: 0.14450684189796448
2023-01-21 09:39:17.391477: step: 2068/527, loss: 0.20723925530910492
2023-01-21 09:39:18.520692: step: 2072/527, loss: 0.20739594101905823
2023-01-21 09:39:19.633809: step: 2076/527, loss: 0.28867560625076294
2023-01-21 09:39:20.780884: step: 2080/527, loss: 0.21481843292713165
2023-01-21 09:39:21.879904: step: 2084/527, loss: 0.019100641831755638
2023-01-21 09:39:22.992668: step: 2088/527, loss: 0.2526426315307617
2023-01-21 09:39:24.088843: step: 2092/527, loss: 0.04098861292004585
2023-01-21 09:39:25.208001: step: 2096/527, loss: 0.06792278587818146
2023-01-21 09:39:26.314231: step: 2100/527, loss: 0.4546794891357422
2023-01-21 09:39:27.414123: step: 2104/527, loss: 0.0580110177397728
2023-01-21 09:39:28.524035: step: 2108/527, loss: 0.07411313056945801
==================================================
Loss: 0.216
--------------------
Dev: {'event': {'p': 0.5728542914171657, 'r': 0.7643142476697736, 'f1': 0.6548773531089561}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6}
Test: {'event': {'p': 0.6153136531365314, 'r': 0.7622857142857142, 'f1': 0.6809596733027055}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6}
Chinese: {'event': {'p': 0.5764705882352941, 'r': 0.9074074074074074, 'f1': 0.7050359712230215}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6}
Korean: {'event': {'p': 0.5833333333333334, 'r': 0.5555555555555556, 'f1': 0.5691056910569107}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6}
Russian: {'event': {'p': 0.4222222222222222, 'r': 0.5277777777777778, 'f1': 0.46913580246913583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.623059866962306, 'r': 0.748335552596538, 'f1': 0.6799758015728978}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Eng Test for Korean: {'event': {'p': 0.6303294573643411, 'r': 0.7434285714285714, 'f1': 0.6822233875196644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
Sample Korean: {'event': {'p': 0.725, 'r': 0.4603174603174603, 'f1': 0.5631067961165048}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 09:40:10.051795: step: 4/527, loss: 0.4163857698440552
2023-01-21 09:40:11.173102: step: 8/527, loss: 0.1529022753238678
2023-01-21 09:40:12.284385: step: 12/527, loss: 0.05762157216668129
2023-01-21 09:40:13.386396: step: 16/527, loss: 0.08841285854578018
2023-01-21 09:40:14.498230: step: 20/527, loss: 0.0973958969116211
2023-01-21 09:40:15.618108: step: 24/527, loss: 0.08074665069580078
2023-01-21 09:40:16.755292: step: 28/527, loss: 0.041546061635017395
2023-01-21 09:40:17.878070: step: 32/527, loss: 0.08659076690673828
2023-01-21 09:40:18.985616: step: 36/527, loss: 0.03707285225391388
2023-01-21 09:40:20.094904: step: 40/527, loss: 0.034918975085020065
2023-01-21 09:40:21.216512: step: 44/527, loss: 1.883135437965393
2023-01-21 09:40:22.343063: step: 48/527, loss: 0.04940805584192276
2023-01-21 09:40:23.461758: step: 52/527, loss: 0.47926968336105347
2023-01-21 09:40:24.565761: step: 56/527, loss: 0.4219369888305664
2023-01-21 09:40:25.676914: step: 60/527, loss: 0.165618896484375
2023-01-21 09:40:26.779701: step: 64/527, loss: 0.09352541714906693
2023-01-21 09:40:27.887896: step: 68/527, loss: 0.10504341125488281
2023-01-21 09:40:29.016143: step: 72/527, loss: 0.06129646301269531
2023-01-21 09:40:30.118792: step: 76/527, loss: 0.09223870933055878
2023-01-21 09:40:31.243218: step: 80/527, loss: 0.041403770446777344
2023-01-21 09:40:32.365966: step: 84/527, loss: 0.18043842911720276
2023-01-21 09:40:33.479560: step: 88/527, loss: 0.05343092605471611
2023-01-21 09:40:34.624463: step: 92/527, loss: 0.1469569355249405
2023-01-21 09:40:35.762541: step: 96/527, loss: 0.049339961260557175
2023-01-21 09:40:36.890690: step: 100/527, loss: 0.3157622516155243
2023-01-21 09:40:38.019200: step: 104/527, loss: 0.060452986508607864
2023-01-21 09:40:39.118268: step: 108/527, loss: 0.08834151923656464
2023-01-21 09:40:40.226950: step: 112/527, loss: 0.0646904930472374
2023-01-21 09:40:41.398516: step: 116/527, loss: 0.18475113809108734
2023-01-21 09:40:42.509512: step: 120/527, loss: 0.06487848609685898
2023-01-21 09:40:43.664453: step: 124/527, loss: 0.018282031640410423
2023-01-21 09:40:44.793561: step: 128/527, loss: 0.72438645362854
2023-01-21 09:40:45.930344: step: 132/527, loss: 0.05598144605755806
2023-01-21 09:40:47.053074: step: 136/527, loss: 0.2264978438615799
2023-01-21 09:40:48.162967: step: 140/527, loss: 0.03982143476605415
2023-01-21 09:40:49.280251: step: 144/527, loss: 0.14689961075782776
2023-01-21 09:40:50.418272: step: 148/527, loss: 0.4916355311870575
2023-01-21 09:40:51.524222: step: 152/527, loss: 0.07981789112091064
2023-01-21 09:40:52.667074: step: 156/527, loss: 0.18129000067710876
2023-01-21 09:40:53.794329: step: 160/527, loss: 0.2122432291507721
2023-01-21 09:40:54.920826: step: 164/527, loss: 0.16031265258789062
2023-01-21 09:40:56.071747: step: 168/527, loss: 0.09366440773010254
2023-01-21 09:40:57.189351: step: 172/527, loss: 0.06383056938648224
2023-01-21 09:40:58.309317: step: 176/527, loss: 0.02323140949010849
2023-01-21 09:40:59.435388: step: 180/527, loss: 0.0427170991897583
2023-01-21 09:41:00.575830: step: 184/527, loss: 0.02111082151532173
2023-01-21 09:41:01.670961: step: 188/527, loss: 0.1825898289680481
2023-01-21 09:41:02.797003: step: 192/527, loss: 0.06691794097423553
2023-01-21 09:41:03.973640: step: 196/527, loss: 0.03627481311559677
2023-01-21 09:41:05.102077: step: 200/527, loss: 0.2426256686449051
2023-01-21 09:41:06.232455: step: 204/527, loss: 0.08774900436401367
2023-01-21 09:41:07.333509: step: 208/527, loss: 0.015432262793183327
2023-01-21 09:41:08.439604: step: 212/527, loss: 0.1281810998916626
2023-01-21 09:41:09.555419: step: 216/527, loss: 0.11239566653966904
2023-01-21 09:41:10.657406: step: 220/527, loss: 0.14200712740421295
2023-01-21 09:41:11.763592: step: 224/527, loss: 0.020856954157352448
2023-01-21 09:41:12.866097: step: 228/527, loss: 0.42918795347213745
2023-01-21 09:41:13.984629: step: 232/527, loss: 0.3628288507461548
2023-01-21 09:41:15.120128: step: 236/527, loss: 0.0947122573852539
2023-01-21 09:41:16.219014: step: 240/527, loss: 0.19834065437316895
2023-01-21 09:41:17.369300: step: 244/527, loss: 0.4911825954914093
2023-01-21 09:41:18.466484: step: 248/527, loss: 0.09635010361671448
2023-01-21 09:41:19.575354: step: 252/527, loss: 0.17898845672607422
2023-01-21 09:41:20.677382: step: 256/527, loss: 0.022247934713959694
2023-01-21 09:41:21.794291: step: 260/527, loss: 0.03129100799560547
2023-01-21 09:41:22.935531: step: 264/527, loss: 0.4095683991909027
2023-01-21 09:41:24.043786: step: 268/527, loss: 0.40701723098754883
2023-01-21 09:41:25.147282: step: 272/527, loss: 0.04488027095794678
2023-01-21 09:41:26.278359: step: 276/527, loss: 0.09322738647460938
2023-01-21 09:41:27.393727: step: 280/527, loss: 0.03998060151934624
2023-01-21 09:41:28.486213: step: 284/527, loss: 0.1009332686662674
2023-01-21 09:41:29.625618: step: 288/527, loss: 0.093017578125
2023-01-21 09:41:30.703814: step: 292/527, loss: 0.10761814564466476
2023-01-21 09:41:31.811645: step: 296/527, loss: 0.12481708824634552
2023-01-21 09:41:32.929573: step: 300/527, loss: 0.22298498451709747
2023-01-21 09:41:34.059300: step: 304/527, loss: 0.03874704986810684
2023-01-21 09:41:35.188250: step: 308/527, loss: 0.10267810523509979
2023-01-21 09:41:36.290354: step: 312/527, loss: 0.11853323131799698
2023-01-21 09:41:37.457227: step: 316/527, loss: 0.18502922356128693
2023-01-21 09:41:38.561301: step: 320/527, loss: 0.23269948363304138
2023-01-21 09:41:39.680500: step: 324/527, loss: 0.16119326651096344
2023-01-21 09:41:40.770815: step: 328/527, loss: 0.2053823471069336
2023-01-21 09:41:41.888753: step: 332/527, loss: 0.20369234681129456
2023-01-21 09:41:43.034491: step: 336/527, loss: 0.0752011314034462
2023-01-21 09:41:44.156519: step: 340/527, loss: 0.23168422281742096
2023-01-21 09:41:45.300713: step: 344/527, loss: 0.11187683045864105
2023-01-21 09:41:46.426395: step: 348/527, loss: 0.07700939476490021
2023-01-21 09:41:47.549387: step: 352/527, loss: 0.09568607807159424
2023-01-21 09:41:48.649391: step: 356/527, loss: 0.09235820919275284
2023-01-21 09:41:49.759547: step: 360/527, loss: 0.04071841016411781
2023-01-21 09:41:50.887634: step: 364/527, loss: 0.3631020486354828
2023-01-21 09:41:52.035545: step: 368/527, loss: 0.08378183841705322
2023-01-21 09:41:53.161431: step: 372/527, loss: 0.10895195603370667
2023-01-21 09:41:54.277432: step: 376/527, loss: 0.2277248501777649
2023-01-21 09:41:55.411719: step: 380/527, loss: 0.08755837380886078
2023-01-21 09:41:56.502781: step: 384/527, loss: 0.021001672372221947
2023-01-21 09:41:57.611400: step: 388/527, loss: 0.03189681097865105
2023-01-21 09:41:58.708446: step: 392/527, loss: 0.07882877439260483
2023-01-21 09:41:59.799388: step: 396/527, loss: 0.051482390612363815
2023-01-21 09:42:00.913651: step: 400/527, loss: 0.05202658474445343
2023-01-21 09:42:02.037224: step: 404/527, loss: 0.010090017691254616
2023-01-21 09:42:03.160085: step: 408/527, loss: 0.04227237403392792
2023-01-21 09:42:04.280682: step: 412/527, loss: 0.0582159049808979
2023-01-21 09:42:05.364915: step: 416/527, loss: 0.04844675213098526
2023-01-21 09:42:06.511649: step: 420/527, loss: 0.1275196075439453
2023-01-21 09:42:07.654423: step: 424/527, loss: 0.03432321548461914
2023-01-21 09:42:08.802901: step: 428/527, loss: 0.027138233184814453
2023-01-21 09:42:09.927160: step: 432/527, loss: 0.15999102592468262
2023-01-21 09:42:11.044487: step: 436/527, loss: 0.0599401481449604
2023-01-21 09:42:12.191686: step: 440/527, loss: 0.13749723136425018
2023-01-21 09:42:13.302553: step: 444/527, loss: 0.1396523416042328
2023-01-21 09:42:14.418609: step: 448/527, loss: 0.1558721512556076
2023-01-21 09:42:15.526654: step: 452/527, loss: 0.0588221549987793
2023-01-21 09:42:16.645634: step: 456/527, loss: 0.0807713121175766
2023-01-21 09:42:17.738830: step: 460/527, loss: 0.2337636947631836
2023-01-21 09:42:18.855392: step: 464/527, loss: 0.353346586227417
2023-01-21 09:42:19.946891: step: 468/527, loss: 0.3557613492012024
2023-01-21 09:42:21.049065: step: 472/527, loss: 0.0360901840031147
2023-01-21 09:42:22.166576: step: 476/527, loss: 0.057752180844545364
2023-01-21 09:42:23.270383: step: 480/527, loss: 0.05274629592895508
2023-01-21 09:42:24.404226: step: 484/527, loss: 0.21701183915138245
2023-01-21 09:42:25.525048: step: 488/527, loss: 0.11842170357704163
2023-01-21 09:42:26.658948: step: 492/527, loss: 0.07372350990772247
2023-01-21 09:42:27.776777: step: 496/527, loss: 0.1916522979736328
2023-01-21 09:42:28.912971: step: 500/527, loss: 0.026337623596191406
2023-01-21 09:42:30.045796: step: 504/527, loss: 0.1076316386461258
2023-01-21 09:42:31.151489: step: 508/527, loss: 0.0957251638174057
2023-01-21 09:42:32.278925: step: 512/527, loss: 0.1499813199043274
2023-01-21 09:42:33.376215: step: 516/527, loss: 0.08520378917455673
2023-01-21 09:42:34.490976: step: 520/527, loss: 0.016389036551117897
2023-01-21 09:42:35.596739: step: 524/527, loss: 0.18101529777050018
2023-01-21 09:42:36.718237: step: 528/527, loss: 0.867286205291748
2023-01-21 09:42:37.822333: step: 532/527, loss: 0.13916663825511932
2023-01-21 09:42:38.929877: step: 536/527, loss: 0.027466487139463425
2023-01-21 09:42:40.034756: step: 540/527, loss: 0.09202560782432556
2023-01-21 09:42:41.149952: step: 544/527, loss: 0.056922055780887604
2023-01-21 09:42:42.310790: step: 548/527, loss: 0.03030090220272541
2023-01-21 09:42:43.443383: step: 552/527, loss: 0.08011779934167862
2023-01-21 09:42:44.542360: step: 556/527, loss: 0.2224745750427246
2023-01-21 09:42:45.658347: step: 560/527, loss: 1.09000825881958
2023-01-21 09:42:46.775640: step: 564/527, loss: 0.11000881344079971
2023-01-21 09:42:47.903669: step: 568/527, loss: 0.10341700911521912
2023-01-21 09:42:48.992470: step: 572/527, loss: 0.09961043298244476
2023-01-21 09:42:50.104438: step: 576/527, loss: 0.08434166759252548
2023-01-21 09:42:51.200429: step: 580/527, loss: 1.3565940856933594
2023-01-21 09:42:52.320789: step: 584/527, loss: 0.061338040977716446
2023-01-21 09:42:53.449248: step: 588/527, loss: 0.09565553814172745
2023-01-21 09:42:54.544183: step: 592/527, loss: 0.058181095868349075
2023-01-21 09:42:55.661947: step: 596/527, loss: 0.028829671442508698
2023-01-21 09:42:56.773603: step: 600/527, loss: 0.624508798122406
2023-01-21 09:42:57.885236: step: 604/527, loss: 0.13610239326953888
2023-01-21 09:42:59.034390: step: 608/527, loss: 0.20758108794689178
2023-01-21 09:43:00.146896: step: 612/527, loss: 0.049506377428770065
2023-01-21 09:43:01.291901: step: 616/527, loss: 0.10307340323925018
2023-01-21 09:43:02.398947: step: 620/527, loss: 0.2556304931640625
2023-01-21 09:43:03.521637: step: 624/527, loss: 0.027541637420654297
2023-01-21 09:43:04.650161: step: 628/527, loss: 0.009839916601777077
2023-01-21 09:43:05.785758: step: 632/527, loss: 0.10135894268751144
2023-01-21 09:43:06.894795: step: 636/527, loss: 0.10231619328260422
2023-01-21 09:43:07.985777: step: 640/527, loss: 0.06040620803833008
2023-01-21 09:43:09.100029: step: 644/527, loss: 0.027965473011136055
2023-01-21 09:43:10.261782: step: 648/527, loss: 0.046386051923036575
2023-01-21 09:43:11.345589: step: 652/527, loss: 0.06775007396936417
2023-01-21 09:43:12.441739: step: 656/527, loss: 0.06058797985315323
2023-01-21 09:43:13.565956: step: 660/527, loss: 0.090027816593647
2023-01-21 09:43:14.671597: step: 664/527, loss: 0.07801564782857895
2023-01-21 09:43:15.806398: step: 668/527, loss: 0.08385339379310608
2023-01-21 09:43:16.895025: step: 672/527, loss: 0.08331829309463501
2023-01-21 09:43:17.991937: step: 676/527, loss: 0.6659325361251831
2023-01-21 09:43:19.105043: step: 680/527, loss: 0.7293831706047058
2023-01-21 09:43:20.286530: step: 684/527, loss: 0.05969591438770294
2023-01-21 09:43:21.388352: step: 688/527, loss: 0.6861038208007812
2023-01-21 09:43:22.483411: step: 692/527, loss: 0.04558753967285156
2023-01-21 09:43:23.613748: step: 696/527, loss: 0.01676201820373535
2023-01-21 09:43:24.691931: step: 700/527, loss: 0.09361982345581055
2023-01-21 09:43:25.795842: step: 704/527, loss: 0.09973125159740448
2023-01-21 09:43:26.916073: step: 708/527, loss: 0.06737489998340607
2023-01-21 09:43:28.048050: step: 712/527, loss: 0.19065189361572266
2023-01-21 09:43:29.160566: step: 716/527, loss: 0.030136965215206146
2023-01-21 09:43:30.268843: step: 720/527, loss: 0.19711259007453918
2023-01-21 09:43:31.365327: step: 724/527, loss: 0.10855312645435333
2023-01-21 09:43:32.533838: step: 728/527, loss: 0.19465523958206177
2023-01-21 09:43:33.638187: step: 732/527, loss: 0.26685845851898193
2023-01-21 09:43:34.757480: step: 736/527, loss: 0.22611409425735474
2023-01-21 09:43:35.868944: step: 740/527, loss: 0.1645619422197342
2023-01-21 09:43:36.975959: step: 744/527, loss: 0.20419436693191528
2023-01-21 09:43:38.111651: step: 748/527, loss: 0.3499457836151123
2023-01-21 09:43:39.225576: step: 752/527, loss: 0.06739039719104767
2023-01-21 09:43:40.345223: step: 756/527, loss: 0.06154441833496094
2023-01-21 09:43:41.463962: step: 760/527, loss: 0.12735189497470856
2023-01-21 09:43:42.581140: step: 764/527, loss: 0.15564127266407013
2023-01-21 09:43:43.690382: step: 768/527, loss: 0.08485320955514908
2023-01-21 09:43:44.823204: step: 772/527, loss: 0.11284656822681427
2023-01-21 09:43:45.969566: step: 776/527, loss: 0.06927724182605743
2023-01-21 09:43:47.058317: step: 780/527, loss: 0.32635122537612915
2023-01-21 09:43:48.180573: step: 784/527, loss: 0.10320340096950531
2023-01-21 09:43:49.279996: step: 788/527, loss: 0.10767364501953125
2023-01-21 09:43:50.397617: step: 792/527, loss: 0.09046483039855957
2023-01-21 09:43:51.525133: step: 796/527, loss: 0.11691588908433914
2023-01-21 09:43:52.631075: step: 800/527, loss: 0.3364820182323456
2023-01-21 09:43:53.777894: step: 804/527, loss: 0.0500788688659668
2023-01-21 09:43:54.881447: step: 808/527, loss: 0.12557896971702576
2023-01-21 09:43:55.960366: step: 812/527, loss: 0.054504796862602234
2023-01-21 09:43:57.100704: step: 816/527, loss: 0.1100677102804184
2023-01-21 09:43:58.219381: step: 820/527, loss: 0.3233882784843445
2023-01-21 09:43:59.334145: step: 824/527, loss: 0.09544192254543304
2023-01-21 09:44:00.465539: step: 828/527, loss: 0.7357593774795532
2023-01-21 09:44:01.561978: step: 832/527, loss: 0.2747243642807007
2023-01-21 09:44:02.690406: step: 836/527, loss: 0.08035686612129211
2023-01-21 09:44:03.825197: step: 840/527, loss: 0.08558931201696396
2023-01-21 09:44:04.965922: step: 844/527, loss: 0.13456383347511292
2023-01-21 09:44:06.080203: step: 848/527, loss: 0.21272344887256622
2023-01-21 09:44:07.201201: step: 852/527, loss: 0.17726688086986542
2023-01-21 09:44:08.323987: step: 856/527, loss: 0.12897071242332458
2023-01-21 09:44:09.449191: step: 860/527, loss: 0.8192094564437866
2023-01-21 09:44:10.581880: step: 864/527, loss: 0.16395854949951172
2023-01-21 09:44:11.698401: step: 868/527, loss: 0.12887009978294373
2023-01-21 09:44:12.841847: step: 872/527, loss: 0.13416939973831177
2023-01-21 09:44:13.934575: step: 876/527, loss: 0.055257562547922134
2023-01-21 09:44:15.050420: step: 880/527, loss: 0.24006567895412445
2023-01-21 09:44:16.135141: step: 884/527, loss: 2.345228910446167
2023-01-21 09:44:17.244564: step: 888/527, loss: 0.10308703780174255
2023-01-21 09:44:18.393469: step: 892/527, loss: 0.01845252513885498
2023-01-21 09:44:19.491247: step: 896/527, loss: 0.04147949442267418
2023-01-21 09:44:20.597409: step: 900/527, loss: 0.0654093325138092
2023-01-21 09:44:21.736874: step: 904/527, loss: 0.08820018917322159
2023-01-21 09:44:22.847060: step: 908/527, loss: 0.05852871015667915
2023-01-21 09:44:23.946773: step: 912/527, loss: 0.18398791551589966
2023-01-21 09:44:25.069302: step: 916/527, loss: 0.2626205384731293
2023-01-21 09:44:26.171633: step: 920/527, loss: 0.029584836214780807
2023-01-21 09:44:27.294950: step: 924/527, loss: 0.046279050409793854
2023-01-21 09:44:28.429145: step: 928/527, loss: 0.042564962059259415
2023-01-21 09:44:29.519623: step: 932/527, loss: 0.14204120635986328
2023-01-21 09:44:30.616387: step: 936/527, loss: 0.2715371251106262
2023-01-21 09:44:31.712686: step: 940/527, loss: 0.028046369552612305
2023-01-21 09:44:32.807554: step: 944/527, loss: 0.11839208751916885
2023-01-21 09:44:33.918096: step: 948/527, loss: 0.03891754150390625
2023-01-21 09:44:35.035220: step: 952/527, loss: 0.06709346920251846
2023-01-21 09:44:36.136994: step: 956/527, loss: 0.16514796018600464
2023-01-21 09:44:37.262420: step: 960/527, loss: 0.2718348503112793
2023-01-21 09:44:38.360148: step: 964/527, loss: 0.04165635257959366
2023-01-21 09:44:39.472863: step: 968/527, loss: 0.10216931998729706
2023-01-21 09:44:40.596830: step: 972/527, loss: 0.7923317551612854
2023-01-21 09:44:41.710881: step: 976/527, loss: 0.11979684978723526
2023-01-21 09:44:42.835565: step: 980/527, loss: 0.03786206617951393
2023-01-21 09:44:43.949414: step: 984/527, loss: 0.051749035716056824
2023-01-21 09:44:45.060720: step: 988/527, loss: 0.12070188671350479
2023-01-21 09:44:46.187243: step: 992/527, loss: 0.09267584979534149
2023-01-21 09:44:47.302086: step: 996/527, loss: 0.03320141136646271
2023-01-21 09:44:48.415799: step: 1000/527, loss: 0.06649112701416016
2023-01-21 09:44:49.530933: step: 1004/527, loss: 0.010967493057250977
2023-01-21 09:44:50.653864: step: 1008/527, loss: 0.057869818061590195
2023-01-21 09:44:51.759115: step: 1012/527, loss: 0.06803999841213226
2023-01-21 09:44:52.877622: step: 1016/527, loss: 0.08660201728343964
2023-01-21 09:44:53.972229: step: 1020/527, loss: 0.013811302371323109
2023-01-21 09:44:55.069060: step: 1024/527, loss: 0.1948748081922531
2023-01-21 09:44:56.189172: step: 1028/527, loss: 0.20716705918312073
2023-01-21 09:44:57.322669: step: 1032/527, loss: 0.08804524689912796
2023-01-21 09:44:58.453109: step: 1036/527, loss: 0.7179480791091919
2023-01-21 09:44:59.613466: step: 1040/527, loss: 0.055435944348573685
2023-01-21 09:45:00.730541: step: 1044/527, loss: 0.1197391003370285
2023-01-21 09:45:01.848956: step: 1048/527, loss: 0.03375749662518501
2023-01-21 09:45:02.958652: step: 1052/527, loss: 0.09264302253723145
2023-01-21 09:45:04.084881: step: 1056/527, loss: 0.06527318805456161
2023-01-21 09:45:05.217083: step: 1060/527, loss: 0.1340409219264984
2023-01-21 09:45:06.325190: step: 1064/527, loss: 0.0313778892159462
2023-01-21 09:45:07.447529: step: 1068/527, loss: 0.018294524401426315
2023-01-21 09:45:08.544646: step: 1072/527, loss: 0.25492602586746216
2023-01-21 09:45:09.632153: step: 1076/527, loss: 0.08791275322437286
2023-01-21 09:45:10.758764: step: 1080/527, loss: 0.061396028846502304
2023-01-21 09:45:11.863996: step: 1084/527, loss: 0.09266863018274307
2023-01-21 09:45:12.979671: step: 1088/527, loss: 0.6098822355270386
2023-01-21 09:45:14.098811: step: 1092/527, loss: 0.05562954023480415
2023-01-21 09:45:15.201025: step: 1096/527, loss: 0.08444322645664215
2023-01-21 09:45:16.324836: step: 1100/527, loss: 0.026590727269649506
2023-01-21 09:45:17.465121: step: 1104/527, loss: 0.2448311746120453
2023-01-21 09:45:18.585818: step: 1108/527, loss: 0.030560877174139023
2023-01-21 09:45:19.705712: step: 1112/527, loss: 0.12442392855882645
2023-01-21 09:45:20.841874: step: 1116/527, loss: 0.18386010825634003
2023-01-21 09:45:21.985745: step: 1120/527, loss: 0.05577525869011879
2023-01-21 09:45:23.123756: step: 1124/527, loss: 0.27393731474876404
2023-01-21 09:45:24.278947: step: 1128/527, loss: 0.12740007042884827
2023-01-21 09:45:25.416034: step: 1132/527, loss: 0.07875290513038635
2023-01-21 09:45:26.541979: step: 1136/527, loss: 0.08007045090198517
2023-01-21 09:45:27.636424: step: 1140/527, loss: 0.061043668538331985
2023-01-21 09:45:28.750394: step: 1144/527, loss: 0.5175440311431885
2023-01-21 09:45:29.851249: step: 1148/527, loss: 0.04177398979663849
2023-01-21 09:45:30.969706: step: 1152/527, loss: 0.11078719794750214
2023-01-21 09:45:32.117661: step: 1156/527, loss: 0.03102750889956951
2023-01-21 09:45:33.264044: step: 1160/527, loss: 1.415436029434204
2023-01-21 09:45:34.386060: step: 1164/527, loss: 0.1366496980190277
2023-01-21 09:45:35.490526: step: 1168/527, loss: 0.07057762145996094
2023-01-21 09:45:36.606788: step: 1172/527, loss: 0.08305162191390991
2023-01-21 09:45:37.730635: step: 1176/527, loss: 0.3974289000034332
2023-01-21 09:45:38.827222: step: 1180/527, loss: 0.12361840903759003
2023-01-21 09:45:39.936681: step: 1184/527, loss: 0.09007033705711365
2023-01-21 09:45:41.056586: step: 1188/527, loss: 0.0807943046092987
2023-01-21 09:45:42.164870: step: 1192/527, loss: 0.05922601372003555
2023-01-21 09:45:43.272293: step: 1196/527, loss: 0.12947070598602295
2023-01-21 09:45:44.373070: step: 1200/527, loss: 0.004709864035248756
2023-01-21 09:45:45.472641: step: 1204/527, loss: 0.1928752213716507
2023-01-21 09:45:46.595027: step: 1208/527, loss: 0.22018328309059143
2023-01-21 09:45:47.733330: step: 1212/527, loss: 0.11299343407154083
2023-01-21 09:45:48.852304: step: 1216/527, loss: 0.04332923889160156
2023-01-21 09:45:49.978405: step: 1220/527, loss: 0.30185890197753906
2023-01-21 09:45:51.121513: step: 1224/527, loss: 0.06659789383411407
2023-01-21 09:45:52.273794: step: 1228/527, loss: 0.13564452528953552
2023-01-21 09:45:53.393976: step: 1232/527, loss: 0.06867778301239014
2023-01-21 09:45:54.493977: step: 1236/527, loss: 0.11100301891565323
2023-01-21 09:45:55.607358: step: 1240/527, loss: 0.0544523224234581
2023-01-21 09:45:56.728389: step: 1244/527, loss: 0.0866757407784462
2023-01-21 09:45:57.845489: step: 1248/527, loss: 0.07921364158391953
2023-01-21 09:45:58.956748: step: 1252/527, loss: 0.05777778476476669
2023-01-21 09:46:00.072466: step: 1256/527, loss: 0.14223213493824005
2023-01-21 09:46:01.195284: step: 1260/527, loss: 0.030316973105072975
2023-01-21 09:46:02.325318: step: 1264/527, loss: 0.12432222068309784
2023-01-21 09:46:03.431856: step: 1268/527, loss: 0.0868983268737793
2023-01-21 09:46:04.556582: step: 1272/527, loss: 0.07804088294506073
2023-01-21 09:46:05.682724: step: 1276/527, loss: 0.04638509824872017
2023-01-21 09:46:06.785388: step: 1280/527, loss: 0.12446976453065872
2023-01-21 09:46:07.912817: step: 1284/527, loss: 0.09551630169153214
2023-01-21 09:46:09.015790: step: 1288/527, loss: 0.06230277940630913
2023-01-21 09:46:10.141930: step: 1292/527, loss: 0.09850476682186127
2023-01-21 09:46:11.257982: step: 1296/527, loss: 0.15821805596351624
2023-01-21 09:46:12.396353: step: 1300/527, loss: 0.2274303436279297
2023-01-21 09:46:13.503002: step: 1304/527, loss: 0.10555467754602432
2023-01-21 09:46:14.606610: step: 1308/527, loss: 0.14376716315746307
2023-01-21 09:46:15.694045: step: 1312/527, loss: 0.04370608553290367
2023-01-21 09:46:16.785965: step: 1316/527, loss: 0.16287164390087128
2023-01-21 09:46:17.871837: step: 1320/527, loss: 0.0780348852276802
2023-01-21 09:46:18.973176: step: 1324/527, loss: 0.30689454078674316
2023-01-21 09:46:20.092164: step: 1328/527, loss: 0.22151941061019897
2023-01-21 09:46:21.223195: step: 1332/527, loss: 0.11925353854894638
2023-01-21 09:46:22.343547: step: 1336/527, loss: 0.04480161517858505
2023-01-21 09:46:23.460644: step: 1340/527, loss: 0.3220299482345581
2023-01-21 09:46:24.604000: step: 1344/527, loss: 0.05163154751062393
2023-01-21 09:46:25.743393: step: 1348/527, loss: 0.5688139200210571
2023-01-21 09:46:26.839761: step: 1352/527, loss: 0.07556018978357315
2023-01-21 09:46:27.971519: step: 1356/527, loss: 0.18702572584152222
2023-01-21 09:46:29.119875: step: 1360/527, loss: 0.1352296769618988
2023-01-21 09:46:30.217237: step: 1364/527, loss: 0.06931314617395401
2023-01-21 09:46:31.299100: step: 1368/527, loss: 0.07552079856395721
2023-01-21 09:46:32.412936: step: 1372/527, loss: 0.19485560059547424
2023-01-21 09:46:33.558027: step: 1376/527, loss: 0.202061265707016
2023-01-21 09:46:34.639488: step: 1380/527, loss: 0.14096584916114807
2023-01-21 09:46:35.789413: step: 1384/527, loss: 0.06458797305822372
2023-01-21 09:46:36.881467: step: 1388/527, loss: 0.0766574889421463
2023-01-21 09:46:37.988162: step: 1392/527, loss: 0.09638500213623047
2023-01-21 09:46:39.088748: step: 1396/527, loss: 0.032813072204589844
2023-01-21 09:46:40.207598: step: 1400/527, loss: 1.4972236156463623
2023-01-21 09:46:41.323249: step: 1404/527, loss: 0.08110485225915909
2023-01-21 09:46:42.430413: step: 1408/527, loss: 0.03479523956775665
2023-01-21 09:46:43.560171: step: 1412/527, loss: 0.16209346055984497
2023-01-21 09:46:44.695019: step: 1416/527, loss: 0.128150075674057
2023-01-21 09:46:45.868492: step: 1420/527, loss: 0.5813602805137634
2023-01-21 09:46:46.992007: step: 1424/527, loss: 0.032004550099372864
2023-01-21 09:46:48.102205: step: 1428/527, loss: 0.0892861932516098
2023-01-21 09:46:49.240029: step: 1432/527, loss: 0.4798944890499115
2023-01-21 09:46:50.408483: step: 1436/527, loss: 0.15733404457569122
2023-01-21 09:46:51.524377: step: 1440/527, loss: 0.04112134128808975
2023-01-21 09:46:52.630612: step: 1444/527, loss: 0.13937664031982422
2023-01-21 09:46:53.734005: step: 1448/527, loss: 0.12938299775123596
2023-01-21 09:46:54.832497: step: 1452/527, loss: 0.17721039056777954
2023-01-21 09:46:55.906518: step: 1456/527, loss: 0.07398548722267151
2023-01-21 09:46:57.025805: step: 1460/527, loss: 0.07432279735803604
2023-01-21 09:46:58.172119: step: 1464/527, loss: 0.24189713597297668
2023-01-21 09:46:59.292436: step: 1468/527, loss: 0.20307216048240662
2023-01-21 09:47:00.371522: step: 1472/527, loss: 0.07236690819263458
2023-01-21 09:47:01.473865: step: 1476/527, loss: 0.09924187511205673
2023-01-21 09:47:02.583695: step: 1480/527, loss: 0.017452048137784004
2023-01-21 09:47:03.719384: step: 1484/527, loss: 0.21310366690158844
2023-01-21 09:47:04.818203: step: 1488/527, loss: 0.08805880695581436
2023-01-21 09:47:05.941033: step: 1492/527, loss: 0.060295961797237396
2023-01-21 09:47:07.036627: step: 1496/527, loss: 0.5504152774810791
2023-01-21 09:47:08.120185: step: 1500/527, loss: 0.2574518322944641
2023-01-21 09:47:09.226423: step: 1504/527, loss: 0.13807792961597443
2023-01-21 09:47:10.347897: step: 1508/527, loss: 0.1989975869655609
2023-01-21 09:47:11.444027: step: 1512/527, loss: 0.33166128396987915
2023-01-21 09:47:12.539628: step: 1516/527, loss: 0.7003905773162842
2023-01-21 09:47:13.682241: step: 1520/527, loss: 0.05732369422912598
2023-01-21 09:47:14.766861: step: 1524/527, loss: 0.03219633176922798
2023-01-21 09:47:15.881769: step: 1528/527, loss: 0.09347958117723465
2023-01-21 09:47:17.010578: step: 1532/527, loss: 0.6772009134292603
2023-01-21 09:47:18.113018: step: 1536/527, loss: 0.1210743859410286
2023-01-21 09:47:19.213378: step: 1540/527, loss: 0.09642343968153
2023-01-21 09:47:20.338121: step: 1544/527, loss: 0.09271785616874695
2023-01-21 09:47:21.440134: step: 1548/527, loss: 0.011908723041415215
2023-01-21 09:47:22.595027: step: 1552/527, loss: 0.25633230805397034
2023-01-21 09:47:23.734069: step: 1556/527, loss: 0.1170506477355957
2023-01-21 09:47:24.824188: step: 1560/527, loss: 0.14003857970237732
2023-01-21 09:47:25.936776: step: 1564/527, loss: 0.08557999134063721
2023-01-21 09:47:27.061274: step: 1568/527, loss: 0.048417091369628906
2023-01-21 09:47:28.187632: step: 1572/527, loss: 0.15803876519203186
2023-01-21 09:47:29.301922: step: 1576/527, loss: 0.029480792582035065
2023-01-21 09:47:30.413399: step: 1580/527, loss: 0.03046315908432007
2023-01-21 09:47:31.550526: step: 1584/527, loss: 0.1353861689567566
2023-01-21 09:47:32.650855: step: 1588/527, loss: 0.1786683052778244
2023-01-21 09:47:33.770755: step: 1592/527, loss: 0.523049533367157
2023-01-21 09:47:34.893761: step: 1596/527, loss: 0.27000704407691956
2023-01-21 09:47:35.997835: step: 1600/527, loss: 0.5134299993515015
2023-01-21 09:47:37.155772: step: 1604/527, loss: 0.064641572535038
2023-01-21 09:47:38.304273: step: 1608/527, loss: 0.2145313173532486
2023-01-21 09:47:39.427685: step: 1612/527, loss: 0.08705604076385498
2023-01-21 09:47:40.554897: step: 1616/527, loss: 0.02277245558798313
2023-01-21 09:47:41.638916: step: 1620/527, loss: 0.014551687985658646
2023-01-21 09:47:42.762198: step: 1624/527, loss: 0.04566545411944389
2023-01-21 09:47:43.878907: step: 1628/527, loss: 0.14497403800487518
2023-01-21 09:47:44.985404: step: 1632/527, loss: 0.11718883365392685
2023-01-21 09:47:46.106553: step: 1636/527, loss: 0.01563549041748047
2023-01-21 09:47:47.254468: step: 1640/527, loss: 0.03902807459235191
2023-01-21 09:47:48.367438: step: 1644/527, loss: 0.07829437404870987
2023-01-21 09:47:49.464845: step: 1648/527, loss: 0.08640975505113602
2023-01-21 09:47:50.604520: step: 1652/527, loss: 0.12538833916187286
2023-01-21 09:47:51.703781: step: 1656/527, loss: 0.08356447517871857
2023-01-21 09:47:52.799141: step: 1660/527, loss: 0.08952084183692932
2023-01-21 09:47:53.902954: step: 1664/527, loss: 0.09460973739624023
2023-01-21 09:47:54.986758: step: 1668/527, loss: 0.14672046899795532
2023-01-21 09:47:56.078000: step: 1672/527, loss: 0.03532552719116211
2023-01-21 09:47:57.179408: step: 1676/527, loss: 0.49581262469291687
2023-01-21 09:47:58.310143: step: 1680/527, loss: 0.07083387672901154
2023-01-21 09:47:59.431675: step: 1684/527, loss: 0.17162543535232544
2023-01-21 09:48:00.572696: step: 1688/527, loss: 0.22113552689552307
2023-01-21 09:48:01.727889: step: 1692/527, loss: 0.1550757884979248
2023-01-21 09:48:02.841088: step: 1696/527, loss: 0.05860729515552521
2023-01-21 09:48:03.932547: step: 1700/527, loss: 0.035207558423280716
2023-01-21 09:48:05.049434: step: 1704/527, loss: 0.3124457597732544
2023-01-21 09:48:06.137846: step: 1708/527, loss: 0.17580759525299072
2023-01-21 09:48:07.286905: step: 1712/527, loss: 0.0979655310511589
2023-01-21 09:48:08.399968: step: 1716/527, loss: 0.0509522445499897
2023-01-21 09:48:09.496775: step: 1720/527, loss: 0.6463108658790588
2023-01-21 09:48:10.604586: step: 1724/527, loss: 0.08071768283843994
2023-01-21 09:48:11.730581: step: 1728/527, loss: 0.15986299514770508
2023-01-21 09:48:12.864583: step: 1732/527, loss: 0.08089923858642578
2023-01-21 09:48:13.987983: step: 1736/527, loss: 0.025244617834687233
2023-01-21 09:48:15.113728: step: 1740/527, loss: 0.0182831771671772
2023-01-21 09:48:16.209829: step: 1744/527, loss: 0.3164002299308777
2023-01-21 09:48:17.319952: step: 1748/527, loss: 0.20178470015525818
2023-01-21 09:48:18.419183: step: 1752/527, loss: 0.09134826809167862
2023-01-21 09:48:19.514260: step: 1756/527, loss: 0.01455754041671753
2023-01-21 09:48:20.655965: step: 1760/527, loss: 0.08732642978429794
2023-01-21 09:48:21.757705: step: 1764/527, loss: 0.09809169918298721
2023-01-21 09:48:22.877073: step: 1768/527, loss: 0.4199890196323395
2023-01-21 09:48:23.997743: step: 1772/527, loss: 0.04358282312750816
2023-01-21 09:48:25.116985: step: 1776/527, loss: 0.1687115728855133
2023-01-21 09:48:26.270788: step: 1780/527, loss: 0.07507472485303879
2023-01-21 09:48:27.371854: step: 1784/527, loss: 0.2308574616909027
2023-01-21 09:48:28.488418: step: 1788/527, loss: 0.014399625360965729
2023-01-21 09:48:29.602116: step: 1792/527, loss: 0.11223955452442169
2023-01-21 09:48:30.719408: step: 1796/527, loss: 0.08046016842126846
2023-01-21 09:48:31.811695: step: 1800/527, loss: 0.04589500278234482
2023-01-21 09:48:32.931003: step: 1804/527, loss: 0.19525334239006042
2023-01-21 09:48:34.067495: step: 1808/527, loss: 0.0960565134882927
2023-01-21 09:48:35.205267: step: 1812/527, loss: 0.12751737236976624
2023-01-21 09:48:36.311703: step: 1816/527, loss: 0.7705336809158325
2023-01-21 09:48:37.411512: step: 1820/527, loss: 0.11391840130090714
2023-01-21 09:48:38.530758: step: 1824/527, loss: 0.06742487102746964
2023-01-21 09:48:39.668265: step: 1828/527, loss: 0.0063932896591722965
2023-01-21 09:48:40.763903: step: 1832/527, loss: 0.08857765793800354
2023-01-21 09:48:41.871173: step: 1836/527, loss: 0.29340478777885437
2023-01-21 09:48:42.997737: step: 1840/527, loss: 0.027070429176092148
2023-01-21 09:48:44.124427: step: 1844/527, loss: 0.12149587273597717
2023-01-21 09:48:45.226734: step: 1848/527, loss: 0.0981663316488266
2023-01-21 09:48:46.365901: step: 1852/527, loss: 0.016504479572176933
2023-01-21 09:48:47.487230: step: 1856/527, loss: 0.6940372586250305
2023-01-21 09:48:48.617116: step: 1860/527, loss: 0.058921121060848236
2023-01-21 09:48:49.761733: step: 1864/527, loss: 0.2871111035346985
2023-01-21 09:48:50.921226: step: 1868/527, loss: 0.07852087169885635
2023-01-21 09:48:52.010309: step: 1872/527, loss: 0.038393642753362656
2023-01-21 09:48:53.120280: step: 1876/527, loss: 0.02996664121747017
2023-01-21 09:48:54.242392: step: 1880/527, loss: 1.200020670890808
2023-01-21 09:48:55.334889: step: 1884/527, loss: 0.08861713856458664
2023-01-21 09:48:56.428308: step: 1888/527, loss: 0.036472223699092865
2023-01-21 09:48:57.528088: step: 1892/527, loss: 0.10223941504955292
2023-01-21 09:48:58.635621: step: 1896/527, loss: 0.155283123254776
2023-01-21 09:48:59.720823: step: 1900/527, loss: 0.00884790439158678
2023-01-21 09:49:00.867230: step: 1904/527, loss: 0.08166541904211044
2023-01-21 09:49:02.000215: step: 1908/527, loss: 0.0051506999880075455
2023-01-21 09:49:03.126525: step: 1912/527, loss: 0.08418627083301544
2023-01-21 09:49:04.244762: step: 1916/527, loss: 0.0670836865901947
2023-01-21 09:49:05.386140: step: 1920/527, loss: 0.24358788132667542
2023-01-21 09:49:06.485844: step: 1924/527, loss: 0.0745553970336914
2023-01-21 09:49:07.593607: step: 1928/527, loss: 0.14394140243530273
2023-01-21 09:49:08.701870: step: 1932/527, loss: 0.05049104988574982
2023-01-21 09:49:09.807009: step: 1936/527, loss: 0.5881040096282959
2023-01-21 09:49:10.899646: step: 1940/527, loss: 0.08709597587585449
2023-01-21 09:49:12.005248: step: 1944/527, loss: 0.13468265533447266
2023-01-21 09:49:13.152727: step: 1948/527, loss: 0.1144346296787262
2023-01-21 09:49:14.256900: step: 1952/527, loss: 0.1711897850036621
2023-01-21 09:49:15.362406: step: 1956/527, loss: 0.03764772415161133
2023-01-21 09:49:16.490466: step: 1960/527, loss: 0.03174591064453125
2023-01-21 09:49:17.616348: step: 1964/527, loss: 0.03121805191040039
2023-01-21 09:49:18.745802: step: 1968/527, loss: 0.11928224563598633
2023-01-21 09:49:19.865022: step: 1972/527, loss: 0.14947627484798431
2023-01-21 09:49:20.966431: step: 1976/527, loss: 0.1037091314792633
2023-01-21 09:49:22.092618: step: 1980/527, loss: 0.05550746992230415
2023-01-21 09:49:23.196214: step: 1984/527, loss: 0.026887917891144753
2023-01-21 09:49:24.326658: step: 1988/527, loss: 0.11578197777271271
2023-01-21 09:49:25.417067: step: 1992/527, loss: 0.15976767241954803
2023-01-21 09:49:26.543513: step: 1996/527, loss: 0.11969108134508133
2023-01-21 09:49:27.634071: step: 2000/527, loss: 0.055633485317230225
2023-01-21 09:49:28.725511: step: 2004/527, loss: 0.019266510382294655
2023-01-21 09:49:29.845657: step: 2008/527, loss: 0.0732610747218132
2023-01-21 09:49:30.919584: step: 2012/527, loss: 0.07769189029932022
2023-01-21 09:49:32.036547: step: 2016/527, loss: 0.06608028709888458
2023-01-21 09:49:33.148824: step: 2020/527, loss: 0.06438541412353516
2023-01-21 09:49:34.228850: step: 2024/527, loss: 0.02088642120361328
2023-01-21 09:49:35.344159: step: 2028/527, loss: 0.014204931445419788
2023-01-21 09:49:36.454539: step: 2032/527, loss: 0.062448449432849884
2023-01-21 09:49:37.575404: step: 2036/527, loss: 0.2478632926940918
2023-01-21 09:49:38.688244: step: 2040/527, loss: 0.13856564462184906
2023-01-21 09:49:39.804867: step: 2044/527, loss: 0.025881171226501465
2023-01-21 09:49:40.903534: step: 2048/527, loss: 0.06553387641906738
2023-01-21 09:49:42.030908: step: 2052/527, loss: 0.09359939396381378
2023-01-21 09:49:43.140140: step: 2056/527, loss: 0.09992627799510956
2023-01-21 09:49:44.260301: step: 2060/527, loss: 0.044696759432554245
2023-01-21 09:49:45.353923: step: 2064/527, loss: 0.0404149554669857
2023-01-21 09:49:46.440675: step: 2068/527, loss: 0.08958516269922256
2023-01-21 09:49:47.574539: step: 2072/527, loss: 0.03951120376586914
2023-01-21 09:49:48.699552: step: 2076/527, loss: 0.11337490379810333
2023-01-21 09:49:49.809216: step: 2080/527, loss: 0.20729884505271912
2023-01-21 09:49:50.961213: step: 2084/527, loss: 0.09457206726074219
2023-01-21 09:49:52.048000: step: 2088/527, loss: 0.10443481802940369
2023-01-21 09:49:53.151382: step: 2092/527, loss: 0.16657915711402893
2023-01-21 09:49:54.285045: step: 2096/527, loss: 0.15816065669059753
2023-01-21 09:49:55.416010: step: 2100/527, loss: 0.9830401539802551
2023-01-21 09:49:56.546167: step: 2104/527, loss: 0.03774967044591904
2023-01-21 09:49:57.634449: step: 2108/527, loss: 0.08231468498706818
==================================================
Loss: 0.160
--------------------
Dev: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Test: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Chinese: {'event': {'p': 0.5512820512820513, 'r': 0.7962962962962963, 'f1': 0.6515151515151515}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
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': 7}
Russian: {'event': {'p': 0.5555555555555556, 'r': 0.5555555555555556, 'f1': 0.5555555555555556}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
New best korean model...
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 09:50:44.083730: step: 4/527, loss: 0.0662505179643631
2023-01-21 09:50:45.210812: step: 8/527, loss: 0.02305145189166069
2023-01-21 09:50:46.356452: step: 12/527, loss: 0.04371337592601776
2023-01-21 09:50:47.454076: step: 16/527, loss: 0.10356616973876953
2023-01-21 09:50:48.552939: step: 20/527, loss: 0.10588302463293076
2023-01-21 09:50:49.661226: step: 24/527, loss: 0.11706677079200745
2023-01-21 09:50:50.746659: step: 28/527, loss: 0.009508872404694557
2023-01-21 09:50:51.866536: step: 32/527, loss: 0.08286657929420471
2023-01-21 09:50:52.983842: step: 36/527, loss: 0.0660085678100586
2023-01-21 09:50:54.095535: step: 40/527, loss: 0.06506600230932236
2023-01-21 09:50:55.210683: step: 44/527, loss: 0.04293499141931534
2023-01-21 09:50:56.307366: step: 48/527, loss: 0.005032634828239679
2023-01-21 09:50:57.423101: step: 52/527, loss: 0.24146828055381775
2023-01-21 09:50:58.542155: step: 56/527, loss: 0.024000121280550957
2023-01-21 09:50:59.663604: step: 60/527, loss: 0.021087074652314186
2023-01-21 09:51:00.739230: step: 64/527, loss: 0.12115335464477539
2023-01-21 09:51:01.854541: step: 68/527, loss: 0.2353973388671875
2023-01-21 09:51:02.983581: step: 72/527, loss: 0.03964404761791229
2023-01-21 09:51:04.116937: step: 76/527, loss: 0.11088452488183975
2023-01-21 09:51:05.221659: step: 80/527, loss: 0.04457240179181099
2023-01-21 09:51:06.342456: step: 84/527, loss: 0.39419201016426086
2023-01-21 09:51:07.480781: step: 88/527, loss: 0.1277870237827301
2023-01-21 09:51:08.646831: step: 92/527, loss: 0.22232437133789062
2023-01-21 09:51:09.783386: step: 96/527, loss: 0.06264851242303848
2023-01-21 09:51:10.896718: step: 100/527, loss: 0.09544029831886292
2023-01-21 09:51:12.026001: step: 104/527, loss: 0.05071617290377617
2023-01-21 09:51:13.127291: step: 108/527, loss: 0.08496695011854172
2023-01-21 09:51:14.235869: step: 112/527, loss: 0.1718399077653885
2023-01-21 09:51:15.357552: step: 116/527, loss: 0.5687503814697266
2023-01-21 09:51:16.457962: step: 120/527, loss: 0.0852271094918251
2023-01-21 09:51:17.571971: step: 124/527, loss: 0.0876384750008583
2023-01-21 09:51:18.663208: step: 128/527, loss: 0.03890504688024521
2023-01-21 09:51:19.760459: step: 132/527, loss: 0.011740398593246937
2023-01-21 09:51:20.935315: step: 136/527, loss: 0.04613227769732475
2023-01-21 09:51:22.039129: step: 140/527, loss: 0.02544255182147026
2023-01-21 09:51:23.176592: step: 144/527, loss: 0.07478688657283783
2023-01-21 09:51:24.263858: step: 148/527, loss: 0.02361168898642063
2023-01-21 09:51:25.395860: step: 152/527, loss: 0.10690093040466309
2023-01-21 09:51:26.512442: step: 156/527, loss: 0.05181805044412613
2023-01-21 09:51:27.640857: step: 160/527, loss: 0.005225944332778454
2023-01-21 09:51:28.736423: step: 164/527, loss: 0.033143140375614166
2023-01-21 09:51:29.856332: step: 168/527, loss: 0.13845062255859375
2023-01-21 09:51:30.981778: step: 172/527, loss: 0.007614517118781805
2023-01-21 09:51:32.121813: step: 176/527, loss: 0.03459987789392471
2023-01-21 09:51:33.236029: step: 180/527, loss: 0.051847122609615326
2023-01-21 09:51:34.327674: step: 184/527, loss: 0.09796829521656036
2023-01-21 09:51:35.439101: step: 188/527, loss: 2.120713710784912
2023-01-21 09:51:36.538324: step: 192/527, loss: 0.07133102416992188
2023-01-21 09:51:37.658175: step: 196/527, loss: 0.027388669550418854
2023-01-21 09:51:38.763950: step: 200/527, loss: 0.09015445411205292
2023-01-21 09:51:39.882409: step: 204/527, loss: 0.14747075736522675
2023-01-21 09:51:41.010881: step: 208/527, loss: 0.06787262111902237
2023-01-21 09:51:42.162663: step: 212/527, loss: 0.037381600588560104
2023-01-21 09:51:43.300981: step: 216/527, loss: 0.08827958256006241
2023-01-21 09:51:44.407881: step: 220/527, loss: 0.06868445873260498
2023-01-21 09:51:45.537998: step: 224/527, loss: 2.1922600269317627
2023-01-21 09:51:46.669573: step: 228/527, loss: 0.9606868624687195
2023-01-21 09:51:47.787122: step: 232/527, loss: 0.1942111998796463
2023-01-21 09:51:48.899495: step: 236/527, loss: 0.014286041259765625
2023-01-21 09:51:50.020566: step: 240/527, loss: 0.03378438949584961
2023-01-21 09:51:51.150411: step: 244/527, loss: 0.21601086854934692
2023-01-21 09:51:52.250062: step: 248/527, loss: 0.020242786034941673
2023-01-21 09:51:53.326333: step: 252/527, loss: 0.028825949877500534
2023-01-21 09:51:54.418145: step: 256/527, loss: 0.1523379385471344
2023-01-21 09:51:55.524251: step: 260/527, loss: 0.11253499984741211
2023-01-21 09:51:56.631990: step: 264/527, loss: 0.12275314331054688
2023-01-21 09:51:57.777941: step: 268/527, loss: 0.060677219182252884
2023-01-21 09:51:58.901500: step: 272/527, loss: 0.2576586604118347
2023-01-21 09:52:00.021016: step: 276/527, loss: 0.0753047913312912
2023-01-21 09:52:01.139154: step: 280/527, loss: 0.054239656776189804
2023-01-21 09:52:02.242406: step: 284/527, loss: 0.027918243780732155
2023-01-21 09:52:03.345818: step: 288/527, loss: 0.09006933867931366
2023-01-21 09:52:04.455941: step: 292/527, loss: 0.13811007142066956
2023-01-21 09:52:05.610759: step: 296/527, loss: 0.08516483008861542
2023-01-21 09:52:06.704874: step: 300/527, loss: 0.04768932983279228
2023-01-21 09:52:07.822358: step: 304/527, loss: 0.09100484848022461
2023-01-21 09:52:08.925340: step: 308/527, loss: 0.06292619556188583
2023-01-21 09:52:10.050896: step: 312/527, loss: 0.09011035412549973
2023-01-21 09:52:11.185676: step: 316/527, loss: 0.10717058181762695
2023-01-21 09:52:12.273592: step: 320/527, loss: 0.02911229245364666
2023-01-21 09:52:13.378270: step: 324/527, loss: 0.006568622775375843
2023-01-21 09:52:14.503929: step: 328/527, loss: 0.09261999279260635
2023-01-21 09:52:15.611753: step: 332/527, loss: 0.24732133746147156
2023-01-21 09:52:16.751592: step: 336/527, loss: 0.24520960450172424
2023-01-21 09:52:17.838371: step: 340/527, loss: 0.05025935173034668
2023-01-21 09:52:18.935590: step: 344/527, loss: 0.8846839070320129
2023-01-21 09:52:20.031972: step: 348/527, loss: 0.11841163784265518
2023-01-21 09:52:21.206676: step: 352/527, loss: 0.2878393828868866
2023-01-21 09:52:22.322005: step: 356/527, loss: 0.048734474927186966
2023-01-21 09:52:23.462613: step: 360/527, loss: 0.08063755184412003
2023-01-21 09:52:24.566054: step: 364/527, loss: 0.1054808646440506
2023-01-21 09:52:25.682750: step: 368/527, loss: 0.2780340611934662
2023-01-21 09:52:26.760426: step: 372/527, loss: 0.02704916149377823
2023-01-21 09:52:27.858691: step: 376/527, loss: 0.023357870057225227
2023-01-21 09:52:28.985144: step: 380/527, loss: 0.008502197451889515
2023-01-21 09:52:30.106808: step: 384/527, loss: 0.01613173447549343
2023-01-21 09:52:31.212087: step: 388/527, loss: 0.06635679304599762
2023-01-21 09:52:32.343324: step: 392/527, loss: 0.3218851089477539
2023-01-21 09:52:33.462407: step: 396/527, loss: 0.017478466033935547
2023-01-21 09:52:34.581609: step: 400/527, loss: 0.07034631073474884
2023-01-21 09:52:35.727540: step: 404/527, loss: 0.013752173632383347
2023-01-21 09:52:36.836447: step: 408/527, loss: 0.03642425686120987
2023-01-21 09:52:37.932807: step: 412/527, loss: 0.015242003835737705
2023-01-21 09:52:39.051003: step: 416/527, loss: 0.2196635752916336
2023-01-21 09:52:40.128077: step: 420/527, loss: 0.07315365970134735
2023-01-21 09:52:41.241041: step: 424/527, loss: 0.07526598125696182
2023-01-21 09:52:42.353945: step: 428/527, loss: 0.21751748025417328
2023-01-21 09:52:43.480103: step: 432/527, loss: 0.09427070617675781
2023-01-21 09:52:44.566547: step: 436/527, loss: 0.05080757290124893
2023-01-21 09:52:45.708522: step: 440/527, loss: 0.0573611743748188
2023-01-21 09:52:46.817650: step: 444/527, loss: 0.023981476202607155
2023-01-21 09:52:47.957348: step: 448/527, loss: 0.2168409377336502
2023-01-21 09:52:49.082831: step: 452/527, loss: 0.6849062442779541
2023-01-21 09:52:50.227475: step: 456/527, loss: 0.13787278532981873
2023-01-21 09:52:51.330278: step: 460/527, loss: 0.04414663091301918
2023-01-21 09:52:52.449973: step: 464/527, loss: 0.13336792588233948
2023-01-21 09:52:53.547015: step: 468/527, loss: 0.0023910521995276213
2023-01-21 09:52:54.654219: step: 472/527, loss: 0.03397655487060547
2023-01-21 09:52:55.770454: step: 476/527, loss: 0.061437513679265976
2023-01-21 09:52:56.915724: step: 480/527, loss: 0.032884836196899414
2023-01-21 09:52:58.033652: step: 484/527, loss: 0.020434189587831497
2023-01-21 09:52:59.150905: step: 488/527, loss: 0.11455907672643661
2023-01-21 09:53:00.246682: step: 492/527, loss: 0.04375505447387695
2023-01-21 09:53:01.352875: step: 496/527, loss: 0.015088570304214954
2023-01-21 09:53:02.467646: step: 500/527, loss: 0.24854516983032227
2023-01-21 09:53:03.576078: step: 504/527, loss: 0.11807260662317276
2023-01-21 09:53:04.669620: step: 508/527, loss: 0.02047448232769966
2023-01-21 09:53:05.792191: step: 512/527, loss: 0.04716825485229492
2023-01-21 09:53:06.888656: step: 516/527, loss: 0.017246628180146217
2023-01-21 09:53:07.990852: step: 520/527, loss: 0.009725379757583141
2023-01-21 09:53:09.117342: step: 524/527, loss: 0.08599434047937393
2023-01-21 09:53:10.253238: step: 528/527, loss: 0.01509561575949192
2023-01-21 09:53:11.404512: step: 532/527, loss: 0.051949284970760345
2023-01-21 09:53:12.509973: step: 536/527, loss: 0.06595317274332047
2023-01-21 09:53:13.648609: step: 540/527, loss: 0.006334829144179821
2023-01-21 09:53:14.769454: step: 544/527, loss: 0.13365554809570312
2023-01-21 09:53:15.867235: step: 548/527, loss: 0.078265480697155
2023-01-21 09:53:16.942532: step: 552/527, loss: 0.3172203004360199
2023-01-21 09:53:18.046145: step: 556/527, loss: 0.07690558582544327
2023-01-21 09:53:19.139596: step: 560/527, loss: 0.04610862582921982
2023-01-21 09:53:20.282478: step: 564/527, loss: 0.10414648056030273
2023-01-21 09:53:21.390231: step: 568/527, loss: 1.6091641187667847
2023-01-21 09:53:22.522292: step: 572/527, loss: 0.0679062157869339
2023-01-21 09:53:23.661237: step: 576/527, loss: 0.1298515349626541
2023-01-21 09:53:24.801425: step: 580/527, loss: 0.09296919405460358
2023-01-21 09:53:25.940282: step: 584/527, loss: 0.06620721518993378
2023-01-21 09:53:27.062665: step: 588/527, loss: 0.17447753250598907
2023-01-21 09:53:28.191560: step: 592/527, loss: 0.11061620712280273
2023-01-21 09:53:29.340413: step: 596/527, loss: 0.022596264258027077
2023-01-21 09:53:30.416628: step: 600/527, loss: 0.009488009847700596
2023-01-21 09:53:31.511906: step: 604/527, loss: 0.08111634850502014
2023-01-21 09:53:32.630795: step: 608/527, loss: 0.2065192312002182
2023-01-21 09:53:33.733675: step: 612/527, loss: 0.15944424271583557
2023-01-21 09:53:34.810754: step: 616/527, loss: 0.0411435142159462
2023-01-21 09:53:35.938674: step: 620/527, loss: 0.4751095771789551
2023-01-21 09:53:37.138414: step: 624/527, loss: 0.017486954107880592
2023-01-21 09:53:38.264832: step: 628/527, loss: 0.10820036381483078
2023-01-21 09:53:39.375313: step: 632/527, loss: 0.07209649682044983
2023-01-21 09:53:40.484545: step: 636/527, loss: 0.3613487482070923
2023-01-21 09:53:41.596365: step: 640/527, loss: 0.052332207560539246
2023-01-21 09:53:42.720406: step: 644/527, loss: 0.042186737060546875
2023-01-21 09:53:43.846908: step: 648/527, loss: 0.09314222633838654
2023-01-21 09:53:44.950945: step: 652/527, loss: 0.05167350918054581
2023-01-21 09:53:46.065522: step: 656/527, loss: 0.18223591148853302
2023-01-21 09:53:47.191801: step: 660/527, loss: 0.049448300153017044
2023-01-21 09:53:48.402556: step: 664/527, loss: 0.25445327162742615
2023-01-21 09:53:49.499854: step: 668/527, loss: 0.050452426075935364
2023-01-21 09:53:50.632357: step: 672/527, loss: 0.048032473772764206
2023-01-21 09:53:51.740334: step: 676/527, loss: 0.005077171139419079
2023-01-21 09:53:52.835606: step: 680/527, loss: 0.07325573265552521
2023-01-21 09:53:53.937605: step: 684/527, loss: 0.02732701227068901
2023-01-21 09:53:55.050193: step: 688/527, loss: 0.04781174659729004
2023-01-21 09:53:56.175982: step: 692/527, loss: 0.012380600906908512
2023-01-21 09:53:57.321800: step: 696/527, loss: 0.45908185839653015
2023-01-21 09:53:58.443736: step: 700/527, loss: 0.06139354780316353
2023-01-21 09:53:59.568771: step: 704/527, loss: 0.030178356915712357
2023-01-21 09:54:00.714504: step: 708/527, loss: 0.04643435403704643
2023-01-21 09:54:01.801120: step: 712/527, loss: 0.052681878209114075
2023-01-21 09:54:02.928493: step: 716/527, loss: 0.27830156683921814
2023-01-21 09:54:04.052104: step: 720/527, loss: 0.3921051025390625
2023-01-21 09:54:05.191595: step: 724/527, loss: 0.0639440268278122
2023-01-21 09:54:06.322630: step: 728/527, loss: 0.10487000644207001
2023-01-21 09:54:07.383408: step: 732/527, loss: 0.007948875427246094
2023-01-21 09:54:08.475183: step: 736/527, loss: 0.05195184051990509
2023-01-21 09:54:09.581959: step: 740/527, loss: 0.09556989371776581
2023-01-21 09:54:10.716774: step: 744/527, loss: 0.0831850990653038
2023-01-21 09:54:11.816642: step: 748/527, loss: 0.9573855400085449
2023-01-21 09:54:12.977116: step: 752/527, loss: 0.03609342873096466
2023-01-21 09:54:14.065898: step: 756/527, loss: 0.16214951872825623
2023-01-21 09:54:15.186281: step: 760/527, loss: 0.07001495361328125
2023-01-21 09:54:16.319672: step: 764/527, loss: 0.035993482917547226
2023-01-21 09:54:17.422355: step: 768/527, loss: 0.08164720982313156
2023-01-21 09:54:18.524838: step: 772/527, loss: 0.08554460853338242
2023-01-21 09:54:19.622713: step: 776/527, loss: 0.034188173711299896
2023-01-21 09:54:20.747040: step: 780/527, loss: 0.17033347487449646
2023-01-21 09:54:21.853312: step: 784/527, loss: 0.6016769409179688
2023-01-21 09:54:22.963990: step: 788/527, loss: 0.059973862022161484
2023-01-21 09:54:24.084993: step: 792/527, loss: 0.038151081651449203
2023-01-21 09:54:25.190038: step: 796/527, loss: 0.03735866770148277
2023-01-21 09:54:26.303556: step: 800/527, loss: 0.022502470761537552
2023-01-21 09:54:27.426875: step: 804/527, loss: 0.06727603077888489
2023-01-21 09:54:28.589434: step: 808/527, loss: 0.12131595611572266
2023-01-21 09:54:29.696568: step: 812/527, loss: 0.028738070279359818
2023-01-21 09:54:30.850325: step: 816/527, loss: 0.13612107932567596
2023-01-21 09:54:31.954843: step: 820/527, loss: 0.06285762786865234
2023-01-21 09:54:33.078619: step: 824/527, loss: 0.015120506286621094
2023-01-21 09:54:34.204993: step: 828/527, loss: 0.01286010816693306
2023-01-21 09:54:35.278603: step: 832/527, loss: 0.03013935126364231
2023-01-21 09:54:36.424025: step: 836/527, loss: 0.045346833765506744
2023-01-21 09:54:37.557640: step: 840/527, loss: 0.03209085762500763
2023-01-21 09:54:38.656784: step: 844/527, loss: 0.07587175071239471
2023-01-21 09:54:39.752833: step: 848/527, loss: 0.060262441635131836
2023-01-21 09:54:40.867774: step: 852/527, loss: 0.03813629224896431
2023-01-21 09:54:41.969019: step: 856/527, loss: 0.054195404052734375
2023-01-21 09:54:43.071205: step: 860/527, loss: 0.06379751861095428
2023-01-21 09:54:44.194399: step: 864/527, loss: 0.17835655808448792
2023-01-21 09:54:45.326275: step: 868/527, loss: 0.13484200835227966
2023-01-21 09:54:46.456530: step: 872/527, loss: 0.09741239994764328
2023-01-21 09:54:47.558208: step: 876/527, loss: 0.20531876385211945
2023-01-21 09:54:48.671284: step: 880/527, loss: 0.04844360426068306
2023-01-21 09:54:49.784987: step: 884/527, loss: 0.13682469725608826
2023-01-21 09:54:50.950814: step: 888/527, loss: 0.021317481994628906
2023-01-21 09:54:52.090530: step: 892/527, loss: 0.08924126625061035
2023-01-21 09:54:53.200401: step: 896/527, loss: 0.08994856476783752
2023-01-21 09:54:54.321715: step: 900/527, loss: 0.09318752586841583
2023-01-21 09:54:55.449731: step: 904/527, loss: 0.09816179424524307
2023-01-21 09:54:56.565168: step: 908/527, loss: 0.06105213239789009
2023-01-21 09:54:57.639613: step: 912/527, loss: 0.12363509833812714
2023-01-21 09:54:58.737907: step: 916/527, loss: 0.3785005807876587
2023-01-21 09:54:59.823895: step: 920/527, loss: 0.029609298333525658
2023-01-21 09:55:00.950328: step: 924/527, loss: 0.038342855870723724
2023-01-21 09:55:02.066794: step: 928/527, loss: 0.09694366157054901
2023-01-21 09:55:03.182563: step: 932/527, loss: 0.021542739123106003
2023-01-21 09:55:04.343953: step: 936/527, loss: 0.08838529884815216
2023-01-21 09:55:05.443805: step: 940/527, loss: 0.15060663223266602
2023-01-21 09:55:06.548008: step: 944/527, loss: 0.0634339302778244
2023-01-21 09:55:07.677971: step: 948/527, loss: 0.01123666763305664
2023-01-21 09:55:08.797681: step: 952/527, loss: 0.0901309996843338
2023-01-21 09:55:09.885993: step: 956/527, loss: 0.2298179566860199
2023-01-21 09:55:11.008713: step: 960/527, loss: 0.025294208899140358
2023-01-21 09:55:12.111599: step: 964/527, loss: 0.08347129821777344
2023-01-21 09:55:13.208570: step: 968/527, loss: 0.2022933065891266
2023-01-21 09:55:14.346326: step: 972/527, loss: 0.07445468753576279
2023-01-21 09:55:15.431109: step: 976/527, loss: 0.19276714324951172
2023-01-21 09:55:16.543819: step: 980/527, loss: 0.6211450695991516
2023-01-21 09:55:17.637449: step: 984/527, loss: 0.10394144058227539
2023-01-21 09:55:18.768326: step: 988/527, loss: 0.08870887756347656
2023-01-21 09:55:19.873605: step: 992/527, loss: 0.05664863809943199
2023-01-21 09:55:20.988762: step: 996/527, loss: 0.03477358818054199
2023-01-21 09:55:22.103805: step: 1000/527, loss: 0.062087155878543854
2023-01-21 09:55:23.225249: step: 1004/527, loss: 0.20270757377147675
2023-01-21 09:55:24.314059: step: 1008/527, loss: 0.04581394046545029
2023-01-21 09:55:25.426355: step: 1012/527, loss: 0.24383649230003357
2023-01-21 09:55:26.517172: step: 1016/527, loss: 0.07283799350261688
2023-01-21 09:55:27.645652: step: 1020/527, loss: 0.05839822441339493
2023-01-21 09:55:28.754034: step: 1024/527, loss: 0.028377056121826172
2023-01-21 09:55:29.908401: step: 1028/527, loss: 0.02766857109963894
2023-01-21 09:55:31.011243: step: 1032/527, loss: 0.4884529411792755
2023-01-21 09:55:32.110472: step: 1036/527, loss: 0.12447874248027802
2023-01-21 09:55:33.203777: step: 1040/527, loss: 0.07007427513599396
2023-01-21 09:55:34.320054: step: 1044/527, loss: 0.06323814392089844
2023-01-21 09:55:35.419271: step: 1048/527, loss: 0.1427680104970932
2023-01-21 09:55:36.545866: step: 1052/527, loss: 0.04865298420190811
2023-01-21 09:55:37.677171: step: 1056/527, loss: 0.14291763305664062
2023-01-21 09:55:38.813822: step: 1060/527, loss: 6.43531608581543
2023-01-21 09:55:39.934396: step: 1064/527, loss: 0.5440343618392944
2023-01-21 09:55:41.073723: step: 1068/527, loss: 0.11009788513183594
2023-01-21 09:55:42.178798: step: 1072/527, loss: 0.07011604309082031
2023-01-21 09:55:43.270158: step: 1076/527, loss: 0.03946361318230629
2023-01-21 09:55:44.354637: step: 1080/527, loss: 0.8340146541595459
2023-01-21 09:55:45.473515: step: 1084/527, loss: 0.4368932843208313
2023-01-21 09:55:46.610735: step: 1088/527, loss: 0.11110344529151917
2023-01-21 09:55:47.701048: step: 1092/527, loss: 0.05119132995605469
2023-01-21 09:55:48.831751: step: 1096/527, loss: 0.2475181519985199
2023-01-21 09:55:49.950413: step: 1100/527, loss: 0.019387532025575638
2023-01-21 09:55:51.052529: step: 1104/527, loss: 0.05043792724609375
2023-01-21 09:55:52.192234: step: 1108/527, loss: 0.1348879039287567
2023-01-21 09:55:53.358893: step: 1112/527, loss: 0.0670267641544342
2023-01-21 09:55:54.476505: step: 1116/527, loss: 0.37143874168395996
2023-01-21 09:55:55.569393: step: 1120/527, loss: 1.0088536739349365
2023-01-21 09:55:56.719694: step: 1124/527, loss: 0.13461600244045258
2023-01-21 09:55:57.836651: step: 1128/527, loss: 0.08546924591064453
2023-01-21 09:55:58.937119: step: 1132/527, loss: 0.05917949602007866
2023-01-21 09:56:00.068674: step: 1136/527, loss: 0.1073429137468338
2023-01-21 09:56:01.202734: step: 1140/527, loss: 0.057706646621227264
2023-01-21 09:56:02.286789: step: 1144/527, loss: 0.07149829715490341
2023-01-21 09:56:03.424552: step: 1148/527, loss: 0.41345348954200745
2023-01-21 09:56:04.585365: step: 1152/527, loss: 0.04643116146326065
2023-01-21 09:56:05.727188: step: 1156/527, loss: 0.0728122740983963
2023-01-21 09:56:06.839762: step: 1160/527, loss: 0.098331980407238
2023-01-21 09:56:07.938805: step: 1164/527, loss: 0.13591843843460083
2023-01-21 09:56:09.057640: step: 1168/527, loss: 0.07631073147058487
2023-01-21 09:56:10.155785: step: 1172/527, loss: 0.015086389146745205
2023-01-21 09:56:11.284362: step: 1176/527, loss: 0.04788418114185333
2023-01-21 09:56:12.383627: step: 1180/527, loss: 0.017382431775331497
2023-01-21 09:56:13.510628: step: 1184/527, loss: 0.08957886695861816
2023-01-21 09:56:14.640983: step: 1188/527, loss: 0.011996269226074219
2023-01-21 09:56:15.774864: step: 1192/527, loss: 0.0798896849155426
2023-01-21 09:56:16.894854: step: 1196/527, loss: 0.03217463567852974
2023-01-21 09:56:18.001307: step: 1200/527, loss: 0.17385807633399963
2023-01-21 09:56:19.117646: step: 1204/527, loss: 0.07956424355506897
2023-01-21 09:56:20.260203: step: 1208/527, loss: 0.12261418998241425
2023-01-21 09:56:21.382292: step: 1212/527, loss: 0.16994795203208923
2023-01-21 09:56:22.509587: step: 1216/527, loss: 0.08262792229652405
2023-01-21 09:56:23.642964: step: 1220/527, loss: 0.1975012719631195
2023-01-21 09:56:24.769201: step: 1224/527, loss: 0.3106329143047333
2023-01-21 09:56:25.879800: step: 1228/527, loss: 0.1893273890018463
2023-01-21 09:56:26.986922: step: 1232/527, loss: 0.045536234974861145
2023-01-21 09:56:28.109241: step: 1236/527, loss: 0.14887762069702148
2023-01-21 09:56:29.178950: step: 1240/527, loss: 0.012140464968979359
2023-01-21 09:56:30.280296: step: 1244/527, loss: 0.08412811160087585
2023-01-21 09:56:31.420807: step: 1248/527, loss: 0.19463062286376953
2023-01-21 09:56:32.533947: step: 1252/527, loss: 0.16118593513965607
2023-01-21 09:56:33.646608: step: 1256/527, loss: 0.0313691608607769
2023-01-21 09:56:34.748552: step: 1260/527, loss: 0.026839066296815872
2023-01-21 09:56:35.860244: step: 1264/527, loss: 0.25529128313064575
2023-01-21 09:56:36.975312: step: 1268/527, loss: 0.007850361056625843
2023-01-21 09:56:38.088508: step: 1272/527, loss: 0.15709514915943146
2023-01-21 09:56:39.231884: step: 1276/527, loss: 0.040537551045417786
2023-01-21 09:56:40.346329: step: 1280/527, loss: 3.074796676635742
2023-01-21 09:56:41.461469: step: 1284/527, loss: 0.008968162350356579
2023-01-21 09:56:42.567751: step: 1288/527, loss: 0.09053463488817215
2023-01-21 09:56:43.701399: step: 1292/527, loss: 0.1288224160671234
2023-01-21 09:56:44.851293: step: 1296/527, loss: 0.07565007358789444
2023-01-21 09:56:45.991903: step: 1300/527, loss: 0.04186534881591797
2023-01-21 09:56:47.108234: step: 1304/527, loss: 0.08322028815746307
2023-01-21 09:56:48.219636: step: 1308/527, loss: 0.01934673823416233
2023-01-21 09:56:49.326969: step: 1312/527, loss: 0.01737365871667862
2023-01-21 09:56:50.434667: step: 1316/527, loss: 0.9277999997138977
2023-01-21 09:56:51.538055: step: 1320/527, loss: 0.2980154752731323
2023-01-21 09:56:52.698277: step: 1324/527, loss: 0.13807877898216248
2023-01-21 09:56:53.808706: step: 1328/527, loss: 1.058421015739441
2023-01-21 09:56:54.900777: step: 1332/527, loss: 0.10170484334230423
2023-01-21 09:56:56.032176: step: 1336/527, loss: 0.04638261720538139
2023-01-21 09:56:57.172200: step: 1340/527, loss: 0.21932242810726166
2023-01-21 09:56:58.294564: step: 1344/527, loss: 0.06194324418902397
2023-01-21 09:56:59.423547: step: 1348/527, loss: 0.23304852843284607
2023-01-21 09:57:00.481654: step: 1352/527, loss: 0.09255866706371307
2023-01-21 09:57:01.612854: step: 1356/527, loss: 0.09989690780639648
2023-01-21 09:57:02.724870: step: 1360/527, loss: 0.2880953848361969
2023-01-21 09:57:03.809923: step: 1364/527, loss: 0.254019558429718
2023-01-21 09:57:04.925407: step: 1368/527, loss: 0.1618337631225586
2023-01-21 09:57:06.042375: step: 1372/527, loss: 0.07139015197753906
2023-01-21 09:57:07.163481: step: 1376/527, loss: 0.08707895874977112
2023-01-21 09:57:08.280164: step: 1380/527, loss: 0.03160696104168892
2023-01-21 09:57:09.385147: step: 1384/527, loss: 0.07711543887853622
2023-01-21 09:57:10.527740: step: 1388/527, loss: 0.06767632067203522
2023-01-21 09:57:11.629414: step: 1392/527, loss: 0.02076435089111328
2023-01-21 09:57:12.744672: step: 1396/527, loss: 0.1154114231467247
2023-01-21 09:57:13.845221: step: 1400/527, loss: 1.3009544610977173
2023-01-21 09:57:14.962496: step: 1404/527, loss: 0.3502483367919922
2023-01-21 09:57:16.066662: step: 1408/527, loss: 0.012606620788574219
2023-01-21 09:57:17.166910: step: 1412/527, loss: 0.15717962384223938
2023-01-21 09:57:18.261148: step: 1416/527, loss: 0.23339462280273438
2023-01-21 09:57:19.362364: step: 1420/527, loss: 0.07521486282348633
2023-01-21 09:57:20.504714: step: 1424/527, loss: 0.115093894302845
2023-01-21 09:57:21.612703: step: 1428/527, loss: 0.008766556158661842
2023-01-21 09:57:22.727259: step: 1432/527, loss: 0.7830890417098999
2023-01-21 09:57:23.857232: step: 1436/527, loss: 0.011635685339570045
2023-01-21 09:57:24.956372: step: 1440/527, loss: 0.03912472724914551
2023-01-21 09:57:26.052232: step: 1444/527, loss: 0.07074785232543945
2023-01-21 09:57:27.144916: step: 1448/527, loss: 0.08525390923023224
2023-01-21 09:57:28.260100: step: 1452/527, loss: 0.005463886074721813
2023-01-21 09:57:29.394147: step: 1456/527, loss: 0.08118104934692383
2023-01-21 09:57:30.528879: step: 1460/527, loss: 0.07672515511512756
2023-01-21 09:57:31.647224: step: 1464/527, loss: 0.015753936022520065
2023-01-21 09:57:32.760531: step: 1468/527, loss: 0.06355476379394531
2023-01-21 09:57:33.902582: step: 1472/527, loss: 0.04105234146118164
2023-01-21 09:57:35.026835: step: 1476/527, loss: 0.052450992166996
2023-01-21 09:57:36.131449: step: 1480/527, loss: 0.11526508629322052
2023-01-21 09:57:37.258960: step: 1484/527, loss: 0.26573240756988525
2023-01-21 09:57:38.403434: step: 1488/527, loss: 0.02306346967816353
2023-01-21 09:57:39.525954: step: 1492/527, loss: 0.04622216522693634
2023-01-21 09:57:40.640921: step: 1496/527, loss: 0.07649651169776917
2023-01-21 09:57:41.760371: step: 1500/527, loss: 0.0812816172838211
2023-01-21 09:57:42.862240: step: 1504/527, loss: 0.06730327755212784
2023-01-21 09:57:43.970336: step: 1508/527, loss: 0.17887631058692932
2023-01-21 09:57:45.078764: step: 1512/527, loss: 0.08821187168359756
2023-01-21 09:57:46.211130: step: 1516/527, loss: 0.1024264395236969
2023-01-21 09:57:47.307072: step: 1520/527, loss: 0.18608751893043518
2023-01-21 09:57:48.401912: step: 1524/527, loss: 0.10426501929759979
2023-01-21 09:57:49.520475: step: 1528/527, loss: 0.08230405300855637
2023-01-21 09:57:50.679322: step: 1532/527, loss: 0.1259574443101883
2023-01-21 09:57:51.792585: step: 1536/527, loss: 0.06192807853221893
2023-01-21 09:57:52.896834: step: 1540/527, loss: 0.3448143005371094
2023-01-21 09:57:53.995042: step: 1544/527, loss: 0.05571703985333443
2023-01-21 09:57:55.105820: step: 1548/527, loss: 0.12170401215553284
2023-01-21 09:57:56.238209: step: 1552/527, loss: 0.25000256299972534
2023-01-21 09:57:57.337380: step: 1556/527, loss: 0.03165683522820473
2023-01-21 09:57:58.441985: step: 1560/527, loss: 0.42499417066574097
2023-01-21 09:57:59.572463: step: 1564/527, loss: 0.0724034309387207
2023-01-21 09:58:00.687294: step: 1568/527, loss: 0.23406529426574707
2023-01-21 09:58:01.802078: step: 1572/527, loss: 0.1629871428012848
2023-01-21 09:58:02.962566: step: 1576/527, loss: 0.39764395356178284
2023-01-21 09:58:04.053485: step: 1580/527, loss: 0.06674136966466904
2023-01-21 09:58:05.148874: step: 1584/527, loss: 0.23009684681892395
2023-01-21 09:58:06.268271: step: 1588/527, loss: 0.31569528579711914
2023-01-21 09:58:07.381452: step: 1592/527, loss: 0.10268383473157883
2023-01-21 09:58:08.497181: step: 1596/527, loss: 0.23501887917518616
2023-01-21 09:58:09.606931: step: 1600/527, loss: 0.0585942268371582
2023-01-21 09:58:10.717172: step: 1604/527, loss: 0.17699089646339417
2023-01-21 09:58:11.814330: step: 1608/527, loss: 0.07348638027906418
2023-01-21 09:58:12.919923: step: 1612/527, loss: 0.8916938900947571
2023-01-21 09:58:14.022615: step: 1616/527, loss: 0.16198918223381042
2023-01-21 09:58:15.124555: step: 1620/527, loss: 0.11379070580005646
2023-01-21 09:58:16.248615: step: 1624/527, loss: 0.0705508291721344
2023-01-21 09:58:17.359809: step: 1628/527, loss: 0.052710454910993576
2023-01-21 09:58:18.517591: step: 1632/527, loss: 0.04200391843914986
2023-01-21 09:58:19.680240: step: 1636/527, loss: 0.1498866081237793
2023-01-21 09:58:20.819563: step: 1640/527, loss: 0.05908498913049698
2023-01-21 09:58:21.948940: step: 1644/527, loss: 0.06964989006519318
2023-01-21 09:58:23.042455: step: 1648/527, loss: 0.025357915088534355
2023-01-21 09:58:24.159842: step: 1652/527, loss: 0.1391225904226303
2023-01-21 09:58:25.255029: step: 1656/527, loss: 0.0820837989449501
2023-01-21 09:58:26.379464: step: 1660/527, loss: 0.23472963273525238
2023-01-21 09:58:27.484395: step: 1664/527, loss: 0.028299950063228607
2023-01-21 09:58:28.596810: step: 1668/527, loss: 0.05475912243127823
2023-01-21 09:58:29.736729: step: 1672/527, loss: 0.18757067620754242
2023-01-21 09:58:30.817343: step: 1676/527, loss: 0.05311885103583336
2023-01-21 09:58:31.932138: step: 1680/527, loss: 0.045778512954711914
2023-01-21 09:58:33.017140: step: 1684/527, loss: 0.06054845079779625
2023-01-21 09:58:34.130383: step: 1688/527, loss: 0.49073266983032227
2023-01-21 09:58:35.261721: step: 1692/527, loss: 0.016210127621889114
2023-01-21 09:58:36.386760: step: 1696/527, loss: 0.1042594462633133
2023-01-21 09:58:37.526040: step: 1700/527, loss: 0.08621196448802948
2023-01-21 09:58:38.640494: step: 1704/527, loss: 0.025224018841981888
2023-01-21 09:58:39.746716: step: 1708/527, loss: 0.06634263694286346
2023-01-21 09:58:40.845658: step: 1712/527, loss: 0.026540659368038177
2023-01-21 09:58:41.932563: step: 1716/527, loss: 0.010688591748476028
2023-01-21 09:58:43.015521: step: 1720/527, loss: 0.09580745548009872
2023-01-21 09:58:44.164431: step: 1724/527, loss: 0.03691267967224121
2023-01-21 09:58:45.292909: step: 1728/527, loss: 0.03740682452917099
2023-01-21 09:58:46.405792: step: 1732/527, loss: 0.011236190795898438
2023-01-21 09:58:47.504215: step: 1736/527, loss: 0.2811966836452484
2023-01-21 09:58:48.606657: step: 1740/527, loss: 0.05300874635577202
2023-01-21 09:58:49.706475: step: 1744/527, loss: 0.09903955459594727
2023-01-21 09:58:50.823871: step: 1748/527, loss: 0.21401633322238922
2023-01-21 09:58:51.960354: step: 1752/527, loss: 0.09192991256713867
2023-01-21 09:58:53.079367: step: 1756/527, loss: 0.2722126245498657
2023-01-21 09:58:54.200521: step: 1760/527, loss: 0.1748226284980774
2023-01-21 09:58:55.318630: step: 1764/527, loss: 0.01491079293191433
2023-01-21 09:58:56.436952: step: 1768/527, loss: 0.2708101272583008
2023-01-21 09:58:57.588216: step: 1772/527, loss: 0.06429481506347656
2023-01-21 09:58:58.677068: step: 1776/527, loss: 0.0377623587846756
2023-01-21 09:58:59.822516: step: 1780/527, loss: 0.6399953961372375
2023-01-21 09:59:00.936568: step: 1784/527, loss: 0.1107591912150383
2023-01-21 09:59:02.035231: step: 1788/527, loss: 0.07851023972034454
2023-01-21 09:59:03.166117: step: 1792/527, loss: 0.04165377840399742
2023-01-21 09:59:04.248898: step: 1796/527, loss: 0.09852257370948792
2023-01-21 09:59:05.380912: step: 1800/527, loss: 0.07205762714147568
2023-01-21 09:59:06.505003: step: 1804/527, loss: 0.08528709411621094
2023-01-21 09:59:07.630405: step: 1808/527, loss: 0.12615692615509033
2023-01-21 09:59:08.755715: step: 1812/527, loss: 0.06534786522388458
2023-01-21 09:59:09.886198: step: 1816/527, loss: 0.03077840805053711
2023-01-21 09:59:10.994606: step: 1820/527, loss: 0.024987507611513138
2023-01-21 09:59:12.123959: step: 1824/527, loss: 0.0980696976184845
2023-01-21 09:59:13.245325: step: 1828/527, loss: 0.05786591023206711
2023-01-21 09:59:14.398612: step: 1832/527, loss: 0.1295120120048523
2023-01-21 09:59:15.495480: step: 1836/527, loss: 0.027859877794981003
2023-01-21 09:59:16.575634: step: 1840/527, loss: 0.17273612320423126
2023-01-21 09:59:17.706366: step: 1844/527, loss: 0.12468204647302628
2023-01-21 09:59:18.826146: step: 1848/527, loss: 0.4620359539985657
2023-01-21 09:59:19.924658: step: 1852/527, loss: 0.03260040283203125
2023-01-21 09:59:21.050662: step: 1856/527, loss: 0.23308992385864258
2023-01-21 09:59:22.201421: step: 1860/527, loss: 0.1158839762210846
2023-01-21 09:59:23.321986: step: 1864/527, loss: 0.7057000398635864
2023-01-21 09:59:24.450276: step: 1868/527, loss: 0.12176495045423508
2023-01-21 09:59:25.541228: step: 1872/527, loss: 0.25947055220603943
2023-01-21 09:59:26.634199: step: 1876/527, loss: 0.012261772528290749
2023-01-21 09:59:27.769167: step: 1880/527, loss: 0.09982604533433914
2023-01-21 09:59:28.868562: step: 1884/527, loss: 0.34849119186401367
2023-01-21 09:59:29.982140: step: 1888/527, loss: 0.11010236293077469
2023-01-21 09:59:31.090056: step: 1892/527, loss: 0.043065883219242096
2023-01-21 09:59:32.178294: step: 1896/527, loss: 0.00511436490342021
2023-01-21 09:59:33.301827: step: 1900/527, loss: 0.2329554557800293
2023-01-21 09:59:34.433333: step: 1904/527, loss: 0.024020005017518997
2023-01-21 09:59:35.573662: step: 1908/527, loss: 0.03548374027013779
2023-01-21 09:59:36.661000: step: 1912/527, loss: 0.1007990837097168
2023-01-21 09:59:37.752099: step: 1916/527, loss: 0.1330731064081192
2023-01-21 09:59:38.860912: step: 1920/527, loss: 0.05790015682578087
2023-01-21 09:59:40.037338: step: 1924/527, loss: 0.031973741948604584
2023-01-21 09:59:41.127827: step: 1928/527, loss: 0.13544607162475586
2023-01-21 09:59:42.216955: step: 1932/527, loss: 0.10375213623046875
2023-01-21 09:59:43.323133: step: 1936/527, loss: 0.00409278878942132
2023-01-21 09:59:44.430023: step: 1940/527, loss: 0.22216644883155823
2023-01-21 09:59:45.539612: step: 1944/527, loss: 0.08315802365541458
2023-01-21 09:59:46.703061: step: 1948/527, loss: 0.16365459561347961
2023-01-21 09:59:47.802957: step: 1952/527, loss: 0.10047474503517151
2023-01-21 09:59:48.911002: step: 1956/527, loss: 0.023608017712831497
2023-01-21 09:59:50.034836: step: 1960/527, loss: 0.14685048162937164
2023-01-21 09:59:51.159103: step: 1964/527, loss: 0.4481789469718933
2023-01-21 09:59:52.261594: step: 1968/527, loss: 0.01799626462161541
2023-01-21 09:59:53.378515: step: 1972/527, loss: 0.13384322822093964
2023-01-21 09:59:54.521277: step: 1976/527, loss: 0.05929103121161461
2023-01-21 09:59:55.628843: step: 1980/527, loss: 0.04733085632324219
2023-01-21 09:59:56.769036: step: 1984/527, loss: 0.1961507797241211
2023-01-21 09:59:57.898576: step: 1988/527, loss: 0.0054912567138671875
2023-01-21 09:59:58.988532: step: 1992/527, loss: 0.010316896252334118
2023-01-21 10:00:00.091316: step: 1996/527, loss: 0.04214811325073242
2023-01-21 10:00:01.226386: step: 2000/527, loss: 0.02843298949301243
2023-01-21 10:00:02.346019: step: 2004/527, loss: 0.1185968816280365
2023-01-21 10:00:03.457603: step: 2008/527, loss: 0.042241476476192474
2023-01-21 10:00:04.555095: step: 2012/527, loss: 0.035033226013183594
2023-01-21 10:00:05.650694: step: 2016/527, loss: 0.10732424259185791
2023-01-21 10:00:06.773319: step: 2020/527, loss: 0.2094259262084961
2023-01-21 10:00:07.891018: step: 2024/527, loss: 0.029109908267855644
2023-01-21 10:00:09.020982: step: 2028/527, loss: 0.11315536499023438
2023-01-21 10:00:10.125373: step: 2032/527, loss: 0.31974098086357117
2023-01-21 10:00:11.247808: step: 2036/527, loss: 0.17729873955249786
2023-01-21 10:00:12.365314: step: 2040/527, loss: 0.1334826797246933
2023-01-21 10:00:13.479765: step: 2044/527, loss: 0.08637714385986328
2023-01-21 10:00:14.574922: step: 2048/527, loss: 0.7767112255096436
2023-01-21 10:00:15.661216: step: 2052/527, loss: 0.07197284698486328
2023-01-21 10:00:16.780267: step: 2056/527, loss: 0.06646418571472168
2023-01-21 10:00:17.884361: step: 2060/527, loss: 0.09405043721199036
2023-01-21 10:00:18.978181: step: 2064/527, loss: 0.10352182388305664
2023-01-21 10:00:20.080536: step: 2068/527, loss: 0.014022446237504482
2023-01-21 10:00:21.220387: step: 2072/527, loss: 0.18049511313438416
2023-01-21 10:00:22.355274: step: 2076/527, loss: 0.1650785505771637
2023-01-21 10:00:23.496285: step: 2080/527, loss: 0.0005835056654177606
2023-01-21 10:00:24.606177: step: 2084/527, loss: 0.1145477294921875
2023-01-21 10:00:25.704254: step: 2088/527, loss: 0.05340452119708061
2023-01-21 10:00:26.778672: step: 2092/527, loss: 0.04760236665606499
2023-01-21 10:00:27.879638: step: 2096/527, loss: 0.016291523352265358
2023-01-21 10:00:29.014429: step: 2100/527, loss: 0.43703246116638184
2023-01-21 10:00:30.157508: step: 2104/527, loss: 0.033960867673158646
2023-01-21 10:00:31.255724: step: 2108/527, loss: 0.09452714771032333
==================================================
Loss: 0.155
--------------------
Dev: {'event': {'p': 0.6291759465478842, 'r': 0.7523302263648469, 'f1': 0.6852637962401456}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8}
Test: {'event': {'p': 0.6327412385981757, 'r': 0.7531428571428571, 'f1': 0.6877119749543438}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8}
Chinese: {'event': {'p': 0.5443037974683544, 'r': 0.7962962962962963, 'f1': 0.6466165413533835}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8}
Korean: {'event': {'p': 0.5681818181818182, 'r': 0.3968253968253968, 'f1': 0.4672897196261682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8}
Russian: {'event': {'p': 0.5277777777777778, 'r': 0.5277777777777778, 'f1': 0.5277777777777778}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:01:12.383027: step: 4/527, loss: 0.051595307886600494
2023-01-21 10:01:13.515649: step: 8/527, loss: 0.21795311570167542
2023-01-21 10:01:14.613273: step: 12/527, loss: 0.014342594891786575
2023-01-21 10:01:15.741084: step: 16/527, loss: 0.007032871246337891
2023-01-21 10:01:16.901260: step: 20/527, loss: 0.056672193109989166
2023-01-21 10:01:18.029957: step: 24/527, loss: 0.09952010959386826
2023-01-21 10:01:19.169265: step: 28/527, loss: 0.0350005142390728
2023-01-21 10:01:20.280275: step: 32/527, loss: 0.04804582521319389
2023-01-21 10:01:21.380838: step: 36/527, loss: 0.0372166633605957
2023-01-21 10:01:22.480276: step: 40/527, loss: 0.14153394103050232
2023-01-21 10:01:23.590270: step: 44/527, loss: 0.0940588042140007
2023-01-21 10:01:24.688072: step: 48/527, loss: 0.07738695293664932
2023-01-21 10:01:25.808271: step: 52/527, loss: 0.019930411130189896
2023-01-21 10:01:26.918278: step: 56/527, loss: 0.07731838524341583
2023-01-21 10:01:28.043913: step: 60/527, loss: 0.026688480749726295
2023-01-21 10:01:29.150448: step: 64/527, loss: 0.2156534194946289
2023-01-21 10:01:30.267686: step: 68/527, loss: 0.07055337727069855
2023-01-21 10:01:31.369478: step: 72/527, loss: 0.03282523527741432
2023-01-21 10:01:32.510007: step: 76/527, loss: 0.07980633527040482
2023-01-21 10:01:33.629670: step: 80/527, loss: 0.05249176174402237
2023-01-21 10:01:34.773217: step: 84/527, loss: 0.08545837551355362
2023-01-21 10:01:35.893215: step: 88/527, loss: 0.06726416945457458
2023-01-21 10:01:37.001369: step: 92/527, loss: 0.02964954450726509
2023-01-21 10:01:38.118099: step: 96/527, loss: 0.586465060710907
2023-01-21 10:01:39.221743: step: 100/527, loss: 0.06918954849243164
2023-01-21 10:01:40.353282: step: 104/527, loss: 0.03303222730755806
2023-01-21 10:01:41.468524: step: 108/527, loss: 0.11050129681825638
2023-01-21 10:01:42.590350: step: 112/527, loss: 0.13752880692481995
2023-01-21 10:01:43.725303: step: 116/527, loss: 0.013901901431381702
2023-01-21 10:01:44.822418: step: 120/527, loss: 0.020580053329467773
2023-01-21 10:01:45.937773: step: 124/527, loss: 0.0653480589389801
2023-01-21 10:01:47.041302: step: 128/527, loss: 0.02235393412411213
2023-01-21 10:01:48.155491: step: 132/527, loss: 0.0476410910487175
2023-01-21 10:01:49.272164: step: 136/527, loss: 0.1430215835571289
2023-01-21 10:01:50.421183: step: 140/527, loss: 0.6804027557373047
2023-01-21 10:01:51.532362: step: 144/527, loss: 0.052741244435310364
2023-01-21 10:01:52.635970: step: 148/527, loss: 0.11618385463953018
2023-01-21 10:01:53.755349: step: 152/527, loss: 0.028086375445127487
2023-01-21 10:01:54.830710: step: 156/527, loss: 0.01134040392935276
2023-01-21 10:01:55.952936: step: 160/527, loss: 0.10427971184253693
2023-01-21 10:01:57.037721: step: 164/527, loss: 0.037943027913570404
2023-01-21 10:01:58.162754: step: 168/527, loss: 0.024988938122987747
2023-01-21 10:01:59.287151: step: 172/527, loss: 0.01216268539428711
2023-01-21 10:02:00.378414: step: 176/527, loss: 0.2157118022441864
2023-01-21 10:02:01.494614: step: 180/527, loss: 0.007859420962631702
2023-01-21 10:02:02.595105: step: 184/527, loss: 0.04901857674121857
2023-01-21 10:02:03.733636: step: 188/527, loss: 0.05857124179601669
2023-01-21 10:02:04.863773: step: 192/527, loss: 0.0880930945277214
2023-01-21 10:02:05.968311: step: 196/527, loss: 0.08765526115894318
2023-01-21 10:02:07.076386: step: 200/527, loss: 0.012512397952377796
2023-01-21 10:02:08.195307: step: 204/527, loss: 0.08281493186950684
2023-01-21 10:02:09.332784: step: 208/527, loss: 0.2987131178379059
2023-01-21 10:02:10.427675: step: 212/527, loss: 0.03581047058105469
2023-01-21 10:02:11.530011: step: 216/527, loss: 0.04344940185546875
2023-01-21 10:02:12.640800: step: 220/527, loss: 0.038758471608161926
2023-01-21 10:02:13.747972: step: 224/527, loss: 0.016405964270234108
2023-01-21 10:02:14.903628: step: 228/527, loss: 0.009473991580307484
2023-01-21 10:02:16.047573: step: 232/527, loss: 0.01953125
2023-01-21 10:02:17.149844: step: 236/527, loss: 0.5504188537597656
2023-01-21 10:02:18.243907: step: 240/527, loss: 0.13292884826660156
2023-01-21 10:02:19.368955: step: 244/527, loss: 0.0989723950624466
2023-01-21 10:02:20.466673: step: 248/527, loss: 0.024614572525024414
2023-01-21 10:02:21.596996: step: 252/527, loss: 0.05763568729162216
2023-01-21 10:02:22.687463: step: 256/527, loss: 0.031978532671928406
2023-01-21 10:02:23.807553: step: 260/527, loss: 0.059507131576538086
2023-01-21 10:02:24.916861: step: 264/527, loss: 0.11678314208984375
2023-01-21 10:02:26.072422: step: 268/527, loss: 0.05213165283203125
2023-01-21 10:02:27.177724: step: 272/527, loss: 0.3329939842224121
2023-01-21 10:02:28.327594: step: 276/527, loss: 0.0064781904220581055
2023-01-21 10:02:29.453551: step: 280/527, loss: 0.1207832396030426
2023-01-21 10:02:30.579401: step: 284/527, loss: 0.15894566476345062
2023-01-21 10:02:31.675920: step: 288/527, loss: 0.018662452697753906
2023-01-21 10:02:32.825693: step: 292/527, loss: 0.1356135457754135
2023-01-21 10:02:33.946098: step: 296/527, loss: 0.021311473101377487
2023-01-21 10:02:35.039573: step: 300/527, loss: 0.039720773696899414
2023-01-21 10:02:36.148591: step: 304/527, loss: 0.03100752830505371
2023-01-21 10:02:37.262427: step: 308/527, loss: 0.36052972078323364
2023-01-21 10:02:38.378873: step: 312/527, loss: 0.14400386810302734
2023-01-21 10:02:39.482203: step: 316/527, loss: 0.18612852692604065
2023-01-21 10:02:40.576721: step: 320/527, loss: 0.006817149929702282
2023-01-21 10:02:41.694441: step: 324/527, loss: 0.03348485380411148
2023-01-21 10:02:42.798488: step: 328/527, loss: 0.03515777736902237
2023-01-21 10:02:43.915703: step: 332/527, loss: 0.1866414099931717
2023-01-21 10:02:45.062187: step: 336/527, loss: 0.1255602389574051
2023-01-21 10:02:46.192265: step: 340/527, loss: 0.06024637073278427
2023-01-21 10:02:47.289030: step: 344/527, loss: 0.11093979328870773
2023-01-21 10:02:48.410420: step: 348/527, loss: 0.07445193827152252
2023-01-21 10:02:49.510513: step: 352/527, loss: 0.12680628895759583
2023-01-21 10:02:50.654473: step: 356/527, loss: 0.012469196692109108
2023-01-21 10:02:51.762545: step: 360/527, loss: 0.04769272729754448
2023-01-21 10:02:52.886260: step: 364/527, loss: 0.028695013374090195
2023-01-21 10:02:54.013933: step: 368/527, loss: 0.02932453155517578
2023-01-21 10:02:55.119534: step: 372/527, loss: 0.08361206203699112
2023-01-21 10:02:56.235575: step: 376/527, loss: 0.015822220593690872
2023-01-21 10:02:57.383548: step: 380/527, loss: 0.9458737373352051
2023-01-21 10:02:58.517580: step: 384/527, loss: 0.07193779945373535
2023-01-21 10:02:59.624611: step: 388/527, loss: 0.041545186191797256
2023-01-21 10:03:00.743171: step: 392/527, loss: 0.07402548938989639
2023-01-21 10:03:01.884321: step: 396/527, loss: 0.2643594741821289
2023-01-21 10:03:02.985883: step: 400/527, loss: 0.05531501770019531
2023-01-21 10:03:04.089014: step: 404/527, loss: 0.140653595328331
2023-01-21 10:03:05.194830: step: 408/527, loss: 0.19587451219558716
2023-01-21 10:03:06.291085: step: 412/527, loss: 0.10895223915576935
2023-01-21 10:03:07.378760: step: 416/527, loss: 0.03143281862139702
2023-01-21 10:03:08.505233: step: 420/527, loss: 0.1968051940202713
2023-01-21 10:03:09.628435: step: 424/527, loss: 0.13352827727794647
2023-01-21 10:03:10.750525: step: 428/527, loss: 0.06161308288574219
2023-01-21 10:03:11.871511: step: 432/527, loss: 0.05079159513115883
2023-01-21 10:03:12.988700: step: 436/527, loss: 0.027446651831269264
2023-01-21 10:03:14.133066: step: 440/527, loss: 0.893958568572998
2023-01-21 10:03:15.248543: step: 444/527, loss: 0.0050172386690974236
2023-01-21 10:03:16.350215: step: 448/527, loss: 0.17577290534973145
2023-01-21 10:03:17.454394: step: 452/527, loss: 0.07095552980899811
2023-01-21 10:03:18.550058: step: 456/527, loss: 0.046774961054325104
2023-01-21 10:03:19.659183: step: 460/527, loss: 0.03888244554400444
2023-01-21 10:03:20.778847: step: 464/527, loss: 0.07091875374317169
2023-01-21 10:03:21.921784: step: 468/527, loss: 0.05677299574017525
2023-01-21 10:03:23.049854: step: 472/527, loss: 0.028757859021425247
2023-01-21 10:03:24.179630: step: 476/527, loss: 0.0855398178100586
2023-01-21 10:03:25.270874: step: 480/527, loss: 0.018016815185546875
2023-01-21 10:03:26.373897: step: 484/527, loss: 0.723493218421936
2023-01-21 10:03:27.502705: step: 488/527, loss: 0.023238373920321465
2023-01-21 10:03:28.635941: step: 492/527, loss: 0.0139433853328228
2023-01-21 10:03:29.757130: step: 496/527, loss: 0.03217935562133789
2023-01-21 10:03:30.873400: step: 500/527, loss: 0.08146496117115021
2023-01-21 10:03:31.972375: step: 504/527, loss: 0.16402435302734375
2023-01-21 10:03:33.080360: step: 508/527, loss: 0.09143076092004776
2023-01-21 10:03:34.177358: step: 512/527, loss: 0.18203610181808472
2023-01-21 10:03:35.291653: step: 516/527, loss: 0.05439407750964165
2023-01-21 10:03:36.422443: step: 520/527, loss: 0.040465641766786575
2023-01-21 10:03:37.554802: step: 524/527, loss: 0.15628820657730103
2023-01-21 10:03:38.659273: step: 528/527, loss: 0.024280261248350143
2023-01-21 10:03:39.790776: step: 532/527, loss: 0.9812090396881104
2023-01-21 10:03:40.885413: step: 536/527, loss: 0.041773274540901184
2023-01-21 10:03:42.021295: step: 540/527, loss: 0.014004421420395374
2023-01-21 10:03:43.097806: step: 544/527, loss: 0.28797128796577454
2023-01-21 10:03:44.205235: step: 548/527, loss: 0.15383762121200562
2023-01-21 10:03:45.351849: step: 552/527, loss: 0.03226127475500107
2023-01-21 10:03:46.472877: step: 556/527, loss: 0.07663097232580185
2023-01-21 10:03:47.590469: step: 560/527, loss: 0.157141774892807
2023-01-21 10:03:48.707355: step: 564/527, loss: 0.030769873410463333
2023-01-21 10:03:49.853730: step: 568/527, loss: 0.021959971636533737
2023-01-21 10:03:50.995815: step: 572/527, loss: 0.3176236152648926
2023-01-21 10:03:52.103106: step: 576/527, loss: 0.09017715603113174
2023-01-21 10:03:53.231466: step: 580/527, loss: 0.11581927537918091
2023-01-21 10:03:54.334594: step: 584/527, loss: 0.025803565979003906
2023-01-21 10:03:55.422907: step: 588/527, loss: 0.11176472157239914
2023-01-21 10:03:56.573254: step: 592/527, loss: 0.17112714052200317
2023-01-21 10:03:57.702512: step: 596/527, loss: 0.04026642069220543
2023-01-21 10:03:58.827709: step: 600/527, loss: 0.04269104078412056
2023-01-21 10:03:59.944635: step: 604/527, loss: 0.001627254532650113
2023-01-21 10:04:01.070246: step: 608/527, loss: 0.06278057396411896
2023-01-21 10:04:02.180403: step: 612/527, loss: 0.12767010927200317
2023-01-21 10:04:03.307617: step: 616/527, loss: 0.16504952311515808
2023-01-21 10:04:04.415539: step: 620/527, loss: 0.8820794224739075
2023-01-21 10:04:05.562594: step: 624/527, loss: 0.06617297977209091
2023-01-21 10:04:06.672618: step: 628/527, loss: 0.10420079529285431
2023-01-21 10:04:07.763873: step: 632/527, loss: 0.1489923745393753
2023-01-21 10:04:08.864202: step: 636/527, loss: 0.035975027829408646
2023-01-21 10:04:09.968581: step: 640/527, loss: 0.006674480624496937
2023-01-21 10:04:11.077528: step: 644/527, loss: 0.05110277980566025
2023-01-21 10:04:12.202533: step: 648/527, loss: 0.10432196408510208
2023-01-21 10:04:13.307208: step: 652/527, loss: 0.12818804383277893
2023-01-21 10:04:14.451685: step: 656/527, loss: 0.08545484393835068
2023-01-21 10:04:15.578741: step: 660/527, loss: 0.1794532835483551
2023-01-21 10:04:16.672262: step: 664/527, loss: 0.02562990039587021
2023-01-21 10:04:17.783509: step: 668/527, loss: 0.11701121181249619
2023-01-21 10:04:18.896972: step: 672/527, loss: 0.0627840980887413
2023-01-21 10:04:20.009735: step: 676/527, loss: 0.028455354273319244
2023-01-21 10:04:21.212636: step: 680/527, loss: 0.08318942785263062
2023-01-21 10:04:22.311041: step: 684/527, loss: 0.058298543095588684
2023-01-21 10:04:23.412128: step: 688/527, loss: 0.11548461765050888
2023-01-21 10:04:24.504301: step: 692/527, loss: 0.05990896373987198
2023-01-21 10:04:25.631023: step: 696/527, loss: 0.04069337993860245
2023-01-21 10:04:26.748410: step: 700/527, loss: 0.0668603926897049
2023-01-21 10:04:27.831824: step: 704/527, loss: 0.0790155902504921
2023-01-21 10:04:28.964654: step: 708/527, loss: 0.0729002058506012
2023-01-21 10:04:30.067654: step: 712/527, loss: 0.04533720016479492
2023-01-21 10:04:31.178711: step: 716/527, loss: 0.00871977861970663
2023-01-21 10:04:32.286141: step: 720/527, loss: 0.24235667288303375
2023-01-21 10:04:33.407683: step: 724/527, loss: 0.16085177659988403
2023-01-21 10:04:34.506967: step: 728/527, loss: 0.06069374084472656
2023-01-21 10:04:35.610280: step: 732/527, loss: 0.8698010444641113
2023-01-21 10:04:36.718903: step: 736/527, loss: 0.08304710686206818
2023-01-21 10:04:37.852001: step: 740/527, loss: 0.04545240476727486
2023-01-21 10:04:38.986634: step: 744/527, loss: 0.17627717554569244
2023-01-21 10:04:40.132809: step: 748/527, loss: 0.10222113132476807
2023-01-21 10:04:41.257662: step: 752/527, loss: 0.04582156985998154
2023-01-21 10:04:42.381631: step: 756/527, loss: 0.12691383063793182
2023-01-21 10:04:43.495891: step: 760/527, loss: 0.015666961669921875
2023-01-21 10:04:44.634555: step: 764/527, loss: 0.023395538330078125
2023-01-21 10:04:45.735992: step: 768/527, loss: 0.12584558129310608
2023-01-21 10:04:46.838689: step: 772/527, loss: 0.2768898010253906
2023-01-21 10:04:47.940606: step: 776/527, loss: 0.061646413058042526
2023-01-21 10:04:49.032492: step: 780/527, loss: 0.14021873474121094
2023-01-21 10:04:50.172111: step: 784/527, loss: 0.022058581933379173
2023-01-21 10:04:51.315435: step: 788/527, loss: 0.09420184791088104
2023-01-21 10:04:52.465916: step: 792/527, loss: 0.006947755813598633
2023-01-21 10:04:53.593080: step: 796/527, loss: 0.035112954676151276
2023-01-21 10:04:54.699413: step: 800/527, loss: 0.05607881397008896
2023-01-21 10:04:55.841077: step: 804/527, loss: 0.13283759355545044
2023-01-21 10:04:56.954385: step: 808/527, loss: 0.2162027209997177
2023-01-21 10:04:58.035952: step: 812/527, loss: 0.010124064050614834
2023-01-21 10:04:59.120279: step: 816/527, loss: 0.04513978958129883
2023-01-21 10:05:00.237304: step: 820/527, loss: 0.08695545047521591
2023-01-21 10:05:01.365159: step: 824/527, loss: 0.028638172894716263
2023-01-21 10:05:02.462561: step: 828/527, loss: 0.07570729404687881
2023-01-21 10:05:03.585267: step: 832/527, loss: 0.031127024441957474
2023-01-21 10:05:04.727290: step: 836/527, loss: 0.147850900888443
2023-01-21 10:05:05.885830: step: 840/527, loss: 0.13352108001708984
2023-01-21 10:05:07.001836: step: 844/527, loss: 0.008698338642716408
2023-01-21 10:05:08.121307: step: 848/527, loss: 0.06953773647546768
2023-01-21 10:05:09.235374: step: 852/527, loss: 0.19496451318264008
2023-01-21 10:05:10.340479: step: 856/527, loss: 0.012446021661162376
2023-01-21 10:05:11.463378: step: 860/527, loss: 0.016244173049926758
2023-01-21 10:05:12.549278: step: 864/527, loss: 0.02343278005719185
2023-01-21 10:05:13.641082: step: 868/527, loss: 0.06877613067626953
2023-01-21 10:05:14.755146: step: 872/527, loss: 0.08043460547924042
2023-01-21 10:05:15.870560: step: 876/527, loss: 0.10054989159107208
2023-01-21 10:05:16.998929: step: 880/527, loss: 0.022923266515135765
2023-01-21 10:05:18.126481: step: 884/527, loss: 0.008685588836669922
2023-01-21 10:05:19.256241: step: 888/527, loss: 0.04530506208539009
2023-01-21 10:05:20.404158: step: 892/527, loss: 0.05135536193847656
2023-01-21 10:05:21.542455: step: 896/527, loss: 0.12792836129665375
2023-01-21 10:05:22.618584: step: 900/527, loss: 0.1459997296333313
2023-01-21 10:05:23.737956: step: 904/527, loss: 0.012559604831039906
2023-01-21 10:05:24.853982: step: 908/527, loss: 0.049593355506658554
2023-01-21 10:05:25.954138: step: 912/527, loss: 0.06437063217163086
2023-01-21 10:05:27.051582: step: 916/527, loss: 0.31694433093070984
2023-01-21 10:05:28.180606: step: 920/527, loss: 0.06975831836462021
2023-01-21 10:05:29.310137: step: 924/527, loss: 0.0027571157552301884
2023-01-21 10:05:30.425689: step: 928/527, loss: 0.02089233323931694
2023-01-21 10:05:31.527099: step: 932/527, loss: 0.12079944461584091
2023-01-21 10:05:32.622925: step: 936/527, loss: 0.025944948196411133
2023-01-21 10:05:33.738261: step: 940/527, loss: 0.060515787452459335
2023-01-21 10:05:34.836320: step: 944/527, loss: 0.06139860302209854
2023-01-21 10:05:35.920030: step: 948/527, loss: 0.0793663039803505
2023-01-21 10:05:37.019263: step: 952/527, loss: 0.03343363106250763
2023-01-21 10:05:38.149746: step: 956/527, loss: 0.058388423174619675
2023-01-21 10:05:39.258350: step: 960/527, loss: 0.003471088595688343
2023-01-21 10:05:40.328482: step: 964/527, loss: 0.0008508682367391884
2023-01-21 10:05:41.448240: step: 968/527, loss: 0.21581801772117615
2023-01-21 10:05:42.564029: step: 972/527, loss: 0.1659446656703949
2023-01-21 10:05:43.666451: step: 976/527, loss: 0.02426748350262642
2023-01-21 10:05:44.762388: step: 980/527, loss: 0.07811923325061798
2023-01-21 10:05:45.864269: step: 984/527, loss: 0.03855371102690697
2023-01-21 10:05:46.963966: step: 988/527, loss: 0.044640637934207916
2023-01-21 10:05:48.083502: step: 992/527, loss: 0.09066858887672424
2023-01-21 10:05:49.192812: step: 996/527, loss: 0.25727522373199463
2023-01-21 10:05:50.328561: step: 1000/527, loss: 0.030248023569583893
2023-01-21 10:05:51.434519: step: 1004/527, loss: 0.1124546080827713
2023-01-21 10:05:52.547884: step: 1008/527, loss: 0.050176240503787994
2023-01-21 10:05:53.675782: step: 1012/527, loss: 0.023157360032200813
2023-01-21 10:05:54.785861: step: 1016/527, loss: 0.051595114171504974
2023-01-21 10:05:55.933794: step: 1020/527, loss: 0.08982644230127335
2023-01-21 10:05:57.081261: step: 1024/527, loss: 0.056859780102968216
2023-01-21 10:05:58.174355: step: 1028/527, loss: 0.0330626480281353
2023-01-21 10:05:59.275809: step: 1032/527, loss: 0.15032851696014404
2023-01-21 10:06:00.410407: step: 1036/527, loss: 0.013005781918764114
2023-01-21 10:06:01.497209: step: 1040/527, loss: 0.010991288349032402
2023-01-21 10:06:02.616095: step: 1044/527, loss: 0.03911247104406357
2023-01-21 10:06:03.744531: step: 1048/527, loss: 0.21045923233032227
2023-01-21 10:06:04.866311: step: 1052/527, loss: 0.06488379091024399
2023-01-21 10:06:05.976534: step: 1056/527, loss: 0.09807959198951721
2023-01-21 10:06:07.100131: step: 1060/527, loss: 0.10139675438404083
2023-01-21 10:06:08.199129: step: 1064/527, loss: 0.05873613432049751
2023-01-21 10:06:09.341584: step: 1068/527, loss: 0.04523877799510956
2023-01-21 10:06:10.433495: step: 1072/527, loss: 0.03674621507525444
2023-01-21 10:06:11.534327: step: 1076/527, loss: 0.08303405344486237
2023-01-21 10:06:12.660053: step: 1080/527, loss: 0.054634857922792435
2023-01-21 10:06:13.773084: step: 1084/527, loss: 0.01169128529727459
2023-01-21 10:06:14.876581: step: 1088/527, loss: 0.0008047103765420616
2023-01-21 10:06:15.957660: step: 1092/527, loss: 0.034918248653411865
2023-01-21 10:06:17.070422: step: 1096/527, loss: 0.016103506088256836
2023-01-21 10:06:18.191777: step: 1100/527, loss: 0.014906119555234909
2023-01-21 10:06:19.326180: step: 1104/527, loss: 0.4689117670059204
2023-01-21 10:06:20.438041: step: 1108/527, loss: 0.06313753128051758
2023-01-21 10:06:21.533649: step: 1112/527, loss: 0.04947681352496147
2023-01-21 10:06:22.665196: step: 1116/527, loss: 0.1910746544599533
2023-01-21 10:06:23.791195: step: 1120/527, loss: 0.05156998708844185
2023-01-21 10:06:24.905189: step: 1124/527, loss: 0.10549669712781906
2023-01-21 10:06:26.022737: step: 1128/527, loss: 0.08306846767663956
2023-01-21 10:06:27.143228: step: 1132/527, loss: 0.2798956036567688
2023-01-21 10:06:28.279278: step: 1136/527, loss: 0.020055418834090233
2023-01-21 10:06:29.393536: step: 1140/527, loss: 0.12009506672620773
2023-01-21 10:06:30.531159: step: 1144/527, loss: 0.07453088462352753
2023-01-21 10:06:31.644663: step: 1148/527, loss: 0.05657706409692764
2023-01-21 10:06:32.751682: step: 1152/527, loss: 0.009659100323915482
2023-01-21 10:06:33.850457: step: 1156/527, loss: 0.07631902396678925
2023-01-21 10:06:34.952866: step: 1160/527, loss: 0.013288307003676891
2023-01-21 10:06:36.052563: step: 1164/527, loss: 0.03264322504401207
2023-01-21 10:06:37.192236: step: 1168/527, loss: 0.10932694375514984
2023-01-21 10:06:38.304567: step: 1172/527, loss: 0.12802904844284058
2023-01-21 10:06:39.459524: step: 1176/527, loss: 0.03888988494873047
2023-01-21 10:06:40.525537: step: 1180/527, loss: 0.09084005653858185
2023-01-21 10:06:41.623674: step: 1184/527, loss: 0.04941602051258087
2023-01-21 10:06:42.738188: step: 1188/527, loss: 0.204171285033226
2023-01-21 10:06:43.863299: step: 1192/527, loss: 0.10325238853693008
2023-01-21 10:06:44.994853: step: 1196/527, loss: 0.13256892561912537
2023-01-21 10:06:46.099386: step: 1200/527, loss: 0.3588072955608368
2023-01-21 10:06:47.184166: step: 1204/527, loss: 0.0320286750793457
2023-01-21 10:06:48.327131: step: 1208/527, loss: 0.07419176399707794
2023-01-21 10:06:49.466113: step: 1212/527, loss: 0.2340150773525238
2023-01-21 10:06:50.567625: step: 1216/527, loss: 0.0997081771492958
2023-01-21 10:06:51.697617: step: 1220/527, loss: 0.03382454067468643
2023-01-21 10:06:52.818781: step: 1224/527, loss: 0.3190089166164398
2023-01-21 10:06:53.920216: step: 1228/527, loss: 0.06604671478271484
2023-01-21 10:06:55.046606: step: 1232/527, loss: 0.021700192242860794
2023-01-21 10:06:56.176921: step: 1236/527, loss: 0.08422355353832245
2023-01-21 10:06:57.288908: step: 1240/527, loss: 0.04974422603845596
2023-01-21 10:06:58.430026: step: 1244/527, loss: 0.050561144948005676
2023-01-21 10:06:59.506902: step: 1248/527, loss: 0.09664221107959747
2023-01-21 10:07:00.629907: step: 1252/527, loss: 0.07937774807214737
2023-01-21 10:07:01.784614: step: 1256/527, loss: 0.06734037399291992
2023-01-21 10:07:02.886561: step: 1260/527, loss: 0.12246689945459366
2023-01-21 10:07:04.080791: step: 1264/527, loss: 0.16401682794094086
2023-01-21 10:07:05.226613: step: 1268/527, loss: 0.04839286953210831
2023-01-21 10:07:06.343326: step: 1272/527, loss: 0.03941002115607262
2023-01-21 10:07:07.453875: step: 1276/527, loss: 0.406686395406723
2023-01-21 10:07:08.561803: step: 1280/527, loss: 0.10335026681423187
2023-01-21 10:07:09.660010: step: 1284/527, loss: 0.027607250958681107
2023-01-21 10:07:10.740523: step: 1288/527, loss: 0.0010218620300292969
2023-01-21 10:07:11.865604: step: 1292/527, loss: 0.18122625350952148
2023-01-21 10:07:13.018924: step: 1296/527, loss: 0.1968582272529602
2023-01-21 10:07:14.150976: step: 1300/527, loss: 0.3424030840396881
2023-01-21 10:07:15.283796: step: 1304/527, loss: 0.07962293922901154
2023-01-21 10:07:16.397871: step: 1308/527, loss: 0.02208681032061577
2023-01-21 10:07:17.521002: step: 1312/527, loss: 0.06310930848121643
2023-01-21 10:07:18.621492: step: 1316/527, loss: 0.011055802926421165
2023-01-21 10:07:19.733690: step: 1320/527, loss: 0.14942240715026855
2023-01-21 10:07:20.816143: step: 1324/527, loss: 0.02859310992062092
2023-01-21 10:07:21.955747: step: 1328/527, loss: 0.20906075835227966
2023-01-21 10:07:23.072756: step: 1332/527, loss: 0.21689942479133606
2023-01-21 10:07:24.144916: step: 1336/527, loss: 0.0136909494176507
2023-01-21 10:07:25.265474: step: 1340/527, loss: 0.03338046371936798
2023-01-21 10:07:26.391111: step: 1344/527, loss: 0.051303669810295105
2023-01-21 10:07:27.538259: step: 1348/527, loss: 0.32002657651901245
2023-01-21 10:07:28.696451: step: 1352/527, loss: 0.22659428417682648
2023-01-21 10:07:29.818995: step: 1356/527, loss: 0.06208410859107971
2023-01-21 10:07:30.947555: step: 1360/527, loss: 0.13741245865821838
2023-01-21 10:07:32.093081: step: 1364/527, loss: 0.2033875584602356
2023-01-21 10:07:33.232451: step: 1368/527, loss: 0.043901439756155014
2023-01-21 10:07:34.350542: step: 1372/527, loss: 0.0405488982796669
2023-01-21 10:07:35.457549: step: 1376/527, loss: 0.027832698076963425
2023-01-21 10:07:36.561362: step: 1380/527, loss: 0.07191906124353409
2023-01-21 10:07:37.651863: step: 1384/527, loss: 0.03992345556616783
2023-01-21 10:07:38.765572: step: 1388/527, loss: 0.046364497393369675
2023-01-21 10:07:39.901278: step: 1392/527, loss: 0.044442370533943176
2023-01-21 10:07:41.023711: step: 1396/527, loss: 0.16097697615623474
2023-01-21 10:07:42.142944: step: 1400/527, loss: 0.053278401494026184
2023-01-21 10:07:43.239241: step: 1404/527, loss: 0.12819309532642365
2023-01-21 10:07:44.367737: step: 1408/527, loss: 0.008398199453949928
2023-01-21 10:07:45.482552: step: 1412/527, loss: 0.05243821069598198
2023-01-21 10:07:46.627122: step: 1416/527, loss: 0.11450280994176865
2023-01-21 10:07:47.705752: step: 1420/527, loss: 0.11117667704820633
2023-01-21 10:07:48.823807: step: 1424/527, loss: 0.022314930334687233
2023-01-21 10:07:49.933878: step: 1428/527, loss: 0.08874158561229706
2023-01-21 10:07:51.048726: step: 1432/527, loss: 0.08282079547643661
2023-01-21 10:07:52.194376: step: 1436/527, loss: 0.030485058203339577
2023-01-21 10:07:53.338100: step: 1440/527, loss: 0.0038703917525708675
2023-01-21 10:07:54.446962: step: 1444/527, loss: 0.012000465765595436
2023-01-21 10:07:55.547879: step: 1448/527, loss: 0.23242883384227753
2023-01-21 10:07:56.666849: step: 1452/527, loss: 0.1509992629289627
2023-01-21 10:07:57.771180: step: 1456/527, loss: 0.6384310722351074
2023-01-21 10:07:58.903919: step: 1460/527, loss: 0.21907207369804382
2023-01-21 10:08:00.027138: step: 1464/527, loss: 0.021789217367768288
2023-01-21 10:08:01.176460: step: 1468/527, loss: 0.03293295204639435
2023-01-21 10:08:02.312298: step: 1472/527, loss: 0.034748006612062454
2023-01-21 10:08:03.476330: step: 1476/527, loss: 0.3407320976257324
2023-01-21 10:08:04.585785: step: 1480/527, loss: 0.00953826867043972
2023-01-21 10:08:05.711596: step: 1484/527, loss: 0.0968172550201416
2023-01-21 10:08:06.852575: step: 1488/527, loss: 0.20362672209739685
2023-01-21 10:08:07.952208: step: 1492/527, loss: 0.0702076181769371
2023-01-21 10:08:09.048168: step: 1496/527, loss: 0.042742349207401276
2023-01-21 10:08:10.157022: step: 1500/527, loss: 0.07776002585887909
2023-01-21 10:08:11.297981: step: 1504/527, loss: 0.006988907232880592
2023-01-21 10:08:12.403910: step: 1508/527, loss: 0.13713455200195312
2023-01-21 10:08:13.539655: step: 1512/527, loss: 0.11749449372291565
2023-01-21 10:08:14.654925: step: 1516/527, loss: 0.04220447689294815
2023-01-21 10:08:15.793333: step: 1520/527, loss: 0.0352315679192543
2023-01-21 10:08:16.906949: step: 1524/527, loss: 0.21461772918701172
2023-01-21 10:08:18.026415: step: 1528/527, loss: 0.3125115633010864
2023-01-21 10:08:19.146897: step: 1532/527, loss: 0.03313593938946724
2023-01-21 10:08:20.254849: step: 1536/527, loss: 0.16639859974384308
2023-01-21 10:08:21.360398: step: 1540/527, loss: 0.0716957077383995
2023-01-21 10:08:22.461996: step: 1544/527, loss: 0.036715127527713776
2023-01-21 10:08:23.598228: step: 1548/527, loss: 0.09695421904325485
2023-01-21 10:08:24.696651: step: 1552/527, loss: 1.7821764945983887
2023-01-21 10:08:25.834777: step: 1556/527, loss: 0.007926560007035732
2023-01-21 10:08:27.002238: step: 1560/527, loss: 0.2573583126068115
2023-01-21 10:08:28.112480: step: 1564/527, loss: 0.09893312305212021
2023-01-21 10:08:29.206911: step: 1568/527, loss: 0.0410795658826828
2023-01-21 10:08:30.329640: step: 1572/527, loss: 0.01193923968821764
2023-01-21 10:08:31.433303: step: 1576/527, loss: 0.04817276448011398
2023-01-21 10:08:32.553938: step: 1580/527, loss: 0.19141235947608948
2023-01-21 10:08:33.670378: step: 1584/527, loss: 0.09637489914894104
2023-01-21 10:08:34.793803: step: 1588/527, loss: 0.08508166670799255
2023-01-21 10:08:35.947237: step: 1592/527, loss: 0.04195718839764595
2023-01-21 10:08:37.073030: step: 1596/527, loss: 0.01997981034219265
2023-01-21 10:08:38.189945: step: 1600/527, loss: 0.6615036725997925
2023-01-21 10:08:39.315052: step: 1604/527, loss: 0.22085018455982208
2023-01-21 10:08:40.451267: step: 1608/527, loss: 0.11679039150476456
2023-01-21 10:08:41.593746: step: 1612/527, loss: 0.32708579301834106
2023-01-21 10:08:42.751279: step: 1616/527, loss: 0.04775838926434517
2023-01-21 10:08:43.879088: step: 1620/527, loss: 0.11859016120433807
2023-01-21 10:08:44.983697: step: 1624/527, loss: 0.053553201258182526
2023-01-21 10:08:46.110220: step: 1628/527, loss: 0.085336834192276
2023-01-21 10:08:47.236482: step: 1632/527, loss: 0.06629399955272675
2023-01-21 10:08:48.371101: step: 1636/527, loss: 0.0947992354631424
2023-01-21 10:08:49.497273: step: 1640/527, loss: 0.0417722724378109
2023-01-21 10:08:50.621376: step: 1644/527, loss: 0.15947164595127106
2023-01-21 10:08:51.744758: step: 1648/527, loss: 0.12869195640087128
2023-01-21 10:08:52.853914: step: 1652/527, loss: 0.10346432030200958
2023-01-21 10:08:53.985085: step: 1656/527, loss: 0.05017418786883354
2023-01-21 10:08:55.103029: step: 1660/527, loss: 0.030830956995487213
2023-01-21 10:08:56.241508: step: 1664/527, loss: 0.025107383728027344
2023-01-21 10:08:57.365178: step: 1668/527, loss: 0.05479329824447632
2023-01-21 10:08:58.484771: step: 1672/527, loss: 0.025297069922089577
2023-01-21 10:08:59.588977: step: 1676/527, loss: 0.02036609686911106
2023-01-21 10:09:00.728127: step: 1680/527, loss: 0.1163950115442276
2023-01-21 10:09:01.822604: step: 1684/527, loss: 0.10086068511009216
2023-01-21 10:09:02.922888: step: 1688/527, loss: 0.01263041514903307
2023-01-21 10:09:04.028826: step: 1692/527, loss: 0.07123565673828125
2023-01-21 10:09:05.127984: step: 1696/527, loss: 0.24792207777500153
2023-01-21 10:09:06.257794: step: 1700/527, loss: 0.018088627606630325
2023-01-21 10:09:07.373258: step: 1704/527, loss: 0.0056847576051950455
2023-01-21 10:09:08.489986: step: 1708/527, loss: 0.2932374179363251
2023-01-21 10:09:09.633297: step: 1712/527, loss: 0.02307887189090252
2023-01-21 10:09:10.756974: step: 1716/527, loss: 0.022414302453398705
2023-01-21 10:09:11.909094: step: 1720/527, loss: 0.11815033853054047
2023-01-21 10:09:13.037752: step: 1724/527, loss: 0.2428234964609146
2023-01-21 10:09:14.151716: step: 1728/527, loss: 0.016019631177186966
2023-01-21 10:09:15.299287: step: 1732/527, loss: 0.04156522825360298
2023-01-21 10:09:16.441416: step: 1736/527, loss: 0.13247385621070862
2023-01-21 10:09:17.553418: step: 1740/527, loss: 0.03684301674365997
2023-01-21 10:09:18.653589: step: 1744/527, loss: 0.06472799181938171
2023-01-21 10:09:19.768455: step: 1748/527, loss: 0.02149028703570366
2023-01-21 10:09:20.937337: step: 1752/527, loss: 0.18686771392822266
2023-01-21 10:09:22.056663: step: 1756/527, loss: 0.04778309166431427
2023-01-21 10:09:23.179030: step: 1760/527, loss: 0.12145087867975235
2023-01-21 10:09:24.304108: step: 1764/527, loss: 0.1457628309726715
2023-01-21 10:09:25.425084: step: 1768/527, loss: 0.12799444794654846
2023-01-21 10:09:26.588353: step: 1772/527, loss: 0.0712505429983139
2023-01-21 10:09:27.703207: step: 1776/527, loss: 0.04652068391442299
2023-01-21 10:09:28.815734: step: 1780/527, loss: 0.05784625932574272
2023-01-21 10:09:29.898210: step: 1784/527, loss: 0.07610654830932617
2023-01-21 10:09:31.018120: step: 1788/527, loss: 0.0842500701546669
2023-01-21 10:09:32.125224: step: 1792/527, loss: 0.14114198088645935
2023-01-21 10:09:33.229490: step: 1796/527, loss: 0.012464619241654873
2023-01-21 10:09:34.371552: step: 1800/527, loss: 0.060916490852832794
2023-01-21 10:09:35.508179: step: 1804/527, loss: 0.04095430299639702
2023-01-21 10:09:36.621723: step: 1808/527, loss: 0.020702935755252838
2023-01-21 10:09:37.742691: step: 1812/527, loss: 0.07763157039880753
2023-01-21 10:09:38.869105: step: 1816/527, loss: 0.053613949567079544
2023-01-21 10:09:40.002300: step: 1820/527, loss: 0.14593087136745453
2023-01-21 10:09:41.143310: step: 1824/527, loss: 0.1952143758535385
2023-01-21 10:09:42.280418: step: 1828/527, loss: 0.024495698511600494
2023-01-21 10:09:43.386349: step: 1832/527, loss: 0.44712674617767334
2023-01-21 10:09:44.510240: step: 1836/527, loss: 0.20110397040843964
2023-01-21 10:09:45.665007: step: 1840/527, loss: 0.6827542185783386
2023-01-21 10:09:46.825635: step: 1844/527, loss: 0.06904802471399307
2023-01-21 10:09:47.949812: step: 1848/527, loss: 0.1400633305311203
2023-01-21 10:09:49.065128: step: 1852/527, loss: 0.031157206743955612
2023-01-21 10:09:50.185208: step: 1856/527, loss: 0.07655920833349228
2023-01-21 10:09:51.297546: step: 1860/527, loss: 0.17182445526123047
2023-01-21 10:09:52.397095: step: 1864/527, loss: 0.12290334701538086
2023-01-21 10:09:53.526720: step: 1868/527, loss: 0.02534008026123047
2023-01-21 10:09:54.655161: step: 1872/527, loss: 0.1496855765581131
2023-01-21 10:09:55.762042: step: 1876/527, loss: 0.07068276405334473
2023-01-21 10:09:56.892761: step: 1880/527, loss: 0.10964031517505646
2023-01-21 10:09:58.029633: step: 1884/527, loss: 0.07338638603687286
2023-01-21 10:09:59.145773: step: 1888/527, loss: 0.051754143089056015
2023-01-21 10:10:00.298052: step: 1892/527, loss: 0.029028750956058502
2023-01-21 10:10:01.412678: step: 1896/527, loss: 0.03004312515258789
2023-01-21 10:10:02.528310: step: 1900/527, loss: 0.9311063885688782
2023-01-21 10:10:03.646363: step: 1904/527, loss: 0.034515380859375
2023-01-21 10:10:04.769865: step: 1908/527, loss: 0.06688328832387924
2023-01-21 10:10:05.876359: step: 1912/527, loss: 0.08460120856761932
2023-01-21 10:10:06.987043: step: 1916/527, loss: 0.21299296617507935
2023-01-21 10:10:08.079506: step: 1920/527, loss: 3.4084885120391846
2023-01-21 10:10:09.189800: step: 1924/527, loss: 0.06421475857496262
2023-01-21 10:10:10.323504: step: 1928/527, loss: 0.06366176903247833
2023-01-21 10:10:11.411998: step: 1932/527, loss: 0.10714731365442276
2023-01-21 10:10:12.545266: step: 1936/527, loss: 0.0551820769906044
2023-01-21 10:10:13.645313: step: 1940/527, loss: 0.001966929528862238
2023-01-21 10:10:14.733428: step: 1944/527, loss: 0.02919469028711319
2023-01-21 10:10:15.863023: step: 1948/527, loss: 0.04343147575855255
2023-01-21 10:10:16.984914: step: 1952/527, loss: 0.00323486328125
2023-01-21 10:10:18.086148: step: 1956/527, loss: 0.0796491950750351
2023-01-21 10:10:19.199916: step: 1960/527, loss: 0.24456629157066345
2023-01-21 10:10:20.305854: step: 1964/527, loss: 0.06715717166662216
2023-01-21 10:10:21.413773: step: 1968/527, loss: 0.2161039412021637
2023-01-21 10:10:22.519709: step: 1972/527, loss: 0.058657363057136536
2023-01-21 10:10:23.645750: step: 1976/527, loss: 0.04664459079504013
2023-01-21 10:10:24.751029: step: 1980/527, loss: 0.15254350006580353
2023-01-21 10:10:25.877572: step: 1984/527, loss: 0.08862748742103577
2023-01-21 10:10:26.984602: step: 1988/527, loss: 0.12920647859573364
2023-01-21 10:10:28.075240: step: 1992/527, loss: 0.06137437745928764
2023-01-21 10:10:29.182063: step: 1996/527, loss: 0.02515232004225254
2023-01-21 10:10:30.318370: step: 2000/527, loss: 0.03706379234790802
2023-01-21 10:10:31.420366: step: 2004/527, loss: 0.1036413162946701
2023-01-21 10:10:32.541890: step: 2008/527, loss: 0.027640534564852715
2023-01-21 10:10:33.651521: step: 2012/527, loss: 0.05735301971435547
2023-01-21 10:10:34.747119: step: 2016/527, loss: 0.24364295601844788
2023-01-21 10:10:35.854353: step: 2020/527, loss: 0.011335229501128197
2023-01-21 10:10:36.939161: step: 2024/527, loss: 0.010156822390854359
2023-01-21 10:10:38.054349: step: 2028/527, loss: 0.1353379189968109
2023-01-21 10:10:39.180613: step: 2032/527, loss: 0.8149141073226929
2023-01-21 10:10:40.331017: step: 2036/527, loss: 0.05819511413574219
2023-01-21 10:10:41.466820: step: 2040/527, loss: 0.0748991072177887
2023-01-21 10:10:42.577939: step: 2044/527, loss: 0.030807018280029297
2023-01-21 10:10:43.696863: step: 2048/527, loss: 0.1352270096540451
2023-01-21 10:10:44.835865: step: 2052/527, loss: 0.06663999706506729
2023-01-21 10:10:45.955814: step: 2056/527, loss: 0.08670587837696075
2023-01-21 10:10:47.089931: step: 2060/527, loss: 0.48546478152275085
2023-01-21 10:10:48.190073: step: 2064/527, loss: 0.007636452093720436
2023-01-21 10:10:49.326996: step: 2068/527, loss: 0.08466167747974396
2023-01-21 10:10:50.433112: step: 2072/527, loss: 0.06291475147008896
2023-01-21 10:10:51.570945: step: 2076/527, loss: 0.23888997733592987
2023-01-21 10:10:52.683370: step: 2080/527, loss: 0.303070068359375
2023-01-21 10:10:53.793311: step: 2084/527, loss: 0.038504458963871
2023-01-21 10:10:54.917695: step: 2088/527, loss: 0.15526457130908966
2023-01-21 10:10:56.027027: step: 2092/527, loss: 0.03289389610290527
2023-01-21 10:10:57.137909: step: 2096/527, loss: 0.0832575187087059
2023-01-21 10:10:58.228642: step: 2100/527, loss: 0.05071668699383736
2023-01-21 10:10:59.324831: step: 2104/527, loss: 0.07595052570104599
2023-01-21 10:11:00.441220: step: 2108/527, loss: 0.08094806969165802
==================================================
Loss: 0.116
--------------------
Dev: {'event': {'p': 0.590818363273453, 'r': 0.7882822902796272, 'f1': 0.6754135767256132}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9}
Test: {'event': {'p': 0.6228622862286228, 'r': 0.7908571428571428, 'f1': 0.6968781470292045}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9}
Chinese: {'event': {'p': 0.5609756097560976, 'r': 0.8518518518518519, 'f1': 0.6764705882352942}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9}
Korean: {'event': {'p': 0.5423728813559322, 'r': 0.5079365079365079, 'f1': 0.5245901639344263}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9}
Russian: {'event': {'p': 0.40476190476190477, 'r': 0.4722222222222222, 'f1': 0.4358974358974359}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:11:42.550781: step: 4/527, loss: 0.003954839892685413
2023-01-21 10:11:43.666549: step: 8/527, loss: 0.046971939504146576
2023-01-21 10:11:44.794589: step: 12/527, loss: 0.04838142544031143
2023-01-21 10:11:45.931974: step: 16/527, loss: 0.07612638175487518
2023-01-21 10:11:47.051981: step: 20/527, loss: 0.0632898360490799
2023-01-21 10:11:48.155374: step: 24/527, loss: 0.03009018860757351
2023-01-21 10:11:49.276280: step: 28/527, loss: 0.011864805594086647
2023-01-21 10:11:50.440760: step: 32/527, loss: 0.026328755542635918
2023-01-21 10:11:51.524012: step: 36/527, loss: 0.017726421356201172
2023-01-21 10:11:52.643176: step: 40/527, loss: 0.017998887225985527
2023-01-21 10:11:53.761053: step: 44/527, loss: 0.024239160120487213
2023-01-21 10:11:54.924783: step: 48/527, loss: 0.2858274281024933
2023-01-21 10:11:56.057927: step: 52/527, loss: 0.13892985880374908
2023-01-21 10:11:57.165119: step: 56/527, loss: 0.03566465526819229
2023-01-21 10:11:58.282907: step: 60/527, loss: 0.04515719413757324
2023-01-21 10:11:59.383694: step: 64/527, loss: 0.09078221768140793
2023-01-21 10:12:00.494954: step: 68/527, loss: 0.008511638268828392
2023-01-21 10:12:01.602768: step: 72/527, loss: 0.04063911736011505
2023-01-21 10:12:02.732965: step: 76/527, loss: 0.27636852860450745
2023-01-21 10:12:03.855313: step: 80/527, loss: 0.013401126489043236
2023-01-21 10:12:04.960663: step: 84/527, loss: 0.03900165483355522
2023-01-21 10:12:06.051404: step: 88/527, loss: 0.059256602078676224
2023-01-21 10:12:07.161263: step: 92/527, loss: 0.003378374967724085
2023-01-21 10:12:08.259054: step: 96/527, loss: 0.0051441197283566
2023-01-21 10:12:09.362309: step: 100/527, loss: 0.03346522897481918
2023-01-21 10:12:10.462163: step: 104/527, loss: 0.0983375608921051
2023-01-21 10:12:11.566682: step: 108/527, loss: 0.010539722628891468
2023-01-21 10:12:12.677042: step: 112/527, loss: 0.008619308471679688
2023-01-21 10:12:13.793372: step: 116/527, loss: 0.03800782933831215
2023-01-21 10:12:14.923873: step: 120/527, loss: 0.01792139932513237
2023-01-21 10:12:16.041426: step: 124/527, loss: 0.34092769026756287
2023-01-21 10:12:17.179928: step: 128/527, loss: 0.04823474958539009
2023-01-21 10:12:18.283839: step: 132/527, loss: 0.035067055374383926
2023-01-21 10:12:19.396280: step: 136/527, loss: 0.07441334426403046
2023-01-21 10:12:20.529723: step: 140/527, loss: 0.45805856585502625
2023-01-21 10:12:21.627109: step: 144/527, loss: 0.14782772958278656
2023-01-21 10:12:22.726875: step: 148/527, loss: 0.003230953123420477
2023-01-21 10:12:23.842896: step: 152/527, loss: 0.1826217621564865
2023-01-21 10:12:24.938644: step: 156/527, loss: 0.027864886447787285
2023-01-21 10:12:26.037822: step: 160/527, loss: 0.04507408291101456
2023-01-21 10:12:27.159612: step: 164/527, loss: 0.08217363059520721
2023-01-21 10:12:28.269230: step: 168/527, loss: 0.05337662994861603
2023-01-21 10:12:29.377214: step: 172/527, loss: 0.01218261756002903
2023-01-21 10:12:30.495135: step: 176/527, loss: 0.02252178266644478
2023-01-21 10:12:31.588105: step: 180/527, loss: 0.031838610768318176
2023-01-21 10:12:32.730105: step: 184/527, loss: 0.04049744829535484
2023-01-21 10:12:33.848591: step: 188/527, loss: 0.04686784744262695
2023-01-21 10:12:34.959571: step: 192/527, loss: 0.03474388271570206
2023-01-21 10:12:36.060427: step: 196/527, loss: 0.07600340247154236
2023-01-21 10:12:37.180917: step: 200/527, loss: 0.03720531612634659
2023-01-21 10:12:38.296617: step: 204/527, loss: 0.02830514870584011
2023-01-21 10:12:39.404669: step: 208/527, loss: 0.06913509964942932
2023-01-21 10:12:40.506838: step: 212/527, loss: 0.04269905388355255
2023-01-21 10:12:41.631629: step: 216/527, loss: 0.0749143585562706
2023-01-21 10:12:42.767694: step: 220/527, loss: 0.024440767243504524
2023-01-21 10:12:43.906364: step: 224/527, loss: 0.10924110561609268
2023-01-21 10:12:44.991347: step: 228/527, loss: 0.0377313606441021
2023-01-21 10:12:46.123556: step: 232/527, loss: 0.05104675143957138
2023-01-21 10:12:47.239270: step: 236/527, loss: 0.04650368541479111
2023-01-21 10:12:48.371865: step: 240/527, loss: 0.03903999179601669
2023-01-21 10:12:49.508803: step: 244/527, loss: 0.1586379110813141
2023-01-21 10:12:50.633799: step: 248/527, loss: 0.024387264624238014
2023-01-21 10:12:51.798110: step: 252/527, loss: 0.034582994878292084
2023-01-21 10:12:52.927922: step: 256/527, loss: 0.11938118934631348
2023-01-21 10:12:54.046376: step: 260/527, loss: 0.05850648880004883
2023-01-21 10:12:55.155247: step: 264/527, loss: 0.10497408360242844
2023-01-21 10:12:56.248369: step: 268/527, loss: 0.08671379089355469
2023-01-21 10:12:57.354554: step: 272/527, loss: 0.14994125068187714
2023-01-21 10:12:58.440452: step: 276/527, loss: 0.04751171916723251
2023-01-21 10:12:59.556984: step: 280/527, loss: 0.004595470614731312
2023-01-21 10:13:00.652660: step: 284/527, loss: 0.03405923768877983
2023-01-21 10:13:01.765306: step: 288/527, loss: 0.03524913638830185
2023-01-21 10:13:02.871529: step: 292/527, loss: 0.040092941373586655
2023-01-21 10:13:04.012247: step: 296/527, loss: 0.16224071383476257
2023-01-21 10:13:05.117436: step: 300/527, loss: 0.05617475509643555
2023-01-21 10:13:06.228401: step: 304/527, loss: 0.02535400539636612
2023-01-21 10:13:07.338225: step: 308/527, loss: 0.12030944228172302
2023-01-21 10:13:08.456275: step: 312/527, loss: 0.09914655238389969
2023-01-21 10:13:09.564018: step: 316/527, loss: 0.03183479607105255
2023-01-21 10:13:10.684953: step: 320/527, loss: 0.3562818467617035
2023-01-21 10:13:11.772028: step: 324/527, loss: 0.012703514657914639
2023-01-21 10:13:12.897625: step: 328/527, loss: 0.18833065032958984
2023-01-21 10:13:14.024513: step: 332/527, loss: 0.08275914192199707
2023-01-21 10:13:15.143642: step: 336/527, loss: 0.0039010047912597656
2023-01-21 10:13:16.260948: step: 340/527, loss: 0.00875091552734375
2023-01-21 10:13:17.357363: step: 344/527, loss: 0.36065831780433655
2023-01-21 10:13:18.482905: step: 348/527, loss: 0.044191304594278336
2023-01-21 10:13:19.581709: step: 352/527, loss: 0.016362953931093216
2023-01-21 10:13:20.728984: step: 356/527, loss: 0.13473300635814667
2023-01-21 10:13:21.838299: step: 360/527, loss: 0.21554070711135864
2023-01-21 10:13:22.931131: step: 364/527, loss: 0.013247680850327015
2023-01-21 10:13:24.041582: step: 368/527, loss: 0.04358730465173721
2023-01-21 10:13:25.151707: step: 372/527, loss: 0.11040067672729492
2023-01-21 10:13:26.278680: step: 376/527, loss: 0.28858065605163574
2023-01-21 10:13:27.389137: step: 380/527, loss: 0.03753490373492241
2023-01-21 10:13:28.495668: step: 384/527, loss: 0.045523930341005325
2023-01-21 10:13:29.614985: step: 388/527, loss: 0.06086587905883789
2023-01-21 10:13:30.722478: step: 392/527, loss: 0.01715068705379963
2023-01-21 10:13:31.850251: step: 396/527, loss: 0.090460404753685
2023-01-21 10:13:32.932184: step: 400/527, loss: 0.03120708465576172
2023-01-21 10:13:34.041103: step: 404/527, loss: 0.06381330639123917
2023-01-21 10:13:35.151649: step: 408/527, loss: 0.08668927848339081
2023-01-21 10:13:36.241351: step: 412/527, loss: 0.0744566023349762
2023-01-21 10:13:37.362892: step: 416/527, loss: 0.10327544808387756
2023-01-21 10:13:38.465134: step: 420/527, loss: 0.012917709536850452
2023-01-21 10:13:39.575375: step: 424/527, loss: 0.1444985270500183
2023-01-21 10:13:40.705283: step: 428/527, loss: 0.009100962430238724
2023-01-21 10:13:41.810864: step: 432/527, loss: 0.023868083953857422
2023-01-21 10:13:42.902977: step: 436/527, loss: 0.004426383879035711
2023-01-21 10:13:43.985368: step: 440/527, loss: 0.05390510708093643
2023-01-21 10:13:45.118607: step: 444/527, loss: 0.06055927276611328
2023-01-21 10:13:46.219812: step: 448/527, loss: 0.039765167981386185
2023-01-21 10:13:47.364638: step: 452/527, loss: 0.05110931769013405
2023-01-21 10:13:48.487633: step: 456/527, loss: 0.029856158420443535
2023-01-21 10:13:49.614794: step: 460/527, loss: 0.059945348650217056
2023-01-21 10:13:50.728742: step: 464/527, loss: 0.10481533408164978
2023-01-21 10:13:51.853547: step: 468/527, loss: 0.003098726272583008
2023-01-21 10:13:52.967692: step: 472/527, loss: 0.21219158172607422
2023-01-21 10:13:54.108057: step: 476/527, loss: 0.027169417589902878
2023-01-21 10:13:55.202966: step: 480/527, loss: 0.12153196334838867
2023-01-21 10:13:56.304341: step: 484/527, loss: 0.0643930435180664
2023-01-21 10:13:57.424134: step: 488/527, loss: 0.018788719549775124
2023-01-21 10:13:58.554944: step: 492/527, loss: 0.12497644126415253
2023-01-21 10:13:59.649004: step: 496/527, loss: 0.05850248411297798
2023-01-21 10:14:00.760024: step: 500/527, loss: 0.142948716878891
2023-01-21 10:14:01.870435: step: 504/527, loss: 0.06825704872608185
2023-01-21 10:14:02.976914: step: 508/527, loss: 0.05962572246789932
2023-01-21 10:14:04.123487: step: 512/527, loss: 0.17915363609790802
2023-01-21 10:14:05.211905: step: 516/527, loss: 0.0045237066224217415
2023-01-21 10:14:06.333410: step: 520/527, loss: 0.043069079518318176
2023-01-21 10:14:07.453701: step: 524/527, loss: 0.07603655010461807
2023-01-21 10:14:08.579859: step: 528/527, loss: 0.06310272216796875
2023-01-21 10:14:09.695058: step: 532/527, loss: 0.028141213580965996
2023-01-21 10:14:10.820389: step: 536/527, loss: 0.04646243900060654
2023-01-21 10:14:11.958588: step: 540/527, loss: 0.03706321865320206
2023-01-21 10:14:13.065107: step: 544/527, loss: 0.000804901123046875
2023-01-21 10:14:14.190727: step: 548/527, loss: 0.05564718320965767
2023-01-21 10:14:15.297460: step: 552/527, loss: 0.013531303033232689
2023-01-21 10:14:16.419502: step: 556/527, loss: 0.022708892822265625
2023-01-21 10:14:17.517627: step: 560/527, loss: 0.012738799676299095
2023-01-21 10:14:18.635175: step: 564/527, loss: 0.12732820212841034
2023-01-21 10:14:19.774995: step: 568/527, loss: 0.03637869656085968
2023-01-21 10:14:20.922873: step: 572/527, loss: 0.010107708163559437
2023-01-21 10:14:22.040105: step: 576/527, loss: 0.11684437096118927
2023-01-21 10:14:23.173208: step: 580/527, loss: 0.09836645424365997
2023-01-21 10:14:24.291634: step: 584/527, loss: 0.6022077798843384
2023-01-21 10:14:25.420059: step: 588/527, loss: 0.013932609930634499
2023-01-21 10:14:26.518225: step: 592/527, loss: 0.04637398570775986
2023-01-21 10:14:27.633301: step: 596/527, loss: 0.05373978987336159
2023-01-21 10:14:28.730845: step: 600/527, loss: 0.07080250233411789
2023-01-21 10:14:29.836223: step: 604/527, loss: 0.0334104523062706
2023-01-21 10:14:30.952331: step: 608/527, loss: 0.029811764135956764
2023-01-21 10:14:32.105656: step: 612/527, loss: 0.03693685680627823
2023-01-21 10:14:33.198429: step: 616/527, loss: 0.01927213743329048
2023-01-21 10:14:34.356211: step: 620/527, loss: 0.040345385670661926
2023-01-21 10:14:35.517642: step: 624/527, loss: 0.026490308344364166
2023-01-21 10:14:36.634392: step: 628/527, loss: 0.04458937793970108
2023-01-21 10:14:37.721282: step: 632/527, loss: 0.08664774894714355
2023-01-21 10:14:38.813182: step: 636/527, loss: 0.05152931436896324
2023-01-21 10:14:39.941076: step: 640/527, loss: 0.06606302410364151
2023-01-21 10:14:41.049114: step: 644/527, loss: 0.08554963767528534
2023-01-21 10:14:42.157359: step: 648/527, loss: 0.06100320816040039
2023-01-21 10:14:43.268418: step: 652/527, loss: 0.08864001929759979
2023-01-21 10:14:44.388226: step: 656/527, loss: 0.00786581076681614
2023-01-21 10:14:45.496485: step: 660/527, loss: 0.026302147656679153
2023-01-21 10:14:46.583583: step: 664/527, loss: 0.028510475531220436
2023-01-21 10:14:47.690290: step: 668/527, loss: 0.020524216815829277
2023-01-21 10:14:48.837855: step: 672/527, loss: 0.004625129513442516
2023-01-21 10:14:49.946190: step: 676/527, loss: 0.06390111893415451
2023-01-21 10:14:51.081736: step: 680/527, loss: 0.02110319212079048
2023-01-21 10:14:52.207025: step: 684/527, loss: 0.12578515708446503
2023-01-21 10:14:53.328997: step: 688/527, loss: 0.07842636108398438
2023-01-21 10:14:54.488603: step: 692/527, loss: 0.056389905512332916
2023-01-21 10:14:55.589300: step: 696/527, loss: 0.09628009796142578
2023-01-21 10:14:56.675942: step: 700/527, loss: 0.03165140002965927
2023-01-21 10:14:57.796069: step: 704/527, loss: 0.013393402099609375
2023-01-21 10:14:58.915598: step: 708/527, loss: 0.08585198223590851
2023-01-21 10:15:00.026133: step: 712/527, loss: 0.02230224572122097
2023-01-21 10:15:01.152595: step: 716/527, loss: 0.025845587253570557
2023-01-21 10:15:02.252181: step: 720/527, loss: 0.048845671117305756
2023-01-21 10:15:03.385269: step: 724/527, loss: 0.1077117919921875
2023-01-21 10:15:04.525381: step: 728/527, loss: 0.032929327338933945
2023-01-21 10:15:05.663046: step: 732/527, loss: 0.009232711978256702
2023-01-21 10:15:06.780077: step: 736/527, loss: 0.056148529052734375
2023-01-21 10:15:07.901304: step: 740/527, loss: 0.07523298263549805
2023-01-21 10:15:09.006472: step: 744/527, loss: 0.025061702355742455
2023-01-21 10:15:10.125740: step: 748/527, loss: 0.0389246940612793
2023-01-21 10:15:11.218385: step: 752/527, loss: 0.046741485595703125
2023-01-21 10:15:12.321796: step: 756/527, loss: 0.10398922115564346
2023-01-21 10:15:13.487665: step: 760/527, loss: 0.09913244843482971
2023-01-21 10:15:14.594729: step: 764/527, loss: 0.11089535802602768
2023-01-21 10:15:15.721528: step: 768/527, loss: 0.015854155644774437
2023-01-21 10:15:16.840379: step: 772/527, loss: 0.034119509160518646
2023-01-21 10:15:17.969861: step: 776/527, loss: 0.005016374867409468
2023-01-21 10:15:19.082731: step: 780/527, loss: 0.09810104966163635
2023-01-21 10:15:20.183965: step: 784/527, loss: 0.050821878015995026
2023-01-21 10:15:21.273700: step: 788/527, loss: 0.04132270812988281
2023-01-21 10:15:22.385494: step: 792/527, loss: 0.00894010066986084
2023-01-21 10:15:23.486897: step: 796/527, loss: 0.026571275666356087
2023-01-21 10:15:24.606105: step: 800/527, loss: 0.03814687952399254
2023-01-21 10:15:25.705031: step: 804/527, loss: 0.0411250963807106
2023-01-21 10:15:26.824890: step: 808/527, loss: 0.12153854966163635
2023-01-21 10:15:27.971365: step: 812/527, loss: 0.1451946347951889
2023-01-21 10:15:29.087783: step: 816/527, loss: 0.05243368446826935
2023-01-21 10:15:30.201231: step: 820/527, loss: 0.008421516045928001
2023-01-21 10:15:31.334649: step: 824/527, loss: 0.0033766748383641243
2023-01-21 10:15:32.460367: step: 828/527, loss: 0.02733621746301651
2023-01-21 10:15:33.568281: step: 832/527, loss: 0.0715036392211914
2023-01-21 10:15:34.700498: step: 836/527, loss: 0.06504850834608078
2023-01-21 10:15:35.832693: step: 840/527, loss: 0.020229816436767578
2023-01-21 10:15:36.975910: step: 844/527, loss: 0.005469238851219416
2023-01-21 10:15:38.121574: step: 848/527, loss: 0.15612564980983734
2023-01-21 10:15:39.284356: step: 852/527, loss: 0.12590274214744568
2023-01-21 10:15:40.411035: step: 856/527, loss: 0.01723346672952175
2023-01-21 10:15:41.529131: step: 860/527, loss: 0.21716849505901337
2023-01-21 10:15:42.699320: step: 864/527, loss: 0.01270294189453125
2023-01-21 10:15:43.824702: step: 868/527, loss: 0.06347303092479706
2023-01-21 10:15:44.941163: step: 872/527, loss: 0.050580885261297226
2023-01-21 10:15:46.056597: step: 876/527, loss: 0.033860303461551666
2023-01-21 10:15:47.204778: step: 880/527, loss: 0.04086499288678169
2023-01-21 10:15:48.319307: step: 884/527, loss: 0.13762474060058594
2023-01-21 10:15:49.418287: step: 888/527, loss: 0.03726176917552948
2023-01-21 10:15:50.554024: step: 892/527, loss: 0.017440414056181908
2023-01-21 10:15:51.701834: step: 896/527, loss: 0.004000520799309015
2023-01-21 10:15:52.852638: step: 900/527, loss: 0.19654139876365662
2023-01-21 10:15:53.974756: step: 904/527, loss: 0.038222648203372955
2023-01-21 10:15:55.125593: step: 908/527, loss: 0.012129020877182484
2023-01-21 10:15:56.260392: step: 912/527, loss: 0.28940287232398987
2023-01-21 10:15:57.365326: step: 916/527, loss: 0.035015106201171875
2023-01-21 10:15:58.466437: step: 920/527, loss: 0.040676262229681015
2023-01-21 10:15:59.612247: step: 924/527, loss: 0.10807066410779953
2023-01-21 10:16:00.774313: step: 928/527, loss: 0.19691458344459534
2023-01-21 10:16:01.947676: step: 932/527, loss: 0.03291444852948189
2023-01-21 10:16:03.082269: step: 936/527, loss: 0.3481800854206085
2023-01-21 10:16:04.218754: step: 940/527, loss: 0.026858510449528694
2023-01-21 10:16:05.371785: step: 944/527, loss: 0.02380075491964817
2023-01-21 10:16:06.498853: step: 948/527, loss: 0.025948908179998398
2023-01-21 10:16:07.611197: step: 952/527, loss: 0.021019816398620605
2023-01-21 10:16:08.732367: step: 956/527, loss: 0.11807326972484589
2023-01-21 10:16:09.848164: step: 960/527, loss: 0.047972775995731354
2023-01-21 10:16:10.970167: step: 964/527, loss: 0.0017597198020666838
2023-01-21 10:16:12.075017: step: 968/527, loss: 0.05020351707935333
2023-01-21 10:16:13.186923: step: 972/527, loss: 0.46636271476745605
2023-01-21 10:16:14.349928: step: 976/527, loss: 0.017462920397520065
2023-01-21 10:16:15.482209: step: 980/527, loss: 0.06998234242200851
2023-01-21 10:16:16.610682: step: 984/527, loss: 0.0309771541506052
2023-01-21 10:16:17.768912: step: 988/527, loss: 0.04553508758544922
2023-01-21 10:16:18.886700: step: 992/527, loss: 0.08498553931713104
2023-01-21 10:16:19.990651: step: 996/527, loss: 0.0004536628839559853
2023-01-21 10:16:21.081915: step: 1000/527, loss: 0.025867082178592682
2023-01-21 10:16:22.206848: step: 1004/527, loss: 0.04738101735711098
2023-01-21 10:16:23.310487: step: 1008/527, loss: 0.04738654941320419
2023-01-21 10:16:24.423139: step: 1012/527, loss: 0.08932209014892578
2023-01-21 10:16:25.541336: step: 1016/527, loss: 0.21987590193748474
2023-01-21 10:16:26.673369: step: 1020/527, loss: 0.05395636707544327
2023-01-21 10:16:27.840163: step: 1024/527, loss: 0.05484752729535103
2023-01-21 10:16:28.970185: step: 1028/527, loss: 0.02601451985538006
2023-01-21 10:16:30.079786: step: 1032/527, loss: 0.04179678112268448
2023-01-21 10:16:31.181054: step: 1036/527, loss: 0.08130798488855362
2023-01-21 10:16:32.314482: step: 1040/527, loss: 0.09006405621767044
2023-01-21 10:16:33.415769: step: 1044/527, loss: 0.07062435150146484
2023-01-21 10:16:34.532634: step: 1048/527, loss: 0.0372396856546402
2023-01-21 10:16:35.626600: step: 1052/527, loss: 0.06186370551586151
2023-01-21 10:16:36.740779: step: 1056/527, loss: 0.0354766845703125
2023-01-21 10:16:37.851008: step: 1060/527, loss: 0.7298870086669922
2023-01-21 10:16:38.957915: step: 1064/527, loss: 0.08513374626636505
2023-01-21 10:16:40.052074: step: 1068/527, loss: 0.14668826758861542
2023-01-21 10:16:41.194123: step: 1072/527, loss: 0.16648262739181519
2023-01-21 10:16:42.312195: step: 1076/527, loss: 0.06459064781665802
2023-01-21 10:16:43.390567: step: 1080/527, loss: 0.020952749997377396
2023-01-21 10:16:44.488563: step: 1084/527, loss: 0.06892142444849014
2023-01-21 10:16:45.624708: step: 1088/527, loss: 0.01784381829202175
2023-01-21 10:16:46.755163: step: 1092/527, loss: 0.030136872082948685
2023-01-21 10:16:47.859069: step: 1096/527, loss: 0.0029964924324303865
2023-01-21 10:16:48.971292: step: 1100/527, loss: 0.052674200385808945
2023-01-21 10:16:50.069007: step: 1104/527, loss: 0.14742393791675568
2023-01-21 10:16:51.205288: step: 1108/527, loss: 0.043134596198797226
2023-01-21 10:16:52.354797: step: 1112/527, loss: 0.010724497027695179
2023-01-21 10:16:53.465306: step: 1116/527, loss: 0.09624017775058746
2023-01-21 10:16:54.595290: step: 1120/527, loss: 0.02912454679608345
2023-01-21 10:16:55.717364: step: 1124/527, loss: 0.07969585061073303
2023-01-21 10:16:56.827181: step: 1128/527, loss: 0.08891606330871582
2023-01-21 10:16:57.936831: step: 1132/527, loss: 0.01089935377240181
2023-01-21 10:16:59.053277: step: 1136/527, loss: 0.049041748046875
2023-01-21 10:17:00.163417: step: 1140/527, loss: 0.019298363476991653
2023-01-21 10:17:01.282641: step: 1144/527, loss: 0.09605112671852112
2023-01-21 10:17:02.403317: step: 1148/527, loss: 0.06948347389698029
2023-01-21 10:17:03.517013: step: 1152/527, loss: 0.024088477715849876
2023-01-21 10:17:04.631410: step: 1156/527, loss: 0.0025262834969908
2023-01-21 10:17:05.753199: step: 1160/527, loss: 0.01917247660458088
2023-01-21 10:17:06.910042: step: 1164/527, loss: 0.12587738037109375
2023-01-21 10:17:08.026146: step: 1168/527, loss: 0.10613860934972763
2023-01-21 10:17:09.159114: step: 1172/527, loss: 0.025966167449951172
2023-01-21 10:17:10.301222: step: 1176/527, loss: 0.044365692883729935
2023-01-21 10:17:11.397876: step: 1180/527, loss: 0.012591457925736904
2023-01-21 10:17:12.487399: step: 1184/527, loss: 0.419198602437973
2023-01-21 10:17:13.603819: step: 1188/527, loss: 0.032729052007198334
2023-01-21 10:17:14.722567: step: 1192/527, loss: 0.06766434013843536
2023-01-21 10:17:15.853629: step: 1196/527, loss: 0.03297629579901695
2023-01-21 10:17:16.979676: step: 1200/527, loss: 0.03873659670352936
2023-01-21 10:17:18.085040: step: 1204/527, loss: 0.07039690017700195
2023-01-21 10:17:19.209645: step: 1208/527, loss: 0.08883514255285263
2023-01-21 10:17:20.310857: step: 1212/527, loss: 0.09126751869916916
2023-01-21 10:17:21.424906: step: 1216/527, loss: 0.29734936356544495
2023-01-21 10:17:22.544147: step: 1220/527, loss: 0.0026366233360022306
2023-01-21 10:17:23.670368: step: 1224/527, loss: 0.036997318267822266
2023-01-21 10:17:24.786623: step: 1228/527, loss: 0.15535296499729156
2023-01-21 10:17:25.889916: step: 1232/527, loss: 0.21307361125946045
2023-01-21 10:17:27.017798: step: 1236/527, loss: 0.04601650685071945
2023-01-21 10:17:28.134457: step: 1240/527, loss: 0.04376354068517685
2023-01-21 10:17:29.282624: step: 1244/527, loss: 0.01740141026675701
2023-01-21 10:17:30.377134: step: 1248/527, loss: 0.08658389747142792
2023-01-21 10:17:31.511207: step: 1252/527, loss: 0.2184923142194748
2023-01-21 10:17:32.615908: step: 1256/527, loss: 0.021442987024784088
2023-01-21 10:17:33.781348: step: 1260/527, loss: 0.03345203399658203
2023-01-21 10:17:34.882482: step: 1264/527, loss: 0.03281059116125107
2023-01-21 10:17:36.006498: step: 1268/527, loss: 0.03055848926305771
2023-01-21 10:17:37.134650: step: 1272/527, loss: 0.16485300660133362
2023-01-21 10:17:38.228378: step: 1276/527, loss: 0.0439058318734169
2023-01-21 10:17:39.353749: step: 1280/527, loss: 0.006540775299072266
2023-01-21 10:17:40.489014: step: 1284/527, loss: 0.007890892215073109
2023-01-21 10:17:41.599776: step: 1288/527, loss: 0.006539535708725452
2023-01-21 10:17:42.716859: step: 1292/527, loss: 0.04639873653650284
2023-01-21 10:17:43.817226: step: 1296/527, loss: 0.30003756284713745
2023-01-21 10:17:44.920140: step: 1300/527, loss: 0.26014575362205505
2023-01-21 10:17:46.042495: step: 1304/527, loss: 0.04884805530309677
2023-01-21 10:17:47.145338: step: 1308/527, loss: 0.0315403938293457
2023-01-21 10:17:48.248334: step: 1312/527, loss: 0.03044353611767292
2023-01-21 10:17:49.354708: step: 1316/527, loss: 0.05885648727416992
2023-01-21 10:17:50.486839: step: 1320/527, loss: 0.04441490024328232
2023-01-21 10:17:51.622305: step: 1324/527, loss: 0.09061069786548615
2023-01-21 10:17:52.726201: step: 1328/527, loss: 0.016812991350889206
2023-01-21 10:17:53.841337: step: 1332/527, loss: 0.037131693214178085
2023-01-21 10:17:54.952322: step: 1336/527, loss: 0.010467815212905407
2023-01-21 10:17:56.085438: step: 1340/527, loss: 0.08517646789550781
2023-01-21 10:17:57.186066: step: 1344/527, loss: 0.09999943524599075
2023-01-21 10:17:58.308948: step: 1348/527, loss: 0.0068035125732421875
2023-01-21 10:17:59.418346: step: 1352/527, loss: 0.03772592544555664
2023-01-21 10:18:00.534950: step: 1356/527, loss: 0.022836869582533836
2023-01-21 10:18:01.627053: step: 1360/527, loss: 0.03131914138793945
2023-01-21 10:18:02.751080: step: 1364/527, loss: 0.05937071144580841
2023-01-21 10:18:03.882704: step: 1368/527, loss: 0.02063455618917942
2023-01-21 10:18:04.980412: step: 1372/527, loss: 0.0737181305885315
2023-01-21 10:18:06.100171: step: 1376/527, loss: 0.04599037021398544
2023-01-21 10:18:07.209165: step: 1380/527, loss: 0.07567653805017471
2023-01-21 10:18:08.376670: step: 1384/527, loss: 0.06350994110107422
2023-01-21 10:18:09.492816: step: 1388/527, loss: 0.049146365374326706
2023-01-21 10:18:10.606778: step: 1392/527, loss: 0.02857999876141548
2023-01-21 10:18:11.718737: step: 1396/527, loss: 0.08483944088220596
2023-01-21 10:18:12.853882: step: 1400/527, loss: 0.0027028084732592106
2023-01-21 10:18:13.955998: step: 1404/527, loss: 0.030332984402775764
2023-01-21 10:18:15.102087: step: 1408/527, loss: 0.005330848973244429
2023-01-21 10:18:16.220413: step: 1412/527, loss: 0.0036644935607910156
2023-01-21 10:18:17.326338: step: 1416/527, loss: 0.03710651397705078
2023-01-21 10:18:18.432791: step: 1420/527, loss: 0.0002440452662995085
2023-01-21 10:18:19.565951: step: 1424/527, loss: 0.06374912708997726
2023-01-21 10:18:20.682769: step: 1428/527, loss: 0.07393179088830948
2023-01-21 10:18:21.769582: step: 1432/527, loss: 0.021627523005008698
2023-01-21 10:18:22.881685: step: 1436/527, loss: 0.05135336145758629
2023-01-21 10:18:23.989931: step: 1440/527, loss: 0.05861806496977806
2023-01-21 10:18:25.080498: step: 1444/527, loss: 0.011415171436965466
2023-01-21 10:18:26.182535: step: 1448/527, loss: 0.07422810047864914
2023-01-21 10:18:27.315580: step: 1452/527, loss: 0.04888172075152397
2023-01-21 10:18:28.429509: step: 1456/527, loss: 0.033731650561094284
2023-01-21 10:18:29.560977: step: 1460/527, loss: 0.033193159848451614
2023-01-21 10:18:30.699863: step: 1464/527, loss: 0.27129611372947693
2023-01-21 10:18:31.802649: step: 1468/527, loss: 0.13051776587963104
2023-01-21 10:18:32.930518: step: 1472/527, loss: 0.21012958884239197
2023-01-21 10:18:34.034440: step: 1476/527, loss: 0.009187507443130016
2023-01-21 10:18:35.131868: step: 1480/527, loss: 0.05515923723578453
2023-01-21 10:18:36.254316: step: 1484/527, loss: 0.06073570251464844
2023-01-21 10:18:37.372311: step: 1488/527, loss: 0.1984787881374359
2023-01-21 10:18:38.509277: step: 1492/527, loss: 0.19858984649181366
2023-01-21 10:18:39.613166: step: 1496/527, loss: 0.023435020819306374
2023-01-21 10:18:40.735675: step: 1500/527, loss: 0.10996846854686737
2023-01-21 10:18:41.879321: step: 1504/527, loss: 0.061175063252449036
2023-01-21 10:18:42.962137: step: 1508/527, loss: 0.29333925247192383
2023-01-21 10:18:44.063127: step: 1512/527, loss: 0.04473914951086044
2023-01-21 10:18:45.164084: step: 1516/527, loss: 0.054048679769039154
2023-01-21 10:18:46.274484: step: 1520/527, loss: 0.03265733644366264
2023-01-21 10:18:47.375613: step: 1524/527, loss: 0.02030506171286106
2023-01-21 10:18:48.463570: step: 1528/527, loss: 0.0635310709476471
2023-01-21 10:18:49.594190: step: 1532/527, loss: 0.0662098377943039
2023-01-21 10:18:50.685915: step: 1536/527, loss: 0.33434516191482544
2023-01-21 10:18:51.808050: step: 1540/527, loss: 0.14279527962207794
2023-01-21 10:18:52.935872: step: 1544/527, loss: 0.1918402761220932
2023-01-21 10:18:54.044240: step: 1548/527, loss: 0.05609443411231041
2023-01-21 10:18:55.136390: step: 1552/527, loss: 0.035256434231996536
2023-01-21 10:18:56.243760: step: 1556/527, loss: 0.1993013471364975
2023-01-21 10:18:57.360085: step: 1560/527, loss: 0.19217254221439362
2023-01-21 10:18:58.471879: step: 1564/527, loss: 0.049072038382291794
2023-01-21 10:18:59.574240: step: 1568/527, loss: 0.03501834720373154
2023-01-21 10:19:00.683523: step: 1572/527, loss: 0.01261987630277872
2023-01-21 10:19:01.803394: step: 1576/527, loss: 0.12150402367115021
2023-01-21 10:19:02.930418: step: 1580/527, loss: 0.10841389000415802
2023-01-21 10:19:04.086028: step: 1584/527, loss: 0.5932765007019043
2023-01-21 10:19:05.203473: step: 1588/527, loss: 0.08304176479578018
2023-01-21 10:19:06.316271: step: 1592/527, loss: 0.3327920138835907
2023-01-21 10:19:07.440955: step: 1596/527, loss: 0.030359935015439987
2023-01-21 10:19:08.547294: step: 1600/527, loss: 0.24892206490039825
2023-01-21 10:19:09.661036: step: 1604/527, loss: 0.1113288402557373
2023-01-21 10:19:10.795776: step: 1608/527, loss: 0.08322592079639435
2023-01-21 10:19:11.925336: step: 1612/527, loss: 0.06382560729980469
2023-01-21 10:19:13.061577: step: 1616/527, loss: 0.09297456592321396
2023-01-21 10:19:14.191263: step: 1620/527, loss: 0.029099320992827415
2023-01-21 10:19:15.302529: step: 1624/527, loss: 0.071587473154068
2023-01-21 10:19:16.447504: step: 1628/527, loss: 0.0069536687806248665
2023-01-21 10:19:17.607209: step: 1632/527, loss: 0.04498634487390518
2023-01-21 10:19:18.728565: step: 1636/527, loss: 0.007207393646240234
2023-01-21 10:19:19.854361: step: 1640/527, loss: 0.00955953635275364
2023-01-21 10:19:20.948318: step: 1644/527, loss: 0.09483642876148224
2023-01-21 10:19:22.050529: step: 1648/527, loss: 0.08596763759851456
2023-01-21 10:19:23.142030: step: 1652/527, loss: 0.050000667572021484
2023-01-21 10:19:24.258835: step: 1656/527, loss: 0.6716679334640503
2023-01-21 10:19:25.340439: step: 1660/527, loss: 0.03650989755988121
2023-01-21 10:19:26.444515: step: 1664/527, loss: 0.005126953125
2023-01-21 10:19:27.592185: step: 1668/527, loss: 0.08572454750537872
2023-01-21 10:19:28.719737: step: 1672/527, loss: 0.058576539158821106
2023-01-21 10:19:29.838653: step: 1676/527, loss: 0.018717478960752487
2023-01-21 10:19:30.938460: step: 1680/527, loss: 0.05393677204847336
2023-01-21 10:19:32.014897: step: 1684/527, loss: 0.04691710323095322
2023-01-21 10:19:33.118984: step: 1688/527, loss: 0.0034494400024414062
2023-01-21 10:19:34.240953: step: 1692/527, loss: 0.06828327476978302
2023-01-21 10:19:35.338154: step: 1696/527, loss: 0.052118588238954544
2023-01-21 10:19:36.455714: step: 1700/527, loss: 0.14473260939121246
2023-01-21 10:19:37.534376: step: 1704/527, loss: 0.18068094551563263
2023-01-21 10:19:38.638608: step: 1708/527, loss: 0.044092558324337006
2023-01-21 10:19:39.762051: step: 1712/527, loss: 0.13796940445899963
2023-01-21 10:19:40.878642: step: 1716/527, loss: 0.0381770133972168
2023-01-21 10:19:42.014571: step: 1720/527, loss: 0.07415933907032013
2023-01-21 10:19:43.143652: step: 1724/527, loss: 0.06726837158203125
2023-01-21 10:19:44.262061: step: 1728/527, loss: 0.00650444021448493
2023-01-21 10:19:45.360266: step: 1732/527, loss: 0.08492393046617508
2023-01-21 10:19:46.505412: step: 1736/527, loss: 0.01097936648875475
2023-01-21 10:19:47.603599: step: 1740/527, loss: 0.042238976806402206
2023-01-21 10:19:48.691097: step: 1744/527, loss: 0.1559789776802063
2023-01-21 10:19:49.815129: step: 1748/527, loss: 0.12835398316383362
2023-01-21 10:19:50.934175: step: 1752/527, loss: 1.0934780836105347
2023-01-21 10:19:52.056006: step: 1756/527, loss: 0.2530114948749542
2023-01-21 10:19:53.169879: step: 1760/527, loss: 0.020436763763427734
2023-01-21 10:19:54.293342: step: 1764/527, loss: 0.024929428473114967
2023-01-21 10:19:55.443614: step: 1768/527, loss: 0.1303289532661438
2023-01-21 10:19:56.584320: step: 1772/527, loss: 0.07172908633947372
2023-01-21 10:19:57.700311: step: 1776/527, loss: 0.06466171145439148
2023-01-21 10:19:58.821787: step: 1780/527, loss: 0.02956414222717285
2023-01-21 10:19:59.934177: step: 1784/527, loss: 0.09388022869825363
2023-01-21 10:20:01.055121: step: 1788/527, loss: 0.025692511349916458
2023-01-21 10:20:02.156639: step: 1792/527, loss: 0.03701906278729439
2023-01-21 10:20:03.288241: step: 1796/527, loss: 0.025358105078339577
2023-01-21 10:20:04.400375: step: 1800/527, loss: 0.08769816905260086
2023-01-21 10:20:05.540892: step: 1804/527, loss: 0.05394699424505234
2023-01-21 10:20:06.685906: step: 1808/527, loss: 0.08894510567188263
2023-01-21 10:20:07.769619: step: 1812/527, loss: 0.03907956928014755
2023-01-21 10:20:08.859655: step: 1816/527, loss: 0.04012441635131836
2023-01-21 10:20:09.944470: step: 1820/527, loss: 0.029288865625858307
2023-01-21 10:20:11.044697: step: 1824/527, loss: 0.00815882720053196
2023-01-21 10:20:12.156228: step: 1828/527, loss: 0.04130678251385689
2023-01-21 10:20:13.280121: step: 1832/527, loss: 0.0510869026184082
2023-01-21 10:20:14.391823: step: 1836/527, loss: 0.050705622881650925
2023-01-21 10:20:15.512392: step: 1840/527, loss: 0.1016816571354866
2023-01-21 10:20:16.614529: step: 1844/527, loss: 0.5428416132926941
2023-01-21 10:20:17.736333: step: 1848/527, loss: 0.14480562508106232
2023-01-21 10:20:18.828870: step: 1852/527, loss: 0.03458847850561142
2023-01-21 10:20:19.972870: step: 1856/527, loss: 0.0006271362071856856
2023-01-21 10:20:21.077643: step: 1860/527, loss: 0.07497739791870117
2023-01-21 10:20:22.227966: step: 1864/527, loss: 0.7174075245857239
2023-01-21 10:20:23.336842: step: 1868/527, loss: 0.035933688282966614
2023-01-21 10:20:24.458283: step: 1872/527, loss: 0.13449449837207794
2023-01-21 10:20:25.596829: step: 1876/527, loss: 0.023523710668087006
2023-01-21 10:20:26.715014: step: 1880/527, loss: 0.07532148063182831
2023-01-21 10:20:27.811602: step: 1884/527, loss: 0.1139548271894455
2023-01-21 10:20:28.941936: step: 1888/527, loss: 0.05602336302399635
2023-01-21 10:20:30.090944: step: 1892/527, loss: 0.10650572925806046
2023-01-21 10:20:31.204437: step: 1896/527, loss: 0.025278665125370026
2023-01-21 10:20:32.298229: step: 1900/527, loss: 0.16824665665626526
2023-01-21 10:20:33.434291: step: 1904/527, loss: 0.05643191188573837
2023-01-21 10:20:34.572588: step: 1908/527, loss: 0.14879217743873596
2023-01-21 10:20:35.708879: step: 1912/527, loss: 0.20584160089492798
2023-01-21 10:20:36.846034: step: 1916/527, loss: 0.003959703724831343
2023-01-21 10:20:38.000177: step: 1920/527, loss: 0.07251858711242676
2023-01-21 10:20:39.121782: step: 1924/527, loss: 0.2500073313713074
2023-01-21 10:20:40.224574: step: 1928/527, loss: 0.0089385025203228
2023-01-21 10:20:41.364142: step: 1932/527, loss: 0.024903345853090286
2023-01-21 10:20:42.467569: step: 1936/527, loss: 0.033914946019649506
2023-01-21 10:20:43.574233: step: 1940/527, loss: 0.06158857420086861
2023-01-21 10:20:44.691079: step: 1944/527, loss: 0.11178407818078995
2023-01-21 10:20:45.795116: step: 1948/527, loss: 0.003274250077083707
2023-01-21 10:20:46.905759: step: 1952/527, loss: 0.02889099158346653
2023-01-21 10:20:48.022208: step: 1956/527, loss: 0.10508528351783752
2023-01-21 10:20:49.135625: step: 1960/527, loss: 0.05529346689581871
2023-01-21 10:20:50.239505: step: 1964/527, loss: 0.0734872817993164
2023-01-21 10:20:51.361651: step: 1968/527, loss: 0.06625165790319443
2023-01-21 10:20:52.497765: step: 1972/527, loss: 0.006357860751450062
2023-01-21 10:20:53.689888: step: 1976/527, loss: 0.06232185661792755
2023-01-21 10:20:54.815929: step: 1980/527, loss: 0.05676531791687012
2023-01-21 10:20:55.917145: step: 1984/527, loss: 0.21562395989894867
2023-01-21 10:20:57.037578: step: 1988/527, loss: 0.10727329552173615
2023-01-21 10:20:58.140560: step: 1992/527, loss: 0.045992087572813034
2023-01-21 10:20:59.251241: step: 1996/527, loss: 0.013383293524384499
2023-01-21 10:21:00.336764: step: 2000/527, loss: 0.008558845147490501
2023-01-21 10:21:01.432558: step: 2004/527, loss: 0.31395137310028076
2023-01-21 10:21:02.545986: step: 2008/527, loss: 0.006506443023681641
2023-01-21 10:21:03.634236: step: 2012/527, loss: 0.03435507044196129
2023-01-21 10:21:04.734101: step: 2016/527, loss: 0.4994511306285858
2023-01-21 10:21:05.835860: step: 2020/527, loss: 0.0842091292142868
2023-01-21 10:21:06.968935: step: 2024/527, loss: 0.13641434907913208
2023-01-21 10:21:08.088213: step: 2028/527, loss: 0.06823810935020447
2023-01-21 10:21:09.194584: step: 2032/527, loss: 0.2879948616027832
2023-01-21 10:21:10.313247: step: 2036/527, loss: 0.1547890305519104
2023-01-21 10:21:11.440609: step: 2040/527, loss: 0.024220729246735573
2023-01-21 10:21:12.560516: step: 2044/527, loss: 0.10612693428993225
2023-01-21 10:21:13.680141: step: 2048/527, loss: 0.07173619419336319
2023-01-21 10:21:14.767377: step: 2052/527, loss: 0.04888134077191353
2023-01-21 10:21:15.878332: step: 2056/527, loss: 0.04329204559326172
2023-01-21 10:21:16.982964: step: 2060/527, loss: 0.10113182663917542
2023-01-21 10:21:18.087252: step: 2064/527, loss: 0.11236695945262909
2023-01-21 10:21:19.192315: step: 2068/527, loss: 0.01601400412619114
2023-01-21 10:21:20.308012: step: 2072/527, loss: 0.0417846217751503
2023-01-21 10:21:21.393467: step: 2076/527, loss: 0.011335277929902077
2023-01-21 10:21:22.524055: step: 2080/527, loss: 0.6258201003074646
2023-01-21 10:21:23.672669: step: 2084/527, loss: 0.09557032585144043
2023-01-21 10:21:24.787986: step: 2088/527, loss: 0.3243850767612457
2023-01-21 10:21:25.956099: step: 2092/527, loss: 0.06134028360247612
2023-01-21 10:21:27.059198: step: 2096/527, loss: 0.04106311872601509
2023-01-21 10:21:28.165267: step: 2100/527, loss: 0.5914804339408875
2023-01-21 10:21:29.286422: step: 2104/527, loss: 0.01369571778923273
2023-01-21 10:21:30.372779: step: 2108/527, loss: 0.026768207550048828
==================================================
Loss: 0.083
--------------------
Dev: {'event': {'p': 0.6084275436793423, 'r': 0.7882822902796272, 'f1': 0.6867749419953597}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10}
Test: {'event': {'p': 0.6218149307107733, 'r': 0.7948571428571428, 'f1': 0.6977677451718083}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10}
Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10}
Korean: {'event': {'p': 0.6382978723404256, 'r': 0.47619047619047616, 'f1': 0.5454545454545455}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10}
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': 10}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:22:12.736662: step: 4/527, loss: 0.08067798614501953
2023-01-21 10:22:13.845658: step: 8/527, loss: 0.04230527952313423
2023-01-21 10:22:14.985920: step: 12/527, loss: 0.04908175766468048
2023-01-21 10:22:16.108842: step: 16/527, loss: 0.03007793426513672
2023-01-21 10:22:17.255274: step: 20/527, loss: 0.01850900612771511
2023-01-21 10:22:18.382410: step: 24/527, loss: 0.020067691802978516
2023-01-21 10:22:19.504578: step: 28/527, loss: 0.025117874145507812
2023-01-21 10:22:20.645095: step: 32/527, loss: 0.038509272038936615
2023-01-21 10:22:21.763788: step: 36/527, loss: 0.000927352870348841
2023-01-21 10:22:22.897096: step: 40/527, loss: 0.046708106994628906
2023-01-21 10:22:24.022342: step: 44/527, loss: 0.01110992394387722
2023-01-21 10:22:25.167239: step: 48/527, loss: 0.04392719268798828
2023-01-21 10:22:26.272243: step: 52/527, loss: 0.052697114646434784
2023-01-21 10:22:27.394584: step: 56/527, loss: 0.015467071905732155
2023-01-21 10:22:28.542793: step: 60/527, loss: 0.02496185339987278
2023-01-21 10:22:29.668670: step: 64/527, loss: 0.001474666642025113
2023-01-21 10:22:30.823778: step: 68/527, loss: 0.04135601222515106
2023-01-21 10:22:31.994509: step: 72/527, loss: 0.12225857377052307
2023-01-21 10:22:33.139376: step: 76/527, loss: 0.12115774303674698
2023-01-21 10:22:34.256696: step: 80/527, loss: 0.0560920275747776
2023-01-21 10:22:35.401435: step: 84/527, loss: 0.022900868207216263
2023-01-21 10:22:36.566228: step: 88/527, loss: 0.0537288673222065
2023-01-21 10:22:37.713338: step: 92/527, loss: 0.0813688263297081
2023-01-21 10:22:38.844929: step: 96/527, loss: 0.1555161476135254
2023-01-21 10:22:39.983816: step: 100/527, loss: 0.1209169402718544
2023-01-21 10:22:41.095193: step: 104/527, loss: 0.04698324203491211
2023-01-21 10:22:42.240783: step: 108/527, loss: 0.06603773683309555
2023-01-21 10:22:43.372055: step: 112/527, loss: 0.18267536163330078
2023-01-21 10:22:44.513528: step: 116/527, loss: 0.0006688118446618319
2023-01-21 10:22:45.654878: step: 120/527, loss: 0.022388862445950508
2023-01-21 10:22:46.769118: step: 124/527, loss: 0.021190166473388672
2023-01-21 10:22:47.917424: step: 128/527, loss: 0.06566648185253143
2023-01-21 10:22:49.039162: step: 132/527, loss: 0.03349723666906357
2023-01-21 10:22:50.175137: step: 136/527, loss: 0.07213124632835388
2023-01-21 10:22:51.302976: step: 140/527, loss: 0.13203755021095276
2023-01-21 10:22:52.442801: step: 144/527, loss: 0.09245087206363678
2023-01-21 10:22:53.568281: step: 148/527, loss: 0.030596591532230377
2023-01-21 10:22:54.681414: step: 152/527, loss: 0.006259727291762829
2023-01-21 10:22:55.783077: step: 156/527, loss: 0.03528337553143501
2023-01-21 10:22:56.902533: step: 160/527, loss: 0.04161083698272705
2023-01-21 10:22:58.090053: step: 164/527, loss: 0.057959459722042084
2023-01-21 10:22:59.247432: step: 168/527, loss: 0.007411670871078968
2023-01-21 10:23:00.384707: step: 172/527, loss: 0.028269432485103607
2023-01-21 10:23:01.512655: step: 176/527, loss: 0.03533497080206871
2023-01-21 10:23:02.649862: step: 180/527, loss: 0.05302934721112251
2023-01-21 10:23:03.776508: step: 184/527, loss: 0.03431425243616104
2023-01-21 10:23:04.955125: step: 188/527, loss: 0.013964367099106312
2023-01-21 10:23:06.100262: step: 192/527, loss: 0.0025426866486668587
2023-01-21 10:23:07.228181: step: 196/527, loss: 0.014343834482133389
2023-01-21 10:23:08.368471: step: 200/527, loss: 0.04323830455541611
2023-01-21 10:23:09.503902: step: 204/527, loss: 0.022257614880800247
2023-01-21 10:23:10.656270: step: 208/527, loss: 0.05538616329431534
2023-01-21 10:23:11.803996: step: 212/527, loss: 0.015301132574677467
2023-01-21 10:23:12.914266: step: 216/527, loss: 0.0043236734345555305
2023-01-21 10:23:14.044771: step: 220/527, loss: 0.17806226015090942
2023-01-21 10:23:15.186025: step: 224/527, loss: 0.059856511652469635
2023-01-21 10:23:16.321124: step: 228/527, loss: 0.002015733625739813
2023-01-21 10:23:17.433716: step: 232/527, loss: 0.07700204849243164
2023-01-21 10:23:18.563476: step: 236/527, loss: 0.007751799188554287
2023-01-21 10:23:19.727484: step: 240/527, loss: 0.05842466652393341
2023-01-21 10:23:20.858449: step: 244/527, loss: 0.03313078731298447
2023-01-21 10:23:21.977648: step: 248/527, loss: 0.0014502524863928556
2023-01-21 10:23:23.100595: step: 252/527, loss: 0.05644569545984268
2023-01-21 10:23:24.253082: step: 256/527, loss: 0.028081322088837624
2023-01-21 10:23:25.396657: step: 260/527, loss: 0.21366047859191895
2023-01-21 10:23:26.511410: step: 264/527, loss: 0.007308769039809704
2023-01-21 10:23:27.613812: step: 268/527, loss: 0.003892803331837058
2023-01-21 10:23:28.786760: step: 272/527, loss: 0.007554054260253906
2023-01-21 10:23:29.921444: step: 276/527, loss: 0.08761701732873917
2023-01-21 10:23:31.057275: step: 280/527, loss: 0.025003815069794655
2023-01-21 10:23:32.191997: step: 284/527, loss: 0.20140905678272247
2023-01-21 10:23:33.289167: step: 288/527, loss: 0.11489498615264893
2023-01-21 10:23:34.434544: step: 292/527, loss: 0.04347868263721466
2023-01-21 10:23:35.553039: step: 296/527, loss: 0.5309319496154785
2023-01-21 10:23:36.686559: step: 300/527, loss: 0.033547595143318176
2023-01-21 10:23:37.835637: step: 304/527, loss: 0.0021164892241358757
2023-01-21 10:23:38.952377: step: 308/527, loss: 0.02490692213177681
2023-01-21 10:23:40.072507: step: 312/527, loss: 0.030603598803281784
2023-01-21 10:23:41.184796: step: 316/527, loss: 0.16288775205612183
2023-01-21 10:23:42.321334: step: 320/527, loss: 0.13951349258422852
2023-01-21 10:23:43.460270: step: 324/527, loss: 0.026932813227176666
2023-01-21 10:23:44.634152: step: 328/527, loss: 0.07625216990709305
2023-01-21 10:23:45.774871: step: 332/527, loss: 0.046442318707704544
2023-01-21 10:23:46.943294: step: 336/527, loss: 0.08221641182899475
2023-01-21 10:23:48.073407: step: 340/527, loss: 0.036043357104063034
2023-01-21 10:23:49.204775: step: 344/527, loss: 0.5584023594856262
2023-01-21 10:23:50.330687: step: 348/527, loss: 0.036026570945978165
2023-01-21 10:23:51.478901: step: 352/527, loss: 0.04068164899945259
2023-01-21 10:23:52.643681: step: 356/527, loss: 0.05809221416711807
2023-01-21 10:23:53.791923: step: 360/527, loss: 0.06390233337879181
2023-01-21 10:23:54.919970: step: 364/527, loss: 0.017533399164676666
2023-01-21 10:23:56.077154: step: 368/527, loss: 0.017421532422304153
2023-01-21 10:23:57.184180: step: 372/527, loss: 0.048799898475408554
2023-01-21 10:23:58.289607: step: 376/527, loss: 0.04963326454162598
2023-01-21 10:23:59.428075: step: 380/527, loss: 0.9355354309082031
2023-01-21 10:24:00.562390: step: 384/527, loss: 0.05899477005004883
2023-01-21 10:24:01.696499: step: 388/527, loss: 0.051241204142570496
2023-01-21 10:24:02.837806: step: 392/527, loss: 0.08990363776683807
2023-01-21 10:24:03.984569: step: 396/527, loss: 0.033190157264471054
2023-01-21 10:24:05.125731: step: 400/527, loss: 0.06903620064258575
2023-01-21 10:24:06.232720: step: 404/527, loss: 0.016281701624393463
2023-01-21 10:24:07.373146: step: 408/527, loss: 0.04275636747479439
2023-01-21 10:24:08.507569: step: 412/527, loss: 0.09220924228429794
2023-01-21 10:24:09.670019: step: 416/527, loss: 0.7665678262710571
2023-01-21 10:24:10.776799: step: 420/527, loss: 0.0925624817609787
2023-01-21 10:24:11.901424: step: 424/527, loss: 0.04685230180621147
2023-01-21 10:24:13.019692: step: 428/527, loss: 0.07393841445446014
2023-01-21 10:24:14.122747: step: 432/527, loss: 0.019419383257627487
2023-01-21 10:24:15.242060: step: 436/527, loss: 0.06814035773277283
2023-01-21 10:24:16.339258: step: 440/527, loss: 0.00699958810582757
2023-01-21 10:24:17.473911: step: 444/527, loss: 0.03958435356616974
2023-01-21 10:24:18.612641: step: 448/527, loss: 0.0026604654267430305
2023-01-21 10:24:19.723878: step: 452/527, loss: 0.0662652999162674
2023-01-21 10:24:20.850825: step: 456/527, loss: 0.004737568087875843
2023-01-21 10:24:21.985721: step: 460/527, loss: 0.050614356994628906
2023-01-21 10:24:23.088984: step: 464/527, loss: 0.0025129318237304688
2023-01-21 10:24:24.234852: step: 468/527, loss: 0.013509177602827549
2023-01-21 10:24:25.398034: step: 472/527, loss: 0.08616548031568527
2023-01-21 10:24:26.555302: step: 476/527, loss: 0.050011731684207916
2023-01-21 10:24:27.718963: step: 480/527, loss: 0.06006794050335884
2023-01-21 10:24:28.841813: step: 484/527, loss: 0.025157546624541283
2023-01-21 10:24:29.961598: step: 488/527, loss: 0.0459684394299984
2023-01-21 10:24:31.091442: step: 492/527, loss: 0.003686761949211359
2023-01-21 10:24:32.218449: step: 496/527, loss: 0.09003839641809464
2023-01-21 10:24:33.370057: step: 500/527, loss: 0.332754522562027
2023-01-21 10:24:34.491848: step: 504/527, loss: 0.004305028822273016
2023-01-21 10:24:35.624961: step: 508/527, loss: 0.01154098566621542
2023-01-21 10:24:36.760575: step: 512/527, loss: 0.008574867621064186
2023-01-21 10:24:37.869687: step: 516/527, loss: 0.029845476150512695
2023-01-21 10:24:39.012419: step: 520/527, loss: 0.030349839478731155
2023-01-21 10:24:40.139331: step: 524/527, loss: 0.13709792494773865
2023-01-21 10:24:41.267298: step: 528/527, loss: 0.008143424987792969
2023-01-21 10:24:42.423567: step: 532/527, loss: 0.006602668669074774
2023-01-21 10:24:43.579319: step: 536/527, loss: 0.009474802762269974
2023-01-21 10:24:44.688594: step: 540/527, loss: 0.1374446004629135
2023-01-21 10:24:45.843062: step: 544/527, loss: 0.020861387252807617
2023-01-21 10:24:46.943519: step: 548/527, loss: 0.06290445476770401
2023-01-21 10:24:48.046666: step: 552/527, loss: 0.08680963516235352
2023-01-21 10:24:49.132839: step: 556/527, loss: 0.020316505804657936
2023-01-21 10:24:50.262842: step: 560/527, loss: 0.22358588874340057
2023-01-21 10:24:51.390129: step: 564/527, loss: 0.029083536937832832
2023-01-21 10:24:52.499699: step: 568/527, loss: 0.012924958020448685
2023-01-21 10:24:53.624116: step: 572/527, loss: 0.1281568557024002
2023-01-21 10:24:54.713430: step: 576/527, loss: 0.07554297149181366
2023-01-21 10:24:55.849542: step: 580/527, loss: 0.11508617550134659
2023-01-21 10:24:57.007533: step: 584/527, loss: 0.022138023748993874
2023-01-21 10:24:58.104960: step: 588/527, loss: 0.008836365304887295
2023-01-21 10:24:59.225793: step: 592/527, loss: 0.06644859910011292
2023-01-21 10:25:00.399187: step: 596/527, loss: 0.08165421336889267
2023-01-21 10:25:01.515386: step: 600/527, loss: 0.12328466773033142
2023-01-21 10:25:02.657766: step: 604/527, loss: 0.023034285753965378
2023-01-21 10:25:03.772775: step: 608/527, loss: 0.08964891731739044
2023-01-21 10:25:04.858215: step: 612/527, loss: 0.008318711072206497
2023-01-21 10:25:05.984039: step: 616/527, loss: 0.06161961704492569
2023-01-21 10:25:07.067727: step: 620/527, loss: 0.022675370797514915
2023-01-21 10:25:08.163557: step: 624/527, loss: 0.3809185028076172
2023-01-21 10:25:09.264287: step: 628/527, loss: 0.00697560328990221
2023-01-21 10:25:10.391284: step: 632/527, loss: 0.015552139841020107
2023-01-21 10:25:11.503754: step: 636/527, loss: 0.10790614783763885
2023-01-21 10:25:12.634865: step: 640/527, loss: 0.124384306371212
2023-01-21 10:25:13.758557: step: 644/527, loss: 0.02337665669620037
2023-01-21 10:25:14.873776: step: 648/527, loss: 0.05483550950884819
2023-01-21 10:25:15.965834: step: 652/527, loss: 0.0023277283180505037
2023-01-21 10:25:17.112822: step: 656/527, loss: 0.03931853920221329
2023-01-21 10:25:18.232797: step: 660/527, loss: 0.053301528096199036
2023-01-21 10:25:19.372839: step: 664/527, loss: 0.028059815987944603
2023-01-21 10:25:20.519768: step: 668/527, loss: 0.2204572856426239
2023-01-21 10:25:21.674535: step: 672/527, loss: 0.10398168861865997
2023-01-21 10:25:22.793248: step: 676/527, loss: 0.08302703499794006
2023-01-21 10:25:23.917386: step: 680/527, loss: 0.007313197944313288
2023-01-21 10:25:25.027985: step: 684/527, loss: 0.03017606772482395
2023-01-21 10:25:26.129667: step: 688/527, loss: 0.03649766743183136
2023-01-21 10:25:27.253105: step: 692/527, loss: 0.02656860277056694
2023-01-21 10:25:28.377382: step: 696/527, loss: 0.03309822082519531
2023-01-21 10:25:29.494374: step: 700/527, loss: 0.03545217588543892
2023-01-21 10:25:30.614317: step: 704/527, loss: 0.004703187849372625
2023-01-21 10:25:31.743051: step: 708/527, loss: 0.02754044532775879
2023-01-21 10:25:32.855249: step: 712/527, loss: 0.015688514336943626
2023-01-21 10:25:33.989935: step: 716/527, loss: 0.006097698118537664
2023-01-21 10:25:35.103397: step: 720/527, loss: 0.005704117007553577
2023-01-21 10:25:36.231836: step: 724/527, loss: 0.02433185465633869
2023-01-21 10:25:37.331364: step: 728/527, loss: 0.05707826837897301
2023-01-21 10:25:38.450113: step: 732/527, loss: 0.17664051055908203
2023-01-21 10:25:39.571453: step: 736/527, loss: 0.489203542470932
2023-01-21 10:25:40.670081: step: 740/527, loss: 0.03538022190332413
2023-01-21 10:25:41.783241: step: 744/527, loss: 0.30911150574684143
2023-01-21 10:25:42.871747: step: 748/527, loss: 0.12843838334083557
2023-01-21 10:25:44.016194: step: 752/527, loss: 0.17990216612815857
2023-01-21 10:25:45.107875: step: 756/527, loss: 0.07838056236505508
2023-01-21 10:25:46.217647: step: 760/527, loss: 0.3793204426765442
2023-01-21 10:25:47.357911: step: 764/527, loss: 0.012790108099579811
2023-01-21 10:25:48.497413: step: 768/527, loss: 0.04151914268732071
2023-01-21 10:25:49.636911: step: 772/527, loss: 0.025007151067256927
2023-01-21 10:25:50.766533: step: 776/527, loss: 0.06588192284107208
2023-01-21 10:25:51.918544: step: 780/527, loss: 0.006550026126205921
2023-01-21 10:25:53.070475: step: 784/527, loss: 0.11618877202272415
2023-01-21 10:25:54.173959: step: 788/527, loss: 0.0011536121601238847
2023-01-21 10:25:55.327708: step: 792/527, loss: 0.009385443292558193
2023-01-21 10:25:56.446785: step: 796/527, loss: 0.09038610756397247
2023-01-21 10:25:57.615660: step: 800/527, loss: 0.48065185546875
2023-01-21 10:25:58.776034: step: 804/527, loss: 0.03906438499689102
2023-01-21 10:25:59.881643: step: 808/527, loss: 0.043680667877197266
2023-01-21 10:26:01.008684: step: 812/527, loss: 0.010559607297182083
2023-01-21 10:26:02.126216: step: 816/527, loss: 0.04532623291015625
2023-01-21 10:26:03.244759: step: 820/527, loss: 0.06716423481702805
2023-01-21 10:26:04.356703: step: 824/527, loss: 0.03185443952679634
2023-01-21 10:26:05.474064: step: 828/527, loss: 0.050204064697027206
2023-01-21 10:26:06.602812: step: 832/527, loss: 0.013174057006835938
2023-01-21 10:26:07.736967: step: 836/527, loss: 0.029108811169862747
2023-01-21 10:26:08.816810: step: 840/527, loss: 0.00281867990270257
2023-01-21 10:26:10.002396: step: 844/527, loss: 0.09427967667579651
2023-01-21 10:26:11.142670: step: 848/527, loss: 0.014589118771255016
2023-01-21 10:26:12.265602: step: 852/527, loss: 0.12413787841796875
2023-01-21 10:26:13.389545: step: 856/527, loss: 0.0020724297501146793
2023-01-21 10:26:14.509520: step: 860/527, loss: 0.13680876791477203
2023-01-21 10:26:15.622762: step: 864/527, loss: 0.0005034446949139237
2023-01-21 10:26:16.766913: step: 868/527, loss: 0.005815601442009211
2023-01-21 10:26:17.880947: step: 872/527, loss: 0.08673496544361115
2023-01-21 10:26:19.028572: step: 876/527, loss: 0.05387873575091362
2023-01-21 10:26:20.182517: step: 880/527, loss: 0.0957273468375206
2023-01-21 10:26:21.289217: step: 884/527, loss: 0.12776851654052734
2023-01-21 10:26:22.452446: step: 888/527, loss: 0.034682273864746094
2023-01-21 10:26:23.589174: step: 892/527, loss: 0.013392925262451172
2023-01-21 10:26:24.758111: step: 896/527, loss: 0.1638377159833908
2023-01-21 10:26:25.870589: step: 900/527, loss: 0.03973083943128586
2023-01-21 10:26:26.996158: step: 904/527, loss: 0.01538238488137722
2023-01-21 10:26:28.112875: step: 908/527, loss: 0.021627331152558327
2023-01-21 10:26:29.241267: step: 912/527, loss: 0.005932795815169811
2023-01-21 10:26:30.373022: step: 916/527, loss: 0.01120824832469225
2023-01-21 10:26:31.513532: step: 920/527, loss: 0.28511470556259155
2023-01-21 10:26:32.683547: step: 924/527, loss: 0.007882118225097656
2023-01-21 10:26:33.809547: step: 928/527, loss: 0.033103276044130325
2023-01-21 10:26:34.983924: step: 932/527, loss: 0.019672680646181107
2023-01-21 10:26:36.158011: step: 936/527, loss: 0.005472755525261164
2023-01-21 10:26:37.287455: step: 940/527, loss: 0.08514904975891113
2023-01-21 10:26:38.415610: step: 944/527, loss: 0.711452305316925
2023-01-21 10:26:39.544334: step: 948/527, loss: 0.04166736826300621
2023-01-21 10:26:40.735295: step: 952/527, loss: 0.013927007094025612
2023-01-21 10:26:41.864482: step: 956/527, loss: 0.28360825777053833
2023-01-21 10:26:42.976690: step: 960/527, loss: 0.03162822872400284
2023-01-21 10:26:44.157563: step: 964/527, loss: 0.005666160956025124
2023-01-21 10:26:45.332742: step: 968/527, loss: 0.08374099433422089
2023-01-21 10:26:46.470392: step: 972/527, loss: 0.005027103237807751
2023-01-21 10:26:47.571550: step: 976/527, loss: 0.043906304985284805
2023-01-21 10:26:48.681163: step: 980/527, loss: 0.01409902609884739
2023-01-21 10:26:49.794462: step: 984/527, loss: 0.08769810199737549
2023-01-21 10:26:50.954298: step: 988/527, loss: 0.35960617661476135
2023-01-21 10:26:52.087634: step: 992/527, loss: 0.001975440885871649
2023-01-21 10:26:53.205781: step: 996/527, loss: 0.0035919665824621916
2023-01-21 10:26:54.349438: step: 1000/527, loss: 0.04291825369000435
2023-01-21 10:26:55.489263: step: 1004/527, loss: 0.019007397815585136
2023-01-21 10:26:56.627216: step: 1008/527, loss: 0.030747700482606888
2023-01-21 10:26:57.776486: step: 1012/527, loss: 0.05249347910284996
2023-01-21 10:26:58.903976: step: 1016/527, loss: 0.00330104841850698
2023-01-21 10:27:00.093169: step: 1020/527, loss: 0.05303707346320152
2023-01-21 10:27:01.217048: step: 1024/527, loss: 0.002799081616103649
2023-01-21 10:27:02.382776: step: 1028/527, loss: 0.13503780961036682
2023-01-21 10:27:03.535579: step: 1032/527, loss: 0.04094085842370987
2023-01-21 10:27:04.680525: step: 1036/527, loss: 0.06127309799194336
2023-01-21 10:27:05.857531: step: 1040/527, loss: 0.10088405758142471
2023-01-21 10:27:07.010579: step: 1044/527, loss: 0.038503360003232956
2023-01-21 10:27:08.174229: step: 1048/527, loss: 0.31427013874053955
2023-01-21 10:27:09.325809: step: 1052/527, loss: 0.022220849990844727
2023-01-21 10:27:10.450799: step: 1056/527, loss: 0.06624408066272736
2023-01-21 10:27:11.586206: step: 1060/527, loss: 0.03269600868225098
2023-01-21 10:27:12.735489: step: 1064/527, loss: 0.015769578516483307
2023-01-21 10:27:13.850900: step: 1068/527, loss: 0.12198818475008011
2023-01-21 10:27:14.955368: step: 1072/527, loss: 0.00968790054321289
2023-01-21 10:27:16.075354: step: 1076/527, loss: 0.007360649295151234
2023-01-21 10:27:17.202249: step: 1080/527, loss: 0.09726858139038086
2023-01-21 10:27:18.339526: step: 1084/527, loss: 0.02572495862841606
2023-01-21 10:27:19.489341: step: 1088/527, loss: 0.06369371712207794
2023-01-21 10:27:20.669097: step: 1092/527, loss: 0.14304561913013458
2023-01-21 10:27:21.804590: step: 1096/527, loss: 0.07329888641834259
2023-01-21 10:27:22.900173: step: 1100/527, loss: 0.009450912475585938
2023-01-21 10:27:24.035261: step: 1104/527, loss: 0.17298002541065216
2023-01-21 10:27:25.163796: step: 1108/527, loss: 0.0022864341735839844
2023-01-21 10:27:26.321533: step: 1112/527, loss: 0.006158983800560236
2023-01-21 10:27:27.462532: step: 1116/527, loss: 0.014201736077666283
2023-01-21 10:27:28.615327: step: 1120/527, loss: 0.013581514358520508
2023-01-21 10:27:29.740691: step: 1124/527, loss: 0.07646389305591583
2023-01-21 10:27:30.858847: step: 1128/527, loss: 0.6779316067695618
2023-01-21 10:27:31.993073: step: 1132/527, loss: 0.14257989823818207
2023-01-21 10:27:33.119582: step: 1136/527, loss: 0.11525392532348633
2023-01-21 10:27:34.272108: step: 1140/527, loss: 0.07272128760814667
2023-01-21 10:27:35.376208: step: 1144/527, loss: 0.030729390680789948
2023-01-21 10:27:36.513238: step: 1148/527, loss: 0.01055831927806139
2023-01-21 10:27:37.660872: step: 1152/527, loss: 0.015939807519316673
2023-01-21 10:27:38.801302: step: 1156/527, loss: 0.03135376051068306
2023-01-21 10:27:39.949738: step: 1160/527, loss: 0.0008707046508789062
2023-01-21 10:27:41.089485: step: 1164/527, loss: 0.016486549749970436
2023-01-21 10:27:42.224080: step: 1168/527, loss: 0.2746197283267975
2023-01-21 10:27:43.386308: step: 1172/527, loss: 0.21655531227588654
2023-01-21 10:27:44.520153: step: 1176/527, loss: 0.3059563636779785
2023-01-21 10:27:45.642399: step: 1180/527, loss: 0.03283210098743439
2023-01-21 10:27:46.801379: step: 1184/527, loss: 0.033354759216308594
2023-01-21 10:27:47.923542: step: 1188/527, loss: 0.05663085728883743
2023-01-21 10:27:49.064447: step: 1192/527, loss: 0.024155616760253906
2023-01-21 10:27:50.230936: step: 1196/527, loss: 0.042307090014219284
2023-01-21 10:27:51.347874: step: 1200/527, loss: 0.04537970945239067
2023-01-21 10:27:52.506522: step: 1204/527, loss: 0.014900971204042435
2023-01-21 10:27:53.642557: step: 1208/527, loss: 0.016940975561738014
2023-01-21 10:27:54.789066: step: 1212/527, loss: 0.019524481147527695
2023-01-21 10:27:55.916075: step: 1216/527, loss: 0.0033661366906017065
2023-01-21 10:27:57.069259: step: 1220/527, loss: 0.024695778265595436
2023-01-21 10:27:58.216051: step: 1224/527, loss: 0.011055564507842064
2023-01-21 10:27:59.349755: step: 1228/527, loss: 0.09283050894737244
2023-01-21 10:28:00.457453: step: 1232/527, loss: 0.19912710785865784
2023-01-21 10:28:01.600310: step: 1236/527, loss: 0.04730741307139397
2023-01-21 10:28:02.732487: step: 1240/527, loss: 0.09515562653541565
2023-01-21 10:28:03.840861: step: 1244/527, loss: 0.09216327965259552
2023-01-21 10:28:04.968405: step: 1248/527, loss: 0.009247112087905407
2023-01-21 10:28:06.089276: step: 1252/527, loss: 0.056338027119636536
2023-01-21 10:28:07.262727: step: 1256/527, loss: 0.05807933956384659
2023-01-21 10:28:08.410170: step: 1260/527, loss: 0.012791823595762253
2023-01-21 10:28:09.525801: step: 1264/527, loss: 0.018391896039247513
2023-01-21 10:28:10.681288: step: 1268/527, loss: 0.03221721574664116
2023-01-21 10:28:11.836292: step: 1272/527, loss: 0.052113912999629974
2023-01-21 10:28:12.973074: step: 1276/527, loss: 0.055801428854465485
2023-01-21 10:28:14.086750: step: 1280/527, loss: 0.007173895835876465
2023-01-21 10:28:15.215972: step: 1284/527, loss: 0.04002103954553604
2023-01-21 10:28:16.308499: step: 1288/527, loss: 0.012015294283628464
2023-01-21 10:28:17.444311: step: 1292/527, loss: 0.032076645642519
2023-01-21 10:28:18.583414: step: 1296/527, loss: 0.22438450157642365
2023-01-21 10:28:19.746064: step: 1300/527, loss: 0.16272085905075073
2023-01-21 10:28:20.868134: step: 1304/527, loss: 0.02991161309182644
2023-01-21 10:28:22.001570: step: 1308/527, loss: 0.06720085442066193
2023-01-21 10:28:23.133084: step: 1312/527, loss: 0.623330295085907
2023-01-21 10:28:24.246472: step: 1316/527, loss: 0.0005444050184451044
2023-01-21 10:28:25.355564: step: 1320/527, loss: 0.020223617553710938
2023-01-21 10:28:26.513142: step: 1324/527, loss: 0.031185531988739967
2023-01-21 10:28:27.655730: step: 1328/527, loss: 0.022219086065888405
2023-01-21 10:28:28.802990: step: 1332/527, loss: 0.031058311462402344
2023-01-21 10:28:29.965088: step: 1336/527, loss: 0.4829002320766449
2023-01-21 10:28:31.105337: step: 1340/527, loss: 0.0422300361096859
2023-01-21 10:28:32.232852: step: 1344/527, loss: 0.01845226250588894
2023-01-21 10:28:33.358985: step: 1348/527, loss: 0.01965484581887722
2023-01-21 10:28:34.481573: step: 1352/527, loss: 0.013964558020234108
2023-01-21 10:28:35.601493: step: 1356/527, loss: 0.018819045275449753
2023-01-21 10:28:36.750097: step: 1360/527, loss: 0.07471399754285812
2023-01-21 10:28:37.888956: step: 1364/527, loss: 0.06723332405090332
2023-01-21 10:28:39.042285: step: 1368/527, loss: 0.12190437316894531
2023-01-21 10:28:40.142032: step: 1372/527, loss: 0.07430973649024963
2023-01-21 10:28:41.303458: step: 1376/527, loss: 0.10568561404943466
2023-01-21 10:28:42.412719: step: 1380/527, loss: 0.05476503074169159
2023-01-21 10:28:43.534565: step: 1384/527, loss: 0.04120798036456108
2023-01-21 10:28:44.672122: step: 1388/527, loss: 0.17752857506275177
2023-01-21 10:28:45.806262: step: 1392/527, loss: 0.09932174533605576
2023-01-21 10:28:46.929328: step: 1396/527, loss: 0.0008258819580078125
2023-01-21 10:28:48.073252: step: 1400/527, loss: 0.025541117414832115
2023-01-21 10:28:49.238474: step: 1404/527, loss: 0.011778187938034534
2023-01-21 10:28:50.382286: step: 1408/527, loss: 0.23266057670116425
2023-01-21 10:28:51.481418: step: 1412/527, loss: 0.02049694024026394
2023-01-21 10:28:52.592546: step: 1416/527, loss: 0.03534431755542755
2023-01-21 10:28:53.717712: step: 1420/527, loss: 0.0729069784283638
2023-01-21 10:28:54.838040: step: 1424/527, loss: 0.09496541321277618
2023-01-21 10:28:55.950253: step: 1428/527, loss: 0.017676448449492455
2023-01-21 10:28:57.086195: step: 1432/527, loss: 0.011694718152284622
2023-01-21 10:28:58.182856: step: 1436/527, loss: 0.02203083038330078
2023-01-21 10:28:59.336979: step: 1440/527, loss: 0.08038368076086044
2023-01-21 10:29:00.463393: step: 1444/527, loss: 0.07521028816699982
2023-01-21 10:29:01.597350: step: 1448/527, loss: 0.11142826080322266
2023-01-21 10:29:02.716289: step: 1452/527, loss: 0.016503525897860527
2023-01-21 10:29:03.917519: step: 1456/527, loss: 0.07093420624732971
2023-01-21 10:29:05.079153: step: 1460/527, loss: 0.009356880560517311
2023-01-21 10:29:06.221746: step: 1464/527, loss: 0.04651136323809624
2023-01-21 10:29:07.354949: step: 1468/527, loss: 0.046209149062633514
2023-01-21 10:29:08.465024: step: 1472/527, loss: 0.07755289226770401
2023-01-21 10:29:09.616007: step: 1476/527, loss: 0.0940558910369873
2023-01-21 10:29:10.783657: step: 1480/527, loss: 0.058814577758312225
2023-01-21 10:29:11.884824: step: 1484/527, loss: 0.03854780271649361
2023-01-21 10:29:13.009259: step: 1488/527, loss: 0.04230472818017006
2023-01-21 10:29:14.154303: step: 1492/527, loss: 0.026504946872591972
2023-01-21 10:29:15.282625: step: 1496/527, loss: 0.015051460824906826
2023-01-21 10:29:16.432953: step: 1500/527, loss: 0.08310193568468094
2023-01-21 10:29:17.589471: step: 1504/527, loss: 0.10364484786987305
2023-01-21 10:29:18.701070: step: 1508/527, loss: 0.0023200989235192537
2023-01-21 10:29:19.809050: step: 1512/527, loss: 0.030957406386733055
2023-01-21 10:29:20.945511: step: 1516/527, loss: 0.034165095537900925
2023-01-21 10:29:22.086402: step: 1520/527, loss: 0.571515679359436
2023-01-21 10:29:23.228876: step: 1524/527, loss: 0.08922509849071503
2023-01-21 10:29:24.364350: step: 1528/527, loss: 0.0038133144844323397
2023-01-21 10:29:25.506028: step: 1532/527, loss: 0.03225760906934738
2023-01-21 10:29:26.615196: step: 1536/527, loss: 0.014249228872358799
2023-01-21 10:29:27.730293: step: 1540/527, loss: 0.04622488096356392
2023-01-21 10:29:28.855867: step: 1544/527, loss: 0.12354183197021484
2023-01-21 10:29:29.974857: step: 1548/527, loss: 0.019068825989961624
2023-01-21 10:29:31.093111: step: 1552/527, loss: 0.09823817759752274
2023-01-21 10:29:32.209162: step: 1556/527, loss: 0.03968248516321182
2023-01-21 10:29:33.361555: step: 1560/527, loss: 0.16510942578315735
2023-01-21 10:29:34.484922: step: 1564/527, loss: 0.087456613779068
2023-01-21 10:29:35.608613: step: 1568/527, loss: 0.015306759625673294
2023-01-21 10:29:36.710848: step: 1572/527, loss: 0.04706630855798721
2023-01-21 10:29:37.848076: step: 1576/527, loss: 0.019093656912446022
2023-01-21 10:29:38.975924: step: 1580/527, loss: 0.11068210750818253
2023-01-21 10:29:40.088656: step: 1584/527, loss: 0.04693346470594406
2023-01-21 10:29:41.219706: step: 1588/527, loss: 0.03644957393407822
2023-01-21 10:29:42.353116: step: 1592/527, loss: 0.009153318591415882
2023-01-21 10:29:43.483588: step: 1596/527, loss: 0.015159226022660732
2023-01-21 10:29:44.637025: step: 1600/527, loss: 0.22866153717041016
2023-01-21 10:29:45.759699: step: 1604/527, loss: 0.01996641233563423
2023-01-21 10:29:46.896183: step: 1608/527, loss: 0.014081764966249466
2023-01-21 10:29:48.032757: step: 1612/527, loss: 0.02429504320025444
2023-01-21 10:29:49.179624: step: 1616/527, loss: 0.0254758819937706
2023-01-21 10:29:50.307020: step: 1620/527, loss: 0.00548896798864007
2023-01-21 10:29:51.417025: step: 1624/527, loss: 0.01192316971719265
2023-01-21 10:29:52.567095: step: 1628/527, loss: 0.029481984674930573
2023-01-21 10:29:53.708125: step: 1632/527, loss: 0.10827651619911194
2023-01-21 10:29:54.824537: step: 1636/527, loss: 0.03793678060173988
2023-01-21 10:29:55.957475: step: 1640/527, loss: 0.017445897683501244
2023-01-21 10:29:57.088548: step: 1644/527, loss: 0.13484449684619904
2023-01-21 10:29:58.239525: step: 1648/527, loss: 0.030888747423887253
2023-01-21 10:29:59.365300: step: 1652/527, loss: 0.014526368118822575
2023-01-21 10:30:00.496267: step: 1656/527, loss: 0.008530902676284313
2023-01-21 10:30:01.632302: step: 1660/527, loss: 0.05581941455602646
2023-01-21 10:30:02.768591: step: 1664/527, loss: 0.0075164795853197575
2023-01-21 10:30:03.958325: step: 1668/527, loss: 0.016440771520137787
2023-01-21 10:30:05.120293: step: 1672/527, loss: 0.06240687519311905
2023-01-21 10:30:06.252558: step: 1676/527, loss: 0.022716714069247246
2023-01-21 10:30:07.376560: step: 1680/527, loss: 0.0809703841805458
2023-01-21 10:30:08.518356: step: 1684/527, loss: 0.113915354013443
2023-01-21 10:30:09.646086: step: 1688/527, loss: 0.025805557146668434
2023-01-21 10:30:10.765951: step: 1692/527, loss: 0.014315415173768997
2023-01-21 10:30:11.898814: step: 1696/527, loss: 0.013329410925507545
2023-01-21 10:30:13.054712: step: 1700/527, loss: 0.0910153016448021
2023-01-21 10:30:14.192459: step: 1704/527, loss: 0.11901970207691193
2023-01-21 10:30:15.292143: step: 1708/527, loss: 0.013755107298493385
2023-01-21 10:30:16.423383: step: 1712/527, loss: 0.04601598158478737
2023-01-21 10:30:17.573203: step: 1716/527, loss: 0.07870922237634659
2023-01-21 10:30:18.675528: step: 1720/527, loss: 0.013075113296508789
2023-01-21 10:30:19.823758: step: 1724/527, loss: 0.035787202417850494
2023-01-21 10:30:20.946866: step: 1728/527, loss: 0.04321441799402237
2023-01-21 10:30:22.120618: step: 1732/527, loss: 0.08494377136230469
2023-01-21 10:30:23.241147: step: 1736/527, loss: 0.04179525375366211
2023-01-21 10:30:24.367401: step: 1740/527, loss: 0.05915093421936035
2023-01-21 10:30:25.471931: step: 1744/527, loss: 0.023685265332460403
2023-01-21 10:30:26.646893: step: 1748/527, loss: 0.031377315521240234
2023-01-21 10:30:27.760088: step: 1752/527, loss: 0.005942630581557751
2023-01-21 10:30:28.886591: step: 1756/527, loss: 0.04503745958209038
2023-01-21 10:30:30.019620: step: 1760/527, loss: 0.04873838648200035
2023-01-21 10:30:31.143780: step: 1764/527, loss: 0.007004499435424805
2023-01-21 10:30:32.262019: step: 1768/527, loss: 0.04896259307861328
2023-01-21 10:30:33.382463: step: 1772/527, loss: 0.2488558292388916
2023-01-21 10:30:34.490177: step: 1776/527, loss: 0.12036170810461044
2023-01-21 10:30:35.624322: step: 1780/527, loss: 0.06050090864300728
2023-01-21 10:30:36.736205: step: 1784/527, loss: 0.07888466864824295
2023-01-21 10:30:37.860527: step: 1788/527, loss: 0.050199463963508606
2023-01-21 10:30:38.984864: step: 1792/527, loss: 0.011244607158005238
2023-01-21 10:30:40.121323: step: 1796/527, loss: 0.049538515508174896
2023-01-21 10:30:41.215945: step: 1800/527, loss: 0.09805078059434891
2023-01-21 10:30:42.377676: step: 1804/527, loss: 0.04809122160077095
2023-01-21 10:30:43.504402: step: 1808/527, loss: 0.035685352981090546
2023-01-21 10:30:44.661030: step: 1812/527, loss: 0.05141716077923775
2023-01-21 10:30:45.809510: step: 1816/527, loss: 0.21552257239818573
2023-01-21 10:30:46.947901: step: 1820/527, loss: 0.06490226089954376
2023-01-21 10:30:48.087422: step: 1824/527, loss: 0.06039810553193092
2023-01-21 10:30:49.225144: step: 1828/527, loss: 0.021173859015107155
2023-01-21 10:30:50.338219: step: 1832/527, loss: 0.06303635239601135
2023-01-21 10:30:51.457020: step: 1836/527, loss: 0.010195828042924404
2023-01-21 10:30:52.601591: step: 1840/527, loss: 0.00405507069081068
2023-01-21 10:30:53.777039: step: 1844/527, loss: 0.005920981988310814
2023-01-21 10:30:54.934715: step: 1848/527, loss: 0.36158448457717896
2023-01-21 10:30:56.084607: step: 1852/527, loss: 0.0351228229701519
2023-01-21 10:30:57.216637: step: 1856/527, loss: 0.09659843146800995
2023-01-21 10:30:58.347531: step: 1860/527, loss: 0.027343153953552246
2023-01-21 10:30:59.473474: step: 1864/527, loss: 0.04684047773480415
2023-01-21 10:31:00.607628: step: 1868/527, loss: 0.10664357990026474
2023-01-21 10:31:01.699818: step: 1872/527, loss: 4.849433753406629e-05
2023-01-21 10:31:02.830188: step: 1876/527, loss: 0.08119240403175354
2023-01-21 10:31:03.971136: step: 1880/527, loss: 0.11827459931373596
2023-01-21 10:31:05.104048: step: 1884/527, loss: 0.06516905128955841
2023-01-21 10:31:06.239862: step: 1888/527, loss: 0.2907160818576813
2023-01-21 10:31:07.349123: step: 1892/527, loss: 0.013770675286650658
2023-01-21 10:31:08.496790: step: 1896/527, loss: 0.4469691216945648
2023-01-21 10:31:09.597909: step: 1900/527, loss: 0.00012354851060081273
2023-01-21 10:31:10.680333: step: 1904/527, loss: 0.00045781134394928813
2023-01-21 10:31:11.799323: step: 1908/527, loss: 0.09767188876867294
2023-01-21 10:31:12.944231: step: 1912/527, loss: 0.10007724910974503
2023-01-21 10:31:14.122339: step: 1916/527, loss: 0.2843441963195801
2023-01-21 10:31:15.273687: step: 1920/527, loss: 0.007694816682487726
2023-01-21 10:31:16.396824: step: 1924/527, loss: 0.08538040518760681
2023-01-21 10:31:17.568799: step: 1928/527, loss: 0.05739450454711914
2023-01-21 10:31:18.677746: step: 1932/527, loss: 0.05072937160730362
2023-01-21 10:31:19.842587: step: 1936/527, loss: 0.009163284674286842
2023-01-21 10:31:21.000414: step: 1940/527, loss: 0.10809822380542755
2023-01-21 10:31:22.152023: step: 1944/527, loss: 0.042459968477487564
2023-01-21 10:31:23.289897: step: 1948/527, loss: 0.01316680945456028
2023-01-21 10:31:24.408340: step: 1952/527, loss: 0.004237270448356867
2023-01-21 10:31:25.521964: step: 1956/527, loss: 0.046647265553474426
2023-01-21 10:31:26.685765: step: 1960/527, loss: 0.10077590495347977
2023-01-21 10:31:27.811418: step: 1964/527, loss: 0.10086269676685333
2023-01-21 10:31:28.951151: step: 1968/527, loss: 0.004143047612160444
2023-01-21 10:31:30.070254: step: 1972/527, loss: 0.14455413818359375
2023-01-21 10:31:31.216084: step: 1976/527, loss: 0.10469808429479599
2023-01-21 10:31:32.373255: step: 1980/527, loss: 0.1698492020368576
2023-01-21 10:31:33.513890: step: 1984/527, loss: 0.020453739911317825
2023-01-21 10:31:34.685578: step: 1988/527, loss: 0.0641474723815918
2023-01-21 10:31:35.797602: step: 1992/527, loss: 0.03961925581097603
2023-01-21 10:31:36.909683: step: 1996/527, loss: 0.04423222318291664
2023-01-21 10:31:38.045322: step: 2000/527, loss: 0.15131813287734985
2023-01-21 10:31:39.165880: step: 2004/527, loss: 0.004822921939194202
2023-01-21 10:31:40.276689: step: 2008/527, loss: 0.03845958784222603
2023-01-21 10:31:41.415503: step: 2012/527, loss: 0.09592628479003906
2023-01-21 10:31:42.550523: step: 2016/527, loss: 0.083295539021492
2023-01-21 10:31:43.674225: step: 2020/527, loss: 0.03820948675274849
2023-01-21 10:31:44.837233: step: 2024/527, loss: 0.08798827975988388
2023-01-21 10:31:45.977036: step: 2028/527, loss: 0.13839396834373474
2023-01-21 10:31:47.117729: step: 2032/527, loss: 0.0405183807015419
2023-01-21 10:31:48.240170: step: 2036/527, loss: 0.046586133539676666
2023-01-21 10:31:49.363353: step: 2040/527, loss: 0.01942768134176731
2023-01-21 10:31:50.513512: step: 2044/527, loss: 0.015798378735780716
2023-01-21 10:31:51.668598: step: 2048/527, loss: 0.009473991580307484
2023-01-21 10:31:52.790704: step: 2052/527, loss: 0.0479372963309288
2023-01-21 10:31:53.924203: step: 2056/527, loss: 0.13619175553321838
2023-01-21 10:31:55.037668: step: 2060/527, loss: 0.3218957781791687
2023-01-21 10:31:56.184843: step: 2064/527, loss: 0.08143134415149689
2023-01-21 10:31:57.298270: step: 2068/527, loss: 0.037137579172849655
2023-01-21 10:31:58.472647: step: 2072/527, loss: 0.04535358399152756
2023-01-21 10:31:59.597580: step: 2076/527, loss: 0.08916588127613068
2023-01-21 10:32:00.745960: step: 2080/527, loss: 0.09756284207105637
2023-01-21 10:32:01.878135: step: 2084/527, loss: 0.11427002400159836
2023-01-21 10:32:02.999616: step: 2088/527, loss: 0.08184423297643661
2023-01-21 10:32:04.124818: step: 2092/527, loss: 0.009721899405121803
2023-01-21 10:32:05.253790: step: 2096/527, loss: 0.06943559646606445
2023-01-21 10:32:06.365912: step: 2100/527, loss: 0.06615275889635086
2023-01-21 10:32:07.514106: step: 2104/527, loss: 0.05557527393102646
2023-01-21 10:32:08.636454: step: 2108/527, loss: 0.10674263536930084
==================================================
Loss: 0.072
--------------------
Dev: {'event': {'p': 0.6122004357298475, 'r': 0.748335552596538, 'f1': 0.6734571599760335}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11}
Test: {'event': {'p': 0.652636671504596, 'r': 0.7708571428571429, 'f1': 0.706837830757139}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11}
Chinese: {'event': {'p': 0.5769230769230769, 'r': 0.8333333333333334, 'f1': 0.6818181818181818}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11}
Korean: {'event': {'p': 0.603448275862069, 'r': 0.5555555555555556, 'f1': 0.5785123966942148}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11}
Russian: {'event': {'p': 0.5833333333333334, 'r': 0.5833333333333334, 'f1': 0.5833333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:32:51.915411: step: 4/527, loss: 0.0090796472504735
2023-01-21 10:32:53.042255: step: 8/527, loss: 1.1015418767929077
2023-01-21 10:32:54.170189: step: 12/527, loss: 0.038718223571777344
2023-01-21 10:32:55.295898: step: 16/527, loss: 0.014827347360551357
2023-01-21 10:32:56.409694: step: 20/527, loss: 0.04376039654016495
2023-01-21 10:32:57.902898: step: 24/527, loss: 0.22693099081516266
2023-01-21 10:32:59.024071: step: 28/527, loss: 0.08030052483081818
2023-01-21 10:33:00.163275: step: 32/527, loss: 0.08119454979896545
2023-01-21 10:33:01.288546: step: 36/527, loss: 0.007630682084709406
2023-01-21 10:33:02.402727: step: 40/527, loss: 0.024507999420166016
2023-01-21 10:33:03.553272: step: 44/527, loss: 0.04251156002283096
2023-01-21 10:33:04.673325: step: 48/527, loss: 0.0012950897216796875
2023-01-21 10:33:05.821676: step: 52/527, loss: 0.009382450953125954
2023-01-21 10:33:06.974011: step: 56/527, loss: 0.07246322929859161
2023-01-21 10:33:08.078134: step: 60/527, loss: 0.1203889399766922
2023-01-21 10:33:09.200311: step: 64/527, loss: 0.050376035273075104
2023-01-21 10:33:10.336160: step: 68/527, loss: 0.07015237957239151
2023-01-21 10:33:11.486049: step: 72/527, loss: 0.029198456555604935
2023-01-21 10:33:12.631571: step: 76/527, loss: 0.03815422207117081
2023-01-21 10:33:13.725941: step: 80/527, loss: 0.06336124241352081
2023-01-21 10:33:14.846397: step: 84/527, loss: 0.06620607525110245
2023-01-21 10:33:15.985348: step: 88/527, loss: 0.023171139881014824
2023-01-21 10:33:17.127913: step: 92/527, loss: 0.1522786170244217
2023-01-21 10:33:18.309433: step: 96/527, loss: 0.025907421484589577
2023-01-21 10:33:19.466405: step: 100/527, loss: 0.020841384306550026
2023-01-21 10:33:20.581751: step: 104/527, loss: 0.019943714141845703
2023-01-21 10:33:21.702244: step: 108/527, loss: 0.05508747324347496
2023-01-21 10:33:22.828662: step: 112/527, loss: 0.01639232598245144
2023-01-21 10:33:23.962177: step: 116/527, loss: 0.006540966220200062
2023-01-21 10:33:25.084031: step: 120/527, loss: 0.006244754884392023
2023-01-21 10:33:26.225659: step: 124/527, loss: 0.3628656268119812
2023-01-21 10:33:27.354186: step: 128/527, loss: 0.007829857058823109
2023-01-21 10:33:28.569331: step: 132/527, loss: 0.015318060293793678
2023-01-21 10:33:29.701956: step: 136/527, loss: 0.0073646544478833675
2023-01-21 10:33:30.849682: step: 140/527, loss: 0.04042835161089897
2023-01-21 10:33:31.983726: step: 144/527, loss: 0.04568615183234215
2023-01-21 10:33:33.079209: step: 148/527, loss: 0.004329228773713112
2023-01-21 10:33:34.212064: step: 152/527, loss: 0.03024768829345703
2023-01-21 10:33:35.367965: step: 156/527, loss: 0.07341212779283524
2023-01-21 10:33:36.496223: step: 160/527, loss: 0.034290507435798645
2023-01-21 10:33:37.618268: step: 164/527, loss: 0.0008838654030114412
2023-01-21 10:33:38.734650: step: 168/527, loss: 0.024438858032226562
2023-01-21 10:33:39.857974: step: 172/527, loss: 0.017620373517274857
2023-01-21 10:33:40.977066: step: 176/527, loss: 0.0392850898206234
2023-01-21 10:33:42.115867: step: 180/527, loss: 0.0028553963638842106
2023-01-21 10:33:43.251867: step: 184/527, loss: 0.0015016555553302169
2023-01-21 10:33:44.405666: step: 188/527, loss: 0.016286849975585938
2023-01-21 10:33:45.522504: step: 192/527, loss: 0.16449928283691406
2023-01-21 10:33:46.663934: step: 196/527, loss: 0.023264314979314804
2023-01-21 10:33:47.786054: step: 200/527, loss: 0.0547269843518734
2023-01-21 10:33:48.954278: step: 204/527, loss: 0.014321994967758656
2023-01-21 10:33:50.095813: step: 208/527, loss: 0.04382476955652237
2023-01-21 10:33:51.236531: step: 212/527, loss: 0.045938681811094284
2023-01-21 10:33:52.401452: step: 216/527, loss: 0.019191646948456764
2023-01-21 10:33:53.561718: step: 220/527, loss: 0.045766159892082214
2023-01-21 10:33:54.684722: step: 224/527, loss: 0.20786495506763458
2023-01-21 10:33:55.813166: step: 228/527, loss: 0.326227605342865
2023-01-21 10:33:56.934781: step: 232/527, loss: 0.026268385350704193
2023-01-21 10:33:58.053656: step: 236/527, loss: 0.015949726104736328
2023-01-21 10:33:59.187048: step: 240/527, loss: 0.03268437460064888
2023-01-21 10:34:00.320364: step: 244/527, loss: 0.01395649928599596
2023-01-21 10:34:01.431334: step: 248/527, loss: 0.04610037803649902
2023-01-21 10:34:02.582042: step: 252/527, loss: 0.047167327255010605
2023-01-21 10:34:03.712070: step: 256/527, loss: 0.34132567048072815
2023-01-21 10:34:04.862487: step: 260/527, loss: 0.06189899146556854
2023-01-21 10:34:05.990686: step: 264/527, loss: 0.0043406011536717415
2023-01-21 10:34:07.099899: step: 268/527, loss: 0.007656860630959272
2023-01-21 10:34:08.240357: step: 272/527, loss: 0.07890200614929199
2023-01-21 10:34:09.391441: step: 276/527, loss: 0.015041542239487171
2023-01-21 10:34:10.490473: step: 280/527, loss: 0.018963146954774857
2023-01-21 10:34:11.660790: step: 284/527, loss: 0.06734047830104828
2023-01-21 10:34:12.789173: step: 288/527, loss: 0.0008373260498046875
2023-01-21 10:34:13.905805: step: 292/527, loss: 0.03211965784430504
2023-01-21 10:34:15.030319: step: 296/527, loss: 0.04786090925335884
2023-01-21 10:34:16.180718: step: 300/527, loss: 0.06224679946899414
2023-01-21 10:34:17.326270: step: 304/527, loss: 0.0392850898206234
2023-01-21 10:34:18.454604: step: 308/527, loss: 0.012391472235321999
2023-01-21 10:34:19.579658: step: 312/527, loss: 0.015016973949968815
2023-01-21 10:34:20.714708: step: 316/527, loss: 0.04218854755163193
2023-01-21 10:34:21.828704: step: 320/527, loss: 0.06763195991516113
2023-01-21 10:34:22.955620: step: 324/527, loss: 0.04276075214147568
2023-01-21 10:34:24.104520: step: 328/527, loss: 0.06996660679578781
2023-01-21 10:34:25.231563: step: 332/527, loss: 0.004106283187866211
2023-01-21 10:34:26.379711: step: 336/527, loss: 0.004105043597519398
2023-01-21 10:34:27.505160: step: 340/527, loss: 0.10584679245948792
2023-01-21 10:34:28.634013: step: 344/527, loss: 0.6630697250366211
2023-01-21 10:34:29.776887: step: 348/527, loss: 0.06550198048353195
2023-01-21 10:34:30.920633: step: 352/527, loss: 0.013426780700683594
2023-01-21 10:34:32.024920: step: 356/527, loss: 0.047193121165037155
2023-01-21 10:34:33.157389: step: 360/527, loss: 0.015097999945282936
2023-01-21 10:34:34.284231: step: 364/527, loss: 0.010131550021469593
2023-01-21 10:34:35.398398: step: 368/527, loss: 0.059766773134469986
2023-01-21 10:34:36.523191: step: 372/527, loss: 0.04051628336310387
2023-01-21 10:34:37.660193: step: 376/527, loss: 0.11368946731090546
2023-01-21 10:34:38.825807: step: 380/527, loss: 0.010639620013535023
2023-01-21 10:34:39.973272: step: 384/527, loss: 0.05690011754631996
2023-01-21 10:34:41.110294: step: 388/527, loss: 0.02503385581076145
2023-01-21 10:34:42.239300: step: 392/527, loss: 0.030107783153653145
2023-01-21 10:34:43.385078: step: 396/527, loss: 0.0432429313659668
2023-01-21 10:34:44.544223: step: 400/527, loss: 0.062355659902095795
2023-01-21 10:34:45.670858: step: 404/527, loss: 0.004106044769287109
2023-01-21 10:34:46.830154: step: 408/527, loss: 0.0394282341003418
2023-01-21 10:34:47.930357: step: 412/527, loss: 0.016356086358428
2023-01-21 10:34:49.053911: step: 416/527, loss: 0.02437877655029297
2023-01-21 10:34:50.194779: step: 420/527, loss: 0.06344123184680939
2023-01-21 10:34:51.326382: step: 424/527, loss: 0.008895492181181908
2023-01-21 10:34:52.432797: step: 428/527, loss: 0.018272925168275833
2023-01-21 10:34:53.544185: step: 432/527, loss: 0.010533380322158337
2023-01-21 10:34:54.648694: step: 436/527, loss: 0.006784630008041859
2023-01-21 10:34:55.774066: step: 440/527, loss: 0.06552910804748535
2023-01-21 10:34:56.911231: step: 444/527, loss: 0.030863476917147636
2023-01-21 10:34:58.052876: step: 448/527, loss: 0.0025686263106763363
2023-01-21 10:34:59.200911: step: 452/527, loss: 0.025992775335907936
2023-01-21 10:35:00.349117: step: 456/527, loss: 0.06317071616649628
2023-01-21 10:35:01.489625: step: 460/527, loss: 0.03235797956585884
2023-01-21 10:35:02.641616: step: 464/527, loss: 0.004155921749770641
2023-01-21 10:35:03.779490: step: 468/527, loss: 0.007954597473144531
2023-01-21 10:35:04.905013: step: 472/527, loss: 0.005334401037544012
2023-01-21 10:35:06.035915: step: 476/527, loss: 0.015418529510498047
2023-01-21 10:35:07.213500: step: 480/527, loss: 0.05910225212574005
2023-01-21 10:35:08.341297: step: 484/527, loss: 0.03519077226519585
2023-01-21 10:35:09.428880: step: 488/527, loss: 0.009540462866425514
2023-01-21 10:35:10.601917: step: 492/527, loss: 0.03551778569817543
2023-01-21 10:35:11.724241: step: 496/527, loss: 0.010079383850097656
2023-01-21 10:35:12.856390: step: 500/527, loss: 0.03679781034588814
2023-01-21 10:35:13.995748: step: 504/527, loss: 0.10027079284191132
2023-01-21 10:35:15.134761: step: 508/527, loss: 0.05423159524798393
2023-01-21 10:35:16.263509: step: 512/527, loss: 0.11605750024318695
2023-01-21 10:35:17.378759: step: 516/527, loss: 0.03766822814941406
2023-01-21 10:35:18.492833: step: 520/527, loss: 0.01925363577902317
2023-01-21 10:35:19.633331: step: 524/527, loss: 0.037119291722774506
2023-01-21 10:35:20.758521: step: 528/527, loss: 0.0038480341900140047
2023-01-21 10:35:21.878992: step: 532/527, loss: 0.11628389358520508
2023-01-21 10:35:23.033784: step: 536/527, loss: 0.01611914671957493
2023-01-21 10:35:24.171655: step: 540/527, loss: 0.012683678418397903
2023-01-21 10:35:25.281623: step: 544/527, loss: 0.09335164725780487
2023-01-21 10:35:26.394211: step: 548/527, loss: 0.001932764076627791
2023-01-21 10:35:27.508537: step: 552/527, loss: 0.007652664091438055
2023-01-21 10:35:28.641735: step: 556/527, loss: 0.009312629699707031
2023-01-21 10:35:29.764204: step: 560/527, loss: 0.033277321606874466
2023-01-21 10:35:30.869819: step: 564/527, loss: 0.03681612014770508
2023-01-21 10:35:32.031305: step: 568/527, loss: 0.14951543509960175
2023-01-21 10:35:33.155722: step: 572/527, loss: 0.03504066541790962
2023-01-21 10:35:34.267688: step: 576/527, loss: 0.032768476754426956
2023-01-21 10:35:35.385473: step: 580/527, loss: 0.02105121500790119
2023-01-21 10:35:36.520962: step: 584/527, loss: 0.01645069196820259
2023-01-21 10:35:37.673580: step: 588/527, loss: 0.01897439919412136
2023-01-21 10:35:38.795335: step: 592/527, loss: 0.07848014682531357
2023-01-21 10:35:39.913281: step: 596/527, loss: 0.03618254512548447
2023-01-21 10:35:41.034916: step: 600/527, loss: 0.004900741390883923
2023-01-21 10:35:42.194939: step: 604/527, loss: 0.015542412176728249
2023-01-21 10:35:43.340841: step: 608/527, loss: 0.028797341510653496
2023-01-21 10:35:44.445688: step: 612/527, loss: 0.1684170812368393
2023-01-21 10:35:45.570564: step: 616/527, loss: 0.13710841536521912
2023-01-21 10:35:46.708112: step: 620/527, loss: 0.08882565796375275
2023-01-21 10:35:47.860291: step: 624/527, loss: 0.03710470348596573
2023-01-21 10:35:49.009861: step: 628/527, loss: 0.039847325533628464
2023-01-21 10:35:50.180363: step: 632/527, loss: 0.06346473842859268
2023-01-21 10:35:51.316702: step: 636/527, loss: 0.1964806616306305
2023-01-21 10:35:52.410494: step: 640/527, loss: 0.025510216131806374
2023-01-21 10:35:53.583511: step: 644/527, loss: 0.2508466839790344
2023-01-21 10:35:54.693192: step: 648/527, loss: 0.019208097830414772
2023-01-21 10:35:55.861223: step: 652/527, loss: 0.02165374904870987
2023-01-21 10:35:57.011286: step: 656/527, loss: 0.024995042011141777
2023-01-21 10:35:58.184620: step: 660/527, loss: 0.0688634142279625
2023-01-21 10:35:59.312667: step: 664/527, loss: 0.06251154094934464
2023-01-21 10:36:00.434850: step: 668/527, loss: 0.002140820026397705
2023-01-21 10:36:01.580629: step: 672/527, loss: 0.568588376045227
2023-01-21 10:36:02.725686: step: 676/527, loss: 0.012835502624511719
2023-01-21 10:36:03.861389: step: 680/527, loss: 0.01817493513226509
2023-01-21 10:36:05.039284: step: 684/527, loss: 0.009250259958207607
2023-01-21 10:36:06.157759: step: 688/527, loss: 0.00530743645504117
2023-01-21 10:36:07.312393: step: 692/527, loss: 0.01853466033935547
2023-01-21 10:36:08.429960: step: 696/527, loss: 0.020940016955137253
2023-01-21 10:36:09.571373: step: 700/527, loss: 0.033652354031801224
2023-01-21 10:36:10.714996: step: 704/527, loss: 0.014206647872924805
2023-01-21 10:36:11.851274: step: 708/527, loss: 0.01654529571533203
2023-01-21 10:36:12.989687: step: 712/527, loss: 0.06393428146839142
2023-01-21 10:36:14.125332: step: 716/527, loss: 0.04209652170538902
2023-01-21 10:36:15.277754: step: 720/527, loss: 0.008309269323945045
2023-01-21 10:36:16.418031: step: 724/527, loss: 0.1493232101202011
2023-01-21 10:36:17.536804: step: 728/527, loss: 0.08231983333826065
2023-01-21 10:36:18.670718: step: 732/527, loss: 0.006633186247199774
2023-01-21 10:36:19.831410: step: 736/527, loss: 0.04080267250537872
2023-01-21 10:36:20.947810: step: 740/527, loss: 0.0752757117152214
2023-01-21 10:36:22.087900: step: 744/527, loss: 0.05169324949383736
2023-01-21 10:36:23.219670: step: 748/527, loss: 0.19992037117481232
2023-01-21 10:36:24.357718: step: 752/527, loss: 0.00885772705078125
2023-01-21 10:36:25.467605: step: 756/527, loss: 0.04186449199914932
2023-01-21 10:36:26.580048: step: 760/527, loss: 0.0403934046626091
2023-01-21 10:36:27.708809: step: 764/527, loss: 0.07358036190271378
2023-01-21 10:36:28.803294: step: 768/527, loss: 0.060792163014411926
2023-01-21 10:36:29.945512: step: 772/527, loss: 0.1698535829782486
2023-01-21 10:36:31.089232: step: 776/527, loss: 0.06591920554637909
2023-01-21 10:36:32.204607: step: 780/527, loss: 0.03862934187054634
2023-01-21 10:36:33.308292: step: 784/527, loss: 0.03184995800256729
2023-01-21 10:36:34.462790: step: 788/527, loss: 0.17610104382038116
2023-01-21 10:36:35.587892: step: 792/527, loss: 0.0256805419921875
2023-01-21 10:36:36.754968: step: 796/527, loss: 0.05252866819500923
2023-01-21 10:36:37.915256: step: 800/527, loss: 0.06755819916725159
2023-01-21 10:36:39.032414: step: 804/527, loss: 0.028908347710967064
2023-01-21 10:36:40.199690: step: 808/527, loss: 0.05783233791589737
2023-01-21 10:36:41.351184: step: 812/527, loss: 0.005163860507309437
2023-01-21 10:36:42.478904: step: 816/527, loss: 0.047847796231508255
2023-01-21 10:36:43.581121: step: 820/527, loss: 0.3479452133178711
2023-01-21 10:36:44.737908: step: 824/527, loss: 0.03642444685101509
2023-01-21 10:36:45.868594: step: 828/527, loss: 0.03324108198285103
2023-01-21 10:36:46.988876: step: 832/527, loss: 0.06421055644750595
2023-01-21 10:36:48.110589: step: 836/527, loss: 0.01081104390323162
2023-01-21 10:36:49.255277: step: 840/527, loss: 0.012990022078156471
2023-01-21 10:36:50.400049: step: 844/527, loss: 0.02893071249127388
2023-01-21 10:36:51.567111: step: 848/527, loss: 0.2745036482810974
2023-01-21 10:36:52.708005: step: 852/527, loss: 0.06306762993335724
2023-01-21 10:36:53.824408: step: 856/527, loss: 0.05304155498743057
2023-01-21 10:36:54.962853: step: 860/527, loss: 0.009493160992860794
2023-01-21 10:36:56.111331: step: 864/527, loss: 0.05348839983344078
2023-01-21 10:36:57.229619: step: 868/527, loss: 0.015718460083007812
2023-01-21 10:36:58.352916: step: 872/527, loss: 0.04415760189294815
2023-01-21 10:36:59.493495: step: 876/527, loss: 0.030023908242583275
2023-01-21 10:37:00.634568: step: 880/527, loss: 0.01096811331808567
2023-01-21 10:37:01.754511: step: 884/527, loss: 0.0032626152969896793
2023-01-21 10:37:02.895134: step: 888/527, loss: 0.06071672588586807
2023-01-21 10:37:04.027337: step: 892/527, loss: 0.1628277748823166
2023-01-21 10:37:05.128155: step: 896/527, loss: 0.0060142045840620995
2023-01-21 10:37:06.285254: step: 900/527, loss: 0.04828405752778053
2023-01-21 10:37:07.437345: step: 904/527, loss: 0.030677415430545807
2023-01-21 10:37:08.576236: step: 908/527, loss: 0.05100831016898155
2023-01-21 10:37:09.713800: step: 912/527, loss: 0.007142952177673578
2023-01-21 10:37:10.837991: step: 916/527, loss: 0.015167188830673695
2023-01-21 10:37:11.987044: step: 920/527, loss: 0.028451919555664062
2023-01-21 10:37:13.139690: step: 924/527, loss: 0.01021499652415514
2023-01-21 10:37:14.249162: step: 928/527, loss: 0.06605949252843857
2023-01-21 10:37:15.369035: step: 932/527, loss: 0.0361669547855854
2023-01-21 10:37:16.488000: step: 936/527, loss: 0.0746254026889801
2023-01-21 10:37:17.639761: step: 940/527, loss: 0.08664393424987793
2023-01-21 10:37:18.745039: step: 944/527, loss: 0.04396582022309303
2023-01-21 10:37:19.871772: step: 948/527, loss: 0.09045977890491486
2023-01-21 10:37:21.014220: step: 952/527, loss: 0.024817848578095436
2023-01-21 10:37:22.123453: step: 956/527, loss: 0.39009198546409607
2023-01-21 10:37:23.213650: step: 960/527, loss: 0.046346187591552734
2023-01-21 10:37:24.346491: step: 964/527, loss: 0.05146045982837677
2023-01-21 10:37:25.473699: step: 968/527, loss: 0.040400173515081406
2023-01-21 10:37:26.610602: step: 972/527, loss: 0.020724773406982422
2023-01-21 10:37:27.746715: step: 976/527, loss: 0.014262771233916283
2023-01-21 10:37:28.872054: step: 980/527, loss: 0.00963220652192831
2023-01-21 10:37:29.998814: step: 984/527, loss: 0.0022412778344005346
2023-01-21 10:37:31.149150: step: 988/527, loss: 0.030391409993171692
2023-01-21 10:37:32.303060: step: 992/527, loss: 0.07612724602222443
2023-01-21 10:37:33.446112: step: 996/527, loss: 0.02101168781518936
2023-01-21 10:37:34.626154: step: 1000/527, loss: 0.0004650593036785722
2023-01-21 10:37:35.778408: step: 1004/527, loss: 0.1222929060459137
2023-01-21 10:37:36.931671: step: 1008/527, loss: 0.016438627615571022
2023-01-21 10:37:38.086927: step: 1012/527, loss: 0.04703950881958008
2023-01-21 10:37:39.230798: step: 1016/527, loss: 0.23692750930786133
2023-01-21 10:37:40.336020: step: 1020/527, loss: 0.03495221212506294
2023-01-21 10:37:41.453532: step: 1024/527, loss: 0.04385042190551758
2023-01-21 10:37:42.581620: step: 1028/527, loss: 0.06741132587194443
2023-01-21 10:37:43.721382: step: 1032/527, loss: 0.014417458325624466
2023-01-21 10:37:44.860853: step: 1036/527, loss: 0.0696842223405838
2023-01-21 10:37:45.971137: step: 1040/527, loss: 0.0762234702706337
2023-01-21 10:37:47.094988: step: 1044/527, loss: 0.026877976953983307
2023-01-21 10:37:48.204248: step: 1048/527, loss: 0.5681314468383789
2023-01-21 10:37:49.311165: step: 1052/527, loss: 0.027744673192501068
2023-01-21 10:37:50.449137: step: 1056/527, loss: 0.033165834844112396
2023-01-21 10:37:51.597619: step: 1060/527, loss: 0.04122314229607582
2023-01-21 10:37:52.730343: step: 1064/527, loss: 0.11802501231431961
2023-01-21 10:37:53.851826: step: 1068/527, loss: 0.058196358382701874
2023-01-21 10:37:54.989235: step: 1072/527, loss: 0.03922281414270401
2023-01-21 10:37:56.134956: step: 1076/527, loss: 0.07330923527479172
2023-01-21 10:37:57.268658: step: 1080/527, loss: 0.07492885738611221
2023-01-21 10:37:58.388303: step: 1084/527, loss: 0.048386767506599426
2023-01-21 10:37:59.508617: step: 1088/527, loss: 0.0008638381841592491
2023-01-21 10:38:00.681339: step: 1092/527, loss: 0.01000371016561985
2023-01-21 10:38:01.823805: step: 1096/527, loss: 0.060792066156864166
2023-01-21 10:38:03.008848: step: 1100/527, loss: 0.08706007152795792
2023-01-21 10:38:04.158855: step: 1104/527, loss: 0.05498151481151581
2023-01-21 10:38:05.263122: step: 1108/527, loss: 0.04267864301800728
2023-01-21 10:38:06.398948: step: 1112/527, loss: 0.07260684669017792
2023-01-21 10:38:07.521235: step: 1116/527, loss: 0.07278890907764435
2023-01-21 10:38:08.664342: step: 1120/527, loss: 0.9936366677284241
2023-01-21 10:38:09.779847: step: 1124/527, loss: 0.0083169462159276
2023-01-21 10:38:10.934957: step: 1128/527, loss: 0.02300419844686985
2023-01-21 10:38:12.047759: step: 1132/527, loss: 0.08628582954406738
2023-01-21 10:38:13.222285: step: 1136/527, loss: 0.11872711032629013
2023-01-21 10:38:14.310116: step: 1140/527, loss: 0.004105567932128906
2023-01-21 10:38:15.448077: step: 1144/527, loss: 0.011553764343261719
2023-01-21 10:38:16.595727: step: 1148/527, loss: 0.12733431160449982
2023-01-21 10:38:17.717606: step: 1152/527, loss: 0.0121329789981246
2023-01-21 10:38:18.836511: step: 1156/527, loss: 0.07633152604103088
2023-01-21 10:38:19.954983: step: 1160/527, loss: 0.01623370125889778
2023-01-21 10:38:21.079772: step: 1164/527, loss: 0.013297795318067074
2023-01-21 10:38:22.206871: step: 1168/527, loss: 0.019066954031586647
2023-01-21 10:38:23.323186: step: 1172/527, loss: 0.033713530749082565
2023-01-21 10:38:24.460886: step: 1176/527, loss: 0.08863174170255661
2023-01-21 10:38:25.629840: step: 1180/527, loss: 0.059471987187862396
2023-01-21 10:38:26.810998: step: 1184/527, loss: 0.038346003741025925
2023-01-21 10:38:27.943715: step: 1188/527, loss: 0.8626817464828491
2023-01-21 10:38:29.077718: step: 1192/527, loss: 0.014910697937011719
2023-01-21 10:38:30.202497: step: 1196/527, loss: 0.008986949920654297
2023-01-21 10:38:31.348771: step: 1200/527, loss: 0.02667694166302681
2023-01-21 10:38:32.471859: step: 1204/527, loss: 0.01373071689158678
2023-01-21 10:38:33.629522: step: 1208/527, loss: 0.0757112056016922
2023-01-21 10:38:34.756559: step: 1212/527, loss: 0.6679500937461853
2023-01-21 10:38:35.898817: step: 1216/527, loss: 0.06474065780639648
2023-01-21 10:38:37.011608: step: 1220/527, loss: 0.005532789509743452
2023-01-21 10:38:38.147960: step: 1224/527, loss: 0.002811431884765625
2023-01-21 10:38:39.285681: step: 1228/527, loss: 0.04421323910355568
2023-01-21 10:38:40.437501: step: 1232/527, loss: 0.01188964769244194
2023-01-21 10:38:41.567445: step: 1236/527, loss: 0.029691221192479134
2023-01-21 10:38:42.676612: step: 1240/527, loss: 0.0054779052734375
2023-01-21 10:38:43.868245: step: 1244/527, loss: 0.9309937357902527
2023-01-21 10:38:44.974767: step: 1248/527, loss: 0.06322555989027023
2023-01-21 10:38:46.122025: step: 1252/527, loss: 0.01869661919772625
2023-01-21 10:38:47.234324: step: 1256/527, loss: 0.05436287075281143
2023-01-21 10:38:48.376843: step: 1260/527, loss: 0.5027929544448853
2023-01-21 10:38:49.503720: step: 1264/527, loss: 0.0004455566522665322
2023-01-21 10:38:50.646361: step: 1268/527, loss: 0.02612585946917534
2023-01-21 10:38:51.844719: step: 1272/527, loss: 0.1235879436135292
2023-01-21 10:38:52.958606: step: 1276/527, loss: 0.006830072030425072
2023-01-21 10:38:54.079852: step: 1280/527, loss: 0.006117630284279585
2023-01-21 10:38:55.238062: step: 1284/527, loss: 0.032392311841249466
2023-01-21 10:38:56.328635: step: 1288/527, loss: 0.08437366783618927
2023-01-21 10:38:57.502357: step: 1292/527, loss: 0.014499664306640625
2023-01-21 10:38:58.659064: step: 1296/527, loss: 0.024416828528046608
2023-01-21 10:38:59.789503: step: 1300/527, loss: 0.08685169368982315
2023-01-21 10:39:00.913721: step: 1304/527, loss: 0.0749754011631012
2023-01-21 10:39:02.054443: step: 1308/527, loss: 0.06223135069012642
2023-01-21 10:39:03.193799: step: 1312/527, loss: 0.019977666437625885
2023-01-21 10:39:04.296555: step: 1316/527, loss: 0.04885587841272354
2023-01-21 10:39:05.431283: step: 1320/527, loss: 0.09427537769079208
2023-01-21 10:39:06.542198: step: 1324/527, loss: 0.01026086788624525
2023-01-21 10:39:07.642333: step: 1328/527, loss: 0.003559684846550226
2023-01-21 10:39:08.752759: step: 1332/527, loss: 0.02853412553668022
2023-01-21 10:39:09.889966: step: 1336/527, loss: 0.04844551160931587
2023-01-21 10:39:11.007799: step: 1340/527, loss: 0.06234912946820259
2023-01-21 10:39:12.144243: step: 1344/527, loss: 0.09103145450353622
2023-01-21 10:39:13.255691: step: 1348/527, loss: 0.12149371951818466
2023-01-21 10:39:14.349450: step: 1352/527, loss: 0.011648845858871937
2023-01-21 10:39:15.449309: step: 1356/527, loss: 0.024466516450047493
2023-01-21 10:39:16.588062: step: 1360/527, loss: 0.0091400146484375
2023-01-21 10:39:17.714304: step: 1364/527, loss: 0.008623885922133923
2023-01-21 10:39:18.847865: step: 1368/527, loss: 0.0035786153748631477
2023-01-21 10:39:19.968157: step: 1372/527, loss: 0.05435457453131676
2023-01-21 10:39:21.089573: step: 1376/527, loss: 0.02626619301736355
2023-01-21 10:39:22.228549: step: 1380/527, loss: 0.0870644599199295
2023-01-21 10:39:23.364374: step: 1384/527, loss: 0.016962861642241478
2023-01-21 10:39:24.494052: step: 1388/527, loss: 0.03856506198644638
2023-01-21 10:39:25.642314: step: 1392/527, loss: 0.2629574239253998
2023-01-21 10:39:26.752983: step: 1396/527, loss: 0.01939401775598526
2023-01-21 10:39:27.864243: step: 1400/527, loss: 0.004716205410659313
2023-01-21 10:39:29.021184: step: 1404/527, loss: 0.07821989059448242
2023-01-21 10:39:30.137481: step: 1408/527, loss: 0.0006430625799112022
2023-01-21 10:39:31.257064: step: 1412/527, loss: 0.018964242190122604
2023-01-21 10:39:32.368545: step: 1416/527, loss: 0.04407491534948349
2023-01-21 10:39:33.496958: step: 1420/527, loss: 0.0819365456700325
2023-01-21 10:39:34.621536: step: 1424/527, loss: 0.06758375465869904
2023-01-21 10:39:35.770523: step: 1428/527, loss: 0.0044998168013989925
2023-01-21 10:39:36.878702: step: 1432/527, loss: 0.5633466243743896
2023-01-21 10:39:38.017003: step: 1436/527, loss: 0.06056986004114151
2023-01-21 10:39:39.154418: step: 1440/527, loss: 0.0574704185128212
2023-01-21 10:39:40.257435: step: 1444/527, loss: 0.23752641677856445
2023-01-21 10:39:41.413740: step: 1448/527, loss: 0.4880952537059784
2023-01-21 10:39:42.557847: step: 1452/527, loss: 0.04465198516845703
2023-01-21 10:39:43.685545: step: 1456/527, loss: 0.004951906390488148
2023-01-21 10:39:44.830839: step: 1460/527, loss: 0.010083580389618874
2023-01-21 10:39:45.970153: step: 1464/527, loss: 0.04278833791613579
2023-01-21 10:39:47.110633: step: 1468/527, loss: 0.049361709505319595
2023-01-21 10:39:48.229333: step: 1472/527, loss: 0.004602623172104359
2023-01-21 10:39:49.341319: step: 1476/527, loss: 0.04504900053143501
2023-01-21 10:39:50.495156: step: 1480/527, loss: 0.05486660078167915
2023-01-21 10:39:51.589833: step: 1484/527, loss: 0.04359569400548935
2023-01-21 10:39:52.736876: step: 1488/527, loss: 0.09672889858484268
2023-01-21 10:39:53.881092: step: 1492/527, loss: 0.06715402752161026
2023-01-21 10:39:55.015451: step: 1496/527, loss: 0.030863190069794655
2023-01-21 10:39:56.140659: step: 1500/527, loss: 0.042561959475278854
2023-01-21 10:39:57.284415: step: 1504/527, loss: 0.25667861104011536
2023-01-21 10:39:58.398423: step: 1508/527, loss: 0.03336620330810547
2023-01-21 10:39:59.529777: step: 1512/527, loss: 0.03314628452062607
2023-01-21 10:40:00.664320: step: 1516/527, loss: 0.020035363733768463
2023-01-21 10:40:01.771942: step: 1520/527, loss: 0.026305008679628372
2023-01-21 10:40:02.885412: step: 1524/527, loss: 0.029089193791151047
2023-01-21 10:40:04.013877: step: 1528/527, loss: 0.02731633372604847
2023-01-21 10:40:05.138008: step: 1532/527, loss: 0.04406256601214409
2023-01-21 10:40:06.254530: step: 1536/527, loss: 0.07775793224573135
2023-01-21 10:40:07.394186: step: 1540/527, loss: 0.06643657386302948
2023-01-21 10:40:08.514376: step: 1544/527, loss: 0.012332677841186523
2023-01-21 10:40:09.663662: step: 1548/527, loss: 0.012622261419892311
2023-01-21 10:40:10.795993: step: 1552/527, loss: 0.08494973182678223
2023-01-21 10:40:11.936647: step: 1556/527, loss: 0.10484905540943146
2023-01-21 10:40:13.071557: step: 1560/527, loss: 0.020217323675751686
2023-01-21 10:40:14.208038: step: 1564/527, loss: 0.027220916002988815
2023-01-21 10:40:15.358446: step: 1568/527, loss: 0.05368633568286896
2023-01-21 10:40:16.510942: step: 1572/527, loss: 0.04664745181798935
2023-01-21 10:40:17.638304: step: 1576/527, loss: 0.024153614416718483
2023-01-21 10:40:18.753458: step: 1580/527, loss: 0.010813355445861816
2023-01-21 10:40:19.871374: step: 1584/527, loss: 0.028812408447265625
2023-01-21 10:40:21.014140: step: 1588/527, loss: 0.05734281614422798
2023-01-21 10:40:22.135192: step: 1592/527, loss: 0.03375907242298126
2023-01-21 10:40:23.245298: step: 1596/527, loss: 0.014395713806152344
2023-01-21 10:40:24.364299: step: 1600/527, loss: 0.03240685537457466
2023-01-21 10:40:25.472449: step: 1604/527, loss: 0.7524027824401855
2023-01-21 10:40:26.611323: step: 1608/527, loss: 0.00562329264357686
2023-01-21 10:40:27.783914: step: 1612/527, loss: 0.06875276565551758
2023-01-21 10:40:28.922204: step: 1616/527, loss: 0.061720944941043854
2023-01-21 10:40:30.050234: step: 1620/527, loss: 0.022563554346561432
2023-01-21 10:40:31.192845: step: 1624/527, loss: 0.07205381244421005
2023-01-21 10:40:32.341723: step: 1628/527, loss: 0.029694175347685814
2023-01-21 10:40:33.499604: step: 1632/527, loss: 0.07472991943359375
2023-01-21 10:40:34.643952: step: 1636/527, loss: 0.054024696350097656
2023-01-21 10:40:35.742515: step: 1640/527, loss: 0.0029277324210852385
2023-01-21 10:40:36.873886: step: 1644/527, loss: 0.0031118392944335938
2023-01-21 10:40:38.008296: step: 1648/527, loss: 0.08872871100902557
2023-01-21 10:40:39.131521: step: 1652/527, loss: 0.08338575810194016
2023-01-21 10:40:40.287001: step: 1656/527, loss: 0.020903684198856354
2023-01-21 10:40:41.447192: step: 1660/527, loss: 0.016211891546845436
2023-01-21 10:40:42.555841: step: 1664/527, loss: 0.029464852064847946
2023-01-21 10:40:43.734970: step: 1668/527, loss: 0.2940831184387207
2023-01-21 10:40:44.870607: step: 1672/527, loss: 1.0269176959991455
2023-01-21 10:40:45.988709: step: 1676/527, loss: 0.03849220275878906
2023-01-21 10:40:47.137899: step: 1680/527, loss: 0.018545055761933327
2023-01-21 10:40:48.265131: step: 1684/527, loss: 0.0028781890869140625
2023-01-21 10:40:49.378488: step: 1688/527, loss: 0.07716947048902512
2023-01-21 10:40:50.502339: step: 1692/527, loss: 0.01875319518148899
2023-01-21 10:40:51.639548: step: 1696/527, loss: 0.005181122105568647
2023-01-21 10:40:52.759370: step: 1700/527, loss: 0.04525575786828995
2023-01-21 10:40:53.881968: step: 1704/527, loss: 0.02023792453110218
2023-01-21 10:40:55.039856: step: 1708/527, loss: 0.026078414171934128
2023-01-21 10:40:56.209671: step: 1712/527, loss: 0.04098577797412872
2023-01-21 10:40:57.336723: step: 1716/527, loss: 0.008429097943007946
2023-01-21 10:40:58.480704: step: 1720/527, loss: 0.09411545097827911
2023-01-21 10:40:59.592796: step: 1724/527, loss: 0.019701480865478516
2023-01-21 10:41:00.730057: step: 1728/527, loss: 0.0583980567753315
2023-01-21 10:41:01.848548: step: 1732/527, loss: 0.028992796316742897
2023-01-21 10:41:03.006394: step: 1736/527, loss: 0.09212350845336914
2023-01-21 10:41:04.154643: step: 1740/527, loss: 0.00423011789098382
2023-01-21 10:41:05.305519: step: 1744/527, loss: 0.10672563314437866
2023-01-21 10:41:06.468235: step: 1748/527, loss: 0.056700803339481354
2023-01-21 10:41:07.593567: step: 1752/527, loss: 0.05688953399658203
2023-01-21 10:41:08.719579: step: 1756/527, loss: 0.04488696902990341
2023-01-21 10:41:09.838124: step: 1760/527, loss: 0.04690838232636452
2023-01-21 10:41:10.990471: step: 1764/527, loss: 0.005692005157470703
2023-01-21 10:41:12.104113: step: 1768/527, loss: 0.014497661963105202
2023-01-21 10:41:13.225784: step: 1772/527, loss: 0.021654700860381126
2023-01-21 10:41:14.377485: step: 1776/527, loss: 0.03479165956377983
2023-01-21 10:41:15.485853: step: 1780/527, loss: 0.015383625403046608
2023-01-21 10:41:16.595153: step: 1784/527, loss: 0.05962171405553818
2023-01-21 10:41:17.722499: step: 1788/527, loss: 0.03440684825181961
2023-01-21 10:41:18.859138: step: 1792/527, loss: 0.022205591201782227
2023-01-21 10:41:19.999313: step: 1796/527, loss: 0.025052262470126152
2023-01-21 10:41:21.112746: step: 1800/527, loss: 0.041197583079338074
2023-01-21 10:41:22.265453: step: 1804/527, loss: 0.011138200759887695
2023-01-21 10:41:23.406751: step: 1808/527, loss: 0.005200004205107689
2023-01-21 10:41:24.574493: step: 1812/527, loss: 0.0211184024810791
2023-01-21 10:41:25.692564: step: 1816/527, loss: 0.0179640781134367
2023-01-21 10:41:26.857722: step: 1820/527, loss: 0.05691700056195259
2023-01-21 10:41:28.001671: step: 1824/527, loss: 0.0329742468893528
2023-01-21 10:41:29.115389: step: 1828/527, loss: 0.007965756580233574
2023-01-21 10:41:30.256249: step: 1832/527, loss: 0.03431396186351776
2023-01-21 10:41:31.379950: step: 1836/527, loss: 0.00861892756074667
2023-01-21 10:41:32.519627: step: 1840/527, loss: 0.0369756706058979
2023-01-21 10:41:33.674043: step: 1844/527, loss: 0.003796482225880027
2023-01-21 10:41:34.805591: step: 1848/527, loss: 0.004118728451430798
2023-01-21 10:41:35.963626: step: 1852/527, loss: 0.028210068121552467
2023-01-21 10:41:37.101184: step: 1856/527, loss: 0.05536289140582085
2023-01-21 10:41:38.214081: step: 1860/527, loss: 0.022899245843291283
2023-01-21 10:41:39.366791: step: 1864/527, loss: 0.06281552463769913
2023-01-21 10:41:40.511215: step: 1868/527, loss: 0.03542738035321236
2023-01-21 10:41:41.668867: step: 1872/527, loss: 0.04755659028887749
2023-01-21 10:41:42.782729: step: 1876/527, loss: 0.05237589031457901
2023-01-21 10:41:43.936599: step: 1880/527, loss: 0.026126384735107422
2023-01-21 10:41:45.045432: step: 1884/527, loss: 0.03187618404626846
2023-01-21 10:41:46.199543: step: 1888/527, loss: 0.16256046295166016
2023-01-21 10:41:47.323693: step: 1892/527, loss: 0.02219209633767605
2023-01-21 10:41:48.473618: step: 1896/527, loss: 0.0382721908390522
2023-01-21 10:41:49.610271: step: 1900/527, loss: 0.024348163977265358
2023-01-21 10:41:50.757955: step: 1904/527, loss: 0.019045734778046608
2023-01-21 10:41:51.900037: step: 1908/527, loss: 0.0006413937080651522
2023-01-21 10:41:53.029354: step: 1912/527, loss: 0.3515731990337372
2023-01-21 10:41:54.162566: step: 1916/527, loss: 0.04123964160680771
2023-01-21 10:41:55.307843: step: 1920/527, loss: 0.047490693628787994
2023-01-21 10:41:56.432083: step: 1924/527, loss: 0.01776914671063423
2023-01-21 10:41:57.563590: step: 1928/527, loss: 0.009844970889389515
2023-01-21 10:41:58.647916: step: 1932/527, loss: 0.009848546236753464
2023-01-21 10:41:59.760783: step: 1936/527, loss: 0.006745052058249712
2023-01-21 10:42:00.895321: step: 1940/527, loss: 0.08625411987304688
2023-01-21 10:42:02.007164: step: 1944/527, loss: 0.012703323736786842
2023-01-21 10:42:03.158644: step: 1948/527, loss: 0.05798950046300888
2023-01-21 10:42:04.310869: step: 1952/527, loss: 0.0014046193100512028
2023-01-21 10:42:05.448469: step: 1956/527, loss: 0.744663417339325
2023-01-21 10:42:06.575468: step: 1960/527, loss: 0.0025683403946459293
2023-01-21 10:42:07.708367: step: 1964/527, loss: 0.0832613930106163
2023-01-21 10:42:08.878924: step: 1968/527, loss: 0.033837415277957916
2023-01-21 10:42:10.013471: step: 1972/527, loss: 0.03206272050738335
2023-01-21 10:42:11.128722: step: 1976/527, loss: 0.11563168466091156
2023-01-21 10:42:12.227564: step: 1980/527, loss: 0.01674327813088894
2023-01-21 10:42:13.371645: step: 1984/527, loss: 0.055724527686834335
2023-01-21 10:42:14.499067: step: 1988/527, loss: 0.07718362659215927
2023-01-21 10:42:15.657502: step: 1992/527, loss: 0.49698877334594727
2023-01-21 10:42:16.758492: step: 1996/527, loss: 0.0918058454990387
2023-01-21 10:42:17.876010: step: 2000/527, loss: 0.07412996888160706
2023-01-21 10:42:19.020745: step: 2004/527, loss: 0.8807516694068909
2023-01-21 10:42:20.115162: step: 2008/527, loss: 0.012332153506577015
2023-01-21 10:42:21.282108: step: 2012/527, loss: 0.015344763174653053
2023-01-21 10:42:22.449868: step: 2016/527, loss: 0.04353685304522514
2023-01-21 10:42:23.555532: step: 2020/527, loss: 0.02442960813641548
2023-01-21 10:42:24.701352: step: 2024/527, loss: 0.0745411366224289
2023-01-21 10:42:25.828207: step: 2028/527, loss: 0.020720958709716797
2023-01-21 10:42:26.985155: step: 2032/527, loss: 0.039470769464969635
2023-01-21 10:42:28.129414: step: 2036/527, loss: 0.020329762250185013
2023-01-21 10:42:29.234930: step: 2040/527, loss: 0.024169350042939186
2023-01-21 10:42:30.388843: step: 2044/527, loss: 0.0059812068939208984
2023-01-21 10:42:31.513751: step: 2048/527, loss: 0.011644172482192516
2023-01-21 10:42:32.640808: step: 2052/527, loss: 0.046210192143917084
2023-01-21 10:42:33.803164: step: 2056/527, loss: 0.03106365166604519
2023-01-21 10:42:34.932854: step: 2060/527, loss: 0.05386839061975479
2023-01-21 10:42:36.065028: step: 2064/527, loss: 0.17499685287475586
2023-01-21 10:42:37.196084: step: 2068/527, loss: 0.05699767917394638
2023-01-21 10:42:38.323843: step: 2072/527, loss: 0.10650528967380524
2023-01-21 10:42:39.469553: step: 2076/527, loss: 0.01688671112060547
2023-01-21 10:42:40.640758: step: 2080/527, loss: 0.0027322769165039062
2023-01-21 10:42:41.795005: step: 2084/527, loss: 0.16313126683235168
2023-01-21 10:42:42.914151: step: 2088/527, loss: 0.11428767442703247
2023-01-21 10:42:44.030919: step: 2092/527, loss: 0.015866708010435104
2023-01-21 10:42:45.202462: step: 2096/527, loss: 0.019983459264039993
2023-01-21 10:42:46.340486: step: 2100/527, loss: 0.05142416059970856
2023-01-21 10:42:47.465678: step: 2104/527, loss: 0.007344258017838001
2023-01-21 10:42:48.595114: step: 2108/527, loss: 0.1389904022216797
==================================================
Loss: 0.070
--------------------
Dev: {'event': {'p': 0.592479674796748, 'r': 0.7762982689747004, 'f1': 0.6720461095100865}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12}
Test: {'event': {'p': 0.6183136899365367, 'r': 0.7794285714285715, 'f1': 0.6895854398382204}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12}
Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12}
Korean: {'event': {'p': 0.603448275862069, 'r': 0.5555555555555556, 'f1': 0.5785123966942148}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12}
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': 12}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:43:32.206835: step: 4/527, loss: 0.05483436957001686
2023-01-21 10:43:33.358344: step: 8/527, loss: 0.013237237930297852
2023-01-21 10:43:34.530259: step: 12/527, loss: 0.048719264566898346
2023-01-21 10:43:35.672661: step: 16/527, loss: 0.0151824951171875
2023-01-21 10:43:36.820289: step: 20/527, loss: 0.021438216790556908
2023-01-21 10:43:37.979028: step: 24/527, loss: 0.0351351723074913
2023-01-21 10:43:39.098013: step: 28/527, loss: 0.23648852109909058
2023-01-21 10:43:40.213220: step: 32/527, loss: 0.07371443510055542
2023-01-21 10:43:41.351796: step: 36/527, loss: 0.031215669587254524
2023-01-21 10:43:42.484396: step: 40/527, loss: 0.0012390136253088713
2023-01-21 10:43:43.629866: step: 44/527, loss: 0.141058549284935
2023-01-21 10:43:44.790760: step: 48/527, loss: 0.0750507339835167
2023-01-21 10:43:45.935708: step: 52/527, loss: 0.033423613756895065
2023-01-21 10:43:47.066028: step: 56/527, loss: 0.28667691349983215
2023-01-21 10:43:48.219403: step: 60/527, loss: 0.018096160143613815
2023-01-21 10:43:49.337111: step: 64/527, loss: 0.040366366505622864
2023-01-21 10:43:50.498195: step: 68/527, loss: 0.04597120359539986
2023-01-21 10:43:51.637273: step: 72/527, loss: 0.05049419403076172
2023-01-21 10:43:52.758938: step: 76/527, loss: 0.032770346850156784
2023-01-21 10:43:53.898007: step: 80/527, loss: 0.033068373799324036
2023-01-21 10:43:55.061887: step: 84/527, loss: 0.025774337351322174
2023-01-21 10:43:56.176053: step: 88/527, loss: 0.1166144385933876
2023-01-21 10:43:57.320345: step: 92/527, loss: 0.21611842513084412
2023-01-21 10:43:58.420782: step: 96/527, loss: 0.03145551681518555
2023-01-21 10:43:59.571982: step: 100/527, loss: 0.012934256345033646
2023-01-21 10:44:00.718919: step: 104/527, loss: 0.01688366010785103
2023-01-21 10:44:01.834436: step: 108/527, loss: 0.01282196119427681
2023-01-21 10:44:02.987292: step: 112/527, loss: 0.0035398483742028475
2023-01-21 10:44:04.123137: step: 116/527, loss: 0.021576501429080963
2023-01-21 10:44:05.264822: step: 120/527, loss: 0.031232357025146484
2023-01-21 10:44:06.387749: step: 124/527, loss: 0.012781786732375622
2023-01-21 10:44:07.505369: step: 128/527, loss: 0.006638812832534313
2023-01-21 10:44:08.649314: step: 132/527, loss: 0.0774417445063591
2023-01-21 10:44:09.810257: step: 136/527, loss: 0.031858157366514206
2023-01-21 10:44:10.933430: step: 140/527, loss: 0.020217036828398705
2023-01-21 10:44:12.071532: step: 144/527, loss: 0.34765854477882385
2023-01-21 10:44:13.222250: step: 148/527, loss: 0.004458904266357422
2023-01-21 10:44:14.389989: step: 152/527, loss: 0.04680920019745827
2023-01-21 10:44:15.530554: step: 156/527, loss: 0.001400089357048273
2023-01-21 10:44:16.679346: step: 160/527, loss: 0.010438013821840286
2023-01-21 10:44:17.805115: step: 164/527, loss: 0.03862609714269638
2023-01-21 10:44:18.957680: step: 168/527, loss: 0.017327118664979935
2023-01-21 10:44:20.122511: step: 172/527, loss: 0.11928883194923401
2023-01-21 10:44:21.279661: step: 176/527, loss: 0.002925062319263816
2023-01-21 10:44:22.399650: step: 180/527, loss: 0.035975027829408646
2023-01-21 10:44:23.533898: step: 184/527, loss: 0.008519936352968216
2023-01-21 10:44:24.653634: step: 188/527, loss: 0.009121417999267578
2023-01-21 10:44:25.815415: step: 192/527, loss: 0.007589531131088734
2023-01-21 10:44:26.935826: step: 196/527, loss: 0.0016632080078125
2023-01-21 10:44:28.121632: step: 200/527, loss: 0.018468666821718216
2023-01-21 10:44:29.270636: step: 204/527, loss: 0.07905006408691406
2023-01-21 10:44:30.399505: step: 208/527, loss: 0.05135173723101616
2023-01-21 10:44:31.541345: step: 212/527, loss: 0.008401012979447842
2023-01-21 10:44:32.705920: step: 216/527, loss: 0.011823464184999466
2023-01-21 10:44:33.820937: step: 220/527, loss: 0.03426084667444229
2023-01-21 10:44:34.941213: step: 224/527, loss: 0.0034529685508459806
2023-01-21 10:44:36.092836: step: 228/527, loss: 0.07303076237440109
2023-01-21 10:44:37.233108: step: 232/527, loss: 0.0059465887024998665
2023-01-21 10:44:38.351280: step: 236/527, loss: 0.010770061053335667
2023-01-21 10:44:39.451522: step: 240/527, loss: 0.0003796577802859247
2023-01-21 10:44:40.573978: step: 244/527, loss: 0.033928871154785156
2023-01-21 10:44:41.693253: step: 248/527, loss: 0.019697191193699837
2023-01-21 10:44:42.842001: step: 252/527, loss: 0.005572700873017311
2023-01-21 10:44:43.969518: step: 256/527, loss: 0.04180946201086044
2023-01-21 10:44:45.110556: step: 260/527, loss: 0.017916489392518997
2023-01-21 10:44:46.274786: step: 264/527, loss: 0.2511996328830719
2023-01-21 10:44:47.387923: step: 268/527, loss: 0.00015192032151389867
2023-01-21 10:44:48.506726: step: 272/527, loss: 0.025321578606963158
2023-01-21 10:44:49.647539: step: 276/527, loss: 0.09013523906469345
2023-01-21 10:44:50.776666: step: 280/527, loss: 0.8055596351623535
2023-01-21 10:44:51.909268: step: 284/527, loss: 0.0018808366730809212
2023-01-21 10:44:53.037196: step: 288/527, loss: 0.038468360900878906
2023-01-21 10:44:54.137352: step: 292/527, loss: 0.05327153205871582
2023-01-21 10:44:55.291015: step: 296/527, loss: 0.0043884990736842155
2023-01-21 10:44:56.434651: step: 300/527, loss: 0.01793527603149414
2023-01-21 10:44:57.553079: step: 304/527, loss: 0.0072252992540597916
2023-01-21 10:44:58.704451: step: 308/527, loss: 0.02978992462158203
2023-01-21 10:44:59.821216: step: 312/527, loss: 0.14135456085205078
2023-01-21 10:45:00.986235: step: 316/527, loss: 0.038083840161561966
2023-01-21 10:45:02.133688: step: 320/527, loss: 0.06750774383544922
2023-01-21 10:45:03.292463: step: 324/527, loss: 0.0014193535316735506
2023-01-21 10:45:04.402758: step: 328/527, loss: 0.031321145594120026
2023-01-21 10:45:05.531377: step: 332/527, loss: 0.015298556536436081
2023-01-21 10:45:06.680999: step: 336/527, loss: 0.10010266304016113
2023-01-21 10:45:07.827485: step: 340/527, loss: 0.021171577274799347
2023-01-21 10:45:09.010186: step: 344/527, loss: 0.0022035599686205387
2023-01-21 10:45:10.149733: step: 348/527, loss: 0.4072292447090149
2023-01-21 10:45:11.254905: step: 352/527, loss: 0.0101292310282588
2023-01-21 10:45:12.411178: step: 356/527, loss: 0.014102363958954811
2023-01-21 10:45:13.543267: step: 360/527, loss: 0.0020759582985192537
2023-01-21 10:45:14.678145: step: 364/527, loss: 0.00048065185546875
2023-01-21 10:45:15.782128: step: 368/527, loss: 0.011178016662597656
2023-01-21 10:45:16.888850: step: 372/527, loss: 0.043828632682561874
2023-01-21 10:45:18.000479: step: 376/527, loss: 0.001505184220150113
2023-01-21 10:45:19.103619: step: 380/527, loss: 0.298762708902359
2023-01-21 10:45:20.239694: step: 384/527, loss: 0.00553207378834486
2023-01-21 10:45:21.354899: step: 388/527, loss: 0.01676025427877903
2023-01-21 10:45:22.537307: step: 392/527, loss: 0.0258955005556345
2023-01-21 10:45:23.638007: step: 396/527, loss: 0.020168591290712357
2023-01-21 10:45:24.753369: step: 400/527, loss: 0.022164441645145416
2023-01-21 10:45:25.893557: step: 404/527, loss: 0.014228916727006435
2023-01-21 10:45:27.004112: step: 408/527, loss: 0.0035377503372728825
2023-01-21 10:45:28.107450: step: 412/527, loss: 0.09039898216724396
2023-01-21 10:45:29.263391: step: 416/527, loss: 0.07017679512500763
2023-01-21 10:45:30.362317: step: 420/527, loss: 0.057897478342056274
2023-01-21 10:45:31.486463: step: 424/527, loss: 0.03418402746319771
2023-01-21 10:45:32.616637: step: 428/527, loss: 0.11395101994276047
2023-01-21 10:45:33.757046: step: 432/527, loss: 0.005800294689834118
2023-01-21 10:45:34.892191: step: 436/527, loss: 0.06565552204847336
2023-01-21 10:45:36.013419: step: 440/527, loss: 0.015845298767089844
2023-01-21 10:45:37.134851: step: 444/527, loss: 0.09393219649791718
2023-01-21 10:45:38.278007: step: 448/527, loss: 0.021622370928525925
2023-01-21 10:45:39.413335: step: 452/527, loss: 0.024625588208436966
2023-01-21 10:45:40.579711: step: 456/527, loss: 0.013593673706054688
2023-01-21 10:45:41.740779: step: 460/527, loss: 0.04725513607263565
2023-01-21 10:45:42.861235: step: 464/527, loss: 0.015504170209169388
2023-01-21 10:45:43.989680: step: 468/527, loss: 0.0030059814453125
2023-01-21 10:45:45.103874: step: 472/527, loss: 0.044334981590509415
2023-01-21 10:45:46.235496: step: 476/527, loss: 0.04174797609448433
2023-01-21 10:45:47.380095: step: 480/527, loss: 0.05864086374640465
2023-01-21 10:45:48.506408: step: 484/527, loss: 0.04286251589655876
2023-01-21 10:45:49.610008: step: 488/527, loss: 0.01697230339050293
2023-01-21 10:45:50.790472: step: 492/527, loss: 0.07252369076013565
2023-01-21 10:45:51.912207: step: 496/527, loss: 0.012589598074555397
2023-01-21 10:45:53.056818: step: 500/527, loss: 0.01417403295636177
2023-01-21 10:45:54.185377: step: 504/527, loss: 0.049610234797000885
2023-01-21 10:45:55.328748: step: 508/527, loss: 0.057517051696777344
2023-01-21 10:45:56.439340: step: 512/527, loss: 0.009934711270034313
2023-01-21 10:45:57.569379: step: 516/527, loss: 0.01191701926290989
2023-01-21 10:45:58.744443: step: 520/527, loss: 0.06930327415466309
2023-01-21 10:45:59.858974: step: 524/527, loss: 0.09046545624732971
2023-01-21 10:46:01.002971: step: 528/527, loss: 0.12750712037086487
2023-01-21 10:46:02.125242: step: 532/527, loss: 0.023652076721191406
2023-01-21 10:46:03.257961: step: 536/527, loss: 0.023132897913455963
2023-01-21 10:46:04.393042: step: 540/527, loss: 0.03845147788524628
2023-01-21 10:46:05.512886: step: 544/527, loss: 0.0004531860467977822
2023-01-21 10:46:06.656558: step: 548/527, loss: 0.014815902337431908
2023-01-21 10:46:07.816907: step: 552/527, loss: 0.057401467114686966
2023-01-21 10:46:08.917949: step: 556/527, loss: 0.0022819519508630037
2023-01-21 10:46:10.066878: step: 560/527, loss: 0.000146770486026071
2023-01-21 10:46:11.200059: step: 564/527, loss: 0.004998969845473766
2023-01-21 10:46:12.358230: step: 568/527, loss: 0.006308412179350853
2023-01-21 10:46:13.478767: step: 572/527, loss: 0.007084369659423828
2023-01-21 10:46:14.606385: step: 576/527, loss: 0.004778003320097923
2023-01-21 10:46:15.734400: step: 580/527, loss: 0.016982747241854668
2023-01-21 10:46:16.872941: step: 584/527, loss: 0.0042404173873364925
2023-01-21 10:46:18.052969: step: 588/527, loss: 0.00701332138851285
2023-01-21 10:46:19.209487: step: 592/527, loss: 0.03866443783044815
2023-01-21 10:46:20.327983: step: 596/527, loss: 0.02811145968735218
2023-01-21 10:46:21.476002: step: 600/527, loss: 0.07745261490345001
2023-01-21 10:46:22.580112: step: 604/527, loss: 0.08996114879846573
2023-01-21 10:46:23.724195: step: 608/527, loss: 0.003554916474968195
2023-01-21 10:46:24.875291: step: 612/527, loss: 0.036653995513916016
2023-01-21 10:46:25.991902: step: 616/527, loss: 0.00877599697560072
2023-01-21 10:46:27.129238: step: 620/527, loss: 0.03270683437585831
2023-01-21 10:46:28.255810: step: 624/527, loss: 0.10150251537561417
2023-01-21 10:46:29.417433: step: 628/527, loss: 0.018178559839725494
2023-01-21 10:46:30.519718: step: 632/527, loss: 0.014668917283415794
2023-01-21 10:46:31.673734: step: 636/527, loss: 0.0009556770673952997
2023-01-21 10:46:32.794729: step: 640/527, loss: 0.034028053283691406
2023-01-21 10:46:33.917723: step: 644/527, loss: 0.01180257834494114
2023-01-21 10:46:35.035562: step: 648/527, loss: 0.04315795749425888
2023-01-21 10:46:36.162713: step: 652/527, loss: 0.04329271614551544
2023-01-21 10:46:37.286474: step: 656/527, loss: 0.0748906135559082
2023-01-21 10:46:38.405383: step: 660/527, loss: 0.7598684430122375
2023-01-21 10:46:39.524303: step: 664/527, loss: 0.03290081024169922
2023-01-21 10:46:40.657769: step: 668/527, loss: 0.003654098603874445
2023-01-21 10:46:41.797110: step: 672/527, loss: 0.019790077582001686
2023-01-21 10:46:42.926524: step: 676/527, loss: 0.042586613446474075
2023-01-21 10:46:44.062054: step: 680/527, loss: 0.01973877102136612
2023-01-21 10:46:45.186729: step: 684/527, loss: 0.0228181853890419
2023-01-21 10:46:46.289487: step: 688/527, loss: 0.05579328536987305
2023-01-21 10:46:47.460769: step: 692/527, loss: 0.04766101762652397
2023-01-21 10:46:48.632725: step: 696/527, loss: 0.02511434629559517
2023-01-21 10:46:49.759606: step: 700/527, loss: 0.023438608273863792
2023-01-21 10:46:50.891437: step: 704/527, loss: 0.08914432674646378
2023-01-21 10:46:51.980175: step: 708/527, loss: 0.03259143978357315
2023-01-21 10:46:53.119026: step: 712/527, loss: 0.0074783326126635075
2023-01-21 10:46:54.235957: step: 716/527, loss: 0.038152217864990234
2023-01-21 10:46:55.390212: step: 720/527, loss: 0.022386979311704636
2023-01-21 10:46:56.497721: step: 724/527, loss: 0.0526885986328125
2023-01-21 10:46:57.627044: step: 728/527, loss: 0.09961071610450745
2023-01-21 10:46:58.763022: step: 732/527, loss: 0.0810524970293045
2023-01-21 10:46:59.918388: step: 736/527, loss: 0.01377859152853489
2023-01-21 10:47:01.025940: step: 740/527, loss: 0.011016559787094593
2023-01-21 10:47:02.197787: step: 744/527, loss: 0.06735400855541229
2023-01-21 10:47:03.333383: step: 748/527, loss: 0.01777200773358345
2023-01-21 10:47:04.456809: step: 752/527, loss: 0.0016444266075268388
2023-01-21 10:47:05.618295: step: 756/527, loss: 0.04064963012933731
2023-01-21 10:47:06.763595: step: 760/527, loss: 0.0959063470363617
2023-01-21 10:47:07.916116: step: 764/527, loss: 0.05010666698217392
2023-01-21 10:47:09.044467: step: 768/527, loss: 0.007413243874907494
2023-01-21 10:47:10.145335: step: 772/527, loss: 0.03229045867919922
2023-01-21 10:47:11.280643: step: 776/527, loss: 0.0047776224091649055
2023-01-21 10:47:12.393684: step: 780/527, loss: 0.02356776036322117
2023-01-21 10:47:13.499795: step: 784/527, loss: 0.0036592960823327303
2023-01-21 10:47:14.615900: step: 788/527, loss: 0.06131782382726669
2023-01-21 10:47:15.781571: step: 792/527, loss: 0.022019004449248314
2023-01-21 10:47:16.926789: step: 796/527, loss: 0.006955373100936413
2023-01-21 10:47:18.034625: step: 800/527, loss: 0.0364038422703743
2023-01-21 10:47:19.156299: step: 804/527, loss: 0.07245779037475586
2023-01-21 10:47:20.289262: step: 808/527, loss: 0.039669133722782135
2023-01-21 10:47:21.394660: step: 812/527, loss: 0.018953992053866386
2023-01-21 10:47:22.510564: step: 816/527, loss: 0.060761354863643646
2023-01-21 10:47:23.636311: step: 820/527, loss: 0.02682756446301937
2023-01-21 10:47:24.750735: step: 824/527, loss: 0.011416817083954811
2023-01-21 10:47:25.880348: step: 828/527, loss: 0.029276657849550247
2023-01-21 10:47:27.030082: step: 832/527, loss: 0.06891937553882599
2023-01-21 10:47:28.153944: step: 836/527, loss: 0.011512089520692825
2023-01-21 10:47:29.264068: step: 840/527, loss: 0.022316837683320045
2023-01-21 10:47:30.356323: step: 844/527, loss: 0.0008707523229531944
2023-01-21 10:47:31.491549: step: 848/527, loss: 0.04833660274744034
2023-01-21 10:47:32.632225: step: 852/527, loss: 0.009812736883759499
2023-01-21 10:47:33.755336: step: 856/527, loss: 0.07840891182422638
2023-01-21 10:47:34.913880: step: 860/527, loss: 0.09594688564538956
2023-01-21 10:47:36.053418: step: 864/527, loss: 0.015311798080801964
2023-01-21 10:47:37.193278: step: 868/527, loss: 0.1030879020690918
2023-01-21 10:47:38.307951: step: 872/527, loss: 0.031411126255989075
2023-01-21 10:47:39.428877: step: 876/527, loss: 0.025075819343328476
2023-01-21 10:47:40.601340: step: 880/527, loss: 0.09888773411512375
2023-01-21 10:47:41.729643: step: 884/527, loss: 0.01148681715130806
2023-01-21 10:47:42.841617: step: 888/527, loss: 0.0035987854935228825
2023-01-21 10:47:43.972456: step: 892/527, loss: 0.012640666216611862
2023-01-21 10:47:45.111345: step: 896/527, loss: 0.09748916327953339
2023-01-21 10:47:46.211020: step: 900/527, loss: 0.02198486216366291
2023-01-21 10:47:47.323669: step: 904/527, loss: 0.04246368631720543
2023-01-21 10:47:48.468080: step: 908/527, loss: 0.03166789934039116
2023-01-21 10:47:49.619581: step: 912/527, loss: 0.03403482213616371
2023-01-21 10:47:50.759003: step: 916/527, loss: 0.0325227752327919
2023-01-21 10:47:51.934809: step: 920/527, loss: 0.04322462156414986
2023-01-21 10:47:53.071460: step: 924/527, loss: 0.01076965406537056
2023-01-21 10:47:54.219700: step: 928/527, loss: 0.057614900171756744
2023-01-21 10:47:55.361706: step: 932/527, loss: 0.21050433814525604
2023-01-21 10:47:56.465793: step: 936/527, loss: 0.014586257748305798
2023-01-21 10:47:57.599950: step: 940/527, loss: 0.06573820114135742
2023-01-21 10:47:58.735207: step: 944/527, loss: 0.28446221351623535
2023-01-21 10:47:59.864825: step: 948/527, loss: 0.035993292927742004
2023-01-21 10:48:00.997023: step: 952/527, loss: 0.08791609108448029
2023-01-21 10:48:02.138376: step: 956/527, loss: 0.023986244574189186
2023-01-21 10:48:03.272474: step: 960/527, loss: 0.05573015660047531
2023-01-21 10:48:04.382486: step: 964/527, loss: 0.11212072521448135
2023-01-21 10:48:05.530982: step: 968/527, loss: 0.08363381028175354
2023-01-21 10:48:06.669008: step: 972/527, loss: 0.009811019524931908
2023-01-21 10:48:07.805691: step: 976/527, loss: 0.03224515914916992
2023-01-21 10:48:08.913964: step: 980/527, loss: 0.018446924164891243
2023-01-21 10:48:10.049839: step: 984/527, loss: 0.06971893459558487
2023-01-21 10:48:11.209642: step: 988/527, loss: 0.04734306409955025
2023-01-21 10:48:12.347393: step: 992/527, loss: 0.020210934802889824
2023-01-21 10:48:13.474798: step: 996/527, loss: 0.08323478698730469
2023-01-21 10:48:14.605740: step: 1000/527, loss: 0.03024921379983425
2023-01-21 10:48:15.720880: step: 1004/527, loss: 0.017646122723817825
2023-01-21 10:48:16.868620: step: 1008/527, loss: 0.6664426922798157
2023-01-21 10:48:18.000720: step: 1012/527, loss: 0.019197940826416016
2023-01-21 10:48:19.147173: step: 1016/527, loss: 0.0016107559204101562
2023-01-21 10:48:20.256532: step: 1020/527, loss: 0.01998577080667019
2023-01-21 10:48:21.380171: step: 1024/527, loss: 0.049959614872932434
2023-01-21 10:48:22.537905: step: 1028/527, loss: 0.19796809554100037
2023-01-21 10:48:23.645922: step: 1032/527, loss: 0.03255829960107803
2023-01-21 10:48:24.812054: step: 1036/527, loss: 0.06451807171106339
2023-01-21 10:48:25.945418: step: 1040/527, loss: 0.003612697124481201
2023-01-21 10:48:27.068124: step: 1044/527, loss: 0.00577092170715332
2023-01-21 10:48:28.201987: step: 1048/527, loss: 0.07512908428907394
2023-01-21 10:48:29.329930: step: 1052/527, loss: 0.00082230573752895
2023-01-21 10:48:30.465598: step: 1056/527, loss: 0.06961727142333984
2023-01-21 10:48:31.603396: step: 1060/527, loss: 0.724550724029541
2023-01-21 10:48:32.747293: step: 1064/527, loss: 0.04136085510253906
2023-01-21 10:48:33.881463: step: 1068/527, loss: 0.07345481216907501
2023-01-21 10:48:35.005608: step: 1072/527, loss: 0.008295250125229359
2023-01-21 10:48:36.133818: step: 1076/527, loss: 0.04828685149550438
2023-01-21 10:48:37.300731: step: 1080/527, loss: 0.007529068272560835
2023-01-21 10:48:38.436067: step: 1084/527, loss: 0.004834223072975874
2023-01-21 10:48:39.549663: step: 1088/527, loss: 0.0325281135737896
2023-01-21 10:48:40.673904: step: 1092/527, loss: 0.1525605171918869
2023-01-21 10:48:41.865732: step: 1096/527, loss: 0.07212257385253906
2023-01-21 10:48:43.007103: step: 1100/527, loss: 0.05296153947710991
2023-01-21 10:48:44.145582: step: 1104/527, loss: 0.05182475969195366
2023-01-21 10:48:45.285350: step: 1108/527, loss: 0.00960607547312975
2023-01-21 10:48:46.425004: step: 1112/527, loss: 0.003126526018604636
2023-01-21 10:48:47.544900: step: 1116/527, loss: 0.03246298059821129
2023-01-21 10:48:48.693158: step: 1120/527, loss: 0.015102053061127663
2023-01-21 10:48:49.806930: step: 1124/527, loss: 0.009367561899125576
2023-01-21 10:48:50.931166: step: 1128/527, loss: 0.07105594873428345
2023-01-21 10:48:52.086191: step: 1132/527, loss: 0.03155851364135742
2023-01-21 10:48:53.185377: step: 1136/527, loss: 0.013150978833436966
2023-01-21 10:48:54.305433: step: 1140/527, loss: 0.035373687744140625
2023-01-21 10:48:55.421366: step: 1144/527, loss: 0.03545961529016495
2023-01-21 10:48:56.544528: step: 1148/527, loss: 0.020163822919130325
2023-01-21 10:48:57.695065: step: 1152/527, loss: 0.008792400360107422
2023-01-21 10:48:58.817916: step: 1156/527, loss: 0.19859322905540466
2023-01-21 10:48:59.947701: step: 1160/527, loss: 0.009354878216981888
2023-01-21 10:49:01.079897: step: 1164/527, loss: 0.020555878058075905
2023-01-21 10:49:02.246277: step: 1168/527, loss: 0.03493805229663849
2023-01-21 10:49:03.373216: step: 1172/527, loss: 0.20771950483322144
2023-01-21 10:49:04.514376: step: 1176/527, loss: 0.007491398137062788
2023-01-21 10:49:05.631415: step: 1180/527, loss: 0.008032703772187233
2023-01-21 10:49:06.746028: step: 1184/527, loss: 0.09722200036048889
2023-01-21 10:49:07.910297: step: 1188/527, loss: 0.5823045969009399
2023-01-21 10:49:09.011170: step: 1192/527, loss: 0.03454260900616646
2023-01-21 10:49:10.157027: step: 1196/527, loss: 0.01978473737835884
2023-01-21 10:49:11.296334: step: 1200/527, loss: 0.015386868268251419
2023-01-21 10:49:12.397162: step: 1204/527, loss: 0.005093097686767578
2023-01-21 10:49:13.540853: step: 1208/527, loss: 0.03135328367352486
2023-01-21 10:49:14.649075: step: 1212/527, loss: 0.032625965774059296
2023-01-21 10:49:15.779753: step: 1216/527, loss: 0.03395795822143555
2023-01-21 10:49:16.923318: step: 1220/527, loss: 0.005505943670868874
2023-01-21 10:49:18.069868: step: 1224/527, loss: 0.05205840989947319
2023-01-21 10:49:19.211989: step: 1228/527, loss: 0.0622289665043354
2023-01-21 10:49:20.341511: step: 1232/527, loss: 0.041649818420410156
2023-01-21 10:49:21.462093: step: 1236/527, loss: 0.013238049112260342
2023-01-21 10:49:22.643790: step: 1240/527, loss: 0.057599641382694244
2023-01-21 10:49:23.783173: step: 1244/527, loss: 0.012153577990829945
2023-01-21 10:49:24.901249: step: 1248/527, loss: 0.005779457278549671
2023-01-21 10:49:26.010409: step: 1252/527, loss: 0.04761715233325958
2023-01-21 10:49:27.126963: step: 1256/527, loss: 0.004703808110207319
2023-01-21 10:49:28.244404: step: 1260/527, loss: 0.057230666279792786
2023-01-21 10:49:29.374133: step: 1264/527, loss: 0.12317228317260742
2023-01-21 10:49:30.502780: step: 1268/527, loss: 0.02155895158648491
2023-01-21 10:49:31.607591: step: 1272/527, loss: 0.01656317710876465
2023-01-21 10:49:32.724839: step: 1276/527, loss: 0.2647598385810852
2023-01-21 10:49:33.866340: step: 1280/527, loss: 0.003936290740966797
2023-01-21 10:49:34.992964: step: 1284/527, loss: 0.01997528038918972
2023-01-21 10:49:36.135735: step: 1288/527, loss: 0.15362955629825592
2023-01-21 10:49:37.248735: step: 1292/527, loss: 0.013242244720458984
2023-01-21 10:49:38.378501: step: 1296/527, loss: 0.022014617919921875
2023-01-21 10:49:39.499110: step: 1300/527, loss: 0.038667868822813034
2023-01-21 10:49:40.641845: step: 1304/527, loss: 0.23070096969604492
2023-01-21 10:49:41.781343: step: 1308/527, loss: 0.08120041340589523
2023-01-21 10:49:42.926958: step: 1312/527, loss: 0.0051898956298828125
2023-01-21 10:49:44.071468: step: 1316/527, loss: 0.07891617715358734
2023-01-21 10:49:45.194913: step: 1320/527, loss: 0.0029237749986350536
2023-01-21 10:49:46.333914: step: 1324/527, loss: 0.02219226583838463
2023-01-21 10:49:47.468447: step: 1328/527, loss: 0.004751873202621937
2023-01-21 10:49:48.630526: step: 1332/527, loss: 0.009826470166444778
2023-01-21 10:49:49.801079: step: 1336/527, loss: 0.0006240367656573653
2023-01-21 10:49:50.957040: step: 1340/527, loss: 0.02756500244140625
2023-01-21 10:49:52.083342: step: 1344/527, loss: 0.03516406938433647
2023-01-21 10:49:53.207487: step: 1348/527, loss: 0.00695037841796875
2023-01-21 10:49:54.373964: step: 1352/527, loss: 0.024383973330259323
2023-01-21 10:49:55.486377: step: 1356/527, loss: 0.002441930817440152
2023-01-21 10:49:56.603600: step: 1360/527, loss: 0.0034097672905772924
2023-01-21 10:49:57.778361: step: 1364/527, loss: 0.0023772239219397306
2023-01-21 10:49:58.864616: step: 1368/527, loss: 0.03637723997235298
2023-01-21 10:49:59.994378: step: 1372/527, loss: 0.04784221947193146
2023-01-21 10:50:01.142916: step: 1376/527, loss: 0.0019015312427654862
2023-01-21 10:50:02.266450: step: 1380/527, loss: 0.27636557817459106
2023-01-21 10:50:03.390899: step: 1384/527, loss: 0.016938496381044388
2023-01-21 10:50:04.505961: step: 1388/527, loss: 0.0158888828009367
2023-01-21 10:50:05.620153: step: 1392/527, loss: 0.002195119857788086
2023-01-21 10:50:06.749597: step: 1396/527, loss: 0.006830978207290173
2023-01-21 10:50:07.900074: step: 1400/527, loss: 0.018581580370664597
2023-01-21 10:50:09.038019: step: 1404/527, loss: 0.016751576215028763
2023-01-21 10:50:10.177356: step: 1408/527, loss: 0.04961090162396431
2023-01-21 10:50:11.331587: step: 1412/527, loss: 0.06777743995189667
2023-01-21 10:50:12.482714: step: 1416/527, loss: 0.021691275760531425
2023-01-21 10:50:13.607938: step: 1420/527, loss: 0.29374438524246216
2023-01-21 10:50:14.698384: step: 1424/527, loss: 0.014352274127304554
2023-01-21 10:50:15.829837: step: 1428/527, loss: 0.03940172120928764
2023-01-21 10:50:16.920651: step: 1432/527, loss: 0.010601520538330078
2023-01-21 10:50:18.037983: step: 1436/527, loss: 0.6946795582771301
2023-01-21 10:50:19.206534: step: 1440/527, loss: 0.051271721720695496
2023-01-21 10:50:20.323225: step: 1444/527, loss: 0.049335479736328125
2023-01-21 10:50:21.450883: step: 1448/527, loss: 0.006586647126823664
2023-01-21 10:50:22.652916: step: 1452/527, loss: 0.11700229346752167
2023-01-21 10:50:23.797204: step: 1456/527, loss: 0.0926612839102745
2023-01-21 10:50:24.947021: step: 1460/527, loss: 0.07294356822967529
2023-01-21 10:50:26.102435: step: 1464/527, loss: 0.028957556933164597
2023-01-21 10:50:27.192836: step: 1468/527, loss: 0.02010526694357395
2023-01-21 10:50:28.355547: step: 1472/527, loss: 0.007172584533691406
2023-01-21 10:50:29.504506: step: 1476/527, loss: 0.04345288500189781
2023-01-21 10:50:30.636115: step: 1480/527, loss: 0.04587268829345703
2023-01-21 10:50:31.758697: step: 1484/527, loss: 0.005192121956497431
2023-01-21 10:50:32.874726: step: 1488/527, loss: 0.029194356873631477
2023-01-21 10:50:34.021671: step: 1492/527, loss: 0.022762108594179153
2023-01-21 10:50:35.166626: step: 1496/527, loss: 0.0031383514869958162
2023-01-21 10:50:36.293794: step: 1500/527, loss: 0.06720896065235138
2023-01-21 10:50:37.411393: step: 1504/527, loss: 0.03612027317285538
2023-01-21 10:50:38.530690: step: 1508/527, loss: 0.035645853728055954
2023-01-21 10:50:39.650498: step: 1512/527, loss: 0.06742215156555176
2023-01-21 10:50:40.762366: step: 1516/527, loss: 0.0030162811744958162
2023-01-21 10:50:41.877419: step: 1520/527, loss: 0.006982946302741766
2023-01-21 10:50:43.031086: step: 1524/527, loss: 0.10296420753002167
2023-01-21 10:50:44.142933: step: 1528/527, loss: 0.015073109418153763
2023-01-21 10:50:45.290522: step: 1532/527, loss: 0.007985735312104225
2023-01-21 10:50:46.412978: step: 1536/527, loss: 0.022669600322842598
2023-01-21 10:50:47.530901: step: 1540/527, loss: 0.004206657409667969
2023-01-21 10:50:48.655751: step: 1544/527, loss: 0.004276275634765625
2023-01-21 10:50:49.810354: step: 1548/527, loss: 0.015130805782973766
2023-01-21 10:50:50.973431: step: 1552/527, loss: 0.0010839462047442794
2023-01-21 10:50:52.103120: step: 1556/527, loss: 0.0053723338060081005
2023-01-21 10:50:53.238168: step: 1560/527, loss: 0.06706143170595169
2023-01-21 10:50:54.371187: step: 1564/527, loss: 0.03491344675421715
2023-01-21 10:50:55.481910: step: 1568/527, loss: 0.025592470541596413
2023-01-21 10:50:56.612588: step: 1572/527, loss: 0.005748748779296875
2023-01-21 10:50:57.743398: step: 1576/527, loss: 0.005154228303581476
2023-01-21 10:50:58.877576: step: 1580/527, loss: 0.01882152445614338
2023-01-21 10:50:59.980154: step: 1584/527, loss: 0.05542793497443199
2023-01-21 10:51:01.115612: step: 1588/527, loss: 0.013289166614413261
2023-01-21 10:51:02.247805: step: 1592/527, loss: 0.009035682305693626
2023-01-21 10:51:03.374087: step: 1596/527, loss: 0.005931854248046875
2023-01-21 10:51:04.491292: step: 1600/527, loss: 0.0595739409327507
2023-01-21 10:51:05.654993: step: 1604/527, loss: 0.05165109783411026
2023-01-21 10:51:06.778959: step: 1608/527, loss: 0.004200268071144819
2023-01-21 10:51:07.933223: step: 1612/527, loss: 0.08597545325756073
2023-01-21 10:51:09.045912: step: 1616/527, loss: 0.05741174519062042
2023-01-21 10:51:10.189339: step: 1620/527, loss: 0.0024640082847326994
2023-01-21 10:51:11.279223: step: 1624/527, loss: 0.01656198501586914
2023-01-21 10:51:12.408005: step: 1628/527, loss: 0.07062244415283203
2023-01-21 10:51:13.541796: step: 1632/527, loss: 0.0074443817138671875
2023-01-21 10:51:14.669962: step: 1636/527, loss: 0.16391582787036896
2023-01-21 10:51:15.810819: step: 1640/527, loss: 0.011956358328461647
2023-01-21 10:51:16.942066: step: 1644/527, loss: 0.2724836468696594
2023-01-21 10:51:18.058588: step: 1648/527, loss: 0.009928036481142044
2023-01-21 10:51:19.171803: step: 1652/527, loss: 0.04166774824261665
2023-01-21 10:51:20.321513: step: 1656/527, loss: 0.04453125223517418
2023-01-21 10:51:21.460008: step: 1660/527, loss: 0.00957473460584879
2023-01-21 10:51:22.566856: step: 1664/527, loss: 0.0003952026308979839
2023-01-21 10:51:23.675692: step: 1668/527, loss: 0.1094079464673996
2023-01-21 10:51:24.791544: step: 1672/527, loss: 0.004000854678452015
2023-01-21 10:51:25.947196: step: 1676/527, loss: 0.046807195991277695
2023-01-21 10:51:27.036024: step: 1680/527, loss: 0.0013566971756517887
2023-01-21 10:51:28.156931: step: 1684/527, loss: 0.04736337810754776
2023-01-21 10:51:29.307824: step: 1688/527, loss: 0.05189934000372887
2023-01-21 10:51:30.447939: step: 1692/527, loss: 0.00387496966868639
2023-01-21 10:51:31.581940: step: 1696/527, loss: 0.036879539489746094
2023-01-21 10:51:32.695034: step: 1700/527, loss: 0.0012104511260986328
2023-01-21 10:51:33.892712: step: 1704/527, loss: 0.061257075518369675
2023-01-21 10:51:35.032030: step: 1708/527, loss: 0.015995407477021217
2023-01-21 10:51:36.171901: step: 1712/527, loss: 0.03021831624209881
2023-01-21 10:51:37.275292: step: 1716/527, loss: 0.051409244537353516
2023-01-21 10:51:38.450990: step: 1720/527, loss: 0.021474361419677734
2023-01-21 10:51:39.592220: step: 1724/527, loss: 0.012036371044814587
2023-01-21 10:51:40.712159: step: 1728/527, loss: 0.007368040271103382
2023-01-21 10:51:41.795959: step: 1732/527, loss: 0.09078263491392136
2023-01-21 10:51:42.889374: step: 1736/527, loss: 0.0004385948122944683
2023-01-21 10:51:44.078914: step: 1740/527, loss: 0.028565645217895508
2023-01-21 10:51:45.203703: step: 1744/527, loss: 0.02421112172305584
2023-01-21 10:51:46.348926: step: 1748/527, loss: 0.07216129451990128
2023-01-21 10:51:47.497701: step: 1752/527, loss: 0.08621430397033691
2023-01-21 10:51:48.632418: step: 1756/527, loss: 0.017140865325927734
2023-01-21 10:51:49.781455: step: 1760/527, loss: 0.003023004624992609
2023-01-21 10:51:50.971727: step: 1764/527, loss: 0.06977224349975586
2023-01-21 10:51:52.093556: step: 1768/527, loss: 0.0037202835083007812
2023-01-21 10:51:53.251369: step: 1772/527, loss: 0.04111452028155327
2023-01-21 10:51:54.411369: step: 1776/527, loss: 0.008955001831054688
2023-01-21 10:51:55.526261: step: 1780/527, loss: 0.018445873633027077
2023-01-21 10:51:56.643265: step: 1784/527, loss: 0.06114606931805611
2023-01-21 10:51:57.776160: step: 1788/527, loss: 0.004582786466926336
2023-01-21 10:51:58.914523: step: 1792/527, loss: 0.05067186802625656
2023-01-21 10:52:00.046379: step: 1796/527, loss: 0.007205224130302668
2023-01-21 10:52:01.179440: step: 1800/527, loss: 0.05734921991825104
2023-01-21 10:52:02.297848: step: 1804/527, loss: 0.025539016351103783
2023-01-21 10:52:03.410215: step: 1808/527, loss: 0.046971406787633896
2023-01-21 10:52:04.546574: step: 1812/527, loss: 0.6912172436714172
2023-01-21 10:52:05.674752: step: 1816/527, loss: 0.06328163295984268
2023-01-21 10:52:06.799621: step: 1820/527, loss: 0.2096502184867859
2023-01-21 10:52:07.940251: step: 1824/527, loss: 0.009600304998457432
2023-01-21 10:52:09.067019: step: 1828/527, loss: 0.0025440691970288754
2023-01-21 10:52:10.211965: step: 1832/527, loss: 0.06278133392333984
2023-01-21 10:52:11.351561: step: 1836/527, loss: 0.012899207882583141
2023-01-21 10:52:12.503561: step: 1840/527, loss: 0.032385826110839844
2023-01-21 10:52:13.605177: step: 1844/527, loss: 0.036482714116573334
2023-01-21 10:52:14.742645: step: 1848/527, loss: 0.002086067106574774
2023-01-21 10:52:15.862565: step: 1852/527, loss: 0.041127681732177734
2023-01-21 10:52:17.003744: step: 1856/527, loss: 0.13921909034252167
2023-01-21 10:52:18.146384: step: 1860/527, loss: 0.04081106185913086
2023-01-21 10:52:19.303276: step: 1864/527, loss: 0.04165401682257652
2023-01-21 10:52:20.443922: step: 1868/527, loss: 0.0629599541425705
2023-01-21 10:52:21.565316: step: 1872/527, loss: 0.01438918150961399
2023-01-21 10:52:22.662989: step: 1876/527, loss: 0.02493257448077202
2023-01-21 10:52:23.804751: step: 1880/527, loss: 0.0367732048034668
2023-01-21 10:52:24.909776: step: 1884/527, loss: 0.0055550173856318
2023-01-21 10:52:26.033383: step: 1888/527, loss: 0.0018085002666339278
2023-01-21 10:52:27.136084: step: 1892/527, loss: 0.02495403401553631
2023-01-21 10:52:28.307204: step: 1896/527, loss: 0.046590615063905716
2023-01-21 10:52:29.438731: step: 1900/527, loss: 0.00687255896627903
2023-01-21 10:52:30.570045: step: 1904/527, loss: 0.6215629577636719
2023-01-21 10:52:31.700087: step: 1908/527, loss: 0.039576247334480286
2023-01-21 10:52:32.793224: step: 1912/527, loss: 0.003080749651417136
2023-01-21 10:52:33.935650: step: 1916/527, loss: 0.012486744672060013
2023-01-21 10:52:35.087420: step: 1920/527, loss: 0.041993334889411926
2023-01-21 10:52:36.201304: step: 1924/527, loss: 0.006340312771499157
2023-01-21 10:52:37.350448: step: 1928/527, loss: 0.09931173175573349
2023-01-21 10:52:38.479770: step: 1932/527, loss: 0.1064685806632042
2023-01-21 10:52:39.600051: step: 1936/527, loss: 0.07896146178245544
2023-01-21 10:52:40.731683: step: 1940/527, loss: 0.05091896280646324
2023-01-21 10:52:41.861799: step: 1944/527, loss: 0.02133970335125923
2023-01-21 10:52:42.975589: step: 1948/527, loss: 0.02213573455810547
2023-01-21 10:52:44.145354: step: 1952/527, loss: 0.031040765345096588
2023-01-21 10:52:45.230159: step: 1956/527, loss: 0.014957714825868607
2023-01-21 10:52:46.349482: step: 1960/527, loss: 0.03968362882733345
2023-01-21 10:52:47.485383: step: 1964/527, loss: 0.03171835094690323
2023-01-21 10:52:48.598984: step: 1968/527, loss: 0.006890201475471258
2023-01-21 10:52:49.722674: step: 1972/527, loss: 0.015164995566010475
2023-01-21 10:52:50.847036: step: 1976/527, loss: 0.01812286488711834
2023-01-21 10:52:51.974387: step: 1980/527, loss: 0.017806529998779297
2023-01-21 10:52:53.118279: step: 1984/527, loss: 0.058237459510564804
2023-01-21 10:52:54.214699: step: 1988/527, loss: 0.01932344399392605
2023-01-21 10:52:55.379457: step: 1992/527, loss: 0.009842300787568092
2023-01-21 10:52:56.512060: step: 1996/527, loss: 0.025014685466885567
2023-01-21 10:52:57.630155: step: 2000/527, loss: 0.012192345224320889
2023-01-21 10:52:58.749032: step: 2004/527, loss: 0.02413005754351616
2023-01-21 10:52:59.884723: step: 2008/527, loss: 0.0463964082300663
2023-01-21 10:53:00.988974: step: 2012/527, loss: 0.06092128902673721
2023-01-21 10:53:02.100853: step: 2016/527, loss: 0.03946990892291069
2023-01-21 10:53:03.281008: step: 2020/527, loss: 0.0009015083778649569
2023-01-21 10:53:04.379765: step: 2024/527, loss: 0.009719084948301315
2023-01-21 10:53:05.505788: step: 2028/527, loss: 0.06445427238941193
2023-01-21 10:53:06.627013: step: 2032/527, loss: 0.08793125301599503
2023-01-21 10:53:07.754204: step: 2036/527, loss: 0.10137882083654404
2023-01-21 10:53:08.914027: step: 2040/527, loss: 0.05763416364789009
2023-01-21 10:53:10.023513: step: 2044/527, loss: 0.5947321653366089
2023-01-21 10:53:11.137227: step: 2048/527, loss: 0.05582108721137047
2023-01-21 10:53:12.265314: step: 2052/527, loss: 0.0015582561027258635
2023-01-21 10:53:13.392953: step: 2056/527, loss: 0.007390642538666725
2023-01-21 10:53:14.507185: step: 2060/527, loss: 0.01080179214477539
2023-01-21 10:53:15.658471: step: 2064/527, loss: 0.008870887570083141
2023-01-21 10:53:16.797055: step: 2068/527, loss: 0.010660935193300247
2023-01-21 10:53:17.922654: step: 2072/527, loss: 0.0019012928241863847
2023-01-21 10:53:19.066110: step: 2076/527, loss: 0.02033062092959881
2023-01-21 10:53:20.226846: step: 2080/527, loss: 0.04858198016881943
2023-01-21 10:53:21.333145: step: 2084/527, loss: 0.006595945917069912
2023-01-21 10:53:22.467410: step: 2088/527, loss: 0.06519432365894318
2023-01-21 10:53:23.616718: step: 2092/527, loss: 0.07375431060791016
2023-01-21 10:53:24.735343: step: 2096/527, loss: 0.024370575323700905
2023-01-21 10:53:25.817302: step: 2100/527, loss: 0.0045320987701416016
2023-01-21 10:53:26.927649: step: 2104/527, loss: 0.03608770668506622
2023-01-21 10:53:28.048284: step: 2108/527, loss: 0.007403564173728228
==================================================
Loss: 0.052
--------------------
Dev: {'event': {'p': 0.6024716786817713, 'r': 0.7789613848202397, 'f1': 0.6794425087108013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13}
Test: {'event': {'p': 0.6284801460520311, 'r': 0.7868571428571428, 'f1': 0.6988074092869829}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13}
Chinese: {'event': {'p': 0.5454545454545454, 'r': 0.8888888888888888, 'f1': 0.676056338028169}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13}
Korean: {'event': {'p': 0.6071428571428571, 'r': 0.5396825396825397, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13}
Russian: {'event': {'p': 0.43902439024390244, 'r': 0.5, 'f1': 0.4675324675324676}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 10:54:11.427927: step: 4/527, loss: 0.025342464447021484
2023-01-21 10:54:12.569081: step: 8/527, loss: 0.015662766993045807
2023-01-21 10:54:13.710311: step: 12/527, loss: 0.02370300330221653
2023-01-21 10:54:14.839162: step: 16/527, loss: 0.03798742592334747
2023-01-21 10:54:15.980437: step: 20/527, loss: 0.011620378121733665
2023-01-21 10:54:17.117771: step: 24/527, loss: 0.04960274696350098
2023-01-21 10:54:18.202105: step: 28/527, loss: 0.0037850383669137955
2023-01-21 10:54:19.337008: step: 32/527, loss: 0.03542933613061905
2023-01-21 10:54:20.491076: step: 36/527, loss: 0.04210076108574867
2023-01-21 10:54:21.616639: step: 40/527, loss: 0.03690643236041069
2023-01-21 10:54:22.762158: step: 44/527, loss: 0.0018833160866051912
2023-01-21 10:54:23.881872: step: 48/527, loss: 0.001009273575618863
2023-01-21 10:54:25.000084: step: 52/527, loss: 0.034002162516117096
2023-01-21 10:54:26.126842: step: 56/527, loss: 0.008019638247787952
2023-01-21 10:54:27.249543: step: 60/527, loss: 0.0033050538040697575
2023-01-21 10:54:28.374571: step: 64/527, loss: 0.05381298065185547
2023-01-21 10:54:29.491275: step: 68/527, loss: 0.0032682418823242188
2023-01-21 10:54:30.600459: step: 72/527, loss: 0.034348104149103165
2023-01-21 10:54:31.733613: step: 76/527, loss: 0.018347548320889473
2023-01-21 10:54:32.853319: step: 80/527, loss: 0.007658672519028187
2023-01-21 10:54:33.986492: step: 84/527, loss: 0.012154294177889824
2023-01-21 10:54:35.138525: step: 88/527, loss: 0.005300331395119429
2023-01-21 10:54:36.260352: step: 92/527, loss: 0.004478359594941139
2023-01-21 10:54:37.391230: step: 96/527, loss: 0.005043506622314453
2023-01-21 10:54:38.532271: step: 100/527, loss: 0.006303215399384499
2023-01-21 10:54:39.636023: step: 104/527, loss: 0.2584305703639984
2023-01-21 10:54:40.773325: step: 108/527, loss: 0.01438894309103489
2023-01-21 10:54:41.873399: step: 112/527, loss: 0.012206220999360085
2023-01-21 10:54:42.993576: step: 116/527, loss: 0.009972000494599342
2023-01-21 10:54:44.138261: step: 120/527, loss: 0.059607695788145065
2023-01-21 10:54:45.242592: step: 124/527, loss: 0.04517402499914169
2023-01-21 10:54:46.364626: step: 128/527, loss: 0.010956143960356712
2023-01-21 10:54:47.489947: step: 132/527, loss: 0.05373172461986542
2023-01-21 10:54:48.603872: step: 136/527, loss: 0.05832987278699875
2023-01-21 10:54:49.715798: step: 140/527, loss: 0.03638887405395508
2023-01-21 10:54:50.810367: step: 144/527, loss: 0.026297666132450104
2023-01-21 10:54:51.938230: step: 148/527, loss: 0.005062866490334272
2023-01-21 10:54:53.055525: step: 152/527, loss: 0.026729965582489967
2023-01-21 10:54:54.177835: step: 156/527, loss: 0.0020202635787427425
2023-01-21 10:54:55.310742: step: 160/527, loss: 0.00012168884131824598
2023-01-21 10:54:56.461689: step: 164/527, loss: 0.02196211740374565
2023-01-21 10:54:57.566300: step: 168/527, loss: 0.0033535019028931856
2023-01-21 10:54:58.674472: step: 172/527, loss: 0.0831810012459755
2023-01-21 10:54:59.818475: step: 176/527, loss: 0.06321487575769424
2023-01-21 10:55:00.940577: step: 180/527, loss: 0.008753872476518154
2023-01-21 10:55:02.065933: step: 184/527, loss: 0.023058701306581497
2023-01-21 10:55:03.189353: step: 188/527, loss: 0.1346489042043686
2023-01-21 10:55:04.313985: step: 192/527, loss: 0.0027726651169359684
2023-01-21 10:55:05.448174: step: 196/527, loss: 0.07787355780601501
2023-01-21 10:55:06.575293: step: 200/527, loss: 0.029454898089170456
2023-01-21 10:55:07.755185: step: 204/527, loss: 0.010012245737016201
2023-01-21 10:55:08.903253: step: 208/527, loss: 0.1282539665699005
2023-01-21 10:55:10.021607: step: 212/527, loss: 0.04250774532556534
2023-01-21 10:55:11.163305: step: 216/527, loss: 0.00432662945240736
2023-01-21 10:55:12.286627: step: 220/527, loss: 0.004070854280143976
2023-01-21 10:55:13.406189: step: 224/527, loss: 0.0023965835571289062
2023-01-21 10:55:14.524704: step: 228/527, loss: 0.050461579114198685
2023-01-21 10:55:15.653806: step: 232/527, loss: 0.6652324795722961
2023-01-21 10:55:16.777907: step: 236/527, loss: 0.10006442666053772
2023-01-21 10:55:17.916340: step: 240/527, loss: 0.011378193274140358
2023-01-21 10:55:19.047281: step: 244/527, loss: 0.011311912909150124
2023-01-21 10:55:20.234941: step: 248/527, loss: 0.003432846162468195
2023-01-21 10:55:21.366255: step: 252/527, loss: 0.6147485375404358
2023-01-21 10:55:22.515140: step: 256/527, loss: 0.004846381954848766
2023-01-21 10:55:23.645196: step: 260/527, loss: 0.004418564029037952
2023-01-21 10:55:24.786934: step: 264/527, loss: 0.4266657531261444
2023-01-21 10:55:25.885367: step: 268/527, loss: 0.04324007034301758
2023-01-21 10:55:26.996449: step: 272/527, loss: 0.001559400581754744
2023-01-21 10:55:28.128076: step: 276/527, loss: 0.030055999755859375
2023-01-21 10:55:29.275672: step: 280/527, loss: 0.46589404344558716
2023-01-21 10:55:30.401992: step: 284/527, loss: 0.018753718584775925
2023-01-21 10:55:31.546182: step: 288/527, loss: 0.038478851318359375
2023-01-21 10:55:32.665850: step: 292/527, loss: 0.015268707647919655
2023-01-21 10:55:33.797245: step: 296/527, loss: 0.028193378821015358
2023-01-21 10:55:34.934273: step: 300/527, loss: 0.019567109644412994
2023-01-21 10:55:36.073390: step: 304/527, loss: 0.04838895797729492
2023-01-21 10:55:37.229056: step: 308/527, loss: 0.03289680555462837
2023-01-21 10:55:38.375649: step: 312/527, loss: 0.005839347839355469
2023-01-21 10:55:39.507831: step: 316/527, loss: 0.005610847845673561
2023-01-21 10:55:40.635333: step: 320/527, loss: 0.02583608590066433
2023-01-21 10:55:41.743259: step: 324/527, loss: 7.62939453125e-05
2023-01-21 10:55:42.900079: step: 328/527, loss: 0.03259086608886719
2023-01-21 10:55:44.019767: step: 332/527, loss: 0.07112522423267365
2023-01-21 10:55:45.133075: step: 336/527, loss: 0.0004938125493936241
2023-01-21 10:55:46.352825: step: 340/527, loss: 0.0396239273250103
2023-01-21 10:55:47.485553: step: 344/527, loss: 0.03540802001953125
2023-01-21 10:55:48.640690: step: 348/527, loss: 0.014509011059999466
2023-01-21 10:55:49.769059: step: 352/527, loss: 0.004350089933723211
2023-01-21 10:55:50.915048: step: 356/527, loss: 0.006381416693329811
2023-01-21 10:55:52.060096: step: 360/527, loss: 0.1478293389081955
2023-01-21 10:55:53.198568: step: 364/527, loss: 0.05781584233045578
2023-01-21 10:55:54.350432: step: 368/527, loss: 0.015013408847153187
2023-01-21 10:55:55.482589: step: 372/527, loss: 0.014759588986635208
2023-01-21 10:55:56.598517: step: 376/527, loss: 0.012675666250288486
2023-01-21 10:55:57.716218: step: 380/527, loss: 0.3324183225631714
2023-01-21 10:55:58.863960: step: 384/527, loss: 0.006377982906997204
2023-01-21 10:55:59.972568: step: 388/527, loss: 0.017656756564974785
2023-01-21 10:56:01.104057: step: 392/527, loss: 0.11679382622241974
2023-01-21 10:56:02.231181: step: 396/527, loss: 0.03275261074304581
2023-01-21 10:56:03.371287: step: 400/527, loss: 6.4849853515625e-05
2023-01-21 10:56:04.510552: step: 404/527, loss: 0.00860381219536066
2023-01-21 10:56:05.617896: step: 408/527, loss: 0.07595701515674591
2023-01-21 10:56:06.757262: step: 412/527, loss: 0.010304831899702549
2023-01-21 10:56:07.861381: step: 416/527, loss: 1.087188684323337e-05
2023-01-21 10:56:08.973263: step: 420/527, loss: 0.12250528484582901
2023-01-21 10:56:10.082801: step: 424/527, loss: 0.006567221600562334
2023-01-21 10:56:11.195080: step: 428/527, loss: 0.0013615608913823962
2023-01-21 10:56:12.297680: step: 432/527, loss: 0.00019998550124000758
2023-01-21 10:56:13.473029: step: 436/527, loss: 0.018865682184696198
2023-01-21 10:56:14.610106: step: 440/527, loss: 0.04039287567138672
2023-01-21 10:56:15.756370: step: 444/527, loss: 0.03710479661822319
2023-01-21 10:56:16.898059: step: 448/527, loss: 0.30073174834251404
2023-01-21 10:56:18.008665: step: 452/527, loss: 0.009966278448700905
2023-01-21 10:56:19.122087: step: 456/527, loss: 0.005640125367790461
2023-01-21 10:56:20.239052: step: 460/527, loss: 0.04480600357055664
2023-01-21 10:56:21.379997: step: 464/527, loss: 0.00991830788552761
2023-01-21 10:56:22.534781: step: 468/527, loss: 0.020785141736268997
2023-01-21 10:56:23.656324: step: 472/527, loss: 0.016439056023955345
2023-01-21 10:56:24.808147: step: 476/527, loss: 0.06115417554974556
2023-01-21 10:56:25.931975: step: 480/527, loss: 0.0018594504799693823
2023-01-21 10:56:27.048748: step: 484/527, loss: 0.015018082223832607
2023-01-21 10:56:28.190805: step: 488/527, loss: 0.13724756240844727
2023-01-21 10:56:29.324234: step: 492/527, loss: 0.049356844276189804
2023-01-21 10:56:30.436881: step: 496/527, loss: 0.001153802964836359
2023-01-21 10:56:31.547244: step: 500/527, loss: 0.014649583026766777
2023-01-21 10:56:32.666963: step: 504/527, loss: 0.059430789202451706
2023-01-21 10:56:33.811892: step: 508/527, loss: 0.012227917090058327
2023-01-21 10:56:34.922349: step: 512/527, loss: 0.03984313830733299
2023-01-21 10:56:36.068321: step: 516/527, loss: 0.015184975229203701
2023-01-21 10:56:37.222412: step: 520/527, loss: 0.0008680344326421618
2023-01-21 10:56:38.355181: step: 524/527, loss: 0.047830674797296524
2023-01-21 10:56:39.485197: step: 528/527, loss: 0.029259586706757545
2023-01-21 10:56:40.640199: step: 532/527, loss: 0.004869365599006414
2023-01-21 10:56:41.805224: step: 536/527, loss: 0.0010447502136230469
2023-01-21 10:56:42.905450: step: 540/527, loss: 0.004124260041862726
2023-01-21 10:56:44.041161: step: 544/527, loss: 0.11318397521972656
2023-01-21 10:56:45.160912: step: 548/527, loss: 0.03995170816779137
2023-01-21 10:56:46.304348: step: 552/527, loss: 0.01352317351847887
2023-01-21 10:56:47.438276: step: 556/527, loss: 0.22084417939186096
2023-01-21 10:56:48.565397: step: 560/527, loss: 0.07673053443431854
2023-01-21 10:56:49.686377: step: 564/527, loss: 0.00282459263689816
2023-01-21 10:56:50.829146: step: 568/527, loss: 0.01804647594690323
2023-01-21 10:56:51.968541: step: 572/527, loss: 0.005524539854377508
2023-01-21 10:56:53.067625: step: 576/527, loss: 0.0006015778053551912
2023-01-21 10:56:54.222461: step: 580/527, loss: 0.08562207967042923
2023-01-21 10:56:55.368186: step: 584/527, loss: 0.026229478418827057
2023-01-21 10:56:56.511385: step: 588/527, loss: 0.09473800659179688
2023-01-21 10:56:57.626535: step: 592/527, loss: 0.048650264739990234
2023-01-21 10:56:58.749514: step: 596/527, loss: 0.0312257781624794
2023-01-21 10:56:59.884455: step: 600/527, loss: 0.009479904547333717
2023-01-21 10:57:01.032613: step: 604/527, loss: 0.05222473293542862
2023-01-21 10:57:02.159158: step: 608/527, loss: 0.003637027693912387
2023-01-21 10:57:03.284845: step: 612/527, loss: 0.009443949908018112
2023-01-21 10:57:04.417046: step: 616/527, loss: 0.003423309186473489
2023-01-21 10:57:05.617372: step: 620/527, loss: 0.016902543604373932
2023-01-21 10:57:06.721114: step: 624/527, loss: 0.06863898783922195
2023-01-21 10:57:07.878159: step: 628/527, loss: 0.44794130325317383
2023-01-21 10:57:08.997482: step: 632/527, loss: 0.006907748989760876
2023-01-21 10:57:10.140648: step: 636/527, loss: 0.030873488634824753
2023-01-21 10:57:11.256238: step: 640/527, loss: 0.03454799950122833
2023-01-21 10:57:12.380131: step: 644/527, loss: 0.010816765949130058
2023-01-21 10:57:13.533562: step: 648/527, loss: 0.02709946781396866
2023-01-21 10:57:14.655627: step: 652/527, loss: 0.007910347543656826
2023-01-21 10:57:15.810711: step: 656/527, loss: 0.10797033458948135
2023-01-21 10:57:16.909771: step: 660/527, loss: 0.01495513878762722
2023-01-21 10:57:18.047171: step: 664/527, loss: 0.014375876635313034
2023-01-21 10:57:19.180913: step: 668/527, loss: 0.00589065533131361
2023-01-21 10:57:20.318352: step: 672/527, loss: 0.03317613527178764
2023-01-21 10:57:21.470200: step: 676/527, loss: 0.15463848412036896
2023-01-21 10:57:22.611545: step: 680/527, loss: 0.013688469305634499
2023-01-21 10:57:23.768887: step: 684/527, loss: 0.0020911216270178556
2023-01-21 10:57:24.865974: step: 688/527, loss: 0.016767119988799095
2023-01-21 10:57:25.958519: step: 692/527, loss: 0.022665690630674362
2023-01-21 10:57:27.122431: step: 696/527, loss: 0.025562833994627
2023-01-21 10:57:28.281020: step: 700/527, loss: 0.026140881702303886
2023-01-21 10:57:29.395666: step: 704/527, loss: 0.05078735202550888
2023-01-21 10:57:30.527246: step: 708/527, loss: 0.42712095379829407
2023-01-21 10:57:31.659972: step: 712/527, loss: 0.0013141154777258635
2023-01-21 10:57:32.793695: step: 716/527, loss: 0.020473767071962357
2023-01-21 10:57:33.914306: step: 720/527, loss: 0.022878218442201614
2023-01-21 10:57:35.039215: step: 724/527, loss: 0.012124061584472656
2023-01-21 10:57:36.175685: step: 728/527, loss: 0.07249081879854202
2023-01-21 10:57:37.307048: step: 732/527, loss: 0.07187262177467346
2023-01-21 10:57:38.447280: step: 736/527, loss: 0.16689807176589966
2023-01-21 10:57:39.570521: step: 740/527, loss: 0.011294281110167503
2023-01-21 10:57:40.703416: step: 744/527, loss: 0.02014617994427681
2023-01-21 10:57:41.823982: step: 748/527, loss: 0.009103154763579369
2023-01-21 10:57:42.949914: step: 752/527, loss: 0.020002175122499466
2023-01-21 10:57:44.100786: step: 756/527, loss: 0.042470645159482956
2023-01-21 10:57:45.218402: step: 760/527, loss: 0.004442119505256414
2023-01-21 10:57:46.406836: step: 764/527, loss: 0.009454727172851562
2023-01-21 10:57:47.520684: step: 768/527, loss: 0.01426792237907648
2023-01-21 10:57:48.666369: step: 772/527, loss: 0.018658161163330078
2023-01-21 10:57:49.792769: step: 776/527, loss: 0.0053310394287109375
2023-01-21 10:57:50.917700: step: 780/527, loss: 0.10296133160591125
2023-01-21 10:57:52.097373: step: 784/527, loss: 0.04624900966882706
2023-01-21 10:57:53.250131: step: 788/527, loss: 0.021212387830018997
2023-01-21 10:57:54.361663: step: 792/527, loss: 0.02985220029950142
2023-01-21 10:57:55.480201: step: 796/527, loss: 0.06775122135877609
2023-01-21 10:57:56.636194: step: 800/527, loss: 0.01259927824139595
2023-01-21 10:57:57.770231: step: 804/527, loss: 0.004502058029174805
2023-01-21 10:57:58.900544: step: 808/527, loss: 0.017753221094608307
2023-01-21 10:58:00.018115: step: 812/527, loss: 0.025107765570282936
2023-01-21 10:58:01.177384: step: 816/527, loss: 0.005241298582404852
2023-01-21 10:58:02.311440: step: 820/527, loss: 0.03649468347430229
2023-01-21 10:58:03.437420: step: 824/527, loss: 0.0032578466925770044
2023-01-21 10:58:04.586971: step: 828/527, loss: 0.021381378173828125
2023-01-21 10:58:05.730654: step: 832/527, loss: 0.03470878675580025
2023-01-21 10:58:06.877573: step: 836/527, loss: 0.007831478491425514
2023-01-21 10:58:08.013059: step: 840/527, loss: 0.030228901654481888
2023-01-21 10:58:09.152297: step: 844/527, loss: 0.03634333983063698
2023-01-21 10:58:10.282468: step: 848/527, loss: 0.024999428540468216
2023-01-21 10:58:11.406683: step: 852/527, loss: 0.02039337158203125
2023-01-21 10:58:12.540674: step: 856/527, loss: 0.014800739474594593
2023-01-21 10:58:13.665487: step: 860/527, loss: 0.05066204071044922
2023-01-21 10:58:14.795810: step: 864/527, loss: 0.00866842269897461
2023-01-21 10:58:15.918549: step: 868/527, loss: 0.011360406875610352
2023-01-21 10:58:17.052098: step: 872/527, loss: 0.028253793716430664
2023-01-21 10:58:18.154363: step: 876/527, loss: 0.00018615722365211695
2023-01-21 10:58:19.283109: step: 880/527, loss: 0.08463191986083984
2023-01-21 10:58:20.429110: step: 884/527, loss: 0.0015666962135583162
2023-01-21 10:58:21.554231: step: 888/527, loss: 0.00656938599422574
2023-01-21 10:58:22.666041: step: 892/527, loss: 0.009884834289550781
2023-01-21 10:58:23.790023: step: 896/527, loss: 0.006913280580192804
2023-01-21 10:58:24.915993: step: 900/527, loss: 0.0542207732796669
2023-01-21 10:58:26.052512: step: 904/527, loss: 0.013064814731478691
2023-01-21 10:58:27.168368: step: 908/527, loss: 0.01010751724243164
2023-01-21 10:58:28.282229: step: 912/527, loss: 0.05079488828778267
2023-01-21 10:58:29.451820: step: 916/527, loss: 0.0058464049361646175
2023-01-21 10:58:30.591514: step: 920/527, loss: 0.01996450498700142
2023-01-21 10:58:31.738220: step: 924/527, loss: 0.07865677028894424
2023-01-21 10:58:32.833631: step: 928/527, loss: 0.021787548437714577
2023-01-21 10:58:33.970863: step: 932/527, loss: 0.015448665246367455
2023-01-21 10:58:35.105632: step: 936/527, loss: 0.0018733978504315019
2023-01-21 10:58:36.232447: step: 940/527, loss: 0.05511321872472763
2023-01-21 10:58:37.368212: step: 944/527, loss: 0.0708896666765213
2023-01-21 10:58:38.484317: step: 948/527, loss: 0.05402517318725586
2023-01-21 10:58:39.623073: step: 952/527, loss: 0.03214244917035103
2023-01-21 10:58:40.758258: step: 956/527, loss: 0.05144014582037926
2023-01-21 10:58:41.870849: step: 960/527, loss: 0.004785824101418257
2023-01-21 10:58:42.980125: step: 964/527, loss: 0.014788723550736904
2023-01-21 10:58:44.126012: step: 968/527, loss: 0.019421331584453583
2023-01-21 10:58:45.256203: step: 972/527, loss: 0.01750476472079754
2023-01-21 10:58:46.407998: step: 976/527, loss: 0.007056832779198885
2023-01-21 10:58:47.511953: step: 980/527, loss: 0.012920666486024857
2023-01-21 10:58:48.642339: step: 984/527, loss: 0.0067899227142333984
2023-01-21 10:58:49.813624: step: 988/527, loss: 0.0632476806640625
2023-01-21 10:58:50.980249: step: 992/527, loss: 0.02522573620080948
2023-01-21 10:58:52.092579: step: 996/527, loss: 0.0188446044921875
2023-01-21 10:58:53.260341: step: 1000/527, loss: 0.045894622802734375
2023-01-21 10:58:54.380557: step: 1004/527, loss: 0.0014340400230139494
2023-01-21 10:58:55.509173: step: 1008/527, loss: 0.007226801011711359
2023-01-21 10:58:56.654042: step: 1012/527, loss: 0.05158586800098419
2023-01-21 10:58:57.758535: step: 1016/527, loss: 0.0014204978942871094
2023-01-21 10:58:58.886506: step: 1020/527, loss: 0.0084984777495265
2023-01-21 10:59:00.013182: step: 1024/527, loss: 0.01755690574645996
2023-01-21 10:59:01.141874: step: 1028/527, loss: 0.07794008404016495
2023-01-21 10:59:02.284374: step: 1032/527, loss: 0.029554082080721855
2023-01-21 10:59:03.410889: step: 1036/527, loss: 0.029317475855350494
2023-01-21 10:59:04.551692: step: 1040/527, loss: 0.0023838996421545744
2023-01-21 10:59:05.676406: step: 1044/527, loss: 0.03493490070104599
2023-01-21 10:59:06.795501: step: 1048/527, loss: 0.0038268091157078743
2023-01-21 10:59:07.885993: step: 1052/527, loss: 0.009655284695327282
2023-01-21 10:59:09.023028: step: 1056/527, loss: 0.0029845237731933594
2023-01-21 10:59:10.187170: step: 1060/527, loss: 0.013245295733213425
2023-01-21 10:59:11.329898: step: 1064/527, loss: 0.00010099411883857101
2023-01-21 10:59:12.470083: step: 1068/527, loss: 0.021912433207035065
2023-01-21 10:59:13.587818: step: 1072/527, loss: 0.00048253536806441844
2023-01-21 10:59:14.757613: step: 1076/527, loss: 0.10121440887451172
2023-01-21 10:59:15.860497: step: 1080/527, loss: 0.05427970737218857
2023-01-21 10:59:16.991420: step: 1084/527, loss: 0.006516552530229092
2023-01-21 10:59:18.118946: step: 1088/527, loss: 0.016476059332489967
2023-01-21 10:59:19.248795: step: 1092/527, loss: 0.004337024874985218
2023-01-21 10:59:20.399174: step: 1096/527, loss: 0.030361175537109375
2023-01-21 10:59:21.542011: step: 1100/527, loss: 0.018692590296268463
2023-01-21 10:59:22.689280: step: 1104/527, loss: 0.000209808349609375
2023-01-21 10:59:23.823828: step: 1108/527, loss: 0.038996122777462006
2023-01-21 10:59:24.960709: step: 1112/527, loss: 0.25211867690086365
2023-01-21 10:59:26.124929: step: 1116/527, loss: 0.02198343351483345
2023-01-21 10:59:27.264963: step: 1120/527, loss: 0.05424303933978081
2023-01-21 10:59:28.405783: step: 1124/527, loss: 0.19440126419067383
2023-01-21 10:59:29.527676: step: 1128/527, loss: 0.02487773820757866
2023-01-21 10:59:30.660046: step: 1132/527, loss: 0.15693818032741547
2023-01-21 10:59:31.779824: step: 1136/527, loss: 0.0010755539406090975
2023-01-21 10:59:32.899518: step: 1140/527, loss: 0.01673154905438423
2023-01-21 10:59:34.015012: step: 1144/527, loss: 0.07730689644813538
2023-01-21 10:59:35.166946: step: 1148/527, loss: 0.0026268004439771175
2023-01-21 10:59:36.325926: step: 1152/527, loss: 0.12117500603199005
2023-01-21 10:59:37.463538: step: 1156/527, loss: 0.0006269455188885331
2023-01-21 10:59:38.594112: step: 1160/527, loss: 0.0002209663507528603
2023-01-21 10:59:39.697079: step: 1164/527, loss: 0.15881307423114777
2023-01-21 10:59:40.840478: step: 1168/527, loss: 0.021887589246034622
2023-01-21 10:59:41.983921: step: 1172/527, loss: 0.0023250579833984375
2023-01-21 10:59:43.126389: step: 1176/527, loss: 0.03418560326099396
2023-01-21 10:59:44.255338: step: 1180/527, loss: 0.0031715393997728825
2023-01-21 10:59:45.406338: step: 1184/527, loss: 0.0027289390563964844
2023-01-21 10:59:46.548071: step: 1188/527, loss: 0.019376277923583984
2023-01-21 10:59:47.696563: step: 1192/527, loss: 0.07493028044700623
2023-01-21 10:59:48.828917: step: 1196/527, loss: 0.0033516883850097656
2023-01-21 10:59:49.973368: step: 1200/527, loss: 0.0548737533390522
2023-01-21 10:59:51.149926: step: 1204/527, loss: 0.008812451735138893
2023-01-21 10:59:52.255319: step: 1208/527, loss: 0.12735100090503693
2023-01-21 10:59:53.422087: step: 1212/527, loss: 0.06548700481653214
2023-01-21 10:59:54.583047: step: 1216/527, loss: 0.010318947024643421
2023-01-21 10:59:55.745123: step: 1220/527, loss: 0.024636270478367805
2023-01-21 10:59:56.869107: step: 1224/527, loss: 0.04885859787464142
2023-01-21 10:59:57.980944: step: 1228/527, loss: 0.021091079339385033
2023-01-21 10:59:59.102271: step: 1232/527, loss: 0.054725077003240585
2023-01-21 11:00:00.211811: step: 1236/527, loss: 0.019669819623231888
2023-01-21 11:00:01.326425: step: 1240/527, loss: 0.06914804130792618
2023-01-21 11:00:02.451212: step: 1244/527, loss: 0.018675709143280983
2023-01-21 11:00:03.580112: step: 1248/527, loss: 0.0005324840312823653
2023-01-21 11:00:04.710841: step: 1252/527, loss: 0.02186889573931694
2023-01-21 11:00:05.830006: step: 1256/527, loss: 0.000992825604043901
2023-01-21 11:00:06.996879: step: 1260/527, loss: 0.019094567745923996
2023-01-21 11:00:08.154605: step: 1264/527, loss: 0.03783464431762695
2023-01-21 11:00:09.264308: step: 1268/527, loss: 0.0125885009765625
2023-01-21 11:00:10.378745: step: 1272/527, loss: 0.002647781278938055
2023-01-21 11:00:11.471220: step: 1276/527, loss: 0.07914400845766068
2023-01-21 11:00:12.586415: step: 1280/527, loss: 0.014637947082519531
2023-01-21 11:00:13.728210: step: 1284/527, loss: 0.016568852588534355
2023-01-21 11:00:14.855760: step: 1288/527, loss: 0.037230875343084335
2023-01-21 11:00:15.991875: step: 1292/527, loss: 0.01924166828393936
2023-01-21 11:00:17.123811: step: 1296/527, loss: 0.02688770368695259
2023-01-21 11:00:18.269169: step: 1300/527, loss: 0.33169764280319214
2023-01-21 11:00:19.410558: step: 1304/527, loss: 0.05361328274011612
2023-01-21 11:00:20.554143: step: 1308/527, loss: 0.0058762077242136
2023-01-21 11:00:21.685757: step: 1312/527, loss: 0.05802622064948082
2023-01-21 11:00:22.819699: step: 1316/527, loss: 0.011061620898544788
2023-01-21 11:00:23.936903: step: 1320/527, loss: 0.06392412632703781
2023-01-21 11:00:25.068504: step: 1324/527, loss: 0.06148626282811165
2023-01-21 11:00:26.213560: step: 1328/527, loss: 0.0288877971470356
2023-01-21 11:00:27.367128: step: 1332/527, loss: 0.03205270692706108
2023-01-21 11:00:28.477519: step: 1336/527, loss: 0.00723114050924778
2023-01-21 11:00:29.611206: step: 1340/527, loss: 0.031041037291288376
2023-01-21 11:00:30.723775: step: 1344/527, loss: 0.03764476999640465
2023-01-21 11:00:31.876542: step: 1348/527, loss: 0.018552685156464577
2023-01-21 11:00:33.033278: step: 1352/527, loss: 0.0029336928855627775
2023-01-21 11:00:34.197135: step: 1356/527, loss: 0.04499192163348198
2023-01-21 11:00:35.333122: step: 1360/527, loss: 0.012230491265654564
2023-01-21 11:00:36.468840: step: 1364/527, loss: 0.06836624443531036
2023-01-21 11:00:37.604376: step: 1368/527, loss: 0.04246025159955025
2023-01-21 11:00:38.731434: step: 1372/527, loss: 0.07367686927318573
2023-01-21 11:00:39.858319: step: 1376/527, loss: 0.03567257151007652
2023-01-21 11:00:41.004808: step: 1380/527, loss: 0.0007926941034384072
2023-01-21 11:00:42.135314: step: 1384/527, loss: 0.009362125769257545
2023-01-21 11:00:43.268149: step: 1388/527, loss: 0.04927559196949005
2023-01-21 11:00:44.405979: step: 1392/527, loss: 0.013686568476259708
2023-01-21 11:00:45.529136: step: 1396/527, loss: 0.006678963080048561
2023-01-21 11:00:46.649904: step: 1400/527, loss: 0.022643184289336205
2023-01-21 11:00:47.816790: step: 1404/527, loss: 0.005195808596909046
2023-01-21 11:00:48.935207: step: 1408/527, loss: 0.06345569342374802
2023-01-21 11:00:50.052552: step: 1412/527, loss: 0.007413959130644798
2023-01-21 11:00:51.249930: step: 1416/527, loss: 0.0896310806274414
2023-01-21 11:00:52.360297: step: 1420/527, loss: 0.40943050384521484
2023-01-21 11:00:53.491621: step: 1424/527, loss: 0.015190697275102139
2023-01-21 11:00:54.613604: step: 1428/527, loss: 0.02352161519229412
2023-01-21 11:00:55.728551: step: 1432/527, loss: 0.022275350987911224
2023-01-21 11:00:56.849937: step: 1436/527, loss: 0.03592429310083389
2023-01-21 11:00:58.004840: step: 1440/527, loss: 0.0030696869362145662
2023-01-21 11:00:59.142039: step: 1444/527, loss: 0.01106948871165514
2023-01-21 11:01:00.281694: step: 1448/527, loss: 0.0014821052318438888
2023-01-21 11:01:01.395570: step: 1452/527, loss: 0.010616493411362171
2023-01-21 11:01:02.487868: step: 1456/527, loss: 0.049814701080322266
2023-01-21 11:01:03.614221: step: 1460/527, loss: 0.40736040472984314
2023-01-21 11:01:04.741480: step: 1464/527, loss: 0.7366933822631836
2023-01-21 11:01:05.878813: step: 1468/527, loss: 0.03532399982213974
2023-01-21 11:01:07.015040: step: 1472/527, loss: 0.01801624335348606
2023-01-21 11:01:08.130874: step: 1476/527, loss: 0.019459152594208717
2023-01-21 11:01:09.259198: step: 1480/527, loss: 0.024257086217403412
2023-01-21 11:01:10.411843: step: 1484/527, loss: 0.02486143261194229
2023-01-21 11:01:11.555159: step: 1488/527, loss: 0.003638839814811945
2023-01-21 11:01:12.647100: step: 1492/527, loss: 0.009184408001601696
2023-01-21 11:01:13.764876: step: 1496/527, loss: 0.02962617762386799
2023-01-21 11:01:14.908747: step: 1500/527, loss: 0.023097610101103783
2023-01-21 11:01:16.049413: step: 1504/527, loss: 0.011059141717851162
2023-01-21 11:01:17.176093: step: 1508/527, loss: 0.017860984429717064
2023-01-21 11:01:18.284655: step: 1512/527, loss: 0.04872875660657883
2023-01-21 11:01:19.410205: step: 1516/527, loss: 0.13870486617088318
2023-01-21 11:01:20.507936: step: 1520/527, loss: 0.008144950494170189
2023-01-21 11:01:21.628424: step: 1524/527, loss: 0.009200001135468483
2023-01-21 11:01:22.757649: step: 1528/527, loss: 0.01956958696246147
2023-01-21 11:01:23.871802: step: 1532/527, loss: 0.007416916079819202
2023-01-21 11:01:25.008636: step: 1536/527, loss: 0.010880470275878906
2023-01-21 11:01:26.119543: step: 1540/527, loss: 0.007791805546730757
2023-01-21 11:01:27.233097: step: 1544/527, loss: 0.00788569450378418
2023-01-21 11:01:28.345910: step: 1548/527, loss: 0.012665462680161
2023-01-21 11:01:29.498071: step: 1552/527, loss: 0.05111370235681534
2023-01-21 11:01:30.651626: step: 1556/527, loss: 0.09236469864845276
2023-01-21 11:01:31.787238: step: 1560/527, loss: 0.0020856859628111124
2023-01-21 11:01:32.884409: step: 1564/527, loss: 0.007236766628921032
2023-01-21 11:01:34.006619: step: 1568/527, loss: 0.0023843764793127775
2023-01-21 11:01:35.149300: step: 1572/527, loss: 0.014798736199736595
2023-01-21 11:01:36.283053: step: 1576/527, loss: 0.002789783524349332
2023-01-21 11:01:37.423503: step: 1580/527, loss: 0.07930745929479599
2023-01-21 11:01:38.548755: step: 1584/527, loss: 0.19496233761310577
2023-01-21 11:01:39.645098: step: 1588/527, loss: 0.02212362363934517
2023-01-21 11:01:40.774562: step: 1592/527, loss: 0.0036204815842211246
2023-01-21 11:01:41.907367: step: 1596/527, loss: 0.002599430037662387
2023-01-21 11:01:43.042286: step: 1600/527, loss: 0.00913381576538086
2023-01-21 11:01:44.137247: step: 1604/527, loss: 0.05161609873175621
2023-01-21 11:01:45.278624: step: 1608/527, loss: 0.006849098484963179
2023-01-21 11:01:46.424109: step: 1612/527, loss: 0.00035114289494231343
2023-01-21 11:01:47.537906: step: 1616/527, loss: 0.026199722662568092
2023-01-21 11:01:48.658188: step: 1620/527, loss: 0.00015335083298850805
2023-01-21 11:01:49.829261: step: 1624/527, loss: 0.003810882568359375
2023-01-21 11:01:50.952378: step: 1628/527, loss: 0.010427093133330345
2023-01-21 11:01:52.083096: step: 1632/527, loss: 0.024481844156980515
2023-01-21 11:01:53.217085: step: 1636/527, loss: 0.022313212975859642
2023-01-21 11:01:54.329892: step: 1640/527, loss: 0.00387744908221066
2023-01-21 11:01:55.440145: step: 1644/527, loss: 0.23459607362747192
2023-01-21 11:01:56.575207: step: 1648/527, loss: 0.009972477331757545
2023-01-21 11:01:57.722939: step: 1652/527, loss: 0.038657426834106445
2023-01-21 11:01:58.865176: step: 1656/527, loss: 0.0717834010720253
2023-01-21 11:01:59.998200: step: 1660/527, loss: 0.008960723876953125
2023-01-21 11:02:01.129159: step: 1664/527, loss: 0.018883515149354935
2023-01-21 11:02:02.258019: step: 1668/527, loss: 0.07325001060962677
2023-01-21 11:02:03.360076: step: 1672/527, loss: 0.007762718480080366
2023-01-21 11:02:04.492214: step: 1676/527, loss: 0.01847999170422554
2023-01-21 11:02:05.634855: step: 1680/527, loss: 0.0014606475597247481
2023-01-21 11:02:06.758190: step: 1684/527, loss: 0.00630222586914897
2023-01-21 11:02:07.913851: step: 1688/527, loss: 0.024070357903838158
2023-01-21 11:02:09.057886: step: 1692/527, loss: 0.08104170858860016
2023-01-21 11:02:10.172140: step: 1696/527, loss: 0.016492461785674095
2023-01-21 11:02:11.359688: step: 1700/527, loss: 0.0006495476118288934
2023-01-21 11:02:12.484470: step: 1704/527, loss: 0.0002934455987997353
2023-01-21 11:02:13.654925: step: 1708/527, loss: 0.036316778510808945
2023-01-21 11:02:14.822422: step: 1712/527, loss: 0.03200960159301758
2023-01-21 11:02:15.966636: step: 1716/527, loss: 0.05724811553955078
2023-01-21 11:02:17.112096: step: 1720/527, loss: 0.0046852584928274155
2023-01-21 11:02:18.262039: step: 1724/527, loss: 0.009562206454575062
2023-01-21 11:02:19.380097: step: 1728/527, loss: 0.0020021439995616674
2023-01-21 11:02:20.513682: step: 1732/527, loss: 0.002619171282276511
2023-01-21 11:02:21.670066: step: 1736/527, loss: 0.14416447281837463
2023-01-21 11:02:22.808309: step: 1740/527, loss: 0.13138779997825623
2023-01-21 11:02:23.928896: step: 1744/527, loss: 0.036812592297792435
2023-01-21 11:02:25.058410: step: 1748/527, loss: 0.0043649678118526936
2023-01-21 11:02:26.187741: step: 1752/527, loss: 0.022001124918460846
2023-01-21 11:02:27.326323: step: 1756/527, loss: 0.07706394046545029
2023-01-21 11:02:28.466181: step: 1760/527, loss: 0.009550285525619984
2023-01-21 11:02:29.634727: step: 1764/527, loss: 0.020464133471250534
2023-01-21 11:02:30.802170: step: 1768/527, loss: 0.04759788513183594
2023-01-21 11:02:31.945684: step: 1772/527, loss: 0.005223751533776522
2023-01-21 11:02:33.091517: step: 1776/527, loss: 0.03288135677576065
2023-01-21 11:02:34.242239: step: 1780/527, loss: 0.055021099746227264
2023-01-21 11:02:35.359631: step: 1784/527, loss: 0.008260106667876244
2023-01-21 11:02:36.491250: step: 1788/527, loss: 0.05603409186005592
2023-01-21 11:02:37.614830: step: 1792/527, loss: 0.02227792702615261
2023-01-21 11:02:38.721614: step: 1796/527, loss: 0.00020914076594635844
2023-01-21 11:02:39.900368: step: 1800/527, loss: 0.012109661474823952
2023-01-21 11:02:41.056586: step: 1804/527, loss: 0.016881417483091354
2023-01-21 11:02:42.186657: step: 1808/527, loss: 0.025268077850341797
2023-01-21 11:02:43.321336: step: 1812/527, loss: 0.0006565094226971269
2023-01-21 11:02:44.495394: step: 1816/527, loss: 0.26937222480773926
2023-01-21 11:02:45.635311: step: 1820/527, loss: 0.0032444957178086042
2023-01-21 11:02:46.762913: step: 1824/527, loss: 0.0413576140999794
2023-01-21 11:02:47.918642: step: 1828/527, loss: 0.23703232407569885
2023-01-21 11:02:49.042085: step: 1832/527, loss: 0.017943954095244408
2023-01-21 11:02:50.176953: step: 1836/527, loss: 0.0021665573585778475
2023-01-21 11:02:51.324146: step: 1840/527, loss: 0.025844955816864967
2023-01-21 11:02:52.469128: step: 1844/527, loss: 0.027740098536014557
2023-01-21 11:02:53.635901: step: 1848/527, loss: 0.08496513962745667
2023-01-21 11:02:54.741495: step: 1852/527, loss: 0.01747722551226616
2023-01-21 11:02:55.879537: step: 1856/527, loss: 0.00048122406587935984
2023-01-21 11:02:57.031950: step: 1860/527, loss: 0.07278356701135635
2023-01-21 11:02:58.172477: step: 1864/527, loss: 0.062247514724731445
2023-01-21 11:02:59.330441: step: 1868/527, loss: 0.019217777997255325
2023-01-21 11:03:00.465994: step: 1872/527, loss: 0.08824024349451065
2023-01-21 11:03:01.592964: step: 1876/527, loss: 0.13959059119224548
2023-01-21 11:03:02.727852: step: 1880/527, loss: 0.01808653026819229
2023-01-21 11:03:03.855373: step: 1884/527, loss: 0.0028315545059740543
2023-01-21 11:03:04.965137: step: 1888/527, loss: 0.02536029927432537
2023-01-21 11:03:06.118166: step: 1892/527, loss: 0.10564308613538742
2023-01-21 11:03:07.247954: step: 1896/527, loss: 0.060558609664440155
2023-01-21 11:03:08.383442: step: 1900/527, loss: 0.004756450653076172
2023-01-21 11:03:09.512596: step: 1904/527, loss: 0.04950676113367081
2023-01-21 11:03:10.685152: step: 1908/527, loss: 0.019383717328310013
2023-01-21 11:03:11.805319: step: 1912/527, loss: 0.09795437753200531
2023-01-21 11:03:12.941202: step: 1916/527, loss: 0.04067039489746094
2023-01-21 11:03:14.054441: step: 1920/527, loss: 0.2678912878036499
2023-01-21 11:03:15.179920: step: 1924/527, loss: 0.019498825073242188
2023-01-21 11:03:16.306176: step: 1928/527, loss: 0.13204097747802734
2023-01-21 11:03:17.432318: step: 1932/527, loss: 0.05231637880206108
2023-01-21 11:03:18.603714: step: 1936/527, loss: 0.050676919519901276
2023-01-21 11:03:19.687866: step: 1940/527, loss: 0.026834391057491302
2023-01-21 11:03:20.832897: step: 1944/527, loss: 0.003731632139533758
2023-01-21 11:03:21.961859: step: 1948/527, loss: 0.012117576785385609
2023-01-21 11:03:23.144445: step: 1952/527, loss: 0.045914556831121445
2023-01-21 11:03:24.270978: step: 1956/527, loss: 0.026911545544862747
2023-01-21 11:03:25.415809: step: 1960/527, loss: 0.010535622015595436
2023-01-21 11:03:26.540389: step: 1964/527, loss: 0.0117524154484272
2023-01-21 11:03:27.680826: step: 1968/527, loss: 0.002340126084163785
2023-01-21 11:03:28.838868: step: 1972/527, loss: 0.00016446113295387477
2023-01-21 11:03:29.973333: step: 1976/527, loss: 0.04923286288976669
2023-01-21 11:03:31.107492: step: 1980/527, loss: 0.04620952531695366
2023-01-21 11:03:32.219945: step: 1984/527, loss: 0.004095935728400946
2023-01-21 11:03:33.383127: step: 1988/527, loss: 0.006879043765366077
2023-01-21 11:03:34.528045: step: 1992/527, loss: 0.046857260167598724
2023-01-21 11:03:35.653277: step: 1996/527, loss: 0.3629687428474426
2023-01-21 11:03:36.785313: step: 2000/527, loss: 0.07240734249353409
2023-01-21 11:03:37.917216: step: 2004/527, loss: 0.012729418464004993
2023-01-21 11:03:39.068742: step: 2008/527, loss: 0.02142629772424698
2023-01-21 11:03:40.226531: step: 2012/527, loss: 0.07787895202636719
2023-01-21 11:03:41.373155: step: 2016/527, loss: 0.008347177878022194
2023-01-21 11:03:42.488772: step: 2020/527, loss: 0.010721921920776367
2023-01-21 11:03:43.607298: step: 2024/527, loss: 0.008940315805375576
2023-01-21 11:03:44.723462: step: 2028/527, loss: 0.01839742809534073
2023-01-21 11:03:45.866165: step: 2032/527, loss: 0.008983040228486061
2023-01-21 11:03:47.008850: step: 2036/527, loss: 0.2618073523044586
2023-01-21 11:03:48.148792: step: 2040/527, loss: 0.10872535407543182
2023-01-21 11:03:49.281066: step: 2044/527, loss: 0.004775047302246094
2023-01-21 11:03:50.429886: step: 2048/527, loss: 0.0004932403680868447
2023-01-21 11:03:51.565928: step: 2052/527, loss: 0.07493095844984055
2023-01-21 11:03:52.683278: step: 2056/527, loss: 0.012756919488310814
2023-01-21 11:03:53.804218: step: 2060/527, loss: 0.10483141243457794
2023-01-21 11:03:54.933192: step: 2064/527, loss: 0.07885026931762695
2023-01-21 11:03:56.089669: step: 2068/527, loss: 0.008841801434755325
2023-01-21 11:03:57.248836: step: 2072/527, loss: 0.14572273194789886
2023-01-21 11:03:58.384915: step: 2076/527, loss: 0.011713081039488316
2023-01-21 11:03:59.514690: step: 2080/527, loss: 0.020029496401548386
2023-01-21 11:04:00.617250: step: 2084/527, loss: 0.013035583309829235
2023-01-21 11:04:01.755489: step: 2088/527, loss: 0.043305281549692154
2023-01-21 11:04:02.910086: step: 2092/527, loss: 0.0033678056206554174
2023-01-21 11:04:04.017764: step: 2096/527, loss: 0.03669004514813423
2023-01-21 11:04:05.153649: step: 2100/527, loss: 0.06149006262421608
2023-01-21 11:04:06.289133: step: 2104/527, loss: 0.02208099514245987
2023-01-21 11:04:07.430830: step: 2108/527, loss: 0.0014815330505371094
==================================================
Loss: 0.044
--------------------
Dev: {'event': {'p': 0.556390977443609, 'r': 0.7882822902796272, 'f1': 0.6523415977961432}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14}
Test: {'event': {'p': 0.5934856175972927, 'r': 0.8017142857142857, 'f1': 0.6820612542537676}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14}
Chinese: {'event': {'p': 0.4725274725274725, 'r': 0.7962962962962963, 'f1': 0.593103448275862}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14}
Korean: {'event': {'p': 0.5633802816901409, 'r': 0.6349206349206349, 'f1': 0.5970149253731343}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14}
Russian: {'event': {'p': 0.43478260869565216, 'r': 0.5555555555555556, 'f1': 0.4878048780487805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:04:51.487689: step: 4/527, loss: 0.22581176459789276
2023-01-21 11:04:52.589211: step: 8/527, loss: 0.02170557901263237
2023-01-21 11:04:53.705411: step: 12/527, loss: 0.007999849505722523
2023-01-21 11:04:54.847293: step: 16/527, loss: 0.22186845541000366
2023-01-21 11:04:56.001859: step: 20/527, loss: 0.016269873827695847
2023-01-21 11:04:57.108877: step: 24/527, loss: 0.011776065453886986
2023-01-21 11:04:58.259783: step: 28/527, loss: 0.02494525909423828
2023-01-21 11:04:59.394682: step: 32/527, loss: 0.0640285462141037
2023-01-21 11:05:00.539237: step: 36/527, loss: 0.04752330482006073
2023-01-21 11:05:01.673109: step: 40/527, loss: 0.153310164809227
2023-01-21 11:05:02.833479: step: 44/527, loss: 0.0013426780933514237
2023-01-21 11:05:03.981987: step: 48/527, loss: 0.027942562475800514
2023-01-21 11:05:05.109345: step: 52/527, loss: 0.030162621289491653
2023-01-21 11:05:06.238337: step: 56/527, loss: 0.04598961025476456
2023-01-21 11:05:07.379631: step: 60/527, loss: 0.035779476165771484
2023-01-21 11:05:08.486595: step: 64/527, loss: 0.028416920453310013
2023-01-21 11:05:09.599812: step: 68/527, loss: 0.23880743980407715
2023-01-21 11:05:10.695206: step: 72/527, loss: 0.04108009114861488
2023-01-21 11:05:11.816085: step: 76/527, loss: 0.001728153321892023
2023-01-21 11:05:12.932601: step: 80/527, loss: 0.011955452151596546
2023-01-21 11:05:14.089416: step: 84/527, loss: 0.03989720344543457
2023-01-21 11:05:15.239979: step: 88/527, loss: 0.4120454788208008
2023-01-21 11:05:16.359123: step: 92/527, loss: 0.006381893530488014
2023-01-21 11:05:17.471560: step: 96/527, loss: 0.010520553216338158
2023-01-21 11:05:18.608117: step: 100/527, loss: 0.05761122703552246
2023-01-21 11:05:19.723062: step: 104/527, loss: 0.029200460761785507
2023-01-21 11:05:20.831436: step: 108/527, loss: 0.01035852450877428
2023-01-21 11:05:21.943268: step: 112/527, loss: 0.006377363111823797
2023-01-21 11:05:23.058223: step: 116/527, loss: 0.31469297409057617
2023-01-21 11:05:24.187683: step: 120/527, loss: 0.15226240456104279
2023-01-21 11:05:25.326787: step: 124/527, loss: 0.031447507441043854
2023-01-21 11:05:26.481784: step: 128/527, loss: 0.015896225348114967
2023-01-21 11:05:27.629622: step: 132/527, loss: 0.00046215057955123484
2023-01-21 11:05:28.727789: step: 136/527, loss: 0.035303495824337006
2023-01-21 11:05:29.858107: step: 140/527, loss: 0.0394381508231163
2023-01-21 11:05:30.975725: step: 144/527, loss: 0.03318657726049423
2023-01-21 11:05:32.148018: step: 148/527, loss: 0.009058475494384766
2023-01-21 11:05:33.289152: step: 152/527, loss: 0.0007450580596923828
2023-01-21 11:05:34.417460: step: 156/527, loss: 0.0013151168823242188
2023-01-21 11:05:35.548220: step: 160/527, loss: 0.025045014917850494
2023-01-21 11:05:36.669461: step: 164/527, loss: 0.0030714033637195826
2023-01-21 11:05:37.788617: step: 168/527, loss: 0.0380069725215435
2023-01-21 11:05:38.911305: step: 172/527, loss: 0.017401695251464844
2023-01-21 11:05:40.053768: step: 176/527, loss: 0.0007852554554119706
2023-01-21 11:05:41.176571: step: 180/527, loss: 0.007587719243019819
2023-01-21 11:05:42.312464: step: 184/527, loss: 0.01526489295065403
2023-01-21 11:05:43.467273: step: 188/527, loss: 0.020406054332852364
2023-01-21 11:05:44.614140: step: 192/527, loss: 0.00583915738388896
2023-01-21 11:05:45.750442: step: 196/527, loss: 0.002165126847103238
2023-01-21 11:05:46.884471: step: 200/527, loss: 0.030858324840664864
2023-01-21 11:05:48.001398: step: 204/527, loss: 0.00779070844873786
2023-01-21 11:05:49.118614: step: 208/527, loss: 0.025826454162597656
2023-01-21 11:05:50.246518: step: 212/527, loss: 0.0029317857697606087
2023-01-21 11:05:51.395679: step: 216/527, loss: 0.01831207238137722
2023-01-21 11:05:52.496945: step: 220/527, loss: 0.0293342936784029
2023-01-21 11:05:53.651414: step: 224/527, loss: 0.008865833282470703
2023-01-21 11:05:54.767771: step: 228/527, loss: 0.0002617835998535156
2023-01-21 11:05:55.903502: step: 232/527, loss: 0.02391691319644451
2023-01-21 11:05:57.037955: step: 236/527, loss: 0.0868292823433876
2023-01-21 11:05:58.175719: step: 240/527, loss: 0.041964102536439896
2023-01-21 11:05:59.336717: step: 244/527, loss: 0.017502497881650925
2023-01-21 11:06:00.475216: step: 248/527, loss: 0.0001388549862895161
2023-01-21 11:06:01.612990: step: 252/527, loss: 0.02922849729657173
2023-01-21 11:06:02.742056: step: 256/527, loss: 0.0016942977672442794
2023-01-21 11:06:03.852305: step: 260/527, loss: 0.0062961578369140625
2023-01-21 11:06:04.983381: step: 264/527, loss: 0.0023731710389256477
2023-01-21 11:06:06.097706: step: 268/527, loss: 0.004225730895996094
2023-01-21 11:06:07.209382: step: 272/527, loss: 0.02932577021420002
2023-01-21 11:06:08.350181: step: 276/527, loss: 0.01079249382019043
2023-01-21 11:06:09.483696: step: 280/527, loss: 0.0006466865306720138
2023-01-21 11:06:10.629381: step: 284/527, loss: 0.013340759091079235
2023-01-21 11:06:11.780058: step: 288/527, loss: 0.046209193766117096
2023-01-21 11:06:12.881243: step: 292/527, loss: 0.0004779815790243447
2023-01-21 11:06:14.004954: step: 296/527, loss: 0.004534339997917414
2023-01-21 11:06:15.144827: step: 300/527, loss: 0.004288864322006702
2023-01-21 11:06:16.287029: step: 304/527, loss: 0.040387917309999466
2023-01-21 11:06:17.482824: step: 308/527, loss: 2.307891918462701e-05
2023-01-21 11:06:18.592415: step: 312/527, loss: 0.004644202999770641
2023-01-21 11:06:19.715931: step: 316/527, loss: 0.07197723537683487
2023-01-21 11:06:20.836650: step: 320/527, loss: 0.00131816859357059
2023-01-21 11:06:21.968644: step: 324/527, loss: 0.0008647919166833162
2023-01-21 11:06:23.115133: step: 328/527, loss: 0.0026178359985351562
2023-01-21 11:06:24.295846: step: 332/527, loss: 0.03224804252386093
2023-01-21 11:06:25.400559: step: 336/527, loss: 0.07398414611816406
2023-01-21 11:06:26.514651: step: 340/527, loss: 0.008480453863739967
2023-01-21 11:06:27.638013: step: 344/527, loss: 0.02752704545855522
2023-01-21 11:06:28.842794: step: 348/527, loss: 0.028559494763612747
2023-01-21 11:06:29.972726: step: 352/527, loss: 0.014121342450380325
2023-01-21 11:06:31.096530: step: 356/527, loss: 0.014368820935487747
2023-01-21 11:06:32.221064: step: 360/527, loss: 0.07667160034179688
2023-01-21 11:06:33.343906: step: 364/527, loss: 0.021698379889130592
2023-01-21 11:06:34.465943: step: 368/527, loss: 0.0033740997314453125
2023-01-21 11:06:35.575103: step: 372/527, loss: 0.0021216394379734993
2023-01-21 11:06:36.698603: step: 376/527, loss: 0.0009344101417809725
2023-01-21 11:06:37.875705: step: 380/527, loss: 0.08845214545726776
2023-01-21 11:06:39.030484: step: 384/527, loss: 0.06588821858167648
2023-01-21 11:06:40.161221: step: 388/527, loss: 0.0014370918506756425
2023-01-21 11:06:41.260942: step: 392/527, loss: 0.19271382689476013
2023-01-21 11:06:42.416412: step: 396/527, loss: 0.01190652884542942
2023-01-21 11:06:43.548306: step: 400/527, loss: 0.0011715888977050781
2023-01-21 11:06:44.675562: step: 404/527, loss: 0.0005519866826944053
2023-01-21 11:06:45.817538: step: 408/527, loss: 0.009817123413085938
2023-01-21 11:06:46.952850: step: 412/527, loss: 0.2533043920993805
2023-01-21 11:06:48.076488: step: 416/527, loss: 6.67572021484375e-05
2023-01-21 11:06:49.206477: step: 420/527, loss: 0.06502113491296768
2023-01-21 11:06:50.386875: step: 424/527, loss: 0.020995713770389557
2023-01-21 11:06:51.513206: step: 428/527, loss: 0.01771850697696209
2023-01-21 11:06:52.645934: step: 432/527, loss: 0.009616279974579811
2023-01-21 11:06:53.786558: step: 436/527, loss: 0.002470398088917136
2023-01-21 11:06:54.966464: step: 440/527, loss: 0.02363281324505806
2023-01-21 11:06:56.101926: step: 444/527, loss: 0.07266445457935333
2023-01-21 11:06:57.258982: step: 448/527, loss: 0.009250020608305931
2023-01-21 11:06:58.388966: step: 452/527, loss: 0.0042115929536521435
2023-01-21 11:06:59.544570: step: 456/527, loss: 0.0006821156130172312
2023-01-21 11:07:00.666413: step: 460/527, loss: 0.028563881292939186
2023-01-21 11:07:01.810029: step: 464/527, loss: 0.032236672937870026
2023-01-21 11:07:02.940114: step: 468/527, loss: 0.013486957177519798
2023-01-21 11:07:04.063272: step: 472/527, loss: 0.028739070519804955
2023-01-21 11:07:05.199011: step: 476/527, loss: 0.029852628707885742
2023-01-21 11:07:06.355199: step: 480/527, loss: 0.003395938780158758
2023-01-21 11:07:07.479817: step: 484/527, loss: 0.029524995014071465
2023-01-21 11:07:08.645076: step: 488/527, loss: 0.03154907375574112
2023-01-21 11:07:09.777780: step: 492/527, loss: 0.001433944795280695
2023-01-21 11:07:10.890553: step: 496/527, loss: 0.01949033886194229
2023-01-21 11:07:12.015437: step: 500/527, loss: 0.23448696732521057
2023-01-21 11:07:13.135885: step: 504/527, loss: 0.004198408219963312
2023-01-21 11:07:14.264411: step: 508/527, loss: 0.00396652240306139
2023-01-21 11:07:15.406653: step: 512/527, loss: 0.026943014934659004
2023-01-21 11:07:16.544120: step: 516/527, loss: 0.005537987221032381
2023-01-21 11:07:17.696455: step: 520/527, loss: 0.1422453075647354
2023-01-21 11:07:18.793881: step: 524/527, loss: 0.0011471748584881425
2023-01-21 11:07:19.932457: step: 528/527, loss: 0.01975860632956028
2023-01-21 11:07:21.041787: step: 532/527, loss: 0.014313125051558018
2023-01-21 11:07:22.171210: step: 536/527, loss: 0.03721790388226509
2023-01-21 11:07:23.298516: step: 540/527, loss: 0.025241805240511894
2023-01-21 11:07:24.421829: step: 544/527, loss: 0.006464958656579256
2023-01-21 11:07:25.539439: step: 548/527, loss: 0.003733062883839011
2023-01-21 11:07:26.680636: step: 552/527, loss: 0.010371970944106579
2023-01-21 11:07:27.791823: step: 556/527, loss: 0.04620685800909996
2023-01-21 11:07:28.932463: step: 560/527, loss: 0.2306041717529297
2023-01-21 11:07:30.075673: step: 564/527, loss: 0.005190277472138405
2023-01-21 11:07:31.215532: step: 568/527, loss: 0.010545730590820312
2023-01-21 11:07:32.363571: step: 572/527, loss: 0.027575969696044922
2023-01-21 11:07:33.496883: step: 576/527, loss: 0.014630794525146484
2023-01-21 11:07:34.615038: step: 580/527, loss: 0.004957115743309259
2023-01-21 11:07:35.747021: step: 584/527, loss: 0.008568287827074528
2023-01-21 11:07:36.895831: step: 588/527, loss: 0.0028623580001294613
2023-01-21 11:07:38.065256: step: 592/527, loss: 0.11433592438697815
2023-01-21 11:07:39.199077: step: 596/527, loss: 0.016902972012758255
2023-01-21 11:07:40.391396: step: 600/527, loss: 0.038194846361875534
2023-01-21 11:07:41.521026: step: 604/527, loss: 0.018544532358646393
2023-01-21 11:07:42.622917: step: 608/527, loss: 0.004984951112419367
2023-01-21 11:07:43.747762: step: 612/527, loss: 0.002406501676887274
2023-01-21 11:07:44.866317: step: 616/527, loss: 0.055287934839725494
2023-01-21 11:07:45.959258: step: 620/527, loss: 0.0013731956714764237
2023-01-21 11:07:47.088798: step: 624/527, loss: 0.004456761293113232
2023-01-21 11:07:48.235760: step: 628/527, loss: 0.08327207714319229
2023-01-21 11:07:49.377549: step: 632/527, loss: 0.049032118171453476
2023-01-21 11:07:50.529673: step: 636/527, loss: 0.02480325661599636
2023-01-21 11:07:51.647830: step: 640/527, loss: 0.03411727026104927
2023-01-21 11:07:52.752600: step: 644/527, loss: 0.0045403484255075455
2023-01-21 11:07:53.894253: step: 648/527, loss: 0.028308678418397903
2023-01-21 11:07:55.039152: step: 652/527, loss: 0.0013042927021160722
2023-01-21 11:07:56.152764: step: 656/527, loss: 0.0010877609020099044
2023-01-21 11:07:57.280058: step: 660/527, loss: 0.00903244037181139
2023-01-21 11:07:58.433008: step: 664/527, loss: 0.3747520446777344
2023-01-21 11:07:59.578652: step: 668/527, loss: 0.09205179661512375
2023-01-21 11:08:00.724211: step: 672/527, loss: 0.00017471313185524195
2023-01-21 11:08:01.855407: step: 676/527, loss: 0.04350528493523598
2023-01-21 11:08:02.972834: step: 680/527, loss: 0.006512451451271772
2023-01-21 11:08:04.118473: step: 684/527, loss: 0.002341025974601507
2023-01-21 11:08:05.240289: step: 688/527, loss: 0.019007539376616478
2023-01-21 11:08:06.337219: step: 692/527, loss: 0.05936779826879501
2023-01-21 11:08:07.503723: step: 696/527, loss: 0.002118778182193637
2023-01-21 11:08:08.668594: step: 700/527, loss: 0.04339752346277237
2023-01-21 11:08:09.798465: step: 704/527, loss: 0.011522864922881126
2023-01-21 11:08:10.988785: step: 708/527, loss: 0.09353247284889221
2023-01-21 11:08:12.074469: step: 712/527, loss: 0.0065042972564697266
2023-01-21 11:08:13.227800: step: 716/527, loss: 0.09559492766857147
2023-01-21 11:08:14.357681: step: 720/527, loss: 0.009997940622270107
2023-01-21 11:08:15.500746: step: 724/527, loss: 0.005124092102050781
2023-01-21 11:08:16.670687: step: 728/527, loss: 0.0015773654449731112
2023-01-21 11:08:17.782372: step: 732/527, loss: 0.0029996871016919613
2023-01-21 11:08:18.921812: step: 736/527, loss: 0.037795402109622955
2023-01-21 11:08:20.024789: step: 740/527, loss: 0.01293411199003458
2023-01-21 11:08:21.198413: step: 744/527, loss: 0.010936308652162552
2023-01-21 11:08:22.324927: step: 748/527, loss: 0.05385913699865341
2023-01-21 11:08:23.481675: step: 752/527, loss: 0.04314403608441353
2023-01-21 11:08:24.591937: step: 756/527, loss: 0.045433998107910156
2023-01-21 11:08:25.705586: step: 760/527, loss: 0.0537446066737175
2023-01-21 11:08:26.844112: step: 764/527, loss: 0.03394460678100586
2023-01-21 11:08:27.980179: step: 768/527, loss: 0.005544853396713734
2023-01-21 11:08:29.103949: step: 772/527, loss: 0.0033130645751953125
2023-01-21 11:08:30.239523: step: 776/527, loss: 0.04084720462560654
2023-01-21 11:08:31.364556: step: 780/527, loss: 0.04300842434167862
2023-01-21 11:08:32.530657: step: 784/527, loss: 0.010472392663359642
2023-01-21 11:08:33.697356: step: 788/527, loss: 0.004683780949562788
2023-01-21 11:08:34.827312: step: 792/527, loss: 0.02362346649169922
2023-01-21 11:08:35.970059: step: 796/527, loss: 0.019580459222197533
2023-01-21 11:08:37.066787: step: 800/527, loss: 0.0004859924374613911
2023-01-21 11:08:38.191138: step: 804/527, loss: 0.008272933773696423
2023-01-21 11:08:39.286882: step: 808/527, loss: 0.00017070770263671875
2023-01-21 11:08:40.444569: step: 812/527, loss: 0.03326749801635742
2023-01-21 11:08:41.578568: step: 816/527, loss: 0.07345166057348251
2023-01-21 11:08:42.712266: step: 820/527, loss: 0.012792587280273438
2023-01-21 11:08:43.857759: step: 824/527, loss: 0.0047397613525390625
2023-01-21 11:08:44.965253: step: 828/527, loss: 0.1512743979692459
2023-01-21 11:08:46.122766: step: 832/527, loss: 0.06097078323364258
2023-01-21 11:08:47.237052: step: 836/527, loss: 0.00021858215040992945
2023-01-21 11:08:48.358501: step: 840/527, loss: 0.0123291015625
2023-01-21 11:08:49.479304: step: 844/527, loss: 0.02930602990090847
2023-01-21 11:08:50.635327: step: 848/527, loss: 0.09752263873815536
2023-01-21 11:08:51.805512: step: 852/527, loss: 0.002418327145278454
2023-01-21 11:08:52.948698: step: 856/527, loss: 0.020143888890743256
2023-01-21 11:08:54.041639: step: 860/527, loss: 0.004431247711181641
2023-01-21 11:08:55.175218: step: 864/527, loss: 0.0016469955444335938
2023-01-21 11:08:56.290659: step: 868/527, loss: 0.08239050209522247
2023-01-21 11:08:57.456885: step: 872/527, loss: 0.05172882229089737
2023-01-21 11:08:58.591532: step: 876/527, loss: 0.030521774664521217
2023-01-21 11:08:59.735600: step: 880/527, loss: 0.11122265458106995
2023-01-21 11:09:00.884661: step: 884/527, loss: 0.0822354331612587
2023-01-21 11:09:02.027497: step: 888/527, loss: 0.029909705743193626
2023-01-21 11:09:03.144016: step: 892/527, loss: 0.0001676559477346018
2023-01-21 11:09:04.298991: step: 896/527, loss: 0.00345611572265625
2023-01-21 11:09:05.435280: step: 900/527, loss: 0.025514651089906693
2023-01-21 11:09:06.577095: step: 904/527, loss: 0.3412491977214813
2023-01-21 11:09:07.717338: step: 908/527, loss: 0.002812004182487726
2023-01-21 11:09:08.866482: step: 912/527, loss: 0.00018939972505904734
2023-01-21 11:09:10.007725: step: 916/527, loss: 0.006989955902099609
2023-01-21 11:09:11.133306: step: 920/527, loss: 0.038678836077451706
2023-01-21 11:09:12.240647: step: 924/527, loss: 0.0037471773102879524
2023-01-21 11:09:13.392563: step: 928/527, loss: 0.011445618234574795
2023-01-21 11:09:14.516997: step: 932/527, loss: 0.018006229773163795
2023-01-21 11:09:15.665755: step: 936/527, loss: 0.013768625445663929
2023-01-21 11:09:16.809429: step: 940/527, loss: 0.00183448800817132
2023-01-21 11:09:17.931309: step: 944/527, loss: 0.011353719048202038
2023-01-21 11:09:19.069480: step: 948/527, loss: 0.01029129046946764
2023-01-21 11:09:20.237880: step: 952/527, loss: 0.3560040593147278
2023-01-21 11:09:21.385003: step: 956/527, loss: 0.008546257391571999
2023-01-21 11:09:22.499511: step: 960/527, loss: 0.0031757354736328125
2023-01-21 11:09:23.611162: step: 964/527, loss: 0.003930854611098766
2023-01-21 11:09:24.740834: step: 968/527, loss: 0.008129405789077282
2023-01-21 11:09:25.867670: step: 972/527, loss: 0.06368846446275711
2023-01-21 11:09:26.991555: step: 976/527, loss: 0.021250534802675247
2023-01-21 11:09:28.128526: step: 980/527, loss: 0.10567241162061691
2023-01-21 11:09:29.251487: step: 984/527, loss: 0.04129352420568466
2023-01-21 11:09:30.376933: step: 988/527, loss: 0.04807844012975693
2023-01-21 11:09:31.498571: step: 992/527, loss: 0.030477095395326614
2023-01-21 11:09:32.628024: step: 996/527, loss: 0.005741024389863014
2023-01-21 11:09:33.748991: step: 1000/527, loss: 0.0018512726528570056
2023-01-21 11:09:34.856850: step: 1004/527, loss: 0.037592507898807526
2023-01-21 11:09:36.009744: step: 1008/527, loss: 0.013914298266172409
2023-01-21 11:09:37.185820: step: 1012/527, loss: 0.06752672046422958
2023-01-21 11:09:38.306219: step: 1016/527, loss: 0.007949447259306908
2023-01-21 11:09:39.448768: step: 1020/527, loss: 0.003915214445441961
2023-01-21 11:09:40.554712: step: 1024/527, loss: 0.01251897867769003
2023-01-21 11:09:41.679309: step: 1028/527, loss: 0.005850601010024548
2023-01-21 11:09:42.818440: step: 1032/527, loss: 0.0817023292183876
2023-01-21 11:09:43.913042: step: 1036/527, loss: 0.008228874765336514
2023-01-21 11:09:45.032001: step: 1040/527, loss: 0.009776497259736061
2023-01-21 11:09:46.182489: step: 1044/527, loss: 0.0006307124858722091
2023-01-21 11:09:47.328292: step: 1048/527, loss: 0.00175647740252316
2023-01-21 11:09:48.481758: step: 1052/527, loss: 0.017972279340028763
2023-01-21 11:09:49.618780: step: 1056/527, loss: 0.008592414669692516
2023-01-21 11:09:50.755907: step: 1060/527, loss: 0.03538760915398598
2023-01-21 11:09:51.894325: step: 1064/527, loss: 0.004697036929428577
2023-01-21 11:09:53.042918: step: 1068/527, loss: 0.13068370521068573
2023-01-21 11:09:54.203268: step: 1072/527, loss: 0.0009929656516760588
2023-01-21 11:09:55.339979: step: 1076/527, loss: 0.0016904830699786544
2023-01-21 11:09:56.466690: step: 1080/527, loss: 0.009094525128602982
2023-01-21 11:09:57.585152: step: 1084/527, loss: 0.02324857749044895
2023-01-21 11:09:58.715155: step: 1088/527, loss: 0.035799790173769
2023-01-21 11:09:59.838520: step: 1092/527, loss: 0.07102776318788528
2023-01-21 11:10:00.977047: step: 1096/527, loss: 0.005430793855339289
2023-01-21 11:10:02.101499: step: 1100/527, loss: 0.0266552921384573
2023-01-21 11:10:03.219795: step: 1104/527, loss: 0.006526744458824396
2023-01-21 11:10:04.345280: step: 1108/527, loss: 0.019536161795258522
2023-01-21 11:10:05.481069: step: 1112/527, loss: 0.035013582557439804
2023-01-21 11:10:06.619578: step: 1116/527, loss: 0.011952400207519531
2023-01-21 11:10:07.774532: step: 1120/527, loss: 0.02453918568789959
2023-01-21 11:10:08.920000: step: 1124/527, loss: 0.013444328680634499
2023-01-21 11:10:10.073860: step: 1128/527, loss: 0.0037784576416015625
2023-01-21 11:10:11.213602: step: 1132/527, loss: 0.06779623031616211
2023-01-21 11:10:12.330550: step: 1136/527, loss: 0.005590343382209539
2023-01-21 11:10:13.436285: step: 1140/527, loss: 0.051160432398319244
2023-01-21 11:10:14.577121: step: 1144/527, loss: 0.02247796021401882
2023-01-21 11:10:15.695838: step: 1148/527, loss: 0.003013419918715954
2023-01-21 11:10:16.830364: step: 1152/527, loss: 0.00958948116749525
2023-01-21 11:10:17.945318: step: 1156/527, loss: 0.01073455810546875
2023-01-21 11:10:19.040876: step: 1160/527, loss: 0.0077984812669456005
2023-01-21 11:10:20.134247: step: 1164/527, loss: 0.005529975984245539
2023-01-21 11:10:21.302645: step: 1168/527, loss: 0.003546428633853793
2023-01-21 11:10:22.425743: step: 1172/527, loss: 0.014021635986864567
2023-01-21 11:10:23.563848: step: 1176/527, loss: 0.014738941565155983
2023-01-21 11:10:24.724264: step: 1180/527, loss: 0.009393453598022461
2023-01-21 11:10:25.882116: step: 1184/527, loss: 0.0003514289855957031
2023-01-21 11:10:27.015122: step: 1188/527, loss: 0.0018528937362134457
2023-01-21 11:10:28.144076: step: 1192/527, loss: 0.008082438260316849
2023-01-21 11:10:29.232577: step: 1196/527, loss: 0.0009001732105389237
2023-01-21 11:10:30.354801: step: 1200/527, loss: 0.004994392395019531
2023-01-21 11:10:31.496639: step: 1204/527, loss: 0.08577661216259003
2023-01-21 11:10:32.661926: step: 1208/527, loss: 0.012539064511656761
2023-01-21 11:10:33.795589: step: 1212/527, loss: 0.023921776562929153
2023-01-21 11:10:34.915501: step: 1216/527, loss: 0.04786815866827965
2023-01-21 11:10:36.067837: step: 1220/527, loss: 0.02133164368569851
2023-01-21 11:10:37.212888: step: 1224/527, loss: 0.028213834390044212
2023-01-21 11:10:38.320506: step: 1228/527, loss: 0.041399385780096054
2023-01-21 11:10:39.442674: step: 1232/527, loss: 0.023358918726444244
2023-01-21 11:10:40.568105: step: 1236/527, loss: 0.02531290054321289
2023-01-21 11:10:41.678787: step: 1240/527, loss: 0.06339474022388458
2023-01-21 11:10:42.798784: step: 1244/527, loss: 3.576399087905884
2023-01-21 11:10:43.954194: step: 1248/527, loss: 0.004817867651581764
2023-01-21 11:10:45.107762: step: 1252/527, loss: 0.08507785946130753
2023-01-21 11:10:46.226389: step: 1256/527, loss: 0.015553951263427734
2023-01-21 11:10:47.341698: step: 1260/527, loss: 0.0004887580871582031
2023-01-21 11:10:48.475401: step: 1264/527, loss: 0.0008686066139489412
2023-01-21 11:10:49.581817: step: 1268/527, loss: 0.022221755236387253
2023-01-21 11:10:50.694370: step: 1272/527, loss: 0.0034784318413585424
2023-01-21 11:10:51.800834: step: 1276/527, loss: 0.0016757011180743575
2023-01-21 11:10:52.942952: step: 1280/527, loss: 0.025454092770814896
2023-01-21 11:10:54.073007: step: 1284/527, loss: 0.003937912173569202
2023-01-21 11:10:55.200563: step: 1288/527, loss: 0.011016273871064186
2023-01-21 11:10:56.333837: step: 1292/527, loss: 0.02684783935546875
2023-01-21 11:10:57.458658: step: 1296/527, loss: 0.0024065019097179174
2023-01-21 11:10:58.577141: step: 1300/527, loss: 0.03681793436408043
2023-01-21 11:10:59.695449: step: 1304/527, loss: 0.017932891845703125
2023-01-21 11:11:00.827377: step: 1308/527, loss: 0.0014888762962073088
2023-01-21 11:11:01.957479: step: 1312/527, loss: 0.02482910081744194
2023-01-21 11:11:03.078154: step: 1316/527, loss: 0.020549679175019264
2023-01-21 11:11:04.188541: step: 1320/527, loss: 0.023158837109804153
2023-01-21 11:11:05.332782: step: 1324/527, loss: 0.060376547276973724
2023-01-21 11:11:06.484866: step: 1328/527, loss: 0.00930476188659668
2023-01-21 11:11:07.594068: step: 1332/527, loss: 0.021881485357880592
2023-01-21 11:11:08.735769: step: 1336/527, loss: 0.009923458099365234
2023-01-21 11:11:09.861011: step: 1340/527, loss: 0.005238723941147327
2023-01-21 11:11:10.990530: step: 1344/527, loss: 0.6213347315788269
2023-01-21 11:11:12.125686: step: 1348/527, loss: 0.028046799823641777
2023-01-21 11:11:13.242099: step: 1352/527, loss: 0.00892176665365696
2023-01-21 11:11:14.371074: step: 1356/527, loss: 0.170710027217865
2023-01-21 11:11:15.514834: step: 1360/527, loss: 0.025066375732421875
2023-01-21 11:11:16.642927: step: 1364/527, loss: 0.0086218835785985
2023-01-21 11:11:17.777583: step: 1368/527, loss: 0.09632397443056107
2023-01-21 11:11:18.921589: step: 1372/527, loss: 0.05324440076947212
2023-01-21 11:11:20.047929: step: 1376/527, loss: 0.0177596565335989
2023-01-21 11:11:21.184544: step: 1380/527, loss: 0.006068420596420765
2023-01-21 11:11:22.317399: step: 1384/527, loss: 0.042371559888124466
2023-01-21 11:11:23.445250: step: 1388/527, loss: 0.0009443283197470009
2023-01-21 11:11:24.574041: step: 1392/527, loss: 0.00021743775869254023
2023-01-21 11:11:25.723368: step: 1396/527, loss: 0.0001392364501953125
2023-01-21 11:11:26.856731: step: 1400/527, loss: 0.019303608685731888
2023-01-21 11:11:27.969337: step: 1404/527, loss: 0.00343151087872684
2023-01-21 11:11:29.138248: step: 1408/527, loss: 0.028829192742705345
2023-01-21 11:11:30.262771: step: 1412/527, loss: 0.01144785899668932
2023-01-21 11:11:31.379125: step: 1416/527, loss: 0.015566635876893997
2023-01-21 11:11:32.557686: step: 1420/527, loss: 0.0021302225068211555
2023-01-21 11:11:33.711657: step: 1424/527, loss: 0.0002058029203908518
2023-01-21 11:11:34.843265: step: 1428/527, loss: 0.0196699146181345
2023-01-21 11:11:36.004224: step: 1432/527, loss: 0.02204723283648491
2023-01-21 11:11:37.121947: step: 1436/527, loss: 0.003985119052231312
2023-01-21 11:11:38.275961: step: 1440/527, loss: 0.08211441338062286
2023-01-21 11:11:39.391801: step: 1444/527, loss: 0.026621095836162567
2023-01-21 11:11:40.548957: step: 1448/527, loss: 0.11128316074609756
2023-01-21 11:11:41.668952: step: 1452/527, loss: 0.023581314831972122
2023-01-21 11:11:42.807741: step: 1456/527, loss: 0.0001388549862895161
2023-01-21 11:11:43.939027: step: 1460/527, loss: 0.00271263113245368
2023-01-21 11:11:45.055120: step: 1464/527, loss: 0.04239244386553764
2023-01-21 11:11:46.191450: step: 1468/527, loss: 0.02157735824584961
2023-01-21 11:11:47.313248: step: 1472/527, loss: 0.0339265838265419
2023-01-21 11:11:48.451411: step: 1476/527, loss: 0.0025482177734375
2023-01-21 11:11:49.546250: step: 1480/527, loss: 0.003231144044548273
2023-01-21 11:11:50.689182: step: 1484/527, loss: 0.026413630694150925
2023-01-21 11:11:51.816367: step: 1488/527, loss: 0.013781165704131126
2023-01-21 11:11:52.948588: step: 1492/527, loss: 0.00566444406285882
2023-01-21 11:11:54.106995: step: 1496/527, loss: 0.007185888476669788
2023-01-21 11:11:55.237174: step: 1500/527, loss: 0.050115205347537994
2023-01-21 11:11:56.371049: step: 1504/527, loss: 0.021830463781952858
2023-01-21 11:11:57.504007: step: 1508/527, loss: 0.014794539660215378
2023-01-21 11:11:58.621421: step: 1512/527, loss: 0.0015491485828533769
2023-01-21 11:11:59.799116: step: 1516/527, loss: 0.04486217349767685
2023-01-21 11:12:00.918226: step: 1520/527, loss: 0.04860267788171768
2023-01-21 11:12:02.057077: step: 1524/527, loss: 0.04024339094758034
2023-01-21 11:12:03.211714: step: 1528/527, loss: 0.00019292831711936742
2023-01-21 11:12:04.327450: step: 1532/527, loss: 0.011592579074203968
2023-01-21 11:12:05.453738: step: 1536/527, loss: 0.006886291783303022
2023-01-21 11:12:06.585576: step: 1540/527, loss: 0.0837029442191124
2023-01-21 11:12:07.722824: step: 1544/527, loss: 0.05651798099279404
2023-01-21 11:12:08.844793: step: 1548/527, loss: 0.019631672650575638
2023-01-21 11:12:09.980207: step: 1552/527, loss: 0.021854113787412643
2023-01-21 11:12:11.100409: step: 1556/527, loss: 0.018078995868563652
2023-01-21 11:12:12.222381: step: 1560/527, loss: 0.016287613660097122
2023-01-21 11:12:13.362826: step: 1564/527, loss: 0.00038185121957212687
2023-01-21 11:12:14.530413: step: 1568/527, loss: 0.004917717073112726
2023-01-21 11:12:15.649989: step: 1572/527, loss: 0.028669262304902077
2023-01-21 11:12:16.774128: step: 1576/527, loss: 0.019704438745975494
2023-01-21 11:12:17.908617: step: 1580/527, loss: 0.007667827419936657
2023-01-21 11:12:19.036886: step: 1584/527, loss: 0.02697582356631756
2023-01-21 11:12:20.181871: step: 1588/527, loss: 0.02042541466653347
2023-01-21 11:12:21.277618: step: 1592/527, loss: 0.030007362365722656
2023-01-21 11:12:22.406222: step: 1596/527, loss: 0.001234936760738492
2023-01-21 11:12:23.554735: step: 1600/527, loss: 0.021441077813506126
2023-01-21 11:12:24.709033: step: 1604/527, loss: 0.05410599708557129
2023-01-21 11:12:25.860532: step: 1608/527, loss: 0.007120895199477673
2023-01-21 11:12:27.000131: step: 1612/527, loss: 0.014181185513734818
2023-01-21 11:12:28.157301: step: 1616/527, loss: 0.016245556995272636
2023-01-21 11:12:29.311165: step: 1620/527, loss: 0.26022282242774963
2023-01-21 11:12:30.486264: step: 1624/527, loss: 0.015086937695741653
2023-01-21 11:12:31.638311: step: 1628/527, loss: 0.04329581931233406
2023-01-21 11:12:32.771468: step: 1632/527, loss: 0.0022755623795092106
2023-01-21 11:12:33.911747: step: 1636/527, loss: 0.0029495239723473787
2023-01-21 11:12:35.059033: step: 1640/527, loss: 0.16014838218688965
2023-01-21 11:12:36.201967: step: 1644/527, loss: 0.10472507774829865
2023-01-21 11:12:37.315768: step: 1648/527, loss: 0.004257449880242348
2023-01-21 11:12:38.432043: step: 1652/527, loss: 0.009605741128325462
2023-01-21 11:12:39.574370: step: 1656/527, loss: 0.014307642355561256
2023-01-21 11:12:40.704498: step: 1660/527, loss: 0.0010386466747149825
2023-01-21 11:12:41.832743: step: 1664/527, loss: 0.012788581661880016
2023-01-21 11:12:42.974251: step: 1668/527, loss: 0.004460525698959827
2023-01-21 11:12:44.105562: step: 1672/527, loss: 0.06954727321863174
2023-01-21 11:12:45.247410: step: 1676/527, loss: 0.0713268294930458
2023-01-21 11:12:46.372791: step: 1680/527, loss: 0.2839442491531372
2023-01-21 11:12:47.491973: step: 1684/527, loss: 0.004411220550537109
2023-01-21 11:12:48.610138: step: 1688/527, loss: 0.016005326062440872
2023-01-21 11:12:49.704225: step: 1692/527, loss: 0.007894039154052734
2023-01-21 11:12:50.834108: step: 1696/527, loss: 0.004533099941909313
2023-01-21 11:12:51.949602: step: 1700/527, loss: 0.004893636330962181
2023-01-21 11:12:53.060088: step: 1704/527, loss: 0.001990080112591386
2023-01-21 11:12:54.206158: step: 1708/527, loss: 0.0029388905968517065
2023-01-21 11:12:55.325516: step: 1712/527, loss: 0.000920200371183455
2023-01-21 11:12:56.439991: step: 1716/527, loss: 0.11655457317829132
2023-01-21 11:12:57.568197: step: 1720/527, loss: 0.0239060390740633
2023-01-21 11:12:58.717140: step: 1724/527, loss: 0.0016626358265057206
2023-01-21 11:12:59.839180: step: 1728/527, loss: 0.012367535382509232
2023-01-21 11:13:00.969288: step: 1732/527, loss: 0.00020959675021003932
2023-01-21 11:13:02.102686: step: 1736/527, loss: 0.002056598663330078
2023-01-21 11:13:03.243221: step: 1740/527, loss: 0.14610633254051208
2023-01-21 11:13:04.365982: step: 1744/527, loss: 0.06456060707569122
2023-01-21 11:13:05.515346: step: 1748/527, loss: 0.08133984357118607
2023-01-21 11:13:06.632242: step: 1752/527, loss: 0.08122367411851883
2023-01-21 11:13:07.804559: step: 1756/527, loss: 0.0542144775390625
2023-01-21 11:13:08.924233: step: 1760/527, loss: 0.0044806962832808495
2023-01-21 11:13:10.064317: step: 1764/527, loss: 0.14074520766735077
2023-01-21 11:13:11.205048: step: 1768/527, loss: 0.03765411674976349
2023-01-21 11:13:12.335097: step: 1772/527, loss: 0.006587505340576172
2023-01-21 11:13:13.463023: step: 1776/527, loss: 0.0027670860290527344
2023-01-21 11:13:14.592889: step: 1780/527, loss: 0.003584956983104348
2023-01-21 11:13:15.732673: step: 1784/527, loss: 0.010698128491640091
2023-01-21 11:13:16.927054: step: 1788/527, loss: 0.03421001508831978
2023-01-21 11:13:18.072993: step: 1792/527, loss: 0.009135819040238857
2023-01-21 11:13:19.190528: step: 1796/527, loss: 0.03324737399816513
2023-01-21 11:13:20.315241: step: 1800/527, loss: 0.0672907829284668
2023-01-21 11:13:21.432443: step: 1804/527, loss: 0.12344705313444138
2023-01-21 11:13:22.569202: step: 1808/527, loss: 0.09816322475671768
2023-01-21 11:13:23.679194: step: 1812/527, loss: 0.0033520699944347143
2023-01-21 11:13:24.809785: step: 1816/527, loss: 0.0016025543445721269
2023-01-21 11:13:25.973526: step: 1820/527, loss: 0.024089908227324486
2023-01-21 11:13:27.105371: step: 1824/527, loss: 0.00564842252060771
2023-01-21 11:13:28.202870: step: 1828/527, loss: 0.03233394771814346
2023-01-21 11:13:29.334677: step: 1832/527, loss: 0.019582130014896393
2023-01-21 11:13:30.465352: step: 1836/527, loss: 0.0007551193702965975
2023-01-21 11:13:31.605078: step: 1840/527, loss: 0.04893064498901367
2023-01-21 11:13:32.713404: step: 1844/527, loss: 0.08800573647022247
2023-01-21 11:13:33.850888: step: 1848/527, loss: 0.03629150241613388
2023-01-21 11:13:34.975749: step: 1852/527, loss: 0.005641841795295477
2023-01-21 11:13:36.112033: step: 1856/527, loss: 0.022507095709443092
2023-01-21 11:13:37.276406: step: 1860/527, loss: 0.6415468454360962
2023-01-21 11:13:38.420602: step: 1864/527, loss: 0.05641660839319229
2023-01-21 11:13:39.537168: step: 1868/527, loss: 0.08890552818775177
2023-01-21 11:13:40.666112: step: 1872/527, loss: 0.0191650390625
2023-01-21 11:13:41.803615: step: 1876/527, loss: 0.1533823013305664
2023-01-21 11:13:42.956481: step: 1880/527, loss: 0.011720657348632812
2023-01-21 11:13:44.067847: step: 1884/527, loss: 0.02922229655086994
2023-01-21 11:13:45.190433: step: 1888/527, loss: 0.0018682957161217928
2023-01-21 11:13:46.330745: step: 1892/527, loss: 0.08601666241884232
2023-01-21 11:13:47.493187: step: 1896/527, loss: 0.026267433539032936
2023-01-21 11:13:48.616710: step: 1900/527, loss: 0.0032966614235192537
2023-01-21 11:13:49.747390: step: 1904/527, loss: 0.0005238056182861328
2023-01-21 11:13:50.840180: step: 1908/527, loss: 0.0026113989297300577
2023-01-21 11:13:51.962031: step: 1912/527, loss: 0.006095409393310547
2023-01-21 11:13:53.087398: step: 1916/527, loss: 0.001840448472648859
2023-01-21 11:13:54.260894: step: 1920/527, loss: 0.0058341980911791325
2023-01-21 11:13:55.418135: step: 1924/527, loss: 0.24067649245262146
2023-01-21 11:13:56.557570: step: 1928/527, loss: 0.05374440923333168
2023-01-21 11:13:57.714556: step: 1932/527, loss: 0.008095169439911842
2023-01-21 11:13:58.832465: step: 1936/527, loss: 0.027173995971679688
2023-01-21 11:13:59.944880: step: 1940/527, loss: 0.009208775125443935
2023-01-21 11:14:01.070982: step: 1944/527, loss: 0.07383685559034348
2023-01-21 11:14:02.186364: step: 1948/527, loss: 0.0012336254585534334
2023-01-21 11:14:03.284782: step: 1952/527, loss: 0.013745879754424095
2023-01-21 11:14:04.414925: step: 1956/527, loss: 0.0004837036249227822
2023-01-21 11:14:05.552422: step: 1960/527, loss: 0.003995132632553577
2023-01-21 11:14:06.690171: step: 1964/527, loss: 0.0010470390552654862
2023-01-21 11:14:07.826958: step: 1968/527, loss: 0.0015604018699377775
2023-01-21 11:14:08.957659: step: 1972/527, loss: 0.027089644223451614
2023-01-21 11:14:10.087143: step: 1976/527, loss: 0.03243446722626686
2023-01-21 11:14:11.256504: step: 1980/527, loss: 0.028348732739686966
2023-01-21 11:14:12.352334: step: 1984/527, loss: 0.006929552648216486
2023-01-21 11:14:13.495288: step: 1988/527, loss: 0.00692596472799778
2023-01-21 11:14:14.612957: step: 1992/527, loss: 0.02545328438282013
2023-01-21 11:14:15.746235: step: 1996/527, loss: 0.007973099127411842
2023-01-21 11:14:16.907015: step: 2000/527, loss: 0.009815883822739124
2023-01-21 11:14:18.048981: step: 2004/527, loss: 0.01253671757876873
2023-01-21 11:14:19.203810: step: 2008/527, loss: 0.008331346325576305
2023-01-21 11:14:20.410692: step: 2012/527, loss: 0.10944052040576935
2023-01-21 11:14:21.522070: step: 2016/527, loss: 0.15852084755897522
2023-01-21 11:14:22.652126: step: 2020/527, loss: 0.000957489013671875
2023-01-21 11:14:23.823123: step: 2024/527, loss: 0.03190651163458824
2023-01-21 11:14:24.999129: step: 2028/527, loss: 0.021916961297392845
2023-01-21 11:14:26.094844: step: 2032/527, loss: 0.14196662604808807
2023-01-21 11:14:27.255700: step: 2036/527, loss: 0.0024091722443699837
2023-01-21 11:14:28.423022: step: 2040/527, loss: 0.07789897918701172
2023-01-21 11:14:29.543433: step: 2044/527, loss: 0.048813819885253906
2023-01-21 11:14:30.676929: step: 2048/527, loss: 0.02327709272503853
2023-01-21 11:14:31.796945: step: 2052/527, loss: 0.018462779000401497
2023-01-21 11:14:32.960887: step: 2056/527, loss: 0.09607505798339844
2023-01-21 11:14:34.097539: step: 2060/527, loss: 0.027627278119325638
2023-01-21 11:14:35.228693: step: 2064/527, loss: 0.002675390336662531
2023-01-21 11:14:36.350262: step: 2068/527, loss: 0.0028102875221520662
2023-01-21 11:14:37.478034: step: 2072/527, loss: 0.0029817582108080387
2023-01-21 11:14:38.618952: step: 2076/527, loss: 0.03514309227466583
2023-01-21 11:14:39.749053: step: 2080/527, loss: 0.02349252812564373
2023-01-21 11:14:40.872406: step: 2084/527, loss: 0.04308328405022621
2023-01-21 11:14:42.003907: step: 2088/527, loss: 0.3347829282283783
2023-01-21 11:14:43.119879: step: 2092/527, loss: 0.03971891477704048
2023-01-21 11:14:44.228472: step: 2096/527, loss: 0.03657674789428711
2023-01-21 11:14:45.337946: step: 2100/527, loss: 0.012680244632065296
2023-01-21 11:14:46.493568: step: 2104/527, loss: 0.007027816958725452
2023-01-21 11:14:47.639778: step: 2108/527, loss: 0.12496252357959747
==================================================
Loss: 0.043
--------------------
Dev: {'event': {'p': 0.579, 'r': 0.7709720372836218, 'f1': 0.6613363792118789}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15}
Test: {'event': {'p': 0.6074235807860262, 'r': 0.7948571428571428, 'f1': 0.6886138613861386}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15}
Chinese: {'event': {'p': 0.5348837209302325, 'r': 0.8518518518518519, 'f1': 0.6571428571428571}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15}
Korean: {'event': {'p': 0.5606060606060606, 'r': 0.5873015873015873, 'f1': 0.5736434108527131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15}
Russian: {'event': {'p': 0.4146341463414634, 'r': 0.4722222222222222, 'f1': 0.4415584415584415}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6232044198895028, 'r': 0.7509986684420772, 'f1': 0.6811594202898552}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Eng Test for Korean: {'event': {'p': 0.614123006833713, 'r': 0.7702857142857142, 'f1': 0.6833967046894803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7}
Sample 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': 7}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:15:31.468277: step: 4/527, loss: 0.01796550862491131
2023-01-21 11:15:32.630083: step: 8/527, loss: 0.003760433290153742
2023-01-21 11:15:33.758676: step: 12/527, loss: 0.0010778427822515368
2023-01-21 11:15:34.864146: step: 16/527, loss: 0.00014400482177734375
2023-01-21 11:15:36.021142: step: 20/527, loss: 0.02779836766421795
2023-01-21 11:15:37.188115: step: 24/527, loss: 0.02455463446676731
2023-01-21 11:15:38.324000: step: 28/527, loss: 0.0013832091353833675
2023-01-21 11:15:39.446243: step: 32/527, loss: 0.0024348259903490543
2023-01-21 11:15:40.584650: step: 36/527, loss: 0.03339195251464844
2023-01-21 11:15:41.741921: step: 40/527, loss: 0.09266510605812073
2023-01-21 11:15:42.869440: step: 44/527, loss: 0.002629089169204235
2023-01-21 11:15:43.996911: step: 48/527, loss: 0.0015424728626385331
2023-01-21 11:15:45.122313: step: 52/527, loss: 0.0314350388944149
2023-01-21 11:15:46.241180: step: 56/527, loss: 0.029792023822665215
2023-01-21 11:15:47.356033: step: 60/527, loss: 0.03655251860618591
2023-01-21 11:15:48.501336: step: 64/527, loss: 0.018351269885897636
2023-01-21 11:15:49.599171: step: 68/527, loss: 0.018922902643680573
2023-01-21 11:15:50.745606: step: 72/527, loss: 0.010276054963469505
2023-01-21 11:15:51.907196: step: 76/527, loss: 0.010428142733871937
2023-01-21 11:15:53.038948: step: 80/527, loss: 0.0011554717784747481
2023-01-21 11:15:54.154271: step: 84/527, loss: 0.061025045812129974
2023-01-21 11:15:55.258807: step: 88/527, loss: 0.014474892988801003
2023-01-21 11:15:56.396507: step: 92/527, loss: 7.530748553108424e-05
2023-01-21 11:15:57.526600: step: 96/527, loss: 0.014089679345488548
2023-01-21 11:15:58.671375: step: 100/527, loss: 0.032923128455877304
2023-01-21 11:15:59.807894: step: 104/527, loss: 0.004480648320168257
2023-01-21 11:16:00.926266: step: 108/527, loss: 0.0306764617562294
2023-01-21 11:16:02.032313: step: 112/527, loss: 0.003929805941879749
2023-01-21 11:16:03.182300: step: 116/527, loss: 0.03920435905456543
2023-01-21 11:16:04.282131: step: 120/527, loss: 0.005666923709213734
2023-01-21 11:16:05.407479: step: 124/527, loss: 0.0018841744167730212
2023-01-21 11:16:06.530276: step: 128/527, loss: 0.004389571957290173
2023-01-21 11:16:07.668558: step: 132/527, loss: 0.0018618584144860506
2023-01-21 11:16:08.835721: step: 136/527, loss: 0.022548485547304153
2023-01-21 11:16:09.967056: step: 140/527, loss: 0.07386932522058487
2023-01-21 11:16:11.105887: step: 144/527, loss: 0.036187365651130676
2023-01-21 11:16:12.194290: step: 148/527, loss: 0.00187263498082757
2023-01-21 11:16:13.314209: step: 152/527, loss: 0.007227516267448664
2023-01-21 11:16:14.450127: step: 156/527, loss: 0.02659015730023384
2023-01-21 11:16:15.597763: step: 160/527, loss: 0.08658161014318466
2023-01-21 11:16:16.738356: step: 164/527, loss: 0.003027343889698386
2023-01-21 11:16:17.893088: step: 168/527, loss: 0.041294097900390625
2023-01-21 11:16:19.037769: step: 172/527, loss: 0.01939373090863228
2023-01-21 11:16:20.186723: step: 176/527, loss: 0.003173828125
2023-01-21 11:16:21.315343: step: 180/527, loss: 0.045679762959480286
2023-01-21 11:16:22.429727: step: 184/527, loss: 0.017844010144472122
2023-01-21 11:16:23.559565: step: 188/527, loss: 0.0351131409406662
2023-01-21 11:16:24.695529: step: 192/527, loss: 0.0023085596039891243
2023-01-21 11:16:25.827861: step: 196/527, loss: 0.02093084529042244
2023-01-21 11:16:26.978464: step: 200/527, loss: 0.007523871026933193
2023-01-21 11:16:28.097374: step: 204/527, loss: 0.013370228931307793
2023-01-21 11:16:29.220198: step: 208/527, loss: 0.0009019851568154991
2023-01-21 11:16:30.326900: step: 212/527, loss: 0.0714532881975174
2023-01-21 11:16:31.488962: step: 216/527, loss: 0.06932735443115234
2023-01-21 11:16:32.619929: step: 220/527, loss: 0.004470443818718195
2023-01-21 11:16:33.725232: step: 224/527, loss: 0.01931924745440483
2023-01-21 11:16:34.900337: step: 228/527, loss: 0.0025319100823253393
2023-01-21 11:16:36.030250: step: 232/527, loss: 0.0019164086552336812
2023-01-21 11:16:37.141485: step: 236/527, loss: 0.052451133728027344
2023-01-21 11:16:38.273585: step: 240/527, loss: 0.05847921222448349
2023-01-21 11:16:39.389556: step: 244/527, loss: 0.0020477562211453915
2023-01-21 11:16:40.484705: step: 248/527, loss: 0.023990154266357422
2023-01-21 11:16:41.630349: step: 252/527, loss: 0.00993657112121582
2023-01-21 11:16:42.778620: step: 256/527, loss: 0.009394359774887562
2023-01-21 11:16:43.921669: step: 260/527, loss: 0.007433701306581497
2023-01-21 11:16:45.079472: step: 264/527, loss: 0.028497029095888138
2023-01-21 11:16:46.232722: step: 268/527, loss: 0.019512366503477097
2023-01-21 11:16:47.354307: step: 272/527, loss: 0.00151910784188658
2023-01-21 11:16:48.469093: step: 276/527, loss: 0.009914685040712357
2023-01-21 11:16:49.612626: step: 280/527, loss: 0.0026863098610192537
2023-01-21 11:16:50.756582: step: 284/527, loss: 0.02004718966782093
2023-01-21 11:16:51.881194: step: 288/527, loss: 0.003618431044742465
2023-01-21 11:16:53.009525: step: 292/527, loss: 0.022746752947568893
2023-01-21 11:16:54.134543: step: 296/527, loss: 0.024396896362304688
2023-01-21 11:16:55.283279: step: 300/527, loss: 0.03436164930462837
2023-01-21 11:16:56.445989: step: 304/527, loss: 0.004638195037841797
2023-01-21 11:16:57.545324: step: 308/527, loss: 0.013498115353286266
2023-01-21 11:16:58.676493: step: 312/527, loss: 0.07237797230482101
2023-01-21 11:16:59.811482: step: 316/527, loss: 0.022986816242337227
2023-01-21 11:17:00.997772: step: 320/527, loss: 0.009342384524643421
2023-01-21 11:17:02.106944: step: 324/527, loss: 0.0014060974353924394
2023-01-21 11:17:03.237593: step: 328/527, loss: 0.007357978727668524
2023-01-21 11:17:04.378516: step: 332/527, loss: 0.016836928203701973
2023-01-21 11:17:05.495847: step: 336/527, loss: 0.007280445192009211
2023-01-21 11:17:06.618755: step: 340/527, loss: 0.013801097869873047
2023-01-21 11:17:07.738272: step: 344/527, loss: 0.004681110382080078
2023-01-21 11:17:08.862730: step: 348/527, loss: 0.0008754730224609375
2023-01-21 11:17:10.020187: step: 352/527, loss: 0.01829557493329048
2023-01-21 11:17:11.150196: step: 356/527, loss: 0.03895444795489311
2023-01-21 11:17:12.288064: step: 360/527, loss: 0.033132076263427734
2023-01-21 11:17:13.406882: step: 364/527, loss: 0.002926349639892578
2023-01-21 11:17:14.559951: step: 368/527, loss: 0.03482809290289879
2023-01-21 11:17:15.677876: step: 372/527, loss: 0.025053169578313828
2023-01-21 11:17:16.800193: step: 376/527, loss: 0.03913097456097603
2023-01-21 11:17:17.933591: step: 380/527, loss: 0.0005289077525958419
2023-01-21 11:17:19.044780: step: 384/527, loss: 0.00032129290048033
2023-01-21 11:17:20.194559: step: 388/527, loss: 0.0016092300647869706
2023-01-21 11:17:21.272357: step: 392/527, loss: 0.002812719438225031
2023-01-21 11:17:22.359755: step: 396/527, loss: 0.012699365615844727
2023-01-21 11:17:23.477443: step: 400/527, loss: 0.01876373402774334
2023-01-21 11:17:24.593400: step: 404/527, loss: 0.00021696090698242188
2023-01-21 11:17:25.718658: step: 408/527, loss: 0.004645347595214844
2023-01-21 11:17:26.826397: step: 412/527, loss: 0.013503074645996094
2023-01-21 11:17:27.950870: step: 416/527, loss: 0.0683530792593956
2023-01-21 11:17:29.060249: step: 420/527, loss: 0.0010607719887048006
2023-01-21 11:17:30.191155: step: 424/527, loss: 0.015596198849380016
2023-01-21 11:17:31.321190: step: 428/527, loss: 0.004405117128044367
2023-01-21 11:17:32.446166: step: 432/527, loss: 0.013647508807480335
2023-01-21 11:17:33.597098: step: 436/527, loss: 0.01462634839117527
2023-01-21 11:17:34.728946: step: 440/527, loss: 0.004271316342055798
2023-01-21 11:17:35.882009: step: 444/527, loss: 0.002661037491634488
2023-01-21 11:17:37.030936: step: 448/527, loss: 0.13009434938430786
2023-01-21 11:17:38.166799: step: 452/527, loss: 0.02267303504049778
2023-01-21 11:17:39.276617: step: 456/527, loss: 0.04173927754163742
2023-01-21 11:17:40.448647: step: 460/527, loss: 0.012711978517472744
2023-01-21 11:17:41.609290: step: 464/527, loss: 0.012140464968979359
2023-01-21 11:17:42.786863: step: 468/527, loss: 0.00994110107421875
2023-01-21 11:17:43.911434: step: 472/527, loss: 0.035683441907167435
2023-01-21 11:17:45.068955: step: 476/527, loss: 0.10478676855564117
2023-01-21 11:17:46.197291: step: 480/527, loss: 0.0024392367340624332
2023-01-21 11:17:47.306006: step: 484/527, loss: 0.003881168318912387
2023-01-21 11:17:48.500019: step: 488/527, loss: 0.03218808025121689
2023-01-21 11:17:49.607556: step: 492/527, loss: 0.027152251452207565
2023-01-21 11:17:50.755581: step: 496/527, loss: 0.0036808967124670744
2023-01-21 11:17:51.864802: step: 500/527, loss: 0.005822372622787952
2023-01-21 11:17:52.985485: step: 504/527, loss: 0.022550487890839577
2023-01-21 11:17:54.110841: step: 508/527, loss: 0.03886609151959419
2023-01-21 11:17:55.245129: step: 512/527, loss: 0.0007715702522546053
2023-01-21 11:17:56.388017: step: 516/527, loss: 0.021832657977938652
2023-01-21 11:17:57.501601: step: 520/527, loss: 0.03423614427447319
2023-01-21 11:17:58.632873: step: 524/527, loss: 0.02031841315329075
2023-01-21 11:17:59.783491: step: 528/527, loss: 0.0026806830428540707
2023-01-21 11:18:00.903117: step: 532/527, loss: 0.005516243167221546
2023-01-21 11:18:02.035519: step: 536/527, loss: 0.00011639595322776586
2023-01-21 11:18:03.148204: step: 540/527, loss: 0.0018218994373455644
2023-01-21 11:18:04.266691: step: 544/527, loss: 0.0581180565059185
2023-01-21 11:18:05.410619: step: 548/527, loss: 0.00033597947913222015
2023-01-21 11:18:06.537710: step: 552/527, loss: 0.345529168844223
2023-01-21 11:18:07.680818: step: 556/527, loss: 0.010593319311738014
2023-01-21 11:18:08.807233: step: 560/527, loss: 0.013450169004499912
2023-01-21 11:18:09.953205: step: 564/527, loss: 0.004492378327995539
2023-01-21 11:18:11.080864: step: 568/527, loss: 0.0871075689792633
2023-01-21 11:18:12.206437: step: 572/527, loss: 0.00222606654278934
2023-01-21 11:18:13.315501: step: 576/527, loss: 0.0002362251398153603
2023-01-21 11:18:14.469695: step: 580/527, loss: 0.02419462241232395
2023-01-21 11:18:15.651986: step: 584/527, loss: 0.04224071651697159
2023-01-21 11:18:16.789465: step: 588/527, loss: 0.0037647245917469263
2023-01-21 11:18:17.920454: step: 592/527, loss: 0.029642868787050247
2023-01-21 11:18:19.045059: step: 596/527, loss: 0.001161861466243863
2023-01-21 11:18:20.186988: step: 600/527, loss: 0.0027507306076586246
2023-01-21 11:18:21.309849: step: 604/527, loss: 0.013688409700989723
2023-01-21 11:18:22.435381: step: 608/527, loss: 0.04600029066205025
2023-01-21 11:18:23.566314: step: 612/527, loss: 0.015257549472153187
2023-01-21 11:18:24.699177: step: 616/527, loss: 0.0034721374977380037
2023-01-21 11:18:25.833573: step: 620/527, loss: 0.0031791210640221834
2023-01-21 11:18:26.936021: step: 624/527, loss: 0.0018707276321947575
2023-01-21 11:18:28.039497: step: 628/527, loss: 0.08830833435058594
2023-01-21 11:18:29.183370: step: 632/527, loss: 0.003943443298339844
2023-01-21 11:18:30.332085: step: 636/527, loss: 0.047086525708436966
2023-01-21 11:18:31.457297: step: 640/527, loss: 0.017644787207245827
2023-01-21 11:18:32.574888: step: 644/527, loss: 0.012405682355165482
2023-01-21 11:18:33.717578: step: 648/527, loss: 0.00797119177877903
2023-01-21 11:18:34.879751: step: 652/527, loss: 0.049462128430604935
2023-01-21 11:18:35.999917: step: 656/527, loss: 0.08826261013746262
2023-01-21 11:18:37.145910: step: 660/527, loss: 0.023029519245028496
2023-01-21 11:18:38.254209: step: 664/527, loss: 0.05416812747716904
2023-01-21 11:18:39.408432: step: 668/527, loss: 0.04210786521434784
2023-01-21 11:18:40.532502: step: 672/527, loss: 0.49136239290237427
2023-01-21 11:18:41.699580: step: 676/527, loss: 0.033941127359867096
2023-01-21 11:18:42.828960: step: 680/527, loss: 0.00445899972692132
2023-01-21 11:18:43.961023: step: 684/527, loss: 0.05203769728541374
2023-01-21 11:18:45.096211: step: 688/527, loss: 0.0015582084888592362
2023-01-21 11:18:46.205239: step: 692/527, loss: 0.023854637518525124
2023-01-21 11:18:47.370931: step: 696/527, loss: 0.0023053172044456005
2023-01-21 11:18:48.539914: step: 700/527, loss: 0.009936237707734108
2023-01-21 11:18:49.693290: step: 704/527, loss: 0.010214841924607754
2023-01-21 11:18:50.811884: step: 708/527, loss: 0.010755729861557484
2023-01-21 11:18:51.944264: step: 712/527, loss: 0.018483353778719902
2023-01-21 11:18:53.050051: step: 716/527, loss: 0.002727699466049671
2023-01-21 11:18:54.162391: step: 720/527, loss: 0.009651947766542435
2023-01-21 11:18:55.315432: step: 724/527, loss: 0.003670978592708707
2023-01-21 11:18:56.417871: step: 728/527, loss: 0.021419240161776543
2023-01-21 11:18:57.561617: step: 732/527, loss: 0.005337810609489679
2023-01-21 11:18:58.704345: step: 736/527, loss: 0.0007686137687414885
2023-01-21 11:18:59.851764: step: 740/527, loss: 0.016780495643615723
2023-01-21 11:19:00.976942: step: 744/527, loss: 0.00103168492205441
2023-01-21 11:19:02.091148: step: 748/527, loss: 0.01735544204711914
2023-01-21 11:19:03.217128: step: 752/527, loss: 0.011169195175170898
2023-01-21 11:19:04.332809: step: 756/527, loss: 0.04162712022662163
2023-01-21 11:19:05.455926: step: 760/527, loss: 0.003524875734001398
2023-01-21 11:19:06.590753: step: 764/527, loss: 0.008033180609345436
2023-01-21 11:19:07.706268: step: 768/527, loss: 0.006381560117006302
2023-01-21 11:19:08.846734: step: 772/527, loss: 0.007855224423110485
2023-01-21 11:19:10.000373: step: 776/527, loss: 0.36501750349998474
2023-01-21 11:19:11.128006: step: 780/527, loss: 0.08331408351659775
2023-01-21 11:19:12.263878: step: 784/527, loss: 0.012855243869125843
2023-01-21 11:19:13.380758: step: 788/527, loss: 0.000392723100958392
2023-01-21 11:19:14.529053: step: 792/527, loss: 0.023824501782655716
2023-01-21 11:19:15.653062: step: 796/527, loss: 0.10647717118263245
2023-01-21 11:19:16.794603: step: 800/527, loss: 0.03512544557452202
2023-01-21 11:19:17.903752: step: 804/527, loss: 0.011204028502106667
2023-01-21 11:19:19.030719: step: 808/527, loss: 0.0025238990783691406
2023-01-21 11:19:20.166758: step: 812/527, loss: 0.007000732235610485
2023-01-21 11:19:21.316086: step: 816/527, loss: 0.02518615871667862
2023-01-21 11:19:22.435475: step: 820/527, loss: 0.0374029166996479
2023-01-21 11:19:23.590594: step: 824/527, loss: 0.054720211774110794
2023-01-21 11:19:24.721779: step: 828/527, loss: 0.0014748573303222656
2023-01-21 11:19:25.813485: step: 832/527, loss: 0.0024270296562463045
2023-01-21 11:19:26.904403: step: 836/527, loss: 0.02218623086810112
2023-01-21 11:19:28.043749: step: 840/527, loss: 0.07157979160547256
2023-01-21 11:19:29.161260: step: 844/527, loss: 0.040227461606264114
2023-01-21 11:19:30.298754: step: 848/527, loss: 0.0009369850158691406
2023-01-21 11:19:31.453798: step: 852/527, loss: 0.0375455841422081
2023-01-21 11:19:32.623759: step: 856/527, loss: 0.03953418880701065
2023-01-21 11:19:33.757185: step: 860/527, loss: 0.03898897394537926
2023-01-21 11:19:34.875572: step: 864/527, loss: 0.007123851682990789
2023-01-21 11:19:35.987065: step: 868/527, loss: 0.033642008900642395
2023-01-21 11:19:37.106768: step: 872/527, loss: 0.03359575197100639
2023-01-21 11:19:38.251525: step: 876/527, loss: 5.254745337879285e-05
2023-01-21 11:19:39.368086: step: 880/527, loss: 0.02740020863711834
2023-01-21 11:19:40.490221: step: 884/527, loss: 0.03701953962445259
2023-01-21 11:19:41.638478: step: 888/527, loss: 0.030324744060635567
2023-01-21 11:19:42.785349: step: 892/527, loss: 0.0009420395363122225
2023-01-21 11:19:43.944383: step: 896/527, loss: 0.024372100830078125
2023-01-21 11:19:45.040967: step: 900/527, loss: 0.06078539043664932
2023-01-21 11:19:46.169303: step: 904/527, loss: 0.017445659264922142
2023-01-21 11:19:47.305078: step: 908/527, loss: 0.06255665421485901
2023-01-21 11:19:48.434251: step: 912/527, loss: 0.07086696475744247
2023-01-21 11:19:49.569170: step: 916/527, loss: 0.030018115416169167
2023-01-21 11:19:50.696815: step: 920/527, loss: 0.01457139104604721
2023-01-21 11:19:51.875200: step: 924/527, loss: 0.004891872406005859
2023-01-21 11:19:53.036796: step: 928/527, loss: 0.07119999080896378
2023-01-21 11:19:54.198827: step: 932/527, loss: 0.0032543183770030737
2023-01-21 11:19:55.361874: step: 936/527, loss: 0.03839254379272461
2023-01-21 11:19:56.490100: step: 940/527, loss: 0.008921097964048386
2023-01-21 11:19:57.635509: step: 944/527, loss: 0.0021088600624352694
2023-01-21 11:19:58.777677: step: 948/527, loss: 0.036472320556640625
2023-01-21 11:19:59.892899: step: 952/527, loss: 0.0025005340576171875
2023-01-21 11:20:01.035741: step: 956/527, loss: 0.011091423220932484
2023-01-21 11:20:02.176554: step: 960/527, loss: 0.107201486825943
2023-01-21 11:20:03.296212: step: 964/527, loss: 0.014519404619932175
2023-01-21 11:20:04.422670: step: 968/527, loss: 0.005887222476303577
2023-01-21 11:20:05.585970: step: 972/527, loss: 0.010346031747758389
2023-01-21 11:20:06.710297: step: 976/527, loss: 0.03791503980755806
2023-01-21 11:20:07.876423: step: 980/527, loss: 0.04982910305261612
2023-01-21 11:20:09.009109: step: 984/527, loss: 0.016743946820497513
2023-01-21 11:20:10.153975: step: 988/527, loss: 0.01588287390768528
2023-01-21 11:20:11.277735: step: 992/527, loss: 0.008346558548510075
2023-01-21 11:20:12.408805: step: 996/527, loss: 0.012001896277070045
2023-01-21 11:20:13.567687: step: 1000/527, loss: 0.003459155559539795
2023-01-21 11:20:14.718147: step: 1004/527, loss: 0.004527187906205654
2023-01-21 11:20:15.856533: step: 1008/527, loss: 0.002144432161003351
2023-01-21 11:20:16.955408: step: 1012/527, loss: 0.01035156287252903
2023-01-21 11:20:18.092149: step: 1016/527, loss: 0.060842517763376236
2023-01-21 11:20:19.235929: step: 1020/527, loss: 0.01167592965066433
2023-01-21 11:20:20.392108: step: 1024/527, loss: 0.02686777152121067
2023-01-21 11:20:21.527537: step: 1028/527, loss: 0.06627483665943146
2023-01-21 11:20:22.642891: step: 1032/527, loss: 0.0034605979453772306
2023-01-21 11:20:23.768431: step: 1036/527, loss: 0.031024957075715065
2023-01-21 11:20:24.917950: step: 1040/527, loss: 0.10779333114624023
2023-01-21 11:20:26.028489: step: 1044/527, loss: 0.011530781164765358
2023-01-21 11:20:27.145141: step: 1048/527, loss: 0.0011577607365325093
2023-01-21 11:20:28.256293: step: 1052/527, loss: 0.06679125130176544
2023-01-21 11:20:29.374746: step: 1056/527, loss: 0.00014677047147415578
2023-01-21 11:20:30.487003: step: 1060/527, loss: 0.0007278919219970703
2023-01-21 11:20:31.627836: step: 1064/527, loss: 0.022234534844756126
2023-01-21 11:20:32.757448: step: 1068/527, loss: 0.011813163757324219
2023-01-21 11:20:33.917432: step: 1072/527, loss: 0.04335355758666992
2023-01-21 11:20:35.057687: step: 1076/527, loss: 0.0291106216609478
2023-01-21 11:20:36.198636: step: 1080/527, loss: 0.0002305984526174143
2023-01-21 11:20:37.324806: step: 1084/527, loss: 0.022954082116484642
2023-01-21 11:20:38.451597: step: 1088/527, loss: 0.014372778125107288
2023-01-21 11:20:39.599857: step: 1092/527, loss: 0.025388337671756744
2023-01-21 11:20:40.715293: step: 1096/527, loss: 0.004535865969955921
2023-01-21 11:20:41.814508: step: 1100/527, loss: 0.015138912945985794
2023-01-21 11:20:42.938438: step: 1104/527, loss: 0.004838848020881414
2023-01-21 11:20:44.061203: step: 1108/527, loss: 0.006296730134636164
2023-01-21 11:20:45.170175: step: 1112/527, loss: 0.0045442585833370686
2023-01-21 11:20:46.306068: step: 1116/527, loss: 0.0018225193489342928
2023-01-21 11:20:47.430319: step: 1120/527, loss: 5.817413693876006e-05
2023-01-21 11:20:48.563910: step: 1124/527, loss: 0.051497459411621094
2023-01-21 11:20:49.686159: step: 1128/527, loss: 0.035170771181583405
2023-01-21 11:20:50.834639: step: 1132/527, loss: 0.046622373163700104
2023-01-21 11:20:51.957360: step: 1136/527, loss: 0.02722950093448162
2023-01-21 11:20:53.114911: step: 1140/527, loss: -1.9073468138230965e-07
2023-01-21 11:20:54.207100: step: 1144/527, loss: 0.00045289992704056203
2023-01-21 11:20:55.319099: step: 1148/527, loss: 0.013576650060713291
2023-01-21 11:20:56.449540: step: 1152/527, loss: 0.030684662982821465
2023-01-21 11:20:57.606414: step: 1156/527, loss: 0.03546333312988281
2023-01-21 11:20:58.750347: step: 1160/527, loss: 0.0027689458802342415
2023-01-21 11:20:59.857906: step: 1164/527, loss: 0.0054517751559615135
2023-01-21 11:21:00.975340: step: 1168/527, loss: 0.008447457104921341
2023-01-21 11:21:02.068450: step: 1172/527, loss: 0.026525402441620827
2023-01-21 11:21:03.180227: step: 1176/527, loss: 0.004005050752311945
2023-01-21 11:21:04.300190: step: 1180/527, loss: 0.001748037408106029
2023-01-21 11:21:05.408568: step: 1184/527, loss: 0.037444498389959335
2023-01-21 11:21:06.543842: step: 1188/527, loss: 0.06391939520835876
2023-01-21 11:21:07.679416: step: 1192/527, loss: 0.01869945600628853
2023-01-21 11:21:08.815416: step: 1196/527, loss: 0.00645866384729743
2023-01-21 11:21:09.936704: step: 1200/527, loss: 0.022257041186094284
2023-01-21 11:21:11.077863: step: 1204/527, loss: 0.07166080921888351
2023-01-21 11:21:12.180325: step: 1208/527, loss: 0.018420221284031868
2023-01-21 11:21:13.297195: step: 1212/527, loss: 0.0022628784645348787
2023-01-21 11:21:14.418330: step: 1216/527, loss: 0.0005194902187213302
2023-01-21 11:21:15.524132: step: 1220/527, loss: 0.00015821456327103078
2023-01-21 11:21:16.666478: step: 1224/527, loss: 0.008310413919389248
2023-01-21 11:21:17.809073: step: 1228/527, loss: 0.006076240912079811
2023-01-21 11:21:18.932293: step: 1232/527, loss: 0.11685733497142792
2023-01-21 11:21:20.045935: step: 1236/527, loss: 0.006088638212531805
2023-01-21 11:21:21.200486: step: 1240/527, loss: 0.0011143683223053813
2023-01-21 11:21:22.328145: step: 1244/527, loss: 0.0017307281959801912
2023-01-21 11:21:23.454431: step: 1248/527, loss: 0.00015778541273903102
2023-01-21 11:21:24.565785: step: 1252/527, loss: 0.09545579552650452
2023-01-21 11:21:25.700100: step: 1256/527, loss: 0.014463711529970169
2023-01-21 11:21:26.827564: step: 1260/527, loss: 0.008971309289336205
2023-01-21 11:21:27.928689: step: 1264/527, loss: 0.007992362603545189
2023-01-21 11:21:29.066234: step: 1268/527, loss: 0.020517636090517044
2023-01-21 11:21:30.208108: step: 1272/527, loss: 0.042435456067323685
2023-01-21 11:21:31.352411: step: 1276/527, loss: 0.03025665320456028
2023-01-21 11:21:32.476509: step: 1280/527, loss: 0.0012463569873943925
2023-01-21 11:21:33.606837: step: 1284/527, loss: 0.009975815191864967
2023-01-21 11:21:34.738970: step: 1288/527, loss: 0.016057778149843216
2023-01-21 11:21:35.863748: step: 1292/527, loss: 0.05933056026697159
2023-01-21 11:21:37.002873: step: 1296/527, loss: 0.012752151116728783
2023-01-21 11:21:38.141523: step: 1300/527, loss: 0.022022247314453125
2023-01-21 11:21:39.265979: step: 1304/527, loss: 0.0022297382820397615
2023-01-21 11:21:40.400922: step: 1308/527, loss: 0.0024038313422352076
2023-01-21 11:21:41.547626: step: 1312/527, loss: 0.0033733369782567024
2023-01-21 11:21:42.690236: step: 1316/527, loss: 0.03678445890545845
2023-01-21 11:21:43.780734: step: 1320/527, loss: 0.0043600560165941715
2023-01-21 11:21:44.901058: step: 1324/527, loss: 0.01308136060833931
2023-01-21 11:21:46.047674: step: 1328/527, loss: 0.015399932861328125
2023-01-21 11:21:47.230645: step: 1332/527, loss: 0.0035995482467114925
2023-01-21 11:21:48.361446: step: 1336/527, loss: 0.005376053042709827
2023-01-21 11:21:49.474066: step: 1340/527, loss: 0.31654825806617737
2023-01-21 11:21:50.646645: step: 1344/527, loss: 0.0012459754943847656
2023-01-21 11:21:51.804917: step: 1348/527, loss: 0.0011630058288574219
2023-01-21 11:21:52.975699: step: 1352/527, loss: 0.022261619567871094
2023-01-21 11:21:54.080163: step: 1356/527, loss: 0.004092406947165728
2023-01-21 11:21:55.209333: step: 1360/527, loss: 0.0007270813221111894
2023-01-21 11:21:56.381855: step: 1364/527, loss: 0.00796198844909668
2023-01-21 11:21:57.535996: step: 1368/527, loss: 0.009603118523955345
2023-01-21 11:21:58.646347: step: 1372/527, loss: 0.0005840301746502519
2023-01-21 11:21:59.752626: step: 1376/527, loss: 0.921379804611206
2023-01-21 11:22:00.866158: step: 1380/527, loss: 0.007695198059082031
2023-01-21 11:22:02.006139: step: 1384/527, loss: 0.010226774029433727
2023-01-21 11:22:03.135649: step: 1388/527, loss: 0.020283127203583717
2023-01-21 11:22:04.267343: step: 1392/527, loss: 0.043895721435546875
2023-01-21 11:22:05.401582: step: 1396/527, loss: 0.060689929872751236
2023-01-21 11:22:06.530428: step: 1400/527, loss: 0.014291572384536266
2023-01-21 11:22:07.635763: step: 1404/527, loss: 0.03801288455724716
2023-01-21 11:22:08.746694: step: 1408/527, loss: 0.010149812325835228
2023-01-21 11:22:09.899485: step: 1412/527, loss: 0.001574784517288208
2023-01-21 11:22:11.018185: step: 1416/527, loss: 0.0871967077255249
2023-01-21 11:22:12.151511: step: 1420/527, loss: 0.00023174285888671875
2023-01-21 11:22:13.279905: step: 1424/527, loss: 0.04286012426018715
2023-01-21 11:22:14.396192: step: 1428/527, loss: 0.007236289791762829
2023-01-21 11:22:15.535116: step: 1432/527, loss: 0.0009522438049316406
2023-01-21 11:22:16.713292: step: 1436/527, loss: 0.008500671945512295
2023-01-21 11:22:17.846190: step: 1440/527, loss: 0.0034437179565429688
2023-01-21 11:22:18.983371: step: 1444/527, loss: 0.07065653800964355
2023-01-21 11:22:20.169932: step: 1448/527, loss: 0.011742210015654564
2023-01-21 11:22:21.295481: step: 1452/527, loss: 0.00027484894962981343
2023-01-21 11:22:22.414943: step: 1456/527, loss: 0.003152942517772317
2023-01-21 11:22:23.543279: step: 1460/527, loss: 0.06954765319824219
2023-01-21 11:22:24.677998: step: 1464/527, loss: 0.006107473745942116
2023-01-21 11:22:25.829043: step: 1468/527, loss: 0.004796219058334827
2023-01-21 11:22:26.920296: step: 1472/527, loss: 0.007786941714584827
2023-01-21 11:22:28.054508: step: 1476/527, loss: 0.007374858949333429
2023-01-21 11:22:29.198788: step: 1480/527, loss: 0.021729031577706337
2023-01-21 11:22:30.324880: step: 1484/527, loss: 0.06575126200914383
2023-01-21 11:22:31.432999: step: 1488/527, loss: 0.06252014636993408
2023-01-21 11:22:32.548364: step: 1492/527, loss: 0.0047200205735862255
2023-01-21 11:22:33.646813: step: 1496/527, loss: 0.0017782404320314527
2023-01-21 11:22:34.756532: step: 1500/527, loss: 0.00934600830078125
2023-01-21 11:22:35.866120: step: 1504/527, loss: 0.0001566886785440147
2023-01-21 11:22:37.027155: step: 1508/527, loss: 0.032178688794374466
2023-01-21 11:22:38.171900: step: 1512/527, loss: 0.0026976587250828743
2023-01-21 11:22:39.278819: step: 1516/527, loss: 0.01969432830810547
2023-01-21 11:22:40.409765: step: 1520/527, loss: 0.016445541754364967
2023-01-21 11:22:41.557018: step: 1524/527, loss: 0.057711124420166016
2023-01-21 11:22:42.679647: step: 1528/527, loss: 0.00487098703160882
2023-01-21 11:22:43.831532: step: 1532/527, loss: 0.005397987086325884
2023-01-21 11:22:44.978646: step: 1536/527, loss: 0.005328846164047718
2023-01-21 11:22:46.121825: step: 1540/527, loss: 0.022876929491758347
2023-01-21 11:22:47.243765: step: 1544/527, loss: 0.0001852035493357107
2023-01-21 11:22:48.369669: step: 1548/527, loss: 0.0034211156889796257
2023-01-21 11:22:49.490135: step: 1552/527, loss: 0.014150619506835938
2023-01-21 11:22:50.646711: step: 1556/527, loss: 0.0007873058784753084
2023-01-21 11:22:51.758004: step: 1560/527, loss: 0.004472828004509211
2023-01-21 11:22:52.893033: step: 1564/527, loss: 0.013537311926484108
2023-01-21 11:22:54.009080: step: 1568/527, loss: 0.08655796945095062
2023-01-21 11:22:55.146592: step: 1572/527, loss: 0.019203854724764824
2023-01-21 11:22:56.286139: step: 1576/527, loss: 0.007057666778564453
2023-01-21 11:22:57.410815: step: 1580/527, loss: 0.06712310016155243
2023-01-21 11:22:58.548928: step: 1584/527, loss: 0.0003762722190003842
2023-01-21 11:22:59.733679: step: 1588/527, loss: 0.09232282638549805
2023-01-21 11:23:00.862423: step: 1592/527, loss: 0.002775049302726984
2023-01-21 11:23:01.997331: step: 1596/527, loss: 0.011726761236786842
2023-01-21 11:23:03.119571: step: 1600/527, loss: 0.015575027093291283
2023-01-21 11:23:04.267586: step: 1604/527, loss: 0.02930011786520481
2023-01-21 11:23:05.416316: step: 1608/527, loss: 0.012111186981201172
2023-01-21 11:23:06.519269: step: 1612/527, loss: 0.00037741661071777344
2023-01-21 11:23:07.658957: step: 1616/527, loss: 0.0036823274567723274
2023-01-21 11:23:08.821412: step: 1620/527, loss: 0.045983076095581055
2023-01-21 11:23:09.934044: step: 1624/527, loss: 0.0010075569152832031
2023-01-21 11:23:11.040900: step: 1628/527, loss: 0.015353393740952015
2023-01-21 11:23:12.173034: step: 1632/527, loss: 0.017975617200136185
2023-01-21 11:23:13.312468: step: 1636/527, loss: 0.06711041927337646
2023-01-21 11:23:14.438048: step: 1640/527, loss: 0.0044731139205396175
2023-01-21 11:23:15.547071: step: 1644/527, loss: 0.005176353268325329
2023-01-21 11:23:16.678485: step: 1648/527, loss: 0.019369127228856087
2023-01-21 11:23:17.799848: step: 1652/527, loss: 0.00023555755615234375
2023-01-21 11:23:18.932173: step: 1656/527, loss: 0.0156721118837595
2023-01-21 11:23:20.058252: step: 1660/527, loss: 0.04265708848834038
2023-01-21 11:23:21.224420: step: 1664/527, loss: 0.016286659985780716
2023-01-21 11:23:22.327203: step: 1668/527, loss: 0.00028324127197265625
2023-01-21 11:23:23.486054: step: 1672/527, loss: 0.14527744054794312
2023-01-21 11:23:24.615324: step: 1676/527, loss: 0.010168267413973808
2023-01-21 11:23:25.749617: step: 1680/527, loss: 0.00461611757054925
2023-01-21 11:23:26.872373: step: 1684/527, loss: 0.0010302544105798006
2023-01-21 11:23:28.014519: step: 1688/527, loss: 0.004906869027763605
2023-01-21 11:23:29.142798: step: 1692/527, loss: 0.009611368179321289
2023-01-21 11:23:30.271569: step: 1696/527, loss: 0.0039535523392260075
2023-01-21 11:23:31.388300: step: 1700/527, loss: 0.05470981448888779
2023-01-21 11:23:32.556518: step: 1704/527, loss: 0.01680011674761772
2023-01-21 11:23:33.683411: step: 1708/527, loss: 0.10050592571496964
2023-01-21 11:23:34.841301: step: 1712/527, loss: 0.0409025177359581
2023-01-21 11:23:35.999671: step: 1716/527, loss: 0.0009572983253747225
2023-01-21 11:23:37.120403: step: 1720/527, loss: 0.02409229427576065
2023-01-21 11:23:38.256728: step: 1724/527, loss: 0.00488967914134264
2023-01-21 11:23:39.373273: step: 1728/527, loss: 0.014126110821962357
2023-01-21 11:23:40.552131: step: 1732/527, loss: 0.02013702504336834
2023-01-21 11:23:41.676248: step: 1736/527, loss: 0.04024486616253853
2023-01-21 11:23:42.806036: step: 1740/527, loss: 0.0014992713695392013
2023-01-21 11:23:43.923252: step: 1744/527, loss: 0.007378387730568647
2023-01-21 11:23:45.040206: step: 1748/527, loss: 0.1129918098449707
2023-01-21 11:23:46.158441: step: 1752/527, loss: 0.024541517719626427
2023-01-21 11:23:47.286058: step: 1756/527, loss: 0.03152618557214737
2023-01-21 11:23:48.429388: step: 1760/527, loss: 0.0646120086312294
2023-01-21 11:23:49.559689: step: 1764/527, loss: 0.006183242425322533
2023-01-21 11:23:50.735975: step: 1768/527, loss: 0.025460053235292435
2023-01-21 11:23:51.888269: step: 1772/527, loss: 0.002890634583309293
2023-01-21 11:23:53.076973: step: 1776/527, loss: 0.0033100128639489412
2023-01-21 11:23:54.178394: step: 1780/527, loss: 0.41713735461235046
2023-01-21 11:23:55.287371: step: 1784/527, loss: 0.00674018869176507
2023-01-21 11:23:56.420764: step: 1788/527, loss: 0.011173821054399014
2023-01-21 11:23:57.533888: step: 1792/527, loss: 1.888275073724799e-05
2023-01-21 11:23:58.651895: step: 1796/527, loss: 0.11146698147058487
2023-01-21 11:23:59.812772: step: 1800/527, loss: 0.06440496444702148
2023-01-21 11:24:00.961559: step: 1804/527, loss: 0.06115322187542915
2023-01-21 11:24:02.115258: step: 1808/527, loss: 0.01685056835412979
2023-01-21 11:24:03.247749: step: 1812/527, loss: 0.026769066229462624
2023-01-21 11:24:04.372825: step: 1816/527, loss: 0.006432008929550648
2023-01-21 11:24:05.507728: step: 1820/527, loss: 0.03898449242115021
2023-01-21 11:24:06.652031: step: 1824/527, loss: 0.02519378624856472
2023-01-21 11:24:07.797399: step: 1828/527, loss: 0.01451888121664524
2023-01-21 11:24:08.943609: step: 1832/527, loss: 0.015449142083525658
2023-01-21 11:24:10.113094: step: 1836/527, loss: 0.002057361649349332
2023-01-21 11:24:11.225677: step: 1840/527, loss: 0.0005893707275390625
2023-01-21 11:24:12.358986: step: 1844/527, loss: 0.007954454980790615
2023-01-21 11:24:13.464182: step: 1848/527, loss: 0.0884731262922287
2023-01-21 11:24:14.613480: step: 1852/527, loss: 0.034000396728515625
2023-01-21 11:24:15.754267: step: 1856/527, loss: 0.683038055896759
2023-01-21 11:24:16.884542: step: 1860/527, loss: 0.00017032623873092234
2023-01-21 11:24:17.975009: step: 1864/527, loss: 4.692077709478326e-05
2023-01-21 11:24:19.111908: step: 1868/527, loss: 0.03025531768798828
2023-01-21 11:24:20.263040: step: 1872/527, loss: 0.012592792510986328
2023-01-21 11:24:21.420783: step: 1876/527, loss: 0.0018447877373546362
2023-01-21 11:24:22.553005: step: 1880/527, loss: 0.0005530357593670487
2023-01-21 11:24:23.683200: step: 1884/527, loss: 0.09721268713474274
2023-01-21 11:24:24.774147: step: 1888/527, loss: 0.03811340406537056
2023-01-21 11:24:25.901751: step: 1892/527, loss: 0.003135299775749445
2023-01-21 11:24:27.048386: step: 1896/527, loss: 0.0025246620643883944
2023-01-21 11:24:28.165026: step: 1900/527, loss: 0.41839560866355896
2023-01-21 11:24:29.290913: step: 1904/527, loss: 0.00585174560546875
2023-01-21 11:24:30.386410: step: 1908/527, loss: 0.04412689432501793
2023-01-21 11:24:31.519223: step: 1912/527, loss: 0.007361793890595436
2023-01-21 11:24:32.659041: step: 1916/527, loss: 0.0014139175182208419
2023-01-21 11:24:33.793135: step: 1920/527, loss: 0.058358386158943176
2023-01-21 11:24:34.918376: step: 1924/527, loss: 0.0028151513542979956
2023-01-21 11:24:36.034059: step: 1928/527, loss: 0.13248196244239807
2023-01-21 11:24:37.170357: step: 1932/527, loss: 0.011483192443847656
2023-01-21 11:24:38.273516: step: 1936/527, loss: 0.02354288101196289
2023-01-21 11:24:39.392680: step: 1940/527, loss: 0.28597211837768555
2023-01-21 11:24:40.492187: step: 1944/527, loss: 0.022793864831328392
2023-01-21 11:24:41.610732: step: 1948/527, loss: 0.05590400844812393
2023-01-21 11:24:42.779085: step: 1952/527, loss: 0.019139863550662994
2023-01-21 11:24:43.887302: step: 1956/527, loss: 0.029625702649354935
2023-01-21 11:24:45.008864: step: 1960/527, loss: 0.069280244410038
2023-01-21 11:24:46.159411: step: 1964/527, loss: 0.007741022389382124
2023-01-21 11:24:47.297976: step: 1968/527, loss: 0.023952770978212357
2023-01-21 11:24:48.447307: step: 1972/527, loss: 0.009385023266077042
2023-01-21 11:24:49.625015: step: 1976/527, loss: 0.004924774169921875
2023-01-21 11:24:50.761918: step: 1980/527, loss: 0.008701229467988014
2023-01-21 11:24:51.896753: step: 1984/527, loss: 0.025692177936434746
2023-01-21 11:24:52.991193: step: 1988/527, loss: 0.1431722640991211
2023-01-21 11:24:54.135362: step: 1992/527, loss: 0.0931774154305458
2023-01-21 11:24:55.287487: step: 1996/527, loss: 0.003478527069091797
2023-01-21 11:24:56.403375: step: 2000/527, loss: 0.015098334290087223
2023-01-21 11:24:57.553716: step: 2004/527, loss: 0.010980415157973766
2023-01-21 11:24:58.703642: step: 2008/527, loss: 0.049023061990737915
2023-01-21 11:24:59.822200: step: 2012/527, loss: 0.015363835729658604
2023-01-21 11:25:00.970886: step: 2016/527, loss: 0.03264818340539932
2023-01-21 11:25:02.070756: step: 2020/527, loss: 0.0023262263275682926
2023-01-21 11:25:03.213207: step: 2024/527, loss: 0.018448637798428535
2023-01-21 11:25:04.334361: step: 2028/527, loss: 0.5469331741333008
2023-01-21 11:25:05.463815: step: 2032/527, loss: 0.42796844244003296
2023-01-21 11:25:06.585635: step: 2036/527, loss: 0.005140590947121382
2023-01-21 11:25:07.739207: step: 2040/527, loss: 0.024213504046201706
2023-01-21 11:25:08.864033: step: 2044/527, loss: 0.03206014633178711
2023-01-21 11:25:09.973282: step: 2048/527, loss: 0.03523874282836914
2023-01-21 11:25:11.106418: step: 2052/527, loss: 0.06481237709522247
2023-01-21 11:25:12.259262: step: 2056/527, loss: 0.001087188720703125
2023-01-21 11:25:13.409790: step: 2060/527, loss: 0.0037787912879139185
2023-01-21 11:25:14.550679: step: 2064/527, loss: 0.06499803066253662
2023-01-21 11:25:15.659750: step: 2068/527, loss: 0.033544253557920456
2023-01-21 11:25:16.792268: step: 2072/527, loss: 0.0015295982593670487
2023-01-21 11:25:17.899790: step: 2076/527, loss: 0.008977080695331097
2023-01-21 11:25:19.054490: step: 2080/527, loss: 0.02858905866742134
2023-01-21 11:25:20.183537: step: 2084/527, loss: 0.01874542236328125
2023-01-21 11:25:21.293302: step: 2088/527, loss: 0.036957550793886185
2023-01-21 11:25:22.401554: step: 2092/527, loss: 0.0003108978271484375
2023-01-21 11:25:23.530604: step: 2096/527, loss: 0.025326158851385117
2023-01-21 11:25:24.658584: step: 2100/527, loss: 0.008788109757006168
2023-01-21 11:25:25.786210: step: 2104/527, loss: 0.024109981954097748
2023-01-21 11:25:26.979695: step: 2108/527, loss: 0.05748730152845383
==================================================
Loss: 0.032
--------------------
Dev: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Test: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Chinese: {'event': {'p': 0.5581395348837209, 'r': 0.8888888888888888, 'f1': 0.6857142857142857}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Russian: {'event': {'p': 0.5135135135135135, 'r': 0.5277777777777778, 'f1': 0.5205479452054794}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
New best korean model...
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:26:17.098629: step: 4/527, loss: 0.012023543938994408
2023-01-21 11:26:18.220252: step: 8/527, loss: 0.04094543308019638
2023-01-21 11:26:19.322799: step: 12/527, loss: 0.0006013870588503778
2023-01-21 11:26:20.436479: step: 16/527, loss: 0.011975479312241077
2023-01-21 11:26:21.552306: step: 20/527, loss: 0.04260425642132759
2023-01-21 11:26:22.693795: step: 24/527, loss: 0.010041999630630016
2023-01-21 11:26:23.793969: step: 28/527, loss: 0.01732778549194336
2023-01-21 11:26:24.995232: step: 32/527, loss: 0.017427541315555573
2023-01-21 11:26:26.089821: step: 36/527, loss: 0.0002480506955180317
2023-01-21 11:26:27.204465: step: 40/527, loss: 0.0013387680519372225
2023-01-21 11:26:28.322155: step: 44/527, loss: 0.0036637308076024055
2023-01-21 11:26:29.487308: step: 48/527, loss: 0.009945297613739967
2023-01-21 11:26:30.581888: step: 52/527, loss: 0.002210330916568637
2023-01-21 11:26:31.712114: step: 56/527, loss: 0.00011782647197833285
2023-01-21 11:26:32.841721: step: 60/527, loss: 0.010311508551239967
2023-01-21 11:26:33.971342: step: 64/527, loss: 0.02317028120160103
2023-01-21 11:26:35.073878: step: 68/527, loss: 0.00042667388333939016
2023-01-21 11:26:36.208355: step: 72/527, loss: 0.0027159214951097965
2023-01-21 11:26:37.334244: step: 76/527, loss: 0.01687854528427124
2023-01-21 11:26:38.479619: step: 80/527, loss: 0.006239509675651789
2023-01-21 11:26:39.609021: step: 84/527, loss: 0.00021395683870650828
2023-01-21 11:26:40.753350: step: 88/527, loss: 0.00017232896061614156
2023-01-21 11:26:41.869085: step: 92/527, loss: 0.007473564241081476
2023-01-21 11:26:42.996782: step: 96/527, loss: 0.0033813477493822575
2023-01-21 11:26:44.117973: step: 100/527, loss: 0.011180972680449486
2023-01-21 11:26:45.233770: step: 104/527, loss: 0.017066096886992455
2023-01-21 11:26:46.355285: step: 108/527, loss: 0.13654617965221405
2023-01-21 11:26:47.505180: step: 112/527, loss: 0.011764908209443092
2023-01-21 11:26:48.634517: step: 116/527, loss: 0.030293095856904984
2023-01-21 11:26:49.755607: step: 120/527, loss: 0.0072259907610714436
2023-01-21 11:26:50.903584: step: 124/527, loss: 0.0020763396751135588
2023-01-21 11:26:52.017364: step: 128/527, loss: 0.0016427993541583419
2023-01-21 11:26:53.151555: step: 132/527, loss: 0.012667465023696423
2023-01-21 11:26:54.261222: step: 136/527, loss: 0.00536799430847168
2023-01-21 11:26:55.375304: step: 140/527, loss: 0.02067899890244007
2023-01-21 11:26:56.483169: step: 144/527, loss: 0.04240426793694496
2023-01-21 11:26:57.613070: step: 148/527, loss: 0.006385612301528454
2023-01-21 11:26:58.743552: step: 152/527, loss: 0.0005168914794921875
2023-01-21 11:26:59.848631: step: 156/527, loss: 0.05644655600190163
2023-01-21 11:27:00.983722: step: 160/527, loss: 0.015594673343002796
2023-01-21 11:27:02.124486: step: 164/527, loss: 0.0006369590992107987
2023-01-21 11:27:03.256120: step: 168/527, loss: 0.018337726593017578
2023-01-21 11:27:04.418416: step: 172/527, loss: 0.043580248951911926
2023-01-21 11:27:05.517093: step: 176/527, loss: 0.00038948061410337687
2023-01-21 11:27:06.639446: step: 180/527, loss: 0.053795814514160156
2023-01-21 11:27:07.809071: step: 184/527, loss: 0.0003503799671307206
2023-01-21 11:27:08.931477: step: 188/527, loss: 0.0020420073997229338
2023-01-21 11:27:10.025371: step: 192/527, loss: 0.014146232977509499
2023-01-21 11:27:11.152524: step: 196/527, loss: 0.03922691196203232
2023-01-21 11:27:12.278292: step: 200/527, loss: 0.0014178275596350431
2023-01-21 11:27:13.422464: step: 204/527, loss: 0.023692702874541283
2023-01-21 11:27:14.537841: step: 208/527, loss: 0.0002446174621582031
2023-01-21 11:27:15.693167: step: 212/527, loss: 0.00988307036459446
2023-01-21 11:27:16.837484: step: 216/527, loss: 0.2236575186252594
2023-01-21 11:27:18.027049: step: 220/527, loss: 0.02660999447107315
2023-01-21 11:27:19.175993: step: 224/527, loss: 0.03299436718225479
2023-01-21 11:27:20.302228: step: 228/527, loss: 0.0026733397971838713
2023-01-21 11:27:21.446841: step: 232/527, loss: 0.4415349066257477
2023-01-21 11:27:22.558281: step: 236/527, loss: 0.014021254144608974
2023-01-21 11:27:23.655856: step: 240/527, loss: 0.011941814795136452
2023-01-21 11:27:24.805380: step: 244/527, loss: 0.03391771391034126
2023-01-21 11:27:25.928264: step: 248/527, loss: 0.008073044009506702
2023-01-21 11:27:27.057690: step: 252/527, loss: 0.002171135041862726
2023-01-21 11:27:28.191776: step: 256/527, loss: 0.026909636333584785
2023-01-21 11:27:29.344430: step: 260/527, loss: 0.012856770306825638
2023-01-21 11:27:30.468685: step: 264/527, loss: 0.005203152075409889
2023-01-21 11:27:31.568274: step: 268/527, loss: 0.0035745620261877775
2023-01-21 11:27:32.686153: step: 272/527, loss: 0.00018396376981399953
2023-01-21 11:27:33.799636: step: 276/527, loss: 0.5419387817382812
2023-01-21 11:27:34.943620: step: 280/527, loss: 0.04282200336456299
2023-01-21 11:27:36.049500: step: 284/527, loss: -1.4305115314527939e-07
2023-01-21 11:27:37.175822: step: 288/527, loss: 0.01764402538537979
2023-01-21 11:27:38.320524: step: 292/527, loss: 0.00885772705078125
2023-01-21 11:27:39.442681: step: 296/527, loss: 0.04000279679894447
2023-01-21 11:27:40.571579: step: 300/527, loss: 0.00015449525380972773
2023-01-21 11:27:41.726222: step: 304/527, loss: 0.00456314068287611
2023-01-21 11:27:42.842955: step: 308/527, loss: 0.02991771697998047
2023-01-21 11:27:43.972586: step: 312/527, loss: 0.03943290933966637
2023-01-21 11:27:45.065074: step: 316/527, loss: 0.006877040956169367
2023-01-21 11:27:46.269523: step: 320/527, loss: 0.026555204764008522
2023-01-21 11:27:47.388592: step: 324/527, loss: 3.44276413670741e-05
2023-01-21 11:27:48.546752: step: 328/527, loss: 0.009870767593383789
2023-01-21 11:27:49.688795: step: 332/527, loss: 0.0012714386684820056
2023-01-21 11:27:50.808535: step: 336/527, loss: 0.006802749820053577
2023-01-21 11:27:51.982736: step: 340/527, loss: 0.011634444817900658
2023-01-21 11:27:53.108744: step: 344/527, loss: 0.0029994011856615543
2023-01-21 11:27:54.227524: step: 348/527, loss: 0.015510296449065208
2023-01-21 11:27:55.375196: step: 352/527, loss: 0.017762470990419388
2023-01-21 11:27:56.487967: step: 356/527, loss: 0.0029795647133141756
2023-01-21 11:27:57.608579: step: 360/527, loss: 0.004479885566979647
2023-01-21 11:27:58.739180: step: 364/527, loss: 0.03564911335706711
2023-01-21 11:27:59.881610: step: 368/527, loss: 0.018178272992372513
2023-01-21 11:28:00.985972: step: 372/527, loss: 0.004526329226791859
2023-01-21 11:28:02.104046: step: 376/527, loss: 0.0023242949973791838
2023-01-21 11:28:03.254352: step: 380/527, loss: 0.01954975165426731
2023-01-21 11:28:04.434503: step: 384/527, loss: 0.0011478423839434981
2023-01-21 11:28:05.564123: step: 388/527, loss: 0.03871011734008789
2023-01-21 11:28:06.720480: step: 392/527, loss: 0.03529510647058487
2023-01-21 11:28:07.826307: step: 396/527, loss: 0.005182838533073664
2023-01-21 11:28:09.004734: step: 400/527, loss: 0.001241493271663785
2023-01-21 11:28:10.112496: step: 404/527, loss: 0.006157398223876953
2023-01-21 11:28:11.235282: step: 408/527, loss: 0.011606884188950062
2023-01-21 11:28:12.327833: step: 412/527, loss: 0.0034851073287427425
2023-01-21 11:28:13.469028: step: 416/527, loss: 0.00500755337998271
2023-01-21 11:28:14.596207: step: 420/527, loss: 0.05559854954481125
2023-01-21 11:28:15.725572: step: 424/527, loss: 0.0035361291375011206
2023-01-21 11:28:16.856426: step: 428/527, loss: 0.0012041092850267887
2023-01-21 11:28:17.970866: step: 432/527, loss: 0.03157329931855202
2023-01-21 11:28:19.119009: step: 436/527, loss: 0.008916949853301048
2023-01-21 11:28:20.237283: step: 440/527, loss: 0.0005478858947753906
2023-01-21 11:28:21.367456: step: 444/527, loss: 0.0008813858148641884
2023-01-21 11:28:22.503953: step: 448/527, loss: 0.0001125335693359375
2023-01-21 11:28:23.651355: step: 452/527, loss: 0.0183684341609478
2023-01-21 11:28:24.755224: step: 456/527, loss: 0.038498688489198685
2023-01-21 11:28:25.871995: step: 460/527, loss: 0.0067369937896728516
2023-01-21 11:28:26.999220: step: 464/527, loss: 0.008778715506196022
2023-01-21 11:28:28.126590: step: 468/527, loss: 0.008963823318481445
2023-01-21 11:28:29.270278: step: 472/527, loss: 0.00019350051297806203
2023-01-21 11:28:30.382189: step: 476/527, loss: 0.04838857799768448
2023-01-21 11:28:31.507745: step: 480/527, loss: 0.017499923706054688
2023-01-21 11:28:32.650223: step: 484/527, loss: 0.0019974708557128906
2023-01-21 11:28:33.766430: step: 488/527, loss: 7.62939453125e-05
2023-01-21 11:28:34.875999: step: 492/527, loss: 2.040863000729587e-05
2023-01-21 11:28:35.998671: step: 496/527, loss: 0.04352226108312607
2023-01-21 11:28:37.139849: step: 500/527, loss: 0.013031196780502796
2023-01-21 11:28:38.238742: step: 504/527, loss: 0.017792606726288795
2023-01-21 11:28:39.349353: step: 508/527, loss: 0.041040610522031784
2023-01-21 11:28:40.481484: step: 512/527, loss: 0.007233142852783203
2023-01-21 11:28:41.592596: step: 516/527, loss: 0.005320454016327858
2023-01-21 11:28:42.695042: step: 520/527, loss: 0.019151879474520683
2023-01-21 11:28:43.869350: step: 524/527, loss: 0.0002973705413751304
2023-01-21 11:28:45.002050: step: 528/527, loss: 0.02302999421954155
2023-01-21 11:28:46.137057: step: 532/527, loss: 0.00029754641582258046
2023-01-21 11:28:47.287662: step: 536/527, loss: 0.006839322857558727
2023-01-21 11:28:48.426760: step: 540/527, loss: 0.03465862572193146
2023-01-21 11:28:49.564631: step: 544/527, loss: 0.04239311069250107
2023-01-21 11:28:50.686676: step: 548/527, loss: 0.009869957342743874
2023-01-21 11:28:51.836653: step: 552/527, loss: 0.03346814960241318
2023-01-21 11:28:52.956322: step: 556/527, loss: 0.006892109289765358
2023-01-21 11:28:54.100579: step: 560/527, loss: 0.012940024957060814
2023-01-21 11:28:55.197490: step: 564/527, loss: 0.03335418552160263
2023-01-21 11:28:56.361818: step: 568/527, loss: 0.009668446145951748
2023-01-21 11:28:57.490455: step: 572/527, loss: 0.008803081698715687
2023-01-21 11:28:58.607700: step: 576/527, loss: 0.0006324768182821572
2023-01-21 11:28:59.743986: step: 580/527, loss: 0.008044051937758923
2023-01-21 11:29:00.875063: step: 584/527, loss: 0.012132071889936924
2023-01-21 11:29:01.994802: step: 588/527, loss: 0.003170299343764782
2023-01-21 11:29:03.115661: step: 592/527, loss: 0.0005458832019940019
2023-01-21 11:29:04.225580: step: 596/527, loss: 0.044548988342285156
2023-01-21 11:29:05.363508: step: 600/527, loss: 0.031710244715213776
2023-01-21 11:29:06.465163: step: 604/527, loss: 0.046830371022224426
2023-01-21 11:29:07.600305: step: 608/527, loss: 0.0058609009720385075
2023-01-21 11:29:08.739548: step: 612/527, loss: 0.003025627229362726
2023-01-21 11:29:09.883557: step: 616/527, loss: 0.015439033508300781
2023-01-21 11:29:11.003909: step: 620/527, loss: 0.00023517609224654734
2023-01-21 11:29:12.121274: step: 624/527, loss: 0.017304515466094017
2023-01-21 11:29:13.279703: step: 628/527, loss: 0.00124187464825809
2023-01-21 11:29:14.395922: step: 632/527, loss: 0.6089746356010437
2023-01-21 11:29:15.530134: step: 636/527, loss: 0.0002651214599609375
2023-01-21 11:29:16.652660: step: 640/527, loss: 0.12038660049438477
2023-01-21 11:29:17.802743: step: 644/527, loss: 0.04125823825597763
2023-01-21 11:29:18.956016: step: 648/527, loss: 0.002989357803016901
2023-01-21 11:29:20.101802: step: 652/527, loss: 0.0663488358259201
2023-01-21 11:29:21.264804: step: 656/527, loss: 0.021756362169981003
2023-01-21 11:29:22.401074: step: 660/527, loss: 0.050377894192934036
2023-01-21 11:29:23.542698: step: 664/527, loss: 0.0016487122047692537
2023-01-21 11:29:24.649024: step: 668/527, loss: 5.831718590343371e-05
2023-01-21 11:29:25.785664: step: 672/527, loss: 0.010119056329131126
2023-01-21 11:29:26.906836: step: 676/527, loss: 0.00011010170419467613
2023-01-21 11:29:28.021753: step: 680/527, loss: 0.01700625568628311
2023-01-21 11:29:29.164069: step: 684/527, loss: 0.008041572757065296
2023-01-21 11:29:30.312926: step: 688/527, loss: 0.0005945205921307206
2023-01-21 11:29:31.404351: step: 692/527, loss: 0.0014999390114098787
2023-01-21 11:29:32.534066: step: 696/527, loss: 0.006946563720703125
2023-01-21 11:29:33.680906: step: 700/527, loss: 0.0009490966331213713
2023-01-21 11:29:34.824927: step: 704/527, loss: 0.0010449886322021484
2023-01-21 11:29:35.953165: step: 708/527, loss: 0.0012478828430175781
2023-01-21 11:29:37.112223: step: 712/527, loss: 0.008593942038714886
2023-01-21 11:29:38.229556: step: 716/527, loss: 0.028827382251620293
2023-01-21 11:29:39.375995: step: 720/527, loss: 0.003876304719597101
2023-01-21 11:29:40.508068: step: 724/527, loss: 0.1774301528930664
2023-01-21 11:29:41.637217: step: 728/527, loss: 0.0007627487066201866
2023-01-21 11:29:42.780834: step: 732/527, loss: 0.0036920548882335424
2023-01-21 11:29:43.908863: step: 736/527, loss: 0.014010143466293812
2023-01-21 11:29:45.030862: step: 740/527, loss: 0.016585636883974075
2023-01-21 11:29:46.157562: step: 744/527, loss: 0.002349674701690674
2023-01-21 11:29:47.279734: step: 748/527, loss: 0.018917512148618698
2023-01-21 11:29:48.382680: step: 752/527, loss: 0.020033836364746094
2023-01-21 11:29:49.493975: step: 756/527, loss: 0.0008847237331792712
2023-01-21 11:29:50.658155: step: 760/527, loss: 0.04032916948199272
2023-01-21 11:29:51.800010: step: 764/527, loss: 0.00039119721623137593
2023-01-21 11:29:52.913026: step: 768/527, loss: 0.009390830993652344
2023-01-21 11:29:54.071525: step: 772/527, loss: 0.0029379846528172493
2023-01-21 11:29:55.206927: step: 776/527, loss: 0.0011009216541424394
2023-01-21 11:29:56.341878: step: 780/527, loss: 0.0897190049290657
2023-01-21 11:29:57.454786: step: 784/527, loss: 0.014131307601928711
2023-01-21 11:29:58.584921: step: 788/527, loss: 0.012793255038559437
2023-01-21 11:29:59.702121: step: 792/527, loss: 0.010584450326859951
2023-01-21 11:30:00.828282: step: 796/527, loss: 0.012411785311996937
2023-01-21 11:30:01.965773: step: 800/527, loss: 0.01278076134622097
2023-01-21 11:30:03.072526: step: 804/527, loss: 0.01890554465353489
2023-01-21 11:30:04.198243: step: 808/527, loss: 0.016850853338837624
2023-01-21 11:30:05.349842: step: 812/527, loss: 0.026802444830536842
2023-01-21 11:30:06.490399: step: 816/527, loss: 0.011115193367004395
2023-01-21 11:30:07.627548: step: 820/527, loss: 0.0012163162464275956
2023-01-21 11:30:08.754967: step: 824/527, loss: 0.012557792477309704
2023-01-21 11:30:09.884295: step: 828/527, loss: 0.017565656453371048
2023-01-21 11:30:10.999640: step: 832/527, loss: 0.0028841018211096525
2023-01-21 11:30:12.135234: step: 836/527, loss: 0.013046360574662685
2023-01-21 11:30:13.292985: step: 840/527, loss: 0.012713813222944736
2023-01-21 11:30:14.439269: step: 844/527, loss: 0.04854559898376465
2023-01-21 11:30:15.555816: step: 848/527, loss: 0.006130218971520662
2023-01-21 11:30:16.693196: step: 852/527, loss: 0.010632324032485485
2023-01-21 11:30:17.804296: step: 856/527, loss: 0.007375145331025124
2023-01-21 11:30:18.961281: step: 860/527, loss: 0.009763908572494984
2023-01-21 11:30:20.103558: step: 864/527, loss: 0.032332804054021835
2023-01-21 11:30:21.241909: step: 868/527, loss: 0.02527923695743084
2023-01-21 11:30:22.371247: step: 872/527, loss: 0.03210477903485298
2023-01-21 11:30:23.487674: step: 876/527, loss: 0.0006395339732989669
2023-01-21 11:30:24.642848: step: 880/527, loss: 0.09718990325927734
2023-01-21 11:30:25.760828: step: 884/527, loss: 0.00027589796809479594
2023-01-21 11:30:26.855363: step: 888/527, loss: 0.002987766172736883
2023-01-21 11:30:27.971831: step: 892/527, loss: 0.012757696211338043
2023-01-21 11:30:29.110164: step: 896/527, loss: 0.12178345024585724
2023-01-21 11:30:30.287317: step: 900/527, loss: 0.06779947876930237
2023-01-21 11:30:31.413717: step: 904/527, loss: 0.010059070773422718
2023-01-21 11:30:32.564514: step: 908/527, loss: 0.0022483826614916325
2023-01-21 11:30:33.688875: step: 912/527, loss: 0.017208267003297806
2023-01-21 11:30:34.824625: step: 916/527, loss: 0.011341189965605736
2023-01-21 11:30:35.940830: step: 920/527, loss: 0.008183193393051624
2023-01-21 11:30:37.064014: step: 924/527, loss: 0.0021278380881994963
2023-01-21 11:30:38.182580: step: 928/527, loss: 0.014651966281235218
2023-01-21 11:30:39.327273: step: 932/527, loss: 0.09289512783288956
2023-01-21 11:30:40.477087: step: 936/527, loss: 0.025646591559052467
2023-01-21 11:30:41.603656: step: 940/527, loss: 0.15113377571105957
2023-01-21 11:30:42.750972: step: 944/527, loss: 0.01959419436752796
2023-01-21 11:30:43.934037: step: 948/527, loss: 0.04614443704485893
2023-01-21 11:30:45.042456: step: 952/527, loss: 0.0031536102760583162
2023-01-21 11:30:46.164072: step: 956/527, loss: 0.03664245456457138
2023-01-21 11:30:47.296076: step: 960/527, loss: 0.0006259441724978387
2023-01-21 11:30:48.462054: step: 964/527, loss: 0.00119953160174191
2023-01-21 11:30:49.622461: step: 968/527, loss: 0.00019273758516646922
2023-01-21 11:30:50.760051: step: 972/527, loss: 0.013826752081513405
2023-01-21 11:30:51.925094: step: 976/527, loss: 6.198883056640625e-05
2023-01-21 11:30:53.068493: step: 980/527, loss: 0.0074058775790035725
2023-01-21 11:30:54.188616: step: 984/527, loss: 0.0036906241439282894
2023-01-21 11:30:55.338039: step: 988/527, loss: 0.02922978438436985
2023-01-21 11:30:56.437918: step: 992/527, loss: 0.0187088493257761
2023-01-21 11:30:57.591754: step: 996/527, loss: 0.002597617916762829
2023-01-21 11:30:58.758392: step: 1000/527, loss: 0.19203153252601624
2023-01-21 11:30:59.919763: step: 1004/527, loss: 0.0067153931595385075
2023-01-21 11:31:01.040644: step: 1008/527, loss: 0.0017892837058752775
2023-01-21 11:31:02.184253: step: 1012/527, loss: 0.0003285408020019531
2023-01-21 11:31:03.357740: step: 1016/527, loss: 0.004096603486686945
2023-01-21 11:31:04.542949: step: 1020/527, loss: 0.013352966867387295
2023-01-21 11:31:05.631180: step: 1024/527, loss: 0.0005763053777627647
2023-01-21 11:31:06.754171: step: 1028/527, loss: 0.03412031754851341
2023-01-21 11:31:07.903268: step: 1032/527, loss: 0.00015039443678688258
2023-01-21 11:31:09.044106: step: 1036/527, loss: 0.007493781857192516
2023-01-21 11:31:10.190885: step: 1040/527, loss: 0.008114052005112171
2023-01-21 11:31:11.341254: step: 1044/527, loss: 0.04449958726763725
2023-01-21 11:31:12.457798: step: 1048/527, loss: 0.0006418228149414062
2023-01-21 11:31:13.606912: step: 1052/527, loss: 0.030861472710967064
2023-01-21 11:31:14.721303: step: 1056/527, loss: 0.017779922112822533
2023-01-21 11:31:15.835192: step: 1060/527, loss: 0.004881381988525391
2023-01-21 11:31:16.942568: step: 1064/527, loss: 0.01088571548461914
2023-01-21 11:31:18.048343: step: 1068/527, loss: 0.007139015477150679
2023-01-21 11:31:19.154855: step: 1072/527, loss: 0.004183827433735132
2023-01-21 11:31:20.265861: step: 1076/527, loss: 0.028364038094878197
2023-01-21 11:31:21.410454: step: 1080/527, loss: 0.00625190744176507
2023-01-21 11:31:22.582988: step: 1084/527, loss: 0.022876977920532227
2023-01-21 11:31:23.726434: step: 1088/527, loss: 0.1670890897512436
2023-01-21 11:31:24.875785: step: 1092/527, loss: 0.005206489935517311
2023-01-21 11:31:26.000906: step: 1096/527, loss: 0.025931548327207565
2023-01-21 11:31:27.149595: step: 1100/527, loss: 0.0007472038269042969
2023-01-21 11:31:28.297274: step: 1104/527, loss: 0.0023030161391943693
2023-01-21 11:31:29.428249: step: 1108/527, loss: 0.007352924905717373
2023-01-21 11:31:30.576261: step: 1112/527, loss: 0.02576141245663166
2023-01-21 11:31:31.694108: step: 1116/527, loss: 0.017031479626893997
2023-01-21 11:31:32.814032: step: 1120/527, loss: 0.014708328992128372
2023-01-21 11:31:33.933175: step: 1124/527, loss: 0.08752937614917755
2023-01-21 11:31:35.095169: step: 1128/527, loss: 0.04369544982910156
2023-01-21 11:31:36.218617: step: 1132/527, loss: 0.027330685406923294
2023-01-21 11:31:37.349629: step: 1136/527, loss: 6.36577678960748e-05
2023-01-21 11:31:38.503953: step: 1140/527, loss: 0.0549774169921875
2023-01-21 11:31:39.656673: step: 1144/527, loss: 0.0032377243041992188
2023-01-21 11:31:40.813629: step: 1148/527, loss: 0.045659683644771576
2023-01-21 11:31:41.917208: step: 1152/527, loss: 0.01936493068933487
2023-01-21 11:31:43.046716: step: 1156/527, loss: 0.0019378185970708728
2023-01-21 11:31:44.187410: step: 1160/527, loss: 0.002151107881218195
2023-01-21 11:31:45.311504: step: 1164/527, loss: 0.002830600831657648
2023-01-21 11:31:46.422645: step: 1168/527, loss: 0.0023395537864416838
2023-01-21 11:31:47.578616: step: 1172/527, loss: 0.6695175170898438
2023-01-21 11:31:48.687001: step: 1176/527, loss: 0.048874858766794205
2023-01-21 11:31:49.829093: step: 1180/527, loss: 0.0002659797901287675
2023-01-21 11:31:50.961635: step: 1184/527, loss: 0.0029108047019690275
2023-01-21 11:31:52.105269: step: 1188/527, loss: 0.09607252478599548
2023-01-21 11:31:53.251630: step: 1192/527, loss: 0.02913341484963894
2023-01-21 11:31:54.400933: step: 1196/527, loss: 0.0024205208756029606
2023-01-21 11:31:55.542914: step: 1200/527, loss: 0.006854248233139515
2023-01-21 11:31:56.675703: step: 1204/527, loss: 0.00104522705078125
2023-01-21 11:31:57.868917: step: 1208/527, loss: 0.00017652512178756297
2023-01-21 11:31:59.028152: step: 1212/527, loss: 0.00683021591976285
2023-01-21 11:32:00.144554: step: 1216/527, loss: 0.0012537003494799137
2023-01-21 11:32:01.277099: step: 1220/527, loss: 0.005291557405143976
2023-01-21 11:32:02.405331: step: 1224/527, loss: 0.001508522080257535
2023-01-21 11:32:03.532552: step: 1228/527, loss: 0.0008301734924316406
2023-01-21 11:32:04.677286: step: 1232/527, loss: 0.009391403757035732
2023-01-21 11:32:05.791407: step: 1236/527, loss: 0.0007136345375329256
2023-01-21 11:32:06.917337: step: 1240/527, loss: 2.1266936528263614e-05
2023-01-21 11:32:08.035062: step: 1244/527, loss: 0.003770160721614957
2023-01-21 11:32:09.185806: step: 1248/527, loss: 0.0062957764603197575
2023-01-21 11:32:10.340857: step: 1252/527, loss: 0.003944778349250555
2023-01-21 11:32:11.469356: step: 1256/527, loss: 0.04713840410113335
2023-01-21 11:32:12.595873: step: 1260/527, loss: 0.0015480995643883944
2023-01-21 11:32:13.727799: step: 1264/527, loss: 9.126662916969508e-05
2023-01-21 11:32:14.842366: step: 1268/527, loss: 0.021884823217988014
2023-01-21 11:32:16.006970: step: 1272/527, loss: 0.04333152994513512
2023-01-21 11:32:17.151804: step: 1276/527, loss: 0.029973983764648438
2023-01-21 11:32:18.324866: step: 1280/527, loss: 0.016769981011748314
2023-01-21 11:32:19.450023: step: 1284/527, loss: 0.008774567395448685
2023-01-21 11:32:20.573122: step: 1288/527, loss: 0.007098579313606024
2023-01-21 11:32:21.705882: step: 1292/527, loss: 0.025838637724518776
2023-01-21 11:32:22.810502: step: 1296/527, loss: 0.01569361612200737
2023-01-21 11:32:23.931019: step: 1300/527, loss: 0.002582645509392023
2023-01-21 11:32:25.054205: step: 1304/527, loss: 0.0036819458473473787
2023-01-21 11:32:26.172599: step: 1308/527, loss: 0.0008226395002566278
2023-01-21 11:32:27.319551: step: 1312/527, loss: 0.009534453973174095
2023-01-21 11:32:28.456558: step: 1316/527, loss: 0.001239776611328125
2023-01-21 11:32:29.567331: step: 1320/527, loss: 0.025084828957915306
2023-01-21 11:32:30.715330: step: 1324/527, loss: 0.033122241497039795
2023-01-21 11:32:31.825420: step: 1328/527, loss: 0.005719661712646484
2023-01-21 11:32:32.927754: step: 1332/527, loss: 0.0009204388479702175
2023-01-21 11:32:34.049822: step: 1336/527, loss: 0.0003449440118856728
2023-01-21 11:32:35.165805: step: 1340/527, loss: 0.0166518222540617
2023-01-21 11:32:36.304099: step: 1344/527, loss: 0.01487827394157648
2023-01-21 11:32:37.415039: step: 1348/527, loss: 0.004667234607040882
2023-01-21 11:32:38.519680: step: 1352/527, loss: 0.003722286317497492
2023-01-21 11:32:39.610114: step: 1356/527, loss: 0.018856525421142578
2023-01-21 11:32:40.715388: step: 1360/527, loss: 0.00654869107529521
2023-01-21 11:32:41.863563: step: 1364/527, loss: 0.008455371484160423
2023-01-21 11:32:43.014970: step: 1368/527, loss: 0.007340908050537109
2023-01-21 11:32:44.124075: step: 1372/527, loss: 0.010366343893110752
2023-01-21 11:32:45.303021: step: 1376/527, loss: 0.03050079569220543
2023-01-21 11:32:46.430751: step: 1380/527, loss: 0.0019420147873461246
2023-01-21 11:32:47.561943: step: 1384/527, loss: 0.01141667366027832
2023-01-21 11:32:48.705074: step: 1388/527, loss: 0.0011735915904864669
2023-01-21 11:32:49.865040: step: 1392/527, loss: 0.0193634033203125
2023-01-21 11:32:51.006179: step: 1396/527, loss: 2.822876012942288e-05
2023-01-21 11:32:52.161548: step: 1400/527, loss: 0.026920510455965996
2023-01-21 11:32:53.292358: step: 1404/527, loss: 0.0067053320817649364
2023-01-21 11:32:54.405926: step: 1408/527, loss: 0.01545019168406725
2023-01-21 11:32:55.539439: step: 1412/527, loss: 0.07587843388319016
2023-01-21 11:32:56.676002: step: 1416/527, loss: 0.026473617181181908
2023-01-21 11:32:57.788526: step: 1420/527, loss: 0.0024975778069347143
2023-01-21 11:32:58.915364: step: 1424/527, loss: 0.03621387854218483
2023-01-21 11:33:00.033526: step: 1428/527, loss: 0.003635978791862726
2023-01-21 11:33:01.169288: step: 1432/527, loss: 0.022147178649902344
2023-01-21 11:33:02.291735: step: 1436/527, loss: 0.0013586044078692794
2023-01-21 11:33:03.422577: step: 1440/527, loss: 0.014281844720244408
2023-01-21 11:33:04.540139: step: 1444/527, loss: 0.01669750176370144
2023-01-21 11:33:05.694348: step: 1448/527, loss: 0.0029286385979503393
2023-01-21 11:33:06.843644: step: 1452/527, loss: 0.08057241141796112
2023-01-21 11:33:07.975175: step: 1456/527, loss: 0.0029115676879882812
2023-01-21 11:33:09.129440: step: 1460/527, loss: 0.012459207326173782
2023-01-21 11:33:10.265131: step: 1464/527, loss: 0.01417398452758789
2023-01-21 11:33:11.387169: step: 1468/527, loss: 0.04093170166015625
2023-01-21 11:33:12.534753: step: 1472/527, loss: 0.026829909533262253
2023-01-21 11:33:13.676139: step: 1476/527, loss: 0.00031766892061568797
2023-01-21 11:33:14.791961: step: 1480/527, loss: 0.00047388076200149953
2023-01-21 11:33:15.921968: step: 1484/527, loss: 0.0019450187683105469
2023-01-21 11:33:17.070633: step: 1488/527, loss: 0.01916656456887722
2023-01-21 11:33:18.216861: step: 1492/527, loss: 0.04605579748749733
2023-01-21 11:33:19.361666: step: 1496/527, loss: 0.002811384154483676
2023-01-21 11:33:20.488601: step: 1500/527, loss: 0.018135739490389824
2023-01-21 11:33:21.613563: step: 1504/527, loss: 0.0009221077198162675
2023-01-21 11:33:22.743733: step: 1508/527, loss: 0.00060272216796875
2023-01-21 11:33:23.915277: step: 1512/527, loss: 0.05436153709888458
2023-01-21 11:33:25.070502: step: 1516/527, loss: 0.0025850296951830387
2023-01-21 11:33:26.205667: step: 1520/527, loss: 0.03289975970983505
2023-01-21 11:33:27.339880: step: 1524/527, loss: 0.11478567868471146
2023-01-21 11:33:28.464233: step: 1528/527, loss: 0.00229644775390625
2023-01-21 11:33:29.611901: step: 1532/527, loss: 0.11428437381982803
2023-01-21 11:33:30.728435: step: 1536/527, loss: 0.03006310574710369
2023-01-21 11:33:31.830997: step: 1540/527, loss: 0.0005461692926473916
2023-01-21 11:33:32.988022: step: 1544/527, loss: 0.019603919237852097
2023-01-21 11:33:34.137749: step: 1548/527, loss: 0.03254871442914009
2023-01-21 11:33:35.249489: step: 1552/527, loss: 0.008860398083925247
2023-01-21 11:33:36.356098: step: 1556/527, loss: 0.010446262545883656
2023-01-21 11:33:37.502345: step: 1560/527, loss: 0.1184714287519455
2023-01-21 11:33:38.632318: step: 1564/527, loss: 0.05498543009161949
2023-01-21 11:33:39.742131: step: 1568/527, loss: 0.005354690831154585
2023-01-21 11:33:40.883996: step: 1572/527, loss: 0.0021903037559241056
2023-01-21 11:33:41.980559: step: 1576/527, loss: 2.9850007194909267e-05
2023-01-21 11:33:43.125697: step: 1580/527, loss: 0.00227947230450809
2023-01-21 11:33:44.264405: step: 1584/527, loss: 0.002041339874267578
2023-01-21 11:33:45.398113: step: 1588/527, loss: 0.08831587433815002
2023-01-21 11:33:46.511582: step: 1592/527, loss: 0.019071388989686966
2023-01-21 11:33:47.639269: step: 1596/527, loss: 0.013457775115966797
2023-01-21 11:33:48.792491: step: 1600/527, loss: 0.013419151306152344
2023-01-21 11:33:49.921915: step: 1604/527, loss: 0.004733848385512829
2023-01-21 11:33:51.080376: step: 1608/527, loss: 0.022649575024843216
2023-01-21 11:33:52.211250: step: 1612/527, loss: 0.032288696616888046
2023-01-21 11:33:53.320324: step: 1616/527, loss: 0.0032791136763989925
2023-01-21 11:33:54.483369: step: 1620/527, loss: 0.004955387208610773
2023-01-21 11:33:55.597843: step: 1624/527, loss: 0.027863217517733574
2023-01-21 11:33:56.710422: step: 1628/527, loss: 0.010656165890395641
2023-01-21 11:33:57.846414: step: 1632/527, loss: 0.03336024284362793
2023-01-21 11:33:59.006346: step: 1636/527, loss: 0.019741438329219818
2023-01-21 11:34:00.145768: step: 1640/527, loss: 0.022695159539580345
2023-01-21 11:34:01.263750: step: 1644/527, loss: 0.00014290810213424265
2023-01-21 11:34:02.371688: step: 1648/527, loss: 0.00019950867863371968
2023-01-21 11:34:03.509592: step: 1652/527, loss: 0.0030302046798169613
2023-01-21 11:34:04.620897: step: 1656/527, loss: 0.00017089844914153218
2023-01-21 11:34:05.776871: step: 1660/527, loss: 0.014669609256088734
2023-01-21 11:34:06.875778: step: 1664/527, loss: 0.08985739201307297
2023-01-21 11:34:08.029807: step: 1668/527, loss: 0.03051757998764515
2023-01-21 11:34:09.200926: step: 1672/527, loss: 0.06256332993507385
2023-01-21 11:34:10.323617: step: 1676/527, loss: 0.03703415393829346
2023-01-21 11:34:11.441999: step: 1680/527, loss: 0.08514900505542755
2023-01-21 11:34:12.552871: step: 1684/527, loss: 0.04565849527716637
2023-01-21 11:34:13.671782: step: 1688/527, loss: 0.006635952275246382
2023-01-21 11:34:14.766652: step: 1692/527, loss: 0.008678436279296875
2023-01-21 11:34:15.881472: step: 1696/527, loss: 0.027866745367646217
2023-01-21 11:34:16.998567: step: 1700/527, loss: 0.042009733617305756
2023-01-21 11:34:18.129200: step: 1704/527, loss: 0.0009476661798544228
2023-01-21 11:34:19.243408: step: 1708/527, loss: 0.025545883923768997
2023-01-21 11:34:20.395679: step: 1712/527, loss: 0.014864349737763405
2023-01-21 11:34:21.503510: step: 1716/527, loss: 0.0021449089981615543
2023-01-21 11:34:22.641826: step: 1720/527, loss: 0.023329783231019974
2023-01-21 11:34:23.810380: step: 1724/527, loss: 0.045185379683971405
2023-01-21 11:34:24.943754: step: 1728/527, loss: 0.00017251967801712453
2023-01-21 11:34:26.032393: step: 1732/527, loss: 0.015349388122558594
2023-01-21 11:34:27.160101: step: 1736/527, loss: 0.013393688946962357
2023-01-21 11:34:28.285257: step: 1740/527, loss: 0.001576328300870955
2023-01-21 11:34:29.458986: step: 1744/527, loss: 0.006050777621567249
2023-01-21 11:34:30.560093: step: 1748/527, loss: 0.0046484945341944695
2023-01-21 11:34:31.698959: step: 1752/527, loss: 0.022418595850467682
2023-01-21 11:34:32.833960: step: 1756/527, loss: 0.015752125531435013
2023-01-21 11:34:33.942775: step: 1760/527, loss: 0.0025797844864428043
2023-01-21 11:34:35.091711: step: 1764/527, loss: 0.05885200574994087
2023-01-21 11:34:36.213093: step: 1768/527, loss: 0.006206226535141468
2023-01-21 11:34:37.349634: step: 1772/527, loss: 0.002802944276481867
2023-01-21 11:34:38.485877: step: 1776/527, loss: 0.01677875593304634
2023-01-21 11:34:39.596906: step: 1780/527, loss: 0.00016489028348587453
2023-01-21 11:34:40.679804: step: 1784/527, loss: 0.0174605380743742
2023-01-21 11:34:41.842963: step: 1788/527, loss: 0.007315254304558039
2023-01-21 11:34:42.967374: step: 1792/527, loss: 0.006932068150490522
2023-01-21 11:34:44.056903: step: 1796/527, loss: 0.008213234134018421
2023-01-21 11:34:45.206851: step: 1800/527, loss: 0.00031769275665283203
2023-01-21 11:34:46.338361: step: 1804/527, loss: 0.0651666596531868
2023-01-21 11:34:47.487128: step: 1808/527, loss: 0.022185945883393288
2023-01-21 11:34:48.605632: step: 1812/527, loss: 0.01569500006735325
2023-01-21 11:34:49.722581: step: 1816/527, loss: 0.018458176404237747
2023-01-21 11:34:50.853319: step: 1820/527, loss: 0.051366619765758514
2023-01-21 11:34:52.002524: step: 1824/527, loss: 0.007671976462006569
2023-01-21 11:34:53.115379: step: 1828/527, loss: 0.0042264461517333984
2023-01-21 11:34:54.250825: step: 1832/527, loss: 0.008039474487304688
2023-01-21 11:34:55.371697: step: 1836/527, loss: 0.0223586093634367
2023-01-21 11:34:56.480523: step: 1840/527, loss: 0.014232253655791283
2023-01-21 11:34:57.624323: step: 1844/527, loss: 0.036349013447761536
2023-01-21 11:34:58.741118: step: 1848/527, loss: 0.007012367248535156
2023-01-21 11:34:59.857194: step: 1852/527, loss: 4.1484832763671875e-05
2023-01-21 11:35:01.002227: step: 1856/527, loss: 0.017586613073945045
2023-01-21 11:35:02.121200: step: 1860/527, loss: 0.009233379736542702
2023-01-21 11:35:03.255749: step: 1864/527, loss: 9.050369408214465e-05
2023-01-21 11:35:04.411472: step: 1868/527, loss: 0.0008281707996502519
2023-01-21 11:35:05.604059: step: 1872/527, loss: 0.045211028307676315
2023-01-21 11:35:06.746674: step: 1876/527, loss: 0.0004749298095703125
2023-01-21 11:35:07.860194: step: 1880/527, loss: 0.011823082342743874
2023-01-21 11:35:09.024102: step: 1884/527, loss: 0.0005674362182617188
2023-01-21 11:35:10.158348: step: 1888/527, loss: 0.04866781085729599
2023-01-21 11:35:11.286788: step: 1892/527, loss: 0.0034399032592773438
2023-01-21 11:35:12.452048: step: 1896/527, loss: 0.004403114318847656
2023-01-21 11:35:13.631828: step: 1900/527, loss: 0.029995013028383255
2023-01-21 11:35:14.781080: step: 1904/527, loss: 0.005824279971420765
2023-01-21 11:35:15.894165: step: 1908/527, loss: 0.001561021781526506
2023-01-21 11:35:17.008955: step: 1912/527, loss: 0.006012153811752796
2023-01-21 11:35:18.140450: step: 1916/527, loss: 0.001210117363370955
2023-01-21 11:35:19.268373: step: 1920/527, loss: 0.015359211713075638
2023-01-21 11:35:20.411737: step: 1924/527, loss: 0.00070018763653934
2023-01-21 11:35:21.519388: step: 1928/527, loss: 0.012458897195756435
2023-01-21 11:35:22.640603: step: 1932/527, loss: 0.008224201388657093
2023-01-21 11:35:23.753417: step: 1936/527, loss: 0.002225685166195035
2023-01-21 11:35:24.887625: step: 1940/527, loss: 0.04883261024951935
2023-01-21 11:35:26.034638: step: 1944/527, loss: 0.025039149448275566
2023-01-21 11:35:27.180215: step: 1948/527, loss: 0.0023357393220067024
2023-01-21 11:35:28.318677: step: 1952/527, loss: 0.011423682793974876
2023-01-21 11:35:29.421578: step: 1956/527, loss: 0.00042676928569562733
2023-01-21 11:35:30.552672: step: 1960/527, loss: 0.00040721893310546875
2023-01-21 11:35:31.667277: step: 1964/527, loss: 0.03636159747838974
2023-01-21 11:35:32.800634: step: 1968/527, loss: 0.001659393310546875
2023-01-21 11:35:33.897483: step: 1972/527, loss: 0.03954753652215004
2023-01-21 11:35:35.031226: step: 1976/527, loss: 0.0014842988457530737
2023-01-21 11:35:36.140416: step: 1980/527, loss: 0.01186308916658163
2023-01-21 11:35:37.288782: step: 1984/527, loss: 0.06276760250329971
2023-01-21 11:35:38.407281: step: 1988/527, loss: 0.008978081867098808
2023-01-21 11:35:39.520996: step: 1992/527, loss: 0.0027392387855798006
2023-01-21 11:35:40.646650: step: 1996/527, loss: 0.00047826769878156483
2023-01-21 11:35:41.786605: step: 2000/527, loss: 0.0005186080816201866
2023-01-21 11:35:42.900782: step: 2004/527, loss: 0.0012513160472735763
2023-01-21 11:35:44.019530: step: 2008/527, loss: 0.0024038313422352076
2023-01-21 11:35:45.149733: step: 2012/527, loss: 0.011249733157455921
2023-01-21 11:35:46.292361: step: 2016/527, loss: 0.0002080917329294607
2023-01-21 11:35:47.424513: step: 2020/527, loss: 0.05853691324591637
2023-01-21 11:35:48.556609: step: 2024/527, loss: 0.10398578643798828
2023-01-21 11:35:49.683720: step: 2028/527, loss: 0.016141893342137337
2023-01-21 11:35:50.809595: step: 2032/527, loss: 7.305145118152723e-05
2023-01-21 11:35:51.961634: step: 2036/527, loss: 0.0618865042924881
2023-01-21 11:35:53.076202: step: 2040/527, loss: 0.004568052478134632
2023-01-21 11:35:54.202046: step: 2044/527, loss: 0.050662897527217865
2023-01-21 11:35:55.353607: step: 2048/527, loss: 0.009787654504179955
2023-01-21 11:35:56.483875: step: 2052/527, loss: 0.20280705392360687
2023-01-21 11:35:57.636775: step: 2056/527, loss: 0.1688835173845291
2023-01-21 11:35:58.761008: step: 2060/527, loss: 0.008786487393081188
2023-01-21 11:35:59.936727: step: 2064/527, loss: 0.0009190559503622353
2023-01-21 11:36:01.098123: step: 2068/527, loss: 0.00034971238346770406
2023-01-21 11:36:02.245730: step: 2072/527, loss: 0.006160736549645662
2023-01-21 11:36:03.376762: step: 2076/527, loss: 0.018799399957060814
2023-01-21 11:36:04.490715: step: 2080/527, loss: 0.05475263670086861
2023-01-21 11:36:05.614103: step: 2084/527, loss: 0.0007116794586181641
2023-01-21 11:36:06.724806: step: 2088/527, loss: 0.0013433456188067794
2023-01-21 11:36:07.859923: step: 2092/527, loss: 0.048093464225530624
2023-01-21 11:36:08.970456: step: 2096/527, loss: 0.07046828418970108
2023-01-21 11:36:10.091717: step: 2100/527, loss: 0.002475023502483964
2023-01-21 11:36:11.253716: step: 2104/527, loss: 0.2958931028842926
2023-01-21 11:36:12.413820: step: 2108/527, loss: 0.11183682084083557
==================================================
Loss: 0.024
--------------------
Dev: {'event': {'p': 0.584, 'r': 0.7776298268974701, 'f1': 0.6670474014848657}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17}
Test: {'event': {'p': 0.6253430924062214, 'r': 0.7811428571428571, 'f1': 0.6946138211382115}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17}
Chinese: {'event': {'p': 0.5609756097560976, 'r': 0.8518518518518519, 'f1': 0.6764705882352942}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17}
Korean: {'event': {'p': 0.559322033898305, 'r': 0.5238095238095238, 'f1': 0.5409836065573771}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17}
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': 17}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:36:56.222098: step: 4/527, loss: 0.040186502039432526
2023-01-21 11:36:57.349778: step: 8/527, loss: 0.0672091469168663
2023-01-21 11:36:58.491924: step: 12/527, loss: 0.027982331812381744
2023-01-21 11:36:59.648734: step: 16/527, loss: 0.0005046845180913806
2023-01-21 11:37:00.781352: step: 20/527, loss: 0.022781657055020332
2023-01-21 11:37:01.951710: step: 24/527, loss: 0.024577999487519264
2023-01-21 11:37:03.051281: step: 28/527, loss: 0.001166784786619246
2023-01-21 11:37:04.187850: step: 32/527, loss: 0.006028938107192516
2023-01-21 11:37:05.317985: step: 36/527, loss: 0.02263965643942356
2023-01-21 11:37:06.441320: step: 40/527, loss: 0.021024418994784355
2023-01-21 11:37:07.552524: step: 44/527, loss: 0.013406133279204369
2023-01-21 11:37:08.660241: step: 48/527, loss: 3.4761429560603574e-05
2023-01-21 11:37:09.808953: step: 52/527, loss: 0.016150474548339844
2023-01-21 11:37:10.945966: step: 56/527, loss: 0.02635030634701252
2023-01-21 11:37:12.069214: step: 60/527, loss: 0.017991255968809128
2023-01-21 11:37:13.192636: step: 64/527, loss: 0.0016645431751385331
2023-01-21 11:37:14.296503: step: 68/527, loss: 0.0035830498673021793
2023-01-21 11:37:15.417770: step: 72/527, loss: 0.03319501876831055
2023-01-21 11:37:16.529887: step: 76/527, loss: 0.0007450103294104338
2023-01-21 11:37:17.656864: step: 80/527, loss: 0.04156055673956871
2023-01-21 11:37:18.863798: step: 84/527, loss: 0.011931514367461205
2023-01-21 11:37:20.008856: step: 88/527, loss: 0.011863517574965954
2023-01-21 11:37:21.165455: step: 92/527, loss: 0.003681659698486328
2023-01-21 11:37:22.269742: step: 96/527, loss: 0.001239872071892023
2023-01-21 11:37:23.375859: step: 100/527, loss: 0.0032170298509299755
2023-01-21 11:37:24.502482: step: 104/527, loss: 0.010618782602250576
2023-01-21 11:37:25.641303: step: 108/527, loss: 0.002019691513851285
2023-01-21 11:37:26.815852: step: 112/527, loss: 0.030876733362674713
2023-01-21 11:37:27.942763: step: 116/527, loss: 0.009692382998764515
2023-01-21 11:37:29.081984: step: 120/527, loss: 0.03450927883386612
2023-01-21 11:37:30.212953: step: 124/527, loss: 0.0016699791885912418
2023-01-21 11:37:31.336527: step: 128/527, loss: 0.0008918762323446572
2023-01-21 11:37:32.435257: step: 132/527, loss: 0.013242721557617188
2023-01-21 11:37:33.572634: step: 136/527, loss: 0.007526874542236328
2023-01-21 11:37:34.725269: step: 140/527, loss: 0.008176612667739391
2023-01-21 11:37:35.848221: step: 144/527, loss: 0.007441902533173561
2023-01-21 11:37:36.955216: step: 148/527, loss: 6.408691115211695e-05
2023-01-21 11:37:38.065713: step: 152/527, loss: 0.04285097122192383
2023-01-21 11:37:39.213363: step: 156/527, loss: 0.0005171775701455772
2023-01-21 11:37:40.355932: step: 160/527, loss: 0.0024738311767578125
2023-01-21 11:37:41.464881: step: 164/527, loss: 0.01603221893310547
2023-01-21 11:37:42.528953: step: 168/527, loss: 0.006485557649284601
2023-01-21 11:37:43.666190: step: 172/527, loss: 0.001798486802726984
2023-01-21 11:37:44.788161: step: 176/527, loss: 0.008454036898911
2023-01-21 11:37:45.908726: step: 180/527, loss: 0.0202668197453022
2023-01-21 11:37:47.086726: step: 184/527, loss: 0.011136150918900967
2023-01-21 11:37:48.213541: step: 188/527, loss: 0.00013980866060592234
2023-01-21 11:37:49.362051: step: 192/527, loss: 0.0009034157264977694
2023-01-21 11:37:50.491100: step: 196/527, loss: 0.0030145645141601562
2023-01-21 11:37:51.606592: step: 200/527, loss: 0.0327632911503315
2023-01-21 11:37:52.744221: step: 204/527, loss: 0.0033899308182299137
2023-01-21 11:37:53.850074: step: 208/527, loss: 0.1989058405160904
2023-01-21 11:37:54.969210: step: 212/527, loss: 0.0010187149746343493
2023-01-21 11:37:56.116941: step: 216/527, loss: 0.0012699126964434981
2023-01-21 11:37:57.269762: step: 220/527, loss: 0.0005788326379843056
2023-01-21 11:37:58.401817: step: 224/527, loss: 0.004528141114860773
2023-01-21 11:37:59.504502: step: 228/527, loss: 0.0027210235130041838
2023-01-21 11:38:00.634874: step: 232/527, loss: 0.010525750927627087
2023-01-21 11:38:01.781706: step: 236/527, loss: 0.062098123133182526
2023-01-21 11:38:02.892460: step: 240/527, loss: 0.007528495974838734
2023-01-21 11:38:04.024623: step: 244/527, loss: 0.021370315924286842
2023-01-21 11:38:05.127093: step: 248/527, loss: 0.01905994489789009
2023-01-21 11:38:06.270395: step: 252/527, loss: 0.0005049228784628212
2023-01-21 11:38:07.421502: step: 256/527, loss: 0.0009790421463549137
2023-01-21 11:38:08.546378: step: 260/527, loss: 0.00026769639225676656
2023-01-21 11:38:09.682482: step: 264/527, loss: 0.0027523040771484375
2023-01-21 11:38:10.804474: step: 268/527, loss: 0.0002803802490234375
2023-01-21 11:38:11.960612: step: 272/527, loss: 0.10391931235790253
2023-01-21 11:38:13.079193: step: 276/527, loss: 0.00026702880859375
2023-01-21 11:38:14.212896: step: 280/527, loss: 0.00011491775512695312
2023-01-21 11:38:15.378729: step: 284/527, loss: 0.0007890701526775956
2023-01-21 11:38:16.473271: step: 288/527, loss: 0.004887866787612438
2023-01-21 11:38:17.588484: step: 292/527, loss: 0.005933189298957586
2023-01-21 11:38:18.708156: step: 296/527, loss: 0.0014652431709691882
2023-01-21 11:38:19.834368: step: 300/527, loss: 0.03130292892456055
2023-01-21 11:38:20.935335: step: 304/527, loss: 0.015362311154603958
2023-01-21 11:38:22.036897: step: 308/527, loss: 0.011118436232209206
2023-01-21 11:38:23.141898: step: 312/527, loss: 0.04197206348180771
2023-01-21 11:38:24.290700: step: 316/527, loss: 0.0005729199037887156
2023-01-21 11:38:25.381487: step: 320/527, loss: 0.020746994763612747
2023-01-21 11:38:26.491952: step: 324/527, loss: 0.012500714510679245
2023-01-21 11:38:27.595587: step: 328/527, loss: 0.02304520644247532
2023-01-21 11:38:28.686190: step: 332/527, loss: 0.0002158164861612022
2023-01-21 11:38:29.813928: step: 336/527, loss: 0.4049331843852997
2023-01-21 11:38:30.922195: step: 340/527, loss: 0.05718555673956871
2023-01-21 11:38:32.047551: step: 344/527, loss: 0.01338043250143528
2023-01-21 11:38:33.172596: step: 348/527, loss: 0.014764976687729359
2023-01-21 11:38:34.305378: step: 352/527, loss: 0.0810842514038086
2023-01-21 11:38:35.435980: step: 356/527, loss: 0.03073091432452202
2023-01-21 11:38:36.587662: step: 360/527, loss: 0.0013738631969317794
2023-01-21 11:38:37.739224: step: 364/527, loss: 0.0003941059112548828
2023-01-21 11:38:38.900944: step: 368/527, loss: 0.06422261893749237
2023-01-21 11:38:40.023466: step: 372/527, loss: 0.04136824607849121
2023-01-21 11:38:41.140667: step: 376/527, loss: 0.07424090802669525
2023-01-21 11:38:42.272014: step: 380/527, loss: 0.0432492271065712
2023-01-21 11:38:43.485131: step: 384/527, loss: 0.05633983761072159
2023-01-21 11:38:44.600294: step: 388/527, loss: 0.03162508085370064
2023-01-21 11:38:45.721497: step: 392/527, loss: 0.04622936621308327
2023-01-21 11:38:46.858892: step: 396/527, loss: 6.642937660217285e-05
2023-01-21 11:38:47.972216: step: 400/527, loss: 0.022319983690977097
2023-01-21 11:38:49.136550: step: 404/527, loss: 0.014006996527314186
2023-01-21 11:38:50.301591: step: 408/527, loss: 0.04410495609045029
2023-01-21 11:38:51.411660: step: 412/527, loss: 0.029336167499423027
2023-01-21 11:38:52.531706: step: 416/527, loss: 0.003387451171875
2023-01-21 11:38:53.693979: step: 420/527, loss: 0.02224436029791832
2023-01-21 11:38:54.815167: step: 424/527, loss: 0.027769852429628372
2023-01-21 11:38:55.938936: step: 428/527, loss: 0.017683982849121094
2023-01-21 11:38:57.075800: step: 432/527, loss: 0.015288162976503372
2023-01-21 11:38:58.214958: step: 436/527, loss: 0.030890941619873047
2023-01-21 11:38:59.373904: step: 440/527, loss: 0.33327770233154297
2023-01-21 11:39:00.519821: step: 444/527, loss: 0.07016811519861221
2023-01-21 11:39:01.648828: step: 448/527, loss: 0.04353253170847893
2023-01-21 11:39:02.804295: step: 452/527, loss: 3.261566234868951e-05
2023-01-21 11:39:03.983619: step: 456/527, loss: 0.3369169235229492
2023-01-21 11:39:05.099732: step: 460/527, loss: 0.002392006106674671
2023-01-21 11:39:06.203959: step: 464/527, loss: 0.006875419523566961
2023-01-21 11:39:07.369857: step: 468/527, loss: 0.03046569786965847
2023-01-21 11:39:08.503404: step: 472/527, loss: 0.006514167878776789
2023-01-21 11:39:09.656088: step: 476/527, loss: 0.008658219128847122
2023-01-21 11:39:10.795674: step: 480/527, loss: 0.0027048110496252775
2023-01-21 11:39:11.924993: step: 484/527, loss: 0.0037185668479651213
2023-01-21 11:39:13.061068: step: 488/527, loss: 0.01454925537109375
2023-01-21 11:39:14.199862: step: 492/527, loss: 0.010749674402177334
2023-01-21 11:39:15.313011: step: 496/527, loss: 0.00014638900756835938
2023-01-21 11:39:16.473579: step: 500/527, loss: 0.02398385852575302
2023-01-21 11:39:17.573362: step: 504/527, loss: 0.00013920068158768117
2023-01-21 11:39:18.700440: step: 508/527, loss: 0.0033648491371423006
2023-01-21 11:39:19.821082: step: 512/527, loss: 0.017268039286136627
2023-01-21 11:39:20.953320: step: 516/527, loss: 0.0032089711166918278
2023-01-21 11:39:22.081790: step: 520/527, loss: 0.014353180304169655
2023-01-21 11:39:23.239403: step: 524/527, loss: 0.020438384264707565
2023-01-21 11:39:24.413724: step: 528/527, loss: 0.02605152130126953
2023-01-21 11:39:25.598926: step: 532/527, loss: 0.004709052853286266
2023-01-21 11:39:26.710748: step: 536/527, loss: 0.0019268036121502519
2023-01-21 11:39:27.828902: step: 540/527, loss: 0.0007682800642214715
2023-01-21 11:39:28.967261: step: 544/527, loss: 0.025227930396795273
2023-01-21 11:39:30.107728: step: 548/527, loss: 8.37326078908518e-05
2023-01-21 11:39:31.215982: step: 552/527, loss: 0.013734627515077591
2023-01-21 11:39:32.352505: step: 556/527, loss: 0.05985736846923828
2023-01-21 11:39:33.502984: step: 560/527, loss: 0.0016315460670739412
2023-01-21 11:39:34.621139: step: 564/527, loss: 0.0042414190247654915
2023-01-21 11:39:35.738597: step: 568/527, loss: 0.06443500518798828
2023-01-21 11:39:36.887818: step: 572/527, loss: 0.012848282232880592
2023-01-21 11:39:38.004880: step: 576/527, loss: 0.025479793548583984
2023-01-21 11:39:39.144035: step: 580/527, loss: 0.00021419525728560984
2023-01-21 11:39:40.272193: step: 584/527, loss: 0.0008415222400799394
2023-01-21 11:39:41.403029: step: 588/527, loss: 0.02218952216207981
2023-01-21 11:39:42.541031: step: 592/527, loss: 0.004956436343491077
2023-01-21 11:39:43.715356: step: 596/527, loss: 0.0010934829479083419
2023-01-21 11:39:44.865826: step: 600/527, loss: 0.016013814136385918
2023-01-21 11:39:45.992652: step: 604/527, loss: 6.909370858920738e-05
2023-01-21 11:39:47.166309: step: 608/527, loss: 0.0004241943533997983
2023-01-21 11:39:48.261580: step: 612/527, loss: 0.040265657007694244
2023-01-21 11:39:49.436875: step: 616/527, loss: 0.03331737592816353
2023-01-21 11:39:50.536570: step: 620/527, loss: 0.004127550404518843
2023-01-21 11:39:51.664994: step: 624/527, loss: 0.011080646887421608
2023-01-21 11:39:52.781514: step: 628/527, loss: 0.004027175717055798
2023-01-21 11:39:53.908278: step: 632/527, loss: 0.003848457243293524
2023-01-21 11:39:55.039206: step: 636/527, loss: 0.027270127087831497
2023-01-21 11:39:56.156893: step: 640/527, loss: 0.022963905707001686
2023-01-21 11:39:57.305791: step: 644/527, loss: 0.0001430511474609375
2023-01-21 11:39:58.434836: step: 648/527, loss: 0.3130761981010437
2023-01-21 11:39:59.579180: step: 652/527, loss: 0.004669570829719305
2023-01-21 11:40:00.723335: step: 656/527, loss: 0.0062469481490552425
2023-01-21 11:40:01.828178: step: 660/527, loss: 0.0010677337413653731
2023-01-21 11:40:02.986956: step: 664/527, loss: 0.014051247388124466
2023-01-21 11:40:04.117893: step: 668/527, loss: 3.1948089599609375e-05
2023-01-21 11:40:05.258269: step: 672/527, loss: 0.0029112815391272306
2023-01-21 11:40:06.359366: step: 676/527, loss: 0.002524852752685547
2023-01-21 11:40:07.485176: step: 680/527, loss: 0.0013834952842444181
2023-01-21 11:40:08.605705: step: 684/527, loss: 0.009224272333085537
2023-01-21 11:40:09.765487: step: 688/527, loss: 0.0004795074346475303
2023-01-21 11:40:10.885081: step: 692/527, loss: 0.04675617441534996
2023-01-21 11:40:12.011463: step: 696/527, loss: 0.018195820972323418
2023-01-21 11:40:13.195686: step: 700/527, loss: 0.03568840026855469
2023-01-21 11:40:14.315155: step: 704/527, loss: 0.007313728332519531
2023-01-21 11:40:15.422230: step: 708/527, loss: 0.018531514331698418
2023-01-21 11:40:16.537403: step: 712/527, loss: 0.0020042420364916325
2023-01-21 11:40:17.668219: step: 716/527, loss: 0.005865669343620539
2023-01-21 11:40:18.796213: step: 720/527, loss: 0.005946349818259478
2023-01-21 11:40:19.909900: step: 724/527, loss: 0.46851426362991333
2023-01-21 11:40:21.047777: step: 728/527, loss: 0.026128005236387253
2023-01-21 11:40:22.165481: step: 732/527, loss: 0.060364533215761185
2023-01-21 11:40:23.285012: step: 736/527, loss: 0.006414318457245827
2023-01-21 11:40:24.418012: step: 740/527, loss: 0.004105186555534601
2023-01-21 11:40:25.595183: step: 744/527, loss: 0.035622406750917435
2023-01-21 11:40:26.727582: step: 748/527, loss: 0.01057128980755806
2023-01-21 11:40:27.842558: step: 752/527, loss: 0.020526790991425514
2023-01-21 11:40:28.969486: step: 756/527, loss: 0.009049797430634499
2023-01-21 11:40:30.097541: step: 760/527, loss: 0.0011502265697345138
2023-01-21 11:40:31.206862: step: 764/527, loss: 0.03863391652703285
2023-01-21 11:40:32.332941: step: 768/527, loss: 0.0030647278763353825
2023-01-21 11:40:33.493924: step: 772/527, loss: 0.03179154545068741
2023-01-21 11:40:34.601724: step: 776/527, loss: 0.0024461746215820312
2023-01-21 11:40:35.701824: step: 780/527, loss: 0.0003772735653910786
2023-01-21 11:40:36.835775: step: 784/527, loss: 0.02175169065594673
2023-01-21 11:40:37.973363: step: 788/527, loss: 0.04328460991382599
2023-01-21 11:40:39.095446: step: 792/527, loss: 0.023350238800048828
2023-01-21 11:40:40.240336: step: 796/527, loss: 0.007850075140595436
2023-01-21 11:40:41.388444: step: 800/527, loss: 0.0018194199074059725
2023-01-21 11:40:42.506505: step: 804/527, loss: 0.0034483911003917456
2023-01-21 11:40:43.631398: step: 808/527, loss: 0.011926270090043545
2023-01-21 11:40:44.750806: step: 812/527, loss: 0.0006452560191974044
2023-01-21 11:40:45.923935: step: 816/527, loss: 0.016267968341708183
2023-01-21 11:40:47.064266: step: 820/527, loss: 0.018735790625214577
2023-01-21 11:40:48.174098: step: 824/527, loss: 0.020946310833096504
2023-01-21 11:40:49.326408: step: 828/527, loss: 0.05528726428747177
2023-01-21 11:40:50.474092: step: 832/527, loss: 0.01360416505485773
2023-01-21 11:40:51.577649: step: 836/527, loss: 0.0006442070007324219
2023-01-21 11:40:52.691745: step: 840/527, loss: 0.06638355553150177
2023-01-21 11:40:53.880934: step: 844/527, loss: 0.00012216568575240672
2023-01-21 11:40:55.029671: step: 848/527, loss: 0.027976226061582565
2023-01-21 11:40:56.183522: step: 852/527, loss: 0.0016086578834801912
2023-01-21 11:40:57.292003: step: 856/527, loss: 0.014592933468520641
2023-01-21 11:40:58.434224: step: 860/527, loss: 0.022863317281007767
2023-01-21 11:40:59.571491: step: 864/527, loss: 0.004327702801674604
2023-01-21 11:41:00.697492: step: 868/527, loss: 0.009133911691606045
2023-01-21 11:41:01.840829: step: 872/527, loss: 0.6425380706787109
2023-01-21 11:41:02.980849: step: 876/527, loss: 0.00011920928955078125
2023-01-21 11:41:04.102471: step: 880/527, loss: 0.08990326523780823
2023-01-21 11:41:05.251047: step: 884/527, loss: 0.00058155064471066
2023-01-21 11:41:06.378926: step: 888/527, loss: 0.0012882233131676912
2023-01-21 11:41:07.516550: step: 892/527, loss: 0.08394451439380646
2023-01-21 11:41:08.655129: step: 896/527, loss: 0.003975295927375555
2023-01-21 11:41:09.798558: step: 900/527, loss: 0.0011583161540329456
2023-01-21 11:41:10.939923: step: 904/527, loss: 0.003781271167099476
2023-01-21 11:41:12.051249: step: 908/527, loss: 0.002442479133605957
2023-01-21 11:41:13.189005: step: 912/527, loss: 0.03841390460729599
2023-01-21 11:41:14.337819: step: 916/527, loss: 0.04734230041503906
2023-01-21 11:41:15.469799: step: 920/527, loss: 0.006579494569450617
2023-01-21 11:41:16.582762: step: 924/527, loss: 0.0032499313820153475
2023-01-21 11:41:17.710953: step: 928/527, loss: 0.04964857175946236
2023-01-21 11:41:18.890510: step: 932/527, loss: 0.006509637925773859
2023-01-21 11:41:19.999614: step: 936/527, loss: 0.0014746190281584859
2023-01-21 11:41:21.131166: step: 940/527, loss: 0.019540976732969284
2023-01-21 11:41:22.246794: step: 944/527, loss: 0.019904911518096924
2023-01-21 11:41:23.357655: step: 948/527, loss: 0.018244266510009766
2023-01-21 11:41:24.468438: step: 952/527, loss: 0.0028182982932776213
2023-01-21 11:41:25.575078: step: 956/527, loss: 0.008939648047089577
2023-01-21 11:41:26.698151: step: 960/527, loss: 0.0893101766705513
2023-01-21 11:41:27.840779: step: 964/527, loss: 0.19413211941719055
2023-01-21 11:41:28.968383: step: 968/527, loss: 0.0056614927016198635
2023-01-21 11:41:30.092967: step: 972/527, loss: 0.0016385079361498356
2023-01-21 11:41:31.199365: step: 976/527, loss: 0.012868690304458141
2023-01-21 11:41:32.332718: step: 980/527, loss: 0.004455185029655695
2023-01-21 11:41:33.496719: step: 984/527, loss: 0.0041336058638989925
2023-01-21 11:41:34.605587: step: 988/527, loss: 0.001928138779476285
2023-01-21 11:41:35.710268: step: 992/527, loss: 0.05878520384430885
2023-01-21 11:41:36.876623: step: 996/527, loss: 0.011075782589614391
2023-01-21 11:41:37.961513: step: 1000/527, loss: 0.0019528151024132967
2023-01-21 11:41:39.050178: step: 1004/527, loss: 0.0164534579962492
2023-01-21 11:41:40.152575: step: 1008/527, loss: 0.0019310475327074528
2023-01-21 11:41:41.269613: step: 1012/527, loss: 0.019474124535918236
2023-01-21 11:41:42.396121: step: 1016/527, loss: 0.005499840248376131
2023-01-21 11:41:43.479660: step: 1020/527, loss: 0.009505176916718483
2023-01-21 11:41:44.617209: step: 1024/527, loss: 0.000293731689453125
2023-01-21 11:41:45.711733: step: 1028/527, loss: 0.029192542657256126
2023-01-21 11:41:46.884467: step: 1032/527, loss: 0.0027622224297374487
2023-01-21 11:41:48.046942: step: 1036/527, loss: 0.03670082241296768
2023-01-21 11:41:49.183537: step: 1040/527, loss: 0.00013669728650711477
2023-01-21 11:41:50.328742: step: 1044/527, loss: 0.002657604170963168
2023-01-21 11:41:51.499398: step: 1048/527, loss: 0.007655620574951172
2023-01-21 11:41:52.690639: step: 1052/527, loss: 0.009197616018354893
2023-01-21 11:41:53.813732: step: 1056/527, loss: 0.6586270928382874
2023-01-21 11:41:54.925639: step: 1060/527, loss: 0.0012735367054119706
2023-01-21 11:41:56.036725: step: 1064/527, loss: 0.035904884338378906
2023-01-21 11:41:57.173000: step: 1068/527, loss: 0.05943875387310982
2023-01-21 11:41:58.298405: step: 1072/527, loss: 0.07975101470947266
2023-01-21 11:41:59.439534: step: 1076/527, loss: 0.009791184216737747
2023-01-21 11:42:00.570505: step: 1080/527, loss: 0.0922277420759201
2023-01-21 11:42:01.655693: step: 1084/527, loss: 0.007860183715820312
2023-01-21 11:42:02.753136: step: 1088/527, loss: 0.018537236377596855
2023-01-21 11:42:03.849271: step: 1092/527, loss: 0.02582836151123047
2023-01-21 11:42:04.934705: step: 1096/527, loss: 0.0018868923652917147
2023-01-21 11:42:06.062116: step: 1100/527, loss: 0.008661841973662376
2023-01-21 11:42:07.186894: step: 1104/527, loss: 0.041791535913944244
2023-01-21 11:42:08.307909: step: 1108/527, loss: 0.025666045024991035
2023-01-21 11:42:09.424623: step: 1112/527, loss: 0.0021966935601085424
2023-01-21 11:42:10.539531: step: 1116/527, loss: 0.10806134343147278
2023-01-21 11:42:11.657229: step: 1120/527, loss: 0.0062334537506103516
2023-01-21 11:42:12.759425: step: 1124/527, loss: 0.0002914428769145161
2023-01-21 11:42:13.872836: step: 1128/527, loss: 0.024120425805449486
2023-01-21 11:42:15.000866: step: 1132/527, loss: 0.0005186080816201866
2023-01-21 11:42:16.141198: step: 1136/527, loss: 0.007250594906508923
2023-01-21 11:42:17.219762: step: 1140/527, loss: 0.0031056401785463095
2023-01-21 11:42:18.350399: step: 1144/527, loss: 0.01995544508099556
2023-01-21 11:42:19.475514: step: 1148/527, loss: 0.13898760080337524
2023-01-21 11:42:20.578742: step: 1152/527, loss: 0.029405975714325905
2023-01-21 11:42:21.686052: step: 1156/527, loss: 0.2844974398612976
2023-01-21 11:42:22.828992: step: 1160/527, loss: 0.0012763501144945621
2023-01-21 11:42:23.981511: step: 1164/527, loss: 0.019233321771025658
2023-01-21 11:42:25.120363: step: 1168/527, loss: 0.045714668929576874
2023-01-21 11:42:26.229198: step: 1172/527, loss: 0.0005618572467938066
2023-01-21 11:42:27.331970: step: 1176/527, loss: 0.021721458062529564
2023-01-21 11:42:28.483962: step: 1180/527, loss: 0.01964702643454075
2023-01-21 11:42:29.639201: step: 1184/527, loss: 0.004763889592140913
2023-01-21 11:42:30.765341: step: 1188/527, loss: 0.04556236043572426
2023-01-21 11:42:31.905153: step: 1192/527, loss: 0.007729625795036554
2023-01-21 11:42:33.028712: step: 1196/527, loss: 0.004032897762954235
2023-01-21 11:42:34.175404: step: 1200/527, loss: 0.044917963445186615
2023-01-21 11:42:35.262671: step: 1204/527, loss: 0.003471946809440851
2023-01-21 11:42:36.411337: step: 1208/527, loss: 0.010382939130067825
2023-01-21 11:42:37.539521: step: 1212/527, loss: 0.007255256175994873
2023-01-21 11:42:38.658892: step: 1216/527, loss: 0.018157387152314186
2023-01-21 11:42:39.756273: step: 1220/527, loss: 0.00022411346435546875
2023-01-21 11:42:40.856276: step: 1224/527, loss: 0.00933299120515585
2023-01-21 11:42:41.938489: step: 1228/527, loss: 0.26342129707336426
2023-01-21 11:42:43.026926: step: 1232/527, loss: 0.0018136025173589587
2023-01-21 11:42:44.159260: step: 1236/527, loss: 0.0006340026739053428
2023-01-21 11:42:45.270312: step: 1240/527, loss: 0.018594931811094284
2023-01-21 11:42:46.384354: step: 1244/527, loss: 0.0012929915683344007
2023-01-21 11:42:47.499780: step: 1248/527, loss: 0.025081824511289597
2023-01-21 11:42:48.649364: step: 1252/527, loss: 0.006076335906982422
2023-01-21 11:42:49.765031: step: 1256/527, loss: 0.0028447150252759457
2023-01-21 11:42:50.882720: step: 1260/527, loss: 0.04791984334588051
2023-01-21 11:42:52.009324: step: 1264/527, loss: 0.007804680150002241
2023-01-21 11:42:53.149584: step: 1268/527, loss: 0.10680066049098969
2023-01-21 11:42:54.266488: step: 1272/527, loss: 0.3834855258464813
2023-01-21 11:42:55.372559: step: 1276/527, loss: 0.010625124908983707
2023-01-21 11:42:56.480158: step: 1280/527, loss: 0.05281725153326988
2023-01-21 11:42:57.618051: step: 1284/527, loss: 0.007564067840576172
2023-01-21 11:42:58.737454: step: 1288/527, loss: 0.005911350250244141
2023-01-21 11:42:59.847098: step: 1292/527, loss: 0.0004930496215820312
2023-01-21 11:43:00.948533: step: 1296/527, loss: 0.0060134888626635075
2023-01-21 11:43:02.078682: step: 1300/527, loss: 0.0027416229713708162
2023-01-21 11:43:03.191824: step: 1304/527, loss: 0.013891208916902542
2023-01-21 11:43:04.283568: step: 1308/527, loss: 0.004548025317490101
2023-01-21 11:43:05.392010: step: 1312/527, loss: 0.03413820266723633
2023-01-21 11:43:06.496189: step: 1316/527, loss: 0.005631828214973211
2023-01-21 11:43:07.591613: step: 1320/527, loss: 0.0016065121162682772
2023-01-21 11:43:08.707429: step: 1324/527, loss: 0.003700304077938199
2023-01-21 11:43:09.861182: step: 1328/527, loss: 0.004315471742302179
2023-01-21 11:43:10.976991: step: 1332/527, loss: -3.814697265625e-06
2023-01-21 11:43:12.088361: step: 1336/527, loss: 0.043228913098573685
2023-01-21 11:43:13.203975: step: 1340/527, loss: 4.375918388366699
2023-01-21 11:43:14.309620: step: 1344/527, loss: 0.03010549396276474
2023-01-21 11:43:15.421977: step: 1348/527, loss: 0.004519939422607422
2023-01-21 11:43:16.507131: step: 1352/527, loss: 0.0009884834289550781
2023-01-21 11:43:17.615830: step: 1356/527, loss: 0.031243612989783287
2023-01-21 11:43:18.750341: step: 1360/527, loss: 3.14712519866589e-06
2023-01-21 11:43:19.847613: step: 1364/527, loss: 0.00040941237239167094
2023-01-21 11:43:20.971440: step: 1368/527, loss: 0.006172371096909046
2023-01-21 11:43:22.101774: step: 1372/527, loss: 0.0034801485016942024
2023-01-21 11:43:23.224560: step: 1376/527, loss: 0.07715263217687607
2023-01-21 11:43:24.354790: step: 1380/527, loss: 0.01746826246380806
2023-01-21 11:43:25.448548: step: 1384/527, loss: 0.09030818939208984
2023-01-21 11:43:26.546236: step: 1388/527, loss: 0.007637977600097656
2023-01-21 11:43:27.674547: step: 1392/527, loss: 0.007068062201142311
2023-01-21 11:43:28.784470: step: 1396/527, loss: 0.03342771530151367
2023-01-21 11:43:29.882823: step: 1400/527, loss: 0.014216804876923561
2023-01-21 11:43:30.969584: step: 1404/527, loss: 0.0035140991676598787
2023-01-21 11:43:32.106265: step: 1408/527, loss: 0.007841682061553001
2023-01-21 11:43:33.203910: step: 1412/527, loss: 0.12781424820423126
2023-01-21 11:43:34.317934: step: 1416/527, loss: 0.026134967803955078
2023-01-21 11:43:35.431553: step: 1420/527, loss: 0.016472624614834785
2023-01-21 11:43:36.542645: step: 1424/527, loss: 0.010421419516205788
2023-01-21 11:43:37.668352: step: 1428/527, loss: 0.018986130133271217
2023-01-21 11:43:38.778397: step: 1432/527, loss: 0.005397701170295477
2023-01-21 11:43:39.909070: step: 1436/527, loss: 0.0007076740730553865
2023-01-21 11:43:41.009350: step: 1440/527, loss: 0.011234093457460403
2023-01-21 11:43:42.102186: step: 1444/527, loss: 0.00044460297795012593
2023-01-21 11:43:43.244397: step: 1448/527, loss: 0.06940049678087234
2023-01-21 11:43:44.340525: step: 1452/527, loss: 0.02171926572918892
2023-01-21 11:43:45.465700: step: 1456/527, loss: 0.02141146920621395
2023-01-21 11:43:46.573700: step: 1460/527, loss: 0.004588508978486061
2023-01-21 11:43:47.681752: step: 1464/527, loss: 0.01808171346783638
2023-01-21 11:43:48.762429: step: 1468/527, loss: 0.020713090896606445
2023-01-21 11:43:49.851299: step: 1472/527, loss: 0.014095068909227848
2023-01-21 11:43:50.974595: step: 1476/527, loss: 0.0026751516852527857
2023-01-21 11:43:52.088058: step: 1480/527, loss: 4.301071021473035e-05
2023-01-21 11:43:53.219897: step: 1484/527, loss: 0.022134114056825638
2023-01-21 11:43:54.341961: step: 1488/527, loss: 0.08616063743829727
2023-01-21 11:43:55.483414: step: 1492/527, loss: 0.03980650752782822
2023-01-21 11:43:56.597762: step: 1496/527, loss: 0.006954193580895662
2023-01-21 11:43:57.707248: step: 1500/527, loss: 0.03614349663257599
2023-01-21 11:43:58.821447: step: 1504/527, loss: 0.0035955430939793587
2023-01-21 11:43:59.951109: step: 1508/527, loss: 0.0005060196272097528
2023-01-21 11:44:01.092205: step: 1512/527, loss: 0.009517860598862171
2023-01-21 11:44:02.209366: step: 1516/527, loss: 0.004303359892219305
2023-01-21 11:44:03.336936: step: 1520/527, loss: 0.05650768429040909
2023-01-21 11:44:04.444520: step: 1524/527, loss: 0.06997299194335938
2023-01-21 11:44:05.573449: step: 1528/527, loss: 0.00019397735013626516
2023-01-21 11:44:06.701629: step: 1532/527, loss: 0.0017027854919433594
2023-01-21 11:44:07.822401: step: 1536/527, loss: 0.011641979217529297
2023-01-21 11:44:08.926207: step: 1540/527, loss: 0.055008694529533386
2023-01-21 11:44:10.054600: step: 1544/527, loss: 0.024526499211788177
2023-01-21 11:44:11.156063: step: 1548/527, loss: 0.001004982041195035
2023-01-21 11:44:12.285345: step: 1552/527, loss: 0.010524654760956764
2023-01-21 11:44:13.389051: step: 1556/527, loss: 0.0004533767933025956
2023-01-21 11:44:14.499953: step: 1560/527, loss: 0.018765592947602272
2023-01-21 11:44:15.608435: step: 1564/527, loss: 0.004154527094215155
2023-01-21 11:44:16.720678: step: 1568/527, loss: 0.03361377865076065
2023-01-21 11:44:17.820775: step: 1572/527, loss: 0.01734332926571369
2023-01-21 11:44:18.918753: step: 1576/527, loss: 0.0003952026308979839
2023-01-21 11:44:20.029565: step: 1580/527, loss: 0.008944297209382057
2023-01-21 11:44:21.200946: step: 1584/527, loss: 0.04268951341509819
2023-01-21 11:44:22.320155: step: 1588/527, loss: 0.008399104699492455
2023-01-21 11:44:23.432806: step: 1592/527, loss: 0.008795356377959251
2023-01-21 11:44:24.538696: step: 1596/527, loss: 0.012418795377016068
2023-01-21 11:44:25.677808: step: 1600/527, loss: 0.0019072532886639237
2023-01-21 11:44:26.784608: step: 1604/527, loss: 8.267239172710106e-05
2023-01-21 11:44:27.903972: step: 1608/527, loss: 0.0005527496105059981
2023-01-21 11:44:29.028511: step: 1612/527, loss: 0.010962486267089844
2023-01-21 11:44:30.140277: step: 1616/527, loss: 0.0006602287758141756
2023-01-21 11:44:31.248657: step: 1620/527, loss: 0.031920626759529114
2023-01-21 11:44:32.374568: step: 1624/527, loss: 0.011629152111709118
2023-01-21 11:44:33.505560: step: 1628/527, loss: 0.0025787353515625
2023-01-21 11:44:34.600822: step: 1632/527, loss: 0.01799631118774414
2023-01-21 11:44:35.725535: step: 1636/527, loss: 0.007521533872932196
2023-01-21 11:44:36.835900: step: 1640/527, loss: 0.12186841666698456
2023-01-21 11:44:37.958633: step: 1644/527, loss: 0.011512279510498047
2023-01-21 11:44:39.082448: step: 1648/527, loss: 0.12494973838329315
2023-01-21 11:44:40.203588: step: 1652/527, loss: 0.011986018158495426
2023-01-21 11:44:41.327518: step: 1656/527, loss: 0.005015564151108265
2023-01-21 11:44:42.462444: step: 1660/527, loss: 0.007798528298735619
2023-01-21 11:44:43.591106: step: 1664/527, loss: 0.002794170519337058
2023-01-21 11:44:44.695054: step: 1668/527, loss: 0.02684185467660427
2023-01-21 11:44:45.802022: step: 1672/527, loss: 0.023714445531368256
2023-01-21 11:44:46.942782: step: 1676/527, loss: 0.013370160013437271
2023-01-21 11:44:48.053672: step: 1680/527, loss: 0.023885631933808327
2023-01-21 11:44:49.189141: step: 1684/527, loss: 0.00039606093196198344
2023-01-21 11:44:50.300364: step: 1688/527, loss: 0.0024572850670665503
2023-01-21 11:44:51.413216: step: 1692/527, loss: 0.003660488175228238
2023-01-21 11:44:52.523490: step: 1696/527, loss: 0.0020572184585034847
2023-01-21 11:44:53.611449: step: 1700/527, loss: 0.0004181861877441406
2023-01-21 11:44:54.718337: step: 1704/527, loss: 0.0007913351291790605
2023-01-21 11:44:55.855668: step: 1708/527, loss: 0.0002601623418740928
2023-01-21 11:44:57.029328: step: 1712/527, loss: 0.017992781475186348
2023-01-21 11:44:58.153858: step: 1716/527, loss: 0.009121155366301537
2023-01-21 11:44:59.254970: step: 1720/527, loss: 0.027771567925810814
2023-01-21 11:45:00.358967: step: 1724/527, loss: 0.005588436499238014
2023-01-21 11:45:01.474701: step: 1728/527, loss: 0.003928852267563343
2023-01-21 11:45:02.571632: step: 1732/527, loss: 0.005960989277809858
2023-01-21 11:45:03.701877: step: 1736/527, loss: 0.010779953561723232
2023-01-21 11:45:04.801372: step: 1740/527, loss: 0.0044099269434809685
2023-01-21 11:45:05.916347: step: 1744/527, loss: 0.0006672203307971358
2023-01-21 11:45:07.011710: step: 1748/527, loss: 0.0036901473067700863
2023-01-21 11:45:08.131522: step: 1752/527, loss: 0.027920342981815338
2023-01-21 11:45:09.251324: step: 1756/527, loss: 0.002324771834537387
2023-01-21 11:45:10.349731: step: 1760/527, loss: 0.10286064445972443
2023-01-21 11:45:11.454411: step: 1764/527, loss: 0.03215789794921875
2023-01-21 11:45:12.618380: step: 1768/527, loss: 0.015442848205566406
2023-01-21 11:45:13.724347: step: 1772/527, loss: 0.00025053025456145406
2023-01-21 11:45:14.818782: step: 1776/527, loss: 0.00043425560579635203
2023-01-21 11:45:15.931396: step: 1780/527, loss: 0.00130462646484375
2023-01-21 11:45:17.032646: step: 1784/527, loss: 0.0009635924943722785
2023-01-21 11:45:18.177954: step: 1788/527, loss: 0.01930699311196804
2023-01-21 11:45:19.338593: step: 1792/527, loss: 0.0031836749985814095
2023-01-21 11:45:20.487466: step: 1796/527, loss: 0.00045108795166015625
2023-01-21 11:45:21.625106: step: 1800/527, loss: 0.0006071091047488153
2023-01-21 11:45:22.712433: step: 1804/527, loss: 0.009622382931411266
2023-01-21 11:45:23.858527: step: 1808/527, loss: 0.10963650047779083
2023-01-21 11:45:24.988678: step: 1812/527, loss: 0.0015749931335449219
2023-01-21 11:45:26.098274: step: 1816/527, loss: 0.0002948761102743447
2023-01-21 11:45:27.223641: step: 1820/527, loss: 0.01230402011424303
2023-01-21 11:45:28.309147: step: 1824/527, loss: 0.022346878424286842
2023-01-21 11:45:29.419856: step: 1828/527, loss: 0.0017177582485601306
2023-01-21 11:45:30.529728: step: 1832/527, loss: 0.09004173427820206
2023-01-21 11:45:31.647118: step: 1836/527, loss: 0.043418265879154205
2023-01-21 11:45:32.779150: step: 1840/527, loss: 0.0003833770751953125
2023-01-21 11:45:33.893970: step: 1844/527, loss: 0.0007290840148925781
2023-01-21 11:45:35.020861: step: 1848/527, loss: 0.0024450302589684725
2023-01-21 11:45:36.103315: step: 1852/527, loss: 0.024298476055264473
2023-01-21 11:45:37.212525: step: 1856/527, loss: 0.007344054989516735
2023-01-21 11:45:38.332345: step: 1860/527, loss: 0.0016605377895757556
2023-01-21 11:45:39.458097: step: 1864/527, loss: 0.007188892923295498
2023-01-21 11:45:40.577221: step: 1868/527, loss: 0.040312767028808594
2023-01-21 11:45:41.706699: step: 1872/527, loss: 0.003297233721241355
2023-01-21 11:45:42.851598: step: 1876/527, loss: 0.018662549555301666
2023-01-21 11:45:44.007890: step: 1880/527, loss: 0.0031269071623682976
2023-01-21 11:45:45.118319: step: 1884/527, loss: 0.001056766603142023
2023-01-21 11:45:46.214333: step: 1888/527, loss: 0.09863471984863281
2023-01-21 11:45:47.335858: step: 1892/527, loss: 0.0003509760135784745
2023-01-21 11:45:48.470361: step: 1896/527, loss: 0.0010200501419603825
2023-01-21 11:45:49.584589: step: 1900/527, loss: 0.0005661010509356856
2023-01-21 11:45:50.688245: step: 1904/527, loss: 0.0011453628540039062
2023-01-21 11:45:51.768188: step: 1908/527, loss: 0.00019297302060294896
2023-01-21 11:45:52.863697: step: 1912/527, loss: 0.03646891564130783
2023-01-21 11:45:53.956982: step: 1916/527, loss: 0.019242573529481888
2023-01-21 11:45:55.077015: step: 1920/527, loss: 0.0015337944496423006
2023-01-21 11:45:56.197543: step: 1924/527, loss: 0.00052642822265625
2023-01-21 11:45:57.325857: step: 1928/527, loss: 0.0002960205019917339
2023-01-21 11:45:58.455224: step: 1932/527, loss: 0.00031070708064362407
2023-01-21 11:45:59.563333: step: 1936/527, loss: 0.00678596505895257
2023-01-21 11:46:00.676685: step: 1940/527, loss: 0.08375845104455948
2023-01-21 11:46:01.793474: step: 1944/527, loss: 0.014844894409179688
2023-01-21 11:46:02.911870: step: 1948/527, loss: 0.0428524985909462
2023-01-21 11:46:04.038816: step: 1952/527, loss: 0.050026893615722656
2023-01-21 11:46:05.150021: step: 1956/527, loss: 1.4208256006240845
2023-01-21 11:46:06.260381: step: 1960/527, loss: 0.005116653628647327
2023-01-21 11:46:07.382997: step: 1964/527, loss: 0.019698143005371094
2023-01-21 11:46:08.492282: step: 1968/527, loss: 0.0010331631638109684
2023-01-21 11:46:09.598964: step: 1972/527, loss: 0.02993779256939888
2023-01-21 11:46:10.709035: step: 1976/527, loss: 0.000225067138671875
2023-01-21 11:46:11.823482: step: 1980/527, loss: 0.08067512512207031
2023-01-21 11:46:12.926140: step: 1984/527, loss: 0.04687700420618057
2023-01-21 11:46:14.044425: step: 1988/527, loss: 0.018857955932617188
2023-01-21 11:46:15.169510: step: 1992/527, loss: 0.01280226744711399
2023-01-21 11:46:16.273774: step: 1996/527, loss: 0.02402694895863533
2023-01-21 11:46:17.404748: step: 2000/527, loss: 0.010155045427381992
2023-01-21 11:46:18.500334: step: 2004/527, loss: 0.08439507335424423
2023-01-21 11:46:19.616800: step: 2008/527, loss: 0.0087890625
2023-01-21 11:46:20.751789: step: 2012/527, loss: 0.003377723740413785
2023-01-21 11:46:21.882259: step: 2016/527, loss: 0.07241944968700409
2023-01-21 11:46:23.019158: step: 2020/527, loss: 0.042044639587402344
2023-01-21 11:46:24.144847: step: 2024/527, loss: 0.029285239055752754
2023-01-21 11:46:25.218269: step: 2028/527, loss: 0.058226823806762695
2023-01-21 11:46:26.347336: step: 2032/527, loss: 0.02170858345925808
2023-01-21 11:46:27.477267: step: 2036/527, loss: 0.0789773017168045
2023-01-21 11:46:28.570935: step: 2040/527, loss: 0.0006162643549032509
2023-01-21 11:46:29.668069: step: 2044/527, loss: 0.0015653610462322831
2023-01-21 11:46:30.787167: step: 2048/527, loss: 0.05936603620648384
2023-01-21 11:46:31.912517: step: 2052/527, loss: 0.6620147824287415
2023-01-21 11:46:33.073677: step: 2056/527, loss: 3.204345557605848e-05
2023-01-21 11:46:34.175602: step: 2060/527, loss: 0.0005727768293581903
2023-01-21 11:46:35.279620: step: 2064/527, loss: 0.0011310577392578125
2023-01-21 11:46:36.395943: step: 2068/527, loss: 0.0053021907806396484
2023-01-21 11:46:37.486522: step: 2072/527, loss: 0.006241464521735907
2023-01-21 11:46:38.618658: step: 2076/527, loss: 0.003083133604377508
2023-01-21 11:46:39.745800: step: 2080/527, loss: 0.0012887001503258944
2023-01-21 11:46:40.864036: step: 2084/527, loss: 9.527205838821828e-05
2023-01-21 11:46:41.994412: step: 2088/527, loss: 0.016990184783935547
2023-01-21 11:46:43.068470: step: 2092/527, loss: 0.01983466185629368
2023-01-21 11:46:44.177207: step: 2096/527, loss: 0.0018999100429937243
2023-01-21 11:46:45.259906: step: 2100/527, loss: 0.0005417823558673263
2023-01-21 11:46:46.391038: step: 2104/527, loss: 0.00032806396484375
2023-01-21 11:46:47.509742: step: 2108/527, loss: 0.021654987707734108
==================================================
Loss: 0.038
--------------------
Dev: {'event': {'p': 0.5776458951533135, 'r': 0.7776298268974701, 'f1': 0.662883087400681}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}
Test: {'event': {'p': 0.6356088560885609, 'r': 0.7874285714285715, 'f1': 0.703420112302195}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}
Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}
Korean: {'event': {'p': 0.5892857142857143, 'r': 0.5238095238095238, 'f1': 0.5546218487394958}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}
Russian: {'event': {'p': 0.5135135135135135, 'r': 0.5277777777777778, 'f1': 0.5205479452054794}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:47:29.291185: step: 4/527, loss: 0.0007443904760293663
2023-01-21 11:47:30.387069: step: 8/527, loss: 0.008268166333436966
2023-01-21 11:47:31.512872: step: 12/527, loss: 0.044470977038145065
2023-01-21 11:47:32.625340: step: 16/527, loss: 0.0016821861499920487
2023-01-21 11:47:33.693370: step: 20/527, loss: 0.0004517078632488847
2023-01-21 11:47:34.788683: step: 24/527, loss: 0.006455803290009499
2023-01-21 11:47:35.898598: step: 28/527, loss: 0.09223899990320206
2023-01-21 11:47:37.035048: step: 32/527, loss: 0.01903076097369194
2023-01-21 11:47:38.151944: step: 36/527, loss: 0.004425621125847101
2023-01-21 11:47:39.281205: step: 40/527, loss: 0.003958797547966242
2023-01-21 11:47:40.392166: step: 44/527, loss: 0.04101219400763512
2023-01-21 11:47:41.491810: step: 48/527, loss: 0.0017236709827557206
2023-01-21 11:47:42.609380: step: 52/527, loss: 0.02956390380859375
2023-01-21 11:47:43.722819: step: 56/527, loss: 0.01581869088113308
2023-01-21 11:47:44.835065: step: 60/527, loss: 0.07888869941234589
2023-01-21 11:47:45.921804: step: 64/527, loss: 0.0004093647003173828
2023-01-21 11:47:47.040127: step: 68/527, loss: 0.007456970401108265
2023-01-21 11:47:48.196758: step: 72/527, loss: 0.07659869641065598
2023-01-21 11:47:49.302833: step: 76/527, loss: 0.001312255859375
2023-01-21 11:47:50.421489: step: 80/527, loss: 0.0057531362399458885
2023-01-21 11:47:51.562517: step: 84/527, loss: 0.02678375318646431
2023-01-21 11:47:52.671311: step: 88/527, loss: 0.008371162228286266
2023-01-21 11:47:53.791664: step: 92/527, loss: 0.015362453646957874
2023-01-21 11:47:54.902573: step: 96/527, loss: 7.62939453125e-05
2023-01-21 11:47:56.022582: step: 100/527, loss: 0.06545820832252502
2023-01-21 11:47:57.109316: step: 104/527, loss: 0.007288122083991766
2023-01-21 11:47:58.233127: step: 108/527, loss: 0.03342590481042862
2023-01-21 11:47:59.366037: step: 112/527, loss: 0.0038074494805186987
2023-01-21 11:48:00.463129: step: 116/527, loss: 0.0034792902879416943
2023-01-21 11:48:01.585662: step: 120/527, loss: 0.1414480209350586
2023-01-21 11:48:02.718054: step: 124/527, loss: 0.009324932470917702
2023-01-21 11:48:03.837131: step: 128/527, loss: 0.024091817438602448
2023-01-21 11:48:04.965322: step: 132/527, loss: 0.0003219604550395161
2023-01-21 11:48:06.090919: step: 136/527, loss: 0.0018285751575604081
2023-01-21 11:48:07.187773: step: 140/527, loss: 0.028315259143710136
2023-01-21 11:48:08.328776: step: 144/527, loss: 0.007821083068847656
2023-01-21 11:48:09.458327: step: 148/527, loss: 0.002789115998893976
2023-01-21 11:48:10.572335: step: 152/527, loss: 0.0004796028370037675
2023-01-21 11:48:11.684999: step: 156/527, loss: 0.0440179817378521
2023-01-21 11:48:12.812952: step: 160/527, loss: 0.003673553466796875
2023-01-21 11:48:13.949865: step: 164/527, loss: 0.019334029406309128
2023-01-21 11:48:15.060359: step: 168/527, loss: 0.00721054058521986
2023-01-21 11:48:16.122903: step: 172/527, loss: 0.0033194541465491056
2023-01-21 11:48:17.239358: step: 176/527, loss: 2.37941749219317e-05
2023-01-21 11:48:18.326203: step: 180/527, loss: 0.0004592418554238975
2023-01-21 11:48:19.418966: step: 184/527, loss: 0.004555463790893555
2023-01-21 11:48:20.523130: step: 188/527, loss: 0.02675657346844673
2023-01-21 11:48:21.630835: step: 192/527, loss: 0.052603721618652344
2023-01-21 11:48:22.753927: step: 196/527, loss: 0.02210082858800888
2023-01-21 11:48:23.876669: step: 200/527, loss: 0.05299912020564079
2023-01-21 11:48:25.002710: step: 204/527, loss: 0.04600124433636665
2023-01-21 11:48:26.112356: step: 208/527, loss: 0.00979003868997097
2023-01-21 11:48:27.222482: step: 212/527, loss: 0.0033321380615234375
2023-01-21 11:48:28.333280: step: 216/527, loss: 0.00018072128295898438
2023-01-21 11:48:29.482658: step: 220/527, loss: 0.0023645402397960424
2023-01-21 11:48:30.566263: step: 224/527, loss: 0.0015435219975188375
2023-01-21 11:48:31.694606: step: 228/527, loss: 0.007020092569291592
2023-01-21 11:48:32.816476: step: 232/527, loss: 3.929138256353326e-05
2023-01-21 11:48:33.903851: step: 236/527, loss: 0.015948820859193802
2023-01-21 11:48:35.054595: step: 240/527, loss: 0.034228041768074036
2023-01-21 11:48:36.141672: step: 244/527, loss: 0.007886076346039772
2023-01-21 11:48:37.251762: step: 248/527, loss: 0.0001703262241790071
2023-01-21 11:48:38.359603: step: 252/527, loss: 0.01219644583761692
2023-01-21 11:48:39.447843: step: 256/527, loss: 0.0360623374581337
2023-01-21 11:48:40.573565: step: 260/527, loss: 0.00242443080060184
2023-01-21 11:48:41.686943: step: 264/527, loss: 0.005789470858871937
2023-01-21 11:48:42.749974: step: 268/527, loss: 0.05938892811536789
2023-01-21 11:48:43.853459: step: 272/527, loss: 0.000931215297896415
2023-01-21 11:48:44.944635: step: 276/527, loss: 0.0023778914473950863
2023-01-21 11:48:46.051761: step: 280/527, loss: 0.0001423835929017514
2023-01-21 11:48:47.172444: step: 284/527, loss: 0.0033958435524255037
2023-01-21 11:48:48.298512: step: 288/527, loss: 0.1252867877483368
2023-01-21 11:48:49.421824: step: 292/527, loss: 7.848739915061742e-05
2023-01-21 11:48:50.535036: step: 296/527, loss: 2.2266287803649902
2023-01-21 11:48:51.629472: step: 300/527, loss: 3.7956240703351796e-05
2023-01-21 11:48:52.755222: step: 304/527, loss: 0.0019287110771983862
2023-01-21 11:48:53.848785: step: 308/527, loss: 0.00761337298899889
2023-01-21 11:48:54.952257: step: 312/527, loss: 0.040021561086177826
2023-01-21 11:48:56.078611: step: 316/527, loss: 0.22668485343456268
2023-01-21 11:48:57.199684: step: 320/527, loss: 0.0023276328574866056
2023-01-21 11:48:58.324950: step: 324/527, loss: 0.0016707897884771228
2023-01-21 11:48:59.462042: step: 328/527, loss: 0.0007989883306436241
2023-01-21 11:49:00.582144: step: 332/527, loss: 0.016665171831846237
2023-01-21 11:49:01.733589: step: 336/527, loss: 0.022960377857089043
2023-01-21 11:49:02.840963: step: 340/527, loss: 0.0011849404545500875
2023-01-21 11:49:03.972592: step: 344/527, loss: 0.00054845807608217
2023-01-21 11:49:05.109521: step: 348/527, loss: 0.016692733392119408
2023-01-21 11:49:06.214526: step: 352/527, loss: 0.026497984305024147
2023-01-21 11:49:07.313135: step: 356/527, loss: 0.0018697739578783512
2023-01-21 11:49:08.442324: step: 360/527, loss: 0.02386179007589817
2023-01-21 11:49:09.538296: step: 364/527, loss: 0.011486626230180264
2023-01-21 11:49:10.658378: step: 368/527, loss: 0.02828803099691868
2023-01-21 11:49:11.780165: step: 372/527, loss: 0.0007494926685467362
2023-01-21 11:49:12.891167: step: 376/527, loss: 0.009190560318529606
2023-01-21 11:49:14.015752: step: 380/527, loss: 0.02438640594482422
2023-01-21 11:49:15.168522: step: 384/527, loss: 0.0008827209239825606
2023-01-21 11:49:16.279907: step: 388/527, loss: 0.0014810562133789062
2023-01-21 11:49:17.391390: step: 392/527, loss: 0.0005075454828329384
2023-01-21 11:49:18.521270: step: 396/527, loss: 0.002871942473575473
2023-01-21 11:49:19.650923: step: 400/527, loss: 0.003062725067138672
2023-01-21 11:49:20.797985: step: 404/527, loss: 0.013873100280761719
2023-01-21 11:49:21.885639: step: 408/527, loss: 0.00836792029440403
2023-01-21 11:49:22.986510: step: 412/527, loss: 0.0010294914245605469
2023-01-21 11:49:24.081012: step: 416/527, loss: 0.015017509460449219
2023-01-21 11:49:25.194864: step: 420/527, loss: 0.03229961544275284
2023-01-21 11:49:26.317986: step: 424/527, loss: 0.0203291904181242
2023-01-21 11:49:27.408585: step: 428/527, loss: 0.013685441575944424
2023-01-21 11:49:28.533062: step: 432/527, loss: 0.012811565771698952
2023-01-21 11:49:29.652075: step: 436/527, loss: 0.02293987385928631
2023-01-21 11:49:30.756313: step: 440/527, loss: 0.0366411916911602
2023-01-21 11:49:31.836804: step: 444/527, loss: 0.0037747861351817846
2023-01-21 11:49:32.957862: step: 448/527, loss: 0.03017454221844673
2023-01-21 11:49:34.065138: step: 452/527, loss: 0.023659897968173027
2023-01-21 11:49:35.158791: step: 456/527, loss: 0.0641777515411377
2023-01-21 11:49:36.268769: step: 460/527, loss: 0.0356692336499691
2023-01-21 11:49:37.387711: step: 464/527, loss: 0.007748937699943781
2023-01-21 11:49:38.497710: step: 468/527, loss: 0.00025978090707212687
2023-01-21 11:49:39.614538: step: 472/527, loss: 0.015533447265625
2023-01-21 11:49:40.771314: step: 476/527, loss: 0.0330720916390419
2023-01-21 11:49:41.881067: step: 480/527, loss: 0.011567306704819202
2023-01-21 11:49:43.036831: step: 484/527, loss: 0.014894485473632812
2023-01-21 11:49:44.168322: step: 488/527, loss: 0.008662509731948376
2023-01-21 11:49:45.287053: step: 492/527, loss: 0.00592122133821249
2023-01-21 11:49:46.397121: step: 496/527, loss: 0.6933574676513672
2023-01-21 11:49:47.521678: step: 500/527, loss: 0.029668331146240234
2023-01-21 11:49:48.632230: step: 504/527, loss: 0.036104775965213776
2023-01-21 11:49:49.756637: step: 508/527, loss: 0.008000469766557217
2023-01-21 11:49:50.846005: step: 512/527, loss: 0.0020822526421397924
2023-01-21 11:49:51.983875: step: 516/527, loss: 0.00012855530076194555
2023-01-21 11:49:53.093699: step: 520/527, loss: 0.11603926867246628
2023-01-21 11:49:54.224386: step: 524/527, loss: 0.019771242514252663
2023-01-21 11:49:55.391194: step: 528/527, loss: 0.008343124762177467
2023-01-21 11:49:56.505728: step: 532/527, loss: 0.004056167788803577
2023-01-21 11:49:57.607859: step: 536/527, loss: 0.005884981248527765
2023-01-21 11:49:58.739258: step: 540/527, loss: 0.002049064729362726
2023-01-21 11:49:59.833976: step: 544/527, loss: 0.0664072260260582
2023-01-21 11:50:00.940273: step: 548/527, loss: 0.010943364351987839
2023-01-21 11:50:02.045864: step: 552/527, loss: 0.0001542091486044228
2023-01-21 11:50:03.163959: step: 556/527, loss: 0.0034339905250817537
2023-01-21 11:50:04.244301: step: 560/527, loss: 0.005725670140236616
2023-01-21 11:50:05.362079: step: 564/527, loss: 0.0372682586312294
2023-01-21 11:50:06.467640: step: 568/527, loss: 9.71794142969884e-05
2023-01-21 11:50:07.581858: step: 572/527, loss: 0.24436703324317932
2023-01-21 11:50:08.705351: step: 576/527, loss: 0.00011959076800849289
2023-01-21 11:50:09.808637: step: 580/527, loss: 0.001471710274927318
2023-01-21 11:50:10.900932: step: 584/527, loss: 0.09455184638500214
2023-01-21 11:50:12.006322: step: 588/527, loss: 0.0020543099381029606
2023-01-21 11:50:13.106246: step: 592/527, loss: 0.015607167035341263
2023-01-21 11:50:14.188591: step: 596/527, loss: 0.00012907982454635203
2023-01-21 11:50:15.319647: step: 600/527, loss: 0.15989524126052856
2023-01-21 11:50:16.447723: step: 604/527, loss: 0.00011978149996139109
2023-01-21 11:50:17.561238: step: 608/527, loss: 0.003145408583804965
2023-01-21 11:50:18.667402: step: 612/527, loss: 0.18141384422779083
2023-01-21 11:50:19.799807: step: 616/527, loss: 0.04190473631024361
2023-01-21 11:50:20.950514: step: 620/527, loss: 0.003602695418521762
2023-01-21 11:50:22.074608: step: 624/527, loss: 0.06306210160255432
2023-01-21 11:50:23.194615: step: 628/527, loss: 0.03174161911010742
2023-01-21 11:50:24.280163: step: 632/527, loss: 0.0453639030456543
2023-01-21 11:50:25.372260: step: 636/527, loss: 0.015012543648481369
2023-01-21 11:50:26.486760: step: 640/527, loss: 0.07665643841028214
2023-01-21 11:50:27.581887: step: 644/527, loss: 0.008855295367538929
2023-01-21 11:50:28.676767: step: 648/527, loss: 0.015130424872040749
2023-01-21 11:50:29.786060: step: 652/527, loss: 0.004867649171501398
2023-01-21 11:50:30.890959: step: 656/527, loss: 0.036391355097293854
2023-01-21 11:50:32.027545: step: 660/527, loss: 0.026813887059688568
2023-01-21 11:50:33.128297: step: 664/527, loss: 0.00071887974627316
2023-01-21 11:50:34.244649: step: 668/527, loss: 0.023047685623168945
2023-01-21 11:50:35.383186: step: 672/527, loss: 0.006439399905502796
2023-01-21 11:50:36.513876: step: 676/527, loss: 0.04015917703509331
2023-01-21 11:50:37.611427: step: 680/527, loss: 0.012956619262695312
2023-01-21 11:50:38.713253: step: 684/527, loss: 0.004407978616654873
2023-01-21 11:50:39.815078: step: 688/527, loss: 0.06499719619750977
2023-01-21 11:50:40.952973: step: 692/527, loss: 0.0036669732071459293
2023-01-21 11:50:42.062123: step: 696/527, loss: 0.004702758975327015
2023-01-21 11:50:43.199913: step: 700/527, loss: 0.0019788744393736124
2023-01-21 11:50:44.344867: step: 704/527, loss: 0.0012704849941655993
2023-01-21 11:50:45.464665: step: 708/527, loss: 0.013256454840302467
2023-01-21 11:50:46.576273: step: 712/527, loss: 0.0035755636636167765
2023-01-21 11:50:47.689757: step: 716/527, loss: 0.0015518188010901213
2023-01-21 11:50:48.798201: step: 720/527, loss: 0.01841564290225506
2023-01-21 11:50:49.906986: step: 724/527, loss: 1.9073513612966053e-07
2023-01-21 11:50:51.037527: step: 728/527, loss: 0.002987384796142578
2023-01-21 11:50:52.149604: step: 732/527, loss: 0.0014947891468182206
2023-01-21 11:50:53.250167: step: 736/527, loss: 0.029780007898807526
2023-01-21 11:50:54.372242: step: 740/527, loss: 0.02634258382022381
2023-01-21 11:50:55.496451: step: 744/527, loss: 0.008105278015136719
2023-01-21 11:50:56.632293: step: 748/527, loss: 0.007285094819962978
2023-01-21 11:50:57.771038: step: 752/527, loss: 0.03876161575317383
2023-01-21 11:50:58.912515: step: 756/527, loss: 0.016056442633271217
2023-01-21 11:51:00.022481: step: 760/527, loss: 0.06254110485315323
2023-01-21 11:51:01.145946: step: 764/527, loss: 0.0197372455149889
2023-01-21 11:51:02.248860: step: 768/527, loss: 0.02198362536728382
2023-01-21 11:51:03.409617: step: 772/527, loss: 0.00655817985534668
2023-01-21 11:51:04.566277: step: 776/527, loss: 0.0031541825737804174
2023-01-21 11:51:05.679827: step: 780/527, loss: 0.02676544152200222
2023-01-21 11:51:06.787986: step: 784/527, loss: 0.07407913357019424
2023-01-21 11:51:07.923050: step: 788/527, loss: 0.006284618284553289
2023-01-21 11:51:09.021666: step: 792/527, loss: 0.02416963502764702
2023-01-21 11:51:10.167334: step: 796/527, loss: 0.004728412721306086
2023-01-21 11:51:11.270732: step: 800/527, loss: 0.0034944533836096525
2023-01-21 11:51:12.383592: step: 804/527, loss: 0.005403477232903242
2023-01-21 11:51:13.486872: step: 808/527, loss: 0.007008838467299938
2023-01-21 11:51:14.553342: step: 812/527, loss: 0.0005703926435671747
2023-01-21 11:51:15.629342: step: 816/527, loss: 0.00057306292001158
2023-01-21 11:51:16.758777: step: 820/527, loss: 5.3215029765851796e-05
2023-01-21 11:51:17.874153: step: 824/527, loss: 4.57763671875e-05
2023-01-21 11:51:18.964805: step: 828/527, loss: 0.016205692663788795
2023-01-21 11:51:20.059471: step: 832/527, loss: 0.00018529892258811742
2023-01-21 11:51:21.172149: step: 836/527, loss: 0.0020374299492686987
2023-01-21 11:51:22.293032: step: 840/527, loss: 0.005025672726333141
2023-01-21 11:51:23.420128: step: 844/527, loss: 0.008652115240693092
2023-01-21 11:51:24.523316: step: 848/527, loss: 0.20478478074073792
2023-01-21 11:51:25.661435: step: 852/527, loss: 0.018135596066713333
2023-01-21 11:51:26.783577: step: 856/527, loss: 0.000851535820402205
2023-01-21 11:51:27.878790: step: 860/527, loss: 0.04741621017456055
2023-01-21 11:51:28.999299: step: 864/527, loss: 0.009001445956528187
2023-01-21 11:51:30.119245: step: 868/527, loss: 0.0010059355990961194
2023-01-21 11:51:31.218502: step: 872/527, loss: 0.043395139276981354
2023-01-21 11:51:32.312509: step: 876/527, loss: 0.005222130101174116
2023-01-21 11:51:33.410934: step: 880/527, loss: 0.050980761647224426
2023-01-21 11:51:34.507639: step: 884/527, loss: 0.08708000928163528
2023-01-21 11:51:35.640713: step: 888/527, loss: 0.010383606888353825
2023-01-21 11:51:36.743469: step: 892/527, loss: 0.015842437744140625
2023-01-21 11:51:37.856085: step: 896/527, loss: 0.0002993583620991558
2023-01-21 11:51:38.989468: step: 900/527, loss: 0.016800498589873314
2023-01-21 11:51:40.115834: step: 904/527, loss: 0.015013694763183594
2023-01-21 11:51:41.266310: step: 908/527, loss: 0.004151248838752508
2023-01-21 11:51:42.390268: step: 912/527, loss: 0.6020898818969727
2023-01-21 11:51:43.510421: step: 916/527, loss: 0.013149452395737171
2023-01-21 11:51:44.618035: step: 920/527, loss: 0.005343389697372913
2023-01-21 11:51:45.729183: step: 924/527, loss: 0.12648582458496094
2023-01-21 11:51:46.857161: step: 928/527, loss: 0.0008691787952557206
2023-01-21 11:51:47.970669: step: 932/527, loss: 0.012345362454652786
2023-01-21 11:51:49.066523: step: 936/527, loss: 0.005781460087746382
2023-01-21 11:51:50.165593: step: 940/527, loss: 0.0277864933013916
2023-01-21 11:51:51.274916: step: 944/527, loss: 0.005160999018698931
2023-01-21 11:51:52.409084: step: 948/527, loss: 0.0010903358925133944
2023-01-21 11:51:53.494046: step: 952/527, loss: 0.0011754990555346012
2023-01-21 11:51:54.599173: step: 956/527, loss: 0.00010132789611816406
2023-01-21 11:51:55.693814: step: 960/527, loss: 0.00043659209040924907
2023-01-21 11:51:56.801137: step: 964/527, loss: 0.00053491594735533
2023-01-21 11:51:57.898918: step: 968/527, loss: 0.012289525009691715
2023-01-21 11:51:59.008265: step: 972/527, loss: 0.06495705246925354
2023-01-21 11:52:00.118294: step: 976/527, loss: 0.0006214663153514266
2023-01-21 11:52:01.252905: step: 980/527, loss: 0.0313752181828022
2023-01-21 11:52:02.359126: step: 984/527, loss: 0.004421997349709272
2023-01-21 11:52:03.483252: step: 988/527, loss: 0.0265058521181345
2023-01-21 11:52:04.575740: step: 992/527, loss: 0.004988765809684992
2023-01-21 11:52:05.653774: step: 996/527, loss: 0.0012411593925207853
2023-01-21 11:52:06.765956: step: 1000/527, loss: 1.573562803969253e-05
2023-01-21 11:52:07.868782: step: 1004/527, loss: 0.004400825593620539
2023-01-21 11:52:08.984775: step: 1008/527, loss: 0.02025900036096573
2023-01-21 11:52:10.103375: step: 1012/527, loss: 0.007733821868896484
2023-01-21 11:52:11.182469: step: 1016/527, loss: 0.00010827779624378309
2023-01-21 11:52:12.291726: step: 1020/527, loss: 0.02309245988726616
2023-01-21 11:52:13.413210: step: 1024/527, loss: 0.00011391640146030113
2023-01-21 11:52:14.513973: step: 1028/527, loss: 0.0014451027382165194
2023-01-21 11:52:15.650033: step: 1032/527, loss: 5.050436973571777
2023-01-21 11:52:16.763671: step: 1036/527, loss: 0.0002795219188556075
2023-01-21 11:52:17.865545: step: 1040/527, loss: 0.0002375125914113596
2023-01-21 11:52:19.001658: step: 1044/527, loss: 0.00471420306712389
2023-01-21 11:52:20.128093: step: 1048/527, loss: 0.04861316829919815
2023-01-21 11:52:21.269297: step: 1052/527, loss: 0.0007120132795535028
2023-01-21 11:52:22.370757: step: 1056/527, loss: 0.01694030873477459
2023-01-21 11:52:23.469818: step: 1060/527, loss: 0.06779942661523819
2023-01-21 11:52:24.574663: step: 1064/527, loss: 0.017715072259306908
2023-01-21 11:52:25.723922: step: 1068/527, loss: 0.0008481026161462069
2023-01-21 11:52:26.856263: step: 1072/527, loss: 0.0011031389003619552
2023-01-21 11:52:27.945665: step: 1076/527, loss: 0.0016887665260583162
2023-01-21 11:52:29.061851: step: 1080/527, loss: 0.006601429078727961
2023-01-21 11:52:30.174426: step: 1084/527, loss: 0.058020591735839844
2023-01-21 11:52:31.274384: step: 1088/527, loss: 0.013167820870876312
2023-01-21 11:52:32.413013: step: 1092/527, loss: 0.018180467188358307
2023-01-21 11:52:33.521555: step: 1096/527, loss: 0.022410202771425247
2023-01-21 11:52:34.605300: step: 1100/527, loss: 0.001178741455078125
2023-01-21 11:52:35.705284: step: 1104/527, loss: 0.00799331720918417
2023-01-21 11:52:36.797558: step: 1108/527, loss: 0.00014337897300720215
2023-01-21 11:52:37.931038: step: 1112/527, loss: 0.008269499987363815
2023-01-21 11:52:39.038710: step: 1116/527, loss: 0.009488487616181374
2023-01-21 11:52:40.138514: step: 1120/527, loss: 0.022758865728974342
2023-01-21 11:52:41.253811: step: 1124/527, loss: 0.0002480506955180317
2023-01-21 11:52:42.362481: step: 1128/527, loss: 0.007994461804628372
2023-01-21 11:52:43.480645: step: 1132/527, loss: 0.010111856274306774
2023-01-21 11:52:44.586748: step: 1136/527, loss: 0.33087652921676636
2023-01-21 11:52:45.695445: step: 1140/527, loss: 0.04574241489171982
2023-01-21 11:52:46.782564: step: 1144/527, loss: 0.002463054610416293
2023-01-21 11:52:47.880795: step: 1148/527, loss: 0.23794154822826385
2023-01-21 11:52:48.967059: step: 1152/527, loss: 0.0018032073276117444
2023-01-21 11:52:50.093872: step: 1156/527, loss: 0.013797283172607422
2023-01-21 11:52:51.229489: step: 1160/527, loss: 0.04929351806640625
2023-01-21 11:52:52.331547: step: 1164/527, loss: 0.003780269529670477
2023-01-21 11:52:53.479155: step: 1168/527, loss: 0.025474930182099342
2023-01-21 11:52:54.573725: step: 1172/527, loss: 0.026148319244384766
2023-01-21 11:52:55.697559: step: 1176/527, loss: 0.004789590835571289
2023-01-21 11:52:56.830351: step: 1180/527, loss: 0.06521129608154297
2023-01-21 11:52:57.926789: step: 1184/527, loss: 0.06087551265954971
2023-01-21 11:52:59.038740: step: 1188/527, loss: 0.006834506988525391
2023-01-21 11:53:00.160909: step: 1192/527, loss: 0.011268138885498047
2023-01-21 11:53:01.257235: step: 1196/527, loss: 0.03778543323278427
2023-01-21 11:53:02.408808: step: 1200/527, loss: 0.003999519627541304
2023-01-21 11:53:03.498097: step: 1204/527, loss: 0.001208496163599193
2023-01-21 11:53:04.612625: step: 1208/527, loss: 0.008374596014618874
2023-01-21 11:53:05.752949: step: 1212/527, loss: 0.03540964052081108
2023-01-21 11:53:06.863254: step: 1216/527, loss: 0.01873922348022461
2023-01-21 11:53:07.972186: step: 1220/527, loss: 0.0034141542855650187
2023-01-21 11:53:09.074062: step: 1224/527, loss: 0.005496692843735218
2023-01-21 11:53:10.185982: step: 1228/527, loss: 0.0016101838555186987
2023-01-21 11:53:11.336371: step: 1232/527, loss: 0.004622363951057196
2023-01-21 11:53:12.474893: step: 1236/527, loss: 0.0005783558008261025
2023-01-21 11:53:13.622806: step: 1240/527, loss: 0.006184959318488836
2023-01-21 11:53:14.752724: step: 1244/527, loss: 0.022483063861727715
2023-01-21 11:53:15.866908: step: 1248/527, loss: 1.0204315913142636e-05
2023-01-21 11:53:16.971817: step: 1252/527, loss: 0.00045022962149232626
2023-01-21 11:53:18.089361: step: 1256/527, loss: 0.015847396105527878
2023-01-21 11:53:19.167810: step: 1260/527, loss: 0.0023582458961755037
2023-01-21 11:53:20.264052: step: 1264/527, loss: 0.011016941629350185
2023-01-21 11:53:21.388488: step: 1268/527, loss: 0.015591287985444069
2023-01-21 11:53:22.495327: step: 1272/527, loss: 6.637573096668348e-05
2023-01-21 11:53:23.588923: step: 1276/527, loss: 4.6443943574558944e-05
2023-01-21 11:53:24.710314: step: 1280/527, loss: 0.00028514862060546875
2023-01-21 11:53:25.835686: step: 1284/527, loss: 0.03325338289141655
2023-01-21 11:53:26.976346: step: 1288/527, loss: 0.017682362347841263
2023-01-21 11:53:28.111477: step: 1292/527, loss: 0.0007347106584347785
2023-01-21 11:53:29.183338: step: 1296/527, loss: 0.00153436663094908
2023-01-21 11:53:30.313251: step: 1300/527, loss: 0.049478720873594284
2023-01-21 11:53:31.370973: step: 1304/527, loss: 0.0007441520574502647
2023-01-21 11:53:32.492642: step: 1308/527, loss: 8.449555025435984e-05
2023-01-21 11:53:33.615165: step: 1312/527, loss: 0.013600158505141735
2023-01-21 11:53:34.711921: step: 1316/527, loss: 0.004617118742316961
2023-01-21 11:53:35.823680: step: 1320/527, loss: 7.476806786144152e-05
2023-01-21 11:53:36.939326: step: 1324/527, loss: 0.02285308949649334
2023-01-21 11:53:38.074758: step: 1328/527, loss: 0.001990032382309437
2023-01-21 11:53:39.190353: step: 1332/527, loss: 0.002467632293701172
2023-01-21 11:53:40.300814: step: 1336/527, loss: 1.7833710444392636e-05
2023-01-21 11:53:41.428778: step: 1340/527, loss: 0.04860043525695801
2023-01-21 11:53:42.545453: step: 1344/527, loss: 0.5575911998748779
2023-01-21 11:53:43.664874: step: 1348/527, loss: 0.003898048307746649
2023-01-21 11:53:44.772803: step: 1352/527, loss: 0.0007877349853515625
2023-01-21 11:53:45.892201: step: 1356/527, loss: 0.0015155792934820056
2023-01-21 11:53:46.971340: step: 1360/527, loss: 0.06365413963794708
2023-01-21 11:53:48.081127: step: 1364/527, loss: 0.0220368392765522
2023-01-21 11:53:49.185618: step: 1368/527, loss: 0.00032845736132003367
2023-01-21 11:53:50.300731: step: 1372/527, loss: 0.050695229321718216
2023-01-21 11:53:51.391212: step: 1376/527, loss: 0.04951038211584091
2023-01-21 11:53:52.503380: step: 1380/527, loss: 8.392335075768642e-06
2023-01-21 11:53:53.598553: step: 1384/527, loss: 0.15241539478302002
2023-01-21 11:53:54.699261: step: 1388/527, loss: 0.0016799926524981856
2023-01-21 11:53:55.822264: step: 1392/527, loss: 0.007445049472153187
2023-01-21 11:53:56.933025: step: 1396/527, loss: 0.08291473984718323
2023-01-21 11:53:58.063890: step: 1400/527, loss: 8.468628220725805e-05
2023-01-21 11:53:59.211662: step: 1404/527, loss: 0.10357294231653214
2023-01-21 11:54:00.321145: step: 1408/527, loss: 0.0027973174583166838
2023-01-21 11:54:01.470662: step: 1412/527, loss: 0.007335376925766468
2023-01-21 11:54:02.629796: step: 1416/527, loss: 0.004297447390854359
2023-01-21 11:54:03.732207: step: 1420/527, loss: 0.0010617256630212069
2023-01-21 11:54:04.856778: step: 1424/527, loss: 0.010309696197509766
2023-01-21 11:54:05.966278: step: 1428/527, loss: 0.0004587173752952367
2023-01-21 11:54:07.078210: step: 1432/527, loss: 0.12448596954345703
2023-01-21 11:54:08.204820: step: 1436/527, loss: 0.014418245293200016
2023-01-21 11:54:09.313798: step: 1440/527, loss: 0.044605061411857605
2023-01-21 11:54:10.398398: step: 1444/527, loss: 5.3024294174974784e-05
2023-01-21 11:54:11.502189: step: 1448/527, loss: 0.10175151377916336
2023-01-21 11:54:12.591455: step: 1452/527, loss: 0.015194701962172985
2023-01-21 11:54:13.697494: step: 1456/527, loss: 0.0005909919855184853
2023-01-21 11:54:14.806140: step: 1460/527, loss: 0.04135628044605255
2023-01-21 11:54:15.909505: step: 1464/527, loss: 0.028210163116455078
2023-01-21 11:54:17.002744: step: 1468/527, loss: 0.0016727447509765625
2023-01-21 11:54:18.126027: step: 1472/527, loss: 0.0003479004080872983
2023-01-21 11:54:19.200294: step: 1476/527, loss: 0.00041565895662643015
2023-01-21 11:54:20.321334: step: 1480/527, loss: 0.001474666642025113
2023-01-21 11:54:21.413841: step: 1484/527, loss: 0.009755611419677734
2023-01-21 11:54:22.498266: step: 1488/527, loss: 0.016039704903960228
2023-01-21 11:54:23.606934: step: 1492/527, loss: 0.006434249691665173
2023-01-21 11:54:24.724573: step: 1496/527, loss: 0.016633987426757812
2023-01-21 11:54:25.834244: step: 1500/527, loss: 0.004258060362190008
2023-01-21 11:54:26.958571: step: 1504/527, loss: 0.006874752230942249
2023-01-21 11:54:28.065577: step: 1508/527, loss: 0.026272868737578392
2023-01-21 11:54:29.207865: step: 1512/527, loss: 0.000469398481072858
2023-01-21 11:54:30.336549: step: 1516/527, loss: 0.00040988920954987407
2023-01-21 11:54:31.447341: step: 1520/527, loss: 0.0004192829073872417
2023-01-21 11:54:32.593536: step: 1524/527, loss: 0.0008742331992834806
2023-01-21 11:54:33.691031: step: 1528/527, loss: 0.4072204530239105
2023-01-21 11:54:34.828314: step: 1532/527, loss: 0.020650483667850494
2023-01-21 11:54:35.938849: step: 1536/527, loss: 0.007707118988037109
2023-01-21 11:54:37.053046: step: 1540/527, loss: 0.006175232119858265
2023-01-21 11:54:38.187099: step: 1544/527, loss: 0.004504489712417126
2023-01-21 11:54:39.282122: step: 1548/527, loss: 0.0011019706726074219
2023-01-21 11:54:40.362526: step: 1552/527, loss: 0.009310722351074219
2023-01-21 11:54:41.512359: step: 1556/527, loss: 0.033133696764707565
2023-01-21 11:54:42.616819: step: 1560/527, loss: 0.0004373550764285028
2023-01-21 11:54:43.748012: step: 1564/527, loss: 0.0021636963356286287
2023-01-21 11:54:44.833439: step: 1568/527, loss: 0.005114078521728516
2023-01-21 11:54:45.940832: step: 1572/527, loss: 0.005603027530014515
2023-01-21 11:54:47.031103: step: 1576/527, loss: 0.0014803410740569234
2023-01-21 11:54:48.099746: step: 1580/527, loss: 0.00031299592228606343
2023-01-21 11:54:49.228621: step: 1584/527, loss: 0.004423046018928289
2023-01-21 11:54:50.334018: step: 1588/527, loss: 0.00604591378942132
2023-01-21 11:54:51.455184: step: 1592/527, loss: 0.06521540135145187
2023-01-21 11:54:52.571626: step: 1596/527, loss: 0.0016664505237713456
2023-01-21 11:54:53.681592: step: 1600/527, loss: 0.0035290243104100227
2023-01-21 11:54:54.810306: step: 1604/527, loss: 0.11249971389770508
2023-01-21 11:54:55.939275: step: 1608/527, loss: 0.03651590272784233
2023-01-21 11:54:57.052351: step: 1612/527, loss: 0.008687973022460938
2023-01-21 11:54:58.180812: step: 1616/527, loss: 0.053115081042051315
2023-01-21 11:54:59.290558: step: 1620/527, loss: 0.0007285118335857987
2023-01-21 11:55:00.380492: step: 1624/527, loss: 0.02037658728659153
2023-01-21 11:55:01.537462: step: 1628/527, loss: 0.002291774842888117
2023-01-21 11:55:02.629825: step: 1632/527, loss: 0.007607841398566961
2023-01-21 11:55:03.741230: step: 1636/527, loss: 0.024045085534453392
2023-01-21 11:55:04.836859: step: 1640/527, loss: 0.004744434729218483
2023-01-21 11:55:05.954323: step: 1644/527, loss: 0.00019474030705168843
2023-01-21 11:55:07.059664: step: 1648/527, loss: 0.07070579379796982
2023-01-21 11:55:08.199382: step: 1652/527, loss: 0.0008762359502725303
2023-01-21 11:55:09.302195: step: 1656/527, loss: 0.003662967821583152
2023-01-21 11:55:10.399423: step: 1660/527, loss: 0.0003383636358194053
2023-01-21 11:55:11.518890: step: 1664/527, loss: 6.895065598655492e-05
2023-01-21 11:55:12.646730: step: 1668/527, loss: 0.017535973340272903
2023-01-21 11:55:13.779367: step: 1672/527, loss: 0.10617122054100037
2023-01-21 11:55:14.907752: step: 1676/527, loss: 0.0004863739013671875
2023-01-21 11:55:16.033353: step: 1680/527, loss: 0.011040115728974342
2023-01-21 11:55:17.142275: step: 1684/527, loss: 0.00426063546910882
2023-01-21 11:55:18.326174: step: 1688/527, loss: 0.0628824234008789
2023-01-21 11:55:19.442710: step: 1692/527, loss: 0.01290226075798273
2023-01-21 11:55:20.567979: step: 1696/527, loss: 0.00955276470631361
2023-01-21 11:55:21.660641: step: 1700/527, loss: 0.00026407241239212453
2023-01-21 11:55:22.792899: step: 1704/527, loss: 0.05009784922003746
2023-01-21 11:55:23.900328: step: 1708/527, loss: 0.04191351309418678
2023-01-21 11:55:24.997231: step: 1712/527, loss: 0.0027766230050474405
2023-01-21 11:55:26.127554: step: 1716/527, loss: 0.01951727829873562
2023-01-21 11:55:27.235218: step: 1720/527, loss: 1.62124638336536e-06
2023-01-21 11:55:28.307731: step: 1724/527, loss: 0.0009394646040163934
2023-01-21 11:55:29.409523: step: 1728/527, loss: 0.0062885284423828125
2023-01-21 11:55:30.538981: step: 1732/527, loss: 0.0038815499283373356
2023-01-21 11:55:31.654312: step: 1736/527, loss: 4.6539309551008046e-05
2023-01-21 11:55:32.766932: step: 1740/527, loss: 0.0009376525995321572
2023-01-21 11:55:33.931973: step: 1744/527, loss: 0.000656509364489466
2023-01-21 11:55:35.048528: step: 1748/527, loss: 0.0020809650886803865
2023-01-21 11:55:36.157330: step: 1752/527, loss: 0.0075584412552416325
2023-01-21 11:55:37.251198: step: 1756/527, loss: 0.016242504119873047
2023-01-21 11:55:38.356474: step: 1760/527, loss: 0.015456486493349075
2023-01-21 11:55:39.485578: step: 1764/527, loss: 0.00047168732271529734
2023-01-21 11:55:40.611040: step: 1768/527, loss: 0.011204337701201439
2023-01-21 11:55:41.727917: step: 1772/527, loss: 0.026050280779600143
2023-01-21 11:55:42.840349: step: 1776/527, loss: 3.213882155250758e-05
2023-01-21 11:55:43.934865: step: 1780/527, loss: 0.07073793560266495
2023-01-21 11:55:45.067774: step: 1784/527, loss: 0.00127582554705441
2023-01-21 11:55:46.173944: step: 1788/527, loss: 0.01715879514813423
2023-01-21 11:55:47.281863: step: 1792/527, loss: 0.0037208558060228825
2023-01-21 11:55:48.412157: step: 1796/527, loss: 0.008762359619140625
2023-01-21 11:55:49.526561: step: 1800/527, loss: 0.02195739932358265
2023-01-21 11:55:50.677519: step: 1804/527, loss: 0.00017976760864257812
2023-01-21 11:55:51.800410: step: 1808/527, loss: 0.06419678032398224
2023-01-21 11:55:52.915525: step: 1812/527, loss: 0.0017028630245476961
2023-01-21 11:55:54.053846: step: 1816/527, loss: 0.008764171972870827
2023-01-21 11:55:55.173659: step: 1820/527, loss: 0.0010064125526696444
2023-01-21 11:55:56.295811: step: 1824/527, loss: 0.0009965896606445312
2023-01-21 11:55:57.413086: step: 1828/527, loss: 0.03653106838464737
2023-01-21 11:55:58.516408: step: 1832/527, loss: 0.0012649536365643144
2023-01-21 11:55:59.600073: step: 1836/527, loss: 0.04128437116742134
2023-01-21 11:56:00.701581: step: 1840/527, loss: 0.009892940521240234
2023-01-21 11:56:01.798041: step: 1844/527, loss: 0.005511665251106024
2023-01-21 11:56:02.914697: step: 1848/527, loss: 0.0699525699019432
2023-01-21 11:56:04.037725: step: 1852/527, loss: 0.04248318821191788
2023-01-21 11:56:05.153748: step: 1856/527, loss: 0.017084650695323944
2023-01-21 11:56:06.268343: step: 1860/527, loss: 0.010862540453672409
2023-01-21 11:56:07.381789: step: 1864/527, loss: 0.09968248009681702
2023-01-21 11:56:08.494228: step: 1868/527, loss: 0.024834442883729935
2023-01-21 11:56:09.568046: step: 1872/527, loss: 0.010555506683886051
2023-01-21 11:56:10.681683: step: 1876/527, loss: 0.0026895522605627775
2023-01-21 11:56:11.818995: step: 1880/527, loss: 0.012289523147046566
2023-01-21 11:56:12.945448: step: 1884/527, loss: 0.002155780792236328
2023-01-21 11:56:14.094014: step: 1888/527, loss: 0.005321407690644264
2023-01-21 11:56:15.222520: step: 1892/527, loss: 0.0006985664367675781
2023-01-21 11:56:16.344173: step: 1896/527, loss: 0.018883895128965378
2023-01-21 11:56:17.420459: step: 1900/527, loss: 1.3208389646024443e-05
2023-01-21 11:56:18.534332: step: 1904/527, loss: 0.004491257481276989
2023-01-21 11:56:19.633116: step: 1908/527, loss: 0.0019748688209801912
2023-01-21 11:56:20.734250: step: 1912/527, loss: 0.05485835298895836
2023-01-21 11:56:21.820231: step: 1916/527, loss: 0.00040683746919967234
2023-01-21 11:56:22.938336: step: 1920/527, loss: 0.014313507825136185
2023-01-21 11:56:24.055581: step: 1924/527, loss: 0.0024770735763013363
2023-01-21 11:56:25.173919: step: 1928/527, loss: 0.012394332326948643
2023-01-21 11:56:26.292023: step: 1932/527, loss: 0.016316033899784088
2023-01-21 11:56:27.413340: step: 1936/527, loss: 0.004023408982902765
2023-01-21 11:56:28.509948: step: 1940/527, loss: 2.8133392333984375e-05
2023-01-21 11:56:29.657513: step: 1944/527, loss: 0.0406283363699913
2023-01-21 11:56:30.741865: step: 1948/527, loss: 0.0002730369451455772
2023-01-21 11:56:31.848531: step: 1952/527, loss: 0.0045524598099291325
2023-01-21 11:56:32.992834: step: 1956/527, loss: 0.003428173018619418
2023-01-21 11:56:34.122110: step: 1960/527, loss: 0.0014935494400560856
2023-01-21 11:56:35.242250: step: 1964/527, loss: 0.0003504753112792969
2023-01-21 11:56:36.367050: step: 1968/527, loss: 0.03166026994585991
2023-01-21 11:56:37.462133: step: 1972/527, loss: 0.006565952207893133
2023-01-21 11:56:38.562548: step: 1976/527, loss: 0.0001829147367971018
2023-01-21 11:56:39.668334: step: 1980/527, loss: 0.0005499362596310675
2023-01-21 11:56:40.760999: step: 1984/527, loss: 0.00029096603975631297
2023-01-21 11:56:41.923939: step: 1988/527, loss: 0.0003496169811114669
2023-01-21 11:56:43.038579: step: 1992/527, loss: 0.004754447843879461
2023-01-21 11:56:44.167676: step: 1996/527, loss: 0.03133583068847656
2023-01-21 11:56:45.298077: step: 2000/527, loss: 0.0011091232299804688
2023-01-21 11:56:46.412668: step: 2004/527, loss: 0.021156834438443184
2023-01-21 11:56:47.531941: step: 2008/527, loss: 0.002025127410888672
2023-01-21 11:56:48.650484: step: 2012/527, loss: 0.5858330726623535
2023-01-21 11:56:49.747578: step: 2016/527, loss: 0.02097921445965767
2023-01-21 11:56:50.903684: step: 2020/527, loss: 0.007396125700324774
2023-01-21 11:56:51.986638: step: 2024/527, loss: 0.00014514924259856343
2023-01-21 11:56:53.107979: step: 2028/527, loss: 0.007003307342529297
2023-01-21 11:56:54.181064: step: 2032/527, loss: 0.0002628088113851845
2023-01-21 11:56:55.270676: step: 2036/527, loss: 0.00835494976490736
2023-01-21 11:56:56.373268: step: 2040/527, loss: 0.0005269050598144531
2023-01-21 11:56:57.512419: step: 2044/527, loss: 0.0018352508777752519
2023-01-21 11:56:58.677618: step: 2048/527, loss: 0.029383469372987747
2023-01-21 11:56:59.826862: step: 2052/527, loss: 0.0007335185655392706
2023-01-21 11:57:00.909590: step: 2056/527, loss: 3.585815284168348e-05
2023-01-21 11:57:02.040203: step: 2060/527, loss: 2.346038854739163e-05
2023-01-21 11:57:03.151426: step: 2064/527, loss: 0.006548976991325617
2023-01-21 11:57:04.269506: step: 2068/527, loss: 0.002814674284309149
2023-01-21 11:57:05.387082: step: 2072/527, loss: 0.001100254012271762
2023-01-21 11:57:06.504652: step: 2076/527, loss: 0.0017435074551030993
2023-01-21 11:57:07.601421: step: 2080/527, loss: 0.014860248193144798
2023-01-21 11:57:08.720860: step: 2084/527, loss: 0.02179727517068386
2023-01-21 11:57:09.804044: step: 2088/527, loss: 0.0015678404597565532
2023-01-21 11:57:10.921109: step: 2092/527, loss: 0.21712498366832733
2023-01-21 11:57:12.037606: step: 2096/527, loss: 0.004669570829719305
2023-01-21 11:57:13.123354: step: 2100/527, loss: 3.80516066798009e-05
2023-01-21 11:57:14.291125: step: 2104/527, loss: 0.055962562561035156
2023-01-21 11:57:15.399751: step: 2108/527, loss: 0.018947506323456764
==================================================
Loss: 0.039
--------------------
Dev: {'event': {'p': 0.626410835214447, 'r': 0.7390146471371505, 'f1': 0.678069639584606}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19}
Test: {'event': {'p': 0.6311047889995258, 'r': 0.7605714285714286, 'f1': 0.6898160145115314}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19}
Chinese: {'event': {'p': 0.5657894736842105, 'r': 0.7962962962962963, 'f1': 0.6615384615384615}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19}
Korean: {'event': {'p': 0.6170212765957447, 'r': 0.4603174603174603, 'f1': 0.5272727272727272}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19}
Russian: {'event': {'p': 0.4722222222222222, 'r': 0.4722222222222222, 'f1': 0.4722222222222222}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 11:57:56.423194: step: 4/527, loss: 3.4332275390625e-05
2023-01-21 11:57:57.503051: step: 8/527, loss: 3.547668529790826e-05
2023-01-21 11:57:58.626440: step: 12/527, loss: 0.0012718200450763106
2023-01-21 11:57:59.745985: step: 16/527, loss: 0.13088759779930115
2023-01-21 11:58:00.835289: step: 20/527, loss: 0.007852173410356045
2023-01-21 11:58:02.265889: step: 24/527, loss: 8.296966552734375e-05
2023-01-21 11:58:03.359063: step: 28/527, loss: 0.0007997513166628778
2023-01-21 11:58:04.476661: step: 32/527, loss: 0.00011363029625499621
2023-01-21 11:58:05.576243: step: 36/527, loss: 0.0019147873390465975
2023-01-21 11:58:06.670743: step: 40/527, loss: 7.109642319846898e-05
2023-01-21 11:58:07.797890: step: 44/527, loss: 0.008765220642089844
2023-01-21 11:58:08.929324: step: 48/527, loss: 0.09263362735509872
2023-01-21 11:58:10.075811: step: 52/527, loss: 0.04531841352581978
2023-01-21 11:58:11.175254: step: 56/527, loss: 2.927780406025704e-05
2023-01-21 11:58:12.312712: step: 60/527, loss: 0.0006204604869708419
2023-01-21 11:58:13.402496: step: 64/527, loss: 7.004738290561363e-05
2023-01-21 11:58:14.536863: step: 68/527, loss: 0.0672679916024208
2023-01-21 11:58:15.659087: step: 72/527, loss: 0.0907646119594574
2023-01-21 11:58:16.741650: step: 76/527, loss: 0.0064262389205396175
2023-01-21 11:58:17.857466: step: 80/527, loss: 0.015196133404970169
2023-01-21 11:58:18.976963: step: 84/527, loss: 0.0004684448358602822
2023-01-21 11:58:20.090022: step: 88/527, loss: 0.0026604891754686832
2023-01-21 11:58:21.198871: step: 92/527, loss: 0.00017480850510764867
2023-01-21 11:58:22.335937: step: 96/527, loss: 0.000293731689453125
2023-01-21 11:58:23.477367: step: 100/527, loss: 0.0004915237659588456
2023-01-21 11:58:24.556480: step: 104/527, loss: 0.0005902289994992316
2023-01-21 11:58:25.641288: step: 108/527, loss: 0.020103169605135918
2023-01-21 11:58:26.773969: step: 112/527, loss: 4.482269287109375e-05
2023-01-21 11:58:27.884307: step: 116/527, loss: 7.62939453125e-06
2023-01-21 11:58:28.993768: step: 120/527, loss: 5.426407005870715e-05
2023-01-21 11:58:30.096747: step: 124/527, loss: 0.17228764295578003
2023-01-21 11:58:31.192005: step: 128/527, loss: 0.0009261191007681191
2023-01-21 11:58:32.323569: step: 132/527, loss: 2.193450927734375e-05
2023-01-21 11:58:33.440871: step: 136/527, loss: 0.0014945983421057463
2023-01-21 11:58:34.557547: step: 140/527, loss: 0.0003391265927348286
2023-01-21 11:58:35.672665: step: 144/527, loss: 0.00045261383638717234
2023-01-21 11:58:36.768385: step: 148/527, loss: 1.5258790426742053e-06
2023-01-21 11:58:37.949060: step: 152/527, loss: 0.04457683488726616
2023-01-21 11:58:39.067612: step: 156/527, loss: 7.62939453125e-05
2023-01-21 11:58:40.185863: step: 160/527, loss: -1.8119810647476697e-06
2023-01-21 11:58:41.250913: step: 164/527, loss: 1.9454957509879023e-05
2023-01-21 11:58:42.365508: step: 168/527, loss: 0.05544090270996094
2023-01-21 11:58:43.457382: step: 172/527, loss: 0.003456211183220148
2023-01-21 11:58:44.590620: step: 176/527, loss: 0.008368110284209251
2023-01-21 11:58:45.707517: step: 180/527, loss: 0.0011278152232989669
2023-01-21 11:58:46.792170: step: 184/527, loss: 0.0002906799491029233
2023-01-21 11:58:47.867111: step: 188/527, loss: 0.03186435624957085
2023-01-21 11:58:48.963274: step: 192/527, loss: 0.011450577527284622
2023-01-21 11:58:50.073317: step: 196/527, loss: 0.0017066001892089844
2023-01-21 11:58:51.216284: step: 200/527, loss: 0.03494129329919815
2023-01-21 11:58:52.315864: step: 204/527, loss: 4.673004150390625e-05
2023-01-21 11:58:53.407038: step: 208/527, loss: 0.0001520156947663054
2023-01-21 11:58:54.517729: step: 212/527, loss: 0.010197353549301624
2023-01-21 11:58:55.626780: step: 216/527, loss: 0.00014171600923873484
2023-01-21 11:58:56.715539: step: 220/527, loss: 0.0008772850269451737
2023-01-21 11:58:57.861179: step: 224/527, loss: 0.0011756897438317537
2023-01-21 11:58:58.967982: step: 228/527, loss: 0.2603588104248047
2023-01-21 11:59:00.126117: step: 232/527, loss: 0.021329879760742188
2023-01-21 11:59:01.238152: step: 236/527, loss: 0.06982040405273438
2023-01-21 11:59:02.348098: step: 240/527, loss: 0.00021991730318404734
2023-01-21 11:59:03.468963: step: 244/527, loss: 0.04594488441944122
2023-01-21 11:59:04.596259: step: 248/527, loss: 0.0031266212463378906
2023-01-21 11:59:05.715616: step: 252/527, loss: 0.0005491256597451866
2023-01-21 11:59:06.853098: step: 256/527, loss: 0.00034198761568404734
2023-01-21 11:59:07.957443: step: 260/527, loss: 0.03650055080652237
2023-01-21 11:59:09.069936: step: 264/527, loss: 0.02214033715426922
2023-01-21 11:59:10.188618: step: 268/527, loss: 0.00029859543428756297
2023-01-21 11:59:11.295671: step: 272/527, loss: 9.002685692394152e-05
2023-01-21 11:59:12.409990: step: 276/527, loss: 0.0001253128139069304
2023-01-21 11:59:13.543135: step: 280/527, loss: 0.0002082824648823589
2023-01-21 11:59:14.638251: step: 284/527, loss: 0.009705162607133389
2023-01-21 11:59:15.762920: step: 288/527, loss: 1.144409225162235e-06
2023-01-21 11:59:16.906118: step: 292/527, loss: 0.010892581194639206
2023-01-21 11:59:18.000125: step: 296/527, loss: 0.025179386138916016
2023-01-21 11:59:19.133721: step: 300/527, loss: 0.0012948036892339587
2023-01-21 11:59:20.274016: step: 304/527, loss: 0.0010653496719896793
2023-01-21 11:59:21.397294: step: 308/527, loss: 0.00011682510375976562
2023-01-21 11:59:22.514122: step: 312/527, loss: 0.002443504286929965
2023-01-21 11:59:23.658784: step: 316/527, loss: 0.007702446077018976
2023-01-21 11:59:24.793219: step: 320/527, loss: 0.007098579779267311
2023-01-21 11:59:25.909070: step: 324/527, loss: 0.006360197439789772
2023-01-21 11:59:27.023759: step: 328/527, loss: 0.00254497560672462
2023-01-21 11:59:28.166186: step: 332/527, loss: 0.01922626607120037
2023-01-21 11:59:29.258870: step: 336/527, loss: 0.00019149779109284282
2023-01-21 11:59:30.376807: step: 340/527, loss: 0.024216175079345703
2023-01-21 11:59:31.557907: step: 344/527, loss: 0.027865981683135033
2023-01-21 11:59:32.676823: step: 348/527, loss: 0.007172203157097101
2023-01-21 11:59:33.800027: step: 352/527, loss: 0.05468282848596573
2023-01-21 11:59:34.912088: step: 356/527, loss: 0.00286350236274302
2023-01-21 11:59:36.040694: step: 360/527, loss: 0.026781558990478516
2023-01-21 11:59:37.177187: step: 364/527, loss: 0.0016329765785485506
2023-01-21 11:59:38.269892: step: 368/527, loss: 0.01660633087158203
2023-01-21 11:59:39.367435: step: 372/527, loss: 0.02549591101706028
2023-01-21 11:59:40.496057: step: 376/527, loss: 6.29425048828125e-05
2023-01-21 11:59:41.648373: step: 380/527, loss: 0.0035185813903808594
2023-01-21 11:59:42.757287: step: 384/527, loss: 0.002018308499827981
2023-01-21 11:59:43.907421: step: 388/527, loss: 0.0009639739873819053
2023-01-21 11:59:45.004591: step: 392/527, loss: 0.003463316010311246
2023-01-21 11:59:46.124264: step: 396/527, loss: 0.004067802801728249
2023-01-21 11:59:47.240081: step: 400/527, loss: 0.04203357547521591
2023-01-21 11:59:48.338554: step: 404/527, loss: 0.0005035400390625
2023-01-21 11:59:49.463561: step: 408/527, loss: 0.00013189316086936742
2023-01-21 11:59:50.598082: step: 412/527, loss: 0.0017994879744946957
2023-01-21 11:59:51.706640: step: 416/527, loss: 0.047010134905576706
2023-01-21 11:59:52.822642: step: 420/527, loss: 0.0003540039178915322
2023-01-21 11:59:53.899137: step: 424/527, loss: 0.0002872943878173828
2023-01-21 11:59:55.000652: step: 428/527, loss: 0.0030436518136411905
2023-01-21 11:59:56.125231: step: 432/527, loss: 0.0007802963373251259
2023-01-21 11:59:57.203249: step: 436/527, loss: 0.019963454455137253
2023-01-21 11:59:58.318418: step: 440/527, loss: 0.04179992899298668
2023-01-21 11:59:59.436908: step: 444/527, loss: 0.0018626212840899825
2023-01-21 12:00:00.533705: step: 448/527, loss: 0.02898721769452095
2023-01-21 12:00:01.630034: step: 452/527, loss: 6.0749058320652694e-05
2023-01-21 12:00:02.737080: step: 456/527, loss: 0.0012759207747876644
2023-01-21 12:00:03.865239: step: 460/527, loss: 0.0002582549932412803
2023-01-21 12:00:04.976370: step: 464/527, loss: 0.030570032075047493
2023-01-21 12:00:06.075493: step: 468/527, loss: 0.0023875238839536905
2023-01-21 12:00:07.200164: step: 472/527, loss: 0.011507606133818626
2023-01-21 12:00:08.317690: step: 476/527, loss: 0.004371261689811945
2023-01-21 12:00:09.421988: step: 480/527, loss: 0.007986831478774548
2023-01-21 12:00:10.548794: step: 484/527, loss: 0.0025652884505689144
2023-01-21 12:00:11.663070: step: 488/527, loss: 0.014095497317612171
2023-01-21 12:00:12.798468: step: 492/527, loss: 0.011564064770936966
2023-01-21 12:00:13.915370: step: 496/527, loss: 0.004274559207260609
2023-01-21 12:00:15.043157: step: 500/527, loss: 0.017993737012147903
2023-01-21 12:00:16.171116: step: 504/527, loss: 0.000107812877104152
2023-01-21 12:00:17.289856: step: 508/527, loss: 0.003637599991634488
2023-01-21 12:00:18.392848: step: 512/527, loss: 0.0009717941284179688
2023-01-21 12:00:19.536773: step: 516/527, loss: 0.037944599986076355
2023-01-21 12:00:20.676816: step: 520/527, loss: 0.0600128173828125
2023-01-21 12:00:21.790535: step: 524/527, loss: 0.01699652150273323
2023-01-21 12:00:22.876852: step: 528/527, loss: 0.008876705542206764
2023-01-21 12:00:23.960280: step: 532/527, loss: 0.00016136169142555445
2023-01-21 12:00:25.051967: step: 536/527, loss: 0.04047584533691406
2023-01-21 12:00:26.169438: step: 540/527, loss: 0.021314620971679688
2023-01-21 12:00:27.242893: step: 544/527, loss: 0.01648111455142498
2023-01-21 12:00:28.334819: step: 548/527, loss: 0.03600750118494034
2023-01-21 12:00:29.454990: step: 552/527, loss: 0.0039390563033521175
2023-01-21 12:00:30.556729: step: 556/527, loss: 3.471374657237902e-05
2023-01-21 12:00:31.711776: step: 560/527, loss: 0.029330160468816757
2023-01-21 12:00:32.816111: step: 564/527, loss: 0.04944801330566406
2023-01-21 12:00:33.939544: step: 568/527, loss: 0.008247566409409046
2023-01-21 12:00:35.092270: step: 572/527, loss: 0.011773395352065563
2023-01-21 12:00:36.185929: step: 576/527, loss: 7.591248140670359e-05
2023-01-21 12:00:37.296986: step: 580/527, loss: 0.04059505835175514
2023-01-21 12:00:38.403576: step: 584/527, loss: 3.337860107421875e-05
2023-01-21 12:00:39.536886: step: 588/527, loss: 0.0006385803571902215
2023-01-21 12:00:40.648030: step: 592/527, loss: 0.043839357793331146
2023-01-21 12:00:41.739685: step: 596/527, loss: 0.00633692741394043
2023-01-21 12:00:42.848651: step: 600/527, loss: 0.00019965172396041453
2023-01-21 12:00:43.942045: step: 604/527, loss: 0.00012164115469204262
2023-01-21 12:00:45.022681: step: 608/527, loss: 0.00040493012056685984
2023-01-21 12:00:46.125707: step: 612/527, loss: 0.05111751705408096
2023-01-21 12:00:47.236167: step: 616/527, loss: 0.01307830773293972
2023-01-21 12:00:48.343998: step: 620/527, loss: 0.00023560522822663188
2023-01-21 12:00:49.481474: step: 624/527, loss: 0.00023727417283225805
2023-01-21 12:00:50.573393: step: 628/527, loss: 0.0669642984867096
2023-01-21 12:00:51.700704: step: 632/527, loss: 0.05983276292681694
2023-01-21 12:00:52.821985: step: 636/527, loss: 0.000390899193007499
2023-01-21 12:00:53.948612: step: 640/527, loss: 0.04865474998950958
2023-01-21 12:00:55.038876: step: 644/527, loss: 0.0013248443137854338
2023-01-21 12:00:56.136934: step: 648/527, loss: 0.0006049156654626131
2023-01-21 12:00:57.270280: step: 652/527, loss: 0.0002872467157430947
2023-01-21 12:00:58.350264: step: 656/527, loss: 0.0272811409085989
2023-01-21 12:00:59.472595: step: 660/527, loss: 0.25415247678756714
2023-01-21 12:01:00.583223: step: 664/527, loss: 0.029935359954833984
2023-01-21 12:01:01.703108: step: 668/527, loss: 0.04670019447803497
2023-01-21 12:01:02.818694: step: 672/527, loss: 0.00045957567635923624
2023-01-21 12:01:03.933100: step: 676/527, loss: 0.0037940978072583675
2023-01-21 12:01:05.027858: step: 680/527, loss: 0.040196992456912994
2023-01-21 12:01:06.118957: step: 684/527, loss: 0.011786842718720436
2023-01-21 12:01:07.248572: step: 688/527, loss: 0.02827005460858345
2023-01-21 12:01:08.375979: step: 692/527, loss: 0.02662963978946209
2023-01-21 12:01:09.492786: step: 696/527, loss: 0.21253737807273865
2023-01-21 12:01:10.660885: step: 700/527, loss: 0.0035976411309093237
2023-01-21 12:01:11.776363: step: 704/527, loss: 0.0009593010181561112
2023-01-21 12:01:12.891685: step: 708/527, loss: 0.008925819769501686
2023-01-21 12:01:13.994982: step: 712/527, loss: 0.0007654189830645919
2023-01-21 12:01:15.128454: step: 716/527, loss: 0.006103515625
2023-01-21 12:01:16.248942: step: 720/527, loss: 0.03998818248510361
2023-01-21 12:01:17.371381: step: 724/527, loss: 0.0035259248688817024
2023-01-21 12:01:18.477995: step: 728/527, loss: 0.01492223795503378
2023-01-21 12:01:19.609738: step: 732/527, loss: 0.001525116036646068
2023-01-21 12:01:20.707042: step: 736/527, loss: 0.0023641586303710938
2023-01-21 12:01:21.806538: step: 740/527, loss: 0.00135297782253474
2023-01-21 12:01:22.882604: step: 744/527, loss: 0.006973076146095991
2023-01-21 12:01:24.012543: step: 748/527, loss: 0.0015350342728197575
2023-01-21 12:01:25.107832: step: 752/527, loss: 0.019895363599061966
2023-01-21 12:01:26.209141: step: 756/527, loss: 0.003822136204689741
2023-01-21 12:01:27.318331: step: 760/527, loss: 0.0054184915497899055
2023-01-21 12:01:28.447829: step: 764/527, loss: 0.02983722649514675
2023-01-21 12:01:29.538111: step: 768/527, loss: 0.0013895034790039062
2023-01-21 12:01:30.655218: step: 772/527, loss: 2.2029875253792852e-05
2023-01-21 12:01:31.777759: step: 776/527, loss: 0.00964050367474556
2023-01-21 12:01:32.899191: step: 780/527, loss: 0.0019660950638353825
2023-01-21 12:01:34.011502: step: 784/527, loss: 0.012362576089799404
2023-01-21 12:01:35.127713: step: 788/527, loss: 0.001563215279020369
2023-01-21 12:01:36.248818: step: 792/527, loss: 0.0016529083950445056
2023-01-21 12:01:37.358868: step: 796/527, loss: 0.013444995507597923
2023-01-21 12:01:38.464563: step: 800/527, loss: 0.03771228715777397
2023-01-21 12:01:39.612341: step: 804/527, loss: 0.02162233740091324
2023-01-21 12:01:40.741556: step: 808/527, loss: 0.00601959228515625
2023-01-21 12:01:41.838719: step: 812/527, loss: 0.08127345889806747
2023-01-21 12:01:42.964317: step: 816/527, loss: 0.00650787353515625
2023-01-21 12:01:44.072150: step: 820/527, loss: 0.0005459785461425781
2023-01-21 12:01:45.166171: step: 824/527, loss: 0.0197772029787302
2023-01-21 12:01:46.298934: step: 828/527, loss: 0.0019708634354174137
2023-01-21 12:01:47.403154: step: 832/527, loss: 0.0007402420160360634
2023-01-21 12:01:48.518731: step: 836/527, loss: 9.5367431640625e-07
2023-01-21 12:01:49.632011: step: 840/527, loss: 0.0016109467251226306
2023-01-21 12:01:50.767587: step: 844/527, loss: 0.020163822919130325
2023-01-21 12:01:51.879211: step: 848/527, loss: 0.010096645914018154
2023-01-21 12:01:52.953982: step: 852/527, loss: 0.02176341973245144
2023-01-21 12:01:54.088566: step: 856/527, loss: 0.16528044641017914
2023-01-21 12:01:55.190501: step: 860/527, loss: 0.0001447677641408518
2023-01-21 12:01:56.288152: step: 864/527, loss: 0.0037090301048010588
2023-01-21 12:01:57.404461: step: 868/527, loss: 0.001271915389224887
2023-01-21 12:01:58.521589: step: 872/527, loss: 0.0018568038940429688
2023-01-21 12:01:59.617657: step: 876/527, loss: 0.0020357132889330387
2023-01-21 12:02:00.725732: step: 880/527, loss: 0.02779207192361355
2023-01-21 12:02:01.852361: step: 884/527, loss: 0.008112144656479359
2023-01-21 12:02:02.984297: step: 888/527, loss: 0.006324195768684149
2023-01-21 12:02:04.094595: step: 892/527, loss: 4.9591064453125e-05
2023-01-21 12:02:05.236892: step: 896/527, loss: 0.05727434158325195
2023-01-21 12:02:06.337542: step: 900/527, loss: 5.3596493671648204e-05
2023-01-21 12:02:07.418394: step: 904/527, loss: 0.007550620939582586
2023-01-21 12:02:08.518171: step: 908/527, loss: 0.029595421627163887
2023-01-21 12:02:09.613237: step: 912/527, loss: 0.03451580926775932
2023-01-21 12:02:10.731676: step: 916/527, loss: 0.38295039534568787
2023-01-21 12:02:11.874201: step: 920/527, loss: 0.00825195387005806
2023-01-21 12:02:12.963526: step: 924/527, loss: 0.004049396608024836
2023-01-21 12:02:14.070068: step: 928/527, loss: 0.0031559946946799755
2023-01-21 12:02:15.186873: step: 932/527, loss: 0.02059326134622097
2023-01-21 12:02:16.334201: step: 936/527, loss: 0.7522391676902771
2023-01-21 12:02:17.463630: step: 940/527, loss: 0.015984343364834785
2023-01-21 12:02:18.579184: step: 944/527, loss: 0.0006873130332678556
2023-01-21 12:02:19.695708: step: 948/527, loss: 0.0054479604586958885
2023-01-21 12:02:20.815031: step: 952/527, loss: 0.001815700437873602
2023-01-21 12:02:21.911253: step: 956/527, loss: 0.0024441718123853207
2023-01-21 12:02:23.048406: step: 960/527, loss: 0.0016455650329589844
2023-01-21 12:02:24.161515: step: 964/527, loss: 0.00030040740966796875
2023-01-21 12:02:25.264205: step: 968/527, loss: 0.021976470947265625
2023-01-21 12:02:26.366441: step: 972/527, loss: 0.002544403076171875
2023-01-21 12:02:27.478771: step: 976/527, loss: 0.05015239864587784
2023-01-21 12:02:28.585628: step: 980/527, loss: 0.006952381227165461
2023-01-21 12:02:29.692923: step: 984/527, loss: 0.0010344506008550525
2023-01-21 12:02:30.814842: step: 988/527, loss: 0.002208113670349121
2023-01-21 12:02:31.887775: step: 992/527, loss: 0.00028774738893844187
2023-01-21 12:02:32.982322: step: 996/527, loss: 0.00961084384471178
2023-01-21 12:02:34.091678: step: 1000/527, loss: 0.0016227723099291325
2023-01-21 12:02:35.195639: step: 1004/527, loss: 0.007226848509162664
2023-01-21 12:02:36.304831: step: 1008/527, loss: 0.046210192143917084
2023-01-21 12:02:37.400572: step: 1012/527, loss: 0.000362634687917307
2023-01-21 12:02:38.547200: step: 1016/527, loss: 0.006450653076171875
2023-01-21 12:02:39.653884: step: 1020/527, loss: 0.017424391582608223
2023-01-21 12:02:40.758197: step: 1024/527, loss: 0.010802841745316982
2023-01-21 12:02:41.884150: step: 1028/527, loss: 0.03613724932074547
2023-01-21 12:02:43.001467: step: 1032/527, loss: 0.007034587673842907
2023-01-21 12:02:44.089376: step: 1036/527, loss: 0.0004828452947549522
2023-01-21 12:02:45.223959: step: 1040/527, loss: 0.0061883931048214436
2023-01-21 12:02:46.336636: step: 1044/527, loss: 0.006003951653838158
2023-01-21 12:02:47.436759: step: 1048/527, loss: 0.0011585236061364412
2023-01-21 12:02:48.545753: step: 1052/527, loss: 0.002670479007065296
2023-01-21 12:02:49.662851: step: 1056/527, loss: 0.002221822738647461
2023-01-21 12:02:50.755479: step: 1060/527, loss: 0.008598614484071732
2023-01-21 12:02:51.840416: step: 1064/527, loss: 0.029587937518954277
2023-01-21 12:02:52.931649: step: 1068/527, loss: 8.630752745375503e-06
2023-01-21 12:02:54.033731: step: 1072/527, loss: 0.023863792419433594
2023-01-21 12:02:55.144511: step: 1076/527, loss: 0.015695666894316673
2023-01-21 12:02:56.259005: step: 1080/527, loss: 0.005527401342988014
2023-01-21 12:02:57.387758: step: 1084/527, loss: 0.01373825129121542
2023-01-21 12:02:58.513808: step: 1088/527, loss: 0.016524458304047585
2023-01-21 12:02:59.659819: step: 1092/527, loss: 0.007532882504165173
2023-01-21 12:03:00.808775: step: 1096/527, loss: 0.02640705183148384
2023-01-21 12:03:01.930166: step: 1100/527, loss: 0.0006666183471679688
2023-01-21 12:03:03.045088: step: 1104/527, loss: 0.012154245749115944
2023-01-21 12:03:04.150770: step: 1108/527, loss: 0.0009609222179278731
2023-01-21 12:03:05.301962: step: 1112/527, loss: 0.00024280548677779734
2023-01-21 12:03:06.416842: step: 1116/527, loss: 0.007675552275031805
2023-01-21 12:03:07.537957: step: 1120/527, loss: 0.008370542898774147
2023-01-21 12:03:08.653287: step: 1124/527, loss: 0.00086722377454862
2023-01-21 12:03:09.763862: step: 1128/527, loss: 0.008390497416257858
2023-01-21 12:03:10.872515: step: 1132/527, loss: 0.026456832885742188
2023-01-21 12:03:11.968310: step: 1136/527, loss: 0.0008254528511315584
2023-01-21 12:03:13.078029: step: 1140/527, loss: 0.012426377274096012
2023-01-21 12:03:14.202993: step: 1144/527, loss: 0.014566803351044655
2023-01-21 12:03:15.335509: step: 1148/527, loss: 0.012318992987275124
2023-01-21 12:03:16.456413: step: 1152/527, loss: 0.0003888130304403603
2023-01-21 12:03:17.563885: step: 1156/527, loss: 0.013282394036650658
2023-01-21 12:03:18.663025: step: 1160/527, loss: 0.002064991043880582
2023-01-21 12:03:19.785231: step: 1164/527, loss: 0.009982109069824219
2023-01-21 12:03:20.934054: step: 1168/527, loss: 0.017261316999793053
2023-01-21 12:03:22.038183: step: 1172/527, loss: 0.011104965582489967
2023-01-21 12:03:23.122176: step: 1176/527, loss: 0.004796791356056929
2023-01-21 12:03:24.229514: step: 1180/527, loss: 2.0122528439969756e-05
2023-01-21 12:03:25.334095: step: 1184/527, loss: 0.02539539337158203
2023-01-21 12:03:26.466230: step: 1188/527, loss: 0.0001730919029796496
2023-01-21 12:03:27.564339: step: 1192/527, loss: 0.0024204254150390625
2023-01-21 12:03:28.633679: step: 1196/527, loss: 0.03469066694378853
2023-01-21 12:03:29.757611: step: 1200/527, loss: 0.009428691118955612
2023-01-21 12:03:30.862455: step: 1204/527, loss: 0.015006923116743565
2023-01-21 12:03:31.981435: step: 1208/527, loss: 0.0003141403431072831
2023-01-21 12:03:33.096799: step: 1212/527, loss: 0.003090381622314453
2023-01-21 12:03:34.203171: step: 1216/527, loss: 0.01769104041159153
2023-01-21 12:03:35.356978: step: 1220/527, loss: 0.0014437675708904862
2023-01-21 12:03:36.486709: step: 1224/527, loss: 0.03375072404742241
2023-01-21 12:03:37.569106: step: 1228/527, loss: 0.0007547378772869706
2023-01-21 12:03:38.688707: step: 1232/527, loss: 0.04166841506958008
2023-01-21 12:03:39.824939: step: 1236/527, loss: 0.00985939521342516
2023-01-21 12:03:40.909849: step: 1240/527, loss: 0.00032000543433241546
2023-01-21 12:03:42.053743: step: 1244/527, loss: 0.0033652307465672493
2023-01-21 12:03:43.178496: step: 1248/527, loss: 0.015625953674316406
2023-01-21 12:03:44.282298: step: 1252/527, loss: 0.010489463806152344
2023-01-21 12:03:45.401338: step: 1256/527, loss: 0.00011768341209972277
2023-01-21 12:03:46.505172: step: 1260/527, loss: 0.04278545454144478
2023-01-21 12:03:47.634407: step: 1264/527, loss: 7.123947580112144e-05
2023-01-21 12:03:48.764078: step: 1268/527, loss: 0.008060836233198643
2023-01-21 12:03:49.875681: step: 1272/527, loss: 0.011568451300263405
2023-01-21 12:03:51.016834: step: 1276/527, loss: 0.03861665725708008
2023-01-21 12:03:52.135359: step: 1280/527, loss: 0.0035602569114416838
2023-01-21 12:03:53.256725: step: 1284/527, loss: 0.018339728936553
2023-01-21 12:03:54.398258: step: 1288/527, loss: 0.00015249251737259328
2023-01-21 12:03:55.535523: step: 1292/527, loss: 0.009997749701142311
2023-01-21 12:03:56.645226: step: 1296/527, loss: 0.07165279984474182
2023-01-21 12:03:57.772356: step: 1300/527, loss: 0.007937287911772728
2023-01-21 12:03:58.932516: step: 1304/527, loss: 0.00026302336482331157
2023-01-21 12:04:00.043454: step: 1308/527, loss: 0.0006668090936727822
2023-01-21 12:04:01.165475: step: 1312/527, loss: 0.04505300521850586
2023-01-21 12:04:02.276539: step: 1316/527, loss: 0.016960715875029564
2023-01-21 12:04:03.420914: step: 1320/527, loss: 3.4618376957951114e-05
2023-01-21 12:04:04.550086: step: 1324/527, loss: 0.00042247772216796875
2023-01-21 12:04:05.656163: step: 1328/527, loss: 0.0004337310965638608
2023-01-21 12:04:06.787586: step: 1332/527, loss: 0.004719734191894531
2023-01-21 12:04:07.939565: step: 1336/527, loss: 0.026227571070194244
2023-01-21 12:04:09.108184: step: 1340/527, loss: 0.028843021020293236
2023-01-21 12:04:10.226036: step: 1344/527, loss: 0.009689902886748314
2023-01-21 12:04:11.342583: step: 1348/527, loss: 0.00145721435546875
2023-01-21 12:04:12.433870: step: 1352/527, loss: 0.009174251928925514
2023-01-21 12:04:13.541904: step: 1356/527, loss: 0.0005667686928063631
2023-01-21 12:04:14.643498: step: 1360/527, loss: 0.002259874250739813
2023-01-21 12:04:15.767337: step: 1364/527, loss: 0.030611135065555573
2023-01-21 12:04:16.870889: step: 1368/527, loss: 0.011901665478944778
2023-01-21 12:04:17.999506: step: 1372/527, loss: 0.004137611482292414
2023-01-21 12:04:19.140285: step: 1376/527, loss: 0.005522060673683882
2023-01-21 12:04:20.273820: step: 1380/527, loss: 0.001278781914152205
2023-01-21 12:04:21.395341: step: 1384/527, loss: 0.0007923126104287803
2023-01-21 12:04:22.505334: step: 1388/527, loss: 0.02739124186336994
2023-01-21 12:04:23.632260: step: 1392/527, loss: 0.0018606185913085938
2023-01-21 12:04:24.740134: step: 1396/527, loss: 0.014475155621767044
2023-01-21 12:04:25.840919: step: 1400/527, loss: 0.0003971099795307964
2023-01-21 12:04:26.980010: step: 1404/527, loss: 0.01198568381369114
2023-01-21 12:04:28.106043: step: 1408/527, loss: 0.004773902706801891
2023-01-21 12:04:29.215021: step: 1412/527, loss: 0.003279113909229636
2023-01-21 12:04:30.322247: step: 1416/527, loss: 0.022846031934022903
2023-01-21 12:04:31.432922: step: 1420/527, loss: 0.0006052017561160028
2023-01-21 12:04:32.551247: step: 1424/527, loss: 0.007365536410361528
2023-01-21 12:04:33.669121: step: 1428/527, loss: 0.0043697357177734375
2023-01-21 12:04:34.795975: step: 1432/527, loss: 0.002794933505356312
2023-01-21 12:04:35.966146: step: 1436/527, loss: 0.036139871925115585
2023-01-21 12:04:37.098456: step: 1440/527, loss: 0.028938675299286842
2023-01-21 12:04:38.238460: step: 1444/527, loss: 0.010556983761489391
2023-01-21 12:04:39.359384: step: 1448/527, loss: 0.011183071881532669
2023-01-21 12:04:40.453960: step: 1452/527, loss: 0.0015433788066729903
2023-01-21 12:04:41.561931: step: 1456/527, loss: 0.00016460419283248484
2023-01-21 12:04:42.678027: step: 1460/527, loss: 0.17852649092674255
2023-01-21 12:04:43.819869: step: 1464/527, loss: 0.014683246612548828
2023-01-21 12:04:44.973602: step: 1468/527, loss: 0.0011581419967114925
2023-01-21 12:04:46.099451: step: 1472/527, loss: 0.008943319320678711
2023-01-21 12:04:47.224237: step: 1476/527, loss: 0.036548420786857605
2023-01-21 12:04:48.329199: step: 1480/527, loss: 0.003646278288215399
2023-01-21 12:04:49.454331: step: 1484/527, loss: 0.00031147003755904734
2023-01-21 12:04:50.558080: step: 1488/527, loss: 0.0014142035506665707
2023-01-21 12:04:51.664631: step: 1492/527, loss: 0.0134903434664011
2023-01-21 12:04:52.788236: step: 1496/527, loss: 0.003542137099429965
2023-01-21 12:04:53.883530: step: 1500/527, loss: 0.0004935264587402344
2023-01-21 12:04:54.997315: step: 1504/527, loss: 0.0002878189261537045
2023-01-21 12:04:56.111825: step: 1508/527, loss: 0.025111103430390358
2023-01-21 12:04:57.214833: step: 1512/527, loss: 0.0046749114990234375
2023-01-21 12:04:58.312080: step: 1516/527, loss: 0.00018749237642623484
2023-01-21 12:04:59.419196: step: 1520/527, loss: 0.04882211610674858
2023-01-21 12:05:00.528324: step: 1524/527, loss: 0.00012826919555664062
2023-01-21 12:05:01.644960: step: 1528/527, loss: 0.005302906036376953
2023-01-21 12:05:02.800043: step: 1532/527, loss: 0.01220398023724556
2023-01-21 12:05:03.941293: step: 1536/527, loss: 0.07104186713695526
2023-01-21 12:05:05.048375: step: 1540/527, loss: 0.006728076841682196
2023-01-21 12:05:06.147571: step: 1544/527, loss: 0.016280079260468483
2023-01-21 12:05:07.260781: step: 1548/527, loss: 0.0017538070678710938
2023-01-21 12:05:08.355317: step: 1552/527, loss: 0.0673857256770134
2023-01-21 12:05:09.487179: step: 1556/527, loss: 0.03412008285522461
2023-01-21 12:05:10.572731: step: 1560/527, loss: 0.0036411285400390625
2023-01-21 12:05:11.669883: step: 1564/527, loss: 0.00609746016561985
2023-01-21 12:05:12.769758: step: 1568/527, loss: 0.01572742499411106
2023-01-21 12:05:13.911454: step: 1572/527, loss: 1.33514404296875e-05
2023-01-21 12:05:15.033671: step: 1576/527, loss: 0.01256332453340292
2023-01-21 12:05:16.164767: step: 1580/527, loss: 0.06875848770141602
2023-01-21 12:05:17.275407: step: 1584/527, loss: 0.024183178320527077
2023-01-21 12:05:18.374358: step: 1588/527, loss: 0.0007033824804238975
2023-01-21 12:05:19.488554: step: 1592/527, loss: 0.014115714468061924
2023-01-21 12:05:20.615343: step: 1596/527, loss: 0.007302188780158758
2023-01-21 12:05:21.713738: step: 1600/527, loss: 4.482269287109375e-05
2023-01-21 12:05:22.825627: step: 1604/527, loss: 0.0024428367614746094
2023-01-21 12:05:23.935193: step: 1608/527, loss: 0.00018796921358443797
2023-01-21 12:05:25.060952: step: 1612/527, loss: 0.0011894701747223735
2023-01-21 12:05:26.204103: step: 1616/527, loss: 0.0028945922385901213
2023-01-21 12:05:27.338156: step: 1620/527, loss: 0.0006848335615359247
2023-01-21 12:05:28.446550: step: 1624/527, loss: 0.005323886871337891
2023-01-21 12:05:29.556485: step: 1628/527, loss: 0.01225967425853014
2023-01-21 12:05:30.668832: step: 1632/527, loss: 0.0009273529867641628
2023-01-21 12:05:31.788041: step: 1636/527, loss: 0.002427005907520652
2023-01-21 12:05:32.902282: step: 1640/527, loss: 0.00592384347692132
2023-01-21 12:05:33.988756: step: 1644/527, loss: 0.003995371051132679
2023-01-21 12:05:35.118049: step: 1648/527, loss: 4.882812572759576e-05
2023-01-21 12:05:36.221051: step: 1652/527, loss: 0.0020814896561205387
2023-01-21 12:05:37.330978: step: 1656/527, loss: 0.0005338669288903475
2023-01-21 12:05:38.468166: step: 1660/527, loss: 0.025425149127840996
2023-01-21 12:05:39.572773: step: 1664/527, loss: 0.0006948232767172158
2023-01-21 12:05:40.706897: step: 1668/527, loss: 0.015040207654237747
2023-01-21 12:05:41.807396: step: 1672/527, loss: 0.05722980573773384
2023-01-21 12:05:42.936034: step: 1676/527, loss: 0.013215922750532627
2023-01-21 12:05:44.086527: step: 1680/527, loss: 0.0012277603382244706
2023-01-21 12:05:45.198153: step: 1684/527, loss: 0.009001731872558594
2023-01-21 12:05:46.313433: step: 1688/527, loss: 0.0006127357482910156
2023-01-21 12:05:47.434935: step: 1692/527, loss: 0.00022125244140625
2023-01-21 12:05:48.552948: step: 1696/527, loss: 0.02458496019244194
2023-01-21 12:05:49.640105: step: 1700/527, loss: 0.0021808624733239412
2023-01-21 12:05:50.752390: step: 1704/527, loss: 0.0029230117797851562
2023-01-21 12:05:51.883820: step: 1708/527, loss: 0.002620697021484375
2023-01-21 12:05:53.004956: step: 1712/527, loss: 0.00016393660916946828
2023-01-21 12:05:54.134387: step: 1716/527, loss: 0.0006011963123455644
2023-01-21 12:05:55.250573: step: 1720/527, loss: 0.005690097808837891
2023-01-21 12:05:56.353806: step: 1724/527, loss: 0.00026326178340241313
2023-01-21 12:05:57.483354: step: 1728/527, loss: 0.02092151716351509
2023-01-21 12:05:58.615348: step: 1732/527, loss: 0.011891174130141735
2023-01-21 12:05:59.725310: step: 1736/527, loss: 0.007113742642104626
2023-01-21 12:06:00.901521: step: 1740/527, loss: 0.04443969950079918
2023-01-21 12:06:01.977797: step: 1744/527, loss: 0.03164253383874893
2023-01-21 12:06:03.134901: step: 1748/527, loss: 0.020047379657626152
2023-01-21 12:06:04.249447: step: 1752/527, loss: 0.0007253646617755294
2023-01-21 12:06:05.379500: step: 1756/527, loss: 0.030295561999082565
2023-01-21 12:06:06.482430: step: 1760/527, loss: 0.013912391848862171
2023-01-21 12:06:07.604248: step: 1764/527, loss: 0.005179405212402344
2023-01-21 12:06:08.749397: step: 1768/527, loss: 0.002670574001967907
2023-01-21 12:06:09.867973: step: 1772/527, loss: 0.0011358261108398438
2023-01-21 12:06:10.987311: step: 1776/527, loss: 0.040836237370967865
2023-01-21 12:06:12.071872: step: 1780/527, loss: 0.00030236245947889984
2023-01-21 12:06:13.196788: step: 1784/527, loss: 0.030338477343320847
2023-01-21 12:06:14.296612: step: 1788/527, loss: 0.017296411097049713
2023-01-21 12:06:15.412749: step: 1792/527, loss: 0.0016822816105559468
2023-01-21 12:06:16.556920: step: 1796/527, loss: 9.174347360385582e-05
2023-01-21 12:06:17.699941: step: 1800/527, loss: 0.012796211056411266
2023-01-21 12:06:18.828468: step: 1804/527, loss: 0.0012359619140625
2023-01-21 12:06:19.934380: step: 1808/527, loss: 0.010367202572524548
2023-01-21 12:06:21.087399: step: 1812/527, loss: 0.02257404290139675
2023-01-21 12:06:22.238014: step: 1816/527, loss: 0.07814665138721466
2023-01-21 12:06:23.398730: step: 1820/527, loss: 0.021368028596043587
2023-01-21 12:06:24.532764: step: 1824/527, loss: 0.00029582978459075093
2023-01-21 12:06:25.623470: step: 1828/527, loss: 2.2554399038199335e-05
2023-01-21 12:06:26.747651: step: 1832/527, loss: 0.0027396203950047493
2023-01-21 12:06:27.848278: step: 1836/527, loss: 0.0002101898135151714
2023-01-21 12:06:29.004357: step: 1840/527, loss: 0.007755280006676912
2023-01-21 12:06:30.095615: step: 1844/527, loss: 0.055211640894412994
2023-01-21 12:06:31.177741: step: 1848/527, loss: 0.0020311353728175163
2023-01-21 12:06:32.277099: step: 1852/527, loss: 0.04140214994549751
2023-01-21 12:06:33.403906: step: 1856/527, loss: 7.877349707996473e-05
2023-01-21 12:06:34.527786: step: 1860/527, loss: 0.0020973205100744963
2023-01-21 12:06:35.642732: step: 1864/527, loss: 0.002594089601188898
2023-01-21 12:06:36.765271: step: 1868/527, loss: 0.0022974968887865543
2023-01-21 12:06:37.879179: step: 1872/527, loss: 6.966591172385961e-05
2023-01-21 12:06:38.973573: step: 1876/527, loss: 0.06346073746681213
2023-01-21 12:06:40.078956: step: 1880/527, loss: 0.020610475912690163
2023-01-21 12:06:41.208581: step: 1884/527, loss: 0.0008833885076455772
2023-01-21 12:06:42.330787: step: 1888/527, loss: 0.02894468419253826
2023-01-21 12:06:43.428603: step: 1892/527, loss: 0.012651252560317516
2023-01-21 12:06:44.561238: step: 1896/527, loss: 0.03854770585894585
2023-01-21 12:06:45.657339: step: 1900/527, loss: 0.0008977890247479081
2023-01-21 12:06:46.749099: step: 1904/527, loss: 0.0003326415899209678
2023-01-21 12:06:47.856647: step: 1908/527, loss: 0.0006975174183025956
2023-01-21 12:06:48.973344: step: 1912/527, loss: 0.00723800715059042
2023-01-21 12:06:50.109767: step: 1916/527, loss: 0.5513867139816284
2023-01-21 12:06:51.226013: step: 1920/527, loss: 0.14393559098243713
2023-01-21 12:06:52.340264: step: 1924/527, loss: 0.009127616882324219
2023-01-21 12:06:53.441778: step: 1928/527, loss: 0.000559902167879045
2023-01-21 12:06:54.549810: step: 1932/527, loss: 0.01709294319152832
2023-01-21 12:06:55.659260: step: 1936/527, loss: 0.007925224490463734
2023-01-21 12:06:56.769766: step: 1940/527, loss: 0.01940007321536541
2023-01-21 12:06:57.875849: step: 1944/527, loss: 0.004201698116958141
2023-01-21 12:06:59.012873: step: 1948/527, loss: 0.001144838286563754
2023-01-21 12:07:00.129326: step: 1952/527, loss: 0.003654670901596546
2023-01-21 12:07:01.240908: step: 1956/527, loss: 0.02370905876159668
2023-01-21 12:07:02.369412: step: 1960/527, loss: 0.001982021378353238
2023-01-21 12:07:03.498318: step: 1964/527, loss: 0.02887563779950142
2023-01-21 12:07:04.625813: step: 1968/527, loss: 0.0001663207949604839
2023-01-21 12:07:05.746621: step: 1972/527, loss: 0.0005970001220703125
2023-01-21 12:07:06.870162: step: 1976/527, loss: 0.0275744441896677
2023-01-21 12:07:07.956966: step: 1980/527, loss: 8.18252592580393e-05
2023-01-21 12:07:09.057201: step: 1984/527, loss: 0.004429054446518421
2023-01-21 12:07:10.177968: step: 1988/527, loss: 0.00226780166849494
2023-01-21 12:07:11.300279: step: 1992/527, loss: 0.0028839111328125
2023-01-21 12:07:12.385516: step: 1996/527, loss: 0.015825461596250534
2023-01-21 12:07:13.474528: step: 2000/527, loss: 0.00030078887357376516
2023-01-21 12:07:14.575515: step: 2004/527, loss: 0.0005528450128622353
2023-01-21 12:07:15.666610: step: 2008/527, loss: 0.0010038375621661544
2023-01-21 12:07:16.773838: step: 2012/527, loss: 0.011748886667191982
2023-01-21 12:07:17.874616: step: 2016/527, loss: 0.007998275570571423
2023-01-21 12:07:18.973406: step: 2020/527, loss: 0.04023732990026474
2023-01-21 12:07:20.060842: step: 2024/527, loss: 0.0004654884396586567
2023-01-21 12:07:21.206560: step: 2028/527, loss: 0.09385576099157333
2023-01-21 12:07:22.319392: step: 2032/527, loss: 0.0006269931909628212
2023-01-21 12:07:23.423067: step: 2036/527, loss: 0.014314842410385609
2023-01-21 12:07:24.566573: step: 2040/527, loss: 0.005443572998046875
2023-01-21 12:07:25.673122: step: 2044/527, loss: 0.023126699030399323
2023-01-21 12:07:26.750012: step: 2048/527, loss: 0.0008742331992834806
2023-01-21 12:07:27.881256: step: 2052/527, loss: 0.00200653076171875
2023-01-21 12:07:28.991835: step: 2056/527, loss: 0.01771564409136772
2023-01-21 12:07:30.091684: step: 2060/527, loss: 0.0006164073711261153
2023-01-21 12:07:31.210615: step: 2064/527, loss: 0.02081775665283203
2023-01-21 12:07:32.352688: step: 2068/527, loss: 0.00015945434279274195
2023-01-21 12:07:33.454517: step: 2072/527, loss: 0.004548454191535711
2023-01-21 12:07:34.574244: step: 2076/527, loss: 0.004913234617561102
2023-01-21 12:07:35.671603: step: 2080/527, loss: 0.05552806705236435
2023-01-21 12:07:36.788568: step: 2084/527, loss: 0.005438041873276234
2023-01-21 12:07:37.899935: step: 2088/527, loss: 0.025446033105254173
2023-01-21 12:07:39.018578: step: 2092/527, loss: 0.024950265884399414
2023-01-21 12:07:40.112959: step: 2096/527, loss: 0.28835418820381165
2023-01-21 12:07:41.241606: step: 2100/527, loss: 0.004086685366928577
2023-01-21 12:07:42.367072: step: 2104/527, loss: 0.07227544486522675
2023-01-21 12:07:43.459336: step: 2108/527, loss: 0.024315834045410156
==================================================
Loss: 0.019
--------------------
Dev: {'event': {'p': 0.5776892430278885, 'r': 0.7723035952063915, 'f1': 0.6609686609686609}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20}
Test: {'event': {'p': 0.6184327693677649, 'r': 0.7937142857142857, 'f1': 0.6951951951951952}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20}
Chinese: {'event': {'p': 0.5697674418604651, 'r': 0.9074074074074074, 'f1': 0.7}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20}
Korean: {'event': {'p': 0.5740740740740741, 'r': 0.49206349206349204, 'f1': 0.5299145299145299}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20}
Russian: {'event': {'p': 0.425, 'r': 0.4722222222222222, 'f1': 0.4473684210526316}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 12:08:24.819603: step: 4/527, loss: 0.0003424167516641319
2023-01-21 12:08:25.942309: step: 8/527, loss: 0.014571189880371094
2023-01-21 12:08:27.091286: step: 12/527, loss: 0.03218331187963486
2023-01-21 12:08:28.217812: step: 16/527, loss: 0.00018224716768600047
2023-01-21 12:08:29.358520: step: 20/527, loss: 0.029027651995420456
2023-01-21 12:08:30.484843: step: 24/527, loss: 0.0012723446125164628
2023-01-21 12:08:31.619479: step: 28/527, loss: 0.004144716076552868
2023-01-21 12:08:32.753314: step: 32/527, loss: 0.0002601623418740928
2023-01-21 12:08:33.871957: step: 36/527, loss: 0.018167970702052116
2023-01-21 12:08:34.973923: step: 40/527, loss: 0.0001473009615438059
2023-01-21 12:08:36.059685: step: 44/527, loss: 0.005300998687744141
2023-01-21 12:08:37.184729: step: 48/527, loss: 0.002562999725341797
2023-01-21 12:08:38.276770: step: 52/527, loss: 1.1634827387752011e-05
2023-01-21 12:08:39.391554: step: 56/527, loss: 0.00010318756540073082
2023-01-21 12:08:40.528888: step: 60/527, loss: 0.0003001093864440918
2023-01-21 12:08:41.641901: step: 64/527, loss: 8.59260544530116e-05
2023-01-21 12:08:42.739586: step: 68/527, loss: 0.0017410516738891602
2023-01-21 12:08:43.844637: step: 72/527, loss: 0.009443092159926891
2023-01-21 12:08:44.963954: step: 76/527, loss: 1.9073486612342094e-07
2023-01-21 12:08:46.049371: step: 80/527, loss: 0.0007999420049600303
2023-01-21 12:08:47.190780: step: 84/527, loss: 0.0010536194313317537
2023-01-21 12:08:48.329663: step: 88/527, loss: 0.047819897532463074
2023-01-21 12:08:49.445492: step: 92/527, loss: -2.574920654296875e-05
2023-01-21 12:08:50.545602: step: 96/527, loss: 0.0007807731744833291
2023-01-21 12:08:51.659747: step: 100/527, loss: 0.03480224683880806
2023-01-21 12:08:52.794922: step: 104/527, loss: 0.0004161834658589214
2023-01-21 12:08:53.902806: step: 108/527, loss: 1.869201696536038e-05
2023-01-21 12:08:55.055399: step: 112/527, loss: 0.042073819786310196
2023-01-21 12:08:56.173270: step: 116/527, loss: 0.00010776519775390625
2023-01-21 12:08:57.285576: step: 120/527, loss: 0.0020597458351403475
2023-01-21 12:08:58.400405: step: 124/527, loss: 3.8146959013829473e-07
2023-01-21 12:08:59.495210: step: 128/527, loss: 0.003101825714111328
2023-01-21 12:09:00.614724: step: 132/527, loss: 0.00022038817405700684
2023-01-21 12:09:01.712774: step: 136/527, loss: 0.035575199872255325
2023-01-21 12:09:02.819385: step: 140/527, loss: 1.2683868590102065e-05
2023-01-21 12:09:03.903224: step: 144/527, loss: 0.023530179634690285
2023-01-21 12:09:05.033869: step: 148/527, loss: 0.06080367788672447
2023-01-21 12:09:06.136981: step: 152/527, loss: 0.01611633412539959
2023-01-21 12:09:07.243885: step: 156/527, loss: 0.000623512314632535
2023-01-21 12:09:08.372560: step: 160/527, loss: 0.01434326171875
2023-01-21 12:09:09.476749: step: 164/527, loss: 0.014948463067412376
2023-01-21 12:09:10.589536: step: 168/527, loss: 0.01594219170510769
2023-01-21 12:09:11.716008: step: 172/527, loss: 0.0072536468505859375
2023-01-21 12:09:12.823830: step: 176/527, loss: 0.037351515144109726
2023-01-21 12:09:13.920735: step: 180/527, loss: 0.006028270814567804
2023-01-21 12:09:14.980912: step: 184/527, loss: 0.012895393185317516
2023-01-21 12:09:16.072124: step: 188/527, loss: 0.3417227864265442
2023-01-21 12:09:17.181946: step: 192/527, loss: 0.020996762439608574
2023-01-21 12:09:18.292812: step: 196/527, loss: 0.0029195784591138363
2023-01-21 12:09:19.431891: step: 200/527, loss: 0.0004861831839662045
2023-01-21 12:09:20.541706: step: 204/527, loss: 0.03167114406824112
2023-01-21 12:09:21.631712: step: 208/527, loss: 0.013806914910674095
2023-01-21 12:09:22.755789: step: 212/527, loss: 0.022811222821474075
2023-01-21 12:09:23.851497: step: 216/527, loss: 0.012236405164003372
2023-01-21 12:09:24.937043: step: 220/527, loss: 0.0006712913746014237
2023-01-21 12:09:26.041807: step: 224/527, loss: 0.0008067131275311112
2023-01-21 12:09:27.179215: step: 228/527, loss: 0.00014848708815407008
2023-01-21 12:09:28.285738: step: 232/527, loss: 0.00014181138249114156
2023-01-21 12:09:29.393500: step: 236/527, loss: 9.31739850784652e-05
2023-01-21 12:09:30.518421: step: 240/527, loss: 0.0009866715408861637
2023-01-21 12:09:31.622557: step: 244/527, loss: 0.03864326328039169
2023-01-21 12:09:32.747817: step: 248/527, loss: 0.0003999710315838456
2023-01-21 12:09:33.871008: step: 252/527, loss: 5.9127810345671605e-06
2023-01-21 12:09:34.991871: step: 256/527, loss: 0.005319785792380571
2023-01-21 12:09:36.093042: step: 260/527, loss: 0.021911241114139557
2023-01-21 12:09:37.210384: step: 264/527, loss: 0.00759735144674778
2023-01-21 12:09:38.336446: step: 268/527, loss: 0.02007455937564373
2023-01-21 12:09:39.434381: step: 272/527, loss: 0.0002645492786541581
2023-01-21 12:09:40.546978: step: 276/527, loss: 0.005757856648415327
2023-01-21 12:09:41.673244: step: 280/527, loss: 0.012510204687714577
2023-01-21 12:09:42.797925: step: 284/527, loss: 8.37326078908518e-05
2023-01-21 12:09:43.895767: step: 288/527, loss: 0.010239219292998314
2023-01-21 12:09:44.980508: step: 292/527, loss: 0.005147362127900124
2023-01-21 12:09:46.115602: step: 296/527, loss: 2.021789623540826e-05
2023-01-21 12:09:47.199040: step: 300/527, loss: 0.00013427734666038305
2023-01-21 12:09:48.309769: step: 304/527, loss: 0.008954429998993874
2023-01-21 12:09:49.435730: step: 308/527, loss: 0.01438894309103489
2023-01-21 12:09:50.565485: step: 312/527, loss: 0.03666668012738228
2023-01-21 12:09:51.665986: step: 316/527, loss: 0.0001791954127838835
2023-01-21 12:09:52.763523: step: 320/527, loss: 0.019196892157197
2023-01-21 12:09:53.848554: step: 324/527, loss: 0.0019220353569835424
2023-01-21 12:09:54.974512: step: 328/527, loss: 0.009200858883559704
2023-01-21 12:09:56.100183: step: 332/527, loss: 0.03407297283411026
2023-01-21 12:09:57.203814: step: 336/527, loss: 0.002361297607421875
2023-01-21 12:09:58.316412: step: 340/527, loss: 0.02950306050479412
2023-01-21 12:09:59.450476: step: 344/527, loss: 0.06600666046142578
2023-01-21 12:10:00.568310: step: 348/527, loss: 3.6048892070539296e-05
2023-01-21 12:10:01.699082: step: 352/527, loss: 0.00510482769459486
2023-01-21 12:10:02.801698: step: 356/527, loss: 0.0010512591106817126
2023-01-21 12:10:03.905710: step: 360/527, loss: 0.002140045166015625
2023-01-21 12:10:05.023259: step: 364/527, loss: 0.0003147125244140625
2023-01-21 12:10:06.148850: step: 368/527, loss: 0.00014820098294876516
2023-01-21 12:10:07.256301: step: 372/527, loss: 0.0005359649658203125
2023-01-21 12:10:08.350023: step: 376/527, loss: 0.00019359588623046875
2023-01-21 12:10:09.459350: step: 380/527, loss: 0.20133666694164276
2023-01-21 12:10:10.555629: step: 384/527, loss: 0.0010221959091722965
2023-01-21 12:10:11.656119: step: 388/527, loss: 0.003403282258659601
2023-01-21 12:10:12.794252: step: 392/527, loss: 0.011954116635024548
2023-01-21 12:10:13.895894: step: 396/527, loss: 0.03757696598768234
2023-01-21 12:10:15.000392: step: 400/527, loss: 0.002617359161376953
2023-01-21 12:10:16.117951: step: 404/527, loss: 0.1376585066318512
2023-01-21 12:10:17.259640: step: 408/527, loss: 0.0005266189691610634
2023-01-21 12:10:18.374727: step: 412/527, loss: 0.005866050720214844
2023-01-21 12:10:19.471099: step: 416/527, loss: 6.017684791004285e-05
2023-01-21 12:10:20.564247: step: 420/527, loss: 0.0006245553377084434
2023-01-21 12:10:21.659094: step: 424/527, loss: 0.03697452321648598
2023-01-21 12:10:22.778260: step: 428/527, loss: 0.006893730256706476
2023-01-21 12:10:23.874171: step: 432/527, loss: 0.030045127496123314
2023-01-21 12:10:24.999574: step: 436/527, loss: 0.0015117645962163806
2023-01-21 12:10:26.131296: step: 440/527, loss: 0.0017983437282964587
2023-01-21 12:10:27.246972: step: 444/527, loss: 0.0009522438631393015
2023-01-21 12:10:28.353478: step: 448/527, loss: 0.03591423109173775
2023-01-21 12:10:29.488523: step: 452/527, loss: 0.03116321563720703
2023-01-21 12:10:30.624015: step: 456/527, loss: 0.030733203515410423
2023-01-21 12:10:31.713142: step: 460/527, loss: 0.018243025988340378
2023-01-21 12:10:32.830802: step: 464/527, loss: 0.06861267983913422
2023-01-21 12:10:33.951873: step: 468/527, loss: 0.00024852753267623484
2023-01-21 12:10:35.119361: step: 472/527, loss: 0.5412971377372742
2023-01-21 12:10:36.225933: step: 476/527, loss: 0.025266362354159355
2023-01-21 12:10:37.359575: step: 480/527, loss: 0.0005380630609579384
2023-01-21 12:10:38.470926: step: 484/527, loss: 0.01474151574075222
2023-01-21 12:10:39.604931: step: 488/527, loss: 0.0031697272788733244
2023-01-21 12:10:40.746915: step: 492/527, loss: 0.023918725550174713
2023-01-21 12:10:41.909906: step: 496/527, loss: 0.02289428748190403
2023-01-21 12:10:43.006127: step: 500/527, loss: 0.024773407727479935
2023-01-21 12:10:44.144729: step: 504/527, loss: 0.0009824753506109118
2023-01-21 12:10:45.275002: step: 508/527, loss: 0.012175941839814186
2023-01-21 12:10:46.385897: step: 512/527, loss: 0.022690391167998314
2023-01-21 12:10:47.487438: step: 516/527, loss: 0.005814170930534601
2023-01-21 12:10:48.595086: step: 520/527, loss: 0.018480967730283737
2023-01-21 12:10:49.726360: step: 524/527, loss: 0.00024271011352539062
2023-01-21 12:10:50.848559: step: 528/527, loss: 0.00148601527325809
2023-01-21 12:10:51.957470: step: 532/527, loss: 0.0037738799583166838
2023-01-21 12:10:53.085891: step: 536/527, loss: 0.005963135045021772
2023-01-21 12:10:54.213853: step: 540/527, loss: 0.025493431836366653
2023-01-21 12:10:55.330965: step: 544/527, loss: 1.4925002687959932e-05
2023-01-21 12:10:56.444422: step: 548/527, loss: 0.03330955654382706
2023-01-21 12:10:57.568812: step: 552/527, loss: 0.005772400181740522
2023-01-21 12:10:58.696768: step: 556/527, loss: 0.02334127388894558
2023-01-21 12:10:59.798668: step: 560/527, loss: 0.00025272369384765625
2023-01-21 12:11:00.934194: step: 564/527, loss: 0.0021568299271166325
2023-01-21 12:11:02.042097: step: 568/527, loss: 0.0005167007329873741
2023-01-21 12:11:03.152762: step: 572/527, loss: 0.07779093086719513
2023-01-21 12:11:04.228678: step: 576/527, loss: 2.4652481442899443e-05
2023-01-21 12:11:05.350435: step: 580/527, loss: 0.00013809204392600805
2023-01-21 12:11:06.455255: step: 584/527, loss: 0.0014862060779705644
2023-01-21 12:11:07.530698: step: 588/527, loss: 7.724761962890625e-05
2023-01-21 12:11:08.643860: step: 592/527, loss: 0.00010795592970680445
2023-01-21 12:11:09.773065: step: 596/527, loss: 0.0028567316476255655
2023-01-21 12:11:10.901047: step: 600/527, loss: 0.13070030510425568
2023-01-21 12:11:12.045564: step: 604/527, loss: 0.0011511803604662418
2023-01-21 12:11:13.174746: step: 608/527, loss: 0.024424076080322266
2023-01-21 12:11:14.263995: step: 612/527, loss: 0.014412213116884232
2023-01-21 12:11:15.393970: step: 616/527, loss: 1.983642505365424e-05
2023-01-21 12:11:16.537713: step: 620/527, loss: 0.03953971713781357
2023-01-21 12:11:17.655616: step: 624/527, loss: 0.00085024832515046
2023-01-21 12:11:18.791271: step: 628/527, loss: 0.00047512055607512593
2023-01-21 12:11:19.892115: step: 632/527, loss: 0.0013311386574059725
2023-01-21 12:11:20.994694: step: 636/527, loss: 0.00192003243137151
2023-01-21 12:11:22.094203: step: 640/527, loss: 0.015132216736674309
2023-01-21 12:11:23.188073: step: 644/527, loss: 0.02239055559039116
2023-01-21 12:11:24.279191: step: 648/527, loss: 0.005105114076286554
2023-01-21 12:11:25.397054: step: 652/527, loss: 0.0017021656967699528
2023-01-21 12:11:26.523752: step: 656/527, loss: 0.042104244232177734
2023-01-21 12:11:27.683730: step: 660/527, loss: 0.0001312255917582661
2023-01-21 12:11:28.822908: step: 664/527, loss: 0.021743202582001686
2023-01-21 12:11:29.942422: step: 668/527, loss: 0.0001617431698832661
2023-01-21 12:11:31.031944: step: 672/527, loss: 0.021367549896240234
2023-01-21 12:11:32.149578: step: 676/527, loss: 0.009956836700439453
2023-01-21 12:11:33.277999: step: 680/527, loss: 0.029199600219726562
2023-01-21 12:11:34.399860: step: 684/527, loss: 0.017909621819853783
2023-01-21 12:11:35.516191: step: 688/527, loss: 0.003249645233154297
2023-01-21 12:11:36.631957: step: 692/527, loss: 0.02673787996172905
2023-01-21 12:11:37.737542: step: 696/527, loss: 9.74655122263357e-05
2023-01-21 12:11:38.818836: step: 700/527, loss: 0.0008546352037228644
2023-01-21 12:11:39.946660: step: 704/527, loss: 0.0018526078201830387
2023-01-21 12:11:41.050583: step: 708/527, loss: 0.006857538595795631
2023-01-21 12:11:42.221317: step: 712/527, loss: 0.03363924100995064
2023-01-21 12:11:43.359526: step: 716/527, loss: 0.00023956299992278218
2023-01-21 12:11:44.448548: step: 720/527, loss: 0.029749106615781784
2023-01-21 12:11:45.564825: step: 724/527, loss: 0.0023924827110022306
2023-01-21 12:11:46.719300: step: 728/527, loss: 0.027650736272335052
2023-01-21 12:11:47.832832: step: 732/527, loss: 1.7452239262638614e-05
2023-01-21 12:11:48.985728: step: 736/527, loss: 0.0034271241165697575
2023-01-21 12:11:50.086976: step: 740/527, loss: 0.004085445310920477
2023-01-21 12:11:51.211647: step: 744/527, loss: 0.003986025229096413
2023-01-21 12:11:52.305070: step: 748/527, loss: 6.341934204101562e-05
2023-01-21 12:11:53.384631: step: 752/527, loss: 0.004427909851074219
2023-01-21 12:11:54.494723: step: 756/527, loss: 0.023706817999482155
2023-01-21 12:11:55.587655: step: 760/527, loss: 0.0002662658516783267
2023-01-21 12:11:56.693614: step: 764/527, loss: 0.00347137451171875
2023-01-21 12:11:57.795688: step: 768/527, loss: 0.00011596680269576609
2023-01-21 12:11:58.869479: step: 772/527, loss: 0.00022678376990370452
2023-01-21 12:12:00.010764: step: 776/527, loss: 0.02955188788473606
2023-01-21 12:12:01.122563: step: 780/527, loss: 0.0014501571422442794
2023-01-21 12:12:02.227403: step: 784/527, loss: 0.00016574858454987407
2023-01-21 12:12:03.331476: step: 788/527, loss: 4.9591067181609105e-06
2023-01-21 12:12:04.441617: step: 792/527, loss: 0.00017614364332985133
2023-01-21 12:12:05.534643: step: 796/527, loss: 0.03747148439288139
2023-01-21 12:12:06.668308: step: 800/527, loss: 0.004490852355957031
2023-01-21 12:12:07.760136: step: 804/527, loss: 0.00029907224234193563
2023-01-21 12:12:08.849127: step: 808/527, loss: 0.00022411346435546875
2023-01-21 12:12:09.992921: step: 812/527, loss: 0.0029441833030432463
2023-01-21 12:12:11.095064: step: 816/527, loss: 0.00038137438241392374
2023-01-21 12:12:12.226821: step: 820/527, loss: 0.00714111328125
2023-01-21 12:12:13.355601: step: 824/527, loss: 0.024107933044433594
2023-01-21 12:12:14.479751: step: 828/527, loss: 0.0009182632202282548
2023-01-21 12:12:15.586162: step: 832/527, loss: 0.012457084842026234
2023-01-21 12:12:16.692942: step: 836/527, loss: 0.00029621124849654734
2023-01-21 12:12:17.808231: step: 840/527, loss: 0.009798049926757812
2023-01-21 12:12:18.915717: step: 844/527, loss: 0.0024993896950036287
2023-01-21 12:12:20.036939: step: 848/527, loss: 0.005990410223603249
2023-01-21 12:12:21.168473: step: 852/527, loss: 4.825591895496473e-05
2023-01-21 12:12:22.323062: step: 856/527, loss: 0.00346794119104743
2023-01-21 12:12:23.417849: step: 860/527, loss: 0.003063201904296875
2023-01-21 12:12:24.512135: step: 864/527, loss: 1.3446808225126006e-05
2023-01-21 12:12:25.617474: step: 868/527, loss: 0.0010539054637774825
2023-01-21 12:12:26.753590: step: 872/527, loss: 0.03994961082935333
2023-01-21 12:12:27.877884: step: 876/527, loss: 0.008144378662109375
2023-01-21 12:12:28.998724: step: 880/527, loss: 0.00452690152451396
2023-01-21 12:12:30.110749: step: 884/527, loss: 0.016897965222597122
2023-01-21 12:12:31.244809: step: 888/527, loss: 0.0018390655750408769
2023-01-21 12:12:32.353218: step: 892/527, loss: 0.3178071975708008
2023-01-21 12:12:33.448376: step: 896/527, loss: 0.1257917433977127
2023-01-21 12:12:34.561076: step: 900/527, loss: 0.012685203924775124
2023-01-21 12:12:35.666102: step: 904/527, loss: 0.050023604184389114
2023-01-21 12:12:36.761426: step: 908/527, loss: 0.00380287179723382
2023-01-21 12:12:37.866842: step: 912/527, loss: 4.062652442371473e-05
2023-01-21 12:12:39.004453: step: 916/527, loss: 0.014753341674804688
2023-01-21 12:12:40.137065: step: 920/527, loss: 0.054183200001716614
2023-01-21 12:12:41.258594: step: 924/527, loss: 0.002619075821712613
2023-01-21 12:12:42.365121: step: 928/527, loss: 0.0003921508905477822
2023-01-21 12:12:43.472330: step: 932/527, loss: 3.6191944673191756e-05
2023-01-21 12:12:44.552773: step: 936/527, loss: 0.00014381408982444555
2023-01-21 12:12:45.661981: step: 940/527, loss: 0.0014183043967932463
2023-01-21 12:12:46.765812: step: 944/527, loss: 0.007401895243674517
2023-01-21 12:12:47.887033: step: 948/527, loss: 8.37326078908518e-05
2023-01-21 12:12:48.997157: step: 952/527, loss: 6.0749054682673886e-05
2023-01-21 12:12:50.133046: step: 956/527, loss: 0.009656048379838467
2023-01-21 12:12:51.256964: step: 960/527, loss: 0.011701774783432484
2023-01-21 12:12:52.399168: step: 964/527, loss: 0.03260479122400284
2023-01-21 12:12:53.508033: step: 968/527, loss: 0.0006240844959393144
2023-01-21 12:12:54.633137: step: 972/527, loss: 5.440712266135961e-05
2023-01-21 12:12:55.780953: step: 976/527, loss: 0.0031078339088708162
2023-01-21 12:12:56.890194: step: 980/527, loss: 7.572174217784777e-05
2023-01-21 12:12:57.995275: step: 984/527, loss: 0.028738977387547493
2023-01-21 12:12:59.141152: step: 988/527, loss: 0.028688622638583183
2023-01-21 12:13:00.231348: step: 992/527, loss: 0.03795475885272026
2023-01-21 12:13:01.338939: step: 996/527, loss: 0.008068228140473366
2023-01-21 12:13:02.465746: step: 1000/527, loss: 0.010193252936005592
2023-01-21 12:13:03.572352: step: 1004/527, loss: 7.724762326688506e-06
2023-01-21 12:13:04.716920: step: 1008/527, loss: 0.014235115610063076
2023-01-21 12:13:05.814459: step: 1012/527, loss: 0.010060882195830345
2023-01-21 12:13:06.952357: step: 1016/527, loss: 0.00010719299461925402
2023-01-21 12:13:08.052340: step: 1020/527, loss: 0.024750327691435814
2023-01-21 12:13:09.200991: step: 1024/527, loss: 0.000522613525390625
2023-01-21 12:13:10.326724: step: 1028/527, loss: 0.0025341035798192024
2023-01-21 12:13:11.444643: step: 1032/527, loss: 0.017375566065311432
2023-01-21 12:13:12.540184: step: 1036/527, loss: 0.09704332053661346
2023-01-21 12:13:13.626087: step: 1040/527, loss: 0.029554367065429688
2023-01-21 12:13:14.706219: step: 1044/527, loss: 0.007841014303267002
2023-01-21 12:13:15.819274: step: 1048/527, loss: 0.006303596775978804
2023-01-21 12:13:16.951965: step: 1052/527, loss: 0.001470375107601285
2023-01-21 12:13:18.040830: step: 1056/527, loss: 0.02668152004480362
2023-01-21 12:13:19.154346: step: 1060/527, loss: 0.0004788398800883442
2023-01-21 12:13:20.294667: step: 1064/527, loss: 0.008269691839814186
2023-01-21 12:13:21.394333: step: 1068/527, loss: 0.00036458970862440765
2023-01-21 12:13:22.502788: step: 1072/527, loss: 0.0019134521717205644
2023-01-21 12:13:23.600245: step: 1076/527, loss: 0.03135652840137482
2023-01-21 12:13:24.723984: step: 1080/527, loss: 0.012887001037597656
2023-01-21 12:13:25.835360: step: 1084/527, loss: 0.03917905315756798
2023-01-21 12:13:26.927875: step: 1088/527, loss: 0.03343725576996803
2023-01-21 12:13:28.012581: step: 1092/527, loss: 0.0103880874812603
2023-01-21 12:13:29.148389: step: 1096/527, loss: 0.017304420471191406
2023-01-21 12:13:30.248198: step: 1100/527, loss: 0.017217446118593216
2023-01-21 12:13:31.380812: step: 1104/527, loss: 0.0013104439713060856
2023-01-21 12:13:32.481454: step: 1108/527, loss: 0.014308547601103783
2023-01-21 12:13:33.575116: step: 1112/527, loss: 0.00029239655123092234
2023-01-21 12:13:34.690247: step: 1116/527, loss: 0.0008804321405477822
2023-01-21 12:13:35.844768: step: 1120/527, loss: 0.008081817999482155
2023-01-21 12:13:36.972177: step: 1124/527, loss: 0.010495948605239391
2023-01-21 12:13:38.078713: step: 1128/527, loss: 0.007045269012451172
2023-01-21 12:13:39.170997: step: 1132/527, loss: 0.019229698926210403
2023-01-21 12:13:40.268538: step: 1136/527, loss: 0.04969377443194389
2023-01-21 12:13:41.372602: step: 1140/527, loss: 0.0026632307562977076
2023-01-21 12:13:42.489955: step: 1144/527, loss: 0.007572222035378218
2023-01-21 12:13:43.626689: step: 1148/527, loss: 0.030561067163944244
2023-01-21 12:13:44.735706: step: 1152/527, loss: 0.0003498077276162803
2023-01-21 12:13:45.871844: step: 1156/527, loss: 0.6666473150253296
2023-01-21 12:13:46.963564: step: 1160/527, loss: 0.0023136138916015625
2023-01-21 12:13:48.094136: step: 1164/527, loss: 0.26929742097854614
2023-01-21 12:13:49.220960: step: 1168/527, loss: 0.02358074113726616
2023-01-21 12:13:50.341638: step: 1172/527, loss: 0.0012399672996252775
2023-01-21 12:13:51.513552: step: 1176/527, loss: 0.042407989501953125
2023-01-21 12:13:52.632775: step: 1180/527, loss: 0.0009212493896484375
2023-01-21 12:13:53.694661: step: 1184/527, loss: 0.0014943123096600175
2023-01-21 12:13:54.840234: step: 1188/527, loss: 0.0006429672357626259
2023-01-21 12:13:55.966389: step: 1192/527, loss: 0.001831150148063898
2023-01-21 12:13:57.102100: step: 1196/527, loss: 0.012487792409956455
2023-01-21 12:13:58.197900: step: 1200/527, loss: 0.027876663953065872
2023-01-21 12:13:59.345915: step: 1204/527, loss: 0.0005977630498819053
2023-01-21 12:14:00.451323: step: 1208/527, loss: 0.004257583525031805
2023-01-21 12:14:01.590353: step: 1212/527, loss: 0.06620216369628906
2023-01-21 12:14:02.675674: step: 1216/527, loss: 0.00024099351139739156
2023-01-21 12:14:03.805195: step: 1220/527, loss: 0.03276100009679794
2023-01-21 12:14:04.892356: step: 1224/527, loss: 0.020774461328983307
2023-01-21 12:14:06.006564: step: 1228/527, loss: 0.001195716904476285
2023-01-21 12:14:07.115796: step: 1232/527, loss: 0.026764871552586555
2023-01-21 12:14:08.206521: step: 1236/527, loss: 0.00016899110050871968
2023-01-21 12:14:09.322502: step: 1240/527, loss: 0.033670518547296524
2023-01-21 12:14:10.434204: step: 1244/527, loss: 0.010154534131288528
2023-01-21 12:14:11.549193: step: 1248/527, loss: 0.009581947699189186
2023-01-21 12:14:12.686800: step: 1252/527, loss: 0.01013793982565403
2023-01-21 12:14:13.810066: step: 1256/527, loss: 0.0011025428539142013
2023-01-21 12:14:14.945049: step: 1260/527, loss: 0.008715820498764515
2023-01-21 12:14:16.049563: step: 1264/527, loss: 0.0013222694396972656
2023-01-21 12:14:17.220001: step: 1268/527, loss: 0.0010924339294433594
2023-01-21 12:14:18.312445: step: 1272/527, loss: 0.017766855657100677
2023-01-21 12:14:19.436460: step: 1276/527, loss: 0.001348304795101285
2023-01-21 12:14:20.554681: step: 1280/527, loss: 0.013893604278564453
2023-01-21 12:14:21.634959: step: 1284/527, loss: 0.07488150894641876
2023-01-21 12:14:22.756019: step: 1288/527, loss: 0.04465227201581001
2023-01-21 12:14:23.865559: step: 1292/527, loss: 0.0003437042178120464
2023-01-21 12:14:25.004973: step: 1296/527, loss: 0.01599426381289959
2023-01-21 12:14:26.124433: step: 1300/527, loss: 0.0031141280196607113
2023-01-21 12:14:27.216449: step: 1304/527, loss: 6.532669067382812e-05
2023-01-21 12:14:28.318766: step: 1308/527, loss: 0.00902261771261692
2023-01-21 12:14:29.412040: step: 1312/527, loss: 0.005781936924904585
2023-01-21 12:14:30.530045: step: 1316/527, loss: 0.00010032653517555445
2023-01-21 12:14:31.695086: step: 1320/527, loss: 0.010472106747329235
2023-01-21 12:14:32.797199: step: 1324/527, loss: 0.008134746924042702
2023-01-21 12:14:33.903364: step: 1328/527, loss: 0.007324791047722101
2023-01-21 12:14:35.007261: step: 1332/527, loss: 0.027659133076667786
2023-01-21 12:14:36.114532: step: 1336/527, loss: 0.0016972542507573962
2023-01-21 12:14:37.225427: step: 1340/527, loss: 0.32235851883888245
2023-01-21 12:14:38.321772: step: 1344/527, loss: 0.031434059143066406
2023-01-21 12:14:39.433661: step: 1348/527, loss: 0.0015349150635302067
2023-01-21 12:14:40.566579: step: 1352/527, loss: 0.001064300537109375
2023-01-21 12:14:41.675146: step: 1356/527, loss: 0.005130887031555176
2023-01-21 12:14:42.813274: step: 1360/527, loss: 0.0289827361702919
2023-01-21 12:14:43.966743: step: 1364/527, loss: 0.00982291717082262
2023-01-21 12:14:45.074620: step: 1368/527, loss: 0.016693115234375
2023-01-21 12:14:46.185104: step: 1372/527, loss: 0.01905345916748047
2023-01-21 12:14:47.321857: step: 1376/527, loss: 0.04285154491662979
2023-01-21 12:14:48.446732: step: 1380/527, loss: 0.0007675171364098787
2023-01-21 12:14:49.561354: step: 1384/527, loss: 0.0008178711286745965
2023-01-21 12:14:50.667185: step: 1388/527, loss: 0.0005775452009402215
2023-01-21 12:14:51.767932: step: 1392/527, loss: 2.3508073354605585e-05
2023-01-21 12:14:52.880322: step: 1396/527, loss: 0.004079436883330345
2023-01-21 12:14:53.997018: step: 1400/527, loss: 0.004896759986877441
2023-01-21 12:14:55.125513: step: 1404/527, loss: 0.013026141561567783
2023-01-21 12:14:56.256321: step: 1408/527, loss: 0.03258190304040909
2023-01-21 12:14:57.361431: step: 1412/527, loss: 0.023769784718751907
2023-01-21 12:14:58.479655: step: 1416/527, loss: 0.0005627929931506515
2023-01-21 12:14:59.602168: step: 1420/527, loss: 0.004569435026496649
2023-01-21 12:15:00.737096: step: 1424/527, loss: 0.0072904592379927635
2023-01-21 12:15:01.846424: step: 1428/527, loss: 0.0033453465439379215
2023-01-21 12:15:02.945927: step: 1432/527, loss: 0.0017184257740154862
2023-01-21 12:15:04.056942: step: 1436/527, loss: 0.02543344348669052
2023-01-21 12:15:05.152797: step: 1440/527, loss: 0.0004535675107035786
2023-01-21 12:15:06.258867: step: 1444/527, loss: 0.0018549920059740543
2023-01-21 12:15:07.411993: step: 1448/527, loss: 0.014914131723344326
2023-01-21 12:15:08.524284: step: 1452/527, loss: 0.0006571770063601434
2023-01-21 12:15:09.630742: step: 1456/527, loss: 4.053860902786255e-05
2023-01-21 12:15:10.752931: step: 1460/527, loss: 0.02966766431927681
2023-01-21 12:15:11.871195: step: 1464/527, loss: 9.088516526389867e-05
2023-01-21 12:15:13.062519: step: 1468/527, loss: 0.01150426920503378
2023-01-21 12:15:14.161357: step: 1472/527, loss: 0.8885663747787476
2023-01-21 12:15:15.249374: step: 1476/527, loss: 0.002996444934979081
2023-01-21 12:15:16.370822: step: 1480/527, loss: 0.00023703573970124125
2023-01-21 12:15:17.503222: step: 1484/527, loss: 0.002443981356918812
2023-01-21 12:15:18.601433: step: 1488/527, loss: 1.4495850336970761e-05
2023-01-21 12:15:19.706527: step: 1492/527, loss: 0.0004406929074320942
2023-01-21 12:15:20.833268: step: 1496/527, loss: 0.03025798872113228
2023-01-21 12:15:21.937556: step: 1500/527, loss: 0.0012924193870276213
2023-01-21 12:15:23.013781: step: 1504/527, loss: 0.2954823970794678
2023-01-21 12:15:24.108731: step: 1508/527, loss: 0.0004878044128417969
2023-01-21 12:15:25.202350: step: 1512/527, loss: 0.004134750459343195
2023-01-21 12:15:26.316809: step: 1516/527, loss: 0.20269259810447693
2023-01-21 12:15:27.440817: step: 1520/527, loss: 6.446838960982859e-05
2023-01-21 12:15:28.547650: step: 1524/527, loss: 3.25202927342616e-05
2023-01-21 12:15:29.683725: step: 1528/527, loss: 0.031868983060121536
2023-01-21 12:15:30.763325: step: 1532/527, loss: 0.0007028579711914062
2023-01-21 12:15:31.849496: step: 1536/527, loss: 0.00023546218289993703
2023-01-21 12:15:32.962017: step: 1540/527, loss: 0.005890607833862305
2023-01-21 12:15:34.059665: step: 1544/527, loss: 0.0072532654739916325
2023-01-21 12:15:35.175371: step: 1548/527, loss: 0.006248283665627241
2023-01-21 12:15:36.312353: step: 1552/527, loss: 0.0033077241387218237
2023-01-21 12:15:37.447974: step: 1556/527, loss: 0.0021356106735765934
2023-01-21 12:15:38.540808: step: 1560/527, loss: 0.09970169514417648
2023-01-21 12:15:39.669417: step: 1564/527, loss: 0.0006385803571902215
2023-01-21 12:15:40.791313: step: 1568/527, loss: 0.04650163650512695
2023-01-21 12:15:41.897819: step: 1572/527, loss: 0.0005164146423339844
2023-01-21 12:15:42.979307: step: 1576/527, loss: 0.007397461216896772
2023-01-21 12:15:44.097383: step: 1580/527, loss: 6.008148375258315e-06
2023-01-21 12:15:45.207355: step: 1584/527, loss: 0.0014300346374511719
2023-01-21 12:15:46.329445: step: 1588/527, loss: 0.02004106156527996
2023-01-21 12:15:47.433857: step: 1592/527, loss: 0.0013718605041503906
2023-01-21 12:15:48.550681: step: 1596/527, loss: 0.002473801374435425
2023-01-21 12:15:49.645660: step: 1600/527, loss: 0.0032970430329442024
2023-01-21 12:15:50.759472: step: 1604/527, loss: 0.003894805908203125
2023-01-21 12:15:51.865580: step: 1608/527, loss: 0.12098999321460724
2023-01-21 12:15:52.952985: step: 1612/527, loss: 0.005961323156952858
2023-01-21 12:15:54.048370: step: 1616/527, loss: 0.012486887164413929
2023-01-21 12:15:55.157187: step: 1620/527, loss: 0.013773728162050247
2023-01-21 12:15:56.259223: step: 1624/527, loss: 0.0011380196083337069
2023-01-21 12:15:57.360596: step: 1628/527, loss: 0.0025688172318041325
2023-01-21 12:15:58.487265: step: 1632/527, loss: 0.021014787256717682
2023-01-21 12:15:59.630868: step: 1636/527, loss: 1.773834264895413e-05
2023-01-21 12:16:00.731889: step: 1640/527, loss: 0.006216621492058039
2023-01-21 12:16:01.874509: step: 1644/527, loss: 0.06589861214160919
2023-01-21 12:16:02.999219: step: 1648/527, loss: 0.00445170421153307
2023-01-21 12:16:04.117113: step: 1652/527, loss: 0.028729820623993874
2023-01-21 12:16:05.249746: step: 1656/527, loss: 0.010041999630630016
2023-01-21 12:16:06.356340: step: 1660/527, loss: 0.0007791519165039062
2023-01-21 12:16:07.457756: step: 1664/527, loss: 0.001821804093196988
2023-01-21 12:16:08.582759: step: 1668/527, loss: 0.0048088072799146175
2023-01-21 12:16:09.673794: step: 1672/527, loss: 0.00033655166043899953
2023-01-21 12:16:10.824970: step: 1676/527, loss: 0.03624897450208664
2023-01-21 12:16:12.001969: step: 1680/527, loss: 0.0005367278936319053
2023-01-21 12:16:13.117976: step: 1684/527, loss: 0.022812461480498314
2023-01-21 12:16:14.240298: step: 1688/527, loss: 0.00011477470980025828
2023-01-21 12:16:15.327627: step: 1692/527, loss: 0.010447025299072266
2023-01-21 12:16:16.448752: step: 1696/527, loss: 0.013906383886933327
2023-01-21 12:16:17.569251: step: 1700/527, loss: 0.0009957790607586503
2023-01-21 12:16:18.695703: step: 1704/527, loss: 0.016587449237704277
2023-01-21 12:16:19.812150: step: 1708/527, loss: 0.0001066207914846018
2023-01-21 12:16:20.956576: step: 1712/527, loss: 0.008736801333725452
2023-01-21 12:16:22.091731: step: 1716/527, loss: 0.02396678924560547
2023-01-21 12:16:23.200858: step: 1720/527, loss: 0.0004646301385946572
2023-01-21 12:16:24.319652: step: 1724/527, loss: 0.0338197723031044
2023-01-21 12:16:25.467550: step: 1728/527, loss: 0.011371993459761143
2023-01-21 12:16:26.571872: step: 1732/527, loss: 0.0016618729569017887
2023-01-21 12:16:27.672468: step: 1736/527, loss: 0.11559267342090607
2023-01-21 12:16:28.759273: step: 1740/527, loss: 0.2119472473859787
2023-01-21 12:16:29.859434: step: 1744/527, loss: 0.00013999939255882055
2023-01-21 12:16:30.967363: step: 1748/527, loss: 0.025621414184570312
2023-01-21 12:16:32.086715: step: 1752/527, loss: 0.002800035523250699
2023-01-21 12:16:33.243896: step: 1756/527, loss: 0.003209733869880438
2023-01-21 12:16:34.351346: step: 1760/527, loss: 0.0009115219581872225
2023-01-21 12:16:35.450333: step: 1764/527, loss: 0.012387752532958984
2023-01-21 12:16:36.581674: step: 1768/527, loss: 0.0020169259514659643
2023-01-21 12:16:37.682161: step: 1772/527, loss: 0.05519409477710724
2023-01-21 12:16:38.795208: step: 1776/527, loss: 0.0023979186080396175
2023-01-21 12:16:39.904230: step: 1780/527, loss: 0.03604469448328018
2023-01-21 12:16:41.016289: step: 1784/527, loss: 0.0015453338855877519
2023-01-21 12:16:42.136809: step: 1788/527, loss: 0.0059226988814771175
2023-01-21 12:16:43.289450: step: 1792/527, loss: 0.0022039413452148438
2023-01-21 12:16:44.403214: step: 1796/527, loss: 0.015506553463637829
2023-01-21 12:16:45.503492: step: 1800/527, loss: 0.0003748893504962325
2023-01-21 12:16:46.627778: step: 1804/527, loss: 0.021599579602479935
2023-01-21 12:16:47.727862: step: 1808/527, loss: 0.005167102906852961
2023-01-21 12:16:48.863701: step: 1812/527, loss: 0.008883094415068626
2023-01-21 12:16:49.977379: step: 1816/527, loss: 0.0042368886061012745
2023-01-21 12:16:51.078207: step: 1820/527, loss: 0.0001442909415345639
2023-01-21 12:16:52.183974: step: 1824/527, loss: 0.001965141389518976
2023-01-21 12:16:53.301374: step: 1828/527, loss: 0.16837435960769653
2023-01-21 12:16:54.417209: step: 1832/527, loss: 0.04644737392663956
2023-01-21 12:16:55.541162: step: 1836/527, loss: 0.001495456788688898
2023-01-21 12:16:56.641583: step: 1840/527, loss: 0.02749796025454998
2023-01-21 12:16:57.767579: step: 1844/527, loss: 0.024663161486387253
2023-01-21 12:16:58.842721: step: 1848/527, loss: 0.00043544769869185984
2023-01-21 12:16:59.945629: step: 1852/527, loss: 0.0008251190301962197
2023-01-21 12:17:01.086886: step: 1856/527, loss: 0.00959930382668972
2023-01-21 12:17:02.195622: step: 1860/527, loss: 0.030899692326784134
2023-01-21 12:17:03.326586: step: 1864/527, loss: 0.019400596618652344
2023-01-21 12:17:04.445042: step: 1868/527, loss: 0.0005040168762207031
2023-01-21 12:17:05.578505: step: 1872/527, loss: 0.01944141462445259
2023-01-21 12:17:06.699444: step: 1876/527, loss: 0.024878978729248047
2023-01-21 12:17:07.802903: step: 1880/527, loss: 0.0004665374872274697
2023-01-21 12:17:08.927744: step: 1884/527, loss: 0.004247569944709539
2023-01-21 12:17:10.006048: step: 1888/527, loss: 0.01380624808371067
2023-01-21 12:17:11.096758: step: 1892/527, loss: 6.904602196300402e-05
2023-01-21 12:17:12.193837: step: 1896/527, loss: 0.022628404200077057
2023-01-21 12:17:13.304149: step: 1900/527, loss: 0.03297929838299751
2023-01-21 12:17:14.425527: step: 1904/527, loss: 0.00018376113439444453
2023-01-21 12:17:15.558465: step: 1908/527, loss: 0.0005106925964355469
2023-01-21 12:17:16.636664: step: 1912/527, loss: 0.2286311537027359
2023-01-21 12:17:17.773020: step: 1916/527, loss: 0.003556633135303855
2023-01-21 12:17:18.904747: step: 1920/527, loss: 0.00367660541087389
2023-01-21 12:17:20.024714: step: 1924/527, loss: 0.004868888761848211
2023-01-21 12:17:21.115050: step: 1928/527, loss: 0.0060253143310546875
2023-01-21 12:17:22.231140: step: 1932/527, loss: 0.01381072960793972
2023-01-21 12:17:23.353835: step: 1936/527, loss: 0.0003581047058105469
2023-01-21 12:17:24.467636: step: 1940/527, loss: 0.01874103769659996
2023-01-21 12:17:25.600355: step: 1944/527, loss: 0.0175323486328125
2023-01-21 12:17:26.685113: step: 1948/527, loss: 0.029023267328739166
2023-01-21 12:17:27.816973: step: 1952/527, loss: 0.0393308661878109
2023-01-21 12:17:28.903291: step: 1956/527, loss: 0.004464888479560614
2023-01-21 12:17:30.050137: step: 1960/527, loss: 0.0200418159365654
2023-01-21 12:17:31.181000: step: 1964/527, loss: 0.00018434523371979594
2023-01-21 12:17:32.262173: step: 1968/527, loss: 0.006373309995979071
2023-01-21 12:17:33.360487: step: 1972/527, loss: 0.003254509065300226
2023-01-21 12:17:34.477001: step: 1976/527, loss: 0.0027070045471191406
2023-01-21 12:17:35.569334: step: 1980/527, loss: 0.0002590179501567036
2023-01-21 12:17:36.700083: step: 1984/527, loss: 0.0009178161853924394
2023-01-21 12:17:37.843645: step: 1988/527, loss: 0.012150573544204235
2023-01-21 12:17:38.973760: step: 1992/527, loss: 0.001127052353695035
2023-01-21 12:17:40.102587: step: 1996/527, loss: 0.0005780697101727128
2023-01-21 12:17:41.200968: step: 2000/527, loss: 0.0018157005542889237
2023-01-21 12:17:42.321041: step: 2004/527, loss: 0.016777994111180305
2023-01-21 12:17:43.436269: step: 2008/527, loss: 0.00564079312607646
2023-01-21 12:17:44.541367: step: 2012/527, loss: 3.256797936046496e-05
2023-01-21 12:17:45.674600: step: 2016/527, loss: 0.0010238647228106856
2023-01-21 12:17:46.803184: step: 2020/527, loss: 0.002127408981323242
2023-01-21 12:17:47.936462: step: 2024/527, loss: 0.041478537023067474
2023-01-21 12:17:49.089730: step: 2028/527, loss: 0.03844613954424858
2023-01-21 12:17:50.212953: step: 2032/527, loss: 0.0011966705787926912
2023-01-21 12:17:51.320205: step: 2036/527, loss: 0.16422395408153534
2023-01-21 12:17:52.440323: step: 2040/527, loss: 0.03650417551398277
2023-01-21 12:17:53.586951: step: 2044/527, loss: 0.0003999710315838456
2023-01-21 12:17:54.701786: step: 2048/527, loss: 0.003025627229362726
2023-01-21 12:17:55.802366: step: 2052/527, loss: 0.0007245063898153603
2023-01-21 12:17:56.917784: step: 2056/527, loss: 0.005990314297378063
2023-01-21 12:17:58.038567: step: 2060/527, loss: 0.0006962775951251388
2023-01-21 12:17:59.155443: step: 2064/527, loss: 0.00025615692720748484
2023-01-21 12:18:00.267076: step: 2068/527, loss: 0.0011220932938158512
2023-01-21 12:18:01.395048: step: 2072/527, loss: 0.04845075681805611
2023-01-21 12:18:02.516685: step: 2076/527, loss: 0.0014861106174066663
2023-01-21 12:18:03.631777: step: 2080/527, loss: 0.014460563659667969
2023-01-21 12:18:04.736246: step: 2084/527, loss: 0.02891988679766655
2023-01-21 12:18:05.833207: step: 2088/527, loss: 0.00039224623469635844
2023-01-21 12:18:06.930964: step: 2092/527, loss: 0.05199460685253143
2023-01-21 12:18:08.046658: step: 2096/527, loss: 0.013969993218779564
2023-01-21 12:18:09.171399: step: 2100/527, loss: 0.002536487765610218
2023-01-21 12:18:10.299201: step: 2104/527, loss: 0.04461870342493057
2023-01-21 12:18:11.415925: step: 2108/527, loss: 0.006943607237190008
==================================================
Loss: 0.021
--------------------
Dev: {'event': {'p': 0.5830845771144278, 'r': 0.7802929427430093, 'f1': 0.6674259681093394}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21}
Test: {'event': {'p': 0.6141975308641975, 'r': 0.796, 'f1': 0.6933797909407665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21}
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': 21}
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': 21}
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': 21}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 12:18:52.858699: step: 4/527, loss: 0.0005558967823162675
2023-01-21 12:18:53.980219: step: 8/527, loss: 0.0004991531604900956
2023-01-21 12:18:55.098662: step: 12/527, loss: 0.07324127852916718
2023-01-21 12:18:56.183991: step: 16/527, loss: 0.000827789306640625
2023-01-21 12:18:57.295200: step: 20/527, loss: 0.007728958036750555
2023-01-21 12:18:58.383384: step: 24/527, loss: 0.00080957415048033
2023-01-21 12:18:59.474060: step: 28/527, loss: 0.0006683349492959678
2023-01-21 12:19:00.568593: step: 32/527, loss: 0.008914565667510033
2023-01-21 12:19:01.657448: step: 36/527, loss: 0.0027980327140539885
2023-01-21 12:19:02.771803: step: 40/527, loss: 0.0003648757701739669
2023-01-21 12:19:03.879453: step: 44/527, loss: 0.00959391612559557
2023-01-21 12:19:04.978029: step: 48/527, loss: 3.6239625842426904e-06
2023-01-21 12:19:06.106607: step: 52/527, loss: 0.001872253487817943
2023-01-21 12:19:07.223390: step: 56/527, loss: 0.00933923665434122
2023-01-21 12:19:08.323620: step: 60/527, loss: 9.641647920943797e-05
2023-01-21 12:19:09.422219: step: 64/527, loss: 0.019780471920967102
2023-01-21 12:19:10.552284: step: 68/527, loss: 0.0026884081307798624
2023-01-21 12:19:11.665624: step: 72/527, loss: 0.011364745907485485
2023-01-21 12:19:12.796751: step: 76/527, loss: 0.03705139085650444
2023-01-21 12:19:13.926623: step: 80/527, loss: 0.018434714525938034
2023-01-21 12:19:15.015070: step: 84/527, loss: 0.00048060418339446187
2023-01-21 12:19:16.140949: step: 88/527, loss: -3.528594970703125e-05
2023-01-21 12:19:17.291707: step: 92/527, loss: 0.18018054962158203
2023-01-21 12:19:18.429501: step: 96/527, loss: 0.0020869255531579256
2023-01-21 12:19:19.542726: step: 100/527, loss: 0.0002636909775901586
2023-01-21 12:19:20.657778: step: 104/527, loss: 0.01960763894021511
2023-01-21 12:19:21.759603: step: 108/527, loss: 0.00822839792817831
2023-01-21 12:19:22.877540: step: 112/527, loss: 0.0005994796520099044
2023-01-21 12:19:23.968075: step: 116/527, loss: 1.029968279908644e-05
2023-01-21 12:19:25.095912: step: 120/527, loss: 0.0008114815573208034
2023-01-21 12:19:26.209165: step: 124/527, loss: 0.00018100740271620452
2023-01-21 12:19:27.344311: step: 128/527, loss: 0.00969605427235365
2023-01-21 12:19:28.461615: step: 132/527, loss: 3.3378603347955504e-07
2023-01-21 12:19:29.611565: step: 136/527, loss: 0.011018563061952591
2023-01-21 12:19:30.713261: step: 140/527, loss: 0.002222251845523715
2023-01-21 12:19:31.839520: step: 144/527, loss: 0.0002647399960551411
2023-01-21 12:19:32.947581: step: 148/527, loss: 0.00054168701171875
2023-01-21 12:19:34.055825: step: 152/527, loss: 0.01643352583050728
2023-01-21 12:19:35.183389: step: 156/527, loss: 0.011928845196962357
2023-01-21 12:19:36.325591: step: 160/527, loss: 0.005833625793457031
2023-01-21 12:19:37.441839: step: 164/527, loss: 0.000813388847745955
2023-01-21 12:19:38.532774: step: 168/527, loss: 0.004551601596176624
2023-01-21 12:19:39.612674: step: 172/527, loss: 7.62939453125e-06
2023-01-21 12:19:40.739590: step: 176/527, loss: 0.010396957397460938
2023-01-21 12:19:41.852140: step: 180/527, loss: 0.00013246535672806203
2023-01-21 12:19:42.958901: step: 184/527, loss: 0.0003122329944744706
2023-01-21 12:19:44.073700: step: 188/527, loss: 0.0016575813060626388
2023-01-21 12:19:45.201087: step: 192/527, loss: 8.277893357444555e-05
2023-01-21 12:19:46.326156: step: 196/527, loss: 0.0002503395080566406
2023-01-21 12:19:47.470376: step: 200/527, loss: 0.010983943939208984
2023-01-21 12:19:48.585158: step: 204/527, loss: 0.023090267553925514
2023-01-21 12:19:49.715494: step: 208/527, loss: 7.696151442360133e-05
2023-01-21 12:19:50.821061: step: 212/527, loss: 0.013471603393554688
2023-01-21 12:19:51.938094: step: 216/527, loss: 0.0002400398370809853
2023-01-21 12:19:53.064969: step: 220/527, loss: 0.009569359011948109
2023-01-21 12:19:54.158128: step: 224/527, loss: 0.004719185642898083
2023-01-21 12:19:55.270220: step: 228/527, loss: 3.9577484130859375e-05
2023-01-21 12:19:56.369099: step: 232/527, loss: 8.206367056118324e-05
2023-01-21 12:19:57.512511: step: 236/527, loss: 0.0002610206720419228
2023-01-21 12:19:58.623049: step: 240/527, loss: 0.0025115013122558594
2023-01-21 12:19:59.798095: step: 244/527, loss: 0.0019284249283373356
2023-01-21 12:20:00.924481: step: 248/527, loss: 0.00011281967454124242
2023-01-21 12:20:02.013549: step: 252/527, loss: 0.09685325622558594
2023-01-21 12:20:03.130725: step: 256/527, loss: 0.0008448601001873612
2023-01-21 12:20:04.234178: step: 260/527, loss: 0.00021781922259833664
2023-01-21 12:20:05.327617: step: 264/527, loss: 0.014787578955292702
2023-01-21 12:20:06.451954: step: 268/527, loss: 0.0017694473499432206
2023-01-21 12:20:07.563652: step: 272/527, loss: 4.444122168933973e-05
2023-01-21 12:20:08.715579: step: 276/527, loss: 0.0029649734497070312
2023-01-21 12:20:09.823661: step: 280/527, loss: 0.00012035370309604332
2023-01-21 12:20:10.938185: step: 284/527, loss: 0.007779121398925781
2023-01-21 12:20:12.043074: step: 288/527, loss: 0.03949422761797905
2023-01-21 12:20:13.170753: step: 292/527, loss: 0.00016922950453590602
2023-01-21 12:20:14.299885: step: 296/527, loss: 0.2985950708389282
2023-01-21 12:20:15.373183: step: 300/527, loss: 0.014481925405561924
2023-01-21 12:20:16.476243: step: 304/527, loss: 0.00300769810564816
2023-01-21 12:20:17.596127: step: 308/527, loss: 0.015477752313017845
2023-01-21 12:20:18.724047: step: 312/527, loss: 0.0036468505859375
2023-01-21 12:20:19.846870: step: 316/527, loss: 0.019165420904755592
2023-01-21 12:20:20.964069: step: 320/527, loss: 0.00016679763211868703
2023-01-21 12:20:22.031704: step: 324/527, loss: 0.000958347343839705
2023-01-21 12:20:23.109741: step: 328/527, loss: 0.002056217286735773
2023-01-21 12:20:24.274320: step: 332/527, loss: 0.030867289751768112
2023-01-21 12:20:25.353310: step: 336/527, loss: 7.05719003235572e-06
2023-01-21 12:20:26.449870: step: 340/527, loss: 0.020842932164669037
2023-01-21 12:20:27.543093: step: 344/527, loss: 0.009637641720473766
2023-01-21 12:20:28.637019: step: 348/527, loss: 0.01580219343304634
2023-01-21 12:20:29.740422: step: 352/527, loss: 0.0066525936126708984
2023-01-21 12:20:30.834096: step: 356/527, loss: 0.0009056210983544588
2023-01-21 12:20:31.935818: step: 360/527, loss: 0.005259132478386164
2023-01-21 12:20:33.054644: step: 364/527, loss: 0.0005645751953125
2023-01-21 12:20:34.196100: step: 368/527, loss: 0.01938645914196968
2023-01-21 12:20:35.290368: step: 372/527, loss: 0.0196057315915823
2023-01-21 12:20:36.407514: step: 376/527, loss: 3.0517578125e-05
2023-01-21 12:20:37.496482: step: 380/527, loss: 0.004533624742180109
2023-01-21 12:20:38.646749: step: 384/527, loss: 0.02848644368350506
2023-01-21 12:20:39.759906: step: 388/527, loss: 0.00016021728515625
2023-01-21 12:20:40.864011: step: 392/527, loss: 0.0010353089310228825
2023-01-21 12:20:41.996756: step: 396/527, loss: 0.005118751898407936
2023-01-21 12:20:43.124718: step: 400/527, loss: 0.06099729239940643
2023-01-21 12:20:44.242439: step: 404/527, loss: 0.0004983901744708419
2023-01-21 12:20:45.347573: step: 408/527, loss: 0.01264648512005806
2023-01-21 12:20:46.476935: step: 412/527, loss: 0.0031671524047851562
2023-01-21 12:20:47.567790: step: 416/527, loss: 0.013668537139892578
2023-01-21 12:20:48.650372: step: 420/527, loss: 0.00015203953080344945
2023-01-21 12:20:49.756340: step: 424/527, loss: 8.840560622047633e-05
2023-01-21 12:20:50.911851: step: 428/527, loss: 0.023430252447724342
2023-01-21 12:20:52.057177: step: 432/527, loss: 0.00021762849064543843
2023-01-21 12:20:53.183049: step: 436/527, loss: 0.0005807876586914062
2023-01-21 12:20:54.291221: step: 440/527, loss: 0.0004954338073730469
2023-01-21 12:20:55.398514: step: 444/527, loss: 0.8372736573219299
2023-01-21 12:20:56.495883: step: 448/527, loss: 0.009445381350815296
2023-01-21 12:20:57.597005: step: 452/527, loss: 0.009234047494828701
2023-01-21 12:20:58.715788: step: 456/527, loss: 0.0020020485389977694
2023-01-21 12:20:59.818338: step: 460/527, loss: 0.0003277778741903603
2023-01-21 12:21:00.939013: step: 464/527, loss: 0.000448036240413785
2023-01-21 12:21:02.055171: step: 468/527, loss: 0.005885219667106867
2023-01-21 12:21:03.193226: step: 472/527, loss: 0.0005820274818688631
2023-01-21 12:21:04.308992: step: 476/527, loss: 0.0004261017020326108
2023-01-21 12:21:05.443301: step: 480/527, loss: 0.0004280567227397114
2023-01-21 12:21:06.545477: step: 484/527, loss: 0.004059505648910999
2023-01-21 12:21:07.682309: step: 488/527, loss: 0.004740810953080654
2023-01-21 12:21:08.783126: step: 492/527, loss: 0.001324558281339705
2023-01-21 12:21:09.904429: step: 496/527, loss: 0.0010866165393963456
2023-01-21 12:21:10.991853: step: 500/527, loss: 0.011626482009887695
2023-01-21 12:21:12.119857: step: 504/527, loss: 0.0015188216930255294
2023-01-21 12:21:13.234747: step: 508/527, loss: 0.003599930088967085
2023-01-21 12:21:14.343903: step: 512/527, loss: 0.00293140416033566
2023-01-21 12:21:15.441143: step: 516/527, loss: 0.016948889940977097
2023-01-21 12:21:16.570682: step: 520/527, loss: 0.0037254809867590666
2023-01-21 12:21:17.717784: step: 524/527, loss: 0.017209814861416817
2023-01-21 12:21:18.825953: step: 528/527, loss: 0.0046545034274458885
2023-01-21 12:21:19.939769: step: 532/527, loss: 0.004320716951042414
2023-01-21 12:21:21.057312: step: 536/527, loss: 0.00037631989107467234
2023-01-21 12:21:22.167553: step: 540/527, loss: 0.02561493031680584
2023-01-21 12:21:23.273050: step: 544/527, loss: 0.0033080100547522306
2023-01-21 12:21:24.405900: step: 548/527, loss: 0.026097487658262253
2023-01-21 12:21:25.543687: step: 552/527, loss: 0.15177659690380096
2023-01-21 12:21:26.669514: step: 556/527, loss: 0.00023493765911553055
2023-01-21 12:21:27.812964: step: 560/527, loss: 0.03329582139849663
2023-01-21 12:21:28.947853: step: 564/527, loss: 0.00020818710618186742
2023-01-21 12:21:30.062946: step: 568/527, loss: 0.00498156575486064
2023-01-21 12:21:31.196732: step: 572/527, loss: 0.011979294009506702
2023-01-21 12:21:32.332744: step: 576/527, loss: 0.00028972624568268657
2023-01-21 12:21:33.442075: step: 580/527, loss: 0.04299907758831978
2023-01-21 12:21:34.596269: step: 584/527, loss: 0.024701213464140892
2023-01-21 12:21:35.708132: step: 588/527, loss: 0.035338591784238815
2023-01-21 12:21:36.793647: step: 592/527, loss: 0.016944551840424538
2023-01-21 12:21:37.921384: step: 596/527, loss: 0.005271721165627241
2023-01-21 12:21:39.046524: step: 600/527, loss: 0.018448447808623314
2023-01-21 12:21:40.169835: step: 604/527, loss: 0.04133720323443413
2023-01-21 12:21:41.267458: step: 608/527, loss: 0.0001282215234823525
2023-01-21 12:21:42.379565: step: 612/527, loss: 0.1305810511112213
2023-01-21 12:21:43.522330: step: 616/527, loss: 0.017786789685487747
2023-01-21 12:21:44.631618: step: 620/527, loss: 0.12646332383155823
2023-01-21 12:21:45.783063: step: 624/527, loss: 0.0006554127321578562
2023-01-21 12:21:46.896092: step: 628/527, loss: 0.03694958612322807
2023-01-21 12:21:47.998554: step: 632/527, loss: 0.0008533477666787803
2023-01-21 12:21:49.103561: step: 636/527, loss: 2.2935870219953358e-05
2023-01-21 12:21:50.236909: step: 640/527, loss: 0.017215536907315254
2023-01-21 12:21:51.322431: step: 644/527, loss: 0.005025959108024836
2023-01-21 12:21:52.439889: step: 648/527, loss: 0.0014410018920898438
2023-01-21 12:21:53.576505: step: 652/527, loss: 0.0013002397026866674
2023-01-21 12:21:54.702331: step: 656/527, loss: 0.03106250800192356
2023-01-21 12:21:55.839545: step: 660/527, loss: 0.0011707305675372481
2023-01-21 12:21:56.982536: step: 664/527, loss: 0.0026319504249840975
2023-01-21 12:21:58.060467: step: 668/527, loss: 1.926422191900201e-05
2023-01-21 12:21:59.180499: step: 672/527, loss: 0.001215267227962613
2023-01-21 12:22:00.312175: step: 676/527, loss: 0.0006723403930664062
2023-01-21 12:22:01.439669: step: 680/527, loss: 0.05684547498822212
2023-01-21 12:22:02.597517: step: 684/527, loss: 0.019237495958805084
2023-01-21 12:22:03.736056: step: 688/527, loss: 0.00013465881056617945
2023-01-21 12:22:04.848882: step: 692/527, loss: 0.05452442169189453
2023-01-21 12:22:05.958514: step: 696/527, loss: 0.00859603937715292
2023-01-21 12:22:07.039594: step: 700/527, loss: 9.489059448242188e-05
2023-01-21 12:22:08.116518: step: 704/527, loss: 0.0008376121986657381
2023-01-21 12:22:09.224120: step: 708/527, loss: 0.08675622940063477
2023-01-21 12:22:10.336240: step: 712/527, loss: 0.006838035769760609
2023-01-21 12:22:11.471745: step: 716/527, loss: 5.722046262235381e-05
2023-01-21 12:22:12.601633: step: 720/527, loss: 0.01479492150247097
2023-01-21 12:22:13.719425: step: 724/527, loss: 0.01570282131433487
2023-01-21 12:22:14.824978: step: 728/527, loss: 0.4493824541568756
2023-01-21 12:22:15.959907: step: 732/527, loss: 0.0002429485321044922
2023-01-21 12:22:17.048839: step: 736/527, loss: 0.046500302851200104
2023-01-21 12:22:18.147564: step: 740/527, loss: 0.0026671886444091797
2023-01-21 12:22:19.248860: step: 744/527, loss: 0.013757658191025257
2023-01-21 12:22:20.344175: step: 748/527, loss: 9.078979201149195e-05
2023-01-21 12:22:21.454535: step: 752/527, loss: 0.03950929641723633
2023-01-21 12:22:22.628080: step: 756/527, loss: 0.034891605377197266
2023-01-21 12:22:23.732762: step: 760/527, loss: 0.007219600956887007
2023-01-21 12:22:24.817394: step: 764/527, loss: 0.000857925449963659
2023-01-21 12:22:25.951207: step: 768/527, loss: 0.021681785583496094
2023-01-21 12:22:27.069296: step: 772/527, loss: 0.012350846081972122
2023-01-21 12:22:28.171413: step: 776/527, loss: 0.045530129224061966
2023-01-21 12:22:29.277871: step: 780/527, loss: 0.06528101116418839
2023-01-21 12:22:30.420687: step: 784/527, loss: 0.00722503662109375
2023-01-21 12:22:31.533859: step: 788/527, loss: 0.02055356465280056
2023-01-21 12:22:32.629250: step: 792/527, loss: 0.00019121170043945312
2023-01-21 12:22:33.749631: step: 796/527, loss: 0.0017677306896075606
2023-01-21 12:22:34.893667: step: 800/527, loss: 0.07141242176294327
2023-01-21 12:22:36.004553: step: 804/527, loss: -4.0054324017546605e-06
2023-01-21 12:22:37.112678: step: 808/527, loss: 0.010314560495316982
2023-01-21 12:22:38.213538: step: 812/527, loss: 0.026074696332216263
2023-01-21 12:22:39.329926: step: 816/527, loss: 0.14199551939964294
2023-01-21 12:22:40.450692: step: 820/527, loss: 0.02960786782205105
2023-01-21 12:22:41.552567: step: 824/527, loss: 3.252029637224041e-05
2023-01-21 12:22:42.674386: step: 828/527, loss: 0.016623497009277344
2023-01-21 12:22:43.766678: step: 832/527, loss: 0.004015207290649414
2023-01-21 12:22:44.900501: step: 836/527, loss: 0.0009103774791583419
2023-01-21 12:22:45.985560: step: 840/527, loss: 0.008807850070297718
2023-01-21 12:22:47.100456: step: 844/527, loss: 0.002062892774119973
2023-01-21 12:22:48.175655: step: 848/527, loss: 0.011046218685805798
2023-01-21 12:22:49.310265: step: 852/527, loss: 0.005708885379135609
2023-01-21 12:22:50.405218: step: 856/527, loss: 0.0010912418365478516
2023-01-21 12:22:51.496636: step: 860/527, loss: 2.86102294921875e-06
2023-01-21 12:22:52.596922: step: 864/527, loss: 0.04565735161304474
2023-01-21 12:22:53.731320: step: 868/527, loss: 0.010359669104218483
2023-01-21 12:22:54.870891: step: 872/527, loss: 0.005309867672622204
2023-01-21 12:22:56.005908: step: 876/527, loss: 0.0011375427711755037
2023-01-21 12:22:57.116878: step: 880/527, loss: 0.0030698776245117188
2023-01-21 12:22:58.217569: step: 884/527, loss: 0.00040431023808196187
2023-01-21 12:22:59.330634: step: 888/527, loss: 0.0009482384193688631
2023-01-21 12:23:00.470024: step: 892/527, loss: 0.0012921332381665707
2023-01-21 12:23:01.572264: step: 896/527, loss: 0.0362180694937706
2023-01-21 12:23:02.671810: step: 900/527, loss: 0.008481407538056374
2023-01-21 12:23:03.772847: step: 904/527, loss: 0.0001273631933145225
2023-01-21 12:23:04.900284: step: 908/527, loss: 0.004111862741410732
2023-01-21 12:23:05.989996: step: 912/527, loss: 0.0025197984650731087
2023-01-21 12:23:07.089649: step: 916/527, loss: 0.0653291642665863
2023-01-21 12:23:08.208008: step: 920/527, loss: 0.015781832858920097
2023-01-21 12:23:09.387985: step: 924/527, loss: 0.0046710968017578125
2023-01-21 12:23:10.493374: step: 928/527, loss: 1.182556115963962e-05
2023-01-21 12:23:11.605301: step: 932/527, loss: 0.008518218994140625
2023-01-21 12:23:12.704007: step: 936/527, loss: 0.062433816492557526
2023-01-21 12:23:13.853106: step: 940/527, loss: 6.27517729299143e-05
2023-01-21 12:23:14.988864: step: 944/527, loss: 0.0009416580433025956
2023-01-21 12:23:16.107697: step: 948/527, loss: 1.9073486328125e-05
2023-01-21 12:23:17.239818: step: 952/527, loss: 0.03326931223273277
2023-01-21 12:23:18.337415: step: 956/527, loss: 0.0005405425908975303
2023-01-21 12:23:19.441110: step: 960/527, loss: 0.03508793190121651
2023-01-21 12:23:20.568823: step: 964/527, loss: 0.6289817690849304
2023-01-21 12:23:21.711524: step: 968/527, loss: 0.012804890051484108
2023-01-21 12:23:22.859725: step: 972/527, loss: 0.0019335746765136719
2023-01-21 12:23:23.993062: step: 976/527, loss: 0.004665756598114967
2023-01-21 12:23:25.075913: step: 980/527, loss: 0.0023023607209324837
2023-01-21 12:23:26.203215: step: 984/527, loss: 0.00524950074031949
2023-01-21 12:23:27.338185: step: 988/527, loss: 0.028874970972537994
2023-01-21 12:23:28.450173: step: 992/527, loss: 0.00019741058349609375
2023-01-21 12:23:29.588433: step: 996/527, loss: 1.125335711549269e-05
2023-01-21 12:23:30.752740: step: 1000/527, loss: 0.0014068603049963713
2023-01-21 12:23:31.862743: step: 1004/527, loss: 0.0010789871448650956
2023-01-21 12:23:32.980171: step: 1008/527, loss: 0.00011510849435580894
2023-01-21 12:23:34.062523: step: 1012/527, loss: 0.01061267964541912
2023-01-21 12:23:35.172705: step: 1016/527, loss: 0.0029767754022032022
2023-01-21 12:23:36.268303: step: 1020/527, loss: 0.05207844078540802
2023-01-21 12:23:37.392860: step: 1024/527, loss: 0.0007111072773113847
2023-01-21 12:23:38.511883: step: 1028/527, loss: 0.044226959347724915
2023-01-21 12:23:39.616356: step: 1032/527, loss: 0.0357481949031353
2023-01-21 12:23:40.721450: step: 1036/527, loss: 0.0006395339732989669
2023-01-21 12:23:41.831849: step: 1040/527, loss: 0.004071617498993874
2023-01-21 12:23:42.997144: step: 1044/527, loss: 0.054274849593639374
2023-01-21 12:23:44.130024: step: 1048/527, loss: 2.3746490114717744e-05
2023-01-21 12:23:45.218663: step: 1052/527, loss: 0.0006860733265057206
2023-01-21 12:23:46.348588: step: 1056/527, loss: 0.022799300029873848
2023-01-21 12:23:47.445911: step: 1060/527, loss: 0.018406962975859642
2023-01-21 12:23:48.557790: step: 1064/527, loss: 0.004471207037568092
2023-01-21 12:23:49.684691: step: 1068/527, loss: 0.00031604766263626516
2023-01-21 12:23:50.799612: step: 1072/527, loss: 0.0063516139052808285
2023-01-21 12:23:51.912117: step: 1076/527, loss: 0.00361804966814816
2023-01-21 12:23:53.048804: step: 1080/527, loss: 0.03751087188720703
2023-01-21 12:23:54.166483: step: 1084/527, loss: 0.001027870224788785
2023-01-21 12:23:55.266633: step: 1088/527, loss: 0.007287454791367054
2023-01-21 12:23:56.381652: step: 1092/527, loss: 0.025456812232732773
2023-01-21 12:23:57.458676: step: 1096/527, loss: 0.006733942311257124
2023-01-21 12:23:58.552343: step: 1100/527, loss: 0.00207958254031837
2023-01-21 12:23:59.699065: step: 1104/527, loss: 6.32286028121598e-05
2023-01-21 12:24:00.800103: step: 1108/527, loss: 7.009506134636467e-06
2023-01-21 12:24:01.918793: step: 1112/527, loss: 1.697540210443549e-05
2023-01-21 12:24:03.057387: step: 1116/527, loss: 0.017817117273807526
2023-01-21 12:24:04.189980: step: 1120/527, loss: 0.0007919311756268144
2023-01-21 12:24:05.307848: step: 1124/527, loss: 0.015481948852539062
2023-01-21 12:24:06.399183: step: 1128/527, loss: 0.00013103484525345266
2023-01-21 12:24:07.533431: step: 1132/527, loss: 0.022413061931729317
2023-01-21 12:24:08.684460: step: 1136/527, loss: 0.0003266334533691406
2023-01-21 12:24:09.810772: step: 1140/527, loss: 0.00014848708815407008
2023-01-21 12:24:10.933287: step: 1144/527, loss: 0.09833307564258575
2023-01-21 12:24:12.037981: step: 1148/527, loss: 0.02944183349609375
2023-01-21 12:24:13.195535: step: 1152/527, loss: 0.15351122617721558
2023-01-21 12:24:14.320755: step: 1156/527, loss: 0.00013799667067360133
2023-01-21 12:24:15.426425: step: 1160/527, loss: 0.05336713790893555
2023-01-21 12:24:16.553427: step: 1164/527, loss: 7.63893112889491e-05
2023-01-21 12:24:17.705579: step: 1168/527, loss: 0.0007970810402184725
2023-01-21 12:24:18.838482: step: 1172/527, loss: 0.004556465428322554
2023-01-21 12:24:19.957086: step: 1176/527, loss: 0.0010605811839923263
2023-01-21 12:24:21.075820: step: 1180/527, loss: 0.0002649307425599545
2023-01-21 12:24:22.168693: step: 1184/527, loss: 0.060370828956365585
2023-01-21 12:24:23.300663: step: 1188/527, loss: 0.0013395309215411544
2023-01-21 12:24:24.417500: step: 1192/527, loss: 0.001082992646843195
2023-01-21 12:24:25.516810: step: 1196/527, loss: 0.014884661883115768
2023-01-21 12:24:26.623838: step: 1200/527, loss: 0.009298896417021751
2023-01-21 12:24:27.716793: step: 1204/527, loss: 2.96115867968183e-05
2023-01-21 12:24:28.807539: step: 1208/527, loss: 0.0247941967099905
2023-01-21 12:24:29.902558: step: 1212/527, loss: 0.0055252076126635075
2023-01-21 12:24:31.014201: step: 1216/527, loss: 0.001121139619499445
2023-01-21 12:24:32.139300: step: 1220/527, loss: 0.017994403839111328
2023-01-21 12:24:33.246894: step: 1224/527, loss: 0.02340412139892578
2023-01-21 12:24:34.383500: step: 1228/527, loss: 0.00017147065955214202
2023-01-21 12:24:35.492282: step: 1232/527, loss: 0.009339428506791592
2023-01-21 12:24:36.585311: step: 1236/527, loss: 0.0006149769178591669
2023-01-21 12:24:37.703653: step: 1240/527, loss: 0.00137500767596066
2023-01-21 12:24:38.791418: step: 1244/527, loss: 0.043062400072813034
2023-01-21 12:24:39.904345: step: 1248/527, loss: 0.023932933807373047
2023-01-21 12:24:41.015642: step: 1252/527, loss: 0.0021259307395666838
2023-01-21 12:24:42.118266: step: 1256/527, loss: 0.00031986235990189016
2023-01-21 12:24:43.208594: step: 1260/527, loss: 0.0002758025948423892
2023-01-21 12:24:44.315700: step: 1264/527, loss: 0.002234077313914895
2023-01-21 12:24:45.409278: step: 1268/527, loss: 0.05120287090539932
2023-01-21 12:24:46.559088: step: 1272/527, loss: 0.022077560424804688
2023-01-21 12:24:47.677006: step: 1276/527, loss: 0.03183254972100258
2023-01-21 12:24:48.768748: step: 1280/527, loss: 0.027191162109375
2023-01-21 12:24:49.881620: step: 1284/527, loss: 0.0013658524258062243
2023-01-21 12:24:51.007000: step: 1288/527, loss: 0.01043081283569336
2023-01-21 12:24:52.115516: step: 1292/527, loss: 2.7275087632006034e-05
2023-01-21 12:24:53.256386: step: 1296/527, loss: 0.02287764847278595
2023-01-21 12:24:54.324741: step: 1300/527, loss: 0.00013208389282226562
2023-01-21 12:24:55.438774: step: 1304/527, loss: 0.08474164456129074
2023-01-21 12:24:56.544846: step: 1308/527, loss: 0.022499753162264824
2023-01-21 12:24:57.658976: step: 1312/527, loss: 0.015335272997617722
2023-01-21 12:24:58.760035: step: 1316/527, loss: 0.13820815086364746
2023-01-21 12:24:59.876470: step: 1320/527, loss: 0.00022163391986396164
2023-01-21 12:25:01.002635: step: 1324/527, loss: 0.011668014340102673
2023-01-21 12:25:02.103261: step: 1328/527, loss: 0.00010423660569358617
2023-01-21 12:25:03.245931: step: 1332/527, loss: 0.0004512786981649697
2023-01-21 12:25:04.369265: step: 1336/527, loss: 0.012129020877182484
2023-01-21 12:25:05.498110: step: 1340/527, loss: 0.0027151110116392374
2023-01-21 12:25:06.597666: step: 1344/527, loss: 0.0013480186462402344
2023-01-21 12:25:07.726711: step: 1348/527, loss: 0.0011337280739098787
2023-01-21 12:25:08.827010: step: 1352/527, loss: 0.02316303364932537
2023-01-21 12:25:09.951390: step: 1356/527, loss: 0.005997275933623314
2023-01-21 12:25:11.059809: step: 1360/527, loss: 0.010057831183075905
2023-01-21 12:25:12.188470: step: 1364/527, loss: 0.0036478997208178043
2023-01-21 12:25:13.302573: step: 1368/527, loss: 0.10908007621765137
2023-01-21 12:25:14.407454: step: 1372/527, loss: 0.0001846313534770161
2023-01-21 12:25:15.517700: step: 1376/527, loss: 0.0015518426662310958
2023-01-21 12:25:16.623855: step: 1380/527, loss: 0.00038137438241392374
2023-01-21 12:25:17.758665: step: 1384/527, loss: 0.043082524091005325
2023-01-21 12:25:18.851042: step: 1388/527, loss: 7.715225365245715e-05
2023-01-21 12:25:19.956994: step: 1392/527, loss: 0.0017282486660405993
2023-01-21 12:25:21.070104: step: 1396/527, loss: 0.058941081166267395
2023-01-21 12:25:22.188899: step: 1400/527, loss: 1.2063979738741182e-05
2023-01-21 12:25:23.289030: step: 1404/527, loss: 0.06037035211920738
2023-01-21 12:25:24.431108: step: 1408/527, loss: 7.82012921263231e-06
2023-01-21 12:25:25.553499: step: 1412/527, loss: 0.0002521514834370464
2023-01-21 12:25:26.661798: step: 1416/527, loss: 0.009723186492919922
2023-01-21 12:25:27.781990: step: 1420/527, loss: 0.0006910323863849044
2023-01-21 12:25:28.886766: step: 1424/527, loss: 0.0027882575523108244
2023-01-21 12:25:29.986993: step: 1428/527, loss: 0.0022771835792809725
2023-01-21 12:25:31.116039: step: 1432/527, loss: 0.0024145126808434725
2023-01-21 12:25:32.231945: step: 1436/527, loss: 6.67572021484375e-06
2023-01-21 12:25:33.342564: step: 1440/527, loss: 0.011468315497040749
2023-01-21 12:25:34.425167: step: 1444/527, loss: 0.0697694718837738
2023-01-21 12:25:35.534795: step: 1448/527, loss: 5.2928924560546875e-05
2023-01-21 12:25:36.649189: step: 1452/527, loss: 0.007643222808837891
2023-01-21 12:25:37.768749: step: 1456/527, loss: 0.0215485580265522
2023-01-21 12:25:38.843200: step: 1460/527, loss: 0.00822000578045845
2023-01-21 12:25:39.959928: step: 1464/527, loss: 0.0030698776245117188
2023-01-21 12:25:41.049429: step: 1468/527, loss: 0.004712950903922319
2023-01-21 12:25:42.188831: step: 1472/527, loss: 0.0005050658946856856
2023-01-21 12:25:43.325286: step: 1476/527, loss: -2.1457672119140625e-06
2023-01-21 12:25:44.466156: step: 1480/527, loss: 0.0072879791259765625
2023-01-21 12:25:45.554647: step: 1484/527, loss: 0.0498899444937706
2023-01-21 12:25:46.670735: step: 1488/527, loss: 0.002757358830422163
2023-01-21 12:25:47.761412: step: 1492/527, loss: 0.013018417172133923
2023-01-21 12:25:48.859013: step: 1496/527, loss: 0.0013814927078783512
2023-01-21 12:25:49.991565: step: 1500/527, loss: 0.0156676284968853
2023-01-21 12:25:51.161593: step: 1504/527, loss: 0.010724353604018688
2023-01-21 12:25:52.251469: step: 1508/527, loss: 0.04039440304040909
2023-01-21 12:25:53.373648: step: 1512/527, loss: 0.012734705582261086
2023-01-21 12:25:54.497487: step: 1516/527, loss: 0.0029209135100245476
2023-01-21 12:25:55.603874: step: 1520/527, loss: 0.02089405059814453
2023-01-21 12:25:56.728849: step: 1524/527, loss: 0.01139287929981947
2023-01-21 12:25:57.880041: step: 1528/527, loss: 0.03298530727624893
2023-01-21 12:25:58.988708: step: 1532/527, loss: 0.535767138004303
2023-01-21 12:26:00.089901: step: 1536/527, loss: 0.0004978179931640625
2023-01-21 12:26:01.205679: step: 1540/527, loss: 0.015664197504520416
2023-01-21 12:26:02.312122: step: 1544/527, loss: 0.008694302290678024
2023-01-21 12:26:03.412686: step: 1548/527, loss: 0.0002738952753134072
2023-01-21 12:26:04.529080: step: 1552/527, loss: 0.0004563808615785092
2023-01-21 12:26:05.627549: step: 1556/527, loss: 2.527237120375503e-05
2023-01-21 12:26:06.718387: step: 1560/527, loss: 0.00012454987154342234
2023-01-21 12:26:07.824890: step: 1564/527, loss: 0.0661611557006836
2023-01-21 12:26:08.943518: step: 1568/527, loss: 0.0479188933968544
2023-01-21 12:26:10.094423: step: 1572/527, loss: 0.06435756385326385
2023-01-21 12:26:11.190336: step: 1576/527, loss: 0.0600128173828125
2023-01-21 12:26:12.329827: step: 1580/527, loss: 0.007936668582260609
2023-01-21 12:26:13.454341: step: 1584/527, loss: 0.16411086916923523
2023-01-21 12:26:14.538677: step: 1588/527, loss: 2.2758544218959287e-05
2023-01-21 12:26:15.627009: step: 1592/527, loss: 0.0006431579240597785
2023-01-21 12:26:16.762454: step: 1596/527, loss: 0.010871505364775658
2023-01-21 12:26:17.852284: step: 1600/527, loss: 0.0017646790947765112
2023-01-21 12:26:18.950918: step: 1604/527, loss: 0.013074017129838467
2023-01-21 12:26:20.066440: step: 1608/527, loss: 0.0011531829368323088
2023-01-21 12:26:21.196853: step: 1612/527, loss: 0.00038280486478470266
2023-01-21 12:26:22.305141: step: 1616/527, loss: 0.007318305782973766
2023-01-21 12:26:23.404084: step: 1620/527, loss: 0.01648383028805256
2023-01-21 12:26:24.511096: step: 1624/527, loss: 0.006229210179299116
2023-01-21 12:26:25.618940: step: 1628/527, loss: 0.0038916587363928556
2023-01-21 12:26:26.733610: step: 1632/527, loss: 0.013828659430146217
2023-01-21 12:26:27.842777: step: 1636/527, loss: 0.25910741090774536
2023-01-21 12:26:28.946647: step: 1640/527, loss: 3.128051685052924e-05
2023-01-21 12:26:30.042247: step: 1644/527, loss: 0.049620676785707474
2023-01-21 12:26:31.170437: step: 1648/527, loss: 0.003712183330208063
2023-01-21 12:26:32.346069: step: 1652/527, loss: 0.12752917408943176
2023-01-21 12:26:33.440739: step: 1656/527, loss: 0.025815939530730247
2023-01-21 12:26:34.544535: step: 1660/527, loss: 0.0006519794696941972
2023-01-21 12:26:35.639863: step: 1664/527, loss: 0.008729267865419388
2023-01-21 12:26:36.734715: step: 1668/527, loss: 0.006938266567885876
2023-01-21 12:26:37.896406: step: 1672/527, loss: 0.00023117066302802414
2023-01-21 12:26:39.043095: step: 1676/527, loss: 0.0013781548477709293
2023-01-21 12:26:40.163911: step: 1680/527, loss: 0.013305379077792168
2023-01-21 12:26:41.264007: step: 1684/527, loss: 6.866455078125e-05
2023-01-21 12:26:42.403763: step: 1688/527, loss: 0.0016859056195244193
2023-01-21 12:26:43.507046: step: 1692/527, loss: 3.719329833984375e-05
2023-01-21 12:26:44.616379: step: 1696/527, loss: 0.0001548767031636089
2023-01-21 12:26:45.726181: step: 1700/527, loss: 3.719329924933845e-06
2023-01-21 12:26:46.804068: step: 1704/527, loss: 0.028162576258182526
2023-01-21 12:26:47.940002: step: 1708/527, loss: 3.5953522456111386e-05
2023-01-21 12:26:49.052946: step: 1712/527, loss: 0.0257783904671669
2023-01-21 12:26:50.176461: step: 1716/527, loss: 1.3542176020564511e-05
2023-01-21 12:26:51.276428: step: 1720/527, loss: 0.00014452934556175023
2023-01-21 12:26:52.407867: step: 1724/527, loss: 0.0040236469358205795
2023-01-21 12:26:53.542822: step: 1728/527, loss: 1.7929078239831142e-05
2023-01-21 12:26:54.682159: step: 1732/527, loss: 0.0007322788005694747
2023-01-21 12:26:55.784198: step: 1736/527, loss: 0.0008134841918945312
2023-01-21 12:26:56.902778: step: 1740/527, loss: 0.0024483681190758944
2023-01-21 12:26:58.017815: step: 1744/527, loss: 0.0016332627274096012
2023-01-21 12:26:59.128220: step: 1748/527, loss: 0.004801368806511164
2023-01-21 12:27:00.226660: step: 1752/527, loss: 0.0023527145385742188
2023-01-21 12:27:01.373997: step: 1756/527, loss: 0.17571859061717987
2023-01-21 12:27:02.476065: step: 1760/527, loss: 0.0013370513916015625
2023-01-21 12:27:03.611166: step: 1764/527, loss: 0.00010089873831020668
2023-01-21 12:27:04.713642: step: 1768/527, loss: 0.00666465749964118
2023-01-21 12:27:05.827497: step: 1772/527, loss: 0.010736274532973766
2023-01-21 12:27:06.911607: step: 1776/527, loss: 0.0024148940574377775
2023-01-21 12:27:08.044208: step: 1780/527, loss: 0.011734199710190296
2023-01-21 12:27:09.146826: step: 1784/527, loss: 0.013148784637451172
2023-01-21 12:27:10.243939: step: 1788/527, loss: 0.003081083297729492
2023-01-21 12:27:11.378893: step: 1792/527, loss: 0.006851387210190296
2023-01-21 12:27:12.486166: step: 1796/527, loss: 0.002324390457943082
2023-01-21 12:27:13.625370: step: 1800/527, loss: 0.0006364822038449347
2023-01-21 12:27:14.777306: step: 1804/527, loss: 0.00011000633821822703
2023-01-21 12:27:15.879687: step: 1808/527, loss: 0.015353584662079811
2023-01-21 12:27:16.967012: step: 1812/527, loss: 0.016373634338378906
2023-01-21 12:27:18.080601: step: 1816/527, loss: 0.04403867945075035
2023-01-21 12:27:19.214523: step: 1820/527, loss: 0.17885570228099823
2023-01-21 12:27:20.330491: step: 1824/527, loss: 0.012077808380126953
2023-01-21 12:27:21.413034: step: 1828/527, loss: 0.002914238255470991
2023-01-21 12:27:22.529473: step: 1832/527, loss: 0.0011190414661541581
2023-01-21 12:27:23.611922: step: 1836/527, loss: 0.003102922346442938
2023-01-21 12:27:24.747542: step: 1840/527, loss: 0.0005242824554443359
2023-01-21 12:27:25.869334: step: 1844/527, loss: 0.05102214962244034
2023-01-21 12:27:27.016587: step: 1848/527, loss: 0.013502693735063076
2023-01-21 12:27:28.134920: step: 1852/527, loss: 0.015835857018828392
2023-01-21 12:27:29.258084: step: 1856/527, loss: 0.008668041788041592
2023-01-21 12:27:30.374643: step: 1860/527, loss: 0.0017417907947674394
2023-01-21 12:27:31.487917: step: 1864/527, loss: 0.0009460925939492881
2023-01-21 12:27:32.612480: step: 1868/527, loss: 1.5735626220703125e-05
2023-01-21 12:27:33.744733: step: 1872/527, loss: 0.024711990728974342
2023-01-21 12:27:34.839399: step: 1876/527, loss: 0.008853626437485218
2023-01-21 12:27:35.948778: step: 1880/527, loss: 0.008028030395507812
2023-01-21 12:27:37.082658: step: 1884/527, loss: 6.313324411166832e-05
2023-01-21 12:27:38.196626: step: 1888/527, loss: 0.006133460905402899
2023-01-21 12:27:39.341792: step: 1892/527, loss: 0.03036346472799778
2023-01-21 12:27:40.414525: step: 1896/527, loss: 0.00024700164794921875
2023-01-21 12:27:41.505380: step: 1900/527, loss: 0.006574248895049095
2023-01-21 12:27:42.645902: step: 1904/527, loss: 0.004259156994521618
2023-01-21 12:27:43.751659: step: 1908/527, loss: 0.001476716948673129
2023-01-21 12:27:44.922233: step: 1912/527, loss: 0.01608562469482422
2023-01-21 12:27:46.051904: step: 1916/527, loss: 0.020825671032071114
2023-01-21 12:27:47.186804: step: 1920/527, loss: 0.05935373157262802
2023-01-21 12:27:48.326187: step: 1924/527, loss: 0.0001262664736714214
2023-01-21 12:27:49.415901: step: 1928/527, loss: 0.002974605420604348
2023-01-21 12:27:50.528352: step: 1932/527, loss: 1.3923644473834429e-05
2023-01-21 12:27:51.625357: step: 1936/527, loss: 0.00023336410231422633
2023-01-21 12:27:52.724249: step: 1940/527, loss: 0.0008708000532351434
2023-01-21 12:27:53.849261: step: 1944/527, loss: 6.46591215627268e-05
2023-01-21 12:27:54.946650: step: 1948/527, loss: 0.0011211395030841231
2023-01-21 12:27:56.064706: step: 1952/527, loss: 0.004268551245331764
2023-01-21 12:27:57.154346: step: 1956/527, loss: 0.015630245208740234
2023-01-21 12:27:58.280241: step: 1960/527, loss: 0.010922718793153763
2023-01-21 12:27:59.375155: step: 1964/527, loss: 0.0007017135503701866
2023-01-21 12:28:00.510571: step: 1968/527, loss: 0.0008071899646893144
2023-01-21 12:28:01.603436: step: 1972/527, loss: 0.0003615856112446636
2023-01-21 12:28:02.719929: step: 1976/527, loss: 0.0011043548583984375
2023-01-21 12:28:03.859659: step: 1980/527, loss: 0.017474746331572533
2023-01-21 12:28:04.994901: step: 1984/527, loss: 0.00515251187607646
2023-01-21 12:28:06.104524: step: 1988/527, loss: 0.00406303396448493
2023-01-21 12:28:07.200960: step: 1992/527, loss: 2.7227401005802676e-05
2023-01-21 12:28:08.308965: step: 1996/527, loss: 0.003277254058048129
2023-01-21 12:28:09.415265: step: 2000/527, loss: 9.098053124034777e-05
2023-01-21 12:28:10.522037: step: 2004/527, loss: 0.0605618953704834
2023-01-21 12:28:11.617848: step: 2008/527, loss: 2.6702882678364404e-05
2023-01-21 12:28:12.713040: step: 2012/527, loss: 0.00025215151254087687
2023-01-21 12:28:13.808000: step: 2016/527, loss: 0.0003925323544535786
2023-01-21 12:28:14.901864: step: 2020/527, loss: 0.006507683079689741
2023-01-21 12:28:15.998693: step: 2024/527, loss: 0.0029703141190111637
2023-01-21 12:28:17.111134: step: 2028/527, loss: 0.04039516672492027
2023-01-21 12:28:18.229687: step: 2032/527, loss: 0.010406875982880592
2023-01-21 12:28:19.337566: step: 2036/527, loss: 0.013894462957978249
2023-01-21 12:28:20.408182: step: 2040/527, loss: 5.0067901611328125e-06
2023-01-21 12:28:21.516179: step: 2044/527, loss: 0.00015916825213935226
2023-01-21 12:28:22.652579: step: 2048/527, loss: 7.028579420875758e-05
2023-01-21 12:28:23.747912: step: 2052/527, loss: 3.318786548334174e-05
2023-01-21 12:28:24.876840: step: 2056/527, loss: 0.01747283898293972
2023-01-21 12:28:25.994736: step: 2060/527, loss: 0.01985015906393528
2023-01-21 12:28:27.096167: step: 2064/527, loss: 0.0006216049077920616
2023-01-21 12:28:28.207478: step: 2068/527, loss: 0.0006938934093341231
2023-01-21 12:28:29.357145: step: 2072/527, loss: 5.4168704082258046e-05
2023-01-21 12:28:30.470389: step: 2076/527, loss: 0.03682155907154083
2023-01-21 12:28:31.576867: step: 2080/527, loss: 0.010202979668974876
2023-01-21 12:28:32.686338: step: 2084/527, loss: 0.07242298126220703
2023-01-21 12:28:33.810837: step: 2088/527, loss: 0.04967985302209854
2023-01-21 12:28:34.930266: step: 2092/527, loss: 0.23714673519134521
2023-01-21 12:28:36.049734: step: 2096/527, loss: 0.4237120747566223
2023-01-21 12:28:37.169781: step: 2100/527, loss: 0.025271130725741386
2023-01-21 12:28:38.270965: step: 2104/527, loss: 0.022679520770907402
2023-01-21 12:28:39.417575: step: 2108/527, loss: 0.0006148338434286416
==================================================
Loss: 0.021
--------------------
Dev: {'event': {'p': 0.6010526315789474, 'r': 0.7603195739014648, 'f1': 0.6713697824808936}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22}
Test: {'event': {'p': 0.6353974121996303, 'r': 0.7857142857142857, 'f1': 0.7026060296371998}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22}
Chinese: {'event': {'p': 0.5875, 'r': 0.8703703703703703, 'f1': 0.7014925373134329}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22}
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': 22}
Russian: {'event': {'p': 0.4857142857142857, 'r': 0.4722222222222222, 'f1': 0.47887323943661975}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 12:29:20.902796: step: 4/527, loss: 0.009901237674057484
2023-01-21 12:29:22.010976: step: 8/527, loss: 0.00029506682767532766
2023-01-21 12:29:23.100562: step: 12/527, loss: 0.011362170800566673
2023-01-21 12:29:24.227423: step: 16/527, loss: 0.00234565744176507
2023-01-21 12:29:25.329877: step: 20/527, loss: 0.0014167786575853825
2023-01-21 12:29:26.447518: step: 24/527, loss: 0.02457427978515625
2023-01-21 12:29:27.580555: step: 28/527, loss: 0.10041961073875427
2023-01-21 12:29:28.692640: step: 32/527, loss: 0.30272847414016724
2023-01-21 12:29:29.819480: step: 36/527, loss: 0.0029817582108080387
2023-01-21 12:29:30.905559: step: 40/527, loss: 0.009951401501893997
2023-01-21 12:29:32.041896: step: 44/527, loss: 0.009654236026108265
2023-01-21 12:29:33.126711: step: 48/527, loss: 0.004967212677001953
2023-01-21 12:29:34.237974: step: 52/527, loss: 0.0023446083068847656
2023-01-21 12:29:35.368422: step: 56/527, loss: 0.00047435759915970266
2023-01-21 12:29:36.446511: step: 60/527, loss: 0.003908920101821423
2023-01-21 12:29:37.606467: step: 64/527, loss: 4.978180004400201e-05
2023-01-21 12:29:38.776256: step: 68/527, loss: 0.00013604163541458547
2023-01-21 12:29:39.878119: step: 72/527, loss: 0.01138296164572239
2023-01-21 12:29:40.985765: step: 76/527, loss: 0.0014940262772142887
2023-01-21 12:29:42.119348: step: 80/527, loss: 2.441406286379788e-05
2023-01-21 12:29:43.237348: step: 84/527, loss: 0.0004268646298442036
2023-01-21 12:29:44.343674: step: 88/527, loss: 0.0005876541254110634
2023-01-21 12:29:45.474092: step: 92/527, loss: 0.0018032074440270662
2023-01-21 12:29:46.579204: step: 96/527, loss: 0.0010448455577716231
2023-01-21 12:29:47.699022: step: 100/527, loss: 1.52587890625e-05
2023-01-21 12:29:48.801159: step: 104/527, loss: 0.0008096694946289062
2023-01-21 12:29:49.923072: step: 108/527, loss: 0.0017808914417400956
2023-01-21 12:29:51.015377: step: 112/527, loss: 0.00030155183048918843
2023-01-21 12:29:52.106096: step: 116/527, loss: 0.04763726890087128
2023-01-21 12:29:53.215438: step: 120/527, loss: 0.02572164684534073
2023-01-21 12:29:54.310398: step: 124/527, loss: 0.0004935264587402344
2023-01-21 12:29:55.406824: step: 128/527, loss: 0.00024518967256881297
2023-01-21 12:29:56.506923: step: 132/527, loss: 0.0001848220854299143
2023-01-21 12:29:57.610313: step: 136/527, loss: 0.06323890388011932
2023-01-21 12:29:58.731615: step: 140/527, loss: 4.844665454584174e-05
2023-01-21 12:29:59.858505: step: 144/527, loss: 0.007431411650031805
2023-01-21 12:30:00.983566: step: 148/527, loss: 5.397796485340223e-05
2023-01-21 12:30:02.132307: step: 152/527, loss: 0.01949443854391575
2023-01-21 12:30:03.254824: step: 156/527, loss: 0.030301380902528763
2023-01-21 12:30:04.351895: step: 160/527, loss: 0.0008165359613485634
2023-01-21 12:30:05.499948: step: 164/527, loss: 0.004011916927993298
2023-01-21 12:30:06.609168: step: 168/527, loss: 0.007187986746430397
2023-01-21 12:30:07.702439: step: 172/527, loss: 0.00019750595674850047
2023-01-21 12:30:08.800923: step: 176/527, loss: 2.6321413315599784e-05
2023-01-21 12:30:09.900702: step: 180/527, loss: 0.0015926361083984375
2023-01-21 12:30:11.057183: step: 184/527, loss: -3.051757857974735e-06
2023-01-21 12:30:12.186524: step: 188/527, loss: 0.002456235932186246
2023-01-21 12:30:13.280479: step: 192/527, loss: 0.0038700581062585115
2023-01-21 12:30:14.393186: step: 196/527, loss: 0.004583549685776234
2023-01-21 12:30:15.503971: step: 200/527, loss: 0.0001430511474609375
2023-01-21 12:30:16.592904: step: 204/527, loss: 0.04462156444787979
2023-01-21 12:30:17.694058: step: 208/527, loss: 0.022519491612911224
2023-01-21 12:30:18.821802: step: 212/527, loss: 0.004508304409682751
2023-01-21 12:30:19.940536: step: 216/527, loss: 0.0004117012140341103
2023-01-21 12:30:21.056701: step: 220/527, loss: 0.006398010067641735
2023-01-21 12:30:22.201500: step: 224/527, loss: 8.79287690622732e-05
2023-01-21 12:30:23.320892: step: 228/527, loss: 0.00012488366337493062
2023-01-21 12:30:24.424913: step: 232/527, loss: 0.0002541542053222656
2023-01-21 12:30:25.548244: step: 236/527, loss: 0.0023065567947924137
2023-01-21 12:30:26.656999: step: 240/527, loss: 0.0010856628650799394
2023-01-21 12:30:27.753234: step: 244/527, loss: 0.005016136448830366
2023-01-21 12:30:28.848721: step: 248/527, loss: 0.012585067190229893
2023-01-21 12:30:30.013728: step: 252/527, loss: 0.030712127685546875
2023-01-21 12:30:31.092393: step: 256/527, loss: 0.03115549124777317
2023-01-21 12:30:32.205776: step: 260/527, loss: 0.005206203553825617
2023-01-21 12:30:33.319013: step: 264/527, loss: 0.06761131435632706
2023-01-21 12:30:34.441557: step: 268/527, loss: 0.023745572194457054
2023-01-21 12:30:35.604327: step: 272/527, loss: 0.0018802642589434981
2023-01-21 12:30:36.715719: step: 276/527, loss: 0.009649086743593216
2023-01-21 12:30:37.828665: step: 280/527, loss: 0.006797504145652056
2023-01-21 12:30:38.950395: step: 284/527, loss: 0.006094837561249733
2023-01-21 12:30:40.074554: step: 288/527, loss: 0.000583267246838659
2023-01-21 12:30:41.182733: step: 292/527, loss: -4.9591067181609105e-06
2023-01-21 12:30:42.321704: step: 296/527, loss: 0.010476112365722656
2023-01-21 12:30:43.448948: step: 300/527, loss: 0.024883801117539406
2023-01-21 12:30:44.538746: step: 304/527, loss: 0.00036878586979582906
2023-01-21 12:30:45.664199: step: 308/527, loss: 0.007901382632553577
2023-01-21 12:30:46.785588: step: 312/527, loss: 0.09342136234045029
2023-01-21 12:30:47.875495: step: 316/527, loss: 1.3256073543743696e-05
2023-01-21 12:30:49.015337: step: 320/527, loss: 0.0015665054088458419
2023-01-21 12:30:50.108973: step: 324/527, loss: 0.00010986328561557457
2023-01-21 12:30:51.223811: step: 328/527, loss: 0.2275102585554123
2023-01-21 12:30:52.338494: step: 332/527, loss: 0.010584068484604359
2023-01-21 12:30:53.461066: step: 336/527, loss: 0.0490722693502903
2023-01-21 12:30:54.590952: step: 340/527, loss: 0.004690027330070734
2023-01-21 12:30:55.721901: step: 344/527, loss: 0.042484235018491745
2023-01-21 12:30:56.824655: step: 348/527, loss: 0.01852741278707981
2023-01-21 12:30:57.919819: step: 352/527, loss: 0.0001046180768753402
2023-01-21 12:30:59.036086: step: 356/527, loss: 0.00253639230504632
2023-01-21 12:31:00.139654: step: 360/527, loss: 0.005726623814553022
2023-01-21 12:31:01.261778: step: 364/527, loss: 0.0002918243408203125
2023-01-21 12:31:02.376226: step: 368/527, loss: 0.013079452328383923
2023-01-21 12:31:03.495907: step: 372/527, loss: 0.0007347107166424394
2023-01-21 12:31:04.611133: step: 376/527, loss: 2.002716064453125e-05
2023-01-21 12:31:05.715999: step: 380/527, loss: 0.021936701610684395
2023-01-21 12:31:06.829197: step: 384/527, loss: 0.0001683235022937879
2023-01-21 12:31:07.933364: step: 388/527, loss: 0.022719383239746094
2023-01-21 12:31:09.030094: step: 392/527, loss: 0.0021144866477698088
2023-01-21 12:31:10.142703: step: 396/527, loss: 1.1205673217773438e-05
2023-01-21 12:31:11.253815: step: 400/527, loss: 0.002982711885124445
2023-01-21 12:31:12.382580: step: 404/527, loss: 0.0012072563404217362
2023-01-21 12:31:13.524561: step: 408/527, loss: 0.03755536302924156
2023-01-21 12:31:14.625639: step: 412/527, loss: 0.015459251590073109
2023-01-21 12:31:15.808086: step: 416/527, loss: 0.006657600402832031
2023-01-21 12:31:16.913127: step: 420/527, loss: 0.00037388800410553813
2023-01-21 12:31:18.019968: step: 424/527, loss: 0.017155839130282402
2023-01-21 12:31:19.108883: step: 428/527, loss: 0.9744386672973633
2023-01-21 12:31:20.228582: step: 432/527, loss: 0.00257701869122684
2023-01-21 12:31:21.325319: step: 436/527, loss: 4.825591895496473e-05
2023-01-21 12:31:22.434797: step: 440/527, loss: 0.32267218828201294
2023-01-21 12:31:23.549284: step: 444/527, loss: 0.026952551677823067
2023-01-21 12:31:24.659967: step: 448/527, loss: 0.00016851424879860133
2023-01-21 12:31:25.777292: step: 452/527, loss: 0.0002056121884379536
2023-01-21 12:31:26.886606: step: 456/527, loss: 0.004977417178452015
2023-01-21 12:31:28.006662: step: 460/527, loss: 2.307891918462701e-05
2023-01-21 12:31:29.124350: step: 464/527, loss: 0.003697967389598489
2023-01-21 12:31:30.241284: step: 468/527, loss: 0.00037741661071777344
2023-01-21 12:31:31.364083: step: 472/527, loss: 0.003179276129230857
2023-01-21 12:31:32.533748: step: 476/527, loss: 0.0006736755603924394
2023-01-21 12:31:33.680325: step: 480/527, loss: 0.017903519794344902
2023-01-21 12:31:34.784247: step: 484/527, loss: 0.000255393999395892
2023-01-21 12:31:35.871539: step: 488/527, loss: 0.04050245136022568
2023-01-21 12:31:36.987913: step: 492/527, loss: 0.025121737271547318
2023-01-21 12:31:38.096346: step: 496/527, loss: 0.0025573731400072575
2023-01-21 12:31:39.239318: step: 500/527, loss: 0.1347423940896988
2023-01-21 12:31:40.381380: step: 504/527, loss: 0.026486970484256744
2023-01-21 12:31:41.494308: step: 508/527, loss: 0.020560551434755325
2023-01-21 12:31:42.611661: step: 512/527, loss: 0.01280517689883709
2023-01-21 12:31:43.734227: step: 516/527, loss: 0.009318160824477673
2023-01-21 12:31:44.869906: step: 520/527, loss: 0.08470630645751953
2023-01-21 12:31:45.979762: step: 524/527, loss: 0.0004506111145019531
2023-01-21 12:31:47.123893: step: 528/527, loss: 1.4495850336970761e-05
2023-01-21 12:31:48.234914: step: 532/527, loss: 0.0031223297119140625
2023-01-21 12:31:49.358989: step: 536/527, loss: 0.001815033028833568
2023-01-21 12:31:50.512486: step: 540/527, loss: 0.2849872410297394
2023-01-21 12:31:51.666315: step: 544/527, loss: 0.002042627427726984
2023-01-21 12:31:52.773171: step: 548/527, loss: 7.805824134266004e-05
2023-01-21 12:31:53.876602: step: 552/527, loss: 0.009734439663589
2023-01-21 12:31:54.986069: step: 556/527, loss: 0.011613654904067516
2023-01-21 12:31:56.093287: step: 560/527, loss: 0.01763901859521866
2023-01-21 12:31:57.232212: step: 564/527, loss: 0.00961456261575222
2023-01-21 12:31:58.330206: step: 568/527, loss: 0.00048232078552246094
2023-01-21 12:31:59.456477: step: 572/527, loss: 0.0001677513209870085
2023-01-21 12:32:00.566509: step: 576/527, loss: 0.1305990219116211
2023-01-21 12:32:01.655932: step: 580/527, loss: 0.5182012319564819
2023-01-21 12:32:02.789860: step: 584/527, loss: 0.008178330026566982
2023-01-21 12:32:03.925874: step: 588/527, loss: 0.011240148916840553
2023-01-21 12:32:05.053321: step: 592/527, loss: 1.144409225162235e-06
2023-01-21 12:32:06.172439: step: 596/527, loss: 0.000789260899182409
2023-01-21 12:32:07.319434: step: 600/527, loss: 0.038184165954589844
2023-01-21 12:32:08.473759: step: 604/527, loss: 0.005100298207253218
2023-01-21 12:32:09.606612: step: 608/527, loss: 1.1539459592313506e-05
2023-01-21 12:32:10.722240: step: 612/527, loss: 0.05793943256139755
2023-01-21 12:32:11.811019: step: 616/527, loss: 0.0009870529174804688
2023-01-21 12:32:12.902688: step: 620/527, loss: 0.00041837693424895406
2023-01-21 12:32:14.028364: step: 624/527, loss: 0.05193343386054039
2023-01-21 12:32:15.159911: step: 628/527, loss: 0.5337886810302734
2023-01-21 12:32:16.275932: step: 632/527, loss: 0.0038826942909508944
2023-01-21 12:32:17.408553: step: 636/527, loss: 4.472732325666584e-05
2023-01-21 12:32:18.532157: step: 640/527, loss: 0.0014015197521075606
2023-01-21 12:32:19.642186: step: 644/527, loss: 0.00017395020404364914
2023-01-21 12:32:20.748814: step: 648/527, loss: 0.00020732879056595266
2023-01-21 12:32:21.867094: step: 652/527, loss: 0.00011711120896507055
2023-01-21 12:32:22.989495: step: 656/527, loss: 0.0033273696899414062
2023-01-21 12:32:24.133021: step: 660/527, loss: 0.029459573328495026
2023-01-21 12:32:25.214984: step: 664/527, loss: 0.01702442206442356
2023-01-21 12:32:26.306452: step: 668/527, loss: 0.03598365932703018
2023-01-21 12:32:27.401435: step: 672/527, loss: 0.009150505065917969
2023-01-21 12:32:28.521169: step: 676/527, loss: 0.003360843751579523
2023-01-21 12:32:29.638592: step: 680/527, loss: 0.008288050070405006
2023-01-21 12:32:30.723845: step: 684/527, loss: 7.62939453125e-06
2023-01-21 12:32:31.836877: step: 688/527, loss: 0.031882286071777344
2023-01-21 12:32:32.950867: step: 692/527, loss: 0.02600417099893093
2023-01-21 12:32:34.077674: step: 696/527, loss: 0.008329391479492188
2023-01-21 12:32:35.182857: step: 700/527, loss: 0.0022581100929528475
2023-01-21 12:32:36.317011: step: 704/527, loss: 0.04228067398071289
2023-01-21 12:32:37.471609: step: 708/527, loss: 0.00466203736141324
2023-01-21 12:32:38.603958: step: 712/527, loss: 0.0011475563514977694
2023-01-21 12:32:39.681831: step: 716/527, loss: 0.0196685791015625
2023-01-21 12:32:40.822041: step: 720/527, loss: 0.0029592516366392374
2023-01-21 12:32:41.932334: step: 724/527, loss: 0.00022468566021416336
2023-01-21 12:32:43.040683: step: 728/527, loss: 0.004058646969497204
2023-01-21 12:32:44.181259: step: 732/527, loss: 0.0015620231861248612
2023-01-21 12:32:45.299409: step: 736/527, loss: 3.8909915019758046e-05
2023-01-21 12:32:46.390538: step: 740/527, loss: 0.2221468985080719
2023-01-21 12:32:47.511916: step: 744/527, loss: 0.0001697540283203125
2023-01-21 12:32:48.655766: step: 748/527, loss: 0.018655015155673027
2023-01-21 12:32:49.749032: step: 752/527, loss: 0.022371292114257812
2023-01-21 12:32:50.840386: step: 756/527, loss: 2.689361645025201e-05
2023-01-21 12:32:51.971059: step: 760/527, loss: 0.00014228820509742945
2023-01-21 12:32:53.101539: step: 764/527, loss: 6.246566954359878e-06
2023-01-21 12:32:54.217453: step: 768/527, loss: 9.1552734375e-05
2023-01-21 12:32:55.331400: step: 772/527, loss: 0.000668430351652205
2023-01-21 12:32:56.482026: step: 776/527, loss: 0.04863777384161949
2023-01-21 12:32:57.592151: step: 780/527, loss: 0.009463215246796608
2023-01-21 12:32:58.697613: step: 784/527, loss: 0.00013408661470748484
2023-01-21 12:32:59.804556: step: 788/527, loss: 7.53879503463395e-05
2023-01-21 12:33:00.919917: step: 792/527, loss: 0.0006320953834801912
2023-01-21 12:33:02.061938: step: 796/527, loss: 0.0005182266468182206
2023-01-21 12:33:03.215985: step: 800/527, loss: 0.018663501366972923
2023-01-21 12:33:04.299389: step: 804/527, loss: 0.023381328210234642
2023-01-21 12:33:05.410349: step: 808/527, loss: 9.670257713878527e-05
2023-01-21 12:33:06.520110: step: 812/527, loss: 0.009738064371049404
2023-01-21 12:33:07.614889: step: 816/527, loss: 0.0011090278858318925
2023-01-21 12:33:08.728479: step: 820/527, loss: 0.024940870702266693
2023-01-21 12:33:09.858573: step: 824/527, loss: 0.00779304513707757
2023-01-21 12:33:10.953461: step: 828/527, loss: 0.007205199915915728
2023-01-21 12:33:12.059044: step: 832/527, loss: 3.347396705066785e-05
2023-01-21 12:33:13.162438: step: 836/527, loss: 0.007741928566247225
2023-01-21 12:33:14.259426: step: 840/527, loss: 2.2315980459097773e-05
2023-01-21 12:33:15.381725: step: 844/527, loss: 0.06525001674890518
2023-01-21 12:33:16.490969: step: 848/527, loss: 0.008314132690429688
2023-01-21 12:33:17.596307: step: 852/527, loss: 2.5987625122070312e-05
2023-01-21 12:33:18.720979: step: 856/527, loss: 0.015436649322509766
2023-01-21 12:33:19.874491: step: 860/527, loss: 0.021979333832859993
2023-01-21 12:33:21.024106: step: 864/527, loss: 0.001811218331567943
2023-01-21 12:33:22.141962: step: 868/527, loss: 0.005060672760009766
2023-01-21 12:33:23.228489: step: 872/527, loss: 0.0005823135143145919
2023-01-21 12:33:24.326627: step: 876/527, loss: 0.0008657455327920616
2023-01-21 12:33:25.460650: step: 880/527, loss: 0.017195511609315872
2023-01-21 12:33:26.570565: step: 884/527, loss: 0.0010107994312420487
2023-01-21 12:33:27.679754: step: 888/527, loss: 0.004335975740104914
2023-01-21 12:33:28.781272: step: 892/527, loss: 0.0005720138433389366
2023-01-21 12:33:29.893311: step: 896/527, loss: 0.0061013223603367805
2023-01-21 12:33:31.014986: step: 900/527, loss: 0.00035772324190475047
2023-01-21 12:33:32.156022: step: 904/527, loss: 0.0610651969909668
2023-01-21 12:33:33.233443: step: 908/527, loss: 1.7070769899873994e-05
2023-01-21 12:33:34.370156: step: 912/527, loss: 0.017061617225408554
2023-01-21 12:33:35.461207: step: 916/527, loss: 0.00043668749276548624
2023-01-21 12:33:36.552984: step: 920/527, loss: 0.024418117478489876
2023-01-21 12:33:37.631947: step: 924/527, loss: 0.0016036033630371094
2023-01-21 12:33:38.840644: step: 928/527, loss: 0.0005802154773846269
2023-01-21 12:33:39.961386: step: 932/527, loss: 0.0049932003021240234
2023-01-21 12:33:41.069645: step: 936/527, loss: 0.0005226611974649131
2023-01-21 12:33:42.185215: step: 940/527, loss: 0.00019984245591331273
2023-01-21 12:33:43.297313: step: 944/527, loss: 7.734298560535535e-05
2023-01-21 12:33:44.387272: step: 948/527, loss: 0.0006738663068972528
2023-01-21 12:33:45.517802: step: 952/527, loss: 0.03094034269452095
2023-01-21 12:33:46.622553: step: 956/527, loss: 0.03349189832806587
2023-01-21 12:33:47.703891: step: 960/527, loss: 0.025272751227021217
2023-01-21 12:33:48.848537: step: 964/527, loss: 0.005921745672821999
2023-01-21 12:33:49.957896: step: 968/527, loss: 0.00013170242891646922
2023-01-21 12:33:51.070981: step: 972/527, loss: 0.002598381135612726
2023-01-21 12:33:52.200096: step: 976/527, loss: 0.00021800995455123484
2023-01-21 12:33:53.318911: step: 980/527, loss: 0.00479049701243639
2023-01-21 12:33:54.422063: step: 984/527, loss: 0.012603998184204102
2023-01-21 12:33:55.546424: step: 988/527, loss: 0.0005575180402956903
2023-01-21 12:33:56.665087: step: 992/527, loss: 0.0004016876337118447
2023-01-21 12:33:57.766028: step: 996/527, loss: 0.025344902649521828
2023-01-21 12:33:58.877788: step: 1000/527, loss: 0.0005071639898233116
2023-01-21 12:34:00.011978: step: 1004/527, loss: 0.003490447998046875
2023-01-21 12:34:01.139964: step: 1008/527, loss: 0.004206848330795765
2023-01-21 12:34:02.270302: step: 1012/527, loss: 0.008918190374970436
2023-01-21 12:34:03.425909: step: 1016/527, loss: 0.05567416921257973
2023-01-21 12:34:04.495149: step: 1020/527, loss: 0.01316604670137167
2023-01-21 12:34:05.622086: step: 1024/527, loss: 0.00023536683875136077
2023-01-21 12:34:06.757626: step: 1028/527, loss: 0.0027950287330895662
2023-01-21 12:34:07.859795: step: 1032/527, loss: 0.002307796385139227
2023-01-21 12:34:08.960659: step: 1036/527, loss: 3.566742088878527e-05
2023-01-21 12:34:10.088104: step: 1040/527, loss: 0.004254436586052179
2023-01-21 12:34:11.227593: step: 1044/527, loss: -2.846717688953504e-05
2023-01-21 12:34:12.357078: step: 1048/527, loss: 0.003581428434699774
2023-01-21 12:34:13.510968: step: 1052/527, loss: 0.0429597869515419
2023-01-21 12:34:14.605107: step: 1056/527, loss: 0.0001621246337890625
2023-01-21 12:34:15.741334: step: 1060/527, loss: 0.0004070282157044858
2023-01-21 12:34:16.867899: step: 1064/527, loss: 0.0016332149971276522
2023-01-21 12:34:17.949304: step: 1068/527, loss: 6.189346458995715e-05
2023-01-21 12:34:19.061165: step: 1072/527, loss: 9.160042100120336e-05
2023-01-21 12:34:20.183999: step: 1076/527, loss: 0.013260078616440296
2023-01-21 12:34:21.281833: step: 1080/527, loss: 0.0001428604155080393
2023-01-21 12:34:22.375189: step: 1084/527, loss: 0.008313274942338467
2023-01-21 12:34:23.485822: step: 1088/527, loss: 0.024941731244325638
2023-01-21 12:34:24.600322: step: 1092/527, loss: 0.009051322937011719
2023-01-21 12:34:25.713589: step: 1096/527, loss: 0.00039281847421079874
2023-01-21 12:34:26.811202: step: 1100/527, loss: 0.00046443939208984375
2023-01-21 12:34:27.948492: step: 1104/527, loss: 0.00027065275935456157
2023-01-21 12:34:29.055872: step: 1108/527, loss: 0.03258323669433594
2023-01-21 12:34:30.175278: step: 1112/527, loss: 0.13284483551979065
2023-01-21 12:34:31.313232: step: 1116/527, loss: 0.058045580983161926
2023-01-21 12:34:32.423767: step: 1120/527, loss: 4.7683710135970614e-07
2023-01-21 12:34:33.529726: step: 1124/527, loss: 0.0050223348662257195
2023-01-21 12:34:34.637131: step: 1128/527, loss: 0.002240181202068925
2023-01-21 12:34:35.748899: step: 1132/527, loss: 0.0042243958450853825
2023-01-21 12:34:36.856448: step: 1136/527, loss: 0.017250681295990944
2023-01-21 12:34:37.970313: step: 1140/527, loss: 0.04178063943982124
2023-01-21 12:34:39.065260: step: 1144/527, loss: 2.059936559817288e-05
2023-01-21 12:34:40.193869: step: 1148/527, loss: 0.0011625289916992188
2023-01-21 12:34:41.313950: step: 1152/527, loss: 0.0033068659249693155
2023-01-21 12:34:42.412770: step: 1156/527, loss: 0.00032396314782090485
2023-01-21 12:34:43.504195: step: 1160/527, loss: 0.0016888618702068925
2023-01-21 12:34:44.617969: step: 1164/527, loss: 0.02366485819220543
2023-01-21 12:34:45.720251: step: 1168/527, loss: 0.0009849548805505037
2023-01-21 12:34:46.813053: step: 1172/527, loss: 0.02193584479391575
2023-01-21 12:34:47.892391: step: 1176/527, loss: 0.0001876354217529297
2023-01-21 12:34:49.020578: step: 1180/527, loss: 0.000842189765535295
2023-01-21 12:34:50.118680: step: 1184/527, loss: 0.0014688492519780993
2023-01-21 12:34:51.260914: step: 1188/527, loss: 0.0004326820489950478
2023-01-21 12:34:52.368485: step: 1192/527, loss: 0.0015844345325604081
2023-01-21 12:34:53.482767: step: 1196/527, loss: 0.0005743980291299522
2023-01-21 12:34:54.620449: step: 1200/527, loss: 0.012994195334613323
2023-01-21 12:34:55.741155: step: 1204/527, loss: 0.005030632019042969
2023-01-21 12:34:56.875411: step: 1208/527, loss: 0.0003017425478901714
2023-01-21 12:34:57.979521: step: 1212/527, loss: 0.0003208160342182964
2023-01-21 12:34:59.075958: step: 1216/527, loss: 0.0008462906116619706
2023-01-21 12:35:00.203973: step: 1220/527, loss: 0.004182625096291304
2023-01-21 12:35:01.318368: step: 1224/527, loss: 0.14578190445899963
2023-01-21 12:35:02.435736: step: 1228/527, loss: 7.62939453125e-06
2023-01-21 12:35:03.543476: step: 1232/527, loss: 0.3640851080417633
2023-01-21 12:35:04.629028: step: 1236/527, loss: 0.0024131773971021175
2023-01-21 12:35:05.730987: step: 1240/527, loss: 0.014840316958725452
2023-01-21 12:35:06.840856: step: 1244/527, loss: 0.004340744111686945
2023-01-21 12:35:07.955701: step: 1248/527, loss: 0.01746845245361328
2023-01-21 12:35:09.093069: step: 1252/527, loss: 0.0006083488697186112
2023-01-21 12:35:10.181180: step: 1256/527, loss: 0.0009681701194494963
2023-01-21 12:35:11.303372: step: 1260/527, loss: 0.00046262741670943797
2023-01-21 12:35:12.416523: step: 1264/527, loss: 0.000327634799759835
2023-01-21 12:35:13.532803: step: 1268/527, loss: 0.06470489501953125
2023-01-21 12:35:14.650911: step: 1272/527, loss: 0.0003659248468466103
2023-01-21 12:35:15.747664: step: 1276/527, loss: 6.95705457474105e-05
2023-01-21 12:35:16.851497: step: 1280/527, loss: 0.005387497134506702
2023-01-21 12:35:17.948931: step: 1284/527, loss: 0.0010467530228197575
2023-01-21 12:35:19.082800: step: 1288/527, loss: 0.009755134582519531
2023-01-21 12:35:20.234903: step: 1292/527, loss: 0.04974942281842232
2023-01-21 12:35:21.329599: step: 1296/527, loss: 0.00016479492478538305
2023-01-21 12:35:22.429533: step: 1300/527, loss: 4.0626528061693534e-05
2023-01-21 12:35:23.550290: step: 1304/527, loss: 0.0004245758173055947
2023-01-21 12:35:24.678644: step: 1308/527, loss: 8.406639244640246e-05
2023-01-21 12:35:25.808818: step: 1312/527, loss: 0.0012098312145099044
2023-01-21 12:35:26.919725: step: 1316/527, loss: 0.0032777786254882812
2023-01-21 12:35:28.061899: step: 1320/527, loss: 0.020536424592137337
2023-01-21 12:35:29.169837: step: 1324/527, loss: 0.00918130949139595
2023-01-21 12:35:30.274865: step: 1328/527, loss: 0.0006234169122762978
2023-01-21 12:35:31.387731: step: 1332/527, loss: 0.0008804321405477822
2023-01-21 12:35:32.496104: step: 1336/527, loss: 0.28104686737060547
2023-01-21 12:35:33.576694: step: 1340/527, loss: 0.07377391308546066
2023-01-21 12:35:34.673567: step: 1344/527, loss: 0.0016926765674725175
2023-01-21 12:35:35.751824: step: 1348/527, loss: 0.01854095607995987
2023-01-21 12:35:36.842330: step: 1352/527, loss: 0.0393003486096859
2023-01-21 12:35:37.959400: step: 1356/527, loss: 0.003024578094482422
2023-01-21 12:35:39.061375: step: 1360/527, loss: 0.01980285719037056
2023-01-21 12:35:40.158944: step: 1364/527, loss: 0.007147598080337048
2023-01-21 12:35:41.300576: step: 1368/527, loss: 9.546280489303172e-05
2023-01-21 12:35:42.424579: step: 1372/527, loss: 0.004654502961784601
2023-01-21 12:35:43.589654: step: 1376/527, loss: 0.0020509720779955387
2023-01-21 12:35:44.694763: step: 1380/527, loss: 0.002462244126945734
2023-01-21 12:35:45.823613: step: 1384/527, loss: 5.2833554946118966e-05
2023-01-21 12:35:46.952795: step: 1388/527, loss: 0.04148240387439728
2023-01-21 12:35:48.061741: step: 1392/527, loss: 0.04448547214269638
2023-01-21 12:35:49.168494: step: 1396/527, loss: 0.00029735564021393657
2023-01-21 12:35:50.239631: step: 1400/527, loss: 0.016714954748749733
2023-01-21 12:35:51.342346: step: 1404/527, loss: 0.00013999939255882055
2023-01-21 12:35:52.460626: step: 1408/527, loss: 0.02235722541809082
2023-01-21 12:35:53.578065: step: 1412/527, loss: 0.014533234760165215
2023-01-21 12:35:54.672673: step: 1416/527, loss: 7.815361459506676e-05
2023-01-21 12:35:55.789531: step: 1420/527, loss: 0.05853080749511719
2023-01-21 12:35:56.947967: step: 1424/527, loss: 0.011784744448959827
2023-01-21 12:35:58.081979: step: 1428/527, loss: 0.012887001037597656
2023-01-21 12:35:59.259619: step: 1432/527, loss: 0.0005016326904296875
2023-01-21 12:36:00.373254: step: 1436/527, loss: 0.022765539586544037
2023-01-21 12:36:01.459956: step: 1440/527, loss: 8.163452002918348e-05
2023-01-21 12:36:02.578358: step: 1444/527, loss: 1.3637541997013614e-05
2023-01-21 12:36:03.689219: step: 1448/527, loss: 0.005961514078080654
2023-01-21 12:36:04.796622: step: 1452/527, loss: 0.022536564618349075
2023-01-21 12:36:05.901291: step: 1456/527, loss: 0.005604791920632124
2023-01-21 12:36:07.026312: step: 1460/527, loss: 2.9277802241267636e-05
2023-01-21 12:36:08.121641: step: 1464/527, loss: 0.0050637247040867805
2023-01-21 12:36:09.237849: step: 1468/527, loss: 0.023099135607481003
2023-01-21 12:36:10.352901: step: 1472/527, loss: 0.010835266672074795
2023-01-21 12:36:11.462984: step: 1476/527, loss: 0.00014810562424827367
2023-01-21 12:36:12.586896: step: 1480/527, loss: 0.012093067169189453
2023-01-21 12:36:13.710226: step: 1484/527, loss: 7.090569124557078e-05
2023-01-21 12:36:14.828141: step: 1488/527, loss: 4.730224463855848e-05
2023-01-21 12:36:15.935488: step: 1492/527, loss: 4.425048973644152e-05
2023-01-21 12:36:17.039928: step: 1496/527, loss: 9.474754187976941e-05
2023-01-21 12:36:18.178331: step: 1500/527, loss: 0.0016165734268724918
2023-01-21 12:36:19.282287: step: 1504/527, loss: 0.0005517005920410156
2023-01-21 12:36:20.409391: step: 1508/527, loss: 0.0015367508167400956
2023-01-21 12:36:21.520792: step: 1512/527, loss: 0.0001291275111725554
2023-01-21 12:36:22.625394: step: 1516/527, loss: 0.05298595875501633
2023-01-21 12:36:23.736693: step: 1520/527, loss: 0.0010730742942541838
2023-01-21 12:36:24.852400: step: 1524/527, loss: 0.0013320923317223787
2023-01-21 12:36:25.974208: step: 1528/527, loss: 0.00012006760516669601
2023-01-21 12:36:27.086287: step: 1532/527, loss: 0.0010648728348314762
2023-01-21 12:36:28.190749: step: 1536/527, loss: 7.534027099609375e-05
2023-01-21 12:36:29.279759: step: 1540/527, loss: 0.009952736087143421
2023-01-21 12:36:30.386667: step: 1544/527, loss: 0.0007658004760742188
2023-01-21 12:36:31.491961: step: 1548/527, loss: 0.05119170993566513
2023-01-21 12:36:32.643836: step: 1552/527, loss: 0.008610057644546032
2023-01-21 12:36:33.756397: step: 1556/527, loss: 1.9931794668082148e-05
2023-01-21 12:36:34.860295: step: 1560/527, loss: 0.0005798340425826609
2023-01-21 12:36:35.980568: step: 1564/527, loss: 0.0007037163013592362
2023-01-21 12:36:37.078999: step: 1568/527, loss: 0.0017225266201421618
2023-01-21 12:36:38.223839: step: 1572/527, loss: 0.013992118649184704
2023-01-21 12:36:39.347498: step: 1576/527, loss: 0.0002193450927734375
2023-01-21 12:36:40.461908: step: 1580/527, loss: 0.006794357672333717
2023-01-21 12:36:41.552050: step: 1584/527, loss: 0.00011739730689441785
2023-01-21 12:36:42.655625: step: 1588/527, loss: 0.0023143768776208162
2023-01-21 12:36:43.736426: step: 1592/527, loss: 3.156661841785535e-05
2023-01-21 12:36:44.837570: step: 1596/527, loss: 0.005384969525039196
2023-01-21 12:36:45.924424: step: 1600/527, loss: 0.0023557664826512337
2023-01-21 12:36:47.021850: step: 1604/527, loss: 0.00022239684767555445
2023-01-21 12:36:48.137488: step: 1608/527, loss: 0.0004681587452068925
2023-01-21 12:36:49.235878: step: 1612/527, loss: 0.01723327860236168
2023-01-21 12:36:50.400629: step: 1616/527, loss: 0.01791858673095703
2023-01-21 12:36:51.522789: step: 1620/527, loss: 0.02886676974594593
2023-01-21 12:36:52.656889: step: 1624/527, loss: 0.011512375436723232
2023-01-21 12:36:53.754538: step: 1628/527, loss: 0.014509772881865501
2023-01-21 12:36:54.880482: step: 1632/527, loss: 0.03004760853946209
2023-01-21 12:36:55.966452: step: 1636/527, loss: 0.00030193329439498484
2023-01-21 12:36:57.063133: step: 1640/527, loss: 0.00290088658221066
2023-01-21 12:36:58.144912: step: 1644/527, loss: 0.012646389193832874
2023-01-21 12:36:59.266112: step: 1648/527, loss: 0.05897979810833931
2023-01-21 12:37:00.345394: step: 1652/527, loss: 3.051757857974735e-06
2023-01-21 12:37:01.470313: step: 1656/527, loss: 0.02617206610739231
2023-01-21 12:37:02.581788: step: 1660/527, loss: 0.014951134100556374
2023-01-21 12:37:03.713514: step: 1664/527, loss: 0.0002899169921875
2023-01-21 12:37:04.809455: step: 1668/527, loss: 0.019494343549013138
2023-01-21 12:37:05.954403: step: 1672/527, loss: 0.000683689140714705
2023-01-21 12:37:07.056755: step: 1676/527, loss: 0.010310744866728783
2023-01-21 12:37:08.160437: step: 1680/527, loss: 0.00010395050048828125
2023-01-21 12:37:09.293980: step: 1684/527, loss: 0.00030202866764739156
2023-01-21 12:37:10.383270: step: 1688/527, loss: 0.004313755314797163
2023-01-21 12:37:11.493127: step: 1692/527, loss: 0.026384973898530006
2023-01-21 12:37:12.605788: step: 1696/527, loss: 0.00076379778329283
2023-01-21 12:37:13.728450: step: 1700/527, loss: 0.00023665429034736007
2023-01-21 12:37:14.846370: step: 1704/527, loss: 0.0001924514799611643
2023-01-21 12:37:15.942294: step: 1708/527, loss: 0.0020292282570153475
2023-01-21 12:37:17.030136: step: 1712/527, loss: 0.09808941185474396
2023-01-21 12:37:18.160199: step: 1716/527, loss: 0.0018439769046381116
2023-01-21 12:37:19.279964: step: 1720/527, loss: 0.00846626702696085
2023-01-21 12:37:20.405008: step: 1724/527, loss: 0.0005520820850506425
2023-01-21 12:37:21.551331: step: 1728/527, loss: 0.0039535523392260075
2023-01-21 12:37:22.669269: step: 1732/527, loss: 0.0019207954173907638
2023-01-21 12:37:23.759481: step: 1736/527, loss: 0.009162330999970436
2023-01-21 12:37:24.861188: step: 1740/527, loss: 0.0017773628933355212
2023-01-21 12:37:25.980768: step: 1744/527, loss: 0.0011793136363849044
2023-01-21 12:37:27.088328: step: 1748/527, loss: 0.0002019882231252268
2023-01-21 12:37:28.179415: step: 1752/527, loss: 0.7095305919647217
2023-01-21 12:37:29.281154: step: 1756/527, loss: 1.7547608877066523e-05
2023-01-21 12:37:30.396278: step: 1760/527, loss: 0.0032100677490234375
2023-01-21 12:37:31.524945: step: 1764/527, loss: 0.03240266069769859
2023-01-21 12:37:32.644291: step: 1768/527, loss: 0.018138503655791283
2023-01-21 12:37:33.745821: step: 1772/527, loss: 7.62939453125e-06
2023-01-21 12:37:34.859729: step: 1776/527, loss: 0.00108251569326967
2023-01-21 12:37:36.031250: step: 1780/527, loss: 0.003858471056446433
2023-01-21 12:37:37.167120: step: 1784/527, loss: 0.00011882782564498484
2023-01-21 12:37:38.272407: step: 1788/527, loss: 0.00010032654245151207
2023-01-21 12:37:39.376085: step: 1792/527, loss: 0.00035119056701660156
2023-01-21 12:37:40.505848: step: 1796/527, loss: 0.3447505831718445
2023-01-21 12:37:41.610954: step: 1800/527, loss: 8.010864803509321e-06
2023-01-21 12:37:42.712910: step: 1804/527, loss: 0.00012130737013649195
2023-01-21 12:37:43.812140: step: 1808/527, loss: 0.0038904191460460424
2023-01-21 12:37:44.910066: step: 1812/527, loss: 3.10870361328125
2023-01-21 12:37:46.020086: step: 1816/527, loss: 0.003504037857055664
2023-01-21 12:37:47.151762: step: 1820/527, loss: 0.00035572052001953125
2023-01-21 12:37:48.262456: step: 1824/527, loss: 0.0002574920654296875
2023-01-21 12:37:49.400842: step: 1828/527, loss: 0.0004428863467182964
2023-01-21 12:37:50.488043: step: 1832/527, loss: 0.005035591311752796
2023-01-21 12:37:51.592844: step: 1836/527, loss: 0.008076191879808903
2023-01-21 12:37:52.711123: step: 1840/527, loss: 0.05245323106646538
2023-01-21 12:37:53.811881: step: 1844/527, loss: 0.024606704711914062
2023-01-21 12:37:54.928913: step: 1848/527, loss: 0.004636859986931086
2023-01-21 12:37:56.018292: step: 1852/527, loss: 1.0204315003647935e-05
2023-01-21 12:37:57.118313: step: 1856/527, loss: 0.013503646478056908
2023-01-21 12:37:58.257835: step: 1860/527, loss: 0.0010825158096849918
2023-01-21 12:37:59.368422: step: 1864/527, loss: 0.0010453223949298263
2023-01-21 12:38:00.485837: step: 1868/527, loss: 0.004374981392174959
2023-01-21 12:38:01.597972: step: 1872/527, loss: 0.006274318788200617
2023-01-21 12:38:02.702481: step: 1876/527, loss: 0.0005322456127032638
2023-01-21 12:38:03.808553: step: 1880/527, loss: 0.005789279937744141
2023-01-21 12:38:04.958363: step: 1884/527, loss: 0.10395793616771698
2023-01-21 12:38:06.058704: step: 1888/527, loss: 0.0005834579933434725
2023-01-21 12:38:07.158601: step: 1892/527, loss: 0.0004730224609375
2023-01-21 12:38:08.250908: step: 1896/527, loss: 0.015216255560517311
2023-01-21 12:38:09.383912: step: 1900/527, loss: 0.00018506050400901586
2023-01-21 12:38:10.489956: step: 1904/527, loss: 0.0008146286127157509
2023-01-21 12:38:11.571884: step: 1908/527, loss: 2.2268295651883818e-05
2023-01-21 12:38:12.695704: step: 1912/527, loss: 0.0007706641918048263
2023-01-21 12:38:13.804487: step: 1916/527, loss: 0.012499618344008923
2023-01-21 12:38:14.916800: step: 1920/527, loss: 0.017778636887669563
2023-01-21 12:38:16.034912: step: 1924/527, loss: 8.296966552734375e-05
2023-01-21 12:38:17.147994: step: 1928/527, loss: 0.04794588312506676
2023-01-21 12:38:18.286145: step: 1932/527, loss: 0.18900719285011292
2023-01-21 12:38:19.424157: step: 1936/527, loss: 0.4262973666191101
2023-01-21 12:38:20.563939: step: 1940/527, loss: 0.009500885382294655
2023-01-21 12:38:21.676330: step: 1944/527, loss: 0.0010272025829181075
2023-01-21 12:38:22.791708: step: 1948/527, loss: 0.004021644592285156
2023-01-21 12:38:23.903989: step: 1952/527, loss: 3.52859501617786e-06
2023-01-21 12:38:25.010833: step: 1956/527, loss: 1.373290979245212e-05
2023-01-21 12:38:26.089458: step: 1960/527, loss: 1.106262243411038e-05
2023-01-21 12:38:27.191553: step: 1964/527, loss: 2.8038026357535273e-05
2023-01-21 12:38:28.299970: step: 1968/527, loss: 0.09510540962219238
2023-01-21 12:38:29.415836: step: 1972/527, loss: 0.0002437591610942036
2023-01-21 12:38:30.524271: step: 1976/527, loss: 0.0005998611450195312
2023-01-21 12:38:31.617609: step: 1980/527, loss: 0.0020483972039073706
2023-01-21 12:38:32.702515: step: 1984/527, loss: 0.08438543975353241
2023-01-21 12:38:33.816331: step: 1988/527, loss: 0.004636954981833696
2023-01-21 12:38:34.933651: step: 1992/527, loss: 2.2029875253792852e-05
2023-01-21 12:38:36.040984: step: 1996/527, loss: 0.0010011672275140882
2023-01-21 12:38:37.161350: step: 2000/527, loss: 0.02397174760699272
2023-01-21 12:38:38.264754: step: 2004/527, loss: 0.03462624549865723
2023-01-21 12:38:39.359109: step: 2008/527, loss: 9.632110959501006e-06
2023-01-21 12:38:40.453563: step: 2012/527, loss: 0.02156086079776287
2023-01-21 12:38:41.530527: step: 2016/527, loss: 0.00016870498075149953
2023-01-21 12:38:42.645145: step: 2020/527, loss: 0.008963203057646751
2023-01-21 12:38:43.751422: step: 2024/527, loss: 0.0026958466041833162
2023-01-21 12:38:44.861771: step: 2028/527, loss: 0.043246079236269
2023-01-21 12:38:45.984554: step: 2032/527, loss: 0.005075711291283369
2023-01-21 12:38:47.100526: step: 2036/527, loss: 0.0039914133958518505
2023-01-21 12:38:48.224457: step: 2040/527, loss: 1.4342349767684937
2023-01-21 12:38:49.356944: step: 2044/527, loss: 0.0011192321544513106
2023-01-21 12:38:50.477608: step: 2048/527, loss: 1.3446808225126006e-05
2023-01-21 12:38:51.606275: step: 2052/527, loss: 0.006104087922722101
2023-01-21 12:38:52.719488: step: 2056/527, loss: 0.04354248195886612
2023-01-21 12:38:53.829756: step: 2060/527, loss: 0.00016536712064407766
2023-01-21 12:38:54.951555: step: 2064/527, loss: 0.004340839572250843
2023-01-21 12:38:56.079815: step: 2068/527, loss: 0.0004107475106138736
2023-01-21 12:38:57.200273: step: 2072/527, loss: 0.054785825312137604
2023-01-21 12:38:58.307783: step: 2076/527, loss: 0.011180782690644264
2023-01-21 12:38:59.422579: step: 2080/527, loss: 0.0047626495361328125
2023-01-21 12:39:00.527161: step: 2084/527, loss: 0.014306164346635342
2023-01-21 12:39:01.608863: step: 2088/527, loss: 0.00014019012451171875
2023-01-21 12:39:02.748272: step: 2092/527, loss: 0.0126221664249897
2023-01-21 12:39:03.846638: step: 2096/527, loss: 0.00021162032498978078
2023-01-21 12:39:04.956963: step: 2100/527, loss: 0.01700897328555584
2023-01-21 12:39:06.062584: step: 2104/527, loss: 0.030184125527739525
2023-01-21 12:39:07.159816: step: 2108/527, loss: 1.5735627130197827e-06
==================================================
Loss: 0.030
--------------------
Dev: {'event': {'p': 0.5964912280701754, 'r': 0.7696404793608522, 'f1': 0.6720930232558139}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}
Test: {'event': {'p': 0.6298819255222525, 'r': 0.7925714285714286, 'f1': 0.701923076923077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}
Chinese: {'event': {'p': 0.5609756097560976, 'r': 0.8518518518518519, 'f1': 0.6764705882352942}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}
Korean: {'event': {'p': 0.5689655172413793, 'r': 0.5238095238095238, 'f1': 0.5454545454545455}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}
Russian: {'event': {'p': 0.425, 'r': 0.4722222222222222, 'f1': 0.4473684210526316}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6066252587991718, 'r': 0.7802929427430093, 'f1': 0.6825859056493885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Eng Test for Korean: {'event': {'p': 0.62580054894785, 'r': 0.7817142857142857, 'f1': 0.6951219512195121}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
Sample Korean: {'event': {'p': 0.6730769230769231, 'r': 0.5555555555555556, 'f1': 0.6086956521739131}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 12:39:48.475973: step: 4/527, loss: 0.0009110451210290194
2023-01-21 12:39:49.559378: step: 8/527, loss: 7.848739915061742e-05
2023-01-21 12:39:50.664787: step: 12/527, loss: 0.010296916589140892
2023-01-21 12:39:51.757849: step: 16/527, loss: 1.2874603271484375e-05
2023-01-21 12:39:52.831566: step: 20/527, loss: 0.005050516687333584
2023-01-21 12:39:53.972817: step: 24/527, loss: 0.0032618525438010693
2023-01-21 12:39:55.082745: step: 28/527, loss: 0.00013599396334029734
2023-01-21 12:39:56.174820: step: 32/527, loss: 0.0010828971862792969
2023-01-21 12:39:57.261106: step: 36/527, loss: 5.836487252963707e-05
2023-01-21 12:39:58.368742: step: 40/527, loss: 0.00027561187744140625
2023-01-21 12:39:59.493567: step: 44/527, loss: 0.00583992013707757
2023-01-21 12:40:00.613850: step: 48/527, loss: 0.002434349153190851
2023-01-21 12:40:01.726059: step: 52/527, loss: 0.0006800651899538934
2023-01-21 12:40:02.834654: step: 56/527, loss: 0.0007221222040243447
2023-01-21 12:40:03.957654: step: 60/527, loss: 0.05299406126141548
2023-01-21 12:40:05.084684: step: 64/527, loss: 0.0016513824230059981
2023-01-21 12:40:06.182768: step: 68/527, loss: 0.006466483697295189
2023-01-21 12:40:07.303580: step: 72/527, loss: 0.004902172368019819
2023-01-21 12:40:08.394974: step: 76/527, loss: 0.028306199237704277
2023-01-21 12:40:09.509429: step: 80/527, loss: 0.0033044815063476562
2023-01-21 12:40:10.641092: step: 84/527, loss: 0.0001655578671488911
2023-01-21 12:40:11.749437: step: 88/527, loss: 0.005522537510842085
2023-01-21 12:40:12.849381: step: 92/527, loss: 0.00021667480177711695
2023-01-21 12:40:13.945701: step: 96/527, loss: 2.6226043701171875e-05
2023-01-21 12:40:15.036434: step: 100/527, loss: 0.010250759311020374
2023-01-21 12:40:16.139481: step: 104/527, loss: 0.0047595021314918995
2023-01-21 12:40:17.257663: step: 108/527, loss: 0.00010604858107399195
2023-01-21 12:40:18.355724: step: 112/527, loss: 0.011271094903349876
2023-01-21 12:40:19.479557: step: 116/527, loss: 0.024996565654873848
2023-01-21 12:40:20.635302: step: 120/527, loss: 0.0003570079570636153
2023-01-21 12:40:21.737580: step: 124/527, loss: 0.0013793945545330644
2023-01-21 12:40:22.850450: step: 128/527, loss: 8.630751835880801e-06
2023-01-21 12:40:23.945587: step: 132/527, loss: 0.00012445449829101562
2023-01-21 12:40:25.080004: step: 136/527, loss: 0.0003688812139444053
2023-01-21 12:40:26.213420: step: 140/527, loss: 0.007424688432365656
2023-01-21 12:40:27.321410: step: 144/527, loss: 0.007475471124053001
2023-01-21 12:40:28.425603: step: 148/527, loss: 0.0024349212180823088
2023-01-21 12:40:29.537842: step: 152/527, loss: 0.0001924514799611643
2023-01-21 12:40:30.662276: step: 156/527, loss: 0.0019601343665271997
2023-01-21 12:40:31.786755: step: 160/527, loss: 0.006792736239731312
2023-01-21 12:40:32.909809: step: 164/527, loss: 0.8472946286201477
2023-01-21 12:40:34.011900: step: 168/527, loss: 2.6702882678364404e-06
2023-01-21 12:40:35.137066: step: 172/527, loss: 0.0067005157470703125
2023-01-21 12:40:36.232710: step: 176/527, loss: 0.00017433166794944555
2023-01-21 12:40:37.344226: step: 180/527, loss: 0.021106814965605736
2023-01-21 12:40:38.441538: step: 184/527, loss: 0.0003991127014160156
2023-01-21 12:40:39.539767: step: 188/527, loss: 0.0012279510265216231
2023-01-21 12:40:40.642700: step: 192/527, loss: 7.4386593951203395e-06
2023-01-21 12:40:41.767962: step: 196/527, loss: 0.0017589569324627519
2023-01-21 12:40:42.874822: step: 200/527, loss: 0.009861993603408337
2023-01-21 12:40:44.018155: step: 204/527, loss: 0.01757373847067356
2023-01-21 12:40:45.153640: step: 208/527, loss: 0.0030303956009447575
2023-01-21 12:40:46.236100: step: 212/527, loss: 2.498626781743951e-05
2023-01-21 12:40:47.358947: step: 216/527, loss: 0.03145027160644531
2023-01-21 12:40:48.457550: step: 220/527, loss: 0.00015382767014671117
2023-01-21 12:40:49.609379: step: 224/527, loss: 0.02449817769229412
2023-01-21 12:40:50.784821: step: 228/527, loss: 0.0024200440384447575
2023-01-21 12:40:51.869101: step: 232/527, loss: 0.01563587225973606
2023-01-21 12:40:52.991118: step: 236/527, loss: 0.001042270683683455
2023-01-21 12:40:54.065403: step: 240/527, loss: 3.0040740966796875e-05
2023-01-21 12:40:55.182972: step: 244/527, loss: 5.111694190418348e-05
2023-01-21 12:40:56.309371: step: 248/527, loss: 2.784729076665826e-05
2023-01-21 12:40:57.385856: step: 252/527, loss: -1.907349087559851e-07
2023-01-21 12:40:58.491112: step: 256/527, loss: 0.0005649566883221269
2023-01-21 12:40:59.609260: step: 260/527, loss: 0.0253111831843853
2023-01-21 12:41:00.728778: step: 264/527, loss: 0.0010314941173419356
2023-01-21 12:41:01.824160: step: 268/527, loss: 0.0006244659889489412
2023-01-21 12:41:02.969145: step: 272/527, loss: 0.028299523517489433
2023-01-21 12:41:04.107898: step: 276/527, loss: 0.00043258670484647155
2023-01-21 12:41:05.201157: step: 280/527, loss: 0.04387526586651802
2023-01-21 12:41:06.289229: step: 284/527, loss: 0.0035158158279955387
2023-01-21 12:41:07.409355: step: 288/527, loss: 0.00032806396484375
2023-01-21 12:41:08.535803: step: 292/527, loss: 2.746581958490424e-05
2023-01-21 12:41:09.666542: step: 296/527, loss: 0.00024552346440032125
2023-01-21 12:41:10.763766: step: 300/527, loss: 3.166198803228326e-05
2023-01-21 12:41:11.876843: step: 304/527, loss: 0.004077625460922718
2023-01-21 12:41:12.980555: step: 308/527, loss: 0.0012128830421715975
2023-01-21 12:41:14.078707: step: 312/527, loss: 3.662109520519152e-05
2023-01-21 12:41:15.207480: step: 316/527, loss: 0.01729745790362358
2023-01-21 12:41:16.327663: step: 320/527, loss: 4.024505687993951e-05
2023-01-21 12:41:17.442593: step: 324/527, loss: 0.0010309219360351562
2023-01-21 12:41:18.548276: step: 328/527, loss: 0.05305824428796768
2023-01-21 12:41:19.625270: step: 332/527, loss: 0.011519813910126686
2023-01-21 12:41:20.754298: step: 336/527, loss: 0.0031494139693677425
2023-01-21 12:41:21.897646: step: 340/527, loss: 0.000179290771484375
2023-01-21 12:41:23.033640: step: 344/527, loss: 0.006348800845444202
2023-01-21 12:41:24.132445: step: 348/527, loss: 0.0012062073219567537
2023-01-21 12:41:25.284827: step: 352/527, loss: 1.277923547604587e-05
2023-01-21 12:41:26.398192: step: 356/527, loss: 0.05075817182660103
2023-01-21 12:41:27.505898: step: 360/527, loss: 0.001114320824854076
2023-01-21 12:41:28.603993: step: 364/527, loss: 8.449555025435984e-05
2023-01-21 12:41:29.719531: step: 368/527, loss: 0.016541672870516777
2023-01-21 12:41:30.815578: step: 372/527, loss: 0.004622983746230602
2023-01-21 12:41:31.937038: step: 376/527, loss: 0.00474814185872674
2023-01-21 12:41:33.038503: step: 380/527, loss: 0.005918407812714577
2023-01-21 12:41:34.175367: step: 384/527, loss: 0.0004203796270303428
2023-01-21 12:41:35.289699: step: 388/527, loss: 0.2650686204433441
2023-01-21 12:41:36.384802: step: 392/527, loss: 0.00012855530076194555
2023-01-21 12:41:37.495517: step: 396/527, loss: 0.023317432031035423
2023-01-21 12:41:38.588846: step: 400/527, loss: 0.0048354147002100945
2023-01-21 12:41:39.721264: step: 404/527, loss: 0.007964467629790306
2023-01-21 12:41:40.822242: step: 408/527, loss: 0.04084453731775284
2023-01-21 12:41:41.941353: step: 412/527, loss: 0.03669404983520508
2023-01-21 12:41:43.056787: step: 416/527, loss: 0.028662966564297676
2023-01-21 12:41:44.169958: step: 420/527, loss: 0.00024499892606399953
2023-01-21 12:41:45.292506: step: 424/527, loss: 0.003098870161920786
2023-01-21 12:41:46.395920: step: 428/527, loss: 0.0018156052101403475
2023-01-21 12:41:47.482926: step: 432/527, loss: 0.007095145992934704
2023-01-21 12:41:48.590849: step: 436/527, loss: 0.00019741058349609375
2023-01-21 12:41:49.698205: step: 440/527, loss: 0.002078437712043524
2023-01-21 12:41:50.823199: step: 444/527, loss: 0.005266761872917414
2023-01-21 12:41:51.923260: step: 448/527, loss: 0.0018029690254479647
2023-01-21 12:41:53.049801: step: 452/527, loss: 0.03513755649328232
2023-01-21 12:41:54.137803: step: 456/527, loss: 0.0023438455536961555
2023-01-21 12:41:55.257438: step: 460/527, loss: 0.006919098552316427
2023-01-21 12:41:56.360198: step: 464/527, loss: 0.0007852555136196315
2023-01-21 12:41:57.449285: step: 468/527, loss: 0.013456153683364391
2023-01-21 12:41:58.535659: step: 472/527, loss: 0.000621795654296875
2023-01-21 12:41:59.704942: step: 476/527, loss: 0.06559744477272034
2023-01-21 12:42:00.822329: step: 480/527, loss: 0.0035276412963867188
2023-01-21 12:42:01.940338: step: 484/527, loss: 0.02904348261654377
2023-01-21 12:42:03.065077: step: 488/527, loss: 0.003094196319580078
2023-01-21 12:42:04.154290: step: 492/527, loss: 0.0008033752674236894
2023-01-21 12:42:05.272374: step: 496/527, loss: 0.00444374093785882
2023-01-21 12:42:06.410290: step: 500/527, loss: 0.014609336853027344
2023-01-21 12:42:07.549619: step: 504/527, loss: 4.720688230008818e-05
2023-01-21 12:42:08.639672: step: 508/527, loss: 0.005415153689682484
2023-01-21 12:42:09.779270: step: 512/527, loss: 0.004528617952018976
2023-01-21 12:42:10.888692: step: 516/527, loss: 0.0003110409015789628
2023-01-21 12:42:11.990749: step: 520/527, loss: 0.00036683081998489797
2023-01-21 12:42:13.097515: step: 524/527, loss: 0.010301709175109863
2023-01-21 12:42:14.226030: step: 528/527, loss: 0.0007524490938521922
2023-01-21 12:42:15.309620: step: 532/527, loss: 0.0020704269409179688
2023-01-21 12:42:16.424606: step: 536/527, loss: 0.031176377087831497
2023-01-21 12:42:17.551624: step: 540/527, loss: 0.01903695985674858
2023-01-21 12:42:18.674903: step: 544/527, loss: 0.004793358035385609
2023-01-21 12:42:19.810207: step: 548/527, loss: 0.005286025814712048
2023-01-21 12:42:20.923934: step: 552/527, loss: 0.0006929397932253778
2023-01-21 12:42:22.047538: step: 556/527, loss: 0.0001911163271870464
2023-01-21 12:42:23.141813: step: 560/527, loss: 0.0026998519897460938
2023-01-21 12:42:24.258994: step: 564/527, loss: 0.06305485218763351
2023-01-21 12:42:25.378654: step: 568/527, loss: 0.028919601812958717
2023-01-21 12:42:26.511175: step: 572/527, loss: 0.00035381317138671875
2023-01-21 12:42:27.597038: step: 576/527, loss: 0.0006907462957315147
2023-01-21 12:42:28.700547: step: 580/527, loss: 0.0004951477167196572
2023-01-21 12:42:29.812560: step: 584/527, loss: 0.00132837297860533
2023-01-21 12:42:30.909434: step: 588/527, loss: 0.06281270831823349
2023-01-21 12:42:32.014164: step: 592/527, loss: 0.0032138824462890625
2023-01-21 12:42:33.150633: step: 596/527, loss: 0.0018033981323242188
2023-01-21 12:42:34.274826: step: 600/527, loss: 0.000164031982421875
2023-01-21 12:42:35.379712: step: 604/527, loss: 0.00018787384033203125
2023-01-21 12:42:36.497992: step: 608/527, loss: 0.0001367092045256868
2023-01-21 12:42:37.589628: step: 612/527, loss: 0.0002193450927734375
2023-01-21 12:42:38.716835: step: 616/527, loss: 7.905960228526965e-05
2023-01-21 12:42:39.812690: step: 620/527, loss: 0.002499866532161832
2023-01-21 12:42:40.924854: step: 624/527, loss: 0.0008541106944903731
2023-01-21 12:42:42.054844: step: 628/527, loss: 0.005959892179816961
2023-01-21 12:42:43.177306: step: 632/527, loss: 0.0015912533272057772
2023-01-21 12:42:44.287454: step: 636/527, loss: 0.001168584800325334
2023-01-21 12:42:45.378608: step: 640/527, loss: 0.003045368241146207
2023-01-21 12:42:46.488236: step: 644/527, loss: 0.031163597479462624
2023-01-21 12:42:47.621270: step: 648/527, loss: 0.0002529144403524697
2023-01-21 12:42:48.716944: step: 652/527, loss: 0.015043544583022594
2023-01-21 12:42:49.832505: step: 656/527, loss: 0.004358482547104359
2023-01-21 12:42:50.931386: step: 660/527, loss: 0.005823039915412664
2023-01-21 12:42:52.057314: step: 664/527, loss: 5.779266211902723e-05
2023-01-21 12:42:53.186895: step: 668/527, loss: 0.0238476749509573
2023-01-21 12:42:54.324835: step: 672/527, loss: 0.015002441592514515
2023-01-21 12:42:55.428939: step: 676/527, loss: 0.00017404557729605585
2023-01-21 12:42:56.515410: step: 680/527, loss: 6.031990051269531e-05
2023-01-21 12:42:57.605357: step: 684/527, loss: 4.291534423828125e-06
2023-01-21 12:42:58.704663: step: 688/527, loss: 0.02594575844705105
2023-01-21 12:42:59.824854: step: 692/527, loss: 0.001591157983057201
2023-01-21 12:43:00.954978: step: 696/527, loss: 0.0002475738583598286
2023-01-21 12:43:02.057814: step: 700/527, loss: 7.534027827205136e-05
2023-01-21 12:43:03.173529: step: 704/527, loss: 1.049041748046875e-05
2023-01-21 12:43:04.278460: step: 708/527, loss: 0.0005067825550213456
2023-01-21 12:43:05.380897: step: 712/527, loss: 0.00020160674466751516
2023-01-21 12:43:06.493163: step: 716/527, loss: 4.57763690064894e-06
2023-01-21 12:43:07.599423: step: 720/527, loss: 0.02637786790728569
2023-01-21 12:43:08.699478: step: 724/527, loss: -9.536743306171047e-08
2023-01-21 12:43:09.832446: step: 728/527, loss: -3.0517576306010596e-06
2023-01-21 12:43:10.975843: step: 732/527, loss: 0.03326826170086861
2023-01-21 12:43:12.088649: step: 736/527, loss: 0.00024242402287200093
2023-01-21 12:43:13.165150: step: 740/527, loss: 0.000480842572869733
2023-01-21 12:43:14.288730: step: 744/527, loss: 0.14041289687156677
2023-01-21 12:43:15.439061: step: 748/527, loss: 0.0016294479137286544
2023-01-21 12:43:16.521987: step: 752/527, loss: 0.00039472582284361124
2023-01-21 12:43:17.615177: step: 756/527, loss: 0.0006849289056845009
2023-01-21 12:43:18.746738: step: 760/527, loss: 0.055683329701423645
2023-01-21 12:43:19.870532: step: 764/527, loss: -5.149840944795869e-06
2023-01-21 12:43:20.981115: step: 768/527, loss: 0.015389394015073776
2023-01-21 12:43:22.099470: step: 772/527, loss: 0.0018640519119799137
2023-01-21 12:43:23.211485: step: 776/527, loss: 0.0010543823009356856
2023-01-21 12:43:24.300479: step: 780/527, loss: 0.0002099990815622732
2023-01-21 12:43:25.404172: step: 784/527, loss: 2.689361645025201e-05
2023-01-21 12:43:26.515888: step: 788/527, loss: 0.004417181480675936
2023-01-21 12:43:27.614657: step: 792/527, loss: 0.018381882458925247
2023-01-21 12:43:28.744273: step: 796/527, loss: 0.01084060687571764
2023-01-21 12:43:29.859699: step: 800/527, loss: 0.004812145140022039
2023-01-21 12:43:30.965551: step: 804/527, loss: 0.08924102783203125
2023-01-21 12:43:32.083359: step: 808/527, loss: 0.014422702603042126
2023-01-21 12:43:33.179346: step: 812/527, loss: 0.030928421765565872
2023-01-21 12:43:34.282516: step: 816/527, loss: 0.01819891855120659
2023-01-21 12:43:35.425129: step: 820/527, loss: 0.003727388335391879
2023-01-21 12:43:36.562985: step: 824/527, loss: 0.06298418343067169
2023-01-21 12:43:37.703826: step: 828/527, loss: 0.0014245033962652087
2023-01-21 12:43:38.799666: step: 832/527, loss: 0.038451578468084335
2023-01-21 12:43:39.928582: step: 836/527, loss: 0.03311176598072052
2023-01-21 12:43:41.048679: step: 840/527, loss: 0.0009989738464355469
2023-01-21 12:43:42.177087: step: 844/527, loss: 0.0011274338467046618
2023-01-21 12:43:43.329785: step: 848/527, loss: 0.00535850552842021
2023-01-21 12:43:44.455964: step: 852/527, loss: 0.007448769174516201
2023-01-21 12:43:45.562209: step: 856/527, loss: 0.010066986083984375
2023-01-21 12:43:46.671953: step: 860/527, loss: 0.0008013487095013261
2023-01-21 12:43:47.783981: step: 864/527, loss: 0.00015277863712981343
2023-01-21 12:43:48.918235: step: 868/527, loss: 9.336470975540578e-05
2023-01-21 12:43:50.022435: step: 872/527, loss: 0.0009654522291384637
2023-01-21 12:43:51.205511: step: 876/527, loss: 0.0037525177467614412
2023-01-21 12:43:52.346759: step: 880/527, loss: 0.025702476501464844
2023-01-21 12:43:53.506590: step: 884/527, loss: 0.0015096664428710938
2023-01-21 12:43:54.677848: step: 888/527, loss: 0.009095096960663795
2023-01-21 12:43:55.813612: step: 892/527, loss: 0.048610687255859375
2023-01-21 12:43:56.950417: step: 896/527, loss: 0.0021051408257335424
2023-01-21 12:43:58.048524: step: 900/527, loss: 0.003425789065659046
2023-01-21 12:43:59.168045: step: 904/527, loss: 0.04430227726697922
2023-01-21 12:44:00.272470: step: 908/527, loss: 0.0002478599490132183
2023-01-21 12:44:01.391421: step: 912/527, loss: 3.1471254260395654e-06
2023-01-21 12:44:02.506299: step: 916/527, loss: 0.0033757209312170744
2023-01-21 12:44:03.635565: step: 920/527, loss: 0.003626060439273715
2023-01-21 12:44:04.763247: step: 924/527, loss: 0.03672304376959801
2023-01-21 12:44:05.931264: step: 928/527, loss: 0.00027179718017578125
2023-01-21 12:44:07.036124: step: 932/527, loss: 0.0011697768932208419
2023-01-21 12:44:08.133427: step: 936/527, loss: 4.291534423828125e-06
2023-01-21 12:44:09.205435: step: 940/527, loss: 0.00032830238342285156
2023-01-21 12:44:10.308522: step: 944/527, loss: 4.425048973644152e-05
2023-01-21 12:44:11.417674: step: 948/527, loss: 5.340576535672881e-06
2023-01-21 12:44:12.520572: step: 952/527, loss: 1.430511474609375e-05
2023-01-21 12:44:13.632501: step: 956/527, loss: 0.00011329651169944555
2023-01-21 12:44:14.750100: step: 960/527, loss: 0.0019049644470214844
2023-01-21 12:44:15.898744: step: 964/527, loss: 0.07014808803796768
2023-01-21 12:44:17.012482: step: 968/527, loss: 0.007043933961540461
2023-01-21 12:44:18.116504: step: 972/527, loss: 0.10995063930749893
2023-01-21 12:44:19.206588: step: 976/527, loss: 0.00012874603271484375
2023-01-21 12:44:20.293597: step: 980/527, loss: 0.013449382968246937
2023-01-21 12:44:21.414727: step: 984/527, loss: 0.01301345881074667
2023-01-21 12:44:22.536923: step: 988/527, loss: 0.007336426060646772
2023-01-21 12:44:23.619448: step: 992/527, loss: -1.5258789289873675e-06
2023-01-21 12:44:24.767136: step: 996/527, loss: 0.011358261108398438
2023-01-21 12:44:25.879391: step: 1000/527, loss: 0.00013909340486861765
2023-01-21 12:44:26.983228: step: 1004/527, loss: 0.0001066207914846018
2023-01-21 12:44:28.081457: step: 1008/527, loss: 0.571603536605835
2023-01-21 12:44:29.169159: step: 1012/527, loss: 0.03208484873175621
2023-01-21 12:44:30.277106: step: 1016/527, loss: 0.00943231675773859
2023-01-21 12:44:31.379851: step: 1020/527, loss: 0.003540611360222101
2023-01-21 12:44:32.519817: step: 1024/527, loss: 0.1077755019068718
2023-01-21 12:44:33.631387: step: 1028/527, loss: 0.04181881248950958
2023-01-21 12:44:34.766212: step: 1032/527, loss: 0.04215993732213974
2023-01-21 12:44:35.864778: step: 1036/527, loss: 0.0014624595642089844
2023-01-21 12:44:36.942829: step: 1040/527, loss: 6.48498553346144e-06
2023-01-21 12:44:38.030342: step: 1044/527, loss: 0.005657482426613569
2023-01-21 12:44:39.150635: step: 1048/527, loss: 0.0014853953616693616
2023-01-21 12:44:40.283678: step: 1052/527, loss: 0.13561153411865234
2023-01-21 12:44:41.379909: step: 1056/527, loss: 0.00016446114750579
2023-01-21 12:44:42.539481: step: 1060/527, loss: 0.11530762165784836
2023-01-21 12:44:43.650360: step: 1064/527, loss: 0.01401300448924303
2023-01-21 12:44:44.751572: step: 1068/527, loss: 0.0003203391970600933
2023-01-21 12:44:45.853686: step: 1072/527, loss: 0.0015596390003338456
2023-01-21 12:44:46.939142: step: 1076/527, loss: 0.0009634971502237022
2023-01-21 12:44:48.056604: step: 1080/527, loss: 6.513595872092992e-05
2023-01-21 12:44:49.202902: step: 1084/527, loss: 0.015500831417739391
2023-01-21 12:44:50.308949: step: 1088/527, loss: 0.007788181304931641
2023-01-21 12:44:51.428289: step: 1092/527, loss: 0.006949997041374445
2023-01-21 12:44:52.515040: step: 1096/527, loss: 0.008608246222138405
2023-01-21 12:44:53.656678: step: 1100/527, loss: 0.00017547607421875
2023-01-21 12:44:54.757037: step: 1104/527, loss: 0.006443119142204523
2023-01-21 12:44:55.868097: step: 1108/527, loss: 0.0023887634743005037
2023-01-21 12:44:56.949259: step: 1112/527, loss: 0.0005298614269122481
2023-01-21 12:44:58.097345: step: 1116/527, loss: 0.0007083893287926912
2023-01-21 12:44:59.246515: step: 1120/527, loss: 0.0016560554504394531
2023-01-21 12:45:00.427549: step: 1124/527, loss: 0.01673126220703125
2023-01-21 12:45:01.526613: step: 1128/527, loss: 0.027311135083436966
2023-01-21 12:45:02.638189: step: 1132/527, loss: 0.01328296773135662
2023-01-21 12:45:03.729575: step: 1136/527, loss: 4.291534423828125e-06
2023-01-21 12:45:04.857044: step: 1140/527, loss: 0.0005595207330770791
2023-01-21 12:45:05.962932: step: 1144/527, loss: 0.0008642196771688759
2023-01-21 12:45:07.109080: step: 1148/527, loss: 0.0006767273298464715
2023-01-21 12:45:08.208488: step: 1152/527, loss: 0.007454490754753351
2023-01-21 12:45:09.344104: step: 1156/527, loss: -3.0517576306010596e-06
2023-01-21 12:45:10.466662: step: 1160/527, loss: 0.0201263427734375
2023-01-21 12:45:11.578898: step: 1164/527, loss: 0.008500671945512295
2023-01-21 12:45:12.685207: step: 1168/527, loss: 0.02196650579571724
2023-01-21 12:45:13.773878: step: 1172/527, loss: 0.0015062332386150956
2023-01-21 12:45:14.865639: step: 1176/527, loss: 3.175735764671117e-05
2023-01-21 12:45:15.984673: step: 1180/527, loss: 0.02458667755126953
2023-01-21 12:45:17.089082: step: 1184/527, loss: 0.006309890653938055
2023-01-21 12:45:18.199220: step: 1188/527, loss: 0.11926989257335663
2023-01-21 12:45:19.286834: step: 1192/527, loss: 0.00024309159198310226
2023-01-21 12:45:20.424544: step: 1196/527, loss: 0.00034198761568404734
2023-01-21 12:45:21.523752: step: 1200/527, loss: 0.0009774207137525082
2023-01-21 12:45:22.665327: step: 1204/527, loss: 0.00083074567373842
2023-01-21 12:45:23.772255: step: 1208/527, loss: 0.012092972174286842
2023-01-21 12:45:24.877418: step: 1212/527, loss: 0.17910784482955933
2023-01-21 12:45:25.965349: step: 1216/527, loss: 8.287429955089465e-05
2023-01-21 12:45:27.088427: step: 1220/527, loss: 0.00026597976102493703
2023-01-21 12:45:28.181144: step: 1224/527, loss: 0.0031692981719970703
2023-01-21 12:45:29.301577: step: 1228/527, loss: 1.621246337890625e-05
2023-01-21 12:45:30.398078: step: 1232/527, loss: 0.0018568038940429688
2023-01-21 12:45:31.515402: step: 1236/527, loss: 0.0008809566497802734
2023-01-21 12:45:32.613029: step: 1240/527, loss: 0.0026684761978685856
2023-01-21 12:45:33.709297: step: 1244/527, loss: 0.010835838504135609
2023-01-21 12:45:34.809630: step: 1248/527, loss: 0.08521823585033417
2023-01-21 12:45:35.930146: step: 1252/527, loss: 0.03731956332921982
2023-01-21 12:45:37.038636: step: 1256/527, loss: 0.00023627281188964844
2023-01-21 12:45:38.156421: step: 1260/527, loss: 6.065368506824598e-05
2023-01-21 12:45:39.296615: step: 1264/527, loss: 0.0001829147367971018
2023-01-21 12:45:40.451700: step: 1268/527, loss: 0.00393333425745368
2023-01-21 12:45:41.574943: step: 1272/527, loss: 0.00014209747314453125
2023-01-21 12:45:42.719102: step: 1276/527, loss: 0.0066648488864302635
2023-01-21 12:45:43.822112: step: 1280/527, loss: 0.002902126405388117
2023-01-21 12:45:44.938312: step: 1284/527, loss: 0.0014165878528729081
2023-01-21 12:45:46.042766: step: 1288/527, loss: 1.735687328618951e-05
2023-01-21 12:45:47.145768: step: 1292/527, loss: 0.07302512973546982
2023-01-21 12:45:48.267343: step: 1296/527, loss: 0.0147247314453125
2023-01-21 12:45:49.361127: step: 1300/527, loss: 5.493163916980848e-05
2023-01-21 12:45:50.489836: step: 1304/527, loss: 2.3746491933707148e-05
2023-01-21 12:45:51.591466: step: 1308/527, loss: 0.009915444999933243
2023-01-21 12:45:52.707224: step: 1312/527, loss: 0.00031299592228606343
2023-01-21 12:45:53.804336: step: 1316/527, loss: 0.008557701483368874
2023-01-21 12:45:54.931133: step: 1320/527, loss: 0.0022821428719908
2023-01-21 12:45:56.040308: step: 1324/527, loss: 0.0010303497547283769
2023-01-21 12:45:57.140029: step: 1328/527, loss: 0.0002884864807128906
2023-01-21 12:45:58.239861: step: 1332/527, loss: 0.008757400326430798
2023-01-21 12:45:59.413477: step: 1336/527, loss: 0.07017135620117188
2023-01-21 12:46:00.516221: step: 1340/527, loss: 0.004674673080444336
2023-01-21 12:46:01.633331: step: 1344/527, loss: 0.07266692817211151
2023-01-21 12:46:02.755371: step: 1348/527, loss: 0.00021200180344749242
2023-01-21 12:46:03.856476: step: 1352/527, loss: 0.00012035370309604332
2023-01-21 12:46:04.972746: step: 1356/527, loss: 0.0030529021751135588
2023-01-21 12:46:06.098743: step: 1360/527, loss: 3.337860107421875e-05
2023-01-21 12:46:07.204575: step: 1364/527, loss: 0.005961894989013672
2023-01-21 12:46:08.321536: step: 1368/527, loss: 1.621246337890625e-05
2023-01-21 12:46:09.450401: step: 1372/527, loss: 0.00015954971604514867
2023-01-21 12:46:10.549819: step: 1376/527, loss: 2.708435022213962e-05
2023-01-21 12:46:11.657731: step: 1380/527, loss: 8.792877633823082e-05
2023-01-21 12:46:12.771448: step: 1384/527, loss: 0.0005931854830123484
2023-01-21 12:46:13.872373: step: 1388/527, loss: 0.004020023159682751
2023-01-21 12:46:14.992861: step: 1392/527, loss: 0.001962375594303012
2023-01-21 12:46:16.129633: step: 1396/527, loss: 0.0006391525384970009
2023-01-21 12:46:17.242501: step: 1400/527, loss: 4.329681542003527e-05
2023-01-21 12:46:18.369755: step: 1404/527, loss: 1.296997106692288e-05
2023-01-21 12:46:19.497136: step: 1408/527, loss: 0.0001470565766794607
2023-01-21 12:46:20.600562: step: 1412/527, loss: 0.013037562370300293
2023-01-21 12:46:21.719843: step: 1416/527, loss: 0.0006107330555096269
2023-01-21 12:46:22.825340: step: 1420/527, loss: 0.0036802294198423624
2023-01-21 12:46:23.961612: step: 1424/527, loss: 2.9659271604032256e-05
2023-01-21 12:46:25.041185: step: 1428/527, loss: 3.027916136488784e-05
2023-01-21 12:46:26.156623: step: 1432/527, loss: 0.0028078078757971525
2023-01-21 12:46:27.252991: step: 1436/527, loss: 0.0001264572056243196
2023-01-21 12:46:28.340832: step: 1440/527, loss: 0.010053062811493874
2023-01-21 12:46:29.449253: step: 1444/527, loss: 3.814697265625e-05
2023-01-21 12:46:30.583818: step: 1448/527, loss: 0.015074157156050205
2023-01-21 12:46:31.709503: step: 1452/527, loss: 0.06313496083021164
2023-01-21 12:46:32.845307: step: 1456/527, loss: 0.00028133392333984375
2023-01-21 12:46:33.994806: step: 1460/527, loss: 0.033357810229063034
2023-01-21 12:46:35.109492: step: 1464/527, loss: 7.505416579078883e-05
2023-01-21 12:46:36.223669: step: 1468/527, loss: 0.0013175965286791325
2023-01-21 12:46:37.386759: step: 1472/527, loss: 0.018235208466649055
2023-01-21 12:46:38.512033: step: 1476/527, loss: 7.152557373046875e-06
2023-01-21 12:46:39.605077: step: 1480/527, loss: 0.020801354199647903
2023-01-21 12:46:40.709435: step: 1484/527, loss: 0.00021457672119140625
2023-01-21 12:46:41.806959: step: 1488/527, loss: 0.0011400223011150956
2023-01-21 12:46:42.917807: step: 1492/527, loss: 0.06320180743932724
2023-01-21 12:46:44.017191: step: 1496/527, loss: 0.0013172149192541838
2023-01-21 12:46:45.155959: step: 1500/527, loss: 0.009605550207197666
2023-01-21 12:46:46.240024: step: 1504/527, loss: 0.0001027107282425277
2023-01-21 12:46:47.384409: step: 1508/527, loss: 0.011391830630600452
2023-01-21 12:46:48.495210: step: 1512/527, loss: 2.441406286379788e-05
2023-01-21 12:46:49.644916: step: 1516/527, loss: 0.003962898626923561
2023-01-21 12:46:50.755623: step: 1520/527, loss: 0.0012102127075195312
2023-01-21 12:46:51.888053: step: 1524/527, loss: 6.29425039733178e-06
2023-01-21 12:46:53.024969: step: 1528/527, loss: 4.6539309551008046e-05
2023-01-21 12:46:54.122173: step: 1532/527, loss: 0.00022726060706190765
2023-01-21 12:46:55.196439: step: 1536/527, loss: 9.202956789522432e-06
2023-01-21 12:46:56.323297: step: 1540/527, loss: 0.03403463587164879
2023-01-21 12:46:57.435694: step: 1544/527, loss: 0.03675422817468643
2023-01-21 12:46:58.535414: step: 1548/527, loss: 0.00046062469482421875
2023-01-21 12:46:59.615080: step: 1552/527, loss: 0.0014482499100267887
2023-01-21 12:47:00.737501: step: 1556/527, loss: 0.006739807315170765
2023-01-21 12:47:01.860950: step: 1560/527, loss: 4.682540748035535e-05
2023-01-21 12:47:02.982117: step: 1564/527, loss: 8.010864803509321e-06
2023-01-21 12:47:04.099463: step: 1568/527, loss: 0.007746315095573664
2023-01-21 12:47:05.243586: step: 1572/527, loss: 0.011471820063889027
2023-01-21 12:47:06.354253: step: 1576/527, loss: 0.0018236160976812243
2023-01-21 12:47:07.475489: step: 1580/527, loss: 9.794235666049644e-05
2023-01-21 12:47:08.676153: step: 1584/527, loss: 0.014097117818892002
2023-01-21 12:47:09.818838: step: 1588/527, loss: 0.054868318140506744
2023-01-21 12:47:10.912406: step: 1592/527, loss: 0.0016113758319988847
2023-01-21 12:47:12.035827: step: 1596/527, loss: 0.000562477158382535
2023-01-21 12:47:13.149046: step: 1600/527, loss: 0.0017742158379405737
2023-01-21 12:47:14.263871: step: 1604/527, loss: 0.007762241642922163
2023-01-21 12:47:15.391192: step: 1608/527, loss: 0.005294323433190584
2023-01-21 12:47:16.497192: step: 1612/527, loss: 0.0006788253667764366
2023-01-21 12:47:17.607433: step: 1616/527, loss: 0.00012226105900481343
2023-01-21 12:47:18.701416: step: 1620/527, loss: 0.00026607513427734375
2023-01-21 12:47:19.812879: step: 1624/527, loss: 0.0003145218070130795
2023-01-21 12:47:20.947454: step: 1628/527, loss: 0.004463577177375555
2023-01-21 12:47:22.085494: step: 1632/527, loss: 3.585815284168348e-05
2023-01-21 12:47:23.203851: step: 1636/527, loss: 0.03625917434692383
2023-01-21 12:47:24.321836: step: 1640/527, loss: 0.0003758430539164692
2023-01-21 12:47:25.452963: step: 1644/527, loss: 0.0035732747055590153
2023-01-21 12:47:26.563790: step: 1648/527, loss: 0.0001291275111725554
2023-01-21 12:47:27.681883: step: 1652/527, loss: 0.005347824189811945
2023-01-21 12:47:28.828774: step: 1656/527, loss: 0.0002368926943745464
2023-01-21 12:47:29.966767: step: 1660/527, loss: 0.00099773402325809
2023-01-21 12:47:31.107432: step: 1664/527, loss: 0.07575778663158417
2023-01-21 12:47:32.213872: step: 1668/527, loss: 9.965896606445312e-05
2023-01-21 12:47:33.375128: step: 1672/527, loss: 0.0036961555015295744
2023-01-21 12:47:34.477105: step: 1676/527, loss: 0.00030803680419921875
2023-01-21 12:47:35.596406: step: 1680/527, loss: 1.602172778802924e-05
2023-01-21 12:47:36.694772: step: 1684/527, loss: 0.0006830215570516884
2023-01-21 12:47:37.803614: step: 1688/527, loss: 0.00043444635230116546
2023-01-21 12:47:38.954063: step: 1692/527, loss: 3.318786548334174e-05
2023-01-21 12:47:40.045727: step: 1696/527, loss: 3.3855438232421875e-05
2023-01-21 12:47:41.146575: step: 1700/527, loss: 0.0003883361641783267
2023-01-21 12:47:42.312445: step: 1704/527, loss: 0.00017051698523573577
2023-01-21 12:47:43.433326: step: 1708/527, loss: 0.00034885405329987407
2023-01-21 12:47:44.578490: step: 1712/527, loss: 0.020542718470096588
2023-01-21 12:47:45.697328: step: 1716/527, loss: 3.5572051274357364e-05
2023-01-21 12:47:46.832152: step: 1720/527, loss: 0.00017442702664993703
2023-01-21 12:47:47.938140: step: 1724/527, loss: 0.0002727508544921875
2023-01-21 12:47:49.030428: step: 1728/527, loss: 0.0009077072609215975
2023-01-21 12:47:50.184724: step: 1732/527, loss: 0.017871523275971413
2023-01-21 12:47:51.285564: step: 1736/527, loss: 0.00011997222463833168
2023-01-21 12:47:52.415040: step: 1740/527, loss: 0.29222604632377625
2023-01-21 12:47:53.504481: step: 1744/527, loss: 0.00028753283550031483
2023-01-21 12:47:54.583601: step: 1748/527, loss: 0.00018997193546965718
2023-01-21 12:47:55.683935: step: 1752/527, loss: 0.0015269280411303043
2023-01-21 12:47:56.789505: step: 1756/527, loss: 0.00023727417283225805
2023-01-21 12:47:57.871724: step: 1760/527, loss: 1.2731552487821318e-05
2023-01-21 12:47:58.981037: step: 1764/527, loss: 0.0012496948475018144
2023-01-21 12:48:00.088196: step: 1768/527, loss: 0.0013566971756517887
2023-01-21 12:48:01.210256: step: 1772/527, loss: 0.0017074585193768144
2023-01-21 12:48:02.341157: step: 1776/527, loss: 0.0005752564175054431
2023-01-21 12:48:03.460890: step: 1780/527, loss: 0.0003880501026287675
2023-01-21 12:48:04.614120: step: 1784/527, loss: 0.5561831593513489
2023-01-21 12:48:05.740380: step: 1788/527, loss: 0.000362205522833392
2023-01-21 12:48:06.883433: step: 1792/527, loss: 0.004054451361298561
2023-01-21 12:48:07.992950: step: 1796/527, loss: 1.5306473869713955e-05
2023-01-21 12:48:09.130287: step: 1800/527, loss: 0.0003002166631631553
2023-01-21 12:48:10.229082: step: 1804/527, loss: 0.00337390904314816
2023-01-21 12:48:11.335718: step: 1808/527, loss: 0.004103279206901789
2023-01-21 12:48:12.480573: step: 1812/527, loss: 0.03952770307660103
2023-01-21 12:48:13.593361: step: 1816/527, loss: 0.07254792004823685
2023-01-21 12:48:14.708739: step: 1820/527, loss: 0.0006922244792804122
2023-01-21 12:48:15.832852: step: 1824/527, loss: 0.00016803742619231343
2023-01-21 12:48:16.938720: step: 1828/527, loss: 0.00020990372286178172
2023-01-21 12:48:18.059788: step: 1832/527, loss: 0.007465553469955921
2023-01-21 12:48:19.208478: step: 1836/527, loss: 0.002485084580257535
2023-01-21 12:48:20.303538: step: 1840/527, loss: 0.00030832289485260844
2023-01-21 12:48:21.423662: step: 1844/527, loss: 0.006155014503747225
2023-01-21 12:48:22.574193: step: 1848/527, loss: 0.01597149483859539
2023-01-21 12:48:23.679794: step: 1852/527, loss: 0.010368538089096546
2023-01-21 12:48:24.793879: step: 1856/527, loss: 0.0009034157264977694
2023-01-21 12:48:25.909145: step: 1860/527, loss: 0.008312702178955078
2023-01-21 12:48:27.048053: step: 1864/527, loss: 0.00018615722365211695
2023-01-21 12:48:28.162935: step: 1868/527, loss: 0.014773559756577015
2023-01-21 12:48:29.263458: step: 1872/527, loss: 0.0020902634132653475
2023-01-21 12:48:30.369948: step: 1876/527, loss: 0.000776100205257535
2023-01-21 12:48:31.515169: step: 1880/527, loss: 3.452301461948082e-05
2023-01-21 12:48:32.635824: step: 1884/527, loss: 0.0002249717799713835
2023-01-21 12:48:33.775900: step: 1888/527, loss: 0.0007179260719567537
2023-01-21 12:48:34.892851: step: 1892/527, loss: 0.0015614510048180819
2023-01-21 12:48:35.986745: step: 1896/527, loss: 0.0034096718300133944
2023-01-21 12:48:37.084872: step: 1900/527, loss: 0.036923788487911224
2023-01-21 12:48:38.194328: step: 1904/527, loss: 0.02938261069357395
2023-01-21 12:48:39.321445: step: 1908/527, loss: 0.001229620072990656
2023-01-21 12:48:40.477163: step: 1912/527, loss: 0.0003037929709535092
2023-01-21 12:48:41.579346: step: 1916/527, loss: 2.7322770620230585e-05
2023-01-21 12:48:42.681869: step: 1920/527, loss: 5.435943421616685e-06
2023-01-21 12:48:43.786353: step: 1924/527, loss: 0.0018215179443359375
2023-01-21 12:48:44.901280: step: 1928/527, loss: 0.024213576689362526
2023-01-21 12:48:46.019075: step: 1932/527, loss: 0.09364891052246094
2023-01-21 12:48:47.127137: step: 1936/527, loss: 8.468628220725805e-05
2023-01-21 12:48:48.237220: step: 1940/527, loss: 0.004463482182472944
2023-01-21 12:48:49.351639: step: 1944/527, loss: 0.0866088941693306
2023-01-21 12:48:50.451793: step: 1948/527, loss: 0.00044760701712220907
2023-01-21 12:48:51.573021: step: 1952/527, loss: 1.564025842526462e-05
2023-01-21 12:48:52.696380: step: 1956/527, loss: 0.048040393739938736
2023-01-21 12:48:53.807184: step: 1960/527, loss: 0.00012545585923362523
2023-01-21 12:48:54.923803: step: 1964/527, loss: 0.0009127140510827303
2023-01-21 12:48:56.002515: step: 1968/527, loss: 0.0009654044988565147
2023-01-21 12:48:57.118076: step: 1972/527, loss: 0.02923450618982315
2023-01-21 12:48:58.221426: step: 1976/527, loss: 0.0028626921121031046
2023-01-21 12:48:59.332078: step: 1980/527, loss: 2.021789623540826e-05
2023-01-21 12:49:00.420107: step: 1984/527, loss: 7.848739915061742e-05
2023-01-21 12:49:01.550983: step: 1988/527, loss: 0.0007068634149618447
2023-01-21 12:49:02.665697: step: 1992/527, loss: 0.0002561568981036544
2023-01-21 12:49:03.787586: step: 1996/527, loss: 0.0010416507720947266
2023-01-21 12:49:04.893493: step: 2000/527, loss: 0.0003029823419637978
2023-01-21 12:49:05.983043: step: 2004/527, loss: 0.01583237573504448
2023-01-21 12:49:07.082051: step: 2008/527, loss: 0.00916681345552206
2023-01-21 12:49:08.165935: step: 2012/527, loss: 9.622573998058215e-05
2023-01-21 12:49:09.242051: step: 2016/527, loss: 0.006411457434296608
2023-01-21 12:49:10.352993: step: 2020/527, loss: 3.82423386326991e-05
2023-01-21 12:49:11.446287: step: 2024/527, loss: 0.009990692138671875
2023-01-21 12:49:12.557645: step: 2028/527, loss: 0.005894470028579235
2023-01-21 12:49:13.682862: step: 2032/527, loss: 0.005634975619614124
2023-01-21 12:49:14.808983: step: 2036/527, loss: 0.005612373352050781
2023-01-21 12:49:15.939156: step: 2040/527, loss: 0.0033271312713623047
2023-01-21 12:49:17.020975: step: 2044/527, loss: 0.09671249985694885
2023-01-21 12:49:18.136821: step: 2048/527, loss: 0.004402351565659046
2023-01-21 12:49:19.198776: step: 2052/527, loss: 0.00016813278489280492
2023-01-21 12:49:20.319873: step: 2056/527, loss: 0.001800537109375
2023-01-21 12:49:21.422731: step: 2060/527, loss: 0.002048015594482422
2023-01-21 12:49:22.536770: step: 2064/527, loss: 0.00409011822193861
2023-01-21 12:49:23.656212: step: 2068/527, loss: 0.02006559446454048
2023-01-21 12:49:24.755433: step: 2072/527, loss: 0.0004901885986328125
2023-01-21 12:49:25.850846: step: 2076/527, loss: 2.6226043701171875e-06
2023-01-21 12:49:26.966610: step: 2080/527, loss: 0.007480430882424116
2023-01-21 12:49:28.057449: step: 2084/527, loss: 0.0010358811123296618
2023-01-21 12:49:29.208049: step: 2088/527, loss: 0.014516926370561123
2023-01-21 12:49:30.330503: step: 2092/527, loss: 0.0018707275157794356
2023-01-21 12:49:31.458240: step: 2096/527, loss: -3.8146959013829473e-07
2023-01-21 12:49:32.599255: step: 2100/527, loss: 0.039951324462890625
2023-01-21 12:49:33.703631: step: 2104/527, loss: 0.00030040740966796875
2023-01-21 12:49:34.840546: step: 2108/527, loss: 0.0022982596419751644
==================================================
Loss: 0.015
--------------------
Dev: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Test: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Chinese: {'event': {'p': 0.5454545454545454, 'r': 0.8888888888888888, 'f1': 0.676056338028169}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Russian: {'event': {'p': 0.4634146341463415, 'r': 0.5277777777777778, 'f1': 0.4935064935064935}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
New best korean model...
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 12:50:21.240165: step: 4/527, loss: 0.001766204833984375
2023-01-21 12:50:22.351810: step: 8/527, loss: 0.0018249511485919356
2023-01-21 12:50:23.446351: step: 12/527, loss: 0.00418467540293932
2023-01-21 12:50:24.580461: step: 16/527, loss: 0.6899878978729248
2023-01-21 12:50:25.699825: step: 20/527, loss: 0.00040187835111282766
2023-01-21 12:50:26.815283: step: 24/527, loss: 0.027169324457645416
2023-01-21 12:50:27.919176: step: 28/527, loss: 0.00012073517427779734
2023-01-21 12:50:29.016040: step: 32/527, loss: 0.0022459030151367188
2023-01-21 12:50:30.131979: step: 36/527, loss: 0.2458038330078125
2023-01-21 12:50:31.275903: step: 40/527, loss: 0.0003391265927348286
2023-01-21 12:50:32.354792: step: 44/527, loss: 0.0024442672729492188
2023-01-21 12:50:33.489521: step: 48/527, loss: 0.0007211684715002775
2023-01-21 12:50:34.578973: step: 52/527, loss: 6.10351571594947e-06
2023-01-21 12:50:35.691323: step: 56/527, loss: 0.06713838875293732
2023-01-21 12:50:36.769764: step: 60/527, loss: 0.042611028999090195
2023-01-21 12:50:37.897863: step: 64/527, loss: 0.016329767182469368
2023-01-21 12:50:39.015206: step: 68/527, loss: 0.00041255951509810984
2023-01-21 12:50:40.144905: step: 72/527, loss: 0.000431060791015625
2023-01-21 12:50:41.273400: step: 76/527, loss: 0.0003604888916015625
2023-01-21 12:50:42.389567: step: 80/527, loss: 0.008054065518081188
2023-01-21 12:50:43.536480: step: 84/527, loss: 7.266998727573082e-05
2023-01-21 12:50:44.647436: step: 88/527, loss: 0.0009488582727499306
2023-01-21 12:50:45.750166: step: 92/527, loss: 0.0487423874437809
2023-01-21 12:50:46.839298: step: 96/527, loss: 0.022173713892698288
2023-01-21 12:50:47.978591: step: 100/527, loss: 0.0002749442937783897
2023-01-21 12:50:49.103394: step: 104/527, loss: 0.04202428087592125
2023-01-21 12:50:50.234280: step: 108/527, loss: 0.00040903090848587453
2023-01-21 12:50:51.334071: step: 112/527, loss: 0.0008162498706951737
2023-01-21 12:50:52.441489: step: 116/527, loss: 0.005974674131721258
2023-01-21 12:50:53.560629: step: 120/527, loss: 0.013529110699892044
2023-01-21 12:50:54.642068: step: 124/527, loss: 1.0824203855008818e-05
2023-01-21 12:50:55.780901: step: 128/527, loss: 8.37326078908518e-05
2023-01-21 12:50:56.873794: step: 132/527, loss: 0.008611869998276234
2023-01-21 12:50:57.994356: step: 136/527, loss: 0.015115928836166859
2023-01-21 12:50:59.082726: step: 140/527, loss: 0.0003482818719930947
2023-01-21 12:51:00.182066: step: 144/527, loss: 0.00032539371750317514
2023-01-21 12:51:01.288408: step: 148/527, loss: 0.0002548217889852822
2023-01-21 12:51:02.487001: step: 152/527, loss: 0.006144905462861061
2023-01-21 12:51:03.615502: step: 156/527, loss: 0.09411268681287766
2023-01-21 12:51:04.728067: step: 160/527, loss: 0.043656542897224426
2023-01-21 12:51:05.828658: step: 164/527, loss: 0.0013572692405432463
2023-01-21 12:51:06.935973: step: 168/527, loss: 0.0023288726806640625
2023-01-21 12:51:08.013409: step: 172/527, loss: 9.498596773482859e-05
2023-01-21 12:51:09.148026: step: 176/527, loss: 0.013548469170928001
2023-01-21 12:51:10.258130: step: 180/527, loss: 0.012019157409667969
2023-01-21 12:51:11.423822: step: 184/527, loss: 0.0277081485837698
2023-01-21 12:51:12.540858: step: 188/527, loss: 0.0001622468262212351
2023-01-21 12:51:13.671509: step: 192/527, loss: 1.6593934560660273e-05
2023-01-21 12:51:14.756960: step: 196/527, loss: 0.00302047748118639
2023-01-21 12:51:15.843963: step: 200/527, loss: 6.866455805720761e-05
2023-01-21 12:51:16.941969: step: 204/527, loss: 0.00016002656775526702
2023-01-21 12:51:18.031329: step: 208/527, loss: 0.006379508879035711
2023-01-21 12:51:19.170877: step: 212/527, loss: 0.000263023393927142
2023-01-21 12:51:20.260084: step: 216/527, loss: 7.629395213371026e-07
2023-01-21 12:51:21.393058: step: 220/527, loss: 4.580021050060168e-05
2023-01-21 12:51:22.494134: step: 224/527, loss: 1.0156631105928682e-05
2023-01-21 12:51:23.624947: step: 228/527, loss: 0.00014934540376998484
2023-01-21 12:51:24.774534: step: 232/527, loss: 0.015249443240463734
2023-01-21 12:51:25.913855: step: 236/527, loss: 0.00617370568215847
2023-01-21 12:51:27.017025: step: 240/527, loss: -8.96453821042087e-06
2023-01-21 12:51:28.172603: step: 244/527, loss: 0.017554283142089844
2023-01-21 12:51:29.297904: step: 248/527, loss: 0.0010103225940838456
2023-01-21 12:51:30.432383: step: 252/527, loss: 0.0008919715764932334
2023-01-21 12:51:31.583961: step: 256/527, loss: 0.00015735626220703125
2023-01-21 12:51:32.708338: step: 260/527, loss: 0.06901970505714417
2023-01-21 12:51:33.822402: step: 264/527, loss: 0.001715803169645369
2023-01-21 12:51:34.935794: step: 268/527, loss: 0.0003888607316184789
2023-01-21 12:51:36.060721: step: 272/527, loss: 8.78334030858241e-05
2023-01-21 12:51:37.168642: step: 276/527, loss: 0.013718223199248314
2023-01-21 12:51:38.328748: step: 280/527, loss: 0.023568343371152878
2023-01-21 12:51:39.489653: step: 284/527, loss: 0.04051399230957031
2023-01-21 12:51:40.632753: step: 288/527, loss: 0.015955161303281784
2023-01-21 12:51:41.751253: step: 292/527, loss: 5.626678466796875e-05
2023-01-21 12:51:42.843287: step: 296/527, loss: 0.03216380998492241
2023-01-21 12:51:43.969687: step: 300/527, loss: 0.45355138182640076
2023-01-21 12:51:45.092197: step: 304/527, loss: 0.0025525183882564306
2023-01-21 12:51:46.178510: step: 308/527, loss: 0.000495356332976371
2023-01-21 12:51:47.296466: step: 312/527, loss: 0.06972599029541016
2023-01-21 12:51:48.401420: step: 316/527, loss: 0.0017675398848950863
2023-01-21 12:51:49.508909: step: 320/527, loss: 2.136230432370212e-05
2023-01-21 12:51:50.641941: step: 324/527, loss: 0.023733139038085938
2023-01-21 12:51:51.788396: step: 328/527, loss: 0.0008713722345419228
2023-01-21 12:51:52.929274: step: 332/527, loss: 0.004764080047607422
2023-01-21 12:51:54.058142: step: 336/527, loss: 0.00717086810618639
2023-01-21 12:51:55.167535: step: 340/527, loss: 0.013479900546371937
2023-01-21 12:51:56.251886: step: 344/527, loss: 0.029793405905365944
2023-01-21 12:51:57.354755: step: 348/527, loss: 0.047705892473459244
2023-01-21 12:51:58.454838: step: 352/527, loss: 8.39233416627394e-06
2023-01-21 12:51:59.570481: step: 356/527, loss: 0.010256004519760609
2023-01-21 12:52:00.691313: step: 360/527, loss: 0.005286789033561945
2023-01-21 12:52:01.796429: step: 364/527, loss: 0.0057773590087890625
2023-01-21 12:52:02.892113: step: 368/527, loss: 0.01985187642276287
2023-01-21 12:52:04.031625: step: 372/527, loss: 0.00016670227341819555
2023-01-21 12:52:05.151940: step: 376/527, loss: 0.01763916201889515
2023-01-21 12:52:06.284240: step: 380/527, loss: 0.0009875297546386719
2023-01-21 12:52:07.396396: step: 384/527, loss: 0.0005135536193847656
2023-01-21 12:52:08.541844: step: 388/527, loss: 0.00759391812607646
2023-01-21 12:52:09.642158: step: 392/527, loss: 0.05786771699786186
2023-01-21 12:52:10.774259: step: 396/527, loss: 0.12877942621707916
2023-01-21 12:52:11.879897: step: 400/527, loss: 0.004709959030151367
2023-01-21 12:52:12.993566: step: 404/527, loss: 3.147125244140625e-05
2023-01-21 12:52:14.116338: step: 408/527, loss: 0.020836830139160156
2023-01-21 12:52:15.249065: step: 412/527, loss: 5.6743621826171875e-05
2023-01-21 12:52:16.371162: step: 416/527, loss: 0.0009277343633584678
2023-01-21 12:52:17.470695: step: 420/527, loss: 0.007067299447953701
2023-01-21 12:52:18.574005: step: 424/527, loss: 0.012654590420424938
2023-01-21 12:52:19.686117: step: 428/527, loss: 0.0031277656089514494
2023-01-21 12:52:20.778502: step: 432/527, loss: 0.039862822741270065
2023-01-21 12:52:21.907203: step: 436/527, loss: 0.00020027162099722773
2023-01-21 12:52:22.988005: step: 440/527, loss: 0.00017857553029898554
2023-01-21 12:52:24.101098: step: 444/527, loss: 0.06750774383544922
2023-01-21 12:52:25.221040: step: 448/527, loss: 0.011782455258071423
2023-01-21 12:52:26.336990: step: 452/527, loss: 2.8991700673941523e-05
2023-01-21 12:52:27.450960: step: 456/527, loss: 0.0020721436012536287
2023-01-21 12:52:28.547178: step: 460/527, loss: 0.05566387251019478
2023-01-21 12:52:29.681828: step: 464/527, loss: 0.00024700164794921875
2023-01-21 12:52:30.786937: step: 468/527, loss: 0.0006452560191974044
2023-01-21 12:52:31.898555: step: 472/527, loss: 0.005136394407600164
2023-01-21 12:52:33.003046: step: 476/527, loss: 0.0005501746782101691
2023-01-21 12:52:34.117958: step: 480/527, loss: 0.0069564348086714745
2023-01-21 12:52:35.212990: step: 484/527, loss: 0.00016579628572799265
2023-01-21 12:52:36.328675: step: 488/527, loss: 1.3947486877441406e-05
2023-01-21 12:52:37.431416: step: 492/527, loss: 0.0009250640869140625
2023-01-21 12:52:38.554344: step: 496/527, loss: 0.0001674652157817036
2023-01-21 12:52:39.710405: step: 500/527, loss: 0.015229320153594017
2023-01-21 12:52:40.838577: step: 504/527, loss: 0.1258329451084137
2023-01-21 12:52:41.939915: step: 508/527, loss: 0.0037277699448168278
2023-01-21 12:52:43.029014: step: 512/527, loss: 0.0003524780331645161
2023-01-21 12:52:44.142704: step: 516/527, loss: 0.00025873183039948344
2023-01-21 12:52:45.250252: step: 520/527, loss: 0.00721473665907979
2023-01-21 12:52:46.377062: step: 524/527, loss: 0.026122570037841797
2023-01-21 12:52:47.525596: step: 528/527, loss: 0.00026359560433775187
2023-01-21 12:52:48.645737: step: 532/527, loss: 0.00775489816442132
2023-01-21 12:52:49.734869: step: 536/527, loss: 0.00010251998901367188
2023-01-21 12:52:50.844244: step: 540/527, loss: 0.03396415710449219
2023-01-21 12:52:51.955467: step: 544/527, loss: 0.00018768310837913305
2023-01-21 12:52:53.074872: step: 548/527, loss: 0.03217964246869087
2023-01-21 12:52:54.191038: step: 552/527, loss: 6.4373016357421875e-06
2023-01-21 12:52:55.286048: step: 556/527, loss: 0.0004201889387331903
2023-01-21 12:52:56.445839: step: 560/527, loss: 0.0004066467226948589
2023-01-21 12:52:57.583783: step: 564/527, loss: 0.05708971247076988
2023-01-21 12:52:58.698413: step: 568/527, loss: 0.0031879425514489412
2023-01-21 12:52:59.848858: step: 572/527, loss: 0.002782726427540183
2023-01-21 12:53:00.964420: step: 576/527, loss: 0.005340910051018
2023-01-21 12:53:02.070264: step: 580/527, loss: 0.0038767815567553043
2023-01-21 12:53:03.179093: step: 584/527, loss: 0.000177288064151071
2023-01-21 12:53:04.297530: step: 588/527, loss: 0.014242363162338734
2023-01-21 12:53:05.392304: step: 592/527, loss: 5.378723290050402e-05
2023-01-21 12:53:06.539594: step: 596/527, loss: 0.008321666158735752
2023-01-21 12:53:07.647065: step: 600/527, loss: 0.0017938613891601562
2023-01-21 12:53:08.791690: step: 604/527, loss: 0.0003831863577943295
2023-01-21 12:53:09.913425: step: 608/527, loss: 3.24249267578125e-05
2023-01-21 12:53:11.007005: step: 612/527, loss: 0.0170148853212595
2023-01-21 12:53:12.114872: step: 616/527, loss: 0.0013532638549804688
2023-01-21 12:53:13.209071: step: 620/527, loss: 0.0026283501647412777
2023-01-21 12:53:14.332442: step: 624/527, loss: 6.437302363337949e-05
2023-01-21 12:53:15.440756: step: 628/527, loss: 1.8501283193472773e-05
2023-01-21 12:53:16.567666: step: 632/527, loss: 0.045168209820985794
2023-01-21 12:53:17.684226: step: 636/527, loss: 0.0015952109824866056
2023-01-21 12:53:18.808784: step: 640/527, loss: 0.0007852554554119706
2023-01-21 12:53:19.936287: step: 644/527, loss: 0.1273820847272873
2023-01-21 12:53:21.026941: step: 648/527, loss: 0.10925483703613281
2023-01-21 12:53:22.105072: step: 652/527, loss: 0.0014833450550213456
2023-01-21 12:53:23.196896: step: 656/527, loss: 0.0017621993320062757
2023-01-21 12:53:24.319184: step: 660/527, loss: 0.0082283029332757
2023-01-21 12:53:25.467478: step: 664/527, loss: 3.051760359085165e-06
2023-01-21 12:53:26.555515: step: 668/527, loss: 0.0025742053985595703
2023-01-21 12:53:27.662542: step: 672/527, loss: 0.018035316839814186
2023-01-21 12:53:28.792061: step: 676/527, loss: 0.010709285736083984
2023-01-21 12:53:29.953085: step: 680/527, loss: 0.027624178677797318
2023-01-21 12:53:31.081926: step: 684/527, loss: 0.00012254714965820312
2023-01-21 12:53:32.224404: step: 688/527, loss: 5.313157816999592e-05
2023-01-21 12:53:33.331683: step: 692/527, loss: 0.00733718927949667
2023-01-21 12:53:34.460603: step: 696/527, loss: 0.0012041091686114669
2023-01-21 12:53:35.601115: step: 700/527, loss: 0.0008429526933468878
2023-01-21 12:53:36.701927: step: 704/527, loss: 0.29358309507369995
2023-01-21 12:53:37.794143: step: 708/527, loss: 0.0001614093780517578
2023-01-21 12:53:38.897011: step: 712/527, loss: 2.918243444582913e-05
2023-01-21 12:53:40.003337: step: 716/527, loss: 0.0034881592728197575
2023-01-21 12:53:41.122459: step: 720/527, loss: 0.00041294097900390625
2023-01-21 12:53:42.239861: step: 724/527, loss: 1.5258789289873675e-06
2023-01-21 12:53:43.368110: step: 728/527, loss: 0.00028905869112350047
2023-01-21 12:53:44.500160: step: 732/527, loss: 0.0005399704095907509
2023-01-21 12:53:45.637468: step: 736/527, loss: 0.0018918991554528475
2023-01-21 12:53:46.765453: step: 740/527, loss: 0.0028125762473791838
2023-01-21 12:53:47.900834: step: 744/527, loss: 0.0004868984397035092
2023-01-21 12:53:49.023727: step: 748/527, loss: 0.00047664641169831157
2023-01-21 12:53:50.157304: step: 752/527, loss: 0.0010887146927416325
2023-01-21 12:53:51.252580: step: 756/527, loss: 0.012437820434570312
2023-01-21 12:53:52.353397: step: 760/527, loss: 1.8024444216280244e-05
2023-01-21 12:53:53.471693: step: 764/527, loss: 0.00011005402484443039
2023-01-21 12:53:54.573759: step: 768/527, loss: -1.1444091796875e-05
2023-01-21 12:53:55.673100: step: 772/527, loss: 0.05443020164966583
2023-01-21 12:53:56.762162: step: 776/527, loss: 3.0040739602554822e-06
2023-01-21 12:53:57.862310: step: 780/527, loss: 0.00017395020404364914
2023-01-21 12:53:58.943183: step: 784/527, loss: 0.0008347511175088584
2023-01-21 12:54:00.062961: step: 788/527, loss: 0.004263973794877529
2023-01-21 12:54:01.199515: step: 792/527, loss: 0.006775474641472101
2023-01-21 12:54:02.302391: step: 796/527, loss: 7.761418237350881e-05
2023-01-21 12:54:03.429188: step: 800/527, loss: -5.34057608092553e-06
2023-01-21 12:54:04.512766: step: 804/527, loss: 0.0009716272470541298
2023-01-21 12:54:05.624086: step: 808/527, loss: 0.0003389358753338456
2023-01-21 12:54:06.745495: step: 812/527, loss: 1.8441548347473145
2023-01-21 12:54:07.843437: step: 816/527, loss: 0.00016536712064407766
2023-01-21 12:54:08.927227: step: 820/527, loss: 0.010838508605957031
2023-01-21 12:54:10.033514: step: 824/527, loss: 0.020152568817138672
2023-01-21 12:54:11.122550: step: 828/527, loss: 0.6454117298126221
2023-01-21 12:54:12.237346: step: 832/527, loss: 0.01870269887149334
2023-01-21 12:54:13.363669: step: 836/527, loss: 0.05858135223388672
2023-01-21 12:54:14.497376: step: 840/527, loss: 0.00013771058002021164
2023-01-21 12:54:15.640970: step: 844/527, loss: 0.8861551284790039
2023-01-21 12:54:16.797664: step: 848/527, loss: 0.021503638476133347
2023-01-21 12:54:17.897910: step: 852/527, loss: 0.0011535168159753084
2023-01-21 12:54:19.027434: step: 856/527, loss: 0.013278389349579811
2023-01-21 12:54:20.175408: step: 860/527, loss: 0.0007536888588219881
2023-01-21 12:54:21.287393: step: 864/527, loss: 0.010644722729921341
2023-01-21 12:54:22.403588: step: 868/527, loss: 0.0001066207914846018
2023-01-21 12:54:23.536967: step: 872/527, loss: 0.023304367437958717
2023-01-21 12:54:24.632586: step: 876/527, loss: 0.002553558209910989
2023-01-21 12:54:25.722471: step: 880/527, loss: 0.0006555200088769197
2023-01-21 12:54:26.824405: step: 884/527, loss: 0.014714241027832031
2023-01-21 12:54:27.956475: step: 888/527, loss: 0.012941170483827591
2023-01-21 12:54:29.109894: step: 892/527, loss: 0.0006715774652548134
2023-01-21 12:54:30.258159: step: 896/527, loss: 5.6552886235294864e-05
2023-01-21 12:54:31.388624: step: 900/527, loss: 0.00010981559898937121
2023-01-21 12:54:32.522108: step: 904/527, loss: 1.754760705807712e-05
2023-01-21 12:54:33.606554: step: 908/527, loss: 0.03206386789679527
2023-01-21 12:54:34.696013: step: 912/527, loss: 0.0006968498346395791
2023-01-21 12:54:35.798895: step: 916/527, loss: 0.0064668660052120686
2023-01-21 12:54:36.918155: step: 920/527, loss: 0.05810356140136719
2023-01-21 12:54:38.005942: step: 924/527, loss: 0.035750579088926315
2023-01-21 12:54:39.098161: step: 928/527, loss: 0.00388336181640625
2023-01-21 12:54:40.224526: step: 932/527, loss: 0.0009361266857013106
2023-01-21 12:54:41.314439: step: 936/527, loss: 0.0003735542413778603
2023-01-21 12:54:42.415421: step: 940/527, loss: 0.0012674810132011771
2023-01-21 12:54:43.522184: step: 944/527, loss: 0.0068503376096487045
2023-01-21 12:54:44.662922: step: 948/527, loss: 0.0006660461658611894
2023-01-21 12:54:45.752917: step: 952/527, loss: 2.2506712411995977e-05
2023-01-21 12:54:46.874207: step: 956/527, loss: 0.007409381680190563
2023-01-21 12:54:47.970153: step: 960/527, loss: 0.00022554397583007812
2023-01-21 12:54:49.102780: step: 964/527, loss: 0.048370361328125
2023-01-21 12:54:50.237356: step: 968/527, loss: 0.005554294679313898
2023-01-21 12:54:51.334389: step: 972/527, loss: 0.0011470795143395662
2023-01-21 12:54:52.446857: step: 976/527, loss: 0.005016994196921587
2023-01-21 12:54:53.560141: step: 980/527, loss: 1.219009280204773
2023-01-21 12:54:54.690454: step: 984/527, loss: 0.0021209719125181437
2023-01-21 12:54:55.805108: step: 988/527, loss: 0.011798858642578125
2023-01-21 12:54:56.912915: step: 992/527, loss: 0.0017727374797686934
2023-01-21 12:54:57.995270: step: 996/527, loss: 0.0008016586070880294
2023-01-21 12:54:59.118201: step: 1000/527, loss: 0.03096199221909046
2023-01-21 12:55:00.239295: step: 1004/527, loss: 0.0004514694446697831
2023-01-21 12:55:01.351456: step: 1008/527, loss: 0.08549900352954865
2023-01-21 12:55:02.461464: step: 1012/527, loss: 0.009777641855180264
2023-01-21 12:55:03.551555: step: 1016/527, loss: 0.09717349708080292
2023-01-21 12:55:04.664445: step: 1020/527, loss: 0.04846920818090439
2023-01-21 12:55:05.761571: step: 1024/527, loss: 0.0003048896905966103
2023-01-21 12:55:06.825521: step: 1028/527, loss: 2.0122528439969756e-05
2023-01-21 12:55:07.912795: step: 1032/527, loss: 0.00011587143671931699
2023-01-21 12:55:09.016082: step: 1036/527, loss: 0.00012168884859420359
2023-01-21 12:55:10.128460: step: 1040/527, loss: 0.011916160583496094
2023-01-21 12:55:11.254900: step: 1044/527, loss: 0.00013866423978470266
2023-01-21 12:55:12.375977: step: 1048/527, loss: 5.53131121705519e-06
2023-01-21 12:55:13.509639: step: 1052/527, loss: 0.00994567945599556
2023-01-21 12:55:14.665534: step: 1056/527, loss: 0.005994987674057484
2023-01-21 12:55:15.812833: step: 1060/527, loss: 0.10977286845445633
2023-01-21 12:55:16.936601: step: 1064/527, loss: 4.7016143071232364e-05
2023-01-21 12:55:18.070354: step: 1068/527, loss: 0.0015819550026208162
2023-01-21 12:55:19.174873: step: 1072/527, loss: 0.0017896651988849044
2023-01-21 12:55:20.269545: step: 1076/527, loss: 0.02225627936422825
2023-01-21 12:55:21.384454: step: 1080/527, loss: 6.008148193359375e-05
2023-01-21 12:55:22.511084: step: 1084/527, loss: 0.00034160615177825093
2023-01-21 12:55:23.619342: step: 1088/527, loss: 0.0885867103934288
2023-01-21 12:55:24.713301: step: 1092/527, loss: 0.0009214401943609118
2023-01-21 12:55:25.826991: step: 1096/527, loss: 0.0040187835693359375
2023-01-21 12:55:26.935777: step: 1100/527, loss: 0.027106191962957382
2023-01-21 12:55:28.042826: step: 1104/527, loss: 0.00024180412583518773
2023-01-21 12:55:29.187601: step: 1108/527, loss: 0.0012324333656579256
2023-01-21 12:55:30.299620: step: 1112/527, loss: 0.04183988645672798
2023-01-21 12:55:31.391442: step: 1116/527, loss: 0.0172333475202322
2023-01-21 12:55:32.539452: step: 1120/527, loss: 0.011845970526337624
2023-01-21 12:55:33.642286: step: 1124/527, loss: 0.0007647991296835244
2023-01-21 12:55:34.764819: step: 1128/527, loss: 0.061222173273563385
2023-01-21 12:55:35.890250: step: 1132/527, loss: 0.015550518408417702
2023-01-21 12:55:36.999432: step: 1136/527, loss: 6.0749054682673886e-05
2023-01-21 12:55:38.124919: step: 1140/527, loss: 0.006541633512824774
2023-01-21 12:55:39.223378: step: 1144/527, loss: 0.0026005746331065893
2023-01-21 12:55:40.342421: step: 1148/527, loss: 0.0006946563953533769
2023-01-21 12:55:41.496803: step: 1152/527, loss: 8.802414231467992e-05
2023-01-21 12:55:42.625040: step: 1156/527, loss: 0.001219034194946289
2023-01-21 12:55:43.716147: step: 1160/527, loss: 5.6743621826171875e-05
2023-01-21 12:55:44.851931: step: 1164/527, loss: 8.296967280330136e-06
2023-01-21 12:55:46.000975: step: 1168/527, loss: 0.011019611731171608
2023-01-21 12:55:47.137622: step: 1172/527, loss: 0.42521238327026367
2023-01-21 12:55:48.241833: step: 1176/527, loss: 0.0013048171531409025
2023-01-21 12:55:49.380708: step: 1180/527, loss: 0.007025241851806641
2023-01-21 12:55:50.513727: step: 1184/527, loss: 0.004977226257324219
2023-01-21 12:55:51.641711: step: 1188/527, loss: 0.055217742919921875
2023-01-21 12:55:52.739788: step: 1192/527, loss: 0.0002927780442405492
2023-01-21 12:55:53.822830: step: 1196/527, loss: 0.0019769668579101562
2023-01-21 12:55:54.967307: step: 1200/527, loss: 1.296997106692288e-05
2023-01-21 12:55:56.091394: step: 1204/527, loss: 2.18868262891192e-05
2023-01-21 12:55:57.246081: step: 1208/527, loss: 9.34600830078125e-05
2023-01-21 12:55:58.347225: step: 1212/527, loss: 0.01757803000509739
2023-01-21 12:55:59.478128: step: 1216/527, loss: 0.0003582954523153603
2023-01-21 12:56:00.603922: step: 1220/527, loss: 0.00045013427734375
2023-01-21 12:56:01.712627: step: 1224/527, loss: 0.0015214920276775956
2023-01-21 12:56:02.812156: step: 1228/527, loss: 0.001737403916195035
2023-01-21 12:56:03.908798: step: 1232/527, loss: 0.047087907791137695
2023-01-21 12:56:05.008611: step: 1236/527, loss: 2.86102294921875e-06
2023-01-21 12:56:06.096676: step: 1240/527, loss: 0.0061502461321651936
2023-01-21 12:56:07.199820: step: 1244/527, loss: 0.0016082762740552425
2023-01-21 12:56:08.300398: step: 1248/527, loss: 0.000289726274786517
2023-01-21 12:56:09.408314: step: 1252/527, loss: 0.0002201557217631489
2023-01-21 12:56:10.497449: step: 1256/527, loss: 0.013474225997924805
2023-01-21 12:56:11.605719: step: 1260/527, loss: 0.002567243529483676
2023-01-21 12:56:12.740441: step: 1264/527, loss: 0.00024967192439362407
2023-01-21 12:56:13.858699: step: 1268/527, loss: 0.012307548895478249
2023-01-21 12:56:14.979158: step: 1272/527, loss: 0.00017652512178756297
2023-01-21 12:56:16.092893: step: 1276/527, loss: 2.3937225705594756e-05
2023-01-21 12:56:17.181365: step: 1280/527, loss: 2.4080276489257812e-05
2023-01-21 12:56:18.289814: step: 1284/527, loss: 0.048406124114990234
2023-01-21 12:56:19.370311: step: 1288/527, loss: 8.94546537892893e-05
2023-01-21 12:56:20.452697: step: 1292/527, loss: 0.02050619199872017
2023-01-21 12:56:21.552547: step: 1296/527, loss: 0.00960388220846653
2023-01-21 12:56:22.652194: step: 1300/527, loss: 0.0002726554812397808
2023-01-21 12:56:23.761445: step: 1304/527, loss: 0.0005056381342001259
2023-01-21 12:56:24.877715: step: 1308/527, loss: 0.0070314412005245686
2023-01-21 12:56:25.995152: step: 1312/527, loss: 0.00028743743314407766
2023-01-21 12:56:27.066658: step: 1316/527, loss: 0.005698060616850853
2023-01-21 12:56:28.175540: step: 1320/527, loss: 0.00011711120168911293
2023-01-21 12:56:29.288633: step: 1324/527, loss: 4.310607982915826e-05
2023-01-21 12:56:30.436171: step: 1328/527, loss: 0.05271758884191513
2023-01-21 12:56:31.552387: step: 1332/527, loss: 0.0021932125091552734
2023-01-21 12:56:32.636043: step: 1336/527, loss: 9.078979201149195e-05
2023-01-21 12:56:33.806677: step: 1340/527, loss: 0.04548444598913193
2023-01-21 12:56:34.952771: step: 1344/527, loss: 0.016104459762573242
2023-01-21 12:56:36.042992: step: 1348/527, loss: 3.34262840624433e-05
2023-01-21 12:56:37.148954: step: 1352/527, loss: 0.0012344360584393144
2023-01-21 12:56:38.289665: step: 1356/527, loss: 7.467270188499242e-05
2023-01-21 12:56:39.388740: step: 1360/527, loss: 0.002158641815185547
2023-01-21 12:56:40.499238: step: 1364/527, loss: 0.0477568618953228
2023-01-21 12:56:41.628627: step: 1368/527, loss: 0.059859372675418854
2023-01-21 12:56:42.770528: step: 1372/527, loss: 0.09371032565832138
2023-01-21 12:56:43.878794: step: 1376/527, loss: 0.000169658669619821
2023-01-21 12:56:45.011060: step: 1380/527, loss: 0.002362537197768688
2023-01-21 12:56:46.105185: step: 1384/527, loss: 0.0005471229669637978
2023-01-21 12:56:47.224340: step: 1388/527, loss: 0.07060647755861282
2023-01-21 12:56:48.342037: step: 1392/527, loss: 0.00016651154146529734
2023-01-21 12:56:49.436105: step: 1396/527, loss: 0.0010497093899175525
2023-01-21 12:56:50.596263: step: 1400/527, loss: 0.03735800087451935
2023-01-21 12:56:51.710921: step: 1404/527, loss: 0.0015563011402264237
2023-01-21 12:56:52.794830: step: 1408/527, loss: 0.0012318610679358244
2023-01-21 12:56:53.922693: step: 1412/527, loss: 0.03470268473029137
2023-01-21 12:56:55.021782: step: 1416/527, loss: 0.012385845184326172
2023-01-21 12:56:56.121010: step: 1420/527, loss: 0.00012826919555664062
2023-01-21 12:56:57.237571: step: 1424/527, loss: 0.012933493591845036
2023-01-21 12:56:58.335638: step: 1428/527, loss: 0.03939790651202202
2023-01-21 12:56:59.432204: step: 1432/527, loss: 0.000185108176083304
2023-01-21 12:57:00.537963: step: 1436/527, loss: 0.04531509801745415
2023-01-21 12:57:01.648514: step: 1440/527, loss: 0.014312363229691982
2023-01-21 12:57:02.777312: step: 1444/527, loss: 0.021743202582001686
2023-01-21 12:57:03.905294: step: 1448/527, loss: 0.0236084945499897
2023-01-21 12:57:05.022534: step: 1452/527, loss: 0.009135056287050247
2023-01-21 12:57:06.147963: step: 1456/527, loss: 8.316039748024195e-05
2023-01-21 12:57:07.253749: step: 1460/527, loss: 3.046989513677545e-05
2023-01-21 12:57:08.349673: step: 1464/527, loss: 4.8160552978515625e-05
2023-01-21 12:57:09.440473: step: 1468/527, loss: 0.003084564348682761
2023-01-21 12:57:10.557163: step: 1472/527, loss: 0.013629436492919922
2023-01-21 12:57:11.685031: step: 1476/527, loss: 0.0019363403553143144
2023-01-21 12:57:12.770625: step: 1480/527, loss: 7.057189577608369e-06
2023-01-21 12:57:13.863442: step: 1484/527, loss: 1.945495569088962e-05
2023-01-21 12:57:14.987322: step: 1488/527, loss: 0.002985095838084817
2023-01-21 12:57:16.088894: step: 1492/527, loss: 0.004210758022964001
2023-01-21 12:57:17.207590: step: 1496/527, loss: 0.0003247261047363281
2023-01-21 12:57:18.305150: step: 1500/527, loss: 2.6798248654813506e-05
2023-01-21 12:57:19.454932: step: 1504/527, loss: 9.956360736396164e-05
2023-01-21 12:57:20.556870: step: 1508/527, loss: 0.0003939390298910439
2023-01-21 12:57:21.656612: step: 1512/527, loss: 0.0005599975702352822
2023-01-21 12:57:22.737901: step: 1516/527, loss: 0.03544482961297035
2023-01-21 12:57:23.850459: step: 1520/527, loss: -4.95910626341356e-06
2023-01-21 12:57:24.952524: step: 1524/527, loss: 0.0007112980238161981
2023-01-21 12:57:26.044542: step: 1528/527, loss: 1.3089180356473662e-05
2023-01-21 12:57:27.136081: step: 1532/527, loss: 0.004083013627678156
2023-01-21 12:57:28.247613: step: 1536/527, loss: 3.24249276673072e-06
2023-01-21 12:57:29.352262: step: 1540/527, loss: 0.0021806717850267887
2023-01-21 12:57:30.451892: step: 1544/527, loss: 2.5558472771081142e-05
2023-01-21 12:57:31.555945: step: 1548/527, loss: 0.0030982017051428556
2023-01-21 12:57:32.685678: step: 1552/527, loss: 0.007536888588219881
2023-01-21 12:57:33.790601: step: 1556/527, loss: 0.02356109581887722
2023-01-21 12:57:34.903686: step: 1560/527, loss: 0.006760883145034313
2023-01-21 12:57:36.026701: step: 1564/527, loss: 0.01896200142800808
2023-01-21 12:57:37.148667: step: 1568/527, loss: 0.0005858421791344881
2023-01-21 12:57:38.237770: step: 1572/527, loss: 0.003640842391178012
2023-01-21 12:57:39.342113: step: 1576/527, loss: 6.251335435081273e-05
2023-01-21 12:57:40.469260: step: 1580/527, loss: 3.0326846172101796e-05
2023-01-21 12:57:41.608788: step: 1584/527, loss: 0.0047454833984375
2023-01-21 12:57:42.729022: step: 1588/527, loss: 0.00029392243595793843
2023-01-21 12:57:43.830490: step: 1592/527, loss: 0.07987356185913086
2023-01-21 12:57:44.918836: step: 1596/527, loss: 8.583068620282575e-07
2023-01-21 12:57:46.059236: step: 1600/527, loss: 0.0006755829090252519
2023-01-21 12:57:47.177638: step: 1604/527, loss: 0.006155490875244141
2023-01-21 12:57:48.278702: step: 1608/527, loss: 1.735687328618951e-05
2023-01-21 12:57:49.410877: step: 1612/527, loss: 0.0001844406215241179
2023-01-21 12:57:50.510927: step: 1616/527, loss: 0.06156330183148384
2023-01-21 12:57:51.612217: step: 1620/527, loss: 3.814698175119702e-07
2023-01-21 12:57:52.724556: step: 1624/527, loss: 0.014499282464385033
2023-01-21 12:57:53.838811: step: 1628/527, loss: 0.0004937172052450478
2023-01-21 12:57:54.933892: step: 1632/527, loss: 0.023496342822909355
2023-01-21 12:57:56.069313: step: 1636/527, loss: 0.013883400708436966
2023-01-21 12:57:57.187259: step: 1640/527, loss: 0.00010795593698276207
2023-01-21 12:57:58.286685: step: 1644/527, loss: 0.0007696152315475047
2023-01-21 12:57:59.392983: step: 1648/527, loss: 0.00011668205115711316
2023-01-21 12:58:00.486360: step: 1652/527, loss: 0.0470890998840332
2023-01-21 12:58:01.601792: step: 1656/527, loss: 0.0008905887953005731
2023-01-21 12:58:02.746735: step: 1660/527, loss: 7.336139969993383e-05
2023-01-21 12:58:03.853501: step: 1664/527, loss: 0.014044332318007946
2023-01-21 12:58:04.991057: step: 1668/527, loss: 0.03412990644574165
2023-01-21 12:58:06.063865: step: 1672/527, loss: 0.0004258155822753906
2023-01-21 12:58:07.218214: step: 1676/527, loss: 0.17165128886699677
2023-01-21 12:58:08.352854: step: 1680/527, loss: 2.040863000729587e-05
2023-01-21 12:58:09.465918: step: 1684/527, loss: 0.007773781195282936
2023-01-21 12:58:10.565487: step: 1688/527, loss: 0.001656913897022605
2023-01-21 12:58:11.681930: step: 1692/527, loss: 0.00026006699772551656
2023-01-21 12:58:12.821854: step: 1696/527, loss: 0.0008150100475177169
2023-01-21 12:58:13.919541: step: 1700/527, loss: 0.00010719299461925402
2023-01-21 12:58:15.053813: step: 1704/527, loss: 0.0021101476158946753
2023-01-21 12:58:16.185899: step: 1708/527, loss: 1.1444091796875e-05
2023-01-21 12:58:17.313468: step: 1712/527, loss: 1.3256072634248994e-05
2023-01-21 12:58:18.417546: step: 1716/527, loss: 0.09243179112672806
2023-01-21 12:58:19.526528: step: 1720/527, loss: 0.004834366030991077
2023-01-21 12:58:20.658005: step: 1724/527, loss: 0.02069120481610298
2023-01-21 12:58:21.750557: step: 1728/527, loss: 4.749298022943549e-05
2023-01-21 12:58:22.879140: step: 1732/527, loss: 0.044396206736564636
2023-01-21 12:58:24.004841: step: 1736/527, loss: 0.07142200320959091
2023-01-21 12:58:25.133005: step: 1740/527, loss: 2.098083541568485e-06
2023-01-21 12:58:26.243109: step: 1744/527, loss: 0.023014020174741745
2023-01-21 12:58:27.344857: step: 1748/527, loss: 0.0004960060468874872
2023-01-21 12:58:28.459575: step: 1752/527, loss: 0.004168892279267311
2023-01-21 12:58:29.573942: step: 1756/527, loss: 0.023257827386260033
2023-01-21 12:58:30.665660: step: 1760/527, loss: 0.02803640440106392
2023-01-21 12:58:31.807010: step: 1764/527, loss: 0.0006607056129723787
2023-01-21 12:58:32.910273: step: 1768/527, loss: 0.0049835206009447575
2023-01-21 12:58:34.031829: step: 1772/527, loss: 0.00020532608323264867
2023-01-21 12:58:35.138923: step: 1776/527, loss: 0.000135326394229196
2023-01-21 12:58:36.239697: step: 1780/527, loss: 4.57763690064894e-06
2023-01-21 12:58:37.336307: step: 1784/527, loss: 1.7547608877066523e-05
2023-01-21 12:58:38.447558: step: 1788/527, loss: 0.005622482392936945
2023-01-21 12:58:39.570890: step: 1792/527, loss: 0.03742418438196182
2023-01-21 12:58:40.697451: step: 1796/527, loss: 0.04316005855798721
2023-01-21 12:58:41.816763: step: 1800/527, loss: 0.11108584702014923
2023-01-21 12:58:42.926970: step: 1804/527, loss: 0.024437524378299713
2023-01-21 12:58:44.038447: step: 1808/527, loss: 0.01724720001220703
2023-01-21 12:58:45.149243: step: 1812/527, loss: 0.005657291505485773
2023-01-21 12:58:46.284797: step: 1816/527, loss: 0.0009641647338867188
2023-01-21 12:58:47.392189: step: 1820/527, loss: 0.008989906869828701
2023-01-21 12:58:48.486246: step: 1824/527, loss: 0.0002552032528910786
2023-01-21 12:58:49.562345: step: 1828/527, loss: 7.190704491222277e-05
2023-01-21 12:58:50.684578: step: 1832/527, loss: 0.00037364958552643657
2023-01-21 12:58:51.795461: step: 1836/527, loss: 0.00011749268014682457
2023-01-21 12:58:52.893183: step: 1840/527, loss: 0.0005306244129315019
2023-01-21 12:58:53.994344: step: 1844/527, loss: 7.953643944347277e-05
2023-01-21 12:58:55.149145: step: 1848/527, loss: 0.006087684538215399
2023-01-21 12:58:56.254193: step: 1852/527, loss: 0.0018959998851642013
2023-01-21 12:58:57.352096: step: 1856/527, loss: 2.6178360712947324e-05
2023-01-21 12:58:58.441802: step: 1860/527, loss: 0.0012657642364501953
2023-01-21 12:58:59.523802: step: 1864/527, loss: 0.0009847640758380294
2023-01-21 12:59:00.617519: step: 1868/527, loss: 0.13579325377941132
2023-01-21 12:59:01.738755: step: 1872/527, loss: 7.162093970691785e-05
2023-01-21 12:59:02.895312: step: 1876/527, loss: 0.002620744751766324
2023-01-21 12:59:04.020425: step: 1880/527, loss: 0.0008141041034832597
2023-01-21 12:59:05.112029: step: 1884/527, loss: 0.001517009804956615
2023-01-21 12:59:06.250708: step: 1888/527, loss: 0.00018656253814697266
2023-01-21 12:59:07.381661: step: 1892/527, loss: 3.452301098150201e-05
2023-01-21 12:59:08.495516: step: 1896/527, loss: 0.004050922580063343
2023-01-21 12:59:09.622922: step: 1900/527, loss: 0.0002058982936432585
2023-01-21 12:59:10.735569: step: 1904/527, loss: 0.015165328979492188
2023-01-21 12:59:11.857784: step: 1908/527, loss: 0.0002885818830691278
2023-01-21 12:59:12.985769: step: 1912/527, loss: 0.04956522583961487
2023-01-21 12:59:14.072464: step: 1916/527, loss: 0.006954193580895662
2023-01-21 12:59:15.195199: step: 1920/527, loss: 1.0776519957289565e-05
2023-01-21 12:59:16.306599: step: 1924/527, loss: 5.52177443751134e-05
2023-01-21 12:59:17.410286: step: 1928/527, loss: 0.0011095048394054174
2023-01-21 12:59:18.507925: step: 1932/527, loss: 1.52587890625e-05
2023-01-21 12:59:19.627577: step: 1936/527, loss: 0.00528373708948493
2023-01-21 12:59:20.735114: step: 1940/527, loss: 0.033998776227235794
2023-01-21 12:59:21.882866: step: 1944/527, loss: 0.011694718152284622
2023-01-21 12:59:22.972643: step: 1948/527, loss: 0.021518802270293236
2023-01-21 12:59:24.078475: step: 1952/527, loss: 0.0004793167463503778
2023-01-21 12:59:25.176702: step: 1956/527, loss: 0.0004134178161621094
2023-01-21 12:59:26.288430: step: 1960/527, loss: 5.340576535672881e-05
2023-01-21 12:59:27.383927: step: 1964/527, loss: 0.0014345169765874743
2023-01-21 12:59:28.479487: step: 1968/527, loss: 0.002166366670280695
2023-01-21 12:59:29.579777: step: 1972/527, loss: 0.01573152467608452
2023-01-21 12:59:30.661241: step: 1976/527, loss: 4.673004241340095e-06
2023-01-21 12:59:31.797675: step: 1980/527, loss: 0.017440224066376686
2023-01-21 12:59:32.905555: step: 1984/527, loss: 0.00034971238346770406
2023-01-21 12:59:34.015528: step: 1988/527, loss: 6.036758713889867e-05
2023-01-21 12:59:35.126714: step: 1992/527, loss: 2.3746490114717744e-05
2023-01-21 12:59:36.227474: step: 1996/527, loss: 0.0002203464537160471
2023-01-21 12:59:37.341351: step: 2000/527, loss: 0.0018532752292230725
2023-01-21 12:59:38.456137: step: 2004/527, loss: 0.03355293348431587
2023-01-21 12:59:39.580627: step: 2008/527, loss: 0.00022020340838935226
2023-01-21 12:59:40.698721: step: 2012/527, loss: 0.017255593091249466
2023-01-21 12:59:41.817266: step: 2016/527, loss: 0.0003689766163006425
2023-01-21 12:59:42.949842: step: 2020/527, loss: 0.005496406927704811
2023-01-21 12:59:44.068288: step: 2024/527, loss: 0.0512942299246788
2023-01-21 12:59:45.177903: step: 2028/527, loss: 0.036402132362127304
2023-01-21 12:59:46.258576: step: 2032/527, loss: 0.0056473733857274055
2023-01-21 12:59:47.369550: step: 2036/527, loss: 0.032465457916259766
2023-01-21 12:59:48.477647: step: 2040/527, loss: 0.06357435882091522
2023-01-21 12:59:49.600832: step: 2044/527, loss: 0.05133533477783203
2023-01-21 12:59:50.686712: step: 2048/527, loss: 0.0052405837923288345
2023-01-21 12:59:51.791052: step: 2052/527, loss: 0.0002544402959756553
2023-01-21 12:59:52.926340: step: 2056/527, loss: 0.060882568359375
2023-01-21 12:59:54.042923: step: 2060/527, loss: 0.007435703184455633
2023-01-21 12:59:55.147748: step: 2064/527, loss: 0.00031194687471725047
2023-01-21 12:59:56.281199: step: 2068/527, loss: 0.00021085739717818797
2023-01-21 12:59:57.408375: step: 2072/527, loss: 0.03925285115838051
2023-01-21 12:59:58.513670: step: 2076/527, loss: 0.0006368637550622225
2023-01-21 12:59:59.632580: step: 2080/527, loss: 0.00699958810582757
2023-01-21 13:00:00.730397: step: 2084/527, loss: 0.0008995056268759072
2023-01-21 13:00:01.831260: step: 2088/527, loss: 5.9604644775390625e-06
2023-01-21 13:00:02.948965: step: 2092/527, loss: 0.0002870559983421117
2023-01-21 13:00:04.053038: step: 2096/527, loss: 0.015148449689149857
2023-01-21 13:00:05.203617: step: 2100/527, loss: 0.0008975506061688066
2023-01-21 13:00:06.343773: step: 2104/527, loss: 0.007875824347138405
2023-01-21 13:00:07.472771: step: 2108/527, loss: 5.3977968491381034e-05
==================================================
Loss: 0.025
--------------------
Dev: {'event': {'p': 0.6041884816753926, 'r': 0.7683089214380826, 'f1': 0.6764361078546307}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25}
Test: {'event': {'p': 0.631651376146789, 'r': 0.7868571428571428, 'f1': 0.7007633587786259}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25}
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': 25}
Korean: {'event': {'p': 0.5555555555555556, 'r': 0.47619047619047616, 'f1': 0.5128205128205129}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25}
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': 25}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 13:00:48.910218: step: 4/527, loss: 0.0010183810954913497
2023-01-21 13:00:50.070238: step: 8/527, loss: 0.00020923613919876516
2023-01-21 13:00:51.189466: step: 12/527, loss: 0.5303603410720825
2023-01-21 13:00:52.280372: step: 16/527, loss: 6.079673767089844e-05
2023-01-21 13:00:53.424505: step: 20/527, loss: 0.01933899149298668
2023-01-21 13:00:54.531694: step: 24/527, loss: 0.002817249158397317
2023-01-21 13:00:55.662175: step: 28/527, loss: 0.0016889572143554688
2023-01-21 13:00:56.793612: step: 32/527, loss: 0.0008220672607421875
2023-01-21 13:00:57.898613: step: 36/527, loss: 0.00550766009837389
2023-01-21 13:00:59.037383: step: 40/527, loss: 0.006059456150978804
2023-01-21 13:01:00.160124: step: 44/527, loss: 5.588531348621473e-05
2023-01-21 13:01:01.279625: step: 48/527, loss: 0.1003929153084755
2023-01-21 13:01:02.406543: step: 52/527, loss: 0.00033965110196731985
2023-01-21 13:01:03.516153: step: 56/527, loss: 0.0001031875581247732
2023-01-21 13:01:04.655082: step: 60/527, loss: 0.0007342338794842362
2023-01-21 13:01:05.763652: step: 64/527, loss: 0.0012675285106524825
2023-01-21 13:01:06.854521: step: 68/527, loss: 0.03591899946331978
2023-01-21 13:01:07.963576: step: 72/527, loss: 0.0001434326113667339
2023-01-21 13:01:09.062203: step: 76/527, loss: 1.010894811770413e-05
2023-01-21 13:01:10.176108: step: 80/527, loss: 0.026571083813905716
2023-01-21 13:01:11.242658: step: 84/527, loss: 1.5592575437040068e-05
2023-01-21 13:01:12.348688: step: 88/527, loss: 0.052752308547496796
2023-01-21 13:01:13.468973: step: 92/527, loss: 0.000523471855558455
2023-01-21 13:01:14.582709: step: 96/527, loss: 0.003355789463967085
2023-01-21 13:01:15.685160: step: 100/527, loss: 0.0023502351250499487
2023-01-21 13:01:16.812651: step: 104/527, loss: 2.4461745852022432e-05
2023-01-21 13:01:17.934754: step: 108/527, loss: 0.0006574630388058722
2023-01-21 13:01:19.087484: step: 112/527, loss: 0.00421750545501709
2023-01-21 13:01:20.187160: step: 116/527, loss: 5.34057608092553e-06
2023-01-21 13:01:21.276744: step: 120/527, loss: 0.0005490303155966103
2023-01-21 13:01:22.412486: step: 124/527, loss: 0.03997693210840225
2023-01-21 13:01:23.534603: step: 128/527, loss: 1.568794141348917e-05
2023-01-21 13:01:24.620330: step: 132/527, loss: 0.00014743805513717234
2023-01-21 13:01:25.745332: step: 136/527, loss: 0.0006561279296875
2023-01-21 13:01:26.828389: step: 140/527, loss: 0.022336388006806374
2023-01-21 13:01:27.938215: step: 144/527, loss: 0.0007814407581463456
2023-01-21 13:01:29.023930: step: 148/527, loss: 2.0599363779183477e-05
2023-01-21 13:01:30.109081: step: 152/527, loss: 0.0001749038783600554
2023-01-21 13:01:31.229395: step: 156/527, loss: 5.3977968491381034e-05
2023-01-21 13:01:32.315771: step: 160/527, loss: -1.62124638336536e-06
2023-01-21 13:01:33.421517: step: 164/527, loss: 0.016738606616854668
2023-01-21 13:01:34.526940: step: 168/527, loss: 0.0006752967601642013
2023-01-21 13:01:35.657544: step: 172/527, loss: 0.5214818120002747
2023-01-21 13:01:36.793064: step: 176/527, loss: 0.062439534813165665
2023-01-21 13:01:37.881931: step: 180/527, loss: 0.0007843017810955644
2023-01-21 13:01:38.983652: step: 184/527, loss: 0.0010223388671875
2023-01-21 13:01:40.109757: step: 188/527, loss: 1.125335711549269e-05
2023-01-21 13:01:41.191236: step: 192/527, loss: 0.00023536682419944555
2023-01-21 13:01:42.284680: step: 196/527, loss: 0.5942993760108948
2023-01-21 13:01:43.395495: step: 200/527, loss: 0.022959327325224876
2023-01-21 13:01:44.467075: step: 204/527, loss: 0.0006401776918210089
2023-01-21 13:01:45.557371: step: 208/527, loss: 0.00012798310490325093
2023-01-21 13:01:46.663622: step: 212/527, loss: 0.024925947189331055
2023-01-21 13:01:47.763844: step: 216/527, loss: 0.00012664795212913305
2023-01-21 13:01:48.855426: step: 220/527, loss: 0.0014165878528729081
2023-01-21 13:01:49.957742: step: 224/527, loss: 0.003051948733627796
2023-01-21 13:01:51.114854: step: 228/527, loss: 1.4877320609230082e-05
2023-01-21 13:01:52.268597: step: 232/527, loss: 0.004054069519042969
2023-01-21 13:01:53.381606: step: 236/527, loss: 7.691383507335559e-05
2023-01-21 13:01:54.513966: step: 240/527, loss: 0.00014896393986418843
2023-01-21 13:01:55.615837: step: 244/527, loss: 0.0003648281272035092
2023-01-21 13:01:56.741750: step: 248/527, loss: 0.006998110096901655
2023-01-21 13:01:57.834607: step: 252/527, loss: 1.0728836059570312e-05
2023-01-21 13:01:58.971429: step: 256/527, loss: 0.013386535458266735
2023-01-21 13:02:00.088435: step: 260/527, loss: 0.043252088129520416
2023-01-21 13:02:01.272679: step: 264/527, loss: 0.019922543317079544
2023-01-21 13:02:02.416930: step: 268/527, loss: 0.00846567191183567
2023-01-21 13:02:03.527879: step: 272/527, loss: 0.08620081096887589
2023-01-21 13:02:04.644846: step: 276/527, loss: 0.0004235267697367817
2023-01-21 13:02:05.746699: step: 280/527, loss: 0.006584834773093462
2023-01-21 13:02:06.888994: step: 284/527, loss: 0.00011687279038596898
2023-01-21 13:02:07.995043: step: 288/527, loss: 0.0021186829544603825
2023-01-21 13:02:09.089826: step: 292/527, loss: 0.00813441351056099
2023-01-21 13:02:10.201036: step: 296/527, loss: 0.02946958690881729
2023-01-21 13:02:11.327460: step: 300/527, loss: 0.04097137600183487
2023-01-21 13:02:12.431608: step: 304/527, loss: 0.0002875328063964844
2023-01-21 13:02:13.582236: step: 308/527, loss: 0.0013885498046875
2023-01-21 13:02:14.700375: step: 312/527, loss: 0.0017648697830736637
2023-01-21 13:02:15.789381: step: 316/527, loss: 0.00010538101196289062
2023-01-21 13:02:16.893670: step: 320/527, loss: 0.001785659696906805
2023-01-21 13:02:18.058247: step: 324/527, loss: 2.5558472771081142e-05
2023-01-21 13:02:19.182682: step: 328/527, loss: 0.0002652168332133442
2023-01-21 13:02:20.281123: step: 332/527, loss: 1.106262243411038e-05
2023-01-21 13:02:21.387842: step: 336/527, loss: 0.00024337769718840718
2023-01-21 13:02:22.465543: step: 340/527, loss: 0.0001246929168701172
2023-01-21 13:02:23.594996: step: 344/527, loss: 0.3799140453338623
2023-01-21 13:02:24.726548: step: 348/527, loss: 0.0017257691361010075
2023-01-21 13:02:25.887456: step: 352/527, loss: 0.09048710018396378
2023-01-21 13:02:26.992713: step: 356/527, loss: 0.07144976407289505
2023-01-21 13:02:28.130946: step: 360/527, loss: 0.0017267226940020919
2023-01-21 13:02:29.272747: step: 364/527, loss: 0.004371833987534046
2023-01-21 13:02:30.409279: step: 368/527, loss: 0.0002040863037109375
2023-01-21 13:02:31.491254: step: 372/527, loss: 0.13012218475341797
2023-01-21 13:02:32.583397: step: 376/527, loss: 0.039885327219963074
2023-01-21 13:02:33.684456: step: 380/527, loss: 0.00011472702317405492
2023-01-21 13:02:34.786494: step: 384/527, loss: 5.865097136847908e-06
2023-01-21 13:02:35.878681: step: 388/527, loss: 7.65800432418473e-05
2023-01-21 13:02:36.961709: step: 392/527, loss: 8.659363084007055e-05
2023-01-21 13:02:38.076239: step: 396/527, loss: 0.00011491775512695312
2023-01-21 13:02:39.198911: step: 400/527, loss: 0.015918731689453125
2023-01-21 13:02:40.312785: step: 404/527, loss: 0.03349361568689346
2023-01-21 13:02:41.447517: step: 408/527, loss: 2.5844572519417852e-05
2023-01-21 13:02:42.532287: step: 412/527, loss: 3.852844383800402e-05
2023-01-21 13:02:43.634426: step: 416/527, loss: 0.0044349669478833675
2023-01-21 13:02:44.743167: step: 420/527, loss: 0.0073455809615552425
2023-01-21 13:02:45.886595: step: 424/527, loss: 0.006294441409409046
2023-01-21 13:02:47.008623: step: 428/527, loss: -4.434585662238533e-06
2023-01-21 13:02:48.075871: step: 432/527, loss: 2.3365021206700476e-06
2023-01-21 13:02:49.163838: step: 436/527, loss: 0.00024366378784179688
2023-01-21 13:02:50.290716: step: 440/527, loss: 0.0005542278522625566
2023-01-21 13:02:51.407088: step: 444/527, loss: 0.0002375602925894782
2023-01-21 13:02:52.510853: step: 448/527, loss: 0.00012903213792014867
2023-01-21 13:02:53.602812: step: 452/527, loss: 0.0015468597412109375
2023-01-21 13:02:54.734605: step: 456/527, loss: 0.02023620717227459
2023-01-21 13:02:55.832847: step: 460/527, loss: 0.0038557052612304688
2023-01-21 13:02:56.965998: step: 464/527, loss: 0.015799619257450104
2023-01-21 13:02:58.096264: step: 468/527, loss: 0.005543041508644819
2023-01-21 13:02:59.194189: step: 472/527, loss: 0.006198692135512829
2023-01-21 13:03:00.282311: step: 476/527, loss: 0.0052689556032419205
2023-01-21 13:03:01.388763: step: 480/527, loss: 6.732940528308973e-05
2023-01-21 13:03:02.557324: step: 484/527, loss: 9.822846550378017e-06
2023-01-21 13:03:03.676875: step: 488/527, loss: 0.0009836197132244706
2023-01-21 13:03:04.784100: step: 492/527, loss: 0.02634620852768421
2023-01-21 13:03:05.865938: step: 496/527, loss: 0.6064134240150452
2023-01-21 13:03:06.997926: step: 500/527, loss: 0.0001789093075785786
2023-01-21 13:03:08.166619: step: 504/527, loss: 0.0005716324085369706
2023-01-21 13:03:09.284172: step: 508/527, loss: 1.163482647825731e-05
2023-01-21 13:03:10.398130: step: 512/527, loss: 0.0005960464477539062
2023-01-21 13:03:11.478205: step: 516/527, loss: 0.0020146372262388468
2023-01-21 13:03:12.598351: step: 520/527, loss: 0.06055726855993271
2023-01-21 13:03:13.698750: step: 524/527, loss: 0.026302719488739967
2023-01-21 13:03:14.793356: step: 528/527, loss: 7.629396350239404e-07
2023-01-21 13:03:15.892407: step: 532/527, loss: 4.94956984766759e-05
2023-01-21 13:03:17.018637: step: 536/527, loss: 0.05432319641113281
2023-01-21 13:03:18.166062: step: 540/527, loss: 0.012856483459472656
2023-01-21 13:03:19.281793: step: 544/527, loss: 5.054473876953125e-05
2023-01-21 13:03:20.397936: step: 548/527, loss: 0.0013012886047363281
2023-01-21 13:03:21.491103: step: 552/527, loss: 0.0007833242416381836
2023-01-21 13:03:22.650443: step: 556/527, loss: 0.012685108929872513
2023-01-21 13:03:23.745443: step: 560/527, loss: 0.03389434888958931
2023-01-21 13:03:24.833035: step: 564/527, loss: 2.2888185412739404e-05
2023-01-21 13:03:25.945877: step: 568/527, loss: 4.425048973644152e-05
2023-01-21 13:03:27.096067: step: 572/527, loss: 0.01061716116964817
2023-01-21 13:03:28.197982: step: 576/527, loss: 7.915496826171875e-05
2023-01-21 13:03:29.321194: step: 580/527, loss: 0.002999782795086503
2023-01-21 13:03:30.422820: step: 584/527, loss: 0.02184000238776207
2023-01-21 13:03:31.555468: step: 588/527, loss: 0.00471344031393528
2023-01-21 13:03:32.702944: step: 592/527, loss: 4.3106076191179454e-05
2023-01-21 13:03:33.826120: step: 596/527, loss: -1.2779236385540571e-05
2023-01-21 13:03:34.930012: step: 600/527, loss: 0.0029702666215598583
2023-01-21 13:03:36.013235: step: 604/527, loss: 0.00016841889009810984
2023-01-21 13:03:37.146656: step: 608/527, loss: 1.9073486612342094e-07
2023-01-21 13:03:38.244927: step: 612/527, loss: 3.6239625842426904e-06
2023-01-21 13:03:39.333465: step: 616/527, loss: -1.0585785275907256e-05
2023-01-21 13:03:40.448260: step: 620/527, loss: 0.0013824462657794356
2023-01-21 13:03:41.559877: step: 624/527, loss: 2.3269654775504023e-05
2023-01-21 13:03:42.661010: step: 628/527, loss: 4.291534423828125e-06
2023-01-21 13:03:43.788048: step: 632/527, loss: 3.61442580469884e-05
2023-01-21 13:03:44.987983: step: 636/527, loss: 0.012423706240952015
2023-01-21 13:03:46.122468: step: 640/527, loss: 0.0020893097389489412
2023-01-21 13:03:47.208744: step: 644/527, loss: 0.0013291359646245837
2023-01-21 13:03:48.319492: step: 648/527, loss: 0.00030770304147154093
2023-01-21 13:03:49.453421: step: 652/527, loss: 0.02173643186688423
2023-01-21 13:03:50.553441: step: 656/527, loss: 5.1116945542162284e-05
2023-01-21 13:03:51.670770: step: 660/527, loss: 0.004600333981215954
2023-01-21 13:03:52.759489: step: 664/527, loss: 0.000610732997301966
2023-01-21 13:03:53.883243: step: 668/527, loss: 0.04113402217626572
2023-01-21 13:03:55.017116: step: 672/527, loss: 0.015576934441924095
2023-01-21 13:03:56.129925: step: 676/527, loss: 0.01519107911735773
2023-01-21 13:03:57.242094: step: 680/527, loss: 0.03730297088623047
2023-01-21 13:03:58.407950: step: 684/527, loss: 0.010211181826889515
2023-01-21 13:03:59.516172: step: 688/527, loss: 0.00141315464861691
2023-01-21 13:04:00.643046: step: 692/527, loss: 0.02148304134607315
2023-01-21 13:04:01.716742: step: 696/527, loss: 0.015772532671689987
2023-01-21 13:04:02.852122: step: 700/527, loss: -3.0994415283203125e-05
2023-01-21 13:04:03.974589: step: 704/527, loss: 0.00022048949904274195
2023-01-21 13:04:05.117308: step: 708/527, loss: 3.24249267578125e-05
2023-01-21 13:04:06.192841: step: 712/527, loss: 0.0003799438418354839
2023-01-21 13:04:07.288932: step: 716/527, loss: 0.000244140625
2023-01-21 13:04:08.407657: step: 720/527, loss: -5.7220458984375e-06
2023-01-21 13:04:09.513791: step: 724/527, loss: 0.002777671907097101
2023-01-21 13:04:10.601200: step: 728/527, loss: 0.0003267287975177169
2023-01-21 13:04:11.710413: step: 732/527, loss: 0.03407011181116104
2023-01-21 13:04:12.879004: step: 736/527, loss: 0.0017590522766113281
2023-01-21 13:04:13.997249: step: 740/527, loss: 3.7860871088923886e-05
2023-01-21 13:04:15.127115: step: 744/527, loss: 0.044820211827754974
2023-01-21 13:04:16.257645: step: 748/527, loss: 0.04671459272503853
2023-01-21 13:04:17.368625: step: 752/527, loss: 1.125335711549269e-05
2023-01-21 13:04:18.509565: step: 756/527, loss: 0.00063323974609375
2023-01-21 13:04:19.636383: step: 760/527, loss: 0.0019325256580486894
2023-01-21 13:04:20.765859: step: 764/527, loss: 0.014039326459169388
2023-01-21 13:04:21.872283: step: 768/527, loss: 3.4046173823298886e-05
2023-01-21 13:04:22.960497: step: 772/527, loss: 4.816055025003152e-06
2023-01-21 13:04:24.056684: step: 776/527, loss: 0.04074068367481232
2023-01-21 13:04:25.144879: step: 780/527, loss: 7.867813110351562e-06
2023-01-21 13:04:26.270718: step: 784/527, loss: 0.0003578186151571572
2023-01-21 13:04:27.376804: step: 788/527, loss: 0.002159976866096258
2023-01-21 13:04:28.508981: step: 792/527, loss: 0.00041131972102448344
2023-01-21 13:04:29.637292: step: 796/527, loss: 2.193450927734375e-05
2023-01-21 13:04:30.761145: step: 800/527, loss: 0.025296688079833984
2023-01-21 13:04:31.875150: step: 804/527, loss: 1.6784666513558477e-05
2023-01-21 13:04:32.998956: step: 808/527, loss: 0.008298492059111595
2023-01-21 13:04:34.102883: step: 812/527, loss: 0.002295970916748047
2023-01-21 13:04:35.221954: step: 816/527, loss: 0.000373649614630267
2023-01-21 13:04:36.347823: step: 820/527, loss: 0.0001390457182424143
2023-01-21 13:04:37.459851: step: 824/527, loss: 0.0032508850563317537
2023-01-21 13:04:38.527438: step: 828/527, loss: 8.144378807628527e-05
2023-01-21 13:04:39.652969: step: 832/527, loss: 0.008536816574633121
2023-01-21 13:04:40.765256: step: 836/527, loss: 0.0018991470569744706
2023-01-21 13:04:41.904064: step: 840/527, loss: 0.0004013061407022178
2023-01-21 13:04:43.004501: step: 844/527, loss: 0.0006505966302938759
2023-01-21 13:04:44.115987: step: 848/527, loss: 0.0023069381713867188
2023-01-21 13:04:45.227217: step: 852/527, loss: 0.0037331103812903166
2023-01-21 13:04:46.333984: step: 856/527, loss: 0.03898124769330025
2023-01-21 13:04:47.442912: step: 860/527, loss: 0.030019378289580345
2023-01-21 13:04:48.553406: step: 864/527, loss: 0.0012832642532885075
2023-01-21 13:04:49.638408: step: 868/527, loss: 0.0006417751428671181
2023-01-21 13:04:50.740867: step: 872/527, loss: 0.008279609493911266
2023-01-21 13:04:51.851142: step: 876/527, loss: 0.004338455386459827
2023-01-21 13:04:52.945039: step: 880/527, loss: -6.9618222369172145e-06
2023-01-21 13:04:54.048631: step: 884/527, loss: 0.005635071080178022
2023-01-21 13:04:55.179613: step: 888/527, loss: 0.03139848634600639
2023-01-21 13:04:56.269553: step: 892/527, loss: 0.003057193709537387
2023-01-21 13:04:57.373214: step: 896/527, loss: 0.0011819839710369706
2023-01-21 13:04:58.494573: step: 900/527, loss: 0.0004093170282430947
2023-01-21 13:04:59.584798: step: 904/527, loss: 0.00039157868013717234
2023-01-21 13:05:00.691445: step: 908/527, loss: 0.005717468447983265
2023-01-21 13:05:01.835480: step: 912/527, loss: 0.0017227173084393144
2023-01-21 13:05:02.950575: step: 916/527, loss: 0.04964065924286842
2023-01-21 13:05:04.108967: step: 920/527, loss: -3.1471249712922145e-06
2023-01-21 13:05:05.249440: step: 924/527, loss: 8.20159948489163e-06
2023-01-21 13:05:06.363124: step: 928/527, loss: 0.01500558853149414
2023-01-21 13:05:07.499392: step: 932/527, loss: 0.01852703094482422
2023-01-21 13:05:08.596189: step: 936/527, loss: 0.011164474301040173
2023-01-21 13:05:09.698487: step: 940/527, loss: 0.04160509258508682
2023-01-21 13:05:10.786349: step: 944/527, loss: 4.596710641635582e-05
2023-01-21 13:05:11.908206: step: 948/527, loss: 0.01620650291442871
2023-01-21 13:05:13.003284: step: 952/527, loss: 2.09808349609375e-05
2023-01-21 13:05:14.128073: step: 956/527, loss: 0.027567576617002487
2023-01-21 13:05:15.245188: step: 960/527, loss: 0.0025661946274340153
2023-01-21 13:05:16.381633: step: 964/527, loss: 0.019494246691465378
2023-01-21 13:05:17.531368: step: 968/527, loss: 0.00299663539044559
2023-01-21 13:05:18.661742: step: 972/527, loss: 0.00027294160099700093
2023-01-21 13:05:19.766844: step: 976/527, loss: 0.021616365760564804
2023-01-21 13:05:20.893293: step: 980/527, loss: 0.0001171588955912739
2023-01-21 13:05:21.999758: step: 984/527, loss: 0.000850296055432409
2023-01-21 13:05:23.106228: step: 988/527, loss: 0.0005934715154580772
2023-01-21 13:05:24.238978: step: 992/527, loss: 0.000865936279296875
2023-01-21 13:05:25.348510: step: 996/527, loss: 1.049041748046875e-05
2023-01-21 13:05:26.473404: step: 1000/527, loss: 0.0022062300704419613
2023-01-21 13:05:27.560149: step: 1004/527, loss: 0.04919185861945152
2023-01-21 13:05:28.664962: step: 1008/527, loss: 0.04399757459759712
2023-01-21 13:05:29.781184: step: 1012/527, loss: 1.678466833254788e-05
2023-01-21 13:05:30.888131: step: 1016/527, loss: 3.7193296975601697e-06
2023-01-21 13:05:32.002505: step: 1020/527, loss: 2.86102294921875e-06
2023-01-21 13:05:33.103059: step: 1024/527, loss: 0.013213539496064186
2023-01-21 13:05:34.218272: step: 1028/527, loss: 0.00796361081302166
2023-01-21 13:05:35.319533: step: 1032/527, loss: 0.03281097486615181
2023-01-21 13:05:36.441939: step: 1036/527, loss: 0.0012125015491619706
2023-01-21 13:05:37.581415: step: 1040/527, loss: 0.052804186940193176
2023-01-21 13:05:38.715543: step: 1044/527, loss: 0.0011572837829589844
2023-01-21 13:05:39.820024: step: 1048/527, loss: 0.00017013550677802414
2023-01-21 13:05:40.931126: step: 1052/527, loss: 2.0980837689421605e-06
2023-01-21 13:05:42.039639: step: 1056/527, loss: 4.38690176451928e-06
2023-01-21 13:05:43.161150: step: 1060/527, loss: 0.007182884030044079
2023-01-21 13:05:44.294597: step: 1064/527, loss: 2.498626781743951e-05
2023-01-21 13:05:45.386907: step: 1068/527, loss: 1.0395049685030244e-05
2023-01-21 13:05:46.492036: step: 1072/527, loss: 0.00010671615018509328
2023-01-21 13:05:47.595037: step: 1076/527, loss: 0.015565109439194202
2023-01-21 13:05:48.695420: step: 1080/527, loss: 1.101493944588583e-05
2023-01-21 13:05:49.817928: step: 1084/527, loss: 1.1825562069134321e-05
2023-01-21 13:05:50.938698: step: 1088/527, loss: 3.9386748539982364e-05
2023-01-21 13:05:52.076864: step: 1092/527, loss: 3.5953522456111386e-05
2023-01-21 13:05:53.167607: step: 1096/527, loss: 5.2261355449445546e-05
2023-01-21 13:05:54.287577: step: 1100/527, loss: 1.926422191900201e-05
2023-01-21 13:05:55.395539: step: 1104/527, loss: 1.4019013178767636e-05
2023-01-21 13:05:56.497735: step: 1108/527, loss: 0.000263214111328125
2023-01-21 13:05:57.619517: step: 1112/527, loss: 0.011109352111816406
2023-01-21 13:05:58.734535: step: 1116/527, loss: 0.09749965369701385
2023-01-21 13:05:59.855506: step: 1120/527, loss: 0.001010894775390625
2023-01-21 13:06:00.964573: step: 1124/527, loss: 0.0011451244354248047
2023-01-21 13:06:02.059232: step: 1128/527, loss: 0.00038824082002975047
2023-01-21 13:06:03.157171: step: 1132/527, loss: 2.1934512915322557e-05
2023-01-21 13:06:04.267318: step: 1136/527, loss: 4.682540748035535e-05
2023-01-21 13:06:05.345099: step: 1140/527, loss: 3.376007225597277e-05
2023-01-21 13:06:06.465766: step: 1144/527, loss: 2.9373170036706142e-05
2023-01-21 13:06:07.598086: step: 1148/527, loss: 7.247925168485381e-06
2023-01-21 13:06:08.712939: step: 1152/527, loss: -1.1444091796875e-05
2023-01-21 13:06:09.835505: step: 1156/527, loss: 0.000469207763671875
2023-01-21 13:06:10.962184: step: 1160/527, loss: 0.0017791748978197575
2023-01-21 13:06:12.066985: step: 1164/527, loss: 0.00014553070650435984
2023-01-21 13:06:13.175770: step: 1168/527, loss: 0.0007131577003747225
2023-01-21 13:06:14.340843: step: 1172/527, loss: 4.3201445805607364e-05
2023-01-21 13:06:15.459611: step: 1176/527, loss: 0.0005456924554891884
2023-01-21 13:06:16.578083: step: 1180/527, loss: 0.01765928417444229
2023-01-21 13:06:17.682197: step: 1184/527, loss: 0.11211776733398438
2023-01-21 13:06:18.819436: step: 1188/527, loss: 3.9768219721736386e-05
2023-01-21 13:06:19.930792: step: 1192/527, loss: 9.937286813510582e-05
2023-01-21 13:06:21.036544: step: 1196/527, loss: 1.3256073543743696e-05
2023-01-21 13:06:22.160393: step: 1200/527, loss: 0.044927217066287994
2023-01-21 13:06:23.266765: step: 1204/527, loss: 2.689361645025201e-05
2023-01-21 13:06:24.393046: step: 1208/527, loss: 0.00011577606346691027
2023-01-21 13:06:25.516904: step: 1212/527, loss: 0.08493976294994354
2023-01-21 13:06:26.647285: step: 1216/527, loss: 0.05260677635669708
2023-01-21 13:06:27.770241: step: 1220/527, loss: 0.0002827644348144531
2023-01-21 13:06:28.930334: step: 1224/527, loss: 0.00042552949162200093
2023-01-21 13:06:30.038023: step: 1228/527, loss: 0.013245105743408203
2023-01-21 13:06:31.142200: step: 1232/527, loss: 0.0005317687755450606
2023-01-21 13:06:32.291808: step: 1236/527, loss: 4.854202416026965e-05
2023-01-21 13:06:33.392332: step: 1240/527, loss: 0.10594673454761505
2023-01-21 13:06:34.524620: step: 1244/527, loss: -2.86102294921875e-06
2023-01-21 13:06:35.630741: step: 1248/527, loss: 0.0040740966796875
2023-01-21 13:06:36.752616: step: 1252/527, loss: 1.8596649624669226e-06
2023-01-21 13:06:37.876236: step: 1256/527, loss: 8.392333256779239e-05
2023-01-21 13:06:38.980297: step: 1260/527, loss: 0.002207183977589011
2023-01-21 13:06:40.085592: step: 1264/527, loss: 7.45773286325857e-05
2023-01-21 13:06:41.197617: step: 1268/527, loss: 0.003558445256203413
2023-01-21 13:06:42.291433: step: 1272/527, loss: 0.02456035651266575
2023-01-21 13:06:43.382973: step: 1276/527, loss: 2.5129318601102568e-05
2023-01-21 13:06:44.499484: step: 1280/527, loss: 5.34057608092553e-06
2023-01-21 13:06:45.606070: step: 1284/527, loss: 0.0025812149979174137
2023-01-21 13:06:46.716721: step: 1288/527, loss: 0.005806351080536842
2023-01-21 13:06:47.852573: step: 1292/527, loss: 4.27722952736076e-05
2023-01-21 13:06:48.973248: step: 1296/527, loss: 1.5258789289873675e-06
2023-01-21 13:06:50.077597: step: 1300/527, loss: 5.3977968491381034e-05
2023-01-21 13:06:51.171938: step: 1304/527, loss: -3.24249267578125e-05
2023-01-21 13:06:52.264489: step: 1308/527, loss: -9.536740890325746e-07
2023-01-21 13:06:53.383096: step: 1312/527, loss: 0.005859804339706898
2023-01-21 13:06:54.485592: step: 1316/527, loss: 3.14712519866589e-06
2023-01-21 13:06:55.598028: step: 1320/527, loss: 0.010534477420151234
2023-01-21 13:06:56.708129: step: 1324/527, loss: 0.011196709237992764
2023-01-21 13:06:57.807632: step: 1328/527, loss: 0.0004432440036907792
2023-01-21 13:06:58.922460: step: 1332/527, loss: 0.00010795592970680445
2023-01-21 13:07:00.017655: step: 1336/527, loss: 0.023223018273711205
2023-01-21 13:07:01.134444: step: 1340/527, loss: 6.008148557157256e-05
2023-01-21 13:07:02.240582: step: 1344/527, loss: 0.0035771848633885384
2023-01-21 13:07:03.328264: step: 1348/527, loss: 9.346009392174892e-06
2023-01-21 13:07:04.432150: step: 1352/527, loss: 0.03470730409026146
2023-01-21 13:07:05.572525: step: 1356/527, loss: 0.007178306579589844
2023-01-21 13:07:06.662827: step: 1360/527, loss: 0.003159904619678855
2023-01-21 13:07:07.767498: step: 1364/527, loss: 0.0009855270618572831
2023-01-21 13:07:08.888989: step: 1368/527, loss: 0.0003097057342529297
2023-01-21 13:07:10.020266: step: 1372/527, loss: 0.0016050338745117188
2023-01-21 13:07:11.124327: step: 1376/527, loss: 0.002596471458673477
2023-01-21 13:07:12.262958: step: 1380/527, loss: 0.062470052391290665
2023-01-21 13:07:13.377058: step: 1384/527, loss: 0.24535445868968964
2023-01-21 13:07:14.479565: step: 1388/527, loss: 3.0231476557673886e-05
2023-01-21 13:07:15.614472: step: 1392/527, loss: 0.009205102920532227
2023-01-21 13:07:16.715977: step: 1396/527, loss: 0.061215974390506744
2023-01-21 13:07:17.807260: step: 1400/527, loss: 0.011213780380785465
2023-01-21 13:07:18.919820: step: 1404/527, loss: 0.00296440115198493
2023-01-21 13:07:20.017765: step: 1408/527, loss: 0.03813991695642471
2023-01-21 13:07:21.130143: step: 1412/527, loss: 0.0005735396989621222
2023-01-21 13:07:22.261162: step: 1416/527, loss: 0.018656635656952858
2023-01-21 13:07:23.354498: step: 1420/527, loss: 2.117157055181451e-05
2023-01-21 13:07:24.451665: step: 1424/527, loss: 0.0003582954523153603
2023-01-21 13:07:25.557513: step: 1428/527, loss: 0.018941879272460938
2023-01-21 13:07:26.663742: step: 1432/527, loss: 0.01845235750079155
2023-01-21 13:07:27.755778: step: 1436/527, loss: -4.5776364459015895e-06
2023-01-21 13:07:28.858512: step: 1440/527, loss: 0.002491474151611328
2023-01-21 13:07:29.971667: step: 1444/527, loss: 0.0017177581321448088
2023-01-21 13:07:31.071840: step: 1448/527, loss: 0.010931205935776234
2023-01-21 13:07:32.166872: step: 1452/527, loss: 0.0004507213889155537
2023-01-21 13:07:33.261960: step: 1456/527, loss: 5.073547436040826e-05
2023-01-21 13:07:34.367654: step: 1460/527, loss: 3.132819983875379e-05
2023-01-21 13:07:35.463349: step: 1464/527, loss: 0.01725759543478489
2023-01-21 13:07:36.578149: step: 1468/527, loss: 0.0009572029230184853
2023-01-21 13:07:37.699156: step: 1472/527, loss: 4.3487551010912284e-05
2023-01-21 13:07:38.797325: step: 1476/527, loss: -2.6702882678364404e-06
2023-01-21 13:07:39.918460: step: 1480/527, loss: 0.7006246447563171
2023-01-21 13:07:41.043416: step: 1484/527, loss: 0.002730798674747348
2023-01-21 13:07:42.173260: step: 1488/527, loss: 0.0753277838230133
2023-01-21 13:07:43.287462: step: 1492/527, loss: 0.00551605224609375
2023-01-21 13:07:44.393253: step: 1496/527, loss: 0.003438663436099887
2023-01-21 13:07:45.523781: step: 1500/527, loss: 0.027538299560546875
2023-01-21 13:07:46.642791: step: 1504/527, loss: 0.009008407592773438
2023-01-21 13:07:47.761115: step: 1508/527, loss: 0.0002904892317019403
2023-01-21 13:07:48.865880: step: 1512/527, loss: 0.0008724212530069053
2023-01-21 13:07:49.982857: step: 1516/527, loss: 0.0010391712421551347
2023-01-21 13:07:51.082082: step: 1520/527, loss: 0.0008005142444744706
2023-01-21 13:07:52.224264: step: 1524/527, loss: 0.0010977268684655428
2023-01-21 13:07:53.334210: step: 1528/527, loss: 0.0004170417960267514
2023-01-21 13:07:54.473812: step: 1532/527, loss: 6.294250852079131e-06
2023-01-21 13:07:55.596088: step: 1536/527, loss: 0.00033254624577239156
2023-01-21 13:07:56.697721: step: 1540/527, loss: 0.0002576351398602128
2023-01-21 13:07:57.823857: step: 1544/527, loss: 0.000232696533203125
2023-01-21 13:07:58.930029: step: 1548/527, loss: 0.033135127276182175
2023-01-21 13:08:00.026790: step: 1552/527, loss: 0.0002883911074604839
2023-01-21 13:08:01.145826: step: 1556/527, loss: 0.00010004044452216476
2023-01-21 13:08:02.244574: step: 1560/527, loss: 6.189346458995715e-05
2023-01-21 13:08:03.379395: step: 1564/527, loss: 0.05324821546673775
2023-01-21 13:08:04.492095: step: 1568/527, loss: 0.0395817756652832
2023-01-21 13:08:05.610496: step: 1572/527, loss: 0.013029957190155983
2023-01-21 13:08:06.735252: step: 1576/527, loss: 0.08099603652954102
2023-01-21 13:08:07.835945: step: 1580/527, loss: 0.03337879106402397
2023-01-21 13:08:08.925018: step: 1584/527, loss: 0.0006295204511843622
2023-01-21 13:08:10.023445: step: 1588/527, loss: 0.0069828033447265625
2023-01-21 13:08:11.135311: step: 1592/527, loss: 0.005482864566147327
2023-01-21 13:08:12.235558: step: 1596/527, loss: 0.004457187373191118
2023-01-21 13:08:13.321079: step: 1600/527, loss: 0.03366050869226456
2023-01-21 13:08:14.414743: step: 1604/527, loss: 0.002764129778370261
2023-01-21 13:08:15.564463: step: 1608/527, loss: 0.03771534189581871
2023-01-21 13:08:16.658871: step: 1612/527, loss: 0.011908340267837048
2023-01-21 13:08:17.801436: step: 1616/527, loss: 6.160735938465223e-05
2023-01-21 13:08:18.901180: step: 1620/527, loss: 0.00014801026554778218
2023-01-21 13:08:19.994672: step: 1624/527, loss: 9.813308861339465e-05
2023-01-21 13:08:21.105413: step: 1628/527, loss: 0.00015125275240279734
2023-01-21 13:08:22.226027: step: 1632/527, loss: 0.0003359794500283897
2023-01-21 13:08:23.386760: step: 1636/527, loss: 0.0027188300155103207
2023-01-21 13:08:24.506402: step: 1640/527, loss: 0.00016899108595680445
2023-01-21 13:08:25.569128: step: 1644/527, loss: 0.00023493767366744578
2023-01-21 13:08:26.711142: step: 1648/527, loss: 0.007884550839662552
2023-01-21 13:08:27.843725: step: 1652/527, loss: 0.002260875655338168
2023-01-21 13:08:28.937036: step: 1656/527, loss: 5.8650970458984375e-05
2023-01-21 13:08:30.046104: step: 1660/527, loss: 1.5354156857938506e-05
2023-01-21 13:08:31.138845: step: 1664/527, loss: 0.00017910002497956157
2023-01-21 13:08:32.262880: step: 1668/527, loss: 0.015706252306699753
2023-01-21 13:08:33.369542: step: 1672/527, loss: 3.42369094141759e-05
2023-01-21 13:08:34.476379: step: 1676/527, loss: 0.0006195068708620965
2023-01-21 13:08:35.567985: step: 1680/527, loss: 0.00016307830810546875
2023-01-21 13:08:36.699101: step: 1684/527, loss: 0.06390685588121414
2023-01-21 13:08:37.797071: step: 1688/527, loss: 2.1266936528263614e-05
2023-01-21 13:08:38.915772: step: 1692/527, loss: 0.037119198590517044
2023-01-21 13:08:40.051234: step: 1696/527, loss: 0.0014846802223473787
2023-01-21 13:08:41.171728: step: 1700/527, loss: 0.006541252601891756
2023-01-21 13:08:42.311141: step: 1704/527, loss: 0.011773491278290749
2023-01-21 13:08:43.440759: step: 1708/527, loss: 0.0006527423392981291
2023-01-21 13:08:44.553726: step: 1712/527, loss: 2.365112231927924e-05
2023-01-21 13:08:45.654476: step: 1716/527, loss: 0.00015468598576262593
2023-01-21 13:08:46.758796: step: 1720/527, loss: 9.91821252682712e-06
2023-01-21 13:08:47.866379: step: 1724/527, loss: 1.33514404296875e-05
2023-01-21 13:08:48.998217: step: 1728/527, loss: 1.33514404296875e-05
2023-01-21 13:08:50.117782: step: 1732/527, loss: 0.005729102995246649
2023-01-21 13:08:51.192960: step: 1736/527, loss: 0.0004498004855122417
2023-01-21 13:08:52.314532: step: 1740/527, loss: 0.005129242315888405
2023-01-21 13:08:53.434713: step: 1744/527, loss: -6.48498553346144e-06
2023-01-21 13:08:54.554551: step: 1748/527, loss: 0.004732513800263405
2023-01-21 13:08:55.654452: step: 1752/527, loss: 3.738403393072076e-05
2023-01-21 13:08:56.765604: step: 1756/527, loss: 0.0002151489316020161
2023-01-21 13:08:57.889372: step: 1760/527, loss: 0.00011606217594817281
2023-01-21 13:08:59.005018: step: 1764/527, loss: 2.460479663568549e-05
2023-01-21 13:09:00.126102: step: 1768/527, loss: 0.00811548251658678
2023-01-21 13:09:01.294807: step: 1772/527, loss: 0.0005331039428710938
2023-01-21 13:09:02.397186: step: 1776/527, loss: 8.80241350387223e-05
2023-01-21 13:09:03.537575: step: 1780/527, loss: 1.888275073724799e-05
2023-01-21 13:09:04.684250: step: 1784/527, loss: 0.06308841705322266
2023-01-21 13:09:05.791987: step: 1788/527, loss: 0.02843456342816353
2023-01-21 13:09:06.903081: step: 1792/527, loss: 0.0036882401909679174
2023-01-21 13:09:08.020251: step: 1796/527, loss: 0.002012062119320035
2023-01-21 13:09:09.176729: step: 1800/527, loss: 0.00254650111310184
2023-01-21 13:09:10.296764: step: 1804/527, loss: 0.0010662077693268657
2023-01-21 13:09:11.390327: step: 1808/527, loss: 0.020600700750947
2023-01-21 13:09:12.507079: step: 1812/527, loss: 0.05712928622961044
2023-01-21 13:09:13.646607: step: 1816/527, loss: 0.03398609533905983
2023-01-21 13:09:14.743488: step: 1820/527, loss: 0.00028486252995207906
2023-01-21 13:09:15.857583: step: 1824/527, loss: 0.004505062010139227
2023-01-21 13:09:16.965929: step: 1828/527, loss: 0.0001009941115626134
2023-01-21 13:09:18.053583: step: 1832/527, loss: 1.4591216313419864e-05
2023-01-21 13:09:19.165270: step: 1836/527, loss: 0.00011768341209972277
2023-01-21 13:09:20.307336: step: 1840/527, loss: 0.0017599106067791581
2023-01-21 13:09:21.405645: step: 1844/527, loss: 0.009038162417709827
2023-01-21 13:09:22.537227: step: 1848/527, loss: 5.0258637202205136e-05
2023-01-21 13:09:23.627467: step: 1852/527, loss: 0.0001902580406749621
2023-01-21 13:09:24.754912: step: 1856/527, loss: 0.0007579803350381553
2023-01-21 13:09:25.854569: step: 1860/527, loss: 0.015275382436811924
2023-01-21 13:09:26.965991: step: 1864/527, loss: 0.0007547378772869706
2023-01-21 13:09:28.086394: step: 1868/527, loss: -9.5367431640625e-07
2023-01-21 13:09:29.190583: step: 1872/527, loss: 0.0003952980332542211
2023-01-21 13:09:30.314632: step: 1876/527, loss: 8.907318260753527e-05
2023-01-21 13:09:31.418559: step: 1880/527, loss: 0.008313274011015892
2023-01-21 13:09:32.536878: step: 1884/527, loss: 0.003382110735401511
2023-01-21 13:09:33.664147: step: 1888/527, loss: 0.6619184613227844
2023-01-21 13:09:34.761148: step: 1892/527, loss: 3.610849307733588e-05
2023-01-21 13:09:35.854033: step: 1896/527, loss: 0.0012271881569176912
2023-01-21 13:09:36.960906: step: 1900/527, loss: 0.00267200474627316
2023-01-21 13:09:38.090036: step: 1904/527, loss: 0.15486736595630646
2023-01-21 13:09:39.187560: step: 1908/527, loss: 0.0021168231032788754
2023-01-21 13:09:40.266613: step: 1912/527, loss: 0.00607223529368639
2023-01-21 13:09:41.363812: step: 1916/527, loss: 0.0002665519714355469
2023-01-21 13:09:42.463057: step: 1920/527, loss: 0.10925684869289398
2023-01-21 13:09:43.611963: step: 1924/527, loss: 5.9700014389818534e-05
2023-01-21 13:09:44.718110: step: 1928/527, loss: 3.6048892070539296e-05
2023-01-21 13:09:45.880183: step: 1932/527, loss: 0.0004403114435262978
2023-01-21 13:09:47.019100: step: 1936/527, loss: 0.00011177062697242945
2023-01-21 13:09:48.144169: step: 1940/527, loss: 0.017100144177675247
2023-01-21 13:09:49.268905: step: 1944/527, loss: 0.007148170378059149
2023-01-21 13:09:50.386842: step: 1948/527, loss: 6.67572021484375e-06
2023-01-21 13:09:51.494875: step: 1952/527, loss: 0.0005556106334552169
2023-01-21 13:09:52.609498: step: 1956/527, loss: 0.009818457998335361
2023-01-21 13:09:53.714547: step: 1960/527, loss: 0.07918872684240341
2023-01-21 13:09:54.840847: step: 1964/527, loss: 0.001754379365593195
2023-01-21 13:09:55.944973: step: 1968/527, loss: 0.0010378838051110506
2023-01-21 13:09:57.067687: step: 1972/527, loss: 0.017582321539521217
2023-01-21 13:09:58.179524: step: 1976/527, loss: 9.479522850597277e-05
2023-01-21 13:09:59.280231: step: 1980/527, loss: 8.535385131835938e-05
2023-01-21 13:10:00.381312: step: 1984/527, loss: 3.075600034208037e-05
2023-01-21 13:10:01.475272: step: 1988/527, loss: 4.0817263652570546e-05
2023-01-21 13:10:02.634954: step: 1992/527, loss: 0.0017286300426349044
2023-01-21 13:10:03.727579: step: 1996/527, loss: 0.00537528982385993
2023-01-21 13:10:04.856199: step: 2000/527, loss: 0.02022418938577175
2023-01-21 13:10:05.968209: step: 2004/527, loss: 0.00023727417283225805
2023-01-21 13:10:07.075926: step: 2008/527, loss: 9.822845458984375e-05
2023-01-21 13:10:08.213914: step: 2012/527, loss: 0.024771880358457565
2023-01-21 13:10:09.327800: step: 2016/527, loss: 4.2533876694506034e-05
2023-01-21 13:10:10.434024: step: 2020/527, loss: 0.0013359070289880037
2023-01-21 13:10:11.559403: step: 2024/527, loss: 0.0007856369484215975
2023-01-21 13:10:12.683104: step: 2028/527, loss: 0.005607319064438343
2023-01-21 13:10:13.777860: step: 2032/527, loss: 0.0003445625479798764
2023-01-21 13:10:14.916403: step: 2036/527, loss: 0.012133216485381126
2023-01-21 13:10:16.000753: step: 2040/527, loss: 0.0001442909415345639
2023-01-21 13:10:17.090218: step: 2044/527, loss: 0.01182627584785223
2023-01-21 13:10:18.180333: step: 2048/527, loss: 0.00011940002150367945
2023-01-21 13:10:19.300231: step: 2052/527, loss: 0.07088293880224228
2023-01-21 13:10:20.405571: step: 2056/527, loss: 0.0004131317255087197
2023-01-21 13:10:21.553986: step: 2060/527, loss: 0.000994873116724193
2023-01-21 13:10:22.684726: step: 2064/527, loss: 9.822845458984375e-05
2023-01-21 13:10:23.801827: step: 2068/527, loss: 0.002620220184326172
2023-01-21 13:10:24.918120: step: 2072/527, loss: 0.00023956298537086695
2023-01-21 13:10:26.032162: step: 2076/527, loss: 0.09332828223705292
2023-01-21 13:10:27.135525: step: 2080/527, loss: 0.0019372940296307206
2023-01-21 13:10:28.270687: step: 2084/527, loss: 0.00027756692725233734
2023-01-21 13:10:29.377701: step: 2088/527, loss: 0.0053806304931640625
2023-01-21 13:10:30.471570: step: 2092/527, loss: 2.09808349609375e-05
2023-01-21 13:10:31.632429: step: 2096/527, loss: 0.0071763996966183186
2023-01-21 13:10:32.728085: step: 2100/527, loss: 0.04844836890697479
2023-01-21 13:10:33.850817: step: 2104/527, loss: 5.0973892939509824e-05
2023-01-21 13:10:35.010078: step: 2108/527, loss: 0.00014972686767578125
==================================================
Loss: 0.018
--------------------
Dev: {'event': {'p': 0.5886524822695035, 'r': 0.7736351531291611, 'f1': 0.668584579976985}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26}
Test: {'event': {'p': 0.6139184397163121, 'r': 0.7914285714285715, 'f1': 0.691462805791313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26}
Chinese: {'event': {'p': 0.5393258426966292, 'r': 0.8888888888888888, 'f1': 0.6713286713286712}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26}
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': 26}
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': 26}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 13:11:16.561004: step: 4/527, loss: 0.04659261927008629
2023-01-21 13:11:17.675781: step: 8/527, loss: 0.00010786057100631297
2023-01-21 13:11:18.782079: step: 12/527, loss: 0.007655430119484663
2023-01-21 13:11:19.915747: step: 16/527, loss: 0.004889965057373047
2023-01-21 13:11:21.018830: step: 20/527, loss: 0.000278282182989642
2023-01-21 13:11:22.127486: step: 24/527, loss: 0.01794900745153427
2023-01-21 13:11:23.234445: step: 28/527, loss: 0.0062577249482274055
2023-01-21 13:11:24.360624: step: 32/527, loss: 0.0012624741066247225
2023-01-21 13:11:25.484816: step: 36/527, loss: 3.910064697265625e-05
2023-01-21 13:11:26.613360: step: 40/527, loss: 0.0017003059620037675
2023-01-21 13:11:27.718308: step: 44/527, loss: 0.004430055618286133
2023-01-21 13:11:28.813295: step: 48/527, loss: 0.007754278369247913
2023-01-21 13:11:29.927859: step: 52/527, loss: 0.0007696151733398438
2023-01-21 13:11:31.049875: step: 56/527, loss: 0.019739866256713867
2023-01-21 13:11:32.172431: step: 60/527, loss: 0.058429960161447525
2023-01-21 13:11:33.293358: step: 64/527, loss: 7.514953904319555e-05
2023-01-21 13:11:34.401216: step: 68/527, loss: 0.010947608388960361
2023-01-21 13:11:35.558551: step: 72/527, loss: 0.008004666306078434
2023-01-21 13:11:36.665995: step: 76/527, loss: 3.3092499506892636e-05
2023-01-21 13:11:37.805998: step: 80/527, loss: 9.984970529330894e-05
2023-01-21 13:11:38.897363: step: 84/527, loss: 0.002942848252132535
2023-01-21 13:11:40.017416: step: 88/527, loss: 0.01419687271118164
2023-01-21 13:11:41.146645: step: 92/527, loss: 0.009990310296416283
2023-01-21 13:11:42.273319: step: 96/527, loss: 0.002063322113826871
2023-01-21 13:11:43.375382: step: 100/527, loss: 8.130073547363281e-05
2023-01-21 13:11:44.494331: step: 104/527, loss: -1.5068053471622989e-05
2023-01-21 13:11:45.574819: step: 108/527, loss: 0.018062258139252663
2023-01-21 13:11:46.680172: step: 112/527, loss: 0.027595043182373047
2023-01-21 13:11:47.779529: step: 116/527, loss: 0.00021009445481467992
2023-01-21 13:11:48.886954: step: 120/527, loss: 0.0014997005928307772
2023-01-21 13:11:49.982063: step: 124/527, loss: 3.0040739602554822e-06
2023-01-21 13:11:51.105011: step: 128/527, loss: 0.002384185791015625
2023-01-21 13:11:52.235963: step: 132/527, loss: 0.0009252548334188759
2023-01-21 13:11:53.331017: step: 136/527, loss: 0.00022974015155341476
2023-01-21 13:11:54.423404: step: 140/527, loss: 0.000465202349005267
2023-01-21 13:11:55.529495: step: 144/527, loss: 3.910064697265625e-05
2023-01-21 13:11:56.673339: step: 148/527, loss: 9.72747784544481e-06
2023-01-21 13:11:57.780618: step: 152/527, loss: 0.0008228778606280684
2023-01-21 13:11:58.871181: step: 156/527, loss: 0.01465139351785183
2023-01-21 13:11:59.984628: step: 160/527, loss: 3.1089784897631034e-05
2023-01-21 13:12:01.098078: step: 164/527, loss: 0.06945057213306427
2023-01-21 13:12:02.229375: step: 168/527, loss: 1.5163422176556196e-05
2023-01-21 13:12:03.340120: step: 172/527, loss: 0.03608589246869087
2023-01-21 13:12:04.462987: step: 176/527, loss: 7.486343383789062e-05
2023-01-21 13:12:05.603118: step: 180/527, loss: 0.00010023116919910535
2023-01-21 13:12:06.707572: step: 184/527, loss: 0.011927126906812191
2023-01-21 13:12:07.794911: step: 188/527, loss: 0.051439668983221054
2023-01-21 13:12:08.920595: step: 192/527, loss: 0.021615982055664062
2023-01-21 13:12:10.045081: step: 196/527, loss: 0.009925270453095436
2023-01-21 13:12:11.144595: step: 200/527, loss: 0.03155670315027237
2023-01-21 13:12:12.289887: step: 204/527, loss: 0.00040836335392668843
2023-01-21 13:12:13.417888: step: 208/527, loss: 8.95500197657384e-05
2023-01-21 13:12:14.518651: step: 212/527, loss: 0.0009998321766033769
2023-01-21 13:12:15.633878: step: 216/527, loss: 0.017232513055205345
2023-01-21 13:12:16.768714: step: 220/527, loss: 6.971359835006297e-05
2023-01-21 13:12:17.849712: step: 224/527, loss: 0.0012765884166583419
2023-01-21 13:12:18.952682: step: 228/527, loss: 0.0014699936145916581
2023-01-21 13:12:20.103767: step: 232/527, loss: 0.007521629333496094
2023-01-21 13:12:21.213763: step: 236/527, loss: 0.008309364318847656
2023-01-21 13:12:22.329332: step: 240/527, loss: 0.00019435881404206157
2023-01-21 13:12:23.449666: step: 244/527, loss: 0.027112485840916634
2023-01-21 13:12:24.576880: step: 248/527, loss: 0.0009975910652428865
2023-01-21 13:12:25.666688: step: 252/527, loss: 0.007272863294929266
2023-01-21 13:12:26.823305: step: 256/527, loss: 0.08685588836669922
2023-01-21 13:12:27.928177: step: 260/527, loss: 0.0005687713855877519
2023-01-21 13:12:29.013992: step: 264/527, loss: 2.241134825453628e-05
2023-01-21 13:12:30.106836: step: 268/527, loss: 0.0033886912278831005
2023-01-21 13:12:31.236041: step: 272/527, loss: 0.0006528854137286544
2023-01-21 13:12:32.336696: step: 276/527, loss: 2.555847095209174e-05
2023-01-21 13:12:33.417963: step: 280/527, loss: -4.0531085687689483e-07
2023-01-21 13:12:34.522194: step: 284/527, loss: 5.5313107623078395e-06
2023-01-21 13:12:35.631685: step: 288/527, loss: 0.0001583099365234375
2023-01-21 13:12:36.748728: step: 292/527, loss: 0.010297585278749466
2023-01-21 13:12:37.863378: step: 296/527, loss: 0.0018918036948889494
2023-01-21 13:12:38.951665: step: 300/527, loss: 0.00018196106248069555
2023-01-21 13:12:40.067140: step: 304/527, loss: 0.03812684863805771
2023-01-21 13:12:41.193954: step: 308/527, loss: 0.008020401000976562
2023-01-21 13:12:42.319330: step: 312/527, loss: 0.00012130737013649195
2023-01-21 13:12:43.426405: step: 316/527, loss: 0.0010629653697833419
2023-01-21 13:12:44.570699: step: 320/527, loss: 0.0002644539054017514
2023-01-21 13:12:45.698149: step: 324/527, loss: 0.14924278855323792
2023-01-21 13:12:46.798669: step: 328/527, loss: 0.0007692337385378778
2023-01-21 13:12:47.892072: step: 332/527, loss: 0.002264023059979081
2023-01-21 13:12:48.998873: step: 336/527, loss: 0.009740781970322132
2023-01-21 13:12:50.118844: step: 340/527, loss: 4.100799742445815e-06
2023-01-21 13:12:51.235741: step: 344/527, loss: 3.833770824712701e-05
2023-01-21 13:12:52.362019: step: 348/527, loss: 0.14693059027194977
2023-01-21 13:12:53.484072: step: 352/527, loss: 1.09672544112982e-06
2023-01-21 13:12:54.599328: step: 356/527, loss: 6.67572021484375e-06
2023-01-21 13:12:55.717701: step: 360/527, loss: 0.0019189835293218493
2023-01-21 13:12:56.833898: step: 364/527, loss: 0.0016914368607103825
2023-01-21 13:12:57.953832: step: 368/527, loss: -1.716613724056515e-06
2023-01-21 13:12:59.088577: step: 372/527, loss: 0.017615128308534622
2023-01-21 13:13:00.188082: step: 376/527, loss: 0.0002153396635549143
2023-01-21 13:13:01.349768: step: 380/527, loss: 0.0014480591053143144
2023-01-21 13:13:02.443725: step: 384/527, loss: 0.00011329651169944555
2023-01-21 13:13:03.546672: step: 388/527, loss: 0.006583022885024548
2023-01-21 13:13:04.690956: step: 392/527, loss: 0.02748594433069229
2023-01-21 13:13:05.790143: step: 396/527, loss: 0.027559852227568626
2023-01-21 13:13:06.917991: step: 400/527, loss: 7.410049147438258e-05
2023-01-21 13:13:08.020979: step: 404/527, loss: 0.005669021513313055
2023-01-21 13:13:09.144925: step: 408/527, loss: 0.00027675629826262593
2023-01-21 13:13:10.255858: step: 412/527, loss: 1.621246337890625e-05
2023-01-21 13:13:11.346892: step: 416/527, loss: -3.623962356869015e-06
2023-01-21 13:13:12.418966: step: 420/527, loss: 0.026864241808652878
2023-01-21 13:13:13.522933: step: 424/527, loss: 0.030648043379187584
2023-01-21 13:13:14.613178: step: 428/527, loss: 0.00031280517578125
2023-01-21 13:13:15.712226: step: 432/527, loss: 0.00034165382385253906
2023-01-21 13:13:16.836605: step: 436/527, loss: 2.765655517578125e-05
2023-01-21 13:13:17.963853: step: 440/527, loss: 0.0003564834769349545
2023-01-21 13:13:19.089317: step: 444/527, loss: 0.001743316650390625
2023-01-21 13:13:20.250655: step: 448/527, loss: 0.0004419327015057206
2023-01-21 13:13:21.388281: step: 452/527, loss: 0.0009684563265182078
2023-01-21 13:13:22.497960: step: 456/527, loss: 0.011153602041304111
2023-01-21 13:13:23.579282: step: 460/527, loss: 0.039178136736154556
2023-01-21 13:13:24.697254: step: 464/527, loss: 0.004218864254653454
2023-01-21 13:13:25.777332: step: 468/527, loss: 4.596710277837701e-05
2023-01-21 13:13:26.891079: step: 472/527, loss: 0.4207834303379059
2023-01-21 13:13:28.021376: step: 476/527, loss: 2.7513504392118193e-05
2023-01-21 13:13:29.162537: step: 480/527, loss: 0.005084228236228228
2023-01-21 13:13:30.283366: step: 484/527, loss: 0.030003929510712624
2023-01-21 13:13:31.427237: step: 488/527, loss: 3.9768219721736386e-05
2023-01-21 13:13:32.545741: step: 492/527, loss: 7.343292054429185e-06
2023-01-21 13:13:33.664343: step: 496/527, loss: 0.010692168027162552
2023-01-21 13:13:34.757536: step: 500/527, loss: 0.0036265135277062654
2023-01-21 13:13:35.862123: step: 504/527, loss: 0.001107549760490656
2023-01-21 13:13:36.969832: step: 508/527, loss: 7.53879503463395e-05
2023-01-21 13:13:38.048327: step: 512/527, loss: 0.008500195108354092
2023-01-21 13:13:39.131693: step: 516/527, loss: 9.975433204090223e-05
2023-01-21 13:13:40.220917: step: 520/527, loss: 0.0020113945938646793
2023-01-21 13:13:41.344588: step: 524/527, loss: 0.0007612229092046618
2023-01-21 13:13:42.458257: step: 528/527, loss: 0.007023429963737726
2023-01-21 13:13:43.610561: step: 532/527, loss: 0.03042755089700222
2023-01-21 13:13:44.732053: step: 536/527, loss: 0.00017652510723564774
2023-01-21 13:13:45.827098: step: 540/527, loss: 1.088216781616211
2023-01-21 13:13:46.943920: step: 544/527, loss: 0.0007103919633664191
2023-01-21 13:13:48.086869: step: 548/527, loss: 0.036032866686582565
2023-01-21 13:13:49.214274: step: 552/527, loss: 3.833770824712701e-05
2023-01-21 13:13:50.370340: step: 556/527, loss: 1.4400482541532256e-05
2023-01-21 13:13:51.501595: step: 560/527, loss: 0.013247203081846237
2023-01-21 13:13:52.614393: step: 564/527, loss: 2.956390289909905e-06
2023-01-21 13:13:53.713257: step: 568/527, loss: 0.00011405945406295359
2023-01-21 13:13:54.804633: step: 572/527, loss: 0.0040070535615086555
2023-01-21 13:13:55.923404: step: 576/527, loss: 0.018068790435791016
2023-01-21 13:13:57.039491: step: 580/527, loss: 0.0018008708721026778
2023-01-21 13:13:58.142115: step: 584/527, loss: 2.47955322265625e-05
2023-01-21 13:13:59.233265: step: 588/527, loss: 0.0020657540298998356
2023-01-21 13:14:00.365776: step: 592/527, loss: 0.14208821952342987
2023-01-21 13:14:01.456080: step: 596/527, loss: 1.220703143189894e-05
2023-01-21 13:14:02.596948: step: 600/527, loss: 0.005465984344482422
2023-01-21 13:14:03.737093: step: 604/527, loss: 0.0006066322093829513
2023-01-21 13:14:04.849799: step: 608/527, loss: 0.0057464600540697575
2023-01-21 13:14:05.991595: step: 612/527, loss: 0.0015774727798998356
2023-01-21 13:14:07.115165: step: 616/527, loss: 0.002112960908561945
2023-01-21 13:14:08.200496: step: 620/527, loss: 1.678466833254788e-05
2023-01-21 13:14:09.310249: step: 624/527, loss: 0.006539439782500267
2023-01-21 13:14:10.403938: step: 628/527, loss: 0.0017723083728924394
2023-01-21 13:14:11.516928: step: 632/527, loss: 0.0009443283197470009
2023-01-21 13:14:12.632557: step: 636/527, loss: 2.1839141481905244e-05
2023-01-21 13:14:13.744912: step: 640/527, loss: 0.004162025637924671
2023-01-21 13:14:14.823451: step: 644/527, loss: 0.0003446579212322831
2023-01-21 13:14:15.945606: step: 648/527, loss: 0.0057426453568041325
2023-01-21 13:14:17.064768: step: 652/527, loss: 0.011994934640824795
2023-01-21 13:14:18.179362: step: 656/527, loss: 0.00012216568575240672
2023-01-21 13:14:19.327705: step: 660/527, loss: 1.9073486328125e-06
2023-01-21 13:14:20.432721: step: 664/527, loss: -1.9073486328125e-06
2023-01-21 13:14:21.623492: step: 668/527, loss: 2.536773718020413e-05
2023-01-21 13:14:22.741037: step: 672/527, loss: 0.01849053055047989
2023-01-21 13:14:23.898286: step: 676/527, loss: 1.9073486328125e-05
2023-01-21 13:14:25.018567: step: 680/527, loss: 1.8024444216280244e-05
2023-01-21 13:14:26.135507: step: 684/527, loss: 0.0004604339774232358
2023-01-21 13:14:27.257683: step: 688/527, loss: 0.034531306475400925
2023-01-21 13:14:28.371138: step: 692/527, loss: 0.00037136077298782766
2023-01-21 13:14:29.521494: step: 696/527, loss: 0.0023142099380493164
2023-01-21 13:14:30.650594: step: 700/527, loss: 0.0022110939025878906
2023-01-21 13:14:31.754094: step: 704/527, loss: 7.62939453125e-05
2023-01-21 13:14:32.907918: step: 708/527, loss: 4.806518700206652e-05
2023-01-21 13:14:34.019238: step: 712/527, loss: 7.925033423816785e-05
2023-01-21 13:14:35.156482: step: 716/527, loss: 0.00011453629122115672
2023-01-21 13:14:36.296958: step: 720/527, loss: 0.0009848595364019275
2023-01-21 13:14:37.427544: step: 724/527, loss: 0.004429626744240522
2023-01-21 13:14:38.549345: step: 728/527, loss: 0.0004675865639001131
2023-01-21 13:14:39.648060: step: 732/527, loss: 0.02022247388958931
2023-01-21 13:14:40.742411: step: 736/527, loss: 0.0001201629638671875
2023-01-21 13:14:41.856748: step: 740/527, loss: 0.00289573660120368
2023-01-21 13:14:42.975414: step: 744/527, loss: 0.0012001037830486894
2023-01-21 13:14:44.098551: step: 748/527, loss: 0.002192688174545765
2023-01-21 13:14:45.200722: step: 752/527, loss: 6.322861008811742e-05
2023-01-21 13:14:46.306689: step: 756/527, loss: 0.012088966555893421
2023-01-21 13:14:47.412193: step: 760/527, loss: 0.326596200466156
2023-01-21 13:14:48.555381: step: 764/527, loss: 1.087188684323337e-05
2023-01-21 13:14:49.649944: step: 768/527, loss: 4.482269105210435e-06
2023-01-21 13:14:50.749681: step: 772/527, loss: 0.010147858411073685
2023-01-21 13:14:51.846805: step: 776/527, loss: 0.0014352798461914062
2023-01-21 13:14:52.958806: step: 780/527, loss: 1.201629675051663e-05
2023-01-21 13:14:54.058129: step: 784/527, loss: 5.722046125811175e-07
2023-01-21 13:14:55.167576: step: 788/527, loss: 0.0015343456761911511
2023-01-21 13:14:56.278833: step: 792/527, loss: 0.02372293546795845
2023-01-21 13:14:57.415920: step: 796/527, loss: 0.04090938717126846
2023-01-21 13:14:58.540664: step: 800/527, loss: 0.09455318748950958
2023-01-21 13:14:59.665160: step: 804/527, loss: 0.024419786408543587
2023-01-21 13:15:00.753030: step: 808/527, loss: 0.0001714706450002268
2023-01-21 13:15:01.852784: step: 812/527, loss: 0.0367090217769146
2023-01-21 13:15:02.970760: step: 816/527, loss: 0.0023056031204760075
2023-01-21 13:15:04.092853: step: 820/527, loss: 2.670288040462765e-06
2023-01-21 13:15:05.229575: step: 824/527, loss: 0.0015481949085369706
2023-01-21 13:15:06.305406: step: 828/527, loss: 0.002475929446518421
2023-01-21 13:15:07.440953: step: 832/527, loss: 0.0013447761302813888
2023-01-21 13:15:08.533153: step: 836/527, loss: 0.0020334243308752775
2023-01-21 13:15:09.618862: step: 840/527, loss: 0.002803945681080222
2023-01-21 13:15:10.716452: step: 844/527, loss: 0.04653206095099449
2023-01-21 13:15:11.813400: step: 848/527, loss: 1.8501283193472773e-05
2023-01-21 13:15:12.949983: step: 852/527, loss: 0.0213457103818655
2023-01-21 13:15:14.076401: step: 856/527, loss: 0.04106597974896431
2023-01-21 13:15:15.219473: step: 860/527, loss: 0.0006138801109045744
2023-01-21 13:15:16.313882: step: 864/527, loss: 0.005485725589096546
2023-01-21 13:15:17.413965: step: 868/527, loss: 0.0023593902587890625
2023-01-21 13:15:18.557427: step: 872/527, loss: 0.020300962030887604
2023-01-21 13:15:19.674129: step: 876/527, loss: 0.011713982559740543
2023-01-21 13:15:20.774951: step: 880/527, loss: 0.0007172584882937372
2023-01-21 13:15:21.866093: step: 884/527, loss: 2.918243444582913e-05
2023-01-21 13:15:22.992964: step: 888/527, loss: 0.05857863277196884
2023-01-21 13:15:24.102235: step: 892/527, loss: 0.0005195617559365928
2023-01-21 13:15:25.234164: step: 896/527, loss: 0.00010604858107399195
2023-01-21 13:15:26.356325: step: 900/527, loss: 8.087158494163305e-05
2023-01-21 13:15:27.473044: step: 904/527, loss: 0.00024223329091910273
2023-01-21 13:15:28.541937: step: 908/527, loss: 0.0002206802455475554
2023-01-21 13:15:29.669101: step: 912/527, loss: 0.009944559074938297
2023-01-21 13:15:30.793961: step: 916/527, loss: 0.01507263258099556
2023-01-21 13:15:31.940294: step: 920/527, loss: 0.05059022828936577
2023-01-21 13:15:33.078037: step: 924/527, loss: 0.00016579628572799265
2023-01-21 13:15:34.167200: step: 928/527, loss: 0.010430526919662952
2023-01-21 13:15:35.312103: step: 932/527, loss: 0.05597362667322159
2023-01-21 13:15:36.400888: step: 936/527, loss: 0.011325836181640625
2023-01-21 13:15:37.492241: step: 940/527, loss: 0.002430153079330921
2023-01-21 13:15:38.584549: step: 944/527, loss: 6.86645489622606e-06
2023-01-21 13:15:39.702393: step: 948/527, loss: 0.00010547637793933973
2023-01-21 13:15:40.825917: step: 952/527, loss: 0.06659820675849915
2023-01-21 13:15:41.934891: step: 956/527, loss: 0.003160381456837058
2023-01-21 13:15:43.051020: step: 960/527, loss: 0.00043816567631438375
2023-01-21 13:15:44.154636: step: 964/527, loss: 0.0023853301536291838
2023-01-21 13:15:45.285905: step: 968/527, loss: 0.01980781741440296
2023-01-21 13:15:46.364579: step: 972/527, loss: 0.022972488775849342
2023-01-21 13:15:47.446486: step: 976/527, loss: 0.0002794266038108617
2023-01-21 13:15:48.572348: step: 980/527, loss: 0.00012836456880904734
2023-01-21 13:15:49.679729: step: 984/527, loss: 0.0029923440888524055
2023-01-21 13:15:50.782824: step: 988/527, loss: 0.00010814666893566027
2023-01-21 13:15:51.905372: step: 992/527, loss: 0.0006361007690429688
2023-01-21 13:15:53.001973: step: 996/527, loss: 0.0024063109885901213
2023-01-21 13:15:54.100572: step: 1000/527, loss: 0.0034011839888989925
2023-01-21 13:15:55.237421: step: 1004/527, loss: 0.035730551928281784
2023-01-21 13:15:56.448402: step: 1008/527, loss: 0.0005638122675009072
2023-01-21 13:15:57.571806: step: 1012/527, loss: 1.010894811770413e-05
2023-01-21 13:15:58.691079: step: 1016/527, loss: 0.028648758307099342
2023-01-21 13:15:59.794048: step: 1020/527, loss: 1.7452241081628017e-05
2023-01-21 13:16:00.911953: step: 1024/527, loss: 0.0160935390740633
2023-01-21 13:16:02.029666: step: 1028/527, loss: 0.003731346223503351
2023-01-21 13:16:03.146425: step: 1032/527, loss: 0.38848456740379333
2023-01-21 13:16:04.256794: step: 1036/527, loss: 0.013659859076142311
2023-01-21 13:16:05.356397: step: 1040/527, loss: 0.0004177093505859375
2023-01-21 13:16:06.454475: step: 1044/527, loss: 0.0009197235340252519
2023-01-21 13:16:07.587412: step: 1048/527, loss: 0.02584228478372097
2023-01-21 13:16:08.698296: step: 1052/527, loss: 0.002713489346206188
2023-01-21 13:16:09.788679: step: 1056/527, loss: -2.117157055181451e-05
2023-01-21 13:16:10.889561: step: 1060/527, loss: 1.1992455256404355e-05
2023-01-21 13:16:11.963267: step: 1064/527, loss: 0.0023584368173033
2023-01-21 13:16:13.087616: step: 1068/527, loss: 0.0005010604509152472
2023-01-21 13:16:14.199712: step: 1072/527, loss: 0.009381484240293503
2023-01-21 13:16:15.305280: step: 1076/527, loss: 0.0001177787926280871
2023-01-21 13:16:16.428264: step: 1080/527, loss: 0.020886898040771484
2023-01-21 13:16:17.536548: step: 1084/527, loss: 0.007814550772309303
2023-01-21 13:16:18.634401: step: 1088/527, loss: 0.021423721686005592
2023-01-21 13:16:19.750979: step: 1092/527, loss: 0.0026113511994481087
2023-01-21 13:16:20.858085: step: 1096/527, loss: 0.0763712003827095
2023-01-21 13:16:21.989394: step: 1100/527, loss: 0.0005100250127725303
2023-01-21 13:16:23.108766: step: 1104/527, loss: 0.00018730164447333664
2023-01-21 13:16:24.235496: step: 1108/527, loss: 0.00315685267560184
2023-01-21 13:16:25.367819: step: 1112/527, loss: 0.00045299530029296875
2023-01-21 13:16:26.472127: step: 1116/527, loss: 0.2409219741821289
2023-01-21 13:16:27.593584: step: 1120/527, loss: 0.012544346041977406
2023-01-21 13:16:28.721577: step: 1124/527, loss: 0.0019412994151934981
2023-01-21 13:16:29.817428: step: 1128/527, loss: 5.531311398954131e-05
2023-01-21 13:16:30.934348: step: 1132/527, loss: 0.0935780480504036
2023-01-21 13:16:32.082815: step: 1136/527, loss: 0.006203460972756147
2023-01-21 13:16:33.162018: step: 1140/527, loss: 0.00879664532840252
2023-01-21 13:16:34.290144: step: 1144/527, loss: 0.07566747069358826
2023-01-21 13:16:35.444097: step: 1148/527, loss: 0.00037384036113508046
2023-01-21 13:16:36.576033: step: 1152/527, loss: 7.629394644936838e-07
2023-01-21 13:16:37.699068: step: 1156/527, loss: 0.0005529404152184725
2023-01-21 13:16:38.800136: step: 1160/527, loss: 0.005059623625129461
2023-01-21 13:16:39.926035: step: 1164/527, loss: 9.632110959501006e-06
2023-01-21 13:16:41.011945: step: 1168/527, loss: 0.00043125153752043843
2023-01-21 13:16:42.128493: step: 1172/527, loss: 0.003337860107421875
2023-01-21 13:16:43.261298: step: 1176/527, loss: 0.019627952948212624
2023-01-21 13:16:44.356830: step: 1180/527, loss: 0.01567087136209011
2023-01-21 13:16:45.455066: step: 1184/527, loss: 0.0031160356011241674
2023-01-21 13:16:46.577380: step: 1188/527, loss: 0.00018334388732910156
2023-01-21 13:16:47.693925: step: 1192/527, loss: 0.09085607528686523
2023-01-21 13:16:48.803019: step: 1196/527, loss: 0.0003784179862122983
2023-01-21 13:16:49.908607: step: 1200/527, loss: 0.00029811859712935984
2023-01-21 13:16:51.024899: step: 1204/527, loss: 0.0007266998873092234
2023-01-21 13:16:52.166119: step: 1208/527, loss: 0.000263023393927142
2023-01-21 13:16:53.277929: step: 1212/527, loss: 1.2111663636460435e-05
2023-01-21 13:16:54.420133: step: 1216/527, loss: 3.662109520519152e-05
2023-01-21 13:16:55.555464: step: 1220/527, loss: 0.001142740249633789
2023-01-21 13:16:56.664992: step: 1224/527, loss: 0.18275032937526703
2023-01-21 13:16:57.765516: step: 1228/527, loss: 0.002254486083984375
2023-01-21 13:16:58.876912: step: 1232/527, loss: 9.107588994083926e-06
2023-01-21 13:16:59.984802: step: 1236/527, loss: 0.0022802352905273438
2023-01-21 13:17:01.074566: step: 1240/527, loss: 0.0007385254139080644
2023-01-21 13:17:02.156559: step: 1244/527, loss: 0.0002507210010662675
2023-01-21 13:17:03.281651: step: 1248/527, loss: 0.018927669152617455
2023-01-21 13:17:04.426339: step: 1252/527, loss: 0.0023344040382653475
2023-01-21 13:17:05.561211: step: 1256/527, loss: 0.0031208039727061987
2023-01-21 13:17:06.674374: step: 1260/527, loss: 0.052913546562194824
2023-01-21 13:17:07.787691: step: 1264/527, loss: 5.006790615880163e-06
2023-01-21 13:17:08.912053: step: 1268/527, loss: 5.054473876953125e-05
2023-01-21 13:17:10.033638: step: 1272/527, loss: 0.00620956439524889
2023-01-21 13:17:11.169902: step: 1276/527, loss: 0.02844705618917942
2023-01-21 13:17:12.320076: step: 1280/527, loss: 0.00705032330006361
2023-01-21 13:17:13.398557: step: 1284/527, loss: 0.00817260704934597
2023-01-21 13:17:14.511970: step: 1288/527, loss: 0.0027587893418967724
2023-01-21 13:17:15.626110: step: 1292/527, loss: 0.0001911163271870464
2023-01-21 13:17:16.736455: step: 1296/527, loss: 0.003289127256721258
2023-01-21 13:17:17.839650: step: 1300/527, loss: 0.07440871745347977
2023-01-21 13:17:18.947869: step: 1304/527, loss: 0.0045642852783203125
2023-01-21 13:17:20.049566: step: 1308/527, loss: 7.743835885776207e-05
2023-01-21 13:17:21.179804: step: 1312/527, loss: 0.00011272430856479332
2023-01-21 13:17:22.315618: step: 1316/527, loss: 8.382798114325851e-05
2023-01-21 13:17:23.439469: step: 1320/527, loss: 0.0015691756270825863
2023-01-21 13:17:24.537867: step: 1324/527, loss: 0.0030309678986668587
2023-01-21 13:17:25.655930: step: 1328/527, loss: 0.00040874481783248484
2023-01-21 13:17:26.766566: step: 1332/527, loss: 0.00029354094294831157
2023-01-21 13:17:27.885586: step: 1336/527, loss: 0.005395126529037952
2023-01-21 13:17:28.995529: step: 1340/527, loss: 0.04798431321978569
2023-01-21 13:17:30.094820: step: 1344/527, loss: 3.52859501617786e-06
2023-01-21 13:17:31.214913: step: 1348/527, loss: 0.03640594705939293
2023-01-21 13:17:32.314237: step: 1352/527, loss: 1.1253356205997989e-05
2023-01-21 13:17:33.447703: step: 1356/527, loss: 0.009305287152528763
2023-01-21 13:17:34.539371: step: 1360/527, loss: 0.00022411346435546875
2023-01-21 13:17:35.672378: step: 1364/527, loss: 0.011685753241181374
2023-01-21 13:17:36.769749: step: 1368/527, loss: 0.0002561330911703408
2023-01-21 13:17:37.860414: step: 1372/527, loss: 0.00043125153752043843
2023-01-21 13:17:38.980652: step: 1376/527, loss: 0.0005689620738849044
2023-01-21 13:17:40.087624: step: 1380/527, loss: 7.4386593951203395e-06
2023-01-21 13:17:41.213368: step: 1384/527, loss: 0.004019069951027632
2023-01-21 13:17:42.337709: step: 1388/527, loss: 0.006441688630729914
2023-01-21 13:17:43.442700: step: 1392/527, loss: 1.71661376953125e-05
2023-01-21 13:17:44.570617: step: 1396/527, loss: 0.00011672973778331652
2023-01-21 13:17:45.660119: step: 1400/527, loss: 0.006808471865952015
2023-01-21 13:17:46.755345: step: 1404/527, loss: 0.017092324793338776
2023-01-21 13:17:47.858197: step: 1408/527, loss: 0.0018543243641033769
2023-01-21 13:17:48.951481: step: 1412/527, loss: 7.543564424850047e-05
2023-01-21 13:17:50.060032: step: 1416/527, loss: 0.018691254779696465
2023-01-21 13:17:51.199228: step: 1420/527, loss: 0.47512856125831604
2023-01-21 13:17:52.307607: step: 1424/527, loss: 3.366470627952367e-05
2023-01-21 13:17:53.408901: step: 1428/527, loss: 1.411438006471144e-05
2023-01-21 13:17:54.532574: step: 1432/527, loss: 7.24792471373803e-06
2023-01-21 13:17:55.637550: step: 1436/527, loss: 0.012579345144331455
2023-01-21 13:17:56.757803: step: 1440/527, loss: 0.00023937225341796875
2023-01-21 13:17:57.868805: step: 1444/527, loss: 0.03071027249097824
2023-01-21 13:17:58.971893: step: 1448/527, loss: 0.009813499636948109
2023-01-21 13:18:00.091178: step: 1452/527, loss: 0.03610806539654732
2023-01-21 13:18:01.222332: step: 1456/527, loss: 2.307891918462701e-05
2023-01-21 13:18:02.354909: step: 1460/527, loss: 0.01668224297463894
2023-01-21 13:18:03.471456: step: 1464/527, loss: 0.0004642486746888608
2023-01-21 13:18:04.564692: step: 1468/527, loss: 0.027607537806034088
2023-01-21 13:18:05.697264: step: 1472/527, loss: 0.012082099914550781
2023-01-21 13:18:06.800465: step: 1476/527, loss: 0.00813837070018053
2023-01-21 13:18:07.891700: step: 1480/527, loss: 0.0001066207914846018
2023-01-21 13:18:08.978174: step: 1484/527, loss: 2.870559910661541e-05
2023-01-21 13:18:10.096105: step: 1488/527, loss: 0.010634804144501686
2023-01-21 13:18:11.203000: step: 1492/527, loss: 0.018246840685606003
2023-01-21 13:18:12.303964: step: 1496/527, loss: 0.010386180132627487
2023-01-21 13:18:13.403049: step: 1500/527, loss: 0.028856659308075905
2023-01-21 13:18:14.506461: step: 1504/527, loss: 1.678466833254788e-05
2023-01-21 13:18:15.628162: step: 1508/527, loss: 0.020030593499541283
2023-01-21 13:18:16.728856: step: 1512/527, loss: 0.00045185087947174907
2023-01-21 13:18:17.861065: step: 1516/527, loss: 0.0005821228260174394
2023-01-21 13:18:18.966428: step: 1520/527, loss: 3.547668529790826e-05
2023-01-21 13:18:20.088140: step: 1524/527, loss: 0.0023143768776208162
2023-01-21 13:18:21.221237: step: 1528/527, loss: 0.021855641156435013
2023-01-21 13:18:22.324744: step: 1532/527, loss: 1.583099401614163e-05
2023-01-21 13:18:23.425313: step: 1536/527, loss: -1.9550323031580774e-06
2023-01-21 13:18:24.530083: step: 1540/527, loss: 0.0011399269569665194
2023-01-21 13:18:25.640636: step: 1544/527, loss: 0.0050754547119140625
2023-01-21 13:18:26.743387: step: 1548/527, loss: 0.0017654896946623921
2023-01-21 13:18:27.877821: step: 1552/527, loss: -2.3269651137525216e-05
2023-01-21 13:18:28.978111: step: 1556/527, loss: 7.314682443393394e-05
2023-01-21 13:18:30.066310: step: 1560/527, loss: 3.6239621294953395e-06
2023-01-21 13:18:31.179410: step: 1564/527, loss: 0.00015573501877952367
2023-01-21 13:18:32.309109: step: 1568/527, loss: 0.00044527053250931203
2023-01-21 13:18:33.411763: step: 1572/527, loss: 2.6709365844726562
2023-01-21 13:18:34.519829: step: 1576/527, loss: 0.002544450806453824
2023-01-21 13:18:35.661438: step: 1580/527, loss: 0.024519825354218483
2023-01-21 13:18:36.789719: step: 1584/527, loss: 0.0005123138544149697
2023-01-21 13:18:37.901421: step: 1588/527, loss: 0.0003116607549600303
2023-01-21 13:18:39.008949: step: 1592/527, loss: 0.018262101337313652
2023-01-21 13:18:40.134437: step: 1596/527, loss: 0.0007053375011309981
2023-01-21 13:18:41.275267: step: 1600/527, loss: 0.0007838249439373612
2023-01-21 13:18:42.396755: step: 1604/527, loss: 0.0001655578671488911
2023-01-21 13:18:43.496094: step: 1608/527, loss: 0.0026958466041833162
2023-01-21 13:18:44.592853: step: 1612/527, loss: 0.0035102847032248974
2023-01-21 13:18:45.724633: step: 1616/527, loss: 0.004958248697221279
2023-01-21 13:18:46.876706: step: 1620/527, loss: 0.004815483465790749
2023-01-21 13:18:47.982132: step: 1624/527, loss: 6.294250852079131e-06
2023-01-21 13:18:49.118051: step: 1628/527, loss: 4.816055661649443e-05
2023-01-21 13:18:50.227998: step: 1632/527, loss: 0.0018827916355803609
2023-01-21 13:18:51.351033: step: 1636/527, loss: 3.843307786155492e-05
2023-01-21 13:18:52.479857: step: 1640/527, loss: 0.0004249572812113911
2023-01-21 13:18:53.591538: step: 1644/527, loss: 0.0005807877168990672
2023-01-21 13:18:54.705056: step: 1648/527, loss: 0.10231515020132065
2023-01-21 13:18:55.792123: step: 1652/527, loss: 0.001157569931820035
2023-01-21 13:18:56.929172: step: 1656/527, loss: 0.002237987471744418
2023-01-21 13:18:58.040734: step: 1660/527, loss: 9.574890282237902e-05
2023-01-21 13:18:59.132831: step: 1664/527, loss: 5.197525297262473e-06
2023-01-21 13:19:00.225927: step: 1668/527, loss: 0.00015010833158157766
2023-01-21 13:19:01.369602: step: 1672/527, loss: 0.005751943681389093
2023-01-21 13:19:02.469261: step: 1676/527, loss: 4.329681542003527e-05
2023-01-21 13:19:03.623903: step: 1680/527, loss: 0.02666759490966797
2023-01-21 13:19:04.745111: step: 1684/527, loss: 3.0490757126244716e-05
2023-01-21 13:19:05.892374: step: 1688/527, loss: 0.058798789978027344
2023-01-21 13:19:06.990728: step: 1692/527, loss: 9.994507126975805e-05
2023-01-21 13:19:08.100563: step: 1696/527, loss: 8.20159948489163e-06
2023-01-21 13:19:09.222269: step: 1700/527, loss: 4.76837158203125e-05
2023-01-21 13:19:10.343997: step: 1704/527, loss: 0.34508514404296875
2023-01-21 13:19:11.450281: step: 1708/527, loss: 0.11628475040197372
2023-01-21 13:19:12.567329: step: 1712/527, loss: 0.0008120537386275828
2023-01-21 13:19:13.665426: step: 1716/527, loss: 0.0002593994140625
2023-01-21 13:19:14.781042: step: 1720/527, loss: 0.0031949998810887337
2023-01-21 13:19:15.883681: step: 1724/527, loss: 0.0020835876930505037
2023-01-21 13:19:17.001196: step: 1728/527, loss: 0.0005289077525958419
2023-01-21 13:19:18.104593: step: 1732/527, loss: 0.004260825924575329
2023-01-21 13:19:19.234340: step: 1736/527, loss: 0.016506386920809746
2023-01-21 13:19:20.385360: step: 1740/527, loss: 0.010212182998657227
2023-01-21 13:19:21.474938: step: 1744/527, loss: 0.00013117790513206273
2023-01-21 13:19:22.584921: step: 1748/527, loss: 0.009903526864945889
2023-01-21 13:19:23.695936: step: 1752/527, loss: 0.006413173396140337
2023-01-21 13:19:24.788148: step: 1756/527, loss: 0.00016367435455322266
2023-01-21 13:19:25.911258: step: 1760/527, loss: 0.0014652252430096269
2023-01-21 13:19:27.042631: step: 1764/527, loss: 0.0017039298545569181
2023-01-21 13:19:28.194979: step: 1768/527, loss: 0.00058832170907408
2023-01-21 13:19:29.302644: step: 1772/527, loss: 0.0031113626901060343
2023-01-21 13:19:30.387777: step: 1776/527, loss: 0.014738750644028187
2023-01-21 13:19:31.528992: step: 1780/527, loss: 0.006545448210090399
2023-01-21 13:19:32.617030: step: 1784/527, loss: 0.009801817126572132
2023-01-21 13:19:33.728992: step: 1788/527, loss: 0.001978302141651511
2023-01-21 13:19:34.854672: step: 1792/527, loss: 0.011224651709198952
2023-01-21 13:19:35.976907: step: 1796/527, loss: 0.01539001613855362
2023-01-21 13:19:37.068680: step: 1800/527, loss: 0.0004193306085653603
2023-01-21 13:19:38.180081: step: 1804/527, loss: 0.0024080276489257812
2023-01-21 13:19:39.296987: step: 1808/527, loss: 0.23451752960681915
2023-01-21 13:19:40.391830: step: 1812/527, loss: 0.002439308213070035
2023-01-21 13:19:41.502825: step: 1816/527, loss: 0.13236045837402344
2023-01-21 13:19:42.611984: step: 1820/527, loss: 0.051760233938694
2023-01-21 13:19:43.712373: step: 1824/527, loss: -3.8146970382513246e-07
2023-01-21 13:19:44.820984: step: 1828/527, loss: 0.0007528305286541581
2023-01-21 13:19:45.909769: step: 1832/527, loss: 0.026163483038544655
2023-01-21 13:19:47.047321: step: 1836/527, loss: 5.6838991440599784e-05
2023-01-21 13:19:48.179042: step: 1840/527, loss: 0.0010646820301190019
2023-01-21 13:19:49.290888: step: 1844/527, loss: 0.0003292083565611392
2023-01-21 13:19:50.412249: step: 1848/527, loss: 0.010122299194335938
2023-01-21 13:19:51.539078: step: 1852/527, loss: 0.15897493064403534
2023-01-21 13:19:52.668709: step: 1856/527, loss: 0.08352112770080566
2023-01-21 13:19:53.789550: step: 1860/527, loss: 0.04018288105726242
2023-01-21 13:19:54.870718: step: 1864/527, loss: 2.2220612663659267e-05
2023-01-21 13:19:55.971598: step: 1868/527, loss: 0.002647113986313343
2023-01-21 13:19:57.063049: step: 1872/527, loss: 0.0006129264947958291
2023-01-21 13:19:58.180993: step: 1876/527, loss: 0.0002506256278138608
2023-01-21 13:19:59.329193: step: 1880/527, loss: 0.00805673561990261
2023-01-21 13:20:00.456014: step: 1884/527, loss: -2.6702882678364404e-06
2023-01-21 13:20:01.558231: step: 1888/527, loss: 2.365112231927924e-05
2023-01-21 13:20:02.674942: step: 1892/527, loss: 0.03178234398365021
2023-01-21 13:20:03.767715: step: 1896/527, loss: 0.0005556106334552169
2023-01-21 13:20:04.864218: step: 1900/527, loss: 0.00022506712411995977
2023-01-21 13:20:06.017524: step: 1904/527, loss: 0.00249404925853014
2023-01-21 13:20:07.114379: step: 1908/527, loss: 0.0011820793151855469
2023-01-21 13:20:08.257828: step: 1912/527, loss: 0.00015459061251021922
2023-01-21 13:20:09.356166: step: 1916/527, loss: 0.00032720566377975047
2023-01-21 13:20:10.487489: step: 1920/527, loss: 0.00040073395939543843
2023-01-21 13:20:11.605908: step: 1924/527, loss: 0.00025081634521484375
2023-01-21 13:20:12.714887: step: 1928/527, loss: 0.0005310058477334678
2023-01-21 13:20:13.802232: step: 1932/527, loss: 0.00023217200941871852
2023-01-21 13:20:14.929112: step: 1936/527, loss: 0.009486199356615543
2023-01-21 13:20:16.053243: step: 1940/527, loss: 7.057190487103071e-06
2023-01-21 13:20:17.162991: step: 1944/527, loss: 0.00887689646333456
2023-01-21 13:20:18.252529: step: 1948/527, loss: 0.0026689530350267887
2023-01-21 13:20:19.365131: step: 1952/527, loss: 0.0011229516239836812
2023-01-21 13:20:20.504675: step: 1956/527, loss: 0.01138229388743639
2023-01-21 13:20:21.603333: step: 1960/527, loss: 0.42775943875312805
2023-01-21 13:20:22.715828: step: 1964/527, loss: 6.341934204101562e-05
2023-01-21 13:20:23.827611: step: 1968/527, loss: 0.0005316734313964844
2023-01-21 13:20:24.946278: step: 1972/527, loss: 0.01154871005564928
2023-01-21 13:20:26.064851: step: 1976/527, loss: 0.0087296012789011
2023-01-21 13:20:27.170049: step: 1980/527, loss: 0.00046939851017668843
2023-01-21 13:20:28.311228: step: 1984/527, loss: 0.0003887176571879536
2023-01-21 13:20:29.472590: step: 1988/527, loss: 0.0837644562125206
2023-01-21 13:20:30.657527: step: 1992/527, loss: 0.01765899732708931
2023-01-21 13:20:31.771395: step: 1996/527, loss: 0.1999751180410385
2023-01-21 13:20:32.872587: step: 2000/527, loss: 0.006393528077751398
2023-01-21 13:20:34.003305: step: 2004/527, loss: 5.798339770990424e-05
2023-01-21 13:20:35.108709: step: 2008/527, loss: 0.0020574568770825863
2023-01-21 13:20:36.188542: step: 2012/527, loss: 0.00037631989107467234
2023-01-21 13:20:37.318878: step: 2016/527, loss: 0.01455764751881361
2023-01-21 13:20:38.409233: step: 2020/527, loss: 6.961822691664565e-06
2023-01-21 13:20:39.507848: step: 2024/527, loss: 0.019783783704042435
2023-01-21 13:20:40.612925: step: 2028/527, loss: 0.007009411230683327
2023-01-21 13:20:41.707779: step: 2032/527, loss: 2.713203321036417e-05
2023-01-21 13:20:42.800566: step: 2036/527, loss: 0.004231167491525412
2023-01-21 13:20:43.926554: step: 2040/527, loss: 0.003732299664989114
2023-01-21 13:20:45.042764: step: 2044/527, loss: 0.010844802483916283
2023-01-21 13:20:46.128496: step: 2048/527, loss: -7.05719003235572e-06
2023-01-21 13:20:47.230460: step: 2052/527, loss: 0.045352935791015625
2023-01-21 13:20:48.332398: step: 2056/527, loss: 6.580352783203125e-05
2023-01-21 13:20:49.441415: step: 2060/527, loss: 0.10742244869470596
2023-01-21 13:20:50.583358: step: 2064/527, loss: 0.0011796951293945312
2023-01-21 13:20:51.680297: step: 2068/527, loss: 0.001395416329614818
2023-01-21 13:20:52.775976: step: 2072/527, loss: 0.001425325870513916
2023-01-21 13:20:53.900834: step: 2076/527, loss: 0.019658468663692474
2023-01-21 13:20:55.003390: step: 2080/527, loss: 0.0008678436279296875
2023-01-21 13:20:56.094130: step: 2084/527, loss: 0.0005279540782794356
2023-01-21 13:20:57.203295: step: 2088/527, loss: 0.007942581549286842
2023-01-21 13:20:58.300004: step: 2092/527, loss: 0.03498439863324165
2023-01-21 13:20:59.429324: step: 2096/527, loss: 0.0013710500206798315
2023-01-21 13:21:00.531206: step: 2100/527, loss: 0.004145812708884478
2023-01-21 13:21:01.660356: step: 2104/527, loss: 0.024820709601044655
2023-01-21 13:21:02.779658: step: 2108/527, loss: 2.95162208203692e-05
==================================================
Loss: 0.023
--------------------
Dev: {'event': {'p': 0.5995785036880927, 'r': 0.7576564580559254, 'f1': 0.6694117647058823}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27}
Test: {'event': {'p': 0.6422574626865671, 'r': 0.7868571428571428, 'f1': 0.707241910631741}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27}
Chinese: {'event': {'p': 0.5714285714285714, 'r': 0.8148148148148148, 'f1': 0.6717557251908397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27}
Korean: {'event': {'p': 0.6363636363636364, 'r': 0.5555555555555556, 'f1': 0.5932203389830508}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27}
Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 13:21:44.309017: step: 4/527, loss: 0.04735551029443741
2023-01-21 13:21:45.427619: step: 8/527, loss: 0.010111046023666859
2023-01-21 13:21:46.511141: step: 12/527, loss: 0.00013046263484284282
2023-01-21 13:21:47.614906: step: 16/527, loss: 0.0005805015680380166
2023-01-21 13:21:48.711764: step: 20/527, loss: 0.00017433168250136077
2023-01-21 13:21:49.823885: step: 24/527, loss: 0.006561470218002796
2023-01-21 13:21:50.926029: step: 28/527, loss: 0.001269531319849193
2023-01-21 13:21:52.051678: step: 32/527, loss: 0.0016639232635498047
2023-01-21 13:21:53.193290: step: 36/527, loss: 0.0001375198335153982
2023-01-21 13:21:54.304318: step: 40/527, loss: 0.0001443863002350554
2023-01-21 13:21:55.417492: step: 44/527, loss: 0.012665940448641777
2023-01-21 13:21:56.526712: step: 48/527, loss: 0.012413167394697666
2023-01-21 13:21:57.598940: step: 52/527, loss: 4.529953366727568e-05
2023-01-21 13:21:58.724808: step: 56/527, loss: 0.000675201416015625
2023-01-21 13:21:59.848752: step: 60/527, loss: 0.0002800941583700478
2023-01-21 13:22:00.983961: step: 64/527, loss: 0.00045499802217818797
2023-01-21 13:22:02.115857: step: 68/527, loss: 0.002941703889518976
2023-01-21 13:22:03.214065: step: 72/527, loss: 0.03275804594159126
2023-01-21 13:22:04.322241: step: 76/527, loss: 0.00012168884859420359
2023-01-21 13:22:05.444732: step: 80/527, loss: 9.822845458984375e-05
2023-01-21 13:22:06.541343: step: 84/527, loss: 0.0013421059120446444
2023-01-21 13:22:07.651911: step: 88/527, loss: 0.0009275436750613153
2023-01-21 13:22:08.775513: step: 92/527, loss: 0.07247848808765411
2023-01-21 13:22:09.903245: step: 96/527, loss: 0.0656684935092926
2023-01-21 13:22:11.036242: step: 100/527, loss: 0.025267362594604492
2023-01-21 13:22:12.150496: step: 104/527, loss: 7.805824861861765e-05
2023-01-21 13:22:13.285826: step: 108/527, loss: 0.0002262592315673828
2023-01-21 13:22:14.371894: step: 112/527, loss: 7.724761962890625e-05
2023-01-21 13:22:15.454201: step: 116/527, loss: 0.005851077847182751
2023-01-21 13:22:16.575341: step: 120/527, loss: 0.1161341741681099
2023-01-21 13:22:17.674412: step: 124/527, loss: 7.686614844715223e-05
2023-01-21 13:22:18.775508: step: 128/527, loss: 4.76837158203125e-05
2023-01-21 13:22:19.895220: step: 132/527, loss: 0.0004421234189067036
2023-01-21 13:22:21.018415: step: 136/527, loss: 0.013034629635512829
2023-01-21 13:22:22.132465: step: 140/527, loss: 0.0051322937943041325
2023-01-21 13:22:23.246759: step: 144/527, loss: 0.00010607243166305125
2023-01-21 13:22:24.360075: step: 148/527, loss: 0.0015447615878656507
2023-01-21 13:22:25.463202: step: 152/527, loss: -3.910065061063506e-06
2023-01-21 13:22:26.554931: step: 156/527, loss: 0.017389679327607155
2023-01-21 13:22:27.688214: step: 160/527, loss: 0.0005512237548828125
2023-01-21 13:22:28.792789: step: 164/527, loss: 0.023341991007328033
2023-01-21 13:22:29.897867: step: 168/527, loss: 0.0038568496238440275
2023-01-21 13:22:31.015650: step: 172/527, loss: 1.4400482541532256e-05
2023-01-21 13:22:32.134112: step: 176/527, loss: 0.0002906799491029233
2023-01-21 13:22:33.249740: step: 180/527, loss: 0.4488166868686676
2023-01-21 13:22:34.355633: step: 184/527, loss: 2.6702882678364404e-06
2023-01-21 13:22:35.446416: step: 188/527, loss: 9.965896424546372e-06
2023-01-21 13:22:36.604840: step: 192/527, loss: 3.06129441014491e-05
2023-01-21 13:22:37.741795: step: 196/527, loss: 0.0010047913528978825
2023-01-21 13:22:38.839625: step: 200/527, loss: 0.0002020835963776335
2023-01-21 13:22:39.936841: step: 204/527, loss: 0.0021049499046057463
2023-01-21 13:22:41.051867: step: 208/527, loss: 0.00012683868408203125
2023-01-21 13:22:42.153900: step: 212/527, loss: 0.004429721739143133
2023-01-21 13:22:43.284230: step: 216/527, loss: 0.009278679266571999
2023-01-21 13:22:44.395932: step: 220/527, loss: 0.0005752563592977822
2023-01-21 13:22:45.492010: step: 224/527, loss: 0.02311573177576065
2023-01-21 13:22:46.601750: step: 228/527, loss: 8.544922457076609e-05
2023-01-21 13:22:47.702981: step: 232/527, loss: 0.006171036045998335
2023-01-21 13:22:48.842946: step: 236/527, loss: 0.0009214401361532509
2023-01-21 13:22:49.959642: step: 240/527, loss: 1.945495569088962e-05
2023-01-21 13:22:51.047252: step: 244/527, loss: 0.001588541199453175
2023-01-21 13:22:52.134040: step: 248/527, loss: 9.012222290039062e-05
2023-01-21 13:22:53.236186: step: 252/527, loss: 0.00021457672119140625
2023-01-21 13:22:54.350571: step: 256/527, loss: 6.4849853515625e-05
2023-01-21 13:22:55.474280: step: 260/527, loss: 9.040832810569555e-05
2023-01-21 13:22:56.561879: step: 264/527, loss: 6.85691848048009e-05
2023-01-21 13:22:57.665378: step: 268/527, loss: 0.0003631591971497983
2023-01-21 13:22:58.802924: step: 272/527, loss: 0.0002639770391397178
2023-01-21 13:22:59.910691: step: 276/527, loss: 4.472732689464465e-05
2023-01-21 13:23:00.990943: step: 280/527, loss: 0.00015144348435569555
2023-01-21 13:23:02.094101: step: 284/527, loss: 0.06241626664996147
2023-01-21 13:23:03.199455: step: 288/527, loss: 1.487731969973538e-05
2023-01-21 13:23:04.304421: step: 292/527, loss: 0.4360926151275635
2023-01-21 13:23:05.455510: step: 296/527, loss: 0.012171745300292969
2023-01-21 13:23:06.544280: step: 300/527, loss: 0.2054843008518219
2023-01-21 13:23:07.612008: step: 304/527, loss: 0.0017966270679607987
2023-01-21 13:23:08.721574: step: 308/527, loss: 0.0009456634288653731
2023-01-21 13:23:09.823430: step: 312/527, loss: 0.0014430999290198088
2023-01-21 13:23:10.924896: step: 316/527, loss: 0.0018686294788494706
2023-01-21 13:23:12.037749: step: 320/527, loss: 2.3746491933707148e-05
2023-01-21 13:23:13.148046: step: 324/527, loss: 0.04205608367919922
2023-01-21 13:23:14.242647: step: 328/527, loss: 5.7983401347883046e-05
2023-01-21 13:23:15.343508: step: 332/527, loss: 0.00044040678767487407
2023-01-21 13:23:16.460603: step: 336/527, loss: 0.0015327453147619963
2023-01-21 13:23:17.568987: step: 340/527, loss: 0.001429653144441545
2023-01-21 13:23:18.666332: step: 344/527, loss: 0.0008526802412234247
2023-01-21 13:23:19.793701: step: 348/527, loss: 0.010652541182935238
2023-01-21 13:23:20.935423: step: 352/527, loss: 0.00011787413677666336
2023-01-21 13:23:22.034098: step: 356/527, loss: 3.719329833984375e-05
2023-01-21 13:23:23.129061: step: 360/527, loss: 0.010147476568818092
2023-01-21 13:23:24.227414: step: 364/527, loss: 0.040294744074344635
2023-01-21 13:23:25.355487: step: 368/527, loss: 1.316070574830519e-05
2023-01-21 13:23:26.472415: step: 372/527, loss: 0.0001508712739450857
2023-01-21 13:23:27.582913: step: 376/527, loss: 0.000629425048828125
2023-01-21 13:23:28.678916: step: 380/527, loss: 0.0002520561101846397
2023-01-21 13:23:29.820117: step: 384/527, loss: 4.9781796406023204e-05
2023-01-21 13:23:30.946602: step: 388/527, loss: 0.00012855530076194555
2023-01-21 13:23:32.065816: step: 392/527, loss: 0.02536744996905327
2023-01-21 13:23:33.178894: step: 396/527, loss: 1.9741059077205136e-05
2023-01-21 13:23:34.296712: step: 400/527, loss: 0.011190605349838734
2023-01-21 13:23:35.413713: step: 404/527, loss: 0.0005252361297607422
2023-01-21 13:23:36.522577: step: 408/527, loss: 5.102157501823967e-06
2023-01-21 13:23:37.643730: step: 412/527, loss: 1.239776611328125e-05
2023-01-21 13:23:38.760911: step: 416/527, loss: 0.016800880432128906
2023-01-21 13:23:39.855261: step: 420/527, loss: 0.0001065254255081527
2023-01-21 13:23:41.014363: step: 424/527, loss: 5.760193380410783e-05
2023-01-21 13:23:42.112714: step: 428/527, loss: 2.47955322265625e-05
2023-01-21 13:23:43.243521: step: 432/527, loss: 0.09670142829418182
2023-01-21 13:23:44.350020: step: 436/527, loss: 3.604888843256049e-05
2023-01-21 13:23:45.463789: step: 440/527, loss: 5.91278057981981e-06
2023-01-21 13:23:46.562955: step: 444/527, loss: 2.4795533590804553e-06
2023-01-21 13:23:47.657895: step: 448/527, loss: 0.0638696700334549
2023-01-21 13:23:48.786893: step: 452/527, loss: 2.47955322265625e-05
2023-01-21 13:23:49.896511: step: 456/527, loss: 0.0022597312927246094
2023-01-21 13:23:51.037970: step: 460/527, loss: 2.574920654296875e-05
2023-01-21 13:23:52.146003: step: 464/527, loss: 0.03964271768927574
2023-01-21 13:23:53.272700: step: 468/527, loss: 8.811951556708664e-05
2023-01-21 13:23:54.397239: step: 472/527, loss: 0.047460176050662994
2023-01-21 13:23:55.533370: step: 476/527, loss: 0.008591747842729092
2023-01-21 13:23:56.639470: step: 480/527, loss: 0.015171051025390625
2023-01-21 13:23:57.763759: step: 484/527, loss: 0.0016754150856286287
2023-01-21 13:23:58.870357: step: 488/527, loss: 0.001619052840396762
2023-01-21 13:23:59.982781: step: 492/527, loss: 8.58306884765625e-06
2023-01-21 13:24:01.093530: step: 496/527, loss: 0.0002715110604185611
2023-01-21 13:24:02.202002: step: 500/527, loss: 0.010997009463608265
2023-01-21 13:24:03.323398: step: 504/527, loss: 0.03769245371222496
2023-01-21 13:24:04.420854: step: 508/527, loss: 0.012016678228974342
2023-01-21 13:24:05.570432: step: 512/527, loss: 2.212524486822076e-05
2023-01-21 13:24:06.697235: step: 516/527, loss: 0.01716017723083496
2023-01-21 13:24:07.805023: step: 520/527, loss: 0.0001218795805471018
2023-01-21 13:24:08.888750: step: 524/527, loss: 1.33514404296875e-05
2023-01-21 13:24:10.039744: step: 528/527, loss: 0.019282149150967598
2023-01-21 13:24:11.138520: step: 532/527, loss: 9.956360008800402e-05
2023-01-21 13:24:12.216876: step: 536/527, loss: 0.0005784988170489669
2023-01-21 13:24:13.326928: step: 540/527, loss: 0.00015506744966842234
2023-01-21 13:24:14.486608: step: 544/527, loss: 0.00017766954260878265
2023-01-21 13:24:15.570002: step: 548/527, loss: -1.9073513612966053e-07
2023-01-21 13:24:16.681897: step: 552/527, loss: 0.022127151489257812
2023-01-21 13:24:17.776825: step: 556/527, loss: 0.0011636734707280993
2023-01-21 13:24:18.885021: step: 560/527, loss: 0.009599637240171432
2023-01-21 13:24:20.013689: step: 564/527, loss: 0.0012530327076092362
2023-01-21 13:24:21.188632: step: 568/527, loss: 0.013079453259706497
2023-01-21 13:24:22.294508: step: 572/527, loss: 1.1444091796875e-05
2023-01-21 13:24:23.433414: step: 576/527, loss: 0.01215906161814928
2023-01-21 13:24:24.539708: step: 580/527, loss: 0.004487800411880016
2023-01-21 13:24:25.643729: step: 584/527, loss: 0.00047855376033112407
2023-01-21 13:24:26.783358: step: 588/527, loss: 0.018651390448212624
2023-01-21 13:24:27.914966: step: 592/527, loss: 0.06014423444867134
2023-01-21 13:24:29.021201: step: 596/527, loss: 0.0021610737312585115
2023-01-21 13:24:30.103879: step: 600/527, loss: 3.089904930675402e-05
2023-01-21 13:24:31.232272: step: 604/527, loss: 0.017348574474453926
2023-01-21 13:24:32.326135: step: 608/527, loss: 8.94546537892893e-05
2023-01-21 13:24:33.463241: step: 612/527, loss: 0.00970010831952095
2023-01-21 13:24:34.596643: step: 616/527, loss: 5.34057608092553e-06
2023-01-21 13:24:35.741733: step: 620/527, loss: 0.08303900063037872
2023-01-21 13:24:36.876017: step: 624/527, loss: 0.028952598571777344
2023-01-21 13:24:37.968983: step: 628/527, loss: 3.719329924933845e-06
2023-01-21 13:24:39.059419: step: 632/527, loss: 2.5749204723979346e-06
2023-01-21 13:24:40.188252: step: 636/527, loss: 0.0001691818324616179
2023-01-21 13:24:41.287275: step: 640/527, loss: 0.015911294147372246
2023-01-21 13:24:42.382568: step: 644/527, loss: 0.0034130571875721216
2023-01-21 13:24:43.525010: step: 648/527, loss: 0.0005619049188680947
2023-01-21 13:24:44.652668: step: 652/527, loss: 0.010981273837387562
2023-01-21 13:24:45.760471: step: 656/527, loss: 0.0025578499771654606
2023-01-21 13:24:46.871794: step: 660/527, loss: 2.3698807126493193e-05
2023-01-21 13:24:47.996860: step: 664/527, loss: 0.002362918807193637
2023-01-21 13:24:49.094743: step: 668/527, loss: 0.017154216766357422
2023-01-21 13:24:50.226266: step: 672/527, loss: 1.0490417707842425e-06
2023-01-21 13:24:51.338859: step: 676/527, loss: 0.02232055552303791
2023-01-21 13:24:52.471276: step: 680/527, loss: 0.032080650329589844
2023-01-21 13:24:53.610086: step: 684/527, loss: 9.250640869140625e-05
2023-01-21 13:24:54.687487: step: 688/527, loss: 6.48498553346144e-06
2023-01-21 13:24:55.826430: step: 692/527, loss: 0.0012264251708984375
2023-01-21 13:24:56.963975: step: 696/527, loss: 2.9563905627583154e-05
2023-01-21 13:24:58.052257: step: 700/527, loss: 0.0020650862716138363
2023-01-21 13:24:59.171805: step: 704/527, loss: 1.8787384760798886e-05
2023-01-21 13:25:00.269633: step: 708/527, loss: 0.00014705659123137593
2023-01-21 13:25:01.397075: step: 712/527, loss: 4.57763690064894e-06
2023-01-21 13:25:02.536873: step: 716/527, loss: 3.051757857974735e-06
2023-01-21 13:25:03.640242: step: 720/527, loss: 0.00011768341937568039
2023-01-21 13:25:04.728861: step: 724/527, loss: 0.012250233441591263
2023-01-21 13:25:05.852859: step: 728/527, loss: 1.163482647825731e-05
2023-01-21 13:25:07.024849: step: 732/527, loss: 0.0001356124848825857
2023-01-21 13:25:08.150937: step: 736/527, loss: 0.011827087961137295
2023-01-21 13:25:09.300856: step: 740/527, loss: 0.02020740695297718
2023-01-21 13:25:10.408540: step: 744/527, loss: 0.02274017408490181
2023-01-21 13:25:11.522256: step: 748/527, loss: 0.00014557837857864797
2023-01-21 13:25:12.663609: step: 752/527, loss: 0.005122947972267866
2023-01-21 13:25:13.804231: step: 756/527, loss: 3.910065061063506e-06
2023-01-21 13:25:14.915262: step: 760/527, loss: 0.00037021638127043843
2023-01-21 13:25:16.050215: step: 764/527, loss: 0.01791534572839737
2023-01-21 13:25:17.194017: step: 768/527, loss: 0.0015121460892260075
2023-01-21 13:25:18.296689: step: 772/527, loss: 1.1682510375976562e-05
2023-01-21 13:25:19.493060: step: 776/527, loss: 0.0676511749625206
2023-01-21 13:25:20.602409: step: 780/527, loss: 0.05192718654870987
2023-01-21 13:25:21.728321: step: 784/527, loss: 0.03401775285601616
2023-01-21 13:25:22.857470: step: 788/527, loss: 0.005163765046745539
2023-01-21 13:25:23.952402: step: 792/527, loss: 0.0007431030389852822
2023-01-21 13:25:25.097573: step: 796/527, loss: 6.418228440452367e-05
2023-01-21 13:25:26.217279: step: 800/527, loss: 0.0024358273949474096
2023-01-21 13:25:27.319427: step: 804/527, loss: 0.00030803680419921875
2023-01-21 13:25:28.447831: step: 808/527, loss: 0.0018108368385583162
2023-01-21 13:25:29.549096: step: 812/527, loss: 0.032341767102479935
2023-01-21 13:25:30.660121: step: 816/527, loss: 0.0004055977042298764
2023-01-21 13:25:31.777086: step: 820/527, loss: 0.012580872513353825
2023-01-21 13:25:32.863567: step: 824/527, loss: 0.0013387680519372225
2023-01-21 13:25:33.989300: step: 828/527, loss: 0.0004814147832803428
2023-01-21 13:25:35.128279: step: 832/527, loss: 0.0003933906555175781
2023-01-21 13:25:36.226870: step: 836/527, loss: 0.11872200667858124
2023-01-21 13:25:37.326201: step: 840/527, loss: 0.00018415450176689774
2023-01-21 13:25:38.426703: step: 844/527, loss: 0.012525368481874466
2023-01-21 13:25:39.500728: step: 848/527, loss: 0.014970016665756702
2023-01-21 13:25:40.606686: step: 852/527, loss: 0.026124905794858932
2023-01-21 13:25:41.720168: step: 856/527, loss: 0.0007524490356445312
2023-01-21 13:25:42.847234: step: 860/527, loss: 0.005696296691894531
2023-01-21 13:25:43.949069: step: 864/527, loss: 0.0031391142401844263
2023-01-21 13:25:45.034255: step: 868/527, loss: 0.0009866715408861637
2023-01-21 13:25:46.137783: step: 872/527, loss: 0.01913614384829998
2023-01-21 13:25:47.247554: step: 876/527, loss: 0.020315933972597122
2023-01-21 13:25:48.362189: step: 880/527, loss: 0.01634521596133709
2023-01-21 13:25:49.497579: step: 884/527, loss: 0.00026378632173873484
2023-01-21 13:25:50.605464: step: 888/527, loss: 0.008613204583525658
2023-01-21 13:25:51.732552: step: 892/527, loss: 0.0049607278779149055
2023-01-21 13:25:52.857722: step: 896/527, loss: 0.002787971403449774
2023-01-21 13:25:53.963432: step: 900/527, loss: 0.005481529049575329
2023-01-21 13:25:55.092646: step: 904/527, loss: 3.2901763916015625e-05
2023-01-21 13:25:56.220395: step: 908/527, loss: 0.003894519992172718
2023-01-21 13:25:57.320931: step: 912/527, loss: 0.0002590179501567036
2023-01-21 13:25:58.406297: step: 916/527, loss: 0.016374588012695312
2023-01-21 13:25:59.524517: step: 920/527, loss: 0.00014839171490166336
2023-01-21 13:26:00.640008: step: 924/527, loss: 3.910064697265625e-05
2023-01-21 13:26:01.748583: step: 928/527, loss: 0.029592515900731087
2023-01-21 13:26:02.842337: step: 932/527, loss: 0.0004029273986816406
2023-01-21 13:26:03.936980: step: 936/527, loss: 0.0037467002402991056
2023-01-21 13:26:05.026570: step: 940/527, loss: 1.0633469173626509e-05
2023-01-21 13:26:06.158019: step: 944/527, loss: 0.0017711640102788806
2023-01-21 13:26:07.314152: step: 948/527, loss: 3.509521411615424e-05
2023-01-21 13:26:08.425224: step: 952/527, loss: 0.0012497903080657125
2023-01-21 13:26:09.533271: step: 956/527, loss: 7.190704491222277e-05
2023-01-21 13:26:10.653941: step: 960/527, loss: 0.005519104190170765
2023-01-21 13:26:11.778923: step: 964/527, loss: 0.00021429063053801656
2023-01-21 13:26:12.896323: step: 968/527, loss: 3.337860107421875e-06
2023-01-21 13:26:14.022846: step: 972/527, loss: 0.016888046637177467
2023-01-21 13:26:15.129157: step: 976/527, loss: 0.10534238815307617
2023-01-21 13:26:16.214833: step: 980/527, loss: -5.340576535672881e-06
2023-01-21 13:26:17.333378: step: 984/527, loss: -1.3065337952866685e-05
2023-01-21 13:26:18.441832: step: 988/527, loss: 0.00045223237248137593
2023-01-21 13:26:19.578111: step: 992/527, loss: 0.027906036004424095
2023-01-21 13:26:20.676778: step: 996/527, loss: 0.002360534854233265
2023-01-21 13:26:21.775921: step: 1000/527, loss: 0.0001746654452290386
2023-01-21 13:26:22.876721: step: 1004/527, loss: 4.76837158203125e-07
2023-01-21 13:26:23.988486: step: 1008/527, loss: 0.0008020401000976562
2023-01-21 13:26:25.114107: step: 1012/527, loss: 0.10755062103271484
2023-01-21 13:26:26.208636: step: 1016/527, loss: 0.005852890200912952
2023-01-21 13:26:27.341219: step: 1020/527, loss: 0.0003234863106627017
2023-01-21 13:26:28.425153: step: 1024/527, loss: 0.006233310792595148
2023-01-21 13:26:29.520226: step: 1028/527, loss: 0.025316238403320312
2023-01-21 13:26:30.643036: step: 1032/527, loss: 0.005932998843491077
2023-01-21 13:26:31.772411: step: 1036/527, loss: 0.00048189167864620686
2023-01-21 13:26:32.861306: step: 1040/527, loss: 0.00022716523380950093
2023-01-21 13:26:33.985232: step: 1044/527, loss: 3.843307786155492e-05
2023-01-21 13:26:35.114368: step: 1048/527, loss: 0.011029244400560856
2023-01-21 13:26:36.220742: step: 1052/527, loss: 0.0005187034839764237
2023-01-21 13:26:37.325978: step: 1056/527, loss: 0.0021114349365234375
2023-01-21 13:26:38.446568: step: 1060/527, loss: 0.0026578903198242188
2023-01-21 13:26:39.567221: step: 1064/527, loss: 0.04144611582159996
2023-01-21 13:26:40.674085: step: 1068/527, loss: 0.0001848220854299143
2023-01-21 13:26:41.761855: step: 1072/527, loss: 1.1825562069134321e-05
2023-01-21 13:26:42.885054: step: 1076/527, loss: 1.4305115882962127e-06
2023-01-21 13:26:44.003949: step: 1080/527, loss: 0.0007661819690838456
2023-01-21 13:26:45.127893: step: 1084/527, loss: 0.12997588515281677
2023-01-21 13:26:46.251919: step: 1088/527, loss: 1.950263867911417e-05
2023-01-21 13:26:47.342371: step: 1092/527, loss: 1.4209747860149946e-05
2023-01-21 13:26:48.428434: step: 1096/527, loss: -5.817413693876006e-06
2023-01-21 13:26:49.536255: step: 1100/527, loss: 4.1484832763671875e-05
2023-01-21 13:26:50.647428: step: 1104/527, loss: 2.4509430659236386e-05
2023-01-21 13:26:51.752387: step: 1108/527, loss: 7.696151442360133e-05
2023-01-21 13:26:52.860682: step: 1112/527, loss: 5.645752025884576e-05
2023-01-21 13:26:53.976461: step: 1116/527, loss: 0.008178806863725185
2023-01-21 13:26:55.069228: step: 1120/527, loss: 0.004252815619111061
2023-01-21 13:26:56.178448: step: 1124/527, loss: 0.0008460998651571572
2023-01-21 13:26:57.279569: step: 1128/527, loss: 0.07693938910961151
2023-01-21 13:26:58.399792: step: 1132/527, loss: 0.0008541106944903731
2023-01-21 13:26:59.520583: step: 1136/527, loss: 4.863739013671875e-05
2023-01-21 13:27:00.634337: step: 1140/527, loss: 0.0014945983421057463
2023-01-21 13:27:01.779363: step: 1144/527, loss: 0.004836368374526501
2023-01-21 13:27:02.916649: step: 1148/527, loss: 0.0009296417701989412
2023-01-21 13:27:04.034412: step: 1152/527, loss: 0.0004594803031068295
2023-01-21 13:27:05.141517: step: 1156/527, loss: 6.67572021484375e-06
2023-01-21 13:27:06.241643: step: 1160/527, loss: -1.8119812921213452e-06
2023-01-21 13:27:07.345966: step: 1164/527, loss: 0.0023652436211705208
2023-01-21 13:27:08.450627: step: 1168/527, loss: 0.0007623672718182206
2023-01-21 13:27:09.591763: step: 1172/527, loss: 0.020679855719208717
2023-01-21 13:27:10.705932: step: 1176/527, loss: 1.0681153071345761e-05
2023-01-21 13:27:11.808483: step: 1180/527, loss: 1.316070574830519e-05
2023-01-21 13:27:12.928155: step: 1184/527, loss: 1.449584942747606e-05
2023-01-21 13:27:14.050039: step: 1188/527, loss: 0.0017274857964366674
2023-01-21 13:27:15.147251: step: 1192/527, loss: 6.85691848048009e-05
2023-01-21 13:27:16.254134: step: 1196/527, loss: -2.155303991457913e-05
2023-01-21 13:27:17.358269: step: 1200/527, loss: 0.007727622985839844
2023-01-21 13:27:18.481776: step: 1204/527, loss: 0.0018629074329510331
2023-01-21 13:27:19.588296: step: 1208/527, loss: 3.566742088878527e-05
2023-01-21 13:27:20.735829: step: 1212/527, loss: 2.765655608527595e-06
2023-01-21 13:27:21.842861: step: 1216/527, loss: 0.021380042657256126
2023-01-21 13:27:22.925713: step: 1220/527, loss: 0.002145195147022605
2023-01-21 13:27:24.048069: step: 1224/527, loss: 0.13016262650489807
2023-01-21 13:27:25.152482: step: 1228/527, loss: 0.0023868561256676912
2023-01-21 13:27:26.286875: step: 1232/527, loss: 0.025242233648896217
2023-01-21 13:27:27.393243: step: 1236/527, loss: 0.0061647770926356316
2023-01-21 13:27:28.527227: step: 1240/527, loss: 0.0017040251987054944
2023-01-21 13:27:29.624348: step: 1244/527, loss: 1.8072127204504795e-05
2023-01-21 13:27:30.743754: step: 1248/527, loss: 0.0004831314436160028
2023-01-21 13:27:31.832399: step: 1252/527, loss: 0.0008007049327716231
2023-01-21 13:27:32.942589: step: 1256/527, loss: 0.00021262170048430562
2023-01-21 13:27:34.104831: step: 1260/527, loss: 0.0006473064422607422
2023-01-21 13:27:35.225748: step: 1264/527, loss: -2.2983551389188506e-05
2023-01-21 13:27:36.326175: step: 1268/527, loss: 2.079009937006049e-05
2023-01-21 13:27:37.428591: step: 1272/527, loss: 0.05006871372461319
2023-01-21 13:27:38.543626: step: 1276/527, loss: 2.00748436327558e-05
2023-01-21 13:27:39.643045: step: 1280/527, loss: 3.3092499506892636e-05
2023-01-21 13:27:40.780519: step: 1284/527, loss: 0.07275257259607315
2023-01-21 13:27:41.905559: step: 1288/527, loss: 0.04023456573486328
2023-01-21 13:27:43.008881: step: 1292/527, loss: 0.002548122312873602
2023-01-21 13:27:44.123397: step: 1296/527, loss: 0.0006945609929971397
2023-01-21 13:27:45.237157: step: 1300/527, loss: 0.03854179382324219
2023-01-21 13:27:46.357153: step: 1304/527, loss: 0.0011818886268883944
2023-01-21 13:27:47.496061: step: 1308/527, loss: 0.003962325863540173
2023-01-21 13:27:48.590166: step: 1312/527, loss: 5.34057608092553e-06
2023-01-21 13:27:49.705351: step: 1316/527, loss: 0.00035710333031602204
2023-01-21 13:27:50.798130: step: 1320/527, loss: 0.00016040803166106343
2023-01-21 13:27:51.950000: step: 1324/527, loss: 0.010517120361328125
2023-01-21 13:27:53.079055: step: 1328/527, loss: 3.471374657237902e-05
2023-01-21 13:27:54.166287: step: 1332/527, loss: 0.0023520467802882195
2023-01-21 13:27:55.289265: step: 1336/527, loss: 0.00012979508028365672
2023-01-21 13:27:56.411363: step: 1340/527, loss: 0.0011652946704998612
2023-01-21 13:27:57.545514: step: 1344/527, loss: 1.8119812921213452e-06
2023-01-21 13:27:58.684762: step: 1348/527, loss: 0.007854747585952282
2023-01-21 13:27:59.782049: step: 1352/527, loss: 7.677078428969253e-06
2023-01-21 13:28:00.888932: step: 1356/527, loss: 0.005829811096191406
2023-01-21 13:28:02.008011: step: 1360/527, loss: 0.0025429725646972656
2023-01-21 13:28:03.100209: step: 1364/527, loss: 1.0013580322265625e-05
2023-01-21 13:28:04.210497: step: 1368/527, loss: 0.0012989044189453125
2023-01-21 13:28:05.360988: step: 1372/527, loss: 0.0002666473446879536
2023-01-21 13:28:06.528203: step: 1376/527, loss: 0.0013496398460119963
2023-01-21 13:28:07.644995: step: 1380/527, loss: 0.007849693298339844
2023-01-21 13:28:08.749435: step: 1384/527, loss: 0.0001634597865631804
2023-01-21 13:28:09.856717: step: 1388/527, loss: 0.0018383980495855212
2023-01-21 13:28:10.975473: step: 1392/527, loss: 0.009204864501953125
2023-01-21 13:28:12.085037: step: 1396/527, loss: 0.0009677887428551912
2023-01-21 13:28:13.210857: step: 1400/527, loss: 0.003944301512092352
2023-01-21 13:28:14.311584: step: 1404/527, loss: 0.0002063751162495464
2023-01-21 13:28:15.445412: step: 1408/527, loss: 0.0030055048409849405
2023-01-21 13:28:16.557476: step: 1412/527, loss: 0.009550857357680798
2023-01-21 13:28:17.677151: step: 1416/527, loss: 0.00971241109073162
2023-01-21 13:28:18.797915: step: 1420/527, loss: 0.03001384809613228
2023-01-21 13:28:19.926758: step: 1424/527, loss: 0.0031227110885083675
2023-01-21 13:28:21.034677: step: 1428/527, loss: 0.9736614227294922
2023-01-21 13:28:22.165221: step: 1432/527, loss: 0.018283559009432793
2023-01-21 13:28:23.285889: step: 1436/527, loss: 0.0017353057628497481
2023-01-21 13:28:24.413137: step: 1440/527, loss: 1.6593934560660273e-05
2023-01-21 13:28:25.535214: step: 1444/527, loss: 0.0002774238819256425
2023-01-21 13:28:26.646133: step: 1448/527, loss: 0.006188774015754461
2023-01-21 13:28:27.780798: step: 1452/527, loss: 5.044937279308215e-05
2023-01-21 13:28:28.897349: step: 1456/527, loss: 0.0005575180402956903
2023-01-21 13:28:29.977110: step: 1460/527, loss: 0.00047149660531431437
2023-01-21 13:28:31.104020: step: 1464/527, loss: 7.295608520507812e-05
2023-01-21 13:28:32.188642: step: 1468/527, loss: 0.03987989202141762
2023-01-21 13:28:33.279358: step: 1472/527, loss: 0.0005826950073242188
2023-01-21 13:28:34.383983: step: 1476/527, loss: 0.08063941448926926
2023-01-21 13:28:35.533138: step: 1480/527, loss: 0.0001127243012888357
2023-01-21 13:28:36.681629: step: 1484/527, loss: 0.01488494873046875
2023-01-21 13:28:37.771036: step: 1488/527, loss: 0.00038776398287154734
2023-01-21 13:28:38.863579: step: 1492/527, loss: 0.00019121171499136835
2023-01-21 13:28:39.977662: step: 1496/527, loss: 0.0426454097032547
2023-01-21 13:28:41.077830: step: 1500/527, loss: 0.00022297502437140793
2023-01-21 13:28:42.172705: step: 1504/527, loss: 0.019687939435243607
2023-01-21 13:28:43.296575: step: 1508/527, loss: 0.0024658204056322575
2023-01-21 13:28:44.410655: step: 1512/527, loss: 0.00297813443467021
2023-01-21 13:28:45.515288: step: 1516/527, loss: 0.015665054321289062
2023-01-21 13:28:46.610327: step: 1520/527, loss: 0.00045242311898618937
2023-01-21 13:28:47.722754: step: 1524/527, loss: 1.2350083125056699e-05
2023-01-21 13:28:48.851099: step: 1528/527, loss: 0.11341685801744461
2023-01-21 13:28:49.959368: step: 1532/527, loss: 1.8596649169921875e-05
2023-01-21 13:28:51.097117: step: 1536/527, loss: 0.0017905235290527344
2023-01-21 13:28:52.231396: step: 1540/527, loss: 5.550384958041832e-05
2023-01-21 13:28:53.400384: step: 1544/527, loss: 0.26640960574150085
2023-01-21 13:28:54.536785: step: 1548/527, loss: 5.7220458984375e-06
2023-01-21 13:28:55.656277: step: 1552/527, loss: 0.07365904003381729
2023-01-21 13:28:56.799511: step: 1556/527, loss: 0.00035109519376419485
2023-01-21 13:28:57.940740: step: 1560/527, loss: 1.583099401614163e-05
2023-01-21 13:28:59.070424: step: 1564/527, loss: 0.0001985549897653982
2023-01-21 13:29:00.183212: step: 1568/527, loss: 0.01279306411743164
2023-01-21 13:29:01.273809: step: 1572/527, loss: 0.00038471221341751516
2023-01-21 13:29:02.356612: step: 1576/527, loss: 0.008066177368164062
2023-01-21 13:29:03.501423: step: 1580/527, loss: 0.0946468785405159
2023-01-21 13:29:04.635689: step: 1584/527, loss: 0.0013032435672357678
2023-01-21 13:29:05.773472: step: 1588/527, loss: 0.3102518320083618
2023-01-21 13:29:06.885807: step: 1592/527, loss: 0.009136534295976162
2023-01-21 13:29:07.999747: step: 1596/527, loss: 0.032277487218379974
2023-01-21 13:29:09.100580: step: 1600/527, loss: 0.003069591475650668
2023-01-21 13:29:10.282378: step: 1604/527, loss: 0.00023527145094703883
2023-01-21 13:29:11.370618: step: 1608/527, loss: 0.0001155376376118511
2023-01-21 13:29:12.504819: step: 1612/527, loss: 0.0011808396084234118
2023-01-21 13:29:13.648675: step: 1616/527, loss: 7.22885160939768e-05
2023-01-21 13:29:14.752632: step: 1620/527, loss: 0.00042495731031522155
2023-01-21 13:29:15.861725: step: 1624/527, loss: 0.0016637801891192794
2023-01-21 13:29:16.984815: step: 1628/527, loss: 0.0006689071306027472
2023-01-21 13:29:18.133823: step: 1632/527, loss: 1.1444092706369702e-05
2023-01-21 13:29:19.234967: step: 1636/527, loss: 3.728866431629285e-05
2023-01-21 13:29:20.392167: step: 1640/527, loss: 0.00015182494826149195
2023-01-21 13:29:21.488267: step: 1644/527, loss: 0.03316822275519371
2023-01-21 13:29:22.606436: step: 1648/527, loss: 0.006784438621252775
2023-01-21 13:29:23.697904: step: 1652/527, loss: 0.006494045257568359
2023-01-21 13:29:24.821024: step: 1656/527, loss: 0.014848328195512295
2023-01-21 13:29:25.927864: step: 1660/527, loss: -7.629394644936838e-07
2023-01-21 13:29:27.031590: step: 1664/527, loss: 0.015440178103744984
2023-01-21 13:29:28.144806: step: 1668/527, loss: 0.0447477325797081
2023-01-21 13:29:29.267244: step: 1672/527, loss: 0.8014277219772339
2023-01-21 13:29:30.346439: step: 1676/527, loss: 1.6498564946232364e-05
2023-01-21 13:29:31.479191: step: 1680/527, loss: 0.0001235961972270161
2023-01-21 13:29:32.582841: step: 1684/527, loss: 0.00016412735567428172
2023-01-21 13:29:33.690573: step: 1688/527, loss: 0.023527145385742188
2023-01-21 13:29:34.784982: step: 1692/527, loss: 0.003637599991634488
2023-01-21 13:29:35.875233: step: 1696/527, loss: 0.0001808166562113911
2023-01-21 13:29:36.990639: step: 1700/527, loss: 0.003204345703125
2023-01-21 13:29:38.144905: step: 1704/527, loss: 0.002167606493458152
2023-01-21 13:29:39.260346: step: 1708/527, loss: 0.0015536308055743575
2023-01-21 13:29:40.363979: step: 1712/527, loss: 0.015065384097397327
2023-01-21 13:29:41.488914: step: 1716/527, loss: 6.299018423305824e-05
2023-01-21 13:29:42.607451: step: 1720/527, loss: 0.04913024976849556
2023-01-21 13:29:43.748638: step: 1724/527, loss: 0.0004507065168581903
2023-01-21 13:29:44.886208: step: 1728/527, loss: 6.208419654285535e-05
2023-01-21 13:29:46.002787: step: 1732/527, loss: 0.019384240731596947
2023-01-21 13:29:47.101252: step: 1736/527, loss: 0.30012303590774536
2023-01-21 13:29:48.226209: step: 1740/527, loss: 0.004107857123017311
2023-01-21 13:29:49.341898: step: 1744/527, loss: 0.0014113426441326737
2023-01-21 13:29:50.445294: step: 1748/527, loss: 9.670257713878527e-05
2023-01-21 13:29:51.566195: step: 1752/527, loss: 0.00024690627469681203
2023-01-21 13:29:52.655492: step: 1756/527, loss: 0.00108680734410882
2023-01-21 13:29:53.783767: step: 1760/527, loss: 0.0006296157953329384
2023-01-21 13:29:54.897493: step: 1764/527, loss: 0.01290679071098566
2023-01-21 13:29:56.024002: step: 1768/527, loss: 0.0018275261390954256
2023-01-21 13:29:57.155313: step: 1772/527, loss: 0.0036556245759129524
2023-01-21 13:29:58.272553: step: 1776/527, loss: 0.0007310867076739669
2023-01-21 13:29:59.383563: step: 1780/527, loss: 7.467270188499242e-05
2023-01-21 13:30:00.502802: step: 1784/527, loss: 0.0006832123035565019
2023-01-21 13:30:01.610244: step: 1788/527, loss: 2.1553041733568534e-05
2023-01-21 13:30:02.719945: step: 1792/527, loss: 0.0006723403930664062
2023-01-21 13:30:03.836784: step: 1796/527, loss: 2.365112231927924e-05
2023-01-21 13:30:04.937236: step: 1800/527, loss: 2.908706846938003e-06
2023-01-21 13:30:06.072228: step: 1804/527, loss: 0.013794518075883389
2023-01-21 13:30:07.193480: step: 1808/527, loss: 0.0018934250110760331
2023-01-21 13:30:08.293825: step: 1812/527, loss: 0.0011570453643798828
2023-01-21 13:30:09.419589: step: 1816/527, loss: 0.024321556091308594
2023-01-21 13:30:10.577783: step: 1820/527, loss: 0.0009059906005859375
2023-01-21 13:30:11.704785: step: 1824/527, loss: 0.05170383304357529
2023-01-21 13:30:12.822079: step: 1828/527, loss: 0.003927039913833141
2023-01-21 13:30:13.920919: step: 1832/527, loss: 0.0013512612786144018
2023-01-21 13:30:15.011310: step: 1836/527, loss: 0.035167694091796875
2023-01-21 13:30:16.103157: step: 1840/527, loss: 0.09160614758729935
2023-01-21 13:30:17.232690: step: 1844/527, loss: 0.03256836161017418
2023-01-21 13:30:18.351119: step: 1848/527, loss: 0.00025653839111328125
2023-01-21 13:30:19.474732: step: 1852/527, loss: 0.00012006759061478078
2023-01-21 13:30:20.577650: step: 1856/527, loss: 3.347397068864666e-05
2023-01-21 13:30:21.697371: step: 1860/527, loss: 0.0015501022571697831
2023-01-21 13:30:22.803457: step: 1864/527, loss: 0.00230998988263309
2023-01-21 13:30:23.919420: step: 1868/527, loss: 0.0005203246837481856
2023-01-21 13:30:25.027229: step: 1872/527, loss: 0.01584930531680584
2023-01-21 13:30:26.136234: step: 1876/527, loss: 5.321502612787299e-05
2023-01-21 13:30:27.269026: step: 1880/527, loss: 0.003977966494858265
2023-01-21 13:30:28.404819: step: 1884/527, loss: 0.04806585609912872
2023-01-21 13:30:29.513073: step: 1888/527, loss: 0.016765404492616653
2023-01-21 13:30:30.625279: step: 1892/527, loss: 0.0013014794094488025
2023-01-21 13:30:31.722592: step: 1896/527, loss: 8.726120722712949e-05
2023-01-21 13:30:32.832551: step: 1900/527, loss: 0.0025468827225267887
2023-01-21 13:30:33.927086: step: 1904/527, loss: 0.001960182096809149
2023-01-21 13:30:35.054682: step: 1908/527, loss: 0.044085029512643814
2023-01-21 13:30:36.187250: step: 1912/527, loss: 0.0008513450738973916
2023-01-21 13:30:37.302312: step: 1916/527, loss: 0.005043411627411842
2023-01-21 13:30:38.421053: step: 1920/527, loss: 0.0001661777641857043
2023-01-21 13:30:39.495734: step: 1924/527, loss: 0.012706947512924671
2023-01-21 13:30:40.598554: step: 1928/527, loss: 0.0038674355018883944
2023-01-21 13:30:41.691026: step: 1932/527, loss: 4.1580202378099784e-05
2023-01-21 13:30:42.811762: step: 1936/527, loss: 0.0007404327625408769
2023-01-21 13:30:43.977692: step: 1940/527, loss: 0.0009922027820721269
2023-01-21 13:30:45.108226: step: 1944/527, loss: 1.5060726404190063
2023-01-21 13:30:46.234929: step: 1948/527, loss: 0.0019271851051598787
2023-01-21 13:30:47.383328: step: 1952/527, loss: 9.5367431640625e-06
2023-01-21 13:30:48.504442: step: 1956/527, loss: 0.6008445024490356
2023-01-21 13:30:49.588040: step: 1960/527, loss: 0.0003595828893594444
2023-01-21 13:30:50.710316: step: 1964/527, loss: 0.00016765593318268657
2023-01-21 13:30:51.803748: step: 1968/527, loss: 0.00015335083298850805
2023-01-21 13:30:52.908336: step: 1972/527, loss: 0.0002494812069926411
2023-01-21 13:30:54.025798: step: 1976/527, loss: 1.087188684323337e-05
2023-01-21 13:30:55.148473: step: 1980/527, loss: 0.0006203174707479775
2023-01-21 13:30:56.301886: step: 1984/527, loss: 0.003362560411915183
2023-01-21 13:30:57.428792: step: 1988/527, loss: 0.009557723999023438
2023-01-21 13:30:58.559728: step: 1992/527, loss: 0.0008581161382608116
2023-01-21 13:30:59.642819: step: 1996/527, loss: 2.2029877072782256e-05
2023-01-21 13:31:00.770874: step: 2000/527, loss: 5.6743621826171875e-05
2023-01-21 13:31:01.882302: step: 2004/527, loss: 0.03742380067706108
2023-01-21 13:31:02.958358: step: 2008/527, loss: 0.0015326023567467928
2023-01-21 13:31:04.120063: step: 2012/527, loss: 0.010592365637421608
2023-01-21 13:31:05.256003: step: 2016/527, loss: 0.03694877773523331
2023-01-21 13:31:06.364335: step: 2020/527, loss: 0.0010503769153729081
2023-01-21 13:31:07.496284: step: 2024/527, loss: 0.0022166252601891756
2023-01-21 13:31:08.617056: step: 2028/527, loss: 0.0246169101446867
2023-01-21 13:31:09.726773: step: 2032/527, loss: 0.0015837668906897306
2023-01-21 13:31:10.817684: step: 2036/527, loss: 0.001325893448665738
2023-01-21 13:31:11.931134: step: 2040/527, loss: 0.001745843910612166
2023-01-21 13:31:13.051726: step: 2044/527, loss: 0.00905466079711914
2023-01-21 13:31:14.161866: step: 2048/527, loss: 0.007166385650634766
2023-01-21 13:31:15.282087: step: 2052/527, loss: 0.012956047430634499
2023-01-21 13:31:16.371577: step: 2056/527, loss: 0.012605667114257812
2023-01-21 13:31:17.506221: step: 2060/527, loss: 1.7166138377433526e-06
2023-01-21 13:31:18.619813: step: 2064/527, loss: 0.0010304928291589022
2023-01-21 13:31:19.714654: step: 2068/527, loss: 7.62939453125e-05
2023-01-21 13:31:20.823317: step: 2072/527, loss: 0.0009903907775878906
2023-01-21 13:31:21.942293: step: 2076/527, loss: 3.395080420887098e-05
2023-01-21 13:31:23.062211: step: 2080/527, loss: 0.0005800247308798134
2023-01-21 13:31:24.158173: step: 2084/527, loss: 0.0012081146705895662
2023-01-21 13:31:25.272808: step: 2088/527, loss: 0.004014301113784313
2023-01-21 13:31:26.371549: step: 2092/527, loss: 0.0012030601501464844
2023-01-21 13:31:27.466790: step: 2096/527, loss: 0.00019807815260719508
2023-01-21 13:31:28.584187: step: 2100/527, loss: 0.003503823187202215
2023-01-21 13:31:29.699124: step: 2104/527, loss: 0.0002639770391397178
2023-01-21 13:31:30.808081: step: 2108/527, loss: 4.205703953630291e-05
==================================================
Loss: 0.020
--------------------
Dev: {'event': {'p': 0.5827482447342026, 'r': 0.7736351531291611, 'f1': 0.6647597254004577}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28}
Test: {'event': {'p': 0.6218034993270525, 'r': 0.792, 'f1': 0.6966574516210103}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28}
Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28}
Korean: {'event': {'p': 0.6226415094339622, 'r': 0.5238095238095238, 'f1': 0.5689655172413793}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28}
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': 28}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
******************************
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 13:32:12.311658: step: 4/527, loss: 2.145767393813003e-05
2023-01-21 13:32:13.434971: step: 8/527, loss: 0.0010246277088299394
2023-01-21 13:32:14.566407: step: 12/527, loss: 0.003389644669368863
2023-01-21 13:32:15.692349: step: 16/527, loss: 0.0010288238991051912
2023-01-21 13:32:16.835244: step: 20/527, loss: 0.04168825224041939
2023-01-21 13:32:17.942703: step: 24/527, loss: -3.337860107421875e-06
2023-01-21 13:32:19.065663: step: 28/527, loss: -4.57763690064894e-06
2023-01-21 13:32:20.185226: step: 32/527, loss: 0.02978038787841797
2023-01-21 13:32:21.315421: step: 36/527, loss: 0.0001316070556640625
2023-01-21 13:32:22.453004: step: 40/527, loss: 0.00010643005953170359
2023-01-21 13:32:23.575507: step: 44/527, loss: 0.000716137932613492
2023-01-21 13:32:24.658811: step: 48/527, loss: 0.011913490481674671
2023-01-21 13:32:25.823938: step: 52/527, loss: 0.0021017075050622225
2023-01-21 13:32:26.934062: step: 56/527, loss: 0.0006391525384970009
2023-01-21 13:32:28.029350: step: 60/527, loss: 6.408691842807457e-05
2023-01-21 13:32:29.140076: step: 64/527, loss: 0.0007295608520507812
2023-01-21 13:32:30.294096: step: 68/527, loss: 2.8228761948412284e-05
2023-01-21 13:32:31.390998: step: 72/527, loss: 0.0016901969211176038
2023-01-21 13:32:32.511132: step: 76/527, loss: 0.0009433747036382556
2023-01-21 13:32:33.615750: step: 80/527, loss: 1.5878678823355585e-05
2023-01-21 13:32:34.734242: step: 84/527, loss: 0.0006093978881835938
2023-01-21 13:32:35.843376: step: 88/527, loss: 2.293586658197455e-05
2023-01-21 13:32:36.946858: step: 92/527, loss: 0.04685802385210991
2023-01-21 13:32:38.057609: step: 96/527, loss: 0.004385948181152344
2023-01-21 13:32:39.157650: step: 100/527, loss: 4.882812572759576e-05
2023-01-21 13:32:40.271897: step: 104/527, loss: 0.0015909194480627775
2023-01-21 13:32:41.389951: step: 108/527, loss: 0.000127696999697946
2023-01-21 13:32:42.552456: step: 112/527, loss: 0.002540683839470148
2023-01-21 13:32:43.691764: step: 116/527, loss: 0.002883529756218195
2023-01-21 13:32:44.780987: step: 120/527, loss: 0.004446601960808039
2023-01-21 13:32:45.912641: step: 124/527, loss: 0.034989356994628906
2023-01-21 13:32:47.021843: step: 128/527, loss: 1.9073486328125e-05
2023-01-21 13:32:48.113278: step: 132/527, loss: 0.00025768281193450093
2023-01-21 13:32:49.256736: step: 136/527, loss: 0.003114318707957864
2023-01-21 13:32:50.341548: step: 140/527, loss: 0.03297596052289009
2023-01-21 13:32:51.479385: step: 144/527, loss: 4.482269287109375e-05
2023-01-21 13:32:52.596941: step: 148/527, loss: 0.0004102706734556705
2023-01-21 13:32:53.696206: step: 152/527, loss: 0.0004593849298544228
2023-01-21 13:32:54.812082: step: 156/527, loss: 0.03136596828699112
2023-01-21 13:32:55.925833: step: 160/527, loss: 0.00022411346435546875
2023-01-21 13:32:57.056154: step: 164/527, loss: 0.003842306323349476
2023-01-21 13:32:58.189766: step: 168/527, loss: 0.00010113716416526586
2023-01-21 13:32:59.290236: step: 172/527, loss: 0.0002590179501567036
2023-01-21 13:33:00.390033: step: 176/527, loss: 0.00016012191190384328
2023-01-21 13:33:01.478901: step: 180/527, loss: 0.07830315083265305
2023-01-21 13:33:02.604672: step: 184/527, loss: 0.0003841400030069053
2023-01-21 13:33:03.704847: step: 188/527, loss: 4.386902219266631e-06
2023-01-21 13:33:04.824104: step: 192/527, loss: 0.001582336495630443
2023-01-21 13:33:05.941553: step: 196/527, loss: -8.82148754044465e-07
2023-01-21 13:33:07.066905: step: 200/527, loss: 0.00022182465181685984
2023-01-21 13:33:08.212474: step: 204/527, loss: 0.00016613007755950093
2023-01-21 13:33:09.315019: step: 208/527, loss: 0.017932463437318802
2023-01-21 13:33:10.422593: step: 212/527, loss: 0.00012130737741244957
2023-01-21 13:33:11.535144: step: 216/527, loss: 0.00566520681604743
2023-01-21 13:33:12.645547: step: 220/527, loss: 0.011272668838500977
2023-01-21 13:33:13.762999: step: 224/527, loss: 0.03747711330652237
2023-01-21 13:33:14.875822: step: 228/527, loss: 0.001617383910343051
2023-01-21 13:33:15.982683: step: 232/527, loss: 0.0033159255981445312
2023-01-21 13:33:17.085175: step: 236/527, loss: 0.010769272223114967
2023-01-21 13:33:18.188272: step: 240/527, loss: 0.019602585583925247
2023-01-21 13:33:19.339467: step: 244/527, loss: 0.0008128166664391756
2023-01-21 13:33:20.478485: step: 248/527, loss: 0.0005959987756796181
2023-01-21 13:33:21.596432: step: 252/527, loss: 0.0047279358841478825
2023-01-21 13:33:22.702385: step: 256/527, loss: 0.011433792300522327
2023-01-21 13:33:23.820562: step: 260/527, loss: 8.068084571277723e-05
2023-01-21 13:33:24.948892: step: 264/527, loss: 0.03424587473273277
2023-01-21 13:33:26.046527: step: 268/527, loss: 0.06713536381721497
2023-01-21 13:33:27.143788: step: 272/527, loss: 0.050714682787656784
2023-01-21 13:33:28.245878: step: 276/527, loss: 0.0012892247177660465
2023-01-21 13:33:29.349469: step: 280/527, loss: 0.022847890853881836
2023-01-21 13:33:30.453709: step: 284/527, loss: 0.0016241073608398438
2023-01-21 13:33:31.566958: step: 288/527, loss: 0.002081298967823386
2023-01-21 13:33:32.688214: step: 292/527, loss: 0.00010938645573332906
2023-01-21 13:33:33.791082: step: 296/527, loss: 4.58717331639491e-05
2023-01-21 13:33:34.891579: step: 300/527, loss: -1.220703143189894e-05
2023-01-21 13:33:36.006993: step: 304/527, loss: 0.019287873059511185
2023-01-21 13:33:37.104899: step: 308/527, loss: 0.00047740936861373484
2023-01-21 13:33:38.218300: step: 312/527, loss: 2.584457615739666e-05
2023-01-21 13:33:39.319733: step: 316/527, loss: 0.00015068055654410273
2023-01-21 13:33:40.439923: step: 320/527, loss: -7.62939453125e-06
2023-01-21 13:33:41.547695: step: 324/527, loss: 0.008928490802645683
2023-01-21 13:33:42.666280: step: 328/527, loss: 3.8146970382513246e-07
2023-01-21 13:33:43.769701: step: 332/527, loss: 7.114410254871473e-05
2023-01-21 13:33:44.879475: step: 336/527, loss: 0.037627603858709335
2023-01-21 13:33:45.994650: step: 340/527, loss: 0.0003310203901492059
2023-01-21 13:33:47.095388: step: 344/527, loss: 2.5224686396541074e-05
2023-01-21 13:33:48.191905: step: 348/527, loss: 0.011365699581801891
2023-01-21 13:33:49.295946: step: 352/527, loss: 0.005747413728386164
2023-01-21 13:33:50.416741: step: 356/527, loss: 6.389617919921875e-05
2023-01-21 13:33:51.538523: step: 360/527, loss: 9.250641596736386e-06
2023-01-21 13:33:52.638575: step: 364/527, loss: 8.20159912109375e-05
2023-01-21 13:33:53.733635: step: 368/527, loss: 1.006126512947958e-05
2023-01-21 13:33:54.856197: step: 372/527, loss: 0.0001655578671488911
2023-01-21 13:33:55.970105: step: 376/527, loss: 0.001965808914974332
2023-01-21 13:33:57.079969: step: 380/527, loss: 0.0002801895316224545
2023-01-21 13:33:58.213553: step: 384/527, loss: 0.0008577347034588456
2023-01-21 13:33:59.313363: step: 388/527, loss: 7.400512549793348e-05
2023-01-21 13:34:00.446602: step: 392/527, loss: 0.07707786560058594
2023-01-21 13:34:01.583923: step: 396/527, loss: 9.72747802734375e-05
2023-01-21 13:34:02.700792: step: 400/527, loss: -1.7166138377433526e-06
2023-01-21 13:34:03.794774: step: 404/527, loss: -1.001357759378152e-06
2023-01-21 13:34:04.898241: step: 408/527, loss: 4.491805884754285e-05
2023-01-21 13:34:06.069311: step: 412/527, loss: 0.009221792221069336
2023-01-21 13:34:07.178369: step: 416/527, loss: 0.002175617264583707
2023-01-21 13:34:08.275353: step: 420/527, loss: 6.65664701955393e-05
2023-01-21 13:34:09.376782: step: 424/527, loss: 0.0028001891914755106
2023-01-21 13:34:10.497351: step: 428/527, loss: 0.0042473794892430305
2023-01-21 13:34:11.622242: step: 432/527, loss: 0.00014686585927847773
2023-01-21 13:34:12.753276: step: 436/527, loss: 0.072443388402462
2023-01-21 13:34:13.861660: step: 440/527, loss: 8.792877633823082e-05
2023-01-21 13:34:14.987163: step: 444/527, loss: 0.08109913021326065
2023-01-21 13:34:16.114387: step: 448/527, loss: 0.0019282341236248612
2023-01-21 13:34:17.192245: step: 452/527, loss: 0.0811772346496582
2023-01-21 13:34:18.309321: step: 456/527, loss: 0.009087467566132545
2023-01-21 13:34:19.428404: step: 460/527, loss: 0.0022134780883789062
2023-01-21 13:34:20.547149: step: 464/527, loss: 0.005785178858786821
2023-01-21 13:34:21.683774: step: 468/527, loss: 0.0011878968216478825
2023-01-21 13:34:22.824854: step: 472/527, loss: 0.00027332306490279734
2023-01-21 13:34:23.943692: step: 476/527, loss: 0.0026000975631177425
2023-01-21 13:34:25.044327: step: 480/527, loss: 0.01197595614939928
2023-01-21 13:34:26.151277: step: 484/527, loss: 0.006413173861801624
2023-01-21 13:34:27.240241: step: 488/527, loss: 0.0011042595142498612
2023-01-21 13:34:28.325202: step: 492/527, loss: 0.0049867150373756886
2023-01-21 13:34:29.470922: step: 496/527, loss: 9.779930405784398e-05
2023-01-21 13:34:30.593677: step: 500/527, loss: 0.001121091889217496
2023-01-21 13:34:31.685538: step: 504/527, loss: 0.00024118424335028976
2023-01-21 13:34:32.806824: step: 508/527, loss: 8.678435733600054e-06
2023-01-21 13:34:33.952892: step: 512/527, loss: 0.013912391848862171
2023-01-21 13:34:35.069576: step: 516/527, loss: 0.0006174087757244706
2023-01-21 13:34:36.193139: step: 520/527, loss: 6.189346458995715e-05
2023-01-21 13:34:37.301721: step: 524/527, loss: 2.212524486822076e-05
2023-01-21 13:34:38.448428: step: 528/527, loss: 0.00011215210543014109
2023-01-21 13:34:39.560947: step: 532/527, loss: 2.2554397219209932e-05
2023-01-21 13:34:40.683348: step: 536/527, loss: 0.007109927944839001
2023-01-21 13:34:41.793498: step: 540/527, loss: 0.005600738804787397
2023-01-21 13:34:42.908335: step: 544/527, loss: 0.006458759307861328
2023-01-21 13:34:44.052766: step: 548/527, loss: 0.02709989622235298
2023-01-21 13:34:45.169135: step: 552/527, loss: 0.00013465881056617945
2023-01-21 13:34:46.283538: step: 556/527, loss: 0.0007377624860964715
2023-01-21 13:34:47.408315: step: 560/527, loss: 0.0046651363372802734
2023-01-21 13:34:48.504571: step: 564/527, loss: 4.3201445805607364e-05
2023-01-21 13:34:49.619124: step: 568/527, loss: 0.0001491546572651714
2023-01-21 13:34:50.717332: step: 572/527, loss: 4.0483475459041074e-05
2023-01-21 13:34:51.830933: step: 576/527, loss: 0.0001430511474609375
2023-01-21 13:34:52.948737: step: 580/527, loss: 0.0030183792114257812
2023-01-21 13:34:54.085947: step: 584/527, loss: 0.012265205383300781
2023-01-21 13:34:55.167205: step: 588/527, loss: 0.03284502029418945
2023-01-21 13:34:56.296225: step: 592/527, loss: 0.007442665286362171
2023-01-21 13:34:57.410707: step: 596/527, loss: 0.0010752677917480469
2023-01-21 13:34:58.518995: step: 600/527, loss: -4.863739377469756e-06
2023-01-21 13:34:59.652054: step: 604/527, loss: 0.006377029232680798
2023-01-21 13:35:00.762826: step: 608/527, loss: 0.0013196945656090975
2023-01-21 13:35:01.887538: step: 612/527, loss: 0.00010347366333007812
2023-01-21 13:35:03.003013: step: 616/527, loss: 3.24249267578125e-05
2023-01-21 13:35:04.098042: step: 620/527, loss: 0.019945908337831497
2023-01-21 13:35:05.200597: step: 624/527, loss: 0.00022659300884697586
2023-01-21 13:35:06.314776: step: 628/527, loss: 0.09983577579259872
2023-01-21 13:35:07.428783: step: 632/527, loss: 0.004530525766313076
2023-01-21 13:35:08.533495: step: 636/527, loss: 0.00013809204392600805
2023-01-21 13:35:09.655778: step: 640/527, loss: 0.35550469160079956
2023-01-21 13:35:10.768641: step: 644/527, loss: 8.459090895485133e-05
2023-01-21 13:35:11.871105: step: 648/527, loss: 0.05952663719654083
2023-01-21 13:35:12.994699: step: 652/527, loss: 0.0003520965692587197
2023-01-21 13:35:14.075277: step: 656/527, loss: 0.0001398086460540071
2023-01-21 13:35:15.184098: step: 660/527, loss: 4.768372036778601e-06
2023-01-21 13:35:16.329162: step: 664/527, loss: 0.005389214027673006
2023-01-21 13:35:17.456998: step: 668/527, loss: 2.021789623540826e-05
2023-01-21 13:35:18.566889: step: 672/527, loss: 0.00022754669771529734
2023-01-21 13:35:19.665518: step: 676/527, loss: 0.0014707566006109118
2023-01-21 13:35:20.775237: step: 680/527, loss: 0.006165695376694202
2023-01-21 13:35:21.892961: step: 684/527, loss: 0.000705337559338659
2023-01-21 13:35:22.995594: step: 688/527, loss: -3.14712519866589e-06
2023-01-21 13:35:24.107769: step: 692/527, loss: 1.2397766795402276e-06
2023-01-21 13:35:25.207317: step: 696/527, loss: 0.0009949684608727694
2023-01-21 13:35:26.294191: step: 700/527, loss: 0.0005684852949343622
2023-01-21 13:35:27.473246: step: 704/527, loss: 9.422302537132055e-05
2023-01-21 13:35:28.588572: step: 708/527, loss: 2.441406286379788e-05
2023-01-21 13:35:29.720426: step: 712/527, loss: 0.7490178942680359
2023-01-21 13:35:30.859767: step: 716/527, loss: 9.250640687241685e-06
2023-01-21 13:35:31.999521: step: 720/527, loss: 0.0010239601833745837
2023-01-21 13:35:33.097880: step: 724/527, loss: 0.00017337797908112407
2023-01-21 13:35:34.196908: step: 728/527, loss: 0.0012812615605071187
2023-01-21 13:35:35.345119: step: 732/527, loss: 3.6811830796068534e-05
2023-01-21 13:35:36.471288: step: 736/527, loss: 0.0006567001109942794
2023-01-21 13:35:37.589554: step: 740/527, loss: 7.724762326688506e-06
2023-01-21 13:35:38.728989: step: 744/527, loss: 0.0004010200500488281
2023-01-21 13:35:39.861116: step: 748/527, loss: 0.00028896331787109375
2023-01-21 13:35:40.950682: step: 752/527, loss: 0.0002511024649720639
2023-01-21 13:35:42.069028: step: 756/527, loss: 3.9577484130859375e-05
2023-01-21 13:35:43.177541: step: 760/527, loss: 2.3365020751953125e-05
2023-01-21 13:35:44.269583: step: 764/527, loss: 8.02993745310232e-05
2023-01-21 13:35:45.390326: step: 768/527, loss: 5.054474058852065e-06
2023-01-21 13:35:46.516807: step: 772/527, loss: 0.0001241683930857107
2023-01-21 13:35:47.651745: step: 776/527, loss: 0.6181806921958923
2023-01-21 13:35:48.785404: step: 780/527, loss: 0.0011331558926030993
2023-01-21 13:35:49.885759: step: 784/527, loss: 0.0010840415488928556
2023-01-21 13:35:51.027339: step: 788/527, loss: 0.00508227339014411
2023-01-21 13:35:52.148823: step: 792/527, loss: 0.045961376279592514
2023-01-21 13:35:53.259792: step: 796/527, loss: 0.00026483534020371735
2023-01-21 13:35:54.379240: step: 800/527, loss: 0.02138996124267578
2023-01-21 13:35:55.535202: step: 804/527, loss: 0.0574214905500412
2023-01-21 13:35:56.640670: step: 808/527, loss: 0.004658699035644531
2023-01-21 13:35:57.732806: step: 812/527, loss: 0.0013451576232910156
2023-01-21 13:35:58.817954: step: 816/527, loss: 0.0007417678716592491
2023-01-21 13:35:59.946582: step: 820/527, loss: 0.00014381408982444555
2023-01-21 13:36:01.066809: step: 824/527, loss: 0.00038967133150435984
2023-01-21 13:36:02.174559: step: 828/527, loss: 0.0013612747425213456
2023-01-21 13:36:03.343447: step: 832/527, loss: 0.00958862341940403
2023-01-21 13:36:04.454844: step: 836/527, loss: 0.005013084504753351
2023-01-21 13:36:05.549072: step: 840/527, loss: 0.0001201629638671875
2023-01-21 13:36:06.649366: step: 844/527, loss: 0.009236717596650124
2023-01-21 13:36:07.775058: step: 848/527, loss: 1.201629675051663e-05
2023-01-21 13:36:08.882022: step: 852/527, loss: 0.009479904547333717
2023-01-21 13:36:09.959394: step: 856/527, loss: -8.296966370835435e-06
2023-01-21 13:36:11.069384: step: 860/527, loss: 7.23838820704259e-05
2023-01-21 13:36:12.185485: step: 864/527, loss: 9.4366077973973e-05
2023-01-21 13:36:13.315728: step: 868/527, loss: 0.00796804390847683
2023-01-21 13:36:14.411444: step: 872/527, loss: 4.615783836925402e-05
2023-01-21 13:36:15.545092: step: 876/527, loss: 0.0001146316499216482
2023-01-21 13:36:16.673059: step: 880/527, loss: 0.00045642853365279734
2023-01-21 13:36:17.779938: step: 884/527, loss: 0.07573939114809036
2023-01-21 13:36:18.887986: step: 888/527, loss: 0.0003253936883993447
2023-01-21 13:36:20.035270: step: 892/527, loss: 0.015353680588304996
2023-01-21 13:36:21.171568: step: 896/527, loss: 0.045088961720466614
2023-01-21 13:36:22.297178: step: 900/527, loss: 0.002724266145378351
2023-01-21 13:36:23.412323: step: 904/527, loss: 0.00017547607421875
2023-01-21 13:36:24.512848: step: 908/527, loss: 4.849433753406629e-05
2023-01-21 13:36:25.637608: step: 912/527, loss: 0.02476043812930584
2023-01-21 13:36:26.750940: step: 916/527, loss: 0.009362602606415749
2023-01-21 13:36:27.875852: step: 920/527, loss: 0.0013145447010174394
2023-01-21 13:36:28.997080: step: 924/527, loss: 0.00038332940312102437
2023-01-21 13:36:30.154719: step: 928/527, loss: 0.004160690121352673
2023-01-21 13:36:31.266696: step: 932/527, loss: 0.4128970205783844
2023-01-21 13:36:32.389539: step: 936/527, loss: 0.0025076866149902344
2023-01-21 13:36:33.512148: step: 940/527, loss: 2.9563905172835803e-06
2023-01-21 13:36:34.654431: step: 944/527, loss: 2.6082992917508818e-05
2023-01-21 13:36:35.743619: step: 948/527, loss: 0.002614307450130582
2023-01-21 13:36:36.852553: step: 952/527, loss: 0.0018732547760009766
2023-01-21 13:36:37.991233: step: 956/527, loss: 0.006592655088752508
2023-01-21 13:36:39.097619: step: 960/527, loss: 0.0005663871997967362
2023-01-21 13:36:40.190040: step: 964/527, loss: 6.198883056640625e-05
2023-01-21 13:36:41.305847: step: 968/527, loss: 0.05710233375430107
2023-01-21 13:36:42.404188: step: 972/527, loss: 0.003173637669533491
2023-01-21 13:36:43.541022: step: 976/527, loss: 0.009129524230957031
2023-01-21 13:36:44.678140: step: 980/527, loss: 0.0004795074346475303
2023-01-21 13:36:45.791565: step: 984/527, loss: 6.66618361719884e-05
2023-01-21 13:36:46.901801: step: 988/527, loss: 1.3256073543743696e-05
2023-01-21 13:36:48.030424: step: 992/527, loss: 3.347396705066785e-05
2023-01-21 13:36:49.151304: step: 996/527, loss: 0.0008980275015346706
2023-01-21 13:36:50.256611: step: 1000/527, loss: 0.0008060455438680947
2023-01-21 13:36:51.409037: step: 1004/527, loss: 1.2063979738741182e-05
2023-01-21 13:36:52.513596: step: 1008/527, loss: 0.0016671180492267013
2023-01-21 13:36:53.630360: step: 1012/527, loss: 0.0011191368103027344
2023-01-21 13:36:54.747116: step: 1016/527, loss: 0.01555933989584446
2023-01-21 13:36:55.845300: step: 1020/527, loss: 3.852844383800402e-05
2023-01-21 13:36:56.942093: step: 1024/527, loss: 0.0727960616350174
2023-01-21 13:36:58.059886: step: 1028/527, loss: 0.03388824686408043
2023-01-21 13:36:59.180998: step: 1032/527, loss: 0.010137557983398438
2023-01-21 13:37:00.301424: step: 1036/527, loss: 0.00033054352388717234
2023-01-21 13:37:01.432690: step: 1040/527, loss: 0.0014032364124432206
2023-01-21 13:37:02.546304: step: 1044/527, loss: 0.008458137512207031
2023-01-21 13:37:03.675372: step: 1048/527, loss: 0.01613616943359375
2023-01-21 13:37:04.746642: step: 1052/527, loss: 2.899169885495212e-05
2023-01-21 13:37:05.911552: step: 1056/527, loss: 0.0011610031360760331
2023-01-21 13:37:07.048759: step: 1060/527, loss: 0.0020305633079260588
2023-01-21 13:37:08.186745: step: 1064/527, loss: 0.00011873245239257812
2023-01-21 13:37:09.303458: step: 1068/527, loss: 2.555847095209174e-05
2023-01-21 13:37:10.403348: step: 1072/527, loss: 0.0011371612781658769
2023-01-21 13:37:11.502370: step: 1076/527, loss: 0.035219814628362656
2023-01-21 13:37:12.615965: step: 1080/527, loss: 8.58306884765625e-06
2023-01-21 13:37:13.761709: step: 1084/527, loss: 0.003634643740952015
2023-01-21 13:37:14.875095: step: 1088/527, loss: 0.012740135192871094
2023-01-21 13:37:15.956811: step: 1092/527, loss: 0.026059437543153763
2023-01-21 13:37:17.064791: step: 1096/527, loss: 0.00043926239595748484
2023-01-21 13:37:18.159281: step: 1100/527, loss: 1.9073486328125e-06
2023-01-21 13:37:19.267277: step: 1104/527, loss: 0.0035418986808508635
2023-01-21 13:37:20.403335: step: 1108/527, loss: 6.4373016357421875e-06
2023-01-21 13:37:21.529669: step: 1112/527, loss: 0.0604364387691021
2023-01-21 13:37:22.633148: step: 1116/527, loss: 0.0003274917835369706
2023-01-21 13:37:23.735348: step: 1120/527, loss: 0.0017480850219726562
2023-01-21 13:37:24.860549: step: 1124/527, loss: -1.7261503671761602e-05
2023-01-21 13:37:25.983384: step: 1128/527, loss: 6.961823146411916e-06
2023-01-21 13:37:27.105609: step: 1132/527, loss: 0.024448012933135033
2023-01-21 13:37:28.217609: step: 1136/527, loss: 9.880065044853836e-05
2023-01-21 13:37:29.372917: step: 1140/527, loss: 0.009129619225859642
2023-01-21 13:37:30.479606: step: 1144/527, loss: 0.001784515450708568
2023-01-21 13:37:31.625096: step: 1148/527, loss: -1.144409225162235e-06
2023-01-21 13:37:32.751385: step: 1152/527, loss: 1.1348724001436494e-05
2023-01-21 13:37:33.882081: step: 1156/527, loss: 0.00376129150390625
2023-01-21 13:37:34.994026: step: 1160/527, loss: 0.0003010749933309853
2023-01-21 13:37:36.103796: step: 1164/527, loss: 0.0007301330333575606
2023-01-21 13:37:37.201248: step: 1168/527, loss: 0.001605224679224193
2023-01-21 13:37:38.336032: step: 1172/527, loss: 2.9087066195643274e-06
2023-01-21 13:37:39.448403: step: 1176/527, loss: 0.0001768112269928679
2023-01-21 13:37:40.572067: step: 1180/527, loss: 0.0001123428373830393
2023-01-21 13:37:41.694930: step: 1184/527, loss: 0.18721266090869904
2023-01-21 13:37:42.835350: step: 1188/527, loss: 0.0021316527854651213
2023-01-21 13:37:43.929260: step: 1192/527, loss: 0.04095039516687393
2023-01-21 13:37:45.046213: step: 1196/527, loss: 3.62396240234375e-05
2023-01-21 13:37:46.180142: step: 1200/527, loss: 0.008442497812211514
2023-01-21 13:37:47.275896: step: 1204/527, loss: 0.00034561159554868937
2023-01-21 13:37:48.427675: step: 1208/527, loss: 0.0002574920654296875
2023-01-21 13:37:49.525980: step: 1212/527, loss: 4.4536594941746444e-05
2023-01-21 13:37:50.677384: step: 1216/527, loss: 0.0007654189830645919
2023-01-21 13:37:51.789134: step: 1220/527, loss: 7.581710815429688e-05
2023-01-21 13:37:52.900430: step: 1224/527, loss: 0.0023818970657885075
2023-01-21 13:37:54.003103: step: 1228/527, loss: 0.023397158831357956
2023-01-21 13:37:55.102372: step: 1232/527, loss: 0.00258560199290514
2023-01-21 13:37:56.214317: step: 1236/527, loss: 0.020354939624667168
2023-01-21 13:37:57.337650: step: 1240/527, loss: 0.01276092603802681
2023-01-21 13:37:58.440061: step: 1244/527, loss: 0.0013586045242846012
2023-01-21 13:37:59.538450: step: 1248/527, loss: 1.1444091796875e-05
2023-01-21 13:38:00.637640: step: 1252/527, loss: 7.095336331985891e-05
2023-01-21 13:38:01.756611: step: 1256/527, loss: 0.026103973388671875
2023-01-21 13:38:02.879901: step: 1260/527, loss: 2.2602080207434483e-05
2023-01-21 13:38:03.998951: step: 1264/527, loss: 0.0014027596917003393
2023-01-21 13:38:05.127460: step: 1268/527, loss: 0.00012235641770530492
2023-01-21 13:38:06.248545: step: 1272/527, loss: 0.011886787600815296
2023-01-21 13:38:07.351623: step: 1276/527, loss: 0.0004477500915527344
2023-01-21 13:38:08.470025: step: 1280/527, loss: 9.231567673850805e-05
2023-01-21 13:38:09.559924: step: 1284/527, loss: 3.261566234868951e-05
2023-01-21 13:38:10.656305: step: 1288/527, loss: 0.004856491461396217
2023-01-21 13:38:11.763591: step: 1292/527, loss: 1.773834264895413e-05
2023-01-21 13:38:12.904357: step: 1296/527, loss: 5.53131121705519e-06
2023-01-21 13:38:14.015597: step: 1300/527, loss: 0.0018771172035485506
2023-01-21 13:38:15.135082: step: 1304/527, loss: -3.957748504035408e-06
2023-01-21 13:38:16.295662: step: 1308/527, loss: 0.0009178638574667275
2023-01-21 13:38:17.378002: step: 1312/527, loss: 0.0008335114107467234
2023-01-21 13:38:18.495673: step: 1316/527, loss: 2.746581958490424e-05
2023-01-21 13:38:19.597727: step: 1320/527, loss: 0.0013227462768554688
2023-01-21 13:38:20.706216: step: 1324/527, loss: 9.5367431640625e-07
2023-01-21 13:38:21.850037: step: 1328/527, loss: 0.0010253905784338713
2023-01-21 13:38:22.944147: step: 1332/527, loss: 2.57492069977161e-06
2023-01-21 13:38:24.038247: step: 1336/527, loss: 0.008300495333969593
2023-01-21 13:38:25.143273: step: 1340/527, loss: 4.0626528061693534e-05
2023-01-21 13:38:26.229693: step: 1344/527, loss: 0.00031108857365325093
2023-01-21 13:38:27.324095: step: 1348/527, loss: 2.3651125957258046e-05
2023-01-21 13:38:28.425813: step: 1352/527, loss: 0.029747294262051582
2023-01-21 13:38:29.543785: step: 1356/527, loss: 4.100799742445815e-06
2023-01-21 13:38:30.646528: step: 1360/527, loss: 0.0008672713884152472
2023-01-21 13:38:31.759390: step: 1364/527, loss: 0.22238807380199432
2023-01-21 13:38:32.857211: step: 1368/527, loss: 0.325967013835907
2023-01-21 13:38:33.994170: step: 1372/527, loss: 4.509836344368523e-06
2023-01-21 13:38:35.093333: step: 1376/527, loss: 0.0001333236723439768
2023-01-21 13:38:36.186425: step: 1380/527, loss: 1.544952465337701e-05
2023-01-21 13:38:37.317628: step: 1384/527, loss: 4.19616708313697e-06
2023-01-21 13:38:38.431863: step: 1388/527, loss: 3.013610876223538e-05
2023-01-21 13:38:39.536607: step: 1392/527, loss: 0.03813190758228302
2023-01-21 13:38:40.632480: step: 1396/527, loss: 1.9836426872643642e-05
2023-01-21 13:38:41.740993: step: 1400/527, loss: 7.004737562965602e-05
2023-01-21 13:38:42.868110: step: 1404/527, loss: 2.47955313170678e-06
2023-01-21 13:38:44.004423: step: 1408/527, loss: 0.03603959083557129
2023-01-21 13:38:45.108185: step: 1412/527, loss: 2.19821922655683e-05
2023-01-21 13:38:46.190997: step: 1416/527, loss: 1.8596649169921875e-05
2023-01-21 13:38:47.296760: step: 1420/527, loss: 0.00026979445829056203
2023-01-21 13:38:48.409009: step: 1424/527, loss: 0.0006380081176757812
2023-01-21 13:38:49.527233: step: 1428/527, loss: 0.0005802154773846269
2023-01-21 13:38:50.691963: step: 1432/527, loss: 0.0001510620058979839
2023-01-21 13:38:51.805050: step: 1436/527, loss: 0.004744911566376686
2023-01-21 13:38:52.911340: step: 1440/527, loss: 3.519058373058215e-05
2023-01-21 13:38:54.042211: step: 1444/527, loss: 2.784729076665826e-05
2023-01-21 13:38:55.139865: step: 1448/527, loss: 1.049041748046875e-05
2023-01-21 13:38:56.266632: step: 1452/527, loss: -7.62939453125e-06
2023-01-21 13:38:57.389956: step: 1456/527, loss: 0.020812224596738815
2023-01-21 13:38:58.509373: step: 1460/527, loss: 0.034818269312381744
2023-01-21 13:38:59.632917: step: 1464/527, loss: 1.716613724056515e-06
2023-01-21 13:39:00.760898: step: 1468/527, loss: 0.0001184463471872732
2023-01-21 13:39:01.926530: step: 1472/527, loss: 0.0003986358642578125
2023-01-21 13:39:03.004011: step: 1476/527, loss: 1.3542176020564511e-05
2023-01-21 13:39:04.118952: step: 1480/527, loss: 2.6035308110294864e-05
2023-01-21 13:39:05.216520: step: 1484/527, loss: 0.00019655228243209422
2023-01-21 13:39:06.314874: step: 1488/527, loss: 0.016224002465605736
2023-01-21 13:39:07.424633: step: 1492/527, loss: 0.0003681182861328125
2023-01-21 13:39:08.553079: step: 1496/527, loss: 0.000533676182385534
2023-01-21 13:39:09.654719: step: 1500/527, loss: 0.02605123445391655
2023-01-21 13:39:10.782572: step: 1504/527, loss: 1.4686585018353071e-05
2023-01-21 13:39:11.911652: step: 1508/527, loss: 0.007518958766013384
2023-01-21 13:39:13.003230: step: 1512/527, loss: 0.0003223419189453125
2023-01-21 13:39:14.130163: step: 1516/527, loss: 0.00869207363575697
2023-01-21 13:39:15.232405: step: 1520/527, loss: 0.01059265062212944
2023-01-21 13:39:16.351634: step: 1524/527, loss: 6.122589547885582e-05
2023-01-21 13:39:17.445969: step: 1528/527, loss: 5.34057608092553e-06
2023-01-21 13:39:18.575285: step: 1532/527, loss: 0.0017091037007048726
2023-01-21 13:39:19.701258: step: 1536/527, loss: 7.686614844715223e-05
2023-01-21 13:39:20.810720: step: 1540/527, loss: 0.0002845764101948589
2023-01-21 13:39:21.942421: step: 1544/527, loss: 0.03704690933227539
2023-01-21 13:39:23.057967: step: 1548/527, loss: 0.010331916622817516
2023-01-21 13:39:24.170446: step: 1552/527, loss: -4.24384961661417e-06
2023-01-21 13:39:25.257669: step: 1556/527, loss: 0.01786823198199272
2023-01-21 13:39:26.359271: step: 1560/527, loss: 6.50405854685232e-05
2023-01-21 13:39:27.502943: step: 1564/527, loss: 2.47955322265625e-05
2023-01-21 13:39:28.592801: step: 1568/527, loss: 2.8610230629055877e-07
2023-01-21 13:39:29.673893: step: 1572/527, loss: 0.011174154467880726
2023-01-21 13:39:30.757083: step: 1576/527, loss: 0.012739181518554688
2023-01-21 13:39:31.865070: step: 1580/527, loss: 5.626678102998994e-06
2023-01-21 13:39:32.989869: step: 1584/527, loss: 0.060941699892282486
2023-01-21 13:39:34.111240: step: 1588/527, loss: 3.814697265625e-06
2023-01-21 13:39:35.233600: step: 1592/527, loss: 0.0009203911758959293
2023-01-21 13:39:36.322030: step: 1596/527, loss: 1.163482647825731e-05
2023-01-21 13:39:37.471868: step: 1600/527, loss: 0.5229610204696655
2023-01-21 13:39:38.581447: step: 1604/527, loss: 2.098083541568485e-06
2023-01-21 13:39:39.701625: step: 1608/527, loss: 3.795623706537299e-05
2023-01-21 13:39:40.814629: step: 1612/527, loss: 0.0004807472287211567
2023-01-21 13:39:41.931650: step: 1616/527, loss: 4.00543194700731e-06
2023-01-21 13:39:43.043251: step: 1620/527, loss: 0.002050018170848489
2023-01-21 13:39:44.136876: step: 1624/527, loss: -4.95910626341356e-06
2023-01-21 13:39:45.291810: step: 1628/527, loss: 0.0001560211239848286
2023-01-21 13:39:46.403442: step: 1632/527, loss: 0.021405315026640892
2023-01-21 13:39:47.506995: step: 1636/527, loss: 7.629394076502649e-07
2023-01-21 13:39:48.597542: step: 1640/527, loss: 0.005777645390480757
2023-01-21 13:39:49.698098: step: 1644/527, loss: 0.0004177093505859375
2023-01-21 13:39:50.789131: step: 1648/527, loss: 0.006636237725615501
2023-01-21 13:39:51.900237: step: 1652/527, loss: 2.3180245989351533e-05
2023-01-21 13:39:53.018553: step: 1656/527, loss: 0.0038425445090979338
2023-01-21 13:39:54.121507: step: 1660/527, loss: 0.02945546992123127
2023-01-21 13:39:55.224415: step: 1664/527, loss: 2.28881845032447e-06
2023-01-21 13:39:56.366978: step: 1668/527, loss: 0.0005195617559365928
2023-01-21 13:39:57.458663: step: 1672/527, loss: 2.5081635612878017e-05
2023-01-21 13:39:58.569648: step: 1676/527, loss: 0.0001655578671488911
2023-01-21 13:39:59.683224: step: 1680/527, loss: 0.0026155475061386824
2023-01-21 13:40:00.790163: step: 1684/527, loss: 3.6811830796068534e-05
2023-01-21 13:40:01.899020: step: 1688/527, loss: -9.536779543850571e-08
2023-01-21 13:40:02.995000: step: 1692/527, loss: 9.832382056629285e-05
2023-01-21 13:40:04.091828: step: 1696/527, loss: 0.002795314881950617
2023-01-21 13:40:05.215295: step: 1700/527, loss: 0.00029754638671875
2023-01-21 13:40:06.356900: step: 1704/527, loss: 0.004522895906120539
2023-01-21 13:40:07.428847: step: 1708/527, loss: 8.869172233971767e-06
2023-01-21 13:40:08.517876: step: 1712/527, loss: 0.00041866302490234375
2023-01-21 13:40:09.674525: step: 1716/527, loss: 0.006893634796142578
2023-01-21 13:40:10.771805: step: 1720/527, loss: 0.12302589416503906
2023-01-21 13:40:11.863030: step: 1724/527, loss: 0.0013587952125817537
2023-01-21 13:40:12.977767: step: 1728/527, loss: 0.00021495818509720266
2023-01-21 13:40:14.095754: step: 1732/527, loss: -1.716613724056515e-06
2023-01-21 13:40:15.242842: step: 1736/527, loss: 5.245208740234375e-06
2023-01-21 13:40:16.347733: step: 1740/527, loss: 0.0020215988624840975
2023-01-21 13:40:17.465845: step: 1744/527, loss: 0.05284080654382706
2023-01-21 13:40:18.594154: step: 1748/527, loss: 0.00064935686532408
2023-01-21 13:40:19.702221: step: 1752/527, loss: 0.0012876511318609118
2023-01-21 13:40:20.808244: step: 1756/527, loss: 0.0003529548703227192
2023-01-21 13:40:21.968516: step: 1760/527, loss: 0.009379196912050247
2023-01-21 13:40:23.101015: step: 1764/527, loss: 0.046181678771972656
2023-01-21 13:40:24.219807: step: 1768/527, loss: 0.0025760652497410774
2023-01-21 13:40:25.349875: step: 1772/527, loss: 0.0003704070986714214
2023-01-21 13:40:26.459812: step: 1776/527, loss: 1.716613724056515e-06
2023-01-21 13:40:27.560792: step: 1780/527, loss: 0.00019168853759765625
2023-01-21 13:40:28.645203: step: 1784/527, loss: 0.08904609829187393
2023-01-21 13:40:29.756787: step: 1788/527, loss: 2.8514861696748994e-05
2023-01-21 13:40:30.858989: step: 1792/527, loss: 0.023518182337284088
2023-01-21 13:40:31.973730: step: 1796/527, loss: 0.014324379153549671
2023-01-21 13:40:33.089197: step: 1800/527, loss: -1.7642973944020923e-06
2023-01-21 13:40:34.208560: step: 1804/527, loss: 0.0011955262161791325
2023-01-21 13:40:35.337981: step: 1808/527, loss: 0.0001396179141011089
2023-01-21 13:40:36.462770: step: 1812/527, loss: 1.678466833254788e-05
2023-01-21 13:40:37.553065: step: 1816/527, loss: 0.00011739729961846024
2023-01-21 13:40:38.655150: step: 1820/527, loss: 0.02970914915204048
2023-01-21 13:40:39.785656: step: 1824/527, loss: 1.4686585018353071e-05
2023-01-21 13:40:40.890639: step: 1828/527, loss: 0.022895528003573418
2023-01-21 13:40:42.010857: step: 1832/527, loss: 0.0023525238502770662
2023-01-21 13:40:43.124151: step: 1836/527, loss: 0.0007162094116210938
2023-01-21 13:40:44.217959: step: 1840/527, loss: 0.016622615978121758
2023-01-21 13:40:45.353607: step: 1844/527, loss: 0.0005181312444619834
2023-01-21 13:40:46.460077: step: 1848/527, loss: 8.010864803509321e-06
2023-01-21 13:40:47.574940: step: 1852/527, loss: 0.019637491554021835
2023-01-21 13:40:48.689746: step: 1856/527, loss: 0.007338047027587891
2023-01-21 13:40:49.787813: step: 1860/527, loss: 0.0011165142059326172
2023-01-21 13:40:50.936325: step: 1864/527, loss: 0.022028256207704544
2023-01-21 13:40:52.057860: step: 1868/527, loss: 3.662109520519152e-05
2023-01-21 13:40:53.166376: step: 1872/527, loss: 2.86102294921875e-06
2023-01-21 13:40:54.264961: step: 1876/527, loss: 0.00021104812913108617
2023-01-21 13:40:55.365735: step: 1880/527, loss: 6.647109694313258e-05
2023-01-21 13:40:56.490202: step: 1884/527, loss: 0.025273799896240234
2023-01-21 13:40:57.610550: step: 1888/527, loss: 0.000213623046875
2023-01-21 13:40:58.740307: step: 1892/527, loss: 0.3446022868156433
2023-01-21 13:40:59.873251: step: 1896/527, loss: 0.001374244806356728
2023-01-21 13:41:01.016830: step: 1900/527, loss: 3.123283386230469e-05
2023-01-21 13:41:02.138435: step: 1904/527, loss: 0.004478550050407648
2023-01-21 13:41:03.265331: step: 1908/527, loss: 0.014357281848788261
2023-01-21 13:41:04.388643: step: 1912/527, loss: 0.047567371279001236
2023-01-21 13:41:05.504025: step: 1916/527, loss: 0.03014068678021431
2023-01-21 13:41:06.607875: step: 1920/527, loss: 0.0006321906694211066
2023-01-21 13:41:07.700578: step: 1924/527, loss: 2.09808349609375e-05
2023-01-21 13:41:08.851472: step: 1928/527, loss: 1.0013580322265625e-05
2023-01-21 13:41:09.960432: step: 1932/527, loss: 0.0006980896578170359
2023-01-21 13:41:11.047791: step: 1936/527, loss: 4.615783836925402e-05
2023-01-21 13:41:12.137271: step: 1940/527, loss: 0.0006071567768231034
2023-01-21 13:41:13.252262: step: 1944/527, loss: 0.00034227370633743703
2023-01-21 13:41:14.353144: step: 1948/527, loss: 0.04827642813324928
2023-01-21 13:41:15.519237: step: 1952/527, loss: 0.004059219267219305
2023-01-21 13:41:16.658930: step: 1956/527, loss: 0.004584193229675293
2023-01-21 13:41:17.783861: step: 1960/527, loss: 3.070831371587701e-05
2023-01-21 13:41:18.909003: step: 1964/527, loss: 0.0015237807529047132
2023-01-21 13:41:20.011602: step: 1968/527, loss: 0.0026638987474143505
2023-01-21 13:41:21.146021: step: 1972/527, loss: 0.00034313200740143657
2023-01-21 13:41:22.270399: step: 1976/527, loss: 0.016387557610869408
2023-01-21 13:41:23.390192: step: 1980/527, loss: 0.0035765646025538445
2023-01-21 13:41:24.503106: step: 1984/527, loss: 0.00015559197345282882
2023-01-21 13:41:25.573678: step: 1988/527, loss: 0.0016122817760333419
2023-01-21 13:41:26.676575: step: 1992/527, loss: 0.01073303259909153
2023-01-21 13:41:27.784190: step: 1996/527, loss: 0.022158240899443626
2023-01-21 13:41:28.908343: step: 2000/527, loss: 0.029164601117372513
2023-01-21 13:41:29.988345: step: 2004/527, loss: 0.00039505958557128906
2023-01-21 13:41:31.116171: step: 2008/527, loss: 9.460449655307457e-05
2023-01-21 13:41:32.227292: step: 2012/527, loss: 0.0002344131498830393
2023-01-21 13:41:33.338600: step: 2016/527, loss: 0.01551132183521986
2023-01-21 13:41:34.440069: step: 2020/527, loss: 0.0062233926728367805
2023-01-21 13:41:35.537723: step: 2024/527, loss: 0.0006309509626589715
2023-01-21 13:41:36.648346: step: 2028/527, loss: 0.00014123917208053172
2023-01-21 13:41:37.770461: step: 2032/527, loss: 6.384849257301539e-05
2023-01-21 13:41:38.887387: step: 2036/527, loss: 0.002294349716976285
2023-01-21 13:41:39.974222: step: 2040/527, loss: 0.0001321315940003842
2023-01-21 13:41:41.076964: step: 2044/527, loss: 0.016176223754882812
2023-01-21 13:41:42.177241: step: 2048/527, loss: 0.00023927689471747726
2023-01-21 13:41:43.273215: step: 2052/527, loss: 0.01070108450949192
2023-01-21 13:41:44.402788: step: 2056/527, loss: 0.0030249597039073706
2023-01-21 13:41:45.542110: step: 2060/527, loss: 0.0004129410081077367
2023-01-21 13:41:46.662444: step: 2064/527, loss: 3.871917579090223e-05
2023-01-21 13:41:47.805315: step: 2068/527, loss: 0.0024806975852698088
2023-01-21 13:41:48.907719: step: 2072/527, loss: 0.023093605414032936
2023-01-21 13:41:50.002999: step: 2076/527, loss: 0.0001046180768753402
2023-01-21 13:41:51.121419: step: 2080/527, loss: 0.0001642227143747732
2023-01-21 13:41:52.223817: step: 2084/527, loss: 7.224082946777344e-05
2023-01-21 13:41:53.341973: step: 2088/527, loss: 0.13199271261692047
2023-01-21 13:41:54.490594: step: 2092/527, loss: 0.013424873352050781
2023-01-21 13:41:55.588717: step: 2096/527, loss: 0.0001222610444528982
2023-01-21 13:41:56.696402: step: 2100/527, loss: 0.0011671066749840975
2023-01-21 13:41:57.781258: step: 2104/527, loss: 0.0020716667640954256
2023-01-21 13:41:58.883470: step: 2108/527, loss: 0.0015638350741937757
==================================================
Loss: 0.015
--------------------
Dev: {'event': {'p': 0.6049129989764586, 'r': 0.7869507323568575, 'f1': 0.6840277777777778}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29}
Test: {'event': {'p': 0.6380996739636703, 'r': 0.7828571428571428, 'f1': 0.7031049525275853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29}
Chinese: {'event': {'p': 0.5393258426966292, 'r': 0.8888888888888888, 'f1': 0.6713286713286712}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29}
Korean: {'event': {'p': 0.7, 'r': 0.4444444444444444, 'f1': 0.5436893203883495}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29}
Russian: {'event': {'p': 0.4473684210526316, 'r': 0.4722222222222222, 'f1': 0.4594594594594595}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29}
==================================================
Current best result:
--------------------
Eng Dev for Chinese: {'event': {'p': 0.6241758241758242, 'r': 0.7563249001331558, 'f1': 0.6839253461770018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Eng Test for Chinese: {'event': {'p': 0.6433059449009183, 'r': 0.7605714285714286, 'f1': 0.6970411102382822}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
Sample Chinese: {'event': {'p': 0.5949367088607594, 'r': 0.8703703703703703, 'f1': 0.706766917293233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5}
--------------------
Eng Dev for Korean: {'event': {'p': 0.6063157894736843, 'r': 0.7669773635153129, 'f1': 0.6772486772486773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Eng Test for Korean: {'event': {'p': 0.6427238805970149, 'r': 0.7874285714285715, 'f1': 0.7077555213148434}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
Sample Korean: {'event': {'p': 0.6792452830188679, 'r': 0.5714285714285714, 'f1': 0.6206896551724137}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24}
--------------------
Eng Dev for Russian: {'event': {'p': 0.6400462962962963, 'r': 0.7363515312916112, 'f1': 0.6848297213622292}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Eng Test for Russian: {'event': {'p': 0.6463168516649849, 'r': 0.732, 'f1': 0.6864951768488746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}
Sample Russian: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3}