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 08:38:13.081781: step: 4/529, loss: 12.749869346618652 2023-01-21 08:38:14.297009: step: 8/529, loss: 15.06234359741211 2023-01-21 08:38:15.474537: step: 12/529, loss: 3.739328384399414 2023-01-21 08:38:16.643842: step: 16/529, loss: 3.6788172721862793 2023-01-21 08:38:17.842853: step: 20/529, loss: 4.1812567710876465 2023-01-21 08:38:18.994746: step: 24/529, loss: 3.3902077674865723 2023-01-21 08:38:20.186324: step: 28/529, loss: 12.677481651306152 2023-01-21 08:38:21.367818: step: 32/529, loss: 3.1846489906311035 2023-01-21 08:38:22.541411: step: 36/529, loss: 18.274188995361328 2023-01-21 08:38:23.709519: step: 40/529, loss: 2.864719867706299 2023-01-21 08:38:24.912553: step: 44/529, loss: 19.781169891357422 2023-01-21 08:38:26.109398: step: 48/529, loss: 8.725495338439941 2023-01-21 08:38:27.271640: step: 52/529, loss: 18.469144821166992 2023-01-21 08:38:28.434759: step: 56/529, loss: 2.492431163787842 2023-01-21 08:38:29.624532: step: 60/529, loss: 13.137426376342773 2023-01-21 08:38:30.830022: step: 64/529, loss: 18.803770065307617 2023-01-21 08:38:31.997623: step: 68/529, loss: 17.356889724731445 2023-01-21 08:38:33.188824: step: 72/529, loss: 6.832083225250244 2023-01-21 08:38:34.363211: step: 76/529, loss: 12.855040550231934 2023-01-21 08:38:35.540514: step: 80/529, loss: 4.13387393951416 2023-01-21 08:38:36.738253: step: 84/529, loss: 8.929308891296387 2023-01-21 08:38:37.923288: step: 88/529, loss: 8.180465698242188 2023-01-21 08:38:39.118000: step: 92/529, loss: 3.007103204727173 2023-01-21 08:38:40.283257: step: 96/529, loss: 3.807387113571167 2023-01-21 08:38:41.454532: step: 100/529, loss: 22.779010772705078 2023-01-21 08:38:42.689147: step: 104/529, loss: 12.340360641479492 2023-01-21 08:38:43.903596: step: 108/529, loss: 15.410866737365723 2023-01-21 08:38:45.080330: step: 112/529, loss: 20.466880798339844 2023-01-21 08:38:46.222256: step: 116/529, loss: 3.780665159225464 2023-01-21 08:38:47.413822: step: 120/529, loss: 2.9077563285827637 2023-01-21 08:38:48.664743: step: 124/529, loss: 21.12575340270996 2023-01-21 08:38:49.882684: step: 128/529, loss: 18.735261917114258 2023-01-21 08:38:51.091553: step: 132/529, loss: 24.76099395751953 2023-01-21 08:38:52.298975: step: 136/529, loss: 6.642233848571777 2023-01-21 08:38:53.474042: step: 140/529, loss: 7.240755081176758 2023-01-21 08:38:54.678015: step: 144/529, loss: 13.34777545928955 2023-01-21 08:38:55.838529: step: 148/529, loss: 12.940428733825684 2023-01-21 08:38:57.029684: step: 152/529, loss: 32.486122131347656 2023-01-21 08:38:58.207258: step: 156/529, loss: 11.29046630859375 2023-01-21 08:38:59.376279: step: 160/529, loss: 18.595088958740234 2023-01-21 08:39:00.563980: step: 164/529, loss: 22.335182189941406 2023-01-21 08:39:01.741341: step: 168/529, loss: 2.133139133453369 2023-01-21 08:39:02.940876: step: 172/529, loss: 4.195262908935547 2023-01-21 08:39:04.161366: step: 176/529, loss: 6.215989112854004 2023-01-21 08:39:05.368471: step: 180/529, loss: 12.389152526855469 2023-01-21 08:39:06.561302: step: 184/529, loss: 21.475690841674805 2023-01-21 08:39:07.776966: step: 188/529, loss: 6.198763847351074 2023-01-21 08:39:09.008452: step: 192/529, loss: 21.22145652770996 2023-01-21 08:39:10.226370: step: 196/529, loss: 4.188440322875977 2023-01-21 08:39:11.418327: step: 200/529, loss: 6.47695255279541 2023-01-21 08:39:12.585117: step: 204/529, loss: 2.6838021278381348 2023-01-21 08:39:13.754662: step: 208/529, loss: 2.681820869445801 2023-01-21 08:39:14.965813: step: 212/529, loss: 3.313990592956543 2023-01-21 08:39:16.201795: step: 216/529, loss: 8.061814308166504 2023-01-21 08:39:17.380018: step: 220/529, loss: 3.0597293376922607 2023-01-21 08:39:18.612336: step: 224/529, loss: 3.2819080352783203 2023-01-21 08:39:19.859329: step: 228/529, loss: 25.673660278320312 2023-01-21 08:39:21.048713: step: 232/529, loss: 17.186492919921875 2023-01-21 08:39:22.245788: step: 236/529, loss: 24.30300521850586 2023-01-21 08:39:23.462899: step: 240/529, loss: 25.181787490844727 2023-01-21 08:39:24.648063: step: 244/529, loss: 3.3175296783447266 2023-01-21 08:39:25.860879: step: 248/529, loss: 16.950958251953125 2023-01-21 08:39:27.046625: step: 252/529, loss: 16.225004196166992 2023-01-21 08:39:28.208006: step: 256/529, loss: 2.3727657794952393 2023-01-21 08:39:29.373625: step: 260/529, loss: 3.617187976837158 2023-01-21 08:39:30.620872: step: 264/529, loss: 12.824539184570312 2023-01-21 08:39:31.821681: step: 268/529, loss: 17.41595458984375 2023-01-21 08:39:33.074004: step: 272/529, loss: 3.492107629776001 2023-01-21 08:39:34.293843: step: 276/529, loss: 7.402348518371582 2023-01-21 08:39:35.462954: step: 280/529, loss: 2.952519655227661 2023-01-21 08:39:36.670059: step: 284/529, loss: 8.797836303710938 2023-01-21 08:39:37.898979: step: 288/529, loss: 14.972488403320312 2023-01-21 08:39:39.089293: step: 292/529, loss: 4.770545482635498 2023-01-21 08:39:40.263499: step: 296/529, loss: 9.256494522094727 2023-01-21 08:39:41.445264: step: 300/529, loss: 3.4210028648376465 2023-01-21 08:39:42.682314: step: 304/529, loss: 2.9110093116760254 2023-01-21 08:39:43.893546: step: 308/529, loss: 3.7728018760681152 2023-01-21 08:39:45.109941: step: 312/529, loss: 3.450716733932495 2023-01-21 08:39:46.299341: step: 316/529, loss: 12.856974601745605 2023-01-21 08:39:47.492096: step: 320/529, loss: 11.36719799041748 2023-01-21 08:39:48.731562: step: 324/529, loss: 3.657066583633423 2023-01-21 08:39:49.925957: step: 328/529, loss: 10.196720123291016 2023-01-21 08:39:51.083466: step: 332/529, loss: 4.825366020202637 2023-01-21 08:39:52.298710: step: 336/529, loss: 8.756556510925293 2023-01-21 08:39:53.498805: step: 340/529, loss: 17.305171966552734 2023-01-21 08:39:54.724070: step: 344/529, loss: 30.628881454467773 2023-01-21 08:39:55.896887: step: 348/529, loss: 2.285128593444824 2023-01-21 08:39:57.087644: step: 352/529, loss: 2.856727123260498 2023-01-21 08:39:58.273201: step: 356/529, loss: 15.390984535217285 2023-01-21 08:39:59.436510: step: 360/529, loss: 11.84009075164795 2023-01-21 08:40:00.618845: step: 364/529, loss: 2.512152671813965 2023-01-21 08:40:01.818354: step: 368/529, loss: 11.980447769165039 2023-01-21 08:40:03.041833: step: 372/529, loss: 10.839179039001465 2023-01-21 08:40:04.204168: step: 376/529, loss: 3.08955979347229 2023-01-21 08:40:05.393659: step: 380/529, loss: 21.360130310058594 2023-01-21 08:40:06.556382: step: 384/529, loss: 1.0728583335876465 2023-01-21 08:40:07.749766: step: 388/529, loss: 15.276307106018066 2023-01-21 08:40:08.918929: step: 392/529, loss: 17.153968811035156 2023-01-21 08:40:10.114739: step: 396/529, loss: 1.5041100978851318 2023-01-21 08:40:11.326151: step: 400/529, loss: 3.5005576610565186 2023-01-21 08:40:12.523431: step: 404/529, loss: 11.276963233947754 2023-01-21 08:40:13.728362: step: 408/529, loss: 6.307821273803711 2023-01-21 08:40:14.930668: step: 412/529, loss: 9.759897232055664 2023-01-21 08:40:16.120526: step: 416/529, loss: 10.426053047180176 2023-01-21 08:40:17.287267: step: 420/529, loss: 11.970014572143555 2023-01-21 08:40:18.472969: step: 424/529, loss: 4.747678756713867 2023-01-21 08:40:19.680316: step: 428/529, loss: 12.482945442199707 2023-01-21 08:40:20.883940: step: 432/529, loss: 17.28499984741211 2023-01-21 08:40:22.078780: step: 436/529, loss: 7.931006908416748 2023-01-21 08:40:23.241198: step: 440/529, loss: 4.634056568145752 2023-01-21 08:40:24.413717: step: 444/529, loss: 2.4623448848724365 2023-01-21 08:40:25.654140: step: 448/529, loss: 3.8997740745544434 2023-01-21 08:40:26.864279: step: 452/529, loss: 10.992735862731934 2023-01-21 08:40:28.072958: step: 456/529, loss: 3.3993771076202393 2023-01-21 08:40:29.266551: step: 460/529, loss: 2.355959892272949 2023-01-21 08:40:30.463957: step: 464/529, loss: 17.227130889892578 2023-01-21 08:40:31.687426: step: 468/529, loss: 3.8131279945373535 2023-01-21 08:40:32.883666: step: 472/529, loss: 1.347888469696045 2023-01-21 08:40:34.070073: step: 476/529, loss: 10.548364639282227 2023-01-21 08:40:35.261416: step: 480/529, loss: 2.078490972518921 2023-01-21 08:40:36.477107: step: 484/529, loss: 7.301095962524414 2023-01-21 08:40:37.671482: step: 488/529, loss: 12.59904670715332 2023-01-21 08:40:38.881561: step: 492/529, loss: 13.853641510009766 2023-01-21 08:40:40.112899: step: 496/529, loss: 5.629919052124023 2023-01-21 08:40:41.373023: step: 500/529, loss: 5.239670276641846 2023-01-21 08:40:42.530304: step: 504/529, loss: 4.342263698577881 2023-01-21 08:40:43.700201: step: 508/529, loss: 1.4194350242614746 2023-01-21 08:40:44.904116: step: 512/529, loss: 4.309412002563477 2023-01-21 08:40:46.112499: step: 516/529, loss: 6.315372943878174 2023-01-21 08:40:47.350960: step: 520/529, loss: 1.605266809463501 2023-01-21 08:40:48.541743: step: 524/529, loss: 1.7205760478973389 2023-01-21 08:40:49.768161: step: 528/529, loss: 4.640732765197754 2023-01-21 08:40:50.929219: step: 532/529, loss: 4.062249660491943 2023-01-21 08:40:52.127596: step: 536/529, loss: 0.7312127947807312 2023-01-21 08:40:53.300572: step: 540/529, loss: 1.187849760055542 2023-01-21 08:40:54.466569: step: 544/529, loss: 1.3783167600631714 2023-01-21 08:40:55.663654: step: 548/529, loss: 3.480274200439453 2023-01-21 08:40:56.864183: step: 552/529, loss: 1.71378755569458 2023-01-21 08:40:58.041833: step: 556/529, loss: 3.9147372245788574 2023-01-21 08:40:59.280133: step: 560/529, loss: 0.8801016211509705 2023-01-21 08:41:00.473241: step: 564/529, loss: 2.5718111991882324 2023-01-21 08:41:01.663073: step: 568/529, loss: 3.6981399059295654 2023-01-21 08:41:02.886734: step: 572/529, loss: 10.200998306274414 2023-01-21 08:41:04.084934: step: 576/529, loss: 0.8777449131011963 2023-01-21 08:41:05.257790: step: 580/529, loss: 1.061792016029358 2023-01-21 08:41:06.445654: step: 584/529, loss: 4.125081539154053 2023-01-21 08:41:07.664425: step: 588/529, loss: 0.5751572847366333 2023-01-21 08:41:08.867126: step: 592/529, loss: 0.34267377853393555 2023-01-21 08:41:10.060157: step: 596/529, loss: 0.7616095542907715 2023-01-21 08:41:11.198766: step: 600/529, loss: 1.309316635131836 2023-01-21 08:41:12.399277: step: 604/529, loss: 2.819707155227661 2023-01-21 08:41:13.653162: step: 608/529, loss: 4.250582218170166 2023-01-21 08:41:14.848825: step: 612/529, loss: 0.4889598786830902 2023-01-21 08:41:16.011974: step: 616/529, loss: 0.5848647952079773 2023-01-21 08:41:17.214629: step: 620/529, loss: 2.201519727706909 2023-01-21 08:41:18.403806: step: 624/529, loss: 1.4721647500991821 2023-01-21 08:41:19.574509: step: 628/529, loss: 0.5933083295822144 2023-01-21 08:41:20.751520: step: 632/529, loss: 6.337012767791748 2023-01-21 08:41:21.946703: step: 636/529, loss: 3.5360021591186523 2023-01-21 08:41:23.188170: step: 640/529, loss: 0.5196281671524048 2023-01-21 08:41:24.362214: step: 644/529, loss: 3.1214585304260254 2023-01-21 08:41:25.542865: step: 648/529, loss: 0.7267407178878784 2023-01-21 08:41:26.777088: step: 652/529, loss: 2.674508810043335 2023-01-21 08:41:27.965570: step: 656/529, loss: 0.6259007453918457 2023-01-21 08:41:29.158514: step: 660/529, loss: 1.703769564628601 2023-01-21 08:41:30.344445: step: 664/529, loss: 2.3324241638183594 2023-01-21 08:41:31.518368: step: 668/529, loss: 5.2529215812683105 2023-01-21 08:41:32.690325: step: 672/529, loss: 2.914316415786743 2023-01-21 08:41:33.854073: step: 676/529, loss: 3.5068469047546387 2023-01-21 08:41:35.047277: step: 680/529, loss: 0.3638465404510498 2023-01-21 08:41:36.265581: step: 684/529, loss: 1.2935245037078857 2023-01-21 08:41:37.468918: step: 688/529, loss: 2.0234553813934326 2023-01-21 08:41:38.636649: step: 692/529, loss: 0.3249831199645996 2023-01-21 08:41:39.856841: step: 696/529, loss: 0.5393751859664917 2023-01-21 08:41:41.023158: step: 700/529, loss: 1.056457757949829 2023-01-21 08:41:42.226215: step: 704/529, loss: 0.29491373896598816 2023-01-21 08:41:43.380209: step: 708/529, loss: 0.6376936435699463 2023-01-21 08:41:44.639130: step: 712/529, loss: 0.7085771560668945 2023-01-21 08:41:45.840073: step: 716/529, loss: 0.9203178882598877 2023-01-21 08:41:46.999736: step: 720/529, loss: 2.2862708568573 2023-01-21 08:41:48.230981: step: 724/529, loss: 1.162123441696167 2023-01-21 08:41:49.468760: step: 728/529, loss: 5.987920761108398 2023-01-21 08:41:50.658594: step: 732/529, loss: 1.1734355688095093 2023-01-21 08:41:51.833036: step: 736/529, loss: 0.4497973620891571 2023-01-21 08:41:53.013277: step: 740/529, loss: 4.0211100578308105 2023-01-21 08:41:54.243589: step: 744/529, loss: 3.999445676803589 2023-01-21 08:41:55.435589: step: 748/529, loss: 0.9228594899177551 2023-01-21 08:41:56.613755: step: 752/529, loss: 3.7844667434692383 2023-01-21 08:41:57.792428: step: 756/529, loss: 1.117471694946289 2023-01-21 08:41:58.997467: step: 760/529, loss: 0.5585752725601196 2023-01-21 08:42:00.164448: step: 764/529, loss: 3.0315425395965576 2023-01-21 08:42:01.391253: step: 768/529, loss: 0.6082629561424255 2023-01-21 08:42:02.573836: step: 772/529, loss: 1.1708859205245972 2023-01-21 08:42:03.743136: step: 776/529, loss: 1.9448699951171875 2023-01-21 08:42:04.903283: step: 780/529, loss: 1.0008831024169922 2023-01-21 08:42:06.112186: step: 784/529, loss: 0.40201902389526367 2023-01-21 08:42:07.289983: step: 788/529, loss: 0.5174985527992249 2023-01-21 08:42:08.480297: step: 792/529, loss: 0.4186003804206848 2023-01-21 08:42:09.677797: step: 796/529, loss: 0.17413930594921112 2023-01-21 08:42:10.859611: step: 800/529, loss: 0.8548687100410461 2023-01-21 08:42:11.998592: step: 804/529, loss: 0.6543086767196655 2023-01-21 08:42:13.200609: step: 808/529, loss: 1.0080170631408691 2023-01-21 08:42:14.399505: step: 812/529, loss: 0.7564957141876221 2023-01-21 08:42:15.573748: step: 816/529, loss: 0.8593447208404541 2023-01-21 08:42:16.783478: step: 820/529, loss: 1.6572167873382568 2023-01-21 08:42:18.015007: step: 824/529, loss: 0.639053463935852 2023-01-21 08:42:19.207117: step: 828/529, loss: 2.0464696884155273 2023-01-21 08:42:20.409451: step: 832/529, loss: 0.535546064376831 2023-01-21 08:42:21.604507: step: 836/529, loss: 2.7077503204345703 2023-01-21 08:42:22.791199: step: 840/529, loss: 0.6438050866127014 2023-01-21 08:42:23.974479: step: 844/529, loss: 0.9946825504302979 2023-01-21 08:42:25.166124: step: 848/529, loss: 0.47921955585479736 2023-01-21 08:42:26.336821: step: 852/529, loss: 1.7623847723007202 2023-01-21 08:42:27.558944: step: 856/529, loss: 2.768583297729492 2023-01-21 08:42:28.766414: step: 860/529, loss: 10.428298950195312 2023-01-21 08:42:29.974241: step: 864/529, loss: 6.824013710021973 2023-01-21 08:42:31.154568: step: 868/529, loss: 1.9721920490264893 2023-01-21 08:42:32.398202: step: 872/529, loss: 2.8358988761901855 2023-01-21 08:42:33.544141: step: 876/529, loss: 0.11737990379333496 2023-01-21 08:42:34.721657: step: 880/529, loss: 0.35182350873947144 2023-01-21 08:42:35.923611: step: 884/529, loss: 0.21775510907173157 2023-01-21 08:42:37.149892: step: 888/529, loss: 0.48986291885375977 2023-01-21 08:42:38.333033: step: 892/529, loss: 0.4439157247543335 2023-01-21 08:42:39.544437: step: 896/529, loss: 1.2020750045776367 2023-01-21 08:42:40.747840: step: 900/529, loss: 0.6157292723655701 2023-01-21 08:42:42.022781: step: 904/529, loss: 1.170060396194458 2023-01-21 08:42:43.185400: step: 908/529, loss: 0.5921035408973694 2023-01-21 08:42:44.418286: step: 912/529, loss: 1.0608599185943604 2023-01-21 08:42:45.622187: step: 916/529, loss: 2.3250577449798584 2023-01-21 08:42:46.830380: step: 920/529, loss: 5.517724990844727 2023-01-21 08:42:48.008545: step: 924/529, loss: 1.0833771228790283 2023-01-21 08:42:49.159495: step: 928/529, loss: 0.7859120965003967 2023-01-21 08:42:50.374360: step: 932/529, loss: 0.740260124206543 2023-01-21 08:42:51.572828: step: 936/529, loss: 0.5986397862434387 2023-01-21 08:42:52.758391: step: 940/529, loss: 2.626059055328369 2023-01-21 08:42:53.961591: step: 944/529, loss: 0.9404835104942322 2023-01-21 08:42:55.135445: step: 948/529, loss: 2.2421154975891113 2023-01-21 08:42:56.351978: step: 952/529, loss: 0.9911721348762512 2023-01-21 08:42:57.549774: step: 956/529, loss: 2.63169264793396 2023-01-21 08:42:58.783634: step: 960/529, loss: 0.5073085427284241 2023-01-21 08:42:59.959851: step: 964/529, loss: 2.324716329574585 2023-01-21 08:43:01.124473: step: 968/529, loss: 2.047309398651123 2023-01-21 08:43:02.324424: step: 972/529, loss: 4.869109153747559 2023-01-21 08:43:03.511981: step: 976/529, loss: 0.9011397957801819 2023-01-21 08:43:04.742046: step: 980/529, loss: 2.1416425704956055 2023-01-21 08:43:05.977877: step: 984/529, loss: 3.2345199584960938 2023-01-21 08:43:07.159718: step: 988/529, loss: 0.2511562705039978 2023-01-21 08:43:08.324750: step: 992/529, loss: 0.3559562861919403 2023-01-21 08:43:09.521674: step: 996/529, loss: 1.0940420627593994 2023-01-21 08:43:10.667666: step: 1000/529, loss: 1.4848049879074097 2023-01-21 08:43:11.823583: step: 1004/529, loss: 2.4509477615356445 2023-01-21 08:43:13.000106: step: 1008/529, loss: 1.0848877429962158 2023-01-21 08:43:14.225946: step: 1012/529, loss: 1.456751823425293 2023-01-21 08:43:15.407380: step: 1016/529, loss: 0.43307602405548096 2023-01-21 08:43:16.631166: step: 1020/529, loss: 1.3844835758209229 2023-01-21 08:43:17.808608: step: 1024/529, loss: 0.38100549578666687 2023-01-21 08:43:19.030917: step: 1028/529, loss: 0.9712100028991699 2023-01-21 08:43:20.239067: step: 1032/529, loss: 1.7322989702224731 2023-01-21 08:43:21.424909: step: 1036/529, loss: 1.124739170074463 2023-01-21 08:43:22.717292: step: 1040/529, loss: 0.41187784075737 2023-01-21 08:43:23.999289: step: 1044/529, loss: 1.361517071723938 2023-01-21 08:43:25.148542: step: 1048/529, loss: 1.112091064453125 2023-01-21 08:43:26.341936: step: 1052/529, loss: 0.3043155074119568 2023-01-21 08:43:27.562905: step: 1056/529, loss: 0.4209948778152466 2023-01-21 08:43:28.743365: step: 1060/529, loss: 0.9845127463340759 2023-01-21 08:43:29.936397: step: 1064/529, loss: 0.36977529525756836 2023-01-21 08:43:31.151258: step: 1068/529, loss: 0.7348709106445312 2023-01-21 08:43:32.371712: step: 1072/529, loss: 0.35695570707321167 2023-01-21 08:43:33.561778: step: 1076/529, loss: 2.1273000240325928 2023-01-21 08:43:34.792924: step: 1080/529, loss: 1.4644229412078857 2023-01-21 08:43:35.995419: step: 1084/529, loss: 0.829424262046814 2023-01-21 08:43:37.223180: step: 1088/529, loss: 0.5012081265449524 2023-01-21 08:43:38.420475: step: 1092/529, loss: 8.467086791992188 2023-01-21 08:43:39.637530: step: 1096/529, loss: 2.9554128646850586 2023-01-21 08:43:40.833022: step: 1100/529, loss: 0.7192343473434448 2023-01-21 08:43:42.017563: step: 1104/529, loss: 1.7786085605621338 2023-01-21 08:43:43.253418: step: 1108/529, loss: 0.4857497811317444 2023-01-21 08:43:44.426883: step: 1112/529, loss: 0.77387934923172 2023-01-21 08:43:45.598052: step: 1116/529, loss: 1.4289132356643677 2023-01-21 08:43:46.805757: step: 1120/529, loss: 0.43532103300094604 2023-01-21 08:43:47.993366: step: 1124/529, loss: 2.104881525039673 2023-01-21 08:43:49.224735: step: 1128/529, loss: 0.6311591267585754 2023-01-21 08:43:50.396393: step: 1132/529, loss: 2.038137912750244 2023-01-21 08:43:51.579286: step: 1136/529, loss: 1.050110101699829 2023-01-21 08:43:52.810869: step: 1140/529, loss: 1.4612703323364258 2023-01-21 08:43:53.994481: step: 1144/529, loss: 0.8107897043228149 2023-01-21 08:43:55.211057: step: 1148/529, loss: 2.7279953956604004 2023-01-21 08:43:56.415610: step: 1152/529, loss: 0.9125210642814636 2023-01-21 08:43:57.627223: step: 1156/529, loss: 0.7746531963348389 2023-01-21 08:43:58.919420: step: 1160/529, loss: 2.0470569133758545 2023-01-21 08:44:00.111809: step: 1164/529, loss: 2.166719436645508 2023-01-21 08:44:01.303761: step: 1168/529, loss: 6.890704154968262 2023-01-21 08:44:02.541639: step: 1172/529, loss: 1.9865775108337402 2023-01-21 08:44:03.729767: step: 1176/529, loss: 0.26850149035453796 2023-01-21 08:44:04.947511: step: 1180/529, loss: 0.5415112972259521 2023-01-21 08:44:06.163016: step: 1184/529, loss: 2.698915958404541 2023-01-21 08:44:07.322217: step: 1188/529, loss: 3.7939517498016357 2023-01-21 08:44:08.529087: step: 1192/529, loss: 0.6123003363609314 2023-01-21 08:44:09.688210: step: 1196/529, loss: 1.086844563484192 2023-01-21 08:44:10.893534: step: 1200/529, loss: 2.7950470447540283 2023-01-21 08:44:12.109193: step: 1204/529, loss: 1.1869001388549805 2023-01-21 08:44:13.307005: step: 1208/529, loss: 0.4582124948501587 2023-01-21 08:44:14.496768: step: 1212/529, loss: 1.6696866750717163 2023-01-21 08:44:15.693022: step: 1216/529, loss: 0.6265381574630737 2023-01-21 08:44:16.921780: step: 1220/529, loss: 0.604323148727417 2023-01-21 08:44:18.154935: step: 1224/529, loss: 3.478773593902588 2023-01-21 08:44:19.372934: step: 1228/529, loss: 0.2246539145708084 2023-01-21 08:44:20.544541: step: 1232/529, loss: 0.44249528646469116 2023-01-21 08:44:21.756281: step: 1236/529, loss: 1.106412410736084 2023-01-21 08:44:22.948324: step: 1240/529, loss: 0.1966862678527832 2023-01-21 08:44:24.109231: step: 1244/529, loss: 0.4916536509990692 2023-01-21 08:44:25.319362: step: 1248/529, loss: 7.338951110839844 2023-01-21 08:44:26.489780: step: 1252/529, loss: 0.18807964026927948 2023-01-21 08:44:27.704164: step: 1256/529, loss: 1.0040178298950195 2023-01-21 08:44:28.910178: step: 1260/529, loss: 0.4104175567626953 2023-01-21 08:44:30.131608: step: 1264/529, loss: 0.6464443206787109 2023-01-21 08:44:31.337409: step: 1268/529, loss: 1.4857327938079834 2023-01-21 08:44:32.513294: step: 1272/529, loss: 0.4584709107875824 2023-01-21 08:44:33.688928: step: 1276/529, loss: 0.2848011255264282 2023-01-21 08:44:34.893529: step: 1280/529, loss: 0.7101318836212158 2023-01-21 08:44:36.090735: step: 1284/529, loss: 0.8263359069824219 2023-01-21 08:44:37.319807: step: 1288/529, loss: 0.5674878358840942 2023-01-21 08:44:38.499290: step: 1292/529, loss: 0.4534398317337036 2023-01-21 08:44:39.697068: step: 1296/529, loss: 0.7932007312774658 2023-01-21 08:44:40.891430: step: 1300/529, loss: 0.3314913511276245 2023-01-21 08:44:42.052471: step: 1304/529, loss: 3.8802387714385986 2023-01-21 08:44:43.296629: step: 1308/529, loss: 0.36882850527763367 2023-01-21 08:44:44.491788: step: 1312/529, loss: 0.40705710649490356 2023-01-21 08:44:45.697953: step: 1316/529, loss: 0.439749538898468 2023-01-21 08:44:46.880824: step: 1320/529, loss: 0.7902179956436157 2023-01-21 08:44:48.096323: step: 1324/529, loss: 0.16738028824329376 2023-01-21 08:44:49.299544: step: 1328/529, loss: 0.39321136474609375 2023-01-21 08:44:50.538141: step: 1332/529, loss: 1.1230003833770752 2023-01-21 08:44:51.774441: step: 1336/529, loss: 2.9127755165100098 2023-01-21 08:44:52.987250: step: 1340/529, loss: 2.699781894683838 2023-01-21 08:44:54.160636: step: 1344/529, loss: 1.6985002756118774 2023-01-21 08:44:55.347292: step: 1348/529, loss: 1.84834623336792 2023-01-21 08:44:56.580509: step: 1352/529, loss: 3.946357250213623 2023-01-21 08:44:57.827685: step: 1356/529, loss: 0.4236244559288025 2023-01-21 08:44:59.040392: step: 1360/529, loss: 0.9585791826248169 2023-01-21 08:45:00.210560: step: 1364/529, loss: 1.753646969795227 2023-01-21 08:45:01.371470: step: 1368/529, loss: 0.43207404017448425 2023-01-21 08:45:02.558459: step: 1372/529, loss: 1.5062307119369507 2023-01-21 08:45:03.761248: step: 1376/529, loss: 0.12238531559705734 2023-01-21 08:45:04.929508: step: 1380/529, loss: 1.0887718200683594 2023-01-21 08:45:06.125979: step: 1384/529, loss: 0.5628679394721985 2023-01-21 08:45:07.347640: step: 1388/529, loss: 4.088003158569336 2023-01-21 08:45:08.552092: step: 1392/529, loss: 1.3439148664474487 2023-01-21 08:45:09.785333: step: 1396/529, loss: 1.5312730073928833 2023-01-21 08:45:10.985102: step: 1400/529, loss: 0.7861891984939575 2023-01-21 08:45:12.211540: step: 1404/529, loss: 0.25290799140930176 2023-01-21 08:45:13.430190: step: 1408/529, loss: 0.28866657614707947 2023-01-21 08:45:14.610226: step: 1412/529, loss: 0.42304468154907227 2023-01-21 08:45:15.789095: step: 1416/529, loss: 1.8015735149383545 2023-01-21 08:45:16.976967: step: 1420/529, loss: 2.2703049182891846 2023-01-21 08:45:18.156003: step: 1424/529, loss: 0.1433177888393402 2023-01-21 08:45:19.329802: step: 1428/529, loss: 0.2984890937805176 2023-01-21 08:45:20.544656: step: 1432/529, loss: 0.4579434394836426 2023-01-21 08:45:21.765926: step: 1436/529, loss: 1.3581758737564087 2023-01-21 08:45:22.936714: step: 1440/529, loss: 0.5189139246940613 2023-01-21 08:45:24.098134: step: 1444/529, loss: 0.43071165680885315 2023-01-21 08:45:25.298994: step: 1448/529, loss: 2.156521797180176 2023-01-21 08:45:26.550965: step: 1452/529, loss: 0.5254663228988647 2023-01-21 08:45:27.734199: step: 1456/529, loss: 1.1736807823181152 2023-01-21 08:45:28.914693: step: 1460/529, loss: 0.4148498475551605 2023-01-21 08:45:30.139660: step: 1464/529, loss: 0.8098351955413818 2023-01-21 08:45:31.294592: step: 1468/529, loss: 0.5285078287124634 2023-01-21 08:45:32.544588: step: 1472/529, loss: 0.3717874586582184 2023-01-21 08:45:33.706249: step: 1476/529, loss: 0.45098644495010376 2023-01-21 08:45:34.909711: step: 1480/529, loss: 0.9672597646713257 2023-01-21 08:45:36.139557: step: 1484/529, loss: 0.5948973894119263 2023-01-21 08:45:37.319015: step: 1488/529, loss: 0.47759419679641724 2023-01-21 08:45:38.526911: step: 1492/529, loss: 1.4880621433258057 2023-01-21 08:45:39.723158: step: 1496/529, loss: 0.18346476554870605 2023-01-21 08:45:40.959156: step: 1500/529, loss: 0.3813942074775696 2023-01-21 08:45:42.172649: step: 1504/529, loss: 1.7668808698654175 2023-01-21 08:45:43.378254: step: 1508/529, loss: 0.24518389999866486 2023-01-21 08:45:44.572387: step: 1512/529, loss: 0.26255813241004944 2023-01-21 08:45:45.751307: step: 1516/529, loss: 0.5573141574859619 2023-01-21 08:45:46.998734: step: 1520/529, loss: 2.8365182876586914 2023-01-21 08:45:48.190092: step: 1524/529, loss: 0.8681997656822205 2023-01-21 08:45:49.410302: step: 1528/529, loss: 0.3937135636806488 2023-01-21 08:45:50.601719: step: 1532/529, loss: 2.150111675262451 2023-01-21 08:45:51.801712: step: 1536/529, loss: 2.356813430786133 2023-01-21 08:45:52.991929: step: 1540/529, loss: 7.773714542388916 2023-01-21 08:45:54.179178: step: 1544/529, loss: 0.4997863173484802 2023-01-21 08:45:55.378871: step: 1548/529, loss: 0.2765703499317169 2023-01-21 08:45:56.595251: step: 1552/529, loss: 3.496220827102661 2023-01-21 08:45:57.802375: step: 1556/529, loss: 0.8893903493881226 2023-01-21 08:45:58.993340: step: 1560/529, loss: 0.4950907826423645 2023-01-21 08:46:00.177025: step: 1564/529, loss: 0.3308749198913574 2023-01-21 08:46:01.393011: step: 1568/529, loss: 0.3531980514526367 2023-01-21 08:46:02.571760: step: 1572/529, loss: 0.32358330488204956 2023-01-21 08:46:03.816323: step: 1576/529, loss: 1.8346972465515137 2023-01-21 08:46:05.021521: step: 1580/529, loss: 0.9833787679672241 2023-01-21 08:46:06.267207: step: 1584/529, loss: 0.31879687309265137 2023-01-21 08:46:07.499937: step: 1588/529, loss: 1.6434193849563599 2023-01-21 08:46:08.716212: step: 1592/529, loss: 0.46918296813964844 2023-01-21 08:46:09.928672: step: 1596/529, loss: 0.42517009377479553 2023-01-21 08:46:11.177119: step: 1600/529, loss: 0.6232078671455383 2023-01-21 08:46:12.339177: step: 1604/529, loss: 1.3762348890304565 2023-01-21 08:46:13.550298: step: 1608/529, loss: 1.7724915742874146 2023-01-21 08:46:14.726189: step: 1612/529, loss: 0.4551715850830078 2023-01-21 08:46:15.926366: step: 1616/529, loss: 2.9664227962493896 2023-01-21 08:46:17.113857: step: 1620/529, loss: 0.6250960826873779 2023-01-21 08:46:18.282005: step: 1624/529, loss: 0.12852254509925842 2023-01-21 08:46:19.526493: step: 1628/529, loss: 0.408147931098938 2023-01-21 08:46:20.704904: step: 1632/529, loss: 0.5588890314102173 2023-01-21 08:46:21.929570: step: 1636/529, loss: 0.3242533802986145 2023-01-21 08:46:23.090829: step: 1640/529, loss: 1.6165903806686401 2023-01-21 08:46:24.337920: step: 1644/529, loss: 0.3288017511367798 2023-01-21 08:46:25.563608: step: 1648/529, loss: 0.45196086168289185 2023-01-21 08:46:26.762228: step: 1652/529, loss: 0.8736375570297241 2023-01-21 08:46:27.972197: step: 1656/529, loss: 0.4038153290748596 2023-01-21 08:46:29.154281: step: 1660/529, loss: 1.762068271636963 2023-01-21 08:46:30.374803: step: 1664/529, loss: 1.0941978693008423 2023-01-21 08:46:31.600204: step: 1668/529, loss: 1.5832854509353638 2023-01-21 08:46:32.831195: step: 1672/529, loss: 1.5226056575775146 2023-01-21 08:46:33.993099: step: 1676/529, loss: 0.5419333577156067 2023-01-21 08:46:35.154895: step: 1680/529, loss: 0.2136530876159668 2023-01-21 08:46:36.360833: step: 1684/529, loss: 3.474647045135498 2023-01-21 08:46:37.566578: step: 1688/529, loss: 1.676355004310608 2023-01-21 08:46:38.785688: step: 1692/529, loss: 0.3030652105808258 2023-01-21 08:46:39.994315: step: 1696/529, loss: 0.5344671010971069 2023-01-21 08:46:41.187646: step: 1700/529, loss: 0.5074743628501892 2023-01-21 08:46:42.445584: step: 1704/529, loss: 0.30652475357055664 2023-01-21 08:46:43.638172: step: 1708/529, loss: 0.42206335067749023 2023-01-21 08:46:44.830442: step: 1712/529, loss: 0.6886497735977173 2023-01-21 08:46:46.038965: step: 1716/529, loss: 0.36699724197387695 2023-01-21 08:46:47.297631: step: 1720/529, loss: 0.5546180009841919 2023-01-21 08:46:48.529707: step: 1724/529, loss: 1.0773897171020508 2023-01-21 08:46:49.681215: step: 1728/529, loss: 0.47667115926742554 2023-01-21 08:46:50.870436: step: 1732/529, loss: 0.1742452085018158 2023-01-21 08:46:52.085487: step: 1736/529, loss: 0.48943957686424255 2023-01-21 08:46:53.284225: step: 1740/529, loss: 0.5446436405181885 2023-01-21 08:46:54.496369: step: 1744/529, loss: 3.94978666305542 2023-01-21 08:46:55.761164: step: 1748/529, loss: 1.9191770553588867 2023-01-21 08:46:56.995381: step: 1752/529, loss: 7.2219061851501465 2023-01-21 08:46:58.184799: step: 1756/529, loss: 0.9037526845932007 2023-01-21 08:46:59.369142: step: 1760/529, loss: 1.1580250263214111 2023-01-21 08:47:00.599178: step: 1764/529, loss: 0.5487180948257446 2023-01-21 08:47:01.792271: step: 1768/529, loss: 0.6091910004615784 2023-01-21 08:47:03.007500: step: 1772/529, loss: 0.5213258862495422 2023-01-21 08:47:04.212979: step: 1776/529, loss: 0.14984159171581268 2023-01-21 08:47:05.417654: step: 1780/529, loss: 0.5182837247848511 2023-01-21 08:47:06.628535: step: 1784/529, loss: 0.4881172180175781 2023-01-21 08:47:07.804881: step: 1788/529, loss: 1.3183894157409668 2023-01-21 08:47:09.021720: step: 1792/529, loss: 0.8680009841918945 2023-01-21 08:47:10.222808: step: 1796/529, loss: 0.9324303269386292 2023-01-21 08:47:11.441618: step: 1800/529, loss: 0.2602888345718384 2023-01-21 08:47:12.641355: step: 1804/529, loss: 1.0053061246871948 2023-01-21 08:47:13.815817: step: 1808/529, loss: 0.6792802810668945 2023-01-21 08:47:14.985707: step: 1812/529, loss: 0.9812270998954773 2023-01-21 08:47:16.190499: step: 1816/529, loss: 0.8634405732154846 2023-01-21 08:47:17.404565: step: 1820/529, loss: 1.580744743347168 2023-01-21 08:47:18.615356: step: 1824/529, loss: 0.3972037732601166 2023-01-21 08:47:19.829127: step: 1828/529, loss: 0.5509463548660278 2023-01-21 08:47:21.028240: step: 1832/529, loss: 0.16157567501068115 2023-01-21 08:47:22.202363: step: 1836/529, loss: 1.9963773488998413 2023-01-21 08:47:23.401238: step: 1840/529, loss: 1.3623301982879639 2023-01-21 08:47:24.587374: step: 1844/529, loss: 0.1316138356924057 2023-01-21 08:47:25.817095: step: 1848/529, loss: 0.5144818425178528 2023-01-21 08:47:27.010675: step: 1852/529, loss: 0.579322338104248 2023-01-21 08:47:28.202385: step: 1856/529, loss: 2.6421186923980713 2023-01-21 08:47:29.392013: step: 1860/529, loss: 3.7761054039001465 2023-01-21 08:47:30.604211: step: 1864/529, loss: 0.9291985630989075 2023-01-21 08:47:31.865567: step: 1868/529, loss: 0.21208734810352325 2023-01-21 08:47:33.033749: step: 1872/529, loss: 6.948488712310791 2023-01-21 08:47:34.207260: step: 1876/529, loss: 0.29366040229797363 2023-01-21 08:47:35.410829: step: 1880/529, loss: 1.0625988245010376 2023-01-21 08:47:36.574505: step: 1884/529, loss: 0.9475798606872559 2023-01-21 08:47:37.764870: step: 1888/529, loss: 0.20837651193141937 2023-01-21 08:47:38.963129: step: 1892/529, loss: 1.3488779067993164 2023-01-21 08:47:40.183748: step: 1896/529, loss: 0.2797640562057495 2023-01-21 08:47:41.391906: step: 1900/529, loss: 1.3218498229980469 2023-01-21 08:47:42.577703: step: 1904/529, loss: 0.5894363522529602 2023-01-21 08:47:43.795719: step: 1908/529, loss: 1.4543956518173218 2023-01-21 08:47:44.956251: step: 1912/529, loss: 0.4089512825012207 2023-01-21 08:47:46.147575: step: 1916/529, loss: 3.1534934043884277 2023-01-21 08:47:47.303418: step: 1920/529, loss: 0.39066389203071594 2023-01-21 08:47:48.494626: step: 1924/529, loss: 0.8841107487678528 2023-01-21 08:47:49.756793: step: 1928/529, loss: 0.5840445756912231 2023-01-21 08:47:51.003665: step: 1932/529, loss: 3.8740904331207275 2023-01-21 08:47:52.233837: step: 1936/529, loss: 1.0114915370941162 2023-01-21 08:47:53.493330: step: 1940/529, loss: 0.9592664837837219 2023-01-21 08:47:54.717381: step: 1944/529, loss: 0.6349629163742065 2023-01-21 08:47:55.953487: step: 1948/529, loss: 0.7871348261833191 2023-01-21 08:47:57.132307: step: 1952/529, loss: 3.2203218936920166 2023-01-21 08:47:58.336122: step: 1956/529, loss: 0.26926127076148987 2023-01-21 08:47:59.526196: step: 1960/529, loss: 0.09062681347131729 2023-01-21 08:48:00.732828: step: 1964/529, loss: 0.575076699256897 2023-01-21 08:48:01.942823: step: 1968/529, loss: 0.22789162397384644 2023-01-21 08:48:03.107949: step: 1972/529, loss: 0.9957990050315857 2023-01-21 08:48:04.293534: step: 1976/529, loss: 0.19148635864257812 2023-01-21 08:48:05.487290: step: 1980/529, loss: 0.3897726237773895 2023-01-21 08:48:06.710661: step: 1984/529, loss: 0.4202050566673279 2023-01-21 08:48:07.905528: step: 1988/529, loss: 1.2259410619735718 2023-01-21 08:48:09.101664: step: 1992/529, loss: 1.153421401977539 2023-01-21 08:48:10.298007: step: 1996/529, loss: 0.9048040509223938 2023-01-21 08:48:11.481560: step: 2000/529, loss: 0.8668357729911804 2023-01-21 08:48:12.688760: step: 2004/529, loss: 1.147754430770874 2023-01-21 08:48:13.907544: step: 2008/529, loss: 0.3463985025882721 2023-01-21 08:48:15.096243: step: 2012/529, loss: 0.4243202805519104 2023-01-21 08:48:16.303909: step: 2016/529, loss: 2.1583595275878906 2023-01-21 08:48:17.492395: step: 2020/529, loss: 0.2030327320098877 2023-01-21 08:48:18.696181: step: 2024/529, loss: 1.4683680534362793 2023-01-21 08:48:19.943332: step: 2028/529, loss: 0.38081997632980347 2023-01-21 08:48:21.127093: step: 2032/529, loss: 0.27679815888404846 2023-01-21 08:48:22.331082: step: 2036/529, loss: 1.3873214721679688 2023-01-21 08:48:23.527194: step: 2040/529, loss: 0.5157045125961304 2023-01-21 08:48:24.740273: step: 2044/529, loss: 1.4614315032958984 2023-01-21 08:48:25.967482: step: 2048/529, loss: 0.8679043054580688 2023-01-21 08:48:27.164432: step: 2052/529, loss: 0.2709521949291229 2023-01-21 08:48:28.319142: step: 2056/529, loss: 0.2642352283000946 2023-01-21 08:48:29.493431: step: 2060/529, loss: 1.4407376050949097 2023-01-21 08:48:30.721855: step: 2064/529, loss: 0.4164312481880188 2023-01-21 08:48:31.934936: step: 2068/529, loss: 0.34282445907592773 2023-01-21 08:48:33.128954: step: 2072/529, loss: 0.7626917958259583 2023-01-21 08:48:34.369209: step: 2076/529, loss: 2.033196449279785 2023-01-21 08:48:35.637828: step: 2080/529, loss: 3.6910836696624756 2023-01-21 08:48:36.889557: step: 2084/529, loss: 0.5010618567466736 2023-01-21 08:48:38.108879: step: 2088/529, loss: 0.6785228848457336 2023-01-21 08:48:39.316131: step: 2092/529, loss: 0.3297010064125061 2023-01-21 08:48:40.505299: step: 2096/529, loss: 0.8777534365653992 2023-01-21 08:48:41.730578: step: 2100/529, loss: 0.5245559215545654 2023-01-21 08:48:42.935749: step: 2104/529, loss: 2.2352466583251953 2023-01-21 08:48:44.161470: step: 2108/529, loss: 0.39593085646629333 2023-01-21 08:48:45.344468: step: 2112/529, loss: 0.20904827117919922 2023-01-21 08:48:46.545532: step: 2116/529, loss: 0.5683928728103638 ================================================== Loss: 3.433 -------------------- Dev: {'event': {'p': 0.5544692737430168, 'r': 0.5286284953395473, 'f1': 0.5412406271301976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Test: {'event': {'p': 0.4727408513816281, 'r': 0.3792690233672858, 'f1': 0.4208776595744681}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Chinese: {'event': {'p': 0.5, 'r': 0.4074074074074074, 'f1': 0.44897959183673464}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Korean: {'event': {'p': 0.7307692307692307, 'r': 0.30158730158730157, 'f1': 0.42696629213483145}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Russian: {'event': {'p': 0.47058823529411764, 'r': 0.2222222222222222, 'f1': 0.3018867924528302}, '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': 0.5544692737430168, 'r': 0.5286284953395473, 'f1': 0.5412406271301976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Chinese: {'event': {'p': 0.4727408513816281, 'r': 0.3792690233672858, 'f1': 0.4208776595744681}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Chinese: {'event': {'p': 0.5, 'r': 0.4074074074074074, 'f1': 0.44897959183673464}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Korean: {'event': {'p': 0.5544692737430168, 'r': 0.5286284953395473, 'f1': 0.5412406271301976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Korean: {'event': {'p': 0.4727408513816281, 'r': 0.3792690233672858, 'f1': 0.4208776595744681}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Korean: {'event': {'p': 0.7307692307692307, 'r': 0.30158730158730157, 'f1': 0.42696629213483145}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Russian: {'event': {'p': 0.5544692737430168, 'r': 0.5286284953395473, 'f1': 0.5412406271301976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Russian: {'event': {'p': 0.4727408513816281, 'r': 0.3792690233672858, 'f1': 0.4208776595744681}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Russian: {'event': {'p': 0.47058823529411764, 'r': 0.2222222222222222, 'f1': 0.3018867924528302}, '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 08:49:44.026873: step: 4/529, loss: 0.5469247102737427 2023-01-21 08:49:45.226462: step: 8/529, loss: 0.29029884934425354 2023-01-21 08:49:46.444695: step: 12/529, loss: 2.2856011390686035 2023-01-21 08:49:47.602557: step: 16/529, loss: 0.26024913787841797 2023-01-21 08:49:48.780880: step: 20/529, loss: 0.2523704171180725 2023-01-21 08:49:49.997512: step: 24/529, loss: 0.7009915709495544 2023-01-21 08:49:51.169139: step: 28/529, loss: 0.6086962819099426 2023-01-21 08:49:52.413833: step: 32/529, loss: 1.2982348203659058 2023-01-21 08:49:53.577185: step: 36/529, loss: 1.8348171710968018 2023-01-21 08:49:54.760348: step: 40/529, loss: 0.47031474113464355 2023-01-21 08:49:55.961642: step: 44/529, loss: 0.9115546941757202 2023-01-21 08:49:57.151943: step: 48/529, loss: 0.38211649656295776 2023-01-21 08:49:58.338744: step: 52/529, loss: 0.13293424248695374 2023-01-21 08:49:59.574403: step: 56/529, loss: 0.26782137155532837 2023-01-21 08:50:00.801681: step: 60/529, loss: 0.33807554841041565 2023-01-21 08:50:02.039161: step: 64/529, loss: 0.5734786987304688 2023-01-21 08:50:03.223699: step: 68/529, loss: 0.2140342742204666 2023-01-21 08:50:04.378671: step: 72/529, loss: 0.21558427810668945 2023-01-21 08:50:05.602309: step: 76/529, loss: 0.17573431134223938 2023-01-21 08:50:06.767184: step: 80/529, loss: 0.955725908279419 2023-01-21 08:50:08.001186: step: 84/529, loss: 0.5600137710571289 2023-01-21 08:50:09.182376: step: 88/529, loss: 0.39307862520217896 2023-01-21 08:50:10.417824: step: 92/529, loss: 0.8976966738700867 2023-01-21 08:50:11.614927: step: 96/529, loss: 0.7666264176368713 2023-01-21 08:50:12.801445: step: 100/529, loss: 0.31468749046325684 2023-01-21 08:50:14.031047: step: 104/529, loss: 0.6417840123176575 2023-01-21 08:50:15.222657: step: 108/529, loss: 0.30608415603637695 2023-01-21 08:50:16.432163: step: 112/529, loss: 2.5712764263153076 2023-01-21 08:50:17.738543: step: 116/529, loss: 0.40532737970352173 2023-01-21 08:50:18.929620: step: 120/529, loss: 0.3818085789680481 2023-01-21 08:50:20.140002: step: 124/529, loss: 0.28497064113616943 2023-01-21 08:50:21.357718: step: 128/529, loss: 0.6045374274253845 2023-01-21 08:50:22.531961: step: 132/529, loss: 0.35619086027145386 2023-01-21 08:50:23.712118: step: 136/529, loss: 0.13047190010547638 2023-01-21 08:50:24.940762: step: 140/529, loss: 1.2081345319747925 2023-01-21 08:50:26.158369: step: 144/529, loss: 1.1790626049041748 2023-01-21 08:50:27.352573: step: 148/529, loss: 1.0163335800170898 2023-01-21 08:50:28.549312: step: 152/529, loss: 1.6506152153015137 2023-01-21 08:50:29.741298: step: 156/529, loss: 0.5505706071853638 2023-01-21 08:50:30.946768: step: 160/529, loss: 0.4442530870437622 2023-01-21 08:50:32.141355: step: 164/529, loss: 0.21369346976280212 2023-01-21 08:50:33.341987: step: 168/529, loss: 0.2938529849052429 2023-01-21 08:50:34.480110: step: 172/529, loss: 0.1406106948852539 2023-01-21 08:50:35.663549: step: 176/529, loss: 0.2674335539340973 2023-01-21 08:50:36.855150: step: 180/529, loss: 0.7235414385795593 2023-01-21 08:50:38.069890: step: 184/529, loss: 0.2027774453163147 2023-01-21 08:50:39.281259: step: 188/529, loss: 2.71345591545105 2023-01-21 08:50:40.530483: step: 192/529, loss: 1.8358243703842163 2023-01-21 08:50:41.754067: step: 196/529, loss: 1.1490973234176636 2023-01-21 08:50:42.970029: step: 200/529, loss: 0.7929194569587708 2023-01-21 08:50:44.163082: step: 204/529, loss: 0.24325361847877502 2023-01-21 08:50:45.354147: step: 208/529, loss: 0.8279992341995239 2023-01-21 08:50:46.577383: step: 212/529, loss: 1.5655983686447144 2023-01-21 08:50:47.782660: step: 216/529, loss: 0.5837101936340332 2023-01-21 08:50:48.976840: step: 220/529, loss: 0.2387804090976715 2023-01-21 08:50:50.166582: step: 224/529, loss: 0.9586046934127808 2023-01-21 08:50:51.362732: step: 228/529, loss: 1.573075294494629 2023-01-21 08:50:52.550588: step: 232/529, loss: 0.8786381483078003 2023-01-21 08:50:53.734289: step: 236/529, loss: 0.30920690298080444 2023-01-21 08:50:55.001669: step: 240/529, loss: 0.8637639284133911 2023-01-21 08:50:56.182203: step: 244/529, loss: 1.4042590856552124 2023-01-21 08:50:57.341298: step: 248/529, loss: 0.7386443614959717 2023-01-21 08:50:58.528638: step: 252/529, loss: 0.8712807893753052 2023-01-21 08:50:59.706857: step: 256/529, loss: 0.4034702777862549 2023-01-21 08:51:00.938276: step: 260/529, loss: 0.3156128525733948 2023-01-21 08:51:02.178252: step: 264/529, loss: 0.7282426357269287 2023-01-21 08:51:03.398935: step: 268/529, loss: 0.10016379505395889 2023-01-21 08:51:04.609073: step: 272/529, loss: 0.3434237539768219 2023-01-21 08:51:05.802422: step: 276/529, loss: 0.30866673588752747 2023-01-21 08:51:07.007007: step: 280/529, loss: 0.130573570728302 2023-01-21 08:51:08.246054: step: 284/529, loss: 1.5664870738983154 2023-01-21 08:51:09.411386: step: 288/529, loss: 0.7725616693496704 2023-01-21 08:51:10.619780: step: 292/529, loss: 0.2995683550834656 2023-01-21 08:51:11.858380: step: 296/529, loss: 0.28513187170028687 2023-01-21 08:51:13.149792: step: 300/529, loss: 8.51626968383789 2023-01-21 08:51:14.351938: step: 304/529, loss: 1.3881690502166748 2023-01-21 08:51:15.528402: step: 308/529, loss: 0.319705605506897 2023-01-21 08:51:16.743177: step: 312/529, loss: 0.7271748781204224 2023-01-21 08:51:17.938028: step: 316/529, loss: 0.39506995677948 2023-01-21 08:51:19.146926: step: 320/529, loss: 1.8276278972625732 2023-01-21 08:51:20.328024: step: 324/529, loss: 0.4516967833042145 2023-01-21 08:51:21.499750: step: 328/529, loss: 0.7862209677696228 2023-01-21 08:51:22.706226: step: 332/529, loss: 0.7785571813583374 2023-01-21 08:51:23.917433: step: 336/529, loss: 0.9727303981781006 2023-01-21 08:51:25.107245: step: 340/529, loss: 2.3505945205688477 2023-01-21 08:51:26.270654: step: 344/529, loss: 0.4268733263015747 2023-01-21 08:51:27.472861: step: 348/529, loss: 0.4842044711112976 2023-01-21 08:51:28.674875: step: 352/529, loss: 0.08876566588878632 2023-01-21 08:51:29.877495: step: 356/529, loss: 0.8700810670852661 2023-01-21 08:51:31.077424: step: 360/529, loss: 0.7326598167419434 2023-01-21 08:51:32.279547: step: 364/529, loss: 0.4604230225086212 2023-01-21 08:51:33.516959: step: 368/529, loss: 0.29568353295326233 2023-01-21 08:51:34.679151: step: 372/529, loss: 0.6302188634872437 2023-01-21 08:51:35.842898: step: 376/529, loss: 0.6179839372634888 2023-01-21 08:51:37.025031: step: 380/529, loss: 0.2163233757019043 2023-01-21 08:51:38.251615: step: 384/529, loss: 1.1559795141220093 2023-01-21 08:51:39.455244: step: 388/529, loss: 1.140015721321106 2023-01-21 08:51:40.641032: step: 392/529, loss: 0.9983171224594116 2023-01-21 08:51:41.852201: step: 396/529, loss: 0.29577934741973877 2023-01-21 08:51:43.069365: step: 400/529, loss: 0.16053524613380432 2023-01-21 08:51:44.268335: step: 404/529, loss: 0.29787904024124146 2023-01-21 08:51:45.447312: step: 408/529, loss: 8.647085189819336 2023-01-21 08:51:46.629886: step: 412/529, loss: 0.2223643809556961 2023-01-21 08:51:47.827335: step: 416/529, loss: 0.5561660528182983 2023-01-21 08:51:49.064942: step: 420/529, loss: 0.4473477602005005 2023-01-21 08:51:50.252644: step: 424/529, loss: 0.18202051520347595 2023-01-21 08:51:51.432781: step: 428/529, loss: 0.3774683177471161 2023-01-21 08:51:52.637444: step: 432/529, loss: 1.8667852878570557 2023-01-21 08:51:53.893171: step: 436/529, loss: 0.22732192277908325 2023-01-21 08:51:55.119905: step: 440/529, loss: 1.1841437816619873 2023-01-21 08:51:56.338898: step: 444/529, loss: 0.7162110805511475 2023-01-21 08:51:57.534740: step: 448/529, loss: 0.19190216064453125 2023-01-21 08:51:58.707679: step: 452/529, loss: 0.13999691605567932 2023-01-21 08:51:59.887766: step: 456/529, loss: 0.7005730271339417 2023-01-21 08:52:01.092555: step: 460/529, loss: 1.731724500656128 2023-01-21 08:52:02.283476: step: 464/529, loss: 0.2750546336174011 2023-01-21 08:52:03.470333: step: 468/529, loss: 1.6575648784637451 2023-01-21 08:52:04.659340: step: 472/529, loss: 0.2552250921726227 2023-01-21 08:52:05.869961: step: 476/529, loss: 1.4123848676681519 2023-01-21 08:52:07.100587: step: 480/529, loss: 1.641983985900879 2023-01-21 08:52:08.274351: step: 484/529, loss: 0.21824704110622406 2023-01-21 08:52:09.466131: step: 488/529, loss: 0.4703761339187622 2023-01-21 08:52:10.645559: step: 492/529, loss: 0.29395991563796997 2023-01-21 08:52:11.827362: step: 496/529, loss: 0.3383379578590393 2023-01-21 08:52:13.044279: step: 500/529, loss: 0.40370863676071167 2023-01-21 08:52:14.217172: step: 504/529, loss: 0.9405983090400696 2023-01-21 08:52:15.442137: step: 508/529, loss: 0.6169747114181519 2023-01-21 08:52:16.640949: step: 512/529, loss: 1.9449288845062256 2023-01-21 08:52:17.803861: step: 516/529, loss: 0.20881938934326172 2023-01-21 08:52:19.001446: step: 520/529, loss: 0.22269250452518463 2023-01-21 08:52:20.182552: step: 524/529, loss: 0.4363614320755005 2023-01-21 08:52:21.373221: step: 528/529, loss: 0.11099929362535477 2023-01-21 08:52:22.594855: step: 532/529, loss: 0.24352195858955383 2023-01-21 08:52:23.780560: step: 536/529, loss: 0.4304102957248688 2023-01-21 08:52:24.994713: step: 540/529, loss: 0.8257529735565186 2023-01-21 08:52:26.153049: step: 544/529, loss: 0.261637419462204 2023-01-21 08:52:27.480459: step: 548/529, loss: 1.3856748342514038 2023-01-21 08:52:28.691065: step: 552/529, loss: 0.07403989136219025 2023-01-21 08:52:29.854931: step: 556/529, loss: 0.1703689694404602 2023-01-21 08:52:31.082042: step: 560/529, loss: 0.32421523332595825 2023-01-21 08:52:32.317061: step: 564/529, loss: 0.3813760280609131 2023-01-21 08:52:33.555161: step: 568/529, loss: 2.0316994190216064 2023-01-21 08:52:34.713605: step: 572/529, loss: 0.6919819712638855 2023-01-21 08:52:35.967485: step: 576/529, loss: 0.2741979658603668 2023-01-21 08:52:37.150007: step: 580/529, loss: 0.18556852638721466 2023-01-21 08:52:38.349233: step: 584/529, loss: 3.65915584564209 2023-01-21 08:52:39.525920: step: 588/529, loss: 2.030156135559082 2023-01-21 08:52:40.711106: step: 592/529, loss: 0.41374799609184265 2023-01-21 08:52:41.899757: step: 596/529, loss: 1.6483994722366333 2023-01-21 08:52:43.131018: step: 600/529, loss: 0.37363871932029724 2023-01-21 08:52:44.312807: step: 604/529, loss: 0.2248472273349762 2023-01-21 08:52:45.544593: step: 608/529, loss: 0.4030396342277527 2023-01-21 08:52:46.778402: step: 612/529, loss: 0.703781008720398 2023-01-21 08:52:48.038311: step: 616/529, loss: 0.8241526484489441 2023-01-21 08:52:49.226121: step: 620/529, loss: 1.5269960165023804 2023-01-21 08:52:50.422053: step: 624/529, loss: 0.3522728979587555 2023-01-21 08:52:51.612920: step: 628/529, loss: 0.10946894437074661 2023-01-21 08:52:52.820705: step: 632/529, loss: 0.14697766304016113 2023-01-21 08:52:54.012420: step: 636/529, loss: 0.795026957988739 2023-01-21 08:52:55.205986: step: 640/529, loss: 0.2673977017402649 2023-01-21 08:52:56.375718: step: 644/529, loss: 1.2692445516586304 2023-01-21 08:52:57.574678: step: 648/529, loss: 1.2542775869369507 2023-01-21 08:52:58.781445: step: 652/529, loss: 2.4184679985046387 2023-01-21 08:52:59.985518: step: 656/529, loss: 0.8977761268615723 2023-01-21 08:53:01.162409: step: 660/529, loss: 0.1103479415178299 2023-01-21 08:53:02.442649: step: 664/529, loss: 7.595412254333496 2023-01-21 08:53:03.653465: step: 668/529, loss: 2.7754533290863037 2023-01-21 08:53:04.854347: step: 672/529, loss: 1.4375044107437134 2023-01-21 08:53:06.074100: step: 676/529, loss: 0.2556249499320984 2023-01-21 08:53:07.292118: step: 680/529, loss: 8.585527420043945 2023-01-21 08:53:08.503067: step: 684/529, loss: 0.3497764766216278 2023-01-21 08:53:09.689734: step: 688/529, loss: 0.8973914384841919 2023-01-21 08:53:10.919148: step: 692/529, loss: 7.146213054656982 2023-01-21 08:53:12.091890: step: 696/529, loss: 0.44740742444992065 2023-01-21 08:53:13.280780: step: 700/529, loss: 0.4297164976596832 2023-01-21 08:53:14.501571: step: 704/529, loss: 0.2591429352760315 2023-01-21 08:53:15.660921: step: 708/529, loss: 0.08609075844287872 2023-01-21 08:53:16.873630: step: 712/529, loss: 0.29313117265701294 2023-01-21 08:53:18.044302: step: 716/529, loss: 0.40758365392684937 2023-01-21 08:53:19.235371: step: 720/529, loss: 0.20712977647781372 2023-01-21 08:53:20.481812: step: 724/529, loss: 0.4123830795288086 2023-01-21 08:53:21.651882: step: 728/529, loss: 2.760382652282715 2023-01-21 08:53:22.834363: step: 732/529, loss: 0.32448187470436096 2023-01-21 08:53:24.017150: step: 736/529, loss: 0.24875859916210175 2023-01-21 08:53:25.222034: step: 740/529, loss: 2.406853675842285 2023-01-21 08:53:26.383297: step: 744/529, loss: 0.7407734990119934 2023-01-21 08:53:27.597212: step: 748/529, loss: 1.4668359756469727 2023-01-21 08:53:28.792311: step: 752/529, loss: 0.8690974712371826 2023-01-21 08:53:29.944559: step: 756/529, loss: 2.563384532928467 2023-01-21 08:53:31.126820: step: 760/529, loss: 1.0204862356185913 2023-01-21 08:53:32.383778: step: 764/529, loss: 0.36617690324783325 2023-01-21 08:53:33.588827: step: 768/529, loss: 0.5697610974311829 2023-01-21 08:53:34.761477: step: 772/529, loss: 0.23933817446231842 2023-01-21 08:53:35.965899: step: 776/529, loss: 0.2333511859178543 2023-01-21 08:53:37.173409: step: 780/529, loss: 0.5242372751235962 2023-01-21 08:53:38.379232: step: 784/529, loss: 0.5164096355438232 2023-01-21 08:53:39.566667: step: 788/529, loss: 0.3202282786369324 2023-01-21 08:53:40.758365: step: 792/529, loss: 1.4534978866577148 2023-01-21 08:53:41.982927: step: 796/529, loss: 0.8111439943313599 2023-01-21 08:53:43.186303: step: 800/529, loss: 0.34762126207351685 2023-01-21 08:53:44.419486: step: 804/529, loss: 0.4840986430644989 2023-01-21 08:53:45.578790: step: 808/529, loss: 0.6049836874008179 2023-01-21 08:53:46.786978: step: 812/529, loss: 0.23891717195510864 2023-01-21 08:53:48.011713: step: 816/529, loss: 0.5742517709732056 2023-01-21 08:53:49.251654: step: 820/529, loss: 0.33200645446777344 2023-01-21 08:53:50.405064: step: 824/529, loss: 0.2666700482368469 2023-01-21 08:53:51.666944: step: 828/529, loss: 2.03470516204834 2023-01-21 08:53:52.914160: step: 832/529, loss: 6.3137359619140625 2023-01-21 08:53:54.119674: step: 836/529, loss: 2.77896785736084 2023-01-21 08:53:55.328122: step: 840/529, loss: 0.08535037189722061 2023-01-21 08:53:56.509168: step: 844/529, loss: 0.9909968972206116 2023-01-21 08:53:57.715253: step: 848/529, loss: 1.377442717552185 2023-01-21 08:53:58.911263: step: 852/529, loss: 0.8108502626419067 2023-01-21 08:54:00.132253: step: 856/529, loss: 0.8727612495422363 2023-01-21 08:54:01.346848: step: 860/529, loss: 0.20590420067310333 2023-01-21 08:54:02.523125: step: 864/529, loss: 0.1079469695687294 2023-01-21 08:54:03.699977: step: 868/529, loss: 0.2608259916305542 2023-01-21 08:54:04.889877: step: 872/529, loss: 1.2493480443954468 2023-01-21 08:54:06.125974: step: 876/529, loss: 2.429363250732422 2023-01-21 08:54:07.384599: step: 880/529, loss: 0.3172902464866638 2023-01-21 08:54:08.618802: step: 884/529, loss: 0.3162820041179657 2023-01-21 08:54:09.830867: step: 888/529, loss: 1.5097625255584717 2023-01-21 08:54:11.021331: step: 892/529, loss: 0.16014719009399414 2023-01-21 08:54:12.219092: step: 896/529, loss: 1.0490915775299072 2023-01-21 08:54:13.404858: step: 900/529, loss: 0.20512184500694275 2023-01-21 08:54:14.594541: step: 904/529, loss: 0.623953104019165 2023-01-21 08:54:15.784809: step: 908/529, loss: 0.6367088556289673 2023-01-21 08:54:16.940872: step: 912/529, loss: 0.15613150596618652 2023-01-21 08:54:18.186063: step: 916/529, loss: 0.5094121694564819 2023-01-21 08:54:19.383761: step: 920/529, loss: 1.450291395187378 2023-01-21 08:54:20.613015: step: 924/529, loss: 0.6057009696960449 2023-01-21 08:54:21.807309: step: 928/529, loss: 1.3076021671295166 2023-01-21 08:54:23.015185: step: 932/529, loss: 2.4547219276428223 2023-01-21 08:54:24.237958: step: 936/529, loss: 1.0458277463912964 2023-01-21 08:54:25.450763: step: 940/529, loss: 0.7264676094055176 2023-01-21 08:54:26.641670: step: 944/529, loss: 0.4123757481575012 2023-01-21 08:54:27.816358: step: 948/529, loss: 0.18185758590698242 2023-01-21 08:54:29.027184: step: 952/529, loss: 0.3374711871147156 2023-01-21 08:54:30.220926: step: 956/529, loss: 0.4082280993461609 2023-01-21 08:54:31.404775: step: 960/529, loss: 0.5859688520431519 2023-01-21 08:54:32.686228: step: 964/529, loss: 0.5211753249168396 2023-01-21 08:54:33.865667: step: 968/529, loss: 0.7855747938156128 2023-01-21 08:54:35.055222: step: 972/529, loss: 0.5818166732788086 2023-01-21 08:54:36.262429: step: 976/529, loss: 0.434374064207077 2023-01-21 08:54:37.468574: step: 980/529, loss: 7.275872230529785 2023-01-21 08:54:38.726168: step: 984/529, loss: 0.47187793254852295 2023-01-21 08:54:39.953289: step: 988/529, loss: 1.1535205841064453 2023-01-21 08:54:41.162080: step: 992/529, loss: 1.2490960359573364 2023-01-21 08:54:42.336539: step: 996/529, loss: 0.25483590364456177 2023-01-21 08:54:43.554380: step: 1000/529, loss: 0.15983963012695312 2023-01-21 08:54:44.773191: step: 1004/529, loss: 2.033478021621704 2023-01-21 08:54:45.996813: step: 1008/529, loss: 1.1238760948181152 2023-01-21 08:54:47.174186: step: 1012/529, loss: 0.43764179944992065 2023-01-21 08:54:48.439544: step: 1016/529, loss: 1.9681974649429321 2023-01-21 08:54:49.684816: step: 1020/529, loss: 1.421175241470337 2023-01-21 08:54:50.892725: step: 1024/529, loss: 0.14086714386940002 2023-01-21 08:54:52.110166: step: 1028/529, loss: 0.26029786467552185 2023-01-21 08:54:53.267249: step: 1032/529, loss: 0.48388442397117615 2023-01-21 08:54:54.535188: step: 1036/529, loss: 0.12781819701194763 2023-01-21 08:54:55.715458: step: 1040/529, loss: 0.980811595916748 2023-01-21 08:54:56.905866: step: 1044/529, loss: 0.5740975737571716 2023-01-21 08:54:58.067207: step: 1048/529, loss: 1.688405156135559 2023-01-21 08:54:59.255865: step: 1052/529, loss: 0.18498817086219788 2023-01-21 08:55:00.429286: step: 1056/529, loss: 0.8830058574676514 2023-01-21 08:55:01.624760: step: 1060/529, loss: 0.7573413848876953 2023-01-21 08:55:02.839127: step: 1064/529, loss: 0.11612348258495331 2023-01-21 08:55:04.028564: step: 1068/529, loss: 0.5016762018203735 2023-01-21 08:55:05.255194: step: 1072/529, loss: 1.8578795194625854 2023-01-21 08:55:06.470063: step: 1076/529, loss: 1.0674693584442139 2023-01-21 08:55:07.652008: step: 1080/529, loss: 0.7718737721443176 2023-01-21 08:55:08.926610: step: 1084/529, loss: 0.20034612715244293 2023-01-21 08:55:10.132539: step: 1088/529, loss: 0.12779387831687927 2023-01-21 08:55:11.324392: step: 1092/529, loss: 0.9483867287635803 2023-01-21 08:55:12.538854: step: 1096/529, loss: 1.1970624923706055 2023-01-21 08:55:13.777591: step: 1100/529, loss: 0.26124781370162964 2023-01-21 08:55:14.996412: step: 1104/529, loss: 0.34428417682647705 2023-01-21 08:55:16.190179: step: 1108/529, loss: 0.23632851243019104 2023-01-21 08:55:17.386474: step: 1112/529, loss: 1.1896905899047852 2023-01-21 08:55:18.544679: step: 1116/529, loss: 0.10548386722803116 2023-01-21 08:55:19.726368: step: 1120/529, loss: 0.30458417534828186 2023-01-21 08:55:20.917630: step: 1124/529, loss: 0.14331656694412231 2023-01-21 08:55:22.104731: step: 1128/529, loss: 0.2040661871433258 2023-01-21 08:55:23.314204: step: 1132/529, loss: 0.23754163086414337 2023-01-21 08:55:24.534308: step: 1136/529, loss: 3.95654559135437 2023-01-21 08:55:25.724655: step: 1140/529, loss: 0.19373951852321625 2023-01-21 08:55:26.960797: step: 1144/529, loss: 1.4854968786239624 2023-01-21 08:55:28.135591: step: 1148/529, loss: 0.18933773040771484 2023-01-21 08:55:29.344469: step: 1152/529, loss: 0.37844783067703247 2023-01-21 08:55:30.524034: step: 1156/529, loss: 0.4130516052246094 2023-01-21 08:55:31.697175: step: 1160/529, loss: 1.2527437210083008 2023-01-21 08:55:32.901499: step: 1164/529, loss: 0.1148277297616005 2023-01-21 08:55:34.105447: step: 1168/529, loss: 1.7610738277435303 2023-01-21 08:55:35.285589: step: 1172/529, loss: 0.6940605640411377 2023-01-21 08:55:36.495095: step: 1176/529, loss: 3.0194191932678223 2023-01-21 08:55:37.685597: step: 1180/529, loss: 0.13224294781684875 2023-01-21 08:55:38.916708: step: 1184/529, loss: 0.19396701455116272 2023-01-21 08:55:40.112083: step: 1188/529, loss: 0.2696632742881775 2023-01-21 08:55:41.317578: step: 1192/529, loss: 0.336747407913208 2023-01-21 08:55:42.568475: step: 1196/529, loss: 1.589300513267517 2023-01-21 08:55:43.776459: step: 1200/529, loss: 0.06356213241815567 2023-01-21 08:55:45.016892: step: 1204/529, loss: 0.15005962550640106 2023-01-21 08:55:46.190116: step: 1208/529, loss: 0.27747613191604614 2023-01-21 08:55:47.377393: step: 1212/529, loss: 0.12508916854858398 2023-01-21 08:55:48.573640: step: 1216/529, loss: 0.24230414628982544 2023-01-21 08:55:49.746629: step: 1220/529, loss: 0.5309888124465942 2023-01-21 08:55:50.931496: step: 1224/529, loss: 0.8018829226493835 2023-01-21 08:55:52.128694: step: 1228/529, loss: 0.6966332197189331 2023-01-21 08:55:53.327166: step: 1232/529, loss: 0.8874320387840271 2023-01-21 08:55:54.542379: step: 1236/529, loss: 1.4590981006622314 2023-01-21 08:55:55.728211: step: 1240/529, loss: 0.6870986223220825 2023-01-21 08:55:56.895603: step: 1244/529, loss: 0.4094250798225403 2023-01-21 08:55:58.071383: step: 1248/529, loss: 0.6817470788955688 2023-01-21 08:55:59.284505: step: 1252/529, loss: 0.12213325500488281 2023-01-21 08:56:00.482284: step: 1256/529, loss: 1.6356189250946045 2023-01-21 08:56:01.687313: step: 1260/529, loss: 0.211894229054451 2023-01-21 08:56:02.892717: step: 1264/529, loss: 0.10953550785779953 2023-01-21 08:56:04.073142: step: 1268/529, loss: 0.661410927772522 2023-01-21 08:56:05.258910: step: 1272/529, loss: 0.852221667766571 2023-01-21 08:56:06.447660: step: 1276/529, loss: 0.17647600173950195 2023-01-21 08:56:07.653806: step: 1280/529, loss: 0.289465993642807 2023-01-21 08:56:08.842968: step: 1284/529, loss: 0.623895525932312 2023-01-21 08:56:10.070651: step: 1288/529, loss: 0.2133682519197464 2023-01-21 08:56:11.309528: step: 1292/529, loss: 2.2674007415771484 2023-01-21 08:56:12.520249: step: 1296/529, loss: 1.1101926565170288 2023-01-21 08:56:13.709589: step: 1300/529, loss: 1.4672174453735352 2023-01-21 08:56:14.907800: step: 1304/529, loss: 0.30789339542388916 2023-01-21 08:56:16.071826: step: 1308/529, loss: 0.8448333144187927 2023-01-21 08:56:17.283836: step: 1312/529, loss: 0.7105619311332703 2023-01-21 08:56:18.484087: step: 1316/529, loss: 0.4730682969093323 2023-01-21 08:56:19.718461: step: 1320/529, loss: 0.2151210904121399 2023-01-21 08:56:20.914665: step: 1324/529, loss: 0.7272160649299622 2023-01-21 08:56:22.129903: step: 1328/529, loss: 0.8214249610900879 2023-01-21 08:56:23.319020: step: 1332/529, loss: 0.6123987436294556 2023-01-21 08:56:24.519796: step: 1336/529, loss: 0.25961440801620483 2023-01-21 08:56:25.710170: step: 1340/529, loss: 0.3758208155632019 2023-01-21 08:56:26.915620: step: 1344/529, loss: 0.14473819732666016 2023-01-21 08:56:28.127640: step: 1348/529, loss: 0.18829727172851562 2023-01-21 08:56:29.290396: step: 1352/529, loss: 0.6719884872436523 2023-01-21 08:56:30.515280: step: 1356/529, loss: 0.1795245260000229 2023-01-21 08:56:31.717879: step: 1360/529, loss: 0.8639701008796692 2023-01-21 08:56:32.927173: step: 1364/529, loss: 1.3646658658981323 2023-01-21 08:56:34.118797: step: 1368/529, loss: 0.35387736558914185 2023-01-21 08:56:35.296494: step: 1372/529, loss: 0.682648777961731 2023-01-21 08:56:36.475678: step: 1376/529, loss: 0.27142849564552307 2023-01-21 08:56:37.677721: step: 1380/529, loss: 0.2741974890232086 2023-01-21 08:56:38.864628: step: 1384/529, loss: 0.15044985711574554 2023-01-21 08:56:40.080572: step: 1388/529, loss: 0.5265138745307922 2023-01-21 08:56:41.348164: step: 1392/529, loss: 0.16304011642932892 2023-01-21 08:56:42.552367: step: 1396/529, loss: 0.06310291588306427 2023-01-21 08:56:43.723051: step: 1400/529, loss: 0.17664042115211487 2023-01-21 08:56:44.929177: step: 1404/529, loss: 0.6299788951873779 2023-01-21 08:56:46.109163: step: 1408/529, loss: 0.7734225988388062 2023-01-21 08:56:47.284966: step: 1412/529, loss: 0.49233388900756836 2023-01-21 08:56:48.460919: step: 1416/529, loss: 0.3346596360206604 2023-01-21 08:56:49.657912: step: 1420/529, loss: 4.376696586608887 2023-01-21 08:56:50.871277: step: 1424/529, loss: 0.05406666174530983 2023-01-21 08:56:52.100568: step: 1428/529, loss: 0.24404850602149963 2023-01-21 08:56:53.279234: step: 1432/529, loss: 0.13517819344997406 2023-01-21 08:56:54.438771: step: 1436/529, loss: 0.10635744035243988 2023-01-21 08:56:55.637478: step: 1440/529, loss: 0.11953391879796982 2023-01-21 08:56:56.828562: step: 1444/529, loss: 2.097682237625122 2023-01-21 08:56:58.011811: step: 1448/529, loss: 0.07378740608692169 2023-01-21 08:56:59.264661: step: 1452/529, loss: 0.3262103199958801 2023-01-21 08:57:00.444616: step: 1456/529, loss: 0.13043051958084106 2023-01-21 08:57:01.642816: step: 1460/529, loss: 0.09592628479003906 2023-01-21 08:57:02.827420: step: 1464/529, loss: 0.3350904583930969 2023-01-21 08:57:04.022091: step: 1468/529, loss: 0.7176365852355957 2023-01-21 08:57:05.198434: step: 1472/529, loss: 0.32632073760032654 2023-01-21 08:57:06.369401: step: 1476/529, loss: 0.13802537322044373 2023-01-21 08:57:07.583386: step: 1480/529, loss: 0.2311009019613266 2023-01-21 08:57:08.806544: step: 1484/529, loss: 0.4850480258464813 2023-01-21 08:57:10.019436: step: 1488/529, loss: 0.9704471826553345 2023-01-21 08:57:11.189662: step: 1492/529, loss: 1.374579668045044 2023-01-21 08:57:12.370132: step: 1496/529, loss: 0.9018918871879578 2023-01-21 08:57:13.574336: step: 1500/529, loss: 0.07233548164367676 2023-01-21 08:57:14.790546: step: 1504/529, loss: 0.1383988857269287 2023-01-21 08:57:15.931528: step: 1508/529, loss: 1.6935938596725464 2023-01-21 08:57:17.131887: step: 1512/529, loss: 0.8001202940940857 2023-01-21 08:57:18.300159: step: 1516/529, loss: 0.27359557151794434 2023-01-21 08:57:19.463501: step: 1520/529, loss: 0.4031650722026825 2023-01-21 08:57:20.697696: step: 1524/529, loss: 0.1616169959306717 2023-01-21 08:57:21.891326: step: 1528/529, loss: 0.14935246109962463 2023-01-21 08:57:23.073424: step: 1532/529, loss: 0.8295471668243408 2023-01-21 08:57:24.256131: step: 1536/529, loss: 0.25358933210372925 2023-01-21 08:57:25.450336: step: 1540/529, loss: 0.11138921231031418 2023-01-21 08:57:26.699051: step: 1544/529, loss: 0.17383775115013123 2023-01-21 08:57:27.871425: step: 1548/529, loss: 0.5011741518974304 2023-01-21 08:57:29.071827: step: 1552/529, loss: 0.1803244650363922 2023-01-21 08:57:30.236013: step: 1556/529, loss: 0.9834560751914978 2023-01-21 08:57:31.413865: step: 1560/529, loss: 0.17623892426490784 2023-01-21 08:57:32.645040: step: 1564/529, loss: 0.15104246139526367 2023-01-21 08:57:33.837579: step: 1568/529, loss: 0.19454827904701233 2023-01-21 08:57:35.037555: step: 1572/529, loss: 0.3268570899963379 2023-01-21 08:57:36.210457: step: 1576/529, loss: 0.2185623198747635 2023-01-21 08:57:37.406262: step: 1580/529, loss: 1.2287840843200684 2023-01-21 08:57:38.634418: step: 1584/529, loss: 0.18187841773033142 2023-01-21 08:57:39.815051: step: 1588/529, loss: 0.13649415969848633 2023-01-21 08:57:41.023924: step: 1592/529, loss: 0.799708366394043 2023-01-21 08:57:42.210144: step: 1596/529, loss: 0.2217964082956314 2023-01-21 08:57:43.392180: step: 1600/529, loss: 0.02056713029742241 2023-01-21 08:57:44.590241: step: 1604/529, loss: 1.126817226409912 2023-01-21 08:57:45.805912: step: 1608/529, loss: 0.2664267420768738 2023-01-21 08:57:47.017663: step: 1612/529, loss: 0.7710850238800049 2023-01-21 08:57:48.244141: step: 1616/529, loss: 0.7347552180290222 2023-01-21 08:57:49.468607: step: 1620/529, loss: 0.22203917801380157 2023-01-21 08:57:50.671425: step: 1624/529, loss: 0.07638435810804367 2023-01-21 08:57:51.915608: step: 1628/529, loss: 0.07383427768945694 2023-01-21 08:57:53.154550: step: 1632/529, loss: 0.244424968957901 2023-01-21 08:57:54.361605: step: 1636/529, loss: 0.8498695492744446 2023-01-21 08:57:55.583192: step: 1640/529, loss: 0.369972825050354 2023-01-21 08:57:56.787525: step: 1644/529, loss: 0.5501283407211304 2023-01-21 08:57:58.014425: step: 1648/529, loss: 0.9238247871398926 2023-01-21 08:57:59.202335: step: 1652/529, loss: 0.3014441132545471 2023-01-21 08:58:00.378803: step: 1656/529, loss: 0.10882674157619476 2023-01-21 08:58:01.569868: step: 1660/529, loss: 0.2592444121837616 2023-01-21 08:58:02.789065: step: 1664/529, loss: 0.5186513662338257 2023-01-21 08:58:04.000337: step: 1668/529, loss: 0.3612937927246094 2023-01-21 08:58:05.200809: step: 1672/529, loss: 0.1474134474992752 2023-01-21 08:58:06.459097: step: 1676/529, loss: 0.4401039183139801 2023-01-21 08:58:07.647566: step: 1680/529, loss: 0.2526678442955017 2023-01-21 08:58:08.872289: step: 1684/529, loss: 0.2809045910835266 2023-01-21 08:58:10.090996: step: 1688/529, loss: 1.2654640674591064 2023-01-21 08:58:11.298119: step: 1692/529, loss: 0.22502213716506958 2023-01-21 08:58:12.492407: step: 1696/529, loss: 0.639827311038971 2023-01-21 08:58:13.712548: step: 1700/529, loss: 0.8159974813461304 2023-01-21 08:58:14.945905: step: 1704/529, loss: 0.5960533022880554 2023-01-21 08:58:16.163070: step: 1708/529, loss: 0.1602187156677246 2023-01-21 08:58:17.401029: step: 1712/529, loss: 0.21204166114330292 2023-01-21 08:58:18.638498: step: 1716/529, loss: 0.046489618718624115 2023-01-21 08:58:19.828736: step: 1720/529, loss: 0.8839322924613953 2023-01-21 08:58:21.014111: step: 1724/529, loss: 0.3805540204048157 2023-01-21 08:58:22.183612: step: 1728/529, loss: 0.13201037049293518 2023-01-21 08:58:23.360036: step: 1732/529, loss: 0.10013015568256378 2023-01-21 08:58:24.566829: step: 1736/529, loss: 0.12067604809999466 2023-01-21 08:58:25.781495: step: 1740/529, loss: 0.1981954574584961 2023-01-21 08:58:26.956047: step: 1744/529, loss: 0.15985354781150818 2023-01-21 08:58:28.154669: step: 1748/529, loss: 0.02093651331961155 2023-01-21 08:58:29.344710: step: 1752/529, loss: 1.675818920135498 2023-01-21 08:58:30.566372: step: 1756/529, loss: 0.1608387529850006 2023-01-21 08:58:31.759211: step: 1760/529, loss: 0.6206117272377014 2023-01-21 08:58:32.980312: step: 1764/529, loss: 1.0477927923202515 2023-01-21 08:58:34.232915: step: 1768/529, loss: 0.2547837793827057 2023-01-21 08:58:35.445122: step: 1772/529, loss: 0.8503913879394531 2023-01-21 08:58:36.669597: step: 1776/529, loss: 0.19042368233203888 2023-01-21 08:58:37.860097: step: 1780/529, loss: 0.6307576894760132 2023-01-21 08:58:39.083630: step: 1784/529, loss: 2.148552179336548 2023-01-21 08:58:40.252043: step: 1788/529, loss: 0.8480708003044128 2023-01-21 08:58:41.443370: step: 1792/529, loss: 0.04899773746728897 2023-01-21 08:58:42.642661: step: 1796/529, loss: 0.4089071750640869 2023-01-21 08:58:43.851304: step: 1800/529, loss: 1.2604007720947266 2023-01-21 08:58:45.068553: step: 1804/529, loss: 1.228386640548706 2023-01-21 08:58:46.299162: step: 1808/529, loss: 0.2881526052951813 2023-01-21 08:58:47.504287: step: 1812/529, loss: 0.2277495414018631 2023-01-21 08:58:48.674038: step: 1816/529, loss: 0.9178861379623413 2023-01-21 08:58:49.847151: step: 1820/529, loss: 0.14514775574207306 2023-01-21 08:58:51.033124: step: 1824/529, loss: 0.19144979119300842 2023-01-21 08:58:52.220352: step: 1828/529, loss: 0.2163538932800293 2023-01-21 08:58:53.395451: step: 1832/529, loss: 14.092599868774414 2023-01-21 08:58:54.592512: step: 1836/529, loss: 0.1104976162314415 2023-01-21 08:58:55.763825: step: 1840/529, loss: 1.156100869178772 2023-01-21 08:58:56.921309: step: 1844/529, loss: 0.9208877682685852 2023-01-21 08:58:58.145773: step: 1848/529, loss: 0.5374807119369507 2023-01-21 08:58:59.334018: step: 1852/529, loss: 0.26244187355041504 2023-01-21 08:59:00.554797: step: 1856/529, loss: 0.2787407636642456 2023-01-21 08:59:01.760478: step: 1860/529, loss: 0.26392149925231934 2023-01-21 08:59:03.022869: step: 1864/529, loss: 0.39164304733276367 2023-01-21 08:59:04.205616: step: 1868/529, loss: 0.1400972455739975 2023-01-21 08:59:05.409020: step: 1872/529, loss: 0.0908488780260086 2023-01-21 08:59:06.664000: step: 1876/529, loss: 0.32914039492607117 2023-01-21 08:59:07.834331: step: 1880/529, loss: 0.2589985728263855 2023-01-21 08:59:09.004070: step: 1884/529, loss: 0.12929029762744904 2023-01-21 08:59:10.227499: step: 1888/529, loss: 0.25012731552124023 2023-01-21 08:59:11.393620: step: 1892/529, loss: 0.20882892608642578 2023-01-21 08:59:12.651856: step: 1896/529, loss: 1.1133434772491455 2023-01-21 08:59:13.816745: step: 1900/529, loss: 0.5980363488197327 2023-01-21 08:59:15.059682: step: 1904/529, loss: 0.4736517071723938 2023-01-21 08:59:16.279588: step: 1908/529, loss: 1.8925607204437256 2023-01-21 08:59:17.487734: step: 1912/529, loss: 0.2998947203159332 2023-01-21 08:59:18.699221: step: 1916/529, loss: 0.23366856575012207 2023-01-21 08:59:19.896161: step: 1920/529, loss: 0.279222697019577 2023-01-21 08:59:21.105321: step: 1924/529, loss: 0.21860238909721375 2023-01-21 08:59:22.298460: step: 1928/529, loss: 0.1360282003879547 2023-01-21 08:59:23.474455: step: 1932/529, loss: 0.24403348565101624 2023-01-21 08:59:24.662570: step: 1936/529, loss: 0.21614505350589752 2023-01-21 08:59:25.850716: step: 1940/529, loss: 0.2609427571296692 2023-01-21 08:59:27.078602: step: 1944/529, loss: 0.34300652146339417 2023-01-21 08:59:28.252446: step: 1948/529, loss: 0.49694690108299255 2023-01-21 08:59:29.429276: step: 1952/529, loss: 0.22034835815429688 2023-01-21 08:59:30.628180: step: 1956/529, loss: 0.13345685601234436 2023-01-21 08:59:31.806310: step: 1960/529, loss: 0.26830148696899414 2023-01-21 08:59:32.966991: step: 1964/529, loss: 1.3214449882507324 2023-01-21 08:59:34.174346: step: 1968/529, loss: 0.8397705554962158 2023-01-21 08:59:35.346080: step: 1972/529, loss: 1.6396328210830688 2023-01-21 08:59:36.544594: step: 1976/529, loss: 1.3403518199920654 2023-01-21 08:59:37.763470: step: 1980/529, loss: 0.3076045513153076 2023-01-21 08:59:38.918114: step: 1984/529, loss: 0.2148914337158203 2023-01-21 08:59:40.080715: step: 1988/529, loss: 0.05349435657262802 2023-01-21 08:59:41.295452: step: 1992/529, loss: 0.16458760201931 2023-01-21 08:59:42.473009: step: 1996/529, loss: 0.44762229919433594 2023-01-21 08:59:43.670512: step: 2000/529, loss: 0.32211047410964966 2023-01-21 08:59:44.882339: step: 2004/529, loss: 0.3049635887145996 2023-01-21 08:59:46.108191: step: 2008/529, loss: 1.9458744525909424 2023-01-21 08:59:47.329228: step: 2012/529, loss: 0.5219520330429077 2023-01-21 08:59:48.517299: step: 2016/529, loss: 0.2289627194404602 2023-01-21 08:59:49.716975: step: 2020/529, loss: 0.19515438377857208 2023-01-21 08:59:50.867665: step: 2024/529, loss: 0.8890830874443054 2023-01-21 08:59:52.072278: step: 2028/529, loss: 0.6588819622993469 2023-01-21 08:59:53.232437: step: 2032/529, loss: 0.2609597444534302 2023-01-21 08:59:54.468426: step: 2036/529, loss: 0.5175431370735168 2023-01-21 08:59:55.645243: step: 2040/529, loss: 0.8420371413230896 2023-01-21 08:59:56.879104: step: 2044/529, loss: 1.3401002883911133 2023-01-21 08:59:58.147709: step: 2048/529, loss: 0.9962199926376343 2023-01-21 08:59:59.333011: step: 2052/529, loss: 0.15557122230529785 2023-01-21 09:00:00.514309: step: 2056/529, loss: 0.3028806746006012 2023-01-21 09:00:01.747622: step: 2060/529, loss: 0.18216124176979065 2023-01-21 09:00:02.966554: step: 2064/529, loss: 0.2800995707511902 2023-01-21 09:00:04.189907: step: 2068/529, loss: 0.5926333665847778 2023-01-21 09:00:05.354220: step: 2072/529, loss: 1.703115701675415 2023-01-21 09:00:06.571722: step: 2076/529, loss: 0.30661749839782715 2023-01-21 09:00:07.792500: step: 2080/529, loss: 0.27016106247901917 2023-01-21 09:00:09.006171: step: 2084/529, loss: 0.22211942076683044 2023-01-21 09:00:10.227762: step: 2088/529, loss: 0.6704018115997314 2023-01-21 09:00:11.461613: step: 2092/529, loss: 0.8512510061264038 2023-01-21 09:00:12.695615: step: 2096/529, loss: 0.4487777054309845 2023-01-21 09:00:13.876925: step: 2100/529, loss: 0.667074978351593 2023-01-21 09:00:15.104028: step: 2104/529, loss: 0.19050435721874237 2023-01-21 09:00:16.291144: step: 2108/529, loss: 0.6942664384841919 2023-01-21 09:00:17.461477: step: 2112/529, loss: 0.518520176410675 2023-01-21 09:00:18.667698: step: 2116/529, loss: 0.5513672828674316 ================================================== Loss: 0.767 -------------------- Dev: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Test: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Chinese: {'event': {'p': 0.4675324675324675, 'r': 0.6666666666666666, 'f1': 0.549618320610687}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Korean: {'event': {'p': 0.6904761904761905, 'r': 0.4603174603174603, 'f1': 0.5523809523809524}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, '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': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Chinese: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Chinese: {'event': {'p': 0.4675324675324675, 'r': 0.6666666666666666, 'f1': 0.549618320610687}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Korean: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.6904761904761905, 'r': 0.4603174603174603, 'f1': 0.5523809523809524}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, '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 09:01:27.329625: step: 4/529, loss: 0.3145027756690979 2023-01-21 09:01:28.514892: step: 8/529, loss: 0.3824879825115204 2023-01-21 09:01:29.706757: step: 12/529, loss: 0.21834233403205872 2023-01-21 09:01:30.879298: step: 16/529, loss: 0.1918337345123291 2023-01-21 09:01:32.041920: step: 20/529, loss: 0.32676324248313904 2023-01-21 09:01:33.271975: step: 24/529, loss: 0.3381422162055969 2023-01-21 09:01:34.469435: step: 28/529, loss: 0.19183596968650818 2023-01-21 09:01:35.658585: step: 32/529, loss: 0.213566392660141 2023-01-21 09:01:36.842795: step: 36/529, loss: 0.1891981065273285 2023-01-21 09:01:38.004647: step: 40/529, loss: 0.2683035731315613 2023-01-21 09:01:39.215089: step: 44/529, loss: 0.21231085062026978 2023-01-21 09:01:40.402532: step: 48/529, loss: 0.7255871295928955 2023-01-21 09:01:41.583207: step: 52/529, loss: 0.6977856755256653 2023-01-21 09:01:42.759575: step: 56/529, loss: 0.4053492546081543 2023-01-21 09:01:43.960099: step: 60/529, loss: 0.2516906261444092 2023-01-21 09:01:45.166439: step: 64/529, loss: 0.19978752732276917 2023-01-21 09:01:46.407167: step: 68/529, loss: 0.7886441946029663 2023-01-21 09:01:47.570130: step: 72/529, loss: 1.855760097503662 2023-01-21 09:01:48.751090: step: 76/529, loss: 0.18097572028636932 2023-01-21 09:01:49.924533: step: 80/529, loss: 1.7791774272918701 2023-01-21 09:01:51.113098: step: 84/529, loss: 0.48580026626586914 2023-01-21 09:01:52.334085: step: 88/529, loss: 6.689284324645996 2023-01-21 09:01:53.537943: step: 92/529, loss: 0.10029307007789612 2023-01-21 09:01:54.714880: step: 96/529, loss: 0.211774542927742 2023-01-21 09:01:55.964359: step: 100/529, loss: 0.18067194521427155 2023-01-21 09:01:57.174951: step: 104/529, loss: 0.22729197144508362 2023-01-21 09:01:58.387343: step: 108/529, loss: 0.13150405883789062 2023-01-21 09:01:59.539449: step: 112/529, loss: 0.8847576379776001 2023-01-21 09:02:00.776865: step: 116/529, loss: 0.08883801102638245 2023-01-21 09:02:01.945709: step: 120/529, loss: 0.6762876510620117 2023-01-21 09:02:03.118180: step: 124/529, loss: 0.39829427003860474 2023-01-21 09:02:04.308355: step: 128/529, loss: 0.04410243034362793 2023-01-21 09:02:05.461040: step: 132/529, loss: 0.7645156979560852 2023-01-21 09:02:06.702881: step: 136/529, loss: 0.6208375692367554 2023-01-21 09:02:07.904594: step: 140/529, loss: 0.45884692668914795 2023-01-21 09:02:09.089213: step: 144/529, loss: 1.2470149993896484 2023-01-21 09:02:10.259506: step: 148/529, loss: 0.28963661193847656 2023-01-21 09:02:11.477032: step: 152/529, loss: 0.3370674252510071 2023-01-21 09:02:12.649359: step: 156/529, loss: 1.81912362575531 2023-01-21 09:02:13.852788: step: 160/529, loss: 0.46266573667526245 2023-01-21 09:02:15.022700: step: 164/529, loss: 0.2175680249929428 2023-01-21 09:02:16.229969: step: 168/529, loss: 0.26922425627708435 2023-01-21 09:02:17.450554: step: 172/529, loss: 0.41409599781036377 2023-01-21 09:02:18.656043: step: 176/529, loss: 0.1715511977672577 2023-01-21 09:02:19.872791: step: 180/529, loss: 0.7273315191268921 2023-01-21 09:02:21.052821: step: 184/529, loss: 0.2191057801246643 2023-01-21 09:02:22.234662: step: 188/529, loss: 0.586996853351593 2023-01-21 09:02:23.432051: step: 192/529, loss: 0.15371084213256836 2023-01-21 09:02:24.631439: step: 196/529, loss: 0.49319761991500854 2023-01-21 09:02:25.786118: step: 200/529, loss: 6.446598052978516 2023-01-21 09:02:26.960423: step: 204/529, loss: 0.2858756482601166 2023-01-21 09:02:28.143440: step: 208/529, loss: 0.17778797447681427 2023-01-21 09:02:29.388114: step: 212/529, loss: 0.9940707087516785 2023-01-21 09:02:30.577743: step: 216/529, loss: 0.21211454272270203 2023-01-21 09:02:31.758185: step: 220/529, loss: 0.6821246147155762 2023-01-21 09:02:32.984520: step: 224/529, loss: 0.37775611877441406 2023-01-21 09:02:34.160796: step: 228/529, loss: 0.1789262741804123 2023-01-21 09:02:35.353464: step: 232/529, loss: 0.6681526303291321 2023-01-21 09:02:36.560252: step: 236/529, loss: 0.16765213012695312 2023-01-21 09:02:37.739104: step: 240/529, loss: 0.18701642751693726 2023-01-21 09:02:38.923589: step: 244/529, loss: 0.18661737442016602 2023-01-21 09:02:40.146232: step: 248/529, loss: 0.38548851013183594 2023-01-21 09:02:41.344538: step: 252/529, loss: 0.7864240407943726 2023-01-21 09:02:42.508785: step: 256/529, loss: 0.1659766286611557 2023-01-21 09:02:43.699426: step: 260/529, loss: 0.7087134122848511 2023-01-21 09:02:44.947857: step: 264/529, loss: 6.761756896972656 2023-01-21 09:02:46.147495: step: 268/529, loss: 0.04154014587402344 2023-01-21 09:02:47.330523: step: 272/529, loss: 0.18368521332740784 2023-01-21 09:02:48.526577: step: 276/529, loss: 0.35539257526397705 2023-01-21 09:02:49.711888: step: 280/529, loss: 0.6105610728263855 2023-01-21 09:02:50.903334: step: 284/529, loss: 0.24418431520462036 2023-01-21 09:02:52.104565: step: 288/529, loss: 0.42339026927948 2023-01-21 09:02:53.326398: step: 292/529, loss: 0.11974182724952698 2023-01-21 09:02:54.509088: step: 296/529, loss: 0.14005470275878906 2023-01-21 09:02:55.717408: step: 300/529, loss: 1.0959142446517944 2023-01-21 09:02:56.913167: step: 304/529, loss: 0.37148499488830566 2023-01-21 09:02:58.104451: step: 308/529, loss: 0.7891260981559753 2023-01-21 09:02:59.378865: step: 312/529, loss: 1.8357021808624268 2023-01-21 09:03:00.535219: step: 316/529, loss: 0.08673782646656036 2023-01-21 09:03:01.854334: step: 320/529, loss: 2.700627326965332 2023-01-21 09:03:03.052456: step: 324/529, loss: 0.08230238407850266 2023-01-21 09:03:04.250864: step: 328/529, loss: 0.3465553820133209 2023-01-21 09:03:05.413624: step: 332/529, loss: 0.343706876039505 2023-01-21 09:03:06.587706: step: 336/529, loss: 0.05547685548663139 2023-01-21 09:03:07.764977: step: 340/529, loss: 0.4214421212673187 2023-01-21 09:03:08.923757: step: 344/529, loss: 0.07134180516004562 2023-01-21 09:03:10.109628: step: 348/529, loss: 0.23021790385246277 2023-01-21 09:03:11.298289: step: 352/529, loss: 0.15429440140724182 2023-01-21 09:03:12.549241: step: 356/529, loss: 0.21846051514148712 2023-01-21 09:03:13.752101: step: 360/529, loss: 0.07128031551837921 2023-01-21 09:03:14.925566: step: 364/529, loss: 0.2953076958656311 2023-01-21 09:03:16.120530: step: 368/529, loss: 0.6530542373657227 2023-01-21 09:03:17.307312: step: 372/529, loss: 0.4478347897529602 2023-01-21 09:03:18.505670: step: 376/529, loss: 0.24450521171092987 2023-01-21 09:03:19.712616: step: 380/529, loss: 0.3753108084201813 2023-01-21 09:03:20.871134: step: 384/529, loss: 0.11954517662525177 2023-01-21 09:03:22.035127: step: 388/529, loss: 0.38516682386398315 2023-01-21 09:03:23.198319: step: 392/529, loss: 0.2172565460205078 2023-01-21 09:03:24.405520: step: 396/529, loss: 0.24366196990013123 2023-01-21 09:03:25.614867: step: 400/529, loss: 0.1932273805141449 2023-01-21 09:03:26.884690: step: 404/529, loss: 1.0306894779205322 2023-01-21 09:03:28.045901: step: 408/529, loss: 1.2246931791305542 2023-01-21 09:03:29.219400: step: 412/529, loss: 0.15378323197364807 2023-01-21 09:03:30.454587: step: 416/529, loss: 0.3526252806186676 2023-01-21 09:03:31.626717: step: 420/529, loss: 0.5678306818008423 2023-01-21 09:03:32.846179: step: 424/529, loss: 0.41032248735427856 2023-01-21 09:03:34.022312: step: 428/529, loss: 0.8143123984336853 2023-01-21 09:03:35.226972: step: 432/529, loss: 0.16842976212501526 2023-01-21 09:03:36.450707: step: 436/529, loss: 0.09926962107419968 2023-01-21 09:03:37.651187: step: 440/529, loss: 0.09357257187366486 2023-01-21 09:03:38.827233: step: 444/529, loss: 0.17700693011283875 2023-01-21 09:03:40.051000: step: 448/529, loss: 1.9232968091964722 2023-01-21 09:03:41.206212: step: 452/529, loss: 0.11117162555456161 2023-01-21 09:03:42.405855: step: 456/529, loss: 0.17162255942821503 2023-01-21 09:03:43.635811: step: 460/529, loss: 0.8817241787910461 2023-01-21 09:03:44.846427: step: 464/529, loss: 0.6451841592788696 2023-01-21 09:03:46.070785: step: 468/529, loss: 0.6476221084594727 2023-01-21 09:03:47.335057: step: 472/529, loss: 0.397754967212677 2023-01-21 09:03:48.528792: step: 476/529, loss: 0.2320270538330078 2023-01-21 09:03:49.722174: step: 480/529, loss: 0.32145872712135315 2023-01-21 09:03:50.918831: step: 484/529, loss: 1.3022773265838623 2023-01-21 09:03:52.118569: step: 488/529, loss: 0.18829499185085297 2023-01-21 09:03:53.326762: step: 492/529, loss: 0.15562638640403748 2023-01-21 09:03:54.509141: step: 496/529, loss: 0.17442092299461365 2023-01-21 09:03:55.734852: step: 500/529, loss: 0.24912777543067932 2023-01-21 09:03:56.982238: step: 504/529, loss: 0.8057572245597839 2023-01-21 09:03:58.151658: step: 508/529, loss: 0.7955676317214966 2023-01-21 09:03:59.327350: step: 512/529, loss: 0.32331976294517517 2023-01-21 09:04:00.554114: step: 516/529, loss: 0.6123404502868652 2023-01-21 09:04:01.746887: step: 520/529, loss: 0.1517670750617981 2023-01-21 09:04:02.941672: step: 524/529, loss: 0.6838817000389099 2023-01-21 09:04:04.171598: step: 528/529, loss: 0.21943378448486328 2023-01-21 09:04:05.390407: step: 532/529, loss: 0.38667726516723633 2023-01-21 09:04:06.581868: step: 536/529, loss: 0.0918995812535286 2023-01-21 09:04:07.783557: step: 540/529, loss: 0.1284637451171875 2023-01-21 09:04:09.024117: step: 544/529, loss: 0.6208086609840393 2023-01-21 09:04:10.242937: step: 548/529, loss: 0.08790130913257599 2023-01-21 09:04:11.422180: step: 552/529, loss: 0.7548690438270569 2023-01-21 09:04:12.637928: step: 556/529, loss: 1.6029006242752075 2023-01-21 09:04:13.828120: step: 560/529, loss: 0.2915436923503876 2023-01-21 09:04:14.993385: step: 564/529, loss: 0.08351273834705353 2023-01-21 09:04:16.236288: step: 568/529, loss: 0.9850178360939026 2023-01-21 09:04:17.451080: step: 572/529, loss: 0.07333894073963165 2023-01-21 09:04:18.635691: step: 576/529, loss: 1.386325478553772 2023-01-21 09:04:19.882400: step: 580/529, loss: 0.22081860899925232 2023-01-21 09:04:21.077179: step: 584/529, loss: 0.1628166139125824 2023-01-21 09:04:22.331822: step: 588/529, loss: 0.21540433168411255 2023-01-21 09:04:23.555554: step: 592/529, loss: 0.06566419452428818 2023-01-21 09:04:24.777871: step: 596/529, loss: 0.6674219369888306 2023-01-21 09:04:25.967827: step: 600/529, loss: 6.23017692565918 2023-01-21 09:04:27.207885: step: 604/529, loss: 0.8146122097969055 2023-01-21 09:04:28.379003: step: 608/529, loss: 0.05529747158288956 2023-01-21 09:04:29.537340: step: 612/529, loss: 0.23919087648391724 2023-01-21 09:04:30.741637: step: 616/529, loss: 0.2634842097759247 2023-01-21 09:04:31.944280: step: 620/529, loss: 0.6310187578201294 2023-01-21 09:04:33.129193: step: 624/529, loss: 0.6703919172286987 2023-01-21 09:04:34.303261: step: 628/529, loss: 1.0428012609481812 2023-01-21 09:04:35.489254: step: 632/529, loss: 2.3234193325042725 2023-01-21 09:04:36.686106: step: 636/529, loss: 0.3630928099155426 2023-01-21 09:04:37.905717: step: 640/529, loss: 0.33125293254852295 2023-01-21 09:04:39.094076: step: 644/529, loss: 0.2506866157054901 2023-01-21 09:04:40.295922: step: 648/529, loss: 1.1564433574676514 2023-01-21 09:04:41.576263: step: 652/529, loss: 0.18596597015857697 2023-01-21 09:04:42.748430: step: 656/529, loss: 0.1827954351902008 2023-01-21 09:04:43.973904: step: 660/529, loss: 0.6522228717803955 2023-01-21 09:04:45.154554: step: 664/529, loss: 0.11323872208595276 2023-01-21 09:04:46.377895: step: 668/529, loss: 0.4669274389743805 2023-01-21 09:04:47.625396: step: 672/529, loss: 0.8580198287963867 2023-01-21 09:04:48.867407: step: 676/529, loss: 0.6306808590888977 2023-01-21 09:04:50.046903: step: 680/529, loss: 0.6906743049621582 2023-01-21 09:04:51.277717: step: 684/529, loss: 0.2813708186149597 2023-01-21 09:04:52.537048: step: 688/529, loss: 0.22030572593212128 2023-01-21 09:04:53.750815: step: 692/529, loss: 0.22473736107349396 2023-01-21 09:04:54.914030: step: 696/529, loss: 0.9583730101585388 2023-01-21 09:04:56.113522: step: 700/529, loss: 0.10873080044984818 2023-01-21 09:04:57.297791: step: 704/529, loss: 0.21155062317848206 2023-01-21 09:04:58.495879: step: 708/529, loss: 6.62742280960083 2023-01-21 09:04:59.727781: step: 712/529, loss: 0.6515638828277588 2023-01-21 09:05:00.946986: step: 716/529, loss: 0.2883855998516083 2023-01-21 09:05:02.139532: step: 720/529, loss: 0.27828481793403625 2023-01-21 09:05:03.372495: step: 724/529, loss: 1.0037860870361328 2023-01-21 09:05:04.614006: step: 728/529, loss: 0.7955505847930908 2023-01-21 09:05:05.799047: step: 732/529, loss: 0.11061549186706543 2023-01-21 09:05:07.044557: step: 736/529, loss: 0.6887736916542053 2023-01-21 09:05:08.273540: step: 740/529, loss: 0.31038370728492737 2023-01-21 09:05:09.479364: step: 744/529, loss: 0.7637366056442261 2023-01-21 09:05:10.670558: step: 748/529, loss: 0.6992408633232117 2023-01-21 09:05:11.860325: step: 752/529, loss: 0.23954148590564728 2023-01-21 09:05:13.085045: step: 756/529, loss: 0.7993958592414856 2023-01-21 09:05:14.283029: step: 760/529, loss: 0.100660540163517 2023-01-21 09:05:15.468916: step: 764/529, loss: 0.20956745743751526 2023-01-21 09:05:16.759937: step: 768/529, loss: 0.34549540281295776 2023-01-21 09:05:17.968070: step: 772/529, loss: 0.3190248906612396 2023-01-21 09:05:19.132076: step: 776/529, loss: 0.07983627915382385 2023-01-21 09:05:20.309866: step: 780/529, loss: 0.18287816643714905 2023-01-21 09:05:21.499656: step: 784/529, loss: 0.1891632080078125 2023-01-21 09:05:22.680814: step: 788/529, loss: 0.7350719571113586 2023-01-21 09:05:23.876114: step: 792/529, loss: 1.3146165609359741 2023-01-21 09:05:25.038543: step: 796/529, loss: 0.6332480311393738 2023-01-21 09:05:26.216632: step: 800/529, loss: 0.19985853135585785 2023-01-21 09:05:27.408832: step: 804/529, loss: 0.3407343029975891 2023-01-21 09:05:28.570429: step: 808/529, loss: 0.09236927330493927 2023-01-21 09:05:29.784817: step: 812/529, loss: 1.0310064554214478 2023-01-21 09:05:30.947603: step: 816/529, loss: 0.49313926696777344 2023-01-21 09:05:32.186616: step: 820/529, loss: 0.8960789442062378 2023-01-21 09:05:33.381229: step: 824/529, loss: 0.22117196023464203 2023-01-21 09:05:34.563740: step: 828/529, loss: 0.5449997186660767 2023-01-21 09:05:35.758565: step: 832/529, loss: 0.3623240888118744 2023-01-21 09:05:36.980525: step: 836/529, loss: 0.3891538381576538 2023-01-21 09:05:38.193843: step: 840/529, loss: 0.20476743578910828 2023-01-21 09:05:39.446648: step: 844/529, loss: 0.321729838848114 2023-01-21 09:05:40.652154: step: 848/529, loss: 0.951215386390686 2023-01-21 09:05:41.852926: step: 852/529, loss: 0.18274226784706116 2023-01-21 09:05:43.075348: step: 856/529, loss: 0.6644158363342285 2023-01-21 09:05:44.237405: step: 860/529, loss: 0.230614572763443 2023-01-21 09:05:45.448024: step: 864/529, loss: 1.8907235860824585 2023-01-21 09:05:46.681755: step: 868/529, loss: 0.32483169436454773 2023-01-21 09:05:47.904118: step: 872/529, loss: 0.23649196326732635 2023-01-21 09:05:49.099314: step: 876/529, loss: 0.37153130769729614 2023-01-21 09:05:50.280851: step: 880/529, loss: 1.2816922664642334 2023-01-21 09:05:51.487327: step: 884/529, loss: 0.22412481904029846 2023-01-21 09:05:52.722006: step: 888/529, loss: 0.6680564880371094 2023-01-21 09:05:53.901483: step: 892/529, loss: 0.17364007234573364 2023-01-21 09:05:55.109369: step: 896/529, loss: 0.26113802194595337 2023-01-21 09:05:56.299126: step: 900/529, loss: 0.11926255375146866 2023-01-21 09:05:57.470637: step: 904/529, loss: 0.9900862574577332 2023-01-21 09:05:58.637705: step: 908/529, loss: 0.10782718658447266 2023-01-21 09:05:59.821863: step: 912/529, loss: 0.8222464323043823 2023-01-21 09:06:01.016659: step: 916/529, loss: 0.11105190217494965 2023-01-21 09:06:02.226396: step: 920/529, loss: 0.06733784824609756 2023-01-21 09:06:03.461448: step: 924/529, loss: 0.2141837179660797 2023-01-21 09:06:04.657632: step: 928/529, loss: 0.2782951295375824 2023-01-21 09:06:05.830329: step: 932/529, loss: 0.3528013229370117 2023-01-21 09:06:07.005206: step: 936/529, loss: 0.3250707983970642 2023-01-21 09:06:08.206922: step: 940/529, loss: 0.7469584941864014 2023-01-21 09:06:09.383340: step: 944/529, loss: 0.44888103008270264 2023-01-21 09:06:10.570375: step: 948/529, loss: 0.0593172088265419 2023-01-21 09:06:11.763923: step: 952/529, loss: 0.12156205624341965 2023-01-21 09:06:12.979422: step: 956/529, loss: 0.17953495681285858 2023-01-21 09:06:14.199242: step: 960/529, loss: 0.8286677598953247 2023-01-21 09:06:15.432174: step: 964/529, loss: 0.7979264855384827 2023-01-21 09:06:16.628323: step: 968/529, loss: 0.4452919065952301 2023-01-21 09:06:17.812292: step: 972/529, loss: 0.06292891502380371 2023-01-21 09:06:18.983422: step: 976/529, loss: 0.234127476811409 2023-01-21 09:06:20.175416: step: 980/529, loss: 0.3322582244873047 2023-01-21 09:06:21.339485: step: 984/529, loss: 0.6477928757667542 2023-01-21 09:06:22.521671: step: 988/529, loss: 0.18503180146217346 2023-01-21 09:06:23.705180: step: 992/529, loss: 0.4200851619243622 2023-01-21 09:06:24.922762: step: 996/529, loss: 0.8120313882827759 2023-01-21 09:06:26.126952: step: 1000/529, loss: 0.17442971467971802 2023-01-21 09:06:27.325636: step: 1004/529, loss: 0.23557721078395844 2023-01-21 09:06:28.556110: step: 1008/529, loss: 1.2806655168533325 2023-01-21 09:06:29.720552: step: 1012/529, loss: 0.8511902689933777 2023-01-21 09:06:30.892235: step: 1016/529, loss: 0.11861763149499893 2023-01-21 09:06:32.145491: step: 1020/529, loss: 0.07488727569580078 2023-01-21 09:06:33.385905: step: 1024/529, loss: 0.22666026651859283 2023-01-21 09:06:34.599595: step: 1028/529, loss: 0.18276947736740112 2023-01-21 09:06:35.778124: step: 1032/529, loss: 0.20344991981983185 2023-01-21 09:06:36.936555: step: 1036/529, loss: 0.7284892797470093 2023-01-21 09:06:38.146070: step: 1040/529, loss: 0.1279032826423645 2023-01-21 09:06:39.380548: step: 1044/529, loss: 1.3013429641723633 2023-01-21 09:06:40.578405: step: 1048/529, loss: 0.4718008041381836 2023-01-21 09:06:41.808116: step: 1052/529, loss: 0.5408033132553101 2023-01-21 09:06:42.982618: step: 1056/529, loss: 0.15192851424217224 2023-01-21 09:06:44.162078: step: 1060/529, loss: 0.13008061051368713 2023-01-21 09:06:45.371940: step: 1064/529, loss: 0.4026821255683899 2023-01-21 09:06:46.569204: step: 1068/529, loss: 0.14857178926467896 2023-01-21 09:06:47.794864: step: 1072/529, loss: 0.11449375003576279 2023-01-21 09:06:48.979245: step: 1076/529, loss: 0.27501630783081055 2023-01-21 09:06:50.191444: step: 1080/529, loss: 0.9491043090820312 2023-01-21 09:06:51.389189: step: 1084/529, loss: 0.6536704897880554 2023-01-21 09:06:52.575078: step: 1088/529, loss: 0.2978436350822449 2023-01-21 09:06:53.760656: step: 1092/529, loss: 0.16854123771190643 2023-01-21 09:06:54.989782: step: 1096/529, loss: 0.21513214707374573 2023-01-21 09:06:56.218452: step: 1100/529, loss: 0.9925499558448792 2023-01-21 09:06:57.422828: step: 1104/529, loss: 0.10279889404773712 2023-01-21 09:06:58.583062: step: 1108/529, loss: 0.241289421916008 2023-01-21 09:06:59.764418: step: 1112/529, loss: 0.271989643573761 2023-01-21 09:07:00.957523: step: 1116/529, loss: 0.27203720808029175 2023-01-21 09:07:02.153321: step: 1120/529, loss: 0.06574294716119766 2023-01-21 09:07:03.323524: step: 1124/529, loss: 1.1330227851867676 2023-01-21 09:07:04.580691: step: 1128/529, loss: 0.17437037825584412 2023-01-21 09:07:05.821472: step: 1132/529, loss: 0.13036957383155823 2023-01-21 09:07:07.027766: step: 1136/529, loss: 0.23241043090820312 2023-01-21 09:07:08.224080: step: 1140/529, loss: 1.4849367141723633 2023-01-21 09:07:09.439763: step: 1144/529, loss: 0.22056794166564941 2023-01-21 09:07:10.675060: step: 1148/529, loss: 0.225905179977417 2023-01-21 09:07:11.886099: step: 1152/529, loss: 0.2188694030046463 2023-01-21 09:07:13.097835: step: 1156/529, loss: 0.38801446557044983 2023-01-21 09:07:14.298363: step: 1160/529, loss: 0.3670467734336853 2023-01-21 09:07:15.529669: step: 1164/529, loss: 7.1593804359436035 2023-01-21 09:07:16.741790: step: 1168/529, loss: 0.3402881622314453 2023-01-21 09:07:17.927827: step: 1172/529, loss: 0.3054807782173157 2023-01-21 09:07:19.118780: step: 1176/529, loss: 0.6274528503417969 2023-01-21 09:07:20.267983: step: 1180/529, loss: 0.7883661985397339 2023-01-21 09:07:21.438665: step: 1184/529, loss: 0.10384626686573029 2023-01-21 09:07:22.646474: step: 1188/529, loss: 1.4313936233520508 2023-01-21 09:07:23.792191: step: 1192/529, loss: 0.25158172845840454 2023-01-21 09:07:25.008871: step: 1196/529, loss: 0.28510063886642456 2023-01-21 09:07:26.153000: step: 1200/529, loss: 0.10683298856019974 2023-01-21 09:07:27.374250: step: 1204/529, loss: 1.1170963048934937 2023-01-21 09:07:28.562764: step: 1208/529, loss: 0.19515009224414825 2023-01-21 09:07:29.802414: step: 1212/529, loss: 1.0336393117904663 2023-01-21 09:07:30.992114: step: 1216/529, loss: 3.859346389770508 2023-01-21 09:07:32.233302: step: 1220/529, loss: 0.31862545013427734 2023-01-21 09:07:33.422405: step: 1224/529, loss: 0.1679462492465973 2023-01-21 09:07:34.618595: step: 1228/529, loss: 0.10264472663402557 2023-01-21 09:07:35.781411: step: 1232/529, loss: 1.085010290145874 2023-01-21 09:07:36.994865: step: 1236/529, loss: 0.24031372368335724 2023-01-21 09:07:38.252897: step: 1240/529, loss: 0.3001592755317688 2023-01-21 09:07:39.487222: step: 1244/529, loss: 3.470602512359619 2023-01-21 09:07:40.683608: step: 1248/529, loss: 0.23282481729984283 2023-01-21 09:07:41.890886: step: 1252/529, loss: 0.0997127816081047 2023-01-21 09:07:43.096773: step: 1256/529, loss: 0.059206150472164154 2023-01-21 09:07:44.300466: step: 1260/529, loss: 7.076552867889404 2023-01-21 09:07:45.500701: step: 1264/529, loss: 0.33907976746559143 2023-01-21 09:07:46.729027: step: 1268/529, loss: 0.6730745434761047 2023-01-21 09:07:47.939566: step: 1272/529, loss: 0.12665286660194397 2023-01-21 09:07:49.179746: step: 1276/529, loss: 0.3525141477584839 2023-01-21 09:07:50.351342: step: 1280/529, loss: 0.875916600227356 2023-01-21 09:07:51.537041: step: 1284/529, loss: 0.21333380043506622 2023-01-21 09:07:52.745402: step: 1288/529, loss: 0.1325697898864746 2023-01-21 09:07:53.939678: step: 1292/529, loss: 0.5195043683052063 2023-01-21 09:07:55.183340: step: 1296/529, loss: 0.16322728991508484 2023-01-21 09:07:56.376961: step: 1300/529, loss: 0.162730410695076 2023-01-21 09:07:57.570936: step: 1304/529, loss: 0.7093981504440308 2023-01-21 09:07:58.751052: step: 1308/529, loss: 0.3268682360649109 2023-01-21 09:07:59.955423: step: 1312/529, loss: 0.08718939125537872 2023-01-21 09:08:01.166756: step: 1316/529, loss: 6.731115818023682 2023-01-21 09:08:02.409000: step: 1320/529, loss: 2.189631223678589 2023-01-21 09:08:03.594956: step: 1324/529, loss: 0.12101832032203674 2023-01-21 09:08:04.778971: step: 1328/529, loss: 0.8480373620986938 2023-01-21 09:08:06.002991: step: 1332/529, loss: 0.26181966066360474 2023-01-21 09:08:07.173971: step: 1336/529, loss: 0.2927878499031067 2023-01-21 09:08:08.394143: step: 1340/529, loss: 0.37417975068092346 2023-01-21 09:08:09.635199: step: 1344/529, loss: 0.11520853638648987 2023-01-21 09:08:10.809574: step: 1348/529, loss: 0.07925625145435333 2023-01-21 09:08:12.002599: step: 1352/529, loss: 0.0692962259054184 2023-01-21 09:08:13.212779: step: 1356/529, loss: 0.9576336741447449 2023-01-21 09:08:14.425256: step: 1360/529, loss: 0.7248705625534058 2023-01-21 09:08:15.658178: step: 1364/529, loss: 0.4641962945461273 2023-01-21 09:08:16.874603: step: 1368/529, loss: 0.5704681277275085 2023-01-21 09:08:18.099512: step: 1372/529, loss: 0.4670063257217407 2023-01-21 09:08:19.314734: step: 1376/529, loss: 0.16036167740821838 2023-01-21 09:08:20.486566: step: 1380/529, loss: 0.20905956625938416 2023-01-21 09:08:21.731938: step: 1384/529, loss: 0.49120253324508667 2023-01-21 09:08:22.882757: step: 1388/529, loss: 0.8013932704925537 2023-01-21 09:08:24.091143: step: 1392/529, loss: 0.23077034950256348 2023-01-21 09:08:25.308549: step: 1396/529, loss: 0.15353356301784515 2023-01-21 09:08:26.494453: step: 1400/529, loss: 0.5879707932472229 2023-01-21 09:08:27.692248: step: 1404/529, loss: 0.18980297446250916 2023-01-21 09:08:28.925271: step: 1408/529, loss: 0.3202105760574341 2023-01-21 09:08:30.093123: step: 1412/529, loss: 1.3615949153900146 2023-01-21 09:08:31.260714: step: 1416/529, loss: 0.0819651186466217 2023-01-21 09:08:32.468668: step: 1420/529, loss: 0.12039222568273544 2023-01-21 09:08:33.709448: step: 1424/529, loss: 0.6147060394287109 2023-01-21 09:08:34.902438: step: 1428/529, loss: 0.393871009349823 2023-01-21 09:08:36.110516: step: 1432/529, loss: 0.27766990661621094 2023-01-21 09:08:37.325547: step: 1436/529, loss: 0.7187163233757019 2023-01-21 09:08:38.589554: step: 1440/529, loss: 0.23915091156959534 2023-01-21 09:08:39.766027: step: 1444/529, loss: 6.786542892456055 2023-01-21 09:08:40.978417: step: 1448/529, loss: 0.32126158475875854 2023-01-21 09:08:42.175914: step: 1452/529, loss: 0.6184698939323425 2023-01-21 09:08:43.363175: step: 1456/529, loss: 0.20836858451366425 2023-01-21 09:08:44.541801: step: 1460/529, loss: 0.11626176536083221 2023-01-21 09:08:45.732849: step: 1464/529, loss: 0.09093180298805237 2023-01-21 09:08:46.955483: step: 1468/529, loss: 0.35342901945114136 2023-01-21 09:08:48.127760: step: 1472/529, loss: 0.14807267487049103 2023-01-21 09:08:49.318154: step: 1476/529, loss: 0.935273289680481 2023-01-21 09:08:50.470884: step: 1480/529, loss: 0.9925022125244141 2023-01-21 09:08:51.651277: step: 1484/529, loss: 0.5748090744018555 2023-01-21 09:08:52.856606: step: 1488/529, loss: 0.1639421582221985 2023-01-21 09:08:54.063241: step: 1492/529, loss: 0.24910077452659607 2023-01-21 09:08:55.268712: step: 1496/529, loss: 0.11983104050159454 2023-01-21 09:08:56.472137: step: 1500/529, loss: 0.2815249562263489 2023-01-21 09:08:57.698286: step: 1504/529, loss: 0.10525989532470703 2023-01-21 09:08:58.872247: step: 1508/529, loss: 0.08625535666942596 2023-01-21 09:09:00.057578: step: 1512/529, loss: 0.7407762408256531 2023-01-21 09:09:01.297157: step: 1516/529, loss: 6.597129821777344 2023-01-21 09:09:02.510792: step: 1520/529, loss: 0.4166269898414612 2023-01-21 09:09:03.809922: step: 1524/529, loss: 0.17616614699363708 2023-01-21 09:09:05.066191: step: 1528/529, loss: 0.15646935999393463 2023-01-21 09:09:06.278233: step: 1532/529, loss: 0.7311524152755737 2023-01-21 09:09:07.466239: step: 1536/529, loss: 0.2112540304660797 2023-01-21 09:09:08.674704: step: 1540/529, loss: 0.18518924713134766 2023-01-21 09:09:09.883517: step: 1544/529, loss: 0.3983272612094879 2023-01-21 09:09:11.098272: step: 1548/529, loss: 0.17392587661743164 2023-01-21 09:09:12.287896: step: 1552/529, loss: 0.10151156783103943 2023-01-21 09:09:13.575774: step: 1556/529, loss: 0.23852606117725372 2023-01-21 09:09:14.754103: step: 1560/529, loss: 0.25653284788131714 2023-01-21 09:09:15.980752: step: 1564/529, loss: 0.39179837703704834 2023-01-21 09:09:17.186387: step: 1568/529, loss: 0.11140461266040802 2023-01-21 09:09:18.434134: step: 1572/529, loss: 0.9269602298736572 2023-01-21 09:09:19.629154: step: 1576/529, loss: 0.8949348330497742 2023-01-21 09:09:20.862581: step: 1580/529, loss: 0.18786931037902832 2023-01-21 09:09:22.072536: step: 1584/529, loss: 0.9753708243370056 2023-01-21 09:09:23.244636: step: 1588/529, loss: 0.06753573566675186 2023-01-21 09:09:24.432244: step: 1592/529, loss: 0.8378344774246216 2023-01-21 09:09:25.597668: step: 1596/529, loss: 0.12011952698230743 2023-01-21 09:09:26.790420: step: 1600/529, loss: 0.243224635720253 2023-01-21 09:09:28.013310: step: 1604/529, loss: 0.04439534991979599 2023-01-21 09:09:29.202283: step: 1608/529, loss: 0.32140398025512695 2023-01-21 09:09:30.376545: step: 1612/529, loss: 0.9675464034080505 2023-01-21 09:09:31.628498: step: 1616/529, loss: 0.8656686544418335 2023-01-21 09:09:32.818936: step: 1620/529, loss: 0.32668519020080566 2023-01-21 09:09:34.038738: step: 1624/529, loss: 0.13679972290992737 2023-01-21 09:09:35.199497: step: 1628/529, loss: 0.2460399568080902 2023-01-21 09:09:36.377516: step: 1632/529, loss: 0.5962568521499634 2023-01-21 09:09:37.601682: step: 1636/529, loss: 0.2564878463745117 2023-01-21 09:09:38.865515: step: 1640/529, loss: 0.16419373452663422 2023-01-21 09:09:40.068328: step: 1644/529, loss: 0.6162596940994263 2023-01-21 09:09:41.283923: step: 1648/529, loss: 0.32858067750930786 2023-01-21 09:09:42.471427: step: 1652/529, loss: 0.3730602264404297 2023-01-21 09:09:43.679162: step: 1656/529, loss: 0.13982048630714417 2023-01-21 09:09:44.880768: step: 1660/529, loss: 0.14858514070510864 2023-01-21 09:09:46.106152: step: 1664/529, loss: 0.3491641879081726 2023-01-21 09:09:47.313462: step: 1668/529, loss: 0.5224711298942566 2023-01-21 09:09:48.539626: step: 1672/529, loss: 0.8076386451721191 2023-01-21 09:09:49.745963: step: 1676/529, loss: 0.028235293924808502 2023-01-21 09:09:50.969072: step: 1680/529, loss: 0.6606581211090088 2023-01-21 09:09:52.160756: step: 1684/529, loss: 0.14222599565982819 2023-01-21 09:09:53.348133: step: 1688/529, loss: 1.2723143100738525 2023-01-21 09:09:54.540979: step: 1692/529, loss: 0.16839638352394104 2023-01-21 09:09:55.712119: step: 1696/529, loss: 0.5446098446846008 2023-01-21 09:09:56.928937: step: 1700/529, loss: 0.10425370186567307 2023-01-21 09:09:58.160997: step: 1704/529, loss: 1.117645263671875 2023-01-21 09:09:59.436496: step: 1708/529, loss: 0.5030900835990906 2023-01-21 09:10:00.613048: step: 1712/529, loss: 0.11759386211633682 2023-01-21 09:10:01.813579: step: 1716/529, loss: 0.0623415969312191 2023-01-21 09:10:03.023571: step: 1720/529, loss: 0.17961858212947845 2023-01-21 09:10:04.206994: step: 1724/529, loss: 0.702845573425293 2023-01-21 09:10:05.433649: step: 1728/529, loss: 0.26210615038871765 2023-01-21 09:10:06.612115: step: 1732/529, loss: 0.24270649254322052 2023-01-21 09:10:07.800450: step: 1736/529, loss: 0.16498994827270508 2023-01-21 09:10:09.056098: step: 1740/529, loss: 0.4635719358921051 2023-01-21 09:10:10.252789: step: 1744/529, loss: 0.23091131448745728 2023-01-21 09:10:11.471864: step: 1748/529, loss: 0.2698693871498108 2023-01-21 09:10:12.633478: step: 1752/529, loss: 0.03807687759399414 2023-01-21 09:10:13.816688: step: 1756/529, loss: 0.1386251002550125 2023-01-21 09:10:14.989737: step: 1760/529, loss: 0.3056437373161316 2023-01-21 09:10:16.181522: step: 1764/529, loss: 1.1966632604599 2023-01-21 09:10:17.397658: step: 1768/529, loss: 0.1674070805311203 2023-01-21 09:10:18.656474: step: 1772/529, loss: 0.3627280592918396 2023-01-21 09:10:19.851663: step: 1776/529, loss: 0.10936679691076279 2023-01-21 09:10:21.063105: step: 1780/529, loss: 0.14010506868362427 2023-01-21 09:10:22.246201: step: 1784/529, loss: 0.5304542779922485 2023-01-21 09:10:23.434642: step: 1788/529, loss: 0.1013612300157547 2023-01-21 09:10:24.649876: step: 1792/529, loss: 0.15223565697669983 2023-01-21 09:10:25.856560: step: 1796/529, loss: 0.1163141280412674 2023-01-21 09:10:27.073107: step: 1800/529, loss: 0.38689059019088745 2023-01-21 09:10:28.251908: step: 1804/529, loss: 0.15308718383312225 2023-01-21 09:10:29.459742: step: 1808/529, loss: 0.29952287673950195 2023-01-21 09:10:30.642409: step: 1812/529, loss: 1.1570935249328613 2023-01-21 09:10:31.820499: step: 1816/529, loss: 0.11791582405567169 2023-01-21 09:10:33.044623: step: 1820/529, loss: 0.9169925451278687 2023-01-21 09:10:34.255292: step: 1824/529, loss: 0.6506383419036865 2023-01-21 09:10:35.439088: step: 1828/529, loss: 0.14234380424022675 2023-01-21 09:10:36.632669: step: 1832/529, loss: 0.4231431484222412 2023-01-21 09:10:37.804343: step: 1836/529, loss: 0.9142149090766907 2023-01-21 09:10:38.993911: step: 1840/529, loss: 0.17554154992103577 2023-01-21 09:10:40.215668: step: 1844/529, loss: 0.07488904148340225 2023-01-21 09:10:41.394102: step: 1848/529, loss: 0.2322222888469696 2023-01-21 09:10:42.578100: step: 1852/529, loss: 0.1308133602142334 2023-01-21 09:10:43.793804: step: 1856/529, loss: 0.37691235542297363 2023-01-21 09:10:45.037952: step: 1860/529, loss: 0.15815718472003937 2023-01-21 09:10:46.265935: step: 1864/529, loss: 0.14444312453269958 2023-01-21 09:10:47.573117: step: 1868/529, loss: 0.18990585207939148 2023-01-21 09:10:48.764031: step: 1872/529, loss: 0.18873673677444458 2023-01-21 09:10:49.990300: step: 1876/529, loss: 0.3121578097343445 2023-01-21 09:10:51.218105: step: 1880/529, loss: 0.37086376547813416 2023-01-21 09:10:52.432879: step: 1884/529, loss: 0.5689072608947754 2023-01-21 09:10:53.662254: step: 1888/529, loss: 1.9780168533325195 2023-01-21 09:10:54.873283: step: 1892/529, loss: 0.42991095781326294 2023-01-21 09:10:56.040683: step: 1896/529, loss: 1.3082197904586792 2023-01-21 09:10:57.265998: step: 1900/529, loss: 0.331615149974823 2023-01-21 09:10:58.444697: step: 1904/529, loss: 0.8489152789115906 2023-01-21 09:10:59.669029: step: 1908/529, loss: 0.17250938713550568 2023-01-21 09:11:00.858052: step: 1912/529, loss: 0.6576183438301086 2023-01-21 09:11:02.056995: step: 1916/529, loss: 0.7333192825317383 2023-01-21 09:11:03.215318: step: 1920/529, loss: 0.037929631769657135 2023-01-21 09:11:04.423847: step: 1924/529, loss: 0.2748446464538574 2023-01-21 09:11:05.587123: step: 1928/529, loss: 0.15305586159229279 2023-01-21 09:11:06.809642: step: 1932/529, loss: 0.6235495805740356 2023-01-21 09:11:07.984442: step: 1936/529, loss: 0.16210369765758514 2023-01-21 09:11:09.222562: step: 1940/529, loss: 0.6080190539360046 2023-01-21 09:11:10.417633: step: 1944/529, loss: 0.029820919036865234 2023-01-21 09:11:11.607082: step: 1948/529, loss: 0.3276333808898926 2023-01-21 09:11:12.793327: step: 1952/529, loss: 0.7028959393501282 2023-01-21 09:11:13.945163: step: 1956/529, loss: 0.2764837145805359 2023-01-21 09:11:15.127658: step: 1960/529, loss: 0.11809177696704865 2023-01-21 09:11:16.335182: step: 1964/529, loss: 0.33259010314941406 2023-01-21 09:11:17.539904: step: 1968/529, loss: 0.08955440670251846 2023-01-21 09:11:18.734516: step: 1972/529, loss: 0.2231583595275879 2023-01-21 09:11:19.958944: step: 1976/529, loss: 0.1280420422554016 2023-01-21 09:11:21.131862: step: 1980/529, loss: 0.7044577598571777 2023-01-21 09:11:22.309195: step: 1984/529, loss: 1.7870354652404785 2023-01-21 09:11:23.531609: step: 1988/529, loss: 0.7346959114074707 2023-01-21 09:11:24.730896: step: 1992/529, loss: 0.15680542588233948 2023-01-21 09:11:25.950280: step: 1996/529, loss: 6.2414374351501465 2023-01-21 09:11:27.167230: step: 2000/529, loss: 6.797178745269775 2023-01-21 09:11:28.355088: step: 2004/529, loss: 0.9583694934844971 2023-01-21 09:11:29.550231: step: 2008/529, loss: 0.15111970901489258 2023-01-21 09:11:30.765015: step: 2012/529, loss: 0.21521183848381042 2023-01-21 09:11:31.991017: step: 2016/529, loss: 0.2288244366645813 2023-01-21 09:11:33.217522: step: 2020/529, loss: 0.3413427472114563 2023-01-21 09:11:34.409904: step: 2024/529, loss: 0.20657825469970703 2023-01-21 09:11:35.634366: step: 2028/529, loss: 0.6548783183097839 2023-01-21 09:11:36.829363: step: 2032/529, loss: 0.23776665329933167 2023-01-21 09:11:38.073527: step: 2036/529, loss: 1.865551233291626 2023-01-21 09:11:39.270045: step: 2040/529, loss: 2.972569465637207 2023-01-21 09:11:40.484626: step: 2044/529, loss: 0.3656849265098572 2023-01-21 09:11:41.677959: step: 2048/529, loss: 0.3478592038154602 2023-01-21 09:11:42.861250: step: 2052/529, loss: 0.6147340536117554 2023-01-21 09:11:44.063866: step: 2056/529, loss: 0.21546894311904907 2023-01-21 09:11:45.332202: step: 2060/529, loss: 0.7534360289573669 2023-01-21 09:11:46.534235: step: 2064/529, loss: 0.18866434693336487 2023-01-21 09:11:47.709660: step: 2068/529, loss: 1.1781119108200073 2023-01-21 09:11:48.888974: step: 2072/529, loss: 0.19519153237342834 2023-01-21 09:11:50.095462: step: 2076/529, loss: 0.7405006885528564 2023-01-21 09:11:51.334038: step: 2080/529, loss: 0.15303802490234375 2023-01-21 09:11:52.521382: step: 2084/529, loss: 0.11230993270874023 2023-01-21 09:11:53.697441: step: 2088/529, loss: 2.185525894165039 2023-01-21 09:11:54.870214: step: 2092/529, loss: 0.1410335898399353 2023-01-21 09:11:56.081148: step: 2096/529, loss: 0.10806441307067871 2023-01-21 09:11:57.274590: step: 2100/529, loss: 0.3362579345703125 2023-01-21 09:11:58.473708: step: 2104/529, loss: 1.2092134952545166 2023-01-21 09:11:59.714261: step: 2108/529, loss: 0.45085421204566956 2023-01-21 09:12:00.925935: step: 2112/529, loss: 0.5155454874038696 2023-01-21 09:12:02.136478: step: 2116/529, loss: 1.3302206993103027 ================================================== Loss: 0.602 -------------------- Dev: {'event': {'p': 0.5331491712707183, 'r': 0.7709720372836218, 'f1': 0.6303756124115406}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Test: {'event': {'p': 0.538004246284501, 'r': 0.7591372079089275, 'f1': 0.6297216699801192}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Chinese: {'event': {'p': 0.5632183908045977, 'r': 0.9074074074074074, 'f1': 0.6950354609929078}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Korean: {'event': {'p': 0.625, 'r': 0.47619047619047616, 'f1': 0.5405405405405405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} 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': 2} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.5331491712707183, 'r': 0.7709720372836218, 'f1': 0.6303756124115406}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.538004246284501, 'r': 0.7591372079089275, 'f1': 0.6297216699801192}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.5632183908045977, 'r': 0.9074074074074074, 'f1': 0.6950354609929078}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.6904761904761905, 'r': 0.4603174603174603, 'f1': 0.5523809523809524}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} ****************************** Epoch: 3 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:12:55.520464: step: 4/529, loss: 0.5109049081802368 2023-01-21 09:12:56.785371: step: 8/529, loss: 0.5732133984565735 2023-01-21 09:12:57.977302: step: 12/529, loss: 0.34648609161376953 2023-01-21 09:12:59.191994: step: 16/529, loss: 6.525545120239258 2023-01-21 09:13:00.395827: step: 20/529, loss: 0.314157634973526 2023-01-21 09:13:01.542361: step: 24/529, loss: 0.016307448968291283 2023-01-21 09:13:02.705123: step: 28/529, loss: 0.15647979080677032 2023-01-21 09:13:03.889429: step: 32/529, loss: 0.12410131096839905 2023-01-21 09:13:05.071627: step: 36/529, loss: 0.02915172651410103 2023-01-21 09:13:06.306930: step: 40/529, loss: 0.11873731017112732 2023-01-21 09:13:07.513203: step: 44/529, loss: 0.7078371047973633 2023-01-21 09:13:08.758312: step: 48/529, loss: 0.2989673912525177 2023-01-21 09:13:09.919649: step: 52/529, loss: 0.46930694580078125 2023-01-21 09:13:11.118042: step: 56/529, loss: 0.4022432565689087 2023-01-21 09:13:12.326233: step: 60/529, loss: 0.2874813973903656 2023-01-21 09:13:13.569133: step: 64/529, loss: 0.9133665561676025 2023-01-21 09:13:14.806007: step: 68/529, loss: 0.4046599268913269 2023-01-21 09:13:15.994688: step: 72/529, loss: 0.24358081817626953 2023-01-21 09:13:17.227898: step: 76/529, loss: 0.09161201119422913 2023-01-21 09:13:18.404284: step: 80/529, loss: 0.4488402009010315 2023-01-21 09:13:19.581994: step: 84/529, loss: 0.22898206114768982 2023-01-21 09:13:20.784541: step: 88/529, loss: 1.3305326700210571 2023-01-21 09:13:22.010741: step: 92/529, loss: 0.5690761804580688 2023-01-21 09:13:23.178436: step: 96/529, loss: 0.7966560125350952 2023-01-21 09:13:24.390340: step: 100/529, loss: 1.448947548866272 2023-01-21 09:13:25.604552: step: 104/529, loss: 0.5288487672805786 2023-01-21 09:13:26.785978: step: 108/529, loss: 1.1439018249511719 2023-01-21 09:13:28.033657: step: 112/529, loss: 6.854997634887695 2023-01-21 09:13:29.263912: step: 116/529, loss: 0.464391827583313 2023-01-21 09:13:30.448109: step: 120/529, loss: 0.14924831688404083 2023-01-21 09:13:31.610988: step: 124/529, loss: 0.08640303462743759 2023-01-21 09:13:32.801086: step: 128/529, loss: 0.6751454472541809 2023-01-21 09:13:34.039617: step: 132/529, loss: 0.19832420349121094 2023-01-21 09:13:35.230599: step: 136/529, loss: 1.6694527864456177 2023-01-21 09:13:36.426629: step: 140/529, loss: 0.2284616380929947 2023-01-21 09:13:37.611654: step: 144/529, loss: 0.36430272459983826 2023-01-21 09:13:38.803403: step: 148/529, loss: 0.25114500522613525 2023-01-21 09:13:40.057688: step: 152/529, loss: 0.05569835007190704 2023-01-21 09:13:41.258857: step: 156/529, loss: 0.13806477189064026 2023-01-21 09:13:42.477107: step: 160/529, loss: 0.3269232511520386 2023-01-21 09:13:43.644705: step: 164/529, loss: 1.2911128997802734 2023-01-21 09:13:44.836879: step: 168/529, loss: 0.8094059228897095 2023-01-21 09:13:45.979756: step: 172/529, loss: 0.27050113677978516 2023-01-21 09:13:47.148681: step: 176/529, loss: 0.21918554604053497 2023-01-21 09:13:48.363659: step: 180/529, loss: 0.21964406967163086 2023-01-21 09:13:49.540457: step: 184/529, loss: 0.09067407250404358 2023-01-21 09:13:50.717520: step: 188/529, loss: 0.5115076303482056 2023-01-21 09:13:51.887448: step: 192/529, loss: 0.21389245986938477 2023-01-21 09:13:53.097274: step: 196/529, loss: 0.7584950923919678 2023-01-21 09:13:54.305445: step: 200/529, loss: 0.05163879692554474 2023-01-21 09:13:55.468613: step: 204/529, loss: 0.2306777983903885 2023-01-21 09:13:56.663236: step: 208/529, loss: 0.07892732322216034 2023-01-21 09:13:57.871794: step: 212/529, loss: 0.12173176556825638 2023-01-21 09:13:59.065033: step: 216/529, loss: 0.2150571048259735 2023-01-21 09:14:00.282666: step: 220/529, loss: 0.04045877605676651 2023-01-21 09:14:01.469910: step: 224/529, loss: 0.1015826165676117 2023-01-21 09:14:02.666871: step: 228/529, loss: 0.16814184188842773 2023-01-21 09:14:03.863520: step: 232/529, loss: 6.8595662117004395 2023-01-21 09:14:05.014833: step: 236/529, loss: 6.207767486572266 2023-01-21 09:14:06.226771: step: 240/529, loss: 0.08937664330005646 2023-01-21 09:14:07.438902: step: 244/529, loss: 0.41951608657836914 2023-01-21 09:14:08.646463: step: 248/529, loss: 0.08552493900060654 2023-01-21 09:14:09.822360: step: 252/529, loss: 0.2039077877998352 2023-01-21 09:14:10.980385: step: 256/529, loss: 0.056791454553604126 2023-01-21 09:14:12.153219: step: 260/529, loss: 0.44299763441085815 2023-01-21 09:14:13.315786: step: 264/529, loss: 0.05849161744117737 2023-01-21 09:14:14.495749: step: 268/529, loss: 0.7345892786979675 2023-01-21 09:14:15.753973: step: 272/529, loss: 0.7938448190689087 2023-01-21 09:14:16.960972: step: 276/529, loss: 0.9983171224594116 2023-01-21 09:14:18.138995: step: 280/529, loss: 0.08468132466077805 2023-01-21 09:14:19.321060: step: 284/529, loss: 0.17036715149879456 2023-01-21 09:14:20.521505: step: 288/529, loss: 0.20651274919509888 2023-01-21 09:14:21.746474: step: 292/529, loss: 0.09717493504285812 2023-01-21 09:14:22.964932: step: 296/529, loss: 0.7852643132209778 2023-01-21 09:14:24.139964: step: 300/529, loss: 0.2795930504798889 2023-01-21 09:14:25.321581: step: 304/529, loss: 0.13673889636993408 2023-01-21 09:14:26.504309: step: 308/529, loss: 0.5701743960380554 2023-01-21 09:14:27.731968: step: 312/529, loss: 1.238905429840088 2023-01-21 09:14:28.959858: step: 316/529, loss: 0.2844606339931488 2023-01-21 09:14:30.160925: step: 320/529, loss: 0.1601576805114746 2023-01-21 09:14:31.336824: step: 324/529, loss: 0.25467196106910706 2023-01-21 09:14:32.537381: step: 328/529, loss: 2.1542980670928955 2023-01-21 09:14:33.751187: step: 332/529, loss: 0.35070475935935974 2023-01-21 09:14:34.979174: step: 336/529, loss: 0.14309945702552795 2023-01-21 09:14:36.205047: step: 340/529, loss: 0.4852376878261566 2023-01-21 09:14:37.426574: step: 344/529, loss: 0.12140484154224396 2023-01-21 09:14:38.626905: step: 348/529, loss: 0.10886368900537491 2023-01-21 09:14:39.843546: step: 352/529, loss: 0.19722777605056763 2023-01-21 09:14:41.028750: step: 356/529, loss: 6.47122049331665 2023-01-21 09:14:42.249540: step: 360/529, loss: 0.22177237272262573 2023-01-21 09:14:43.417546: step: 364/529, loss: 0.1820840835571289 2023-01-21 09:14:44.620620: step: 368/529, loss: 0.16937704384326935 2023-01-21 09:14:45.787474: step: 372/529, loss: 0.39619603753089905 2023-01-21 09:14:47.011933: step: 376/529, loss: 0.2773638963699341 2023-01-21 09:14:48.213366: step: 380/529, loss: 0.18199768662452698 2023-01-21 09:14:49.414372: step: 384/529, loss: 0.12582755088806152 2023-01-21 09:14:50.624221: step: 388/529, loss: 0.15641355514526367 2023-01-21 09:14:51.841935: step: 392/529, loss: 1.1052515506744385 2023-01-21 09:14:53.031680: step: 396/529, loss: 0.3339536786079407 2023-01-21 09:14:54.194881: step: 400/529, loss: 0.036341190338134766 2023-01-21 09:14:55.394479: step: 404/529, loss: 0.3012176752090454 2023-01-21 09:14:56.597057: step: 408/529, loss: 0.179648756980896 2023-01-21 09:14:57.779723: step: 412/529, loss: 0.060222819447517395 2023-01-21 09:14:58.954255: step: 416/529, loss: 0.7835181951522827 2023-01-21 09:15:00.125451: step: 420/529, loss: 0.2005150020122528 2023-01-21 09:15:01.286646: step: 424/529, loss: 0.49489861726760864 2023-01-21 09:15:02.490293: step: 428/529, loss: 0.12319083511829376 2023-01-21 09:15:03.689321: step: 432/529, loss: 0.12356844544410706 2023-01-21 09:15:04.905292: step: 436/529, loss: 0.21137410402297974 2023-01-21 09:15:06.083768: step: 440/529, loss: 0.24912777543067932 2023-01-21 09:15:07.270376: step: 444/529, loss: 0.9127544164657593 2023-01-21 09:15:08.467430: step: 448/529, loss: 0.2940702438354492 2023-01-21 09:15:09.640826: step: 452/529, loss: 1.3996059894561768 2023-01-21 09:15:10.825544: step: 456/529, loss: 0.7008556723594666 2023-01-21 09:15:12.026914: step: 460/529, loss: 0.1283189356327057 2023-01-21 09:15:13.234177: step: 464/529, loss: 0.05786309391260147 2023-01-21 09:15:14.458576: step: 468/529, loss: 6.357701778411865 2023-01-21 09:15:15.677637: step: 472/529, loss: 0.838046669960022 2023-01-21 09:15:16.863497: step: 476/529, loss: 0.20604373514652252 2023-01-21 09:15:18.040116: step: 480/529, loss: 0.12215200066566467 2023-01-21 09:15:19.224836: step: 484/529, loss: 0.4220316708087921 2023-01-21 09:15:20.405277: step: 488/529, loss: 0.6431341171264648 2023-01-21 09:15:21.568129: step: 492/529, loss: 0.309932142496109 2023-01-21 09:15:22.792905: step: 496/529, loss: 6.305740833282471 2023-01-21 09:15:24.014638: step: 500/529, loss: 0.5798084139823914 2023-01-21 09:15:25.235364: step: 504/529, loss: 1.1165878772735596 2023-01-21 09:15:26.439973: step: 508/529, loss: 0.42484939098358154 2023-01-21 09:15:27.627520: step: 512/529, loss: 0.14822569489479065 2023-01-21 09:15:28.816019: step: 516/529, loss: 0.6631903052330017 2023-01-21 09:15:30.047897: step: 520/529, loss: 0.10189656913280487 2023-01-21 09:15:31.284022: step: 524/529, loss: 0.10649529099464417 2023-01-21 09:15:32.487802: step: 528/529, loss: 0.244944229722023 2023-01-21 09:15:33.648666: step: 532/529, loss: 0.18202033638954163 2023-01-21 09:15:34.831199: step: 536/529, loss: 0.08590392768383026 2023-01-21 09:15:36.034610: step: 540/529, loss: 0.14160548150539398 2023-01-21 09:15:37.261736: step: 544/529, loss: 2.385071277618408 2023-01-21 09:15:38.473865: step: 548/529, loss: 0.11255516856908798 2023-01-21 09:15:39.670956: step: 552/529, loss: 0.8202965259552002 2023-01-21 09:15:40.854070: step: 556/529, loss: 0.15509234368801117 2023-01-21 09:15:42.086038: step: 560/529, loss: 0.10815811157226562 2023-01-21 09:15:43.279842: step: 564/529, loss: 0.14583225548267365 2023-01-21 09:15:44.474659: step: 568/529, loss: 0.061933234333992004 2023-01-21 09:15:45.690953: step: 572/529, loss: 0.22751104831695557 2023-01-21 09:15:46.893511: step: 576/529, loss: 6.163327217102051 2023-01-21 09:15:48.081801: step: 580/529, loss: 0.9687484502792358 2023-01-21 09:15:49.309135: step: 584/529, loss: 0.6361277103424072 2023-01-21 09:15:50.514385: step: 588/529, loss: 0.14939336478710175 2023-01-21 09:15:51.665761: step: 592/529, loss: 0.5203262567520142 2023-01-21 09:15:52.878474: step: 596/529, loss: 0.23648720979690552 2023-01-21 09:15:54.059837: step: 600/529, loss: 0.17689228057861328 2023-01-21 09:15:55.220803: step: 604/529, loss: 0.920056164264679 2023-01-21 09:15:56.474409: step: 608/529, loss: 0.5570315718650818 2023-01-21 09:15:57.689257: step: 612/529, loss: 0.2973543107509613 2023-01-21 09:15:58.867432: step: 616/529, loss: 1.263750433921814 2023-01-21 09:16:00.047348: step: 620/529, loss: 0.3595085144042969 2023-01-21 09:16:01.228314: step: 624/529, loss: 0.5976070165634155 2023-01-21 09:16:02.426267: step: 628/529, loss: 0.4708814322948456 2023-01-21 09:16:03.630379: step: 632/529, loss: 0.2147449553012848 2023-01-21 09:16:04.777349: step: 636/529, loss: 0.4652779698371887 2023-01-21 09:16:06.018272: step: 640/529, loss: 0.17160634696483612 2023-01-21 09:16:07.182246: step: 644/529, loss: 0.13353107869625092 2023-01-21 09:16:08.371237: step: 648/529, loss: 0.20135146379470825 2023-01-21 09:16:09.535263: step: 652/529, loss: 1.2545086145401 2023-01-21 09:16:10.708856: step: 656/529, loss: 6.808806896209717 2023-01-21 09:16:11.896744: step: 660/529, loss: 0.1724664717912674 2023-01-21 09:16:13.109949: step: 664/529, loss: 0.6020802855491638 2023-01-21 09:16:14.331635: step: 668/529, loss: 0.20385518670082092 2023-01-21 09:16:15.540834: step: 672/529, loss: 0.0566895455121994 2023-01-21 09:16:16.786827: step: 676/529, loss: 0.15117855370044708 2023-01-21 09:16:18.045429: step: 680/529, loss: 0.30817556381225586 2023-01-21 09:16:19.227046: step: 684/529, loss: 0.20230546593666077 2023-01-21 09:16:20.405998: step: 688/529, loss: 0.56947922706604 2023-01-21 09:16:21.589832: step: 692/529, loss: 0.7333808541297913 2023-01-21 09:16:22.778398: step: 696/529, loss: 0.21116027235984802 2023-01-21 09:16:23.977716: step: 700/529, loss: 0.3762238621711731 2023-01-21 09:16:25.154998: step: 704/529, loss: 0.1824939250946045 2023-01-21 09:16:26.327026: step: 708/529, loss: 0.04479799419641495 2023-01-21 09:16:27.539542: step: 712/529, loss: 1.4403314590454102 2023-01-21 09:16:28.719125: step: 716/529, loss: 0.14453621208667755 2023-01-21 09:16:29.899010: step: 720/529, loss: 0.027975894510746002 2023-01-21 09:16:31.030035: step: 724/529, loss: 0.0838829055428505 2023-01-21 09:16:32.206117: step: 728/529, loss: 0.06790085136890411 2023-01-21 09:16:33.415603: step: 732/529, loss: 0.02364349365234375 2023-01-21 09:16:34.654340: step: 736/529, loss: 0.2607499957084656 2023-01-21 09:16:35.816459: step: 740/529, loss: 0.7614808678627014 2023-01-21 09:16:36.989162: step: 744/529, loss: 0.0895475521683693 2023-01-21 09:16:38.153720: step: 748/529, loss: 0.04825344309210777 2023-01-21 09:16:39.322531: step: 752/529, loss: 0.17991089820861816 2023-01-21 09:16:40.514841: step: 756/529, loss: 0.411781370639801 2023-01-21 09:16:41.707934: step: 760/529, loss: 0.2736542820930481 2023-01-21 09:16:42.881270: step: 764/529, loss: 0.24587230384349823 2023-01-21 09:16:44.074186: step: 768/529, loss: 0.2465108036994934 2023-01-21 09:16:45.240089: step: 772/529, loss: 0.21385689079761505 2023-01-21 09:16:46.445068: step: 776/529, loss: 0.6135229468345642 2023-01-21 09:16:47.659611: step: 780/529, loss: 0.1008480042219162 2023-01-21 09:16:48.863346: step: 784/529, loss: 0.05660128593444824 2023-01-21 09:16:50.039951: step: 788/529, loss: 0.21203604340553284 2023-01-21 09:16:51.224947: step: 792/529, loss: 0.4665027856826782 2023-01-21 09:16:52.433636: step: 796/529, loss: 1.4546465873718262 2023-01-21 09:16:53.617795: step: 800/529, loss: 0.12697425484657288 2023-01-21 09:16:54.810080: step: 804/529, loss: 0.1567641794681549 2023-01-21 09:16:56.028145: step: 808/529, loss: 1.7793183326721191 2023-01-21 09:16:57.223286: step: 812/529, loss: 0.3820658326148987 2023-01-21 09:16:58.406898: step: 816/529, loss: 6.272670269012451 2023-01-21 09:16:59.597199: step: 820/529, loss: 0.18555909395217896 2023-01-21 09:17:00.792578: step: 824/529, loss: 0.15330100059509277 2023-01-21 09:17:01.993608: step: 828/529, loss: 0.2847210764884949 2023-01-21 09:17:03.229537: step: 832/529, loss: 0.11364109814167023 2023-01-21 09:17:04.474368: step: 836/529, loss: 0.6854779124259949 2023-01-21 09:17:05.693436: step: 840/529, loss: 1.2999851703643799 2023-01-21 09:17:06.947679: step: 844/529, loss: 0.8045494556427002 2023-01-21 09:17:08.119384: step: 848/529, loss: 0.2060120701789856 2023-01-21 09:17:09.326243: step: 852/529, loss: 0.06410017609596252 2023-01-21 09:17:10.512391: step: 856/529, loss: 0.16923925280570984 2023-01-21 09:17:11.731873: step: 860/529, loss: 0.5960256457328796 2023-01-21 09:17:12.903717: step: 864/529, loss: 0.20848403871059418 2023-01-21 09:17:14.108695: step: 868/529, loss: 0.17049500346183777 2023-01-21 09:17:15.291823: step: 872/529, loss: 0.11787271499633789 2023-01-21 09:17:16.486690: step: 876/529, loss: 0.09072265774011612 2023-01-21 09:17:17.689614: step: 880/529, loss: 0.21228435635566711 2023-01-21 09:17:18.890136: step: 884/529, loss: 0.5185469388961792 2023-01-21 09:17:20.097985: step: 888/529, loss: 0.1307760775089264 2023-01-21 09:17:21.293843: step: 892/529, loss: 0.2846911549568176 2023-01-21 09:17:22.501508: step: 896/529, loss: 1.2598540782928467 2023-01-21 09:17:23.723245: step: 900/529, loss: 0.793027937412262 2023-01-21 09:17:24.914733: step: 904/529, loss: 0.05815587192773819 2023-01-21 09:17:26.108913: step: 908/529, loss: 0.8849951028823853 2023-01-21 09:17:27.257664: step: 912/529, loss: 0.6899126768112183 2023-01-21 09:17:28.442125: step: 916/529, loss: 0.07221993803977966 2023-01-21 09:17:29.678064: step: 920/529, loss: 0.11165991425514221 2023-01-21 09:17:30.902396: step: 924/529, loss: 0.2791863679885864 2023-01-21 09:17:32.136563: step: 928/529, loss: 0.19381389021873474 2023-01-21 09:17:33.338878: step: 932/529, loss: 1.0754708051681519 2023-01-21 09:17:34.554851: step: 936/529, loss: 0.29282352328300476 2023-01-21 09:17:35.752708: step: 940/529, loss: 1.066131353378296 2023-01-21 09:17:36.948342: step: 944/529, loss: 0.20834693312644958 2023-01-21 09:17:38.152527: step: 948/529, loss: 0.24618284404277802 2023-01-21 09:17:39.379877: step: 952/529, loss: 0.7845458984375 2023-01-21 09:17:40.589293: step: 956/529, loss: 0.09872083365917206 2023-01-21 09:17:41.750385: step: 960/529, loss: 0.08556786179542542 2023-01-21 09:17:42.993447: step: 964/529, loss: 1.088285207748413 2023-01-21 09:17:44.163233: step: 968/529, loss: 0.07125687599182129 2023-01-21 09:17:45.348685: step: 972/529, loss: 1.1183327436447144 2023-01-21 09:17:46.561289: step: 976/529, loss: 0.2211429625749588 2023-01-21 09:17:47.751794: step: 980/529, loss: 0.941512405872345 2023-01-21 09:17:48.920135: step: 984/529, loss: 0.14278888702392578 2023-01-21 09:17:50.149405: step: 988/529, loss: 0.20018434524536133 2023-01-21 09:17:51.342686: step: 992/529, loss: 1.1467530727386475 2023-01-21 09:17:52.573674: step: 996/529, loss: 0.0738748088479042 2023-01-21 09:17:53.746493: step: 1000/529, loss: 0.15200920403003693 2023-01-21 09:17:54.952045: step: 1004/529, loss: 0.15242335200309753 2023-01-21 09:17:56.088243: step: 1008/529, loss: 0.710332989692688 2023-01-21 09:17:57.265518: step: 1012/529, loss: 0.3448978364467621 2023-01-21 09:17:58.447671: step: 1016/529, loss: 0.07349491119384766 2023-01-21 09:17:59.663478: step: 1020/529, loss: 0.8348375558853149 2023-01-21 09:18:00.864080: step: 1024/529, loss: 1.107804536819458 2023-01-21 09:18:02.091937: step: 1028/529, loss: 0.11502466350793839 2023-01-21 09:18:03.332138: step: 1032/529, loss: 0.32818615436553955 2023-01-21 09:18:04.571748: step: 1036/529, loss: 1.8135936260223389 2023-01-21 09:18:05.753937: step: 1040/529, loss: 0.30856990814208984 2023-01-21 09:18:06.984514: step: 1044/529, loss: 0.38151323795318604 2023-01-21 09:18:08.188158: step: 1048/529, loss: 0.6319208145141602 2023-01-21 09:18:09.380222: step: 1052/529, loss: 0.665233850479126 2023-01-21 09:18:10.628441: step: 1056/529, loss: 0.07123690098524094 2023-01-21 09:18:11.859632: step: 1060/529, loss: 0.13324937224388123 2023-01-21 09:18:13.053949: step: 1064/529, loss: 0.048981428146362305 2023-01-21 09:18:14.265380: step: 1068/529, loss: 0.1499921828508377 2023-01-21 09:18:15.455509: step: 1072/529, loss: 0.29684311151504517 2023-01-21 09:18:16.643514: step: 1076/529, loss: 0.10205487906932831 2023-01-21 09:18:17.860047: step: 1080/529, loss: 1.0443146228790283 2023-01-21 09:18:19.072816: step: 1084/529, loss: 0.4533035159111023 2023-01-21 09:18:20.302607: step: 1088/529, loss: 0.28314828872680664 2023-01-21 09:18:21.626161: step: 1092/529, loss: 0.2152790129184723 2023-01-21 09:18:22.822839: step: 1096/529, loss: 0.25698530673980713 2023-01-21 09:18:23.991431: step: 1100/529, loss: 0.07423897087574005 2023-01-21 09:18:25.211388: step: 1104/529, loss: 0.20360970497131348 2023-01-21 09:18:26.402828: step: 1108/529, loss: 0.07388553768396378 2023-01-21 09:18:27.634908: step: 1112/529, loss: 0.4620504081249237 2023-01-21 09:18:28.874757: step: 1116/529, loss: 6.0797295570373535 2023-01-21 09:18:30.052075: step: 1120/529, loss: 0.10023251175880432 2023-01-21 09:18:31.280566: step: 1124/529, loss: 0.5048360824584961 2023-01-21 09:18:32.489456: step: 1128/529, loss: 0.10539598762989044 2023-01-21 09:18:33.709252: step: 1132/529, loss: 0.3162137269973755 2023-01-21 09:18:34.910587: step: 1136/529, loss: 0.202396959066391 2023-01-21 09:18:36.114441: step: 1140/529, loss: 0.8612620830535889 2023-01-21 09:18:37.359202: step: 1144/529, loss: 0.12645836174488068 2023-01-21 09:18:38.572476: step: 1148/529, loss: 6.365048408508301 2023-01-21 09:18:39.784186: step: 1152/529, loss: 0.6691771745681763 2023-01-21 09:18:40.937893: step: 1156/529, loss: 0.5278523564338684 2023-01-21 09:18:42.162633: step: 1160/529, loss: 0.4883405566215515 2023-01-21 09:18:43.330150: step: 1164/529, loss: 6.404985427856445 2023-01-21 09:18:44.529476: step: 1168/529, loss: 0.15984183549880981 2023-01-21 09:18:45.721531: step: 1172/529, loss: 0.04248318821191788 2023-01-21 09:18:46.977740: step: 1176/529, loss: 0.11476249992847443 2023-01-21 09:18:48.213585: step: 1180/529, loss: 0.15053291618824005 2023-01-21 09:18:49.411135: step: 1184/529, loss: 0.07252518832683563 2023-01-21 09:18:50.605843: step: 1188/529, loss: 0.8370453119277954 2023-01-21 09:18:51.783652: step: 1192/529, loss: 0.18597184121608734 2023-01-21 09:18:52.998432: step: 1196/529, loss: 0.18255853652954102 2023-01-21 09:18:54.194286: step: 1200/529, loss: 0.8022556304931641 2023-01-21 09:18:55.376426: step: 1204/529, loss: 0.12340659648180008 2023-01-21 09:18:56.545309: step: 1208/529, loss: 0.15646138787269592 2023-01-21 09:18:57.753444: step: 1212/529, loss: 0.39518415927886963 2023-01-21 09:18:58.945342: step: 1216/529, loss: 0.18684397637844086 2023-01-21 09:19:00.154655: step: 1220/529, loss: 0.012744761072099209 2023-01-21 09:19:01.375073: step: 1224/529, loss: 1.641706109046936 2023-01-21 09:19:02.594215: step: 1228/529, loss: 0.35822969675064087 2023-01-21 09:19:03.784590: step: 1232/529, loss: 0.05193042755126953 2023-01-21 09:19:04.954932: step: 1236/529, loss: 1.405439853668213 2023-01-21 09:19:06.141813: step: 1240/529, loss: 0.02931075170636177 2023-01-21 09:19:07.353569: step: 1244/529, loss: 0.10154262185096741 2023-01-21 09:19:08.576524: step: 1248/529, loss: 0.5561035871505737 2023-01-21 09:19:09.768147: step: 1252/529, loss: 0.13672709465026855 2023-01-21 09:19:10.976370: step: 1256/529, loss: 0.2545683681964874 2023-01-21 09:19:12.137149: step: 1260/529, loss: 0.20950919389724731 2023-01-21 09:19:13.337715: step: 1264/529, loss: 0.15473446249961853 2023-01-21 09:19:14.553564: step: 1268/529, loss: 0.13981465995311737 2023-01-21 09:19:15.811302: step: 1272/529, loss: 0.7003820538520813 2023-01-21 09:19:17.016281: step: 1276/529, loss: 0.07366523891687393 2023-01-21 09:19:18.239182: step: 1280/529, loss: 0.06368014961481094 2023-01-21 09:19:19.413400: step: 1284/529, loss: 0.19977210462093353 2023-01-21 09:19:20.611703: step: 1288/529, loss: 0.2320655882358551 2023-01-21 09:19:21.833361: step: 1292/529, loss: 0.04660029709339142 2023-01-21 09:19:23.035975: step: 1296/529, loss: 0.06706514209508896 2023-01-21 09:19:24.223424: step: 1300/529, loss: 0.1906927227973938 2023-01-21 09:19:25.390960: step: 1304/529, loss: 0.3095458149909973 2023-01-21 09:19:26.566271: step: 1308/529, loss: 0.19476346671581268 2023-01-21 09:19:27.741731: step: 1312/529, loss: 0.4397938847541809 2023-01-21 09:19:28.906289: step: 1316/529, loss: 0.6443344354629517 2023-01-21 09:19:30.106257: step: 1320/529, loss: 0.15377309918403625 2023-01-21 09:19:31.316886: step: 1324/529, loss: 0.28709226846694946 2023-01-21 09:19:32.497594: step: 1328/529, loss: 0.20887838304042816 2023-01-21 09:19:33.718847: step: 1332/529, loss: 0.762951672077179 2023-01-21 09:19:34.929622: step: 1336/529, loss: 0.12815412878990173 2023-01-21 09:19:36.133870: step: 1340/529, loss: 0.17199945449829102 2023-01-21 09:19:37.343772: step: 1344/529, loss: 0.03209657967090607 2023-01-21 09:19:38.609071: step: 1348/529, loss: 0.24765682220458984 2023-01-21 09:19:39.844498: step: 1352/529, loss: 0.11222286522388458 2023-01-21 09:19:41.059456: step: 1356/529, loss: 1.4803955554962158 2023-01-21 09:19:42.252279: step: 1360/529, loss: 0.058991432189941406 2023-01-21 09:19:43.439718: step: 1364/529, loss: 0.1906137466430664 2023-01-21 09:19:44.634836: step: 1368/529, loss: 0.14589159190654755 2023-01-21 09:19:45.832982: step: 1372/529, loss: 0.7356292009353638 2023-01-21 09:19:47.020592: step: 1376/529, loss: 0.0292937271296978 2023-01-21 09:19:48.198464: step: 1380/529, loss: 0.1148996651172638 2023-01-21 09:19:49.397243: step: 1384/529, loss: 0.4950922429561615 2023-01-21 09:19:50.568447: step: 1388/529, loss: 0.6311390995979309 2023-01-21 09:19:51.767626: step: 1392/529, loss: 0.22987575829029083 2023-01-21 09:19:52.972696: step: 1396/529, loss: 0.10704431682825089 2023-01-21 09:19:54.177259: step: 1400/529, loss: 0.1649210900068283 2023-01-21 09:19:55.332692: step: 1404/529, loss: 2.6152503490448 2023-01-21 09:19:56.525177: step: 1408/529, loss: 0.6494420766830444 2023-01-21 09:19:57.737152: step: 1412/529, loss: 0.8145421743392944 2023-01-21 09:19:58.975295: step: 1416/529, loss: 0.5473567843437195 2023-01-21 09:20:00.182563: step: 1420/529, loss: 0.1390179544687271 2023-01-21 09:20:01.359570: step: 1424/529, loss: 0.8674865365028381 2023-01-21 09:20:02.603436: step: 1428/529, loss: 0.10704083740711212 2023-01-21 09:20:03.822414: step: 1432/529, loss: 0.3758760690689087 2023-01-21 09:20:05.061969: step: 1436/529, loss: 0.6295431852340698 2023-01-21 09:20:06.255205: step: 1440/529, loss: 0.38997578620910645 2023-01-21 09:20:07.465485: step: 1444/529, loss: 0.11759305000305176 2023-01-21 09:20:08.661754: step: 1448/529, loss: 0.7263033986091614 2023-01-21 09:20:09.855242: step: 1452/529, loss: 0.4058622121810913 2023-01-21 09:20:11.086136: step: 1456/529, loss: 0.2892892062664032 2023-01-21 09:20:12.289951: step: 1460/529, loss: 0.26381224393844604 2023-01-21 09:20:13.495299: step: 1464/529, loss: 0.8017733693122864 2023-01-21 09:20:14.697546: step: 1468/529, loss: 0.34554481506347656 2023-01-21 09:20:15.902979: step: 1472/529, loss: 0.18896789848804474 2023-01-21 09:20:17.066712: step: 1476/529, loss: 0.31995201110839844 2023-01-21 09:20:18.238749: step: 1480/529, loss: 0.7046374678611755 2023-01-21 09:20:19.486948: step: 1484/529, loss: 0.2419232428073883 2023-01-21 09:20:20.665658: step: 1488/529, loss: 0.6109528541564941 2023-01-21 09:20:21.828977: step: 1492/529, loss: 0.1533583700656891 2023-01-21 09:20:23.056588: step: 1496/529, loss: 0.15195946395397186 2023-01-21 09:20:24.329526: step: 1500/529, loss: 0.7536163926124573 2023-01-21 09:20:25.518329: step: 1504/529, loss: 0.048946477472782135 2023-01-21 09:20:26.715229: step: 1508/529, loss: 0.1295931339263916 2023-01-21 09:20:27.898746: step: 1512/529, loss: 0.5034002661705017 2023-01-21 09:20:29.098319: step: 1516/529, loss: 0.177537739276886 2023-01-21 09:20:30.283240: step: 1520/529, loss: 0.762010931968689 2023-01-21 09:20:31.474258: step: 1524/529, loss: 0.3654606342315674 2023-01-21 09:20:32.711122: step: 1528/529, loss: 0.24029389023780823 2023-01-21 09:20:33.885999: step: 1532/529, loss: 0.902274489402771 2023-01-21 09:20:35.091464: step: 1536/529, loss: 0.1744929850101471 2023-01-21 09:20:36.312568: step: 1540/529, loss: 0.26031026244163513 2023-01-21 09:20:37.547893: step: 1544/529, loss: 0.08439760655164719 2023-01-21 09:20:38.806675: step: 1548/529, loss: 0.03405600041151047 2023-01-21 09:20:39.994322: step: 1552/529, loss: 0.0916815772652626 2023-01-21 09:20:41.193006: step: 1556/529, loss: 0.3352050185203552 2023-01-21 09:20:42.395493: step: 1560/529, loss: 0.3663368225097656 2023-01-21 09:20:43.592533: step: 1564/529, loss: 0.05466461181640625 2023-01-21 09:20:44.800622: step: 1568/529, loss: 0.49147921800613403 2023-01-21 09:20:46.023605: step: 1572/529, loss: 0.2640061378479004 2023-01-21 09:20:47.251212: step: 1576/529, loss: 0.3838962912559509 2023-01-21 09:20:48.483687: step: 1580/529, loss: 0.24831877648830414 2023-01-21 09:20:49.703205: step: 1584/529, loss: 0.11662021279335022 2023-01-21 09:20:50.927587: step: 1588/529, loss: 0.20084114372730255 2023-01-21 09:20:52.120446: step: 1592/529, loss: 0.309184730052948 2023-01-21 09:20:53.297233: step: 1596/529, loss: 0.43160516023635864 2023-01-21 09:20:54.495767: step: 1600/529, loss: 0.2824947237968445 2023-01-21 09:20:55.707006: step: 1604/529, loss: 0.8066858649253845 2023-01-21 09:20:56.888572: step: 1608/529, loss: 0.19391365349292755 2023-01-21 09:20:58.081857: step: 1612/529, loss: 0.5352584719657898 2023-01-21 09:20:59.325037: step: 1616/529, loss: 0.1670810729265213 2023-01-21 09:21:00.502865: step: 1620/529, loss: 1.385312795639038 2023-01-21 09:21:01.681700: step: 1624/529, loss: 0.13863128423690796 2023-01-21 09:21:02.863319: step: 1628/529, loss: 0.7014477252960205 2023-01-21 09:21:04.079627: step: 1632/529, loss: 0.47930222749710083 2023-01-21 09:21:05.262733: step: 1636/529, loss: 0.3952499330043793 2023-01-21 09:21:06.485699: step: 1640/529, loss: 1.3406263589859009 2023-01-21 09:21:07.720694: step: 1644/529, loss: 0.11925850063562393 2023-01-21 09:21:08.972062: step: 1648/529, loss: 0.7813748121261597 2023-01-21 09:21:10.165559: step: 1652/529, loss: 0.25027528405189514 2023-01-21 09:21:11.352286: step: 1656/529, loss: 0.4950025677680969 2023-01-21 09:21:12.551805: step: 1660/529, loss: 1.7744572162628174 2023-01-21 09:21:13.791334: step: 1664/529, loss: 0.14767208695411682 2023-01-21 09:21:14.965709: step: 1668/529, loss: 0.43377143144607544 2023-01-21 09:21:16.227400: step: 1672/529, loss: 0.27826356887817383 2023-01-21 09:21:17.438774: step: 1676/529, loss: 0.55902099609375 2023-01-21 09:21:18.634832: step: 1680/529, loss: 0.08716002106666565 2023-01-21 09:21:19.919872: step: 1684/529, loss: 0.12237024307250977 2023-01-21 09:21:21.119514: step: 1688/529, loss: 0.14731532335281372 2023-01-21 09:21:22.298283: step: 1692/529, loss: 0.10474033653736115 2023-01-21 09:21:23.523134: step: 1696/529, loss: 0.030937576666474342 2023-01-21 09:21:24.690789: step: 1700/529, loss: 0.04436516761779785 2023-01-21 09:21:25.882315: step: 1704/529, loss: 0.3761083781719208 2023-01-21 09:21:27.090789: step: 1708/529, loss: 0.4886413514614105 2023-01-21 09:21:28.297032: step: 1712/529, loss: 0.1376006305217743 2023-01-21 09:21:29.515657: step: 1716/529, loss: 0.054845474660396576 2023-01-21 09:21:30.694750: step: 1720/529, loss: 0.7890036702156067 2023-01-21 09:21:31.888378: step: 1724/529, loss: 0.2800571322441101 2023-01-21 09:21:33.062732: step: 1728/529, loss: 0.1427145004272461 2023-01-21 09:21:34.268076: step: 1732/529, loss: 0.10956144332885742 2023-01-21 09:21:35.448354: step: 1736/529, loss: 0.11455106735229492 2023-01-21 09:21:36.647094: step: 1740/529, loss: 0.39195671677589417 2023-01-21 09:21:37.871346: step: 1744/529, loss: 0.12060967087745667 2023-01-21 09:21:39.102769: step: 1748/529, loss: 0.6424474716186523 2023-01-21 09:21:40.287462: step: 1752/529, loss: 0.6103021502494812 2023-01-21 09:21:41.480623: step: 1756/529, loss: 0.3288912773132324 2023-01-21 09:21:42.715526: step: 1760/529, loss: 0.13594570755958557 2023-01-21 09:21:43.904207: step: 1764/529, loss: 0.06404094398021698 2023-01-21 09:21:45.133975: step: 1768/529, loss: 0.0946381539106369 2023-01-21 09:21:46.322702: step: 1772/529, loss: 0.1461222767829895 2023-01-21 09:21:47.530972: step: 1776/529, loss: 0.08773450553417206 2023-01-21 09:21:48.726878: step: 1780/529, loss: 0.575681209564209 2023-01-21 09:21:49.965905: step: 1784/529, loss: 0.1255619078874588 2023-01-21 09:21:51.178585: step: 1788/529, loss: 0.7396607398986816 2023-01-21 09:21:52.344956: step: 1792/529, loss: 0.0425536148250103 2023-01-21 09:21:53.539219: step: 1796/529, loss: 0.10895557701587677 2023-01-21 09:21:54.736940: step: 1800/529, loss: 0.4857218861579895 2023-01-21 09:21:55.954609: step: 1804/529, loss: 0.5225170850753784 2023-01-21 09:21:57.141623: step: 1808/529, loss: 0.281673401594162 2023-01-21 09:21:58.336561: step: 1812/529, loss: 1.1940937042236328 2023-01-21 09:21:59.565857: step: 1816/529, loss: 0.210743248462677 2023-01-21 09:22:00.768919: step: 1820/529, loss: 0.43428170680999756 2023-01-21 09:22:01.986255: step: 1824/529, loss: 0.06261620670557022 2023-01-21 09:22:03.157521: step: 1828/529, loss: 0.21472720801830292 2023-01-21 09:22:04.349959: step: 1832/529, loss: 0.27495822310447693 2023-01-21 09:22:05.559960: step: 1836/529, loss: 0.326531320810318 2023-01-21 09:22:06.795156: step: 1840/529, loss: 0.1997019350528717 2023-01-21 09:22:08.025290: step: 1844/529, loss: 0.1821242868900299 2023-01-21 09:22:09.207821: step: 1848/529, loss: 0.15181970596313477 2023-01-21 09:22:10.421062: step: 1852/529, loss: 0.6286935806274414 2023-01-21 09:22:11.600601: step: 1856/529, loss: 0.14447203278541565 2023-01-21 09:22:12.748139: step: 1860/529, loss: 0.15174522995948792 2023-01-21 09:22:13.942204: step: 1864/529, loss: 0.2580508589744568 2023-01-21 09:22:15.118117: step: 1868/529, loss: 0.10839777439832687 2023-01-21 09:22:16.301274: step: 1872/529, loss: 0.07493095099925995 2023-01-21 09:22:17.517734: step: 1876/529, loss: 0.15180546045303345 2023-01-21 09:22:18.735856: step: 1880/529, loss: 0.45923691987991333 2023-01-21 09:22:19.938062: step: 1884/529, loss: 0.11712709069252014 2023-01-21 09:22:21.177410: step: 1888/529, loss: 0.6078892946243286 2023-01-21 09:22:22.418536: step: 1892/529, loss: 0.0914234146475792 2023-01-21 09:22:23.610631: step: 1896/529, loss: 0.8523756861686707 2023-01-21 09:22:24.790167: step: 1900/529, loss: 0.9335610866546631 2023-01-21 09:22:26.012984: step: 1904/529, loss: 0.23920392990112305 2023-01-21 09:22:27.219526: step: 1908/529, loss: 0.09496049582958221 2023-01-21 09:22:28.406894: step: 1912/529, loss: 0.3865712285041809 2023-01-21 09:22:29.623568: step: 1916/529, loss: 0.12209845334291458 2023-01-21 09:22:30.827196: step: 1920/529, loss: 0.3588685393333435 2023-01-21 09:22:32.025044: step: 1924/529, loss: 1.01535964012146 2023-01-21 09:22:33.260751: step: 1928/529, loss: 0.12728175520896912 2023-01-21 09:22:34.476696: step: 1932/529, loss: 0.13523274660110474 2023-01-21 09:22:35.706071: step: 1936/529, loss: 0.3225170075893402 2023-01-21 09:22:36.909539: step: 1940/529, loss: 0.15277346968650818 2023-01-21 09:22:38.106881: step: 1944/529, loss: 1.0891169309616089 2023-01-21 09:22:39.399508: step: 1948/529, loss: 0.4466021656990051 2023-01-21 09:22:40.585679: step: 1952/529, loss: 0.9465929865837097 2023-01-21 09:22:41.801630: step: 1956/529, loss: 0.23988103866577148 2023-01-21 09:22:42.994260: step: 1960/529, loss: 0.05404920503497124 2023-01-21 09:22:44.193600: step: 1964/529, loss: 0.6631994247436523 2023-01-21 09:22:45.415870: step: 1968/529, loss: 0.17227205634117126 2023-01-21 09:22:46.606875: step: 1972/529, loss: 0.10587625950574875 2023-01-21 09:22:47.809330: step: 1976/529, loss: 0.46818360686302185 2023-01-21 09:22:49.026759: step: 1980/529, loss: 0.38124608993530273 2023-01-21 09:22:50.234046: step: 1984/529, loss: 0.03677845001220703 2023-01-21 09:22:51.448736: step: 1988/529, loss: 6.685095310211182 2023-01-21 09:22:52.646642: step: 1992/529, loss: 0.22513896226882935 2023-01-21 09:22:53.844637: step: 1996/529, loss: 0.1617441177368164 2023-01-21 09:22:55.026851: step: 2000/529, loss: 0.03118271939456463 2023-01-21 09:22:56.241293: step: 2004/529, loss: 0.18503373861312866 2023-01-21 09:22:57.399129: step: 2008/529, loss: 0.8965415954589844 2023-01-21 09:22:58.599795: step: 2012/529, loss: 0.5357754230499268 2023-01-21 09:22:59.841780: step: 2016/529, loss: 1.1179760694503784 2023-01-21 09:23:01.079339: step: 2020/529, loss: 0.09247355908155441 2023-01-21 09:23:02.317252: step: 2024/529, loss: 0.11763171851634979 2023-01-21 09:23:03.535299: step: 2028/529, loss: 0.1918288767337799 2023-01-21 09:23:04.721131: step: 2032/529, loss: 0.16145601868629456 2023-01-21 09:23:05.944963: step: 2036/529, loss: 0.8493215441703796 2023-01-21 09:23:07.163982: step: 2040/529, loss: 1.5145994424819946 2023-01-21 09:23:08.390802: step: 2044/529, loss: 0.24307653307914734 2023-01-21 09:23:09.586242: step: 2048/529, loss: 0.20988675951957703 2023-01-21 09:23:10.811544: step: 2052/529, loss: 0.8853633999824524 2023-01-21 09:23:12.036179: step: 2056/529, loss: 0.1410486251115799 2023-01-21 09:23:13.246501: step: 2060/529, loss: 0.11293602734804153 2023-01-21 09:23:14.469555: step: 2064/529, loss: 0.2375413030385971 2023-01-21 09:23:15.705631: step: 2068/529, loss: 0.7124490737915039 2023-01-21 09:23:16.986699: step: 2072/529, loss: 0.1207837164402008 2023-01-21 09:23:18.157405: step: 2076/529, loss: 0.04087388515472412 2023-01-21 09:23:19.353203: step: 2080/529, loss: 1.069754719734192 2023-01-21 09:23:20.573015: step: 2084/529, loss: 0.08507518470287323 2023-01-21 09:23:21.831531: step: 2088/529, loss: 0.11504700779914856 2023-01-21 09:23:23.031128: step: 2092/529, loss: 0.24900245666503906 2023-01-21 09:23:24.216606: step: 2096/529, loss: 0.6266627311706543 2023-01-21 09:23:25.418419: step: 2100/529, loss: 0.022880934178829193 2023-01-21 09:23:26.589410: step: 2104/529, loss: 0.2722981572151184 2023-01-21 09:23:27.748778: step: 2108/529, loss: 0.4525381922721863 2023-01-21 09:23:28.931897: step: 2112/529, loss: 0.621144711971283 2023-01-21 09:23:30.127302: step: 2116/529, loss: 0.058031752705574036 ================================================== Loss: 0.545 -------------------- Dev: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Test: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Korean: {'event': {'p': 0.65, 'r': 0.4126984126984127, 'f1': 0.5048543689320388}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, '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': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5942350332594235, 'r': 0.7137150466045273, 'f1': 0.6485178463399879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5454545454545454, 'r': 0.6578789694427801, 'f1': 0.5964149918522542}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.6904761904761905, 'r': 0.4603174603174603, 'f1': 0.5523809523809524}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Russian: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, '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 09:24:30.164093: step: 4/529, loss: 0.2502412796020508 2023-01-21 09:24:31.319489: step: 8/529, loss: 0.21559953689575195 2023-01-21 09:24:32.546572: step: 12/529, loss: 0.19210587441921234 2023-01-21 09:24:33.803072: step: 16/529, loss: 0.03894238546490669 2023-01-21 09:24:34.976829: step: 20/529, loss: 0.2251969277858734 2023-01-21 09:24:36.154992: step: 24/529, loss: 0.06638240814208984 2023-01-21 09:24:37.417361: step: 28/529, loss: 0.23995256423950195 2023-01-21 09:24:38.624649: step: 32/529, loss: 0.06886114925146103 2023-01-21 09:24:39.881750: step: 36/529, loss: 0.12067346274852753 2023-01-21 09:24:41.083321: step: 40/529, loss: 0.16794681549072266 2023-01-21 09:24:42.259715: step: 44/529, loss: 0.07075369358062744 2023-01-21 09:24:43.463986: step: 48/529, loss: 0.12292347103357315 2023-01-21 09:24:44.653476: step: 52/529, loss: 0.6521369218826294 2023-01-21 09:24:45.838197: step: 56/529, loss: 0.05216493830084801 2023-01-21 09:24:47.015461: step: 60/529, loss: 0.09195118397474289 2023-01-21 09:24:48.205763: step: 64/529, loss: 0.09234986454248428 2023-01-21 09:24:49.381908: step: 68/529, loss: 0.17367076873779297 2023-01-21 09:24:50.599442: step: 72/529, loss: 0.22387734055519104 2023-01-21 09:24:51.838300: step: 76/529, loss: 0.08996899425983429 2023-01-21 09:24:53.055658: step: 80/529, loss: 0.21819201111793518 2023-01-21 09:24:54.240101: step: 84/529, loss: 0.11473675072193146 2023-01-21 09:24:55.458600: step: 88/529, loss: 0.082275390625 2023-01-21 09:24:56.682502: step: 92/529, loss: 0.19086790084838867 2023-01-21 09:24:57.881644: step: 96/529, loss: 0.04892168194055557 2023-01-21 09:24:59.068076: step: 100/529, loss: 0.8130882978439331 2023-01-21 09:25:00.285061: step: 104/529, loss: 0.14128953218460083 2023-01-21 09:25:01.484287: step: 108/529, loss: 0.656221866607666 2023-01-21 09:25:02.706592: step: 112/529, loss: 0.20366695523262024 2023-01-21 09:25:03.947824: step: 116/529, loss: 0.25114861130714417 2023-01-21 09:25:05.106618: step: 120/529, loss: 0.10173463821411133 2023-01-21 09:25:06.322553: step: 124/529, loss: 0.1308383047580719 2023-01-21 09:25:07.498927: step: 128/529, loss: 0.6807863116264343 2023-01-21 09:25:08.659857: step: 132/529, loss: 0.2657139301300049 2023-01-21 09:25:09.880266: step: 136/529, loss: 0.16858011484146118 2023-01-21 09:25:11.146834: step: 140/529, loss: 0.17533636093139648 2023-01-21 09:25:12.350212: step: 144/529, loss: 0.15527839958667755 2023-01-21 09:25:13.599747: step: 148/529, loss: 0.13998566567897797 2023-01-21 09:25:14.784875: step: 152/529, loss: 6.27352237701416 2023-01-21 09:25:16.000117: step: 156/529, loss: 0.7151943445205688 2023-01-21 09:25:17.196272: step: 160/529, loss: 0.3012429177761078 2023-01-21 09:25:18.439306: step: 164/529, loss: 0.11990289390087128 2023-01-21 09:25:19.613608: step: 168/529, loss: 0.13066883385181427 2023-01-21 09:25:20.814921: step: 172/529, loss: 0.629902720451355 2023-01-21 09:25:22.031888: step: 176/529, loss: 0.14125214517116547 2023-01-21 09:25:23.213586: step: 180/529, loss: 0.11176681518554688 2023-01-21 09:25:24.402409: step: 184/529, loss: 0.27583953738212585 2023-01-21 09:25:25.588564: step: 188/529, loss: 0.07218732684850693 2023-01-21 09:25:26.773375: step: 192/529, loss: 6.145091533660889 2023-01-21 09:25:27.960775: step: 196/529, loss: 0.09435997903347015 2023-01-21 09:25:29.183358: step: 200/529, loss: 0.20767942070960999 2023-01-21 09:25:30.391239: step: 204/529, loss: 0.08072328567504883 2023-01-21 09:25:31.631126: step: 208/529, loss: 0.06381388008594513 2023-01-21 09:25:32.818451: step: 212/529, loss: 0.025726724416017532 2023-01-21 09:25:33.999690: step: 216/529, loss: 5.529368877410889 2023-01-21 09:25:35.208776: step: 220/529, loss: 0.28097963333129883 2023-01-21 09:25:36.400205: step: 224/529, loss: 0.10538144409656525 2023-01-21 09:25:37.596042: step: 228/529, loss: 0.6860580444335938 2023-01-21 09:25:38.789328: step: 232/529, loss: 0.44918498396873474 2023-01-21 09:25:39.940633: step: 236/529, loss: 0.17048045992851257 2023-01-21 09:25:41.167679: step: 240/529, loss: 0.10401153564453125 2023-01-21 09:25:42.340610: step: 244/529, loss: 0.18155184388160706 2023-01-21 09:25:43.547387: step: 248/529, loss: 0.6845030784606934 2023-01-21 09:25:44.734558: step: 252/529, loss: 0.3077487051486969 2023-01-21 09:25:45.951682: step: 256/529, loss: 0.09060707688331604 2023-01-21 09:25:47.157200: step: 260/529, loss: 0.30717048048973083 2023-01-21 09:25:48.347548: step: 264/529, loss: 0.08343382179737091 2023-01-21 09:25:49.608391: step: 268/529, loss: 0.47527599334716797 2023-01-21 09:25:50.766262: step: 272/529, loss: 0.024470090866088867 2023-01-21 09:25:51.952733: step: 276/529, loss: 0.30808818340301514 2023-01-21 09:25:53.148277: step: 280/529, loss: 0.7827953100204468 2023-01-21 09:25:54.347554: step: 284/529, loss: 0.7774907350540161 2023-01-21 09:25:55.515126: step: 288/529, loss: 0.03212566301226616 2023-01-21 09:25:56.720625: step: 292/529, loss: 0.9486326575279236 2023-01-21 09:25:57.924820: step: 296/529, loss: 0.2746638357639313 2023-01-21 09:25:59.084268: step: 300/529, loss: 0.16681557893753052 2023-01-21 09:26:00.296847: step: 304/529, loss: 0.08173046261072159 2023-01-21 09:26:01.501400: step: 308/529, loss: 6.288697719573975 2023-01-21 09:26:02.704148: step: 312/529, loss: 0.14964056015014648 2023-01-21 09:26:03.942715: step: 316/529, loss: 0.2742324471473694 2023-01-21 09:26:05.105183: step: 320/529, loss: 0.14716453850269318 2023-01-21 09:26:06.266959: step: 324/529, loss: 0.5346121191978455 2023-01-21 09:26:07.486919: step: 328/529, loss: 0.5574740171432495 2023-01-21 09:26:08.682125: step: 332/529, loss: 0.2730085253715515 2023-01-21 09:26:09.883204: step: 336/529, loss: 0.12539997696876526 2023-01-21 09:26:11.051168: step: 340/529, loss: 0.04188213497400284 2023-01-21 09:26:12.243218: step: 344/529, loss: 0.14948859810829163 2023-01-21 09:26:13.453434: step: 348/529, loss: 0.2436954528093338 2023-01-21 09:26:14.693480: step: 352/529, loss: 1.1195698976516724 2023-01-21 09:26:15.890638: step: 356/529, loss: 0.2025630921125412 2023-01-21 09:26:17.082715: step: 360/529, loss: 0.5436082482337952 2023-01-21 09:26:18.272334: step: 364/529, loss: 0.20505733788013458 2023-01-21 09:26:19.497223: step: 368/529, loss: 0.6219170689582825 2023-01-21 09:26:20.648542: step: 372/529, loss: 0.9745227098464966 2023-01-21 09:26:21.832284: step: 376/529, loss: 0.09567232429981232 2023-01-21 09:26:22.988033: step: 380/529, loss: 0.10994787514209747 2023-01-21 09:26:24.172045: step: 384/529, loss: 0.06637553870677948 2023-01-21 09:26:25.389521: step: 388/529, loss: 0.38644522428512573 2023-01-21 09:26:26.571301: step: 392/529, loss: 0.12255477905273438 2023-01-21 09:26:27.745439: step: 396/529, loss: 0.5542092323303223 2023-01-21 09:26:28.970515: step: 400/529, loss: 0.053899191319942474 2023-01-21 09:26:30.184843: step: 404/529, loss: 0.4054171144962311 2023-01-21 09:26:31.402508: step: 408/529, loss: 0.24461831152439117 2023-01-21 09:26:32.667989: step: 412/529, loss: 0.07134656608104706 2023-01-21 09:26:33.836964: step: 416/529, loss: 0.043059565126895905 2023-01-21 09:26:35.003568: step: 420/529, loss: 0.08745727688074112 2023-01-21 09:26:36.176190: step: 424/529, loss: 0.1223725825548172 2023-01-21 09:26:37.362771: step: 428/529, loss: 0.5513986349105835 2023-01-21 09:26:38.539930: step: 432/529, loss: 1.014221429824829 2023-01-21 09:26:39.756329: step: 436/529, loss: 0.7200706005096436 2023-01-21 09:26:40.913214: step: 440/529, loss: 0.04349174350500107 2023-01-21 09:26:42.051252: step: 444/529, loss: 0.137525275349617 2023-01-21 09:26:43.257302: step: 448/529, loss: 0.7859518527984619 2023-01-21 09:26:44.436307: step: 452/529, loss: 0.6683461666107178 2023-01-21 09:26:45.681280: step: 456/529, loss: 0.2929232716560364 2023-01-21 09:26:46.901848: step: 460/529, loss: 0.3526670038700104 2023-01-21 09:26:48.096969: step: 464/529, loss: 0.12022700905799866 2023-01-21 09:26:49.280743: step: 468/529, loss: 0.17588362097740173 2023-01-21 09:26:50.478176: step: 472/529, loss: 0.08686542510986328 2023-01-21 09:26:51.678124: step: 476/529, loss: 0.23461675643920898 2023-01-21 09:26:52.905906: step: 480/529, loss: 0.14042958617210388 2023-01-21 09:26:54.079047: step: 484/529, loss: 0.07853622734546661 2023-01-21 09:26:55.273695: step: 488/529, loss: 1.4251697063446045 2023-01-21 09:26:56.497475: step: 492/529, loss: 5.416125774383545 2023-01-21 09:26:57.678139: step: 496/529, loss: 0.16637592017650604 2023-01-21 09:26:58.857889: step: 500/529, loss: 0.04373054578900337 2023-01-21 09:27:00.036476: step: 504/529, loss: 0.7165427803993225 2023-01-21 09:27:01.248707: step: 508/529, loss: 0.09049635380506516 2023-01-21 09:27:02.459609: step: 512/529, loss: 0.25090402364730835 2023-01-21 09:27:03.638473: step: 516/529, loss: 0.18650618195533752 2023-01-21 09:27:04.848443: step: 520/529, loss: 0.9127770662307739 2023-01-21 09:27:06.025939: step: 524/529, loss: 0.1572188436985016 2023-01-21 09:27:07.232862: step: 528/529, loss: 0.4635128974914551 2023-01-21 09:27:08.426617: step: 532/529, loss: 0.7403554320335388 2023-01-21 09:27:09.658822: step: 536/529, loss: 0.1569177657365799 2023-01-21 09:27:10.904996: step: 540/529, loss: 0.6742302179336548 2023-01-21 09:27:12.169675: step: 544/529, loss: 0.1890222579240799 2023-01-21 09:27:13.354020: step: 548/529, loss: 0.06375548988580704 2023-01-21 09:27:14.569070: step: 552/529, loss: 0.1572054922580719 2023-01-21 09:27:15.766557: step: 556/529, loss: 0.4539460241794586 2023-01-21 09:27:16.964400: step: 560/529, loss: 0.07919806987047195 2023-01-21 09:27:18.169452: step: 564/529, loss: 0.06230345368385315 2023-01-21 09:27:19.373762: step: 568/529, loss: 0.2389349341392517 2023-01-21 09:27:20.568157: step: 572/529, loss: 0.18917961418628693 2023-01-21 09:27:21.758355: step: 576/529, loss: 0.6497324109077454 2023-01-21 09:27:22.988899: step: 580/529, loss: 0.4957512319087982 2023-01-21 09:27:24.202129: step: 584/529, loss: 0.6842037439346313 2023-01-21 09:27:25.421069: step: 588/529, loss: 0.10157127678394318 2023-01-21 09:27:26.613275: step: 592/529, loss: 0.10126753151416779 2023-01-21 09:27:27.832145: step: 596/529, loss: 0.0788264274597168 2023-01-21 09:27:29.034655: step: 600/529, loss: 0.14935550093650818 2023-01-21 09:27:30.228375: step: 604/529, loss: 0.07220669090747833 2023-01-21 09:27:31.419425: step: 608/529, loss: 0.36246538162231445 2023-01-21 09:27:32.629800: step: 612/529, loss: 0.21938246488571167 2023-01-21 09:27:33.849182: step: 616/529, loss: 0.12169580906629562 2023-01-21 09:27:35.028182: step: 620/529, loss: 0.1293935328722 2023-01-21 09:27:36.246402: step: 624/529, loss: 0.17244437336921692 2023-01-21 09:27:37.500507: step: 628/529, loss: 0.3392561674118042 2023-01-21 09:27:38.741849: step: 632/529, loss: 0.11422643065452576 2023-01-21 09:27:39.980440: step: 636/529, loss: 0.126190185546875 2023-01-21 09:27:41.159076: step: 640/529, loss: 0.37342149019241333 2023-01-21 09:27:42.353612: step: 644/529, loss: 0.24555817246437073 2023-01-21 09:27:43.548132: step: 648/529, loss: 0.6092405319213867 2023-01-21 09:27:44.772048: step: 652/529, loss: 0.05914516746997833 2023-01-21 09:27:45.959922: step: 656/529, loss: 1.1573021411895752 2023-01-21 09:27:47.152307: step: 660/529, loss: 1.499564528465271 2023-01-21 09:27:48.348440: step: 664/529, loss: 0.12421464920043945 2023-01-21 09:27:49.546038: step: 668/529, loss: 0.1295979619026184 2023-01-21 09:27:50.732163: step: 672/529, loss: 0.06339740753173828 2023-01-21 09:27:51.951026: step: 676/529, loss: 0.4435332119464874 2023-01-21 09:27:53.108385: step: 680/529, loss: 0.2729329466819763 2023-01-21 09:27:54.295252: step: 684/529, loss: 0.15566444396972656 2023-01-21 09:27:55.511595: step: 688/529, loss: 0.2621266543865204 2023-01-21 09:27:56.738025: step: 692/529, loss: 0.5348869562149048 2023-01-21 09:27:57.950679: step: 696/529, loss: 1.1988469362258911 2023-01-21 09:27:59.133674: step: 700/529, loss: 0.2524147033691406 2023-01-21 09:28:00.360954: step: 704/529, loss: 0.10733585059642792 2023-01-21 09:28:01.524064: step: 708/529, loss: 0.07933025062084198 2023-01-21 09:28:02.746889: step: 712/529, loss: 0.028967857360839844 2023-01-21 09:28:03.939034: step: 716/529, loss: 0.9275743365287781 2023-01-21 09:28:05.118089: step: 720/529, loss: 0.20581255853176117 2023-01-21 09:28:06.286904: step: 724/529, loss: 0.12088622897863388 2023-01-21 09:28:07.454736: step: 728/529, loss: 0.7049477100372314 2023-01-21 09:28:08.619503: step: 732/529, loss: 0.7286319136619568 2023-01-21 09:28:09.813920: step: 736/529, loss: 0.6800567507743835 2023-01-21 09:28:10.999895: step: 740/529, loss: 0.13806940615177155 2023-01-21 09:28:12.184685: step: 744/529, loss: 0.7309344410896301 2023-01-21 09:28:13.380178: step: 748/529, loss: 0.24343052506446838 2023-01-21 09:28:14.612694: step: 752/529, loss: 0.07532835006713867 2023-01-21 09:28:15.825134: step: 756/529, loss: 0.1570483148097992 2023-01-21 09:28:17.009039: step: 760/529, loss: 0.11138339340686798 2023-01-21 09:28:18.203649: step: 764/529, loss: 0.1902478188276291 2023-01-21 09:28:19.400873: step: 768/529, loss: 0.5234995484352112 2023-01-21 09:28:20.618694: step: 772/529, loss: 0.07961826026439667 2023-01-21 09:28:21.838120: step: 776/529, loss: 0.1992332488298416 2023-01-21 09:28:23.025202: step: 780/529, loss: 0.30524635314941406 2023-01-21 09:28:24.235344: step: 784/529, loss: 0.2621254026889801 2023-01-21 09:28:25.435594: step: 788/529, loss: 6.731590270996094 2023-01-21 09:28:26.657483: step: 792/529, loss: 1.0632907152175903 2023-01-21 09:28:27.872876: step: 796/529, loss: 0.2712801992893219 2023-01-21 09:28:29.078327: step: 800/529, loss: 0.33300885558128357 2023-01-21 09:28:30.283077: step: 804/529, loss: 0.07124600559473038 2023-01-21 09:28:31.490775: step: 808/529, loss: 0.289437860250473 2023-01-21 09:28:32.745907: step: 812/529, loss: 0.10650768131017685 2023-01-21 09:28:33.953603: step: 816/529, loss: 0.2907988429069519 2023-01-21 09:28:35.153538: step: 820/529, loss: 0.15293750166893005 2023-01-21 09:28:36.340574: step: 824/529, loss: 0.04618760570883751 2023-01-21 09:28:37.533619: step: 828/529, loss: 0.7582330703735352 2023-01-21 09:28:38.741389: step: 832/529, loss: 0.1023097038269043 2023-01-21 09:28:39.931020: step: 836/529, loss: 0.19602380692958832 2023-01-21 09:28:41.149279: step: 840/529, loss: 0.39763709902763367 2023-01-21 09:28:42.364188: step: 844/529, loss: 0.16959291696548462 2023-01-21 09:28:43.552172: step: 848/529, loss: 0.11945943534374237 2023-01-21 09:28:44.728731: step: 852/529, loss: 0.7064674496650696 2023-01-21 09:28:45.910751: step: 856/529, loss: 0.07346732914447784 2023-01-21 09:28:47.085239: step: 860/529, loss: 0.7056807279586792 2023-01-21 09:28:48.287068: step: 864/529, loss: 0.11672163754701614 2023-01-21 09:28:49.490731: step: 868/529, loss: 0.8193079233169556 2023-01-21 09:28:50.717526: step: 872/529, loss: 0.17722482979297638 2023-01-21 09:28:51.904738: step: 876/529, loss: 0.5032776594161987 2023-01-21 09:28:53.148255: step: 880/529, loss: 0.22781753540039062 2023-01-21 09:28:54.387282: step: 884/529, loss: 0.33599910140037537 2023-01-21 09:28:55.550647: step: 888/529, loss: 0.24913644790649414 2023-01-21 09:28:56.730799: step: 892/529, loss: 1.0517165660858154 2023-01-21 09:28:57.926206: step: 896/529, loss: 0.023081159219145775 2023-01-21 09:28:59.070522: step: 900/529, loss: 0.079694464802742 2023-01-21 09:29:00.277041: step: 904/529, loss: 0.13987857103347778 2023-01-21 09:29:01.459547: step: 908/529, loss: 0.04917202144861221 2023-01-21 09:29:02.711588: step: 912/529, loss: 0.38371506333351135 2023-01-21 09:29:03.911611: step: 916/529, loss: 0.14735527336597443 2023-01-21 09:29:05.133056: step: 920/529, loss: 0.19159266352653503 2023-01-21 09:29:06.343457: step: 924/529, loss: 0.02148113213479519 2023-01-21 09:29:07.552439: step: 928/529, loss: 0.6060795783996582 2023-01-21 09:29:08.775624: step: 932/529, loss: 0.03179464489221573 2023-01-21 09:29:09.989904: step: 936/529, loss: 0.5949546694755554 2023-01-21 09:29:11.159898: step: 940/529, loss: 0.0662682056427002 2023-01-21 09:29:12.370095: step: 944/529, loss: 0.15894919633865356 2023-01-21 09:29:13.555490: step: 948/529, loss: 0.6855624914169312 2023-01-21 09:29:14.769217: step: 952/529, loss: 0.9729734063148499 2023-01-21 09:29:15.974402: step: 956/529, loss: 0.5432029962539673 2023-01-21 09:29:17.127093: step: 960/529, loss: 0.033452510833740234 2023-01-21 09:29:18.349644: step: 964/529, loss: 0.5826089382171631 2023-01-21 09:29:19.549333: step: 968/529, loss: 0.16668090224266052 2023-01-21 09:29:20.739700: step: 972/529, loss: 0.4389062821865082 2023-01-21 09:29:21.912223: step: 976/529, loss: 0.627124547958374 2023-01-21 09:29:23.061341: step: 980/529, loss: 0.1279161423444748 2023-01-21 09:29:24.230744: step: 984/529, loss: 0.5198164582252502 2023-01-21 09:29:25.465240: step: 988/529, loss: 0.8815890550613403 2023-01-21 09:29:26.645637: step: 992/529, loss: 0.5943320393562317 2023-01-21 09:29:27.880719: step: 996/529, loss: 0.1565295308828354 2023-01-21 09:29:29.049880: step: 1000/529, loss: 0.054601289331912994 2023-01-21 09:29:30.290630: step: 1004/529, loss: 0.33045950531959534 2023-01-21 09:29:31.495598: step: 1008/529, loss: 4.906688213348389 2023-01-21 09:29:32.698576: step: 1012/529, loss: 0.16284996271133423 2023-01-21 09:29:33.917007: step: 1016/529, loss: 0.08170967549085617 2023-01-21 09:29:35.107018: step: 1020/529, loss: 0.07424584031105042 2023-01-21 09:29:36.257288: step: 1024/529, loss: 0.34577685594558716 2023-01-21 09:29:37.420479: step: 1028/529, loss: 0.23452281951904297 2023-01-21 09:29:38.558606: step: 1032/529, loss: 0.3462425172328949 2023-01-21 09:29:39.749946: step: 1036/529, loss: 0.6544706225395203 2023-01-21 09:29:40.921411: step: 1040/529, loss: 0.09467440098524094 2023-01-21 09:29:42.147795: step: 1044/529, loss: 0.2865774929523468 2023-01-21 09:29:43.350187: step: 1048/529, loss: 0.6097820997238159 2023-01-21 09:29:44.559670: step: 1052/529, loss: 0.22973862290382385 2023-01-21 09:29:45.700029: step: 1056/529, loss: 6.003236770629883 2023-01-21 09:29:46.909355: step: 1060/529, loss: 0.16612310707569122 2023-01-21 09:29:48.128570: step: 1064/529, loss: 0.3027915060520172 2023-01-21 09:29:49.316624: step: 1068/529, loss: 0.10988998413085938 2023-01-21 09:29:50.523467: step: 1072/529, loss: 0.2076096534729004 2023-01-21 09:29:51.759811: step: 1076/529, loss: 0.6231120228767395 2023-01-21 09:29:52.965840: step: 1080/529, loss: 0.11262819916009903 2023-01-21 09:29:54.155845: step: 1084/529, loss: 0.10338087379932404 2023-01-21 09:29:55.396784: step: 1088/529, loss: 0.28512048721313477 2023-01-21 09:29:56.616165: step: 1092/529, loss: 0.3338623046875 2023-01-21 09:29:57.805371: step: 1096/529, loss: 0.154316246509552 2023-01-21 09:29:58.993433: step: 1100/529, loss: 0.16560226678848267 2023-01-21 09:30:00.167102: step: 1104/529, loss: 0.21936655044555664 2023-01-21 09:30:01.361959: step: 1108/529, loss: 0.1365387886762619 2023-01-21 09:30:02.539430: step: 1112/529, loss: 0.273162841796875 2023-01-21 09:30:03.725483: step: 1116/529, loss: 0.19150061905384064 2023-01-21 09:30:04.923391: step: 1120/529, loss: 0.14278630912303925 2023-01-21 09:30:06.131009: step: 1124/529, loss: 0.10513119399547577 2023-01-21 09:30:07.350981: step: 1128/529, loss: 0.24738502502441406 2023-01-21 09:30:08.588933: step: 1132/529, loss: 0.14306268095970154 2023-01-21 09:30:09.767421: step: 1136/529, loss: 0.04414534568786621 2023-01-21 09:30:11.013860: step: 1140/529, loss: 0.17272883653640747 2023-01-21 09:30:12.178854: step: 1144/529, loss: 0.05614333599805832 2023-01-21 09:30:13.391190: step: 1148/529, loss: 0.9135578274726868 2023-01-21 09:30:14.568114: step: 1152/529, loss: 0.14959993958473206 2023-01-21 09:30:15.782675: step: 1156/529, loss: 0.17412757873535156 2023-01-21 09:30:16.956232: step: 1160/529, loss: 0.12987175583839417 2023-01-21 09:30:18.140161: step: 1164/529, loss: 0.06797285377979279 2023-01-21 09:30:19.388894: step: 1168/529, loss: 0.1130923479795456 2023-01-21 09:30:20.607163: step: 1172/529, loss: 0.4426470398902893 2023-01-21 09:30:21.818754: step: 1176/529, loss: 0.19118309020996094 2023-01-21 09:30:23.008883: step: 1180/529, loss: 0.1117292195558548 2023-01-21 09:30:24.284343: step: 1184/529, loss: 0.6640169620513916 2023-01-21 09:30:25.497040: step: 1188/529, loss: 4.48140811920166 2023-01-21 09:30:26.661445: step: 1192/529, loss: 0.08659081161022186 2023-01-21 09:30:27.823233: step: 1196/529, loss: 0.03147587925195694 2023-01-21 09:30:29.035561: step: 1200/529, loss: 0.11511115729808807 2023-01-21 09:30:30.248533: step: 1204/529, loss: 0.1763749122619629 2023-01-21 09:30:31.414622: step: 1208/529, loss: 0.16321153938770294 2023-01-21 09:30:32.618306: step: 1212/529, loss: 0.4295472204685211 2023-01-21 09:30:33.779185: step: 1216/529, loss: 0.6507922410964966 2023-01-21 09:30:34.966778: step: 1220/529, loss: 0.26718711853027344 2023-01-21 09:30:36.156672: step: 1224/529, loss: 0.044921256601810455 2023-01-21 09:30:37.335756: step: 1228/529, loss: 0.08841486275196075 2023-01-21 09:30:38.521636: step: 1232/529, loss: 0.10089216381311417 2023-01-21 09:30:39.755583: step: 1236/529, loss: 0.7087823748588562 2023-01-21 09:30:40.970505: step: 1240/529, loss: 0.1036825180053711 2023-01-21 09:30:42.193882: step: 1244/529, loss: 0.20612198114395142 2023-01-21 09:30:43.430858: step: 1248/529, loss: 0.17646685242652893 2023-01-21 09:30:44.644031: step: 1252/529, loss: 0.1127084270119667 2023-01-21 09:30:45.840898: step: 1256/529, loss: 0.46867480874061584 2023-01-21 09:30:47.053005: step: 1260/529, loss: 0.457366406917572 2023-01-21 09:30:48.234514: step: 1264/529, loss: 1.3867372274398804 2023-01-21 09:30:49.451101: step: 1268/529, loss: 0.07479186356067657 2023-01-21 09:30:50.606926: step: 1272/529, loss: 0.2538675367832184 2023-01-21 09:30:51.803421: step: 1276/529, loss: 0.1473599076271057 2023-01-21 09:30:53.018414: step: 1280/529, loss: 0.10108103603124619 2023-01-21 09:30:54.189816: step: 1284/529, loss: 0.372117817401886 2023-01-21 09:30:55.358633: step: 1288/529, loss: 0.12571612000465393 2023-01-21 09:30:56.550115: step: 1292/529, loss: 0.07035856693983078 2023-01-21 09:30:57.735259: step: 1296/529, loss: 0.030086757615208626 2023-01-21 09:30:58.999943: step: 1300/529, loss: 0.21377216279506683 2023-01-21 09:31:00.189889: step: 1304/529, loss: 0.09911099076271057 2023-01-21 09:31:01.467460: step: 1308/529, loss: 0.617196798324585 2023-01-21 09:31:02.660325: step: 1312/529, loss: 0.8181546926498413 2023-01-21 09:31:03.855343: step: 1316/529, loss: 0.2112378031015396 2023-01-21 09:31:05.066525: step: 1320/529, loss: 0.10260926187038422 2023-01-21 09:31:06.248901: step: 1324/529, loss: 0.05346975475549698 2023-01-21 09:31:07.408688: step: 1328/529, loss: 0.3387274742126465 2023-01-21 09:31:08.564345: step: 1332/529, loss: 0.0790284126996994 2023-01-21 09:31:09.747958: step: 1336/529, loss: 0.1004907637834549 2023-01-21 09:31:10.940181: step: 1340/529, loss: 0.18727150559425354 2023-01-21 09:31:12.152747: step: 1344/529, loss: 6.199986934661865 2023-01-21 09:31:13.361231: step: 1348/529, loss: 0.23256847262382507 2023-01-21 09:31:14.566859: step: 1352/529, loss: 0.12678948044776917 2023-01-21 09:31:15.781175: step: 1356/529, loss: 0.16418799757957458 2023-01-21 09:31:16.961617: step: 1360/529, loss: 0.0450749397277832 2023-01-21 09:31:18.205152: step: 1364/529, loss: 0.4133574366569519 2023-01-21 09:31:19.364915: step: 1368/529, loss: 0.16582107543945312 2023-01-21 09:31:20.579930: step: 1372/529, loss: 0.17802810668945312 2023-01-21 09:31:21.753349: step: 1376/529, loss: 0.22307071089744568 2023-01-21 09:31:22.942811: step: 1380/529, loss: 2.0557503700256348 2023-01-21 09:31:24.162426: step: 1384/529, loss: 0.08926315605640411 2023-01-21 09:31:25.337052: step: 1388/529, loss: 0.38156142830848694 2023-01-21 09:31:26.562324: step: 1392/529, loss: 0.6394654512405396 2023-01-21 09:31:27.764845: step: 1396/529, loss: 0.08124236762523651 2023-01-21 09:31:28.954155: step: 1400/529, loss: 0.1441776305437088 2023-01-21 09:31:30.107763: step: 1404/529, loss: 0.20134636759757996 2023-01-21 09:31:31.247382: step: 1408/529, loss: 0.17993207275867462 2023-01-21 09:31:32.431228: step: 1412/529, loss: 0.09152422100305557 2023-01-21 09:31:33.611715: step: 1416/529, loss: 0.40726202726364136 2023-01-21 09:31:34.797831: step: 1420/529, loss: 0.3616032302379608 2023-01-21 09:31:35.968590: step: 1424/529, loss: 0.23492255806922913 2023-01-21 09:31:37.182628: step: 1428/529, loss: 0.05406615510582924 2023-01-21 09:31:38.358832: step: 1432/529, loss: 0.5799874663352966 2023-01-21 09:31:39.522095: step: 1436/529, loss: 0.04527115821838379 2023-01-21 09:31:40.732802: step: 1440/529, loss: 0.5468133091926575 2023-01-21 09:31:41.920315: step: 1444/529, loss: 0.11244764924049377 2023-01-21 09:31:43.119984: step: 1448/529, loss: 0.266893208026886 2023-01-21 09:31:44.309818: step: 1452/529, loss: 0.15133695304393768 2023-01-21 09:31:45.475372: step: 1456/529, loss: 0.189447820186615 2023-01-21 09:31:46.657074: step: 1460/529, loss: 0.05978555977344513 2023-01-21 09:31:47.825474: step: 1464/529, loss: 0.529640257358551 2023-01-21 09:31:48.987213: step: 1468/529, loss: 0.6634248495101929 2023-01-21 09:31:50.205405: step: 1472/529, loss: 0.07440290600061417 2023-01-21 09:31:51.433700: step: 1476/529, loss: 0.11741209030151367 2023-01-21 09:31:52.642914: step: 1480/529, loss: 0.8374691009521484 2023-01-21 09:31:53.835347: step: 1484/529, loss: 0.17753085494041443 2023-01-21 09:31:55.020405: step: 1488/529, loss: 0.3507664203643799 2023-01-21 09:31:56.191967: step: 1492/529, loss: 1.8072805404663086 2023-01-21 09:31:57.371682: step: 1496/529, loss: 1.472825527191162 2023-01-21 09:31:58.566131: step: 1500/529, loss: 0.6865641474723816 2023-01-21 09:31:59.798634: step: 1504/529, loss: 1.0736228227615356 2023-01-21 09:32:00.986876: step: 1508/529, loss: 0.0917641669511795 2023-01-21 09:32:02.152115: step: 1512/529, loss: 0.3716105818748474 2023-01-21 09:32:03.337910: step: 1516/529, loss: 0.3487027585506439 2023-01-21 09:32:04.537298: step: 1520/529, loss: 0.8320213556289673 2023-01-21 09:32:05.708180: step: 1524/529, loss: 0.4757848381996155 2023-01-21 09:32:06.932055: step: 1528/529, loss: 0.18038229644298553 2023-01-21 09:32:08.125749: step: 1532/529, loss: 0.2668381631374359 2023-01-21 09:32:09.316909: step: 1536/529, loss: 0.07612647861242294 2023-01-21 09:32:10.510857: step: 1540/529, loss: 0.21797865629196167 2023-01-21 09:32:11.662479: step: 1544/529, loss: 0.058518316596746445 2023-01-21 09:32:12.854466: step: 1548/529, loss: 0.18121612071990967 2023-01-21 09:32:14.024877: step: 1552/529, loss: 0.1679491102695465 2023-01-21 09:32:15.275536: step: 1556/529, loss: 0.32042181491851807 2023-01-21 09:32:16.462157: step: 1560/529, loss: 0.22156062722206116 2023-01-21 09:32:17.629657: step: 1564/529, loss: 0.3879724144935608 2023-01-21 09:32:18.811142: step: 1568/529, loss: 0.14725805819034576 2023-01-21 09:32:20.002281: step: 1572/529, loss: 0.16188357770442963 2023-01-21 09:32:21.171580: step: 1576/529, loss: 0.12782125174999237 2023-01-21 09:32:22.374590: step: 1580/529, loss: 1.4023945331573486 2023-01-21 09:32:23.571973: step: 1584/529, loss: 0.16383209824562073 2023-01-21 09:32:24.786651: step: 1588/529, loss: 0.4181675910949707 2023-01-21 09:32:25.962782: step: 1592/529, loss: 0.217121422290802 2023-01-21 09:32:27.144624: step: 1596/529, loss: 0.43179693818092346 2023-01-21 09:32:28.353214: step: 1600/529, loss: 0.28129568696022034 2023-01-21 09:32:29.573644: step: 1604/529, loss: 0.1418551355600357 2023-01-21 09:32:30.793457: step: 1608/529, loss: 1.4765256643295288 2023-01-21 09:32:32.009060: step: 1612/529, loss: 0.10199747234582901 2023-01-21 09:32:33.177994: step: 1616/529, loss: 0.4221336543560028 2023-01-21 09:32:34.378753: step: 1620/529, loss: 0.2339315414428711 2023-01-21 09:32:35.586050: step: 1624/529, loss: 5.125759124755859 2023-01-21 09:32:36.736556: step: 1628/529, loss: 0.6543909907341003 2023-01-21 09:32:37.930042: step: 1632/529, loss: 0.0964723601937294 2023-01-21 09:32:39.124842: step: 1636/529, loss: 0.039095688611269 2023-01-21 09:32:40.302236: step: 1640/529, loss: 0.1001126766204834 2023-01-21 09:32:41.475722: step: 1644/529, loss: 0.3113107681274414 2023-01-21 09:32:42.652546: step: 1648/529, loss: 0.24315276741981506 2023-01-21 09:32:43.871889: step: 1652/529, loss: 0.16151729226112366 2023-01-21 09:32:45.044108: step: 1656/529, loss: 1.046234369277954 2023-01-21 09:32:46.252871: step: 1660/529, loss: 0.28707069158554077 2023-01-21 09:32:47.490905: step: 1664/529, loss: 0.4638436436653137 2023-01-21 09:32:48.694509: step: 1668/529, loss: 0.25943848490715027 2023-01-21 09:32:49.893676: step: 1672/529, loss: 0.12710922956466675 2023-01-21 09:32:51.117452: step: 1676/529, loss: 0.08610612154006958 2023-01-21 09:32:52.274098: step: 1680/529, loss: 0.6280315518379211 2023-01-21 09:32:53.449940: step: 1684/529, loss: 0.6920431852340698 2023-01-21 09:32:54.643059: step: 1688/529, loss: 0.18153858184814453 2023-01-21 09:32:55.823751: step: 1692/529, loss: 0.24365907907485962 2023-01-21 09:32:57.035198: step: 1696/529, loss: 0.2582958936691284 2023-01-21 09:32:58.277539: step: 1700/529, loss: 0.7487557530403137 2023-01-21 09:32:59.512682: step: 1704/529, loss: 0.618736982345581 2023-01-21 09:33:00.674624: step: 1708/529, loss: 0.09239549934864044 2023-01-21 09:33:01.863302: step: 1712/529, loss: 0.30387306213378906 2023-01-21 09:33:03.047049: step: 1716/529, loss: 0.9698070287704468 2023-01-21 09:33:04.258381: step: 1720/529, loss: 0.1637330949306488 2023-01-21 09:33:05.450123: step: 1724/529, loss: 0.9275952577590942 2023-01-21 09:33:06.616311: step: 1728/529, loss: 0.3570396304130554 2023-01-21 09:33:07.835338: step: 1732/529, loss: 0.23173418641090393 2023-01-21 09:33:09.033132: step: 1736/529, loss: 0.08866572380065918 2023-01-21 09:33:10.209768: step: 1740/529, loss: 1.144148826599121 2023-01-21 09:33:11.389586: step: 1744/529, loss: 0.16854286193847656 2023-01-21 09:33:12.578348: step: 1748/529, loss: 0.290663480758667 2023-01-21 09:33:13.790889: step: 1752/529, loss: 0.21626943349838257 2023-01-21 09:33:14.991628: step: 1756/529, loss: 0.1097053587436676 2023-01-21 09:33:16.179092: step: 1760/529, loss: 0.18987546861171722 2023-01-21 09:33:17.377746: step: 1764/529, loss: 0.12195196747779846 2023-01-21 09:33:18.563588: step: 1768/529, loss: 0.17239037156105042 2023-01-21 09:33:19.752979: step: 1772/529, loss: 0.04777822643518448 2023-01-21 09:33:20.960005: step: 1776/529, loss: 0.231571763753891 2023-01-21 09:33:22.122033: step: 1780/529, loss: 0.18764953315258026 2023-01-21 09:33:23.297435: step: 1784/529, loss: 0.13542090356349945 2023-01-21 09:33:24.494340: step: 1788/529, loss: 0.13952189683914185 2023-01-21 09:33:25.659327: step: 1792/529, loss: 1.1039947271347046 2023-01-21 09:33:26.825096: step: 1796/529, loss: 0.31469279527664185 2023-01-21 09:33:28.050684: step: 1800/529, loss: 0.141163632273674 2023-01-21 09:33:29.263054: step: 1804/529, loss: 0.13917790353298187 2023-01-21 09:33:30.480872: step: 1808/529, loss: 0.05063953623175621 2023-01-21 09:33:31.762995: step: 1812/529, loss: 0.1642959713935852 2023-01-21 09:33:32.995353: step: 1816/529, loss: 0.018425656482577324 2023-01-21 09:33:34.207390: step: 1820/529, loss: 0.3701033592224121 2023-01-21 09:33:35.418932: step: 1824/529, loss: 0.0518035888671875 2023-01-21 09:33:36.600193: step: 1828/529, loss: 0.02172534540295601 2023-01-21 09:33:37.776825: step: 1832/529, loss: 0.1204461082816124 2023-01-21 09:33:38.967266: step: 1836/529, loss: 1.6124284267425537 2023-01-21 09:33:40.158203: step: 1840/529, loss: 0.6887340545654297 2023-01-21 09:33:41.327878: step: 1844/529, loss: 0.2187584936618805 2023-01-21 09:33:42.490933: step: 1848/529, loss: 0.3393929600715637 2023-01-21 09:33:43.666741: step: 1852/529, loss: 0.13632269203662872 2023-01-21 09:33:44.869512: step: 1856/529, loss: 0.39690670371055603 2023-01-21 09:33:46.075325: step: 1860/529, loss: 0.17722836136817932 2023-01-21 09:33:47.237642: step: 1864/529, loss: 0.1351182907819748 2023-01-21 09:33:48.479227: step: 1868/529, loss: 1.1313598155975342 2023-01-21 09:33:49.671830: step: 1872/529, loss: 0.15648508071899414 2023-01-21 09:33:50.829854: step: 1876/529, loss: 0.1422465443611145 2023-01-21 09:33:51.976256: step: 1880/529, loss: 0.09015731513500214 2023-01-21 09:33:53.176634: step: 1884/529, loss: 0.1626604050397873 2023-01-21 09:33:54.385502: step: 1888/529, loss: 0.10304880142211914 2023-01-21 09:33:55.554444: step: 1892/529, loss: 0.17091742157936096 2023-01-21 09:33:56.753777: step: 1896/529, loss: 0.1544683575630188 2023-01-21 09:33:57.947900: step: 1900/529, loss: 0.13257665932178497 2023-01-21 09:33:59.171862: step: 1904/529, loss: 0.4589420258998871 2023-01-21 09:34:00.384635: step: 1908/529, loss: 0.060416512191295624 2023-01-21 09:34:01.542911: step: 1912/529, loss: 0.1425163298845291 2023-01-21 09:34:02.702470: step: 1916/529, loss: 0.41728153824806213 2023-01-21 09:34:03.918555: step: 1920/529, loss: 0.1548786610364914 2023-01-21 09:34:05.124396: step: 1924/529, loss: 0.13955669105052948 2023-01-21 09:34:06.322592: step: 1928/529, loss: 0.23480787873268127 2023-01-21 09:34:07.488018: step: 1932/529, loss: 0.12438789010047913 2023-01-21 09:34:08.662170: step: 1936/529, loss: 0.3161556124687195 2023-01-21 09:34:09.860405: step: 1940/529, loss: 0.1263858824968338 2023-01-21 09:34:11.081526: step: 1944/529, loss: 0.22066307067871094 2023-01-21 09:34:12.262533: step: 1948/529, loss: 0.12270255386829376 2023-01-21 09:34:13.499484: step: 1952/529, loss: 0.593161940574646 2023-01-21 09:34:14.721158: step: 1956/529, loss: 0.08518419414758682 2023-01-21 09:34:15.926451: step: 1960/529, loss: 0.5108179450035095 2023-01-21 09:34:17.136918: step: 1964/529, loss: 0.6039228439331055 2023-01-21 09:34:18.334794: step: 1968/529, loss: 0.2617681622505188 2023-01-21 09:34:19.552998: step: 1972/529, loss: 0.17124386131763458 2023-01-21 09:34:20.722635: step: 1976/529, loss: 0.07970023155212402 2023-01-21 09:34:21.931618: step: 1980/529, loss: 0.47440657019615173 2023-01-21 09:34:23.126402: step: 1984/529, loss: 0.21349477767944336 2023-01-21 09:34:24.321083: step: 1988/529, loss: 0.10600729286670685 2023-01-21 09:34:25.495657: step: 1992/529, loss: 0.14141102135181427 2023-01-21 09:34:26.703757: step: 1996/529, loss: 0.33906498551368713 2023-01-21 09:34:27.953395: step: 2000/529, loss: 1.9259979724884033 2023-01-21 09:34:29.160623: step: 2004/529, loss: 0.7772596478462219 2023-01-21 09:34:30.340458: step: 2008/529, loss: 0.26714658737182617 2023-01-21 09:34:31.587445: step: 2012/529, loss: 0.26636749505996704 2023-01-21 09:34:32.760181: step: 2016/529, loss: 0.1128692626953125 2023-01-21 09:34:33.956368: step: 2020/529, loss: 0.08774447441101074 2023-01-21 09:34:35.134811: step: 2024/529, loss: 0.34954845905303955 2023-01-21 09:34:36.338352: step: 2028/529, loss: 0.36431077122688293 2023-01-21 09:34:37.536325: step: 2032/529, loss: 0.11794634163379669 2023-01-21 09:34:38.711904: step: 2036/529, loss: 0.1983383744955063 2023-01-21 09:34:39.891571: step: 2040/529, loss: 0.05095238983631134 2023-01-21 09:34:41.082874: step: 2044/529, loss: 0.22152921557426453 2023-01-21 09:34:42.331972: step: 2048/529, loss: 0.14781494438648224 2023-01-21 09:34:43.536532: step: 2052/529, loss: 0.6407216787338257 2023-01-21 09:34:44.726008: step: 2056/529, loss: 0.14101961255073547 2023-01-21 09:34:45.931108: step: 2060/529, loss: 0.42766687273979187 2023-01-21 09:34:47.133686: step: 2064/529, loss: 0.08741907775402069 2023-01-21 09:34:48.313821: step: 2068/529, loss: 0.6224902868270874 2023-01-21 09:34:49.489470: step: 2072/529, loss: 0.4738982319831848 2023-01-21 09:34:50.719039: step: 2076/529, loss: 0.27600136399269104 2023-01-21 09:34:51.939930: step: 2080/529, loss: 0.6062637567520142 2023-01-21 09:34:53.148253: step: 2084/529, loss: 0.09425287693738937 2023-01-21 09:34:54.363405: step: 2088/529, loss: 0.1627206802368164 2023-01-21 09:34:55.560892: step: 2092/529, loss: 0.13426536321640015 2023-01-21 09:34:56.772667: step: 2096/529, loss: 0.08971855789422989 2023-01-21 09:34:58.004122: step: 2100/529, loss: 0.18097352981567383 2023-01-21 09:34:59.174811: step: 2104/529, loss: 0.2610015869140625 2023-01-21 09:35:00.385072: step: 2108/529, loss: 0.7071977853775024 2023-01-21 09:35:01.582645: step: 2112/529, loss: 0.1372658759355545 2023-01-21 09:35:02.757475: step: 2116/529, loss: 0.03658909723162651 ================================================== Loss: 0.422 -------------------- Dev: {'event': {'p': 0.5972944849115505, 'r': 0.7643142476697736, 'f1': 0.6705607476635514}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Test: {'event': {'p': 0.6162162162162163, 'r': 0.7513481126423007, 'f1': 0.6771058315334773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Chinese: {'event': {'p': 0.64, 'r': 0.8888888888888888, 'f1': 0.7441860465116279}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} 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': 4} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5972944849115505, 'r': 0.7643142476697736, 'f1': 0.6705607476635514}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6162162162162163, 'r': 0.7513481126423007, 'f1': 0.6771058315334773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Russian: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Russian: {'event': {'p': 0.6296296296296297, 'r': 0.4722222222222222, 'f1': 0.5396825396825397}, '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 09:35:51.262030: step: 4/529, loss: 0.3368922173976898 2023-01-21 09:35:52.517884: step: 8/529, loss: 0.2498549371957779 2023-01-21 09:35:53.641423: step: 12/529, loss: 0.13210654258728027 2023-01-21 09:35:54.860806: step: 16/529, loss: 0.04694252088665962 2023-01-21 09:35:56.050566: step: 20/529, loss: 0.1581013947725296 2023-01-21 09:35:57.246681: step: 24/529, loss: 0.08795900642871857 2023-01-21 09:35:58.426756: step: 28/529, loss: 0.06205299496650696 2023-01-21 09:35:59.603946: step: 32/529, loss: 0.07818803936243057 2023-01-21 09:36:00.761318: step: 36/529, loss: 0.03181157261133194 2023-01-21 09:36:01.939156: step: 40/529, loss: 0.0711650401353836 2023-01-21 09:36:03.117917: step: 44/529, loss: 0.06321640312671661 2023-01-21 09:36:04.306687: step: 48/529, loss: 0.16439934074878693 2023-01-21 09:36:05.504297: step: 52/529, loss: 0.012914848513901234 2023-01-21 09:36:06.670440: step: 56/529, loss: 0.1339038610458374 2023-01-21 09:36:07.870669: step: 60/529, loss: 0.013389921747148037 2023-01-21 09:36:09.056916: step: 64/529, loss: 0.16260270774364471 2023-01-21 09:36:10.219134: step: 68/529, loss: 0.2935546040534973 2023-01-21 09:36:11.422349: step: 72/529, loss: 0.09496411681175232 2023-01-21 09:36:12.614957: step: 76/529, loss: 0.877576470375061 2023-01-21 09:36:13.789258: step: 80/529, loss: 0.19418397545814514 2023-01-21 09:36:14.948663: step: 84/529, loss: 0.0757625624537468 2023-01-21 09:36:16.085067: step: 88/529, loss: 0.23217211663722992 2023-01-21 09:36:17.243259: step: 92/529, loss: 0.245915025472641 2023-01-21 09:36:18.434014: step: 96/529, loss: 0.4205031394958496 2023-01-21 09:36:19.630898: step: 100/529, loss: 0.0636201873421669 2023-01-21 09:36:20.812096: step: 104/529, loss: 0.20632687211036682 2023-01-21 09:36:22.010142: step: 108/529, loss: 0.06781215965747833 2023-01-21 09:36:23.222366: step: 112/529, loss: 0.23506608605384827 2023-01-21 09:36:24.429273: step: 116/529, loss: 0.03888101875782013 2023-01-21 09:36:25.586114: step: 120/529, loss: 0.27391594648361206 2023-01-21 09:36:26.773954: step: 124/529, loss: 0.13371562957763672 2023-01-21 09:36:27.959097: step: 128/529, loss: 0.7427403926849365 2023-01-21 09:36:29.144446: step: 132/529, loss: 0.19109773635864258 2023-01-21 09:36:30.329956: step: 136/529, loss: 0.11806771904230118 2023-01-21 09:36:31.528538: step: 140/529, loss: 0.0564023032784462 2023-01-21 09:36:32.736916: step: 144/529, loss: 0.17895051836967468 2023-01-21 09:36:33.940876: step: 148/529, loss: 0.09865856170654297 2023-01-21 09:36:35.136500: step: 152/529, loss: 0.02326202392578125 2023-01-21 09:36:36.334715: step: 156/529, loss: 0.14999417960643768 2023-01-21 09:36:37.511778: step: 160/529, loss: 0.104170061647892 2023-01-21 09:36:38.694568: step: 164/529, loss: 0.03118734434247017 2023-01-21 09:36:39.842255: step: 168/529, loss: 0.14105692505836487 2023-01-21 09:36:41.040778: step: 172/529, loss: 1.0737955570220947 2023-01-21 09:36:42.251007: step: 176/529, loss: 0.14666888117790222 2023-01-21 09:36:43.409563: step: 180/529, loss: 0.16274519264698029 2023-01-21 09:36:44.586916: step: 184/529, loss: 0.07790937274694443 2023-01-21 09:36:45.829196: step: 188/529, loss: 0.9040259718894958 2023-01-21 09:36:46.987170: step: 192/529, loss: 0.10623179376125336 2023-01-21 09:36:48.169121: step: 196/529, loss: 0.154289111495018 2023-01-21 09:36:49.363596: step: 200/529, loss: 0.2605861723423004 2023-01-21 09:36:50.568796: step: 204/529, loss: 0.2002088725566864 2023-01-21 09:36:51.756332: step: 208/529, loss: 0.3796895444393158 2023-01-21 09:36:52.954332: step: 212/529, loss: 0.406202495098114 2023-01-21 09:36:54.128064: step: 216/529, loss: 0.44357118010520935 2023-01-21 09:36:55.332665: step: 220/529, loss: 0.10600404441356659 2023-01-21 09:36:56.508834: step: 224/529, loss: 0.006806469056755304 2023-01-21 09:36:57.738049: step: 228/529, loss: 0.054962921887636185 2023-01-21 09:36:58.967323: step: 232/529, loss: 0.4608858823776245 2023-01-21 09:37:00.140745: step: 236/529, loss: 0.03203745186328888 2023-01-21 09:37:01.320648: step: 240/529, loss: 0.33352941274642944 2023-01-21 09:37:02.533669: step: 244/529, loss: 0.11124230176210403 2023-01-21 09:37:03.694417: step: 248/529, loss: 1.1060606241226196 2023-01-21 09:37:04.881014: step: 252/529, loss: 2.1010019779205322 2023-01-21 09:37:06.092172: step: 256/529, loss: 0.10182676464319229 2023-01-21 09:37:07.289254: step: 260/529, loss: 0.19599036872386932 2023-01-21 09:37:08.479026: step: 264/529, loss: 0.23356997966766357 2023-01-21 09:37:09.664683: step: 268/529, loss: 0.15966132283210754 2023-01-21 09:37:10.822584: step: 272/529, loss: 0.40934592485427856 2023-01-21 09:37:12.024926: step: 276/529, loss: 0.40123826265335083 2023-01-21 09:37:13.206773: step: 280/529, loss: 0.15164980292320251 2023-01-21 09:37:14.446827: step: 284/529, loss: 0.10599565505981445 2023-01-21 09:37:15.623659: step: 288/529, loss: 0.3209083676338196 2023-01-21 09:37:16.776521: step: 292/529, loss: 0.10308961570262909 2023-01-21 09:37:17.981054: step: 296/529, loss: 0.3025496006011963 2023-01-21 09:37:19.158702: step: 300/529, loss: 0.42719945311546326 2023-01-21 09:37:20.361424: step: 304/529, loss: 0.17302380502223969 2023-01-21 09:37:21.532682: step: 308/529, loss: 0.32263821363449097 2023-01-21 09:37:22.761717: step: 312/529, loss: 0.293412983417511 2023-01-21 09:37:23.938738: step: 316/529, loss: 0.14710655808448792 2023-01-21 09:37:25.128577: step: 320/529, loss: 0.07069597393274307 2023-01-21 09:37:26.300913: step: 324/529, loss: 0.07201728969812393 2023-01-21 09:37:27.533654: step: 328/529, loss: 0.0860690101981163 2023-01-21 09:37:28.713777: step: 332/529, loss: 0.26549968123435974 2023-01-21 09:37:29.887185: step: 336/529, loss: 0.23117035627365112 2023-01-21 09:37:31.090435: step: 340/529, loss: 0.5455995202064514 2023-01-21 09:37:32.282979: step: 344/529, loss: 0.07775802910327911 2023-01-21 09:37:33.474269: step: 348/529, loss: 0.992932915687561 2023-01-21 09:37:34.696193: step: 352/529, loss: 0.25192126631736755 2023-01-21 09:37:35.890544: step: 356/529, loss: 0.19559749960899353 2023-01-21 09:37:37.106752: step: 360/529, loss: 0.10951748490333557 2023-01-21 09:37:38.275720: step: 364/529, loss: 0.3038797080516815 2023-01-21 09:37:39.499406: step: 368/529, loss: 0.08814611285924911 2023-01-21 09:37:40.725390: step: 372/529, loss: 0.6434275507926941 2023-01-21 09:37:41.950291: step: 376/529, loss: 0.0830361396074295 2023-01-21 09:37:43.126246: step: 380/529, loss: 0.09309542179107666 2023-01-21 09:37:44.304113: step: 384/529, loss: 0.2797517478466034 2023-01-21 09:37:45.540635: step: 388/529, loss: 0.03623028099536896 2023-01-21 09:37:46.734085: step: 392/529, loss: 0.41027534008026123 2023-01-21 09:37:47.919911: step: 396/529, loss: 0.24451595544815063 2023-01-21 09:37:49.079835: step: 400/529, loss: 0.08677756786346436 2023-01-21 09:37:50.331224: step: 404/529, loss: 0.18799933791160583 2023-01-21 09:37:51.555250: step: 408/529, loss: 1.2076491117477417 2023-01-21 09:37:52.776782: step: 412/529, loss: 0.19329233467578888 2023-01-21 09:37:53.986862: step: 416/529, loss: 0.16854381561279297 2023-01-21 09:37:55.178901: step: 420/529, loss: 0.05669412761926651 2023-01-21 09:37:56.387923: step: 424/529, loss: 0.10756617039442062 2023-01-21 09:37:57.597513: step: 428/529, loss: 0.19117392599582672 2023-01-21 09:37:58.772001: step: 432/529, loss: 1.4733712673187256 2023-01-21 09:37:59.958122: step: 436/529, loss: 0.2350231111049652 2023-01-21 09:38:01.141904: step: 440/529, loss: 1.290475845336914 2023-01-21 09:38:02.304229: step: 444/529, loss: 0.08296690136194229 2023-01-21 09:38:03.504013: step: 448/529, loss: 0.43990230560302734 2023-01-21 09:38:04.688902: step: 452/529, loss: 0.19779834151268005 2023-01-21 09:38:05.904683: step: 456/529, loss: 0.10606088489294052 2023-01-21 09:38:07.137471: step: 460/529, loss: 0.06485767662525177 2023-01-21 09:38:08.323849: step: 464/529, loss: 0.25423184037208557 2023-01-21 09:38:09.523612: step: 468/529, loss: 0.15825147926807404 2023-01-21 09:38:10.714241: step: 472/529, loss: 0.23665395379066467 2023-01-21 09:38:11.910031: step: 476/529, loss: 0.5980239510536194 2023-01-21 09:38:13.103283: step: 480/529, loss: 0.07079296559095383 2023-01-21 09:38:14.310857: step: 484/529, loss: 0.2878645956516266 2023-01-21 09:38:15.472560: step: 488/529, loss: 0.1929466277360916 2023-01-21 09:38:16.639691: step: 492/529, loss: 0.015117883682250977 2023-01-21 09:38:17.815803: step: 496/529, loss: 0.12071068584918976 2023-01-21 09:38:18.965533: step: 500/529, loss: 0.3531966805458069 2023-01-21 09:38:20.205735: step: 504/529, loss: 0.4819986820220947 2023-01-21 09:38:21.380658: step: 508/529, loss: 0.2964286804199219 2023-01-21 09:38:22.549382: step: 512/529, loss: 0.16189280152320862 2023-01-21 09:38:23.716455: step: 516/529, loss: 0.0845704972743988 2023-01-21 09:38:24.935991: step: 520/529, loss: 0.2002183049917221 2023-01-21 09:38:26.109714: step: 524/529, loss: 0.014196586795151234 2023-01-21 09:38:27.294624: step: 528/529, loss: 0.544185996055603 2023-01-21 09:38:28.481083: step: 532/529, loss: 0.0861654281616211 2023-01-21 09:38:29.672200: step: 536/529, loss: 0.07937517017126083 2023-01-21 09:38:30.874090: step: 540/529, loss: 0.1246723160147667 2023-01-21 09:38:32.120116: step: 544/529, loss: 0.09098978340625763 2023-01-21 09:38:33.262457: step: 548/529, loss: 0.051527928560972214 2023-01-21 09:38:34.432271: step: 552/529, loss: 0.18235445022583008 2023-01-21 09:38:35.600581: step: 556/529, loss: 0.09235215187072754 2023-01-21 09:38:36.763719: step: 560/529, loss: 0.09273795783519745 2023-01-21 09:38:37.970476: step: 564/529, loss: 0.4784258306026459 2023-01-21 09:38:39.179113: step: 568/529, loss: 0.40999284386634827 2023-01-21 09:38:40.369851: step: 572/529, loss: 0.11626291275024414 2023-01-21 09:38:41.618577: step: 576/529, loss: 0.4656146168708801 2023-01-21 09:38:42.776415: step: 580/529, loss: 0.06487789750099182 2023-01-21 09:38:43.972682: step: 584/529, loss: 0.23979537189006805 2023-01-21 09:38:45.193508: step: 588/529, loss: 0.11722078919410706 2023-01-21 09:38:46.394242: step: 592/529, loss: 0.12201094627380371 2023-01-21 09:38:47.588671: step: 596/529, loss: 0.0864621102809906 2023-01-21 09:38:48.747771: step: 600/529, loss: 0.426220566034317 2023-01-21 09:38:49.913137: step: 604/529, loss: 0.06975936889648438 2023-01-21 09:38:51.115361: step: 608/529, loss: 0.27815327048301697 2023-01-21 09:38:52.282153: step: 612/529, loss: 0.4007023572921753 2023-01-21 09:38:53.480741: step: 616/529, loss: 0.14910751581192017 2023-01-21 09:38:54.667139: step: 620/529, loss: 0.09513149410486221 2023-01-21 09:38:55.847344: step: 624/529, loss: 0.07580986618995667 2023-01-21 09:38:57.032168: step: 628/529, loss: 0.05873217433691025 2023-01-21 09:38:58.250856: step: 632/529, loss: 0.32759779691696167 2023-01-21 09:38:59.410338: step: 636/529, loss: 0.12283525615930557 2023-01-21 09:39:00.634224: step: 640/529, loss: 0.0848771184682846 2023-01-21 09:39:01.815860: step: 644/529, loss: 0.29472485184669495 2023-01-21 09:39:03.005093: step: 648/529, loss: 0.4429854452610016 2023-01-21 09:39:04.193631: step: 652/529, loss: 0.06603134423494339 2023-01-21 09:39:05.419246: step: 656/529, loss: 0.32188865542411804 2023-01-21 09:39:06.587492: step: 660/529, loss: 1.5369899272918701 2023-01-21 09:39:07.755896: step: 664/529, loss: 0.9638265371322632 2023-01-21 09:39:08.937334: step: 668/529, loss: 0.012149143032729626 2023-01-21 09:39:10.144636: step: 672/529, loss: 0.16493645310401917 2023-01-21 09:39:11.343875: step: 676/529, loss: 0.07542910426855087 2023-01-21 09:39:12.482889: step: 680/529, loss: 0.2924463748931885 2023-01-21 09:39:13.714322: step: 684/529, loss: 0.10549774020910263 2023-01-21 09:39:14.927090: step: 688/529, loss: 0.2298925518989563 2023-01-21 09:39:16.163875: step: 692/529, loss: 2.7476649284362793 2023-01-21 09:39:17.363052: step: 696/529, loss: 0.11038355529308319 2023-01-21 09:39:18.554594: step: 700/529, loss: 0.7959449887275696 2023-01-21 09:39:19.746691: step: 704/529, loss: 0.15166091918945312 2023-01-21 09:39:20.915781: step: 708/529, loss: 0.8752228021621704 2023-01-21 09:39:22.101595: step: 712/529, loss: 0.0954316109418869 2023-01-21 09:39:23.312206: step: 716/529, loss: 0.4553895890712738 2023-01-21 09:39:24.468418: step: 720/529, loss: 0.03252401575446129 2023-01-21 09:39:25.612828: step: 724/529, loss: 0.34170404076576233 2023-01-21 09:39:26.788935: step: 728/529, loss: 0.06364612281322479 2023-01-21 09:39:28.027671: step: 732/529, loss: 0.8943905830383301 2023-01-21 09:39:29.225278: step: 736/529, loss: 0.23584184050559998 2023-01-21 09:39:30.478235: step: 740/529, loss: 0.4798815846443176 2023-01-21 09:39:31.692005: step: 744/529, loss: 0.16191625595092773 2023-01-21 09:39:32.914585: step: 748/529, loss: 0.08670292049646378 2023-01-21 09:39:34.099616: step: 752/529, loss: 0.09526281803846359 2023-01-21 09:39:35.280616: step: 756/529, loss: 0.04325075447559357 2023-01-21 09:39:36.440894: step: 760/529, loss: 0.07826395332813263 2023-01-21 09:39:37.633942: step: 764/529, loss: 0.2919696867465973 2023-01-21 09:39:38.865577: step: 768/529, loss: 0.19557294249534607 2023-01-21 09:39:40.038991: step: 772/529, loss: 0.11964817345142365 2023-01-21 09:39:41.230468: step: 776/529, loss: 0.08191376179456711 2023-01-21 09:39:42.398958: step: 780/529, loss: 0.1510431319475174 2023-01-21 09:39:43.608335: step: 784/529, loss: 0.3965839445590973 2023-01-21 09:39:44.783381: step: 788/529, loss: 0.08628664165735245 2023-01-21 09:39:46.015052: step: 792/529, loss: 0.6498953104019165 2023-01-21 09:39:47.212275: step: 796/529, loss: 0.1226348727941513 2023-01-21 09:39:48.406928: step: 800/529, loss: 0.03941622003912926 2023-01-21 09:39:49.617711: step: 804/529, loss: 0.16135630011558533 2023-01-21 09:39:50.855404: step: 808/529, loss: 0.07068949192762375 2023-01-21 09:39:52.045928: step: 812/529, loss: 0.1697521060705185 2023-01-21 09:39:53.227290: step: 816/529, loss: 0.1488906741142273 2023-01-21 09:39:54.436319: step: 820/529, loss: 0.09741143882274628 2023-01-21 09:39:55.655132: step: 824/529, loss: 0.22455954551696777 2023-01-21 09:39:56.887126: step: 828/529, loss: 0.06436805427074432 2023-01-21 09:39:58.122240: step: 832/529, loss: 0.1164170503616333 2023-01-21 09:39:59.312465: step: 836/529, loss: 0.07019386440515518 2023-01-21 09:40:00.523485: step: 840/529, loss: 0.15635260939598083 2023-01-21 09:40:01.746746: step: 844/529, loss: 0.0648404136300087 2023-01-21 09:40:02.935262: step: 848/529, loss: 0.1756184995174408 2023-01-21 09:40:04.150274: step: 852/529, loss: 0.06272697448730469 2023-01-21 09:40:05.366829: step: 856/529, loss: 0.20437927544116974 2023-01-21 09:40:06.565697: step: 860/529, loss: 0.05671968311071396 2023-01-21 09:40:07.736658: step: 864/529, loss: 0.3450182378292084 2023-01-21 09:40:08.919464: step: 868/529, loss: 0.033079661428928375 2023-01-21 09:40:10.130716: step: 872/529, loss: 0.09306202083826065 2023-01-21 09:40:11.328842: step: 876/529, loss: 0.8007081747055054 2023-01-21 09:40:12.524956: step: 880/529, loss: 0.07276339828968048 2023-01-21 09:40:13.726316: step: 884/529, loss: 0.2312414050102234 2023-01-21 09:40:14.930021: step: 888/529, loss: 0.060846708714962006 2023-01-21 09:40:16.096475: step: 892/529, loss: 0.09002557396888733 2023-01-21 09:40:17.293374: step: 896/529, loss: 0.2715110182762146 2023-01-21 09:40:18.491809: step: 900/529, loss: 0.18790817260742188 2023-01-21 09:40:19.740825: step: 904/529, loss: 0.0989135205745697 2023-01-21 09:40:20.996111: step: 908/529, loss: 0.07950916141271591 2023-01-21 09:40:22.177576: step: 912/529, loss: 0.11917848885059357 2023-01-21 09:40:23.384736: step: 916/529, loss: 0.842942476272583 2023-01-21 09:40:24.588053: step: 920/529, loss: 0.035099223256111145 2023-01-21 09:40:25.782393: step: 924/529, loss: 0.06903371959924698 2023-01-21 09:40:26.951849: step: 928/529, loss: 0.10572004318237305 2023-01-21 09:40:28.141241: step: 932/529, loss: 0.06524658203125 2023-01-21 09:40:29.351140: step: 936/529, loss: 0.003122329944744706 2023-01-21 09:40:30.559918: step: 940/529, loss: 0.06736727058887482 2023-01-21 09:40:31.734007: step: 944/529, loss: 0.05806684494018555 2023-01-21 09:40:32.965688: step: 948/529, loss: 0.1061316505074501 2023-01-21 09:40:34.163604: step: 952/529, loss: 0.11372075229883194 2023-01-21 09:40:35.315438: step: 956/529, loss: 0.10416851192712784 2023-01-21 09:40:36.539551: step: 960/529, loss: 0.1355249434709549 2023-01-21 09:40:37.727741: step: 964/529, loss: 0.06759805977344513 2023-01-21 09:40:38.937876: step: 968/529, loss: 0.06649474799633026 2023-01-21 09:40:40.160525: step: 972/529, loss: 0.42227956652641296 2023-01-21 09:40:41.342562: step: 976/529, loss: 0.0725914016366005 2023-01-21 09:40:42.528687: step: 980/529, loss: 0.0722140297293663 2023-01-21 09:40:43.742797: step: 984/529, loss: 0.3616194725036621 2023-01-21 09:40:44.903769: step: 988/529, loss: 0.16700276732444763 2023-01-21 09:40:46.098370: step: 992/529, loss: 0.16695700585842133 2023-01-21 09:40:47.270709: step: 996/529, loss: 0.11583714187145233 2023-01-21 09:40:48.458312: step: 1000/529, loss: 0.17715506255626678 2023-01-21 09:40:49.660446: step: 1004/529, loss: 0.13055677711963654 2023-01-21 09:40:50.841401: step: 1008/529, loss: 0.45041924715042114 2023-01-21 09:40:52.014165: step: 1012/529, loss: 0.07279562950134277 2023-01-21 09:40:53.205254: step: 1016/529, loss: 0.07018999755382538 2023-01-21 09:40:54.399752: step: 1020/529, loss: 0.052582837641239166 2023-01-21 09:40:55.619132: step: 1024/529, loss: 0.07242272794246674 2023-01-21 09:40:56.813146: step: 1028/529, loss: 0.44716396927833557 2023-01-21 09:40:58.003122: step: 1032/529, loss: 0.2427215576171875 2023-01-21 09:40:59.203079: step: 1036/529, loss: 0.11899687349796295 2023-01-21 09:41:00.414603: step: 1040/529, loss: 0.05931386724114418 2023-01-21 09:41:01.618104: step: 1044/529, loss: 0.42251184582710266 2023-01-21 09:41:02.807787: step: 1048/529, loss: 0.04243669658899307 2023-01-21 09:41:04.006515: step: 1052/529, loss: 0.4443243741989136 2023-01-21 09:41:05.183874: step: 1056/529, loss: 0.3045470416545868 2023-01-21 09:41:06.399972: step: 1060/529, loss: 0.49162760376930237 2023-01-21 09:41:07.576480: step: 1064/529, loss: 0.08958248049020767 2023-01-21 09:41:08.781228: step: 1068/529, loss: 0.1678430587053299 2023-01-21 09:41:09.969408: step: 1072/529, loss: 0.3991091847419739 2023-01-21 09:41:11.110625: step: 1076/529, loss: 0.14095774292945862 2023-01-21 09:41:12.279190: step: 1080/529, loss: 0.04410099983215332 2023-01-21 09:41:13.501835: step: 1084/529, loss: 0.12525591254234314 2023-01-21 09:41:14.660219: step: 1088/529, loss: 0.1044493168592453 2023-01-21 09:41:15.873082: step: 1092/529, loss: 0.4947187602519989 2023-01-21 09:41:17.105614: step: 1096/529, loss: 0.060716915875673294 2023-01-21 09:41:18.273676: step: 1100/529, loss: 0.018504096195101738 2023-01-21 09:41:19.493617: step: 1104/529, loss: 1.369433045387268 2023-01-21 09:41:20.707110: step: 1108/529, loss: 0.19306893646717072 2023-01-21 09:41:21.900736: step: 1112/529, loss: 0.16246303915977478 2023-01-21 09:41:23.066798: step: 1116/529, loss: 0.711046040058136 2023-01-21 09:41:24.327549: step: 1120/529, loss: 0.04261474311351776 2023-01-21 09:41:25.508939: step: 1124/529, loss: 0.3693057894706726 2023-01-21 09:41:26.735872: step: 1128/529, loss: 0.23279480636119843 2023-01-21 09:41:27.910601: step: 1132/529, loss: 0.13697485625743866 2023-01-21 09:41:29.110765: step: 1136/529, loss: 0.044436171650886536 2023-01-21 09:41:30.321318: step: 1140/529, loss: 0.16294154524803162 2023-01-21 09:41:31.493253: step: 1144/529, loss: 0.4666368365287781 2023-01-21 09:41:32.673686: step: 1148/529, loss: 0.14895009994506836 2023-01-21 09:41:33.852758: step: 1152/529, loss: 0.29820647835731506 2023-01-21 09:41:35.041917: step: 1156/529, loss: 0.020008422434329987 2023-01-21 09:41:36.256395: step: 1160/529, loss: 0.41993483901023865 2023-01-21 09:41:37.446774: step: 1164/529, loss: 0.4133132994174957 2023-01-21 09:41:38.646004: step: 1168/529, loss: 0.3883378803730011 2023-01-21 09:41:39.836755: step: 1172/529, loss: 0.3750898241996765 2023-01-21 09:41:41.049055: step: 1176/529, loss: 0.14187221229076385 2023-01-21 09:41:42.244162: step: 1180/529, loss: 0.24963007867336273 2023-01-21 09:41:43.444341: step: 1184/529, loss: 0.057286933064460754 2023-01-21 09:41:44.685358: step: 1188/529, loss: 0.12287655472755432 2023-01-21 09:41:45.935455: step: 1192/529, loss: 1.093945860862732 2023-01-21 09:41:47.160949: step: 1196/529, loss: 0.24185018241405487 2023-01-21 09:41:48.335102: step: 1200/529, loss: 0.03350668027997017 2023-01-21 09:41:49.526669: step: 1204/529, loss: 0.10707263648509979 2023-01-21 09:41:50.725996: step: 1208/529, loss: 0.6292132139205933 2023-01-21 09:41:51.915445: step: 1212/529, loss: 0.12353439629077911 2023-01-21 09:41:53.110666: step: 1216/529, loss: 0.03813161700963974 2023-01-21 09:41:54.328991: step: 1220/529, loss: 0.31402310729026794 2023-01-21 09:41:55.519291: step: 1224/529, loss: 0.17011003196239471 2023-01-21 09:41:56.722079: step: 1228/529, loss: 0.09156350791454315 2023-01-21 09:41:57.916692: step: 1232/529, loss: 0.1751844435930252 2023-01-21 09:41:59.098026: step: 1236/529, loss: 0.2659228444099426 2023-01-21 09:42:00.338390: step: 1240/529, loss: 0.3951171040534973 2023-01-21 09:42:01.508012: step: 1244/529, loss: 0.07096286118030548 2023-01-21 09:42:02.703144: step: 1248/529, loss: 0.17628613114356995 2023-01-21 09:42:03.889911: step: 1252/529, loss: 0.1602945327758789 2023-01-21 09:42:05.062841: step: 1256/529, loss: 0.027808405458927155 2023-01-21 09:42:06.218936: step: 1260/529, loss: 0.061707306653261185 2023-01-21 09:42:07.452668: step: 1264/529, loss: 0.2329026758670807 2023-01-21 09:42:08.651929: step: 1268/529, loss: 0.12378893047571182 2023-01-21 09:42:09.871793: step: 1272/529, loss: 0.39805689454078674 2023-01-21 09:42:11.063612: step: 1276/529, loss: 0.611111581325531 2023-01-21 09:42:12.338737: step: 1280/529, loss: 0.5031253695487976 2023-01-21 09:42:13.545757: step: 1284/529, loss: 0.16896553337574005 2023-01-21 09:42:14.724915: step: 1288/529, loss: 0.07348137348890305 2023-01-21 09:42:15.922836: step: 1292/529, loss: 0.09705796837806702 2023-01-21 09:42:17.130573: step: 1296/529, loss: 0.20655202865600586 2023-01-21 09:42:18.358472: step: 1300/529, loss: 0.05219273641705513 2023-01-21 09:42:19.554256: step: 1304/529, loss: 0.024089336395263672 2023-01-21 09:42:20.759748: step: 1308/529, loss: 0.3057098388671875 2023-01-21 09:42:21.942098: step: 1312/529, loss: 0.07620535045862198 2023-01-21 09:42:23.163739: step: 1316/529, loss: 1.2700896263122559 2023-01-21 09:42:24.327951: step: 1320/529, loss: 0.10412168502807617 2023-01-21 09:42:25.543199: step: 1324/529, loss: 0.062221430242061615 2023-01-21 09:42:26.752853: step: 1328/529, loss: 0.10426702350378036 2023-01-21 09:42:27.977211: step: 1332/529, loss: 0.16047362983226776 2023-01-21 09:42:29.120491: step: 1336/529, loss: 0.03934831917285919 2023-01-21 09:42:30.302355: step: 1340/529, loss: 0.24802951514720917 2023-01-21 09:42:31.484704: step: 1344/529, loss: 0.02524590492248535 2023-01-21 09:42:32.705656: step: 1348/529, loss: 0.08456268906593323 2023-01-21 09:42:33.889442: step: 1352/529, loss: 0.04311637952923775 2023-01-21 09:42:35.055848: step: 1356/529, loss: 0.07134446501731873 2023-01-21 09:42:36.228847: step: 1360/529, loss: 0.40787267684936523 2023-01-21 09:42:37.440824: step: 1364/529, loss: 0.10543957352638245 2023-01-21 09:42:38.603507: step: 1368/529, loss: 0.12224235385656357 2023-01-21 09:42:39.820472: step: 1372/529, loss: 0.7178272008895874 2023-01-21 09:42:41.020468: step: 1376/529, loss: 0.056239090859889984 2023-01-21 09:42:42.171638: step: 1380/529, loss: 0.05826739966869354 2023-01-21 09:42:43.325103: step: 1384/529, loss: 0.36607295274734497 2023-01-21 09:42:44.516450: step: 1388/529, loss: 0.11008220165967941 2023-01-21 09:42:45.715903: step: 1392/529, loss: 0.23859873414039612 2023-01-21 09:42:46.900492: step: 1396/529, loss: 0.02314014360308647 2023-01-21 09:42:48.104255: step: 1400/529, loss: 0.18839998543262482 2023-01-21 09:42:49.310583: step: 1404/529, loss: 0.6206230521202087 2023-01-21 09:42:50.503568: step: 1408/529, loss: 0.09049182385206223 2023-01-21 09:42:51.689978: step: 1412/529, loss: 0.2897721230983734 2023-01-21 09:42:52.904501: step: 1416/529, loss: 0.2700129449367523 2023-01-21 09:42:54.095660: step: 1420/529, loss: 0.19877544045448303 2023-01-21 09:42:55.284959: step: 1424/529, loss: 0.0664820671081543 2023-01-21 09:42:56.491372: step: 1428/529, loss: 0.28000402450561523 2023-01-21 09:42:57.675671: step: 1432/529, loss: 0.074510857462883 2023-01-21 09:42:58.874628: step: 1436/529, loss: 0.1641642153263092 2023-01-21 09:43:00.073650: step: 1440/529, loss: 0.15378575026988983 2023-01-21 09:43:01.285155: step: 1444/529, loss: 0.1630905121564865 2023-01-21 09:43:02.436672: step: 1448/529, loss: 0.4707326889038086 2023-01-21 09:43:03.648173: step: 1452/529, loss: 0.3298848271369934 2023-01-21 09:43:04.841045: step: 1456/529, loss: 0.20145931839942932 2023-01-21 09:43:06.049827: step: 1460/529, loss: 0.134362131357193 2023-01-21 09:43:07.264614: step: 1464/529, loss: 0.03931255266070366 2023-01-21 09:43:08.490225: step: 1468/529, loss: 0.10354576259851456 2023-01-21 09:43:09.683223: step: 1472/529, loss: 0.11603575199842453 2023-01-21 09:43:10.859065: step: 1476/529, loss: 0.2502133548259735 2023-01-21 09:43:12.069225: step: 1480/529, loss: 0.06853652000427246 2023-01-21 09:43:13.296866: step: 1484/529, loss: 0.10526428371667862 2023-01-21 09:43:14.516807: step: 1488/529, loss: 0.3817511796951294 2023-01-21 09:43:15.752533: step: 1492/529, loss: 0.346446692943573 2023-01-21 09:43:16.937109: step: 1496/529, loss: 0.9585376977920532 2023-01-21 09:43:18.132348: step: 1500/529, loss: 0.6719222664833069 2023-01-21 09:43:19.315123: step: 1504/529, loss: 0.05418987199664116 2023-01-21 09:43:20.527297: step: 1508/529, loss: 0.20145903527736664 2023-01-21 09:43:21.695747: step: 1512/529, loss: 0.3391339182853699 2023-01-21 09:43:22.883132: step: 1516/529, loss: 0.11726198345422745 2023-01-21 09:43:24.105172: step: 1520/529, loss: 0.23056840896606445 2023-01-21 09:43:25.296865: step: 1524/529, loss: 0.1521472930908203 2023-01-21 09:43:26.456305: step: 1528/529, loss: 0.8188669681549072 2023-01-21 09:43:27.652532: step: 1532/529, loss: 0.16956940293312073 2023-01-21 09:43:28.855552: step: 1536/529, loss: 0.06618394702672958 2023-01-21 09:43:30.061205: step: 1540/529, loss: 0.23513688147068024 2023-01-21 09:43:31.298725: step: 1544/529, loss: 0.05753679573535919 2023-01-21 09:43:32.491337: step: 1548/529, loss: 0.3507155478000641 2023-01-21 09:43:33.695532: step: 1552/529, loss: 0.11184482276439667 2023-01-21 09:43:34.865850: step: 1556/529, loss: 0.12555654346942902 2023-01-21 09:43:36.129088: step: 1560/529, loss: 0.17332783341407776 2023-01-21 09:43:37.294247: step: 1564/529, loss: 0.48004570603370667 2023-01-21 09:43:38.501303: step: 1568/529, loss: 0.2001146674156189 2023-01-21 09:43:39.662027: step: 1572/529, loss: 0.11530779302120209 2023-01-21 09:43:40.862253: step: 1576/529, loss: 0.4709669053554535 2023-01-21 09:43:42.013447: step: 1580/529, loss: 0.21258020401000977 2023-01-21 09:43:43.221471: step: 1584/529, loss: 1.6643543243408203 2023-01-21 09:43:44.396511: step: 1588/529, loss: 0.1019502654671669 2023-01-21 09:43:45.576164: step: 1592/529, loss: 0.6992985606193542 2023-01-21 09:43:46.811115: step: 1596/529, loss: 0.5678982734680176 2023-01-21 09:43:47.995970: step: 1600/529, loss: 0.16735419631004333 2023-01-21 09:43:49.212746: step: 1604/529, loss: 0.2393384575843811 2023-01-21 09:43:50.413191: step: 1608/529, loss: 0.13760320842266083 2023-01-21 09:43:51.589646: step: 1612/529, loss: 0.04805126413702965 2023-01-21 09:43:52.777883: step: 1616/529, loss: 0.3238641619682312 2023-01-21 09:43:54.017270: step: 1620/529, loss: 0.5406720638275146 2023-01-21 09:43:55.233622: step: 1624/529, loss: 0.20301905274391174 2023-01-21 09:43:56.439621: step: 1628/529, loss: 0.3603173494338989 2023-01-21 09:43:57.654336: step: 1632/529, loss: 0.255856990814209 2023-01-21 09:43:58.860444: step: 1636/529, loss: 0.3791975975036621 2023-01-21 09:44:00.004091: step: 1640/529, loss: 0.08984789252281189 2023-01-21 09:44:01.183989: step: 1644/529, loss: 0.15052680671215057 2023-01-21 09:44:02.392320: step: 1648/529, loss: 0.0904684066772461 2023-01-21 09:44:03.557432: step: 1652/529, loss: 0.07586755603551865 2023-01-21 09:44:04.739913: step: 1656/529, loss: 0.08873328566551208 2023-01-21 09:44:05.959074: step: 1660/529, loss: 0.2630959451198578 2023-01-21 09:44:07.094524: step: 1664/529, loss: 0.04309406131505966 2023-01-21 09:44:08.328493: step: 1668/529, loss: 0.2471274435520172 2023-01-21 09:44:09.553960: step: 1672/529, loss: 0.7920916676521301 2023-01-21 09:44:10.771773: step: 1676/529, loss: 0.17498460412025452 2023-01-21 09:44:11.946284: step: 1680/529, loss: 0.298117995262146 2023-01-21 09:44:13.144666: step: 1684/529, loss: 0.5017746686935425 2023-01-21 09:44:14.343903: step: 1688/529, loss: 0.08158578723669052 2023-01-21 09:44:15.522381: step: 1692/529, loss: 0.03380145877599716 2023-01-21 09:44:16.716669: step: 1696/529, loss: 0.5202838182449341 2023-01-21 09:44:17.880907: step: 1700/529, loss: 0.23248635232448578 2023-01-21 09:44:19.048082: step: 1704/529, loss: 0.032161712646484375 2023-01-21 09:44:20.236204: step: 1708/529, loss: 0.017815638333559036 2023-01-21 09:44:21.456902: step: 1712/529, loss: 1.1783852577209473 2023-01-21 09:44:22.617106: step: 1716/529, loss: 0.1213672012090683 2023-01-21 09:44:23.773023: step: 1720/529, loss: 0.04418399557471275 2023-01-21 09:44:24.956619: step: 1724/529, loss: 0.1420421600341797 2023-01-21 09:44:26.160182: step: 1728/529, loss: 0.15293598175048828 2023-01-21 09:44:27.367911: step: 1732/529, loss: 0.14314956963062286 2023-01-21 09:44:28.588771: step: 1736/529, loss: 0.848331093788147 2023-01-21 09:44:29.790382: step: 1740/529, loss: 0.1606319546699524 2023-01-21 09:44:30.989616: step: 1744/529, loss: 1.2704627513885498 2023-01-21 09:44:32.212428: step: 1748/529, loss: 0.28026026487350464 2023-01-21 09:44:33.457182: step: 1752/529, loss: 0.1323471963405609 2023-01-21 09:44:34.666575: step: 1756/529, loss: 0.12400197982788086 2023-01-21 09:44:35.903330: step: 1760/529, loss: 0.18167515099048615 2023-01-21 09:44:37.106949: step: 1764/529, loss: 0.612712025642395 2023-01-21 09:44:38.311007: step: 1768/529, loss: 0.08404932171106339 2023-01-21 09:44:39.516071: step: 1772/529, loss: 0.1690712422132492 2023-01-21 09:44:40.695720: step: 1776/529, loss: 1.1270406246185303 2023-01-21 09:44:41.899473: step: 1780/529, loss: 0.1806797981262207 2023-01-21 09:44:43.108001: step: 1784/529, loss: 0.3205658495426178 2023-01-21 09:44:44.346360: step: 1788/529, loss: 0.47389286756515503 2023-01-21 09:44:45.530100: step: 1792/529, loss: 0.11828117072582245 2023-01-21 09:44:46.701454: step: 1796/529, loss: 0.0999017208814621 2023-01-21 09:44:47.997005: step: 1800/529, loss: 0.43107879161834717 2023-01-21 09:44:49.187844: step: 1804/529, loss: 0.1836099624633789 2023-01-21 09:44:50.366402: step: 1808/529, loss: 0.3601379692554474 2023-01-21 09:44:51.551349: step: 1812/529, loss: 0.06813059002161026 2023-01-21 09:44:52.763847: step: 1816/529, loss: 0.21963539719581604 2023-01-21 09:44:53.955162: step: 1820/529, loss: 0.1584382951259613 2023-01-21 09:44:55.119131: step: 1824/529, loss: 0.16205815970897675 2023-01-21 09:44:56.331841: step: 1828/529, loss: 0.3298429846763611 2023-01-21 09:44:57.571007: step: 1832/529, loss: 0.09752638638019562 2023-01-21 09:44:58.787999: step: 1836/529, loss: 0.24355296790599823 2023-01-21 09:44:59.986403: step: 1840/529, loss: 0.3982164263725281 2023-01-21 09:45:01.160176: step: 1844/529, loss: 0.1338053196668625 2023-01-21 09:45:02.389194: step: 1848/529, loss: 0.1026037186384201 2023-01-21 09:45:03.622759: step: 1852/529, loss: 0.2660253643989563 2023-01-21 09:45:04.783803: step: 1856/529, loss: 0.08683805912733078 2023-01-21 09:45:05.970994: step: 1860/529, loss: 0.09859263896942139 2023-01-21 09:45:07.183450: step: 1864/529, loss: 0.18862372636795044 2023-01-21 09:45:08.370824: step: 1868/529, loss: 0.2622078061103821 2023-01-21 09:45:09.569467: step: 1872/529, loss: 1.7078286409378052 2023-01-21 09:45:10.785280: step: 1876/529, loss: 0.5195013284683228 2023-01-21 09:45:12.018391: step: 1880/529, loss: 0.1314561814069748 2023-01-21 09:45:13.190674: step: 1884/529, loss: 0.13290110230445862 2023-01-21 09:45:14.358939: step: 1888/529, loss: 0.1165735274553299 2023-01-21 09:45:15.531113: step: 1892/529, loss: 0.1444697380065918 2023-01-21 09:45:16.785746: step: 1896/529, loss: 0.325215607881546 2023-01-21 09:45:17.988001: step: 1900/529, loss: 0.2552858293056488 2023-01-21 09:45:19.178567: step: 1904/529, loss: 0.12094173580408096 2023-01-21 09:45:20.357911: step: 1908/529, loss: 0.07685938477516174 2023-01-21 09:45:21.546350: step: 1912/529, loss: 0.041986849159002304 2023-01-21 09:45:22.699649: step: 1916/529, loss: 0.13326062262058258 2023-01-21 09:45:23.872437: step: 1920/529, loss: 0.08943261951208115 2023-01-21 09:45:25.089766: step: 1924/529, loss: 0.33203163743019104 2023-01-21 09:45:26.294941: step: 1928/529, loss: 0.19435596466064453 2023-01-21 09:45:27.518944: step: 1932/529, loss: 0.32233601808547974 2023-01-21 09:45:28.673401: step: 1936/529, loss: 0.06961946189403534 2023-01-21 09:45:29.838210: step: 1940/529, loss: 0.4815933406352997 2023-01-21 09:45:31.031348: step: 1944/529, loss: 3.1408493518829346 2023-01-21 09:45:32.256256: step: 1948/529, loss: 1.1807891130447388 2023-01-21 09:45:33.476958: step: 1952/529, loss: 0.05921440199017525 2023-01-21 09:45:34.739124: step: 1956/529, loss: 0.6824713945388794 2023-01-21 09:45:35.941345: step: 1960/529, loss: 0.1524069756269455 2023-01-21 09:45:37.118871: step: 1964/529, loss: 0.035631559789180756 2023-01-21 09:45:38.326754: step: 1968/529, loss: 0.00496253976598382 2023-01-21 09:45:39.485076: step: 1972/529, loss: 0.27055779099464417 2023-01-21 09:45:40.657851: step: 1976/529, loss: 0.046446993947029114 2023-01-21 09:45:41.852065: step: 1980/529, loss: 0.2582065463066101 2023-01-21 09:45:43.083443: step: 1984/529, loss: 0.11286059021949768 2023-01-21 09:45:44.264018: step: 1988/529, loss: 0.12138843536376953 2023-01-21 09:45:45.415466: step: 1992/529, loss: 0.6295469403266907 2023-01-21 09:45:46.603922: step: 1996/529, loss: 0.07660198211669922 2023-01-21 09:45:47.753111: step: 2000/529, loss: 0.5249840021133423 2023-01-21 09:45:48.950392: step: 2004/529, loss: 0.21708515286445618 2023-01-21 09:45:50.123909: step: 2008/529, loss: 0.08994626998901367 2023-01-21 09:45:51.322275: step: 2012/529, loss: 0.1991274356842041 2023-01-21 09:45:52.501459: step: 2016/529, loss: 0.03229856491088867 2023-01-21 09:45:53.725483: step: 2020/529, loss: 0.11284971237182617 2023-01-21 09:45:54.911932: step: 2024/529, loss: 0.1971236765384674 2023-01-21 09:45:56.078380: step: 2028/529, loss: 0.029836272820830345 2023-01-21 09:45:57.286708: step: 2032/529, loss: 0.6303896307945251 2023-01-21 09:45:58.485788: step: 2036/529, loss: 0.05511905997991562 2023-01-21 09:45:59.728792: step: 2040/529, loss: 0.37411537766456604 2023-01-21 09:46:00.904986: step: 2044/529, loss: 0.0494970828294754 2023-01-21 09:46:02.145185: step: 2048/529, loss: 0.05961489677429199 2023-01-21 09:46:03.344362: step: 2052/529, loss: 0.14190036058425903 2023-01-21 09:46:04.544623: step: 2056/529, loss: 0.15941821038722992 2023-01-21 09:46:05.708328: step: 2060/529, loss: 0.04620103910565376 2023-01-21 09:46:06.934752: step: 2064/529, loss: 0.1399737000465393 2023-01-21 09:46:08.159693: step: 2068/529, loss: 0.2921361029148102 2023-01-21 09:46:09.343909: step: 2072/529, loss: 0.3828907012939453 2023-01-21 09:46:10.538654: step: 2076/529, loss: 0.18975695967674255 2023-01-21 09:46:11.742136: step: 2080/529, loss: 0.053339533507823944 2023-01-21 09:46:12.907959: step: 2084/529, loss: 1.0323189496994019 2023-01-21 09:46:14.102623: step: 2088/529, loss: 0.08985204249620438 2023-01-21 09:46:15.271804: step: 2092/529, loss: 0.5369731187820435 2023-01-21 09:46:16.459066: step: 2096/529, loss: 0.05523400753736496 2023-01-21 09:46:17.668700: step: 2100/529, loss: 0.12967653572559357 2023-01-21 09:46:18.934567: step: 2104/529, loss: 0.4421672821044922 2023-01-21 09:46:20.121266: step: 2108/529, loss: 0.07011689990758896 2023-01-21 09:46:21.309097: step: 2112/529, loss: 0.2283024787902832 2023-01-21 09:46:22.527252: step: 2116/529, loss: 0.23045654594898224 ================================================== Loss: 0.258 -------------------- Dev: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Test: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Chinese: {'event': {'p': 0.6, 'r': 0.7777777777777778, 'f1': 0.6774193548387097}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Korean: {'event': {'p': 0.5396825396825397, 'r': 0.5396825396825397, 'f1': 0.5396825396825397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5972944849115505, 'r': 0.7643142476697736, 'f1': 0.6705607476635514}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6162162162162163, 'r': 0.7513481126423007, 'f1': 0.6771058315334773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 6 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:47:10.595553: step: 4/529, loss: 0.10589957237243652 2023-01-21 09:47:11.764354: step: 8/529, loss: 0.21908380091190338 2023-01-21 09:47:12.940520: step: 12/529, loss: 0.07282567024230957 2023-01-21 09:47:14.126206: step: 16/529, loss: 0.6927698850631714 2023-01-21 09:47:15.292674: step: 20/529, loss: 0.16808386147022247 2023-01-21 09:47:16.496776: step: 24/529, loss: 0.3427462577819824 2023-01-21 09:47:17.676528: step: 28/529, loss: 0.07606158405542374 2023-01-21 09:47:18.860888: step: 32/529, loss: 0.17398712038993835 2023-01-21 09:47:20.054370: step: 36/529, loss: 0.08700942993164062 2023-01-21 09:47:21.219520: step: 40/529, loss: 0.16824179887771606 2023-01-21 09:47:22.405274: step: 44/529, loss: 0.03893642500042915 2023-01-21 09:47:23.601907: step: 48/529, loss: 0.18418365716934204 2023-01-21 09:47:24.826974: step: 52/529, loss: 0.14820843935012817 2023-01-21 09:47:26.031748: step: 56/529, loss: 0.02458428218960762 2023-01-21 09:47:27.199068: step: 60/529, loss: 0.07334604114294052 2023-01-21 09:47:28.433012: step: 64/529, loss: 0.13620567321777344 2023-01-21 09:47:29.653546: step: 68/529, loss: 0.04479523003101349 2023-01-21 09:47:30.837386: step: 72/529, loss: 0.10311079025268555 2023-01-21 09:47:32.010626: step: 76/529, loss: 0.17127513885498047 2023-01-21 09:47:33.224944: step: 80/529, loss: 0.12035751342773438 2023-01-21 09:47:34.435521: step: 84/529, loss: 0.07973573356866837 2023-01-21 09:47:35.604742: step: 88/529, loss: 0.17489276826381683 2023-01-21 09:47:36.820990: step: 92/529, loss: 0.04985666275024414 2023-01-21 09:47:37.988093: step: 96/529, loss: 0.12287703156471252 2023-01-21 09:47:39.193682: step: 100/529, loss: 0.22918052971363068 2023-01-21 09:47:40.369766: step: 104/529, loss: 0.07233639061450958 2023-01-21 09:47:41.522168: step: 108/529, loss: 0.09398073703050613 2023-01-21 09:47:42.722666: step: 112/529, loss: 0.23602047562599182 2023-01-21 09:47:43.884696: step: 116/529, loss: 0.09541550278663635 2023-01-21 09:47:45.050453: step: 120/529, loss: 0.017404623329639435 2023-01-21 09:47:46.261589: step: 124/529, loss: 0.6073402762413025 2023-01-21 09:47:47.447629: step: 128/529, loss: 0.09669695794582367 2023-01-21 09:47:48.606930: step: 132/529, loss: 0.44639524817466736 2023-01-21 09:47:49.792944: step: 136/529, loss: 0.3745598793029785 2023-01-21 09:47:50.993687: step: 140/529, loss: 0.16365870833396912 2023-01-21 09:47:52.146174: step: 144/529, loss: 0.2182145118713379 2023-01-21 09:47:53.367767: step: 148/529, loss: 0.12828445434570312 2023-01-21 09:47:54.568310: step: 152/529, loss: 0.09437441825866699 2023-01-21 09:47:55.766958: step: 156/529, loss: 0.10758233070373535 2023-01-21 09:47:56.940306: step: 160/529, loss: 0.08210039138793945 2023-01-21 09:47:58.128299: step: 164/529, loss: 0.14318743348121643 2023-01-21 09:47:59.372949: step: 168/529, loss: 0.761003851890564 2023-01-21 09:48:00.530341: step: 172/529, loss: 0.17744722962379456 2023-01-21 09:48:01.765375: step: 176/529, loss: 0.05552082136273384 2023-01-21 09:48:02.954813: step: 180/529, loss: 0.040445901453495026 2023-01-21 09:48:04.120323: step: 184/529, loss: 0.22667713463306427 2023-01-21 09:48:05.342223: step: 188/529, loss: 0.09442977607250214 2023-01-21 09:48:06.525437: step: 192/529, loss: 0.04959920048713684 2023-01-21 09:48:07.706530: step: 196/529, loss: 0.12227264046669006 2023-01-21 09:48:08.867374: step: 200/529, loss: 0.08564300835132599 2023-01-21 09:48:10.039158: step: 204/529, loss: 0.2260417342185974 2023-01-21 09:48:11.274235: step: 208/529, loss: 0.3497160077095032 2023-01-21 09:48:12.447598: step: 212/529, loss: 0.08734989166259766 2023-01-21 09:48:13.634353: step: 216/529, loss: 0.3192896842956543 2023-01-21 09:48:14.847425: step: 220/529, loss: 0.07175035774707794 2023-01-21 09:48:16.012062: step: 224/529, loss: 0.05217437818646431 2023-01-21 09:48:17.205682: step: 228/529, loss: 0.22653542459011078 2023-01-21 09:48:18.396571: step: 232/529, loss: 0.05051755905151367 2023-01-21 09:48:19.592666: step: 236/529, loss: 0.09453191608190536 2023-01-21 09:48:20.781779: step: 240/529, loss: 0.3061576783657074 2023-01-21 09:48:22.001212: step: 244/529, loss: 0.5732786059379578 2023-01-21 09:48:23.156419: step: 248/529, loss: 0.08783912658691406 2023-01-21 09:48:24.334939: step: 252/529, loss: 0.08129101246595383 2023-01-21 09:48:25.517238: step: 256/529, loss: 0.10382424294948578 2023-01-21 09:48:26.695512: step: 260/529, loss: 0.03250408172607422 2023-01-21 09:48:27.888172: step: 264/529, loss: 0.25739455223083496 2023-01-21 09:48:29.088823: step: 268/529, loss: 0.4122230112552643 2023-01-21 09:48:30.274348: step: 272/529, loss: 0.0798913985490799 2023-01-21 09:48:31.457340: step: 276/529, loss: 0.12691421806812286 2023-01-21 09:48:32.673624: step: 280/529, loss: 0.035292983055114746 2023-01-21 09:48:33.878670: step: 284/529, loss: 0.2380600869655609 2023-01-21 09:48:35.037442: step: 288/529, loss: 0.044199660420417786 2023-01-21 09:48:36.263269: step: 292/529, loss: 0.14047423005104065 2023-01-21 09:48:37.495003: step: 296/529, loss: 0.13608382642269135 2023-01-21 09:48:38.690579: step: 300/529, loss: 0.7455171346664429 2023-01-21 09:48:39.852405: step: 304/529, loss: 0.04918579012155533 2023-01-21 09:48:41.049488: step: 308/529, loss: 0.23515090346336365 2023-01-21 09:48:42.251031: step: 312/529, loss: 0.3390515446662903 2023-01-21 09:48:43.486190: step: 316/529, loss: 0.12649232149124146 2023-01-21 09:48:44.653629: step: 320/529, loss: 0.07180137932300568 2023-01-21 09:48:45.847676: step: 324/529, loss: 0.13320092856884003 2023-01-21 09:48:47.012443: step: 328/529, loss: 0.0655854195356369 2023-01-21 09:48:48.252909: step: 332/529, loss: 0.222435861825943 2023-01-21 09:48:49.456892: step: 336/529, loss: 0.03322477638721466 2023-01-21 09:48:50.620146: step: 340/529, loss: 0.09303388744592667 2023-01-21 09:48:51.826026: step: 344/529, loss: 0.6150559186935425 2023-01-21 09:48:52.993055: step: 348/529, loss: 0.22516921162605286 2023-01-21 09:48:54.211581: step: 352/529, loss: 0.06781677901744843 2023-01-21 09:48:55.378993: step: 356/529, loss: 0.3233199119567871 2023-01-21 09:48:56.593127: step: 360/529, loss: 1.7418625354766846 2023-01-21 09:48:57.752315: step: 364/529, loss: 0.11892616003751755 2023-01-21 09:48:58.966768: step: 368/529, loss: 0.08918438106775284 2023-01-21 09:49:00.185710: step: 372/529, loss: 0.4355054795742035 2023-01-21 09:49:01.415820: step: 376/529, loss: 0.2471255660057068 2023-01-21 09:49:02.622733: step: 380/529, loss: 0.12259600311517715 2023-01-21 09:49:03.819440: step: 384/529, loss: 0.1358955353498459 2023-01-21 09:49:05.093124: step: 388/529, loss: 0.779319167137146 2023-01-21 09:49:06.307630: step: 392/529, loss: 0.21276387572288513 2023-01-21 09:49:07.505772: step: 396/529, loss: 0.17627926170825958 2023-01-21 09:49:08.716471: step: 400/529, loss: 0.014339685440063477 2023-01-21 09:49:09.898419: step: 404/529, loss: 0.42780694365501404 2023-01-21 09:49:11.142774: step: 408/529, loss: 0.15504422783851624 2023-01-21 09:49:12.300569: step: 412/529, loss: 0.08078651875257492 2023-01-21 09:49:13.488593: step: 416/529, loss: 0.10272950679063797 2023-01-21 09:49:14.684237: step: 420/529, loss: 0.03373727947473526 2023-01-21 09:49:15.852957: step: 424/529, loss: 0.051468562334775925 2023-01-21 09:49:17.054661: step: 428/529, loss: 0.21358510851860046 2023-01-21 09:49:18.273812: step: 432/529, loss: 0.14000442624092102 2023-01-21 09:49:19.459813: step: 436/529, loss: 0.18310546875 2023-01-21 09:49:20.639117: step: 440/529, loss: 0.046625807881355286 2023-01-21 09:49:21.822178: step: 444/529, loss: 0.10094237327575684 2023-01-21 09:49:22.983456: step: 448/529, loss: 0.10158684104681015 2023-01-21 09:49:24.180070: step: 452/529, loss: 0.45576751232147217 2023-01-21 09:49:25.395400: step: 456/529, loss: 0.04575367271900177 2023-01-21 09:49:26.586718: step: 460/529, loss: 0.03559841960668564 2023-01-21 09:49:27.774503: step: 464/529, loss: 0.03655586391687393 2023-01-21 09:49:28.972060: step: 468/529, loss: 0.07966232299804688 2023-01-21 09:49:30.176394: step: 472/529, loss: 0.06193351745605469 2023-01-21 09:49:31.378075: step: 476/529, loss: 0.18272274732589722 2023-01-21 09:49:32.575467: step: 480/529, loss: 0.045558929443359375 2023-01-21 09:49:33.761384: step: 484/529, loss: 0.2142782211303711 2023-01-21 09:49:34.946567: step: 488/529, loss: 0.11159686744213104 2023-01-21 09:49:36.123236: step: 492/529, loss: 1.1090492010116577 2023-01-21 09:49:37.308667: step: 496/529, loss: 0.44914522767066956 2023-01-21 09:49:38.527322: step: 500/529, loss: 0.12861518561840057 2023-01-21 09:49:39.749743: step: 504/529, loss: 0.0794256180524826 2023-01-21 09:49:40.912839: step: 508/529, loss: 1.1489883661270142 2023-01-21 09:49:42.078573: step: 512/529, loss: 0.3028290271759033 2023-01-21 09:49:43.252257: step: 516/529, loss: 0.4474167823791504 2023-01-21 09:49:44.421620: step: 520/529, loss: 0.05286760628223419 2023-01-21 09:49:45.642091: step: 524/529, loss: 0.0742516964673996 2023-01-21 09:49:46.812350: step: 528/529, loss: 0.13036441802978516 2023-01-21 09:49:47.989405: step: 532/529, loss: 0.0628504753112793 2023-01-21 09:49:49.162418: step: 536/529, loss: 0.26407405734062195 2023-01-21 09:49:50.383099: step: 540/529, loss: 0.1905006766319275 2023-01-21 09:49:51.595033: step: 544/529, loss: 0.09538622200489044 2023-01-21 09:49:52.801158: step: 548/529, loss: 0.06817150115966797 2023-01-21 09:49:54.012862: step: 552/529, loss: 0.09002959728240967 2023-01-21 09:49:55.152232: step: 556/529, loss: 0.06637854874134064 2023-01-21 09:49:56.369941: step: 560/529, loss: 0.14218568801879883 2023-01-21 09:49:57.540759: step: 564/529, loss: 0.10075482726097107 2023-01-21 09:49:58.703468: step: 568/529, loss: 0.0855649933218956 2023-01-21 09:49:59.893247: step: 572/529, loss: 0.05151872709393501 2023-01-21 09:50:01.112751: step: 576/529, loss: 0.09031935036182404 2023-01-21 09:50:02.254571: step: 580/529, loss: 0.09525956958532333 2023-01-21 09:50:03.490944: step: 584/529, loss: 0.10757496953010559 2023-01-21 09:50:04.638537: step: 588/529, loss: 0.06247062608599663 2023-01-21 09:50:05.815804: step: 592/529, loss: 0.1461045742034912 2023-01-21 09:50:06.997719: step: 596/529, loss: 0.05201916769146919 2023-01-21 09:50:08.187755: step: 600/529, loss: 0.1091095432639122 2023-01-21 09:50:09.419280: step: 604/529, loss: 0.13102392852306366 2023-01-21 09:50:10.597034: step: 608/529, loss: 0.1188901960849762 2023-01-21 09:50:11.769685: step: 612/529, loss: 0.031339120119810104 2023-01-21 09:50:12.964929: step: 616/529, loss: 0.17421989142894745 2023-01-21 09:50:14.145072: step: 620/529, loss: 0.2147742360830307 2023-01-21 09:50:15.386145: step: 624/529, loss: 0.028879739344120026 2023-01-21 09:50:16.637803: step: 628/529, loss: 0.048896029591560364 2023-01-21 09:50:17.855095: step: 632/529, loss: 0.06603431701660156 2023-01-21 09:50:19.022179: step: 636/529, loss: 0.2266365885734558 2023-01-21 09:50:20.231532: step: 640/529, loss: 0.517035186290741 2023-01-21 09:50:21.419898: step: 644/529, loss: 0.10150003433227539 2023-01-21 09:50:22.623425: step: 648/529, loss: 0.18182729184627533 2023-01-21 09:50:23.848692: step: 652/529, loss: 0.08189000934362411 2023-01-21 09:50:25.025067: step: 656/529, loss: 0.746856689453125 2023-01-21 09:50:26.220057: step: 660/529, loss: 0.16903123259544373 2023-01-21 09:50:27.438393: step: 664/529, loss: 0.27688655257225037 2023-01-21 09:50:28.629139: step: 668/529, loss: 0.13444343209266663 2023-01-21 09:50:29.819597: step: 672/529, loss: 0.09752359986305237 2023-01-21 09:50:30.993921: step: 676/529, loss: 0.14398424327373505 2023-01-21 09:50:32.218246: step: 680/529, loss: 0.0674186646938324 2023-01-21 09:50:33.396329: step: 684/529, loss: 0.05894365906715393 2023-01-21 09:50:34.607694: step: 688/529, loss: 0.15008696913719177 2023-01-21 09:50:35.783748: step: 692/529, loss: 0.05433502420783043 2023-01-21 09:50:36.955033: step: 696/529, loss: 0.06244697794318199 2023-01-21 09:50:38.154729: step: 700/529, loss: 0.11346426606178284 2023-01-21 09:50:39.362926: step: 704/529, loss: 0.31059932708740234 2023-01-21 09:50:40.534307: step: 708/529, loss: 0.42036062479019165 2023-01-21 09:50:41.706399: step: 712/529, loss: 0.17736978828907013 2023-01-21 09:50:42.899800: step: 716/529, loss: 0.20522668957710266 2023-01-21 09:50:44.075543: step: 720/529, loss: 0.13032397627830505 2023-01-21 09:50:45.293511: step: 724/529, loss: 0.15366163849830627 2023-01-21 09:50:46.526311: step: 728/529, loss: 0.17785649001598358 2023-01-21 09:50:47.730145: step: 732/529, loss: 0.07725667953491211 2023-01-21 09:50:48.929661: step: 736/529, loss: 0.327543169260025 2023-01-21 09:50:50.076486: step: 740/529, loss: 0.11206570267677307 2023-01-21 09:50:51.252119: step: 744/529, loss: 0.1620674431324005 2023-01-21 09:50:52.494235: step: 748/529, loss: 0.09441623091697693 2023-01-21 09:50:53.701584: step: 752/529, loss: 0.08023242652416229 2023-01-21 09:50:54.907118: step: 756/529, loss: 0.14856243133544922 2023-01-21 09:50:56.118211: step: 760/529, loss: 0.20588092505931854 2023-01-21 09:50:57.320919: step: 764/529, loss: 0.35215798020362854 2023-01-21 09:50:58.560660: step: 768/529, loss: 0.08385610580444336 2023-01-21 09:50:59.759937: step: 772/529, loss: 0.45255520939826965 2023-01-21 09:51:00.943400: step: 776/529, loss: 0.17122679948806763 2023-01-21 09:51:02.130744: step: 780/529, loss: 0.1269122064113617 2023-01-21 09:51:03.318204: step: 784/529, loss: 0.19283480942249298 2023-01-21 09:51:04.493996: step: 788/529, loss: 0.3809739947319031 2023-01-21 09:51:05.664993: step: 792/529, loss: 0.17116416990756989 2023-01-21 09:51:06.811867: step: 796/529, loss: 0.09271755069494247 2023-01-21 09:51:08.025835: step: 800/529, loss: 0.3224680423736572 2023-01-21 09:51:09.225304: step: 804/529, loss: 0.08264753967523575 2023-01-21 09:51:10.361855: step: 808/529, loss: 0.1487569361925125 2023-01-21 09:51:11.588689: step: 812/529, loss: 0.21787682175636292 2023-01-21 09:51:12.761615: step: 816/529, loss: 0.14486685395240784 2023-01-21 09:51:13.964481: step: 820/529, loss: 0.06020841747522354 2023-01-21 09:51:15.138686: step: 824/529, loss: 0.08215351402759552 2023-01-21 09:51:16.339170: step: 828/529, loss: 0.29529374837875366 2023-01-21 09:51:17.524535: step: 832/529, loss: 0.09386768937110901 2023-01-21 09:51:18.718819: step: 836/529, loss: 0.10929146409034729 2023-01-21 09:51:19.882280: step: 840/529, loss: 0.10478658974170685 2023-01-21 09:51:21.125077: step: 844/529, loss: 0.04086117446422577 2023-01-21 09:51:22.319031: step: 848/529, loss: 0.14262591302394867 2023-01-21 09:51:23.503850: step: 852/529, loss: 0.2966577112674713 2023-01-21 09:51:24.685250: step: 856/529, loss: 0.07949323952198029 2023-01-21 09:51:25.852496: step: 860/529, loss: 0.06210136413574219 2023-01-21 09:51:27.054265: step: 864/529, loss: 0.12779837846755981 2023-01-21 09:51:28.235483: step: 868/529, loss: 0.0899849385023117 2023-01-21 09:51:29.471549: step: 872/529, loss: 0.19754600524902344 2023-01-21 09:51:30.660101: step: 876/529, loss: 0.06834087520837784 2023-01-21 09:51:31.885964: step: 880/529, loss: 0.39172792434692383 2023-01-21 09:51:33.074193: step: 884/529, loss: 0.2842188775539398 2023-01-21 09:51:34.254169: step: 888/529, loss: 1.0451123714447021 2023-01-21 09:51:35.410961: step: 892/529, loss: 0.0974455326795578 2023-01-21 09:51:36.595078: step: 896/529, loss: 0.08310942351818085 2023-01-21 09:51:37.809123: step: 900/529, loss: 0.3428930342197418 2023-01-21 09:51:39.037424: step: 904/529, loss: 0.6722579598426819 2023-01-21 09:51:40.201531: step: 908/529, loss: 0.09764671325683594 2023-01-21 09:51:41.375404: step: 912/529, loss: 0.13239383697509766 2023-01-21 09:51:42.553840: step: 916/529, loss: 0.7763820886611938 2023-01-21 09:51:43.743500: step: 920/529, loss: 0.20144206285476685 2023-01-21 09:51:44.965069: step: 924/529, loss: 0.05530283600091934 2023-01-21 09:51:46.142376: step: 928/529, loss: 1.1358426809310913 2023-01-21 09:51:47.329221: step: 932/529, loss: 0.04838709905743599 2023-01-21 09:51:48.476012: step: 936/529, loss: 0.17898578941822052 2023-01-21 09:51:49.669275: step: 940/529, loss: 0.0015744924312457442 2023-01-21 09:51:50.883577: step: 944/529, loss: 0.3382924497127533 2023-01-21 09:51:52.090956: step: 948/529, loss: 0.11519765853881836 2023-01-21 09:51:53.282470: step: 952/529, loss: 0.6631711721420288 2023-01-21 09:51:54.453890: step: 956/529, loss: 0.09338932484388351 2023-01-21 09:51:55.636768: step: 960/529, loss: 0.0464324951171875 2023-01-21 09:51:56.821382: step: 964/529, loss: 0.09477519989013672 2023-01-21 09:51:57.978495: step: 968/529, loss: 0.07203254848718643 2023-01-21 09:51:59.172920: step: 972/529, loss: 0.28010159730911255 2023-01-21 09:52:00.347382: step: 976/529, loss: 0.1176910474896431 2023-01-21 09:52:01.613534: step: 980/529, loss: 0.21471108496189117 2023-01-21 09:52:02.759919: step: 984/529, loss: 0.038515470921993256 2023-01-21 09:52:03.935044: step: 988/529, loss: 0.1317310333251953 2023-01-21 09:52:05.080438: step: 992/529, loss: 0.018651485443115234 2023-01-21 09:52:06.247040: step: 996/529, loss: 0.06031942367553711 2023-01-21 09:52:07.474983: step: 1000/529, loss: 0.17538222670555115 2023-01-21 09:52:08.677368: step: 1004/529, loss: 0.3069891333580017 2023-01-21 09:52:09.883905: step: 1008/529, loss: 0.0990542396903038 2023-01-21 09:52:11.101154: step: 1012/529, loss: 0.0547361820936203 2023-01-21 09:52:12.324313: step: 1016/529, loss: 0.11041097342967987 2023-01-21 09:52:13.544266: step: 1020/529, loss: 0.1792929768562317 2023-01-21 09:52:14.742251: step: 1024/529, loss: 0.06711836159229279 2023-01-21 09:52:15.937490: step: 1028/529, loss: 0.05936441570520401 2023-01-21 09:52:17.087303: step: 1032/529, loss: 0.09092765301465988 2023-01-21 09:52:18.275570: step: 1036/529, loss: 0.03158579021692276 2023-01-21 09:52:19.443868: step: 1040/529, loss: 0.10615473240613937 2023-01-21 09:52:20.654232: step: 1044/529, loss: 0.09594269096851349 2023-01-21 09:52:21.850123: step: 1048/529, loss: 0.12677517533302307 2023-01-21 09:52:23.052455: step: 1052/529, loss: 0.1462719887495041 2023-01-21 09:52:24.234808: step: 1056/529, loss: 0.14660054445266724 2023-01-21 09:52:25.412669: step: 1060/529, loss: 0.10974879562854767 2023-01-21 09:52:26.618321: step: 1064/529, loss: 0.10343937575817108 2023-01-21 09:52:27.767848: step: 1068/529, loss: 0.23514434695243835 2023-01-21 09:52:28.955855: step: 1072/529, loss: 0.2881399095058441 2023-01-21 09:52:30.099448: step: 1076/529, loss: 0.1503990739583969 2023-01-21 09:52:31.308552: step: 1080/529, loss: 0.21785756945610046 2023-01-21 09:52:32.496094: step: 1084/529, loss: 0.16295868158340454 2023-01-21 09:52:33.678980: step: 1088/529, loss: 0.06543765217065811 2023-01-21 09:52:34.863694: step: 1092/529, loss: 0.13311271369457245 2023-01-21 09:52:36.066541: step: 1096/529, loss: 0.6653415560722351 2023-01-21 09:52:37.285005: step: 1100/529, loss: 0.0646912157535553 2023-01-21 09:52:38.623137: step: 1104/529, loss: 0.17172202467918396 2023-01-21 09:52:39.855179: step: 1108/529, loss: 0.07701635360717773 2023-01-21 09:52:41.055866: step: 1112/529, loss: 0.1330009400844574 2023-01-21 09:52:42.238242: step: 1116/529, loss: 0.15148693323135376 2023-01-21 09:52:43.441801: step: 1120/529, loss: 0.19542741775512695 2023-01-21 09:52:44.636269: step: 1124/529, loss: 0.059992220252752304 2023-01-21 09:52:45.867293: step: 1128/529, loss: 0.8736928105354309 2023-01-21 09:52:47.076521: step: 1132/529, loss: 0.08895877003669739 2023-01-21 09:52:48.231315: step: 1136/529, loss: 0.036873627454042435 2023-01-21 09:52:49.418773: step: 1140/529, loss: 0.3747602701187134 2023-01-21 09:52:50.630689: step: 1144/529, loss: 0.241233229637146 2023-01-21 09:52:51.835732: step: 1148/529, loss: 0.10447006672620773 2023-01-21 09:52:52.994046: step: 1152/529, loss: 0.16195747256278992 2023-01-21 09:52:54.215725: step: 1156/529, loss: 0.3099810481071472 2023-01-21 09:52:55.437785: step: 1160/529, loss: 0.012358666397631168 2023-01-21 09:52:56.668527: step: 1164/529, loss: 0.061627957969903946 2023-01-21 09:52:57.857957: step: 1168/529, loss: 0.4493255615234375 2023-01-21 09:52:59.033020: step: 1172/529, loss: 0.06605024635791779 2023-01-21 09:53:00.295559: step: 1176/529, loss: 0.0601743683218956 2023-01-21 09:53:01.457848: step: 1180/529, loss: 0.5440307855606079 2023-01-21 09:53:02.746423: step: 1184/529, loss: 0.4135196805000305 2023-01-21 09:53:03.949622: step: 1188/529, loss: 0.10755673050880432 2023-01-21 09:53:05.134704: step: 1192/529, loss: 0.09541235119104385 2023-01-21 09:53:06.356165: step: 1196/529, loss: 0.2487073838710785 2023-01-21 09:53:07.546464: step: 1200/529, loss: 0.09223347157239914 2023-01-21 09:53:08.703832: step: 1204/529, loss: 0.1353287696838379 2023-01-21 09:53:09.900223: step: 1208/529, loss: 0.12909898161888123 2023-01-21 09:53:11.099762: step: 1212/529, loss: 0.18661917746067047 2023-01-21 09:53:12.316596: step: 1216/529, loss: 0.09025654941797256 2023-01-21 09:53:13.551385: step: 1220/529, loss: 0.1750471144914627 2023-01-21 09:53:14.745726: step: 1224/529, loss: 0.04515476152300835 2023-01-21 09:53:15.950237: step: 1228/529, loss: 0.5536761283874512 2023-01-21 09:53:17.113894: step: 1232/529, loss: 0.09854154288768768 2023-01-21 09:53:18.303286: step: 1236/529, loss: 0.10595427453517914 2023-01-21 09:53:19.489916: step: 1240/529, loss: 0.054291535168886185 2023-01-21 09:53:20.658596: step: 1244/529, loss: 0.064898781478405 2023-01-21 09:53:21.827231: step: 1248/529, loss: 0.09864282608032227 2023-01-21 09:53:23.064232: step: 1252/529, loss: 0.34795936942100525 2023-01-21 09:53:24.242991: step: 1256/529, loss: 0.03506441414356232 2023-01-21 09:53:25.454966: step: 1260/529, loss: 0.16099444031715393 2023-01-21 09:53:26.625002: step: 1264/529, loss: 0.036310866475105286 2023-01-21 09:53:27.794164: step: 1268/529, loss: 0.0830836296081543 2023-01-21 09:53:28.997182: step: 1272/529, loss: 0.11686515808105469 2023-01-21 09:53:30.171946: step: 1276/529, loss: 0.1329561173915863 2023-01-21 09:53:31.373002: step: 1280/529, loss: 0.08197034895420074 2023-01-21 09:53:32.637347: step: 1284/529, loss: 0.23507671058177948 2023-01-21 09:53:33.827424: step: 1288/529, loss: 0.11753444373607635 2023-01-21 09:53:35.016635: step: 1292/529, loss: 0.14726237952709198 2023-01-21 09:53:36.233618: step: 1296/529, loss: 0.15031376481056213 2023-01-21 09:53:37.432475: step: 1300/529, loss: 0.5373073816299438 2023-01-21 09:53:38.579843: step: 1304/529, loss: 0.03905687481164932 2023-01-21 09:53:39.814608: step: 1308/529, loss: 0.1880422830581665 2023-01-21 09:53:41.012906: step: 1312/529, loss: 0.17137441039085388 2023-01-21 09:53:42.212086: step: 1316/529, loss: 0.28714749217033386 2023-01-21 09:53:43.426009: step: 1320/529, loss: 0.14100919663906097 2023-01-21 09:53:44.621586: step: 1324/529, loss: 0.017325926572084427 2023-01-21 09:53:45.811127: step: 1328/529, loss: 0.051026154309511185 2023-01-21 09:53:47.019813: step: 1332/529, loss: 0.12988539040088654 2023-01-21 09:53:48.220274: step: 1336/529, loss: 0.11078184098005295 2023-01-21 09:53:49.400502: step: 1340/529, loss: 0.06654338538646698 2023-01-21 09:53:50.582492: step: 1344/529, loss: 0.21339836716651917 2023-01-21 09:53:51.770520: step: 1348/529, loss: 0.03113708645105362 2023-01-21 09:53:52.968782: step: 1352/529, loss: 0.06716771423816681 2023-01-21 09:53:54.126061: step: 1356/529, loss: 0.44783973693847656 2023-01-21 09:53:55.321768: step: 1360/529, loss: 0.2566145956516266 2023-01-21 09:53:56.498410: step: 1364/529, loss: 0.14547625184059143 2023-01-21 09:53:57.737219: step: 1368/529, loss: 0.10531790554523468 2023-01-21 09:53:58.939489: step: 1372/529, loss: 0.12327289581298828 2023-01-21 09:54:00.111658: step: 1376/529, loss: 0.0153217613697052 2023-01-21 09:54:01.352184: step: 1380/529, loss: 0.21671715378761292 2023-01-21 09:54:02.553996: step: 1384/529, loss: 0.11157159507274628 2023-01-21 09:54:03.749912: step: 1388/529, loss: 0.42696744203567505 2023-01-21 09:54:04.953517: step: 1392/529, loss: 0.2715120315551758 2023-01-21 09:54:06.191915: step: 1396/529, loss: 0.26450294256210327 2023-01-21 09:54:07.392247: step: 1400/529, loss: 0.6119434833526611 2023-01-21 09:54:08.580414: step: 1404/529, loss: 0.18399199843406677 2023-01-21 09:54:09.768844: step: 1408/529, loss: 0.2501608729362488 2023-01-21 09:54:10.984025: step: 1412/529, loss: 0.4303356111049652 2023-01-21 09:54:12.186498: step: 1416/529, loss: 0.05102033540606499 2023-01-21 09:54:13.405871: step: 1420/529, loss: 0.481133371591568 2023-01-21 09:54:14.577492: step: 1424/529, loss: 0.14962634444236755 2023-01-21 09:54:15.758919: step: 1428/529, loss: 0.03391450643539429 2023-01-21 09:54:16.932207: step: 1432/529, loss: 0.2718815803527832 2023-01-21 09:54:18.157466: step: 1436/529, loss: 0.39867112040519714 2023-01-21 09:54:19.344551: step: 1440/529, loss: 0.22161683440208435 2023-01-21 09:54:20.557164: step: 1444/529, loss: 0.6567305326461792 2023-01-21 09:54:21.743005: step: 1448/529, loss: 0.04203090816736221 2023-01-21 09:54:22.938270: step: 1452/529, loss: 0.08172011375427246 2023-01-21 09:54:24.136587: step: 1456/529, loss: 0.23961743712425232 2023-01-21 09:54:25.353008: step: 1460/529, loss: 0.06445465236902237 2023-01-21 09:54:26.556436: step: 1464/529, loss: 0.07300949096679688 2023-01-21 09:54:27.698088: step: 1468/529, loss: 0.1495160162448883 2023-01-21 09:54:28.932415: step: 1472/529, loss: 0.09207125008106232 2023-01-21 09:54:30.081318: step: 1476/529, loss: 0.3349166512489319 2023-01-21 09:54:31.310185: step: 1480/529, loss: 0.5212050676345825 2023-01-21 09:54:32.489294: step: 1484/529, loss: 0.18425656855106354 2023-01-21 09:54:33.686157: step: 1488/529, loss: 0.48989734053611755 2023-01-21 09:54:34.835901: step: 1492/529, loss: 0.0749574676156044 2023-01-21 09:54:35.996466: step: 1496/529, loss: 0.3424680233001709 2023-01-21 09:54:37.162413: step: 1500/529, loss: 0.18373823165893555 2023-01-21 09:54:38.363296: step: 1504/529, loss: 0.2712888717651367 2023-01-21 09:54:39.539091: step: 1508/529, loss: 0.048180483281612396 2023-01-21 09:54:40.737107: step: 1512/529, loss: 0.06005420535802841 2023-01-21 09:54:41.889936: step: 1516/529, loss: 0.14139194786548615 2023-01-21 09:54:43.133495: step: 1520/529, loss: 0.0999637097120285 2023-01-21 09:54:44.322329: step: 1524/529, loss: 0.06776027381420135 2023-01-21 09:54:45.519648: step: 1528/529, loss: 0.11130230128765106 2023-01-21 09:54:46.688864: step: 1532/529, loss: 0.10155268013477325 2023-01-21 09:54:47.895418: step: 1536/529, loss: 0.530655562877655 2023-01-21 09:54:49.065018: step: 1540/529, loss: 0.03936328738927841 2023-01-21 09:54:50.253425: step: 1544/529, loss: 0.0909823477268219 2023-01-21 09:54:51.438161: step: 1548/529, loss: 0.05777755007147789 2023-01-21 09:54:52.669559: step: 1552/529, loss: 0.13439245522022247 2023-01-21 09:54:53.897239: step: 1556/529, loss: 0.14082685112953186 2023-01-21 09:54:55.066018: step: 1560/529, loss: 0.3213765025138855 2023-01-21 09:54:56.286395: step: 1564/529, loss: 0.06599275767803192 2023-01-21 09:54:57.462661: step: 1568/529, loss: 0.18704243004322052 2023-01-21 09:54:58.619835: step: 1572/529, loss: 0.06092967838048935 2023-01-21 09:54:59.790212: step: 1576/529, loss: 0.14499612152576447 2023-01-21 09:55:00.969986: step: 1580/529, loss: 0.17739105224609375 2023-01-21 09:55:02.128166: step: 1584/529, loss: 0.48578178882598877 2023-01-21 09:55:03.302066: step: 1588/529, loss: 0.06537933647632599 2023-01-21 09:55:04.506387: step: 1592/529, loss: 0.06096382066607475 2023-01-21 09:55:05.711457: step: 1596/529, loss: 0.0891498550772667 2023-01-21 09:55:06.907811: step: 1600/529, loss: 0.3035891652107239 2023-01-21 09:55:08.110212: step: 1604/529, loss: 0.17341357469558716 2023-01-21 09:55:09.305407: step: 1608/529, loss: 0.15328426659107208 2023-01-21 09:55:10.493445: step: 1612/529, loss: 0.10283203423023224 2023-01-21 09:55:11.675961: step: 1616/529, loss: 0.1294664442539215 2023-01-21 09:55:12.832279: step: 1620/529, loss: 0.08666791766881943 2023-01-21 09:55:14.103961: step: 1624/529, loss: 0.18237236142158508 2023-01-21 09:55:15.305132: step: 1628/529, loss: 0.7318611741065979 2023-01-21 09:55:16.480395: step: 1632/529, loss: 0.0477423220872879 2023-01-21 09:55:17.688881: step: 1636/529, loss: 0.16483289003372192 2023-01-21 09:55:18.890935: step: 1640/529, loss: 1.1591278314590454 2023-01-21 09:55:20.089155: step: 1644/529, loss: 0.10539735108613968 2023-01-21 09:55:21.245350: step: 1648/529, loss: 0.4011775255203247 2023-01-21 09:55:22.444710: step: 1652/529, loss: 0.12273788452148438 2023-01-21 09:55:23.643693: step: 1656/529, loss: 0.19339217245578766 2023-01-21 09:55:24.852178: step: 1660/529, loss: 0.06434116512537003 2023-01-21 09:55:26.079502: step: 1664/529, loss: 0.12760713696479797 2023-01-21 09:55:27.294980: step: 1668/529, loss: 0.3298415243625641 2023-01-21 09:55:28.499235: step: 1672/529, loss: 0.26326093077659607 2023-01-21 09:55:29.747150: step: 1676/529, loss: 0.19430240988731384 2023-01-21 09:55:30.936210: step: 1680/529, loss: 0.07733235508203506 2023-01-21 09:55:32.129875: step: 1684/529, loss: 0.08207711577415466 2023-01-21 09:55:33.353213: step: 1688/529, loss: 0.07611904293298721 2023-01-21 09:55:34.543807: step: 1692/529, loss: 0.7262917757034302 2023-01-21 09:55:35.735576: step: 1696/529, loss: 0.38938167691230774 2023-01-21 09:55:36.912784: step: 1700/529, loss: 0.1283935308456421 2023-01-21 09:55:38.061144: step: 1704/529, loss: 0.10140834003686905 2023-01-21 09:55:39.269123: step: 1708/529, loss: 0.20819798111915588 2023-01-21 09:55:40.490651: step: 1712/529, loss: 0.020225979387760162 2023-01-21 09:55:41.706913: step: 1716/529, loss: 0.1567997932434082 2023-01-21 09:55:42.865364: step: 1720/529, loss: 0.15373000502586365 2023-01-21 09:55:44.091291: step: 1724/529, loss: 0.09842424094676971 2023-01-21 09:55:45.283662: step: 1728/529, loss: 0.03168511390686035 2023-01-21 09:55:46.505245: step: 1732/529, loss: 0.04384488984942436 2023-01-21 09:55:47.700875: step: 1736/529, loss: 0.21343976259231567 2023-01-21 09:55:48.882564: step: 1740/529, loss: 0.06888332217931747 2023-01-21 09:55:50.098212: step: 1744/529, loss: 0.0893133133649826 2023-01-21 09:55:51.259171: step: 1748/529, loss: 0.10008268803358078 2023-01-21 09:55:52.464175: step: 1752/529, loss: 0.167872816324234 2023-01-21 09:55:53.669733: step: 1756/529, loss: 0.4022245407104492 2023-01-21 09:55:54.869316: step: 1760/529, loss: 0.04134855419397354 2023-01-21 09:55:56.069638: step: 1764/529, loss: 0.030942248180508614 2023-01-21 09:55:57.240401: step: 1768/529, loss: 0.09721393883228302 2023-01-21 09:55:58.415257: step: 1772/529, loss: 0.03240518644452095 2023-01-21 09:55:59.556616: step: 1776/529, loss: 0.9947913885116577 2023-01-21 09:56:00.782776: step: 1780/529, loss: 0.19643336534500122 2023-01-21 09:56:01.971691: step: 1784/529, loss: 0.01492605172097683 2023-01-21 09:56:03.196498: step: 1788/529, loss: 0.09687414020299911 2023-01-21 09:56:04.414835: step: 1792/529, loss: 0.28758639097213745 2023-01-21 09:56:05.570150: step: 1796/529, loss: 0.22570115327835083 2023-01-21 09:56:06.749575: step: 1800/529, loss: 0.04280395433306694 2023-01-21 09:56:07.921370: step: 1804/529, loss: 0.09879274666309357 2023-01-21 09:56:09.116223: step: 1808/529, loss: 0.6870439052581787 2023-01-21 09:56:10.273055: step: 1812/529, loss: 0.06153907999396324 2023-01-21 09:56:11.501703: step: 1816/529, loss: 0.08938350528478622 2023-01-21 09:56:12.663320: step: 1820/529, loss: 0.10924364626407623 2023-01-21 09:56:13.850846: step: 1824/529, loss: 0.5274261832237244 2023-01-21 09:56:15.047191: step: 1828/529, loss: 0.4776703715324402 2023-01-21 09:56:16.284130: step: 1832/529, loss: 0.019423581659793854 2023-01-21 09:56:17.475536: step: 1836/529, loss: 0.2219340205192566 2023-01-21 09:56:18.647153: step: 1840/529, loss: 0.6077440977096558 2023-01-21 09:56:19.835235: step: 1844/529, loss: 0.25719720125198364 2023-01-21 09:56:21.045869: step: 1848/529, loss: 0.22958669066429138 2023-01-21 09:56:22.265339: step: 1852/529, loss: 0.25654125213623047 2023-01-21 09:56:23.446317: step: 1856/529, loss: 0.05707988515496254 2023-01-21 09:56:24.659915: step: 1860/529, loss: 0.0982334166765213 2023-01-21 09:56:25.836359: step: 1864/529, loss: 0.11462493240833282 2023-01-21 09:56:27.047434: step: 1868/529, loss: 0.2678220868110657 2023-01-21 09:56:28.252096: step: 1872/529, loss: 0.08859243988990784 2023-01-21 09:56:29.472734: step: 1876/529, loss: 0.5402352213859558 2023-01-21 09:56:30.664368: step: 1880/529, loss: 0.06526260077953339 2023-01-21 09:56:31.886939: step: 1884/529, loss: 0.6797018051147461 2023-01-21 09:56:33.085257: step: 1888/529, loss: 0.2423418164253235 2023-01-21 09:56:34.301700: step: 1892/529, loss: 0.1985209584236145 2023-01-21 09:56:35.465596: step: 1896/529, loss: 0.08824422210454941 2023-01-21 09:56:36.659442: step: 1900/529, loss: 0.26627302169799805 2023-01-21 09:56:37.808476: step: 1904/529, loss: 0.1284509301185608 2023-01-21 09:56:39.012817: step: 1908/529, loss: 0.6680278182029724 2023-01-21 09:56:40.217233: step: 1912/529, loss: 0.4039851129055023 2023-01-21 09:56:41.399428: step: 1916/529, loss: 0.08700904995203018 2023-01-21 09:56:42.602048: step: 1920/529, loss: 0.13818468153476715 2023-01-21 09:56:43.800679: step: 1924/529, loss: 0.4956548810005188 2023-01-21 09:56:44.985486: step: 1928/529, loss: 0.6594530940055847 2023-01-21 09:56:46.152649: step: 1932/529, loss: 0.11204744875431061 2023-01-21 09:56:47.344244: step: 1936/529, loss: 0.09674148261547089 2023-01-21 09:56:48.538329: step: 1940/529, loss: 0.16983680427074432 2023-01-21 09:56:49.721354: step: 1944/529, loss: 0.0009875536197796464 2023-01-21 09:56:50.921296: step: 1948/529, loss: 0.1586267054080963 2023-01-21 09:56:52.120567: step: 1952/529, loss: 0.42432236671447754 2023-01-21 09:56:53.285266: step: 1956/529, loss: 0.0937003642320633 2023-01-21 09:56:54.498999: step: 1960/529, loss: 0.14340762794017792 2023-01-21 09:56:55.684950: step: 1964/529, loss: 0.47758838534355164 2023-01-21 09:56:56.927953: step: 1968/529, loss: 0.1976001262664795 2023-01-21 09:56:58.110875: step: 1972/529, loss: 0.046059273183345795 2023-01-21 09:56:59.294371: step: 1976/529, loss: 0.048142239451408386 2023-01-21 09:57:00.543433: step: 1980/529, loss: 0.20206299424171448 2023-01-21 09:57:01.719173: step: 1984/529, loss: 0.07489385455846786 2023-01-21 09:57:02.964854: step: 1988/529, loss: 0.1461162567138672 2023-01-21 09:57:04.165620: step: 1992/529, loss: 1.400456190109253 2023-01-21 09:57:05.351387: step: 1996/529, loss: 0.005507993511855602 2023-01-21 09:57:06.540264: step: 2000/529, loss: 0.21512573957443237 2023-01-21 09:57:07.744096: step: 2004/529, loss: 0.09254741668701172 2023-01-21 09:57:08.954879: step: 2008/529, loss: 0.09168457984924316 2023-01-21 09:57:10.205352: step: 2012/529, loss: 0.0891714096069336 2023-01-21 09:57:11.434535: step: 2016/529, loss: 0.05012502893805504 2023-01-21 09:57:12.701268: step: 2020/529, loss: 0.05382196605205536 2023-01-21 09:57:13.904663: step: 2024/529, loss: 0.29301711916923523 2023-01-21 09:57:15.088292: step: 2028/529, loss: 0.05165357515215874 2023-01-21 09:57:16.271538: step: 2032/529, loss: 0.0846046507358551 2023-01-21 09:57:17.447019: step: 2036/529, loss: 0.18431416153907776 2023-01-21 09:57:18.655079: step: 2040/529, loss: 0.09120287746191025 2023-01-21 09:57:19.875809: step: 2044/529, loss: 0.1373794674873352 2023-01-21 09:57:21.085967: step: 2048/529, loss: 0.17447508871555328 2023-01-21 09:57:22.297516: step: 2052/529, loss: 0.13549461960792542 2023-01-21 09:57:23.480430: step: 2056/529, loss: 0.03647775575518608 2023-01-21 09:57:24.712281: step: 2060/529, loss: 0.08494958281517029 2023-01-21 09:57:25.927040: step: 2064/529, loss: 0.7453889846801758 2023-01-21 09:57:27.124894: step: 2068/529, loss: 0.19579295814037323 2023-01-21 09:57:28.288896: step: 2072/529, loss: 0.1647653877735138 2023-01-21 09:57:29.462454: step: 2076/529, loss: 0.30069634318351746 2023-01-21 09:57:30.655536: step: 2080/529, loss: 0.12725667655467987 2023-01-21 09:57:31.879905: step: 2084/529, loss: 0.11948347091674805 2023-01-21 09:57:33.069780: step: 2088/529, loss: 0.09413222968578339 2023-01-21 09:57:34.271696: step: 2092/529, loss: 0.01352386549115181 2023-01-21 09:57:35.491671: step: 2096/529, loss: 0.08415775746107101 2023-01-21 09:57:36.672780: step: 2100/529, loss: 0.08150902390480042 2023-01-21 09:57:37.893516: step: 2104/529, loss: 0.16767293214797974 2023-01-21 09:57:39.083269: step: 2108/529, loss: 0.907122015953064 2023-01-21 09:57:40.280957: step: 2112/529, loss: 0.07126370072364807 2023-01-21 09:57:41.491069: step: 2116/529, loss: 0.1488230675458908 ================================================== Loss: 0.196 -------------------- Dev: {'event': {'p': 0.564922480620155, 'r': 0.7762982689747004, 'f1': 0.6539540100953449}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Test: {'event': {'p': 0.6181046676096181, 'r': 0.7855002995805872, 'f1': 0.691820580474934}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Chinese: {'event': {'p': 0.573170731707317, 'r': 0.8703703703703703, 'f1': 0.6911764705882353}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Korean: {'event': {'p': 0.5223880597014925, 'r': 0.5555555555555556, 'f1': 0.5384615384615385}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Russian: {'event': {'p': 0.3958333333333333, 'r': 0.5277777777777778, 'f1': 0.45238095238095233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5972944849115505, 'r': 0.7643142476697736, 'f1': 0.6705607476635514}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6162162162162163, 'r': 0.7513481126423007, 'f1': 0.6771058315334773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 7 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:58:25.129798: step: 4/529, loss: 0.14414720237255096 2023-01-21 09:58:26.321722: step: 8/529, loss: 0.08610916882753372 2023-01-21 09:58:27.499210: step: 12/529, loss: 0.09214423596858978 2023-01-21 09:58:28.689502: step: 16/529, loss: 0.10569429397583008 2023-01-21 09:58:29.884213: step: 20/529, loss: 0.012187529355287552 2023-01-21 09:58:31.079585: step: 24/529, loss: 0.04768495634198189 2023-01-21 09:58:32.315393: step: 28/529, loss: 0.17110490798950195 2023-01-21 09:58:33.490800: step: 32/529, loss: 0.05441169813275337 2023-01-21 09:58:34.662955: step: 36/529, loss: 0.07184362411499023 2023-01-21 09:58:35.861891: step: 40/529, loss: 0.08034954220056534 2023-01-21 09:58:37.040663: step: 44/529, loss: 0.5869011282920837 2023-01-21 09:58:38.273854: step: 48/529, loss: 0.10149791091680527 2023-01-21 09:58:39.456345: step: 52/529, loss: 0.12109003216028214 2023-01-21 09:58:40.653426: step: 56/529, loss: 0.027198124676942825 2023-01-21 09:58:41.832671: step: 60/529, loss: 0.04667243734002113 2023-01-21 09:58:42.996050: step: 64/529, loss: 0.020627975463867188 2023-01-21 09:58:44.195629: step: 68/529, loss: 0.07522383332252502 2023-01-21 09:58:45.356977: step: 72/529, loss: 0.1055065169930458 2023-01-21 09:58:46.539026: step: 76/529, loss: 0.198052778840065 2023-01-21 09:58:47.704580: step: 80/529, loss: 0.034584809094667435 2023-01-21 09:58:48.897785: step: 84/529, loss: 0.16676485538482666 2023-01-21 09:58:50.104983: step: 88/529, loss: 0.14330492913722992 2023-01-21 09:58:51.354263: step: 92/529, loss: 0.1588471382856369 2023-01-21 09:58:52.533312: step: 96/529, loss: 0.07598753273487091 2023-01-21 09:58:53.720751: step: 100/529, loss: 0.401014119386673 2023-01-21 09:58:54.870719: step: 104/529, loss: 0.22215361893177032 2023-01-21 09:58:56.050782: step: 108/529, loss: 0.06700734794139862 2023-01-21 09:58:57.260115: step: 112/529, loss: 0.2054046094417572 2023-01-21 09:58:58.440358: step: 116/529, loss: 0.11804036796092987 2023-01-21 09:58:59.633486: step: 120/529, loss: 0.09916677325963974 2023-01-21 09:59:00.853174: step: 124/529, loss: 0.04561276361346245 2023-01-21 09:59:02.040805: step: 128/529, loss: 0.08148641884326935 2023-01-21 09:59:03.195113: step: 132/529, loss: 0.03351778909564018 2023-01-21 09:59:04.369279: step: 136/529, loss: 0.04172353446483612 2023-01-21 09:59:05.537898: step: 140/529, loss: 0.01796731911599636 2023-01-21 09:59:06.726455: step: 144/529, loss: 0.09609613567590714 2023-01-21 09:59:07.920852: step: 148/529, loss: 0.06445484608411789 2023-01-21 09:59:09.081710: step: 152/529, loss: 0.0369747169315815 2023-01-21 09:59:10.280810: step: 156/529, loss: 0.016296863555908203 2023-01-21 09:59:11.524619: step: 160/529, loss: 0.0977272093296051 2023-01-21 09:59:12.724110: step: 164/529, loss: 0.04573802649974823 2023-01-21 09:59:13.916315: step: 168/529, loss: 0.22275123000144958 2023-01-21 09:59:15.095344: step: 172/529, loss: 0.10831575095653534 2023-01-21 09:59:16.269326: step: 176/529, loss: 0.17365506291389465 2023-01-21 09:59:17.491911: step: 180/529, loss: 0.4810195863246918 2023-01-21 09:59:18.725465: step: 184/529, loss: 0.0827573835849762 2023-01-21 09:59:19.920079: step: 188/529, loss: 0.1930348426103592 2023-01-21 09:59:21.103085: step: 192/529, loss: 0.025715328752994537 2023-01-21 09:59:22.299621: step: 196/529, loss: 0.09359245747327805 2023-01-21 09:59:23.485911: step: 200/529, loss: 0.020565176382660866 2023-01-21 09:59:24.661497: step: 204/529, loss: 0.025641536340117455 2023-01-21 09:59:25.825844: step: 208/529, loss: 0.18171711266040802 2023-01-21 09:59:27.043473: step: 212/529, loss: 0.0633297935128212 2023-01-21 09:59:28.265480: step: 216/529, loss: 0.354599267244339 2023-01-21 09:59:29.453823: step: 220/529, loss: 0.07756538689136505 2023-01-21 09:59:30.640126: step: 224/529, loss: 0.33542659878730774 2023-01-21 09:59:31.846909: step: 228/529, loss: 0.3240453898906708 2023-01-21 09:59:33.022208: step: 232/529, loss: 0.04538364335894585 2023-01-21 09:59:34.249335: step: 236/529, loss: 0.04080557823181152 2023-01-21 09:59:35.429511: step: 240/529, loss: 0.7638151049613953 2023-01-21 09:59:36.657797: step: 244/529, loss: 0.3556199073791504 2023-01-21 09:59:37.865944: step: 248/529, loss: 0.26836538314819336 2023-01-21 09:59:39.063715: step: 252/529, loss: 0.1899253875017166 2023-01-21 09:59:40.274094: step: 256/529, loss: 0.050566673278808594 2023-01-21 09:59:41.478802: step: 260/529, loss: 0.28201359510421753 2023-01-21 09:59:42.675241: step: 264/529, loss: 0.031092606484889984 2023-01-21 09:59:43.848791: step: 268/529, loss: 0.0398009791970253 2023-01-21 09:59:45.025282: step: 272/529, loss: 0.09260912239551544 2023-01-21 09:59:46.199356: step: 276/529, loss: 0.04423123225569725 2023-01-21 09:59:47.378839: step: 280/529, loss: 0.05221528932452202 2023-01-21 09:59:48.532187: step: 284/529, loss: 0.004040050320327282 2023-01-21 09:59:49.713103: step: 288/529, loss: 0.8165051341056824 2023-01-21 09:59:50.917039: step: 292/529, loss: 0.3493371903896332 2023-01-21 09:59:52.127002: step: 296/529, loss: 0.11197242885828018 2023-01-21 09:59:53.353756: step: 300/529, loss: 0.295889675617218 2023-01-21 09:59:54.563917: step: 304/529, loss: 0.29164400696754456 2023-01-21 09:59:55.729502: step: 308/529, loss: 0.06157274171710014 2023-01-21 09:59:56.922547: step: 312/529, loss: 0.0882442444562912 2023-01-21 09:59:58.133630: step: 316/529, loss: 0.1500554084777832 2023-01-21 09:59:59.311605: step: 320/529, loss: 0.2247953861951828 2023-01-21 10:00:00.516179: step: 324/529, loss: 0.10559125244617462 2023-01-21 10:00:01.725079: step: 328/529, loss: 0.19618292152881622 2023-01-21 10:00:02.951868: step: 332/529, loss: 0.12368012219667435 2023-01-21 10:00:04.107054: step: 336/529, loss: 0.09588590264320374 2023-01-21 10:00:05.310441: step: 340/529, loss: 0.11881313472986221 2023-01-21 10:00:06.535614: step: 344/529, loss: 0.42184609174728394 2023-01-21 10:00:07.722014: step: 348/529, loss: 0.025215625762939453 2023-01-21 10:00:08.953386: step: 352/529, loss: 0.4897242486476898 2023-01-21 10:00:10.167434: step: 356/529, loss: 0.020172975957393646 2023-01-21 10:00:11.368454: step: 360/529, loss: 0.29419660568237305 2023-01-21 10:00:12.580818: step: 364/529, loss: 0.08933086693286896 2023-01-21 10:00:13.752078: step: 368/529, loss: 0.02380390092730522 2023-01-21 10:00:14.927960: step: 372/529, loss: 0.24956035614013672 2023-01-21 10:00:16.121192: step: 376/529, loss: 0.07972851395606995 2023-01-21 10:00:17.340896: step: 380/529, loss: 0.11400480568408966 2023-01-21 10:00:18.577415: step: 384/529, loss: 0.07235798984766006 2023-01-21 10:00:19.731629: step: 388/529, loss: 0.032538414001464844 2023-01-21 10:00:20.956033: step: 392/529, loss: 0.0408930778503418 2023-01-21 10:00:22.189575: step: 396/529, loss: 0.3430141508579254 2023-01-21 10:00:23.372600: step: 400/529, loss: 0.030090905725955963 2023-01-21 10:00:24.593647: step: 404/529, loss: 0.2875344753265381 2023-01-21 10:00:25.795041: step: 408/529, loss: 0.011258221231400967 2023-01-21 10:00:26.943161: step: 412/529, loss: 0.031578682363033295 2023-01-21 10:00:28.119286: step: 416/529, loss: 0.31611311435699463 2023-01-21 10:00:29.327202: step: 420/529, loss: 0.05510731041431427 2023-01-21 10:00:30.502751: step: 424/529, loss: 0.07624687999486923 2023-01-21 10:00:31.713355: step: 428/529, loss: 0.06054506450891495 2023-01-21 10:00:32.932932: step: 432/529, loss: 0.03580637276172638 2023-01-21 10:00:34.089645: step: 436/529, loss: 0.20370140671730042 2023-01-21 10:00:35.306497: step: 440/529, loss: 0.08785905689001083 2023-01-21 10:00:36.496548: step: 444/529, loss: 0.056346990168094635 2023-01-21 10:00:37.673266: step: 448/529, loss: 0.06762728840112686 2023-01-21 10:00:38.833030: step: 452/529, loss: 0.00726971635594964 2023-01-21 10:00:39.997649: step: 456/529, loss: 0.06746310740709305 2023-01-21 10:00:41.227984: step: 460/529, loss: 0.13362936675548553 2023-01-21 10:00:42.433860: step: 464/529, loss: 0.025434209033846855 2023-01-21 10:00:43.605163: step: 468/529, loss: 0.06456603854894638 2023-01-21 10:00:44.851100: step: 472/529, loss: 0.08219128102064133 2023-01-21 10:00:46.018868: step: 476/529, loss: 0.3151577413082123 2023-01-21 10:00:47.233564: step: 480/529, loss: 0.06379556655883789 2023-01-21 10:00:48.418465: step: 484/529, loss: 0.027253054082393646 2023-01-21 10:00:49.644385: step: 488/529, loss: 0.04433002322912216 2023-01-21 10:00:50.821893: step: 492/529, loss: 0.1069088950753212 2023-01-21 10:00:52.024101: step: 496/529, loss: 0.13772830367088318 2023-01-21 10:00:53.182788: step: 500/529, loss: 0.058483317494392395 2023-01-21 10:00:54.430154: step: 504/529, loss: 1.2549943923950195 2023-01-21 10:00:55.580945: step: 508/529, loss: 0.07459201663732529 2023-01-21 10:00:56.783015: step: 512/529, loss: 0.2625943124294281 2023-01-21 10:00:57.974765: step: 516/529, loss: 1.8364765644073486 2023-01-21 10:00:59.151159: step: 520/529, loss: 0.39650508761405945 2023-01-21 10:01:00.365547: step: 524/529, loss: 0.5529074668884277 2023-01-21 10:01:01.507901: step: 528/529, loss: 0.04726262390613556 2023-01-21 10:01:02.710558: step: 532/529, loss: 0.07336633652448654 2023-01-21 10:01:03.911443: step: 536/529, loss: 0.37702199816703796 2023-01-21 10:01:05.117763: step: 540/529, loss: 0.12614865601062775 2023-01-21 10:01:06.305796: step: 544/529, loss: 0.18821199238300323 2023-01-21 10:01:07.552164: step: 548/529, loss: 0.02433795854449272 2023-01-21 10:01:08.745534: step: 552/529, loss: 0.05870981514453888 2023-01-21 10:01:09.938104: step: 556/529, loss: 0.3052510619163513 2023-01-21 10:01:11.131986: step: 560/529, loss: 0.10687504708766937 2023-01-21 10:01:12.363221: step: 564/529, loss: 0.1900564730167389 2023-01-21 10:01:13.547950: step: 568/529, loss: 0.062425900250673294 2023-01-21 10:01:14.734478: step: 572/529, loss: 0.04736023396253586 2023-01-21 10:01:15.928367: step: 576/529, loss: 0.0904124304652214 2023-01-21 10:01:17.111165: step: 580/529, loss: 0.18770459294319153 2023-01-21 10:01:18.290486: step: 584/529, loss: 0.056600190699100494 2023-01-21 10:01:19.495017: step: 588/529, loss: 0.09418735653162003 2023-01-21 10:01:20.694856: step: 592/529, loss: 0.5272855162620544 2023-01-21 10:01:21.865625: step: 596/529, loss: 0.10469424724578857 2023-01-21 10:01:23.070369: step: 600/529, loss: 0.01532745361328125 2023-01-21 10:01:24.243458: step: 604/529, loss: 0.05669808387756348 2023-01-21 10:01:25.438720: step: 608/529, loss: 0.055425144731998444 2023-01-21 10:01:26.620157: step: 612/529, loss: 0.008281421847641468 2023-01-21 10:01:27.844338: step: 616/529, loss: 0.05983608216047287 2023-01-21 10:01:29.057284: step: 620/529, loss: 0.11513213813304901 2023-01-21 10:01:30.230551: step: 624/529, loss: 0.0886169895529747 2023-01-21 10:01:31.384539: step: 628/529, loss: 0.0753655880689621 2023-01-21 10:01:32.608076: step: 632/529, loss: 0.21485203504562378 2023-01-21 10:01:33.829284: step: 636/529, loss: 0.33699098229408264 2023-01-21 10:01:35.057859: step: 640/529, loss: 0.15715599060058594 2023-01-21 10:01:36.294667: step: 644/529, loss: 0.14431075751781464 2023-01-21 10:01:37.498857: step: 648/529, loss: 0.15808573365211487 2023-01-21 10:01:38.703633: step: 652/529, loss: 0.2170001119375229 2023-01-21 10:01:39.872474: step: 656/529, loss: 0.07208108901977539 2023-01-21 10:01:41.056067: step: 660/529, loss: 0.08158054202795029 2023-01-21 10:01:42.306876: step: 664/529, loss: 0.18894919753074646 2023-01-21 10:01:43.497306: step: 668/529, loss: 0.5570565462112427 2023-01-21 10:01:44.725943: step: 672/529, loss: 0.1003580093383789 2023-01-21 10:01:45.908152: step: 676/529, loss: 0.04667682945728302 2023-01-21 10:01:47.143940: step: 680/529, loss: 0.05080671235918999 2023-01-21 10:01:48.320961: step: 684/529, loss: 0.1406419277191162 2023-01-21 10:01:49.505996: step: 688/529, loss: 0.05003051832318306 2023-01-21 10:01:50.713167: step: 692/529, loss: 0.1397019922733307 2023-01-21 10:01:51.914318: step: 696/529, loss: 0.22290164232254028 2023-01-21 10:01:53.118927: step: 700/529, loss: 0.5983518958091736 2023-01-21 10:01:54.317063: step: 704/529, loss: 0.038890838623046875 2023-01-21 10:01:55.498853: step: 708/529, loss: 0.060469530522823334 2023-01-21 10:01:56.654691: step: 712/529, loss: 0.07994937896728516 2023-01-21 10:01:57.849460: step: 716/529, loss: 0.09934130311012268 2023-01-21 10:01:59.051720: step: 720/529, loss: 0.11925573647022247 2023-01-21 10:02:00.277676: step: 724/529, loss: 0.10380133986473083 2023-01-21 10:02:01.444091: step: 728/529, loss: 0.1414201706647873 2023-01-21 10:02:02.693261: step: 732/529, loss: 0.10719519108533859 2023-01-21 10:02:03.841191: step: 736/529, loss: 0.08757490664720535 2023-01-21 10:02:05.051157: step: 740/529, loss: 0.23866048455238342 2023-01-21 10:02:06.281260: step: 744/529, loss: 0.7229957580566406 2023-01-21 10:02:07.514601: step: 748/529, loss: 0.025325536727905273 2023-01-21 10:02:08.694984: step: 752/529, loss: 0.1644696742296219 2023-01-21 10:02:09.922415: step: 756/529, loss: 0.4008598327636719 2023-01-21 10:02:11.129864: step: 760/529, loss: 0.019062042236328125 2023-01-21 10:02:12.316888: step: 764/529, loss: 0.7790459990501404 2023-01-21 10:02:13.516829: step: 768/529, loss: 0.06344500184059143 2023-01-21 10:02:14.771010: step: 772/529, loss: 0.09370222687721252 2023-01-21 10:02:15.975284: step: 776/529, loss: 0.0684460699558258 2023-01-21 10:02:17.158560: step: 780/529, loss: 0.09711484611034393 2023-01-21 10:02:18.351553: step: 784/529, loss: 0.01529760379344225 2023-01-21 10:02:19.534558: step: 788/529, loss: 0.36080384254455566 2023-01-21 10:02:20.755136: step: 792/529, loss: 0.08460836112499237 2023-01-21 10:02:21.925850: step: 796/529, loss: 0.17579208314418793 2023-01-21 10:02:23.118999: step: 800/529, loss: 0.18806973099708557 2023-01-21 10:02:24.303920: step: 804/529, loss: 0.03975820541381836 2023-01-21 10:02:25.505111: step: 808/529, loss: 0.006653547286987305 2023-01-21 10:02:26.706309: step: 812/529, loss: 0.07395439594984055 2023-01-21 10:02:27.869376: step: 816/529, loss: 0.046633053570985794 2023-01-21 10:02:29.047673: step: 820/529, loss: 0.29164236783981323 2023-01-21 10:02:30.298942: step: 824/529, loss: 0.05899238586425781 2023-01-21 10:02:31.486743: step: 828/529, loss: 0.36911869049072266 2023-01-21 10:02:32.708511: step: 832/529, loss: 0.06191730499267578 2023-01-21 10:02:33.962148: step: 836/529, loss: 0.11162882298231125 2023-01-21 10:02:35.181050: step: 840/529, loss: 0.6514137983322144 2023-01-21 10:02:36.364459: step: 844/529, loss: 0.12691307067871094 2023-01-21 10:02:37.553031: step: 848/529, loss: 0.6717785596847534 2023-01-21 10:02:38.694756: step: 852/529, loss: 0.15295261144638062 2023-01-21 10:02:39.933107: step: 856/529, loss: 0.05758490785956383 2023-01-21 10:02:41.137778: step: 860/529, loss: 0.02046995237469673 2023-01-21 10:02:42.337465: step: 864/529, loss: 0.05305027961730957 2023-01-21 10:02:43.494218: step: 868/529, loss: 0.05044245719909668 2023-01-21 10:02:44.707553: step: 872/529, loss: 0.2777186632156372 2023-01-21 10:02:45.895013: step: 876/529, loss: 0.2014492005109787 2023-01-21 10:02:47.090224: step: 880/529, loss: 0.41537055373191833 2023-01-21 10:02:48.296356: step: 884/529, loss: 0.14407005906105042 2023-01-21 10:02:49.506910: step: 888/529, loss: 0.03330142796039581 2023-01-21 10:02:50.712488: step: 892/529, loss: 0.09800411015748978 2023-01-21 10:02:51.901929: step: 896/529, loss: 0.09380187839269638 2023-01-21 10:02:53.083252: step: 900/529, loss: 0.2185775637626648 2023-01-21 10:02:54.267035: step: 904/529, loss: 0.05491933599114418 2023-01-21 10:02:55.465252: step: 908/529, loss: 0.06201772764325142 2023-01-21 10:02:56.619883: step: 912/529, loss: 0.19165100157260895 2023-01-21 10:02:57.835013: step: 916/529, loss: 0.042129408568143845 2023-01-21 10:02:58.996355: step: 920/529, loss: 0.04134722054004669 2023-01-21 10:03:00.186760: step: 924/529, loss: 0.5099632143974304 2023-01-21 10:03:01.411755: step: 928/529, loss: 0.12240594625473022 2023-01-21 10:03:02.578661: step: 932/529, loss: 0.08612065017223358 2023-01-21 10:03:03.734766: step: 936/529, loss: 0.06671981513500214 2023-01-21 10:03:04.924714: step: 940/529, loss: 0.20419836044311523 2023-01-21 10:03:06.119700: step: 944/529, loss: 0.08661937713623047 2023-01-21 10:03:07.301032: step: 948/529, loss: 0.15728655457496643 2023-01-21 10:03:08.530851: step: 952/529, loss: 0.09781036525964737 2023-01-21 10:03:09.736686: step: 956/529, loss: 0.07027482986450195 2023-01-21 10:03:10.941387: step: 960/529, loss: 0.06562285125255585 2023-01-21 10:03:12.156012: step: 964/529, loss: 0.15864458680152893 2023-01-21 10:03:13.353424: step: 968/529, loss: 0.17225085198879242 2023-01-21 10:03:14.582716: step: 972/529, loss: 0.010514688678085804 2023-01-21 10:03:15.831968: step: 976/529, loss: 0.21594390273094177 2023-01-21 10:03:16.996883: step: 980/529, loss: 0.08297464996576309 2023-01-21 10:03:18.247125: step: 984/529, loss: 0.07915416359901428 2023-01-21 10:03:19.437484: step: 988/529, loss: 0.10100136697292328 2023-01-21 10:03:20.662190: step: 992/529, loss: 0.16859151422977448 2023-01-21 10:03:21.889297: step: 996/529, loss: 0.38472771644592285 2023-01-21 10:03:23.126528: step: 1000/529, loss: 0.3493598997592926 2023-01-21 10:03:24.397162: step: 1004/529, loss: 0.2528516948223114 2023-01-21 10:03:25.694754: step: 1008/529, loss: 0.08789043128490448 2023-01-21 10:03:26.899006: step: 1012/529, loss: 0.131754070520401 2023-01-21 10:03:28.138113: step: 1016/529, loss: 0.372668594121933 2023-01-21 10:03:29.315299: step: 1020/529, loss: 0.09886160492897034 2023-01-21 10:03:30.619331: step: 1024/529, loss: 0.2696099281311035 2023-01-21 10:03:31.807067: step: 1028/529, loss: 0.0637538880109787 2023-01-21 10:03:33.009506: step: 1032/529, loss: 0.05749655142426491 2023-01-21 10:03:34.168790: step: 1036/529, loss: 0.09321737289428711 2023-01-21 10:03:35.359775: step: 1040/529, loss: 0.12446780502796173 2023-01-21 10:03:36.548582: step: 1044/529, loss: 0.026815414428710938 2023-01-21 10:03:37.731219: step: 1048/529, loss: 0.01471786480396986 2023-01-21 10:03:38.902548: step: 1052/529, loss: 0.051738835871219635 2023-01-21 10:03:40.124848: step: 1056/529, loss: 0.3338344395160675 2023-01-21 10:03:41.333317: step: 1060/529, loss: 0.1210559606552124 2023-01-21 10:03:42.556673: step: 1064/529, loss: 0.04723243787884712 2023-01-21 10:03:43.717294: step: 1068/529, loss: 0.10160288959741592 2023-01-21 10:03:44.893937: step: 1072/529, loss: 0.1779688000679016 2023-01-21 10:03:46.082676: step: 1076/529, loss: 0.0451601967215538 2023-01-21 10:03:47.297689: step: 1080/529, loss: 0.18579836189746857 2023-01-21 10:03:48.467953: step: 1084/529, loss: 0.1562459021806717 2023-01-21 10:03:49.636630: step: 1088/529, loss: 0.10708599537611008 2023-01-21 10:03:50.814940: step: 1092/529, loss: 0.06045665964484215 2023-01-21 10:03:52.016688: step: 1096/529, loss: 0.04016532748937607 2023-01-21 10:03:53.195381: step: 1100/529, loss: 0.02961740642786026 2023-01-21 10:03:54.398144: step: 1104/529, loss: 0.13050255179405212 2023-01-21 10:03:55.616492: step: 1108/529, loss: 0.11328859627246857 2023-01-21 10:03:56.818296: step: 1112/529, loss: 0.2272767424583435 2023-01-21 10:03:58.008194: step: 1116/529, loss: 0.2151515930891037 2023-01-21 10:03:59.251623: step: 1120/529, loss: 0.20970459282398224 2023-01-21 10:04:00.458017: step: 1124/529, loss: 0.27438580989837646 2023-01-21 10:04:01.662685: step: 1128/529, loss: 0.0640861839056015 2023-01-21 10:04:02.870906: step: 1132/529, loss: 0.025502096861600876 2023-01-21 10:04:04.085979: step: 1136/529, loss: 0.07590790092945099 2023-01-21 10:04:05.300191: step: 1140/529, loss: 0.1393032968044281 2023-01-21 10:04:06.541098: step: 1144/529, loss: 0.3347460627555847 2023-01-21 10:04:07.690715: step: 1148/529, loss: 0.10406798869371414 2023-01-21 10:04:08.857205: step: 1152/529, loss: 1.4244558811187744 2023-01-21 10:04:10.038877: step: 1156/529, loss: 0.10619769245386124 2023-01-21 10:04:11.216379: step: 1160/529, loss: 0.09533052146434784 2023-01-21 10:04:12.393508: step: 1164/529, loss: 0.12391719967126846 2023-01-21 10:04:13.603599: step: 1168/529, loss: 0.10453243553638458 2023-01-21 10:04:14.804292: step: 1172/529, loss: 0.4716840982437134 2023-01-21 10:04:15.978117: step: 1176/529, loss: 0.08279738575220108 2023-01-21 10:04:17.155584: step: 1180/529, loss: 0.01758880726993084 2023-01-21 10:04:18.346215: step: 1184/529, loss: 0.03442859649658203 2023-01-21 10:04:19.514947: step: 1188/529, loss: 0.10302762687206268 2023-01-21 10:04:20.715379: step: 1192/529, loss: 0.055144499987363815 2023-01-21 10:04:21.914967: step: 1196/529, loss: 0.17081815004348755 2023-01-21 10:04:23.139397: step: 1200/529, loss: 0.0809200331568718 2023-01-21 10:04:24.315518: step: 1204/529, loss: 0.06702928245067596 2023-01-21 10:04:25.490637: step: 1208/529, loss: 0.04754498228430748 2023-01-21 10:04:26.721656: step: 1212/529, loss: 0.2841157913208008 2023-01-21 10:04:27.908631: step: 1216/529, loss: 0.5948737859725952 2023-01-21 10:04:29.084445: step: 1220/529, loss: 0.12462782859802246 2023-01-21 10:04:30.256464: step: 1224/529, loss: 0.09428625553846359 2023-01-21 10:04:31.464546: step: 1228/529, loss: 0.12613964080810547 2023-01-21 10:04:32.684364: step: 1232/529, loss: 0.04598255455493927 2023-01-21 10:04:33.880074: step: 1236/529, loss: 0.0697137862443924 2023-01-21 10:04:35.092630: step: 1240/529, loss: 0.10657081753015518 2023-01-21 10:04:36.311725: step: 1244/529, loss: 0.37994813919067383 2023-01-21 10:04:37.516202: step: 1248/529, loss: 0.45272141695022583 2023-01-21 10:04:38.680809: step: 1252/529, loss: 0.07155018299818039 2023-01-21 10:04:39.883530: step: 1256/529, loss: 0.1282035857439041 2023-01-21 10:04:41.048612: step: 1260/529, loss: 0.15995995700359344 2023-01-21 10:04:42.224025: step: 1264/529, loss: 0.34280338883399963 2023-01-21 10:04:43.414466: step: 1268/529, loss: 0.021961044520139694 2023-01-21 10:04:44.599998: step: 1272/529, loss: 0.041260723024606705 2023-01-21 10:04:45.767355: step: 1276/529, loss: 0.10261498391628265 2023-01-21 10:04:46.957630: step: 1280/529, loss: 0.10749664157629013 2023-01-21 10:04:48.195448: step: 1284/529, loss: 0.3650610148906708 2023-01-21 10:04:49.455371: step: 1288/529, loss: 0.12068872153759003 2023-01-21 10:04:50.632478: step: 1292/529, loss: 0.1430201530456543 2023-01-21 10:04:51.850737: step: 1296/529, loss: 1.3118350505828857 2023-01-21 10:04:53.039937: step: 1300/529, loss: 0.012200355529785156 2023-01-21 10:04:54.239069: step: 1304/529, loss: 0.057032011449337006 2023-01-21 10:04:55.444641: step: 1308/529, loss: 0.1295982301235199 2023-01-21 10:04:56.625432: step: 1312/529, loss: 0.07049560546875 2023-01-21 10:04:57.803842: step: 1316/529, loss: 0.04867897182703018 2023-01-21 10:04:58.983878: step: 1320/529, loss: 0.1340874582529068 2023-01-21 10:05:00.136822: step: 1324/529, loss: 0.021107196807861328 2023-01-21 10:05:01.308860: step: 1328/529, loss: 0.008972419425845146 2023-01-21 10:05:02.520547: step: 1332/529, loss: 0.11829090118408203 2023-01-21 10:05:03.734028: step: 1336/529, loss: 0.06536245346069336 2023-01-21 10:05:04.995714: step: 1340/529, loss: 0.04923496022820473 2023-01-21 10:05:06.216771: step: 1344/529, loss: 0.34935712814331055 2023-01-21 10:05:07.431556: step: 1348/529, loss: 0.027606630697846413 2023-01-21 10:05:08.614922: step: 1352/529, loss: 1.0833959579467773 2023-01-21 10:05:09.794334: step: 1356/529, loss: 0.017308330163359642 2023-01-21 10:05:10.951459: step: 1360/529, loss: 0.3455123007297516 2023-01-21 10:05:12.196025: step: 1364/529, loss: 0.037766262888908386 2023-01-21 10:05:13.406249: step: 1368/529, loss: 0.11279010772705078 2023-01-21 10:05:14.617835: step: 1372/529, loss: 0.022310638800263405 2023-01-21 10:05:15.801310: step: 1376/529, loss: 0.0952509418129921 2023-01-21 10:05:16.974104: step: 1380/529, loss: 0.11090002208948135 2023-01-21 10:05:18.144030: step: 1384/529, loss: 0.08343953639268875 2023-01-21 10:05:19.357386: step: 1388/529, loss: 0.04089689254760742 2023-01-21 10:05:20.525512: step: 1392/529, loss: 0.30994436144828796 2023-01-21 10:05:21.724553: step: 1396/529, loss: 0.11132745444774628 2023-01-21 10:05:22.894465: step: 1400/529, loss: 0.05379399657249451 2023-01-21 10:05:24.108158: step: 1404/529, loss: 0.11103429645299911 2023-01-21 10:05:25.334164: step: 1408/529, loss: 0.09269294887781143 2023-01-21 10:05:26.543487: step: 1412/529, loss: 0.03979680687189102 2023-01-21 10:05:27.709758: step: 1416/529, loss: 0.12062730640172958 2023-01-21 10:05:28.921524: step: 1420/529, loss: 0.3484598696231842 2023-01-21 10:05:30.091111: step: 1424/529, loss: 0.06316041946411133 2023-01-21 10:05:31.306730: step: 1428/529, loss: 0.12038116157054901 2023-01-21 10:05:32.494842: step: 1432/529, loss: 0.050589751452207565 2023-01-21 10:05:33.697748: step: 1436/529, loss: 0.12125988304615021 2023-01-21 10:05:34.872518: step: 1440/529, loss: 0.20059843361377716 2023-01-21 10:05:36.059929: step: 1444/529, loss: 0.019537828862667084 2023-01-21 10:05:37.265882: step: 1448/529, loss: 0.14411191642284393 2023-01-21 10:05:38.515479: step: 1452/529, loss: 0.04714317247271538 2023-01-21 10:05:39.722119: step: 1456/529, loss: 0.3148542642593384 2023-01-21 10:05:40.897993: step: 1460/529, loss: 0.24474793672561646 2023-01-21 10:05:42.078486: step: 1464/529, loss: 0.05517106130719185 2023-01-21 10:05:43.267083: step: 1468/529, loss: 0.225184828042984 2023-01-21 10:05:44.428930: step: 1472/529, loss: 0.06958065181970596 2023-01-21 10:05:45.608772: step: 1476/529, loss: 0.1354474127292633 2023-01-21 10:05:46.795574: step: 1480/529, loss: 0.02657771110534668 2023-01-21 10:05:47.999467: step: 1484/529, loss: 0.09428386390209198 2023-01-21 10:05:49.214476: step: 1488/529, loss: 0.19674520194530487 2023-01-21 10:05:50.419125: step: 1492/529, loss: 0.11501121520996094 2023-01-21 10:05:51.589889: step: 1496/529, loss: 0.19871121644973755 2023-01-21 10:05:52.817819: step: 1500/529, loss: 0.4033264219760895 2023-01-21 10:05:54.000525: step: 1504/529, loss: 0.12177863717079163 2023-01-21 10:05:55.200640: step: 1508/529, loss: 0.08350207656621933 2023-01-21 10:05:56.384811: step: 1512/529, loss: 0.019609831273555756 2023-01-21 10:05:57.561091: step: 1516/529, loss: 0.3669924736022949 2023-01-21 10:05:58.761689: step: 1520/529, loss: 0.03388643264770508 2023-01-21 10:05:59.948943: step: 1524/529, loss: 0.18395265936851501 2023-01-21 10:06:01.157018: step: 1528/529, loss: 0.1612405776977539 2023-01-21 10:06:02.363248: step: 1532/529, loss: 0.13534298539161682 2023-01-21 10:06:03.545615: step: 1536/529, loss: 0.07960841804742813 2023-01-21 10:06:04.730290: step: 1540/529, loss: 0.12602612376213074 2023-01-21 10:06:05.940772: step: 1544/529, loss: 0.0558650977909565 2023-01-21 10:06:07.122551: step: 1548/529, loss: 0.03162384033203125 2023-01-21 10:06:08.346798: step: 1552/529, loss: 0.3522595465183258 2023-01-21 10:06:09.526719: step: 1556/529, loss: 0.2920095920562744 2023-01-21 10:06:10.716333: step: 1560/529, loss: 0.12839755415916443 2023-01-21 10:06:11.903020: step: 1564/529, loss: 0.04806976392865181 2023-01-21 10:06:13.071028: step: 1568/529, loss: 0.0692567378282547 2023-01-21 10:06:14.266202: step: 1572/529, loss: 0.10409679263830185 2023-01-21 10:06:15.447217: step: 1576/529, loss: 0.06270952522754669 2023-01-21 10:06:16.605191: step: 1580/529, loss: 0.11875371634960175 2023-01-21 10:06:17.800943: step: 1584/529, loss: 0.05838203802704811 2023-01-21 10:06:18.994700: step: 1588/529, loss: 0.06086568534374237 2023-01-21 10:06:20.193309: step: 1592/529, loss: 0.14741821587085724 2023-01-21 10:06:21.411389: step: 1596/529, loss: 0.10667534172534943 2023-01-21 10:06:22.591559: step: 1600/529, loss: 0.08822593837976456 2023-01-21 10:06:23.796504: step: 1604/529, loss: 0.05639181286096573 2023-01-21 10:06:25.013772: step: 1608/529, loss: 0.17751845717430115 2023-01-21 10:06:26.197028: step: 1612/529, loss: 0.02821202203631401 2023-01-21 10:06:27.474783: step: 1616/529, loss: 0.17321842908859253 2023-01-21 10:06:28.732402: step: 1620/529, loss: 0.19982223212718964 2023-01-21 10:06:29.947589: step: 1624/529, loss: 0.08125362545251846 2023-01-21 10:06:31.127138: step: 1628/529, loss: 0.02132434770464897 2023-01-21 10:06:32.301247: step: 1632/529, loss: 0.26734429597854614 2023-01-21 10:06:33.467906: step: 1636/529, loss: 0.06552913039922714 2023-01-21 10:06:34.640721: step: 1640/529, loss: 0.0243743434548378 2023-01-21 10:06:35.825718: step: 1644/529, loss: 0.03976898267865181 2023-01-21 10:06:37.013009: step: 1648/529, loss: 0.07415466755628586 2023-01-21 10:06:38.232664: step: 1652/529, loss: 0.15793848037719727 2023-01-21 10:06:39.440361: step: 1656/529, loss: 0.20690852403640747 2023-01-21 10:06:40.623329: step: 1660/529, loss: 0.05217461660504341 2023-01-21 10:06:41.811712: step: 1664/529, loss: 0.0488986037671566 2023-01-21 10:06:43.012220: step: 1668/529, loss: 0.3758572041988373 2023-01-21 10:06:44.181711: step: 1672/529, loss: 0.1287458837032318 2023-01-21 10:06:45.381093: step: 1676/529, loss: 0.03412642702460289 2023-01-21 10:06:46.566097: step: 1680/529, loss: 0.017566537484526634 2023-01-21 10:06:47.765871: step: 1684/529, loss: 0.22134742140769958 2023-01-21 10:06:48.963524: step: 1688/529, loss: 0.1414697766304016 2023-01-21 10:06:50.186025: step: 1692/529, loss: 0.1128886267542839 2023-01-21 10:06:51.357379: step: 1696/529, loss: 0.2069406658411026 2023-01-21 10:06:52.529018: step: 1700/529, loss: 0.07560800760984421 2023-01-21 10:06:53.723804: step: 1704/529, loss: 0.07611308246850967 2023-01-21 10:06:54.889247: step: 1708/529, loss: 0.07005896419286728 2023-01-21 10:06:56.115551: step: 1712/529, loss: 0.17178316414356232 2023-01-21 10:06:57.359614: step: 1716/529, loss: 0.05612926557660103 2023-01-21 10:06:58.583785: step: 1720/529, loss: 0.04977273941040039 2023-01-21 10:06:59.781049: step: 1724/529, loss: 0.15518073737621307 2023-01-21 10:07:00.964766: step: 1728/529, loss: 0.04040198028087616 2023-01-21 10:07:02.136072: step: 1732/529, loss: 0.08940983563661575 2023-01-21 10:07:03.303790: step: 1736/529, loss: 0.05588426813483238 2023-01-21 10:07:04.479271: step: 1740/529, loss: 0.024835502728819847 2023-01-21 10:07:05.662616: step: 1744/529, loss: 0.6345804333686829 2023-01-21 10:07:06.841262: step: 1748/529, loss: 0.47183215618133545 2023-01-21 10:07:08.050498: step: 1752/529, loss: 0.1791291981935501 2023-01-21 10:07:09.257513: step: 1756/529, loss: 0.037714697420597076 2023-01-21 10:07:10.473569: step: 1760/529, loss: 0.06665611267089844 2023-01-21 10:07:11.662840: step: 1764/529, loss: 0.6147672533988953 2023-01-21 10:07:12.867721: step: 1768/529, loss: 0.11704973876476288 2023-01-21 10:07:14.040812: step: 1772/529, loss: 0.04602517932653427 2023-01-21 10:07:15.239890: step: 1776/529, loss: 0.1902427077293396 2023-01-21 10:07:16.395395: step: 1780/529, loss: 0.08552742004394531 2023-01-21 10:07:17.573955: step: 1784/529, loss: 0.2217470109462738 2023-01-21 10:07:18.745343: step: 1788/529, loss: 0.1756361871957779 2023-01-21 10:07:19.948143: step: 1792/529, loss: 0.00928716640919447 2023-01-21 10:07:21.170477: step: 1796/529, loss: 0.07677898555994034 2023-01-21 10:07:22.353989: step: 1800/529, loss: 0.15860281884670258 2023-01-21 10:07:23.519324: step: 1804/529, loss: 0.1334318220615387 2023-01-21 10:07:24.733696: step: 1808/529, loss: 0.007172394078224897 2023-01-21 10:07:25.898466: step: 1812/529, loss: 0.022566795349121094 2023-01-21 10:07:27.098814: step: 1816/529, loss: 0.11337108910083771 2023-01-21 10:07:28.302938: step: 1820/529, loss: 0.05176963657140732 2023-01-21 10:07:29.468166: step: 1824/529, loss: 0.09789667278528214 2023-01-21 10:07:30.643606: step: 1828/529, loss: 0.20572558045387268 2023-01-21 10:07:31.881217: step: 1832/529, loss: 0.06294755637645721 2023-01-21 10:07:33.078776: step: 1836/529, loss: 0.08909421414136887 2023-01-21 10:07:34.263204: step: 1840/529, loss: 0.08354387432336807 2023-01-21 10:07:35.477299: step: 1844/529, loss: 0.072461798787117 2023-01-21 10:07:36.668967: step: 1848/529, loss: 0.13327093422412872 2023-01-21 10:07:37.869733: step: 1852/529, loss: 0.08729010075330734 2023-01-21 10:07:39.051862: step: 1856/529, loss: 0.09080667048692703 2023-01-21 10:07:40.302253: step: 1860/529, loss: 0.16696929931640625 2023-01-21 10:07:41.483742: step: 1864/529, loss: 0.16344699263572693 2023-01-21 10:07:42.679177: step: 1868/529, loss: 0.056607436388731 2023-01-21 10:07:43.890132: step: 1872/529, loss: 0.20858383178710938 2023-01-21 10:07:45.113170: step: 1876/529, loss: 0.15799474716186523 2023-01-21 10:07:46.303333: step: 1880/529, loss: 0.10217438638210297 2023-01-21 10:07:47.489545: step: 1884/529, loss: 0.024437617510557175 2023-01-21 10:07:48.717012: step: 1888/529, loss: 0.05741949379444122 2023-01-21 10:07:49.895808: step: 1892/529, loss: 0.09978371113538742 2023-01-21 10:07:51.093693: step: 1896/529, loss: 0.07088422775268555 2023-01-21 10:07:52.306154: step: 1900/529, loss: 0.7726289629936218 2023-01-21 10:07:53.512458: step: 1904/529, loss: 0.05080311372876167 2023-01-21 10:07:54.696746: step: 1908/529, loss: 0.19627884030342102 2023-01-21 10:07:55.911228: step: 1912/529, loss: 0.4477725923061371 2023-01-21 10:07:57.109336: step: 1916/529, loss: 0.5173094868659973 2023-01-21 10:07:58.313308: step: 1920/529, loss: 0.1326911896467209 2023-01-21 10:07:59.519123: step: 1924/529, loss: 0.2749665081501007 2023-01-21 10:08:00.698926: step: 1928/529, loss: 0.07716703414916992 2023-01-21 10:08:01.886416: step: 1932/529, loss: 0.19308114051818848 2023-01-21 10:08:03.067725: step: 1936/529, loss: 0.15743032097816467 2023-01-21 10:08:04.282317: step: 1940/529, loss: 0.23950806260108948 2023-01-21 10:08:05.448949: step: 1944/529, loss: 0.08223400264978409 2023-01-21 10:08:06.646613: step: 1948/529, loss: 0.0815381109714508 2023-01-21 10:08:07.824251: step: 1952/529, loss: 0.2271292805671692 2023-01-21 10:08:09.061230: step: 1956/529, loss: 0.030028581619262695 2023-01-21 10:08:10.225516: step: 1960/529, loss: 0.005736017599701881 2023-01-21 10:08:11.422247: step: 1964/529, loss: 0.12222181260585785 2023-01-21 10:08:12.579284: step: 1968/529, loss: 0.096971794962883 2023-01-21 10:08:13.737076: step: 1972/529, loss: 0.0944160521030426 2023-01-21 10:08:14.913279: step: 1976/529, loss: 0.18898820877075195 2023-01-21 10:08:16.096532: step: 1980/529, loss: 0.09631462395191193 2023-01-21 10:08:17.293395: step: 1984/529, loss: 0.07906635105609894 2023-01-21 10:08:18.443130: step: 1988/529, loss: 0.4967004358768463 2023-01-21 10:08:19.625586: step: 1992/529, loss: 0.04845285415649414 2023-01-21 10:08:20.815154: step: 1996/529, loss: 0.052182961255311966 2023-01-21 10:08:22.015863: step: 2000/529, loss: 0.419796347618103 2023-01-21 10:08:23.208800: step: 2004/529, loss: 0.10645198822021484 2023-01-21 10:08:24.368537: step: 2008/529, loss: 0.3282504975795746 2023-01-21 10:08:25.551840: step: 2012/529, loss: 0.06817064434289932 2023-01-21 10:08:26.765956: step: 2016/529, loss: 0.0825490951538086 2023-01-21 10:08:27.965957: step: 2020/529, loss: 0.30041542649269104 2023-01-21 10:08:29.137760: step: 2024/529, loss: 0.11773135513067245 2023-01-21 10:08:30.373475: step: 2028/529, loss: 0.07731147110462189 2023-01-21 10:08:31.586686: step: 2032/529, loss: 0.13552813231945038 2023-01-21 10:08:32.765288: step: 2036/529, loss: 0.0021768570877611637 2023-01-21 10:08:33.947041: step: 2040/529, loss: 0.15446814894676208 2023-01-21 10:08:35.097813: step: 2044/529, loss: 0.07837028801441193 2023-01-21 10:08:36.281417: step: 2048/529, loss: 0.1547018587589264 2023-01-21 10:08:37.485845: step: 2052/529, loss: 0.03462862968444824 2023-01-21 10:08:38.740925: step: 2056/529, loss: 0.2763025164604187 2023-01-21 10:08:39.950669: step: 2060/529, loss: 0.06091451644897461 2023-01-21 10:08:41.161659: step: 2064/529, loss: 0.06399951130151749 2023-01-21 10:08:42.344843: step: 2068/529, loss: 0.030105162411928177 2023-01-21 10:08:43.518110: step: 2072/529, loss: 0.07769875228404999 2023-01-21 10:08:44.679472: step: 2076/529, loss: 0.13661842048168182 2023-01-21 10:08:45.858300: step: 2080/529, loss: 0.0390084870159626 2023-01-21 10:08:47.065738: step: 2084/529, loss: 0.5840644836425781 2023-01-21 10:08:48.263707: step: 2088/529, loss: 0.09293203800916672 2023-01-21 10:08:49.476413: step: 2092/529, loss: 0.2014886885881424 2023-01-21 10:08:50.687844: step: 2096/529, loss: 0.16322669386863708 2023-01-21 10:08:51.898723: step: 2100/529, loss: 0.45108363032341003 2023-01-21 10:08:53.128110: step: 2104/529, loss: 0.24104803800582886 2023-01-21 10:08:54.341531: step: 2108/529, loss: 0.04788923263549805 2023-01-21 10:08:55.575007: step: 2112/529, loss: 0.10468444973230362 2023-01-21 10:08:56.773628: step: 2116/529, loss: 0.1076471358537674 ================================================== Loss: 0.156 -------------------- Dev: {'event': {'p': 0.6077586206896551, 'r': 0.7509986684420772, 'f1': 0.6718284693269803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Test: {'event': {'p': 0.6372795969773299, 'r': 0.7579388855602157, 'f1': 0.692391899288451}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Chinese: {'event': {'p': 0.6388888888888888, 'r': 0.8518518518518519, 'f1': 0.7301587301587301}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Korean: {'event': {'p': 0.58, 'r': 0.4603174603174603, 'f1': 0.5132743362831858}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Russian: {'event': {'p': 0.46153846153846156, 'r': 0.5, 'f1': 0.48000000000000004}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5972944849115505, 'r': 0.7643142476697736, 'f1': 0.6705607476635514}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.6162162162162163, 'r': 0.7513481126423007, 'f1': 0.6771058315334773}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} 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': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 8 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:09:42.119012: step: 4/529, loss: 0.04047767072916031 2023-01-21 10:09:43.308576: step: 8/529, loss: 0.07821826636791229 2023-01-21 10:09:44.511032: step: 12/529, loss: 0.07875081896781921 2023-01-21 10:09:45.701113: step: 16/529, loss: 0.04651594161987305 2023-01-21 10:09:46.844207: step: 20/529, loss: 0.1396748572587967 2023-01-21 10:09:48.017920: step: 24/529, loss: 0.013078117743134499 2023-01-21 10:09:49.182431: step: 28/529, loss: 0.08061981201171875 2023-01-21 10:09:50.422361: step: 32/529, loss: 0.0347994789481163 2023-01-21 10:09:51.683312: step: 36/529, loss: 0.08932581543922424 2023-01-21 10:09:52.887811: step: 40/529, loss: 0.1444837599992752 2023-01-21 10:09:54.041386: step: 44/529, loss: 0.011849736794829369 2023-01-21 10:09:55.235487: step: 48/529, loss: 0.06010847166180611 2023-01-21 10:09:56.419114: step: 52/529, loss: 0.16181078553199768 2023-01-21 10:09:57.599837: step: 56/529, loss: 0.09997274726629257 2023-01-21 10:09:58.814702: step: 60/529, loss: 0.08434572070837021 2023-01-21 10:10:00.007470: step: 64/529, loss: 0.1226482167840004 2023-01-21 10:10:01.195704: step: 68/529, loss: 0.04693803936243057 2023-01-21 10:10:02.433468: step: 72/529, loss: 0.07839357107877731 2023-01-21 10:10:03.628433: step: 76/529, loss: 0.13694123923778534 2023-01-21 10:10:04.806107: step: 80/529, loss: 0.05220532417297363 2023-01-21 10:10:05.964722: step: 84/529, loss: 0.16351184248924255 2023-01-21 10:10:07.153390: step: 88/529, loss: 0.029614927247166634 2023-01-21 10:10:08.337439: step: 92/529, loss: 0.2026420533657074 2023-01-21 10:10:09.544230: step: 96/529, loss: 0.008451461791992188 2023-01-21 10:10:10.718134: step: 100/529, loss: 0.053057052195072174 2023-01-21 10:10:11.889091: step: 104/529, loss: 0.01918935775756836 2023-01-21 10:10:13.105789: step: 108/529, loss: 0.04503031075000763 2023-01-21 10:10:14.291448: step: 112/529, loss: 0.024651335552334785 2023-01-21 10:10:15.489140: step: 116/529, loss: 0.29497799277305603 2023-01-21 10:10:16.681272: step: 120/529, loss: 0.21508580446243286 2023-01-21 10:10:17.916158: step: 124/529, loss: 0.05541248247027397 2023-01-21 10:10:19.105383: step: 128/529, loss: 0.012671947479248047 2023-01-21 10:10:20.323034: step: 132/529, loss: 0.10966930538415909 2023-01-21 10:10:21.539616: step: 136/529, loss: 0.0374639518558979 2023-01-21 10:10:22.737179: step: 140/529, loss: 0.049568939954042435 2023-01-21 10:10:23.917867: step: 144/529, loss: 0.08577098697423935 2023-01-21 10:10:25.077040: step: 148/529, loss: 0.013721656054258347 2023-01-21 10:10:26.298312: step: 152/529, loss: 0.3713884949684143 2023-01-21 10:10:27.461492: step: 156/529, loss: 0.03838930279016495 2023-01-21 10:10:28.686242: step: 160/529, loss: 0.11151674389839172 2023-01-21 10:10:29.891338: step: 164/529, loss: 0.15473294258117676 2023-01-21 10:10:31.110573: step: 168/529, loss: 0.14717569947242737 2023-01-21 10:10:32.266259: step: 172/529, loss: 0.0909331813454628 2023-01-21 10:10:33.425498: step: 176/529, loss: 0.017193149775266647 2023-01-21 10:10:34.622067: step: 180/529, loss: 0.008180523291230202 2023-01-21 10:10:35.850777: step: 184/529, loss: 0.07138733565807343 2023-01-21 10:10:37.023698: step: 188/529, loss: 0.012599324807524681 2023-01-21 10:10:38.190963: step: 192/529, loss: 0.1007021889090538 2023-01-21 10:10:39.385714: step: 196/529, loss: 0.11016488075256348 2023-01-21 10:10:40.594036: step: 200/529, loss: 0.07014036178588867 2023-01-21 10:10:41.762558: step: 204/529, loss: 0.4783112406730652 2023-01-21 10:10:42.937932: step: 208/529, loss: 0.03167133033275604 2023-01-21 10:10:44.119766: step: 212/529, loss: 0.1298878788948059 2023-01-21 10:10:45.285297: step: 216/529, loss: 0.11687121540307999 2023-01-21 10:10:46.466543: step: 220/529, loss: 0.021193360909819603 2023-01-21 10:10:47.691499: step: 224/529, loss: 0.17543688416481018 2023-01-21 10:10:48.959140: step: 228/529, loss: 0.17486333847045898 2023-01-21 10:10:50.123139: step: 232/529, loss: 0.06752453744411469 2023-01-21 10:10:51.318007: step: 236/529, loss: 0.07916069775819778 2023-01-21 10:10:52.513953: step: 240/529, loss: 0.06988201290369034 2023-01-21 10:10:53.709385: step: 244/529, loss: 0.025316622108221054 2023-01-21 10:10:54.952422: step: 248/529, loss: 0.13482967019081116 2023-01-21 10:10:56.132345: step: 252/529, loss: 0.11785188317298889 2023-01-21 10:10:57.350090: step: 256/529, loss: 0.1267215758562088 2023-01-21 10:10:58.552447: step: 260/529, loss: 0.2078627347946167 2023-01-21 10:10:59.755708: step: 264/529, loss: 1.0372698307037354 2023-01-21 10:11:00.955902: step: 268/529, loss: 0.090336374938488 2023-01-21 10:11:02.134133: step: 272/529, loss: 0.040514782071113586 2023-01-21 10:11:03.348344: step: 276/529, loss: 0.0604795441031456 2023-01-21 10:11:04.610303: step: 280/529, loss: 0.13312676548957825 2023-01-21 10:11:05.812391: step: 284/529, loss: 0.04873685911297798 2023-01-21 10:11:07.019428: step: 288/529, loss: 0.04018230736255646 2023-01-21 10:11:08.226430: step: 292/529, loss: 0.08473005890846252 2023-01-21 10:11:09.432811: step: 296/529, loss: 0.0846981555223465 2023-01-21 10:11:10.611735: step: 300/529, loss: 0.10076389461755753 2023-01-21 10:11:11.815341: step: 304/529, loss: 0.041455648839473724 2023-01-21 10:11:13.013481: step: 308/529, loss: 0.07964906841516495 2023-01-21 10:11:14.194081: step: 312/529, loss: 0.2698894441127777 2023-01-21 10:11:15.384070: step: 316/529, loss: 0.02723999135196209 2023-01-21 10:11:16.541135: step: 320/529, loss: 0.4251673221588135 2023-01-21 10:11:17.721763: step: 324/529, loss: 0.1616990566253662 2023-01-21 10:11:18.931560: step: 328/529, loss: 0.07225904613733292 2023-01-21 10:11:20.146619: step: 332/529, loss: 0.039491843432188034 2023-01-21 10:11:21.347225: step: 336/529, loss: 0.347970575094223 2023-01-21 10:11:22.575549: step: 340/529, loss: 0.48211658000946045 2023-01-21 10:11:23.747422: step: 344/529, loss: 0.08665800094604492 2023-01-21 10:11:24.951873: step: 348/529, loss: 0.037229862064123154 2023-01-21 10:11:26.133029: step: 352/529, loss: 0.05745267868041992 2023-01-21 10:11:27.315598: step: 356/529, loss: 0.07717876881361008 2023-01-21 10:11:28.506902: step: 360/529, loss: 0.09231309592723846 2023-01-21 10:11:29.705396: step: 364/529, loss: 0.07006492465734482 2023-01-21 10:11:30.895357: step: 368/529, loss: 0.45909157395362854 2023-01-21 10:11:32.111331: step: 372/529, loss: 0.07901191711425781 2023-01-21 10:11:33.315033: step: 376/529, loss: 0.06261923909187317 2023-01-21 10:11:34.527657: step: 380/529, loss: 0.1023443266749382 2023-01-21 10:11:35.716047: step: 384/529, loss: 0.18097075819969177 2023-01-21 10:11:36.876579: step: 388/529, loss: 0.027232695370912552 2023-01-21 10:11:38.087173: step: 392/529, loss: 0.27917614579200745 2023-01-21 10:11:39.251984: step: 396/529, loss: 0.03713860362768173 2023-01-21 10:11:40.470258: step: 400/529, loss: 0.1250859797000885 2023-01-21 10:11:41.723033: step: 404/529, loss: 0.07704105973243713 2023-01-21 10:11:42.886041: step: 408/529, loss: 0.04263181611895561 2023-01-21 10:11:44.096336: step: 412/529, loss: 0.04491834715008736 2023-01-21 10:11:45.280986: step: 416/529, loss: 0.21583300828933716 2023-01-21 10:11:46.546752: step: 420/529, loss: 0.16193485260009766 2023-01-21 10:11:47.764786: step: 424/529, loss: 0.48976191878318787 2023-01-21 10:11:48.947443: step: 428/529, loss: 0.03811531513929367 2023-01-21 10:11:50.129908: step: 432/529, loss: 0.08802290260791779 2023-01-21 10:11:51.321648: step: 436/529, loss: 0.09434080123901367 2023-01-21 10:11:52.479691: step: 440/529, loss: 0.10715632140636444 2023-01-21 10:11:53.683292: step: 444/529, loss: 0.09228632599115372 2023-01-21 10:11:54.911589: step: 448/529, loss: 0.11972329765558243 2023-01-21 10:11:56.092081: step: 452/529, loss: 0.0010260582203045487 2023-01-21 10:11:57.294384: step: 456/529, loss: 0.04023704677820206 2023-01-21 10:11:58.445217: step: 460/529, loss: 0.004379749298095703 2023-01-21 10:11:59.585492: step: 464/529, loss: 0.07423634827136993 2023-01-21 10:12:00.778286: step: 468/529, loss: 0.03722720220685005 2023-01-21 10:12:01.979146: step: 472/529, loss: 0.03283347934484482 2023-01-21 10:12:03.154664: step: 476/529, loss: 0.7855027914047241 2023-01-21 10:12:04.370893: step: 480/529, loss: 0.06754789501428604 2023-01-21 10:12:05.507914: step: 484/529, loss: 0.020860005170106888 2023-01-21 10:12:06.693552: step: 488/529, loss: 0.023627091199159622 2023-01-21 10:12:07.866672: step: 492/529, loss: 0.005011129658669233 2023-01-21 10:12:09.058112: step: 496/529, loss: 0.06558027118444443 2023-01-21 10:12:10.248165: step: 500/529, loss: 0.06105814129114151 2023-01-21 10:12:11.409435: step: 504/529, loss: 0.16027694940567017 2023-01-21 10:12:12.650169: step: 508/529, loss: 0.08924088627099991 2023-01-21 10:12:13.874284: step: 512/529, loss: 0.39572152495384216 2023-01-21 10:12:15.097324: step: 516/529, loss: 0.10617408901453018 2023-01-21 10:12:16.289933: step: 520/529, loss: 0.13908430933952332 2023-01-21 10:12:17.489000: step: 524/529, loss: 0.0780036449432373 2023-01-21 10:12:18.692892: step: 528/529, loss: 0.3542849123477936 2023-01-21 10:12:19.884581: step: 532/529, loss: 0.03689432144165039 2023-01-21 10:12:21.100177: step: 536/529, loss: 0.16381417214870453 2023-01-21 10:12:22.306281: step: 540/529, loss: 0.05505218356847763 2023-01-21 10:12:23.529297: step: 544/529, loss: 1.293656349182129 2023-01-21 10:12:24.706330: step: 548/529, loss: 0.12761688232421875 2023-01-21 10:12:25.892169: step: 552/529, loss: 0.19070672988891602 2023-01-21 10:12:27.060322: step: 556/529, loss: 0.013144303113222122 2023-01-21 10:12:28.313062: step: 560/529, loss: 0.08128806203603745 2023-01-21 10:12:29.467747: step: 564/529, loss: 0.05706482008099556 2023-01-21 10:12:30.690503: step: 568/529, loss: 0.1587892472743988 2023-01-21 10:12:31.886089: step: 572/529, loss: 0.10761166363954544 2023-01-21 10:12:33.067703: step: 576/529, loss: 0.06638222187757492 2023-01-21 10:12:34.266376: step: 580/529, loss: 0.09939480572938919 2023-01-21 10:12:35.470094: step: 584/529, loss: 0.14611834287643433 2023-01-21 10:12:36.651221: step: 588/529, loss: 0.007876873016357422 2023-01-21 10:12:37.859974: step: 592/529, loss: 0.12069825828075409 2023-01-21 10:12:39.020400: step: 596/529, loss: 0.0007598876836709678 2023-01-21 10:12:40.251520: step: 600/529, loss: 0.026388835161924362 2023-01-21 10:12:41.448008: step: 604/529, loss: 0.09614834189414978 2023-01-21 10:12:42.620755: step: 608/529, loss: 0.06546439975500107 2023-01-21 10:12:43.796189: step: 612/529, loss: 0.02298755757510662 2023-01-21 10:12:44.961759: step: 616/529, loss: 0.05813789367675781 2023-01-21 10:12:46.108566: step: 620/529, loss: 0.03316373750567436 2023-01-21 10:12:47.286801: step: 624/529, loss: 0.024449540302157402 2023-01-21 10:12:48.460592: step: 628/529, loss: 0.3957507908344269 2023-01-21 10:12:49.689127: step: 632/529, loss: 0.3080708384513855 2023-01-21 10:12:50.895479: step: 636/529, loss: 0.09192705154418945 2023-01-21 10:12:52.086451: step: 640/529, loss: 0.11697450280189514 2023-01-21 10:12:53.299922: step: 644/529, loss: 0.04845685884356499 2023-01-21 10:12:54.494376: step: 648/529, loss: 0.16040655970573425 2023-01-21 10:12:55.695238: step: 652/529, loss: 0.1269499808549881 2023-01-21 10:12:56.878598: step: 656/529, loss: 0.024476148188114166 2023-01-21 10:12:58.046767: step: 660/529, loss: 0.007953787222504616 2023-01-21 10:12:59.276851: step: 664/529, loss: 0.0908367931842804 2023-01-21 10:13:00.510481: step: 668/529, loss: 0.013273334130644798 2023-01-21 10:13:01.722686: step: 672/529, loss: 0.06754917651414871 2023-01-21 10:13:02.992306: step: 676/529, loss: 0.024636365473270416 2023-01-21 10:13:04.172291: step: 680/529, loss: 0.01600360870361328 2023-01-21 10:13:05.400792: step: 684/529, loss: 0.012657023034989834 2023-01-21 10:13:06.603922: step: 688/529, loss: 0.03296785056591034 2023-01-21 10:13:07.803774: step: 692/529, loss: 0.08263721317052841 2023-01-21 10:13:08.996224: step: 696/529, loss: 0.024071549996733665 2023-01-21 10:13:10.190892: step: 700/529, loss: 0.09579934924840927 2023-01-21 10:13:11.367042: step: 704/529, loss: 0.016787387430667877 2023-01-21 10:13:12.540546: step: 708/529, loss: 0.05615215376019478 2023-01-21 10:13:13.730303: step: 712/529, loss: 0.02637634426355362 2023-01-21 10:13:14.927370: step: 716/529, loss: 0.14427319169044495 2023-01-21 10:13:16.133049: step: 720/529, loss: 0.053510282188653946 2023-01-21 10:13:17.345214: step: 724/529, loss: 0.08245811611413956 2023-01-21 10:13:18.540376: step: 728/529, loss: 0.0813908576965332 2023-01-21 10:13:19.762103: step: 732/529, loss: 0.5283936262130737 2023-01-21 10:13:20.950768: step: 736/529, loss: 0.29870322346687317 2023-01-21 10:13:22.114946: step: 740/529, loss: 0.022414540871977806 2023-01-21 10:13:23.296762: step: 744/529, loss: 0.1062687486410141 2023-01-21 10:13:24.460282: step: 748/529, loss: 0.04527187719941139 2023-01-21 10:13:25.646669: step: 752/529, loss: 0.08696842193603516 2023-01-21 10:13:26.858867: step: 756/529, loss: 0.10458926856517792 2023-01-21 10:13:28.075973: step: 760/529, loss: 0.06515693664550781 2023-01-21 10:13:29.298652: step: 764/529, loss: 0.22541780769824982 2023-01-21 10:13:30.466213: step: 768/529, loss: 0.14605307579040527 2023-01-21 10:13:31.640068: step: 772/529, loss: 0.026258373633027077 2023-01-21 10:13:32.830961: step: 776/529, loss: 0.12572507560253143 2023-01-21 10:13:34.038049: step: 780/529, loss: 0.16668052971363068 2023-01-21 10:13:35.210280: step: 784/529, loss: 0.0823274701833725 2023-01-21 10:13:36.421368: step: 788/529, loss: 0.11530504375696182 2023-01-21 10:13:37.629823: step: 792/529, loss: 0.29284152388572693 2023-01-21 10:13:38.850908: step: 796/529, loss: 0.08278483897447586 2023-01-21 10:13:40.020980: step: 800/529, loss: 0.12059229612350464 2023-01-21 10:13:41.203752: step: 804/529, loss: 0.00380191789008677 2023-01-21 10:13:42.407629: step: 808/529, loss: 0.10135946422815323 2023-01-21 10:13:43.647033: step: 812/529, loss: 0.09400882571935654 2023-01-21 10:13:44.835012: step: 816/529, loss: 0.06097893789410591 2023-01-21 10:13:46.044144: step: 820/529, loss: 0.012851476669311523 2023-01-21 10:13:47.238049: step: 824/529, loss: 0.22244702279567719 2023-01-21 10:13:48.406541: step: 828/529, loss: 0.07800417393445969 2023-01-21 10:13:49.576251: step: 832/529, loss: 0.04298420250415802 2023-01-21 10:13:50.702300: step: 836/529, loss: 0.0540955513715744 2023-01-21 10:13:51.919518: step: 840/529, loss: 0.05601468309760094 2023-01-21 10:13:53.118297: step: 844/529, loss: 0.016303157433867455 2023-01-21 10:13:54.310226: step: 848/529, loss: 0.03289385139942169 2023-01-21 10:13:55.518430: step: 852/529, loss: 0.05658779293298721 2023-01-21 10:13:56.731722: step: 856/529, loss: 0.05487480014562607 2023-01-21 10:13:57.926188: step: 860/529, loss: 0.34701091051101685 2023-01-21 10:13:59.132763: step: 864/529, loss: 0.05912923812866211 2023-01-21 10:14:00.341595: step: 868/529, loss: 0.16625633835792542 2023-01-21 10:14:01.573332: step: 872/529, loss: 0.09264211356639862 2023-01-21 10:14:02.794881: step: 876/529, loss: 0.027820684015750885 2023-01-21 10:14:03.974040: step: 880/529, loss: 0.01377172488719225 2023-01-21 10:14:05.193719: step: 884/529, loss: 0.39780083298683167 2023-01-21 10:14:06.411043: step: 888/529, loss: 0.07220707088708878 2023-01-21 10:14:07.626546: step: 892/529, loss: 0.21084824204444885 2023-01-21 10:14:08.814246: step: 896/529, loss: 0.02554459497332573 2023-01-21 10:14:10.019047: step: 900/529, loss: 0.09172508865594864 2023-01-21 10:14:11.222598: step: 904/529, loss: 0.054864026606082916 2023-01-21 10:14:12.425398: step: 908/529, loss: 0.05406989902257919 2023-01-21 10:14:13.601918: step: 912/529, loss: 0.0252551082521677 2023-01-21 10:14:14.807504: step: 916/529, loss: 0.5439997315406799 2023-01-21 10:14:16.010476: step: 920/529, loss: 0.0982782393693924 2023-01-21 10:14:17.188096: step: 924/529, loss: 0.1017635390162468 2023-01-21 10:14:18.414014: step: 928/529, loss: 0.06407535821199417 2023-01-21 10:14:19.600754: step: 932/529, loss: 0.043857574462890625 2023-01-21 10:14:20.787196: step: 936/529, loss: 0.06193418800830841 2023-01-21 10:14:22.018338: step: 940/529, loss: 0.09406929463148117 2023-01-21 10:14:23.219329: step: 944/529, loss: 0.0677550807595253 2023-01-21 10:14:24.411387: step: 948/529, loss: 0.05266690254211426 2023-01-21 10:14:25.657024: step: 952/529, loss: 0.09897145628929138 2023-01-21 10:14:26.832397: step: 956/529, loss: 0.00474205007776618 2023-01-21 10:14:28.016584: step: 960/529, loss: 0.05670957267284393 2023-01-21 10:14:29.196323: step: 964/529, loss: 0.4810495972633362 2023-01-21 10:14:30.383697: step: 968/529, loss: 0.09206457436084747 2023-01-21 10:14:31.629054: step: 972/529, loss: 0.03547511249780655 2023-01-21 10:14:32.811199: step: 976/529, loss: 0.3007730543613434 2023-01-21 10:14:34.002817: step: 980/529, loss: 0.035579536110162735 2023-01-21 10:14:35.202225: step: 984/529, loss: 0.22938866913318634 2023-01-21 10:14:36.450370: step: 988/529, loss: 0.059946633875370026 2023-01-21 10:14:37.614851: step: 992/529, loss: 0.04224582016468048 2023-01-21 10:14:38.864665: step: 996/529, loss: 0.18574008345603943 2023-01-21 10:14:40.097852: step: 1000/529, loss: 0.14251522719860077 2023-01-21 10:14:41.332103: step: 1004/529, loss: 0.14577370882034302 2023-01-21 10:14:42.512315: step: 1008/529, loss: 0.01887531206011772 2023-01-21 10:14:43.692522: step: 1012/529, loss: 0.0874418243765831 2023-01-21 10:14:44.899017: step: 1016/529, loss: 0.8079484105110168 2023-01-21 10:14:46.077673: step: 1020/529, loss: 0.172323077917099 2023-01-21 10:14:47.237106: step: 1024/529, loss: 0.11157283931970596 2023-01-21 10:14:48.407023: step: 1028/529, loss: 0.036309339106082916 2023-01-21 10:14:49.636974: step: 1032/529, loss: 0.11312361061573029 2023-01-21 10:14:50.842208: step: 1036/529, loss: 0.07403126358985901 2023-01-21 10:14:52.029679: step: 1040/529, loss: 0.09309511631727219 2023-01-21 10:14:53.208026: step: 1044/529, loss: 0.13171759247779846 2023-01-21 10:14:54.404138: step: 1048/529, loss: 0.10741788148880005 2023-01-21 10:14:55.607387: step: 1052/529, loss: 0.11655206978321075 2023-01-21 10:14:56.810296: step: 1056/529, loss: 0.07213926315307617 2023-01-21 10:14:58.017633: step: 1060/529, loss: 0.07369041442871094 2023-01-21 10:14:59.184475: step: 1064/529, loss: 0.07184381037950516 2023-01-21 10:15:00.365837: step: 1068/529, loss: 0.060051776468753815 2023-01-21 10:15:01.535047: step: 1072/529, loss: 0.7109342813491821 2023-01-21 10:15:02.777591: step: 1076/529, loss: 0.03996581956744194 2023-01-21 10:15:03.946302: step: 1080/529, loss: 0.013586855493485928 2023-01-21 10:15:05.164063: step: 1084/529, loss: 0.014812374487519264 2023-01-21 10:15:06.332167: step: 1088/529, loss: 0.1448853611946106 2023-01-21 10:15:07.539111: step: 1092/529, loss: 0.08025312423706055 2023-01-21 10:15:08.720410: step: 1096/529, loss: 0.03768768161535263 2023-01-21 10:15:09.916400: step: 1100/529, loss: 0.17361049354076385 2023-01-21 10:15:11.084561: step: 1104/529, loss: 0.05589780956506729 2023-01-21 10:15:12.277425: step: 1108/529, loss: 0.1783219277858734 2023-01-21 10:15:13.444099: step: 1112/529, loss: 0.0425073616206646 2023-01-21 10:15:14.669305: step: 1116/529, loss: 0.10130815953016281 2023-01-21 10:15:15.909406: step: 1120/529, loss: 0.6068741083145142 2023-01-21 10:15:17.100682: step: 1124/529, loss: 0.13269658386707306 2023-01-21 10:15:18.279845: step: 1128/529, loss: 0.12661753594875336 2023-01-21 10:15:19.466841: step: 1132/529, loss: 0.05074606090784073 2023-01-21 10:15:20.677856: step: 1136/529, loss: 0.02085275761783123 2023-01-21 10:15:21.843315: step: 1140/529, loss: 0.13153213262557983 2023-01-21 10:15:23.031942: step: 1144/529, loss: 0.3103060722351074 2023-01-21 10:15:24.188099: step: 1148/529, loss: 0.025852108374238014 2023-01-21 10:15:25.414111: step: 1152/529, loss: 0.10076145827770233 2023-01-21 10:15:26.628630: step: 1156/529, loss: 0.05120129883289337 2023-01-21 10:15:27.766209: step: 1160/529, loss: 0.02644982375204563 2023-01-21 10:15:28.981918: step: 1164/529, loss: 0.19872704148292542 2023-01-21 10:15:30.178926: step: 1168/529, loss: 0.1585426777601242 2023-01-21 10:15:31.349430: step: 1172/529, loss: 0.32388150691986084 2023-01-21 10:15:32.553951: step: 1176/529, loss: 0.015065575018525124 2023-01-21 10:15:33.714352: step: 1180/529, loss: 0.08784134685993195 2023-01-21 10:15:34.930846: step: 1184/529, loss: 0.0757235512137413 2023-01-21 10:15:36.102631: step: 1188/529, loss: 0.035753894597291946 2023-01-21 10:15:37.299415: step: 1192/529, loss: 0.04402007907629013 2023-01-21 10:15:38.518887: step: 1196/529, loss: 0.10342884063720703 2023-01-21 10:15:39.722659: step: 1200/529, loss: 0.6368496417999268 2023-01-21 10:15:40.933898: step: 1204/529, loss: 0.24611268937587738 2023-01-21 10:15:42.197761: step: 1208/529, loss: 0.1868046224117279 2023-01-21 10:15:43.405815: step: 1212/529, loss: 0.04192471504211426 2023-01-21 10:15:44.606518: step: 1216/529, loss: 0.38344359397888184 2023-01-21 10:15:45.766135: step: 1220/529, loss: 0.09596691280603409 2023-01-21 10:15:46.964346: step: 1224/529, loss: 0.16268451511859894 2023-01-21 10:15:48.154068: step: 1228/529, loss: 0.07357574254274368 2023-01-21 10:15:49.346115: step: 1232/529, loss: 0.04232821613550186 2023-01-21 10:15:50.545328: step: 1236/529, loss: 0.029508303850889206 2023-01-21 10:15:51.759256: step: 1240/529, loss: 0.05965328589081764 2023-01-21 10:15:52.931879: step: 1244/529, loss: 0.19239473342895508 2023-01-21 10:15:54.135985: step: 1248/529, loss: 0.0510801300406456 2023-01-21 10:15:55.314757: step: 1252/529, loss: 0.0352511890232563 2023-01-21 10:15:56.520785: step: 1256/529, loss: 0.3062821328639984 2023-01-21 10:15:57.759988: step: 1260/529, loss: 0.13641229271888733 2023-01-21 10:15:58.995790: step: 1264/529, loss: 0.23968066275119781 2023-01-21 10:16:00.200361: step: 1268/529, loss: 0.0965091735124588 2023-01-21 10:16:01.398949: step: 1272/529, loss: 0.21080684661865234 2023-01-21 10:16:02.613370: step: 1276/529, loss: 0.32785558700561523 2023-01-21 10:16:03.798709: step: 1280/529, loss: 0.16244274377822876 2023-01-21 10:16:04.981324: step: 1284/529, loss: 0.06273765861988068 2023-01-21 10:16:06.162986: step: 1288/529, loss: 0.09987225383520126 2023-01-21 10:16:07.390608: step: 1292/529, loss: 0.17433367669582367 2023-01-21 10:16:08.616458: step: 1296/529, loss: 0.2575455904006958 2023-01-21 10:16:09.848245: step: 1300/529, loss: 0.08590006828308105 2023-01-21 10:16:11.051152: step: 1304/529, loss: 0.08878374099731445 2023-01-21 10:16:12.275219: step: 1308/529, loss: 0.1368647664785385 2023-01-21 10:16:13.451093: step: 1312/529, loss: 0.0055213929153978825 2023-01-21 10:16:14.699956: step: 1316/529, loss: 0.09987707436084747 2023-01-21 10:16:15.885285: step: 1320/529, loss: 0.04167604818940163 2023-01-21 10:16:17.109416: step: 1324/529, loss: 0.17418260872364044 2023-01-21 10:16:18.347113: step: 1328/529, loss: 0.20379528403282166 2023-01-21 10:16:19.530111: step: 1332/529, loss: 0.12248249351978302 2023-01-21 10:16:20.715760: step: 1336/529, loss: 0.1846325844526291 2023-01-21 10:16:21.923214: step: 1340/529, loss: 0.09751148521900177 2023-01-21 10:16:23.093374: step: 1344/529, loss: 0.05286912992596626 2023-01-21 10:16:24.295829: step: 1348/529, loss: 0.09309554100036621 2023-01-21 10:16:25.531907: step: 1352/529, loss: 0.16922931373119354 2023-01-21 10:16:26.713889: step: 1356/529, loss: 0.13260327279567719 2023-01-21 10:16:27.890195: step: 1360/529, loss: 0.07543373107910156 2023-01-21 10:16:29.086023: step: 1364/529, loss: 0.11243017017841339 2023-01-21 10:16:30.271204: step: 1368/529, loss: 0.042791225016117096 2023-01-21 10:16:31.426493: step: 1372/529, loss: 0.12579889595508575 2023-01-21 10:16:32.632102: step: 1376/529, loss: 0.025398921221494675 2023-01-21 10:16:33.882380: step: 1380/529, loss: 0.11367006599903107 2023-01-21 10:16:35.038260: step: 1384/529, loss: 0.020359564572572708 2023-01-21 10:16:36.195927: step: 1388/529, loss: 0.02606039121747017 2023-01-21 10:16:37.388056: step: 1392/529, loss: 0.054610732942819595 2023-01-21 10:16:38.594405: step: 1396/529, loss: 0.3439517319202423 2023-01-21 10:16:39.757809: step: 1400/529, loss: 0.01565117947757244 2023-01-21 10:16:40.946465: step: 1404/529, loss: 0.0865788459777832 2023-01-21 10:16:42.137083: step: 1408/529, loss: 0.04938087612390518 2023-01-21 10:16:43.303853: step: 1412/529, loss: 0.04691317304968834 2023-01-21 10:16:44.467265: step: 1416/529, loss: 0.07006168365478516 2023-01-21 10:16:45.652912: step: 1420/529, loss: 0.11272773146629333 2023-01-21 10:16:46.882447: step: 1424/529, loss: 0.03806428983807564 2023-01-21 10:16:48.103866: step: 1428/529, loss: 0.03891425207257271 2023-01-21 10:16:49.276056: step: 1432/529, loss: 0.06599149852991104 2023-01-21 10:16:50.438500: step: 1436/529, loss: 0.0348605178296566 2023-01-21 10:16:51.616434: step: 1440/529, loss: 0.07947144657373428 2023-01-21 10:16:52.807628: step: 1444/529, loss: 0.06252908706665039 2023-01-21 10:16:53.975426: step: 1448/529, loss: 0.046996310353279114 2023-01-21 10:16:55.173566: step: 1452/529, loss: 0.1292612999677658 2023-01-21 10:16:56.311385: step: 1456/529, loss: 0.023672515526413918 2023-01-21 10:16:57.498359: step: 1460/529, loss: 0.41750413179397583 2023-01-21 10:16:58.720553: step: 1464/529, loss: 0.22391678392887115 2023-01-21 10:16:59.906744: step: 1468/529, loss: 0.18273591995239258 2023-01-21 10:17:01.084204: step: 1472/529, loss: 0.6467920541763306 2023-01-21 10:17:02.264016: step: 1476/529, loss: 0.06702125817537308 2023-01-21 10:17:03.455140: step: 1480/529, loss: 0.251875102519989 2023-01-21 10:17:04.663598: step: 1484/529, loss: 0.06994829326868057 2023-01-21 10:17:05.893744: step: 1488/529, loss: 0.3780898153781891 2023-01-21 10:17:07.068269: step: 1492/529, loss: 0.1128087043762207 2023-01-21 10:17:08.333060: step: 1496/529, loss: 0.18003129959106445 2023-01-21 10:17:09.506534: step: 1500/529, loss: 0.16708484292030334 2023-01-21 10:17:10.712972: step: 1504/529, loss: 0.050896644592285156 2023-01-21 10:17:11.889947: step: 1508/529, loss: 0.0634942501783371 2023-01-21 10:17:13.094009: step: 1512/529, loss: 0.11012440174818039 2023-01-21 10:17:14.311965: step: 1516/529, loss: 0.1159524917602539 2023-01-21 10:17:15.495499: step: 1520/529, loss: 0.09098625183105469 2023-01-21 10:17:16.692100: step: 1524/529, loss: 0.14877291023731232 2023-01-21 10:17:17.869539: step: 1528/529, loss: 0.027217866852879524 2023-01-21 10:17:19.107030: step: 1532/529, loss: 0.1360592246055603 2023-01-21 10:17:20.265862: step: 1536/529, loss: 0.11871004104614258 2023-01-21 10:17:21.425433: step: 1540/529, loss: 0.024248875677585602 2023-01-21 10:17:22.590225: step: 1544/529, loss: 0.04238881915807724 2023-01-21 10:17:23.814445: step: 1548/529, loss: 0.07538357377052307 2023-01-21 10:17:25.007133: step: 1552/529, loss: 0.26502570509910583 2023-01-21 10:17:26.217546: step: 1556/529, loss: 0.08521862328052521 2023-01-21 10:17:27.398753: step: 1560/529, loss: 0.10298766940832138 2023-01-21 10:17:28.605098: step: 1564/529, loss: 0.056702613830566406 2023-01-21 10:17:29.800483: step: 1568/529, loss: 0.31682783365249634 2023-01-21 10:17:31.025480: step: 1572/529, loss: 0.03723783418536186 2023-01-21 10:17:32.264135: step: 1576/529, loss: 0.034201718866825104 2023-01-21 10:17:33.447118: step: 1580/529, loss: 0.014857674017548561 2023-01-21 10:17:34.626341: step: 1584/529, loss: 0.08071684837341309 2023-01-21 10:17:35.839513: step: 1588/529, loss: 0.04908592998981476 2023-01-21 10:17:37.038610: step: 1592/529, loss: 0.057427406311035156 2023-01-21 10:17:38.244751: step: 1596/529, loss: 0.14947891235351562 2023-01-21 10:17:39.422038: step: 1600/529, loss: 0.13462170958518982 2023-01-21 10:17:40.568781: step: 1604/529, loss: 0.35665810108184814 2023-01-21 10:17:41.749281: step: 1608/529, loss: 0.034613847732543945 2023-01-21 10:17:42.943577: step: 1612/529, loss: 0.03477802127599716 2023-01-21 10:17:44.152731: step: 1616/529, loss: 0.2647930085659027 2023-01-21 10:17:45.330658: step: 1620/529, loss: 0.12565088272094727 2023-01-21 10:17:46.494745: step: 1624/529, loss: 0.895087480545044 2023-01-21 10:17:47.670667: step: 1628/529, loss: 0.23116064071655273 2023-01-21 10:17:48.847312: step: 1632/529, loss: 0.09947586059570312 2023-01-21 10:17:50.062921: step: 1636/529, loss: 0.09858989715576172 2023-01-21 10:17:51.290256: step: 1640/529, loss: 0.053867340087890625 2023-01-21 10:17:52.462279: step: 1644/529, loss: 0.01315016858279705 2023-01-21 10:17:53.701907: step: 1648/529, loss: 0.20471249520778656 2023-01-21 10:17:54.862718: step: 1652/529, loss: 0.05195903778076172 2023-01-21 10:17:56.092113: step: 1656/529, loss: 0.2735646069049835 2023-01-21 10:17:57.280766: step: 1660/529, loss: 0.15480592846870422 2023-01-21 10:17:58.447578: step: 1664/529, loss: 0.05930893495678902 2023-01-21 10:17:59.651767: step: 1668/529, loss: 0.12114687263965607 2023-01-21 10:18:00.857842: step: 1672/529, loss: 0.03629293292760849 2023-01-21 10:18:02.102304: step: 1676/529, loss: 0.10183429718017578 2023-01-21 10:18:03.265818: step: 1680/529, loss: 0.14595910906791687 2023-01-21 10:18:04.426723: step: 1684/529, loss: 0.14450454711914062 2023-01-21 10:18:05.629089: step: 1688/529, loss: 0.028574656695127487 2023-01-21 10:18:06.801214: step: 1692/529, loss: 0.03656186908483505 2023-01-21 10:18:08.064258: step: 1696/529, loss: 0.1851770430803299 2023-01-21 10:18:09.252432: step: 1700/529, loss: 0.037909604609012604 2023-01-21 10:18:10.403745: step: 1704/529, loss: 0.12020893394947052 2023-01-21 10:18:11.608757: step: 1708/529, loss: 0.013139819726347923 2023-01-21 10:18:12.838960: step: 1712/529, loss: 0.06804962456226349 2023-01-21 10:18:14.081530: step: 1716/529, loss: 0.20287780463695526 2023-01-21 10:18:15.283004: step: 1720/529, loss: 0.036954592913389206 2023-01-21 10:18:16.453866: step: 1724/529, loss: 0.13216161727905273 2023-01-21 10:18:17.639425: step: 1728/529, loss: 0.014828872866928577 2023-01-21 10:18:18.783972: step: 1732/529, loss: 0.01449351292103529 2023-01-21 10:18:19.963413: step: 1736/529, loss: 0.0587100088596344 2023-01-21 10:18:21.188377: step: 1740/529, loss: 0.12489938735961914 2023-01-21 10:18:22.359403: step: 1744/529, loss: 0.3651566505432129 2023-01-21 10:18:23.566908: step: 1748/529, loss: 0.0887083113193512 2023-01-21 10:18:24.753822: step: 1752/529, loss: 0.0992402583360672 2023-01-21 10:18:25.927441: step: 1756/529, loss: 0.12959490716457367 2023-01-21 10:18:27.165925: step: 1760/529, loss: 0.043139033019542694 2023-01-21 10:18:28.374533: step: 1764/529, loss: 0.12684936821460724 2023-01-21 10:18:29.584982: step: 1768/529, loss: 0.10574893653392792 2023-01-21 10:18:30.780521: step: 1772/529, loss: 0.004454374313354492 2023-01-21 10:18:31.944097: step: 1776/529, loss: 0.004342270083725452 2023-01-21 10:18:33.157204: step: 1780/529, loss: 0.04948420822620392 2023-01-21 10:18:34.346973: step: 1784/529, loss: 0.04153924062848091 2023-01-21 10:18:35.521965: step: 1788/529, loss: 0.706120491027832 2023-01-21 10:18:36.705193: step: 1792/529, loss: 0.0879400223493576 2023-01-21 10:18:37.908214: step: 1796/529, loss: 0.10746678709983826 2023-01-21 10:18:39.091503: step: 1800/529, loss: 0.06811189651489258 2023-01-21 10:18:40.257109: step: 1804/529, loss: 0.16225960850715637 2023-01-21 10:18:41.435469: step: 1808/529, loss: 0.26645317673683167 2023-01-21 10:18:42.649677: step: 1812/529, loss: 0.13903197646141052 2023-01-21 10:18:43.825866: step: 1816/529, loss: 0.10329466313123703 2023-01-21 10:18:45.057748: step: 1820/529, loss: 0.0913720577955246 2023-01-21 10:18:46.249275: step: 1824/529, loss: 0.16671404242515564 2023-01-21 10:18:47.447183: step: 1828/529, loss: 0.027530908584594727 2023-01-21 10:18:48.599629: step: 1832/529, loss: 0.6448995471000671 2023-01-21 10:18:49.748214: step: 1836/529, loss: 0.07307520508766174 2023-01-21 10:18:50.982894: step: 1840/529, loss: 0.10081490874290466 2023-01-21 10:18:52.179221: step: 1844/529, loss: 0.0912538543343544 2023-01-21 10:18:53.333361: step: 1848/529, loss: 0.11710606515407562 2023-01-21 10:18:54.513493: step: 1852/529, loss: 0.05176488310098648 2023-01-21 10:18:55.678148: step: 1856/529, loss: 0.2116812765598297 2023-01-21 10:18:56.880278: step: 1860/529, loss: 0.0839686393737793 2023-01-21 10:18:58.078668: step: 1864/529, loss: 0.07875747978687286 2023-01-21 10:18:59.319791: step: 1868/529, loss: 0.2987893521785736 2023-01-21 10:19:00.500799: step: 1872/529, loss: 0.03684444725513458 2023-01-21 10:19:01.688019: step: 1876/529, loss: 0.07712383568286896 2023-01-21 10:19:02.845723: step: 1880/529, loss: 0.010545921511948109 2023-01-21 10:19:04.014656: step: 1884/529, loss: 0.06599533557891846 2023-01-21 10:19:05.209646: step: 1888/529, loss: 0.025420378893613815 2023-01-21 10:19:06.425772: step: 1892/529, loss: 0.14610300958156586 2023-01-21 10:19:07.586484: step: 1896/529, loss: 0.05847196653485298 2023-01-21 10:19:08.770271: step: 1900/529, loss: 0.09752541035413742 2023-01-21 10:19:09.931377: step: 1904/529, loss: 0.026617813855409622 2023-01-21 10:19:11.116483: step: 1908/529, loss: 0.38210946321487427 2023-01-21 10:19:12.279103: step: 1912/529, loss: 0.042977333068847656 2023-01-21 10:19:13.505640: step: 1916/529, loss: 0.016511917114257812 2023-01-21 10:19:14.755920: step: 1920/529, loss: 0.07485198974609375 2023-01-21 10:19:15.973454: step: 1924/529, loss: 0.061173632740974426 2023-01-21 10:19:17.165166: step: 1928/529, loss: 0.05647630989551544 2023-01-21 10:19:18.343842: step: 1932/529, loss: 0.13428688049316406 2023-01-21 10:19:19.568484: step: 1936/529, loss: 0.056896306574344635 2023-01-21 10:19:20.788266: step: 1940/529, loss: 0.18245278298854828 2023-01-21 10:19:21.910615: step: 1944/529, loss: 0.06295967102050781 2023-01-21 10:19:23.114342: step: 1948/529, loss: 0.08530721813440323 2023-01-21 10:19:24.270678: step: 1952/529, loss: 0.17455103993415833 2023-01-21 10:19:25.450745: step: 1956/529, loss: 0.0513579398393631 2023-01-21 10:19:26.637085: step: 1960/529, loss: 0.06473426520824432 2023-01-21 10:19:27.810412: step: 1964/529, loss: 0.005498173646628857 2023-01-21 10:19:28.985194: step: 1968/529, loss: 0.0337066650390625 2023-01-21 10:19:30.173511: step: 1972/529, loss: 0.5874265432357788 2023-01-21 10:19:31.323679: step: 1976/529, loss: 0.12416896969079971 2023-01-21 10:19:32.548873: step: 1980/529, loss: 0.07729692757129669 2023-01-21 10:19:33.770436: step: 1984/529, loss: 0.11703939735889435 2023-01-21 10:19:34.963849: step: 1988/529, loss: 0.055304817855358124 2023-01-21 10:19:36.147438: step: 1992/529, loss: 0.09929418563842773 2023-01-21 10:19:37.361122: step: 1996/529, loss: 0.0662347823381424 2023-01-21 10:19:38.524930: step: 2000/529, loss: 0.09534512460231781 2023-01-21 10:19:39.752680: step: 2004/529, loss: 0.1967918872833252 2023-01-21 10:19:41.016060: step: 2008/529, loss: 0.08169937133789062 2023-01-21 10:19:42.210449: step: 2012/529, loss: 0.25774747133255005 2023-01-21 10:19:43.393675: step: 2016/529, loss: 0.10631094127893448 2023-01-21 10:19:44.562789: step: 2020/529, loss: 0.44311124086380005 2023-01-21 10:19:45.736239: step: 2024/529, loss: 0.023267174139618874 2023-01-21 10:19:46.886584: step: 2028/529, loss: 0.09495649486780167 2023-01-21 10:19:48.043763: step: 2032/529, loss: 0.03013906627893448 2023-01-21 10:19:49.220017: step: 2036/529, loss: 0.09901189804077148 2023-01-21 10:19:50.372923: step: 2040/529, loss: 0.011135769076645374 2023-01-21 10:19:51.555683: step: 2044/529, loss: 0.06298866868019104 2023-01-21 10:19:52.733729: step: 2048/529, loss: 0.019121933728456497 2023-01-21 10:19:53.926951: step: 2052/529, loss: 0.031234467402100563 2023-01-21 10:19:55.145536: step: 2056/529, loss: 0.017626380547881126 2023-01-21 10:19:56.335496: step: 2060/529, loss: 0.22762995958328247 2023-01-21 10:19:57.534002: step: 2064/529, loss: 0.14104080200195312 2023-01-21 10:19:58.682352: step: 2068/529, loss: 0.07647185027599335 2023-01-21 10:19:59.860821: step: 2072/529, loss: 0.40145397186279297 2023-01-21 10:20:01.093338: step: 2076/529, loss: 0.2017030417919159 2023-01-21 10:20:02.308583: step: 2080/529, loss: 0.1204218864440918 2023-01-21 10:20:03.483696: step: 2084/529, loss: 0.01604294776916504 2023-01-21 10:20:04.666750: step: 2088/529, loss: 0.8274124264717102 2023-01-21 10:20:05.859465: step: 2092/529, loss: 0.15177275240421295 2023-01-21 10:20:07.059061: step: 2096/529, loss: 0.9513418078422546 2023-01-21 10:20:08.251360: step: 2100/529, loss: 0.14701539278030396 2023-01-21 10:20:09.406109: step: 2104/529, loss: 0.3254031240940094 2023-01-21 10:20:10.611118: step: 2108/529, loss: 0.06617040932178497 2023-01-21 10:20:11.827469: step: 2112/529, loss: 0.7396628856658936 2023-01-21 10:20:13.020692: step: 2116/529, loss: 0.09895887225866318 ================================================== Loss: 0.127 -------------------- Dev: {'event': {'p': 0.5877106045589693, 'r': 0.7896138482023968, 'f1': 0.6738636363636363}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Test: {'event': {'p': 0.6253574833174452, 'r': 0.7860994607549431, 'f1': 0.6965755242898859}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} 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': 8} Korean: {'event': {'p': 0.6271186440677966, 'r': 0.5873015873015873, 'f1': 0.6065573770491803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Russian: {'event': {'p': 0.4883720930232558, 'r': 0.5833333333333334, 'f1': 0.5316455696202531}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5877106045589693, 'r': 0.7896138482023968, 'f1': 0.6738636363636363}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6253574833174452, 'r': 0.7860994607549431, 'f1': 0.6965755242898859}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.6271186440677966, 'r': 0.5873015873015873, 'f1': 0.6065573770491803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 9 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:21:02.128511: step: 4/529, loss: 0.07706203311681747 2023-01-21 10:21:03.336790: step: 8/529, loss: 0.3609369397163391 2023-01-21 10:21:04.505594: step: 12/529, loss: 0.02501654624938965 2023-01-21 10:21:05.712477: step: 16/529, loss: 0.20460176467895508 2023-01-21 10:21:06.916267: step: 20/529, loss: 0.05868654325604439 2023-01-21 10:21:08.138649: step: 24/529, loss: 0.1241549551486969 2023-01-21 10:21:09.321001: step: 28/529, loss: 0.027274703606963158 2023-01-21 10:21:10.498712: step: 32/529, loss: 0.09330911934375763 2023-01-21 10:21:11.643476: step: 36/529, loss: 0.10482750833034515 2023-01-21 10:21:12.846273: step: 40/529, loss: 0.09188003093004227 2023-01-21 10:21:14.014386: step: 44/529, loss: 0.042665816843509674 2023-01-21 10:21:15.184345: step: 48/529, loss: 0.10282287746667862 2023-01-21 10:21:16.408048: step: 52/529, loss: 0.04815168306231499 2023-01-21 10:21:17.641899: step: 56/529, loss: 0.002200603485107422 2023-01-21 10:21:18.857159: step: 60/529, loss: 0.08510227501392365 2023-01-21 10:21:20.051330: step: 64/529, loss: 0.05405187979340553 2023-01-21 10:21:21.227484: step: 68/529, loss: 0.04417545348405838 2023-01-21 10:21:22.419982: step: 72/529, loss: 0.2788141071796417 2023-01-21 10:21:23.637751: step: 76/529, loss: 0.07424087822437286 2023-01-21 10:21:24.799068: step: 80/529, loss: 0.011209297925233841 2023-01-21 10:21:25.979876: step: 84/529, loss: 0.014039039611816406 2023-01-21 10:21:27.174086: step: 88/529, loss: 0.014297867193818092 2023-01-21 10:21:28.355719: step: 92/529, loss: 0.029778480529785156 2023-01-21 10:21:29.559166: step: 96/529, loss: 0.09650173783302307 2023-01-21 10:21:30.725133: step: 100/529, loss: 0.021039199084043503 2023-01-21 10:21:31.920722: step: 104/529, loss: 0.3834657073020935 2023-01-21 10:21:33.146892: step: 108/529, loss: 0.054160404950380325 2023-01-21 10:21:34.320278: step: 112/529, loss: 0.029475783929228783 2023-01-21 10:21:35.491857: step: 116/529, loss: 0.15277805924415588 2023-01-21 10:21:36.700007: step: 120/529, loss: 0.12655220925807953 2023-01-21 10:21:37.937698: step: 124/529, loss: 0.05424446985125542 2023-01-21 10:21:39.128698: step: 128/529, loss: 0.020809365436434746 2023-01-21 10:21:40.299564: step: 132/529, loss: 0.0649077445268631 2023-01-21 10:21:41.472111: step: 136/529, loss: 0.052417658269405365 2023-01-21 10:21:42.695306: step: 140/529, loss: 0.026404570788145065 2023-01-21 10:21:43.858712: step: 144/529, loss: 0.11163578182458878 2023-01-21 10:21:45.056610: step: 148/529, loss: 0.06529025733470917 2023-01-21 10:21:46.246181: step: 152/529, loss: 0.29639601707458496 2023-01-21 10:21:47.432190: step: 156/529, loss: 0.07414093613624573 2023-01-21 10:21:48.600542: step: 160/529, loss: 0.0452701598405838 2023-01-21 10:21:49.823916: step: 164/529, loss: 0.1126430481672287 2023-01-21 10:21:50.996476: step: 168/529, loss: 0.06600398570299149 2023-01-21 10:21:52.167491: step: 172/529, loss: 0.05362091213464737 2023-01-21 10:21:53.363266: step: 176/529, loss: 0.14277419447898865 2023-01-21 10:21:54.555383: step: 180/529, loss: 0.0647861510515213 2023-01-21 10:21:55.759372: step: 184/529, loss: 0.07490501552820206 2023-01-21 10:21:56.963355: step: 188/529, loss: 0.08984088897705078 2023-01-21 10:21:58.194117: step: 192/529, loss: 0.2683661878108978 2023-01-21 10:21:59.431093: step: 196/529, loss: 0.04118680953979492 2023-01-21 10:22:00.609841: step: 200/529, loss: 0.12113752961158752 2023-01-21 10:22:01.868556: step: 204/529, loss: 0.13915586471557617 2023-01-21 10:22:03.071947: step: 208/529, loss: 0.3384190797805786 2023-01-21 10:22:04.247065: step: 212/529, loss: 0.028758669272065163 2023-01-21 10:22:05.433082: step: 216/529, loss: 0.03902549669146538 2023-01-21 10:22:06.646522: step: 220/529, loss: 0.05488548427820206 2023-01-21 10:22:07.840995: step: 224/529, loss: 0.12658318877220154 2023-01-21 10:22:09.001203: step: 228/529, loss: 0.019839953631162643 2023-01-21 10:22:10.237769: step: 232/529, loss: 0.19139966368675232 2023-01-21 10:22:11.425275: step: 236/529, loss: 0.08031488955020905 2023-01-21 10:22:12.622587: step: 240/529, loss: 0.2018882781267166 2023-01-21 10:22:13.816445: step: 244/529, loss: 0.04423055797815323 2023-01-21 10:22:15.041777: step: 248/529, loss: 0.0513455867767334 2023-01-21 10:22:16.251212: step: 252/529, loss: 0.05306091159582138 2023-01-21 10:22:17.469253: step: 256/529, loss: 0.07549601048231125 2023-01-21 10:22:18.680757: step: 260/529, loss: 0.27129754424095154 2023-01-21 10:22:19.885916: step: 264/529, loss: 0.34771308302879333 2023-01-21 10:22:21.102311: step: 268/529, loss: 0.04903598129749298 2023-01-21 10:22:22.279527: step: 272/529, loss: 0.03504333645105362 2023-01-21 10:22:23.482397: step: 276/529, loss: 0.133867084980011 2023-01-21 10:22:24.705017: step: 280/529, loss: 0.5706217885017395 2023-01-21 10:22:25.882247: step: 284/529, loss: 0.09402342140674591 2023-01-21 10:22:27.119217: step: 288/529, loss: 0.07075071334838867 2023-01-21 10:22:28.336233: step: 292/529, loss: 0.027519797906279564 2023-01-21 10:22:29.520077: step: 296/529, loss: 0.11909227818250656 2023-01-21 10:22:30.738604: step: 300/529, loss: 0.08387766033411026 2023-01-21 10:22:31.958547: step: 304/529, loss: 0.263397216796875 2023-01-21 10:22:33.145950: step: 308/529, loss: 0.020185183733701706 2023-01-21 10:22:34.351310: step: 312/529, loss: 0.03851518779993057 2023-01-21 10:22:35.510365: step: 316/529, loss: 0.056471921503543854 2023-01-21 10:22:36.692138: step: 320/529, loss: 0.14544352889060974 2023-01-21 10:22:37.858950: step: 324/529, loss: 0.04146280884742737 2023-01-21 10:22:39.060731: step: 328/529, loss: 0.03757286071777344 2023-01-21 10:22:40.272244: step: 332/529, loss: 0.17466364800930023 2023-01-21 10:22:41.470345: step: 336/529, loss: 0.17639651894569397 2023-01-21 10:22:42.655069: step: 340/529, loss: 0.1868622750043869 2023-01-21 10:22:43.864629: step: 344/529, loss: 0.14407387375831604 2023-01-21 10:22:45.046019: step: 348/529, loss: 0.026997661218047142 2023-01-21 10:22:46.300533: step: 352/529, loss: 0.21264734864234924 2023-01-21 10:22:47.486472: step: 356/529, loss: 0.1343534290790558 2023-01-21 10:22:48.641655: step: 360/529, loss: 0.23456978797912598 2023-01-21 10:22:49.797192: step: 364/529, loss: 0.04469137266278267 2023-01-21 10:22:50.998547: step: 368/529, loss: 0.009378815069794655 2023-01-21 10:22:52.222645: step: 372/529, loss: 0.24064597487449646 2023-01-21 10:22:53.461368: step: 376/529, loss: 0.0624295249581337 2023-01-21 10:22:54.656320: step: 380/529, loss: 0.49790287017822266 2023-01-21 10:22:55.922817: step: 384/529, loss: 0.07479248195886612 2023-01-21 10:22:57.136575: step: 388/529, loss: 0.08195725083351135 2023-01-21 10:22:58.324122: step: 392/529, loss: 0.07117652893066406 2023-01-21 10:22:59.502997: step: 396/529, loss: 0.01633152924478054 2023-01-21 10:23:00.748652: step: 400/529, loss: 0.02432394027709961 2023-01-21 10:23:01.970026: step: 404/529, loss: 0.04968385770916939 2023-01-21 10:23:03.136071: step: 408/529, loss: 0.0017351150745525956 2023-01-21 10:23:04.335785: step: 412/529, loss: 0.12085647881031036 2023-01-21 10:23:05.522520: step: 416/529, loss: 0.02084331586956978 2023-01-21 10:23:06.717982: step: 420/529, loss: 0.030452441424131393 2023-01-21 10:23:07.880546: step: 424/529, loss: 0.2088872790336609 2023-01-21 10:23:09.079664: step: 428/529, loss: 0.03261137008666992 2023-01-21 10:23:10.325271: step: 432/529, loss: 0.10151596367359161 2023-01-21 10:23:11.557349: step: 436/529, loss: 0.08158822357654572 2023-01-21 10:23:12.737415: step: 440/529, loss: 0.01679687574505806 2023-01-21 10:23:13.942152: step: 444/529, loss: 0.056443214416503906 2023-01-21 10:23:15.158937: step: 448/529, loss: 0.08200817555189133 2023-01-21 10:23:16.366519: step: 452/529, loss: 0.01014404371380806 2023-01-21 10:23:17.561235: step: 456/529, loss: 0.03626766428351402 2023-01-21 10:23:18.742867: step: 460/529, loss: 0.05161895602941513 2023-01-21 10:23:19.948428: step: 464/529, loss: 0.1402040421962738 2023-01-21 10:23:21.141225: step: 468/529, loss: 0.1648157238960266 2023-01-21 10:23:22.372244: step: 472/529, loss: 0.015301037579774857 2023-01-21 10:23:23.571646: step: 476/529, loss: 0.48414841294288635 2023-01-21 10:23:24.745782: step: 480/529, loss: 0.11328811943531036 2023-01-21 10:23:25.960041: step: 484/529, loss: 0.05878782272338867 2023-01-21 10:23:27.152931: step: 488/529, loss: 0.08626174926757812 2023-01-21 10:23:28.308966: step: 492/529, loss: 1.5473535060882568 2023-01-21 10:23:29.519999: step: 496/529, loss: 0.025368690490722656 2023-01-21 10:23:30.697424: step: 500/529, loss: 1.169867753982544 2023-01-21 10:23:31.942033: step: 504/529, loss: 0.5612404942512512 2023-01-21 10:23:33.183133: step: 508/529, loss: 0.09032583236694336 2023-01-21 10:23:34.380854: step: 512/529, loss: 0.03566131740808487 2023-01-21 10:23:35.527475: step: 516/529, loss: 0.014281749725341797 2023-01-21 10:23:36.714750: step: 520/529, loss: 0.006334591191262007 2023-01-21 10:23:37.920418: step: 524/529, loss: 0.06767146289348602 2023-01-21 10:23:39.100613: step: 528/529, loss: 0.13928499817848206 2023-01-21 10:23:40.406356: step: 532/529, loss: 0.10244293510913849 2023-01-21 10:23:41.593020: step: 536/529, loss: 0.019744396209716797 2023-01-21 10:23:42.758915: step: 540/529, loss: 0.036473989486694336 2023-01-21 10:23:43.943177: step: 544/529, loss: 0.09676847606897354 2023-01-21 10:23:45.158985: step: 548/529, loss: 0.10120868682861328 2023-01-21 10:23:46.411963: step: 552/529, loss: 0.13388314843177795 2023-01-21 10:23:47.598430: step: 556/529, loss: 0.014387417584657669 2023-01-21 10:23:48.780708: step: 560/529, loss: 0.10456142574548721 2023-01-21 10:23:49.968125: step: 564/529, loss: 0.13018293678760529 2023-01-21 10:23:51.124821: step: 568/529, loss: 0.2190563678741455 2023-01-21 10:23:52.311029: step: 572/529, loss: 0.09677457809448242 2023-01-21 10:23:53.505113: step: 576/529, loss: 0.129651740193367 2023-01-21 10:23:54.695278: step: 580/529, loss: 0.02453765831887722 2023-01-21 10:23:55.914642: step: 584/529, loss: 0.17237071692943573 2023-01-21 10:23:57.120747: step: 588/529, loss: 0.12866467237472534 2023-01-21 10:23:58.317432: step: 592/529, loss: 0.1281770020723343 2023-01-21 10:23:59.523501: step: 596/529, loss: 0.11868076771497726 2023-01-21 10:24:00.702785: step: 600/529, loss: 0.005851459689438343 2023-01-21 10:24:01.923289: step: 604/529, loss: 0.06365146487951279 2023-01-21 10:24:03.145006: step: 608/529, loss: 0.025307275354862213 2023-01-21 10:24:04.308338: step: 612/529, loss: 0.05476546287536621 2023-01-21 10:24:05.503689: step: 616/529, loss: 0.032396603375673294 2023-01-21 10:24:06.717049: step: 620/529, loss: 0.33760377764701843 2023-01-21 10:24:07.887889: step: 624/529, loss: 0.11570902168750763 2023-01-21 10:24:09.115156: step: 628/529, loss: 0.04723663628101349 2023-01-21 10:24:10.305923: step: 632/529, loss: 0.021453188732266426 2023-01-21 10:24:11.518162: step: 636/529, loss: 0.04021596908569336 2023-01-21 10:24:12.743475: step: 640/529, loss: 0.04281273111701012 2023-01-21 10:24:14.003388: step: 644/529, loss: 0.060198500752449036 2023-01-21 10:24:15.224326: step: 648/529, loss: 0.289010226726532 2023-01-21 10:24:16.428605: step: 652/529, loss: 0.05201692879199982 2023-01-21 10:24:17.629521: step: 656/529, loss: 0.04979963228106499 2023-01-21 10:24:18.829957: step: 660/529, loss: 0.8255425095558167 2023-01-21 10:24:20.032404: step: 664/529, loss: 0.025757886469364166 2023-01-21 10:24:21.206060: step: 668/529, loss: 0.1857587695121765 2023-01-21 10:24:22.415473: step: 672/529, loss: 0.0813564658164978 2023-01-21 10:24:23.626806: step: 676/529, loss: 0.34784936904907227 2023-01-21 10:24:24.866526: step: 680/529, loss: 0.0848187580704689 2023-01-21 10:24:26.032078: step: 684/529, loss: 0.06092672422528267 2023-01-21 10:24:27.267356: step: 688/529, loss: 1.097140908241272 2023-01-21 10:24:28.471267: step: 692/529, loss: 0.01776568964123726 2023-01-21 10:24:29.682764: step: 696/529, loss: 0.12852688133716583 2023-01-21 10:24:30.866592: step: 700/529, loss: 0.050512123852968216 2023-01-21 10:24:32.107668: step: 704/529, loss: 0.07791423797607422 2023-01-21 10:24:33.339994: step: 708/529, loss: 0.03385234251618385 2023-01-21 10:24:34.575869: step: 712/529, loss: 0.02483396604657173 2023-01-21 10:24:35.780876: step: 716/529, loss: 0.15987731516361237 2023-01-21 10:24:36.978303: step: 720/529, loss: 0.04217751324176788 2023-01-21 10:24:38.162547: step: 724/529, loss: 0.07350321114063263 2023-01-21 10:24:39.372489: step: 728/529, loss: 0.1448248028755188 2023-01-21 10:24:40.555367: step: 732/529, loss: 0.0734289139509201 2023-01-21 10:24:41.723382: step: 736/529, loss: 0.0869404524564743 2023-01-21 10:24:42.898065: step: 740/529, loss: 0.04873237758874893 2023-01-21 10:24:44.110682: step: 744/529, loss: 0.05395345389842987 2023-01-21 10:24:45.327584: step: 748/529, loss: 0.03888511657714844 2023-01-21 10:24:46.517685: step: 752/529, loss: 0.05777006223797798 2023-01-21 10:24:47.730363: step: 756/529, loss: 0.2934258282184601 2023-01-21 10:24:48.957604: step: 760/529, loss: 0.05800657346844673 2023-01-21 10:24:50.192531: step: 764/529, loss: 0.08716487884521484 2023-01-21 10:24:51.398672: step: 768/529, loss: 0.02558775059878826 2023-01-21 10:24:52.601414: step: 772/529, loss: 0.008740568533539772 2023-01-21 10:24:53.776205: step: 776/529, loss: 0.015002918429672718 2023-01-21 10:24:55.034442: step: 780/529, loss: 0.07867622375488281 2023-01-21 10:24:56.243526: step: 784/529, loss: 0.15878896415233612 2023-01-21 10:24:57.413170: step: 788/529, loss: 0.06344223022460938 2023-01-21 10:24:58.609752: step: 792/529, loss: 0.21710500121116638 2023-01-21 10:24:59.771439: step: 796/529, loss: 0.054468631744384766 2023-01-21 10:25:00.954115: step: 800/529, loss: 0.09668131172657013 2023-01-21 10:25:02.133101: step: 804/529, loss: 0.03438092768192291 2023-01-21 10:25:03.331473: step: 808/529, loss: 0.06402941048145294 2023-01-21 10:25:04.548788: step: 812/529, loss: 0.11263751983642578 2023-01-21 10:25:05.734442: step: 816/529, loss: 0.020251015201210976 2023-01-21 10:25:06.937717: step: 820/529, loss: 0.10936126857995987 2023-01-21 10:25:08.191691: step: 824/529, loss: 0.019695425406098366 2023-01-21 10:25:09.373437: step: 828/529, loss: 0.05466928705573082 2023-01-21 10:25:10.538393: step: 832/529, loss: 0.16966834664344788 2023-01-21 10:25:11.730858: step: 836/529, loss: 0.12107186019420624 2023-01-21 10:25:12.898681: step: 840/529, loss: 0.03607664257287979 2023-01-21 10:25:14.071824: step: 844/529, loss: 0.10125122219324112 2023-01-21 10:25:15.261963: step: 848/529, loss: 0.015216827392578125 2023-01-21 10:25:16.431295: step: 852/529, loss: 0.08168697357177734 2023-01-21 10:25:17.617463: step: 856/529, loss: 0.06559620052576065 2023-01-21 10:25:18.806923: step: 860/529, loss: 0.0023474693298339844 2023-01-21 10:25:20.005879: step: 864/529, loss: 0.09548597037792206 2023-01-21 10:25:21.191943: step: 868/529, loss: 0.02702934667468071 2023-01-21 10:25:22.404737: step: 872/529, loss: 0.043364908546209335 2023-01-21 10:25:23.612312: step: 876/529, loss: 0.033113572746515274 2023-01-21 10:25:24.845272: step: 880/529, loss: 0.04623126983642578 2023-01-21 10:25:26.037173: step: 884/529, loss: 0.06952495872974396 2023-01-21 10:25:27.177899: step: 888/529, loss: 0.054462432861328125 2023-01-21 10:25:28.388096: step: 892/529, loss: 0.23846206068992615 2023-01-21 10:25:29.578493: step: 896/529, loss: 0.03779583051800728 2023-01-21 10:25:30.759347: step: 900/529, loss: 0.12914705276489258 2023-01-21 10:25:31.968285: step: 904/529, loss: 0.06119070202112198 2023-01-21 10:25:33.148865: step: 908/529, loss: 0.002698421711102128 2023-01-21 10:25:34.380026: step: 912/529, loss: 0.38810914754867554 2023-01-21 10:25:35.590433: step: 916/529, loss: 0.04940776899456978 2023-01-21 10:25:36.767685: step: 920/529, loss: 0.03824205324053764 2023-01-21 10:25:37.975861: step: 924/529, loss: 0.06851714849472046 2023-01-21 10:25:39.189827: step: 928/529, loss: 0.050595857203006744 2023-01-21 10:25:40.435417: step: 932/529, loss: 0.12183570861816406 2023-01-21 10:25:41.592062: step: 936/529, loss: 0.06306600570678711 2023-01-21 10:25:42.771005: step: 940/529, loss: 0.37963250279426575 2023-01-21 10:25:44.042880: step: 944/529, loss: 0.12578324973583221 2023-01-21 10:25:45.218738: step: 948/529, loss: 0.034774210304021835 2023-01-21 10:25:46.397358: step: 952/529, loss: 0.0017559052212163806 2023-01-21 10:25:47.570618: step: 956/529, loss: 0.0008977412944659591 2023-01-21 10:25:48.822191: step: 960/529, loss: 0.037227824330329895 2023-01-21 10:25:50.020511: step: 964/529, loss: 0.06928663700819016 2023-01-21 10:25:51.185635: step: 968/529, loss: 0.04085521399974823 2023-01-21 10:25:52.386590: step: 972/529, loss: 0.026029014959931374 2023-01-21 10:25:53.588824: step: 976/529, loss: 0.6921762228012085 2023-01-21 10:25:54.772830: step: 980/529, loss: 0.03453264385461807 2023-01-21 10:25:55.980804: step: 984/529, loss: 0.0025396347045898438 2023-01-21 10:25:57.201735: step: 988/529, loss: 0.14246463775634766 2023-01-21 10:25:58.420924: step: 992/529, loss: 0.1719975471496582 2023-01-21 10:25:59.589495: step: 996/529, loss: 0.012391663156449795 2023-01-21 10:26:00.830195: step: 1000/529, loss: 0.5715689659118652 2023-01-21 10:26:02.039649: step: 1004/529, loss: 0.2206006944179535 2023-01-21 10:26:03.246172: step: 1008/529, loss: 0.05662384256720543 2023-01-21 10:26:04.456818: step: 1012/529, loss: 0.08283748477697372 2023-01-21 10:26:05.622433: step: 1016/529, loss: 0.13019829988479614 2023-01-21 10:26:06.810668: step: 1020/529, loss: 0.04871978610754013 2023-01-21 10:26:07.995150: step: 1024/529, loss: 0.009431696496903896 2023-01-21 10:26:09.176041: step: 1028/529, loss: 0.09745875000953674 2023-01-21 10:26:10.381564: step: 1032/529, loss: 3.7824597358703613 2023-01-21 10:26:11.630328: step: 1036/529, loss: 0.05119842663407326 2023-01-21 10:26:12.824561: step: 1040/529, loss: 0.05995120853185654 2023-01-21 10:26:14.061920: step: 1044/529, loss: 0.22861547768115997 2023-01-21 10:26:15.243155: step: 1048/529, loss: 0.11673326790332794 2023-01-21 10:26:16.448982: step: 1052/529, loss: 0.02841949462890625 2023-01-21 10:26:17.643314: step: 1056/529, loss: 0.06642589718103409 2023-01-21 10:26:18.840589: step: 1060/529, loss: 0.08536510914564133 2023-01-21 10:26:20.038590: step: 1064/529, loss: 0.057392265647649765 2023-01-21 10:26:21.211680: step: 1068/529, loss: 0.018297864124178886 2023-01-21 10:26:22.418643: step: 1072/529, loss: 0.16273346543312073 2023-01-21 10:26:23.575760: step: 1076/529, loss: 0.03767561912536621 2023-01-21 10:26:24.753951: step: 1080/529, loss: 0.06970615684986115 2023-01-21 10:26:25.927035: step: 1084/529, loss: 0.0776124969124794 2023-01-21 10:26:27.098318: step: 1088/529, loss: 0.7339916229248047 2023-01-21 10:26:28.356759: step: 1092/529, loss: 0.03493137285113335 2023-01-21 10:26:29.603542: step: 1096/529, loss: 0.1935720443725586 2023-01-21 10:26:30.837394: step: 1100/529, loss: 0.04453315585851669 2023-01-21 10:26:32.024834: step: 1104/529, loss: 0.07666020095348358 2023-01-21 10:26:33.209744: step: 1108/529, loss: 0.042481180280447006 2023-01-21 10:26:34.445025: step: 1112/529, loss: 0.032961394637823105 2023-01-21 10:26:35.650378: step: 1116/529, loss: 0.051497265696525574 2023-01-21 10:26:36.862802: step: 1120/529, loss: 0.058836936950683594 2023-01-21 10:26:38.123533: step: 1124/529, loss: 0.055160000920295715 2023-01-21 10:26:39.303796: step: 1128/529, loss: 0.10562935471534729 2023-01-21 10:26:40.500585: step: 1132/529, loss: 0.0020885944832116365 2023-01-21 10:26:41.711616: step: 1136/529, loss: 0.03945513069629669 2023-01-21 10:26:42.914373: step: 1140/529, loss: 0.022968102246522903 2023-01-21 10:26:44.073983: step: 1144/529, loss: 0.11372071504592896 2023-01-21 10:26:45.243615: step: 1148/529, loss: 0.11726979911327362 2023-01-21 10:26:46.453207: step: 1152/529, loss: 0.008947182446718216 2023-01-21 10:26:47.630993: step: 1156/529, loss: 0.035971928387880325 2023-01-21 10:26:48.873917: step: 1160/529, loss: 0.11392173916101456 2023-01-21 10:26:50.031388: step: 1164/529, loss: 0.05277242511510849 2023-01-21 10:26:51.221106: step: 1168/529, loss: 0.03427296131849289 2023-01-21 10:26:52.422195: step: 1172/529, loss: 0.13313226401805878 2023-01-21 10:26:53.676034: step: 1176/529, loss: 0.045962050557136536 2023-01-21 10:26:54.904305: step: 1180/529, loss: 0.10334751754999161 2023-01-21 10:26:56.086329: step: 1184/529, loss: 0.028791427612304688 2023-01-21 10:26:57.289931: step: 1188/529, loss: 0.2083265483379364 2023-01-21 10:26:58.470745: step: 1192/529, loss: 0.044665709137916565 2023-01-21 10:26:59.657628: step: 1196/529, loss: 0.12532596290111542 2023-01-21 10:27:00.885460: step: 1200/529, loss: 0.06838683784008026 2023-01-21 10:27:02.104746: step: 1204/529, loss: 0.03884610906243324 2023-01-21 10:27:03.313955: step: 1208/529, loss: 0.026630450040102005 2023-01-21 10:27:04.531238: step: 1212/529, loss: 0.03295183181762695 2023-01-21 10:27:05.760033: step: 1216/529, loss: 0.1562710702419281 2023-01-21 10:27:07.009621: step: 1220/529, loss: 0.09037284553050995 2023-01-21 10:27:08.290794: step: 1224/529, loss: 0.1379532814025879 2023-01-21 10:27:09.460387: step: 1228/529, loss: 0.03228042274713516 2023-01-21 10:27:10.644595: step: 1232/529, loss: 0.03690236434340477 2023-01-21 10:27:11.851871: step: 1236/529, loss: 0.029663560912013054 2023-01-21 10:27:13.065955: step: 1240/529, loss: 0.07394389808177948 2023-01-21 10:27:14.235624: step: 1244/529, loss: 0.821942150592804 2023-01-21 10:27:15.484669: step: 1248/529, loss: 0.047072794288396835 2023-01-21 10:27:16.721488: step: 1252/529, loss: 0.1581771820783615 2023-01-21 10:27:17.949364: step: 1256/529, loss: 0.04433155059814453 2023-01-21 10:27:19.167625: step: 1260/529, loss: 0.023726653307676315 2023-01-21 10:27:20.387526: step: 1264/529, loss: 0.10929131507873535 2023-01-21 10:27:21.583207: step: 1268/529, loss: 0.021021651104092598 2023-01-21 10:27:22.783021: step: 1272/529, loss: 0.12015724182128906 2023-01-21 10:27:23.996915: step: 1276/529, loss: 0.1456432342529297 2023-01-21 10:27:25.234211: step: 1280/529, loss: 0.2869764268398285 2023-01-21 10:27:26.411534: step: 1284/529, loss: 0.06210336834192276 2023-01-21 10:27:27.584955: step: 1288/529, loss: 0.4272199869155884 2023-01-21 10:27:28.797327: step: 1292/529, loss: 0.08444433659315109 2023-01-21 10:27:30.028882: step: 1296/529, loss: 0.05749301612377167 2023-01-21 10:27:31.261969: step: 1300/529, loss: 0.10277929902076721 2023-01-21 10:27:32.412013: step: 1304/529, loss: 0.02456817775964737 2023-01-21 10:27:33.602456: step: 1308/529, loss: 0.012319850735366344 2023-01-21 10:27:34.788246: step: 1312/529, loss: 0.053867489099502563 2023-01-21 10:27:35.971805: step: 1316/529, loss: 0.1931808590888977 2023-01-21 10:27:37.181285: step: 1320/529, loss: 0.042501356452703476 2023-01-21 10:27:38.354071: step: 1324/529, loss: 0.030437849462032318 2023-01-21 10:27:39.568895: step: 1328/529, loss: 0.018532896414399147 2023-01-21 10:27:40.720196: step: 1332/529, loss: 0.04071979224681854 2023-01-21 10:27:41.907687: step: 1336/529, loss: 0.03715820237994194 2023-01-21 10:27:43.142551: step: 1340/529, loss: 0.12394581735134125 2023-01-21 10:27:44.332878: step: 1344/529, loss: 0.2526525557041168 2023-01-21 10:27:45.533581: step: 1348/529, loss: 0.10094404220581055 2023-01-21 10:27:46.724326: step: 1352/529, loss: 0.17400990426540375 2023-01-21 10:27:47.904915: step: 1356/529, loss: 0.1548875868320465 2023-01-21 10:27:49.090375: step: 1360/529, loss: 0.06449642032384872 2023-01-21 10:27:50.271047: step: 1364/529, loss: 0.16042232513427734 2023-01-21 10:27:51.467889: step: 1368/529, loss: 0.05946556478738785 2023-01-21 10:27:52.687055: step: 1372/529, loss: 0.05468807369470596 2023-01-21 10:27:53.861072: step: 1376/529, loss: 0.060177993029356 2023-01-21 10:27:55.066855: step: 1380/529, loss: 0.3314034342765808 2023-01-21 10:27:56.319943: step: 1384/529, loss: 0.10789716988801956 2023-01-21 10:27:57.511268: step: 1388/529, loss: 0.013530587777495384 2023-01-21 10:27:58.726460: step: 1392/529, loss: 0.02457904815673828 2023-01-21 10:27:59.880881: step: 1396/529, loss: 0.15012827515602112 2023-01-21 10:28:01.149004: step: 1400/529, loss: 0.15897050499916077 2023-01-21 10:28:02.369996: step: 1404/529, loss: 0.09443187713623047 2023-01-21 10:28:03.543139: step: 1408/529, loss: 0.06389837712049484 2023-01-21 10:28:04.714949: step: 1412/529, loss: 0.0392460823059082 2023-01-21 10:28:05.908535: step: 1416/529, loss: 0.03802771493792534 2023-01-21 10:28:07.071725: step: 1420/529, loss: 0.017196549102663994 2023-01-21 10:28:08.238995: step: 1424/529, loss: 0.0019979476928710938 2023-01-21 10:28:09.435773: step: 1428/529, loss: 0.05449257045984268 2023-01-21 10:28:10.680253: step: 1432/529, loss: 0.10396003723144531 2023-01-21 10:28:11.863473: step: 1436/529, loss: 0.11388321220874786 2023-01-21 10:28:13.063510: step: 1440/529, loss: 0.03711528703570366 2023-01-21 10:28:14.252363: step: 1444/529, loss: 0.39723843336105347 2023-01-21 10:28:15.500523: step: 1448/529, loss: 0.5783222317695618 2023-01-21 10:28:16.710928: step: 1452/529, loss: 0.27517127990722656 2023-01-21 10:28:17.910467: step: 1456/529, loss: 0.03254413604736328 2023-01-21 10:28:19.148079: step: 1460/529, loss: 0.11783038079738617 2023-01-21 10:28:20.311694: step: 1464/529, loss: 0.06386937946081161 2023-01-21 10:28:21.517424: step: 1468/529, loss: 0.0928686112165451 2023-01-21 10:28:22.680297: step: 1472/529, loss: 0.07659082859754562 2023-01-21 10:28:23.833815: step: 1476/529, loss: 0.1755451112985611 2023-01-21 10:28:25.025186: step: 1480/529, loss: 0.23981857299804688 2023-01-21 10:28:26.190404: step: 1484/529, loss: 0.023630524054169655 2023-01-21 10:28:27.362990: step: 1488/529, loss: 0.134020134806633 2023-01-21 10:28:28.598145: step: 1492/529, loss: 0.05197105556726456 2023-01-21 10:28:29.801280: step: 1496/529, loss: 0.025871945545077324 2023-01-21 10:28:30.990133: step: 1500/529, loss: 0.12018966674804688 2023-01-21 10:28:32.183961: step: 1504/529, loss: 0.04415726661682129 2023-01-21 10:28:33.358886: step: 1508/529, loss: 0.020696640014648438 2023-01-21 10:28:34.570941: step: 1512/529, loss: 0.1421644240617752 2023-01-21 10:28:35.743253: step: 1516/529, loss: 0.3093307614326477 2023-01-21 10:28:36.999297: step: 1520/529, loss: 0.03205394744873047 2023-01-21 10:28:38.208279: step: 1524/529, loss: 0.03148508071899414 2023-01-21 10:28:39.390367: step: 1528/529, loss: 0.05031432956457138 2023-01-21 10:28:40.586388: step: 1532/529, loss: 0.05000114440917969 2023-01-21 10:28:41.782713: step: 1536/529, loss: 0.02052149921655655 2023-01-21 10:28:42.971001: step: 1540/529, loss: 0.044590190052986145 2023-01-21 10:28:44.186357: step: 1544/529, loss: 0.06557836383581161 2023-01-21 10:28:45.345958: step: 1548/529, loss: 0.07163619995117188 2023-01-21 10:28:46.510105: step: 1552/529, loss: 0.16393285989761353 2023-01-21 10:28:47.725858: step: 1556/529, loss: 0.09536054730415344 2023-01-21 10:28:48.969304: step: 1560/529, loss: 0.2355089783668518 2023-01-21 10:28:50.150595: step: 1564/529, loss: 0.27672088146209717 2023-01-21 10:28:51.372826: step: 1568/529, loss: 0.08388538658618927 2023-01-21 10:28:52.581540: step: 1572/529, loss: 0.1348729133605957 2023-01-21 10:28:53.792862: step: 1576/529, loss: 0.06362266838550568 2023-01-21 10:28:54.954213: step: 1580/529, loss: 0.07376833260059357 2023-01-21 10:28:56.177554: step: 1584/529, loss: 0.1292678862810135 2023-01-21 10:28:57.400621: step: 1588/529, loss: 0.05223522335290909 2023-01-21 10:28:58.620975: step: 1592/529, loss: 0.037822578102350235 2023-01-21 10:28:59.822648: step: 1596/529, loss: 0.05437326431274414 2023-01-21 10:29:01.029039: step: 1600/529, loss: 0.10252627730369568 2023-01-21 10:29:02.200311: step: 1604/529, loss: 0.03990483283996582 2023-01-21 10:29:03.411546: step: 1608/529, loss: 0.05997011438012123 2023-01-21 10:29:04.624586: step: 1612/529, loss: 0.010653710924088955 2023-01-21 10:29:05.838037: step: 1616/529, loss: 0.12889929115772247 2023-01-21 10:29:07.044774: step: 1620/529, loss: 0.03674058988690376 2023-01-21 10:29:08.285621: step: 1624/529, loss: 0.09569883346557617 2023-01-21 10:29:09.482459: step: 1628/529, loss: 0.026285553351044655 2023-01-21 10:29:10.699286: step: 1632/529, loss: 0.07412166148424149 2023-01-21 10:29:11.859007: step: 1636/529, loss: 0.09094281494617462 2023-01-21 10:29:13.037025: step: 1640/529, loss: 0.08064451068639755 2023-01-21 10:29:14.237733: step: 1644/529, loss: 0.07012634724378586 2023-01-21 10:29:15.422800: step: 1648/529, loss: 0.009239577688276768 2023-01-21 10:29:16.711456: step: 1652/529, loss: 1.6439318656921387 2023-01-21 10:29:17.923659: step: 1656/529, loss: 0.07777824997901917 2023-01-21 10:29:19.094249: step: 1660/529, loss: 0.01565084606409073 2023-01-21 10:29:20.291435: step: 1664/529, loss: 0.0075900075025856495 2023-01-21 10:29:21.488940: step: 1668/529, loss: 0.028315259143710136 2023-01-21 10:29:22.717412: step: 1672/529, loss: 0.10052014142274857 2023-01-21 10:29:23.904041: step: 1676/529, loss: 0.0331057533621788 2023-01-21 10:29:25.094582: step: 1680/529, loss: 0.04570808634161949 2023-01-21 10:29:26.342925: step: 1684/529, loss: 0.09786806255578995 2023-01-21 10:29:27.487620: step: 1688/529, loss: 1.1208122968673706 2023-01-21 10:29:28.717696: step: 1692/529, loss: 0.012130022048950195 2023-01-21 10:29:29.928344: step: 1696/529, loss: 0.05638742446899414 2023-01-21 10:29:31.108981: step: 1700/529, loss: 0.03684711456298828 2023-01-21 10:29:32.306996: step: 1704/529, loss: 0.02162313461303711 2023-01-21 10:29:33.500852: step: 1708/529, loss: 0.07683763653039932 2023-01-21 10:29:34.708462: step: 1712/529, loss: 0.2494373321533203 2023-01-21 10:29:35.898014: step: 1716/529, loss: 0.13256961107254028 2023-01-21 10:29:37.124483: step: 1720/529, loss: 0.02953662909567356 2023-01-21 10:29:38.301048: step: 1724/529, loss: 0.09087281674146652 2023-01-21 10:29:39.521172: step: 1728/529, loss: 0.10782899707555771 2023-01-21 10:29:40.707229: step: 1732/529, loss: 0.006664848420768976 2023-01-21 10:29:41.904507: step: 1736/529, loss: 0.016600513830780983 2023-01-21 10:29:43.123145: step: 1740/529, loss: 0.15614071488380432 2023-01-21 10:29:44.315381: step: 1744/529, loss: 0.5339058637619019 2023-01-21 10:29:45.505109: step: 1748/529, loss: 0.15408745408058167 2023-01-21 10:29:46.739265: step: 1752/529, loss: 0.12845849990844727 2023-01-21 10:29:47.934328: step: 1756/529, loss: 0.040496826171875 2023-01-21 10:29:49.134412: step: 1760/529, loss: 0.04293489456176758 2023-01-21 10:29:50.346717: step: 1764/529, loss: 0.05974116548895836 2023-01-21 10:29:51.556653: step: 1768/529, loss: 0.1178036704659462 2023-01-21 10:29:52.778624: step: 1772/529, loss: 0.05413856357336044 2023-01-21 10:29:53.972000: step: 1776/529, loss: 0.013603448867797852 2023-01-21 10:29:55.198013: step: 1780/529, loss: 0.11620178818702698 2023-01-21 10:29:56.363955: step: 1784/529, loss: 0.04439087212085724 2023-01-21 10:29:57.533752: step: 1788/529, loss: 0.17102183401584625 2023-01-21 10:29:58.718374: step: 1792/529, loss: 0.00329933175817132 2023-01-21 10:29:59.931485: step: 1796/529, loss: 0.051810551434755325 2023-01-21 10:30:01.118211: step: 1800/529, loss: 0.1280687302350998 2023-01-21 10:30:02.301267: step: 1804/529, loss: 0.09831380844116211 2023-01-21 10:30:03.485587: step: 1808/529, loss: 0.03563718870282173 2023-01-21 10:30:04.686838: step: 1812/529, loss: 0.08889111876487732 2023-01-21 10:30:05.902517: step: 1816/529, loss: 0.22061900794506073 2023-01-21 10:30:07.067376: step: 1820/529, loss: 0.04422557353973389 2023-01-21 10:30:08.268887: step: 1824/529, loss: 0.0134124755859375 2023-01-21 10:30:09.437361: step: 1828/529, loss: 0.9438523650169373 2023-01-21 10:30:10.633312: step: 1832/529, loss: 0.030322838574647903 2023-01-21 10:30:11.821931: step: 1836/529, loss: 0.01393957156687975 2023-01-21 10:30:13.001826: step: 1840/529, loss: 0.8380979299545288 2023-01-21 10:30:14.193791: step: 1844/529, loss: 0.1119009256362915 2023-01-21 10:30:15.361666: step: 1848/529, loss: 0.10191259533166885 2023-01-21 10:30:16.586401: step: 1852/529, loss: 0.22992515563964844 2023-01-21 10:30:17.752737: step: 1856/529, loss: 0.03986644744873047 2023-01-21 10:30:18.934502: step: 1860/529, loss: 0.055548667907714844 2023-01-21 10:30:20.123071: step: 1864/529, loss: 0.10059471428394318 2023-01-21 10:30:21.316824: step: 1868/529, loss: 0.0764986127614975 2023-01-21 10:30:22.482407: step: 1872/529, loss: 0.11834988743066788 2023-01-21 10:30:23.657540: step: 1876/529, loss: 0.0778445228934288 2023-01-21 10:30:24.933263: step: 1880/529, loss: 0.09871644526720047 2023-01-21 10:30:26.100706: step: 1884/529, loss: 0.24607856571674347 2023-01-21 10:30:27.366030: step: 1888/529, loss: 0.09327040612697601 2023-01-21 10:30:28.546924: step: 1892/529, loss: 0.05929937586188316 2023-01-21 10:30:29.751975: step: 1896/529, loss: 0.15408927202224731 2023-01-21 10:30:30.960737: step: 1900/529, loss: 0.07512674480676651 2023-01-21 10:30:32.180343: step: 1904/529, loss: 0.12093200534582138 2023-01-21 10:30:33.341980: step: 1908/529, loss: 0.036788176745176315 2023-01-21 10:30:34.504512: step: 1912/529, loss: 0.03445596620440483 2023-01-21 10:30:35.685508: step: 1916/529, loss: 0.04038117080926895 2023-01-21 10:30:36.906138: step: 1920/529, loss: 0.04441061243414879 2023-01-21 10:30:38.067067: step: 1924/529, loss: 0.18578501045703888 2023-01-21 10:30:39.232034: step: 1928/529, loss: 0.023291492834687233 2023-01-21 10:30:40.436327: step: 1932/529, loss: 0.01905536837875843 2023-01-21 10:30:41.653829: step: 1936/529, loss: 0.07259535789489746 2023-01-21 10:30:42.849658: step: 1940/529, loss: 0.07072114944458008 2023-01-21 10:30:44.092698: step: 1944/529, loss: 0.12202244251966476 2023-01-21 10:30:45.299788: step: 1948/529, loss: 0.021721171215176582 2023-01-21 10:30:46.512567: step: 1952/529, loss: 0.021088410168886185 2023-01-21 10:30:47.660641: step: 1956/529, loss: 0.14635547995567322 2023-01-21 10:30:48.847043: step: 1960/529, loss: 0.021374035626649857 2023-01-21 10:30:50.036152: step: 1964/529, loss: 0.03525595739483833 2023-01-21 10:30:51.228071: step: 1968/529, loss: 0.1727035492658615 2023-01-21 10:30:52.446598: step: 1972/529, loss: 0.0465879961848259 2023-01-21 10:30:53.679506: step: 1976/529, loss: 0.027051258832216263 2023-01-21 10:30:54.879192: step: 1980/529, loss: 0.015650654211640358 2023-01-21 10:30:56.094847: step: 1984/529, loss: 0.028936101123690605 2023-01-21 10:30:57.343150: step: 1988/529, loss: 0.07564010471105576 2023-01-21 10:30:58.560382: step: 1992/529, loss: 0.017669105902314186 2023-01-21 10:30:59.750167: step: 1996/529, loss: 0.11853981018066406 2023-01-21 10:31:00.938272: step: 2000/529, loss: 0.006445837207138538 2023-01-21 10:31:02.120658: step: 2004/529, loss: 0.043023206293582916 2023-01-21 10:31:03.313731: step: 2008/529, loss: 0.0902436301112175 2023-01-21 10:31:04.528005: step: 2012/529, loss: 0.03726549446582794 2023-01-21 10:31:05.703819: step: 2016/529, loss: 0.0626346617937088 2023-01-21 10:31:06.944297: step: 2020/529, loss: 0.027577782049775124 2023-01-21 10:31:08.108494: step: 2024/529, loss: 0.13318586349487305 2023-01-21 10:31:09.326774: step: 2028/529, loss: 0.5146718621253967 2023-01-21 10:31:10.549063: step: 2032/529, loss: 0.12210503220558167 2023-01-21 10:31:11.762605: step: 2036/529, loss: 0.08855452388525009 2023-01-21 10:31:12.981055: step: 2040/529, loss: 0.05148293823003769 2023-01-21 10:31:14.196116: step: 2044/529, loss: 0.07322102040052414 2023-01-21 10:31:15.383680: step: 2048/529, loss: 0.07995501160621643 2023-01-21 10:31:16.585903: step: 2052/529, loss: 0.004712438676506281 2023-01-21 10:31:17.775155: step: 2056/529, loss: 0.22446632385253906 2023-01-21 10:31:18.987476: step: 2060/529, loss: 0.05641670525074005 2023-01-21 10:31:20.178843: step: 2064/529, loss: 0.0193037036806345 2023-01-21 10:31:21.345099: step: 2068/529, loss: 0.11494479328393936 2023-01-21 10:31:22.510371: step: 2072/529, loss: 0.02848968468606472 2023-01-21 10:31:23.715351: step: 2076/529, loss: 0.09234781563282013 2023-01-21 10:31:24.931256: step: 2080/529, loss: 0.32560861110687256 2023-01-21 10:31:26.152108: step: 2084/529, loss: 0.15645551681518555 2023-01-21 10:31:27.394454: step: 2088/529, loss: 0.035015396773815155 2023-01-21 10:31:28.584305: step: 2092/529, loss: 0.04155368730425835 2023-01-21 10:31:29.790784: step: 2096/529, loss: 0.17655391991138458 2023-01-21 10:31:31.005429: step: 2100/529, loss: 0.21046391129493713 2023-01-21 10:31:32.224864: step: 2104/529, loss: 0.006680441088974476 2023-01-21 10:31:33.406927: step: 2108/529, loss: 0.15230941772460938 2023-01-21 10:31:34.611773: step: 2112/529, loss: 0.040708351880311966 2023-01-21 10:31:35.864131: step: 2116/529, loss: 0.11756172776222229 ================================================== Loss: 0.121 -------------------- Dev: {'event': {'p': 0.5754245754245755, 'r': 0.7669773635153129, 'f1': 0.6575342465753425}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Test: {'event': {'p': 0.6193181818181818, 'r': 0.7837028160575195, 'f1': 0.6918804549061094}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Chinese: {'event': {'p': 0.5529411764705883, 'r': 0.8703703703703703, 'f1': 0.6762589928057553}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Korean: {'event': {'p': 0.5178571428571429, 'r': 0.4603174603174603, 'f1': 0.48739495798319327}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Russian: {'event': {'p': 0.47619047619047616, 'r': 0.5555555555555556, 'f1': 0.5128205128205129}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5877106045589693, 'r': 0.7896138482023968, 'f1': 0.6738636363636363}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6253574833174452, 'r': 0.7860994607549431, 'f1': 0.6965755242898859}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.6271186440677966, 'r': 0.5873015873015873, 'f1': 0.6065573770491803}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 10 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:32:21.994265: step: 4/529, loss: 0.13319548964500427 2023-01-21 10:32:23.170300: step: 8/529, loss: 0.018059730529785156 2023-01-21 10:32:24.386988: step: 12/529, loss: 0.06347141414880753 2023-01-21 10:32:25.577628: step: 16/529, loss: 0.7278034687042236 2023-01-21 10:32:26.770594: step: 20/529, loss: 0.03369617462158203 2023-01-21 10:32:27.956153: step: 24/529, loss: 0.004032040014863014 2023-01-21 10:32:29.148055: step: 28/529, loss: 0.021892739459872246 2023-01-21 10:32:30.374348: step: 32/529, loss: 0.548066258430481 2023-01-21 10:32:31.565823: step: 36/529, loss: 0.049533747136592865 2023-01-21 10:32:32.762417: step: 40/529, loss: 0.4965149760246277 2023-01-21 10:32:33.989447: step: 44/529, loss: 0.06300202012062073 2023-01-21 10:32:35.153389: step: 48/529, loss: 0.015262365341186523 2023-01-21 10:32:36.389801: step: 52/529, loss: 0.08438215404748917 2023-01-21 10:32:37.608329: step: 56/529, loss: 0.06253781914710999 2023-01-21 10:32:38.766155: step: 60/529, loss: 0.025278473272919655 2023-01-21 10:32:39.947836: step: 64/529, loss: 0.07637815922498703 2023-01-21 10:32:41.108879: step: 68/529, loss: 0.017382239922881126 2023-01-21 10:32:42.323816: step: 72/529, loss: 0.04782142862677574 2023-01-21 10:32:43.529247: step: 76/529, loss: 0.03657874837517738 2023-01-21 10:32:44.720039: step: 80/529, loss: 0.05125589668750763 2023-01-21 10:32:45.913406: step: 84/529, loss: 0.08427486568689346 2023-01-21 10:32:47.109242: step: 88/529, loss: 0.05002489313483238 2023-01-21 10:32:48.278051: step: 92/529, loss: 0.053534530103206635 2023-01-21 10:32:49.513163: step: 96/529, loss: 0.013772344216704369 2023-01-21 10:32:50.687006: step: 100/529, loss: 0.14294010400772095 2023-01-21 10:32:51.867627: step: 104/529, loss: 0.03716163709759712 2023-01-21 10:32:53.087654: step: 108/529, loss: 0.07859639823436737 2023-01-21 10:32:54.229927: step: 112/529, loss: 0.04225420951843262 2023-01-21 10:32:55.465135: step: 116/529, loss: 0.04295364022254944 2023-01-21 10:32:56.658696: step: 120/529, loss: 0.010546302422881126 2023-01-21 10:32:57.827390: step: 124/529, loss: 0.0484435111284256 2023-01-21 10:32:59.042516: step: 128/529, loss: 0.0320165641605854 2023-01-21 10:33:00.247497: step: 132/529, loss: 0.05277876928448677 2023-01-21 10:33:01.418991: step: 136/529, loss: 0.0038738250732421875 2023-01-21 10:33:02.633487: step: 140/529, loss: 1.21299147605896 2023-01-21 10:33:03.848126: step: 144/529, loss: 0.05450868606567383 2023-01-21 10:33:05.053034: step: 148/529, loss: 0.053360890597105026 2023-01-21 10:33:06.256235: step: 152/529, loss: 0.0466761589050293 2023-01-21 10:33:07.439231: step: 156/529, loss: 0.03779389336705208 2023-01-21 10:33:08.632770: step: 160/529, loss: 0.31127986311912537 2023-01-21 10:33:09.826722: step: 164/529, loss: 0.009652900509536266 2023-01-21 10:33:11.024171: step: 168/529, loss: 0.025745011866092682 2023-01-21 10:33:12.232158: step: 172/529, loss: 1.607777714729309 2023-01-21 10:33:13.387509: step: 176/529, loss: 0.033055976033210754 2023-01-21 10:33:14.599618: step: 180/529, loss: 0.03438835218548775 2023-01-21 10:33:15.786510: step: 184/529, loss: 0.03239021450281143 2023-01-21 10:33:16.993851: step: 188/529, loss: 0.0017548560863360763 2023-01-21 10:33:18.191090: step: 192/529, loss: 0.07959385216236115 2023-01-21 10:33:19.424890: step: 196/529, loss: 0.05726489797234535 2023-01-21 10:33:20.619886: step: 200/529, loss: 0.05039789527654648 2023-01-21 10:33:21.830107: step: 204/529, loss: 0.05500612035393715 2023-01-21 10:33:23.003787: step: 208/529, loss: 0.02331404760479927 2023-01-21 10:33:24.284131: step: 212/529, loss: 0.09467992931604385 2023-01-21 10:33:25.486182: step: 216/529, loss: 0.02184772491455078 2023-01-21 10:33:26.682447: step: 220/529, loss: 0.04513363912701607 2023-01-21 10:33:27.920147: step: 224/529, loss: 0.029505157843232155 2023-01-21 10:33:29.105578: step: 228/529, loss: 0.05676870420575142 2023-01-21 10:33:30.351321: step: 232/529, loss: 0.1100500151515007 2023-01-21 10:33:31.547196: step: 236/529, loss: 0.007414055056869984 2023-01-21 10:33:32.749102: step: 240/529, loss: 0.05604495853185654 2023-01-21 10:33:33.977264: step: 244/529, loss: 0.2492111325263977 2023-01-21 10:33:35.166886: step: 248/529, loss: 0.17275163531303406 2023-01-21 10:33:36.347692: step: 252/529, loss: 0.03466100990772247 2023-01-21 10:33:37.518373: step: 256/529, loss: 0.0047782897017896175 2023-01-21 10:33:38.711647: step: 260/529, loss: 0.0862194076180458 2023-01-21 10:33:39.871859: step: 264/529, loss: 0.015404606238007545 2023-01-21 10:33:41.055861: step: 268/529, loss: 0.012108184397220612 2023-01-21 10:33:42.246084: step: 272/529, loss: 0.002204513642936945 2023-01-21 10:33:43.413769: step: 276/529, loss: 0.10799828171730042 2023-01-21 10:33:44.613695: step: 280/529, loss: 0.18516984581947327 2023-01-21 10:33:45.822980: step: 284/529, loss: 0.037203602492809296 2023-01-21 10:33:47.029512: step: 288/529, loss: 0.04303140565752983 2023-01-21 10:33:48.267494: step: 292/529, loss: 0.13160762190818787 2023-01-21 10:33:49.449460: step: 296/529, loss: 0.01925215683877468 2023-01-21 10:33:50.650422: step: 300/529, loss: 0.0520690456032753 2023-01-21 10:33:51.826745: step: 304/529, loss: 0.025791360065340996 2023-01-21 10:33:53.082529: step: 308/529, loss: 0.1583041250705719 2023-01-21 10:33:54.279519: step: 312/529, loss: 0.043757058680057526 2023-01-21 10:33:55.462508: step: 316/529, loss: 0.09216461330652237 2023-01-21 10:33:56.651452: step: 320/529, loss: 0.04335489124059677 2023-01-21 10:33:57.880192: step: 324/529, loss: 0.08223362267017365 2023-01-21 10:33:59.088585: step: 328/529, loss: 0.0650109276175499 2023-01-21 10:34:00.287158: step: 332/529, loss: 0.011264611035585403 2023-01-21 10:34:01.509142: step: 336/529, loss: 0.14198246598243713 2023-01-21 10:34:02.734526: step: 340/529, loss: 0.03590850904583931 2023-01-21 10:34:03.920283: step: 344/529, loss: 0.028302384540438652 2023-01-21 10:34:05.105081: step: 348/529, loss: 0.053537748754024506 2023-01-21 10:34:06.300291: step: 352/529, loss: 0.033713921904563904 2023-01-21 10:34:07.480645: step: 356/529, loss: 0.11537113040685654 2023-01-21 10:34:08.693079: step: 360/529, loss: 0.042710211127996445 2023-01-21 10:34:09.867152: step: 364/529, loss: 0.0415978878736496 2023-01-21 10:34:11.036766: step: 368/529, loss: 0.01885662041604519 2023-01-21 10:34:12.273281: step: 372/529, loss: 0.008861875161528587 2023-01-21 10:34:13.463282: step: 376/529, loss: 0.057954125106334686 2023-01-21 10:34:14.677618: step: 380/529, loss: 0.07978086173534393 2023-01-21 10:34:15.855955: step: 384/529, loss: 0.40618085861206055 2023-01-21 10:34:17.047339: step: 388/529, loss: 0.011972714215517044 2023-01-21 10:34:18.240129: step: 392/529, loss: 0.039052676409482956 2023-01-21 10:34:19.402626: step: 396/529, loss: 0.12208671867847443 2023-01-21 10:34:20.573567: step: 400/529, loss: 0.007897424511611462 2023-01-21 10:34:21.772782: step: 404/529, loss: 0.08028239756822586 2023-01-21 10:34:22.974459: step: 408/529, loss: 0.035206567496061325 2023-01-21 10:34:24.239059: step: 412/529, loss: 0.06528787314891815 2023-01-21 10:34:25.406650: step: 416/529, loss: 0.0524505153298378 2023-01-21 10:34:26.593962: step: 420/529, loss: 0.05712347477674484 2023-01-21 10:34:27.789440: step: 424/529, loss: 0.040287233889102936 2023-01-21 10:34:29.004356: step: 428/529, loss: 0.10688810795545578 2023-01-21 10:34:30.204560: step: 432/529, loss: 0.0826304480433464 2023-01-21 10:34:31.427664: step: 436/529, loss: 0.029454803094267845 2023-01-21 10:34:32.685255: step: 440/529, loss: 0.06835965812206268 2023-01-21 10:34:33.924962: step: 444/529, loss: 0.11208248138427734 2023-01-21 10:34:35.112205: step: 448/529, loss: 0.06697436422109604 2023-01-21 10:34:36.377512: step: 452/529, loss: 0.08378639817237854 2023-01-21 10:34:37.562539: step: 456/529, loss: 0.07583846896886826 2023-01-21 10:34:38.781813: step: 460/529, loss: 0.07790136337280273 2023-01-21 10:34:39.991684: step: 464/529, loss: 0.02196331135928631 2023-01-21 10:34:41.175334: step: 468/529, loss: 0.05186839401721954 2023-01-21 10:34:42.398843: step: 472/529, loss: 0.025539018213748932 2023-01-21 10:34:43.613383: step: 476/529, loss: 0.10688219219446182 2023-01-21 10:34:44.840288: step: 480/529, loss: 0.12453468143939972 2023-01-21 10:34:46.097616: step: 484/529, loss: 0.05586185306310654 2023-01-21 10:34:47.259130: step: 488/529, loss: 0.76120525598526 2023-01-21 10:34:48.466277: step: 492/529, loss: 0.05771160498261452 2023-01-21 10:34:49.657988: step: 496/529, loss: 0.044410038739442825 2023-01-21 10:34:50.864539: step: 500/529, loss: 0.007841015234589577 2023-01-21 10:34:52.064754: step: 504/529, loss: 0.03519125282764435 2023-01-21 10:34:53.265011: step: 508/529, loss: 0.04109792783856392 2023-01-21 10:34:54.457362: step: 512/529, loss: 0.003284883452579379 2023-01-21 10:34:55.633268: step: 516/529, loss: 0.16335734724998474 2023-01-21 10:34:56.803015: step: 520/529, loss: 0.0568576343357563 2023-01-21 10:34:57.984724: step: 524/529, loss: 0.024648571386933327 2023-01-21 10:34:59.242027: step: 528/529, loss: 0.07842740416526794 2023-01-21 10:35:00.409307: step: 532/529, loss: 0.0037682533729821444 2023-01-21 10:35:01.568401: step: 536/529, loss: 0.012028366327285767 2023-01-21 10:35:02.739547: step: 540/529, loss: 0.023465871810913086 2023-01-21 10:35:03.950937: step: 544/529, loss: 0.04558572918176651 2023-01-21 10:35:05.186313: step: 548/529, loss: 0.4863826036453247 2023-01-21 10:35:06.364837: step: 552/529, loss: 0.1269817352294922 2023-01-21 10:35:07.556752: step: 556/529, loss: 0.0556950606405735 2023-01-21 10:35:08.756475: step: 560/529, loss: 0.026216890662908554 2023-01-21 10:35:09.949144: step: 564/529, loss: 0.011079978197813034 2023-01-21 10:35:11.168122: step: 568/529, loss: 0.0196835994720459 2023-01-21 10:35:12.366082: step: 572/529, loss: 0.01098709087818861 2023-01-21 10:35:13.608697: step: 576/529, loss: 0.2602923512458801 2023-01-21 10:35:14.848942: step: 580/529, loss: 0.02339344099164009 2023-01-21 10:35:16.019601: step: 584/529, loss: 0.043868161737918854 2023-01-21 10:35:17.195708: step: 588/529, loss: 0.060431480407714844 2023-01-21 10:35:18.370324: step: 592/529, loss: 0.1173524409532547 2023-01-21 10:35:19.595355: step: 596/529, loss: 0.38689398765563965 2023-01-21 10:35:20.782495: step: 600/529, loss: 0.47778576612472534 2023-01-21 10:35:21.931812: step: 604/529, loss: 0.005942249670624733 2023-01-21 10:35:23.120781: step: 608/529, loss: 0.07339668273925781 2023-01-21 10:35:24.380090: step: 612/529, loss: 0.18461008369922638 2023-01-21 10:35:25.587012: step: 616/529, loss: 0.03339235857129097 2023-01-21 10:35:26.771254: step: 620/529, loss: 0.028765009716153145 2023-01-21 10:35:27.986319: step: 624/529, loss: 0.0717369094491005 2023-01-21 10:35:29.184213: step: 628/529, loss: 0.035436443984508514 2023-01-21 10:35:30.377932: step: 632/529, loss: 0.30106741189956665 2023-01-21 10:35:31.597100: step: 636/529, loss: 0.00485916156321764 2023-01-21 10:35:32.799892: step: 640/529, loss: 0.08416728675365448 2023-01-21 10:35:34.040950: step: 644/529, loss: 0.04262995719909668 2023-01-21 10:35:35.220168: step: 648/529, loss: 0.00021152498084120452 2023-01-21 10:35:36.435543: step: 652/529, loss: 0.11014995723962784 2023-01-21 10:35:37.647184: step: 656/529, loss: 0.1337263137102127 2023-01-21 10:35:38.826492: step: 660/529, loss: 0.020613862201571465 2023-01-21 10:35:40.014287: step: 664/529, loss: 0.3737545907497406 2023-01-21 10:35:41.208882: step: 668/529, loss: 0.03664112091064453 2023-01-21 10:35:42.446930: step: 672/529, loss: 0.030140304937958717 2023-01-21 10:35:43.620128: step: 676/529, loss: 0.025378037244081497 2023-01-21 10:35:44.842902: step: 680/529, loss: 0.016735846176743507 2023-01-21 10:35:46.064322: step: 684/529, loss: 0.2162659764289856 2023-01-21 10:35:47.303331: step: 688/529, loss: 0.11829642951488495 2023-01-21 10:35:48.537713: step: 692/529, loss: 0.020205307751893997 2023-01-21 10:35:49.753161: step: 696/529, loss: 0.06990861892700195 2023-01-21 10:35:50.955723: step: 700/529, loss: 0.05043850094079971 2023-01-21 10:35:52.141868: step: 704/529, loss: 0.0879116803407669 2023-01-21 10:35:53.318187: step: 708/529, loss: 0.11114483326673508 2023-01-21 10:35:54.515985: step: 712/529, loss: 0.050248242914676666 2023-01-21 10:35:55.788481: step: 716/529, loss: 0.016098951920866966 2023-01-21 10:35:56.990043: step: 720/529, loss: 0.08850259333848953 2023-01-21 10:35:58.151757: step: 724/529, loss: 0.8642034530639648 2023-01-21 10:35:59.353656: step: 728/529, loss: 0.22455640137195587 2023-01-21 10:36:00.548100: step: 732/529, loss: 0.02417612075805664 2023-01-21 10:36:01.786284: step: 736/529, loss: 0.3426995277404785 2023-01-21 10:36:02.974817: step: 740/529, loss: 0.11350803077220917 2023-01-21 10:36:04.139680: step: 744/529, loss: 0.14121656119823456 2023-01-21 10:36:05.323560: step: 748/529, loss: 0.06158151477575302 2023-01-21 10:36:06.513214: step: 752/529, loss: 0.010848474688827991 2023-01-21 10:36:07.698329: step: 756/529, loss: 0.06322264671325684 2023-01-21 10:36:08.866601: step: 760/529, loss: 0.02086639404296875 2023-01-21 10:36:10.113143: step: 764/529, loss: 0.044230081140995026 2023-01-21 10:36:11.307548: step: 768/529, loss: 0.14699527621269226 2023-01-21 10:36:12.467299: step: 772/529, loss: 0.04583549126982689 2023-01-21 10:36:13.654037: step: 776/529, loss: 0.0786447525024414 2023-01-21 10:36:14.865303: step: 780/529, loss: 0.059853557497262955 2023-01-21 10:36:16.050500: step: 784/529, loss: 0.10635185241699219 2023-01-21 10:36:17.269581: step: 788/529, loss: 0.10776054859161377 2023-01-21 10:36:18.472122: step: 792/529, loss: 0.1211184486746788 2023-01-21 10:36:19.683240: step: 796/529, loss: 0.057624056935310364 2023-01-21 10:36:20.903898: step: 800/529, loss: 0.14387273788452148 2023-01-21 10:36:22.076745: step: 804/529, loss: 0.04469814524054527 2023-01-21 10:36:23.343321: step: 808/529, loss: 0.46169471740722656 2023-01-21 10:36:24.566483: step: 812/529, loss: 0.05056142807006836 2023-01-21 10:36:25.739552: step: 816/529, loss: 0.9103809595108032 2023-01-21 10:36:26.942691: step: 820/529, loss: 0.06847725063562393 2023-01-21 10:36:28.119827: step: 824/529, loss: 0.03443317487835884 2023-01-21 10:36:29.304315: step: 828/529, loss: 0.006948280148208141 2023-01-21 10:36:30.530033: step: 832/529, loss: 0.017670728266239166 2023-01-21 10:36:31.734702: step: 836/529, loss: 0.06516246497631073 2023-01-21 10:36:32.923788: step: 840/529, loss: 0.07942891120910645 2023-01-21 10:36:34.087189: step: 844/529, loss: 0.02394256554543972 2023-01-21 10:36:35.308349: step: 848/529, loss: 0.015009118244051933 2023-01-21 10:36:36.498407: step: 852/529, loss: 0.015423298813402653 2023-01-21 10:36:37.730291: step: 856/529, loss: 0.06465931236743927 2023-01-21 10:36:38.941020: step: 860/529, loss: 0.08004292845726013 2023-01-21 10:36:40.158493: step: 864/529, loss: 0.04202743619680405 2023-01-21 10:36:41.321526: step: 868/529, loss: 0.06557998806238174 2023-01-21 10:36:42.523501: step: 872/529, loss: 0.07551483809947968 2023-01-21 10:36:43.716024: step: 876/529, loss: 0.0211213119328022 2023-01-21 10:36:44.914106: step: 880/529, loss: 0.06236915662884712 2023-01-21 10:36:46.106214: step: 884/529, loss: 0.14247122406959534 2023-01-21 10:36:47.275376: step: 888/529, loss: 0.019756030291318893 2023-01-21 10:36:48.480539: step: 892/529, loss: 0.06666717678308487 2023-01-21 10:36:49.665757: step: 896/529, loss: 0.008554745465517044 2023-01-21 10:36:50.880471: step: 900/529, loss: 0.053200624883174896 2023-01-21 10:36:52.077599: step: 904/529, loss: 0.37125349044799805 2023-01-21 10:36:53.269982: step: 908/529, loss: 0.040814876556396484 2023-01-21 10:36:54.483524: step: 912/529, loss: 0.014794540591537952 2023-01-21 10:36:55.685035: step: 916/529, loss: 0.029592324048280716 2023-01-21 10:36:56.860982: step: 920/529, loss: 0.121874138712883 2023-01-21 10:36:58.100319: step: 924/529, loss: 0.019934751093387604 2023-01-21 10:36:59.384228: step: 928/529, loss: 0.05108080059289932 2023-01-21 10:37:00.576219: step: 932/529, loss: 0.04032840579748154 2023-01-21 10:37:01.822849: step: 936/529, loss: 0.08637695014476776 2023-01-21 10:37:03.058407: step: 940/529, loss: 0.064073845744133 2023-01-21 10:37:04.243643: step: 944/529, loss: 0.029075955972075462 2023-01-21 10:37:05.428510: step: 948/529, loss: 0.0765417143702507 2023-01-21 10:37:06.619358: step: 952/529, loss: 0.04902248457074165 2023-01-21 10:37:07.808013: step: 956/529, loss: 0.03183465078473091 2023-01-21 10:37:09.041456: step: 960/529, loss: 0.06572423130273819 2023-01-21 10:37:10.231394: step: 964/529, loss: 0.009702682495117188 2023-01-21 10:37:11.389431: step: 968/529, loss: 0.018375016748905182 2023-01-21 10:37:12.585489: step: 972/529, loss: 0.12438878417015076 2023-01-21 10:37:13.788166: step: 976/529, loss: 0.03125810623168945 2023-01-21 10:37:14.972283: step: 980/529, loss: 0.04159488528966904 2023-01-21 10:37:16.189200: step: 984/529, loss: 0.12201671302318573 2023-01-21 10:37:17.373114: step: 988/529, loss: 0.018727876245975494 2023-01-21 10:37:18.543103: step: 992/529, loss: 0.012703276239335537 2023-01-21 10:37:19.752450: step: 996/529, loss: 0.04838228598237038 2023-01-21 10:37:20.945016: step: 1000/529, loss: 0.023961449041962624 2023-01-21 10:37:22.136392: step: 1004/529, loss: 0.09158802777528763 2023-01-21 10:37:23.328509: step: 1008/529, loss: 0.04957609251141548 2023-01-21 10:37:24.504854: step: 1012/529, loss: 0.021778298541903496 2023-01-21 10:37:25.706120: step: 1016/529, loss: 0.013942575082182884 2023-01-21 10:37:26.876378: step: 1020/529, loss: 0.285350501537323 2023-01-21 10:37:28.063000: step: 1024/529, loss: 0.1779472380876541 2023-01-21 10:37:29.260817: step: 1028/529, loss: 0.06171026453375816 2023-01-21 10:37:30.445816: step: 1032/529, loss: 0.09072227776050568 2023-01-21 10:37:31.655415: step: 1036/529, loss: 0.3835482597351074 2023-01-21 10:37:32.853118: step: 1040/529, loss: 0.06224212795495987 2023-01-21 10:37:34.008216: step: 1044/529, loss: 0.06522088497877121 2023-01-21 10:37:35.216667: step: 1048/529, loss: 0.1870567500591278 2023-01-21 10:37:36.448467: step: 1052/529, loss: 0.05466442182660103 2023-01-21 10:37:37.606342: step: 1056/529, loss: 0.16986331343650818 2023-01-21 10:37:38.794634: step: 1060/529, loss: 0.002873706864193082 2023-01-21 10:37:40.015149: step: 1064/529, loss: 0.05595429241657257 2023-01-21 10:37:41.242714: step: 1068/529, loss: 0.1596684455871582 2023-01-21 10:37:42.471742: step: 1072/529, loss: 0.04306383430957794 2023-01-21 10:37:43.637564: step: 1076/529, loss: 0.050516318529844284 2023-01-21 10:37:44.844583: step: 1080/529, loss: 0.060624219477176666 2023-01-21 10:37:46.027351: step: 1084/529, loss: 0.04683075100183487 2023-01-21 10:37:47.229223: step: 1088/529, loss: 0.15315961837768555 2023-01-21 10:37:48.423134: step: 1092/529, loss: 0.0033500194549560547 2023-01-21 10:37:49.612591: step: 1096/529, loss: 0.0043374537490308285 2023-01-21 10:37:50.792699: step: 1100/529, loss: 0.05547018349170685 2023-01-21 10:37:51.981610: step: 1104/529, loss: 0.06650204211473465 2023-01-21 10:37:53.175948: step: 1108/529, loss: 0.3064367175102234 2023-01-21 10:37:54.378028: step: 1112/529, loss: 0.04076967388391495 2023-01-21 10:37:55.586903: step: 1116/529, loss: 0.045052669942379 2023-01-21 10:37:56.767825: step: 1120/529, loss: 0.09005539864301682 2023-01-21 10:37:57.933830: step: 1124/529, loss: 0.5244442224502563 2023-01-21 10:37:59.139780: step: 1128/529, loss: 0.08465595543384552 2023-01-21 10:38:00.330422: step: 1132/529, loss: 0.024016762152314186 2023-01-21 10:38:01.506961: step: 1136/529, loss: 0.03293666988611221 2023-01-21 10:38:02.696939: step: 1140/529, loss: 0.07514485716819763 2023-01-21 10:38:03.906573: step: 1144/529, loss: 0.016962332651019096 2023-01-21 10:38:05.107146: step: 1148/529, loss: 0.045172788202762604 2023-01-21 10:38:06.325724: step: 1152/529, loss: 0.2070116102695465 2023-01-21 10:38:07.489521: step: 1156/529, loss: 0.04378795623779297 2023-01-21 10:38:08.696740: step: 1160/529, loss: 0.03152036666870117 2023-01-21 10:38:09.872941: step: 1164/529, loss: 0.02256651036441326 2023-01-21 10:38:11.082562: step: 1168/529, loss: 0.032332513481378555 2023-01-21 10:38:12.283851: step: 1172/529, loss: 0.04702835530042648 2023-01-21 10:38:13.483550: step: 1176/529, loss: 0.10719041526317596 2023-01-21 10:38:14.685537: step: 1180/529, loss: 0.11200113594532013 2023-01-21 10:38:15.928604: step: 1184/529, loss: 0.11812859028577805 2023-01-21 10:38:17.104212: step: 1188/529, loss: 0.004242801573127508 2023-01-21 10:38:18.310530: step: 1192/529, loss: 0.034917548298835754 2023-01-21 10:38:19.505149: step: 1196/529, loss: 0.18195569515228271 2023-01-21 10:38:20.726657: step: 1200/529, loss: 0.5743680000305176 2023-01-21 10:38:21.949565: step: 1204/529, loss: 0.0611509308218956 2023-01-21 10:38:23.109172: step: 1208/529, loss: 0.021572113037109375 2023-01-21 10:38:24.304222: step: 1212/529, loss: 0.00331878662109375 2023-01-21 10:38:25.500871: step: 1216/529, loss: 0.005379676818847656 2023-01-21 10:38:26.758826: step: 1220/529, loss: 0.05204496532678604 2023-01-21 10:38:27.956260: step: 1224/529, loss: 0.11636243015527725 2023-01-21 10:38:29.156122: step: 1228/529, loss: 0.07349061965942383 2023-01-21 10:38:30.385704: step: 1232/529, loss: 0.03672294691205025 2023-01-21 10:38:31.569988: step: 1236/529, loss: 0.04184398800134659 2023-01-21 10:38:32.776426: step: 1240/529, loss: 0.07160405814647675 2023-01-21 10:38:33.972720: step: 1244/529, loss: 0.04400477558374405 2023-01-21 10:38:35.158465: step: 1248/529, loss: 0.03861656412482262 2023-01-21 10:38:36.372881: step: 1252/529, loss: 0.08048562705516815 2023-01-21 10:38:37.533009: step: 1256/529, loss: 0.0093552116304636 2023-01-21 10:38:38.716716: step: 1260/529, loss: 0.07427319884300232 2023-01-21 10:38:39.899770: step: 1264/529, loss: 0.0884058028459549 2023-01-21 10:38:41.065286: step: 1268/529, loss: 0.10429945588111877 2023-01-21 10:38:42.227558: step: 1272/529, loss: 0.01878499984741211 2023-01-21 10:38:43.421403: step: 1276/529, loss: 0.005427074618637562 2023-01-21 10:38:44.601475: step: 1280/529, loss: 0.06605371832847595 2023-01-21 10:38:45.816532: step: 1284/529, loss: 0.009358787909150124 2023-01-21 10:38:47.030726: step: 1288/529, loss: 0.1979314684867859 2023-01-21 10:38:48.250271: step: 1292/529, loss: 0.0626378059387207 2023-01-21 10:38:49.450814: step: 1296/529, loss: 0.00027866364689543843 2023-01-21 10:38:50.690244: step: 1300/529, loss: 0.01588420942425728 2023-01-21 10:38:51.948090: step: 1304/529, loss: 0.011727476492524147 2023-01-21 10:38:53.185419: step: 1308/529, loss: 0.022165682166814804 2023-01-21 10:38:54.361684: step: 1312/529, loss: 0.47613105177879333 2023-01-21 10:38:55.616899: step: 1316/529, loss: 0.042531583458185196 2023-01-21 10:38:56.799264: step: 1320/529, loss: 0.24019335210323334 2023-01-21 10:38:57.996931: step: 1324/529, loss: 0.04384305700659752 2023-01-21 10:38:59.244738: step: 1328/529, loss: 0.3339088559150696 2023-01-21 10:39:00.425300: step: 1332/529, loss: 0.6282937526702881 2023-01-21 10:39:01.595674: step: 1336/529, loss: 0.07174015045166016 2023-01-21 10:39:02.848749: step: 1340/529, loss: 0.08688221126794815 2023-01-21 10:39:04.058019: step: 1344/529, loss: 0.007117795757949352 2023-01-21 10:39:05.243263: step: 1348/529, loss: 0.06505441665649414 2023-01-21 10:39:06.422900: step: 1352/529, loss: 0.09686699509620667 2023-01-21 10:39:07.627905: step: 1356/529, loss: 0.030857086181640625 2023-01-21 10:39:08.837619: step: 1360/529, loss: 0.013363170437514782 2023-01-21 10:39:10.004162: step: 1364/529, loss: 0.002827262971550226 2023-01-21 10:39:11.145593: step: 1368/529, loss: 0.09202156215906143 2023-01-21 10:39:12.316530: step: 1372/529, loss: 0.06125626713037491 2023-01-21 10:39:13.487550: step: 1376/529, loss: 0.19162344932556152 2023-01-21 10:39:14.694849: step: 1380/529, loss: 0.05741758272051811 2023-01-21 10:39:15.869188: step: 1384/529, loss: 0.0636628195643425 2023-01-21 10:39:17.087133: step: 1388/529, loss: 0.16150379180908203 2023-01-21 10:39:18.312708: step: 1392/529, loss: 0.06197939068078995 2023-01-21 10:39:19.512099: step: 1396/529, loss: 0.09217548370361328 2023-01-21 10:39:20.698825: step: 1400/529, loss: 0.12160249054431915 2023-01-21 10:39:21.875859: step: 1404/529, loss: 0.09355239570140839 2023-01-21 10:39:23.047384: step: 1408/529, loss: 0.0180224422365427 2023-01-21 10:39:24.235925: step: 1412/529, loss: 0.03396472707390785 2023-01-21 10:39:25.413053: step: 1416/529, loss: 0.010427569970488548 2023-01-21 10:39:26.626148: step: 1420/529, loss: 0.03540525585412979 2023-01-21 10:39:27.825673: step: 1424/529, loss: 0.06806745380163193 2023-01-21 10:39:29.028270: step: 1428/529, loss: 0.019762182608246803 2023-01-21 10:39:30.216081: step: 1432/529, loss: 0.018743038177490234 2023-01-21 10:39:31.410811: step: 1436/529, loss: 0.09737863391637802 2023-01-21 10:39:32.681570: step: 1440/529, loss: 0.024572182446718216 2023-01-21 10:39:33.849394: step: 1444/529, loss: 0.04427499696612358 2023-01-21 10:39:35.061743: step: 1448/529, loss: 0.010946846567094326 2023-01-21 10:39:36.271825: step: 1452/529, loss: 0.059969618916511536 2023-01-21 10:39:37.531657: step: 1456/529, loss: 0.017001723870635033 2023-01-21 10:39:38.718111: step: 1460/529, loss: 0.6445391178131104 2023-01-21 10:39:39.957423: step: 1464/529, loss: 0.04973822087049484 2023-01-21 10:39:41.165020: step: 1468/529, loss: 0.010976076126098633 2023-01-21 10:39:42.333329: step: 1472/529, loss: 0.00947415828704834 2023-01-21 10:39:43.541610: step: 1476/529, loss: 0.05750136822462082 2023-01-21 10:39:44.764794: step: 1480/529, loss: 0.11432208865880966 2023-01-21 10:39:45.928289: step: 1484/529, loss: 0.02811429463326931 2023-01-21 10:39:47.221065: step: 1488/529, loss: 0.2895934283733368 2023-01-21 10:39:48.428844: step: 1492/529, loss: 0.03552551567554474 2023-01-21 10:39:49.671468: step: 1496/529, loss: 0.1318141520023346 2023-01-21 10:39:50.825354: step: 1500/529, loss: 0.13489246368408203 2023-01-21 10:39:52.026986: step: 1504/529, loss: 0.031130218878388405 2023-01-21 10:39:53.224713: step: 1508/529, loss: 0.13389912247657776 2023-01-21 10:39:54.392937: step: 1512/529, loss: 0.07240858674049377 2023-01-21 10:39:55.595681: step: 1516/529, loss: 0.027633430436253548 2023-01-21 10:39:56.810719: step: 1520/529, loss: 0.2103595733642578 2023-01-21 10:39:58.061075: step: 1524/529, loss: 0.011047554202377796 2023-01-21 10:39:59.266452: step: 1528/529, loss: 0.051865674555301666 2023-01-21 10:40:00.478708: step: 1532/529, loss: 0.08383378386497498 2023-01-21 10:40:01.689404: step: 1536/529, loss: 0.16126574575901031 2023-01-21 10:40:02.872995: step: 1540/529, loss: 0.04711761325597763 2023-01-21 10:40:04.087442: step: 1544/529, loss: 0.1483105719089508 2023-01-21 10:40:05.259857: step: 1548/529, loss: 0.09462644159793854 2023-01-21 10:40:06.441677: step: 1552/529, loss: 0.01610751263797283 2023-01-21 10:40:07.646551: step: 1556/529, loss: 0.047036781907081604 2023-01-21 10:40:08.806758: step: 1560/529, loss: 0.035237133502960205 2023-01-21 10:40:10.019564: step: 1564/529, loss: 0.08369751274585724 2023-01-21 10:40:11.218485: step: 1568/529, loss: 0.019838809967041016 2023-01-21 10:40:12.455494: step: 1572/529, loss: 0.06706152111291885 2023-01-21 10:40:13.635216: step: 1576/529, loss: 0.009032058529555798 2023-01-21 10:40:14.826802: step: 1580/529, loss: 0.04050722345709801 2023-01-21 10:40:16.065994: step: 1584/529, loss: 0.027249623090028763 2023-01-21 10:40:17.266487: step: 1588/529, loss: 0.05882978439331055 2023-01-21 10:40:18.481872: step: 1592/529, loss: 0.005846119020134211 2023-01-21 10:40:19.674949: step: 1596/529, loss: 0.028565408661961555 2023-01-21 10:40:20.885256: step: 1600/529, loss: 0.008575011044740677 2023-01-21 10:40:22.075075: step: 1604/529, loss: 0.03824224695563316 2023-01-21 10:40:23.233158: step: 1608/529, loss: 0.05434904247522354 2023-01-21 10:40:24.470540: step: 1612/529, loss: 0.03480052947998047 2023-01-21 10:40:25.654136: step: 1616/529, loss: 0.08869180828332901 2023-01-21 10:40:26.896610: step: 1620/529, loss: 0.04703493416309357 2023-01-21 10:40:28.132854: step: 1624/529, loss: 0.04728660732507706 2023-01-21 10:40:29.332499: step: 1628/529, loss: 0.25898608565330505 2023-01-21 10:40:30.524230: step: 1632/529, loss: 0.07266082614660263 2023-01-21 10:40:31.660835: step: 1636/529, loss: 0.006017875857651234 2023-01-21 10:40:32.857816: step: 1640/529, loss: 0.005083751864731312 2023-01-21 10:40:34.053029: step: 1644/529, loss: 0.06535936146974564 2023-01-21 10:40:35.230709: step: 1648/529, loss: 0.026714229956269264 2023-01-21 10:40:36.494960: step: 1652/529, loss: 0.014203835278749466 2023-01-21 10:40:37.732733: step: 1656/529, loss: 0.17198285460472107 2023-01-21 10:40:38.897532: step: 1660/529, loss: 0.6538625955581665 2023-01-21 10:40:40.106667: step: 1664/529, loss: 0.0641627311706543 2023-01-21 10:40:41.317525: step: 1668/529, loss: 0.012610369361937046 2023-01-21 10:40:42.528626: step: 1672/529, loss: 0.0954427719116211 2023-01-21 10:40:43.693765: step: 1676/529, loss: 0.07017917931079865 2023-01-21 10:40:44.912706: step: 1680/529, loss: 0.07186637073755264 2023-01-21 10:40:46.078928: step: 1684/529, loss: 0.09809704124927521 2023-01-21 10:40:47.252009: step: 1688/529, loss: 0.021207522600889206 2023-01-21 10:40:48.441994: step: 1692/529, loss: 0.008821630850434303 2023-01-21 10:40:49.629380: step: 1696/529, loss: 0.18709488213062286 2023-01-21 10:40:50.858929: step: 1700/529, loss: 0.09484080970287323 2023-01-21 10:40:52.024125: step: 1704/529, loss: 0.09167365729808807 2023-01-21 10:40:53.245920: step: 1708/529, loss: 0.0692354217171669 2023-01-21 10:40:54.455826: step: 1712/529, loss: 0.09218921512365341 2023-01-21 10:40:55.597321: step: 1716/529, loss: 0.11082544922828674 2023-01-21 10:40:56.781587: step: 1720/529, loss: 0.4856891632080078 2023-01-21 10:40:58.022431: step: 1724/529, loss: 0.173588365316391 2023-01-21 10:40:59.202352: step: 1728/529, loss: 0.09916222095489502 2023-01-21 10:41:00.422718: step: 1732/529, loss: 0.33995094895362854 2023-01-21 10:41:01.617240: step: 1736/529, loss: 0.019084453582763672 2023-01-21 10:41:02.805794: step: 1740/529, loss: 0.10552337765693665 2023-01-21 10:41:03.986698: step: 1744/529, loss: 0.0319819450378418 2023-01-21 10:41:05.199540: step: 1748/529, loss: 0.07181663811206818 2023-01-21 10:41:06.407997: step: 1752/529, loss: 0.05091724544763565 2023-01-21 10:41:07.597774: step: 1756/529, loss: 0.04818334802985191 2023-01-21 10:41:08.767658: step: 1760/529, loss: 0.05908724293112755 2023-01-21 10:41:09.949293: step: 1764/529, loss: 0.08158192783594131 2023-01-21 10:41:11.169074: step: 1768/529, loss: 0.16746819019317627 2023-01-21 10:41:12.370095: step: 1772/529, loss: 0.005501556675881147 2023-01-21 10:41:13.600667: step: 1776/529, loss: 0.1229754388332367 2023-01-21 10:41:14.775348: step: 1780/529, loss: 0.04066481441259384 2023-01-21 10:41:15.985661: step: 1784/529, loss: 0.2833024859428406 2023-01-21 10:41:17.180774: step: 1788/529, loss: 0.05566740036010742 2023-01-21 10:41:18.369531: step: 1792/529, loss: 0.0060708047822117805 2023-01-21 10:41:19.579245: step: 1796/529, loss: 0.07069225609302521 2023-01-21 10:41:20.774976: step: 1800/529, loss: 0.038553811609745026 2023-01-21 10:41:22.050374: step: 1804/529, loss: 0.08105144649744034 2023-01-21 10:41:23.227154: step: 1808/529, loss: 0.040831662714481354 2023-01-21 10:41:24.397828: step: 1812/529, loss: 0.012246990576386452 2023-01-21 10:41:25.572020: step: 1816/529, loss: 0.005950260441750288 2023-01-21 10:41:26.779082: step: 1820/529, loss: 0.20272284746170044 2023-01-21 10:41:28.013308: step: 1824/529, loss: 0.04891033470630646 2023-01-21 10:41:29.223674: step: 1828/529, loss: 0.1917746514081955 2023-01-21 10:41:30.444969: step: 1832/529, loss: 0.23157139122486115 2023-01-21 10:41:31.667631: step: 1836/529, loss: 0.017279338091611862 2023-01-21 10:41:32.848185: step: 1840/529, loss: 0.006729698274284601 2023-01-21 10:41:34.062144: step: 1844/529, loss: 0.021752070635557175 2023-01-21 10:41:35.278623: step: 1848/529, loss: 0.05968036875128746 2023-01-21 10:41:36.446425: step: 1852/529, loss: 0.020246410742402077 2023-01-21 10:41:37.647388: step: 1856/529, loss: 0.02618122100830078 2023-01-21 10:41:38.819262: step: 1860/529, loss: 0.020935581997036934 2023-01-21 10:41:40.018427: step: 1864/529, loss: 0.2516254782676697 2023-01-21 10:41:41.279380: step: 1868/529, loss: 0.0006624698871746659 2023-01-21 10:41:42.489521: step: 1872/529, loss: 0.1720069795846939 2023-01-21 10:41:43.666752: step: 1876/529, loss: 0.006412982940673828 2023-01-21 10:41:44.875130: step: 1880/529, loss: 0.09994864463806152 2023-01-21 10:41:46.070201: step: 1884/529, loss: 0.6097953915596008 2023-01-21 10:41:47.291626: step: 1888/529, loss: 0.11386509239673615 2023-01-21 10:41:48.475940: step: 1892/529, loss: 0.036307334899902344 2023-01-21 10:41:49.709026: step: 1896/529, loss: 0.07576818019151688 2023-01-21 10:41:50.899891: step: 1900/529, loss: 0.09280133992433548 2023-01-21 10:41:52.078240: step: 1904/529, loss: 0.04515876621007919 2023-01-21 10:41:53.212994: step: 1908/529, loss: 0.032294463366270065 2023-01-21 10:41:54.411977: step: 1912/529, loss: 0.010451888665556908 2023-01-21 10:41:55.593773: step: 1916/529, loss: 0.19697609543800354 2023-01-21 10:41:56.730255: step: 1920/529, loss: 0.007030535023659468 2023-01-21 10:41:57.888153: step: 1924/529, loss: 0.078119657933712 2023-01-21 10:41:59.088354: step: 1928/529, loss: 0.054036520421504974 2023-01-21 10:42:00.296891: step: 1932/529, loss: 0.031043052673339844 2023-01-21 10:42:01.496697: step: 1936/529, loss: 0.04542751610279083 2023-01-21 10:42:02.716062: step: 1940/529, loss: 0.06074027717113495 2023-01-21 10:42:03.914717: step: 1944/529, loss: 0.04008760303258896 2023-01-21 10:42:05.111938: step: 1948/529, loss: 0.02817593887448311 2023-01-21 10:42:06.293363: step: 1952/529, loss: 0.03502235561609268 2023-01-21 10:42:07.473643: step: 1956/529, loss: 0.17737768590450287 2023-01-21 10:42:08.649706: step: 1960/529, loss: 0.05894641950726509 2023-01-21 10:42:09.858822: step: 1964/529, loss: 0.06258449703454971 2023-01-21 10:42:11.085270: step: 1968/529, loss: 0.09734315425157547 2023-01-21 10:42:12.317188: step: 1972/529, loss: 0.4017411172389984 2023-01-21 10:42:13.522560: step: 1976/529, loss: 0.06903314590454102 2023-01-21 10:42:14.717885: step: 1980/529, loss: 0.07976827770471573 2023-01-21 10:42:15.910093: step: 1984/529, loss: 0.05909710004925728 2023-01-21 10:42:17.100190: step: 1988/529, loss: 0.09733323752880096 2023-01-21 10:42:18.294227: step: 1992/529, loss: 0.0556214302778244 2023-01-21 10:42:19.531824: step: 1996/529, loss: 0.16538219153881073 2023-01-21 10:42:20.712258: step: 2000/529, loss: 0.10899905860424042 2023-01-21 10:42:21.900910: step: 2004/529, loss: 0.029842281714081764 2023-01-21 10:42:23.083269: step: 2008/529, loss: 0.042408086359500885 2023-01-21 10:42:24.315157: step: 2012/529, loss: 0.03325076401233673 2023-01-21 10:42:25.505638: step: 2016/529, loss: 0.026003647595643997 2023-01-21 10:42:26.735327: step: 2020/529, loss: 0.035254668444395065 2023-01-21 10:42:27.950561: step: 2024/529, loss: 0.04162921756505966 2023-01-21 10:42:29.140660: step: 2028/529, loss: 0.03584732860326767 2023-01-21 10:42:30.338958: step: 2032/529, loss: 0.024262525141239166 2023-01-21 10:42:31.517981: step: 2036/529, loss: 0.11120691895484924 2023-01-21 10:42:32.741508: step: 2040/529, loss: 0.04747423902153969 2023-01-21 10:42:33.958076: step: 2044/529, loss: 0.026966571807861328 2023-01-21 10:42:35.184920: step: 2048/529, loss: 0.13409996032714844 2023-01-21 10:42:36.376487: step: 2052/529, loss: 0.24730148911476135 2023-01-21 10:42:37.533217: step: 2056/529, loss: 0.08357515186071396 2023-01-21 10:42:38.733534: step: 2060/529, loss: 0.0038730620872229338 2023-01-21 10:42:39.919407: step: 2064/529, loss: 0.008508061990141869 2023-01-21 10:42:41.136949: step: 2068/529, loss: 0.05024733766913414 2023-01-21 10:42:42.324311: step: 2072/529, loss: 0.09975495934486389 2023-01-21 10:42:43.522559: step: 2076/529, loss: 0.48831596970558167 2023-01-21 10:42:44.760922: step: 2080/529, loss: 0.01406846009194851 2023-01-21 10:42:45.961870: step: 2084/529, loss: 0.8734437227249146 2023-01-21 10:42:47.160150: step: 2088/529, loss: 0.017843054607510567 2023-01-21 10:42:48.347234: step: 2092/529, loss: 0.1170232743024826 2023-01-21 10:42:49.528826: step: 2096/529, loss: 0.013381004333496094 2023-01-21 10:42:50.799705: step: 2100/529, loss: 0.05179738998413086 2023-01-21 10:42:52.024186: step: 2104/529, loss: 0.027214527130126953 2023-01-21 10:42:53.215376: step: 2108/529, loss: 0.11447896808385849 2023-01-21 10:42:54.407484: step: 2112/529, loss: 0.330289751291275 2023-01-21 10:42:55.603440: step: 2116/529, loss: 0.36585643887519836 ================================================== Loss: 0.096 -------------------- Dev: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Test: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Chinese: {'event': {'p': 0.5529411764705883, 'r': 0.8703703703703703, 'f1': 0.6762589928057553}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Russian: {'event': {'p': 0.46, 'r': 0.6388888888888888, 'f1': 0.5348837209302325}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 11 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:43:47.968702: step: 4/529, loss: 0.0415768176317215 2023-01-21 10:43:49.199756: step: 8/529, loss: 0.16287919878959656 2023-01-21 10:43:50.379989: step: 12/529, loss: 0.0070549012161791325 2023-01-21 10:43:51.583464: step: 16/529, loss: 0.03151078149676323 2023-01-21 10:43:52.819991: step: 20/529, loss: 0.03748636692762375 2023-01-21 10:43:54.003707: step: 24/529, loss: 0.0657106414437294 2023-01-21 10:43:55.206856: step: 28/529, loss: 0.00724449148401618 2023-01-21 10:43:56.436211: step: 32/529, loss: 0.010496187955141068 2023-01-21 10:43:57.651986: step: 36/529, loss: 0.11836367100477219 2023-01-21 10:43:58.838997: step: 40/529, loss: 0.06352093815803528 2023-01-21 10:44:00.038358: step: 44/529, loss: 0.007104778196662664 2023-01-21 10:44:01.270746: step: 48/529, loss: 0.017368067055940628 2023-01-21 10:44:02.479628: step: 52/529, loss: 0.01285247877240181 2023-01-21 10:44:03.672087: step: 56/529, loss: 0.35003361105918884 2023-01-21 10:44:04.853014: step: 60/529, loss: 0.06347475200891495 2023-01-21 10:44:06.070306: step: 64/529, loss: 0.05626125633716583 2023-01-21 10:44:07.260785: step: 68/529, loss: 0.021786022931337357 2023-01-21 10:44:08.457903: step: 72/529, loss: 0.028809165582060814 2023-01-21 10:44:09.650053: step: 76/529, loss: 0.059075068682432175 2023-01-21 10:44:10.873201: step: 80/529, loss: 0.14954443275928497 2023-01-21 10:44:12.102740: step: 84/529, loss: 0.03666548803448677 2023-01-21 10:44:13.323364: step: 88/529, loss: 0.028780652210116386 2023-01-21 10:44:14.542399: step: 92/529, loss: 0.006824493408203125 2023-01-21 10:44:15.730349: step: 96/529, loss: 0.18172197043895721 2023-01-21 10:44:16.948856: step: 100/529, loss: 0.05139312893152237 2023-01-21 10:44:18.127475: step: 104/529, loss: 0.0803680419921875 2023-01-21 10:44:19.331963: step: 108/529, loss: 0.14311495423316956 2023-01-21 10:44:20.506240: step: 112/529, loss: 0.07278792560100555 2023-01-21 10:44:21.669303: step: 116/529, loss: 0.03270377963781357 2023-01-21 10:44:22.846995: step: 120/529, loss: 0.0043490887619555 2023-01-21 10:44:24.039508: step: 124/529, loss: 0.012589359655976295 2023-01-21 10:44:25.202102: step: 128/529, loss: 0.03191699832677841 2023-01-21 10:44:26.429487: step: 132/529, loss: 0.06626572459936142 2023-01-21 10:44:27.618160: step: 136/529, loss: 0.011196709237992764 2023-01-21 10:44:28.830846: step: 140/529, loss: 0.018349885940551758 2023-01-21 10:44:30.033225: step: 144/529, loss: 0.037526704370975494 2023-01-21 10:44:31.235289: step: 148/529, loss: 0.018480874598026276 2023-01-21 10:44:32.437010: step: 152/529, loss: 0.010340118780732155 2023-01-21 10:44:33.634354: step: 156/529, loss: 0.07742618024349213 2023-01-21 10:44:34.837576: step: 160/529, loss: 0.05386161804199219 2023-01-21 10:44:36.029453: step: 164/529, loss: 0.04370565339922905 2023-01-21 10:44:37.208288: step: 168/529, loss: 0.05732927471399307 2023-01-21 10:44:38.442120: step: 172/529, loss: 0.10836195200681686 2023-01-21 10:44:39.640307: step: 176/529, loss: 0.02942199818789959 2023-01-21 10:44:40.796277: step: 180/529, loss: 0.04060402140021324 2023-01-21 10:44:41.993629: step: 184/529, loss: 0.01614408567547798 2023-01-21 10:44:43.173499: step: 188/529, loss: 0.18061956763267517 2023-01-21 10:44:44.376082: step: 192/529, loss: 0.0369415283203125 2023-01-21 10:44:45.539784: step: 196/529, loss: 0.1239471435546875 2023-01-21 10:44:46.789480: step: 200/529, loss: 0.058680009096860886 2023-01-21 10:44:47.983354: step: 204/529, loss: 0.05028572306036949 2023-01-21 10:44:49.175395: step: 208/529, loss: 0.060007862746715546 2023-01-21 10:44:50.384455: step: 212/529, loss: 0.03610801696777344 2023-01-21 10:44:51.563137: step: 216/529, loss: 0.042501404881477356 2023-01-21 10:44:52.752669: step: 220/529, loss: 0.02070922777056694 2023-01-21 10:44:53.936372: step: 224/529, loss: 0.010623836889863014 2023-01-21 10:44:55.142283: step: 228/529, loss: 0.024311255663633347 2023-01-21 10:44:56.371491: step: 232/529, loss: 0.05710411071777344 2023-01-21 10:44:57.588931: step: 236/529, loss: 0.07958227396011353 2023-01-21 10:44:58.774681: step: 240/529, loss: 0.05572567135095596 2023-01-21 10:44:59.953739: step: 244/529, loss: 0.331353098154068 2023-01-21 10:45:01.148372: step: 248/529, loss: 0.053750231862068176 2023-01-21 10:45:02.368351: step: 252/529, loss: 0.03795952722430229 2023-01-21 10:45:03.555658: step: 256/529, loss: 0.007666397374123335 2023-01-21 10:45:04.778081: step: 260/529, loss: 0.16255168616771698 2023-01-21 10:45:06.002360: step: 264/529, loss: 0.008515920490026474 2023-01-21 10:45:07.231252: step: 268/529, loss: 0.056786537170410156 2023-01-21 10:45:08.410316: step: 272/529, loss: 0.0397094264626503 2023-01-21 10:45:09.610628: step: 276/529, loss: 0.02472515031695366 2023-01-21 10:45:10.743324: step: 280/529, loss: 0.00625221710652113 2023-01-21 10:45:11.906082: step: 284/529, loss: 0.02401280403137207 2023-01-21 10:45:13.146625: step: 288/529, loss: 0.053014516830444336 2023-01-21 10:45:14.401270: step: 292/529, loss: 0.007286358159035444 2023-01-21 10:45:15.586228: step: 296/529, loss: 0.0013787508942186832 2023-01-21 10:45:16.767841: step: 300/529, loss: 0.24921369552612305 2023-01-21 10:45:17.946032: step: 304/529, loss: 0.05410442501306534 2023-01-21 10:45:19.134401: step: 308/529, loss: 0.038373470306396484 2023-01-21 10:45:20.346044: step: 312/529, loss: 0.44612064957618713 2023-01-21 10:45:21.533377: step: 316/529, loss: 0.06152915954589844 2023-01-21 10:45:22.669350: step: 320/529, loss: 0.09439195692539215 2023-01-21 10:45:23.883745: step: 324/529, loss: 0.018712997436523438 2023-01-21 10:45:25.085536: step: 328/529, loss: 0.002446127124130726 2023-01-21 10:45:26.242521: step: 332/529, loss: 0.014412927441298962 2023-01-21 10:45:27.438442: step: 336/529, loss: 0.08691225200891495 2023-01-21 10:45:28.608538: step: 340/529, loss: 0.014534950256347656 2023-01-21 10:45:29.766647: step: 344/529, loss: 0.11884526908397675 2023-01-21 10:45:30.981383: step: 348/529, loss: 0.03379569202661514 2023-01-21 10:45:32.192003: step: 352/529, loss: 0.020771600306034088 2023-01-21 10:45:33.366793: step: 356/529, loss: 0.26751813292503357 2023-01-21 10:45:34.571779: step: 360/529, loss: 0.10104808956384659 2023-01-21 10:45:35.784320: step: 364/529, loss: 0.11772260814905167 2023-01-21 10:45:36.934068: step: 368/529, loss: 0.06868879497051239 2023-01-21 10:45:38.109573: step: 372/529, loss: 0.09869156032800674 2023-01-21 10:45:39.312538: step: 376/529, loss: 0.042247869074344635 2023-01-21 10:45:40.565320: step: 380/529, loss: 0.03956031799316406 2023-01-21 10:45:41.741823: step: 384/529, loss: 0.07672491669654846 2023-01-21 10:45:42.992579: step: 388/529, loss: 0.11837844550609589 2023-01-21 10:45:44.202736: step: 392/529, loss: 0.012391806580126286 2023-01-21 10:45:45.475785: step: 396/529, loss: 0.2483467012643814 2023-01-21 10:45:46.669206: step: 400/529, loss: 0.005062580108642578 2023-01-21 10:45:47.843133: step: 404/529, loss: 0.07758307456970215 2023-01-21 10:45:49.029014: step: 408/529, loss: 0.08309879899024963 2023-01-21 10:45:50.216644: step: 412/529, loss: 0.0534786693751812 2023-01-21 10:45:51.409377: step: 416/529, loss: 0.04766378924250603 2023-01-21 10:45:52.650895: step: 420/529, loss: 0.017609978094697 2023-01-21 10:45:53.868259: step: 424/529, loss: 0.03704414516687393 2023-01-21 10:45:55.072825: step: 428/529, loss: 0.027111627161502838 2023-01-21 10:45:56.260587: step: 432/529, loss: 0.02130155637860298 2023-01-21 10:45:57.469098: step: 436/529, loss: 0.007975387386977673 2023-01-21 10:45:58.653747: step: 440/529, loss: 0.071660615503788 2023-01-21 10:45:59.872155: step: 444/529, loss: 0.05532436445355415 2023-01-21 10:46:01.044862: step: 448/529, loss: 0.08488073199987411 2023-01-21 10:46:02.302075: step: 452/529, loss: 0.0222688689827919 2023-01-21 10:46:03.519827: step: 456/529, loss: 0.04647884517908096 2023-01-21 10:46:04.732269: step: 460/529, loss: 0.17887821793556213 2023-01-21 10:46:05.937016: step: 464/529, loss: 0.10830917209386826 2023-01-21 10:46:07.137307: step: 468/529, loss: 0.03001413494348526 2023-01-21 10:46:08.328879: step: 472/529, loss: 0.059232331812381744 2023-01-21 10:46:09.536325: step: 476/529, loss: 0.0866231918334961 2023-01-21 10:46:10.710750: step: 480/529, loss: 0.001173305558040738 2023-01-21 10:46:11.957458: step: 484/529, loss: 0.04395580664277077 2023-01-21 10:46:13.152836: step: 488/529, loss: 0.012170601636171341 2023-01-21 10:46:14.328866: step: 492/529, loss: 0.006155920214951038 2023-01-21 10:46:15.527009: step: 496/529, loss: 0.009749412536621094 2023-01-21 10:46:16.789899: step: 500/529, loss: 0.08456535637378693 2023-01-21 10:46:18.032375: step: 504/529, loss: 0.07859258353710175 2023-01-21 10:46:19.205112: step: 508/529, loss: 0.044638726860284805 2023-01-21 10:46:20.398213: step: 512/529, loss: 0.03965149074792862 2023-01-21 10:46:21.608604: step: 516/529, loss: 0.3025369644165039 2023-01-21 10:46:22.860560: step: 520/529, loss: 0.005192232318222523 2023-01-21 10:46:24.047745: step: 524/529, loss: 0.016986466944217682 2023-01-21 10:46:25.265802: step: 528/529, loss: 0.018117427825927734 2023-01-21 10:46:26.425507: step: 532/529, loss: 0.01415863074362278 2023-01-21 10:46:27.621064: step: 536/529, loss: 0.1638137847185135 2023-01-21 10:46:28.826361: step: 540/529, loss: 2.179351806640625 2023-01-21 10:46:30.004847: step: 544/529, loss: 0.004407837055623531 2023-01-21 10:46:31.209364: step: 548/529, loss: 0.061171628534793854 2023-01-21 10:46:32.397340: step: 552/529, loss: 0.08730974793434143 2023-01-21 10:46:33.589838: step: 556/529, loss: 0.1683538407087326 2023-01-21 10:46:34.804021: step: 560/529, loss: 0.10898485779762268 2023-01-21 10:46:36.057367: step: 564/529, loss: 0.06438169628381729 2023-01-21 10:46:37.278327: step: 568/529, loss: 0.014602613635361195 2023-01-21 10:46:38.487787: step: 572/529, loss: 0.025857355445623398 2023-01-21 10:46:39.681538: step: 576/529, loss: 0.04417915269732475 2023-01-21 10:46:40.925762: step: 580/529, loss: 0.012629318982362747 2023-01-21 10:46:42.104381: step: 584/529, loss: 0.09801841527223587 2023-01-21 10:46:43.320047: step: 588/529, loss: 0.03852539137005806 2023-01-21 10:46:44.518535: step: 592/529, loss: 0.04214029386639595 2023-01-21 10:46:45.721160: step: 596/529, loss: 0.20783615112304688 2023-01-21 10:46:46.887135: step: 600/529, loss: 0.0742349624633789 2023-01-21 10:46:48.139008: step: 604/529, loss: 0.061101723462343216 2023-01-21 10:46:49.323258: step: 608/529, loss: 0.06343875080347061 2023-01-21 10:46:50.556360: step: 612/529, loss: 0.04918226972222328 2023-01-21 10:46:51.760555: step: 616/529, loss: 0.1318499594926834 2023-01-21 10:46:52.971541: step: 620/529, loss: 0.018779754638671875 2023-01-21 10:46:54.169490: step: 624/529, loss: 0.11114434897899628 2023-01-21 10:46:55.404097: step: 628/529, loss: 0.08607196807861328 2023-01-21 10:46:56.593019: step: 632/529, loss: 0.0006310463068075478 2023-01-21 10:46:57.811627: step: 636/529, loss: 0.0015189171535894275 2023-01-21 10:46:59.058768: step: 640/529, loss: 0.006538200657814741 2023-01-21 10:47:00.313257: step: 644/529, loss: 0.08150138705968857 2023-01-21 10:47:01.492807: step: 648/529, loss: 0.08642778545618057 2023-01-21 10:47:02.707875: step: 652/529, loss: 0.04719109833240509 2023-01-21 10:47:03.923334: step: 656/529, loss: 0.17432785034179688 2023-01-21 10:47:05.131763: step: 660/529, loss: 0.10804205387830734 2023-01-21 10:47:06.324978: step: 664/529, loss: 0.0521734245121479 2023-01-21 10:47:07.499006: step: 668/529, loss: 0.049578286707401276 2023-01-21 10:47:08.698162: step: 672/529, loss: 0.6951566934585571 2023-01-21 10:47:09.873204: step: 676/529, loss: 0.16706885397434235 2023-01-21 10:47:11.103588: step: 680/529, loss: 0.056014157831668854 2023-01-21 10:47:12.407503: step: 684/529, loss: 0.059819985181093216 2023-01-21 10:47:13.583216: step: 688/529, loss: 0.0017995834350585938 2023-01-21 10:47:14.838352: step: 692/529, loss: 0.013722610659897327 2023-01-21 10:47:16.028841: step: 696/529, loss: 0.348165899515152 2023-01-21 10:47:17.195601: step: 700/529, loss: 0.059636928141117096 2023-01-21 10:47:18.384756: step: 704/529, loss: 0.07060060650110245 2023-01-21 10:47:19.546013: step: 708/529, loss: 0.00022745133901480585 2023-01-21 10:47:20.715209: step: 712/529, loss: 0.002770645311102271 2023-01-21 10:47:21.927427: step: 716/529, loss: 0.05276527628302574 2023-01-21 10:47:23.116695: step: 720/529, loss: 0.02288804203271866 2023-01-21 10:47:24.309800: step: 724/529, loss: 0.025639917701482773 2023-01-21 10:47:25.533608: step: 728/529, loss: 0.013970565050840378 2023-01-21 10:47:26.709451: step: 732/529, loss: 0.03739595413208008 2023-01-21 10:47:27.915721: step: 736/529, loss: 0.03491213172674179 2023-01-21 10:47:29.129009: step: 740/529, loss: 0.048519134521484375 2023-01-21 10:47:30.354696: step: 744/529, loss: 0.0017409325810149312 2023-01-21 10:47:31.513984: step: 748/529, loss: 0.015275574289262295 2023-01-21 10:47:32.720400: step: 752/529, loss: 0.06074018403887749 2023-01-21 10:47:33.957353: step: 756/529, loss: 0.08562984317541122 2023-01-21 10:47:35.196882: step: 760/529, loss: 0.004721450619399548 2023-01-21 10:47:36.346804: step: 764/529, loss: 0.05725431442260742 2023-01-21 10:47:37.576992: step: 768/529, loss: 0.11611500382423401 2023-01-21 10:47:38.738480: step: 772/529, loss: 0.006209087558090687 2023-01-21 10:47:39.930610: step: 776/529, loss: 0.08293185383081436 2023-01-21 10:47:41.128328: step: 780/529, loss: 0.0747164785861969 2023-01-21 10:47:42.303043: step: 784/529, loss: 0.053025342524051666 2023-01-21 10:47:43.502256: step: 788/529, loss: 0.022747613489627838 2023-01-21 10:47:44.665964: step: 792/529, loss: 0.011130904778838158 2023-01-21 10:47:45.824668: step: 796/529, loss: 0.028368093073368073 2023-01-21 10:47:47.017674: step: 800/529, loss: 0.07959599792957306 2023-01-21 10:47:48.193712: step: 804/529, loss: 0.10140614956617355 2023-01-21 10:47:49.365753: step: 808/529, loss: 0.0361296683549881 2023-01-21 10:47:50.561447: step: 812/529, loss: 0.06905317306518555 2023-01-21 10:47:51.794282: step: 816/529, loss: 0.08662247657775879 2023-01-21 10:47:53.034924: step: 820/529, loss: 0.07569456100463867 2023-01-21 10:47:54.216000: step: 824/529, loss: 0.027566147968173027 2023-01-21 10:47:55.421746: step: 828/529, loss: 0.03365898132324219 2023-01-21 10:47:56.605723: step: 832/529, loss: 0.05202503129839897 2023-01-21 10:47:57.797827: step: 836/529, loss: 0.059766486287117004 2023-01-21 10:47:59.042016: step: 840/529, loss: 0.3832658529281616 2023-01-21 10:48:00.246514: step: 844/529, loss: 0.031035233289003372 2023-01-21 10:48:01.444534: step: 848/529, loss: 0.014853572472929955 2023-01-21 10:48:02.640899: step: 852/529, loss: 0.0037977220490574837 2023-01-21 10:48:03.864578: step: 856/529, loss: 0.10528374463319778 2023-01-21 10:48:05.029643: step: 860/529, loss: 0.0754704475402832 2023-01-21 10:48:06.196392: step: 864/529, loss: 0.0284087173640728 2023-01-21 10:48:07.369321: step: 868/529, loss: 0.015226030722260475 2023-01-21 10:48:08.544366: step: 872/529, loss: 0.04141269251704216 2023-01-21 10:48:09.751609: step: 876/529, loss: 0.03359193727374077 2023-01-21 10:48:10.977404: step: 880/529, loss: 0.0811983048915863 2023-01-21 10:48:12.153540: step: 884/529, loss: 0.2063777893781662 2023-01-21 10:48:13.357322: step: 888/529, loss: 0.011458205990493298 2023-01-21 10:48:14.553340: step: 892/529, loss: 0.0023745535872876644 2023-01-21 10:48:15.769606: step: 896/529, loss: 0.07005605101585388 2023-01-21 10:48:16.938597: step: 900/529, loss: 0.009489346295595169 2023-01-21 10:48:18.145039: step: 904/529, loss: 0.0004330635129008442 2023-01-21 10:48:19.347909: step: 908/529, loss: 0.028146887198090553 2023-01-21 10:48:20.557393: step: 912/529, loss: 0.06805019080638885 2023-01-21 10:48:21.760412: step: 916/529, loss: 0.013579512014985085 2023-01-21 10:48:22.984454: step: 920/529, loss: 0.02916431427001953 2023-01-21 10:48:24.184679: step: 924/529, loss: 0.08250255882740021 2023-01-21 10:48:25.416317: step: 928/529, loss: 0.03976269066333771 2023-01-21 10:48:26.648670: step: 932/529, loss: 0.009790134616196156 2023-01-21 10:48:27.830081: step: 936/529, loss: 0.014933109283447266 2023-01-21 10:48:29.053771: step: 940/529, loss: 0.013311576098203659 2023-01-21 10:48:30.268906: step: 944/529, loss: 0.045032598078250885 2023-01-21 10:48:31.470856: step: 948/529, loss: 0.2498399317264557 2023-01-21 10:48:32.669288: step: 952/529, loss: 0.005346202757209539 2023-01-21 10:48:33.861926: step: 956/529, loss: 0.039979077875614166 2023-01-21 10:48:35.109895: step: 960/529, loss: 0.14465287327766418 2023-01-21 10:48:36.325874: step: 964/529, loss: 0.13699102401733398 2023-01-21 10:48:37.538803: step: 968/529, loss: 0.04071617126464844 2023-01-21 10:48:38.715241: step: 972/529, loss: 0.08028783649206161 2023-01-21 10:48:39.938328: step: 976/529, loss: 0.0539124496281147 2023-01-21 10:48:41.120646: step: 980/529, loss: 0.03078932873904705 2023-01-21 10:48:42.304313: step: 984/529, loss: 0.008827686309814453 2023-01-21 10:48:43.493426: step: 988/529, loss: 0.004320716951042414 2023-01-21 10:48:44.682913: step: 992/529, loss: 0.03374020755290985 2023-01-21 10:48:45.864971: step: 996/529, loss: 0.04356823116540909 2023-01-21 10:48:47.067712: step: 1000/529, loss: 0.007504081819206476 2023-01-21 10:48:48.282000: step: 1004/529, loss: 0.2338055670261383 2023-01-21 10:48:49.472920: step: 1008/529, loss: 0.010401297360658646 2023-01-21 10:48:50.714530: step: 1012/529, loss: 0.7151467800140381 2023-01-21 10:48:51.923891: step: 1016/529, loss: 0.007266903296113014 2023-01-21 10:48:53.134413: step: 1020/529, loss: 0.03681335598230362 2023-01-21 10:48:54.336296: step: 1024/529, loss: 0.046021413058042526 2023-01-21 10:48:55.553099: step: 1028/529, loss: 0.5442251563072205 2023-01-21 10:48:56.764764: step: 1032/529, loss: 0.11134085804224014 2023-01-21 10:48:57.989044: step: 1036/529, loss: 0.03705864027142525 2023-01-21 10:48:59.165496: step: 1040/529, loss: 0.10039444267749786 2023-01-21 10:49:00.370952: step: 1044/529, loss: 0.035482216626405716 2023-01-21 10:49:01.606822: step: 1048/529, loss: 0.1307574212551117 2023-01-21 10:49:02.819602: step: 1052/529, loss: 0.0779542475938797 2023-01-21 10:49:04.002231: step: 1056/529, loss: 0.026824284344911575 2023-01-21 10:49:05.210037: step: 1060/529, loss: 0.06595049053430557 2023-01-21 10:49:06.404093: step: 1064/529, loss: 0.27425500750541687 2023-01-21 10:49:07.583423: step: 1068/529, loss: 0.8390282392501831 2023-01-21 10:49:08.771412: step: 1072/529, loss: 0.10883913189172745 2023-01-21 10:49:09.918707: step: 1076/529, loss: 0.4379534125328064 2023-01-21 10:49:11.077678: step: 1080/529, loss: 0.10718196630477905 2023-01-21 10:49:12.288411: step: 1084/529, loss: 0.6153775453567505 2023-01-21 10:49:13.517276: step: 1088/529, loss: 0.037024691700935364 2023-01-21 10:49:14.727124: step: 1092/529, loss: 0.09814572334289551 2023-01-21 10:49:15.893351: step: 1096/529, loss: 0.13121357560157776 2023-01-21 10:49:17.089130: step: 1100/529, loss: 0.004143333528190851 2023-01-21 10:49:18.263259: step: 1104/529, loss: 0.003635501954704523 2023-01-21 10:49:19.455425: step: 1108/529, loss: 0.09390535205602646 2023-01-21 10:49:20.673734: step: 1112/529, loss: 0.13246221840381622 2023-01-21 10:49:21.854340: step: 1116/529, loss: 0.04397711902856827 2023-01-21 10:49:23.027357: step: 1120/529, loss: 0.07568073272705078 2023-01-21 10:49:24.262341: step: 1124/529, loss: 0.024960899725556374 2023-01-21 10:49:25.488774: step: 1128/529, loss: 0.06967039406299591 2023-01-21 10:49:26.702092: step: 1132/529, loss: 0.03537330776453018 2023-01-21 10:49:27.915633: step: 1136/529, loss: 0.2469595968723297 2023-01-21 10:49:29.099079: step: 1140/529, loss: 0.03331012651324272 2023-01-21 10:49:30.244241: step: 1144/529, loss: 0.17991404235363007 2023-01-21 10:49:31.446029: step: 1148/529, loss: 0.0007800101884640753 2023-01-21 10:49:32.651374: step: 1152/529, loss: 0.03366527706384659 2023-01-21 10:49:33.832610: step: 1156/529, loss: 0.06735801696777344 2023-01-21 10:49:35.027024: step: 1160/529, loss: 0.01091608963906765 2023-01-21 10:49:36.227443: step: 1164/529, loss: 0.04776306077837944 2023-01-21 10:49:37.393502: step: 1168/529, loss: 0.020830631256103516 2023-01-21 10:49:38.565707: step: 1172/529, loss: 0.07038145512342453 2023-01-21 10:49:39.788068: step: 1176/529, loss: 0.035063743591308594 2023-01-21 10:49:40.982471: step: 1180/529, loss: 0.003455638885498047 2023-01-21 10:49:42.175808: step: 1184/529, loss: 0.024196529760956764 2023-01-21 10:49:43.364187: step: 1188/529, loss: 0.5926642417907715 2023-01-21 10:49:44.533739: step: 1192/529, loss: 0.016666889190673828 2023-01-21 10:49:45.726000: step: 1196/529, loss: 0.007498359773308039 2023-01-21 10:49:46.894002: step: 1200/529, loss: 0.08083172142505646 2023-01-21 10:49:48.071828: step: 1204/529, loss: 0.011226177215576172 2023-01-21 10:49:49.275969: step: 1208/529, loss: 0.026195431128144264 2023-01-21 10:49:50.423324: step: 1212/529, loss: 0.14075350761413574 2023-01-21 10:49:51.617253: step: 1216/529, loss: 0.01112380065023899 2023-01-21 10:49:52.824276: step: 1220/529, loss: 0.08782787621021271 2023-01-21 10:49:54.038266: step: 1224/529, loss: 0.045603275299072266 2023-01-21 10:49:55.238923: step: 1228/529, loss: 0.0511232390999794 2023-01-21 10:49:56.502299: step: 1232/529, loss: 0.033591464161872864 2023-01-21 10:49:57.679455: step: 1236/529, loss: 0.07966890186071396 2023-01-21 10:49:58.847311: step: 1240/529, loss: 0.08808381855487823 2023-01-21 10:50:00.046859: step: 1244/529, loss: 0.08162736892700195 2023-01-21 10:50:01.251481: step: 1248/529, loss: 0.004642772488296032 2023-01-21 10:50:02.450432: step: 1252/529, loss: 0.004563713446259499 2023-01-21 10:50:03.642876: step: 1256/529, loss: 0.04097886383533478 2023-01-21 10:50:04.821137: step: 1260/529, loss: 0.015616036020219326 2023-01-21 10:50:06.049865: step: 1264/529, loss: 0.2817387580871582 2023-01-21 10:50:07.233339: step: 1268/529, loss: 0.0014008521102368832 2023-01-21 10:50:08.409990: step: 1272/529, loss: 0.2954476475715637 2023-01-21 10:50:09.640343: step: 1276/529, loss: 0.016694659367203712 2023-01-21 10:50:10.845905: step: 1280/529, loss: 0.010402154177427292 2023-01-21 10:50:12.076814: step: 1284/529, loss: 0.010169655084609985 2023-01-21 10:50:13.308405: step: 1288/529, loss: 0.7191836833953857 2023-01-21 10:50:14.549707: step: 1292/529, loss: 0.04813079908490181 2023-01-21 10:50:15.790629: step: 1296/529, loss: 0.16865968704223633 2023-01-21 10:50:17.027609: step: 1300/529, loss: 0.0674375593662262 2023-01-21 10:50:18.294785: step: 1304/529, loss: 0.040503308176994324 2023-01-21 10:50:19.535745: step: 1308/529, loss: 0.009086275473237038 2023-01-21 10:50:20.781054: step: 1312/529, loss: 0.05688035488128662 2023-01-21 10:50:22.026509: step: 1316/529, loss: 0.13068923354148865 2023-01-21 10:50:23.231480: step: 1320/529, loss: 0.026905059814453125 2023-01-21 10:50:24.458948: step: 1324/529, loss: 0.02541036531329155 2023-01-21 10:50:25.647111: step: 1328/529, loss: 0.02647991105914116 2023-01-21 10:50:26.835992: step: 1332/529, loss: 0.03379230573773384 2023-01-21 10:50:28.045888: step: 1336/529, loss: 0.04011640325188637 2023-01-21 10:50:29.246846: step: 1340/529, loss: 0.051969338208436966 2023-01-21 10:50:30.429264: step: 1344/529, loss: 0.03831997141242027 2023-01-21 10:50:31.614518: step: 1348/529, loss: 0.0036433220375329256 2023-01-21 10:50:32.812538: step: 1352/529, loss: 0.031075477600097656 2023-01-21 10:50:34.044373: step: 1356/529, loss: 0.2768142819404602 2023-01-21 10:50:35.306554: step: 1360/529, loss: 0.11124163120985031 2023-01-21 10:50:36.491452: step: 1364/529, loss: 0.05142412334680557 2023-01-21 10:50:37.684381: step: 1368/529, loss: 0.06866578757762909 2023-01-21 10:50:38.900022: step: 1372/529, loss: 0.06152145937085152 2023-01-21 10:50:40.076700: step: 1376/529, loss: 0.04783432558178902 2023-01-21 10:50:41.262688: step: 1380/529, loss: 0.013824463821947575 2023-01-21 10:50:42.445521: step: 1384/529, loss: 0.04217877611517906 2023-01-21 10:50:43.644100: step: 1388/529, loss: 0.019161606207489967 2023-01-21 10:50:44.873326: step: 1392/529, loss: 0.014749813824892044 2023-01-21 10:50:46.058209: step: 1396/529, loss: 0.08193669468164444 2023-01-21 10:50:47.243085: step: 1400/529, loss: 1.2768535614013672 2023-01-21 10:50:48.441848: step: 1404/529, loss: 0.060097888112068176 2023-01-21 10:50:49.641508: step: 1408/529, loss: 0.03291139751672745 2023-01-21 10:50:50.839622: step: 1412/529, loss: 0.0447445884346962 2023-01-21 10:50:52.031956: step: 1416/529, loss: 0.06698312610387802 2023-01-21 10:50:53.249639: step: 1420/529, loss: 0.005542087368667126 2023-01-21 10:50:54.400102: step: 1424/529, loss: 0.0012058259453624487 2023-01-21 10:50:55.568246: step: 1428/529, loss: 0.01199884433299303 2023-01-21 10:50:56.777373: step: 1432/529, loss: 0.005538749508559704 2023-01-21 10:50:57.946819: step: 1436/529, loss: 0.1967674195766449 2023-01-21 10:50:59.185818: step: 1440/529, loss: 0.08329162746667862 2023-01-21 10:51:00.398964: step: 1444/529, loss: 0.09773287922143936 2023-01-21 10:51:01.611390: step: 1448/529, loss: 0.08748719841241837 2023-01-21 10:51:02.811567: step: 1452/529, loss: 0.11403703689575195 2023-01-21 10:51:04.034268: step: 1456/529, loss: 0.04840588569641113 2023-01-21 10:51:05.190037: step: 1460/529, loss: 0.021442366763949394 2023-01-21 10:51:06.371233: step: 1464/529, loss: 0.3614417016506195 2023-01-21 10:51:07.546975: step: 1468/529, loss: 0.05172135308384895 2023-01-21 10:51:08.706146: step: 1472/529, loss: 0.11756746470928192 2023-01-21 10:51:09.898233: step: 1476/529, loss: 0.03296380117535591 2023-01-21 10:51:11.080248: step: 1480/529, loss: 0.00482444791123271 2023-01-21 10:51:12.279650: step: 1484/529, loss: 0.04635229334235191 2023-01-21 10:51:13.502080: step: 1488/529, loss: 0.10678057372570038 2023-01-21 10:51:14.689214: step: 1492/529, loss: 0.08385252952575684 2023-01-21 10:51:15.863829: step: 1496/529, loss: 0.09917765110731125 2023-01-21 10:51:17.066364: step: 1500/529, loss: 0.0008244514465332031 2023-01-21 10:51:18.232124: step: 1504/529, loss: 0.006707263179123402 2023-01-21 10:51:19.437796: step: 1508/529, loss: 0.04261207580566406 2023-01-21 10:51:20.641531: step: 1512/529, loss: 0.038673970848321915 2023-01-21 10:51:21.832544: step: 1516/529, loss: 0.1708984375 2023-01-21 10:51:23.056547: step: 1520/529, loss: 0.02282085455954075 2023-01-21 10:51:24.280492: step: 1524/529, loss: 0.003697013948112726 2023-01-21 10:51:25.426319: step: 1528/529, loss: 0.1574891060590744 2023-01-21 10:51:26.591322: step: 1532/529, loss: 0.02370772324502468 2023-01-21 10:51:27.795299: step: 1536/529, loss: 0.1049346923828125 2023-01-21 10:51:29.035555: step: 1540/529, loss: 0.10836047679185867 2023-01-21 10:51:30.221014: step: 1544/529, loss: 0.06298549473285675 2023-01-21 10:51:31.453636: step: 1548/529, loss: 0.05909767374396324 2023-01-21 10:51:32.658265: step: 1552/529, loss: 0.22140489518642426 2023-01-21 10:51:33.843077: step: 1556/529, loss: 1.6520860195159912 2023-01-21 10:51:35.015763: step: 1560/529, loss: 0.03864097595214844 2023-01-21 10:51:36.228324: step: 1564/529, loss: 0.06007556617259979 2023-01-21 10:51:37.439249: step: 1568/529, loss: 0.0972478911280632 2023-01-21 10:51:38.587142: step: 1572/529, loss: 0.04356813430786133 2023-01-21 10:51:39.780734: step: 1576/529, loss: 0.023343658074736595 2023-01-21 10:51:40.985870: step: 1580/529, loss: 0.044876955449581146 2023-01-21 10:51:42.189834: step: 1584/529, loss: 0.09529328346252441 2023-01-21 10:51:43.396551: step: 1588/529, loss: 0.007829666137695312 2023-01-21 10:51:44.557207: step: 1592/529, loss: 0.009492016397416592 2023-01-21 10:51:45.799169: step: 1596/529, loss: 0.06847629696130753 2023-01-21 10:51:46.999308: step: 1600/529, loss: 0.04010649025440216 2023-01-21 10:51:48.160794: step: 1604/529, loss: 0.0021521567832678556 2023-01-21 10:51:49.331868: step: 1608/529, loss: 0.08057518303394318 2023-01-21 10:51:50.535135: step: 1612/529, loss: 0.0020606995094567537 2023-01-21 10:51:51.755623: step: 1616/529, loss: 0.011684322729706764 2023-01-21 10:51:53.025610: step: 1620/529, loss: 0.013513803482055664 2023-01-21 10:51:54.202837: step: 1624/529, loss: 0.049569036811590195 2023-01-21 10:51:55.429688: step: 1628/529, loss: 0.07808075100183487 2023-01-21 10:51:56.668151: step: 1632/529, loss: 0.0656675398349762 2023-01-21 10:51:57.893929: step: 1636/529, loss: 0.08072996884584427 2023-01-21 10:51:59.070279: step: 1640/529, loss: 0.10784606635570526 2023-01-21 10:52:00.285900: step: 1644/529, loss: 0.008269643411040306 2023-01-21 10:52:01.543956: step: 1648/529, loss: 0.02915988117456436 2023-01-21 10:52:02.753481: step: 1652/529, loss: 0.01055297814309597 2023-01-21 10:52:03.995253: step: 1656/529, loss: 0.16715869307518005 2023-01-21 10:52:05.220456: step: 1660/529, loss: 0.02644510380923748 2023-01-21 10:52:06.417548: step: 1664/529, loss: 0.039705850183963776 2023-01-21 10:52:07.600257: step: 1668/529, loss: 0.15847139060497284 2023-01-21 10:52:08.805675: step: 1672/529, loss: 0.0022668838500976562 2023-01-21 10:52:10.012962: step: 1676/529, loss: 0.05460897833108902 2023-01-21 10:52:11.199355: step: 1680/529, loss: 0.028580807149410248 2023-01-21 10:52:12.410009: step: 1684/529, loss: 0.009059716016054153 2023-01-21 10:52:13.589593: step: 1688/529, loss: 0.02703390084207058 2023-01-21 10:52:14.773916: step: 1692/529, loss: 0.11716632544994354 2023-01-21 10:52:15.979483: step: 1696/529, loss: 0.033617496490478516 2023-01-21 10:52:17.156822: step: 1700/529, loss: 0.014165449887514114 2023-01-21 10:52:18.377948: step: 1704/529, loss: 0.28714197874069214 2023-01-21 10:52:19.622993: step: 1708/529, loss: 0.06220531836152077 2023-01-21 10:52:20.845929: step: 1712/529, loss: 0.05443773418664932 2023-01-21 10:52:22.081565: step: 1716/529, loss: 0.05654316022992134 2023-01-21 10:52:23.261688: step: 1720/529, loss: 0.01900176890194416 2023-01-21 10:52:24.547053: step: 1724/529, loss: 0.05625953897833824 2023-01-21 10:52:25.729715: step: 1728/529, loss: 0.03941211849451065 2023-01-21 10:52:26.908662: step: 1732/529, loss: 0.047432586550712585 2023-01-21 10:52:28.161378: step: 1736/529, loss: 0.0619785338640213 2023-01-21 10:52:29.343934: step: 1740/529, loss: 0.03973264992237091 2023-01-21 10:52:30.518799: step: 1744/529, loss: 0.04565148428082466 2023-01-21 10:52:31.677527: step: 1748/529, loss: 0.002802229020744562 2023-01-21 10:52:32.878204: step: 1752/529, loss: 0.004509186837822199 2023-01-21 10:52:34.088720: step: 1756/529, loss: 0.07708778977394104 2023-01-21 10:52:35.267940: step: 1760/529, loss: 0.04762439802289009 2023-01-21 10:52:36.485995: step: 1764/529, loss: 0.023673629388213158 2023-01-21 10:52:37.690795: step: 1768/529, loss: 0.08362588286399841 2023-01-21 10:52:38.857253: step: 1772/529, loss: 0.032880593091249466 2023-01-21 10:52:40.040193: step: 1776/529, loss: 0.0415492057800293 2023-01-21 10:52:41.221739: step: 1780/529, loss: 0.020442772656679153 2023-01-21 10:52:42.381109: step: 1784/529, loss: 0.0461428165435791 2023-01-21 10:52:43.615681: step: 1788/529, loss: 0.005391550250351429 2023-01-21 10:52:44.835433: step: 1792/529, loss: 0.05739030987024307 2023-01-21 10:52:46.058595: step: 1796/529, loss: 0.08875790238380432 2023-01-21 10:52:47.225164: step: 1800/529, loss: 0.021050645038485527 2023-01-21 10:52:48.454237: step: 1804/529, loss: 0.11907606571912766 2023-01-21 10:52:49.645953: step: 1808/529, loss: 0.017148304730653763 2023-01-21 10:52:50.842914: step: 1812/529, loss: 0.010773849673569202 2023-01-21 10:52:52.065821: step: 1816/529, loss: 0.010222816839814186 2023-01-21 10:52:53.285931: step: 1820/529, loss: 0.0512174628674984 2023-01-21 10:52:54.498019: step: 1824/529, loss: 0.09181147068738937 2023-01-21 10:52:55.715036: step: 1828/529, loss: 0.0018573760753497481 2023-01-21 10:52:56.929328: step: 1832/529, loss: 0.007009983062744141 2023-01-21 10:52:58.130088: step: 1836/529, loss: 0.004858588799834251 2023-01-21 10:52:59.328625: step: 1840/529, loss: 0.03976621478796005 2023-01-21 10:53:00.526435: step: 1844/529, loss: 0.011035728268325329 2023-01-21 10:53:01.730900: step: 1848/529, loss: 0.010021209716796875 2023-01-21 10:53:02.958577: step: 1852/529, loss: 0.12872228026390076 2023-01-21 10:53:04.181592: step: 1856/529, loss: 0.36261507868766785 2023-01-21 10:53:05.355326: step: 1860/529, loss: 0.0037521361373364925 2023-01-21 10:53:06.522095: step: 1864/529, loss: 0.09352226555347443 2023-01-21 10:53:07.702749: step: 1868/529, loss: 0.04515848308801651 2023-01-21 10:53:08.910430: step: 1872/529, loss: 0.05004854127764702 2023-01-21 10:53:10.109199: step: 1876/529, loss: 0.025104999542236328 2023-01-21 10:53:11.291507: step: 1880/529, loss: 0.09868564456701279 2023-01-21 10:53:12.485336: step: 1884/529, loss: 0.06416326016187668 2023-01-21 10:53:13.656366: step: 1888/529, loss: 0.030042458325624466 2023-01-21 10:53:14.836134: step: 1892/529, loss: 0.032268330454826355 2023-01-21 10:53:16.030125: step: 1896/529, loss: 0.14064764976501465 2023-01-21 10:53:17.219302: step: 1900/529, loss: 0.021680736914277077 2023-01-21 10:53:18.415390: step: 1904/529, loss: 0.0785006508231163 2023-01-21 10:53:19.604870: step: 1908/529, loss: 0.14607620239257812 2023-01-21 10:53:20.859084: step: 1912/529, loss: 0.10666332393884659 2023-01-21 10:53:22.060383: step: 1916/529, loss: 0.05374584347009659 2023-01-21 10:53:23.224402: step: 1920/529, loss: 0.030852699652314186 2023-01-21 10:53:24.423870: step: 1924/529, loss: 0.012407065369188786 2023-01-21 10:53:25.600055: step: 1928/529, loss: 0.02901306189596653 2023-01-21 10:53:26.800768: step: 1932/529, loss: 0.0940331518650055 2023-01-21 10:53:27.998853: step: 1936/529, loss: 0.10514755547046661 2023-01-21 10:53:29.203992: step: 1940/529, loss: 0.005821514409035444 2023-01-21 10:53:30.395650: step: 1944/529, loss: 0.0033780098892748356 2023-01-21 10:53:31.598543: step: 1948/529, loss: 0.08220081031322479 2023-01-21 10:53:32.790243: step: 1952/529, loss: 0.10253963619470596 2023-01-21 10:53:33.999515: step: 1956/529, loss: 0.012839126400649548 2023-01-21 10:53:35.187246: step: 1960/529, loss: 0.0481351837515831 2023-01-21 10:53:36.372624: step: 1964/529, loss: 0.004809475038200617 2023-01-21 10:53:37.569265: step: 1968/529, loss: 0.05212026834487915 2023-01-21 10:53:38.759632: step: 1972/529, loss: 0.039101503789424896 2023-01-21 10:53:40.010712: step: 1976/529, loss: 0.021471597254276276 2023-01-21 10:53:41.212918: step: 1980/529, loss: 0.038008108735084534 2023-01-21 10:53:42.408407: step: 1984/529, loss: 0.0005153656238690019 2023-01-21 10:53:43.579285: step: 1988/529, loss: 0.05422983318567276 2023-01-21 10:53:44.773655: step: 1992/529, loss: 0.05885887145996094 2023-01-21 10:53:45.964275: step: 1996/529, loss: 0.09132575988769531 2023-01-21 10:53:47.128183: step: 2000/529, loss: 0.04573478922247887 2023-01-21 10:53:48.344469: step: 2004/529, loss: 0.031400490552186966 2023-01-21 10:53:49.554521: step: 2008/529, loss: 0.1244727075099945 2023-01-21 10:53:50.742481: step: 2012/529, loss: 0.030265428125858307 2023-01-21 10:53:51.975808: step: 2016/529, loss: 0.10329704731702805 2023-01-21 10:53:53.139516: step: 2020/529, loss: 0.0018930435180664062 2023-01-21 10:53:54.361693: step: 2024/529, loss: 0.02173147350549698 2023-01-21 10:53:55.535555: step: 2028/529, loss: 0.05556526407599449 2023-01-21 10:53:56.671109: step: 2032/529, loss: 0.01640152931213379 2023-01-21 10:53:57.942161: step: 2036/529, loss: 0.061226844787597656 2023-01-21 10:53:59.132838: step: 2040/529, loss: 0.0216490738093853 2023-01-21 10:54:00.372272: step: 2044/529, loss: 0.03312535211443901 2023-01-21 10:54:01.575876: step: 2048/529, loss: 0.030582714825868607 2023-01-21 10:54:02.797556: step: 2052/529, loss: 0.01811254397034645 2023-01-21 10:54:04.009438: step: 2056/529, loss: 0.0998251885175705 2023-01-21 10:54:05.191128: step: 2060/529, loss: 0.03838043287396431 2023-01-21 10:54:06.334966: step: 2064/529, loss: 0.04695744439959526 2023-01-21 10:54:07.542065: step: 2068/529, loss: 0.09503249824047089 2023-01-21 10:54:08.735714: step: 2072/529, loss: 0.016979455947875977 2023-01-21 10:54:09.958801: step: 2076/529, loss: 0.5666231513023376 2023-01-21 10:54:11.181357: step: 2080/529, loss: 0.043134406208992004 2023-01-21 10:54:12.369434: step: 2084/529, loss: 0.004335975740104914 2023-01-21 10:54:13.559327: step: 2088/529, loss: 0.005875587463378906 2023-01-21 10:54:14.738484: step: 2092/529, loss: 0.052643489092588425 2023-01-21 10:54:15.929013: step: 2096/529, loss: 0.08310270309448242 2023-01-21 10:54:17.129709: step: 2100/529, loss: 0.24817276000976562 2023-01-21 10:54:18.333813: step: 2104/529, loss: 0.030518721789121628 2023-01-21 10:54:19.523026: step: 2108/529, loss: 0.001597070717252791 2023-01-21 10:54:20.715692: step: 2112/529, loss: 0.10433516651391983 2023-01-21 10:54:21.891309: step: 2116/529, loss: 0.054807666689157486 ================================================== Loss: 0.080 -------------------- Dev: {'event': {'p': 0.5893037336024218, 'r': 0.7776298268974701, 'f1': 0.6704936854190586}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Test: {'event': {'p': 0.6291547958214625, 'r': 0.7938885560215698, 'f1': 0.7019867549668874}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Chinese: {'event': {'p': 0.5662650602409639, 'r': 0.8703703703703703, 'f1': 0.6861313868613138}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Korean: {'event': {'p': 0.5818181818181818, 'r': 0.5079365079365079, 'f1': 0.5423728813559321}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} 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': 11} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 12 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:55:06.567751: step: 4/529, loss: 0.09461308270692825 2023-01-21 10:55:07.788519: step: 8/529, loss: 0.015323924832046032 2023-01-21 10:55:09.002526: step: 12/529, loss: 0.050940513610839844 2023-01-21 10:55:10.167641: step: 16/529, loss: 0.06520853191614151 2023-01-21 10:55:11.309148: step: 20/529, loss: 0.19915488362312317 2023-01-21 10:55:12.471321: step: 24/529, loss: 0.0025695799849927425 2023-01-21 10:55:13.678199: step: 28/529, loss: 0.015318823046982288 2023-01-21 10:55:14.908613: step: 32/529, loss: 0.06139631196856499 2023-01-21 10:55:16.073005: step: 36/529, loss: 0.006859683897346258 2023-01-21 10:55:17.248656: step: 40/529, loss: 0.0289854034781456 2023-01-21 10:55:18.467732: step: 44/529, loss: 0.12239294499158859 2023-01-21 10:55:19.678297: step: 48/529, loss: 0.022112753242254257 2023-01-21 10:55:20.885969: step: 52/529, loss: 0.004193782806396484 2023-01-21 10:55:22.061159: step: 56/529, loss: 0.0712035596370697 2023-01-21 10:55:23.266413: step: 60/529, loss: 0.016473818570375443 2023-01-21 10:55:24.510948: step: 64/529, loss: 0.04520554840564728 2023-01-21 10:55:25.675370: step: 68/529, loss: 0.00995492935180664 2023-01-21 10:55:26.842320: step: 72/529, loss: 0.019214224070310593 2023-01-21 10:55:28.016886: step: 76/529, loss: 0.2561284005641937 2023-01-21 10:55:29.187956: step: 80/529, loss: 0.059084296226501465 2023-01-21 10:55:30.367868: step: 84/529, loss: 0.0697411596775055 2023-01-21 10:55:31.585164: step: 88/529, loss: 5.209192752838135 2023-01-21 10:55:32.796737: step: 92/529, loss: 0.06205282360315323 2023-01-21 10:55:33.989331: step: 96/529, loss: 0.027886293828487396 2023-01-21 10:55:35.166482: step: 100/529, loss: 0.010370826348662376 2023-01-21 10:55:36.343633: step: 104/529, loss: 0.006428814493119717 2023-01-21 10:55:37.522067: step: 108/529, loss: 0.10677352547645569 2023-01-21 10:55:38.744073: step: 112/529, loss: 0.13863268494606018 2023-01-21 10:55:39.899179: step: 116/529, loss: 0.09578684717416763 2023-01-21 10:55:41.093663: step: 120/529, loss: 0.020717715844511986 2023-01-21 10:55:42.290200: step: 124/529, loss: 0.014867973513901234 2023-01-21 10:55:43.510598: step: 128/529, loss: 0.06393737345933914 2023-01-21 10:55:44.717037: step: 132/529, loss: 0.054084040224552155 2023-01-21 10:55:45.926589: step: 136/529, loss: 0.01900775544345379 2023-01-21 10:55:47.097449: step: 140/529, loss: 0.0064868927001953125 2023-01-21 10:55:48.298078: step: 144/529, loss: 0.018323326483368874 2023-01-21 10:55:49.491453: step: 148/529, loss: 0.007177924737334251 2023-01-21 10:55:50.681496: step: 152/529, loss: 0.05831766128540039 2023-01-21 10:55:51.867640: step: 156/529, loss: 0.013208579272031784 2023-01-21 10:55:53.072774: step: 160/529, loss: 0.014266777783632278 2023-01-21 10:55:54.287950: step: 164/529, loss: 0.12816008925437927 2023-01-21 10:55:55.485225: step: 168/529, loss: 0.019687842577695847 2023-01-21 10:55:56.686838: step: 172/529, loss: 0.025074752047657967 2023-01-21 10:55:57.900978: step: 176/529, loss: 0.041172221302986145 2023-01-21 10:55:59.076406: step: 180/529, loss: 0.14975497126579285 2023-01-21 10:56:00.276516: step: 184/529, loss: 0.10053076595067978 2023-01-21 10:56:01.464211: step: 188/529, loss: 0.03291922062635422 2023-01-21 10:56:02.718373: step: 192/529, loss: 0.024123001843690872 2023-01-21 10:56:03.917569: step: 196/529, loss: 0.09668521583080292 2023-01-21 10:56:05.129475: step: 200/529, loss: 0.006633949466049671 2023-01-21 10:56:06.317949: step: 204/529, loss: 0.01633734628558159 2023-01-21 10:56:07.492604: step: 208/529, loss: 0.016956424340605736 2023-01-21 10:56:08.729155: step: 212/529, loss: 0.07556218653917313 2023-01-21 10:56:09.950631: step: 216/529, loss: 0.07795467972755432 2023-01-21 10:56:11.162725: step: 220/529, loss: 0.010879325680434704 2023-01-21 10:56:12.309776: step: 224/529, loss: 0.0018048763740807772 2023-01-21 10:56:13.489663: step: 228/529, loss: 0.0389968603849411 2023-01-21 10:56:14.694777: step: 232/529, loss: 0.016606902703642845 2023-01-21 10:56:15.909051: step: 236/529, loss: 0.0254350658506155 2023-01-21 10:56:17.175652: step: 240/529, loss: 0.21078383922576904 2023-01-21 10:56:18.395775: step: 244/529, loss: 0.11606959998607635 2023-01-21 10:56:19.546461: step: 248/529, loss: 0.022133732214570045 2023-01-21 10:56:20.716657: step: 252/529, loss: 0.06801538169384003 2023-01-21 10:56:21.876219: step: 256/529, loss: 0.0504608154296875 2023-01-21 10:56:23.062611: step: 260/529, loss: 0.07034663110971451 2023-01-21 10:56:24.260663: step: 264/529, loss: 0.03427903354167938 2023-01-21 10:56:25.466801: step: 268/529, loss: 0.0036689518019557 2023-01-21 10:56:26.666663: step: 272/529, loss: 0.20312373340129852 2023-01-21 10:56:27.874872: step: 276/529, loss: 0.06532511860132217 2023-01-21 10:56:29.086123: step: 280/529, loss: 0.003205204149708152 2023-01-21 10:56:30.283180: step: 284/529, loss: 0.24573364853858948 2023-01-21 10:56:31.459012: step: 288/529, loss: 0.040628623217344284 2023-01-21 10:56:32.679428: step: 292/529, loss: 0.07421312481164932 2023-01-21 10:56:33.845387: step: 296/529, loss: 0.014757443219423294 2023-01-21 10:56:35.009773: step: 300/529, loss: 0.030996274203062057 2023-01-21 10:56:36.225507: step: 304/529, loss: 0.01604442670941353 2023-01-21 10:56:37.396005: step: 308/529, loss: 0.022764777764678 2023-01-21 10:56:38.566238: step: 312/529, loss: 0.0264129638671875 2023-01-21 10:56:39.761436: step: 316/529, loss: 0.02089815028011799 2023-01-21 10:56:41.005693: step: 320/529, loss: 0.019779682159423828 2023-01-21 10:56:42.172679: step: 324/529, loss: 0.002960694022476673 2023-01-21 10:56:43.320271: step: 328/529, loss: 0.0029618265107274055 2023-01-21 10:56:44.554336: step: 332/529, loss: 0.05655193328857422 2023-01-21 10:56:45.739539: step: 336/529, loss: 0.030414726585149765 2023-01-21 10:56:46.908642: step: 340/529, loss: 0.19958622753620148 2023-01-21 10:56:48.121345: step: 344/529, loss: 0.06624484062194824 2023-01-21 10:56:49.339835: step: 348/529, loss: 0.03439150005578995 2023-01-21 10:56:50.553311: step: 352/529, loss: 0.07798977196216583 2023-01-21 10:56:51.776085: step: 356/529, loss: 0.2156461775302887 2023-01-21 10:56:53.007797: step: 360/529, loss: 0.010379552841186523 2023-01-21 10:56:54.173105: step: 364/529, loss: 0.01006460189819336 2023-01-21 10:56:55.387879: step: 368/529, loss: 0.017821503803133965 2023-01-21 10:56:56.584480: step: 372/529, loss: 0.006402206607162952 2023-01-21 10:56:57.762154: step: 376/529, loss: 0.0037534714210778475 2023-01-21 10:56:58.940746: step: 380/529, loss: 0.0677327886223793 2023-01-21 10:57:00.123394: step: 384/529, loss: 0.038045406341552734 2023-01-21 10:57:01.301052: step: 388/529, loss: 0.024000264704227448 2023-01-21 10:57:02.492230: step: 392/529, loss: 0.12070684134960175 2023-01-21 10:57:03.697708: step: 396/529, loss: 0.0043558599427342415 2023-01-21 10:57:04.899648: step: 400/529, loss: 0.0075893402099609375 2023-01-21 10:57:06.107098: step: 404/529, loss: 0.007226181216537952 2023-01-21 10:57:07.343744: step: 408/529, loss: 0.046715546399354935 2023-01-21 10:57:08.561181: step: 412/529, loss: 0.0024368285667151213 2023-01-21 10:57:09.765316: step: 416/529, loss: 0.02151932753622532 2023-01-21 10:57:10.950569: step: 420/529, loss: 0.0023601532448083162 2023-01-21 10:57:12.139657: step: 424/529, loss: 0.03203096613287926 2023-01-21 10:57:13.330478: step: 428/529, loss: 0.003885936690494418 2023-01-21 10:57:14.504186: step: 432/529, loss: 0.07177035510540009 2023-01-21 10:57:15.692430: step: 436/529, loss: 0.012391472235321999 2023-01-21 10:57:16.851587: step: 440/529, loss: 0.05973052978515625 2023-01-21 10:57:18.057970: step: 444/529, loss: 0.06989746540784836 2023-01-21 10:57:19.294211: step: 448/529, loss: 0.004841280169785023 2023-01-21 10:57:20.479170: step: 452/529, loss: 0.09497766941785812 2023-01-21 10:57:21.709881: step: 456/529, loss: 0.03509116172790527 2023-01-21 10:57:22.935938: step: 460/529, loss: 0.10291356593370438 2023-01-21 10:57:24.100982: step: 464/529, loss: 0.013241196051239967 2023-01-21 10:57:25.302809: step: 468/529, loss: 0.019420623779296875 2023-01-21 10:57:26.498011: step: 472/529, loss: 0.022420763969421387 2023-01-21 10:57:27.702488: step: 476/529, loss: 0.07654790580272675 2023-01-21 10:57:28.904017: step: 480/529, loss: 0.09972763061523438 2023-01-21 10:57:30.114233: step: 484/529, loss: 0.007969999685883522 2023-01-21 10:57:31.315863: step: 488/529, loss: 0.07571058720350266 2023-01-21 10:57:32.510967: step: 492/529, loss: 0.0266784206032753 2023-01-21 10:57:33.680346: step: 496/529, loss: 0.015119075775146484 2023-01-21 10:57:34.914306: step: 500/529, loss: 0.014446831308305264 2023-01-21 10:57:36.120304: step: 504/529, loss: 0.0249054916203022 2023-01-21 10:57:37.333575: step: 508/529, loss: 0.08558405935764313 2023-01-21 10:57:38.508029: step: 512/529, loss: 0.014135551638901234 2023-01-21 10:57:39.737292: step: 516/529, loss: 0.019354963675141335 2023-01-21 10:57:40.959444: step: 520/529, loss: 0.2923146188259125 2023-01-21 10:57:42.214894: step: 524/529, loss: 0.026810120791196823 2023-01-21 10:57:43.420642: step: 528/529, loss: 0.11260776966810226 2023-01-21 10:57:44.616039: step: 532/529, loss: 0.04780283197760582 2023-01-21 10:57:45.831037: step: 536/529, loss: 0.0428677573800087 2023-01-21 10:57:47.027481: step: 540/529, loss: 0.005142593290656805 2023-01-21 10:57:48.213884: step: 544/529, loss: 0.01918959617614746 2023-01-21 10:57:49.378772: step: 548/529, loss: 0.011128854937851429 2023-01-21 10:57:50.587335: step: 552/529, loss: 0.054070379585027695 2023-01-21 10:57:51.856982: step: 556/529, loss: 0.03426065668463707 2023-01-21 10:57:53.065348: step: 560/529, loss: 0.01594552956521511 2023-01-21 10:57:54.252425: step: 564/529, loss: 0.04134702682495117 2023-01-21 10:57:55.450310: step: 568/529, loss: 0.0501188300549984 2023-01-21 10:57:56.645151: step: 572/529, loss: 0.09141997992992401 2023-01-21 10:57:57.822067: step: 576/529, loss: 0.028127051889896393 2023-01-21 10:57:59.002833: step: 580/529, loss: 0.026869487017393112 2023-01-21 10:58:00.169883: step: 584/529, loss: 0.030240248888731003 2023-01-21 10:58:01.387513: step: 588/529, loss: 0.02818460389971733 2023-01-21 10:58:02.550709: step: 592/529, loss: 0.023508453741669655 2023-01-21 10:58:03.746830: step: 596/529, loss: 0.05320854112505913 2023-01-21 10:58:04.980881: step: 600/529, loss: 0.07365011423826218 2023-01-21 10:58:06.160753: step: 604/529, loss: 0.0428253635764122 2023-01-21 10:58:07.369139: step: 608/529, loss: 0.06732264161109924 2023-01-21 10:58:08.539210: step: 612/529, loss: 0.03494701161980629 2023-01-21 10:58:09.700302: step: 616/529, loss: 0.004226016812026501 2023-01-21 10:58:10.897870: step: 620/529, loss: 0.03827972337603569 2023-01-21 10:58:12.096883: step: 624/529, loss: 0.21730975806713104 2023-01-21 10:58:13.297836: step: 628/529, loss: 0.0005623817560262978 2023-01-21 10:58:14.551955: step: 632/529, loss: 0.07962741702795029 2023-01-21 10:58:15.769815: step: 636/529, loss: 0.044205572456121445 2023-01-21 10:58:16.979583: step: 640/529, loss: 0.05206260830163956 2023-01-21 10:58:18.175239: step: 644/529, loss: 0.11605257540941238 2023-01-21 10:58:19.356812: step: 648/529, loss: 0.011652708053588867 2023-01-21 10:58:20.573853: step: 652/529, loss: 0.056532762944698334 2023-01-21 10:58:21.754731: step: 656/529, loss: 0.01340098399668932 2023-01-21 10:58:22.951470: step: 660/529, loss: 0.07416544109582901 2023-01-21 10:58:24.128248: step: 664/529, loss: 0.00959248561412096 2023-01-21 10:58:25.310086: step: 668/529, loss: 0.07944647967815399 2023-01-21 10:58:26.474552: step: 672/529, loss: 0.01790943183004856 2023-01-21 10:58:27.730045: step: 676/529, loss: 0.4197238087654114 2023-01-21 10:58:28.927257: step: 680/529, loss: 0.003906631376594305 2023-01-21 10:58:30.113099: step: 684/529, loss: 0.02599181979894638 2023-01-21 10:58:31.326373: step: 688/529, loss: 0.013612094335258007 2023-01-21 10:58:32.539081: step: 692/529, loss: 0.005812168121337891 2023-01-21 10:58:33.796965: step: 696/529, loss: 0.0794641524553299 2023-01-21 10:58:34.977136: step: 700/529, loss: 0.007331180386245251 2023-01-21 10:58:36.211459: step: 704/529, loss: 0.0018688678974285722 2023-01-21 10:58:37.415939: step: 708/529, loss: 0.04258701950311661 2023-01-21 10:58:38.619643: step: 712/529, loss: 0.17342452704906464 2023-01-21 10:58:39.804667: step: 716/529, loss: 0.0268572811037302 2023-01-21 10:58:41.019209: step: 720/529, loss: 0.3645104765892029 2023-01-21 10:58:42.217038: step: 724/529, loss: 0.05303540453314781 2023-01-21 10:58:43.438081: step: 728/529, loss: 0.01623992994427681 2023-01-21 10:58:44.655713: step: 732/529, loss: 0.17442341148853302 2023-01-21 10:58:45.842190: step: 736/529, loss: 0.05026240646839142 2023-01-21 10:58:47.056165: step: 740/529, loss: 0.06288299709558487 2023-01-21 10:58:48.294604: step: 744/529, loss: 0.0065247537568211555 2023-01-21 10:58:49.504263: step: 748/529, loss: 0.04794654995203018 2023-01-21 10:58:50.654510: step: 752/529, loss: 0.014880466274917126 2023-01-21 10:58:51.878451: step: 756/529, loss: 0.037469103932380676 2023-01-21 10:58:53.094418: step: 760/529, loss: 0.018707560375332832 2023-01-21 10:58:54.311725: step: 764/529, loss: 0.019561195746064186 2023-01-21 10:58:55.485892: step: 768/529, loss: 0.0009921550517901778 2023-01-21 10:58:56.735747: step: 772/529, loss: 0.25392797589302063 2023-01-21 10:58:57.945068: step: 776/529, loss: 0.33541297912597656 2023-01-21 10:58:59.133743: step: 780/529, loss: 0.01707172393798828 2023-01-21 10:59:00.331370: step: 784/529, loss: 0.02098911441862583 2023-01-21 10:59:01.581676: step: 788/529, loss: 0.3321826756000519 2023-01-21 10:59:02.808828: step: 792/529, loss: 0.3049025535583496 2023-01-21 10:59:03.997115: step: 796/529, loss: 0.10816683620214462 2023-01-21 10:59:05.136009: step: 800/529, loss: 0.025606347247958183 2023-01-21 10:59:06.331443: step: 804/529, loss: 0.02271747589111328 2023-01-21 10:59:07.569132: step: 808/529, loss: 0.010996055789291859 2023-01-21 10:59:08.770553: step: 812/529, loss: 0.006251621060073376 2023-01-21 10:59:09.933477: step: 816/529, loss: 0.013600422069430351 2023-01-21 10:59:11.143969: step: 820/529, loss: 0.019065190106630325 2023-01-21 10:59:12.344699: step: 824/529, loss: 0.04062776640057564 2023-01-21 10:59:13.529785: step: 828/529, loss: 0.019218921661376953 2023-01-21 10:59:14.706629: step: 832/529, loss: 0.013275337405502796 2023-01-21 10:59:15.917443: step: 836/529, loss: 0.012666894122958183 2023-01-21 10:59:17.117570: step: 840/529, loss: 0.0011449814774096012 2023-01-21 10:59:18.325131: step: 844/529, loss: 0.031485844403505325 2023-01-21 10:59:19.522568: step: 848/529, loss: 0.02624540403485298 2023-01-21 10:59:20.738286: step: 852/529, loss: 0.009102249518036842 2023-01-21 10:59:21.931384: step: 856/529, loss: 0.06545954197645187 2023-01-21 10:59:23.128164: step: 860/529, loss: 0.321106493473053 2023-01-21 10:59:24.312950: step: 864/529, loss: 0.028829121962189674 2023-01-21 10:59:25.560242: step: 868/529, loss: 0.021967221051454544 2023-01-21 10:59:26.777475: step: 872/529, loss: 0.061124518513679504 2023-01-21 10:59:27.977196: step: 876/529, loss: 0.06485958397388458 2023-01-21 10:59:29.199753: step: 880/529, loss: 0.022169413045048714 2023-01-21 10:59:30.397958: step: 884/529, loss: 0.005156898405402899 2023-01-21 10:59:31.696637: step: 888/529, loss: 0.03813190385699272 2023-01-21 10:59:32.908499: step: 892/529, loss: 0.0007540702936239541 2023-01-21 10:59:34.147858: step: 896/529, loss: 0.03882799297571182 2023-01-21 10:59:35.335493: step: 900/529, loss: 0.0030347825959324837 2023-01-21 10:59:36.558504: step: 904/529, loss: 0.013276767916977406 2023-01-21 10:59:37.750826: step: 908/529, loss: 0.06771676242351532 2023-01-21 10:59:38.954216: step: 912/529, loss: 0.012628268450498581 2023-01-21 10:59:40.121054: step: 916/529, loss: 0.11901235580444336 2023-01-21 10:59:41.311084: step: 920/529, loss: 0.3809569478034973 2023-01-21 10:59:42.486595: step: 924/529, loss: 0.030087757855653763 2023-01-21 10:59:43.729473: step: 928/529, loss: 0.17925891280174255 2023-01-21 10:59:44.926354: step: 932/529, loss: 0.018484115600585938 2023-01-21 10:59:46.147590: step: 936/529, loss: 0.04119310528039932 2023-01-21 10:59:47.324158: step: 940/529, loss: 0.05986042320728302 2023-01-21 10:59:48.579381: step: 944/529, loss: 0.046485330909490585 2023-01-21 10:59:49.777572: step: 948/529, loss: 0.004480266477912664 2023-01-21 10:59:50.946101: step: 952/529, loss: 0.05463600158691406 2023-01-21 10:59:52.166834: step: 956/529, loss: 0.0003727913135662675 2023-01-21 10:59:53.335541: step: 960/529, loss: 0.0339718796312809 2023-01-21 10:59:54.504667: step: 964/529, loss: 1.669961929321289 2023-01-21 10:59:55.694406: step: 968/529, loss: 0.006590652279555798 2023-01-21 10:59:56.912723: step: 972/529, loss: 0.052147865295410156 2023-01-21 10:59:58.101716: step: 976/529, loss: 0.011987876147031784 2023-01-21 10:59:59.280406: step: 980/529, loss: 0.026206208392977715 2023-01-21 11:00:00.463852: step: 984/529, loss: 0.031342603266239166 2023-01-21 11:00:01.661659: step: 988/529, loss: 0.03202629089355469 2023-01-21 11:00:02.848059: step: 992/529, loss: 0.030991649255156517 2023-01-21 11:00:04.055849: step: 996/529, loss: 0.026134680956602097 2023-01-21 11:00:05.262019: step: 1000/529, loss: 0.05096714198589325 2023-01-21 11:00:06.489241: step: 1004/529, loss: 0.2423064261674881 2023-01-21 11:00:07.678735: step: 1008/529, loss: 0.013319539837539196 2023-01-21 11:00:08.857279: step: 1012/529, loss: 0.07091889530420303 2023-01-21 11:00:10.017291: step: 1016/529, loss: 0.0694112777709961 2023-01-21 11:00:11.188527: step: 1020/529, loss: 0.11578676104545593 2023-01-21 11:00:12.424537: step: 1024/529, loss: 0.15999144315719604 2023-01-21 11:00:13.609148: step: 1028/529, loss: 0.006967830937355757 2023-01-21 11:00:14.824209: step: 1032/529, loss: 0.05098381265997887 2023-01-21 11:00:16.064102: step: 1036/529, loss: 0.08201351016759872 2023-01-21 11:00:17.282375: step: 1040/529, loss: 0.03741617500782013 2023-01-21 11:00:18.520368: step: 1044/529, loss: 0.017806481570005417 2023-01-21 11:00:19.726803: step: 1048/529, loss: 0.08050355315208435 2023-01-21 11:00:20.921672: step: 1052/529, loss: 0.5405857563018799 2023-01-21 11:00:22.081975: step: 1056/529, loss: 0.11757411807775497 2023-01-21 11:00:23.329061: step: 1060/529, loss: 0.21716323494911194 2023-01-21 11:00:24.520019: step: 1064/529, loss: 0.03898058086633682 2023-01-21 11:00:25.723573: step: 1068/529, loss: 0.017492199316620827 2023-01-21 11:00:26.919667: step: 1072/529, loss: 0.025894690304994583 2023-01-21 11:00:28.097530: step: 1076/529, loss: 0.018796445801854134 2023-01-21 11:00:29.325364: step: 1080/529, loss: 0.03311324119567871 2023-01-21 11:00:30.489436: step: 1084/529, loss: 0.06111498177051544 2023-01-21 11:00:31.730427: step: 1088/529, loss: 0.006524813361465931 2023-01-21 11:00:32.938911: step: 1092/529, loss: 0.043233491480350494 2023-01-21 11:00:34.102118: step: 1096/529, loss: 0.070864237844944 2023-01-21 11:00:35.287745: step: 1100/529, loss: 0.007418537046760321 2023-01-21 11:00:36.466960: step: 1104/529, loss: 0.0047269826754927635 2023-01-21 11:00:37.654188: step: 1108/529, loss: 0.022239113226532936 2023-01-21 11:00:38.859214: step: 1112/529, loss: 0.03437929227948189 2023-01-21 11:00:40.082435: step: 1116/529, loss: 0.05128903687000275 2023-01-21 11:00:41.271059: step: 1120/529, loss: 0.03506822511553764 2023-01-21 11:00:42.446460: step: 1124/529, loss: 0.024320315569639206 2023-01-21 11:00:43.669422: step: 1128/529, loss: 0.054823778569698334 2023-01-21 11:00:44.847910: step: 1132/529, loss: 0.06581497192382812 2023-01-21 11:00:46.051602: step: 1136/529, loss: 0.041820526123046875 2023-01-21 11:00:47.251829: step: 1140/529, loss: 0.12122364342212677 2023-01-21 11:00:48.425527: step: 1144/529, loss: 0.031093694269657135 2023-01-21 11:00:49.595988: step: 1148/529, loss: 0.8840776681900024 2023-01-21 11:00:50.789486: step: 1152/529, loss: 0.10295772552490234 2023-01-21 11:00:51.989854: step: 1156/529, loss: 0.01616945117712021 2023-01-21 11:00:53.194191: step: 1160/529, loss: 0.05126610025763512 2023-01-21 11:00:54.448697: step: 1164/529, loss: 0.0327458381652832 2023-01-21 11:00:55.617021: step: 1168/529, loss: 0.05950140953063965 2023-01-21 11:00:56.823378: step: 1172/529, loss: 0.09811064600944519 2023-01-21 11:00:58.020684: step: 1176/529, loss: 0.017319535836577415 2023-01-21 11:00:59.205100: step: 1180/529, loss: 0.013949966058135033 2023-01-21 11:01:00.422285: step: 1184/529, loss: 0.021736908704042435 2023-01-21 11:01:01.627285: step: 1188/529, loss: 0.008223820477724075 2023-01-21 11:01:02.825755: step: 1192/529, loss: 0.026146888732910156 2023-01-21 11:01:04.017861: step: 1196/529, loss: 0.07489853352308273 2023-01-21 11:01:05.229861: step: 1200/529, loss: 0.28294438123703003 2023-01-21 11:01:06.419438: step: 1204/529, loss: 0.09280635416507721 2023-01-21 11:01:07.607220: step: 1208/529, loss: 0.05431985855102539 2023-01-21 11:01:08.843724: step: 1212/529, loss: 0.015847016125917435 2023-01-21 11:01:10.076695: step: 1216/529, loss: 0.015099907293915749 2023-01-21 11:01:11.231527: step: 1220/529, loss: 0.018815994262695312 2023-01-21 11:01:12.421147: step: 1224/529, loss: 0.19454345107078552 2023-01-21 11:01:13.665719: step: 1228/529, loss: 0.03185071796178818 2023-01-21 11:01:14.831043: step: 1232/529, loss: 0.08119087666273117 2023-01-21 11:01:16.055769: step: 1236/529, loss: 0.028986692428588867 2023-01-21 11:01:17.241291: step: 1240/529, loss: 0.08334074169397354 2023-01-21 11:01:18.462515: step: 1244/529, loss: 0.0006021976005285978 2023-01-21 11:01:19.621238: step: 1248/529, loss: 0.017824841663241386 2023-01-21 11:01:20.776423: step: 1252/529, loss: 0.019753120839595795 2023-01-21 11:01:21.996996: step: 1256/529, loss: 0.057345200330019 2023-01-21 11:01:23.193097: step: 1260/529, loss: 0.08658619225025177 2023-01-21 11:01:24.384558: step: 1264/529, loss: 0.016444897279143333 2023-01-21 11:01:25.619856: step: 1268/529, loss: 0.01755542680621147 2023-01-21 11:01:26.827597: step: 1272/529, loss: 0.018363334238529205 2023-01-21 11:01:28.059389: step: 1276/529, loss: 0.05356493219733238 2023-01-21 11:01:29.224039: step: 1280/529, loss: 0.021189548075199127 2023-01-21 11:01:30.520158: step: 1284/529, loss: 0.07716875523328781 2023-01-21 11:01:31.751149: step: 1288/529, loss: 0.030627060681581497 2023-01-21 11:01:32.975061: step: 1292/529, loss: 0.04808850213885307 2023-01-21 11:01:34.158497: step: 1296/529, loss: 0.05766897276043892 2023-01-21 11:01:35.328899: step: 1300/529, loss: 0.1416155844926834 2023-01-21 11:01:36.525628: step: 1304/529, loss: 0.06356658786535263 2023-01-21 11:01:37.727246: step: 1308/529, loss: 0.006972694303840399 2023-01-21 11:01:38.926923: step: 1312/529, loss: 0.0775662362575531 2023-01-21 11:01:40.137004: step: 1316/529, loss: 0.021193789318203926 2023-01-21 11:01:41.309631: step: 1320/529, loss: 0.006855368614196777 2023-01-21 11:01:42.499687: step: 1324/529, loss: 0.05127792805433273 2023-01-21 11:01:43.715073: step: 1328/529, loss: 0.054251864552497864 2023-01-21 11:01:44.912642: step: 1332/529, loss: 0.016922971233725548 2023-01-21 11:01:46.074219: step: 1336/529, loss: 0.008650684729218483 2023-01-21 11:01:47.267909: step: 1340/529, loss: 0.004607868380844593 2023-01-21 11:01:48.490603: step: 1344/529, loss: 0.12435702979564667 2023-01-21 11:01:49.685467: step: 1348/529, loss: 0.015988444909453392 2023-01-21 11:01:50.845576: step: 1352/529, loss: 0.015291977673768997 2023-01-21 11:01:52.068425: step: 1356/529, loss: 0.03930168226361275 2023-01-21 11:01:53.289892: step: 1360/529, loss: 0.20888499915599823 2023-01-21 11:01:54.477885: step: 1364/529, loss: 0.03146400302648544 2023-01-21 11:01:55.733336: step: 1368/529, loss: 0.0033580781891942024 2023-01-21 11:01:56.980441: step: 1372/529, loss: 0.013326454907655716 2023-01-21 11:01:58.174750: step: 1376/529, loss: 0.013088989071547985 2023-01-21 11:01:59.349119: step: 1380/529, loss: 0.528385579586029 2023-01-21 11:02:00.545520: step: 1384/529, loss: 0.03378696367144585 2023-01-21 11:02:01.724851: step: 1388/529, loss: 0.006578922271728516 2023-01-21 11:02:02.950837: step: 1392/529, loss: 0.1777210235595703 2023-01-21 11:02:04.101477: step: 1396/529, loss: 0.04436364397406578 2023-01-21 11:02:05.357596: step: 1400/529, loss: 0.08823862671852112 2023-01-21 11:02:06.519119: step: 1404/529, loss: 0.15832966566085815 2023-01-21 11:02:07.738747: step: 1408/529, loss: 0.0797787681221962 2023-01-21 11:02:08.936615: step: 1412/529, loss: 0.0033970833756029606 2023-01-21 11:02:10.168753: step: 1416/529, loss: 0.059432413429021835 2023-01-21 11:02:11.379599: step: 1420/529, loss: 0.06300163269042969 2023-01-21 11:02:12.563738: step: 1424/529, loss: 0.09147863835096359 2023-01-21 11:02:13.756519: step: 1428/529, loss: 0.07947196811437607 2023-01-21 11:02:14.938174: step: 1432/529, loss: 0.02444438822567463 2023-01-21 11:02:16.129167: step: 1436/529, loss: 0.07819309830665588 2023-01-21 11:02:17.302997: step: 1440/529, loss: 0.009305763989686966 2023-01-21 11:02:18.487347: step: 1444/529, loss: 0.037786152213811874 2023-01-21 11:02:19.704345: step: 1448/529, loss: 0.16643677651882172 2023-01-21 11:02:20.880489: step: 1452/529, loss: 0.008370399475097656 2023-01-21 11:02:22.087003: step: 1456/529, loss: 0.04152803495526314 2023-01-21 11:02:23.263270: step: 1460/529, loss: 0.03420862928032875 2023-01-21 11:02:24.444198: step: 1464/529, loss: 0.006215715780854225 2023-01-21 11:02:25.634136: step: 1468/529, loss: 0.11815135180950165 2023-01-21 11:02:26.833203: step: 1472/529, loss: 0.08371029049158096 2023-01-21 11:02:28.054231: step: 1476/529, loss: 0.07180075347423553 2023-01-21 11:02:29.228809: step: 1480/529, loss: 0.08303123712539673 2023-01-21 11:02:30.430686: step: 1484/529, loss: 0.11439651995897293 2023-01-21 11:02:31.649511: step: 1488/529, loss: 0.13230475783348083 2023-01-21 11:02:32.838982: step: 1492/529, loss: 0.03144483640789986 2023-01-21 11:02:34.024290: step: 1496/529, loss: 0.121607206761837 2023-01-21 11:02:35.232633: step: 1500/529, loss: 0.030336331576108932 2023-01-21 11:02:36.410093: step: 1504/529, loss: 0.026038551703095436 2023-01-21 11:02:37.597712: step: 1508/529, loss: 0.13881778717041016 2023-01-21 11:02:38.766061: step: 1512/529, loss: 0.011060286313295364 2023-01-21 11:02:39.979891: step: 1516/529, loss: 0.022604942321777344 2023-01-21 11:02:41.151021: step: 1520/529, loss: 0.03933410719037056 2023-01-21 11:02:42.344324: step: 1524/529, loss: 0.007920932956039906 2023-01-21 11:02:43.544679: step: 1528/529, loss: 0.03963327407836914 2023-01-21 11:02:44.794300: step: 1532/529, loss: 0.05646514892578125 2023-01-21 11:02:45.976776: step: 1536/529, loss: 0.02344970777630806 2023-01-21 11:02:47.197376: step: 1540/529, loss: 0.039815474301576614 2023-01-21 11:02:48.361682: step: 1544/529, loss: 0.045206259936094284 2023-01-21 11:02:49.582488: step: 1548/529, loss: 0.37295645475387573 2023-01-21 11:02:50.807468: step: 1552/529, loss: 0.009088659659028053 2023-01-21 11:02:52.007459: step: 1556/529, loss: 0.06314563751220703 2023-01-21 11:02:53.187634: step: 1560/529, loss: 0.0035879136994481087 2023-01-21 11:02:54.381670: step: 1564/529, loss: 0.01016540452837944 2023-01-21 11:02:55.567813: step: 1568/529, loss: 0.03444109112024307 2023-01-21 11:02:56.785083: step: 1572/529, loss: 0.04573829472064972 2023-01-21 11:02:57.954947: step: 1576/529, loss: 0.0017147064208984375 2023-01-21 11:02:59.130931: step: 1580/529, loss: 0.19392293691635132 2023-01-21 11:03:00.316986: step: 1584/529, loss: 0.08233385533094406 2023-01-21 11:03:01.555248: step: 1588/529, loss: 0.8740646243095398 2023-01-21 11:03:02.758850: step: 1592/529, loss: 0.31442394852638245 2023-01-21 11:03:03.986073: step: 1596/529, loss: 0.055511474609375 2023-01-21 11:03:05.173007: step: 1600/529, loss: 0.05942859873175621 2023-01-21 11:03:06.349816: step: 1604/529, loss: 0.12933655083179474 2023-01-21 11:03:07.548079: step: 1608/529, loss: 0.061638880521059036 2023-01-21 11:03:08.714218: step: 1612/529, loss: 0.035477831959724426 2023-01-21 11:03:09.852316: step: 1616/529, loss: 0.037639666348695755 2023-01-21 11:03:11.038742: step: 1620/529, loss: 0.005637597758322954 2023-01-21 11:03:12.245380: step: 1624/529, loss: 0.04678850248456001 2023-01-21 11:03:13.474200: step: 1628/529, loss: 0.016644954681396484 2023-01-21 11:03:14.668536: step: 1632/529, loss: 0.035111334174871445 2023-01-21 11:03:15.844152: step: 1636/529, loss: 0.02014904096722603 2023-01-21 11:03:17.038837: step: 1640/529, loss: 0.031966306269168854 2023-01-21 11:03:18.265325: step: 1644/529, loss: 0.062006283551454544 2023-01-21 11:03:19.467562: step: 1648/529, loss: 0.05241985619068146 2023-01-21 11:03:20.708874: step: 1652/529, loss: 0.0505254752933979 2023-01-21 11:03:21.902156: step: 1656/529, loss: 0.004819107241928577 2023-01-21 11:03:23.080822: step: 1660/529, loss: 0.06722517311573029 2023-01-21 11:03:24.279056: step: 1664/529, loss: 0.007700538262724876 2023-01-21 11:03:25.438888: step: 1668/529, loss: 0.020280171185731888 2023-01-21 11:03:26.642266: step: 1672/529, loss: 0.08656089007854462 2023-01-21 11:03:27.830037: step: 1676/529, loss: 0.051488880068063736 2023-01-21 11:03:29.055014: step: 1680/529, loss: 0.025290392339229584 2023-01-21 11:03:30.222572: step: 1684/529, loss: 0.039223767817020416 2023-01-21 11:03:31.363994: step: 1688/529, loss: 0.025942638516426086 2023-01-21 11:03:32.579479: step: 1692/529, loss: 0.04711208492517471 2023-01-21 11:03:33.755961: step: 1696/529, loss: 0.012575579807162285 2023-01-21 11:03:34.985095: step: 1700/529, loss: 0.07909746468067169 2023-01-21 11:03:36.149884: step: 1704/529, loss: 0.0576418861746788 2023-01-21 11:03:37.374207: step: 1708/529, loss: 0.0006843090523034334 2023-01-21 11:03:38.539381: step: 1712/529, loss: 0.05928444862365723 2023-01-21 11:03:39.753108: step: 1716/529, loss: 0.009669494815170765 2023-01-21 11:03:40.981553: step: 1720/529, loss: 0.058765411376953125 2023-01-21 11:03:42.156326: step: 1724/529, loss: 0.03457631915807724 2023-01-21 11:03:43.352183: step: 1728/529, loss: 0.2439233809709549 2023-01-21 11:03:44.527053: step: 1732/529, loss: 0.07965365052223206 2023-01-21 11:03:45.707232: step: 1736/529, loss: 0.06890296936035156 2023-01-21 11:03:46.909416: step: 1740/529, loss: 0.1105751022696495 2023-01-21 11:03:48.112369: step: 1744/529, loss: 0.036740779876708984 2023-01-21 11:03:49.279623: step: 1748/529, loss: 0.03487520292401314 2023-01-21 11:03:50.496586: step: 1752/529, loss: 0.0848359614610672 2023-01-21 11:03:51.711047: step: 1756/529, loss: 0.024629021063447 2023-01-21 11:03:52.906313: step: 1760/529, loss: 0.06233549118041992 2023-01-21 11:03:54.142042: step: 1764/529, loss: 0.022668268531560898 2023-01-21 11:03:55.348560: step: 1768/529, loss: 0.031013870611786842 2023-01-21 11:03:56.557098: step: 1772/529, loss: 0.08588772267103195 2023-01-21 11:03:57.747075: step: 1776/529, loss: 0.019756412133574486 2023-01-21 11:03:58.931004: step: 1780/529, loss: 0.03923831135034561 2023-01-21 11:04:00.107567: step: 1784/529, loss: 0.04715175926685333 2023-01-21 11:04:01.339134: step: 1788/529, loss: 0.03348913416266441 2023-01-21 11:04:02.563047: step: 1792/529, loss: 0.24279536306858063 2023-01-21 11:04:03.754241: step: 1796/529, loss: 0.019339561462402344 2023-01-21 11:04:04.926906: step: 1800/529, loss: 0.13360892236232758 2023-01-21 11:04:06.132058: step: 1804/529, loss: 0.02291727066040039 2023-01-21 11:04:07.349133: step: 1808/529, loss: 0.03660411760210991 2023-01-21 11:04:08.589217: step: 1812/529, loss: 0.07315082848072052 2023-01-21 11:04:09.864381: step: 1816/529, loss: 0.037412360310554504 2023-01-21 11:04:11.102503: step: 1820/529, loss: 0.006944465916603804 2023-01-21 11:04:12.339315: step: 1824/529, loss: 0.06970392167568207 2023-01-21 11:04:13.537030: step: 1828/529, loss: 0.01299066562205553 2023-01-21 11:04:14.729894: step: 1832/529, loss: 0.07199078053236008 2023-01-21 11:04:15.924560: step: 1836/529, loss: 0.011887121014297009 2023-01-21 11:04:17.155880: step: 1840/529, loss: 0.00045189858064986765 2023-01-21 11:04:18.358878: step: 1844/529, loss: 0.02391376718878746 2023-01-21 11:04:19.590669: step: 1848/529, loss: 0.23020629584789276 2023-01-21 11:04:20.787539: step: 1852/529, loss: 0.012273884378373623 2023-01-21 11:04:21.992034: step: 1856/529, loss: 0.005406570620834827 2023-01-21 11:04:23.200234: step: 1860/529, loss: 0.03417768329381943 2023-01-21 11:04:24.386352: step: 1864/529, loss: 0.052526094019412994 2023-01-21 11:04:25.570155: step: 1868/529, loss: 0.009327221661806107 2023-01-21 11:04:26.842249: step: 1872/529, loss: 0.30849599838256836 2023-01-21 11:04:28.051219: step: 1876/529, loss: 0.04136858135461807 2023-01-21 11:04:29.222071: step: 1880/529, loss: 0.023511648178100586 2023-01-21 11:04:30.432936: step: 1884/529, loss: 0.007583570666611195 2023-01-21 11:04:31.613370: step: 1888/529, loss: 0.06826065480709076 2023-01-21 11:04:32.813934: step: 1892/529, loss: 0.06730309128761292 2023-01-21 11:04:33.995752: step: 1896/529, loss: 0.016716431826353073 2023-01-21 11:04:35.203690: step: 1900/529, loss: 0.014785576611757278 2023-01-21 11:04:36.391675: step: 1904/529, loss: 0.04808998107910156 2023-01-21 11:04:37.573836: step: 1908/529, loss: 0.03860732913017273 2023-01-21 11:04:38.748556: step: 1912/529, loss: 0.00348491664044559 2023-01-21 11:04:39.932651: step: 1916/529, loss: 0.005905247293412685 2023-01-21 11:04:41.140856: step: 1920/529, loss: 0.015135146677494049 2023-01-21 11:04:42.321539: step: 1924/529, loss: 0.024395275861024857 2023-01-21 11:04:43.511474: step: 1928/529, loss: 0.008846092969179153 2023-01-21 11:04:44.731903: step: 1932/529, loss: 0.03268633037805557 2023-01-21 11:04:45.941055: step: 1936/529, loss: 0.24961014091968536 2023-01-21 11:04:47.191018: step: 1940/529, loss: 0.007929516024887562 2023-01-21 11:04:48.372922: step: 1944/529, loss: 0.05357445776462555 2023-01-21 11:04:49.550023: step: 1948/529, loss: 0.029566098004579544 2023-01-21 11:04:50.765750: step: 1952/529, loss: 0.008351707831025124 2023-01-21 11:04:51.972980: step: 1956/529, loss: 0.008855628781020641 2023-01-21 11:04:53.168080: step: 1960/529, loss: 0.10262642055749893 2023-01-21 11:04:54.391856: step: 1964/529, loss: 0.05252294987440109 2023-01-21 11:04:55.599986: step: 1968/529, loss: 0.004825973883271217 2023-01-21 11:04:56.782609: step: 1972/529, loss: 0.00816421490162611 2023-01-21 11:04:57.983700: step: 1976/529, loss: 0.06373405456542969 2023-01-21 11:04:59.174390: step: 1980/529, loss: 0.016758346930146217 2023-01-21 11:05:00.383548: step: 1984/529, loss: 0.034069158136844635 2023-01-21 11:05:01.589631: step: 1988/529, loss: 0.06388463824987411 2023-01-21 11:05:02.799015: step: 1992/529, loss: 0.015813924372196198 2023-01-21 11:05:04.030476: step: 1996/529, loss: 0.024594688788056374 2023-01-21 11:05:05.247374: step: 2000/529, loss: 0.014341259375214577 2023-01-21 11:05:06.397958: step: 2004/529, loss: 0.028731895610690117 2023-01-21 11:05:07.593024: step: 2008/529, loss: 0.013081742450594902 2023-01-21 11:05:08.818045: step: 2012/529, loss: 0.05651276186108589 2023-01-21 11:05:10.022292: step: 2016/529, loss: 0.025107288733124733 2023-01-21 11:05:11.199271: step: 2020/529, loss: 0.019038867205381393 2023-01-21 11:05:12.386986: step: 2024/529, loss: 0.0070785521529614925 2023-01-21 11:05:13.606328: step: 2028/529, loss: 0.04467587545514107 2023-01-21 11:05:14.774262: step: 2032/529, loss: 0.0174710750579834 2023-01-21 11:05:15.951657: step: 2036/529, loss: 0.15973225235939026 2023-01-21 11:05:17.126017: step: 2040/529, loss: 0.018823623657226562 2023-01-21 11:05:18.303465: step: 2044/529, loss: 0.008092689327895641 2023-01-21 11:05:19.484013: step: 2048/529, loss: 0.04268283769488335 2023-01-21 11:05:20.708798: step: 2052/529, loss: 0.004303073976188898 2023-01-21 11:05:21.903903: step: 2056/529, loss: 0.03626861423254013 2023-01-21 11:05:23.114482: step: 2060/529, loss: 0.030824948102235794 2023-01-21 11:05:24.290917: step: 2064/529, loss: 0.077679343521595 2023-01-21 11:05:25.507423: step: 2068/529, loss: 0.021583938971161842 2023-01-21 11:05:26.778604: step: 2072/529, loss: 0.0035893439780920744 2023-01-21 11:05:27.987974: step: 2076/529, loss: 0.027521992102265358 2023-01-21 11:05:29.207401: step: 2080/529, loss: 0.0003895759582519531 2023-01-21 11:05:30.381687: step: 2084/529, loss: 0.008096123114228249 2023-01-21 11:05:31.560299: step: 2088/529, loss: 0.05906124413013458 2023-01-21 11:05:32.728110: step: 2092/529, loss: 0.026816463097929955 2023-01-21 11:05:33.925027: step: 2096/529, loss: 0.047417543828487396 2023-01-21 11:05:35.181986: step: 2100/529, loss: 0.03231191635131836 2023-01-21 11:05:36.367880: step: 2104/529, loss: 0.08727569878101349 2023-01-21 11:05:37.585509: step: 2108/529, loss: 0.10679206252098083 2023-01-21 11:05:38.756431: step: 2112/529, loss: 0.03921813890337944 2023-01-21 11:05:39.946556: step: 2116/529, loss: 0.046384263783693314 ================================================== Loss: 0.072 -------------------- Dev: {'event': {'p': 0.5710144927536231, 'r': 0.7869507323568575, 'f1': 0.6618141097424413}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Test: {'event': {'p': 0.6354166666666666, 'r': 0.8040742959856201, 'f1': 0.7098651150489288}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Chinese: {'event': {'p': 0.5161290322580645, 'r': 0.8888888888888888, 'f1': 0.6530612244897959}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} 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': 12} Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 13 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:06:24.881688: step: 4/529, loss: 0.03589210286736488 2023-01-21 11:06:26.068707: step: 8/529, loss: 0.04646158218383789 2023-01-21 11:06:27.264051: step: 12/529, loss: 0.0015617371536791325 2023-01-21 11:06:28.474362: step: 16/529, loss: 0.009456157684326172 2023-01-21 11:06:29.685908: step: 20/529, loss: 0.036652661859989166 2023-01-21 11:06:30.889254: step: 24/529, loss: 0.07016907632350922 2023-01-21 11:06:32.049915: step: 28/529, loss: 0.007915496826171875 2023-01-21 11:06:33.267699: step: 32/529, loss: 0.04337654262781143 2023-01-21 11:06:34.446906: step: 36/529, loss: 0.0009256363264285028 2023-01-21 11:06:35.670307: step: 40/529, loss: 0.08891038596630096 2023-01-21 11:06:36.889175: step: 44/529, loss: 0.06703028827905655 2023-01-21 11:06:38.070210: step: 48/529, loss: 0.03881406784057617 2023-01-21 11:06:39.247387: step: 52/529, loss: 0.05977373197674751 2023-01-21 11:06:40.455850: step: 56/529, loss: 0.0228405948728323 2023-01-21 11:06:41.676380: step: 60/529, loss: 0.04689016565680504 2023-01-21 11:06:42.887677: step: 64/529, loss: 0.04889635741710663 2023-01-21 11:06:44.062224: step: 68/529, loss: 0.09251828491687775 2023-01-21 11:06:45.232370: step: 72/529, loss: 0.06995969265699387 2023-01-21 11:06:46.423396: step: 76/529, loss: 0.042516373097896576 2023-01-21 11:06:47.634527: step: 80/529, loss: 1.0364328622817993 2023-01-21 11:06:48.841983: step: 84/529, loss: 0.006809329614043236 2023-01-21 11:06:50.054891: step: 88/529, loss: 0.1515457183122635 2023-01-21 11:06:51.296545: step: 92/529, loss: 0.01153254508972168 2023-01-21 11:06:52.527222: step: 96/529, loss: 0.04926338046789169 2023-01-21 11:06:53.718133: step: 100/529, loss: 0.02107715606689453 2023-01-21 11:06:54.944131: step: 104/529, loss: 0.04326820373535156 2023-01-21 11:06:56.145542: step: 108/529, loss: 0.04518928378820419 2023-01-21 11:06:57.371910: step: 112/529, loss: 0.22474442422389984 2023-01-21 11:06:58.546006: step: 116/529, loss: 0.043840885162353516 2023-01-21 11:06:59.735574: step: 120/529, loss: 0.04966764152050018 2023-01-21 11:07:00.901051: step: 124/529, loss: 0.025442887097597122 2023-01-21 11:07:02.139715: step: 128/529, loss: 0.030176257714629173 2023-01-21 11:07:03.386590: step: 132/529, loss: 0.02025279961526394 2023-01-21 11:07:04.583144: step: 136/529, loss: 0.029550839215517044 2023-01-21 11:07:05.764829: step: 140/529, loss: 0.017325783148407936 2023-01-21 11:07:06.949373: step: 144/529, loss: 0.013940048404037952 2023-01-21 11:07:08.141364: step: 148/529, loss: 0.02227191999554634 2023-01-21 11:07:09.332929: step: 152/529, loss: 0.021846294403076172 2023-01-21 11:07:10.515029: step: 156/529, loss: 0.0671592727303505 2023-01-21 11:07:11.701861: step: 160/529, loss: 0.02972097508609295 2023-01-21 11:07:12.933708: step: 164/529, loss: 0.023504257202148438 2023-01-21 11:07:14.106310: step: 168/529, loss: 0.05240507051348686 2023-01-21 11:07:15.290708: step: 172/529, loss: 0.012106704525649548 2023-01-21 11:07:16.441276: step: 176/529, loss: 0.00878896750509739 2023-01-21 11:07:17.627950: step: 180/529, loss: 0.011750412173569202 2023-01-21 11:07:18.823634: step: 184/529, loss: 0.680425226688385 2023-01-21 11:07:20.013910: step: 188/529, loss: 0.02919158898293972 2023-01-21 11:07:21.264704: step: 192/529, loss: 0.361136257648468 2023-01-21 11:07:22.478136: step: 196/529, loss: 0.06529875099658966 2023-01-21 11:07:23.673562: step: 200/529, loss: 0.16266211867332458 2023-01-21 11:07:24.872077: step: 204/529, loss: 0.0011018753284588456 2023-01-21 11:07:26.090866: step: 208/529, loss: 0.04203929752111435 2023-01-21 11:07:27.292257: step: 212/529, loss: 0.0029926300048828125 2023-01-21 11:07:28.457220: step: 216/529, loss: 0.05895552784204483 2023-01-21 11:07:29.662705: step: 220/529, loss: 0.0029110193718224764 2023-01-21 11:07:30.873265: step: 224/529, loss: 0.018513107672333717 2023-01-21 11:07:32.051995: step: 228/529, loss: 0.01807880401611328 2023-01-21 11:07:33.237290: step: 232/529, loss: 0.0023907662834972143 2023-01-21 11:07:34.426079: step: 236/529, loss: 0.31831127405166626 2023-01-21 11:07:35.637203: step: 240/529, loss: 0.011489678174257278 2023-01-21 11:07:36.830361: step: 244/529, loss: 0.028061963617801666 2023-01-21 11:07:38.009325: step: 248/529, loss: 0.037703897804021835 2023-01-21 11:07:39.194806: step: 252/529, loss: 0.008648396469652653 2023-01-21 11:07:40.424415: step: 256/529, loss: 0.007590389344841242 2023-01-21 11:07:41.606855: step: 260/529, loss: 0.06069403141736984 2023-01-21 11:07:42.799006: step: 264/529, loss: 0.05303707346320152 2023-01-21 11:07:44.022784: step: 268/529, loss: 0.010276603512465954 2023-01-21 11:07:45.227511: step: 272/529, loss: 0.011726569384336472 2023-01-21 11:07:46.395312: step: 276/529, loss: 0.04112198203802109 2023-01-21 11:07:47.627983: step: 280/529, loss: 0.0410066619515419 2023-01-21 11:07:48.829699: step: 284/529, loss: 0.010034275241196156 2023-01-21 11:07:50.030842: step: 288/529, loss: 0.11113253235816956 2023-01-21 11:07:51.230043: step: 292/529, loss: 0.05635128170251846 2023-01-21 11:07:52.419041: step: 296/529, loss: 0.013919925317168236 2023-01-21 11:07:53.624586: step: 300/529, loss: 0.023853302001953125 2023-01-21 11:07:54.826452: step: 304/529, loss: 0.05160675197839737 2023-01-21 11:07:55.999808: step: 308/529, loss: 0.0018610000843182206 2023-01-21 11:07:57.200262: step: 312/529, loss: 0.0082855224609375 2023-01-21 11:07:58.358856: step: 316/529, loss: 0.09461002796888351 2023-01-21 11:07:59.537824: step: 320/529, loss: 0.0010989189613610506 2023-01-21 11:08:00.762841: step: 324/529, loss: 0.012009715661406517 2023-01-21 11:08:01.969165: step: 328/529, loss: 0.034279823303222656 2023-01-21 11:08:03.148415: step: 332/529, loss: 0.0033151626121252775 2023-01-21 11:08:04.326328: step: 336/529, loss: 0.0009092331165447831 2023-01-21 11:08:05.473429: step: 340/529, loss: 0.08475551754236221 2023-01-21 11:08:06.652098: step: 344/529, loss: 0.021069718524813652 2023-01-21 11:08:07.838226: step: 348/529, loss: 0.05076861381530762 2023-01-21 11:08:09.047175: step: 352/529, loss: 0.02338237874209881 2023-01-21 11:08:10.218653: step: 356/529, loss: 0.029043961316347122 2023-01-21 11:08:11.400822: step: 360/529, loss: 0.0011564254527911544 2023-01-21 11:08:12.612711: step: 364/529, loss: 0.0520382896065712 2023-01-21 11:08:13.831581: step: 368/529, loss: 0.0035355566069483757 2023-01-21 11:08:15.023112: step: 372/529, loss: 0.02729806862771511 2023-01-21 11:08:16.247087: step: 376/529, loss: 0.016955947503447533 2023-01-21 11:08:17.441982: step: 380/529, loss: 0.045243073254823685 2023-01-21 11:08:18.611601: step: 384/529, loss: 0.0070375921204686165 2023-01-21 11:08:19.741360: step: 388/529, loss: 0.01606140285730362 2023-01-21 11:08:20.922620: step: 392/529, loss: 0.7768144011497498 2023-01-21 11:08:22.136436: step: 396/529, loss: 0.013397599570453167 2023-01-21 11:08:23.327845: step: 400/529, loss: 0.01559677068144083 2023-01-21 11:08:24.485370: step: 404/529, loss: 0.002626609755679965 2023-01-21 11:08:25.670508: step: 408/529, loss: 0.041284941136837006 2023-01-21 11:08:26.821508: step: 412/529, loss: 0.04969969019293785 2023-01-21 11:08:28.003786: step: 416/529, loss: 0.22172483801841736 2023-01-21 11:08:29.154705: step: 420/529, loss: 0.040380001068115234 2023-01-21 11:08:30.388131: step: 424/529, loss: 0.004055786412209272 2023-01-21 11:08:31.578969: step: 428/529, loss: 0.005878734402358532 2023-01-21 11:08:32.785188: step: 432/529, loss: 0.015458201989531517 2023-01-21 11:08:33.953246: step: 436/529, loss: 0.12731723487377167 2023-01-21 11:08:35.148384: step: 440/529, loss: 0.05409574508666992 2023-01-21 11:08:36.365403: step: 444/529, loss: 0.008396720513701439 2023-01-21 11:08:37.532274: step: 448/529, loss: 0.019727088510990143 2023-01-21 11:08:38.729311: step: 452/529, loss: 0.09304781258106232 2023-01-21 11:08:39.915022: step: 456/529, loss: 0.02880227565765381 2023-01-21 11:08:41.111324: step: 460/529, loss: 0.009797525592148304 2023-01-21 11:08:42.298945: step: 464/529, loss: 0.015048407949507236 2023-01-21 11:08:43.503664: step: 468/529, loss: 0.03133077546954155 2023-01-21 11:08:44.668859: step: 472/529, loss: 0.0329648032784462 2023-01-21 11:08:45.868148: step: 476/529, loss: 0.07482966780662537 2023-01-21 11:08:47.054790: step: 480/529, loss: 0.002013778779655695 2023-01-21 11:08:48.241332: step: 484/529, loss: 0.06957130879163742 2023-01-21 11:08:49.429418: step: 488/529, loss: 0.03861398622393608 2023-01-21 11:08:50.600137: step: 492/529, loss: 0.04085197299718857 2023-01-21 11:08:51.770142: step: 496/529, loss: 0.04988241195678711 2023-01-21 11:08:52.975726: step: 500/529, loss: 0.0883125364780426 2023-01-21 11:08:54.244124: step: 504/529, loss: 0.019052695482969284 2023-01-21 11:08:55.435406: step: 508/529, loss: 0.039664555341005325 2023-01-21 11:08:56.626046: step: 512/529, loss: 0.08069401234388351 2023-01-21 11:08:57.864961: step: 516/529, loss: 0.007090187165886164 2023-01-21 11:08:59.069849: step: 520/529, loss: 0.03291063383221626 2023-01-21 11:09:00.275966: step: 524/529, loss: 0.05287685617804527 2023-01-21 11:09:01.453977: step: 528/529, loss: 0.00021538735018111765 2023-01-21 11:09:02.661984: step: 532/529, loss: 0.07650041580200195 2023-01-21 11:09:03.865514: step: 536/529, loss: 0.024096012115478516 2023-01-21 11:09:05.115745: step: 540/529, loss: 0.03077392652630806 2023-01-21 11:09:06.328208: step: 544/529, loss: 0.04125552251935005 2023-01-21 11:09:07.535128: step: 548/529, loss: 0.047049619257450104 2023-01-21 11:09:08.728616: step: 552/529, loss: 0.03869781643152237 2023-01-21 11:09:09.986236: step: 556/529, loss: 0.00636062677949667 2023-01-21 11:09:11.192802: step: 560/529, loss: 0.0735289603471756 2023-01-21 11:09:12.361008: step: 564/529, loss: 0.027564240619540215 2023-01-21 11:09:13.555459: step: 568/529, loss: 0.016397427767515182 2023-01-21 11:09:14.793888: step: 572/529, loss: 0.061690427362918854 2023-01-21 11:09:16.000920: step: 576/529, loss: 0.13451018929481506 2023-01-21 11:09:17.185658: step: 580/529, loss: 0.09298533201217651 2023-01-21 11:09:18.393752: step: 584/529, loss: 0.023816490545868874 2023-01-21 11:09:19.572194: step: 588/529, loss: 0.10072250664234161 2023-01-21 11:09:20.803643: step: 592/529, loss: 0.006880283355712891 2023-01-21 11:09:22.031448: step: 596/529, loss: 0.018852615728974342 2023-01-21 11:09:23.214750: step: 600/529, loss: 0.30449360609054565 2023-01-21 11:09:24.358491: step: 604/529, loss: 0.0024212838616222143 2023-01-21 11:09:25.555388: step: 608/529, loss: 0.013472843915224075 2023-01-21 11:09:26.750876: step: 612/529, loss: 0.030254507437348366 2023-01-21 11:09:27.942436: step: 616/529, loss: 0.009787368588149548 2023-01-21 11:09:29.185887: step: 620/529, loss: 0.013148308731615543 2023-01-21 11:09:30.367781: step: 624/529, loss: 0.016123484820127487 2023-01-21 11:09:31.577447: step: 628/529, loss: 0.001380824949592352 2023-01-21 11:09:32.811691: step: 632/529, loss: 0.0006970405811443925 2023-01-21 11:09:34.038123: step: 636/529, loss: 0.09678631275892258 2023-01-21 11:09:35.192608: step: 640/529, loss: 0.08303698897361755 2023-01-21 11:09:36.362968: step: 644/529, loss: 0.0005417346837930381 2023-01-21 11:09:37.546018: step: 648/529, loss: 0.02665691450238228 2023-01-21 11:09:38.742884: step: 652/529, loss: 0.12355299293994904 2023-01-21 11:09:39.948073: step: 656/529, loss: 0.002891921903938055 2023-01-21 11:09:41.185038: step: 660/529, loss: 0.08119087666273117 2023-01-21 11:09:42.398109: step: 664/529, loss: 0.07087497413158417 2023-01-21 11:09:43.569756: step: 668/529, loss: 0.010590840131044388 2023-01-21 11:09:44.770914: step: 672/529, loss: 0.016165543347597122 2023-01-21 11:09:45.940912: step: 676/529, loss: 0.0018718719948083162 2023-01-21 11:09:47.119175: step: 680/529, loss: 0.009228992275893688 2023-01-21 11:09:48.289844: step: 684/529, loss: 0.04632072150707245 2023-01-21 11:09:49.524364: step: 688/529, loss: 0.03412752225995064 2023-01-21 11:09:50.722756: step: 692/529, loss: 0.030023861676454544 2023-01-21 11:09:51.915903: step: 696/529, loss: 0.01071634329855442 2023-01-21 11:09:53.092924: step: 700/529, loss: 0.154022216796875 2023-01-21 11:09:54.255026: step: 704/529, loss: 0.10907518118619919 2023-01-21 11:09:55.442895: step: 708/529, loss: 0.00128002162091434 2023-01-21 11:09:56.651691: step: 712/529, loss: 0.05320291593670845 2023-01-21 11:09:57.833458: step: 716/529, loss: 0.0026769163087010384 2023-01-21 11:09:59.007067: step: 720/529, loss: 0.004432487301528454 2023-01-21 11:10:00.232025: step: 724/529, loss: 0.011027717962861061 2023-01-21 11:10:01.450865: step: 728/529, loss: 0.06812162697315216 2023-01-21 11:10:02.653771: step: 732/529, loss: 0.03999815136194229 2023-01-21 11:10:03.881488: step: 736/529, loss: 0.0667715072631836 2023-01-21 11:10:05.128579: step: 740/529, loss: 0.013537026010453701 2023-01-21 11:10:06.371307: step: 744/529, loss: 0.0009384155855514109 2023-01-21 11:10:07.550360: step: 748/529, loss: 0.014869261533021927 2023-01-21 11:10:08.746074: step: 752/529, loss: 0.006194686982780695 2023-01-21 11:10:09.957239: step: 756/529, loss: 0.0077079772017896175 2023-01-21 11:10:11.160169: step: 760/529, loss: 0.03536491468548775 2023-01-21 11:10:12.400711: step: 764/529, loss: 0.006357002072036266 2023-01-21 11:10:13.615287: step: 768/529, loss: 0.03909120708703995 2023-01-21 11:10:14.811018: step: 772/529, loss: 0.0005373954772949219 2023-01-21 11:10:15.978093: step: 776/529, loss: 0.04561972990632057 2023-01-21 11:10:17.171396: step: 780/529, loss: 0.023701095953583717 2023-01-21 11:10:18.362273: step: 784/529, loss: 0.030454445630311966 2023-01-21 11:10:19.575764: step: 788/529, loss: 0.00836334191262722 2023-01-21 11:10:20.738200: step: 792/529, loss: 0.018549060449004173 2023-01-21 11:10:21.963870: step: 796/529, loss: 0.03867530822753906 2023-01-21 11:10:23.146959: step: 800/529, loss: 0.04585866630077362 2023-01-21 11:10:24.325067: step: 804/529, loss: 0.01351385097950697 2023-01-21 11:10:25.513949: step: 808/529, loss: 0.016197824850678444 2023-01-21 11:10:26.693030: step: 812/529, loss: 0.5049812197685242 2023-01-21 11:10:27.857436: step: 816/529, loss: 0.004399907775223255 2023-01-21 11:10:29.035698: step: 820/529, loss: 0.03361711651086807 2023-01-21 11:10:30.185257: step: 824/529, loss: 0.028964996337890625 2023-01-21 11:10:31.376172: step: 828/529, loss: 0.05644054338335991 2023-01-21 11:10:32.563988: step: 832/529, loss: 0.06460876762866974 2023-01-21 11:10:33.763284: step: 836/529, loss: 0.007114887237548828 2023-01-21 11:10:35.020762: step: 840/529, loss: 0.07699642330408096 2023-01-21 11:10:36.172829: step: 844/529, loss: 0.004122877027839422 2023-01-21 11:10:37.382093: step: 848/529, loss: 0.04374675825238228 2023-01-21 11:10:38.573743: step: 852/529, loss: 0.038248348981142044 2023-01-21 11:10:39.726480: step: 856/529, loss: 0.034829046577215195 2023-01-21 11:10:40.926567: step: 860/529, loss: 0.046213533729314804 2023-01-21 11:10:42.142317: step: 864/529, loss: 0.026385022327303886 2023-01-21 11:10:43.334528: step: 868/529, loss: 0.005799293518066406 2023-01-21 11:10:44.553907: step: 872/529, loss: 0.2844853401184082 2023-01-21 11:10:45.752070: step: 876/529, loss: 0.020926428958773613 2023-01-21 11:10:46.945800: step: 880/529, loss: 0.02928609773516655 2023-01-21 11:10:48.133720: step: 884/529, loss: 0.03726644441485405 2023-01-21 11:10:49.306606: step: 888/529, loss: 0.4652038514614105 2023-01-21 11:10:50.505050: step: 892/529, loss: 0.02466583251953125 2023-01-21 11:10:51.726685: step: 896/529, loss: 0.011637592688202858 2023-01-21 11:10:52.883604: step: 900/529, loss: 0.0012027741177007556 2023-01-21 11:10:54.065924: step: 904/529, loss: 0.005543517880141735 2023-01-21 11:10:55.307142: step: 908/529, loss: 0.001992166042327881 2023-01-21 11:10:56.505550: step: 912/529, loss: 0.01972498930990696 2023-01-21 11:10:57.691569: step: 916/529, loss: 0.016737990081310272 2023-01-21 11:10:58.920096: step: 920/529, loss: 0.09065389633178711 2023-01-21 11:11:00.089698: step: 924/529, loss: 0.016427231952548027 2023-01-21 11:11:01.281073: step: 928/529, loss: 0.03529768064618111 2023-01-21 11:11:02.502478: step: 932/529, loss: 0.021167373284697533 2023-01-21 11:11:03.704792: step: 936/529, loss: 0.044316768646240234 2023-01-21 11:11:04.955867: step: 940/529, loss: 0.12078666687011719 2023-01-21 11:11:06.139715: step: 944/529, loss: 0.0019444465870037675 2023-01-21 11:11:07.373046: step: 948/529, loss: 0.0016371726524084806 2023-01-21 11:11:08.564403: step: 952/529, loss: 0.006275392137467861 2023-01-21 11:11:09.826414: step: 956/529, loss: 0.029324818402528763 2023-01-21 11:11:11.006605: step: 960/529, loss: 0.002056622412055731 2023-01-21 11:11:12.267611: step: 964/529, loss: 0.19268283247947693 2023-01-21 11:11:13.492335: step: 968/529, loss: 0.11185389012098312 2023-01-21 11:11:14.662761: step: 972/529, loss: 0.010258197784423828 2023-01-21 11:11:15.868789: step: 976/529, loss: 0.03547201305627823 2023-01-21 11:11:17.048768: step: 980/529, loss: 0.04166984558105469 2023-01-21 11:11:18.215958: step: 984/529, loss: 0.019840337336063385 2023-01-21 11:11:19.437512: step: 988/529, loss: 0.019434357061982155 2023-01-21 11:11:20.637880: step: 992/529, loss: 0.05450878292322159 2023-01-21 11:11:21.799781: step: 996/529, loss: 0.036869145929813385 2023-01-21 11:11:22.985883: step: 1000/529, loss: 0.011790085583925247 2023-01-21 11:11:24.140064: step: 1004/529, loss: 0.029982566833496094 2023-01-21 11:11:25.357064: step: 1008/529, loss: 0.019115637987852097 2023-01-21 11:11:26.550638: step: 1012/529, loss: 0.00112323765642941 2023-01-21 11:11:27.712390: step: 1016/529, loss: 0.0033578872680664062 2023-01-21 11:11:28.887823: step: 1020/529, loss: 0.03182220458984375 2023-01-21 11:11:30.105178: step: 1024/529, loss: 0.04552431404590607 2023-01-21 11:11:31.301992: step: 1028/529, loss: 0.04718761518597603 2023-01-21 11:11:32.487289: step: 1032/529, loss: 0.002457237336784601 2023-01-21 11:11:33.715168: step: 1036/529, loss: 7.613883018493652 2023-01-21 11:11:34.909998: step: 1040/529, loss: 0.019194984808564186 2023-01-21 11:11:36.060381: step: 1044/529, loss: 0.020956039428710938 2023-01-21 11:11:37.281771: step: 1048/529, loss: 0.0018651008140295744 2023-01-21 11:11:38.459232: step: 1052/529, loss: 0.0804869681596756 2023-01-21 11:11:39.674095: step: 1056/529, loss: 0.007631206884980202 2023-01-21 11:11:40.877966: step: 1060/529, loss: 0.003436565399169922 2023-01-21 11:11:42.082877: step: 1064/529, loss: 0.01638183556497097 2023-01-21 11:11:43.338402: step: 1068/529, loss: 0.0275771152228117 2023-01-21 11:11:44.558754: step: 1072/529, loss: 0.03643946722149849 2023-01-21 11:11:45.747948: step: 1076/529, loss: 0.0037660598754882812 2023-01-21 11:11:46.997190: step: 1080/529, loss: 0.04119758680462837 2023-01-21 11:11:48.217167: step: 1084/529, loss: 0.06853781640529633 2023-01-21 11:11:49.411206: step: 1088/529, loss: 0.00012159347534179688 2023-01-21 11:11:50.642393: step: 1092/529, loss: 0.013943481259047985 2023-01-21 11:11:51.849165: step: 1096/529, loss: 0.03274545818567276 2023-01-21 11:11:53.035558: step: 1100/529, loss: 0.08214292675256729 2023-01-21 11:11:54.208854: step: 1104/529, loss: 0.0075247762724757195 2023-01-21 11:11:55.406671: step: 1108/529, loss: 0.03893442451953888 2023-01-21 11:11:56.566426: step: 1112/529, loss: 0.05002059042453766 2023-01-21 11:11:57.758721: step: 1116/529, loss: 0.06410379707813263 2023-01-21 11:11:58.943187: step: 1120/529, loss: 0.008359337225556374 2023-01-21 11:12:00.174908: step: 1124/529, loss: 0.0009662627708166838 2023-01-21 11:12:01.360744: step: 1128/529, loss: 0.022592831403017044 2023-01-21 11:12:02.591259: step: 1132/529, loss: 0.6187344789505005 2023-01-21 11:12:03.782899: step: 1136/529, loss: 0.029236221686005592 2023-01-21 11:12:04.976839: step: 1140/529, loss: 0.01214456558227539 2023-01-21 11:12:06.148706: step: 1144/529, loss: 0.0078095439821481705 2023-01-21 11:12:07.403309: step: 1148/529, loss: 0.052944280207157135 2023-01-21 11:12:08.575087: step: 1152/529, loss: 0.06216888502240181 2023-01-21 11:12:09.748667: step: 1156/529, loss: 8.769035775912926e-05 2023-01-21 11:12:10.945815: step: 1160/529, loss: 0.03388276323676109 2023-01-21 11:12:12.153366: step: 1164/529, loss: 0.058298349380493164 2023-01-21 11:12:13.316258: step: 1168/529, loss: 0.2575405240058899 2023-01-21 11:12:14.498960: step: 1172/529, loss: 0.021151067689061165 2023-01-21 11:12:15.694901: step: 1176/529, loss: 0.01527795847505331 2023-01-21 11:12:16.880104: step: 1180/529, loss: 0.01699075847864151 2023-01-21 11:12:18.088515: step: 1184/529, loss: 0.09953327476978302 2023-01-21 11:12:19.329818: step: 1188/529, loss: 0.33709201216697693 2023-01-21 11:12:20.523567: step: 1192/529, loss: 0.013056564144790173 2023-01-21 11:12:21.711769: step: 1196/529, loss: 0.020695878192782402 2023-01-21 11:12:22.901182: step: 1200/529, loss: 0.08869123458862305 2023-01-21 11:12:24.161929: step: 1204/529, loss: 0.012712287716567516 2023-01-21 11:12:25.381430: step: 1208/529, loss: 0.06995334476232529 2023-01-21 11:12:26.557446: step: 1212/529, loss: 0.005582476034760475 2023-01-21 11:12:27.762693: step: 1216/529, loss: 0.016435623168945312 2023-01-21 11:12:28.966617: step: 1220/529, loss: 0.18591423332691193 2023-01-21 11:12:30.182969: step: 1224/529, loss: 0.0067001343704760075 2023-01-21 11:12:31.407636: step: 1228/529, loss: 0.020901108160614967 2023-01-21 11:12:32.610527: step: 1232/529, loss: 0.029928207397460938 2023-01-21 11:12:33.794320: step: 1236/529, loss: 0.01280994527041912 2023-01-21 11:12:35.036672: step: 1240/529, loss: 0.05920276790857315 2023-01-21 11:12:36.252161: step: 1244/529, loss: 0.012746095657348633 2023-01-21 11:12:37.428979: step: 1248/529, loss: 0.002952480223029852 2023-01-21 11:12:38.613402: step: 1252/529, loss: 0.09848232567310333 2023-01-21 11:12:39.820119: step: 1256/529, loss: 0.09303627908229828 2023-01-21 11:12:40.996856: step: 1260/529, loss: 0.028832150623202324 2023-01-21 11:12:42.217860: step: 1264/529, loss: 0.005545425694435835 2023-01-21 11:12:43.376465: step: 1268/529, loss: 0.16472569108009338 2023-01-21 11:12:44.584946: step: 1272/529, loss: 0.027658844366669655 2023-01-21 11:12:45.797736: step: 1276/529, loss: 0.12092819809913635 2023-01-21 11:12:46.990107: step: 1280/529, loss: 0.00525136012583971 2023-01-21 11:12:48.228350: step: 1284/529, loss: 0.006985223386436701 2023-01-21 11:12:49.408388: step: 1288/529, loss: 0.0033423902932554483 2023-01-21 11:12:50.570534: step: 1292/529, loss: 0.038785506039857864 2023-01-21 11:12:51.738259: step: 1296/529, loss: 0.009946918115019798 2023-01-21 11:12:52.904259: step: 1300/529, loss: 0.0059906006790697575 2023-01-21 11:12:54.154744: step: 1304/529, loss: 0.03427105024456978 2023-01-21 11:12:55.348160: step: 1308/529, loss: 0.05811386555433273 2023-01-21 11:12:56.546370: step: 1312/529, loss: 0.01921234093606472 2023-01-21 11:12:57.756608: step: 1316/529, loss: 0.0275576114654541 2023-01-21 11:12:58.925921: step: 1320/529, loss: 0.006468010134994984 2023-01-21 11:13:00.133112: step: 1324/529, loss: 0.03248901665210724 2023-01-21 11:13:01.307486: step: 1328/529, loss: 0.020908953621983528 2023-01-21 11:13:02.532503: step: 1332/529, loss: 0.019912149757146835 2023-01-21 11:13:03.741212: step: 1336/529, loss: 0.008447027765214443 2023-01-21 11:13:04.981715: step: 1340/529, loss: 0.3015109896659851 2023-01-21 11:13:06.144710: step: 1344/529, loss: 0.03302764892578125 2023-01-21 11:13:07.395644: step: 1348/529, loss: 0.033263400197029114 2023-01-21 11:13:08.601489: step: 1352/529, loss: 0.007282543461769819 2023-01-21 11:13:09.855203: step: 1356/529, loss: 0.06480860710144043 2023-01-21 11:13:11.071414: step: 1360/529, loss: 0.07706928998231888 2023-01-21 11:13:12.285941: step: 1364/529, loss: 0.00021967888460494578 2023-01-21 11:13:13.496714: step: 1368/529, loss: 0.007479286286979914 2023-01-21 11:13:14.723552: step: 1372/529, loss: 0.0020242691971361637 2023-01-21 11:13:15.912010: step: 1376/529, loss: 0.0025319100823253393 2023-01-21 11:13:17.150231: step: 1380/529, loss: 0.028531935065984726 2023-01-21 11:13:18.322946: step: 1384/529, loss: 0.05704698711633682 2023-01-21 11:13:19.522635: step: 1388/529, loss: 0.0183913242071867 2023-01-21 11:13:20.714892: step: 1392/529, loss: 0.007855034433305264 2023-01-21 11:13:21.898054: step: 1396/529, loss: 0.0016820431919768453 2023-01-21 11:13:23.124143: step: 1400/529, loss: 0.06636278331279755 2023-01-21 11:13:24.343783: step: 1404/529, loss: 0.01532160583883524 2023-01-21 11:13:25.585590: step: 1408/529, loss: 0.024730205535888672 2023-01-21 11:13:26.810800: step: 1412/529, loss: 0.0005437612417154014 2023-01-21 11:13:28.006415: step: 1416/529, loss: 0.04557623714208603 2023-01-21 11:13:29.216078: step: 1420/529, loss: 0.026226473972201347 2023-01-21 11:13:30.428562: step: 1424/529, loss: 0.0005801916122436523 2023-01-21 11:13:31.635016: step: 1428/529, loss: 0.03095417097210884 2023-01-21 11:13:32.822629: step: 1432/529, loss: 0.004605341237038374 2023-01-21 11:13:34.005870: step: 1436/529, loss: 0.05617962405085564 2023-01-21 11:13:35.209115: step: 1440/529, loss: 0.010184098035097122 2023-01-21 11:13:36.469132: step: 1444/529, loss: 0.004900550935417414 2023-01-21 11:13:37.669258: step: 1448/529, loss: 0.20014648139476776 2023-01-21 11:13:38.844167: step: 1452/529, loss: 0.011930370703339577 2023-01-21 11:13:40.018392: step: 1456/529, loss: 0.0005268097156658769 2023-01-21 11:13:41.240338: step: 1460/529, loss: 0.02560530975461006 2023-01-21 11:13:42.467652: step: 1464/529, loss: 0.06348982453346252 2023-01-21 11:13:43.653575: step: 1468/529, loss: 0.08747158199548721 2023-01-21 11:13:44.848746: step: 1472/529, loss: 0.022947883233428 2023-01-21 11:13:46.026013: step: 1476/529, loss: 0.04778118431568146 2023-01-21 11:13:47.259725: step: 1480/529, loss: 0.0037987353280186653 2023-01-21 11:13:48.470080: step: 1484/529, loss: 0.025248384103178978 2023-01-21 11:13:49.689342: step: 1488/529, loss: 0.011888314038515091 2023-01-21 11:13:50.885425: step: 1492/529, loss: 0.06573381274938583 2023-01-21 11:13:52.100742: step: 1496/529, loss: 0.012727736495435238 2023-01-21 11:13:53.298852: step: 1500/529, loss: 0.066846564412117 2023-01-21 11:13:54.491410: step: 1504/529, loss: 0.016164684668183327 2023-01-21 11:13:55.659666: step: 1508/529, loss: 0.01792125776410103 2023-01-21 11:13:56.857946: step: 1512/529, loss: 0.01788330264389515 2023-01-21 11:13:58.070965: step: 1516/529, loss: 0.07105641067028046 2023-01-21 11:13:59.295285: step: 1520/529, loss: 0.01463176030665636 2023-01-21 11:14:00.498691: step: 1524/529, loss: 0.07115145027637482 2023-01-21 11:14:01.699134: step: 1528/529, loss: 0.1553487777709961 2023-01-21 11:14:02.918771: step: 1532/529, loss: 0.033963393419981 2023-01-21 11:14:04.091558: step: 1536/529, loss: 0.013622378930449486 2023-01-21 11:14:05.277649: step: 1540/529, loss: 0.03125820308923721 2023-01-21 11:14:06.443951: step: 1544/529, loss: 0.004850661847740412 2023-01-21 11:14:07.618454: step: 1548/529, loss: 0.0060867308638989925 2023-01-21 11:14:08.781307: step: 1552/529, loss: 0.007416534703224897 2023-01-21 11:14:09.951390: step: 1556/529, loss: 0.0038619041442871094 2023-01-21 11:14:11.152039: step: 1560/529, loss: 0.0263079646974802 2023-01-21 11:14:12.381103: step: 1564/529, loss: 0.036171529442071915 2023-01-21 11:14:13.559412: step: 1568/529, loss: 0.046029187738895416 2023-01-21 11:14:14.816687: step: 1572/529, loss: 0.015340805053710938 2023-01-21 11:14:16.053841: step: 1576/529, loss: 0.06995902210474014 2023-01-21 11:14:17.278924: step: 1580/529, loss: 0.005814552307128906 2023-01-21 11:14:18.493032: step: 1584/529, loss: 0.055701542645692825 2023-01-21 11:14:19.640468: step: 1588/529, loss: 0.016504669561982155 2023-01-21 11:14:20.856804: step: 1592/529, loss: 0.06160459667444229 2023-01-21 11:14:22.056805: step: 1596/529, loss: 0.027900028973817825 2023-01-21 11:14:23.304510: step: 1600/529, loss: 0.006214332766830921 2023-01-21 11:14:24.478785: step: 1604/529, loss: 0.003419304033741355 2023-01-21 11:14:25.672785: step: 1608/529, loss: 0.0074745179153978825 2023-01-21 11:14:26.840933: step: 1612/529, loss: 0.017643345519900322 2023-01-21 11:14:28.037989: step: 1616/529, loss: 0.0021614551078528166 2023-01-21 11:14:29.202124: step: 1620/529, loss: 0.07348699867725372 2023-01-21 11:14:30.388959: step: 1624/529, loss: 0.130061075091362 2023-01-21 11:14:31.693129: step: 1628/529, loss: 0.05086164548993111 2023-01-21 11:14:32.934370: step: 1632/529, loss: 0.010193824768066406 2023-01-21 11:14:34.157067: step: 1636/529, loss: 0.005540275946259499 2023-01-21 11:14:35.358294: step: 1640/529, loss: 0.022570276632905006 2023-01-21 11:14:36.559142: step: 1644/529, loss: 0.045505695044994354 2023-01-21 11:14:37.726679: step: 1648/529, loss: 0.05347108840942383 2023-01-21 11:14:38.947615: step: 1652/529, loss: 0.06545610725879669 2023-01-21 11:14:40.140059: step: 1656/529, loss: 0.031281281262636185 2023-01-21 11:14:41.325594: step: 1660/529, loss: 0.04181130230426788 2023-01-21 11:14:42.492975: step: 1664/529, loss: 0.027374744415283203 2023-01-21 11:14:43.684781: step: 1668/529, loss: 0.008363437838852406 2023-01-21 11:14:44.861268: step: 1672/529, loss: 0.03581733629107475 2023-01-21 11:14:46.090587: step: 1676/529, loss: 0.05150566250085831 2023-01-21 11:14:47.304906: step: 1680/529, loss: 0.0016584396362304688 2023-01-21 11:14:48.500602: step: 1684/529, loss: 0.00958938617259264 2023-01-21 11:14:49.719908: step: 1688/529, loss: 0.03881445154547691 2023-01-21 11:14:50.929855: step: 1692/529, loss: 0.05616569519042969 2023-01-21 11:14:52.140521: step: 1696/529, loss: 0.003646993776783347 2023-01-21 11:14:53.321764: step: 1700/529, loss: 0.07463932037353516 2023-01-21 11:14:54.565503: step: 1704/529, loss: 0.007560920435935259 2023-01-21 11:14:55.766089: step: 1708/529, loss: 0.09490852802991867 2023-01-21 11:14:56.990374: step: 1712/529, loss: 0.02349872514605522 2023-01-21 11:14:58.203723: step: 1716/529, loss: 0.0493219755589962 2023-01-21 11:14:59.369868: step: 1720/529, loss: 0.048899270594120026 2023-01-21 11:15:00.577693: step: 1724/529, loss: 0.013747883029282093 2023-01-21 11:15:01.789448: step: 1728/529, loss: 0.027751732617616653 2023-01-21 11:15:02.973017: step: 1732/529, loss: 0.05959749221801758 2023-01-21 11:15:04.183201: step: 1736/529, loss: 0.021447371691465378 2023-01-21 11:15:05.391990: step: 1740/529, loss: 0.0032948970329016447 2023-01-21 11:15:06.608013: step: 1744/529, loss: 0.0014051944017410278 2023-01-21 11:15:07.773388: step: 1748/529, loss: 0.03415260463953018 2023-01-21 11:15:08.980658: step: 1752/529, loss: 0.07967357337474823 2023-01-21 11:15:10.156715: step: 1756/529, loss: 0.042795706540346146 2023-01-21 11:15:11.323194: step: 1760/529, loss: 0.07079362869262695 2023-01-21 11:15:12.512531: step: 1764/529, loss: 0.011539077386260033 2023-01-21 11:15:13.696643: step: 1768/529, loss: 0.014110947027802467 2023-01-21 11:15:14.871730: step: 1772/529, loss: 0.04072609171271324 2023-01-21 11:15:16.103023: step: 1776/529, loss: 0.03273210674524307 2023-01-21 11:15:17.345894: step: 1780/529, loss: 0.08004264533519745 2023-01-21 11:15:18.546700: step: 1784/529, loss: 0.04920263588428497 2023-01-21 11:15:19.774495: step: 1788/529, loss: 0.10527591407299042 2023-01-21 11:15:20.947412: step: 1792/529, loss: 0.0065056802704930305 2023-01-21 11:15:22.156695: step: 1796/529, loss: 0.056125592440366745 2023-01-21 11:15:23.362652: step: 1800/529, loss: 0.005870628636330366 2023-01-21 11:15:24.527619: step: 1804/529, loss: 0.01254568062722683 2023-01-21 11:15:25.744297: step: 1808/529, loss: 0.0728035494685173 2023-01-21 11:15:26.938865: step: 1812/529, loss: 0.08295102417469025 2023-01-21 11:15:28.116220: step: 1816/529, loss: 0.07871723175048828 2023-01-21 11:15:29.315521: step: 1820/529, loss: 0.010755443945527077 2023-01-21 11:15:30.518784: step: 1824/529, loss: 0.015051651746034622 2023-01-21 11:15:31.691461: step: 1828/529, loss: 0.008225775323808193 2023-01-21 11:15:32.880635: step: 1832/529, loss: 0.01578162983059883 2023-01-21 11:15:34.036455: step: 1836/529, loss: 0.0046412465162575245 2023-01-21 11:15:35.217693: step: 1840/529, loss: 0.1570996195077896 2023-01-21 11:15:36.427882: step: 1844/529, loss: 0.06905098259449005 2023-01-21 11:15:37.600943: step: 1848/529, loss: 0.02801513671875 2023-01-21 11:15:38.837004: step: 1852/529, loss: 0.06198396533727646 2023-01-21 11:15:40.003410: step: 1856/529, loss: 0.036179397255182266 2023-01-21 11:15:41.208669: step: 1860/529, loss: 0.012075996957719326 2023-01-21 11:15:42.413806: step: 1864/529, loss: 0.18381473422050476 2023-01-21 11:15:43.587774: step: 1868/529, loss: 0.04402294382452965 2023-01-21 11:15:44.819593: step: 1872/529, loss: 0.036356642842292786 2023-01-21 11:15:45.989292: step: 1876/529, loss: 0.018994569778442383 2023-01-21 11:15:47.263439: step: 1880/529, loss: 0.04035930708050728 2023-01-21 11:15:48.482375: step: 1884/529, loss: 0.024074364453554153 2023-01-21 11:15:49.705579: step: 1888/529, loss: 0.018526840955018997 2023-01-21 11:15:50.929827: step: 1892/529, loss: 0.010426235385239124 2023-01-21 11:15:52.140764: step: 1896/529, loss: 0.051320839673280716 2023-01-21 11:15:53.346401: step: 1900/529, loss: 0.12202034145593643 2023-01-21 11:15:54.580605: step: 1904/529, loss: 0.05969180911779404 2023-01-21 11:15:55.813538: step: 1908/529, loss: 0.07862873375415802 2023-01-21 11:15:56.990374: step: 1912/529, loss: 0.02726149559020996 2023-01-21 11:15:58.190588: step: 1916/529, loss: 0.03670492395758629 2023-01-21 11:15:59.373168: step: 1920/529, loss: 0.013945579528808594 2023-01-21 11:16:00.564348: step: 1924/529, loss: 0.02425394207239151 2023-01-21 11:16:01.814066: step: 1928/529, loss: 0.06108541786670685 2023-01-21 11:16:03.021633: step: 1932/529, loss: 0.1501600295305252 2023-01-21 11:16:04.188637: step: 1936/529, loss: 0.03265504911541939 2023-01-21 11:16:05.355721: step: 1940/529, loss: 0.16952934861183167 2023-01-21 11:16:06.582404: step: 1944/529, loss: 0.01171646174043417 2023-01-21 11:16:07.757322: step: 1948/529, loss: 0.12484703212976456 2023-01-21 11:16:08.945156: step: 1952/529, loss: 0.34257006645202637 2023-01-21 11:16:10.204157: step: 1956/529, loss: 0.056500911712646484 2023-01-21 11:16:11.430391: step: 1960/529, loss: 0.025409795343875885 2023-01-21 11:16:12.656392: step: 1964/529, loss: 0.036218930035829544 2023-01-21 11:16:13.865120: step: 1968/529, loss: 0.05441112443804741 2023-01-21 11:16:15.038174: step: 1972/529, loss: 0.025200797244906425 2023-01-21 11:16:16.207580: step: 1976/529, loss: 0.015030860900878906 2023-01-21 11:16:17.434706: step: 1980/529, loss: 0.020447731018066406 2023-01-21 11:16:18.651820: step: 1984/529, loss: 0.04184160754084587 2023-01-21 11:16:19.818872: step: 1988/529, loss: 0.028174782171845436 2023-01-21 11:16:21.063177: step: 1992/529, loss: 0.2870357632637024 2023-01-21 11:16:22.280612: step: 1996/529, loss: 0.1252630203962326 2023-01-21 11:16:23.466746: step: 2000/529, loss: 0.028281403705477715 2023-01-21 11:16:24.659630: step: 2004/529, loss: 0.02722320705652237 2023-01-21 11:16:25.864704: step: 2008/529, loss: 0.03438367694616318 2023-01-21 11:16:27.065555: step: 2012/529, loss: 0.014744091778993607 2023-01-21 11:16:28.272360: step: 2016/529, loss: 0.07052536308765411 2023-01-21 11:16:29.476080: step: 2020/529, loss: 0.11352807283401489 2023-01-21 11:16:30.676616: step: 2024/529, loss: 0.059533145278692245 2023-01-21 11:16:31.868276: step: 2028/529, loss: 0.007819080725312233 2023-01-21 11:16:33.070641: step: 2032/529, loss: 0.005525302607566118 2023-01-21 11:16:34.215932: step: 2036/529, loss: 0.023994911462068558 2023-01-21 11:16:35.401188: step: 2040/529, loss: 0.011910355649888515 2023-01-21 11:16:36.636655: step: 2044/529, loss: 0.00218791957013309 2023-01-21 11:16:37.810648: step: 2048/529, loss: 0.01730489730834961 2023-01-21 11:16:38.982205: step: 2052/529, loss: 0.03709264099597931 2023-01-21 11:16:40.182152: step: 2056/529, loss: 0.02214631997048855 2023-01-21 11:16:41.355329: step: 2060/529, loss: 0.027809666469693184 2023-01-21 11:16:42.500441: step: 2064/529, loss: 0.01901989057660103 2023-01-21 11:16:43.714061: step: 2068/529, loss: 0.01893148384988308 2023-01-21 11:16:44.966432: step: 2072/529, loss: 0.008374262601137161 2023-01-21 11:16:46.225963: step: 2076/529, loss: 0.02663717232644558 2023-01-21 11:16:47.443478: step: 2080/529, loss: 0.044006966054439545 2023-01-21 11:16:48.597843: step: 2084/529, loss: 0.09519825130701065 2023-01-21 11:16:49.800009: step: 2088/529, loss: 0.037520598620176315 2023-01-21 11:16:51.020591: step: 2092/529, loss: 0.0752742812037468 2023-01-21 11:16:52.209792: step: 2096/529, loss: 0.04630632326006889 2023-01-21 11:16:53.431574: step: 2100/529, loss: 0.026484372094273567 2023-01-21 11:16:54.631150: step: 2104/529, loss: 0.002570724580436945 2023-01-21 11:16:55.832900: step: 2108/529, loss: 0.01820087432861328 2023-01-21 11:16:56.997205: step: 2112/529, loss: 0.05377311632037163 2023-01-21 11:16:58.209330: step: 2116/529, loss: 0.06346340477466583 ================================================== Loss: 0.063 -------------------- Dev: {'event': {'p': 0.5783730158730159, 'r': 0.7762982689747004, 'f1': 0.6628766344513929}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Test: {'event': {'p': 0.6390361445783133, 'r': 0.7944877171959257, 'f1': 0.7083333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} 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': 13} Korean: {'event': {'p': 0.4782608695652174, 'r': 0.5238095238095238, 'f1': 0.5}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Russian: {'event': {'p': 0.45652173913043476, 'r': 0.5833333333333334, 'f1': 0.5121951219512195}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 14 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:17:43.182385: step: 4/529, loss: 0.006434726994484663 2023-01-21 11:17:44.404295: step: 8/529, loss: 0.0008400916703976691 2023-01-21 11:17:45.571779: step: 12/529, loss: 0.11036419868469238 2023-01-21 11:17:46.798347: step: 16/529, loss: 0.11412505805492401 2023-01-21 11:17:48.010045: step: 20/529, loss: 0.002006101654842496 2023-01-21 11:17:49.225480: step: 24/529, loss: 0.01931934431195259 2023-01-21 11:17:50.420539: step: 28/529, loss: 0.0006698131328448653 2023-01-21 11:17:51.624764: step: 32/529, loss: 0.00022406579228118062 2023-01-21 11:17:52.808362: step: 36/529, loss: 0.05135917663574219 2023-01-21 11:17:53.982042: step: 40/529, loss: 0.013453959487378597 2023-01-21 11:17:55.189401: step: 44/529, loss: 0.1395878791809082 2023-01-21 11:17:56.439550: step: 48/529, loss: 0.006522846408188343 2023-01-21 11:17:57.623212: step: 52/529, loss: 0.039248086512088776 2023-01-21 11:17:58.815155: step: 56/529, loss: 0.001166057656519115 2023-01-21 11:18:00.002876: step: 60/529, loss: 0.015667341649532318 2023-01-21 11:18:01.162777: step: 64/529, loss: 0.002609044313430786 2023-01-21 11:18:02.370901: step: 68/529, loss: 0.017549704760313034 2023-01-21 11:18:03.699602: step: 72/529, loss: 0.01764049381017685 2023-01-21 11:18:04.892685: step: 76/529, loss: 0.015106964856386185 2023-01-21 11:18:06.102946: step: 80/529, loss: 0.05608396977186203 2023-01-21 11:18:07.309266: step: 84/529, loss: 0.004139518830925226 2023-01-21 11:18:08.460907: step: 88/529, loss: 0.004285716917365789 2023-01-21 11:18:09.649262: step: 92/529, loss: 0.03477032855153084 2023-01-21 11:18:10.864454: step: 96/529, loss: 0.07379131019115448 2023-01-21 11:18:12.065517: step: 100/529, loss: 0.03617975860834122 2023-01-21 11:18:13.313875: step: 104/529, loss: 0.0012842179276049137 2023-01-21 11:18:14.520352: step: 108/529, loss: 0.10076847672462463 2023-01-21 11:18:15.750806: step: 112/529, loss: 0.027474403381347656 2023-01-21 11:18:16.983573: step: 116/529, loss: 0.018334198743104935 2023-01-21 11:18:18.160642: step: 120/529, loss: 0.04866437986493111 2023-01-21 11:18:19.395224: step: 124/529, loss: 0.04026756435632706 2023-01-21 11:18:20.579556: step: 128/529, loss: 0.08572883903980255 2023-01-21 11:18:21.755452: step: 132/529, loss: 0.006194686982780695 2023-01-21 11:18:22.960431: step: 136/529, loss: 0.053522590547800064 2023-01-21 11:18:24.126375: step: 140/529, loss: 0.03320331871509552 2023-01-21 11:18:25.328208: step: 144/529, loss: 0.009608983993530273 2023-01-21 11:18:26.512578: step: 148/529, loss: 0.0009262085077352822 2023-01-21 11:18:27.680332: step: 152/529, loss: 0.0314909927546978 2023-01-21 11:18:28.882943: step: 156/529, loss: 0.0453365333378315 2023-01-21 11:18:30.063646: step: 160/529, loss: 0.06576023250818253 2023-01-21 11:18:31.250504: step: 164/529, loss: 0.004456997383385897 2023-01-21 11:18:32.486995: step: 168/529, loss: 0.0060249329544603825 2023-01-21 11:18:33.718590: step: 172/529, loss: 0.09245280921459198 2023-01-21 11:18:34.959257: step: 176/529, loss: 0.1105566993355751 2023-01-21 11:18:36.165506: step: 180/529, loss: 0.019758939743041992 2023-01-21 11:18:37.358543: step: 184/529, loss: 0.015011405572295189 2023-01-21 11:18:38.540374: step: 188/529, loss: 0.024150991812348366 2023-01-21 11:18:39.775222: step: 192/529, loss: 0.06746292114257812 2023-01-21 11:18:40.941635: step: 196/529, loss: 0.016742898151278496 2023-01-21 11:18:42.158056: step: 200/529, loss: 0.04550713673233986 2023-01-21 11:18:43.365394: step: 204/529, loss: 0.03450794145464897 2023-01-21 11:18:44.556471: step: 208/529, loss: 0.09060563892126083 2023-01-21 11:18:45.760238: step: 212/529, loss: 0.5808812379837036 2023-01-21 11:18:46.938557: step: 216/529, loss: 0.020499326288700104 2023-01-21 11:18:48.157407: step: 220/529, loss: 0.0982338935136795 2023-01-21 11:18:49.349014: step: 224/529, loss: 0.0031575202010571957 2023-01-21 11:18:50.532218: step: 228/529, loss: 0.0011027336586266756 2023-01-21 11:18:51.694286: step: 232/529, loss: 0.0063127996399998665 2023-01-21 11:18:52.881465: step: 236/529, loss: 0.02760915644466877 2023-01-21 11:18:54.054222: step: 240/529, loss: 2.726727247238159 2023-01-21 11:18:55.251559: step: 244/529, loss: 0.021411515772342682 2023-01-21 11:18:56.462410: step: 248/529, loss: 0.0002449035819154233 2023-01-21 11:18:57.710969: step: 252/529, loss: 0.055960461497306824 2023-01-21 11:18:58.917966: step: 256/529, loss: 0.01124725304543972 2023-01-21 11:19:00.089932: step: 260/529, loss: 0.0079659940674901 2023-01-21 11:19:01.313743: step: 264/529, loss: 0.03912172466516495 2023-01-21 11:19:02.503381: step: 268/529, loss: 0.0008550644270144403 2023-01-21 11:19:03.684114: step: 272/529, loss: 0.025252103805541992 2023-01-21 11:19:04.905448: step: 276/529, loss: 0.01659221760928631 2023-01-21 11:19:06.069975: step: 280/529, loss: 0.015396738424897194 2023-01-21 11:19:07.247898: step: 284/529, loss: 0.01416015625 2023-01-21 11:19:08.480505: step: 288/529, loss: 0.0008448601001873612 2023-01-21 11:19:09.656126: step: 292/529, loss: 0.02195291593670845 2023-01-21 11:19:10.847699: step: 296/529, loss: 0.011858178302645683 2023-01-21 11:19:12.052517: step: 300/529, loss: 0.020637130364775658 2023-01-21 11:19:13.293654: step: 304/529, loss: 0.02291898801922798 2023-01-21 11:19:14.510457: step: 308/529, loss: 0.012097597122192383 2023-01-21 11:19:15.666088: step: 312/529, loss: 0.0002045154687948525 2023-01-21 11:19:16.927233: step: 316/529, loss: 0.033960580825805664 2023-01-21 11:19:18.151574: step: 320/529, loss: 0.0037948605604469776 2023-01-21 11:19:19.339086: step: 324/529, loss: 0.000901722873095423 2023-01-21 11:19:20.552241: step: 328/529, loss: 0.026515580713748932 2023-01-21 11:19:21.736874: step: 332/529, loss: 0.0035042765084654093 2023-01-21 11:19:22.977568: step: 336/529, loss: 0.0032387732062488794 2023-01-21 11:19:24.183006: step: 340/529, loss: 0.09098567813634872 2023-01-21 11:19:25.386161: step: 344/529, loss: 0.03037567250430584 2023-01-21 11:19:26.621732: step: 348/529, loss: 0.009226626716554165 2023-01-21 11:19:27.852083: step: 352/529, loss: 0.01393747329711914 2023-01-21 11:19:29.049233: step: 356/529, loss: 0.00024585722712799907 2023-01-21 11:19:30.253980: step: 360/529, loss: 0.08063382655382156 2023-01-21 11:19:31.443264: step: 364/529, loss: 0.018328571692109108 2023-01-21 11:19:32.688755: step: 368/529, loss: 0.07306937873363495 2023-01-21 11:19:33.918760: step: 372/529, loss: 0.04751329496502876 2023-01-21 11:19:35.089346: step: 376/529, loss: 0.0012422561412677169 2023-01-21 11:19:36.265214: step: 380/529, loss: 0.07454396039247513 2023-01-21 11:19:37.465300: step: 384/529, loss: 0.0024165152572095394 2023-01-21 11:19:38.666562: step: 388/529, loss: 0.019887257367372513 2023-01-21 11:19:39.900560: step: 392/529, loss: 0.07866191864013672 2023-01-21 11:19:41.084451: step: 396/529, loss: 0.03600006178021431 2023-01-21 11:19:42.289957: step: 400/529, loss: 0.00916605070233345 2023-01-21 11:19:43.496796: step: 404/529, loss: 0.038260556757450104 2023-01-21 11:19:44.709000: step: 408/529, loss: 0.05882992595434189 2023-01-21 11:19:45.874699: step: 412/529, loss: 0.48201170563697815 2023-01-21 11:19:47.062508: step: 416/529, loss: 0.008504676632583141 2023-01-21 11:19:48.237436: step: 420/529, loss: 0.01841001585125923 2023-01-21 11:19:49.439179: step: 424/529, loss: 0.20601196587085724 2023-01-21 11:19:50.656088: step: 428/529, loss: 0.0026581764686852694 2023-01-21 11:19:51.848031: step: 432/529, loss: 0.018877126276493073 2023-01-21 11:19:53.016371: step: 436/529, loss: 0.06688833236694336 2023-01-21 11:19:54.204019: step: 440/529, loss: 0.04808664321899414 2023-01-21 11:19:55.433091: step: 444/529, loss: 0.022965431213378906 2023-01-21 11:19:56.623626: step: 448/529, loss: 0.005961370654404163 2023-01-21 11:19:57.792298: step: 452/529, loss: 0.012044096365571022 2023-01-21 11:19:58.995009: step: 456/529, loss: 0.025146102532744408 2023-01-21 11:20:00.182162: step: 460/529, loss: 0.03997688367962837 2023-01-21 11:20:01.365551: step: 464/529, loss: 0.0025017261505126953 2023-01-21 11:20:02.562811: step: 468/529, loss: 0.004339027684181929 2023-01-21 11:20:03.776897: step: 472/529, loss: 0.005260562989860773 2023-01-21 11:20:04.945275: step: 476/529, loss: 0.022475672885775566 2023-01-21 11:20:06.122457: step: 480/529, loss: 0.008837452158331871 2023-01-21 11:20:07.335906: step: 484/529, loss: 0.015871619805693626 2023-01-21 11:20:08.587383: step: 488/529, loss: 0.08900375664234161 2023-01-21 11:20:09.775223: step: 492/529, loss: 0.015891743823885918 2023-01-21 11:20:10.995542: step: 496/529, loss: 0.016161728650331497 2023-01-21 11:20:12.201186: step: 500/529, loss: 0.12802700698375702 2023-01-21 11:20:13.422163: step: 504/529, loss: 0.0031542659271508455 2023-01-21 11:20:14.618183: step: 508/529, loss: 0.0010539054637774825 2023-01-21 11:20:15.815517: step: 512/529, loss: 0.05295724794268608 2023-01-21 11:20:17.023607: step: 516/529, loss: 0.09179496765136719 2023-01-21 11:20:18.213284: step: 520/529, loss: 0.08914504200220108 2023-01-21 11:20:19.437834: step: 524/529, loss: 0.00392074603587389 2023-01-21 11:20:20.679178: step: 528/529, loss: 0.0016963481903076172 2023-01-21 11:20:21.912819: step: 532/529, loss: 0.003937530796974897 2023-01-21 11:20:23.152926: step: 536/529, loss: 0.03847523033618927 2023-01-21 11:20:24.392277: step: 540/529, loss: 0.01657574251294136 2023-01-21 11:20:25.559189: step: 544/529, loss: 0.0651828795671463 2023-01-21 11:20:26.755790: step: 548/529, loss: 0.007713747210800648 2023-01-21 11:20:27.929222: step: 552/529, loss: 0.025628376752138138 2023-01-21 11:20:29.073652: step: 556/529, loss: 0.0032600881531834602 2023-01-21 11:20:30.242795: step: 560/529, loss: 0.06552676856517792 2023-01-21 11:20:31.478856: step: 564/529, loss: 0.5084815621376038 2023-01-21 11:20:32.701203: step: 568/529, loss: 0.005863571539521217 2023-01-21 11:20:33.905621: step: 572/529, loss: 0.06808986514806747 2023-01-21 11:20:35.091861: step: 576/529, loss: 0.18612103164196014 2023-01-21 11:20:36.266247: step: 580/529, loss: 0.033710576593875885 2023-01-21 11:20:37.495722: step: 584/529, loss: 0.026553058996796608 2023-01-21 11:20:38.705775: step: 588/529, loss: 0.06056251376867294 2023-01-21 11:20:39.901093: step: 592/529, loss: 0.002029514405876398 2023-01-21 11:20:41.079777: step: 596/529, loss: 0.01723318174481392 2023-01-21 11:20:42.261529: step: 600/529, loss: 0.0005194663535803556 2023-01-21 11:20:43.446923: step: 604/529, loss: 0.02321462519466877 2023-01-21 11:20:44.642202: step: 608/529, loss: 0.0063208104111254215 2023-01-21 11:20:45.852236: step: 612/529, loss: 0.029242612421512604 2023-01-21 11:20:47.078706: step: 616/529, loss: 0.02488880231976509 2023-01-21 11:20:48.325451: step: 620/529, loss: 0.02666320651769638 2023-01-21 11:20:49.512907: step: 624/529, loss: 0.017958641052246094 2023-01-21 11:20:50.710328: step: 628/529, loss: 0.03376655653119087 2023-01-21 11:20:51.906964: step: 632/529, loss: 0.08083219826221466 2023-01-21 11:20:53.079017: step: 636/529, loss: 0.00589599646627903 2023-01-21 11:20:54.302982: step: 640/529, loss: 0.003988933749496937 2023-01-21 11:20:55.448552: step: 644/529, loss: 0.043036654591560364 2023-01-21 11:20:56.596313: step: 648/529, loss: 0.0315793976187706 2023-01-21 11:20:57.788601: step: 652/529, loss: 0.03280750662088394 2023-01-21 11:20:58.961774: step: 656/529, loss: 0.0444762259721756 2023-01-21 11:21:00.169717: step: 660/529, loss: 0.2329617589712143 2023-01-21 11:21:01.336661: step: 664/529, loss: 0.009524727240204811 2023-01-21 11:21:02.533995: step: 668/529, loss: 0.022509289905428886 2023-01-21 11:21:03.741344: step: 672/529, loss: 0.037732888013124466 2023-01-21 11:21:04.959312: step: 676/529, loss: 0.03295574337244034 2023-01-21 11:21:06.145920: step: 680/529, loss: 0.02532811276614666 2023-01-21 11:21:07.325966: step: 684/529, loss: 0.07379178702831268 2023-01-21 11:21:08.527288: step: 688/529, loss: 0.008675575256347656 2023-01-21 11:21:09.712238: step: 692/529, loss: 0.02624817006289959 2023-01-21 11:21:10.890124: step: 696/529, loss: 0.004151344299316406 2023-01-21 11:21:12.086145: step: 700/529, loss: 0.003677845001220703 2023-01-21 11:21:13.275120: step: 704/529, loss: 0.1063636764883995 2023-01-21 11:21:14.466594: step: 708/529, loss: 0.03356208652257919 2023-01-21 11:21:15.686748: step: 712/529, loss: 0.0001941680966410786 2023-01-21 11:21:16.883864: step: 716/529, loss: 0.21397551894187927 2023-01-21 11:21:18.060977: step: 720/529, loss: 0.0016517640324309468 2023-01-21 11:21:19.289729: step: 724/529, loss: 0.05578498914837837 2023-01-21 11:21:20.509573: step: 728/529, loss: 0.07641620934009552 2023-01-21 11:21:21.706089: step: 732/529, loss: 0.01671614684164524 2023-01-21 11:21:22.880457: step: 736/529, loss: 0.003430795855820179 2023-01-21 11:21:24.055440: step: 740/529, loss: 0.9274988174438477 2023-01-21 11:21:25.249504: step: 744/529, loss: 0.07541751861572266 2023-01-21 11:21:26.413364: step: 748/529, loss: 0.053116895258426666 2023-01-21 11:21:27.615709: step: 752/529, loss: 0.011973142623901367 2023-01-21 11:21:28.778233: step: 756/529, loss: 0.06626713275909424 2023-01-21 11:21:29.967256: step: 760/529, loss: 0.030478190630674362 2023-01-21 11:21:31.178246: step: 764/529, loss: 0.025000954046845436 2023-01-21 11:21:32.412717: step: 768/529, loss: 0.042382147163152695 2023-01-21 11:21:33.610026: step: 772/529, loss: 0.018733978271484375 2023-01-21 11:21:34.784402: step: 776/529, loss: 0.020431995391845703 2023-01-21 11:21:35.981793: step: 780/529, loss: 0.004310572519898415 2023-01-21 11:21:37.167348: step: 784/529, loss: 0.0185564998537302 2023-01-21 11:21:38.345602: step: 788/529, loss: 0.038434840738773346 2023-01-21 11:21:39.546556: step: 792/529, loss: 0.019561195746064186 2023-01-21 11:21:40.725116: step: 796/529, loss: 0.06027545779943466 2023-01-21 11:21:41.946228: step: 800/529, loss: 0.0046752928756177425 2023-01-21 11:21:43.116151: step: 804/529, loss: 0.019846344366669655 2023-01-21 11:21:44.303674: step: 808/529, loss: 0.006213807966560125 2023-01-21 11:21:45.528798: step: 812/529, loss: 0.11909833550453186 2023-01-21 11:21:46.722531: step: 816/529, loss: 0.13234928250312805 2023-01-21 11:21:47.913255: step: 820/529, loss: 0.011299706064164639 2023-01-21 11:21:49.090431: step: 824/529, loss: 0.33778563141822815 2023-01-21 11:21:50.279274: step: 828/529, loss: 0.002963733859360218 2023-01-21 11:21:51.523059: step: 832/529, loss: 0.016438674181699753 2023-01-21 11:21:52.691093: step: 836/529, loss: 0.13288964331150055 2023-01-21 11:21:53.866229: step: 840/529, loss: 0.05501976236701012 2023-01-21 11:21:55.079009: step: 844/529, loss: 0.011426448822021484 2023-01-21 11:21:56.271670: step: 848/529, loss: 0.014819907955825329 2023-01-21 11:21:57.503659: step: 852/529, loss: 0.033202577382326126 2023-01-21 11:21:58.703087: step: 856/529, loss: 0.03636999428272247 2023-01-21 11:21:59.913844: step: 860/529, loss: 0.012090111151337624 2023-01-21 11:22:01.141344: step: 864/529, loss: 0.005740070249885321 2023-01-21 11:22:02.390063: step: 868/529, loss: 0.0006246566772460938 2023-01-21 11:22:03.541330: step: 872/529, loss: 0.0747474730014801 2023-01-21 11:22:04.725485: step: 876/529, loss: 0.010081958957016468 2023-01-21 11:22:05.902618: step: 880/529, loss: 0.024919986724853516 2023-01-21 11:22:07.125916: step: 884/529, loss: 0.02447795867919922 2023-01-21 11:22:08.308806: step: 888/529, loss: 0.0512942336499691 2023-01-21 11:22:09.519415: step: 892/529, loss: 0.023643113672733307 2023-01-21 11:22:10.700611: step: 896/529, loss: 0.051764197647571564 2023-01-21 11:22:11.907382: step: 900/529, loss: 0.046014074236154556 2023-01-21 11:22:13.090513: step: 904/529, loss: 0.006800270173698664 2023-01-21 11:22:14.332271: step: 908/529, loss: 0.03627128526568413 2023-01-21 11:22:15.490641: step: 912/529, loss: 0.006840634159743786 2023-01-21 11:22:16.680515: step: 916/529, loss: 0.07601527869701385 2023-01-21 11:22:17.901774: step: 920/529, loss: 0.008344730362296104 2023-01-21 11:22:19.069505: step: 924/529, loss: 0.08563299477100372 2023-01-21 11:22:20.288615: step: 928/529, loss: 0.004166221711784601 2023-01-21 11:22:21.454226: step: 932/529, loss: 0.0009152413113042712 2023-01-21 11:22:22.670163: step: 936/529, loss: 0.06440772861242294 2023-01-21 11:22:23.881158: step: 940/529, loss: 0.10162906348705292 2023-01-21 11:22:25.107440: step: 944/529, loss: 0.025232411921024323 2023-01-21 11:22:26.340788: step: 948/529, loss: 0.002295208163559437 2023-01-21 11:22:27.547826: step: 952/529, loss: 0.009371376596391201 2023-01-21 11:22:28.725489: step: 956/529, loss: 0.010438680648803711 2023-01-21 11:22:29.912902: step: 960/529, loss: 0.04470653831958771 2023-01-21 11:22:31.088983: step: 964/529, loss: 0.042991068214178085 2023-01-21 11:22:32.293122: step: 968/529, loss: 0.02541675604879856 2023-01-21 11:22:33.481317: step: 972/529, loss: 0.0003917694266419858 2023-01-21 11:22:34.647496: step: 976/529, loss: 0.06460399925708771 2023-01-21 11:22:35.868497: step: 980/529, loss: 0.018777180463075638 2023-01-21 11:22:37.078270: step: 984/529, loss: 0.013294219970703125 2023-01-21 11:22:38.331242: step: 988/529, loss: 0.0044689178466796875 2023-01-21 11:22:39.545204: step: 992/529, loss: 0.01676197163760662 2023-01-21 11:22:40.726623: step: 996/529, loss: 0.0016300200950354338 2023-01-21 11:22:41.916398: step: 1000/529, loss: 0.06373615562915802 2023-01-21 11:22:43.124911: step: 1004/529, loss: 0.011961079202592373 2023-01-21 11:22:44.345798: step: 1008/529, loss: 0.016974449157714844 2023-01-21 11:22:45.541199: step: 1012/529, loss: 0.04460439831018448 2023-01-21 11:22:46.720368: step: 1016/529, loss: 0.0014761924976482987 2023-01-21 11:22:47.923024: step: 1020/529, loss: 0.002438449999317527 2023-01-21 11:22:49.123970: step: 1024/529, loss: 0.06784658879041672 2023-01-21 11:22:50.342131: step: 1028/529, loss: 5.356073233997449e-05 2023-01-21 11:22:51.521243: step: 1032/529, loss: 0.0034419295843690634 2023-01-21 11:22:52.709193: step: 1036/529, loss: 0.04433612525463104 2023-01-21 11:22:53.864300: step: 1040/529, loss: 0.021614840254187584 2023-01-21 11:22:55.058454: step: 1044/529, loss: 0.00931086577475071 2023-01-21 11:22:56.255878: step: 1048/529, loss: 0.00672181835398078 2023-01-21 11:22:57.462971: step: 1052/529, loss: 0.011656666174530983 2023-01-21 11:22:58.678449: step: 1056/529, loss: 0.003907013218849897 2023-01-21 11:22:59.880645: step: 1060/529, loss: 0.014815330505371094 2023-01-21 11:23:01.105565: step: 1064/529, loss: 0.04444169998168945 2023-01-21 11:23:02.312758: step: 1068/529, loss: 0.022906780242919922 2023-01-21 11:23:03.531320: step: 1072/529, loss: 0.04141044616699219 2023-01-21 11:23:04.782437: step: 1076/529, loss: 0.027387570589780807 2023-01-21 11:23:06.038408: step: 1080/529, loss: 0.0660918653011322 2023-01-21 11:23:07.271199: step: 1084/529, loss: 0.07434835284948349 2023-01-21 11:23:08.469202: step: 1088/529, loss: 0.009459877386689186 2023-01-21 11:23:09.632855: step: 1092/529, loss: 0.00022692680067848414 2023-01-21 11:23:10.841219: step: 1096/529, loss: 0.04892406240105629 2023-01-21 11:23:12.015348: step: 1100/529, loss: 0.00038533209590241313 2023-01-21 11:23:13.204678: step: 1104/529, loss: 0.00266265869140625 2023-01-21 11:23:14.381560: step: 1108/529, loss: 0.006957817357033491 2023-01-21 11:23:15.598631: step: 1112/529, loss: 0.018420647829771042 2023-01-21 11:23:16.818415: step: 1116/529, loss: 0.00014655590348411351 2023-01-21 11:23:18.017565: step: 1120/529, loss: 0.020693015307188034 2023-01-21 11:23:19.228085: step: 1124/529, loss: 0.062240030616521835 2023-01-21 11:23:20.382414: step: 1128/529, loss: 0.0307219997048378 2023-01-21 11:23:21.543908: step: 1132/529, loss: 0.054282285273075104 2023-01-21 11:23:22.747846: step: 1136/529, loss: 0.060857199132442474 2023-01-21 11:23:23.953208: step: 1140/529, loss: 0.011554241180419922 2023-01-21 11:23:25.174261: step: 1144/529, loss: 0.09443559497594833 2023-01-21 11:23:26.364796: step: 1148/529, loss: 0.006712627597153187 2023-01-21 11:23:27.592754: step: 1152/529, loss: 0.0019847869407385588 2023-01-21 11:23:28.834778: step: 1156/529, loss: 0.04674654081463814 2023-01-21 11:23:30.051010: step: 1160/529, loss: 0.10736479610204697 2023-01-21 11:23:31.208505: step: 1164/529, loss: 0.06859998404979706 2023-01-21 11:23:32.440432: step: 1168/529, loss: 0.04300422593951225 2023-01-21 11:23:33.608111: step: 1172/529, loss: 0.003788185305893421 2023-01-21 11:23:34.801287: step: 1176/529, loss: 0.05814613029360771 2023-01-21 11:23:35.976721: step: 1180/529, loss: 0.01941204071044922 2023-01-21 11:23:37.219407: step: 1184/529, loss: 0.02299346961081028 2023-01-21 11:23:38.385193: step: 1188/529, loss: 0.029853487387299538 2023-01-21 11:23:39.608250: step: 1192/529, loss: 0.012223467230796814 2023-01-21 11:23:40.818088: step: 1196/529, loss: 0.009244346991181374 2023-01-21 11:23:41.968645: step: 1200/529, loss: 1.0814969539642334 2023-01-21 11:23:43.142122: step: 1204/529, loss: 0.07040288299322128 2023-01-21 11:23:44.352324: step: 1208/529, loss: 0.022809602320194244 2023-01-21 11:23:45.558664: step: 1212/529, loss: 0.03156089782714844 2023-01-21 11:23:46.771683: step: 1216/529, loss: 0.04622860252857208 2023-01-21 11:23:47.956102: step: 1220/529, loss: 0.003847408341243863 2023-01-21 11:23:49.171298: step: 1224/529, loss: 0.03626842424273491 2023-01-21 11:23:50.390401: step: 1228/529, loss: 0.028188131749629974 2023-01-21 11:23:51.593137: step: 1232/529, loss: 0.03349714353680611 2023-01-21 11:23:52.805410: step: 1236/529, loss: 0.0012277603382244706 2023-01-21 11:23:54.024144: step: 1240/529, loss: 0.01671772077679634 2023-01-21 11:23:55.198702: step: 1244/529, loss: 0.016425132751464844 2023-01-21 11:23:56.421092: step: 1248/529, loss: 0.022949744015932083 2023-01-21 11:23:57.596186: step: 1252/529, loss: 0.00038061142549850047 2023-01-21 11:23:58.806411: step: 1256/529, loss: 0.023551560938358307 2023-01-21 11:23:59.969092: step: 1260/529, loss: 0.00201835623010993 2023-01-21 11:24:01.135634: step: 1264/529, loss: 0.05218858644366264 2023-01-21 11:24:02.349228: step: 1268/529, loss: 0.01800518110394478 2023-01-21 11:24:03.561515: step: 1272/529, loss: 0.005594968795776367 2023-01-21 11:24:04.718856: step: 1276/529, loss: 0.000411796587286517 2023-01-21 11:24:05.915919: step: 1280/529, loss: 0.017890071496367455 2023-01-21 11:24:07.150047: step: 1284/529, loss: 0.036007121205329895 2023-01-21 11:24:08.344138: step: 1288/529, loss: 0.008306790143251419 2023-01-21 11:24:09.499086: step: 1292/529, loss: 0.003742170287296176 2023-01-21 11:24:10.663688: step: 1296/529, loss: 0.013090801425278187 2023-01-21 11:24:11.874584: step: 1300/529, loss: 0.0004913330194540322 2023-01-21 11:24:13.038351: step: 1304/529, loss: 0.10652267932891846 2023-01-21 11:24:14.275009: step: 1308/529, loss: 0.03825564309954643 2023-01-21 11:24:15.471011: step: 1312/529, loss: 0.07320308685302734 2023-01-21 11:24:16.708497: step: 1316/529, loss: 0.17192836105823517 2023-01-21 11:24:17.943116: step: 1320/529, loss: 0.03330964967608452 2023-01-21 11:24:19.155103: step: 1324/529, loss: 0.014474963769316673 2023-01-21 11:24:20.354626: step: 1328/529, loss: 0.11852865666151047 2023-01-21 11:24:21.527741: step: 1332/529, loss: 0.0608641617000103 2023-01-21 11:24:22.722303: step: 1336/529, loss: 0.001866340753622353 2023-01-21 11:24:23.946298: step: 1340/529, loss: 0.006926822941750288 2023-01-21 11:24:25.142281: step: 1344/529, loss: 0.00719451904296875 2023-01-21 11:24:26.378554: step: 1348/529, loss: 0.008491134271025658 2023-01-21 11:24:27.567019: step: 1352/529, loss: 0.04129600524902344 2023-01-21 11:24:28.756482: step: 1356/529, loss: 0.00010333061800338328 2023-01-21 11:24:29.944527: step: 1360/529, loss: 0.0492795929312706 2023-01-21 11:24:31.121436: step: 1364/529, loss: 0.0018196107121184468 2023-01-21 11:24:32.394636: step: 1368/529, loss: 0.03764224052429199 2023-01-21 11:24:33.670351: step: 1372/529, loss: 0.10187163949012756 2023-01-21 11:24:34.866763: step: 1376/529, loss: 0.08095722645521164 2023-01-21 11:24:36.102779: step: 1380/529, loss: 0.04181652143597603 2023-01-21 11:24:37.305703: step: 1384/529, loss: 0.05039720609784126 2023-01-21 11:24:38.481075: step: 1388/529, loss: 0.0014057159423828125 2023-01-21 11:24:39.691211: step: 1392/529, loss: 0.009013366885483265 2023-01-21 11:24:40.862088: step: 1396/529, loss: 0.03609433025121689 2023-01-21 11:24:42.087220: step: 1400/529, loss: 0.005547666922211647 2023-01-21 11:24:43.301065: step: 1404/529, loss: 0.06917114555835724 2023-01-21 11:24:44.480421: step: 1408/529, loss: 0.018605804070830345 2023-01-21 11:24:45.655518: step: 1412/529, loss: 0.03199462965130806 2023-01-21 11:24:46.887919: step: 1416/529, loss: 0.005234909243881702 2023-01-21 11:24:48.061767: step: 1420/529, loss: 0.015880774706602097 2023-01-21 11:24:49.273164: step: 1424/529, loss: 0.0042021749541163445 2023-01-21 11:24:50.500508: step: 1428/529, loss: 0.016536809504032135 2023-01-21 11:24:51.697995: step: 1432/529, loss: 0.03170328214764595 2023-01-21 11:24:52.860797: step: 1436/529, loss: 0.02615375444293022 2023-01-21 11:24:54.090086: step: 1440/529, loss: 0.056096553802490234 2023-01-21 11:24:55.260976: step: 1444/529, loss: 0.03340788185596466 2023-01-21 11:24:56.454783: step: 1448/529, loss: 0.19283849000930786 2023-01-21 11:24:57.670004: step: 1452/529, loss: 0.23090897500514984 2023-01-21 11:24:58.867791: step: 1456/529, loss: 0.0010876655578613281 2023-01-21 11:25:00.093019: step: 1460/529, loss: 0.014848804101347923 2023-01-21 11:25:01.299652: step: 1464/529, loss: 0.07609906047582626 2023-01-21 11:25:02.488133: step: 1468/529, loss: 0.09213419258594513 2023-01-21 11:25:03.690331: step: 1472/529, loss: 0.0009335518116131425 2023-01-21 11:25:04.848912: step: 1476/529, loss: 0.07358980178833008 2023-01-21 11:25:06.062745: step: 1480/529, loss: 0.06184225156903267 2023-01-21 11:25:07.317779: step: 1484/529, loss: 0.017884254455566406 2023-01-21 11:25:08.526867: step: 1488/529, loss: 0.017476225271821022 2023-01-21 11:25:09.737199: step: 1492/529, loss: 0.004854679573327303 2023-01-21 11:25:10.905335: step: 1496/529, loss: 0.010556602850556374 2023-01-21 11:25:12.074952: step: 1500/529, loss: 0.00023078919912222773 2023-01-21 11:25:13.251902: step: 1504/529, loss: 0.02437286451458931 2023-01-21 11:25:14.479997: step: 1508/529, loss: 0.03038806840777397 2023-01-21 11:25:15.665341: step: 1512/529, loss: 0.00694322632625699 2023-01-21 11:25:16.853488: step: 1516/529, loss: 0.004525661468505859 2023-01-21 11:25:18.077192: step: 1520/529, loss: 0.03191728889942169 2023-01-21 11:25:19.251226: step: 1524/529, loss: 0.020081328228116035 2023-01-21 11:25:20.422959: step: 1528/529, loss: 0.02995023876428604 2023-01-21 11:25:21.609686: step: 1532/529, loss: 0.006091165356338024 2023-01-21 11:25:22.797600: step: 1536/529, loss: 0.04897952079772949 2023-01-21 11:25:23.975784: step: 1540/529, loss: 0.016426658257842064 2023-01-21 11:25:25.181915: step: 1544/529, loss: 0.011411094106733799 2023-01-21 11:25:26.334596: step: 1548/529, loss: 0.01201553363353014 2023-01-21 11:25:27.564537: step: 1552/529, loss: 0.02086820639669895 2023-01-21 11:25:28.815313: step: 1556/529, loss: 0.027756690979003906 2023-01-21 11:25:30.005005: step: 1560/529, loss: 0.039585113525390625 2023-01-21 11:25:31.204912: step: 1564/529, loss: 0.00830750446766615 2023-01-21 11:25:32.444170: step: 1568/529, loss: 0.1655474603176117 2023-01-21 11:25:33.640466: step: 1572/529, loss: 0.023420382291078568 2023-01-21 11:25:34.848724: step: 1576/529, loss: 0.03487744182348251 2023-01-21 11:25:36.073498: step: 1580/529, loss: 0.021817969158291817 2023-01-21 11:25:37.283274: step: 1584/529, loss: 0.21566659212112427 2023-01-21 11:25:38.459057: step: 1588/529, loss: 0.008579397574067116 2023-01-21 11:25:39.639263: step: 1592/529, loss: 0.027110768482089043 2023-01-21 11:25:40.820975: step: 1596/529, loss: 0.006395912263542414 2023-01-21 11:25:41.991215: step: 1600/529, loss: 0.037157345563173294 2023-01-21 11:25:43.151308: step: 1604/529, loss: 0.0024506808258593082 2023-01-21 11:25:44.356896: step: 1608/529, loss: 0.007738590240478516 2023-01-21 11:25:45.546810: step: 1612/529, loss: 0.043433211743831635 2023-01-21 11:25:46.732116: step: 1616/529, loss: 0.06994209438562393 2023-01-21 11:25:47.930712: step: 1620/529, loss: 0.03247642517089844 2023-01-21 11:25:49.126234: step: 1624/529, loss: 0.01715526543557644 2023-01-21 11:25:50.310106: step: 1628/529, loss: 0.09717759490013123 2023-01-21 11:25:51.575441: step: 1632/529, loss: 0.0369015708565712 2023-01-21 11:25:52.807667: step: 1636/529, loss: 0.008671188727021217 2023-01-21 11:25:54.030014: step: 1640/529, loss: 0.028925515711307526 2023-01-21 11:25:55.234881: step: 1644/529, loss: 0.06007962301373482 2023-01-21 11:25:56.402119: step: 1648/529, loss: 0.10519365966320038 2023-01-21 11:25:57.646064: step: 1652/529, loss: 0.023768329992890358 2023-01-21 11:25:58.815536: step: 1656/529, loss: 0.008392333984375 2023-01-21 11:26:00.023826: step: 1660/529, loss: 0.027727318927645683 2023-01-21 11:26:01.226484: step: 1664/529, loss: 0.04635639116168022 2023-01-21 11:26:02.472422: step: 1668/529, loss: 0.053844548761844635 2023-01-21 11:26:03.635783: step: 1672/529, loss: 0.00657577533274889 2023-01-21 11:26:04.842690: step: 1676/529, loss: 0.0302339568734169 2023-01-21 11:26:06.056680: step: 1680/529, loss: 0.05921907722949982 2023-01-21 11:26:07.229070: step: 1684/529, loss: 0.13537339866161346 2023-01-21 11:26:08.452979: step: 1688/529, loss: 0.012246895581483841 2023-01-21 11:26:09.680684: step: 1692/529, loss: 0.05579128488898277 2023-01-21 11:26:10.853939: step: 1696/529, loss: 0.04338718578219414 2023-01-21 11:26:12.049338: step: 1700/529, loss: 0.026494218036532402 2023-01-21 11:26:13.233023: step: 1704/529, loss: 0.001966858049854636 2023-01-21 11:26:14.401572: step: 1708/529, loss: 0.05786151811480522 2023-01-21 11:26:15.576061: step: 1712/529, loss: 0.005179214756935835 2023-01-21 11:26:16.771870: step: 1716/529, loss: 0.0737430602312088 2023-01-21 11:26:18.014238: step: 1720/529, loss: 0.021807575598359108 2023-01-21 11:26:19.207280: step: 1724/529, loss: 0.0010194778442382812 2023-01-21 11:26:20.426624: step: 1728/529, loss: 0.061908818781375885 2023-01-21 11:26:21.657095: step: 1732/529, loss: 0.0765039250254631 2023-01-21 11:26:22.877249: step: 1736/529, loss: 0.005922412965446711 2023-01-21 11:26:24.067405: step: 1740/529, loss: 0.3252026438713074 2023-01-21 11:26:25.272166: step: 1744/529, loss: 0.0926109328866005 2023-01-21 11:26:26.459087: step: 1748/529, loss: 0.027489136904478073 2023-01-21 11:26:27.659248: step: 1752/529, loss: 0.0010555267799645662 2023-01-21 11:26:28.865059: step: 1756/529, loss: 0.016663696616888046 2023-01-21 11:26:30.055537: step: 1760/529, loss: 0.0016357898712158203 2023-01-21 11:26:31.251755: step: 1764/529, loss: 0.014433860778808594 2023-01-21 11:26:32.465080: step: 1768/529, loss: 0.03584950044751167 2023-01-21 11:26:33.684325: step: 1772/529, loss: 0.0376676581799984 2023-01-21 11:26:34.824915: step: 1776/529, loss: 0.0016571879386901855 2023-01-21 11:26:36.003071: step: 1780/529, loss: 0.04992799460887909 2023-01-21 11:26:37.200576: step: 1784/529, loss: 0.0011640548473224044 2023-01-21 11:26:38.376810: step: 1788/529, loss: 0.00037755968514829874 2023-01-21 11:26:39.545195: step: 1792/529, loss: 0.0003247261047363281 2023-01-21 11:26:40.770955: step: 1796/529, loss: 0.06704292446374893 2023-01-21 11:26:41.985794: step: 1800/529, loss: 0.020928382873535156 2023-01-21 11:26:43.174046: step: 1804/529, loss: 0.02228565141558647 2023-01-21 11:26:44.414687: step: 1808/529, loss: 0.041188813745975494 2023-01-21 11:26:45.613706: step: 1812/529, loss: 0.02476959303021431 2023-01-21 11:26:46.815074: step: 1816/529, loss: 0.04090118780732155 2023-01-21 11:26:48.032999: step: 1820/529, loss: 0.09428396075963974 2023-01-21 11:26:49.247350: step: 1824/529, loss: 0.016428424045443535 2023-01-21 11:26:50.413985: step: 1828/529, loss: 0.1583634465932846 2023-01-21 11:26:51.623503: step: 1832/529, loss: 0.002458715345710516 2023-01-21 11:26:52.808477: step: 1836/529, loss: 0.5011193752288818 2023-01-21 11:26:53.946048: step: 1840/529, loss: 0.004469394683837891 2023-01-21 11:26:55.135546: step: 1844/529, loss: 0.026998329907655716 2023-01-21 11:26:56.341944: step: 1848/529, loss: 0.0014792443253099918 2023-01-21 11:26:57.545051: step: 1852/529, loss: 0.07603378593921661 2023-01-21 11:26:58.728494: step: 1856/529, loss: 0.020480729639530182 2023-01-21 11:26:59.944587: step: 1860/529, loss: 0.06016102060675621 2023-01-21 11:27:01.131480: step: 1864/529, loss: 0.05631885677576065 2023-01-21 11:27:02.304660: step: 1868/529, loss: 0.02674713358283043 2023-01-21 11:27:03.542412: step: 1872/529, loss: 0.016243457794189453 2023-01-21 11:27:04.753855: step: 1876/529, loss: 0.03381548076868057 2023-01-21 11:27:05.939586: step: 1880/529, loss: 0.11983232200145721 2023-01-21 11:27:07.103668: step: 1884/529, loss: 0.004639292135834694 2023-01-21 11:27:08.316104: step: 1888/529, loss: 0.01834392547607422 2023-01-21 11:27:09.475624: step: 1892/529, loss: 0.025397300720214844 2023-01-21 11:27:10.657027: step: 1896/529, loss: 0.044028472155332565 2023-01-21 11:27:11.849912: step: 1900/529, loss: 0.025841571390628815 2023-01-21 11:27:13.077939: step: 1904/529, loss: 0.06521806865930557 2023-01-21 11:27:14.310274: step: 1908/529, loss: 0.0029126168228685856 2023-01-21 11:27:15.479596: step: 1912/529, loss: 0.029267311096191406 2023-01-21 11:27:16.675274: step: 1916/529, loss: 0.0072495462372899055 2023-01-21 11:27:17.884477: step: 1920/529, loss: 0.019012833014130592 2023-01-21 11:27:19.096874: step: 1924/529, loss: 0.05298023298382759 2023-01-21 11:27:20.316631: step: 1928/529, loss: 0.03023510053753853 2023-01-21 11:27:21.522240: step: 1932/529, loss: 0.010569954290986061 2023-01-21 11:27:22.712279: step: 1936/529, loss: 0.023093413561582565 2023-01-21 11:27:23.908498: step: 1940/529, loss: 0.033271122723817825 2023-01-21 11:27:25.113334: step: 1944/529, loss: 0.03945844620466232 2023-01-21 11:27:26.292176: step: 1948/529, loss: 0.025533199310302734 2023-01-21 11:27:27.491519: step: 1952/529, loss: 0.00640106201171875 2023-01-21 11:27:28.734885: step: 1956/529, loss: 0.0009051322704181075 2023-01-21 11:27:29.920784: step: 1960/529, loss: 0.028422977775335312 2023-01-21 11:27:31.118572: step: 1964/529, loss: 0.00532188406214118 2023-01-21 11:27:32.304677: step: 1968/529, loss: 0.021473407745361328 2023-01-21 11:27:33.516275: step: 1972/529, loss: 0.014600562863051891 2023-01-21 11:27:34.703629: step: 1976/529, loss: 0.004134654998779297 2023-01-21 11:27:35.909297: step: 1980/529, loss: 0.01762828789651394 2023-01-21 11:27:37.127894: step: 1984/529, loss: 0.0854034423828125 2023-01-21 11:27:38.342181: step: 1988/529, loss: 0.013809156604111195 2023-01-21 11:27:39.535109: step: 1992/529, loss: 0.007717037573456764 2023-01-21 11:27:40.732592: step: 1996/529, loss: 0.05528444051742554 2023-01-21 11:27:41.906948: step: 2000/529, loss: 0.013115406967699528 2023-01-21 11:27:43.119207: step: 2004/529, loss: 0.007703018374741077 2023-01-21 11:27:44.312125: step: 2008/529, loss: 0.09631047397851944 2023-01-21 11:27:45.484546: step: 2012/529, loss: 0.0621306411921978 2023-01-21 11:27:46.697093: step: 2016/529, loss: 0.0782390609383583 2023-01-21 11:27:47.897951: step: 2020/529, loss: 0.02251298539340496 2023-01-21 11:27:49.078909: step: 2024/529, loss: 0.04134407266974449 2023-01-21 11:27:50.321597: step: 2028/529, loss: 0.004563379101455212 2023-01-21 11:27:51.482862: step: 2032/529, loss: 0.015045547857880592 2023-01-21 11:27:52.703737: step: 2036/529, loss: 0.028399229049682617 2023-01-21 11:27:53.915872: step: 2040/529, loss: 0.023426437750458717 2023-01-21 11:27:55.128642: step: 2044/529, loss: 0.003564548445865512 2023-01-21 11:27:56.377723: step: 2048/529, loss: 0.19866323471069336 2023-01-21 11:27:57.579327: step: 2052/529, loss: 0.02984476089477539 2023-01-21 11:27:58.737736: step: 2056/529, loss: 0.006458664312958717 2023-01-21 11:27:59.958012: step: 2060/529, loss: 0.021860789507627487 2023-01-21 11:28:01.191903: step: 2064/529, loss: 0.038942813873291016 2023-01-21 11:28:02.478472: step: 2068/529, loss: 0.0885000228881836 2023-01-21 11:28:03.668438: step: 2072/529, loss: 0.03750648349523544 2023-01-21 11:28:04.853447: step: 2076/529, loss: 0.03317384794354439 2023-01-21 11:28:06.008827: step: 2080/529, loss: 0.05186500772833824 2023-01-21 11:28:07.185173: step: 2084/529, loss: 0.5616950988769531 2023-01-21 11:28:08.342947: step: 2088/529, loss: 0.00017333030700683594 2023-01-21 11:28:09.576016: step: 2092/529, loss: 0.024314500391483307 2023-01-21 11:28:10.735279: step: 2096/529, loss: 0.010793877765536308 2023-01-21 11:28:11.939590: step: 2100/529, loss: 0.14183488488197327 2023-01-21 11:28:13.155287: step: 2104/529, loss: 0.022354889661073685 2023-01-21 11:28:14.381525: step: 2108/529, loss: 0.24488191306591034 2023-01-21 11:28:15.533018: step: 2112/529, loss: 0.01973886415362358 2023-01-21 11:28:16.722050: step: 2116/529, loss: 0.023023415356874466 ================================================== Loss: 0.049 -------------------- Dev: {'event': {'p': 0.6347634763476347, 'r': 0.7683089214380826, 'f1': 0.6951807228915663}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Test: {'event': {'p': 0.6570852244074634, 'r': 0.78070701018574, 'f1': 0.7135815991237677}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Chinese: {'event': {'p': 0.6164383561643836, 'r': 0.8333333333333334, 'f1': 0.7086614173228347}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Korean: {'event': {'p': 0.5769230769230769, 'r': 0.47619047619047616, 'f1': 0.5217391304347826}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} 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': 14} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6, 'r': 0.7470039946737683, 'f1': 0.6654804270462634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Russian: {'event': {'p': 0.630667345899134, 'r': 0.7417615338526063, 'f1': 0.6817180616740088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ****************************** Epoch: 15 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:29:00.670588: step: 4/529, loss: 0.03386259078979492 2023-01-21 11:29:01.912116: step: 8/529, loss: 0.01895732991397381 2023-01-21 11:29:03.143808: step: 12/529, loss: 0.004780387971550226 2023-01-21 11:29:04.350630: step: 16/529, loss: 0.0059111593291163445 2023-01-21 11:29:05.565087: step: 20/529, loss: 0.009102058596909046 2023-01-21 11:29:06.765554: step: 24/529, loss: 0.0008537292596884072 2023-01-21 11:29:07.936205: step: 28/529, loss: 0.030220700427889824 2023-01-21 11:29:09.159819: step: 32/529, loss: 0.01554260216653347 2023-01-21 11:29:10.364272: step: 36/529, loss: 0.03621234744787216 2023-01-21 11:29:11.576889: step: 40/529, loss: 0.007046795450150967 2023-01-21 11:29:12.769766: step: 44/529, loss: 0.041281700134277344 2023-01-21 11:29:13.973369: step: 48/529, loss: 0.016413306817412376 2023-01-21 11:29:15.207086: step: 52/529, loss: 0.313218891620636 2023-01-21 11:29:16.451382: step: 56/529, loss: 0.007563972845673561 2023-01-21 11:29:17.654106: step: 60/529, loss: 0.0021259309723973274 2023-01-21 11:29:18.846376: step: 64/529, loss: 0.014450454153120518 2023-01-21 11:29:20.037463: step: 68/529, loss: 0.013047791086137295 2023-01-21 11:29:21.237426: step: 72/529, loss: 0.025946902111172676 2023-01-21 11:29:22.402115: step: 76/529, loss: 0.0008395194890908897 2023-01-21 11:29:23.586932: step: 80/529, loss: 0.027770616114139557 2023-01-21 11:29:24.802724: step: 84/529, loss: 0.043586160987615585 2023-01-21 11:29:26.029245: step: 88/529, loss: 0.011481857858598232 2023-01-21 11:29:27.213486: step: 92/529, loss: 0.042740631848573685 2023-01-21 11:29:28.404304: step: 96/529, loss: 0.054631710052490234 2023-01-21 11:29:29.608447: step: 100/529, loss: 0.0072231292724609375 2023-01-21 11:29:30.790411: step: 104/529, loss: 0.045874498784542084 2023-01-21 11:29:31.956126: step: 108/529, loss: 0.031756602227687836 2023-01-21 11:29:33.137631: step: 112/529, loss: 0.00864114798605442 2023-01-21 11:29:34.295104: step: 116/529, loss: 0.0062046051025390625 2023-01-21 11:29:35.527055: step: 120/529, loss: 0.09919299930334091 2023-01-21 11:29:36.697180: step: 124/529, loss: 0.02214345894753933 2023-01-21 11:29:37.920433: step: 128/529, loss: 0.021587181836366653 2023-01-21 11:29:39.108075: step: 132/529, loss: 0.009229565039277077 2023-01-21 11:29:40.326979: step: 136/529, loss: 0.0022395134437829256 2023-01-21 11:29:41.547075: step: 140/529, loss: 0.008449936285614967 2023-01-21 11:29:42.711960: step: 144/529, loss: 0.04003726318478584 2023-01-21 11:29:43.917053: step: 148/529, loss: 0.03970833122730255 2023-01-21 11:29:45.069810: step: 152/529, loss: 0.003884220262989402 2023-01-21 11:29:46.210355: step: 156/529, loss: 0.02139425277709961 2023-01-21 11:29:47.400773: step: 160/529, loss: 0.053163837641477585 2023-01-21 11:29:48.618351: step: 164/529, loss: 0.01021342258900404 2023-01-21 11:29:49.809457: step: 168/529, loss: 0.00022326708131004125 2023-01-21 11:29:50.987862: step: 172/529, loss: 0.01319131813943386 2023-01-21 11:29:52.188294: step: 176/529, loss: 0.006054425146430731 2023-01-21 11:29:53.340488: step: 180/529, loss: 0.0011057853698730469 2023-01-21 11:29:54.558097: step: 184/529, loss: 0.0019320010906085372 2023-01-21 11:29:55.753850: step: 188/529, loss: 0.025048447772860527 2023-01-21 11:29:56.897070: step: 192/529, loss: 0.020913315936923027 2023-01-21 11:29:58.083946: step: 196/529, loss: 0.0015975474379956722 2023-01-21 11:29:59.292011: step: 200/529, loss: 0.0023563385475426912 2023-01-21 11:30:00.468731: step: 204/529, loss: 0.011036348529160023 2023-01-21 11:30:01.641092: step: 208/529, loss: 0.017014121636748314 2023-01-21 11:30:02.836780: step: 212/529, loss: 0.037658121436834335 2023-01-21 11:30:04.128667: step: 216/529, loss: 0.012936020269989967 2023-01-21 11:30:05.301963: step: 220/529, loss: 0.08248496055603027 2023-01-21 11:30:06.478510: step: 224/529, loss: 0.02506725676357746 2023-01-21 11:30:07.706006: step: 228/529, loss: 0.014095116406679153 2023-01-21 11:30:08.851473: step: 232/529, loss: 0.001775503158569336 2023-01-21 11:30:10.038223: step: 236/529, loss: 0.03330402448773384 2023-01-21 11:30:11.253022: step: 240/529, loss: 0.01157617662101984 2023-01-21 11:30:12.442806: step: 244/529, loss: 0.009611034765839577 2023-01-21 11:30:13.667730: step: 248/529, loss: 0.002068805741146207 2023-01-21 11:30:14.836812: step: 252/529, loss: 0.026804160326719284 2023-01-21 11:30:16.007183: step: 256/529, loss: 0.000242447858909145 2023-01-21 11:30:17.198094: step: 260/529, loss: 0.013714027591049671 2023-01-21 11:30:18.407983: step: 264/529, loss: 0.016179945319890976 2023-01-21 11:30:19.600175: step: 268/529, loss: 0.004040074534714222 2023-01-21 11:30:20.760230: step: 272/529, loss: 0.034638214856386185 2023-01-21 11:30:21.986783: step: 276/529, loss: 0.06983423233032227 2023-01-21 11:30:23.178092: step: 280/529, loss: 0.04929351806640625 2023-01-21 11:30:24.394697: step: 284/529, loss: 0.001996803330257535 2023-01-21 11:30:25.589746: step: 288/529, loss: 0.0008342742803506553 2023-01-21 11:30:26.758233: step: 292/529, loss: 0.0026855471078306437 2023-01-21 11:30:27.990440: step: 296/529, loss: 0.09072045981884003 2023-01-21 11:30:29.143285: step: 300/529, loss: 0.016280174255371094 2023-01-21 11:30:30.320060: step: 304/529, loss: 0.02234954945743084 2023-01-21 11:30:31.536194: step: 308/529, loss: 0.04812812805175781 2023-01-21 11:30:32.790153: step: 312/529, loss: 0.004421234130859375 2023-01-21 11:30:34.010307: step: 316/529, loss: 0.005058002658188343 2023-01-21 11:30:35.180594: step: 320/529, loss: 0.003060245420783758 2023-01-21 11:30:36.387285: step: 324/529, loss: 0.0555814728140831 2023-01-21 11:30:37.557638: step: 328/529, loss: 0.05863399803638458 2023-01-21 11:30:38.738249: step: 332/529, loss: 0.2641967833042145 2023-01-21 11:30:39.903389: step: 336/529, loss: 0.005670833867043257 2023-01-21 11:30:41.077532: step: 340/529, loss: 0.02503070794045925 2023-01-21 11:30:42.281560: step: 344/529, loss: 0.019318629056215286 2023-01-21 11:30:43.489395: step: 348/529, loss: 0.2950594127178192 2023-01-21 11:30:44.688946: step: 352/529, loss: 0.047513388097286224 2023-01-21 11:30:45.865881: step: 356/529, loss: 0.0666167289018631 2023-01-21 11:30:47.074070: step: 360/529, loss: 0.003142166417092085 2023-01-21 11:30:48.281209: step: 364/529, loss: 0.010245705023407936 2023-01-21 11:30:49.478604: step: 368/529, loss: 0.015403127297759056 2023-01-21 11:30:50.693956: step: 372/529, loss: 0.04624128341674805 2023-01-21 11:30:51.830833: step: 376/529, loss: 0.015455389395356178 2023-01-21 11:30:53.016748: step: 380/529, loss: 0.13157230615615845 2023-01-21 11:30:54.210205: step: 384/529, loss: 0.013043022714555264 2023-01-21 11:30:55.378410: step: 388/529, loss: 0.3139774203300476 2023-01-21 11:30:56.552878: step: 392/529, loss: 0.010184002108871937 2023-01-21 11:30:57.742653: step: 396/529, loss: 0.06964874267578125 2023-01-21 11:30:58.943073: step: 400/529, loss: 0.021155167371034622 2023-01-21 11:31:00.147715: step: 404/529, loss: 0.014400673098862171 2023-01-21 11:31:01.360283: step: 408/529, loss: 0.008025742135941982 2023-01-21 11:31:02.580332: step: 412/529, loss: 0.030499696731567383 2023-01-21 11:31:03.780981: step: 416/529, loss: 0.03894786909222603 2023-01-21 11:31:04.957760: step: 420/529, loss: 0.00033605098724365234 2023-01-21 11:31:06.142646: step: 424/529, loss: 0.0067351339384913445 2023-01-21 11:31:07.354455: step: 428/529, loss: 0.013858509249985218 2023-01-21 11:31:08.576545: step: 432/529, loss: 5.1021575927734375e-05 2023-01-21 11:31:09.761346: step: 436/529, loss: 0.0002719878975767642 2023-01-21 11:31:10.956142: step: 440/529, loss: 0.0006118774181231856 2023-01-21 11:31:12.149774: step: 444/529, loss: 0.017720509320497513 2023-01-21 11:31:13.328591: step: 448/529, loss: 0.0334598533809185 2023-01-21 11:31:14.508102: step: 452/529, loss: 0.017281150445342064 2023-01-21 11:31:15.729856: step: 456/529, loss: 0.0006261348607949913 2023-01-21 11:31:16.976380: step: 460/529, loss: 0.04263028874993324 2023-01-21 11:31:18.181386: step: 464/529, loss: 0.06358852237462997 2023-01-21 11:31:19.386593: step: 468/529, loss: 0.004293775651603937 2023-01-21 11:31:20.607426: step: 472/529, loss: 0.011333084665238857 2023-01-21 11:31:21.842646: step: 476/529, loss: 0.0314447395503521 2023-01-21 11:31:23.045448: step: 480/529, loss: 0.009497452527284622 2023-01-21 11:31:24.222651: step: 484/529, loss: 0.004403853323310614 2023-01-21 11:31:25.433941: step: 488/529, loss: 0.024588823318481445 2023-01-21 11:31:26.621320: step: 492/529, loss: 0.000617599522229284 2023-01-21 11:31:27.827429: step: 496/529, loss: 0.028503989800810814 2023-01-21 11:31:29.028308: step: 500/529, loss: 0.025380326434969902 2023-01-21 11:31:30.304541: step: 504/529, loss: 0.04590282589197159 2023-01-21 11:31:31.511725: step: 508/529, loss: 0.005432748701423407 2023-01-21 11:31:32.669703: step: 512/529, loss: 0.007571840658783913 2023-01-21 11:31:33.859318: step: 516/529, loss: 0.002200651215389371 2023-01-21 11:31:35.112121: step: 520/529, loss: 0.10663548111915588 2023-01-21 11:31:36.330815: step: 524/529, loss: 0.00011873245239257812 2023-01-21 11:31:37.527471: step: 528/529, loss: 0.0043000816367566586 2023-01-21 11:31:38.717353: step: 532/529, loss: 0.013962030410766602 2023-01-21 11:31:39.925498: step: 536/529, loss: 0.03724203258752823 2023-01-21 11:31:41.101099: step: 540/529, loss: 0.01220398023724556 2023-01-21 11:31:42.317056: step: 544/529, loss: 0.00094013218767941 2023-01-21 11:31:43.502716: step: 548/529, loss: 0.11208763718605042 2023-01-21 11:31:44.728826: step: 552/529, loss: 0.00269584683701396 2023-01-21 11:31:45.924743: step: 556/529, loss: 0.008362388238310814 2023-01-21 11:31:47.056941: step: 560/529, loss: 7.03811674611643e-05 2023-01-21 11:31:48.213033: step: 564/529, loss: 0.02459421195089817 2023-01-21 11:31:49.421607: step: 568/529, loss: 0.0016084671951830387 2023-01-21 11:31:50.649317: step: 572/529, loss: 0.00010249615297652781 2023-01-21 11:31:51.849577: step: 576/529, loss: 0.003461170243099332 2023-01-21 11:31:53.037105: step: 580/529, loss: 0.0967632308602333 2023-01-21 11:31:54.229411: step: 584/529, loss: 0.004319381900131702 2023-01-21 11:31:55.452041: step: 588/529, loss: 0.17025050520896912 2023-01-21 11:31:56.625189: step: 592/529, loss: 0.02927989885210991 2023-01-21 11:31:57.824699: step: 596/529, loss: 0.010028552263975143 2023-01-21 11:31:59.016738: step: 600/529, loss: 0.0898953527212143 2023-01-21 11:32:00.200198: step: 604/529, loss: 0.0050258636474609375 2023-01-21 11:32:01.409742: step: 608/529, loss: 0.007094001863151789 2023-01-21 11:32:02.601751: step: 612/529, loss: 0.014985848218202591 2023-01-21 11:32:03.775196: step: 616/529, loss: 0.005885475780814886 2023-01-21 11:32:04.978279: step: 620/529, loss: 0.07012100517749786 2023-01-21 11:32:06.149127: step: 624/529, loss: 0.0042441366240382195 2023-01-21 11:32:07.341088: step: 628/529, loss: 0.007571220397949219 2023-01-21 11:32:08.499177: step: 632/529, loss: 0.007518816273659468 2023-01-21 11:32:09.692106: step: 636/529, loss: 0.01237402018159628 2023-01-21 11:32:10.896697: step: 640/529, loss: 0.004335212521255016 2023-01-21 11:32:12.135858: step: 644/529, loss: 0.012503623962402344 2023-01-21 11:32:13.311360: step: 648/529, loss: 1.3706327676773071 2023-01-21 11:32:14.484990: step: 652/529, loss: 0.07127085328102112 2023-01-21 11:32:15.657642: step: 656/529, loss: 0.004528427496552467 2023-01-21 11:32:16.851660: step: 660/529, loss: 0.0044953348115086555 2023-01-21 11:32:18.005604: step: 664/529, loss: 0.017867375165224075 2023-01-21 11:32:19.244471: step: 668/529, loss: 0.00947660207748413 2023-01-21 11:32:20.458257: step: 672/529, loss: 0.0017441869713366032 2023-01-21 11:32:21.683943: step: 676/529, loss: 0.00891189556568861 2023-01-21 11:32:22.893794: step: 680/529, loss: 0.05151014402508736 2023-01-21 11:32:24.116350: step: 684/529, loss: 0.06192026287317276 2023-01-21 11:32:25.319273: step: 688/529, loss: 0.015614127740263939 2023-01-21 11:32:26.488318: step: 692/529, loss: 0.00044994353083893657 2023-01-21 11:32:27.709615: step: 696/529, loss: 0.02568059042096138 2023-01-21 11:32:28.936540: step: 700/529, loss: 0.012043857015669346 2023-01-21 11:32:30.130659: step: 704/529, loss: 0.00026807785616256297 2023-01-21 11:32:31.309685: step: 708/529, loss: 0.0009771824115887284 2023-01-21 11:32:32.540486: step: 712/529, loss: 0.001566219376400113 2023-01-21 11:32:33.719986: step: 716/529, loss: 0.026256466284394264 2023-01-21 11:32:34.929220: step: 720/529, loss: 0.05051736906170845 2023-01-21 11:32:36.132346: step: 724/529, loss: 0.034284114837646484 2023-01-21 11:32:37.353329: step: 728/529, loss: 0.04318294674158096 2023-01-21 11:32:38.546005: step: 732/529, loss: 0.0006236076587811112 2023-01-21 11:32:39.729360: step: 736/529, loss: 0.0037344933953136206 2023-01-21 11:32:40.913642: step: 740/529, loss: 0.021323634311556816 2023-01-21 11:32:42.075974: step: 744/529, loss: 0.1035049706697464 2023-01-21 11:32:43.335216: step: 748/529, loss: 0.0002919197140727192 2023-01-21 11:32:44.522502: step: 752/529, loss: 0.0005622864118777215 2023-01-21 11:32:45.703430: step: 756/529, loss: 0.12760527431964874 2023-01-21 11:32:46.897579: step: 760/529, loss: 0.0017379760975018144 2023-01-21 11:32:48.130416: step: 764/529, loss: 0.04026851803064346 2023-01-21 11:32:49.369447: step: 768/529, loss: 0.04715137928724289 2023-01-21 11:32:50.548398: step: 772/529, loss: 0.042139194905757904 2023-01-21 11:32:51.781249: step: 776/529, loss: 0.046198081225156784 2023-01-21 11:32:52.957339: step: 780/529, loss: 0.015376472845673561 2023-01-21 11:32:54.199648: step: 784/529, loss: 0.49414387345314026 2023-01-21 11:32:55.365872: step: 788/529, loss: 0.04853539541363716 2023-01-21 11:32:56.527967: step: 792/529, loss: 0.001589059829711914 2023-01-21 11:32:57.701095: step: 796/529, loss: 0.014024067670106888 2023-01-21 11:32:58.879305: step: 800/529, loss: 0.02421126328408718 2023-01-21 11:33:00.054230: step: 804/529, loss: 0.02999694272875786 2023-01-21 11:33:01.313944: step: 808/529, loss: 0.01574125327169895 2023-01-21 11:33:02.509818: step: 812/529, loss: 0.000994777656160295 2023-01-21 11:33:03.803291: step: 816/529, loss: 0.006306457333266735 2023-01-21 11:33:05.028756: step: 820/529, loss: 0.00876693706959486 2023-01-21 11:33:06.210529: step: 824/529, loss: 0.04562978819012642 2023-01-21 11:33:07.419937: step: 828/529, loss: 0.01352615375071764 2023-01-21 11:33:08.578214: step: 832/529, loss: 0.011071777902543545 2023-01-21 11:33:09.752224: step: 836/529, loss: 0.0028502466157078743 2023-01-21 11:33:10.957232: step: 840/529, loss: 0.007470130920410156 2023-01-21 11:33:12.147058: step: 844/529, loss: 0.03695087134838104 2023-01-21 11:33:13.315326: step: 848/529, loss: 0.040552519261837006 2023-01-21 11:33:14.543200: step: 852/529, loss: 0.006839561276137829 2023-01-21 11:33:15.723057: step: 856/529, loss: 0.025505127385258675 2023-01-21 11:33:16.973561: step: 860/529, loss: 0.01203093584626913 2023-01-21 11:33:18.175939: step: 864/529, loss: 0.031674861907958984 2023-01-21 11:33:19.352048: step: 868/529, loss: 0.015520190820097923 2023-01-21 11:33:20.579257: step: 872/529, loss: 0.0310027115046978 2023-01-21 11:33:21.772199: step: 876/529, loss: 0.028268050402402878 2023-01-21 11:33:22.984345: step: 880/529, loss: 0.017487049102783203 2023-01-21 11:33:24.176208: step: 884/529, loss: 0.022662067785859108 2023-01-21 11:33:25.372716: step: 888/529, loss: 0.0008187294006347656 2023-01-21 11:33:26.571400: step: 892/529, loss: 0.006766903679817915 2023-01-21 11:33:27.746981: step: 896/529, loss: 0.06830596923828125 2023-01-21 11:33:28.961633: step: 900/529, loss: 0.05154714733362198 2023-01-21 11:33:30.151041: step: 904/529, loss: 0.020001031458377838 2023-01-21 11:33:31.378059: step: 908/529, loss: 0.009759045206010342 2023-01-21 11:33:32.635964: step: 912/529, loss: 0.02165517769753933 2023-01-21 11:33:33.856752: step: 916/529, loss: 0.11270676553249359 2023-01-21 11:33:35.049978: step: 920/529, loss: 0.024031830951571465 2023-01-21 11:33:36.283592: step: 924/529, loss: 0.018333053216338158 2023-01-21 11:33:37.470872: step: 928/529, loss: 0.02349414862692356 2023-01-21 11:33:38.737619: step: 932/529, loss: 0.0009970664978027344 2023-01-21 11:33:39.908876: step: 936/529, loss: 0.013111591339111328 2023-01-21 11:33:41.100337: step: 940/529, loss: 0.002722835633903742 2023-01-21 11:33:42.313162: step: 944/529, loss: 0.08323125541210175 2023-01-21 11:33:43.535701: step: 948/529, loss: 0.11373920738697052 2023-01-21 11:33:44.710872: step: 952/529, loss: 0.03552971035242081 2023-01-21 11:33:45.938543: step: 956/529, loss: 0.00485689053311944 2023-01-21 11:33:47.149858: step: 960/529, loss: 0.017791032791137695 2023-01-21 11:33:48.351397: step: 964/529, loss: 0.03160820156335831 2023-01-21 11:33:49.590675: step: 968/529, loss: 0.0551326759159565 2023-01-21 11:33:50.779676: step: 972/529, loss: 0.072378970682621 2023-01-21 11:33:51.957942: step: 976/529, loss: 0.009146403521299362 2023-01-21 11:33:53.115808: step: 980/529, loss: 0.014388322830200195 2023-01-21 11:33:54.304331: step: 984/529, loss: 0.0001085281401174143 2023-01-21 11:33:55.488531: step: 988/529, loss: 0.0009907246567308903 2023-01-21 11:33:56.698491: step: 992/529, loss: 0.022311020642518997 2023-01-21 11:33:57.874813: step: 996/529, loss: 0.04650535807013512 2023-01-21 11:33:59.060709: step: 1000/529, loss: 0.0392637737095356 2023-01-21 11:34:00.298241: step: 1004/529, loss: 0.13808879256248474 2023-01-21 11:34:01.461411: step: 1008/529, loss: 0.0325559638440609 2023-01-21 11:34:02.657615: step: 1012/529, loss: 0.003709411481395364 2023-01-21 11:34:03.894336: step: 1016/529, loss: 0.0242327693849802 2023-01-21 11:34:05.088171: step: 1020/529, loss: 0.009673071093857288 2023-01-21 11:34:06.287228: step: 1024/529, loss: 0.012622070498764515 2023-01-21 11:34:07.435866: step: 1028/529, loss: 0.01915273815393448 2023-01-21 11:34:08.683239: step: 1032/529, loss: 0.032759569585323334 2023-01-21 11:34:09.880717: step: 1036/529, loss: 0.05134544521570206 2023-01-21 11:34:11.073551: step: 1040/529, loss: 0.04553031921386719 2023-01-21 11:34:12.262771: step: 1044/529, loss: 0.02095937728881836 2023-01-21 11:34:13.445948: step: 1048/529, loss: 0.0012256145710125566 2023-01-21 11:34:14.639000: step: 1052/529, loss: 0.036905381828546524 2023-01-21 11:34:15.814274: step: 1056/529, loss: 0.00185565953142941 2023-01-21 11:34:16.998291: step: 1060/529, loss: 0.012190436944365501 2023-01-21 11:34:18.220262: step: 1064/529, loss: 0.11687707901000977 2023-01-21 11:34:19.408791: step: 1068/529, loss: 0.010545921511948109 2023-01-21 11:34:20.587456: step: 1072/529, loss: 0.017496777698397636 2023-01-21 11:34:21.734552: step: 1076/529, loss: 0.026461221277713776 2023-01-21 11:34:22.939372: step: 1080/529, loss: 0.0037640570662915707 2023-01-21 11:34:24.131476: step: 1084/529, loss: 0.07564334571361542 2023-01-21 11:34:25.313443: step: 1088/529, loss: 0.0267194751650095 2023-01-21 11:34:26.548185: step: 1092/529, loss: 0.013687754049897194 2023-01-21 11:34:27.717451: step: 1096/529, loss: 0.0001232624053955078 2023-01-21 11:34:28.892452: step: 1100/529, loss: 0.005167103372514248 2023-01-21 11:34:30.096390: step: 1104/529, loss: 0.030566884204745293 2023-01-21 11:34:31.261251: step: 1108/529, loss: 0.006701088044792414 2023-01-21 11:34:32.468254: step: 1112/529, loss: 0.03737516701221466 2023-01-21 11:34:33.661184: step: 1116/529, loss: 0.0008534193621017039 2023-01-21 11:34:34.864694: step: 1120/529, loss: 0.01563892513513565 2023-01-21 11:34:36.048193: step: 1124/529, loss: 0.04441662132740021 2023-01-21 11:34:37.220219: step: 1128/529, loss: 0.004844713024795055 2023-01-21 11:34:38.431073: step: 1132/529, loss: 0.07932376861572266 2023-01-21 11:34:39.649784: step: 1136/529, loss: 0.0013658523093909025 2023-01-21 11:34:40.860420: step: 1140/529, loss: 0.0016350746154785156 2023-01-21 11:34:42.061830: step: 1144/529, loss: 0.005328559782356024 2023-01-21 11:34:43.287731: step: 1148/529, loss: 0.06714649498462677 2023-01-21 11:34:44.464682: step: 1152/529, loss: 0.0002677917364053428 2023-01-21 11:34:45.637274: step: 1156/529, loss: 0.06665830314159393 2023-01-21 11:34:46.840586: step: 1160/529, loss: 0.02783813513815403 2023-01-21 11:34:48.032135: step: 1164/529, loss: 0.0013402937911450863 2023-01-21 11:34:49.215228: step: 1168/529, loss: 0.04208948835730553 2023-01-21 11:34:50.406480: step: 1172/529, loss: 1.7547608877066523e-05 2023-01-21 11:34:51.619261: step: 1176/529, loss: 0.12692709267139435 2023-01-21 11:34:52.807399: step: 1180/529, loss: 0.48442211747169495 2023-01-21 11:34:54.016679: step: 1184/529, loss: 0.0024298669304698706 2023-01-21 11:34:55.255386: step: 1188/529, loss: 0.003436184022575617 2023-01-21 11:34:56.466367: step: 1192/529, loss: 0.003067588899284601 2023-01-21 11:34:57.688666: step: 1196/529, loss: 0.00865182839334011 2023-01-21 11:34:58.914199: step: 1200/529, loss: 0.016780853271484375 2023-01-21 11:35:00.137553: step: 1204/529, loss: 0.0481531135737896 2023-01-21 11:35:01.320227: step: 1208/529, loss: 0.007971477694809437 2023-01-21 11:35:02.512312: step: 1212/529, loss: 0.004215812310576439 2023-01-21 11:35:03.686901: step: 1216/529, loss: 0.004649353213608265 2023-01-21 11:35:04.849531: step: 1220/529, loss: 0.004528045654296875 2023-01-21 11:35:06.050341: step: 1224/529, loss: 0.0052315713837742805 2023-01-21 11:35:07.305462: step: 1228/529, loss: 0.01217193715274334 2023-01-21 11:35:08.514383: step: 1232/529, loss: 0.004373741336166859 2023-01-21 11:35:09.675858: step: 1236/529, loss: 0.006182289216667414 2023-01-21 11:35:10.832426: step: 1240/529, loss: 0.00027313234750181437 2023-01-21 11:35:12.019529: step: 1244/529, loss: 0.0034875872079283 2023-01-21 11:35:13.197275: step: 1248/529, loss: 0.2646411955356598 2023-01-21 11:35:14.418573: step: 1252/529, loss: 0.002607727190479636 2023-01-21 11:35:15.646412: step: 1256/529, loss: 0.7531660199165344 2023-01-21 11:35:16.823296: step: 1260/529, loss: 0.03487901762127876 2023-01-21 11:35:18.024680: step: 1264/529, loss: 0.03135108947753906 2023-01-21 11:35:19.265222: step: 1268/529, loss: 0.004093504045158625 2023-01-21 11:35:20.489024: step: 1272/529, loss: 0.0968753769993782 2023-01-21 11:35:21.690875: step: 1276/529, loss: 0.0015981674659997225 2023-01-21 11:35:22.909176: step: 1280/529, loss: 0.00389785785228014 2023-01-21 11:35:24.196904: step: 1284/529, loss: 0.07193365693092346 2023-01-21 11:35:25.420426: step: 1288/529, loss: 0.022795582190155983 2023-01-21 11:35:26.616467: step: 1292/529, loss: 1.197209358215332 2023-01-21 11:35:27.810939: step: 1296/529, loss: 0.02820768393576145 2023-01-21 11:35:28.974804: step: 1300/529, loss: 0.017236996442079544 2023-01-21 11:35:30.135813: step: 1304/529, loss: 0.008360767737030983 2023-01-21 11:35:31.338182: step: 1308/529, loss: 0.033712100237607956 2023-01-21 11:35:32.576189: step: 1312/529, loss: 0.022310351952910423 2023-01-21 11:35:33.779370: step: 1316/529, loss: 0.002975940704345703 2023-01-21 11:35:34.972762: step: 1320/529, loss: 0.007573175244033337 2023-01-21 11:35:36.158034: step: 1324/529, loss: 0.020482826977968216 2023-01-21 11:35:37.344572: step: 1328/529, loss: 0.0724654346704483 2023-01-21 11:35:38.513325: step: 1332/529, loss: 0.0022208928130567074 2023-01-21 11:35:39.698278: step: 1336/529, loss: 0.008019829168915749 2023-01-21 11:35:40.916836: step: 1340/529, loss: 0.01813511922955513 2023-01-21 11:35:42.072761: step: 1344/529, loss: 0.004330349154770374 2023-01-21 11:35:43.254922: step: 1348/529, loss: 0.06065215915441513 2023-01-21 11:35:44.458952: step: 1352/529, loss: 0.004653835669159889 2023-01-21 11:35:45.642478: step: 1356/529, loss: 0.023238133639097214 2023-01-21 11:35:46.872802: step: 1360/529, loss: 0.0682673454284668 2023-01-21 11:35:48.065226: step: 1364/529, loss: 0.006237936206161976 2023-01-21 11:35:49.306326: step: 1368/529, loss: 0.03821459040045738 2023-01-21 11:35:50.502992: step: 1372/529, loss: 0.006769227795302868 2023-01-21 11:35:51.709628: step: 1376/529, loss: 0.05889396741986275 2023-01-21 11:35:52.898762: step: 1380/529, loss: 0.0033165933564305305 2023-01-21 11:35:54.102915: step: 1384/529, loss: 0.03815498203039169 2023-01-21 11:35:55.305865: step: 1388/529, loss: 0.006605911068618298 2023-01-21 11:35:56.473522: step: 1392/529, loss: 0.005694007966667414 2023-01-21 11:35:57.613597: step: 1396/529, loss: 0.00038361549377441406 2023-01-21 11:35:58.762723: step: 1400/529, loss: 0.0003912925603799522 2023-01-21 11:35:59.985624: step: 1404/529, loss: 0.015763377770781517 2023-01-21 11:36:01.144611: step: 1408/529, loss: 0.06012139469385147 2023-01-21 11:36:02.396989: step: 1412/529, loss: 0.004133415408432484 2023-01-21 11:36:03.602572: step: 1416/529, loss: 0.041353560984134674 2023-01-21 11:36:04.762997: step: 1420/529, loss: 0.017238235101103783 2023-01-21 11:36:05.950846: step: 1424/529, loss: 0.044570159167051315 2023-01-21 11:36:07.177427: step: 1428/529, loss: 0.0005863189580850303 2023-01-21 11:36:08.348390: step: 1432/529, loss: 0.01669340208172798 2023-01-21 11:36:09.547183: step: 1436/529, loss: 0.007371616084128618 2023-01-21 11:36:10.750661: step: 1440/529, loss: 0.028499890118837357 2023-01-21 11:36:11.948825: step: 1444/529, loss: 0.04242124781012535 2023-01-21 11:36:13.118837: step: 1448/529, loss: 0.0008799553033895791 2023-01-21 11:36:14.323349: step: 1452/529, loss: 0.023804377764463425 2023-01-21 11:36:15.517231: step: 1456/529, loss: 0.00970306433737278 2023-01-21 11:36:16.654438: step: 1460/529, loss: 0.06566095352172852 2023-01-21 11:36:17.819292: step: 1464/529, loss: 0.026435375213623047 2023-01-21 11:36:19.036277: step: 1468/529, loss: 0.006897640414535999 2023-01-21 11:36:20.216025: step: 1472/529, loss: 0.038277819752693176 2023-01-21 11:36:21.428507: step: 1476/529, loss: 0.033819008618593216 2023-01-21 11:36:22.653681: step: 1480/529, loss: 0.03408203274011612 2023-01-21 11:36:23.874331: step: 1484/529, loss: 0.041475772857666016 2023-01-21 11:36:25.041002: step: 1488/529, loss: 0.133219912648201 2023-01-21 11:36:26.279793: step: 1492/529, loss: 0.018799876794219017 2023-01-21 11:36:27.482847: step: 1496/529, loss: 0.044335030019283295 2023-01-21 11:36:28.647752: step: 1500/529, loss: 0.02014923095703125 2023-01-21 11:36:29.857592: step: 1504/529, loss: 0.08037119358778 2023-01-21 11:36:31.071160: step: 1508/529, loss: 1.0089977979660034 2023-01-21 11:36:32.275107: step: 1512/529, loss: 0.046825215220451355 2023-01-21 11:36:33.487252: step: 1516/529, loss: 0.020081140100955963 2023-01-21 11:36:34.700119: step: 1520/529, loss: 0.042453862726688385 2023-01-21 11:36:35.887459: step: 1524/529, loss: 0.024216270074248314 2023-01-21 11:36:37.063437: step: 1528/529, loss: 0.05180168151855469 2023-01-21 11:36:38.286921: step: 1532/529, loss: 0.041919708251953125 2023-01-21 11:36:39.528406: step: 1536/529, loss: 0.017532825469970703 2023-01-21 11:36:40.725862: step: 1540/529, loss: 0.018871881067752838 2023-01-21 11:36:41.924652: step: 1544/529, loss: 0.012684965506196022 2023-01-21 11:36:43.138967: step: 1548/529, loss: 0.06335287541151047 2023-01-21 11:36:44.357404: step: 1552/529, loss: 0.007520914543420076 2023-01-21 11:36:45.553151: step: 1556/529, loss: 0.0012094497215002775 2023-01-21 11:36:46.748309: step: 1560/529, loss: 0.005062485113739967 2023-01-21 11:36:47.965247: step: 1564/529, loss: 0.0018834115471690893 2023-01-21 11:36:49.152127: step: 1568/529, loss: 0.039183810353279114 2023-01-21 11:36:50.362038: step: 1572/529, loss: 0.006548118777573109 2023-01-21 11:36:51.563576: step: 1576/529, loss: 0.04161491245031357 2023-01-21 11:36:52.777054: step: 1580/529, loss: 0.06325483322143555 2023-01-21 11:36:53.992212: step: 1584/529, loss: 0.13219738006591797 2023-01-21 11:36:55.182767: step: 1588/529, loss: 0.04285898432135582 2023-01-21 11:36:56.369426: step: 1592/529, loss: 0.005718231201171875 2023-01-21 11:36:57.559396: step: 1596/529, loss: 0.0009244919056072831 2023-01-21 11:36:58.720031: step: 1600/529, loss: 0.0001794815034372732 2023-01-21 11:36:59.949491: step: 1604/529, loss: 0.01938762702047825 2023-01-21 11:37:01.124915: step: 1608/529, loss: 0.00012264252291060984 2023-01-21 11:37:02.382990: step: 1612/529, loss: 0.0076004983857274055 2023-01-21 11:37:03.565802: step: 1616/529, loss: 0.008213996887207031 2023-01-21 11:37:04.754588: step: 1620/529, loss: 0.03521881252527237 2023-01-21 11:37:05.941071: step: 1624/529, loss: 0.05723857879638672 2023-01-21 11:37:07.113014: step: 1628/529, loss: 0.0016631126636639237 2023-01-21 11:37:08.289914: step: 1632/529, loss: 0.002435874892398715 2023-01-21 11:37:09.486426: step: 1636/529, loss: 0.0012926101917400956 2023-01-21 11:37:10.706481: step: 1640/529, loss: 0.20914383232593536 2023-01-21 11:37:11.944877: step: 1644/529, loss: 0.00137414934579283 2023-01-21 11:37:13.177876: step: 1648/529, loss: 0.11086144298315048 2023-01-21 11:37:14.433400: step: 1652/529, loss: 0.021106624975800514 2023-01-21 11:37:15.635133: step: 1656/529, loss: 0.07170820981264114 2023-01-21 11:37:16.815885: step: 1660/529, loss: 0.027965163812041283 2023-01-21 11:37:17.993956: step: 1664/529, loss: 0.011815452948212624 2023-01-21 11:37:19.165340: step: 1668/529, loss: 0.0005886077997274697 2023-01-21 11:37:20.375324: step: 1672/529, loss: 0.028856562450528145 2023-01-21 11:37:21.586997: step: 1676/529, loss: 0.036635592579841614 2023-01-21 11:37:22.772074: step: 1680/529, loss: 0.011177443899214268 2023-01-21 11:37:23.955131: step: 1684/529, loss: 0.015666676685214043 2023-01-21 11:37:25.153418: step: 1688/529, loss: 0.011370849795639515 2023-01-21 11:37:26.309812: step: 1692/529, loss: 0.029594993218779564 2023-01-21 11:37:27.494873: step: 1696/529, loss: 6.642795562744141 2023-01-21 11:37:28.728207: step: 1700/529, loss: 1.9073486328125e-06 2023-01-21 11:37:29.902996: step: 1704/529, loss: 0.0684257447719574 2023-01-21 11:37:31.112226: step: 1708/529, loss: 0.006921291351318359 2023-01-21 11:37:32.321988: step: 1712/529, loss: 0.0013270378112792969 2023-01-21 11:37:33.546000: step: 1716/529, loss: 0.00669174175709486 2023-01-21 11:37:34.740005: step: 1720/529, loss: 0.021431589499115944 2023-01-21 11:37:36.012677: step: 1724/529, loss: 0.056902315467596054 2023-01-21 11:37:37.231414: step: 1728/529, loss: 0.015398931689560413 2023-01-21 11:37:38.407172: step: 1732/529, loss: 0.05008867010474205 2023-01-21 11:37:39.576075: step: 1736/529, loss: 0.0014361381763592362 2023-01-21 11:37:40.744385: step: 1740/529, loss: 0.002488803816959262 2023-01-21 11:37:41.935961: step: 1744/529, loss: 0.0064025879837572575 2023-01-21 11:37:43.156919: step: 1748/529, loss: 0.002174949739128351 2023-01-21 11:37:44.362053: step: 1752/529, loss: 0.054010674357414246 2023-01-21 11:37:45.569671: step: 1756/529, loss: 0.01879558525979519 2023-01-21 11:37:46.745846: step: 1760/529, loss: 0.040892407298088074 2023-01-21 11:37:47.930558: step: 1764/529, loss: 0.015877151861786842 2023-01-21 11:37:49.141315: step: 1768/529, loss: 0.0018224716186523438 2023-01-21 11:37:50.362935: step: 1772/529, loss: 0.024832725524902344 2023-01-21 11:37:51.579587: step: 1776/529, loss: 0.010940122418105602 2023-01-21 11:37:52.746860: step: 1780/529, loss: 0.007358360569924116 2023-01-21 11:37:53.969390: step: 1784/529, loss: 0.03130703046917915 2023-01-21 11:37:55.174288: step: 1788/529, loss: 0.004744672682136297 2023-01-21 11:37:56.437336: step: 1792/529, loss: 0.0041048284620046616 2023-01-21 11:37:57.629038: step: 1796/529, loss: 0.011958122253417969 2023-01-21 11:37:58.861920: step: 1800/529, loss: 0.00035005214158445597 2023-01-21 11:38:00.083873: step: 1804/529, loss: 0.06836128234863281 2023-01-21 11:38:01.265843: step: 1808/529, loss: 0.01869945414364338 2023-01-21 11:38:02.440485: step: 1812/529, loss: 0.0030375481583178043 2023-01-21 11:38:03.654144: step: 1816/529, loss: 0.029547788202762604 2023-01-21 11:38:04.877766: step: 1820/529, loss: 0.014499188400804996 2023-01-21 11:38:06.077209: step: 1824/529, loss: 0.08833573758602142 2023-01-21 11:38:07.252139: step: 1828/529, loss: 0.23115572333335876 2023-01-21 11:38:08.449381: step: 1832/529, loss: 0.002206707140430808 2023-01-21 11:38:09.665817: step: 1836/529, loss: 0.015276908874511719 2023-01-21 11:38:10.911663: step: 1840/529, loss: 0.03730583190917969 2023-01-21 11:38:12.144457: step: 1844/529, loss: 0.0075813294388353825 2023-01-21 11:38:13.295300: step: 1848/529, loss: 0.028373051434755325 2023-01-21 11:38:14.474405: step: 1852/529, loss: 0.8360044360160828 2023-01-21 11:38:15.635060: step: 1856/529, loss: 0.03366518020629883 2023-01-21 11:38:16.820953: step: 1860/529, loss: 0.24306488037109375 2023-01-21 11:38:18.053398: step: 1864/529, loss: 0.005439949221909046 2023-01-21 11:38:19.295429: step: 1868/529, loss: 0.043221186846494675 2023-01-21 11:38:20.504048: step: 1872/529, loss: 0.005711936857551336 2023-01-21 11:38:21.702744: step: 1876/529, loss: 0.003545427229255438 2023-01-21 11:38:22.884320: step: 1880/529, loss: 0.045832060277462006 2023-01-21 11:38:24.099837: step: 1884/529, loss: 0.08937826752662659 2023-01-21 11:38:25.270737: step: 1888/529, loss: 0.01433253288269043 2023-01-21 11:38:26.467016: step: 1892/529, loss: 0.0729837417602539 2023-01-21 11:38:27.692492: step: 1896/529, loss: 0.00558319129049778 2023-01-21 11:38:28.919281: step: 1900/529, loss: 0.005814933683723211 2023-01-21 11:38:30.121763: step: 1904/529, loss: 0.004852867219597101 2023-01-21 11:38:31.286558: step: 1908/529, loss: 0.002795314881950617 2023-01-21 11:38:32.495470: step: 1912/529, loss: 0.18903236091136932 2023-01-21 11:38:33.687455: step: 1916/529, loss: 0.0008465528371743858 2023-01-21 11:38:34.850805: step: 1920/529, loss: 0.018373871222138405 2023-01-21 11:38:36.046666: step: 1924/529, loss: 0.00037751198397018015 2023-01-21 11:38:37.251594: step: 1928/529, loss: 0.04945354908704758 2023-01-21 11:38:38.537700: step: 1932/529, loss: 0.0007461548084393144 2023-01-21 11:38:39.738464: step: 1936/529, loss: 0.06285524368286133 2023-01-21 11:38:40.914278: step: 1940/529, loss: 0.023958398029208183 2023-01-21 11:38:42.135894: step: 1944/529, loss: 0.023633766919374466 2023-01-21 11:38:43.336870: step: 1948/529, loss: 0.00016717911057639867 2023-01-21 11:38:44.555287: step: 1952/529, loss: 0.005665207281708717 2023-01-21 11:38:45.715878: step: 1956/529, loss: 0.01394033432006836 2023-01-21 11:38:46.951183: step: 1960/529, loss: 0.002126026200130582 2023-01-21 11:38:48.187956: step: 1964/529, loss: 0.008430815301835537 2023-01-21 11:38:49.442958: step: 1968/529, loss: 0.014192390255630016 2023-01-21 11:38:50.616018: step: 1972/529, loss: 0.0002094268857035786 2023-01-21 11:38:51.848214: step: 1976/529, loss: 0.06872482597827911 2023-01-21 11:38:53.083577: step: 1980/529, loss: 0.009262656792998314 2023-01-21 11:38:54.292002: step: 1984/529, loss: 0.034957073628902435 2023-01-21 11:38:55.470783: step: 1988/529, loss: 0.017655039206147194 2023-01-21 11:38:56.657015: step: 1992/529, loss: 0.015191060490906239 2023-01-21 11:38:57.857903: step: 1996/529, loss: 0.022785378620028496 2023-01-21 11:38:59.088225: step: 2000/529, loss: 0.037732984870672226 2023-01-21 11:39:00.306717: step: 2004/529, loss: 0.017534922808408737 2023-01-21 11:39:01.521667: step: 2008/529, loss: 0.0014892579056322575 2023-01-21 11:39:02.773724: step: 2012/529, loss: 0.09685364365577698 2023-01-21 11:39:03.975738: step: 2016/529, loss: 0.005210495088249445 2023-01-21 11:39:05.194323: step: 2020/529, loss: 0.004749489016830921 2023-01-21 11:39:06.433655: step: 2024/529, loss: 0.012339066714048386 2023-01-21 11:39:07.636951: step: 2028/529, loss: 6.151199340820312e-05 2023-01-21 11:39:08.803181: step: 2032/529, loss: 0.03139953687787056 2023-01-21 11:39:09.994504: step: 2036/529, loss: 0.023186493664979935 2023-01-21 11:39:11.167282: step: 2040/529, loss: 0.022021150216460228 2023-01-21 11:39:12.378598: step: 2044/529, loss: 0.001293611596338451 2023-01-21 11:39:13.577669: step: 2048/529, loss: 0.030144404619932175 2023-01-21 11:39:14.750857: step: 2052/529, loss: 0.0006465911865234375 2023-01-21 11:39:15.936659: step: 2056/529, loss: 0.005116557702422142 2023-01-21 11:39:17.135948: step: 2060/529, loss: 0.018648911267518997 2023-01-21 11:39:18.322693: step: 2064/529, loss: 0.03609009087085724 2023-01-21 11:39:19.504931: step: 2068/529, loss: 0.002794933272525668 2023-01-21 11:39:20.659942: step: 2072/529, loss: 0.003845310304313898 2023-01-21 11:39:21.831040: step: 2076/529, loss: 0.10613413155078888 2023-01-21 11:39:23.050704: step: 2080/529, loss: 0.006571575999259949 2023-01-21 11:39:24.227366: step: 2084/529, loss: 0.0341605544090271 2023-01-21 11:39:25.434186: step: 2088/529, loss: 0.11511025577783585 2023-01-21 11:39:26.655375: step: 2092/529, loss: 0.002692603971809149 2023-01-21 11:39:27.813555: step: 2096/529, loss: 0.0010046481620520353 2023-01-21 11:39:28.965819: step: 2100/529, loss: 0.36170119047164917 2023-01-21 11:39:30.158778: step: 2104/529, loss: 0.35221290588378906 2023-01-21 11:39:31.373480: step: 2108/529, loss: 0.022685719653964043 2023-01-21 11:39:32.579256: step: 2112/529, loss: 0.02852764166891575 2023-01-21 11:39:33.742222: step: 2116/529, loss: 0.008012199774384499 ================================================== Loss: 0.053 -------------------- Dev: {'event': {'p': 0.5977249224405378, 'r': 0.7696404793608522, 'f1': 0.6728754365541327}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Test: {'event': {'p': 0.6421001926782274, 'r': 0.7986818454164171, 'f1': 0.7118825100133512}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Chinese: {'event': {'p': 0.5945945945945946, 'r': 0.8148148148148148, 'f1': 0.6875}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Korean: {'event': {'p': 0.6326530612244898, 'r': 0.49206349206349204, 'f1': 0.5535714285714285}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977249224405378, 'r': 0.7696404793608522, 'f1': 0.6728754365541327}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Russian: {'event': {'p': 0.6421001926782274, 'r': 0.7986818454164171, 'f1': 0.7118825100133512}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} ****************************** 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 11:40:26.339557: step: 4/529, loss: 0.01121821440756321 2023-01-21 11:40:27.506836: step: 8/529, loss: 0.009226560592651367 2023-01-21 11:40:28.685522: step: 12/529, loss: 0.0846099853515625 2023-01-21 11:40:29.907498: step: 16/529, loss: 0.012137413024902344 2023-01-21 11:40:31.092368: step: 20/529, loss: 0.012349986471235752 2023-01-21 11:40:32.258301: step: 24/529, loss: 0.0023848055861890316 2023-01-21 11:40:33.435063: step: 28/529, loss: 0.008556938730180264 2023-01-21 11:40:34.633078: step: 32/529, loss: 0.0001464843808207661 2023-01-21 11:40:35.839693: step: 36/529, loss: 0.027249909937381744 2023-01-21 11:40:37.036879: step: 40/529, loss: 0.0024625780060887337 2023-01-21 11:40:38.196676: step: 44/529, loss: 0.02579193189740181 2023-01-21 11:40:39.420225: step: 48/529, loss: 0.002651691436767578 2023-01-21 11:40:40.674694: step: 52/529, loss: 0.40329664945602417 2023-01-21 11:40:41.862810: step: 56/529, loss: 0.0604059211909771 2023-01-21 11:40:43.009172: step: 60/529, loss: 0.0022771835792809725 2023-01-21 11:40:44.201057: step: 64/529, loss: 0.004975414369255304 2023-01-21 11:40:45.429776: step: 68/529, loss: 0.0007012367132119834 2023-01-21 11:40:46.646055: step: 72/529, loss: 0.0647285208106041 2023-01-21 11:40:47.826138: step: 76/529, loss: 0.004416084382683039 2023-01-21 11:40:49.023035: step: 80/529, loss: 0.32259073853492737 2023-01-21 11:40:50.265495: step: 84/529, loss: 0.006053447723388672 2023-01-21 11:40:51.473176: step: 88/529, loss: 0.012158012948930264 2023-01-21 11:40:52.731475: step: 92/529, loss: 0.05882968753576279 2023-01-21 11:40:53.874166: step: 96/529, loss: 0.005947113502770662 2023-01-21 11:40:55.064082: step: 100/529, loss: 0.0033861161209642887 2023-01-21 11:40:56.294651: step: 104/529, loss: 0.0119476318359375 2023-01-21 11:40:57.470863: step: 108/529, loss: 0.05767965316772461 2023-01-21 11:40:58.661442: step: 112/529, loss: 0.021436691284179688 2023-01-21 11:40:59.868475: step: 116/529, loss: 0.02227325364947319 2023-01-21 11:41:01.050525: step: 120/529, loss: 0.0013520241482183337 2023-01-21 11:41:02.219534: step: 124/529, loss: 0.010037422180175781 2023-01-21 11:41:03.440467: step: 128/529, loss: 0.0115540511906147 2023-01-21 11:41:04.615745: step: 132/529, loss: 0.016535472124814987 2023-01-21 11:41:05.822531: step: 136/529, loss: 0.07580481469631195 2023-01-21 11:41:07.016752: step: 140/529, loss: 0.00099773402325809 2023-01-21 11:41:08.167475: step: 144/529, loss: 0.022234437987208366 2023-01-21 11:41:09.357533: step: 148/529, loss: 0.03079976886510849 2023-01-21 11:41:10.632442: step: 152/529, loss: 0.04285158962011337 2023-01-21 11:41:11.819463: step: 156/529, loss: 0.0010192871559411287 2023-01-21 11:41:13.018829: step: 160/529, loss: 0.012722802348434925 2023-01-21 11:41:14.191853: step: 164/529, loss: 0.00026292799157090485 2023-01-21 11:41:15.399637: step: 168/529, loss: 0.08401966094970703 2023-01-21 11:41:16.593218: step: 172/529, loss: 0.026950262486934662 2023-01-21 11:41:17.764963: step: 176/529, loss: 0.0032828333787620068 2023-01-21 11:41:18.966359: step: 180/529, loss: 0.0003248214488849044 2023-01-21 11:41:20.127948: step: 184/529, loss: 0.01965932734310627 2023-01-21 11:41:21.297684: step: 188/529, loss: 0.003544426057487726 2023-01-21 11:41:22.483399: step: 192/529, loss: 0.007383918855339289 2023-01-21 11:41:23.707560: step: 196/529, loss: 0.0006082534673623741 2023-01-21 11:41:24.873251: step: 200/529, loss: 0.0032835006713867188 2023-01-21 11:41:26.076626: step: 204/529, loss: 0.0021963119506835938 2023-01-21 11:41:27.276354: step: 208/529, loss: 0.0751006156206131 2023-01-21 11:41:28.483635: step: 212/529, loss: 0.006998253054916859 2023-01-21 11:41:29.691021: step: 216/529, loss: 0.11058690398931503 2023-01-21 11:41:30.866259: step: 220/529, loss: 0.011524009518325329 2023-01-21 11:41:32.106935: step: 224/529, loss: 0.12669506669044495 2023-01-21 11:41:33.299576: step: 228/529, loss: 0.00801916141062975 2023-01-21 11:41:34.504293: step: 232/529, loss: 0.003075027372688055 2023-01-21 11:41:35.711533: step: 236/529, loss: 0.0575687438249588 2023-01-21 11:41:36.887921: step: 240/529, loss: 0.00046362876310013235 2023-01-21 11:41:38.126406: step: 244/529, loss: 0.002476978348568082 2023-01-21 11:41:39.317328: step: 248/529, loss: 0.01172714214771986 2023-01-21 11:41:40.521933: step: 252/529, loss: 0.020298385992646217 2023-01-21 11:41:41.737386: step: 256/529, loss: 0.013495922088623047 2023-01-21 11:41:42.936554: step: 260/529, loss: 0.01186294574290514 2023-01-21 11:41:44.105399: step: 264/529, loss: 0.006715011782944202 2023-01-21 11:41:45.275110: step: 268/529, loss: 0.006658363621681929 2023-01-21 11:41:46.490272: step: 272/529, loss: 0.03999576345086098 2023-01-21 11:41:47.699108: step: 276/529, loss: 0.006159305572509766 2023-01-21 11:41:48.889981: step: 280/529, loss: 0.0008301734924316406 2023-01-21 11:41:50.121816: step: 284/529, loss: 0.005973624996840954 2023-01-21 11:41:51.314952: step: 288/529, loss: 0.019219208508729935 2023-01-21 11:41:52.478463: step: 292/529, loss: 0.0027778628282248974 2023-01-21 11:41:53.702138: step: 296/529, loss: 0.006109809968620539 2023-01-21 11:41:54.884108: step: 300/529, loss: 0.017147399485111237 2023-01-21 11:41:56.119274: step: 304/529, loss: 0.0014553070068359375 2023-01-21 11:41:57.290121: step: 308/529, loss: 0.00022627711587119848 2023-01-21 11:41:58.553429: step: 312/529, loss: 0.00605201767757535 2023-01-21 11:41:59.728851: step: 316/529, loss: 0.0034310820046812296 2023-01-21 11:42:00.942872: step: 320/529, loss: 0.022520065307617188 2023-01-21 11:42:02.123032: step: 324/529, loss: 0.6154108047485352 2023-01-21 11:42:03.311933: step: 328/529, loss: 7.524489774368703e-05 2023-01-21 11:42:04.566213: step: 332/529, loss: 7.32898770365864e-05 2023-01-21 11:42:05.756328: step: 336/529, loss: 0.004454994574189186 2023-01-21 11:42:06.942684: step: 340/529, loss: 0.0006483077886514366 2023-01-21 11:42:08.085598: step: 344/529, loss: 0.01694812811911106 2023-01-21 11:42:09.284021: step: 348/529, loss: 0.038036346435546875 2023-01-21 11:42:10.489999: step: 352/529, loss: 0.048842813819646835 2023-01-21 11:42:11.723444: step: 356/529, loss: 0.03309059143066406 2023-01-21 11:42:12.938886: step: 360/529, loss: 0.0056591033935546875 2023-01-21 11:42:14.113239: step: 364/529, loss: 0.01780109480023384 2023-01-21 11:42:15.315608: step: 368/529, loss: 0.015294170007109642 2023-01-21 11:42:16.541830: step: 372/529, loss: 0.016104508191347122 2023-01-21 11:42:17.748605: step: 376/529, loss: 0.014001751318573952 2023-01-21 11:42:18.937321: step: 380/529, loss: 0.02171015739440918 2023-01-21 11:42:20.116998: step: 384/529, loss: 0.01760873757302761 2023-01-21 11:42:21.295724: step: 388/529, loss: 0.0004352569521870464 2023-01-21 11:42:22.538963: step: 392/529, loss: 0.031203269958496094 2023-01-21 11:42:23.742328: step: 396/529, loss: 0.02887716330587864 2023-01-21 11:42:24.938778: step: 400/529, loss: 0.030919646844267845 2023-01-21 11:42:26.144288: step: 404/529, loss: 0.020296763628721237 2023-01-21 11:42:27.316616: step: 408/529, loss: 0.008166313171386719 2023-01-21 11:42:28.486334: step: 412/529, loss: 0.019211482256650925 2023-01-21 11:42:29.665061: step: 416/529, loss: 0.04082345962524414 2023-01-21 11:42:30.868970: step: 420/529, loss: 0.06339378654956818 2023-01-21 11:42:32.103218: step: 424/529, loss: 3.919601658708416e-05 2023-01-21 11:42:33.302854: step: 428/529, loss: 0.05355234444141388 2023-01-21 11:42:34.499063: step: 432/529, loss: 0.012090301141142845 2023-01-21 11:42:35.683912: step: 436/529, loss: 0.07747115939855576 2023-01-21 11:42:36.836846: step: 440/529, loss: 0.009994077496230602 2023-01-21 11:42:38.014851: step: 444/529, loss: 0.0014617920387536287 2023-01-21 11:42:39.210893: step: 448/529, loss: 5.14984130859375e-05 2023-01-21 11:42:40.388018: step: 452/529, loss: 0.0010608673328533769 2023-01-21 11:42:41.546848: step: 456/529, loss: 0.006717490963637829 2023-01-21 11:42:42.702164: step: 460/529, loss: 0.0038290023803710938 2023-01-21 11:42:43.909095: step: 464/529, loss: 0.0036141397431492805 2023-01-21 11:42:45.095766: step: 468/529, loss: 0.004974365234375 2023-01-21 11:42:46.297952: step: 472/529, loss: 0.03995056450366974 2023-01-21 11:42:47.419495: step: 476/529, loss: 0.034803010523319244 2023-01-21 11:42:48.629880: step: 480/529, loss: 0.04414510726928711 2023-01-21 11:42:49.831541: step: 484/529, loss: 0.0013128280406817794 2023-01-21 11:42:51.011001: step: 488/529, loss: 0.0024770735763013363 2023-01-21 11:42:52.192710: step: 492/529, loss: 0.0034605024848133326 2023-01-21 11:42:53.365703: step: 496/529, loss: 0.008593273349106312 2023-01-21 11:42:54.550199: step: 500/529, loss: 0.00027976036653853953 2023-01-21 11:42:55.750514: step: 504/529, loss: 0.08622293919324875 2023-01-21 11:42:56.944535: step: 508/529, loss: 0.008352423086762428 2023-01-21 11:42:58.164492: step: 512/529, loss: 0.015434456057846546 2023-01-21 11:42:59.361012: step: 516/529, loss: 0.030466269701719284 2023-01-21 11:43:00.533024: step: 520/529, loss: 0.00214729318395257 2023-01-21 11:43:01.756651: step: 524/529, loss: 0.0069138528779149055 2023-01-21 11:43:02.956481: step: 528/529, loss: 0.0016923904186114669 2023-01-21 11:43:04.180422: step: 532/529, loss: 0.0012170791160315275 2023-01-21 11:43:05.389353: step: 536/529, loss: 0.007384300697594881 2023-01-21 11:43:06.572504: step: 540/529, loss: 0.02920837327837944 2023-01-21 11:43:07.773558: step: 544/529, loss: 0.00040912628173828125 2023-01-21 11:43:08.959879: step: 548/529, loss: 0.0011096715461462736 2023-01-21 11:43:10.156878: step: 552/529, loss: 0.038210179656744 2023-01-21 11:43:11.351375: step: 556/529, loss: 0.0002511024649720639 2023-01-21 11:43:12.529401: step: 560/529, loss: 0.004841613583266735 2023-01-21 11:43:13.691475: step: 564/529, loss: 0.008134794421494007 2023-01-21 11:43:14.892014: step: 568/529, loss: 0.01677103154361248 2023-01-21 11:43:16.052249: step: 572/529, loss: 0.03830432891845703 2023-01-21 11:43:17.236982: step: 576/529, loss: 0.0004141807439737022 2023-01-21 11:43:18.496683: step: 580/529, loss: 0.006304312031716108 2023-01-21 11:43:19.701865: step: 584/529, loss: 0.07688693702220917 2023-01-21 11:43:20.956401: step: 588/529, loss: 0.07493872940540314 2023-01-21 11:43:22.144543: step: 592/529, loss: 0.021212458610534668 2023-01-21 11:43:23.352153: step: 596/529, loss: 0.040657807141542435 2023-01-21 11:43:24.555589: step: 600/529, loss: 0.029030513018369675 2023-01-21 11:43:25.769799: step: 604/529, loss: 0.015389442443847656 2023-01-21 11:43:26.981949: step: 608/529, loss: 0.0003103256458416581 2023-01-21 11:43:28.213992: step: 612/529, loss: 0.004745113663375378 2023-01-21 11:43:29.361242: step: 616/529, loss: 0.012839317321777344 2023-01-21 11:43:30.537066: step: 620/529, loss: 0.026345444843173027 2023-01-21 11:43:31.722956: step: 624/529, loss: 0.030898286029696465 2023-01-21 11:43:32.952594: step: 628/529, loss: 0.0033539296127855778 2023-01-21 11:43:34.140782: step: 632/529, loss: 0.003013992216438055 2023-01-21 11:43:35.341594: step: 636/529, loss: 0.0014978409744799137 2023-01-21 11:43:36.558608: step: 640/529, loss: 0.04772148281335831 2023-01-21 11:43:37.722029: step: 644/529, loss: 0.13713550567626953 2023-01-21 11:43:38.939143: step: 648/529, loss: 0.018921375274658203 2023-01-21 11:43:40.136896: step: 652/529, loss: 4.57763671875e-05 2023-01-21 11:43:41.337399: step: 656/529, loss: 0.0009199142223224044 2023-01-21 11:43:42.530750: step: 660/529, loss: 0.003127574920654297 2023-01-21 11:43:43.729729: step: 664/529, loss: 0.00039501191349700093 2023-01-21 11:43:44.917536: step: 668/529, loss: 0.0005623817560262978 2023-01-21 11:43:46.080824: step: 672/529, loss: 0.12925167381763458 2023-01-21 11:43:47.241398: step: 676/529, loss: 0.0006808757316321135 2023-01-21 11:43:48.370824: step: 680/529, loss: 0.012927723117172718 2023-01-21 11:43:49.561325: step: 684/529, loss: 0.0007337093120440841 2023-01-21 11:43:50.745680: step: 688/529, loss: 0.006521463394165039 2023-01-21 11:43:51.953776: step: 692/529, loss: 0.0020774840377271175 2023-01-21 11:43:53.114438: step: 696/529, loss: 0.000921058701351285 2023-01-21 11:43:54.299880: step: 700/529, loss: 0.001961231231689453 2023-01-21 11:43:55.481007: step: 704/529, loss: 0.016443252563476562 2023-01-21 11:43:56.708720: step: 708/529, loss: 0.014692497439682484 2023-01-21 11:43:57.936232: step: 712/529, loss: 0.0019512177677825093 2023-01-21 11:43:59.164736: step: 716/529, loss: 0.0249189380556345 2023-01-21 11:44:00.361217: step: 720/529, loss: 0.0023110867477953434 2023-01-21 11:44:01.633318: step: 724/529, loss: 0.04198732599616051 2023-01-21 11:44:02.802016: step: 728/529, loss: 0.02201404608786106 2023-01-21 11:44:04.027705: step: 732/529, loss: 0.009539724327623844 2023-01-21 11:44:05.205815: step: 736/529, loss: 0.007643270771950483 2023-01-21 11:44:06.394780: step: 740/529, loss: 0.0024990083184093237 2023-01-21 11:44:07.583375: step: 744/529, loss: 0.34504929184913635 2023-01-21 11:44:08.789899: step: 748/529, loss: 0.03697576746344566 2023-01-21 11:44:09.966430: step: 752/529, loss: 0.0015136718284338713 2023-01-21 11:44:11.170974: step: 756/529, loss: 0.016345692798495293 2023-01-21 11:44:12.324314: step: 760/529, loss: 0.004855346865952015 2023-01-21 11:44:13.542697: step: 764/529, loss: 0.01900806464254856 2023-01-21 11:44:14.701974: step: 768/529, loss: 0.02727355994284153 2023-01-21 11:44:15.915251: step: 772/529, loss: 0.0007274628151208162 2023-01-21 11:44:17.087622: step: 776/529, loss: 0.023694420233368874 2023-01-21 11:44:18.253621: step: 780/529, loss: 0.00031871796818450093 2023-01-21 11:44:19.434683: step: 784/529, loss: 0.04489602893590927 2023-01-21 11:44:20.647397: step: 788/529, loss: 0.057988740503787994 2023-01-21 11:44:21.845927: step: 792/529, loss: 0.004422807600349188 2023-01-21 11:44:23.017529: step: 796/529, loss: 0.025780821219086647 2023-01-21 11:44:24.188254: step: 800/529, loss: 0.02680225297808647 2023-01-21 11:44:25.393903: step: 804/529, loss: 0.005648136604577303 2023-01-21 11:44:26.546153: step: 808/529, loss: 0.011340761557221413 2023-01-21 11:44:27.737933: step: 812/529, loss: 6.45637555862777e-05 2023-01-21 11:44:28.939106: step: 816/529, loss: 0.0002510070626158267 2023-01-21 11:44:30.157646: step: 820/529, loss: 0.03329353407025337 2023-01-21 11:44:31.352109: step: 824/529, loss: 0.0010089874267578125 2023-01-21 11:44:32.512055: step: 828/529, loss: 0.0292326919734478 2023-01-21 11:44:33.758944: step: 832/529, loss: 0.0004772186221089214 2023-01-21 11:44:34.976381: step: 836/529, loss: 0.023160552605986595 2023-01-21 11:44:36.122207: step: 840/529, loss: 0.0014319419860839844 2023-01-21 11:44:37.289022: step: 844/529, loss: 0.004728889558464289 2023-01-21 11:44:38.466862: step: 848/529, loss: 0.0009454727405682206 2023-01-21 11:44:39.684045: step: 852/529, loss: 0.028275156393647194 2023-01-21 11:44:40.852326: step: 856/529, loss: 0.00031557082547806203 2023-01-21 11:44:42.040779: step: 860/529, loss: 0.0010160446399822831 2023-01-21 11:44:43.178543: step: 864/529, loss: 0.009815978817641735 2023-01-21 11:44:44.380269: step: 868/529, loss: 0.008692741394042969 2023-01-21 11:44:45.590203: step: 872/529, loss: 0.12085652351379395 2023-01-21 11:44:46.790229: step: 876/529, loss: 0.0038149834144860506 2023-01-21 11:44:47.970684: step: 880/529, loss: 0.26933154463768005 2023-01-21 11:44:49.140258: step: 884/529, loss: 0.0072152139618992805 2023-01-21 11:44:50.330498: step: 888/529, loss: 0.021677399054169655 2023-01-21 11:44:51.509532: step: 892/529, loss: 0.007256889715790749 2023-01-21 11:44:52.714832: step: 896/529, loss: 0.0056095123291015625 2023-01-21 11:44:53.945749: step: 900/529, loss: 0.0011039734818041325 2023-01-21 11:44:55.150282: step: 904/529, loss: 0.007837391458451748 2023-01-21 11:44:56.338253: step: 908/529, loss: 0.056517791002988815 2023-01-21 11:44:57.600130: step: 912/529, loss: 0.09569063782691956 2023-01-21 11:44:58.774899: step: 916/529, loss: 0.006251240149140358 2023-01-21 11:44:59.938217: step: 920/529, loss: 0.0009443163871765137 2023-01-21 11:45:01.126608: step: 924/529, loss: 0.0035196305252611637 2023-01-21 11:45:02.277180: step: 928/529, loss: 0.0057975295931100845 2023-01-21 11:45:03.441200: step: 932/529, loss: 0.0019431114196777344 2023-01-21 11:45:04.668380: step: 936/529, loss: 0.010368729010224342 2023-01-21 11:45:05.841821: step: 940/529, loss: 0.008254623971879482 2023-01-21 11:45:07.032803: step: 944/529, loss: 0.20224666595458984 2023-01-21 11:45:08.187425: step: 948/529, loss: 0.0019812583923339844 2023-01-21 11:45:09.403285: step: 952/529, loss: 0.001178741455078125 2023-01-21 11:45:10.593685: step: 956/529, loss: 0.06243705749511719 2023-01-21 11:45:11.793699: step: 960/529, loss: 0.04425458982586861 2023-01-21 11:45:12.952472: step: 964/529, loss: 0.010663795284926891 2023-01-21 11:45:14.154022: step: 968/529, loss: 0.022866439074277878 2023-01-21 11:45:15.333680: step: 972/529, loss: 0.0005489349132403731 2023-01-21 11:45:16.503297: step: 976/529, loss: 0.015569902025163174 2023-01-21 11:45:17.697887: step: 980/529, loss: 0.0002842903195414692 2023-01-21 11:45:18.875096: step: 984/529, loss: 0.06867074966430664 2023-01-21 11:45:20.062650: step: 988/529, loss: 0.0024886131286621094 2023-01-21 11:45:21.240232: step: 992/529, loss: 0.023933792486786842 2023-01-21 11:45:22.411130: step: 996/529, loss: 0.08007517457008362 2023-01-21 11:45:23.616472: step: 1000/529, loss: 0.009482097811996937 2023-01-21 11:45:24.799559: step: 1004/529, loss: 0.09728232026100159 2023-01-21 11:45:26.000313: step: 1008/529, loss: 0.03349161148071289 2023-01-21 11:45:27.185252: step: 1012/529, loss: 0.06912766396999359 2023-01-21 11:45:28.387099: step: 1016/529, loss: 0.004979133605957031 2023-01-21 11:45:29.583466: step: 1020/529, loss: 0.005501747131347656 2023-01-21 11:45:30.741780: step: 1024/529, loss: 0.0006181717035360634 2023-01-21 11:45:31.950368: step: 1028/529, loss: 0.1310766190290451 2023-01-21 11:45:33.111545: step: 1032/529, loss: 0.025401020422577858 2023-01-21 11:45:34.313431: step: 1036/529, loss: 0.009587478823959827 2023-01-21 11:45:35.485877: step: 1040/529, loss: 0.005969238467514515 2023-01-21 11:45:36.660095: step: 1044/529, loss: 0.0011276245350018144 2023-01-21 11:45:37.819572: step: 1048/529, loss: 0.009228324517607689 2023-01-21 11:45:39.027581: step: 1052/529, loss: 0.024904441088438034 2023-01-21 11:45:40.213095: step: 1056/529, loss: 0.06702175736427307 2023-01-21 11:45:41.406336: step: 1060/529, loss: 0.04661712795495987 2023-01-21 11:45:42.579893: step: 1064/529, loss: 0.0012708663707599044 2023-01-21 11:45:43.775864: step: 1068/529, loss: 0.4381944537162781 2023-01-21 11:45:44.947038: step: 1072/529, loss: 0.0014874637126922607 2023-01-21 11:45:46.096465: step: 1076/529, loss: 0.010035132989287376 2023-01-21 11:45:47.298796: step: 1080/529, loss: 0.00543665885925293 2023-01-21 11:45:48.490149: step: 1084/529, loss: 0.005423927679657936 2023-01-21 11:45:49.669667: step: 1088/529, loss: 0.0071464539505541325 2023-01-21 11:45:50.848357: step: 1092/529, loss: 0.015397739596664906 2023-01-21 11:45:52.065574: step: 1096/529, loss: 0.0010281562572345138 2023-01-21 11:45:53.302316: step: 1100/529, loss: 0.012851142324507236 2023-01-21 11:45:54.504820: step: 1104/529, loss: 0.00019159317889716476 2023-01-21 11:45:55.716063: step: 1108/529, loss: 0.06314335018396378 2023-01-21 11:45:56.917003: step: 1112/529, loss: 0.003363990690559149 2023-01-21 11:45:58.125145: step: 1116/529, loss: 0.0033940793946385384 2023-01-21 11:45:59.322419: step: 1120/529, loss: 0.024219417944550514 2023-01-21 11:46:00.479191: step: 1124/529, loss: 0.013112019747495651 2023-01-21 11:46:01.677281: step: 1128/529, loss: 0.04424247890710831 2023-01-21 11:46:02.855053: step: 1132/529, loss: 0.0023622512817382812 2023-01-21 11:46:04.014267: step: 1136/529, loss: 0.0319644957780838 2023-01-21 11:46:05.215397: step: 1140/529, loss: 9.479522850597277e-05 2023-01-21 11:46:06.453228: step: 1144/529, loss: 0.028423786163330078 2023-01-21 11:46:07.651916: step: 1148/529, loss: 0.03562774509191513 2023-01-21 11:46:08.828095: step: 1152/529, loss: 0.002202588366344571 2023-01-21 11:46:10.026308: step: 1156/529, loss: 0.02062663994729519 2023-01-21 11:46:11.217560: step: 1160/529, loss: 0.18561343848705292 2023-01-21 11:46:12.455121: step: 1164/529, loss: 0.012226772494614124 2023-01-21 11:46:13.678747: step: 1168/529, loss: 0.013913918286561966 2023-01-21 11:46:14.915589: step: 1172/529, loss: 0.00038404465885832906 2023-01-21 11:46:16.077019: step: 1176/529, loss: 0.029262160882353783 2023-01-21 11:46:17.254433: step: 1180/529, loss: 0.04159623757004738 2023-01-21 11:46:18.446084: step: 1184/529, loss: 0.0004094124014955014 2023-01-21 11:46:19.614081: step: 1188/529, loss: 0.010108851827681065 2023-01-21 11:46:20.797061: step: 1192/529, loss: 0.024247266352176666 2023-01-21 11:46:21.988696: step: 1196/529, loss: 0.006991386413574219 2023-01-21 11:46:23.154103: step: 1200/529, loss: 9.250640869140625e-05 2023-01-21 11:46:24.334926: step: 1204/529, loss: 0.007433605846017599 2023-01-21 11:46:25.513342: step: 1208/529, loss: 0.025221824645996094 2023-01-21 11:46:26.665265: step: 1212/529, loss: 0.009494495578110218 2023-01-21 11:46:27.888752: step: 1216/529, loss: 0.004331540782004595 2023-01-21 11:46:29.068238: step: 1220/529, loss: 0.03277161344885826 2023-01-21 11:46:30.301591: step: 1224/529, loss: 0.007504463195800781 2023-01-21 11:46:31.485393: step: 1228/529, loss: 0.01615610159933567 2023-01-21 11:46:32.670315: step: 1232/529, loss: 0.008563614450395107 2023-01-21 11:46:33.886777: step: 1236/529, loss: 0.03475017473101616 2023-01-21 11:46:35.098936: step: 1240/529, loss: 0.005573368165642023 2023-01-21 11:46:36.286475: step: 1244/529, loss: 0.027919579297304153 2023-01-21 11:46:37.458912: step: 1248/529, loss: 0.0002548217889852822 2023-01-21 11:46:38.633981: step: 1252/529, loss: 0.0007903099176473916 2023-01-21 11:46:39.825661: step: 1256/529, loss: 0.01017465628683567 2023-01-21 11:46:41.006391: step: 1260/529, loss: 0.007338523864746094 2023-01-21 11:46:42.203571: step: 1264/529, loss: 0.003681659698486328 2023-01-21 11:46:43.398875: step: 1268/529, loss: 0.009714031592011452 2023-01-21 11:46:44.602335: step: 1272/529, loss: 0.004846715833991766 2023-01-21 11:46:45.803166: step: 1276/529, loss: 0.01438795868307352 2023-01-21 11:46:46.980575: step: 1280/529, loss: 0.018788862973451614 2023-01-21 11:46:48.182394: step: 1284/529, loss: 0.006714058108627796 2023-01-21 11:46:49.353699: step: 1288/529, loss: 0.0031226160936057568 2023-01-21 11:46:50.545959: step: 1292/529, loss: 0.01138153113424778 2023-01-21 11:46:51.711860: step: 1296/529, loss: 0.013680171221494675 2023-01-21 11:46:52.923417: step: 1300/529, loss: 0.06703425198793411 2023-01-21 11:46:54.092174: step: 1304/529, loss: 0.04135749489068985 2023-01-21 11:46:55.325313: step: 1308/529, loss: 0.01202306803315878 2023-01-21 11:46:56.571268: step: 1312/529, loss: 0.0014865875709801912 2023-01-21 11:46:57.730710: step: 1316/529, loss: 0.0002413749898551032 2023-01-21 11:46:58.930620: step: 1320/529, loss: 0.040582943707704544 2023-01-21 11:47:00.086115: step: 1324/529, loss: 0.0029956817161291838 2023-01-21 11:47:01.287221: step: 1328/529, loss: 0.01362705323845148 2023-01-21 11:47:02.492993: step: 1332/529, loss: 0.05243435129523277 2023-01-21 11:47:03.684529: step: 1336/529, loss: 0.006578254513442516 2023-01-21 11:47:04.870398: step: 1340/529, loss: 0.013654327020049095 2023-01-21 11:47:06.055930: step: 1344/529, loss: 0.040641263127326965 2023-01-21 11:47:07.262878: step: 1348/529, loss: 0.00038814544677734375 2023-01-21 11:47:08.458366: step: 1352/529, loss: 0.1281740367412567 2023-01-21 11:47:09.629363: step: 1356/529, loss: 0.0008365631802007556 2023-01-21 11:47:10.804333: step: 1360/529, loss: 0.0012022495502606034 2023-01-21 11:47:11.971456: step: 1364/529, loss: 0.005544376093894243 2023-01-21 11:47:13.190881: step: 1368/529, loss: 0.00048084260197356343 2023-01-21 11:47:14.402724: step: 1372/529, loss: 0.0035402297507971525 2023-01-21 11:47:15.570411: step: 1376/529, loss: 0.00027799609233625233 2023-01-21 11:47:16.773094: step: 1380/529, loss: 0.01206054724752903 2023-01-21 11:47:17.975982: step: 1384/529, loss: 0.024425983428955078 2023-01-21 11:47:19.158278: step: 1388/529, loss: 0.0038602829445153475 2023-01-21 11:47:20.391330: step: 1392/529, loss: 0.0025281906127929688 2023-01-21 11:47:21.589743: step: 1396/529, loss: 0.03226060792803764 2023-01-21 11:47:22.798559: step: 1400/529, loss: 0.04071927070617676 2023-01-21 11:47:23.964163: step: 1404/529, loss: 0.0009320259559899569 2023-01-21 11:47:25.203724: step: 1408/529, loss: 0.07824049144983292 2023-01-21 11:47:26.429595: step: 1412/529, loss: 0.003738880157470703 2023-01-21 11:47:27.595647: step: 1416/529, loss: 0.05905161052942276 2023-01-21 11:47:28.770596: step: 1420/529, loss: 0.01171798724681139 2023-01-21 11:47:29.933367: step: 1424/529, loss: 0.012498283758759499 2023-01-21 11:47:31.163654: step: 1428/529, loss: 0.002703094622120261 2023-01-21 11:47:32.412733: step: 1432/529, loss: 0.011316108517348766 2023-01-21 11:47:33.661510: step: 1436/529, loss: 0.06114530935883522 2023-01-21 11:47:34.860170: step: 1440/529, loss: 0.0012956619029864669 2023-01-21 11:47:36.039147: step: 1444/529, loss: 0.019363021478056908 2023-01-21 11:47:37.264709: step: 1448/529, loss: 0.02879038080573082 2023-01-21 11:47:38.468579: step: 1452/529, loss: 0.047258805483579636 2023-01-21 11:47:39.649442: step: 1456/529, loss: 0.0016565322875976562 2023-01-21 11:47:40.878874: step: 1460/529, loss: 0.0010992051102221012 2023-01-21 11:47:42.042401: step: 1464/529, loss: 0.026791954413056374 2023-01-21 11:47:43.256687: step: 1468/529, loss: 0.008302116766571999 2023-01-21 11:47:44.434530: step: 1472/529, loss: 0.049391746520996094 2023-01-21 11:47:45.642798: step: 1476/529, loss: 0.08932819217443466 2023-01-21 11:47:46.870210: step: 1480/529, loss: 0.021849442273378372 2023-01-21 11:47:48.041994: step: 1484/529, loss: 0.013622093945741653 2023-01-21 11:47:49.274512: step: 1488/529, loss: 0.022086476907134056 2023-01-21 11:47:50.457519: step: 1492/529, loss: 0.0036905291490256786 2023-01-21 11:47:51.652469: step: 1496/529, loss: 0.01504507102072239 2023-01-21 11:47:52.813018: step: 1500/529, loss: 0.012854481115937233 2023-01-21 11:47:54.005606: step: 1504/529, loss: 0.0008449077722616494 2023-01-21 11:47:55.202480: step: 1508/529, loss: 0.015509843826293945 2023-01-21 11:47:56.407371: step: 1512/529, loss: 0.00522689800709486 2023-01-21 11:47:57.613384: step: 1516/529, loss: 0.030788518488407135 2023-01-21 11:47:58.815388: step: 1520/529, loss: 0.0058441162109375 2023-01-21 11:47:59.996827: step: 1524/529, loss: 0.0012895583640784025 2023-01-21 11:48:01.167480: step: 1528/529, loss: 0.0012809752952307463 2023-01-21 11:48:02.424884: step: 1532/529, loss: 0.04959602281451225 2023-01-21 11:48:03.590190: step: 1536/529, loss: 0.001171112060546875 2023-01-21 11:48:04.812602: step: 1540/529, loss: 0.02509183995425701 2023-01-21 11:48:05.999802: step: 1544/529, loss: 0.0034193038009107113 2023-01-21 11:48:07.212287: step: 1548/529, loss: 9.689330909168348e-05 2023-01-21 11:48:08.357514: step: 1552/529, loss: 0.005097580142319202 2023-01-21 11:48:09.569679: step: 1556/529, loss: 0.035167478024959564 2023-01-21 11:48:10.742813: step: 1560/529, loss: 0.027625370770692825 2023-01-21 11:48:11.916768: step: 1564/529, loss: 0.004566192626953125 2023-01-21 11:48:13.115224: step: 1568/529, loss: 0.0054801939986646175 2023-01-21 11:48:14.339780: step: 1572/529, loss: 1.7290217876434326 2023-01-21 11:48:15.506049: step: 1576/529, loss: 0.028194140642881393 2023-01-21 11:48:16.696440: step: 1580/529, loss: 0.007608318235725164 2023-01-21 11:48:17.900230: step: 1584/529, loss: 0.01574697531759739 2023-01-21 11:48:19.074620: step: 1588/529, loss: 0.026951074600219727 2023-01-21 11:48:20.269204: step: 1592/529, loss: 0.005021381191909313 2023-01-21 11:48:21.452899: step: 1596/529, loss: 0.006332588382065296 2023-01-21 11:48:22.658128: step: 1600/529, loss: 0.000997257186099887 2023-01-21 11:48:23.808813: step: 1604/529, loss: 0.006206321530044079 2023-01-21 11:48:25.017586: step: 1608/529, loss: 0.023686503991484642 2023-01-21 11:48:26.280531: step: 1612/529, loss: 0.022844314575195312 2023-01-21 11:48:27.456930: step: 1616/529, loss: 0.007186699192970991 2023-01-21 11:48:28.692949: step: 1620/529, loss: 0.00924606341868639 2023-01-21 11:48:29.937000: step: 1624/529, loss: 0.004923725500702858 2023-01-21 11:48:31.149421: step: 1628/529, loss: 0.025643253698945045 2023-01-21 11:48:32.336261: step: 1632/529, loss: 0.018687628209590912 2023-01-21 11:48:33.555847: step: 1636/529, loss: 0.024279212579131126 2023-01-21 11:48:34.750077: step: 1640/529, loss: 0.027161790058016777 2023-01-21 11:48:35.957867: step: 1644/529, loss: 0.003999567124992609 2023-01-21 11:48:37.142978: step: 1648/529, loss: 0.00629425048828125 2023-01-21 11:48:38.369501: step: 1652/529, loss: 0.0024860859848558903 2023-01-21 11:48:39.592953: step: 1656/529, loss: 0.04794752970337868 2023-01-21 11:48:40.815837: step: 1660/529, loss: 0.0013815879356116056 2023-01-21 11:48:42.039207: step: 1664/529, loss: 0.051721714437007904 2023-01-21 11:48:43.264424: step: 1668/529, loss: 0.0013279914855957031 2023-01-21 11:48:44.448320: step: 1672/529, loss: 0.045391276478767395 2023-01-21 11:48:45.629286: step: 1676/529, loss: 0.002954673720523715 2023-01-21 11:48:46.844847: step: 1680/529, loss: 0.013586617074906826 2023-01-21 11:48:48.021213: step: 1684/529, loss: 0.04551806300878525 2023-01-21 11:48:49.205606: step: 1688/529, loss: 0.006103039253503084 2023-01-21 11:48:50.415429: step: 1692/529, loss: 0.02467222325503826 2023-01-21 11:48:51.586716: step: 1696/529, loss: 0.014585686847567558 2023-01-21 11:48:52.784188: step: 1700/529, loss: 0.0015466214390471578 2023-01-21 11:48:54.000217: step: 1704/529, loss: 0.09454059600830078 2023-01-21 11:48:55.201064: step: 1708/529, loss: 0.020899580791592598 2023-01-21 11:48:56.442316: step: 1712/529, loss: 0.008185243234038353 2023-01-21 11:48:57.618047: step: 1716/529, loss: 0.001505184220150113 2023-01-21 11:48:58.766548: step: 1720/529, loss: 0.003099441761150956 2023-01-21 11:48:59.949097: step: 1724/529, loss: 0.011951684951782227 2023-01-21 11:49:01.141197: step: 1728/529, loss: 0.007378101348876953 2023-01-21 11:49:02.326410: step: 1732/529, loss: 0.00152587890625 2023-01-21 11:49:03.521510: step: 1736/529, loss: 0.0012115954887121916 2023-01-21 11:49:04.741714: step: 1740/529, loss: 0.03050861321389675 2023-01-21 11:49:05.907957: step: 1744/529, loss: 0.00444717425853014 2023-01-21 11:49:07.096647: step: 1748/529, loss: 0.04294319078326225 2023-01-21 11:49:08.272551: step: 1752/529, loss: 0.13906900584697723 2023-01-21 11:49:09.469724: step: 1756/529, loss: 0.004921626765280962 2023-01-21 11:49:10.676597: step: 1760/529, loss: 0.005080700386315584 2023-01-21 11:49:11.874916: step: 1764/529, loss: 0.012706183828413486 2023-01-21 11:49:13.100103: step: 1768/529, loss: 3.662109520519152e-05 2023-01-21 11:49:14.270584: step: 1772/529, loss: 0.03647499158978462 2023-01-21 11:49:15.493782: step: 1776/529, loss: 0.04907546192407608 2023-01-21 11:49:16.673938: step: 1780/529, loss: 0.3328498601913452 2023-01-21 11:49:17.882030: step: 1784/529, loss: 0.002761268522590399 2023-01-21 11:49:19.118209: step: 1788/529, loss: 0.004390048794448376 2023-01-21 11:49:20.345908: step: 1792/529, loss: 0.006041717249900103 2023-01-21 11:49:21.581061: step: 1796/529, loss: 0.017708588391542435 2023-01-21 11:49:22.758641: step: 1800/529, loss: 0.07054843753576279 2023-01-21 11:49:23.944797: step: 1804/529, loss: 0.009965325705707073 2023-01-21 11:49:25.125244: step: 1808/529, loss: 0.054119400680065155 2023-01-21 11:49:26.338045: step: 1812/529, loss: 0.06457634270191193 2023-01-21 11:49:27.532198: step: 1816/529, loss: 0.01677706278860569 2023-01-21 11:49:28.742383: step: 1820/529, loss: 0.007974434643983841 2023-01-21 11:49:29.907431: step: 1824/529, loss: 0.019582368433475494 2023-01-21 11:49:31.125780: step: 1828/529, loss: 0.019602203741669655 2023-01-21 11:49:32.320831: step: 1832/529, loss: 0.0006181717035360634 2023-01-21 11:49:33.494523: step: 1836/529, loss: 0.6808400750160217 2023-01-21 11:49:34.672210: step: 1840/529, loss: 0.0051863668486475945 2023-01-21 11:49:35.825579: step: 1844/529, loss: 0.055155351758003235 2023-01-21 11:49:36.987119: step: 1848/529, loss: 0.02918853983283043 2023-01-21 11:49:38.157680: step: 1852/529, loss: 0.024594496935606003 2023-01-21 11:49:39.345800: step: 1856/529, loss: 0.0910254493355751 2023-01-21 11:49:40.548777: step: 1860/529, loss: 0.05084395408630371 2023-01-21 11:49:41.757657: step: 1864/529, loss: 0.060060784220695496 2023-01-21 11:49:42.922075: step: 1868/529, loss: 0.05107574164867401 2023-01-21 11:49:44.137027: step: 1872/529, loss: 0.04668428748846054 2023-01-21 11:49:45.311147: step: 1876/529, loss: 0.009514618664979935 2023-01-21 11:49:46.512941: step: 1880/529, loss: 0.006371926981955767 2023-01-21 11:49:47.769683: step: 1884/529, loss: 0.1806192398071289 2023-01-21 11:49:48.937372: step: 1888/529, loss: 4.792213439941406e-05 2023-01-21 11:49:50.165801: step: 1892/529, loss: 0.2598453462123871 2023-01-21 11:49:51.364188: step: 1896/529, loss: 0.0001903533993754536 2023-01-21 11:49:52.577722: step: 1900/529, loss: 7.43865966796875e-05 2023-01-21 11:49:53.787697: step: 1904/529, loss: 0.012529182247817516 2023-01-21 11:49:54.965442: step: 1908/529, loss: 0.04476690664887428 2023-01-21 11:49:56.159375: step: 1912/529, loss: 0.005770206451416016 2023-01-21 11:49:57.403732: step: 1916/529, loss: 0.0027514458633959293 2023-01-21 11:49:58.566896: step: 1920/529, loss: 0.005097389221191406 2023-01-21 11:49:59.769338: step: 1924/529, loss: 0.009049415588378906 2023-01-21 11:50:00.966583: step: 1928/529, loss: 0.12448372691869736 2023-01-21 11:50:02.211258: step: 1932/529, loss: 0.017469406127929688 2023-01-21 11:50:03.380006: step: 1936/529, loss: 0.005369949154555798 2023-01-21 11:50:04.587389: step: 1940/529, loss: 0.054808903485536575 2023-01-21 11:50:05.817206: step: 1944/529, loss: 0.009992980398237705 2023-01-21 11:50:07.029995: step: 1948/529, loss: 0.040560342371463776 2023-01-21 11:50:08.234123: step: 1952/529, loss: 0.21151982247829437 2023-01-21 11:50:09.406144: step: 1956/529, loss: 0.14898987114429474 2023-01-21 11:50:10.577320: step: 1960/529, loss: 0.10263834148645401 2023-01-21 11:50:11.768231: step: 1964/529, loss: 0.03355102613568306 2023-01-21 11:50:12.944667: step: 1968/529, loss: 0.008838415145874023 2023-01-21 11:50:14.198377: step: 1972/529, loss: 0.00287456507794559 2023-01-21 11:50:15.383180: step: 1976/529, loss: 0.02179746702313423 2023-01-21 11:50:16.569716: step: 1980/529, loss: 0.0022665977012366056 2023-01-21 11:50:17.793980: step: 1984/529, loss: 0.0020546913146972656 2023-01-21 11:50:18.988342: step: 1988/529, loss: 0.02467956580221653 2023-01-21 11:50:20.137738: step: 1992/529, loss: 0.0016218662494793534 2023-01-21 11:50:21.338156: step: 1996/529, loss: 0.0455591194331646 2023-01-21 11:50:22.505954: step: 2000/529, loss: 0.012937354855239391 2023-01-21 11:50:23.711345: step: 2004/529, loss: 0.00029430389986373484 2023-01-21 11:50:24.902647: step: 2008/529, loss: 0.018345260992646217 2023-01-21 11:50:26.095181: step: 2012/529, loss: 0.01963214948773384 2023-01-21 11:50:27.255979: step: 2016/529, loss: 0.0324590690433979 2023-01-21 11:50:28.418963: step: 2020/529, loss: 0.003579521318897605 2023-01-21 11:50:29.608311: step: 2024/529, loss: 0.012766839005053043 2023-01-21 11:50:30.830858: step: 2028/529, loss: 0.07166843861341476 2023-01-21 11:50:32.070009: step: 2032/529, loss: 0.046289682388305664 2023-01-21 11:50:33.254066: step: 2036/529, loss: 0.0010502815712243319 2023-01-21 11:50:34.426846: step: 2040/529, loss: 0.0020748137030750513 2023-01-21 11:50:35.596729: step: 2044/529, loss: 4.320144944358617e-05 2023-01-21 11:50:36.822765: step: 2048/529, loss: 0.01425857562571764 2023-01-21 11:50:38.026795: step: 2052/529, loss: 0.0026103975251317024 2023-01-21 11:50:39.226201: step: 2056/529, loss: 0.03824196010828018 2023-01-21 11:50:40.437333: step: 2060/529, loss: 0.0035764696076512337 2023-01-21 11:50:41.617136: step: 2064/529, loss: 0.023006059229373932 2023-01-21 11:50:42.851525: step: 2068/529, loss: 0.03257961571216583 2023-01-21 11:50:44.081162: step: 2072/529, loss: 0.004044151399284601 2023-01-21 11:50:45.304269: step: 2076/529, loss: 0.00021162032498978078 2023-01-21 11:50:46.524019: step: 2080/529, loss: 0.11506490409374237 2023-01-21 11:50:47.745247: step: 2084/529, loss: 0.001313495566137135 2023-01-21 11:50:48.946984: step: 2088/529, loss: 0.012626123614609241 2023-01-21 11:50:50.187626: step: 2092/529, loss: 0.00032138824462890625 2023-01-21 11:50:51.353312: step: 2096/529, loss: 0.005664253141731024 2023-01-21 11:50:52.591211: step: 2100/529, loss: 0.03302164003252983 2023-01-21 11:50:53.779184: step: 2104/529, loss: 0.05964851751923561 2023-01-21 11:50:54.950649: step: 2108/529, loss: 0.006173897068947554 2023-01-21 11:50:56.102752: step: 2112/529, loss: 0.006352138239890337 2023-01-21 11:50:57.310742: step: 2116/529, loss: 0.09593772888183594 ================================================== Loss: 0.031 -------------------- Dev: {'event': {'p': 0.6079664570230608, 'r': 0.7723035952063915, 'f1': 0.6803519061583578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Test: {'event': {'p': 0.6472317491425772, 'r': 0.7914919113241462, 'f1': 0.7121293800539084}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Chinese: {'event': {'p': 0.5915492957746479, 'r': 0.7777777777777778, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Korean: {'event': {'p': 0.5538461538461539, 'r': 0.5714285714285714, 'f1': 0.5625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Russian: {'event': {'p': 0.391304347826087, 'r': 0.5, 'f1': 0.4390243902439025}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.5977249224405378, 'r': 0.7696404793608522, 'f1': 0.6728754365541327}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Russian: {'event': {'p': 0.6421001926782274, 'r': 0.7986818454164171, 'f1': 0.7118825100133512}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Russian: {'event': {'p': 0.5365853658536586, 'r': 0.6111111111111112, 'f1': 0.5714285714285715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} ****************************** 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 11:51:40.380660: step: 4/529, loss: 0.006571960635483265 2023-01-21 11:51:41.603733: step: 8/529, loss: 0.3789542317390442 2023-01-21 11:51:42.779793: step: 12/529, loss: 0.10134448856115341 2023-01-21 11:51:44.005472: step: 16/529, loss: 0.012534618377685547 2023-01-21 11:51:45.182210: step: 20/529, loss: 0.025675201788544655 2023-01-21 11:51:46.329445: step: 24/529, loss: 0.008911609649658203 2023-01-21 11:51:47.537924: step: 28/529, loss: 0.05687293782830238 2023-01-21 11:51:48.745289: step: 32/529, loss: 0.005775260739028454 2023-01-21 11:51:49.913039: step: 36/529, loss: 0.00014379025378730148 2023-01-21 11:51:51.142487: step: 40/529, loss: 0.015054893679916859 2023-01-21 11:51:52.273079: step: 44/529, loss: 0.005792808718979359 2023-01-21 11:51:53.584447: step: 48/529, loss: 0.003933334723114967 2023-01-21 11:51:54.787793: step: 52/529, loss: 0.00013227463932707906 2023-01-21 11:51:56.018239: step: 56/529, loss: 0.010695314034819603 2023-01-21 11:51:57.206883: step: 60/529, loss: 0.01738920249044895 2023-01-21 11:51:58.433893: step: 64/529, loss: 0.0016285896999761462 2023-01-21 11:51:59.630548: step: 68/529, loss: 0.0009321690304204822 2023-01-21 11:52:00.868095: step: 72/529, loss: 0.039693549275398254 2023-01-21 11:52:02.045976: step: 76/529, loss: 0.002345752902328968 2023-01-21 11:52:03.204325: step: 80/529, loss: 0.00020928384037688375 2023-01-21 11:52:04.406022: step: 84/529, loss: 0.009134196676313877 2023-01-21 11:52:05.573993: step: 88/529, loss: 0.0166186336427927 2023-01-21 11:52:06.773955: step: 92/529, loss: 0.0007817268487997353 2023-01-21 11:52:07.949125: step: 96/529, loss: 0.011416149325668812 2023-01-21 11:52:09.153304: step: 100/529, loss: 0.003180980682373047 2023-01-21 11:52:10.314304: step: 104/529, loss: 0.04525613784790039 2023-01-21 11:52:11.496917: step: 108/529, loss: 0.0002058029203908518 2023-01-21 11:52:12.692062: step: 112/529, loss: 0.0019274712540209293 2023-01-21 11:52:13.903889: step: 116/529, loss: 0.037876129150390625 2023-01-21 11:52:15.072968: step: 120/529, loss: 0.0005941391573287547 2023-01-21 11:52:16.264117: step: 124/529, loss: 0.0012660056818276644 2023-01-21 11:52:17.446714: step: 128/529, loss: 0.012882089242339134 2023-01-21 11:52:18.679372: step: 132/529, loss: 0.02698535844683647 2023-01-21 11:52:19.870350: step: 136/529, loss: 0.003217506455257535 2023-01-21 11:52:21.063775: step: 140/529, loss: 0.0021598339080810547 2023-01-21 11:52:22.243188: step: 144/529, loss: 0.06418894976377487 2023-01-21 11:52:23.388955: step: 148/529, loss: 0.03130693361163139 2023-01-21 11:52:24.598897: step: 152/529, loss: 0.011578226462006569 2023-01-21 11:52:25.788904: step: 156/529, loss: 0.0030247687827795744 2023-01-21 11:52:26.982301: step: 160/529, loss: 0.012471770867705345 2023-01-21 11:52:28.158671: step: 164/529, loss: 0.03228330612182617 2023-01-21 11:52:29.302750: step: 168/529, loss: 0.009148389101028442 2023-01-21 11:52:30.495249: step: 172/529, loss: 0.02418060228228569 2023-01-21 11:52:31.660189: step: 176/529, loss: 0.011884975247085094 2023-01-21 11:52:32.843399: step: 180/529, loss: 0.0002838134823832661 2023-01-21 11:52:34.013856: step: 184/529, loss: 0.0006986617809161544 2023-01-21 11:52:35.198324: step: 188/529, loss: 0.04712171480059624 2023-01-21 11:52:36.344884: step: 192/529, loss: 0.0741826519370079 2023-01-21 11:52:37.537050: step: 196/529, loss: 0.005088806618005037 2023-01-21 11:52:38.691944: step: 200/529, loss: 0.00757675152271986 2023-01-21 11:52:39.879748: step: 204/529, loss: 0.002328396076336503 2023-01-21 11:52:41.121683: step: 208/529, loss: 0.0028010366950184107 2023-01-21 11:52:42.292895: step: 212/529, loss: 0.01111297681927681 2023-01-21 11:52:43.493785: step: 216/529, loss: 0.01621084287762642 2023-01-21 11:52:44.717709: step: 220/529, loss: 0.005317973904311657 2023-01-21 11:52:45.877943: step: 224/529, loss: 0.00105457310564816 2023-01-21 11:52:47.065406: step: 228/529, loss: 0.00031681061955168843 2023-01-21 11:52:48.268618: step: 232/529, loss: 0.016783427447080612 2023-01-21 11:52:49.448918: step: 236/529, loss: 0.05583992227911949 2023-01-21 11:52:50.649683: step: 240/529, loss: 0.031087016686797142 2023-01-21 11:52:51.854185: step: 244/529, loss: 0.0019330024952068925 2023-01-21 11:52:53.023172: step: 248/529, loss: 0.00423278845846653 2023-01-21 11:52:54.219013: step: 252/529, loss: 0.00016126631817314774 2023-01-21 11:52:55.452034: step: 256/529, loss: 0.041787147521972656 2023-01-21 11:52:56.683839: step: 260/529, loss: 0.00141315464861691 2023-01-21 11:52:57.884465: step: 264/529, loss: 0.0016763686435297132 2023-01-21 11:52:59.138071: step: 268/529, loss: 0.021828651428222656 2023-01-21 11:53:00.337404: step: 272/529, loss: 0.0238539706915617 2023-01-21 11:53:01.489767: step: 276/529, loss: 0.0015583992935717106 2023-01-21 11:53:02.660075: step: 280/529, loss: 0.0005940437549725175 2023-01-21 11:53:03.873862: step: 284/529, loss: 0.0038047791458666325 2023-01-21 11:53:05.069953: step: 288/529, loss: 0.07631397247314453 2023-01-21 11:53:06.283748: step: 292/529, loss: 0.0028490067925304174 2023-01-21 11:53:07.485390: step: 296/529, loss: 0.0024476051330566406 2023-01-21 11:53:08.694329: step: 300/529, loss: 0.0007816314464434981 2023-01-21 11:53:09.875943: step: 304/529, loss: 0.014049815945327282 2023-01-21 11:53:11.096794: step: 308/529, loss: 0.004992294125258923 2023-01-21 11:53:12.278262: step: 312/529, loss: 0.010159874334931374 2023-01-21 11:53:13.497603: step: 316/529, loss: 0.00039901735726743937 2023-01-21 11:53:14.647770: step: 320/529, loss: 0.0036937713157385588 2023-01-21 11:53:15.854534: step: 324/529, loss: 0.02211008034646511 2023-01-21 11:53:17.046982: step: 328/529, loss: 0.0004680156707763672 2023-01-21 11:53:18.260964: step: 332/529, loss: 0.018275832757353783 2023-01-21 11:53:19.472432: step: 336/529, loss: 0.007916641421616077 2023-01-21 11:53:20.648772: step: 340/529, loss: 0.0007257461547851562 2023-01-21 11:53:21.849738: step: 344/529, loss: 0.007826900109648705 2023-01-21 11:53:23.031571: step: 348/529, loss: 0.01871032826602459 2023-01-21 11:53:24.243032: step: 352/529, loss: 9.880066500045359e-05 2023-01-21 11:53:25.439959: step: 356/529, loss: 0.0033885003067553043 2023-01-21 11:53:26.601725: step: 360/529, loss: 0.004598522093147039 2023-01-21 11:53:27.804824: step: 364/529, loss: 0.02944660186767578 2023-01-21 11:53:29.019104: step: 368/529, loss: 0.001666212105192244 2023-01-21 11:53:30.199820: step: 372/529, loss: 0.007786560337990522 2023-01-21 11:53:31.397851: step: 376/529, loss: 0.014383077621459961 2023-01-21 11:53:32.623115: step: 380/529, loss: 0.00168695452157408 2023-01-21 11:53:33.767208: step: 384/529, loss: 3.4856795537052676e-05 2023-01-21 11:53:34.936366: step: 388/529, loss: 9.841918654274195e-05 2023-01-21 11:53:36.114937: step: 392/529, loss: 0.00204048166051507 2023-01-21 11:53:37.297477: step: 396/529, loss: 0.00855102576315403 2023-01-21 11:53:38.478890: step: 400/529, loss: 0.022398853674530983 2023-01-21 11:53:39.628004: step: 404/529, loss: 0.024639606475830078 2023-01-21 11:53:40.816640: step: 408/529, loss: 0.0036026001907885075 2023-01-21 11:53:42.009668: step: 412/529, loss: 0.018871117383241653 2023-01-21 11:53:43.172342: step: 416/529, loss: 0.03730565309524536 2023-01-21 11:53:44.326781: step: 420/529, loss: 0.01784696616232395 2023-01-21 11:53:45.509544: step: 424/529, loss: 0.004335498902946711 2023-01-21 11:53:46.679055: step: 428/529, loss: 0.07183676213026047 2023-01-21 11:53:47.894316: step: 432/529, loss: 0.011652946472167969 2023-01-21 11:53:49.077061: step: 436/529, loss: 0.0003651619190350175 2023-01-21 11:53:50.299669: step: 440/529, loss: 0.03558044135570526 2023-01-21 11:53:51.454214: step: 444/529, loss: 0.024232864379882812 2023-01-21 11:53:52.648250: step: 448/529, loss: 0.045076750218868256 2023-01-21 11:53:53.843852: step: 452/529, loss: 0.003825378604233265 2023-01-21 11:53:55.069167: step: 456/529, loss: 0.04449443891644478 2023-01-21 11:53:56.267362: step: 460/529, loss: 0.003377723740413785 2023-01-21 11:53:57.462767: step: 464/529, loss: 0.0029262544121593237 2023-01-21 11:53:58.673024: step: 468/529, loss: 0.06795196235179901 2023-01-21 11:53:59.825511: step: 472/529, loss: 0.05705471336841583 2023-01-21 11:54:01.036396: step: 476/529, loss: 0.05007028579711914 2023-01-21 11:54:02.226464: step: 480/529, loss: 0.02044752985239029 2023-01-21 11:54:03.421543: step: 484/529, loss: 0.05402488633990288 2023-01-21 11:54:04.641101: step: 488/529, loss: 0.004245567601174116 2023-01-21 11:54:05.785344: step: 492/529, loss: 0.0018449783092364669 2023-01-21 11:54:06.985577: step: 496/529, loss: 0.02396526373922825 2023-01-21 11:54:08.160660: step: 500/529, loss: 0.055263713002204895 2023-01-21 11:54:09.368408: step: 504/529, loss: 0.02273092418909073 2023-01-21 11:54:10.583876: step: 508/529, loss: 0.014951705932617188 2023-01-21 11:54:11.771538: step: 512/529, loss: 0.010766410268843174 2023-01-21 11:54:12.974487: step: 516/529, loss: 0.005733966827392578 2023-01-21 11:54:14.185967: step: 520/529, loss: 0.011838722042739391 2023-01-21 11:54:15.386977: step: 524/529, loss: 0.01317259669303894 2023-01-21 11:54:16.611620: step: 528/529, loss: 0.03067016787827015 2023-01-21 11:54:17.784250: step: 532/529, loss: 0.004012489225715399 2023-01-21 11:54:18.999340: step: 536/529, loss: 0.04266476631164551 2023-01-21 11:54:20.221908: step: 540/529, loss: 0.6610426902770996 2023-01-21 11:54:21.431283: step: 544/529, loss: 0.014409112744033337 2023-01-21 11:54:22.633819: step: 548/529, loss: 1.2493132999225054e-05 2023-01-21 11:54:23.856638: step: 552/529, loss: 0.007291411980986595 2023-01-21 11:54:25.034488: step: 556/529, loss: 0.03713931888341904 2023-01-21 11:54:26.253121: step: 560/529, loss: 0.010931110940873623 2023-01-21 11:54:27.449035: step: 564/529, loss: 0.0015510559314861894 2023-01-21 11:54:28.675008: step: 568/529, loss: 0.007613754365593195 2023-01-21 11:54:29.817150: step: 572/529, loss: 0.05505962669849396 2023-01-21 11:54:31.022581: step: 576/529, loss: 0.0015285492409020662 2023-01-21 11:54:32.205643: step: 580/529, loss: 0.07072468101978302 2023-01-21 11:54:33.387907: step: 584/529, loss: 0.03504753112792969 2023-01-21 11:54:34.644143: step: 588/529, loss: 0.011196804232895374 2023-01-21 11:54:35.829081: step: 592/529, loss: 0.058238983154296875 2023-01-21 11:54:37.022631: step: 596/529, loss: 0.002355909440666437 2023-01-21 11:54:38.213392: step: 600/529, loss: 0.003253936767578125 2023-01-21 11:54:39.425078: step: 604/529, loss: 0.03973989561200142 2023-01-21 11:54:40.623316: step: 608/529, loss: 0.17533054947853088 2023-01-21 11:54:41.808590: step: 612/529, loss: 0.00117664341814816 2023-01-21 11:54:43.014625: step: 616/529, loss: 0.0010062218643724918 2023-01-21 11:54:44.244072: step: 620/529, loss: 0.0017116547096520662 2023-01-21 11:54:45.431041: step: 624/529, loss: 0.002258873078972101 2023-01-21 11:54:46.625205: step: 628/529, loss: 0.025373173877596855 2023-01-21 11:54:47.768869: step: 632/529, loss: 0.10393805056810379 2023-01-21 11:54:48.944498: step: 636/529, loss: 0.009538006968796253 2023-01-21 11:54:50.119016: step: 640/529, loss: 0.0037597655318677425 2023-01-21 11:54:51.349262: step: 644/529, loss: 0.0028879165183752775 2023-01-21 11:54:52.538196: step: 648/529, loss: 0.0011927604209631681 2023-01-21 11:54:53.735612: step: 652/529, loss: 0.003087806748226285 2023-01-21 11:54:54.941426: step: 656/529, loss: 0.053670503199100494 2023-01-21 11:54:56.113611: step: 660/529, loss: 0.019797706976532936 2023-01-21 11:54:57.307562: step: 664/529, loss: 0.0404604896903038 2023-01-21 11:54:58.482526: step: 668/529, loss: 0.0015056610573083162 2023-01-21 11:54:59.676974: step: 672/529, loss: 0.028074074536561966 2023-01-21 11:55:00.852472: step: 676/529, loss: 0.008553934283554554 2023-01-21 11:55:02.033419: step: 680/529, loss: 0.041733644902706146 2023-01-21 11:55:03.227014: step: 684/529, loss: 0.01942129246890545 2023-01-21 11:55:04.427131: step: 688/529, loss: 0.009439945220947266 2023-01-21 11:55:05.625532: step: 692/529, loss: 0.005882549099624157 2023-01-21 11:55:06.845441: step: 696/529, loss: 0.006945610046386719 2023-01-21 11:55:08.025247: step: 700/529, loss: 0.009824943728744984 2023-01-21 11:55:09.245708: step: 704/529, loss: 0.03566179424524307 2023-01-21 11:55:10.494907: step: 708/529, loss: 0.006748676300048828 2023-01-21 11:55:11.701822: step: 712/529, loss: 0.0012460709549486637 2023-01-21 11:55:12.903089: step: 716/529, loss: 0.002067947294563055 2023-01-21 11:55:14.108167: step: 720/529, loss: 0.00664520263671875 2023-01-21 11:55:15.295157: step: 724/529, loss: 0.03570222854614258 2023-01-21 11:55:16.552529: step: 728/529, loss: 0.010402202606201172 2023-01-21 11:55:17.719263: step: 732/529, loss: 0.016822339966893196 2023-01-21 11:55:18.919531: step: 736/529, loss: 0.000281858432572335 2023-01-21 11:55:20.111323: step: 740/529, loss: 0.026053808629512787 2023-01-21 11:55:21.370870: step: 744/529, loss: 0.03105487860739231 2023-01-21 11:55:22.531470: step: 748/529, loss: 1.71661376953125e-05 2023-01-21 11:55:23.703857: step: 752/529, loss: 9.121894981944934e-05 2023-01-21 11:55:24.888325: step: 756/529, loss: 0.0020624161697924137 2023-01-21 11:55:26.099492: step: 760/529, loss: 0.0011945724254474044 2023-01-21 11:55:27.280998: step: 764/529, loss: 0.0013222694396972656 2023-01-21 11:55:28.455342: step: 768/529, loss: 0.05161752924323082 2023-01-21 11:55:29.648547: step: 772/529, loss: 0.00734558142721653 2023-01-21 11:55:30.822270: step: 776/529, loss: 0.004113960545510054 2023-01-21 11:55:32.022490: step: 780/529, loss: 0.00046529772225767374 2023-01-21 11:55:33.224574: step: 784/529, loss: 0.03204412758350372 2023-01-21 11:55:34.434583: step: 788/529, loss: 0.0036285400856286287 2023-01-21 11:55:35.635332: step: 792/529, loss: 0.011181068606674671 2023-01-21 11:55:36.783252: step: 796/529, loss: 0.0006835937383584678 2023-01-21 11:55:37.959235: step: 800/529, loss: 0.006413078401237726 2023-01-21 11:55:39.158323: step: 804/529, loss: 0.31526806950569153 2023-01-21 11:55:40.313110: step: 808/529, loss: 0.26873940229415894 2023-01-21 11:55:41.463722: step: 812/529, loss: 0.013759994879364967 2023-01-21 11:55:42.696667: step: 816/529, loss: 0.0015799523098394275 2023-01-21 11:55:43.869252: step: 820/529, loss: 0.06830139458179474 2023-01-21 11:55:45.040992: step: 824/529, loss: 0.2790340185165405 2023-01-21 11:55:46.238221: step: 828/529, loss: 0.006582784466445446 2023-01-21 11:55:47.379895: step: 832/529, loss: 0.09185931086540222 2023-01-21 11:55:48.597756: step: 836/529, loss: 0.0755043476819992 2023-01-21 11:55:49.838237: step: 840/529, loss: 0.013502979651093483 2023-01-21 11:55:51.040624: step: 844/529, loss: 0.08826322853565216 2023-01-21 11:55:52.251437: step: 848/529, loss: 0.024544810876250267 2023-01-21 11:55:53.431206: step: 852/529, loss: 0.00805587787181139 2023-01-21 11:55:54.639428: step: 856/529, loss: 0.012436485849320889 2023-01-21 11:55:55.828339: step: 860/529, loss: 0.008637333288788795 2023-01-21 11:55:56.984852: step: 864/529, loss: 0.012133502401411533 2023-01-21 11:55:58.155979: step: 868/529, loss: 0.014812324196100235 2023-01-21 11:55:59.308982: step: 872/529, loss: 0.015912819653749466 2023-01-21 11:56:00.490446: step: 876/529, loss: 0.07575374096632004 2023-01-21 11:56:01.681374: step: 880/529, loss: 0.03498053550720215 2023-01-21 11:56:02.875994: step: 884/529, loss: 0.0006704330444335938 2023-01-21 11:56:04.086901: step: 888/529, loss: 0.00042639972525648773 2023-01-21 11:56:05.278357: step: 892/529, loss: 0.009276200085878372 2023-01-21 11:56:06.484911: step: 896/529, loss: 0.001554775284603238 2023-01-21 11:56:07.693390: step: 900/529, loss: 0.010940838605165482 2023-01-21 11:56:08.859550: step: 904/529, loss: 0.008047867566347122 2023-01-21 11:56:10.033886: step: 908/529, loss: 0.01586151123046875 2023-01-21 11:56:11.259798: step: 912/529, loss: 0.012513875029981136 2023-01-21 11:56:12.423767: step: 916/529, loss: 0.03560008853673935 2023-01-21 11:56:13.638437: step: 920/529, loss: 0.04224357753992081 2023-01-21 11:56:14.852598: step: 924/529, loss: 0.01746082305908203 2023-01-21 11:56:16.036937: step: 928/529, loss: 0.024683572351932526 2023-01-21 11:56:17.187775: step: 932/529, loss: 0.0329832062125206 2023-01-21 11:56:18.340825: step: 936/529, loss: 0.002388095948845148 2023-01-21 11:56:19.521083: step: 940/529, loss: 0.0007612705230712891 2023-01-21 11:56:20.715880: step: 944/529, loss: 0.043321993201971054 2023-01-21 11:56:21.885435: step: 948/529, loss: 0.0029769898392260075 2023-01-21 11:56:23.056920: step: 952/529, loss: 0.0005672454717569053 2023-01-21 11:56:24.237461: step: 956/529, loss: 0.04524478688836098 2023-01-21 11:56:25.452353: step: 960/529, loss: 0.0048018451780080795 2023-01-21 11:56:26.617744: step: 964/529, loss: 0.006884098052978516 2023-01-21 11:56:27.791311: step: 968/529, loss: 0.022586917504668236 2023-01-21 11:56:29.004512: step: 972/529, loss: 0.5738434195518494 2023-01-21 11:56:30.227190: step: 976/529, loss: 0.04804258421063423 2023-01-21 11:56:31.424367: step: 980/529, loss: 0.0022665977012366056 2023-01-21 11:56:32.606302: step: 984/529, loss: 0.0009860992431640625 2023-01-21 11:56:33.772395: step: 988/529, loss: 0.003027630038559437 2023-01-21 11:56:34.990118: step: 992/529, loss: 0.0001685619354248047 2023-01-21 11:56:36.205789: step: 996/529, loss: 0.0057128253392875195 2023-01-21 11:56:37.420549: step: 1000/529, loss: 0.08522529900074005 2023-01-21 11:56:38.575055: step: 1004/529, loss: 0.01741781271994114 2023-01-21 11:56:39.799750: step: 1008/529, loss: 0.007825016975402832 2023-01-21 11:56:40.979111: step: 1012/529, loss: 0.0022407532669603825 2023-01-21 11:56:42.170955: step: 1016/529, loss: 0.02085113525390625 2023-01-21 11:56:43.353966: step: 1020/529, loss: 0.007826042361557484 2023-01-21 11:56:44.565715: step: 1024/529, loss: 0.00010738372657215223 2023-01-21 11:56:45.730886: step: 1028/529, loss: 0.017702292650938034 2023-01-21 11:56:46.918488: step: 1032/529, loss: 0.0039351461455225945 2023-01-21 11:56:48.101918: step: 1036/529, loss: 0.013628054410219193 2023-01-21 11:56:49.320520: step: 1040/529, loss: 0.0008265495416708291 2023-01-21 11:56:50.550584: step: 1044/529, loss: 0.01953139342367649 2023-01-21 11:56:51.766689: step: 1048/529, loss: 0.0017587661277502775 2023-01-21 11:56:52.920970: step: 1052/529, loss: 0.0026020051445811987 2023-01-21 11:56:54.124877: step: 1056/529, loss: 0.006898117251694202 2023-01-21 11:56:55.320767: step: 1060/529, loss: 0.05593309551477432 2023-01-21 11:56:56.484517: step: 1064/529, loss: 0.06625232845544815 2023-01-21 11:56:57.677676: step: 1068/529, loss: 0.0061355591751635075 2023-01-21 11:56:58.845133: step: 1072/529, loss: 0.029186248779296875 2023-01-21 11:57:00.071288: step: 1076/529, loss: 7.495879981433973e-05 2023-01-21 11:57:01.240169: step: 1080/529, loss: 0.014934826642274857 2023-01-21 11:57:02.420325: step: 1084/529, loss: 1.7833710444392636e-05 2023-01-21 11:57:03.635428: step: 1088/529, loss: 0.005803585052490234 2023-01-21 11:57:04.857121: step: 1092/529, loss: 0.01610098034143448 2023-01-21 11:57:06.038212: step: 1096/529, loss: 0.01956787146627903 2023-01-21 11:57:07.223708: step: 1100/529, loss: 1.0681153071345761e-05 2023-01-21 11:57:08.422332: step: 1104/529, loss: 0.002038669539615512 2023-01-21 11:57:09.670600: step: 1108/529, loss: 0.02741718292236328 2023-01-21 11:57:10.864608: step: 1112/529, loss: 0.12127208709716797 2023-01-21 11:57:12.005775: step: 1116/529, loss: 0.0003486633358988911 2023-01-21 11:57:13.210685: step: 1120/529, loss: 0.0025796890258789062 2023-01-21 11:57:14.393470: step: 1124/529, loss: 0.0004753112734761089 2023-01-21 11:57:15.553807: step: 1128/529, loss: 0.0010111809242516756 2023-01-21 11:57:16.718660: step: 1132/529, loss: 0.01136474683880806 2023-01-21 11:57:17.911823: step: 1136/529, loss: 0.013244867324829102 2023-01-21 11:57:19.074137: step: 1140/529, loss: 0.015136336907744408 2023-01-21 11:57:20.335149: step: 1144/529, loss: 0.04565753787755966 2023-01-21 11:57:21.550540: step: 1148/529, loss: 0.03288231045007706 2023-01-21 11:57:22.748546: step: 1152/529, loss: 0.07918472588062286 2023-01-21 11:57:23.966305: step: 1156/529, loss: 0.009158515371382236 2023-01-21 11:57:25.164275: step: 1160/529, loss: 1.014816403388977 2023-01-21 11:57:26.368550: step: 1164/529, loss: 0.05476246029138565 2023-01-21 11:57:27.530860: step: 1168/529, loss: 0.0002135276881745085 2023-01-21 11:57:28.694770: step: 1172/529, loss: 0.028290940448641777 2023-01-21 11:57:29.864006: step: 1176/529, loss: 0.05173978954553604 2023-01-21 11:57:31.050522: step: 1180/529, loss: 0.04988740757107735 2023-01-21 11:57:32.277648: step: 1184/529, loss: 0.018652820959687233 2023-01-21 11:57:33.460459: step: 1188/529, loss: 0.03699808195233345 2023-01-21 11:57:34.650338: step: 1192/529, loss: 0.03100281022489071 2023-01-21 11:57:35.826040: step: 1196/529, loss: 0.02783365361392498 2023-01-21 11:57:37.016577: step: 1200/529, loss: 0.027106190100312233 2023-01-21 11:57:38.211315: step: 1204/529, loss: 0.002590274903923273 2023-01-21 11:57:39.378536: step: 1208/529, loss: 0.16605910658836365 2023-01-21 11:57:40.577956: step: 1212/529, loss: 0.3674212396144867 2023-01-21 11:57:41.857328: step: 1216/529, loss: 0.028063582256436348 2023-01-21 11:57:43.036272: step: 1220/529, loss: 0.016549110412597656 2023-01-21 11:57:44.206126: step: 1224/529, loss: 0.005625104531645775 2023-01-21 11:57:45.399008: step: 1228/529, loss: 0.00025601388188079 2023-01-21 11:57:46.561095: step: 1232/529, loss: 0.004915332887321711 2023-01-21 11:57:47.728980: step: 1236/529, loss: 0.004533672239631414 2023-01-21 11:57:48.912118: step: 1240/529, loss: 0.006754970643669367 2023-01-21 11:57:50.206624: step: 1244/529, loss: 0.03239727020263672 2023-01-21 11:57:51.425895: step: 1248/529, loss: 0.07681388407945633 2023-01-21 11:57:52.630957: step: 1252/529, loss: 0.005131632089614868 2023-01-21 11:57:53.853106: step: 1256/529, loss: 0.0021423338912427425 2023-01-21 11:57:55.042015: step: 1260/529, loss: 0.0013616561191156507 2023-01-21 11:57:56.222876: step: 1264/529, loss: 0.015633296221494675 2023-01-21 11:57:57.433646: step: 1268/529, loss: 0.04625868797302246 2023-01-21 11:57:58.601645: step: 1272/529, loss: 0.00016145705012604594 2023-01-21 11:57:59.807998: step: 1276/529, loss: 6.4849853515625e-05 2023-01-21 11:58:01.011098: step: 1280/529, loss: 0.20325584709644318 2023-01-21 11:58:02.221938: step: 1284/529, loss: 0.20736046135425568 2023-01-21 11:58:03.433711: step: 1288/529, loss: 0.003837489988654852 2023-01-21 11:58:04.613704: step: 1292/529, loss: 0.0018584252102300525 2023-01-21 11:58:05.756526: step: 1296/529, loss: 0.00408592214807868 2023-01-21 11:58:06.948620: step: 1300/529, loss: 0.011039352975785732 2023-01-21 11:58:08.136509: step: 1304/529, loss: 0.0006231308216229081 2023-01-21 11:58:09.361847: step: 1308/529, loss: 0.05428466945886612 2023-01-21 11:58:10.595936: step: 1312/529, loss: 0.009240532293915749 2023-01-21 11:58:11.777107: step: 1316/529, loss: 0.003713846206665039 2023-01-21 11:58:12.997066: step: 1320/529, loss: 0.02912120893597603 2023-01-21 11:58:14.191152: step: 1324/529, loss: 0.00027823448181152344 2023-01-21 11:58:15.396164: step: 1328/529, loss: 0.0011169433128088713 2023-01-21 11:58:16.598785: step: 1332/529, loss: 0.0014785766834393144 2023-01-21 11:58:17.791246: step: 1336/529, loss: 0.000480842572869733 2023-01-21 11:58:18.951061: step: 1340/529, loss: 0.018392659723758698 2023-01-21 11:58:20.172008: step: 1344/529, loss: 0.004159736912697554 2023-01-21 11:58:21.378193: step: 1348/529, loss: 0.021472740918397903 2023-01-21 11:58:22.577151: step: 1352/529, loss: 0.03484220430254936 2023-01-21 11:58:23.762298: step: 1356/529, loss: 0.0028735161758959293 2023-01-21 11:58:24.951133: step: 1360/529, loss: 0.0011415481567382812 2023-01-21 11:58:26.118177: step: 1364/529, loss: 0.005647468846291304 2023-01-21 11:58:27.296802: step: 1368/529, loss: 0.008494806475937366 2023-01-21 11:58:28.477583: step: 1372/529, loss: 0.0046915714628994465 2023-01-21 11:58:29.646691: step: 1376/529, loss: 0.013881778344511986 2023-01-21 11:58:30.885294: step: 1380/529, loss: 0.0013718605041503906 2023-01-21 11:58:32.051853: step: 1384/529, loss: 0.007982289418578148 2023-01-21 11:58:33.241178: step: 1388/529, loss: 5.145073009771295e-05 2023-01-21 11:58:34.408933: step: 1392/529, loss: 0.0008903503767214715 2023-01-21 11:58:35.628101: step: 1396/529, loss: 0.0009665489196777344 2023-01-21 11:58:36.787357: step: 1400/529, loss: 0.0009942532051354647 2023-01-21 11:58:37.992470: step: 1404/529, loss: 0.04839128255844116 2023-01-21 11:58:39.146295: step: 1408/529, loss: 0.00291271205060184 2023-01-21 11:58:40.301656: step: 1412/529, loss: 0.03854217752814293 2023-01-21 11:58:41.460355: step: 1416/529, loss: 0.000541305576916784 2023-01-21 11:58:42.671514: step: 1420/529, loss: 0.01825394481420517 2023-01-21 11:58:43.870149: step: 1424/529, loss: 0.1423528641462326 2023-01-21 11:58:45.062386: step: 1428/529, loss: 0.00010032653517555445 2023-01-21 11:58:46.234568: step: 1432/529, loss: 0.0003559827746357769 2023-01-21 11:58:47.405803: step: 1436/529, loss: 7.781982276355848e-05 2023-01-21 11:58:48.620058: step: 1440/529, loss: 0.0014945983421057463 2023-01-21 11:58:49.841444: step: 1444/529, loss: 3.376007225597277e-05 2023-01-21 11:58:51.028088: step: 1448/529, loss: 0.01575806923210621 2023-01-21 11:58:52.190748: step: 1452/529, loss: 0.0018788338638842106 2023-01-21 11:58:53.369939: step: 1456/529, loss: 0.002624702639877796 2023-01-21 11:58:54.580889: step: 1460/529, loss: 0.0018415451049804688 2023-01-21 11:58:55.761579: step: 1464/529, loss: 0.004680562298744917 2023-01-21 11:58:57.000844: step: 1468/529, loss: 0.05300407484173775 2023-01-21 11:58:58.176575: step: 1472/529, loss: 0.0304412841796875 2023-01-21 11:58:59.412247: step: 1476/529, loss: 0.121337890625 2023-01-21 11:59:00.595849: step: 1480/529, loss: 0.027698229998350143 2023-01-21 11:59:01.801014: step: 1484/529, loss: 0.04926319420337677 2023-01-21 11:59:03.010411: step: 1488/529, loss: 0.022179411724209785 2023-01-21 11:59:04.192025: step: 1492/529, loss: 0.00044317246647551656 2023-01-21 11:59:05.380457: step: 1496/529, loss: 0.0031815350521355867 2023-01-21 11:59:06.587173: step: 1500/529, loss: 0.0795368179678917 2023-01-21 11:59:07.823053: step: 1504/529, loss: 0.02361736260354519 2023-01-21 11:59:09.024554: step: 1508/529, loss: 0.06574898213148117 2023-01-21 11:59:10.260433: step: 1512/529, loss: 0.1653156280517578 2023-01-21 11:59:11.416758: step: 1516/529, loss: 0.021730568259954453 2023-01-21 11:59:12.680513: step: 1520/529, loss: 0.003413581755012274 2023-01-21 11:59:13.904470: step: 1524/529, loss: 0.00037059784517623484 2023-01-21 11:59:15.076024: step: 1528/529, loss: 0.09672079235315323 2023-01-21 11:59:16.323073: step: 1532/529, loss: 0.051881350576877594 2023-01-21 11:59:17.536593: step: 1536/529, loss: 0.004330158233642578 2023-01-21 11:59:18.728535: step: 1540/529, loss: 0.011902237311005592 2023-01-21 11:59:19.899586: step: 1544/529, loss: 0.0005717277526855469 2023-01-21 11:59:21.094491: step: 1548/529, loss: 0.29159051179885864 2023-01-21 11:59:22.294008: step: 1552/529, loss: 0.009038162417709827 2023-01-21 11:59:23.463531: step: 1556/529, loss: 0.001659661647863686 2023-01-21 11:59:24.668038: step: 1560/529, loss: 0.059976864606142044 2023-01-21 11:59:25.871298: step: 1564/529, loss: 0.01754627376794815 2023-01-21 11:59:27.064817: step: 1568/529, loss: 0.00023312569828704 2023-01-21 11:59:28.251185: step: 1572/529, loss: 0.005563163664191961 2023-01-21 11:59:29.427853: step: 1576/529, loss: 0.0010837077861651778 2023-01-21 11:59:30.582822: step: 1580/529, loss: 0.00657157925888896 2023-01-21 11:59:31.768612: step: 1584/529, loss: 0.008053207769989967 2023-01-21 11:59:32.929018: step: 1588/529, loss: 0.0010655403602868319 2023-01-21 11:59:34.108172: step: 1592/529, loss: 0.0046104430221021175 2023-01-21 11:59:35.309803: step: 1596/529, loss: 0.005394125357270241 2023-01-21 11:59:36.480562: step: 1600/529, loss: 0.017582226544618607 2023-01-21 11:59:37.669978: step: 1604/529, loss: 0.03404712677001953 2023-01-21 11:59:38.852225: step: 1608/529, loss: 0.02000255510210991 2023-01-21 11:59:40.035952: step: 1612/529, loss: 0.004681883845478296 2023-01-21 11:59:41.219245: step: 1616/529, loss: 0.0028125762473791838 2023-01-21 11:59:42.368169: step: 1620/529, loss: 0.016115987673401833 2023-01-21 11:59:43.584289: step: 1624/529, loss: 0.009394645690917969 2023-01-21 11:59:44.774290: step: 1628/529, loss: 0.00820016860961914 2023-01-21 11:59:45.937747: step: 1632/529, loss: 0.09341640770435333 2023-01-21 11:59:47.165461: step: 1636/529, loss: 0.019188690930604935 2023-01-21 11:59:48.354722: step: 1640/529, loss: 0.09687347710132599 2023-01-21 11:59:49.517826: step: 1644/529, loss: 0.00403518695384264 2023-01-21 11:59:50.725755: step: 1648/529, loss: 0.03035411797463894 2023-01-21 11:59:51.916374: step: 1652/529, loss: 0.0021083832252770662 2023-01-21 11:59:53.091549: step: 1656/529, loss: 0.016806984320282936 2023-01-21 11:59:54.273639: step: 1660/529, loss: 0.14373797178268433 2023-01-21 11:59:55.466989: step: 1664/529, loss: 0.01669750176370144 2023-01-21 11:59:56.649636: step: 1668/529, loss: 0.07492275536060333 2023-01-21 11:59:57.808267: step: 1672/529, loss: 0.021019555628299713 2023-01-21 11:59:59.068807: step: 1676/529, loss: 0.04082183539867401 2023-01-21 12:00:00.286121: step: 1680/529, loss: 0.0009871483780443668 2023-01-21 12:00:01.430417: step: 1684/529, loss: 0.0035179615952074528 2023-01-21 12:00:02.595722: step: 1688/529, loss: 0.0017014980548992753 2023-01-21 12:00:03.774621: step: 1692/529, loss: 0.056725598871707916 2023-01-21 12:00:04.993683: step: 1696/529, loss: 0.18515071272850037 2023-01-21 12:00:06.170153: step: 1700/529, loss: 0.04341382905840874 2023-01-21 12:00:07.340714: step: 1704/529, loss: 0.021404171362519264 2023-01-21 12:00:08.522218: step: 1708/529, loss: 3.337860107421875e-05 2023-01-21 12:00:09.694900: step: 1712/529, loss: 0.002731704618781805 2023-01-21 12:00:10.889462: step: 1716/529, loss: 0.012829303741455078 2023-01-21 12:00:12.042859: step: 1720/529, loss: 0.000878453254699707 2023-01-21 12:00:13.235846: step: 1724/529, loss: 0.02314412221312523 2023-01-21 12:00:14.469570: step: 1728/529, loss: 0.0098114013671875 2023-01-21 12:00:15.697339: step: 1732/529, loss: 0.037679292261600494 2023-01-21 12:00:16.894191: step: 1736/529, loss: 0.02869434282183647 2023-01-21 12:00:18.091932: step: 1740/529, loss: 0.014662742614746094 2023-01-21 12:00:19.251080: step: 1744/529, loss: 0.047613147646188736 2023-01-21 12:00:20.454434: step: 1748/529, loss: 0.007218551822006702 2023-01-21 12:00:21.631586: step: 1752/529, loss: 0.05199103429913521 2023-01-21 12:00:22.832956: step: 1756/529, loss: 0.00269489292986691 2023-01-21 12:00:24.018412: step: 1760/529, loss: 0.0016673087375238538 2023-01-21 12:00:25.189323: step: 1764/529, loss: 0.012362862005829811 2023-01-21 12:00:26.388622: step: 1768/529, loss: 0.1458592563867569 2023-01-21 12:00:27.612077: step: 1772/529, loss: 0.05312805250287056 2023-01-21 12:00:28.794250: step: 1776/529, loss: 0.0008801460498943925 2023-01-21 12:00:29.986299: step: 1780/529, loss: 0.054697465151548386 2023-01-21 12:00:31.149251: step: 1784/529, loss: 0.007571316324174404 2023-01-21 12:00:32.356446: step: 1788/529, loss: 0.0024704933166503906 2023-01-21 12:00:33.556760: step: 1792/529, loss: 0.0009280681842938066 2023-01-21 12:00:34.770405: step: 1796/529, loss: 0.003054714296013117 2023-01-21 12:00:35.949916: step: 1800/529, loss: 0.006800842471420765 2023-01-21 12:00:37.181247: step: 1804/529, loss: 0.01835303194820881 2023-01-21 12:00:38.321995: step: 1808/529, loss: 0.03291856124997139 2023-01-21 12:00:39.542327: step: 1812/529, loss: 0.007248306181281805 2023-01-21 12:00:40.720264: step: 1816/529, loss: 0.006799554452300072 2023-01-21 12:00:41.963143: step: 1820/529, loss: 0.012285137549042702 2023-01-21 12:00:43.131626: step: 1824/529, loss: 0.008073901757597923 2023-01-21 12:00:44.274893: step: 1828/529, loss: 0.008564376272261143 2023-01-21 12:00:45.447235: step: 1832/529, loss: 0.04788989946246147 2023-01-21 12:00:46.694365: step: 1836/529, loss: 0.0033513070084154606 2023-01-21 12:00:47.909706: step: 1840/529, loss: 0.07816791534423828 2023-01-21 12:00:49.113928: step: 1844/529, loss: 0.0001598358212504536 2023-01-21 12:00:50.335364: step: 1848/529, loss: 0.008773136883974075 2023-01-21 12:00:51.503595: step: 1852/529, loss: 0.004462623503059149 2023-01-21 12:00:52.700077: step: 1856/529, loss: 0.008317852392792702 2023-01-21 12:00:53.892796: step: 1860/529, loss: 0.006545543670654297 2023-01-21 12:00:55.089669: step: 1864/529, loss: 0.14648190140724182 2023-01-21 12:00:56.315352: step: 1868/529, loss: 0.024390220642089844 2023-01-21 12:00:57.495782: step: 1872/529, loss: 0.017086029052734375 2023-01-21 12:00:58.691035: step: 1876/529, loss: 0.02382678911089897 2023-01-21 12:00:59.884210: step: 1880/529, loss: 0.004903741180896759 2023-01-21 12:01:01.068573: step: 1884/529, loss: 0.00018033981905318797 2023-01-21 12:01:02.250135: step: 1888/529, loss: 0.03513813391327858 2023-01-21 12:01:03.395962: step: 1892/529, loss: 0.000362205522833392 2023-01-21 12:01:04.574046: step: 1896/529, loss: 0.0208453182131052 2023-01-21 12:01:05.784196: step: 1900/529, loss: 0.009539700113236904 2023-01-21 12:01:07.013991: step: 1904/529, loss: 0.01561746560037136 2023-01-21 12:01:08.219297: step: 1908/529, loss: 0.09768714755773544 2023-01-21 12:01:09.446409: step: 1912/529, loss: 0.025345230475068092 2023-01-21 12:01:10.625611: step: 1916/529, loss: 0.027213675901293755 2023-01-21 12:01:11.869279: step: 1920/529, loss: 0.002090645022690296 2023-01-21 12:01:13.039757: step: 1924/529, loss: 0.026171397417783737 2023-01-21 12:01:14.224553: step: 1928/529, loss: 0.010178041644394398 2023-01-21 12:01:15.431427: step: 1932/529, loss: 0.05351848527789116 2023-01-21 12:01:16.606064: step: 1936/529, loss: 0.0030673982109874487 2023-01-21 12:01:17.770739: step: 1940/529, loss: 0.00229988107457757 2023-01-21 12:01:18.927966: step: 1944/529, loss: 0.011982298456132412 2023-01-21 12:01:20.097905: step: 1948/529, loss: 0.09695949405431747 2023-01-21 12:01:21.267267: step: 1952/529, loss: 0.00252456683665514 2023-01-21 12:01:22.440505: step: 1956/529, loss: 0.00902624148875475 2023-01-21 12:01:23.623685: step: 1960/529, loss: 0.009364367462694645 2023-01-21 12:01:24.800787: step: 1964/529, loss: 0.005198335740715265 2023-01-21 12:01:26.007546: step: 1968/529, loss: 0.0012603760696947575 2023-01-21 12:01:27.245690: step: 1972/529, loss: 0.00810775812715292 2023-01-21 12:01:28.455379: step: 1976/529, loss: 0.0907234251499176 2023-01-21 12:01:29.635136: step: 1980/529, loss: 0.001408433890901506 2023-01-21 12:01:30.812541: step: 1984/529, loss: 0.0027451515197753906 2023-01-21 12:01:32.048535: step: 1988/529, loss: 0.011291027069091797 2023-01-21 12:01:33.258537: step: 1992/529, loss: 0.002602529479190707 2023-01-21 12:01:34.441808: step: 1996/529, loss: 0.005759811960160732 2023-01-21 12:01:35.641006: step: 2000/529, loss: 0.01916332170367241 2023-01-21 12:01:36.815902: step: 2004/529, loss: 0.004253959748893976 2023-01-21 12:01:37.960767: step: 2008/529, loss: 0.00305099505931139 2023-01-21 12:01:39.141881: step: 2012/529, loss: 0.0014254569541662931 2023-01-21 12:01:40.304692: step: 2016/529, loss: 0.006015682127326727 2023-01-21 12:01:41.456941: step: 2020/529, loss: 0.014016724191606045 2023-01-21 12:01:42.702623: step: 2024/529, loss: 0.0018743515247479081 2023-01-21 12:01:43.920446: step: 2028/529, loss: 0.011777544394135475 2023-01-21 12:01:45.107696: step: 2032/529, loss: 0.020840167999267578 2023-01-21 12:01:46.278811: step: 2036/529, loss: 0.0001446724054403603 2023-01-21 12:01:47.469279: step: 2040/529, loss: 0.05276113003492355 2023-01-21 12:01:48.641551: step: 2044/529, loss: 0.04874706268310547 2023-01-21 12:01:49.844680: step: 2048/529, loss: 0.021726323291659355 2023-01-21 12:01:51.036826: step: 2052/529, loss: 0.0004444122314453125 2023-01-21 12:01:52.233819: step: 2056/529, loss: 0.0022496222518384457 2023-01-21 12:01:53.506503: step: 2060/529, loss: 0.018486356362700462 2023-01-21 12:01:54.738878: step: 2064/529, loss: 0.03096456453204155 2023-01-21 12:01:55.903263: step: 2068/529, loss: 0.010410308837890625 2023-01-21 12:01:57.121047: step: 2072/529, loss: 0.00010929107520496473 2023-01-21 12:01:58.311905: step: 2076/529, loss: 0.027348138391971588 2023-01-21 12:01:59.508981: step: 2080/529, loss: 0.015237999148666859 2023-01-21 12:02:00.685685: step: 2084/529, loss: 0.021816540509462357 2023-01-21 12:02:01.922976: step: 2088/529, loss: 0.004836464300751686 2023-01-21 12:02:03.165799: step: 2092/529, loss: 0.0010181904071941972 2023-01-21 12:02:04.350197: step: 2096/529, loss: 0.006256484892219305 2023-01-21 12:02:05.541076: step: 2100/529, loss: 0.03036174736917019 2023-01-21 12:02:06.798872: step: 2104/529, loss: 0.36529675126075745 2023-01-21 12:02:07.999069: step: 2108/529, loss: 0.010082244873046875 2023-01-21 12:02:09.184193: step: 2112/529, loss: 0.03462104871869087 2023-01-21 12:02:10.378516: step: 2116/529, loss: 0.00874700490385294 ================================================== Loss: 0.029 -------------------- Dev: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Test: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Chinese: {'event': {'p': 0.6376811594202898, 'r': 0.8148148148148148, 'f1': 0.7154471544715448}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Korean: {'event': {'p': 0.6739130434782609, 'r': 0.49206349206349204, 'f1': 0.5688073394495413}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 12:03:01.455813: step: 4/529, loss: 0.06424503028392792 2023-01-21 12:03:02.653628: step: 8/529, loss: 0.008706474676728249 2023-01-21 12:03:03.873349: step: 12/529, loss: 0.007932472042739391 2023-01-21 12:03:05.057615: step: 16/529, loss: 7.572174217784777e-05 2023-01-21 12:03:06.238038: step: 20/529, loss: 0.09052447974681854 2023-01-21 12:03:07.450292: step: 24/529, loss: 8.277893357444555e-05 2023-01-21 12:03:08.614211: step: 28/529, loss: 0.004005813505500555 2023-01-21 12:03:09.797831: step: 32/529, loss: 0.005010413937270641 2023-01-21 12:03:11.004372: step: 36/529, loss: 0.015244293957948685 2023-01-21 12:03:12.170095: step: 40/529, loss: 0.028431154787540436 2023-01-21 12:03:13.382718: step: 44/529, loss: 0.00016918181790970266 2023-01-21 12:03:14.571012: step: 48/529, loss: 0.04810895770788193 2023-01-21 12:03:15.775872: step: 52/529, loss: 0.020151138305664062 2023-01-21 12:03:16.988134: step: 56/529, loss: 0.0078071593306958675 2023-01-21 12:03:18.121571: step: 60/529, loss: 0.002602195832878351 2023-01-21 12:03:19.337865: step: 64/529, loss: 0.0021851540077477694 2023-01-21 12:03:20.540825: step: 68/529, loss: 0.028033830225467682 2023-01-21 12:03:21.718093: step: 72/529, loss: 0.019364167004823685 2023-01-21 12:03:22.882206: step: 76/529, loss: 0.05893249809741974 2023-01-21 12:03:24.053555: step: 80/529, loss: 0.137858584523201 2023-01-21 12:03:25.250384: step: 84/529, loss: 0.0037191391456872225 2023-01-21 12:03:26.482968: step: 88/529, loss: 0.0005077362293377519 2023-01-21 12:03:27.675022: step: 92/529, loss: 0.021845554932951927 2023-01-21 12:03:28.837778: step: 96/529, loss: 0.004063797183334827 2023-01-21 12:03:29.984013: step: 100/529, loss: 0.007959174923598766 2023-01-21 12:03:31.178304: step: 104/529, loss: 0.0042365072295069695 2023-01-21 12:03:32.380894: step: 108/529, loss: 0.02066984213888645 2023-01-21 12:03:33.587425: step: 112/529, loss: 0.016972161829471588 2023-01-21 12:03:34.804284: step: 116/529, loss: 0.17690548300743103 2023-01-21 12:03:35.992775: step: 120/529, loss: 0.07662391662597656 2023-01-21 12:03:37.134788: step: 124/529, loss: 0.04297695308923721 2023-01-21 12:03:38.330855: step: 128/529, loss: 0.0019210815662518144 2023-01-21 12:03:39.543632: step: 132/529, loss: 0.0053751468658447266 2023-01-21 12:03:40.745173: step: 136/529, loss: 0.00046062469482421875 2023-01-21 12:03:41.923461: step: 140/529, loss: 0.008402824401855469 2023-01-21 12:03:43.097104: step: 144/529, loss: 0.009838677011430264 2023-01-21 12:03:44.286946: step: 148/529, loss: 0.003105735871940851 2023-01-21 12:03:45.478117: step: 152/529, loss: 0.25977858901023865 2023-01-21 12:03:46.696028: step: 156/529, loss: 0.009750175289809704 2023-01-21 12:03:47.865158: step: 160/529, loss: 0.003467082977294922 2023-01-21 12:03:49.065233: step: 164/529, loss: 0.0028938292525708675 2023-01-21 12:03:50.291156: step: 168/529, loss: 0.004630994983017445 2023-01-21 12:03:51.464150: step: 172/529, loss: 0.0011221885215491056 2023-01-21 12:03:52.682199: step: 176/529, loss: 0.002478695008903742 2023-01-21 12:03:53.862379: step: 180/529, loss: 0.0029655457474291325 2023-01-21 12:03:55.073434: step: 184/529, loss: 0.0015304565895348787 2023-01-21 12:03:56.255928: step: 188/529, loss: 0.004024410620331764 2023-01-21 12:03:57.399523: step: 192/529, loss: 0.017668360844254494 2023-01-21 12:03:58.604969: step: 196/529, loss: 0.006122017279267311 2023-01-21 12:03:59.780752: step: 200/529, loss: 0.05918922647833824 2023-01-21 12:04:00.992352: step: 204/529, loss: 0.004817390814423561 2023-01-21 12:04:02.178033: step: 208/529, loss: 0.008568000979721546 2023-01-21 12:04:03.337718: step: 212/529, loss: 0.0013031960697844625 2023-01-21 12:04:04.542162: step: 216/529, loss: 0.0086517333984375 2023-01-21 12:04:05.740081: step: 220/529, loss: 0.023004816845059395 2023-01-21 12:04:06.937385: step: 224/529, loss: 0.005378055851906538 2023-01-21 12:04:08.080089: step: 228/529, loss: 0.03881673887372017 2023-01-21 12:04:09.261765: step: 232/529, loss: 0.007745838724076748 2023-01-21 12:04:10.463391: step: 236/529, loss: 0.08488941937685013 2023-01-21 12:04:11.642335: step: 240/529, loss: 0.016089629381895065 2023-01-21 12:04:12.827704: step: 244/529, loss: 0.008154870010912418 2023-01-21 12:04:13.999177: step: 248/529, loss: 0.06566081196069717 2023-01-21 12:04:15.204475: step: 252/529, loss: 2.7370453608455136e-05 2023-01-21 12:04:16.358090: step: 256/529, loss: 4.715919931186363e-05 2023-01-21 12:04:17.546838: step: 260/529, loss: 0.02048931084573269 2023-01-21 12:04:18.732605: step: 264/529, loss: 0.0015218734042719007 2023-01-21 12:04:19.894859: step: 268/529, loss: 0.03358745574951172 2023-01-21 12:04:21.090258: step: 272/529, loss: 0.00013828277587890625 2023-01-21 12:04:22.306052: step: 276/529, loss: 0.004203987307846546 2023-01-21 12:04:23.520857: step: 280/529, loss: 0.01325836218893528 2023-01-21 12:04:24.722208: step: 284/529, loss: 0.0995725691318512 2023-01-21 12:04:25.910298: step: 288/529, loss: 0.0031786919571459293 2023-01-21 12:04:27.134244: step: 292/529, loss: 0.00556869525462389 2023-01-21 12:04:28.360474: step: 296/529, loss: 0.002055597258731723 2023-01-21 12:04:29.554419: step: 300/529, loss: 0.001387095544487238 2023-01-21 12:04:30.743475: step: 304/529, loss: 0.001844215439632535 2023-01-21 12:04:31.950102: step: 308/529, loss: 0.015028762631118298 2023-01-21 12:04:33.117157: step: 312/529, loss: 0.036978576332330704 2023-01-21 12:04:34.301773: step: 316/529, loss: 0.006365966983139515 2023-01-21 12:04:35.560721: step: 320/529, loss: 0.004104995634406805 2023-01-21 12:04:36.744411: step: 324/529, loss: 0.009681510739028454 2023-01-21 12:04:37.902601: step: 328/529, loss: 0.0002933502255473286 2023-01-21 12:04:39.101305: step: 332/529, loss: 0.0018339157104492188 2023-01-21 12:04:40.270147: step: 336/529, loss: 0.0024933815002441406 2023-01-21 12:04:41.499388: step: 340/529, loss: 0.014305305667221546 2023-01-21 12:04:42.730296: step: 344/529, loss: 0.007672500796616077 2023-01-21 12:04:43.909566: step: 348/529, loss: 9.870529174804688e-05 2023-01-21 12:04:45.082821: step: 352/529, loss: 0.0012907981872558594 2023-01-21 12:04:46.306973: step: 356/529, loss: 0.000255584716796875 2023-01-21 12:04:47.515396: step: 360/529, loss: 0.010764884762465954 2023-01-21 12:04:48.703167: step: 364/529, loss: 0.01815338060259819 2023-01-21 12:04:49.907144: step: 368/529, loss: 0.06235694885253906 2023-01-21 12:04:51.157007: step: 372/529, loss: 0.011397361755371094 2023-01-21 12:04:52.323312: step: 376/529, loss: 0.00359687814489007 2023-01-21 12:04:53.542690: step: 380/529, loss: 0.05834845080971718 2023-01-21 12:04:54.707388: step: 384/529, loss: 0.001628875732421875 2023-01-21 12:04:55.911503: step: 388/529, loss: 0.013346291147172451 2023-01-21 12:04:57.170356: step: 392/529, loss: 0.09892668575048447 2023-01-21 12:04:58.427027: step: 396/529, loss: 0.0017574310768395662 2023-01-21 12:04:59.592475: step: 400/529, loss: 0.0038386343512684107 2023-01-21 12:05:00.812175: step: 404/529, loss: 0.003656578017398715 2023-01-21 12:05:02.013809: step: 408/529, loss: 0.0422358512878418 2023-01-21 12:05:03.225408: step: 412/529, loss: 0.008557033725082874 2023-01-21 12:05:04.414096: step: 416/529, loss: 0.0542333610355854 2023-01-21 12:05:05.630740: step: 420/529, loss: 0.028049660846590996 2023-01-21 12:05:06.824481: step: 424/529, loss: 0.0005784034729003906 2023-01-21 12:05:07.999066: step: 428/529, loss: 0.0033512115478515625 2023-01-21 12:05:09.191106: step: 432/529, loss: 0.055254556238651276 2023-01-21 12:05:10.392804: step: 436/529, loss: 0.0023679733276367188 2023-01-21 12:05:11.574495: step: 440/529, loss: 0.0010782241588458419 2023-01-21 12:05:12.770630: step: 444/529, loss: 8.7738037109375e-05 2023-01-21 12:05:13.968337: step: 448/529, loss: 0.006907558999955654 2023-01-21 12:05:15.157991: step: 452/529, loss: 0.00539774913340807 2023-01-21 12:05:16.393312: step: 456/529, loss: 0.01805396005511284 2023-01-21 12:05:17.662030: step: 460/529, loss: 0.060446739196777344 2023-01-21 12:05:18.819541: step: 464/529, loss: 0.0008045196300372481 2023-01-21 12:05:19.987459: step: 468/529, loss: 0.13915549218654633 2023-01-21 12:05:21.152264: step: 472/529, loss: 0.025353431701660156 2023-01-21 12:05:22.318212: step: 476/529, loss: 0.018467998132109642 2023-01-21 12:05:23.493765: step: 480/529, loss: 0.014942443929612637 2023-01-21 12:05:24.714106: step: 484/529, loss: 0.015442563220858574 2023-01-21 12:05:25.896354: step: 488/529, loss: 0.00036154090776108205 2023-01-21 12:05:27.077045: step: 492/529, loss: 2.0790101189049892e-05 2023-01-21 12:05:28.247548: step: 496/529, loss: 0.000743865966796875 2023-01-21 12:05:29.425413: step: 500/529, loss: 0.00969996489584446 2023-01-21 12:05:30.609870: step: 504/529, loss: 6.122589547885582e-05 2023-01-21 12:05:31.793174: step: 508/529, loss: 0.00043911935063079 2023-01-21 12:05:33.010092: step: 512/529, loss: 0.025482559576630592 2023-01-21 12:05:34.189874: step: 516/529, loss: 0.015529823489487171 2023-01-21 12:05:35.400172: step: 520/529, loss: 0.007698011584579945 2023-01-21 12:05:36.593725: step: 524/529, loss: 0.005531501956284046 2023-01-21 12:05:37.763951: step: 528/529, loss: 0.04183664545416832 2023-01-21 12:05:38.961117: step: 532/529, loss: 0.002524757292121649 2023-01-21 12:05:40.138705: step: 536/529, loss: 0.0004337310965638608 2023-01-21 12:05:41.349842: step: 540/529, loss: 0.027965927496552467 2023-01-21 12:05:42.523145: step: 544/529, loss: 0.01026067789644003 2023-01-21 12:05:43.742579: step: 548/529, loss: 0.003141164779663086 2023-01-21 12:05:44.910088: step: 552/529, loss: 0.055869486182928085 2023-01-21 12:05:46.114676: step: 556/529, loss: 0.0028190254233777523 2023-01-21 12:05:47.299626: step: 560/529, loss: 0.00136909494176507 2023-01-21 12:05:48.485471: step: 564/529, loss: 0.0025454522110521793 2023-01-21 12:05:49.639823: step: 568/529, loss: 0.008861064910888672 2023-01-21 12:05:50.811360: step: 572/529, loss: 0.007780647370964289 2023-01-21 12:05:52.024508: step: 576/529, loss: 0.003621101612225175 2023-01-21 12:05:53.167964: step: 580/529, loss: 0.0003100395551882684 2023-01-21 12:05:54.403867: step: 584/529, loss: 0.005069160368293524 2023-01-21 12:05:55.613001: step: 588/529, loss: 0.0027904033195227385 2023-01-21 12:05:56.803923: step: 592/529, loss: 0.002065372420474887 2023-01-21 12:05:57.995165: step: 596/529, loss: 0.013754081912338734 2023-01-21 12:05:59.186991: step: 600/529, loss: 0.0006410598871298134 2023-01-21 12:06:00.391197: step: 604/529, loss: 0.007453632541000843 2023-01-21 12:06:01.568406: step: 608/529, loss: 0.43801918625831604 2023-01-21 12:06:02.737638: step: 612/529, loss: 0.012390327639877796 2023-01-21 12:06:03.976975: step: 616/529, loss: 0.09776792675256729 2023-01-21 12:06:05.159446: step: 620/529, loss: 0.008025741204619408 2023-01-21 12:06:06.356316: step: 624/529, loss: 0.01973114162683487 2023-01-21 12:06:07.569391: step: 628/529, loss: 0.010404204949736595 2023-01-21 12:06:08.763183: step: 632/529, loss: 0.0007291794172488153 2023-01-21 12:06:09.962311: step: 636/529, loss: 0.0015153884887695312 2023-01-21 12:06:11.120992: step: 640/529, loss: 0.0011337280739098787 2023-01-21 12:06:12.296602: step: 644/529, loss: 7.243156142067164e-05 2023-01-21 12:06:13.456047: step: 648/529, loss: 0.008921433240175247 2023-01-21 12:06:14.612132: step: 652/529, loss: 9.346007573185489e-06 2023-01-21 12:06:15.765286: step: 656/529, loss: 0.0009045600891113281 2023-01-21 12:06:16.932618: step: 660/529, loss: 0.004174423404037952 2023-01-21 12:06:18.119469: step: 664/529, loss: 0.008622074499726295 2023-01-21 12:06:19.346890: step: 668/529, loss: 0.0034662247635424137 2023-01-21 12:06:20.590509: step: 672/529, loss: 0.015787601470947266 2023-01-21 12:06:21.790495: step: 676/529, loss: 0.00700383260846138 2023-01-21 12:06:22.950317: step: 680/529, loss: 0.06572704762220383 2023-01-21 12:06:24.116637: step: 684/529, loss: 0.035642433911561966 2023-01-21 12:06:25.288118: step: 688/529, loss: 0.004046916961669922 2023-01-21 12:06:26.513031: step: 692/529, loss: 0.021069765090942383 2023-01-21 12:06:27.707108: step: 696/529, loss: 5.4550167988054454e-05 2023-01-21 12:06:28.945775: step: 700/529, loss: 0.004474544897675514 2023-01-21 12:06:30.117773: step: 704/529, loss: 0.037020303308963776 2023-01-21 12:06:31.308000: step: 708/529, loss: 0.04548950493335724 2023-01-21 12:06:32.501634: step: 712/529, loss: 0.02918701246380806 2023-01-21 12:06:33.695375: step: 716/529, loss: 0.003761101048439741 2023-01-21 12:06:34.885315: step: 720/529, loss: 0.0030232430435717106 2023-01-21 12:06:36.088007: step: 724/529, loss: 0.013779353350400925 2023-01-21 12:06:37.233883: step: 728/529, loss: 0.04705391079187393 2023-01-21 12:06:38.437151: step: 732/529, loss: 0.012912321835756302 2023-01-21 12:06:39.620433: step: 736/529, loss: 0.6670275926589966 2023-01-21 12:06:40.799456: step: 740/529, loss: 0.051375675946474075 2023-01-21 12:06:41.999270: step: 744/529, loss: 0.015016174875199795 2023-01-21 12:06:43.200514: step: 748/529, loss: 0.012148857116699219 2023-01-21 12:06:44.435441: step: 752/529, loss: 0.057822417467832565 2023-01-21 12:06:45.676045: step: 756/529, loss: 0.028007127344608307 2023-01-21 12:06:46.820067: step: 760/529, loss: 0.0005052089691162109 2023-01-21 12:06:48.025330: step: 764/529, loss: 0.002269458957016468 2023-01-21 12:06:49.233479: step: 768/529, loss: 6.842613220214844e-05 2023-01-21 12:06:50.401089: step: 772/529, loss: 0.0006751060136593878 2023-01-21 12:06:51.629827: step: 776/529, loss: 0.0009168624528683722 2023-01-21 12:06:52.799296: step: 780/529, loss: 0.007680845446884632 2023-01-21 12:06:53.950465: step: 784/529, loss: 0.004424571990966797 2023-01-21 12:06:55.158991: step: 788/529, loss: 0.006457233801484108 2023-01-21 12:06:56.363718: step: 792/529, loss: 0.0023097037337720394 2023-01-21 12:06:57.566736: step: 796/529, loss: 0.014001274481415749 2023-01-21 12:06:58.742360: step: 800/529, loss: 0.001825094223022461 2023-01-21 12:06:59.938437: step: 804/529, loss: 0.15867634117603302 2023-01-21 12:07:01.193086: step: 808/529, loss: 0.006685495376586914 2023-01-21 12:07:02.391806: step: 812/529, loss: 0.34852904081344604 2023-01-21 12:07:03.638201: step: 816/529, loss: 0.009759140200912952 2023-01-21 12:07:04.810190: step: 820/529, loss: 0.007753563113510609 2023-01-21 12:07:06.000934: step: 824/529, loss: 0.004956340882927179 2023-01-21 12:07:07.198001: step: 828/529, loss: 0.015973662957549095 2023-01-21 12:07:08.361019: step: 832/529, loss: 0.002685642335563898 2023-01-21 12:07:09.505608: step: 836/529, loss: 0.0029163360595703125 2023-01-21 12:07:10.703760: step: 840/529, loss: 0.00093669886700809 2023-01-21 12:07:11.897845: step: 844/529, loss: 0.03747301176190376 2023-01-21 12:07:13.105974: step: 848/529, loss: 0.021499251946806908 2023-01-21 12:07:14.268085: step: 852/529, loss: 0.000241184257902205 2023-01-21 12:07:15.455466: step: 856/529, loss: 0.03581847995519638 2023-01-21 12:07:16.597505: step: 860/529, loss: 0.0035420418716967106 2023-01-21 12:07:17.821287: step: 864/529, loss: 0.024512864649295807 2023-01-21 12:07:18.995143: step: 868/529, loss: 0.006778290029615164 2023-01-21 12:07:20.125118: step: 872/529, loss: 0.0019012452103197575 2023-01-21 12:07:21.295901: step: 876/529, loss: 0.0024512293748557568 2023-01-21 12:07:22.500030: step: 880/529, loss: 0.001793003175407648 2023-01-21 12:07:23.704806: step: 884/529, loss: 0.05764207988977432 2023-01-21 12:07:24.917714: step: 888/529, loss: 0.01868429221212864 2023-01-21 12:07:26.107563: step: 892/529, loss: 0.00026149750920012593 2023-01-21 12:07:27.286410: step: 896/529, loss: 0.0016706467140465975 2023-01-21 12:07:28.434026: step: 900/529, loss: 0.0296328067779541 2023-01-21 12:07:29.619338: step: 904/529, loss: 0.034564971923828125 2023-01-21 12:07:30.790140: step: 908/529, loss: 0.04131045192480087 2023-01-21 12:07:31.957864: step: 912/529, loss: 0.009833813644945621 2023-01-21 12:07:33.108515: step: 916/529, loss: 0.019705582410097122 2023-01-21 12:07:34.319926: step: 920/529, loss: 0.013294219970703125 2023-01-21 12:07:35.512834: step: 924/529, loss: 0.04758806526660919 2023-01-21 12:07:36.725824: step: 928/529, loss: 0.22303810715675354 2023-01-21 12:07:37.906892: step: 932/529, loss: 0.07901981472969055 2023-01-21 12:07:39.078758: step: 936/529, loss: 0.0025468827225267887 2023-01-21 12:07:40.273870: step: 940/529, loss: 0.0006996154552325606 2023-01-21 12:07:41.514782: step: 944/529, loss: 0.028627874329686165 2023-01-21 12:07:42.712205: step: 948/529, loss: 0.004445123486220837 2023-01-21 12:07:43.909545: step: 952/529, loss: 0.013493537902832031 2023-01-21 12:07:45.070209: step: 956/529, loss: 0.019879436120390892 2023-01-21 12:07:46.286696: step: 960/529, loss: 0.05215025320649147 2023-01-21 12:07:47.497251: step: 964/529, loss: 0.000286102294921875 2023-01-21 12:07:48.719969: step: 968/529, loss: 0.03295727074146271 2023-01-21 12:07:49.938739: step: 972/529, loss: 0.005147552117705345 2023-01-21 12:07:51.111419: step: 976/529, loss: 0.00522613525390625 2023-01-21 12:07:52.284364: step: 980/529, loss: 0.0008656025165691972 2023-01-21 12:07:53.469786: step: 984/529, loss: 0.00034236907958984375 2023-01-21 12:07:54.665038: step: 988/529, loss: 0.024534225463867188 2023-01-21 12:07:55.828512: step: 992/529, loss: 0.01809110678732395 2023-01-21 12:07:57.017218: step: 996/529, loss: 0.01804046705365181 2023-01-21 12:07:58.217719: step: 1000/529, loss: 0.02171487733721733 2023-01-21 12:07:59.414649: step: 1004/529, loss: 0.04370298236608505 2023-01-21 12:08:00.616752: step: 1008/529, loss: 0.022652246057987213 2023-01-21 12:08:01.803517: step: 1012/529, loss: 0.00015854835510253906 2023-01-21 12:08:03.026903: step: 1016/529, loss: 0.010126590728759766 2023-01-21 12:08:04.212519: step: 1020/529, loss: 0.02750396728515625 2023-01-21 12:08:05.369101: step: 1024/529, loss: 0.0030849457252770662 2023-01-21 12:08:06.574644: step: 1028/529, loss: 0.021029474213719368 2023-01-21 12:08:07.777034: step: 1032/529, loss: 0.0019851208198815584 2023-01-21 12:08:09.016202: step: 1036/529, loss: 0.0017242431640625 2023-01-21 12:08:10.231111: step: 1040/529, loss: 0.05150318518280983 2023-01-21 12:08:11.410126: step: 1044/529, loss: 0.046482231467962265 2023-01-21 12:08:12.589535: step: 1048/529, loss: 0.03938732296228409 2023-01-21 12:08:13.789735: step: 1052/529, loss: 0.004544639959931374 2023-01-21 12:08:14.998946: step: 1056/529, loss: 0.012930489145219326 2023-01-21 12:08:16.185131: step: 1060/529, loss: 0.06606350094079971 2023-01-21 12:08:17.378543: step: 1064/529, loss: 0.13102301955223083 2023-01-21 12:08:18.553330: step: 1068/529, loss: 0.018883515149354935 2023-01-21 12:08:19.735430: step: 1072/529, loss: 0.00043230055598542094 2023-01-21 12:08:20.938148: step: 1076/529, loss: 0.061090562492609024 2023-01-21 12:08:22.131929: step: 1080/529, loss: 0.01868267171084881 2023-01-21 12:08:23.308677: step: 1084/529, loss: 0.02054271660745144 2023-01-21 12:08:24.486571: step: 1088/529, loss: 0.014475250616669655 2023-01-21 12:08:25.674309: step: 1092/529, loss: 0.013321589678525925 2023-01-21 12:08:26.869308: step: 1096/529, loss: 0.0029973508790135384 2023-01-21 12:08:28.056520: step: 1100/529, loss: 0.05721712112426758 2023-01-21 12:08:29.259767: step: 1104/529, loss: 0.0016152382595464587 2023-01-21 12:08:30.491625: step: 1108/529, loss: 0.007339096628129482 2023-01-21 12:08:31.672914: step: 1112/529, loss: 0.01019744947552681 2023-01-21 12:08:32.873155: step: 1116/529, loss: 0.016485881060361862 2023-01-21 12:08:34.025300: step: 1120/529, loss: 0.004796076100319624 2023-01-21 12:08:35.237280: step: 1124/529, loss: 0.0078108785673975945 2023-01-21 12:08:36.449391: step: 1128/529, loss: 0.005593681707978249 2023-01-21 12:08:37.645941: step: 1132/529, loss: 0.04302167892456055 2023-01-21 12:08:38.829136: step: 1136/529, loss: 0.0029899119399487972 2023-01-21 12:08:39.999336: step: 1140/529, loss: 0.0039408206939697266 2023-01-21 12:08:41.194330: step: 1144/529, loss: 0.015243053436279297 2023-01-21 12:08:42.388872: step: 1148/529, loss: 0.008402634412050247 2023-01-21 12:08:43.612270: step: 1152/529, loss: 0.0356656089425087 2023-01-21 12:08:44.768271: step: 1156/529, loss: 0.02371673658490181 2023-01-21 12:08:45.989805: step: 1160/529, loss: 0.004023551940917969 2023-01-21 12:08:47.176625: step: 1164/529, loss: 0.0035541534889489412 2023-01-21 12:08:48.437253: step: 1168/529, loss: 0.0019771575462073088 2023-01-21 12:08:49.626245: step: 1172/529, loss: 0.012814903631806374 2023-01-21 12:08:50.825028: step: 1176/529, loss: 0.00015735626220703125 2023-01-21 12:08:52.006482: step: 1180/529, loss: 0.005934262182563543 2023-01-21 12:08:53.167592: step: 1184/529, loss: 4.7492983867414296e-05 2023-01-21 12:08:54.405427: step: 1188/529, loss: 0.000568389892578125 2023-01-21 12:08:55.562043: step: 1192/529, loss: 0.0047317505814135075 2023-01-21 12:08:56.698079: step: 1196/529, loss: 0.0015502930618822575 2023-01-21 12:08:57.901278: step: 1200/529, loss: 0.04354400932788849 2023-01-21 12:08:59.075269: step: 1204/529, loss: 0.0015049935318529606 2023-01-21 12:09:00.251552: step: 1208/529, loss: 0.01749744452536106 2023-01-21 12:09:01.441020: step: 1212/529, loss: 0.1285255402326584 2023-01-21 12:09:02.614985: step: 1216/529, loss: 0.00011482239642646164 2023-01-21 12:09:03.762925: step: 1220/529, loss: 0.007799911312758923 2023-01-21 12:09:04.932419: step: 1224/529, loss: 0.00016622543625999242 2023-01-21 12:09:06.054956: step: 1228/529, loss: 0.0008256912115029991 2023-01-21 12:09:07.226982: step: 1232/529, loss: 0.008492779918015003 2023-01-21 12:09:08.437742: step: 1236/529, loss: 0.0019851685501635075 2023-01-21 12:09:09.600298: step: 1240/529, loss: 0.060060691088438034 2023-01-21 12:09:10.828405: step: 1244/529, loss: 0.0013676644302904606 2023-01-21 12:09:12.039469: step: 1248/529, loss: 0.03038635291159153 2023-01-21 12:09:13.276617: step: 1252/529, loss: 0.046868182718753815 2023-01-21 12:09:14.481620: step: 1256/529, loss: 0.046543315052986145 2023-01-21 12:09:15.730518: step: 1260/529, loss: 0.00159120571333915 2023-01-21 12:09:16.911926: step: 1264/529, loss: 0.0002974033122882247 2023-01-21 12:09:18.120476: step: 1268/529, loss: 0.009661865420639515 2023-01-21 12:09:19.335957: step: 1272/529, loss: 0.028136014938354492 2023-01-21 12:09:20.502344: step: 1276/529, loss: 0.00021605491929221898 2023-01-21 12:09:21.715255: step: 1280/529, loss: 0.009549331851303577 2023-01-21 12:09:22.927300: step: 1284/529, loss: 0.009549617767333984 2023-01-21 12:09:24.091739: step: 1288/529, loss: 0.00022563934908248484 2023-01-21 12:09:25.340287: step: 1292/529, loss: 0.006129646208137274 2023-01-21 12:09:26.531443: step: 1296/529, loss: 4.2343137465650216e-05 2023-01-21 12:09:27.724598: step: 1300/529, loss: 0.0018351555336266756 2023-01-21 12:09:28.949949: step: 1304/529, loss: 0.001064300537109375 2023-01-21 12:09:30.172469: step: 1308/529, loss: 0.047364044934511185 2023-01-21 12:09:31.347499: step: 1312/529, loss: 0.027245236560702324 2023-01-21 12:09:32.542402: step: 1316/529, loss: 0.04909687116742134 2023-01-21 12:09:33.735255: step: 1320/529, loss: 0.008952331729233265 2023-01-21 12:09:34.957859: step: 1324/529, loss: 0.008603191934525967 2023-01-21 12:09:36.176665: step: 1328/529, loss: 0.006449508480727673 2023-01-21 12:09:37.399329: step: 1332/529, loss: 0.007208824623376131 2023-01-21 12:09:38.624721: step: 1336/529, loss: 0.0018554687267169356 2023-01-21 12:09:39.816723: step: 1340/529, loss: 0.013210153207182884 2023-01-21 12:09:41.019851: step: 1344/529, loss: 0.03131254017353058 2023-01-21 12:09:42.244742: step: 1348/529, loss: 0.021416282281279564 2023-01-21 12:09:43.446608: step: 1352/529, loss: 0.021112823858857155 2023-01-21 12:09:44.641265: step: 1356/529, loss: 0.024181555956602097 2023-01-21 12:09:45.840206: step: 1360/529, loss: 0.0011964797740802169 2023-01-21 12:09:47.041065: step: 1364/529, loss: 0.0003619194030761719 2023-01-21 12:09:48.222841: step: 1368/529, loss: 0.03746175765991211 2023-01-21 12:09:49.445805: step: 1372/529, loss: 0.002320241881534457 2023-01-21 12:09:50.619407: step: 1376/529, loss: 0.03711114078760147 2023-01-21 12:09:51.822056: step: 1380/529, loss: 0.03472557291388512 2023-01-21 12:09:53.001007: step: 1384/529, loss: 0.004736900329589844 2023-01-21 12:09:54.227898: step: 1388/529, loss: 0.0034609795548021793 2023-01-21 12:09:55.470652: step: 1392/529, loss: 0.0391940101981163 2023-01-21 12:09:56.659630: step: 1396/529, loss: 0.006446361541748047 2023-01-21 12:09:57.873034: step: 1400/529, loss: 0.0011011123424395919 2023-01-21 12:09:59.100105: step: 1404/529, loss: 0.02707061730325222 2023-01-21 12:10:00.295521: step: 1408/529, loss: 0.008593464270234108 2023-01-21 12:10:01.532087: step: 1412/529, loss: 0.005849075503647327 2023-01-21 12:10:02.692643: step: 1416/529, loss: 0.008972168900072575 2023-01-21 12:10:03.863927: step: 1420/529, loss: 0.047736648470163345 2023-01-21 12:10:05.099533: step: 1424/529, loss: 0.0008117676479741931 2023-01-21 12:10:06.305696: step: 1428/529, loss: 0.026935484260320663 2023-01-21 12:10:07.506037: step: 1432/529, loss: 0.041068557649850845 2023-01-21 12:10:08.691301: step: 1436/529, loss: 0.00026874541072174907 2023-01-21 12:10:09.854027: step: 1440/529, loss: 0.031628336757421494 2023-01-21 12:10:11.021002: step: 1444/529, loss: 0.006427955813705921 2023-01-21 12:10:12.243421: step: 1448/529, loss: 0.005690575111657381 2023-01-21 12:10:13.422451: step: 1452/529, loss: 0.0006567955133505166 2023-01-21 12:10:14.599655: step: 1456/529, loss: 0.0030237198807299137 2023-01-21 12:10:15.737297: step: 1460/529, loss: 0.0068683624267578125 2023-01-21 12:10:16.952553: step: 1464/529, loss: 0.0023954391945153475 2023-01-21 12:10:18.148567: step: 1468/529, loss: 0.003706264542415738 2023-01-21 12:10:19.340301: step: 1472/529, loss: 0.010748482309281826 2023-01-21 12:10:20.581027: step: 1476/529, loss: 0.018119478598237038 2023-01-21 12:10:21.724763: step: 1480/529, loss: 0.010569977574050426 2023-01-21 12:10:22.911476: step: 1484/529, loss: 0.04724063724279404 2023-01-21 12:10:24.072586: step: 1488/529, loss: 1.792907642084174e-05 2023-01-21 12:10:25.242176: step: 1492/529, loss: 0.0736178383231163 2023-01-21 12:10:26.430747: step: 1496/529, loss: 0.00263385777361691 2023-01-21 12:10:27.618815: step: 1500/529, loss: 0.000102996826171875 2023-01-21 12:10:28.802502: step: 1504/529, loss: 0.08344592899084091 2023-01-21 12:10:29.974679: step: 1508/529, loss: 0.008185386657714844 2023-01-21 12:10:31.180970: step: 1512/529, loss: 0.02715597115457058 2023-01-21 12:10:32.361031: step: 1516/529, loss: 0.004849815275520086 2023-01-21 12:10:33.541468: step: 1520/529, loss: 0.005395602900534868 2023-01-21 12:10:34.723961: step: 1524/529, loss: 0.004115152172744274 2023-01-21 12:10:35.905374: step: 1528/529, loss: 0.0003519534948281944 2023-01-21 12:10:37.116881: step: 1532/529, loss: 0.3924638628959656 2023-01-21 12:10:38.294703: step: 1536/529, loss: 0.0459442175924778 2023-01-21 12:10:39.531720: step: 1540/529, loss: 0.2564008831977844 2023-01-21 12:10:40.730696: step: 1544/529, loss: 0.0105011947453022 2023-01-21 12:10:41.889883: step: 1548/529, loss: 0.0003972053527832031 2023-01-21 12:10:43.063867: step: 1552/529, loss: 0.0016611098544672132 2023-01-21 12:10:44.246999: step: 1556/529, loss: 0.0011737823951989412 2023-01-21 12:10:45.449374: step: 1560/529, loss: 0.009410095401108265 2023-01-21 12:10:46.659958: step: 1564/529, loss: 0.03652916103601456 2023-01-21 12:10:47.841492: step: 1568/529, loss: 0.019153403118252754 2023-01-21 12:10:49.004299: step: 1572/529, loss: 0.07477583736181259 2023-01-21 12:10:50.207036: step: 1576/529, loss: 0.03913898766040802 2023-01-21 12:10:51.395966: step: 1580/529, loss: 0.008938598446547985 2023-01-21 12:10:52.587126: step: 1584/529, loss: 0.0014506339794024825 2023-01-21 12:10:53.772020: step: 1588/529, loss: 9.212493750965223e-05 2023-01-21 12:10:54.936765: step: 1592/529, loss: 0.0002627372741699219 2023-01-21 12:10:56.147177: step: 1596/529, loss: 0.036153409630060196 2023-01-21 12:10:57.352861: step: 1600/529, loss: 0.024897100403904915 2023-01-21 12:10:58.534601: step: 1604/529, loss: 0.0018195391166955233 2023-01-21 12:10:59.695650: step: 1608/529, loss: 0.0006594657897949219 2023-01-21 12:11:00.904039: step: 1612/529, loss: 0.018125344067811966 2023-01-21 12:11:02.109508: step: 1616/529, loss: 0.023831749334931374 2023-01-21 12:11:03.301695: step: 1620/529, loss: 0.7519766688346863 2023-01-21 12:11:04.471401: step: 1624/529, loss: 0.012073611840605736 2023-01-21 12:11:05.677538: step: 1628/529, loss: 0.021049117669463158 2023-01-21 12:11:06.875649: step: 1632/529, loss: 3.356933666509576e-05 2023-01-21 12:11:08.067458: step: 1636/529, loss: 0.012299228459596634 2023-01-21 12:11:09.281567: step: 1640/529, loss: 0.03125343099236488 2023-01-21 12:11:10.474451: step: 1644/529, loss: 0.03994322195649147 2023-01-21 12:11:11.672088: step: 1648/529, loss: 0.010166454128921032 2023-01-21 12:11:12.842010: step: 1652/529, loss: 0.011269474402070045 2023-01-21 12:11:14.019338: step: 1656/529, loss: 0.004206943325698376 2023-01-21 12:11:15.279859: step: 1660/529, loss: 0.015709497034549713 2023-01-21 12:11:16.478382: step: 1664/529, loss: 0.08432827144861221 2023-01-21 12:11:17.674563: step: 1668/529, loss: 0.03844795376062393 2023-01-21 12:11:18.860850: step: 1672/529, loss: 0.007732391357421875 2023-01-21 12:11:20.057910: step: 1676/529, loss: 0.0016179084777832031 2023-01-21 12:11:21.256314: step: 1680/529, loss: 0.009468006901443005 2023-01-21 12:11:22.438798: step: 1684/529, loss: 0.0014778138138353825 2023-01-21 12:11:23.759504: step: 1688/529, loss: 0.20578722655773163 2023-01-21 12:11:24.969225: step: 1692/529, loss: 0.01151266135275364 2023-01-21 12:11:26.182043: step: 1696/529, loss: 0.010944557376205921 2023-01-21 12:11:27.405474: step: 1700/529, loss: 0.05595417320728302 2023-01-21 12:11:28.581627: step: 1704/529, loss: 0.018660927191376686 2023-01-21 12:11:29.735943: step: 1708/529, loss: 0.0008911133045330644 2023-01-21 12:11:30.958945: step: 1712/529, loss: 0.06262359023094177 2023-01-21 12:11:32.144926: step: 1716/529, loss: 0.0032632825896143913 2023-01-21 12:11:33.351862: step: 1720/529, loss: 0.006718921475112438 2023-01-21 12:11:34.648248: step: 1724/529, loss: 0.0031825066544115543 2023-01-21 12:11:35.888416: step: 1728/529, loss: 0.030719662085175514 2023-01-21 12:11:37.080049: step: 1732/529, loss: 0.022855568677186966 2023-01-21 12:11:38.254997: step: 1736/529, loss: 0.011527299880981445 2023-01-21 12:11:39.436050: step: 1740/529, loss: 0.028424454852938652 2023-01-21 12:11:40.594266: step: 1744/529, loss: 0.00027675629826262593 2023-01-21 12:11:41.800778: step: 1748/529, loss: 0.011684417724609375 2023-01-21 12:11:43.046693: step: 1752/529, loss: 0.08835625648498535 2023-01-21 12:11:44.220030: step: 1756/529, loss: 0.08548679947853088 2023-01-21 12:11:45.389483: step: 1760/529, loss: 0.003707838011905551 2023-01-21 12:11:46.570525: step: 1764/529, loss: 0.0014142035506665707 2023-01-21 12:11:47.800832: step: 1768/529, loss: 0.010040379129350185 2023-01-21 12:11:49.024310: step: 1772/529, loss: 0.011329079046845436 2023-01-21 12:11:50.236472: step: 1776/529, loss: 0.009504318237304688 2023-01-21 12:11:51.430391: step: 1780/529, loss: 0.09586277604103088 2023-01-21 12:11:52.611877: step: 1784/529, loss: 0.044762708246707916 2023-01-21 12:11:53.815148: step: 1788/529, loss: 0.009194565005600452 2023-01-21 12:11:55.003809: step: 1792/529, loss: 0.019130611792206764 2023-01-21 12:11:56.230632: step: 1796/529, loss: 0.032532550394535065 2023-01-21 12:11:57.414600: step: 1800/529, loss: 0.002416324568912387 2023-01-21 12:11:58.593862: step: 1804/529, loss: 0.12907905876636505 2023-01-21 12:11:59.781158: step: 1808/529, loss: 0.001695060869678855 2023-01-21 12:12:00.990136: step: 1812/529, loss: 0.026753615587949753 2023-01-21 12:12:02.189713: step: 1816/529, loss: 0.044733572751283646 2023-01-21 12:12:03.412053: step: 1820/529, loss: 0.017317676916718483 2023-01-21 12:12:04.615626: step: 1824/529, loss: 0.040297508239746094 2023-01-21 12:12:05.789489: step: 1828/529, loss: 0.02081015147268772 2023-01-21 12:12:06.956433: step: 1832/529, loss: 0.00019874573627021164 2023-01-21 12:12:08.173962: step: 1836/529, loss: 0.015726901590824127 2023-01-21 12:12:09.383262: step: 1840/529, loss: 0.010347366333007812 2023-01-21 12:12:10.582598: step: 1844/529, loss: 0.00031410454539582133 2023-01-21 12:12:11.750504: step: 1848/529, loss: 0.012937640771269798 2023-01-21 12:12:12.954884: step: 1852/529, loss: 0.0008021831745281816 2023-01-21 12:12:14.143203: step: 1856/529, loss: 0.02051115222275257 2023-01-21 12:12:15.356400: step: 1860/529, loss: 0.020203400403261185 2023-01-21 12:12:16.517175: step: 1864/529, loss: 0.002439212752506137 2023-01-21 12:12:17.673801: step: 1868/529, loss: 0.0023755072616040707 2023-01-21 12:12:18.874434: step: 1872/529, loss: 0.016912128776311874 2023-01-21 12:12:20.084635: step: 1876/529, loss: 0.05541591718792915 2023-01-21 12:12:21.264645: step: 1880/529, loss: 0.010518074035644531 2023-01-21 12:12:22.495470: step: 1884/529, loss: 0.005601120181381702 2023-01-21 12:12:23.697860: step: 1888/529, loss: 0.00017547607421875 2023-01-21 12:12:24.884535: step: 1892/529, loss: 0.009705543518066406 2023-01-21 12:12:26.045711: step: 1896/529, loss: 1.3732910701946821e-05 2023-01-21 12:12:27.212560: step: 1900/529, loss: 0.0019033431308344007 2023-01-21 12:12:28.401459: step: 1904/529, loss: 0.03345794975757599 2023-01-21 12:12:29.588757: step: 1908/529, loss: 7.83920258982107e-05 2023-01-21 12:12:30.768202: step: 1912/529, loss: 0.0019386291969567537 2023-01-21 12:12:31.938680: step: 1916/529, loss: 0.017402220517396927 2023-01-21 12:12:33.106313: step: 1920/529, loss: 0.03869824483990669 2023-01-21 12:12:34.353903: step: 1924/529, loss: 0.0214520450681448 2023-01-21 12:12:35.566851: step: 1928/529, loss: 0.0009203911176882684 2023-01-21 12:12:36.775436: step: 1932/529, loss: 0.06093807518482208 2023-01-21 12:12:37.937247: step: 1936/529, loss: 0.0027781487442553043 2023-01-21 12:12:39.123775: step: 1940/529, loss: 0.0013230800395831466 2023-01-21 12:12:40.321876: step: 1944/529, loss: 0.00035896303597837687 2023-01-21 12:12:41.479323: step: 1948/529, loss: 0.049846649169921875 2023-01-21 12:12:42.656531: step: 1952/529, loss: 0.0010044098598882556 2023-01-21 12:12:43.888235: step: 1956/529, loss: 0.0958099439740181 2023-01-21 12:12:45.097465: step: 1960/529, loss: 0.00028228759765625 2023-01-21 12:12:46.272459: step: 1964/529, loss: 0.029911233112215996 2023-01-21 12:12:47.445421: step: 1968/529, loss: 0.0011051774490624666 2023-01-21 12:12:48.642988: step: 1972/529, loss: 0.0001562118559377268 2023-01-21 12:12:49.829894: step: 1976/529, loss: 0.006267261691391468 2023-01-21 12:12:51.026896: step: 1980/529, loss: 0.0007879257318563759 2023-01-21 12:12:52.229057: step: 1984/529, loss: 0.0077056884765625 2023-01-21 12:12:53.469515: step: 1988/529, loss: 0.003443432040512562 2023-01-21 12:12:54.666421: step: 1992/529, loss: 0.02061910554766655 2023-01-21 12:12:55.951228: step: 1996/529, loss: 0.00135040283203125 2023-01-21 12:12:57.148684: step: 2000/529, loss: 0.04771852120757103 2023-01-21 12:12:58.367933: step: 2004/529, loss: 0.007236003875732422 2023-01-21 12:12:59.537541: step: 2008/529, loss: 0.0054528238251805305 2023-01-21 12:13:00.681632: step: 2012/529, loss: 0.03294019773602486 2023-01-21 12:13:01.866069: step: 2016/529, loss: 9.13620024221018e-05 2023-01-21 12:13:03.053687: step: 2020/529, loss: 0.010421467013657093 2023-01-21 12:13:04.226906: step: 2024/529, loss: 0.020175648853182793 2023-01-21 12:13:05.424091: step: 2028/529, loss: 0.01042032241821289 2023-01-21 12:13:06.620529: step: 2032/529, loss: 0.014084530994296074 2023-01-21 12:13:07.802723: step: 2036/529, loss: 0.02331419102847576 2023-01-21 12:13:09.034771: step: 2040/529, loss: 0.04138088598847389 2023-01-21 12:13:10.242545: step: 2044/529, loss: 0.007630777545273304 2023-01-21 12:13:11.439349: step: 2048/529, loss: 0.0072886464186012745 2023-01-21 12:13:12.639022: step: 2052/529, loss: 0.005963707342743874 2023-01-21 12:13:13.821991: step: 2056/529, loss: 0.024396324530243874 2023-01-21 12:13:14.976119: step: 2060/529, loss: 0.023745249956846237 2023-01-21 12:13:16.179953: step: 2064/529, loss: 0.005225372035056353 2023-01-21 12:13:17.324650: step: 2068/529, loss: 0.01992173306643963 2023-01-21 12:13:18.527396: step: 2072/529, loss: 0.3842557966709137 2023-01-21 12:13:19.686243: step: 2076/529, loss: 0.000415802001953125 2023-01-21 12:13:20.887810: step: 2080/529, loss: 0.039803698658943176 2023-01-21 12:13:22.116000: step: 2084/529, loss: 0.0005195617559365928 2023-01-21 12:13:23.301030: step: 2088/529, loss: 0.0003741264226846397 2023-01-21 12:13:24.506329: step: 2092/529, loss: 0.03693675994873047 2023-01-21 12:13:25.706946: step: 2096/529, loss: 0.005920505616813898 2023-01-21 12:13:26.896539: step: 2100/529, loss: 0.029214000329375267 2023-01-21 12:13:28.092420: step: 2104/529, loss: 0.0033657075837254524 2023-01-21 12:13:29.295448: step: 2108/529, loss: 0.04295841604471207 2023-01-21 12:13:30.524257: step: 2112/529, loss: 0.04494164139032364 2023-01-21 12:13:31.714916: step: 2116/529, loss: 0.0019300461281090975 ================================================== Loss: 0.025 -------------------- Dev: {'event': {'p': 0.6002066115702479, 'r': 0.7736351531291611, 'f1': 0.6759744037230948}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Test: {'event': {'p': 0.6511056511056511, 'r': 0.7938885560215698, 'f1': 0.7154427645788337}, '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': 0.5636363636363636, 'r': 0.49206349206349204, 'f1': 0.5254237288135594}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} 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': 18} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** Epoch: 19 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:14:14.733084: step: 4/529, loss: 0.042935751378536224 2023-01-21 12:14:15.878601: step: 8/529, loss: 0.004815959837287664 2023-01-21 12:14:17.036477: step: 12/529, loss: 0.005259990692138672 2023-01-21 12:14:18.283154: step: 16/529, loss: 0.6996700167655945 2023-01-21 12:14:19.481493: step: 20/529, loss: 0.06484298408031464 2023-01-21 12:14:20.659404: step: 24/529, loss: 0.0007749557844363153 2023-01-21 12:14:21.828861: step: 28/529, loss: 0.017015792429447174 2023-01-21 12:14:23.016884: step: 32/529, loss: 0.0003442764573264867 2023-01-21 12:14:24.183519: step: 36/529, loss: 0.002808952471241355 2023-01-21 12:14:25.412722: step: 40/529, loss: 0.006525897886604071 2023-01-21 12:14:26.621974: step: 44/529, loss: 0.028618622571229935 2023-01-21 12:14:27.799675: step: 48/529, loss: 0.0017216682899743319 2023-01-21 12:14:28.989526: step: 52/529, loss: 0.019769860431551933 2023-01-21 12:14:30.219520: step: 56/529, loss: 3.2424923119833693e-06 2023-01-21 12:14:31.383171: step: 60/529, loss: 0.0063193319365382195 2023-01-21 12:14:32.609772: step: 64/529, loss: 0.00045680999755859375 2023-01-21 12:14:33.800172: step: 68/529, loss: 0.0002552986261434853 2023-01-21 12:14:34.963596: step: 72/529, loss: 0.06339512020349503 2023-01-21 12:14:36.128023: step: 76/529, loss: 0.006579685024917126 2023-01-21 12:14:37.278766: step: 80/529, loss: 0.00028629304142668843 2023-01-21 12:14:38.491123: step: 84/529, loss: 0.0038661002181470394 2023-01-21 12:14:39.684500: step: 88/529, loss: 0.0009481430170126259 2023-01-21 12:14:40.878049: step: 92/529, loss: 0.0023017884232103825 2023-01-21 12:14:42.031664: step: 96/529, loss: 0.0019096970790997148 2023-01-21 12:14:43.200470: step: 100/529, loss: 0.00011987685866188258 2023-01-21 12:14:44.382228: step: 104/529, loss: 0.009096241556107998 2023-01-21 12:14:45.584911: step: 108/529, loss: 0.0023687363136559725 2023-01-21 12:14:46.814480: step: 112/529, loss: 0.015666484832763672 2023-01-21 12:14:48.001512: step: 116/529, loss: 0.010844516567885876 2023-01-21 12:14:49.213688: step: 120/529, loss: 0.00016021728515625 2023-01-21 12:14:50.402453: step: 124/529, loss: 0.00017271040997002274 2023-01-21 12:14:51.643487: step: 128/529, loss: 0.014075851067900658 2023-01-21 12:14:52.857811: step: 132/529, loss: 0.0013660431141033769 2023-01-21 12:14:54.035525: step: 136/529, loss: 1.9168852304574102e-05 2023-01-21 12:14:55.216275: step: 140/529, loss: 0.0013269424671307206 2023-01-21 12:14:56.441498: step: 144/529, loss: 0.005456352140754461 2023-01-21 12:14:57.651119: step: 148/529, loss: 0.001750087714754045 2023-01-21 12:14:58.831972: step: 152/529, loss: 0.026131058111786842 2023-01-21 12:15:00.060101: step: 156/529, loss: 0.04542860761284828 2023-01-21 12:15:01.290554: step: 160/529, loss: 4.386902219266631e-06 2023-01-21 12:15:02.526243: step: 164/529, loss: 0.009830665774643421 2023-01-21 12:15:03.722661: step: 168/529, loss: 0.0005489349132403731 2023-01-21 12:15:04.919877: step: 172/529, loss: 0.007158470340073109 2023-01-21 12:15:06.117615: step: 176/529, loss: 0.008014869876205921 2023-01-21 12:15:07.316552: step: 180/529, loss: 0.035913754254579544 2023-01-21 12:15:08.509763: step: 184/529, loss: 0.02865428850054741 2023-01-21 12:15:09.688832: step: 188/529, loss: 0.10077238082885742 2023-01-21 12:15:10.859866: step: 192/529, loss: 0.004261970520019531 2023-01-21 12:15:12.078486: step: 196/529, loss: 0.000240325927734375 2023-01-21 12:15:13.309278: step: 200/529, loss: 0.011649513617157936 2023-01-21 12:15:14.520648: step: 204/529, loss: 0.0006336212391033769 2023-01-21 12:15:15.679257: step: 208/529, loss: 0.01825389824807644 2023-01-21 12:15:16.895123: step: 212/529, loss: 0.0030647278763353825 2023-01-21 12:15:18.091358: step: 216/529, loss: 0.01799612119793892 2023-01-21 12:15:19.256928: step: 220/529, loss: 0.040863037109375 2023-01-21 12:15:20.451629: step: 224/529, loss: 0.01732940785586834 2023-01-21 12:15:21.634283: step: 228/529, loss: 0.007949638180434704 2023-01-21 12:15:22.832611: step: 232/529, loss: 0.025159548968076706 2023-01-21 12:15:24.041172: step: 236/529, loss: 0.028084373101592064 2023-01-21 12:15:25.210915: step: 240/529, loss: 0.009926891885697842 2023-01-21 12:15:26.392685: step: 244/529, loss: 0.004693221766501665 2023-01-21 12:15:27.542847: step: 248/529, loss: 0.007546487729996443 2023-01-21 12:15:28.715165: step: 252/529, loss: 0.00022850035747978836 2023-01-21 12:15:29.938472: step: 256/529, loss: 0.0015692710876464844 2023-01-21 12:15:31.108277: step: 260/529, loss: 0.021524859592318535 2023-01-21 12:15:32.318303: step: 264/529, loss: 0.0020837783813476562 2023-01-21 12:15:33.494853: step: 268/529, loss: 0.000560379063244909 2023-01-21 12:15:34.663327: step: 272/529, loss: 0.0011016845237463713 2023-01-21 12:15:35.838896: step: 276/529, loss: 0.0015305518172681332 2023-01-21 12:15:37.000879: step: 280/529, loss: 0.0006735801580362022 2023-01-21 12:15:38.181094: step: 284/529, loss: 0.0008599281427450478 2023-01-21 12:15:39.417399: step: 288/529, loss: 0.010336684994399548 2023-01-21 12:15:40.607360: step: 292/529, loss: 0.0006685256958007812 2023-01-21 12:15:41.790900: step: 296/529, loss: 0.0060256957076489925 2023-01-21 12:15:42.958485: step: 300/529, loss: 0.000815963780041784 2023-01-21 12:15:44.125655: step: 304/529, loss: 0.02645416185259819 2023-01-21 12:15:45.321807: step: 308/529, loss: 0.01834697835147381 2023-01-21 12:15:46.515524: step: 312/529, loss: 0.008624554611742496 2023-01-21 12:15:47.719555: step: 316/529, loss: 0.007229137234389782 2023-01-21 12:15:48.956184: step: 320/529, loss: 0.008791828528046608 2023-01-21 12:15:50.147281: step: 324/529, loss: 7.247925168485381e-06 2023-01-21 12:15:51.362007: step: 328/529, loss: 0.04197730869054794 2023-01-21 12:15:52.582847: step: 332/529, loss: 0.015083122067153454 2023-01-21 12:15:53.779889: step: 336/529, loss: 0.00303230294957757 2023-01-21 12:15:54.970573: step: 340/529, loss: 0.00012350082397460938 2023-01-21 12:15:56.183892: step: 344/529, loss: 0.000244140625 2023-01-21 12:15:57.368646: step: 348/529, loss: 0.0011779784690588713 2023-01-21 12:15:58.539008: step: 352/529, loss: 0.0009716034401208162 2023-01-21 12:15:59.719777: step: 356/529, loss: 0.024477768689393997 2023-01-21 12:16:00.910768: step: 360/529, loss: 0.005537224002182484 2023-01-21 12:16:02.107222: step: 364/529, loss: 0.0011005401611328125 2023-01-21 12:16:03.317587: step: 368/529, loss: 0.0023327828384935856 2023-01-21 12:16:04.553448: step: 372/529, loss: 0.012629889883100986 2023-01-21 12:16:05.748541: step: 376/529, loss: 0.000884628330823034 2023-01-21 12:16:06.946430: step: 380/529, loss: 0.0020687105134129524 2023-01-21 12:16:08.149187: step: 384/529, loss: 0.004502487368881702 2023-01-21 12:16:09.367243: step: 388/529, loss: 0.008827496320009232 2023-01-21 12:16:10.518611: step: 392/529, loss: 0.028681088238954544 2023-01-21 12:16:11.679610: step: 396/529, loss: 0.004202461335808039 2023-01-21 12:16:12.814203: step: 400/529, loss: 0.07386750727891922 2023-01-21 12:16:14.011087: step: 404/529, loss: 0.0651487410068512 2023-01-21 12:16:15.177184: step: 408/529, loss: 0.0022944926749914885 2023-01-21 12:16:16.287322: step: 412/529, loss: 0.015975380316376686 2023-01-21 12:16:17.505091: step: 416/529, loss: 0.004264259245246649 2023-01-21 12:16:18.677107: step: 420/529, loss: 0.0006412506336346269 2023-01-21 12:16:19.867540: step: 424/529, loss: 0.0010412216652184725 2023-01-21 12:16:21.035502: step: 428/529, loss: 0.00019407272338867188 2023-01-21 12:16:22.296266: step: 432/529, loss: 0.0026771544944494963 2023-01-21 12:16:23.493488: step: 436/529, loss: 0.0003685951232910156 2023-01-21 12:16:24.673018: step: 440/529, loss: 0.019832421094179153 2023-01-21 12:16:25.910978: step: 444/529, loss: 0.00048246385995298624 2023-01-21 12:16:27.094321: step: 448/529, loss: 0.00011801719665527344 2023-01-21 12:16:28.257079: step: 452/529, loss: 0.005002498626708984 2023-01-21 12:16:29.466424: step: 456/529, loss: 0.02692241594195366 2023-01-21 12:16:30.632379: step: 460/529, loss: 0.0001867294340627268 2023-01-21 12:16:31.868520: step: 464/529, loss: 0.02612285688519478 2023-01-21 12:16:33.097008: step: 468/529, loss: 0.007007598876953125 2023-01-21 12:16:34.287957: step: 472/529, loss: 0.003475189208984375 2023-01-21 12:16:35.509814: step: 476/529, loss: 0.012428474612534046 2023-01-21 12:16:36.687376: step: 480/529, loss: 0.0009718418004922569 2023-01-21 12:16:37.878805: step: 484/529, loss: 0.00010185241990257055 2023-01-21 12:16:39.052286: step: 488/529, loss: 0.000911331211682409 2023-01-21 12:16:40.205513: step: 492/529, loss: 0.00024662018404342234 2023-01-21 12:16:41.385500: step: 496/529, loss: 0.04441995546221733 2023-01-21 12:16:42.602128: step: 500/529, loss: 0.004452705383300781 2023-01-21 12:16:43.749830: step: 504/529, loss: 0.0008153438684530556 2023-01-21 12:16:44.917195: step: 508/529, loss: 0.018802309408783913 2023-01-21 12:16:46.100683: step: 512/529, loss: 0.004047775641083717 2023-01-21 12:16:47.285462: step: 516/529, loss: 0.005288982763886452 2023-01-21 12:16:48.425187: step: 520/529, loss: 0.03788948059082031 2023-01-21 12:16:49.624942: step: 524/529, loss: 0.0015561580657958984 2023-01-21 12:16:50.827860: step: 528/529, loss: 0.6090884804725647 2023-01-21 12:16:52.055957: step: 532/529, loss: 0.02444629557430744 2023-01-21 12:16:53.229237: step: 536/529, loss: 0.025970458984375 2023-01-21 12:16:54.414415: step: 540/529, loss: 0.0029148103203624487 2023-01-21 12:16:55.656305: step: 544/529, loss: 0.00287132291123271 2023-01-21 12:16:56.868868: step: 548/529, loss: 0.005879211239516735 2023-01-21 12:16:58.052803: step: 552/529, loss: 0.00017366409883834422 2023-01-21 12:16:59.271267: step: 556/529, loss: 0.00036563872708939016 2023-01-21 12:17:00.423755: step: 560/529, loss: 0.008824491873383522 2023-01-21 12:17:01.583627: step: 564/529, loss: 0.00010423660569358617 2023-01-21 12:17:02.742284: step: 568/529, loss: 6.141662743175402e-05 2023-01-21 12:17:03.924593: step: 572/529, loss: 0.0029921531677246094 2023-01-21 12:17:05.097297: step: 576/529, loss: 0.031603239476680756 2023-01-21 12:17:06.313114: step: 580/529, loss: 0.008437538519501686 2023-01-21 12:17:07.526516: step: 584/529, loss: 0.0007762908935546875 2023-01-21 12:17:08.654323: step: 588/529, loss: 6.008148375258315e-06 2023-01-21 12:17:09.867015: step: 592/529, loss: 0.020706748589873314 2023-01-21 12:17:11.050795: step: 596/529, loss: 0.0021098137367516756 2023-01-21 12:17:12.245885: step: 600/529, loss: 0.026454545557498932 2023-01-21 12:17:13.449432: step: 604/529, loss: 0.38733500242233276 2023-01-21 12:17:14.603217: step: 608/529, loss: 0.040145114064216614 2023-01-21 12:17:15.776615: step: 612/529, loss: 0.02242717705667019 2023-01-21 12:17:16.975619: step: 616/529, loss: 0.011246300302445889 2023-01-21 12:17:18.140062: step: 620/529, loss: 0.10426407307386398 2023-01-21 12:17:19.375212: step: 624/529, loss: 0.031542014330625534 2023-01-21 12:17:20.574917: step: 628/529, loss: 0.01098785363137722 2023-01-21 12:17:21.814917: step: 632/529, loss: 0.029364967718720436 2023-01-21 12:17:23.040717: step: 636/529, loss: 0.031473252922296524 2023-01-21 12:17:24.225547: step: 640/529, loss: 0.0009755134815350175 2023-01-21 12:17:25.373337: step: 644/529, loss: 0.003891182132065296 2023-01-21 12:17:26.576273: step: 648/529, loss: 0.01086578331887722 2023-01-21 12:17:27.767782: step: 652/529, loss: 0.040348101407289505 2023-01-21 12:17:28.947097: step: 656/529, loss: 0.0060707093216478825 2023-01-21 12:17:30.141189: step: 660/529, loss: 0.002191019244492054 2023-01-21 12:17:31.364249: step: 664/529, loss: 0.0010872840648517013 2023-01-21 12:17:32.556115: step: 668/529, loss: 0.049207497388124466 2023-01-21 12:17:33.756890: step: 672/529, loss: 0.015920164063572884 2023-01-21 12:17:34.971165: step: 676/529, loss: 0.03878040239214897 2023-01-21 12:17:36.202834: step: 680/529, loss: 0.03665771707892418 2023-01-21 12:17:37.430559: step: 684/529, loss: 0.00017604828462935984 2023-01-21 12:17:38.609291: step: 688/529, loss: 0.0013577460777014494 2023-01-21 12:17:39.791055: step: 692/529, loss: 0.008386707864701748 2023-01-21 12:17:40.963893: step: 696/529, loss: 5.655288987327367e-05 2023-01-21 12:17:42.150312: step: 700/529, loss: 0.02446003071963787 2023-01-21 12:17:43.364871: step: 704/529, loss: 0.004346084780991077 2023-01-21 12:17:44.523580: step: 708/529, loss: 0.004009819123893976 2023-01-21 12:17:45.737952: step: 712/529, loss: 0.021407699212431908 2023-01-21 12:17:46.978381: step: 716/529, loss: 0.0011683463817462325 2023-01-21 12:17:48.166957: step: 720/529, loss: 0.006407547276467085 2023-01-21 12:17:49.380083: step: 724/529, loss: 0.0028341293800622225 2023-01-21 12:17:50.576853: step: 728/529, loss: 0.12858662009239197 2023-01-21 12:17:51.742015: step: 732/529, loss: 0.001036643865518272 2023-01-21 12:17:52.922118: step: 736/529, loss: 0.0035652159713208675 2023-01-21 12:17:54.100477: step: 740/529, loss: 0.002392721129581332 2023-01-21 12:17:55.329558: step: 744/529, loss: 1.487731969973538e-05 2023-01-21 12:17:56.482743: step: 748/529, loss: 0.03475785255432129 2023-01-21 12:17:57.714788: step: 752/529, loss: 0.0001604080171091482 2023-01-21 12:17:58.963146: step: 756/529, loss: 0.0992010086774826 2023-01-21 12:18:00.127371: step: 760/529, loss: 0.005539322271943092 2023-01-21 12:18:01.294248: step: 764/529, loss: 0.006421280093491077 2023-01-21 12:18:02.505511: step: 768/529, loss: 0.0068187713623046875 2023-01-21 12:18:03.708920: step: 772/529, loss: 0.00850753765553236 2023-01-21 12:18:04.883970: step: 776/529, loss: 0.0023262023460119963 2023-01-21 12:18:06.050310: step: 780/529, loss: 0.10414247959852219 2023-01-21 12:18:07.264780: step: 784/529, loss: 0.06970927864313126 2023-01-21 12:18:08.466648: step: 788/529, loss: 0.1101720854640007 2023-01-21 12:18:09.650143: step: 792/529, loss: 0.0008069038740359247 2023-01-21 12:18:10.833700: step: 796/529, loss: 0.006174373906105757 2023-01-21 12:18:12.032257: step: 800/529, loss: 0.07027263194322586 2023-01-21 12:18:13.260916: step: 804/529, loss: 0.012847852893173695 2023-01-21 12:18:14.437908: step: 808/529, loss: 0.008003043942153454 2023-01-21 12:18:15.643427: step: 812/529, loss: 0.03995443508028984 2023-01-21 12:18:16.858369: step: 816/529, loss: 0.03222246095538139 2023-01-21 12:18:18.002665: step: 820/529, loss: 0.02943878062069416 2023-01-21 12:18:19.195720: step: 824/529, loss: 0.0002769470156636089 2023-01-21 12:18:20.418973: step: 828/529, loss: 0.0018831252818927169 2023-01-21 12:18:21.604755: step: 832/529, loss: 0.03874276205897331 2023-01-21 12:18:22.810246: step: 836/529, loss: 0.0018135070567950606 2023-01-21 12:18:24.003439: step: 840/529, loss: 0.015957452356815338 2023-01-21 12:18:25.154745: step: 844/529, loss: 0.004188633523881435 2023-01-21 12:18:26.312606: step: 848/529, loss: 0.004436301998794079 2023-01-21 12:18:27.504163: step: 852/529, loss: 0.04503602907061577 2023-01-21 12:18:28.721373: step: 856/529, loss: 0.013290406204760075 2023-01-21 12:18:29.906049: step: 860/529, loss: 0.0005834579933434725 2023-01-21 12:18:31.105284: step: 864/529, loss: 0.0037806988693773746 2023-01-21 12:18:32.315818: step: 868/529, loss: 0.004334593191742897 2023-01-21 12:18:33.512238: step: 872/529, loss: 0.0030536651611328125 2023-01-21 12:18:34.723665: step: 876/529, loss: 0.0018623352516442537 2023-01-21 12:18:35.910422: step: 880/529, loss: 0.00014553070650435984 2023-01-21 12:18:37.083417: step: 884/529, loss: 0.006325721740722656 2023-01-21 12:18:38.260277: step: 888/529, loss: 8.525848534191027e-05 2023-01-21 12:18:39.480582: step: 892/529, loss: 0.11257939785718918 2023-01-21 12:18:40.684000: step: 896/529, loss: 0.0007785797351971269 2023-01-21 12:18:41.879248: step: 900/529, loss: 0.02764301188290119 2023-01-21 12:18:43.074193: step: 904/529, loss: 0.007180404849350452 2023-01-21 12:18:44.319724: step: 908/529, loss: 0.032185934484004974 2023-01-21 12:18:45.516686: step: 912/529, loss: 7.05718994140625e-05 2023-01-21 12:18:46.708705: step: 916/529, loss: 0.021619606763124466 2023-01-21 12:18:47.861943: step: 920/529, loss: 0.00023937225341796875 2023-01-21 12:18:49.041902: step: 924/529, loss: 0.0003494262637104839 2023-01-21 12:18:50.248069: step: 928/529, loss: 0.009297179989516735 2023-01-21 12:18:51.460561: step: 932/529, loss: 0.0012050628429278731 2023-01-21 12:18:52.691338: step: 936/529, loss: 0.04403753578662872 2023-01-21 12:18:53.882218: step: 940/529, loss: 0.00047931671724654734 2023-01-21 12:18:55.054554: step: 944/529, loss: 0.0030799866653978825 2023-01-21 12:18:56.221704: step: 948/529, loss: 9.698867506813258e-05 2023-01-21 12:18:57.434375: step: 952/529, loss: 0.016660690307617188 2023-01-21 12:18:58.593213: step: 956/529, loss: 0.025107193738222122 2023-01-21 12:18:59.786635: step: 960/529, loss: 0.010676764883100986 2023-01-21 12:19:00.981314: step: 964/529, loss: 0.028946496546268463 2023-01-21 12:19:02.158054: step: 968/529, loss: 0.02181243896484375 2023-01-21 12:19:03.367209: step: 972/529, loss: 0.07496700435876846 2023-01-21 12:19:04.543795: step: 976/529, loss: 0.004200423136353493 2023-01-21 12:19:05.699513: step: 980/529, loss: 0.005382442381232977 2023-01-21 12:19:06.848614: step: 984/529, loss: 0.0029206275939941406 2023-01-21 12:19:08.029279: step: 988/529, loss: 0.10541171580553055 2023-01-21 12:19:09.213968: step: 992/529, loss: 0.0001316070556640625 2023-01-21 12:19:10.427094: step: 996/529, loss: 0.0007550239679403603 2023-01-21 12:19:11.607060: step: 1000/529, loss: 0.00869069155305624 2023-01-21 12:19:12.792123: step: 1004/529, loss: 0.029651977121829987 2023-01-21 12:19:14.037817: step: 1008/529, loss: 0.006114387419074774 2023-01-21 12:19:15.176925: step: 1012/529, loss: 0.06513696163892746 2023-01-21 12:19:16.368879: step: 1016/529, loss: 0.03226127475500107 2023-01-21 12:19:17.571992: step: 1020/529, loss: 0.03772401809692383 2023-01-21 12:19:18.803755: step: 1024/529, loss: 0.03134498745203018 2023-01-21 12:19:19.985451: step: 1028/529, loss: 0.015622902661561966 2023-01-21 12:19:21.195510: step: 1032/529, loss: 0.003602599957957864 2023-01-21 12:19:22.398789: step: 1036/529, loss: 0.0030490877106785774 2023-01-21 12:19:23.590670: step: 1040/529, loss: 0.01727619208395481 2023-01-21 12:19:24.812847: step: 1044/529, loss: 0.03474845737218857 2023-01-21 12:19:26.059142: step: 1048/529, loss: 0.016892720013856888 2023-01-21 12:19:27.287463: step: 1052/529, loss: 0.0053730010986328125 2023-01-21 12:19:28.466220: step: 1056/529, loss: 0.0010399818420410156 2023-01-21 12:19:29.671340: step: 1060/529, loss: 0.016312886029481888 2023-01-21 12:19:30.859865: step: 1064/529, loss: 0.00034084319486282766 2023-01-21 12:19:32.100817: step: 1068/529, loss: 0.04338092729449272 2023-01-21 12:19:33.282064: step: 1072/529, loss: 0.04091205447912216 2023-01-21 12:19:34.451524: step: 1076/529, loss: 0.018723249435424805 2023-01-21 12:19:35.651011: step: 1080/529, loss: 0.037447549402713776 2023-01-21 12:19:36.838892: step: 1084/529, loss: 5.149841672391631e-05 2023-01-21 12:19:38.034491: step: 1088/529, loss: 0.11782610416412354 2023-01-21 12:19:39.226083: step: 1092/529, loss: 0.011681366711854935 2023-01-21 12:19:40.403485: step: 1096/529, loss: 0.03158588334918022 2023-01-21 12:19:41.622524: step: 1100/529, loss: 0.0027848242316395044 2023-01-21 12:19:42.839527: step: 1104/529, loss: 0.0015151978004723787 2023-01-21 12:19:44.019566: step: 1108/529, loss: 0.0001126289353123866 2023-01-21 12:19:45.224795: step: 1112/529, loss: 0.2589547038078308 2023-01-21 12:19:46.444272: step: 1116/529, loss: 0.006248379126191139 2023-01-21 12:19:47.671133: step: 1120/529, loss: 0.0002654075506143272 2023-01-21 12:19:48.870894: step: 1124/529, loss: 0.00021305083646439016 2023-01-21 12:19:50.063823: step: 1128/529, loss: 0.035358332097530365 2023-01-21 12:19:51.219576: step: 1132/529, loss: 0.002123546553775668 2023-01-21 12:19:52.411131: step: 1136/529, loss: 0.00122661585919559 2023-01-21 12:19:53.626088: step: 1140/529, loss: 9.36508149607107e-05 2023-01-21 12:19:54.835804: step: 1144/529, loss: 0.00020828247943427414 2023-01-21 12:19:56.034298: step: 1148/529, loss: 0.0339202880859375 2023-01-21 12:19:57.220647: step: 1152/529, loss: 0.0015138626331463456 2023-01-21 12:19:58.396722: step: 1156/529, loss: 0.012713718228042126 2023-01-21 12:19:59.643218: step: 1160/529, loss: 0.05988325923681259 2023-01-21 12:20:00.828715: step: 1164/529, loss: 6.065368506824598e-05 2023-01-21 12:20:02.023773: step: 1168/529, loss: 0.025690842419862747 2023-01-21 12:20:03.229011: step: 1172/529, loss: 0.024699974805116653 2023-01-21 12:20:04.435924: step: 1176/529, loss: 0.01584930345416069 2023-01-21 12:20:05.625936: step: 1180/529, loss: 0.0004261970752850175 2023-01-21 12:20:06.834571: step: 1184/529, loss: 0.0013479233020916581 2023-01-21 12:20:07.994046: step: 1188/529, loss: 0.0002750396670307964 2023-01-21 12:20:09.153804: step: 1192/529, loss: 0.001307678292505443 2023-01-21 12:20:10.343348: step: 1196/529, loss: 0.013512516394257545 2023-01-21 12:20:11.503700: step: 1200/529, loss: 0.016615010797977448 2023-01-21 12:20:12.695813: step: 1204/529, loss: 0.002271938370540738 2023-01-21 12:20:13.853834: step: 1208/529, loss: 0.05526590347290039 2023-01-21 12:20:15.058920: step: 1212/529, loss: 0.021343043074011803 2023-01-21 12:20:16.200394: step: 1216/529, loss: 0.0004173278866801411 2023-01-21 12:20:17.404380: step: 1220/529, loss: 0.006449985783547163 2023-01-21 12:20:18.590169: step: 1224/529, loss: 0.027311896905303 2023-01-21 12:20:19.783813: step: 1228/529, loss: 0.01453247107565403 2023-01-21 12:20:20.971020: step: 1232/529, loss: 0.0019279479747638106 2023-01-21 12:20:22.113848: step: 1236/529, loss: 0.0004570961173158139 2023-01-21 12:20:23.315827: step: 1240/529, loss: 0.00037975312443450093 2023-01-21 12:20:24.539418: step: 1244/529, loss: 0.016918564215302467 2023-01-21 12:20:25.736665: step: 1248/529, loss: 0.011882019229233265 2023-01-21 12:20:26.938664: step: 1252/529, loss: 0.022782515734434128 2023-01-21 12:20:28.132683: step: 1256/529, loss: 0.2287009209394455 2023-01-21 12:20:29.385578: step: 1260/529, loss: 0.0557008758187294 2023-01-21 12:20:30.556402: step: 1264/529, loss: 0.004025841131806374 2023-01-21 12:20:31.739011: step: 1268/529, loss: 0.002757263369858265 2023-01-21 12:20:33.075720: step: 1272/529, loss: 0.002262687776237726 2023-01-21 12:20:34.244015: step: 1276/529, loss: 0.17201462388038635 2023-01-21 12:20:35.488575: step: 1280/529, loss: 0.0007619857788085938 2023-01-21 12:20:36.679942: step: 1284/529, loss: 0.038854315876960754 2023-01-21 12:20:37.886286: step: 1288/529, loss: 0.010829162783920765 2023-01-21 12:20:39.068786: step: 1292/529, loss: 0.028765391558408737 2023-01-21 12:20:40.267038: step: 1296/529, loss: 0.004512500949203968 2023-01-21 12:20:41.509873: step: 1300/529, loss: 0.01488342322409153 2023-01-21 12:20:42.801459: step: 1304/529, loss: 0.05646705627441406 2023-01-21 12:20:43.988063: step: 1308/529, loss: 0.0418243408203125 2023-01-21 12:20:45.142804: step: 1312/529, loss: 0.007448387332260609 2023-01-21 12:20:46.327083: step: 1316/529, loss: 0.004742145538330078 2023-01-21 12:20:47.546478: step: 1320/529, loss: 0.005156040191650391 2023-01-21 12:20:48.727273: step: 1324/529, loss: 0.1142582893371582 2023-01-21 12:20:49.906359: step: 1328/529, loss: 0.003535938449203968 2023-01-21 12:20:51.091745: step: 1332/529, loss: 8.296966552734375e-05 2023-01-21 12:20:52.302998: step: 1336/529, loss: 0.0034512518905103207 2023-01-21 12:20:53.521647: step: 1340/529, loss: 0.00480918912217021 2023-01-21 12:20:54.678523: step: 1344/529, loss: 0.014030265621840954 2023-01-21 12:20:55.892773: step: 1348/529, loss: 0.016479969024658203 2023-01-21 12:20:57.079228: step: 1352/529, loss: 0.1699983775615692 2023-01-21 12:20:58.320636: step: 1356/529, loss: 0.03920488432049751 2023-01-21 12:20:59.511576: step: 1360/529, loss: 0.007481765933334827 2023-01-21 12:21:00.674612: step: 1364/529, loss: 0.00048913958016783 2023-01-21 12:21:01.890591: step: 1368/529, loss: 0.08634891360998154 2023-01-21 12:21:03.079395: step: 1372/529, loss: 0.00015211105346679688 2023-01-21 12:21:04.275188: step: 1376/529, loss: 0.036019135266542435 2023-01-21 12:21:05.450969: step: 1380/529, loss: 3.0517578125e-05 2023-01-21 12:21:06.668606: step: 1384/529, loss: 0.008962249383330345 2023-01-21 12:21:07.885284: step: 1388/529, loss: 0.2295665740966797 2023-01-21 12:21:09.115902: step: 1392/529, loss: 0.027381135150790215 2023-01-21 12:21:10.311029: step: 1396/529, loss: 0.0017087936867028475 2023-01-21 12:21:11.493203: step: 1400/529, loss: 0.00073666573734954 2023-01-21 12:21:12.694354: step: 1404/529, loss: 0.07298936694860458 2023-01-21 12:21:13.838406: step: 1408/529, loss: 0.0032261849846690893 2023-01-21 12:21:15.031212: step: 1412/529, loss: 0.0341595895588398 2023-01-21 12:21:16.212119: step: 1416/529, loss: 0.022269297391176224 2023-01-21 12:21:17.357521: step: 1420/529, loss: 4.100799742445815e-06 2023-01-21 12:21:18.547132: step: 1424/529, loss: 0.011155366897583008 2023-01-21 12:21:19.729355: step: 1428/529, loss: 0.001195764634758234 2023-01-21 12:21:20.903562: step: 1432/529, loss: 0.0007196426158770919 2023-01-21 12:21:22.098770: step: 1436/529, loss: 0.03333578258752823 2023-01-21 12:21:23.312875: step: 1440/529, loss: 0.03145332634449005 2023-01-21 12:21:24.510496: step: 1444/529, loss: 0.0593264065682888 2023-01-21 12:21:25.719236: step: 1448/529, loss: 0.01563272438943386 2023-01-21 12:21:26.903158: step: 1452/529, loss: 0.0015298366779461503 2023-01-21 12:21:28.078232: step: 1456/529, loss: 0.023151110857725143 2023-01-21 12:21:29.246097: step: 1460/529, loss: 0.008292580023407936 2023-01-21 12:21:30.429216: step: 1464/529, loss: 0.0032569884788244963 2023-01-21 12:21:31.664066: step: 1468/529, loss: 0.055326174944639206 2023-01-21 12:21:32.858474: step: 1472/529, loss: 0.0008455276256427169 2023-01-21 12:21:34.036272: step: 1476/529, loss: 0.0005357742193154991 2023-01-21 12:21:35.200583: step: 1480/529, loss: 0.03813495859503746 2023-01-21 12:21:36.419895: step: 1484/529, loss: 0.002634143689647317 2023-01-21 12:21:37.637657: step: 1488/529, loss: 0.006098556332290173 2023-01-21 12:21:38.822311: step: 1492/529, loss: 2.28881845032447e-06 2023-01-21 12:21:40.039780: step: 1496/529, loss: 0.017171192914247513 2023-01-21 12:21:41.223820: step: 1500/529, loss: 0.0027893066871911287 2023-01-21 12:21:42.423707: step: 1504/529, loss: 0.019234228879213333 2023-01-21 12:21:43.620009: step: 1508/529, loss: 0.00029468536376953125 2023-01-21 12:21:44.810300: step: 1512/529, loss: 0.00987706147134304 2023-01-21 12:21:45.972596: step: 1516/529, loss: 0.04477691650390625 2023-01-21 12:21:47.184792: step: 1520/529, loss: 0.000858306884765625 2023-01-21 12:21:48.379926: step: 1524/529, loss: 0.021194934844970703 2023-01-21 12:21:49.609483: step: 1528/529, loss: 0.04408779367804527 2023-01-21 12:21:50.807177: step: 1532/529, loss: 0.002058029407635331 2023-01-21 12:21:52.059418: step: 1536/529, loss: 0.025771522894501686 2023-01-21 12:21:53.258372: step: 1540/529, loss: 0.05279712751507759 2023-01-21 12:21:54.437268: step: 1544/529, loss: 0.0425536148250103 2023-01-21 12:21:55.639947: step: 1548/529, loss: 0.05156688764691353 2023-01-21 12:21:56.802164: step: 1552/529, loss: 0.005675602238625288 2023-01-21 12:21:57.996391: step: 1556/529, loss: 0.00023174285888671875 2023-01-21 12:21:59.189720: step: 1560/529, loss: 0.0016486168606206775 2023-01-21 12:22:00.403549: step: 1564/529, loss: 0.0038089752197265625 2023-01-21 12:22:01.553281: step: 1568/529, loss: 0.0075532435439527035 2023-01-21 12:22:02.744054: step: 1572/529, loss: 0.002165222307667136 2023-01-21 12:22:03.931172: step: 1576/529, loss: 0.05667886510491371 2023-01-21 12:22:05.095539: step: 1580/529, loss: 0.00043935776920989156 2023-01-21 12:22:06.325548: step: 1584/529, loss: 0.00920791644603014 2023-01-21 12:22:07.502747: step: 1588/529, loss: 0.0038474085740745068 2023-01-21 12:22:08.681028: step: 1592/529, loss: 0.07811546325683594 2023-01-21 12:22:09.849980: step: 1596/529, loss: 0.05807380750775337 2023-01-21 12:22:11.042023: step: 1600/529, loss: 0.00041942595271393657 2023-01-21 12:22:12.229277: step: 1604/529, loss: 0.02204456366598606 2023-01-21 12:22:13.479946: step: 1608/529, loss: 0.005208015441894531 2023-01-21 12:22:14.650032: step: 1612/529, loss: 0.008244229480624199 2023-01-21 12:22:15.823963: step: 1616/529, loss: 0.0006495475536212325 2023-01-21 12:22:17.001612: step: 1620/529, loss: 0.0012501715682446957 2023-01-21 12:22:18.143353: step: 1624/529, loss: 0.011770821176469326 2023-01-21 12:22:19.305580: step: 1628/529, loss: 0.0430939681828022 2023-01-21 12:22:20.538199: step: 1632/529, loss: 0.0009360313997603953 2023-01-21 12:22:21.739390: step: 1636/529, loss: 0.03535766527056694 2023-01-21 12:22:22.939281: step: 1640/529, loss: 0.007501030340790749 2023-01-21 12:22:24.125213: step: 1644/529, loss: 0.015155602246522903 2023-01-21 12:22:25.294723: step: 1648/529, loss: 0.025030899792909622 2023-01-21 12:22:26.465328: step: 1652/529, loss: 0.005852699279785156 2023-01-21 12:22:27.630708: step: 1656/529, loss: 0.001203346299007535 2023-01-21 12:22:28.868262: step: 1660/529, loss: 0.0012742995750159025 2023-01-21 12:22:30.087678: step: 1664/529, loss: 0.01425943337380886 2023-01-21 12:22:31.302927: step: 1668/529, loss: 0.0005098342662677169 2023-01-21 12:22:32.530248: step: 1672/529, loss: 0.0005107164615765214 2023-01-21 12:22:33.693116: step: 1676/529, loss: 0.0010564328404143453 2023-01-21 12:22:34.901533: step: 1680/529, loss: 0.00022201538376975805 2023-01-21 12:22:36.090112: step: 1684/529, loss: 0.007825469598174095 2023-01-21 12:22:37.291726: step: 1688/529, loss: 0.05650081858038902 2023-01-21 12:22:38.441782: step: 1692/529, loss: 0.0008904456626623869 2023-01-21 12:22:39.635162: step: 1696/529, loss: 0.003869724227115512 2023-01-21 12:22:40.838602: step: 1700/529, loss: 0.02690863609313965 2023-01-21 12:22:42.022452: step: 1704/529, loss: 0.00015697479830123484 2023-01-21 12:22:43.209599: step: 1708/529, loss: 0.0013691901694983244 2023-01-21 12:22:44.407331: step: 1712/529, loss: 0.0009058952564373612 2023-01-21 12:22:45.629501: step: 1716/529, loss: 0.06457547843456268 2023-01-21 12:22:46.809329: step: 1720/529, loss: 0.004027175717055798 2023-01-21 12:22:48.020628: step: 1724/529, loss: 0.0747649222612381 2023-01-21 12:22:49.246484: step: 1728/529, loss: 0.011796379461884499 2023-01-21 12:22:50.451332: step: 1732/529, loss: 0.09886636584997177 2023-01-21 12:22:51.621793: step: 1736/529, loss: 0.009797287173569202 2023-01-21 12:22:52.804703: step: 1740/529, loss: 0.0022899629548192024 2023-01-21 12:22:54.034160: step: 1744/529, loss: 0.0005227565998211503 2023-01-21 12:22:55.213090: step: 1748/529, loss: 0.004146575927734375 2023-01-21 12:22:56.440102: step: 1752/529, loss: 0.0010561943054199219 2023-01-21 12:22:57.603842: step: 1756/529, loss: 0.016541479155421257 2023-01-21 12:22:58.814385: step: 1760/529, loss: 0.03152790293097496 2023-01-21 12:23:00.002835: step: 1764/529, loss: 0.05371475592255592 2023-01-21 12:23:01.184626: step: 1768/529, loss: 0.11742812395095825 2023-01-21 12:23:02.438440: step: 1772/529, loss: 0.13532543182373047 2023-01-21 12:23:03.609384: step: 1776/529, loss: 0.06766372174024582 2023-01-21 12:23:04.797468: step: 1780/529, loss: 0.37517157196998596 2023-01-21 12:23:05.989662: step: 1784/529, loss: 0.046967267990112305 2023-01-21 12:23:07.185107: step: 1788/529, loss: 0.07056532055139542 2023-01-21 12:23:08.364089: step: 1792/529, loss: 0.0005147933843545616 2023-01-21 12:23:09.561587: step: 1796/529, loss: 0.0560787171125412 2023-01-21 12:23:10.811791: step: 1800/529, loss: 0.006079196929931641 2023-01-21 12:23:12.012682: step: 1804/529, loss: 0.013102340511977673 2023-01-21 12:23:13.179826: step: 1808/529, loss: 0.44049468636512756 2023-01-21 12:23:14.367061: step: 1812/529, loss: 0.027129841968417168 2023-01-21 12:23:15.557686: step: 1816/529, loss: 0.012971687130630016 2023-01-21 12:23:16.757602: step: 1820/529, loss: 0.038538359105587006 2023-01-21 12:23:17.919131: step: 1824/529, loss: 0.00048389434232376516 2023-01-21 12:23:19.106428: step: 1828/529, loss: 0.0005072593921795487 2023-01-21 12:23:20.308980: step: 1832/529, loss: 0.002433872316032648 2023-01-21 12:23:21.544183: step: 1836/529, loss: 0.024733543395996094 2023-01-21 12:23:22.720716: step: 1840/529, loss: 0.004181766416877508 2023-01-21 12:23:23.906422: step: 1844/529, loss: 0.0002925872977357358 2023-01-21 12:23:25.072554: step: 1848/529, loss: 0.02150707319378853 2023-01-21 12:23:26.279930: step: 1852/529, loss: 0.0015194893348962069 2023-01-21 12:23:27.453686: step: 1856/529, loss: 0.0004215836524963379 2023-01-21 12:23:28.627402: step: 1860/529, loss: 0.052111152559518814 2023-01-21 12:23:29.847853: step: 1864/529, loss: 0.0054779052734375 2023-01-21 12:23:31.010699: step: 1868/529, loss: 0.005977058317512274 2023-01-21 12:23:32.205999: step: 1872/529, loss: 0.011924266815185547 2023-01-21 12:23:33.387002: step: 1876/529, loss: 0.020780611783266068 2023-01-21 12:23:34.572151: step: 1880/529, loss: 0.09207862615585327 2023-01-21 12:23:35.758116: step: 1884/529, loss: 0.00015788078599143773 2023-01-21 12:23:37.027632: step: 1888/529, loss: 0.003840923309326172 2023-01-21 12:23:38.200780: step: 1892/529, loss: 0.0019676208030432463 2023-01-21 12:23:39.399406: step: 1896/529, loss: 0.00837850570678711 2023-01-21 12:23:40.578821: step: 1900/529, loss: 0.011746288277208805 2023-01-21 12:23:41.750975: step: 1904/529, loss: 0.041113436222076416 2023-01-21 12:23:42.917724: step: 1908/529, loss: 0.002162837889045477 2023-01-21 12:23:44.116220: step: 1912/529, loss: 0.11363878101110458 2023-01-21 12:23:45.310546: step: 1916/529, loss: 0.0003292083856649697 2023-01-21 12:23:46.500426: step: 1920/529, loss: 0.1145562157034874 2023-01-21 12:23:47.670673: step: 1924/529, loss: 0.006500816438347101 2023-01-21 12:23:48.851721: step: 1928/529, loss: 0.07559080421924591 2023-01-21 12:23:50.056564: step: 1932/529, loss: 0.050786685198545456 2023-01-21 12:23:51.232152: step: 1936/529, loss: 1.2361985445022583 2023-01-21 12:23:52.455960: step: 1940/529, loss: 0.007229328155517578 2023-01-21 12:23:53.709934: step: 1944/529, loss: 0.03435201942920685 2023-01-21 12:23:54.932533: step: 1948/529, loss: 0.017311954870820045 2023-01-21 12:23:56.120429: step: 1952/529, loss: 0.005193233489990234 2023-01-21 12:23:57.302760: step: 1956/529, loss: 0.00022010804968886077 2023-01-21 12:23:58.485511: step: 1960/529, loss: 0.015543175861239433 2023-01-21 12:23:59.639466: step: 1964/529, loss: 0.003014469286426902 2023-01-21 12:24:00.772242: step: 1968/529, loss: 9.55581635935232e-05 2023-01-21 12:24:01.962912: step: 1972/529, loss: 0.014536094851791859 2023-01-21 12:24:03.135532: step: 1976/529, loss: 0.03141927719116211 2023-01-21 12:24:04.325057: step: 1980/529, loss: 0.0026701928582042456 2023-01-21 12:24:05.492580: step: 1984/529, loss: 0.002510261721909046 2023-01-21 12:24:06.670028: step: 1988/529, loss: 0.03645653650164604 2023-01-21 12:24:07.863911: step: 1992/529, loss: 0.00030179024906829 2023-01-21 12:24:09.056522: step: 1996/529, loss: 0.0101776123046875 2023-01-21 12:24:10.233043: step: 2000/529, loss: 0.006832325365394354 2023-01-21 12:24:11.458611: step: 2004/529, loss: 0.007663822267204523 2023-01-21 12:24:12.688488: step: 2008/529, loss: 0.009458637796342373 2023-01-21 12:24:13.863705: step: 2012/529, loss: 0.016492033377289772 2023-01-21 12:24:15.019494: step: 2016/529, loss: 0.005748271942138672 2023-01-21 12:24:16.187959: step: 2020/529, loss: 0.0003354072687216103 2023-01-21 12:24:17.355917: step: 2024/529, loss: 0.030788470059633255 2023-01-21 12:24:18.547292: step: 2028/529, loss: 0.0008027076837606728 2023-01-21 12:24:19.708367: step: 2032/529, loss: 0.006003665737807751 2023-01-21 12:24:20.899621: step: 2036/529, loss: 0.00010275841486873105 2023-01-21 12:24:22.075906: step: 2040/529, loss: 0.004646110814064741 2023-01-21 12:24:23.303604: step: 2044/529, loss: 0.015318943187594414 2023-01-21 12:24:24.449659: step: 2048/529, loss: 0.017773056402802467 2023-01-21 12:24:25.631215: step: 2052/529, loss: 0.01988658867776394 2023-01-21 12:24:26.816647: step: 2056/529, loss: 0.00151405343785882 2023-01-21 12:24:27.995443: step: 2060/529, loss: 0.001674652099609375 2023-01-21 12:24:29.199774: step: 2064/529, loss: 0.015311241149902344 2023-01-21 12:24:30.429910: step: 2068/529, loss: 0.007903862744569778 2023-01-21 12:24:31.654984: step: 2072/529, loss: 0.01455917302519083 2023-01-21 12:24:32.830504: step: 2076/529, loss: 0.0010942459339275956 2023-01-21 12:24:34.001368: step: 2080/529, loss: 0.0020085335709154606 2023-01-21 12:24:35.219649: step: 2084/529, loss: 0.002347290515899658 2023-01-21 12:24:36.403451: step: 2088/529, loss: 0.009985732845962048 2023-01-21 12:24:37.547470: step: 2092/529, loss: 1.0842952728271484 2023-01-21 12:24:38.700009: step: 2096/529, loss: 0.0009742736583575606 2023-01-21 12:24:39.898903: step: 2100/529, loss: 0.005417156033217907 2023-01-21 12:24:41.081156: step: 2104/529, loss: 4.9591064453125e-05 2023-01-21 12:24:42.277881: step: 2108/529, loss: 0.005000579636543989 2023-01-21 12:24:43.470680: step: 2112/529, loss: 0.004596090875566006 2023-01-21 12:24:44.638815: step: 2116/529, loss: 0.04891243204474449 ================================================== Loss: 0.028 -------------------- Dev: {'event': {'p': 0.6046747967479674, 'r': 0.7922769640479361, 'f1': 0.685878962536023}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Test: {'event': {'p': 0.6377990430622009, 'r': 0.7986818454164171, 'f1': 0.7092311785049216}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Chinese: {'event': {'p': 0.631578947368421, 'r': 0.8888888888888888, 'f1': 0.7384615384615385}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Korean: {'event': {'p': 0.5, 'r': 0.5079365079365079, 'f1': 0.5039370078740157}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} 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': 19} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** Epoch: 20 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:25:27.658220: step: 4/529, loss: 0.0016073227161541581 2023-01-21 12:25:28.810549: step: 8/529, loss: 0.0013536930782720447 2023-01-21 12:25:29.985689: step: 12/529, loss: 0.03565731272101402 2023-01-21 12:25:31.191785: step: 16/529, loss: 0.02318735234439373 2023-01-21 12:25:32.402370: step: 20/529, loss: 0.00046825408935546875 2023-01-21 12:25:33.615318: step: 24/529, loss: 0.011806678958237171 2023-01-21 12:25:34.760673: step: 28/529, loss: 0.002562230918556452 2023-01-21 12:25:35.971012: step: 32/529, loss: 7.643699791515246e-05 2023-01-21 12:25:37.118544: step: 36/529, loss: 0.09265337139368057 2023-01-21 12:25:38.302096: step: 40/529, loss: 9.698867506813258e-05 2023-01-21 12:25:39.505516: step: 44/529, loss: 0.00016403199697379023 2023-01-21 12:25:40.693705: step: 48/529, loss: 0.003382873721420765 2023-01-21 12:25:41.955983: step: 52/529, loss: 0.006996918004006147 2023-01-21 12:25:43.186593: step: 56/529, loss: 0.0020665170159190893 2023-01-21 12:25:44.394412: step: 60/529, loss: 0.0033197402954101562 2023-01-21 12:25:45.592758: step: 64/529, loss: 0.028774453327059746 2023-01-21 12:25:46.788901: step: 68/529, loss: 0.010026264004409313 2023-01-21 12:25:47.954604: step: 72/529, loss: 0.0035764218773692846 2023-01-21 12:25:49.146866: step: 76/529, loss: 0.00037221910315565765 2023-01-21 12:25:50.328303: step: 80/529, loss: 0.046616170555353165 2023-01-21 12:25:51.506020: step: 84/529, loss: 0.016783332452178 2023-01-21 12:25:52.696022: step: 88/529, loss: 0.014735984615981579 2023-01-21 12:25:53.891876: step: 92/529, loss: 0.0016683578724041581 2023-01-21 12:25:55.085046: step: 96/529, loss: 0.00018568038649391383 2023-01-21 12:25:56.262962: step: 100/529, loss: 0.0022497177124023438 2023-01-21 12:25:57.462154: step: 104/529, loss: 0.028211116790771484 2023-01-21 12:25:58.713860: step: 108/529, loss: 0.04281797632575035 2023-01-21 12:25:59.894678: step: 112/529, loss: 0.010030555538833141 2023-01-21 12:26:01.055433: step: 116/529, loss: 0.04286494478583336 2023-01-21 12:26:02.248402: step: 120/529, loss: 0.016987990587949753 2023-01-21 12:26:03.419404: step: 124/529, loss: 0.000446128862677142 2023-01-21 12:26:04.627352: step: 128/529, loss: 0.019921589642763138 2023-01-21 12:26:05.809458: step: 132/529, loss: 0.04748887941241264 2023-01-21 12:26:07.011100: step: 136/529, loss: 0.00899276789277792 2023-01-21 12:26:08.195694: step: 140/529, loss: 0.0006554603460244834 2023-01-21 12:26:09.382072: step: 144/529, loss: 1.354217511106981e-05 2023-01-21 12:26:10.549471: step: 148/529, loss: 0.0031435966957360506 2023-01-21 12:26:11.716846: step: 152/529, loss: 0.00028028490487486124 2023-01-21 12:26:12.902819: step: 156/529, loss: 0.00738182058557868 2023-01-21 12:26:14.081888: step: 160/529, loss: 0.013612747192382812 2023-01-21 12:26:15.270134: step: 164/529, loss: 0.00028209685115143657 2023-01-21 12:26:16.457050: step: 168/529, loss: 0.0022464515641331673 2023-01-21 12:26:17.625121: step: 172/529, loss: 0.00041670797509141266 2023-01-21 12:26:18.860994: step: 176/529, loss: 0.0002637386496644467 2023-01-21 12:26:20.012096: step: 180/529, loss: 0.00023403167142532766 2023-01-21 12:26:21.212973: step: 184/529, loss: 0.0037538527976721525 2023-01-21 12:26:22.391640: step: 188/529, loss: 0.0004145622078794986 2023-01-21 12:26:23.566014: step: 192/529, loss: 0.011241721920669079 2023-01-21 12:26:24.755017: step: 196/529, loss: 0.007048130035400391 2023-01-21 12:26:25.951148: step: 200/529, loss: 0.0032802580390125513 2023-01-21 12:26:27.124273: step: 204/529, loss: 0.02246570587158203 2023-01-21 12:26:28.317481: step: 208/529, loss: 0.02220172993838787 2023-01-21 12:26:29.526355: step: 212/529, loss: 0.7484796047210693 2023-01-21 12:26:30.695560: step: 216/529, loss: 0.01000671461224556 2023-01-21 12:26:31.886865: step: 220/529, loss: 0.006572246551513672 2023-01-21 12:26:33.115221: step: 224/529, loss: 0.008432101458311081 2023-01-21 12:26:34.301890: step: 228/529, loss: 0.004733157344162464 2023-01-21 12:26:35.492656: step: 232/529, loss: 0.010437775403261185 2023-01-21 12:26:36.653494: step: 236/529, loss: 0.0018742561805993319 2023-01-21 12:26:37.823096: step: 240/529, loss: 6.837844557594508e-05 2023-01-21 12:26:39.010340: step: 244/529, loss: 0.02784719690680504 2023-01-21 12:26:40.153559: step: 248/529, loss: 0.0012046813499182463 2023-01-21 12:26:41.350679: step: 252/529, loss: 0.0001548767031636089 2023-01-21 12:26:42.531758: step: 256/529, loss: 0.0013825417263433337 2023-01-21 12:26:43.735381: step: 260/529, loss: 0.0012588024837896228 2023-01-21 12:26:44.917311: step: 264/529, loss: 0.005891228094696999 2023-01-21 12:26:46.123680: step: 268/529, loss: 0.0007252931827679276 2023-01-21 12:26:47.288736: step: 272/529, loss: 0.015035438351333141 2023-01-21 12:26:48.509988: step: 276/529, loss: 0.09604117274284363 2023-01-21 12:26:49.700780: step: 280/529, loss: 0.023737384006381035 2023-01-21 12:26:50.900486: step: 284/529, loss: 0.0026497840881347656 2023-01-21 12:26:52.161367: step: 288/529, loss: 0.07260475307703018 2023-01-21 12:26:53.296657: step: 292/529, loss: 0.001001119613647461 2023-01-21 12:26:54.477167: step: 296/529, loss: 0.002080631209537387 2023-01-21 12:26:55.663184: step: 300/529, loss: 0.05675039440393448 2023-01-21 12:26:56.820958: step: 304/529, loss: 0.0003025055048055947 2023-01-21 12:26:57.995114: step: 308/529, loss: 0.002463150303810835 2023-01-21 12:26:59.190773: step: 312/529, loss: 0.07567157596349716 2023-01-21 12:27:00.394335: step: 316/529, loss: 0.005502796731889248 2023-01-21 12:27:01.577547: step: 320/529, loss: 0.007103347685188055 2023-01-21 12:27:02.789322: step: 324/529, loss: 0.043245889246463776 2023-01-21 12:27:03.982555: step: 328/529, loss: 0.05011434480547905 2023-01-21 12:27:05.184726: step: 332/529, loss: 0.05334950238466263 2023-01-21 12:27:06.368496: step: 336/529, loss: 0.020284080877900124 2023-01-21 12:27:07.547805: step: 340/529, loss: 6.294250852079131e-06 2023-01-21 12:27:08.738483: step: 344/529, loss: 0.593254804611206 2023-01-21 12:27:09.876460: step: 348/529, loss: 9.775161743164062e-05 2023-01-21 12:27:11.101253: step: 352/529, loss: 0.0025234222412109375 2023-01-21 12:27:12.280376: step: 356/529, loss: 0.22762832045555115 2023-01-21 12:27:13.467469: step: 360/529, loss: 0.0017623902531340718 2023-01-21 12:27:14.683246: step: 364/529, loss: 0.007865333929657936 2023-01-21 12:27:15.836593: step: 368/529, loss: 0.08284921199083328 2023-01-21 12:27:17.015572: step: 372/529, loss: 0.0011682510375976562 2023-01-21 12:27:18.221069: step: 376/529, loss: 1.4209747860149946e-05 2023-01-21 12:27:19.409160: step: 380/529, loss: 0.01260232925415039 2023-01-21 12:27:20.583868: step: 384/529, loss: 0.006535148713737726 2023-01-21 12:27:21.767752: step: 388/529, loss: 0.0010840416653081775 2023-01-21 12:27:22.971321: step: 392/529, loss: 0.11127223819494247 2023-01-21 12:27:24.141715: step: 396/529, loss: 0.004363822750747204 2023-01-21 12:27:25.296003: step: 400/529, loss: 0.0313815139234066 2023-01-21 12:27:26.516151: step: 404/529, loss: 0.0976623073220253 2023-01-21 12:27:27.723952: step: 408/529, loss: 0.030428696423768997 2023-01-21 12:27:28.890280: step: 412/529, loss: 0.9066535234451294 2023-01-21 12:27:30.086118: step: 416/529, loss: 0.00017480850510764867 2023-01-21 12:27:31.328137: step: 420/529, loss: 0.00018277167691849172 2023-01-21 12:27:32.536489: step: 424/529, loss: 0.00012407303438521922 2023-01-21 12:27:33.721138: step: 428/529, loss: 7.80105619924143e-05 2023-01-21 12:27:34.979603: step: 432/529, loss: 0.1299879550933838 2023-01-21 12:27:36.193511: step: 436/529, loss: 0.007201099302619696 2023-01-21 12:27:37.417977: step: 440/529, loss: 0.030178260058164597 2023-01-21 12:27:38.627670: step: 444/529, loss: 0.0011638641590252519 2023-01-21 12:27:39.822667: step: 448/529, loss: 0.01890125311911106 2023-01-21 12:27:41.029460: step: 452/529, loss: 0.0255903247743845 2023-01-21 12:27:42.285084: step: 456/529, loss: 0.0007987022399902344 2023-01-21 12:27:43.451171: step: 460/529, loss: 0.0005666256183758378 2023-01-21 12:27:44.674813: step: 464/529, loss: 0.0009533882257528603 2023-01-21 12:27:45.843358: step: 468/529, loss: 0.009092355147004128 2023-01-21 12:27:47.006237: step: 472/529, loss: 0.057149793952703476 2023-01-21 12:27:48.178487: step: 476/529, loss: 0.07583598792552948 2023-01-21 12:27:49.355649: step: 480/529, loss: 0.029883574694395065 2023-01-21 12:27:50.551634: step: 484/529, loss: 0.02776813507080078 2023-01-21 12:27:51.730530: step: 488/529, loss: 0.004580259323120117 2023-01-21 12:27:52.932973: step: 492/529, loss: 0.019835853949189186 2023-01-21 12:27:54.193560: step: 496/529, loss: 0.0029943466652184725 2023-01-21 12:27:55.372803: step: 500/529, loss: 0.010996437631547451 2023-01-21 12:27:56.569835: step: 504/529, loss: 0.014081097207963467 2023-01-21 12:27:57.754343: step: 508/529, loss: 0.0014564513694494963 2023-01-21 12:27:59.017473: step: 512/529, loss: 0.00030517578125 2023-01-21 12:28:00.198030: step: 516/529, loss: 0.040574267506599426 2023-01-21 12:28:01.449924: step: 520/529, loss: 0.00870523415505886 2023-01-21 12:28:02.657730: step: 524/529, loss: 0.07299929112195969 2023-01-21 12:28:03.842649: step: 528/529, loss: 0.0005697250599041581 2023-01-21 12:28:05.008686: step: 532/529, loss: 0.006549358367919922 2023-01-21 12:28:06.218618: step: 536/529, loss: 0.007571410853415728 2023-01-21 12:28:07.442065: step: 540/529, loss: 0.03962249681353569 2023-01-21 12:28:08.650578: step: 544/529, loss: 0.00168523786123842 2023-01-21 12:28:09.859620: step: 548/529, loss: 0.04717826843261719 2023-01-21 12:28:11.080788: step: 552/529, loss: 5.130767749506049e-05 2023-01-21 12:28:12.260826: step: 556/529, loss: 0.002742862794548273 2023-01-21 12:28:13.470297: step: 560/529, loss: 0.052001193165779114 2023-01-21 12:28:14.662150: step: 564/529, loss: 0.04471731185913086 2023-01-21 12:28:15.851863: step: 568/529, loss: 0.018649626523256302 2023-01-21 12:28:17.055834: step: 572/529, loss: 0.0024925232864916325 2023-01-21 12:28:18.298533: step: 576/529, loss: 0.00191841134801507 2023-01-21 12:28:19.512790: step: 580/529, loss: 0.018245507031679153 2023-01-21 12:28:20.685686: step: 584/529, loss: 0.0001753807155182585 2023-01-21 12:28:21.897038: step: 588/529, loss: 0.1297696977853775 2023-01-21 12:28:23.085405: step: 592/529, loss: 0.0013309955829754472 2023-01-21 12:28:24.302247: step: 596/529, loss: 0.02813739888370037 2023-01-21 12:28:25.485527: step: 600/529, loss: 0.01161866169422865 2023-01-21 12:28:26.648735: step: 604/529, loss: 0.004985237494111061 2023-01-21 12:28:27.859035: step: 608/529, loss: 0.013849449343979359 2023-01-21 12:28:29.047730: step: 612/529, loss: 0.05014057457447052 2023-01-21 12:28:30.194949: step: 616/529, loss: 0.0005031585460528731 2023-01-21 12:28:31.335347: step: 620/529, loss: 0.00379600515589118 2023-01-21 12:28:32.505838: step: 624/529, loss: 0.0028319358825683594 2023-01-21 12:28:33.686416: step: 628/529, loss: 0.0027130127418786287 2023-01-21 12:28:34.886536: step: 632/529, loss: 0.029138755053281784 2023-01-21 12:28:36.114503: step: 636/529, loss: 0.0012501716846600175 2023-01-21 12:28:37.303043: step: 640/529, loss: 0.0007723808521404862 2023-01-21 12:28:38.508686: step: 644/529, loss: 0.012537766247987747 2023-01-21 12:28:39.712388: step: 648/529, loss: 0.003552055452018976 2023-01-21 12:28:40.892833: step: 652/529, loss: 0.0008320808410644531 2023-01-21 12:28:42.084564: step: 656/529, loss: 0.022772502154111862 2023-01-21 12:28:43.292547: step: 660/529, loss: 0.0010007858509197831 2023-01-21 12:28:44.481020: step: 664/529, loss: 0.537615954875946 2023-01-21 12:28:45.651840: step: 668/529, loss: 5.7220458984375e-06 2023-01-21 12:28:46.867165: step: 672/529, loss: 0.023189926519989967 2023-01-21 12:28:48.071514: step: 676/529, loss: 0.020878886803984642 2023-01-21 12:28:49.256001: step: 680/529, loss: 0.00383338937535882 2023-01-21 12:28:50.441381: step: 684/529, loss: 0.0005691051483154297 2023-01-21 12:28:51.648160: step: 688/529, loss: 0.12039170414209366 2023-01-21 12:28:52.804121: step: 692/529, loss: 0.0004737853887490928 2023-01-21 12:28:53.978215: step: 696/529, loss: 0.03276491165161133 2023-01-21 12:28:55.183085: step: 700/529, loss: 0.019254494458436966 2023-01-21 12:28:56.353601: step: 704/529, loss: 0.001360988593660295 2023-01-21 12:28:57.578048: step: 708/529, loss: 5.3894520533503965e-05 2023-01-21 12:28:58.781828: step: 712/529, loss: 0.004268837161362171 2023-01-21 12:29:00.013859: step: 716/529, loss: 0.011077880859375 2023-01-21 12:29:01.251574: step: 720/529, loss: 0.0006185531965456903 2023-01-21 12:29:02.413351: step: 724/529, loss: 0.018194008618593216 2023-01-21 12:29:03.559212: step: 728/529, loss: 0.00045809749281033874 2023-01-21 12:29:04.744060: step: 732/529, loss: 0.010229205712676048 2023-01-21 12:29:05.978157: step: 736/529, loss: 0.00107488629873842 2023-01-21 12:29:07.148921: step: 740/529, loss: 0.14359036087989807 2023-01-21 12:29:08.336445: step: 744/529, loss: 7.514953904319555e-05 2023-01-21 12:29:09.503343: step: 748/529, loss: 0.092382051050663 2023-01-21 12:29:10.674289: step: 752/529, loss: 0.0001911163271870464 2023-01-21 12:29:11.886166: step: 756/529, loss: 0.1937045156955719 2023-01-21 12:29:13.069341: step: 760/529, loss: 0.1064656674861908 2023-01-21 12:29:14.247538: step: 764/529, loss: 1.8215179807157256e-05 2023-01-21 12:29:15.440295: step: 768/529, loss: 0.030310630798339844 2023-01-21 12:29:16.637403: step: 772/529, loss: 0.012318420223891735 2023-01-21 12:29:17.821542: step: 776/529, loss: 0.19332891702651978 2023-01-21 12:29:18.986301: step: 780/529, loss: 0.005836868658661842 2023-01-21 12:29:20.158070: step: 784/529, loss: 0.002092361683025956 2023-01-21 12:29:21.371659: step: 788/529, loss: 0.00038909912109375 2023-01-21 12:29:22.595129: step: 792/529, loss: 0.037246037274599075 2023-01-21 12:29:23.777110: step: 796/529, loss: 0.02420949935913086 2023-01-21 12:29:24.919428: step: 800/529, loss: 0.01276335772126913 2023-01-21 12:29:26.101168: step: 804/529, loss: 0.045938681811094284 2023-01-21 12:29:27.282609: step: 808/529, loss: 0.0020398139022290707 2023-01-21 12:29:28.471179: step: 812/529, loss: 0.004785728175193071 2023-01-21 12:29:29.647517: step: 816/529, loss: 0.020598506554961205 2023-01-21 12:29:30.795295: step: 820/529, loss: 0.005712604615837336 2023-01-21 12:29:31.964624: step: 824/529, loss: 0.004669592250138521 2023-01-21 12:29:33.136673: step: 828/529, loss: 0.002628326416015625 2023-01-21 12:29:34.295216: step: 832/529, loss: 0.00013209581084083766 2023-01-21 12:29:35.477526: step: 836/529, loss: 0.01651954837143421 2023-01-21 12:29:36.668283: step: 840/529, loss: 0.0256989486515522 2023-01-21 12:29:37.885328: step: 844/529, loss: 0.005681419279426336 2023-01-21 12:29:39.108409: step: 848/529, loss: 0.004430532455444336 2023-01-21 12:29:40.320316: step: 852/529, loss: 0.021818209439516068 2023-01-21 12:29:41.497406: step: 856/529, loss: 0.001922511961311102 2023-01-21 12:29:42.713855: step: 860/529, loss: 0.0014760971534997225 2023-01-21 12:29:43.929680: step: 864/529, loss: 0.0018540859455242753 2023-01-21 12:29:45.108740: step: 868/529, loss: 0.7207501530647278 2023-01-21 12:29:46.303818: step: 872/529, loss: 0.010114670731127262 2023-01-21 12:29:47.474023: step: 876/529, loss: 0.0010590553283691406 2023-01-21 12:29:48.634007: step: 880/529, loss: 0.0004748344363179058 2023-01-21 12:29:49.822840: step: 884/529, loss: 0.03907651826739311 2023-01-21 12:29:51.005698: step: 888/529, loss: 0.0006536960718221962 2023-01-21 12:29:52.216220: step: 892/529, loss: 0.10833549499511719 2023-01-21 12:29:53.407608: step: 896/529, loss: 0.0037919997703284025 2023-01-21 12:29:54.613177: step: 900/529, loss: 0.006622600834816694 2023-01-21 12:29:55.820403: step: 904/529, loss: 0.0005246162181720138 2023-01-21 12:29:57.039873: step: 908/529, loss: 0.02417621575295925 2023-01-21 12:29:58.217222: step: 912/529, loss: 0.02222600020468235 2023-01-21 12:29:59.409156: step: 916/529, loss: 0.0001682281435932964 2023-01-21 12:30:00.623503: step: 920/529, loss: 6.418228440452367e-05 2023-01-21 12:30:01.796812: step: 924/529, loss: 0.05570679157972336 2023-01-21 12:30:03.003794: step: 928/529, loss: 0.0003883361932821572 2023-01-21 12:30:04.163891: step: 932/529, loss: 0.00046939851017668843 2023-01-21 12:30:05.374838: step: 936/529, loss: 0.014226722531020641 2023-01-21 12:30:06.568293: step: 940/529, loss: 0.028089523315429688 2023-01-21 12:30:07.741705: step: 944/529, loss: 0.017992019653320312 2023-01-21 12:30:08.947764: step: 948/529, loss: 0.042827751487493515 2023-01-21 12:30:10.116916: step: 952/529, loss: 0.0011246681679040194 2023-01-21 12:30:11.305082: step: 956/529, loss: 0.027071570977568626 2023-01-21 12:30:12.475446: step: 960/529, loss: 0.002462100936099887 2023-01-21 12:30:13.685831: step: 964/529, loss: 0.001001596450805664 2023-01-21 12:30:14.894305: step: 968/529, loss: 0.002326583955436945 2023-01-21 12:30:16.094900: step: 972/529, loss: 0.006061935797333717 2023-01-21 12:30:17.252260: step: 976/529, loss: 0.0010293960804119706 2023-01-21 12:30:18.457146: step: 980/529, loss: 0.028292465955018997 2023-01-21 12:30:19.664548: step: 984/529, loss: 0.018955135717988014 2023-01-21 12:30:20.837169: step: 988/529, loss: 5.9795380366267636e-05 2023-01-21 12:30:22.066565: step: 992/529, loss: 0.00214385986328125 2023-01-21 12:30:23.230317: step: 996/529, loss: 0.010997582226991653 2023-01-21 12:30:24.452553: step: 1000/529, loss: 0.051388930529356 2023-01-21 12:30:25.617591: step: 1004/529, loss: 0.0022928237449377775 2023-01-21 12:30:26.818016: step: 1008/529, loss: 0.0025238036178052425 2023-01-21 12:30:27.977074: step: 1012/529, loss: 0.01413431204855442 2023-01-21 12:30:29.198837: step: 1016/529, loss: 0.0011574745876714587 2023-01-21 12:30:30.394034: step: 1020/529, loss: 5.9318539570085704e-05 2023-01-21 12:30:31.569836: step: 1024/529, loss: 0.027594758197665215 2023-01-21 12:30:32.751281: step: 1028/529, loss: 1.8978118532686494e-05 2023-01-21 12:30:33.938122: step: 1032/529, loss: 0.002118015196174383 2023-01-21 12:30:35.120742: step: 1036/529, loss: 5.7220458984375e-06 2023-01-21 12:30:36.324966: step: 1040/529, loss: 0.00027694704476743937 2023-01-21 12:30:37.524948: step: 1044/529, loss: 0.00014371873112395406 2023-01-21 12:30:38.767512: step: 1048/529, loss: 0.030681420117616653 2023-01-21 12:30:39.953283: step: 1052/529, loss: 0.02742939069867134 2023-01-21 12:30:41.161106: step: 1056/529, loss: 0.027376748621463776 2023-01-21 12:30:42.348056: step: 1060/529, loss: 0.00011324882507324219 2023-01-21 12:30:43.547234: step: 1064/529, loss: 0.00013806819333694875 2023-01-21 12:30:44.800181: step: 1068/529, loss: 2.09808349609375e-05 2023-01-21 12:30:45.971896: step: 1072/529, loss: 0.0022203922271728516 2023-01-21 12:30:47.137820: step: 1076/529, loss: 0.0006958008161745965 2023-01-21 12:30:48.357155: step: 1080/529, loss: 0.0008769988780841231 2023-01-21 12:30:49.534760: step: 1084/529, loss: 0.0034041404724121094 2023-01-21 12:30:50.721130: step: 1088/529, loss: 0.010478496551513672 2023-01-21 12:30:51.878239: step: 1092/529, loss: 0.0035492898896336555 2023-01-21 12:30:53.033125: step: 1096/529, loss: 0.03338112682104111 2023-01-21 12:30:54.237350: step: 1100/529, loss: 0.0004745483456645161 2023-01-21 12:30:55.420444: step: 1104/529, loss: 0.009274768643081188 2023-01-21 12:30:56.595087: step: 1108/529, loss: 0.0006011963123455644 2023-01-21 12:30:57.812697: step: 1112/529, loss: 0.001428794814273715 2023-01-21 12:30:58.987884: step: 1116/529, loss: 0.008805084973573685 2023-01-21 12:31:00.183895: step: 1120/529, loss: 0.00511245708912611 2023-01-21 12:31:01.366288: step: 1124/529, loss: 0.0023772239219397306 2023-01-21 12:31:02.585920: step: 1128/529, loss: 0.0047760009765625 2023-01-21 12:31:03.767553: step: 1132/529, loss: 0.05124321207404137 2023-01-21 12:31:04.924449: step: 1136/529, loss: 0.027296971529722214 2023-01-21 12:31:06.094191: step: 1140/529, loss: 0.015580369159579277 2023-01-21 12:31:07.273042: step: 1144/529, loss: 1.905228853225708 2023-01-21 12:31:08.481754: step: 1148/529, loss: 0.004398536402732134 2023-01-21 12:31:09.686487: step: 1152/529, loss: 0.0006052016979083419 2023-01-21 12:31:10.834865: step: 1156/529, loss: 0.08208908885717392 2023-01-21 12:31:12.022218: step: 1160/529, loss: 0.021008683368563652 2023-01-21 12:31:13.207074: step: 1164/529, loss: 0.013992881402373314 2023-01-21 12:31:14.401468: step: 1168/529, loss: 0.02656869776546955 2023-01-21 12:31:15.570461: step: 1172/529, loss: 6.647109694313258e-05 2023-01-21 12:31:16.760906: step: 1176/529, loss: 0.0025585652329027653 2023-01-21 12:31:17.968052: step: 1180/529, loss: 0.06045074760913849 2023-01-21 12:31:19.122493: step: 1184/529, loss: 0.002610874129459262 2023-01-21 12:31:20.329221: step: 1188/529, loss: 0.7269910573959351 2023-01-21 12:31:21.551426: step: 1192/529, loss: 0.1248149424791336 2023-01-21 12:31:22.736353: step: 1196/529, loss: 0.003365802811458707 2023-01-21 12:31:23.919524: step: 1200/529, loss: 0.001822996186092496 2023-01-21 12:31:25.134104: step: 1204/529, loss: 0.02316865883767605 2023-01-21 12:31:26.345687: step: 1208/529, loss: 0.0001424789516022429 2023-01-21 12:31:27.628590: step: 1212/529, loss: 0.0001937389315571636 2023-01-21 12:31:28.795858: step: 1216/529, loss: 0.006111335940659046 2023-01-21 12:31:30.040464: step: 1220/529, loss: 0.00031118394690565765 2023-01-21 12:31:31.284085: step: 1224/529, loss: 0.020068645477294922 2023-01-21 12:31:32.446756: step: 1228/529, loss: 0.01991257630288601 2023-01-21 12:31:33.652717: step: 1232/529, loss: 0.039063453674316406 2023-01-21 12:31:34.801325: step: 1236/529, loss: 2.7942656743107364e-05 2023-01-21 12:31:35.976003: step: 1240/529, loss: 0.06479320675134659 2023-01-21 12:31:37.191036: step: 1244/529, loss: 0.00020675660925917327 2023-01-21 12:31:38.336552: step: 1248/529, loss: 0.0023190020583570004 2023-01-21 12:31:39.537257: step: 1252/529, loss: 0.01610546186566353 2023-01-21 12:31:40.766977: step: 1256/529, loss: 0.014467049390077591 2023-01-21 12:31:41.920869: step: 1260/529, loss: 0.0018232346046715975 2023-01-21 12:31:43.092431: step: 1264/529, loss: 0.01195917185395956 2023-01-21 12:31:44.327335: step: 1268/529, loss: 0.0013654709327965975 2023-01-21 12:31:45.524316: step: 1272/529, loss: 0.029822159558534622 2023-01-21 12:31:46.690230: step: 1276/529, loss: 0.0003397941472940147 2023-01-21 12:31:47.987323: step: 1280/529, loss: 0.00039768218994140625 2023-01-21 12:31:49.183493: step: 1284/529, loss: 0.00290088658221066 2023-01-21 12:31:50.376982: step: 1288/529, loss: 0.004054450895637274 2023-01-21 12:31:51.572604: step: 1292/529, loss: 0.0015881540020927787 2023-01-21 12:31:52.758367: step: 1296/529, loss: 0.08943118900060654 2023-01-21 12:31:53.949138: step: 1300/529, loss: 0.00017538070096634328 2023-01-21 12:31:55.168666: step: 1304/529, loss: 0.15942907333374023 2023-01-21 12:31:56.351948: step: 1308/529, loss: 0.0014765739906579256 2023-01-21 12:31:57.570567: step: 1312/529, loss: 0.01698584482073784 2023-01-21 12:31:58.770008: step: 1316/529, loss: 0.014727783389389515 2023-01-21 12:31:59.961544: step: 1320/529, loss: 0.0006592273712158203 2023-01-21 12:32:01.244392: step: 1324/529, loss: 0.00010104179818881676 2023-01-21 12:32:02.425928: step: 1328/529, loss: 0.009118080139160156 2023-01-21 12:32:03.588255: step: 1332/529, loss: 0.03366592153906822 2023-01-21 12:32:04.738615: step: 1336/529, loss: 0.0008112907526083291 2023-01-21 12:32:05.900952: step: 1340/529, loss: 0.018190480768680573 2023-01-21 12:32:07.113494: step: 1344/529, loss: 0.037294767796993256 2023-01-21 12:32:08.296393: step: 1348/529, loss: 0.029923629015684128 2023-01-21 12:32:09.451483: step: 1352/529, loss: 0.01628437079489231 2023-01-21 12:32:10.704657: step: 1356/529, loss: 0.0019905089866369963 2023-01-21 12:32:11.889906: step: 1360/529, loss: 0.006911182310432196 2023-01-21 12:32:13.069876: step: 1364/529, loss: 0.03901968151330948 2023-01-21 12:32:14.250134: step: 1368/529, loss: 0.02261362038552761 2023-01-21 12:32:15.469254: step: 1372/529, loss: 0.0022178650833666325 2023-01-21 12:32:16.659970: step: 1376/529, loss: 0.0070931436493992805 2023-01-21 12:32:17.870267: step: 1380/529, loss: 0.04878511279821396 2023-01-21 12:32:19.062496: step: 1384/529, loss: 0.01128082349896431 2023-01-21 12:32:20.232014: step: 1388/529, loss: 0.0001964569091796875 2023-01-21 12:32:21.409861: step: 1392/529, loss: 0.0005277633899822831 2023-01-21 12:32:22.623718: step: 1396/529, loss: 0.003966141026467085 2023-01-21 12:32:23.790262: step: 1400/529, loss: 0.021597862243652344 2023-01-21 12:32:25.002125: step: 1404/529, loss: 0.013552665710449219 2023-01-21 12:32:26.187580: step: 1408/529, loss: 0.0006745815626345575 2023-01-21 12:32:27.371369: step: 1412/529, loss: 0.019646883010864258 2023-01-21 12:32:28.612161: step: 1416/529, loss: 0.010751819238066673 2023-01-21 12:32:29.810404: step: 1420/529, loss: 0.017238998785614967 2023-01-21 12:32:30.972386: step: 1424/529, loss: 3.900528099620715e-05 2023-01-21 12:32:32.182805: step: 1428/529, loss: 0.0016060828929767013 2023-01-21 12:32:33.352145: step: 1432/529, loss: 0.002705526538193226 2023-01-21 12:32:34.563123: step: 1436/529, loss: 0.002124214079231024 2023-01-21 12:32:35.806083: step: 1440/529, loss: 0.00521926861256361 2023-01-21 12:32:36.985243: step: 1444/529, loss: 0.24625560641288757 2023-01-21 12:32:38.164680: step: 1448/529, loss: 0.0007156372303143144 2023-01-21 12:32:39.360055: step: 1452/529, loss: 0.0061820982955396175 2023-01-21 12:32:40.554215: step: 1456/529, loss: 0.0010063170921057463 2023-01-21 12:32:41.750395: step: 1460/529, loss: 0.3570823669433594 2023-01-21 12:32:42.943644: step: 1464/529, loss: 0.0005876541254110634 2023-01-21 12:32:44.132336: step: 1468/529, loss: 0.0010268212063238025 2023-01-21 12:32:45.330188: step: 1472/529, loss: 0.00039510728674940765 2023-01-21 12:32:46.476162: step: 1476/529, loss: 0.00904550589621067 2023-01-21 12:32:47.667802: step: 1480/529, loss: 0.013424396514892578 2023-01-21 12:32:48.856320: step: 1484/529, loss: 0.001997566083446145 2023-01-21 12:32:50.053197: step: 1488/529, loss: 0.01690368726849556 2023-01-21 12:32:51.249325: step: 1492/529, loss: 0.01404027920216322 2023-01-21 12:32:52.469482: step: 1496/529, loss: 0.00025777818518690765 2023-01-21 12:32:53.629233: step: 1500/529, loss: 0.0012628554832190275 2023-01-21 12:32:54.783040: step: 1504/529, loss: 0.003780364990234375 2023-01-21 12:32:55.993520: step: 1508/529, loss: 0.0034401954617351294 2023-01-21 12:32:57.220401: step: 1512/529, loss: 0.00096044538076967 2023-01-21 12:32:58.394949: step: 1516/529, loss: 0.017782973125576973 2023-01-21 12:32:59.571626: step: 1520/529, loss: 0.021206380799412727 2023-01-21 12:33:00.797411: step: 1524/529, loss: 0.029020071029663086 2023-01-21 12:33:01.975862: step: 1528/529, loss: 0.0018839358817785978 2023-01-21 12:33:03.212014: step: 1532/529, loss: 0.001266098115593195 2023-01-21 12:33:04.373739: step: 1536/529, loss: 0.004473114386200905 2023-01-21 12:33:05.567436: step: 1540/529, loss: 0.07688990235328674 2023-01-21 12:33:06.794866: step: 1544/529, loss: 0.0014057159423828125 2023-01-21 12:33:07.988545: step: 1548/529, loss: 0.03889627382159233 2023-01-21 12:33:09.198892: step: 1552/529, loss: 2.4414061044808477e-05 2023-01-21 12:33:10.444262: step: 1556/529, loss: 0.01679687574505806 2023-01-21 12:33:11.637632: step: 1560/529, loss: 0.003276014467701316 2023-01-21 12:33:12.850711: step: 1564/529, loss: 0.012526988983154297 2023-01-21 12:33:14.036287: step: 1568/529, loss: 0.0015130043029785156 2023-01-21 12:33:15.191632: step: 1572/529, loss: 0.00183448800817132 2023-01-21 12:33:16.352766: step: 1576/529, loss: 0.02165050618350506 2023-01-21 12:33:17.519996: step: 1580/529, loss: 0.01760711707174778 2023-01-21 12:33:18.730510: step: 1584/529, loss: 0.007588768377900124 2023-01-21 12:33:19.931380: step: 1588/529, loss: 1.71661376953125e-05 2023-01-21 12:33:21.112815: step: 1592/529, loss: 0.008407974615693092 2023-01-21 12:33:22.272369: step: 1596/529, loss: 0.0005439758533611894 2023-01-21 12:33:23.440441: step: 1600/529, loss: 0.03986234590411186 2023-01-21 12:33:24.687013: step: 1604/529, loss: 0.000877761805895716 2023-01-21 12:33:25.874142: step: 1608/529, loss: 0.016019631177186966 2023-01-21 12:33:27.052258: step: 1612/529, loss: 0.002694988390430808 2023-01-21 12:33:28.243431: step: 1616/529, loss: 0.0005079269176349044 2023-01-21 12:33:29.411800: step: 1620/529, loss: 0.0021526336204260588 2023-01-21 12:33:30.662225: step: 1624/529, loss: 0.035662271082401276 2023-01-21 12:33:31.896438: step: 1628/529, loss: 1.068115216185106e-05 2023-01-21 12:33:33.099784: step: 1632/529, loss: 0.00200653076171875 2023-01-21 12:33:34.318058: step: 1636/529, loss: 0.013348388485610485 2023-01-21 12:33:35.523836: step: 1640/529, loss: 0.6704956889152527 2023-01-21 12:33:36.702819: step: 1644/529, loss: 0.007969331927597523 2023-01-21 12:33:37.889082: step: 1648/529, loss: 0.0027492523659020662 2023-01-21 12:33:39.034209: step: 1652/529, loss: 0.002999591873958707 2023-01-21 12:33:40.212243: step: 1656/529, loss: 0.047217655926942825 2023-01-21 12:33:41.406122: step: 1660/529, loss: 0.07541093975305557 2023-01-21 12:33:42.618536: step: 1664/529, loss: 0.0128333093598485 2023-01-21 12:33:43.820031: step: 1668/529, loss: 0.018995190039277077 2023-01-21 12:33:45.028867: step: 1672/529, loss: 0.00039949416532181203 2023-01-21 12:33:46.209105: step: 1676/529, loss: 0.0001434326113667339 2023-01-21 12:33:47.428377: step: 1680/529, loss: 0.011221122927963734 2023-01-21 12:33:48.603479: step: 1684/529, loss: 0.0006533623090945184 2023-01-21 12:33:49.810361: step: 1688/529, loss: 0.026187706738710403 2023-01-21 12:33:50.982973: step: 1692/529, loss: 0.051987841725349426 2023-01-21 12:33:52.150375: step: 1696/529, loss: 0.0020817758049815893 2023-01-21 12:33:53.339330: step: 1700/529, loss: 0.014611244201660156 2023-01-21 12:33:54.499617: step: 1704/529, loss: 0.015263748355209827 2023-01-21 12:33:55.660266: step: 1708/529, loss: 0.0015641212230548263 2023-01-21 12:33:56.850070: step: 1712/529, loss: 0.003006553743034601 2023-01-21 12:33:58.052774: step: 1716/529, loss: 0.04374561086297035 2023-01-21 12:33:59.222837: step: 1720/529, loss: 0.0007204055436886847 2023-01-21 12:34:00.410606: step: 1724/529, loss: 0.003359937807545066 2023-01-21 12:34:01.584343: step: 1728/529, loss: 0.005859375465661287 2023-01-21 12:34:02.852427: step: 1732/529, loss: 0.03243722766637802 2023-01-21 12:34:04.049601: step: 1736/529, loss: 7.114410254871473e-05 2023-01-21 12:34:05.231568: step: 1740/529, loss: 0.1302306205034256 2023-01-21 12:34:06.396457: step: 1744/529, loss: 0.03613262251019478 2023-01-21 12:34:07.594152: step: 1748/529, loss: 0.0005180359003134072 2023-01-21 12:34:08.795970: step: 1752/529, loss: 0.007503319066017866 2023-01-21 12:34:10.036456: step: 1756/529, loss: 0.04974937438964844 2023-01-21 12:34:11.216454: step: 1760/529, loss: 0.01761932484805584 2023-01-21 12:34:12.381128: step: 1764/529, loss: 0.0006916046258993447 2023-01-21 12:34:13.573012: step: 1768/529, loss: 0.005828857421875 2023-01-21 12:34:14.797243: step: 1772/529, loss: 0.006870460696518421 2023-01-21 12:34:15.959611: step: 1776/529, loss: 0.0004364967462606728 2023-01-21 12:34:17.150463: step: 1780/529, loss: 0.0002768516424112022 2023-01-21 12:34:18.320660: step: 1784/529, loss: 0.0021639822516590357 2023-01-21 12:34:19.522056: step: 1788/529, loss: 0.006115913391113281 2023-01-21 12:34:20.679710: step: 1792/529, loss: 0.0001923561212606728 2023-01-21 12:34:21.890153: step: 1796/529, loss: 0.00017061234393622726 2023-01-21 12:34:23.057073: step: 1800/529, loss: 0.006600570864975452 2023-01-21 12:34:24.262207: step: 1804/529, loss: 0.0006340027321130037 2023-01-21 12:34:25.442353: step: 1808/529, loss: 0.0011976242531090975 2023-01-21 12:34:26.687561: step: 1812/529, loss: 8.459091623080894e-05 2023-01-21 12:34:27.883223: step: 1816/529, loss: 0.03202095255255699 2023-01-21 12:34:29.079688: step: 1820/529, loss: 0.019614791497588158 2023-01-21 12:34:30.247331: step: 1824/529, loss: 3.604888843256049e-05 2023-01-21 12:34:31.479674: step: 1828/529, loss: 0.00018167495727539062 2023-01-21 12:34:32.713583: step: 1832/529, loss: 0.0374908447265625 2023-01-21 12:34:33.915002: step: 1836/529, loss: 0.017992116510868073 2023-01-21 12:34:35.130367: step: 1840/529, loss: 0.04459953308105469 2023-01-21 12:34:36.340137: step: 1844/529, loss: 0.007378006353974342 2023-01-21 12:34:37.609124: step: 1848/529, loss: 0.0009414672967977822 2023-01-21 12:34:38.795550: step: 1852/529, loss: 0.005203342530876398 2023-01-21 12:34:39.988050: step: 1856/529, loss: 1.5543813705444336 2023-01-21 12:34:41.218469: step: 1860/529, loss: 0.0008533477666787803 2023-01-21 12:34:42.384055: step: 1864/529, loss: 0.004583168309181929 2023-01-21 12:34:43.578647: step: 1868/529, loss: 0.00023059846716932952 2023-01-21 12:34:44.738340: step: 1872/529, loss: 0.06282377243041992 2023-01-21 12:34:45.943306: step: 1876/529, loss: 0.0007351875538006425 2023-01-21 12:34:47.121837: step: 1880/529, loss: 0.0001617431698832661 2023-01-21 12:34:48.315716: step: 1884/529, loss: -4.863738922722405e-06 2023-01-21 12:34:49.474711: step: 1888/529, loss: 0.013444977812469006 2023-01-21 12:34:50.666241: step: 1892/529, loss: 0.06164093315601349 2023-01-21 12:34:51.834180: step: 1896/529, loss: 0.009922981262207031 2023-01-21 12:34:53.071816: step: 1900/529, loss: 0.026669027283787727 2023-01-21 12:34:54.279407: step: 1904/529, loss: 0.012551403604447842 2023-01-21 12:34:55.505616: step: 1908/529, loss: 0.2914290428161621 2023-01-21 12:34:56.687298: step: 1912/529, loss: 0.012662124820053577 2023-01-21 12:34:57.856283: step: 1916/529, loss: 0.0052946568466722965 2023-01-21 12:34:59.014825: step: 1920/529, loss: 0.07859744876623154 2023-01-21 12:35:00.198138: step: 1924/529, loss: 0.046630859375 2023-01-21 12:35:01.385722: step: 1928/529, loss: 0.0011096000671386719 2023-01-21 12:35:02.555413: step: 1932/529, loss: 0.013671446591615677 2023-01-21 12:35:03.785799: step: 1936/529, loss: 0.03882698714733124 2023-01-21 12:35:04.988587: step: 1940/529, loss: 0.016082847490906715 2023-01-21 12:35:06.179544: step: 1944/529, loss: 0.020233154296875 2023-01-21 12:35:07.413762: step: 1948/529, loss: 0.0036792755126953125 2023-01-21 12:35:08.593788: step: 1952/529, loss: 0.00017132758512161672 2023-01-21 12:35:09.785095: step: 1956/529, loss: 0.00024623872013762593 2023-01-21 12:35:11.014731: step: 1960/529, loss: 0.02321949042379856 2023-01-21 12:35:12.236749: step: 1964/529, loss: 0.008112525567412376 2023-01-21 12:35:13.401939: step: 1968/529, loss: 0.008886623196303844 2023-01-21 12:35:14.556447: step: 1972/529, loss: 0.000754880893509835 2023-01-21 12:35:15.757928: step: 1976/529, loss: 0.0007356643909588456 2023-01-21 12:35:16.977928: step: 1980/529, loss: 0.0013558388454839587 2023-01-21 12:35:18.140681: step: 1984/529, loss: 0.008496666327118874 2023-01-21 12:35:19.319469: step: 1988/529, loss: 0.019655991345643997 2023-01-21 12:35:20.496677: step: 1992/529, loss: 0.03364581987261772 2023-01-21 12:35:21.673084: step: 1996/529, loss: 0.00404667854309082 2023-01-21 12:35:22.845583: step: 2000/529, loss: 0.001017761300317943 2023-01-21 12:35:24.054995: step: 2004/529, loss: 0.001706790877506137 2023-01-21 12:35:25.254111: step: 2008/529, loss: 0.006466579623520374 2023-01-21 12:35:26.454366: step: 2012/529, loss: 0.03728752210736275 2023-01-21 12:35:27.645973: step: 2016/529, loss: 0.09564084559679031 2023-01-21 12:35:28.841290: step: 2020/529, loss: 0.0026894567999988794 2023-01-21 12:35:30.021554: step: 2024/529, loss: 0.0015523910988122225 2023-01-21 12:35:31.189577: step: 2028/529, loss: 0.001751708914525807 2023-01-21 12:35:32.371210: step: 2032/529, loss: 0.002993297530338168 2023-01-21 12:35:33.560809: step: 2036/529, loss: 0.0010641098488122225 2023-01-21 12:35:34.747582: step: 2040/529, loss: 0.10790844261646271 2023-01-21 12:35:35.911924: step: 2044/529, loss: 0.0006014824029989541 2023-01-21 12:35:37.087643: step: 2048/529, loss: 0.00023059846716932952 2023-01-21 12:35:38.270549: step: 2052/529, loss: 0.0009954452980309725 2023-01-21 12:35:39.445604: step: 2056/529, loss: 0.04278812184929848 2023-01-21 12:35:40.630703: step: 2060/529, loss: 0.0279101375490427 2023-01-21 12:35:41.838045: step: 2064/529, loss: 0.0009799957042559981 2023-01-21 12:35:43.042149: step: 2068/529, loss: 0.5962082147598267 2023-01-21 12:35:44.204755: step: 2072/529, loss: 0.0005461692926473916 2023-01-21 12:35:45.419888: step: 2076/529, loss: 0.00013847352238371968 2023-01-21 12:35:46.596950: step: 2080/529, loss: 0.017247533425688744 2023-01-21 12:35:47.857918: step: 2084/529, loss: 0.02177906036376953 2023-01-21 12:35:49.010976: step: 2088/529, loss: 0.0005646705394610763 2023-01-21 12:35:50.172408: step: 2092/529, loss: 0.0012795449001714587 2023-01-21 12:35:51.361483: step: 2096/529, loss: 0.005169391632080078 2023-01-21 12:35:52.594474: step: 2100/529, loss: 0.005059337709099054 2023-01-21 12:35:53.775904: step: 2104/529, loss: 0.0015271187294274569 2023-01-21 12:35:54.946893: step: 2108/529, loss: 0.00047559739323332906 2023-01-21 12:35:56.103033: step: 2112/529, loss: 0.004547024145722389 2023-01-21 12:35:57.255317: step: 2116/529, loss: 0.004264927469193935 ================================================== Loss: 0.035 -------------------- Dev: {'event': {'p': 0.5826693227091634, 'r': 0.7789613848202397, 'f1': 0.6666666666666667}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Test: {'event': {'p': 0.6177145479577788, 'r': 0.8064709406830437, 'f1': 0.6995841995841996}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Chinese: {'event': {'p': 0.5487804878048781, 'r': 0.8333333333333334, 'f1': 0.6617647058823529}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Korean: {'event': {'p': 0.5245901639344263, 'r': 0.5079365079365079, 'f1': 0.5161290322580645}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} 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': 20} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 12:36:40.758122: step: 4/529, loss: 0.003536258591338992 2023-01-21 12:36:41.951072: step: 8/529, loss: 0.00401649484410882 2023-01-21 12:36:43.141033: step: 12/529, loss: 0.009791756048798561 2023-01-21 12:36:44.321274: step: 16/529, loss: 0.03188343346118927 2023-01-21 12:36:45.483485: step: 20/529, loss: 5.500912811839953e-05 2023-01-21 12:36:46.670740: step: 24/529, loss: 0.008395016193389893 2023-01-21 12:36:47.855460: step: 28/529, loss: 0.0025276183150708675 2023-01-21 12:36:49.064272: step: 32/529, loss: 0.02211780659854412 2023-01-21 12:36:50.254009: step: 36/529, loss: 0.001338195870630443 2023-01-21 12:36:51.425957: step: 40/529, loss: 0.0004429817199707031 2023-01-21 12:36:52.650311: step: 44/529, loss: 0.022576140239834785 2023-01-21 12:36:53.843147: step: 48/529, loss: 0.01085510291159153 2023-01-21 12:36:55.038928: step: 52/529, loss: 1.411438006471144e-05 2023-01-21 12:36:56.233276: step: 56/529, loss: 0.043985746800899506 2023-01-21 12:36:57.430823: step: 60/529, loss: 0.0011208534706383944 2023-01-21 12:36:58.620431: step: 64/529, loss: 0.02612333372235298 2023-01-21 12:36:59.836969: step: 68/529, loss: 0.005660533905029297 2023-01-21 12:37:01.050301: step: 72/529, loss: 0.015165328979492188 2023-01-21 12:37:02.234486: step: 76/529, loss: 0.0007496834150515497 2023-01-21 12:37:03.427293: step: 80/529, loss: 0.00016832351684570312 2023-01-21 12:37:04.644979: step: 84/529, loss: 0.011369897052645683 2023-01-21 12:37:05.801479: step: 88/529, loss: 0.00012662410154007375 2023-01-21 12:37:06.981166: step: 92/529, loss: 0.0035753250122070312 2023-01-21 12:37:08.174054: step: 96/529, loss: 0.00040302277193404734 2023-01-21 12:37:09.350838: step: 100/529, loss: 0.0001674652157817036 2023-01-21 12:37:10.540464: step: 104/529, loss: 0.00047588348388671875 2023-01-21 12:37:11.769646: step: 108/529, loss: 0.029589273035526276 2023-01-21 12:37:12.914700: step: 112/529, loss: 4.854202416026965e-05 2023-01-21 12:37:14.083829: step: 116/529, loss: 0.005323315039277077 2023-01-21 12:37:15.268060: step: 120/529, loss: 0.0034019469749182463 2023-01-21 12:37:16.457159: step: 124/529, loss: 0.00132579798810184 2023-01-21 12:37:17.662395: step: 128/529, loss: 0.0004345893976278603 2023-01-21 12:37:18.836127: step: 132/529, loss: 0.00021781922259833664 2023-01-21 12:37:20.025207: step: 136/529, loss: 3.6239621294953395e-06 2023-01-21 12:37:21.247856: step: 140/529, loss: 0.004038429353386164 2023-01-21 12:37:22.446822: step: 144/529, loss: 2.6416779292048886e-05 2023-01-21 12:37:23.640078: step: 148/529, loss: 0.03313455358147621 2023-01-21 12:37:24.786647: step: 152/529, loss: 0.0026945113204419613 2023-01-21 12:37:25.986472: step: 156/529, loss: 0.0006809235201217234 2023-01-21 12:37:27.130374: step: 160/529, loss: 0.01350097730755806 2023-01-21 12:37:28.294816: step: 164/529, loss: 0.00012521744065452367 2023-01-21 12:37:29.450542: step: 168/529, loss: 0.011375904083251953 2023-01-21 12:37:30.641689: step: 172/529, loss: 0.010783863253891468 2023-01-21 12:37:31.887504: step: 176/529, loss: 0.005101298913359642 2023-01-21 12:37:33.085158: step: 180/529, loss: 0.03699932247400284 2023-01-21 12:37:34.270472: step: 184/529, loss: 0.04721546173095703 2023-01-21 12:37:35.482715: step: 188/529, loss: 0.0003075599961448461 2023-01-21 12:37:36.772172: step: 192/529, loss: 0.008887100033462048 2023-01-21 12:37:37.949949: step: 196/529, loss: 0.0007946014520712197 2023-01-21 12:37:39.101312: step: 200/529, loss: 0.00021104812913108617 2023-01-21 12:37:40.295936: step: 204/529, loss: 0.00012912749662064016 2023-01-21 12:37:41.471696: step: 208/529, loss: 7.162988367781509e-06 2023-01-21 12:37:42.671916: step: 212/529, loss: 0.0015114308334887028 2023-01-21 12:37:43.846303: step: 216/529, loss: 0.019399071112275124 2023-01-21 12:37:45.035580: step: 220/529, loss: 0.006062602624297142 2023-01-21 12:37:46.196609: step: 224/529, loss: 0.0007863998180255294 2023-01-21 12:37:47.406068: step: 228/529, loss: 0.0009471893426962197 2023-01-21 12:37:48.589877: step: 232/529, loss: 0.025431107729673386 2023-01-21 12:37:49.796960: step: 236/529, loss: 0.03290128707885742 2023-01-21 12:37:50.960642: step: 240/529, loss: 0.02950897254049778 2023-01-21 12:37:52.164183: step: 244/529, loss: 0.0008078575483523309 2023-01-21 12:37:53.376236: step: 248/529, loss: 0.0052238465286791325 2023-01-21 12:37:54.539173: step: 252/529, loss: 0.0016533373855054379 2023-01-21 12:37:55.707214: step: 256/529, loss: 0.009891225025057793 2023-01-21 12:37:56.862837: step: 260/529, loss: 0.00014419555373024195 2023-01-21 12:37:58.092933: step: 264/529, loss: 0.03104419820010662 2023-01-21 12:37:59.304175: step: 268/529, loss: 0.013837910257279873 2023-01-21 12:38:00.486643: step: 272/529, loss: 0.00020356177992653102 2023-01-21 12:38:01.631003: step: 276/529, loss: 0.0030412673950195312 2023-01-21 12:38:02.841762: step: 280/529, loss: 0.004071045201271772 2023-01-21 12:38:04.062674: step: 284/529, loss: 0.012658214196562767 2023-01-21 12:38:05.214356: step: 288/529, loss: 0.0003450393851380795 2023-01-21 12:38:06.372817: step: 292/529, loss: 0.10960173606872559 2023-01-21 12:38:07.534241: step: 296/529, loss: 8.034706115722656e-05 2023-01-21 12:38:08.719161: step: 300/529, loss: 0.0008501053089275956 2023-01-21 12:38:09.897155: step: 304/529, loss: 0.0015618324978277087 2023-01-21 12:38:11.099882: step: 308/529, loss: 0.00710296630859375 2023-01-21 12:38:12.353704: step: 312/529, loss: 6.50405854685232e-05 2023-01-21 12:38:13.534602: step: 316/529, loss: 0.1365833580493927 2023-01-21 12:38:14.722257: step: 320/529, loss: 0.0015281677478924394 2023-01-21 12:38:15.954876: step: 324/529, loss: 0.005007743835449219 2023-01-21 12:38:17.148819: step: 328/529, loss: 0.0002971649228129536 2023-01-21 12:38:18.355987: step: 332/529, loss: 0.004324436187744141 2023-01-21 12:38:19.543205: step: 336/529, loss: 0.0028727531898766756 2023-01-21 12:38:20.708458: step: 340/529, loss: 0.002777671907097101 2023-01-21 12:38:21.919976: step: 344/529, loss: 0.014994049444794655 2023-01-21 12:38:23.100319: step: 348/529, loss: 0.0023178099654614925 2023-01-21 12:38:24.331319: step: 352/529, loss: 0.023823166266083717 2023-01-21 12:38:25.488020: step: 356/529, loss: 0.006887626834213734 2023-01-21 12:38:26.683325: step: 360/529, loss: 0.00013637542724609375 2023-01-21 12:38:27.880756: step: 364/529, loss: 0.001359653426334262 2023-01-21 12:38:29.059676: step: 368/529, loss: 0.06855068355798721 2023-01-21 12:38:30.235539: step: 372/529, loss: -2.47955313170678e-06 2023-01-21 12:38:31.434691: step: 376/529, loss: 0.0009726047283038497 2023-01-21 12:38:32.634236: step: 380/529, loss: 0.0002743721124716103 2023-01-21 12:38:33.818241: step: 384/529, loss: 0.19169625639915466 2023-01-21 12:38:35.007218: step: 388/529, loss: 0.0013978957431390882 2023-01-21 12:38:36.210421: step: 392/529, loss: 0.003904151963070035 2023-01-21 12:38:37.442800: step: 396/529, loss: 0.0008531571365892887 2023-01-21 12:38:38.616272: step: 400/529, loss: 0.00040855410043150187 2023-01-21 12:38:39.811103: step: 404/529, loss: 0.04409322887659073 2023-01-21 12:38:40.993060: step: 408/529, loss: 0.00435562152415514 2023-01-21 12:38:42.182346: step: 412/529, loss: 0.003629970597103238 2023-01-21 12:38:43.357264: step: 416/529, loss: 1.6975403923424892e-05 2023-01-21 12:38:44.511582: step: 420/529, loss: 0.00011405944678699598 2023-01-21 12:38:45.760275: step: 424/529, loss: 0.0049426318146288395 2023-01-21 12:38:46.932329: step: 428/529, loss: 0.007988596335053444 2023-01-21 12:38:48.104371: step: 432/529, loss: 0.00614242535084486 2023-01-21 12:38:49.281310: step: 436/529, loss: 0.028187084943056107 2023-01-21 12:38:50.489816: step: 440/529, loss: 0.013258266262710094 2023-01-21 12:38:51.662722: step: 444/529, loss: 0.0049720765091478825 2023-01-21 12:38:52.890928: step: 448/529, loss: 0.013359450735151768 2023-01-21 12:38:54.087905: step: 452/529, loss: 0.0017400265205651522 2023-01-21 12:38:55.299086: step: 456/529, loss: 0.0007940292125567794 2023-01-21 12:38:56.487257: step: 460/529, loss: 0.008706379681825638 2023-01-21 12:38:57.646924: step: 464/529, loss: 0.0002285003720317036 2023-01-21 12:38:58.849599: step: 468/529, loss: 0.009459304623305798 2023-01-21 12:39:00.044059: step: 472/529, loss: 0.0020677566062659025 2023-01-21 12:39:01.199120: step: 476/529, loss: 2.1123885744600557e-05 2023-01-21 12:39:02.377476: step: 480/529, loss: 0.08410850167274475 2023-01-21 12:39:03.602135: step: 484/529, loss: 0.00011711120168911293 2023-01-21 12:39:04.845570: step: 488/529, loss: 0.0040624141693115234 2023-01-21 12:39:06.047668: step: 492/529, loss: 0.0008524894947186112 2023-01-21 12:39:07.237652: step: 496/529, loss: 0.0017962456913664937 2023-01-21 12:39:08.427682: step: 500/529, loss: 0.02494506910443306 2023-01-21 12:39:09.634084: step: 504/529, loss: 0.02695484273135662 2023-01-21 12:39:10.821529: step: 508/529, loss: 0.004578590393066406 2023-01-21 12:39:12.030114: step: 512/529, loss: 0.0003187179390806705 2023-01-21 12:39:13.201160: step: 516/529, loss: 0.00011186600022483617 2023-01-21 12:39:14.392196: step: 520/529, loss: 0.08090553432703018 2023-01-21 12:39:15.581342: step: 524/529, loss: 0.0005354881286621094 2023-01-21 12:39:16.732598: step: 528/529, loss: 0.0222603902220726 2023-01-21 12:39:17.921746: step: 532/529, loss: 0.04537668451666832 2023-01-21 12:39:19.121139: step: 536/529, loss: 0.0017190934158861637 2023-01-21 12:39:20.296594: step: 540/529, loss: 0.004778957460075617 2023-01-21 12:39:21.489929: step: 544/529, loss: 0.0038251876831054688 2023-01-21 12:39:22.662162: step: 548/529, loss: 0.009939384646713734 2023-01-21 12:39:23.863908: step: 552/529, loss: 0.032839205116033554 2023-01-21 12:39:25.058905: step: 556/529, loss: 0.09469451755285263 2023-01-21 12:39:26.248519: step: 560/529, loss: 0.00033283233642578125 2023-01-21 12:39:27.493394: step: 564/529, loss: 0.005995369050651789 2023-01-21 12:39:28.702192: step: 568/529, loss: 0.27593690156936646 2023-01-21 12:39:29.897395: step: 572/529, loss: 0.010373879224061966 2023-01-21 12:39:31.082433: step: 576/529, loss: 8.659363084007055e-05 2023-01-21 12:39:32.317134: step: 580/529, loss: 0.002545595169067383 2023-01-21 12:39:33.502506: step: 584/529, loss: 0.0017949105240404606 2023-01-21 12:39:34.707018: step: 588/529, loss: 0.00016803742619231343 2023-01-21 12:39:35.923411: step: 592/529, loss: 0.00021395683870650828 2023-01-21 12:39:37.084954: step: 596/529, loss: 0.0022533417213708162 2023-01-21 12:39:38.244887: step: 600/529, loss: 0.0010963439708575606 2023-01-21 12:39:39.421945: step: 604/529, loss: 3.8576126826228574e-05 2023-01-21 12:39:40.655752: step: 608/529, loss: 0.2497566193342209 2023-01-21 12:39:41.865121: step: 612/529, loss: 7.24792471373803e-06 2023-01-21 12:39:43.106252: step: 616/529, loss: 0.017292022705078125 2023-01-21 12:39:44.278405: step: 620/529, loss: 0.046401023864746094 2023-01-21 12:39:45.480259: step: 624/529, loss: 0.001372909639030695 2023-01-21 12:39:46.703452: step: 628/529, loss: 9.613037400413305e-05 2023-01-21 12:39:47.888537: step: 632/529, loss: 0.002762699266895652 2023-01-21 12:39:49.073870: step: 636/529, loss: 0.04123172536492348 2023-01-21 12:39:50.246578: step: 640/529, loss: 0.0005984306335449219 2023-01-21 12:39:51.448698: step: 644/529, loss: 0.01367807388305664 2023-01-21 12:39:52.649268: step: 648/529, loss: 0.8617509603500366 2023-01-21 12:39:53.841173: step: 652/529, loss: 0.010860348120331764 2023-01-21 12:39:54.981245: step: 656/529, loss: 0.00017023086547851562 2023-01-21 12:39:56.165114: step: 660/529, loss: 0.0034050941467285156 2023-01-21 12:39:57.383775: step: 664/529, loss: 0.018602848052978516 2023-01-21 12:39:58.568960: step: 668/529, loss: 0.0005172729725018144 2023-01-21 12:39:59.800055: step: 672/529, loss: 0.0017038345104083419 2023-01-21 12:40:00.927596: step: 676/529, loss: 0.010099983774125576 2023-01-21 12:40:02.145117: step: 680/529, loss: 0.021842384710907936 2023-01-21 12:40:03.358957: step: 684/529, loss: 0.0013540268409997225 2023-01-21 12:40:04.557870: step: 688/529, loss: 0.009274386800825596 2023-01-21 12:40:05.775169: step: 692/529, loss: 0.0017917632358148694 2023-01-21 12:40:06.949700: step: 696/529, loss: 0.00023126603628043085 2023-01-21 12:40:08.168768: step: 700/529, loss: 0.001645830343477428 2023-01-21 12:40:09.367050: step: 704/529, loss: 0.0015456199180334806 2023-01-21 12:40:10.550761: step: 708/529, loss: 0.0009533882257528603 2023-01-21 12:40:11.769500: step: 712/529, loss: 0.0005386352422647178 2023-01-21 12:40:12.947127: step: 716/529, loss: 0.000213623046875 2023-01-21 12:40:14.115333: step: 720/529, loss: 0.014083003625273705 2023-01-21 12:40:15.292359: step: 724/529, loss: 5.207061622058973e-05 2023-01-21 12:40:16.503196: step: 728/529, loss: 0.0012840271228924394 2023-01-21 12:40:17.691140: step: 732/529, loss: 0.001161861466243863 2023-01-21 12:40:18.849742: step: 736/529, loss: 3.519058373058215e-05 2023-01-21 12:40:20.091534: step: 740/529, loss: 0.02496509626507759 2023-01-21 12:40:21.320657: step: 744/529, loss: 3.0517578125e-05 2023-01-21 12:40:22.545319: step: 748/529, loss: -1.5258788153005298e-06 2023-01-21 12:40:23.743134: step: 752/529, loss: 0.06933784484863281 2023-01-21 12:40:24.959334: step: 756/529, loss: 0.004029512405395508 2023-01-21 12:40:26.184239: step: 760/529, loss: 0.00035624505835585296 2023-01-21 12:40:27.364361: step: 764/529, loss: 0.030359458178281784 2023-01-21 12:40:28.580584: step: 768/529, loss: 4.2724612285383046e-05 2023-01-21 12:40:29.810249: step: 772/529, loss: 0.09139309078454971 2023-01-21 12:40:31.000337: step: 776/529, loss: 0.0003488540824037045 2023-01-21 12:40:32.241213: step: 780/529, loss: 0.037384990602731705 2023-01-21 12:40:33.431394: step: 784/529, loss: 0.04303741455078125 2023-01-21 12:40:34.625224: step: 788/529, loss: 0.0006812095525674522 2023-01-21 12:40:35.788125: step: 792/529, loss: 0.0005521774291992188 2023-01-21 12:40:36.932827: step: 796/529, loss: 0.0057732341811060905 2023-01-21 12:40:38.143850: step: 800/529, loss: 0.002682304475456476 2023-01-21 12:40:39.328935: step: 804/529, loss: 0.0033542632590979338 2023-01-21 12:40:40.519914: step: 808/529, loss: 0.01537857111543417 2023-01-21 12:40:41.679102: step: 812/529, loss: 0.004690742585808039 2023-01-21 12:40:42.937111: step: 816/529, loss: 0.008952331729233265 2023-01-21 12:40:44.216326: step: 820/529, loss: 0.0034716606605798006 2023-01-21 12:40:45.414107: step: 824/529, loss: 0.0033850669860839844 2023-01-21 12:40:46.655184: step: 828/529, loss: 0.001586818601936102 2023-01-21 12:40:47.808445: step: 832/529, loss: 0.00036578180151991546 2023-01-21 12:40:49.021650: step: 836/529, loss: 0.040811728686094284 2023-01-21 12:40:50.189870: step: 840/529, loss: 0.007379340939223766 2023-01-21 12:40:51.390825: step: 844/529, loss: 0.0026514052879065275 2023-01-21 12:40:52.614309: step: 848/529, loss: 0.011712265200912952 2023-01-21 12:40:53.831863: step: 852/529, loss: 0.5419204235076904 2023-01-21 12:40:54.990285: step: 856/529, loss: 0.008180045522749424 2023-01-21 12:40:56.217753: step: 860/529, loss: 0.00390453333966434 2023-01-21 12:40:57.394116: step: 864/529, loss: 0.04202689975500107 2023-01-21 12:40:58.562402: step: 868/529, loss: 0.0007396817090921104 2023-01-21 12:40:59.790898: step: 872/529, loss: 0.00846948567777872 2023-01-21 12:41:01.006962: step: 876/529, loss: 0.0719936341047287 2023-01-21 12:41:02.202892: step: 880/529, loss: 0.00221843714825809 2023-01-21 12:41:03.466087: step: 884/529, loss: 0.004226875491440296 2023-01-21 12:41:04.668408: step: 888/529, loss: 0.021477794274687767 2023-01-21 12:41:05.853639: step: 892/529, loss: 0.03520527109503746 2023-01-21 12:41:07.047960: step: 896/529, loss: 0.00015630721463821828 2023-01-21 12:41:08.200769: step: 900/529, loss: 0.012874030508100986 2023-01-21 12:41:09.336179: step: 904/529, loss: 5.3882598876953125e-05 2023-01-21 12:41:10.500259: step: 908/529, loss: 0.03362884745001793 2023-01-21 12:41:11.694156: step: 912/529, loss: 0.0075553893111646175 2023-01-21 12:41:12.874667: step: 916/529, loss: 0.0011791229480877519 2023-01-21 12:41:14.046771: step: 920/529, loss: 0.3843654692173004 2023-01-21 12:41:15.220688: step: 924/529, loss: 0.023404695093631744 2023-01-21 12:41:16.425707: step: 928/529, loss: 5.455017162603326e-05 2023-01-21 12:41:17.652864: step: 932/529, loss: 0.039823438972234726 2023-01-21 12:41:18.862057: step: 936/529, loss: 0.003694439074024558 2023-01-21 12:41:20.028785: step: 940/529, loss: 0.021706486120820045 2023-01-21 12:41:21.188190: step: 944/529, loss: 7.934570021461695e-05 2023-01-21 12:41:22.379349: step: 948/529, loss: 0.008457946591079235 2023-01-21 12:41:23.520480: step: 952/529, loss: 0.0008961677667684853 2023-01-21 12:41:24.742508: step: 956/529, loss: 0.00037422182504087687 2023-01-21 12:41:25.895753: step: 960/529, loss: 0.005460643675178289 2023-01-21 12:41:27.078805: step: 964/529, loss: 0.02651066705584526 2023-01-21 12:41:28.282303: step: 968/529, loss: 7.629394644936838e-07 2023-01-21 12:41:29.443658: step: 972/529, loss: 6.637573096668348e-05 2023-01-21 12:41:30.682041: step: 976/529, loss: 0.002395438961684704 2023-01-21 12:41:31.888965: step: 980/529, loss: 0.0043611531145870686 2023-01-21 12:41:33.072127: step: 984/529, loss: 0.010830879211425781 2023-01-21 12:41:34.286191: step: 988/529, loss: 0.017606543377041817 2023-01-21 12:41:35.467060: step: 992/529, loss: 0.012253857217729092 2023-01-21 12:41:36.611427: step: 996/529, loss: 4.653930591302924e-05 2023-01-21 12:41:37.851788: step: 1000/529, loss: 9.5367431640625e-06 2023-01-21 12:41:39.077239: step: 1004/529, loss: 0.025669097900390625 2023-01-21 12:41:40.263424: step: 1008/529, loss: 0.0023054121993482113 2023-01-21 12:41:41.437768: step: 1012/529, loss: 0.0008772372966632247 2023-01-21 12:41:42.619585: step: 1016/529, loss: 0.00021200180344749242 2023-01-21 12:41:43.805177: step: 1020/529, loss: 0.0025230408646166325 2023-01-21 12:41:45.012869: step: 1024/529, loss: 0.012245750986039639 2023-01-21 12:41:46.197885: step: 1028/529, loss: 0.022622013464570045 2023-01-21 12:41:47.411490: step: 1032/529, loss: 0.0014062881236895919 2023-01-21 12:41:48.584032: step: 1036/529, loss: 0.00652656564489007 2023-01-21 12:41:49.736354: step: 1040/529, loss: 0.0006211280706338584 2023-01-21 12:41:50.998920: step: 1044/529, loss: 0.05848102644085884 2023-01-21 12:41:52.139486: step: 1048/529, loss: 0.0018608197569847107 2023-01-21 12:41:53.355425: step: 1052/529, loss: 0.005982398986816406 2023-01-21 12:41:54.546583: step: 1056/529, loss: 0.013392448425292969 2023-01-21 12:41:55.805962: step: 1060/529, loss: 0.012042808346450329 2023-01-21 12:41:57.019617: step: 1064/529, loss: 0.00027179718017578125 2023-01-21 12:41:58.189145: step: 1068/529, loss: 1.716613724056515e-06 2023-01-21 12:41:59.401006: step: 1072/529, loss: 0.00745887728407979 2023-01-21 12:42:00.641998: step: 1076/529, loss: 0.05139007419347763 2023-01-21 12:42:01.823071: step: 1080/529, loss: 0.0003081798495259136 2023-01-21 12:42:03.023560: step: 1084/529, loss: 0.004885197151452303 2023-01-21 12:42:04.249428: step: 1088/529, loss: 0.01110916119068861 2023-01-21 12:42:05.439835: step: 1092/529, loss: 0.0004084586980752647 2023-01-21 12:42:06.662679: step: 1096/529, loss: 9.14573683985509e-05 2023-01-21 12:42:07.856458: step: 1100/529, loss: 0.0012966155773028731 2023-01-21 12:42:09.048331: step: 1104/529, loss: 0.0002846241113729775 2023-01-21 12:42:10.217191: step: 1108/529, loss: 0.025937365368008614 2023-01-21 12:42:11.411965: step: 1112/529, loss: 0.030796300619840622 2023-01-21 12:42:12.606563: step: 1116/529, loss: 0.0004399299796205014 2023-01-21 12:42:13.791302: step: 1120/529, loss: 0.00017662048048805445 2023-01-21 12:42:14.982625: step: 1124/529, loss: 0.019646836444735527 2023-01-21 12:42:16.186071: step: 1128/529, loss: 0.008589553646743298 2023-01-21 12:42:17.391018: step: 1132/529, loss: 0.0024746893905103207 2023-01-21 12:42:18.555382: step: 1136/529, loss: 0.00013790131197310984 2023-01-21 12:42:19.754399: step: 1140/529, loss: 0.015877723693847656 2023-01-21 12:42:20.934991: step: 1144/529, loss: 0.0019639015663415194 2023-01-21 12:42:22.154409: step: 1148/529, loss: 0.00019588469876907766 2023-01-21 12:42:23.387755: step: 1152/529, loss: 0.0010656355880200863 2023-01-21 12:42:24.564009: step: 1156/529, loss: 3.4809113458322827e-06 2023-01-21 12:42:25.743633: step: 1160/529, loss: 0.017291545867919922 2023-01-21 12:42:26.909687: step: 1164/529, loss: 0.08784690499305725 2023-01-21 12:42:28.101972: step: 1168/529, loss: 0.006003189366310835 2023-01-21 12:42:29.375091: step: 1172/529, loss: 0.06397876888513565 2023-01-21 12:42:30.516965: step: 1176/529, loss: 4.5108794438419864e-05 2023-01-21 12:42:31.698016: step: 1180/529, loss: 0.03741512447595596 2023-01-21 12:42:32.930430: step: 1184/529, loss: 0.03997192531824112 2023-01-21 12:42:34.127880: step: 1188/529, loss: 0.00814743060618639 2023-01-21 12:42:35.392666: step: 1192/529, loss: 0.0019296647515147924 2023-01-21 12:42:36.562225: step: 1196/529, loss: 0.007316398434340954 2023-01-21 12:42:37.775285: step: 1200/529, loss: 0.005738449282944202 2023-01-21 12:42:38.976798: step: 1204/529, loss: 0.011988830752670765 2023-01-21 12:42:40.169198: step: 1208/529, loss: 0.0002483367861714214 2023-01-21 12:42:41.363144: step: 1212/529, loss: 0.012173938564956188 2023-01-21 12:42:42.563340: step: 1216/529, loss: 0.008593368344008923 2023-01-21 12:42:43.757314: step: 1220/529, loss: 0.002732848981395364 2023-01-21 12:42:44.906065: step: 1224/529, loss: 0.001075363252311945 2023-01-21 12:42:46.124786: step: 1228/529, loss: 0.0035504342522472143 2023-01-21 12:42:47.288234: step: 1232/529, loss: 0.01102523785084486 2023-01-21 12:42:48.479550: step: 1236/529, loss: 1.0433621406555176 2023-01-21 12:42:49.716922: step: 1240/529, loss: 0.005492592230439186 2023-01-21 12:42:50.899504: step: 1244/529, loss: 0.009107494726777077 2023-01-21 12:42:52.171085: step: 1248/529, loss: 0.0006118774181231856 2023-01-21 12:42:53.320754: step: 1252/529, loss: 0.0015988349914550781 2023-01-21 12:42:54.527235: step: 1256/529, loss: 0.0009424209711141884 2023-01-21 12:42:55.688503: step: 1260/529, loss: 0.0021330832969397306 2023-01-21 12:42:56.860270: step: 1264/529, loss: 0.0034543040674179792 2023-01-21 12:42:58.073076: step: 1268/529, loss: 0.02487774007022381 2023-01-21 12:42:59.249693: step: 1272/529, loss: 0.001723480294458568 2023-01-21 12:43:00.468718: step: 1276/529, loss: -2.9563905172835803e-06 2023-01-21 12:43:01.638114: step: 1280/529, loss: 0.1012289971113205 2023-01-21 12:43:02.826568: step: 1284/529, loss: 0.012979316525161266 2023-01-21 12:43:04.025567: step: 1288/529, loss: 0.03601064905524254 2023-01-21 12:43:05.187721: step: 1292/529, loss: 0.017980385571718216 2023-01-21 12:43:06.390549: step: 1296/529, loss: 0.029918862506747246 2023-01-21 12:43:07.550993: step: 1300/529, loss: 0.04778280481696129 2023-01-21 12:43:08.730146: step: 1304/529, loss: 0.0007806778303347528 2023-01-21 12:43:09.964355: step: 1308/529, loss: 0.013626289553940296 2023-01-21 12:43:11.125532: step: 1312/529, loss: 2.117157055181451e-05 2023-01-21 12:43:12.288443: step: 1316/529, loss: 0.015629006549715996 2023-01-21 12:43:13.490619: step: 1320/529, loss: 0.00102491385769099 2023-01-21 12:43:14.694894: step: 1324/529, loss: 0.03168588876724243 2023-01-21 12:43:15.907251: step: 1328/529, loss: 0.08364677429199219 2023-01-21 12:43:17.106673: step: 1332/529, loss: 6.54130744934082 2023-01-21 12:43:18.248174: step: 1336/529, loss: 0.10628020763397217 2023-01-21 12:43:19.419916: step: 1340/529, loss: 0.0005728721152991056 2023-01-21 12:43:20.615747: step: 1344/529, loss: 0.0016220093239098787 2023-01-21 12:43:21.779448: step: 1348/529, loss: 0.003975295927375555 2023-01-21 12:43:22.980569: step: 1352/529, loss: 0.0009122848859988153 2023-01-21 12:43:24.160196: step: 1356/529, loss: 0.006723451893776655 2023-01-21 12:43:25.341042: step: 1360/529, loss: 0.00216426863335073 2023-01-21 12:43:26.551877: step: 1364/529, loss: 0.003072547959163785 2023-01-21 12:43:27.722904: step: 1368/529, loss: 0.0002906322479248047 2023-01-21 12:43:28.884466: step: 1372/529, loss: 0.007688426878303289 2023-01-21 12:43:30.070163: step: 1376/529, loss: 0.03008747100830078 2023-01-21 12:43:31.270114: step: 1380/529, loss: 0.09815044701099396 2023-01-21 12:43:32.496567: step: 1384/529, loss: 0.00253047957085073 2023-01-21 12:43:33.698252: step: 1388/529, loss: 0.031522177159786224 2023-01-21 12:43:34.871484: step: 1392/529, loss: 0.004557228181511164 2023-01-21 12:43:36.046061: step: 1396/529, loss: 0.0005568504566326737 2023-01-21 12:43:37.229050: step: 1400/529, loss: 0.019248582422733307 2023-01-21 12:43:38.444887: step: 1404/529, loss: 3.757476952159777e-05 2023-01-21 12:43:39.610966: step: 1408/529, loss: 0.3026461601257324 2023-01-21 12:43:40.800138: step: 1412/529, loss: 0.033855997025966644 2023-01-21 12:43:41.963843: step: 1416/529, loss: 0.001484012696892023 2023-01-21 12:43:43.148451: step: 1420/529, loss: 0.0004883766523562372 2023-01-21 12:43:44.349742: step: 1424/529, loss: 0.00010242462303722277 2023-01-21 12:43:45.529895: step: 1428/529, loss: 0.014437961392104626 2023-01-21 12:43:46.724403: step: 1432/529, loss: 0.025298595428466797 2023-01-21 12:43:47.945919: step: 1436/529, loss: 0.00977945327758789 2023-01-21 12:43:49.122736: step: 1440/529, loss: 0.02181701548397541 2023-01-21 12:43:50.318066: step: 1444/529, loss: 0.01446991041302681 2023-01-21 12:43:51.554152: step: 1448/529, loss: 0.002718687057495117 2023-01-21 12:43:52.764693: step: 1452/529, loss: 0.00040612221346236765 2023-01-21 12:43:53.991785: step: 1456/529, loss: 0.0014866827987134457 2023-01-21 12:43:55.241912: step: 1460/529, loss: 0.025051381438970566 2023-01-21 12:43:56.422138: step: 1464/529, loss: 0.005203437991440296 2023-01-21 12:43:57.661503: step: 1468/529, loss: 0.0014740824699401855 2023-01-21 12:43:58.822636: step: 1472/529, loss: 0.02445850521326065 2023-01-21 12:44:00.004766: step: 1476/529, loss: 0.0006958484882488847 2023-01-21 12:44:01.219872: step: 1480/529, loss: 0.06531887501478195 2023-01-21 12:44:02.407993: step: 1484/529, loss: 0.044643305242061615 2023-01-21 12:44:03.627119: step: 1488/529, loss: 0.004179954528808594 2023-01-21 12:44:04.837817: step: 1492/529, loss: 0.0019033432472497225 2023-01-21 12:44:06.010311: step: 1496/529, loss: 0.0009115219581872225 2023-01-21 12:44:07.211907: step: 1500/529, loss: 0.01475915964692831 2023-01-21 12:44:08.414209: step: 1504/529, loss: 0.01742544211447239 2023-01-21 12:44:09.569515: step: 1508/529, loss: 6.160736666060984e-05 2023-01-21 12:44:10.727121: step: 1512/529, loss: 0.0008149623754434288 2023-01-21 12:44:11.935655: step: 1516/529, loss: 0.0003214835887774825 2023-01-21 12:44:13.125720: step: 1520/529, loss: 0.0006517410511150956 2023-01-21 12:44:14.283128: step: 1524/529, loss: 0.01327667199075222 2023-01-21 12:44:15.470327: step: 1528/529, loss: 0.0025823593605309725 2023-01-21 12:44:16.674057: step: 1532/529, loss: 0.013388633728027344 2023-01-21 12:44:17.914701: step: 1536/529, loss: 0.031649161130189896 2023-01-21 12:44:19.121621: step: 1540/529, loss: 0.0002270221884828061 2023-01-21 12:44:20.300663: step: 1544/529, loss: 0.0012959480518475175 2023-01-21 12:44:21.489445: step: 1548/529, loss: 0.005197572521865368 2023-01-21 12:44:22.679356: step: 1552/529, loss: 0.0027899742126464844 2023-01-21 12:44:23.823433: step: 1556/529, loss: 0.0001034975066431798 2023-01-21 12:44:25.019292: step: 1560/529, loss: 0.004379653837531805 2023-01-21 12:44:26.217159: step: 1564/529, loss: 0.004017258062958717 2023-01-21 12:44:27.425401: step: 1568/529, loss: 0.003856086637824774 2023-01-21 12:44:28.608299: step: 1572/529, loss: 0.001081275986507535 2023-01-21 12:44:29.828114: step: 1576/529, loss: 0.0038760663010179996 2023-01-21 12:44:31.038208: step: 1580/529, loss: 0.0024871828500181437 2023-01-21 12:44:32.222776: step: 1584/529, loss: 0.047409821301698685 2023-01-21 12:44:33.409291: step: 1588/529, loss: 0.0003452301025390625 2023-01-21 12:44:34.553700: step: 1592/529, loss: 2.8038026357535273e-05 2023-01-21 12:44:35.723282: step: 1596/529, loss: 0.0019240856636315584 2023-01-21 12:44:36.871063: step: 1600/529, loss: 0.0010791778331622481 2023-01-21 12:44:38.053324: step: 1604/529, loss: 0.013415718451142311 2023-01-21 12:44:39.263083: step: 1608/529, loss: 0.2130260467529297 2023-01-21 12:44:40.430173: step: 1612/529, loss: 0.001004123711027205 2023-01-21 12:44:41.583884: step: 1616/529, loss: 0.0004294872051104903 2023-01-21 12:44:42.760275: step: 1620/529, loss: 0.001323795411735773 2023-01-21 12:44:43.930451: step: 1624/529, loss: 4.997253563487902e-05 2023-01-21 12:44:45.138350: step: 1628/529, loss: 0.05893135070800781 2023-01-21 12:44:46.295371: step: 1632/529, loss: 0.0002254486025776714 2023-01-21 12:44:47.513523: step: 1636/529, loss: 0.004370022099465132 2023-01-21 12:44:48.684530: step: 1640/529, loss: 0.0016426086658611894 2023-01-21 12:44:49.903995: step: 1644/529, loss: 0.013583755120635033 2023-01-21 12:44:51.110939: step: 1648/529, loss: 0.01568617857992649 2023-01-21 12:44:52.338101: step: 1652/529, loss: 0.10471649467945099 2023-01-21 12:44:53.533754: step: 1656/529, loss: 0.0027340890374034643 2023-01-21 12:44:54.731838: step: 1660/529, loss: 0.003850650740787387 2023-01-21 12:44:55.934733: step: 1664/529, loss: 0.8167505264282227 2023-01-21 12:44:57.098057: step: 1668/529, loss: 0.0005524397129192948 2023-01-21 12:44:58.267147: step: 1672/529, loss: 0.0005867957952432334 2023-01-21 12:44:59.445484: step: 1676/529, loss: 0.004097366705536842 2023-01-21 12:45:00.648012: step: 1680/529, loss: 0.004999065771698952 2023-01-21 12:45:01.864735: step: 1684/529, loss: 0.02600574679672718 2023-01-21 12:45:03.058972: step: 1688/529, loss: 0.012535858899354935 2023-01-21 12:45:04.271511: step: 1692/529, loss: 0.022948648780584335 2023-01-21 12:45:05.475478: step: 1696/529, loss: 0.026111412793397903 2023-01-21 12:45:06.680829: step: 1700/529, loss: 0.01734142377972603 2023-01-21 12:45:07.883780: step: 1704/529, loss: 0.1176968589425087 2023-01-21 12:45:09.025097: step: 1708/529, loss: 0.004532337188720703 2023-01-21 12:45:10.197494: step: 1712/529, loss: 0.012049293145537376 2023-01-21 12:45:11.355340: step: 1716/529, loss: 0.011378765106201172 2023-01-21 12:45:12.593766: step: 1720/529, loss: 0.01756277121603489 2023-01-21 12:45:13.822859: step: 1724/529, loss: 0.010570907965302467 2023-01-21 12:45:14.976884: step: 1728/529, loss: 3.7670135498046875e-05 2023-01-21 12:45:16.149506: step: 1732/529, loss: 1.735687328618951e-05 2023-01-21 12:45:17.349803: step: 1736/529, loss: 0.0002173423854401335 2023-01-21 12:45:18.581295: step: 1740/529, loss: 0.0018852234352380037 2023-01-21 12:45:19.752442: step: 1744/529, loss: 0.003607082413509488 2023-01-21 12:45:20.932000: step: 1748/529, loss: 0.012110329233109951 2023-01-21 12:45:22.126341: step: 1752/529, loss: 0.00013599396334029734 2023-01-21 12:45:23.358380: step: 1756/529, loss: 0.021680641919374466 2023-01-21 12:45:24.559744: step: 1760/529, loss: 0.007324123755097389 2023-01-21 12:45:25.763830: step: 1764/529, loss: 0.008916569873690605 2023-01-21 12:45:27.023044: step: 1768/529, loss: 0.01064910925924778 2023-01-21 12:45:28.200922: step: 1772/529, loss: 0.012298869900405407 2023-01-21 12:45:29.369205: step: 1776/529, loss: 0.0027667046524584293 2023-01-21 12:45:30.580312: step: 1780/529, loss: 4.978180368198082e-05 2023-01-21 12:45:31.822271: step: 1784/529, loss: 0.013062858022749424 2023-01-21 12:45:33.051974: step: 1788/529, loss: 0.04527664557099342 2023-01-21 12:45:34.244783: step: 1792/529, loss: 0.00011978149996139109 2023-01-21 12:45:35.441197: step: 1796/529, loss: 0.004878378007560968 2023-01-21 12:45:36.606468: step: 1800/529, loss: 0.0009832382202148438 2023-01-21 12:45:37.805915: step: 1804/529, loss: 0.007679652888327837 2023-01-21 12:45:39.047270: step: 1808/529, loss: 0.010171796195209026 2023-01-21 12:45:40.230111: step: 1812/529, loss: 0.00025091171846725047 2023-01-21 12:45:41.388024: step: 1816/529, loss: 0.00021839141845703125 2023-01-21 12:45:42.565198: step: 1820/529, loss: 0.009160232730209827 2023-01-21 12:45:43.708846: step: 1824/529, loss: 7.157326035667211e-05 2023-01-21 12:45:44.898543: step: 1828/529, loss: 0.0010103225940838456 2023-01-21 12:45:46.117196: step: 1832/529, loss: 0.121434785425663 2023-01-21 12:45:47.379545: step: 1836/529, loss: 0.007035255432128906 2023-01-21 12:45:48.584062: step: 1840/529, loss: 0.023754501715302467 2023-01-21 12:45:49.782356: step: 1844/529, loss: 0.055586766451597214 2023-01-21 12:45:50.958192: step: 1848/529, loss: 0.03387375921010971 2023-01-21 12:45:52.148827: step: 1852/529, loss: 0.027310943230986595 2023-01-21 12:45:53.336651: step: 1856/529, loss: 0.004269313998520374 2023-01-21 12:45:54.480375: step: 1860/529, loss: 0.02469787560403347 2023-01-21 12:45:55.703208: step: 1864/529, loss: 0.020456695929169655 2023-01-21 12:45:56.952152: step: 1868/529, loss: 0.004259491339325905 2023-01-21 12:45:58.139247: step: 1872/529, loss: 0.011418819427490234 2023-01-21 12:45:59.349000: step: 1876/529, loss: 0.00019569396681617945 2023-01-21 12:46:00.523536: step: 1880/529, loss: 5.722049536416307e-07 2023-01-21 12:46:01.715476: step: 1884/529, loss: 0.0038892747834324837 2023-01-21 12:46:02.883477: step: 1888/529, loss: 0.009188652038574219 2023-01-21 12:46:04.074909: step: 1892/529, loss: 0.04315633699297905 2023-01-21 12:46:05.237224: step: 1896/529, loss: 0.0001201629638671875 2023-01-21 12:46:06.445110: step: 1900/529, loss: 0.00011901855759788305 2023-01-21 12:46:07.649304: step: 1904/529, loss: 0.0020957947708666325 2023-01-21 12:46:08.805152: step: 1908/529, loss: 0.07537918537855148 2023-01-21 12:46:09.980110: step: 1912/529, loss: 0.008517456240952015 2023-01-21 12:46:11.160357: step: 1916/529, loss: 0.011315489187836647 2023-01-21 12:46:12.346533: step: 1920/529, loss: 0.027272367849946022 2023-01-21 12:46:13.527236: step: 1924/529, loss: 0.003683662274852395 2023-01-21 12:46:14.702260: step: 1928/529, loss: 8.544922457076609e-05 2023-01-21 12:46:15.914662: step: 1932/529, loss: 0.024008655920624733 2023-01-21 12:46:17.084302: step: 1936/529, loss: 0.0006074905395507812 2023-01-21 12:46:18.257414: step: 1940/529, loss: 0.0015086650382727385 2023-01-21 12:46:19.440448: step: 1944/529, loss: 0.005527210421860218 2023-01-21 12:46:20.644118: step: 1948/529, loss: 0.01891803927719593 2023-01-21 12:46:21.859230: step: 1952/529, loss: 0.3365371823310852 2023-01-21 12:46:23.008389: step: 1956/529, loss: 0.0031655312050133944 2023-01-21 12:46:24.202808: step: 1960/529, loss: 4.56810012110509e-05 2023-01-21 12:46:25.378913: step: 1964/529, loss: 0.004829073324799538 2023-01-21 12:46:26.616714: step: 1968/529, loss: 0.016674233600497246 2023-01-21 12:46:27.811876: step: 1972/529, loss: 0.0006406784523278475 2023-01-21 12:46:29.001623: step: 1976/529, loss: 5.216598219703883e-05 2023-01-21 12:46:30.175892: step: 1980/529, loss: 0.002842521760612726 2023-01-21 12:46:31.377871: step: 1984/529, loss: 0.008170604705810547 2023-01-21 12:46:32.589485: step: 1988/529, loss: 0.0002967834589071572 2023-01-21 12:46:33.791554: step: 1992/529, loss: 0.0002035141078522429 2023-01-21 12:46:34.935257: step: 1996/529, loss: 2.6226043701171875e-06 2023-01-21 12:46:36.131106: step: 2000/529, loss: 0.00036640168400481343 2023-01-21 12:46:37.392290: step: 2004/529, loss: 0.0018598557217046618 2023-01-21 12:46:38.588711: step: 2008/529, loss: 0.026102256029844284 2023-01-21 12:46:39.792850: step: 2012/529, loss: 0.005317592527717352 2023-01-21 12:46:40.960128: step: 2016/529, loss: 0.010424232110381126 2023-01-21 12:46:42.127961: step: 2020/529, loss: 0.0006548881647177041 2023-01-21 12:46:43.322472: step: 2024/529, loss: 0.02035045623779297 2023-01-21 12:46:44.494325: step: 2028/529, loss: 0.0013819695450365543 2023-01-21 12:46:45.669820: step: 2032/529, loss: 0.011353731155395508 2023-01-21 12:46:46.867558: step: 2036/529, loss: 0.00028066636878065765 2023-01-21 12:46:48.002023: step: 2040/529, loss: 0.0007247448083944619 2023-01-21 12:46:49.216740: step: 2044/529, loss: 0.029674530029296875 2023-01-21 12:46:50.394322: step: 2048/529, loss: 0.0016052245628088713 2023-01-21 12:46:51.586876: step: 2052/529, loss: 0.0005439758533611894 2023-01-21 12:46:52.787854: step: 2056/529, loss: 0.009554862976074219 2023-01-21 12:46:53.945786: step: 2060/529, loss: 0.02063455618917942 2023-01-21 12:46:55.169335: step: 2064/529, loss: 0.0025295258965343237 2023-01-21 12:46:56.342514: step: 2068/529, loss: 0.0031034471467137337 2023-01-21 12:46:57.582292: step: 2072/529, loss: 0.0003180503845214844 2023-01-21 12:46:58.784194: step: 2076/529, loss: 0.00011911392357433215 2023-01-21 12:46:59.963019: step: 2080/529, loss: 0.0004371166287455708 2023-01-21 12:47:01.154340: step: 2084/529, loss: 0.017603684216737747 2023-01-21 12:47:02.355191: step: 2088/529, loss: 0.037097737193107605 2023-01-21 12:47:03.521883: step: 2092/529, loss: 0.0011396408081054688 2023-01-21 12:47:04.720725: step: 2096/529, loss: 6.519556336570531e-05 2023-01-21 12:47:05.909827: step: 2100/529, loss: 0.032398417592048645 2023-01-21 12:47:07.050140: step: 2104/529, loss: 0.016867637634277344 2023-01-21 12:47:08.245649: step: 2108/529, loss: 0.0005841254605911672 2023-01-21 12:47:09.408270: step: 2112/529, loss: 0.0011454150080680847 2023-01-21 12:47:10.587927: step: 2116/529, loss: 9.72747784544481e-06 ================================================== Loss: 0.034 -------------------- Dev: {'event': {'p': 0.6016427104722792, 'r': 0.7802929427430093, 'f1': 0.6794202898550724}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Test: {'event': {'p': 0.6448007774538387, 'r': 0.7950868783702816, 'f1': 0.7121008854306413}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Chinese: {'event': {'p': 0.5679012345679012, 'r': 0.8518518518518519, 'f1': 0.6814814814814814}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Korean: {'event': {'p': 0.5396825396825397, 'r': 0.5396825396825397, 'f1': 0.5396825396825397}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Russian: {'event': {'p': 0.4318181818181818, 'r': 0.5277777777777778, 'f1': 0.47500000000000003}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 12:47:53.706694: step: 4/529, loss: 0.000898170517757535 2023-01-21 12:47:54.859790: step: 8/529, loss: 0.0036108018830418587 2023-01-21 12:47:56.058888: step: 12/529, loss: 0.00015649796114303172 2023-01-21 12:47:57.249404: step: 16/529, loss: 0.003478622529655695 2023-01-21 12:47:58.403594: step: 20/529, loss: 0.000854110752698034 2023-01-21 12:47:59.602287: step: 24/529, loss: 0.0006662368541583419 2023-01-21 12:48:00.854393: step: 28/529, loss: 0.0652950257062912 2023-01-21 12:48:02.107855: step: 32/529, loss: 0.0008264541975222528 2023-01-21 12:48:03.293479: step: 36/529, loss: 0.00046195986215025187 2023-01-21 12:48:04.520688: step: 40/529, loss: 0.00012683868408203125 2023-01-21 12:48:05.700886: step: 44/529, loss: 4.0721897676121444e-05 2023-01-21 12:48:06.862332: step: 48/529, loss: 4.339218776294729e-06 2023-01-21 12:48:08.057148: step: 52/529, loss: 0.0124785415828228 2023-01-21 12:48:09.233458: step: 56/529, loss: 0.0054172039963305 2023-01-21 12:48:10.372207: step: 60/529, loss: 5.474090721691027e-05 2023-01-21 12:48:11.601992: step: 64/529, loss: 0.04378366470336914 2023-01-21 12:48:12.839931: step: 68/529, loss: 0.0018190382979810238 2023-01-21 12:48:14.004076: step: 72/529, loss: 0.005840111058205366 2023-01-21 12:48:15.187886: step: 76/529, loss: 0.0023975849617272615 2023-01-21 12:48:16.408072: step: 80/529, loss: 0.004075098317116499 2023-01-21 12:48:17.575952: step: 84/529, loss: 0.3560231328010559 2023-01-21 12:48:18.769838: step: 88/529, loss: 0.0016739845741540194 2023-01-21 12:48:19.943820: step: 92/529, loss: 0.4803546071052551 2023-01-21 12:48:21.155649: step: 96/529, loss: 1.9788742065429688e-05 2023-01-21 12:48:22.326483: step: 100/529, loss: 0.010644244961440563 2023-01-21 12:48:23.512086: step: 104/529, loss: 0.05007629469037056 2023-01-21 12:48:24.695734: step: 108/529, loss: 0.0007342934841290116 2023-01-21 12:48:25.899107: step: 112/529, loss: 0.0019673823844641447 2023-01-21 12:48:27.128087: step: 116/529, loss: 0.007548236753791571 2023-01-21 12:48:28.294121: step: 120/529, loss: -6.675722943327855e-07 2023-01-21 12:48:29.463794: step: 124/529, loss: 0.010536003857851028 2023-01-21 12:48:30.645916: step: 128/529, loss: 0.002724266145378351 2023-01-21 12:48:31.881193: step: 132/529, loss: 0.0025578499771654606 2023-01-21 12:48:33.082989: step: 136/529, loss: 0.0001903533993754536 2023-01-21 12:48:34.265124: step: 140/529, loss: 0.0005716323503293097 2023-01-21 12:48:35.484672: step: 144/529, loss: 0.010661887936294079 2023-01-21 12:48:36.704826: step: 148/529, loss: 9.479522850597277e-05 2023-01-21 12:48:37.889918: step: 152/529, loss: 0.00019254683866165578 2023-01-21 12:48:39.066104: step: 156/529, loss: 0.0006583214271813631 2023-01-21 12:48:40.278937: step: 160/529, loss: 0.021617794409394264 2023-01-21 12:48:41.443274: step: 164/529, loss: 0.007967852987349033 2023-01-21 12:48:42.620246: step: 168/529, loss: 0.022089196369051933 2023-01-21 12:48:43.814606: step: 172/529, loss: 0.00033512114896439016 2023-01-21 12:48:44.973423: step: 176/529, loss: 0.0001146316499216482 2023-01-21 12:48:46.140052: step: 180/529, loss: 0.008104896172881126 2023-01-21 12:48:47.296408: step: 184/529, loss: 0.00011768341937568039 2023-01-21 12:48:48.461568: step: 188/529, loss: 0.008573532104492188 2023-01-21 12:48:49.602257: step: 192/529, loss: 0.04063277319073677 2023-01-21 12:48:50.786502: step: 196/529, loss: 0.0036703108344227076 2023-01-21 12:48:51.989743: step: 200/529, loss: 0.0321260467171669 2023-01-21 12:48:53.182844: step: 204/529, loss: 0.004272747319191694 2023-01-21 12:48:54.314782: step: 208/529, loss: 0.00010042190842796117 2023-01-21 12:48:55.529905: step: 212/529, loss: 0.0007357597351074219 2023-01-21 12:48:56.705564: step: 216/529, loss: 0.01537170447409153 2023-01-21 12:48:57.882177: step: 220/529, loss: 0.0023218155838549137 2023-01-21 12:48:59.080184: step: 224/529, loss: 0.00044889451237395406 2023-01-21 12:49:00.222463: step: 228/529, loss: 0.0013660431141033769 2023-01-21 12:49:01.399269: step: 232/529, loss: 0.0004330635129008442 2023-01-21 12:49:02.582998: step: 236/529, loss: 6.29425048828125e-05 2023-01-21 12:49:03.748353: step: 240/529, loss: 0.005469989497214556 2023-01-21 12:49:04.907936: step: 244/529, loss: 0.009460163302719593 2023-01-21 12:49:06.096028: step: 248/529, loss: 0.00032253266545012593 2023-01-21 12:49:07.306928: step: 252/529, loss: 0.001751041505485773 2023-01-21 12:49:08.517480: step: 256/529, loss: 0.0013683318393304944 2023-01-21 12:49:09.725104: step: 260/529, loss: 0.020910168066620827 2023-01-21 12:49:10.886826: step: 264/529, loss: 0.030348969623446465 2023-01-21 12:49:12.088904: step: 268/529, loss: 0.019161511212587357 2023-01-21 12:49:13.325459: step: 272/529, loss: 0.020816612988710403 2023-01-21 12:49:14.619300: step: 276/529, loss: 0.0016132354503497481 2023-01-21 12:49:15.794123: step: 280/529, loss: 0.016295909881591797 2023-01-21 12:49:17.022486: step: 284/529, loss: 0.025266075506806374 2023-01-21 12:49:18.202008: step: 288/529, loss: 0.011342430487275124 2023-01-21 12:49:19.383640: step: 292/529, loss: 0.02742328681051731 2023-01-21 12:49:20.578243: step: 296/529, loss: 0.00030794143094681203 2023-01-21 12:49:21.747747: step: 300/529, loss: 0.00015134812565520406 2023-01-21 12:49:22.901334: step: 304/529, loss: 0.00656933756545186 2023-01-21 12:49:24.090972: step: 308/529, loss: 4.3582916987361386e-05 2023-01-21 12:49:25.295030: step: 312/529, loss: 0.012262344360351562 2023-01-21 12:49:26.485771: step: 316/529, loss: 0.0012790679465979338 2023-01-21 12:49:27.686619: step: 320/529, loss: 0.011958600021898746 2023-01-21 12:49:28.880594: step: 324/529, loss: 0.00024366378784179688 2023-01-21 12:49:30.064550: step: 328/529, loss: 0.0012674331665039062 2023-01-21 12:49:31.280909: step: 332/529, loss: 0.050702571868896484 2023-01-21 12:49:32.501016: step: 336/529, loss: 0.0012767792213708162 2023-01-21 12:49:33.696044: step: 340/529, loss: 0.012340545654296875 2023-01-21 12:49:34.879858: step: 344/529, loss: 0.001984119415283203 2023-01-21 12:49:36.062659: step: 348/529, loss: 0.01017541904002428 2023-01-21 12:49:37.236757: step: 352/529, loss: 0.00038437845068983734 2023-01-21 12:49:38.497492: step: 356/529, loss: 0.003945684526115656 2023-01-21 12:49:39.731316: step: 360/529, loss: 0.1507761925458908 2023-01-21 12:49:40.915923: step: 364/529, loss: 4.081726001459174e-05 2023-01-21 12:49:42.118749: step: 368/529, loss: 0.016002465039491653 2023-01-21 12:49:43.306858: step: 372/529, loss: 0.1385457068681717 2023-01-21 12:49:44.480325: step: 376/529, loss: 0.0004816055588889867 2023-01-21 12:49:45.693517: step: 380/529, loss: 0.01354208029806614 2023-01-21 12:49:46.911640: step: 384/529, loss: 0.0003362655406817794 2023-01-21 12:49:48.125597: step: 388/529, loss: 6.103515625e-05 2023-01-21 12:49:49.327690: step: 392/529, loss: 5.435943421616685e-06 2023-01-21 12:49:50.519743: step: 396/529, loss: 0.0013409615494310856 2023-01-21 12:49:51.699921: step: 400/529, loss: 0.007124709896743298 2023-01-21 12:49:52.884654: step: 404/529, loss: 0.0001832962007028982 2023-01-21 12:49:54.051029: step: 408/529, loss: 0.004058551974594593 2023-01-21 12:49:55.271731: step: 412/529, loss: 0.0012286186683923006 2023-01-21 12:49:56.463955: step: 416/529, loss: 0.004363155458122492 2023-01-21 12:49:57.628640: step: 420/529, loss: 0.00022745132446289062 2023-01-21 12:49:58.836427: step: 424/529, loss: 0.030775094404816628 2023-01-21 12:49:59.990547: step: 428/529, loss: 0.006182861514389515 2023-01-21 12:50:01.184153: step: 432/529, loss: -7.152557373046875e-06 2023-01-21 12:50:02.384683: step: 436/529, loss: 0.02369880862534046 2023-01-21 12:50:03.528036: step: 440/529, loss: 0.013905716128647327 2023-01-21 12:50:04.708120: step: 444/529, loss: 0.0012062073219567537 2023-01-21 12:50:05.904940: step: 448/529, loss: 0.0031617165077477694 2023-01-21 12:50:07.115380: step: 452/529, loss: 0.0003533363342285156 2023-01-21 12:50:08.340233: step: 456/529, loss: 0.0004265785391908139 2023-01-21 12:50:09.519201: step: 460/529, loss: 0.001143455388955772 2023-01-21 12:50:10.800104: step: 464/529, loss: 0.006726074498146772 2023-01-21 12:50:12.067528: step: 468/529, loss: 0.103388212621212 2023-01-21 12:50:13.230881: step: 472/529, loss: 0.00012693404278252274 2023-01-21 12:50:14.398361: step: 476/529, loss: 0.033284951001405716 2023-01-21 12:50:15.600201: step: 480/529, loss: 0.0007293701637536287 2023-01-21 12:50:16.829883: step: 484/529, loss: 0.001493167830631137 2023-01-21 12:50:18.066557: step: 488/529, loss: 0.12736888229846954 2023-01-21 12:50:19.254452: step: 492/529, loss: 0.0004247665638104081 2023-01-21 12:50:20.467671: step: 496/529, loss: 0.007828759960830212 2023-01-21 12:50:21.655771: step: 500/529, loss: 0.0004994392511434853 2023-01-21 12:50:22.811319: step: 504/529, loss: 5.1307681133039296e-05 2023-01-21 12:50:24.014427: step: 508/529, loss: 0.009360289201140404 2023-01-21 12:50:25.211424: step: 512/529, loss: 0.0005950927734375 2023-01-21 12:50:26.412380: step: 516/529, loss: 0.0006711006280966103 2023-01-21 12:50:27.591638: step: 520/529, loss: 0.0017589569324627519 2023-01-21 12:50:28.814365: step: 524/529, loss: 0.03187408670783043 2023-01-21 12:50:30.026747: step: 528/529, loss: 0.016418838873505592 2023-01-21 12:50:31.246547: step: 532/529, loss: 0.003821468446403742 2023-01-21 12:50:32.376343: step: 536/529, loss: 0.0004405975341796875 2023-01-21 12:50:33.536291: step: 540/529, loss: 0.017548855394124985 2023-01-21 12:50:34.727079: step: 544/529, loss: 0.0020092965569347143 2023-01-21 12:50:35.940873: step: 548/529, loss: 0.004520607180893421 2023-01-21 12:50:37.164776: step: 552/529, loss: 0.05471072345972061 2023-01-21 12:50:38.370277: step: 556/529, loss: 0.03413067013025284 2023-01-21 12:50:39.595611: step: 560/529, loss: 0.00017023086547851562 2023-01-21 12:50:40.815768: step: 564/529, loss: 0.018899155780673027 2023-01-21 12:50:41.989140: step: 568/529, loss: 1.3637543816003017e-05 2023-01-21 12:50:43.198245: step: 572/529, loss: 0.0420990027487278 2023-01-21 12:50:44.427168: step: 576/529, loss: 0.0003098964807577431 2023-01-21 12:50:45.624318: step: 580/529, loss: 0.07922515273094177 2023-01-21 12:50:46.814308: step: 584/529, loss: 0.0020321845076978207 2023-01-21 12:50:47.982935: step: 588/529, loss: 0.12905311584472656 2023-01-21 12:50:49.174189: step: 592/529, loss: 0.00039615630521439016 2023-01-21 12:50:50.387353: step: 596/529, loss: 0.00038280486478470266 2023-01-21 12:50:51.558828: step: 600/529, loss: 0.0007277489057742059 2023-01-21 12:50:52.723895: step: 604/529, loss: 0.007420635316520929 2023-01-21 12:50:53.964852: step: 608/529, loss: 0.018919754773378372 2023-01-21 12:50:55.192294: step: 612/529, loss: 9.32693510549143e-05 2023-01-21 12:50:56.362950: step: 616/529, loss: 0.023487472906708717 2023-01-21 12:50:57.584323: step: 620/529, loss: 0.020222092047333717 2023-01-21 12:50:58.776333: step: 624/529, loss: 0.028501415625214577 2023-01-21 12:50:59.937311: step: 628/529, loss: 9.16481003514491e-05 2023-01-21 12:51:01.143944: step: 632/529, loss: 0.03372354432940483 2023-01-21 12:51:02.315206: step: 636/529, loss: 0.0003410339413676411 2023-01-21 12:51:03.533447: step: 640/529, loss: 0.0005878924857825041 2023-01-21 12:51:04.758783: step: 644/529, loss: 0.03943615034222603 2023-01-21 12:51:05.960086: step: 648/529, loss: 0.011771392077207565 2023-01-21 12:51:07.188295: step: 652/529, loss: 0.01586771011352539 2023-01-21 12:51:08.349191: step: 656/529, loss: 0.002084159990772605 2023-01-21 12:51:09.589397: step: 660/529, loss: 0.023773955181241035 2023-01-21 12:51:10.771605: step: 664/529, loss: 2.8705597287626006e-05 2023-01-21 12:51:11.930911: step: 668/529, loss: 8.172988600563258e-05 2023-01-21 12:51:13.103920: step: 672/529, loss: 0.06975989788770676 2023-01-21 12:51:14.303759: step: 676/529, loss: 7.677078247070312e-05 2023-01-21 12:51:15.514120: step: 680/529, loss: 0.0027942657470703125 2023-01-21 12:51:16.749867: step: 684/529, loss: 0.011702632531523705 2023-01-21 12:51:17.943979: step: 688/529, loss: 0.04537677764892578 2023-01-21 12:51:19.131600: step: 692/529, loss: 0.0002615928533487022 2023-01-21 12:51:20.310170: step: 696/529, loss: 0.005172061733901501 2023-01-21 12:51:21.491445: step: 700/529, loss: 0.019414568319916725 2023-01-21 12:51:22.688294: step: 704/529, loss: 0.0023340226616710424 2023-01-21 12:51:23.984793: step: 708/529, loss: 0.023093223571777344 2023-01-21 12:51:25.142628: step: 712/529, loss: 0.07595959305763245 2023-01-21 12:51:26.336056: step: 716/529, loss: 5.970001075183973e-05 2023-01-21 12:51:27.531463: step: 720/529, loss: 0.024978255853056908 2023-01-21 12:51:28.714988: step: 724/529, loss: 0.004244994837790728 2023-01-21 12:51:29.892581: step: 728/529, loss: 0.04862070083618164 2023-01-21 12:51:31.070169: step: 732/529, loss: 4.4441225327318534e-05 2023-01-21 12:51:32.224156: step: 736/529, loss: 0.002824401715770364 2023-01-21 12:51:33.383013: step: 740/529, loss: 0.0015510559314861894 2023-01-21 12:51:34.571449: step: 744/529, loss: 0.0005427360301837325 2023-01-21 12:51:35.768400: step: 748/529, loss: 0.016252899542450905 2023-01-21 12:51:36.989680: step: 752/529, loss: 0.08495263010263443 2023-01-21 12:51:38.179847: step: 756/529, loss: 0.003172254655510187 2023-01-21 12:51:39.374672: step: 760/529, loss: 0.0015342712868005037 2023-01-21 12:51:40.592596: step: 764/529, loss: 0.00022029876708984375 2023-01-21 12:51:41.782085: step: 768/529, loss: 0.00035686494084075093 2023-01-21 12:51:42.993204: step: 772/529, loss: 0.0008879661909304559 2023-01-21 12:51:44.190087: step: 776/529, loss: 0.0001375198335153982 2023-01-21 12:51:45.309323: step: 780/529, loss: 0.011178922839462757 2023-01-21 12:51:46.488096: step: 784/529, loss: 0.0025388719514012337 2023-01-21 12:51:47.644099: step: 788/529, loss: 0.002006149385124445 2023-01-21 12:51:48.930585: step: 792/529, loss: 0.0040136342868208885 2023-01-21 12:51:50.110510: step: 796/529, loss: 0.06336746364831924 2023-01-21 12:51:51.333924: step: 800/529, loss: 0.03689708933234215 2023-01-21 12:51:52.530125: step: 804/529, loss: 0.016944121569395065 2023-01-21 12:51:53.720247: step: 808/529, loss: 0.0004863739013671875 2023-01-21 12:51:54.909944: step: 812/529, loss: 0.019335174933075905 2023-01-21 12:51:56.119002: step: 816/529, loss: 0.054054830223321915 2023-01-21 12:51:57.320010: step: 820/529, loss: 0.0015951156383380294 2023-01-21 12:51:58.499483: step: 824/529, loss: 0.04383373260498047 2023-01-21 12:51:59.688737: step: 828/529, loss: 0.0003765106375794858 2023-01-21 12:52:00.858559: step: 832/529, loss: 0.05890718474984169 2023-01-21 12:52:02.127573: step: 836/529, loss: 0.0010253905784338713 2023-01-21 12:52:03.288973: step: 840/529, loss: 0.07326450943946838 2023-01-21 12:52:04.491240: step: 844/529, loss: 6.394386582542211e-05 2023-01-21 12:52:05.708715: step: 848/529, loss: 6.518364534713328e-05 2023-01-21 12:52:06.898181: step: 852/529, loss: 0.00027294160099700093 2023-01-21 12:52:08.093887: step: 856/529, loss: 0.0026446343399584293 2023-01-21 12:52:09.291246: step: 860/529, loss: 0.009461402893066406 2023-01-21 12:52:10.478632: step: 864/529, loss: 0.025462724268436432 2023-01-21 12:52:11.685235: step: 868/529, loss: 0.011704540811479092 2023-01-21 12:52:12.887086: step: 872/529, loss: 4.9400332500226796e-05 2023-01-21 12:52:14.046977: step: 876/529, loss: 0.0022995946928858757 2023-01-21 12:52:15.238873: step: 880/529, loss: 0.004719448275864124 2023-01-21 12:52:16.418426: step: 884/529, loss: 0.14319248497486115 2023-01-21 12:52:17.592669: step: 888/529, loss: 1.850128137448337e-05 2023-01-21 12:52:18.788700: step: 892/529, loss: 0.03753848001360893 2023-01-21 12:52:20.000510: step: 896/529, loss: 0.012701893225312233 2023-01-21 12:52:21.229532: step: 900/529, loss: 0.0013249159092083573 2023-01-21 12:52:22.423812: step: 904/529, loss: 0.00012512208195403218 2023-01-21 12:52:23.590401: step: 908/529, loss: 0.00013785957708023489 2023-01-21 12:52:24.789436: step: 912/529, loss: 0.030365657061338425 2023-01-21 12:52:25.955097: step: 916/529, loss: 8.258819434558973e-05 2023-01-21 12:52:27.133532: step: 920/529, loss: 0.0012405395973473787 2023-01-21 12:52:28.344493: step: 924/529, loss: 0.013542176224291325 2023-01-21 12:52:29.537451: step: 928/529, loss: 0.0004238128603901714 2023-01-21 12:52:30.726680: step: 932/529, loss: 0.0019830705132335424 2023-01-21 12:52:31.891895: step: 936/529, loss: 0.0016424179775640368 2023-01-21 12:52:33.056413: step: 940/529, loss: 4.348754737293348e-05 2023-01-21 12:52:34.231078: step: 944/529, loss: 0.03330859914422035 2023-01-21 12:52:35.385870: step: 948/529, loss: 0.0001504898100392893 2023-01-21 12:52:36.552319: step: 952/529, loss: 0.002265429589897394 2023-01-21 12:52:37.753651: step: 956/529, loss: 2.059936559817288e-05 2023-01-21 12:52:38.912258: step: 960/529, loss: 0.0007035851594991982 2023-01-21 12:52:40.087195: step: 964/529, loss: 0.0048697469756007195 2023-01-21 12:52:41.276169: step: 968/529, loss: 0.0015480995643883944 2023-01-21 12:52:42.484442: step: 972/529, loss: 0.002645349595695734 2023-01-21 12:52:43.693934: step: 976/529, loss: 0.000987148261629045 2023-01-21 12:52:44.855781: step: 980/529, loss: 0.011107087135314941 2023-01-21 12:52:46.080958: step: 984/529, loss: 0.16761799156665802 2023-01-21 12:52:47.283753: step: 988/529, loss: 0.00363674177788198 2023-01-21 12:52:48.466114: step: 992/529, loss: 0.0010346651542931795 2023-01-21 12:52:49.662553: step: 996/529, loss: 0.022006798535585403 2023-01-21 12:52:50.857212: step: 1000/529, loss: 0.01878967322409153 2023-01-21 12:52:52.058300: step: 1004/529, loss: 0.001506519503891468 2023-01-21 12:52:53.261019: step: 1008/529, loss: 1.1587143490032759e-05 2023-01-21 12:52:54.402840: step: 1012/529, loss: 0.033208273351192474 2023-01-21 12:52:55.544751: step: 1016/529, loss: 0.04389353096485138 2023-01-21 12:52:56.719315: step: 1020/529, loss: 0.0032369615510106087 2023-01-21 12:52:57.912035: step: 1024/529, loss: 0.05927124246954918 2023-01-21 12:52:59.061992: step: 1028/529, loss: 0.06422028690576553 2023-01-21 12:53:00.220158: step: 1032/529, loss: 9.937286813510582e-05 2023-01-21 12:53:01.443682: step: 1036/529, loss: 0.005011939909309149 2023-01-21 12:53:02.682476: step: 1040/529, loss: 0.005518722347915173 2023-01-21 12:53:03.907310: step: 1044/529, loss: 0.09006920456886292 2023-01-21 12:53:05.068224: step: 1048/529, loss: 0.003423976944759488 2023-01-21 12:53:06.272377: step: 1052/529, loss: 0.016025066375732422 2023-01-21 12:53:07.480917: step: 1056/529, loss: 1.1253356205997989e-05 2023-01-21 12:53:08.686153: step: 1060/529, loss: 0.009554862976074219 2023-01-21 12:53:09.895904: step: 1064/529, loss: 0.0010534287430346012 2023-01-21 12:53:11.085625: step: 1068/529, loss: 0.000583648681640625 2023-01-21 12:53:12.272123: step: 1072/529, loss: 0.0005009651649743319 2023-01-21 12:53:13.456751: step: 1076/529, loss: 0.0002645492786541581 2023-01-21 12:53:14.669220: step: 1080/529, loss: 0.019129563122987747 2023-01-21 12:53:15.867776: step: 1084/529, loss: 0.0005270004621706903 2023-01-21 12:53:17.052055: step: 1088/529, loss: 0.00031490327091887593 2023-01-21 12:53:18.272365: step: 1092/529, loss: 0.0001180648832814768 2023-01-21 12:53:19.465240: step: 1096/529, loss: 0.017127513885498047 2023-01-21 12:53:20.689527: step: 1100/529, loss: 0.0007427215459756553 2023-01-21 12:53:21.859669: step: 1104/529, loss: 0.010594367980957031 2023-01-21 12:53:23.092171: step: 1108/529, loss: 0.004141903016716242 2023-01-21 12:53:24.264821: step: 1112/529, loss: 0.5100986361503601 2023-01-21 12:53:25.442925: step: 1116/529, loss: 0.0015306472778320312 2023-01-21 12:53:26.697233: step: 1120/529, loss: 0.07941456139087677 2023-01-21 12:53:27.909577: step: 1124/529, loss: 0.004347038455307484 2023-01-21 12:53:29.110425: step: 1128/529, loss: 0.03517203405499458 2023-01-21 12:53:30.288113: step: 1132/529, loss: 0.00018129349336959422 2023-01-21 12:53:31.475373: step: 1136/529, loss: 0.003693962236866355 2023-01-21 12:53:32.678976: step: 1140/529, loss: 0.026404380798339844 2023-01-21 12:53:33.908549: step: 1144/529, loss: 6.046295311534777e-05 2023-01-21 12:53:35.147511: step: 1148/529, loss: 0.01632366143167019 2023-01-21 12:53:36.318851: step: 1152/529, loss: 0.015412474051117897 2023-01-21 12:53:37.511768: step: 1156/529, loss: 0.001611566636711359 2023-01-21 12:53:38.713751: step: 1160/529, loss: 0.0003528594970703125 2023-01-21 12:53:39.893317: step: 1164/529, loss: 0.06460094451904297 2023-01-21 12:53:41.108271: step: 1168/529, loss: 0.002927589463070035 2023-01-21 12:53:42.298484: step: 1172/529, loss: 0.028927041217684746 2023-01-21 12:53:43.505468: step: 1176/529, loss: 0.03572044521570206 2023-01-21 12:53:44.681592: step: 1180/529, loss: 0.019527077674865723 2023-01-21 12:53:45.838825: step: 1184/529, loss: 0.014702415093779564 2023-01-21 12:53:47.024640: step: 1188/529, loss: 0.010488605126738548 2023-01-21 12:53:48.278227: step: 1192/529, loss: 0.001035499619320035 2023-01-21 12:53:49.479867: step: 1196/529, loss: 0.04453869163990021 2023-01-21 12:53:50.673008: step: 1200/529, loss: 0.006957626435905695 2023-01-21 12:53:51.845284: step: 1204/529, loss: 0.002349948976188898 2023-01-21 12:53:53.066890: step: 1208/529, loss: 0.0002538680855650455 2023-01-21 12:53:54.214599: step: 1212/529, loss: 0.0028120994102209806 2023-01-21 12:53:55.415205: step: 1216/529, loss: 0.004297828767448664 2023-01-21 12:53:56.627175: step: 1220/529, loss: 0.003631210420280695 2023-01-21 12:53:57.789285: step: 1224/529, loss: 0.00024385451979469508 2023-01-21 12:53:58.992648: step: 1228/529, loss: 0.02272777631878853 2023-01-21 12:54:00.168191: step: 1232/529, loss: 0.27537012100219727 2023-01-21 12:54:01.359038: step: 1236/529, loss: 0.015215015970170498 2023-01-21 12:54:02.550562: step: 1240/529, loss: 0.007067585363984108 2023-01-21 12:54:03.711770: step: 1244/529, loss: 0.0032721522729843855 2023-01-21 12:54:04.938342: step: 1248/529, loss: 0.002231597900390625 2023-01-21 12:54:06.168742: step: 1252/529, loss: 0.0018440247513353825 2023-01-21 12:54:07.377369: step: 1256/529, loss: 6.961822509765625e-05 2023-01-21 12:54:08.536767: step: 1260/529, loss: 3.719329833984375e-05 2023-01-21 12:54:09.755861: step: 1264/529, loss: 0.008255720138549805 2023-01-21 12:54:10.971802: step: 1268/529, loss: 0.18575401604175568 2023-01-21 12:54:12.168594: step: 1272/529, loss: 0.0010348320938646793 2023-01-21 12:54:13.388803: step: 1276/529, loss: 0.0012639999622479081 2023-01-21 12:54:14.641893: step: 1280/529, loss: 0.002755546709522605 2023-01-21 12:54:15.776115: step: 1284/529, loss: 0.011878490447998047 2023-01-21 12:54:16.938257: step: 1288/529, loss: 0.030170299112796783 2023-01-21 12:54:18.148240: step: 1292/529, loss: 0.05582933500409126 2023-01-21 12:54:19.431540: step: 1296/529, loss: 0.014387655071914196 2023-01-21 12:54:20.629618: step: 1300/529, loss: 0.0024883272126317024 2023-01-21 12:54:21.799552: step: 1304/529, loss: 0.025716353207826614 2023-01-21 12:54:22.967802: step: 1308/529, loss: 7.648468454135582e-05 2023-01-21 12:54:24.116962: step: 1312/529, loss: 2.1266936528263614e-05 2023-01-21 12:54:25.304183: step: 1316/529, loss: 0.00329341902397573 2023-01-21 12:54:26.469453: step: 1320/529, loss: 6.847381155239418e-05 2023-01-21 12:54:27.672635: step: 1324/529, loss: 0.00016870498075149953 2023-01-21 12:54:28.882859: step: 1328/529, loss: 5.14984130859375e-05 2023-01-21 12:54:30.031645: step: 1332/529, loss: 0.036052536219358444 2023-01-21 12:54:31.242418: step: 1336/529, loss: 0.000446319580078125 2023-01-21 12:54:32.423908: step: 1340/529, loss: 0.003314495086669922 2023-01-21 12:54:33.554885: step: 1344/529, loss: 0.0031042098999023438 2023-01-21 12:54:34.733111: step: 1348/529, loss: 0.0008060456020757556 2023-01-21 12:54:35.941011: step: 1352/529, loss: 0.001272010849788785 2023-01-21 12:54:37.174961: step: 1356/529, loss: 0.06508941948413849 2023-01-21 12:54:38.374224: step: 1360/529, loss: 0.00029096603975631297 2023-01-21 12:54:39.553870: step: 1364/529, loss: 0.00904998742043972 2023-01-21 12:54:40.747410: step: 1368/529, loss: 0.00498199462890625 2023-01-21 12:54:41.965250: step: 1372/529, loss: 0.002743625780567527 2023-01-21 12:54:43.206888: step: 1376/529, loss: 0.07878484576940536 2023-01-21 12:54:44.400445: step: 1380/529, loss: 9.855032112682238e-05 2023-01-21 12:54:45.608079: step: 1384/529, loss: 2.269744800287299e-05 2023-01-21 12:54:46.822210: step: 1388/529, loss: 0.0020182610023766756 2023-01-21 12:54:48.004238: step: 1392/529, loss: 0.008685683831572533 2023-01-21 12:54:49.221031: step: 1396/529, loss: 0.0031312943901866674 2023-01-21 12:54:50.438772: step: 1400/529, loss: 0.034381963312625885 2023-01-21 12:54:51.653189: step: 1404/529, loss: 0.01310033816844225 2023-01-21 12:54:52.835157: step: 1408/529, loss: 0.0005679130554199219 2023-01-21 12:54:54.013272: step: 1412/529, loss: 3.128051685052924e-05 2023-01-21 12:54:55.207031: step: 1416/529, loss: 0.06314654648303986 2023-01-21 12:54:56.407028: step: 1420/529, loss: 0.00017089843458961695 2023-01-21 12:54:57.615273: step: 1424/529, loss: 0.0020983696449548006 2023-01-21 12:54:58.840623: step: 1428/529, loss: 0.004392432980239391 2023-01-21 12:55:00.034741: step: 1432/529, loss: 0.5976591110229492 2023-01-21 12:55:01.181588: step: 1436/529, loss: 0.00067224504891783 2023-01-21 12:55:02.380138: step: 1440/529, loss: 0.0054450989700853825 2023-01-21 12:55:03.576596: step: 1444/529, loss: 0.023593807592988014 2023-01-21 12:55:04.808856: step: 1448/529, loss: 0.0015541077591478825 2023-01-21 12:55:06.003150: step: 1452/529, loss: 0.051627349108457565 2023-01-21 12:55:07.189040: step: 1456/529, loss: 0.0075782774947583675 2023-01-21 12:55:08.386141: step: 1460/529, loss: 0.02765197679400444 2023-01-21 12:55:09.535171: step: 1464/529, loss: 0.0008444786071777344 2023-01-21 12:55:10.714086: step: 1468/529, loss: 0.00035037993802689016 2023-01-21 12:55:11.912832: step: 1472/529, loss: 0.00997009314596653 2023-01-21 12:55:13.103920: step: 1476/529, loss: 0.11177940666675568 2023-01-21 12:55:14.312503: step: 1480/529, loss: 0.008936119265854359 2023-01-21 12:55:15.498377: step: 1484/529, loss: 0.0004549503210000694 2023-01-21 12:55:16.726757: step: 1488/529, loss: 0.0005990028730593622 2023-01-21 12:55:17.859164: step: 1492/529, loss: 0.03421497344970703 2023-01-21 12:55:19.027774: step: 1496/529, loss: 0.0045372010208666325 2023-01-21 12:55:20.200674: step: 1500/529, loss: 0.2617638409137726 2023-01-21 12:55:21.355839: step: 1504/529, loss: 0.0007524490938521922 2023-01-21 12:55:22.540223: step: 1508/529, loss: 1.6832351320772432e-05 2023-01-21 12:55:23.745077: step: 1512/529, loss: 0.0016531944274902344 2023-01-21 12:55:24.935734: step: 1516/529, loss: 0.06861049681901932 2023-01-21 12:55:26.111741: step: 1520/529, loss: 0.0664815902709961 2023-01-21 12:55:27.331153: step: 1524/529, loss: 0.001583099365234375 2023-01-21 12:55:28.489127: step: 1528/529, loss: 0.0028020860627293587 2023-01-21 12:55:29.657631: step: 1532/529, loss: 0.0009150505065917969 2023-01-21 12:55:30.857270: step: 1536/529, loss: 0.009389019571244717 2023-01-21 12:55:32.040454: step: 1540/529, loss: 0.0004385947831906378 2023-01-21 12:55:33.255069: step: 1544/529, loss: 0.005563164129853249 2023-01-21 12:55:34.409586: step: 1548/529, loss: 0.00017862320237327367 2023-01-21 12:55:35.586851: step: 1552/529, loss: 0.0131384851410985 2023-01-21 12:55:36.805314: step: 1556/529, loss: 0.013698577880859375 2023-01-21 12:55:37.997616: step: 1560/529, loss: 0.00808407086879015 2023-01-21 12:55:39.218662: step: 1564/529, loss: 0.00026960374088957906 2023-01-21 12:55:40.422349: step: 1568/529, loss: 0.004306221380829811 2023-01-21 12:55:41.602706: step: 1572/529, loss: 0.004226732533425093 2023-01-21 12:55:42.789380: step: 1576/529, loss: 0.0013734340900555253 2023-01-21 12:55:43.963500: step: 1580/529, loss: 0.007473850157111883 2023-01-21 12:55:45.156246: step: 1584/529, loss: 0.017133522778749466 2023-01-21 12:55:46.296562: step: 1588/529, loss: 0.007586574647575617 2023-01-21 12:55:47.454482: step: 1592/529, loss: 6.198883056640625e-05 2023-01-21 12:55:48.658779: step: 1596/529, loss: 0.0015925407642498612 2023-01-21 12:55:49.832843: step: 1600/529, loss: 0.021272659301757812 2023-01-21 12:55:51.023792: step: 1604/529, loss: 0.013446665368974209 2023-01-21 12:55:52.222225: step: 1608/529, loss: 0.032831527292728424 2023-01-21 12:55:53.460126: step: 1612/529, loss: 0.07146281749010086 2023-01-21 12:55:54.642497: step: 1616/529, loss: 0.009949398227036 2023-01-21 12:55:55.818343: step: 1620/529, loss: 9.441375732421875e-05 2023-01-21 12:55:57.026423: step: 1624/529, loss: 0.037702418863773346 2023-01-21 12:55:58.221581: step: 1628/529, loss: 0.0012516975402832031 2023-01-21 12:55:59.409824: step: 1632/529, loss: 7.219314284157008e-05 2023-01-21 12:56:00.610573: step: 1636/529, loss: 0.03187829628586769 2023-01-21 12:56:01.788898: step: 1640/529, loss: 0.006687736604362726 2023-01-21 12:56:02.993876: step: 1644/529, loss: 0.002936935517936945 2023-01-21 12:56:04.166379: step: 1648/529, loss: 0.003667736193165183 2023-01-21 12:56:05.364588: step: 1652/529, loss: 0.002071523806080222 2023-01-21 12:56:06.590071: step: 1656/529, loss: 0.013837242498993874 2023-01-21 12:56:07.790443: step: 1660/529, loss: 0.44162482023239136 2023-01-21 12:56:08.964132: step: 1664/529, loss: 6.227493577171117e-05 2023-01-21 12:56:10.160606: step: 1668/529, loss: 0.007802486419677734 2023-01-21 12:56:11.361277: step: 1672/529, loss: 0.009524345397949219 2023-01-21 12:56:12.555594: step: 1676/529, loss: 0.007470607757568359 2023-01-21 12:56:13.730642: step: 1680/529, loss: 0.005115986336022615 2023-01-21 12:56:14.919549: step: 1684/529, loss: 6.771088010282256e-06 2023-01-21 12:56:16.092840: step: 1688/529, loss: 0.0024892569053918123 2023-01-21 12:56:17.289550: step: 1692/529, loss: 0.0007131576421670616 2023-01-21 12:56:18.451841: step: 1696/529, loss: 0.034295033663511276 2023-01-21 12:56:19.612465: step: 1700/529, loss: 6.942749314475805e-05 2023-01-21 12:56:20.871643: step: 1704/529, loss: 0.013085603713989258 2023-01-21 12:56:22.099950: step: 1708/529, loss: 0.019481658935546875 2023-01-21 12:56:23.282387: step: 1712/529, loss: 0.00022064447693992406 2023-01-21 12:56:24.460487: step: 1716/529, loss: 0.1772640198469162 2023-01-21 12:56:25.653507: step: 1720/529, loss: 0.0050002094358205795 2023-01-21 12:56:26.832537: step: 1724/529, loss: 0.004858207888901234 2023-01-21 12:56:28.021414: step: 1728/529, loss: 0.009202385321259499 2023-01-21 12:56:29.183378: step: 1732/529, loss: 2.2602080207434483e-05 2023-01-21 12:56:30.387377: step: 1736/529, loss: -7.629394644936838e-07 2023-01-21 12:56:31.548379: step: 1740/529, loss: 0.007745265960693359 2023-01-21 12:56:32.695064: step: 1744/529, loss: 0.00013227463932707906 2023-01-21 12:56:33.944574: step: 1748/529, loss: 0.0014761924976482987 2023-01-21 12:56:35.164864: step: 1752/529, loss: 0.00010704994201660156 2023-01-21 12:56:36.372394: step: 1756/529, loss: 0.0009677886846475303 2023-01-21 12:56:37.582018: step: 1760/529, loss: 0.0005519867409020662 2023-01-21 12:56:38.725203: step: 1764/529, loss: 0.02731151506304741 2023-01-21 12:56:39.892018: step: 1768/529, loss: 0.0007440566550940275 2023-01-21 12:56:41.086393: step: 1772/529, loss: 0.017073439434170723 2023-01-21 12:56:42.276071: step: 1776/529, loss: 0.016886521130800247 2023-01-21 12:56:43.446647: step: 1780/529, loss: 0.0002614021359477192 2023-01-21 12:56:44.628141: step: 1784/529, loss: 0.012561511248350143 2023-01-21 12:56:45.864735: step: 1788/529, loss: 1.0352295637130737 2023-01-21 12:56:47.092065: step: 1792/529, loss: 1.6075966358184814 2023-01-21 12:56:48.258662: step: 1796/529, loss: 1.2111664545955136e-05 2023-01-21 12:56:49.472904: step: 1800/529, loss: 5.1212311518611386e-05 2023-01-21 12:56:50.678771: step: 1804/529, loss: 6.089210728532635e-05 2023-01-21 12:56:51.847905: step: 1808/529, loss: 0.0014251709217205644 2023-01-21 12:56:53.053862: step: 1812/529, loss: 0.013620949350297451 2023-01-21 12:56:54.257869: step: 1816/529, loss: 0.018726635724306107 2023-01-21 12:56:55.435429: step: 1820/529, loss: 0.0004659652477130294 2023-01-21 12:56:56.653753: step: 1824/529, loss: 0.0474826842546463 2023-01-21 12:56:57.861548: step: 1828/529, loss: 2.1457672119140625e-06 2023-01-21 12:56:59.070435: step: 1832/529, loss: 0.26096248626708984 2023-01-21 12:57:00.255204: step: 1836/529, loss: 0.0018550873501226306 2023-01-21 12:57:01.461641: step: 1840/529, loss: 0.02120218425989151 2023-01-21 12:57:02.678883: step: 1844/529, loss: 0.001298332354053855 2023-01-21 12:57:03.856612: step: 1848/529, loss: 0.0010971069568768144 2023-01-21 12:57:05.073509: step: 1852/529, loss: 8.497238741256297e-05 2023-01-21 12:57:06.270978: step: 1856/529, loss: 6.923675391590223e-05 2023-01-21 12:57:07.467241: step: 1860/529, loss: 0.0001624107244424522 2023-01-21 12:57:08.649869: step: 1864/529, loss: 0.00016980171494651586 2023-01-21 12:57:09.836158: step: 1868/529, loss: 0.00019006729417014867 2023-01-21 12:57:11.026104: step: 1872/529, loss: 0.08499736338853836 2023-01-21 12:57:12.216836: step: 1876/529, loss: 0.003906059078872204 2023-01-21 12:57:13.415449: step: 1880/529, loss: 0.025139618664979935 2023-01-21 12:57:14.589595: step: 1884/529, loss: 0.044082071632146835 2023-01-21 12:57:15.766426: step: 1888/529, loss: 0.004578971769660711 2023-01-21 12:57:16.967458: step: 1892/529, loss: 0.013745593838393688 2023-01-21 12:57:18.162578: step: 1896/529, loss: 0.0011387825943529606 2023-01-21 12:57:19.332568: step: 1900/529, loss: 0.0006008148193359375 2023-01-21 12:57:20.515915: step: 1904/529, loss: 0.016133690252900124 2023-01-21 12:57:21.714811: step: 1908/529, loss: 0.0001794815034372732 2023-01-21 12:57:22.897078: step: 1912/529, loss: 9.493828110862523e-05 2023-01-21 12:57:24.119793: step: 1916/529, loss: 0.006560373120009899 2023-01-21 12:57:25.326473: step: 1920/529, loss: 0.12894611060619354 2023-01-21 12:57:26.513346: step: 1924/529, loss: 4.7349931264761835e-05 2023-01-21 12:57:27.705090: step: 1928/529, loss: 0.001413536025211215 2023-01-21 12:57:28.879176: step: 1932/529, loss: 0.0014220238663256168 2023-01-21 12:57:30.066098: step: 1936/529, loss: 0.00019788742065429688 2023-01-21 12:57:31.268905: step: 1940/529, loss: 0.0005460738902911544 2023-01-21 12:57:32.465814: step: 1944/529, loss: 0.0019762038718909025 2023-01-21 12:57:33.653045: step: 1948/529, loss: 0.042997173964977264 2023-01-21 12:57:34.832278: step: 1952/529, loss: 0.04138369485735893 2023-01-21 12:57:36.095341: step: 1956/529, loss: 0.006378364749252796 2023-01-21 12:57:37.267835: step: 1960/529, loss: 0.0005731344572268426 2023-01-21 12:57:38.472633: step: 1964/529, loss: 0.007376671303063631 2023-01-21 12:57:39.706716: step: 1968/529, loss: 0.0012138367164880037 2023-01-21 12:57:40.875914: step: 1972/529, loss: 0.00034012793912552297 2023-01-21 12:57:42.080846: step: 1976/529, loss: 0.33268243074417114 2023-01-21 12:57:43.297768: step: 1980/529, loss: -5.483627774083288e-07 2023-01-21 12:57:44.514925: step: 1984/529, loss: 0.016938399523496628 2023-01-21 12:57:45.681829: step: 1988/529, loss: 0.009025001898407936 2023-01-21 12:57:46.857511: step: 1992/529, loss: 0.005885410588234663 2023-01-21 12:57:48.025555: step: 1996/529, loss: 9.384154691360891e-05 2023-01-21 12:57:49.249006: step: 2000/529, loss: 0.08298740535974503 2023-01-21 12:57:50.453024: step: 2004/529, loss: 0.00032634736271575093 2023-01-21 12:57:51.633317: step: 2008/529, loss: 0.13261787593364716 2023-01-21 12:57:52.798280: step: 2012/529, loss: 0.033293724060058594 2023-01-21 12:57:53.976686: step: 2016/529, loss: 0.00035409926204010844 2023-01-21 12:57:55.162548: step: 2020/529, loss: 0.0011157034896314144 2023-01-21 12:57:56.331546: step: 2024/529, loss: 0.001242351601831615 2023-01-21 12:57:57.552420: step: 2028/529, loss: 0.009975624270737171 2023-01-21 12:57:58.777667: step: 2032/529, loss: 0.00020103454880882055 2023-01-21 12:57:59.936229: step: 2036/529, loss: 0.42387503385543823 2023-01-21 12:58:01.129140: step: 2040/529, loss: 0.018227767199277878 2023-01-21 12:58:02.338273: step: 2044/529, loss: 0.00041770937968976796 2023-01-21 12:58:03.532242: step: 2048/529, loss: 0.08130493015050888 2023-01-21 12:58:04.726890: step: 2052/529, loss: 0.023687362670898438 2023-01-21 12:58:05.918675: step: 2056/529, loss: 0.0005147934425622225 2023-01-21 12:58:07.108882: step: 2060/529, loss: 0.009633636102080345 2023-01-21 12:58:08.290758: step: 2064/529, loss: 0.009859466925263405 2023-01-21 12:58:09.455301: step: 2068/529, loss: 0.0032457353081554174 2023-01-21 12:58:10.651216: step: 2072/529, loss: 0.0010078430641442537 2023-01-21 12:58:11.853997: step: 2076/529, loss: 0.013354111462831497 2023-01-21 12:58:13.034329: step: 2080/529, loss: 0.0013794898986816406 2023-01-21 12:58:14.219625: step: 2084/529, loss: 0.0003688812139444053 2023-01-21 12:58:15.386794: step: 2088/529, loss: 0.00066123012220487 2023-01-21 12:58:16.530113: step: 2092/529, loss: 2.5558472771081142e-05 2023-01-21 12:58:17.705070: step: 2096/529, loss: 0.021580219268798828 2023-01-21 12:58:18.939798: step: 2100/529, loss: 0.0540035255253315 2023-01-21 12:58:20.144535: step: 2104/529, loss: 0.012967204675078392 2023-01-21 12:58:21.320910: step: 2108/529, loss: 0.044558335095644 2023-01-21 12:58:22.514936: step: 2112/529, loss: 0.005290317814797163 2023-01-21 12:58:23.726150: step: 2116/529, loss: 0.004627418704330921 ================================================== Loss: 0.026 -------------------- Dev: {'event': {'p': 0.5983772819472617, 'r': 0.7856191744340879, 'f1': 0.6793321819228555}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Test: {'event': {'p': 0.6215456674473068, 'r': 0.7950868783702816, 'f1': 0.6976866456361724}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Chinese: {'event': {'p': 0.6103896103896104, 'r': 0.8703703703703703, 'f1': 0.7175572519083969}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Korean: {'event': {'p': 0.5645161290322581, 'r': 0.5555555555555556, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Russian: {'event': {'p': 0.4418604651162791, 'r': 0.5277777777777778, 'f1': 0.4810126582278481}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 12:59:07.310505: step: 4/529, loss: 0.0011466980213299394 2023-01-21 12:59:08.527474: step: 8/529, loss: 0.011140919290482998 2023-01-21 12:59:09.717719: step: 12/529, loss: 0.021004866808652878 2023-01-21 12:59:10.924431: step: 16/529, loss: 0.0010696412064135075 2023-01-21 12:59:12.113857: step: 20/529, loss: 5.855560448253527e-05 2023-01-21 12:59:13.344283: step: 24/529, loss: -1.106262243411038e-05 2023-01-21 12:59:14.542330: step: 28/529, loss: 0.001407623291015625 2023-01-21 12:59:15.717504: step: 32/529, loss: 0.0002246856747660786 2023-01-21 12:59:16.909750: step: 36/529, loss: 0.0006568909157067537 2023-01-21 12:59:18.089794: step: 40/529, loss: 0.0017924308776855469 2023-01-21 12:59:19.290695: step: 44/529, loss: 0.0008139610290527344 2023-01-21 12:59:20.482542: step: 48/529, loss: 0.000766754150390625 2023-01-21 12:59:21.684442: step: 52/529, loss: 0.00024299623328261077 2023-01-21 12:59:22.877742: step: 56/529, loss: 0.019057463854551315 2023-01-21 12:59:24.053547: step: 60/529, loss: 0.0006607055547647178 2023-01-21 12:59:25.288260: step: 64/529, loss: 0.002446079161018133 2023-01-21 12:59:26.472782: step: 68/529, loss: 2.384185791015625e-05 2023-01-21 12:59:27.652418: step: 72/529, loss: 0.0002698898606467992 2023-01-21 12:59:28.856320: step: 76/529, loss: 0.002190876053646207 2023-01-21 12:59:30.025081: step: 80/529, loss: 0.0023322582710534334 2023-01-21 12:59:31.187815: step: 84/529, loss: 0.0057866573333740234 2023-01-21 12:59:32.396542: step: 88/529, loss: 0.0017610550858080387 2023-01-21 12:59:33.549218: step: 92/529, loss: 0.007925319485366344 2023-01-21 12:59:34.774510: step: 96/529, loss: -3.261566234868951e-05 2023-01-21 12:59:35.964882: step: 100/529, loss: 0.08399781584739685 2023-01-21 12:59:37.174696: step: 104/529, loss: 0.0006403923034667969 2023-01-21 12:59:38.346471: step: 108/529, loss: 5.207061622058973e-05 2023-01-21 12:59:39.510360: step: 112/529, loss: 0.0011971474159508944 2023-01-21 12:59:40.682220: step: 116/529, loss: 0.006357765290886164 2023-01-21 12:59:41.832134: step: 120/529, loss: 0.00043311118497513235 2023-01-21 12:59:43.028166: step: 124/529, loss: 0.000164031982421875 2023-01-21 12:59:44.272243: step: 128/529, loss: 0.022243691608309746 2023-01-21 12:59:45.467164: step: 132/529, loss: 0.0003643035888671875 2023-01-21 12:59:46.663404: step: 136/529, loss: 0.19191579520702362 2023-01-21 12:59:47.950216: step: 140/529, loss: 0.0017140388954430819 2023-01-21 12:59:49.103410: step: 144/529, loss: 0.004242515657097101 2023-01-21 12:59:50.293246: step: 148/529, loss: 0.04884643480181694 2023-01-21 12:59:51.489339: step: 152/529, loss: 0.05515914410352707 2023-01-21 12:59:52.701382: step: 156/529, loss: 0.005044078920036554 2023-01-21 12:59:53.875406: step: 160/529, loss: 0.00034618377685546875 2023-01-21 12:59:55.097315: step: 164/529, loss: 0.00033988954965025187 2023-01-21 12:59:56.276944: step: 168/529, loss: 2.136230432370212e-05 2023-01-21 12:59:57.466801: step: 172/529, loss: 0.00537109375 2023-01-21 12:59:58.675668: step: 176/529, loss: 0.0037658691871911287 2023-01-21 12:59:59.857689: step: 180/529, loss: 0.005347347352653742 2023-01-21 13:00:01.106709: step: 184/529, loss: 6.732941255904734e-05 2023-01-21 13:00:02.285170: step: 188/529, loss: 2.441406286379788e-05 2023-01-21 13:00:03.449468: step: 192/529, loss: 0.02805481106042862 2023-01-21 13:00:04.614111: step: 196/529, loss: 0.0001373291015625 2023-01-21 13:00:05.800444: step: 200/529, loss: 0.00017805099196266383 2023-01-21 13:00:06.997436: step: 204/529, loss: 0.015758514404296875 2023-01-21 13:00:08.236978: step: 208/529, loss: 0.21355171501636505 2023-01-21 13:00:09.399082: step: 212/529, loss: 0.00011086463928222656 2023-01-21 13:00:10.568356: step: 216/529, loss: 0.006464862730354071 2023-01-21 13:00:11.763505: step: 220/529, loss: 0.0763818770647049 2023-01-21 13:00:12.972037: step: 224/529, loss: 0.08663111180067062 2023-01-21 13:00:14.128417: step: 228/529, loss: 0.010535096749663353 2023-01-21 13:00:15.283385: step: 232/529, loss: 0.020877551287412643 2023-01-21 13:00:16.507741: step: 236/529, loss: 0.01925039477646351 2023-01-21 13:00:17.669287: step: 240/529, loss: 0.004262924659997225 2023-01-21 13:00:18.841815: step: 244/529, loss: 0.03810299560427666 2023-01-21 13:00:20.009424: step: 248/529, loss: 0.00286521902307868 2023-01-21 13:00:21.240260: step: 252/529, loss: 0.08436579257249832 2023-01-21 13:00:22.428373: step: 256/529, loss: 0.03114328533411026 2023-01-21 13:00:23.610130: step: 260/529, loss: 0.006331252865493298 2023-01-21 13:00:24.841388: step: 264/529, loss: 0.01859569549560547 2023-01-21 13:00:25.998582: step: 268/529, loss: 0.0009475707774981856 2023-01-21 13:00:27.169611: step: 272/529, loss: 0.0001999855157919228 2023-01-21 13:00:28.353944: step: 276/529, loss: 0.014757824130356312 2023-01-21 13:00:29.520545: step: 280/529, loss: 0.00096893310546875 2023-01-21 13:00:30.689230: step: 284/529, loss: 0.00014920235844329 2023-01-21 13:00:31.939105: step: 288/529, loss: 0.031137848272919655 2023-01-21 13:00:33.108584: step: 292/529, loss: 0.010267352685332298 2023-01-21 13:00:34.286947: step: 296/529, loss: 0.012238883413374424 2023-01-21 13:00:35.524721: step: 300/529, loss: 0.00220832834020257 2023-01-21 13:00:36.699487: step: 304/529, loss: 0.02190227434039116 2023-01-21 13:00:37.898657: step: 308/529, loss: 0.001356363296508789 2023-01-21 13:00:39.135978: step: 312/529, loss: 0.01667766645550728 2023-01-21 13:00:40.334534: step: 316/529, loss: 0.003408050397410989 2023-01-21 13:00:41.534434: step: 320/529, loss: 0.010883904062211514 2023-01-21 13:00:42.758351: step: 324/529, loss: 0.025311851873993874 2023-01-21 13:00:43.936904: step: 328/529, loss: 0.00088329310528934 2023-01-21 13:00:45.093438: step: 332/529, loss: 0.007094192318618298 2023-01-21 13:00:46.297170: step: 336/529, loss: 0.00018911361985374242 2023-01-21 13:00:47.449469: step: 340/529, loss: 0.00045862197293899953 2023-01-21 13:00:48.630931: step: 344/529, loss: 7.638931856490672e-05 2023-01-21 13:00:49.786662: step: 348/529, loss: 0.0004475593741517514 2023-01-21 13:00:50.971813: step: 352/529, loss: 0.0477084182202816 2023-01-21 13:00:52.088647: step: 356/529, loss: 0.00011663437180686742 2023-01-21 13:00:53.300905: step: 360/529, loss: 0.004853058140724897 2023-01-21 13:00:54.620130: step: 364/529, loss: 0.0001011848435155116 2023-01-21 13:00:55.825795: step: 368/529, loss: 0.0011312485439702868 2023-01-21 13:00:57.004596: step: 372/529, loss: 0.0007665634038858116 2023-01-21 13:00:58.172290: step: 376/529, loss: 0.008740234188735485 2023-01-21 13:00:59.369297: step: 380/529, loss: 0.010465908795595169 2023-01-21 13:01:00.525668: step: 384/529, loss: 0.00942773837596178 2023-01-21 13:01:01.731172: step: 388/529, loss: 0.0003105163632426411 2023-01-21 13:01:02.924339: step: 392/529, loss: 0.003458309220150113 2023-01-21 13:01:04.105133: step: 396/529, loss: 0.0013598442310467362 2023-01-21 13:01:05.296551: step: 400/529, loss: 0.011349869892001152 2023-01-21 13:01:06.496283: step: 404/529, loss: 0.00021533964900299907 2023-01-21 13:01:07.633217: step: 408/529, loss: 0.00022144318791106343 2023-01-21 13:01:08.815666: step: 412/529, loss: 9.15527380129788e-06 2023-01-21 13:01:10.026126: step: 416/529, loss: 0.005234050564467907 2023-01-21 13:01:11.242299: step: 420/529, loss: 0.03963203728199005 2023-01-21 13:01:12.433143: step: 424/529, loss: 1.735687328618951e-05 2023-01-21 13:01:13.593561: step: 428/529, loss: 0.0015305520500987768 2023-01-21 13:01:14.840694: step: 432/529, loss: 0.020304633304476738 2023-01-21 13:01:16.037554: step: 436/529, loss: 0.04639263078570366 2023-01-21 13:01:17.196370: step: 440/529, loss: 0.00025539397029206157 2023-01-21 13:01:18.389159: step: 444/529, loss: 0.056534577161073685 2023-01-21 13:01:19.569849: step: 448/529, loss: 8.440018063993193e-06 2023-01-21 13:01:20.741661: step: 452/529, loss: 0.025882070884108543 2023-01-21 13:01:21.920105: step: 456/529, loss: 0.0001871108979685232 2023-01-21 13:01:23.134769: step: 460/529, loss: 0.001416015555150807 2023-01-21 13:01:24.343850: step: 464/529, loss: 6.904602196300402e-05 2023-01-21 13:01:25.470244: step: 468/529, loss: 0.00015163421630859375 2023-01-21 13:01:26.678613: step: 472/529, loss: 0.0022642137482762337 2023-01-21 13:01:27.859329: step: 476/529, loss: 0.0030195238068699837 2023-01-21 13:01:29.019169: step: 480/529, loss: 0.08980374783277512 2023-01-21 13:01:30.213732: step: 484/529, loss: 0.0006435394170694053 2023-01-21 13:01:31.415646: step: 488/529, loss: 0.07520389556884766 2023-01-21 13:01:32.661790: step: 492/529, loss: 0.09205436706542969 2023-01-21 13:01:33.855288: step: 496/529, loss: 9.675026376498863e-05 2023-01-21 13:01:35.045890: step: 500/529, loss: 0.006104469299316406 2023-01-21 13:01:36.211173: step: 504/529, loss: 0.0002510070917196572 2023-01-21 13:01:37.428347: step: 508/529, loss: 0.00519142160192132 2023-01-21 13:01:38.626892: step: 512/529, loss: 0.0221099853515625 2023-01-21 13:01:39.833553: step: 516/529, loss: 0.02217893674969673 2023-01-21 13:01:41.044819: step: 520/529, loss: 0.00015125275240279734 2023-01-21 13:01:42.200537: step: 524/529, loss: 0.008015727624297142 2023-01-21 13:01:43.407250: step: 528/529, loss: 0.011707497760653496 2023-01-21 13:01:44.571656: step: 532/529, loss: 0.003997230436652899 2023-01-21 13:01:45.768247: step: 536/529, loss: -3.814697265625e-06 2023-01-21 13:01:46.976911: step: 540/529, loss: 0.0002864837588276714 2023-01-21 13:01:48.180630: step: 544/529, loss: 0.0034162523224949837 2023-01-21 13:01:49.389645: step: 548/529, loss: 0.0002442359982524067 2023-01-21 13:01:50.588559: step: 552/529, loss: 0.01571192778646946 2023-01-21 13:01:51.766534: step: 556/529, loss: 0.00012807846360374242 2023-01-21 13:01:52.973692: step: 560/529, loss: 0.00701751746237278 2023-01-21 13:01:54.175999: step: 564/529, loss: 0.0005651474348269403 2023-01-21 13:01:55.352639: step: 568/529, loss: 0.030129052698612213 2023-01-21 13:01:56.550552: step: 572/529, loss: 0.00363845843821764 2023-01-21 13:01:57.738794: step: 576/529, loss: 0.0014683724148198962 2023-01-21 13:01:58.890009: step: 580/529, loss: 0.0019275664817541838 2023-01-21 13:02:00.082464: step: 584/529, loss: 0.0018230437999591231 2023-01-21 13:02:01.337478: step: 588/529, loss: 0.0016445161309093237 2023-01-21 13:02:02.520242: step: 592/529, loss: 5.9604644775390625e-05 2023-01-21 13:02:03.719128: step: 596/529, loss: 0.0002191543608205393 2023-01-21 13:02:04.897816: step: 600/529, loss: 0.0007568836444988847 2023-01-21 13:02:06.116101: step: 604/529, loss: 0.00040407179039902985 2023-01-21 13:02:07.302806: step: 608/529, loss: 0.0002759933704510331 2023-01-21 13:02:08.480936: step: 612/529, loss: 0.0007089138380251825 2023-01-21 13:02:09.674514: step: 616/529, loss: 6.532669794978574e-05 2023-01-21 13:02:10.859526: step: 620/529, loss: 0.0008990287315100431 2023-01-21 13:02:12.063213: step: 624/529, loss: 0.002867126604542136 2023-01-21 13:02:13.261972: step: 628/529, loss: 0.0003772735653910786 2023-01-21 13:02:14.439598: step: 632/529, loss: 0.021743880584836006 2023-01-21 13:02:15.642147: step: 636/529, loss: 0.004989433567970991 2023-01-21 13:02:16.801794: step: 640/529, loss: 0.00010070801363326609 2023-01-21 13:02:18.034797: step: 644/529, loss: 0.026337958872318268 2023-01-21 13:02:19.255494: step: 648/529, loss: 0.0022365569602698088 2023-01-21 13:02:20.434769: step: 652/529, loss: 0.0008566856267862022 2023-01-21 13:02:21.605080: step: 656/529, loss: 0.10277990996837616 2023-01-21 13:02:22.743406: step: 660/529, loss: 0.0002665519714355469 2023-01-21 13:02:24.018824: step: 664/529, loss: 0.0032622814178466797 2023-01-21 13:02:25.206277: step: 668/529, loss: 0.00014422535605262965 2023-01-21 13:02:26.386039: step: 672/529, loss: 0.003916407003998756 2023-01-21 13:02:27.596701: step: 676/529, loss: 0.00019788742065429688 2023-01-21 13:02:28.809606: step: 680/529, loss: 0.019771194085478783 2023-01-21 13:02:29.992662: step: 684/529, loss: 0.004765319637954235 2023-01-21 13:02:31.189471: step: 688/529, loss: 0.0004483223019633442 2023-01-21 13:02:32.401242: step: 692/529, loss: 0.0023601532448083162 2023-01-21 13:02:33.617122: step: 696/529, loss: 0.00390968332067132 2023-01-21 13:02:34.794284: step: 700/529, loss: 0.04851856082677841 2023-01-21 13:02:35.940647: step: 704/529, loss: 1.6212465197895654e-05 2023-01-21 13:02:37.124014: step: 708/529, loss: 0.0015943527687340975 2023-01-21 13:02:38.357957: step: 712/529, loss: 0.0713178738951683 2023-01-21 13:02:39.562889: step: 716/529, loss: 0.0019653320778161287 2023-01-21 13:02:40.756210: step: 720/529, loss: 0.0008353710290975869 2023-01-21 13:02:41.957165: step: 724/529, loss: 0.00015573501877952367 2023-01-21 13:02:43.141367: step: 728/529, loss: 0.005900764837861061 2023-01-21 13:02:44.314897: step: 732/529, loss: 0.07257615029811859 2023-01-21 13:02:45.479085: step: 736/529, loss: 0.008382988162338734 2023-01-21 13:02:46.656536: step: 740/529, loss: 0.006828689482063055 2023-01-21 13:02:47.823315: step: 744/529, loss: 0.008426666259765625 2023-01-21 13:02:49.057650: step: 748/529, loss: 0.00099945068359375 2023-01-21 13:02:50.224940: step: 752/529, loss: 0.033737894147634506 2023-01-21 13:02:51.441282: step: 756/529, loss: 0.00043392181396484375 2023-01-21 13:02:52.671819: step: 760/529, loss: 0.028609275817871094 2023-01-21 13:02:53.924413: step: 764/529, loss: 0.0004980087396688759 2023-01-21 13:02:55.126294: step: 768/529, loss: 0.05525636672973633 2023-01-21 13:02:56.317841: step: 772/529, loss: 0.026274489238858223 2023-01-21 13:02:57.474014: step: 776/529, loss: 4.215240915073082e-05 2023-01-21 13:02:58.644809: step: 780/529, loss: 0.021368885412812233 2023-01-21 13:02:59.801092: step: 784/529, loss: 0.002225208329036832 2023-01-21 13:03:00.973169: step: 788/529, loss: 0.016387177631258965 2023-01-21 13:03:02.163540: step: 792/529, loss: 0.012323761358857155 2023-01-21 13:03:03.365162: step: 796/529, loss: 0.0009130478138104081 2023-01-21 13:03:04.528460: step: 800/529, loss: 0.00011219978478038684 2023-01-21 13:03:05.731985: step: 804/529, loss: 0.04990873485803604 2023-01-21 13:03:06.936810: step: 808/529, loss: 3.690719313453883e-05 2023-01-21 13:03:08.112767: step: 812/529, loss: 0.0005104065057821572 2023-01-21 13:03:09.280857: step: 816/529, loss: 0.0008589744684286416 2023-01-21 13:03:10.441026: step: 820/529, loss: 0.003920650575309992 2023-01-21 13:03:11.654660: step: 824/529, loss: 0.012481831945478916 2023-01-21 13:03:12.892087: step: 828/529, loss: 0.05159268528223038 2023-01-21 13:03:14.057627: step: 832/529, loss: 0.0016345500480383635 2023-01-21 13:03:15.210126: step: 836/529, loss: 0.0011128425830975175 2023-01-21 13:03:16.402023: step: 840/529, loss: 0.0003391265927348286 2023-01-21 13:03:17.616734: step: 844/529, loss: 0.0008003473049029708 2023-01-21 13:03:18.815321: step: 848/529, loss: 0.003372669219970703 2023-01-21 13:03:20.036186: step: 852/529, loss: 0.005484771449118853 2023-01-21 13:03:21.248671: step: 856/529, loss: 0.020690681412816048 2023-01-21 13:03:22.432125: step: 860/529, loss: 0.0008691787952557206 2023-01-21 13:03:23.609474: step: 864/529, loss: 0.0003770828479900956 2023-01-21 13:03:24.870069: step: 868/529, loss: 0.002551555633544922 2023-01-21 13:03:26.081057: step: 872/529, loss: 0.0023283003829419613 2023-01-21 13:03:27.211219: step: 876/529, loss: 0.0013247966999188066 2023-01-21 13:03:28.379748: step: 880/529, loss: 0.013599300757050514 2023-01-21 13:03:29.574859: step: 884/529, loss: 0.0004140853707212955 2023-01-21 13:03:30.730403: step: 888/529, loss: 0.0003833770751953125 2023-01-21 13:03:31.939908: step: 892/529, loss: 0.010223388671875 2023-01-21 13:03:33.133262: step: 896/529, loss: 0.021367836743593216 2023-01-21 13:03:34.317148: step: 900/529, loss: 0.006887626834213734 2023-01-21 13:03:35.536316: step: 904/529, loss: 0.025962544605135918 2023-01-21 13:03:36.719315: step: 908/529, loss: 0.11311989277601242 2023-01-21 13:03:37.892786: step: 912/529, loss: 9.860992577159777e-05 2023-01-21 13:03:39.080156: step: 916/529, loss: 0.0002941131533589214 2023-01-21 13:03:40.310877: step: 920/529, loss: 0.008574104867875576 2023-01-21 13:03:41.512354: step: 924/529, loss: 0.014598656445741653 2023-01-21 13:03:42.701822: step: 928/529, loss: 0.008367538452148438 2023-01-21 13:03:43.927191: step: 932/529, loss: 0.007159137632697821 2023-01-21 13:03:45.147120: step: 936/529, loss: 0.00037450791569426656 2023-01-21 13:03:46.315732: step: 940/529, loss: 0.008912897668778896 2023-01-21 13:03:47.517487: step: 944/529, loss: 0.0006434440729208291 2023-01-21 13:03:48.723041: step: 948/529, loss: 0.0003526687796693295 2023-01-21 13:03:49.907481: step: 952/529, loss: 0.00012640953355003148 2023-01-21 13:03:51.090003: step: 956/529, loss: 0.004246425349265337 2023-01-21 13:03:52.291909: step: 960/529, loss: 0.0005268097156658769 2023-01-21 13:03:53.504975: step: 964/529, loss: 0.004040718078613281 2023-01-21 13:03:54.703497: step: 968/529, loss: 0.00014801025099586695 2023-01-21 13:03:55.905509: step: 972/529, loss: 0.024769021198153496 2023-01-21 13:03:57.077403: step: 976/529, loss: 1.010894811770413e-05 2023-01-21 13:03:58.311298: step: 980/529, loss: 0.02168731763958931 2023-01-21 13:03:59.502497: step: 984/529, loss: 0.0597570426762104 2023-01-21 13:04:00.655415: step: 988/529, loss: 8.544921729480848e-05 2023-01-21 13:04:01.881584: step: 992/529, loss: 0.009358787909150124 2023-01-21 13:04:03.077416: step: 996/529, loss: 8.888245065463707e-05 2023-01-21 13:04:04.274491: step: 1000/529, loss: 0.001436996622942388 2023-01-21 13:04:05.464447: step: 1004/529, loss: 0.003125953720882535 2023-01-21 13:04:06.641008: step: 1008/529, loss: 0.027190685272216797 2023-01-21 13:04:07.828672: step: 1012/529, loss: 0.0006450653309002519 2023-01-21 13:04:09.018048: step: 1016/529, loss: 0.2021610289812088 2023-01-21 13:04:10.216961: step: 1020/529, loss: 1.659393274167087e-05 2023-01-21 13:04:11.414175: step: 1024/529, loss: 0.00047588348388671875 2023-01-21 13:04:12.591539: step: 1028/529, loss: 5.512237476068549e-05 2023-01-21 13:04:13.788958: step: 1032/529, loss: 0.060749053955078125 2023-01-21 13:04:14.985833: step: 1036/529, loss: 0.0056081777438521385 2023-01-21 13:04:16.164703: step: 1040/529, loss: 0.0012121201725676656 2023-01-21 13:04:17.345872: step: 1044/529, loss: 0.0002593994140625 2023-01-21 13:04:18.534718: step: 1048/529, loss: 0.04466743767261505 2023-01-21 13:04:19.723048: step: 1052/529, loss: 0.007094955537468195 2023-01-21 13:04:20.900117: step: 1056/529, loss: 0.016281509771943092 2023-01-21 13:04:22.111509: step: 1060/529, loss: -9.34600848268019e-06 2023-01-21 13:04:23.291980: step: 1064/529, loss: 0.00012645722017623484 2023-01-21 13:04:24.473564: step: 1068/529, loss: 0.06765823811292648 2023-01-21 13:04:25.633396: step: 1072/529, loss: -7.295608156709932e-06 2023-01-21 13:04:26.795784: step: 1076/529, loss: 0.0007905960083007812 2023-01-21 13:04:27.994211: step: 1080/529, loss: 0.05456504970788956 2023-01-21 13:04:29.222973: step: 1084/529, loss: 0.008597373962402344 2023-01-21 13:04:30.381587: step: 1088/529, loss: 0.0007982254028320312 2023-01-21 13:04:31.558305: step: 1092/529, loss: 0.0019685744773596525 2023-01-21 13:04:32.743325: step: 1096/529, loss: 0.00016858578601386398 2023-01-21 13:04:33.912417: step: 1100/529, loss: 2.6607514882925898e-05 2023-01-21 13:04:35.082631: step: 1104/529, loss: 0.1957935392856598 2023-01-21 13:04:36.239570: step: 1108/529, loss: 7.247925168485381e-06 2023-01-21 13:04:37.385825: step: 1112/529, loss: 0.0011564255692064762 2023-01-21 13:04:38.573080: step: 1116/529, loss: 0.0030940056312829256 2023-01-21 13:04:39.742718: step: 1120/529, loss: 0.010408020578324795 2023-01-21 13:04:40.882156: step: 1124/529, loss: 9.13620024221018e-05 2023-01-21 13:04:42.085255: step: 1128/529, loss: 0.0021070479415357113 2023-01-21 13:04:43.274053: step: 1132/529, loss: 0.0001659393310546875 2023-01-21 13:04:44.457087: step: 1136/529, loss: 2.8419495720299892e-05 2023-01-21 13:04:45.679169: step: 1140/529, loss: 0.03549323230981827 2023-01-21 13:04:46.845132: step: 1144/529, loss: 0.000324249267578125 2023-01-21 13:04:48.006864: step: 1148/529, loss: 0.0008911133045330644 2023-01-21 13:04:49.223699: step: 1152/529, loss: 0.04113483428955078 2023-01-21 13:04:50.402151: step: 1156/529, loss: 0.0001880168856587261 2023-01-21 13:04:51.576333: step: 1160/529, loss: 0.001344966934993863 2023-01-21 13:04:52.735353: step: 1164/529, loss: 0.10941438376903534 2023-01-21 13:04:53.904348: step: 1168/529, loss: 0.01562347449362278 2023-01-21 13:04:55.063543: step: 1172/529, loss: 0.0001163482666015625 2023-01-21 13:04:56.233730: step: 1176/529, loss: 0.000244140625 2023-01-21 13:04:57.431588: step: 1180/529, loss: 0.002197265625 2023-01-21 13:04:58.614996: step: 1184/529, loss: 0.0017303467029705644 2023-01-21 13:04:59.801093: step: 1188/529, loss: 1.735687328618951e-05 2023-01-21 13:05:01.058380: step: 1192/529, loss: 0.023471450433135033 2023-01-21 13:05:02.292636: step: 1196/529, loss: 0.0016420366009697318 2023-01-21 13:05:03.492197: step: 1200/529, loss: 2.7084352041129023e-05 2023-01-21 13:05:04.671811: step: 1204/529, loss: 0.05576743930578232 2023-01-21 13:05:05.851405: step: 1208/529, loss: 0.00030298231285996735 2023-01-21 13:05:07.046471: step: 1212/529, loss: 0.00066375732421875 2023-01-21 13:05:08.291075: step: 1216/529, loss: 0.0050605772994458675 2023-01-21 13:05:09.455989: step: 1220/529, loss: 0.00010914802987826988 2023-01-21 13:05:10.621188: step: 1224/529, loss: 0.0546233169734478 2023-01-21 13:05:11.824427: step: 1228/529, loss: 0.0021585465874522924 2023-01-21 13:05:13.003186: step: 1232/529, loss: 0.004947852808982134 2023-01-21 13:05:14.210236: step: 1236/529, loss: 0.005428123287856579 2023-01-21 13:05:15.388705: step: 1240/529, loss: 0.0010595321655273438 2023-01-21 13:05:16.551059: step: 1244/529, loss: 0.007367610931396484 2023-01-21 13:05:17.762024: step: 1248/529, loss: 0.0010976791381835938 2023-01-21 13:05:18.963777: step: 1252/529, loss: 0.007058906368911266 2023-01-21 13:05:20.161736: step: 1256/529, loss: -5.5599211918888614e-05 2023-01-21 13:05:21.344155: step: 1260/529, loss: 0.005295943934470415 2023-01-21 13:05:22.513806: step: 1264/529, loss: 0.0005245208740234375 2023-01-21 13:05:23.654241: step: 1268/529, loss: 5.53131121705519e-06 2023-01-21 13:05:24.901684: step: 1272/529, loss: 0.029079819098114967 2023-01-21 13:05:26.113244: step: 1276/529, loss: 0.08116894215345383 2023-01-21 13:05:27.357658: step: 1280/529, loss: 0.0007322311284951866 2023-01-21 13:05:28.540639: step: 1284/529, loss: 0.002185344696044922 2023-01-21 13:05:29.700968: step: 1288/529, loss: 0.13346920907497406 2023-01-21 13:05:30.884387: step: 1292/529, loss: 0.005793952848762274 2023-01-21 13:05:32.054577: step: 1296/529, loss: 0.008227539248764515 2023-01-21 13:05:33.233201: step: 1300/529, loss: 7.696152169955894e-05 2023-01-21 13:05:34.460904: step: 1304/529, loss: 0.03481044992804527 2023-01-21 13:05:35.645955: step: 1308/529, loss: 0.02016420289874077 2023-01-21 13:05:36.852343: step: 1312/529, loss: 0.0004611969052348286 2023-01-21 13:05:38.043489: step: 1316/529, loss: 0.00869522150605917 2023-01-21 13:05:39.229862: step: 1320/529, loss: 0.0008018493535928428 2023-01-21 13:05:40.418319: step: 1324/529, loss: 0.0005743026849813759 2023-01-21 13:05:41.650848: step: 1328/529, loss: 0.031172847375273705 2023-01-21 13:05:42.875459: step: 1332/529, loss: 0.001120912958867848 2023-01-21 13:05:44.090787: step: 1336/529, loss: 0.00335445418022573 2023-01-21 13:05:45.253684: step: 1340/529, loss: 0.0003461838059592992 2023-01-21 13:05:46.426791: step: 1344/529, loss: 1.430511474609375e-05 2023-01-21 13:05:47.609464: step: 1348/529, loss: 0.00023345947556663305 2023-01-21 13:05:48.808553: step: 1352/529, loss: 0.00012245179095771164 2023-01-21 13:05:49.979712: step: 1356/529, loss: 0.01572895050048828 2023-01-21 13:05:51.144432: step: 1360/529, loss: 0.0003319740353617817 2023-01-21 13:05:52.357608: step: 1364/529, loss: 7.820130122127011e-06 2023-01-21 13:05:53.523594: step: 1368/529, loss: 0.046427156776189804 2023-01-21 13:05:54.789682: step: 1372/529, loss: 0.03627300634980202 2023-01-21 13:05:55.970838: step: 1376/529, loss: 0.0005508423200808465 2023-01-21 13:05:57.180646: step: 1380/529, loss: 0.010284805670380592 2023-01-21 13:05:58.397363: step: 1384/529, loss: 0.004490375518798828 2023-01-21 13:05:59.608977: step: 1388/529, loss: 0.0024746893905103207 2023-01-21 13:06:00.827210: step: 1392/529, loss: 0.00020141602726653218 2023-01-21 13:06:02.010646: step: 1396/529, loss: 0.0029833794105798006 2023-01-21 13:06:03.208885: step: 1400/529, loss: 0.00792026612907648 2023-01-21 13:06:04.381852: step: 1404/529, loss: 4.4631960918195546e-05 2023-01-21 13:06:05.549657: step: 1408/529, loss: 0.022495746612548828 2023-01-21 13:06:06.692879: step: 1412/529, loss: 0.0024471760261803865 2023-01-21 13:06:07.871169: step: 1416/529, loss: 0.003971099853515625 2023-01-21 13:06:09.100360: step: 1420/529, loss: 0.016007328405976295 2023-01-21 13:06:10.297411: step: 1424/529, loss: 7.090568396961316e-05 2023-01-21 13:06:11.478672: step: 1428/529, loss: 0.010274888016283512 2023-01-21 13:06:12.652158: step: 1432/529, loss: 0.021803809329867363 2023-01-21 13:06:13.815641: step: 1436/529, loss: 0.00027141571626998484 2023-01-21 13:06:15.033880: step: 1440/529, loss: 0.01190719660371542 2023-01-21 13:06:16.219854: step: 1444/529, loss: 0.003740501357242465 2023-01-21 13:06:17.438255: step: 1448/529, loss: 0.00031185150146484375 2023-01-21 13:06:18.598581: step: 1452/529, loss: 0.06364526599645615 2023-01-21 13:06:19.764248: step: 1456/529, loss: 0.0003293037589173764 2023-01-21 13:06:20.962470: step: 1460/529, loss: 0.004374980926513672 2023-01-21 13:06:22.144624: step: 1464/529, loss: 3.4046173823298886e-05 2023-01-21 13:06:23.337771: step: 1468/529, loss: 2.574920654296875e-05 2023-01-21 13:06:24.505754: step: 1472/529, loss: 0.0002283096400788054 2023-01-21 13:06:25.675087: step: 1476/529, loss: 2.6988982426701114e-05 2023-01-21 13:06:26.865147: step: 1480/529, loss: 0.0028427124489098787 2023-01-21 13:06:28.031842: step: 1484/529, loss: 0.024480247870087624 2023-01-21 13:06:29.178515: step: 1488/529, loss: 0.0013875007862225175 2023-01-21 13:06:30.363142: step: 1492/529, loss: 0.00023746490478515625 2023-01-21 13:06:31.566347: step: 1496/529, loss: 0.02492804452776909 2023-01-21 13:06:32.779028: step: 1500/529, loss: 0.01005945261567831 2023-01-21 13:06:33.947693: step: 1504/529, loss: 0.02878589555621147 2023-01-21 13:06:35.108840: step: 1508/529, loss: 7.62939453125e-06 2023-01-21 13:06:36.321262: step: 1512/529, loss: 5.426407005870715e-05 2023-01-21 13:06:37.500008: step: 1516/529, loss: 0.000453948974609375 2023-01-21 13:06:38.680712: step: 1520/529, loss: 0.01130456943064928 2023-01-21 13:06:39.904346: step: 1524/529, loss: 0.006641769781708717 2023-01-21 13:06:41.074802: step: 1528/529, loss: 2.408623549854383e-05 2023-01-21 13:06:42.240180: step: 1532/529, loss: 0.0009686946868896484 2023-01-21 13:06:43.444035: step: 1536/529, loss: 0.01239156723022461 2023-01-21 13:06:44.640360: step: 1540/529, loss: 0.012543296441435814 2023-01-21 13:06:45.847573: step: 1544/529, loss: 0.0011988639598712325 2023-01-21 13:06:47.023964: step: 1548/529, loss: 0.01417740061879158 2023-01-21 13:06:48.214896: step: 1552/529, loss: 1.8119812921213452e-06 2023-01-21 13:06:49.440891: step: 1556/529, loss: 0.006872177589684725 2023-01-21 13:06:50.607964: step: 1560/529, loss: 0.02960510365664959 2023-01-21 13:06:51.797495: step: 1564/529, loss: 0.00018749237642623484 2023-01-21 13:06:52.993643: step: 1568/529, loss: 1.010894811770413e-05 2023-01-21 13:06:54.191833: step: 1572/529, loss: 0.042542073875665665 2023-01-21 13:06:55.364667: step: 1576/529, loss: 0.05537395551800728 2023-01-21 13:06:56.596185: step: 1580/529, loss: 0.000194549560546875 2023-01-21 13:06:57.791559: step: 1584/529, loss: 0.0018932819366455078 2023-01-21 13:06:58.997754: step: 1588/529, loss: 0.00416336115449667 2023-01-21 13:07:00.193535: step: 1592/529, loss: 0.025518132373690605 2023-01-21 13:07:01.364516: step: 1596/529, loss: 0.009344673715531826 2023-01-21 13:07:02.528385: step: 1600/529, loss: 0.0032118798699229956 2023-01-21 13:07:03.710221: step: 1604/529, loss: 0.00014743805513717234 2023-01-21 13:07:04.870968: step: 1608/529, loss: 6.575584120582789e-05 2023-01-21 13:07:06.062441: step: 1612/529, loss: 0.00054931640625 2023-01-21 13:07:07.246098: step: 1616/529, loss: 0.013928604312241077 2023-01-21 13:07:08.441276: step: 1620/529, loss: 0.0024173737037926912 2023-01-21 13:07:09.626903: step: 1624/529, loss: 0.009382057934999466 2023-01-21 13:07:10.785947: step: 1628/529, loss: 4.38690185546875e-05 2023-01-21 13:07:11.984345: step: 1632/529, loss: 0.008113289251923561 2023-01-21 13:07:13.185717: step: 1636/529, loss: 0.0033277510665357113 2023-01-21 13:07:14.352623: step: 1640/529, loss: 0.12257251888513565 2023-01-21 13:07:15.534046: step: 1644/529, loss: 0.006795788183808327 2023-01-21 13:07:16.728142: step: 1648/529, loss: 0.005118942353874445 2023-01-21 13:07:17.950963: step: 1652/529, loss: 0.019062327221035957 2023-01-21 13:07:19.108910: step: 1656/529, loss: 0.0044578551314771175 2023-01-21 13:07:20.248655: step: 1660/529, loss: 9.040832082973793e-05 2023-01-21 13:07:21.466169: step: 1664/529, loss: 0.023558713495731354 2023-01-21 13:07:22.669086: step: 1668/529, loss: 0.006600379943847656 2023-01-21 13:07:23.827359: step: 1672/529, loss: 0.030307531356811523 2023-01-21 13:07:25.014154: step: 1676/529, loss: 4.5776364459015895e-06 2023-01-21 13:07:26.224240: step: 1680/529, loss: 0.018062591552734375 2023-01-21 13:07:27.387912: step: 1684/529, loss: 0.0003994941944256425 2023-01-21 13:07:28.584848: step: 1688/529, loss: 0.09516115486621857 2023-01-21 13:07:29.752540: step: 1692/529, loss: 0.0005546570173464715 2023-01-21 13:07:30.945325: step: 1696/529, loss: 6.8187719080015086e-06 2023-01-21 13:07:32.145577: step: 1700/529, loss: 0.029366517439484596 2023-01-21 13:07:33.370066: step: 1704/529, loss: 0.00013456343731377274 2023-01-21 13:07:34.569841: step: 1708/529, loss: 0.0027837753295898438 2023-01-21 13:07:35.765484: step: 1712/529, loss: 0.012783337384462357 2023-01-21 13:07:36.996492: step: 1716/529, loss: -7.629394076502649e-07 2023-01-21 13:07:38.192608: step: 1720/529, loss: 0.006575966253876686 2023-01-21 13:07:39.420510: step: 1724/529, loss: 0.00048255923320539296 2023-01-21 13:07:40.667283: step: 1728/529, loss: 0.0001010894775390625 2023-01-21 13:07:41.845401: step: 1732/529, loss: 0.0012860298156738281 2023-01-21 13:07:43.043966: step: 1736/529, loss: 1.4138221558823716e-05 2023-01-21 13:07:44.279511: step: 1740/529, loss: 0.0001581192045705393 2023-01-21 13:07:45.450286: step: 1744/529, loss: 0.13606788218021393 2023-01-21 13:07:46.662535: step: 1748/529, loss: 0.07141780853271484 2023-01-21 13:07:47.844503: step: 1752/529, loss: 0.0007764697074890137 2023-01-21 13:07:49.010625: step: 1756/529, loss: 0.0003563881036825478 2023-01-21 13:07:50.181784: step: 1760/529, loss: 0.00023126602172851562 2023-01-21 13:07:51.374417: step: 1764/529, loss: 0.20197181403636932 2023-01-21 13:07:52.582950: step: 1768/529, loss: 0.0018333016196265817 2023-01-21 13:07:53.757555: step: 1772/529, loss: 0.0001388549862895161 2023-01-21 13:07:54.910328: step: 1776/529, loss: -2.8610227218450746e-06 2023-01-21 13:07:56.081278: step: 1780/529, loss: 0.0002076149103231728 2023-01-21 13:07:57.291789: step: 1784/529, loss: 0.0037496089935302734 2023-01-21 13:07:58.461457: step: 1788/529, loss: 8.00132766016759e-05 2023-01-21 13:07:59.671808: step: 1792/529, loss: 0.001179695245809853 2023-01-21 13:08:00.854553: step: 1796/529, loss: 0.007172584533691406 2023-01-21 13:08:02.043377: step: 1800/529, loss: 0.00010862351336982101 2023-01-21 13:08:03.192144: step: 1804/529, loss: 3.957748504035408e-06 2023-01-21 13:08:04.384360: step: 1808/529, loss: 0.0012466430198401213 2023-01-21 13:08:05.570247: step: 1812/529, loss: 8.840560622047633e-05 2023-01-21 13:08:06.739717: step: 1816/529, loss: 0.0003914833068847656 2023-01-21 13:08:07.883962: step: 1820/529, loss: 0.018556786701083183 2023-01-21 13:08:09.108879: step: 1824/529, loss: 0.0011154175736010075 2023-01-21 13:08:10.308072: step: 1828/529, loss: -1.7690657841740176e-05 2023-01-21 13:08:11.484212: step: 1832/529, loss: 0.04803028330206871 2023-01-21 13:08:12.699719: step: 1836/529, loss: 0.0010268688201904297 2023-01-21 13:08:13.879797: step: 1840/529, loss: 0.01755809783935547 2023-01-21 13:08:15.102644: step: 1844/529, loss: 0.0047210692428052425 2023-01-21 13:08:16.296713: step: 1848/529, loss: 0.0007061004871502519 2023-01-21 13:08:17.527235: step: 1852/529, loss: 0.00408935546875 2023-01-21 13:08:18.715179: step: 1856/529, loss: 0.006187486927956343 2023-01-21 13:08:19.946560: step: 1860/529, loss: 2.002716064453125e-05 2023-01-21 13:08:21.158168: step: 1864/529, loss: 0.01383819617331028 2023-01-21 13:08:22.347704: step: 1868/529, loss: 6.66618361719884e-05 2023-01-21 13:08:23.553344: step: 1872/529, loss: 0.0011068343883380294 2023-01-21 13:08:24.776927: step: 1876/529, loss: 0.00041456223698332906 2023-01-21 13:08:25.978775: step: 1880/529, loss: 0.003167057177051902 2023-01-21 13:08:27.156302: step: 1884/529, loss: 1.6307829355355352e-05 2023-01-21 13:08:28.326851: step: 1888/529, loss: 0.0013562202220782638 2023-01-21 13:08:29.515240: step: 1892/529, loss: 0.00125207903329283 2023-01-21 13:08:30.718634: step: 1896/529, loss: 0.018916606903076172 2023-01-21 13:08:31.940101: step: 1900/529, loss: 0.0005151748773641884 2023-01-21 13:08:33.131796: step: 1904/529, loss: 0.009308434091508389 2023-01-21 13:08:34.325018: step: 1908/529, loss: 0.00052642822265625 2023-01-21 13:08:35.496071: step: 1912/529, loss: 6.771087555534905e-06 2023-01-21 13:08:36.680094: step: 1916/529, loss: 0.011631297878921032 2023-01-21 13:08:37.922065: step: 1920/529, loss: 0.009518814273178577 2023-01-21 13:08:39.129225: step: 1924/529, loss: 0.02677593193948269 2023-01-21 13:08:40.312102: step: 1928/529, loss: 0.008015250787138939 2023-01-21 13:08:41.498844: step: 1932/529, loss: 0.004521655850112438 2023-01-21 13:08:42.647602: step: 1936/529, loss: 6.914138793945312e-05 2023-01-21 13:08:43.856044: step: 1940/529, loss: 0.0019124031532555819 2023-01-21 13:08:45.082311: step: 1944/529, loss: 0.03182869032025337 2023-01-21 13:08:46.256632: step: 1948/529, loss: 0.001189804170280695 2023-01-21 13:08:47.453327: step: 1952/529, loss: 0.00010261536226607859 2023-01-21 13:08:48.669502: step: 1956/529, loss: 0.00010595321509754285 2023-01-21 13:08:49.873807: step: 1960/529, loss: 0.025307130068540573 2023-01-21 13:08:51.057211: step: 1964/529, loss: 0.0002987861807923764 2023-01-21 13:08:52.213024: step: 1968/529, loss: 0.00019760133000090718 2023-01-21 13:08:53.438439: step: 1972/529, loss: 2.4127959477482364e-05 2023-01-21 13:08:54.670011: step: 1976/529, loss: 0.0005468368763104081 2023-01-21 13:08:55.881614: step: 1980/529, loss: 0.007959079928696156 2023-01-21 13:08:57.098171: step: 1984/529, loss: 0.021266557276248932 2023-01-21 13:08:58.307047: step: 1988/529, loss: 0.002390289446339011 2023-01-21 13:08:59.575662: step: 1992/529, loss: 0.00016326904005836695 2023-01-21 13:09:00.778814: step: 1996/529, loss: 0.0011688232189044356 2023-01-21 13:09:01.963612: step: 2000/529, loss: 1.4591216313419864e-05 2023-01-21 13:09:03.162268: step: 2004/529, loss: 0.005731773562729359 2023-01-21 13:09:04.350733: step: 2008/529, loss: 0.0011002063984051347 2023-01-21 13:09:05.578196: step: 2012/529, loss: 0.0024938583374023438 2023-01-21 13:09:06.742122: step: 2016/529, loss: 4.920959327137098e-05 2023-01-21 13:09:07.943516: step: 2020/529, loss: 0.019359970465302467 2023-01-21 13:09:09.118038: step: 2024/529, loss: 0.0009479522705078125 2023-01-21 13:09:10.294012: step: 2028/529, loss: 0.008326435461640358 2023-01-21 13:09:11.464790: step: 2032/529, loss: 2.0503997802734375e-05 2023-01-21 13:09:12.688299: step: 2036/529, loss: 0.005489062983542681 2023-01-21 13:09:13.873329: step: 2040/529, loss: 3.929138256353326e-05 2023-01-21 13:09:15.083287: step: 2044/529, loss: 0.011785602197051048 2023-01-21 13:09:16.274854: step: 2048/529, loss: 0.006059455685317516 2023-01-21 13:09:17.478772: step: 2052/529, loss: 0.0043305400758981705 2023-01-21 13:09:18.657923: step: 2056/529, loss: 8.735657320357859e-05 2023-01-21 13:09:19.847552: step: 2060/529, loss: 3.013610876223538e-05 2023-01-21 13:09:21.012871: step: 2064/529, loss: 0.06370048969984055 2023-01-21 13:09:22.221032: step: 2068/529, loss: 0.0014255524147301912 2023-01-21 13:09:23.445540: step: 2072/529, loss: 0.0006597518804483116 2023-01-21 13:09:24.640094: step: 2076/529, loss: 0.00036649705725722015 2023-01-21 13:09:25.833866: step: 2080/529, loss: 0.0121612548828125 2023-01-21 13:09:27.049397: step: 2084/529, loss: 0.6038684844970703 2023-01-21 13:09:28.194790: step: 2088/529, loss: 2.861022494471399e-07 2023-01-21 13:09:29.392988: step: 2092/529, loss: 0.0030422210693359375 2023-01-21 13:09:30.585305: step: 2096/529, loss: 0.00014095308142714202 2023-01-21 13:09:31.777512: step: 2100/529, loss: 0.0013132572639733553 2023-01-21 13:09:32.973741: step: 2104/529, loss: 0.0006562233320437372 2023-01-21 13:09:34.177661: step: 2108/529, loss: 0.01683693192899227 2023-01-21 13:09:35.348888: step: 2112/529, loss: 0.0032360078766942024 2023-01-21 13:09:36.565076: step: 2116/529, loss: 0.01158595085144043 ================================================== Loss: 0.014 -------------------- Dev: {'event': {'p': 0.5989637305699482, 'r': 0.7696404793608522, 'f1': 0.6736596736596737}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Test: {'event': {'p': 0.6320076665069477, 'r': 0.7902935889754344, 'f1': 0.7023429179978701}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Chinese: {'event': {'p': 0.5569620253164557, 'r': 0.8148148148148148, 'f1': 0.6616541353383458}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Korean: {'event': {'p': 0.515625, 'r': 0.5238095238095238, 'f1': 0.5196850393700788}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Russian: {'event': {'p': 0.4888888888888889, 'r': 0.6111111111111112, 'f1': 0.5432098765432098}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 13:10:19.619135: step: 4/529, loss: 0.08703899383544922 2023-01-21 13:10:20.753314: step: 8/529, loss: 0.022186851128935814 2023-01-21 13:10:21.920509: step: 12/529, loss: 0.001086425851099193 2023-01-21 13:10:23.096486: step: 16/529, loss: 4.0721894038142636e-05 2023-01-21 13:10:24.331859: step: 20/529, loss: 0.0002344131498830393 2023-01-21 13:10:25.507184: step: 24/529, loss: 6.608963303733617e-05 2023-01-21 13:10:26.696950: step: 28/529, loss: 0.010137557983398438 2023-01-21 13:10:27.883601: step: 32/529, loss: 0.00024929046048782766 2023-01-21 13:10:29.067953: step: 36/529, loss: -1.602172778802924e-05 2023-01-21 13:10:30.251133: step: 40/529, loss: 6.990433030296117e-05 2023-01-21 13:10:31.448194: step: 44/529, loss: 0.00010929108248092234 2023-01-21 13:10:32.672048: step: 48/529, loss: 0.007225608453154564 2023-01-21 13:10:33.874129: step: 52/529, loss: 0.034627340734004974 2023-01-21 13:10:35.082543: step: 56/529, loss: 0.00010499954805709422 2023-01-21 13:10:36.286150: step: 60/529, loss: 0.014532947912812233 2023-01-21 13:10:37.457060: step: 64/529, loss: 0.002875900361686945 2023-01-21 13:10:38.625349: step: 68/529, loss: 1.33514404296875e-05 2023-01-21 13:10:39.808588: step: 72/529, loss: 0.015816498547792435 2023-01-21 13:10:41.016597: step: 76/529, loss: 0.0013780116569250822 2023-01-21 13:10:42.186188: step: 80/529, loss: 0.003924178890883923 2023-01-21 13:10:43.365098: step: 84/529, loss: 4.9114227294921875e-05 2023-01-21 13:10:44.541682: step: 88/529, loss: 7.691383507335559e-05 2023-01-21 13:10:45.761478: step: 92/529, loss: 0.005752182099968195 2023-01-21 13:10:46.962957: step: 96/529, loss: 0.04823489487171173 2023-01-21 13:10:48.144348: step: 100/529, loss: 0.00444793701171875 2023-01-21 13:10:49.315995: step: 104/529, loss: 0.0017443657852709293 2023-01-21 13:10:50.522491: step: 108/529, loss: 3.814697265625e-06 2023-01-21 13:10:51.700984: step: 112/529, loss: 0.003720092587172985 2023-01-21 13:10:52.901638: step: 116/529, loss: 0.00010356902930652723 2023-01-21 13:10:54.154275: step: 120/529, loss: 0.00181922921910882 2023-01-21 13:10:55.362342: step: 124/529, loss: 0.0009958266746252775 2023-01-21 13:10:56.548298: step: 128/529, loss: 0.0438106544315815 2023-01-21 13:10:57.727419: step: 132/529, loss: 0.01500854454934597 2023-01-21 13:10:58.902733: step: 136/529, loss: 0.000351142865838483 2023-01-21 13:11:00.175575: step: 140/529, loss: 0.0001565933198435232 2023-01-21 13:11:01.349540: step: 144/529, loss: 0.0006480694282799959 2023-01-21 13:11:02.557435: step: 148/529, loss: 0.0001642227143747732 2023-01-21 13:11:03.764818: step: 152/529, loss: 0.09603157639503479 2023-01-21 13:11:04.974732: step: 156/529, loss: 2.1171568732825108e-05 2023-01-21 13:11:06.171898: step: 160/529, loss: 0.883753776550293 2023-01-21 13:11:07.370324: step: 164/529, loss: 0.0009885788895189762 2023-01-21 13:11:08.587180: step: 168/529, loss: 0.0007259369012899697 2023-01-21 13:11:09.806781: step: 172/529, loss: 0.014078522101044655 2023-01-21 13:11:11.028869: step: 176/529, loss: 2.803802453854587e-05 2023-01-21 13:11:12.225806: step: 180/529, loss: 0.0028085708618164062 2023-01-21 13:11:13.381644: step: 184/529, loss: 0.0005070686456747353 2023-01-21 13:11:14.594006: step: 188/529, loss: 1.754760705807712e-05 2023-01-21 13:11:15.785160: step: 192/529, loss: 0.061084747314453125 2023-01-21 13:11:16.985609: step: 196/529, loss: 0.004200935363769531 2023-01-21 13:11:18.162226: step: 200/529, loss: 0.00029773713322356343 2023-01-21 13:11:19.350955: step: 204/529, loss: 5.2547460654750466e-05 2023-01-21 13:11:20.522833: step: 208/529, loss: 0.0001352310209767893 2023-01-21 13:11:21.732568: step: 212/529, loss: 0.0026991846971213818 2023-01-21 13:11:22.916743: step: 216/529, loss: 0.023371504619717598 2023-01-21 13:11:24.077696: step: 220/529, loss: 0.004308176226913929 2023-01-21 13:11:25.268697: step: 224/529, loss: 0.0005672455299645662 2023-01-21 13:11:26.434081: step: 228/529, loss: 0.0005071639898233116 2023-01-21 13:11:27.616121: step: 232/529, loss: 0.00012607575627043843 2023-01-21 13:11:28.780907: step: 236/529, loss: 0.0010143279796466231 2023-01-21 13:11:29.969664: step: 240/529, loss: 0.06271801143884659 2023-01-21 13:11:31.209798: step: 244/529, loss: 0.00872650183737278 2023-01-21 13:11:32.419543: step: 248/529, loss: 0.005633544642478228 2023-01-21 13:11:33.651421: step: 252/529, loss: 0.018194103613495827 2023-01-21 13:11:34.807683: step: 256/529, loss: 0.01584930345416069 2023-01-21 13:11:36.002718: step: 260/529, loss: 0.020670508965849876 2023-01-21 13:11:37.189542: step: 264/529, loss: 7.047652616165578e-05 2023-01-21 13:11:38.371262: step: 268/529, loss: 0.00018482207087799907 2023-01-21 13:11:39.609635: step: 272/529, loss: 0.00048351287841796875 2023-01-21 13:11:40.761367: step: 276/529, loss: 0.002191162435337901 2023-01-21 13:11:41.944570: step: 280/529, loss: 0.006224441807717085 2023-01-21 13:11:43.190932: step: 284/529, loss: 0.009140587411820889 2023-01-21 13:11:44.372738: step: 288/529, loss: 2.689361645025201e-05 2023-01-21 13:11:45.534717: step: 292/529, loss: 8.7738037109375e-05 2023-01-21 13:11:46.706174: step: 296/529, loss: 0.03854427486658096 2023-01-21 13:11:47.898937: step: 300/529, loss: 0.0005305290105752647 2023-01-21 13:11:49.078072: step: 304/529, loss: 0.004228019621223211 2023-01-21 13:11:50.255980: step: 308/529, loss: 0.043044187128543854 2023-01-21 13:11:51.412853: step: 312/529, loss: 9.322167170466855e-05 2023-01-21 13:11:52.586493: step: 316/529, loss: 0.013856029137969017 2023-01-21 13:11:53.792020: step: 320/529, loss: 0.0010091782314702868 2023-01-21 13:11:55.030087: step: 324/529, loss: 0.00734214810654521 2023-01-21 13:11:56.196222: step: 328/529, loss: 0.0008930206531658769 2023-01-21 13:11:57.385923: step: 332/529, loss: 0.001455593155696988 2023-01-21 13:11:58.596899: step: 336/529, loss: 0.0024827958550304174 2023-01-21 13:11:59.767423: step: 340/529, loss: 0.000335884076775983 2023-01-21 13:12:00.969073: step: 344/529, loss: 0.00793552491813898 2023-01-21 13:12:02.159434: step: 348/529, loss: 0.0026943206321448088 2023-01-21 13:12:03.340280: step: 352/529, loss: 7.2479248046875e-05 2023-01-21 13:12:04.500798: step: 356/529, loss: 0.004124355502426624 2023-01-21 13:12:05.681549: step: 360/529, loss: 0.03957557678222656 2023-01-21 13:12:06.849773: step: 364/529, loss: 0.0004838943714275956 2023-01-21 13:12:08.049877: step: 368/529, loss: 0.0059379576705396175 2023-01-21 13:12:09.327906: step: 372/529, loss: 0.052053071558475494 2023-01-21 13:12:10.505488: step: 376/529, loss: 0.00016708375187590718 2023-01-21 13:12:11.680245: step: 380/529, loss: 0.01997966691851616 2023-01-21 13:12:12.873719: step: 384/529, loss: 0.00010795592970680445 2023-01-21 13:12:14.151992: step: 388/529, loss: 0.0069217681884765625 2023-01-21 13:12:15.349403: step: 392/529, loss: 0.0005355835310183465 2023-01-21 13:12:16.529532: step: 396/529, loss: 0.03647136688232422 2023-01-21 13:12:17.698537: step: 400/529, loss: 0.033852577209472656 2023-01-21 13:12:18.897254: step: 404/529, loss: 0.00032672882662154734 2023-01-21 13:12:20.058848: step: 408/529, loss: 0.003079724498093128 2023-01-21 13:12:21.252735: step: 412/529, loss: 0.00234222412109375 2023-01-21 13:12:22.452005: step: 416/529, loss: 0.04577350988984108 2023-01-21 13:12:23.686959: step: 420/529, loss: 0.010225106030702591 2023-01-21 13:12:24.904991: step: 424/529, loss: 0.00011091232590842992 2023-01-21 13:12:26.094410: step: 428/529, loss: 0.0007096290355548263 2023-01-21 13:12:27.256801: step: 432/529, loss: 0.0010703086154535413 2023-01-21 13:12:28.456084: step: 436/529, loss: 0.0004926681285724044 2023-01-21 13:12:29.636934: step: 440/529, loss: 8.869171870173886e-05 2023-01-21 13:12:30.846925: step: 444/529, loss: 0.013087654486298561 2023-01-21 13:12:32.036797: step: 448/529, loss: 3.2901763916015625e-05 2023-01-21 13:12:33.184437: step: 452/529, loss: 0.04663858935236931 2023-01-21 13:12:34.382833: step: 456/529, loss: 0.0007803440093994141 2023-01-21 13:12:35.608899: step: 460/529, loss: 0.000606155430432409 2023-01-21 13:12:36.830638: step: 464/529, loss: 0.004592132288962603 2023-01-21 13:12:38.005069: step: 468/529, loss: 0.0033407211303710938 2023-01-21 13:12:39.218473: step: 472/529, loss: 0.0011446953285485506 2023-01-21 13:12:40.438325: step: 476/529, loss: 0.001308441162109375 2023-01-21 13:12:41.639217: step: 480/529, loss: 0.00011310577974654734 2023-01-21 13:12:42.813172: step: 484/529, loss: 0.0013108253479003906 2023-01-21 13:12:44.037024: step: 488/529, loss: -4.57763690064894e-06 2023-01-21 13:12:45.196269: step: 492/529, loss: 0.0006173133733682334 2023-01-21 13:12:46.404900: step: 496/529, loss: 2.527237120375503e-05 2023-01-21 13:12:47.586505: step: 500/529, loss: 6.866455078125e-05 2023-01-21 13:12:48.820326: step: 504/529, loss: 0.0836639404296875 2023-01-21 13:12:50.014542: step: 508/529, loss: 0.0005289078108035028 2023-01-21 13:12:51.219085: step: 512/529, loss: 0.016691779717803 2023-01-21 13:12:52.435368: step: 516/529, loss: 0.04110231623053551 2023-01-21 13:12:53.643456: step: 520/529, loss: -2.0599363779183477e-05 2023-01-21 13:12:54.830948: step: 524/529, loss: 0.00023736954608466476 2023-01-21 13:12:56.020707: step: 528/529, loss: 0.0008762360084801912 2023-01-21 13:12:57.193534: step: 532/529, loss: 0.000423431396484375 2023-01-21 13:12:58.375696: step: 536/529, loss: 0.054270461201667786 2023-01-21 13:12:59.563273: step: 540/529, loss: 0.08389253914356232 2023-01-21 13:13:00.737572: step: 544/529, loss: 0.03192710876464844 2023-01-21 13:13:02.004256: step: 548/529, loss: 0.005930709652602673 2023-01-21 13:13:03.170867: step: 552/529, loss: 4.901886131847277e-05 2023-01-21 13:13:04.344059: step: 556/529, loss: 0.03146667405962944 2023-01-21 13:13:05.536882: step: 560/529, loss: 3.814697265625e-05 2023-01-21 13:13:06.713620: step: 564/529, loss: 0.043440911918878555 2023-01-21 13:13:07.925931: step: 568/529, loss: 0.028762057423591614 2023-01-21 13:13:09.127549: step: 572/529, loss: 0.0034703253768384457 2023-01-21 13:13:10.333130: step: 576/529, loss: 0.01987428590655327 2023-01-21 13:13:11.568868: step: 580/529, loss: 0.04110870510339737 2023-01-21 13:13:12.741797: step: 584/529, loss: 0.026866436004638672 2023-01-21 13:13:13.928079: step: 588/529, loss: 0.001018524169921875 2023-01-21 13:13:15.143570: step: 592/529, loss: 0.017357636243104935 2023-01-21 13:13:16.275119: step: 596/529, loss: 0.00027208327082917094 2023-01-21 13:13:17.504206: step: 600/529, loss: 0.05283012241125107 2023-01-21 13:13:18.678331: step: 604/529, loss: 0.026448631659150124 2023-01-21 13:13:19.826858: step: 608/529, loss: 0.0029283524490892887 2023-01-21 13:13:21.040159: step: 612/529, loss: 0.024448204785585403 2023-01-21 13:13:22.261090: step: 616/529, loss: 2.174377368646674e-05 2023-01-21 13:13:23.455266: step: 620/529, loss: 0.08898010104894638 2023-01-21 13:13:24.683534: step: 624/529, loss: 0.00012140273611294106 2023-01-21 13:13:25.879925: step: 628/529, loss: 0.0007497787592001259 2023-01-21 13:13:27.061757: step: 632/529, loss: 0.0002910614130087197 2023-01-21 13:13:28.260652: step: 636/529, loss: 0.0003046512429136783 2023-01-21 13:13:29.416464: step: 640/529, loss: 0.006114721763879061 2023-01-21 13:13:30.614694: step: 644/529, loss: 0.0002195358247263357 2023-01-21 13:13:31.781377: step: 648/529, loss: 7.152557373046875e-05 2023-01-21 13:13:32.941071: step: 652/529, loss: 0.49724096059799194 2023-01-21 13:13:34.135587: step: 656/529, loss: 0.0011238098377361894 2023-01-21 13:13:35.292797: step: 660/529, loss: 0.005605077836662531 2023-01-21 13:13:36.500696: step: 664/529, loss: 0.008432770147919655 2023-01-21 13:13:37.683956: step: 668/529, loss: 0.0006137847667559981 2023-01-21 13:13:38.904982: step: 672/529, loss: 0.001155090401880443 2023-01-21 13:13:40.115887: step: 676/529, loss: 0.11891335994005203 2023-01-21 13:13:41.314129: step: 680/529, loss: 0.0037322998978197575 2023-01-21 13:13:42.480816: step: 684/529, loss: 0.0049332622438669205 2023-01-21 13:13:43.692159: step: 688/529, loss: 0.036225512623786926 2023-01-21 13:13:44.886091: step: 692/529, loss: 0.0023182868026196957 2023-01-21 13:13:46.089708: step: 696/529, loss: 0.00010499954805709422 2023-01-21 13:13:47.279836: step: 700/529, loss: 0.01994462125003338 2023-01-21 13:13:48.481773: step: 704/529, loss: 0.01813526265323162 2023-01-21 13:13:49.753215: step: 708/529, loss: 0.007596397772431374 2023-01-21 13:13:50.969070: step: 712/529, loss: 0.03473644703626633 2023-01-21 13:13:52.121513: step: 716/529, loss: 0.0009654999012127519 2023-01-21 13:13:53.314174: step: 720/529, loss: 0.0001998901425395161 2023-01-21 13:13:54.461885: step: 724/529, loss: 2.3746490114717744e-05 2023-01-21 13:13:55.642208: step: 728/529, loss: -1.010894811770413e-05 2023-01-21 13:13:56.819607: step: 732/529, loss: 0.00028247834416106343 2023-01-21 13:13:58.040493: step: 736/529, loss: 0.009356784634292126 2023-01-21 13:13:59.224561: step: 740/529, loss: 0.004090213682502508 2023-01-21 13:14:00.430616: step: 744/529, loss: 0.0009544372442178428 2023-01-21 13:14:01.574758: step: 748/529, loss: 1.3351441339182202e-06 2023-01-21 13:14:02.798826: step: 752/529, loss: 0.081451416015625 2023-01-21 13:14:03.954592: step: 756/529, loss: 0.005661583039909601 2023-01-21 13:14:05.149421: step: 760/529, loss: 0.3019602596759796 2023-01-21 13:14:06.367516: step: 764/529, loss: 3.9386748539982364e-05 2023-01-21 13:14:07.520641: step: 768/529, loss: 0.00024328232393600047 2023-01-21 13:14:08.706867: step: 772/529, loss: 0.0945829451084137 2023-01-21 13:14:09.917949: step: 776/529, loss: 0.0008716583251953125 2023-01-21 13:14:11.094704: step: 780/529, loss: 0.00021438598923850805 2023-01-21 13:14:12.282743: step: 784/529, loss: 0.0028961182106286287 2023-01-21 13:14:13.501987: step: 788/529, loss: 9.50813337112777e-05 2023-01-21 13:14:14.664379: step: 792/529, loss: -5.7220458984375e-06 2023-01-21 13:14:15.857730: step: 796/529, loss: 0.00040807726327329874 2023-01-21 13:14:17.043840: step: 800/529, loss: 0.000682449375744909 2023-01-21 13:14:18.244906: step: 804/529, loss: 8.75473051564768e-05 2023-01-21 13:14:19.444407: step: 808/529, loss: 0.02306804619729519 2023-01-21 13:14:20.577014: step: 812/529, loss: 0.0014967918395996094 2023-01-21 13:14:21.749474: step: 816/529, loss: 2.6702882678364404e-05 2023-01-21 13:14:22.913548: step: 820/529, loss: 7.43865984986769e-06 2023-01-21 13:14:24.077960: step: 824/529, loss: 0.0003253698523622006 2023-01-21 13:14:25.258407: step: 828/529, loss: 0.003597450442612171 2023-01-21 13:14:26.459113: step: 832/529, loss: 0.00011729002289939672 2023-01-21 13:14:27.615638: step: 836/529, loss: 1.773834264895413e-05 2023-01-21 13:14:28.832295: step: 840/529, loss: 0.0001504898100392893 2023-01-21 13:14:30.049903: step: 844/529, loss: 0.0023234367836266756 2023-01-21 13:14:31.273431: step: 848/529, loss: 0.004158973693847656 2023-01-21 13:14:32.506737: step: 852/529, loss: 0.05565262213349342 2023-01-21 13:14:33.645036: step: 856/529, loss: 0.0014417648781090975 2023-01-21 13:14:34.851598: step: 860/529, loss: 0.010286903008818626 2023-01-21 13:14:36.041645: step: 864/529, loss: 0.004706859588623047 2023-01-21 13:14:37.228223: step: 868/529, loss: 0.00017786027456168085 2023-01-21 13:14:38.400669: step: 872/529, loss: 0.01874399185180664 2023-01-21 13:14:39.592569: step: 876/529, loss: 0.04812603071331978 2023-01-21 13:14:40.773799: step: 880/529, loss: 2.1600722902803682e-05 2023-01-21 13:14:41.962368: step: 884/529, loss: 0.003119087079539895 2023-01-21 13:14:43.214994: step: 888/529, loss: 0.005294895265251398 2023-01-21 13:14:44.391602: step: 892/529, loss: 0.004654693882912397 2023-01-21 13:14:45.556024: step: 896/529, loss: 0.00036237240419723094 2023-01-21 13:14:46.711170: step: 900/529, loss: 0.003823757404461503 2023-01-21 13:14:47.918314: step: 904/529, loss: 1.3732910701946821e-05 2023-01-21 13:14:49.099240: step: 908/529, loss: 0.09959020465612411 2023-01-21 13:14:50.274661: step: 912/529, loss: 7.43865984986769e-06 2023-01-21 13:14:51.448903: step: 916/529, loss: 0.0012193679576739669 2023-01-21 13:14:52.597165: step: 920/529, loss: 0.004172801971435547 2023-01-21 13:14:53.786519: step: 924/529, loss: 0.03820228576660156 2023-01-21 13:14:54.979331: step: 928/529, loss: 0.0003452301025390625 2023-01-21 13:14:56.125430: step: 932/529, loss: 0.0024466991890221834 2023-01-21 13:14:57.308071: step: 936/529, loss: 0.062316037714481354 2023-01-21 13:14:58.487415: step: 940/529, loss: 0.0018846511375159025 2023-01-21 13:14:59.683415: step: 944/529, loss: 0.006036949343979359 2023-01-21 13:15:00.930511: step: 948/529, loss: 0.03917961195111275 2023-01-21 13:15:02.143257: step: 952/529, loss: 0.009608173742890358 2023-01-21 13:15:03.323760: step: 956/529, loss: 0.0013042449718341231 2023-01-21 13:15:04.519170: step: 960/529, loss: 0.00272541050799191 2023-01-21 13:15:05.667487: step: 964/529, loss: 0.006847381591796875 2023-01-21 13:15:06.851336: step: 968/529, loss: 0.000492954277433455 2023-01-21 13:15:08.071785: step: 972/529, loss: 0.006552505772560835 2023-01-21 13:15:09.314965: step: 976/529, loss: 0.0002483367861714214 2023-01-21 13:15:10.499658: step: 980/529, loss: 0.006560325622558594 2023-01-21 13:15:11.744852: step: 984/529, loss: -4.76837158203125e-06 2023-01-21 13:15:12.966564: step: 988/529, loss: 0.0001720428408589214 2023-01-21 13:15:14.166790: step: 992/529, loss: 0.014256476424634457 2023-01-21 13:15:15.335867: step: 996/529, loss: 4.081726001459174e-05 2023-01-21 13:15:16.498341: step: 1000/529, loss: 0.0009751320467330515 2023-01-21 13:15:17.737981: step: 1004/529, loss: 0.00298061384819448 2023-01-21 13:15:18.975077: step: 1008/529, loss: 0.0008645057678222656 2023-01-21 13:15:20.141515: step: 1012/529, loss: 0.013348006643354893 2023-01-21 13:15:21.361024: step: 1016/529, loss: 0.022020913660526276 2023-01-21 13:15:22.546891: step: 1020/529, loss: 0.004765987396240234 2023-01-21 13:15:23.717544: step: 1024/529, loss: 0.6906064748764038 2023-01-21 13:15:24.885829: step: 1028/529, loss: 0.00296707171946764 2023-01-21 13:15:26.075226: step: 1032/529, loss: 0.026301002129912376 2023-01-21 13:15:27.245701: step: 1036/529, loss: 0.05818118900060654 2023-01-21 13:15:28.385043: step: 1040/529, loss: 1.9454957509879023e-05 2023-01-21 13:15:29.564689: step: 1044/529, loss: 0.01805725134909153 2023-01-21 13:15:30.790982: step: 1048/529, loss: 0.023810388520359993 2023-01-21 13:15:31.989294: step: 1052/529, loss: 0.009112549014389515 2023-01-21 13:15:33.181480: step: 1056/529, loss: 3.337860107421875e-05 2023-01-21 13:15:34.384413: step: 1060/529, loss: 0.0013909341068938375 2023-01-21 13:15:35.552564: step: 1064/529, loss: 0.0026483535766601562 2023-01-21 13:15:36.732512: step: 1068/529, loss: 0.0010196685325354338 2023-01-21 13:15:37.937552: step: 1072/529, loss: 0.0009142875787802041 2023-01-21 13:15:39.160105: step: 1076/529, loss: 0.26565074920654297 2023-01-21 13:15:40.438726: step: 1080/529, loss: 0.01162881962954998 2023-01-21 13:15:41.610244: step: 1084/529, loss: 0.0004436492745298892 2023-01-21 13:15:42.833520: step: 1088/529, loss: 0.0006002426380291581 2023-01-21 13:15:44.008418: step: 1092/529, loss: 0.21182385087013245 2023-01-21 13:15:45.221338: step: 1096/529, loss: 0.024146651849150658 2023-01-21 13:15:46.414894: step: 1100/529, loss: 0.0009248733404092491 2023-01-21 13:15:47.581139: step: 1104/529, loss: 0.015993213281035423 2023-01-21 13:15:48.788923: step: 1108/529, loss: 0.0316954143345356 2023-01-21 13:15:49.959106: step: 1112/529, loss: 0.008948231115937233 2023-01-21 13:15:51.153068: step: 1116/529, loss: 0.0018062591552734375 2023-01-21 13:15:52.341809: step: 1120/529, loss: 0.012559986673295498 2023-01-21 13:15:53.511985: step: 1124/529, loss: 0.041483186185359955 2023-01-21 13:15:54.677916: step: 1128/529, loss: 0.19248619675636292 2023-01-21 13:15:55.863188: step: 1132/529, loss: 0.0011848449939861894 2023-01-21 13:15:57.068530: step: 1136/529, loss: 0.0025873184204101562 2023-01-21 13:15:58.278893: step: 1140/529, loss: 0.0001985549897653982 2023-01-21 13:15:59.492455: step: 1144/529, loss: 3.376007225597277e-05 2023-01-21 13:16:00.710024: step: 1148/529, loss: 0.001006698701530695 2023-01-21 13:16:01.933136: step: 1152/529, loss: 0.0023355484008789062 2023-01-21 13:16:03.095552: step: 1156/529, loss: 0.00019531250291038305 2023-01-21 13:16:04.244764: step: 1160/529, loss: 0.0009933472611010075 2023-01-21 13:16:05.441945: step: 1164/529, loss: 0.000141143798828125 2023-01-21 13:16:06.622784: step: 1168/529, loss: 0.007078933529555798 2023-01-21 13:16:07.832329: step: 1172/529, loss: 0.015691757202148438 2023-01-21 13:16:09.023005: step: 1176/529, loss: 0.004620718769729137 2023-01-21 13:16:10.206322: step: 1180/529, loss: 0.00032892229501158 2023-01-21 13:16:11.380089: step: 1184/529, loss: 0.008436775766313076 2023-01-21 13:16:12.574684: step: 1188/529, loss: 0.1825597882270813 2023-01-21 13:16:13.787253: step: 1192/529, loss: 0.006682968232780695 2023-01-21 13:16:14.992818: step: 1196/529, loss: 0.0018310546875 2023-01-21 13:16:16.168071: step: 1200/529, loss: 0.0028045654762536287 2023-01-21 13:16:17.352217: step: 1204/529, loss: 0.04111327975988388 2023-01-21 13:16:18.531845: step: 1208/529, loss: 0.007266044616699219 2023-01-21 13:16:19.670851: step: 1212/529, loss: 0.00042110084905289114 2023-01-21 13:16:20.841508: step: 1216/529, loss: 0.023906517773866653 2023-01-21 13:16:22.037546: step: 1220/529, loss: 0.026366423815488815 2023-01-21 13:16:23.249965: step: 1224/529, loss: 0.06181488186120987 2023-01-21 13:16:24.500927: step: 1228/529, loss: 0.004440403077751398 2023-01-21 13:16:25.712138: step: 1232/529, loss: 0.25809335708618164 2023-01-21 13:16:26.970981: step: 1236/529, loss: 0.00036640168400481343 2023-01-21 13:16:28.192937: step: 1240/529, loss: 0.02675599977374077 2023-01-21 13:16:29.389446: step: 1244/529, loss: 0.044527627527713776 2023-01-21 13:16:30.570436: step: 1248/529, loss: 0.00044384002103470266 2023-01-21 13:16:31.774365: step: 1252/529, loss: 0.0016265868907794356 2023-01-21 13:16:32.973060: step: 1256/529, loss: 0.0017253875266760588 2023-01-21 13:16:34.137176: step: 1260/529, loss: 0.0009447097545489669 2023-01-21 13:16:35.304139: step: 1264/529, loss: 0.0037884714547544718 2023-01-21 13:16:36.492532: step: 1268/529, loss: 0.0015060901641845703 2023-01-21 13:16:37.721316: step: 1272/529, loss: 0.0515931136906147 2023-01-21 13:16:38.888377: step: 1276/529, loss: 7.62939453125e-06 2023-01-21 13:16:40.068877: step: 1280/529, loss: 0.00015697479830123484 2023-01-21 13:16:41.271179: step: 1284/529, loss: 0.0009894848335534334 2023-01-21 13:16:42.489299: step: 1288/529, loss: 0.0008552551735192537 2023-01-21 13:16:43.717952: step: 1292/529, loss: 0.001204204629175365 2023-01-21 13:16:44.921142: step: 1296/529, loss: 0.009193611331284046 2023-01-21 13:16:46.132106: step: 1300/529, loss: 0.0029575349763035774 2023-01-21 13:16:47.309637: step: 1304/529, loss: 2.307891918462701e-05 2023-01-21 13:16:48.489628: step: 1308/529, loss: 0.0016100883949548006 2023-01-21 13:16:49.681748: step: 1312/529, loss: 0.0003833770751953125 2023-01-21 13:16:50.904430: step: 1316/529, loss: 0.00750656146556139 2023-01-21 13:16:52.089153: step: 1320/529, loss: 0.11062035709619522 2023-01-21 13:16:53.302450: step: 1324/529, loss: 0.0009062409517355263 2023-01-21 13:16:54.514960: step: 1328/529, loss: 9.603500802768394e-05 2023-01-21 13:16:55.695696: step: 1332/529, loss: 0.01048889197409153 2023-01-21 13:16:56.898623: step: 1336/529, loss: 0.0009457588312216103 2023-01-21 13:16:58.116005: step: 1340/529, loss: 0.013356066308915615 2023-01-21 13:16:59.308554: step: 1344/529, loss: 0.006854057312011719 2023-01-21 13:17:00.514938: step: 1348/529, loss: 6.313324411166832e-05 2023-01-21 13:17:01.716704: step: 1352/529, loss: 0.0018568038940429688 2023-01-21 13:17:02.939383: step: 1356/529, loss: 0.0003631591680459678 2023-01-21 13:17:04.125564: step: 1360/529, loss: 0.004102802369743586 2023-01-21 13:17:05.277426: step: 1364/529, loss: 0.00030736924964003265 2023-01-21 13:17:06.461965: step: 1368/529, loss: 0.03762407600879669 2023-01-21 13:17:07.657235: step: 1372/529, loss: 0.009470652788877487 2023-01-21 13:17:08.839570: step: 1376/529, loss: 0.0012943744659423828 2023-01-21 13:17:10.054848: step: 1380/529, loss: 0.017116928473114967 2023-01-21 13:17:11.234896: step: 1384/529, loss: 0.0026426315307617188 2023-01-21 13:17:12.463807: step: 1388/529, loss: 0.009540367871522903 2023-01-21 13:17:13.639492: step: 1392/529, loss: 0.0001813888520700857 2023-01-21 13:17:14.819224: step: 1396/529, loss: 0.010789012536406517 2023-01-21 13:17:16.051911: step: 1400/529, loss: 0.04092216491699219 2023-01-21 13:17:17.262260: step: 1404/529, loss: 0.003507232526317239 2023-01-21 13:17:18.448374: step: 1408/529, loss: 0.03182784095406532 2023-01-21 13:17:19.708636: step: 1412/529, loss: 0.028965570032596588 2023-01-21 13:17:20.909213: step: 1416/529, loss: 0.013645696453750134 2023-01-21 13:17:22.068644: step: 1420/529, loss: 6.752014451194555e-05 2023-01-21 13:17:23.267285: step: 1424/529, loss: 0.04342842102050781 2023-01-21 13:17:24.475795: step: 1428/529, loss: 0.0003849029599223286 2023-01-21 13:17:25.663124: step: 1432/529, loss: 0.041029881685972214 2023-01-21 13:17:26.878650: step: 1436/529, loss: 0.0007427216041833162 2023-01-21 13:17:28.049695: step: 1440/529, loss: 0.00022239684767555445 2023-01-21 13:17:29.221812: step: 1444/529, loss: 5.9413909184513614e-05 2023-01-21 13:17:30.399920: step: 1448/529, loss: 0.0018434524536132812 2023-01-21 13:17:31.569426: step: 1452/529, loss: 0.0016441346378996968 2023-01-21 13:17:32.745003: step: 1456/529, loss: 0.020009947940707207 2023-01-21 13:17:33.964477: step: 1460/529, loss: 2.4223329091910273e-05 2023-01-21 13:17:35.162264: step: 1464/529, loss: 4.2724612285383046e-05 2023-01-21 13:17:36.343650: step: 1468/529, loss: 0.00018644332885742188 2023-01-21 13:17:37.518731: step: 1472/529, loss: 5.760193016612902e-05 2023-01-21 13:17:38.706216: step: 1476/529, loss: 0.1125604659318924 2023-01-21 13:17:39.898557: step: 1480/529, loss: 0.005111026577651501 2023-01-21 13:17:41.149190: step: 1484/529, loss: 0.0013748168712481856 2023-01-21 13:17:42.330416: step: 1488/529, loss: 0.00010166168067371473 2023-01-21 13:17:43.547449: step: 1492/529, loss: 0.00165386195294559 2023-01-21 13:17:44.703652: step: 1496/529, loss: 0.06788130104541779 2023-01-21 13:17:45.886822: step: 1500/529, loss: 0.00046043400652706623 2023-01-21 13:17:47.063470: step: 1504/529, loss: 5.364418029785156e-05 2023-01-21 13:17:48.225955: step: 1508/529, loss: 0.002368831541389227 2023-01-21 13:17:49.429839: step: 1512/529, loss: 0.0002368926943745464 2023-01-21 13:17:50.644927: step: 1516/529, loss: 1.2254715329618193e-05 2023-01-21 13:17:51.851194: step: 1520/529, loss: 0.03748512640595436 2023-01-21 13:17:53.020240: step: 1524/529, loss: 5.15937790623866e-05 2023-01-21 13:17:54.217489: step: 1528/529, loss: 0.03810310363769531 2023-01-21 13:17:55.444261: step: 1532/529, loss: 0.0005926132434979081 2023-01-21 13:17:56.625360: step: 1536/529, loss: 0.0004899025079794228 2023-01-21 13:17:57.822156: step: 1540/529, loss: 0.03801784664392471 2023-01-21 13:17:59.033162: step: 1544/529, loss: 7.82012939453125e-05 2023-01-21 13:18:00.247932: step: 1548/529, loss: 0.005653571803122759 2023-01-21 13:18:01.454208: step: 1552/529, loss: 4.7159195673884824e-05 2023-01-21 13:18:02.698561: step: 1556/529, loss: 0.0006330489995889366 2023-01-21 13:18:03.902687: step: 1560/529, loss: 0.05439510568976402 2023-01-21 13:18:05.096509: step: 1564/529, loss: 3.62396240234375e-05 2023-01-21 13:18:06.276862: step: 1568/529, loss: 0.0010322571033611894 2023-01-21 13:18:07.479248: step: 1572/529, loss: 0.0022098540794104338 2023-01-21 13:18:08.658584: step: 1576/529, loss: 0.0029603957664221525 2023-01-21 13:18:09.844216: step: 1580/529, loss: 0.008899307809770107 2023-01-21 13:18:11.052427: step: 1584/529, loss: 0.004836654290556908 2023-01-21 13:18:12.200076: step: 1588/529, loss: 0.0004618644597940147 2023-01-21 13:18:13.413853: step: 1592/529, loss: 0.0250917449593544 2023-01-21 13:18:14.577793: step: 1596/529, loss: 0.03251810371875763 2023-01-21 13:18:15.765573: step: 1600/529, loss: 0.0005973815568722785 2023-01-21 13:18:16.998341: step: 1604/529, loss: 0.015264606103301048 2023-01-21 13:18:18.202407: step: 1608/529, loss: 0.006385994143784046 2023-01-21 13:18:19.401873: step: 1612/529, loss: 0.04300248995423317 2023-01-21 13:18:20.581511: step: 1616/529, loss: 0.00016651154146529734 2023-01-21 13:18:21.755591: step: 1620/529, loss: 0.00012578963651321828 2023-01-21 13:18:22.920058: step: 1624/529, loss: 0.0005056381342001259 2023-01-21 13:18:24.122985: step: 1628/529, loss: 0.09431781619787216 2023-01-21 13:18:25.293890: step: 1632/529, loss: 0.005275917239487171 2023-01-21 13:18:26.472176: step: 1636/529, loss: 0.0025323869194835424 2023-01-21 13:18:27.679407: step: 1640/529, loss: 0.011876821517944336 2023-01-21 13:18:28.878830: step: 1644/529, loss: 0.00017843247042037547 2023-01-21 13:18:30.104401: step: 1648/529, loss: 0.0916026160120964 2023-01-21 13:18:31.307052: step: 1652/529, loss: 0.01324691902846098 2023-01-21 13:18:32.515607: step: 1656/529, loss: 0.0011009216541424394 2023-01-21 13:18:33.687508: step: 1660/529, loss: 0.0004036426544189453 2023-01-21 13:18:34.879466: step: 1664/529, loss: 0.0001712799130473286 2023-01-21 13:18:36.064858: step: 1668/529, loss: 0.002393055008724332 2023-01-21 13:18:37.237053: step: 1672/529, loss: 0.023511027917265892 2023-01-21 13:18:38.422194: step: 1676/529, loss: 0.0009181976784020662 2023-01-21 13:18:39.578131: step: 1680/529, loss: 1.4400482541532256e-05 2023-01-21 13:18:40.779503: step: 1684/529, loss: 0.005416488740593195 2023-01-21 13:18:41.963748: step: 1688/529, loss: 0.00471992464736104 2023-01-21 13:18:43.163273: step: 1692/529, loss: 2.4032595320022665e-05 2023-01-21 13:18:44.356357: step: 1696/529, loss: 0.11666784435510635 2023-01-21 13:18:45.547346: step: 1700/529, loss: 0.002605962799862027 2023-01-21 13:18:46.731214: step: 1704/529, loss: 0.004446220584213734 2023-01-21 13:18:47.916559: step: 1708/529, loss: 0.006130027584731579 2023-01-21 13:18:49.093280: step: 1712/529, loss: 0.0010610581375658512 2023-01-21 13:18:50.287540: step: 1716/529, loss: 0.0014236450660973787 2023-01-21 13:18:51.467314: step: 1720/529, loss: 0.0008883953560143709 2023-01-21 13:18:52.659665: step: 1724/529, loss: 0.0034356117248535156 2023-01-21 13:18:53.833681: step: 1728/529, loss: 0.0008444786071777344 2023-01-21 13:18:55.093047: step: 1732/529, loss: 0.005627060309052467 2023-01-21 13:18:56.289987: step: 1736/529, loss: 8.02993745310232e-05 2023-01-21 13:18:57.467333: step: 1740/529, loss: 8.39233416627394e-06 2023-01-21 13:18:58.660146: step: 1744/529, loss: 0.001068210694938898 2023-01-21 13:18:59.844538: step: 1748/529, loss: 0.013585949316620827 2023-01-21 13:19:01.048655: step: 1752/529, loss: 0.0006834983360022306 2023-01-21 13:19:02.214830: step: 1756/529, loss: 0.011636734008789062 2023-01-21 13:19:03.429481: step: 1760/529, loss: 2.5081633793888614e-05 2023-01-21 13:19:04.596932: step: 1764/529, loss: 0.0002938270627055317 2023-01-21 13:19:05.776969: step: 1768/529, loss: 0.00014762878709007055 2023-01-21 13:19:06.933461: step: 1772/529, loss: 0.003143119625747204 2023-01-21 13:19:08.078176: step: 1776/529, loss: 0.0004108429129701108 2023-01-21 13:19:09.258454: step: 1780/529, loss: 0.009210300631821156 2023-01-21 13:19:10.442845: step: 1784/529, loss: 0.005858707707375288 2023-01-21 13:19:11.612585: step: 1788/529, loss: 0.005080032162368298 2023-01-21 13:19:12.834347: step: 1792/529, loss: 0.007211446762084961 2023-01-21 13:19:14.021110: step: 1796/529, loss: 0.0016631126636639237 2023-01-21 13:19:15.194896: step: 1800/529, loss: 0.00011739730689441785 2023-01-21 13:19:16.396290: step: 1804/529, loss: 0.0029680251609534025 2023-01-21 13:19:17.598940: step: 1808/529, loss: 0.04195919260382652 2023-01-21 13:19:18.838354: step: 1812/529, loss: 0.0007523060194216669 2023-01-21 13:19:20.013332: step: 1816/529, loss: 0.00557632464915514 2023-01-21 13:19:21.210330: step: 1820/529, loss: 0.0006479262956418097 2023-01-21 13:19:22.413102: step: 1824/529, loss: 0.0002773284795694053 2023-01-21 13:19:23.568266: step: 1828/529, loss: 0.2148026078939438 2023-01-21 13:19:24.726506: step: 1832/529, loss: 0.00019674301438499242 2023-01-21 13:19:25.923104: step: 1836/529, loss: 0.00010042190842796117 2023-01-21 13:19:27.110710: step: 1840/529, loss: 0.0002117156982421875 2023-01-21 13:19:28.289472: step: 1844/529, loss: 0.0005591392400674522 2023-01-21 13:19:29.449396: step: 1848/529, loss: 0.00020494460477493703 2023-01-21 13:19:30.589706: step: 1852/529, loss: 0.019325638189911842 2023-01-21 13:19:31.788680: step: 1856/529, loss: 0.004228496924042702 2023-01-21 13:19:32.975958: step: 1860/529, loss: 0.00027484894962981343 2023-01-21 13:19:34.156441: step: 1864/529, loss: 0.0004433155118022114 2023-01-21 13:19:35.336844: step: 1868/529, loss: 0.01247315388172865 2023-01-21 13:19:36.512216: step: 1872/529, loss: 0.0008173942333087325 2023-01-21 13:19:37.679494: step: 1876/529, loss: 0.020439432933926582 2023-01-21 13:19:38.856435: step: 1880/529, loss: 0.00034008026705123484 2023-01-21 13:19:40.042797: step: 1884/529, loss: 0.00028285980806685984 2023-01-21 13:19:41.268707: step: 1888/529, loss: 0.0009370804182253778 2023-01-21 13:19:42.478698: step: 1892/529, loss: 0.03917722776532173 2023-01-21 13:19:43.680282: step: 1896/529, loss: 0.0004302978341002017 2023-01-21 13:19:44.908765: step: 1900/529, loss: 0.00016093255544546992 2023-01-21 13:19:46.077794: step: 1904/529, loss: 0.0004818916495423764 2023-01-21 13:19:47.289769: step: 1908/529, loss: 4.615783836925402e-05 2023-01-21 13:19:48.495560: step: 1912/529, loss: 0.004528045654296875 2023-01-21 13:19:49.721302: step: 1916/529, loss: 0.0008831024169921875 2023-01-21 13:19:50.877865: step: 1920/529, loss: 0.001818323158659041 2023-01-21 13:19:52.109581: step: 1924/529, loss: 7.02858014847152e-05 2023-01-21 13:19:53.331479: step: 1928/529, loss: 0.00012321471876930445 2023-01-21 13:19:54.553300: step: 1932/529, loss: 0.003964233677834272 2023-01-21 13:19:55.718869: step: 1936/529, loss: 0.0038354399148374796 2023-01-21 13:19:56.883291: step: 1940/529, loss: 0.0005627631908282638 2023-01-21 13:19:58.071098: step: 1944/529, loss: 0.00045032502384856343 2023-01-21 13:19:59.265549: step: 1948/529, loss: 2.155303991457913e-05 2023-01-21 13:20:00.472527: step: 1952/529, loss: 0.00028438569279387593 2023-01-21 13:20:01.761431: step: 1956/529, loss: 0.005705928895622492 2023-01-21 13:20:02.964616: step: 1960/529, loss: 0.0015460015274584293 2023-01-21 13:20:04.145972: step: 1964/529, loss: 0.01661682315170765 2023-01-21 13:20:05.412327: step: 1968/529, loss: 0.08536114543676376 2023-01-21 13:20:06.579790: step: 1972/529, loss: 0.0005023956764489412 2023-01-21 13:20:07.755308: step: 1976/529, loss: 0.0014404297107830644 2023-01-21 13:20:08.903426: step: 1980/529, loss: 0.003703594207763672 2023-01-21 13:20:10.096610: step: 1984/529, loss: 0.004171181004494429 2023-01-21 13:20:11.301286: step: 1988/529, loss: 0.005911064334213734 2023-01-21 13:20:12.498248: step: 1992/529, loss: 0.01422643568366766 2023-01-21 13:20:13.669273: step: 1996/529, loss: 0.03371906280517578 2023-01-21 13:20:14.862869: step: 2000/529, loss: 0.00043234825716353953 2023-01-21 13:20:16.064494: step: 2004/529, loss: 1.602172778802924e-05 2023-01-21 13:20:17.267137: step: 2008/529, loss: 1.0490421118447557e-06 2023-01-21 13:20:18.438428: step: 2012/529, loss: 8.335113670909777e-05 2023-01-21 13:20:19.619716: step: 2016/529, loss: 0.00013484954251907766 2023-01-21 13:20:20.819537: step: 2020/529, loss: 0.007437038701027632 2023-01-21 13:20:21.977962: step: 2024/529, loss: 0.0023869038559496403 2023-01-21 13:20:23.181316: step: 2028/529, loss: 0.04378543049097061 2023-01-21 13:20:24.363339: step: 2032/529, loss: 0.0057586668990552425 2023-01-21 13:20:25.582398: step: 2036/529, loss: 0.010715675540268421 2023-01-21 13:20:26.772059: step: 2040/529, loss: 0.0005029678577557206 2023-01-21 13:20:27.951731: step: 2044/529, loss: 0.004349899478256702 2023-01-21 13:20:29.137732: step: 2048/529, loss: 4.625320343620842e-06 2023-01-21 13:20:30.319645: step: 2052/529, loss: 0.0016422271728515625 2023-01-21 13:20:31.509557: step: 2056/529, loss: 3.24249267578125e-05 2023-01-21 13:20:32.708052: step: 2060/529, loss: 0.003667163662612438 2023-01-21 13:20:33.931746: step: 2064/529, loss: 5.664825584972277e-05 2023-01-21 13:20:35.149897: step: 2068/529, loss: 1.8215179807157256e-05 2023-01-21 13:20:36.338008: step: 2072/529, loss: 0.00015563966007903218 2023-01-21 13:20:37.513122: step: 2076/529, loss: 0.0001848220854299143 2023-01-21 13:20:38.737811: step: 2080/529, loss: 0.03695087507367134 2023-01-21 13:20:39.947670: step: 2084/529, loss: 3.490447852527723e-05 2023-01-21 13:20:41.174716: step: 2088/529, loss: 0.007520962040871382 2023-01-21 13:20:42.340243: step: 2092/529, loss: 0.00028758050757460296 2023-01-21 13:20:43.536908: step: 2096/529, loss: 0.00033397675724700093 2023-01-21 13:20:44.711724: step: 2100/529, loss: 3.390312485862523e-05 2023-01-21 13:20:45.859798: step: 2104/529, loss: 0.00029048920259810984 2023-01-21 13:20:47.115819: step: 2108/529, loss: 0.006049156654626131 2023-01-21 13:20:48.294312: step: 2112/529, loss: 0.019324827939271927 2023-01-21 13:20:49.511584: step: 2116/529, loss: 0.023595044389367104 ================================================== Loss: 0.017 -------------------- Dev: {'event': {'p': 0.5989847715736041, 'r': 0.7856191744340879, 'f1': 0.6797235023041475}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Test: {'event': {'p': 0.6251175917215428, 'r': 0.7962852007189934, 'f1': 0.700395256916996}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Chinese: {'event': {'p': 0.5679012345679012, 'r': 0.8518518518518519, 'f1': 0.6814814814814814}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Korean: {'event': {'p': 0.5362318840579711, 'r': 0.5873015873015873, 'f1': 0.5606060606060606}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Russian: {'event': {'p': 0.4523809523809524, 'r': 0.5277777777777778, 'f1': 0.4871794871794871}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** Epoch: 25 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:21:32.665241: step: 4/529, loss: 3.519058373058215e-05 2023-01-21 13:21:33.860697: step: 8/529, loss: 0.589938759803772 2023-01-21 13:21:35.114173: step: 12/529, loss: 0.0015676498878747225 2023-01-21 13:21:36.333794: step: 16/529, loss: 0.006231784820556641 2023-01-21 13:21:37.517282: step: 20/529, loss: 0.0007458686595782638 2023-01-21 13:21:38.713239: step: 24/529, loss: 0.010531043633818626 2023-01-21 13:21:39.908872: step: 28/529, loss: 0.0051742554642260075 2023-01-21 13:21:41.096520: step: 32/529, loss: 0.002394771669059992 2023-01-21 13:21:42.286768: step: 36/529, loss: 1.163482647825731e-05 2023-01-21 13:21:43.487255: step: 40/529, loss: 4.043579247081652e-05 2023-01-21 13:21:44.651936: step: 44/529, loss: 0.014004579745233059 2023-01-21 13:21:45.872280: step: 48/529, loss: 0.0005135536193847656 2023-01-21 13:21:47.069927: step: 52/529, loss: 0.0007102966192178428 2023-01-21 13:21:48.279558: step: 56/529, loss: 0.03771953657269478 2023-01-21 13:21:49.462880: step: 60/529, loss: 0.0005418777582235634 2023-01-21 13:21:50.623656: step: 64/529, loss: 0.0013505936367437243 2023-01-21 13:21:51.806203: step: 68/529, loss: 0.023321151733398438 2023-01-21 13:21:52.998546: step: 72/529, loss: 0.048165515065193176 2023-01-21 13:21:54.151456: step: 76/529, loss: 0.00012664795212913305 2023-01-21 13:21:55.382896: step: 80/529, loss: 0.2819991111755371 2023-01-21 13:21:56.581975: step: 84/529, loss: 0.037427328526973724 2023-01-21 13:21:57.751067: step: 88/529, loss: 0.007565689273178577 2023-01-21 13:21:58.957105: step: 92/529, loss: 0.00262451171875 2023-01-21 13:22:00.134666: step: 96/529, loss: 0.004447984509170055 2023-01-21 13:22:01.323845: step: 100/529, loss: 0.0008533477666787803 2023-01-21 13:22:02.519956: step: 104/529, loss: 0.0024695396423339844 2023-01-21 13:22:03.714814: step: 108/529, loss: 1.3971329281048384e-05 2023-01-21 13:22:04.885264: step: 112/529, loss: 0.0001579284726176411 2023-01-21 13:22:06.046336: step: 116/529, loss: 0.009340382181107998 2023-01-21 13:22:07.242171: step: 120/529, loss: 0.22913393378257751 2023-01-21 13:22:08.422520: step: 124/529, loss: 0.0002875328063964844 2023-01-21 13:22:09.633056: step: 128/529, loss: 0.0009663582313805819 2023-01-21 13:22:10.772092: step: 132/529, loss: 0.019967293366789818 2023-01-21 13:22:11.981277: step: 136/529, loss: 0.007351875305175781 2023-01-21 13:22:13.207206: step: 140/529, loss: 7.305145845748484e-05 2023-01-21 13:22:14.404214: step: 144/529, loss: 0.0020355223678052425 2023-01-21 13:22:15.600495: step: 148/529, loss: 0.0007492065196856856 2023-01-21 13:22:16.777899: step: 152/529, loss: 0.005029201507568359 2023-01-21 13:22:17.951576: step: 156/529, loss: 0.0002533912775106728 2023-01-21 13:22:19.133503: step: 160/529, loss: 2.307891918462701e-05 2023-01-21 13:22:20.327596: step: 164/529, loss: 0.030162431299686432 2023-01-21 13:22:21.533101: step: 168/529, loss: 0.008948802947998047 2023-01-21 13:22:22.704194: step: 172/529, loss: 1.8882752556237392e-05 2023-01-21 13:22:23.916482: step: 176/529, loss: 0.007852554321289062 2023-01-21 13:22:25.127215: step: 180/529, loss: 0.0001201629638671875 2023-01-21 13:22:26.327771: step: 184/529, loss: 0.00443191546946764 2023-01-21 13:22:27.530795: step: 188/529, loss: 0.35907572507858276 2023-01-21 13:22:28.715253: step: 192/529, loss: 5.245208740234375e-05 2023-01-21 13:22:29.878297: step: 196/529, loss: 0.013924050144851208 2023-01-21 13:22:31.072560: step: 200/529, loss: 0.007835770025849342 2023-01-21 13:22:32.349909: step: 204/529, loss: 0.024692537263035774 2023-01-21 13:22:33.549508: step: 208/529, loss: 0.0005146980402059853 2023-01-21 13:22:34.739864: step: 212/529, loss: 0.0010484695667400956 2023-01-21 13:22:35.919327: step: 216/529, loss: 0.0032876969780772924 2023-01-21 13:22:37.079154: step: 220/529, loss: 0.00010967254638671875 2023-01-21 13:22:38.316713: step: 224/529, loss: 0.012119054794311523 2023-01-21 13:22:39.562208: step: 228/529, loss: 0.002966261003166437 2023-01-21 13:22:40.757000: step: 232/529, loss: 0.0023641586303710938 2023-01-21 13:22:41.935488: step: 236/529, loss: 0.00012807846360374242 2023-01-21 13:22:43.147317: step: 240/529, loss: 0.00020973682694602758 2023-01-21 13:22:44.333516: step: 244/529, loss: 0.0053067682310938835 2023-01-21 13:22:45.520573: step: 248/529, loss: 0.0009965896606445312 2023-01-21 13:22:46.768730: step: 252/529, loss: 0.0003490447998046875 2023-01-21 13:22:47.912472: step: 256/529, loss: 1.258850079466356e-05 2023-01-21 13:22:49.123776: step: 260/529, loss: 0.03919544070959091 2023-01-21 13:22:50.348693: step: 264/529, loss: 0.002490520477294922 2023-01-21 13:22:51.495976: step: 268/529, loss: 0.006438255775719881 2023-01-21 13:22:52.663762: step: 272/529, loss: 0.00010890961129916832 2023-01-21 13:22:53.850596: step: 276/529, loss: 0.008707045577466488 2023-01-21 13:22:55.019210: step: 280/529, loss: 4.9591067181609105e-06 2023-01-21 13:22:56.194495: step: 284/529, loss: 0.0011163712479174137 2023-01-21 13:22:57.391025: step: 288/529, loss: 0.06883106380701065 2023-01-21 13:22:58.595338: step: 292/529, loss: 0.0002906799491029233 2023-01-21 13:22:59.828136: step: 296/529, loss: 0.03744640573859215 2023-01-21 13:23:01.033541: step: 300/529, loss: 0.0022716522216796875 2023-01-21 13:23:02.256787: step: 304/529, loss: 0.00017333030700683594 2023-01-21 13:23:03.429958: step: 308/529, loss: 0.0006372451898641884 2023-01-21 13:23:04.649121: step: 312/529, loss: 5.1784514653263614e-05 2023-01-21 13:23:05.842172: step: 316/529, loss: 5.91278057981981e-06 2023-01-21 13:23:07.026248: step: 320/529, loss: 0.00038051605224609375 2023-01-21 13:23:08.200454: step: 324/529, loss: 0.00400776881724596 2023-01-21 13:23:09.400089: step: 328/529, loss: 0.019878387451171875 2023-01-21 13:23:10.594082: step: 332/529, loss: 0.0005832671886309981 2023-01-21 13:23:11.749519: step: 336/529, loss: 0.00011296272714389488 2023-01-21 13:23:12.977697: step: 340/529, loss: 0.012678910046815872 2023-01-21 13:23:14.172337: step: 344/529, loss: 1.5354156857938506e-05 2023-01-21 13:23:15.318910: step: 348/529, loss: 0.02221374586224556 2023-01-21 13:23:16.502149: step: 352/529, loss: 0.0004319667932577431 2023-01-21 13:23:17.713605: step: 356/529, loss: 0.0012866019969806075 2023-01-21 13:23:18.927704: step: 360/529, loss: 0.0014366151299327612 2023-01-21 13:23:20.096274: step: 364/529, loss: 0.009512138552963734 2023-01-21 13:23:21.360501: step: 368/529, loss: 0.0007678985712118447 2023-01-21 13:23:22.576509: step: 372/529, loss: 5.7601926528150216e-05 2023-01-21 13:23:23.776811: step: 376/529, loss: 0.0007490158313885331 2023-01-21 13:23:24.993648: step: 380/529, loss: 0.33468008041381836 2023-01-21 13:23:26.204598: step: 384/529, loss: 0.003107261611148715 2023-01-21 13:23:27.446314: step: 388/529, loss: 0.0002445220889057964 2023-01-21 13:23:28.654077: step: 392/529, loss: 2.136230432370212e-05 2023-01-21 13:23:29.848641: step: 396/529, loss: 0.0033927918411791325 2023-01-21 13:23:31.047361: step: 400/529, loss: 0.0386258102953434 2023-01-21 13:23:32.216807: step: 404/529, loss: -4.1365624383615796e-06 2023-01-21 13:23:33.361483: step: 408/529, loss: 0.005389547441154718 2023-01-21 13:23:34.576254: step: 412/529, loss: 0.02464141882956028 2023-01-21 13:23:35.785935: step: 416/529, loss: 0.04879140853881836 2023-01-21 13:23:36.920416: step: 420/529, loss: 0.00010566711716819555 2023-01-21 13:23:38.128077: step: 424/529, loss: 0.029463957995176315 2023-01-21 13:23:39.335543: step: 428/529, loss: 0.013134574517607689 2023-01-21 13:23:40.514652: step: 432/529, loss: 0.0006608963012695312 2023-01-21 13:23:41.715803: step: 436/529, loss: 0.05553503334522247 2023-01-21 13:23:42.886145: step: 440/529, loss: 0.03685379400849342 2023-01-21 13:23:44.083233: step: 444/529, loss: 0.00010395050776423886 2023-01-21 13:23:45.271469: step: 448/529, loss: 9.5367431640625e-07 2023-01-21 13:23:46.468750: step: 452/529, loss: 0.00016485452943015844 2023-01-21 13:23:47.645050: step: 456/529, loss: 0.004799461457878351 2023-01-21 13:23:48.833519: step: 460/529, loss: 7.152557373046875e-06 2023-01-21 13:23:50.020470: step: 464/529, loss: 0.005683708004653454 2023-01-21 13:23:51.227965: step: 468/529, loss: 0.01776904985308647 2023-01-21 13:23:52.425342: step: 472/529, loss: 2.384185791015625e-06 2023-01-21 13:23:53.603044: step: 476/529, loss: 0.00020236970158293843 2023-01-21 13:23:54.826902: step: 480/529, loss: 0.006550169084221125 2023-01-21 13:23:56.050386: step: 484/529, loss: 0.0015988351078704 2023-01-21 13:23:57.278951: step: 488/529, loss: 0.00024127960205078125 2023-01-21 13:23:58.448758: step: 492/529, loss: 2.803802453854587e-05 2023-01-21 13:23:59.676015: step: 496/529, loss: 0.00527877826243639 2023-01-21 13:24:00.867070: step: 500/529, loss: 0.004704189486801624 2023-01-21 13:24:02.113202: step: 504/529, loss: 0.0002384185791015625 2023-01-21 13:24:03.313386: step: 508/529, loss: 0.0070091248489916325 2023-01-21 13:24:04.521230: step: 512/529, loss: 0.03541393578052521 2023-01-21 13:24:05.706143: step: 516/529, loss: 0.0001183509812108241 2023-01-21 13:24:06.869475: step: 520/529, loss: 0.0006942272302694619 2023-01-21 13:24:08.050255: step: 524/529, loss: 0.01378555316478014 2023-01-21 13:24:09.184664: step: 528/529, loss: 0.0132325179874897 2023-01-21 13:24:10.356541: step: 532/529, loss: 0.00013618469529319555 2023-01-21 13:24:11.553539: step: 536/529, loss: 0.1562759429216385 2023-01-21 13:24:12.709502: step: 540/529, loss: 9.08970832824707e-05 2023-01-21 13:24:13.855417: step: 544/529, loss: 0.0010103225940838456 2023-01-21 13:24:15.045308: step: 548/529, loss: 0.0025781632866710424 2023-01-21 13:24:16.272520: step: 552/529, loss: 0.007628345396369696 2023-01-21 13:24:17.467410: step: 556/529, loss: 0.00021572114201262593 2023-01-21 13:24:18.686891: step: 560/529, loss: 4.262923903297633e-05 2023-01-21 13:24:19.870581: step: 564/529, loss: 9.727478754939511e-06 2023-01-21 13:24:21.031882: step: 568/529, loss: 0.0023058413062244654 2023-01-21 13:24:22.280576: step: 572/529, loss: 0.00018901826115325093 2023-01-21 13:24:23.459379: step: 576/529, loss: 2.2697449821862392e-05 2023-01-21 13:24:24.637859: step: 580/529, loss: 4.76837158203125e-05 2023-01-21 13:24:25.846218: step: 584/529, loss: 0.013970708474516869 2023-01-21 13:24:27.030981: step: 588/529, loss: 0.0003679275687318295 2023-01-21 13:24:28.196984: step: 592/529, loss: 1.850128137448337e-05 2023-01-21 13:24:29.384511: step: 596/529, loss: 0.006728744599968195 2023-01-21 13:24:30.606336: step: 600/529, loss: 0.02705860137939453 2023-01-21 13:24:31.785992: step: 604/529, loss: 0.0022388456854969263 2023-01-21 13:24:32.982515: step: 608/529, loss: 6.027221752447076e-05 2023-01-21 13:24:34.136791: step: 612/529, loss: 0.03446245193481445 2023-01-21 13:24:35.338292: step: 616/529, loss: 0.00015945434279274195 2023-01-21 13:24:36.547184: step: 620/529, loss: 0.0026391984429210424 2023-01-21 13:24:37.730271: step: 624/529, loss: 0.0277772918343544 2023-01-21 13:24:38.958650: step: 628/529, loss: 0.011947202496230602 2023-01-21 13:24:40.142352: step: 632/529, loss: 3.409385681152344e-05 2023-01-21 13:24:41.366271: step: 636/529, loss: 0.007177162449806929 2023-01-21 13:24:42.529666: step: 640/529, loss: 0.022189904004335403 2023-01-21 13:24:43.714780: step: 644/529, loss: 0.0005662917974404991 2023-01-21 13:24:44.907157: step: 648/529, loss: 0.001060104463249445 2023-01-21 13:24:46.080517: step: 652/529, loss: 7.658005051780492e-05 2023-01-21 13:24:47.291920: step: 656/529, loss: 0.01444101333618164 2023-01-21 13:24:48.523035: step: 660/529, loss: 5.474090721691027e-05 2023-01-21 13:24:49.704078: step: 664/529, loss: 9.441375004826114e-05 2023-01-21 13:24:50.881822: step: 668/529, loss: 0.017198849469423294 2023-01-21 13:24:52.027832: step: 672/529, loss: 6.48498553346144e-06 2023-01-21 13:24:53.230843: step: 676/529, loss: 0.00018825530423782766 2023-01-21 13:24:54.442431: step: 680/529, loss: 0.0014240265591070056 2023-01-21 13:24:55.650552: step: 684/529, loss: 0.0015491485828533769 2023-01-21 13:24:56.859478: step: 688/529, loss: 1.0967254638671875e-05 2023-01-21 13:24:58.024746: step: 692/529, loss: 1.8262862795381807e-05 2023-01-21 13:24:59.230832: step: 696/529, loss: 0.0014374733436852694 2023-01-21 13:25:00.417087: step: 700/529, loss: 0.00024976732674986124 2023-01-21 13:25:01.625379: step: 704/529, loss: 0.0003394126833882183 2023-01-21 13:25:02.823711: step: 708/529, loss: 0.0013156890636309981 2023-01-21 13:25:03.992696: step: 712/529, loss: 0.008582497015595436 2023-01-21 13:25:05.147417: step: 716/529, loss: 0.007393646519631147 2023-01-21 13:25:06.345451: step: 720/529, loss: 0.0001539230433991179 2023-01-21 13:25:07.532779: step: 724/529, loss: 0.0030104636680334806 2023-01-21 13:25:08.726995: step: 728/529, loss: 0.041518211364746094 2023-01-21 13:25:09.916450: step: 732/529, loss: 0.033913709223270416 2023-01-21 13:25:11.118552: step: 736/529, loss: 0.0007184982532635331 2023-01-21 13:25:12.347712: step: 740/529, loss: 0.028669072315096855 2023-01-21 13:25:13.539789: step: 744/529, loss: 0.00010452270362293348 2023-01-21 13:25:14.720619: step: 748/529, loss: 0.03687705844640732 2023-01-21 13:25:15.881105: step: 752/529, loss: 0.00035572052001953125 2023-01-21 13:25:17.056685: step: 756/529, loss: 0.0001070022553903982 2023-01-21 13:25:18.215211: step: 760/529, loss: 0.004831123165786266 2023-01-21 13:25:19.394391: step: 764/529, loss: 0.0005943298456259072 2023-01-21 13:25:20.584992: step: 768/529, loss: 0.0012346267467364669 2023-01-21 13:25:21.744061: step: 772/529, loss: 5.0163271225756034e-05 2023-01-21 13:25:22.989134: step: 776/529, loss: 4.1484832763671875e-05 2023-01-21 13:25:24.149116: step: 780/529, loss: 0.002336263656616211 2023-01-21 13:25:25.308965: step: 784/529, loss: 3.4809111184586072e-06 2023-01-21 13:25:26.475967: step: 788/529, loss: 0.012293053790926933 2023-01-21 13:25:27.669525: step: 792/529, loss: 0.004297065548598766 2023-01-21 13:25:28.907859: step: 796/529, loss: 0.006991100497543812 2023-01-21 13:25:30.089074: step: 800/529, loss: 0.0004047393740620464 2023-01-21 13:25:31.311283: step: 804/529, loss: 0.00013511181168723851 2023-01-21 13:25:32.507049: step: 808/529, loss: 0.00046148302499204874 2023-01-21 13:25:33.682423: step: 812/529, loss: 7.2479248046875e-05 2023-01-21 13:25:34.830922: step: 816/529, loss: 0.0024557113647460938 2023-01-21 13:25:35.988937: step: 820/529, loss: 7.62939453125e-05 2023-01-21 13:25:37.193898: step: 824/529, loss: 0.009744835086166859 2023-01-21 13:25:38.390328: step: 828/529, loss: 0.10424967110157013 2023-01-21 13:25:39.599325: step: 832/529, loss: 0.010329628363251686 2023-01-21 13:25:40.824126: step: 836/529, loss: 0.009173011407256126 2023-01-21 13:25:42.006812: step: 840/529, loss: 0.03605499491095543 2023-01-21 13:25:43.215582: step: 844/529, loss: 0.0002420425444142893 2023-01-21 13:25:44.394074: step: 848/529, loss: 0.00012092590623069555 2023-01-21 13:25:45.544347: step: 852/529, loss: 0.006921720691025257 2023-01-21 13:25:46.712244: step: 856/529, loss: 0.01713581010699272 2023-01-21 13:25:47.895028: step: 860/529, loss: 0.0007255554082803428 2023-01-21 13:25:49.096115: step: 864/529, loss: 0.025281716138124466 2023-01-21 13:25:50.311299: step: 868/529, loss: 0.0003278255753684789 2023-01-21 13:25:51.473353: step: 872/529, loss: 2.0313262211857364e-05 2023-01-21 13:25:52.673398: step: 876/529, loss: 0.0006422996520996094 2023-01-21 13:25:53.866492: step: 880/529, loss: 0.001736545586027205 2023-01-21 13:25:55.074304: step: 884/529, loss: 0.15635128319263458 2023-01-21 13:25:56.305350: step: 888/529, loss: 0.00011463165719760582 2023-01-21 13:25:57.468712: step: 892/529, loss: 0.00067052838858217 2023-01-21 13:25:58.700624: step: 896/529, loss: 0.004331111907958984 2023-01-21 13:25:59.910300: step: 900/529, loss: 0.0013283728621900082 2023-01-21 13:26:01.126060: step: 904/529, loss: 0.0056386953219771385 2023-01-21 13:26:02.397909: step: 908/529, loss: 1.2874603271484375e-05 2023-01-21 13:26:03.576677: step: 912/529, loss: 0.00407066335901618 2023-01-21 13:26:04.761784: step: 916/529, loss: 0.003124713897705078 2023-01-21 13:26:05.951282: step: 920/529, loss: 0.00035057071363553405 2023-01-21 13:26:07.121996: step: 924/529, loss: 0.03272724151611328 2023-01-21 13:26:08.358234: step: 928/529, loss: 0.0002471923944540322 2023-01-21 13:26:09.582536: step: 932/529, loss: 0.0017559052212163806 2023-01-21 13:26:10.798867: step: 936/529, loss: 0.00048809050349518657 2023-01-21 13:26:11.997747: step: 940/529, loss: 0.2978772222995758 2023-01-21 13:26:13.202116: step: 944/529, loss: 0.00698776263743639 2023-01-21 13:26:14.389505: step: 948/529, loss: 6.12258882028982e-05 2023-01-21 13:26:15.569009: step: 952/529, loss: 0.00015125275240279734 2023-01-21 13:26:16.741687: step: 956/529, loss: 2.250671423098538e-05 2023-01-21 13:26:17.930412: step: 960/529, loss: 0.0002061843842966482 2023-01-21 13:26:19.126178: step: 964/529, loss: 0.010334777645766735 2023-01-21 13:26:20.330944: step: 968/529, loss: 5.2261355449445546e-05 2023-01-21 13:26:21.553341: step: 972/529, loss: 4.615783836925402e-05 2023-01-21 13:26:22.707616: step: 976/529, loss: 0.006630039773881435 2023-01-21 13:26:23.927894: step: 980/529, loss: 0.003926277160644531 2023-01-21 13:26:25.103734: step: 984/529, loss: 0.00029964448185637593 2023-01-21 13:26:26.296291: step: 988/529, loss: 8.94546537892893e-05 2023-01-21 13:26:27.460550: step: 992/529, loss: 0.003434467129409313 2023-01-21 13:26:28.626476: step: 996/529, loss: 8.144378807628527e-05 2023-01-21 13:26:29.818810: step: 1000/529, loss: 0.07996673882007599 2023-01-21 13:26:31.017409: step: 1004/529, loss: 9.51766996877268e-05 2023-01-21 13:26:32.213393: step: 1008/529, loss: 0.0006629943381994963 2023-01-21 13:26:33.383881: step: 1012/529, loss: 0.021312618628144264 2023-01-21 13:26:34.578083: step: 1016/529, loss: 0.0017731189727783203 2023-01-21 13:26:35.779501: step: 1020/529, loss: 0.004471111577004194 2023-01-21 13:26:36.987628: step: 1024/529, loss: 1.296997106692288e-05 2023-01-21 13:26:38.133235: step: 1028/529, loss: 0.000309848808683455 2023-01-21 13:26:39.300039: step: 1032/529, loss: 5.0830840336857364e-05 2023-01-21 13:26:40.444994: step: 1036/529, loss: 0.0001619339018361643 2023-01-21 13:26:41.646708: step: 1040/529, loss: 0.00020523072453215718 2023-01-21 13:26:42.827880: step: 1044/529, loss: 5.722049536416307e-07 2023-01-21 13:26:43.989465: step: 1048/529, loss: 2.355575270485133e-05 2023-01-21 13:26:45.169560: step: 1052/529, loss: -9.536779543850571e-08 2023-01-21 13:26:46.403637: step: 1056/529, loss: 5.574875831604004 2023-01-21 13:26:47.657272: step: 1060/529, loss: 3.4332276754867053e-06 2023-01-21 13:26:48.878578: step: 1064/529, loss: 0.0006635665777139366 2023-01-21 13:26:50.108476: step: 1068/529, loss: 0.025925828143954277 2023-01-21 13:26:51.280933: step: 1072/529, loss: 0.00012702943058684468 2023-01-21 13:26:52.430669: step: 1076/529, loss: 3.3092499506892636e-05 2023-01-21 13:26:53.589773: step: 1080/529, loss: 0.020606135949492455 2023-01-21 13:26:54.796610: step: 1084/529, loss: 0.0004594803031068295 2023-01-21 13:26:55.986203: step: 1088/529, loss: 0.015452956780791283 2023-01-21 13:26:57.139761: step: 1092/529, loss: 0.002453613094985485 2023-01-21 13:26:58.315574: step: 1096/529, loss: 0.00010023117647506297 2023-01-21 13:26:59.506060: step: 1100/529, loss: 0.0038464071694761515 2023-01-21 13:27:00.699390: step: 1104/529, loss: 0.03686580806970596 2023-01-21 13:27:01.887652: step: 1108/529, loss: 0.007505225948989391 2023-01-21 13:27:03.076025: step: 1112/529, loss: 0.0014426232082769275 2023-01-21 13:27:04.267556: step: 1116/529, loss: 3.61442580469884e-05 2023-01-21 13:27:05.459394: step: 1120/529, loss: 0.0026311874389648438 2023-01-21 13:27:06.633268: step: 1124/529, loss: 0.011591816321015358 2023-01-21 13:27:07.837075: step: 1128/529, loss: 3.6716461181640625e-05 2023-01-21 13:27:09.106372: step: 1132/529, loss: 5.149841854290571e-06 2023-01-21 13:27:10.263985: step: 1136/529, loss: 0.00011005401756847277 2023-01-21 13:27:11.460433: step: 1140/529, loss: 0.00087823870126158 2023-01-21 13:27:12.657918: step: 1144/529, loss: 0.0003364562871865928 2023-01-21 13:27:13.848097: step: 1148/529, loss: 0.002280616667121649 2023-01-21 13:27:15.067569: step: 1152/529, loss: 0.012219048105180264 2023-01-21 13:27:16.272883: step: 1156/529, loss: 0.04020838811993599 2023-01-21 13:27:17.465242: step: 1160/529, loss: 2.136230432370212e-05 2023-01-21 13:27:18.663743: step: 1164/529, loss: 0.009549903683364391 2023-01-21 13:27:19.870648: step: 1168/529, loss: 0.010606002993881702 2023-01-21 13:27:21.057614: step: 1172/529, loss: 0.00090789794921875 2023-01-21 13:27:22.195212: step: 1176/529, loss: 0.011428642086684704 2023-01-21 13:27:23.391068: step: 1180/529, loss: 0.016846848651766777 2023-01-21 13:27:24.622326: step: 1184/529, loss: 0.0007736206171102822 2023-01-21 13:27:25.795833: step: 1188/529, loss: 2.8133392333984375e-05 2023-01-21 13:27:27.015047: step: 1192/529, loss: 0.0010198593372479081 2023-01-21 13:27:28.196921: step: 1196/529, loss: 0.0026692866813391447 2023-01-21 13:27:29.374543: step: 1200/529, loss: 0.00013818741717841476 2023-01-21 13:27:30.578589: step: 1204/529, loss: 8.20159948489163e-06 2023-01-21 13:27:31.734956: step: 1208/529, loss: 0.003239917801693082 2023-01-21 13:27:32.957150: step: 1212/529, loss: 0.0004779815790243447 2023-01-21 13:27:34.135448: step: 1216/529, loss: 0.0011984348529949784 2023-01-21 13:27:35.331897: step: 1220/529, loss: 9.670257713878527e-05 2023-01-21 13:27:36.535003: step: 1224/529, loss: 0.003485966008156538 2023-01-21 13:27:37.767913: step: 1228/529, loss: 0.0010880469344556332 2023-01-21 13:27:38.988275: step: 1232/529, loss: 0.00021276473125908524 2023-01-21 13:27:40.202900: step: 1236/529, loss: 4.76837158203125e-06 2023-01-21 13:27:41.363420: step: 1240/529, loss: -1.3351439065445447e-06 2023-01-21 13:27:42.571925: step: 1244/529, loss: 0.0008756637689657509 2023-01-21 13:27:43.756978: step: 1248/529, loss: 0.02888660691678524 2023-01-21 13:27:44.947493: step: 1252/529, loss: 0.08788514137268066 2023-01-21 13:27:46.220245: step: 1256/529, loss: 0.00033130645169876516 2023-01-21 13:27:47.381435: step: 1260/529, loss: 0.0002189636288676411 2023-01-21 13:27:48.581496: step: 1264/529, loss: 0.0036752698943018913 2023-01-21 13:27:49.779707: step: 1268/529, loss: -2.384185791015625e-07 2023-01-21 13:27:50.994745: step: 1272/529, loss: 0.024285221472382545 2023-01-21 13:27:52.158228: step: 1276/529, loss: 3.814697322468419e-07 2023-01-21 13:27:53.341746: step: 1280/529, loss: 0.00014381408982444555 2023-01-21 13:27:54.497796: step: 1284/529, loss: -2.47955313170678e-06 2023-01-21 13:27:55.717201: step: 1288/529, loss: 0.03478727117180824 2023-01-21 13:27:56.838586: step: 1292/529, loss: 0.0002849579032044858 2023-01-21 13:27:58.036823: step: 1296/529, loss: 0.007300090976059437 2023-01-21 13:27:59.208354: step: 1300/529, loss: 0.003224706742912531 2023-01-21 13:28:00.393366: step: 1304/529, loss: 0.0001447677641408518 2023-01-21 13:28:01.578378: step: 1308/529, loss: 0.000463891017716378 2023-01-21 13:28:02.809937: step: 1312/529, loss: 0.00286026019603014 2023-01-21 13:28:03.979506: step: 1316/529, loss: 0.029608584940433502 2023-01-21 13:28:05.237016: step: 1320/529, loss: 0.008067512884736061 2023-01-21 13:28:06.394775: step: 1324/529, loss: 2.6798248654813506e-05 2023-01-21 13:28:07.603533: step: 1328/529, loss: 3.1936167943058535e-05 2023-01-21 13:28:08.787399: step: 1332/529, loss: 0.030396176502108574 2023-01-21 13:28:09.963522: step: 1336/529, loss: 0.01925182342529297 2023-01-21 13:28:11.132706: step: 1340/529, loss: 0.2518463134765625 2023-01-21 13:28:12.339311: step: 1344/529, loss: 3.8146970382513246e-06 2023-01-21 13:28:13.542656: step: 1348/529, loss: 0.0006702423561364412 2023-01-21 13:28:14.739601: step: 1352/529, loss: 0.00074939732439816 2023-01-21 13:28:15.947415: step: 1356/529, loss: 8.792877633823082e-05 2023-01-21 13:28:17.141223: step: 1360/529, loss: 0.00014200211444403976 2023-01-21 13:28:18.315160: step: 1364/529, loss: 0.00022964477830100805 2023-01-21 13:28:19.455803: step: 1368/529, loss: 0.013141060248017311 2023-01-21 13:28:20.667345: step: 1372/529, loss: 0.1378769427537918 2023-01-21 13:28:21.818332: step: 1376/529, loss: 0.0006799697293899953 2023-01-21 13:28:22.963774: step: 1380/529, loss: 0.00019021033949684352 2023-01-21 13:28:24.175037: step: 1384/529, loss: 0.028212929144501686 2023-01-21 13:28:25.354286: step: 1388/529, loss: 2.7561189199332148e-05 2023-01-21 13:28:26.564840: step: 1392/529, loss: 0.03977527469396591 2023-01-21 13:28:27.740235: step: 1396/529, loss: 5.455017162603326e-05 2023-01-21 13:28:28.920976: step: 1400/529, loss: 6.151199340820312e-05 2023-01-21 13:28:30.097324: step: 1404/529, loss: 0.006925821304321289 2023-01-21 13:28:31.278718: step: 1408/529, loss: 0.04187622293829918 2023-01-21 13:28:32.475519: step: 1412/529, loss: 0.00257282261736691 2023-01-21 13:28:33.673819: step: 1416/529, loss: 0.026589583605527878 2023-01-21 13:28:34.864637: step: 1420/529, loss: 0.007771492004394531 2023-01-21 13:28:36.079947: step: 1424/529, loss: 0.0024022101424634457 2023-01-21 13:28:37.302195: step: 1428/529, loss: 0.0005808830028399825 2023-01-21 13:28:38.506236: step: 1432/529, loss: 0.017254162579774857 2023-01-21 13:28:39.735870: step: 1436/529, loss: 0.38376161456108093 2023-01-21 13:28:40.883043: step: 1440/529, loss: 0.0022794725373387337 2023-01-21 13:28:42.063782: step: 1444/529, loss: 8.20159948489163e-06 2023-01-21 13:28:43.273532: step: 1448/529, loss: 0.0014661789173260331 2023-01-21 13:28:44.485835: step: 1452/529, loss: 4.978180004400201e-05 2023-01-21 13:28:45.658680: step: 1456/529, loss: 0.0010312080848962069 2023-01-21 13:28:46.832276: step: 1460/529, loss: 0.0070625306107103825 2023-01-21 13:28:48.009778: step: 1464/529, loss: 4.501342846197076e-05 2023-01-21 13:28:49.191125: step: 1468/529, loss: 0.004653740208595991 2023-01-21 13:28:50.319112: step: 1472/529, loss: 2.994537317135837e-05 2023-01-21 13:28:51.495532: step: 1476/529, loss: 0.00011477470252430066 2023-01-21 13:28:52.738526: step: 1480/529, loss: 0.0014583587180823088 2023-01-21 13:28:53.949620: step: 1484/529, loss: 0.04657306522130966 2023-01-21 13:28:55.138112: step: 1488/529, loss: 1.1539458682818804e-05 2023-01-21 13:28:56.313895: step: 1492/529, loss: 1.33514404296875e-05 2023-01-21 13:28:57.521208: step: 1496/529, loss: 0.0009698867797851562 2023-01-21 13:28:58.796181: step: 1500/529, loss: 0.0005991935613565147 2023-01-21 13:28:59.982811: step: 1504/529, loss: 0.0002716064336709678 2023-01-21 13:29:01.171782: step: 1508/529, loss: 1.049041748046875e-05 2023-01-21 13:29:02.369745: step: 1512/529, loss: 0.0002399444638285786 2023-01-21 13:29:03.566126: step: 1516/529, loss: 0.00013961792865302414 2023-01-21 13:29:04.723231: step: 1520/529, loss: 4.673004150390625e-05 2023-01-21 13:29:05.953607: step: 1524/529, loss: 0.0004291534423828125 2023-01-21 13:29:07.170806: step: 1528/529, loss: 0.011511802673339844 2023-01-21 13:29:08.336648: step: 1532/529, loss: 0.024472618475556374 2023-01-21 13:29:09.505193: step: 1536/529, loss: 0.00011825562251033261 2023-01-21 13:29:10.688049: step: 1540/529, loss: 0.00026950836763717234 2023-01-21 13:29:11.873870: step: 1544/529, loss: 0.0027468684129416943 2023-01-21 13:29:13.024989: step: 1548/529, loss: 9.269714064430445e-05 2023-01-21 13:29:14.208541: step: 1552/529, loss: 0.023320579901337624 2023-01-21 13:29:15.395130: step: 1556/529, loss: 0.007452297490090132 2023-01-21 13:29:16.625291: step: 1560/529, loss: 0.0007141113164834678 2023-01-21 13:29:17.786045: step: 1564/529, loss: 0.040764614939689636 2023-01-21 13:29:18.953314: step: 1568/529, loss: 0.026871085166931152 2023-01-21 13:29:20.092890: step: 1572/529, loss: 2.6702882678364404e-06 2023-01-21 13:29:21.280425: step: 1576/529, loss: 0.002458763075992465 2023-01-21 13:29:22.466104: step: 1580/529, loss: 0.17653360962867737 2023-01-21 13:29:23.641593: step: 1584/529, loss: 0.0001762867032084614 2023-01-21 13:29:24.844516: step: 1588/529, loss: 0.0001638412504689768 2023-01-21 13:29:26.036063: step: 1592/529, loss: 7.362365431617945e-05 2023-01-21 13:29:27.236075: step: 1596/529, loss: 0.002918815705925226 2023-01-21 13:29:28.439118: step: 1600/529, loss: 0.0003949165402445942 2023-01-21 13:29:29.565506: step: 1604/529, loss: 0.007817196659743786 2023-01-21 13:29:30.786813: step: 1608/529, loss: 0.19501753151416779 2023-01-21 13:29:31.971377: step: 1612/529, loss: 0.00028858185396529734 2023-01-21 13:29:33.155734: step: 1616/529, loss: 0.0014340400230139494 2023-01-21 13:29:34.357313: step: 1620/529, loss: -5.722044988942798e-07 2023-01-21 13:29:35.511096: step: 1624/529, loss: 0.007927799597382545 2023-01-21 13:29:36.682696: step: 1628/529, loss: 0.0007692337385378778 2023-01-21 13:29:37.869132: step: 1632/529, loss: 5.435943421616685e-06 2023-01-21 13:29:39.085489: step: 1636/529, loss: 0.7499170303344727 2023-01-21 13:29:40.300768: step: 1640/529, loss: -1.2302399227337446e-05 2023-01-21 13:29:41.483732: step: 1644/529, loss: 0.007376289460808039 2023-01-21 13:29:42.655221: step: 1648/529, loss: 0.000584459281526506 2023-01-21 13:29:43.880753: step: 1652/529, loss: 0.0014457226498052478 2023-01-21 13:29:45.053987: step: 1656/529, loss: 0.0026224851608276367 2023-01-21 13:29:46.289283: step: 1660/529, loss: 0.04477253183722496 2023-01-21 13:29:47.476800: step: 1664/529, loss: 0.004599952604621649 2023-01-21 13:29:48.688933: step: 1668/529, loss: 0.005660629365593195 2023-01-21 13:29:49.847392: step: 1672/529, loss: 0.0015316962962970138 2023-01-21 13:29:51.059300: step: 1676/529, loss: 0.0003719329833984375 2023-01-21 13:29:52.237000: step: 1680/529, loss: 0.004139423836022615 2023-01-21 13:29:53.401642: step: 1684/529, loss: 0.0035297395661473274 2023-01-21 13:29:54.581714: step: 1688/529, loss: 0.00048160552978515625 2023-01-21 13:29:55.781512: step: 1692/529, loss: 0.0028154372703284025 2023-01-21 13:29:56.966603: step: 1696/529, loss: 0.0015116692520678043 2023-01-21 13:29:58.258612: step: 1700/529, loss: 0.006237888243049383 2023-01-21 13:29:59.422979: step: 1704/529, loss: 0.0020206451881676912 2023-01-21 13:30:00.615729: step: 1708/529, loss: 0.002826500218361616 2023-01-21 13:30:01.826498: step: 1712/529, loss: 7.05719003235572e-06 2023-01-21 13:30:03.064787: step: 1716/529, loss: 2.9945374990347773e-05 2023-01-21 13:30:04.250368: step: 1720/529, loss: 0.0034042359329760075 2023-01-21 13:30:05.481070: step: 1724/529, loss: 0.031420040875673294 2023-01-21 13:30:06.673858: step: 1728/529, loss: 0.0038157464005053043 2023-01-21 13:30:07.862444: step: 1732/529, loss: 0.0002608299255371094 2023-01-21 13:30:09.075038: step: 1736/529, loss: 0.029870223253965378 2023-01-21 13:30:10.229156: step: 1740/529, loss: 0.0625544935464859 2023-01-21 13:30:11.418866: step: 1744/529, loss: 0.0016108036506921053 2023-01-21 13:30:12.601947: step: 1748/529, loss: 0.03388862684369087 2023-01-21 13:30:13.826710: step: 1752/529, loss: 0.00012941360182594508 2023-01-21 13:30:15.024442: step: 1756/529, loss: 0.00017662048048805445 2023-01-21 13:30:16.201130: step: 1760/529, loss: 0.00014102458953857422 2023-01-21 13:30:17.332886: step: 1764/529, loss: 0.011363220401108265 2023-01-21 13:30:18.546401: step: 1768/529, loss: 5.2738188969669864e-05 2023-01-21 13:30:19.729644: step: 1772/529, loss: 0.0006916046258993447 2023-01-21 13:30:20.924006: step: 1776/529, loss: -1.4066696166992188e-05 2023-01-21 13:30:22.124990: step: 1780/529, loss: 0.000110626220703125 2023-01-21 13:30:23.341895: step: 1784/529, loss: 0.016381453722715378 2023-01-21 13:30:24.520497: step: 1788/529, loss: 0.00945901870727539 2023-01-21 13:30:25.742237: step: 1792/529, loss: 0.020259572193026543 2023-01-21 13:30:26.932807: step: 1796/529, loss: 0.0005234241834841669 2023-01-21 13:30:28.110367: step: 1800/529, loss: 7.858276512706652e-05 2023-01-21 13:30:29.322417: step: 1804/529, loss: 0.00010566711716819555 2023-01-21 13:30:30.506298: step: 1808/529, loss: 0.000653171562589705 2023-01-21 13:30:31.678621: step: 1812/529, loss: 0.039186764508485794 2023-01-21 13:30:32.951618: step: 1816/529, loss: 0.00087146763689816 2023-01-21 13:30:34.146299: step: 1820/529, loss: 9.91821252682712e-06 2023-01-21 13:30:35.316909: step: 1824/529, loss: 0.0004175186331849545 2023-01-21 13:30:36.549585: step: 1828/529, loss: 4.510879807639867e-05 2023-01-21 13:30:37.750441: step: 1832/529, loss: 0.000721836113370955 2023-01-21 13:30:38.932469: step: 1836/529, loss: 0.03773384168744087 2023-01-21 13:30:40.116604: step: 1840/529, loss: 0.0001464843808207661 2023-01-21 13:30:41.253413: step: 1844/529, loss: 0.0006329536554403603 2023-01-21 13:30:42.438644: step: 1848/529, loss: 5.5933000112418085e-05 2023-01-21 13:30:43.621724: step: 1852/529, loss: 0.0012897491687908769 2023-01-21 13:30:44.807013: step: 1856/529, loss: 0.02592296525835991 2023-01-21 13:30:45.981995: step: 1860/529, loss: 0.0015470505459234118 2023-01-21 13:30:47.153098: step: 1864/529, loss: 0.049543190747499466 2023-01-21 13:30:48.380015: step: 1868/529, loss: 0.0015035629039630294 2023-01-21 13:30:49.560054: step: 1872/529, loss: 0.0188782699406147 2023-01-21 13:30:50.745230: step: 1876/529, loss: 0.0211073886603117 2023-01-21 13:30:51.933609: step: 1880/529, loss: 0.03014841116964817 2023-01-21 13:30:53.109237: step: 1884/529, loss: 0.003857135772705078 2023-01-21 13:30:54.314056: step: 1888/529, loss: 0.008114052005112171 2023-01-21 13:30:55.476520: step: 1892/529, loss: 5.5789947509765625e-05 2023-01-21 13:30:56.714288: step: 1896/529, loss: 0.053807638585567474 2023-01-21 13:30:57.891187: step: 1900/529, loss: 0.0039161681197583675 2023-01-21 13:30:59.058974: step: 1904/529, loss: 7.762909081066027e-05 2023-01-21 13:31:00.278486: step: 1908/529, loss: 0.0004047393740620464 2023-01-21 13:31:01.477430: step: 1912/529, loss: 0.0036408426240086555 2023-01-21 13:31:02.700774: step: 1916/529, loss: 0.0027837753295898438 2023-01-21 13:31:03.923539: step: 1920/529, loss: 0.014561224728822708 2023-01-21 13:31:05.143290: step: 1924/529, loss: 6.151199340820312e-05 2023-01-21 13:31:06.311260: step: 1928/529, loss: 2.8514863515738398e-05 2023-01-21 13:31:07.480503: step: 1932/529, loss: 0.0009162903297692537 2023-01-21 13:31:08.700192: step: 1936/529, loss: 0.0015294075710698962 2023-01-21 13:31:09.915060: step: 1940/529, loss: 1.354217511106981e-05 2023-01-21 13:31:11.141453: step: 1944/529, loss: 0.001585769816301763 2023-01-21 13:31:12.380642: step: 1948/529, loss: 1.773834264895413e-05 2023-01-21 13:31:13.551953: step: 1952/529, loss: -2.28881845032447e-06 2023-01-21 13:31:14.728602: step: 1956/529, loss: 6.628036771871848e-06 2023-01-21 13:31:15.922589: step: 1960/529, loss: 0.03980560228228569 2023-01-21 13:31:17.126051: step: 1964/529, loss: 0.03052501752972603 2023-01-21 13:31:18.361441: step: 1968/529, loss: 1.2683869499596767e-05 2023-01-21 13:31:19.551316: step: 1972/529, loss: 0.0005664825439453125 2023-01-21 13:31:20.736918: step: 1976/529, loss: 1.2302398317842744e-05 2023-01-21 13:31:21.883881: step: 1980/529, loss: 0.026405811309814453 2023-01-21 13:31:23.036439: step: 1984/529, loss: 0.0452028289437294 2023-01-21 13:31:24.192954: step: 1988/529, loss: 2.098083541568485e-06 2023-01-21 13:31:25.413388: step: 1992/529, loss: 0.03825235366821289 2023-01-21 13:31:26.593237: step: 1996/529, loss: 8.592606172896922e-05 2023-01-21 13:31:27.784247: step: 2000/529, loss: -3.8146970382513246e-07 2023-01-21 13:31:28.956105: step: 2004/529, loss: 0.00030694008455611765 2023-01-21 13:31:30.142691: step: 2008/529, loss: 0.026889991015195847 2023-01-21 13:31:31.362482: step: 2012/529, loss: 0.011952591128647327 2023-01-21 13:31:32.537038: step: 2016/529, loss: 0.0002199172886321321 2023-01-21 13:31:33.703014: step: 2020/529, loss: -4.95910626341356e-06 2023-01-21 13:31:34.845953: step: 2024/529, loss: 0.0004527568817138672 2023-01-21 13:31:36.012196: step: 2028/529, loss: 7.233619544422254e-05 2023-01-21 13:31:37.213987: step: 2032/529, loss: 0.0010230063926428556 2023-01-21 13:31:38.387112: step: 2036/529, loss: 0.0031953812576830387 2023-01-21 13:31:39.589157: step: 2040/529, loss: 0.00025687218294478953 2023-01-21 13:31:40.753364: step: 2044/529, loss: 0.0012039184803143144 2023-01-21 13:31:41.950843: step: 2048/529, loss: 0.059897422790527344 2023-01-21 13:31:43.129703: step: 2052/529, loss: 3.2997133530443534e-05 2023-01-21 13:31:44.299039: step: 2056/529, loss: 0.011304378509521484 2023-01-21 13:31:45.494700: step: 2060/529, loss: 0.0005025863647460938 2023-01-21 13:31:46.684856: step: 2064/529, loss: 2.708435022213962e-05 2023-01-21 13:31:47.891324: step: 2068/529, loss: 9.183883958030492e-05 2023-01-21 13:31:49.116496: step: 2072/529, loss: 0.00016613007755950093 2023-01-21 13:31:50.292685: step: 2076/529, loss: 0.003967666532844305 2023-01-21 13:31:51.465601: step: 2080/529, loss: 0.00012187957327114418 2023-01-21 13:31:52.615322: step: 2084/529, loss: 9.002685692394152e-05 2023-01-21 13:31:53.848606: step: 2088/529, loss: 0.004203033167868853 2023-01-21 13:31:55.036751: step: 2092/529, loss: 0.000575208687223494 2023-01-21 13:31:56.195583: step: 2096/529, loss: 0.003066539764404297 2023-01-21 13:31:57.367547: step: 2100/529, loss: 7.677078247070312e-05 2023-01-21 13:31:58.545673: step: 2104/529, loss: 0.00011692046973621473 2023-01-21 13:31:59.723978: step: 2108/529, loss: 0.009930610656738281 2023-01-21 13:32:00.872668: step: 2112/529, loss: 4.3630603613564745e-05 2023-01-21 13:32:02.105330: step: 2116/529, loss: 0.0004192352353129536 ================================================== Loss: 0.026 -------------------- Dev: {'event': {'p': 0.5885311871227364, 'r': 0.7789613848202397, 'f1': 0.6704871060171921}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Test: {'event': {'p': 0.6141552511415526, 'r': 0.8058717795086878, 'f1': 0.6970717802539519}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Chinese: {'event': {'p': 0.6075949367088608, 'r': 0.8888888888888888, 'f1': 0.7218045112781953}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Korean: {'event': {'p': 0.5588235294117647, 'r': 0.6031746031746031, 'f1': 0.5801526717557252}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Russian: {'event': {'p': 0.4, 'r': 0.5, 'f1': 0.4444444444444445}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 13:32:45.741439: step: 4/529, loss: 4.863739013671875e-05 2023-01-21 13:32:46.969393: step: 8/529, loss: 0.007897567935287952 2023-01-21 13:32:48.127329: step: 12/529, loss: 0.0008023261907510459 2023-01-21 13:32:49.308129: step: 16/529, loss: -1.182556115963962e-05 2023-01-21 13:32:50.485917: step: 20/529, loss: 7.009506225585938e-05 2023-01-21 13:32:51.677144: step: 24/529, loss: 0.04871249198913574 2023-01-21 13:32:52.890005: step: 28/529, loss: 0.0014415740733966231 2023-01-21 13:32:54.051763: step: 32/529, loss: 0.000217437744140625 2023-01-21 13:32:55.263672: step: 36/529, loss: 0.00027027129544876516 2023-01-21 13:32:56.453058: step: 40/529, loss: 0.00011959076073253527 2023-01-21 13:32:57.653602: step: 44/529, loss: 0.00744972238317132 2023-01-21 13:32:58.821985: step: 48/529, loss: 3.395080420887098e-05 2023-01-21 13:33:00.075936: step: 52/529, loss: 1.7166138377433526e-06 2023-01-21 13:33:01.299652: step: 56/529, loss: 0.0001775741548044607 2023-01-21 13:33:02.468688: step: 60/529, loss: 2.3746491933707148e-05 2023-01-21 13:33:03.641650: step: 64/529, loss: 0.012915420345962048 2023-01-21 13:33:04.873305: step: 68/529, loss: 0.008329391479492188 2023-01-21 13:33:06.065925: step: 72/529, loss: 0.0016176223289221525 2023-01-21 13:33:07.258864: step: 76/529, loss: 0.00372905726544559 2023-01-21 13:33:08.476047: step: 80/529, loss: 0.0005328178522177041 2023-01-21 13:33:09.647361: step: 84/529, loss: 0.541806697845459 2023-01-21 13:33:10.842021: step: 88/529, loss: 0.5525747537612915 2023-01-21 13:33:12.066813: step: 92/529, loss: 0.03722648695111275 2023-01-21 13:33:13.276258: step: 96/529, loss: 0.0006057739374227822 2023-01-21 13:33:14.507218: step: 100/529, loss: 0.05169200897216797 2023-01-21 13:33:15.720724: step: 104/529, loss: 0.024047089740633965 2023-01-21 13:33:16.899979: step: 108/529, loss: -4.95910626341356e-06 2023-01-21 13:33:18.080963: step: 112/529, loss: 2.679825047380291e-05 2023-01-21 13:33:19.254442: step: 116/529, loss: 5.264282299322076e-05 2023-01-21 13:33:20.422489: step: 120/529, loss: 7.867813110351562e-05 2023-01-21 13:33:21.580053: step: 124/529, loss: 0.006603050045669079 2023-01-21 13:33:22.747824: step: 128/529, loss: 7.886887033237144e-05 2023-01-21 13:33:23.909500: step: 132/529, loss: 0.01063528098165989 2023-01-21 13:33:25.069720: step: 136/529, loss: 0.00019521712965797633 2023-01-21 13:33:26.355963: step: 140/529, loss: 0.12045975029468536 2023-01-21 13:33:27.515257: step: 144/529, loss: 1.9073486328125e-06 2023-01-21 13:33:28.677483: step: 148/529, loss: 0.052079200744628906 2023-01-21 13:33:29.926190: step: 152/529, loss: 0.0012541771866381168 2023-01-21 13:33:31.130488: step: 156/529, loss: 2.021789623540826e-05 2023-01-21 13:33:32.310302: step: 160/529, loss: 0.0002448081795591861 2023-01-21 13:33:33.491739: step: 164/529, loss: 0.03843631595373154 2023-01-21 13:33:34.668753: step: 168/529, loss: 0.0008377075428143144 2023-01-21 13:33:35.904532: step: 172/529, loss: 0.0003574371512513608 2023-01-21 13:33:37.080064: step: 176/529, loss: 0.004027176182717085 2023-01-21 13:33:38.237782: step: 180/529, loss: 4.391670154291205e-05 2023-01-21 13:33:39.433840: step: 184/529, loss: 0.035072021186351776 2023-01-21 13:33:40.657132: step: 188/529, loss: 1.239776611328125e-05 2023-01-21 13:33:41.858218: step: 192/529, loss: 0.0017299652099609375 2023-01-21 13:33:43.059164: step: 196/529, loss: 0.00040912628173828125 2023-01-21 13:33:44.236139: step: 200/529, loss: -1.1539459592313506e-05 2023-01-21 13:33:45.449461: step: 204/529, loss: 0.00048089027404785156 2023-01-21 13:33:46.650958: step: 208/529, loss: 0.0006774902576580644 2023-01-21 13:33:47.829342: step: 212/529, loss: 0.0016258240211755037 2023-01-21 13:33:48.999178: step: 216/529, loss: 0.00032072069006972015 2023-01-21 13:33:50.248889: step: 220/529, loss: 0.0018663406372070312 2023-01-21 13:33:51.435008: step: 224/529, loss: 7.715224637649953e-05 2023-01-21 13:33:52.620643: step: 228/529, loss: 0.010829126462340355 2023-01-21 13:33:53.786213: step: 232/529, loss: 0.00010375976125942543 2023-01-21 13:33:55.001641: step: 236/529, loss: 0.08788375556468964 2023-01-21 13:33:56.168158: step: 240/529, loss: 0.005086564924567938 2023-01-21 13:33:57.387457: step: 244/529, loss: 1.487731969973538e-05 2023-01-21 13:33:58.565999: step: 248/529, loss: 5.7220458984375e-06 2023-01-21 13:33:59.736314: step: 252/529, loss: 0.026588892564177513 2023-01-21 13:34:00.977712: step: 256/529, loss: 1.7929078239831142e-05 2023-01-21 13:34:02.163257: step: 260/529, loss: 8.20159948489163e-06 2023-01-21 13:34:03.353083: step: 264/529, loss: 0.0015436172252520919 2023-01-21 13:34:04.586202: step: 268/529, loss: 0.0009223938104696572 2023-01-21 13:34:05.779353: step: 272/529, loss: 0.0008216857677325606 2023-01-21 13:34:06.994910: step: 276/529, loss: 0.01021575927734375 2023-01-21 13:34:08.181008: step: 280/529, loss: 0.002016353653743863 2023-01-21 13:34:09.378622: step: 284/529, loss: 0.0005573272937908769 2023-01-21 13:34:10.538838: step: 288/529, loss: 0.00012702943058684468 2023-01-21 13:34:11.676896: step: 292/529, loss: 3.981589907198213e-05 2023-01-21 13:34:12.876648: step: 296/529, loss: 5.283355858409777e-05 2023-01-21 13:34:14.058476: step: 300/529, loss: -1.144409225162235e-06 2023-01-21 13:34:15.267038: step: 304/529, loss: 0.00012073517427779734 2023-01-21 13:34:16.447368: step: 308/529, loss: 0.00012073517427779734 2023-01-21 13:34:17.642884: step: 312/529, loss: 0.0034881592728197575 2023-01-21 13:34:18.857022: step: 316/529, loss: 0.004831505008041859 2023-01-21 13:34:20.051083: step: 320/529, loss: 0.045798301696777344 2023-01-21 13:34:21.222492: step: 324/529, loss: 0.00014333725266624242 2023-01-21 13:34:22.437222: step: 328/529, loss: 0.004484748933464289 2023-01-21 13:34:23.606746: step: 332/529, loss: 0.0028752805665135384 2023-01-21 13:34:24.794560: step: 336/529, loss: 0.005227088928222656 2023-01-21 13:34:26.019744: step: 340/529, loss: 0.015465069562196732 2023-01-21 13:34:27.214554: step: 344/529, loss: 3.299712989246473e-05 2023-01-21 13:34:28.416694: step: 348/529, loss: 0.016699982807040215 2023-01-21 13:34:29.588381: step: 352/529, loss: 0.002591419266536832 2023-01-21 13:34:30.807224: step: 356/529, loss: 5.1975250244140625e-05 2023-01-21 13:34:32.037228: step: 360/529, loss: 0.0003038406139239669 2023-01-21 13:34:33.260961: step: 364/529, loss: 0.014149331487715244 2023-01-21 13:34:34.442214: step: 368/529, loss: 0.021544646471738815 2023-01-21 13:34:35.601321: step: 372/529, loss: 0.007944679819047451 2023-01-21 13:34:36.776830: step: 376/529, loss: 0.0012592315906658769 2023-01-21 13:34:37.964926: step: 380/529, loss: 0.008535576052963734 2023-01-21 13:34:39.205415: step: 384/529, loss: 2.593994213384576e-05 2023-01-21 13:34:40.370957: step: 388/529, loss: 0.0004827499215025455 2023-01-21 13:34:41.560629: step: 392/529, loss: 0.002033233642578125 2023-01-21 13:34:42.758825: step: 396/529, loss: 2.0980833141948096e-06 2023-01-21 13:34:43.930555: step: 400/529, loss: 7.581711543025449e-05 2023-01-21 13:34:45.154398: step: 404/529, loss: 0.0004957198980264366 2023-01-21 13:34:46.387437: step: 408/529, loss: 0.020366763696074486 2023-01-21 13:34:47.587930: step: 412/529, loss: 0.055117081850767136 2023-01-21 13:34:48.731003: step: 416/529, loss: 0.025087451562285423 2023-01-21 13:34:49.919644: step: 420/529, loss: 0.00012531279935501516 2023-01-21 13:34:51.136210: step: 424/529, loss: 0.005813026800751686 2023-01-21 13:34:52.298946: step: 428/529, loss: 0.059925176203250885 2023-01-21 13:34:53.490522: step: 432/529, loss: 0.001520443009212613 2023-01-21 13:34:54.665229: step: 436/529, loss: 0.00107402796857059 2023-01-21 13:34:55.848637: step: 440/529, loss: 0.007129669189453125 2023-01-21 13:34:57.024191: step: 444/529, loss: 0.0007104873657226562 2023-01-21 13:34:58.234989: step: 448/529, loss: 0.009379005990922451 2023-01-21 13:34:59.444558: step: 452/529, loss: 4.186630030744709e-05 2023-01-21 13:35:00.637587: step: 456/529, loss: 0.002368164248764515 2023-01-21 13:35:01.793827: step: 460/529, loss: 0.5788941383361816 2023-01-21 13:35:03.027659: step: 464/529, loss: 0.00020599366689566523 2023-01-21 13:35:04.178536: step: 468/529, loss: 0.0011881828540936112 2023-01-21 13:35:05.373978: step: 472/529, loss: 7.743835158180445e-05 2023-01-21 13:35:06.573612: step: 476/529, loss: 0.0033714293967932463 2023-01-21 13:35:07.737600: step: 480/529, loss: 0.08300705254077911 2023-01-21 13:35:08.967170: step: 484/529, loss: 0.005856513977050781 2023-01-21 13:35:10.135468: step: 488/529, loss: 2.8324127924861386e-05 2023-01-21 13:35:11.302412: step: 492/529, loss: 3.6811830796068534e-05 2023-01-21 13:35:12.509818: step: 496/529, loss: 0.0002978801785502583 2023-01-21 13:35:13.706765: step: 500/529, loss: 6.198883056640625e-05 2023-01-21 13:35:14.919486: step: 504/529, loss: 0.008921146392822266 2023-01-21 13:35:16.097152: step: 508/529, loss: 3.643035961431451e-05 2023-01-21 13:35:17.275907: step: 512/529, loss: 0.09600763767957687 2023-01-21 13:35:18.455501: step: 516/529, loss: 0.0012737273937091231 2023-01-21 13:35:19.637611: step: 520/529, loss: 0.0014580488204956055 2023-01-21 13:35:20.834917: step: 524/529, loss: 0.011813831515610218 2023-01-21 13:35:22.090980: step: 528/529, loss: 0.750552773475647 2023-01-21 13:35:23.245626: step: 532/529, loss: 0.014604568481445312 2023-01-21 13:35:24.437955: step: 536/529, loss: 0.04468338564038277 2023-01-21 13:35:25.641432: step: 540/529, loss: 8.316040475619957e-05 2023-01-21 13:35:26.808126: step: 544/529, loss: 1.9073468138230965e-07 2023-01-21 13:35:28.034110: step: 548/529, loss: 0.0031305314041674137 2023-01-21 13:35:29.236019: step: 552/529, loss: 3.6239625842426904e-06 2023-01-21 13:35:30.457400: step: 556/529, loss: 0.00011024475679732859 2023-01-21 13:35:31.609339: step: 560/529, loss: 0.0006025314214639366 2023-01-21 13:35:32.808095: step: 564/529, loss: 4.272460864740424e-05 2023-01-21 13:35:34.060156: step: 568/529, loss: 0.07542724907398224 2023-01-21 13:35:35.209005: step: 572/529, loss: 5.035400317865424e-05 2023-01-21 13:35:36.380670: step: 576/529, loss: 0.011050987988710403 2023-01-21 13:35:37.583751: step: 580/529, loss: 0.0009958267910405993 2023-01-21 13:35:38.759012: step: 584/529, loss: 0.0032279014121741056 2023-01-21 13:35:39.926925: step: 588/529, loss: 0.012427901849150658 2023-01-21 13:35:41.128430: step: 592/529, loss: 0.0003264427068643272 2023-01-21 13:35:42.347247: step: 596/529, loss: 0.05097704008221626 2023-01-21 13:35:43.532736: step: 600/529, loss: 0.03349733352661133 2023-01-21 13:35:44.709382: step: 604/529, loss: 0.04917879030108452 2023-01-21 13:35:45.956484: step: 608/529, loss: 0.00736732454970479 2023-01-21 13:35:47.165260: step: 612/529, loss: 6.961822509765625e-05 2023-01-21 13:35:48.347316: step: 616/529, loss: 1.087188684323337e-05 2023-01-21 13:35:49.522695: step: 620/529, loss: 3.242493221478071e-06 2023-01-21 13:35:50.689328: step: 624/529, loss: 3.25202927342616e-05 2023-01-21 13:35:51.863139: step: 628/529, loss: -3.814697265625e-06 2023-01-21 13:35:53.050522: step: 632/529, loss: 0.037526510655879974 2023-01-21 13:35:54.225888: step: 636/529, loss: 0.0010787963401526213 2023-01-21 13:35:55.404408: step: 640/529, loss: 0.003557920688763261 2023-01-21 13:35:56.602121: step: 644/529, loss: 0.009724712930619717 2023-01-21 13:35:57.784774: step: 648/529, loss: 0.0016495705349370837 2023-01-21 13:35:58.960195: step: 652/529, loss: 0.0005769729614257812 2023-01-21 13:36:00.208504: step: 656/529, loss: 5.91278057981981e-06 2023-01-21 13:36:01.397324: step: 660/529, loss: 0.001140546752139926 2023-01-21 13:36:02.584538: step: 664/529, loss: 0.011935329996049404 2023-01-21 13:36:03.785534: step: 668/529, loss: 0.00394783029332757 2023-01-21 13:36:04.965168: step: 672/529, loss: 0.0012670516734942794 2023-01-21 13:36:06.165010: step: 676/529, loss: 0.1113218367099762 2023-01-21 13:36:07.388206: step: 680/529, loss: 0.0009527206420898438 2023-01-21 13:36:08.573126: step: 684/529, loss: 0.0025323869194835424 2023-01-21 13:36:09.761844: step: 688/529, loss: -1.52587890625e-05 2023-01-21 13:36:10.937837: step: 692/529, loss: 0.00011291504779364914 2023-01-21 13:36:12.107349: step: 696/529, loss: -3.14712519866589e-06 2023-01-21 13:36:13.347505: step: 700/529, loss: 0.00687065115198493 2023-01-21 13:36:14.480612: step: 704/529, loss: 0.000339508056640625 2023-01-21 13:36:15.663656: step: 708/529, loss: 0.008040428161621094 2023-01-21 13:36:16.857814: step: 712/529, loss: 0.0005170822259970009 2023-01-21 13:36:18.024380: step: 716/529, loss: 0.0002056121884379536 2023-01-21 13:36:19.246082: step: 720/529, loss: 0.024943161755800247 2023-01-21 13:36:20.405572: step: 724/529, loss: 0.0004837989981751889 2023-01-21 13:36:21.607950: step: 728/529, loss: 0.0034859657753258944 2023-01-21 13:36:22.785934: step: 732/529, loss: 0.0008152961963787675 2023-01-21 13:36:23.967358: step: 736/529, loss: 3.757476952159777e-05 2023-01-21 13:36:25.213408: step: 740/529, loss: 0.06342391669750214 2023-01-21 13:36:26.387850: step: 744/529, loss: 0.00021781922259833664 2023-01-21 13:36:27.548476: step: 748/529, loss: 1.3065338862361386e-05 2023-01-21 13:36:28.715928: step: 752/529, loss: 0.004344367887824774 2023-01-21 13:36:29.911558: step: 756/529, loss: 0.053530313074588776 2023-01-21 13:36:31.115454: step: 760/529, loss: 0.802336573600769 2023-01-21 13:36:32.314226: step: 764/529, loss: 0.0009565353393554688 2023-01-21 13:36:33.528855: step: 768/529, loss: 0.0004884243244305253 2023-01-21 13:36:34.746700: step: 772/529, loss: 0.022400524467229843 2023-01-21 13:36:35.935883: step: 776/529, loss: 0.01739025115966797 2023-01-21 13:36:37.094899: step: 780/529, loss: 0.0001124382033594884 2023-01-21 13:36:38.293590: step: 784/529, loss: 0.008530139923095703 2023-01-21 13:36:39.468451: step: 788/529, loss: 1.8596649169921875e-05 2023-01-21 13:36:40.638889: step: 792/529, loss: 0.010828113183379173 2023-01-21 13:36:41.844766: step: 796/529, loss: 0.048886775970458984 2023-01-21 13:36:43.057970: step: 800/529, loss: 0.00014734268188476562 2023-01-21 13:36:44.211323: step: 804/529, loss: 0.0017309188842773438 2023-01-21 13:36:45.378517: step: 808/529, loss: 0.08981933444738388 2023-01-21 13:36:46.589988: step: 812/529, loss: 0.04576887935400009 2023-01-21 13:36:47.774116: step: 816/529, loss: 3.0231476557673886e-05 2023-01-21 13:36:48.936854: step: 820/529, loss: 4.587173680192791e-05 2023-01-21 13:36:50.142851: step: 824/529, loss: 0.01203455962240696 2023-01-21 13:36:51.323347: step: 828/529, loss: 3.0279159545898438e-05 2023-01-21 13:36:52.482591: step: 832/529, loss: 0.0066534047946333885 2023-01-21 13:36:53.686825: step: 836/529, loss: 0.0025269510224461555 2023-01-21 13:36:54.889039: step: 840/529, loss: 0.039559267461299896 2023-01-21 13:36:56.071214: step: 844/529, loss: 9.393692016601562e-05 2023-01-21 13:36:57.283581: step: 848/529, loss: 0.04873852804303169 2023-01-21 13:36:58.461964: step: 852/529, loss: 0.013102197088301182 2023-01-21 13:36:59.633923: step: 856/529, loss: 0.00043487548828125 2023-01-21 13:37:00.848771: step: 860/529, loss: 0.02509746514260769 2023-01-21 13:37:02.095674: step: 864/529, loss: 0.03627157211303711 2023-01-21 13:37:03.321814: step: 868/529, loss: 0.0024751664604991674 2023-01-21 13:37:04.512784: step: 872/529, loss: 0.0014551162021234632 2023-01-21 13:37:05.772416: step: 876/529, loss: 0.0004028320254292339 2023-01-21 13:37:06.972979: step: 880/529, loss: 1.0967254638671875e-05 2023-01-21 13:37:08.145230: step: 884/529, loss: 8.58306884765625e-06 2023-01-21 13:37:09.301656: step: 888/529, loss: 0.0002563476446084678 2023-01-21 13:37:10.512263: step: 892/529, loss: 0.0015506744384765625 2023-01-21 13:37:11.668344: step: 896/529, loss: 0.0005542993894778192 2023-01-21 13:37:12.850236: step: 900/529, loss: 1.5926361811580136e-05 2023-01-21 13:37:14.059199: step: 904/529, loss: 0.011474848724901676 2023-01-21 13:37:15.265217: step: 908/529, loss: 0.031229974702000618 2023-01-21 13:37:16.432158: step: 912/529, loss: 2.4700166250113398e-05 2023-01-21 13:37:17.636806: step: 916/529, loss: 0.00014057158841751516 2023-01-21 13:37:18.848848: step: 920/529, loss: 6.618499901378527e-05 2023-01-21 13:37:20.058142: step: 924/529, loss: -2.346038854739163e-05 2023-01-21 13:37:21.265689: step: 928/529, loss: 0.07287827134132385 2023-01-21 13:37:22.473325: step: 932/529, loss: -2.002715973503655e-06 2023-01-21 13:37:23.715180: step: 936/529, loss: 0.009729957208037376 2023-01-21 13:37:24.915326: step: 940/529, loss: 0.00077905657235533 2023-01-21 13:37:26.107100: step: 944/529, loss: 0.00020761489577125758 2023-01-21 13:37:27.263499: step: 948/529, loss: 0.0003769874747376889 2023-01-21 13:37:28.452175: step: 952/529, loss: 0.0006922722095623612 2023-01-21 13:37:29.632293: step: 956/529, loss: 0.0003673553583212197 2023-01-21 13:37:30.824557: step: 960/529, loss: 2.956390289909905e-06 2023-01-21 13:37:32.028083: step: 964/529, loss: 0.00026359560433775187 2023-01-21 13:37:33.209143: step: 968/529, loss: 1.2969970157428179e-05 2023-01-21 13:37:34.403323: step: 972/529, loss: 0.05458088219165802 2023-01-21 13:37:35.572276: step: 976/529, loss: 1.1444091796875e-05 2023-01-21 13:37:36.805063: step: 980/529, loss: 0.026933861896395683 2023-01-21 13:37:37.975291: step: 984/529, loss: 0.0008135795942507684 2023-01-21 13:37:39.190817: step: 988/529, loss: 2.8991700673941523e-05 2023-01-21 13:37:40.402606: step: 992/529, loss: 0.0004686355823650956 2023-01-21 13:37:41.610980: step: 996/529, loss: 0.02346344105899334 2023-01-21 13:37:42.775210: step: 1000/529, loss: 0.004833845887333155 2023-01-21 13:37:43.995819: step: 1004/529, loss: 0.0020793916191905737 2023-01-21 13:37:45.207118: step: 1008/529, loss: 0.007673073094338179 2023-01-21 13:37:46.387593: step: 1012/529, loss: 5.0830840336857364e-05 2023-01-21 13:37:47.567344: step: 1016/529, loss: 0.0014425277477130294 2023-01-21 13:37:48.738636: step: 1020/529, loss: 4.76837158203125e-07 2023-01-21 13:37:49.915724: step: 1024/529, loss: 6.279945955611765e-05 2023-01-21 13:37:51.087457: step: 1028/529, loss: 0.0030023574363440275 2023-01-21 13:37:52.343621: step: 1032/529, loss: 0.020935440436005592 2023-01-21 13:37:53.509188: step: 1036/529, loss: 0.0015079497825354338 2023-01-21 13:37:54.738263: step: 1040/529, loss: 8.869171324477065e-06 2023-01-21 13:37:55.951750: step: 1044/529, loss: 0.03527965396642685 2023-01-21 13:37:57.124843: step: 1048/529, loss: 1.0490418844710803e-06 2023-01-21 13:37:58.285051: step: 1052/529, loss: 0.0003127098025288433 2023-01-21 13:37:59.451778: step: 1056/529, loss: 0.0001999855157919228 2023-01-21 13:38:00.644986: step: 1060/529, loss: 0.00170307164080441 2023-01-21 13:38:01.819739: step: 1064/529, loss: 0.0006354332435876131 2023-01-21 13:38:02.992399: step: 1068/529, loss: 3.6859513784293085e-05 2023-01-21 13:38:04.157903: step: 1072/529, loss: 7.724761962890625e-05 2023-01-21 13:38:05.322750: step: 1076/529, loss: 4.57763690064894e-06 2023-01-21 13:38:06.476430: step: 1080/529, loss: 2.765656063274946e-06 2023-01-21 13:38:07.641149: step: 1084/529, loss: 0.06073570251464844 2023-01-21 13:38:08.852743: step: 1088/529, loss: 0.0004094124014955014 2023-01-21 13:38:10.002165: step: 1092/529, loss: 3.910064697265625e-05 2023-01-21 13:38:11.191252: step: 1096/529, loss: 0.0017435074551030993 2023-01-21 13:38:12.382753: step: 1100/529, loss: 0.0064581395126879215 2023-01-21 13:38:13.588804: step: 1104/529, loss: 0.02167205885052681 2023-01-21 13:38:14.785949: step: 1108/529, loss: 1.5068053471622989e-05 2023-01-21 13:38:15.946042: step: 1112/529, loss: 0.004329585935920477 2023-01-21 13:38:17.146133: step: 1116/529, loss: 0.009985638782382011 2023-01-21 13:38:18.380355: step: 1120/529, loss: 0.00016508102999068797 2023-01-21 13:38:19.537317: step: 1124/529, loss: 0.0026569366455078125 2023-01-21 13:38:20.692429: step: 1128/529, loss: 0.004138756077736616 2023-01-21 13:38:21.885264: step: 1132/529, loss: 0.008070182986557484 2023-01-21 13:38:23.040772: step: 1136/529, loss: 4.00543194700731e-06 2023-01-21 13:38:24.224954: step: 1140/529, loss: 0.0002991676446981728 2023-01-21 13:38:25.440154: step: 1144/529, loss: 0.016420649364590645 2023-01-21 13:38:26.627143: step: 1148/529, loss: 0.003762149950489402 2023-01-21 13:38:27.803119: step: 1152/529, loss: 0.036241911351680756 2023-01-21 13:38:28.990846: step: 1156/529, loss: 0.011696243658661842 2023-01-21 13:38:30.202824: step: 1160/529, loss: 0.05250797048211098 2023-01-21 13:38:31.394311: step: 1164/529, loss: 6.69479341013357e-05 2023-01-21 13:38:32.651705: step: 1168/529, loss: 0.0013175965286791325 2023-01-21 13:38:33.813200: step: 1172/529, loss: 7.629396350239404e-07 2023-01-21 13:38:35.001134: step: 1176/529, loss: 0.0044708251953125 2023-01-21 13:38:36.183409: step: 1180/529, loss: 0.014200210571289062 2023-01-21 13:38:37.407486: step: 1184/529, loss: 0.0027084352914243937 2023-01-21 13:38:38.600109: step: 1188/529, loss: 0.011113166809082031 2023-01-21 13:38:39.792256: step: 1192/529, loss: 0.0011014938354492188 2023-01-21 13:38:41.005120: step: 1196/529, loss: 0.006846427917480469 2023-01-21 13:38:42.193178: step: 1200/529, loss: 0.0006832123035565019 2023-01-21 13:38:43.372908: step: 1204/529, loss: 0.00381641392596066 2023-01-21 13:38:44.590991: step: 1208/529, loss: 0.02046814002096653 2023-01-21 13:38:45.810424: step: 1212/529, loss: 0.008275222964584827 2023-01-21 13:38:46.982318: step: 1216/529, loss: 0.0030252456199377775 2023-01-21 13:38:48.154502: step: 1220/529, loss: 0.03834524378180504 2023-01-21 13:38:49.347625: step: 1224/529, loss: 0.00010585784912109375 2023-01-21 13:38:50.522445: step: 1228/529, loss: 0.0001394271821482107 2023-01-21 13:38:51.729901: step: 1232/529, loss: 0.00040798186091706157 2023-01-21 13:38:52.871330: step: 1236/529, loss: 0.0002079010009765625 2023-01-21 13:38:54.074855: step: 1240/529, loss: 0.0011264801723882556 2023-01-21 13:38:55.256900: step: 1244/529, loss: 0.007622480392456055 2023-01-21 13:38:56.452254: step: 1248/529, loss: 0.08837050944566727 2023-01-21 13:38:57.686144: step: 1252/529, loss: 0.00016403199697379023 2023-01-21 13:38:58.927055: step: 1256/529, loss: 0.0014233589172363281 2023-01-21 13:39:00.177422: step: 1260/529, loss: 4.234314110362902e-05 2023-01-21 13:39:01.358545: step: 1264/529, loss: 0.021548127755522728 2023-01-21 13:39:02.554254: step: 1268/529, loss: 0.013478565961122513 2023-01-21 13:39:03.762415: step: 1272/529, loss: 0.09491729736328125 2023-01-21 13:39:04.978525: step: 1276/529, loss: 0.04279670864343643 2023-01-21 13:39:06.178602: step: 1280/529, loss: 0.0018508434295654297 2023-01-21 13:39:07.362798: step: 1284/529, loss: 0.02612590789794922 2023-01-21 13:39:08.554931: step: 1288/529, loss: 0.0014410018920898438 2023-01-21 13:39:09.743917: step: 1292/529, loss: 8.1062319168268e-07 2023-01-21 13:39:10.968284: step: 1296/529, loss: 0.00702247628942132 2023-01-21 13:39:12.126260: step: 1300/529, loss: 0.00023112297640182078 2023-01-21 13:39:13.323252: step: 1304/529, loss: 0.0007623672718182206 2023-01-21 13:39:14.523479: step: 1308/529, loss: 0.00037822723970748484 2023-01-21 13:39:15.664867: step: 1312/529, loss: 0.02459869533777237 2023-01-21 13:39:16.886239: step: 1316/529, loss: 0.009889364242553711 2023-01-21 13:39:18.091683: step: 1320/529, loss: 0.00023279190645553172 2023-01-21 13:39:19.247858: step: 1324/529, loss: -1.5258790426742053e-06 2023-01-21 13:39:20.456428: step: 1328/529, loss: 0.4184935688972473 2023-01-21 13:39:21.676742: step: 1332/529, loss: 0.040122415870428085 2023-01-21 13:39:22.842346: step: 1336/529, loss: 0.01235275249928236 2023-01-21 13:39:24.015055: step: 1340/529, loss: 0.0002510070917196572 2023-01-21 13:39:25.209665: step: 1344/529, loss: 9.059906460606726e-07 2023-01-21 13:39:26.412599: step: 1348/529, loss: 0.0001520156947663054 2023-01-21 13:39:27.642099: step: 1352/529, loss: 0.0017858505016192794 2023-01-21 13:39:28.812100: step: 1356/529, loss: 0.001049041748046875 2023-01-21 13:39:29.976257: step: 1360/529, loss: 1.7356871467200108e-05 2023-01-21 13:39:31.196315: step: 1364/529, loss: 2.689361645025201e-05 2023-01-21 13:39:32.440371: step: 1368/529, loss: 0.012994194403290749 2023-01-21 13:39:33.611745: step: 1372/529, loss: 0.00758857699111104 2023-01-21 13:39:34.816045: step: 1376/529, loss: 0.059294987469911575 2023-01-21 13:39:36.006206: step: 1380/529, loss: 0.007464217953383923 2023-01-21 13:39:37.158933: step: 1384/529, loss: 1.068115216185106e-05 2023-01-21 13:39:38.363153: step: 1388/529, loss: -1.0013580322265625e-05 2023-01-21 13:39:39.545696: step: 1392/529, loss: 0.0013828754890710115 2023-01-21 13:39:40.755790: step: 1396/529, loss: 0.0034358978737145662 2023-01-21 13:39:41.920511: step: 1400/529, loss: 1.1777878171415068e-05 2023-01-21 13:39:43.082459: step: 1404/529, loss: 0.0036846159491688013 2023-01-21 13:39:44.255904: step: 1408/529, loss: 0.0008441925165243447 2023-01-21 13:39:45.474947: step: 1412/529, loss: 0.0008647918584756553 2023-01-21 13:39:46.654241: step: 1416/529, loss: 0.03754768148064613 2023-01-21 13:39:47.848837: step: 1420/529, loss: 9.584426152287051e-06 2023-01-21 13:39:49.042119: step: 1424/529, loss: 0.045939065515995026 2023-01-21 13:39:50.324882: step: 1428/529, loss: 0.0003535270516294986 2023-01-21 13:39:51.497521: step: 1432/529, loss: 0.0005018234369345009 2023-01-21 13:39:52.675103: step: 1436/529, loss: 0.0001672744838288054 2023-01-21 13:39:53.879205: step: 1440/529, loss: 0.0007632256019860506 2023-01-21 13:39:55.071147: step: 1444/529, loss: 3.9291378925554454e-05 2023-01-21 13:39:56.274478: step: 1448/529, loss: 0.00011777878535212949 2023-01-21 13:39:57.502919: step: 1452/529, loss: 0.00047130585880950093 2023-01-21 13:39:58.685301: step: 1456/529, loss: 2.050399962172378e-05 2023-01-21 13:39:59.852951: step: 1460/529, loss: 6.370544724632055e-05 2023-01-21 13:40:01.036875: step: 1464/529, loss: 0.0018415451049804688 2023-01-21 13:40:02.210837: step: 1468/529, loss: 8.869171324477065e-06 2023-01-21 13:40:03.435187: step: 1472/529, loss: 0.0006696701166220009 2023-01-21 13:40:04.585367: step: 1476/529, loss: 9.069442603504285e-05 2023-01-21 13:40:05.863589: step: 1480/529, loss: 0.024466706439852715 2023-01-21 13:40:07.097522: step: 1484/529, loss: 0.011666965670883656 2023-01-21 13:40:08.298791: step: 1488/529, loss: 0.011014366522431374 2023-01-21 13:40:09.487206: step: 1492/529, loss: 0.026410484686493874 2023-01-21 13:40:10.682594: step: 1496/529, loss: 0.00019407272338867188 2023-01-21 13:40:11.882545: step: 1500/529, loss: 0.003574156668037176 2023-01-21 13:40:13.045289: step: 1504/529, loss: 0.015203761868178844 2023-01-21 13:40:14.231872: step: 1508/529, loss: 0.0036230087280273438 2023-01-21 13:40:15.434237: step: 1512/529, loss: 0.0001472473086323589 2023-01-21 13:40:16.624420: step: 1516/529, loss: 0.06792354583740234 2023-01-21 13:40:17.896144: step: 1520/529, loss: 0.0014261245960369706 2023-01-21 13:40:19.055148: step: 1524/529, loss: 8.106231689453125e-06 2023-01-21 13:40:20.241538: step: 1528/529, loss: 7.247925623232732e-06 2023-01-21 13:40:21.467787: step: 1532/529, loss: 0.0002766609250102192 2023-01-21 13:40:22.648116: step: 1536/529, loss: 1.640319896978326e-05 2023-01-21 13:40:23.869689: step: 1540/529, loss: 4.978180004400201e-05 2023-01-21 13:40:25.007905: step: 1544/529, loss: 0.0013835907448083162 2023-01-21 13:40:26.269390: step: 1548/529, loss: 0.03432135656476021 2023-01-21 13:40:27.457158: step: 1552/529, loss: 5.517006138688885e-05 2023-01-21 13:40:28.659381: step: 1556/529, loss: 0.0003009796200785786 2023-01-21 13:40:29.844141: step: 1560/529, loss: 0.006619930267333984 2023-01-21 13:40:31.044097: step: 1564/529, loss: 0.02258882485330105 2023-01-21 13:40:32.244507: step: 1568/529, loss: 0.000751876796130091 2023-01-21 13:40:33.430815: step: 1572/529, loss: 0.11988982558250427 2023-01-21 13:40:34.615071: step: 1576/529, loss: 0.0005485772853717208 2023-01-21 13:40:35.840596: step: 1580/529, loss: 0.027642060071229935 2023-01-21 13:40:37.029341: step: 1584/529, loss: 0.014576531015336514 2023-01-21 13:40:38.194309: step: 1588/529, loss: 9.183883958030492e-05 2023-01-21 13:40:39.381347: step: 1592/529, loss: 0.004127406980842352 2023-01-21 13:40:40.537773: step: 1596/529, loss: 3.34262840624433e-05 2023-01-21 13:40:41.752023: step: 1600/529, loss: 6.8664553509734105e-06 2023-01-21 13:40:42.919991: step: 1604/529, loss: 0.030353927984833717 2023-01-21 13:40:44.129491: step: 1608/529, loss: 0.05172233656048775 2023-01-21 13:40:45.331021: step: 1612/529, loss: 0.001163482666015625 2023-01-21 13:40:46.501123: step: 1616/529, loss: 0.0005651473766192794 2023-01-21 13:40:47.684522: step: 1620/529, loss: 0.014611911959946156 2023-01-21 13:40:48.865117: step: 1624/529, loss: 3.814697265625e-05 2023-01-21 13:40:50.070440: step: 1628/529, loss: 0.05935544893145561 2023-01-21 13:40:51.263975: step: 1632/529, loss: 2.5177003408316523e-05 2023-01-21 13:40:52.484733: step: 1636/529, loss: 0.0007702827570028603 2023-01-21 13:40:53.685513: step: 1640/529, loss: 0.002065467881038785 2023-01-21 13:40:54.904701: step: 1644/529, loss: 0.005012321285903454 2023-01-21 13:40:56.064647: step: 1648/529, loss: 0.0012476444244384766 2023-01-21 13:40:57.233793: step: 1652/529, loss: 0.002281284425407648 2023-01-21 13:40:58.398508: step: 1656/529, loss: 0.009362364187836647 2023-01-21 13:40:59.563983: step: 1660/529, loss: 0.037930678576231 2023-01-21 13:41:00.796553: step: 1664/529, loss: 0.243106871843338 2023-01-21 13:41:01.974765: step: 1668/529, loss: 0.0015608788235113025 2023-01-21 13:41:03.155443: step: 1672/529, loss: 9.35554489842616e-05 2023-01-21 13:41:04.320834: step: 1676/529, loss: 7.543563697254285e-05 2023-01-21 13:41:05.489285: step: 1680/529, loss: 0.0008529663318768144 2023-01-21 13:41:06.686968: step: 1684/529, loss: 2.8610247682081535e-07 2023-01-21 13:41:07.842450: step: 1688/529, loss: 0.00015802383131813258 2023-01-21 13:41:09.028242: step: 1692/529, loss: 0.010190105997025967 2023-01-21 13:41:10.205628: step: 1696/529, loss: 0.005534553434699774 2023-01-21 13:41:11.420672: step: 1700/529, loss: 0.0004459381161723286 2023-01-21 13:41:12.584846: step: 1704/529, loss: 4.76837158203125e-06 2023-01-21 13:41:13.776958: step: 1708/529, loss: 0.02378368377685547 2023-01-21 13:41:14.942545: step: 1712/529, loss: 0.006598091218620539 2023-01-21 13:41:16.123293: step: 1716/529, loss: 0.019258594140410423 2023-01-21 13:41:17.294792: step: 1720/529, loss: 3.185272362316027e-05 2023-01-21 13:41:18.520771: step: 1724/529, loss: 0.012248420156538486 2023-01-21 13:41:19.709794: step: 1728/529, loss: 0.0001103401227737777 2023-01-21 13:41:20.910752: step: 1732/529, loss: 0.006299782078713179 2023-01-21 13:41:22.080217: step: 1736/529, loss: 0.006045246031135321 2023-01-21 13:41:23.251533: step: 1740/529, loss: 0.000209808349609375 2023-01-21 13:41:24.437499: step: 1744/529, loss: 8.75473051564768e-05 2023-01-21 13:41:25.634417: step: 1748/529, loss: 0.001033020089380443 2023-01-21 13:41:26.829207: step: 1752/529, loss: 0.02100515365600586 2023-01-21 13:41:28.026963: step: 1756/529, loss: -7.057189577608369e-06 2023-01-21 13:41:29.239069: step: 1760/529, loss: -4.0054324017546605e-06 2023-01-21 13:41:30.379692: step: 1764/529, loss: 0.016878794878721237 2023-01-21 13:41:31.559839: step: 1768/529, loss: 0.0010881423950195312 2023-01-21 13:41:32.743425: step: 1772/529, loss: 1.6880036127986386e-05 2023-01-21 13:41:33.871422: step: 1776/529, loss: 0.06962118297815323 2023-01-21 13:41:35.051227: step: 1780/529, loss: 0.0021721841767430305 2023-01-21 13:41:36.246507: step: 1784/529, loss: 0.024844933301210403 2023-01-21 13:41:37.391318: step: 1788/529, loss: 5.321502612787299e-05 2023-01-21 13:41:38.584336: step: 1792/529, loss: 0.00026512148906476796 2023-01-21 13:41:39.759519: step: 1796/529, loss: 0.038497354835271835 2023-01-21 13:41:40.947536: step: 1800/529, loss: -9.5367431640625e-06 2023-01-21 13:41:42.111924: step: 1804/529, loss: 0.002333736512809992 2023-01-21 13:41:43.292505: step: 1808/529, loss: 9.393692016601562e-05 2023-01-21 13:41:44.510966: step: 1812/529, loss: 0.0052017210982739925 2023-01-21 13:41:45.666264: step: 1816/529, loss: 0.0008275986183434725 2023-01-21 13:41:46.824444: step: 1820/529, loss: 8.707046799827367e-05 2023-01-21 13:41:47.983168: step: 1824/529, loss: 0.008606362156569958 2023-01-21 13:41:49.215512: step: 1828/529, loss: 0.0002910613839048892 2023-01-21 13:41:50.384126: step: 1832/529, loss: 0.011018658056855202 2023-01-21 13:41:51.571107: step: 1836/529, loss: 2.975463939947076e-05 2023-01-21 13:41:52.790864: step: 1840/529, loss: 0.00067224504891783 2023-01-21 13:41:54.003144: step: 1844/529, loss: 0.00022230148897506297 2023-01-21 13:41:55.195218: step: 1848/529, loss: 2.937316821771674e-05 2023-01-21 13:41:56.416777: step: 1852/529, loss: 0.00011758803884731606 2023-01-21 13:41:57.605381: step: 1856/529, loss: 0.002803039737045765 2023-01-21 13:41:58.800932: step: 1860/529, loss: 0.0055374144576489925 2023-01-21 13:42:00.002949: step: 1864/529, loss: 0.0005592346424236894 2023-01-21 13:42:01.172093: step: 1868/529, loss: 0.02321653440594673 2023-01-21 13:42:02.404507: step: 1872/529, loss: 5.416870044427924e-05 2023-01-21 13:42:03.596300: step: 1876/529, loss: 0.002674198243767023 2023-01-21 13:42:04.806011: step: 1880/529, loss: -7.033347628748743e-07 2023-01-21 13:42:05.981821: step: 1884/529, loss: 0.6018469333648682 2023-01-21 13:42:07.173854: step: 1888/529, loss: 2.47955322265625e-05 2023-01-21 13:42:08.369775: step: 1892/529, loss: 0.001531410263851285 2023-01-21 13:42:09.593324: step: 1896/529, loss: 0.0004920900100842118 2023-01-21 13:42:10.771330: step: 1900/529, loss: 0.03541088104248047 2023-01-21 13:42:11.960244: step: 1904/529, loss: 0.00011940002150367945 2023-01-21 13:42:13.154360: step: 1908/529, loss: 0.00033655168954283 2023-01-21 13:42:14.363738: step: 1912/529, loss: 0.007105445954948664 2023-01-21 13:42:15.526121: step: 1916/529, loss: 6.008148375258315e-06 2023-01-21 13:42:16.717888: step: 1920/529, loss: 0.012304354459047318 2023-01-21 13:42:17.893567: step: 1924/529, loss: 9.250640687241685e-06 2023-01-21 13:42:19.079570: step: 1928/529, loss: 0.019771575927734375 2023-01-21 13:42:20.282067: step: 1932/529, loss: 0.0007734298706054688 2023-01-21 13:42:21.498804: step: 1936/529, loss: 0.001318645547144115 2023-01-21 13:42:22.664037: step: 1940/529, loss: 0.00352058419957757 2023-01-21 13:42:23.887936: step: 1944/529, loss: 3.4809113458322827e-06 2023-01-21 13:42:25.081113: step: 1948/529, loss: 0.0061775208450853825 2023-01-21 13:42:26.254948: step: 1952/529, loss: 0.0004104614199604839 2023-01-21 13:42:27.440270: step: 1956/529, loss: 3.0422212148550898e-05 2023-01-21 13:42:28.639475: step: 1960/529, loss: 0.0029163360595703125 2023-01-21 13:42:29.842386: step: 1964/529, loss: 0.0637754499912262 2023-01-21 13:42:31.013521: step: 1968/529, loss: 0.00018959045701194555 2023-01-21 13:42:32.139924: step: 1972/529, loss: 0.0035562515258789062 2023-01-21 13:42:33.295803: step: 1976/529, loss: 0.0045639993622899055 2023-01-21 13:42:34.498544: step: 1980/529, loss: 0.00011138916306663305 2023-01-21 13:42:35.680515: step: 1984/529, loss: 0.0001979827939067036 2023-01-21 13:42:36.885179: step: 1988/529, loss: 0.008693313226103783 2023-01-21 13:42:38.108350: step: 1992/529, loss: 0.029511641710996628 2023-01-21 13:42:39.316439: step: 1996/529, loss: 0.0002187728969147429 2023-01-21 13:42:40.496753: step: 2000/529, loss: 0.0008719444158487022 2023-01-21 13:42:41.709702: step: 2004/529, loss: -1.544952465337701e-05 2023-01-21 13:42:42.932064: step: 2008/529, loss: 0.002560138702392578 2023-01-21 13:42:44.125299: step: 2012/529, loss: 0.0032976153306663036 2023-01-21 13:42:45.345398: step: 2016/529, loss: 2.403259350103326e-05 2023-01-21 13:42:46.514201: step: 2020/529, loss: 0.025091171264648438 2023-01-21 13:42:47.718828: step: 2024/529, loss: -1.2016295841021929e-05 2023-01-21 13:42:48.907989: step: 2028/529, loss: -1.6117095583467744e-05 2023-01-21 13:42:50.112735: step: 2032/529, loss: 0.000480461138067767 2023-01-21 13:42:51.300971: step: 2036/529, loss: 0.003774297423660755 2023-01-21 13:42:52.484425: step: 2040/529, loss: 0.0020401000510901213 2023-01-21 13:42:53.660431: step: 2044/529, loss: 0.0016375541454181075 2023-01-21 13:42:54.830235: step: 2048/529, loss: 3.318786548334174e-05 2023-01-21 13:42:56.013676: step: 2052/529, loss: 0.002202844712883234 2023-01-21 13:42:57.156997: step: 2056/529, loss: 0.018551349639892578 2023-01-21 13:42:58.358609: step: 2060/529, loss: 0.03924274817109108 2023-01-21 13:42:59.526434: step: 2064/529, loss: 0.0285168644040823 2023-01-21 13:43:00.691679: step: 2068/529, loss: 0.0011410713195800781 2023-01-21 13:43:01.890376: step: 2072/529, loss: 9.841918654274195e-05 2023-01-21 13:43:03.093598: step: 2076/529, loss: 0.00019073484872933477 2023-01-21 13:43:04.289856: step: 2080/529, loss: -5.626678102998994e-06 2023-01-21 13:43:05.473906: step: 2084/529, loss: 0.00012016297114314511 2023-01-21 13:43:06.680056: step: 2088/529, loss: 0.005490493960678577 2023-01-21 13:43:07.893444: step: 2092/529, loss: 2.784729076665826e-05 2023-01-21 13:43:09.045736: step: 2096/529, loss: 5.435943421616685e-06 2023-01-21 13:43:10.246963: step: 2100/529, loss: 0.0028486251831054688 2023-01-21 13:43:11.407934: step: 2104/529, loss: 0.0009204864618368447 2023-01-21 13:43:12.610004: step: 2108/529, loss: 0.007889271713793278 2023-01-21 13:43:13.794388: step: 2112/529, loss: 0.00013446807861328125 2023-01-21 13:43:14.964650: step: 2116/529, loss: 9.040832810569555e-05 ================================================== Loss: 0.018 -------------------- Dev: {'event': {'p': 0.6095744680851064, 'r': 0.762982689747004, 'f1': 0.6777054997043169}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Test: {'event': {'p': 0.6495098039215687, 'r': 0.7938885560215698, 'f1': 0.7144782960366675}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Chinese: {'event': {'p': 0.625, 'r': 0.8333333333333334, 'f1': 0.7142857142857143}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Korean: {'event': {'p': 0.6181818181818182, 'r': 0.5396825396825397, 'f1': 0.576271186440678}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Russian: {'event': {'p': 0.4523809523809524, 'r': 0.5277777777777778, 'f1': 0.4871794871794871}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 13:43:57.841835: step: 4/529, loss: 2.19345088225964e-06 2023-01-21 13:43:59.022817: step: 8/529, loss: 0.00020184516324661672 2023-01-21 13:44:00.235428: step: 12/529, loss: 6.198883056640625e-06 2023-01-21 13:44:01.414886: step: 16/529, loss: 0.0007835388532839715 2023-01-21 13:44:02.680278: step: 20/529, loss: 0.044389914721250534 2023-01-21 13:44:03.864236: step: 24/529, loss: 0.039454080164432526 2023-01-21 13:44:05.054289: step: 28/529, loss: 0.00010643005225574598 2023-01-21 13:44:06.299301: step: 32/529, loss: 0.00015716553025413305 2023-01-21 13:44:07.527625: step: 36/529, loss: 0.022363662719726562 2023-01-21 13:44:08.689982: step: 40/529, loss: 0.005863761994987726 2023-01-21 13:44:09.862643: step: 44/529, loss: 0.00014147759065963328 2023-01-21 13:44:11.052693: step: 48/529, loss: 0.08831176906824112 2023-01-21 13:44:12.242669: step: 52/529, loss: 3.814697265625e-05 2023-01-21 13:44:13.440105: step: 56/529, loss: 0.07169074565172195 2023-01-21 13:44:14.630092: step: 60/529, loss: 0.001692104386165738 2023-01-21 13:44:15.792104: step: 64/529, loss: 6.389617919921875e-05 2023-01-21 13:44:17.005629: step: 68/529, loss: 4.57763671875e-05 2023-01-21 13:44:18.132028: step: 72/529, loss: 0.0007342338794842362 2023-01-21 13:44:19.334871: step: 76/529, loss: 0.006405449006706476 2023-01-21 13:44:20.511760: step: 80/529, loss: 0.0009681701776571572 2023-01-21 13:44:21.706551: step: 84/529, loss: 0.04061546549201012 2023-01-21 13:44:22.871952: step: 88/529, loss: 0.0011108398903161287 2023-01-21 13:44:24.042798: step: 92/529, loss: 0.0007769108051434159 2023-01-21 13:44:25.241549: step: 96/529, loss: 0.0421789214015007 2023-01-21 13:44:26.392736: step: 100/529, loss: 0.00013875961303710938 2023-01-21 13:44:27.646043: step: 104/529, loss: 2.0980833141948096e-06 2023-01-21 13:44:28.854880: step: 108/529, loss: 5.0163271225756034e-05 2023-01-21 13:44:30.051637: step: 112/529, loss: 0.0036340714432299137 2023-01-21 13:44:31.214665: step: 116/529, loss: 1.7547608877066523e-05 2023-01-21 13:44:32.427415: step: 120/529, loss: 0.00015993117995094508 2023-01-21 13:44:33.603057: step: 124/529, loss: 0.011015511117875576 2023-01-21 13:44:34.810561: step: 128/529, loss: 0.000583648681640625 2023-01-21 13:44:35.956118: step: 132/529, loss: 3.045797348022461e-05 2023-01-21 13:44:37.123519: step: 136/529, loss: 0.0003021717129740864 2023-01-21 13:44:38.325271: step: 140/529, loss: 0.0018939971923828125 2023-01-21 13:44:39.522068: step: 144/529, loss: 2.365112231927924e-05 2023-01-21 13:44:40.666465: step: 148/529, loss: 1.6450881958007812e-05 2023-01-21 13:44:41.848311: step: 152/529, loss: 5.507469177246094e-05 2023-01-21 13:44:43.018644: step: 156/529, loss: 0.004967355635017157 2023-01-21 13:44:44.263601: step: 160/529, loss: 1.7166138377433526e-06 2023-01-21 13:44:45.438582: step: 164/529, loss: 0.00013780595327261835 2023-01-21 13:44:46.563831: step: 168/529, loss: 0.0017105103470385075 2023-01-21 13:44:47.754719: step: 172/529, loss: 0.015952683985233307 2023-01-21 13:44:48.932488: step: 176/529, loss: 0.002684688428416848 2023-01-21 13:44:50.106766: step: 180/529, loss: 5.092620995128527e-05 2023-01-21 13:44:51.299791: step: 184/529, loss: 3.1280520488508046e-05 2023-01-21 13:44:52.468619: step: 188/529, loss: 0.03620586544275284 2023-01-21 13:44:53.644656: step: 192/529, loss: 1.6307831174344756e-05 2023-01-21 13:44:54.822189: step: 196/529, loss: 3.2424929941043956e-06 2023-01-21 13:44:55.964970: step: 200/529, loss: 0.0006960869068279862 2023-01-21 13:44:57.101291: step: 204/529, loss: 0.00019054414588026702 2023-01-21 13:44:58.269208: step: 208/529, loss: 0.0035186768509447575 2023-01-21 13:44:59.447872: step: 212/529, loss: -2.021789623540826e-05 2023-01-21 13:45:00.676979: step: 216/529, loss: 0.0003067016659770161 2023-01-21 13:45:01.888642: step: 220/529, loss: 0.0003581047058105469 2023-01-21 13:45:03.096127: step: 224/529, loss: 0.010175704956054688 2023-01-21 13:45:04.301271: step: 228/529, loss: 0.00035562514676712453 2023-01-21 13:45:05.513190: step: 232/529, loss: 0.0033266067039221525 2023-01-21 13:45:06.701617: step: 236/529, loss: 0.02343006245791912 2023-01-21 13:45:07.883796: step: 240/529, loss: 0.00012750625319313258 2023-01-21 13:45:09.087594: step: 244/529, loss: 0.00013065338134765625 2023-01-21 13:45:10.283718: step: 248/529, loss: 0.000347137451171875 2023-01-21 13:45:11.481516: step: 252/529, loss: 0.00453033484518528 2023-01-21 13:45:12.665388: step: 256/529, loss: 0.00041685104952193797 2023-01-21 13:45:13.835162: step: 260/529, loss: 0.0001924514799611643 2023-01-21 13:45:15.014966: step: 264/529, loss: 0.00046253204345703125 2023-01-21 13:45:16.179592: step: 268/529, loss: 3.24249267578125e-05 2023-01-21 13:45:17.380132: step: 272/529, loss: 0.0029591561760753393 2023-01-21 13:45:18.564920: step: 276/529, loss: 5.5313107623078395e-06 2023-01-21 13:45:19.726855: step: 280/529, loss: 9.34600848268019e-06 2023-01-21 13:45:20.957768: step: 284/529, loss: 0.00013322829909157008 2023-01-21 13:45:22.102763: step: 288/529, loss: 0.0007555008050985634 2023-01-21 13:45:23.299849: step: 292/529, loss: 0.005958748050034046 2023-01-21 13:45:24.450672: step: 296/529, loss: 0.0029787064995616674 2023-01-21 13:45:25.658842: step: 300/529, loss: 0.000217437744140625 2023-01-21 13:45:26.852569: step: 304/529, loss: 2.765655517578125e-05 2023-01-21 13:45:28.027088: step: 308/529, loss: 0.018631553277373314 2023-01-21 13:45:29.193942: step: 312/529, loss: 0.0018231152789667249 2023-01-21 13:45:30.420615: step: 316/529, loss: 0.0012628554832190275 2023-01-21 13:45:31.635397: step: 320/529, loss: 0.0006187438848428428 2023-01-21 13:45:32.827239: step: 324/529, loss: 0.0005594253307208419 2023-01-21 13:45:33.984016: step: 328/529, loss: 0.00259742746129632 2023-01-21 13:45:35.157328: step: 332/529, loss: 0.0005500793922692537 2023-01-21 13:45:36.358547: step: 336/529, loss: 0.0007852554554119706 2023-01-21 13:45:37.563641: step: 340/529, loss: 7.62939453125e-06 2023-01-21 13:45:38.743753: step: 344/529, loss: 0.002188110491260886 2023-01-21 13:45:39.959771: step: 348/529, loss: 0.0006380081758834422 2023-01-21 13:45:41.165037: step: 352/529, loss: 0.0019458772148936987 2023-01-21 13:45:42.337846: step: 356/529, loss: 2.86102294921875e-06 2023-01-21 13:45:43.499440: step: 360/529, loss: 0.0468907356262207 2023-01-21 13:45:44.677957: step: 364/529, loss: 0.0018169403774663806 2023-01-21 13:45:45.851111: step: 368/529, loss: 0.0011387824779376388 2023-01-21 13:45:47.039884: step: 372/529, loss: 0.0018321990501135588 2023-01-21 13:45:48.207266: step: 376/529, loss: 0.0001605987490620464 2023-01-21 13:45:49.376971: step: 380/529, loss: 0.0016286849277094007 2023-01-21 13:45:50.535492: step: 384/529, loss: 0.0016176224453374743 2023-01-21 13:45:51.741804: step: 388/529, loss: 0.004326724912971258 2023-01-21 13:45:52.920869: step: 392/529, loss: 0.04388561472296715 2023-01-21 13:45:54.181536: step: 396/529, loss: 1.5735626220703125e-05 2023-01-21 13:45:55.404731: step: 400/529, loss: 0.0029001235961914062 2023-01-21 13:45:56.637291: step: 404/529, loss: 0.016232872381806374 2023-01-21 13:45:57.794901: step: 408/529, loss: 0.0008089065668173134 2023-01-21 13:45:58.981822: step: 412/529, loss: 8.707046072231606e-05 2023-01-21 13:46:00.138484: step: 416/529, loss: 0.0002419471857137978 2023-01-21 13:46:01.329079: step: 420/529, loss: 0.0007030487176962197 2023-01-21 13:46:02.537252: step: 424/529, loss: 0.03721580654382706 2023-01-21 13:46:03.763150: step: 428/529, loss: 0.0026134492363780737 2023-01-21 13:46:04.962919: step: 432/529, loss: 0.025157354772090912 2023-01-21 13:46:06.178765: step: 436/529, loss: 0.0016920090420171618 2023-01-21 13:46:07.392710: step: 440/529, loss: 0.0004112243477720767 2023-01-21 13:46:08.618422: step: 444/529, loss: 0.052002910524606705 2023-01-21 13:46:09.778314: step: 448/529, loss: 0.0019020081963390112 2023-01-21 13:46:11.006389: step: 452/529, loss: 0.014834118075668812 2023-01-21 13:46:12.197096: step: 456/529, loss: 3.967285010730848e-05 2023-01-21 13:46:13.395448: step: 460/529, loss: 4.2724612285383046e-05 2023-01-21 13:46:14.558679: step: 464/529, loss: 0.00010051727440441027 2023-01-21 13:46:15.776279: step: 468/529, loss: 0.011730194091796875 2023-01-21 13:46:16.933216: step: 472/529, loss: 6.656646291958168e-05 2023-01-21 13:46:18.137109: step: 476/529, loss: 1.33514404296875e-05 2023-01-21 13:46:19.314167: step: 480/529, loss: 1.0585784366412554e-05 2023-01-21 13:46:20.440871: step: 484/529, loss: 2.47955313170678e-06 2023-01-21 13:46:21.655376: step: 488/529, loss: 0.010639572516083717 2023-01-21 13:46:22.844455: step: 492/529, loss: 0.02635822258889675 2023-01-21 13:46:24.037871: step: 496/529, loss: 1.773834264895413e-05 2023-01-21 13:46:25.206800: step: 500/529, loss: 0.0062191965989768505 2023-01-21 13:46:26.356438: step: 504/529, loss: 0.0003980636829510331 2023-01-21 13:46:27.550673: step: 508/529, loss: 0.0029202462173998356 2023-01-21 13:46:28.742012: step: 512/529, loss: 0.0009311676258221269 2023-01-21 13:46:29.929174: step: 516/529, loss: 5.636215428239666e-05 2023-01-21 13:46:31.139201: step: 520/529, loss: 0.04598722606897354 2023-01-21 13:46:32.323306: step: 524/529, loss: 7.295608611457283e-06 2023-01-21 13:46:33.521664: step: 528/529, loss: 0.011223316192626953 2023-01-21 13:46:34.725889: step: 532/529, loss: 0.003146362490952015 2023-01-21 13:46:35.948233: step: 536/529, loss: 0.00026025774423033 2023-01-21 13:46:37.104388: step: 540/529, loss: 0.00035305024357512593 2023-01-21 13:46:38.299307: step: 544/529, loss: 0.00010547638521529734 2023-01-21 13:46:39.491369: step: 548/529, loss: 2.3078917365637608e-05 2023-01-21 13:46:40.662434: step: 552/529, loss: -1.3351439065445447e-06 2023-01-21 13:46:41.853025: step: 556/529, loss: 0.0007143020629882812 2023-01-21 13:46:43.045273: step: 560/529, loss: 0.0004932403680868447 2023-01-21 13:46:44.217752: step: 564/529, loss: 5.855560448253527e-05 2023-01-21 13:46:45.430393: step: 568/529, loss: 0.008263015188276768 2023-01-21 13:46:46.634609: step: 572/529, loss: 0.0734064131975174 2023-01-21 13:46:47.818066: step: 576/529, loss: 0.0005625725025311112 2023-01-21 13:46:49.050594: step: 580/529, loss: 3.757476952159777e-05 2023-01-21 13:46:50.212334: step: 584/529, loss: 0.0033500671852380037 2023-01-21 13:46:51.398295: step: 588/529, loss: 0.0004800796741619706 2023-01-21 13:46:52.591306: step: 592/529, loss: 0.005219078157097101 2023-01-21 13:46:53.769802: step: 596/529, loss: 0.0011716842418536544 2023-01-21 13:46:54.954725: step: 600/529, loss: 0.0002229690580861643 2023-01-21 13:46:56.160523: step: 604/529, loss: 0.00015525818162132055 2023-01-21 13:46:57.328109: step: 608/529, loss: 1.5258789289873675e-06 2023-01-21 13:46:58.511887: step: 612/529, loss: 0.0039768218994140625 2023-01-21 13:46:59.693677: step: 616/529, loss: 0.00028896331787109375 2023-01-21 13:47:00.852204: step: 620/529, loss: 0.0003613472217693925 2023-01-21 13:47:02.088010: step: 624/529, loss: 0.00010528565326239914 2023-01-21 13:47:03.249579: step: 628/529, loss: 0.018291091546416283 2023-01-21 13:47:04.391277: step: 632/529, loss: 0.005200195126235485 2023-01-21 13:47:05.570640: step: 636/529, loss: 0.001433086465112865 2023-01-21 13:47:06.753185: step: 640/529, loss: 0.0013350486988201737 2023-01-21 13:47:07.958946: step: 644/529, loss: 0.031363680958747864 2023-01-21 13:47:09.121968: step: 648/529, loss: 4.196166628389619e-06 2023-01-21 13:47:10.281619: step: 652/529, loss: 0.0012488365173339844 2023-01-21 13:47:11.443957: step: 656/529, loss: 3.719329833984375e-05 2023-01-21 13:47:12.650086: step: 660/529, loss: 0.005908107850700617 2023-01-21 13:47:13.840851: step: 664/529, loss: 0.00012736320786643773 2023-01-21 13:47:15.018171: step: 668/529, loss: 1.0204315913142636e-05 2023-01-21 13:47:16.204871: step: 672/529, loss: 0.00017976762319449335 2023-01-21 13:47:17.377157: step: 676/529, loss: 0.024214552715420723 2023-01-21 13:47:18.594919: step: 680/529, loss: 0.008968067355453968 2023-01-21 13:47:19.844939: step: 684/529, loss: 0.0034555436577647924 2023-01-21 13:47:21.027599: step: 688/529, loss: 0.012199974618852139 2023-01-21 13:47:22.254670: step: 692/529, loss: 0.0038137435913085938 2023-01-21 13:47:23.510150: step: 696/529, loss: 0.000660324061755091 2023-01-21 13:47:24.683907: step: 700/529, loss: 0.00025405886117368937 2023-01-21 13:47:25.885134: step: 704/529, loss: 0.3003028929233551 2023-01-21 13:47:27.080086: step: 708/529, loss: 1.5735626220703125e-05 2023-01-21 13:47:28.255700: step: 712/529, loss: -6.9618222369172145e-06 2023-01-21 13:47:29.460795: step: 716/529, loss: 1.955032530531753e-05 2023-01-21 13:47:30.657256: step: 720/529, loss: 4.38690176451928e-06 2023-01-21 13:47:31.814639: step: 724/529, loss: 6.675720669591101e-06 2023-01-21 13:47:32.978890: step: 728/529, loss: 0.013900566846132278 2023-01-21 13:47:34.149546: step: 732/529, loss: 0.00011157989501953125 2023-01-21 13:47:35.308107: step: 736/529, loss: 0.005905533209443092 2023-01-21 13:47:36.493153: step: 740/529, loss: 1.2016297660011332e-05 2023-01-21 13:47:37.704698: step: 744/529, loss: 5.3739546274300665e-05 2023-01-21 13:47:38.889912: step: 748/529, loss: 0.11075859516859055 2023-01-21 13:47:40.070512: step: 752/529, loss: 9.603500802768394e-05 2023-01-21 13:47:41.217518: step: 756/529, loss: 0.0001447677641408518 2023-01-21 13:47:42.406480: step: 760/529, loss: 9.059906005859375e-06 2023-01-21 13:47:43.603088: step: 764/529, loss: 0.013219070620834827 2023-01-21 13:47:44.788379: step: 768/529, loss: -9.34600848268019e-06 2023-01-21 13:47:45.985032: step: 772/529, loss: 0.0007427215459756553 2023-01-21 13:47:47.171134: step: 776/529, loss: 0.009543133899569511 2023-01-21 13:47:48.383406: step: 780/529, loss: 8.897781663108617e-05 2023-01-21 13:47:49.582012: step: 784/529, loss: 0.011541747488081455 2023-01-21 13:47:50.735487: step: 788/529, loss: 0.0004888534313067794 2023-01-21 13:47:51.965117: step: 792/529, loss: 0.0622706413269043 2023-01-21 13:47:53.140856: step: 796/529, loss: 1.4114379155216739e-05 2023-01-21 13:47:54.369850: step: 800/529, loss: 0.0014701844193041325 2023-01-21 13:47:55.531544: step: 804/529, loss: 0.0001698732521617785 2023-01-21 13:47:56.721520: step: 808/529, loss: 0.003596401307731867 2023-01-21 13:47:57.935100: step: 812/529, loss: 0.0003406524774618447 2023-01-21 13:47:59.114196: step: 816/529, loss: 0.017597390338778496 2023-01-21 13:48:00.255431: step: 820/529, loss: 1.182556115963962e-05 2023-01-21 13:48:01.440302: step: 824/529, loss: 0.00218715681694448 2023-01-21 13:48:02.598706: step: 828/529, loss: 0.029137611389160156 2023-01-21 13:48:03.851617: step: 832/529, loss: 0.02660808525979519 2023-01-21 13:48:05.042354: step: 836/529, loss: 0.017335321754217148 2023-01-21 13:48:06.240711: step: 840/529, loss: 0.00018177033052779734 2023-01-21 13:48:07.428391: step: 844/529, loss: 0.0019325255416333675 2023-01-21 13:48:08.628423: step: 848/529, loss: 0.00020437240891624242 2023-01-21 13:48:09.820487: step: 852/529, loss: 0.005922890268266201 2023-01-21 13:48:11.032011: step: 856/529, loss: 0.0001552104949951172 2023-01-21 13:48:12.221985: step: 860/529, loss: 0.01691455766558647 2023-01-21 13:48:13.410394: step: 864/529, loss: 0.0006818771362304688 2023-01-21 13:48:14.543795: step: 868/529, loss: -4.291534423828125e-06 2023-01-21 13:48:15.790690: step: 872/529, loss: 0.0014144896995276213 2023-01-21 13:48:16.976589: step: 876/529, loss: 1.3446808225126006e-05 2023-01-21 13:48:18.208636: step: 880/529, loss: 0.00864410400390625 2023-01-21 13:48:19.443686: step: 884/529, loss: 0.01627349853515625 2023-01-21 13:48:20.627357: step: 888/529, loss: 0.00018825530423782766 2023-01-21 13:48:21.871872: step: 892/529, loss: 7.820130122127011e-06 2023-01-21 13:48:23.042401: step: 896/529, loss: 0.7037311792373657 2023-01-21 13:48:24.211209: step: 900/529, loss: 0.0022602081298828125 2023-01-21 13:48:25.452071: step: 904/529, loss: 0.05023917928338051 2023-01-21 13:48:26.644123: step: 908/529, loss: 0.00010004043724620715 2023-01-21 13:48:27.812629: step: 912/529, loss: 0.0032807348761707544 2023-01-21 13:48:29.018906: step: 916/529, loss: 0.01593475230038166 2023-01-21 13:48:30.239020: step: 920/529, loss: 0.00019874573627021164 2023-01-21 13:48:31.476346: step: 924/529, loss: 0.017996979877352715 2023-01-21 13:48:32.682446: step: 928/529, loss: 0.005619811825454235 2023-01-21 13:48:33.899789: step: 932/529, loss: 4.720688230008818e-05 2023-01-21 13:48:35.101103: step: 936/529, loss: 3.43322744811303e-06 2023-01-21 13:48:36.312362: step: 940/529, loss: 6.885529001010582e-05 2023-01-21 13:48:37.439355: step: 944/529, loss: 0.0009704590192995965 2023-01-21 13:48:38.673678: step: 948/529, loss: 0.0012825012672692537 2023-01-21 13:48:39.875160: step: 952/529, loss: 0.0008444786071777344 2023-01-21 13:48:41.076621: step: 956/529, loss: 0.007120418827980757 2023-01-21 13:48:42.261566: step: 960/529, loss: 0.0002723693905863911 2023-01-21 13:48:43.448502: step: 964/529, loss: 0.00042386056156829 2023-01-21 13:48:44.660349: step: 968/529, loss: 1.52587890625e-05 2023-01-21 13:48:45.839881: step: 972/529, loss: 0.00051288609392941 2023-01-21 13:48:47.041208: step: 976/529, loss: 0.00043234828626737 2023-01-21 13:48:48.246338: step: 980/529, loss: 0.0001903533993754536 2023-01-21 13:48:49.461091: step: 984/529, loss: 0.005395126529037952 2023-01-21 13:48:50.634654: step: 988/529, loss: 0.020402908325195312 2023-01-21 13:48:51.823932: step: 992/529, loss: 1.7261505490751006e-05 2023-01-21 13:48:53.043091: step: 996/529, loss: 0.0001523971586721018 2023-01-21 13:48:54.261637: step: 1000/529, loss: 3.089904930675402e-05 2023-01-21 13:48:55.491664: step: 1004/529, loss: 0.00013556479825638235 2023-01-21 13:48:56.677146: step: 1008/529, loss: 0.00023975371732376516 2023-01-21 13:48:57.943265: step: 1012/529, loss: 0.0002521514834370464 2023-01-21 13:48:59.144859: step: 1016/529, loss: 0.05911846086382866 2023-01-21 13:49:00.366716: step: 1020/529, loss: 2.0027162008773303e-06 2023-01-21 13:49:01.576223: step: 1024/529, loss: 5.7220458984375e-06 2023-01-21 13:49:02.778594: step: 1028/529, loss: 0.013902951031923294 2023-01-21 13:49:04.026702: step: 1032/529, loss: 0.0017786026000976562 2023-01-21 13:49:05.216665: step: 1036/529, loss: 0.0005986213800497353 2023-01-21 13:49:06.431078: step: 1040/529, loss: 9.130835678661242e-05 2023-01-21 13:49:07.670653: step: 1044/529, loss: 0.04312162846326828 2023-01-21 13:49:08.870084: step: 1048/529, loss: 0.002151107881218195 2023-01-21 13:49:10.116628: step: 1052/529, loss: 0.027092933654785156 2023-01-21 13:49:11.297172: step: 1056/529, loss: 7.82012921263231e-06 2023-01-21 13:49:12.502058: step: 1060/529, loss: 0.019725419580936432 2023-01-21 13:49:13.706504: step: 1064/529, loss: 0.00010414123971713707 2023-01-21 13:49:14.907535: step: 1068/529, loss: 0.005253935232758522 2023-01-21 13:49:16.152144: step: 1072/529, loss: 0.43947526812553406 2023-01-21 13:49:17.359931: step: 1076/529, loss: 0.004548931494355202 2023-01-21 13:49:18.547132: step: 1080/529, loss: 0.059807490557432175 2023-01-21 13:49:19.789794: step: 1084/529, loss: 0.004552745725959539 2023-01-21 13:49:20.967075: step: 1088/529, loss: 0.010261917486786842 2023-01-21 13:49:22.186493: step: 1092/529, loss: 0.0020591735374182463 2023-01-21 13:49:23.352175: step: 1096/529, loss: 0.06018352508544922 2023-01-21 13:49:24.519993: step: 1100/529, loss: 0.02090167999267578 2023-01-21 13:49:25.768091: step: 1104/529, loss: 0.2931070327758789 2023-01-21 13:49:26.951012: step: 1108/529, loss: -4.95910626341356e-06 2023-01-21 13:49:28.183997: step: 1112/529, loss: 0.00036659243050962687 2023-01-21 13:49:29.404675: step: 1116/529, loss: 0.00027647020760923624 2023-01-21 13:49:30.602662: step: 1120/529, loss: 6.29425048828125e-05 2023-01-21 13:49:31.799494: step: 1124/529, loss: 3.814697265625e-06 2023-01-21 13:49:32.993175: step: 1128/529, loss: 0.003273487091064453 2023-01-21 13:49:34.251352: step: 1132/529, loss: 0.00039396286592818797 2023-01-21 13:49:35.477411: step: 1136/529, loss: 0.000423431396484375 2023-01-21 13:49:36.657747: step: 1140/529, loss: 0.00012474060349632055 2023-01-21 13:49:37.849171: step: 1144/529, loss: 0.008915329352021217 2023-01-21 13:49:39.137373: step: 1148/529, loss: 3.185272362316027e-05 2023-01-21 13:49:40.330816: step: 1152/529, loss: 0.00014762878709007055 2023-01-21 13:49:41.505682: step: 1156/529, loss: 0.031037140637636185 2023-01-21 13:49:42.700514: step: 1160/529, loss: 0.005969238001853228 2023-01-21 13:49:43.896931: step: 1164/529, loss: 0.0012039184803143144 2023-01-21 13:49:45.105199: step: 1168/529, loss: 6.88552827341482e-05 2023-01-21 13:49:46.285951: step: 1172/529, loss: 0.004775619599968195 2023-01-21 13:49:47.458356: step: 1176/529, loss: 0.0002693176211323589 2023-01-21 13:49:48.645424: step: 1180/529, loss: -1.33514404296875e-05 2023-01-21 13:49:49.870402: step: 1184/529, loss: 0.02767951600253582 2023-01-21 13:49:51.084537: step: 1188/529, loss: 0.0003884315665345639 2023-01-21 13:49:52.293131: step: 1192/529, loss: 1.5258789289873675e-06 2023-01-21 13:49:53.503548: step: 1196/529, loss: 1.0871887752728071e-05 2023-01-21 13:49:54.659492: step: 1200/529, loss: 8.678435733600054e-06 2023-01-21 13:49:55.841969: step: 1204/529, loss: 0.01877574808895588 2023-01-21 13:49:57.051543: step: 1208/529, loss: 0.00018076896958518773 2023-01-21 13:49:58.294243: step: 1212/529, loss: 0.00426936149597168 2023-01-21 13:49:59.453458: step: 1216/529, loss: 9.52720656641759e-05 2023-01-21 13:50:00.638069: step: 1220/529, loss: 0.011672401800751686 2023-01-21 13:50:01.805061: step: 1224/529, loss: 2.9563905172835803e-06 2023-01-21 13:50:02.997846: step: 1228/529, loss: 3.290176209702622e-06 2023-01-21 13:50:04.188616: step: 1232/529, loss: 3.814697265625e-06 2023-01-21 13:50:05.410345: step: 1236/529, loss: 0.0011616707779467106 2023-01-21 13:50:06.609041: step: 1240/529, loss: 0.006520652677863836 2023-01-21 13:50:07.810476: step: 1244/529, loss: 2.014822006225586 2023-01-21 13:50:09.030537: step: 1248/529, loss: 0.0009389877086505294 2023-01-21 13:50:10.281018: step: 1252/529, loss: 0.0013923646183684468 2023-01-21 13:50:11.491836: step: 1256/529, loss: 0.011000634171068668 2023-01-21 13:50:12.676851: step: 1260/529, loss: 0.00026683806208893657 2023-01-21 13:50:13.879807: step: 1264/529, loss: 0.00042724609375 2023-01-21 13:50:15.105948: step: 1268/529, loss: 0.0004532337188720703 2023-01-21 13:50:16.296680: step: 1272/529, loss: 0.037917040288448334 2023-01-21 13:50:17.482580: step: 1276/529, loss: 0.002945518586784601 2023-01-21 13:50:18.713617: step: 1280/529, loss: 6.513595872092992e-05 2023-01-21 13:50:19.904476: step: 1284/529, loss: 0.00044422148494049907 2023-01-21 13:50:21.100210: step: 1288/529, loss: 6.732940528308973e-05 2023-01-21 13:50:22.296952: step: 1292/529, loss: 0.010332870297133923 2023-01-21 13:50:23.510161: step: 1296/529, loss: 3.24249267578125e-05 2023-01-21 13:50:24.684312: step: 1300/529, loss: 0.0005731582641601562 2023-01-21 13:50:25.854128: step: 1304/529, loss: 4.405975414556451e-05 2023-01-21 13:50:27.039269: step: 1308/529, loss: 0.009279823862016201 2023-01-21 13:50:28.230602: step: 1312/529, loss: 0.011157035827636719 2023-01-21 13:50:29.400328: step: 1316/529, loss: 0.0017451286548748612 2023-01-21 13:50:30.574920: step: 1320/529, loss: 0.0003120422479696572 2023-01-21 13:50:31.752118: step: 1324/529, loss: 0.0662609115242958 2023-01-21 13:50:32.925520: step: 1328/529, loss: 8.163452002918348e-05 2023-01-21 13:50:34.072188: step: 1332/529, loss: 0.0007122606039047241 2023-01-21 13:50:35.235776: step: 1336/529, loss: 0.00018515587726142257 2023-01-21 13:50:36.438464: step: 1340/529, loss: 0.00019779204740189016 2023-01-21 13:50:37.647043: step: 1344/529, loss: 3.204345557605848e-05 2023-01-21 13:50:38.858627: step: 1348/529, loss: 9.31739850784652e-05 2023-01-21 13:50:40.072894: step: 1352/529, loss: 0.017632009461522102 2023-01-21 13:50:41.275330: step: 1356/529, loss: 0.033524226397275925 2023-01-21 13:50:42.500301: step: 1360/529, loss: 0.001314353896304965 2023-01-21 13:50:43.671446: step: 1364/529, loss: 1.9931794668082148e-05 2023-01-21 13:50:44.896483: step: 1368/529, loss: 0.0006610870477743447 2023-01-21 13:50:46.060630: step: 1372/529, loss: 0.00023546218289993703 2023-01-21 13:50:47.225185: step: 1376/529, loss: 0.0006831169012002647 2023-01-21 13:50:48.447938: step: 1380/529, loss: 0.0016157150967046618 2023-01-21 13:50:49.624544: step: 1384/529, loss: 0.0009798050159588456 2023-01-21 13:50:50.812912: step: 1388/529, loss: 0.00011453629122115672 2023-01-21 13:50:52.039255: step: 1392/529, loss: 0.00032958982046693563 2023-01-21 13:50:53.194205: step: 1396/529, loss: 2.212524486822076e-05 2023-01-21 13:50:54.380807: step: 1400/529, loss: 0.12508058547973633 2023-01-21 13:50:55.561832: step: 1404/529, loss: 0.0006109238020144403 2023-01-21 13:50:56.774869: step: 1408/529, loss: 0.0005239486927166581 2023-01-21 13:50:57.958100: step: 1412/529, loss: 8.621215965831652e-05 2023-01-21 13:50:59.145606: step: 1416/529, loss: 0.001671648002229631 2023-01-21 13:51:00.354226: step: 1420/529, loss: 0.0018751145107671618 2023-01-21 13:51:01.587332: step: 1424/529, loss: 1.9168854123563506e-05 2023-01-21 13:51:02.798269: step: 1428/529, loss: 0.0021585465874522924 2023-01-21 13:51:03.979076: step: 1432/529, loss: 0.0243548396974802 2023-01-21 13:51:05.162419: step: 1436/529, loss: 0.019945433363318443 2023-01-21 13:51:06.401658: step: 1440/529, loss: -4.9591067181609105e-06 2023-01-21 13:51:07.648562: step: 1444/529, loss: 0.001189422677271068 2023-01-21 13:51:08.878145: step: 1448/529, loss: 0.010272311978042126 2023-01-21 13:51:10.054721: step: 1452/529, loss: 0.00023288727970793843 2023-01-21 13:51:11.285073: step: 1456/529, loss: 0.01434249896556139 2023-01-21 13:51:12.458711: step: 1460/529, loss: 0.0005619525909423828 2023-01-21 13:51:13.643695: step: 1464/529, loss: 0.0018398285610601306 2023-01-21 13:51:14.822805: step: 1468/529, loss: 0.0023834705352783203 2023-01-21 13:51:15.975331: step: 1472/529, loss: 0.00046749116154387593 2023-01-21 13:51:17.176692: step: 1476/529, loss: 0.00139198312535882 2023-01-21 13:51:18.382991: step: 1480/529, loss: 0.0017762182978913188 2023-01-21 13:51:19.597983: step: 1484/529, loss: 3.7604360580444336 2023-01-21 13:51:20.743056: step: 1488/529, loss: 0.00020956993103027344 2023-01-21 13:51:21.920712: step: 1492/529, loss: 5.569458153331652e-05 2023-01-21 13:51:23.153744: step: 1496/529, loss: 0.0012048721546307206 2023-01-21 13:51:24.329981: step: 1500/529, loss: 0.0003140449698548764 2023-01-21 13:51:25.521476: step: 1504/529, loss: 0.0001008987455861643 2023-01-21 13:51:26.705206: step: 1508/529, loss: -4.887580871582031e-06 2023-01-21 13:51:27.917763: step: 1512/529, loss: -2.6035308110294864e-05 2023-01-21 13:51:29.110368: step: 1516/529, loss: 0.0016532897716388106 2023-01-21 13:51:30.323398: step: 1520/529, loss: 0.009258843027055264 2023-01-21 13:51:31.541117: step: 1524/529, loss: 0.027378082275390625 2023-01-21 13:51:32.734629: step: 1528/529, loss: 0.12286939471960068 2023-01-21 13:51:33.980623: step: 1532/529, loss: 2.155303991457913e-05 2023-01-21 13:51:35.127211: step: 1536/529, loss: 0.00108509068377316 2023-01-21 13:51:36.322437: step: 1540/529, loss: -8.1300731835654e-06 2023-01-21 13:51:37.507606: step: 1544/529, loss: 5.168914867681451e-05 2023-01-21 13:51:38.685834: step: 1548/529, loss: 0.0027379037346690893 2023-01-21 13:51:39.890965: step: 1552/529, loss: 0.05477733537554741 2023-01-21 13:51:41.076694: step: 1556/529, loss: 2.261141777038574 2023-01-21 13:51:42.234740: step: 1560/529, loss: 6.10351571594947e-06 2023-01-21 13:51:43.409879: step: 1564/529, loss: 0.007138634100556374 2023-01-21 13:51:44.609976: step: 1568/529, loss: 0.004919338505715132 2023-01-21 13:51:45.799808: step: 1572/529, loss: 4.639625331037678e-05 2023-01-21 13:51:46.980966: step: 1576/529, loss: 3.24249276673072e-06 2023-01-21 13:51:48.145121: step: 1580/529, loss: 0.0007768631330691278 2023-01-21 13:51:49.306754: step: 1584/529, loss: 0.000610732997301966 2023-01-21 13:51:50.441013: step: 1588/529, loss: 0.0009853362571448088 2023-01-21 13:51:51.643872: step: 1592/529, loss: 0.004426336381584406 2023-01-21 13:51:52.886169: step: 1596/529, loss: 0.0009174346923828125 2023-01-21 13:51:54.075457: step: 1600/529, loss: 0.0124359130859375 2023-01-21 13:51:55.264801: step: 1604/529, loss: 0.02292346954345703 2023-01-21 13:51:56.472563: step: 1608/529, loss: 0.012574195861816406 2023-01-21 13:51:57.678232: step: 1612/529, loss: 0.0005159854772500694 2023-01-21 13:51:58.825041: step: 1616/529, loss: 0.0016510963905602694 2023-01-21 13:52:00.024602: step: 1620/529, loss: 0.0036033629439771175 2023-01-21 13:52:01.230259: step: 1624/529, loss: 2.899169885495212e-05 2023-01-21 13:52:02.425203: step: 1628/529, loss: 0.0020426749251782894 2023-01-21 13:52:03.622975: step: 1632/529, loss: 4.425048973644152e-05 2023-01-21 13:52:04.781000: step: 1636/529, loss: 0.0011596680851653218 2023-01-21 13:52:05.931407: step: 1640/529, loss: 2.2220612663659267e-05 2023-01-21 13:52:07.120792: step: 1644/529, loss: -1.0776519047794864e-05 2023-01-21 13:52:08.317669: step: 1648/529, loss: 0.025642015039920807 2023-01-21 13:52:09.520681: step: 1652/529, loss: 0.0025005340576171875 2023-01-21 13:52:10.730562: step: 1656/529, loss: 0.07654371112585068 2023-01-21 13:52:11.931507: step: 1660/529, loss: 0.18458519876003265 2023-01-21 13:52:13.110625: step: 1664/529, loss: 8.926392183639109e-05 2023-01-21 13:52:14.338956: step: 1668/529, loss: 0.009643793106079102 2023-01-21 13:52:15.493661: step: 1672/529, loss: 0.0016570091247558594 2023-01-21 13:52:16.683029: step: 1676/529, loss: 0.010727310553193092 2023-01-21 13:52:17.858638: step: 1680/529, loss: 2.237558328488376e-05 2023-01-21 13:52:19.037131: step: 1684/529, loss: 0.2558395564556122 2023-01-21 13:52:20.222946: step: 1688/529, loss: 0.07726965099573135 2023-01-21 13:52:21.447315: step: 1692/529, loss: 0.0008657455327920616 2023-01-21 13:52:22.606212: step: 1696/529, loss: 0.0005592346424236894 2023-01-21 13:52:23.839611: step: 1700/529, loss: 0.02435169368982315 2023-01-21 13:52:25.112148: step: 1704/529, loss: 0.006641578394919634 2023-01-21 13:52:26.304825: step: 1708/529, loss: 0.0005907058948650956 2023-01-21 13:52:27.455499: step: 1712/529, loss: 0.002627659123390913 2023-01-21 13:52:28.634247: step: 1716/529, loss: 0.012280846014618874 2023-01-21 13:52:29.819440: step: 1720/529, loss: 1.6117095583467744e-05 2023-01-21 13:52:30.999730: step: 1724/529, loss: 0.047978781163692474 2023-01-21 13:52:32.225934: step: 1728/529, loss: 0.0007720947614870965 2023-01-21 13:52:33.415186: step: 1732/529, loss: -1.5258788153005298e-06 2023-01-21 13:52:34.589030: step: 1736/529, loss: 2.0885467165498994e-05 2023-01-21 13:52:35.857324: step: 1740/529, loss: 0.01361169759184122 2023-01-21 13:52:37.067310: step: 1744/529, loss: 9.212493750965223e-05 2023-01-21 13:52:38.262919: step: 1748/529, loss: 1.5211105164780747e-05 2023-01-21 13:52:39.480456: step: 1752/529, loss: 1.150965726992581e-05 2023-01-21 13:52:40.669724: step: 1756/529, loss: 0.002271461533382535 2023-01-21 13:52:41.857822: step: 1760/529, loss: 0.01685047149658203 2023-01-21 13:52:43.094732: step: 1764/529, loss: 0.0018886567559093237 2023-01-21 13:52:44.321016: step: 1768/529, loss: 0.015965841710567474 2023-01-21 13:52:45.532944: step: 1772/529, loss: 0.002438068389892578 2023-01-21 13:52:46.698751: step: 1776/529, loss: 0.0003922939649783075 2023-01-21 13:52:47.890029: step: 1780/529, loss: 0.0001739501894917339 2023-01-21 13:52:49.064226: step: 1784/529, loss: 0.00033779145451262593 2023-01-21 13:52:50.262398: step: 1788/529, loss: 0.0017832756275311112 2023-01-21 13:52:51.449791: step: 1792/529, loss: 0.00011749268014682457 2023-01-21 13:52:52.640124: step: 1796/529, loss: 4.730224463855848e-05 2023-01-21 13:52:53.845886: step: 1800/529, loss: 0.0010387421352788806 2023-01-21 13:52:55.027811: step: 1804/529, loss: 0.009067917242646217 2023-01-21 13:52:56.220417: step: 1808/529, loss: 0.0007368087535724044 2023-01-21 13:52:57.440156: step: 1812/529, loss: 6.289482553256676e-05 2023-01-21 13:52:58.640954: step: 1816/529, loss: 5.14984139954322e-06 2023-01-21 13:52:59.820967: step: 1820/529, loss: 0.00011310577974654734 2023-01-21 13:53:00.981078: step: 1824/529, loss: 1.773834264895413e-05 2023-01-21 13:53:02.200131: step: 1828/529, loss: 0.06255068629980087 2023-01-21 13:53:03.373304: step: 1832/529, loss: 0.0003951072576455772 2023-01-21 13:53:04.562732: step: 1836/529, loss: 0.0851929634809494 2023-01-21 13:53:05.759125: step: 1840/529, loss: 0.002176952548325062 2023-01-21 13:53:06.942100: step: 1844/529, loss: 0.0006973266135901213 2023-01-21 13:53:08.149972: step: 1848/529, loss: 0.011502456851303577 2023-01-21 13:53:09.342872: step: 1852/529, loss: 0.007765770424157381 2023-01-21 13:53:10.536036: step: 1856/529, loss: -6.103515261202119e-06 2023-01-21 13:53:11.717196: step: 1860/529, loss: 0.00012321471876930445 2023-01-21 13:53:12.889205: step: 1864/529, loss: 9.059906005859375e-06 2023-01-21 13:53:14.075475: step: 1868/529, loss: 0.015971088781952858 2023-01-21 13:53:15.261270: step: 1872/529, loss: 0.0037128927651792765 2023-01-21 13:53:16.465575: step: 1876/529, loss: 0.0021805763244628906 2023-01-21 13:53:17.667003: step: 1880/529, loss: 0.015220356173813343 2023-01-21 13:53:18.848887: step: 1884/529, loss: 0.00017452239990234375 2023-01-21 13:53:20.030374: step: 1888/529, loss: 0.0010402679909020662 2023-01-21 13:53:21.237577: step: 1892/529, loss: 0.0003410339413676411 2023-01-21 13:53:22.396330: step: 1896/529, loss: -2.9087066195643274e-06 2023-01-21 13:53:23.570908: step: 1900/529, loss: 0.010410022921860218 2023-01-21 13:53:24.746756: step: 1904/529, loss: 0.00027875902014784515 2023-01-21 13:53:25.911236: step: 1908/529, loss: 0.00042304996168240905 2023-01-21 13:53:27.120315: step: 1912/529, loss: 0.0062160491943359375 2023-01-21 13:53:28.348077: step: 1916/529, loss: 9.79900432866998e-05 2023-01-21 13:53:29.545774: step: 1920/529, loss: 8.39233416627394e-06 2023-01-21 13:53:30.698951: step: 1924/529, loss: 3.166198803228326e-05 2023-01-21 13:53:31.912217: step: 1928/529, loss: 0.027146339416503906 2023-01-21 13:53:33.131360: step: 1932/529, loss: 0.004527282901108265 2023-01-21 13:53:34.331127: step: 1936/529, loss: 3.8909915019758046e-05 2023-01-21 13:53:35.482526: step: 1940/529, loss: 5.054474058852065e-06 2023-01-21 13:53:36.655694: step: 1944/529, loss: 0.0021365166176110506 2023-01-21 13:53:37.841853: step: 1948/529, loss: 2.346038854739163e-05 2023-01-21 13:53:39.061763: step: 1952/529, loss: 0.0016041279304772615 2023-01-21 13:53:40.298915: step: 1956/529, loss: 0.06608200073242188 2023-01-21 13:53:41.530697: step: 1960/529, loss: 8.525848534191027e-05 2023-01-21 13:53:42.724705: step: 1964/529, loss: 0.0011838913196697831 2023-01-21 13:53:43.920607: step: 1968/529, loss: 0.0035634993109852076 2023-01-21 13:53:45.076353: step: 1972/529, loss: 0.0005339622730389237 2023-01-21 13:53:46.221267: step: 1976/529, loss: 0.0028240203391760588 2023-01-21 13:53:47.402370: step: 1980/529, loss: 0.00021314620971679688 2023-01-21 13:53:48.592029: step: 1984/529, loss: 0.02833261527121067 2023-01-21 13:53:49.763340: step: 1988/529, loss: 2.77042381640058e-05 2023-01-21 13:53:50.971852: step: 1992/529, loss: 0.09884300082921982 2023-01-21 13:53:52.167586: step: 1996/529, loss: 0.005834007635712624 2023-01-21 13:53:53.375486: step: 2000/529, loss: 0.005459595005959272 2023-01-21 13:53:54.597835: step: 2004/529, loss: 0.012092972174286842 2023-01-21 13:53:55.804393: step: 2008/529, loss: 5.083084397483617e-05 2023-01-21 13:53:57.017989: step: 2012/529, loss: 0.0010726929176598787 2023-01-21 13:53:58.210723: step: 2016/529, loss: 0.0007318496354855597 2023-01-21 13:53:59.428674: step: 2020/529, loss: 2.1839141481905244e-05 2023-01-21 13:54:00.624241: step: 2024/529, loss: 0.0002273559512104839 2023-01-21 13:54:01.811284: step: 2028/529, loss: 8.7738037109375e-05 2023-01-21 13:54:03.019307: step: 2032/529, loss: 0.031193161383271217 2023-01-21 13:54:04.211472: step: 2036/529, loss: 3.6525725590763614e-05 2023-01-21 13:54:05.451050: step: 2040/529, loss: 0.0024789811577647924 2023-01-21 13:54:06.645530: step: 2044/529, loss: 0.00010967255366267636 2023-01-21 13:54:07.861506: step: 2048/529, loss: 0.11280445754528046 2023-01-21 13:54:09.057198: step: 2052/529, loss: 0.00018043517775367945 2023-01-21 13:54:10.266869: step: 2056/529, loss: 0.003935051150619984 2023-01-21 13:54:11.453664: step: 2060/529, loss: 0.000484466552734375 2023-01-21 13:54:12.697241: step: 2064/529, loss: 0.0006269454606808722 2023-01-21 13:54:13.992749: step: 2068/529, loss: 0.0044602397829294205 2023-01-21 13:54:15.173830: step: 2072/529, loss: 0.0001160621686722152 2023-01-21 13:54:16.383636: step: 2076/529, loss: 0.00020751952251885086 2023-01-21 13:54:17.585252: step: 2080/529, loss: 0.00010733604722190648 2023-01-21 13:54:18.734744: step: 2084/529, loss: 0.00017858743376564234 2023-01-21 13:54:19.936112: step: 2088/529, loss: 5.14984130859375e-05 2023-01-21 13:54:21.136540: step: 2092/529, loss: 0.0007474899757653475 2023-01-21 13:54:22.378258: step: 2096/529, loss: 3.0517578125e-05 2023-01-21 13:54:23.553007: step: 2100/529, loss: 0.00019588469876907766 2023-01-21 13:54:24.770482: step: 2104/529, loss: 0.005795764736831188 2023-01-21 13:54:25.919979: step: 2108/529, loss: 3.3664702641544864e-05 2023-01-21 13:54:27.119880: step: 2112/529, loss: 5.35011276951991e-05 2023-01-21 13:54:28.291380: step: 2116/529, loss: 9.632110959501006e-06 ================================================== Loss: 0.027 -------------------- Dev: {'event': {'p': 0.608324439701174, 'r': 0.758988015978695, 'f1': 0.6753554502369669}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Test: {'event': {'p': 0.6355810616929699, 'r': 0.7962852007189934, 'f1': 0.7069148936170213}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Chinese: {'event': {'p': 0.6571428571428571, 'r': 0.8518518518518519, 'f1': 0.7419354838709677}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Korean: {'event': {'p': 0.54, 'r': 0.42857142857142855, 'f1': 0.47787610619469023}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Russian: {'event': {'p': 0.5128205128205128, 'r': 0.5555555555555556, 'f1': 0.5333333333333333}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 13:55:11.347549: step: 4/529, loss: 0.002486610319465399 2023-01-21 13:55:12.519144: step: 8/529, loss: 0.0005454063648357987 2023-01-21 13:55:13.752670: step: 12/529, loss: 1.3828277587890625e-05 2023-01-21 13:55:14.958841: step: 16/529, loss: 0.0005033492925576866 2023-01-21 13:55:16.180380: step: 20/529, loss: 0.0004085540713276714 2023-01-21 13:55:17.306317: step: 24/529, loss: 0.0005671024555340409 2023-01-21 13:55:18.507607: step: 28/529, loss: 0.0001996994105866179 2023-01-21 13:55:19.661960: step: 32/529, loss: 0.009290218353271484 2023-01-21 13:55:20.813585: step: 36/529, loss: 4.405975414556451e-05 2023-01-21 13:55:22.022820: step: 40/529, loss: 1.6975403923424892e-05 2023-01-21 13:55:23.167533: step: 44/529, loss: 3.166198803228326e-05 2023-01-21 13:55:24.406914: step: 48/529, loss: 0.03087139129638672 2023-01-21 13:55:25.636367: step: 52/529, loss: 0.008427905850112438 2023-01-21 13:55:26.815361: step: 56/529, loss: 1.7976761228055693e-05 2023-01-21 13:55:28.021816: step: 60/529, loss: 0.0009828568436205387 2023-01-21 13:55:29.228755: step: 64/529, loss: 0.00016860962205100805 2023-01-21 13:55:30.430988: step: 68/529, loss: 0.00017662048048805445 2023-01-21 13:55:31.593880: step: 72/529, loss: 0.02104807086288929 2023-01-21 13:55:32.774846: step: 76/529, loss: 2.09808349609375e-05 2023-01-21 13:55:33.984315: step: 80/529, loss: 0.0002040863037109375 2023-01-21 13:55:35.195481: step: 84/529, loss: 0.010392571799457073 2023-01-21 13:55:36.370573: step: 88/529, loss: 2.09808349609375e-05 2023-01-21 13:55:37.549712: step: 92/529, loss: 0.0007870674016885459 2023-01-21 13:55:38.775813: step: 96/529, loss: 0.00017404556274414062 2023-01-21 13:55:39.956852: step: 100/529, loss: 8.96453821042087e-06 2023-01-21 13:55:41.256555: step: 104/529, loss: 0.00034046173095703125 2023-01-21 13:55:42.436606: step: 108/529, loss: 0.006067848298698664 2023-01-21 13:55:43.641214: step: 112/529, loss: 0.00023155212693382055 2023-01-21 13:55:44.860540: step: 116/529, loss: 8.7738037109375e-05 2023-01-21 13:55:46.043548: step: 120/529, loss: 0.026311874389648438 2023-01-21 13:55:47.248048: step: 124/529, loss: 4.997253563487902e-05 2023-01-21 13:55:48.398124: step: 128/529, loss: 8.77380352903856e-06 2023-01-21 13:55:49.555732: step: 132/529, loss: 0.006289768498390913 2023-01-21 13:55:50.732408: step: 136/529, loss: 1.678466833254788e-05 2023-01-21 13:55:51.951460: step: 140/529, loss: 0.009939384646713734 2023-01-21 13:55:53.159032: step: 144/529, loss: 0.0007006645319052041 2023-01-21 13:55:54.326999: step: 148/529, loss: 2.603531174827367e-05 2023-01-21 13:55:55.518028: step: 152/529, loss: 0.023067761212587357 2023-01-21 13:55:56.685348: step: 156/529, loss: 0.000702285731676966 2023-01-21 13:55:57.868265: step: 160/529, loss: 4.196167537884321e-06 2023-01-21 13:55:59.044272: step: 164/529, loss: 0.01557083148509264 2023-01-21 13:56:00.200763: step: 168/529, loss: 0.0030640128534287214 2023-01-21 13:56:01.393054: step: 172/529, loss: 8.831024024402723e-05 2023-01-21 13:56:02.571435: step: 176/529, loss: 0.0007307053310796618 2023-01-21 13:56:03.768232: step: 180/529, loss: 0.5614844560623169 2023-01-21 13:56:04.954150: step: 184/529, loss: 5.1116945542162284e-05 2023-01-21 13:56:06.140047: step: 188/529, loss: 0.0006590366829186678 2023-01-21 13:56:07.342052: step: 192/529, loss: 0.024069594219326973 2023-01-21 13:56:08.552282: step: 196/529, loss: 8.525848534191027e-05 2023-01-21 13:56:09.768955: step: 200/529, loss: 5.91278057981981e-06 2023-01-21 13:56:11.001902: step: 204/529, loss: 0.0004283905145712197 2023-01-21 13:56:12.183921: step: 208/529, loss: 0.23826999962329865 2023-01-21 13:56:13.405931: step: 212/529, loss: 0.0013281822903081775 2023-01-21 13:56:14.624304: step: 216/529, loss: 2.6035309929284267e-05 2023-01-21 13:56:15.812230: step: 220/529, loss: 0.00026111601619049907 2023-01-21 13:56:16.975235: step: 224/529, loss: 0.0002505302254576236 2023-01-21 13:56:18.130887: step: 228/529, loss: 0.0002790451399050653 2023-01-21 13:56:19.311829: step: 232/529, loss: 0.0001808166562113911 2023-01-21 13:56:20.559008: step: 236/529, loss: 0.0013794898986816406 2023-01-21 13:56:21.747952: step: 240/529, loss: 3.452301098150201e-05 2023-01-21 13:56:22.921659: step: 244/529, loss: 1.8024444216280244e-05 2023-01-21 13:56:24.061167: step: 248/529, loss: 0.0009590149275027215 2023-01-21 13:56:25.275356: step: 252/529, loss: 0.0034128190018236637 2023-01-21 13:56:26.455762: step: 256/529, loss: 0.0032050132285803556 2023-01-21 13:56:27.634386: step: 260/529, loss: 1.6212465197895654e-05 2023-01-21 13:56:28.804854: step: 264/529, loss: 3.814697322468419e-07 2023-01-21 13:56:30.001239: step: 268/529, loss: 2.937316821771674e-05 2023-01-21 13:56:31.180339: step: 272/529, loss: 0.009503936395049095 2023-01-21 13:56:32.399794: step: 276/529, loss: 9.584426152287051e-06 2023-01-21 13:56:33.575882: step: 280/529, loss: 0.003019094467163086 2023-01-21 13:56:34.742213: step: 284/529, loss: 0.053794100880622864 2023-01-21 13:56:35.975610: step: 288/529, loss: 0.00039386749267578125 2023-01-21 13:56:37.229442: step: 292/529, loss: 0.0002857208310160786 2023-01-21 13:56:38.439886: step: 296/529, loss: 0.00019207000150345266 2023-01-21 13:56:39.621547: step: 300/529, loss: 0.0008145809406414628 2023-01-21 13:56:40.806026: step: 304/529, loss: 1.583099401614163e-05 2023-01-21 13:56:41.968612: step: 308/529, loss: 0.05093097686767578 2023-01-21 13:56:43.134059: step: 312/529, loss: -1.106262243411038e-05 2023-01-21 13:56:44.295655: step: 316/529, loss: 1.564025842526462e-05 2023-01-21 13:56:45.576048: step: 320/529, loss: 0.0007534981123171747 2023-01-21 13:56:46.751270: step: 324/529, loss: 7.724762326688506e-06 2023-01-21 13:56:47.959377: step: 328/529, loss: 0.0011709213722497225 2023-01-21 13:56:49.130537: step: 332/529, loss: 0.0011053085327148438 2023-01-21 13:56:50.295993: step: 336/529, loss: 0.0003216266632080078 2023-01-21 13:56:51.464937: step: 340/529, loss: 0.002346134278923273 2023-01-21 13:56:52.662052: step: 344/529, loss: 3.4046173823298886e-05 2023-01-21 13:56:53.879403: step: 348/529, loss: 0.022385407239198685 2023-01-21 13:56:55.048541: step: 352/529, loss: 0.00017142295837402344 2023-01-21 13:56:56.247615: step: 356/529, loss: 0.011541938409209251 2023-01-21 13:56:57.413457: step: 360/529, loss: 9.775162470759824e-05 2023-01-21 13:56:58.652948: step: 364/529, loss: 0.015950823202729225 2023-01-21 13:56:59.847175: step: 368/529, loss: 0.01206598337739706 2023-01-21 13:57:01.030408: step: 372/529, loss: -2.670288040462765e-06 2023-01-21 13:57:02.213477: step: 376/529, loss: 2.384185791015625e-06 2023-01-21 13:57:03.414035: step: 380/529, loss: 5.197525388211943e-05 2023-01-21 13:57:04.608171: step: 384/529, loss: 2.975463939947076e-05 2023-01-21 13:57:05.781129: step: 388/529, loss: 8.583068620282575e-07 2023-01-21 13:57:06.994121: step: 392/529, loss: 0.002277946565300226 2023-01-21 13:57:08.199088: step: 396/529, loss: 0.0003795623779296875 2023-01-21 13:57:09.402602: step: 400/529, loss: 5.948544185230276e-06 2023-01-21 13:57:10.603032: step: 404/529, loss: 2.264976683363784e-05 2023-01-21 13:57:11.817706: step: 408/529, loss: 0.008116102777421474 2023-01-21 13:57:12.992296: step: 412/529, loss: 0.01571502722799778 2023-01-21 13:57:14.211631: step: 416/529, loss: 0.0037660598754882812 2023-01-21 13:57:15.397733: step: 420/529, loss: 0.004323672968894243 2023-01-21 13:57:16.594038: step: 424/529, loss: -5.722044988942798e-07 2023-01-21 13:57:17.770931: step: 428/529, loss: 0.002426528837531805 2023-01-21 13:57:18.933976: step: 432/529, loss: 0.0012628554832190275 2023-01-21 13:57:20.158488: step: 436/529, loss: 0.0017341614002361894 2023-01-21 13:57:21.318427: step: 440/529, loss: 0.008396483026444912 2023-01-21 13:57:22.505000: step: 444/529, loss: -9.536745437799254e-08 2023-01-21 13:57:23.737865: step: 448/529, loss: 4.587173680192791e-05 2023-01-21 13:57:24.916156: step: 452/529, loss: 8.0108642578125e-05 2023-01-21 13:57:26.137260: step: 456/529, loss: 4.48703758593183e-05 2023-01-21 13:57:27.324299: step: 460/529, loss: 0.0006788254249840975 2023-01-21 13:57:28.525637: step: 464/529, loss: 0.0002368926943745464 2023-01-21 13:57:29.716035: step: 468/529, loss: 0.0015342712868005037 2023-01-21 13:57:30.904221: step: 472/529, loss: 0.03257932886481285 2023-01-21 13:57:32.087163: step: 476/529, loss: 0.000133514404296875 2023-01-21 13:57:33.268625: step: 480/529, loss: 0.00044574736966751516 2023-01-21 13:57:34.492717: step: 484/529, loss: 0.0001923561212606728 2023-01-21 13:57:35.713663: step: 488/529, loss: 0.0010777473216876388 2023-01-21 13:57:36.930566: step: 492/529, loss: 2.9087066195643274e-06 2023-01-21 13:57:38.168958: step: 496/529, loss: 0.0001256942778127268 2023-01-21 13:57:39.345912: step: 500/529, loss: 0.00018177033052779734 2023-01-21 13:57:40.504997: step: 504/529, loss: 0.00011037290096282959 2023-01-21 13:57:41.632860: step: 508/529, loss: 4.875659669778543e-06 2023-01-21 13:57:42.814417: step: 512/529, loss: 0.03704824671149254 2023-01-21 13:57:43.986377: step: 516/529, loss: 0.0005247116205282509 2023-01-21 13:57:45.165603: step: 520/529, loss: 9.5367431640625e-07 2023-01-21 13:57:46.339640: step: 524/529, loss: 1.029968279908644e-05 2023-01-21 13:57:47.534609: step: 528/529, loss: 0.019686318933963776 2023-01-21 13:57:48.711396: step: 532/529, loss: 0.0001111269011744298 2023-01-21 13:57:49.859800: step: 536/529, loss: 7.82012921263231e-06 2023-01-21 13:57:51.038765: step: 540/529, loss: 4.119873119634576e-05 2023-01-21 13:57:52.237593: step: 544/529, loss: 0.0014347076648846269 2023-01-21 13:57:53.437105: step: 548/529, loss: 0.001260089804418385 2023-01-21 13:57:54.596044: step: 552/529, loss: 0.0002948761102743447 2023-01-21 13:57:55.793799: step: 556/529, loss: 0.009482955560088158 2023-01-21 13:57:57.003766: step: 560/529, loss: 0.0001127243012888357 2023-01-21 13:57:58.228183: step: 564/529, loss: 2.574920654296875e-05 2023-01-21 13:57:59.424219: step: 568/529, loss: 0.0022785186301916838 2023-01-21 13:58:00.602561: step: 572/529, loss: 0.003586387727409601 2023-01-21 13:58:01.776506: step: 576/529, loss: 0.009890555404126644 2023-01-21 13:58:02.963036: step: 580/529, loss: 4.863739377469756e-06 2023-01-21 13:58:04.173804: step: 584/529, loss: 0.01574115827679634 2023-01-21 13:58:05.370903: step: 588/529, loss: 0.0031417845748364925 2023-01-21 13:58:06.552006: step: 592/529, loss: 0.07559995353221893 2023-01-21 13:58:07.730365: step: 596/529, loss: 0.0011539459228515625 2023-01-21 13:58:08.906657: step: 600/529, loss: 0.00301017751917243 2023-01-21 13:58:10.122787: step: 604/529, loss: 0.0008828163263387978 2023-01-21 13:58:11.288872: step: 608/529, loss: 0.001103305839933455 2023-01-21 13:58:12.444975: step: 612/529, loss: 7.83920258982107e-05 2023-01-21 13:58:13.637112: step: 616/529, loss: 0.5508025288581848 2023-01-21 13:58:14.814763: step: 620/529, loss: 0.0001636505185160786 2023-01-21 13:58:15.981500: step: 624/529, loss: 0.00022392273240257055 2023-01-21 13:58:17.119688: step: 628/529, loss: 8.487701052217744e-06 2023-01-21 13:58:18.318404: step: 632/529, loss: 8.440018063993193e-06 2023-01-21 13:58:19.569007: step: 636/529, loss: -3.719329924933845e-06 2023-01-21 13:58:20.786374: step: 640/529, loss: 6.215572648216039e-05 2023-01-21 13:58:22.013394: step: 644/529, loss: 0.0003132820129394531 2023-01-21 13:58:23.191828: step: 648/529, loss: 0.0046621328219771385 2023-01-21 13:58:24.378739: step: 652/529, loss: 0.001276302384212613 2023-01-21 13:58:25.563783: step: 656/529, loss: 1.106262243411038e-05 2023-01-21 13:58:26.758679: step: 660/529, loss: 0.0013508796691894531 2023-01-21 13:58:27.969483: step: 664/529, loss: 5.9890749980695546e-05 2023-01-21 13:58:29.175962: step: 668/529, loss: 0.006545639596879482 2023-01-21 13:58:30.336935: step: 672/529, loss: 2.765655608527595e-06 2023-01-21 13:58:31.509651: step: 676/529, loss: 0.00011539459228515625 2023-01-21 13:58:32.705125: step: 680/529, loss: 5.931854320806451e-05 2023-01-21 13:58:33.874070: step: 684/529, loss: 0.00044164658174850047 2023-01-21 13:58:35.102486: step: 688/529, loss: 0.0007810592651367188 2023-01-21 13:58:36.256620: step: 692/529, loss: 3.137588282697834e-05 2023-01-21 13:58:37.440533: step: 696/529, loss: 3.890991138177924e-05 2023-01-21 13:58:38.649837: step: 700/529, loss: 0.0807764008641243 2023-01-21 13:58:39.836778: step: 704/529, loss: 0.07057266682386398 2023-01-21 13:58:41.026543: step: 708/529, loss: 0.004187965299934149 2023-01-21 13:58:42.205429: step: 712/529, loss: 0.006999587640166283 2023-01-21 13:58:43.391411: step: 716/529, loss: 0.00022430421086028218 2023-01-21 13:58:44.556249: step: 720/529, loss: 9.775161743164062e-05 2023-01-21 13:58:45.747435: step: 724/529, loss: 0.00011463165719760582 2023-01-21 13:58:46.954578: step: 728/529, loss: 0.000489568745251745 2023-01-21 13:58:48.146415: step: 732/529, loss: 0.005581521894782782 2023-01-21 13:58:49.308066: step: 736/529, loss: 0.000760841416195035 2023-01-21 13:58:50.506519: step: 740/529, loss: 0.006346225738525391 2023-01-21 13:58:51.672979: step: 744/529, loss: 0.00013961792865302414 2023-01-21 13:58:52.858562: step: 748/529, loss: 0.00010690689668990672 2023-01-21 13:58:54.083372: step: 752/529, loss: 0.637780487537384 2023-01-21 13:58:55.254821: step: 756/529, loss: 0.0035182000137865543 2023-01-21 13:58:56.430325: step: 760/529, loss: 5.91278076171875e-05 2023-01-21 13:58:57.578695: step: 764/529, loss: 0.0017399787902832031 2023-01-21 13:58:58.775976: step: 768/529, loss: 0.0027918340638279915 2023-01-21 13:58:59.989519: step: 772/529, loss: 0.0005471229669637978 2023-01-21 13:59:01.150289: step: 776/529, loss: 0.011491012759506702 2023-01-21 13:59:02.427376: step: 780/529, loss: 0.033760640770196915 2023-01-21 13:59:03.648552: step: 784/529, loss: 0.01816840097308159 2023-01-21 13:59:04.828716: step: 788/529, loss: 0.0046198368072509766 2023-01-21 13:59:05.990988: step: 792/529, loss: 4.5156477426644415e-05 2023-01-21 13:59:07.188155: step: 796/529, loss: 0.00037469863309524953 2023-01-21 13:59:08.373549: step: 800/529, loss: 0.0029431821312755346 2023-01-21 13:59:09.546766: step: 804/529, loss: 0.022354699671268463 2023-01-21 13:59:10.708894: step: 808/529, loss: 0.03462781757116318 2023-01-21 13:59:11.886415: step: 812/529, loss: 0.00015592575073242188 2023-01-21 13:59:13.091113: step: 816/529, loss: 0.000102996826171875 2023-01-21 13:59:14.283488: step: 820/529, loss: 0.0003194332239218056 2023-01-21 13:59:15.441858: step: 824/529, loss: 0.002585983369499445 2023-01-21 13:59:16.603202: step: 828/529, loss: 0.0099938390776515 2023-01-21 13:59:17.779133: step: 832/529, loss: 0.00041918756323866546 2023-01-21 13:59:18.947679: step: 836/529, loss: 0.0003768921014852822 2023-01-21 13:59:20.160531: step: 840/529, loss: 4.18663039454259e-05 2023-01-21 13:59:21.353467: step: 844/529, loss: 0.00011382102820789441 2023-01-21 13:59:22.496325: step: 848/529, loss: 8.535384949936997e-06 2023-01-21 13:59:23.734512: step: 852/529, loss: 0.00019674301438499242 2023-01-21 13:59:24.930487: step: 856/529, loss: 0.0007982254610396922 2023-01-21 13:59:26.114232: step: 860/529, loss: 0.01453480776399374 2023-01-21 13:59:27.324151: step: 864/529, loss: 0.00403099087998271 2023-01-21 13:59:28.493782: step: 868/529, loss: 0.0033857347443699837 2023-01-21 13:59:29.686614: step: 872/529, loss: 5.4168704082258046e-05 2023-01-21 13:59:30.829571: step: 876/529, loss: 0.003027343889698386 2023-01-21 13:59:32.037423: step: 880/529, loss: 0.0005221366882324219 2023-01-21 13:59:33.274388: step: 884/529, loss: 4.95910626341356e-06 2023-01-21 13:59:34.446929: step: 888/529, loss: 0.00013847352238371968 2023-01-21 13:59:35.649032: step: 892/529, loss: 0.0012887954944744706 2023-01-21 13:59:36.905044: step: 896/529, loss: 0.0002054214564850554 2023-01-21 13:59:38.063495: step: 900/529, loss: 1.6689300537109375e-06 2023-01-21 13:59:39.279094: step: 904/529, loss: 0.0005992889637127519 2023-01-21 13:59:40.463769: step: 908/529, loss: 0.00015020370483398438 2023-01-21 13:59:41.678168: step: 912/529, loss: 0.02367849461734295 2023-01-21 13:59:42.900025: step: 916/529, loss: 0.03129472956061363 2023-01-21 13:59:44.076653: step: 920/529, loss: 0.07983321696519852 2023-01-21 13:59:45.297800: step: 924/529, loss: 3.24249276673072e-06 2023-01-21 13:59:46.459678: step: 928/529, loss: 2.1028519768151455e-05 2023-01-21 13:59:47.647053: step: 932/529, loss: 0.0003984451468568295 2023-01-21 13:59:48.817350: step: 936/529, loss: 0.004168891813606024 2023-01-21 13:59:50.030742: step: 940/529, loss: 0.00019202232942916453 2023-01-21 13:59:51.203692: step: 944/529, loss: 0.011140060611069202 2023-01-21 13:59:52.383672: step: 948/529, loss: 0.00029754638671875 2023-01-21 13:59:53.579447: step: 952/529, loss: 0.011180304922163486 2023-01-21 13:59:54.818155: step: 956/529, loss: 2.784729076665826e-05 2023-01-21 13:59:56.053958: step: 960/529, loss: 2.1553038095589727e-05 2023-01-21 13:59:57.233149: step: 964/529, loss: 0.08011846989393234 2023-01-21 13:59:58.427856: step: 968/529, loss: 0.010040283203125 2023-01-21 13:59:59.618819: step: 972/529, loss: 0.025724411010742188 2023-01-21 14:00:00.779635: step: 976/529, loss: 7.3909759521484375e-06 2023-01-21 14:00:01.966417: step: 980/529, loss: 0.0004756927664857358 2023-01-21 14:00:03.165948: step: 984/529, loss: 0.00335521693341434 2023-01-21 14:00:04.351603: step: 988/529, loss: 2.803802453854587e-05 2023-01-21 14:00:05.548627: step: 992/529, loss: 0.0028936387971043587 2023-01-21 14:00:06.737350: step: 996/529, loss: 0.04069824144244194 2023-01-21 14:00:07.930025: step: 1000/529, loss: 0.007387733552604914 2023-01-21 14:00:09.104637: step: 1004/529, loss: 0.0010744095779955387 2023-01-21 14:00:10.301103: step: 1008/529, loss: 0.00016803742619231343 2023-01-21 14:00:11.453647: step: 1012/529, loss: 0.002085113665089011 2023-01-21 14:00:12.605829: step: 1016/529, loss: -6.198883397701138e-07 2023-01-21 14:00:13.815171: step: 1020/529, loss: 2.6702880859375e-05 2023-01-21 14:00:14.987305: step: 1024/529, loss: -2.86102294921875e-06 2023-01-21 14:00:16.177371: step: 1028/529, loss: 4.754066685563885e-05 2023-01-21 14:00:17.413088: step: 1032/529, loss: 0.12673984467983246 2023-01-21 14:00:18.600506: step: 1036/529, loss: 0.0003602981742005795 2023-01-21 14:00:19.773928: step: 1040/529, loss: 2.2602082026423886e-05 2023-01-21 14:00:20.966666: step: 1044/529, loss: 1.2111663636460435e-05 2023-01-21 14:00:22.157613: step: 1048/529, loss: 0.002100181533023715 2023-01-21 14:00:23.409006: step: 1052/529, loss: 0.2902399003505707 2023-01-21 14:00:24.609060: step: 1056/529, loss: -4.38690176451928e-06 2023-01-21 14:00:25.789230: step: 1060/529, loss: 0.005321884527802467 2023-01-21 14:00:26.929278: step: 1064/529, loss: 1.0013580322265625e-05 2023-01-21 14:00:28.148211: step: 1068/529, loss: 0.04849839210510254 2023-01-21 14:00:29.316531: step: 1072/529, loss: 0.0009517669677734375 2023-01-21 14:00:30.512396: step: 1076/529, loss: -8.773804438533261e-06 2023-01-21 14:00:31.658840: step: 1080/529, loss: 0.00018634795560501516 2023-01-21 14:00:32.862751: step: 1084/529, loss: -6.675719532722724e-07 2023-01-21 14:00:34.069208: step: 1088/529, loss: 0.010739946737885475 2023-01-21 14:00:35.268786: step: 1092/529, loss: 4.6205521357478574e-05 2023-01-21 14:00:36.499563: step: 1096/529, loss: 7.743835885776207e-05 2023-01-21 14:00:37.695472: step: 1100/529, loss: 0.2479230910539627 2023-01-21 14:00:38.903514: step: 1104/529, loss: 0.020854569971561432 2023-01-21 14:00:40.066503: step: 1108/529, loss: 0.0001441717176930979 2023-01-21 14:00:41.241062: step: 1112/529, loss: 0.00016117095947265625 2023-01-21 14:00:42.450133: step: 1116/529, loss: 0.03041229397058487 2023-01-21 14:00:43.655839: step: 1120/529, loss: 0.006233597174286842 2023-01-21 14:00:44.831446: step: 1124/529, loss: -1.8835066839528736e-06 2023-01-21 14:00:46.089964: step: 1128/529, loss: 0.0012374878861010075 2023-01-21 14:00:47.285664: step: 1132/529, loss: 1.5258789289873675e-06 2023-01-21 14:00:48.437670: step: 1136/529, loss: 6.67572021484375e-06 2023-01-21 14:00:49.604454: step: 1140/529, loss: 4.57763690064894e-06 2023-01-21 14:00:50.765405: step: 1144/529, loss: 0.04687223210930824 2023-01-21 14:00:51.962422: step: 1148/529, loss: 0.010915899649262428 2023-01-21 14:00:53.099051: step: 1152/529, loss: 1.602172778802924e-05 2023-01-21 14:00:54.306362: step: 1156/529, loss: 0.02811412885785103 2023-01-21 14:00:55.497605: step: 1160/529, loss: 3.919601658708416e-05 2023-01-21 14:00:56.711637: step: 1164/529, loss: 0.0004516601620707661 2023-01-21 14:00:57.896679: step: 1168/529, loss: 2.689361645025201e-05 2023-01-21 14:00:59.112849: step: 1172/529, loss: 0.005347347352653742 2023-01-21 14:01:00.277067: step: 1176/529, loss: 0.007536268327385187 2023-01-21 14:01:01.464803: step: 1180/529, loss: 0.043550778180360794 2023-01-21 14:01:02.643020: step: 1184/529, loss: 2.2029877072782256e-05 2023-01-21 14:01:03.826101: step: 1188/529, loss: 1.201629675051663e-05 2023-01-21 14:01:05.001625: step: 1192/529, loss: 2.555847095209174e-05 2023-01-21 14:01:06.191749: step: 1196/529, loss: 0.0003506660577841103 2023-01-21 14:01:07.384707: step: 1200/529, loss: 0.0031450269743800163 2023-01-21 14:01:08.571663: step: 1204/529, loss: 0.10371074825525284 2023-01-21 14:01:09.795176: step: 1208/529, loss: 0.07595143467187881 2023-01-21 14:01:10.994129: step: 1212/529, loss: 0.0015107154613360763 2023-01-21 14:01:12.209330: step: 1216/529, loss: 0.0004150390741415322 2023-01-21 14:01:13.405888: step: 1220/529, loss: 0.005833053961396217 2023-01-21 14:01:14.581265: step: 1224/529, loss: 0.00012621880159713328 2023-01-21 14:01:15.784211: step: 1228/529, loss: 0.0037235261406749487 2023-01-21 14:01:16.962312: step: 1232/529, loss: 0.04663591459393501 2023-01-21 14:01:18.133958: step: 1236/529, loss: 1.1014938536391128e-05 2023-01-21 14:01:19.339421: step: 1240/529, loss: 3.4332276754867053e-06 2023-01-21 14:01:20.514963: step: 1244/529, loss: 0.013466835021972656 2023-01-21 14:01:21.744047: step: 1248/529, loss: 0.0031305314041674137 2023-01-21 14:01:22.947424: step: 1252/529, loss: 0.0013742446899414062 2023-01-21 14:01:24.121826: step: 1256/529, loss: 0.024256134405732155 2023-01-21 14:01:25.328681: step: 1260/529, loss: 0.0006215095636434853 2023-01-21 14:01:26.542115: step: 1264/529, loss: 0.004816532135009766 2023-01-21 14:01:27.708898: step: 1268/529, loss: 0.0010930061107501388 2023-01-21 14:01:28.889552: step: 1272/529, loss: 0.0013159752124920487 2023-01-21 14:01:30.079413: step: 1276/529, loss: 0.0001308917999267578 2023-01-21 14:01:31.327644: step: 1280/529, loss: 3.352165367687121e-05 2023-01-21 14:01:32.540190: step: 1284/529, loss: 0.0028423310723155737 2023-01-21 14:01:33.758961: step: 1288/529, loss: 0.0013466834789142013 2023-01-21 14:01:34.952425: step: 1292/529, loss: 1.430511474609375e-05 2023-01-21 14:01:36.144039: step: 1296/529, loss: 0.00586891220882535 2023-01-21 14:01:37.334904: step: 1300/529, loss: 0.007724952884018421 2023-01-21 14:01:38.515637: step: 1304/529, loss: 0.12593841552734375 2023-01-21 14:01:39.743225: step: 1308/529, loss: 0.0009386062738485634 2023-01-21 14:01:40.925649: step: 1312/529, loss: 0.0006546020740643144 2023-01-21 14:01:42.129387: step: 1316/529, loss: 5.435943603515625e-05 2023-01-21 14:01:43.318806: step: 1320/529, loss: 0.0003288268926553428 2023-01-21 14:01:44.517877: step: 1324/529, loss: -1.9311905816721264e-06 2023-01-21 14:01:45.746677: step: 1328/529, loss: 4.425048973644152e-05 2023-01-21 14:01:46.950270: step: 1332/529, loss: 6.866455078125e-05 2023-01-21 14:01:48.198737: step: 1336/529, loss: -1.4352798643813003e-05 2023-01-21 14:01:49.384883: step: 1340/529, loss: 0.0009828567272052169 2023-01-21 14:01:50.576509: step: 1344/529, loss: 0.02327137067914009 2023-01-21 14:01:51.760184: step: 1348/529, loss: 0.02124004438519478 2023-01-21 14:01:52.982754: step: 1352/529, loss: 0.00036783216637559235 2023-01-21 14:01:54.154137: step: 1356/529, loss: 0.012428618036210537 2023-01-21 14:01:55.362791: step: 1360/529, loss: 0.0004364013730082661 2023-01-21 14:01:56.571537: step: 1364/529, loss: 0.0009974479908123612 2023-01-21 14:01:57.814451: step: 1368/529, loss: 0.021364212036132812 2023-01-21 14:01:59.019505: step: 1372/529, loss: 0.0004796028370037675 2023-01-21 14:02:00.239311: step: 1376/529, loss: 0.00023784636869095266 2023-01-21 14:02:01.450977: step: 1380/529, loss: 0.010939240455627441 2023-01-21 14:02:02.667295: step: 1384/529, loss: 0.010668087750673294 2023-01-21 14:02:03.859842: step: 1388/529, loss: 0.010875415056943893 2023-01-21 14:02:05.106372: step: 1392/529, loss: 0.0016296387184411287 2023-01-21 14:02:06.295128: step: 1396/529, loss: 0.10114937275648117 2023-01-21 14:02:07.488690: step: 1400/529, loss: 0.00014142990403342992 2023-01-21 14:02:08.684434: step: 1404/529, loss: 0.01692981831729412 2023-01-21 14:02:09.878435: step: 1408/529, loss: 0.0073986053466796875 2023-01-21 14:02:11.062707: step: 1412/529, loss: 0.0018570900429040194 2023-01-21 14:02:12.260587: step: 1416/529, loss: 1.3351441339182202e-05 2023-01-21 14:02:13.451136: step: 1420/529, loss: 6.637573096668348e-05 2023-01-21 14:02:14.674113: step: 1424/529, loss: 4.901886131847277e-05 2023-01-21 14:02:15.899356: step: 1428/529, loss: 6.675720101156912e-07 2023-01-21 14:02:17.092383: step: 1432/529, loss: 4.38690176451928e-06 2023-01-21 14:02:18.330984: step: 1436/529, loss: 0.0013442039489746094 2023-01-21 14:02:19.570013: step: 1440/529, loss: 0.0004611015319824219 2023-01-21 14:02:20.757493: step: 1444/529, loss: -4.95910626341356e-06 2023-01-21 14:02:21.979047: step: 1448/529, loss: 0.002014636993408203 2023-01-21 14:02:23.152963: step: 1452/529, loss: 0.05167656019330025 2023-01-21 14:02:24.337257: step: 1456/529, loss: 0.0015058517456054688 2023-01-21 14:02:25.558453: step: 1460/529, loss: 0.03955860063433647 2023-01-21 14:02:26.758264: step: 1464/529, loss: 0.03722947835922241 2023-01-21 14:02:27.954075: step: 1468/529, loss: 0.01768798939883709 2023-01-21 14:02:29.126844: step: 1472/529, loss: 0.012889672070741653 2023-01-21 14:02:30.364236: step: 1476/529, loss: 0.0015233993763104081 2023-01-21 14:02:31.528055: step: 1480/529, loss: 0.0009507178911007941 2023-01-21 14:02:32.690176: step: 1484/529, loss: 0.0001539230433991179 2023-01-21 14:02:33.846619: step: 1488/529, loss: 2.7465821403893642e-05 2023-01-21 14:02:35.035181: step: 1492/529, loss: 0.00016374587721657008 2023-01-21 14:02:36.276259: step: 1496/529, loss: 0.013057136908173561 2023-01-21 14:02:37.452391: step: 1500/529, loss: 0.00038776398287154734 2023-01-21 14:02:38.648628: step: 1504/529, loss: 0.3891308903694153 2023-01-21 14:02:39.890107: step: 1508/529, loss: 2.4604798454674892e-05 2023-01-21 14:02:41.039612: step: 1512/529, loss: 0.0003338813839945942 2023-01-21 14:02:42.280742: step: 1516/529, loss: 0.00045928958570584655 2023-01-21 14:02:43.472863: step: 1520/529, loss: 1.4495850336970761e-05 2023-01-21 14:02:44.638404: step: 1524/529, loss: 0.0003067016659770161 2023-01-21 14:02:45.783014: step: 1528/529, loss: 0.0018367767333984375 2023-01-21 14:02:46.975598: step: 1532/529, loss: 0.0015243530506268144 2023-01-21 14:02:48.223684: step: 1536/529, loss: 0.008710002526640892 2023-01-21 14:02:49.395939: step: 1540/529, loss: 0.009780501946806908 2023-01-21 14:02:50.629268: step: 1544/529, loss: 6.713867333019152e-05 2023-01-21 14:02:51.841395: step: 1548/529, loss: 0.0002689361572265625 2023-01-21 14:02:53.018650: step: 1552/529, loss: 0.0006654262542724609 2023-01-21 14:02:54.238884: step: 1556/529, loss: 0.014021968469023705 2023-01-21 14:02:55.440793: step: 1560/529, loss: 5.626678466796875e-05 2023-01-21 14:02:56.663160: step: 1564/529, loss: 0.021380044519901276 2023-01-21 14:02:57.872756: step: 1568/529, loss: 0.003910636994987726 2023-01-21 14:02:59.117185: step: 1572/529, loss: 0.042675018310546875 2023-01-21 14:03:00.312229: step: 1576/529, loss: 0.004794502630829811 2023-01-21 14:03:01.479525: step: 1580/529, loss: 0.0016431808471679688 2023-01-21 14:03:02.679476: step: 1584/529, loss: 2.079009937006049e-05 2023-01-21 14:03:03.882640: step: 1588/529, loss: 0.003921222873032093 2023-01-21 14:03:05.044174: step: 1592/529, loss: 0.013204192742705345 2023-01-21 14:03:06.220950: step: 1596/529, loss: 0.00010919570922851562 2023-01-21 14:03:07.413538: step: 1600/529, loss: 0.01206207275390625 2023-01-21 14:03:08.585082: step: 1604/529, loss: 0.016500186175107956 2023-01-21 14:03:09.764266: step: 1608/529, loss: 2.651214526849799e-05 2023-01-21 14:03:10.975123: step: 1612/529, loss: 0.0038360594771802425 2023-01-21 14:03:12.200180: step: 1616/529, loss: 0.00177764892578125 2023-01-21 14:03:13.349059: step: 1620/529, loss: 0.0021054267417639494 2023-01-21 14:03:14.530599: step: 1624/529, loss: 0.0012552260886877775 2023-01-21 14:03:15.722655: step: 1628/529, loss: 0.007490634918212891 2023-01-21 14:03:16.936153: step: 1632/529, loss: 0.03485069423913956 2023-01-21 14:03:18.169036: step: 1636/529, loss: 0.004987811669707298 2023-01-21 14:03:19.364678: step: 1640/529, loss: -7.05719003235572e-06 2023-01-21 14:03:20.549161: step: 1644/529, loss: 0.0034748553298413754 2023-01-21 14:03:21.760323: step: 1648/529, loss: 0.0005661964532919228 2023-01-21 14:03:22.978084: step: 1652/529, loss: 0.022451020777225494 2023-01-21 14:03:24.203690: step: 1656/529, loss: 0.0001357078435830772 2023-01-21 14:03:25.371063: step: 1660/529, loss: 0.02911815606057644 2023-01-21 14:03:26.560634: step: 1664/529, loss: 0.0006277084467001259 2023-01-21 14:03:27.767368: step: 1668/529, loss: 0.00031528473482467234 2023-01-21 14:03:28.976985: step: 1672/529, loss: 8.335112943314016e-05 2023-01-21 14:03:30.191596: step: 1676/529, loss: 0.0021923722233623266 2023-01-21 14:03:31.385981: step: 1680/529, loss: 5.5027008784236386e-05 2023-01-21 14:03:32.541641: step: 1684/529, loss: 0.004140091128647327 2023-01-21 14:03:33.736862: step: 1688/529, loss: 9.078979928744957e-05 2023-01-21 14:03:34.912842: step: 1692/529, loss: 6.31332368357107e-05 2023-01-21 14:03:36.170557: step: 1696/529, loss: 0.0006995201110839844 2023-01-21 14:03:37.376620: step: 1700/529, loss: 0.02968435175716877 2023-01-21 14:03:38.591208: step: 1704/529, loss: 0.05621156841516495 2023-01-21 14:03:39.813225: step: 1708/529, loss: 1.869201696536038e-05 2023-01-21 14:03:40.987560: step: 1712/529, loss: 0.012327956967055798 2023-01-21 14:03:42.164806: step: 1716/529, loss: 0.000583648681640625 2023-01-21 14:03:43.352929: step: 1720/529, loss: -1.049041748046875e-05 2023-01-21 14:03:44.575549: step: 1724/529, loss: -0.0 2023-01-21 14:03:45.760877: step: 1728/529, loss: 0.009617805480957031 2023-01-21 14:03:46.995454: step: 1732/529, loss: 0.006374072749167681 2023-01-21 14:03:48.183206: step: 1736/529, loss: 0.0006165504455566406 2023-01-21 14:03:49.403640: step: 1740/529, loss: 4.7874447773210704e-05 2023-01-21 14:03:50.618166: step: 1744/529, loss: 7.591247413074598e-05 2023-01-21 14:03:51.793206: step: 1748/529, loss: 0.10577736049890518 2023-01-21 14:03:52.939246: step: 1752/529, loss: 0.006121492478996515 2023-01-21 14:03:54.136007: step: 1756/529, loss: 7.743835885776207e-05 2023-01-21 14:03:55.341751: step: 1760/529, loss: 0.0013271331554278731 2023-01-21 14:03:56.530302: step: 1764/529, loss: 0.0006898880237713456 2023-01-21 14:03:57.714352: step: 1768/529, loss: 0.0035812377464026213 2023-01-21 14:03:58.915718: step: 1772/529, loss: 0.0008927345043048263 2023-01-21 14:04:00.109041: step: 1776/529, loss: 0.00018568038649391383 2023-01-21 14:04:01.254025: step: 1780/529, loss: 0.004691887181252241 2023-01-21 14:04:02.417554: step: 1784/529, loss: 2.1743775505456142e-05 2023-01-21 14:04:03.625458: step: 1788/529, loss: 0.4828300476074219 2023-01-21 14:04:04.839309: step: 1792/529, loss: 0.0002576827828306705 2023-01-21 14:04:06.038051: step: 1796/529, loss: 0.004059982020407915 2023-01-21 14:04:07.224275: step: 1800/529, loss: 0.08966103196144104 2023-01-21 14:04:08.398332: step: 1804/529, loss: 0.0018215179443359375 2023-01-21 14:04:09.617604: step: 1808/529, loss: 0.04393348842859268 2023-01-21 14:04:10.799230: step: 1812/529, loss: 0.00022945404634810984 2023-01-21 14:04:11.987501: step: 1816/529, loss: 0.02981853485107422 2023-01-21 14:04:13.158887: step: 1820/529, loss: 0.21706753969192505 2023-01-21 14:04:14.348895: step: 1824/529, loss: 8.010864803509321e-06 2023-01-21 14:04:15.532312: step: 1828/529, loss: 0.010526467114686966 2023-01-21 14:04:16.696222: step: 1832/529, loss: 0.384307861328125 2023-01-21 14:04:17.877593: step: 1836/529, loss: 6.403923180187121e-05 2023-01-21 14:04:19.120858: step: 1840/529, loss: 0.0010183334816247225 2023-01-21 14:04:20.286682: step: 1844/529, loss: 6.580352783203125e-05 2023-01-21 14:04:21.472716: step: 1848/529, loss: 0.0016900539631024003 2023-01-21 14:04:22.642645: step: 1852/529, loss: 0.04127006232738495 2023-01-21 14:04:23.835799: step: 1856/529, loss: 1.4400482541532256e-05 2023-01-21 14:04:25.023228: step: 1860/529, loss: 5.865097136847908e-06 2023-01-21 14:04:26.206425: step: 1864/529, loss: 0.026184655725955963 2023-01-21 14:04:27.419649: step: 1868/529, loss: 0.0017786979442462325 2023-01-21 14:04:28.647016: step: 1872/529, loss: 0.02577819861471653 2023-01-21 14:04:29.853788: step: 1876/529, loss: 7.057190487103071e-06 2023-01-21 14:04:31.069910: step: 1880/529, loss: 1.068115216185106e-05 2023-01-21 14:04:32.298995: step: 1884/529, loss: 0.0005952835199423134 2023-01-21 14:04:33.546889: step: 1888/529, loss: 0.029308224096894264 2023-01-21 14:04:34.735260: step: 1892/529, loss: 0.0006029129144735634 2023-01-21 14:04:35.888087: step: 1896/529, loss: 0.0004171371692791581 2023-01-21 14:04:37.076640: step: 1900/529, loss: 0.01193017978221178 2023-01-21 14:04:38.267191: step: 1904/529, loss: 0.001606154371984303 2023-01-21 14:04:39.439833: step: 1908/529, loss: 0.0003562927304301411 2023-01-21 14:04:40.644713: step: 1912/529, loss: 0.00019655228243209422 2023-01-21 14:04:41.855495: step: 1916/529, loss: 0.010395621880888939 2023-01-21 14:04:43.032211: step: 1920/529, loss: 0.0001253843365702778 2023-01-21 14:04:44.209905: step: 1924/529, loss: 8.77380352903856e-06 2023-01-21 14:04:45.392111: step: 1928/529, loss: 0.015627099201083183 2023-01-21 14:04:46.592020: step: 1932/529, loss: 0.00200996408239007 2023-01-21 14:04:47.788320: step: 1936/529, loss: 0.0002484321594238281 2023-01-21 14:04:48.982014: step: 1940/529, loss: 0.0008121490245684981 2023-01-21 14:04:50.137009: step: 1944/529, loss: 1.5258789289873675e-06 2023-01-21 14:04:51.309382: step: 1948/529, loss: -6.008148375258315e-06 2023-01-21 14:04:52.505918: step: 1952/529, loss: 0.014435959048569202 2023-01-21 14:04:53.670488: step: 1956/529, loss: 0.007031822577118874 2023-01-21 14:04:54.867549: step: 1960/529, loss: 0.00036101340083405375 2023-01-21 14:04:56.023467: step: 1964/529, loss: 0.0004458427429199219 2023-01-21 14:04:57.181081: step: 1968/529, loss: 8.506774611305445e-05 2023-01-21 14:04:58.437626: step: 1972/529, loss: -2.86102294921875e-05 2023-01-21 14:04:59.656609: step: 1976/529, loss: 0.0005737662431783974 2023-01-21 14:05:00.858606: step: 1980/529, loss: 0.0033237459138035774 2023-01-21 14:05:02.020050: step: 1984/529, loss: 6.27517729299143e-05 2023-01-21 14:05:03.273052: step: 1988/529, loss: 0.05934343487024307 2023-01-21 14:05:04.436287: step: 1992/529, loss: 4.76837158203125e-07 2023-01-21 14:05:05.614056: step: 1996/529, loss: 0.00024776457576081157 2023-01-21 14:05:06.802783: step: 2000/529, loss: 0.017863083630800247 2023-01-21 14:05:07.984996: step: 2004/529, loss: 3.933906555175781e-05 2023-01-21 14:05:09.159554: step: 2008/529, loss: 0.32417765259742737 2023-01-21 14:05:10.381552: step: 2012/529, loss: 0.0009929656516760588 2023-01-21 14:05:11.565907: step: 2016/529, loss: 0.007342815399169922 2023-01-21 14:05:12.806587: step: 2020/529, loss: 7.343292236328125e-05 2023-01-21 14:05:13.968089: step: 2024/529, loss: 1.392364538332913e-05 2023-01-21 14:05:15.091108: step: 2028/529, loss: 0.00014572142390534282 2023-01-21 14:05:16.276146: step: 2032/529, loss: 0.006381034851074219 2023-01-21 14:05:17.476686: step: 2036/529, loss: 1.335144588665571e-06 2023-01-21 14:05:18.657992: step: 2040/529, loss: 1.9550323486328125e-05 2023-01-21 14:05:19.884116: step: 2044/529, loss: 0.00033969880314543843 2023-01-21 14:05:21.107217: step: 2048/529, loss: 0.00014381408982444555 2023-01-21 14:05:22.266654: step: 2052/529, loss: 0.00042142870370298624 2023-01-21 14:05:23.477020: step: 2056/529, loss: 0.021132469177246094 2023-01-21 14:05:24.664764: step: 2060/529, loss: 0.06449127197265625 2023-01-21 14:05:25.838823: step: 2064/529, loss: 0.0003295898495707661 2023-01-21 14:05:27.010422: step: 2068/529, loss: 0.0002716064336709678 2023-01-21 14:05:28.193268: step: 2072/529, loss: 0.0024772644974291325 2023-01-21 14:05:29.348937: step: 2076/529, loss: 0.06616067886352539 2023-01-21 14:05:30.555755: step: 2080/529, loss: 0.0008478164672851562 2023-01-21 14:05:31.745845: step: 2084/529, loss: 0.4999699592590332 2023-01-21 14:05:32.963109: step: 2088/529, loss: 9.250640687241685e-06 2023-01-21 14:05:34.162819: step: 2092/529, loss: -1.888275073724799e-05 2023-01-21 14:05:35.351980: step: 2096/529, loss: 2.5844574338407256e-05 2023-01-21 14:05:36.569038: step: 2100/529, loss: 0.004937362857162952 2023-01-21 14:05:37.786979: step: 2104/529, loss: 0.003136062528938055 2023-01-21 14:05:38.994980: step: 2108/529, loss: 0.06406402587890625 2023-01-21 14:05:40.189132: step: 2112/529, loss: 1.1634827387752011e-05 2023-01-21 14:05:41.352992: step: 2116/529, loss: 0.001333046006038785 ================================================== Loss: 0.016 -------------------- Dev: {'event': {'p': 0.6014418125643667, 'r': 0.7776298268974701, 'f1': 0.6782810685249709}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Test: {'event': {'p': 0.6387686387686388, 'r': 0.7956860395446375, 'f1': 0.7086446104589113}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Chinese: {'event': {'p': 0.618421052631579, 'r': 0.8703703703703703, 'f1': 0.723076923076923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Korean: {'event': {'p': 0.5285714285714286, 'r': 0.5873015873015873, 'f1': 0.556390977443609}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Russian: {'event': {'p': 0.4523809523809524, 'r': 0.5277777777777778, 'f1': 0.4871794871794871}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ****************************** 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 14:06:24.999290: step: 4/529, loss: 1.3732910701946821e-05 2023-01-21 14:06:26.174332: step: 8/529, loss: 0.00113086705096066 2023-01-21 14:06:27.386136: step: 12/529, loss: 0.0009518623119220138 2023-01-21 14:06:28.601938: step: 16/529, loss: 0.0012510300148278475 2023-01-21 14:06:29.787171: step: 20/529, loss: 0.0003692627069540322 2023-01-21 14:06:30.965794: step: 24/529, loss: 0.004490328021347523 2023-01-21 14:06:32.151305: step: 28/529, loss: 0.0005512237548828125 2023-01-21 14:06:33.306744: step: 32/529, loss: 2.079009937006049e-05 2023-01-21 14:06:34.460570: step: 36/529, loss: 2.822876012942288e-05 2023-01-21 14:06:35.626004: step: 40/529, loss: 9.059906005859375e-06 2023-01-21 14:06:36.824883: step: 44/529, loss: 0.004665374755859375 2023-01-21 14:06:37.978992: step: 48/529, loss: 4.253387305652723e-05 2023-01-21 14:06:39.128355: step: 52/529, loss: 0.00018796921358443797 2023-01-21 14:06:40.318568: step: 56/529, loss: 3.719329924933845e-06 2023-01-21 14:06:41.523848: step: 60/529, loss: 0.0008264541975222528 2023-01-21 14:06:42.743387: step: 64/529, loss: 0.004531669896095991 2023-01-21 14:06:43.916734: step: 68/529, loss: 0.001506042550317943 2023-01-21 14:06:45.091552: step: 72/529, loss: 0.05139656364917755 2023-01-21 14:06:46.234456: step: 76/529, loss: 0.00015468598576262593 2023-01-21 14:06:47.393014: step: 80/529, loss: 5.245208740234375e-06 2023-01-21 14:06:48.610091: step: 84/529, loss: 0.006305885501205921 2023-01-21 14:06:49.780732: step: 88/529, loss: 0.0002493858337402344 2023-01-21 14:06:50.967979: step: 92/529, loss: 0.00018024446035269648 2023-01-21 14:06:52.214903: step: 96/529, loss: 0.004341077990829945 2023-01-21 14:06:53.395165: step: 100/529, loss: 0.018564321100711823 2023-01-21 14:06:54.574989: step: 104/529, loss: 0.07999381422996521 2023-01-21 14:06:55.773133: step: 108/529, loss: -3.910064151568804e-06 2023-01-21 14:06:56.979968: step: 112/529, loss: 3.0040741876291577e-06 2023-01-21 14:06:58.145908: step: 116/529, loss: 0.00025177001953125 2023-01-21 14:06:59.330759: step: 120/529, loss: 0.00040907858056016266 2023-01-21 14:07:00.563595: step: 124/529, loss: 0.0024687768891453743 2023-01-21 14:07:01.739666: step: 128/529, loss: -2.670288040462765e-06 2023-01-21 14:07:02.980805: step: 132/529, loss: 4.062652442371473e-05 2023-01-21 14:07:04.154258: step: 136/529, loss: 0.00042972565279342234 2023-01-21 14:07:05.339357: step: 140/529, loss: 0.0980004370212555 2023-01-21 14:07:06.536002: step: 144/529, loss: 0.006424331571906805 2023-01-21 14:07:07.724323: step: 148/529, loss: 0.0010761262383311987 2023-01-21 14:07:08.962560: step: 152/529, loss: 7.343292054429185e-06 2023-01-21 14:07:10.146439: step: 156/529, loss: 0.0018783569103106856 2023-01-21 14:07:11.316255: step: 160/529, loss: 0.02445368841290474 2023-01-21 14:07:12.503033: step: 164/529, loss: 0.007605171296745539 2023-01-21 14:07:13.699281: step: 168/529, loss: 4.3487551010912284e-05 2023-01-21 14:07:14.925405: step: 172/529, loss: 0.0003483772452455014 2023-01-21 14:07:16.084534: step: 176/529, loss: 0.00037403107853606343 2023-01-21 14:07:17.277853: step: 180/529, loss: -1.411438006471144e-05 2023-01-21 14:07:18.449298: step: 184/529, loss: 0.00045795441837981343 2023-01-21 14:07:19.621900: step: 188/529, loss: 4.57763690064894e-06 2023-01-21 14:07:20.818329: step: 192/529, loss: 0.0005281448247842491 2023-01-21 14:07:21.975055: step: 196/529, loss: 0.000709152256604284 2023-01-21 14:07:23.145108: step: 200/529, loss: 1.8119812921213452e-06 2023-01-21 14:07:24.314370: step: 204/529, loss: 3.4999848139705136e-05 2023-01-21 14:07:25.504391: step: 208/529, loss: 3.814697265625e-06 2023-01-21 14:07:26.653456: step: 212/529, loss: 0.00015144348435569555 2023-01-21 14:07:27.807385: step: 216/529, loss: 2.098083541568485e-06 2023-01-21 14:07:28.981707: step: 220/529, loss: 2.9182436264818534e-05 2023-01-21 14:07:30.157959: step: 224/529, loss: 0.03857875242829323 2023-01-21 14:07:31.377019: step: 228/529, loss: 0.004447269719094038 2023-01-21 14:07:32.545124: step: 232/529, loss: 0.0024159906897693872 2023-01-21 14:07:33.734736: step: 236/529, loss: 0.0012336730724200606 2023-01-21 14:07:34.951910: step: 240/529, loss: 0.003175163408741355 2023-01-21 14:07:36.121078: step: 244/529, loss: -1.1253356205997989e-05 2023-01-21 14:07:37.323857: step: 248/529, loss: 0.01205539796501398 2023-01-21 14:07:38.567309: step: 252/529, loss: 0.00443878211081028 2023-01-21 14:07:39.773611: step: 256/529, loss: 0.001834773924201727 2023-01-21 14:07:40.992335: step: 260/529, loss: 0.03584964573383331 2023-01-21 14:07:42.220177: step: 264/529, loss: 3.623962811616366e-06 2023-01-21 14:07:43.417611: step: 268/529, loss: 0.10174474120140076 2023-01-21 14:07:44.596093: step: 272/529, loss: 0.00570142874494195 2023-01-21 14:07:45.761076: step: 276/529, loss: 0.00019626619177870452 2023-01-21 14:07:46.976484: step: 280/529, loss: 0.5596234202384949 2023-01-21 14:07:48.151985: step: 284/529, loss: 0.019455909729003906 2023-01-21 14:07:49.330042: step: 288/529, loss: 9.880066500045359e-05 2023-01-21 14:07:50.535676: step: 292/529, loss: 0.0023687363136559725 2023-01-21 14:07:51.757401: step: 296/529, loss: 0.026824951171875 2023-01-21 14:07:52.960458: step: 300/529, loss: 0.0022735595703125 2023-01-21 14:07:54.109147: step: 304/529, loss: 0.0002544402959756553 2023-01-21 14:07:55.298551: step: 308/529, loss: 0.027847671881318092 2023-01-21 14:07:56.494562: step: 312/529, loss: 0.0008834838517941535 2023-01-21 14:07:57.670003: step: 316/529, loss: 0.023713111877441406 2023-01-21 14:07:58.865833: step: 320/529, loss: 0.0009000778663903475 2023-01-21 14:08:00.043797: step: 324/529, loss: 0.005767631810158491 2023-01-21 14:08:01.258035: step: 328/529, loss: 0.04710369184613228 2023-01-21 14:08:02.532227: step: 332/529, loss: 0.0007384300697594881 2023-01-21 14:08:03.694784: step: 336/529, loss: 0.000423240679083392 2023-01-21 14:08:04.920678: step: 340/529, loss: 9.632110595703125e-05 2023-01-21 14:08:06.105628: step: 344/529, loss: 0.05529461055994034 2023-01-21 14:08:07.298076: step: 348/529, loss: 0.00020985603623557836 2023-01-21 14:08:08.488898: step: 352/529, loss: 0.007557106204330921 2023-01-21 14:08:09.644072: step: 356/529, loss: 0.00012722014798782766 2023-01-21 14:08:10.836945: step: 360/529, loss: 0.0001659393310546875 2023-01-21 14:08:12.052012: step: 364/529, loss: 1.4019012269272935e-05 2023-01-21 14:08:13.204932: step: 368/529, loss: 3.623962356869015e-06 2023-01-21 14:08:14.369195: step: 372/529, loss: 0.00018978118896484375 2023-01-21 14:08:15.509108: step: 376/529, loss: 0.23175106942653656 2023-01-21 14:08:16.657378: step: 380/529, loss: 2.9277802241267636e-05 2023-01-21 14:08:17.852885: step: 384/529, loss: -5.722046125811175e-07 2023-01-21 14:08:19.039461: step: 388/529, loss: 0.027511978521943092 2023-01-21 14:08:20.201720: step: 392/529, loss: 4.4441225327318534e-05 2023-01-21 14:08:21.366160: step: 396/529, loss: 0.012281608767807484 2023-01-21 14:08:22.551819: step: 400/529, loss: 1.068115216185106e-05 2023-01-21 14:08:23.755788: step: 404/529, loss: 0.0023773193825036287 2023-01-21 14:08:24.959514: step: 408/529, loss: 0.004778099246323109 2023-01-21 14:08:26.122140: step: 412/529, loss: 2.288818359375e-05 2023-01-21 14:08:27.353613: step: 416/529, loss: 0.03590993955731392 2023-01-21 14:08:28.605043: step: 420/529, loss: 6.847381155239418e-05 2023-01-21 14:08:29.780820: step: 424/529, loss: 0.0026382445357739925 2023-01-21 14:08:31.009857: step: 428/529, loss: 0.018700027838349342 2023-01-21 14:08:32.223792: step: 432/529, loss: 0.004219437018036842 2023-01-21 14:08:33.388120: step: 436/529, loss: 0.007000541780143976 2023-01-21 14:08:34.619261: step: 440/529, loss: 9.965896424546372e-06 2023-01-21 14:08:35.773966: step: 444/529, loss: 3.814697265625e-06 2023-01-21 14:08:36.995343: step: 448/529, loss: 0.005376577842980623 2023-01-21 14:08:38.184045: step: 452/529, loss: 0.0016256332164630294 2023-01-21 14:08:39.385009: step: 456/529, loss: 9.15527380129788e-06 2023-01-21 14:08:40.576381: step: 460/529, loss: 0.00023155211238190532 2023-01-21 14:08:41.786779: step: 464/529, loss: 0.0001810073881642893 2023-01-21 14:08:42.965303: step: 468/529, loss: 0.003034305525943637 2023-01-21 14:08:44.139121: step: 472/529, loss: 0.0027370452880859375 2023-01-21 14:08:45.355686: step: 476/529, loss: 0.010607671923935413 2023-01-21 14:08:46.557474: step: 480/529, loss: 0.00036487579927779734 2023-01-21 14:08:47.753541: step: 484/529, loss: 0.005246544256806374 2023-01-21 14:08:48.935371: step: 488/529, loss: 0.006982612423598766 2023-01-21 14:08:50.130960: step: 492/529, loss: 0.0002725601370912045 2023-01-21 14:08:51.306036: step: 496/529, loss: 0.0001316070556640625 2023-01-21 14:08:52.513705: step: 500/529, loss: 0.0005240440368652344 2023-01-21 14:08:53.693221: step: 504/529, loss: 0.05100107565522194 2023-01-21 14:08:54.901937: step: 508/529, loss: 3.8814545405330136e-05 2023-01-21 14:08:56.079682: step: 512/529, loss: 0.0017307280795648694 2023-01-21 14:08:57.263425: step: 516/529, loss: 0.026123715564608574 2023-01-21 14:08:58.430452: step: 520/529, loss: 0.0002359390346100554 2023-01-21 14:08:59.679551: step: 524/529, loss: 1.9359587895451114e-05 2023-01-21 14:09:00.898769: step: 528/529, loss: 0.04048128426074982 2023-01-21 14:09:02.099250: step: 532/529, loss: 0.0003334045468363911 2023-01-21 14:09:03.252109: step: 536/529, loss: 0.002658843994140625 2023-01-21 14:09:04.473669: step: 540/529, loss: 0.02190380170941353 2023-01-21 14:09:05.670599: step: 544/529, loss: 2.002715973503655e-06 2023-01-21 14:09:06.876819: step: 548/529, loss: 0.005922365467995405 2023-01-21 14:09:08.008647: step: 552/529, loss: 2.7561187380342744e-05 2023-01-21 14:09:09.201891: step: 556/529, loss: 3.0183791750459932e-05 2023-01-21 14:09:10.383692: step: 560/529, loss: 0.031472206115722656 2023-01-21 14:09:11.578642: step: 564/529, loss: 5.950927879894152e-05 2023-01-21 14:09:12.758991: step: 568/529, loss: 9.536743306171047e-08 2023-01-21 14:09:13.922984: step: 572/529, loss: 0.005866241175681353 2023-01-21 14:09:15.081791: step: 576/529, loss: 0.00015716553025413305 2023-01-21 14:09:16.230326: step: 580/529, loss: 0.00015401840209960938 2023-01-21 14:09:17.377598: step: 584/529, loss: 1.468658410885837e-05 2023-01-21 14:09:18.508673: step: 588/529, loss: 0.0015761376125738025 2023-01-21 14:09:19.732981: step: 592/529, loss: 0.00042381288949400187 2023-01-21 14:09:20.953571: step: 596/529, loss: 0.016824722290039062 2023-01-21 14:09:22.174571: step: 600/529, loss: -2.6702880859375e-05 2023-01-21 14:09:23.328107: step: 604/529, loss: 0.009059811010956764 2023-01-21 14:09:24.548549: step: 608/529, loss: 7.624625868629664e-05 2023-01-21 14:09:25.754640: step: 612/529, loss: 0.008725929073989391 2023-01-21 14:09:26.879154: step: 616/529, loss: 6.222724914550781e-05 2023-01-21 14:09:28.091767: step: 620/529, loss: 0.005821752827614546 2023-01-21 14:09:29.314416: step: 624/529, loss: 0.043379951268434525 2023-01-21 14:09:30.498558: step: 628/529, loss: 0.0006338119274005294 2023-01-21 14:09:31.717702: step: 632/529, loss: 0.0015340804820880294 2023-01-21 14:09:32.898293: step: 636/529, loss: 0.0014158248668536544 2023-01-21 14:09:34.064640: step: 640/529, loss: 3.643035961431451e-05 2023-01-21 14:09:35.212156: step: 644/529, loss: -7.057189577608369e-06 2023-01-21 14:09:36.416829: step: 648/529, loss: 0.04231424629688263 2023-01-21 14:09:37.624220: step: 652/529, loss: 0.03902072831988335 2023-01-21 14:09:38.840885: step: 656/529, loss: 0.0033262253273278475 2023-01-21 14:09:40.017304: step: 660/529, loss: 2.8419495720299892e-05 2023-01-21 14:09:41.232323: step: 664/529, loss: 7.724761962890625e-05 2023-01-21 14:09:42.455966: step: 668/529, loss: 0.10927601158618927 2023-01-21 14:09:43.593797: step: 672/529, loss: 0.016883373260498047 2023-01-21 14:09:44.782438: step: 676/529, loss: 0.01273498497903347 2023-01-21 14:09:45.982966: step: 680/529, loss: 0.00022220611572265625 2023-01-21 14:09:47.147703: step: 684/529, loss: -9.5367431640625e-07 2023-01-21 14:09:48.345159: step: 688/529, loss: 0.000442695600213483 2023-01-21 14:09:49.551147: step: 692/529, loss: 5.91278076171875e-05 2023-01-21 14:09:50.735855: step: 696/529, loss: 0.036199573427438736 2023-01-21 14:09:51.907017: step: 700/529, loss: 0.0010829925304278731 2023-01-21 14:09:53.114016: step: 704/529, loss: 1.2445449101505801e-05 2023-01-21 14:09:54.347921: step: 708/529, loss: 0.025235366076231003 2023-01-21 14:09:55.574081: step: 712/529, loss: 0.2874128818511963 2023-01-21 14:09:56.764599: step: 716/529, loss: 0.00248889927752316 2023-01-21 14:09:57.971332: step: 720/529, loss: 0.0007202147971838713 2023-01-21 14:09:59.165117: step: 724/529, loss: 0.011014747433364391 2023-01-21 14:10:00.313890: step: 728/529, loss: 0.0002038955717580393 2023-01-21 14:10:01.493615: step: 732/529, loss: 0.00037536618765443563 2023-01-21 14:10:02.712530: step: 736/529, loss: 0.00024862290592864156 2023-01-21 14:10:03.909540: step: 740/529, loss: 4.6443943574558944e-05 2023-01-21 14:10:05.154125: step: 744/529, loss: 0.04201088100671768 2023-01-21 14:10:06.365890: step: 748/529, loss: 0.00036363600520417094 2023-01-21 14:10:07.564300: step: 752/529, loss: 3.943443516618572e-05 2023-01-21 14:10:08.774713: step: 756/529, loss: 0.0028442381881177425 2023-01-21 14:10:09.958942: step: 760/529, loss: 0.002449417021125555 2023-01-21 14:10:11.175222: step: 764/529, loss: 0.00412063580006361 2023-01-21 14:10:12.345256: step: 768/529, loss: 1.6021729607018642e-05 2023-01-21 14:10:13.540486: step: 772/529, loss: 0.014474011026322842 2023-01-21 14:10:14.748613: step: 776/529, loss: 0.0006295203929767013 2023-01-21 14:10:16.003010: step: 780/529, loss: 0.05057773366570473 2023-01-21 14:10:17.236622: step: 784/529, loss: 0.008190535940229893 2023-01-21 14:10:18.396850: step: 788/529, loss: 0.006695294287055731 2023-01-21 14:10:19.599675: step: 792/529, loss: 7.5292591645848e-05 2023-01-21 14:10:20.767011: step: 796/529, loss: 0.0003414154052734375 2023-01-21 14:10:21.944985: step: 800/529, loss: 0.0004547119315247983 2023-01-21 14:10:23.110612: step: 804/529, loss: 0.00012836456880904734 2023-01-21 14:10:24.303513: step: 808/529, loss: 0.0083503732457757 2023-01-21 14:10:25.443635: step: 812/529, loss: 1.5687943232478574e-05 2023-01-21 14:10:26.608641: step: 816/529, loss: 0.027898598462343216 2023-01-21 14:10:27.815169: step: 820/529, loss: 0.0020931244362145662 2023-01-21 14:10:29.020865: step: 824/529, loss: 0.0005487442249432206 2023-01-21 14:10:30.162527: step: 828/529, loss: 1.4972686585679185e-05 2023-01-21 14:10:31.418676: step: 832/529, loss: 0.0009633064619265497 2023-01-21 14:10:32.625662: step: 836/529, loss: 0.02407073974609375 2023-01-21 14:10:33.798561: step: 840/529, loss: 2.8133392333984375e-05 2023-01-21 14:10:34.993075: step: 844/529, loss: 0.0002899169921875 2023-01-21 14:10:36.195830: step: 848/529, loss: 0.00021638871112372726 2023-01-21 14:10:37.421542: step: 852/529, loss: 0.004264450166374445 2023-01-21 14:10:38.674398: step: 856/529, loss: 0.014208793640136719 2023-01-21 14:10:39.860921: step: 860/529, loss: 0.0017125130398198962 2023-01-21 14:10:41.091192: step: 864/529, loss: 0.0013696671230718493 2023-01-21 14:10:42.288642: step: 868/529, loss: 0.0005676269647665322 2023-01-21 14:10:43.506743: step: 872/529, loss: 0.016567612066864967 2023-01-21 14:10:44.716737: step: 876/529, loss: 0.0013925553066655993 2023-01-21 14:10:45.900611: step: 880/529, loss: 1.4400482541532256e-05 2023-01-21 14:10:47.088125: step: 884/529, loss: 0.0003388405020814389 2023-01-21 14:10:48.247426: step: 888/529, loss: 5.722046125811175e-07 2023-01-21 14:10:49.467652: step: 892/529, loss: 0.0003335952933412045 2023-01-21 14:10:50.663297: step: 896/529, loss: 0.000936317490413785 2023-01-21 14:10:51.842871: step: 900/529, loss: 0.0017990111373364925 2023-01-21 14:10:53.015586: step: 904/529, loss: 0.00029745104257017374 2023-01-21 14:10:54.192834: step: 908/529, loss: 0.08940735459327698 2023-01-21 14:10:55.372698: step: 912/529, loss: -3.52859501617786e-06 2023-01-21 14:10:56.553407: step: 916/529, loss: 0.0001536846102681011 2023-01-21 14:10:57.752303: step: 920/529, loss: -9.72747784544481e-06 2023-01-21 14:10:58.992549: step: 924/529, loss: 0.004598808474838734 2023-01-21 14:11:00.177447: step: 928/529, loss: 0.0011684418423101306 2023-01-21 14:11:01.431573: step: 932/529, loss: -9.5367431640625e-06 2023-01-21 14:11:02.631861: step: 936/529, loss: 8.39233416627394e-06 2023-01-21 14:11:03.794554: step: 940/529, loss: 0.0036809921730309725 2023-01-21 14:11:04.999207: step: 944/529, loss: 0.00024070740619208664 2023-01-21 14:11:06.188080: step: 948/529, loss: 0.0005346298567019403 2023-01-21 14:11:07.382070: step: 952/529, loss: 0.0014528273604810238 2023-01-21 14:11:08.559337: step: 956/529, loss: 0.0001371383696096018 2023-01-21 14:11:09.727896: step: 960/529, loss: 1.201629675051663e-05 2023-01-21 14:11:10.915564: step: 964/529, loss: 0.0548831969499588 2023-01-21 14:11:12.125216: step: 968/529, loss: 0.0005851269233971834 2023-01-21 14:11:13.286433: step: 972/529, loss: 0.013447761535644531 2023-01-21 14:11:14.459438: step: 976/529, loss: -1.4495848517981358e-05 2023-01-21 14:11:15.660591: step: 980/529, loss: 0.022922707721590996 2023-01-21 14:11:16.826937: step: 984/529, loss: 1.983642505365424e-05 2023-01-21 14:11:17.998796: step: 988/529, loss: 0.0060821534134447575 2023-01-21 14:11:19.233322: step: 992/529, loss: 0.000286102294921875 2023-01-21 14:11:20.440393: step: 996/529, loss: 1.907349087559851e-07 2023-01-21 14:11:21.631095: step: 1000/529, loss: 0.0013631820911541581 2023-01-21 14:11:22.782414: step: 1004/529, loss: 0.0034191131126135588 2023-01-21 14:11:23.947622: step: 1008/529, loss: 0.08127987384796143 2023-01-21 14:11:25.148131: step: 1012/529, loss: 0.0008543491712771356 2023-01-21 14:11:26.342925: step: 1016/529, loss: 0.0005819320795126259 2023-01-21 14:11:27.508787: step: 1020/529, loss: 0.0015231609577313066 2023-01-21 14:11:28.701566: step: 1024/529, loss: 0.028425026684999466 2023-01-21 14:11:29.919538: step: 1028/529, loss: 0.007377576548606157 2023-01-21 14:11:31.073665: step: 1032/529, loss: 0.0002226829674327746 2023-01-21 14:11:32.252299: step: 1036/529, loss: 0.0004733086097985506 2023-01-21 14:11:33.444300: step: 1040/529, loss: 0.0069751739501953125 2023-01-21 14:11:34.670294: step: 1044/529, loss: 0.0013044357765465975 2023-01-21 14:11:35.843281: step: 1048/529, loss: 0.008054542355239391 2023-01-21 14:11:37.013842: step: 1052/529, loss: 0.0001636505185160786 2023-01-21 14:11:38.215963: step: 1056/529, loss: 0.0007921219221316278 2023-01-21 14:11:39.361808: step: 1060/529, loss: 0.0005320549244061112 2023-01-21 14:11:40.536274: step: 1064/529, loss: 0.0011274099815636873 2023-01-21 14:11:41.702845: step: 1068/529, loss: 0.010270977392792702 2023-01-21 14:11:42.880815: step: 1072/529, loss: 1.869201696536038e-05 2023-01-21 14:11:44.058307: step: 1076/529, loss: 0.0002075195370707661 2023-01-21 14:11:45.249170: step: 1080/529, loss: 0.7307483553886414 2023-01-21 14:11:46.437163: step: 1084/529, loss: 0.010613000951707363 2023-01-21 14:11:47.635733: step: 1088/529, loss: 8.621215965831652e-05 2023-01-21 14:11:48.805146: step: 1092/529, loss: 0.007846069522202015 2023-01-21 14:11:49.949444: step: 1096/529, loss: 6.523132469737902e-05 2023-01-21 14:11:51.186001: step: 1100/529, loss: 8.931160846259445e-05 2023-01-21 14:11:52.349893: step: 1104/529, loss: 0.0005597114795818925 2023-01-21 14:11:53.550181: step: 1108/529, loss: 2.4795534045551904e-05 2023-01-21 14:11:54.722255: step: 1112/529, loss: 0.00027465823222883046 2023-01-21 14:11:55.887380: step: 1116/529, loss: 3.9529801142634824e-05 2023-01-21 14:11:57.097505: step: 1120/529, loss: 0.007442093454301357 2023-01-21 14:11:58.271535: step: 1124/529, loss: 0.00012378694373182952 2023-01-21 14:11:59.453168: step: 1128/529, loss: 0.028606891632080078 2023-01-21 14:12:00.616387: step: 1132/529, loss: 0.01957225799560547 2023-01-21 14:12:01.789640: step: 1136/529, loss: 0.007449245546013117 2023-01-21 14:12:02.955023: step: 1140/529, loss: 4.9591064453125e-05 2023-01-21 14:12:04.151935: step: 1144/529, loss: 5.254745337879285e-05 2023-01-21 14:12:05.324969: step: 1148/529, loss: 0.0027244091033935547 2023-01-21 14:12:06.624437: step: 1152/529, loss: 0.00439033517614007 2023-01-21 14:12:07.785082: step: 1156/529, loss: 6.294250852079131e-06 2023-01-21 14:12:08.985827: step: 1160/529, loss: 0.000913429306820035 2023-01-21 14:12:10.195285: step: 1164/529, loss: 0.009033155627548695 2023-01-21 14:12:11.394994: step: 1168/529, loss: 0.00177001953125 2023-01-21 14:12:12.568250: step: 1172/529, loss: 0.00033798220101743937 2023-01-21 14:12:13.761791: step: 1176/529, loss: 0.002139186952263117 2023-01-21 14:12:14.980849: step: 1180/529, loss: 0.0073337554931640625 2023-01-21 14:12:16.220239: step: 1184/529, loss: 0.0008852005121298134 2023-01-21 14:12:17.397099: step: 1188/529, loss: 0.0025569917634129524 2023-01-21 14:12:18.608521: step: 1192/529, loss: 0.00018558502779342234 2023-01-21 14:12:19.782049: step: 1196/529, loss: 0.0025222781114280224 2023-01-21 14:12:20.971077: step: 1200/529, loss: 0.002098608063533902 2023-01-21 14:12:22.172397: step: 1204/529, loss: 0.00010204315185546875 2023-01-21 14:12:23.419144: step: 1208/529, loss: 0.002108001848682761 2023-01-21 14:12:24.614879: step: 1212/529, loss: 0.00040225981501862407 2023-01-21 14:12:25.832492: step: 1216/529, loss: 0.0001434326113667339 2023-01-21 14:12:27.007519: step: 1220/529, loss: 2.765655608527595e-06 2023-01-21 14:12:28.195315: step: 1224/529, loss: 0.0033040824346244335 2023-01-21 14:12:29.390668: step: 1228/529, loss: -1.287460349885805e-06 2023-01-21 14:12:30.594668: step: 1232/529, loss: 6.12258882028982e-05 2023-01-21 14:12:31.797920: step: 1236/529, loss: 0.0016671180492267013 2023-01-21 14:12:33.052006: step: 1240/529, loss: 0.628967821598053 2023-01-21 14:12:34.255482: step: 1244/529, loss: 8.335113670909777e-05 2023-01-21 14:12:35.459206: step: 1248/529, loss: 2.28881845032447e-06 2023-01-21 14:12:36.663877: step: 1252/529, loss: 0.0017341615166515112 2023-01-21 14:12:37.807761: step: 1256/529, loss: 0.05816249921917915 2023-01-21 14:12:39.000460: step: 1260/529, loss: 0.00010776520502986386 2023-01-21 14:12:40.166204: step: 1264/529, loss: 0.01769847795367241 2023-01-21 14:12:41.348281: step: 1268/529, loss: 0.06173725426197052 2023-01-21 14:12:42.542536: step: 1272/529, loss: 0.0031499862670898438 2023-01-21 14:12:43.772008: step: 1276/529, loss: 0.09998436272144318 2023-01-21 14:12:45.000118: step: 1280/529, loss: 4.253387305652723e-05 2023-01-21 14:12:46.215951: step: 1284/529, loss: 0.0032931803725659847 2023-01-21 14:12:47.387555: step: 1288/529, loss: 0.002288436982780695 2023-01-21 14:12:48.595993: step: 1292/529, loss: 1.1157990229548886e-05 2023-01-21 14:12:49.745717: step: 1296/529, loss: 3.814697322468419e-07 2023-01-21 14:12:50.944313: step: 1300/529, loss: 8.668899681651965e-05 2023-01-21 14:12:52.137654: step: 1304/529, loss: 0.002167224884033203 2023-01-21 14:12:53.304128: step: 1308/529, loss: 0.05486641079187393 2023-01-21 14:12:54.472907: step: 1312/529, loss: 0.0024099352303892374 2023-01-21 14:12:55.700227: step: 1316/529, loss: 0.020002080127596855 2023-01-21 14:12:56.904450: step: 1320/529, loss: 4.19616708313697e-06 2023-01-21 14:12:58.086304: step: 1324/529, loss: 0.010896253399550915 2023-01-21 14:12:59.344184: step: 1328/529, loss: 0.05603199079632759 2023-01-21 14:13:00.543731: step: 1332/529, loss: 0.00064849853515625 2023-01-21 14:13:01.768675: step: 1336/529, loss: 0.0005826950073242188 2023-01-21 14:13:02.987386: step: 1340/529, loss: 0.04306488111615181 2023-01-21 14:13:04.169608: step: 1344/529, loss: 0.0001544952392578125 2023-01-21 14:13:05.358127: step: 1348/529, loss: 0.07240915298461914 2023-01-21 14:13:06.531527: step: 1352/529, loss: 0.0008665084606036544 2023-01-21 14:13:07.712648: step: 1356/529, loss: 0.10205211490392685 2023-01-21 14:13:08.937550: step: 1360/529, loss: 0.3231346011161804 2023-01-21 14:13:10.110545: step: 1364/529, loss: 0.038396645337343216 2023-01-21 14:13:11.296997: step: 1368/529, loss: 0.10998497158288956 2023-01-21 14:13:12.470082: step: 1372/529, loss: 0.034264590591192245 2023-01-21 14:13:13.664404: step: 1376/529, loss: 0.0009187698597088456 2023-01-21 14:13:14.825851: step: 1380/529, loss: 0.013098526746034622 2023-01-21 14:13:16.021267: step: 1384/529, loss: 0.0007959366193972528 2023-01-21 14:13:17.270751: step: 1388/529, loss: 7.915496098576114e-06 2023-01-21 14:13:18.442363: step: 1392/529, loss: 0.0016950607532635331 2023-01-21 14:13:19.657481: step: 1396/529, loss: 0.009640026837587357 2023-01-21 14:13:20.833738: step: 1400/529, loss: 0.03501405939459801 2023-01-21 14:13:22.026805: step: 1404/529, loss: 0.004276084713637829 2023-01-21 14:13:23.174344: step: 1408/529, loss: 0.005029868800193071 2023-01-21 14:13:24.348493: step: 1412/529, loss: 0.0031041146721690893 2023-01-21 14:13:25.520008: step: 1416/529, loss: 5.054474058852065e-06 2023-01-21 14:13:26.739703: step: 1420/529, loss: 4.00543194700731e-06 2023-01-21 14:13:27.919307: step: 1424/529, loss: 5.7029727031476796e-05 2023-01-21 14:13:29.117745: step: 1428/529, loss: 2.8228761948412284e-05 2023-01-21 14:13:30.306842: step: 1432/529, loss: 0.06308785080909729 2023-01-21 14:13:31.473997: step: 1436/529, loss: 0.0008561611175537109 2023-01-21 14:13:32.696589: step: 1440/529, loss: 2.4700164431123994e-05 2023-01-21 14:13:33.872715: step: 1444/529, loss: 0.003226661588996649 2023-01-21 14:13:35.029246: step: 1448/529, loss: 0.004049873445183039 2023-01-21 14:13:36.185673: step: 1452/529, loss: 0.0008733749273233116 2023-01-21 14:13:37.374445: step: 1456/529, loss: 0.00016326905461028218 2023-01-21 14:13:38.572821: step: 1460/529, loss: 0.00029439927311614156 2023-01-21 14:13:39.847794: step: 1464/529, loss: 0.003342247102409601 2023-01-21 14:13:41.013015: step: 1468/529, loss: 0.023175524547696114 2023-01-21 14:13:42.192159: step: 1472/529, loss: 0.019219398498535156 2023-01-21 14:13:43.356164: step: 1476/529, loss: 0.024181175976991653 2023-01-21 14:13:44.520272: step: 1480/529, loss: 0.00043621062650345266 2023-01-21 14:13:45.724327: step: 1484/529, loss: 0.0004558563232421875 2023-01-21 14:13:46.889117: step: 1488/529, loss: 6.632805161643773e-05 2023-01-21 14:13:48.096474: step: 1492/529, loss: 0.0022317885886877775 2023-01-21 14:13:49.290440: step: 1496/529, loss: 0.0036783218383789062 2023-01-21 14:13:50.500787: step: 1500/529, loss: 0.33480775356292725 2023-01-21 14:13:51.671338: step: 1504/529, loss: 5.722046125811175e-07 2023-01-21 14:13:52.879402: step: 1508/529, loss: 0.0043860916048288345 2023-01-21 14:13:54.052799: step: 1512/529, loss: 2.2506716049974784e-05 2023-01-21 14:13:55.253175: step: 1516/529, loss: 0.0005534648662433028 2023-01-21 14:13:56.444327: step: 1520/529, loss: 0.0030406953301280737 2023-01-21 14:13:57.636618: step: 1524/529, loss: -2.098083541568485e-06 2023-01-21 14:13:58.891485: step: 1528/529, loss: 0.0005517959361895919 2023-01-21 14:14:00.140929: step: 1532/529, loss: 0.056876279413700104 2023-01-21 14:14:01.345437: step: 1536/529, loss: 0.008852005004882812 2023-01-21 14:14:02.522515: step: 1540/529, loss: 4.76837158203125e-06 2023-01-21 14:14:03.726379: step: 1544/529, loss: 0.0021227120887488127 2023-01-21 14:14:04.916969: step: 1548/529, loss: 0.0013967513805255294 2023-01-21 14:14:06.112791: step: 1552/529, loss: 9.71794142969884e-05 2023-01-21 14:14:07.299679: step: 1556/529, loss: 0.03422565385699272 2023-01-21 14:14:08.518340: step: 1560/529, loss: 0.028199482709169388 2023-01-21 14:14:09.733891: step: 1564/529, loss: 0.00015182494826149195 2023-01-21 14:14:10.901270: step: 1568/529, loss: 0.038149263709783554 2023-01-21 14:14:12.108210: step: 1572/529, loss: 0.00139789585955441 2023-01-21 14:14:13.302748: step: 1576/529, loss: 0.00022726059250999242 2023-01-21 14:14:14.483383: step: 1580/529, loss: 0.0001848220854299143 2023-01-21 14:14:15.707339: step: 1584/529, loss: 0.00015487671771552414 2023-01-21 14:14:16.937197: step: 1588/529, loss: 0.00145721435546875 2023-01-21 14:14:18.167855: step: 1592/529, loss: 0.0006259918445721269 2023-01-21 14:14:19.374780: step: 1596/529, loss: 5.14984139954322e-06 2023-01-21 14:14:20.598577: step: 1600/529, loss: 0.0008106232271529734 2023-01-21 14:14:21.780225: step: 1604/529, loss: -2.1076202756376006e-05 2023-01-21 14:14:22.925074: step: 1608/529, loss: 0.0005048752063885331 2023-01-21 14:14:24.120885: step: 1612/529, loss: 0.0017232894897460938 2023-01-21 14:14:25.337828: step: 1616/529, loss: 0.0001220703125 2023-01-21 14:14:26.522960: step: 1620/529, loss: 0.0008788108825683594 2023-01-21 14:14:27.748293: step: 1624/529, loss: 0.0016613006591796875 2023-01-21 14:14:28.949898: step: 1628/529, loss: 0.015953348949551582 2023-01-21 14:14:30.116076: step: 1632/529, loss: 0.00039677618769928813 2023-01-21 14:14:31.315069: step: 1636/529, loss: 0.00019750595674850047 2023-01-21 14:14:32.514744: step: 1640/529, loss: 7.686615572310984e-05 2023-01-21 14:14:33.695707: step: 1644/529, loss: 3.319978759463993e-06 2023-01-21 14:14:34.846082: step: 1648/529, loss: 0.0036615373101085424 2023-01-21 14:14:36.027615: step: 1652/529, loss: 0.006658363621681929 2023-01-21 14:14:37.212067: step: 1656/529, loss: 0.005681514739990234 2023-01-21 14:14:38.456777: step: 1660/529, loss: 0.0007167815929278731 2023-01-21 14:14:39.633003: step: 1664/529, loss: 4.6539309551008046e-05 2023-01-21 14:14:40.784811: step: 1668/529, loss: -5.7220458984375e-06 2023-01-21 14:14:41.984979: step: 1672/529, loss: 0.0235443115234375 2023-01-21 14:14:43.192998: step: 1676/529, loss: 0.002525901887565851 2023-01-21 14:14:44.380568: step: 1680/529, loss: 2.002716064453125e-05 2023-01-21 14:14:45.554509: step: 1684/529, loss: 3.3664702641544864e-05 2023-01-21 14:14:46.684798: step: 1688/529, loss: 0.0007604121929034591 2023-01-21 14:14:47.876716: step: 1692/529, loss: 0.004181290045380592 2023-01-21 14:14:49.079562: step: 1696/529, loss: 0.004848528187721968 2023-01-21 14:14:50.268637: step: 1700/529, loss: 0.006644248962402344 2023-01-21 14:14:51.457636: step: 1704/529, loss: 0.014876174740493298 2023-01-21 14:14:52.640846: step: 1708/529, loss: 0.06999950855970383 2023-01-21 14:14:53.802867: step: 1712/529, loss: 0.0003153800789732486 2023-01-21 14:14:55.019042: step: 1716/529, loss: 0.0019784928299486637 2023-01-21 14:14:56.222488: step: 1720/529, loss: 7.07626313669607e-05 2023-01-21 14:14:57.401385: step: 1724/529, loss: 0.02034597471356392 2023-01-21 14:14:58.591993: step: 1728/529, loss: 0.00048065188457258046 2023-01-21 14:14:59.805822: step: 1732/529, loss: 1.5853469371795654 2023-01-21 14:15:00.998620: step: 1736/529, loss: 0.010175133123993874 2023-01-21 14:15:02.187345: step: 1740/529, loss: 0.0001585006684763357 2023-01-21 14:15:03.379023: step: 1744/529, loss: 0.0015062332386150956 2023-01-21 14:15:04.566398: step: 1748/529, loss: 1.392364538332913e-05 2023-01-21 14:15:05.754310: step: 1752/529, loss: 0.04295406490564346 2023-01-21 14:15:07.004118: step: 1756/529, loss: 0.00018281936354469508 2023-01-21 14:15:08.197554: step: 1760/529, loss: 3.719329924933845e-06 2023-01-21 14:15:09.415343: step: 1764/529, loss: 0.007261753082275391 2023-01-21 14:15:10.621711: step: 1768/529, loss: 0.005169677548110485 2023-01-21 14:15:11.776461: step: 1772/529, loss: 0.0004545212141238153 2023-01-21 14:15:13.015462: step: 1776/529, loss: 0.008730506524443626 2023-01-21 14:15:14.231044: step: 1780/529, loss: 3.43322744811303e-06 2023-01-21 14:15:15.424263: step: 1784/529, loss: 0.0007745742914266884 2023-01-21 14:15:16.616961: step: 1788/529, loss: 2.632141149661038e-05 2023-01-21 14:15:17.791764: step: 1792/529, loss: 0.002716904738917947 2023-01-21 14:15:18.952630: step: 1796/529, loss: 0.00016779899306129664 2023-01-21 14:15:20.109213: step: 1800/529, loss: 3.0708310077898204e-05 2023-01-21 14:15:21.253257: step: 1804/529, loss: 0.0017543792491778731 2023-01-21 14:15:22.442286: step: 1808/529, loss: 0.049134161323308945 2023-01-21 14:15:23.681943: step: 1812/529, loss: 0.2432354986667633 2023-01-21 14:15:24.869897: step: 1816/529, loss: 0.00014219283184502274 2023-01-21 14:15:26.044139: step: 1820/529, loss: 0.03889045864343643 2023-01-21 14:15:27.211960: step: 1824/529, loss: 0.00013351438974495977 2023-01-21 14:15:28.346056: step: 1828/529, loss: 0.08385400474071503 2023-01-21 14:15:29.574963: step: 1832/529, loss: 0.007090759463608265 2023-01-21 14:15:30.759953: step: 1836/529, loss: 0.0006044387700967491 2023-01-21 14:15:31.960481: step: 1840/529, loss: 0.00013494491577148438 2023-01-21 14:15:33.134531: step: 1844/529, loss: 0.0032146451994776726 2023-01-21 14:15:34.292689: step: 1848/529, loss: 9.183883958030492e-05 2023-01-21 14:15:35.452142: step: 1852/529, loss: -5.674362000718247e-06 2023-01-21 14:15:36.685859: step: 1856/529, loss: 0.00024547576322220266 2023-01-21 14:15:37.900544: step: 1860/529, loss: 0.00048656464787200093 2023-01-21 14:15:39.086349: step: 1864/529, loss: 0.0027584077324718237 2023-01-21 14:15:40.262877: step: 1868/529, loss: 0.0001434326113667339 2023-01-21 14:15:41.446032: step: 1872/529, loss: 0.0009277344215661287 2023-01-21 14:15:42.679892: step: 1876/529, loss: 0.006348991766571999 2023-01-21 14:15:43.863946: step: 1880/529, loss: 0.0003717422659974545 2023-01-21 14:15:45.078855: step: 1884/529, loss: 0.036380767822265625 2023-01-21 14:15:46.235739: step: 1888/529, loss: 0.031036807224154472 2023-01-21 14:15:47.441705: step: 1892/529, loss: 0.0053983209654688835 2023-01-21 14:15:48.634536: step: 1896/529, loss: 3.0708310077898204e-05 2023-01-21 14:15:49.837185: step: 1900/529, loss: 0.00629768380895257 2023-01-21 14:15:51.014546: step: 1904/529, loss: 2.555847095209174e-05 2023-01-21 14:15:52.160992: step: 1908/529, loss: 0.0008892059559002519 2023-01-21 14:15:53.403863: step: 1912/529, loss: 0.0058303833939135075 2023-01-21 14:15:54.567388: step: 1916/529, loss: 6.599426706088707e-05 2023-01-21 14:15:55.722500: step: 1920/529, loss: 0.026189804077148438 2023-01-21 14:15:56.909638: step: 1924/529, loss: 0.01587228663265705 2023-01-21 14:15:58.122422: step: 1928/529, loss: 0.0031957626342773438 2023-01-21 14:15:59.294700: step: 1932/529, loss: -1.430511474609375e-06 2023-01-21 14:16:00.543901: step: 1936/529, loss: 0.004824829287827015 2023-01-21 14:16:01.746237: step: 1940/529, loss: 5.054474240751006e-05 2023-01-21 14:16:02.976456: step: 1944/529, loss: 0.03985634073615074 2023-01-21 14:16:04.191373: step: 1948/529, loss: 0.00010471343557583168 2023-01-21 14:16:05.432595: step: 1952/529, loss: 0.012557410635054111 2023-01-21 14:16:06.626870: step: 1956/529, loss: 0.01804809644818306 2023-01-21 14:16:07.852666: step: 1960/529, loss: 0.004710960201919079 2023-01-21 14:16:09.004125: step: 1964/529, loss: 2.555847095209174e-05 2023-01-21 14:16:10.203535: step: 1968/529, loss: 0.03960246965289116 2023-01-21 14:16:11.403102: step: 1972/529, loss: 0.0014725684886798263 2023-01-21 14:16:12.560316: step: 1976/529, loss: 2.956390289909905e-06 2023-01-21 14:16:13.726999: step: 1980/529, loss: 4.482269287109375e-05 2023-01-21 14:16:14.898081: step: 1984/529, loss: 0.00027980803861282766 2023-01-21 14:16:16.079256: step: 1988/529, loss: 0.0002525329473428428 2023-01-21 14:16:17.252617: step: 1992/529, loss: 1.8978118532686494e-05 2023-01-21 14:16:18.463916: step: 1996/529, loss: 0.005339432042092085 2023-01-21 14:16:19.634419: step: 2000/529, loss: 0.0010305405594408512 2023-01-21 14:16:20.826190: step: 2004/529, loss: 1.0681153071345761e-05 2023-01-21 14:16:21.997950: step: 2008/529, loss: -1.8119812921213452e-06 2023-01-21 14:16:23.179361: step: 2012/529, loss: 3.42369094141759e-05 2023-01-21 14:16:24.344930: step: 2016/529, loss: 0.0003417968691792339 2023-01-21 14:16:25.552262: step: 2020/529, loss: 0.0011999130947515368 2023-01-21 14:16:26.718632: step: 2024/529, loss: 0.06895418465137482 2023-01-21 14:16:27.906904: step: 2028/529, loss: 1.8596649169921875e-05 2023-01-21 14:16:29.075509: step: 2032/529, loss: 0.05424537509679794 2023-01-21 14:16:30.230451: step: 2036/529, loss: 0.0005564690218307078 2023-01-21 14:16:31.407804: step: 2040/529, loss: 0.0004893303266726434 2023-01-21 14:16:32.590041: step: 2044/529, loss: 0.0006778478855267167 2023-01-21 14:16:33.800091: step: 2048/529, loss: 0.00010166168794967234 2023-01-21 14:16:34.991291: step: 2052/529, loss: 0.033219531178474426 2023-01-21 14:16:36.163983: step: 2056/529, loss: 0.0014505386352539062 2023-01-21 14:16:37.388816: step: 2060/529, loss: 3.871917579090223e-05 2023-01-21 14:16:38.616058: step: 2064/529, loss: 0.011248588562011719 2023-01-21 14:16:39.815039: step: 2068/529, loss: -1.411438006471144e-05 2023-01-21 14:16:40.980452: step: 2072/529, loss: 0.006740093231201172 2023-01-21 14:16:42.178537: step: 2076/529, loss: 6.198883056640625e-06 2023-01-21 14:16:43.382751: step: 2080/529, loss: 3.24249267578125e-05 2023-01-21 14:16:44.608724: step: 2084/529, loss: 0.17038460075855255 2023-01-21 14:16:45.822358: step: 2088/529, loss: 1.277923547604587e-05 2023-01-21 14:16:47.042928: step: 2092/529, loss: 0.006978320889174938 2023-01-21 14:16:48.231731: step: 2096/529, loss: 3.185272362316027e-05 2023-01-21 14:16:49.447376: step: 2100/529, loss: 0.0004268646298442036 2023-01-21 14:16:50.622637: step: 2104/529, loss: 0.03451395407319069 2023-01-21 14:16:51.848412: step: 2108/529, loss: 0.023598290979862213 2023-01-21 14:16:53.045009: step: 2112/529, loss: 0.443160742521286 2023-01-21 14:16:54.231365: step: 2116/529, loss: 0.0002507686731405556 ================================================== Loss: 0.019 -------------------- Dev: {'event': {'p': 0.5977710233029382, 'r': 0.7856191744340879, 'f1': 0.6789413118527042}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Test: {'event': {'p': 0.6216466234967623, 'r': 0.805272618334332, 'f1': 0.7016444792482381}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Chinese: {'event': {'p': 0.5625, 'r': 0.8333333333333334, 'f1': 0.6716417910447761}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Korean: {'event': {'p': 0.49295774647887325, 'r': 0.5555555555555556, 'f1': 0.5223880597014925}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Russian: {'event': {'p': 0.40816326530612246, 'r': 0.5555555555555556, 'f1': 0.47058823529411764}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6463560334528077, 'r': 0.7203728362183754, 'f1': 0.6813602015113349}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6293375394321766, 'r': 0.7171959257040144, 'f1': 0.6704004480537665}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.7142857142857143, 'r': 0.8333333333333334, 'f1': 0.7692307692307692}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.5911764705882353, 'r': 0.8029294274300932, 'f1': 0.6809712027103332}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Korean: {'event': {'p': 0.6277407054337464, 'r': 0.7890952666267226, 'f1': 0.6992301566233077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} -------------------- Eng Dev for Russian: {'event': {'p': 0.6682464454976303, 'r': 0.7509986684420772, 'f1': 0.7072100313479625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Eng Test for Russian: {'event': {'p': 0.6650968079539508, 'r': 0.7615338526063511, 'f1': 0.7100558659217877}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Sample Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17}