Command that produces this log: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 ---------------------------------------------------------------------------------------------------- > trainable params: >>> xlmr.embeddings.word_embeddings.weight: torch.Size([250002, 1024]) >>> xlmr.embeddings.position_embeddings.weight: torch.Size([514, 1024]) >>> xlmr.embeddings.token_type_embeddings.weight: torch.Size([1, 1024]) >>> xlmr.embeddings.LayerNorm.weight: torch.Size([1024]) >>> xlmr.embeddings.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.0.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.0.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.0.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.0.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.1.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.1.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.1.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.1.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.2.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.2.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.2.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.2.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.3.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.3.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.3.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.3.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.4.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.4.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.4.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.4.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.5.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.5.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.5.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.5.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.6.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.6.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.6.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.6.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.7.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.7.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.7.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.7.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.8.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.8.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.8.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.8.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.9.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.9.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.9.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.9.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.10.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.10.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.10.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.10.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.11.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.11.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.11.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.11.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.12.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.12.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.12.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.12.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.13.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.13.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.13.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.13.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.14.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.14.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.14.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.14.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.15.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.15.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.15.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.15.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.16.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.16.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.16.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.16.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.17.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.17.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.17.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.17.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.18.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.18.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.18.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.18.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.19.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.19.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.19.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.19.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.20.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.20.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.20.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.20.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.21.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.21.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.21.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.21.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.22.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.22.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.22.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.22.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.23.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.23.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.23.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.23.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.pooler.dense.weight: torch.Size([1024, 1024]) >>> xlmr.pooler.dense.bias: torch.Size([1024]) >>> trigger_label_ffn.layers.0.weight: torch.Size([450, 1024]) >>> trigger_label_ffn.layers.0.bias: torch.Size([450]) >>> trigger_label_ffn.layers.1.weight: torch.Size([233, 450]) >>> trigger_label_ffn.layers.1.bias: torch.Size([233]) >>> trigger_crf.transition: torch.Size([235, 235]) n_trainable_params: 560511990, n_nontrainable_params: 0 ---------------------------------------------------------------------------------------------------- ****************************** Epoch: 0 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:25:06.283297: step: 4/529, loss: 3.621532440185547 2023-01-21 08:25:07.411454: step: 8/529, loss: 13.542070388793945 2023-01-21 08:25:08.542650: step: 12/529, loss: 3.117363929748535 2023-01-21 08:25:09.688998: step: 16/529, loss: 4.673288822174072 2023-01-21 08:25:10.863841: step: 20/529, loss: 5.232785224914551 2023-01-21 08:25:12.035812: step: 24/529, loss: 12.63650894165039 2023-01-21 08:25:13.135545: step: 28/529, loss: 2.595937728881836 2023-01-21 08:25:14.266512: step: 32/529, loss: 14.397211074829102 2023-01-21 08:25:15.400051: step: 36/529, loss: 16.40961265563965 2023-01-21 08:25:16.524353: step: 40/529, loss: 16.237842559814453 2023-01-21 08:25:17.646245: step: 44/529, loss: 9.762625694274902 2023-01-21 08:25:18.795819: step: 48/529, loss: 19.22272300720215 2023-01-21 08:25:19.910521: step: 52/529, loss: 14.365507125854492 2023-01-21 08:25:21.040367: step: 56/529, loss: 10.777974128723145 2023-01-21 08:25:22.195003: step: 60/529, loss: 21.444992065429688 2023-01-21 08:25:23.302621: step: 64/529, loss: 2.9161853790283203 2023-01-21 08:25:24.449986: step: 68/529, loss: 12.690162658691406 2023-01-21 08:25:25.550343: step: 72/529, loss: 4.3046112060546875 2023-01-21 08:25:26.674768: step: 76/529, loss: 2.0277481079101562 2023-01-21 08:25:27.793772: step: 80/529, loss: 13.829889297485352 2023-01-21 08:25:28.920236: step: 84/529, loss: 16.591564178466797 2023-01-21 08:25:30.057621: step: 88/529, loss: 3.896984338760376 2023-01-21 08:25:31.187547: step: 92/529, loss: 25.74120330810547 2023-01-21 08:25:32.322717: step: 96/529, loss: 2.3117856979370117 2023-01-21 08:25:33.439180: step: 100/529, loss: 2.5821826457977295 2023-01-21 08:25:34.581796: step: 104/529, loss: 3.8382575511932373 2023-01-21 08:25:35.714989: step: 108/529, loss: 4.201536178588867 2023-01-21 08:25:36.878670: step: 112/529, loss: 3.37551212310791 2023-01-21 08:25:38.030951: step: 116/529, loss: 11.776656150817871 2023-01-21 08:25:39.140648: step: 120/529, loss: 7.709282875061035 2023-01-21 08:25:40.288332: step: 124/529, loss: 33.39939498901367 2023-01-21 08:25:41.433323: step: 128/529, loss: 14.534605026245117 2023-01-21 08:25:42.601950: step: 132/529, loss: 3.7271571159362793 2023-01-21 08:25:43.740721: step: 136/529, loss: 3.990236520767212 2023-01-21 08:25:44.883732: step: 140/529, loss: 2.5393314361572266 2023-01-21 08:25:46.004099: step: 144/529, loss: 3.369417428970337 2023-01-21 08:25:47.114627: step: 148/529, loss: 17.484071731567383 2023-01-21 08:25:48.269315: step: 152/529, loss: 17.277616500854492 2023-01-21 08:25:49.405916: step: 156/529, loss: 15.19094467163086 2023-01-21 08:25:50.553846: step: 160/529, loss: 3.8732707500457764 2023-01-21 08:25:51.744900: step: 164/529, loss: 18.729747772216797 2023-01-21 08:25:52.876804: step: 168/529, loss: 7.309647083282471 2023-01-21 08:25:53.991837: step: 172/529, loss: 12.084159851074219 2023-01-21 08:25:55.138903: step: 176/529, loss: 17.671701431274414 2023-01-21 08:25:56.332458: step: 180/529, loss: 3.7090930938720703 2023-01-21 08:25:57.484252: step: 184/529, loss: 7.771060943603516 2023-01-21 08:25:58.622489: step: 188/529, loss: 11.59591293334961 2023-01-21 08:25:59.779883: step: 192/529, loss: 17.572521209716797 2023-01-21 08:26:00.900632: step: 196/529, loss: 12.50350570678711 2023-01-21 08:26:02.017503: step: 200/529, loss: 2.356724500656128 2023-01-21 08:26:03.159061: step: 204/529, loss: 3.408334493637085 2023-01-21 08:26:04.334705: step: 208/529, loss: 7.889789581298828 2023-01-21 08:26:05.454783: step: 212/529, loss: 2.3980212211608887 2023-01-21 08:26:06.587347: step: 216/529, loss: 3.275615930557251 2023-01-21 08:26:07.716854: step: 220/529, loss: 3.4327845573425293 2023-01-21 08:26:08.855570: step: 224/529, loss: 3.4952778816223145 2023-01-21 08:26:10.036497: step: 228/529, loss: 15.922210693359375 2023-01-21 08:26:11.176212: step: 232/529, loss: 3.977330207824707 2023-01-21 08:26:12.366556: step: 236/529, loss: 16.399816513061523 2023-01-21 08:26:13.500415: step: 240/529, loss: 3.9154746532440186 2023-01-21 08:26:14.639501: step: 244/529, loss: 6.403435707092285 2023-01-21 08:26:15.793621: step: 248/529, loss: 22.584266662597656 2023-01-21 08:26:16.915031: step: 252/529, loss: 3.6529977321624756 2023-01-21 08:26:18.057101: step: 256/529, loss: 4.413402557373047 2023-01-21 08:26:19.196825: step: 260/529, loss: 11.04973030090332 2023-01-21 08:26:20.328674: step: 264/529, loss: 13.011677742004395 2023-01-21 08:26:21.506180: step: 268/529, loss: 21.92221450805664 2023-01-21 08:26:22.656472: step: 272/529, loss: 3.4088897705078125 2023-01-21 08:26:23.789816: step: 276/529, loss: 9.868379592895508 2023-01-21 08:26:24.947083: step: 280/529, loss: 7.998035907745361 2023-01-21 08:26:26.111766: step: 284/529, loss: 7.747998237609863 2023-01-21 08:26:27.236478: step: 288/529, loss: 10.487600326538086 2023-01-21 08:26:28.349029: step: 292/529, loss: 2.0814990997314453 2023-01-21 08:26:29.498759: step: 296/529, loss: 19.663665771484375 2023-01-21 08:26:30.618974: step: 300/529, loss: 18.744508743286133 2023-01-21 08:26:31.771119: step: 304/529, loss: 15.447212219238281 2023-01-21 08:26:32.920566: step: 308/529, loss: 3.228632926940918 2023-01-21 08:26:34.088170: step: 312/529, loss: 9.735000610351562 2023-01-21 08:26:35.227323: step: 316/529, loss: 8.61428451538086 2023-01-21 08:26:36.338751: step: 320/529, loss: 2.3212857246398926 2023-01-21 08:26:37.490903: step: 324/529, loss: 3.5599350929260254 2023-01-21 08:26:38.620474: step: 328/529, loss: 25.49233627319336 2023-01-21 08:26:39.774668: step: 332/529, loss: 14.523412704467773 2023-01-21 08:26:40.919206: step: 336/529, loss: 15.256505966186523 2023-01-21 08:26:42.096097: step: 340/529, loss: 4.101177215576172 2023-01-21 08:26:43.244346: step: 344/529, loss: 13.484951972961426 2023-01-21 08:26:44.344533: step: 348/529, loss: 12.421173095703125 2023-01-21 08:26:45.523033: step: 352/529, loss: 20.946449279785156 2023-01-21 08:26:46.643131: step: 356/529, loss: 5.729172229766846 2023-01-21 08:26:47.808326: step: 360/529, loss: 4.24559211730957 2023-01-21 08:26:48.980152: step: 364/529, loss: 15.634750366210938 2023-01-21 08:26:50.102161: step: 368/529, loss: 3.1929025650024414 2023-01-21 08:26:51.269102: step: 372/529, loss: 4.050047874450684 2023-01-21 08:26:52.389112: step: 376/529, loss: 6.843565940856934 2023-01-21 08:26:53.521930: step: 380/529, loss: 6.251766681671143 2023-01-21 08:26:54.683570: step: 384/529, loss: 2.5218305587768555 2023-01-21 08:26:55.816539: step: 388/529, loss: 4.976187705993652 2023-01-21 08:26:56.960021: step: 392/529, loss: 14.570524215698242 2023-01-21 08:26:58.108347: step: 396/529, loss: 2.7670998573303223 2023-01-21 08:26:59.243851: step: 400/529, loss: 2.284353733062744 2023-01-21 08:27:00.380967: step: 404/529, loss: 2.0368025302886963 2023-01-21 08:27:01.527768: step: 408/529, loss: 11.088184356689453 2023-01-21 08:27:02.693270: step: 412/529, loss: 8.588536262512207 2023-01-21 08:27:03.860665: step: 416/529, loss: 15.744332313537598 2023-01-21 08:27:04.991809: step: 420/529, loss: 2.5162558555603027 2023-01-21 08:27:06.140163: step: 424/529, loss: 12.147928237915039 2023-01-21 08:27:07.275122: step: 428/529, loss: 16.330371856689453 2023-01-21 08:27:08.398972: step: 432/529, loss: 2.4632058143615723 2023-01-21 08:27:09.549472: step: 436/529, loss: 9.33383560180664 2023-01-21 08:27:10.687942: step: 440/529, loss: 10.591821670532227 2023-01-21 08:27:11.864859: step: 444/529, loss: 1.8141107559204102 2023-01-21 08:27:12.991597: step: 448/529, loss: 4.411225318908691 2023-01-21 08:27:14.091604: step: 452/529, loss: 1.4334101676940918 2023-01-21 08:27:15.235756: step: 456/529, loss: 3.184615135192871 2023-01-21 08:27:16.356182: step: 460/529, loss: 4.646284103393555 2023-01-21 08:27:17.475379: step: 464/529, loss: 1.6411666870117188 2023-01-21 08:27:18.621181: step: 468/529, loss: 6.401620864868164 2023-01-21 08:27:19.814951: step: 472/529, loss: 5.5645294189453125 2023-01-21 08:27:20.963091: step: 476/529, loss: 2.019742250442505 2023-01-21 08:27:22.112174: step: 480/529, loss: 1.4654401540756226 2023-01-21 08:27:23.234864: step: 484/529, loss: 1.5957221984863281 2023-01-21 08:27:24.393531: step: 488/529, loss: 11.635149002075195 2023-01-21 08:27:25.536015: step: 492/529, loss: 3.5896639823913574 2023-01-21 08:27:26.702905: step: 496/529, loss: 20.204261779785156 2023-01-21 08:27:27.835573: step: 500/529, loss: 1.2019952535629272 2023-01-21 08:27:28.982129: step: 504/529, loss: 9.574487686157227 2023-01-21 08:27:30.112784: step: 508/529, loss: 1.0506038665771484 2023-01-21 08:27:31.237761: step: 512/529, loss: 1.271968960762024 2023-01-21 08:27:32.382257: step: 516/529, loss: 7.965608596801758 2023-01-21 08:27:33.530114: step: 520/529, loss: 1.824202537536621 2023-01-21 08:27:34.702305: step: 524/529, loss: 1.1979656219482422 2023-01-21 08:27:35.829827: step: 528/529, loss: 0.8139227628707886 2023-01-21 08:27:36.982007: step: 532/529, loss: 5.290958881378174 2023-01-21 08:27:38.116050: step: 536/529, loss: 7.221170902252197 2023-01-21 08:27:39.275545: step: 540/529, loss: 5.015775203704834 2023-01-21 08:27:40.422760: step: 544/529, loss: 3.8171215057373047 2023-01-21 08:27:41.563589: step: 548/529, loss: 1.9852263927459717 2023-01-21 08:27:42.734615: step: 552/529, loss: 1.1681818962097168 2023-01-21 08:27:43.860318: step: 556/529, loss: 0.9781670570373535 2023-01-21 08:27:45.009498: step: 560/529, loss: 1.547278642654419 2023-01-21 08:27:46.146214: step: 564/529, loss: 2.165371894836426 2023-01-21 08:27:47.277045: step: 568/529, loss: 3.466583251953125 2023-01-21 08:27:48.429343: step: 572/529, loss: 2.0548954010009766 2023-01-21 08:27:49.557500: step: 576/529, loss: 1.2156181335449219 2023-01-21 08:27:50.701345: step: 580/529, loss: 1.1908199787139893 2023-01-21 08:27:51.824403: step: 584/529, loss: 3.61586856842041 2023-01-21 08:27:52.979918: step: 588/529, loss: 5.851022720336914 2023-01-21 08:27:54.101639: step: 592/529, loss: 1.3916974067687988 2023-01-21 08:27:55.262657: step: 596/529, loss: 4.2090373039245605 2023-01-21 08:27:56.400876: step: 600/529, loss: 1.3879903554916382 2023-01-21 08:27:57.534736: step: 604/529, loss: 0.8362724781036377 2023-01-21 08:27:58.686592: step: 608/529, loss: 0.513518214225769 2023-01-21 08:27:59.800649: step: 612/529, loss: 1.4912546873092651 2023-01-21 08:28:00.896750: step: 616/529, loss: 10.108013153076172 2023-01-21 08:28:02.072840: step: 620/529, loss: 3.1758322715759277 2023-01-21 08:28:03.193592: step: 624/529, loss: 0.7201522588729858 2023-01-21 08:28:04.367175: step: 628/529, loss: 3.003248929977417 2023-01-21 08:28:05.533880: step: 632/529, loss: 0.8918566703796387 2023-01-21 08:28:06.698015: step: 636/529, loss: 3.295438528060913 2023-01-21 08:28:07.854125: step: 640/529, loss: 3.516563892364502 2023-01-21 08:28:08.995503: step: 644/529, loss: 2.8429579734802246 2023-01-21 08:28:10.144971: step: 648/529, loss: 0.42065197229385376 2023-01-21 08:28:11.304032: step: 652/529, loss: 0.626232922077179 2023-01-21 08:28:12.458283: step: 656/529, loss: 1.2756016254425049 2023-01-21 08:28:13.591678: step: 660/529, loss: 0.5550310611724854 2023-01-21 08:28:14.721097: step: 664/529, loss: 1.5328176021575928 2023-01-21 08:28:15.853885: step: 668/529, loss: 1.6487457752227783 2023-01-21 08:28:17.018347: step: 672/529, loss: 2.7514681816101074 2023-01-21 08:28:18.173613: step: 676/529, loss: 4.128016471862793 2023-01-21 08:28:19.310066: step: 680/529, loss: 0.6200653314590454 2023-01-21 08:28:20.451618: step: 684/529, loss: 3.6971468925476074 2023-01-21 08:28:21.594540: step: 688/529, loss: 0.7104389667510986 2023-01-21 08:28:22.741976: step: 692/529, loss: 0.9314755797386169 2023-01-21 08:28:23.853282: step: 696/529, loss: 1.1598939895629883 2023-01-21 08:28:25.011550: step: 700/529, loss: 0.8454206585884094 2023-01-21 08:28:26.155058: step: 704/529, loss: 1.9795746803283691 2023-01-21 08:28:27.367218: step: 708/529, loss: 1.7655086517333984 2023-01-21 08:28:28.538792: step: 712/529, loss: 0.9487324953079224 2023-01-21 08:28:29.683804: step: 716/529, loss: 1.832971215248108 2023-01-21 08:28:30.807758: step: 720/529, loss: 1.7045546770095825 2023-01-21 08:28:31.949576: step: 724/529, loss: 1.2064987421035767 2023-01-21 08:28:33.075852: step: 728/529, loss: 3.393432378768921 2023-01-21 08:28:34.221575: step: 732/529, loss: 0.3778058886528015 2023-01-21 08:28:35.371612: step: 736/529, loss: 0.8241660594940186 2023-01-21 08:28:36.493553: step: 740/529, loss: 0.511652410030365 2023-01-21 08:28:37.655086: step: 744/529, loss: 3.5488650798797607 2023-01-21 08:28:38.811389: step: 748/529, loss: 1.674997329711914 2023-01-21 08:28:39.951794: step: 752/529, loss: 3.3751368522644043 2023-01-21 08:28:41.102540: step: 756/529, loss: 0.5803846120834351 2023-01-21 08:28:42.257978: step: 760/529, loss: 0.4445878863334656 2023-01-21 08:28:43.401884: step: 764/529, loss: 0.9939770698547363 2023-01-21 08:28:44.532427: step: 768/529, loss: 0.36156535148620605 2023-01-21 08:28:45.715542: step: 772/529, loss: 3.4489002227783203 2023-01-21 08:28:46.858444: step: 776/529, loss: 4.135929584503174 2023-01-21 08:28:47.995811: step: 780/529, loss: 2.089932918548584 2023-01-21 08:28:49.149159: step: 784/529, loss: 0.4640917181968689 2023-01-21 08:28:50.345327: step: 788/529, loss: 1.1669062376022339 2023-01-21 08:28:51.513761: step: 792/529, loss: 0.9698988199234009 2023-01-21 08:28:52.665424: step: 796/529, loss: 1.3405826091766357 2023-01-21 08:28:53.832215: step: 800/529, loss: 2.5966296195983887 2023-01-21 08:28:54.967387: step: 804/529, loss: 0.779301106929779 2023-01-21 08:28:56.104571: step: 808/529, loss: 2.9553210735321045 2023-01-21 08:28:57.253644: step: 812/529, loss: 0.5742433071136475 2023-01-21 08:28:58.400526: step: 816/529, loss: 1.4621717929840088 2023-01-21 08:28:59.554321: step: 820/529, loss: 0.5653336644172668 2023-01-21 08:29:00.700456: step: 824/529, loss: 0.357096791267395 2023-01-21 08:29:01.853467: step: 828/529, loss: 2.403977394104004 2023-01-21 08:29:02.991478: step: 832/529, loss: 3.190617561340332 2023-01-21 08:29:04.168957: step: 836/529, loss: 1.0175585746765137 2023-01-21 08:29:05.307657: step: 840/529, loss: 1.1468260288238525 2023-01-21 08:29:06.453769: step: 844/529, loss: 1.7973920106887817 2023-01-21 08:29:07.600937: step: 848/529, loss: 0.7385352849960327 2023-01-21 08:29:08.752797: step: 852/529, loss: 1.0770177841186523 2023-01-21 08:29:09.881839: step: 856/529, loss: 3.0698933601379395 2023-01-21 08:29:11.031079: step: 860/529, loss: 0.4015979766845703 2023-01-21 08:29:12.227183: step: 864/529, loss: 2.0169858932495117 2023-01-21 08:29:13.364270: step: 868/529, loss: 1.0402926206588745 2023-01-21 08:29:14.521073: step: 872/529, loss: 1.6893560886383057 2023-01-21 08:29:15.701793: step: 876/529, loss: 0.6583280563354492 2023-01-21 08:29:16.841382: step: 880/529, loss: 0.29671764373779297 2023-01-21 08:29:17.983059: step: 884/529, loss: 1.7505154609680176 2023-01-21 08:29:19.124548: step: 888/529, loss: 1.2750203609466553 2023-01-21 08:29:20.298633: step: 892/529, loss: 2.740898609161377 2023-01-21 08:29:21.444802: step: 896/529, loss: 0.40280312299728394 2023-01-21 08:29:22.580071: step: 900/529, loss: 0.29644542932510376 2023-01-21 08:29:23.712592: step: 904/529, loss: 2.701512336730957 2023-01-21 08:29:24.852756: step: 908/529, loss: 0.23906993865966797 2023-01-21 08:29:25.998614: step: 912/529, loss: 0.5065438747406006 2023-01-21 08:29:27.144342: step: 916/529, loss: 0.23993238806724548 2023-01-21 08:29:28.273923: step: 920/529, loss: 1.8112282752990723 2023-01-21 08:29:29.435711: step: 924/529, loss: 0.3414899408817291 2023-01-21 08:29:30.588630: step: 928/529, loss: 0.6074423789978027 2023-01-21 08:29:31.777676: step: 932/529, loss: 0.819431722164154 2023-01-21 08:29:32.895079: step: 936/529, loss: 1.5271637439727783 2023-01-21 08:29:34.035972: step: 940/529, loss: 2.2112016677856445 2023-01-21 08:29:35.171616: step: 944/529, loss: 1.2948899269104004 2023-01-21 08:29:36.333174: step: 948/529, loss: 0.5799522399902344 2023-01-21 08:29:37.501352: step: 952/529, loss: 3.5372352600097656 2023-01-21 08:29:38.642419: step: 956/529, loss: 5.067075729370117 2023-01-21 08:29:39.803046: step: 960/529, loss: 1.584993839263916 2023-01-21 08:29:40.941603: step: 964/529, loss: 0.9113391041755676 2023-01-21 08:29:42.106278: step: 968/529, loss: 3.3932623863220215 2023-01-21 08:29:43.241772: step: 972/529, loss: 1.536823034286499 2023-01-21 08:29:44.391191: step: 976/529, loss: 1.6471267938613892 2023-01-21 08:29:45.564836: step: 980/529, loss: 5.270932197570801 2023-01-21 08:29:46.723628: step: 984/529, loss: 0.5724227428436279 2023-01-21 08:29:47.886261: step: 988/529, loss: 1.2616477012634277 2023-01-21 08:29:49.011071: step: 992/529, loss: 1.6249890327453613 2023-01-21 08:29:50.175098: step: 996/529, loss: 1.4779839515686035 2023-01-21 08:29:51.332791: step: 1000/529, loss: 0.3906409442424774 2023-01-21 08:29:52.464275: step: 1004/529, loss: 2.9598405361175537 2023-01-21 08:29:53.593429: step: 1008/529, loss: 2.644022226333618 2023-01-21 08:29:54.771134: step: 1012/529, loss: 3.001842975616455 2023-01-21 08:29:55.889406: step: 1016/529, loss: 2.463235855102539 2023-01-21 08:29:57.031763: step: 1020/529, loss: 0.45442572236061096 2023-01-21 08:29:58.182929: step: 1024/529, loss: 0.30489426851272583 2023-01-21 08:29:59.330151: step: 1028/529, loss: 0.5068156719207764 2023-01-21 08:30:00.476203: step: 1032/529, loss: 1.2137457132339478 2023-01-21 08:30:01.638683: step: 1036/529, loss: 6.931217193603516 2023-01-21 08:30:02.774634: step: 1040/529, loss: 0.32786035537719727 2023-01-21 08:30:03.911740: step: 1044/529, loss: 2.5096447467803955 2023-01-21 08:30:05.067889: step: 1048/529, loss: 0.4271506667137146 2023-01-21 08:30:06.184350: step: 1052/529, loss: 3.0060503482818604 2023-01-21 08:30:07.341995: step: 1056/529, loss: 0.6822559833526611 2023-01-21 08:30:08.487416: step: 1060/529, loss: 0.718085527420044 2023-01-21 08:30:09.619523: step: 1064/529, loss: 1.7505027055740356 2023-01-21 08:30:10.763767: step: 1068/529, loss: 0.655796229839325 2023-01-21 08:30:11.931715: step: 1072/529, loss: 0.716809868812561 2023-01-21 08:30:13.116715: step: 1076/529, loss: 4.08268928527832 2023-01-21 08:30:14.275490: step: 1080/529, loss: 0.26747018098831177 2023-01-21 08:30:15.406082: step: 1084/529, loss: 0.5302784442901611 2023-01-21 08:30:16.559526: step: 1088/529, loss: 2.5076818466186523 2023-01-21 08:30:17.709474: step: 1092/529, loss: 9.097386360168457 2023-01-21 08:30:18.860051: step: 1096/529, loss: 3.2745413780212402 2023-01-21 08:30:20.007533: step: 1100/529, loss: 2.0416064262390137 2023-01-21 08:30:21.144088: step: 1104/529, loss: 0.5942925214767456 2023-01-21 08:30:22.294796: step: 1108/529, loss: 0.6088376045227051 2023-01-21 08:30:23.438481: step: 1112/529, loss: 1.2197341918945312 2023-01-21 08:30:24.601437: step: 1116/529, loss: 0.43286314606666565 2023-01-21 08:30:25.755927: step: 1120/529, loss: 0.23642738163471222 2023-01-21 08:30:26.871939: step: 1124/529, loss: 2.0354673862457275 2023-01-21 08:30:28.018099: step: 1128/529, loss: 0.7759148478507996 2023-01-21 08:30:29.148510: step: 1132/529, loss: 0.6867160797119141 2023-01-21 08:30:30.303867: step: 1136/529, loss: 0.33035534620285034 2023-01-21 08:30:31.436039: step: 1140/529, loss: 0.30516648292541504 2023-01-21 08:30:32.589379: step: 1144/529, loss: 1.8776904344558716 2023-01-21 08:30:33.730955: step: 1148/529, loss: 1.650612235069275 2023-01-21 08:30:34.876332: step: 1152/529, loss: 0.38097113370895386 2023-01-21 08:30:36.027292: step: 1156/529, loss: 0.5021973848342896 2023-01-21 08:30:37.165298: step: 1160/529, loss: 8.511953353881836 2023-01-21 08:30:38.311607: step: 1164/529, loss: 1.480036735534668 2023-01-21 08:30:39.449978: step: 1168/529, loss: 1.989585041999817 2023-01-21 08:30:40.618182: step: 1172/529, loss: 5.852951526641846 2023-01-21 08:30:41.765966: step: 1176/529, loss: 0.9966527223587036 2023-01-21 08:30:42.911007: step: 1180/529, loss: 2.0300960540771484 2023-01-21 08:30:44.059283: step: 1184/529, loss: 0.49075227975845337 2023-01-21 08:30:45.194179: step: 1188/529, loss: 0.5505347847938538 2023-01-21 08:30:46.354230: step: 1192/529, loss: 1.1302720308303833 2023-01-21 08:30:47.480574: step: 1196/529, loss: 1.0865519046783447 2023-01-21 08:30:48.631867: step: 1200/529, loss: 0.7182204127311707 2023-01-21 08:30:49.777177: step: 1204/529, loss: 1.036863923072815 2023-01-21 08:30:50.931066: step: 1208/529, loss: 1.3710113763809204 2023-01-21 08:30:52.090409: step: 1212/529, loss: 0.3698939383029938 2023-01-21 08:30:53.253345: step: 1216/529, loss: 0.5071775317192078 2023-01-21 08:30:54.404948: step: 1220/529, loss: 2.9697458744049072 2023-01-21 08:30:55.541820: step: 1224/529, loss: 0.794247031211853 2023-01-21 08:30:56.669117: step: 1228/529, loss: 0.8224543333053589 2023-01-21 08:30:57.820521: step: 1232/529, loss: 0.21337567269802094 2023-01-21 08:30:58.978202: step: 1236/529, loss: 0.4057280421257019 2023-01-21 08:31:00.116035: step: 1240/529, loss: 0.4606713354587555 2023-01-21 08:31:01.278417: step: 1244/529, loss: 0.38343098759651184 2023-01-21 08:31:02.427563: step: 1248/529, loss: 0.5494058132171631 2023-01-21 08:31:03.556180: step: 1252/529, loss: 0.7290338277816772 2023-01-21 08:31:04.687725: step: 1256/529, loss: 1.510838508605957 2023-01-21 08:31:05.801430: step: 1260/529, loss: 3.643113374710083 2023-01-21 08:31:06.940782: step: 1264/529, loss: 1.732445240020752 2023-01-21 08:31:08.062539: step: 1268/529, loss: 0.40447598695755005 2023-01-21 08:31:09.186076: step: 1272/529, loss: 0.8668304085731506 2023-01-21 08:31:10.334342: step: 1276/529, loss: 0.25786638259887695 2023-01-21 08:31:11.510001: step: 1280/529, loss: 0.8625131845474243 2023-01-21 08:31:12.676766: step: 1284/529, loss: 0.3239728808403015 2023-01-21 08:31:13.803032: step: 1288/529, loss: 1.5239778757095337 2023-01-21 08:31:14.946582: step: 1292/529, loss: 1.7557837963104248 2023-01-21 08:31:16.102371: step: 1296/529, loss: 1.75142240524292 2023-01-21 08:31:17.239146: step: 1300/529, loss: 0.9475021362304688 2023-01-21 08:31:18.381194: step: 1304/529, loss: 1.3006868362426758 2023-01-21 08:31:19.519379: step: 1308/529, loss: 0.5015181303024292 2023-01-21 08:31:20.680768: step: 1312/529, loss: 1.085694432258606 2023-01-21 08:31:21.806880: step: 1316/529, loss: 0.35742467641830444 2023-01-21 08:31:22.946595: step: 1320/529, loss: 1.0509415864944458 2023-01-21 08:31:24.070923: step: 1324/529, loss: 0.3264305889606476 2023-01-21 08:31:25.209939: step: 1328/529, loss: 0.5333709716796875 2023-01-21 08:31:26.368239: step: 1332/529, loss: 0.4387022852897644 2023-01-21 08:31:27.518288: step: 1336/529, loss: 0.4024922251701355 2023-01-21 08:31:28.659632: step: 1340/529, loss: 1.852999210357666 2023-01-21 08:31:29.766629: step: 1344/529, loss: 0.7516586780548096 2023-01-21 08:31:30.915653: step: 1348/529, loss: 2.232771873474121 2023-01-21 08:31:32.039354: step: 1352/529, loss: 0.2666875720024109 2023-01-21 08:31:33.173948: step: 1356/529, loss: 0.33623045682907104 2023-01-21 08:31:34.314302: step: 1360/529, loss: 0.5571774244308472 2023-01-21 08:31:35.456764: step: 1364/529, loss: 0.48473238945007324 2023-01-21 08:31:36.637799: step: 1368/529, loss: 0.17340508103370667 2023-01-21 08:31:37.777699: step: 1372/529, loss: 0.4033452272415161 2023-01-21 08:31:38.919295: step: 1376/529, loss: 0.6019673347473145 2023-01-21 08:31:40.062538: step: 1380/529, loss: 1.0341713428497314 2023-01-21 08:31:41.181437: step: 1384/529, loss: 0.286648154258728 2023-01-21 08:31:42.323544: step: 1388/529, loss: 4.122661590576172 2023-01-21 08:31:43.436879: step: 1392/529, loss: 0.7338075637817383 2023-01-21 08:31:44.610510: step: 1396/529, loss: 0.6949564218521118 2023-01-21 08:31:45.734733: step: 1400/529, loss: 0.525062620639801 2023-01-21 08:31:46.878465: step: 1404/529, loss: 6.894042015075684 2023-01-21 08:31:48.037191: step: 1408/529, loss: 1.2990083694458008 2023-01-21 08:31:49.175307: step: 1412/529, loss: 2.1497371196746826 2023-01-21 08:31:50.330803: step: 1416/529, loss: 1.2402030229568481 2023-01-21 08:31:51.451845: step: 1420/529, loss: 0.5442218780517578 2023-01-21 08:31:52.643291: step: 1424/529, loss: 2.8404784202575684 2023-01-21 08:31:53.779417: step: 1428/529, loss: 0.2555355131626129 2023-01-21 08:31:54.927570: step: 1432/529, loss: 0.7639809846878052 2023-01-21 08:31:56.065736: step: 1436/529, loss: 1.7471200227737427 2023-01-21 08:31:57.192704: step: 1440/529, loss: 0.23662137985229492 2023-01-21 08:31:58.304280: step: 1444/529, loss: 0.40045350790023804 2023-01-21 08:31:59.471543: step: 1448/529, loss: 1.1397353410720825 2023-01-21 08:32:00.614102: step: 1452/529, loss: 0.4629027545452118 2023-01-21 08:32:01.754840: step: 1456/529, loss: 0.9567396640777588 2023-01-21 08:32:02.913023: step: 1460/529, loss: 0.341137558221817 2023-01-21 08:32:04.033416: step: 1464/529, loss: 0.4640963077545166 2023-01-21 08:32:05.141696: step: 1468/529, loss: 2.211867332458496 2023-01-21 08:32:06.266823: step: 1472/529, loss: 0.585845947265625 2023-01-21 08:32:07.396629: step: 1476/529, loss: 0.4816916882991791 2023-01-21 08:32:08.522309: step: 1480/529, loss: 1.4729294776916504 2023-01-21 08:32:09.645329: step: 1484/529, loss: 0.4252338409423828 2023-01-21 08:32:10.806124: step: 1488/529, loss: 5.345481872558594 2023-01-21 08:32:11.970194: step: 1492/529, loss: 0.8776652812957764 2023-01-21 08:32:13.131996: step: 1496/529, loss: 0.29841554164886475 2023-01-21 08:32:14.283469: step: 1500/529, loss: 0.5287191271781921 2023-01-21 08:32:15.410199: step: 1504/529, loss: 0.5562278628349304 2023-01-21 08:32:16.571828: step: 1508/529, loss: 2.2081239223480225 2023-01-21 08:32:17.694699: step: 1512/529, loss: 0.37848329544067383 2023-01-21 08:32:18.850247: step: 1516/529, loss: 1.870319128036499 2023-01-21 08:32:20.003420: step: 1520/529, loss: 0.24362607300281525 2023-01-21 08:32:21.140408: step: 1524/529, loss: 0.47149384021759033 2023-01-21 08:32:22.302503: step: 1528/529, loss: 0.6378626823425293 2023-01-21 08:32:23.434563: step: 1532/529, loss: 0.5113728046417236 2023-01-21 08:32:24.595061: step: 1536/529, loss: 3.1645052433013916 2023-01-21 08:32:25.728601: step: 1540/529, loss: 1.1048482656478882 2023-01-21 08:32:26.872637: step: 1544/529, loss: 0.9618955850601196 2023-01-21 08:32:28.029397: step: 1548/529, loss: 0.6876150965690613 2023-01-21 08:32:29.150750: step: 1552/529, loss: 0.2438717484474182 2023-01-21 08:32:30.319757: step: 1556/529, loss: 7.936010360717773 2023-01-21 08:32:31.477792: step: 1560/529, loss: 1.4922549724578857 2023-01-21 08:32:32.665308: step: 1564/529, loss: 0.27962514758110046 2023-01-21 08:32:33.815762: step: 1568/529, loss: 7.159943103790283 2023-01-21 08:32:34.940714: step: 1572/529, loss: 1.120267629623413 2023-01-21 08:32:36.086409: step: 1576/529, loss: 0.41146671772003174 2023-01-21 08:32:37.201910: step: 1580/529, loss: 2.4651637077331543 2023-01-21 08:32:38.344127: step: 1584/529, loss: 0.8812049627304077 2023-01-21 08:32:39.466000: step: 1588/529, loss: 0.20512624084949493 2023-01-21 08:32:40.622666: step: 1592/529, loss: 1.12703537940979 2023-01-21 08:32:41.752926: step: 1596/529, loss: 7.471231460571289 2023-01-21 08:32:42.966699: step: 1600/529, loss: 1.0049278736114502 2023-01-21 08:32:44.118191: step: 1604/529, loss: 1.0047423839569092 2023-01-21 08:32:45.263340: step: 1608/529, loss: 0.7438883781433105 2023-01-21 08:32:46.379872: step: 1612/529, loss: 1.006537914276123 2023-01-21 08:32:47.529917: step: 1616/529, loss: 1.6063528060913086 2023-01-21 08:32:48.687164: step: 1620/529, loss: 0.975894570350647 2023-01-21 08:32:49.846360: step: 1624/529, loss: 1.0521715879440308 2023-01-21 08:32:50.959085: step: 1628/529, loss: 1.572803020477295 2023-01-21 08:32:52.092349: step: 1632/529, loss: 0.5591209530830383 2023-01-21 08:32:53.244407: step: 1636/529, loss: 1.0661540031433105 2023-01-21 08:32:54.421218: step: 1640/529, loss: 0.5868942141532898 2023-01-21 08:32:55.550488: step: 1644/529, loss: 1.431801438331604 2023-01-21 08:32:56.700013: step: 1648/529, loss: 1.76229989528656 2023-01-21 08:32:57.857400: step: 1652/529, loss: 0.37676024436950684 2023-01-21 08:32:58.988702: step: 1656/529, loss: 0.30986320972442627 2023-01-21 08:33:00.160993: step: 1660/529, loss: 0.912360668182373 2023-01-21 08:33:01.301440: step: 1664/529, loss: 0.8488156795501709 2023-01-21 08:33:02.470162: step: 1668/529, loss: 0.8135032653808594 2023-01-21 08:33:03.614989: step: 1672/529, loss: 0.07859086990356445 2023-01-21 08:33:04.747916: step: 1676/529, loss: 0.22118231654167175 2023-01-21 08:33:05.885863: step: 1680/529, loss: 0.6065610647201538 2023-01-21 08:33:07.048858: step: 1684/529, loss: 0.18476100265979767 2023-01-21 08:33:08.184519: step: 1688/529, loss: 0.4858241677284241 2023-01-21 08:33:09.345169: step: 1692/529, loss: 0.550195574760437 2023-01-21 08:33:10.489017: step: 1696/529, loss: 0.34195661544799805 2023-01-21 08:33:11.621815: step: 1700/529, loss: 0.7902202606201172 2023-01-21 08:33:12.791378: step: 1704/529, loss: 0.1298881471157074 2023-01-21 08:33:13.943096: step: 1708/529, loss: 0.6398295164108276 2023-01-21 08:33:15.142106: step: 1712/529, loss: 0.20396332442760468 2023-01-21 08:33:16.307058: step: 1716/529, loss: 0.2997762858867645 2023-01-21 08:33:17.486694: step: 1720/529, loss: 1.7543193101882935 2023-01-21 08:33:18.629456: step: 1724/529, loss: 3.327333688735962 2023-01-21 08:33:19.742476: step: 1728/529, loss: 0.3710118532180786 2023-01-21 08:33:20.898009: step: 1732/529, loss: 0.5208996534347534 2023-01-21 08:33:22.026714: step: 1736/529, loss: 0.3774925768375397 2023-01-21 08:33:23.137103: step: 1740/529, loss: 0.5465073585510254 2023-01-21 08:33:24.291741: step: 1744/529, loss: 0.4329710006713867 2023-01-21 08:33:25.413791: step: 1748/529, loss: 0.5405237078666687 2023-01-21 08:33:26.574186: step: 1752/529, loss: 1.8058745861053467 2023-01-21 08:33:27.697470: step: 1756/529, loss: 0.5758148431777954 2023-01-21 08:33:28.818908: step: 1760/529, loss: 0.43531861901283264 2023-01-21 08:33:29.947449: step: 1764/529, loss: 0.37160149216651917 2023-01-21 08:33:31.095560: step: 1768/529, loss: 0.566301167011261 2023-01-21 08:33:32.238069: step: 1772/529, loss: 0.18470755219459534 2023-01-21 08:33:33.374694: step: 1776/529, loss: 0.44613516330718994 2023-01-21 08:33:34.520592: step: 1780/529, loss: 1.723960280418396 2023-01-21 08:33:35.664954: step: 1784/529, loss: 0.4782143831253052 2023-01-21 08:33:36.820764: step: 1788/529, loss: 0.22105231881141663 2023-01-21 08:33:37.948698: step: 1792/529, loss: 0.10841484367847443 2023-01-21 08:33:39.084484: step: 1796/529, loss: 3.045856475830078 2023-01-21 08:33:40.232098: step: 1800/529, loss: 0.6988542675971985 2023-01-21 08:33:41.389682: step: 1804/529, loss: 0.8430399894714355 2023-01-21 08:33:42.550726: step: 1808/529, loss: 0.25225868821144104 2023-01-21 08:33:43.668896: step: 1812/529, loss: 0.5378435850143433 2023-01-21 08:33:44.815126: step: 1816/529, loss: 2.1167407035827637 2023-01-21 08:33:45.912799: step: 1820/529, loss: 0.4301365911960602 2023-01-21 08:33:47.049111: step: 1824/529, loss: 0.14607305824756622 2023-01-21 08:33:48.197585: step: 1828/529, loss: 2.0961008071899414 2023-01-21 08:33:49.353191: step: 1832/529, loss: 2.315335512161255 2023-01-21 08:33:50.479047: step: 1836/529, loss: 0.4069212079048157 2023-01-21 08:33:51.617661: step: 1840/529, loss: 0.12884049117565155 2023-01-21 08:33:52.754553: step: 1844/529, loss: 0.8808757662773132 2023-01-21 08:33:53.878597: step: 1848/529, loss: 0.6215300559997559 2023-01-21 08:33:55.014488: step: 1852/529, loss: 0.3097376227378845 2023-01-21 08:33:56.141371: step: 1856/529, loss: 0.5453568696975708 2023-01-21 08:33:57.269704: step: 1860/529, loss: 0.3228136897087097 2023-01-21 08:33:58.409275: step: 1864/529, loss: 3.5696210861206055 2023-01-21 08:33:59.539393: step: 1868/529, loss: 0.21188050508499146 2023-01-21 08:34:00.696885: step: 1872/529, loss: 1.728649377822876 2023-01-21 08:34:01.837867: step: 1876/529, loss: 1.2658092975616455 2023-01-21 08:34:02.958526: step: 1880/529, loss: 0.8763715028762817 2023-01-21 08:34:04.112485: step: 1884/529, loss: 0.4274565577507019 2023-01-21 08:34:05.315936: step: 1888/529, loss: 0.8709762692451477 2023-01-21 08:34:06.454160: step: 1892/529, loss: 1.7838966846466064 2023-01-21 08:34:07.562565: step: 1896/529, loss: 0.22778764367103577 2023-01-21 08:34:08.708429: step: 1900/529, loss: 2.0915491580963135 2023-01-21 08:34:09.860492: step: 1904/529, loss: 0.33962053060531616 2023-01-21 08:34:10.984903: step: 1908/529, loss: 0.45984387397766113 2023-01-21 08:34:12.140649: step: 1912/529, loss: 2.1153433322906494 2023-01-21 08:34:13.288964: step: 1916/529, loss: 0.30112916231155396 2023-01-21 08:34:14.443481: step: 1920/529, loss: 0.39301174879074097 2023-01-21 08:34:15.601257: step: 1924/529, loss: 0.21990787982940674 2023-01-21 08:34:16.751520: step: 1928/529, loss: 0.30765199661254883 2023-01-21 08:34:17.901537: step: 1932/529, loss: 0.4110768437385559 2023-01-21 08:34:19.003907: step: 1936/529, loss: 1.0211756229400635 2023-01-21 08:34:20.141330: step: 1940/529, loss: 0.8921117782592773 2023-01-21 08:34:21.310378: step: 1944/529, loss: 0.6786177158355713 2023-01-21 08:34:22.461095: step: 1948/529, loss: 0.532731831073761 2023-01-21 08:34:23.602815: step: 1952/529, loss: 0.2940041124820709 2023-01-21 08:34:24.723574: step: 1956/529, loss: 1.8261213302612305 2023-01-21 08:34:25.864169: step: 1960/529, loss: 0.15733389556407928 2023-01-21 08:34:27.029253: step: 1964/529, loss: 0.7331670522689819 2023-01-21 08:34:28.164274: step: 1968/529, loss: 0.8912182450294495 2023-01-21 08:34:29.339867: step: 1972/529, loss: 1.4951698780059814 2023-01-21 08:34:30.516585: step: 1976/529, loss: 0.49952155351638794 2023-01-21 08:34:31.681340: step: 1980/529, loss: 1.2072674036026 2023-01-21 08:34:32.813431: step: 1984/529, loss: 0.20373401045799255 2023-01-21 08:34:33.947629: step: 1988/529, loss: 1.7011696100234985 2023-01-21 08:34:35.069725: step: 1992/529, loss: 0.6681041717529297 2023-01-21 08:34:36.197437: step: 1996/529, loss: 0.22191476821899414 2023-01-21 08:34:37.362705: step: 2000/529, loss: 0.7241841554641724 2023-01-21 08:34:38.498595: step: 2004/529, loss: 1.2355533838272095 2023-01-21 08:34:39.644582: step: 2008/529, loss: 0.36321955919265747 2023-01-21 08:34:40.799737: step: 2012/529, loss: 1.7107552289962769 2023-01-21 08:34:41.940251: step: 2016/529, loss: 1.6967189311981201 2023-01-21 08:34:43.086320: step: 2020/529, loss: 1.6046949625015259 2023-01-21 08:34:44.234859: step: 2024/529, loss: 0.25262928009033203 2023-01-21 08:34:45.381484: step: 2028/529, loss: 0.43971705436706543 2023-01-21 08:34:46.554315: step: 2032/529, loss: 0.12725749611854553 2023-01-21 08:34:47.679912: step: 2036/529, loss: 0.4302833676338196 2023-01-21 08:34:48.832062: step: 2040/529, loss: 0.8647814989089966 2023-01-21 08:34:49.973702: step: 2044/529, loss: 0.7444102764129639 2023-01-21 08:34:51.119813: step: 2048/529, loss: 0.2606492042541504 2023-01-21 08:34:52.260661: step: 2052/529, loss: 0.8823707699775696 2023-01-21 08:34:53.403675: step: 2056/529, loss: 1.8375400304794312 2023-01-21 08:34:54.554952: step: 2060/529, loss: 0.30606240034103394 2023-01-21 08:34:55.708205: step: 2064/529, loss: 0.8164175152778625 2023-01-21 08:34:56.848363: step: 2068/529, loss: 0.25308284163475037 2023-01-21 08:34:58.001648: step: 2072/529, loss: 0.23977404832839966 2023-01-21 08:34:59.120655: step: 2076/529, loss: 0.2210439145565033 2023-01-21 08:35:00.248341: step: 2080/529, loss: 0.463204026222229 2023-01-21 08:35:01.418176: step: 2084/529, loss: 1.7650251388549805 2023-01-21 08:35:02.566439: step: 2088/529, loss: 6.856368064880371 2023-01-21 08:35:03.706235: step: 2092/529, loss: 1.110244870185852 2023-01-21 08:35:04.842146: step: 2096/529, loss: 0.3843016028404236 2023-01-21 08:35:05.970244: step: 2100/529, loss: 0.9868789911270142 2023-01-21 08:35:07.128795: step: 2104/529, loss: 0.808090329170227 2023-01-21 08:35:08.278105: step: 2108/529, loss: 0.1819995492696762 2023-01-21 08:35:09.437332: step: 2112/529, loss: 0.7787676453590393 2023-01-21 08:35:10.590925: step: 2116/529, loss: 2.5226590633392334 ================================================== Loss: 3.208 -------------------- Dev: {'event': {'p': 0.5862068965517241, 'r': 0.5432756324900133, 'f1': 0.5639253628196268}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Test: {'event': {'p': 0.4797191887675507, 'r': 0.3563151796060255, 'f1': 0.4089095744680851}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Chinese: {'event': {'p': 0.5609756097560976, 'r': 0.42592592592592593, 'f1': 0.4842105263157895}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Korean: {'event': {'p': 0.6923076923076923, 'r': 0.2857142857142857, 'f1': 0.4044943820224719}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Russian: {'event': {'p': 0.375, 'r': 0.16666666666666666, 'f1': 0.23076923076923078}, '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': 0.5862068965517241, 'r': 0.5432756324900133, 'f1': 0.5639253628196268}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Chinese: {'event': {'p': 0.4797191887675507, 'r': 0.3563151796060255, 'f1': 0.4089095744680851}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Chinese: {'event': {'p': 0.5609756097560976, 'r': 0.42592592592592593, 'f1': 0.4842105263157895}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Korean: {'event': {'p': 0.5862068965517241, 'r': 0.5432756324900133, 'f1': 0.5639253628196268}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Korean: {'event': {'p': 0.4797191887675507, 'r': 0.3563151796060255, 'f1': 0.4089095744680851}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Korean: {'event': {'p': 0.6923076923076923, 'r': 0.2857142857142857, 'f1': 0.4044943820224719}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Russian: {'event': {'p': 0.5862068965517241, 'r': 0.5432756324900133, 'f1': 0.5639253628196268}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Russian: {'event': {'p': 0.4797191887675507, 'r': 0.3563151796060255, 'f1': 0.4089095744680851}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Russian: {'event': {'p': 0.375, 'r': 0.16666666666666666, 'f1': 0.23076923076923078}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} ****************************** Epoch: 1 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:36:09.723768: step: 4/529, loss: 2.938530921936035 2023-01-21 08:36:10.862601: step: 8/529, loss: 0.35229891538619995 2023-01-21 08:36:12.002158: step: 12/529, loss: 1.1775540113449097 2023-01-21 08:36:13.180897: step: 16/529, loss: 1.7110569477081299 2023-01-21 08:36:14.328648: step: 20/529, loss: 1.1641806364059448 2023-01-21 08:36:15.481993: step: 24/529, loss: 0.19667291641235352 2023-01-21 08:36:16.650695: step: 28/529, loss: 0.8878239393234253 2023-01-21 08:36:17.770787: step: 32/529, loss: 4.964476585388184 2023-01-21 08:36:18.930317: step: 36/529, loss: 1.4169527292251587 2023-01-21 08:36:20.063109: step: 40/529, loss: 1.3267056941986084 2023-01-21 08:36:21.234552: step: 44/529, loss: 0.9741167426109314 2023-01-21 08:36:22.356799: step: 48/529, loss: 0.9785480499267578 2023-01-21 08:36:23.499930: step: 52/529, loss: 6.838603973388672 2023-01-21 08:36:24.638578: step: 56/529, loss: 0.42646974325180054 2023-01-21 08:36:25.776199: step: 60/529, loss: 1.7733484506607056 2023-01-21 08:36:26.913236: step: 64/529, loss: 0.8867408633232117 2023-01-21 08:36:28.059376: step: 68/529, loss: 1.1831772327423096 2023-01-21 08:36:29.209278: step: 72/529, loss: 0.2452596127986908 2023-01-21 08:36:30.348204: step: 76/529, loss: 2.0087647438049316 2023-01-21 08:36:31.492344: step: 80/529, loss: 0.6992994546890259 2023-01-21 08:36:32.606071: step: 84/529, loss: 0.3327838182449341 2023-01-21 08:36:33.751313: step: 88/529, loss: 0.2657693922519684 2023-01-21 08:36:34.895780: step: 92/529, loss: 0.5083276033401489 2023-01-21 08:36:36.039566: step: 96/529, loss: 1.513451099395752 2023-01-21 08:36:37.149827: step: 100/529, loss: 0.8501608967781067 2023-01-21 08:36:38.283572: step: 104/529, loss: 11.079312324523926 2023-01-21 08:36:39.423505: step: 108/529, loss: 0.18545770645141602 2023-01-21 08:36:40.582636: step: 112/529, loss: 0.13908815383911133 2023-01-21 08:36:41.755257: step: 116/529, loss: 0.6973282694816589 2023-01-21 08:36:42.937246: step: 120/529, loss: 0.29902762174606323 2023-01-21 08:36:44.070207: step: 124/529, loss: 0.905144214630127 2023-01-21 08:36:45.198576: step: 128/529, loss: 0.9375255107879639 2023-01-21 08:36:46.329001: step: 132/529, loss: 0.30079782009124756 2023-01-21 08:36:47.508050: step: 136/529, loss: 1.599959135055542 2023-01-21 08:36:48.656504: step: 140/529, loss: 0.6238815784454346 2023-01-21 08:36:49.780099: step: 144/529, loss: 0.9514894485473633 2023-01-21 08:36:50.902523: step: 148/529, loss: 0.25117167830467224 2023-01-21 08:36:52.041872: step: 152/529, loss: 1.3730931282043457 2023-01-21 08:36:53.198583: step: 156/529, loss: 0.5314227342605591 2023-01-21 08:36:54.327226: step: 160/529, loss: 1.093907117843628 2023-01-21 08:36:55.464502: step: 164/529, loss: 0.21213456988334656 2023-01-21 08:36:56.588283: step: 168/529, loss: 0.34765952825546265 2023-01-21 08:36:57.741342: step: 172/529, loss: 0.19783231616020203 2023-01-21 08:36:58.881210: step: 176/529, loss: 1.2791751623153687 2023-01-21 08:37:00.025408: step: 180/529, loss: 0.326324462890625 2023-01-21 08:37:01.175802: step: 184/529, loss: 0.37679052352905273 2023-01-21 08:37:02.307313: step: 188/529, loss: 2.1085660457611084 2023-01-21 08:37:03.439075: step: 192/529, loss: 0.64113450050354 2023-01-21 08:37:04.581710: step: 196/529, loss: 0.4685416519641876 2023-01-21 08:37:05.729830: step: 200/529, loss: 0.49039435386657715 2023-01-21 08:37:06.868541: step: 204/529, loss: 0.23786889016628265 2023-01-21 08:37:08.005965: step: 208/529, loss: 0.3385983407497406 2023-01-21 08:37:09.137865: step: 212/529, loss: 0.40551871061325073 2023-01-21 08:37:10.271747: step: 216/529, loss: 0.9544097781181335 2023-01-21 08:37:11.405068: step: 220/529, loss: 0.3216747045516968 2023-01-21 08:37:12.551679: step: 224/529, loss: 1.4213862419128418 2023-01-21 08:37:13.723353: step: 228/529, loss: 0.3096015155315399 2023-01-21 08:37:14.843998: step: 232/529, loss: 0.7316606044769287 2023-01-21 08:37:15.994722: step: 236/529, loss: 0.821317195892334 2023-01-21 08:37:17.110225: step: 240/529, loss: 0.4682171940803528 2023-01-21 08:37:18.269597: step: 244/529, loss: 0.15753412246704102 2023-01-21 08:37:19.413244: step: 248/529, loss: 0.16952915489673615 2023-01-21 08:37:20.567549: step: 252/529, loss: 2.312734365463257 2023-01-21 08:37:21.731816: step: 256/529, loss: 0.39927130937576294 2023-01-21 08:37:22.856872: step: 260/529, loss: 0.14115343987941742 2023-01-21 08:37:24.005394: step: 264/529, loss: 0.1422864943742752 2023-01-21 08:37:25.157678: step: 268/529, loss: 0.5056096911430359 2023-01-21 08:37:26.283275: step: 272/529, loss: 0.18460464477539062 2023-01-21 08:37:27.429672: step: 276/529, loss: 0.32110393047332764 2023-01-21 08:37:28.582919: step: 280/529, loss: 0.37311363220214844 2023-01-21 08:37:29.713755: step: 284/529, loss: 2.1574289798736572 2023-01-21 08:37:30.845940: step: 288/529, loss: 0.9065800905227661 2023-01-21 08:37:31.987421: step: 292/529, loss: 0.18569979071617126 2023-01-21 08:37:33.140142: step: 296/529, loss: 0.12617188692092896 2023-01-21 08:37:34.273722: step: 300/529, loss: 0.18380004167556763 2023-01-21 08:37:35.417911: step: 304/529, loss: 0.5336170196533203 2023-01-21 08:37:36.570387: step: 308/529, loss: 0.48369672894477844 2023-01-21 08:37:37.730720: step: 312/529, loss: 0.7249584197998047 2023-01-21 08:37:38.916599: step: 316/529, loss: 0.8844048380851746 2023-01-21 08:37:40.060332: step: 320/529, loss: 0.3962167799472809 2023-01-21 08:37:41.229560: step: 324/529, loss: 0.41168874502182007 2023-01-21 08:37:42.380227: step: 328/529, loss: 0.9417327642440796 2023-01-21 08:37:43.524181: step: 332/529, loss: 1.018467664718628 2023-01-21 08:37:44.652524: step: 336/529, loss: 0.09948311001062393 2023-01-21 08:37:45.795993: step: 340/529, loss: 0.11509513854980469 2023-01-21 08:37:46.943770: step: 344/529, loss: 0.8259420394897461 2023-01-21 08:37:48.104544: step: 348/529, loss: 0.4823358654975891 2023-01-21 08:37:49.260652: step: 352/529, loss: 0.5422476530075073 2023-01-21 08:37:50.445673: step: 356/529, loss: 0.7489575147628784 2023-01-21 08:37:51.589126: step: 360/529, loss: 0.7969327569007874 2023-01-21 08:37:52.737764: step: 364/529, loss: 1.4077816009521484 2023-01-21 08:37:53.883579: step: 368/529, loss: 0.323464959859848 2023-01-21 08:37:55.032494: step: 372/529, loss: 0.15628428757190704 2023-01-21 08:37:56.209913: step: 376/529, loss: 0.45432502031326294 2023-01-21 08:37:57.361321: step: 380/529, loss: 0.3573431372642517 2023-01-21 08:37:58.514592: step: 384/529, loss: 1.9641146659851074 2023-01-21 08:37:59.677628: step: 388/529, loss: 0.45423832535743713 2023-01-21 08:38:00.821927: step: 392/529, loss: 0.8016027212142944 2023-01-21 08:38:02.011994: step: 396/529, loss: 0.6931349039077759 2023-01-21 08:38:03.157351: step: 400/529, loss: 7.130451679229736 2023-01-21 08:38:04.291383: step: 404/529, loss: 0.9619883894920349 2023-01-21 08:38:05.448893: step: 408/529, loss: 0.17553609609603882 2023-01-21 08:38:06.589979: step: 412/529, loss: 0.5371543169021606 2023-01-21 08:38:07.779951: step: 416/529, loss: 6.35456657409668 2023-01-21 08:38:08.974698: step: 420/529, loss: 1.8735235929489136 2023-01-21 08:38:10.148177: step: 424/529, loss: 0.31495970487594604 2023-01-21 08:38:11.328750: step: 428/529, loss: 3.0553712844848633 2023-01-21 08:38:12.475148: step: 432/529, loss: 0.32576435804367065 2023-01-21 08:38:13.595858: step: 436/529, loss: 0.578923225402832 2023-01-21 08:38:14.708447: step: 440/529, loss: 0.2525663375854492 2023-01-21 08:38:15.844963: step: 444/529, loss: 0.15729370713233948 2023-01-21 08:38:16.973920: step: 448/529, loss: 0.8245846033096313 2023-01-21 08:38:18.101682: step: 452/529, loss: 0.27101975679397583 2023-01-21 08:38:19.225804: step: 456/529, loss: 0.19498157501220703 2023-01-21 08:38:20.357657: step: 460/529, loss: 0.5603306293487549 2023-01-21 08:38:21.511257: step: 464/529, loss: 0.19061031937599182 2023-01-21 08:38:22.646429: step: 468/529, loss: 0.34526732563972473 2023-01-21 08:38:23.779616: step: 472/529, loss: 0.2579323649406433 2023-01-21 08:38:24.933937: step: 476/529, loss: 0.5323038101196289 2023-01-21 08:38:26.088040: step: 480/529, loss: 1.062608242034912 2023-01-21 08:38:27.225388: step: 484/529, loss: 1.7276530265808105 2023-01-21 08:38:28.362837: step: 488/529, loss: 2.9541053771972656 2023-01-21 08:38:29.485985: step: 492/529, loss: 0.38564595580101013 2023-01-21 08:38:30.612406: step: 496/529, loss: 0.35428586602211 2023-01-21 08:38:31.753696: step: 500/529, loss: 0.7991260886192322 2023-01-21 08:38:32.890971: step: 504/529, loss: 0.7846972942352295 2023-01-21 08:38:33.991624: step: 508/529, loss: 0.6294810771942139 2023-01-21 08:38:35.113762: step: 512/529, loss: 0.8370799422264099 2023-01-21 08:38:36.260393: step: 516/529, loss: 0.34018832445144653 2023-01-21 08:38:37.421458: step: 520/529, loss: 0.8810480237007141 2023-01-21 08:38:38.557154: step: 524/529, loss: 0.3134942948818207 2023-01-21 08:38:39.670755: step: 528/529, loss: 0.29537904262542725 2023-01-21 08:38:40.790357: step: 532/529, loss: 1.2235320806503296 2023-01-21 08:38:41.911960: step: 536/529, loss: 0.9157170653343201 2023-01-21 08:38:43.053669: step: 540/529, loss: 0.9095420241355896 2023-01-21 08:38:44.186160: step: 544/529, loss: 0.38371741771698 2023-01-21 08:38:45.313383: step: 548/529, loss: 0.26568323373794556 2023-01-21 08:38:46.466951: step: 552/529, loss: 0.7462350726127625 2023-01-21 08:38:47.602464: step: 556/529, loss: 0.2976147532463074 2023-01-21 08:38:48.749106: step: 560/529, loss: 0.2737061679363251 2023-01-21 08:38:49.895558: step: 564/529, loss: 6.999796390533447 2023-01-21 08:38:51.043204: step: 568/529, loss: 0.24717697501182556 2023-01-21 08:38:52.181256: step: 572/529, loss: 0.3453161120414734 2023-01-21 08:38:53.287758: step: 576/529, loss: 0.2021385133266449 2023-01-21 08:38:54.419063: step: 580/529, loss: 0.14663821458816528 2023-01-21 08:38:55.550907: step: 584/529, loss: 2.5010900497436523 2023-01-21 08:38:56.658720: step: 588/529, loss: 0.14363646507263184 2023-01-21 08:38:57.802476: step: 592/529, loss: 0.22648312151432037 2023-01-21 08:38:58.948741: step: 596/529, loss: 0.7444033622741699 2023-01-21 08:39:00.076906: step: 600/529, loss: 0.5878614783287048 2023-01-21 08:39:01.200639: step: 604/529, loss: 0.35113203525543213 2023-01-21 08:39:02.328120: step: 608/529, loss: 1.207403302192688 2023-01-21 08:39:03.513079: step: 612/529, loss: 0.9373064041137695 2023-01-21 08:39:04.652131: step: 616/529, loss: 0.3856774568557739 2023-01-21 08:39:05.797589: step: 620/529, loss: 1.5732369422912598 2023-01-21 08:39:06.947836: step: 624/529, loss: 1.8320481777191162 2023-01-21 08:39:08.089890: step: 628/529, loss: 0.4988988935947418 2023-01-21 08:39:09.236120: step: 632/529, loss: 0.2700865864753723 2023-01-21 08:39:10.375100: step: 636/529, loss: 0.9359628558158875 2023-01-21 08:39:11.529436: step: 640/529, loss: 1.0615077018737793 2023-01-21 08:39:12.655404: step: 644/529, loss: 0.056235421448946 2023-01-21 08:39:13.760429: step: 648/529, loss: 0.5603745579719543 2023-01-21 08:39:14.896139: step: 652/529, loss: 0.3206295967102051 2023-01-21 08:39:16.020047: step: 656/529, loss: 1.1626073122024536 2023-01-21 08:39:17.125944: step: 660/529, loss: 0.2522863745689392 2023-01-21 08:39:18.264598: step: 664/529, loss: 0.1974714696407318 2023-01-21 08:39:19.396346: step: 668/529, loss: 0.6952510476112366 2023-01-21 08:39:20.531658: step: 672/529, loss: 0.32597237825393677 2023-01-21 08:39:21.671844: step: 676/529, loss: 0.2884550094604492 2023-01-21 08:39:22.829946: step: 680/529, loss: 2.3008382320404053 2023-01-21 08:39:23.941098: step: 684/529, loss: 1.3631598949432373 2023-01-21 08:39:25.087684: step: 688/529, loss: 1.0244896411895752 2023-01-21 08:39:26.256610: step: 692/529, loss: 1.0603013038635254 2023-01-21 08:39:27.367423: step: 696/529, loss: 0.06894628703594208 2023-01-21 08:39:28.516801: step: 700/529, loss: 2.2512283325195312 2023-01-21 08:39:29.635590: step: 704/529, loss: 0.34750068187713623 2023-01-21 08:39:30.792328: step: 708/529, loss: 0.49512797594070435 2023-01-21 08:39:31.973655: step: 712/529, loss: 1.5319430828094482 2023-01-21 08:39:33.128443: step: 716/529, loss: 0.5386723875999451 2023-01-21 08:39:34.272149: step: 720/529, loss: 0.08035726845264435 2023-01-21 08:39:35.405732: step: 724/529, loss: 0.32039251923561096 2023-01-21 08:39:36.545527: step: 728/529, loss: 0.9971221685409546 2023-01-21 08:39:37.685444: step: 732/529, loss: 0.25700846314430237 2023-01-21 08:39:38.804813: step: 736/529, loss: 0.2144605666399002 2023-01-21 08:39:39.971326: step: 740/529, loss: 2.633221387863159 2023-01-21 08:39:41.105607: step: 744/529, loss: 0.8342758417129517 2023-01-21 08:39:42.251277: step: 748/529, loss: 0.6564019322395325 2023-01-21 08:39:43.376172: step: 752/529, loss: 0.36439400911331177 2023-01-21 08:39:44.514462: step: 756/529, loss: 0.1734904646873474 2023-01-21 08:39:45.658830: step: 760/529, loss: 2.699251174926758 2023-01-21 08:39:46.809004: step: 764/529, loss: 1.4306433200836182 2023-01-21 08:39:47.994853: step: 768/529, loss: 1.356773018836975 2023-01-21 08:39:49.103639: step: 772/529, loss: 0.7788138389587402 2023-01-21 08:39:50.246446: step: 776/529, loss: 0.48847952485084534 2023-01-21 08:39:51.385437: step: 780/529, loss: 0.9458274841308594 2023-01-21 08:39:52.529300: step: 784/529, loss: 0.39781731367111206 2023-01-21 08:39:53.615808: step: 788/529, loss: 0.7113245129585266 2023-01-21 08:39:54.727267: step: 792/529, loss: 0.3420921564102173 2023-01-21 08:39:55.846159: step: 796/529, loss: 0.8405942916870117 2023-01-21 08:39:56.995194: step: 800/529, loss: 0.12295912206172943 2023-01-21 08:39:58.165705: step: 804/529, loss: 0.49430298805236816 2023-01-21 08:39:59.317898: step: 808/529, loss: 0.8668269515037537 2023-01-21 08:40:00.445652: step: 812/529, loss: 0.827810525894165 2023-01-21 08:40:01.552142: step: 816/529, loss: 0.2077282965183258 2023-01-21 08:40:02.692461: step: 820/529, loss: 0.8789294362068176 2023-01-21 08:40:03.844135: step: 824/529, loss: 0.508680522441864 2023-01-21 08:40:04.966613: step: 828/529, loss: 6.765714645385742 2023-01-21 08:40:06.079986: step: 832/529, loss: 0.8015994429588318 2023-01-21 08:40:07.213508: step: 836/529, loss: 0.11436138302087784 2023-01-21 08:40:08.352569: step: 840/529, loss: 0.26067009568214417 2023-01-21 08:40:09.503219: step: 844/529, loss: 0.23226170241832733 2023-01-21 08:40:10.637344: step: 848/529, loss: 0.8968196511268616 2023-01-21 08:40:11.819241: step: 852/529, loss: 0.2878015637397766 2023-01-21 08:40:12.952290: step: 856/529, loss: 6.6833086013793945 2023-01-21 08:40:14.085848: step: 860/529, loss: 0.543400764465332 2023-01-21 08:40:15.179544: step: 864/529, loss: 2.3582990169525146 2023-01-21 08:40:16.301196: step: 868/529, loss: 0.17841282486915588 2023-01-21 08:40:17.415894: step: 872/529, loss: 0.7141642570495605 2023-01-21 08:40:18.541488: step: 876/529, loss: 1.2771152257919312 2023-01-21 08:40:19.685809: step: 880/529, loss: 0.4982832670211792 2023-01-21 08:40:20.867176: step: 884/529, loss: 0.40006905794143677 2023-01-21 08:40:21.989801: step: 888/529, loss: 0.2375883013010025 2023-01-21 08:40:23.112968: step: 892/529, loss: 1.8226087093353271 2023-01-21 08:40:24.247019: step: 896/529, loss: 0.2102208137512207 2023-01-21 08:40:25.371712: step: 900/529, loss: 0.3662320077419281 2023-01-21 08:40:26.490766: step: 904/529, loss: 0.11506939679384232 2023-01-21 08:40:27.611255: step: 908/529, loss: 0.8001874089241028 2023-01-21 08:40:28.717630: step: 912/529, loss: 0.12894801795482635 2023-01-21 08:40:29.856752: step: 916/529, loss: 0.18576350808143616 2023-01-21 08:40:30.990124: step: 920/529, loss: 0.10321883857250214 2023-01-21 08:40:32.100717: step: 924/529, loss: 0.2037675827741623 2023-01-21 08:40:33.234482: step: 928/529, loss: 0.21415062248706818 2023-01-21 08:40:34.326238: step: 932/529, loss: 0.3335719108581543 2023-01-21 08:40:35.494522: step: 936/529, loss: 1.5137025117874146 2023-01-21 08:40:36.636832: step: 940/529, loss: 0.22497105598449707 2023-01-21 08:40:37.743871: step: 944/529, loss: 0.2935396134853363 2023-01-21 08:40:38.889564: step: 948/529, loss: 1.4770570993423462 2023-01-21 08:40:40.040292: step: 952/529, loss: 0.23993521928787231 2023-01-21 08:40:41.170861: step: 956/529, loss: 0.7086160778999329 2023-01-21 08:40:42.318520: step: 960/529, loss: 0.05982933193445206 2023-01-21 08:40:43.467475: step: 964/529, loss: 0.06373167037963867 2023-01-21 08:40:44.601312: step: 968/529, loss: 0.17378145456314087 2023-01-21 08:40:45.737056: step: 972/529, loss: 0.24838675558567047 2023-01-21 08:40:46.877603: step: 976/529, loss: 0.8171840906143188 2023-01-21 08:40:48.024995: step: 980/529, loss: 0.9138569831848145 2023-01-21 08:40:49.162958: step: 984/529, loss: 1.0758111476898193 2023-01-21 08:40:50.310208: step: 988/529, loss: 0.19634565711021423 2023-01-21 08:40:51.445603: step: 992/529, loss: 0.21462908387184143 2023-01-21 08:40:52.580011: step: 996/529, loss: 0.9129133224487305 2023-01-21 08:40:53.685353: step: 1000/529, loss: 2.635676383972168 2023-01-21 08:40:54.835572: step: 1004/529, loss: 1.3215116262435913 2023-01-21 08:40:55.970414: step: 1008/529, loss: 0.224347785115242 2023-01-21 08:40:57.119956: step: 1012/529, loss: 0.718658983707428 2023-01-21 08:40:58.293329: step: 1016/529, loss: 0.24458342790603638 2023-01-21 08:40:59.439584: step: 1020/529, loss: 0.16220471262931824 2023-01-21 08:41:00.609957: step: 1024/529, loss: 0.12832693755626678 2023-01-21 08:41:01.729589: step: 1028/529, loss: 0.17081347107887268 2023-01-21 08:41:02.869118: step: 1032/529, loss: 0.3509206175804138 2023-01-21 08:41:04.035064: step: 1036/529, loss: 0.2427106350660324 2023-01-21 08:41:05.188587: step: 1040/529, loss: 0.18622225522994995 2023-01-21 08:41:06.318124: step: 1044/529, loss: 0.2536700367927551 2023-01-21 08:41:07.456206: step: 1048/529, loss: 0.7693371176719666 2023-01-21 08:41:08.596794: step: 1052/529, loss: 0.24589061737060547 2023-01-21 08:41:09.733445: step: 1056/529, loss: 0.2053605169057846 2023-01-21 08:41:10.883994: step: 1060/529, loss: 0.05548463016748428 2023-01-21 08:41:12.004049: step: 1064/529, loss: 1.7561016082763672 2023-01-21 08:41:13.147635: step: 1068/529, loss: 1.046337604522705 2023-01-21 08:41:14.279338: step: 1072/529, loss: 1.5373632907867432 2023-01-21 08:41:15.408144: step: 1076/529, loss: 1.1018669605255127 2023-01-21 08:41:16.543179: step: 1080/529, loss: 1.7216929197311401 2023-01-21 08:41:17.695355: step: 1084/529, loss: 1.964949607849121 2023-01-21 08:41:18.810873: step: 1088/529, loss: 0.8232311010360718 2023-01-21 08:41:19.995134: step: 1092/529, loss: 0.3604514002799988 2023-01-21 08:41:21.142438: step: 1096/529, loss: 0.8829132318496704 2023-01-21 08:41:22.266989: step: 1100/529, loss: 2.215179443359375 2023-01-21 08:41:23.421324: step: 1104/529, loss: 0.10780338943004608 2023-01-21 08:41:24.554030: step: 1108/529, loss: 0.2431640625 2023-01-21 08:41:25.695810: step: 1112/529, loss: 0.08399734646081924 2023-01-21 08:41:26.850429: step: 1116/529, loss: 1.0450273752212524 2023-01-21 08:41:28.050503: step: 1120/529, loss: 1.693611741065979 2023-01-21 08:41:29.175767: step: 1124/529, loss: 0.597059428691864 2023-01-21 08:41:30.336934: step: 1128/529, loss: 0.8682540059089661 2023-01-21 08:41:31.435582: step: 1132/529, loss: 0.14875009655952454 2023-01-21 08:41:32.598362: step: 1136/529, loss: 0.1848442256450653 2023-01-21 08:41:33.729934: step: 1140/529, loss: 0.5265677571296692 2023-01-21 08:41:34.869814: step: 1144/529, loss: 1.4149702787399292 2023-01-21 08:41:36.009796: step: 1148/529, loss: 0.5851609110832214 2023-01-21 08:41:37.152255: step: 1152/529, loss: 1.226862907409668 2023-01-21 08:41:38.304523: step: 1156/529, loss: 0.3521884083747864 2023-01-21 08:41:39.457521: step: 1160/529, loss: 0.2268299162387848 2023-01-21 08:41:40.567813: step: 1164/529, loss: 1.445133090019226 2023-01-21 08:41:41.718799: step: 1168/529, loss: 1.6017242670059204 2023-01-21 08:41:42.885359: step: 1172/529, loss: 0.2239668071269989 2023-01-21 08:41:44.030946: step: 1176/529, loss: 0.08118319511413574 2023-01-21 08:41:45.186495: step: 1180/529, loss: 0.35108959674835205 2023-01-21 08:41:46.288588: step: 1184/529, loss: 1.0763485431671143 2023-01-21 08:41:47.429957: step: 1188/529, loss: 0.34547722339630127 2023-01-21 08:41:48.578014: step: 1192/529, loss: 0.1340794563293457 2023-01-21 08:41:49.710756: step: 1196/529, loss: 0.22881336510181427 2023-01-21 08:41:50.835447: step: 1200/529, loss: 0.16314508020877838 2023-01-21 08:41:51.988985: step: 1204/529, loss: 0.6221767663955688 2023-01-21 08:41:53.140478: step: 1208/529, loss: 0.18148942291736603 2023-01-21 08:41:54.255649: step: 1212/529, loss: 1.5709794759750366 2023-01-21 08:41:55.381849: step: 1216/529, loss: 0.7450945377349854 2023-01-21 08:41:56.519365: step: 1220/529, loss: 0.5750439763069153 2023-01-21 08:41:57.682109: step: 1224/529, loss: 0.8094569444656372 2023-01-21 08:41:58.814692: step: 1228/529, loss: 0.48620858788490295 2023-01-21 08:41:59.959483: step: 1232/529, loss: 0.25180378556251526 2023-01-21 08:42:01.112402: step: 1236/529, loss: 0.4793487787246704 2023-01-21 08:42:02.262243: step: 1240/529, loss: 0.21152149140834808 2023-01-21 08:42:03.380336: step: 1244/529, loss: 0.26250505447387695 2023-01-21 08:42:04.516790: step: 1248/529, loss: 0.17898836731910706 2023-01-21 08:42:05.630618: step: 1252/529, loss: 0.09999523311853409 2023-01-21 08:42:06.760129: step: 1256/529, loss: 0.1858902871608734 2023-01-21 08:42:07.894660: step: 1260/529, loss: 0.5422772169113159 2023-01-21 08:42:09.049500: step: 1264/529, loss: 0.33639851212501526 2023-01-21 08:42:10.184315: step: 1268/529, loss: 0.9363800883293152 2023-01-21 08:42:11.305849: step: 1272/529, loss: 0.14035215973854065 2023-01-21 08:42:12.472636: step: 1276/529, loss: 1.1116856336593628 2023-01-21 08:42:13.607200: step: 1280/529, loss: 0.9366650581359863 2023-01-21 08:42:14.810737: step: 1284/529, loss: 0.21494418382644653 2023-01-21 08:42:15.964129: step: 1288/529, loss: 0.3201838731765747 2023-01-21 08:42:17.074998: step: 1292/529, loss: 0.4720112681388855 2023-01-21 08:42:18.209426: step: 1296/529, loss: 0.8141045570373535 2023-01-21 08:42:19.332669: step: 1300/529, loss: 0.44637930393218994 2023-01-21 08:42:20.496084: step: 1304/529, loss: 0.1693161129951477 2023-01-21 08:42:21.609462: step: 1308/529, loss: 0.46504440903663635 2023-01-21 08:42:22.760211: step: 1312/529, loss: 0.5144976377487183 2023-01-21 08:42:23.888307: step: 1316/529, loss: 6.5810418128967285 2023-01-21 08:42:25.043971: step: 1320/529, loss: 0.5250568389892578 2023-01-21 08:42:26.163878: step: 1324/529, loss: 0.10729971528053284 2023-01-21 08:42:27.323495: step: 1328/529, loss: 1.054152250289917 2023-01-21 08:42:28.447376: step: 1332/529, loss: 1.365271806716919 2023-01-21 08:42:29.570298: step: 1336/529, loss: 1.442115306854248 2023-01-21 08:42:30.698537: step: 1340/529, loss: 0.20402908325195312 2023-01-21 08:42:31.798455: step: 1344/529, loss: 0.14029577374458313 2023-01-21 08:42:32.925682: step: 1348/529, loss: 0.9411318898200989 2023-01-21 08:42:34.073066: step: 1352/529, loss: 0.1641758531332016 2023-01-21 08:42:35.243189: step: 1356/529, loss: 0.22643594443798065 2023-01-21 08:42:36.358300: step: 1360/529, loss: 0.9889498353004456 2023-01-21 08:42:37.470486: step: 1364/529, loss: 0.2504335343837738 2023-01-21 08:42:38.594026: step: 1368/529, loss: 1.7736448049545288 2023-01-21 08:42:39.708070: step: 1372/529, loss: 0.04648590087890625 2023-01-21 08:42:40.851262: step: 1376/529, loss: 0.5473880171775818 2023-01-21 08:42:41.968894: step: 1380/529, loss: 0.3631100654602051 2023-01-21 08:42:43.117619: step: 1384/529, loss: 0.24835491180419922 2023-01-21 08:42:44.346554: step: 1388/529, loss: 0.11867466568946838 2023-01-21 08:42:45.450545: step: 1392/529, loss: 0.2440529465675354 2023-01-21 08:42:46.583359: step: 1396/529, loss: 0.7663075923919678 2023-01-21 08:42:47.723583: step: 1400/529, loss: 1.7392289638519287 2023-01-21 08:42:48.854237: step: 1404/529, loss: 0.19356174767017365 2023-01-21 08:42:49.999232: step: 1408/529, loss: 0.24853840470314026 2023-01-21 08:42:51.139814: step: 1412/529, loss: 1.6402099132537842 2023-01-21 08:42:52.273650: step: 1416/529, loss: 0.6658819317817688 2023-01-21 08:42:53.408547: step: 1420/529, loss: 0.4736892580986023 2023-01-21 08:42:54.559189: step: 1424/529, loss: 0.31594592332839966 2023-01-21 08:42:55.708956: step: 1428/529, loss: 0.4358847737312317 2023-01-21 08:42:56.842686: step: 1432/529, loss: 1.840773344039917 2023-01-21 08:42:57.986912: step: 1436/529, loss: 0.26872050762176514 2023-01-21 08:42:59.131311: step: 1440/529, loss: 0.09251470863819122 2023-01-21 08:43:00.261329: step: 1444/529, loss: 0.23535729944705963 2023-01-21 08:43:01.424933: step: 1448/529, loss: 0.5126436948776245 2023-01-21 08:43:02.571857: step: 1452/529, loss: 0.15987567603588104 2023-01-21 08:43:03.700289: step: 1456/529, loss: 0.06912603229284286 2023-01-21 08:43:04.815862: step: 1460/529, loss: 1.303166151046753 2023-01-21 08:43:05.977904: step: 1464/529, loss: 0.8047708868980408 2023-01-21 08:43:07.129677: step: 1468/529, loss: 0.31621819734573364 2023-01-21 08:43:08.261788: step: 1472/529, loss: 0.08525316417217255 2023-01-21 08:43:09.381475: step: 1476/529, loss: 6.623735427856445 2023-01-21 08:43:10.506380: step: 1480/529, loss: 0.9232900738716125 2023-01-21 08:43:11.648634: step: 1484/529, loss: 0.4921996295452118 2023-01-21 08:43:12.792610: step: 1488/529, loss: 0.23620501160621643 2023-01-21 08:43:13.921884: step: 1492/529, loss: 0.7446185350418091 2023-01-21 08:43:15.039675: step: 1496/529, loss: 0.7989373207092285 2023-01-21 08:43:16.184144: step: 1500/529, loss: 0.07127318531274796 2023-01-21 08:43:17.329382: step: 1504/529, loss: 0.30706846714019775 2023-01-21 08:43:18.466237: step: 1508/529, loss: 1.816852331161499 2023-01-21 08:43:19.606556: step: 1512/529, loss: 0.8129447102546692 2023-01-21 08:43:20.736801: step: 1516/529, loss: 0.8326485753059387 2023-01-21 08:43:21.867059: step: 1520/529, loss: 0.8802800178527832 2023-01-21 08:43:23.026574: step: 1524/529, loss: 0.2800569534301758 2023-01-21 08:43:24.143666: step: 1528/529, loss: 0.20654058456420898 2023-01-21 08:43:25.276189: step: 1532/529, loss: 0.09551505744457245 2023-01-21 08:43:26.375421: step: 1536/529, loss: 0.08555688709020615 2023-01-21 08:43:27.520940: step: 1540/529, loss: 1.3692773580551147 2023-01-21 08:43:28.638247: step: 1544/529, loss: 1.1589257717132568 2023-01-21 08:43:29.752368: step: 1548/529, loss: 0.7796484231948853 2023-01-21 08:43:30.874030: step: 1552/529, loss: 0.7295999526977539 2023-01-21 08:43:32.013840: step: 1556/529, loss: 0.18316173553466797 2023-01-21 08:43:33.147866: step: 1560/529, loss: 0.6112227439880371 2023-01-21 08:43:34.288085: step: 1564/529, loss: 0.10929808765649796 2023-01-21 08:43:35.421069: step: 1568/529, loss: 6.765477180480957 2023-01-21 08:43:36.572903: step: 1572/529, loss: 1.2350817918777466 2023-01-21 08:43:37.726096: step: 1576/529, loss: 0.07628726959228516 2023-01-21 08:43:38.847004: step: 1580/529, loss: 0.8256204128265381 2023-01-21 08:43:39.992057: step: 1584/529, loss: 0.26255807280540466 2023-01-21 08:43:41.117271: step: 1588/529, loss: 0.2382659912109375 2023-01-21 08:43:42.270635: step: 1592/529, loss: 0.8648579120635986 2023-01-21 08:43:43.407707: step: 1596/529, loss: 0.6343369483947754 2023-01-21 08:43:44.547872: step: 1600/529, loss: 0.8459938764572144 2023-01-21 08:43:45.705220: step: 1604/529, loss: 0.7642489671707153 2023-01-21 08:43:46.825932: step: 1608/529, loss: 0.7206225395202637 2023-01-21 08:43:47.959712: step: 1612/529, loss: 0.5299168825149536 2023-01-21 08:43:49.128276: step: 1616/529, loss: 1.0863205194473267 2023-01-21 08:43:50.251918: step: 1620/529, loss: 0.5170079469680786 2023-01-21 08:43:51.390068: step: 1624/529, loss: 0.5441691279411316 2023-01-21 08:43:52.542318: step: 1628/529, loss: 0.06420707702636719 2023-01-21 08:43:53.709807: step: 1632/529, loss: 6.329038619995117 2023-01-21 08:43:54.819413: step: 1636/529, loss: 0.13987036049365997 2023-01-21 08:43:55.959791: step: 1640/529, loss: 0.12047194689512253 2023-01-21 08:43:57.065152: step: 1644/529, loss: 1.3959094285964966 2023-01-21 08:43:58.210269: step: 1648/529, loss: 0.1417022943496704 2023-01-21 08:43:59.330994: step: 1652/529, loss: 0.26220405101776123 2023-01-21 08:44:00.445385: step: 1656/529, loss: 0.09691682457923889 2023-01-21 08:44:01.606368: step: 1660/529, loss: 0.5304877758026123 2023-01-21 08:44:02.733039: step: 1664/529, loss: 0.8642672300338745 2023-01-21 08:44:03.853012: step: 1668/529, loss: 1.105774164199829 2023-01-21 08:44:04.992943: step: 1672/529, loss: 0.11097776889801025 2023-01-21 08:44:06.126052: step: 1676/529, loss: 0.39009782671928406 2023-01-21 08:44:07.258480: step: 1680/529, loss: 0.9966732263565063 2023-01-21 08:44:08.413550: step: 1684/529, loss: 0.3803766965866089 2023-01-21 08:44:09.568034: step: 1688/529, loss: 0.35071516036987305 2023-01-21 08:44:10.690807: step: 1692/529, loss: 1.15390145778656 2023-01-21 08:44:11.815789: step: 1696/529, loss: 1.5842915773391724 2023-01-21 08:44:12.936908: step: 1700/529, loss: 0.20530223846435547 2023-01-21 08:44:14.031600: step: 1704/529, loss: 0.41568660736083984 2023-01-21 08:44:15.203178: step: 1708/529, loss: 0.4050375819206238 2023-01-21 08:44:16.391781: step: 1712/529, loss: 0.7066041231155396 2023-01-21 08:44:17.546099: step: 1716/529, loss: 0.3591589033603668 2023-01-21 08:44:18.685596: step: 1720/529, loss: 0.1619473397731781 2023-01-21 08:44:19.832963: step: 1724/529, loss: 0.22097691893577576 2023-01-21 08:44:20.965042: step: 1728/529, loss: 0.3964705467224121 2023-01-21 08:44:22.113393: step: 1732/529, loss: 0.19676485657691956 2023-01-21 08:44:23.263437: step: 1736/529, loss: 0.7947767972946167 2023-01-21 08:44:24.374440: step: 1740/529, loss: 0.5161329507827759 2023-01-21 08:44:25.506820: step: 1744/529, loss: 0.8569895625114441 2023-01-21 08:44:26.645960: step: 1748/529, loss: 0.8217809200286865 2023-01-21 08:44:27.787289: step: 1752/529, loss: 1.4282824993133545 2023-01-21 08:44:28.946159: step: 1756/529, loss: 0.6928896903991699 2023-01-21 08:44:30.044669: step: 1760/529, loss: 0.12821082770824432 2023-01-21 08:44:31.146382: step: 1764/529, loss: 7.222170829772949 2023-01-21 08:44:32.290639: step: 1768/529, loss: 0.2352183312177658 2023-01-21 08:44:33.413763: step: 1772/529, loss: 7.337364196777344 2023-01-21 08:44:34.550188: step: 1776/529, loss: 0.10053012520074844 2023-01-21 08:44:35.722572: step: 1780/529, loss: 0.22173920273780823 2023-01-21 08:44:36.863021: step: 1784/529, loss: 0.7717358469963074 2023-01-21 08:44:37.998884: step: 1788/529, loss: 0.150108203291893 2023-01-21 08:44:39.135099: step: 1792/529, loss: 0.2151103913784027 2023-01-21 08:44:40.283978: step: 1796/529, loss: 0.27949219942092896 2023-01-21 08:44:41.415776: step: 1800/529, loss: 0.6257975101470947 2023-01-21 08:44:42.556565: step: 1804/529, loss: 0.6478987336158752 2023-01-21 08:44:43.691151: step: 1808/529, loss: 1.05632746219635 2023-01-21 08:44:44.836539: step: 1812/529, loss: 0.9121267199516296 2023-01-21 08:44:46.021652: step: 1816/529, loss: 0.17341919243335724 2023-01-21 08:44:47.155895: step: 1820/529, loss: 0.7128279209136963 2023-01-21 08:44:48.268220: step: 1824/529, loss: 0.9977483153343201 2023-01-21 08:44:49.428962: step: 1828/529, loss: 1.5038676261901855 2023-01-21 08:44:50.567882: step: 1832/529, loss: 0.9541134834289551 2023-01-21 08:44:51.730667: step: 1836/529, loss: 0.2863430380821228 2023-01-21 08:44:52.860304: step: 1840/529, loss: 0.06166362762451172 2023-01-21 08:44:54.029439: step: 1844/529, loss: 0.17570121586322784 2023-01-21 08:44:55.191143: step: 1848/529, loss: 0.4174157977104187 2023-01-21 08:44:56.358711: step: 1852/529, loss: 1.4150358438491821 2023-01-21 08:44:57.530168: step: 1856/529, loss: 0.8669883012771606 2023-01-21 08:44:58.658478: step: 1860/529, loss: 0.2999749183654785 2023-01-21 08:44:59.814820: step: 1864/529, loss: 0.9722987413406372 2023-01-21 08:45:00.953301: step: 1868/529, loss: 0.18898595869541168 2023-01-21 08:45:02.100418: step: 1872/529, loss: 0.4102143943309784 2023-01-21 08:45:03.233446: step: 1876/529, loss: 0.7225392460823059 2023-01-21 08:45:04.361430: step: 1880/529, loss: 0.6729050874710083 2023-01-21 08:45:05.502534: step: 1884/529, loss: 1.8289964199066162 2023-01-21 08:45:06.653868: step: 1888/529, loss: 0.9194282293319702 2023-01-21 08:45:07.803118: step: 1892/529, loss: 0.14319868385791779 2023-01-21 08:45:08.928279: step: 1896/529, loss: 0.5910947322845459 2023-01-21 08:45:10.078388: step: 1900/529, loss: 1.8206111192703247 2023-01-21 08:45:11.188911: step: 1904/529, loss: 0.12756066024303436 2023-01-21 08:45:12.308406: step: 1908/529, loss: 1.415897250175476 2023-01-21 08:45:13.472368: step: 1912/529, loss: 2.8653218746185303 2023-01-21 08:45:14.601106: step: 1916/529, loss: 0.8677077293395996 2023-01-21 08:45:15.754498: step: 1920/529, loss: 0.42419806122779846 2023-01-21 08:45:16.884064: step: 1924/529, loss: 0.1219908744096756 2023-01-21 08:45:18.022513: step: 1928/529, loss: 0.2571730613708496 2023-01-21 08:45:19.139901: step: 1932/529, loss: 0.7489161491394043 2023-01-21 08:45:20.281862: step: 1936/529, loss: 0.18201252818107605 2023-01-21 08:45:21.432307: step: 1940/529, loss: 3.432046413421631 2023-01-21 08:45:22.584287: step: 1944/529, loss: 0.17417903244495392 2023-01-21 08:45:23.726093: step: 1948/529, loss: 0.45939940214157104 2023-01-21 08:45:24.882811: step: 1952/529, loss: 1.9692161083221436 2023-01-21 08:45:26.022960: step: 1956/529, loss: 0.1637752503156662 2023-01-21 08:45:27.127029: step: 1960/529, loss: 0.2635403871536255 2023-01-21 08:45:28.261158: step: 1964/529, loss: 6.100759506225586 2023-01-21 08:45:29.401039: step: 1968/529, loss: 0.6198025345802307 2023-01-21 08:45:30.535835: step: 1972/529, loss: 0.23389548063278198 2023-01-21 08:45:31.686225: step: 1976/529, loss: 0.2064833641052246 2023-01-21 08:45:32.846959: step: 1980/529, loss: 0.16781291365623474 2023-01-21 08:45:33.976605: step: 1984/529, loss: 0.2910175323486328 2023-01-21 08:45:35.097735: step: 1988/529, loss: 7.207370281219482 2023-01-21 08:45:36.244434: step: 1992/529, loss: 0.35911956429481506 2023-01-21 08:45:37.381024: step: 1996/529, loss: 0.27823057770729065 2023-01-21 08:45:38.535608: step: 2000/529, loss: 0.3757549524307251 2023-01-21 08:45:39.717653: step: 2004/529, loss: 1.7029109001159668 2023-01-21 08:45:40.874952: step: 2008/529, loss: 0.2338418960571289 2023-01-21 08:45:42.027594: step: 2012/529, loss: 0.5623645782470703 2023-01-21 08:45:43.189464: step: 2016/529, loss: 0.3921836018562317 2023-01-21 08:45:44.324321: step: 2020/529, loss: 0.2561310827732086 2023-01-21 08:45:45.456121: step: 2024/529, loss: 1.0950664281845093 2023-01-21 08:45:46.611682: step: 2028/529, loss: 0.35538214445114136 2023-01-21 08:45:47.776723: step: 2032/529, loss: 0.4026249051094055 2023-01-21 08:45:48.912114: step: 2036/529, loss: 0.156175896525383 2023-01-21 08:45:50.062075: step: 2040/529, loss: 1.4233074188232422 2023-01-21 08:45:51.200412: step: 2044/529, loss: 0.4461545944213867 2023-01-21 08:45:52.327649: step: 2048/529, loss: 0.1356600821018219 2023-01-21 08:45:53.484259: step: 2052/529, loss: 0.618218183517456 2023-01-21 08:45:54.645958: step: 2056/529, loss: 1.7878608703613281 2023-01-21 08:45:55.761235: step: 2060/529, loss: 0.48725080490112305 2023-01-21 08:45:56.868295: step: 2064/529, loss: 0.09457807242870331 2023-01-21 08:45:57.999276: step: 2068/529, loss: 0.21639785170555115 2023-01-21 08:45:59.131861: step: 2072/529, loss: 0.37085646390914917 2023-01-21 08:46:00.264360: step: 2076/529, loss: 0.17226460576057434 2023-01-21 08:46:01.389699: step: 2080/529, loss: 0.736858069896698 2023-01-21 08:46:02.541594: step: 2084/529, loss: 1.7456613779067993 2023-01-21 08:46:03.691844: step: 2088/529, loss: 0.16766414046287537 2023-01-21 08:46:04.842533: step: 2092/529, loss: 0.11730727553367615 2023-01-21 08:46:05.982667: step: 2096/529, loss: 0.34722986817359924 2023-01-21 08:46:07.121794: step: 2100/529, loss: 0.12117743492126465 2023-01-21 08:46:08.242635: step: 2104/529, loss: 0.3582167625427246 2023-01-21 08:46:09.370175: step: 2108/529, loss: 0.20710162818431854 2023-01-21 08:46:10.497896: step: 2112/529, loss: 0.06176753342151642 2023-01-21 08:46:11.628741: step: 2116/529, loss: 0.7002752423286438 ================================================== Loss: 0.846 -------------------- Dev: {'event': {'p': 0.5645677694770544, 'r': 0.7043941411451398, 'f1': 0.6267772511848342}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Test: {'event': {'p': 0.5636722606120435, 'r': 0.6616454229432214, 'f1': 0.6087420042643923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Chinese: {'event': {'p': 0.5135135135135135, 'r': 0.7037037037037037, 'f1': 0.59375}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Korean: {'event': {'p': 0.6086956521739131, 'r': 0.4444444444444444, 'f1': 0.5137614678899082}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Russian: {'event': {'p': 0.5, 'r': 0.4444444444444444, 'f1': 0.47058823529411764}, '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': 0.5645677694770544, 'r': 0.7043941411451398, 'f1': 0.6267772511848342}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Chinese: {'event': {'p': 0.5636722606120435, 'r': 0.6616454229432214, 'f1': 0.6087420042643923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Chinese: {'event': {'p': 0.5135135135135135, 'r': 0.7037037037037037, 'f1': 0.59375}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Korean: {'event': {'p': 0.5645677694770544, 'r': 0.7043941411451398, 'f1': 0.6267772511848342}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5636722606120435, 'r': 0.6616454229432214, 'f1': 0.6087420042643923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.6086956521739131, 'r': 0.4444444444444444, 'f1': 0.5137614678899082}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.5645677694770544, 'r': 0.7043941411451398, 'f1': 0.6267772511848342}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5636722606120435, 'r': 0.6616454229432214, 'f1': 0.6087420042643923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.5, 'r': 0.4444444444444444, 'f1': 0.47058823529411764}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} ****************************** Epoch: 2 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:47:08.800498: step: 4/529, loss: 0.16011753678321838 2023-01-21 08:47:09.957742: step: 8/529, loss: 0.3136522173881531 2023-01-21 08:47:11.087841: step: 12/529, loss: 0.5542739033699036 2023-01-21 08:47:12.246614: step: 16/529, loss: 0.12885847687721252 2023-01-21 08:47:13.406044: step: 20/529, loss: 0.17488440871238708 2023-01-21 08:47:14.526464: step: 24/529, loss: 0.16168662905693054 2023-01-21 08:47:15.640789: step: 28/529, loss: 0.12315917015075684 2023-01-21 08:47:16.770181: step: 32/529, loss: 0.6722848415374756 2023-01-21 08:47:17.889527: step: 36/529, loss: 0.5296027064323425 2023-01-21 08:47:19.033177: step: 40/529, loss: 0.2444693148136139 2023-01-21 08:47:20.160280: step: 44/529, loss: 0.6960229277610779 2023-01-21 08:47:21.332321: step: 48/529, loss: 1.7689166069030762 2023-01-21 08:47:22.443041: step: 52/529, loss: 0.17338839173316956 2023-01-21 08:47:23.568073: step: 56/529, loss: 0.11341724544763565 2023-01-21 08:47:24.686507: step: 60/529, loss: 0.13741794228553772 2023-01-21 08:47:25.853838: step: 64/529, loss: 0.13269048929214478 2023-01-21 08:47:27.030086: step: 68/529, loss: 0.17899766564369202 2023-01-21 08:47:28.150390: step: 72/529, loss: 0.6834906339645386 2023-01-21 08:47:29.307840: step: 76/529, loss: 0.23811855912208557 2023-01-21 08:47:30.447666: step: 80/529, loss: 0.21603471040725708 2023-01-21 08:47:31.579747: step: 84/529, loss: 2.2758147716522217 2023-01-21 08:47:32.717158: step: 88/529, loss: 0.26220351457595825 2023-01-21 08:47:33.842548: step: 92/529, loss: 0.16112929582595825 2023-01-21 08:47:34.993601: step: 96/529, loss: 0.12959790229797363 2023-01-21 08:47:36.152159: step: 100/529, loss: 0.7131268978118896 2023-01-21 08:47:37.313269: step: 104/529, loss: 0.23847992718219757 2023-01-21 08:47:38.447560: step: 108/529, loss: 1.1560603380203247 2023-01-21 08:47:39.592080: step: 112/529, loss: 0.32345473766326904 2023-01-21 08:47:40.759436: step: 116/529, loss: 0.17514057457447052 2023-01-21 08:47:41.936441: step: 120/529, loss: 0.7883058786392212 2023-01-21 08:47:43.065153: step: 124/529, loss: 0.3674681782722473 2023-01-21 08:47:44.220570: step: 128/529, loss: 0.9620416760444641 2023-01-21 08:47:45.369540: step: 132/529, loss: 1.2430660724639893 2023-01-21 08:47:46.490581: step: 136/529, loss: 6.324300289154053 2023-01-21 08:47:47.624417: step: 140/529, loss: 0.22229023277759552 2023-01-21 08:47:48.773617: step: 144/529, loss: 0.7625331878662109 2023-01-21 08:47:49.912169: step: 148/529, loss: 0.3625265955924988 2023-01-21 08:47:51.053591: step: 152/529, loss: 0.1423037052154541 2023-01-21 08:47:52.187366: step: 156/529, loss: 0.4354117512702942 2023-01-21 08:47:53.343969: step: 160/529, loss: 0.7554001212120056 2023-01-21 08:47:54.478263: step: 164/529, loss: 0.8097216486930847 2023-01-21 08:47:55.624988: step: 168/529, loss: 0.33486807346343994 2023-01-21 08:47:56.756464: step: 172/529, loss: 0.2294527143239975 2023-01-21 08:47:57.895326: step: 176/529, loss: 0.12679247558116913 2023-01-21 08:47:59.010721: step: 180/529, loss: 0.08267946541309357 2023-01-21 08:48:00.164407: step: 184/529, loss: 0.225310280919075 2023-01-21 08:48:01.303534: step: 188/529, loss: 0.3811991810798645 2023-01-21 08:48:02.459706: step: 192/529, loss: 0.25362318754196167 2023-01-21 08:48:03.591170: step: 196/529, loss: 1.0539844036102295 2023-01-21 08:48:04.736283: step: 200/529, loss: 0.4539848566055298 2023-01-21 08:48:05.897377: step: 204/529, loss: 0.17168235778808594 2023-01-21 08:48:07.036522: step: 208/529, loss: 0.6570345163345337 2023-01-21 08:48:08.156999: step: 212/529, loss: 1.1744686365127563 2023-01-21 08:48:09.278880: step: 216/529, loss: 0.29787909984588623 2023-01-21 08:48:10.414826: step: 220/529, loss: 0.20865727961063385 2023-01-21 08:48:11.547593: step: 224/529, loss: 0.19810311496257782 2023-01-21 08:48:12.684538: step: 228/529, loss: 0.18043336272239685 2023-01-21 08:48:13.837107: step: 232/529, loss: 1.6988426446914673 2023-01-21 08:48:15.001501: step: 236/529, loss: 0.3392007350921631 2023-01-21 08:48:16.149653: step: 240/529, loss: 0.3021160364151001 2023-01-21 08:48:17.277802: step: 244/529, loss: 1.0553247928619385 2023-01-21 08:48:18.449700: step: 248/529, loss: 0.28210583329200745 2023-01-21 08:48:19.595052: step: 252/529, loss: 1.4673709869384766 2023-01-21 08:48:20.752496: step: 256/529, loss: 0.29661792516708374 2023-01-21 08:48:21.897862: step: 260/529, loss: 0.2706844210624695 2023-01-21 08:48:22.999204: step: 264/529, loss: 0.7277727127075195 2023-01-21 08:48:24.125258: step: 268/529, loss: 0.6652582287788391 2023-01-21 08:48:25.293683: step: 272/529, loss: 6.562233924865723 2023-01-21 08:48:26.434341: step: 276/529, loss: 1.0673260688781738 2023-01-21 08:48:27.581693: step: 280/529, loss: 0.10092764347791672 2023-01-21 08:48:28.754755: step: 284/529, loss: 0.4907453656196594 2023-01-21 08:48:29.888560: step: 288/529, loss: 0.09231443703174591 2023-01-21 08:48:31.021853: step: 292/529, loss: 1.4154690504074097 2023-01-21 08:48:32.165473: step: 296/529, loss: 0.11383305490016937 2023-01-21 08:48:33.295741: step: 300/529, loss: 1.0224688053131104 2023-01-21 08:48:34.443012: step: 304/529, loss: 0.1355007290840149 2023-01-21 08:48:35.557025: step: 308/529, loss: 0.682678759098053 2023-01-21 08:48:36.694246: step: 312/529, loss: 1.8023496866226196 2023-01-21 08:48:37.822967: step: 316/529, loss: 0.3560555577278137 2023-01-21 08:48:38.954842: step: 320/529, loss: 0.1748853325843811 2023-01-21 08:48:40.099208: step: 324/529, loss: 0.6619178056716919 2023-01-21 08:48:41.239539: step: 328/529, loss: 0.36525458097457886 2023-01-21 08:48:42.402618: step: 332/529, loss: 1.289191484451294 2023-01-21 08:48:43.542983: step: 336/529, loss: 0.14037008583545685 2023-01-21 08:48:44.691836: step: 340/529, loss: 7.26772403717041 2023-01-21 08:48:45.839999: step: 344/529, loss: 2.330707550048828 2023-01-21 08:48:46.996478: step: 348/529, loss: 1.4237453937530518 2023-01-21 08:48:48.116704: step: 352/529, loss: 0.3275485038757324 2023-01-21 08:48:49.254691: step: 356/529, loss: 0.26687249541282654 2023-01-21 08:48:50.390227: step: 360/529, loss: 0.6357746124267578 2023-01-21 08:48:51.528337: step: 364/529, loss: 0.17802830040454865 2023-01-21 08:48:52.646698: step: 368/529, loss: 0.7167764902114868 2023-01-21 08:48:53.810551: step: 372/529, loss: 0.24290715157985687 2023-01-21 08:48:54.998829: step: 376/529, loss: 0.28208112716674805 2023-01-21 08:48:56.128253: step: 380/529, loss: 0.3086233139038086 2023-01-21 08:48:57.287876: step: 384/529, loss: 0.7601917386054993 2023-01-21 08:48:58.428105: step: 388/529, loss: 0.3973931074142456 2023-01-21 08:48:59.579010: step: 392/529, loss: 0.2916305661201477 2023-01-21 08:49:00.736783: step: 396/529, loss: 0.18600140511989594 2023-01-21 08:49:01.908572: step: 400/529, loss: 0.554033100605011 2023-01-21 08:49:03.072637: step: 404/529, loss: 0.43686628341674805 2023-01-21 08:49:04.203730: step: 408/529, loss: 0.8465619683265686 2023-01-21 08:49:05.352251: step: 412/529, loss: 1.2262346744537354 2023-01-21 08:49:06.507487: step: 416/529, loss: 0.2643786370754242 2023-01-21 08:49:07.643619: step: 420/529, loss: 1.0234525203704834 2023-01-21 08:49:08.805095: step: 424/529, loss: 0.35052046179771423 2023-01-21 08:49:09.965754: step: 428/529, loss: 1.3906512260437012 2023-01-21 08:49:11.101062: step: 432/529, loss: 0.5203295946121216 2023-01-21 08:49:12.220449: step: 436/529, loss: 0.4099210798740387 2023-01-21 08:49:13.340423: step: 440/529, loss: 0.018064402043819427 2023-01-21 08:49:14.463204: step: 444/529, loss: 0.252826988697052 2023-01-21 08:49:15.590413: step: 448/529, loss: 0.159762442111969 2023-01-21 08:49:16.742300: step: 452/529, loss: 0.12250981479883194 2023-01-21 08:49:17.888735: step: 456/529, loss: 0.21491505205631256 2023-01-21 08:49:19.028778: step: 460/529, loss: 0.22326365113258362 2023-01-21 08:49:20.182886: step: 464/529, loss: 1.591773271560669 2023-01-21 08:49:21.356424: step: 468/529, loss: 0.6827014684677124 2023-01-21 08:49:22.504505: step: 472/529, loss: 0.3413742780685425 2023-01-21 08:49:23.659564: step: 476/529, loss: 0.08217153698205948 2023-01-21 08:49:24.837338: step: 480/529, loss: 0.17546749114990234 2023-01-21 08:49:26.002167: step: 484/529, loss: 1.4722208976745605 2023-01-21 08:49:27.159004: step: 488/529, loss: 0.3064320981502533 2023-01-21 08:49:28.303886: step: 492/529, loss: 0.11648903787136078 2023-01-21 08:49:29.480539: step: 496/529, loss: 0.6317901611328125 2023-01-21 08:49:30.651173: step: 500/529, loss: 0.5103319883346558 2023-01-21 08:49:31.787085: step: 504/529, loss: 0.673570454120636 2023-01-21 08:49:32.910494: step: 508/529, loss: 0.6077795028686523 2023-01-21 08:49:34.032286: step: 512/529, loss: 0.10362100601196289 2023-01-21 08:49:35.168883: step: 516/529, loss: 0.1989048421382904 2023-01-21 08:49:36.310089: step: 520/529, loss: 0.15940414369106293 2023-01-21 08:49:37.455662: step: 524/529, loss: 0.49244728684425354 2023-01-21 08:49:38.596072: step: 528/529, loss: 0.11199302226305008 2023-01-21 08:49:39.773381: step: 532/529, loss: 0.35654306411743164 2023-01-21 08:49:40.926831: step: 536/529, loss: 0.38375359773635864 2023-01-21 08:49:42.052830: step: 540/529, loss: 0.2046581357717514 2023-01-21 08:49:43.228495: step: 544/529, loss: 0.2462368756532669 2023-01-21 08:49:44.357213: step: 548/529, loss: 0.730583667755127 2023-01-21 08:49:45.497504: step: 552/529, loss: 1.0266247987747192 2023-01-21 08:49:46.644192: step: 556/529, loss: 0.784329891204834 2023-01-21 08:49:47.764478: step: 560/529, loss: 0.9169520735740662 2023-01-21 08:49:48.923843: step: 564/529, loss: 0.5217580795288086 2023-01-21 08:49:50.069822: step: 568/529, loss: 1.5671823024749756 2023-01-21 08:49:51.233545: step: 572/529, loss: 0.19533872604370117 2023-01-21 08:49:52.353625: step: 576/529, loss: 0.5604658722877502 2023-01-21 08:49:53.480691: step: 580/529, loss: 0.11667013168334961 2023-01-21 08:49:54.632291: step: 584/529, loss: 0.5202959775924683 2023-01-21 08:49:55.772610: step: 588/529, loss: 0.3391176164150238 2023-01-21 08:49:56.910799: step: 592/529, loss: 0.8964076638221741 2023-01-21 08:49:58.026430: step: 596/529, loss: 0.657941997051239 2023-01-21 08:49:59.180669: step: 600/529, loss: 0.588658332824707 2023-01-21 08:50:00.326223: step: 604/529, loss: 0.21640311181545258 2023-01-21 08:50:01.487544: step: 608/529, loss: 0.20219768583774567 2023-01-21 08:50:02.646820: step: 612/529, loss: 0.29336604475975037 2023-01-21 08:50:03.786212: step: 616/529, loss: 0.2765015661716461 2023-01-21 08:50:04.930520: step: 620/529, loss: 0.5687368512153625 2023-01-21 08:50:06.066627: step: 624/529, loss: 0.15059643983840942 2023-01-21 08:50:07.241579: step: 628/529, loss: 0.5599428415298462 2023-01-21 08:50:08.355663: step: 632/529, loss: 0.7464699745178223 2023-01-21 08:50:09.488676: step: 636/529, loss: 1.5670126676559448 2023-01-21 08:50:10.622721: step: 640/529, loss: 0.05278778076171875 2023-01-21 08:50:11.785446: step: 644/529, loss: 0.2779834270477295 2023-01-21 08:50:12.941589: step: 648/529, loss: 0.18645945191383362 2023-01-21 08:50:14.073768: step: 652/529, loss: 0.7625888586044312 2023-01-21 08:50:15.214346: step: 656/529, loss: 0.2967225909233093 2023-01-21 08:50:16.353583: step: 660/529, loss: 0.21448002755641937 2023-01-21 08:50:17.481275: step: 664/529, loss: 0.2279062271118164 2023-01-21 08:50:18.669923: step: 668/529, loss: 0.12128783017396927 2023-01-21 08:50:19.825153: step: 672/529, loss: 1.1565241813659668 2023-01-21 08:50:20.948504: step: 676/529, loss: 0.23948058485984802 2023-01-21 08:50:22.079431: step: 680/529, loss: 0.30275630950927734 2023-01-21 08:50:23.239475: step: 684/529, loss: 0.19882088899612427 2023-01-21 08:50:24.376685: step: 688/529, loss: 0.7269073724746704 2023-01-21 08:50:25.517798: step: 692/529, loss: 0.098810575902462 2023-01-21 08:50:26.634739: step: 696/529, loss: 0.5192068815231323 2023-01-21 08:50:27.803715: step: 700/529, loss: 0.8430692553520203 2023-01-21 08:50:28.968703: step: 704/529, loss: 0.4456140995025635 2023-01-21 08:50:30.110193: step: 708/529, loss: 0.22368068993091583 2023-01-21 08:50:31.263303: step: 712/529, loss: 0.3451046049594879 2023-01-21 08:50:32.387749: step: 716/529, loss: 0.33338505029678345 2023-01-21 08:50:33.489792: step: 720/529, loss: 0.7560536861419678 2023-01-21 08:50:34.618932: step: 724/529, loss: 0.7425456047058105 2023-01-21 08:50:35.736880: step: 728/529, loss: 0.689315676689148 2023-01-21 08:50:36.886336: step: 732/529, loss: 0.31961071491241455 2023-01-21 08:50:38.005982: step: 736/529, loss: 0.08549609035253525 2023-01-21 08:50:39.127270: step: 740/529, loss: 0.4769206643104553 2023-01-21 08:50:40.250488: step: 744/529, loss: 0.1587519347667694 2023-01-21 08:50:41.384574: step: 748/529, loss: 0.28562304377555847 2023-01-21 08:50:42.538144: step: 752/529, loss: 0.4195876121520996 2023-01-21 08:50:43.694927: step: 756/529, loss: 0.13163384795188904 2023-01-21 08:50:44.830339: step: 760/529, loss: 0.07811227440834045 2023-01-21 08:50:45.971924: step: 764/529, loss: 0.23471862077713013 2023-01-21 08:50:47.088355: step: 768/529, loss: 0.503345787525177 2023-01-21 08:50:48.193695: step: 772/529, loss: 0.1733681708574295 2023-01-21 08:50:49.340498: step: 776/529, loss: 0.8007156252861023 2023-01-21 08:50:50.450038: step: 780/529, loss: 0.7773919105529785 2023-01-21 08:50:51.592010: step: 784/529, loss: 0.09906230121850967 2023-01-21 08:50:52.703012: step: 788/529, loss: 0.0492280013859272 2023-01-21 08:50:53.882996: step: 792/529, loss: 0.16614651679992676 2023-01-21 08:50:55.037101: step: 796/529, loss: 0.12801270186901093 2023-01-21 08:50:56.150695: step: 800/529, loss: 0.17420612275600433 2023-01-21 08:50:57.301369: step: 804/529, loss: 0.9932152628898621 2023-01-21 08:50:58.421933: step: 808/529, loss: 0.7149071097373962 2023-01-21 08:50:59.581854: step: 812/529, loss: 0.24240341782569885 2023-01-21 08:51:00.732203: step: 816/529, loss: 0.8995043635368347 2023-01-21 08:51:01.872023: step: 820/529, loss: 0.4197057783603668 2023-01-21 08:51:03.012869: step: 824/529, loss: 0.37795573472976685 2023-01-21 08:51:04.174686: step: 828/529, loss: 0.5616082549095154 2023-01-21 08:51:05.324285: step: 832/529, loss: 0.662162721157074 2023-01-21 08:51:06.464927: step: 836/529, loss: 0.19101658463478088 2023-01-21 08:51:07.615825: step: 840/529, loss: 0.10286326706409454 2023-01-21 08:51:08.745774: step: 844/529, loss: 0.3165660798549652 2023-01-21 08:51:09.892025: step: 848/529, loss: 0.15785284340381622 2023-01-21 08:51:11.024445: step: 852/529, loss: 0.07431665062904358 2023-01-21 08:51:12.134400: step: 856/529, loss: 0.23629015684127808 2023-01-21 08:51:13.273882: step: 860/529, loss: 0.21047420799732208 2023-01-21 08:51:14.401447: step: 864/529, loss: 0.08362946659326553 2023-01-21 08:51:15.535106: step: 868/529, loss: 0.12692776322364807 2023-01-21 08:51:16.660658: step: 872/529, loss: 0.09655408561229706 2023-01-21 08:51:17.799517: step: 876/529, loss: 0.4454402029514313 2023-01-21 08:51:18.948489: step: 880/529, loss: 0.25781354308128357 2023-01-21 08:51:20.112122: step: 884/529, loss: 0.8869230151176453 2023-01-21 08:51:21.224115: step: 888/529, loss: 0.05918631702661514 2023-01-21 08:51:22.332975: step: 892/529, loss: 0.5052001476287842 2023-01-21 08:51:23.519988: step: 896/529, loss: 0.9153575301170349 2023-01-21 08:51:24.659974: step: 900/529, loss: 0.14046001434326172 2023-01-21 08:51:25.793838: step: 904/529, loss: 0.09355268627405167 2023-01-21 08:51:26.962612: step: 908/529, loss: 1.9980554580688477 2023-01-21 08:51:28.108533: step: 912/529, loss: 0.488818883895874 2023-01-21 08:51:29.246288: step: 916/529, loss: 0.13364887237548828 2023-01-21 08:51:30.385503: step: 920/529, loss: 0.8750467896461487 2023-01-21 08:51:31.538217: step: 924/529, loss: 0.5229588747024536 2023-01-21 08:51:32.656108: step: 928/529, loss: 0.6437985301017761 2023-01-21 08:51:33.788005: step: 932/529, loss: 0.17561788856983185 2023-01-21 08:51:34.959037: step: 936/529, loss: 1.4316258430480957 2023-01-21 08:51:36.109909: step: 940/529, loss: 0.09925102442502975 2023-01-21 08:51:37.250303: step: 944/529, loss: 0.13579416275024414 2023-01-21 08:51:38.387501: step: 948/529, loss: 0.21429473161697388 2023-01-21 08:51:39.526844: step: 952/529, loss: 0.2562198340892792 2023-01-21 08:51:40.654419: step: 956/529, loss: 1.0596842765808105 2023-01-21 08:51:41.815610: step: 960/529, loss: 0.26569002866744995 2023-01-21 08:51:42.969316: step: 964/529, loss: 0.17212048172950745 2023-01-21 08:51:44.108313: step: 968/529, loss: 0.3943077027797699 2023-01-21 08:51:45.260157: step: 972/529, loss: 0.9857510328292847 2023-01-21 08:51:46.414593: step: 976/529, loss: 0.8594427108764648 2023-01-21 08:51:47.555056: step: 980/529, loss: 0.75419682264328 2023-01-21 08:51:48.683700: step: 984/529, loss: 2.1439573764801025 2023-01-21 08:51:49.816840: step: 988/529, loss: 0.22956937551498413 2023-01-21 08:51:50.953605: step: 992/529, loss: 0.9550902247428894 2023-01-21 08:51:52.147615: step: 996/529, loss: 0.23789557814598083 2023-01-21 08:51:53.287310: step: 1000/529, loss: 0.7501827478408813 2023-01-21 08:51:54.437648: step: 1004/529, loss: 1.0087507963180542 2023-01-21 08:51:55.570472: step: 1008/529, loss: 0.166591078042984 2023-01-21 08:51:56.707120: step: 1012/529, loss: 0.14258627593517303 2023-01-21 08:51:57.850688: step: 1016/529, loss: 0.3370937407016754 2023-01-21 08:51:59.024203: step: 1020/529, loss: 0.13173122704029083 2023-01-21 08:52:00.140987: step: 1024/529, loss: 0.6296415328979492 2023-01-21 08:52:01.257700: step: 1028/529, loss: 0.02834026701748371 2023-01-21 08:52:02.480222: step: 1032/529, loss: 0.19032001495361328 2023-01-21 08:52:03.613082: step: 1036/529, loss: 0.06784754246473312 2023-01-21 08:52:04.774779: step: 1040/529, loss: 0.35349151492118835 2023-01-21 08:52:05.916648: step: 1044/529, loss: 0.4109952747821808 2023-01-21 08:52:07.050804: step: 1048/529, loss: 0.3357512354850769 2023-01-21 08:52:08.175707: step: 1052/529, loss: 1.1619939804077148 2023-01-21 08:52:09.322959: step: 1056/529, loss: 0.13449373841285706 2023-01-21 08:52:10.489248: step: 1060/529, loss: 0.771314263343811 2023-01-21 08:52:11.647876: step: 1064/529, loss: 1.1261049509048462 2023-01-21 08:52:12.786731: step: 1068/529, loss: 0.6880071759223938 2023-01-21 08:52:13.916431: step: 1072/529, loss: 0.3649477958679199 2023-01-21 08:52:15.078588: step: 1076/529, loss: 1.7755614519119263 2023-01-21 08:52:16.223759: step: 1080/529, loss: 0.1534440666437149 2023-01-21 08:52:17.394669: step: 1084/529, loss: 0.4672432839870453 2023-01-21 08:52:18.529244: step: 1088/529, loss: 0.2176845520734787 2023-01-21 08:52:19.690271: step: 1092/529, loss: 0.23515939712524414 2023-01-21 08:52:20.819577: step: 1096/529, loss: 0.3562561273574829 2023-01-21 08:52:21.926920: step: 1100/529, loss: 0.18892565369606018 2023-01-21 08:52:23.109377: step: 1104/529, loss: 0.24174949526786804 2023-01-21 08:52:24.213744: step: 1108/529, loss: 0.03390555456280708 2023-01-21 08:52:25.328952: step: 1112/529, loss: 0.17155544459819794 2023-01-21 08:52:26.457438: step: 1116/529, loss: 1.00954270362854 2023-01-21 08:52:27.618504: step: 1120/529, loss: 0.2654804289340973 2023-01-21 08:52:28.735533: step: 1124/529, loss: 0.23189465701580048 2023-01-21 08:52:29.853584: step: 1128/529, loss: 0.5436960458755493 2023-01-21 08:52:30.983542: step: 1132/529, loss: 0.30535823106765747 2023-01-21 08:52:32.102073: step: 1136/529, loss: 0.13110123574733734 2023-01-21 08:52:33.252679: step: 1140/529, loss: 0.059475041925907135 2023-01-21 08:52:34.384520: step: 1144/529, loss: 0.14658494293689728 2023-01-21 08:52:35.495909: step: 1148/529, loss: 0.1200718879699707 2023-01-21 08:52:36.647950: step: 1152/529, loss: 0.3168865144252777 2023-01-21 08:52:37.802060: step: 1156/529, loss: 0.6561247110366821 2023-01-21 08:52:38.938669: step: 1160/529, loss: 0.302634060382843 2023-01-21 08:52:40.096754: step: 1164/529, loss: 0.19865378737449646 2023-01-21 08:52:41.241832: step: 1168/529, loss: 0.4789091944694519 2023-01-21 08:52:42.357079: step: 1172/529, loss: 0.8969265222549438 2023-01-21 08:52:43.512697: step: 1176/529, loss: 1.3511192798614502 2023-01-21 08:52:44.624707: step: 1180/529, loss: 0.03264591842889786 2023-01-21 08:52:45.765445: step: 1184/529, loss: 0.25470343232154846 2023-01-21 08:52:46.888947: step: 1188/529, loss: 7.005441188812256 2023-01-21 08:52:48.029986: step: 1192/529, loss: 1.1059383153915405 2023-01-21 08:52:49.157565: step: 1196/529, loss: 0.22476397454738617 2023-01-21 08:52:50.300789: step: 1200/529, loss: 0.8762972354888916 2023-01-21 08:52:51.428278: step: 1204/529, loss: 0.23472939431667328 2023-01-21 08:52:52.571389: step: 1208/529, loss: 0.2966242730617523 2023-01-21 08:52:53.686885: step: 1212/529, loss: 0.22277450561523438 2023-01-21 08:52:54.792736: step: 1216/529, loss: 0.6476133465766907 2023-01-21 08:52:55.963926: step: 1220/529, loss: 0.16679368913173676 2023-01-21 08:52:57.096583: step: 1224/529, loss: 0.2518458962440491 2023-01-21 08:52:58.218167: step: 1228/529, loss: 0.5916452407836914 2023-01-21 08:52:59.336670: step: 1232/529, loss: 0.37040069699287415 2023-01-21 08:53:00.475066: step: 1236/529, loss: 0.1746373176574707 2023-01-21 08:53:01.607583: step: 1240/529, loss: 0.8113880157470703 2023-01-21 08:53:02.747383: step: 1244/529, loss: 0.3751646876335144 2023-01-21 08:53:03.889089: step: 1248/529, loss: 0.5634957551956177 2023-01-21 08:53:05.023796: step: 1252/529, loss: 0.6778720617294312 2023-01-21 08:53:06.173755: step: 1256/529, loss: 0.19265109300613403 2023-01-21 08:53:07.310365: step: 1260/529, loss: 0.18457737565040588 2023-01-21 08:53:08.499945: step: 1264/529, loss: 0.3128463625907898 2023-01-21 08:53:09.677236: step: 1268/529, loss: 0.1766294538974762 2023-01-21 08:53:10.820643: step: 1272/529, loss: 0.27493372559547424 2023-01-21 08:53:11.960456: step: 1276/529, loss: 0.5111462473869324 2023-01-21 08:53:13.095566: step: 1280/529, loss: 0.4772043228149414 2023-01-21 08:53:14.212620: step: 1284/529, loss: 0.10342197865247726 2023-01-21 08:53:15.375682: step: 1288/529, loss: 0.2782687246799469 2023-01-21 08:53:16.524861: step: 1292/529, loss: 0.8069081902503967 2023-01-21 08:53:17.673135: step: 1296/529, loss: 1.9833240509033203 2023-01-21 08:53:18.823719: step: 1300/529, loss: 0.049399662762880325 2023-01-21 08:53:19.985472: step: 1304/529, loss: 0.3477020263671875 2023-01-21 08:53:21.121819: step: 1308/529, loss: 0.5692951083183289 2023-01-21 08:53:22.287714: step: 1312/529, loss: 6.946808338165283 2023-01-21 08:53:23.437052: step: 1316/529, loss: 0.8236302137374878 2023-01-21 08:53:24.584220: step: 1320/529, loss: 0.35909515619277954 2023-01-21 08:53:25.724019: step: 1324/529, loss: 0.1959819793701172 2023-01-21 08:53:26.872781: step: 1328/529, loss: 0.7148933410644531 2023-01-21 08:53:28.013796: step: 1332/529, loss: 0.3113003969192505 2023-01-21 08:53:29.210669: step: 1336/529, loss: 0.12670226395130157 2023-01-21 08:53:30.368404: step: 1340/529, loss: 0.24998851120471954 2023-01-21 08:53:31.508549: step: 1344/529, loss: 1.0941404104232788 2023-01-21 08:53:32.636591: step: 1348/529, loss: 0.8368428349494934 2023-01-21 08:53:33.794551: step: 1352/529, loss: 0.5432189702987671 2023-01-21 08:53:34.954625: step: 1356/529, loss: 1.8350986242294312 2023-01-21 08:53:36.070989: step: 1360/529, loss: 0.18263152241706848 2023-01-21 08:53:37.198440: step: 1364/529, loss: 0.06184806674718857 2023-01-21 08:53:38.323681: step: 1368/529, loss: 1.4730383157730103 2023-01-21 08:53:39.434798: step: 1372/529, loss: 0.1663825958967209 2023-01-21 08:53:40.578479: step: 1376/529, loss: 0.08552595973014832 2023-01-21 08:53:41.707369: step: 1380/529, loss: 0.8457716703414917 2023-01-21 08:53:42.832135: step: 1384/529, loss: 0.8865984082221985 2023-01-21 08:53:43.967491: step: 1388/529, loss: 0.9729822278022766 2023-01-21 08:53:45.085313: step: 1392/529, loss: 0.2225530594587326 2023-01-21 08:53:46.244525: step: 1396/529, loss: 0.23109284043312073 2023-01-21 08:53:47.382392: step: 1400/529, loss: 0.23125630617141724 2023-01-21 08:53:48.529988: step: 1404/529, loss: 0.22260676324367523 2023-01-21 08:53:49.679633: step: 1408/529, loss: 0.2076948583126068 2023-01-21 08:53:50.831492: step: 1412/529, loss: 0.21892911195755005 2023-01-21 08:53:51.995628: step: 1416/529, loss: 1.2129038572311401 2023-01-21 08:53:53.128771: step: 1420/529, loss: 0.14834395051002502 2023-01-21 08:53:54.276481: step: 1424/529, loss: 0.23918914794921875 2023-01-21 08:53:55.396762: step: 1428/529, loss: 0.35814207792282104 2023-01-21 08:53:56.513826: step: 1432/529, loss: 0.8250369429588318 2023-01-21 08:53:57.690942: step: 1436/529, loss: 3.754572868347168 2023-01-21 08:53:58.816185: step: 1440/529, loss: 7.024494171142578 2023-01-21 08:53:59.996226: step: 1444/529, loss: 1.5178947448730469 2023-01-21 08:54:01.101835: step: 1448/529, loss: 0.0458315871655941 2023-01-21 08:54:02.280525: step: 1452/529, loss: 0.22355583310127258 2023-01-21 08:54:03.401664: step: 1456/529, loss: 0.764581561088562 2023-01-21 08:54:04.592563: step: 1460/529, loss: 2.560023307800293 2023-01-21 08:54:05.712626: step: 1464/529, loss: 2.1982405185699463 2023-01-21 08:54:06.863417: step: 1468/529, loss: 0.18284812569618225 2023-01-21 08:54:07.986171: step: 1472/529, loss: 0.22098904848098755 2023-01-21 08:54:09.098671: step: 1476/529, loss: 0.3310770094394684 2023-01-21 08:54:10.274145: step: 1480/529, loss: 0.21488434076309204 2023-01-21 08:54:11.404633: step: 1484/529, loss: 0.24475297331809998 2023-01-21 08:54:12.550181: step: 1488/529, loss: 0.7137027382850647 2023-01-21 08:54:13.678781: step: 1492/529, loss: 0.3269401788711548 2023-01-21 08:54:14.794204: step: 1496/529, loss: 0.2582850456237793 2023-01-21 08:54:15.943634: step: 1500/529, loss: 0.08873625099658966 2023-01-21 08:54:17.077424: step: 1504/529, loss: 0.517112135887146 2023-01-21 08:54:18.197176: step: 1508/529, loss: 0.6780310273170471 2023-01-21 08:54:19.341669: step: 1512/529, loss: 0.19513577222824097 2023-01-21 08:54:20.510565: step: 1516/529, loss: 0.2572776675224304 2023-01-21 08:54:21.622150: step: 1520/529, loss: 1.0514582395553589 2023-01-21 08:54:22.733888: step: 1524/529, loss: 0.19755320250988007 2023-01-21 08:54:23.884057: step: 1528/529, loss: 0.26384878158569336 2023-01-21 08:54:25.011694: step: 1532/529, loss: 0.16261273622512817 2023-01-21 08:54:26.139384: step: 1536/529, loss: 0.1483904868364334 2023-01-21 08:54:27.279999: step: 1540/529, loss: 0.1921371966600418 2023-01-21 08:54:28.451248: step: 1544/529, loss: 1.8900600671768188 2023-01-21 08:54:29.588883: step: 1548/529, loss: 0.16470623016357422 2023-01-21 08:54:30.737532: step: 1552/529, loss: 0.17478471994400024 2023-01-21 08:54:31.851489: step: 1556/529, loss: 0.06803274154663086 2023-01-21 08:54:32.991765: step: 1560/529, loss: 0.18762320280075073 2023-01-21 08:54:34.140668: step: 1564/529, loss: 0.11039657890796661 2023-01-21 08:54:35.287442: step: 1568/529, loss: 0.12208385765552521 2023-01-21 08:54:36.454260: step: 1572/529, loss: 0.18015176057815552 2023-01-21 08:54:37.611612: step: 1576/529, loss: 1.9510889053344727 2023-01-21 08:54:38.749542: step: 1580/529, loss: 0.07773247361183167 2023-01-21 08:54:39.893438: step: 1584/529, loss: 1.0601160526275635 2023-01-21 08:54:41.040386: step: 1588/529, loss: 0.8043298125267029 2023-01-21 08:54:42.199797: step: 1592/529, loss: 0.25013047456741333 2023-01-21 08:54:43.351239: step: 1596/529, loss: 0.5774046778678894 2023-01-21 08:54:44.504772: step: 1600/529, loss: 6.909578323364258 2023-01-21 08:54:45.663069: step: 1604/529, loss: 0.24420300126075745 2023-01-21 08:54:46.798855: step: 1608/529, loss: 0.15039558708667755 2023-01-21 08:54:47.919782: step: 1612/529, loss: 0.6981691122055054 2023-01-21 08:54:49.041186: step: 1616/529, loss: 0.1107785701751709 2023-01-21 08:54:50.167004: step: 1620/529, loss: 0.6491174697875977 2023-01-21 08:54:51.304317: step: 1624/529, loss: 0.14282435178756714 2023-01-21 08:54:52.450085: step: 1628/529, loss: 0.04771938547492027 2023-01-21 08:54:53.598700: step: 1632/529, loss: 0.1727065145969391 2023-01-21 08:54:54.744025: step: 1636/529, loss: 0.37428656220436096 2023-01-21 08:54:55.880856: step: 1640/529, loss: 0.7700635194778442 2023-01-21 08:54:57.017191: step: 1644/529, loss: 0.639403760433197 2023-01-21 08:54:58.186700: step: 1648/529, loss: 0.11602506041526794 2023-01-21 08:54:59.345524: step: 1652/529, loss: 0.8400684595108032 2023-01-21 08:55:00.484253: step: 1656/529, loss: 0.13578824698925018 2023-01-21 08:55:01.635772: step: 1660/529, loss: 0.8040549159049988 2023-01-21 08:55:02.774706: step: 1664/529, loss: 0.0938500389456749 2023-01-21 08:55:03.927889: step: 1668/529, loss: 0.26375675201416016 2023-01-21 08:55:05.056744: step: 1672/529, loss: 0.6129226684570312 2023-01-21 08:55:06.235778: step: 1676/529, loss: 0.06168346479535103 2023-01-21 08:55:07.401723: step: 1680/529, loss: 1.5032662153244019 2023-01-21 08:55:08.542385: step: 1684/529, loss: 0.40120744705200195 2023-01-21 08:55:09.693683: step: 1688/529, loss: 0.5830522775650024 2023-01-21 08:55:10.919888: step: 1692/529, loss: 2.0823144912719727 2023-01-21 08:55:12.090187: step: 1696/529, loss: 0.20201969146728516 2023-01-21 08:55:13.209074: step: 1700/529, loss: 0.1854238510131836 2023-01-21 08:55:14.361020: step: 1704/529, loss: 0.12755604088306427 2023-01-21 08:55:15.497028: step: 1708/529, loss: 0.1539296656847 2023-01-21 08:55:16.654891: step: 1712/529, loss: 0.9455440640449524 2023-01-21 08:55:17.789717: step: 1716/529, loss: 0.44399499893188477 2023-01-21 08:55:18.907449: step: 1720/529, loss: 0.0912221223115921 2023-01-21 08:55:20.056663: step: 1724/529, loss: 0.46971315145492554 2023-01-21 08:55:21.199263: step: 1728/529, loss: 1.4169028997421265 2023-01-21 08:55:22.344698: step: 1732/529, loss: 0.11658430099487305 2023-01-21 08:55:23.498722: step: 1736/529, loss: 0.3496878743171692 2023-01-21 08:55:24.627344: step: 1740/529, loss: 0.08327837288379669 2023-01-21 08:55:25.791921: step: 1744/529, loss: 0.8752743601799011 2023-01-21 08:55:26.943298: step: 1748/529, loss: 0.2983238101005554 2023-01-21 08:55:28.072651: step: 1752/529, loss: 0.11087103188037872 2023-01-21 08:55:29.201281: step: 1756/529, loss: 0.24697303771972656 2023-01-21 08:55:30.343427: step: 1760/529, loss: 0.19345569610595703 2023-01-21 08:55:31.463488: step: 1764/529, loss: 0.9849398136138916 2023-01-21 08:55:32.639882: step: 1768/529, loss: 0.2449812889099121 2023-01-21 08:55:33.790801: step: 1772/529, loss: 0.18153142929077148 2023-01-21 08:55:34.966153: step: 1776/529, loss: 0.27847251296043396 2023-01-21 08:55:36.099747: step: 1780/529, loss: 0.7282560467720032 2023-01-21 08:55:37.227776: step: 1784/529, loss: 0.04258241504430771 2023-01-21 08:55:38.372613: step: 1788/529, loss: 0.22480377554893494 2023-01-21 08:55:39.519009: step: 1792/529, loss: 0.9145678281784058 2023-01-21 08:55:40.649569: step: 1796/529, loss: 2.241812229156494 2023-01-21 08:55:41.794823: step: 1800/529, loss: 0.16483697295188904 2023-01-21 08:55:42.926725: step: 1804/529, loss: 0.1204107254743576 2023-01-21 08:55:44.079304: step: 1808/529, loss: 0.6715775728225708 2023-01-21 08:55:45.238949: step: 1812/529, loss: 0.35002368688583374 2023-01-21 08:55:46.392536: step: 1816/529, loss: 0.7887779474258423 2023-01-21 08:55:47.537362: step: 1820/529, loss: 0.08474154770374298 2023-01-21 08:55:48.690104: step: 1824/529, loss: 0.7663998603820801 2023-01-21 08:55:49.843049: step: 1828/529, loss: 0.23556175827980042 2023-01-21 08:55:50.983686: step: 1832/529, loss: 0.0884857177734375 2023-01-21 08:55:52.113006: step: 1836/529, loss: 0.6963865756988525 2023-01-21 08:55:53.260569: step: 1840/529, loss: 0.5266475677490234 2023-01-21 08:55:54.414284: step: 1844/529, loss: 1.689920425415039 2023-01-21 08:55:55.557479: step: 1848/529, loss: 0.47878068685531616 2023-01-21 08:55:56.707035: step: 1852/529, loss: 0.6409839391708374 2023-01-21 08:55:57.814359: step: 1856/529, loss: 0.7571682333946228 2023-01-21 08:55:58.946912: step: 1860/529, loss: 0.14327040314674377 2023-01-21 08:56:00.066030: step: 1864/529, loss: 0.22532731294631958 2023-01-21 08:56:01.218764: step: 1868/529, loss: 0.7403206825256348 2023-01-21 08:56:02.366653: step: 1872/529, loss: 0.08780007809400558 2023-01-21 08:56:03.476799: step: 1876/529, loss: 0.8261210918426514 2023-01-21 08:56:04.621362: step: 1880/529, loss: 0.42685550451278687 2023-01-21 08:56:05.761113: step: 1884/529, loss: 0.12489864230155945 2023-01-21 08:56:06.925154: step: 1888/529, loss: 0.08488745987415314 2023-01-21 08:56:08.072133: step: 1892/529, loss: 0.17247600853443146 2023-01-21 08:56:09.187463: step: 1896/529, loss: 0.3552228510379791 2023-01-21 08:56:10.300163: step: 1900/529, loss: 0.05527188628911972 2023-01-21 08:56:11.445645: step: 1904/529, loss: 0.1169508546590805 2023-01-21 08:56:12.600212: step: 1908/529, loss: 0.08105888962745667 2023-01-21 08:56:13.747296: step: 1912/529, loss: 0.3276294469833374 2023-01-21 08:56:14.890172: step: 1916/529, loss: 0.7120423316955566 2023-01-21 08:56:16.026351: step: 1920/529, loss: 0.3325200080871582 2023-01-21 08:56:17.157116: step: 1924/529, loss: 0.22089600563049316 2023-01-21 08:56:18.305887: step: 1928/529, loss: 0.12485527992248535 2023-01-21 08:56:19.456870: step: 1932/529, loss: 8.238045692443848 2023-01-21 08:56:20.592601: step: 1936/529, loss: 0.7729694843292236 2023-01-21 08:56:21.714536: step: 1940/529, loss: 0.13526931405067444 2023-01-21 08:56:22.853969: step: 1944/529, loss: 1.0348960161209106 2023-01-21 08:56:24.027780: step: 1948/529, loss: 0.2054780125617981 2023-01-21 08:56:25.168699: step: 1952/529, loss: 0.8352066278457642 2023-01-21 08:56:26.295379: step: 1956/529, loss: 0.8822604417800903 2023-01-21 08:56:27.452634: step: 1960/529, loss: 0.7498101592063904 2023-01-21 08:56:28.593305: step: 1964/529, loss: 6.706881046295166 2023-01-21 08:56:29.710946: step: 1968/529, loss: 0.3287595808506012 2023-01-21 08:56:30.832540: step: 1972/529, loss: 0.15849485993385315 2023-01-21 08:56:31.997473: step: 1976/529, loss: 0.1749209463596344 2023-01-21 08:56:33.136924: step: 1980/529, loss: 0.050691843032836914 2023-01-21 08:56:34.272454: step: 1984/529, loss: 0.14543476700782776 2023-01-21 08:56:35.449388: step: 1988/529, loss: 0.11504574120044708 2023-01-21 08:56:36.565161: step: 1992/529, loss: 0.2853316366672516 2023-01-21 08:56:37.697399: step: 1996/529, loss: 0.9961269497871399 2023-01-21 08:56:38.833206: step: 2000/529, loss: 0.07696647942066193 2023-01-21 08:56:39.959232: step: 2004/529, loss: 0.7733392119407654 2023-01-21 08:56:41.115910: step: 2008/529, loss: 0.6258556246757507 2023-01-21 08:56:42.235073: step: 2012/529, loss: 0.5575348734855652 2023-01-21 08:56:43.360139: step: 2016/529, loss: 0.07689771801233292 2023-01-21 08:56:44.519047: step: 2020/529, loss: 0.11228857189416885 2023-01-21 08:56:45.629817: step: 2024/529, loss: 0.18749314546585083 2023-01-21 08:56:46.769420: step: 2028/529, loss: 0.2973749041557312 2023-01-21 08:56:47.913976: step: 2032/529, loss: 0.12729129195213318 2023-01-21 08:56:49.072423: step: 2036/529, loss: 0.11492805182933807 2023-01-21 08:56:50.194687: step: 2040/529, loss: 0.022746801376342773 2023-01-21 08:56:51.323633: step: 2044/529, loss: 0.2855968475341797 2023-01-21 08:56:52.462698: step: 2048/529, loss: 0.1572956144809723 2023-01-21 08:56:53.612466: step: 2052/529, loss: 1.50773286819458 2023-01-21 08:56:54.739886: step: 2056/529, loss: 0.18963176012039185 2023-01-21 08:56:55.884367: step: 2060/529, loss: 0.13837838172912598 2023-01-21 08:56:57.021121: step: 2064/529, loss: 0.061939336359500885 2023-01-21 08:56:58.182069: step: 2068/529, loss: 0.855870246887207 2023-01-21 08:56:59.351957: step: 2072/529, loss: 1.0453215837478638 2023-01-21 08:57:00.485343: step: 2076/529, loss: 1.4882158041000366 2023-01-21 08:57:01.680504: step: 2080/529, loss: 2.292067050933838 2023-01-21 08:57:02.819456: step: 2084/529, loss: 0.5006760358810425 2023-01-21 08:57:03.928739: step: 2088/529, loss: 0.03980712965130806 2023-01-21 08:57:05.043299: step: 2092/529, loss: 0.14607982337474823 2023-01-21 08:57:06.166339: step: 2096/529, loss: 1.2448679208755493 2023-01-21 08:57:07.305679: step: 2100/529, loss: 0.10721039772033691 2023-01-21 08:57:08.436844: step: 2104/529, loss: 0.7397187948226929 2023-01-21 08:57:09.535898: step: 2108/529, loss: 1.00004243850708 2023-01-21 08:57:10.668825: step: 2112/529, loss: 0.19689975678920746 2023-01-21 08:57:11.866504: step: 2116/529, loss: 0.7404813766479492 ================================================== Loss: 0.603 -------------------- Dev: {'event': {'p': 0.574869109947644, 'r': 0.7310252996005326, 'f1': 0.6436107854630715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Test: {'event': {'p': 0.5623836126629422, 'r': 0.6998841251448435, 'f1': 0.6236448115642746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Chinese: {'event': {'p': 0.5238095238095238, 'r': 0.8148148148148148, 'f1': 0.6376811594202898}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Korean: {'event': {'p': 0.7142857142857143, 'r': 0.47619047619047616, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Russian: {'event': {'p': 0.525, 'r': 0.5833333333333334, 'f1': 0.5526315789473685}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574869109947644, 'r': 0.7310252996005326, 'f1': 0.6436107854630715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.5623836126629422, 'r': 0.6998841251448435, 'f1': 0.6236448115642746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.5238095238095238, 'r': 0.8148148148148148, 'f1': 0.6376811594202898}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.574869109947644, 'r': 0.7310252996005326, 'f1': 0.6436107854630715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.5623836126629422, 'r': 0.6998841251448435, 'f1': 0.6236448115642746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.7142857142857143, 'r': 0.47619047619047616, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.574869109947644, 'r': 0.7310252996005326, 'f1': 0.6436107854630715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Russian: {'event': {'p': 0.5623836126629422, 'r': 0.6998841251448435, 'f1': 0.6236448115642746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Russian: {'event': {'p': 0.525, 'r': 0.5833333333333334, 'f1': 0.5526315789473685}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} ****************************** Epoch: 3 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:58:16.567409: step: 4/529, loss: 0.1127520501613617 2023-01-21 08:58:17.698610: step: 8/529, loss: 0.432242214679718 2023-01-21 08:58:18.816899: step: 12/529, loss: 0.15021859109401703 2023-01-21 08:58:19.937872: step: 16/529, loss: 0.1918453723192215 2023-01-21 08:58:21.061462: step: 20/529, loss: 0.10688343644142151 2023-01-21 08:58:22.174673: step: 24/529, loss: 0.1043185293674469 2023-01-21 08:58:23.295549: step: 28/529, loss: 0.26591378450393677 2023-01-21 08:58:24.429758: step: 32/529, loss: 0.7260921597480774 2023-01-21 08:58:25.576869: step: 36/529, loss: 0.32221031188964844 2023-01-21 08:58:26.708434: step: 40/529, loss: 0.18153077363967896 2023-01-21 08:58:27.813753: step: 44/529, loss: 0.0676766186952591 2023-01-21 08:58:28.933628: step: 48/529, loss: 0.4816628694534302 2023-01-21 08:58:30.050004: step: 52/529, loss: 0.18099379539489746 2023-01-21 08:58:31.186824: step: 56/529, loss: 1.011576771736145 2023-01-21 08:58:32.318603: step: 60/529, loss: 0.383029580116272 2023-01-21 08:58:33.436825: step: 64/529, loss: 0.9920968413352966 2023-01-21 08:58:34.601560: step: 68/529, loss: 0.05940570682287216 2023-01-21 08:58:35.732870: step: 72/529, loss: 0.17299515008926392 2023-01-21 08:58:36.867869: step: 76/529, loss: 0.9448099136352539 2023-01-21 08:58:38.007633: step: 80/529, loss: 0.16690778732299805 2023-01-21 08:58:39.154458: step: 84/529, loss: 0.0600038543343544 2023-01-21 08:58:40.321445: step: 88/529, loss: 0.22403989732265472 2023-01-21 08:58:41.458725: step: 92/529, loss: 0.3614594042301178 2023-01-21 08:58:42.602581: step: 96/529, loss: 0.04655952751636505 2023-01-21 08:58:43.731326: step: 100/529, loss: 6.479700565338135 2023-01-21 08:58:44.861766: step: 104/529, loss: 0.23768778145313263 2023-01-21 08:58:46.009258: step: 108/529, loss: 0.41127586364746094 2023-01-21 08:58:47.135391: step: 112/529, loss: 0.08945274353027344 2023-01-21 08:58:48.273595: step: 116/529, loss: 0.0813365951180458 2023-01-21 08:58:49.422413: step: 120/529, loss: 0.11603689193725586 2023-01-21 08:58:50.555519: step: 124/529, loss: 1.1029725074768066 2023-01-21 08:58:51.689273: step: 128/529, loss: 6.2161865234375 2023-01-21 08:58:52.828751: step: 132/529, loss: 1.9589625597000122 2023-01-21 08:58:53.998278: step: 136/529, loss: 0.19591423869132996 2023-01-21 08:58:55.139884: step: 140/529, loss: 7.002114295959473 2023-01-21 08:58:56.275026: step: 144/529, loss: 1.5503019094467163 2023-01-21 08:58:57.375476: step: 148/529, loss: 6.921739101409912 2023-01-21 08:58:58.481778: step: 152/529, loss: 0.1581367552280426 2023-01-21 08:58:59.603584: step: 156/529, loss: 0.3111073076725006 2023-01-21 08:59:00.729037: step: 160/529, loss: 0.5508758425712585 2023-01-21 08:59:01.857806: step: 164/529, loss: 0.2559228837490082 2023-01-21 08:59:02.984529: step: 168/529, loss: 0.658478319644928 2023-01-21 08:59:04.124244: step: 172/529, loss: 0.445242702960968 2023-01-21 08:59:05.250650: step: 176/529, loss: 0.08531536906957626 2023-01-21 08:59:06.360156: step: 180/529, loss: 0.185246080160141 2023-01-21 08:59:07.506514: step: 184/529, loss: 0.18567711114883423 2023-01-21 08:59:08.688671: step: 188/529, loss: 0.32176923751831055 2023-01-21 08:59:09.798069: step: 192/529, loss: 0.05649113655090332 2023-01-21 08:59:10.953475: step: 196/529, loss: 0.0940040647983551 2023-01-21 08:59:12.110257: step: 200/529, loss: 0.19925037026405334 2023-01-21 08:59:13.237294: step: 204/529, loss: 6.022089958190918 2023-01-21 08:59:14.382552: step: 208/529, loss: 0.7966360449790955 2023-01-21 08:59:15.501050: step: 212/529, loss: 0.16041961312294006 2023-01-21 08:59:16.626487: step: 216/529, loss: 0.7453233003616333 2023-01-21 08:59:17.758104: step: 220/529, loss: 0.046007610857486725 2023-01-21 08:59:18.892945: step: 224/529, loss: 0.07042908668518066 2023-01-21 08:59:20.015744: step: 228/529, loss: 0.25191402435302734 2023-01-21 08:59:21.131994: step: 232/529, loss: 0.03773627430200577 2023-01-21 08:59:22.255644: step: 236/529, loss: 0.25366511940956116 2023-01-21 08:59:23.403600: step: 240/529, loss: 0.22157259285449982 2023-01-21 08:59:24.541489: step: 244/529, loss: 0.3382606506347656 2023-01-21 08:59:25.648205: step: 248/529, loss: 0.353292852640152 2023-01-21 08:59:26.759668: step: 252/529, loss: 0.25949347019195557 2023-01-21 08:59:27.922081: step: 256/529, loss: 0.4028354287147522 2023-01-21 08:59:29.060493: step: 260/529, loss: 0.2683922052383423 2023-01-21 08:59:30.195696: step: 264/529, loss: 0.1497069150209427 2023-01-21 08:59:31.342573: step: 268/529, loss: 0.11594495922327042 2023-01-21 08:59:32.487417: step: 272/529, loss: 0.1656152307987213 2023-01-21 08:59:33.603585: step: 276/529, loss: 0.22007064521312714 2023-01-21 08:59:34.750690: step: 280/529, loss: 0.5109320282936096 2023-01-21 08:59:35.884699: step: 284/529, loss: 0.305112361907959 2023-01-21 08:59:37.010597: step: 288/529, loss: 0.9760399460792542 2023-01-21 08:59:38.145826: step: 292/529, loss: 0.29509663581848145 2023-01-21 08:59:39.350668: step: 296/529, loss: 0.06189975515007973 2023-01-21 08:59:40.479105: step: 300/529, loss: 2.9684865474700928 2023-01-21 08:59:41.606637: step: 304/529, loss: 0.4282251298427582 2023-01-21 08:59:42.750952: step: 308/529, loss: 0.1137821227312088 2023-01-21 08:59:43.872954: step: 312/529, loss: 0.24588042497634888 2023-01-21 08:59:45.022585: step: 316/529, loss: 0.1399907022714615 2023-01-21 08:59:46.150121: step: 320/529, loss: 0.41705432534217834 2023-01-21 08:59:47.314159: step: 324/529, loss: 0.28086891770362854 2023-01-21 08:59:48.462070: step: 328/529, loss: 0.6341606974601746 2023-01-21 08:59:49.584142: step: 332/529, loss: 2.945425033569336 2023-01-21 08:59:50.726929: step: 336/529, loss: 1.951957106590271 2023-01-21 08:59:51.847188: step: 340/529, loss: 0.5843101739883423 2023-01-21 08:59:52.989290: step: 344/529, loss: 0.0877981185913086 2023-01-21 08:59:54.120386: step: 348/529, loss: 0.23725710809230804 2023-01-21 08:59:55.252332: step: 352/529, loss: 0.19671788811683655 2023-01-21 08:59:56.432632: step: 356/529, loss: 0.259087473154068 2023-01-21 08:59:57.592460: step: 360/529, loss: 0.09261293709278107 2023-01-21 08:59:58.727720: step: 364/529, loss: 0.17963162064552307 2023-01-21 08:59:59.854396: step: 368/529, loss: 0.06281280517578125 2023-01-21 09:00:00.983792: step: 372/529, loss: 0.026651956140995026 2023-01-21 09:00:02.131265: step: 376/529, loss: 0.5319308042526245 2023-01-21 09:00:03.282781: step: 380/529, loss: 0.5862182378768921 2023-01-21 09:00:04.433583: step: 384/529, loss: 6.4143524169921875 2023-01-21 09:00:05.574718: step: 388/529, loss: 0.1870521605014801 2023-01-21 09:00:06.689016: step: 392/529, loss: 0.07262754440307617 2023-01-21 09:00:07.842772: step: 396/529, loss: 0.7711917161941528 2023-01-21 09:00:09.000951: step: 400/529, loss: 0.3445551097393036 2023-01-21 09:00:10.159852: step: 404/529, loss: 0.05444631725549698 2023-01-21 09:00:11.304939: step: 408/529, loss: 0.31717413663864136 2023-01-21 09:00:12.479906: step: 412/529, loss: 0.46483713388442993 2023-01-21 09:00:13.600793: step: 416/529, loss: 0.06532096862792969 2023-01-21 09:00:14.720333: step: 420/529, loss: 0.21404992043972015 2023-01-21 09:00:15.857923: step: 424/529, loss: 0.29162129759788513 2023-01-21 09:00:17.019781: step: 428/529, loss: 0.14978857338428497 2023-01-21 09:00:18.144993: step: 432/529, loss: 0.4147473871707916 2023-01-21 09:00:19.270571: step: 436/529, loss: 0.34247875213623047 2023-01-21 09:00:20.402216: step: 440/529, loss: 0.28635191917419434 2023-01-21 09:00:21.553774: step: 444/529, loss: 0.274493545293808 2023-01-21 09:00:22.704013: step: 448/529, loss: 0.1970309317111969 2023-01-21 09:00:23.874929: step: 452/529, loss: 0.20057940483093262 2023-01-21 09:00:25.050997: step: 456/529, loss: 0.10596227645874023 2023-01-21 09:00:26.223756: step: 460/529, loss: 1.2692768573760986 2023-01-21 09:00:27.363261: step: 464/529, loss: 0.32083916664123535 2023-01-21 09:00:28.516768: step: 468/529, loss: 0.16598376631736755 2023-01-21 09:00:29.678984: step: 472/529, loss: 0.1347036361694336 2023-01-21 09:00:30.821655: step: 476/529, loss: 0.4680764079093933 2023-01-21 09:00:31.982799: step: 480/529, loss: 0.19098129868507385 2023-01-21 09:00:33.126927: step: 484/529, loss: 0.15129438042640686 2023-01-21 09:00:34.243022: step: 488/529, loss: 0.1173185333609581 2023-01-21 09:00:35.398682: step: 492/529, loss: 0.15769940614700317 2023-01-21 09:00:36.528363: step: 496/529, loss: 0.13384681940078735 2023-01-21 09:00:37.684571: step: 500/529, loss: 0.08125276863574982 2023-01-21 09:00:38.826585: step: 504/529, loss: 0.15946227312088013 2023-01-21 09:00:39.959954: step: 508/529, loss: 0.09619312733411789 2023-01-21 09:00:41.094734: step: 512/529, loss: 0.5139181613922119 2023-01-21 09:00:42.233801: step: 516/529, loss: 0.18244200944900513 2023-01-21 09:00:43.400250: step: 520/529, loss: 0.6223430633544922 2023-01-21 09:00:44.556425: step: 524/529, loss: 0.328574001789093 2023-01-21 09:00:45.694062: step: 528/529, loss: 0.08651085197925568 2023-01-21 09:00:46.868153: step: 532/529, loss: 0.45230740308761597 2023-01-21 09:00:48.020914: step: 536/529, loss: 0.9030293226242065 2023-01-21 09:00:49.197368: step: 540/529, loss: 0.9087143540382385 2023-01-21 09:00:50.320681: step: 544/529, loss: 0.3730502128601074 2023-01-21 09:00:51.489704: step: 548/529, loss: 0.09509558975696564 2023-01-21 09:00:52.618909: step: 552/529, loss: 0.7257412672042847 2023-01-21 09:00:53.754796: step: 556/529, loss: 0.5058338046073914 2023-01-21 09:00:54.899065: step: 560/529, loss: 0.15334831178188324 2023-01-21 09:00:56.014329: step: 564/529, loss: 0.16440238058567047 2023-01-21 09:00:57.132788: step: 568/529, loss: 1.058837890625 2023-01-21 09:00:58.293366: step: 572/529, loss: 0.31908196210861206 2023-01-21 09:00:59.437186: step: 576/529, loss: 0.13805875182151794 2023-01-21 09:01:00.616281: step: 580/529, loss: 0.9343050122261047 2023-01-21 09:01:01.752804: step: 584/529, loss: 0.7799698710441589 2023-01-21 09:01:02.881489: step: 588/529, loss: 0.08400693535804749 2023-01-21 09:01:04.007529: step: 592/529, loss: 0.19094926118850708 2023-01-21 09:01:05.169678: step: 596/529, loss: 0.2239713817834854 2023-01-21 09:01:06.286073: step: 600/529, loss: 0.19903230667114258 2023-01-21 09:01:07.428215: step: 604/529, loss: 0.07337456196546555 2023-01-21 09:01:08.566201: step: 608/529, loss: 1.0952904224395752 2023-01-21 09:01:09.726348: step: 612/529, loss: 0.061934471130371094 2023-01-21 09:01:10.881990: step: 616/529, loss: 0.02922229841351509 2023-01-21 09:01:12.091302: step: 620/529, loss: 0.07398152351379395 2023-01-21 09:01:13.258484: step: 624/529, loss: 0.6863476037979126 2023-01-21 09:01:14.395164: step: 628/529, loss: 0.10770788043737411 2023-01-21 09:01:15.556617: step: 632/529, loss: 0.14766426384449005 2023-01-21 09:01:16.709297: step: 636/529, loss: 0.5057401657104492 2023-01-21 09:01:17.878738: step: 640/529, loss: 0.17820319533348083 2023-01-21 09:01:18.997768: step: 644/529, loss: 0.08456039428710938 2023-01-21 09:01:20.123787: step: 648/529, loss: 0.8642090559005737 2023-01-21 09:01:21.284556: step: 652/529, loss: 0.13768067955970764 2023-01-21 09:01:22.445868: step: 656/529, loss: 0.3024471402168274 2023-01-21 09:01:23.551146: step: 660/529, loss: 0.14301714301109314 2023-01-21 09:01:24.699015: step: 664/529, loss: 0.9911198019981384 2023-01-21 09:01:25.862303: step: 668/529, loss: 6.947729110717773 2023-01-21 09:01:27.002573: step: 672/529, loss: 0.6631352305412292 2023-01-21 09:01:28.150260: step: 676/529, loss: 0.7771148681640625 2023-01-21 09:01:29.277872: step: 680/529, loss: 1.2801828384399414 2023-01-21 09:01:30.443425: step: 684/529, loss: 0.8821722269058228 2023-01-21 09:01:31.578621: step: 688/529, loss: 0.6851043701171875 2023-01-21 09:01:32.741526: step: 692/529, loss: 0.19688820838928223 2023-01-21 09:01:33.907444: step: 696/529, loss: 0.11057816445827484 2023-01-21 09:01:35.052964: step: 700/529, loss: 0.6198978424072266 2023-01-21 09:01:36.192585: step: 704/529, loss: 6.326823711395264 2023-01-21 09:01:37.334363: step: 708/529, loss: 0.37878283858299255 2023-01-21 09:01:38.454635: step: 712/529, loss: 0.16252851486206055 2023-01-21 09:01:39.562431: step: 716/529, loss: 0.959736704826355 2023-01-21 09:01:40.697741: step: 720/529, loss: 0.6077576279640198 2023-01-21 09:01:41.820397: step: 724/529, loss: 0.5647670030593872 2023-01-21 09:01:42.988174: step: 728/529, loss: 0.24695168435573578 2023-01-21 09:01:44.141453: step: 732/529, loss: 0.7023944854736328 2023-01-21 09:01:45.302322: step: 736/529, loss: 0.15500479936599731 2023-01-21 09:01:46.450166: step: 740/529, loss: 0.47891730070114136 2023-01-21 09:01:47.565179: step: 744/529, loss: 0.6409094929695129 2023-01-21 09:01:48.711003: step: 748/529, loss: 0.7811018228530884 2023-01-21 09:01:49.849626: step: 752/529, loss: 0.8542653322219849 2023-01-21 09:01:50.959538: step: 756/529, loss: 0.4607399106025696 2023-01-21 09:01:52.097904: step: 760/529, loss: 0.7185568809509277 2023-01-21 09:01:53.245198: step: 764/529, loss: 0.4659437835216522 2023-01-21 09:01:54.388559: step: 768/529, loss: 0.09852056205272675 2023-01-21 09:01:55.497497: step: 772/529, loss: 0.4453864097595215 2023-01-21 09:01:56.610735: step: 776/529, loss: 1.1008559465408325 2023-01-21 09:01:57.770315: step: 780/529, loss: 0.13033896684646606 2023-01-21 09:01:58.904298: step: 784/529, loss: 0.9933927655220032 2023-01-21 09:02:00.017920: step: 788/529, loss: 0.47975999116897583 2023-01-21 09:02:01.174638: step: 792/529, loss: 0.08241887390613556 2023-01-21 09:02:02.292964: step: 796/529, loss: 0.18805037438869476 2023-01-21 09:02:03.433546: step: 800/529, loss: 0.1067572683095932 2023-01-21 09:02:04.557345: step: 804/529, loss: 0.6861340999603271 2023-01-21 09:02:05.702712: step: 808/529, loss: 0.7667003870010376 2023-01-21 09:02:06.846192: step: 812/529, loss: 0.07227583229541779 2023-01-21 09:02:07.986860: step: 816/529, loss: 1.3184375762939453 2023-01-21 09:02:09.151188: step: 820/529, loss: 0.06064195558428764 2023-01-21 09:02:10.285621: step: 824/529, loss: 0.11673454940319061 2023-01-21 09:02:11.404371: step: 828/529, loss: 0.08390478789806366 2023-01-21 09:02:12.518618: step: 832/529, loss: 0.19092312455177307 2023-01-21 09:02:13.669334: step: 836/529, loss: 0.7027814984321594 2023-01-21 09:02:14.807123: step: 840/529, loss: 0.1882866770029068 2023-01-21 09:02:15.916621: step: 844/529, loss: 0.12573833763599396 2023-01-21 09:02:17.060237: step: 848/529, loss: 0.40620365738868713 2023-01-21 09:02:18.218809: step: 852/529, loss: 0.07022324204444885 2023-01-21 09:02:19.363806: step: 856/529, loss: 0.2873774766921997 2023-01-21 09:02:20.509122: step: 860/529, loss: 0.6653907299041748 2023-01-21 09:02:21.635372: step: 864/529, loss: 0.16346341371536255 2023-01-21 09:02:22.766263: step: 868/529, loss: 0.054229069501161575 2023-01-21 09:02:23.897075: step: 872/529, loss: 0.1588718444108963 2023-01-21 09:02:25.025983: step: 876/529, loss: 0.09767427295446396 2023-01-21 09:02:26.174204: step: 880/529, loss: 0.9408793449401855 2023-01-21 09:02:27.312270: step: 884/529, loss: 0.1131628081202507 2023-01-21 09:02:28.479141: step: 888/529, loss: 0.5475469827651978 2023-01-21 09:02:29.601285: step: 892/529, loss: 0.8748939633369446 2023-01-21 09:02:30.737114: step: 896/529, loss: 0.14808779954910278 2023-01-21 09:02:31.872665: step: 900/529, loss: 0.8827480673789978 2023-01-21 09:02:32.999534: step: 904/529, loss: 0.6642062664031982 2023-01-21 09:02:34.135591: step: 908/529, loss: 0.1711595058441162 2023-01-21 09:02:35.266558: step: 912/529, loss: 0.2882459759712219 2023-01-21 09:02:36.418642: step: 916/529, loss: 0.16212357580661774 2023-01-21 09:02:37.546559: step: 920/529, loss: 0.08147764205932617 2023-01-21 09:02:38.667378: step: 924/529, loss: 0.15510712563991547 2023-01-21 09:02:39.832425: step: 928/529, loss: 0.5895828604698181 2023-01-21 09:02:40.939568: step: 932/529, loss: 0.28344759345054626 2023-01-21 09:02:42.073580: step: 936/529, loss: 0.1728839874267578 2023-01-21 09:02:43.209541: step: 940/529, loss: 0.7093252539634705 2023-01-21 09:02:44.381841: step: 944/529, loss: 0.49861180782318115 2023-01-21 09:02:45.493437: step: 948/529, loss: 0.4498213827610016 2023-01-21 09:02:46.650569: step: 952/529, loss: 0.13618993759155273 2023-01-21 09:02:47.807169: step: 956/529, loss: 0.07919006794691086 2023-01-21 09:02:48.995483: step: 960/529, loss: 0.4718391001224518 2023-01-21 09:02:50.149446: step: 964/529, loss: 0.24556908011436462 2023-01-21 09:02:51.295928: step: 968/529, loss: 0.16113919019699097 2023-01-21 09:02:52.460740: step: 972/529, loss: 0.3961173892021179 2023-01-21 09:02:53.596379: step: 976/529, loss: 0.10006770491600037 2023-01-21 09:02:54.735797: step: 980/529, loss: 0.8665919303894043 2023-01-21 09:02:55.858928: step: 984/529, loss: 0.16403932869434357 2023-01-21 09:02:57.021353: step: 988/529, loss: 0.20638760924339294 2023-01-21 09:02:58.180294: step: 992/529, loss: 0.45358705520629883 2023-01-21 09:02:59.365048: step: 996/529, loss: 0.10458078980445862 2023-01-21 09:03:00.518137: step: 1000/529, loss: 0.7031070590019226 2023-01-21 09:03:01.647736: step: 1004/529, loss: 0.28158989548683167 2023-01-21 09:03:02.780408: step: 1008/529, loss: 0.20380783081054688 2023-01-21 09:03:03.911062: step: 1012/529, loss: 6.5429277420043945 2023-01-21 09:03:05.054009: step: 1016/529, loss: 0.36283808946609497 2023-01-21 09:03:06.213491: step: 1020/529, loss: 0.2250961810350418 2023-01-21 09:03:07.355823: step: 1024/529, loss: 0.4927643835544586 2023-01-21 09:03:08.504714: step: 1028/529, loss: 0.08336491882801056 2023-01-21 09:03:09.627125: step: 1032/529, loss: 0.13230600953102112 2023-01-21 09:03:10.749811: step: 1036/529, loss: 0.7081431150436401 2023-01-21 09:03:11.915064: step: 1040/529, loss: 0.004212379455566406 2023-01-21 09:03:13.051692: step: 1044/529, loss: 0.1725786179304123 2023-01-21 09:03:14.193986: step: 1048/529, loss: 0.5279135704040527 2023-01-21 09:03:15.326701: step: 1052/529, loss: 0.6752395629882812 2023-01-21 09:03:16.442898: step: 1056/529, loss: 0.20082946121692657 2023-01-21 09:03:17.604091: step: 1060/529, loss: 6.2498459815979 2023-01-21 09:03:18.731312: step: 1064/529, loss: 0.07474994659423828 2023-01-21 09:03:19.840953: step: 1068/529, loss: 0.1972431242465973 2023-01-21 09:03:20.982095: step: 1072/529, loss: 0.04951906204223633 2023-01-21 09:03:22.137933: step: 1076/529, loss: 0.37428781390190125 2023-01-21 09:03:23.271790: step: 1080/529, loss: 0.13798211514949799 2023-01-21 09:03:24.391987: step: 1084/529, loss: 1.4594587087631226 2023-01-21 09:03:25.539505: step: 1088/529, loss: 0.36609047651290894 2023-01-21 09:03:26.653458: step: 1092/529, loss: 0.06333620846271515 2023-01-21 09:03:27.791873: step: 1096/529, loss: 0.3570215702056885 2023-01-21 09:03:28.949241: step: 1100/529, loss: 0.4960615038871765 2023-01-21 09:03:30.118481: step: 1104/529, loss: 0.040868185460567474 2023-01-21 09:03:31.255371: step: 1108/529, loss: 0.1393783688545227 2023-01-21 09:03:32.418748: step: 1112/529, loss: 0.5948085784912109 2023-01-21 09:03:33.584787: step: 1116/529, loss: 0.11437148600816727 2023-01-21 09:03:34.719587: step: 1120/529, loss: 0.7754427194595337 2023-01-21 09:03:35.852299: step: 1124/529, loss: 0.3609389066696167 2023-01-21 09:03:36.983312: step: 1128/529, loss: 0.10810461640357971 2023-01-21 09:03:38.129735: step: 1132/529, loss: 0.8239008784294128 2023-01-21 09:03:39.265731: step: 1136/529, loss: 0.24920830130577087 2023-01-21 09:03:40.423317: step: 1140/529, loss: 0.13289594650268555 2023-01-21 09:03:41.538017: step: 1144/529, loss: 0.16481778025627136 2023-01-21 09:03:42.684417: step: 1148/529, loss: 0.09337092190980911 2023-01-21 09:03:43.817367: step: 1152/529, loss: 7.192674160003662 2023-01-21 09:03:44.969418: step: 1156/529, loss: 0.36509427428245544 2023-01-21 09:03:46.103356: step: 1160/529, loss: 1.1140697002410889 2023-01-21 09:03:47.310997: step: 1164/529, loss: 0.04064493253827095 2023-01-21 09:03:48.469984: step: 1168/529, loss: 0.2842632234096527 2023-01-21 09:03:49.595153: step: 1172/529, loss: 0.20427417755126953 2023-01-21 09:03:50.733711: step: 1176/529, loss: 0.7634838223457336 2023-01-21 09:03:51.843880: step: 1180/529, loss: 0.28395357728004456 2023-01-21 09:03:53.000083: step: 1184/529, loss: 0.21039456129074097 2023-01-21 09:03:54.137713: step: 1188/529, loss: 0.4592967927455902 2023-01-21 09:03:55.253247: step: 1192/529, loss: 0.10056066513061523 2023-01-21 09:03:56.424964: step: 1196/529, loss: 0.4734603762626648 2023-01-21 09:03:57.588995: step: 1200/529, loss: 0.1483844816684723 2023-01-21 09:03:58.733451: step: 1204/529, loss: 0.8957277536392212 2023-01-21 09:03:59.850280: step: 1208/529, loss: 0.8649401664733887 2023-01-21 09:04:00.991653: step: 1212/529, loss: 1.3796403408050537 2023-01-21 09:04:02.110101: step: 1216/529, loss: 0.5873506665229797 2023-01-21 09:04:03.257894: step: 1220/529, loss: 0.37370890378952026 2023-01-21 09:04:04.392614: step: 1224/529, loss: 0.27227482199668884 2023-01-21 09:04:05.512418: step: 1228/529, loss: 0.15552778542041779 2023-01-21 09:04:06.627797: step: 1232/529, loss: 0.12194252014160156 2023-01-21 09:04:07.774623: step: 1236/529, loss: 1.2321577072143555 2023-01-21 09:04:08.907045: step: 1240/529, loss: 0.062488533556461334 2023-01-21 09:04:10.024733: step: 1244/529, loss: 0.1309993863105774 2023-01-21 09:04:11.151704: step: 1248/529, loss: 0.3407500982284546 2023-01-21 09:04:12.280335: step: 1252/529, loss: 1.0729297399520874 2023-01-21 09:04:13.414722: step: 1256/529, loss: 0.6232610940933228 2023-01-21 09:04:14.566181: step: 1260/529, loss: 1.2132762670516968 2023-01-21 09:04:15.715448: step: 1264/529, loss: 0.16887111961841583 2023-01-21 09:04:16.845929: step: 1268/529, loss: 0.13054580986499786 2023-01-21 09:04:17.996273: step: 1272/529, loss: 0.1171242743730545 2023-01-21 09:04:19.151467: step: 1276/529, loss: 0.07810792326927185 2023-01-21 09:04:20.313519: step: 1280/529, loss: 1.12432062625885 2023-01-21 09:04:21.439751: step: 1284/529, loss: 0.7599477767944336 2023-01-21 09:04:22.598367: step: 1288/529, loss: 0.10164737701416016 2023-01-21 09:04:23.731679: step: 1292/529, loss: 1.122117280960083 2023-01-21 09:04:24.903404: step: 1296/529, loss: 0.23339280486106873 2023-01-21 09:04:26.057769: step: 1300/529, loss: 0.825629472732544 2023-01-21 09:04:27.236917: step: 1304/529, loss: 0.26286134123802185 2023-01-21 09:04:28.382518: step: 1308/529, loss: 0.1601598709821701 2023-01-21 09:04:29.516140: step: 1312/529, loss: 0.10929213464260101 2023-01-21 09:04:30.650163: step: 1316/529, loss: 0.6846439242362976 2023-01-21 09:04:31.767894: step: 1320/529, loss: 0.20264720916748047 2023-01-21 09:04:32.915844: step: 1324/529, loss: 0.043024733662605286 2023-01-21 09:04:34.033568: step: 1328/529, loss: 0.08657798916101456 2023-01-21 09:04:35.166495: step: 1332/529, loss: 0.11942777782678604 2023-01-21 09:04:36.307020: step: 1336/529, loss: 0.07697949558496475 2023-01-21 09:04:37.444104: step: 1340/529, loss: 0.07051143795251846 2023-01-21 09:04:38.563435: step: 1344/529, loss: 0.03893470764160156 2023-01-21 09:04:39.721938: step: 1348/529, loss: 0.17344561219215393 2023-01-21 09:04:40.878276: step: 1352/529, loss: 0.19600439071655273 2023-01-21 09:04:41.995472: step: 1356/529, loss: 0.1965889036655426 2023-01-21 09:04:43.149623: step: 1360/529, loss: 0.23195305466651917 2023-01-21 09:04:44.310189: step: 1364/529, loss: 0.5283797383308411 2023-01-21 09:04:45.480599: step: 1368/529, loss: 0.12434768676757812 2023-01-21 09:04:46.614327: step: 1372/529, loss: 0.14588089287281036 2023-01-21 09:04:47.765917: step: 1376/529, loss: 0.12842664122581482 2023-01-21 09:04:48.908511: step: 1380/529, loss: 1.1464896202087402 2023-01-21 09:04:50.045442: step: 1384/529, loss: 0.6363987922668457 2023-01-21 09:04:51.181222: step: 1388/529, loss: 0.3084859848022461 2023-01-21 09:04:52.331424: step: 1392/529, loss: 0.06737995147705078 2023-01-21 09:04:53.487399: step: 1396/529, loss: 1.9096086025238037 2023-01-21 09:04:54.625545: step: 1400/529, loss: 0.14616060256958008 2023-01-21 09:04:55.810303: step: 1404/529, loss: 0.9278597831726074 2023-01-21 09:04:56.943295: step: 1408/529, loss: 0.32750579714775085 2023-01-21 09:04:58.090958: step: 1412/529, loss: 3.3215439319610596 2023-01-21 09:04:59.210207: step: 1416/529, loss: 0.11584663391113281 2023-01-21 09:05:00.338372: step: 1420/529, loss: 0.4824641942977905 2023-01-21 09:05:01.484522: step: 1424/529, loss: 1.3737637996673584 2023-01-21 09:05:02.653635: step: 1428/529, loss: 0.2705325186252594 2023-01-21 09:05:03.785801: step: 1432/529, loss: 0.9200029969215393 2023-01-21 09:05:04.930703: step: 1436/529, loss: 0.05270242691040039 2023-01-21 09:05:06.130256: step: 1440/529, loss: 0.6112867593765259 2023-01-21 09:05:07.254929: step: 1444/529, loss: 0.5234668850898743 2023-01-21 09:05:08.411358: step: 1448/529, loss: 0.8266826868057251 2023-01-21 09:05:09.573564: step: 1452/529, loss: 1.4764163494110107 2023-01-21 09:05:10.712782: step: 1456/529, loss: 0.25212153792381287 2023-01-21 09:05:11.867485: step: 1460/529, loss: 0.1477159559726715 2023-01-21 09:05:12.972833: step: 1464/529, loss: 0.10954266041517258 2023-01-21 09:05:14.149782: step: 1468/529, loss: 0.7533424496650696 2023-01-21 09:05:15.293056: step: 1472/529, loss: 0.6905034184455872 2023-01-21 09:05:16.445175: step: 1476/529, loss: 0.8208313584327698 2023-01-21 09:05:17.601903: step: 1480/529, loss: 0.6493015885353088 2023-01-21 09:05:18.735801: step: 1484/529, loss: 0.3004302978515625 2023-01-21 09:05:19.837040: step: 1488/529, loss: 0.19165191054344177 2023-01-21 09:05:20.992970: step: 1492/529, loss: 0.5054263472557068 2023-01-21 09:05:22.120443: step: 1496/529, loss: 0.1792614907026291 2023-01-21 09:05:23.260622: step: 1500/529, loss: 0.5962066054344177 2023-01-21 09:05:24.414958: step: 1504/529, loss: 1.348507285118103 2023-01-21 09:05:25.561317: step: 1508/529, loss: 0.07748661190271378 2023-01-21 09:05:26.686448: step: 1512/529, loss: 0.5533306002616882 2023-01-21 09:05:27.825167: step: 1516/529, loss: 1.0416698455810547 2023-01-21 09:05:28.958483: step: 1520/529, loss: 0.14721174538135529 2023-01-21 09:05:30.103784: step: 1524/529, loss: 1.5281424522399902 2023-01-21 09:05:31.234993: step: 1528/529, loss: 0.7640345096588135 2023-01-21 09:05:32.392874: step: 1532/529, loss: 0.05257921293377876 2023-01-21 09:05:33.526519: step: 1536/529, loss: 0.1727236807346344 2023-01-21 09:05:34.647798: step: 1540/529, loss: 0.11017632484436035 2023-01-21 09:05:35.793948: step: 1544/529, loss: 1.6037763357162476 2023-01-21 09:05:36.953060: step: 1548/529, loss: 0.3284495770931244 2023-01-21 09:05:38.108803: step: 1552/529, loss: 0.8090665340423584 2023-01-21 09:05:39.262106: step: 1556/529, loss: 0.042127083986997604 2023-01-21 09:05:40.406621: step: 1560/529, loss: 0.14104318618774414 2023-01-21 09:05:41.555091: step: 1564/529, loss: 0.11853523552417755 2023-01-21 09:05:42.735169: step: 1568/529, loss: 0.13271060585975647 2023-01-21 09:05:43.874343: step: 1572/529, loss: 0.20564812421798706 2023-01-21 09:05:45.024899: step: 1576/529, loss: 0.083777517080307 2023-01-21 09:05:46.146848: step: 1580/529, loss: 0.11326833069324493 2023-01-21 09:05:47.306822: step: 1584/529, loss: 0.23062877357006073 2023-01-21 09:05:48.435741: step: 1588/529, loss: 0.13704939186573029 2023-01-21 09:05:49.551408: step: 1592/529, loss: 0.09096350520849228 2023-01-21 09:05:50.727184: step: 1596/529, loss: 0.07349219918251038 2023-01-21 09:05:51.859672: step: 1600/529, loss: 0.09050159901380539 2023-01-21 09:05:53.017967: step: 1604/529, loss: 0.12973442673683167 2023-01-21 09:05:54.163511: step: 1608/529, loss: 0.22155389189720154 2023-01-21 09:05:55.304023: step: 1612/529, loss: 0.10656996071338654 2023-01-21 09:05:56.471570: step: 1616/529, loss: 0.1378362774848938 2023-01-21 09:05:57.595048: step: 1620/529, loss: 0.10250511765480042 2023-01-21 09:05:58.710754: step: 1624/529, loss: 0.8444705605506897 2023-01-21 09:05:59.842937: step: 1628/529, loss: 0.10342703014612198 2023-01-21 09:06:00.978199: step: 1632/529, loss: 0.4166475236415863 2023-01-21 09:06:02.128877: step: 1636/529, loss: 0.15266117453575134 2023-01-21 09:06:03.227504: step: 1640/529, loss: 0.3019540309906006 2023-01-21 09:06:04.377613: step: 1644/529, loss: 0.1784309446811676 2023-01-21 09:06:05.555934: step: 1648/529, loss: 0.43192481994628906 2023-01-21 09:06:06.664205: step: 1652/529, loss: 0.2244025319814682 2023-01-21 09:06:07.804713: step: 1656/529, loss: 0.06013345718383789 2023-01-21 09:06:09.004421: step: 1660/529, loss: 0.0564112663269043 2023-01-21 09:06:10.152276: step: 1664/529, loss: 0.2510750889778137 2023-01-21 09:06:11.327872: step: 1668/529, loss: 0.11374817043542862 2023-01-21 09:06:12.476293: step: 1672/529, loss: 0.11951828002929688 2023-01-21 09:06:13.628527: step: 1676/529, loss: 0.18748603761196136 2023-01-21 09:06:14.799174: step: 1680/529, loss: 0.43095022439956665 2023-01-21 09:06:15.939389: step: 1684/529, loss: 0.7503036260604858 2023-01-21 09:06:17.079108: step: 1688/529, loss: 0.8607301115989685 2023-01-21 09:06:18.199537: step: 1692/529, loss: 0.025034546852111816 2023-01-21 09:06:19.365911: step: 1696/529, loss: 0.11721468716859818 2023-01-21 09:06:20.506617: step: 1700/529, loss: 0.4294639229774475 2023-01-21 09:06:21.655424: step: 1704/529, loss: 0.668768584728241 2023-01-21 09:06:22.819844: step: 1708/529, loss: 0.11543674767017365 2023-01-21 09:06:23.940327: step: 1712/529, loss: 0.23011913895606995 2023-01-21 09:06:25.081168: step: 1716/529, loss: 1.3662277460098267 2023-01-21 09:06:26.229266: step: 1720/529, loss: 0.04422121122479439 2023-01-21 09:06:27.388997: step: 1724/529, loss: 0.14086276292800903 2023-01-21 09:06:28.529367: step: 1728/529, loss: 0.11898299306631088 2023-01-21 09:06:29.670168: step: 1732/529, loss: 0.07863965630531311 2023-01-21 09:06:30.790293: step: 1736/529, loss: 0.17624256014823914 2023-01-21 09:06:31.965024: step: 1740/529, loss: 0.436884343624115 2023-01-21 09:06:33.120672: step: 1744/529, loss: 0.4204428195953369 2023-01-21 09:06:34.254707: step: 1748/529, loss: 0.11820783466100693 2023-01-21 09:06:35.390962: step: 1752/529, loss: 0.14672546088695526 2023-01-21 09:06:36.553689: step: 1756/529, loss: 0.6049914360046387 2023-01-21 09:06:37.688604: step: 1760/529, loss: 0.13614541292190552 2023-01-21 09:06:38.821421: step: 1764/529, loss: 0.2935635447502136 2023-01-21 09:06:39.974927: step: 1768/529, loss: 0.5210188031196594 2023-01-21 09:06:41.102587: step: 1772/529, loss: 0.1947949230670929 2023-01-21 09:06:42.293255: step: 1776/529, loss: 1.035811185836792 2023-01-21 09:06:43.449744: step: 1780/529, loss: 0.8085301518440247 2023-01-21 09:06:44.580565: step: 1784/529, loss: 0.39702004194259644 2023-01-21 09:06:45.717536: step: 1788/529, loss: 0.7795924544334412 2023-01-21 09:06:46.854831: step: 1792/529, loss: 0.17830868065357208 2023-01-21 09:06:47.993263: step: 1796/529, loss: 0.2492694854736328 2023-01-21 09:06:49.102482: step: 1800/529, loss: 0.09401115775108337 2023-01-21 09:06:50.235254: step: 1804/529, loss: 0.18963509798049927 2023-01-21 09:06:51.411384: step: 1808/529, loss: 0.24590949714183807 2023-01-21 09:06:52.580350: step: 1812/529, loss: 0.9545996785163879 2023-01-21 09:06:53.709100: step: 1816/529, loss: 0.3508211374282837 2023-01-21 09:06:54.854776: step: 1820/529, loss: 0.1993359625339508 2023-01-21 09:06:55.988509: step: 1824/529, loss: 0.2797101140022278 2023-01-21 09:06:57.117392: step: 1828/529, loss: 0.30427980422973633 2023-01-21 09:06:58.276498: step: 1832/529, loss: 0.3970484137535095 2023-01-21 09:06:59.427367: step: 1836/529, loss: 0.28484582901000977 2023-01-21 09:07:00.559769: step: 1840/529, loss: 0.09865588694810867 2023-01-21 09:07:01.686240: step: 1844/529, loss: 0.39399316906929016 2023-01-21 09:07:02.818072: step: 1848/529, loss: 0.11836538463830948 2023-01-21 09:07:03.991738: step: 1852/529, loss: 0.2808743119239807 2023-01-21 09:07:05.131883: step: 1856/529, loss: 1.8474198579788208 2023-01-21 09:07:06.276866: step: 1860/529, loss: 0.08241195976734161 2023-01-21 09:07:07.413642: step: 1864/529, loss: 0.45600995421409607 2023-01-21 09:07:08.574210: step: 1868/529, loss: 0.07925812900066376 2023-01-21 09:07:09.707287: step: 1872/529, loss: 0.12548808753490448 2023-01-21 09:07:10.813849: step: 1876/529, loss: 0.2866532802581787 2023-01-21 09:07:11.981799: step: 1880/529, loss: 1.063374400138855 2023-01-21 09:07:13.141384: step: 1884/529, loss: 0.23573750257492065 2023-01-21 09:07:14.314292: step: 1888/529, loss: 0.37855157256126404 2023-01-21 09:07:15.463874: step: 1892/529, loss: 0.6805038452148438 2023-01-21 09:07:16.594341: step: 1896/529, loss: 0.11260871589183807 2023-01-21 09:07:17.740192: step: 1900/529, loss: 0.31062474846839905 2023-01-21 09:07:18.881783: step: 1904/529, loss: 0.7683759331703186 2023-01-21 09:07:20.020757: step: 1908/529, loss: 0.17898626625537872 2023-01-21 09:07:21.162920: step: 1912/529, loss: 0.15403634309768677 2023-01-21 09:07:22.290122: step: 1916/529, loss: 1.7226102352142334 2023-01-21 09:07:23.421124: step: 1920/529, loss: 0.09516830742359161 2023-01-21 09:07:24.564781: step: 1924/529, loss: 0.24629083275794983 2023-01-21 09:07:25.727375: step: 1928/529, loss: 1.373489260673523 2023-01-21 09:07:26.893072: step: 1932/529, loss: 1.7580018043518066 2023-01-21 09:07:28.027027: step: 1936/529, loss: 0.23545204102993011 2023-01-21 09:07:29.155214: step: 1940/529, loss: 0.13541316986083984 2023-01-21 09:07:30.274580: step: 1944/529, loss: 0.1970551609992981 2023-01-21 09:07:31.414040: step: 1948/529, loss: 1.1924164295196533 2023-01-21 09:07:32.548540: step: 1952/529, loss: 0.6954349279403687 2023-01-21 09:07:33.722829: step: 1956/529, loss: 0.11528763920068741 2023-01-21 09:07:34.894700: step: 1960/529, loss: 0.1927940398454666 2023-01-21 09:07:36.049769: step: 1964/529, loss: 0.21880188584327698 2023-01-21 09:07:37.190353: step: 1968/529, loss: 0.19161270558834076 2023-01-21 09:07:38.348923: step: 1972/529, loss: 0.049072980880737305 2023-01-21 09:07:39.468380: step: 1976/529, loss: 0.21237002313137054 2023-01-21 09:07:40.605175: step: 1980/529, loss: 0.7952540516853333 2023-01-21 09:07:41.747179: step: 1984/529, loss: 0.6937505006790161 2023-01-21 09:07:42.893217: step: 1988/529, loss: 0.8313053846359253 2023-01-21 09:07:44.019187: step: 1992/529, loss: 0.5270646810531616 2023-01-21 09:07:45.211315: step: 1996/529, loss: 6.488864421844482 2023-01-21 09:07:46.361400: step: 2000/529, loss: 0.2873687148094177 2023-01-21 09:07:47.513028: step: 2004/529, loss: 0.98978590965271 2023-01-21 09:07:48.676506: step: 2008/529, loss: 0.23303309082984924 2023-01-21 09:07:49.800343: step: 2012/529, loss: 0.2435900717973709 2023-01-21 09:07:50.956083: step: 2016/529, loss: 0.1796712875366211 2023-01-21 09:07:52.092928: step: 2020/529, loss: 2.326051950454712 2023-01-21 09:07:53.242800: step: 2024/529, loss: 0.18379679322242737 2023-01-21 09:07:54.407565: step: 2028/529, loss: 0.09204272925853729 2023-01-21 09:07:55.560717: step: 2032/529, loss: 1.51820969581604 2023-01-21 09:07:56.734819: step: 2036/529, loss: 1.0125430822372437 2023-01-21 09:07:57.892279: step: 2040/529, loss: 0.3211841583251953 2023-01-21 09:07:59.031391: step: 2044/529, loss: 0.21571610867977142 2023-01-21 09:08:00.178871: step: 2048/529, loss: 0.07568865269422531 2023-01-21 09:08:01.312944: step: 2052/529, loss: 0.055057816207408905 2023-01-21 09:08:02.458373: step: 2056/529, loss: 0.9104283452033997 2023-01-21 09:08:03.567716: step: 2060/529, loss: 0.8375106453895569 2023-01-21 09:08:04.730476: step: 2064/529, loss: 0.045627787709236145 2023-01-21 09:08:05.929565: step: 2068/529, loss: 1.3845282793045044 2023-01-21 09:08:07.066694: step: 2072/529, loss: 0.37636032700538635 2023-01-21 09:08:08.211951: step: 2076/529, loss: 0.15654096007347107 2023-01-21 09:08:09.345462: step: 2080/529, loss: 0.7759748101234436 2023-01-21 09:08:10.462940: step: 2084/529, loss: 0.19368892908096313 2023-01-21 09:08:11.602483: step: 2088/529, loss: 0.13523884117603302 2023-01-21 09:08:12.749333: step: 2092/529, loss: 1.2407002449035645 2023-01-21 09:08:13.860507: step: 2096/529, loss: 0.23802976310253143 2023-01-21 09:08:14.983462: step: 2100/529, loss: 0.09033231437206268 2023-01-21 09:08:16.142350: step: 2104/529, loss: 1.3069990873336792 2023-01-21 09:08:17.270916: step: 2108/529, loss: 0.18443596363067627 2023-01-21 09:08:18.419633: step: 2112/529, loss: 0.10947924107313156 2023-01-21 09:08:19.548817: step: 2116/529, loss: 1.9434700012207031 ================================================== Loss: 0.557 -------------------- Dev: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Test: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Chinese: {'event': {'p': 0.6301369863013698, 'r': 0.8518518518518519, 'f1': 0.7244094488188976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Korean: {'event': {'p': 0.6304347826086957, 'r': 0.4603174603174603, 'f1': 0.5321100917431192}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} 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': 3} New best chinese model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.6301369863013698, 'r': 0.8518518518518519, 'f1': 0.7244094488188976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.574869109947644, 'r': 0.7310252996005326, 'f1': 0.6436107854630715}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.5623836126629422, 'r': 0.6998841251448435, 'f1': 0.6236448115642746}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.7142857142857143, 'r': 0.47619047619047616, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Russian: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample 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': 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 09:09:16.068060: step: 4/529, loss: 0.24466048181056976 2023-01-21 09:09:17.201683: step: 8/529, loss: 0.11129050701856613 2023-01-21 09:09:18.330378: step: 12/529, loss: 0.16316938400268555 2023-01-21 09:09:19.465816: step: 16/529, loss: 0.3817647695541382 2023-01-21 09:09:20.586669: step: 20/529, loss: 0.1378255933523178 2023-01-21 09:09:21.733546: step: 24/529, loss: 0.3425389230251312 2023-01-21 09:09:22.853463: step: 28/529, loss: 0.17779813706874847 2023-01-21 09:09:23.968394: step: 32/529, loss: 1.1851680278778076 2023-01-21 09:09:25.096134: step: 36/529, loss: 0.16075149178504944 2023-01-21 09:09:26.220413: step: 40/529, loss: 0.12994928658008575 2023-01-21 09:09:27.361423: step: 44/529, loss: 0.17955541610717773 2023-01-21 09:09:28.515208: step: 48/529, loss: 0.12387075275182724 2023-01-21 09:09:29.670711: step: 52/529, loss: 0.11709528416395187 2023-01-21 09:09:30.807591: step: 56/529, loss: 0.2231781929731369 2023-01-21 09:09:31.940460: step: 60/529, loss: 1.452794075012207 2023-01-21 09:09:33.061283: step: 64/529, loss: 0.05014796555042267 2023-01-21 09:09:34.178719: step: 68/529, loss: 0.06693992763757706 2023-01-21 09:09:35.300077: step: 72/529, loss: 0.1076974868774414 2023-01-21 09:09:36.421959: step: 76/529, loss: 0.1814381182193756 2023-01-21 09:09:37.563734: step: 80/529, loss: 0.14542865753173828 2023-01-21 09:09:38.717139: step: 84/529, loss: 0.0811973586678505 2023-01-21 09:09:39.817199: step: 88/529, loss: 0.1084100529551506 2023-01-21 09:09:40.958952: step: 92/529, loss: 0.22772979736328125 2023-01-21 09:09:42.110676: step: 96/529, loss: 0.021969128400087357 2023-01-21 09:09:43.249797: step: 100/529, loss: 0.1365947723388672 2023-01-21 09:09:44.377015: step: 104/529, loss: 0.10780796408653259 2023-01-21 09:09:45.486377: step: 108/529, loss: 0.45810508728027344 2023-01-21 09:09:46.662573: step: 112/529, loss: 0.39113712310791016 2023-01-21 09:09:47.799609: step: 116/529, loss: 1.3864153623580933 2023-01-21 09:09:48.913865: step: 120/529, loss: 0.3315982222557068 2023-01-21 09:09:50.057044: step: 124/529, loss: 0.6596972346305847 2023-01-21 09:09:51.211601: step: 128/529, loss: 0.21459154784679413 2023-01-21 09:09:52.365048: step: 132/529, loss: 0.4117923676967621 2023-01-21 09:09:53.504022: step: 136/529, loss: 0.14763402938842773 2023-01-21 09:09:54.655138: step: 140/529, loss: 0.657440721988678 2023-01-21 09:09:55.778157: step: 144/529, loss: 0.6973082423210144 2023-01-21 09:09:56.942704: step: 148/529, loss: 0.7258090972900391 2023-01-21 09:09:58.065077: step: 152/529, loss: 0.12133245170116425 2023-01-21 09:09:59.171695: step: 156/529, loss: 0.11962912231683731 2023-01-21 09:10:00.317436: step: 160/529, loss: 0.09799843281507492 2023-01-21 09:10:01.444043: step: 164/529, loss: 0.16272759437561035 2023-01-21 09:10:02.577550: step: 168/529, loss: 1.2910672426223755 2023-01-21 09:10:03.733382: step: 172/529, loss: 0.6917350888252258 2023-01-21 09:10:04.876821: step: 176/529, loss: 0.16712398827075958 2023-01-21 09:10:06.029343: step: 180/529, loss: 0.18724757432937622 2023-01-21 09:10:07.172368: step: 184/529, loss: 0.0921320915222168 2023-01-21 09:10:08.332180: step: 188/529, loss: 0.2076890468597412 2023-01-21 09:10:09.498534: step: 192/529, loss: 0.1354234665632248 2023-01-21 09:10:10.638062: step: 196/529, loss: 0.20211133360862732 2023-01-21 09:10:11.784548: step: 200/529, loss: 1.0623573064804077 2023-01-21 09:10:12.923717: step: 204/529, loss: 0.44535207748413086 2023-01-21 09:10:14.078078: step: 208/529, loss: 0.17244748771190643 2023-01-21 09:10:15.218121: step: 212/529, loss: 0.6709588170051575 2023-01-21 09:10:16.379922: step: 216/529, loss: 0.07567596435546875 2023-01-21 09:10:17.519615: step: 220/529, loss: 0.0934787467122078 2023-01-21 09:10:18.646529: step: 224/529, loss: 0.07409334182739258 2023-01-21 09:10:19.799804: step: 228/529, loss: 0.24432912468910217 2023-01-21 09:10:20.951886: step: 232/529, loss: 0.21654443442821503 2023-01-21 09:10:22.076428: step: 236/529, loss: 0.06525573879480362 2023-01-21 09:10:23.226902: step: 240/529, loss: 0.09920063614845276 2023-01-21 09:10:24.406460: step: 244/529, loss: 0.0917000100016594 2023-01-21 09:10:25.556158: step: 248/529, loss: 0.06005983427166939 2023-01-21 09:10:26.692093: step: 252/529, loss: 1.3319963216781616 2023-01-21 09:10:27.804904: step: 256/529, loss: 0.6950969099998474 2023-01-21 09:10:28.933849: step: 260/529, loss: 0.7448029518127441 2023-01-21 09:10:30.077423: step: 264/529, loss: 0.7736472487449646 2023-01-21 09:10:31.205040: step: 268/529, loss: 0.28931865096092224 2023-01-21 09:10:32.355901: step: 272/529, loss: 0.13168630003929138 2023-01-21 09:10:33.511949: step: 276/529, loss: 0.8730499148368835 2023-01-21 09:10:34.653279: step: 280/529, loss: 0.2902229428291321 2023-01-21 09:10:35.789113: step: 284/529, loss: 0.18720950186252594 2023-01-21 09:10:36.922015: step: 288/529, loss: 0.15021830797195435 2023-01-21 09:10:38.091483: step: 292/529, loss: 0.3634463846683502 2023-01-21 09:10:39.256077: step: 296/529, loss: 0.3067096769809723 2023-01-21 09:10:40.388028: step: 300/529, loss: 0.08461332321166992 2023-01-21 09:10:41.523308: step: 304/529, loss: 0.08676118403673172 2023-01-21 09:10:42.646669: step: 308/529, loss: 0.08008575439453125 2023-01-21 09:10:43.770670: step: 312/529, loss: 0.11048805713653564 2023-01-21 09:10:44.918581: step: 316/529, loss: 0.1701037436723709 2023-01-21 09:10:46.081740: step: 320/529, loss: 0.23445215821266174 2023-01-21 09:10:47.229433: step: 324/529, loss: 0.12070250511169434 2023-01-21 09:10:48.361664: step: 328/529, loss: 0.16680152714252472 2023-01-21 09:10:49.538409: step: 332/529, loss: 0.38611292839050293 2023-01-21 09:10:50.684140: step: 336/529, loss: 0.07200375199317932 2023-01-21 09:10:51.851164: step: 340/529, loss: 0.6658762097358704 2023-01-21 09:10:53.062969: step: 344/529, loss: 0.33177271485328674 2023-01-21 09:10:54.209580: step: 348/529, loss: 0.4771179258823395 2023-01-21 09:10:55.350957: step: 352/529, loss: 0.16518573462963104 2023-01-21 09:10:56.498451: step: 356/529, loss: 0.07805061340332031 2023-01-21 09:10:57.641607: step: 360/529, loss: 0.5546988248825073 2023-01-21 09:10:58.759888: step: 364/529, loss: 0.11429693549871445 2023-01-21 09:10:59.895091: step: 368/529, loss: 0.08749409019947052 2023-01-21 09:11:01.054178: step: 372/529, loss: 0.0869588851928711 2023-01-21 09:11:02.179699: step: 376/529, loss: 0.17761069536209106 2023-01-21 09:11:03.308210: step: 380/529, loss: 0.08462834358215332 2023-01-21 09:11:04.432276: step: 384/529, loss: 0.06631775200366974 2023-01-21 09:11:05.570156: step: 388/529, loss: 0.18535280227661133 2023-01-21 09:11:06.712642: step: 392/529, loss: 0.7874088287353516 2023-01-21 09:11:07.862876: step: 396/529, loss: 0.05974102020263672 2023-01-21 09:11:09.017351: step: 400/529, loss: 0.1407126486301422 2023-01-21 09:11:10.141747: step: 404/529, loss: 0.09984250366687775 2023-01-21 09:11:11.262792: step: 408/529, loss: 0.0883147269487381 2023-01-21 09:11:12.412966: step: 412/529, loss: 0.31385642290115356 2023-01-21 09:11:13.536779: step: 416/529, loss: 0.05303812026977539 2023-01-21 09:11:14.643534: step: 420/529, loss: 0.10931988060474396 2023-01-21 09:11:15.784645: step: 424/529, loss: 0.8752502202987671 2023-01-21 09:11:16.897858: step: 428/529, loss: 0.02527799643576145 2023-01-21 09:11:18.039439: step: 432/529, loss: 0.24102745950222015 2023-01-21 09:11:19.172956: step: 436/529, loss: 0.09306631237268448 2023-01-21 09:11:20.323906: step: 440/529, loss: 0.9254924058914185 2023-01-21 09:11:21.472495: step: 444/529, loss: 0.74078369140625 2023-01-21 09:11:22.609386: step: 448/529, loss: 0.29278564453125 2023-01-21 09:11:23.744848: step: 452/529, loss: 0.8850492238998413 2023-01-21 09:11:24.881129: step: 456/529, loss: 0.16823837161064148 2023-01-21 09:11:26.019083: step: 460/529, loss: 0.41014689207077026 2023-01-21 09:11:27.150947: step: 464/529, loss: 0.017510438337922096 2023-01-21 09:11:28.276400: step: 468/529, loss: 0.47500133514404297 2023-01-21 09:11:29.421432: step: 472/529, loss: 0.6120018362998962 2023-01-21 09:11:30.547653: step: 476/529, loss: 0.08728199452161789 2023-01-21 09:11:31.691942: step: 480/529, loss: 0.11671486496925354 2023-01-21 09:11:32.838138: step: 484/529, loss: 0.6810390949249268 2023-01-21 09:11:33.963145: step: 488/529, loss: 0.0753626823425293 2023-01-21 09:11:35.096742: step: 492/529, loss: 0.787083625793457 2023-01-21 09:11:36.221819: step: 496/529, loss: 0.5044618248939514 2023-01-21 09:11:37.343611: step: 500/529, loss: 0.7071605324745178 2023-01-21 09:11:38.471882: step: 504/529, loss: 0.6244059205055237 2023-01-21 09:11:39.597677: step: 508/529, loss: 0.2614052891731262 2023-01-21 09:11:40.714545: step: 512/529, loss: 0.5192671418190002 2023-01-21 09:11:41.869979: step: 516/529, loss: 0.1447453796863556 2023-01-21 09:11:42.996617: step: 520/529, loss: 0.25025472044944763 2023-01-21 09:11:44.147465: step: 524/529, loss: 0.4523128271102905 2023-01-21 09:11:45.274795: step: 528/529, loss: 0.0843936949968338 2023-01-21 09:11:46.412611: step: 532/529, loss: 0.2642250955104828 2023-01-21 09:11:47.575885: step: 536/529, loss: 0.5646122694015503 2023-01-21 09:11:48.711815: step: 540/529, loss: 0.3771764039993286 2023-01-21 09:11:49.841013: step: 544/529, loss: 0.3648337423801422 2023-01-21 09:11:50.979370: step: 548/529, loss: 0.0735812783241272 2023-01-21 09:11:52.112548: step: 552/529, loss: 0.10009908676147461 2023-01-21 09:11:53.258675: step: 556/529, loss: 0.8021878600120544 2023-01-21 09:11:54.408603: step: 560/529, loss: 0.13122233748435974 2023-01-21 09:11:55.533635: step: 564/529, loss: 0.6014177799224854 2023-01-21 09:11:56.669591: step: 568/529, loss: 0.1372823715209961 2023-01-21 09:11:57.816574: step: 572/529, loss: 0.13991041481494904 2023-01-21 09:11:58.949712: step: 576/529, loss: 0.1183624267578125 2023-01-21 09:12:00.075879: step: 580/529, loss: 0.07020334899425507 2023-01-21 09:12:01.201345: step: 584/529, loss: 0.11784706264734268 2023-01-21 09:12:02.329216: step: 588/529, loss: 0.8573439717292786 2023-01-21 09:12:03.462425: step: 592/529, loss: 0.8647388219833374 2023-01-21 09:12:04.610082: step: 596/529, loss: 0.1837974637746811 2023-01-21 09:12:05.724351: step: 600/529, loss: 0.3412172794342041 2023-01-21 09:12:06.892797: step: 604/529, loss: 0.12396860122680664 2023-01-21 09:12:08.026852: step: 608/529, loss: 0.06313876807689667 2023-01-21 09:12:09.147770: step: 612/529, loss: 0.08566518127918243 2023-01-21 09:12:10.288264: step: 616/529, loss: 0.43108296394348145 2023-01-21 09:12:11.404479: step: 620/529, loss: 0.18287301063537598 2023-01-21 09:12:12.554840: step: 624/529, loss: 0.15974578261375427 2023-01-21 09:12:13.699462: step: 628/529, loss: 0.13344326615333557 2023-01-21 09:12:14.898541: step: 632/529, loss: 0.09007206559181213 2023-01-21 09:12:16.055009: step: 636/529, loss: 0.15565833449363708 2023-01-21 09:12:17.197112: step: 640/529, loss: 0.25013208389282227 2023-01-21 09:12:18.326357: step: 644/529, loss: 5.261640548706055 2023-01-21 09:12:19.483058: step: 648/529, loss: 0.29935896396636963 2023-01-21 09:12:20.608315: step: 652/529, loss: 6.333972454071045 2023-01-21 09:12:21.747355: step: 656/529, loss: 0.08399935066699982 2023-01-21 09:12:22.852220: step: 660/529, loss: 0.08417835831642151 2023-01-21 09:12:23.992480: step: 664/529, loss: 0.5735808610916138 2023-01-21 09:12:25.127750: step: 668/529, loss: 0.21012167632579803 2023-01-21 09:12:26.277179: step: 672/529, loss: 0.18438653647899628 2023-01-21 09:12:27.428251: step: 676/529, loss: 0.07517824321985245 2023-01-21 09:12:28.538301: step: 680/529, loss: 0.25263285636901855 2023-01-21 09:12:29.674700: step: 684/529, loss: 0.19530019164085388 2023-01-21 09:12:30.813034: step: 688/529, loss: 0.153814896941185 2023-01-21 09:12:31.962200: step: 692/529, loss: 0.17285394668579102 2023-01-21 09:12:33.080680: step: 696/529, loss: 0.21799883246421814 2023-01-21 09:12:34.229261: step: 700/529, loss: 0.14127960801124573 2023-01-21 09:12:35.365676: step: 704/529, loss: 0.2035275399684906 2023-01-21 09:12:36.501857: step: 708/529, loss: 0.05994949862360954 2023-01-21 09:12:37.653893: step: 712/529, loss: 0.4114479124546051 2023-01-21 09:12:38.781691: step: 716/529, loss: 0.4014359712600708 2023-01-21 09:12:39.938505: step: 720/529, loss: 1.0294115543365479 2023-01-21 09:12:41.088604: step: 724/529, loss: 0.876505970954895 2023-01-21 09:12:42.226230: step: 728/529, loss: 0.23265314102172852 2023-01-21 09:12:43.364973: step: 732/529, loss: 0.08632984757423401 2023-01-21 09:12:44.502952: step: 736/529, loss: 1.0822360515594482 2023-01-21 09:12:45.686968: step: 740/529, loss: 0.1900378167629242 2023-01-21 09:12:46.810486: step: 744/529, loss: 0.19116735458374023 2023-01-21 09:12:47.948769: step: 748/529, loss: 0.7378531098365784 2023-01-21 09:12:49.127973: step: 752/529, loss: 0.09987787902355194 2023-01-21 09:12:50.274341: step: 756/529, loss: 0.17485705018043518 2023-01-21 09:12:51.455897: step: 760/529, loss: 0.10996036231517792 2023-01-21 09:12:52.609750: step: 764/529, loss: 0.20369598269462585 2023-01-21 09:12:53.782515: step: 768/529, loss: 0.11662064492702484 2023-01-21 09:12:54.915305: step: 772/529, loss: 0.10381467640399933 2023-01-21 09:12:56.026885: step: 776/529, loss: 0.22626467049121857 2023-01-21 09:12:57.154998: step: 780/529, loss: 0.29835304617881775 2023-01-21 09:12:58.286465: step: 784/529, loss: 0.1822047233581543 2023-01-21 09:12:59.453940: step: 788/529, loss: 0.2610871195793152 2023-01-21 09:13:00.611939: step: 792/529, loss: 0.26282861828804016 2023-01-21 09:13:01.760946: step: 796/529, loss: 0.08006184548139572 2023-01-21 09:13:02.911130: step: 800/529, loss: 0.7551807761192322 2023-01-21 09:13:04.039746: step: 804/529, loss: 0.12836794555187225 2023-01-21 09:13:05.167825: step: 808/529, loss: 0.13698796927928925 2023-01-21 09:13:06.324368: step: 812/529, loss: 0.9885527491569519 2023-01-21 09:13:07.475670: step: 816/529, loss: 0.20509730279445648 2023-01-21 09:13:08.603852: step: 820/529, loss: 0.14692345261573792 2023-01-21 09:13:09.747292: step: 824/529, loss: 0.3278070092201233 2023-01-21 09:13:10.885306: step: 828/529, loss: 0.27918368577957153 2023-01-21 09:13:12.055708: step: 832/529, loss: 0.03283729776740074 2023-01-21 09:13:13.195820: step: 836/529, loss: 1.1109706163406372 2023-01-21 09:13:14.322151: step: 840/529, loss: 0.11066427826881409 2023-01-21 09:13:15.460035: step: 844/529, loss: 0.20504646003246307 2023-01-21 09:13:16.596301: step: 848/529, loss: 5.909251689910889 2023-01-21 09:13:17.710960: step: 852/529, loss: 0.6502519249916077 2023-01-21 09:13:18.828759: step: 856/529, loss: 0.14182119071483612 2023-01-21 09:13:19.989406: step: 860/529, loss: 0.09417720139026642 2023-01-21 09:13:21.118675: step: 864/529, loss: 0.7739275097846985 2023-01-21 09:13:22.320746: step: 868/529, loss: 0.1988564431667328 2023-01-21 09:13:23.436262: step: 872/529, loss: 0.09548965096473694 2023-01-21 09:13:24.554656: step: 876/529, loss: 0.0929441973567009 2023-01-21 09:13:25.687009: step: 880/529, loss: 0.10891647636890411 2023-01-21 09:13:26.818698: step: 884/529, loss: 0.6041755676269531 2023-01-21 09:13:27.954122: step: 888/529, loss: 0.7679823637008667 2023-01-21 09:13:29.098951: step: 892/529, loss: 0.19759735465049744 2023-01-21 09:13:30.233394: step: 896/529, loss: 0.20988713204860687 2023-01-21 09:13:31.370637: step: 900/529, loss: 0.929301381111145 2023-01-21 09:13:32.505697: step: 904/529, loss: 0.07646217197179794 2023-01-21 09:13:33.625398: step: 908/529, loss: 1.0134717226028442 2023-01-21 09:13:34.754317: step: 912/529, loss: 0.1411271095275879 2023-01-21 09:13:35.876082: step: 916/529, loss: 0.09640460461378098 2023-01-21 09:13:37.022543: step: 920/529, loss: 0.1452304869890213 2023-01-21 09:13:38.194364: step: 924/529, loss: 0.264639288187027 2023-01-21 09:13:39.348054: step: 928/529, loss: 0.02313246577978134 2023-01-21 09:13:40.496993: step: 932/529, loss: 0.11836281418800354 2023-01-21 09:13:41.621915: step: 936/529, loss: 0.2608543336391449 2023-01-21 09:13:42.767874: step: 940/529, loss: 0.19579553604125977 2023-01-21 09:13:43.883287: step: 944/529, loss: 0.07904257625341415 2023-01-21 09:13:45.016621: step: 948/529, loss: 0.10473886132240295 2023-01-21 09:13:46.169284: step: 952/529, loss: 0.17764106392860413 2023-01-21 09:13:47.341932: step: 956/529, loss: 0.132731631398201 2023-01-21 09:13:48.475446: step: 960/529, loss: 0.18035712838172913 2023-01-21 09:13:49.596099: step: 964/529, loss: 0.14639359712600708 2023-01-21 09:13:50.707929: step: 968/529, loss: 0.18622994422912598 2023-01-21 09:13:51.858103: step: 972/529, loss: 0.266743928194046 2023-01-21 09:13:53.025312: step: 976/529, loss: 0.08481035381555557 2023-01-21 09:13:54.153630: step: 980/529, loss: 0.16922646760940552 2023-01-21 09:13:55.279157: step: 984/529, loss: 0.15957021713256836 2023-01-21 09:13:56.402964: step: 988/529, loss: 0.10932941734790802 2023-01-21 09:13:57.555450: step: 992/529, loss: 0.5267252922058105 2023-01-21 09:13:58.702739: step: 996/529, loss: 0.08189411461353302 2023-01-21 09:13:59.836384: step: 1000/529, loss: 0.09332104027271271 2023-01-21 09:14:00.991761: step: 1004/529, loss: 1.0411455631256104 2023-01-21 09:14:02.108591: step: 1008/529, loss: 0.2185802459716797 2023-01-21 09:14:03.244419: step: 1012/529, loss: 0.10366229712963104 2023-01-21 09:14:04.401288: step: 1016/529, loss: 0.20752954483032227 2023-01-21 09:14:05.530745: step: 1020/529, loss: 0.1996767520904541 2023-01-21 09:14:06.650376: step: 1024/529, loss: 0.8366801738739014 2023-01-21 09:14:07.793614: step: 1028/529, loss: 0.05245428532361984 2023-01-21 09:14:08.907163: step: 1032/529, loss: 0.1484830379486084 2023-01-21 09:14:10.061301: step: 1036/529, loss: 0.10129952430725098 2023-01-21 09:14:11.201704: step: 1040/529, loss: 0.5959730744361877 2023-01-21 09:14:12.373798: step: 1044/529, loss: 0.12547464668750763 2023-01-21 09:14:13.519460: step: 1048/529, loss: 0.8448886871337891 2023-01-21 09:14:14.654001: step: 1052/529, loss: 0.05065946653485298 2023-01-21 09:14:15.806379: step: 1056/529, loss: 0.36637231707572937 2023-01-21 09:14:16.915870: step: 1060/529, loss: 0.5167497396469116 2023-01-21 09:14:18.079188: step: 1064/529, loss: 0.4764404296875 2023-01-21 09:14:19.187217: step: 1068/529, loss: 0.08988809585571289 2023-01-21 09:14:20.317314: step: 1072/529, loss: 0.01889929734170437 2023-01-21 09:14:21.424044: step: 1076/529, loss: 0.1427462249994278 2023-01-21 09:14:22.577953: step: 1080/529, loss: 0.7212098836898804 2023-01-21 09:14:23.694216: step: 1084/529, loss: 0.5810263156890869 2023-01-21 09:14:24.826349: step: 1088/529, loss: 0.33085861802101135 2023-01-21 09:14:25.973003: step: 1092/529, loss: 0.11662331223487854 2023-01-21 09:14:27.094512: step: 1096/529, loss: 0.42196404933929443 2023-01-21 09:14:28.237240: step: 1100/529, loss: 0.7228653430938721 2023-01-21 09:14:29.367289: step: 1104/529, loss: 0.09744454175233841 2023-01-21 09:14:30.510980: step: 1108/529, loss: 0.6256855130195618 2023-01-21 09:14:31.650647: step: 1112/529, loss: 0.05705108493566513 2023-01-21 09:14:32.777345: step: 1116/529, loss: 0.06933460384607315 2023-01-21 09:14:33.905582: step: 1120/529, loss: 5.567978858947754 2023-01-21 09:14:35.055337: step: 1124/529, loss: 0.12601642310619354 2023-01-21 09:14:36.201978: step: 1128/529, loss: 0.06533446162939072 2023-01-21 09:14:37.350135: step: 1132/529, loss: 0.4048699736595154 2023-01-21 09:14:38.498455: step: 1136/529, loss: 0.3857220709323883 2023-01-21 09:14:39.617671: step: 1140/529, loss: 0.0532347671687603 2023-01-21 09:14:40.766555: step: 1144/529, loss: 0.09898900985717773 2023-01-21 09:14:41.893492: step: 1148/529, loss: 0.06811871379613876 2023-01-21 09:14:43.032751: step: 1152/529, loss: 1.375936508178711 2023-01-21 09:14:44.178490: step: 1156/529, loss: 0.07062454521656036 2023-01-21 09:14:45.299107: step: 1160/529, loss: 0.2772492468357086 2023-01-21 09:14:46.452081: step: 1164/529, loss: 0.4373779296875 2023-01-21 09:14:47.583823: step: 1168/529, loss: 0.5396996736526489 2023-01-21 09:14:48.729738: step: 1172/529, loss: 0.5175167322158813 2023-01-21 09:14:49.875729: step: 1176/529, loss: 0.1952734887599945 2023-01-21 09:14:51.019432: step: 1180/529, loss: 0.10334673523902893 2023-01-21 09:14:52.146649: step: 1184/529, loss: 0.16952762007713318 2023-01-21 09:14:53.282011: step: 1188/529, loss: 0.4987879693508148 2023-01-21 09:14:54.433708: step: 1192/529, loss: 0.06664963066577911 2023-01-21 09:14:55.565085: step: 1196/529, loss: 0.552270770072937 2023-01-21 09:14:56.725650: step: 1200/529, loss: 0.39646783471107483 2023-01-21 09:14:57.886356: step: 1204/529, loss: 0.6991478204727173 2023-01-21 09:14:59.031476: step: 1208/529, loss: 0.13083943724632263 2023-01-21 09:15:00.185782: step: 1212/529, loss: 0.0903620719909668 2023-01-21 09:15:01.316683: step: 1216/529, loss: 0.058579400181770325 2023-01-21 09:15:02.443207: step: 1220/529, loss: 0.4212229251861572 2023-01-21 09:15:03.588861: step: 1224/529, loss: 0.042082786560058594 2023-01-21 09:15:04.767399: step: 1228/529, loss: 0.9738116264343262 2023-01-21 09:15:05.910838: step: 1232/529, loss: 0.10584497451782227 2023-01-21 09:15:07.029289: step: 1236/529, loss: 0.5029982924461365 2023-01-21 09:15:08.187063: step: 1240/529, loss: 0.10875015705823898 2023-01-21 09:15:09.320822: step: 1244/529, loss: 0.10504321753978729 2023-01-21 09:15:10.458182: step: 1248/529, loss: 0.18976612389087677 2023-01-21 09:15:11.604866: step: 1252/529, loss: 0.15540161728858948 2023-01-21 09:15:12.719470: step: 1256/529, loss: 0.09569196403026581 2023-01-21 09:15:13.867514: step: 1260/529, loss: 0.05697832256555557 2023-01-21 09:15:15.017656: step: 1264/529, loss: 0.18261757493019104 2023-01-21 09:15:16.141461: step: 1268/529, loss: 0.4580424427986145 2023-01-21 09:15:17.298308: step: 1272/529, loss: 0.058217622339725494 2023-01-21 09:15:18.431414: step: 1276/529, loss: 0.1900755912065506 2023-01-21 09:15:19.577566: step: 1280/529, loss: 0.19807033240795135 2023-01-21 09:15:20.687358: step: 1284/529, loss: 0.23657692968845367 2023-01-21 09:15:21.818224: step: 1288/529, loss: 0.7006770968437195 2023-01-21 09:15:22.965293: step: 1292/529, loss: 0.1827041208744049 2023-01-21 09:15:24.080397: step: 1296/529, loss: 0.5632963180541992 2023-01-21 09:15:25.233476: step: 1300/529, loss: 0.189055398106575 2023-01-21 09:15:26.362863: step: 1304/529, loss: 0.1618351936340332 2023-01-21 09:15:27.512881: step: 1308/529, loss: 0.024189185351133347 2023-01-21 09:15:28.657884: step: 1312/529, loss: 0.21861210465431213 2023-01-21 09:15:29.758233: step: 1316/529, loss: 0.18013858795166016 2023-01-21 09:15:30.935688: step: 1320/529, loss: 0.22302323579788208 2023-01-21 09:15:32.062423: step: 1324/529, loss: 0.07286234200000763 2023-01-21 09:15:33.197402: step: 1328/529, loss: 0.1516880989074707 2023-01-21 09:15:34.315368: step: 1332/529, loss: 0.026779651641845703 2023-01-21 09:15:35.446900: step: 1336/529, loss: 0.19312362372875214 2023-01-21 09:15:36.572108: step: 1340/529, loss: 0.024222230538725853 2023-01-21 09:15:37.698414: step: 1344/529, loss: 0.5090200304985046 2023-01-21 09:15:38.845333: step: 1348/529, loss: 0.1182275339961052 2023-01-21 09:15:40.012959: step: 1352/529, loss: 1.2169787883758545 2023-01-21 09:15:41.161685: step: 1356/529, loss: 1.5662081241607666 2023-01-21 09:15:42.293559: step: 1360/529, loss: 0.854303777217865 2023-01-21 09:15:43.445701: step: 1364/529, loss: 0.09669408947229385 2023-01-21 09:15:44.587708: step: 1368/529, loss: 0.2680419087409973 2023-01-21 09:15:45.746163: step: 1372/529, loss: 0.22132301330566406 2023-01-21 09:15:46.879530: step: 1376/529, loss: 0.08938374370336533 2023-01-21 09:15:47.996902: step: 1380/529, loss: 0.0645257979631424 2023-01-21 09:15:49.141598: step: 1384/529, loss: 0.7656015157699585 2023-01-21 09:15:50.253378: step: 1388/529, loss: 0.08659515529870987 2023-01-21 09:15:51.388899: step: 1392/529, loss: 0.24602393805980682 2023-01-21 09:15:52.511979: step: 1396/529, loss: 0.12929227948188782 2023-01-21 09:15:53.641912: step: 1400/529, loss: 0.09883227944374084 2023-01-21 09:15:54.780346: step: 1404/529, loss: 0.03693590313196182 2023-01-21 09:15:55.885189: step: 1408/529, loss: 0.14979323744773865 2023-01-21 09:15:57.047017: step: 1412/529, loss: 0.27159959077835083 2023-01-21 09:15:58.163091: step: 1416/529, loss: 0.06096029281616211 2023-01-21 09:15:59.269614: step: 1420/529, loss: 0.2789209485054016 2023-01-21 09:16:00.372918: step: 1424/529, loss: 0.13256461918354034 2023-01-21 09:16:01.501393: step: 1428/529, loss: 0.13757982850074768 2023-01-21 09:16:02.628909: step: 1432/529, loss: 0.09792099893093109 2023-01-21 09:16:03.788686: step: 1436/529, loss: 1.123103141784668 2023-01-21 09:16:04.932657: step: 1440/529, loss: 0.5919109582901001 2023-01-21 09:16:06.065964: step: 1444/529, loss: 0.6940789818763733 2023-01-21 09:16:07.238124: step: 1448/529, loss: 0.4919891357421875 2023-01-21 09:16:08.362626: step: 1452/529, loss: 0.1172342374920845 2023-01-21 09:16:09.500621: step: 1456/529, loss: 0.06498537212610245 2023-01-21 09:16:10.656817: step: 1460/529, loss: 0.1114787608385086 2023-01-21 09:16:11.783664: step: 1464/529, loss: 0.29998978972435 2023-01-21 09:16:12.937297: step: 1468/529, loss: 0.12245502322912216 2023-01-21 09:16:14.084999: step: 1472/529, loss: 0.17830339074134827 2023-01-21 09:16:15.228216: step: 1476/529, loss: 1.1929702758789062 2023-01-21 09:16:16.337183: step: 1480/529, loss: 0.5831903219223022 2023-01-21 09:16:17.482795: step: 1484/529, loss: 0.13941699266433716 2023-01-21 09:16:18.616668: step: 1488/529, loss: 0.17368373274803162 2023-01-21 09:16:19.794111: step: 1492/529, loss: 0.6343552470207214 2023-01-21 09:16:20.922036: step: 1496/529, loss: 0.4606724679470062 2023-01-21 09:16:22.067023: step: 1500/529, loss: 0.299577534198761 2023-01-21 09:16:23.218971: step: 1504/529, loss: 0.15684852004051208 2023-01-21 09:16:24.394443: step: 1508/529, loss: 0.04608802869915962 2023-01-21 09:16:25.564984: step: 1512/529, loss: 0.262067049741745 2023-01-21 09:16:26.718770: step: 1516/529, loss: 0.11879158765077591 2023-01-21 09:16:27.857437: step: 1520/529, loss: 0.3135489523410797 2023-01-21 09:16:28.997498: step: 1524/529, loss: 0.1385493278503418 2023-01-21 09:16:30.126331: step: 1528/529, loss: 4.754935264587402 2023-01-21 09:16:31.261301: step: 1532/529, loss: 0.05047278478741646 2023-01-21 09:16:32.377612: step: 1536/529, loss: 0.2229703962802887 2023-01-21 09:16:33.505446: step: 1540/529, loss: 0.1779649704694748 2023-01-21 09:16:34.631840: step: 1544/529, loss: 0.14939361810684204 2023-01-21 09:16:35.785742: step: 1548/529, loss: 0.03045635297894478 2023-01-21 09:16:36.961296: step: 1552/529, loss: 0.2526680827140808 2023-01-21 09:16:38.094918: step: 1556/529, loss: 0.32061004638671875 2023-01-21 09:16:39.240750: step: 1560/529, loss: 0.31971797347068787 2023-01-21 09:16:40.370825: step: 1564/529, loss: 0.8463422656059265 2023-01-21 09:16:41.508833: step: 1568/529, loss: 0.04594426229596138 2023-01-21 09:16:42.642112: step: 1572/529, loss: 0.13086232542991638 2023-01-21 09:16:43.787395: step: 1576/529, loss: 0.18717403709888458 2023-01-21 09:16:44.902978: step: 1580/529, loss: 0.05564568191766739 2023-01-21 09:16:46.063512: step: 1584/529, loss: 0.022755956277251244 2023-01-21 09:16:47.187296: step: 1588/529, loss: 0.5509677529335022 2023-01-21 09:16:48.324427: step: 1592/529, loss: 0.5876039266586304 2023-01-21 09:16:49.464965: step: 1596/529, loss: 0.4976891279220581 2023-01-21 09:16:50.598352: step: 1600/529, loss: 0.7775392532348633 2023-01-21 09:16:51.713709: step: 1604/529, loss: 0.2504066526889801 2023-01-21 09:16:52.842810: step: 1608/529, loss: 0.2704865038394928 2023-01-21 09:16:53.935247: step: 1612/529, loss: 0.1411399394273758 2023-01-21 09:16:55.069267: step: 1616/529, loss: 0.0903095081448555 2023-01-21 09:16:56.186328: step: 1620/529, loss: 0.1270519196987152 2023-01-21 09:16:57.345067: step: 1624/529, loss: 0.2893911302089691 2023-01-21 09:16:58.511670: step: 1628/529, loss: 0.43057534098625183 2023-01-21 09:16:59.648880: step: 1632/529, loss: 0.4903358817100525 2023-01-21 09:17:00.778052: step: 1636/529, loss: 0.04562988132238388 2023-01-21 09:17:01.894000: step: 1640/529, loss: 0.08480499684810638 2023-01-21 09:17:03.032661: step: 1644/529, loss: 0.31595534086227417 2023-01-21 09:17:04.176968: step: 1648/529, loss: 0.1193302646279335 2023-01-21 09:17:05.352287: step: 1652/529, loss: 0.28404197096824646 2023-01-21 09:17:06.514049: step: 1656/529, loss: 0.24520379304885864 2023-01-21 09:17:07.650763: step: 1660/529, loss: 0.09461050480604172 2023-01-21 09:17:08.789921: step: 1664/529, loss: 0.05453546345233917 2023-01-21 09:17:09.904096: step: 1668/529, loss: 0.4951627850532532 2023-01-21 09:17:11.041954: step: 1672/529, loss: 0.3646741807460785 2023-01-21 09:17:12.200767: step: 1676/529, loss: 0.2803020179271698 2023-01-21 09:17:13.363404: step: 1680/529, loss: 0.17202310264110565 2023-01-21 09:17:14.507984: step: 1684/529, loss: 0.1145220696926117 2023-01-21 09:17:15.634011: step: 1688/529, loss: 0.28769055008888245 2023-01-21 09:17:16.780788: step: 1692/529, loss: 0.13950395584106445 2023-01-21 09:17:17.937472: step: 1696/529, loss: 0.934943437576294 2023-01-21 09:17:19.100659: step: 1700/529, loss: 0.06297507882118225 2023-01-21 09:17:20.250279: step: 1704/529, loss: 0.28420257568359375 2023-01-21 09:17:21.378308: step: 1708/529, loss: 0.05071087181568146 2023-01-21 09:17:22.514030: step: 1712/529, loss: 0.23099786043167114 2023-01-21 09:17:23.674797: step: 1716/529, loss: 0.10282459110021591 2023-01-21 09:17:24.818367: step: 1720/529, loss: 0.24735870957374573 2023-01-21 09:17:25.937392: step: 1724/529, loss: 2.310133695602417 2023-01-21 09:17:27.060819: step: 1728/529, loss: 0.10250245034694672 2023-01-21 09:17:28.197228: step: 1732/529, loss: 0.10065250843763351 2023-01-21 09:17:29.352399: step: 1736/529, loss: 0.04280738905072212 2023-01-21 09:17:30.499923: step: 1740/529, loss: 0.5495694875717163 2023-01-21 09:17:31.643440: step: 1744/529, loss: 0.1267234832048416 2023-01-21 09:17:32.786865: step: 1748/529, loss: 0.4049305021762848 2023-01-21 09:17:33.917363: step: 1752/529, loss: 0.18132664263248444 2023-01-21 09:17:35.031621: step: 1756/529, loss: 0.7158254981040955 2023-01-21 09:17:36.167116: step: 1760/529, loss: 0.06081700325012207 2023-01-21 09:17:37.318138: step: 1764/529, loss: 0.14400339126586914 2023-01-21 09:17:38.418346: step: 1768/529, loss: 0.15114939212799072 2023-01-21 09:17:39.548125: step: 1772/529, loss: 0.12647433578968048 2023-01-21 09:17:40.696846: step: 1776/529, loss: 0.19220593571662903 2023-01-21 09:17:41.863594: step: 1780/529, loss: 0.03862974792718887 2023-01-21 09:17:43.008305: step: 1784/529, loss: 0.034238241612911224 2023-01-21 09:17:44.120974: step: 1788/529, loss: 0.39714276790618896 2023-01-21 09:17:45.265009: step: 1792/529, loss: 0.6737889051437378 2023-01-21 09:17:46.395312: step: 1796/529, loss: 0.1121646910905838 2023-01-21 09:17:47.558403: step: 1800/529, loss: 0.24534845352172852 2023-01-21 09:17:48.677475: step: 1804/529, loss: 0.1764148771762848 2023-01-21 09:17:49.807708: step: 1808/529, loss: 0.2621656358242035 2023-01-21 09:17:50.956973: step: 1812/529, loss: 0.3271993398666382 2023-01-21 09:17:52.102758: step: 1816/529, loss: 0.35607290267944336 2023-01-21 09:17:53.240437: step: 1820/529, loss: 0.05315547436475754 2023-01-21 09:17:54.364326: step: 1824/529, loss: 0.15396633744239807 2023-01-21 09:17:55.513190: step: 1828/529, loss: 0.1048869639635086 2023-01-21 09:17:56.640591: step: 1832/529, loss: 0.08683881908655167 2023-01-21 09:17:57.783102: step: 1836/529, loss: 0.4033700227737427 2023-01-21 09:17:58.904049: step: 1840/529, loss: 0.18419551849365234 2023-01-21 09:18:00.074239: step: 1844/529, loss: 0.09755392372608185 2023-01-21 09:18:01.214861: step: 1848/529, loss: 0.34812992811203003 2023-01-21 09:18:02.386334: step: 1852/529, loss: 0.05299287289381027 2023-01-21 09:18:03.517495: step: 1856/529, loss: 0.10171431303024292 2023-01-21 09:18:04.656898: step: 1860/529, loss: 0.08608823269605637 2023-01-21 09:18:05.813984: step: 1864/529, loss: 0.161363422870636 2023-01-21 09:18:06.934708: step: 1868/529, loss: 0.1409951150417328 2023-01-21 09:18:08.062606: step: 1872/529, loss: 0.7660082578659058 2023-01-21 09:18:09.210626: step: 1876/529, loss: 0.15494881570339203 2023-01-21 09:18:10.335333: step: 1880/529, loss: 0.33420228958129883 2023-01-21 09:18:11.489448: step: 1884/529, loss: 0.22919301688671112 2023-01-21 09:18:12.681135: step: 1888/529, loss: 0.07616036385297775 2023-01-21 09:18:13.792588: step: 1892/529, loss: 0.10895605385303497 2023-01-21 09:18:14.925647: step: 1896/529, loss: 0.22049346566200256 2023-01-21 09:18:16.078629: step: 1900/529, loss: 0.0726408064365387 2023-01-21 09:18:17.215476: step: 1904/529, loss: 0.041957952082157135 2023-01-21 09:18:18.344941: step: 1908/529, loss: 0.33600637316703796 2023-01-21 09:18:19.483945: step: 1912/529, loss: 0.27662912011146545 2023-01-21 09:18:20.608237: step: 1916/529, loss: 0.27566593885421753 2023-01-21 09:18:21.765745: step: 1920/529, loss: 0.14633388817310333 2023-01-21 09:18:22.938561: step: 1924/529, loss: 0.14777374267578125 2023-01-21 09:18:24.112410: step: 1928/529, loss: 0.2467377781867981 2023-01-21 09:18:25.245422: step: 1932/529, loss: 0.3071153163909912 2023-01-21 09:18:26.419032: step: 1936/529, loss: 0.0790814459323883 2023-01-21 09:18:27.572970: step: 1940/529, loss: 0.39208564162254333 2023-01-21 09:18:28.704500: step: 1944/529, loss: 0.09246049076318741 2023-01-21 09:18:29.858962: step: 1948/529, loss: 0.15102358162403107 2023-01-21 09:18:30.965696: step: 1952/529, loss: 0.050856590270996094 2023-01-21 09:18:32.120761: step: 1956/529, loss: 0.22097045183181763 2023-01-21 09:18:33.269939: step: 1960/529, loss: 0.7996987104415894 2023-01-21 09:18:34.398409: step: 1964/529, loss: 0.09530887752771378 2023-01-21 09:18:35.562473: step: 1968/529, loss: 0.039115048944950104 2023-01-21 09:18:36.711247: step: 1972/529, loss: 0.24181966483592987 2023-01-21 09:18:37.823833: step: 1976/529, loss: 0.06567387282848358 2023-01-21 09:18:38.976051: step: 1980/529, loss: 0.5974474549293518 2023-01-21 09:18:40.128556: step: 1984/529, loss: 0.2725328505039215 2023-01-21 09:18:41.255859: step: 1988/529, loss: 0.16146451234817505 2023-01-21 09:18:42.433759: step: 1992/529, loss: 0.261522114276886 2023-01-21 09:18:43.576322: step: 1996/529, loss: 0.25922155380249023 2023-01-21 09:18:44.711707: step: 2000/529, loss: 0.37495356798171997 2023-01-21 09:18:45.860780: step: 2004/529, loss: 1.3707140684127808 2023-01-21 09:18:47.037325: step: 2008/529, loss: 6.217792987823486 2023-01-21 09:18:48.154947: step: 2012/529, loss: 0.3286910653114319 2023-01-21 09:18:49.300119: step: 2016/529, loss: 0.22771930694580078 2023-01-21 09:18:50.414793: step: 2020/529, loss: 0.15865278244018555 2023-01-21 09:18:51.536516: step: 2024/529, loss: 0.26538124680519104 2023-01-21 09:18:52.666214: step: 2028/529, loss: 0.0401032455265522 2023-01-21 09:18:53.863832: step: 2032/529, loss: 0.17519842088222504 2023-01-21 09:18:55.002506: step: 2036/529, loss: 0.1319824755191803 2023-01-21 09:18:56.142761: step: 2040/529, loss: 0.12547016143798828 2023-01-21 09:18:57.287022: step: 2044/529, loss: 0.6281712055206299 2023-01-21 09:18:58.418133: step: 2048/529, loss: 0.07938146591186523 2023-01-21 09:18:59.538297: step: 2052/529, loss: 0.0992530807852745 2023-01-21 09:19:00.688504: step: 2056/529, loss: 0.1517200469970703 2023-01-21 09:19:01.816600: step: 2060/529, loss: 0.6488975286483765 2023-01-21 09:19:02.952807: step: 2064/529, loss: 0.111371710896492 2023-01-21 09:19:04.099768: step: 2068/529, loss: 0.4982793927192688 2023-01-21 09:19:05.229739: step: 2072/529, loss: 0.08705510944128036 2023-01-21 09:19:06.333002: step: 2076/529, loss: 0.19659142196178436 2023-01-21 09:19:07.453815: step: 2080/529, loss: 0.11851692199707031 2023-01-21 09:19:08.587613: step: 2084/529, loss: 0.18685206770896912 2023-01-21 09:19:09.709380: step: 2088/529, loss: 0.22805020213127136 2023-01-21 09:19:10.878100: step: 2092/529, loss: 1.03988778591156 2023-01-21 09:19:12.024186: step: 2096/529, loss: 0.23737582564353943 2023-01-21 09:19:13.175550: step: 2100/529, loss: 0.054612256586551666 2023-01-21 09:19:14.345707: step: 2104/529, loss: 0.20533724129199982 2023-01-21 09:19:15.494548: step: 2108/529, loss: 0.25845497846603394 2023-01-21 09:19:16.627061: step: 2112/529, loss: 0.07220889627933502 2023-01-21 09:19:17.774693: step: 2116/529, loss: 0.26004448533058167 ================================================== Loss: 0.354 -------------------- Dev: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Test: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Chinese: {'event': {'p': 0.6024096385542169, 'r': 0.9259259259259259, 'f1': 0.7299270072992702}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Korean: {'event': {'p': 0.6956521739130435, 'r': 0.5079365079365079, 'f1': 0.5871559633027522}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Russian: {'event': {'p': 0.5428571428571428, 'r': 0.5277777777777778, 'f1': 0.5352112676056338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.6301369863013698, 'r': 0.8518518518518519, 'f1': 0.7244094488188976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.6956521739130435, 'r': 0.5079365079365079, 'f1': 0.5871559633027522}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5428571428571428, 'r': 0.5277777777777778, 'f1': 0.5352112676056338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 5 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:20:15.744189: step: 4/529, loss: 0.22427862882614136 2023-01-21 09:20:16.908173: step: 8/529, loss: 0.10048151016235352 2023-01-21 09:20:18.018633: step: 12/529, loss: 0.06975755840539932 2023-01-21 09:20:19.138020: step: 16/529, loss: 0.07934622466564178 2023-01-21 09:20:20.252375: step: 20/529, loss: 0.11244344711303711 2023-01-21 09:20:21.356740: step: 24/529, loss: 0.21492882072925568 2023-01-21 09:20:22.488336: step: 28/529, loss: 0.08214688301086426 2023-01-21 09:20:23.629992: step: 32/529, loss: 0.14411644637584686 2023-01-21 09:20:24.781676: step: 36/529, loss: 0.06832685321569443 2023-01-21 09:20:25.939267: step: 40/529, loss: 0.9251334071159363 2023-01-21 09:20:27.070260: step: 44/529, loss: 0.05976181477308273 2023-01-21 09:20:28.196808: step: 48/529, loss: 0.08708854019641876 2023-01-21 09:20:29.311823: step: 52/529, loss: 0.06360912322998047 2023-01-21 09:20:30.443527: step: 56/529, loss: 0.08166372776031494 2023-01-21 09:20:31.582778: step: 60/529, loss: 0.22660627961158752 2023-01-21 09:20:32.731425: step: 64/529, loss: 0.1092415452003479 2023-01-21 09:20:33.870602: step: 68/529, loss: 0.15970100462436676 2023-01-21 09:20:35.017175: step: 72/529, loss: 0.149301216006279 2023-01-21 09:20:36.187621: step: 76/529, loss: 0.24440088868141174 2023-01-21 09:20:37.349483: step: 80/529, loss: 0.20189151167869568 2023-01-21 09:20:38.480853: step: 84/529, loss: 0.10554523766040802 2023-01-21 09:20:39.623476: step: 88/529, loss: 0.14898566901683807 2023-01-21 09:20:40.741044: step: 92/529, loss: 0.17725713551044464 2023-01-21 09:20:41.888005: step: 96/529, loss: 0.17440198361873627 2023-01-21 09:20:43.043793: step: 100/529, loss: 0.2739085555076599 2023-01-21 09:20:44.162159: step: 104/529, loss: 0.5840579867362976 2023-01-21 09:20:45.291833: step: 108/529, loss: 0.6455684900283813 2023-01-21 09:20:46.423369: step: 112/529, loss: 0.30396878719329834 2023-01-21 09:20:47.558564: step: 116/529, loss: 0.17748594284057617 2023-01-21 09:20:48.715295: step: 120/529, loss: 0.39729365706443787 2023-01-21 09:20:49.852212: step: 124/529, loss: 0.07466993480920792 2023-01-21 09:20:50.998811: step: 128/529, loss: 0.8461507558822632 2023-01-21 09:20:52.130160: step: 132/529, loss: 0.18433362245559692 2023-01-21 09:20:53.278549: step: 136/529, loss: 0.19498604536056519 2023-01-21 09:20:54.420496: step: 140/529, loss: 0.22603902220726013 2023-01-21 09:20:55.557890: step: 144/529, loss: 0.20145784318447113 2023-01-21 09:20:56.692358: step: 148/529, loss: 0.13971567153930664 2023-01-21 09:20:57.834861: step: 152/529, loss: 0.3831354081630707 2023-01-21 09:20:58.993542: step: 156/529, loss: 0.07915820926427841 2023-01-21 09:21:00.107691: step: 160/529, loss: 0.3496702313423157 2023-01-21 09:21:01.246954: step: 164/529, loss: 0.4002492427825928 2023-01-21 09:21:02.370174: step: 168/529, loss: 0.05228271335363388 2023-01-21 09:21:03.487005: step: 172/529, loss: 0.03417503833770752 2023-01-21 09:21:04.617874: step: 176/529, loss: 0.13278627395629883 2023-01-21 09:21:05.751250: step: 180/529, loss: 0.06800059974193573 2023-01-21 09:21:06.870748: step: 184/529, loss: 0.4448899030685425 2023-01-21 09:21:08.023361: step: 188/529, loss: 0.11199875175952911 2023-01-21 09:21:09.170060: step: 192/529, loss: 0.02708606794476509 2023-01-21 09:21:10.327221: step: 196/529, loss: 0.047673940658569336 2023-01-21 09:21:11.445410: step: 200/529, loss: 0.09937945008277893 2023-01-21 09:21:12.619434: step: 204/529, loss: 0.13282537460327148 2023-01-21 09:21:13.798408: step: 208/529, loss: 0.09272298961877823 2023-01-21 09:21:14.928824: step: 212/529, loss: 0.10857147723436356 2023-01-21 09:21:16.091526: step: 216/529, loss: 0.020266152918338776 2023-01-21 09:21:17.214223: step: 220/529, loss: 0.08863510936498642 2023-01-21 09:21:18.356738: step: 224/529, loss: 0.20711678266525269 2023-01-21 09:21:19.513488: step: 228/529, loss: 0.08193827420473099 2023-01-21 09:21:20.640515: step: 232/529, loss: 0.08115196228027344 2023-01-21 09:21:21.783932: step: 236/529, loss: 0.7200329899787903 2023-01-21 09:21:22.938971: step: 240/529, loss: 0.26839351654052734 2023-01-21 09:21:24.088104: step: 244/529, loss: 0.11625099927186966 2023-01-21 09:21:25.240972: step: 248/529, loss: 0.02670593373477459 2023-01-21 09:21:26.383457: step: 252/529, loss: 0.3011418282985687 2023-01-21 09:21:27.553986: step: 256/529, loss: 0.7763880491256714 2023-01-21 09:21:28.689639: step: 260/529, loss: 0.08882975578308105 2023-01-21 09:21:29.828510: step: 264/529, loss: 0.24912786483764648 2023-01-21 09:21:30.978557: step: 268/529, loss: 0.2661977708339691 2023-01-21 09:21:32.106526: step: 272/529, loss: 0.053052619099617004 2023-01-21 09:21:33.224773: step: 276/529, loss: 0.060025885701179504 2023-01-21 09:21:34.411996: step: 280/529, loss: 0.0874544158577919 2023-01-21 09:21:35.544236: step: 284/529, loss: 0.18547393381595612 2023-01-21 09:21:36.665342: step: 288/529, loss: 0.5351834893226624 2023-01-21 09:21:37.781208: step: 292/529, loss: 0.06581669300794601 2023-01-21 09:21:38.902128: step: 296/529, loss: 0.22228574752807617 2023-01-21 09:21:40.055184: step: 300/529, loss: 0.8187047839164734 2023-01-21 09:21:41.186790: step: 304/529, loss: 0.04924288019537926 2023-01-21 09:21:42.317100: step: 308/529, loss: 0.1315666288137436 2023-01-21 09:21:43.456480: step: 312/529, loss: 0.03098888508975506 2023-01-21 09:21:44.622598: step: 316/529, loss: 0.4047360420227051 2023-01-21 09:21:45.755975: step: 320/529, loss: 0.047917917370796204 2023-01-21 09:21:46.889850: step: 324/529, loss: 0.11188822239637375 2023-01-21 09:21:48.025456: step: 328/529, loss: 0.38239985704421997 2023-01-21 09:21:49.134246: step: 332/529, loss: 0.15101128816604614 2023-01-21 09:21:50.237570: step: 336/529, loss: 0.05776116997003555 2023-01-21 09:21:51.383414: step: 340/529, loss: 0.15691126883029938 2023-01-21 09:21:52.526080: step: 344/529, loss: 0.7498137354850769 2023-01-21 09:21:53.669681: step: 348/529, loss: 0.09618520736694336 2023-01-21 09:21:54.803339: step: 352/529, loss: 0.40496835112571716 2023-01-21 09:21:55.922866: step: 356/529, loss: 0.10190154612064362 2023-01-21 09:21:57.055840: step: 360/529, loss: 0.017052555456757545 2023-01-21 09:21:58.207091: step: 364/529, loss: 0.05175848305225372 2023-01-21 09:21:59.338464: step: 368/529, loss: 0.6044270396232605 2023-01-21 09:22:00.477504: step: 372/529, loss: 0.1350676566362381 2023-01-21 09:22:01.633677: step: 376/529, loss: 0.19380322098731995 2023-01-21 09:22:02.789574: step: 380/529, loss: 0.27833104133605957 2023-01-21 09:22:03.923368: step: 384/529, loss: 1.098759651184082 2023-01-21 09:22:05.066452: step: 388/529, loss: 0.27214139699935913 2023-01-21 09:22:06.213747: step: 392/529, loss: 0.3750571310520172 2023-01-21 09:22:07.367668: step: 396/529, loss: 0.3386901915073395 2023-01-21 09:22:08.511357: step: 400/529, loss: 0.11729726940393448 2023-01-21 09:22:09.630087: step: 404/529, loss: 0.2177065908908844 2023-01-21 09:22:10.811925: step: 408/529, loss: 0.508978545665741 2023-01-21 09:22:11.941848: step: 412/529, loss: 0.051433611661195755 2023-01-21 09:22:13.117478: step: 416/529, loss: 0.37486132979393005 2023-01-21 09:22:14.240121: step: 420/529, loss: 0.42103490233421326 2023-01-21 09:22:15.419295: step: 424/529, loss: 0.863083004951477 2023-01-21 09:22:16.552986: step: 428/529, loss: 0.6344043612480164 2023-01-21 09:22:17.707317: step: 432/529, loss: 0.6767671704292297 2023-01-21 09:22:18.861851: step: 436/529, loss: 0.55816251039505 2023-01-21 09:22:20.016538: step: 440/529, loss: 0.14557896554470062 2023-01-21 09:22:21.144935: step: 444/529, loss: 0.2398478388786316 2023-01-21 09:22:22.266111: step: 448/529, loss: 0.0840398296713829 2023-01-21 09:22:23.375318: step: 452/529, loss: 0.04188041761517525 2023-01-21 09:22:24.505572: step: 456/529, loss: 0.1442861557006836 2023-01-21 09:22:25.636356: step: 460/529, loss: 0.06053047627210617 2023-01-21 09:22:26.802989: step: 464/529, loss: 0.01460943277925253 2023-01-21 09:22:27.927340: step: 468/529, loss: 0.35192325711250305 2023-01-21 09:22:29.056806: step: 472/529, loss: 0.07562866806983948 2023-01-21 09:22:30.204106: step: 476/529, loss: 0.22201451659202576 2023-01-21 09:22:31.333960: step: 480/529, loss: 0.33580923080444336 2023-01-21 09:22:32.498368: step: 484/529, loss: 0.48352259397506714 2023-01-21 09:22:33.632508: step: 488/529, loss: 0.06163311004638672 2023-01-21 09:22:34.761365: step: 492/529, loss: 0.09614791721105576 2023-01-21 09:22:35.899854: step: 496/529, loss: 0.1638510823249817 2023-01-21 09:22:37.059739: step: 500/529, loss: 0.05747361481189728 2023-01-21 09:22:38.227163: step: 504/529, loss: 0.08745288848876953 2023-01-21 09:22:39.391266: step: 508/529, loss: 0.1919899880886078 2023-01-21 09:22:40.529206: step: 512/529, loss: 0.04347677156329155 2023-01-21 09:22:41.679896: step: 516/529, loss: 0.062215328216552734 2023-01-21 09:22:42.799363: step: 520/529, loss: 0.18911391496658325 2023-01-21 09:22:43.959675: step: 524/529, loss: 0.5368407368659973 2023-01-21 09:22:45.133749: step: 528/529, loss: 0.32455766201019287 2023-01-21 09:22:46.253504: step: 532/529, loss: 0.18294811248779297 2023-01-21 09:22:47.377655: step: 536/529, loss: 0.12199468910694122 2023-01-21 09:22:48.497553: step: 540/529, loss: 0.4287474751472473 2023-01-21 09:22:49.605853: step: 544/529, loss: 0.5247993469238281 2023-01-21 09:22:50.756078: step: 548/529, loss: 0.09729328006505966 2023-01-21 09:22:51.882838: step: 552/529, loss: 0.09226283431053162 2023-01-21 09:22:53.040791: step: 556/529, loss: 0.05288305506110191 2023-01-21 09:22:54.179568: step: 560/529, loss: 0.3276095390319824 2023-01-21 09:22:55.335765: step: 564/529, loss: 0.026784610003232956 2023-01-21 09:22:56.471350: step: 568/529, loss: 0.0736333355307579 2023-01-21 09:22:57.630991: step: 572/529, loss: 0.7001714110374451 2023-01-21 09:22:58.750479: step: 576/529, loss: 0.1079442948102951 2023-01-21 09:22:59.889573: step: 580/529, loss: 0.14818096160888672 2023-01-21 09:23:01.005761: step: 584/529, loss: 0.09834060072898865 2023-01-21 09:23:02.141650: step: 588/529, loss: 0.40732526779174805 2023-01-21 09:23:03.285558: step: 592/529, loss: 0.3458413779735565 2023-01-21 09:23:04.414351: step: 596/529, loss: 0.2152392864227295 2023-01-21 09:23:05.575410: step: 600/529, loss: 0.2431774139404297 2023-01-21 09:23:06.672837: step: 604/529, loss: 0.04874591529369354 2023-01-21 09:23:07.796538: step: 608/529, loss: 0.14137907326221466 2023-01-21 09:23:08.958470: step: 612/529, loss: 0.05199284479022026 2023-01-21 09:23:10.101401: step: 616/529, loss: 0.1299666464328766 2023-01-21 09:23:11.217343: step: 620/529, loss: 0.06879501044750214 2023-01-21 09:23:12.372137: step: 624/529, loss: 0.4626198709011078 2023-01-21 09:23:13.519709: step: 628/529, loss: 0.40443477034568787 2023-01-21 09:23:14.666960: step: 632/529, loss: 0.0699642151594162 2023-01-21 09:23:15.816762: step: 636/529, loss: 0.6595377922058105 2023-01-21 09:23:16.961848: step: 640/529, loss: 0.2773308753967285 2023-01-21 09:23:18.086160: step: 644/529, loss: 0.19436411559581757 2023-01-21 09:23:19.227115: step: 648/529, loss: 0.38811931014060974 2023-01-21 09:23:20.373006: step: 652/529, loss: 0.13943955302238464 2023-01-21 09:23:21.518793: step: 656/529, loss: 0.03386202082037926 2023-01-21 09:23:22.690199: step: 660/529, loss: 0.06380109488964081 2023-01-21 09:23:23.861279: step: 664/529, loss: 0.10193099826574326 2023-01-21 09:23:24.997840: step: 668/529, loss: 0.3615955114364624 2023-01-21 09:23:26.167118: step: 672/529, loss: 0.06915836036205292 2023-01-21 09:23:27.302000: step: 676/529, loss: 0.026192951947450638 2023-01-21 09:23:28.422720: step: 680/529, loss: 0.16068187355995178 2023-01-21 09:23:29.575451: step: 684/529, loss: 0.9120908975601196 2023-01-21 09:23:30.722092: step: 688/529, loss: 0.18342283368110657 2023-01-21 09:23:31.862614: step: 692/529, loss: 0.09877967834472656 2023-01-21 09:23:32.983693: step: 696/529, loss: 0.4771346151828766 2023-01-21 09:23:34.100771: step: 700/529, loss: 1.077742576599121 2023-01-21 09:23:35.235446: step: 704/529, loss: 1.4014058113098145 2023-01-21 09:23:36.377188: step: 708/529, loss: 0.16131535172462463 2023-01-21 09:23:37.502159: step: 712/529, loss: 0.09649448096752167 2023-01-21 09:23:38.626692: step: 716/529, loss: 0.4106588363647461 2023-01-21 09:23:39.763131: step: 720/529, loss: 0.08541478961706161 2023-01-21 09:23:40.892877: step: 724/529, loss: 0.05950794368982315 2023-01-21 09:23:42.045753: step: 728/529, loss: 0.06594400852918625 2023-01-21 09:23:43.190293: step: 732/529, loss: 0.32339897751808167 2023-01-21 09:23:44.298876: step: 736/529, loss: 0.028058623895049095 2023-01-21 09:23:45.450683: step: 740/529, loss: 0.7276901006698608 2023-01-21 09:23:46.641170: step: 744/529, loss: 0.09411649405956268 2023-01-21 09:23:47.774650: step: 748/529, loss: 0.07759447395801544 2023-01-21 09:23:48.925352: step: 752/529, loss: 0.21213531494140625 2023-01-21 09:23:50.070789: step: 756/529, loss: 0.17913202941417694 2023-01-21 09:23:51.215493: step: 760/529, loss: 0.23558884859085083 2023-01-21 09:23:52.334659: step: 764/529, loss: 0.5390642285346985 2023-01-21 09:23:53.471763: step: 768/529, loss: 0.1820848435163498 2023-01-21 09:23:54.638655: step: 772/529, loss: 0.3198251724243164 2023-01-21 09:23:55.841603: step: 776/529, loss: 0.38268783688545227 2023-01-21 09:23:56.999445: step: 780/529, loss: 0.12136220932006836 2023-01-21 09:23:58.112904: step: 784/529, loss: 0.45440828800201416 2023-01-21 09:23:59.231332: step: 788/529, loss: 0.17436981201171875 2023-01-21 09:24:00.378804: step: 792/529, loss: 0.46746596693992615 2023-01-21 09:24:01.558899: step: 796/529, loss: 0.7969484329223633 2023-01-21 09:24:02.721551: step: 800/529, loss: 0.05847759172320366 2023-01-21 09:24:03.885061: step: 804/529, loss: 0.17216472327709198 2023-01-21 09:24:05.037284: step: 808/529, loss: 0.07059316337108612 2023-01-21 09:24:06.147709: step: 812/529, loss: 0.1346181333065033 2023-01-21 09:24:07.311099: step: 816/529, loss: 0.10951745510101318 2023-01-21 09:24:08.436836: step: 820/529, loss: 0.13720086216926575 2023-01-21 09:24:09.578855: step: 824/529, loss: 0.08416938781738281 2023-01-21 09:24:10.729047: step: 828/529, loss: 0.11608610302209854 2023-01-21 09:24:11.911543: step: 832/529, loss: 0.05738382041454315 2023-01-21 09:24:13.032774: step: 836/529, loss: 0.19938774406909943 2023-01-21 09:24:14.173766: step: 840/529, loss: 0.13680848479270935 2023-01-21 09:24:15.307167: step: 844/529, loss: 0.6348241567611694 2023-01-21 09:24:16.432812: step: 848/529, loss: 0.28561049699783325 2023-01-21 09:24:17.604735: step: 852/529, loss: 0.11899995803833008 2023-01-21 09:24:18.749645: step: 856/529, loss: 0.6188222765922546 2023-01-21 09:24:19.905786: step: 860/529, loss: 0.15651655197143555 2023-01-21 09:24:21.052976: step: 864/529, loss: 0.3003966212272644 2023-01-21 09:24:22.209187: step: 868/529, loss: 0.2617243826389313 2023-01-21 09:24:23.348049: step: 872/529, loss: 0.16764774918556213 2023-01-21 09:24:24.535002: step: 876/529, loss: 0.2268868088722229 2023-01-21 09:24:25.678342: step: 880/529, loss: 0.05832281336188316 2023-01-21 09:24:26.858037: step: 884/529, loss: 0.16580167412757874 2023-01-21 09:24:27.967556: step: 888/529, loss: 0.8168898820877075 2023-01-21 09:24:29.099922: step: 892/529, loss: 0.10663219541311264 2023-01-21 09:24:30.193610: step: 896/529, loss: 0.023827457800507545 2023-01-21 09:24:31.343032: step: 900/529, loss: 0.4531335234642029 2023-01-21 09:24:32.479211: step: 904/529, loss: 0.06965355575084686 2023-01-21 09:24:33.573616: step: 908/529, loss: 0.10445914417505264 2023-01-21 09:24:34.719318: step: 912/529, loss: 0.820561408996582 2023-01-21 09:24:35.819292: step: 916/529, loss: 0.018340062350034714 2023-01-21 09:24:36.979615: step: 920/529, loss: 0.1836930811405182 2023-01-21 09:24:38.108220: step: 924/529, loss: 0.24556264281272888 2023-01-21 09:24:39.206486: step: 928/529, loss: 0.007442331872880459 2023-01-21 09:24:40.348785: step: 932/529, loss: 0.17948399484157562 2023-01-21 09:24:41.481378: step: 936/529, loss: 0.16641484200954437 2023-01-21 09:24:42.644884: step: 940/529, loss: 0.492008775472641 2023-01-21 09:24:43.771247: step: 944/529, loss: 3.918454170227051 2023-01-21 09:24:44.894377: step: 948/529, loss: 0.137139230966568 2023-01-21 09:24:46.026539: step: 952/529, loss: 0.14130143821239471 2023-01-21 09:24:47.173441: step: 956/529, loss: 0.09719114005565643 2023-01-21 09:24:48.310343: step: 960/529, loss: 0.06868916004896164 2023-01-21 09:24:49.455748: step: 964/529, loss: 0.27867984771728516 2023-01-21 09:24:50.611656: step: 968/529, loss: 0.1194157600402832 2023-01-21 09:24:51.736991: step: 972/529, loss: 0.11559510231018066 2023-01-21 09:24:52.876709: step: 976/529, loss: 0.15028801560401917 2023-01-21 09:24:54.029738: step: 980/529, loss: 0.09685392677783966 2023-01-21 09:24:55.162891: step: 984/529, loss: 0.4270941913127899 2023-01-21 09:24:56.294739: step: 988/529, loss: 0.30056682229042053 2023-01-21 09:24:57.398458: step: 992/529, loss: 0.547038197517395 2023-01-21 09:24:58.506732: step: 996/529, loss: 0.12587013840675354 2023-01-21 09:24:59.621474: step: 1000/529, loss: 0.5889007449150085 2023-01-21 09:25:00.765769: step: 1004/529, loss: 0.1895991861820221 2023-01-21 09:25:01.875522: step: 1008/529, loss: 0.08883605152368546 2023-01-21 09:25:03.010680: step: 1012/529, loss: 0.08840571343898773 2023-01-21 09:25:04.175742: step: 1016/529, loss: 1.289722204208374 2023-01-21 09:25:05.317004: step: 1020/529, loss: 0.25252074003219604 2023-01-21 09:25:06.434708: step: 1024/529, loss: 0.06562743335962296 2023-01-21 09:25:07.543964: step: 1028/529, loss: 0.05796804279088974 2023-01-21 09:25:08.656117: step: 1032/529, loss: 1.5989314317703247 2023-01-21 09:25:09.774962: step: 1036/529, loss: 0.09847583621740341 2023-01-21 09:25:10.886482: step: 1040/529, loss: 0.12631091475486755 2023-01-21 09:25:12.029962: step: 1044/529, loss: 0.1590011715888977 2023-01-21 09:25:13.161651: step: 1048/529, loss: 0.03858909755945206 2023-01-21 09:25:14.304314: step: 1052/529, loss: 0.11746521294116974 2023-01-21 09:25:15.454711: step: 1056/529, loss: 0.26975998282432556 2023-01-21 09:25:16.610145: step: 1060/529, loss: 0.15761476755142212 2023-01-21 09:25:17.727383: step: 1064/529, loss: 0.1558375358581543 2023-01-21 09:25:18.878468: step: 1068/529, loss: 0.16191354393959045 2023-01-21 09:25:20.018384: step: 1072/529, loss: 0.16426125168800354 2023-01-21 09:25:21.141550: step: 1076/529, loss: 0.10819950699806213 2023-01-21 09:25:22.276914: step: 1080/529, loss: 0.2586577236652374 2023-01-21 09:25:23.426918: step: 1084/529, loss: 0.1367994248867035 2023-01-21 09:25:24.565013: step: 1088/529, loss: 0.11013297736644745 2023-01-21 09:25:25.704279: step: 1092/529, loss: 0.23391591012477875 2023-01-21 09:25:26.894979: step: 1096/529, loss: 0.1418527215719223 2023-01-21 09:25:28.015648: step: 1100/529, loss: 0.05804424360394478 2023-01-21 09:25:29.145794: step: 1104/529, loss: 0.3440670371055603 2023-01-21 09:25:30.288453: step: 1108/529, loss: 0.10559874027967453 2023-01-21 09:25:31.397666: step: 1112/529, loss: 0.6162834167480469 2023-01-21 09:25:32.545841: step: 1116/529, loss: 0.034951020032167435 2023-01-21 09:25:33.701478: step: 1120/529, loss: 0.056363776326179504 2023-01-21 09:25:34.820760: step: 1124/529, loss: 0.10268769413232803 2023-01-21 09:25:35.952390: step: 1128/529, loss: 0.1996818482875824 2023-01-21 09:25:37.092683: step: 1132/529, loss: 0.4804918169975281 2023-01-21 09:25:38.221176: step: 1136/529, loss: 0.30917951464653015 2023-01-21 09:25:39.354891: step: 1140/529, loss: 0.01994924619793892 2023-01-21 09:25:40.500422: step: 1144/529, loss: 0.08589410781860352 2023-01-21 09:25:41.666446: step: 1148/529, loss: 0.20463895797729492 2023-01-21 09:25:42.812958: step: 1152/529, loss: 0.5892337560653687 2023-01-21 09:25:43.954863: step: 1156/529, loss: 0.15211649239063263 2023-01-21 09:25:45.122371: step: 1160/529, loss: 0.5815248489379883 2023-01-21 09:25:46.270648: step: 1164/529, loss: 0.24427394568920135 2023-01-21 09:25:47.402260: step: 1168/529, loss: 0.05843021720647812 2023-01-21 09:25:48.507288: step: 1172/529, loss: 0.03600635752081871 2023-01-21 09:25:49.656936: step: 1176/529, loss: 1.3941996097564697 2023-01-21 09:25:50.769459: step: 1180/529, loss: 0.08063025772571564 2023-01-21 09:25:51.870560: step: 1184/529, loss: 0.0839991569519043 2023-01-21 09:25:53.028910: step: 1188/529, loss: 0.06887964904308319 2023-01-21 09:25:54.182865: step: 1192/529, loss: 0.17440348863601685 2023-01-21 09:25:55.366777: step: 1196/529, loss: 0.09976129233837128 2023-01-21 09:25:56.491376: step: 1200/529, loss: 0.04871072620153427 2023-01-21 09:25:57.632069: step: 1204/529, loss: 0.08659162372350693 2023-01-21 09:25:58.809505: step: 1208/529, loss: 0.35692697763442993 2023-01-21 09:25:59.953369: step: 1212/529, loss: 0.314456582069397 2023-01-21 09:26:01.101275: step: 1216/529, loss: 0.04938311129808426 2023-01-21 09:26:02.220569: step: 1220/529, loss: 0.39713582396507263 2023-01-21 09:26:03.327735: step: 1224/529, loss: 0.23301349580287933 2023-01-21 09:26:04.469484: step: 1228/529, loss: 0.058625318109989166 2023-01-21 09:26:05.612443: step: 1232/529, loss: 0.2063652127981186 2023-01-21 09:26:06.739111: step: 1236/529, loss: 0.2320031225681305 2023-01-21 09:26:07.873221: step: 1240/529, loss: 0.21889905631542206 2023-01-21 09:26:08.998400: step: 1244/529, loss: 0.10784120857715607 2023-01-21 09:26:10.144583: step: 1248/529, loss: 0.25295644998550415 2023-01-21 09:26:11.308924: step: 1252/529, loss: 0.1691589057445526 2023-01-21 09:26:12.443977: step: 1256/529, loss: 0.5489426255226135 2023-01-21 09:26:13.592236: step: 1260/529, loss: 0.06394462287425995 2023-01-21 09:26:14.738957: step: 1264/529, loss: 0.273446261882782 2023-01-21 09:26:15.874100: step: 1268/529, loss: 0.1923009753227234 2023-01-21 09:26:16.996276: step: 1272/529, loss: 0.15968474745750427 2023-01-21 09:26:18.126247: step: 1276/529, loss: 0.056958820670843124 2023-01-21 09:26:19.263634: step: 1280/529, loss: 1.2160197496414185 2023-01-21 09:26:20.424937: step: 1284/529, loss: 1.1465739011764526 2023-01-21 09:26:21.558666: step: 1288/529, loss: 0.04865061491727829 2023-01-21 09:26:22.682596: step: 1292/529, loss: 0.042046401649713516 2023-01-21 09:26:23.795498: step: 1296/529, loss: 0.014796162024140358 2023-01-21 09:26:24.956639: step: 1300/529, loss: 0.04874224588274956 2023-01-21 09:26:26.099532: step: 1304/529, loss: 0.664114236831665 2023-01-21 09:26:27.262513: step: 1308/529, loss: 0.3814714252948761 2023-01-21 09:26:28.436186: step: 1312/529, loss: 0.07547149807214737 2023-01-21 09:26:29.564185: step: 1316/529, loss: 0.2374231219291687 2023-01-21 09:26:30.705074: step: 1320/529, loss: 0.09163355827331543 2023-01-21 09:26:31.853350: step: 1324/529, loss: 0.1001967042684555 2023-01-21 09:26:33.002646: step: 1328/529, loss: 0.09128275513648987 2023-01-21 09:26:34.150719: step: 1332/529, loss: 0.11172228306531906 2023-01-21 09:26:35.299007: step: 1336/529, loss: 0.18540725111961365 2023-01-21 09:26:36.439389: step: 1340/529, loss: 0.35773998498916626 2023-01-21 09:26:37.567222: step: 1344/529, loss: 0.1189003735780716 2023-01-21 09:26:38.710972: step: 1348/529, loss: 0.12460193037986755 2023-01-21 09:26:39.838631: step: 1352/529, loss: 0.10962381213903427 2023-01-21 09:26:40.962970: step: 1356/529, loss: 0.4055734872817993 2023-01-21 09:26:42.123091: step: 1360/529, loss: 0.14154282212257385 2023-01-21 09:26:43.260754: step: 1364/529, loss: 0.07836923748254776 2023-01-21 09:26:44.417351: step: 1368/529, loss: 0.4551607072353363 2023-01-21 09:26:45.571496: step: 1372/529, loss: 0.0414639450609684 2023-01-21 09:26:46.697416: step: 1376/529, loss: 0.29673391580581665 2023-01-21 09:26:47.833547: step: 1380/529, loss: 0.1371878832578659 2023-01-21 09:26:49.004554: step: 1384/529, loss: 0.11697244644165039 2023-01-21 09:26:50.155838: step: 1388/529, loss: 0.5230741500854492 2023-01-21 09:26:51.303577: step: 1392/529, loss: 0.32486194372177124 2023-01-21 09:26:52.460936: step: 1396/529, loss: 0.21487733721733093 2023-01-21 09:26:53.612910: step: 1400/529, loss: 0.18460053205490112 2023-01-21 09:26:54.710908: step: 1404/529, loss: 0.04629088565707207 2023-01-21 09:26:55.835178: step: 1408/529, loss: 0.29455748200416565 2023-01-21 09:26:56.964231: step: 1412/529, loss: 0.20230531692504883 2023-01-21 09:26:58.092332: step: 1416/529, loss: 0.13287538290023804 2023-01-21 09:26:59.256665: step: 1420/529, loss: 0.6037986874580383 2023-01-21 09:27:00.370108: step: 1424/529, loss: 0.044907521456480026 2023-01-21 09:27:01.494704: step: 1428/529, loss: 0.28747427463531494 2023-01-21 09:27:02.625303: step: 1432/529, loss: 0.08556830883026123 2023-01-21 09:27:03.761681: step: 1436/529, loss: 0.22988833487033844 2023-01-21 09:27:04.908479: step: 1440/529, loss: 0.48885446786880493 2023-01-21 09:27:06.057831: step: 1444/529, loss: 0.2024051696062088 2023-01-21 09:27:07.178611: step: 1448/529, loss: 0.42543095350265503 2023-01-21 09:27:08.282673: step: 1452/529, loss: 0.09166135638952255 2023-01-21 09:27:09.389945: step: 1456/529, loss: 0.2525315284729004 2023-01-21 09:27:10.515795: step: 1460/529, loss: 0.04898281395435333 2023-01-21 09:27:11.635060: step: 1464/529, loss: 0.11130361258983612 2023-01-21 09:27:12.764768: step: 1468/529, loss: 0.11058778315782547 2023-01-21 09:27:13.904954: step: 1472/529, loss: 0.0893896147608757 2023-01-21 09:27:15.020261: step: 1476/529, loss: 0.07288694381713867 2023-01-21 09:27:16.147582: step: 1480/529, loss: 0.06834030151367188 2023-01-21 09:27:17.297985: step: 1484/529, loss: 0.02478962019085884 2023-01-21 09:27:18.459947: step: 1488/529, loss: 0.1446254700422287 2023-01-21 09:27:19.611022: step: 1492/529, loss: 0.45494937896728516 2023-01-21 09:27:20.766167: step: 1496/529, loss: 1.6910068988800049 2023-01-21 09:27:21.933400: step: 1500/529, loss: 0.132727712392807 2023-01-21 09:27:23.073593: step: 1504/529, loss: 0.1272542029619217 2023-01-21 09:27:24.197506: step: 1508/529, loss: 0.2846756875514984 2023-01-21 09:27:25.321334: step: 1512/529, loss: 0.814903736114502 2023-01-21 09:27:26.480520: step: 1516/529, loss: 0.24233828485012054 2023-01-21 09:27:27.607658: step: 1520/529, loss: 0.4306846559047699 2023-01-21 09:27:28.730193: step: 1524/529, loss: 0.054372213780879974 2023-01-21 09:27:29.847057: step: 1528/529, loss: 0.32174140214920044 2023-01-21 09:27:30.971404: step: 1532/529, loss: 0.12267415970563889 2023-01-21 09:27:32.103652: step: 1536/529, loss: 0.07979068905115128 2023-01-21 09:27:33.254319: step: 1540/529, loss: 0.09366913139820099 2023-01-21 09:27:34.392667: step: 1544/529, loss: 0.12291507422924042 2023-01-21 09:27:35.548785: step: 1548/529, loss: 0.854139506816864 2023-01-21 09:27:36.699025: step: 1552/529, loss: 0.23747539520263672 2023-01-21 09:27:37.837725: step: 1556/529, loss: 0.0986103042960167 2023-01-21 09:27:38.979293: step: 1560/529, loss: 0.07693271338939667 2023-01-21 09:27:40.148118: step: 1564/529, loss: 0.02435903623700142 2023-01-21 09:27:41.312830: step: 1568/529, loss: 0.19802474975585938 2023-01-21 09:27:42.466230: step: 1572/529, loss: 0.11087074875831604 2023-01-21 09:27:43.623682: step: 1576/529, loss: 0.14248834550380707 2023-01-21 09:27:44.741589: step: 1580/529, loss: 0.023084450513124466 2023-01-21 09:27:45.853878: step: 1584/529, loss: 0.15884724259376526 2023-01-21 09:27:47.007838: step: 1588/529, loss: 0.06725578755140305 2023-01-21 09:27:48.143042: step: 1592/529, loss: 0.44117504358291626 2023-01-21 09:27:49.247769: step: 1596/529, loss: 0.1392897665500641 2023-01-21 09:27:50.367968: step: 1600/529, loss: 0.08854389190673828 2023-01-21 09:27:51.515360: step: 1604/529, loss: 0.20504769682884216 2023-01-21 09:27:52.658905: step: 1608/529, loss: 1.0618385076522827 2023-01-21 09:27:53.794715: step: 1612/529, loss: 0.2695992887020111 2023-01-21 09:27:54.916058: step: 1616/529, loss: 0.16080017387866974 2023-01-21 09:27:56.086815: step: 1620/529, loss: 0.48127710819244385 2023-01-21 09:27:57.217553: step: 1624/529, loss: 2.445857524871826 2023-01-21 09:27:58.350700: step: 1628/529, loss: 0.05478806793689728 2023-01-21 09:27:59.498780: step: 1632/529, loss: 0.07957783341407776 2023-01-21 09:28:00.618616: step: 1636/529, loss: 0.45860737562179565 2023-01-21 09:28:01.755229: step: 1640/529, loss: 0.15542663633823395 2023-01-21 09:28:02.921058: step: 1644/529, loss: 1.3824344873428345 2023-01-21 09:28:04.081687: step: 1648/529, loss: 0.2135431319475174 2023-01-21 09:28:05.220255: step: 1652/529, loss: 0.08407983183860779 2023-01-21 09:28:06.360389: step: 1656/529, loss: 0.04255995899438858 2023-01-21 09:28:07.489001: step: 1660/529, loss: 0.10629893094301224 2023-01-21 09:28:08.641953: step: 1664/529, loss: 0.18027277290821075 2023-01-21 09:28:09.768389: step: 1668/529, loss: 0.1216583251953125 2023-01-21 09:28:10.890441: step: 1672/529, loss: 0.10211792588233948 2023-01-21 09:28:12.019196: step: 1676/529, loss: 0.23892784118652344 2023-01-21 09:28:13.198498: step: 1680/529, loss: 0.0686337947845459 2023-01-21 09:28:14.337667: step: 1684/529, loss: 0.07210950553417206 2023-01-21 09:28:15.468280: step: 1688/529, loss: 0.3025897741317749 2023-01-21 09:28:16.631345: step: 1692/529, loss: 0.15028133988380432 2023-01-21 09:28:17.779328: step: 1696/529, loss: 0.13339270651340485 2023-01-21 09:28:18.924684: step: 1700/529, loss: 0.10011950135231018 2023-01-21 09:28:20.050836: step: 1704/529, loss: 0.14445707201957703 2023-01-21 09:28:21.183318: step: 1708/529, loss: 0.29445910453796387 2023-01-21 09:28:22.357764: step: 1712/529, loss: 0.3250378668308258 2023-01-21 09:28:23.488433: step: 1716/529, loss: 0.9704848527908325 2023-01-21 09:28:24.604689: step: 1720/529, loss: 0.20039625465869904 2023-01-21 09:28:25.738353: step: 1724/529, loss: 0.08942041546106339 2023-01-21 09:28:26.893748: step: 1728/529, loss: 0.0853240042924881 2023-01-21 09:28:28.038301: step: 1732/529, loss: 0.35689955949783325 2023-01-21 09:28:29.178965: step: 1736/529, loss: 0.5217226147651672 2023-01-21 09:28:30.297073: step: 1740/529, loss: 0.23207783699035645 2023-01-21 09:28:31.444641: step: 1744/529, loss: 0.18612231314182281 2023-01-21 09:28:32.583905: step: 1748/529, loss: 0.13245268166065216 2023-01-21 09:28:33.725839: step: 1752/529, loss: 0.10207509994506836 2023-01-21 09:28:34.887798: step: 1756/529, loss: 0.12094831466674805 2023-01-21 09:28:36.042487: step: 1760/529, loss: 0.27033597230911255 2023-01-21 09:28:37.199552: step: 1764/529, loss: 0.19553199410438538 2023-01-21 09:28:38.364673: step: 1768/529, loss: 1.1943162679672241 2023-01-21 09:28:39.494229: step: 1772/529, loss: 0.07562501728534698 2023-01-21 09:28:40.674542: step: 1776/529, loss: 0.37366580963134766 2023-01-21 09:28:41.804586: step: 1780/529, loss: 0.3176751732826233 2023-01-21 09:28:42.925702: step: 1784/529, loss: 0.08720416575670242 2023-01-21 09:28:44.059775: step: 1788/529, loss: 0.09665494412183762 2023-01-21 09:28:45.213189: step: 1792/529, loss: 0.10573120415210724 2023-01-21 09:28:46.344376: step: 1796/529, loss: 0.09265251457691193 2023-01-21 09:28:47.465605: step: 1800/529, loss: 0.5216751098632812 2023-01-21 09:28:48.571636: step: 1804/529, loss: 0.0836523100733757 2023-01-21 09:28:49.698793: step: 1808/529, loss: 0.0906444564461708 2023-01-21 09:28:50.822658: step: 1812/529, loss: 0.3712461590766907 2023-01-21 09:28:51.961672: step: 1816/529, loss: 0.6995339393615723 2023-01-21 09:28:53.113070: step: 1820/529, loss: 0.11586017161607742 2023-01-21 09:28:54.266541: step: 1824/529, loss: 0.02235245704650879 2023-01-21 09:28:55.428360: step: 1828/529, loss: 0.07299160957336426 2023-01-21 09:28:56.545696: step: 1832/529, loss: 0.7483769655227661 2023-01-21 09:28:57.666223: step: 1836/529, loss: 0.08506083488464355 2023-01-21 09:28:58.826818: step: 1840/529, loss: 0.16500243544578552 2023-01-21 09:28:59.956317: step: 1844/529, loss: 0.16785374283790588 2023-01-21 09:29:01.094274: step: 1848/529, loss: 0.8889985084533691 2023-01-21 09:29:02.249140: step: 1852/529, loss: 0.09586697071790695 2023-01-21 09:29:03.393816: step: 1856/529, loss: 0.16931210458278656 2023-01-21 09:29:04.588610: step: 1860/529, loss: 0.5723984837532043 2023-01-21 09:29:05.770220: step: 1864/529, loss: 0.6708760261535645 2023-01-21 09:29:06.895976: step: 1868/529, loss: 0.11356306076049805 2023-01-21 09:29:08.039431: step: 1872/529, loss: 0.4433327913284302 2023-01-21 09:29:09.213124: step: 1876/529, loss: 0.12220098823308945 2023-01-21 09:29:10.386600: step: 1880/529, loss: 0.25582271814346313 2023-01-21 09:29:11.548166: step: 1884/529, loss: 0.031248951330780983 2023-01-21 09:29:12.660334: step: 1888/529, loss: 0.10533743351697922 2023-01-21 09:29:13.774392: step: 1892/529, loss: 0.07743396610021591 2023-01-21 09:29:14.919597: step: 1896/529, loss: 0.08821959793567657 2023-01-21 09:29:16.054775: step: 1900/529, loss: 0.23253564536571503 2023-01-21 09:29:17.214432: step: 1904/529, loss: 0.09577665477991104 2023-01-21 09:29:18.365633: step: 1908/529, loss: 0.10028591006994247 2023-01-21 09:29:19.486548: step: 1912/529, loss: 0.07080264389514923 2023-01-21 09:29:20.607529: step: 1916/529, loss: 0.30206307768821716 2023-01-21 09:29:21.764523: step: 1920/529, loss: 3.4601290225982666 2023-01-21 09:29:22.889193: step: 1924/529, loss: 0.4781244397163391 2023-01-21 09:29:24.042614: step: 1928/529, loss: 0.1625032126903534 2023-01-21 09:29:25.172641: step: 1932/529, loss: 0.05069418251514435 2023-01-21 09:29:26.314449: step: 1936/529, loss: 0.15439115464687347 2023-01-21 09:29:27.424703: step: 1940/529, loss: 0.06670689582824707 2023-01-21 09:29:28.538488: step: 1944/529, loss: 0.0798218697309494 2023-01-21 09:29:29.649787: step: 1948/529, loss: 0.04812512546777725 2023-01-21 09:29:30.814647: step: 1952/529, loss: 0.314921498298645 2023-01-21 09:29:31.942203: step: 1956/529, loss: 0.053060341626405716 2023-01-21 09:29:33.088457: step: 1960/529, loss: 0.19245347380638123 2023-01-21 09:29:34.212913: step: 1964/529, loss: 0.08722953498363495 2023-01-21 09:29:35.328554: step: 1968/529, loss: 0.1384839415550232 2023-01-21 09:29:36.442340: step: 1972/529, loss: 0.04252929612994194 2023-01-21 09:29:37.601016: step: 1976/529, loss: 0.3670635223388672 2023-01-21 09:29:38.734227: step: 1980/529, loss: 0.03096799924969673 2023-01-21 09:29:39.850734: step: 1984/529, loss: 0.055448152124881744 2023-01-21 09:29:40.996648: step: 1988/529, loss: 0.14109449088573456 2023-01-21 09:29:42.149313: step: 1992/529, loss: 0.1049683541059494 2023-01-21 09:29:43.267375: step: 1996/529, loss: 0.05912170559167862 2023-01-21 09:29:44.427668: step: 2000/529, loss: 0.29120540618896484 2023-01-21 09:29:45.583080: step: 2004/529, loss: 0.7962394952774048 2023-01-21 09:29:46.715373: step: 2008/529, loss: 0.2000894546508789 2023-01-21 09:29:47.838165: step: 2012/529, loss: 0.18997983634471893 2023-01-21 09:29:48.958443: step: 2016/529, loss: 0.16265468299388885 2023-01-21 09:29:50.081692: step: 2020/529, loss: 0.04038066789507866 2023-01-21 09:29:51.194771: step: 2024/529, loss: 0.0757012888789177 2023-01-21 09:29:52.351497: step: 2028/529, loss: 0.2071550041437149 2023-01-21 09:29:53.522405: step: 2032/529, loss: 0.14090615510940552 2023-01-21 09:29:54.688061: step: 2036/529, loss: 0.15868401527404785 2023-01-21 09:29:55.806008: step: 2040/529, loss: 0.12236276268959045 2023-01-21 09:29:56.940052: step: 2044/529, loss: 0.15578851103782654 2023-01-21 09:29:58.053005: step: 2048/529, loss: 0.060861311852931976 2023-01-21 09:29:59.203999: step: 2052/529, loss: 0.03957071527838707 2023-01-21 09:30:00.359025: step: 2056/529, loss: 0.4482616186141968 2023-01-21 09:30:01.496560: step: 2060/529, loss: 0.797402024269104 2023-01-21 09:30:02.616911: step: 2064/529, loss: 0.8909168243408203 2023-01-21 09:30:03.759894: step: 2068/529, loss: 0.07186398655176163 2023-01-21 09:30:04.911937: step: 2072/529, loss: 0.6693910360336304 2023-01-21 09:30:06.038940: step: 2076/529, loss: 2.4048049449920654 2023-01-21 09:30:07.197373: step: 2080/529, loss: 0.23566734790802002 2023-01-21 09:30:08.390114: step: 2084/529, loss: 0.7993472814559937 2023-01-21 09:30:09.534388: step: 2088/529, loss: 0.1624138355255127 2023-01-21 09:30:10.682397: step: 2092/529, loss: 0.19414672255516052 2023-01-21 09:30:11.802529: step: 2096/529, loss: 0.12755393981933594 2023-01-21 09:30:12.946129: step: 2100/529, loss: 0.12087831646203995 2023-01-21 09:30:14.075913: step: 2104/529, loss: 0.09632611274719238 2023-01-21 09:30:15.245104: step: 2108/529, loss: 0.15626157820224762 2023-01-21 09:30:16.383312: step: 2112/529, loss: 0.9749086499214172 2023-01-21 09:30:17.556340: step: 2116/529, loss: 0.05219235271215439 ================================================== Loss: 0.265 -------------------- Dev: {'event': {'p': 0.5289256198347108, 'r': 0.7669773635153129, 'f1': 0.6260869565217392}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Test: {'event': {'p': 0.5726122982991714, 'r': 0.7607184241019699, 'f1': 0.6533963672555362}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Chinese: {'event': {'p': 0.4835164835164835, 'r': 0.8148148148148148, 'f1': 0.6068965517241379}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Korean: {'event': {'p': 0.5, 'r': 0.5238095238095238, 'f1': 0.5116279069767442}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Russian: {'event': {'p': 0.4107142857142857, 'r': 0.6388888888888888, 'f1': 0.5}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.6301369863013698, 'r': 0.8518518518518519, 'f1': 0.7244094488188976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.6956521739130435, 'r': 0.5079365079365079, 'f1': 0.5871559633027522}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5428571428571428, 'r': 0.5277777777777778, 'f1': 0.5352112676056338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 6 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:30:59.436167: step: 4/529, loss: 0.1470019519329071 2023-01-21 09:31:00.577862: step: 8/529, loss: 0.15354004502296448 2023-01-21 09:31:01.693659: step: 12/529, loss: 0.07719042152166367 2023-01-21 09:31:02.818214: step: 16/529, loss: 0.43562179803848267 2023-01-21 09:31:03.968079: step: 20/529, loss: 0.11794624477624893 2023-01-21 09:31:05.100414: step: 24/529, loss: 0.018527984619140625 2023-01-21 09:31:06.230358: step: 28/529, loss: 0.2061825692653656 2023-01-21 09:31:07.339840: step: 32/529, loss: 0.021318960934877396 2023-01-21 09:31:08.459665: step: 36/529, loss: 0.05560178682208061 2023-01-21 09:31:09.619905: step: 40/529, loss: 0.3228421211242676 2023-01-21 09:31:10.761352: step: 44/529, loss: 0.2062358856201172 2023-01-21 09:31:11.913944: step: 48/529, loss: 0.06267385929822922 2023-01-21 09:31:13.090147: step: 52/529, loss: 0.21166104078292847 2023-01-21 09:31:14.191308: step: 56/529, loss: 0.07229993492364883 2023-01-21 09:31:15.335926: step: 60/529, loss: 0.06668376922607422 2023-01-21 09:31:16.472494: step: 64/529, loss: 0.15439891815185547 2023-01-21 09:31:17.598647: step: 68/529, loss: 0.5146299600601196 2023-01-21 09:31:18.709839: step: 72/529, loss: 0.085711769759655 2023-01-21 09:31:19.828246: step: 76/529, loss: 0.3377135694026947 2023-01-21 09:31:21.006711: step: 80/529, loss: 0.4217444360256195 2023-01-21 09:31:22.159657: step: 84/529, loss: 0.17859764397144318 2023-01-21 09:31:23.258452: step: 88/529, loss: 0.03776540607213974 2023-01-21 09:31:24.413603: step: 92/529, loss: 0.3045387268066406 2023-01-21 09:31:25.552816: step: 96/529, loss: 0.11409024894237518 2023-01-21 09:31:26.686501: step: 100/529, loss: 0.234690859913826 2023-01-21 09:31:27.836138: step: 104/529, loss: 0.19339197874069214 2023-01-21 09:31:28.995154: step: 108/529, loss: 0.3174059987068176 2023-01-21 09:31:30.122331: step: 112/529, loss: 0.03271510452032089 2023-01-21 09:31:31.287009: step: 116/529, loss: 0.12669047713279724 2023-01-21 09:31:32.453201: step: 120/529, loss: 0.10272550582885742 2023-01-21 09:31:33.580761: step: 124/529, loss: 0.5136853456497192 2023-01-21 09:31:34.719655: step: 128/529, loss: 0.2828788757324219 2023-01-21 09:31:35.858095: step: 132/529, loss: 0.02475113235414028 2023-01-21 09:31:36.990570: step: 136/529, loss: 0.01852746121585369 2023-01-21 09:31:38.130552: step: 140/529, loss: 0.2424711287021637 2023-01-21 09:31:39.249777: step: 144/529, loss: 0.19414682686328888 2023-01-21 09:31:40.400939: step: 148/529, loss: 0.12833824753761292 2023-01-21 09:31:41.529220: step: 152/529, loss: 0.028827380388975143 2023-01-21 09:31:42.653995: step: 156/529, loss: 0.20758502185344696 2023-01-21 09:31:43.777119: step: 160/529, loss: 0.06466741859912872 2023-01-21 09:31:44.905862: step: 164/529, loss: 0.041637469083070755 2023-01-21 09:31:46.035388: step: 168/529, loss: 0.12986308336257935 2023-01-21 09:31:47.152664: step: 172/529, loss: 0.6346961855888367 2023-01-21 09:31:48.290687: step: 176/529, loss: 0.15828584134578705 2023-01-21 09:31:49.417223: step: 180/529, loss: 0.48843392729759216 2023-01-21 09:31:50.538158: step: 184/529, loss: 0.47235769033432007 2023-01-21 09:31:51.681110: step: 188/529, loss: 0.0854257270693779 2023-01-21 09:31:52.808758: step: 192/529, loss: 0.44334205985069275 2023-01-21 09:31:53.932041: step: 196/529, loss: 0.0744076743721962 2023-01-21 09:31:55.061020: step: 200/529, loss: 0.04431162029504776 2023-01-21 09:31:56.223273: step: 204/529, loss: 0.31575697660446167 2023-01-21 09:31:57.412197: step: 208/529, loss: 0.1267659217119217 2023-01-21 09:31:58.530777: step: 212/529, loss: 0.22790119051933289 2023-01-21 09:31:59.659681: step: 216/529, loss: 0.027790643274784088 2023-01-21 09:32:00.792311: step: 220/529, loss: 0.34567898511886597 2023-01-21 09:32:01.916367: step: 224/529, loss: 0.35775792598724365 2023-01-21 09:32:03.038547: step: 228/529, loss: 0.17790432274341583 2023-01-21 09:32:04.205371: step: 232/529, loss: 0.1421743929386139 2023-01-21 09:32:05.356927: step: 236/529, loss: 0.11125729978084564 2023-01-21 09:32:06.486130: step: 240/529, loss: 0.23817892372608185 2023-01-21 09:32:07.645657: step: 244/529, loss: 1.199149250984192 2023-01-21 09:32:08.776194: step: 248/529, loss: 0.1931035965681076 2023-01-21 09:32:09.903611: step: 252/529, loss: 0.029933644458651543 2023-01-21 09:32:11.023569: step: 256/529, loss: 0.1190594732761383 2023-01-21 09:32:12.151655: step: 260/529, loss: 0.17947152256965637 2023-01-21 09:32:13.284528: step: 264/529, loss: 0.08676370978355408 2023-01-21 09:32:14.424022: step: 268/529, loss: 0.05386481434106827 2023-01-21 09:32:15.597039: step: 272/529, loss: 0.3328613340854645 2023-01-21 09:32:16.739141: step: 276/529, loss: 0.17038804292678833 2023-01-21 09:32:17.867597: step: 280/529, loss: 0.16274502873420715 2023-01-21 09:32:19.032478: step: 284/529, loss: 0.20642271637916565 2023-01-21 09:32:20.133603: step: 288/529, loss: 0.13507691025733948 2023-01-21 09:32:21.265713: step: 292/529, loss: 0.08156852424144745 2023-01-21 09:32:22.377796: step: 296/529, loss: 0.0426325798034668 2023-01-21 09:32:23.498588: step: 300/529, loss: 0.051397159695625305 2023-01-21 09:32:24.658488: step: 304/529, loss: 0.21066169440746307 2023-01-21 09:32:25.812091: step: 308/529, loss: 0.03730187565088272 2023-01-21 09:32:26.956511: step: 312/529, loss: 0.10313472896814346 2023-01-21 09:32:28.083293: step: 316/529, loss: 0.07313623279333115 2023-01-21 09:32:29.187415: step: 320/529, loss: 0.085165835916996 2023-01-21 09:32:30.373722: step: 324/529, loss: 0.10756182670593262 2023-01-21 09:32:31.532465: step: 328/529, loss: 0.09438561648130417 2023-01-21 09:32:32.638459: step: 332/529, loss: 0.038346484303474426 2023-01-21 09:32:33.758147: step: 336/529, loss: 0.02523212507367134 2023-01-21 09:32:34.878269: step: 340/529, loss: 0.11392489075660706 2023-01-21 09:32:36.015081: step: 344/529, loss: 0.08156931400299072 2023-01-21 09:32:37.181035: step: 348/529, loss: 0.16472864151000977 2023-01-21 09:32:38.330185: step: 352/529, loss: 0.20975056290626526 2023-01-21 09:32:39.473690: step: 356/529, loss: 0.042269516736269 2023-01-21 09:32:40.620619: step: 360/529, loss: 0.2373577058315277 2023-01-21 09:32:41.801517: step: 364/529, loss: 0.07121773064136505 2023-01-21 09:32:42.943375: step: 368/529, loss: 0.07017125934362411 2023-01-21 09:32:44.130456: step: 372/529, loss: 0.21531495451927185 2023-01-21 09:32:45.253445: step: 376/529, loss: 0.04931068792939186 2023-01-21 09:32:46.398188: step: 380/529, loss: 0.08102398365736008 2023-01-21 09:32:47.547059: step: 384/529, loss: 0.11685208976268768 2023-01-21 09:32:48.692206: step: 388/529, loss: 0.03999481350183487 2023-01-21 09:32:49.858083: step: 392/529, loss: 0.08942137658596039 2023-01-21 09:32:50.989692: step: 396/529, loss: 0.06027255207300186 2023-01-21 09:32:52.126975: step: 400/529, loss: 0.3250984847545624 2023-01-21 09:32:53.291683: step: 404/529, loss: 0.4859161376953125 2023-01-21 09:32:54.455908: step: 408/529, loss: 0.21767020225524902 2023-01-21 09:32:55.569731: step: 412/529, loss: 0.02526264265179634 2023-01-21 09:32:56.660359: step: 416/529, loss: 0.17558957636356354 2023-01-21 09:32:57.779846: step: 420/529, loss: 0.1540956050157547 2023-01-21 09:32:58.925795: step: 424/529, loss: 0.24311915040016174 2023-01-21 09:33:00.065987: step: 428/529, loss: 0.055359553545713425 2023-01-21 09:33:01.205796: step: 432/529, loss: 0.08109503239393234 2023-01-21 09:33:02.363226: step: 436/529, loss: 0.7565504312515259 2023-01-21 09:33:03.512002: step: 440/529, loss: 0.9818193316459656 2023-01-21 09:33:04.635287: step: 444/529, loss: 0.49908995628356934 2023-01-21 09:33:05.764471: step: 448/529, loss: 0.13468456268310547 2023-01-21 09:33:06.903314: step: 452/529, loss: 0.09451542049646378 2023-01-21 09:33:08.022918: step: 456/529, loss: 0.05398140847682953 2023-01-21 09:33:09.159165: step: 460/529, loss: 0.062566377222538 2023-01-21 09:33:10.312152: step: 464/529, loss: 0.13144922256469727 2023-01-21 09:33:11.468931: step: 468/529, loss: 0.25995445251464844 2023-01-21 09:33:12.605504: step: 472/529, loss: 0.0994727611541748 2023-01-21 09:33:13.735605: step: 476/529, loss: 0.1284542977809906 2023-01-21 09:33:14.876089: step: 480/529, loss: 0.15237684547901154 2023-01-21 09:33:15.989351: step: 484/529, loss: 0.13454779982566833 2023-01-21 09:33:17.103739: step: 488/529, loss: 0.49535244703292847 2023-01-21 09:33:18.235452: step: 492/529, loss: 0.1663494110107422 2023-01-21 09:33:19.385257: step: 496/529, loss: 0.4106990098953247 2023-01-21 09:33:20.523500: step: 500/529, loss: 0.12006096541881561 2023-01-21 09:33:21.637930: step: 504/529, loss: 0.03542385250329971 2023-01-21 09:33:22.774955: step: 508/529, loss: 0.0414612777531147 2023-01-21 09:33:23.921453: step: 512/529, loss: 0.5257449150085449 2023-01-21 09:33:25.047721: step: 516/529, loss: 0.21701198816299438 2023-01-21 09:33:26.186071: step: 520/529, loss: 0.12552958726882935 2023-01-21 09:33:27.323106: step: 524/529, loss: 0.1172323226928711 2023-01-21 09:33:28.462321: step: 528/529, loss: 2.761033535003662 2023-01-21 09:33:29.603157: step: 532/529, loss: 0.27108487486839294 2023-01-21 09:33:30.719258: step: 536/529, loss: 0.17586469650268555 2023-01-21 09:33:31.865396: step: 540/529, loss: 0.043247416615486145 2023-01-21 09:33:32.979915: step: 544/529, loss: 0.018515586853027344 2023-01-21 09:33:34.105553: step: 548/529, loss: 0.02021508291363716 2023-01-21 09:33:35.243815: step: 552/529, loss: 0.0851355567574501 2023-01-21 09:33:36.370797: step: 556/529, loss: 0.17973260581493378 2023-01-21 09:33:37.498874: step: 560/529, loss: 0.4213048815727234 2023-01-21 09:33:38.676435: step: 564/529, loss: 0.17982031404972076 2023-01-21 09:33:39.816266: step: 568/529, loss: 0.12651550769805908 2023-01-21 09:33:40.951952: step: 572/529, loss: 0.09167223423719406 2023-01-21 09:33:42.115277: step: 576/529, loss: 0.08841677010059357 2023-01-21 09:33:43.245014: step: 580/529, loss: 0.18619662523269653 2023-01-21 09:33:44.391309: step: 584/529, loss: 0.5100477933883667 2023-01-21 09:33:45.510175: step: 588/529, loss: 0.09792108088731766 2023-01-21 09:33:46.673963: step: 592/529, loss: 0.09741516411304474 2023-01-21 09:33:47.815264: step: 596/529, loss: 0.015562344342470169 2023-01-21 09:33:48.955823: step: 600/529, loss: 0.15650692582130432 2023-01-21 09:33:50.127702: step: 604/529, loss: 0.19272108376026154 2023-01-21 09:33:51.255938: step: 608/529, loss: 1.4801222085952759 2023-01-21 09:33:52.406590: step: 612/529, loss: 0.263576865196228 2023-01-21 09:33:53.542877: step: 616/529, loss: 0.005211162380874157 2023-01-21 09:33:54.685725: step: 620/529, loss: 0.07423782348632812 2023-01-21 09:33:55.833482: step: 624/529, loss: 0.04071540758013725 2023-01-21 09:33:56.984386: step: 628/529, loss: 0.051209352910518646 2023-01-21 09:33:58.124094: step: 632/529, loss: 0.013059521093964577 2023-01-21 09:33:59.269278: step: 636/529, loss: 0.2136402130126953 2023-01-21 09:34:00.417985: step: 640/529, loss: 0.08834586292505264 2023-01-21 09:34:01.535504: step: 644/529, loss: 0.7316094040870667 2023-01-21 09:34:02.654638: step: 648/529, loss: 0.1162010207772255 2023-01-21 09:34:03.788127: step: 652/529, loss: 0.16994495689868927 2023-01-21 09:34:04.909520: step: 656/529, loss: 0.13997697830200195 2023-01-21 09:34:06.032937: step: 660/529, loss: 0.030808638781309128 2023-01-21 09:34:07.183339: step: 664/529, loss: 0.04116683080792427 2023-01-21 09:34:08.313164: step: 668/529, loss: 0.03182578086853027 2023-01-21 09:34:09.431280: step: 672/529, loss: 0.08994419872760773 2023-01-21 09:34:10.594056: step: 676/529, loss: 0.11245398223400116 2023-01-21 09:34:11.724838: step: 680/529, loss: 0.5464494824409485 2023-01-21 09:34:12.872852: step: 684/529, loss: 0.051720716059207916 2023-01-21 09:34:13.995582: step: 688/529, loss: 0.2325720489025116 2023-01-21 09:34:15.114175: step: 692/529, loss: 0.2228042036294937 2023-01-21 09:34:16.248216: step: 696/529, loss: 0.043380022048950195 2023-01-21 09:34:17.364842: step: 700/529, loss: 0.09498069435358047 2023-01-21 09:34:18.523364: step: 704/529, loss: 0.13558922708034515 2023-01-21 09:34:19.655243: step: 708/529, loss: 0.07393055409193039 2023-01-21 09:34:20.811260: step: 712/529, loss: 0.06666383892297745 2023-01-21 09:34:21.952461: step: 716/529, loss: 0.16336464881896973 2023-01-21 09:34:23.089505: step: 720/529, loss: 0.10356657207012177 2023-01-21 09:34:24.225886: step: 724/529, loss: 0.06547460705041885 2023-01-21 09:34:25.359879: step: 728/529, loss: 0.13290682435035706 2023-01-21 09:34:26.483322: step: 732/529, loss: 0.09052858501672745 2023-01-21 09:34:27.612419: step: 736/529, loss: 0.396187424659729 2023-01-21 09:34:28.757859: step: 740/529, loss: 0.6283569931983948 2023-01-21 09:34:29.900499: step: 744/529, loss: 0.009532547555863857 2023-01-21 09:34:31.071689: step: 748/529, loss: 0.42701712250709534 2023-01-21 09:34:32.190942: step: 752/529, loss: 0.032381534576416016 2023-01-21 09:34:33.344660: step: 756/529, loss: 0.023659182712435722 2023-01-21 09:34:34.486456: step: 760/529, loss: 0.053922273218631744 2023-01-21 09:34:35.666300: step: 764/529, loss: 0.7633985280990601 2023-01-21 09:34:36.810216: step: 768/529, loss: 0.05611744523048401 2023-01-21 09:34:37.977349: step: 772/529, loss: 0.12910637259483337 2023-01-21 09:34:39.115287: step: 776/529, loss: 0.1542777568101883 2023-01-21 09:34:40.263629: step: 780/529, loss: 0.1200251579284668 2023-01-21 09:34:41.393133: step: 784/529, loss: 0.04896056652069092 2023-01-21 09:34:42.544834: step: 788/529, loss: 0.008860170841217041 2023-01-21 09:34:43.699947: step: 792/529, loss: 0.14046001434326172 2023-01-21 09:34:44.823785: step: 796/529, loss: 0.24559718370437622 2023-01-21 09:34:45.960265: step: 800/529, loss: 0.4521062672138214 2023-01-21 09:34:47.089925: step: 804/529, loss: 0.16702929139137268 2023-01-21 09:34:48.238926: step: 808/529, loss: 0.052259065210819244 2023-01-21 09:34:49.369050: step: 812/529, loss: 3.2721288204193115 2023-01-21 09:34:50.519144: step: 816/529, loss: 0.1798514425754547 2023-01-21 09:34:51.642166: step: 820/529, loss: 0.15131349861621857 2023-01-21 09:34:52.764830: step: 824/529, loss: 0.4318869709968567 2023-01-21 09:34:53.931236: step: 828/529, loss: 0.02573556825518608 2023-01-21 09:34:55.074782: step: 832/529, loss: 0.04328899458050728 2023-01-21 09:34:56.223674: step: 836/529, loss: 0.1388297975063324 2023-01-21 09:34:57.401343: step: 840/529, loss: 0.12231426686048508 2023-01-21 09:34:58.547289: step: 844/529, loss: 0.08716468513011932 2023-01-21 09:34:59.699699: step: 848/529, loss: 0.12824907898902893 2023-01-21 09:35:00.823293: step: 852/529, loss: 0.09431162476539612 2023-01-21 09:35:01.938942: step: 856/529, loss: 0.058193303644657135 2023-01-21 09:35:03.129925: step: 860/529, loss: 0.11723414063453674 2023-01-21 09:35:04.281959: step: 864/529, loss: 0.10897369682788849 2023-01-21 09:35:05.416878: step: 868/529, loss: 0.11294145882129669 2023-01-21 09:35:06.535785: step: 872/529, loss: 0.06196422502398491 2023-01-21 09:35:07.687270: step: 876/529, loss: 0.06574240326881409 2023-01-21 09:35:08.833763: step: 880/529, loss: 0.0637269988656044 2023-01-21 09:35:09.963631: step: 884/529, loss: 0.19259533286094666 2023-01-21 09:35:11.099579: step: 888/529, loss: 0.16550368070602417 2023-01-21 09:35:12.221233: step: 892/529, loss: 0.13171425461769104 2023-01-21 09:35:13.367383: step: 896/529, loss: 0.005716037470847368 2023-01-21 09:35:14.521058: step: 900/529, loss: 0.015420913696289062 2023-01-21 09:35:15.643643: step: 904/529, loss: 0.04086365923285484 2023-01-21 09:35:16.779885: step: 908/529, loss: 0.12154017388820648 2023-01-21 09:35:17.914150: step: 912/529, loss: 0.14820271730422974 2023-01-21 09:35:19.039284: step: 916/529, loss: 0.03450622782111168 2023-01-21 09:35:20.168065: step: 920/529, loss: 0.12771210074424744 2023-01-21 09:35:21.294425: step: 924/529, loss: 0.06066742166876793 2023-01-21 09:35:22.437613: step: 928/529, loss: 0.03146343305706978 2023-01-21 09:35:23.564186: step: 932/529, loss: 0.22619009017944336 2023-01-21 09:35:24.687430: step: 936/529, loss: 0.10137005150318146 2023-01-21 09:35:25.828614: step: 940/529, loss: 0.03821449354290962 2023-01-21 09:35:26.944316: step: 944/529, loss: 0.10627007484436035 2023-01-21 09:35:28.051745: step: 948/529, loss: 0.033090949058532715 2023-01-21 09:35:29.168906: step: 952/529, loss: 0.4556516706943512 2023-01-21 09:35:30.298024: step: 956/529, loss: 0.07339987903833389 2023-01-21 09:35:31.478443: step: 960/529, loss: 0.05294237285852432 2023-01-21 09:35:32.606999: step: 964/529, loss: 0.583365797996521 2023-01-21 09:35:33.772261: step: 968/529, loss: 0.03099527209997177 2023-01-21 09:35:34.929047: step: 972/529, loss: 1.66312837600708 2023-01-21 09:35:36.065238: step: 976/529, loss: 0.17345017194747925 2023-01-21 09:35:37.193881: step: 980/529, loss: 0.1968173086643219 2023-01-21 09:35:38.404690: step: 984/529, loss: 0.6370856761932373 2023-01-21 09:35:39.525056: step: 988/529, loss: 0.08437786251306534 2023-01-21 09:35:40.694237: step: 992/529, loss: 0.02574920654296875 2023-01-21 09:35:41.871396: step: 996/529, loss: 0.07056809216737747 2023-01-21 09:35:43.006514: step: 1000/529, loss: 0.2804013192653656 2023-01-21 09:35:44.157908: step: 1004/529, loss: 0.09528064727783203 2023-01-21 09:35:45.283790: step: 1008/529, loss: 0.5438389182090759 2023-01-21 09:35:46.411361: step: 1012/529, loss: 0.12470737099647522 2023-01-21 09:35:47.548380: step: 1016/529, loss: 0.45941734313964844 2023-01-21 09:35:48.687586: step: 1020/529, loss: 0.340905100107193 2023-01-21 09:35:49.823140: step: 1024/529, loss: 0.5324832201004028 2023-01-21 09:35:50.941686: step: 1028/529, loss: 0.0022804916370660067 2023-01-21 09:35:52.094221: step: 1032/529, loss: 0.023268794640898705 2023-01-21 09:35:53.231083: step: 1036/529, loss: 0.20383435487747192 2023-01-21 09:35:54.369459: step: 1040/529, loss: 0.17854785919189453 2023-01-21 09:35:55.520160: step: 1044/529, loss: 0.12450919300317764 2023-01-21 09:35:56.644303: step: 1048/529, loss: 0.1918668895959854 2023-01-21 09:35:57.783793: step: 1052/529, loss: 0.032587431371212006 2023-01-21 09:35:58.929314: step: 1056/529, loss: 0.07682953029870987 2023-01-21 09:36:00.070264: step: 1060/529, loss: 0.7495388984680176 2023-01-21 09:36:01.201111: step: 1064/529, loss: 0.11987963318824768 2023-01-21 09:36:02.340523: step: 1068/529, loss: 0.22625942528247833 2023-01-21 09:36:03.479106: step: 1072/529, loss: 0.021382426843047142 2023-01-21 09:36:04.623455: step: 1076/529, loss: 0.12423534691333771 2023-01-21 09:36:05.731653: step: 1080/529, loss: 0.1948629915714264 2023-01-21 09:36:06.911875: step: 1084/529, loss: 0.1633104383945465 2023-01-21 09:36:08.043896: step: 1088/529, loss: 0.1142793670296669 2023-01-21 09:36:09.137683: step: 1092/529, loss: 0.10774402320384979 2023-01-21 09:36:10.273576: step: 1096/529, loss: 0.13552847504615784 2023-01-21 09:36:11.414501: step: 1100/529, loss: 0.04751596599817276 2023-01-21 09:36:12.534484: step: 1104/529, loss: 0.04301629215478897 2023-01-21 09:36:13.664228: step: 1108/529, loss: 0.2184770554304123 2023-01-21 09:36:14.830827: step: 1112/529, loss: 0.295999139547348 2023-01-21 09:36:15.979252: step: 1116/529, loss: 0.06016087904572487 2023-01-21 09:36:17.136475: step: 1120/529, loss: 0.09768819808959961 2023-01-21 09:36:18.274161: step: 1124/529, loss: 0.13039502501487732 2023-01-21 09:36:19.399863: step: 1128/529, loss: 0.05840778350830078 2023-01-21 09:36:20.538921: step: 1132/529, loss: 0.20190268754959106 2023-01-21 09:36:21.703038: step: 1136/529, loss: 0.08311748504638672 2023-01-21 09:36:22.840374: step: 1140/529, loss: 0.2716061770915985 2023-01-21 09:36:23.995512: step: 1144/529, loss: 0.03608150780200958 2023-01-21 09:36:25.148648: step: 1148/529, loss: 0.12211474031209946 2023-01-21 09:36:26.294860: step: 1152/529, loss: 0.21685010194778442 2023-01-21 09:36:27.414897: step: 1156/529, loss: 0.10879373550415039 2023-01-21 09:36:28.561040: step: 1160/529, loss: 0.1112680435180664 2023-01-21 09:36:29.699843: step: 1164/529, loss: 0.05439148098230362 2023-01-21 09:36:30.836501: step: 1168/529, loss: 0.8142209053039551 2023-01-21 09:36:31.955242: step: 1172/529, loss: 0.042769573628902435 2023-01-21 09:36:33.100653: step: 1176/529, loss: 0.11595716327428818 2023-01-21 09:36:34.250877: step: 1180/529, loss: 0.05971822887659073 2023-01-21 09:36:35.417725: step: 1184/529, loss: 0.17592830955982208 2023-01-21 09:36:36.567443: step: 1188/529, loss: 0.159053236246109 2023-01-21 09:36:37.687850: step: 1192/529, loss: 0.009970713406801224 2023-01-21 09:36:38.862394: step: 1196/529, loss: 0.2734020948410034 2023-01-21 09:36:40.009269: step: 1200/529, loss: 0.41264763474464417 2023-01-21 09:36:41.133833: step: 1204/529, loss: 0.2158103883266449 2023-01-21 09:36:42.276990: step: 1208/529, loss: 0.1070350706577301 2023-01-21 09:36:43.399595: step: 1212/529, loss: 0.09604816883802414 2023-01-21 09:36:44.546102: step: 1216/529, loss: 0.13412058353424072 2023-01-21 09:36:45.680930: step: 1220/529, loss: 0.13900576531887054 2023-01-21 09:36:46.825850: step: 1224/529, loss: 0.17343321442604065 2023-01-21 09:36:47.946315: step: 1228/529, loss: 0.11591120064258575 2023-01-21 09:36:49.064038: step: 1232/529, loss: 1.3875800371170044 2023-01-21 09:36:50.192848: step: 1236/529, loss: 0.3283710479736328 2023-01-21 09:36:51.316285: step: 1240/529, loss: 0.25912365317344666 2023-01-21 09:36:52.440431: step: 1244/529, loss: 0.1069006696343422 2023-01-21 09:36:53.583641: step: 1248/529, loss: 0.52838134765625 2023-01-21 09:36:54.719896: step: 1252/529, loss: 0.1586301177740097 2023-01-21 09:36:55.829781: step: 1256/529, loss: 0.08219738304615021 2023-01-21 09:36:56.974119: step: 1260/529, loss: 0.20823365449905396 2023-01-21 09:36:58.121476: step: 1264/529, loss: 0.13486385345458984 2023-01-21 09:36:59.239320: step: 1268/529, loss: 0.5196982622146606 2023-01-21 09:37:00.401315: step: 1272/529, loss: 0.10855884850025177 2023-01-21 09:37:01.538444: step: 1276/529, loss: 1.1849353313446045 2023-01-21 09:37:02.680144: step: 1280/529, loss: 0.03606891632080078 2023-01-21 09:37:03.815553: step: 1284/529, loss: 0.30629798769950867 2023-01-21 09:37:04.942878: step: 1288/529, loss: 0.16889868676662445 2023-01-21 09:37:06.100502: step: 1292/529, loss: 0.14534083008766174 2023-01-21 09:37:07.227868: step: 1296/529, loss: 0.0436803363263607 2023-01-21 09:37:08.354517: step: 1300/529, loss: 0.05636334419250488 2023-01-21 09:37:09.504222: step: 1304/529, loss: 0.09943743050098419 2023-01-21 09:37:10.639541: step: 1308/529, loss: 0.1236240416765213 2023-01-21 09:37:11.771761: step: 1312/529, loss: 0.7554237246513367 2023-01-21 09:37:12.898203: step: 1316/529, loss: 0.15680323541164398 2023-01-21 09:37:14.058481: step: 1320/529, loss: 0.3036162555217743 2023-01-21 09:37:15.198767: step: 1324/529, loss: 0.13943703472614288 2023-01-21 09:37:16.336040: step: 1328/529, loss: 0.010088921524584293 2023-01-21 09:37:17.475784: step: 1332/529, loss: 0.2421044409275055 2023-01-21 09:37:18.645940: step: 1336/529, loss: 0.022925235331058502 2023-01-21 09:37:19.775941: step: 1340/529, loss: 0.1155308336019516 2023-01-21 09:37:20.945982: step: 1344/529, loss: 0.002062034560367465 2023-01-21 09:37:22.068849: step: 1348/529, loss: 0.17157764732837677 2023-01-21 09:37:23.212308: step: 1352/529, loss: 0.15530209243297577 2023-01-21 09:37:24.347374: step: 1356/529, loss: 0.09535965323448181 2023-01-21 09:37:25.505414: step: 1360/529, loss: 0.0919216126203537 2023-01-21 09:37:26.665132: step: 1364/529, loss: 0.13834786415100098 2023-01-21 09:37:27.820096: step: 1368/529, loss: 0.3966245651245117 2023-01-21 09:37:28.943818: step: 1372/529, loss: 0.2716043293476105 2023-01-21 09:37:30.085877: step: 1376/529, loss: 0.05550813674926758 2023-01-21 09:37:31.214635: step: 1380/529, loss: 0.079542376101017 2023-01-21 09:37:32.347363: step: 1384/529, loss: 0.10487575829029083 2023-01-21 09:37:33.471662: step: 1388/529, loss: 0.20975279808044434 2023-01-21 09:37:34.622749: step: 1392/529, loss: 0.07599058747291565 2023-01-21 09:37:35.739621: step: 1396/529, loss: 0.05106344074010849 2023-01-21 09:37:36.912009: step: 1400/529, loss: 0.5079331994056702 2023-01-21 09:37:38.043695: step: 1404/529, loss: 0.07319359481334686 2023-01-21 09:37:39.169212: step: 1408/529, loss: 0.16860371828079224 2023-01-21 09:37:40.323432: step: 1412/529, loss: 0.14180158078670502 2023-01-21 09:37:41.447295: step: 1416/529, loss: 0.11361794173717499 2023-01-21 09:37:42.614911: step: 1420/529, loss: 0.0701180025935173 2023-01-21 09:37:43.754012: step: 1424/529, loss: 0.0807436928153038 2023-01-21 09:37:44.890623: step: 1428/529, loss: 0.12773476541042328 2023-01-21 09:37:46.014155: step: 1432/529, loss: 0.2757173478603363 2023-01-21 09:37:47.139275: step: 1436/529, loss: 0.03978696092963219 2023-01-21 09:37:48.270441: step: 1440/529, loss: 0.126498281955719 2023-01-21 09:37:49.422741: step: 1444/529, loss: 0.3003924489021301 2023-01-21 09:37:50.548054: step: 1448/529, loss: 0.14830370247364044 2023-01-21 09:37:51.684123: step: 1452/529, loss: 0.17213058471679688 2023-01-21 09:37:52.852598: step: 1456/529, loss: 0.16140413284301758 2023-01-21 09:37:53.987617: step: 1460/529, loss: 0.0833822712302208 2023-01-21 09:37:55.123474: step: 1464/529, loss: 0.4955655038356781 2023-01-21 09:37:56.247485: step: 1468/529, loss: 0.0634758472442627 2023-01-21 09:37:57.404260: step: 1472/529, loss: 0.4631425142288208 2023-01-21 09:37:58.528566: step: 1476/529, loss: 0.3466726243495941 2023-01-21 09:37:59.648833: step: 1480/529, loss: 0.1507301926612854 2023-01-21 09:38:00.785847: step: 1484/529, loss: 0.11795787513256073 2023-01-21 09:38:01.910037: step: 1488/529, loss: 0.13609285652637482 2023-01-21 09:38:03.041095: step: 1492/529, loss: 0.1769440621137619 2023-01-21 09:38:04.147690: step: 1496/529, loss: 0.0699739009141922 2023-01-21 09:38:05.283437: step: 1500/529, loss: 0.3907044231891632 2023-01-21 09:38:06.416419: step: 1504/529, loss: 0.24156685173511505 2023-01-21 09:38:07.572691: step: 1508/529, loss: 0.14699365198612213 2023-01-21 09:38:08.704101: step: 1512/529, loss: 0.38158854842185974 2023-01-21 09:38:09.825515: step: 1516/529, loss: 0.09126892685890198 2023-01-21 09:38:10.945909: step: 1520/529, loss: 0.4601791501045227 2023-01-21 09:38:12.115009: step: 1524/529, loss: 0.12492237240076065 2023-01-21 09:38:13.227577: step: 1528/529, loss: 0.04337163269519806 2023-01-21 09:38:14.362304: step: 1532/529, loss: 0.20987573266029358 2023-01-21 09:38:15.503560: step: 1536/529, loss: 0.09402570873498917 2023-01-21 09:38:16.654934: step: 1540/529, loss: 0.033980369567871094 2023-01-21 09:38:17.781334: step: 1544/529, loss: 0.6974887847900391 2023-01-21 09:38:18.909734: step: 1548/529, loss: 0.35368672013282776 2023-01-21 09:38:20.036793: step: 1552/529, loss: 0.09562735259532928 2023-01-21 09:38:21.167775: step: 1556/529, loss: 0.42046844959259033 2023-01-21 09:38:22.318969: step: 1560/529, loss: 0.625869870185852 2023-01-21 09:38:23.454963: step: 1564/529, loss: 0.6839299201965332 2023-01-21 09:38:24.619579: step: 1568/529, loss: 0.2801129221916199 2023-01-21 09:38:25.759718: step: 1572/529, loss: 0.8438965082168579 2023-01-21 09:38:26.909898: step: 1576/529, loss: 0.12409190833568573 2023-01-21 09:38:28.047420: step: 1580/529, loss: 0.36157163977622986 2023-01-21 09:38:29.199866: step: 1584/529, loss: 0.0032471180893480778 2023-01-21 09:38:30.324240: step: 1588/529, loss: 0.0833933874964714 2023-01-21 09:38:31.473413: step: 1592/529, loss: 0.1490146666765213 2023-01-21 09:38:32.634853: step: 1596/529, loss: 0.2304130494594574 2023-01-21 09:38:33.778641: step: 1600/529, loss: 0.026635361835360527 2023-01-21 09:38:34.878768: step: 1604/529, loss: 0.11579003930091858 2023-01-21 09:38:36.010219: step: 1608/529, loss: 0.06304445117712021 2023-01-21 09:38:37.165756: step: 1612/529, loss: 0.9998999834060669 2023-01-21 09:38:38.296943: step: 1616/529, loss: 0.2079617977142334 2023-01-21 09:38:39.424698: step: 1620/529, loss: 0.3229769766330719 2023-01-21 09:38:40.560508: step: 1624/529, loss: 0.0373837985098362 2023-01-21 09:38:41.695571: step: 1628/529, loss: 0.056791115552186966 2023-01-21 09:38:42.825013: step: 1632/529, loss: 0.08234858512878418 2023-01-21 09:38:43.948805: step: 1636/529, loss: 0.22209934890270233 2023-01-21 09:38:45.097231: step: 1640/529, loss: 0.3499147295951843 2023-01-21 09:38:46.281755: step: 1644/529, loss: 0.018796682357788086 2023-01-21 09:38:47.402345: step: 1648/529, loss: 0.04752139747142792 2023-01-21 09:38:48.547689: step: 1652/529, loss: 0.3127335011959076 2023-01-21 09:38:49.712937: step: 1656/529, loss: 0.0968031957745552 2023-01-21 09:38:50.852261: step: 1660/529, loss: 0.08821725845336914 2023-01-21 09:38:51.965894: step: 1664/529, loss: 0.04978980869054794 2023-01-21 09:38:53.090714: step: 1668/529, loss: 0.42310333251953125 2023-01-21 09:38:54.229707: step: 1672/529, loss: 0.43186110258102417 2023-01-21 09:38:55.386149: step: 1676/529, loss: 0.10648288577795029 2023-01-21 09:38:56.510726: step: 1680/529, loss: 0.11094231903553009 2023-01-21 09:38:57.639794: step: 1684/529, loss: 0.05051689222455025 2023-01-21 09:38:58.780132: step: 1688/529, loss: 0.10724658519029617 2023-01-21 09:38:59.914467: step: 1692/529, loss: 0.6355093121528625 2023-01-21 09:39:01.040573: step: 1696/529, loss: 0.11709442734718323 2023-01-21 09:39:02.174018: step: 1700/529, loss: 0.08421077579259872 2023-01-21 09:39:03.280972: step: 1704/529, loss: 0.057763874530792236 2023-01-21 09:39:04.391608: step: 1708/529, loss: 0.17214146256446838 2023-01-21 09:39:05.527938: step: 1712/529, loss: 0.26311343908309937 2023-01-21 09:39:06.671228: step: 1716/529, loss: 0.5882684588432312 2023-01-21 09:39:07.816772: step: 1720/529, loss: 0.06975355744361877 2023-01-21 09:39:09.001379: step: 1724/529, loss: 0.2308824360370636 2023-01-21 09:39:10.134738: step: 1728/529, loss: 0.06496963649988174 2023-01-21 09:39:11.281329: step: 1732/529, loss: 0.25633737444877625 2023-01-21 09:39:12.418779: step: 1736/529, loss: 0.33822715282440186 2023-01-21 09:39:13.563778: step: 1740/529, loss: 0.2473660111427307 2023-01-21 09:39:14.706855: step: 1744/529, loss: 0.7095004320144653 2023-01-21 09:39:15.828320: step: 1748/529, loss: 0.18230445683002472 2023-01-21 09:39:16.961186: step: 1752/529, loss: 0.21284739673137665 2023-01-21 09:39:18.078870: step: 1756/529, loss: 0.8302550315856934 2023-01-21 09:39:19.208288: step: 1760/529, loss: 0.038268186151981354 2023-01-21 09:39:20.378712: step: 1764/529, loss: 0.05125870928168297 2023-01-21 09:39:21.520701: step: 1768/529, loss: 0.16092367470264435 2023-01-21 09:39:22.649326: step: 1772/529, loss: 0.08133144676685333 2023-01-21 09:39:23.783629: step: 1776/529, loss: 0.11688362061977386 2023-01-21 09:39:24.934072: step: 1780/529, loss: 0.45523911714553833 2023-01-21 09:39:26.075303: step: 1784/529, loss: 0.028124429285526276 2023-01-21 09:39:27.213914: step: 1788/529, loss: 0.06494168937206268 2023-01-21 09:39:28.317072: step: 1792/529, loss: 0.026015616953372955 2023-01-21 09:39:29.439491: step: 1796/529, loss: 0.09396862983703613 2023-01-21 09:39:30.572241: step: 1800/529, loss: 0.17881089448928833 2023-01-21 09:39:31.736377: step: 1804/529, loss: 0.04808454588055611 2023-01-21 09:39:32.881528: step: 1808/529, loss: 0.09634856879711151 2023-01-21 09:39:33.997043: step: 1812/529, loss: 0.03631029278039932 2023-01-21 09:39:35.138535: step: 1816/529, loss: 0.1206122413277626 2023-01-21 09:39:36.320921: step: 1820/529, loss: 0.24403706192970276 2023-01-21 09:39:37.454532: step: 1824/529, loss: 0.1272398978471756 2023-01-21 09:39:38.580184: step: 1828/529, loss: 0.08723344653844833 2023-01-21 09:39:39.696498: step: 1832/529, loss: 0.04036426544189453 2023-01-21 09:39:40.857880: step: 1836/529, loss: 0.04338674619793892 2023-01-21 09:39:41.993517: step: 1840/529, loss: 0.9547991752624512 2023-01-21 09:39:43.144139: step: 1844/529, loss: 0.06135311350226402 2023-01-21 09:39:44.280845: step: 1848/529, loss: 0.14369043707847595 2023-01-21 09:39:45.373711: step: 1852/529, loss: 0.015814971178770065 2023-01-21 09:39:46.512990: step: 1856/529, loss: 0.14163294434547424 2023-01-21 09:39:47.621189: step: 1860/529, loss: 0.020140409469604492 2023-01-21 09:39:48.747557: step: 1864/529, loss: 0.07844791561365128 2023-01-21 09:39:49.880861: step: 1868/529, loss: 0.10875293612480164 2023-01-21 09:39:50.999856: step: 1872/529, loss: 0.352674663066864 2023-01-21 09:39:52.111671: step: 1876/529, loss: 0.05489597097039223 2023-01-21 09:39:53.219301: step: 1880/529, loss: 0.19982747733592987 2023-01-21 09:39:54.361780: step: 1884/529, loss: 0.5023584961891174 2023-01-21 09:39:55.495783: step: 1888/529, loss: 0.1934628039598465 2023-01-21 09:39:56.631750: step: 1892/529, loss: 0.11241722106933594 2023-01-21 09:39:57.774686: step: 1896/529, loss: 0.12496557831764221 2023-01-21 09:39:58.906073: step: 1900/529, loss: 0.8799470663070679 2023-01-21 09:40:00.024328: step: 1904/529, loss: 0.14097614586353302 2023-01-21 09:40:01.171451: step: 1908/529, loss: 0.7329250574111938 2023-01-21 09:40:02.286783: step: 1912/529, loss: 0.04732117801904678 2023-01-21 09:40:03.431549: step: 1916/529, loss: 0.021816158667206764 2023-01-21 09:40:04.563365: step: 1920/529, loss: 0.08509254455566406 2023-01-21 09:40:05.710953: step: 1924/529, loss: 0.09246654808521271 2023-01-21 09:40:06.840752: step: 1928/529, loss: 0.159480482339859 2023-01-21 09:40:07.960698: step: 1932/529, loss: 0.8304230570793152 2023-01-21 09:40:09.089381: step: 1936/529, loss: 0.0987558364868164 2023-01-21 09:40:10.249603: step: 1940/529, loss: 0.480537474155426 2023-01-21 09:40:11.379754: step: 1944/529, loss: 0.05770282447338104 2023-01-21 09:40:12.509858: step: 1948/529, loss: 0.23431549966335297 2023-01-21 09:40:13.665652: step: 1952/529, loss: 0.08774594962596893 2023-01-21 09:40:14.802214: step: 1956/529, loss: 0.09121169149875641 2023-01-21 09:40:15.944822: step: 1960/529, loss: 0.38218921422958374 2023-01-21 09:40:17.097465: step: 1964/529, loss: 0.04716808721423149 2023-01-21 09:40:18.218663: step: 1968/529, loss: 0.11767492443323135 2023-01-21 09:40:19.371166: step: 1972/529, loss: 0.11477241665124893 2023-01-21 09:40:20.537241: step: 1976/529, loss: 1.535006046295166 2023-01-21 09:40:21.655411: step: 1980/529, loss: 0.07560023665428162 2023-01-21 09:40:22.825896: step: 1984/529, loss: 0.3149995803833008 2023-01-21 09:40:23.956713: step: 1988/529, loss: 0.6757254600524902 2023-01-21 09:40:25.090779: step: 1992/529, loss: 1.0960496664047241 2023-01-21 09:40:26.273513: step: 1996/529, loss: 0.13806819915771484 2023-01-21 09:40:27.425513: step: 2000/529, loss: 0.3999154269695282 2023-01-21 09:40:28.554952: step: 2004/529, loss: 0.05291027948260307 2023-01-21 09:40:29.697801: step: 2008/529, loss: 0.1192389503121376 2023-01-21 09:40:30.824482: step: 2012/529, loss: 0.2540108561515808 2023-01-21 09:40:31.956767: step: 2016/529, loss: 0.1288958042860031 2023-01-21 09:40:33.085858: step: 2020/529, loss: 0.04929985851049423 2023-01-21 09:40:34.200398: step: 2024/529, loss: 0.12469659000635147 2023-01-21 09:40:35.370669: step: 2028/529, loss: 1.9697210788726807 2023-01-21 09:40:36.515187: step: 2032/529, loss: 0.14989744126796722 2023-01-21 09:40:37.638800: step: 2036/529, loss: 0.11352329701185226 2023-01-21 09:40:38.797107: step: 2040/529, loss: 0.18717674911022186 2023-01-21 09:40:39.920631: step: 2044/529, loss: 0.023313332349061966 2023-01-21 09:40:41.106852: step: 2048/529, loss: 0.06607413291931152 2023-01-21 09:40:42.229514: step: 2052/529, loss: 0.4923926889896393 2023-01-21 09:40:43.383912: step: 2056/529, loss: 0.2689165771007538 2023-01-21 09:40:44.556267: step: 2060/529, loss: 0.1502152532339096 2023-01-21 09:40:45.679464: step: 2064/529, loss: 0.006889152340590954 2023-01-21 09:40:46.801091: step: 2068/529, loss: 0.6686629056930542 2023-01-21 09:40:47.970708: step: 2072/529, loss: 0.08644552528858185 2023-01-21 09:40:49.117214: step: 2076/529, loss: 0.0826151892542839 2023-01-21 09:40:50.261476: step: 2080/529, loss: 0.09907140582799911 2023-01-21 09:40:51.437098: step: 2084/529, loss: 0.10432948917150497 2023-01-21 09:40:52.568105: step: 2088/529, loss: 0.15374603867530823 2023-01-21 09:40:53.727290: step: 2092/529, loss: 0.04619045555591583 2023-01-21 09:40:54.855212: step: 2096/529, loss: 0.029952311888337135 2023-01-21 09:40:55.974472: step: 2100/529, loss: 0.29914793372154236 2023-01-21 09:40:57.093648: step: 2104/529, loss: 0.14015436172485352 2023-01-21 09:40:58.220923: step: 2108/529, loss: 0.0412413626909256 2023-01-21 09:40:59.371977: step: 2112/529, loss: 0.08781389892101288 2023-01-21 09:41:00.549584: step: 2116/529, loss: 0.06290626525878906 ================================================== Loss: 0.212 -------------------- Dev: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Test: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Chinese: {'event': {'p': 0.6024096385542169, 'r': 0.9259259259259259, 'f1': 0.7299270072992702}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} 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': 6} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6427718040621266, 'r': 0.7163781624500666, 'f1': 0.6775818639798489}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Chinese: {'event': {'p': 0.6065822784810126, 'r': 0.694090382387022, 'f1': 0.6473925965955147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Chinese: {'event': {'p': 0.6301369863013698, 'r': 0.8518518518518519, 'f1': 0.7244094488188976}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Korean: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.6956521739130435, 'r': 0.5079365079365079, 'f1': 0.5871559633027522}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 09:41:46.845051: step: 4/529, loss: 0.042563535273075104 2023-01-21 09:41:47.989755: step: 8/529, loss: 0.039869122207164764 2023-01-21 09:41:49.107778: step: 12/529, loss: 0.1457923948764801 2023-01-21 09:41:50.242851: step: 16/529, loss: 0.10531263798475266 2023-01-21 09:41:51.347699: step: 20/529, loss: 0.00659990357235074 2023-01-21 09:41:52.456747: step: 24/529, loss: 0.03553485870361328 2023-01-21 09:41:53.584755: step: 28/529, loss: 0.19160613417625427 2023-01-21 09:41:54.721525: step: 32/529, loss: 0.14029160141944885 2023-01-21 09:41:55.865529: step: 36/529, loss: 0.8193327188491821 2023-01-21 09:41:56.977362: step: 40/529, loss: 0.05508880689740181 2023-01-21 09:41:58.089191: step: 44/529, loss: 0.20330199599266052 2023-01-21 09:41:59.258564: step: 48/529, loss: 0.07096557319164276 2023-01-21 09:42:00.365733: step: 52/529, loss: 0.2691882252693176 2023-01-21 09:42:01.509752: step: 56/529, loss: 0.2167079895734787 2023-01-21 09:42:02.651466: step: 60/529, loss: 0.2393827736377716 2023-01-21 09:42:03.770461: step: 64/529, loss: 0.08929280936717987 2023-01-21 09:42:04.876324: step: 68/529, loss: 0.24706554412841797 2023-01-21 09:42:06.008410: step: 72/529, loss: 0.04640159755945206 2023-01-21 09:42:07.129142: step: 76/529, loss: 0.15074601769447327 2023-01-21 09:42:08.244455: step: 80/529, loss: 0.13661418855190277 2023-01-21 09:42:09.366746: step: 84/529, loss: 0.23108814656734467 2023-01-21 09:42:10.509557: step: 88/529, loss: 0.053687382489442825 2023-01-21 09:42:11.646132: step: 92/529, loss: 0.08737733960151672 2023-01-21 09:42:12.800625: step: 96/529, loss: 0.07151909172534943 2023-01-21 09:42:13.925815: step: 100/529, loss: 0.052175235003232956 2023-01-21 09:42:15.027283: step: 104/529, loss: 0.1434159278869629 2023-01-21 09:42:16.168032: step: 108/529, loss: 0.06303472816944122 2023-01-21 09:42:17.317264: step: 112/529, loss: 0.08998637646436691 2023-01-21 09:42:18.444315: step: 116/529, loss: 0.08417949825525284 2023-01-21 09:42:19.580907: step: 120/529, loss: 0.020761489868164062 2023-01-21 09:42:20.717585: step: 124/529, loss: 0.016847802326083183 2023-01-21 09:42:21.838348: step: 128/529, loss: 0.07435612380504608 2023-01-21 09:42:22.992766: step: 132/529, loss: 0.28077831864356995 2023-01-21 09:42:24.112609: step: 136/529, loss: 1.048264503479004 2023-01-21 09:42:25.231586: step: 140/529, loss: 0.06646700203418732 2023-01-21 09:42:26.363419: step: 144/529, loss: 0.07954072952270508 2023-01-21 09:42:27.524740: step: 148/529, loss: 0.11353912204504013 2023-01-21 09:42:28.655379: step: 152/529, loss: 0.1407584249973297 2023-01-21 09:42:29.794489: step: 156/529, loss: 0.0325779914855957 2023-01-21 09:42:30.905153: step: 160/529, loss: 0.1824987530708313 2023-01-21 09:42:32.055161: step: 164/529, loss: 0.027321243658661842 2023-01-21 09:42:33.190533: step: 168/529, loss: 0.13722367584705353 2023-01-21 09:42:34.346106: step: 172/529, loss: 0.047692012041807175 2023-01-21 09:42:35.476746: step: 176/529, loss: 0.24363203346729279 2023-01-21 09:42:36.606289: step: 180/529, loss: 0.01588892936706543 2023-01-21 09:42:37.741098: step: 184/529, loss: 0.12598839402198792 2023-01-21 09:42:38.888665: step: 188/529, loss: 0.21481962502002716 2023-01-21 09:42:40.036695: step: 192/529, loss: 0.7324357032775879 2023-01-21 09:42:41.173042: step: 196/529, loss: 0.07340145111083984 2023-01-21 09:42:42.308182: step: 200/529, loss: 0.15468549728393555 2023-01-21 09:42:43.471353: step: 204/529, loss: 0.025687791407108307 2023-01-21 09:42:44.610785: step: 208/529, loss: 0.058331869542598724 2023-01-21 09:42:45.747741: step: 212/529, loss: 0.1435280740261078 2023-01-21 09:42:46.888624: step: 216/529, loss: 0.15259627997875214 2023-01-21 09:42:48.006136: step: 220/529, loss: 0.34549087285995483 2023-01-21 09:42:49.134850: step: 224/529, loss: 0.11227608472108841 2023-01-21 09:42:50.267951: step: 228/529, loss: 0.14857101440429688 2023-01-21 09:42:51.399993: step: 232/529, loss: 0.008654308505356312 2023-01-21 09:42:52.496518: step: 236/529, loss: 0.12001033127307892 2023-01-21 09:42:53.616357: step: 240/529, loss: 0.03143613412976265 2023-01-21 09:42:54.743167: step: 244/529, loss: 0.40172329545021057 2023-01-21 09:42:55.972228: step: 248/529, loss: 0.14287033677101135 2023-01-21 09:42:57.098210: step: 252/529, loss: 0.02498483844101429 2023-01-21 09:42:58.243249: step: 256/529, loss: 0.11949684470891953 2023-01-21 09:42:59.350501: step: 260/529, loss: 0.1006307601928711 2023-01-21 09:43:00.485546: step: 264/529, loss: 0.041253700852394104 2023-01-21 09:43:01.628356: step: 268/529, loss: 0.041181087493896484 2023-01-21 09:43:02.814083: step: 272/529, loss: 0.04162292554974556 2023-01-21 09:43:03.927925: step: 276/529, loss: 0.04879941791296005 2023-01-21 09:43:05.064578: step: 280/529, loss: 0.038159750401973724 2023-01-21 09:43:06.214611: step: 284/529, loss: 0.060675811022520065 2023-01-21 09:43:07.351902: step: 288/529, loss: 0.18230067193508148 2023-01-21 09:43:08.503380: step: 292/529, loss: 0.1592148244380951 2023-01-21 09:43:09.618761: step: 296/529, loss: 0.020122099667787552 2023-01-21 09:43:10.754983: step: 300/529, loss: 0.06520567089319229 2023-01-21 09:43:11.884076: step: 304/529, loss: 0.06264705955982208 2023-01-21 09:43:13.000474: step: 308/529, loss: 0.1522393673658371 2023-01-21 09:43:14.149021: step: 312/529, loss: 0.07093029469251633 2023-01-21 09:43:15.285185: step: 316/529, loss: 0.08137550204992294 2023-01-21 09:43:16.425229: step: 320/529, loss: 0.07664747536182404 2023-01-21 09:43:17.560170: step: 324/529, loss: 0.03860750049352646 2023-01-21 09:43:18.733271: step: 328/529, loss: 0.41961613297462463 2023-01-21 09:43:19.860053: step: 332/529, loss: 0.02822551690042019 2023-01-21 09:43:20.995351: step: 336/529, loss: 0.2965034544467926 2023-01-21 09:43:22.115892: step: 340/529, loss: 0.09459743648767471 2023-01-21 09:43:23.261273: step: 344/529, loss: 0.08964033424854279 2023-01-21 09:43:24.394894: step: 348/529, loss: 0.17884483933448792 2023-01-21 09:43:25.529888: step: 352/529, loss: 0.1376895010471344 2023-01-21 09:43:26.690098: step: 356/529, loss: 0.03124542161822319 2023-01-21 09:43:27.837767: step: 360/529, loss: 0.05669746547937393 2023-01-21 09:43:28.981557: step: 364/529, loss: 0.13595418632030487 2023-01-21 09:43:30.154013: step: 368/529, loss: 0.08331713825464249 2023-01-21 09:43:31.298813: step: 372/529, loss: 0.06101689487695694 2023-01-21 09:43:32.442975: step: 376/529, loss: 0.06221456825733185 2023-01-21 09:43:33.567342: step: 380/529, loss: 0.14734864234924316 2023-01-21 09:43:34.706662: step: 384/529, loss: 0.16810818016529083 2023-01-21 09:43:35.877069: step: 388/529, loss: 0.12999382615089417 2023-01-21 09:43:37.031848: step: 392/529, loss: 0.007696962915360928 2023-01-21 09:43:38.183940: step: 396/529, loss: 0.02716522291302681 2023-01-21 09:43:39.341060: step: 400/529, loss: 0.13542616367340088 2023-01-21 09:43:40.463205: step: 404/529, loss: 0.057915713638067245 2023-01-21 09:43:41.614449: step: 408/529, loss: 0.17980174720287323 2023-01-21 09:43:42.756338: step: 412/529, loss: 0.02211456373333931 2023-01-21 09:43:43.916348: step: 416/529, loss: 0.05041904374957085 2023-01-21 09:43:45.065018: step: 420/529, loss: 0.12386885285377502 2023-01-21 09:43:46.195797: step: 424/529, loss: 0.09354014694690704 2023-01-21 09:43:47.327738: step: 428/529, loss: 0.03186183050274849 2023-01-21 09:43:48.447345: step: 432/529, loss: 0.1892780363559723 2023-01-21 09:43:49.622134: step: 436/529, loss: 0.03945617750287056 2023-01-21 09:43:50.746010: step: 440/529, loss: 0.21628212928771973 2023-01-21 09:43:51.870476: step: 444/529, loss: 0.024228597059845924 2023-01-21 09:43:53.002221: step: 448/529, loss: 0.19096116721630096 2023-01-21 09:43:54.158021: step: 452/529, loss: 0.05030279606580734 2023-01-21 09:43:55.315057: step: 456/529, loss: 0.035592157393693924 2023-01-21 09:43:56.445186: step: 460/529, loss: 0.06591042876243591 2023-01-21 09:43:57.568169: step: 464/529, loss: 0.08670230209827423 2023-01-21 09:43:58.720303: step: 468/529, loss: 0.01954355277121067 2023-01-21 09:43:59.825183: step: 472/529, loss: 0.25123530626296997 2023-01-21 09:44:00.976456: step: 476/529, loss: 0.026798725128173828 2023-01-21 09:44:02.127597: step: 480/529, loss: 0.10264835506677628 2023-01-21 09:44:03.248244: step: 484/529, loss: 0.23255595564842224 2023-01-21 09:44:04.380749: step: 488/529, loss: 0.05419750139117241 2023-01-21 09:44:05.518390: step: 492/529, loss: 0.09384670853614807 2023-01-21 09:44:06.658411: step: 496/529, loss: 0.07152791321277618 2023-01-21 09:44:07.794956: step: 500/529, loss: 0.027454353868961334 2023-01-21 09:44:08.953888: step: 504/529, loss: 0.0911775603890419 2023-01-21 09:44:10.100336: step: 508/529, loss: 0.11702613532543182 2023-01-21 09:44:11.265418: step: 512/529, loss: 0.16320085525512695 2023-01-21 09:44:12.368172: step: 516/529, loss: 0.3054993152618408 2023-01-21 09:44:13.497550: step: 520/529, loss: 0.09896469116210938 2023-01-21 09:44:14.648482: step: 524/529, loss: 0.09951867908239365 2023-01-21 09:44:15.793133: step: 528/529, loss: 0.284894198179245 2023-01-21 09:44:16.905865: step: 532/529, loss: 0.16384926438331604 2023-01-21 09:44:18.045377: step: 536/529, loss: 0.017270183190703392 2023-01-21 09:44:19.172294: step: 540/529, loss: 0.15269556641578674 2023-01-21 09:44:20.334962: step: 544/529, loss: 0.02434062771499157 2023-01-21 09:44:21.467959: step: 548/529, loss: 0.07973742485046387 2023-01-21 09:44:22.613030: step: 552/529, loss: 0.07940101623535156 2023-01-21 09:44:23.767421: step: 556/529, loss: 0.29565659165382385 2023-01-21 09:44:24.948697: step: 560/529, loss: 0.25998440384864807 2023-01-21 09:44:26.104946: step: 564/529, loss: 0.012747240252792835 2023-01-21 09:44:27.236189: step: 568/529, loss: 0.1795019656419754 2023-01-21 09:44:28.361031: step: 572/529, loss: 0.05109043046832085 2023-01-21 09:44:29.483573: step: 576/529, loss: 0.09165702015161514 2023-01-21 09:44:30.618763: step: 580/529, loss: 0.25682029128074646 2023-01-21 09:44:31.746909: step: 584/529, loss: 0.031218720600008965 2023-01-21 09:44:32.875115: step: 588/529, loss: 0.18826337158679962 2023-01-21 09:44:33.992593: step: 592/529, loss: 0.13673192262649536 2023-01-21 09:44:35.132470: step: 596/529, loss: 0.09707927703857422 2023-01-21 09:44:36.276651: step: 600/529, loss: 0.09904041141271591 2023-01-21 09:44:37.387745: step: 604/529, loss: 0.10942697525024414 2023-01-21 09:44:38.515028: step: 608/529, loss: 0.7978367805480957 2023-01-21 09:44:39.666649: step: 612/529, loss: 0.2259882688522339 2023-01-21 09:44:40.821087: step: 616/529, loss: 0.06765404343605042 2023-01-21 09:44:41.959845: step: 620/529, loss: 0.4739961624145508 2023-01-21 09:44:43.108834: step: 624/529, loss: 0.11198749393224716 2023-01-21 09:44:44.257221: step: 628/529, loss: 0.3723900318145752 2023-01-21 09:44:45.381575: step: 632/529, loss: 0.02120819129049778 2023-01-21 09:44:46.519464: step: 636/529, loss: 0.011112213134765625 2023-01-21 09:44:47.659562: step: 640/529, loss: 0.6732972860336304 2023-01-21 09:44:48.797612: step: 644/529, loss: 0.05375304073095322 2023-01-21 09:44:49.910702: step: 648/529, loss: 0.05211582034826279 2023-01-21 09:44:51.053540: step: 652/529, loss: 0.10294894874095917 2023-01-21 09:44:52.199537: step: 656/529, loss: 0.06932105869054794 2023-01-21 09:44:53.305716: step: 660/529, loss: 1.161803960800171 2023-01-21 09:44:54.424834: step: 664/529, loss: 0.11703729629516602 2023-01-21 09:44:55.557502: step: 668/529, loss: 0.19213466346263885 2023-01-21 09:44:56.696709: step: 672/529, loss: 0.19419851899147034 2023-01-21 09:44:57.821452: step: 676/529, loss: 0.05112151801586151 2023-01-21 09:44:58.962816: step: 680/529, loss: 0.11244416236877441 2023-01-21 09:45:00.100056: step: 684/529, loss: 0.15277761220932007 2023-01-21 09:45:01.245014: step: 688/529, loss: 0.07526111602783203 2023-01-21 09:45:02.378806: step: 692/529, loss: 0.23086854815483093 2023-01-21 09:45:03.522795: step: 696/529, loss: 0.14803822338581085 2023-01-21 09:45:04.664112: step: 700/529, loss: 0.10436268150806427 2023-01-21 09:45:05.815972: step: 704/529, loss: 0.028618240728974342 2023-01-21 09:45:06.956696: step: 708/529, loss: 0.10217809677124023 2023-01-21 09:45:08.077999: step: 712/529, loss: 0.08290252834558487 2023-01-21 09:45:09.208327: step: 716/529, loss: 0.21745462715625763 2023-01-21 09:45:10.341990: step: 720/529, loss: 0.34703075885772705 2023-01-21 09:45:11.466551: step: 724/529, loss: 0.14654918015003204 2023-01-21 09:45:12.618850: step: 728/529, loss: 0.21568042039871216 2023-01-21 09:45:13.701599: step: 732/529, loss: 0.08602862805128098 2023-01-21 09:45:14.832739: step: 736/529, loss: 0.040395356714725494 2023-01-21 09:45:15.992790: step: 740/529, loss: 0.09341306984424591 2023-01-21 09:45:17.103805: step: 744/529, loss: 0.06264495849609375 2023-01-21 09:45:18.234044: step: 748/529, loss: 0.1596328765153885 2023-01-21 09:45:19.341831: step: 752/529, loss: 0.17639192938804626 2023-01-21 09:45:20.489469: step: 756/529, loss: 0.533918559551239 2023-01-21 09:45:21.613938: step: 760/529, loss: 0.06819219887256622 2023-01-21 09:45:22.750415: step: 764/529, loss: 0.0642085075378418 2023-01-21 09:45:23.875927: step: 768/529, loss: 0.031195450574159622 2023-01-21 09:45:25.012859: step: 772/529, loss: 0.08208828419446945 2023-01-21 09:45:26.166793: step: 776/529, loss: 0.08131522685289383 2023-01-21 09:45:27.329437: step: 780/529, loss: 0.052170418202877045 2023-01-21 09:45:28.494879: step: 784/529, loss: 0.18491744995117188 2023-01-21 09:45:29.682195: step: 788/529, loss: 0.07410116493701935 2023-01-21 09:45:30.829970: step: 792/529, loss: 0.09056082367897034 2023-01-21 09:45:31.989554: step: 796/529, loss: 0.2675556242465973 2023-01-21 09:45:33.112440: step: 800/529, loss: 0.4297499656677246 2023-01-21 09:45:34.225145: step: 804/529, loss: 0.12766046822071075 2023-01-21 09:45:35.355661: step: 808/529, loss: 0.13021984696388245 2023-01-21 09:45:36.500524: step: 812/529, loss: 0.7451604604721069 2023-01-21 09:45:37.654942: step: 816/529, loss: 0.15032804012298584 2023-01-21 09:45:38.807693: step: 820/529, loss: 0.0913359671831131 2023-01-21 09:45:39.936709: step: 824/529, loss: 0.0726165845990181 2023-01-21 09:45:41.093247: step: 828/529, loss: 0.016353178769350052 2023-01-21 09:45:42.252317: step: 832/529, loss: 0.14222213625907898 2023-01-21 09:45:43.380893: step: 836/529, loss: 0.060122109949588776 2023-01-21 09:45:44.505859: step: 840/529, loss: 0.09133853763341904 2023-01-21 09:45:45.641112: step: 844/529, loss: 0.08547191321849823 2023-01-21 09:45:46.833505: step: 848/529, loss: 0.5536893010139465 2023-01-21 09:45:47.956147: step: 852/529, loss: 0.042575884610414505 2023-01-21 09:45:49.140244: step: 856/529, loss: 0.03320198133587837 2023-01-21 09:45:50.267301: step: 860/529, loss: 0.6481418609619141 2023-01-21 09:45:51.380032: step: 864/529, loss: 0.1351698935031891 2023-01-21 09:45:52.486148: step: 868/529, loss: 0.0561734214425087 2023-01-21 09:45:53.620100: step: 872/529, loss: 0.44409817457199097 2023-01-21 09:45:54.749534: step: 876/529, loss: 0.05455455929040909 2023-01-21 09:45:55.917356: step: 880/529, loss: 0.41209110617637634 2023-01-21 09:45:57.069844: step: 884/529, loss: 0.18684768676757812 2023-01-21 09:45:58.220258: step: 888/529, loss: 0.043112851679325104 2023-01-21 09:45:59.371408: step: 892/529, loss: 0.0607968345284462 2023-01-21 09:46:00.524680: step: 896/529, loss: 0.3999425172805786 2023-01-21 09:46:01.664892: step: 900/529, loss: 0.11191777884960175 2023-01-21 09:46:02.871815: step: 904/529, loss: 0.13103695213794708 2023-01-21 09:46:04.021633: step: 908/529, loss: 0.051504041999578476 2023-01-21 09:46:05.130942: step: 912/529, loss: 0.016675090417265892 2023-01-21 09:46:06.279277: step: 916/529, loss: 0.039655208587646484 2023-01-21 09:46:07.419622: step: 920/529, loss: 0.3030610978603363 2023-01-21 09:46:08.548566: step: 924/529, loss: 0.8533024787902832 2023-01-21 09:46:09.655654: step: 928/529, loss: 0.0611872673034668 2023-01-21 09:46:10.800396: step: 932/529, loss: 0.16871222853660583 2023-01-21 09:46:11.939281: step: 936/529, loss: 0.170294851064682 2023-01-21 09:46:13.079482: step: 940/529, loss: 0.0599120631814003 2023-01-21 09:46:14.208977: step: 944/529, loss: 0.015521145425736904 2023-01-21 09:46:15.344024: step: 948/529, loss: 0.11492423713207245 2023-01-21 09:46:16.455593: step: 952/529, loss: 0.09835801273584366 2023-01-21 09:46:17.593812: step: 956/529, loss: 0.03549613803625107 2023-01-21 09:46:18.732799: step: 960/529, loss: 0.026025010272860527 2023-01-21 09:46:19.872378: step: 964/529, loss: 0.03955962881445885 2023-01-21 09:46:21.009052: step: 968/529, loss: 0.283291220664978 2023-01-21 09:46:22.138626: step: 972/529, loss: 0.06722621619701385 2023-01-21 09:46:23.268295: step: 976/529, loss: 0.13107609748840332 2023-01-21 09:46:24.403693: step: 980/529, loss: 0.2288549542427063 2023-01-21 09:46:25.551770: step: 984/529, loss: 0.07444553822278976 2023-01-21 09:46:26.675371: step: 988/529, loss: 0.07134180516004562 2023-01-21 09:46:27.809974: step: 992/529, loss: 0.8081989884376526 2023-01-21 09:46:28.936959: step: 996/529, loss: 0.07223998010158539 2023-01-21 09:46:30.079862: step: 1000/529, loss: 0.015805674716830254 2023-01-21 09:46:31.195188: step: 1004/529, loss: 0.13688930869102478 2023-01-21 09:46:32.348206: step: 1008/529, loss: 1.0747992992401123 2023-01-21 09:46:33.486755: step: 1012/529, loss: 0.03820457309484482 2023-01-21 09:46:34.636360: step: 1016/529, loss: 0.114777572453022 2023-01-21 09:46:35.800080: step: 1020/529, loss: 0.04070005193352699 2023-01-21 09:46:36.954838: step: 1024/529, loss: 0.12567424774169922 2023-01-21 09:46:38.086909: step: 1028/529, loss: 0.04873304441571236 2023-01-21 09:46:39.210224: step: 1032/529, loss: 0.09399242699146271 2023-01-21 09:46:40.362775: step: 1036/529, loss: 0.04477892071008682 2023-01-21 09:46:41.502129: step: 1040/529, loss: 0.2507326006889343 2023-01-21 09:46:42.629845: step: 1044/529, loss: 0.05149943754076958 2023-01-21 09:46:43.756820: step: 1048/529, loss: 0.10164327919483185 2023-01-21 09:46:44.893167: step: 1052/529, loss: 0.03725767135620117 2023-01-21 09:46:46.042534: step: 1056/529, loss: 0.12715959548950195 2023-01-21 09:46:47.169261: step: 1060/529, loss: 0.1448778212070465 2023-01-21 09:46:48.306628: step: 1064/529, loss: 0.03995705023407936 2023-01-21 09:46:49.442388: step: 1068/529, loss: 0.8035299777984619 2023-01-21 09:46:50.567871: step: 1072/529, loss: 0.07350502163171768 2023-01-21 09:46:51.704558: step: 1076/529, loss: 0.10941322147846222 2023-01-21 09:46:52.844721: step: 1080/529, loss: 0.07037315517663956 2023-01-21 09:46:53.996309: step: 1084/529, loss: 0.31369680166244507 2023-01-21 09:46:55.112331: step: 1088/529, loss: 0.04736471176147461 2023-01-21 09:46:56.229581: step: 1092/529, loss: 0.02118206024169922 2023-01-21 09:46:57.359755: step: 1096/529, loss: 0.22719049453735352 2023-01-21 09:46:58.515117: step: 1100/529, loss: 0.03657474368810654 2023-01-21 09:46:59.653629: step: 1104/529, loss: 0.04487466812133789 2023-01-21 09:47:00.785427: step: 1108/529, loss: 0.035875797271728516 2023-01-21 09:47:01.922554: step: 1112/529, loss: 0.2374078780412674 2023-01-21 09:47:03.081979: step: 1116/529, loss: 0.10019969940185547 2023-01-21 09:47:04.213269: step: 1120/529, loss: 0.025711774826049805 2023-01-21 09:47:05.375300: step: 1124/529, loss: 0.3841056525707245 2023-01-21 09:47:06.532795: step: 1128/529, loss: 0.10552720725536346 2023-01-21 09:47:07.695939: step: 1132/529, loss: 0.031164171174168587 2023-01-21 09:47:08.851413: step: 1136/529, loss: 0.11592216789722443 2023-01-21 09:47:09.973557: step: 1140/529, loss: 0.036431025713682175 2023-01-21 09:47:11.102019: step: 1144/529, loss: 0.12117863446474075 2023-01-21 09:47:12.238791: step: 1148/529, loss: 0.08418798446655273 2023-01-21 09:47:13.391923: step: 1152/529, loss: 0.0654185563325882 2023-01-21 09:47:14.544138: step: 1156/529, loss: 0.17068414390087128 2023-01-21 09:47:15.717907: step: 1160/529, loss: 0.025050830096006393 2023-01-21 09:47:16.850582: step: 1164/529, loss: 0.01633739471435547 2023-01-21 09:47:17.999402: step: 1168/529, loss: 0.13830961287021637 2023-01-21 09:47:19.130478: step: 1172/529, loss: 0.09134463965892792 2023-01-21 09:47:20.252033: step: 1176/529, loss: 0.05947933346033096 2023-01-21 09:47:21.413783: step: 1180/529, loss: 0.3574865460395813 2023-01-21 09:47:22.521117: step: 1184/529, loss: 0.07889652997255325 2023-01-21 09:47:23.631371: step: 1188/529, loss: 0.06050758436322212 2023-01-21 09:47:24.758800: step: 1192/529, loss: 0.03741293027997017 2023-01-21 09:47:25.852822: step: 1196/529, loss: 0.3063947856426239 2023-01-21 09:47:27.000834: step: 1200/529, loss: 0.17077618837356567 2023-01-21 09:47:28.152679: step: 1204/529, loss: 0.30345994234085083 2023-01-21 09:47:29.279759: step: 1208/529, loss: 0.4140658378601074 2023-01-21 09:47:30.407085: step: 1212/529, loss: 0.1676034927368164 2023-01-21 09:47:31.537941: step: 1216/529, loss: 0.49675828218460083 2023-01-21 09:47:32.667446: step: 1220/529, loss: 0.14764443039894104 2023-01-21 09:47:33.784037: step: 1224/529, loss: 0.02834343910217285 2023-01-21 09:47:34.902131: step: 1228/529, loss: 0.14538101851940155 2023-01-21 09:47:36.039546: step: 1232/529, loss: 0.6591930985450745 2023-01-21 09:47:37.162509: step: 1236/529, loss: 0.13059931993484497 2023-01-21 09:47:38.297585: step: 1240/529, loss: 0.11616416275501251 2023-01-21 09:47:39.411354: step: 1244/529, loss: 0.11457046866416931 2023-01-21 09:47:40.538632: step: 1248/529, loss: 1.0732479095458984 2023-01-21 09:47:41.654582: step: 1252/529, loss: 0.018082572147250175 2023-01-21 09:47:42.785808: step: 1256/529, loss: 0.054738424718379974 2023-01-21 09:47:43.913427: step: 1260/529, loss: 0.06846165657043457 2023-01-21 09:47:45.033801: step: 1264/529, loss: 0.027254248037934303 2023-01-21 09:47:46.184446: step: 1268/529, loss: 0.11909093707799911 2023-01-21 09:47:47.308547: step: 1272/529, loss: 1.4714410305023193 2023-01-21 09:47:48.430272: step: 1276/529, loss: 0.06743469834327698 2023-01-21 09:47:49.556039: step: 1280/529, loss: 0.08549775928258896 2023-01-21 09:47:50.693048: step: 1284/529, loss: 0.13805027306079865 2023-01-21 09:47:51.867111: step: 1288/529, loss: 0.09634530544281006 2023-01-21 09:47:53.029485: step: 1292/529, loss: 0.08727006614208221 2023-01-21 09:47:54.162156: step: 1296/529, loss: 0.10100764781236649 2023-01-21 09:47:55.313462: step: 1300/529, loss: 0.07968340069055557 2023-01-21 09:47:56.436269: step: 1304/529, loss: 0.0658191666007042 2023-01-21 09:47:57.559606: step: 1308/529, loss: 0.11240248382091522 2023-01-21 09:47:58.698398: step: 1312/529, loss: 0.150920107960701 2023-01-21 09:47:59.863127: step: 1316/529, loss: 0.015492726117372513 2023-01-21 09:48:01.019086: step: 1320/529, loss: 0.12983761727809906 2023-01-21 09:48:02.165083: step: 1324/529, loss: 0.12339344620704651 2023-01-21 09:48:03.348388: step: 1328/529, loss: 1.1393409967422485 2023-01-21 09:48:04.481569: step: 1332/529, loss: 0.26907700300216675 2023-01-21 09:48:05.606948: step: 1336/529, loss: 0.27794188261032104 2023-01-21 09:48:06.755159: step: 1340/529, loss: 0.0393255241215229 2023-01-21 09:48:07.870530: step: 1344/529, loss: 0.08656424283981323 2023-01-21 09:48:09.027924: step: 1348/529, loss: 0.015587425790727139 2023-01-21 09:48:10.136509: step: 1352/529, loss: 0.1288565695285797 2023-01-21 09:48:11.280062: step: 1356/529, loss: 0.02566203847527504 2023-01-21 09:48:12.424290: step: 1360/529, loss: 0.3174566328525543 2023-01-21 09:48:13.562706: step: 1364/529, loss: 0.12984704971313477 2023-01-21 09:48:14.705011: step: 1368/529, loss: 0.11874476075172424 2023-01-21 09:48:15.821920: step: 1372/529, loss: 0.0717153549194336 2023-01-21 09:48:16.938196: step: 1376/529, loss: 0.11055093258619308 2023-01-21 09:48:18.043222: step: 1380/529, loss: 0.16246142983436584 2023-01-21 09:48:19.204552: step: 1384/529, loss: 0.05267667770385742 2023-01-21 09:48:20.376746: step: 1388/529, loss: 0.3214760720729828 2023-01-21 09:48:21.507210: step: 1392/529, loss: 0.0080725671723485 2023-01-21 09:48:22.658676: step: 1396/529, loss: 0.16526928544044495 2023-01-21 09:48:23.792230: step: 1400/529, loss: 0.17696905136108398 2023-01-21 09:48:24.926033: step: 1404/529, loss: 0.1043233871459961 2023-01-21 09:48:26.058266: step: 1408/529, loss: 0.34566959738731384 2023-01-21 09:48:27.253849: step: 1412/529, loss: 0.15531301498413086 2023-01-21 09:48:28.395925: step: 1416/529, loss: 0.12357378005981445 2023-01-21 09:48:29.544637: step: 1420/529, loss: 0.1280101239681244 2023-01-21 09:48:30.702221: step: 1424/529, loss: 0.07814846187829971 2023-01-21 09:48:31.842826: step: 1428/529, loss: 0.23037277162075043 2023-01-21 09:48:33.007634: step: 1432/529, loss: 0.08404131233692169 2023-01-21 09:48:34.147833: step: 1436/529, loss: 1.2264775037765503 2023-01-21 09:48:35.295492: step: 1440/529, loss: 0.1733301281929016 2023-01-21 09:48:36.451291: step: 1444/529, loss: 0.4530996084213257 2023-01-21 09:48:37.658711: step: 1448/529, loss: 0.15931062400341034 2023-01-21 09:48:38.781169: step: 1452/529, loss: 0.03198566287755966 2023-01-21 09:48:39.905650: step: 1456/529, loss: 0.05705156549811363 2023-01-21 09:48:41.026798: step: 1460/529, loss: 0.13826751708984375 2023-01-21 09:48:42.150044: step: 1464/529, loss: 0.07437744736671448 2023-01-21 09:48:43.261093: step: 1468/529, loss: 0.09493550658226013 2023-01-21 09:48:44.398528: step: 1472/529, loss: 0.1000904068350792 2023-01-21 09:48:45.520076: step: 1476/529, loss: 0.013208961114287376 2023-01-21 09:48:46.647571: step: 1480/529, loss: 0.045256901532411575 2023-01-21 09:48:47.788442: step: 1484/529, loss: 0.14116841554641724 2023-01-21 09:48:48.911505: step: 1488/529, loss: 0.055004313588142395 2023-01-21 09:48:50.038345: step: 1492/529, loss: 0.12032108008861542 2023-01-21 09:48:51.150003: step: 1496/529, loss: 0.11526477336883545 2023-01-21 09:48:52.268004: step: 1500/529, loss: 0.10058292746543884 2023-01-21 09:48:53.398099: step: 1504/529, loss: 0.14667540788650513 2023-01-21 09:48:54.518165: step: 1508/529, loss: 0.09236583858728409 2023-01-21 09:48:55.653594: step: 1512/529, loss: 0.13801270723342896 2023-01-21 09:48:56.805223: step: 1516/529, loss: 0.13479165732860565 2023-01-21 09:48:57.939199: step: 1520/529, loss: 0.7682908177375793 2023-01-21 09:48:59.076088: step: 1524/529, loss: 0.14104799926280975 2023-01-21 09:49:00.195840: step: 1528/529, loss: 0.22073164582252502 2023-01-21 09:49:01.324944: step: 1532/529, loss: 0.06504352390766144 2023-01-21 09:49:02.440758: step: 1536/529, loss: 0.10310936719179153 2023-01-21 09:49:03.571201: step: 1540/529, loss: 0.07471437007188797 2023-01-21 09:49:04.689979: step: 1544/529, loss: 0.04968462139368057 2023-01-21 09:49:05.833007: step: 1548/529, loss: 0.10697784274816513 2023-01-21 09:49:06.970158: step: 1552/529, loss: 0.10752282291650772 2023-01-21 09:49:08.082384: step: 1556/529, loss: 0.08000221103429794 2023-01-21 09:49:09.213013: step: 1560/529, loss: 0.12828406691551208 2023-01-21 09:49:10.351333: step: 1564/529, loss: 0.08738312870264053 2023-01-21 09:49:11.466649: step: 1568/529, loss: 0.643318772315979 2023-01-21 09:49:12.614105: step: 1572/529, loss: 0.10045900195837021 2023-01-21 09:49:13.762800: step: 1576/529, loss: 0.06383457034826279 2023-01-21 09:49:14.886158: step: 1580/529, loss: 0.0026175023522228003 2023-01-21 09:49:16.004381: step: 1584/529, loss: 0.14141488075256348 2023-01-21 09:49:17.161375: step: 1588/529, loss: 0.1619626134634018 2023-01-21 09:49:18.309607: step: 1592/529, loss: 0.08035793155431747 2023-01-21 09:49:19.432411: step: 1596/529, loss: 0.1018514633178711 2023-01-21 09:49:20.564794: step: 1600/529, loss: 0.05981726944446564 2023-01-21 09:49:21.703484: step: 1604/529, loss: 0.08651790767908096 2023-01-21 09:49:22.831400: step: 1608/529, loss: 0.08780994266271591 2023-01-21 09:49:23.985787: step: 1612/529, loss: 0.5993364453315735 2023-01-21 09:49:25.112754: step: 1616/529, loss: 0.12241001427173615 2023-01-21 09:49:26.256570: step: 1620/529, loss: 0.08432440459728241 2023-01-21 09:49:27.389032: step: 1624/529, loss: 0.19328823685646057 2023-01-21 09:49:28.530198: step: 1628/529, loss: 0.06647606194019318 2023-01-21 09:49:29.687862: step: 1632/529, loss: 0.27084845304489136 2023-01-21 09:49:30.804034: step: 1636/529, loss: 0.38451796770095825 2023-01-21 09:49:31.984616: step: 1640/529, loss: 0.9823002815246582 2023-01-21 09:49:33.117245: step: 1644/529, loss: 0.10056290775537491 2023-01-21 09:49:34.281764: step: 1648/529, loss: 0.025450706481933594 2023-01-21 09:49:35.426092: step: 1652/529, loss: 0.06078185886144638 2023-01-21 09:49:36.596036: step: 1656/529, loss: 0.34959831833839417 2023-01-21 09:49:37.709751: step: 1660/529, loss: 0.06246213614940643 2023-01-21 09:49:38.862760: step: 1664/529, loss: 0.10400734096765518 2023-01-21 09:49:39.994488: step: 1668/529, loss: 0.14536552131175995 2023-01-21 09:49:41.161859: step: 1672/529, loss: 0.2501552402973175 2023-01-21 09:49:42.300881: step: 1676/529, loss: 0.02607426606118679 2023-01-21 09:49:43.433402: step: 1680/529, loss: 0.11617942154407501 2023-01-21 09:49:44.567348: step: 1684/529, loss: 0.043949078768491745 2023-01-21 09:49:45.736132: step: 1688/529, loss: 0.020086385309696198 2023-01-21 09:49:46.869697: step: 1692/529, loss: 0.15174122154712677 2023-01-21 09:49:48.037853: step: 1696/529, loss: 0.08995270729064941 2023-01-21 09:49:49.182782: step: 1700/529, loss: 0.12455320358276367 2023-01-21 09:49:50.331459: step: 1704/529, loss: 0.023541593924164772 2023-01-21 09:49:51.477499: step: 1708/529, loss: 0.02675476111471653 2023-01-21 09:49:52.596494: step: 1712/529, loss: 0.0025634765625 2023-01-21 09:49:53.732761: step: 1716/529, loss: 0.14679060876369476 2023-01-21 09:49:54.866529: step: 1720/529, loss: 0.1373869925737381 2023-01-21 09:49:55.988385: step: 1724/529, loss: 0.09653063118457794 2023-01-21 09:49:57.139768: step: 1728/529, loss: 0.23379340767860413 2023-01-21 09:49:58.264414: step: 1732/529, loss: 0.3909895122051239 2023-01-21 09:49:59.408101: step: 1736/529, loss: 0.13762784004211426 2023-01-21 09:50:00.540940: step: 1740/529, loss: 0.030368424952030182 2023-01-21 09:50:01.683396: step: 1744/529, loss: 0.11322517693042755 2023-01-21 09:50:02.847995: step: 1748/529, loss: 0.08092860877513885 2023-01-21 09:50:03.965059: step: 1752/529, loss: 0.03083205223083496 2023-01-21 09:50:05.102019: step: 1756/529, loss: 0.1130671501159668 2023-01-21 09:50:06.244757: step: 1760/529, loss: 0.08137166500091553 2023-01-21 09:50:07.382279: step: 1764/529, loss: 0.18223676085472107 2023-01-21 09:50:08.506349: step: 1768/529, loss: 0.1380632519721985 2023-01-21 09:50:09.638066: step: 1772/529, loss: 0.07590857148170471 2023-01-21 09:50:10.745051: step: 1776/529, loss: 0.06209617853164673 2023-01-21 09:50:11.902438: step: 1780/529, loss: 0.037247609347105026 2023-01-21 09:50:13.072754: step: 1784/529, loss: 0.33542293310165405 2023-01-21 09:50:14.232964: step: 1788/529, loss: 0.17589330673217773 2023-01-21 09:50:15.408807: step: 1792/529, loss: 0.15250429511070251 2023-01-21 09:50:16.542724: step: 1796/529, loss: 0.13087502121925354 2023-01-21 09:50:17.704698: step: 1800/529, loss: 0.06590089946985245 2023-01-21 09:50:18.856651: step: 1804/529, loss: 0.4045642018318176 2023-01-21 09:50:20.006373: step: 1808/529, loss: 0.12438926845788956 2023-01-21 09:50:21.173040: step: 1812/529, loss: 0.20083141326904297 2023-01-21 09:50:22.302324: step: 1816/529, loss: 0.07120934128761292 2023-01-21 09:50:23.457570: step: 1820/529, loss: 0.08995237946510315 2023-01-21 09:50:24.582407: step: 1824/529, loss: 0.06654606014490128 2023-01-21 09:50:25.733653: step: 1828/529, loss: 0.10235595703125 2023-01-21 09:50:26.862495: step: 1832/529, loss: 0.1309341937303543 2023-01-21 09:50:27.987360: step: 1836/529, loss: 0.1873556226491928 2023-01-21 09:50:29.115542: step: 1840/529, loss: 0.10164289176464081 2023-01-21 09:50:30.266991: step: 1844/529, loss: 0.6633205413818359 2023-01-21 09:50:31.390693: step: 1848/529, loss: 0.04476223140954971 2023-01-21 09:50:32.525871: step: 1852/529, loss: 0.04623117670416832 2023-01-21 09:50:33.654086: step: 1856/529, loss: 0.09150433540344238 2023-01-21 09:50:34.786372: step: 1860/529, loss: 0.39013785123825073 2023-01-21 09:50:35.944812: step: 1864/529, loss: 0.15161553025245667 2023-01-21 09:50:37.101582: step: 1868/529, loss: 0.054671287536621094 2023-01-21 09:50:38.230856: step: 1872/529, loss: 0.11773681640625 2023-01-21 09:50:39.362708: step: 1876/529, loss: 0.2029450386762619 2023-01-21 09:50:40.490356: step: 1880/529, loss: 0.08829344809055328 2023-01-21 09:50:41.623609: step: 1884/529, loss: 0.09330368041992188 2023-01-21 09:50:42.760182: step: 1888/529, loss: 0.174309641122818 2023-01-21 09:50:43.886720: step: 1892/529, loss: 0.10527510941028595 2023-01-21 09:50:45.027268: step: 1896/529, loss: 0.1252739429473877 2023-01-21 09:50:46.211356: step: 1900/529, loss: 0.1409069001674652 2023-01-21 09:50:47.378156: step: 1904/529, loss: 0.1402646154165268 2023-01-21 09:50:48.537491: step: 1908/529, loss: 0.2022300660610199 2023-01-21 09:50:49.671629: step: 1912/529, loss: 0.036243438720703125 2023-01-21 09:50:50.819303: step: 1916/529, loss: 0.12417373061180115 2023-01-21 09:50:51.973815: step: 1920/529, loss: 0.3375236392021179 2023-01-21 09:50:53.120459: step: 1924/529, loss: 0.11629372090101242 2023-01-21 09:50:54.276347: step: 1928/529, loss: 0.10912533104419708 2023-01-21 09:50:55.422823: step: 1932/529, loss: 0.03679618984460831 2023-01-21 09:50:56.557680: step: 1936/529, loss: 0.06200075149536133 2023-01-21 09:50:57.677933: step: 1940/529, loss: 0.0762210339307785 2023-01-21 09:50:58.817440: step: 1944/529, loss: 0.04805798828601837 2023-01-21 09:50:59.948071: step: 1948/529, loss: 0.08847646415233612 2023-01-21 09:51:01.073386: step: 1952/529, loss: 0.09774451702833176 2023-01-21 09:51:02.206432: step: 1956/529, loss: 0.024759482592344284 2023-01-21 09:51:03.376682: step: 1960/529, loss: 0.1670069694519043 2023-01-21 09:51:04.531241: step: 1964/529, loss: 0.1777803897857666 2023-01-21 09:51:05.673312: step: 1968/529, loss: 0.04754888266324997 2023-01-21 09:51:06.813865: step: 1972/529, loss: 0.08843345195055008 2023-01-21 09:51:07.958029: step: 1976/529, loss: 0.055100034922361374 2023-01-21 09:51:09.121535: step: 1980/529, loss: 0.0917171984910965 2023-01-21 09:51:10.257505: step: 1984/529, loss: 0.7161492109298706 2023-01-21 09:51:11.399264: step: 1988/529, loss: 0.19107218086719513 2023-01-21 09:51:12.541588: step: 1992/529, loss: 0.04111886024475098 2023-01-21 09:51:13.691211: step: 1996/529, loss: 0.09747834503650665 2023-01-21 09:51:14.823044: step: 2000/529, loss: 0.06688375771045685 2023-01-21 09:51:15.964265: step: 2004/529, loss: 0.0905034989118576 2023-01-21 09:51:17.105865: step: 2008/529, loss: 0.023035384714603424 2023-01-21 09:51:18.248871: step: 2012/529, loss: 0.09108586609363556 2023-01-21 09:51:19.398996: step: 2016/529, loss: 0.13962984085083008 2023-01-21 09:51:20.525912: step: 2020/529, loss: 0.05945110321044922 2023-01-21 09:51:21.656872: step: 2024/529, loss: 0.004268026445060968 2023-01-21 09:51:22.810291: step: 2028/529, loss: 0.1306636929512024 2023-01-21 09:51:23.927119: step: 2032/529, loss: 0.01739818975329399 2023-01-21 09:51:25.061708: step: 2036/529, loss: 0.13418646156787872 2023-01-21 09:51:26.171054: step: 2040/529, loss: 0.032872725278139114 2023-01-21 09:51:27.342920: step: 2044/529, loss: 0.21858596801757812 2023-01-21 09:51:28.476879: step: 2048/529, loss: 0.04718789830803871 2023-01-21 09:51:29.624506: step: 2052/529, loss: 0.25042039155960083 2023-01-21 09:51:30.757003: step: 2056/529, loss: 0.029134273529052734 2023-01-21 09:51:31.895758: step: 2060/529, loss: 0.08246364444494247 2023-01-21 09:51:33.022208: step: 2064/529, loss: 0.04234304651618004 2023-01-21 09:51:34.168833: step: 2068/529, loss: 0.45850732922554016 2023-01-21 09:51:35.324018: step: 2072/529, loss: 0.1547883152961731 2023-01-21 09:51:36.488548: step: 2076/529, loss: 0.07652082294225693 2023-01-21 09:51:37.618201: step: 2080/529, loss: 0.18373946845531464 2023-01-21 09:51:38.751113: step: 2084/529, loss: 0.0656876415014267 2023-01-21 09:51:39.858820: step: 2088/529, loss: 0.022410297766327858 2023-01-21 09:51:40.980851: step: 2092/529, loss: 0.1502913534641266 2023-01-21 09:51:42.127721: step: 2096/529, loss: 0.18248087167739868 2023-01-21 09:51:43.273002: step: 2100/529, loss: 0.08375997841358185 2023-01-21 09:51:44.446512: step: 2104/529, loss: 0.353262335062027 2023-01-21 09:51:45.566572: step: 2108/529, loss: 0.040602684020996094 2023-01-21 09:51:46.704359: step: 2112/529, loss: 0.09877309948205948 2023-01-21 09:51:47.809631: step: 2116/529, loss: 0.07686974853277206 ================================================== Loss: 0.153 -------------------- Dev: {'event': {'p': 0.5991561181434599, 'r': 0.7563249001331558, 'f1': 0.6686286050618011}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Test: {'event': {'p': 0.6213733075435203, 'r': 0.7444959443800695, 'f1': 0.6773853452820242}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Korean: {'event': {'p': 0.6, 'r': 0.47619047619047616, 'f1': 0.5309734513274337}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Russian: {'event': {'p': 0.3684210526315789, 'r': 0.3888888888888889, 'f1': 0.37837837837837834}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.5991561181434599, 'r': 0.7563249001331558, 'f1': 0.6686286050618011}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Chinese: {'event': {'p': 0.6213733075435203, 'r': 0.7444959443800695, 'f1': 0.6773853452820242}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Korean: {'event': {'p': 0.6052344601962922, 'r': 0.7390146471371505, 'f1': 0.6654676258992807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5926622765757291, 'r': 0.7300115874855156, 'f1': 0.6542056074766356}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.6956521739130435, 'r': 0.5079365079365079, 'f1': 0.5871559633027522}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** Epoch: 8 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:52:33.293223: step: 4/529, loss: 0.05318479612469673 2023-01-21 09:52:34.405740: step: 8/529, loss: 0.1949097216129303 2023-01-21 09:52:35.511742: step: 12/529, loss: 0.023500585928559303 2023-01-21 09:52:36.638730: step: 16/529, loss: 0.3665791451931 2023-01-21 09:52:37.762944: step: 20/529, loss: 0.29152393341064453 2023-01-21 09:52:38.887555: step: 24/529, loss: 0.01975259743630886 2023-01-21 09:52:40.047711: step: 28/529, loss: 0.10723571479320526 2023-01-21 09:52:41.184144: step: 32/529, loss: 0.04829874262213707 2023-01-21 09:52:42.318742: step: 36/529, loss: 0.2006334811449051 2023-01-21 09:52:43.466878: step: 40/529, loss: 0.1003212034702301 2023-01-21 09:52:44.604585: step: 44/529, loss: 0.03500232473015785 2023-01-21 09:52:45.720407: step: 48/529, loss: 0.13632678985595703 2023-01-21 09:52:46.864033: step: 52/529, loss: 0.1662098914384842 2023-01-21 09:52:47.970623: step: 56/529, loss: 0.19724826514720917 2023-01-21 09:52:49.100680: step: 60/529, loss: 0.14256420731544495 2023-01-21 09:52:50.249895: step: 64/529, loss: 1.071069359779358 2023-01-21 09:52:51.402051: step: 68/529, loss: 0.09685635566711426 2023-01-21 09:52:52.528900: step: 72/529, loss: 0.3655296266078949 2023-01-21 09:52:53.673448: step: 76/529, loss: 0.04453172907233238 2023-01-21 09:52:54.790817: step: 80/529, loss: 0.3913934826850891 2023-01-21 09:52:55.919675: step: 84/529, loss: 0.07473363727331161 2023-01-21 09:52:57.082499: step: 88/529, loss: 0.08572244644165039 2023-01-21 09:52:58.212667: step: 92/529, loss: 0.03197994455695152 2023-01-21 09:52:59.340230: step: 96/529, loss: 0.12795200943946838 2023-01-21 09:53:00.483115: step: 100/529, loss: 0.11406002193689346 2023-01-21 09:53:01.622405: step: 104/529, loss: 0.12601271271705627 2023-01-21 09:53:02.765943: step: 108/529, loss: 0.04080963134765625 2023-01-21 09:53:03.927329: step: 112/529, loss: 0.002889442490413785 2023-01-21 09:53:05.051986: step: 116/529, loss: 0.09884748607873917 2023-01-21 09:53:06.172272: step: 120/529, loss: 0.24842369556427002 2023-01-21 09:53:07.306753: step: 124/529, loss: 0.04914894327521324 2023-01-21 09:53:08.436688: step: 128/529, loss: 0.03499181196093559 2023-01-21 09:53:09.561166: step: 132/529, loss: 0.2525457441806793 2023-01-21 09:53:10.691905: step: 136/529, loss: 0.038802146911621094 2023-01-21 09:53:11.815534: step: 140/529, loss: 0.1172059029340744 2023-01-21 09:53:12.962465: step: 144/529, loss: 0.05616440623998642 2023-01-21 09:53:14.104750: step: 148/529, loss: 0.05729561299085617 2023-01-21 09:53:15.291765: step: 152/529, loss: 0.21257629990577698 2023-01-21 09:53:16.445683: step: 156/529, loss: 0.16854453086853027 2023-01-21 09:53:17.579566: step: 160/529, loss: 0.1114848181605339 2023-01-21 09:53:18.732807: step: 164/529, loss: 0.9672426581382751 2023-01-21 09:53:19.858813: step: 168/529, loss: 0.12714210152626038 2023-01-21 09:53:20.993350: step: 172/529, loss: 0.023979663848876953 2023-01-21 09:53:22.120748: step: 176/529, loss: 0.040175821632146835 2023-01-21 09:53:23.250914: step: 180/529, loss: 0.04639720916748047 2023-01-21 09:53:24.366573: step: 184/529, loss: 0.09676265716552734 2023-01-21 09:53:25.520985: step: 188/529, loss: 0.23949480056762695 2023-01-21 09:53:26.635028: step: 192/529, loss: 0.09937691688537598 2023-01-21 09:53:27.762631: step: 196/529, loss: 0.04111146926879883 2023-01-21 09:53:28.884291: step: 200/529, loss: 0.03849754482507706 2023-01-21 09:53:29.994273: step: 204/529, loss: 0.05183448642492294 2023-01-21 09:53:31.132053: step: 208/529, loss: 0.1392926275730133 2023-01-21 09:53:32.263893: step: 212/529, loss: 0.022698592394590378 2023-01-21 09:53:33.392089: step: 216/529, loss: 0.09227676689624786 2023-01-21 09:53:34.525433: step: 220/529, loss: 0.08268661797046661 2023-01-21 09:53:35.647964: step: 224/529, loss: 0.05050544813275337 2023-01-21 09:53:36.766320: step: 228/529, loss: 1.5040233135223389 2023-01-21 09:53:37.876647: step: 232/529, loss: 0.032087456434965134 2023-01-21 09:53:39.015321: step: 236/529, loss: 0.2216595709323883 2023-01-21 09:53:40.127777: step: 240/529, loss: 0.06133265793323517 2023-01-21 09:53:41.272055: step: 244/529, loss: 0.14863300323486328 2023-01-21 09:53:42.403904: step: 248/529, loss: 0.13959331810474396 2023-01-21 09:53:43.558644: step: 252/529, loss: 0.05861089378595352 2023-01-21 09:53:44.696048: step: 256/529, loss: 0.023376846686005592 2023-01-21 09:53:45.849520: step: 260/529, loss: 0.14258232712745667 2023-01-21 09:53:46.984509: step: 264/529, loss: 0.015352893620729446 2023-01-21 09:53:48.145118: step: 268/529, loss: 0.4317781329154968 2023-01-21 09:53:49.265241: step: 272/529, loss: 0.008370732888579369 2023-01-21 09:53:50.412695: step: 276/529, loss: 0.09130898118019104 2023-01-21 09:53:51.555518: step: 280/529, loss: 0.043993424624204636 2023-01-21 09:53:52.674208: step: 284/529, loss: 0.032162856310606 2023-01-21 09:53:53.819546: step: 288/529, loss: 0.005527496337890625 2023-01-21 09:53:54.925363: step: 292/529, loss: 0.13753624260425568 2023-01-21 09:53:56.077907: step: 296/529, loss: 0.07669506222009659 2023-01-21 09:53:57.213155: step: 300/529, loss: 0.0482851043343544 2023-01-21 09:53:58.371915: step: 304/529, loss: 0.030687524005770683 2023-01-21 09:53:59.535998: step: 308/529, loss: 0.06084032356739044 2023-01-21 09:54:00.673362: step: 312/529, loss: 0.0035672190133482218 2023-01-21 09:54:01.798975: step: 316/529, loss: 0.024836160242557526 2023-01-21 09:54:02.939323: step: 320/529, loss: 0.03090486489236355 2023-01-21 09:54:04.082661: step: 324/529, loss: 0.053237106651067734 2023-01-21 09:54:05.235566: step: 328/529, loss: 0.20230978727340698 2023-01-21 09:54:06.392811: step: 332/529, loss: 0.06596565246582031 2023-01-21 09:54:07.541086: step: 336/529, loss: 0.12374744564294815 2023-01-21 09:54:08.681000: step: 340/529, loss: 0.2253352254629135 2023-01-21 09:54:09.813098: step: 344/529, loss: 0.09065189212560654 2023-01-21 09:54:10.964406: step: 348/529, loss: 0.005558634176850319 2023-01-21 09:54:12.119444: step: 352/529, loss: 0.10750294476747513 2023-01-21 09:54:13.246305: step: 356/529, loss: 0.04743318259716034 2023-01-21 09:54:14.380235: step: 360/529, loss: 0.18426036834716797 2023-01-21 09:54:15.542534: step: 364/529, loss: 0.1449565887451172 2023-01-21 09:54:16.680902: step: 368/529, loss: 0.07447991520166397 2023-01-21 09:54:17.808227: step: 372/529, loss: 0.07435961067676544 2023-01-21 09:54:18.960955: step: 376/529, loss: 0.08984804153442383 2023-01-21 09:54:20.099380: step: 380/529, loss: 0.12104926258325577 2023-01-21 09:54:21.220068: step: 384/529, loss: 0.011165762320160866 2023-01-21 09:54:22.361265: step: 388/529, loss: 0.029711389914155006 2023-01-21 09:54:23.464501: step: 392/529, loss: 0.18595843017101288 2023-01-21 09:54:24.620423: step: 396/529, loss: 0.15950337052345276 2023-01-21 09:54:25.779056: step: 400/529, loss: 0.03425941616296768 2023-01-21 09:54:26.957918: step: 404/529, loss: 0.03351545333862305 2023-01-21 09:54:28.099237: step: 408/529, loss: 0.06853863596916199 2023-01-21 09:54:29.252887: step: 412/529, loss: 0.17024406790733337 2023-01-21 09:54:30.403264: step: 416/529, loss: 0.03452186658978462 2023-01-21 09:54:31.547908: step: 420/529, loss: 0.41898632049560547 2023-01-21 09:54:32.696539: step: 424/529, loss: 0.19416266679763794 2023-01-21 09:54:33.812516: step: 428/529, loss: 0.03100604936480522 2023-01-21 09:54:34.978612: step: 432/529, loss: 0.2522452473640442 2023-01-21 09:54:36.112530: step: 436/529, loss: 0.08203506469726562 2023-01-21 09:54:37.248189: step: 440/529, loss: 0.047330476343631744 2023-01-21 09:54:38.360823: step: 444/529, loss: 0.04445610195398331 2023-01-21 09:54:39.482075: step: 448/529, loss: 0.25001177191734314 2023-01-21 09:54:40.608753: step: 452/529, loss: 0.086267851293087 2023-01-21 09:54:41.730789: step: 456/529, loss: 0.14265404641628265 2023-01-21 09:54:42.917050: step: 460/529, loss: 0.07036981731653214 2023-01-21 09:54:44.082887: step: 464/529, loss: 0.18255801498889923 2023-01-21 09:54:45.232706: step: 468/529, loss: 0.135430246591568 2023-01-21 09:54:46.384295: step: 472/529, loss: 0.09917138516902924 2023-01-21 09:54:47.512589: step: 476/529, loss: 0.05850672721862793 2023-01-21 09:54:48.674499: step: 480/529, loss: 0.09747008979320526 2023-01-21 09:54:49.844587: step: 484/529, loss: 0.10258684307336807 2023-01-21 09:54:50.958893: step: 488/529, loss: 0.03634071350097656 2023-01-21 09:54:52.089375: step: 492/529, loss: 0.021259833127260208 2023-01-21 09:54:53.217099: step: 496/529, loss: 0.16468815505504608 2023-01-21 09:54:54.351964: step: 500/529, loss: 0.18522053956985474 2023-01-21 09:54:55.475770: step: 504/529, loss: 0.2683284878730774 2023-01-21 09:54:56.624565: step: 508/529, loss: 0.041706085205078125 2023-01-21 09:54:57.760100: step: 512/529, loss: 0.044283390045166016 2023-01-21 09:54:58.903818: step: 516/529, loss: 0.05767402797937393 2023-01-21 09:55:00.013793: step: 520/529, loss: 0.06858015060424805 2023-01-21 09:55:01.125558: step: 524/529, loss: 0.17204904556274414 2023-01-21 09:55:02.242707: step: 528/529, loss: 0.03605666384100914 2023-01-21 09:55:03.368824: step: 532/529, loss: 0.12136439979076385 2023-01-21 09:55:04.518724: step: 536/529, loss: 0.15342359244823456 2023-01-21 09:55:05.632320: step: 540/529, loss: 0.05116262659430504 2023-01-21 09:55:06.750778: step: 544/529, loss: 0.20479270815849304 2023-01-21 09:55:07.876738: step: 548/529, loss: 0.10099725425243378 2023-01-21 09:55:09.014838: step: 552/529, loss: 0.021530630066990852 2023-01-21 09:55:10.135135: step: 556/529, loss: 0.07126083970069885 2023-01-21 09:55:11.285912: step: 560/529, loss: 0.05545482784509659 2023-01-21 09:55:12.438704: step: 564/529, loss: 0.04364471137523651 2023-01-21 09:55:13.546021: step: 568/529, loss: 0.10060080885887146 2023-01-21 09:55:14.699667: step: 572/529, loss: 0.1453665792942047 2023-01-21 09:55:15.832034: step: 576/529, loss: 0.06455536186695099 2023-01-21 09:55:16.972647: step: 580/529, loss: 0.21362534165382385 2023-01-21 09:55:18.105396: step: 584/529, loss: 0.6793594360351562 2023-01-21 09:55:19.215160: step: 588/529, loss: 0.016106892377138138 2023-01-21 09:55:20.388988: step: 592/529, loss: 0.01707904413342476 2023-01-21 09:55:21.526243: step: 596/529, loss: 0.16646508872509003 2023-01-21 09:55:22.646291: step: 600/529, loss: 0.08240914344787598 2023-01-21 09:55:23.765626: step: 604/529, loss: 0.29609212279319763 2023-01-21 09:55:24.899319: step: 608/529, loss: 0.447531133890152 2023-01-21 09:55:26.024829: step: 612/529, loss: 0.12447375804185867 2023-01-21 09:55:27.156252: step: 616/529, loss: 0.016482163220643997 2023-01-21 09:55:28.291998: step: 620/529, loss: 0.05028257519006729 2023-01-21 09:55:29.438435: step: 624/529, loss: 0.08668084442615509 2023-01-21 09:55:30.566286: step: 628/529, loss: 0.05309438705444336 2023-01-21 09:55:31.684940: step: 632/529, loss: 0.1840221881866455 2023-01-21 09:55:32.804391: step: 636/529, loss: 0.036759376525878906 2023-01-21 09:55:33.905487: step: 640/529, loss: 0.08736486732959747 2023-01-21 09:55:35.023441: step: 644/529, loss: 0.5296918749809265 2023-01-21 09:55:36.167677: step: 648/529, loss: 0.07456312328577042 2023-01-21 09:55:37.303407: step: 652/529, loss: 0.01911679469048977 2023-01-21 09:55:38.407982: step: 656/529, loss: 0.08950238674879074 2023-01-21 09:55:39.552476: step: 660/529, loss: 0.08298883587121964 2023-01-21 09:55:40.693034: step: 664/529, loss: 0.26108962297439575 2023-01-21 09:55:41.870032: step: 668/529, loss: 0.03059511072933674 2023-01-21 09:55:42.989544: step: 672/529, loss: 0.518105685710907 2023-01-21 09:55:44.127922: step: 676/529, loss: 1.171210527420044 2023-01-21 09:55:45.286008: step: 680/529, loss: 0.29627352952957153 2023-01-21 09:55:46.457847: step: 684/529, loss: 0.04069194942712784 2023-01-21 09:55:47.579410: step: 688/529, loss: 0.20056581497192383 2023-01-21 09:55:48.735421: step: 692/529, loss: 0.5721842646598816 2023-01-21 09:55:49.853208: step: 696/529, loss: 0.023599527776241302 2023-01-21 09:55:50.978832: step: 700/529, loss: 0.8086647987365723 2023-01-21 09:55:52.111951: step: 704/529, loss: 0.1393848955631256 2023-01-21 09:55:53.283064: step: 708/529, loss: 0.2550451159477234 2023-01-21 09:55:54.413596: step: 712/529, loss: 0.046225450932979584 2023-01-21 09:55:55.561379: step: 716/529, loss: 0.0578191764652729 2023-01-21 09:55:56.698528: step: 720/529, loss: 0.12173286080360413 2023-01-21 09:55:57.828762: step: 724/529, loss: 0.10239486396312714 2023-01-21 09:55:58.973701: step: 728/529, loss: 0.18153181672096252 2023-01-21 09:56:00.109871: step: 732/529, loss: 0.026910115033388138 2023-01-21 09:56:01.299234: step: 736/529, loss: 0.24620114266872406 2023-01-21 09:56:02.444271: step: 740/529, loss: 0.04441988468170166 2023-01-21 09:56:03.583293: step: 744/529, loss: 0.0939784049987793 2023-01-21 09:56:04.729931: step: 748/529, loss: 0.4313790500164032 2023-01-21 09:56:05.870548: step: 752/529, loss: 0.04972870647907257 2023-01-21 09:56:07.002756: step: 756/529, loss: 0.039792679250240326 2023-01-21 09:56:08.145434: step: 760/529, loss: 0.04780202358961105 2023-01-21 09:56:09.272783: step: 764/529, loss: 0.04972591623663902 2023-01-21 09:56:10.406046: step: 768/529, loss: 0.07703933864831924 2023-01-21 09:56:11.538465: step: 772/529, loss: 0.08703837543725967 2023-01-21 09:56:12.689374: step: 776/529, loss: 0.10489273071289062 2023-01-21 09:56:13.872909: step: 780/529, loss: 0.09100428223609924 2023-01-21 09:56:15.011931: step: 784/529, loss: 0.06083212047815323 2023-01-21 09:56:16.132602: step: 788/529, loss: 0.086255744099617 2023-01-21 09:56:17.254556: step: 792/529, loss: 0.05362110212445259 2023-01-21 09:56:18.422747: step: 796/529, loss: 0.17210140824317932 2023-01-21 09:56:19.546718: step: 800/529, loss: 0.0892024040222168 2023-01-21 09:56:20.658446: step: 804/529, loss: 0.2554086446762085 2023-01-21 09:56:21.788779: step: 808/529, loss: 0.29440221190452576 2023-01-21 09:56:22.937275: step: 812/529, loss: 0.08527612686157227 2023-01-21 09:56:24.103099: step: 816/529, loss: 0.048023417592048645 2023-01-21 09:56:25.271333: step: 820/529, loss: 0.1514115333557129 2023-01-21 09:56:26.404608: step: 824/529, loss: 0.03909330442547798 2023-01-21 09:56:27.546390: step: 828/529, loss: 0.2097305804491043 2023-01-21 09:56:28.721032: step: 832/529, loss: 0.06543455272912979 2023-01-21 09:56:29.844818: step: 836/529, loss: 0.10457348078489304 2023-01-21 09:56:31.002429: step: 840/529, loss: 0.0517488494515419 2023-01-21 09:56:32.116160: step: 844/529, loss: 0.056903742253780365 2023-01-21 09:56:33.245364: step: 848/529, loss: 0.09288758784532547 2023-01-21 09:56:34.376875: step: 852/529, loss: 0.02783241495490074 2023-01-21 09:56:35.507241: step: 856/529, loss: 0.11873073130846024 2023-01-21 09:56:36.638999: step: 860/529, loss: 0.18188077211380005 2023-01-21 09:56:37.770068: step: 864/529, loss: 0.07721786201000214 2023-01-21 09:56:38.946560: step: 868/529, loss: 0.161095529794693 2023-01-21 09:56:40.083935: step: 872/529, loss: 0.11576557159423828 2023-01-21 09:56:41.222933: step: 876/529, loss: 0.044678784906864166 2023-01-21 09:56:42.399608: step: 880/529, loss: 0.14981861412525177 2023-01-21 09:56:43.516533: step: 884/529, loss: 0.01584150828421116 2023-01-21 09:56:44.669348: step: 888/529, loss: 0.05137233808636665 2023-01-21 09:56:45.841300: step: 892/529, loss: 0.013538789004087448 2023-01-21 09:56:46.970922: step: 896/529, loss: 0.056047774851322174 2023-01-21 09:56:48.090592: step: 900/529, loss: 0.02088337019085884 2023-01-21 09:56:49.214895: step: 904/529, loss: 0.05288701504468918 2023-01-21 09:56:50.341944: step: 908/529, loss: 0.03825350105762482 2023-01-21 09:56:51.481846: step: 912/529, loss: 0.09216120839118958 2023-01-21 09:56:52.598756: step: 916/529, loss: 0.037940215319395065 2023-01-21 09:56:53.751486: step: 920/529, loss: 0.07928531616926193 2023-01-21 09:56:54.935684: step: 924/529, loss: 0.186113640666008 2023-01-21 09:56:56.067066: step: 928/529, loss: 0.0708475112915039 2023-01-21 09:56:57.229826: step: 932/529, loss: 0.09119148552417755 2023-01-21 09:56:58.359781: step: 936/529, loss: 0.045620083808898926 2023-01-21 09:56:59.519342: step: 940/529, loss: 1.0741746425628662 2023-01-21 09:57:00.658172: step: 944/529, loss: 0.09215126931667328 2023-01-21 09:57:01.771858: step: 948/529, loss: 0.04115419462323189 2023-01-21 09:57:02.908577: step: 952/529, loss: 0.028416156768798828 2023-01-21 09:57:04.046776: step: 956/529, loss: 0.2806165814399719 2023-01-21 09:57:05.167074: step: 960/529, loss: 0.03901674970984459 2023-01-21 09:57:06.308607: step: 964/529, loss: 0.045200541615486145 2023-01-21 09:57:07.462715: step: 968/529, loss: 0.12343569099903107 2023-01-21 09:57:08.582427: step: 972/529, loss: 0.1250072568655014 2023-01-21 09:57:09.742802: step: 976/529, loss: 0.10300364345312119 2023-01-21 09:57:10.868323: step: 980/529, loss: 0.04640607535839081 2023-01-21 09:57:12.001988: step: 984/529, loss: 0.33151257038116455 2023-01-21 09:57:13.103827: step: 988/529, loss: 0.02757744863629341 2023-01-21 09:57:14.247675: step: 992/529, loss: 0.03221692889928818 2023-01-21 09:57:15.378245: step: 996/529, loss: 0.03503725677728653 2023-01-21 09:57:16.505901: step: 1000/529, loss: 0.4943471848964691 2023-01-21 09:57:17.643538: step: 1004/529, loss: 0.48283347487449646 2023-01-21 09:57:18.773941: step: 1008/529, loss: 0.06921110302209854 2023-01-21 09:57:19.908186: step: 1012/529, loss: 0.019154738634824753 2023-01-21 09:57:21.033252: step: 1016/529, loss: 0.08541993796825409 2023-01-21 09:57:22.148716: step: 1020/529, loss: 0.3570098876953125 2023-01-21 09:57:23.293930: step: 1024/529, loss: 0.09803364425897598 2023-01-21 09:57:24.457518: step: 1028/529, loss: 0.026587389409542084 2023-01-21 09:57:25.580626: step: 1032/529, loss: 0.10580854117870331 2023-01-21 09:57:26.724298: step: 1036/529, loss: 0.17923840880393982 2023-01-21 09:57:27.891745: step: 1040/529, loss: 0.09724769741296768 2023-01-21 09:57:29.000979: step: 1044/529, loss: 0.0658857598900795 2023-01-21 09:57:30.166176: step: 1048/529, loss: 0.66107577085495 2023-01-21 09:57:31.312510: step: 1052/529, loss: 0.07817935943603516 2023-01-21 09:57:32.442597: step: 1056/529, loss: 0.07002381980419159 2023-01-21 09:57:33.580213: step: 1060/529, loss: 0.10745124518871307 2023-01-21 09:57:34.738526: step: 1064/529, loss: 0.05089931935071945 2023-01-21 09:57:35.903038: step: 1068/529, loss: 0.04335417598485947 2023-01-21 09:57:37.018285: step: 1072/529, loss: 0.03953962028026581 2023-01-21 09:57:38.150668: step: 1076/529, loss: 0.021099664270877838 2023-01-21 09:57:39.272917: step: 1080/529, loss: 0.05591411888599396 2023-01-21 09:57:40.415341: step: 1084/529, loss: 0.07288141548633575 2023-01-21 09:57:41.542367: step: 1088/529, loss: 0.02204909361898899 2023-01-21 09:57:42.702566: step: 1092/529, loss: 0.22917665541172028 2023-01-21 09:57:43.839903: step: 1096/529, loss: 0.43293654918670654 2023-01-21 09:57:44.987745: step: 1100/529, loss: 0.09718716144561768 2023-01-21 09:57:46.112630: step: 1104/529, loss: 0.18410708010196686 2023-01-21 09:57:47.246371: step: 1108/529, loss: 0.24055281281471252 2023-01-21 09:57:48.396741: step: 1112/529, loss: 0.0636385902762413 2023-01-21 09:57:49.536217: step: 1116/529, loss: 0.03886529058218002 2023-01-21 09:57:50.667139: step: 1120/529, loss: 0.6433650255203247 2023-01-21 09:57:51.804397: step: 1124/529, loss: 0.1143740639090538 2023-01-21 09:57:52.948537: step: 1128/529, loss: 0.05895862728357315 2023-01-21 09:57:54.100850: step: 1132/529, loss: 0.008826113305985928 2023-01-21 09:57:55.285774: step: 1136/529, loss: 0.11787395924329758 2023-01-21 09:57:56.418039: step: 1140/529, loss: 0.42764148116111755 2023-01-21 09:57:57.560004: step: 1144/529, loss: 0.11822395771741867 2023-01-21 09:57:58.690711: step: 1148/529, loss: 0.06025657802820206 2023-01-21 09:57:59.820404: step: 1152/529, loss: 0.05768585205078125 2023-01-21 09:58:00.964896: step: 1156/529, loss: 0.06109189987182617 2023-01-21 09:58:02.119172: step: 1160/529, loss: 0.40720662474632263 2023-01-21 09:58:03.282759: step: 1164/529, loss: 0.03205518797039986 2023-01-21 09:58:04.415508: step: 1168/529, loss: 0.12144317477941513 2023-01-21 09:58:05.539581: step: 1172/529, loss: 0.06864538788795471 2023-01-21 09:58:06.715510: step: 1176/529, loss: 0.9421911239624023 2023-01-21 09:58:07.856793: step: 1180/529, loss: 0.024359513074159622 2023-01-21 09:58:09.008544: step: 1184/529, loss: 0.1171969398856163 2023-01-21 09:58:10.120473: step: 1188/529, loss: 0.05607500299811363 2023-01-21 09:58:11.265930: step: 1192/529, loss: 0.06799764931201935 2023-01-21 09:58:12.392648: step: 1196/529, loss: 0.5128872990608215 2023-01-21 09:58:13.519531: step: 1200/529, loss: 0.05003919452428818 2023-01-21 09:58:14.667544: step: 1204/529, loss: 0.4062836468219757 2023-01-21 09:58:15.799500: step: 1208/529, loss: 0.11244507133960724 2023-01-21 09:58:16.921326: step: 1212/529, loss: 0.08958905190229416 2023-01-21 09:58:18.069479: step: 1216/529, loss: 0.02769956737756729 2023-01-21 09:58:19.219963: step: 1220/529, loss: 0.28001290559768677 2023-01-21 09:58:20.383050: step: 1224/529, loss: 0.008771037682890892 2023-01-21 09:58:21.494255: step: 1228/529, loss: 0.11592188477516174 2023-01-21 09:58:22.625369: step: 1232/529, loss: 0.009035897441208363 2023-01-21 09:58:23.751528: step: 1236/529, loss: 0.04196319729089737 2023-01-21 09:58:24.876037: step: 1240/529, loss: 0.03999403864145279 2023-01-21 09:58:26.027205: step: 1244/529, loss: 0.09356927871704102 2023-01-21 09:58:27.178629: step: 1248/529, loss: 0.0769878402352333 2023-01-21 09:58:28.344292: step: 1252/529, loss: 0.13391000032424927 2023-01-21 09:58:29.515347: step: 1256/529, loss: 0.11210170388221741 2023-01-21 09:58:30.637062: step: 1260/529, loss: 0.18320532143115997 2023-01-21 09:58:31.772194: step: 1264/529, loss: 0.13048622012138367 2023-01-21 09:58:32.915174: step: 1268/529, loss: 0.03078451193869114 2023-01-21 09:58:34.039012: step: 1272/529, loss: 0.038887396454811096 2023-01-21 09:58:35.170526: step: 1276/529, loss: 0.11737289279699326 2023-01-21 09:58:36.291589: step: 1280/529, loss: 0.06679253280162811 2023-01-21 09:58:37.410576: step: 1284/529, loss: 0.07098370045423508 2023-01-21 09:58:38.562885: step: 1288/529, loss: 0.25798606872558594 2023-01-21 09:58:39.657560: step: 1292/529, loss: 0.02176227606832981 2023-01-21 09:58:40.805688: step: 1296/529, loss: 0.6038552522659302 2023-01-21 09:58:41.960271: step: 1300/529, loss: 0.10248793661594391 2023-01-21 09:58:43.110804: step: 1304/529, loss: 0.05112873390316963 2023-01-21 09:58:44.244267: step: 1308/529, loss: 0.19271716475486755 2023-01-21 09:58:45.398144: step: 1312/529, loss: 0.0049221995286643505 2023-01-21 09:58:46.525241: step: 1316/529, loss: 0.04891491308808327 2023-01-21 09:58:47.674359: step: 1320/529, loss: 0.00672416714951396 2023-01-21 09:58:48.811194: step: 1324/529, loss: 0.19254112243652344 2023-01-21 09:58:49.945066: step: 1328/529, loss: 0.06768903881311417 2023-01-21 09:58:51.074767: step: 1332/529, loss: 0.08918926864862442 2023-01-21 09:58:52.206357: step: 1336/529, loss: 0.12589283287525177 2023-01-21 09:58:53.331949: step: 1340/529, loss: 0.1547192633152008 2023-01-21 09:58:54.459121: step: 1344/529, loss: 0.013988470658659935 2023-01-21 09:58:55.598822: step: 1348/529, loss: 0.0335942767560482 2023-01-21 09:58:56.711628: step: 1352/529, loss: 0.12018919736146927 2023-01-21 09:58:57.844569: step: 1356/529, loss: 0.11278744041919708 2023-01-21 09:58:58.968174: step: 1360/529, loss: 0.06362161040306091 2023-01-21 09:59:00.120355: step: 1364/529, loss: 0.6211714148521423 2023-01-21 09:59:01.273276: step: 1368/529, loss: 0.10992946475744247 2023-01-21 09:59:02.411610: step: 1372/529, loss: 0.0833558514714241 2023-01-21 09:59:03.580937: step: 1376/529, loss: 0.088636115193367 2023-01-21 09:59:04.709409: step: 1380/529, loss: 0.013067150488495827 2023-01-21 09:59:05.862338: step: 1384/529, loss: 0.2268669158220291 2023-01-21 09:59:07.006992: step: 1388/529, loss: 0.025382185354828835 2023-01-21 09:59:08.150117: step: 1392/529, loss: 0.04205169528722763 2023-01-21 09:59:09.303717: step: 1396/529, loss: 0.11921052634716034 2023-01-21 09:59:10.431696: step: 1400/529, loss: 0.13757380843162537 2023-01-21 09:59:11.558080: step: 1404/529, loss: 0.21393270790576935 2023-01-21 09:59:12.734421: step: 1408/529, loss: 0.5535637736320496 2023-01-21 09:59:13.877460: step: 1412/529, loss: 0.052686452865600586 2023-01-21 09:59:15.012193: step: 1416/529, loss: 0.12932290136814117 2023-01-21 09:59:16.165951: step: 1420/529, loss: 0.0324755422770977 2023-01-21 09:59:17.295594: step: 1424/529, loss: 0.039873696863651276 2023-01-21 09:59:18.431252: step: 1428/529, loss: 0.06404171139001846 2023-01-21 09:59:19.577143: step: 1432/529, loss: 0.026835300028324127 2023-01-21 09:59:20.717556: step: 1436/529, loss: 0.06509961932897568 2023-01-21 09:59:21.862631: step: 1440/529, loss: 0.042978666722774506 2023-01-21 09:59:22.986047: step: 1444/529, loss: 0.049712374806404114 2023-01-21 09:59:24.113431: step: 1448/529, loss: 0.03558788448572159 2023-01-21 09:59:25.264746: step: 1452/529, loss: 0.13023605942726135 2023-01-21 09:59:26.415718: step: 1456/529, loss: 0.418692946434021 2023-01-21 09:59:27.563949: step: 1460/529, loss: 0.18764762580394745 2023-01-21 09:59:28.723908: step: 1464/529, loss: 0.05810079723596573 2023-01-21 09:59:29.856091: step: 1468/529, loss: 0.1285969763994217 2023-01-21 09:59:31.013143: step: 1472/529, loss: 0.08018741756677628 2023-01-21 09:59:32.127406: step: 1476/529, loss: 0.07754955440759659 2023-01-21 09:59:33.245769: step: 1480/529, loss: 0.08813085407018661 2023-01-21 09:59:34.381208: step: 1484/529, loss: 0.08301258087158203 2023-01-21 09:59:35.510663: step: 1488/529, loss: 0.11304989457130432 2023-01-21 09:59:36.659965: step: 1492/529, loss: 0.4508059024810791 2023-01-21 09:59:37.789008: step: 1496/529, loss: 0.012687206268310547 2023-01-21 09:59:38.935504: step: 1500/529, loss: 0.1006084457039833 2023-01-21 09:59:40.122331: step: 1504/529, loss: 0.5418062210083008 2023-01-21 09:59:41.274067: step: 1508/529, loss: 0.04585056006908417 2023-01-21 09:59:42.441334: step: 1512/529, loss: 0.04876108095049858 2023-01-21 09:59:43.615287: step: 1516/529, loss: 0.07245712727308273 2023-01-21 09:59:44.767126: step: 1520/529, loss: 0.08835025131702423 2023-01-21 09:59:45.916963: step: 1524/529, loss: 0.09246230125427246 2023-01-21 09:59:47.078012: step: 1528/529, loss: 0.4610868990421295 2023-01-21 09:59:48.230246: step: 1532/529, loss: 0.04692067950963974 2023-01-21 09:59:49.372317: step: 1536/529, loss: 0.02602415159344673 2023-01-21 09:59:50.508682: step: 1540/529, loss: 0.19669437408447266 2023-01-21 09:59:51.634062: step: 1544/529, loss: 0.2208385467529297 2023-01-21 09:59:52.797919: step: 1548/529, loss: 0.04646854102611542 2023-01-21 09:59:53.921276: step: 1552/529, loss: 0.08519373089075089 2023-01-21 09:59:55.037296: step: 1556/529, loss: 0.1493523269891739 2023-01-21 09:59:56.173117: step: 1560/529, loss: 0.15319472551345825 2023-01-21 09:59:57.338345: step: 1564/529, loss: 0.43943822383880615 2023-01-21 09:59:58.469977: step: 1568/529, loss: 0.04767151176929474 2023-01-21 09:59:59.649427: step: 1572/529, loss: 0.10989365726709366 2023-01-21 10:00:00.816967: step: 1576/529, loss: 0.14813171327114105 2023-01-21 10:00:01.950172: step: 1580/529, loss: 0.08882666379213333 2023-01-21 10:00:03.112794: step: 1584/529, loss: 0.6751924753189087 2023-01-21 10:00:04.251316: step: 1588/529, loss: 0.13727807998657227 2023-01-21 10:00:05.370160: step: 1592/529, loss: 0.3277858793735504 2023-01-21 10:00:06.490156: step: 1596/529, loss: 0.035420991480350494 2023-01-21 10:00:07.628041: step: 1600/529, loss: 0.058509137481451035 2023-01-21 10:00:08.788992: step: 1604/529, loss: 0.03420314937829971 2023-01-21 10:00:09.919835: step: 1608/529, loss: 0.12462940067052841 2023-01-21 10:00:11.051839: step: 1612/529, loss: 0.0929853692650795 2023-01-21 10:00:12.189164: step: 1616/529, loss: 0.1898341327905655 2023-01-21 10:00:13.328718: step: 1620/529, loss: 0.11131754517555237 2023-01-21 10:00:14.473713: step: 1624/529, loss: 0.052225757390260696 2023-01-21 10:00:15.604204: step: 1628/529, loss: 0.03277263790369034 2023-01-21 10:00:16.723504: step: 1632/529, loss: 0.012453724630177021 2023-01-21 10:00:17.866417: step: 1636/529, loss: 0.1413319706916809 2023-01-21 10:00:18.990989: step: 1640/529, loss: 0.06446418166160583 2023-01-21 10:00:20.182332: step: 1644/529, loss: 0.0094451904296875 2023-01-21 10:00:21.310795: step: 1648/529, loss: 0.3612396717071533 2023-01-21 10:00:22.439894: step: 1652/529, loss: 0.07011563330888748 2023-01-21 10:00:23.590377: step: 1656/529, loss: 0.07222671806812286 2023-01-21 10:00:24.707336: step: 1660/529, loss: 0.11896353214979172 2023-01-21 10:00:25.902344: step: 1664/529, loss: 0.1331164389848709 2023-01-21 10:00:27.055403: step: 1668/529, loss: 0.2884250581264496 2023-01-21 10:00:28.181973: step: 1672/529, loss: 0.0808897539973259 2023-01-21 10:00:29.349348: step: 1676/529, loss: 0.07857723534107208 2023-01-21 10:00:30.482367: step: 1680/529, loss: 0.16002291440963745 2023-01-21 10:00:31.635082: step: 1684/529, loss: 0.20574329793453217 2023-01-21 10:00:32.767169: step: 1688/529, loss: 0.022405290976166725 2023-01-21 10:00:33.885298: step: 1692/529, loss: 0.05014238506555557 2023-01-21 10:00:35.012509: step: 1696/529, loss: 0.028252029791474342 2023-01-21 10:00:36.148682: step: 1700/529, loss: 0.0632355734705925 2023-01-21 10:00:37.289330: step: 1704/529, loss: 0.10403050482273102 2023-01-21 10:00:38.414219: step: 1708/529, loss: 0.027089595794677734 2023-01-21 10:00:39.548679: step: 1712/529, loss: 0.22194623947143555 2023-01-21 10:00:40.704037: step: 1716/529, loss: 0.10052306950092316 2023-01-21 10:00:41.884751: step: 1720/529, loss: 0.12119140475988388 2023-01-21 10:00:43.019917: step: 1724/529, loss: 0.10400199890136719 2023-01-21 10:00:44.150300: step: 1728/529, loss: 0.12181172519922256 2023-01-21 10:00:45.265629: step: 1732/529, loss: 0.05166090652346611 2023-01-21 10:00:46.453450: step: 1736/529, loss: 0.06643734127283096 2023-01-21 10:00:47.581346: step: 1740/529, loss: 0.05976066738367081 2023-01-21 10:00:48.715932: step: 1744/529, loss: 0.05646400526165962 2023-01-21 10:00:49.852436: step: 1748/529, loss: 0.23236894607543945 2023-01-21 10:00:50.974573: step: 1752/529, loss: 0.24543443322181702 2023-01-21 10:00:52.119683: step: 1756/529, loss: 0.031052017584443092 2023-01-21 10:00:53.268959: step: 1760/529, loss: 0.06249818950891495 2023-01-21 10:00:54.386352: step: 1764/529, loss: 0.04362058639526367 2023-01-21 10:00:55.500375: step: 1768/529, loss: 0.042107775807380676 2023-01-21 10:00:56.630825: step: 1772/529, loss: 0.18370705842971802 2023-01-21 10:00:57.758037: step: 1776/529, loss: 0.16098804771900177 2023-01-21 10:00:58.884659: step: 1780/529, loss: 0.05587730556726456 2023-01-21 10:01:00.033964: step: 1784/529, loss: 0.14834336936473846 2023-01-21 10:01:01.145428: step: 1788/529, loss: 0.0771198719739914 2023-01-21 10:01:02.255750: step: 1792/529, loss: 0.03963589668273926 2023-01-21 10:01:03.376263: step: 1796/529, loss: 0.05296192318201065 2023-01-21 10:01:04.529689: step: 1800/529, loss: 0.03234367445111275 2023-01-21 10:01:05.679011: step: 1804/529, loss: 0.13179749250411987 2023-01-21 10:01:06.794725: step: 1808/529, loss: 0.05174579471349716 2023-01-21 10:01:07.919739: step: 1812/529, loss: 0.01923990249633789 2023-01-21 10:01:09.058682: step: 1816/529, loss: 0.048125553876161575 2023-01-21 10:01:10.192419: step: 1820/529, loss: 0.042542293667793274 2023-01-21 10:01:11.322995: step: 1824/529, loss: 0.044344425201416016 2023-01-21 10:01:12.478300: step: 1828/529, loss: 0.10813198238611221 2023-01-21 10:01:13.612107: step: 1832/529, loss: 0.08128070831298828 2023-01-21 10:01:14.747684: step: 1836/529, loss: 0.08430986106395721 2023-01-21 10:01:15.861932: step: 1840/529, loss: 0.1343776285648346 2023-01-21 10:01:17.025636: step: 1844/529, loss: 0.08498091995716095 2023-01-21 10:01:18.164329: step: 1848/529, loss: 0.07745161652565002 2023-01-21 10:01:19.318051: step: 1852/529, loss: 0.10018501430749893 2023-01-21 10:01:20.440816: step: 1856/529, loss: 0.0638589859008789 2023-01-21 10:01:21.568998: step: 1860/529, loss: 0.077477365732193 2023-01-21 10:01:22.675357: step: 1864/529, loss: 0.03726239129900932 2023-01-21 10:01:23.848459: step: 1868/529, loss: 0.16419890522956848 2023-01-21 10:01:25.005918: step: 1872/529, loss: 0.07609482109546661 2023-01-21 10:01:26.129955: step: 1876/529, loss: 0.55999755859375 2023-01-21 10:01:27.268593: step: 1880/529, loss: 0.09526167064905167 2023-01-21 10:01:28.448319: step: 1884/529, loss: 0.04246530681848526 2023-01-21 10:01:29.585821: step: 1888/529, loss: 0.03554539754986763 2023-01-21 10:01:30.701232: step: 1892/529, loss: 0.26806288957595825 2023-01-21 10:01:31.843222: step: 1896/529, loss: 0.04108457639813423 2023-01-21 10:01:33.016639: step: 1900/529, loss: 0.020398426800966263 2023-01-21 10:01:34.159556: step: 1904/529, loss: 0.026177596300840378 2023-01-21 10:01:35.311461: step: 1908/529, loss: 0.10287800431251526 2023-01-21 10:01:36.447520: step: 1912/529, loss: 0.10997362434864044 2023-01-21 10:01:37.590297: step: 1916/529, loss: 0.03597879409790039 2023-01-21 10:01:38.729730: step: 1920/529, loss: 0.08412304520606995 2023-01-21 10:01:39.883398: step: 1924/529, loss: 0.049201205372810364 2023-01-21 10:01:41.027435: step: 1928/529, loss: 0.07876729965209961 2023-01-21 10:01:42.191515: step: 1932/529, loss: 0.11989942193031311 2023-01-21 10:01:43.332934: step: 1936/529, loss: 0.009969890117645264 2023-01-21 10:01:44.472388: step: 1940/529, loss: 0.00480041466653347 2023-01-21 10:01:45.606949: step: 1944/529, loss: 0.28289785981178284 2023-01-21 10:01:46.760664: step: 1948/529, loss: 0.27803611755371094 2023-01-21 10:01:47.884204: step: 1952/529, loss: 0.025960635393857956 2023-01-21 10:01:49.057218: step: 1956/529, loss: 0.22723540663719177 2023-01-21 10:01:50.201280: step: 1960/529, loss: 0.05281104892492294 2023-01-21 10:01:51.329554: step: 1964/529, loss: 0.023982621729373932 2023-01-21 10:01:52.458937: step: 1968/529, loss: 0.015401411801576614 2023-01-21 10:01:53.600963: step: 1972/529, loss: 0.03710746765136719 2023-01-21 10:01:54.711701: step: 1976/529, loss: 0.11982041597366333 2023-01-21 10:01:55.876201: step: 1980/529, loss: 0.09672613441944122 2023-01-21 10:01:57.013748: step: 1984/529, loss: 0.018486738204956055 2023-01-21 10:01:58.151771: step: 1988/529, loss: 0.06714914739131927 2023-01-21 10:01:59.269160: step: 1992/529, loss: 0.07497940212488174 2023-01-21 10:02:00.391115: step: 1996/529, loss: 0.1743757277727127 2023-01-21 10:02:01.502362: step: 2000/529, loss: 0.4676494598388672 2023-01-21 10:02:02.644231: step: 2004/529, loss: 0.12826529145240784 2023-01-21 10:02:03.786632: step: 2008/529, loss: 0.016652679070830345 2023-01-21 10:02:04.937942: step: 2012/529, loss: 0.14339298009872437 2023-01-21 10:02:06.094439: step: 2016/529, loss: 0.23808622360229492 2023-01-21 10:02:07.262586: step: 2020/529, loss: 0.1738269329071045 2023-01-21 10:02:08.395397: step: 2024/529, loss: 0.033908870071172714 2023-01-21 10:02:09.526977: step: 2028/529, loss: 0.13454551994800568 2023-01-21 10:02:10.649592: step: 2032/529, loss: 0.26784712076187134 2023-01-21 10:02:11.785389: step: 2036/529, loss: 0.007646751590073109 2023-01-21 10:02:12.922704: step: 2040/529, loss: 0.027662038803100586 2023-01-21 10:02:14.093041: step: 2044/529, loss: 0.19166623055934906 2023-01-21 10:02:15.214864: step: 2048/529, loss: 0.021297072991728783 2023-01-21 10:02:16.377808: step: 2052/529, loss: 0.11430978775024414 2023-01-21 10:02:17.494566: step: 2056/529, loss: 0.13252010941505432 2023-01-21 10:02:18.636796: step: 2060/529, loss: 0.09151826053857803 2023-01-21 10:02:19.765809: step: 2064/529, loss: 0.0400049202144146 2023-01-21 10:02:20.926754: step: 2068/529, loss: 0.06467743217945099 2023-01-21 10:02:22.081302: step: 2072/529, loss: 0.45453158020973206 2023-01-21 10:02:23.215549: step: 2076/529, loss: 0.020348738878965378 2023-01-21 10:02:24.320923: step: 2080/529, loss: 0.08773956447839737 2023-01-21 10:02:25.446388: step: 2084/529, loss: 0.05284789949655533 2023-01-21 10:02:26.601255: step: 2088/529, loss: 0.3052424490451813 2023-01-21 10:02:27.743206: step: 2092/529, loss: 0.09540295600891113 2023-01-21 10:02:28.875843: step: 2096/529, loss: 0.08626031875610352 2023-01-21 10:02:30.025827: step: 2100/529, loss: 0.03313612937927246 2023-01-21 10:02:31.168535: step: 2104/529, loss: 0.03666281700134277 2023-01-21 10:02:32.306874: step: 2108/529, loss: 0.20233607292175293 2023-01-21 10:02:33.427117: step: 2112/529, loss: 0.023164652287960052 2023-01-21 10:02:34.558816: step: 2116/529, loss: 0.012871170416474342 ================================================== Loss: 0.132 -------------------- Dev: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Test: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Russian: {'event': {'p': 0.47368421052631576, 'r': 0.5, 'f1': 0.4864864864864865}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} New best chinese model... New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 10:03:28.501984: step: 4/529, loss: 0.024635886773467064 2023-01-21 10:03:29.638894: step: 8/529, loss: 0.12102871388196945 2023-01-21 10:03:30.745885: step: 12/529, loss: 0.024912506341934204 2023-01-21 10:03:31.860190: step: 16/529, loss: 0.1407354325056076 2023-01-21 10:03:32.977769: step: 20/529, loss: 0.8606937527656555 2023-01-21 10:03:34.116375: step: 24/529, loss: 0.03574056550860405 2023-01-21 10:03:35.248588: step: 28/529, loss: 0.02708921581506729 2023-01-21 10:03:36.379359: step: 32/529, loss: 0.08230924606323242 2023-01-21 10:03:37.516072: step: 36/529, loss: 0.05329904705286026 2023-01-21 10:03:38.636291: step: 40/529, loss: 0.06985016167163849 2023-01-21 10:03:39.798784: step: 44/529, loss: 0.005886650178581476 2023-01-21 10:03:40.917339: step: 48/529, loss: 0.10070696473121643 2023-01-21 10:03:42.088981: step: 52/529, loss: 0.04015941545367241 2023-01-21 10:03:43.209903: step: 56/529, loss: 0.09081902354955673 2023-01-21 10:03:44.351819: step: 60/529, loss: 0.0830882117152214 2023-01-21 10:03:45.480180: step: 64/529, loss: 0.10325441509485245 2023-01-21 10:03:46.600740: step: 68/529, loss: 0.047469139099121094 2023-01-21 10:03:47.711122: step: 72/529, loss: 0.002521228976547718 2023-01-21 10:03:48.833937: step: 76/529, loss: 0.008720017038285732 2023-01-21 10:03:49.980233: step: 80/529, loss: 0.09637127071619034 2023-01-21 10:03:51.130848: step: 84/529, loss: 0.0161316879093647 2023-01-21 10:03:52.245287: step: 88/529, loss: 0.0007552146562375128 2023-01-21 10:03:53.343910: step: 92/529, loss: 0.00423774728551507 2023-01-21 10:03:54.441557: step: 96/529, loss: 0.10617323219776154 2023-01-21 10:03:55.577637: step: 100/529, loss: 0.07259197533130646 2023-01-21 10:03:56.715166: step: 104/529, loss: 0.172722727060318 2023-01-21 10:03:57.848191: step: 108/529, loss: 0.07465362548828125 2023-01-21 10:03:58.973834: step: 112/529, loss: 0.04421424865722656 2023-01-21 10:04:00.106453: step: 116/529, loss: 0.19773635268211365 2023-01-21 10:04:01.280424: step: 120/529, loss: 0.6913812160491943 2023-01-21 10:04:02.442353: step: 124/529, loss: 0.19183094799518585 2023-01-21 10:04:03.593350: step: 128/529, loss: 0.1485912799835205 2023-01-21 10:04:04.759484: step: 132/529, loss: 0.005334615707397461 2023-01-21 10:04:05.893691: step: 136/529, loss: 0.0032499313820153475 2023-01-21 10:04:07.034317: step: 140/529, loss: 0.12808199226856232 2023-01-21 10:04:08.151122: step: 144/529, loss: 0.16091834008693695 2023-01-21 10:04:09.277563: step: 148/529, loss: 0.22805194556713104 2023-01-21 10:04:10.425272: step: 152/529, loss: 0.04231410101056099 2023-01-21 10:04:11.545730: step: 156/529, loss: 0.09192819893360138 2023-01-21 10:04:12.655387: step: 160/529, loss: 0.1499231457710266 2023-01-21 10:04:13.815203: step: 164/529, loss: 0.1801927089691162 2023-01-21 10:04:14.935272: step: 168/529, loss: 0.015019417740404606 2023-01-21 10:04:16.094224: step: 172/529, loss: 0.08351421356201172 2023-01-21 10:04:17.247985: step: 176/529, loss: 0.17925682663917542 2023-01-21 10:04:18.391271: step: 180/529, loss: 0.02656717225909233 2023-01-21 10:04:19.530138: step: 184/529, loss: 0.05475788190960884 2023-01-21 10:04:20.714199: step: 188/529, loss: 0.08898897469043732 2023-01-21 10:04:21.886767: step: 192/529, loss: 0.3894866108894348 2023-01-21 10:04:23.012666: step: 196/529, loss: 0.0421941764652729 2023-01-21 10:04:24.124604: step: 200/529, loss: 0.0024327754508703947 2023-01-21 10:04:25.285570: step: 204/529, loss: 0.14827638864517212 2023-01-21 10:04:26.449908: step: 208/529, loss: 0.12349166721105576 2023-01-21 10:04:27.578597: step: 212/529, loss: 0.002152395434677601 2023-01-21 10:04:28.722197: step: 216/529, loss: 0.022513294592499733 2023-01-21 10:04:29.834446: step: 220/529, loss: 0.07244625687599182 2023-01-21 10:04:30.955482: step: 224/529, loss: 0.027306701987981796 2023-01-21 10:04:32.111190: step: 228/529, loss: 0.13446694612503052 2023-01-21 10:04:33.255375: step: 232/529, loss: 0.09044504910707474 2023-01-21 10:04:34.370147: step: 236/529, loss: 0.0581950917840004 2023-01-21 10:04:35.482578: step: 240/529, loss: 0.016605664044618607 2023-01-21 10:04:36.625411: step: 244/529, loss: 0.047232963144779205 2023-01-21 10:04:37.779670: step: 248/529, loss: 0.10404881089925766 2023-01-21 10:04:38.908196: step: 252/529, loss: 0.03578891605138779 2023-01-21 10:04:40.061405: step: 256/529, loss: 0.07694482803344727 2023-01-21 10:04:41.200145: step: 260/529, loss: 0.37140488624572754 2023-01-21 10:04:42.359615: step: 264/529, loss: 1.439310073852539 2023-01-21 10:04:43.491263: step: 268/529, loss: 0.02712078206241131 2023-01-21 10:04:44.601482: step: 272/529, loss: 0.034998226910829544 2023-01-21 10:04:45.729430: step: 276/529, loss: 0.011753225699067116 2023-01-21 10:04:46.854880: step: 280/529, loss: 0.21423006057739258 2023-01-21 10:04:48.007123: step: 284/529, loss: 0.1374889314174652 2023-01-21 10:04:49.101653: step: 288/529, loss: 0.024491310119628906 2023-01-21 10:04:50.238389: step: 292/529, loss: 0.005165863316506147 2023-01-21 10:04:51.367242: step: 296/529, loss: 0.1666305661201477 2023-01-21 10:04:52.491181: step: 300/529, loss: 0.012852216139435768 2023-01-21 10:04:53.627303: step: 304/529, loss: 0.03822169452905655 2023-01-21 10:04:54.766962: step: 308/529, loss: 0.06178493797779083 2023-01-21 10:04:55.917962: step: 312/529, loss: 0.04464082792401314 2023-01-21 10:04:57.080810: step: 316/529, loss: 0.03887185826897621 2023-01-21 10:04:58.239237: step: 320/529, loss: 0.07088928669691086 2023-01-21 10:04:59.370558: step: 324/529, loss: 0.06082601472735405 2023-01-21 10:05:00.515189: step: 328/529, loss: 0.6549112200737 2023-01-21 10:05:01.659172: step: 332/529, loss: 0.34678858518600464 2023-01-21 10:05:02.776126: step: 336/529, loss: 0.041258908808231354 2023-01-21 10:05:03.919000: step: 340/529, loss: 0.08799009025096893 2023-01-21 10:05:05.042300: step: 344/529, loss: 0.15504541993141174 2023-01-21 10:05:06.182892: step: 348/529, loss: 0.13681164383888245 2023-01-21 10:05:07.363759: step: 352/529, loss: 0.05576801300048828 2023-01-21 10:05:08.479589: step: 356/529, loss: 0.1775272786617279 2023-01-21 10:05:09.621623: step: 360/529, loss: 0.07686281204223633 2023-01-21 10:05:10.768461: step: 364/529, loss: 0.03118123859167099 2023-01-21 10:05:11.929773: step: 368/529, loss: 0.5335355997085571 2023-01-21 10:05:13.085820: step: 372/529, loss: 0.13557516038417816 2023-01-21 10:05:14.260722: step: 376/529, loss: 1.4577245712280273 2023-01-21 10:05:15.392460: step: 380/529, loss: 0.004966163542121649 2023-01-21 10:05:16.540560: step: 384/529, loss: 0.05156436562538147 2023-01-21 10:05:17.690584: step: 388/529, loss: 0.07223444432020187 2023-01-21 10:05:18.810197: step: 392/529, loss: 0.05274343490600586 2023-01-21 10:05:19.936601: step: 396/529, loss: 0.11118907481431961 2023-01-21 10:05:21.065390: step: 400/529, loss: 0.010439014062285423 2023-01-21 10:05:22.212309: step: 404/529, loss: 0.05379696190357208 2023-01-21 10:05:23.340603: step: 408/529, loss: 0.02378091961145401 2023-01-21 10:05:24.505627: step: 412/529, loss: 0.08376821875572205 2023-01-21 10:05:25.666675: step: 416/529, loss: 0.08409491181373596 2023-01-21 10:05:26.802523: step: 420/529, loss: 0.04938249662518501 2023-01-21 10:05:27.945829: step: 424/529, loss: 0.09655790030956268 2023-01-21 10:05:29.112959: step: 428/529, loss: 0.075322724878788 2023-01-21 10:05:30.228586: step: 432/529, loss: 0.15176595747470856 2023-01-21 10:05:31.365169: step: 436/529, loss: 0.04056701809167862 2023-01-21 10:05:32.523994: step: 440/529, loss: 0.08368691802024841 2023-01-21 10:05:33.669764: step: 444/529, loss: 0.01739966869354248 2023-01-21 10:05:34.837707: step: 448/529, loss: 0.029403090476989746 2023-01-21 10:05:35.961960: step: 452/529, loss: 0.07746572047472 2023-01-21 10:05:37.093862: step: 456/529, loss: 0.025560760870575905 2023-01-21 10:05:38.215813: step: 460/529, loss: 0.056169889867305756 2023-01-21 10:05:39.332561: step: 464/529, loss: 0.07014112174510956 2023-01-21 10:05:40.469086: step: 468/529, loss: 0.06594792008399963 2023-01-21 10:05:41.604460: step: 472/529, loss: 0.02782931551337242 2023-01-21 10:05:42.747253: step: 476/529, loss: 0.06167278066277504 2023-01-21 10:05:43.915259: step: 480/529, loss: 0.03837576135993004 2023-01-21 10:05:45.057601: step: 484/529, loss: 0.09715328365564346 2023-01-21 10:05:46.193274: step: 488/529, loss: 0.19760794937610626 2023-01-21 10:05:47.317812: step: 492/529, loss: 0.08687610924243927 2023-01-21 10:05:48.471181: step: 496/529, loss: 0.010358810424804688 2023-01-21 10:05:49.585202: step: 500/529, loss: 0.05630359798669815 2023-01-21 10:05:50.717341: step: 504/529, loss: 0.014113283716142178 2023-01-21 10:05:51.844905: step: 508/529, loss: 0.11378459632396698 2023-01-21 10:05:52.969072: step: 512/529, loss: 0.09627313911914825 2023-01-21 10:05:54.092388: step: 516/529, loss: 0.06669721752405167 2023-01-21 10:05:55.213908: step: 520/529, loss: 0.004662740044295788 2023-01-21 10:05:56.356484: step: 524/529, loss: 0.09451689571142197 2023-01-21 10:05:57.498070: step: 528/529, loss: 0.0363583080470562 2023-01-21 10:05:58.625981: step: 532/529, loss: 0.04375305026769638 2023-01-21 10:05:59.810769: step: 536/529, loss: 0.67576003074646 2023-01-21 10:06:00.919819: step: 540/529, loss: 0.10921287536621094 2023-01-21 10:06:02.043985: step: 544/529, loss: 0.05100860819220543 2023-01-21 10:06:03.166909: step: 548/529, loss: 0.0291900634765625 2023-01-21 10:06:04.291910: step: 552/529, loss: 0.048766233026981354 2023-01-21 10:06:05.433299: step: 556/529, loss: 1.0434304475784302 2023-01-21 10:06:06.588194: step: 560/529, loss: 0.018218040466308594 2023-01-21 10:06:07.749239: step: 564/529, loss: 0.04148721694946289 2023-01-21 10:06:08.867380: step: 568/529, loss: 0.06952936947345734 2023-01-21 10:06:10.042334: step: 572/529, loss: 0.07440491020679474 2023-01-21 10:06:11.170063: step: 576/529, loss: 0.2156505584716797 2023-01-21 10:06:12.323006: step: 580/529, loss: 0.07615752518177032 2023-01-21 10:06:13.469113: step: 584/529, loss: 0.08578047901391983 2023-01-21 10:06:14.632458: step: 588/529, loss: 0.061104025691747665 2023-01-21 10:06:15.785074: step: 592/529, loss: 0.5516277551651001 2023-01-21 10:06:16.920203: step: 596/529, loss: 0.05072822794318199 2023-01-21 10:06:18.054551: step: 600/529, loss: 0.09885187447071075 2023-01-21 10:06:19.198190: step: 604/529, loss: 0.015060711652040482 2023-01-21 10:06:20.334177: step: 608/529, loss: 0.0293439868837595 2023-01-21 10:06:21.503537: step: 612/529, loss: 0.08143343776464462 2023-01-21 10:06:22.622774: step: 616/529, loss: 0.03999634087085724 2023-01-21 10:06:23.751133: step: 620/529, loss: 0.19652004539966583 2023-01-21 10:06:24.902199: step: 624/529, loss: 0.06015019491314888 2023-01-21 10:06:26.032008: step: 628/529, loss: 0.13016852736473083 2023-01-21 10:06:27.166571: step: 632/529, loss: 0.19950413703918457 2023-01-21 10:06:28.311492: step: 636/529, loss: 0.01291036605834961 2023-01-21 10:06:29.447247: step: 640/529, loss: 0.0941803902387619 2023-01-21 10:06:30.568193: step: 644/529, loss: 0.16976594924926758 2023-01-21 10:06:31.680840: step: 648/529, loss: 0.08327942341566086 2023-01-21 10:06:32.827528: step: 652/529, loss: 0.12168264389038086 2023-01-21 10:06:33.963946: step: 656/529, loss: 0.41181421279907227 2023-01-21 10:06:35.112154: step: 660/529, loss: 0.11356215178966522 2023-01-21 10:06:36.266196: step: 664/529, loss: 0.21660451591014862 2023-01-21 10:06:37.397122: step: 668/529, loss: 0.11194296181201935 2023-01-21 10:06:38.531620: step: 672/529, loss: 0.021971512585878372 2023-01-21 10:06:39.670634: step: 676/529, loss: 0.19974097609519958 2023-01-21 10:06:40.818846: step: 680/529, loss: 0.07890134304761887 2023-01-21 10:06:41.953570: step: 684/529, loss: 0.03074016608297825 2023-01-21 10:06:43.080045: step: 688/529, loss: 0.0827782154083252 2023-01-21 10:06:44.249998: step: 692/529, loss: 0.004318189341574907 2023-01-21 10:06:45.407356: step: 696/529, loss: 0.009356451220810413 2023-01-21 10:06:46.526440: step: 700/529, loss: 0.035997629165649414 2023-01-21 10:06:47.658015: step: 704/529, loss: 0.030234530568122864 2023-01-21 10:06:48.781576: step: 708/529, loss: 0.11812038719654083 2023-01-21 10:06:49.939726: step: 712/529, loss: 0.0879218652844429 2023-01-21 10:06:51.079105: step: 716/529, loss: 0.09496765583753586 2023-01-21 10:06:52.238162: step: 720/529, loss: 0.017397737130522728 2023-01-21 10:06:53.354686: step: 724/529, loss: 0.018061447888612747 2023-01-21 10:06:54.493918: step: 728/529, loss: 0.16862984001636505 2023-01-21 10:06:55.632034: step: 732/529, loss: 0.1033288985490799 2023-01-21 10:06:56.783560: step: 736/529, loss: 0.07814265042543411 2023-01-21 10:06:57.909764: step: 740/529, loss: 0.0056092264130711555 2023-01-21 10:06:59.117805: step: 744/529, loss: 0.10409178584814072 2023-01-21 10:07:00.242392: step: 748/529, loss: 0.2466103583574295 2023-01-21 10:07:01.368437: step: 752/529, loss: 0.05389976501464844 2023-01-21 10:07:02.483994: step: 756/529, loss: 0.02763204649090767 2023-01-21 10:07:03.638512: step: 760/529, loss: 0.014043103903532028 2023-01-21 10:07:04.784121: step: 764/529, loss: 0.03315000608563423 2023-01-21 10:07:05.900630: step: 768/529, loss: 0.045502278953790665 2023-01-21 10:07:07.027580: step: 772/529, loss: 0.004908704664558172 2023-01-21 10:07:08.158318: step: 776/529, loss: 0.1298997849225998 2023-01-21 10:07:09.273195: step: 780/529, loss: 0.14381685853004456 2023-01-21 10:07:10.435386: step: 784/529, loss: 0.23429670929908752 2023-01-21 10:07:11.574857: step: 788/529, loss: 0.1267426460981369 2023-01-21 10:07:12.754437: step: 792/529, loss: 0.16323474049568176 2023-01-21 10:07:13.871528: step: 796/529, loss: 0.07715859264135361 2023-01-21 10:07:15.024097: step: 800/529, loss: 0.19973106682300568 2023-01-21 10:07:16.177241: step: 804/529, loss: 0.06991338729858398 2023-01-21 10:07:17.329233: step: 808/529, loss: 0.1399039775133133 2023-01-21 10:07:18.469462: step: 812/529, loss: 0.06861171871423721 2023-01-21 10:07:19.596455: step: 816/529, loss: 0.011025072075426579 2023-01-21 10:07:20.713431: step: 820/529, loss: 0.26301050186157227 2023-01-21 10:07:21.816011: step: 824/529, loss: 0.1727457046508789 2023-01-21 10:07:22.945030: step: 828/529, loss: 0.20113487541675568 2023-01-21 10:07:24.096327: step: 832/529, loss: 0.0834382027387619 2023-01-21 10:07:25.243614: step: 836/529, loss: 0.045196499675512314 2023-01-21 10:07:26.396723: step: 840/529, loss: 0.0984107106924057 2023-01-21 10:07:27.500010: step: 844/529, loss: 0.0543401725590229 2023-01-21 10:07:28.648254: step: 848/529, loss: 0.08001022785902023 2023-01-21 10:07:29.781535: step: 852/529, loss: 0.12925682961940765 2023-01-21 10:07:30.902443: step: 856/529, loss: 0.014968682080507278 2023-01-21 10:07:32.036134: step: 860/529, loss: 0.026435567066073418 2023-01-21 10:07:33.187755: step: 864/529, loss: 0.14768782258033752 2023-01-21 10:07:34.321499: step: 868/529, loss: 0.14447613060474396 2023-01-21 10:07:35.489940: step: 872/529, loss: 0.056848764419555664 2023-01-21 10:07:36.641841: step: 876/529, loss: 0.05720815435051918 2023-01-21 10:07:37.792675: step: 880/529, loss: 0.029929781332612038 2023-01-21 10:07:38.937105: step: 884/529, loss: 0.06685766577720642 2023-01-21 10:07:40.077402: step: 888/529, loss: 0.15352869033813477 2023-01-21 10:07:41.205290: step: 892/529, loss: 0.08006639778614044 2023-01-21 10:07:42.348756: step: 896/529, loss: 0.0016183853149414062 2023-01-21 10:07:43.462468: step: 900/529, loss: 0.043326523154973984 2023-01-21 10:07:44.570002: step: 904/529, loss: 0.12937259674072266 2023-01-21 10:07:45.720689: step: 908/529, loss: 0.06569299101829529 2023-01-21 10:07:46.895982: step: 912/529, loss: 0.016637422144412994 2023-01-21 10:07:48.020489: step: 916/529, loss: 0.08480334281921387 2023-01-21 10:07:49.157255: step: 920/529, loss: 0.05195445939898491 2023-01-21 10:07:50.273616: step: 924/529, loss: 0.028085611760616302 2023-01-21 10:07:51.421187: step: 928/529, loss: 0.023533152416348457 2023-01-21 10:07:52.556532: step: 932/529, loss: 0.017978288233280182 2023-01-21 10:07:53.723644: step: 936/529, loss: 0.02107391320168972 2023-01-21 10:07:54.849106: step: 940/529, loss: 0.046195317059755325 2023-01-21 10:07:55.984167: step: 944/529, loss: 0.004907131195068359 2023-01-21 10:07:57.125910: step: 948/529, loss: 0.08358898013830185 2023-01-21 10:07:58.272640: step: 952/529, loss: 0.04867877811193466 2023-01-21 10:07:59.393487: step: 956/529, loss: 0.1325804740190506 2023-01-21 10:08:00.536320: step: 960/529, loss: 0.2596103549003601 2023-01-21 10:08:01.684441: step: 964/529, loss: 0.10032215714454651 2023-01-21 10:08:02.812206: step: 968/529, loss: 0.01120071392506361 2023-01-21 10:08:03.963640: step: 972/529, loss: 0.141401007771492 2023-01-21 10:08:05.089892: step: 976/529, loss: 0.20454931259155273 2023-01-21 10:08:06.217460: step: 980/529, loss: 0.00998315867036581 2023-01-21 10:08:07.343107: step: 984/529, loss: 0.024337315931916237 2023-01-21 10:08:08.482642: step: 988/529, loss: 0.21922969818115234 2023-01-21 10:08:09.632090: step: 992/529, loss: 0.016145039349794388 2023-01-21 10:08:10.762774: step: 996/529, loss: 0.04762573540210724 2023-01-21 10:08:11.935332: step: 1000/529, loss: 0.0537199042737484 2023-01-21 10:08:13.073070: step: 1004/529, loss: 0.04943258687853813 2023-01-21 10:08:14.213067: step: 1008/529, loss: 0.028740502893924713 2023-01-21 10:08:15.335842: step: 1012/529, loss: 0.21890640258789062 2023-01-21 10:08:16.486010: step: 1016/529, loss: 0.13252533972263336 2023-01-21 10:08:17.648575: step: 1020/529, loss: 0.05497632175683975 2023-01-21 10:08:18.807181: step: 1024/529, loss: 0.07800378650426865 2023-01-21 10:08:19.953044: step: 1028/529, loss: 0.02326517179608345 2023-01-21 10:08:21.079558: step: 1032/529, loss: 0.00707588205114007 2023-01-21 10:08:22.207088: step: 1036/529, loss: 0.2861449718475342 2023-01-21 10:08:23.341513: step: 1040/529, loss: 0.020989418029785156 2023-01-21 10:08:24.459572: step: 1044/529, loss: 0.07055123150348663 2023-01-21 10:08:25.586387: step: 1048/529, loss: 0.10883531719446182 2023-01-21 10:08:26.708456: step: 1052/529, loss: 0.018304063007235527 2023-01-21 10:08:27.862205: step: 1056/529, loss: 0.7230802774429321 2023-01-21 10:08:28.990847: step: 1060/529, loss: 0.14433537423610687 2023-01-21 10:08:30.105903: step: 1064/529, loss: 0.0849699079990387 2023-01-21 10:08:31.218174: step: 1068/529, loss: 0.2437925785779953 2023-01-21 10:08:32.352179: step: 1072/529, loss: 0.09633298218250275 2023-01-21 10:08:33.506780: step: 1076/529, loss: 0.0748809278011322 2023-01-21 10:08:34.647459: step: 1080/529, loss: 0.051718711853027344 2023-01-21 10:08:35.788904: step: 1084/529, loss: 0.1578083038330078 2023-01-21 10:08:36.929523: step: 1088/529, loss: 0.01725740358233452 2023-01-21 10:08:38.076748: step: 1092/529, loss: 0.1236649602651596 2023-01-21 10:08:39.227061: step: 1096/529, loss: 0.0022564411628991365 2023-01-21 10:08:40.355655: step: 1100/529, loss: 0.03927011415362358 2023-01-21 10:08:41.498859: step: 1104/529, loss: 0.12160759419202805 2023-01-21 10:08:42.645210: step: 1108/529, loss: 0.013162421993911266 2023-01-21 10:08:43.813193: step: 1112/529, loss: 0.045484352856874466 2023-01-21 10:08:44.924810: step: 1116/529, loss: 0.04290657117962837 2023-01-21 10:08:46.063437: step: 1120/529, loss: 0.08926299214363098 2023-01-21 10:08:47.210574: step: 1124/529, loss: 0.04589691385626793 2023-01-21 10:08:48.346423: step: 1128/529, loss: 0.05873260274529457 2023-01-21 10:08:49.499101: step: 1132/529, loss: 0.4263211190700531 2023-01-21 10:08:50.649630: step: 1136/529, loss: 0.022725248709321022 2023-01-21 10:08:51.812551: step: 1140/529, loss: 0.08342447131872177 2023-01-21 10:08:52.928082: step: 1144/529, loss: 0.05119381099939346 2023-01-21 10:08:54.071535: step: 1148/529, loss: 0.0746581107378006 2023-01-21 10:08:55.201863: step: 1152/529, loss: 0.012243843637406826 2023-01-21 10:08:56.326462: step: 1156/529, loss: 0.017125988379120827 2023-01-21 10:08:57.494127: step: 1160/529, loss: 0.12133846431970596 2023-01-21 10:08:58.630730: step: 1164/529, loss: 0.1178840696811676 2023-01-21 10:08:59.786040: step: 1168/529, loss: 0.028528405353426933 2023-01-21 10:09:00.929650: step: 1172/529, loss: 0.11759261786937714 2023-01-21 10:09:02.055456: step: 1176/529, loss: 0.003613757900893688 2023-01-21 10:09:03.209631: step: 1180/529, loss: 0.049944546073675156 2023-01-21 10:09:04.359318: step: 1184/529, loss: 0.023375703021883965 2023-01-21 10:09:05.495737: step: 1188/529, loss: 0.10626917332410812 2023-01-21 10:09:06.632972: step: 1192/529, loss: 0.32783156633377075 2023-01-21 10:09:07.765025: step: 1196/529, loss: 0.0784115344285965 2023-01-21 10:09:08.934041: step: 1200/529, loss: 0.3225267231464386 2023-01-21 10:09:10.067863: step: 1204/529, loss: 0.04196042940020561 2023-01-21 10:09:11.219681: step: 1208/529, loss: 0.10783758759498596 2023-01-21 10:09:12.398046: step: 1212/529, loss: 0.035337068140506744 2023-01-21 10:09:13.538626: step: 1216/529, loss: 0.37195196747779846 2023-01-21 10:09:14.668035: step: 1220/529, loss: 0.11980586498975754 2023-01-21 10:09:15.831746: step: 1224/529, loss: 0.09923744201660156 2023-01-21 10:09:16.981215: step: 1228/529, loss: 0.05331268534064293 2023-01-21 10:09:18.080805: step: 1232/529, loss: 0.06574630737304688 2023-01-21 10:09:19.207675: step: 1236/529, loss: 0.12260589748620987 2023-01-21 10:09:20.325379: step: 1240/529, loss: 0.14013658463954926 2023-01-21 10:09:21.448284: step: 1244/529, loss: 0.10244853794574738 2023-01-21 10:09:22.589745: step: 1248/529, loss: 0.05360110104084015 2023-01-21 10:09:23.735385: step: 1252/529, loss: 0.11024399101734161 2023-01-21 10:09:24.872451: step: 1256/529, loss: 0.10939264297485352 2023-01-21 10:09:26.001534: step: 1260/529, loss: 0.2813161015510559 2023-01-21 10:09:27.147969: step: 1264/529, loss: 0.4367527961730957 2023-01-21 10:09:28.295155: step: 1268/529, loss: 0.14290466904640198 2023-01-21 10:09:29.433558: step: 1272/529, loss: 0.08297085762023926 2023-01-21 10:09:30.592532: step: 1276/529, loss: 0.06573257595300674 2023-01-21 10:09:31.772038: step: 1280/529, loss: 0.11522331833839417 2023-01-21 10:09:32.933520: step: 1284/529, loss: 0.09795427322387695 2023-01-21 10:09:34.077892: step: 1288/529, loss: 0.02776818349957466 2023-01-21 10:09:35.240670: step: 1292/529, loss: 0.0947689414024353 2023-01-21 10:09:36.380689: step: 1296/529, loss: 0.169599249958992 2023-01-21 10:09:37.504254: step: 1300/529, loss: 0.04477834701538086 2023-01-21 10:09:38.665468: step: 1304/529, loss: 0.0932869017124176 2023-01-21 10:09:39.823760: step: 1308/529, loss: 0.00855474453419447 2023-01-21 10:09:40.987453: step: 1312/529, loss: 0.028590798377990723 2023-01-21 10:09:42.183656: step: 1316/529, loss: 0.08530931919813156 2023-01-21 10:09:43.332722: step: 1320/529, loss: 0.020514965057373047 2023-01-21 10:09:44.487658: step: 1324/529, loss: 0.045728109776973724 2023-01-21 10:09:45.633541: step: 1328/529, loss: 0.07422919571399689 2023-01-21 10:09:46.764761: step: 1332/529, loss: 0.027492430061101913 2023-01-21 10:09:47.896255: step: 1336/529, loss: 0.16383209824562073 2023-01-21 10:09:49.019534: step: 1340/529, loss: 0.034011900424957275 2023-01-21 10:09:50.168044: step: 1344/529, loss: 0.04374532774090767 2023-01-21 10:09:51.290647: step: 1348/529, loss: 0.06766939163208008 2023-01-21 10:09:52.434077: step: 1352/529, loss: 0.043773747980594635 2023-01-21 10:09:53.575314: step: 1356/529, loss: 0.08694286644458771 2023-01-21 10:09:54.707430: step: 1360/529, loss: 0.05438380315899849 2023-01-21 10:09:55.848800: step: 1364/529, loss: 0.05547008663415909 2023-01-21 10:09:57.003865: step: 1368/529, loss: 0.07337551563978195 2023-01-21 10:09:58.161747: step: 1372/529, loss: 0.05067472532391548 2023-01-21 10:09:59.279576: step: 1376/529, loss: 0.03581256791949272 2023-01-21 10:10:00.424104: step: 1380/529, loss: 0.009047317318618298 2023-01-21 10:10:01.553802: step: 1384/529, loss: 0.01960754580795765 2023-01-21 10:10:02.689137: step: 1388/529, loss: 0.07275829464197159 2023-01-21 10:10:03.807300: step: 1392/529, loss: 0.04798593744635582 2023-01-21 10:10:04.941458: step: 1396/529, loss: 0.08816394954919815 2023-01-21 10:10:06.081621: step: 1400/529, loss: 0.15449228882789612 2023-01-21 10:10:07.205737: step: 1404/529, loss: 0.02014179341495037 2023-01-21 10:10:08.335949: step: 1408/529, loss: 0.10159559547901154 2023-01-21 10:10:09.476602: step: 1412/529, loss: 0.026207923889160156 2023-01-21 10:10:10.605541: step: 1416/529, loss: 0.03822488710284233 2023-01-21 10:10:11.740170: step: 1420/529, loss: 0.16318555176258087 2023-01-21 10:10:12.905857: step: 1424/529, loss: 0.009918594732880592 2023-01-21 10:10:14.035270: step: 1428/529, loss: 0.026558782905340195 2023-01-21 10:10:15.147465: step: 1432/529, loss: 0.03461778163909912 2023-01-21 10:10:16.267965: step: 1436/529, loss: 0.11733609437942505 2023-01-21 10:10:17.388945: step: 1440/529, loss: 0.07947101444005966 2023-01-21 10:10:18.540503: step: 1444/529, loss: 0.0937531441450119 2023-01-21 10:10:19.688770: step: 1448/529, loss: 0.08047189563512802 2023-01-21 10:10:20.821725: step: 1452/529, loss: 0.007455253973603249 2023-01-21 10:10:21.969986: step: 1456/529, loss: 0.15630942583084106 2023-01-21 10:10:23.128198: step: 1460/529, loss: 0.09452686458826065 2023-01-21 10:10:24.273734: step: 1464/529, loss: 0.0605742409825325 2023-01-21 10:10:25.383686: step: 1468/529, loss: 0.10759225487709045 2023-01-21 10:10:26.509599: step: 1472/529, loss: 0.4065554141998291 2023-01-21 10:10:27.641770: step: 1476/529, loss: 0.08842973411083221 2023-01-21 10:10:28.753946: step: 1480/529, loss: 0.6630769371986389 2023-01-21 10:10:29.900437: step: 1484/529, loss: 0.01212625578045845 2023-01-21 10:10:31.011145: step: 1488/529, loss: 0.07003974914550781 2023-01-21 10:10:32.162616: step: 1492/529, loss: 0.06025858223438263 2023-01-21 10:10:33.298295: step: 1496/529, loss: 0.08622658252716064 2023-01-21 10:10:34.437913: step: 1500/529, loss: 0.010400105267763138 2023-01-21 10:10:35.594119: step: 1504/529, loss: 0.4117165803909302 2023-01-21 10:10:36.726165: step: 1508/529, loss: 0.028048038482666016 2023-01-21 10:10:37.889475: step: 1512/529, loss: 0.09052066504955292 2023-01-21 10:10:38.991125: step: 1516/529, loss: 0.02193927764892578 2023-01-21 10:10:40.146336: step: 1520/529, loss: 0.027385331690311432 2023-01-21 10:10:41.276495: step: 1524/529, loss: 0.0029133083298802376 2023-01-21 10:10:42.419434: step: 1528/529, loss: 0.10590820014476776 2023-01-21 10:10:43.523046: step: 1532/529, loss: 0.0780634880065918 2023-01-21 10:10:44.643855: step: 1536/529, loss: 0.007105112075805664 2023-01-21 10:10:45.782854: step: 1540/529, loss: 0.06804361194372177 2023-01-21 10:10:46.914967: step: 1544/529, loss: 0.08576307445764542 2023-01-21 10:10:48.048628: step: 1548/529, loss: 0.042864322662353516 2023-01-21 10:10:49.205368: step: 1552/529, loss: 0.0626750960946083 2023-01-21 10:10:50.334592: step: 1556/529, loss: 0.058167554438114166 2023-01-21 10:10:51.470705: step: 1560/529, loss: 0.03959827497601509 2023-01-21 10:10:52.589072: step: 1564/529, loss: 0.009998607449233532 2023-01-21 10:10:53.733173: step: 1568/529, loss: 0.16010750830173492 2023-01-21 10:10:54.881032: step: 1572/529, loss: 0.05658111721277237 2023-01-21 10:10:55.996980: step: 1576/529, loss: 0.0739203467965126 2023-01-21 10:10:57.128612: step: 1580/529, loss: 0.10778641700744629 2023-01-21 10:10:58.245708: step: 1584/529, loss: 0.0023717046715319157 2023-01-21 10:10:59.384310: step: 1588/529, loss: 0.0529102087020874 2023-01-21 10:11:00.506871: step: 1592/529, loss: 0.09331293404102325 2023-01-21 10:11:01.655986: step: 1596/529, loss: 0.036615852266550064 2023-01-21 10:11:02.787782: step: 1600/529, loss: 0.0989147201180458 2023-01-21 10:11:03.921850: step: 1604/529, loss: 0.14255276322364807 2023-01-21 10:11:05.063486: step: 1608/529, loss: 0.11164913326501846 2023-01-21 10:11:06.216421: step: 1612/529, loss: 0.36577796936035156 2023-01-21 10:11:07.355400: step: 1616/529, loss: 0.27770549058914185 2023-01-21 10:11:08.502528: step: 1620/529, loss: 0.07905597239732742 2023-01-21 10:11:09.617527: step: 1624/529, loss: 0.11833696067333221 2023-01-21 10:11:10.751890: step: 1628/529, loss: 0.06645546108484268 2023-01-21 10:11:11.912080: step: 1632/529, loss: 0.01667703501880169 2023-01-21 10:11:13.020622: step: 1636/529, loss: 0.04961814731359482 2023-01-21 10:11:14.140709: step: 1640/529, loss: 0.028949355706572533 2023-01-21 10:11:15.270414: step: 1644/529, loss: 0.5641604661941528 2023-01-21 10:11:16.421485: step: 1648/529, loss: 0.0822833999991417 2023-01-21 10:11:17.594274: step: 1652/529, loss: 0.12084569782018661 2023-01-21 10:11:18.753452: step: 1656/529, loss: 0.08424148708581924 2023-01-21 10:11:19.918034: step: 1660/529, loss: 0.24304667115211487 2023-01-21 10:11:21.074650: step: 1664/529, loss: 0.04488034173846245 2023-01-21 10:11:22.209691: step: 1668/529, loss: 0.030668700113892555 2023-01-21 10:11:23.369367: step: 1672/529, loss: 0.11166906356811523 2023-01-21 10:11:24.493715: step: 1676/529, loss: 0.05258913338184357 2023-01-21 10:11:25.614837: step: 1680/529, loss: 0.03966110944747925 2023-01-21 10:11:26.736607: step: 1684/529, loss: 0.05706949159502983 2023-01-21 10:11:27.870077: step: 1688/529, loss: 0.08502653241157532 2023-01-21 10:11:29.002558: step: 1692/529, loss: 0.012454413808882236 2023-01-21 10:11:30.125488: step: 1696/529, loss: 0.12608790397644043 2023-01-21 10:11:31.281150: step: 1700/529, loss: 0.062413401901721954 2023-01-21 10:11:32.430068: step: 1704/529, loss: 0.012491869740188122 2023-01-21 10:11:33.559227: step: 1708/529, loss: 0.06444601714611053 2023-01-21 10:11:34.699854: step: 1712/529, loss: 0.10224554687738419 2023-01-21 10:11:35.862076: step: 1716/529, loss: 0.2857908308506012 2023-01-21 10:11:37.022554: step: 1720/529, loss: 0.02364215813577175 2023-01-21 10:11:38.257458: step: 1724/529, loss: 0.050722405314445496 2023-01-21 10:11:39.407837: step: 1728/529, loss: 0.05851469188928604 2023-01-21 10:11:40.550035: step: 1732/529, loss: 0.036136794835329056 2023-01-21 10:11:41.692493: step: 1736/529, loss: 0.005303668789565563 2023-01-21 10:11:42.817190: step: 1740/529, loss: 0.07457681000232697 2023-01-21 10:11:43.964591: step: 1744/529, loss: 0.018906213343143463 2023-01-21 10:11:45.125838: step: 1748/529, loss: 0.10388734191656113 2023-01-21 10:11:46.250700: step: 1752/529, loss: 0.04726286232471466 2023-01-21 10:11:47.380045: step: 1756/529, loss: 0.01022949256002903 2023-01-21 10:11:48.524083: step: 1760/529, loss: 0.06159830093383789 2023-01-21 10:11:49.654490: step: 1764/529, loss: 0.1197529286146164 2023-01-21 10:11:50.771985: step: 1768/529, loss: 0.0792245864868164 2023-01-21 10:11:51.901394: step: 1772/529, loss: 0.02394075319170952 2023-01-21 10:11:53.057239: step: 1776/529, loss: 0.08460597693920135 2023-01-21 10:11:54.174703: step: 1780/529, loss: 0.04512443393468857 2023-01-21 10:11:55.321140: step: 1784/529, loss: 0.05376272276043892 2023-01-21 10:11:56.475463: step: 1788/529, loss: 0.10175077617168427 2023-01-21 10:11:57.617893: step: 1792/529, loss: 0.006239033304154873 2023-01-21 10:11:58.724822: step: 1796/529, loss: 0.024085475131869316 2023-01-21 10:11:59.911147: step: 1800/529, loss: 0.12385706603527069 2023-01-21 10:12:01.078903: step: 1804/529, loss: 0.1836889386177063 2023-01-21 10:12:02.194617: step: 1808/529, loss: 0.013002490624785423 2023-01-21 10:12:03.326366: step: 1812/529, loss: 0.2278801053762436 2023-01-21 10:12:04.471175: step: 1816/529, loss: 0.07473068684339523 2023-01-21 10:12:05.632017: step: 1820/529, loss: 0.16262513399124146 2023-01-21 10:12:06.785464: step: 1824/529, loss: 0.031132889911532402 2023-01-21 10:12:07.921759: step: 1828/529, loss: 0.018962860107421875 2023-01-21 10:12:09.036159: step: 1832/529, loss: 0.1415359079837799 2023-01-21 10:12:10.171987: step: 1836/529, loss: 0.011631202884018421 2023-01-21 10:12:11.314234: step: 1840/529, loss: 0.009527111425995827 2023-01-21 10:12:12.480082: step: 1844/529, loss: 0.06211052089929581 2023-01-21 10:12:13.604068: step: 1848/529, loss: 0.02534055896103382 2023-01-21 10:12:14.744261: step: 1852/529, loss: 0.056978702545166016 2023-01-21 10:12:15.873035: step: 1856/529, loss: 0.03632063791155815 2023-01-21 10:12:17.036371: step: 1860/529, loss: 0.6903088688850403 2023-01-21 10:12:18.156381: step: 1864/529, loss: 0.13236571848392487 2023-01-21 10:12:19.289313: step: 1868/529, loss: 0.04093503952026367 2023-01-21 10:12:20.410660: step: 1872/529, loss: 0.09035070240497589 2023-01-21 10:12:21.554216: step: 1876/529, loss: 0.1581283062696457 2023-01-21 10:12:22.688520: step: 1880/529, loss: 0.025211524218320847 2023-01-21 10:12:23.809221: step: 1884/529, loss: 0.26080742478370667 2023-01-21 10:12:24.952536: step: 1888/529, loss: 0.5973278880119324 2023-01-21 10:12:26.089385: step: 1892/529, loss: 0.04400506243109703 2023-01-21 10:12:27.245114: step: 1896/529, loss: 0.05986309424042702 2023-01-21 10:12:28.366504: step: 1900/529, loss: 0.051085613667964935 2023-01-21 10:12:29.508958: step: 1904/529, loss: 0.011987018398940563 2023-01-21 10:12:30.638806: step: 1908/529, loss: 0.15386730432510376 2023-01-21 10:12:31.784827: step: 1912/529, loss: 0.01692638359963894 2023-01-21 10:12:32.920262: step: 1916/529, loss: 0.12070226669311523 2023-01-21 10:12:34.052711: step: 1920/529, loss: 0.016756916418671608 2023-01-21 10:12:35.197212: step: 1924/529, loss: 0.022647477686405182 2023-01-21 10:12:36.374632: step: 1928/529, loss: 0.13211488723754883 2023-01-21 10:12:37.467148: step: 1932/529, loss: 0.01573462411761284 2023-01-21 10:12:38.567856: step: 1936/529, loss: 0.039135657250881195 2023-01-21 10:12:39.701415: step: 1940/529, loss: 0.18346408009529114 2023-01-21 10:12:40.835418: step: 1944/529, loss: 0.12774324417114258 2023-01-21 10:12:41.978937: step: 1948/529, loss: 0.003621053881943226 2023-01-21 10:12:43.162523: step: 1952/529, loss: 0.06587381660938263 2023-01-21 10:12:44.318869: step: 1956/529, loss: 0.0050697922706604 2023-01-21 10:12:45.479003: step: 1960/529, loss: 0.05135393515229225 2023-01-21 10:12:46.631952: step: 1964/529, loss: 0.0958009734749794 2023-01-21 10:12:47.752205: step: 1968/529, loss: 0.0972389206290245 2023-01-21 10:12:48.882312: step: 1972/529, loss: 0.02541046030819416 2023-01-21 10:12:50.033636: step: 1976/529, loss: 0.22043271362781525 2023-01-21 10:12:51.168000: step: 1980/529, loss: 0.1106606274843216 2023-01-21 10:12:52.310515: step: 1984/529, loss: 0.020216751843690872 2023-01-21 10:12:53.455459: step: 1988/529, loss: 0.02553238905966282 2023-01-21 10:12:54.585767: step: 1992/529, loss: 0.13589835166931152 2023-01-21 10:12:55.752875: step: 1996/529, loss: 0.07527312636375427 2023-01-21 10:12:56.881340: step: 2000/529, loss: 0.010635661892592907 2023-01-21 10:12:57.989156: step: 2004/529, loss: 0.06710109859704971 2023-01-21 10:12:59.120352: step: 2008/529, loss: 0.04665551334619522 2023-01-21 10:13:00.286169: step: 2012/529, loss: 0.00983352679759264 2023-01-21 10:13:01.457857: step: 2016/529, loss: 0.12732791900634766 2023-01-21 10:13:02.622575: step: 2020/529, loss: 0.038633156567811966 2023-01-21 10:13:03.786617: step: 2024/529, loss: 0.023262787610292435 2023-01-21 10:13:04.928187: step: 2028/529, loss: 0.06486359238624573 2023-01-21 10:13:06.077848: step: 2032/529, loss: 0.029662322252988815 2023-01-21 10:13:07.220003: step: 2036/529, loss: 0.08603730797767639 2023-01-21 10:13:08.355574: step: 2040/529, loss: 0.050253868103027344 2023-01-21 10:13:09.485256: step: 2044/529, loss: 0.06838864833116531 2023-01-21 10:13:10.643145: step: 2048/529, loss: 0.18926943838596344 2023-01-21 10:13:11.819613: step: 2052/529, loss: 0.05205736309289932 2023-01-21 10:13:12.935875: step: 2056/529, loss: 0.051969338208436966 2023-01-21 10:13:14.055992: step: 2060/529, loss: 0.014870643615722656 2023-01-21 10:13:15.186609: step: 2064/529, loss: 0.027440406382083893 2023-01-21 10:13:16.332145: step: 2068/529, loss: 0.004777717869728804 2023-01-21 10:13:17.502993: step: 2072/529, loss: 0.17016807198524475 2023-01-21 10:13:18.667434: step: 2076/529, loss: 0.244316965341568 2023-01-21 10:13:19.838546: step: 2080/529, loss: 0.04060692712664604 2023-01-21 10:13:20.996973: step: 2084/529, loss: 0.05943470075726509 2023-01-21 10:13:22.144200: step: 2088/529, loss: 0.1215766966342926 2023-01-21 10:13:23.269407: step: 2092/529, loss: 0.05379905924201012 2023-01-21 10:13:24.397815: step: 2096/529, loss: 0.008314895443618298 2023-01-21 10:13:25.550168: step: 2100/529, loss: 0.005596065428107977 2023-01-21 10:13:26.732655: step: 2104/529, loss: 0.08714733272790909 2023-01-21 10:13:27.908249: step: 2108/529, loss: 0.16617390513420105 2023-01-21 10:13:29.027937: step: 2112/529, loss: 0.03994948789477348 2023-01-21 10:13:30.174956: step: 2116/529, loss: 0.055564116686582565 ================================================== Loss: 0.102 -------------------- Dev: {'event': {'p': 0.6336633663366337, 'r': 0.7669773635153129, 'f1': 0.6939759036144578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Test: {'event': {'p': 0.6261325703385789, 'r': 0.7607184241019699, 'f1': 0.686895108553492}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Chinese: {'event': {'p': 0.5903614457831325, 'r': 0.9074074074074074, 'f1': 0.7153284671532847}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Korean: {'event': {'p': 0.6458333333333334, 'r': 0.49206349206349204, 'f1': 0.5585585585585586}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Russian: {'event': {'p': 0.5, 'r': 0.5277777777777778, 'f1': 0.5135135135135136}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 10:14:11.615718: step: 4/529, loss: 0.06526575237512589 2023-01-21 10:14:12.779036: step: 8/529, loss: 0.05167045444250107 2023-01-21 10:14:13.922874: step: 12/529, loss: 0.034413862973451614 2023-01-21 10:14:15.036977: step: 16/529, loss: 0.1399787962436676 2023-01-21 10:14:16.190755: step: 20/529, loss: 0.024865586310625076 2023-01-21 10:14:17.354407: step: 24/529, loss: 0.014812469482421875 2023-01-21 10:14:18.532626: step: 28/529, loss: 0.07330341637134552 2023-01-21 10:14:19.651844: step: 32/529, loss: 0.007396126165986061 2023-01-21 10:14:20.771099: step: 36/529, loss: 0.008503532037138939 2023-01-21 10:14:21.890201: step: 40/529, loss: 0.001967144198715687 2023-01-21 10:14:23.004074: step: 44/529, loss: 0.049854375422000885 2023-01-21 10:14:24.109689: step: 48/529, loss: 0.03765811771154404 2023-01-21 10:14:25.257066: step: 52/529, loss: 0.025577712804079056 2023-01-21 10:14:26.400497: step: 56/529, loss: 0.21424388885498047 2023-01-21 10:14:27.514479: step: 60/529, loss: 0.021035075187683105 2023-01-21 10:14:28.646325: step: 64/529, loss: 0.010951709933578968 2023-01-21 10:14:29.759326: step: 68/529, loss: 0.0076386453583836555 2023-01-21 10:14:30.893886: step: 72/529, loss: 0.07369747757911682 2023-01-21 10:14:32.020148: step: 76/529, loss: 0.02851381152868271 2023-01-21 10:14:33.146643: step: 80/529, loss: 0.6189826130867004 2023-01-21 10:14:34.265768: step: 84/529, loss: 0.04281141981482506 2023-01-21 10:14:35.431527: step: 88/529, loss: 0.14122648537158966 2023-01-21 10:14:36.551748: step: 92/529, loss: 0.1223832219839096 2023-01-21 10:14:37.684485: step: 96/529, loss: 0.07379761338233948 2023-01-21 10:14:38.819175: step: 100/529, loss: 0.06322650611400604 2023-01-21 10:14:39.924739: step: 104/529, loss: 0.1651563197374344 2023-01-21 10:14:41.030234: step: 108/529, loss: 0.08313963562250137 2023-01-21 10:14:42.174003: step: 112/529, loss: 0.5426965951919556 2023-01-21 10:14:43.358497: step: 116/529, loss: 0.046189069747924805 2023-01-21 10:14:44.480796: step: 120/529, loss: 0.0735151618719101 2023-01-21 10:14:45.624809: step: 124/529, loss: 0.11353117227554321 2023-01-21 10:14:46.767499: step: 128/529, loss: 0.048750780522823334 2023-01-21 10:14:47.924635: step: 132/529, loss: 0.015401840209960938 2023-01-21 10:14:49.029441: step: 136/529, loss: 0.06352796405553818 2023-01-21 10:14:50.168902: step: 140/529, loss: 0.03906898573040962 2023-01-21 10:14:51.308682: step: 144/529, loss: 0.12227649986743927 2023-01-21 10:14:52.450910: step: 148/529, loss: 0.07670503109693527 2023-01-21 10:14:53.595507: step: 152/529, loss: 0.004206561949104071 2023-01-21 10:14:54.738514: step: 156/529, loss: 0.40554171800613403 2023-01-21 10:14:55.899833: step: 160/529, loss: 0.018439387902617455 2023-01-21 10:14:57.016685: step: 164/529, loss: 0.06138668209314346 2023-01-21 10:14:58.183598: step: 168/529, loss: 0.023206520825624466 2023-01-21 10:14:59.296249: step: 172/529, loss: 9.193420555675402e-05 2023-01-21 10:15:00.441104: step: 176/529, loss: 0.07504411041736603 2023-01-21 10:15:01.564964: step: 180/529, loss: 0.07717952877283096 2023-01-21 10:15:02.702586: step: 184/529, loss: 0.004258966539055109 2023-01-21 10:15:03.856280: step: 188/529, loss: 0.0023467065766453743 2023-01-21 10:15:04.996695: step: 192/529, loss: 0.056708622723817825 2023-01-21 10:15:06.131023: step: 196/529, loss: 0.08248114585876465 2023-01-21 10:15:07.269449: step: 200/529, loss: 0.08621253818273544 2023-01-21 10:15:08.420429: step: 204/529, loss: 0.09565964341163635 2023-01-21 10:15:09.527667: step: 208/529, loss: 0.14741650223731995 2023-01-21 10:15:10.635361: step: 212/529, loss: 0.03267812728881836 2023-01-21 10:15:11.768930: step: 216/529, loss: 0.10312697291374207 2023-01-21 10:15:12.939825: step: 220/529, loss: 0.0288193728774786 2023-01-21 10:15:14.051271: step: 224/529, loss: 0.03588409349322319 2023-01-21 10:15:15.176814: step: 228/529, loss: 0.021371817216277122 2023-01-21 10:15:16.328914: step: 232/529, loss: 0.040978334844112396 2023-01-21 10:15:17.499527: step: 236/529, loss: 0.0374603271484375 2023-01-21 10:15:18.631151: step: 240/529, loss: 0.08695602416992188 2023-01-21 10:15:19.765276: step: 244/529, loss: 0.03203125298023224 2023-01-21 10:15:20.917506: step: 248/529, loss: 0.012463569641113281 2023-01-21 10:15:22.058975: step: 252/529, loss: 0.10618915408849716 2023-01-21 10:15:23.193378: step: 256/529, loss: 0.11084241420030594 2023-01-21 10:15:24.319803: step: 260/529, loss: 0.059847354888916016 2023-01-21 10:15:25.418853: step: 264/529, loss: 0.7478528022766113 2023-01-21 10:15:26.551588: step: 268/529, loss: 0.21398086845874786 2023-01-21 10:15:27.685292: step: 272/529, loss: 0.055710986256599426 2023-01-21 10:15:28.822680: step: 276/529, loss: 0.07299118489027023 2023-01-21 10:15:29.950717: step: 280/529, loss: 0.03334188461303711 2023-01-21 10:15:31.082065: step: 284/529, loss: 0.038349542766809464 2023-01-21 10:15:32.215965: step: 288/529, loss: 0.09635090827941895 2023-01-21 10:15:33.360327: step: 292/529, loss: 0.0373292937874794 2023-01-21 10:15:34.511600: step: 296/529, loss: 0.12191420048475266 2023-01-21 10:15:35.662107: step: 300/529, loss: 0.07888355106115341 2023-01-21 10:15:36.797442: step: 304/529, loss: 0.8172504305839539 2023-01-21 10:15:37.927765: step: 308/529, loss: 0.030851364135742188 2023-01-21 10:15:39.043817: step: 312/529, loss: 0.03205299377441406 2023-01-21 10:15:40.176665: step: 316/529, loss: 0.10401568561792374 2023-01-21 10:15:41.346316: step: 320/529, loss: 0.11300201714038849 2023-01-21 10:15:42.494081: step: 324/529, loss: 0.034682657569646835 2023-01-21 10:15:43.646775: step: 328/529, loss: 0.02340526506304741 2023-01-21 10:15:44.810041: step: 332/529, loss: 0.1827675849199295 2023-01-21 10:15:45.952432: step: 336/529, loss: 0.0405457504093647 2023-01-21 10:15:47.094482: step: 340/529, loss: 0.026951981708407402 2023-01-21 10:15:48.234893: step: 344/529, loss: 0.05903301388025284 2023-01-21 10:15:49.387052: step: 348/529, loss: 0.06084003672003746 2023-01-21 10:15:50.544173: step: 352/529, loss: 0.3267245292663574 2023-01-21 10:15:51.692471: step: 356/529, loss: 0.02098846435546875 2023-01-21 10:15:52.826276: step: 360/529, loss: 0.06029319763183594 2023-01-21 10:15:53.968831: step: 364/529, loss: 0.08244076371192932 2023-01-21 10:15:55.129398: step: 368/529, loss: 0.0160688403993845 2023-01-21 10:15:56.297986: step: 372/529, loss: 0.06547784805297852 2023-01-21 10:15:57.430695: step: 376/529, loss: 0.006495666690170765 2023-01-21 10:15:58.542963: step: 380/529, loss: 0.04197130352258682 2023-01-21 10:15:59.683838: step: 384/529, loss: 0.025869369506835938 2023-01-21 10:16:00.865605: step: 388/529, loss: 0.0416143462061882 2023-01-21 10:16:01.994208: step: 392/529, loss: 0.03596305847167969 2023-01-21 10:16:03.127891: step: 396/529, loss: 0.12299247086048126 2023-01-21 10:16:04.307065: step: 400/529, loss: 0.11730728298425674 2023-01-21 10:16:05.440054: step: 404/529, loss: 0.1666286438703537 2023-01-21 10:16:06.603597: step: 408/529, loss: 0.01699056662619114 2023-01-21 10:16:07.736552: step: 412/529, loss: 0.13448619842529297 2023-01-21 10:16:08.854181: step: 416/529, loss: 0.05790691450238228 2023-01-21 10:16:10.020224: step: 420/529, loss: 0.0315336212515831 2023-01-21 10:16:11.146048: step: 424/529, loss: 0.022933436557650566 2023-01-21 10:16:12.282915: step: 428/529, loss: 0.035239316523075104 2023-01-21 10:16:13.419869: step: 432/529, loss: 0.03359203413128853 2023-01-21 10:16:14.553278: step: 436/529, loss: 0.07688503712415695 2023-01-21 10:16:15.675352: step: 440/529, loss: 0.002285003662109375 2023-01-21 10:16:16.812189: step: 444/529, loss: 0.022823907434940338 2023-01-21 10:16:17.942219: step: 448/529, loss: 0.45081567764282227 2023-01-21 10:16:19.078232: step: 452/529, loss: 1.2924983501434326 2023-01-21 10:16:20.192477: step: 456/529, loss: 0.05244283378124237 2023-01-21 10:16:21.320539: step: 460/529, loss: 0.030366992577910423 2023-01-21 10:16:22.466864: step: 464/529, loss: 0.11951503902673721 2023-01-21 10:16:23.617464: step: 468/529, loss: 0.019057750701904297 2023-01-21 10:16:24.722991: step: 472/529, loss: 0.00408825883641839 2023-01-21 10:16:25.869353: step: 476/529, loss: 0.02265291102230549 2023-01-21 10:16:26.984858: step: 480/529, loss: 0.09745602309703827 2023-01-21 10:16:28.144303: step: 484/529, loss: 0.06776171177625656 2023-01-21 10:16:29.290107: step: 488/529, loss: 0.03466839715838432 2023-01-21 10:16:30.458080: step: 492/529, loss: 0.0844636932015419 2023-01-21 10:16:31.585572: step: 496/529, loss: 0.04363875836133957 2023-01-21 10:16:32.719867: step: 500/529, loss: 0.09243683516979218 2023-01-21 10:16:33.881839: step: 504/529, loss: 0.13318701088428497 2023-01-21 10:16:35.033269: step: 508/529, loss: 0.015395975671708584 2023-01-21 10:16:36.164841: step: 512/529, loss: 0.0700516328215599 2023-01-21 10:16:37.320791: step: 516/529, loss: 0.05809517204761505 2023-01-21 10:16:38.460672: step: 520/529, loss: 0.05484752357006073 2023-01-21 10:16:39.584216: step: 524/529, loss: 0.030734922736883163 2023-01-21 10:16:40.694929: step: 528/529, loss: 0.1467321515083313 2023-01-21 10:16:41.883135: step: 532/529, loss: 0.029723817482590675 2023-01-21 10:16:42.992626: step: 536/529, loss: 0.04259147495031357 2023-01-21 10:16:44.143886: step: 540/529, loss: 0.05704698711633682 2023-01-21 10:16:45.273802: step: 544/529, loss: 0.4296625256538391 2023-01-21 10:16:46.425324: step: 548/529, loss: 0.1550310254096985 2023-01-21 10:16:47.559760: step: 552/529, loss: 0.09001141041517258 2023-01-21 10:16:48.703640: step: 556/529, loss: 0.1141211986541748 2023-01-21 10:16:49.841860: step: 560/529, loss: 0.029521655291318893 2023-01-21 10:16:50.957859: step: 564/529, loss: 0.00520715769380331 2023-01-21 10:16:52.095015: step: 568/529, loss: 0.007484960835427046 2023-01-21 10:16:53.251823: step: 572/529, loss: 0.12971477210521698 2023-01-21 10:16:54.386371: step: 576/529, loss: 0.058158110827207565 2023-01-21 10:16:55.514127: step: 580/529, loss: 0.006542682647705078 2023-01-21 10:16:56.643837: step: 584/529, loss: 0.028951240703463554 2023-01-21 10:16:57.773353: step: 588/529, loss: 0.09232206642627716 2023-01-21 10:16:58.916280: step: 592/529, loss: 0.05073648318648338 2023-01-21 10:17:00.059842: step: 596/529, loss: 0.05542507395148277 2023-01-21 10:17:01.214667: step: 600/529, loss: 0.04432716593146324 2023-01-21 10:17:02.372065: step: 604/529, loss: 0.014589262194931507 2023-01-21 10:17:03.536006: step: 608/529, loss: 0.20675840973854065 2023-01-21 10:17:04.658828: step: 612/529, loss: 0.002327871508896351 2023-01-21 10:17:05.803827: step: 616/529, loss: 0.0313684456050396 2023-01-21 10:17:06.944799: step: 620/529, loss: 1.0805071592330933 2023-01-21 10:17:08.054391: step: 624/529, loss: 0.001956033753231168 2023-01-21 10:17:09.196578: step: 628/529, loss: 0.0279464740306139 2023-01-21 10:17:10.354215: step: 632/529, loss: 0.0613618865609169 2023-01-21 10:17:11.490471: step: 636/529, loss: 0.13872985541820526 2023-01-21 10:17:12.660233: step: 640/529, loss: 0.05197715759277344 2023-01-21 10:17:13.784871: step: 644/529, loss: 0.041597843170166016 2023-01-21 10:17:14.911063: step: 648/529, loss: 0.04060318320989609 2023-01-21 10:17:16.048271: step: 652/529, loss: 0.033745717257261276 2023-01-21 10:17:17.177894: step: 656/529, loss: 0.002406215760856867 2023-01-21 10:17:18.302988: step: 660/529, loss: 0.0020011900924146175 2023-01-21 10:17:19.450389: step: 664/529, loss: 0.0461977943778038 2023-01-21 10:17:20.575475: step: 668/529, loss: 0.024914361536502838 2023-01-21 10:17:21.709812: step: 672/529, loss: 0.11208336055278778 2023-01-21 10:17:22.835548: step: 676/529, loss: 0.022490788251161575 2023-01-21 10:17:23.953424: step: 680/529, loss: 0.003051948733627796 2023-01-21 10:17:25.111869: step: 684/529, loss: 0.03430178388953209 2023-01-21 10:17:26.237942: step: 688/529, loss: 0.0536319725215435 2023-01-21 10:17:27.402541: step: 692/529, loss: 0.017032720148563385 2023-01-21 10:17:28.526938: step: 696/529, loss: 0.015553189441561699 2023-01-21 10:17:29.649901: step: 700/529, loss: 0.08068780601024628 2023-01-21 10:17:30.803762: step: 704/529, loss: 0.14039373397827148 2023-01-21 10:17:31.956503: step: 708/529, loss: 0.021083546802401543 2023-01-21 10:17:33.101115: step: 712/529, loss: 0.06777191162109375 2023-01-21 10:17:34.234520: step: 716/529, loss: 0.03463554382324219 2023-01-21 10:17:35.381402: step: 720/529, loss: 0.023313330486416817 2023-01-21 10:17:36.527563: step: 724/529, loss: 0.2650325894355774 2023-01-21 10:17:37.673657: step: 728/529, loss: 0.02833719365298748 2023-01-21 10:17:38.827741: step: 732/529, loss: 0.038880396634340286 2023-01-21 10:17:39.975192: step: 736/529, loss: 0.10498104989528656 2023-01-21 10:17:41.122008: step: 740/529, loss: 0.08159196376800537 2023-01-21 10:17:42.258682: step: 744/529, loss: 0.0482330322265625 2023-01-21 10:17:43.381927: step: 748/529, loss: 0.20723971724510193 2023-01-21 10:17:44.528296: step: 752/529, loss: 0.04625587537884712 2023-01-21 10:17:45.642888: step: 756/529, loss: 0.007816123776137829 2023-01-21 10:17:46.769276: step: 760/529, loss: 0.03884997218847275 2023-01-21 10:17:47.887111: step: 764/529, loss: 0.012987899594008923 2023-01-21 10:17:49.036865: step: 768/529, loss: 0.20532551407814026 2023-01-21 10:17:50.174792: step: 772/529, loss: 0.11797638237476349 2023-01-21 10:17:51.304295: step: 776/529, loss: 0.05479174107313156 2023-01-21 10:17:52.464446: step: 780/529, loss: 0.08997063338756561 2023-01-21 10:17:53.594288: step: 784/529, loss: 0.09629566967487335 2023-01-21 10:17:54.731555: step: 788/529, loss: 0.12250480055809021 2023-01-21 10:17:55.870690: step: 792/529, loss: 0.18010644614696503 2023-01-21 10:17:56.991903: step: 796/529, loss: 0.49465686082839966 2023-01-21 10:17:58.089030: step: 800/529, loss: 0.030751466751098633 2023-01-21 10:17:59.231620: step: 804/529, loss: 0.10435283929109573 2023-01-21 10:18:00.383552: step: 808/529, loss: 0.13514499366283417 2023-01-21 10:18:01.526564: step: 812/529, loss: 0.036295510828495026 2023-01-21 10:18:02.666353: step: 816/529, loss: 0.03746400028467178 2023-01-21 10:18:03.806375: step: 820/529, loss: 0.026634838432073593 2023-01-21 10:18:04.959765: step: 824/529, loss: 0.157756045460701 2023-01-21 10:18:06.103121: step: 828/529, loss: 0.018427561968564987 2023-01-21 10:18:07.239789: step: 832/529, loss: 0.009665250778198242 2023-01-21 10:18:08.375274: step: 836/529, loss: 0.02650909498333931 2023-01-21 10:18:09.511513: step: 840/529, loss: 0.007261467166244984 2023-01-21 10:18:10.620465: step: 844/529, loss: 0.003098106477409601 2023-01-21 10:18:11.740467: step: 848/529, loss: 0.11651191860437393 2023-01-21 10:18:12.919775: step: 852/529, loss: 0.1394055187702179 2023-01-21 10:18:14.100730: step: 856/529, loss: 0.08608122169971466 2023-01-21 10:18:15.226670: step: 860/529, loss: 0.06292672455310822 2023-01-21 10:18:16.352819: step: 864/529, loss: 0.09668950736522675 2023-01-21 10:18:17.495363: step: 868/529, loss: 0.021326828747987747 2023-01-21 10:18:18.637955: step: 872/529, loss: 0.07423420250415802 2023-01-21 10:18:19.798284: step: 876/529, loss: 0.08232231438159943 2023-01-21 10:18:20.925389: step: 880/529, loss: 0.05654726177453995 2023-01-21 10:18:22.088323: step: 884/529, loss: 0.03529674932360649 2023-01-21 10:18:23.205824: step: 888/529, loss: 0.010225963778793812 2023-01-21 10:18:24.351680: step: 892/529, loss: 0.0780513808131218 2023-01-21 10:18:25.512171: step: 896/529, loss: 0.06091899797320366 2023-01-21 10:18:26.662860: step: 900/529, loss: 0.05104684829711914 2023-01-21 10:18:27.797080: step: 904/529, loss: 0.04106760025024414 2023-01-21 10:18:28.937968: step: 908/529, loss: 0.06950192898511887 2023-01-21 10:18:30.077123: step: 912/529, loss: 0.028328705579042435 2023-01-21 10:18:31.197044: step: 916/529, loss: 0.09161491692066193 2023-01-21 10:18:32.356493: step: 920/529, loss: 0.12742692232131958 2023-01-21 10:18:33.504778: step: 924/529, loss: 0.02413802035152912 2023-01-21 10:18:34.631501: step: 928/529, loss: 0.014517117291688919 2023-01-21 10:18:35.785981: step: 932/529, loss: 0.06157340854406357 2023-01-21 10:18:36.924454: step: 936/529, loss: 0.0881475955247879 2023-01-21 10:18:38.050168: step: 940/529, loss: 0.2531529366970062 2023-01-21 10:18:39.180971: step: 944/529, loss: 0.010202598758041859 2023-01-21 10:18:40.287752: step: 948/529, loss: 0.022090911865234375 2023-01-21 10:18:41.404238: step: 952/529, loss: 0.05865836516022682 2023-01-21 10:18:42.541066: step: 956/529, loss: 0.06268072128295898 2023-01-21 10:18:43.681176: step: 960/529, loss: 0.015076875686645508 2023-01-21 10:18:44.812111: step: 964/529, loss: 0.06994371116161346 2023-01-21 10:18:45.943397: step: 968/529, loss: 0.02480478212237358 2023-01-21 10:18:47.081739: step: 972/529, loss: 0.032300855964422226 2023-01-21 10:18:48.222435: step: 976/529, loss: 0.01724104955792427 2023-01-21 10:18:49.361590: step: 980/529, loss: 0.044585611671209335 2023-01-21 10:18:50.493174: step: 984/529, loss: 0.09066153317689896 2023-01-21 10:18:51.624713: step: 988/529, loss: 0.01212616078555584 2023-01-21 10:18:52.775132: step: 992/529, loss: 0.10081902146339417 2023-01-21 10:18:53.916256: step: 996/529, loss: 0.03439655154943466 2023-01-21 10:18:55.040219: step: 1000/529, loss: 0.015155363827943802 2023-01-21 10:18:56.156930: step: 1004/529, loss: 0.044470883905887604 2023-01-21 10:18:57.279081: step: 1008/529, loss: 0.43867626786231995 2023-01-21 10:18:58.422688: step: 1012/529, loss: 0.05597391352057457 2023-01-21 10:18:59.551300: step: 1016/529, loss: 0.021386291831731796 2023-01-21 10:19:00.698026: step: 1020/529, loss: 0.058826547116041183 2023-01-21 10:19:01.853683: step: 1024/529, loss: 0.09164562076330185 2023-01-21 10:19:02.967501: step: 1028/529, loss: 0.016368960961699486 2023-01-21 10:19:04.107658: step: 1032/529, loss: 0.05447404459118843 2023-01-21 10:19:05.251421: step: 1036/529, loss: 0.10788127779960632 2023-01-21 10:19:06.357780: step: 1040/529, loss: 0.03672537952661514 2023-01-21 10:19:07.493741: step: 1044/529, loss: 0.04853649437427521 2023-01-21 10:19:08.639131: step: 1048/529, loss: 0.022081661969423294 2023-01-21 10:19:09.826012: step: 1052/529, loss: 0.05111103132367134 2023-01-21 10:19:10.976347: step: 1056/529, loss: 0.41411569714546204 2023-01-21 10:19:12.107578: step: 1060/529, loss: 0.02478158473968506 2023-01-21 10:19:13.233121: step: 1064/529, loss: 0.07401961833238602 2023-01-21 10:19:14.365599: step: 1068/529, loss: 0.09418430179357529 2023-01-21 10:19:15.502356: step: 1072/529, loss: 0.01998300477862358 2023-01-21 10:19:16.627101: step: 1076/529, loss: 0.018085192888975143 2023-01-21 10:19:17.723251: step: 1080/529, loss: 0.02526579052209854 2023-01-21 10:19:18.850293: step: 1084/529, loss: 0.009394549764692783 2023-01-21 10:19:19.980197: step: 1088/529, loss: 0.008757686242461205 2023-01-21 10:19:21.110159: step: 1092/529, loss: 0.08566398918628693 2023-01-21 10:19:22.243076: step: 1096/529, loss: 0.027594996616244316 2023-01-21 10:19:23.378908: step: 1100/529, loss: 0.02775411680340767 2023-01-21 10:19:24.517194: step: 1104/529, loss: 0.039984896779060364 2023-01-21 10:19:25.684400: step: 1108/529, loss: 0.022752225399017334 2023-01-21 10:19:26.829403: step: 1112/529, loss: 0.1442001312971115 2023-01-21 10:19:27.930119: step: 1116/529, loss: 0.03279523923993111 2023-01-21 10:19:29.057369: step: 1120/529, loss: 0.05236015468835831 2023-01-21 10:19:30.180465: step: 1124/529, loss: 0.018459510058164597 2023-01-21 10:19:31.331008: step: 1128/529, loss: 0.02420501783490181 2023-01-21 10:19:32.464439: step: 1132/529, loss: 0.10935159027576447 2023-01-21 10:19:33.628673: step: 1136/529, loss: 0.10718335956335068 2023-01-21 10:19:34.759297: step: 1140/529, loss: 0.060411833226680756 2023-01-21 10:19:35.887469: step: 1144/529, loss: 0.03718376159667969 2023-01-21 10:19:36.978666: step: 1148/529, loss: 0.13251176476478577 2023-01-21 10:19:38.110615: step: 1152/529, loss: 0.01127004623413086 2023-01-21 10:19:39.227938: step: 1156/529, loss: 0.09124326705932617 2023-01-21 10:19:40.352130: step: 1160/529, loss: 0.02315673977136612 2023-01-21 10:19:41.499417: step: 1164/529, loss: 0.03303088992834091 2023-01-21 10:19:42.655736: step: 1168/529, loss: 0.1468932330608368 2023-01-21 10:19:43.787064: step: 1172/529, loss: 0.08278445899486542 2023-01-21 10:19:44.934908: step: 1176/529, loss: 0.09086084365844727 2023-01-21 10:19:46.061998: step: 1180/529, loss: 0.06497984379529953 2023-01-21 10:19:47.186326: step: 1184/529, loss: 0.08080454170703888 2023-01-21 10:19:48.334344: step: 1188/529, loss: 0.07291407883167267 2023-01-21 10:19:49.462089: step: 1192/529, loss: 0.18544892966747284 2023-01-21 10:19:50.582941: step: 1196/529, loss: 0.06635923683643341 2023-01-21 10:19:51.701862: step: 1200/529, loss: 0.014287185855209827 2023-01-21 10:19:52.823680: step: 1204/529, loss: 0.007806397043168545 2023-01-21 10:19:53.915744: step: 1208/529, loss: 0.04401030391454697 2023-01-21 10:19:55.043651: step: 1212/529, loss: 0.12251091748476028 2023-01-21 10:19:56.168302: step: 1216/529, loss: 0.10045633465051651 2023-01-21 10:19:57.291639: step: 1220/529, loss: 0.10395517945289612 2023-01-21 10:19:58.421921: step: 1224/529, loss: 0.0873747318983078 2023-01-21 10:19:59.552190: step: 1228/529, loss: 0.05305042862892151 2023-01-21 10:20:00.689818: step: 1232/529, loss: 0.1095348373055458 2023-01-21 10:20:01.816793: step: 1236/529, loss: 0.3450354337692261 2023-01-21 10:20:02.943866: step: 1240/529, loss: 0.0362754724919796 2023-01-21 10:20:04.074510: step: 1244/529, loss: 0.5497283935546875 2023-01-21 10:20:05.187132: step: 1248/529, loss: 0.06034679710865021 2023-01-21 10:20:06.308726: step: 1252/529, loss: 0.10291165858507156 2023-01-21 10:20:07.436860: step: 1256/529, loss: 0.012968205846846104 2023-01-21 10:20:08.582875: step: 1260/529, loss: 1.0556843280792236 2023-01-21 10:20:09.700955: step: 1264/529, loss: 0.11362943798303604 2023-01-21 10:20:10.824257: step: 1268/529, loss: 0.05350985378026962 2023-01-21 10:20:11.963973: step: 1272/529, loss: 0.1688673496246338 2023-01-21 10:20:13.143971: step: 1276/529, loss: 0.4436616897583008 2023-01-21 10:20:14.255800: step: 1280/529, loss: 0.028574658557772636 2023-01-21 10:20:15.369732: step: 1284/529, loss: 0.054915525019168854 2023-01-21 10:20:16.527558: step: 1288/529, loss: 0.19133606553077698 2023-01-21 10:20:17.641433: step: 1292/529, loss: 0.10890035331249237 2023-01-21 10:20:18.751634: step: 1296/529, loss: 0.03509059175848961 2023-01-21 10:20:19.906636: step: 1300/529, loss: 0.003997898194938898 2023-01-21 10:20:21.043084: step: 1304/529, loss: 0.05790267139673233 2023-01-21 10:20:22.161305: step: 1308/529, loss: 0.05169752612709999 2023-01-21 10:20:23.294159: step: 1312/529, loss: 0.04183614253997803 2023-01-21 10:20:24.420668: step: 1316/529, loss: 0.03798015043139458 2023-01-21 10:20:25.585690: step: 1320/529, loss: 0.1893543154001236 2023-01-21 10:20:26.738609: step: 1324/529, loss: 0.09197287261486053 2023-01-21 10:20:27.866458: step: 1328/529, loss: 0.04724083095788956 2023-01-21 10:20:29.004520: step: 1332/529, loss: 0.02349853515625 2023-01-21 10:20:30.148657: step: 1336/529, loss: 0.01756155490875244 2023-01-21 10:20:31.302460: step: 1340/529, loss: 0.0781540647149086 2023-01-21 10:20:32.434137: step: 1344/529, loss: 0.5248399972915649 2023-01-21 10:20:33.553252: step: 1348/529, loss: 0.05299682915210724 2023-01-21 10:20:34.701508: step: 1352/529, loss: 0.018192388117313385 2023-01-21 10:20:35.834033: step: 1356/529, loss: 0.12810096144676208 2023-01-21 10:20:36.963326: step: 1360/529, loss: 0.07665862888097763 2023-01-21 10:20:38.100447: step: 1364/529, loss: 0.04927987977862358 2023-01-21 10:20:39.223057: step: 1368/529, loss: 0.03404684364795685 2023-01-21 10:20:40.343319: step: 1372/529, loss: 0.014595890417695045 2023-01-21 10:20:41.481823: step: 1376/529, loss: 0.17629805207252502 2023-01-21 10:20:42.647441: step: 1380/529, loss: 0.10727634280920029 2023-01-21 10:20:43.780129: step: 1384/529, loss: 0.006274843588471413 2023-01-21 10:20:44.928193: step: 1388/529, loss: 0.13010425865650177 2023-01-21 10:20:46.092760: step: 1392/529, loss: 0.20049165189266205 2023-01-21 10:20:47.188756: step: 1396/529, loss: 0.0012710571754723787 2023-01-21 10:20:48.324122: step: 1400/529, loss: 0.10759425163269043 2023-01-21 10:20:49.497311: step: 1404/529, loss: 0.10323648154735565 2023-01-21 10:20:50.645086: step: 1408/529, loss: 0.029401589184999466 2023-01-21 10:20:51.771433: step: 1412/529, loss: 0.012305354699492455 2023-01-21 10:20:52.905121: step: 1416/529, loss: 0.06985431164503098 2023-01-21 10:20:54.031030: step: 1420/529, loss: 0.04644613340497017 2023-01-21 10:20:55.166684: step: 1424/529, loss: 0.03849058225750923 2023-01-21 10:20:56.311773: step: 1428/529, loss: 0.16459780931472778 2023-01-21 10:20:57.450129: step: 1432/529, loss: 0.20758318901062012 2023-01-21 10:20:58.578523: step: 1436/529, loss: 0.09304933249950409 2023-01-21 10:20:59.711761: step: 1440/529, loss: 0.01618170738220215 2023-01-21 10:21:00.840512: step: 1444/529, loss: 0.029584791511297226 2023-01-21 10:21:01.969210: step: 1448/529, loss: 0.19560737907886505 2023-01-21 10:21:03.094033: step: 1452/529, loss: 0.14860764145851135 2023-01-21 10:21:04.215429: step: 1456/529, loss: 0.024477459490299225 2023-01-21 10:21:05.336866: step: 1460/529, loss: 0.09299030900001526 2023-01-21 10:21:06.481609: step: 1464/529, loss: 0.06084957346320152 2023-01-21 10:21:07.625596: step: 1468/529, loss: 0.09184026718139648 2023-01-21 10:21:08.755209: step: 1472/529, loss: 0.02703237533569336 2023-01-21 10:21:09.901963: step: 1476/529, loss: 0.0022878171876072884 2023-01-21 10:21:11.037359: step: 1480/529, loss: 0.09968023747205734 2023-01-21 10:21:12.166970: step: 1484/529, loss: 0.0010515212779864669 2023-01-21 10:21:13.329685: step: 1488/529, loss: 0.19247189164161682 2023-01-21 10:21:14.474104: step: 1492/529, loss: 0.048993803560733795 2023-01-21 10:21:15.609093: step: 1496/529, loss: 0.059478189796209335 2023-01-21 10:21:16.720702: step: 1500/529, loss: 0.32829493284225464 2023-01-21 10:21:17.872869: step: 1504/529, loss: 0.12595443427562714 2023-01-21 10:21:19.057871: step: 1508/529, loss: 1.102855920791626 2023-01-21 10:21:20.178691: step: 1512/529, loss: 0.03773389011621475 2023-01-21 10:21:21.377734: step: 1516/529, loss: 0.006098154000937939 2023-01-21 10:21:22.491201: step: 1520/529, loss: 0.03117676079273224 2023-01-21 10:21:23.642111: step: 1524/529, loss: 0.009250831790268421 2023-01-21 10:21:24.777311: step: 1528/529, loss: 0.0743291899561882 2023-01-21 10:21:25.924938: step: 1532/529, loss: 0.13249903917312622 2023-01-21 10:21:27.070678: step: 1536/529, loss: 0.7494790554046631 2023-01-21 10:21:28.170095: step: 1540/529, loss: 0.005266284570097923 2023-01-21 10:21:29.260975: step: 1544/529, loss: 0.0011754035949707031 2023-01-21 10:21:30.399855: step: 1548/529, loss: 0.13013802468776703 2023-01-21 10:21:31.526775: step: 1552/529, loss: 0.12646155059337616 2023-01-21 10:21:32.662836: step: 1556/529, loss: 0.02957763522863388 2023-01-21 10:21:33.771313: step: 1560/529, loss: 0.1027521938085556 2023-01-21 10:21:34.914793: step: 1564/529, loss: 0.05476560443639755 2023-01-21 10:21:36.042911: step: 1568/529, loss: 0.046015169471502304 2023-01-21 10:21:37.202596: step: 1572/529, loss: 0.06575965881347656 2023-01-21 10:21:38.340937: step: 1576/529, loss: 0.14659033715724945 2023-01-21 10:21:39.500899: step: 1580/529, loss: 0.259303480386734 2023-01-21 10:21:40.614755: step: 1584/529, loss: 0.021404456347227097 2023-01-21 10:21:41.754261: step: 1588/529, loss: 0.04481077566742897 2023-01-21 10:21:42.907822: step: 1592/529, loss: 0.043441012501716614 2023-01-21 10:21:44.051701: step: 1596/529, loss: 0.05086226761341095 2023-01-21 10:21:45.186415: step: 1600/529, loss: 0.06997766345739365 2023-01-21 10:21:46.294439: step: 1604/529, loss: 0.0350494384765625 2023-01-21 10:21:47.451327: step: 1608/529, loss: 0.02022533491253853 2023-01-21 10:21:48.594025: step: 1612/529, loss: 0.04236030578613281 2023-01-21 10:21:49.722901: step: 1616/529, loss: 0.1416357159614563 2023-01-21 10:21:50.871131: step: 1620/529, loss: 0.03966255113482475 2023-01-21 10:21:52.021815: step: 1624/529, loss: 0.0459684357047081 2023-01-21 10:21:53.156553: step: 1628/529, loss: 0.02164764329791069 2023-01-21 10:21:54.284724: step: 1632/529, loss: 0.043527986854314804 2023-01-21 10:21:55.413310: step: 1636/529, loss: 0.042981650680303574 2023-01-21 10:21:56.546541: step: 1640/529, loss: 0.04893989861011505 2023-01-21 10:21:57.694539: step: 1644/529, loss: 0.3489700257778168 2023-01-21 10:21:58.816107: step: 1648/529, loss: 0.020563887432217598 2023-01-21 10:21:59.918192: step: 1652/529, loss: 0.10988236218690872 2023-01-21 10:22:01.060268: step: 1656/529, loss: 0.0028249742463231087 2023-01-21 10:22:02.179212: step: 1660/529, loss: 0.06804027408361435 2023-01-21 10:22:03.314048: step: 1664/529, loss: 0.07805585861206055 2023-01-21 10:22:04.437399: step: 1668/529, loss: 1.7727779150009155 2023-01-21 10:22:05.570293: step: 1672/529, loss: 0.027508163824677467 2023-01-21 10:22:06.684610: step: 1676/529, loss: 0.030366133898496628 2023-01-21 10:22:07.820152: step: 1680/529, loss: 0.014676189050078392 2023-01-21 10:22:08.954336: step: 1684/529, loss: 0.07325849682092667 2023-01-21 10:22:10.086708: step: 1688/529, loss: 0.0035275935661047697 2023-01-21 10:22:11.249183: step: 1692/529, loss: 0.0010804652702063322 2023-01-21 10:22:12.381913: step: 1696/529, loss: 0.08370047062635422 2023-01-21 10:22:13.498784: step: 1700/529, loss: 0.02211475372314453 2023-01-21 10:22:14.609190: step: 1704/529, loss: 0.03272400051355362 2023-01-21 10:22:15.770062: step: 1708/529, loss: 0.04789257049560547 2023-01-21 10:22:16.935463: step: 1712/529, loss: 0.03433294966816902 2023-01-21 10:22:18.063520: step: 1716/529, loss: 0.035271503031253815 2023-01-21 10:22:19.232645: step: 1720/529, loss: 0.01615114137530327 2023-01-21 10:22:20.360309: step: 1724/529, loss: 0.04921875149011612 2023-01-21 10:22:21.530737: step: 1728/529, loss: 0.3380189836025238 2023-01-21 10:22:22.722268: step: 1732/529, loss: 0.06096029281616211 2023-01-21 10:22:23.917105: step: 1736/529, loss: 0.16447143256664276 2023-01-21 10:22:25.055095: step: 1740/529, loss: 0.015569686889648438 2023-01-21 10:22:26.199999: step: 1744/529, loss: 0.04421548917889595 2023-01-21 10:22:27.353630: step: 1748/529, loss: 0.1116512268781662 2023-01-21 10:22:28.470045: step: 1752/529, loss: 0.03895087167620659 2023-01-21 10:22:29.609501: step: 1756/529, loss: 0.26146629452705383 2023-01-21 10:22:30.755457: step: 1760/529, loss: 0.0629245787858963 2023-01-21 10:22:31.930575: step: 1764/529, loss: 0.0997367799282074 2023-01-21 10:22:33.065511: step: 1768/529, loss: 0.018084336072206497 2023-01-21 10:22:34.198999: step: 1772/529, loss: 0.021899128332734108 2023-01-21 10:22:35.364403: step: 1776/529, loss: 0.11147318035364151 2023-01-21 10:22:36.532245: step: 1780/529, loss: 0.07213706523180008 2023-01-21 10:22:37.672055: step: 1784/529, loss: 0.028705166652798653 2023-01-21 10:22:38.803066: step: 1788/529, loss: 0.14554014801979065 2023-01-21 10:22:39.944822: step: 1792/529, loss: 0.10406913608312607 2023-01-21 10:22:41.122305: step: 1796/529, loss: 0.04659681394696236 2023-01-21 10:22:42.278183: step: 1800/529, loss: 0.1068873405456543 2023-01-21 10:22:43.423316: step: 1804/529, loss: 0.011534976772964 2023-01-21 10:22:44.549961: step: 1808/529, loss: 0.053809307515621185 2023-01-21 10:22:45.686943: step: 1812/529, loss: 0.04740581288933754 2023-01-21 10:22:46.792711: step: 1816/529, loss: 0.03465624153614044 2023-01-21 10:22:47.948275: step: 1820/529, loss: 0.12698392570018768 2023-01-21 10:22:49.090822: step: 1824/529, loss: 0.003928279969841242 2023-01-21 10:22:50.252836: step: 1828/529, loss: 0.08658495545387268 2023-01-21 10:22:51.398269: step: 1832/529, loss: 0.029357576742768288 2023-01-21 10:22:52.522443: step: 1836/529, loss: 0.0035843849182128906 2023-01-21 10:22:53.645781: step: 1840/529, loss: 0.03657648339867592 2023-01-21 10:22:54.785025: step: 1844/529, loss: 0.23565292358398438 2023-01-21 10:22:55.927668: step: 1848/529, loss: 0.06023597717285156 2023-01-21 10:22:57.072453: step: 1852/529, loss: 0.0174531452357769 2023-01-21 10:22:58.221587: step: 1856/529, loss: 0.17921113967895508 2023-01-21 10:22:59.366446: step: 1860/529, loss: 0.024892425164580345 2023-01-21 10:23:00.513096: step: 1864/529, loss: 0.47402411699295044 2023-01-21 10:23:01.675130: step: 1868/529, loss: 0.10864716023206711 2023-01-21 10:23:02.813703: step: 1872/529, loss: 0.0034317015670239925 2023-01-21 10:23:03.936247: step: 1876/529, loss: 0.006154841743409634 2023-01-21 10:23:05.058487: step: 1880/529, loss: 0.20169925689697266 2023-01-21 10:23:06.156063: step: 1884/529, loss: 0.0027984618209302425 2023-01-21 10:23:07.288863: step: 1888/529, loss: 0.026531364768743515 2023-01-21 10:23:08.422814: step: 1892/529, loss: 0.09096936881542206 2023-01-21 10:23:09.573775: step: 1896/529, loss: 0.2370111495256424 2023-01-21 10:23:10.700135: step: 1900/529, loss: 0.007167292293161154 2023-01-21 10:23:11.817699: step: 1904/529, loss: 0.042585086077451706 2023-01-21 10:23:12.983967: step: 1908/529, loss: 0.039690397679805756 2023-01-21 10:23:14.115265: step: 1912/529, loss: 0.0035885809920728207 2023-01-21 10:23:15.216492: step: 1916/529, loss: 0.3277645707130432 2023-01-21 10:23:16.382179: step: 1920/529, loss: 0.09594707190990448 2023-01-21 10:23:17.507963: step: 1924/529, loss: 0.03206615522503853 2023-01-21 10:23:18.653633: step: 1928/529, loss: 0.08674336224794388 2023-01-21 10:23:19.778291: step: 1932/529, loss: 0.011665034107863903 2023-01-21 10:23:20.908697: step: 1936/529, loss: 0.051511384546756744 2023-01-21 10:23:22.054402: step: 1940/529, loss: 0.0838102400302887 2023-01-21 10:23:23.204056: step: 1944/529, loss: 0.05924806743860245 2023-01-21 10:23:24.370806: step: 1948/529, loss: 0.03266000747680664 2023-01-21 10:23:25.514141: step: 1952/529, loss: 0.1516532003879547 2023-01-21 10:23:26.658705: step: 1956/529, loss: 0.09760494530200958 2023-01-21 10:23:27.796088: step: 1960/529, loss: 0.13171225786209106 2023-01-21 10:23:28.932543: step: 1964/529, loss: 0.04678545147180557 2023-01-21 10:23:30.058127: step: 1968/529, loss: 0.08297538757324219 2023-01-21 10:23:31.213734: step: 1972/529, loss: 0.03066234476864338 2023-01-21 10:23:32.355016: step: 1976/529, loss: 0.054668523371219635 2023-01-21 10:23:33.473281: step: 1980/529, loss: 0.09990043938159943 2023-01-21 10:23:34.620142: step: 1984/529, loss: 0.06002431362867355 2023-01-21 10:23:35.758779: step: 1988/529, loss: 0.013646173290908337 2023-01-21 10:23:36.929724: step: 1992/529, loss: 0.09699364006519318 2023-01-21 10:23:38.070697: step: 1996/529, loss: 0.00797348003834486 2023-01-21 10:23:39.209590: step: 2000/529, loss: 0.052484966814517975 2023-01-21 10:23:40.359895: step: 2004/529, loss: 0.17032918334007263 2023-01-21 10:23:41.506669: step: 2008/529, loss: 0.05007191002368927 2023-01-21 10:23:42.662263: step: 2012/529, loss: 0.0706019401550293 2023-01-21 10:23:43.815003: step: 2016/529, loss: 0.012638043612241745 2023-01-21 10:23:44.963081: step: 2020/529, loss: 0.16289833188056946 2023-01-21 10:23:46.099487: step: 2024/529, loss: 0.023499298840761185 2023-01-21 10:23:47.198874: step: 2028/529, loss: 0.02674427069723606 2023-01-21 10:23:48.354980: step: 2032/529, loss: 1.3728324174880981 2023-01-21 10:23:49.480758: step: 2036/529, loss: 0.02027606964111328 2023-01-21 10:23:50.612924: step: 2040/529, loss: 0.0457921028137207 2023-01-21 10:23:51.751973: step: 2044/529, loss: 0.027369117364287376 2023-01-21 10:23:52.869684: step: 2048/529, loss: 0.0031394334509968758 2023-01-21 10:23:54.007210: step: 2052/529, loss: 0.012844085693359375 2023-01-21 10:23:55.137275: step: 2056/529, loss: 0.08121900260448456 2023-01-21 10:23:56.256551: step: 2060/529, loss: 0.09306765347719193 2023-01-21 10:23:57.384420: step: 2064/529, loss: 0.06612720340490341 2023-01-21 10:23:58.543388: step: 2068/529, loss: 0.043155863881111145 2023-01-21 10:23:59.664548: step: 2072/529, loss: 0.11836127936840057 2023-01-21 10:24:00.816549: step: 2076/529, loss: 0.08553294837474823 2023-01-21 10:24:01.971108: step: 2080/529, loss: 0.065460205078125 2023-01-21 10:24:03.114226: step: 2084/529, loss: 0.0066121225245296955 2023-01-21 10:24:04.287297: step: 2088/529, loss: 0.06153126060962677 2023-01-21 10:24:05.442814: step: 2092/529, loss: 0.012886285781860352 2023-01-21 10:24:06.530408: step: 2096/529, loss: 0.1644933670759201 2023-01-21 10:24:07.688038: step: 2100/529, loss: 0.09799116104841232 2023-01-21 10:24:08.826339: step: 2104/529, loss: 0.09508619457483292 2023-01-21 10:24:09.961334: step: 2108/529, loss: 0.171131893992424 2023-01-21 10:24:11.116062: step: 2112/529, loss: 0.07161112129688263 2023-01-21 10:24:12.249763: step: 2116/529, loss: 0.06396923214197159 ================================================== Loss: 0.094 -------------------- Dev: {'event': {'p': 0.5991649269311065, 'r': 0.7643142476697736, 'f1': 0.6717378583967233}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Test: {'event': {'p': 0.6037822878228782, 'r': 0.7584009269988412, 'f1': 0.6723163841807909}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Chinese: {'event': {'p': 0.6024096385542169, 'r': 0.9259259259259259, 'f1': 0.7299270072992702}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Korean: {'event': {'p': 0.6111111111111112, 'r': 0.5238095238095238, 'f1': 0.5641025641025642}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Russian: {'event': {'p': 0.5, 'r': 0.5555555555555556, 'f1': 0.5263157894736842}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 10:24:53.790019: step: 4/529, loss: 0.059105969965457916 2023-01-21 10:24:54.934901: step: 8/529, loss: 0.0645173043012619 2023-01-21 10:24:56.047017: step: 12/529, loss: 0.04948629438877106 2023-01-21 10:24:57.168947: step: 16/529, loss: 0.009715271182358265 2023-01-21 10:24:58.322075: step: 20/529, loss: 0.023418236523866653 2023-01-21 10:24:59.457600: step: 24/529, loss: 0.056659817695617676 2023-01-21 10:25:00.617409: step: 28/529, loss: 0.08656396716833115 2023-01-21 10:25:01.757001: step: 32/529, loss: 0.049924638122320175 2023-01-21 10:25:02.880951: step: 36/529, loss: 0.04948770999908447 2023-01-21 10:25:04.030810: step: 40/529, loss: 0.06343088299036026 2023-01-21 10:25:05.163746: step: 44/529, loss: 0.015443230047821999 2023-01-21 10:25:06.304917: step: 48/529, loss: 0.15336580574512482 2023-01-21 10:25:07.426099: step: 52/529, loss: 0.020012950524687767 2023-01-21 10:25:08.555888: step: 56/529, loss: 0.026934243738651276 2023-01-21 10:25:09.677441: step: 60/529, loss: 0.046183157712221146 2023-01-21 10:25:10.799504: step: 64/529, loss: 0.027323054149746895 2023-01-21 10:25:11.980988: step: 68/529, loss: 0.060655929148197174 2023-01-21 10:25:13.144315: step: 72/529, loss: 0.023775674402713776 2023-01-21 10:25:14.283269: step: 76/529, loss: 0.07388029247522354 2023-01-21 10:25:15.417703: step: 80/529, loss: 0.12838220596313477 2023-01-21 10:25:16.555143: step: 84/529, loss: 0.0717768669128418 2023-01-21 10:25:17.661001: step: 88/529, loss: 0.0027473450172692537 2023-01-21 10:25:18.798927: step: 92/529, loss: 0.04699058458209038 2023-01-21 10:25:19.928702: step: 96/529, loss: 0.08606776595115662 2023-01-21 10:25:21.057365: step: 100/529, loss: 0.043141745030879974 2023-01-21 10:25:22.224386: step: 104/529, loss: 0.09770040214061737 2023-01-21 10:25:23.355269: step: 108/529, loss: 0.05930342525243759 2023-01-21 10:25:24.527744: step: 112/529, loss: 0.16444511711597443 2023-01-21 10:25:25.641138: step: 116/529, loss: 0.045732785016298294 2023-01-21 10:25:26.746686: step: 120/529, loss: 0.12604545056819916 2023-01-21 10:25:27.925792: step: 124/529, loss: 0.013486290350556374 2023-01-21 10:25:29.075369: step: 128/529, loss: 0.04182462766766548 2023-01-21 10:25:30.206778: step: 132/529, loss: 0.05329122394323349 2023-01-21 10:25:31.387997: step: 136/529, loss: 0.08341045677661896 2023-01-21 10:25:32.534325: step: 140/529, loss: 0.014041328802704811 2023-01-21 10:25:33.674111: step: 144/529, loss: 0.02705674059689045 2023-01-21 10:25:34.803761: step: 148/529, loss: 0.04370150715112686 2023-01-21 10:25:35.948789: step: 152/529, loss: 0.010295487008988857 2023-01-21 10:25:37.081734: step: 156/529, loss: 0.02036895789206028 2023-01-21 10:25:38.217464: step: 160/529, loss: 0.016992950811982155 2023-01-21 10:25:39.333212: step: 164/529, loss: 0.020859908312559128 2023-01-21 10:25:40.459559: step: 168/529, loss: 0.07204189151525497 2023-01-21 10:25:41.615113: step: 172/529, loss: 0.02249589003622532 2023-01-21 10:25:42.786242: step: 176/529, loss: 0.02533874660730362 2023-01-21 10:25:43.953482: step: 180/529, loss: 0.08395090699195862 2023-01-21 10:25:45.126017: step: 184/529, loss: 0.010426384396851063 2023-01-21 10:25:46.247038: step: 188/529, loss: 0.03150737285614014 2023-01-21 10:25:47.365880: step: 192/529, loss: 0.042586516588926315 2023-01-21 10:25:48.528498: step: 196/529, loss: 0.04754199832677841 2023-01-21 10:25:49.685460: step: 200/529, loss: 0.008863067254424095 2023-01-21 10:25:50.836058: step: 204/529, loss: 0.005052757449448109 2023-01-21 10:25:51.985797: step: 208/529, loss: 0.04658851400017738 2023-01-21 10:25:53.146954: step: 212/529, loss: 0.0265120267868042 2023-01-21 10:25:54.267684: step: 216/529, loss: 0.0661783218383789 2023-01-21 10:25:55.399803: step: 220/529, loss: 0.032805442810058594 2023-01-21 10:25:56.549486: step: 224/529, loss: 0.06348991394042969 2023-01-21 10:25:57.679570: step: 228/529, loss: 0.035439252853393555 2023-01-21 10:25:58.816576: step: 232/529, loss: 0.2851695716381073 2023-01-21 10:25:59.959174: step: 236/529, loss: 0.04455013573169708 2023-01-21 10:26:01.126081: step: 240/529, loss: 0.0759057104587555 2023-01-21 10:26:02.277994: step: 244/529, loss: 0.06230630725622177 2023-01-21 10:26:03.406153: step: 248/529, loss: 0.05028343200683594 2023-01-21 10:26:04.549313: step: 252/529, loss: 0.025524521246552467 2023-01-21 10:26:05.692909: step: 256/529, loss: 0.10955486446619034 2023-01-21 10:26:06.838620: step: 260/529, loss: 0.048208095133304596 2023-01-21 10:26:07.964196: step: 264/529, loss: 0.033761024475097656 2023-01-21 10:26:09.111066: step: 268/529, loss: 0.057462021708488464 2023-01-21 10:26:10.259805: step: 272/529, loss: 0.06448125839233398 2023-01-21 10:26:11.447952: step: 276/529, loss: 0.04942522197961807 2023-01-21 10:26:12.620907: step: 280/529, loss: 0.00915985181927681 2023-01-21 10:26:13.797646: step: 284/529, loss: 0.024196816608309746 2023-01-21 10:26:14.911959: step: 288/529, loss: 0.060741715133190155 2023-01-21 10:26:16.062389: step: 292/529, loss: 0.03851795196533203 2023-01-21 10:26:17.209235: step: 296/529, loss: 0.055022433400154114 2023-01-21 10:26:18.357740: step: 300/529, loss: 0.004605770111083984 2023-01-21 10:26:19.531437: step: 304/529, loss: 0.04043159633874893 2023-01-21 10:26:20.674022: step: 308/529, loss: 0.046128083020448685 2023-01-21 10:26:21.833266: step: 312/529, loss: 0.010307692922651768 2023-01-21 10:26:22.994266: step: 316/529, loss: 0.5625835657119751 2023-01-21 10:26:24.128115: step: 320/529, loss: 0.040352363139390945 2023-01-21 10:26:25.264527: step: 324/529, loss: 0.018245507031679153 2023-01-21 10:26:26.406282: step: 328/529, loss: 0.03461475297808647 2023-01-21 10:26:27.535822: step: 332/529, loss: 0.04432697221636772 2023-01-21 10:26:28.664849: step: 336/529, loss: 0.033873941749334335 2023-01-21 10:26:29.795444: step: 340/529, loss: 0.06554307788610458 2023-01-21 10:26:30.975883: step: 344/529, loss: 0.09892864525318146 2023-01-21 10:26:32.119546: step: 348/529, loss: 0.04830008000135422 2023-01-21 10:26:33.253431: step: 352/529, loss: 0.0764305591583252 2023-01-21 10:26:34.391586: step: 356/529, loss: 0.003944301512092352 2023-01-21 10:26:35.534097: step: 360/529, loss: 0.11398305743932724 2023-01-21 10:26:36.706462: step: 364/529, loss: 0.05307529866695404 2023-01-21 10:26:37.836300: step: 368/529, loss: 0.04363727569580078 2023-01-21 10:26:38.989645: step: 372/529, loss: 0.061898425221443176 2023-01-21 10:26:40.133669: step: 376/529, loss: 0.0026102066040039062 2023-01-21 10:26:41.246628: step: 380/529, loss: 0.0448877327144146 2023-01-21 10:26:42.369064: step: 384/529, loss: 0.029390431940555573 2023-01-21 10:26:43.502939: step: 388/529, loss: 0.06987170875072479 2023-01-21 10:26:44.632629: step: 392/529, loss: 0.04736804962158203 2023-01-21 10:26:45.772751: step: 396/529, loss: 0.0361480712890625 2023-01-21 10:26:46.893164: step: 400/529, loss: 0.023975946009159088 2023-01-21 10:26:48.043689: step: 404/529, loss: 0.03843478858470917 2023-01-21 10:26:49.171514: step: 408/529, loss: 0.04291677474975586 2023-01-21 10:26:50.332035: step: 412/529, loss: 0.00771255511790514 2023-01-21 10:26:51.498847: step: 416/529, loss: 0.0485166534781456 2023-01-21 10:26:52.624344: step: 420/529, loss: 0.04470081254839897 2023-01-21 10:26:53.847000: step: 424/529, loss: 0.02491469494998455 2023-01-21 10:26:54.975933: step: 428/529, loss: 0.059216879308223724 2023-01-21 10:26:56.110617: step: 432/529, loss: 0.02149954065680504 2023-01-21 10:26:57.253163: step: 436/529, loss: 0.030137062072753906 2023-01-21 10:26:58.362141: step: 440/529, loss: 0.02650637738406658 2023-01-21 10:26:59.528385: step: 444/529, loss: 0.15329113602638245 2023-01-21 10:27:00.673995: step: 448/529, loss: 0.024412918835878372 2023-01-21 10:27:01.833227: step: 452/529, loss: 0.5086908340454102 2023-01-21 10:27:02.963008: step: 456/529, loss: 0.007043266203254461 2023-01-21 10:27:04.078046: step: 460/529, loss: 0.06777582317590714 2023-01-21 10:27:05.214645: step: 464/529, loss: 0.1480318009853363 2023-01-21 10:27:06.320702: step: 468/529, loss: 0.026124000549316406 2023-01-21 10:27:07.451020: step: 472/529, loss: 0.4231296181678772 2023-01-21 10:27:08.581721: step: 476/529, loss: 0.0619107261300087 2023-01-21 10:27:09.733828: step: 480/529, loss: 0.020137976855039597 2023-01-21 10:27:10.864867: step: 484/529, loss: 0.006447935476899147 2023-01-21 10:27:11.975834: step: 488/529, loss: 0.038645461201667786 2023-01-21 10:27:13.117235: step: 492/529, loss: 0.02344489097595215 2023-01-21 10:27:14.256864: step: 496/529, loss: 0.066119484603405 2023-01-21 10:27:15.381824: step: 500/529, loss: 0.014425469562411308 2023-01-21 10:27:16.541458: step: 504/529, loss: 0.08803434669971466 2023-01-21 10:27:17.673161: step: 508/529, loss: 0.0633271262049675 2023-01-21 10:27:18.821504: step: 512/529, loss: 0.029651260003447533 2023-01-21 10:27:19.970032: step: 516/529, loss: 0.019073868170380592 2023-01-21 10:27:21.094725: step: 520/529, loss: 0.06599827110767365 2023-01-21 10:27:22.249736: step: 524/529, loss: 0.02843942679464817 2023-01-21 10:27:23.397057: step: 528/529, loss: 0.0077641489915549755 2023-01-21 10:27:24.556988: step: 532/529, loss: 0.009955024346709251 2023-01-21 10:27:25.712918: step: 536/529, loss: 0.022554589435458183 2023-01-21 10:27:26.866320: step: 540/529, loss: 0.14339271187782288 2023-01-21 10:27:28.012023: step: 544/529, loss: 0.48264798521995544 2023-01-21 10:27:29.155550: step: 548/529, loss: 0.008354758843779564 2023-01-21 10:27:30.269974: step: 552/529, loss: 0.04884080961346626 2023-01-21 10:27:31.383863: step: 556/529, loss: 0.05390520021319389 2023-01-21 10:27:32.517630: step: 560/529, loss: 0.04314251244068146 2023-01-21 10:27:33.649388: step: 564/529, loss: 0.08556537330150604 2023-01-21 10:27:34.803213: step: 568/529, loss: 0.024846553802490234 2023-01-21 10:27:35.939664: step: 572/529, loss: 0.04279651865363121 2023-01-21 10:27:37.105552: step: 576/529, loss: 0.03595714643597603 2023-01-21 10:27:38.235057: step: 580/529, loss: 0.08370581269264221 2023-01-21 10:27:39.389678: step: 584/529, loss: 0.005891323555260897 2023-01-21 10:27:40.516415: step: 588/529, loss: 0.02362527884542942 2023-01-21 10:27:41.641375: step: 592/529, loss: 0.0581483393907547 2023-01-21 10:27:42.782945: step: 596/529, loss: 0.01826615259051323 2023-01-21 10:27:43.910966: step: 600/529, loss: 0.04715690761804581 2023-01-21 10:27:45.063232: step: 604/529, loss: 0.06747522205114365 2023-01-21 10:27:46.186807: step: 608/529, loss: 0.08730635792016983 2023-01-21 10:27:47.310117: step: 612/529, loss: 0.026374198496341705 2023-01-21 10:27:48.442655: step: 616/529, loss: 0.017811108380556107 2023-01-21 10:27:49.612010: step: 620/529, loss: 1.7652653455734253 2023-01-21 10:27:50.780764: step: 624/529, loss: 0.054247546941041946 2023-01-21 10:27:51.938527: step: 628/529, loss: 0.018061399459838867 2023-01-21 10:27:53.066628: step: 632/529, loss: 0.09570875763893127 2023-01-21 10:27:54.186745: step: 636/529, loss: 0.0214083194732666 2023-01-21 10:27:55.315263: step: 640/529, loss: 0.07457075268030167 2023-01-21 10:27:56.455082: step: 644/529, loss: 0.04270830377936363 2023-01-21 10:27:57.593990: step: 648/529, loss: 0.0673549622297287 2023-01-21 10:27:58.740300: step: 652/529, loss: 0.25669556856155396 2023-01-21 10:27:59.893016: step: 656/529, loss: 0.08315477520227432 2023-01-21 10:28:01.027699: step: 660/529, loss: 0.011950301937758923 2023-01-21 10:28:02.163207: step: 664/529, loss: 0.013238906860351562 2023-01-21 10:28:03.331367: step: 668/529, loss: 0.013357162475585938 2023-01-21 10:28:04.509907: step: 672/529, loss: 0.02165069617331028 2023-01-21 10:28:05.648333: step: 676/529, loss: 0.001606035279110074 2023-01-21 10:28:06.786080: step: 680/529, loss: 0.051695965230464935 2023-01-21 10:28:07.945067: step: 684/529, loss: 0.05854034423828125 2023-01-21 10:28:09.064296: step: 688/529, loss: 0.02282686159014702 2023-01-21 10:28:10.214918: step: 692/529, loss: 0.06225090101361275 2023-01-21 10:28:11.364846: step: 696/529, loss: 0.07593603432178497 2023-01-21 10:28:12.492129: step: 700/529, loss: 0.04264011234045029 2023-01-21 10:28:13.633864: step: 704/529, loss: 0.06381864845752716 2023-01-21 10:28:14.770868: step: 708/529, loss: 0.0481809601187706 2023-01-21 10:28:15.900518: step: 712/529, loss: 0.01950225792825222 2023-01-21 10:28:17.054242: step: 716/529, loss: 0.024835586547851562 2023-01-21 10:28:18.186035: step: 720/529, loss: 0.04066123813390732 2023-01-21 10:28:19.337301: step: 724/529, loss: 0.07068872451782227 2023-01-21 10:28:20.472118: step: 728/529, loss: 0.004130077548325062 2023-01-21 10:28:21.614393: step: 732/529, loss: 0.04898247867822647 2023-01-21 10:28:22.742000: step: 736/529, loss: 0.04032273590564728 2023-01-21 10:28:23.879752: step: 740/529, loss: 0.10968603938817978 2023-01-21 10:28:25.052945: step: 744/529, loss: 6.086958408355713 2023-01-21 10:28:26.184523: step: 748/529, loss: 0.10305681079626083 2023-01-21 10:28:27.359099: step: 752/529, loss: 0.06556110084056854 2023-01-21 10:28:28.508984: step: 756/529, loss: 0.03654296323657036 2023-01-21 10:28:29.655033: step: 760/529, loss: 0.03674917295575142 2023-01-21 10:28:30.801063: step: 764/529, loss: 0.10063209384679794 2023-01-21 10:28:31.982638: step: 768/529, loss: 0.01710643619298935 2023-01-21 10:28:33.127642: step: 772/529, loss: 0.009410286322236061 2023-01-21 10:28:34.313405: step: 776/529, loss: 0.034612275660037994 2023-01-21 10:28:35.439182: step: 780/529, loss: 0.012948036193847656 2023-01-21 10:28:36.582178: step: 784/529, loss: 0.009429693222045898 2023-01-21 10:28:37.709684: step: 788/529, loss: 0.03881826624274254 2023-01-21 10:28:38.838672: step: 792/529, loss: 0.39961832761764526 2023-01-21 10:28:39.986640: step: 796/529, loss: 0.028296278789639473 2023-01-21 10:28:41.117530: step: 800/529, loss: 0.2209060788154602 2023-01-21 10:28:42.314459: step: 804/529, loss: 0.02661895751953125 2023-01-21 10:28:43.474548: step: 808/529, loss: 0.008953286334872246 2023-01-21 10:28:44.613224: step: 812/529, loss: 0.12106628715991974 2023-01-21 10:28:45.767190: step: 816/529, loss: 0.008331775665283203 2023-01-21 10:28:46.911532: step: 820/529, loss: 0.002130711218342185 2023-01-21 10:28:48.022884: step: 824/529, loss: 0.08493924140930176 2023-01-21 10:28:49.159003: step: 828/529, loss: 0.006622696295380592 2023-01-21 10:28:50.302803: step: 832/529, loss: 0.07141521573066711 2023-01-21 10:28:51.432621: step: 836/529, loss: 0.12237599492073059 2023-01-21 10:28:52.622451: step: 840/529, loss: 0.05327673256397247 2023-01-21 10:28:53.745131: step: 844/529, loss: 0.01591324806213379 2023-01-21 10:28:54.889101: step: 848/529, loss: 0.05979042127728462 2023-01-21 10:28:56.031211: step: 852/529, loss: 0.08037286251783371 2023-01-21 10:28:57.175660: step: 856/529, loss: 0.02086477354168892 2023-01-21 10:28:58.309620: step: 860/529, loss: 0.0671960860490799 2023-01-21 10:28:59.462836: step: 864/529, loss: 0.062377430498600006 2023-01-21 10:29:00.617817: step: 868/529, loss: 0.04353771358728409 2023-01-21 10:29:01.769136: step: 872/529, loss: 0.02901592291891575 2023-01-21 10:29:02.897593: step: 876/529, loss: 0.012292003259062767 2023-01-21 10:29:04.073695: step: 880/529, loss: 0.00787963904440403 2023-01-21 10:29:05.225284: step: 884/529, loss: 0.007695913314819336 2023-01-21 10:29:06.338893: step: 888/529, loss: 0.03949563577771187 2023-01-21 10:29:07.455278: step: 892/529, loss: 0.07749109715223312 2023-01-21 10:29:08.596122: step: 896/529, loss: 0.01820545271039009 2023-01-21 10:29:09.713782: step: 900/529, loss: 0.004364109132438898 2023-01-21 10:29:10.851447: step: 904/529, loss: 0.09595232456922531 2023-01-21 10:29:12.008530: step: 908/529, loss: 0.03320474550127983 2023-01-21 10:29:13.144956: step: 912/529, loss: 0.16649967432022095 2023-01-21 10:29:14.291453: step: 916/529, loss: 0.18315038084983826 2023-01-21 10:29:15.429892: step: 920/529, loss: 0.07141990959644318 2023-01-21 10:29:16.587624: step: 924/529, loss: 0.037522315979003906 2023-01-21 10:29:17.715054: step: 928/529, loss: 0.12246754765510559 2023-01-21 10:29:18.869698: step: 932/529, loss: 0.028950978070497513 2023-01-21 10:29:20.049078: step: 936/529, loss: 0.0023561001289635897 2023-01-21 10:29:21.187068: step: 940/529, loss: 0.10039768368005753 2023-01-21 10:29:22.301973: step: 944/529, loss: 0.4176711142063141 2023-01-21 10:29:23.427372: step: 948/529, loss: 0.04866151884198189 2023-01-21 10:29:24.563011: step: 952/529, loss: 0.01057291030883789 2023-01-21 10:29:25.701284: step: 956/529, loss: 0.07761001586914062 2023-01-21 10:29:26.838196: step: 960/529, loss: 0.02510242536664009 2023-01-21 10:29:27.973985: step: 964/529, loss: 0.007092857733368874 2023-01-21 10:29:29.129779: step: 968/529, loss: 0.22311726212501526 2023-01-21 10:29:30.252928: step: 972/529, loss: 0.05343981087207794 2023-01-21 10:29:31.404033: step: 976/529, loss: 0.014866352081298828 2023-01-21 10:29:32.531885: step: 980/529, loss: 0.12448649853467941 2023-01-21 10:29:33.690329: step: 984/529, loss: 0.018642805516719818 2023-01-21 10:29:34.816502: step: 988/529, loss: 0.013216591440141201 2023-01-21 10:29:35.959062: step: 992/529, loss: 0.009308052249252796 2023-01-21 10:29:37.107179: step: 996/529, loss: 0.052773188799619675 2023-01-21 10:29:38.226067: step: 1000/529, loss: 0.021493911743164062 2023-01-21 10:29:39.385751: step: 1004/529, loss: 0.021619463339447975 2023-01-21 10:29:40.531001: step: 1008/529, loss: 0.008909416384994984 2023-01-21 10:29:41.639381: step: 1012/529, loss: 0.03968324884772301 2023-01-21 10:29:42.770126: step: 1016/529, loss: 0.01737501472234726 2023-01-21 10:29:43.914350: step: 1020/529, loss: 0.08769450336694717 2023-01-21 10:29:45.064184: step: 1024/529, loss: 0.06746683269739151 2023-01-21 10:29:46.214021: step: 1028/529, loss: 0.04657325893640518 2023-01-21 10:29:47.352307: step: 1032/529, loss: 0.0790410041809082 2023-01-21 10:29:48.516894: step: 1036/529, loss: 0.0008233592379838228 2023-01-21 10:29:49.680570: step: 1040/529, loss: 0.027800561860203743 2023-01-21 10:29:50.805283: step: 1044/529, loss: 0.012504386715590954 2023-01-21 10:29:51.980330: step: 1048/529, loss: 0.14007015526294708 2023-01-21 10:29:53.136218: step: 1052/529, loss: 0.017615031450986862 2023-01-21 10:29:54.286466: step: 1056/529, loss: 0.03188290819525719 2023-01-21 10:29:55.449887: step: 1060/529, loss: 0.03685960918664932 2023-01-21 10:29:56.578688: step: 1064/529, loss: 0.01133031863719225 2023-01-21 10:29:57.748205: step: 1068/529, loss: 0.12951403856277466 2023-01-21 10:29:58.861363: step: 1072/529, loss: 0.13424329459667206 2023-01-21 10:29:59.978472: step: 1076/529, loss: 0.010524488054215908 2023-01-21 10:30:01.109636: step: 1080/529, loss: 0.05392765998840332 2023-01-21 10:30:02.234480: step: 1084/529, loss: 0.1908133625984192 2023-01-21 10:30:03.371434: step: 1088/529, loss: 0.0072705745697021484 2023-01-21 10:30:04.516059: step: 1092/529, loss: 0.9742828607559204 2023-01-21 10:30:05.653932: step: 1096/529, loss: 0.042876627296209335 2023-01-21 10:30:06.812741: step: 1100/529, loss: 0.02855663374066353 2023-01-21 10:30:07.944847: step: 1104/529, loss: 0.010004043579101562 2023-01-21 10:30:09.059719: step: 1108/529, loss: 0.03783893957734108 2023-01-21 10:30:10.187693: step: 1112/529, loss: 0.027332782745361328 2023-01-21 10:30:11.331457: step: 1116/529, loss: 0.026285268366336823 2023-01-21 10:30:12.461986: step: 1120/529, loss: 0.021649695932865143 2023-01-21 10:30:13.601015: step: 1124/529, loss: 0.0799618735909462 2023-01-21 10:30:14.747286: step: 1128/529, loss: 0.11349983513355255 2023-01-21 10:30:15.880442: step: 1132/529, loss: 0.06392087787389755 2023-01-21 10:30:17.013002: step: 1136/529, loss: 0.02036123350262642 2023-01-21 10:30:18.152002: step: 1140/529, loss: 0.009700584225356579 2023-01-21 10:30:19.278257: step: 1144/529, loss: 0.008136940188705921 2023-01-21 10:30:20.417267: step: 1148/529, loss: 0.04749260097742081 2023-01-21 10:30:21.578705: step: 1152/529, loss: 0.036373041570186615 2023-01-21 10:30:22.714166: step: 1156/529, loss: 0.05070476233959198 2023-01-21 10:30:23.844573: step: 1160/529, loss: 0.08789065480232239 2023-01-21 10:30:24.995394: step: 1164/529, loss: 0.015475083142518997 2023-01-21 10:30:26.150336: step: 1168/529, loss: 0.09388384968042374 2023-01-21 10:30:27.280584: step: 1172/529, loss: 0.02700061909854412 2023-01-21 10:30:28.443834: step: 1176/529, loss: 0.03811941295862198 2023-01-21 10:30:29.606663: step: 1180/529, loss: 0.18512848019599915 2023-01-21 10:30:30.783952: step: 1184/529, loss: 0.08280449360609055 2023-01-21 10:30:31.935788: step: 1188/529, loss: 0.2402186393737793 2023-01-21 10:30:33.063633: step: 1192/529, loss: 0.049204353243112564 2023-01-21 10:30:34.182104: step: 1196/529, loss: 0.026955604553222656 2023-01-21 10:30:35.319144: step: 1200/529, loss: 0.035355761647224426 2023-01-21 10:30:36.482370: step: 1204/529, loss: 0.019194602966308594 2023-01-21 10:30:37.631373: step: 1208/529, loss: 0.019014250487089157 2023-01-21 10:30:38.788496: step: 1212/529, loss: 0.07572594285011292 2023-01-21 10:30:39.973341: step: 1216/529, loss: 0.026965808123350143 2023-01-21 10:30:41.099846: step: 1220/529, loss: 0.057595252990722656 2023-01-21 10:30:42.230841: step: 1224/529, loss: 0.07913951575756073 2023-01-21 10:30:43.389346: step: 1228/529, loss: 0.05920295789837837 2023-01-21 10:30:44.545865: step: 1232/529, loss: 0.3227514922618866 2023-01-21 10:30:45.695194: step: 1236/529, loss: 0.04816637188196182 2023-01-21 10:30:46.815903: step: 1240/529, loss: 0.09544067084789276 2023-01-21 10:30:47.954757: step: 1244/529, loss: 0.04865531995892525 2023-01-21 10:30:49.107452: step: 1248/529, loss: 0.05016060173511505 2023-01-21 10:30:50.252861: step: 1252/529, loss: 0.07657375931739807 2023-01-21 10:30:51.381785: step: 1256/529, loss: 0.06572751700878143 2023-01-21 10:30:52.524872: step: 1260/529, loss: 0.021331120282411575 2023-01-21 10:30:53.650104: step: 1264/529, loss: 0.11602268368005753 2023-01-21 10:30:54.813117: step: 1268/529, loss: 0.020692825317382812 2023-01-21 10:30:55.972273: step: 1272/529, loss: 0.06271343678236008 2023-01-21 10:30:57.114532: step: 1276/529, loss: 0.019038628786802292 2023-01-21 10:30:58.245440: step: 1280/529, loss: 0.031415559351444244 2023-01-21 10:30:59.375377: step: 1284/529, loss: 0.11527501046657562 2023-01-21 10:31:00.519398: step: 1288/529, loss: 0.06377792358398438 2023-01-21 10:31:01.661058: step: 1292/529, loss: 0.05818531662225723 2023-01-21 10:31:02.788832: step: 1296/529, loss: 0.0839502364397049 2023-01-21 10:31:03.923762: step: 1300/529, loss: 0.013971042819321156 2023-01-21 10:31:05.069780: step: 1304/529, loss: 0.05349111557006836 2023-01-21 10:31:06.199884: step: 1308/529, loss: 0.07636123150587082 2023-01-21 10:31:07.348907: step: 1312/529, loss: 0.013654423877596855 2023-01-21 10:31:08.485520: step: 1316/529, loss: 0.046025846153497696 2023-01-21 10:31:09.610257: step: 1320/529, loss: 0.08206848800182343 2023-01-21 10:31:10.729555: step: 1324/529, loss: 0.021786976605653763 2023-01-21 10:31:11.894665: step: 1328/529, loss: 0.017472337931394577 2023-01-21 10:31:13.025525: step: 1332/529, loss: 0.00736546516418457 2023-01-21 10:31:14.172649: step: 1336/529, loss: 0.04738273471593857 2023-01-21 10:31:15.295917: step: 1340/529, loss: 0.011144256219267845 2023-01-21 10:31:16.431469: step: 1344/529, loss: 0.013173866085708141 2023-01-21 10:31:17.582318: step: 1348/529, loss: 0.0019618987571448088 2023-01-21 10:31:18.691446: step: 1352/529, loss: 0.03924417495727539 2023-01-21 10:31:19.852186: step: 1356/529, loss: 0.008240891620516777 2023-01-21 10:31:21.002756: step: 1360/529, loss: 0.01332874409854412 2023-01-21 10:31:22.132067: step: 1364/529, loss: 0.012959122657775879 2023-01-21 10:31:23.272215: step: 1368/529, loss: 0.2457553893327713 2023-01-21 10:31:24.421162: step: 1372/529, loss: 0.008721256628632545 2023-01-21 10:31:25.567162: step: 1376/529, loss: 0.06694066524505615 2023-01-21 10:31:26.695066: step: 1380/529, loss: 0.05604109913110733 2023-01-21 10:31:27.830899: step: 1384/529, loss: 0.009931516833603382 2023-01-21 10:31:28.984965: step: 1388/529, loss: 0.045420125126838684 2023-01-21 10:31:30.100979: step: 1392/529, loss: 0.057488251477479935 2023-01-21 10:31:31.218942: step: 1396/529, loss: 0.030385855585336685 2023-01-21 10:31:32.386207: step: 1400/529, loss: 0.03591332212090492 2023-01-21 10:31:33.537981: step: 1404/529, loss: 0.005641937255859375 2023-01-21 10:31:34.695177: step: 1408/529, loss: 0.06142731010913849 2023-01-21 10:31:35.844504: step: 1412/529, loss: 0.03171978145837784 2023-01-21 10:31:36.978031: step: 1416/529, loss: 0.15856285393238068 2023-01-21 10:31:38.121250: step: 1420/529, loss: 0.019669247791171074 2023-01-21 10:31:39.260694: step: 1424/529, loss: 0.00806207675486803 2023-01-21 10:31:40.399047: step: 1428/529, loss: 0.020203400403261185 2023-01-21 10:31:41.517239: step: 1432/529, loss: 0.1023469865322113 2023-01-21 10:31:42.643547: step: 1436/529, loss: 0.09445791691541672 2023-01-21 10:31:43.791236: step: 1440/529, loss: 0.08383483439683914 2023-01-21 10:31:44.911752: step: 1444/529, loss: 2.042593002319336 2023-01-21 10:31:46.046165: step: 1448/529, loss: 0.0007819652673788369 2023-01-21 10:31:47.157001: step: 1452/529, loss: 0.0836271345615387 2023-01-21 10:31:48.286050: step: 1456/529, loss: 0.07840938866138458 2023-01-21 10:31:49.436975: step: 1460/529, loss: 0.05001964420080185 2023-01-21 10:31:50.601013: step: 1464/529, loss: 0.0506071113049984 2023-01-21 10:31:51.777506: step: 1468/529, loss: 0.09818859398365021 2023-01-21 10:31:52.924980: step: 1472/529, loss: 0.03470568731427193 2023-01-21 10:31:54.061158: step: 1476/529, loss: 0.04303798824548721 2023-01-21 10:31:55.180544: step: 1480/529, loss: 0.007410717196762562 2023-01-21 10:31:56.298969: step: 1484/529, loss: 0.03707394376397133 2023-01-21 10:31:57.440581: step: 1488/529, loss: 0.010460233315825462 2023-01-21 10:31:58.571415: step: 1492/529, loss: 0.02482910268008709 2023-01-21 10:31:59.707706: step: 1496/529, loss: 0.1253221482038498 2023-01-21 10:32:00.848988: step: 1500/529, loss: 0.11007185280323029 2023-01-21 10:32:02.006169: step: 1504/529, loss: 0.06298866868019104 2023-01-21 10:32:03.137480: step: 1508/529, loss: 0.013941287994384766 2023-01-21 10:32:04.284499: step: 1512/529, loss: 0.032973095774650574 2023-01-21 10:32:05.406881: step: 1516/529, loss: 0.006217336747795343 2023-01-21 10:32:06.554368: step: 1520/529, loss: 0.02469329908490181 2023-01-21 10:32:07.693468: step: 1524/529, loss: 0.06664124131202698 2023-01-21 10:32:08.831065: step: 1528/529, loss: 0.04446103423833847 2023-01-21 10:32:09.966723: step: 1532/529, loss: 0.055391646921634674 2023-01-21 10:32:11.120832: step: 1536/529, loss: 0.0877685546875 2023-01-21 10:32:12.285183: step: 1540/529, loss: 0.04209556430578232 2023-01-21 10:32:13.406658: step: 1544/529, loss: 0.007458257954567671 2023-01-21 10:32:14.554669: step: 1548/529, loss: 0.05372180789709091 2023-01-21 10:32:15.677017: step: 1552/529, loss: 0.08323769271373749 2023-01-21 10:32:16.835511: step: 1556/529, loss: 0.02480611763894558 2023-01-21 10:32:17.985746: step: 1560/529, loss: 0.03324947506189346 2023-01-21 10:32:19.135244: step: 1564/529, loss: 0.08538541942834854 2023-01-21 10:32:20.292904: step: 1568/529, loss: 0.04229292646050453 2023-01-21 10:32:21.436010: step: 1572/529, loss: 0.047565557062625885 2023-01-21 10:32:22.561824: step: 1576/529, loss: 0.19779334962368011 2023-01-21 10:32:23.702968: step: 1580/529, loss: 0.04985647648572922 2023-01-21 10:32:24.832641: step: 1584/529, loss: 0.26626014709472656 2023-01-21 10:32:25.936619: step: 1588/529, loss: 0.01124420203268528 2023-01-21 10:32:27.097645: step: 1592/529, loss: 0.017621994018554688 2023-01-21 10:32:28.229706: step: 1596/529, loss: 0.004379463382065296 2023-01-21 10:32:29.395627: step: 1600/529, loss: 0.054818347096443176 2023-01-21 10:32:30.529153: step: 1604/529, loss: 0.015379572287201881 2023-01-21 10:32:31.675121: step: 1608/529, loss: 0.028466034680604935 2023-01-21 10:32:32.771781: step: 1612/529, loss: 0.09541749954223633 2023-01-21 10:32:33.912216: step: 1616/529, loss: 0.040041591972112656 2023-01-21 10:32:35.080024: step: 1620/529, loss: 0.17233572900295258 2023-01-21 10:32:36.222286: step: 1624/529, loss: 0.4602371156215668 2023-01-21 10:32:37.366247: step: 1628/529, loss: 0.026013899594545364 2023-01-21 10:32:38.503640: step: 1632/529, loss: 0.1623372584581375 2023-01-21 10:32:39.630329: step: 1636/529, loss: 0.008664751425385475 2023-01-21 10:32:40.755636: step: 1640/529, loss: 0.11099891364574432 2023-01-21 10:32:41.914702: step: 1644/529, loss: 0.025978660210967064 2023-01-21 10:32:43.072859: step: 1648/529, loss: 0.3354078531265259 2023-01-21 10:32:44.225838: step: 1652/529, loss: 0.03018779866397381 2023-01-21 10:32:45.353924: step: 1656/529, loss: 0.004010963253676891 2023-01-21 10:32:46.498851: step: 1660/529, loss: 0.8976025581359863 2023-01-21 10:32:47.616810: step: 1664/529, loss: 0.004965687170624733 2023-01-21 10:32:48.771576: step: 1668/529, loss: 0.10575437545776367 2023-01-21 10:32:49.905617: step: 1672/529, loss: 0.04000701755285263 2023-01-21 10:32:51.057150: step: 1676/529, loss: 0.050385523587465286 2023-01-21 10:32:52.202611: step: 1680/529, loss: 0.022879745811223984 2023-01-21 10:32:53.330252: step: 1684/529, loss: 0.08208112418651581 2023-01-21 10:32:54.463856: step: 1688/529, loss: 0.03307771682739258 2023-01-21 10:32:55.611008: step: 1692/529, loss: 0.07761363685131073 2023-01-21 10:32:56.742635: step: 1696/529, loss: 0.05516853183507919 2023-01-21 10:32:57.862308: step: 1700/529, loss: 0.027624012902379036 2023-01-21 10:32:59.037130: step: 1704/529, loss: 0.016211509704589844 2023-01-21 10:33:00.149101: step: 1708/529, loss: 0.017551232129335403 2023-01-21 10:33:01.278181: step: 1712/529, loss: 0.008517026901245117 2023-01-21 10:33:02.429986: step: 1716/529, loss: 0.05596160888671875 2023-01-21 10:33:03.579537: step: 1720/529, loss: 0.03875589370727539 2023-01-21 10:33:04.728512: step: 1724/529, loss: 0.15320034325122833 2023-01-21 10:33:05.883979: step: 1728/529, loss: 0.07086601853370667 2023-01-21 10:33:07.015597: step: 1732/529, loss: 0.021431587636470795 2023-01-21 10:33:08.144900: step: 1736/529, loss: 0.06189604103565216 2023-01-21 10:33:09.321247: step: 1740/529, loss: 0.05914762616157532 2023-01-21 10:33:10.492946: step: 1744/529, loss: 0.08634558320045471 2023-01-21 10:33:11.624718: step: 1748/529, loss: 0.037447355687618256 2023-01-21 10:33:12.781115: step: 1752/529, loss: 0.085790254175663 2023-01-21 10:33:13.924613: step: 1756/529, loss: 0.02508225478231907 2023-01-21 10:33:15.062904: step: 1760/529, loss: 0.012481116689741611 2023-01-21 10:33:16.191975: step: 1764/529, loss: 0.1250247359275818 2023-01-21 10:33:17.311214: step: 1768/529, loss: 0.01087942160665989 2023-01-21 10:33:18.429821: step: 1772/529, loss: 0.19690103828907013 2023-01-21 10:33:19.563532: step: 1776/529, loss: 0.07402630150318146 2023-01-21 10:33:20.676552: step: 1780/529, loss: 0.07241058349609375 2023-01-21 10:33:21.809739: step: 1784/529, loss: 0.026280594989657402 2023-01-21 10:33:22.935323: step: 1788/529, loss: 0.13185958564281464 2023-01-21 10:33:24.069430: step: 1792/529, loss: 0.06563983112573624 2023-01-21 10:33:25.199547: step: 1796/529, loss: 0.00043745042057707906 2023-01-21 10:33:26.354893: step: 1800/529, loss: 0.02372722700238228 2023-01-21 10:33:27.529444: step: 1804/529, loss: 0.021540069952607155 2023-01-21 10:33:28.683790: step: 1808/529, loss: 0.07707672566175461 2023-01-21 10:33:29.843840: step: 1812/529, loss: 0.02854936011135578 2023-01-21 10:33:30.986637: step: 1816/529, loss: 0.037993669509887695 2023-01-21 10:33:32.122677: step: 1820/529, loss: 0.08730325847864151 2023-01-21 10:33:33.266417: step: 1824/529, loss: 0.05254373699426651 2023-01-21 10:33:34.420875: step: 1828/529, loss: 0.13970327377319336 2023-01-21 10:33:35.561871: step: 1832/529, loss: 0.02880268171429634 2023-01-21 10:33:36.699597: step: 1836/529, loss: 0.008268028497695923 2023-01-21 10:33:37.836127: step: 1840/529, loss: 0.10478801280260086 2023-01-21 10:33:39.003691: step: 1844/529, loss: 0.04279174655675888 2023-01-21 10:33:40.174721: step: 1848/529, loss: 0.12447204440832138 2023-01-21 10:33:41.309837: step: 1852/529, loss: 0.00900888442993164 2023-01-21 10:33:42.460722: step: 1856/529, loss: 0.015284538269042969 2023-01-21 10:33:43.592659: step: 1860/529, loss: 0.0898069441318512 2023-01-21 10:33:44.758686: step: 1864/529, loss: 0.11184167861938477 2023-01-21 10:33:45.906821: step: 1868/529, loss: 0.03919639438390732 2023-01-21 10:33:47.043181: step: 1872/529, loss: 0.03649768978357315 2023-01-21 10:33:48.192306: step: 1876/529, loss: 0.01764148473739624 2023-01-21 10:33:49.352977: step: 1880/529, loss: 0.44543614983558655 2023-01-21 10:33:50.497610: step: 1884/529, loss: 0.0774996280670166 2023-01-21 10:33:51.617049: step: 1888/529, loss: 0.05189252272248268 2023-01-21 10:33:52.837307: step: 1892/529, loss: 0.08855752646923065 2023-01-21 10:33:53.959069: step: 1896/529, loss: 0.04244070127606392 2023-01-21 10:33:55.099890: step: 1900/529, loss: 0.0645488053560257 2023-01-21 10:33:56.222100: step: 1904/529, loss: 0.033716440200805664 2023-01-21 10:33:57.385358: step: 1908/529, loss: 0.012368488125503063 2023-01-21 10:33:58.509869: step: 1912/529, loss: 0.04617509990930557 2023-01-21 10:33:59.626689: step: 1916/529, loss: 0.0032367706298828125 2023-01-21 10:34:00.786013: step: 1920/529, loss: 0.031805798411369324 2023-01-21 10:34:01.941225: step: 1924/529, loss: 0.1205202117562294 2023-01-21 10:34:03.081202: step: 1928/529, loss: 0.024872303009033203 2023-01-21 10:34:04.201164: step: 1932/529, loss: 0.0014486312866210938 2023-01-21 10:34:05.351858: step: 1936/529, loss: 0.009503459557890892 2023-01-21 10:34:06.470096: step: 1940/529, loss: 0.12899132072925568 2023-01-21 10:34:07.589956: step: 1944/529, loss: 0.004157447721809149 2023-01-21 10:34:08.723154: step: 1948/529, loss: 0.0014079094398766756 2023-01-21 10:34:09.835040: step: 1952/529, loss: 0.011880159378051758 2023-01-21 10:34:10.996691: step: 1956/529, loss: 0.025216009467840195 2023-01-21 10:34:12.147060: step: 1960/529, loss: 0.09297056496143341 2023-01-21 10:34:13.269164: step: 1964/529, loss: 0.0857691764831543 2023-01-21 10:34:14.403791: step: 1968/529, loss: 0.9529792666435242 2023-01-21 10:34:15.526157: step: 1972/529, loss: 0.27082690596580505 2023-01-21 10:34:16.665472: step: 1976/529, loss: 0.06855086982250214 2023-01-21 10:34:17.815562: step: 1980/529, loss: 0.2497449815273285 2023-01-21 10:34:18.981777: step: 1984/529, loss: 0.008680534549057484 2023-01-21 10:34:20.122540: step: 1988/529, loss: 0.27282285690307617 2023-01-21 10:34:21.293494: step: 1992/529, loss: 0.04147171974182129 2023-01-21 10:34:22.442191: step: 1996/529, loss: 0.015248488634824753 2023-01-21 10:34:23.598107: step: 2000/529, loss: 0.0354703888297081 2023-01-21 10:34:24.723645: step: 2004/529, loss: 0.0023620130959898233 2023-01-21 10:34:25.919339: step: 2008/529, loss: 0.024892903864383698 2023-01-21 10:34:27.074570: step: 2012/529, loss: 0.0019614219199866056 2023-01-21 10:34:28.214383: step: 2016/529, loss: 0.011834239587187767 2023-01-21 10:34:29.375150: step: 2020/529, loss: 0.014704895205795765 2023-01-21 10:34:30.530267: step: 2024/529, loss: 0.11264859139919281 2023-01-21 10:34:31.688094: step: 2028/529, loss: 0.13860435783863068 2023-01-21 10:34:32.820623: step: 2032/529, loss: 0.14688043296337128 2023-01-21 10:34:33.951960: step: 2036/529, loss: 0.05584259331226349 2023-01-21 10:34:35.107141: step: 2040/529, loss: 0.11217594146728516 2023-01-21 10:34:36.251687: step: 2044/529, loss: 0.1956554353237152 2023-01-21 10:34:37.392656: step: 2048/529, loss: 0.10633993148803711 2023-01-21 10:34:38.523422: step: 2052/529, loss: 0.0359223373234272 2023-01-21 10:34:39.649468: step: 2056/529, loss: 0.011312102898955345 2023-01-21 10:34:40.779098: step: 2060/529, loss: 0.11122951656579971 2023-01-21 10:34:41.936454: step: 2064/529, loss: 0.057933416217565536 2023-01-21 10:34:43.092717: step: 2068/529, loss: 0.019565392285585403 2023-01-21 10:34:44.269122: step: 2072/529, loss: 0.07759909331798553 2023-01-21 10:34:45.433215: step: 2076/529, loss: 0.05175838619470596 2023-01-21 10:34:46.557253: step: 2080/529, loss: 0.010888171382248402 2023-01-21 10:34:47.696016: step: 2084/529, loss: 0.05133819580078125 2023-01-21 10:34:48.821575: step: 2088/529, loss: 0.03733024373650551 2023-01-21 10:34:49.948342: step: 2092/529, loss: 0.015604972839355469 2023-01-21 10:34:51.085498: step: 2096/529, loss: 0.03483233600854874 2023-01-21 10:34:52.227257: step: 2100/529, loss: 0.0181745532900095 2023-01-21 10:34:53.353804: step: 2104/529, loss: 0.012670159339904785 2023-01-21 10:34:54.483420: step: 2108/529, loss: 0.0547761432826519 2023-01-21 10:34:55.643791: step: 2112/529, loss: 0.009130263701081276 2023-01-21 10:34:56.798070: step: 2116/529, loss: 0.11732816696166992 ================================================== Loss: 0.084 -------------------- Dev: {'event': {'p': 0.6038500506585613, 'r': 0.7936085219707057, 'f1': 0.6858457997698505}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Test: {'event': {'p': 0.6094974642692486, 'r': 0.7659327925840093, 'f1': 0.6788189987163029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Chinese: {'event': {'p': 0.5517241379310345, 'r': 0.8888888888888888, 'f1': 0.6808510638297872}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Korean: {'event': {'p': 0.5892857142857143, 'r': 0.5238095238095238, 'f1': 0.5546218487394958}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 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': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 10:35:38.552641: step: 4/529, loss: 0.009599877521395683 2023-01-21 10:35:39.672972: step: 8/529, loss: 0.05222933366894722 2023-01-21 10:35:40.803627: step: 12/529, loss: 0.06327927112579346 2023-01-21 10:35:41.935816: step: 16/529, loss: 0.055777840316295624 2023-01-21 10:35:43.081957: step: 20/529, loss: 0.00335693359375 2023-01-21 10:35:44.211221: step: 24/529, loss: 0.06440359354019165 2023-01-21 10:35:45.358356: step: 28/529, loss: 0.04398956149816513 2023-01-21 10:35:46.499117: step: 32/529, loss: 0.05248670652508736 2023-01-21 10:35:47.602972: step: 36/529, loss: 0.0849311351776123 2023-01-21 10:35:48.750755: step: 40/529, loss: 0.02024099975824356 2023-01-21 10:35:49.885552: step: 44/529, loss: 0.01705489121377468 2023-01-21 10:35:51.040475: step: 48/529, loss: 0.0442129410803318 2023-01-21 10:35:52.162948: step: 52/529, loss: 0.05578432232141495 2023-01-21 10:35:53.293389: step: 56/529, loss: 0.030672885477542877 2023-01-21 10:35:54.425848: step: 60/529, loss: 0.1617903858423233 2023-01-21 10:35:55.579794: step: 64/529, loss: 0.0031072618439793587 2023-01-21 10:35:56.745718: step: 68/529, loss: 0.10301484912633896 2023-01-21 10:35:57.873292: step: 72/529, loss: 0.005670642480254173 2023-01-21 10:35:58.995429: step: 76/529, loss: 0.09278325736522675 2023-01-21 10:36:00.135622: step: 80/529, loss: 0.0005670547834597528 2023-01-21 10:36:01.246854: step: 84/529, loss: 0.03500165790319443 2023-01-21 10:36:02.397838: step: 88/529, loss: 0.09778013825416565 2023-01-21 10:36:03.539739: step: 92/529, loss: 0.08888387680053711 2023-01-21 10:36:04.692275: step: 96/529, loss: 0.03051605261862278 2023-01-21 10:36:05.812590: step: 100/529, loss: 0.0037838458083570004 2023-01-21 10:36:06.931868: step: 104/529, loss: 0.04044532775878906 2023-01-21 10:36:08.074605: step: 108/529, loss: 0.0709843635559082 2023-01-21 10:36:09.204593: step: 112/529, loss: 0.03643383830785751 2023-01-21 10:36:10.330084: step: 116/529, loss: 0.0047227381728589535 2023-01-21 10:36:11.480196: step: 120/529, loss: 0.03100304678082466 2023-01-21 10:36:12.658394: step: 124/529, loss: 0.025669338181614876 2023-01-21 10:36:13.774753: step: 128/529, loss: 0.04012279585003853 2023-01-21 10:36:14.909132: step: 132/529, loss: 0.08746547996997833 2023-01-21 10:36:16.036443: step: 136/529, loss: 0.007591248024255037 2023-01-21 10:36:17.172419: step: 140/529, loss: 0.16226130723953247 2023-01-21 10:36:18.313826: step: 144/529, loss: 0.08277511596679688 2023-01-21 10:36:19.466171: step: 148/529, loss: 0.09894141554832458 2023-01-21 10:36:20.602114: step: 152/529, loss: 0.03770933300256729 2023-01-21 10:36:21.720520: step: 156/529, loss: 0.036876678466796875 2023-01-21 10:36:22.862956: step: 160/529, loss: 0.011736870743334293 2023-01-21 10:36:24.000522: step: 164/529, loss: 0.03053007274866104 2023-01-21 10:36:25.120140: step: 168/529, loss: 0.02890167199075222 2023-01-21 10:36:26.250957: step: 172/529, loss: 0.005604791454970837 2023-01-21 10:36:27.384035: step: 176/529, loss: 0.04822740703821182 2023-01-21 10:36:28.508615: step: 180/529, loss: 0.02780008316040039 2023-01-21 10:36:29.646468: step: 184/529, loss: 0.05961675941944122 2023-01-21 10:36:30.823254: step: 188/529, loss: 0.05668621137738228 2023-01-21 10:36:31.944002: step: 192/529, loss: 0.08477868884801865 2023-01-21 10:36:33.064291: step: 196/529, loss: 0.05289297178387642 2023-01-21 10:36:34.205773: step: 200/529, loss: 0.006886291783303022 2023-01-21 10:36:35.358930: step: 204/529, loss: 0.07418803870677948 2023-01-21 10:36:36.500943: step: 208/529, loss: 0.004151248838752508 2023-01-21 10:36:37.615886: step: 212/529, loss: 0.02871112897992134 2023-01-21 10:36:38.741053: step: 216/529, loss: 0.019187163561582565 2023-01-21 10:36:39.903301: step: 220/529, loss: 0.04193706437945366 2023-01-21 10:36:41.052859: step: 224/529, loss: 0.012210512533783913 2023-01-21 10:36:42.204671: step: 228/529, loss: 0.005512523930519819 2023-01-21 10:36:43.349440: step: 232/529, loss: 0.04316416010260582 2023-01-21 10:36:44.490342: step: 236/529, loss: 0.037366341799497604 2023-01-21 10:36:45.632292: step: 240/529, loss: 0.03855609893798828 2023-01-21 10:36:46.806637: step: 244/529, loss: 0.09604892134666443 2023-01-21 10:36:47.942289: step: 248/529, loss: 0.07038488239049911 2023-01-21 10:36:49.110071: step: 252/529, loss: 0.055687904357910156 2023-01-21 10:36:50.265650: step: 256/529, loss: 0.0026640892028808594 2023-01-21 10:36:51.408396: step: 260/529, loss: 0.06011056900024414 2023-01-21 10:36:52.541030: step: 264/529, loss: 0.026077603921294212 2023-01-21 10:36:53.691584: step: 268/529, loss: 0.027528000995516777 2023-01-21 10:36:54.842175: step: 272/529, loss: 0.17011794447898865 2023-01-21 10:36:55.968782: step: 276/529, loss: 0.012026202864944935 2023-01-21 10:36:57.099170: step: 280/529, loss: 0.047213364392519 2023-01-21 10:36:58.246673: step: 284/529, loss: 0.08218078315258026 2023-01-21 10:36:59.399057: step: 288/529, loss: 0.004873848054558039 2023-01-21 10:37:00.528926: step: 292/529, loss: 0.08538713306188583 2023-01-21 10:37:01.666071: step: 296/529, loss: 0.058156393468379974 2023-01-21 10:37:02.797408: step: 300/529, loss: 0.001339817070402205 2023-01-21 10:37:03.944105: step: 304/529, loss: 0.02874293550848961 2023-01-21 10:37:05.077243: step: 308/529, loss: 0.0029857635963708162 2023-01-21 10:37:06.239605: step: 312/529, loss: 0.009916496463119984 2023-01-21 10:37:07.392701: step: 316/529, loss: 0.08734360337257385 2023-01-21 10:37:08.558101: step: 320/529, loss: 0.05480532720685005 2023-01-21 10:37:09.718989: step: 324/529, loss: 0.008849096484482288 2023-01-21 10:37:10.910861: step: 328/529, loss: 0.01684398576617241 2023-01-21 10:37:12.050684: step: 332/529, loss: 0.006241226103156805 2023-01-21 10:37:13.163049: step: 336/529, loss: 0.007660579867660999 2023-01-21 10:37:14.355654: step: 340/529, loss: 0.0035522461403161287 2023-01-21 10:37:15.493854: step: 344/529, loss: 0.11639384925365448 2023-01-21 10:37:16.634015: step: 348/529, loss: 0.045095253735780716 2023-01-21 10:37:17.769680: step: 352/529, loss: 0.024739742279052734 2023-01-21 10:37:18.910759: step: 356/529, loss: 0.10741166770458221 2023-01-21 10:37:20.067481: step: 360/529, loss: 0.0057547567412257195 2023-01-21 10:37:21.192789: step: 364/529, loss: 0.03356199339032173 2023-01-21 10:37:22.340988: step: 368/529, loss: 0.052814170718193054 2023-01-21 10:37:23.477176: step: 372/529, loss: 0.008664015680551529 2023-01-21 10:37:24.612156: step: 376/529, loss: 0.01374511793255806 2023-01-21 10:37:25.772344: step: 380/529, loss: 0.017842482775449753 2023-01-21 10:37:26.916906: step: 384/529, loss: 0.06266383826732635 2023-01-21 10:37:28.085437: step: 388/529, loss: 0.0030565261840820312 2023-01-21 10:37:29.226659: step: 392/529, loss: 0.005564498715102673 2023-01-21 10:37:30.380168: step: 396/529, loss: 0.0850488692522049 2023-01-21 10:37:31.487427: step: 400/529, loss: 0.008472156710922718 2023-01-21 10:37:32.627002: step: 404/529, loss: 0.032532740384340286 2023-01-21 10:37:33.787634: step: 408/529, loss: 0.04038968309760094 2023-01-21 10:37:34.892257: step: 412/529, loss: 0.04558210074901581 2023-01-21 10:37:36.016765: step: 416/529, loss: 0.10349903255701065 2023-01-21 10:37:37.170553: step: 420/529, loss: 0.003133487654849887 2023-01-21 10:37:38.303707: step: 424/529, loss: 0.06093263626098633 2023-01-21 10:37:39.435056: step: 428/529, loss: 0.007752132602035999 2023-01-21 10:37:40.542584: step: 432/529, loss: 0.007301974575966597 2023-01-21 10:37:41.648934: step: 436/529, loss: 0.04609117656946182 2023-01-21 10:37:42.772705: step: 440/529, loss: 0.00887289084494114 2023-01-21 10:37:43.920202: step: 444/529, loss: 0.010225106030702591 2023-01-21 10:37:45.084304: step: 448/529, loss: 0.010163498111069202 2023-01-21 10:37:46.211478: step: 452/529, loss: 0.032201290130615234 2023-01-21 10:37:47.336958: step: 456/529, loss: 0.07714233547449112 2023-01-21 10:37:48.463151: step: 460/529, loss: 0.050862789154052734 2023-01-21 10:37:49.600193: step: 464/529, loss: 0.16744364798069 2023-01-21 10:37:50.727590: step: 468/529, loss: 0.06182861328125 2023-01-21 10:37:51.877415: step: 472/529, loss: 0.0007011413690634072 2023-01-21 10:37:53.004475: step: 476/529, loss: 0.010732650756835938 2023-01-21 10:37:54.147092: step: 480/529, loss: 0.03118162415921688 2023-01-21 10:37:55.297525: step: 484/529, loss: 0.028308773413300514 2023-01-21 10:37:56.440553: step: 488/529, loss: 0.031081723049283028 2023-01-21 10:37:57.556758: step: 492/529, loss: 0.018332291394472122 2023-01-21 10:37:58.770579: step: 496/529, loss: 0.0164140947163105 2023-01-21 10:37:59.924784: step: 500/529, loss: 0.0637664794921875 2023-01-21 10:38:01.058208: step: 504/529, loss: 0.10507993400096893 2023-01-21 10:38:02.208592: step: 508/529, loss: 0.017353488132357597 2023-01-21 10:38:03.363847: step: 512/529, loss: 0.005373883061110973 2023-01-21 10:38:04.506561: step: 516/529, loss: 0.08379878848791122 2023-01-21 10:38:05.639437: step: 520/529, loss: 0.10570697486400604 2023-01-21 10:38:06.785328: step: 524/529, loss: 0.023064803332090378 2023-01-21 10:38:07.898529: step: 528/529, loss: 0.019742680713534355 2023-01-21 10:38:09.007668: step: 532/529, loss: 0.0810873731970787 2023-01-21 10:38:10.139451: step: 536/529, loss: 0.043680526316165924 2023-01-21 10:38:11.264187: step: 540/529, loss: 0.04802761226892471 2023-01-21 10:38:12.444948: step: 544/529, loss: 0.06572437286376953 2023-01-21 10:38:13.589641: step: 548/529, loss: 0.003919410519301891 2023-01-21 10:38:14.721213: step: 552/529, loss: 0.013157462701201439 2023-01-21 10:38:15.870425: step: 556/529, loss: 0.2778494954109192 2023-01-21 10:38:17.023241: step: 560/529, loss: 0.09738927334547043 2023-01-21 10:38:18.193715: step: 564/529, loss: 0.0027718544006347656 2023-01-21 10:38:19.323350: step: 568/529, loss: 0.07343816757202148 2023-01-21 10:38:20.469548: step: 572/529, loss: 0.06742553412914276 2023-01-21 10:38:21.612567: step: 576/529, loss: 0.03533611074090004 2023-01-21 10:38:22.759329: step: 580/529, loss: 0.0640086680650711 2023-01-21 10:38:23.871369: step: 584/529, loss: 0.012063790112733841 2023-01-21 10:38:25.012539: step: 588/529, loss: 0.03139467537403107 2023-01-21 10:38:26.167162: step: 592/529, loss: 0.023262416943907738 2023-01-21 10:38:27.301706: step: 596/529, loss: 0.019202329218387604 2023-01-21 10:38:28.457753: step: 600/529, loss: 0.015586710534989834 2023-01-21 10:38:29.608784: step: 604/529, loss: 0.047947682440280914 2023-01-21 10:38:30.761085: step: 608/529, loss: 0.11125669628381729 2023-01-21 10:38:31.937466: step: 612/529, loss: 0.005725288297981024 2023-01-21 10:38:33.097671: step: 616/529, loss: 0.10224466770887375 2023-01-21 10:38:34.250152: step: 620/529, loss: 0.03097229078412056 2023-01-21 10:38:35.401783: step: 624/529, loss: 0.0586116798222065 2023-01-21 10:38:36.522522: step: 628/529, loss: 0.03724556043744087 2023-01-21 10:38:37.655400: step: 632/529, loss: 0.004291534423828125 2023-01-21 10:38:38.788630: step: 636/529, loss: 0.07390899211168289 2023-01-21 10:38:39.931356: step: 640/529, loss: 0.06829357147216797 2023-01-21 10:38:41.103735: step: 644/529, loss: 0.04636736214160919 2023-01-21 10:38:42.222455: step: 648/529, loss: 0.01751413382589817 2023-01-21 10:38:43.365865: step: 652/529, loss: 0.017524385824799538 2023-01-21 10:38:44.522927: step: 656/529, loss: 0.009670840576291084 2023-01-21 10:38:45.643204: step: 660/529, loss: 0.014430809766054153 2023-01-21 10:38:46.777403: step: 664/529, loss: 0.021681498736143112 2023-01-21 10:38:47.910699: step: 668/529, loss: 0.06616697460412979 2023-01-21 10:38:49.052637: step: 672/529, loss: 0.06775989383459091 2023-01-21 10:38:50.187349: step: 676/529, loss: 0.051607392728328705 2023-01-21 10:38:51.310396: step: 680/529, loss: 0.05058612674474716 2023-01-21 10:38:52.426095: step: 684/529, loss: 0.002528333803638816 2023-01-21 10:38:53.619065: step: 688/529, loss: 0.09183569252490997 2023-01-21 10:38:54.740495: step: 692/529, loss: 0.08151645958423615 2023-01-21 10:38:55.920237: step: 696/529, loss: 0.02961273491382599 2023-01-21 10:38:57.046577: step: 700/529, loss: 0.0008846282726153731 2023-01-21 10:38:58.208503: step: 704/529, loss: 0.15756511688232422 2023-01-21 10:38:59.355785: step: 708/529, loss: 0.06086301803588867 2023-01-21 10:39:00.512484: step: 712/529, loss: 0.01701936684548855 2023-01-21 10:39:01.659880: step: 716/529, loss: 0.0008195400005206466 2023-01-21 10:39:02.816481: step: 720/529, loss: 0.025336647406220436 2023-01-21 10:39:03.953905: step: 724/529, loss: 0.02629675902426243 2023-01-21 10:39:05.100398: step: 728/529, loss: 0.017276477068662643 2023-01-21 10:39:06.220303: step: 732/529, loss: 0.019921448081731796 2023-01-21 10:39:07.314760: step: 736/529, loss: 0.0005456447834149003 2023-01-21 10:39:08.446663: step: 740/529, loss: 0.030028486624360085 2023-01-21 10:39:09.598850: step: 744/529, loss: 0.09304457157850266 2023-01-21 10:39:10.761330: step: 748/529, loss: 0.03932990878820419 2023-01-21 10:39:11.884594: step: 752/529, loss: 0.06434869766235352 2023-01-21 10:39:13.033804: step: 756/529, loss: 0.06484080106019974 2023-01-21 10:39:14.143856: step: 760/529, loss: 0.05992784723639488 2023-01-21 10:39:15.290817: step: 764/529, loss: 0.027353955432772636 2023-01-21 10:39:16.467121: step: 768/529, loss: 0.05796634033322334 2023-01-21 10:39:17.603775: step: 772/529, loss: 0.0020380974747240543 2023-01-21 10:39:18.777001: step: 776/529, loss: 0.014320945367217064 2023-01-21 10:39:19.906199: step: 780/529, loss: 0.029306985437870026 2023-01-21 10:39:21.033165: step: 784/529, loss: 0.045825958251953125 2023-01-21 10:39:22.198949: step: 788/529, loss: 0.006955146789550781 2023-01-21 10:39:23.344808: step: 792/529, loss: 0.03657341003417969 2023-01-21 10:39:24.455280: step: 796/529, loss: 0.011964459903538227 2023-01-21 10:39:25.605465: step: 800/529, loss: 0.05966358631849289 2023-01-21 10:39:26.748290: step: 804/529, loss: 0.002954536583274603 2023-01-21 10:39:27.888526: step: 808/529, loss: 0.00522656412795186 2023-01-21 10:39:29.052285: step: 812/529, loss: 0.002196025801822543 2023-01-21 10:39:30.204222: step: 816/529, loss: 0.0021085739135742188 2023-01-21 10:39:31.325737: step: 820/529, loss: 0.16720938682556152 2023-01-21 10:39:32.454614: step: 824/529, loss: 0.0009832382202148438 2023-01-21 10:39:33.613565: step: 828/529, loss: 0.011444663628935814 2023-01-21 10:39:34.723997: step: 832/529, loss: 0.02932109870016575 2023-01-21 10:39:35.902658: step: 836/529, loss: 0.016179466620087624 2023-01-21 10:39:37.070006: step: 840/529, loss: 0.45787420868873596 2023-01-21 10:39:38.206227: step: 844/529, loss: 0.14280928671360016 2023-01-21 10:39:39.336513: step: 848/529, loss: 0.05953960120677948 2023-01-21 10:39:40.503086: step: 852/529, loss: 0.05152187496423721 2023-01-21 10:39:41.629106: step: 856/529, loss: 0.09777195751667023 2023-01-21 10:39:42.777120: step: 860/529, loss: 0.09475746005773544 2023-01-21 10:39:43.912570: step: 864/529, loss: 0.005778599064797163 2023-01-21 10:39:45.042988: step: 868/529, loss: 0.0038797378074377775 2023-01-21 10:39:46.171981: step: 872/529, loss: 0.0001352310209767893 2023-01-21 10:39:47.317523: step: 876/529, loss: 0.004854965023696423 2023-01-21 10:39:48.446435: step: 880/529, loss: 0.043163299560546875 2023-01-21 10:39:49.581400: step: 884/529, loss: 0.1254405975341797 2023-01-21 10:39:50.722706: step: 888/529, loss: 0.00938568077981472 2023-01-21 10:39:51.855285: step: 892/529, loss: 0.0840395987033844 2023-01-21 10:39:52.957499: step: 896/529, loss: 0.01646900177001953 2023-01-21 10:39:54.139591: step: 900/529, loss: 0.0025002004113048315 2023-01-21 10:39:55.284858: step: 904/529, loss: 0.01778726652264595 2023-01-21 10:39:56.428690: step: 908/529, loss: 0.014309407211840153 2023-01-21 10:39:57.575498: step: 912/529, loss: 0.14741039276123047 2023-01-21 10:39:58.722447: step: 916/529, loss: 0.04818916320800781 2023-01-21 10:39:59.862447: step: 920/529, loss: 0.005638504400849342 2023-01-21 10:40:01.025590: step: 924/529, loss: 0.06700392067432404 2023-01-21 10:40:02.153506: step: 928/529, loss: 0.011551285162568092 2023-01-21 10:40:03.325929: step: 932/529, loss: 0.009420013055205345 2023-01-21 10:40:04.465174: step: 936/529, loss: 0.03510008007287979 2023-01-21 10:40:05.638782: step: 940/529, loss: 0.04221782460808754 2023-01-21 10:40:06.785447: step: 944/529, loss: 0.045285772532224655 2023-01-21 10:40:07.909134: step: 948/529, loss: 0.022705204784870148 2023-01-21 10:40:09.042130: step: 952/529, loss: 0.035432007163763046 2023-01-21 10:40:10.168747: step: 956/529, loss: 0.013840675354003906 2023-01-21 10:40:11.328335: step: 960/529, loss: 0.04016897827386856 2023-01-21 10:40:12.464986: step: 964/529, loss: 0.0026357651222497225 2023-01-21 10:40:13.581677: step: 968/529, loss: 0.06760015338659286 2023-01-21 10:40:14.768905: step: 972/529, loss: 0.04548444598913193 2023-01-21 10:40:15.886987: step: 976/529, loss: 0.06658191233873367 2023-01-21 10:40:17.035159: step: 980/529, loss: 0.024829866364598274 2023-01-21 10:40:18.150103: step: 984/529, loss: 0.06164813041687012 2023-01-21 10:40:19.299985: step: 988/529, loss: 0.03172893822193146 2023-01-21 10:40:20.432624: step: 992/529, loss: 0.03628425672650337 2023-01-21 10:40:21.552560: step: 996/529, loss: 0.012101937085390091 2023-01-21 10:40:22.727323: step: 1000/529, loss: 0.0015178680187091231 2023-01-21 10:40:23.873101: step: 1004/529, loss: 0.0031326294410973787 2023-01-21 10:40:25.012463: step: 1008/529, loss: 0.0817171111702919 2023-01-21 10:40:26.170273: step: 1012/529, loss: 0.04501743242144585 2023-01-21 10:40:27.301899: step: 1016/529, loss: 0.007783698849380016 2023-01-21 10:40:28.437208: step: 1020/529, loss: 0.00141820905264467 2023-01-21 10:40:29.597702: step: 1024/529, loss: 0.03396713733673096 2023-01-21 10:40:30.751378: step: 1028/529, loss: 0.08785443007946014 2023-01-21 10:40:31.895176: step: 1032/529, loss: 0.005506038665771484 2023-01-21 10:40:33.089585: step: 1036/529, loss: 0.7554685473442078 2023-01-21 10:40:34.244178: step: 1040/529, loss: 0.10742369294166565 2023-01-21 10:40:35.370268: step: 1044/529, loss: 0.042765043675899506 2023-01-21 10:40:36.543894: step: 1048/529, loss: 0.010907554998993874 2023-01-21 10:40:37.656507: step: 1052/529, loss: 0.09366540610790253 2023-01-21 10:40:38.815276: step: 1056/529, loss: 0.0018295288318768144 2023-01-21 10:40:39.953477: step: 1060/529, loss: 0.03758659586310387 2023-01-21 10:40:41.097030: step: 1064/529, loss: 0.0420774444937706 2023-01-21 10:40:42.218304: step: 1068/529, loss: 0.03819551318883896 2023-01-21 10:40:43.352946: step: 1072/529, loss: 0.022656060755252838 2023-01-21 10:40:44.492495: step: 1076/529, loss: 0.004863167181611061 2023-01-21 10:40:45.631408: step: 1080/529, loss: 0.003184795379638672 2023-01-21 10:40:46.789937: step: 1084/529, loss: 0.009426689706742764 2023-01-21 10:40:47.921293: step: 1088/529, loss: 0.060143519192934036 2023-01-21 10:40:49.046950: step: 1092/529, loss: 0.0052474020048975945 2023-01-21 10:40:50.182304: step: 1096/529, loss: 0.007507228758186102 2023-01-21 10:40:51.315124: step: 1100/529, loss: 0.034093573689460754 2023-01-21 10:40:52.487219: step: 1104/529, loss: 0.039986610412597656 2023-01-21 10:40:53.623356: step: 1108/529, loss: 0.027509402483701706 2023-01-21 10:40:54.774917: step: 1112/529, loss: 0.013420677743852139 2023-01-21 10:40:55.925310: step: 1116/529, loss: 0.024685287848114967 2023-01-21 10:40:57.060740: step: 1120/529, loss: 0.05600138008594513 2023-01-21 10:40:58.185912: step: 1124/529, loss: 0.1418040245771408 2023-01-21 10:40:59.314250: step: 1128/529, loss: 0.01038274820894003 2023-01-21 10:41:00.477127: step: 1132/529, loss: 0.036375001072883606 2023-01-21 10:41:01.639054: step: 1136/529, loss: 0.01700592041015625 2023-01-21 10:41:02.763601: step: 1140/529, loss: 0.0056282044388353825 2023-01-21 10:41:03.909154: step: 1144/529, loss: 0.006110859103500843 2023-01-21 10:41:05.037435: step: 1148/529, loss: 0.06296902149915695 2023-01-21 10:41:06.169662: step: 1152/529, loss: 0.11532774567604065 2023-01-21 10:41:07.293604: step: 1156/529, loss: 0.0735454112291336 2023-01-21 10:41:08.431539: step: 1160/529, loss: 0.2641119062900543 2023-01-21 10:41:09.581232: step: 1164/529, loss: 0.0187851432710886 2023-01-21 10:41:10.711487: step: 1168/529, loss: 0.018866728991270065 2023-01-21 10:41:11.831679: step: 1172/529, loss: 0.04534149542450905 2023-01-21 10:41:13.019180: step: 1176/529, loss: 0.037976741790771484 2023-01-21 10:41:14.187550: step: 1180/529, loss: 0.03110940381884575 2023-01-21 10:41:15.339046: step: 1184/529, loss: 0.043977927416563034 2023-01-21 10:41:16.473504: step: 1188/529, loss: 0.016446590423583984 2023-01-21 10:41:17.611500: step: 1192/529, loss: 0.009401321411132812 2023-01-21 10:41:18.748339: step: 1196/529, loss: 0.005697441287338734 2023-01-21 10:41:19.889462: step: 1200/529, loss: 0.054293442517519 2023-01-21 10:41:21.027496: step: 1204/529, loss: 0.02964916080236435 2023-01-21 10:41:22.130931: step: 1208/529, loss: 0.07031688839197159 2023-01-21 10:41:23.292144: step: 1212/529, loss: 0.049356937408447266 2023-01-21 10:41:24.439451: step: 1216/529, loss: 0.017296219244599342 2023-01-21 10:41:25.578693: step: 1220/529, loss: 0.006849432364106178 2023-01-21 10:41:26.732620: step: 1224/529, loss: 0.007527542300522327 2023-01-21 10:41:27.878756: step: 1228/529, loss: 0.06866807490587234 2023-01-21 10:41:29.018031: step: 1232/529, loss: 0.18938341736793518 2023-01-21 10:41:30.175674: step: 1236/529, loss: 0.08155861496925354 2023-01-21 10:41:31.324068: step: 1240/529, loss: 0.029927443712949753 2023-01-21 10:41:32.450272: step: 1244/529, loss: 0.051896288990974426 2023-01-21 10:41:33.575515: step: 1248/529, loss: 0.024506758898496628 2023-01-21 10:41:34.732624: step: 1252/529, loss: 0.05853986740112305 2023-01-21 10:41:35.873095: step: 1256/529, loss: 0.001984023954719305 2023-01-21 10:41:37.011000: step: 1260/529, loss: 0.04619855806231499 2023-01-21 10:41:38.148255: step: 1264/529, loss: 0.047243308275938034 2023-01-21 10:41:39.283711: step: 1268/529, loss: 0.04430665820837021 2023-01-21 10:41:40.409302: step: 1272/529, loss: 0.01496591605246067 2023-01-21 10:41:41.540627: step: 1276/529, loss: 0.09296198189258575 2023-01-21 10:41:42.686468: step: 1280/529, loss: 0.0359838493168354 2023-01-21 10:41:43.812122: step: 1284/529, loss: 0.029023757204413414 2023-01-21 10:41:44.981820: step: 1288/529, loss: 0.9746854901313782 2023-01-21 10:41:46.120961: step: 1292/529, loss: 0.18511143326759338 2023-01-21 10:41:47.281005: step: 1296/529, loss: 0.04672403261065483 2023-01-21 10:41:48.428624: step: 1300/529, loss: 0.03355436399579048 2023-01-21 10:41:49.553295: step: 1304/529, loss: 0.023637771606445312 2023-01-21 10:41:50.704449: step: 1308/529, loss: 0.12433824688196182 2023-01-21 10:41:51.860877: step: 1312/529, loss: 0.0066044810228049755 2023-01-21 10:41:53.010653: step: 1316/529, loss: 0.0436527244746685 2023-01-21 10:41:54.132038: step: 1320/529, loss: 0.0006201743963174522 2023-01-21 10:41:55.248019: step: 1324/529, loss: 0.024755384773015976 2023-01-21 10:41:56.415817: step: 1328/529, loss: 0.07917232811450958 2023-01-21 10:41:57.595763: step: 1332/529, loss: 0.013935089111328125 2023-01-21 10:41:58.736042: step: 1336/529, loss: 0.05790986865758896 2023-01-21 10:41:59.871205: step: 1340/529, loss: 0.013862133957445621 2023-01-21 10:42:01.015650: step: 1344/529, loss: 0.01408529281616211 2023-01-21 10:42:02.133530: step: 1348/529, loss: 0.019444739446043968 2023-01-21 10:42:03.277675: step: 1352/529, loss: 0.06703615188598633 2023-01-21 10:42:04.430290: step: 1356/529, loss: 0.14207935333251953 2023-01-21 10:42:05.587788: step: 1360/529, loss: 0.0772436112165451 2023-01-21 10:42:06.735670: step: 1364/529, loss: 0.047171495854854584 2023-01-21 10:42:07.907134: step: 1368/529, loss: 0.005540227983146906 2023-01-21 10:42:09.067111: step: 1372/529, loss: 0.040296077728271484 2023-01-21 10:42:10.226980: step: 1376/529, loss: 0.007160187233239412 2023-01-21 10:42:11.335819: step: 1380/529, loss: 0.008231687359511852 2023-01-21 10:42:12.467098: step: 1384/529, loss: 0.013277912512421608 2023-01-21 10:42:13.567187: step: 1388/529, loss: 0.23824390769004822 2023-01-21 10:42:14.714035: step: 1392/529, loss: 0.006083154585212469 2023-01-21 10:42:15.834286: step: 1396/529, loss: 0.03039703331887722 2023-01-21 10:42:16.952974: step: 1400/529, loss: 0.040851689875125885 2023-01-21 10:42:18.131180: step: 1404/529, loss: 0.013806152157485485 2023-01-21 10:42:19.302104: step: 1408/529, loss: 0.12582091987133026 2023-01-21 10:42:20.429187: step: 1412/529, loss: 0.04688234254717827 2023-01-21 10:42:21.581687: step: 1416/529, loss: 0.014122772961854935 2023-01-21 10:42:22.684942: step: 1420/529, loss: 0.15463829040527344 2023-01-21 10:42:23.831382: step: 1424/529, loss: 0.06169462203979492 2023-01-21 10:42:24.962759: step: 1428/529, loss: 0.1042117103934288 2023-01-21 10:42:26.114323: step: 1432/529, loss: 0.7648319602012634 2023-01-21 10:42:27.258048: step: 1436/529, loss: 0.006183910649269819 2023-01-21 10:42:28.389920: step: 1440/529, loss: 0.13220424950122833 2023-01-21 10:42:29.530878: step: 1444/529, loss: 0.47135886549949646 2023-01-21 10:42:30.654563: step: 1448/529, loss: 0.3208070695400238 2023-01-21 10:42:31.796601: step: 1452/529, loss: 0.05192179977893829 2023-01-21 10:42:32.957094: step: 1456/529, loss: 0.11350338160991669 2023-01-21 10:42:34.091670: step: 1460/529, loss: 0.023944854736328125 2023-01-21 10:42:35.203948: step: 1464/529, loss: 0.006532740779221058 2023-01-21 10:42:36.362980: step: 1468/529, loss: 0.025796595960855484 2023-01-21 10:42:37.545683: step: 1472/529, loss: 0.01515111979097128 2023-01-21 10:42:38.693311: step: 1476/529, loss: 0.006575917825102806 2023-01-21 10:42:39.854015: step: 1480/529, loss: 0.0012142182094976306 2023-01-21 10:42:41.004798: step: 1484/529, loss: 0.0015543936751782894 2023-01-21 10:42:42.162977: step: 1488/529, loss: 0.6960460543632507 2023-01-21 10:42:43.346681: step: 1492/529, loss: 0.13783079385757446 2023-01-21 10:42:44.495343: step: 1496/529, loss: 0.002954196883365512 2023-01-21 10:42:45.625492: step: 1500/529, loss: 0.04598264768719673 2023-01-21 10:42:46.786806: step: 1504/529, loss: 0.04302208125591278 2023-01-21 10:42:47.926137: step: 1508/529, loss: 0.13760700821876526 2023-01-21 10:42:49.064979: step: 1512/529, loss: 0.08358879387378693 2023-01-21 10:42:50.212488: step: 1516/529, loss: 0.03550577163696289 2023-01-21 10:42:51.310348: step: 1520/529, loss: 0.05867529287934303 2023-01-21 10:42:52.447138: step: 1524/529, loss: 0.015043259598314762 2023-01-21 10:42:53.595401: step: 1528/529, loss: 0.049494028091430664 2023-01-21 10:42:54.753241: step: 1532/529, loss: 0.058661751449108124 2023-01-21 10:42:55.898839: step: 1536/529, loss: 0.0775054469704628 2023-01-21 10:42:57.034556: step: 1540/529, loss: 0.40463846921920776 2023-01-21 10:42:58.147098: step: 1544/529, loss: 0.12960176169872284 2023-01-21 10:42:59.283684: step: 1548/529, loss: 0.11314563453197479 2023-01-21 10:43:00.406512: step: 1552/529, loss: 0.008826208300888538 2023-01-21 10:43:01.562631: step: 1556/529, loss: 0.018711376935243607 2023-01-21 10:43:02.730218: step: 1560/529, loss: 0.4075283110141754 2023-01-21 10:43:03.884671: step: 1564/529, loss: 0.4174271523952484 2023-01-21 10:43:05.054347: step: 1568/529, loss: 0.0032883407548069954 2023-01-21 10:43:06.179918: step: 1572/529, loss: 0.028062917292118073 2023-01-21 10:43:07.348643: step: 1576/529, loss: 0.06418509781360626 2023-01-21 10:43:08.514395: step: 1580/529, loss: 0.39848265051841736 2023-01-21 10:43:09.640984: step: 1584/529, loss: 0.03924448415637016 2023-01-21 10:43:10.786717: step: 1588/529, loss: 0.014502717182040215 2023-01-21 10:43:11.951319: step: 1592/529, loss: 0.023236466571688652 2023-01-21 10:43:13.076407: step: 1596/529, loss: 0.36373335123062134 2023-01-21 10:43:14.263087: step: 1600/529, loss: 0.06777782738208771 2023-01-21 10:43:15.419984: step: 1604/529, loss: 0.05396628752350807 2023-01-21 10:43:16.568867: step: 1608/529, loss: 0.03881688043475151 2023-01-21 10:43:17.722761: step: 1612/529, loss: 0.02200622670352459 2023-01-21 10:43:18.869877: step: 1616/529, loss: 0.04023432731628418 2023-01-21 10:43:19.993472: step: 1620/529, loss: 0.06971673667430878 2023-01-21 10:43:21.103320: step: 1624/529, loss: 0.08448925614356995 2023-01-21 10:43:22.228012: step: 1628/529, loss: 0.04024811089038849 2023-01-21 10:43:23.375674: step: 1632/529, loss: 0.024126671254634857 2023-01-21 10:43:24.500646: step: 1636/529, loss: 0.0516846664249897 2023-01-21 10:43:25.629309: step: 1640/529, loss: 0.04051714017987251 2023-01-21 10:43:26.778892: step: 1644/529, loss: 0.009109973907470703 2023-01-21 10:43:27.937487: step: 1648/529, loss: 0.03788652643561363 2023-01-21 10:43:29.104401: step: 1652/529, loss: 0.011710261926054955 2023-01-21 10:43:30.249882: step: 1656/529, loss: 0.02306833304464817 2023-01-21 10:43:31.436832: step: 1660/529, loss: 0.04119815677404404 2023-01-21 10:43:32.611086: step: 1664/529, loss: 0.0015746117569506168 2023-01-21 10:43:33.741062: step: 1668/529, loss: 0.017468642443418503 2023-01-21 10:43:34.857686: step: 1672/529, loss: 0.002208232879638672 2023-01-21 10:43:35.990388: step: 1676/529, loss: 0.017556989565491676 2023-01-21 10:43:37.132923: step: 1680/529, loss: 0.08904094994068146 2023-01-21 10:43:38.279608: step: 1684/529, loss: 0.02933357283473015 2023-01-21 10:43:39.408848: step: 1688/529, loss: 0.07072000950574875 2023-01-21 10:43:40.546841: step: 1692/529, loss: 0.011268901638686657 2023-01-21 10:43:41.654618: step: 1696/529, loss: 0.06895160675048828 2023-01-21 10:43:42.867551: step: 1700/529, loss: 0.028394699096679688 2023-01-21 10:43:44.072029: step: 1704/529, loss: 1.0956612825393677 2023-01-21 10:43:45.232606: step: 1708/529, loss: 0.042693473398685455 2023-01-21 10:43:46.406071: step: 1712/529, loss: 0.106645368039608 2023-01-21 10:43:47.573459: step: 1716/529, loss: 0.01604471169412136 2023-01-21 10:43:48.713417: step: 1720/529, loss: 0.006686401553452015 2023-01-21 10:43:49.856462: step: 1724/529, loss: 0.19760455191135406 2023-01-21 10:43:51.013023: step: 1728/529, loss: 0.04504687711596489 2023-01-21 10:43:52.167936: step: 1732/529, loss: 0.029277395457029343 2023-01-21 10:43:53.306296: step: 1736/529, loss: 0.05705766752362251 2023-01-21 10:43:54.465560: step: 1740/529, loss: 0.13684529066085815 2023-01-21 10:43:55.626487: step: 1744/529, loss: 0.03550753742456436 2023-01-21 10:43:56.760736: step: 1748/529, loss: 0.04758463054895401 2023-01-21 10:43:57.881253: step: 1752/529, loss: 0.012179756537079811 2023-01-21 10:43:59.009476: step: 1756/529, loss: 0.02662181854248047 2023-01-21 10:44:00.162695: step: 1760/529, loss: 0.08154463768005371 2023-01-21 10:44:01.303892: step: 1764/529, loss: 0.08133916556835175 2023-01-21 10:44:02.434362: step: 1768/529, loss: 0.00993881281465292 2023-01-21 10:44:03.591070: step: 1772/529, loss: 0.06253661960363388 2023-01-21 10:44:04.701068: step: 1776/529, loss: 0.01262588519603014 2023-01-21 10:44:05.849491: step: 1780/529, loss: 0.01673126220703125 2023-01-21 10:44:06.967691: step: 1784/529, loss: 0.015593529678881168 2023-01-21 10:44:08.131854: step: 1788/529, loss: 0.07304706424474716 2023-01-21 10:44:09.247278: step: 1792/529, loss: 0.19124622642993927 2023-01-21 10:44:10.400636: step: 1796/529, loss: 0.04656529426574707 2023-01-21 10:44:11.522744: step: 1800/529, loss: 0.0666409507393837 2023-01-21 10:44:12.682126: step: 1804/529, loss: 0.12291279435157776 2023-01-21 10:44:13.832048: step: 1808/529, loss: 0.03063497692346573 2023-01-21 10:44:15.005176: step: 1812/529, loss: 0.010074997320771217 2023-01-21 10:44:16.168853: step: 1816/529, loss: 0.04221305996179581 2023-01-21 10:44:17.308530: step: 1820/529, loss: 0.04090938717126846 2023-01-21 10:44:18.441629: step: 1824/529, loss: 0.10782096534967422 2023-01-21 10:44:19.585731: step: 1828/529, loss: 0.04785943031311035 2023-01-21 10:44:20.707769: step: 1832/529, loss: 0.00011295080184936523 2023-01-21 10:44:21.838809: step: 1836/529, loss: 0.579179584980011 2023-01-21 10:44:23.010004: step: 1840/529, loss: 0.04755387455224991 2023-01-21 10:44:24.149980: step: 1844/529, loss: 0.013316750526428223 2023-01-21 10:44:25.310741: step: 1848/529, loss: 0.03761430084705353 2023-01-21 10:44:26.463489: step: 1852/529, loss: 0.0030691148713231087 2023-01-21 10:44:27.584140: step: 1856/529, loss: 0.036391064524650574 2023-01-21 10:44:28.719836: step: 1860/529, loss: 0.004856681916862726 2023-01-21 10:44:29.845433: step: 1864/529, loss: 0.039927005767822266 2023-01-21 10:44:31.002946: step: 1868/529, loss: 0.05290422588586807 2023-01-21 10:44:32.143132: step: 1872/529, loss: 0.09407234191894531 2023-01-21 10:44:33.246247: step: 1876/529, loss: 0.024172117933630943 2023-01-21 10:44:34.361450: step: 1880/529, loss: 0.032428935170173645 2023-01-21 10:44:35.477325: step: 1884/529, loss: 0.04099474102258682 2023-01-21 10:44:36.625044: step: 1888/529, loss: 0.0036382675170898438 2023-01-21 10:44:37.745540: step: 1892/529, loss: 0.02134869061410427 2023-01-21 10:44:38.882110: step: 1896/529, loss: 0.02146444469690323 2023-01-21 10:44:40.060466: step: 1900/529, loss: 0.0011233330005779862 2023-01-21 10:44:41.219391: step: 1904/529, loss: 0.14567409455776215 2023-01-21 10:44:42.394679: step: 1908/529, loss: 0.015575027093291283 2023-01-21 10:44:43.511853: step: 1912/529, loss: 0.0479520782828331 2023-01-21 10:44:44.634272: step: 1916/529, loss: 0.02944030798971653 2023-01-21 10:44:45.766288: step: 1920/529, loss: 0.016190147027373314 2023-01-21 10:44:46.899863: step: 1924/529, loss: 0.037545107305049896 2023-01-21 10:44:48.038591: step: 1928/529, loss: 0.02879791334271431 2023-01-21 10:44:49.167218: step: 1932/529, loss: 0.01570892333984375 2023-01-21 10:44:50.322468: step: 1936/529, loss: 0.06906204670667648 2023-01-21 10:44:51.469748: step: 1940/529, loss: 0.023740483447909355 2023-01-21 10:44:52.618351: step: 1944/529, loss: 0.07427291572093964 2023-01-21 10:44:53.754573: step: 1948/529, loss: 0.023793792352080345 2023-01-21 10:44:54.870765: step: 1952/529, loss: 0.027672769501805305 2023-01-21 10:44:56.022555: step: 1956/529, loss: 0.024505263194441795 2023-01-21 10:44:57.168693: step: 1960/529, loss: 0.012414837256073952 2023-01-21 10:44:58.308817: step: 1964/529, loss: 0.026230812072753906 2023-01-21 10:44:59.446891: step: 1968/529, loss: 0.018483638763427734 2023-01-21 10:45:00.573313: step: 1972/529, loss: 9.117127046920359e-05 2023-01-21 10:45:01.715524: step: 1976/529, loss: 0.12292156368494034 2023-01-21 10:45:02.868974: step: 1980/529, loss: 0.013800477609038353 2023-01-21 10:45:04.023564: step: 1984/529, loss: 0.01651925966143608 2023-01-21 10:45:05.158436: step: 1988/529, loss: 0.02746448665857315 2023-01-21 10:45:06.295518: step: 1992/529, loss: 0.20702160894870758 2023-01-21 10:45:07.430983: step: 1996/529, loss: 0.06955838203430176 2023-01-21 10:45:08.542711: step: 2000/529, loss: 0.048099711537361145 2023-01-21 10:45:09.659717: step: 2004/529, loss: 0.0075654988177120686 2023-01-21 10:45:10.792015: step: 2008/529, loss: 0.06574840843677521 2023-01-21 10:45:11.950433: step: 2012/529, loss: 0.01720752939581871 2023-01-21 10:45:13.101815: step: 2016/529, loss: 0.03441762924194336 2023-01-21 10:45:14.219279: step: 2020/529, loss: 0.0008242607582360506 2023-01-21 10:45:15.360416: step: 2024/529, loss: 0.07475261390209198 2023-01-21 10:45:16.485770: step: 2028/529, loss: 0.09648562222719193 2023-01-21 10:45:17.632156: step: 2032/529, loss: 0.03050060383975506 2023-01-21 10:45:18.773623: step: 2036/529, loss: 0.020706940442323685 2023-01-21 10:45:19.918264: step: 2040/529, loss: 0.03740882873535156 2023-01-21 10:45:21.085885: step: 2044/529, loss: 0.047423794865608215 2023-01-21 10:45:22.231805: step: 2048/529, loss: 0.054579734802246094 2023-01-21 10:45:23.356606: step: 2052/529, loss: 0.006700319238007069 2023-01-21 10:45:24.467612: step: 2056/529, loss: 0.00593195017427206 2023-01-21 10:45:25.594305: step: 2060/529, loss: 0.9261052012443542 2023-01-21 10:45:26.760451: step: 2064/529, loss: 0.010789108462631702 2023-01-21 10:45:27.885914: step: 2068/529, loss: 0.01773672178387642 2023-01-21 10:45:29.024252: step: 2072/529, loss: 0.025397587567567825 2023-01-21 10:45:30.144982: step: 2076/529, loss: 0.017604541033506393 2023-01-21 10:45:31.256710: step: 2080/529, loss: 0.0047149658203125 2023-01-21 10:45:32.402313: step: 2084/529, loss: 0.09763198345899582 2023-01-21 10:45:33.543865: step: 2088/529, loss: 0.22763033211231232 2023-01-21 10:45:34.690015: step: 2092/529, loss: 0.12451066821813583 2023-01-21 10:45:35.834713: step: 2096/529, loss: 0.0216109286993742 2023-01-21 10:45:36.970333: step: 2100/529, loss: 0.05923175811767578 2023-01-21 10:45:38.101025: step: 2104/529, loss: 0.08895035088062286 2023-01-21 10:45:39.231602: step: 2108/529, loss: 0.4702809751033783 2023-01-21 10:45:40.403899: step: 2112/529, loss: 0.03024921379983425 2023-01-21 10:45:41.535205: step: 2116/529, loss: 0.03227867931127548 ================================================== Loss: 0.061 -------------------- Dev: {'event': {'p': 0.6184782608695653, 'r': 0.7576564580559254, 'f1': 0.6810293237582286}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Test: {'event': {'p': 0.6239924134660977, 'r': 0.7624565469293163, 'f1': 0.6863102998696219}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Chinese: {'event': {'p': 0.5348837209302325, 'r': 0.8518518518518519, 'f1': 0.6571428571428571}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Korean: {'event': {'p': 0.6851851851851852, 'r': 0.5873015873015873, 'f1': 0.6324786324786325}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} 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': 12} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** Epoch: 13 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:46:24.071370: step: 4/529, loss: 0.019993305206298828 2023-01-21 10:46:25.216737: step: 8/529, loss: 0.02758360095322132 2023-01-21 10:46:26.350460: step: 12/529, loss: 0.0386652834713459 2023-01-21 10:46:27.497660: step: 16/529, loss: 0.04045925289392471 2023-01-21 10:46:28.629434: step: 20/529, loss: 0.005299663636833429 2023-01-21 10:46:29.761328: step: 24/529, loss: 0.03268828243017197 2023-01-21 10:46:30.923528: step: 28/529, loss: 0.006821060553193092 2023-01-21 10:46:32.062822: step: 32/529, loss: 0.060141950845718384 2023-01-21 10:46:33.186466: step: 36/529, loss: 0.08947410434484482 2023-01-21 10:46:34.333019: step: 40/529, loss: 0.005906868260353804 2023-01-21 10:46:35.494669: step: 44/529, loss: 0.005701541900634766 2023-01-21 10:46:36.650674: step: 48/529, loss: 0.05475273355841637 2023-01-21 10:46:37.769543: step: 52/529, loss: 0.049897994846105576 2023-01-21 10:46:38.901786: step: 56/529, loss: 0.0092627527192235 2023-01-21 10:46:40.039280: step: 60/529, loss: 0.003398799803107977 2023-01-21 10:46:41.185582: step: 64/529, loss: 0.04215526580810547 2023-01-21 10:46:42.372591: step: 68/529, loss: 0.0023714066483080387 2023-01-21 10:46:43.548141: step: 72/529, loss: 0.02426781691610813 2023-01-21 10:46:44.670694: step: 76/529, loss: 0.058391764760017395 2023-01-21 10:46:45.782936: step: 80/529, loss: 0.05724773555994034 2023-01-21 10:46:46.904433: step: 84/529, loss: 0.014653205871582031 2023-01-21 10:46:48.071570: step: 88/529, loss: 0.04702577739953995 2023-01-21 10:46:49.219862: step: 92/529, loss: 0.12947197258472443 2023-01-21 10:46:50.386062: step: 96/529, loss: 0.012953282333910465 2023-01-21 10:46:51.559606: step: 100/529, loss: 0.04952354356646538 2023-01-21 10:46:52.712318: step: 104/529, loss: 0.0255808774381876 2023-01-21 10:46:53.883108: step: 108/529, loss: 0.04384269565343857 2023-01-21 10:46:55.017434: step: 112/529, loss: 0.0018137454753741622 2023-01-21 10:46:56.136046: step: 116/529, loss: 0.04469194635748863 2023-01-21 10:46:57.322222: step: 120/529, loss: 0.05742548406124115 2023-01-21 10:46:58.459858: step: 124/529, loss: 0.003180885221809149 2023-01-21 10:46:59.594038: step: 128/529, loss: 0.054470207542181015 2023-01-21 10:47:00.746792: step: 132/529, loss: 0.07437200844287872 2023-01-21 10:47:01.885353: step: 136/529, loss: 0.042356204241514206 2023-01-21 10:47:03.029503: step: 140/529, loss: 0.2078346312046051 2023-01-21 10:47:04.167336: step: 144/529, loss: 0.18840885162353516 2023-01-21 10:47:05.303238: step: 148/529, loss: 0.013764859177172184 2023-01-21 10:47:06.424541: step: 152/529, loss: 0.019231222569942474 2023-01-21 10:47:07.548932: step: 156/529, loss: 0.061971474438905716 2023-01-21 10:47:08.693597: step: 160/529, loss: 0.12830591201782227 2023-01-21 10:47:09.849148: step: 164/529, loss: 0.033579349517822266 2023-01-21 10:47:10.992498: step: 168/529, loss: 0.030616380274295807 2023-01-21 10:47:12.149603: step: 172/529, loss: 0.015859533101320267 2023-01-21 10:47:13.292471: step: 176/529, loss: 0.0979946106672287 2023-01-21 10:47:14.484774: step: 180/529, loss: 0.019607067108154297 2023-01-21 10:47:15.623342: step: 184/529, loss: 0.047873977571725845 2023-01-21 10:47:16.763951: step: 188/529, loss: 0.055992770940065384 2023-01-21 10:47:17.906261: step: 192/529, loss: 0.07307066768407822 2023-01-21 10:47:19.029045: step: 196/529, loss: 0.0008630752563476562 2023-01-21 10:47:20.189390: step: 200/529, loss: 0.02937774732708931 2023-01-21 10:47:21.337947: step: 204/529, loss: 0.07343216240406036 2023-01-21 10:47:22.453966: step: 208/529, loss: 0.014411927200853825 2023-01-21 10:47:23.593621: step: 212/529, loss: 0.05884857103228569 2023-01-21 10:47:24.707448: step: 216/529, loss: 0.004012203309684992 2023-01-21 10:47:25.808960: step: 220/529, loss: 0.05199480056762695 2023-01-21 10:47:26.974448: step: 224/529, loss: 0.031542565673589706 2023-01-21 10:47:28.123784: step: 228/529, loss: 0.0015168190002441406 2023-01-21 10:47:29.273190: step: 232/529, loss: 0.005208874121308327 2023-01-21 10:47:30.422257: step: 236/529, loss: 0.05750169977545738 2023-01-21 10:47:31.539596: step: 240/529, loss: 0.0014677047729492188 2023-01-21 10:47:32.713066: step: 244/529, loss: 0.0011476516956463456 2023-01-21 10:47:33.867341: step: 248/529, loss: 0.06792517006397247 2023-01-21 10:47:35.023361: step: 252/529, loss: 0.0019504786469042301 2023-01-21 10:47:36.172652: step: 256/529, loss: 0.0005694389110431075 2023-01-21 10:47:37.307331: step: 260/529, loss: 0.0578581839799881 2023-01-21 10:47:38.479058: step: 264/529, loss: 0.007916033267974854 2023-01-21 10:47:39.622307: step: 268/529, loss: 0.030933380126953125 2023-01-21 10:47:40.733457: step: 272/529, loss: 0.00893621426075697 2023-01-21 10:47:41.883707: step: 276/529, loss: 0.07375459372997284 2023-01-21 10:47:42.999218: step: 280/529, loss: 0.009297228418290615 2023-01-21 10:47:44.147586: step: 284/529, loss: 0.004671478644013405 2023-01-21 10:47:45.302111: step: 288/529, loss: 0.05834665521979332 2023-01-21 10:47:46.422694: step: 292/529, loss: 0.00677533121779561 2023-01-21 10:47:47.557413: step: 296/529, loss: 0.20710425078868866 2023-01-21 10:47:48.693026: step: 300/529, loss: 0.02287016063928604 2023-01-21 10:47:49.835297: step: 304/529, loss: 0.0005518913385458291 2023-01-21 10:47:50.973473: step: 308/529, loss: 0.06621570885181427 2023-01-21 10:47:52.113916: step: 312/529, loss: 0.01689929887652397 2023-01-21 10:47:53.245068: step: 316/529, loss: 0.09396085143089294 2023-01-21 10:47:54.397639: step: 320/529, loss: 0.012692594900727272 2023-01-21 10:47:55.534982: step: 324/529, loss: 0.0017153739463537931 2023-01-21 10:47:56.676101: step: 328/529, loss: 0.034481048583984375 2023-01-21 10:47:57.792843: step: 332/529, loss: 0.02468099445104599 2023-01-21 10:47:58.924301: step: 336/529, loss: 0.05512991175055504 2023-01-21 10:48:00.063305: step: 340/529, loss: 0.015399790368974209 2023-01-21 10:48:01.214546: step: 344/529, loss: 0.028708746656775475 2023-01-21 10:48:02.348038: step: 348/529, loss: 3.5144460201263428 2023-01-21 10:48:03.475562: step: 352/529, loss: 0.0008200168376788497 2023-01-21 10:48:04.598387: step: 356/529, loss: 0.02463064342737198 2023-01-21 10:48:05.699675: step: 360/529, loss: 0.004363822750747204 2023-01-21 10:48:06.815487: step: 364/529, loss: 0.010788536630570889 2023-01-21 10:48:07.954897: step: 368/529, loss: 0.01725492626428604 2023-01-21 10:48:09.110790: step: 372/529, loss: 0.04038182646036148 2023-01-21 10:48:10.242219: step: 376/529, loss: 0.07231082767248154 2023-01-21 10:48:11.362987: step: 380/529, loss: 0.012486457824707031 2023-01-21 10:48:12.510929: step: 384/529, loss: 0.0014396667247638106 2023-01-21 10:48:13.649252: step: 388/529, loss: 0.021851444616913795 2023-01-21 10:48:14.767332: step: 392/529, loss: 0.021245192736387253 2023-01-21 10:48:15.911979: step: 396/529, loss: 0.08244087547063828 2023-01-21 10:48:17.051166: step: 400/529, loss: 0.04936408996582031 2023-01-21 10:48:18.207477: step: 404/529, loss: 0.00750045757740736 2023-01-21 10:48:19.353913: step: 408/529, loss: 0.017092514783143997 2023-01-21 10:48:20.476808: step: 412/529, loss: 0.03356580808758736 2023-01-21 10:48:21.625151: step: 416/529, loss: 0.01418156735599041 2023-01-21 10:48:22.764514: step: 420/529, loss: 0.03341960906982422 2023-01-21 10:48:23.895609: step: 424/529, loss: 0.2953829765319824 2023-01-21 10:48:25.027314: step: 428/529, loss: 0.0011455535423010588 2023-01-21 10:48:26.179815: step: 432/529, loss: 0.01996469683945179 2023-01-21 10:48:27.356609: step: 436/529, loss: 0.0800243392586708 2023-01-21 10:48:28.494148: step: 440/529, loss: 0.06483154743909836 2023-01-21 10:48:29.622894: step: 444/529, loss: 0.03417787700891495 2023-01-21 10:48:30.760909: step: 448/529, loss: 0.09809355437755585 2023-01-21 10:48:31.898324: step: 452/529, loss: 0.025599099695682526 2023-01-21 10:48:33.035994: step: 456/529, loss: 0.0006878852727822959 2023-01-21 10:48:34.183276: step: 460/529, loss: 0.0422239787876606 2023-01-21 10:48:35.329472: step: 464/529, loss: 0.0010572910541668534 2023-01-21 10:48:36.479062: step: 468/529, loss: 0.024126624688506126 2023-01-21 10:48:37.588901: step: 472/529, loss: 0.016437530517578125 2023-01-21 10:48:38.753181: step: 476/529, loss: 0.09520583599805832 2023-01-21 10:48:39.869509: step: 480/529, loss: 0.008626163005828857 2023-01-21 10:48:41.004217: step: 484/529, loss: 0.19090357422828674 2023-01-21 10:48:42.207380: step: 488/529, loss: 0.00013589859008789062 2023-01-21 10:48:43.369378: step: 492/529, loss: 0.00926132220774889 2023-01-21 10:48:44.487119: step: 496/529, loss: 0.0018499374855309725 2023-01-21 10:48:45.628802: step: 500/529, loss: 0.13141241669654846 2023-01-21 10:48:46.772683: step: 504/529, loss: 0.008917713537812233 2023-01-21 10:48:47.933307: step: 508/529, loss: 0.04187174141407013 2023-01-21 10:48:49.071288: step: 512/529, loss: 0.004901218693703413 2023-01-21 10:48:50.197616: step: 516/529, loss: 0.04365139082074165 2023-01-21 10:48:51.344852: step: 520/529, loss: 0.09695301204919815 2023-01-21 10:48:52.480157: step: 524/529, loss: 0.13254258036613464 2023-01-21 10:48:53.622607: step: 528/529, loss: 0.028083277866244316 2023-01-21 10:48:54.770134: step: 532/529, loss: 0.5566078424453735 2023-01-21 10:48:55.905971: step: 536/529, loss: 0.007655620574951172 2023-01-21 10:48:57.042685: step: 540/529, loss: 0.01911487616598606 2023-01-21 10:48:58.183473: step: 544/529, loss: 0.04500117525458336 2023-01-21 10:48:59.315278: step: 548/529, loss: 0.08489380031824112 2023-01-21 10:49:00.459700: step: 552/529, loss: 0.01476287841796875 2023-01-21 10:49:01.634273: step: 556/529, loss: 0.024106120690703392 2023-01-21 10:49:02.756597: step: 560/529, loss: 0.030239392071962357 2023-01-21 10:49:03.892458: step: 564/529, loss: 0.0031176568008959293 2023-01-21 10:49:05.019772: step: 568/529, loss: 0.07796774059534073 2023-01-21 10:49:06.178224: step: 572/529, loss: 0.026243021711707115 2023-01-21 10:49:07.345428: step: 576/529, loss: 0.02004070207476616 2023-01-21 10:49:08.493581: step: 580/529, loss: 0.008266258984804153 2023-01-21 10:49:09.624475: step: 584/529, loss: 0.011605835519731045 2023-01-21 10:49:10.772164: step: 588/529, loss: 0.0015160085167735815 2023-01-21 10:49:11.936491: step: 592/529, loss: 0.001451778458431363 2023-01-21 10:49:13.081012: step: 596/529, loss: 0.04489459842443466 2023-01-21 10:49:14.193314: step: 600/529, loss: 0.010213185101747513 2023-01-21 10:49:15.381329: step: 604/529, loss: 0.05307454988360405 2023-01-21 10:49:16.522921: step: 608/529, loss: 0.013193107210099697 2023-01-21 10:49:17.665162: step: 612/529, loss: 0.0022746205795556307 2023-01-21 10:49:18.825399: step: 616/529, loss: 0.005787753965705633 2023-01-21 10:49:19.954904: step: 620/529, loss: 0.09382744133472443 2023-01-21 10:49:21.061175: step: 624/529, loss: 0.006672095973044634 2023-01-21 10:49:22.199099: step: 628/529, loss: 0.02024865336716175 2023-01-21 10:49:23.317299: step: 632/529, loss: 0.006074261851608753 2023-01-21 10:49:24.497786: step: 636/529, loss: 0.026778221130371094 2023-01-21 10:49:25.657088: step: 640/529, loss: 0.019365500658750534 2023-01-21 10:49:26.746239: step: 644/529, loss: 0.01812162436544895 2023-01-21 10:49:27.911577: step: 648/529, loss: 0.03174591064453125 2023-01-21 10:49:29.049410: step: 652/529, loss: 0.009531021118164062 2023-01-21 10:49:30.169678: step: 656/529, loss: 0.0018622399074956775 2023-01-21 10:49:31.274566: step: 660/529, loss: 0.01842517778277397 2023-01-21 10:49:32.411659: step: 664/529, loss: 0.008487224578857422 2023-01-21 10:49:33.521800: step: 668/529, loss: 0.00127496721688658 2023-01-21 10:49:34.663388: step: 672/529, loss: 0.08787160366773605 2023-01-21 10:49:35.806906: step: 676/529, loss: 0.0006636619800701737 2023-01-21 10:49:36.960363: step: 680/529, loss: 0.03404254838824272 2023-01-21 10:49:38.112165: step: 684/529, loss: 0.3431926965713501 2023-01-21 10:49:39.294309: step: 688/529, loss: 0.0643855556845665 2023-01-21 10:49:40.456282: step: 692/529, loss: 0.01131210383027792 2023-01-21 10:49:41.609013: step: 696/529, loss: 0.010996866039931774 2023-01-21 10:49:42.766530: step: 700/529, loss: 0.021298695355653763 2023-01-21 10:49:43.914489: step: 704/529, loss: 0.040885165333747864 2023-01-21 10:49:45.045563: step: 708/529, loss: 0.06747809052467346 2023-01-21 10:49:46.179580: step: 712/529, loss: 0.0776216983795166 2023-01-21 10:49:47.314960: step: 716/529, loss: 0.018359661102294922 2023-01-21 10:49:48.463991: step: 720/529, loss: 0.07620649039745331 2023-01-21 10:49:49.602666: step: 724/529, loss: 0.05436286702752113 2023-01-21 10:49:50.762660: step: 728/529, loss: 0.005736351013183594 2023-01-21 10:49:51.886070: step: 732/529, loss: 0.02626323699951172 2023-01-21 10:49:53.045769: step: 736/529, loss: 0.030824948102235794 2023-01-21 10:49:54.196251: step: 740/529, loss: 0.05275621637701988 2023-01-21 10:49:55.334274: step: 744/529, loss: 0.0042005060240626335 2023-01-21 10:49:56.488432: step: 748/529, loss: 0.028584908694028854 2023-01-21 10:49:57.605055: step: 752/529, loss: 0.019593620672822 2023-01-21 10:49:58.754819: step: 756/529, loss: 0.11598511040210724 2023-01-21 10:49:59.903017: step: 760/529, loss: 0.019463349133729935 2023-01-21 10:50:01.045114: step: 764/529, loss: 0.002620697021484375 2023-01-21 10:50:02.182586: step: 768/529, loss: 0.017252827063202858 2023-01-21 10:50:03.323303: step: 772/529, loss: 0.051537707448005676 2023-01-21 10:50:04.450913: step: 776/529, loss: 0.4955669641494751 2023-01-21 10:50:05.592804: step: 780/529, loss: 0.04112367704510689 2023-01-21 10:50:06.780904: step: 784/529, loss: 0.04253892973065376 2023-01-21 10:50:07.942362: step: 788/529, loss: 0.18894128501415253 2023-01-21 10:50:09.078149: step: 792/529, loss: 0.007572841830551624 2023-01-21 10:50:10.213873: step: 796/529, loss: 0.01739196851849556 2023-01-21 10:50:11.375619: step: 800/529, loss: 0.11712293326854706 2023-01-21 10:50:12.512652: step: 804/529, loss: 0.028022432699799538 2023-01-21 10:50:13.634131: step: 808/529, loss: 0.010631752200424671 2023-01-21 10:50:14.759176: step: 812/529, loss: 0.01587390899658203 2023-01-21 10:50:15.943743: step: 816/529, loss: 0.025449277833104134 2023-01-21 10:50:17.114473: step: 820/529, loss: 0.04334645718336105 2023-01-21 10:50:18.261689: step: 824/529, loss: 0.01745166815817356 2023-01-21 10:50:19.378762: step: 828/529, loss: 0.026250077411532402 2023-01-21 10:50:20.504056: step: 832/529, loss: 0.08116170763969421 2023-01-21 10:50:21.684041: step: 836/529, loss: 0.018582439050078392 2023-01-21 10:50:22.829269: step: 840/529, loss: 0.010861778631806374 2023-01-21 10:50:23.951299: step: 844/529, loss: 0.13930387794971466 2023-01-21 10:50:25.104577: step: 848/529, loss: 0.1266312599182129 2023-01-21 10:50:26.232162: step: 852/529, loss: 0.0019449234241619706 2023-01-21 10:50:27.399339: step: 856/529, loss: 0.01752759888768196 2023-01-21 10:50:28.534602: step: 860/529, loss: 0.004564667120575905 2023-01-21 10:50:29.667622: step: 864/529, loss: 0.033022690564394 2023-01-21 10:50:30.800649: step: 868/529, loss: 0.01507644634693861 2023-01-21 10:50:31.953803: step: 872/529, loss: 0.006407356355339289 2023-01-21 10:50:33.080400: step: 876/529, loss: 0.013100052252411842 2023-01-21 10:50:34.238701: step: 880/529, loss: 0.0013195992214605212 2023-01-21 10:50:35.418293: step: 884/529, loss: 0.02054414711892605 2023-01-21 10:50:36.562167: step: 888/529, loss: 0.000996494316495955 2023-01-21 10:50:37.705976: step: 892/529, loss: 0.015393257141113281 2023-01-21 10:50:38.833929: step: 896/529, loss: 0.010095691308379173 2023-01-21 10:50:39.982558: step: 900/529, loss: 0.061825837939977646 2023-01-21 10:50:41.101202: step: 904/529, loss: 0.02988128736615181 2023-01-21 10:50:42.232819: step: 908/529, loss: 0.010990714654326439 2023-01-21 10:50:43.355942: step: 912/529, loss: 0.05737915262579918 2023-01-21 10:50:44.487420: step: 916/529, loss: 0.07438087463378906 2023-01-21 10:50:45.658576: step: 920/529, loss: 0.08790607750415802 2023-01-21 10:50:46.790931: step: 924/529, loss: 0.01807727850973606 2023-01-21 10:50:47.940720: step: 928/529, loss: 0.07593412697315216 2023-01-21 10:50:49.088600: step: 932/529, loss: 0.016252517700195312 2023-01-21 10:50:50.232472: step: 936/529, loss: 0.015817834064364433 2023-01-21 10:50:51.357282: step: 940/529, loss: 0.021053221076726913 2023-01-21 10:50:52.520706: step: 944/529, loss: 0.3982565999031067 2023-01-21 10:50:53.653906: step: 948/529, loss: 0.208054780960083 2023-01-21 10:50:54.771641: step: 952/529, loss: 0.05603466182947159 2023-01-21 10:50:55.866551: step: 956/529, loss: 0.011123323813080788 2023-01-21 10:50:57.027662: step: 960/529, loss: 0.003956890199333429 2023-01-21 10:50:58.154926: step: 964/529, loss: 0.005530357826501131 2023-01-21 10:50:59.304896: step: 968/529, loss: 0.00014495849609375 2023-01-21 10:51:00.476279: step: 972/529, loss: 0.026345601305365562 2023-01-21 10:51:01.646856: step: 976/529, loss: 0.13561353087425232 2023-01-21 10:51:02.796209: step: 980/529, loss: 0.05152587965130806 2023-01-21 10:51:03.925702: step: 984/529, loss: 0.028082944452762604 2023-01-21 10:51:05.083799: step: 988/529, loss: 0.028377152979373932 2023-01-21 10:51:06.254704: step: 992/529, loss: 0.04860544204711914 2023-01-21 10:51:07.418256: step: 996/529, loss: 0.9132544994354248 2023-01-21 10:51:08.554736: step: 1000/529, loss: 0.021569300442934036 2023-01-21 10:51:09.721370: step: 1004/529, loss: 0.08717389404773712 2023-01-21 10:51:10.889194: step: 1008/529, loss: 0.11046136915683746 2023-01-21 10:51:12.025245: step: 1012/529, loss: 0.04021396487951279 2023-01-21 10:51:13.178800: step: 1016/529, loss: 0.006649589631706476 2023-01-21 10:51:14.351372: step: 1020/529, loss: 0.0011124610900878906 2023-01-21 10:51:15.489119: step: 1024/529, loss: 0.06144142150878906 2023-01-21 10:51:16.668810: step: 1028/529, loss: 0.06999292224645615 2023-01-21 10:51:17.788626: step: 1032/529, loss: 0.023520182818174362 2023-01-21 10:51:18.916305: step: 1036/529, loss: 0.058330535888671875 2023-01-21 10:51:20.045125: step: 1040/529, loss: 0.011556053534150124 2023-01-21 10:51:21.187152: step: 1044/529, loss: 0.004688453860580921 2023-01-21 10:51:22.346394: step: 1048/529, loss: 0.03973636403679848 2023-01-21 10:51:23.502794: step: 1052/529, loss: 1.2408208847045898 2023-01-21 10:51:24.645372: step: 1056/529, loss: 0.0006423950544558465 2023-01-21 10:51:25.766171: step: 1060/529, loss: 0.03664274513721466 2023-01-21 10:51:26.887111: step: 1064/529, loss: 0.022694967687129974 2023-01-21 10:51:28.028519: step: 1068/529, loss: 0.0499635711312294 2023-01-21 10:51:29.192361: step: 1072/529, loss: 0.006103706546127796 2023-01-21 10:51:30.317313: step: 1076/529, loss: 0.005550098605453968 2023-01-21 10:51:31.442301: step: 1080/529, loss: 0.04679050296545029 2023-01-21 10:51:32.616113: step: 1084/529, loss: 0.01918802224099636 2023-01-21 10:51:33.753860: step: 1088/529, loss: 0.016031265258789062 2023-01-21 10:51:34.921840: step: 1092/529, loss: 0.20714177191257477 2023-01-21 10:51:36.054953: step: 1096/529, loss: 0.04801655188202858 2023-01-21 10:51:37.188185: step: 1100/529, loss: 0.0032142880372703075 2023-01-21 10:51:38.314475: step: 1104/529, loss: 0.009187364019453526 2023-01-21 10:51:39.501451: step: 1108/529, loss: 0.09246368706226349 2023-01-21 10:51:40.647749: step: 1112/529, loss: 0.02632179483771324 2023-01-21 10:51:41.786824: step: 1116/529, loss: 0.04905581474304199 2023-01-21 10:51:42.944442: step: 1120/529, loss: 0.013914108276367188 2023-01-21 10:51:44.090766: step: 1124/529, loss: 6.427765038097277e-05 2023-01-21 10:51:45.250749: step: 1128/529, loss: 0.0027844428550451994 2023-01-21 10:51:46.401261: step: 1132/529, loss: 0.11400279402732849 2023-01-21 10:51:47.591144: step: 1136/529, loss: 0.0089332340285182 2023-01-21 10:51:48.722804: step: 1140/529, loss: 0.04021177813410759 2023-01-21 10:51:49.850227: step: 1144/529, loss: 0.05507812649011612 2023-01-21 10:51:50.981590: step: 1148/529, loss: 0.0202270969748497 2023-01-21 10:51:52.123435: step: 1152/529, loss: 0.03666200488805771 2023-01-21 10:51:53.242178: step: 1156/529, loss: 0.001990842865779996 2023-01-21 10:51:54.375001: step: 1160/529, loss: 0.04222536459565163 2023-01-21 10:51:55.529775: step: 1164/529, loss: 0.06465263664722443 2023-01-21 10:51:56.691704: step: 1168/529, loss: 0.02528991736471653 2023-01-21 10:51:57.838970: step: 1172/529, loss: 0.10889530181884766 2023-01-21 10:51:59.012602: step: 1176/529, loss: 0.12215165793895721 2023-01-21 10:52:00.153548: step: 1180/529, loss: 0.007244729902595282 2023-01-21 10:52:01.290104: step: 1184/529, loss: 0.11880825459957123 2023-01-21 10:52:02.418162: step: 1188/529, loss: 0.09156408905982971 2023-01-21 10:52:03.561827: step: 1192/529, loss: 0.04717455059289932 2023-01-21 10:52:04.697272: step: 1196/529, loss: 0.04629569128155708 2023-01-21 10:52:05.843518: step: 1200/529, loss: 0.047486212104558945 2023-01-21 10:52:07.022692: step: 1204/529, loss: 0.23759636282920837 2023-01-21 10:52:08.172890: step: 1208/529, loss: 0.03310270234942436 2023-01-21 10:52:09.323453: step: 1212/529, loss: 0.4257963299751282 2023-01-21 10:52:10.426266: step: 1216/529, loss: 0.08411683887243271 2023-01-21 10:52:11.574421: step: 1220/529, loss: 0.06458492577075958 2023-01-21 10:52:12.714323: step: 1224/529, loss: 0.01833038404583931 2023-01-21 10:52:13.850438: step: 1228/529, loss: 0.0045150755904614925 2023-01-21 10:52:14.987837: step: 1232/529, loss: 0.1313977837562561 2023-01-21 10:52:16.150341: step: 1236/529, loss: 0.06871018558740616 2023-01-21 10:52:17.367410: step: 1240/529, loss: 0.014271451160311699 2023-01-21 10:52:18.495318: step: 1244/529, loss: 0.04798565059900284 2023-01-21 10:52:19.640704: step: 1248/529, loss: 0.005623340606689453 2023-01-21 10:52:20.797071: step: 1252/529, loss: 0.04986229166388512 2023-01-21 10:52:21.927314: step: 1256/529, loss: 0.005147885996848345 2023-01-21 10:52:23.033700: step: 1260/529, loss: 0.005708837881684303 2023-01-21 10:52:24.195057: step: 1264/529, loss: 0.013275910168886185 2023-01-21 10:52:25.323440: step: 1268/529, loss: 0.013947105966508389 2023-01-21 10:52:26.453753: step: 1272/529, loss: 0.008233213797211647 2023-01-21 10:52:27.582206: step: 1276/529, loss: 0.010438251309096813 2023-01-21 10:52:28.711482: step: 1280/529, loss: 0.006468772888183594 2023-01-21 10:52:29.864385: step: 1284/529, loss: 0.04680328071117401 2023-01-21 10:52:30.980840: step: 1288/529, loss: 0.0008646011119708419 2023-01-21 10:52:32.100958: step: 1292/529, loss: 0.0692012831568718 2023-01-21 10:52:33.237392: step: 1296/529, loss: 0.02568512037396431 2023-01-21 10:52:34.375453: step: 1300/529, loss: 0.025310229510068893 2023-01-21 10:52:35.515846: step: 1304/529, loss: 0.05374126881361008 2023-01-21 10:52:36.649863: step: 1308/529, loss: 0.007646418176591396 2023-01-21 10:52:37.816833: step: 1312/529, loss: 0.5259893536567688 2023-01-21 10:52:38.951099: step: 1316/529, loss: 0.007978057488799095 2023-01-21 10:52:40.100011: step: 1320/529, loss: 0.09202414005994797 2023-01-21 10:52:41.260162: step: 1324/529, loss: 0.0008187771309167147 2023-01-21 10:52:42.390939: step: 1328/529, loss: 0.010210538282990456 2023-01-21 10:52:43.539799: step: 1332/529, loss: 0.009675407782196999 2023-01-21 10:52:44.664365: step: 1336/529, loss: 0.4917963147163391 2023-01-21 10:52:45.801604: step: 1340/529, loss: 0.02311873435974121 2023-01-21 10:52:46.939006: step: 1344/529, loss: 0.020246410742402077 2023-01-21 10:52:48.076627: step: 1348/529, loss: 0.008223408833146095 2023-01-21 10:52:49.221605: step: 1352/529, loss: 0.024042226374149323 2023-01-21 10:52:50.369995: step: 1356/529, loss: 0.003786277724429965 2023-01-21 10:52:51.481327: step: 1360/529, loss: 0.01281118392944336 2023-01-21 10:52:52.634688: step: 1364/529, loss: 0.10583677142858505 2023-01-21 10:52:53.792562: step: 1368/529, loss: 0.005557632073760033 2023-01-21 10:52:54.941344: step: 1372/529, loss: 0.24628451466560364 2023-01-21 10:52:56.085746: step: 1376/529, loss: 0.029894256964325905 2023-01-21 10:52:57.252946: step: 1380/529, loss: 0.002945518586784601 2023-01-21 10:52:58.391559: step: 1384/529, loss: 0.03356170654296875 2023-01-21 10:52:59.545233: step: 1388/529, loss: 0.1352996826171875 2023-01-21 10:53:00.723378: step: 1392/529, loss: 0.10184673964977264 2023-01-21 10:53:01.850860: step: 1396/529, loss: 0.5471808910369873 2023-01-21 10:53:03.015826: step: 1400/529, loss: 0.0005770206335000694 2023-01-21 10:53:04.186799: step: 1404/529, loss: 0.00571861257776618 2023-01-21 10:53:05.334982: step: 1408/529, loss: 0.02298908121883869 2023-01-21 10:53:06.481037: step: 1412/529, loss: 0.02575836144387722 2023-01-21 10:53:07.617800: step: 1416/529, loss: 0.002178239868953824 2023-01-21 10:53:08.758885: step: 1420/529, loss: 0.1595216691493988 2023-01-21 10:53:09.859635: step: 1424/529, loss: 0.008147621527314186 2023-01-21 10:53:10.984331: step: 1428/529, loss: 0.025105096399784088 2023-01-21 10:53:12.128490: step: 1432/529, loss: 0.8444983959197998 2023-01-21 10:53:13.265115: step: 1436/529, loss: 0.01779031753540039 2023-01-21 10:53:14.451243: step: 1440/529, loss: 0.031321004033088684 2023-01-21 10:53:15.616985: step: 1444/529, loss: 0.08353462815284729 2023-01-21 10:53:16.747559: step: 1448/529, loss: 0.00377235421910882 2023-01-21 10:53:17.861926: step: 1452/529, loss: 0.012806129641830921 2023-01-21 10:53:18.994381: step: 1456/529, loss: 0.011811209842562675 2023-01-21 10:53:20.131176: step: 1460/529, loss: 0.07190532982349396 2023-01-21 10:53:21.260352: step: 1464/529, loss: 0.02921276167035103 2023-01-21 10:53:22.383381: step: 1468/529, loss: 0.02391042746603489 2023-01-21 10:53:23.514231: step: 1472/529, loss: 0.0103295324370265 2023-01-21 10:53:24.676274: step: 1476/529, loss: 0.10181155055761337 2023-01-21 10:53:25.819367: step: 1480/529, loss: 0.019657611846923828 2023-01-21 10:53:26.950165: step: 1484/529, loss: 0.0474882610142231 2023-01-21 10:53:28.068005: step: 1488/529, loss: 0.0377744697034359 2023-01-21 10:53:29.203748: step: 1492/529, loss: 0.01982865482568741 2023-01-21 10:53:30.348582: step: 1496/529, loss: 0.05767975002527237 2023-01-21 10:53:31.480995: step: 1500/529, loss: 0.6384870409965515 2023-01-21 10:53:32.613682: step: 1504/529, loss: 0.008641624823212624 2023-01-21 10:53:33.728943: step: 1508/529, loss: 0.05865902826189995 2023-01-21 10:53:34.896541: step: 1512/529, loss: 0.01728830300271511 2023-01-21 10:53:36.050101: step: 1516/529, loss: 0.030413534492254257 2023-01-21 10:53:37.162691: step: 1520/529, loss: 0.005578422918915749 2023-01-21 10:53:38.314624: step: 1524/529, loss: 0.0035753785632550716 2023-01-21 10:53:39.501862: step: 1528/529, loss: 0.27121925354003906 2023-01-21 10:53:40.640656: step: 1532/529, loss: 0.07558765262365341 2023-01-21 10:53:41.750702: step: 1536/529, loss: 0.018369674682617188 2023-01-21 10:53:42.909649: step: 1540/529, loss: 0.01779937744140625 2023-01-21 10:53:44.024545: step: 1544/529, loss: 0.11485147476196289 2023-01-21 10:53:45.183544: step: 1548/529, loss: 0.009597969241440296 2023-01-21 10:53:46.323812: step: 1552/529, loss: 0.04382495954632759 2023-01-21 10:53:47.468538: step: 1556/529, loss: 0.01810441166162491 2023-01-21 10:53:48.642136: step: 1560/529, loss: 1.122008204460144 2023-01-21 10:53:49.769035: step: 1564/529, loss: 0.0633401870727539 2023-01-21 10:53:50.913963: step: 1568/529, loss: 0.022121049463748932 2023-01-21 10:53:52.043342: step: 1572/529, loss: 0.02047419361770153 2023-01-21 10:53:53.180703: step: 1576/529, loss: 0.013487149029970169 2023-01-21 10:53:54.320361: step: 1580/529, loss: 0.04717302322387695 2023-01-21 10:53:55.475190: step: 1584/529, loss: 0.05500049889087677 2023-01-21 10:53:56.621791: step: 1588/529, loss: 0.013088608160614967 2023-01-21 10:53:57.736784: step: 1592/529, loss: 0.05883502960205078 2023-01-21 10:53:58.893378: step: 1596/529, loss: 0.11016788333654404 2023-01-21 10:54:00.014595: step: 1600/529, loss: 0.10014276206493378 2023-01-21 10:54:01.165686: step: 1604/529, loss: 0.001014757202938199 2023-01-21 10:54:02.308511: step: 1608/529, loss: 0.018223382532596588 2023-01-21 10:54:03.453108: step: 1612/529, loss: 0.010668945498764515 2023-01-21 10:54:04.584985: step: 1616/529, loss: 0.08153381198644638 2023-01-21 10:54:05.715989: step: 1620/529, loss: 0.008607864379882812 2023-01-21 10:54:06.849152: step: 1624/529, loss: 0.05545301362872124 2023-01-21 10:54:07.994507: step: 1628/529, loss: 0.05251751095056534 2023-01-21 10:54:09.104395: step: 1632/529, loss: 0.0032676695846021175 2023-01-21 10:54:10.226055: step: 1636/529, loss: 0.012538433074951172 2023-01-21 10:54:11.366790: step: 1640/529, loss: 0.0659894049167633 2023-01-21 10:54:12.520553: step: 1644/529, loss: 0.026009656488895416 2023-01-21 10:54:13.642979: step: 1648/529, loss: 0.022279931232333183 2023-01-21 10:54:14.787656: step: 1652/529, loss: 0.062365058809518814 2023-01-21 10:54:15.925505: step: 1656/529, loss: 0.049483396112918854 2023-01-21 10:54:17.069871: step: 1660/529, loss: 0.026654861867427826 2023-01-21 10:54:18.223755: step: 1664/529, loss: 0.01733558252453804 2023-01-21 10:54:19.385137: step: 1668/529, loss: 0.06093359366059303 2023-01-21 10:54:20.518940: step: 1672/529, loss: 0.02612752839922905 2023-01-21 10:54:21.659830: step: 1676/529, loss: 0.12475338578224182 2023-01-21 10:54:22.801121: step: 1680/529, loss: 0.047333717346191406 2023-01-21 10:54:23.932993: step: 1684/529, loss: 0.003857517149299383 2023-01-21 10:54:25.083012: step: 1688/529, loss: 0.00394363421946764 2023-01-21 10:54:26.226341: step: 1692/529, loss: 0.03974476084113121 2023-01-21 10:54:27.343422: step: 1696/529, loss: 0.03626508638262749 2023-01-21 10:54:28.494007: step: 1700/529, loss: 0.007637977600097656 2023-01-21 10:54:29.627124: step: 1704/529, loss: 0.01441192626953125 2023-01-21 10:54:30.761824: step: 1708/529, loss: 0.2278156280517578 2023-01-21 10:54:31.865945: step: 1712/529, loss: 0.005734110251069069 2023-01-21 10:54:33.024803: step: 1716/529, loss: 0.08266372978687286 2023-01-21 10:54:34.173648: step: 1720/529, loss: 0.032337188720703125 2023-01-21 10:54:35.307938: step: 1724/529, loss: 0.08831825852394104 2023-01-21 10:54:36.434555: step: 1728/529, loss: 0.0772031769156456 2023-01-21 10:54:37.559627: step: 1732/529, loss: 0.021537208929657936 2023-01-21 10:54:38.698294: step: 1736/529, loss: 0.08185596764087677 2023-01-21 10:54:39.820464: step: 1740/529, loss: 0.0002483367861714214 2023-01-21 10:54:40.952597: step: 1744/529, loss: 0.04616403579711914 2023-01-21 10:54:42.106541: step: 1748/529, loss: 0.0072168828919529915 2023-01-21 10:54:43.249672: step: 1752/529, loss: 0.10313248634338379 2023-01-21 10:54:44.381220: step: 1756/529, loss: 0.041164495050907135 2023-01-21 10:54:45.501165: step: 1760/529, loss: 0.05121174082159996 2023-01-21 10:54:46.647481: step: 1764/529, loss: 0.010016155429184437 2023-01-21 10:54:47.776084: step: 1768/529, loss: 0.003292560577392578 2023-01-21 10:54:48.902112: step: 1772/529, loss: 0.04646673426032066 2023-01-21 10:54:50.031259: step: 1776/529, loss: 0.005523967556655407 2023-01-21 10:54:51.160603: step: 1780/529, loss: 0.15123902261257172 2023-01-21 10:54:52.292332: step: 1784/529, loss: 0.04751448705792427 2023-01-21 10:54:53.465403: step: 1788/529, loss: 0.0642232894897461 2023-01-21 10:54:54.635986: step: 1792/529, loss: 0.028957748785614967 2023-01-21 10:54:55.767884: step: 1796/529, loss: 0.034417059272527695 2023-01-21 10:54:56.906117: step: 1800/529, loss: 0.06205263361334801 2023-01-21 10:54:58.046889: step: 1804/529, loss: 0.006472969427704811 2023-01-21 10:54:59.164424: step: 1808/529, loss: 0.09211353957653046 2023-01-21 10:55:00.292472: step: 1812/529, loss: 0.007915687747299671 2023-01-21 10:55:01.439245: step: 1816/529, loss: 0.06450091302394867 2023-01-21 10:55:02.607635: step: 1820/529, loss: 0.02703409269452095 2023-01-21 10:55:03.751649: step: 1824/529, loss: 0.00462155370041728 2023-01-21 10:55:04.931073: step: 1828/529, loss: 0.5667726397514343 2023-01-21 10:55:06.082721: step: 1832/529, loss: 0.0026515007484704256 2023-01-21 10:55:07.191003: step: 1836/529, loss: 0.0011473179329186678 2023-01-21 10:55:08.303564: step: 1840/529, loss: 0.09822896122932434 2023-01-21 10:55:09.465737: step: 1844/529, loss: 0.0019580840598791838 2023-01-21 10:55:10.616647: step: 1848/529, loss: 0.12257274985313416 2023-01-21 10:55:11.745825: step: 1852/529, loss: 0.014573573134839535 2023-01-21 10:55:12.891453: step: 1856/529, loss: 0.022257328033447266 2023-01-21 10:55:14.026848: step: 1860/529, loss: 0.011684035882353783 2023-01-21 10:55:15.179136: step: 1864/529, loss: 0.03660747408866882 2023-01-21 10:55:16.308724: step: 1868/529, loss: 0.04032135009765625 2023-01-21 10:55:17.441396: step: 1872/529, loss: 0.15340514481067657 2023-01-21 10:55:18.592151: step: 1876/529, loss: 0.017031574621796608 2023-01-21 10:55:19.736405: step: 1880/529, loss: 0.02414569817483425 2023-01-21 10:55:20.868719: step: 1884/529, loss: 0.21909034252166748 2023-01-21 10:55:21.993441: step: 1888/529, loss: 0.052767373621463776 2023-01-21 10:55:23.121659: step: 1892/529, loss: 0.026542089879512787 2023-01-21 10:55:24.293101: step: 1896/529, loss: 0.10610581189393997 2023-01-21 10:55:25.420969: step: 1900/529, loss: 0.01424565352499485 2023-01-21 10:55:26.569723: step: 1904/529, loss: 0.44736775755882263 2023-01-21 10:55:27.730969: step: 1908/529, loss: 0.01866903342306614 2023-01-21 10:55:28.851224: step: 1912/529, loss: 0.06140518561005592 2023-01-21 10:55:29.972358: step: 1916/529, loss: 0.03488168865442276 2023-01-21 10:55:31.129511: step: 1920/529, loss: 0.044077496975660324 2023-01-21 10:55:32.272647: step: 1924/529, loss: 0.045752812176942825 2023-01-21 10:55:33.377110: step: 1928/529, loss: 0.02935056760907173 2023-01-21 10:55:34.522116: step: 1932/529, loss: 0.009860992431640625 2023-01-21 10:55:35.691238: step: 1936/529, loss: 0.1397884339094162 2023-01-21 10:55:36.829701: step: 1940/529, loss: 0.031992532312870026 2023-01-21 10:55:38.009214: step: 1944/529, loss: 0.08644180744886398 2023-01-21 10:55:39.162848: step: 1948/529, loss: 0.04606590420007706 2023-01-21 10:55:40.305324: step: 1952/529, loss: 0.005463266745209694 2023-01-21 10:55:41.405416: step: 1956/529, loss: 0.0010972976451739669 2023-01-21 10:55:42.547802: step: 1960/529, loss: 0.005159950349479914 2023-01-21 10:55:43.691468: step: 1964/529, loss: 0.03960418701171875 2023-01-21 10:55:44.830743: step: 1968/529, loss: 0.017051314935088158 2023-01-21 10:55:45.989193: step: 1972/529, loss: 0.07580963522195816 2023-01-21 10:55:47.111591: step: 1976/529, loss: 0.028292130678892136 2023-01-21 10:55:48.256596: step: 1980/529, loss: 0.004938411992043257 2023-01-21 10:55:49.411460: step: 1984/529, loss: 0.02169984206557274 2023-01-21 10:55:50.555267: step: 1988/529, loss: 0.004240894224494696 2023-01-21 10:55:51.678629: step: 1992/529, loss: 0.06446247547864914 2023-01-21 10:55:52.814156: step: 1996/529, loss: 0.005039024632424116 2023-01-21 10:55:53.962023: step: 2000/529, loss: 0.03675718232989311 2023-01-21 10:55:55.106322: step: 2004/529, loss: 0.1637801229953766 2023-01-21 10:55:56.264265: step: 2008/529, loss: 0.016495706513524055 2023-01-21 10:55:57.390790: step: 2012/529, loss: 0.04091978073120117 2023-01-21 10:55:58.523893: step: 2016/529, loss: 0.08089341968297958 2023-01-21 10:55:59.689491: step: 2020/529, loss: 0.026083802804350853 2023-01-21 10:56:00.806321: step: 2024/529, loss: 0.0022251398768275976 2023-01-21 10:56:01.941166: step: 2028/529, loss: 0.08772774040699005 2023-01-21 10:56:03.067124: step: 2032/529, loss: 0.026732444763183594 2023-01-21 10:56:04.195545: step: 2036/529, loss: 0.016544677317142487 2023-01-21 10:56:05.337072: step: 2040/529, loss: 0.0016375065315514803 2023-01-21 10:56:06.505214: step: 2044/529, loss: 0.011168288998305798 2023-01-21 10:56:07.630819: step: 2048/529, loss: 0.0015014648670330644 2023-01-21 10:56:08.782925: step: 2052/529, loss: 0.23055382072925568 2023-01-21 10:56:09.914137: step: 2056/529, loss: 0.017482947558164597 2023-01-21 10:56:11.047838: step: 2060/529, loss: 0.00787048414349556 2023-01-21 10:56:12.196141: step: 2064/529, loss: 0.08394412696361542 2023-01-21 10:56:13.341600: step: 2068/529, loss: 1.5620718002319336 2023-01-21 10:56:14.445364: step: 2072/529, loss: 0.01348123513162136 2023-01-21 10:56:15.576893: step: 2076/529, loss: 0.020075369626283646 2023-01-21 10:56:16.685523: step: 2080/529, loss: 0.06412658840417862 2023-01-21 10:56:17.881993: step: 2084/529, loss: 0.0004307747003622353 2023-01-21 10:56:19.043299: step: 2088/529, loss: 0.015450209379196167 2023-01-21 10:56:20.148453: step: 2092/529, loss: 0.016770077869296074 2023-01-21 10:56:21.256381: step: 2096/529, loss: 0.1123904213309288 2023-01-21 10:56:22.380006: step: 2100/529, loss: 0.09366931766271591 2023-01-21 10:56:23.521665: step: 2104/529, loss: 0.04261951521039009 2023-01-21 10:56:24.649014: step: 2108/529, loss: 0.028529644012451172 2023-01-21 10:56:25.788162: step: 2112/529, loss: 0.5548986196517944 2023-01-21 10:56:26.938531: step: 2116/529, loss: 0.0037319185212254524 ================================================== Loss: 0.068 -------------------- Dev: {'event': {'p': 0.5853174603174603, 'r': 0.7856191744340879, 'f1': 0.670835702103468}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Test: {'event': {'p': 0.6082845565060784, 'r': 0.7827346465816918, 'f1': 0.6845705599189258}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Chinese: {'event': {'p': 0.5268817204301075, 'r': 0.9074074074074074, 'f1': 0.6666666666666666}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Korean: {'event': {'p': 0.5555555555555556, 'r': 0.5555555555555556, 'f1': 0.5555555555555556}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} 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': 13} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** Epoch: 14 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:57:08.638072: step: 4/529, loss: 0.013891411013901234 2023-01-21 10:57:09.780783: step: 8/529, loss: 0.041683390736579895 2023-01-21 10:57:10.917130: step: 12/529, loss: 0.011958027258515358 2023-01-21 10:57:12.044785: step: 16/529, loss: 0.00681457482278347 2023-01-21 10:57:13.199045: step: 20/529, loss: 0.13926096260547638 2023-01-21 10:57:14.334657: step: 24/529, loss: 0.02781384252011776 2023-01-21 10:57:15.472708: step: 28/529, loss: 0.04537608101963997 2023-01-21 10:57:16.593931: step: 32/529, loss: 0.013766097836196423 2023-01-21 10:57:17.711766: step: 36/529, loss: 0.020801449194550514 2023-01-21 10:57:18.871671: step: 40/529, loss: 0.0009093284606933594 2023-01-21 10:57:20.023299: step: 44/529, loss: 0.014041900634765625 2023-01-21 10:57:21.163004: step: 48/529, loss: 0.006550979800522327 2023-01-21 10:57:22.293091: step: 52/529, loss: 0.02831106260418892 2023-01-21 10:57:23.419822: step: 56/529, loss: 0.6888611912727356 2023-01-21 10:57:24.560322: step: 60/529, loss: 0.009065628051757812 2023-01-21 10:57:25.735543: step: 64/529, loss: 0.0005560875288210809 2023-01-21 10:57:26.852062: step: 68/529, loss: 0.054268453270196915 2023-01-21 10:57:27.998233: step: 72/529, loss: 0.03435225412249565 2023-01-21 10:57:29.127367: step: 76/529, loss: 0.009554672054946423 2023-01-21 10:57:30.309351: step: 80/529, loss: 0.022829245775938034 2023-01-21 10:57:31.455348: step: 84/529, loss: 0.04375457763671875 2023-01-21 10:57:32.608428: step: 88/529, loss: 0.05114898830652237 2023-01-21 10:57:33.739850: step: 92/529, loss: 6.201476097106934 2023-01-21 10:57:34.892103: step: 96/529, loss: 0.008097457699477673 2023-01-21 10:57:36.019427: step: 100/529, loss: 0.03678007051348686 2023-01-21 10:57:37.146914: step: 104/529, loss: 0.023497773334383965 2023-01-21 10:57:38.310300: step: 108/529, loss: 0.01891956478357315 2023-01-21 10:57:39.459139: step: 112/529, loss: 0.05131816864013672 2023-01-21 10:57:40.609673: step: 116/529, loss: 0.05701904371380806 2023-01-21 10:57:41.758520: step: 120/529, loss: 0.0013956070179119706 2023-01-21 10:57:42.899991: step: 124/529, loss: 0.0012221336364746094 2023-01-21 10:57:44.029107: step: 128/529, loss: 0.0005672932020388544 2023-01-21 10:57:45.155100: step: 132/529, loss: 0.012733984738588333 2023-01-21 10:57:46.326369: step: 136/529, loss: 0.020636605098843575 2023-01-21 10:57:47.456902: step: 140/529, loss: 0.024883698672056198 2023-01-21 10:57:48.622897: step: 144/529, loss: 0.009672450833022594 2023-01-21 10:57:49.727117: step: 148/529, loss: 0.014662360772490501 2023-01-21 10:57:50.883328: step: 152/529, loss: 0.03758711740374565 2023-01-21 10:57:52.043529: step: 156/529, loss: 0.1078585609793663 2023-01-21 10:57:53.260109: step: 160/529, loss: 0.036435458809137344 2023-01-21 10:57:54.445351: step: 164/529, loss: 0.058748628944158554 2023-01-21 10:57:55.579377: step: 168/529, loss: 0.011932754889130592 2023-01-21 10:57:56.698342: step: 172/529, loss: 0.005487346556037664 2023-01-21 10:57:57.843789: step: 176/529, loss: 0.060613345354795456 2023-01-21 10:57:58.979969: step: 180/529, loss: 0.024787425994873047 2023-01-21 10:58:00.129991: step: 184/529, loss: 0.12951135635375977 2023-01-21 10:58:01.301833: step: 188/529, loss: 0.0011964798904955387 2023-01-21 10:58:02.443111: step: 192/529, loss: 0.03510608524084091 2023-01-21 10:58:03.604847: step: 196/529, loss: 0.06786441802978516 2023-01-21 10:58:04.739442: step: 200/529, loss: 0.023728180676698685 2023-01-21 10:58:05.860944: step: 204/529, loss: 0.005610323511064053 2023-01-21 10:58:06.984513: step: 208/529, loss: 0.33244168758392334 2023-01-21 10:58:08.129050: step: 212/529, loss: 0.03986950218677521 2023-01-21 10:58:09.233758: step: 216/529, loss: 0.0499635711312294 2023-01-21 10:58:10.391915: step: 220/529, loss: 0.018599702045321465 2023-01-21 10:58:11.547618: step: 224/529, loss: 0.004405784886330366 2023-01-21 10:58:12.678513: step: 228/529, loss: 0.0024839877150952816 2023-01-21 10:58:13.814445: step: 232/529, loss: 0.0013145447010174394 2023-01-21 10:58:14.948872: step: 236/529, loss: 0.005399561021476984 2023-01-21 10:58:16.101384: step: 240/529, loss: 0.005805349443107843 2023-01-21 10:58:17.257879: step: 244/529, loss: 0.0017516136867925525 2023-01-21 10:58:18.412824: step: 248/529, loss: 0.01354761142283678 2023-01-21 10:58:19.553821: step: 252/529, loss: 0.014997268095612526 2023-01-21 10:58:20.670402: step: 256/529, loss: 0.0028181076049804688 2023-01-21 10:58:21.819067: step: 260/529, loss: 0.05532827600836754 2023-01-21 10:58:22.957796: step: 264/529, loss: 0.04148883745074272 2023-01-21 10:58:24.132454: step: 268/529, loss: 0.0029524145647883415 2023-01-21 10:58:25.260615: step: 272/529, loss: 0.01788921281695366 2023-01-21 10:58:26.398970: step: 276/529, loss: 0.017932604998350143 2023-01-21 10:58:27.558720: step: 280/529, loss: 0.013398361392319202 2023-01-21 10:58:28.731844: step: 284/529, loss: 0.003225231310352683 2023-01-21 10:58:29.895035: step: 288/529, loss: 0.012411355972290039 2023-01-21 10:58:31.048852: step: 292/529, loss: 0.04199790954589844 2023-01-21 10:58:32.194031: step: 296/529, loss: 0.008658528327941895 2023-01-21 10:58:33.368795: step: 300/529, loss: 0.010313463397324085 2023-01-21 10:58:34.487279: step: 304/529, loss: 0.002655506134033203 2023-01-21 10:58:35.657615: step: 308/529, loss: 0.027561664581298828 2023-01-21 10:58:36.824950: step: 312/529, loss: 0.0014448165893554688 2023-01-21 10:58:37.959530: step: 316/529, loss: 0.006081891246140003 2023-01-21 10:58:39.098403: step: 320/529, loss: 0.04324827343225479 2023-01-21 10:58:40.241860: step: 324/529, loss: 0.013394307345151901 2023-01-21 10:58:41.370456: step: 328/529, loss: 0.014495849609375 2023-01-21 10:58:42.505655: step: 332/529, loss: 0.012241745367646217 2023-01-21 10:58:43.689449: step: 336/529, loss: 0.005948448553681374 2023-01-21 10:58:44.841896: step: 340/529, loss: 0.015468978323042393 2023-01-21 10:58:46.005435: step: 344/529, loss: 0.011286593042314053 2023-01-21 10:58:47.195130: step: 348/529, loss: 0.020523834973573685 2023-01-21 10:58:48.326700: step: 352/529, loss: 0.04298458248376846 2023-01-21 10:58:49.495370: step: 356/529, loss: 0.04024963825941086 2023-01-21 10:58:50.630811: step: 360/529, loss: 5.0354006816633046e-05 2023-01-21 10:58:51.781704: step: 364/529, loss: 0.01866731606423855 2023-01-21 10:58:52.907946: step: 368/529, loss: 0.007298565004020929 2023-01-21 10:58:54.066147: step: 372/529, loss: 0.0024089813232421875 2023-01-21 10:58:55.213840: step: 376/529, loss: 0.04795151203870773 2023-01-21 10:58:56.348125: step: 380/529, loss: 0.0008177757263183594 2023-01-21 10:58:57.480345: step: 384/529, loss: 0.007581877522170544 2023-01-21 10:58:58.603217: step: 388/529, loss: 0.029965687543153763 2023-01-21 10:58:59.725328: step: 392/529, loss: 0.033069707453250885 2023-01-21 10:59:00.925748: step: 396/529, loss: 0.060196854174137115 2023-01-21 10:59:02.052367: step: 400/529, loss: 0.0030612945556640625 2023-01-21 10:59:03.199522: step: 404/529, loss: 0.007460355758666992 2023-01-21 10:59:04.352349: step: 408/529, loss: 0.001958560897037387 2023-01-21 10:59:05.476830: step: 412/529, loss: 0.025990676134824753 2023-01-21 10:59:06.594842: step: 416/529, loss: 0.004959297366440296 2023-01-21 10:59:07.740210: step: 420/529, loss: 0.0011463165283203125 2023-01-21 10:59:08.867057: step: 424/529, loss: 0.03199481964111328 2023-01-21 10:59:09.983160: step: 428/529, loss: 0.018511200323700905 2023-01-21 10:59:11.129024: step: 432/529, loss: 0.11543647199869156 2023-01-21 10:59:12.264787: step: 436/529, loss: 0.08503293991088867 2023-01-21 10:59:13.403951: step: 440/529, loss: 0.00580596923828125 2023-01-21 10:59:14.540215: step: 444/529, loss: 0.14231853187084198 2023-01-21 10:59:15.656247: step: 448/529, loss: 0.05593481287360191 2023-01-21 10:59:16.809611: step: 452/529, loss: 0.03409595414996147 2023-01-21 10:59:17.957900: step: 456/529, loss: 0.005191040225327015 2023-01-21 10:59:19.082604: step: 460/529, loss: 0.003504848573356867 2023-01-21 10:59:20.204341: step: 464/529, loss: 0.0076157571747899055 2023-01-21 10:59:21.347420: step: 468/529, loss: 0.002480888506397605 2023-01-21 10:59:22.464430: step: 472/529, loss: 0.025597095489501953 2023-01-21 10:59:23.634288: step: 476/529, loss: 0.07940158993005753 2023-01-21 10:59:24.762095: step: 480/529, loss: 0.02297048643231392 2023-01-21 10:59:25.916767: step: 484/529, loss: 0.004005551338195801 2023-01-21 10:59:27.056398: step: 488/529, loss: 0.3205123841762543 2023-01-21 10:59:28.202821: step: 492/529, loss: 0.0022920609917491674 2023-01-21 10:59:29.356577: step: 496/529, loss: 0.07876396179199219 2023-01-21 10:59:30.481903: step: 500/529, loss: 0.000438690185546875 2023-01-21 10:59:31.633543: step: 504/529, loss: 0.0026340484619140625 2023-01-21 10:59:32.790607: step: 508/529, loss: 0.06593503803014755 2023-01-21 10:59:33.930928: step: 512/529, loss: 0.03384266048669815 2023-01-21 10:59:35.059088: step: 516/529, loss: 0.05515794828534126 2023-01-21 10:59:36.210632: step: 520/529, loss: 0.030569840222597122 2023-01-21 10:59:37.339141: step: 524/529, loss: 0.010356427170336246 2023-01-21 10:59:38.475663: step: 528/529, loss: 0.0026908398140221834 2023-01-21 10:59:39.631859: step: 532/529, loss: 0.0008689880487509072 2023-01-21 10:59:40.781880: step: 536/529, loss: 0.044808197766542435 2023-01-21 10:59:41.936023: step: 540/529, loss: 0.046677686274051666 2023-01-21 10:59:43.076895: step: 544/529, loss: 0.06704635918140411 2023-01-21 10:59:44.198979: step: 548/529, loss: 0.0769743025302887 2023-01-21 10:59:45.360541: step: 552/529, loss: 0.034835245460271835 2023-01-21 10:59:46.509456: step: 556/529, loss: 0.17201557755470276 2023-01-21 10:59:47.641288: step: 560/529, loss: 0.004364967346191406 2023-01-21 10:59:48.818346: step: 564/529, loss: 0.0031766891479492188 2023-01-21 10:59:49.934269: step: 568/529, loss: 0.08495932072401047 2023-01-21 10:59:51.099311: step: 572/529, loss: 0.05210399627685547 2023-01-21 10:59:52.222039: step: 576/529, loss: 0.12970809638500214 2023-01-21 10:59:53.366911: step: 580/529, loss: 0.022859763354063034 2023-01-21 10:59:54.517065: step: 584/529, loss: 0.0030261040665209293 2023-01-21 10:59:55.646476: step: 588/529, loss: 0.0009380817646160722 2023-01-21 10:59:56.787577: step: 592/529, loss: 0.22260628640651703 2023-01-21 10:59:57.948500: step: 596/529, loss: 0.02772378921508789 2023-01-21 10:59:59.077216: step: 600/529, loss: 0.0002532959042582661 2023-01-21 11:00:00.211222: step: 604/529, loss: 0.04819688946008682 2023-01-21 11:00:01.347434: step: 608/529, loss: 0.0017631531227380037 2023-01-21 11:00:02.490514: step: 612/529, loss: 0.0020809173583984375 2023-01-21 11:00:03.630109: step: 616/529, loss: 0.0010547637939453125 2023-01-21 11:00:04.764294: step: 620/529, loss: 0.020159149542450905 2023-01-21 11:00:05.921357: step: 624/529, loss: 0.02003917656838894 2023-01-21 11:00:07.028556: step: 628/529, loss: 0.0022153854370117188 2023-01-21 11:00:08.183396: step: 632/529, loss: 0.015769682824611664 2023-01-21 11:00:09.316885: step: 636/529, loss: 0.03011493571102619 2023-01-21 11:00:10.458099: step: 640/529, loss: 0.009197044186294079 2023-01-21 11:00:11.626308: step: 644/529, loss: 0.022007372230291367 2023-01-21 11:00:12.743976: step: 648/529, loss: 0.15374240279197693 2023-01-21 11:00:13.891308: step: 652/529, loss: 0.025133132934570312 2023-01-21 11:00:15.009558: step: 656/529, loss: 0.06175212934613228 2023-01-21 11:00:16.161970: step: 660/529, loss: 0.040116310119628906 2023-01-21 11:00:17.336497: step: 664/529, loss: 0.11048908531665802 2023-01-21 11:00:18.474372: step: 668/529, loss: 0.07229853421449661 2023-01-21 11:00:19.625378: step: 672/529, loss: 0.019942903891205788 2023-01-21 11:00:20.715510: step: 676/529, loss: 0.005787563510239124 2023-01-21 11:00:21.851730: step: 680/529, loss: 0.07249470055103302 2023-01-21 11:00:22.990426: step: 684/529, loss: 0.03927869722247124 2023-01-21 11:00:24.128405: step: 688/529, loss: 0.0736946165561676 2023-01-21 11:00:25.255090: step: 692/529, loss: 0.0028409003280103207 2023-01-21 11:00:26.377601: step: 696/529, loss: 0.0024815560318529606 2023-01-21 11:00:27.530349: step: 700/529, loss: 0.05633602291345596 2023-01-21 11:00:28.657570: step: 704/529, loss: 0.02580423466861248 2023-01-21 11:00:29.819854: step: 708/529, loss: 0.05776557698845863 2023-01-21 11:00:30.947152: step: 712/529, loss: 0.6596037149429321 2023-01-21 11:00:32.073573: step: 716/529, loss: 0.0180421844124794 2023-01-21 11:00:33.209561: step: 720/529, loss: 0.033204078674316406 2023-01-21 11:00:34.392564: step: 724/529, loss: 0.013104247860610485 2023-01-21 11:00:35.526838: step: 728/529, loss: 0.04937624931335449 2023-01-21 11:00:36.670522: step: 732/529, loss: 0.005064773838967085 2023-01-21 11:00:37.792009: step: 736/529, loss: 0.11554157733917236 2023-01-21 11:00:38.924799: step: 740/529, loss: 0.0070552825927734375 2023-01-21 11:00:40.035293: step: 744/529, loss: 0.004991054534912109 2023-01-21 11:00:41.169625: step: 748/529, loss: 0.0004006386152468622 2023-01-21 11:00:42.352627: step: 752/529, loss: 0.003079700516536832 2023-01-21 11:00:43.495317: step: 756/529, loss: 0.006748390384018421 2023-01-21 11:00:44.632656: step: 760/529, loss: 0.007106590084731579 2023-01-21 11:00:45.761582: step: 764/529, loss: 0.009605009108781815 2023-01-21 11:00:46.898735: step: 768/529, loss: 0.012813187204301357 2023-01-21 11:00:48.004327: step: 772/529, loss: 0.008759260177612305 2023-01-21 11:00:49.127212: step: 776/529, loss: 0.011754417791962624 2023-01-21 11:00:50.281590: step: 780/529, loss: 0.13001365959644318 2023-01-21 11:00:51.410010: step: 784/529, loss: 0.03844747692346573 2023-01-21 11:00:52.525793: step: 788/529, loss: 0.02269592322409153 2023-01-21 11:00:53.665216: step: 792/529, loss: 0.032015420496463776 2023-01-21 11:00:54.813078: step: 796/529, loss: 0.009649276733398438 2023-01-21 11:00:55.981206: step: 800/529, loss: 0.0011268615489825606 2023-01-21 11:00:57.132564: step: 804/529, loss: 0.05411524698138237 2023-01-21 11:00:58.291552: step: 808/529, loss: 0.006265067961066961 2023-01-21 11:00:59.447014: step: 812/529, loss: 0.10088777542114258 2023-01-21 11:01:00.581170: step: 816/529, loss: 0.00819706916809082 2023-01-21 11:01:01.716171: step: 820/529, loss: 0.06826849281787872 2023-01-21 11:01:02.847632: step: 824/529, loss: 0.005083942320197821 2023-01-21 11:01:03.977548: step: 828/529, loss: 0.013431024737656116 2023-01-21 11:01:05.085538: step: 832/529, loss: 0.04371538385748863 2023-01-21 11:01:06.199309: step: 836/529, loss: 0.0521882027387619 2023-01-21 11:01:07.333587: step: 840/529, loss: 0.014702796936035156 2023-01-21 11:01:08.488774: step: 844/529, loss: 0.010171031579375267 2023-01-21 11:01:09.622896: step: 848/529, loss: 0.144450381398201 2023-01-21 11:01:10.730921: step: 852/529, loss: 4.5490265620173886e-05 2023-01-21 11:01:11.884581: step: 856/529, loss: 0.029668809846043587 2023-01-21 11:01:13.008634: step: 860/529, loss: 0.008723735809326172 2023-01-21 11:01:14.164620: step: 864/529, loss: 0.0061460500583052635 2023-01-21 11:01:15.314313: step: 868/529, loss: 0.003956222906708717 2023-01-21 11:01:16.467217: step: 872/529, loss: 0.05649156495928764 2023-01-21 11:01:17.595410: step: 876/529, loss: 0.008920669555664062 2023-01-21 11:01:18.707554: step: 880/529, loss: 0.08269424736499786 2023-01-21 11:01:19.826073: step: 884/529, loss: 0.0009930611122399569 2023-01-21 11:01:20.954800: step: 888/529, loss: 0.010860348120331764 2023-01-21 11:01:22.090979: step: 892/529, loss: 0.41957569122314453 2023-01-21 11:01:23.237358: step: 896/529, loss: 0.004673862364143133 2023-01-21 11:01:24.370010: step: 900/529, loss: 0.21123036742210388 2023-01-21 11:01:25.492940: step: 904/529, loss: 0.05917511135339737 2023-01-21 11:01:26.620765: step: 908/529, loss: 0.0036683082580566406 2023-01-21 11:01:27.772556: step: 912/529, loss: 0.00275042955763638 2023-01-21 11:01:28.932666: step: 916/529, loss: 0.005301475524902344 2023-01-21 11:01:30.052903: step: 920/529, loss: 0.01792621612548828 2023-01-21 11:01:31.194531: step: 924/529, loss: 0.0310866367071867 2023-01-21 11:01:32.334071: step: 928/529, loss: 0.011602211743593216 2023-01-21 11:01:33.482663: step: 932/529, loss: 0.06785359233617783 2023-01-21 11:01:34.623163: step: 936/529, loss: 0.27556121349334717 2023-01-21 11:01:35.749436: step: 940/529, loss: 0.04063882678747177 2023-01-21 11:01:36.876950: step: 944/529, loss: 0.007766342256218195 2023-01-21 11:01:38.023614: step: 948/529, loss: 0.012747096829116344 2023-01-21 11:01:39.185479: step: 952/529, loss: 0.4037269055843353 2023-01-21 11:01:40.306471: step: 956/529, loss: 0.06459865719079971 2023-01-21 11:01:41.419312: step: 960/529, loss: 0.03466486930847168 2023-01-21 11:01:42.543362: step: 964/529, loss: 0.017555857077240944 2023-01-21 11:01:43.673280: step: 968/529, loss: 0.036191560328006744 2023-01-21 11:01:44.813697: step: 972/529, loss: 0.017478371039032936 2023-01-21 11:01:45.953469: step: 976/529, loss: 0.0012299061054363847 2023-01-21 11:01:47.098554: step: 980/529, loss: 0.00601587351411581 2023-01-21 11:01:48.217116: step: 984/529, loss: 0.051186561584472656 2023-01-21 11:01:49.318444: step: 988/529, loss: 0.012828731909394264 2023-01-21 11:01:50.466322: step: 992/529, loss: 0.01859283447265625 2023-01-21 11:01:51.589498: step: 996/529, loss: 0.04816108196973801 2023-01-21 11:01:52.732638: step: 1000/529, loss: 0.07851681858301163 2023-01-21 11:01:53.858722: step: 1004/529, loss: 0.022362638264894485 2023-01-21 11:01:54.984645: step: 1008/529, loss: 0.015149594284594059 2023-01-21 11:01:56.116182: step: 1012/529, loss: 0.0010516643524169922 2023-01-21 11:01:57.249349: step: 1016/529, loss: 0.04691944271326065 2023-01-21 11:01:58.392240: step: 1020/529, loss: 0.02227654494345188 2023-01-21 11:01:59.527698: step: 1024/529, loss: 0.02096710354089737 2023-01-21 11:02:00.663708: step: 1028/529, loss: 0.048047445714473724 2023-01-21 11:02:01.782575: step: 1032/529, loss: 0.0027255534660071135 2023-01-21 11:02:02.928152: step: 1036/529, loss: 0.1443762332201004 2023-01-21 11:02:04.060309: step: 1040/529, loss: 0.10750333964824677 2023-01-21 11:02:05.224417: step: 1044/529, loss: 0.0056348321959376335 2023-01-21 11:02:06.362469: step: 1048/529, loss: 0.00713119562715292 2023-01-21 11:02:07.526321: step: 1052/529, loss: 0.009885597974061966 2023-01-21 11:02:08.670866: step: 1056/529, loss: 0.02140035666525364 2023-01-21 11:02:09.794552: step: 1060/529, loss: 0.0005970955244265497 2023-01-21 11:02:10.961604: step: 1064/529, loss: 0.03064143657684326 2023-01-21 11:02:12.111366: step: 1068/529, loss: 0.010861778631806374 2023-01-21 11:02:13.258755: step: 1072/529, loss: 0.01453475933521986 2023-01-21 11:02:14.392516: step: 1076/529, loss: 0.02284078672528267 2023-01-21 11:02:15.518294: step: 1080/529, loss: 0.005233383271843195 2023-01-21 11:02:16.640759: step: 1084/529, loss: 0.05689802020788193 2023-01-21 11:02:17.782356: step: 1088/529, loss: 0.006886959075927734 2023-01-21 11:02:18.952546: step: 1092/529, loss: 0.017435455694794655 2023-01-21 11:02:20.116004: step: 1096/529, loss: 0.06448850780725479 2023-01-21 11:02:21.252901: step: 1100/529, loss: 0.004688596818596125 2023-01-21 11:02:22.359720: step: 1104/529, loss: 0.07476458698511124 2023-01-21 11:02:23.497207: step: 1108/529, loss: 0.0071928976103663445 2023-01-21 11:02:24.632620: step: 1112/529, loss: 0.003978443332016468 2023-01-21 11:02:25.784626: step: 1116/529, loss: 0.03825988993048668 2023-01-21 11:02:26.926476: step: 1120/529, loss: 0.009201622568070889 2023-01-21 11:02:28.052091: step: 1124/529, loss: 0.11412010341882706 2023-01-21 11:02:29.182274: step: 1128/529, loss: 0.04814338684082031 2023-01-21 11:02:30.340096: step: 1132/529, loss: 0.37667930126190186 2023-01-21 11:02:31.466783: step: 1136/529, loss: 0.07248277962207794 2023-01-21 11:02:32.587383: step: 1140/529, loss: 0.005755329038947821 2023-01-21 11:02:33.706858: step: 1144/529, loss: 0.01786527782678604 2023-01-21 11:02:34.816320: step: 1148/529, loss: 0.0021076202392578125 2023-01-21 11:02:35.932801: step: 1152/529, loss: 0.01595510169863701 2023-01-21 11:02:37.104134: step: 1156/529, loss: 0.021775150671601295 2023-01-21 11:02:38.220184: step: 1160/529, loss: 0.008389472961425781 2023-01-21 11:02:39.350936: step: 1164/529, loss: 0.007810688111931086 2023-01-21 11:02:40.512582: step: 1168/529, loss: 0.06326351314783096 2023-01-21 11:02:41.688209: step: 1172/529, loss: 0.013787174597382545 2023-01-21 11:02:42.825977: step: 1176/529, loss: 0.025340938940644264 2023-01-21 11:02:43.969352: step: 1180/529, loss: 0.035582639276981354 2023-01-21 11:02:45.136018: step: 1184/529, loss: 0.02855520322918892 2023-01-21 11:02:46.266510: step: 1188/529, loss: 0.05001568794250488 2023-01-21 11:02:47.412431: step: 1192/529, loss: 0.00042896269587799907 2023-01-21 11:02:48.558956: step: 1196/529, loss: 0.06171979755163193 2023-01-21 11:02:49.675883: step: 1200/529, loss: 0.0286375992000103 2023-01-21 11:02:50.819124: step: 1204/529, loss: 0.003940582275390625 2023-01-21 11:02:51.963721: step: 1208/529, loss: 0.0458616241812706 2023-01-21 11:02:53.091240: step: 1212/529, loss: 0.04186258465051651 2023-01-21 11:02:54.242553: step: 1216/529, loss: 0.02067112922668457 2023-01-21 11:02:55.388234: step: 1220/529, loss: 0.0646023228764534 2023-01-21 11:02:56.513808: step: 1224/529, loss: 0.005819129757583141 2023-01-21 11:02:57.662198: step: 1228/529, loss: 0.010585689917206764 2023-01-21 11:02:58.807114: step: 1232/529, loss: 0.01636333577334881 2023-01-21 11:02:59.923196: step: 1236/529, loss: 0.09576711803674698 2023-01-21 11:03:01.068005: step: 1240/529, loss: 0.004456424620002508 2023-01-21 11:03:02.218334: step: 1244/529, loss: 0.009386253543198109 2023-01-21 11:03:03.331310: step: 1248/529, loss: 0.001676368759945035 2023-01-21 11:03:04.471757: step: 1252/529, loss: 0.0812651664018631 2023-01-21 11:03:05.624138: step: 1256/529, loss: 0.02503356896340847 2023-01-21 11:03:06.762867: step: 1260/529, loss: 0.011914395727217197 2023-01-21 11:03:07.913199: step: 1264/529, loss: 0.04774303734302521 2023-01-21 11:03:09.047674: step: 1268/529, loss: 0.0016178131336346269 2023-01-21 11:03:10.181243: step: 1272/529, loss: 0.1578844040632248 2023-01-21 11:03:11.325510: step: 1276/529, loss: 0.01888093911111355 2023-01-21 11:03:12.458143: step: 1280/529, loss: 0.09535980224609375 2023-01-21 11:03:13.611910: step: 1284/529, loss: 0.0023229599464684725 2023-01-21 11:03:14.760591: step: 1288/529, loss: 0.028337908908724785 2023-01-21 11:03:15.899296: step: 1292/529, loss: 0.002019214676693082 2023-01-21 11:03:17.048536: step: 1296/529, loss: 0.008580779656767845 2023-01-21 11:03:18.200597: step: 1300/529, loss: 0.030537987127900124 2023-01-21 11:03:19.319380: step: 1304/529, loss: 0.030974293127655983 2023-01-21 11:03:20.480505: step: 1308/529, loss: 0.04489422217011452 2023-01-21 11:03:21.634333: step: 1312/529, loss: 0.00021810531325172633 2023-01-21 11:03:22.781464: step: 1316/529, loss: 0.0043315887451171875 2023-01-21 11:03:23.939604: step: 1320/529, loss: 0.04198741912841797 2023-01-21 11:03:25.087966: step: 1324/529, loss: 0.004833888728171587 2023-01-21 11:03:26.205329: step: 1328/529, loss: 0.00950317457318306 2023-01-21 11:03:27.346535: step: 1332/529, loss: 0.07978996634483337 2023-01-21 11:03:28.456496: step: 1336/529, loss: 0.0001537799835205078 2023-01-21 11:03:29.609340: step: 1340/529, loss: 0.03995666652917862 2023-01-21 11:03:30.740057: step: 1344/529, loss: 0.0014316558372229338 2023-01-21 11:03:31.899398: step: 1348/529, loss: 0.07693634182214737 2023-01-21 11:03:33.020840: step: 1352/529, loss: 0.006489467807114124 2023-01-21 11:03:34.144881: step: 1356/529, loss: 0.006716489791870117 2023-01-21 11:03:35.277893: step: 1360/529, loss: 0.010333252139389515 2023-01-21 11:03:36.416246: step: 1364/529, loss: 0.007012748625129461 2023-01-21 11:03:37.563006: step: 1368/529, loss: 0.03144798427820206 2023-01-21 11:03:38.715712: step: 1372/529, loss: 0.021324826404452324 2023-01-21 11:03:39.865027: step: 1376/529, loss: 0.05162010341882706 2023-01-21 11:03:41.004924: step: 1380/529, loss: 0.009546756744384766 2023-01-21 11:03:42.118931: step: 1384/529, loss: 0.005966758821159601 2023-01-21 11:03:43.244694: step: 1388/529, loss: 0.028606224805116653 2023-01-21 11:03:44.387946: step: 1392/529, loss: 0.009893418289721012 2023-01-21 11:03:45.534543: step: 1396/529, loss: 0.003786468645557761 2023-01-21 11:03:46.691419: step: 1400/529, loss: 0.01431961078196764 2023-01-21 11:03:47.794247: step: 1404/529, loss: 0.027814865112304688 2023-01-21 11:03:48.934153: step: 1408/529, loss: 0.056890297681093216 2023-01-21 11:03:50.095020: step: 1412/529, loss: 0.07930316776037216 2023-01-21 11:03:51.227999: step: 1416/529, loss: 0.08392544090747833 2023-01-21 11:03:52.351413: step: 1420/529, loss: 0.030322551727294922 2023-01-21 11:03:53.472902: step: 1424/529, loss: 0.04663574695587158 2023-01-21 11:03:54.634582: step: 1428/529, loss: 0.08130503445863724 2023-01-21 11:03:55.785279: step: 1432/529, loss: 0.008636855520308018 2023-01-21 11:03:56.904912: step: 1436/529, loss: 0.003972244448959827 2023-01-21 11:03:58.016925: step: 1440/529, loss: 0.016019249334931374 2023-01-21 11:03:59.181308: step: 1444/529, loss: 0.0065202717669308186 2023-01-21 11:04:00.333576: step: 1448/529, loss: 0.03464512899518013 2023-01-21 11:04:01.457780: step: 1452/529, loss: 0.020255088806152344 2023-01-21 11:04:02.569732: step: 1456/529, loss: 0.03506002575159073 2023-01-21 11:04:03.707568: step: 1460/529, loss: 0.0009643555385991931 2023-01-21 11:04:04.834604: step: 1464/529, loss: 0.0051961299031972885 2023-01-21 11:04:05.961022: step: 1468/529, loss: 0.035044290125370026 2023-01-21 11:04:07.117015: step: 1472/529, loss: 0.03768463060259819 2023-01-21 11:04:08.242096: step: 1476/529, loss: 0.4639724791049957 2023-01-21 11:04:09.385888: step: 1480/529, loss: 0.04469423368573189 2023-01-21 11:04:10.522732: step: 1484/529, loss: 0.01664752885699272 2023-01-21 11:04:11.644744: step: 1488/529, loss: 0.00614585867151618 2023-01-21 11:04:12.786716: step: 1492/529, loss: 0.02300853654742241 2023-01-21 11:04:13.910569: step: 1496/529, loss: 0.05658302083611488 2023-01-21 11:04:15.111737: step: 1500/529, loss: 0.2596409022808075 2023-01-21 11:04:16.301911: step: 1504/529, loss: 0.04389248043298721 2023-01-21 11:04:17.439368: step: 1508/529, loss: 0.07749929279088974 2023-01-21 11:04:18.548079: step: 1512/529, loss: 0.023305749520659447 2023-01-21 11:04:19.694663: step: 1516/529, loss: 0.023348616436123848 2023-01-21 11:04:20.810465: step: 1520/529, loss: 0.00030350685119628906 2023-01-21 11:04:21.930759: step: 1524/529, loss: 0.0035295486450195312 2023-01-21 11:04:23.088646: step: 1528/529, loss: 0.0034933090209960938 2023-01-21 11:04:24.231162: step: 1532/529, loss: 0.019969414919614792 2023-01-21 11:04:25.356862: step: 1536/529, loss: 0.02688765525817871 2023-01-21 11:04:26.496296: step: 1540/529, loss: 0.04217548668384552 2023-01-21 11:04:27.628808: step: 1544/529, loss: 0.0020812987349927425 2023-01-21 11:04:28.788879: step: 1548/529, loss: 0.08510151505470276 2023-01-21 11:04:29.920014: step: 1552/529, loss: 0.019606687128543854 2023-01-21 11:04:31.050018: step: 1556/529, loss: 0.0450017936527729 2023-01-21 11:04:32.188001: step: 1560/529, loss: 0.01704845391213894 2023-01-21 11:04:33.334853: step: 1564/529, loss: 0.0031768563203513622 2023-01-21 11:04:34.478231: step: 1568/529, loss: 0.01782960817217827 2023-01-21 11:04:35.605732: step: 1572/529, loss: 0.0015869140625 2023-01-21 11:04:36.774444: step: 1576/529, loss: 0.08877535164356232 2023-01-21 11:04:37.918284: step: 1580/529, loss: 0.0011180877918377519 2023-01-21 11:04:39.043247: step: 1584/529, loss: 0.008982563391327858 2023-01-21 11:04:40.214785: step: 1588/529, loss: 0.012991905212402344 2023-01-21 11:04:41.369167: step: 1592/529, loss: 0.07841567695140839 2023-01-21 11:04:42.518236: step: 1596/529, loss: 0.012593078427016735 2023-01-21 11:04:43.648534: step: 1600/529, loss: 0.04396235942840576 2023-01-21 11:04:44.806529: step: 1604/529, loss: 0.007347392849624157 2023-01-21 11:04:45.965536: step: 1608/529, loss: 0.11661539226770401 2023-01-21 11:04:47.114978: step: 1612/529, loss: 0.13773012161254883 2023-01-21 11:04:48.234680: step: 1616/529, loss: 0.08009806275367737 2023-01-21 11:04:49.373186: step: 1620/529, loss: 0.011623095721006393 2023-01-21 11:04:50.499187: step: 1624/529, loss: 0.0416659377515316 2023-01-21 11:04:51.606109: step: 1628/529, loss: 0.04001579433679581 2023-01-21 11:04:52.747526: step: 1632/529, loss: 0.05554688721895218 2023-01-21 11:04:53.856522: step: 1636/529, loss: 0.04513001814484596 2023-01-21 11:04:54.992045: step: 1640/529, loss: 0.01767749898135662 2023-01-21 11:04:56.102308: step: 1644/529, loss: 0.7634038925170898 2023-01-21 11:04:57.226869: step: 1648/529, loss: 0.005371284671127796 2023-01-21 11:04:58.363070: step: 1652/529, loss: 0.004937505815178156 2023-01-21 11:04:59.510723: step: 1656/529, loss: 0.005384254269301891 2023-01-21 11:05:00.644649: step: 1660/529, loss: 0.05162429437041283 2023-01-21 11:05:01.788769: step: 1664/529, loss: 0.00018367767916060984 2023-01-21 11:05:02.923204: step: 1668/529, loss: 0.008079337887465954 2023-01-21 11:05:04.091869: step: 1672/529, loss: 0.11684408783912659 2023-01-21 11:05:05.225012: step: 1676/529, loss: 0.025528907775878906 2023-01-21 11:05:06.375096: step: 1680/529, loss: 0.0022632598411291838 2023-01-21 11:05:07.497092: step: 1684/529, loss: 0.03894843906164169 2023-01-21 11:05:08.644892: step: 1688/529, loss: 0.01730175130069256 2023-01-21 11:05:09.811671: step: 1692/529, loss: 0.019917108118534088 2023-01-21 11:05:10.915671: step: 1696/529, loss: 0.023000001907348633 2023-01-21 11:05:12.080195: step: 1700/529, loss: 0.010028553195297718 2023-01-21 11:05:13.183443: step: 1704/529, loss: 0.1472880244255066 2023-01-21 11:05:14.329899: step: 1708/529, loss: 0.01743769645690918 2023-01-21 11:05:15.459092: step: 1712/529, loss: 0.00885009765625 2023-01-21 11:05:16.610251: step: 1716/529, loss: 0.02431316487491131 2023-01-21 11:05:17.720493: step: 1720/529, loss: 0.09490552544593811 2023-01-21 11:05:18.823794: step: 1724/529, loss: 0.00895319040864706 2023-01-21 11:05:19.960182: step: 1728/529, loss: 0.2750937044620514 2023-01-21 11:05:21.118260: step: 1732/529, loss: 0.03558792918920517 2023-01-21 11:05:22.262975: step: 1736/529, loss: 0.050969600677490234 2023-01-21 11:05:23.379740: step: 1740/529, loss: 0.011721134185791016 2023-01-21 11:05:24.532238: step: 1744/529, loss: 0.023114491254091263 2023-01-21 11:05:25.684556: step: 1748/529, loss: 0.001452541328035295 2023-01-21 11:05:26.794770: step: 1752/529, loss: 0.03354015201330185 2023-01-21 11:05:27.976648: step: 1756/529, loss: 0.08127937465906143 2023-01-21 11:05:29.110810: step: 1760/529, loss: 0.019754886627197266 2023-01-21 11:05:30.250829: step: 1764/529, loss: 0.03828897699713707 2023-01-21 11:05:31.351953: step: 1768/529, loss: 0.010671710595488548 2023-01-21 11:05:32.474083: step: 1772/529, loss: 0.01685638353228569 2023-01-21 11:05:33.598525: step: 1776/529, loss: 0.01693294383585453 2023-01-21 11:05:34.734441: step: 1780/529, loss: 0.00014791489229537547 2023-01-21 11:05:35.864518: step: 1784/529, loss: 0.03846006467938423 2023-01-21 11:05:37.060225: step: 1788/529, loss: 0.6008392572402954 2023-01-21 11:05:38.184854: step: 1792/529, loss: 0.023286307230591774 2023-01-21 11:05:39.321971: step: 1796/529, loss: 0.07066001743078232 2023-01-21 11:05:40.459416: step: 1800/529, loss: 0.016535472124814987 2023-01-21 11:05:41.593296: step: 1804/529, loss: 0.3201374113559723 2023-01-21 11:05:42.764520: step: 1808/529, loss: 0.0007961273076944053 2023-01-21 11:05:43.898542: step: 1812/529, loss: 0.05725574493408203 2023-01-21 11:05:45.034590: step: 1816/529, loss: 0.0387762077152729 2023-01-21 11:05:46.171489: step: 1820/529, loss: 0.0030109407380223274 2023-01-21 11:05:47.319914: step: 1824/529, loss: 0.03799133375287056 2023-01-21 11:05:48.455747: step: 1828/529, loss: 0.01337509136646986 2023-01-21 11:05:49.607507: step: 1832/529, loss: 0.016776561737060547 2023-01-21 11:05:50.731820: step: 1836/529, loss: 0.0409914031624794 2023-01-21 11:05:51.860426: step: 1840/529, loss: 0.000979948090389371 2023-01-21 11:05:52.986990: step: 1844/529, loss: 0.04781847447156906 2023-01-21 11:05:54.143793: step: 1848/529, loss: 0.35043296217918396 2023-01-21 11:05:55.309586: step: 1852/529, loss: 0.013577461242675781 2023-01-21 11:05:56.457122: step: 1856/529, loss: 0.013792228884994984 2023-01-21 11:05:57.588084: step: 1860/529, loss: 0.02039957046508789 2023-01-21 11:05:58.741526: step: 1864/529, loss: 0.022695161402225494 2023-01-21 11:05:59.880059: step: 1868/529, loss: 0.5700778961181641 2023-01-21 11:06:01.016996: step: 1872/529, loss: 0.05107765272259712 2023-01-21 11:06:02.127613: step: 1876/529, loss: 0.03049449995160103 2023-01-21 11:06:03.281463: step: 1880/529, loss: 0.01476974505931139 2023-01-21 11:06:04.394260: step: 1884/529, loss: 0.0025437355507165194 2023-01-21 11:06:05.515096: step: 1888/529, loss: 0.009549331851303577 2023-01-21 11:06:06.671450: step: 1892/529, loss: 0.005546092987060547 2023-01-21 11:06:07.800532: step: 1896/529, loss: 0.004867935553193092 2023-01-21 11:06:08.926390: step: 1900/529, loss: 0.017035389319062233 2023-01-21 11:06:10.053321: step: 1904/529, loss: 0.01043329294770956 2023-01-21 11:06:11.210381: step: 1908/529, loss: 0.0014803409576416016 2023-01-21 11:06:12.383134: step: 1912/529, loss: 0.09273986518383026 2023-01-21 11:06:13.515922: step: 1916/529, loss: 0.015713995322585106 2023-01-21 11:06:14.656548: step: 1920/529, loss: 0.014237403869628906 2023-01-21 11:06:15.815295: step: 1924/529, loss: 0.046131327748298645 2023-01-21 11:06:16.967224: step: 1928/529, loss: 0.05037422105669975 2023-01-21 11:06:18.135726: step: 1932/529, loss: 0.04351005703210831 2023-01-21 11:06:19.295067: step: 1936/529, loss: 0.002090549562126398 2023-01-21 11:06:20.455457: step: 1940/529, loss: 0.10046205669641495 2023-01-21 11:06:21.582026: step: 1944/529, loss: 0.03925180435180664 2023-01-21 11:06:22.756177: step: 1948/529, loss: 0.018270110711455345 2023-01-21 11:06:23.887460: step: 1952/529, loss: 0.002542448230087757 2023-01-21 11:06:25.020806: step: 1956/529, loss: 0.01734628714621067 2023-01-21 11:06:26.202736: step: 1960/529, loss: 0.04117755964398384 2023-01-21 11:06:27.330055: step: 1964/529, loss: 0.024634741246700287 2023-01-21 11:06:28.461005: step: 1968/529, loss: 0.001993560930714011 2023-01-21 11:06:29.630313: step: 1972/529, loss: 0.02664492093026638 2023-01-21 11:06:30.781518: step: 1976/529, loss: 0.21703921258449554 2023-01-21 11:06:31.925084: step: 1980/529, loss: 0.07699108123779297 2023-01-21 11:06:33.060440: step: 1984/529, loss: 0.0547727607190609 2023-01-21 11:06:34.229478: step: 1988/529, loss: 0.07107200473546982 2023-01-21 11:06:35.361946: step: 1992/529, loss: 0.00813512783497572 2023-01-21 11:06:36.517433: step: 1996/529, loss: 0.03665969520807266 2023-01-21 11:06:37.634225: step: 2000/529, loss: 0.013404274359345436 2023-01-21 11:06:38.757198: step: 2004/529, loss: 0.018314838409423828 2023-01-21 11:06:39.916685: step: 2008/529, loss: 0.036168672144412994 2023-01-21 11:06:41.031012: step: 2012/529, loss: 0.04513759911060333 2023-01-21 11:06:42.149086: step: 2016/529, loss: 0.06351499259471893 2023-01-21 11:06:43.296675: step: 2020/529, loss: 0.013169956393539906 2023-01-21 11:06:44.418751: step: 2024/529, loss: 0.0058501241728663445 2023-01-21 11:06:45.558270: step: 2028/529, loss: 0.03847971186041832 2023-01-21 11:06:46.687447: step: 2032/529, loss: 0.08768733590841293 2023-01-21 11:06:47.834791: step: 2036/529, loss: 0.035093024373054504 2023-01-21 11:06:48.943966: step: 2040/529, loss: 0.00420115003362298 2023-01-21 11:06:50.102179: step: 2044/529, loss: 0.0059295655228197575 2023-01-21 11:06:51.233654: step: 2048/529, loss: 0.011991972103714943 2023-01-21 11:06:52.386667: step: 2052/529, loss: 0.03097086027264595 2023-01-21 11:06:53.513586: step: 2056/529, loss: 0.051961518824100494 2023-01-21 11:06:54.666978: step: 2060/529, loss: 0.005309009924530983 2023-01-21 11:06:55.777070: step: 2064/529, loss: 0.04272814095020294 2023-01-21 11:06:56.903852: step: 2068/529, loss: 0.04711475223302841 2023-01-21 11:06:58.034150: step: 2072/529, loss: 0.015694905072450638 2023-01-21 11:06:59.185923: step: 2076/529, loss: 0.012761592864990234 2023-01-21 11:07:00.330635: step: 2080/529, loss: 0.0014646290801465511 2023-01-21 11:07:01.460644: step: 2084/529, loss: 0.050594329833984375 2023-01-21 11:07:02.599851: step: 2088/529, loss: 0.0034246444702148438 2023-01-21 11:07:03.738828: step: 2092/529, loss: 0.5826783776283264 2023-01-21 11:07:04.870654: step: 2096/529, loss: 0.01982097700238228 2023-01-21 11:07:06.034505: step: 2100/529, loss: 0.05595235899090767 2023-01-21 11:07:07.166763: step: 2104/529, loss: 0.049640558660030365 2023-01-21 11:07:08.304103: step: 2108/529, loss: 0.039298154413700104 2023-01-21 11:07:09.431327: step: 2112/529, loss: 0.003610137151554227 2023-01-21 11:07:10.582308: step: 2116/529, loss: 0.039118196815252304 ================================================== Loss: 0.056 -------------------- Dev: {'event': {'p': 0.5752741774675972, 'r': 0.7683089214380826, 'f1': 0.6579247434435577}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Test: {'event': {'p': 0.6139944392956441, 'r': 0.7676709154113557, 'f1': 0.6822863027806385}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Chinese: {'event': {'p': 0.5903614457831325, 'r': 0.9074074074074074, 'f1': 0.7153284671532847}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Korean: {'event': {'p': 0.5757575757575758, 'r': 0.6031746031746031, 'f1': 0.5891472868217055}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Russian: {'event': {'p': 0.4772727272727273, 'r': 0.5833333333333334, 'f1': 0.5250000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 11:07:52.378765: step: 4/529, loss: 0.16996756196022034 2023-01-21 11:07:53.515484: step: 8/529, loss: 0.005076599307358265 2023-01-21 11:07:54.644340: step: 12/529, loss: 0.007590484339743853 2023-01-21 11:07:55.769759: step: 16/529, loss: 0.03187151253223419 2023-01-21 11:07:56.909314: step: 20/529, loss: 0.0015483855968341231 2023-01-21 11:07:58.028972: step: 24/529, loss: 0.0004878044128417969 2023-01-21 11:07:59.133279: step: 28/529, loss: 0.026233481243252754 2023-01-21 11:08:00.278998: step: 32/529, loss: 0.006692695897072554 2023-01-21 11:08:01.410429: step: 36/529, loss: 0.02227935940027237 2023-01-21 11:08:02.567757: step: 40/529, loss: 0.042588043957948685 2023-01-21 11:08:03.714761: step: 44/529, loss: 0.02246375009417534 2023-01-21 11:08:04.842474: step: 48/529, loss: 0.009333801455795765 2023-01-21 11:08:05.963554: step: 52/529, loss: 9.773074998520315e-05 2023-01-21 11:08:07.056812: step: 56/529, loss: 6.961822691664565e-06 2023-01-21 11:08:08.194563: step: 60/529, loss: 0.011077404022216797 2023-01-21 11:08:09.311592: step: 64/529, loss: 0.010853481478989124 2023-01-21 11:08:10.453747: step: 68/529, loss: 0.07768469303846359 2023-01-21 11:08:11.603257: step: 72/529, loss: 0.05700492858886719 2023-01-21 11:08:12.731321: step: 76/529, loss: 0.0006534576532430947 2023-01-21 11:08:13.907627: step: 80/529, loss: 0.01553564053028822 2023-01-21 11:08:15.049309: step: 84/529, loss: 0.02438812330365181 2023-01-21 11:08:16.167095: step: 88/529, loss: 0.00216846470721066 2023-01-21 11:08:17.322369: step: 92/529, loss: 0.0062614441849291325 2023-01-21 11:08:18.459581: step: 96/529, loss: 0.05970897525548935 2023-01-21 11:08:19.598308: step: 100/529, loss: 0.005962848663330078 2023-01-21 11:08:20.764440: step: 104/529, loss: 0.0027582645416259766 2023-01-21 11:08:21.904616: step: 108/529, loss: 0.013190556317567825 2023-01-21 11:08:23.032955: step: 112/529, loss: 0.08990192413330078 2023-01-21 11:08:24.133072: step: 116/529, loss: 0.01794157177209854 2023-01-21 11:08:25.269267: step: 120/529, loss: 0.0024349212180823088 2023-01-21 11:08:26.408320: step: 124/529, loss: 0.01016225852072239 2023-01-21 11:08:27.559664: step: 128/529, loss: 0.007003259379416704 2023-01-21 11:08:28.672171: step: 132/529, loss: 0.010140038095414639 2023-01-21 11:08:29.814225: step: 136/529, loss: 0.05052977055311203 2023-01-21 11:08:30.948328: step: 140/529, loss: 0.6274555325508118 2023-01-21 11:08:32.106030: step: 144/529, loss: 0.00042402741382829845 2023-01-21 11:08:33.225293: step: 148/529, loss: 0.0008934021461755037 2023-01-21 11:08:34.360947: step: 152/529, loss: 0.046714212745428085 2023-01-21 11:08:35.493095: step: 156/529, loss: 0.05329084396362305 2023-01-21 11:08:36.609686: step: 160/529, loss: 0.7044841647148132 2023-01-21 11:08:37.758218: step: 164/529, loss: 0.06576278060674667 2023-01-21 11:08:38.910539: step: 168/529, loss: 0.0871133804321289 2023-01-21 11:08:40.033557: step: 172/529, loss: 0.008204365149140358 2023-01-21 11:08:41.201605: step: 176/529, loss: 0.031652070581912994 2023-01-21 11:08:42.326868: step: 180/529, loss: 0.0943388044834137 2023-01-21 11:08:43.473277: step: 184/529, loss: 0.006318187806755304 2023-01-21 11:08:44.588099: step: 188/529, loss: 0.012644767761230469 2023-01-21 11:08:45.755746: step: 192/529, loss: 0.019297409802675247 2023-01-21 11:08:46.919522: step: 196/529, loss: 0.020193099975585938 2023-01-21 11:08:48.072550: step: 200/529, loss: 0.06245727464556694 2023-01-21 11:08:49.235660: step: 204/529, loss: 0.007662010379135609 2023-01-21 11:08:50.362481: step: 208/529, loss: 0.002058363053947687 2023-01-21 11:08:51.511830: step: 212/529, loss: 0.022742461413145065 2023-01-21 11:08:52.715953: step: 216/529, loss: 0.03687229007482529 2023-01-21 11:08:53.832175: step: 220/529, loss: 0.0328306183218956 2023-01-21 11:08:54.963693: step: 224/529, loss: 0.002059746067970991 2023-01-21 11:08:56.121207: step: 228/529, loss: 0.03150958940386772 2023-01-21 11:08:57.255137: step: 232/529, loss: 0.0026178359985351562 2023-01-21 11:08:58.389795: step: 236/529, loss: 0.04283933714032173 2023-01-21 11:08:59.529403: step: 240/529, loss: 0.08069400489330292 2023-01-21 11:09:00.654511: step: 244/529, loss: 0.05912809073925018 2023-01-21 11:09:01.780268: step: 248/529, loss: 0.008021831512451172 2023-01-21 11:09:02.919816: step: 252/529, loss: 0.03494071960449219 2023-01-21 11:09:04.070677: step: 256/529, loss: 0.06016801297664642 2023-01-21 11:09:05.213551: step: 260/529, loss: 0.0014163017040118575 2023-01-21 11:09:06.380738: step: 264/529, loss: 0.029808808118104935 2023-01-21 11:09:07.515824: step: 268/529, loss: 0.027716446667909622 2023-01-21 11:09:08.656707: step: 272/529, loss: 0.02725806273519993 2023-01-21 11:09:09.804186: step: 276/529, loss: 0.018924428150057793 2023-01-21 11:09:10.930116: step: 280/529, loss: 0.01695718802511692 2023-01-21 11:09:12.092773: step: 284/529, loss: 0.07867870479822159 2023-01-21 11:09:13.229031: step: 288/529, loss: 0.007264202460646629 2023-01-21 11:09:14.369315: step: 292/529, loss: 0.0009866715408861637 2023-01-21 11:09:15.522743: step: 296/529, loss: 0.11060390621423721 2023-01-21 11:09:16.651646: step: 300/529, loss: 0.03944873809814453 2023-01-21 11:09:17.831804: step: 304/529, loss: 0.009007835760712624 2023-01-21 11:09:18.955491: step: 308/529, loss: 0.00017089843458961695 2023-01-21 11:09:20.079628: step: 312/529, loss: 0.02495289035141468 2023-01-21 11:09:21.235360: step: 316/529, loss: 0.008362198248505592 2023-01-21 11:09:22.389427: step: 320/529, loss: 0.04006824642419815 2023-01-21 11:09:23.528464: step: 324/529, loss: 0.2926328480243683 2023-01-21 11:09:24.666430: step: 328/529, loss: 0.0011467934818938375 2023-01-21 11:09:25.808182: step: 332/529, loss: 0.08258338272571564 2023-01-21 11:09:26.928022: step: 336/529, loss: 0.009478378109633923 2023-01-21 11:09:28.067677: step: 340/529, loss: 0.012785816565155983 2023-01-21 11:09:29.242402: step: 344/529, loss: 0.008547211065888405 2023-01-21 11:09:30.368350: step: 348/529, loss: 0.30324259400367737 2023-01-21 11:09:31.496697: step: 352/529, loss: 0.005092620849609375 2023-01-21 11:09:32.653336: step: 356/529, loss: 0.058544449508190155 2023-01-21 11:09:33.771497: step: 360/529, loss: 0.010987663641571999 2023-01-21 11:09:34.905154: step: 364/529, loss: 0.04443359375 2023-01-21 11:09:36.049105: step: 368/529, loss: 0.038179781287908554 2023-01-21 11:09:37.209233: step: 372/529, loss: 0.0011805534595623612 2023-01-21 11:09:38.362685: step: 376/529, loss: 0.012799929827451706 2023-01-21 11:09:39.501513: step: 380/529, loss: 0.002487563993781805 2023-01-21 11:09:40.624275: step: 384/529, loss: 0.016067029908299446 2023-01-21 11:09:41.768160: step: 388/529, loss: 0.0014701009495183825 2023-01-21 11:09:42.940319: step: 392/529, loss: 0.3104632794857025 2023-01-21 11:09:44.088820: step: 396/529, loss: 0.020148087292909622 2023-01-21 11:09:45.223121: step: 400/529, loss: 0.0029549598693847656 2023-01-21 11:09:46.376681: step: 404/529, loss: 0.008225655183196068 2023-01-21 11:09:47.540901: step: 408/529, loss: 0.026165008544921875 2023-01-21 11:09:48.680515: step: 412/529, loss: 0.005286216735839844 2023-01-21 11:09:49.802063: step: 416/529, loss: 0.006191444583237171 2023-01-21 11:09:50.940871: step: 420/529, loss: 0.0009723663097247481 2023-01-21 11:09:52.092077: step: 424/529, loss: 0.018939495086669922 2023-01-21 11:09:53.214072: step: 428/529, loss: 0.01410207711160183 2023-01-21 11:09:54.348286: step: 432/529, loss: 0.020436527207493782 2023-01-21 11:09:55.507807: step: 436/529, loss: 0.08088894188404083 2023-01-21 11:09:56.650807: step: 440/529, loss: 0.004369902890175581 2023-01-21 11:09:57.814526: step: 444/529, loss: 0.007995223626494408 2023-01-21 11:09:58.927202: step: 448/529, loss: 0.000412178022088483 2023-01-21 11:10:00.064910: step: 452/529, loss: 0.055826760828495026 2023-01-21 11:10:01.208121: step: 456/529, loss: 0.030097246170043945 2023-01-21 11:10:02.337343: step: 460/529, loss: 0.7274318337440491 2023-01-21 11:10:03.483103: step: 464/529, loss: 0.002973270369693637 2023-01-21 11:10:04.639807: step: 468/529, loss: 0.012209320440888405 2023-01-21 11:10:05.767292: step: 472/529, loss: 0.03964290767908096 2023-01-21 11:10:06.885686: step: 476/529, loss: 0.0015967368381097913 2023-01-21 11:10:08.046257: step: 480/529, loss: 0.029360389336943626 2023-01-21 11:10:09.190922: step: 484/529, loss: 0.03325538709759712 2023-01-21 11:10:10.357694: step: 488/529, loss: 0.0014679909218102694 2023-01-21 11:10:11.500888: step: 492/529, loss: 0.00027151108952239156 2023-01-21 11:10:12.659243: step: 496/529, loss: 0.0003137588792014867 2023-01-21 11:10:13.802197: step: 500/529, loss: 0.1130281463265419 2023-01-21 11:10:14.954820: step: 504/529, loss: 0.0010694622760638595 2023-01-21 11:10:16.079145: step: 508/529, loss: 0.005390739534050226 2023-01-21 11:10:17.224869: step: 512/529, loss: 0.003537750570103526 2023-01-21 11:10:18.357774: step: 516/529, loss: 0.0022899149917066097 2023-01-21 11:10:19.489956: step: 520/529, loss: 0.012795163318514824 2023-01-21 11:10:20.661544: step: 524/529, loss: 0.023830413818359375 2023-01-21 11:10:21.802689: step: 528/529, loss: 0.08153896033763885 2023-01-21 11:10:22.951328: step: 532/529, loss: 0.003934192471206188 2023-01-21 11:10:24.092877: step: 536/529, loss: 0.04983186721801758 2023-01-21 11:10:25.228867: step: 540/529, loss: 0.024060631170868874 2023-01-21 11:10:26.380392: step: 544/529, loss: 0.00870666466653347 2023-01-21 11:10:27.551425: step: 548/529, loss: 0.0002626419300213456 2023-01-21 11:10:28.700447: step: 552/529, loss: 0.01973896101117134 2023-01-21 11:10:29.837994: step: 556/529, loss: 0.11025281250476837 2023-01-21 11:10:30.971768: step: 560/529, loss: 0.0601528063416481 2023-01-21 11:10:32.114943: step: 564/529, loss: 0.004730796907097101 2023-01-21 11:10:33.266382: step: 568/529, loss: 0.001201629638671875 2023-01-21 11:10:34.414666: step: 572/529, loss: 0.005031776614487171 2023-01-21 11:10:35.560372: step: 576/529, loss: 0.012799644842743874 2023-01-21 11:10:36.680343: step: 580/529, loss: 0.04560399055480957 2023-01-21 11:10:37.813853: step: 584/529, loss: 0.002846240997314453 2023-01-21 11:10:38.960622: step: 588/529, loss: 0.010749626904726028 2023-01-21 11:10:40.111070: step: 592/529, loss: 0.004678439814597368 2023-01-21 11:10:41.219750: step: 596/529, loss: 0.024710988625884056 2023-01-21 11:10:42.379720: step: 600/529, loss: 0.008387947455048561 2023-01-21 11:10:43.510061: step: 604/529, loss: 0.10133734345436096 2023-01-21 11:10:44.626761: step: 608/529, loss: 0.0006291388999670744 2023-01-21 11:10:45.782771: step: 612/529, loss: 0.008056163787841797 2023-01-21 11:10:46.910925: step: 616/529, loss: 0.015163994394242764 2023-01-21 11:10:48.040564: step: 620/529, loss: 0.005123329348862171 2023-01-21 11:10:49.141465: step: 624/529, loss: 0.032633017748594284 2023-01-21 11:10:50.302645: step: 628/529, loss: 0.22910137474536896 2023-01-21 11:10:51.470976: step: 632/529, loss: 0.031112540513277054 2023-01-21 11:10:52.574249: step: 636/529, loss: 0.009467506781220436 2023-01-21 11:10:53.729170: step: 640/529, loss: 0.013737487606704235 2023-01-21 11:10:54.868933: step: 644/529, loss: 0.03164634481072426 2023-01-21 11:10:55.999726: step: 648/529, loss: 0.02077159844338894 2023-01-21 11:10:57.146838: step: 652/529, loss: 0.05575375631451607 2023-01-21 11:10:58.259148: step: 656/529, loss: 0.00039901735726743937 2023-01-21 11:10:59.390991: step: 660/529, loss: 0.005500936880707741 2023-01-21 11:11:00.534144: step: 664/529, loss: 0.0033428191673010588 2023-01-21 11:11:01.706429: step: 668/529, loss: 0.0009239197243005037 2023-01-21 11:11:02.850266: step: 672/529, loss: 0.0018924714531749487 2023-01-21 11:11:03.979989: step: 676/529, loss: 0.008203268051147461 2023-01-21 11:11:05.135921: step: 680/529, loss: 0.021155167371034622 2023-01-21 11:11:06.286967: step: 684/529, loss: 0.10065823048353195 2023-01-21 11:11:07.440729: step: 688/529, loss: 0.04441213980317116 2023-01-21 11:11:08.576893: step: 692/529, loss: 0.09707680344581604 2023-01-21 11:11:09.720249: step: 696/529, loss: 0.014704705215990543 2023-01-21 11:11:10.843978: step: 700/529, loss: 0.006220054812729359 2023-01-21 11:11:11.973909: step: 704/529, loss: 0.04774494096636772 2023-01-21 11:11:13.119278: step: 708/529, loss: 0.0047775269486010075 2023-01-21 11:11:14.252439: step: 712/529, loss: 0.055693067610263824 2023-01-21 11:11:15.410368: step: 716/529, loss: 0.006354904267936945 2023-01-21 11:11:16.544127: step: 720/529, loss: 0.02730097621679306 2023-01-21 11:11:17.662256: step: 724/529, loss: 0.0012255669571459293 2023-01-21 11:11:18.830500: step: 728/529, loss: 0.019093705341219902 2023-01-21 11:11:19.992707: step: 732/529, loss: 0.03645744174718857 2023-01-21 11:11:21.145341: step: 736/529, loss: 0.0029304504860192537 2023-01-21 11:11:22.274894: step: 740/529, loss: 0.012084770947694778 2023-01-21 11:11:23.432718: step: 744/529, loss: 0.017528818920254707 2023-01-21 11:11:24.580681: step: 748/529, loss: 0.06217384338378906 2023-01-21 11:11:25.737126: step: 752/529, loss: 0.1284852921962738 2023-01-21 11:11:26.896084: step: 756/529, loss: 0.053660109639167786 2023-01-21 11:11:28.032190: step: 760/529, loss: 0.08294791728258133 2023-01-21 11:11:29.173362: step: 764/529, loss: 0.08515243977308273 2023-01-21 11:11:30.329421: step: 768/529, loss: 0.02444629743695259 2023-01-21 11:11:31.440323: step: 772/529, loss: 0.01609659381210804 2023-01-21 11:11:32.546930: step: 776/529, loss: 0.003753805300220847 2023-01-21 11:11:33.722538: step: 780/529, loss: 0.8426268100738525 2023-01-21 11:11:34.867912: step: 784/529, loss: 0.011741161346435547 2023-01-21 11:11:36.021661: step: 788/529, loss: 0.016111088916659355 2023-01-21 11:11:37.163162: step: 792/529, loss: 0.10392117500305176 2023-01-21 11:11:38.287183: step: 796/529, loss: 0.004294014070183039 2023-01-21 11:11:39.440374: step: 800/529, loss: 0.04252567142248154 2023-01-21 11:11:40.598182: step: 804/529, loss: 0.03582897037267685 2023-01-21 11:11:41.772848: step: 808/529, loss: 0.003137016436085105 2023-01-21 11:11:42.894789: step: 812/529, loss: 0.00883330125361681 2023-01-21 11:11:44.016287: step: 816/529, loss: 0.0005880355602130294 2023-01-21 11:11:45.142337: step: 820/529, loss: 0.04645185545086861 2023-01-21 11:11:46.271687: step: 824/529, loss: 0.029874280095100403 2023-01-21 11:11:47.401510: step: 828/529, loss: 0.0016699791885912418 2023-01-21 11:11:48.534483: step: 832/529, loss: 0.0007974624750204384 2023-01-21 11:11:49.675193: step: 836/529, loss: 0.04726262018084526 2023-01-21 11:11:50.832883: step: 840/529, loss: 0.004408645443618298 2023-01-21 11:11:51.969547: step: 844/529, loss: 0.011714362539350986 2023-01-21 11:11:53.103072: step: 848/529, loss: 0.027034951373934746 2023-01-21 11:11:54.240168: step: 852/529, loss: 0.07254581898450851 2023-01-21 11:11:55.356582: step: 856/529, loss: 0.023009682074189186 2023-01-21 11:11:56.499814: step: 860/529, loss: 0.041849710047245026 2023-01-21 11:11:57.656185: step: 864/529, loss: 0.0442989356815815 2023-01-21 11:11:58.792635: step: 868/529, loss: 0.025445176288485527 2023-01-21 11:11:59.934280: step: 872/529, loss: 0.13881874084472656 2023-01-21 11:12:01.083515: step: 876/529, loss: 0.07557277381420135 2023-01-21 11:12:02.205518: step: 880/529, loss: 0.005017471499741077 2023-01-21 11:12:03.333347: step: 884/529, loss: 0.01241073664277792 2023-01-21 11:12:04.511251: step: 888/529, loss: 0.017607688903808594 2023-01-21 11:12:05.659861: step: 892/529, loss: 0.00048313141451217234 2023-01-21 11:12:06.805955: step: 896/529, loss: 0.05135700851678848 2023-01-21 11:12:07.926542: step: 900/529, loss: 0.005811596289277077 2023-01-21 11:12:09.108482: step: 904/529, loss: 0.0035961628891527653 2023-01-21 11:12:10.230343: step: 908/529, loss: 0.002309179399162531 2023-01-21 11:12:11.375818: step: 912/529, loss: 0.025805141776800156 2023-01-21 11:12:12.494714: step: 916/529, loss: 0.0017173767555505037 2023-01-21 11:12:13.619357: step: 920/529, loss: 0.002993965055793524 2023-01-21 11:12:14.763971: step: 924/529, loss: 0.054241180419921875 2023-01-21 11:12:15.944077: step: 928/529, loss: 0.017448045313358307 2023-01-21 11:12:17.090790: step: 932/529, loss: 0.007323741912841797 2023-01-21 11:12:18.254554: step: 936/529, loss: 0.00948710460215807 2023-01-21 11:12:19.402488: step: 940/529, loss: 0.31747865676879883 2023-01-21 11:12:20.536774: step: 944/529, loss: 0.08458757400512695 2023-01-21 11:12:21.689623: step: 948/529, loss: 0.22754301130771637 2023-01-21 11:12:22.807563: step: 952/529, loss: 0.03138537332415581 2023-01-21 11:12:23.968162: step: 956/529, loss: 0.02237720601260662 2023-01-21 11:12:25.148913: step: 960/529, loss: 0.13312724232673645 2023-01-21 11:12:26.292630: step: 964/529, loss: 0.057062432169914246 2023-01-21 11:12:27.424294: step: 968/529, loss: 0.01851635053753853 2023-01-21 11:12:28.574714: step: 972/529, loss: 0.0032947540748864412 2023-01-21 11:12:29.731805: step: 976/529, loss: 8.287429955089465e-05 2023-01-21 11:12:30.882896: step: 980/529, loss: 0.00848779734224081 2023-01-21 11:12:32.034177: step: 984/529, loss: 0.000860536121763289 2023-01-21 11:12:33.129521: step: 988/529, loss: 0.06617279350757599 2023-01-21 11:12:34.235745: step: 992/529, loss: 0.03181810677051544 2023-01-21 11:12:35.361368: step: 996/529, loss: 0.19051896035671234 2023-01-21 11:12:36.509611: step: 1000/529, loss: 0.005600929260253906 2023-01-21 11:12:37.641716: step: 1004/529, loss: 0.03881397098302841 2023-01-21 11:12:38.767233: step: 1008/529, loss: 0.04361448436975479 2023-01-21 11:12:39.873743: step: 1012/529, loss: 0.013620281592011452 2023-01-21 11:12:41.046500: step: 1016/529, loss: 0.012393188662827015 2023-01-21 11:12:42.189593: step: 1020/529, loss: 0.023893356323242188 2023-01-21 11:12:43.296957: step: 1024/529, loss: 0.006225681398063898 2023-01-21 11:12:44.445798: step: 1028/529, loss: 0.002750969026237726 2023-01-21 11:12:45.584319: step: 1032/529, loss: 0.00141735072247684 2023-01-21 11:12:46.748509: step: 1036/529, loss: 0.0636373981833458 2023-01-21 11:12:47.884417: step: 1040/529, loss: 0.0546053871512413 2023-01-21 11:12:49.008267: step: 1044/529, loss: 0.0020420073997229338 2023-01-21 11:12:50.133771: step: 1048/529, loss: 0.00898046512156725 2023-01-21 11:12:51.271923: step: 1052/529, loss: 0.004096698481589556 2023-01-21 11:12:52.388005: step: 1056/529, loss: 0.03249092027544975 2023-01-21 11:12:53.541066: step: 1060/529, loss: 0.015903044492006302 2023-01-21 11:12:54.672786: step: 1064/529, loss: 0.013583565130829811 2023-01-21 11:12:55.833484: step: 1068/529, loss: 0.06061878055334091 2023-01-21 11:12:56.972186: step: 1072/529, loss: 0.04557066038250923 2023-01-21 11:12:58.121811: step: 1076/529, loss: 0.09788218140602112 2023-01-21 11:12:59.257807: step: 1080/529, loss: 0.003725338028743863 2023-01-21 11:13:00.409226: step: 1084/529, loss: 0.007439231500029564 2023-01-21 11:13:01.574238: step: 1088/529, loss: 0.002645206404849887 2023-01-21 11:13:02.737444: step: 1092/529, loss: 0.009668922051787376 2023-01-21 11:13:03.878640: step: 1096/529, loss: 0.04261379316449165 2023-01-21 11:13:05.012798: step: 1100/529, loss: 0.10369786620140076 2023-01-21 11:13:06.153742: step: 1104/529, loss: 0.04311056062579155 2023-01-21 11:13:07.357863: step: 1108/529, loss: 0.05699528753757477 2023-01-21 11:13:08.491263: step: 1112/529, loss: 0.013989115133881569 2023-01-21 11:13:09.622811: step: 1116/529, loss: 0.008242416195571423 2023-01-21 11:13:10.788502: step: 1120/529, loss: 0.022354699671268463 2023-01-21 11:13:11.929638: step: 1124/529, loss: 0.006768512539565563 2023-01-21 11:13:13.063198: step: 1128/529, loss: 0.0567021369934082 2023-01-21 11:13:14.161944: step: 1132/529, loss: 0.008379554376006126 2023-01-21 11:13:15.309833: step: 1136/529, loss: 0.021453000605106354 2023-01-21 11:13:16.428566: step: 1140/529, loss: 0.004947090521454811 2023-01-21 11:13:17.566747: step: 1144/529, loss: 0.005310439970344305 2023-01-21 11:13:18.699828: step: 1148/529, loss: 0.01039657648652792 2023-01-21 11:13:19.849747: step: 1152/529, loss: 0.0004621505504474044 2023-01-21 11:13:20.982500: step: 1156/529, loss: 0.017467260360717773 2023-01-21 11:13:22.094863: step: 1160/529, loss: 0.0006465911865234375 2023-01-21 11:13:23.236015: step: 1164/529, loss: 0.03760728985071182 2023-01-21 11:13:24.375150: step: 1168/529, loss: 0.04383974149823189 2023-01-21 11:13:25.522568: step: 1172/529, loss: 0.027658557519316673 2023-01-21 11:13:26.651797: step: 1176/529, loss: 0.03648385778069496 2023-01-21 11:13:27.784697: step: 1180/529, loss: 0.05360241234302521 2023-01-21 11:13:28.909148: step: 1184/529, loss: 0.014761734753847122 2023-01-21 11:13:30.035981: step: 1188/529, loss: 0.015715884044766426 2023-01-21 11:13:31.166859: step: 1192/529, loss: 0.02530203014612198 2023-01-21 11:13:32.298588: step: 1196/529, loss: 0.06698932498693466 2023-01-21 11:13:33.438541: step: 1200/529, loss: 0.04981184005737305 2023-01-21 11:13:34.577481: step: 1204/529, loss: 0.005845737177878618 2023-01-21 11:13:35.716146: step: 1208/529, loss: 0.01490030251443386 2023-01-21 11:13:36.847488: step: 1212/529, loss: 0.017954446375370026 2023-01-21 11:13:37.980379: step: 1216/529, loss: 0.016650009900331497 2023-01-21 11:13:39.122865: step: 1220/529, loss: 0.030155181884765625 2023-01-21 11:13:40.258417: step: 1224/529, loss: 0.003593230154365301 2023-01-21 11:13:41.382759: step: 1228/529, loss: 0.012989449314773083 2023-01-21 11:13:42.509556: step: 1232/529, loss: 0.04081685468554497 2023-01-21 11:13:43.656824: step: 1236/529, loss: 0.0027082921005785465 2023-01-21 11:13:44.878712: step: 1240/529, loss: 0.003213500836864114 2023-01-21 11:13:46.014043: step: 1244/529, loss: 0.016209697350859642 2023-01-21 11:13:47.188969: step: 1248/529, loss: 0.03333749994635582 2023-01-21 11:13:48.337389: step: 1252/529, loss: 0.04004526138305664 2023-01-21 11:13:49.483356: step: 1256/529, loss: 0.05194530636072159 2023-01-21 11:13:50.632902: step: 1260/529, loss: 0.010418986901640892 2023-01-21 11:13:51.793216: step: 1264/529, loss: 0.0014635085826739669 2023-01-21 11:13:52.931498: step: 1268/529, loss: 0.01530599594116211 2023-01-21 11:13:54.081607: step: 1272/529, loss: 0.006389713380485773 2023-01-21 11:13:55.211320: step: 1276/529, loss: 0.18706876039505005 2023-01-21 11:13:56.337690: step: 1280/529, loss: 0.04842853546142578 2023-01-21 11:13:57.484773: step: 1284/529, loss: 0.001101589179597795 2023-01-21 11:13:58.602659: step: 1288/529, loss: 0.1895170956850052 2023-01-21 11:13:59.723036: step: 1292/529, loss: 0.004184150602668524 2023-01-21 11:14:00.844563: step: 1296/529, loss: 0.004011726006865501 2023-01-21 11:14:01.981230: step: 1300/529, loss: 0.029828930273652077 2023-01-21 11:14:03.124855: step: 1304/529, loss: 0.09407033771276474 2023-01-21 11:14:04.263056: step: 1308/529, loss: 0.0029435157775878906 2023-01-21 11:14:05.390816: step: 1312/529, loss: 0.03441968187689781 2023-01-21 11:14:06.530192: step: 1316/529, loss: 0.0010502338409423828 2023-01-21 11:14:07.672714: step: 1320/529, loss: 0.006374835968017578 2023-01-21 11:14:08.791552: step: 1324/529, loss: 0.0007005095249041915 2023-01-21 11:14:09.945205: step: 1328/529, loss: 0.01744675636291504 2023-01-21 11:14:11.045576: step: 1332/529, loss: 0.02995932474732399 2023-01-21 11:14:12.198666: step: 1336/529, loss: 0.012354278936982155 2023-01-21 11:14:13.341278: step: 1340/529, loss: 0.00022754669771529734 2023-01-21 11:14:14.500201: step: 1344/529, loss: 0.00383758544921875 2023-01-21 11:14:15.658382: step: 1348/529, loss: 0.008125782012939453 2023-01-21 11:14:16.810753: step: 1352/529, loss: 0.3304021656513214 2023-01-21 11:14:17.949319: step: 1356/529, loss: 0.02353649027645588 2023-01-21 11:14:19.080844: step: 1360/529, loss: 0.04986514896154404 2023-01-21 11:14:20.217528: step: 1364/529, loss: 0.0038272857200354338 2023-01-21 11:14:21.360844: step: 1368/529, loss: 0.0012815475929528475 2023-01-21 11:14:22.541742: step: 1372/529, loss: 0.0014628410572186112 2023-01-21 11:14:23.673767: step: 1376/529, loss: 0.0026388170663267374 2023-01-21 11:14:24.817226: step: 1380/529, loss: 0.052590563893318176 2023-01-21 11:14:25.948461: step: 1384/529, loss: 0.003288173582404852 2023-01-21 11:14:27.094932: step: 1388/529, loss: 0.01676187478005886 2023-01-21 11:14:28.233760: step: 1392/529, loss: 0.04448271170258522 2023-01-21 11:14:29.382482: step: 1396/529, loss: 0.0007320403819903731 2023-01-21 11:14:30.480222: step: 1400/529, loss: 7.095337059581652e-05 2023-01-21 11:14:31.621700: step: 1404/529, loss: 0.016016578301787376 2023-01-21 11:14:32.767819: step: 1408/529, loss: 0.0011569024063646793 2023-01-21 11:14:33.889693: step: 1412/529, loss: 0.031043626368045807 2023-01-21 11:14:35.014613: step: 1416/529, loss: 0.05520286783576012 2023-01-21 11:14:36.145104: step: 1420/529, loss: 0.022416209802031517 2023-01-21 11:14:37.314937: step: 1424/529, loss: 0.0023966075386852026 2023-01-21 11:14:38.465752: step: 1428/529, loss: 0.0006723403930664062 2023-01-21 11:14:39.609162: step: 1432/529, loss: 0.02556753344833851 2023-01-21 11:14:40.760045: step: 1436/529, loss: 0.00012383461580611765 2023-01-21 11:14:41.917240: step: 1440/529, loss: 0.06725616753101349 2023-01-21 11:14:43.058532: step: 1444/529, loss: 0.08004990220069885 2023-01-21 11:14:44.183100: step: 1448/529, loss: 0.006039905361831188 2023-01-21 11:14:45.335962: step: 1452/529, loss: 0.022419167682528496 2023-01-21 11:14:46.506551: step: 1456/529, loss: 0.008258628658950329 2023-01-21 11:14:47.651020: step: 1460/529, loss: 0.013870335184037685 2023-01-21 11:14:48.802572: step: 1464/529, loss: 0.036595914512872696 2023-01-21 11:14:49.979674: step: 1468/529, loss: 0.005097007844597101 2023-01-21 11:14:51.117448: step: 1472/529, loss: 0.010624503716826439 2023-01-21 11:14:52.258399: step: 1476/529, loss: 0.0012076377170160413 2023-01-21 11:14:53.420598: step: 1480/529, loss: 0.019147682934999466 2023-01-21 11:14:54.548019: step: 1484/529, loss: 0.050473593175411224 2023-01-21 11:14:55.680953: step: 1488/529, loss: 0.0036925317253917456 2023-01-21 11:14:56.786268: step: 1492/529, loss: 0.012013006024062634 2023-01-21 11:14:57.973373: step: 1496/529, loss: 0.04311714321374893 2023-01-21 11:14:59.119470: step: 1500/529, loss: 0.0006580352783203125 2023-01-21 11:15:00.267593: step: 1504/529, loss: 0.043698880821466446 2023-01-21 11:15:01.403799: step: 1508/529, loss: 0.013142204843461514 2023-01-21 11:15:02.569989: step: 1512/529, loss: 0.07252349704504013 2023-01-21 11:15:03.678253: step: 1516/529, loss: 0.04689350351691246 2023-01-21 11:15:04.833480: step: 1520/529, loss: 0.049704551696777344 2023-01-21 11:15:05.974950: step: 1524/529, loss: 0.002729797502979636 2023-01-21 11:15:07.122205: step: 1528/529, loss: 0.021985625848174095 2023-01-21 11:15:08.272288: step: 1532/529, loss: 0.040839388966560364 2023-01-21 11:15:09.435447: step: 1536/529, loss: 0.019243812188506126 2023-01-21 11:15:10.594405: step: 1540/529, loss: 0.03152446821331978 2023-01-21 11:15:11.744265: step: 1544/529, loss: 0.08307524025440216 2023-01-21 11:15:12.883860: step: 1548/529, loss: 0.057274628430604935 2023-01-21 11:15:14.047185: step: 1552/529, loss: 0.027982329949736595 2023-01-21 11:15:15.186240: step: 1556/529, loss: 0.020365620031952858 2023-01-21 11:15:16.326837: step: 1560/529, loss: 0.058455564081668854 2023-01-21 11:15:17.484894: step: 1564/529, loss: 0.015607072040438652 2023-01-21 11:15:18.619849: step: 1568/529, loss: 0.02716102823615074 2023-01-21 11:15:19.767490: step: 1572/529, loss: 0.0123648876324296 2023-01-21 11:15:20.896524: step: 1576/529, loss: 0.011349297128617764 2023-01-21 11:15:22.040027: step: 1580/529, loss: 0.014312363229691982 2023-01-21 11:15:23.192414: step: 1584/529, loss: 0.019231844693422318 2023-01-21 11:15:24.329897: step: 1588/529, loss: 0.0054531097412109375 2023-01-21 11:15:25.434301: step: 1592/529, loss: 0.07473921775817871 2023-01-21 11:15:26.585476: step: 1596/529, loss: 0.029568098485469818 2023-01-21 11:15:27.769450: step: 1600/529, loss: 0.016063785180449486 2023-01-21 11:15:28.904718: step: 1604/529, loss: 0.017739344388246536 2023-01-21 11:15:30.047555: step: 1608/529, loss: 0.028479887172579765 2023-01-21 11:15:31.194657: step: 1612/529, loss: 0.01737537421286106 2023-01-21 11:15:32.333673: step: 1616/529, loss: 0.06229591369628906 2023-01-21 11:15:33.479237: step: 1620/529, loss: 0.009745597839355469 2023-01-21 11:15:34.634868: step: 1624/529, loss: 0.037886809557676315 2023-01-21 11:15:35.785598: step: 1628/529, loss: 0.021085739135742188 2023-01-21 11:15:36.922837: step: 1632/529, loss: 0.062497518956661224 2023-01-21 11:15:38.081632: step: 1636/529, loss: 0.0496278777718544 2023-01-21 11:15:39.211765: step: 1640/529, loss: 0.026555635035037994 2023-01-21 11:15:40.325203: step: 1644/529, loss: 0.016200972720980644 2023-01-21 11:15:41.471632: step: 1648/529, loss: 0.009702110663056374 2023-01-21 11:15:42.621308: step: 1652/529, loss: 0.024866199120879173 2023-01-21 11:15:43.753664: step: 1656/529, loss: 0.00010375976853538305 2023-01-21 11:15:44.900768: step: 1660/529, loss: 0.02479419857263565 2023-01-21 11:15:46.038390: step: 1664/529, loss: 0.005896473303437233 2023-01-21 11:15:47.161334: step: 1668/529, loss: 0.004992770962417126 2023-01-21 11:15:48.294215: step: 1672/529, loss: 0.0034923553466796875 2023-01-21 11:15:49.454679: step: 1676/529, loss: 0.03645153343677521 2023-01-21 11:15:50.595941: step: 1680/529, loss: 0.12226153165102005 2023-01-21 11:15:51.731776: step: 1684/529, loss: 0.004172611515969038 2023-01-21 11:15:52.889309: step: 1688/529, loss: 0.11449889838695526 2023-01-21 11:15:54.017561: step: 1692/529, loss: 0.04302450641989708 2023-01-21 11:15:55.183771: step: 1696/529, loss: 0.016445159912109375 2023-01-21 11:15:56.323200: step: 1700/529, loss: 0.01430602092295885 2023-01-21 11:15:57.518494: step: 1704/529, loss: 0.022727251052856445 2023-01-21 11:15:58.661629: step: 1708/529, loss: 0.0005927563179284334 2023-01-21 11:15:59.793435: step: 1712/529, loss: 0.06709681451320648 2023-01-21 11:16:00.943111: step: 1716/529, loss: 0.08925037086009979 2023-01-21 11:16:02.086973: step: 1720/529, loss: 0.010358238592743874 2023-01-21 11:16:03.195226: step: 1724/529, loss: 0.0007161140674725175 2023-01-21 11:16:04.373248: step: 1728/529, loss: 0.018272781744599342 2023-01-21 11:16:05.510602: step: 1732/529, loss: 0.0007228851318359375 2023-01-21 11:16:06.690298: step: 1736/529, loss: 0.037283290177583694 2023-01-21 11:16:07.827091: step: 1740/529, loss: 0.00985937099903822 2023-01-21 11:16:08.952368: step: 1744/529, loss: 0.019458914175629616 2023-01-21 11:16:10.066351: step: 1748/529, loss: 0.0015424728626385331 2023-01-21 11:16:11.202989: step: 1752/529, loss: 0.03205299377441406 2023-01-21 11:16:12.352638: step: 1756/529, loss: 0.040040016174316406 2023-01-21 11:16:13.476325: step: 1760/529, loss: 0.12323351204395294 2023-01-21 11:16:14.604311: step: 1764/529, loss: 0.019593574106693268 2023-01-21 11:16:15.727911: step: 1768/529, loss: 0.04877519607543945 2023-01-21 11:16:16.866057: step: 1772/529, loss: 0.0030515671242028475 2023-01-21 11:16:17.984335: step: 1776/529, loss: 0.2574901878833771 2023-01-21 11:16:19.111083: step: 1780/529, loss: 0.004457664210349321 2023-01-21 11:16:20.257327: step: 1784/529, loss: 0.038317110389471054 2023-01-21 11:16:21.410666: step: 1788/529, loss: 0.007608700077980757 2023-01-21 11:16:22.557451: step: 1792/529, loss: 0.009356451220810413 2023-01-21 11:16:23.718456: step: 1796/529, loss: 0.014154816046357155 2023-01-21 11:16:24.856952: step: 1800/529, loss: 0.003493881318718195 2023-01-21 11:16:25.995784: step: 1804/529, loss: 0.018837666139006615 2023-01-21 11:16:27.138594: step: 1808/529, loss: 0.004622268490493298 2023-01-21 11:16:28.254817: step: 1812/529, loss: 0.006932544521987438 2023-01-21 11:16:29.356769: step: 1816/529, loss: 0.020758438855409622 2023-01-21 11:16:30.489027: step: 1820/529, loss: 0.0020929337479174137 2023-01-21 11:16:31.612066: step: 1824/529, loss: 0.006173992529511452 2023-01-21 11:16:32.769436: step: 1828/529, loss: 0.0010514259338378906 2023-01-21 11:16:33.898919: step: 1832/529, loss: 0.0023732902482151985 2023-01-21 11:16:35.035810: step: 1836/529, loss: 0.007515525911003351 2023-01-21 11:16:36.147489: step: 1840/529, loss: 0.06514782458543777 2023-01-21 11:16:37.289687: step: 1844/529, loss: 0.03289084881544113 2023-01-21 11:16:38.406397: step: 1848/529, loss: 0.01877593994140625 2023-01-21 11:16:39.531274: step: 1852/529, loss: 0.011627770029008389 2023-01-21 11:16:40.675838: step: 1856/529, loss: 0.0002834320184774697 2023-01-21 11:16:41.824472: step: 1860/529, loss: 0.05669088289141655 2023-01-21 11:16:42.977582: step: 1864/529, loss: 0.023387527093291283 2023-01-21 11:16:44.097510: step: 1868/529, loss: 0.059326935559511185 2023-01-21 11:16:45.242664: step: 1872/529, loss: 0.0014371873112395406 2023-01-21 11:16:46.363360: step: 1876/529, loss: 0.005904007237404585 2023-01-21 11:16:47.497991: step: 1880/529, loss: 0.010495376773178577 2023-01-21 11:16:48.676421: step: 1884/529, loss: 0.02429637871682644 2023-01-21 11:16:49.811756: step: 1888/529, loss: 0.030131960287690163 2023-01-21 11:16:50.968779: step: 1892/529, loss: 0.11261577904224396 2023-01-21 11:16:52.112754: step: 1896/529, loss: 0.13868561387062073 2023-01-21 11:16:53.251099: step: 1900/529, loss: 0.05415001139044762 2023-01-21 11:16:54.390912: step: 1904/529, loss: 0.05102949216961861 2023-01-21 11:16:55.517548: step: 1908/529, loss: 0.024415969848632812 2023-01-21 11:16:56.673619: step: 1912/529, loss: 0.03174114227294922 2023-01-21 11:16:57.812797: step: 1916/529, loss: 0.004187393467873335 2023-01-21 11:16:58.929246: step: 1920/529, loss: 0.028246738016605377 2023-01-21 11:17:00.068798: step: 1924/529, loss: 0.0053529744036495686 2023-01-21 11:17:01.193680: step: 1928/529, loss: 0.0418216697871685 2023-01-21 11:17:02.356366: step: 1932/529, loss: 0.030275821685791016 2023-01-21 11:17:03.495633: step: 1936/529, loss: 0.0005325317615643144 2023-01-21 11:17:04.670964: step: 1940/529, loss: 0.00972747802734375 2023-01-21 11:17:05.816875: step: 1944/529, loss: 0.020929764956235886 2023-01-21 11:17:06.966892: step: 1948/529, loss: 0.6107298135757446 2023-01-21 11:17:08.105940: step: 1952/529, loss: 0.0022125244140625 2023-01-21 11:17:09.251331: step: 1956/529, loss: 0.06037025526165962 2023-01-21 11:17:10.388475: step: 1960/529, loss: 0.20495624840259552 2023-01-21 11:17:11.529887: step: 1964/529, loss: 0.06659393012523651 2023-01-21 11:17:12.655862: step: 1968/529, loss: 0.015048789791762829 2023-01-21 11:17:13.803173: step: 1972/529, loss: 0.05694231390953064 2023-01-21 11:17:14.959612: step: 1976/529, loss: 0.05497036129236221 2023-01-21 11:17:16.108543: step: 1980/529, loss: 0.013063526712357998 2023-01-21 11:17:17.240665: step: 1984/529, loss: 0.014065265655517578 2023-01-21 11:17:18.388382: step: 1988/529, loss: 0.030993271619081497 2023-01-21 11:17:19.542990: step: 1992/529, loss: 0.006761837285012007 2023-01-21 11:17:20.683950: step: 1996/529, loss: 0.0004570961173158139 2023-01-21 11:17:21.816255: step: 2000/529, loss: 0.0019927024841308594 2023-01-21 11:17:22.936629: step: 2004/529, loss: 0.01276693306863308 2023-01-21 11:17:24.072953: step: 2008/529, loss: 0.036294981837272644 2023-01-21 11:17:25.259482: step: 2012/529, loss: 0.07033786177635193 2023-01-21 11:17:26.398871: step: 2016/529, loss: 0.004414939787238836 2023-01-21 11:17:27.534270: step: 2020/529, loss: 0.0027147294022142887 2023-01-21 11:17:28.682540: step: 2024/529, loss: 0.04505958408117294 2023-01-21 11:17:29.837759: step: 2028/529, loss: 0.022596361115574837 2023-01-21 11:17:31.000976: step: 2032/529, loss: 0.013995361514389515 2023-01-21 11:17:32.117488: step: 2036/529, loss: 0.0021209716796875 2023-01-21 11:17:33.271124: step: 2040/529, loss: 0.030959319323301315 2023-01-21 11:17:34.419580: step: 2044/529, loss: 0.0828220322728157 2023-01-21 11:17:35.566628: step: 2048/529, loss: 0.01774454116821289 2023-01-21 11:17:36.748230: step: 2052/529, loss: 0.21185509860515594 2023-01-21 11:17:37.917504: step: 2056/529, loss: 0.018067264929413795 2023-01-21 11:17:39.046553: step: 2060/529, loss: 0.014842080883681774 2023-01-21 11:17:40.184759: step: 2064/529, loss: 0.0048958780243992805 2023-01-21 11:17:41.321781: step: 2068/529, loss: 0.04531888663768768 2023-01-21 11:17:42.468114: step: 2072/529, loss: 0.040624428540468216 2023-01-21 11:17:43.618180: step: 2076/529, loss: 0.035187624394893646 2023-01-21 11:17:44.763128: step: 2080/529, loss: 0.023624802008271217 2023-01-21 11:17:45.894331: step: 2084/529, loss: 0.00090875627938658 2023-01-21 11:17:47.057516: step: 2088/529, loss: 0.08014983683824539 2023-01-21 11:17:48.183266: step: 2092/529, loss: 0.03574104607105255 2023-01-21 11:17:49.307856: step: 2096/529, loss: 0.0019851685501635075 2023-01-21 11:17:50.432329: step: 2100/529, loss: 0.027103662490844727 2023-01-21 11:17:51.575221: step: 2104/529, loss: 0.007000541780143976 2023-01-21 11:17:52.708578: step: 2108/529, loss: 0.008797550573945045 2023-01-21 11:17:53.827102: step: 2112/529, loss: 0.00836801528930664 2023-01-21 11:17:54.967639: step: 2116/529, loss: 0.013452769257128239 ================================================== Loss: 0.038 -------------------- Dev: {'event': {'p': 0.6070686070686071, 'r': 0.7776298268974701, 'f1': 0.6818447168709866}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Test: {'event': {'p': 0.613290263036456, 'r': 0.7699884125144844, 'f1': 0.6827639352684306}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Chinese: {'event': {'p': 0.5813953488372093, 'r': 0.9259259259259259, 'f1': 0.7142857142857142}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Russian: {'event': {'p': 0.4358974358974359, 'r': 0.4722222222222222, 'f1': 0.45333333333333337}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 11:18:36.880889: step: 4/529, loss: 0.001281786011531949 2023-01-21 11:18:38.000506: step: 8/529, loss: 0.0014184952015057206 2023-01-21 11:18:39.120989: step: 12/529, loss: 0.02671823650598526 2023-01-21 11:18:40.237496: step: 16/529, loss: 0.002808952471241355 2023-01-21 11:18:41.356581: step: 20/529, loss: 0.0007003784412518144 2023-01-21 11:18:42.472172: step: 24/529, loss: 0.006023192312568426 2023-01-21 11:18:43.648136: step: 28/529, loss: 0.055132679641246796 2023-01-21 11:18:44.783712: step: 32/529, loss: 0.029755974188447 2023-01-21 11:18:45.926061: step: 36/529, loss: 0.02724306657910347 2023-01-21 11:18:47.049482: step: 40/529, loss: 0.29437941312789917 2023-01-21 11:18:48.163935: step: 44/529, loss: 0.05317831039428711 2023-01-21 11:18:49.306938: step: 48/529, loss: 0.014949607662856579 2023-01-21 11:18:50.427345: step: 52/529, loss: 0.037228774279356 2023-01-21 11:18:51.564449: step: 56/529, loss: 0.027472306042909622 2023-01-21 11:18:52.672243: step: 60/529, loss: 0.05710430070757866 2023-01-21 11:18:53.827016: step: 64/529, loss: 0.05984363704919815 2023-01-21 11:18:54.936987: step: 68/529, loss: 0.818611204624176 2023-01-21 11:18:56.058402: step: 72/529, loss: 0.012174416333436966 2023-01-21 11:18:57.183668: step: 76/529, loss: 0.010378837585449219 2023-01-21 11:18:58.351503: step: 80/529, loss: 0.015028666704893112 2023-01-21 11:18:59.466162: step: 84/529, loss: 0.04272763803601265 2023-01-21 11:19:00.601005: step: 88/529, loss: 0.020542336627840996 2023-01-21 11:19:01.750299: step: 92/529, loss: 0.002444410463795066 2023-01-21 11:19:02.898201: step: 96/529, loss: 0.03330212086439133 2023-01-21 11:19:04.018745: step: 100/529, loss: 0.029061317443847656 2023-01-21 11:19:05.139932: step: 104/529, loss: 0.01877441443502903 2023-01-21 11:19:06.251869: step: 108/529, loss: 0.017064666375517845 2023-01-21 11:19:07.394065: step: 112/529, loss: 0.005825853440910578 2023-01-21 11:19:08.523383: step: 116/529, loss: 0.010346747003495693 2023-01-21 11:19:09.662439: step: 120/529, loss: 0.017881011590361595 2023-01-21 11:19:10.808018: step: 124/529, loss: 0.013179779052734375 2023-01-21 11:19:11.940270: step: 128/529, loss: 0.0008934021461755037 2023-01-21 11:19:13.086051: step: 132/529, loss: 0.008393668569624424 2023-01-21 11:19:14.242521: step: 136/529, loss: 0.020969772711396217 2023-01-21 11:19:15.386584: step: 140/529, loss: 0.0016659736866131425 2023-01-21 11:19:16.549133: step: 144/529, loss: 0.04127998277544975 2023-01-21 11:19:17.703063: step: 148/529, loss: 0.015370846726000309 2023-01-21 11:19:18.848833: step: 152/529, loss: 0.005079889204353094 2023-01-21 11:19:20.001562: step: 156/529, loss: 0.0354466438293457 2023-01-21 11:19:21.145130: step: 160/529, loss: 0.015561294741928577 2023-01-21 11:19:22.260571: step: 164/529, loss: 0.019716214388608932 2023-01-21 11:19:23.370781: step: 168/529, loss: 0.039950087666511536 2023-01-21 11:19:24.550771: step: 172/529, loss: 0.0864294022321701 2023-01-21 11:19:25.695722: step: 176/529, loss: 0.01704864576458931 2023-01-21 11:19:26.829456: step: 180/529, loss: 0.0011204719776287675 2023-01-21 11:19:27.950688: step: 184/529, loss: 0.022375011816620827 2023-01-21 11:19:29.091401: step: 188/529, loss: 0.023101806640625 2023-01-21 11:19:30.254896: step: 192/529, loss: 0.013962794095277786 2023-01-21 11:19:31.401761: step: 196/529, loss: 0.0011806488037109375 2023-01-21 11:19:32.537805: step: 200/529, loss: 0.00344600691460073 2023-01-21 11:19:33.662314: step: 204/529, loss: 0.019872188568115234 2023-01-21 11:19:34.790601: step: 208/529, loss: 0.040847018361091614 2023-01-21 11:19:35.938721: step: 212/529, loss: 0.031068801879882812 2023-01-21 11:19:37.087087: step: 216/529, loss: 0.2228774130344391 2023-01-21 11:19:38.251928: step: 220/529, loss: 0.0023597716353833675 2023-01-21 11:19:39.400709: step: 224/529, loss: 0.049439672380685806 2023-01-21 11:19:40.518974: step: 228/529, loss: 0.12172757089138031 2023-01-21 11:19:41.687616: step: 232/529, loss: 0.0013475418090820312 2023-01-21 11:19:42.803154: step: 236/529, loss: 0.0004077911435160786 2023-01-21 11:19:43.963084: step: 240/529, loss: 0.0029257775750011206 2023-01-21 11:19:45.121460: step: 244/529, loss: 0.0007245540618896484 2023-01-21 11:19:46.274403: step: 248/529, loss: 0.027238082140684128 2023-01-21 11:19:47.402577: step: 252/529, loss: 0.017442703247070312 2023-01-21 11:19:48.555232: step: 256/529, loss: 0.0006719589000567794 2023-01-21 11:19:49.667970: step: 260/529, loss: 0.0037513733841478825 2023-01-21 11:19:50.778946: step: 264/529, loss: 0.0024683952797204256 2023-01-21 11:19:51.962985: step: 268/529, loss: 0.0041824341751635075 2023-01-21 11:19:53.089666: step: 272/529, loss: 0.05124969780445099 2023-01-21 11:19:54.224015: step: 276/529, loss: 0.03738613426685333 2023-01-21 11:19:55.383694: step: 280/529, loss: 0.20684003829956055 2023-01-21 11:19:56.504819: step: 284/529, loss: 0.01148614939302206 2023-01-21 11:19:57.628517: step: 288/529, loss: 0.007374573033303022 2023-01-21 11:19:58.763960: step: 292/529, loss: 0.0002853400947060436 2023-01-21 11:19:59.908939: step: 296/529, loss: 0.010897351428866386 2023-01-21 11:20:01.062997: step: 300/529, loss: 0.004628848750144243 2023-01-21 11:20:02.195184: step: 304/529, loss: 0.030051900073885918 2023-01-21 11:20:03.317297: step: 308/529, loss: 0.03163433074951172 2023-01-21 11:20:04.467093: step: 312/529, loss: 0.0016204833518713713 2023-01-21 11:20:05.648956: step: 316/529, loss: 0.0462338924407959 2023-01-21 11:20:06.781914: step: 320/529, loss: 0.007203841581940651 2023-01-21 11:20:07.939085: step: 324/529, loss: 0.027962829917669296 2023-01-21 11:20:09.072436: step: 328/529, loss: 0.025838088244199753 2023-01-21 11:20:10.219445: step: 332/529, loss: 0.0007618904346600175 2023-01-21 11:20:11.356215: step: 336/529, loss: 0.02007732354104519 2023-01-21 11:20:12.511566: step: 340/529, loss: 0.010866357013583183 2023-01-21 11:20:13.665923: step: 344/529, loss: 0.021649742498993874 2023-01-21 11:20:14.802786: step: 348/529, loss: 0.1568128615617752 2023-01-21 11:20:15.949486: step: 352/529, loss: 0.007755470462143421 2023-01-21 11:20:17.101863: step: 356/529, loss: 0.01465530414134264 2023-01-21 11:20:18.248488: step: 360/529, loss: 0.003658866975456476 2023-01-21 11:20:19.391225: step: 364/529, loss: 0.03625817596912384 2023-01-21 11:20:20.529301: step: 368/529, loss: 0.010960579849779606 2023-01-21 11:20:21.696078: step: 372/529, loss: 0.10849571228027344 2023-01-21 11:20:22.832591: step: 376/529, loss: 0.01132059097290039 2023-01-21 11:20:23.953084: step: 380/529, loss: 0.015038777142763138 2023-01-21 11:20:25.097551: step: 384/529, loss: 0.01786346361041069 2023-01-21 11:20:26.213135: step: 388/529, loss: 0.15416479110717773 2023-01-21 11:20:27.365855: step: 392/529, loss: 0.004218196962028742 2023-01-21 11:20:28.520427: step: 396/529, loss: 0.006312168203294277 2023-01-21 11:20:29.646291: step: 400/529, loss: 0.05187266319990158 2023-01-21 11:20:30.791904: step: 404/529, loss: 0.02471904829144478 2023-01-21 11:20:31.916697: step: 408/529, loss: 0.024489115923643112 2023-01-21 11:20:33.066373: step: 412/529, loss: 0.03293609619140625 2023-01-21 11:20:34.180023: step: 416/529, loss: 0.06496404111385345 2023-01-21 11:20:35.312564: step: 420/529, loss: 0.05026718229055405 2023-01-21 11:20:36.434574: step: 424/529, loss: 0.007446670439094305 2023-01-21 11:20:37.588854: step: 428/529, loss: 0.0482393279671669 2023-01-21 11:20:38.717307: step: 432/529, loss: 0.00011339187767589465 2023-01-21 11:20:39.879507: step: 436/529, loss: 0.016169849783182144 2023-01-21 11:20:41.009010: step: 440/529, loss: 0.0003383636358194053 2023-01-21 11:20:42.147223: step: 444/529, loss: 0.012859517708420753 2023-01-21 11:20:43.301768: step: 448/529, loss: 0.019469834864139557 2023-01-21 11:20:44.443465: step: 452/529, loss: 0.03647594526410103 2023-01-21 11:20:45.557829: step: 456/529, loss: 0.028537657111883163 2023-01-21 11:20:46.704465: step: 460/529, loss: 0.01862783543765545 2023-01-21 11:20:47.828307: step: 464/529, loss: 0.02925877645611763 2023-01-21 11:20:48.961415: step: 468/529, loss: 0.0005485534784384072 2023-01-21 11:20:50.102760: step: 472/529, loss: 0.014030457474291325 2023-01-21 11:20:51.236233: step: 476/529, loss: 0.045073606073856354 2023-01-21 11:20:52.384070: step: 480/529, loss: 0.010161971673369408 2023-01-21 11:20:53.546743: step: 484/529, loss: 2.371072696405463e-05 2023-01-21 11:20:54.683117: step: 488/529, loss: 0.017067719250917435 2023-01-21 11:20:55.795736: step: 492/529, loss: 0.027418233454227448 2023-01-21 11:20:56.935435: step: 496/529, loss: 0.009294891729950905 2023-01-21 11:20:58.046097: step: 500/529, loss: 0.005720901302993298 2023-01-21 11:20:59.197692: step: 504/529, loss: 0.0652192085981369 2023-01-21 11:21:00.360150: step: 508/529, loss: 0.10440702736377716 2023-01-21 11:21:01.534187: step: 512/529, loss: 0.04993128776550293 2023-01-21 11:21:02.702683: step: 516/529, loss: 0.006574440281838179 2023-01-21 11:21:03.856718: step: 520/529, loss: 0.002021408174186945 2023-01-21 11:21:04.973840: step: 524/529, loss: 0.01113739050924778 2023-01-21 11:21:06.130739: step: 528/529, loss: 0.023603392764925957 2023-01-21 11:21:07.263437: step: 532/529, loss: 0.02385425753891468 2023-01-21 11:21:08.377481: step: 536/529, loss: 0.03247556835412979 2023-01-21 11:21:09.527951: step: 540/529, loss: 0.03999767452478409 2023-01-21 11:21:10.678787: step: 544/529, loss: 0.09994792938232422 2023-01-21 11:21:11.827854: step: 548/529, loss: 0.015366173349320889 2023-01-21 11:21:12.974972: step: 552/529, loss: 0.0015928269131109118 2023-01-21 11:21:14.103192: step: 556/529, loss: 0.016782188788056374 2023-01-21 11:21:15.241574: step: 560/529, loss: 0.00019073486328125 2023-01-21 11:21:16.366824: step: 564/529, loss: 0.015441703610122204 2023-01-21 11:21:17.527911: step: 568/529, loss: 0.02672004885971546 2023-01-21 11:21:18.643688: step: 572/529, loss: 7.219314284157008e-05 2023-01-21 11:21:19.738053: step: 576/529, loss: 0.01328277587890625 2023-01-21 11:21:20.870330: step: 580/529, loss: 0.010341835208237171 2023-01-21 11:21:22.005589: step: 584/529, loss: 0.011991810984909534 2023-01-21 11:21:23.159334: step: 588/529, loss: 0.009359002113342285 2023-01-21 11:21:24.287272: step: 592/529, loss: 0.06405448913574219 2023-01-21 11:21:25.434459: step: 596/529, loss: 0.011866187676787376 2023-01-21 11:21:26.572264: step: 600/529, loss: 0.004872036166489124 2023-01-21 11:21:27.670934: step: 604/529, loss: 0.0012449742062017322 2023-01-21 11:21:28.791785: step: 608/529, loss: 0.010881043039262295 2023-01-21 11:21:29.949527: step: 612/529, loss: 0.04543018713593483 2023-01-21 11:21:31.099647: step: 616/529, loss: 0.031232308596372604 2023-01-21 11:21:32.262400: step: 620/529, loss: 0.011626816354691982 2023-01-21 11:21:33.402215: step: 624/529, loss: 6.079673767089844e-05 2023-01-21 11:21:34.559972: step: 628/529, loss: 0.01129298284649849 2023-01-21 11:21:35.697387: step: 632/529, loss: 0.06663388758897781 2023-01-21 11:21:36.850544: step: 636/529, loss: 0.002163791796192527 2023-01-21 11:21:37.996027: step: 640/529, loss: 0.023525238037109375 2023-01-21 11:21:39.114141: step: 644/529, loss: 0.003359127091243863 2023-01-21 11:21:40.250075: step: 648/529, loss: 0.0018508911598473787 2023-01-21 11:21:41.375922: step: 652/529, loss: 0.00677757291123271 2023-01-21 11:21:42.496268: step: 656/529, loss: 0.0023449896834790707 2023-01-21 11:21:43.629779: step: 660/529, loss: 0.0024999617598950863 2023-01-21 11:21:44.758170: step: 664/529, loss: 0.003799057099968195 2023-01-21 11:21:45.905778: step: 668/529, loss: 0.01631317101418972 2023-01-21 11:21:47.082921: step: 672/529, loss: 0.023102665320038795 2023-01-21 11:21:48.222983: step: 676/529, loss: 0.0014797747135162354 2023-01-21 11:21:49.348795: step: 680/529, loss: 0.0326116569340229 2023-01-21 11:21:50.532271: step: 684/529, loss: 0.005245017819106579 2023-01-21 11:21:51.657515: step: 688/529, loss: 0.011355400085449219 2023-01-21 11:21:52.786665: step: 692/529, loss: 0.010459757409989834 2023-01-21 11:21:53.918286: step: 696/529, loss: 0.006140995305031538 2023-01-21 11:21:55.048352: step: 700/529, loss: 0.0497777946293354 2023-01-21 11:21:56.183719: step: 704/529, loss: 0.057834457606077194 2023-01-21 11:21:57.323352: step: 708/529, loss: 0.04412546381354332 2023-01-21 11:21:58.452265: step: 712/529, loss: 0.022747039794921875 2023-01-21 11:21:59.630404: step: 716/529, loss: 0.05197286605834961 2023-01-21 11:22:00.753932: step: 720/529, loss: 0.006995964329689741 2023-01-21 11:22:01.885341: step: 724/529, loss: 0.011628532782196999 2023-01-21 11:22:03.032469: step: 728/529, loss: 0.0024467469193041325 2023-01-21 11:22:04.150149: step: 732/529, loss: 0.012562013231217861 2023-01-21 11:22:05.268500: step: 736/529, loss: 0.03850231319665909 2023-01-21 11:22:06.395496: step: 740/529, loss: 0.019614219665527344 2023-01-21 11:22:07.506956: step: 744/529, loss: 0.0022322656586766243 2023-01-21 11:22:08.645122: step: 748/529, loss: 0.05734901875257492 2023-01-21 11:22:09.768961: step: 752/529, loss: 0.0025336265098303556 2023-01-21 11:22:10.929047: step: 756/529, loss: 0.041613198816776276 2023-01-21 11:22:12.108467: step: 760/529, loss: 0.002742481417953968 2023-01-21 11:22:13.230698: step: 764/529, loss: 0.04124489054083824 2023-01-21 11:22:14.336171: step: 768/529, loss: 0.013933181762695312 2023-01-21 11:22:15.468757: step: 772/529, loss: 0.12521381676197052 2023-01-21 11:22:16.595740: step: 776/529, loss: 0.029155541211366653 2023-01-21 11:22:17.740640: step: 780/529, loss: 0.15854930877685547 2023-01-21 11:22:18.913244: step: 784/529, loss: 0.04188594967126846 2023-01-21 11:22:20.036505: step: 788/529, loss: 0.011525154113769531 2023-01-21 11:22:21.198329: step: 792/529, loss: 0.007031917572021484 2023-01-21 11:22:22.318909: step: 796/529, loss: 0.016255665570497513 2023-01-21 11:22:23.450554: step: 800/529, loss: 0.021538352593779564 2023-01-21 11:22:24.616317: step: 804/529, loss: 0.014235210604965687 2023-01-21 11:22:25.751086: step: 808/529, loss: 0.0015422821743413806 2023-01-21 11:22:26.898768: step: 812/529, loss: 0.0009525298955850303 2023-01-21 11:22:28.038286: step: 816/529, loss: 0.014393424615263939 2023-01-21 11:22:29.154270: step: 820/529, loss: 0.025265691801905632 2023-01-21 11:22:30.286622: step: 824/529, loss: 0.014307593926787376 2023-01-21 11:22:31.427719: step: 828/529, loss: 0.040863037109375 2023-01-21 11:22:32.552749: step: 832/529, loss: 0.06160621717572212 2023-01-21 11:22:33.723451: step: 836/529, loss: 0.039519503712654114 2023-01-21 11:22:34.843810: step: 840/529, loss: 0.04864306375384331 2023-01-21 11:22:35.956258: step: 844/529, loss: 0.006793022155761719 2023-01-21 11:22:37.105481: step: 848/529, loss: 0.028392599895596504 2023-01-21 11:22:38.257678: step: 852/529, loss: 0.04839162901043892 2023-01-21 11:22:39.385059: step: 856/529, loss: 0.025597859174013138 2023-01-21 11:22:40.547362: step: 860/529, loss: 0.011329174041748047 2023-01-21 11:22:41.691305: step: 864/529, loss: 0.05792341008782387 2023-01-21 11:22:42.867461: step: 868/529, loss: 0.0001806020736694336 2023-01-21 11:22:43.985789: step: 872/529, loss: 0.00564651470631361 2023-01-21 11:22:45.138441: step: 876/529, loss: 0.0024355887435376644 2023-01-21 11:22:46.249611: step: 880/529, loss: 0.003887748811393976 2023-01-21 11:22:47.387664: step: 884/529, loss: 0.041213274002075195 2023-01-21 11:22:48.495017: step: 888/529, loss: 0.0001218795805471018 2023-01-21 11:22:49.619501: step: 892/529, loss: 0.02173767238855362 2023-01-21 11:22:50.762892: step: 896/529, loss: 0.00775833148509264 2023-01-21 11:22:51.915022: step: 900/529, loss: 0.01319589652121067 2023-01-21 11:22:53.042002: step: 904/529, loss: 0.08747310936450958 2023-01-21 11:22:54.180732: step: 908/529, loss: 0.04705953598022461 2023-01-21 11:22:55.285750: step: 912/529, loss: 0.00038928986759856343 2023-01-21 11:22:56.426571: step: 916/529, loss: 0.0011153221130371094 2023-01-21 11:22:57.606845: step: 920/529, loss: 0.02843475341796875 2023-01-21 11:22:58.746121: step: 924/529, loss: 0.0027133941184729338 2023-01-21 11:22:59.886509: step: 928/529, loss: 0.003977918531745672 2023-01-21 11:23:00.995873: step: 932/529, loss: 0.014071177691221237 2023-01-21 11:23:02.149127: step: 936/529, loss: 0.00036373137845657766 2023-01-21 11:23:03.290652: step: 940/529, loss: 0.07526607811450958 2023-01-21 11:23:04.464317: step: 944/529, loss: 0.06381382793188095 2023-01-21 11:23:05.611416: step: 948/529, loss: 0.022525405511260033 2023-01-21 11:23:06.757603: step: 952/529, loss: 0.022205352783203125 2023-01-21 11:23:07.927644: step: 956/529, loss: 0.002408695174381137 2023-01-21 11:23:09.065076: step: 960/529, loss: 0.0003517150762490928 2023-01-21 11:23:10.214215: step: 964/529, loss: 0.040834810584783554 2023-01-21 11:23:11.337387: step: 968/529, loss: 0.0012203216319903731 2023-01-21 11:23:12.452530: step: 972/529, loss: 0.0014957428211346269 2023-01-21 11:23:13.622272: step: 976/529, loss: 0.03419971093535423 2023-01-21 11:23:14.760098: step: 980/529, loss: 0.013865947723388672 2023-01-21 11:23:15.877690: step: 984/529, loss: 0.020564651116728783 2023-01-21 11:23:16.989173: step: 988/529, loss: 0.09808731079101562 2023-01-21 11:23:18.143882: step: 992/529, loss: 0.06164246052503586 2023-01-21 11:23:19.283102: step: 996/529, loss: 0.02040834352374077 2023-01-21 11:23:20.393347: step: 1000/529, loss: 0.0250641368329525 2023-01-21 11:23:21.525960: step: 1004/529, loss: 0.008885860443115234 2023-01-21 11:23:22.677951: step: 1008/529, loss: 0.022706128656864166 2023-01-21 11:23:23.819249: step: 1012/529, loss: 0.0010701239807531238 2023-01-21 11:23:24.949214: step: 1016/529, loss: 0.01834726333618164 2023-01-21 11:23:26.088029: step: 1020/529, loss: 0.02515435218811035 2023-01-21 11:23:27.242835: step: 1024/529, loss: 0.04207124933600426 2023-01-21 11:23:28.359572: step: 1028/529, loss: 0.006293106358498335 2023-01-21 11:23:29.521094: step: 1032/529, loss: 0.005190277472138405 2023-01-21 11:23:30.650309: step: 1036/529, loss: 0.0027791978791356087 2023-01-21 11:23:31.778253: step: 1040/529, loss: 0.0037404061295092106 2023-01-21 11:23:32.927042: step: 1044/529, loss: 0.016967011615633965 2023-01-21 11:23:34.072665: step: 1048/529, loss: 0.0010066032409667969 2023-01-21 11:23:35.220796: step: 1052/529, loss: 0.0013590813614428043 2023-01-21 11:23:36.385512: step: 1056/529, loss: 0.028480149805545807 2023-01-21 11:23:37.488254: step: 1060/529, loss: 0.0007292270893231034 2023-01-21 11:23:38.619135: step: 1064/529, loss: 0.021309280768036842 2023-01-21 11:23:39.774785: step: 1068/529, loss: 0.008470727130770683 2023-01-21 11:23:41.010993: step: 1072/529, loss: 0.010136699303984642 2023-01-21 11:23:42.144657: step: 1076/529, loss: 0.013977620750665665 2023-01-21 11:23:43.294125: step: 1080/529, loss: 0.028240211308002472 2023-01-21 11:23:44.415615: step: 1084/529, loss: 0.006890487857162952 2023-01-21 11:23:45.564208: step: 1088/529, loss: 0.0545470267534256 2023-01-21 11:23:46.742722: step: 1092/529, loss: 0.033797264099121094 2023-01-21 11:23:47.861485: step: 1096/529, loss: 0.025826359167695045 2023-01-21 11:23:49.001636: step: 1100/529, loss: 0.02433023601770401 2023-01-21 11:23:50.134927: step: 1104/529, loss: 0.041450027376413345 2023-01-21 11:23:51.280787: step: 1108/529, loss: 0.02288188971579075 2023-01-21 11:23:52.401983: step: 1112/529, loss: 0.00238380441442132 2023-01-21 11:23:53.552368: step: 1116/529, loss: 1.104986310005188 2023-01-21 11:23:54.706509: step: 1120/529, loss: 0.07848186790943146 2023-01-21 11:23:55.837819: step: 1124/529, loss: 0.0017284393543377519 2023-01-21 11:23:56.971400: step: 1128/529, loss: 0.07091350853443146 2023-01-21 11:23:58.099668: step: 1132/529, loss: 0.07041201740503311 2023-01-21 11:23:59.221179: step: 1136/529, loss: 3.9434431528206915e-05 2023-01-21 11:24:00.341381: step: 1140/529, loss: 0.010570049285888672 2023-01-21 11:24:01.503015: step: 1144/529, loss: 0.049651291221380234 2023-01-21 11:24:02.632399: step: 1148/529, loss: 0.04151783138513565 2023-01-21 11:24:03.779072: step: 1152/529, loss: 0.005554008297622204 2023-01-21 11:24:04.888553: step: 1156/529, loss: 0.03578986972570419 2023-01-21 11:24:06.039659: step: 1160/529, loss: 0.01227722130715847 2023-01-21 11:24:07.167636: step: 1164/529, loss: 0.02663097344338894 2023-01-21 11:24:08.310280: step: 1168/529, loss: 0.0036211013793945312 2023-01-21 11:24:09.446677: step: 1172/529, loss: 0.20060434937477112 2023-01-21 11:24:10.583541: step: 1176/529, loss: 0.03645477443933487 2023-01-21 11:24:11.744174: step: 1180/529, loss: 0.0031021118629723787 2023-01-21 11:24:12.903123: step: 1184/529, loss: 0.05019417032599449 2023-01-21 11:24:14.046589: step: 1188/529, loss: 0.01927957497537136 2023-01-21 11:24:15.191458: step: 1192/529, loss: 0.0033814432099461555 2023-01-21 11:24:16.323096: step: 1196/529, loss: 0.00040349960909225047 2023-01-21 11:24:17.464284: step: 1200/529, loss: 0.02581195905804634 2023-01-21 11:24:18.601920: step: 1204/529, loss: 0.28841090202331543 2023-01-21 11:24:19.750210: step: 1208/529, loss: 0.005242729559540749 2023-01-21 11:24:20.886736: step: 1212/529, loss: 0.00017042159743141383 2023-01-21 11:24:22.024550: step: 1216/529, loss: 0.009722709655761719 2023-01-21 11:24:23.151331: step: 1220/529, loss: 0.0021661759819835424 2023-01-21 11:24:24.329591: step: 1224/529, loss: 0.03922243043780327 2023-01-21 11:24:25.459369: step: 1228/529, loss: 0.006839179899543524 2023-01-21 11:24:26.603083: step: 1232/529, loss: 0.0016073227161541581 2023-01-21 11:24:27.770436: step: 1236/529, loss: 0.010824967175722122 2023-01-21 11:24:28.914619: step: 1240/529, loss: 0.03457050397992134 2023-01-21 11:24:30.071145: step: 1244/529, loss: 0.02587719075381756 2023-01-21 11:24:31.224606: step: 1248/529, loss: 0.02059345319867134 2023-01-21 11:24:32.385969: step: 1252/529, loss: 0.3758808672428131 2023-01-21 11:24:33.535419: step: 1256/529, loss: 0.057457152754068375 2023-01-21 11:24:34.700837: step: 1260/529, loss: 0.019969655200839043 2023-01-21 11:24:35.836173: step: 1264/529, loss: 0.0007939815404824913 2023-01-21 11:24:36.981027: step: 1268/529, loss: 0.024445246905088425 2023-01-21 11:24:38.115198: step: 1272/529, loss: 0.008227348327636719 2023-01-21 11:24:39.285086: step: 1276/529, loss: 0.02775583229959011 2023-01-21 11:24:40.449498: step: 1280/529, loss: 0.037984464317560196 2023-01-21 11:24:41.598875: step: 1284/529, loss: 0.0016654968494549394 2023-01-21 11:24:42.720803: step: 1288/529, loss: 0.0008003234979696572 2023-01-21 11:24:43.857750: step: 1292/529, loss: 0.006765366066247225 2023-01-21 11:24:44.992128: step: 1296/529, loss: 0.04855833202600479 2023-01-21 11:24:46.129738: step: 1300/529, loss: 0.020854951813817024 2023-01-21 11:24:47.307739: step: 1304/529, loss: 0.03893337398767471 2023-01-21 11:24:48.429782: step: 1308/529, loss: 0.005517005920410156 2023-01-21 11:24:49.566074: step: 1312/529, loss: 0.0022008896339684725 2023-01-21 11:24:50.746884: step: 1316/529, loss: 0.004822349641472101 2023-01-21 11:24:51.906038: step: 1320/529, loss: 0.009962749667465687 2023-01-21 11:24:53.093062: step: 1324/529, loss: 0.01456222590059042 2023-01-21 11:24:54.205116: step: 1328/529, loss: 0.03274068981409073 2023-01-21 11:24:55.347652: step: 1332/529, loss: 0.010128498077392578 2023-01-21 11:24:56.512846: step: 1336/529, loss: 0.28467294573783875 2023-01-21 11:24:57.674337: step: 1340/529, loss: 0.016507340595126152 2023-01-21 11:24:58.805882: step: 1344/529, loss: 0.03522081300616264 2023-01-21 11:24:59.947060: step: 1348/529, loss: 0.00025696755619719625 2023-01-21 11:25:01.075933: step: 1352/529, loss: 0.0002065070002572611 2023-01-21 11:25:02.240565: step: 1356/529, loss: 0.03253822401165962 2023-01-21 11:25:03.357384: step: 1360/529, loss: 0.020462609827518463 2023-01-21 11:25:04.479798: step: 1364/529, loss: 0.09718303382396698 2023-01-21 11:25:05.596595: step: 1368/529, loss: 0.026201914995908737 2023-01-21 11:25:06.738608: step: 1372/529, loss: 0.001691627549007535 2023-01-21 11:25:07.837686: step: 1376/529, loss: 0.0019495964515954256 2023-01-21 11:25:08.966987: step: 1380/529, loss: 0.0007335662958212197 2023-01-21 11:25:10.100255: step: 1384/529, loss: 0.052747536450624466 2023-01-21 11:25:11.242683: step: 1388/529, loss: 0.06004119664430618 2023-01-21 11:25:12.392666: step: 1392/529, loss: 0.013131809420883656 2023-01-21 11:25:13.518637: step: 1396/529, loss: 0.002691173693165183 2023-01-21 11:25:14.633471: step: 1400/529, loss: 0.015098286792635918 2023-01-21 11:25:15.759493: step: 1404/529, loss: 0.03063373640179634 2023-01-21 11:25:16.885936: step: 1408/529, loss: 0.0003714561462402344 2023-01-21 11:25:18.041126: step: 1412/529, loss: 0.00010094643221236765 2023-01-21 11:25:19.171468: step: 1416/529, loss: 0.0002612113894429058 2023-01-21 11:25:20.321014: step: 1420/529, loss: 0.010074520483613014 2023-01-21 11:25:21.418572: step: 1424/529, loss: 0.008118915371596813 2023-01-21 11:25:22.591883: step: 1428/529, loss: 0.0013757705455645919 2023-01-21 11:25:23.735671: step: 1432/529, loss: 0.0011120797134935856 2023-01-21 11:25:24.862066: step: 1436/529, loss: 0.0006565094226971269 2023-01-21 11:25:25.980879: step: 1440/529, loss: 0.021738719195127487 2023-01-21 11:25:27.096453: step: 1444/529, loss: 0.0002754211309365928 2023-01-21 11:25:28.220253: step: 1448/529, loss: 0.0031210898887366056 2023-01-21 11:25:29.353556: step: 1452/529, loss: 0.06607885658740997 2023-01-21 11:25:30.484762: step: 1456/529, loss: 0.0023772241547703743 2023-01-21 11:25:31.658759: step: 1460/529, loss: 0.0004784107441082597 2023-01-21 11:25:32.780608: step: 1464/529, loss: 0.12763424217700958 2023-01-21 11:25:33.918153: step: 1468/529, loss: 0.0265871062874794 2023-01-21 11:25:35.054217: step: 1472/529, loss: 0.02527751959860325 2023-01-21 11:25:36.185657: step: 1476/529, loss: 0.03329315036535263 2023-01-21 11:25:37.348504: step: 1480/529, loss: 0.10500279068946838 2023-01-21 11:25:38.494737: step: 1484/529, loss: 0.009329795837402344 2023-01-21 11:25:39.617846: step: 1488/529, loss: 0.017876578494906425 2023-01-21 11:25:40.755463: step: 1492/529, loss: 0.0744309350848198 2023-01-21 11:25:41.935957: step: 1496/529, loss: 0.10974683612585068 2023-01-21 11:25:43.076328: step: 1500/529, loss: 0.04722023010253906 2023-01-21 11:25:44.238509: step: 1504/529, loss: 0.002580833388492465 2023-01-21 11:25:45.380545: step: 1508/529, loss: 0.003114557359367609 2023-01-21 11:25:46.502793: step: 1512/529, loss: 0.2487998902797699 2023-01-21 11:25:47.680891: step: 1516/529, loss: 0.7010349035263062 2023-01-21 11:25:48.840712: step: 1520/529, loss: 0.018754005432128906 2023-01-21 11:25:49.961872: step: 1524/529, loss: 0.02645101398229599 2023-01-21 11:25:51.120732: step: 1528/529, loss: 0.01490316353738308 2023-01-21 11:25:52.264827: step: 1532/529, loss: 0.003131866455078125 2023-01-21 11:25:53.404792: step: 1536/529, loss: 0.027715682983398438 2023-01-21 11:25:54.528447: step: 1540/529, loss: 0.0011343001388013363 2023-01-21 11:25:55.665122: step: 1544/529, loss: 0.022390365600585938 2023-01-21 11:25:56.824931: step: 1548/529, loss: 0.01876678504049778 2023-01-21 11:25:57.963480: step: 1552/529, loss: 0.004564475733786821 2023-01-21 11:25:59.095227: step: 1556/529, loss: 0.024617576971650124 2023-01-21 11:26:00.274733: step: 1560/529, loss: 0.00028591154841706157 2023-01-21 11:26:01.411564: step: 1564/529, loss: 0.013111305423080921 2023-01-21 11:26:02.553755: step: 1568/529, loss: 0.008581352420151234 2023-01-21 11:26:03.672851: step: 1572/529, loss: 0.014995861798524857 2023-01-21 11:26:04.786312: step: 1576/529, loss: 0.04624929279088974 2023-01-21 11:26:05.934627: step: 1580/529, loss: 0.06347541511058807 2023-01-21 11:26:07.095738: step: 1584/529, loss: 0.013006210327148438 2023-01-21 11:26:08.233307: step: 1588/529, loss: 0.0006497383583337069 2023-01-21 11:26:09.384041: step: 1592/529, loss: 0.0072428700514137745 2023-01-21 11:26:10.530984: step: 1596/529, loss: 0.002002525143325329 2023-01-21 11:26:11.647541: step: 1600/529, loss: 0.0663658007979393 2023-01-21 11:26:12.786954: step: 1604/529, loss: 0.00048828125 2023-01-21 11:26:13.906640: step: 1608/529, loss: 0.025282716378569603 2023-01-21 11:26:15.085026: step: 1612/529, loss: 0.023212242871522903 2023-01-21 11:26:16.243243: step: 1616/529, loss: 0.01238784845918417 2023-01-21 11:26:17.397266: step: 1620/529, loss: 0.06938133388757706 2023-01-21 11:26:18.547670: step: 1624/529, loss: 0.0457189567387104 2023-01-21 11:26:19.679232: step: 1628/529, loss: 0.00818624533712864 2023-01-21 11:26:20.805122: step: 1632/529, loss: 0.04999618977308273 2023-01-21 11:26:21.939235: step: 1636/529, loss: 0.0023382187355309725 2023-01-21 11:26:23.058879: step: 1640/529, loss: 0.014842129312455654 2023-01-21 11:26:24.208968: step: 1644/529, loss: 0.008029174990952015 2023-01-21 11:26:25.318602: step: 1648/529, loss: 0.008969116024672985 2023-01-21 11:26:26.473813: step: 1652/529, loss: 0.03575191646814346 2023-01-21 11:26:27.637402: step: 1656/529, loss: 0.037931445986032486 2023-01-21 11:26:28.758181: step: 1660/529, loss: 0.029606247320771217 2023-01-21 11:26:29.906268: step: 1664/529, loss: 0.056976318359375 2023-01-21 11:26:31.018776: step: 1668/529, loss: 0.003731155302375555 2023-01-21 11:26:32.176126: step: 1672/529, loss: 0.049599841237068176 2023-01-21 11:26:33.324831: step: 1676/529, loss: 0.020070362836122513 2023-01-21 11:26:34.487267: step: 1680/529, loss: 0.056424617767333984 2023-01-21 11:26:35.638946: step: 1684/529, loss: 0.04691801220178604 2023-01-21 11:26:36.802399: step: 1688/529, loss: 0.0008308410178869963 2023-01-21 11:26:37.937622: step: 1692/529, loss: 0.000640916870906949 2023-01-21 11:26:39.092555: step: 1696/529, loss: 0.010032462887465954 2023-01-21 11:26:40.233631: step: 1700/529, loss: 0.013447379693388939 2023-01-21 11:26:41.375531: step: 1704/529, loss: 0.036274053156375885 2023-01-21 11:26:42.503735: step: 1708/529, loss: 0.5931419134140015 2023-01-21 11:26:43.634859: step: 1712/529, loss: 0.032277777791023254 2023-01-21 11:26:44.740274: step: 1716/529, loss: 0.009038066491484642 2023-01-21 11:26:45.888006: step: 1720/529, loss: 0.010699844919145107 2023-01-21 11:26:47.036759: step: 1724/529, loss: 0.007488823030143976 2023-01-21 11:26:48.174987: step: 1728/529, loss: 0.005389595404267311 2023-01-21 11:26:49.306653: step: 1732/529, loss: 0.023988818749785423 2023-01-21 11:26:50.452384: step: 1736/529, loss: 0.0010326385963708162 2023-01-21 11:26:51.589623: step: 1740/529, loss: 0.1594664603471756 2023-01-21 11:26:52.713712: step: 1744/529, loss: 0.052855681627988815 2023-01-21 11:26:53.845939: step: 1748/529, loss: 0.006522479932755232 2023-01-21 11:26:54.964617: step: 1752/529, loss: 0.013235759921371937 2023-01-21 11:26:56.063218: step: 1756/529, loss: 0.02443857304751873 2023-01-21 11:26:57.172854: step: 1760/529, loss: 0.008518275804817677 2023-01-21 11:26:58.290891: step: 1764/529, loss: 0.004141808021813631 2023-01-21 11:26:59.442452: step: 1768/529, loss: 0.047465730458498 2023-01-21 11:27:00.561702: step: 1772/529, loss: 0.02482481114566326 2023-01-21 11:27:01.730427: step: 1776/529, loss: 0.023076439276337624 2023-01-21 11:27:02.856847: step: 1780/529, loss: 0.02016582526266575 2023-01-21 11:27:04.024137: step: 1784/529, loss: 0.11726856231689453 2023-01-21 11:27:05.168500: step: 1788/529, loss: 0.00041370390681549907 2023-01-21 11:27:06.315887: step: 1792/529, loss: 0.01807870902121067 2023-01-21 11:27:07.450969: step: 1796/529, loss: 0.003523159073665738 2023-01-21 11:27:08.660199: step: 1800/529, loss: 0.024566078558564186 2023-01-21 11:27:09.787419: step: 1804/529, loss: 0.0014314651489257812 2023-01-21 11:27:10.927173: step: 1808/529, loss: 0.09362497925758362 2023-01-21 11:27:12.101705: step: 1812/529, loss: 0.004131317138671875 2023-01-21 11:27:13.254305: step: 1816/529, loss: 0.013601303100585938 2023-01-21 11:27:14.381485: step: 1820/529, loss: 0.028903769329190254 2023-01-21 11:27:15.516854: step: 1824/529, loss: 0.05010054260492325 2023-01-21 11:27:16.671779: step: 1828/529, loss: 0.012841988354921341 2023-01-21 11:27:17.814689: step: 1832/529, loss: 0.04880065843462944 2023-01-21 11:27:18.933250: step: 1836/529, loss: 0.00017623903113417327 2023-01-21 11:27:20.063517: step: 1840/529, loss: 0.037212345749139786 2023-01-21 11:27:21.191895: step: 1844/529, loss: 0.005331802647560835 2023-01-21 11:27:22.337347: step: 1848/529, loss: 0.011832189746201038 2023-01-21 11:27:23.461770: step: 1852/529, loss: 0.19008304178714752 2023-01-21 11:27:24.598003: step: 1856/529, loss: 0.033326104283332825 2023-01-21 11:27:25.744924: step: 1860/529, loss: 0.0009250640869140625 2023-01-21 11:27:26.904450: step: 1864/529, loss: 0.007072639651596546 2023-01-21 11:27:28.037594: step: 1868/529, loss: 0.013271331787109375 2023-01-21 11:27:29.167926: step: 1872/529, loss: 0.018895339220762253 2023-01-21 11:27:30.324296: step: 1876/529, loss: 0.010406112298369408 2023-01-21 11:27:31.472663: step: 1880/529, loss: 0.018147850409150124 2023-01-21 11:27:32.610826: step: 1884/529, loss: 0.00739631662145257 2023-01-21 11:27:33.766134: step: 1888/529, loss: 0.004591751378029585 2023-01-21 11:27:34.914466: step: 1892/529, loss: 0.0007364749908447266 2023-01-21 11:27:36.040113: step: 1896/529, loss: 0.019927598536014557 2023-01-21 11:27:37.209151: step: 1900/529, loss: 0.005155086982995272 2023-01-21 11:27:38.417142: step: 1904/529, loss: 0.013299942016601562 2023-01-21 11:27:39.594037: step: 1908/529, loss: 0.038001637905836105 2023-01-21 11:27:40.751615: step: 1912/529, loss: 0.046645261347293854 2023-01-21 11:27:41.887544: step: 1916/529, loss: 0.0026298523880541325 2023-01-21 11:27:43.058599: step: 1920/529, loss: 0.0358491912484169 2023-01-21 11:27:44.209154: step: 1924/529, loss: 0.05771274492144585 2023-01-21 11:27:45.344161: step: 1928/529, loss: 7.724761962890625e-05 2023-01-21 11:27:46.476475: step: 1932/529, loss: 0.0052055357955396175 2023-01-21 11:27:47.631539: step: 1936/529, loss: 0.0002552032528910786 2023-01-21 11:27:48.758746: step: 1940/529, loss: 0.0004426002560649067 2023-01-21 11:27:49.888802: step: 1944/529, loss: 0.07954387366771698 2023-01-21 11:27:51.034158: step: 1948/529, loss: 0.010173988528549671 2023-01-21 11:27:52.155756: step: 1952/529, loss: 0.10088739544153214 2023-01-21 11:27:53.306751: step: 1956/529, loss: 0.034436989575624466 2023-01-21 11:27:54.456229: step: 1960/529, loss: 0.006924915127456188 2023-01-21 11:27:55.582230: step: 1964/529, loss: 0.016892993822693825 2023-01-21 11:27:56.780563: step: 1968/529, loss: 0.16579914093017578 2023-01-21 11:27:57.951395: step: 1972/529, loss: 0.04669789969921112 2023-01-21 11:27:59.070748: step: 1976/529, loss: 0.028626490384340286 2023-01-21 11:28:00.239777: step: 1980/529, loss: 0.008643150329589844 2023-01-21 11:28:01.377455: step: 1984/529, loss: 0.003363037249073386 2023-01-21 11:28:02.524246: step: 1988/529, loss: 0.03058185614645481 2023-01-21 11:28:03.650620: step: 1992/529, loss: 0.09120196104049683 2023-01-21 11:28:04.791705: step: 1996/529, loss: 0.013782501220703125 2023-01-21 11:28:05.939458: step: 2000/529, loss: 0.06439408659934998 2023-01-21 11:28:07.116358: step: 2004/529, loss: 0.02214536815881729 2023-01-21 11:28:08.270307: step: 2008/529, loss: 0.04889526218175888 2023-01-21 11:28:09.454146: step: 2012/529, loss: 0.0643530860543251 2023-01-21 11:28:10.587435: step: 2016/529, loss: 0.006801176350563765 2023-01-21 11:28:11.738521: step: 2020/529, loss: 0.005294132046401501 2023-01-21 11:28:12.887410: step: 2024/529, loss: 0.005389213562011719 2023-01-21 11:28:14.009371: step: 2028/529, loss: 0.024183249101042747 2023-01-21 11:28:15.173913: step: 2032/529, loss: 0.016169453039765358 2023-01-21 11:28:16.346967: step: 2036/529, loss: 0.01381683349609375 2023-01-21 11:28:17.485103: step: 2040/529, loss: 0.000222945207497105 2023-01-21 11:28:18.629839: step: 2044/529, loss: 0.019466400146484375 2023-01-21 11:28:19.783493: step: 2048/529, loss: 0.008198071271181107 2023-01-21 11:28:20.909627: step: 2052/529, loss: 0.03301544114947319 2023-01-21 11:28:22.048818: step: 2056/529, loss: 0.015151405707001686 2023-01-21 11:28:23.184803: step: 2060/529, loss: 0.07522258907556534 2023-01-21 11:28:24.334204: step: 2064/529, loss: 0.0515805259346962 2023-01-21 11:28:25.495662: step: 2068/529, loss: 0.03762846440076828 2023-01-21 11:28:26.618405: step: 2072/529, loss: 0.0026432513259351254 2023-01-21 11:28:27.746611: step: 2076/529, loss: 0.028045939281582832 2023-01-21 11:28:28.905814: step: 2080/529, loss: 0.0003803253057412803 2023-01-21 11:28:30.043218: step: 2084/529, loss: 0.004448986612260342 2023-01-21 11:28:31.197646: step: 2088/529, loss: 0.005446815863251686 2023-01-21 11:28:32.336167: step: 2092/529, loss: 0.022199535742402077 2023-01-21 11:28:33.456557: step: 2096/529, loss: 0.8220190405845642 2023-01-21 11:28:34.583996: step: 2100/529, loss: 0.01908578909933567 2023-01-21 11:28:35.733779: step: 2104/529, loss: 0.0033995630219578743 2023-01-21 11:28:36.869925: step: 2108/529, loss: 0.002408313797786832 2023-01-21 11:28:37.999127: step: 2112/529, loss: 0.004783153533935547 2023-01-21 11:28:39.141903: step: 2116/529, loss: 0.0001625060976948589 ================================================== Loss: 0.035 -------------------- Dev: {'event': {'p': 0.6213592233009708, 'r': 0.7669773635153129, 'f1': 0.6865315852205005}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Test: {'event': {'p': 0.6436274509803922, 'r': 0.7607184241019699, 'f1': 0.6972915560276155}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Chinese: {'event': {'p': 0.6024096385542169, 'r': 0.9259259259259259, 'f1': 0.7299270072992702}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Korean: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Russian: {'event': {'p': 0.4444444444444444, 'r': 0.5555555555555556, 'f1': 0.49382716049382713}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** 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 11:29:20.579412: step: 4/529, loss: 0.00980291422456503 2023-01-21 11:29:21.706845: step: 8/529, loss: 0.03227834776043892 2023-01-21 11:29:22.828966: step: 12/529, loss: 0.004454899113625288 2023-01-21 11:29:23.952515: step: 16/529, loss: 0.0016155242919921875 2023-01-21 11:29:25.087232: step: 20/529, loss: 0.005145454313606024 2023-01-21 11:29:26.213510: step: 24/529, loss: 0.020261382684111595 2023-01-21 11:29:27.346776: step: 28/529, loss: 0.0001310348598053679 2023-01-21 11:29:28.473958: step: 32/529, loss: 0.06226807087659836 2023-01-21 11:29:29.617908: step: 36/529, loss: 0.008191203698515892 2023-01-21 11:29:30.758300: step: 40/529, loss: 0.002740478375926614 2023-01-21 11:29:31.880903: step: 44/529, loss: 0.024071503430604935 2023-01-21 11:29:33.038715: step: 48/529, loss: 0.014621258713304996 2023-01-21 11:29:34.189479: step: 52/529, loss: 0.03954477235674858 2023-01-21 11:29:35.319061: step: 56/529, loss: 0.3380860686302185 2023-01-21 11:29:36.452684: step: 60/529, loss: 0.0024174691643565893 2023-01-21 11:29:37.586719: step: 64/529, loss: 0.08009643852710724 2023-01-21 11:29:38.702967: step: 68/529, loss: 0.005244204308837652 2023-01-21 11:29:39.870292: step: 72/529, loss: 0.006070995237678289 2023-01-21 11:29:41.006890: step: 76/529, loss: 0.0014183043967932463 2023-01-21 11:29:42.139153: step: 80/529, loss: 0.005898761563003063 2023-01-21 11:29:43.292725: step: 84/529, loss: 0.00016880035400390625 2023-01-21 11:29:44.421632: step: 88/529, loss: 0.012494374066591263 2023-01-21 11:29:45.550659: step: 92/529, loss: 0.18097858130931854 2023-01-21 11:29:46.673049: step: 96/529, loss: 0.04884052276611328 2023-01-21 11:29:47.819841: step: 100/529, loss: 0.004426002502441406 2023-01-21 11:29:48.957983: step: 104/529, loss: 0.01939554139971733 2023-01-21 11:29:50.111618: step: 108/529, loss: 0.022711850702762604 2023-01-21 11:29:51.231867: step: 112/529, loss: 0.003566932864487171 2023-01-21 11:29:52.355559: step: 116/529, loss: 0.012328815646469593 2023-01-21 11:29:53.487644: step: 120/529, loss: 0.07349672168493271 2023-01-21 11:29:54.630511: step: 124/529, loss: 0.02655172534286976 2023-01-21 11:29:55.740879: step: 128/529, loss: 0.000664138759020716 2023-01-21 11:29:56.892567: step: 132/529, loss: 0.02839946746826172 2023-01-21 11:29:58.027798: step: 136/529, loss: 0.0016240120166912675 2023-01-21 11:29:59.150972: step: 140/529, loss: 0.07888088375329971 2023-01-21 11:30:00.280508: step: 144/529, loss: 0.0017282485496252775 2023-01-21 11:30:01.433178: step: 148/529, loss: 0.0030162811744958162 2023-01-21 11:30:02.570139: step: 152/529, loss: 0.026589393615722656 2023-01-21 11:30:03.710981: step: 156/529, loss: 0.06963425129652023 2023-01-21 11:30:04.834266: step: 160/529, loss: 0.001889991806820035 2023-01-21 11:30:05.951962: step: 164/529, loss: 0.0028238771483302116 2023-01-21 11:30:07.100590: step: 168/529, loss: 0.007329368963837624 2023-01-21 11:30:08.234865: step: 172/529, loss: 0.01625976525247097 2023-01-21 11:30:09.374601: step: 176/529, loss: 0.02523813210427761 2023-01-21 11:30:10.520468: step: 180/529, loss: 0.0008698463207110763 2023-01-21 11:30:11.618202: step: 184/529, loss: 0.003505134489387274 2023-01-21 11:30:12.735172: step: 188/529, loss: 0.09457750618457794 2023-01-21 11:30:13.880173: step: 192/529, loss: 0.03552980348467827 2023-01-21 11:30:15.009849: step: 196/529, loss: 0.013662147335708141 2023-01-21 11:30:16.143465: step: 200/529, loss: 0.07314208149909973 2023-01-21 11:30:17.273134: step: 204/529, loss: 0.002875042147934437 2023-01-21 11:30:18.415957: step: 208/529, loss: 0.017232514917850494 2023-01-21 11:30:19.536108: step: 212/529, loss: 0.00427322369068861 2023-01-21 11:30:20.664712: step: 216/529, loss: 0.04822085052728653 2023-01-21 11:30:21.809121: step: 220/529, loss: 9.384155418956652e-05 2023-01-21 11:30:22.947522: step: 224/529, loss: 0.006916237063705921 2023-01-21 11:30:24.082924: step: 228/529, loss: 0.001729774521663785 2023-01-21 11:30:25.239252: step: 232/529, loss: 0.00023651123046875 2023-01-21 11:30:26.387569: step: 236/529, loss: 0.004020023625344038 2023-01-21 11:30:27.522999: step: 240/529, loss: 0.004679298494011164 2023-01-21 11:30:28.642846: step: 244/529, loss: 4.0626528061693534e-05 2023-01-21 11:30:29.799900: step: 248/529, loss: 0.09011640399694443 2023-01-21 11:30:30.917598: step: 252/529, loss: 0.05582733079791069 2023-01-21 11:30:32.042143: step: 256/529, loss: 0.02553529664874077 2023-01-21 11:30:33.203954: step: 260/529, loss: 0.029871942475438118 2023-01-21 11:30:34.313935: step: 264/529, loss: 0.001834773924201727 2023-01-21 11:30:35.459985: step: 268/529, loss: 0.041945457458496094 2023-01-21 11:30:36.580497: step: 272/529, loss: 0.026454783976078033 2023-01-21 11:30:37.707149: step: 276/529, loss: 0.04946298897266388 2023-01-21 11:30:38.865717: step: 280/529, loss: 0.0064636231400072575 2023-01-21 11:30:40.009355: step: 284/529, loss: 0.05516119301319122 2023-01-21 11:30:41.128506: step: 288/529, loss: 0.018744850531220436 2023-01-21 11:30:42.249770: step: 292/529, loss: 0.0003684043767862022 2023-01-21 11:30:43.372301: step: 296/529, loss: 0.00012865067401435226 2023-01-21 11:30:44.521876: step: 300/529, loss: 0.04284706339240074 2023-01-21 11:30:45.639977: step: 304/529, loss: 0.02588510513305664 2023-01-21 11:30:46.795690: step: 308/529, loss: 0.002304458525031805 2023-01-21 11:30:47.940765: step: 312/529, loss: 0.044738009572029114 2023-01-21 11:30:49.047104: step: 316/529, loss: 0.012210655026137829 2023-01-21 11:30:50.200237: step: 320/529, loss: 0.019185829907655716 2023-01-21 11:30:51.322646: step: 324/529, loss: 0.04814085736870766 2023-01-21 11:30:52.460708: step: 328/529, loss: 0.011650276370346546 2023-01-21 11:30:53.603957: step: 332/529, loss: 0.021657563745975494 2023-01-21 11:30:54.754493: step: 336/529, loss: 0.00460438709706068 2023-01-21 11:30:55.881131: step: 340/529, loss: 0.003719830419868231 2023-01-21 11:30:57.036365: step: 344/529, loss: 0.04912605881690979 2023-01-21 11:30:58.172107: step: 348/529, loss: 0.06083059683442116 2023-01-21 11:30:59.323170: step: 352/529, loss: 0.04420957714319229 2023-01-21 11:31:00.461447: step: 356/529, loss: 0.0010129929287359118 2023-01-21 11:31:01.584220: step: 360/529, loss: 0.0008832931634970009 2023-01-21 11:31:02.724648: step: 364/529, loss: 0.002809906145557761 2023-01-21 11:31:03.882775: step: 368/529, loss: 0.07935362309217453 2023-01-21 11:31:05.029213: step: 372/529, loss: 0.0003561020130291581 2023-01-21 11:31:06.171738: step: 376/529, loss: 0.030472850427031517 2023-01-21 11:31:07.315432: step: 380/529, loss: 0.004821205511689186 2023-01-21 11:31:08.425813: step: 384/529, loss: 0.0731317549943924 2023-01-21 11:31:09.553499: step: 388/529, loss: 0.0010470390552654862 2023-01-21 11:31:10.711454: step: 392/529, loss: 0.015868186950683594 2023-01-21 11:31:11.885199: step: 396/529, loss: 0.00107488629873842 2023-01-21 11:31:13.005781: step: 400/529, loss: 0.012009715661406517 2023-01-21 11:31:14.151123: step: 404/529, loss: 0.0049270628951489925 2023-01-21 11:31:15.293248: step: 408/529, loss: 0.0009677887428551912 2023-01-21 11:31:16.447399: step: 412/529, loss: 0.006442260928452015 2023-01-21 11:31:17.556169: step: 416/529, loss: 0.03438606113195419 2023-01-21 11:31:18.694240: step: 420/529, loss: 0.0003637791087385267 2023-01-21 11:31:19.784716: step: 424/529, loss: 0.00032358168391510844 2023-01-21 11:31:20.929800: step: 428/529, loss: 0.005825805943459272 2023-01-21 11:31:22.035481: step: 432/529, loss: 0.0033159255981445312 2023-01-21 11:31:23.212191: step: 436/529, loss: 0.000827789306640625 2023-01-21 11:31:24.357486: step: 440/529, loss: 0.019193649291992188 2023-01-21 11:31:25.533364: step: 444/529, loss: 0.05843019485473633 2023-01-21 11:31:26.655134: step: 448/529, loss: 0.02928762510418892 2023-01-21 11:31:27.809661: step: 452/529, loss: 0.03692684322595596 2023-01-21 11:31:28.972565: step: 456/529, loss: 0.007652568630874157 2023-01-21 11:31:30.103989: step: 460/529, loss: 0.0006719589000567794 2023-01-21 11:31:31.246879: step: 464/529, loss: 0.014843559823930264 2023-01-21 11:31:32.390998: step: 468/529, loss: 0.029177237302064896 2023-01-21 11:31:33.532872: step: 472/529, loss: 0.021376824006438255 2023-01-21 11:31:34.662517: step: 476/529, loss: 0.06917037814855576 2023-01-21 11:31:35.799179: step: 480/529, loss: 0.023467255756258965 2023-01-21 11:31:36.929307: step: 484/529, loss: 0.022478677332401276 2023-01-21 11:31:38.045604: step: 488/529, loss: 0.0926908552646637 2023-01-21 11:31:39.169471: step: 492/529, loss: 0.0012767792213708162 2023-01-21 11:31:40.301779: step: 496/529, loss: 0.028953934088349342 2023-01-21 11:31:41.468252: step: 500/529, loss: 0.01389245968312025 2023-01-21 11:31:42.651821: step: 504/529, loss: 0.013598919846117496 2023-01-21 11:31:43.823604: step: 508/529, loss: 0.0002747535763774067 2023-01-21 11:31:44.989326: step: 512/529, loss: 0.0010767936473712325 2023-01-21 11:31:46.146159: step: 516/529, loss: 0.0036822319962084293 2023-01-21 11:31:47.293276: step: 520/529, loss: 0.02061338536441326 2023-01-21 11:31:48.426407: step: 524/529, loss: 0.0012369155883789062 2023-01-21 11:31:49.564813: step: 528/529, loss: 0.00026416778564453125 2023-01-21 11:31:50.742771: step: 532/529, loss: 0.0058189393021166325 2023-01-21 11:31:51.898806: step: 536/529, loss: 0.01992187462747097 2023-01-21 11:31:53.016559: step: 540/529, loss: 0.0005779266939498484 2023-01-21 11:31:54.178155: step: 544/529, loss: 0.0041221617721021175 2023-01-21 11:31:55.346784: step: 548/529, loss: 0.00023622512526344508 2023-01-21 11:31:56.487524: step: 552/529, loss: 0.004508304409682751 2023-01-21 11:31:57.620056: step: 556/529, loss: 0.03316173702478409 2023-01-21 11:31:58.771217: step: 560/529, loss: 0.004639816470444202 2023-01-21 11:31:59.907988: step: 564/529, loss: 0.004973125644028187 2023-01-21 11:32:01.029485: step: 568/529, loss: 0.04298343509435654 2023-01-21 11:32:02.190942: step: 572/529, loss: 0.0038782120682299137 2023-01-21 11:32:03.294501: step: 576/529, loss: 0.007598876953125 2023-01-21 11:32:04.429738: step: 580/529, loss: 0.09616661071777344 2023-01-21 11:32:05.548215: step: 584/529, loss: 0.010191250592470169 2023-01-21 11:32:06.694797: step: 588/529, loss: 0.033980607986450195 2023-01-21 11:32:07.832078: step: 592/529, loss: 0.07082367688417435 2023-01-21 11:32:09.000203: step: 596/529, loss: 0.011842155829071999 2023-01-21 11:32:10.147304: step: 600/529, loss: 0.003889370011165738 2023-01-21 11:32:11.288598: step: 604/529, loss: 0.01498708687722683 2023-01-21 11:32:12.420975: step: 608/529, loss: 0.004042625427246094 2023-01-21 11:32:13.583846: step: 612/529, loss: 0.010221290402114391 2023-01-21 11:32:14.705562: step: 616/529, loss: 0.0005509377224370837 2023-01-21 11:32:15.843571: step: 620/529, loss: 0.0004811286926269531 2023-01-21 11:32:16.984404: step: 624/529, loss: 0.01910400390625 2023-01-21 11:32:18.117607: step: 628/529, loss: 0.008157063275575638 2023-01-21 11:32:19.281993: step: 632/529, loss: 0.024146152660250664 2023-01-21 11:32:20.404030: step: 636/529, loss: 0.03137988969683647 2023-01-21 11:32:21.562646: step: 640/529, loss: 0.0002980232238769531 2023-01-21 11:32:22.717145: step: 644/529, loss: 0.0009495735284872353 2023-01-21 11:32:23.864916: step: 648/529, loss: 7.381439354503527e-05 2023-01-21 11:32:25.043902: step: 652/529, loss: 0.030804958194494247 2023-01-21 11:32:26.178132: step: 656/529, loss: 0.001996326493099332 2023-01-21 11:32:27.315087: step: 660/529, loss: 0.05277710035443306 2023-01-21 11:32:28.447145: step: 664/529, loss: 0.043516673147678375 2023-01-21 11:32:29.619095: step: 668/529, loss: 0.052054502069950104 2023-01-21 11:32:30.789421: step: 672/529, loss: 0.04818020015954971 2023-01-21 11:32:31.945891: step: 676/529, loss: 0.001868820283561945 2023-01-21 11:32:33.075040: step: 680/529, loss: 0.0366579070687294 2023-01-21 11:32:34.225759: step: 684/529, loss: 0.0008013725164346397 2023-01-21 11:32:35.374532: step: 688/529, loss: 0.02044367790222168 2023-01-21 11:32:36.493380: step: 692/529, loss: 0.002616023877635598 2023-01-21 11:32:37.640001: step: 696/529, loss: 0.023929595947265625 2023-01-21 11:32:38.777493: step: 700/529, loss: 0.03232889249920845 2023-01-21 11:32:39.889476: step: 704/529, loss: 0.0035856247413903475 2023-01-21 11:32:41.036666: step: 708/529, loss: 0.01963496208190918 2023-01-21 11:32:42.205971: step: 712/529, loss: 0.1837204098701477 2023-01-21 11:32:43.327050: step: 716/529, loss: 0.009389782324433327 2023-01-21 11:32:44.496707: step: 720/529, loss: 0.027189921587705612 2023-01-21 11:32:45.621760: step: 724/529, loss: 0.12962722778320312 2023-01-21 11:32:46.761605: step: 728/529, loss: 0.016514873132109642 2023-01-21 11:32:47.903086: step: 732/529, loss: 9.298324584960938e-05 2023-01-21 11:32:49.031699: step: 736/529, loss: 0.012986088171601295 2023-01-21 11:32:50.172601: step: 740/529, loss: 0.003890013787895441 2023-01-21 11:32:51.328404: step: 744/529, loss: 0.0001510620058979839 2023-01-21 11:32:52.468586: step: 748/529, loss: 1.9741059077205136e-05 2023-01-21 11:32:53.595251: step: 752/529, loss: 0.059157274663448334 2023-01-21 11:32:54.748884: step: 756/529, loss: 0.003543091006577015 2023-01-21 11:32:55.896190: step: 760/529, loss: 0.0038873672019690275 2023-01-21 11:32:57.053526: step: 764/529, loss: 0.13239344954490662 2023-01-21 11:32:58.244185: step: 768/529, loss: 0.014187430962920189 2023-01-21 11:32:59.367923: step: 772/529, loss: 0.011584663763642311 2023-01-21 11:33:00.502627: step: 776/529, loss: 0.009246778674423695 2023-01-21 11:33:01.629597: step: 780/529, loss: 0.033936403691768646 2023-01-21 11:33:02.729935: step: 784/529, loss: 0.0004646301094908267 2023-01-21 11:33:03.864299: step: 788/529, loss: 0.021702958270907402 2023-01-21 11:33:04.993542: step: 792/529, loss: 0.031316231936216354 2023-01-21 11:33:06.148433: step: 796/529, loss: 0.05794792249798775 2023-01-21 11:33:07.295395: step: 800/529, loss: 0.03635444492101669 2023-01-21 11:33:08.439730: step: 804/529, loss: 0.04673881456255913 2023-01-21 11:33:09.624240: step: 808/529, loss: 0.008633327670395374 2023-01-21 11:33:10.740987: step: 812/529, loss: 0.006572866812348366 2023-01-21 11:33:11.920323: step: 816/529, loss: 0.016468429937958717 2023-01-21 11:33:13.039967: step: 820/529, loss: 0.034932807087898254 2023-01-21 11:33:14.162144: step: 824/529, loss: 0.014830934815108776 2023-01-21 11:33:15.302667: step: 828/529, loss: 0.005287742707878351 2023-01-21 11:33:16.438292: step: 832/529, loss: 0.07670764625072479 2023-01-21 11:33:17.563148: step: 836/529, loss: 0.005574226379394531 2023-01-21 11:33:18.697268: step: 840/529, loss: 0.019486570730805397 2023-01-21 11:33:19.847024: step: 844/529, loss: 0.010714912787079811 2023-01-21 11:33:20.989339: step: 848/529, loss: 0.0005819320795126259 2023-01-21 11:33:22.141973: step: 852/529, loss: 0.019542885944247246 2023-01-21 11:33:23.268424: step: 856/529, loss: 0.04361124336719513 2023-01-21 11:33:24.393925: step: 860/529, loss: 0.06918258965015411 2023-01-21 11:33:25.566496: step: 864/529, loss: 0.03521308675408363 2023-01-21 11:33:26.703155: step: 868/529, loss: 0.0021944046020507812 2023-01-21 11:33:27.860873: step: 872/529, loss: 0.03416891023516655 2023-01-21 11:33:28.992690: step: 876/529, loss: 0.009745406918227673 2023-01-21 11:33:30.147045: step: 880/529, loss: 0.00012369155592750758 2023-01-21 11:33:31.301009: step: 884/529, loss: 0.001563263009302318 2023-01-21 11:33:32.433009: step: 888/529, loss: 0.007650995627045631 2023-01-21 11:33:33.541794: step: 892/529, loss: 0.004489421844482422 2023-01-21 11:33:34.668088: step: 896/529, loss: 0.032956384122371674 2023-01-21 11:33:35.834578: step: 900/529, loss: 0.2825174331665039 2023-01-21 11:33:36.942544: step: 904/529, loss: 0.019931888207793236 2023-01-21 11:33:38.054798: step: 908/529, loss: 0.005643606185913086 2023-01-21 11:33:39.181931: step: 912/529, loss: 0.0011617183918133378 2023-01-21 11:33:40.301062: step: 916/529, loss: 0.0005805015680380166 2023-01-21 11:33:41.448550: step: 920/529, loss: 0.024542905390262604 2023-01-21 11:33:42.583116: step: 924/529, loss: 0.008698082529008389 2023-01-21 11:33:43.743855: step: 928/529, loss: 0.0010606765281409025 2023-01-21 11:33:44.866356: step: 932/529, loss: 0.0008480072137899697 2023-01-21 11:33:46.012361: step: 936/529, loss: 0.04472217708826065 2023-01-21 11:33:47.151586: step: 940/529, loss: 3.333091444801539e-05 2023-01-21 11:33:48.306328: step: 944/529, loss: 0.37310677766799927 2023-01-21 11:33:49.453916: step: 948/529, loss: 0.0001680374116403982 2023-01-21 11:33:50.604248: step: 952/529, loss: 0.004506015684455633 2023-01-21 11:33:51.728460: step: 956/529, loss: 0.020349882543087006 2023-01-21 11:33:52.881285: step: 960/529, loss: 0.0869840607047081 2023-01-21 11:33:54.023501: step: 964/529, loss: 0.18192100524902344 2023-01-21 11:33:55.164351: step: 968/529, loss: 0.0008014679187908769 2023-01-21 11:33:56.294150: step: 972/529, loss: 0.009219598956406116 2023-01-21 11:33:57.455865: step: 976/529, loss: 0.0006784438737668097 2023-01-21 11:33:58.597995: step: 980/529, loss: 0.06420622020959854 2023-01-21 11:33:59.722746: step: 984/529, loss: 0.010016536340117455 2023-01-21 11:34:00.860582: step: 988/529, loss: 0.019565392285585403 2023-01-21 11:34:02.003798: step: 992/529, loss: 0.012001228518784046 2023-01-21 11:34:03.162807: step: 996/529, loss: 0.00846109353005886 2023-01-21 11:34:04.333947: step: 1000/529, loss: 0.015276432037353516 2023-01-21 11:34:05.451119: step: 1004/529, loss: 0.0004455566522665322 2023-01-21 11:34:06.588348: step: 1008/529, loss: 0.0019134521717205644 2023-01-21 11:34:07.740382: step: 1012/529, loss: 0.09234543144702911 2023-01-21 11:34:08.926867: step: 1016/529, loss: 0.04788951948285103 2023-01-21 11:34:10.060435: step: 1020/529, loss: 0.025681018829345703 2023-01-21 11:34:11.198694: step: 1024/529, loss: 0.00454216031357646 2023-01-21 11:34:12.356478: step: 1028/529, loss: 0.008239269256591797 2023-01-21 11:34:13.490742: step: 1032/529, loss: 0.029389668256044388 2023-01-21 11:34:14.626099: step: 1036/529, loss: 0.003059768583625555 2023-01-21 11:34:15.763087: step: 1040/529, loss: 0.007839965634047985 2023-01-21 11:34:16.919038: step: 1044/529, loss: 0.03391628339886665 2023-01-21 11:34:18.048050: step: 1048/529, loss: 0.018726253882050514 2023-01-21 11:34:19.178463: step: 1052/529, loss: 0.00048160552978515625 2023-01-21 11:34:20.304733: step: 1056/529, loss: 0.0026079178787767887 2023-01-21 11:34:21.466102: step: 1060/529, loss: 0.011533737182617188 2023-01-21 11:34:22.585326: step: 1064/529, loss: 0.0816609188914299 2023-01-21 11:34:23.711293: step: 1068/529, loss: 0.010011862963438034 2023-01-21 11:34:24.884927: step: 1072/529, loss: 0.01912817917764187 2023-01-21 11:34:26.008399: step: 1076/529, loss: 0.003060912946239114 2023-01-21 11:34:27.145092: step: 1080/529, loss: 0.027090836316347122 2023-01-21 11:34:28.273790: step: 1084/529, loss: 0.00986633263528347 2023-01-21 11:34:29.428742: step: 1088/529, loss: 0.0008974075317382812 2023-01-21 11:34:30.572895: step: 1092/529, loss: 0.1181427463889122 2023-01-21 11:34:31.744213: step: 1096/529, loss: 0.0059570311568677425 2023-01-21 11:34:32.909777: step: 1100/529, loss: 0.01003274880349636 2023-01-21 11:34:34.076457: step: 1104/529, loss: 0.0009575844742357731 2023-01-21 11:34:35.211024: step: 1108/529, loss: 0.0012995719444006681 2023-01-21 11:34:36.337517: step: 1112/529, loss: 0.028218938037753105 2023-01-21 11:34:37.489239: step: 1116/529, loss: 0.01837439462542534 2023-01-21 11:34:38.625626: step: 1120/529, loss: 0.04990367963910103 2023-01-21 11:34:39.774470: step: 1124/529, loss: 0.03256044536828995 2023-01-21 11:34:40.909238: step: 1128/529, loss: 0.006563139148056507 2023-01-21 11:34:42.036408: step: 1132/529, loss: 0.008784675970673561 2023-01-21 11:34:43.191754: step: 1136/529, loss: 0.014457417652010918 2023-01-21 11:34:44.346655: step: 1140/529, loss: 0.01504430826753378 2023-01-21 11:34:45.474056: step: 1144/529, loss: 0.14821308851242065 2023-01-21 11:34:46.617412: step: 1148/529, loss: 0.11262250691652298 2023-01-21 11:34:47.752054: step: 1152/529, loss: 0.0006557464366778731 2023-01-21 11:34:48.918681: step: 1156/529, loss: 0.02716846391558647 2023-01-21 11:34:50.050234: step: 1160/529, loss: 0.0027498246636241674 2023-01-21 11:34:51.199139: step: 1164/529, loss: 0.004001808352768421 2023-01-21 11:34:52.327585: step: 1168/529, loss: 0.032067395746707916 2023-01-21 11:34:53.474746: step: 1172/529, loss: 0.023300744593143463 2023-01-21 11:34:54.619145: step: 1176/529, loss: 6.742477125953883e-05 2023-01-21 11:34:55.761991: step: 1180/529, loss: 0.004763603676110506 2023-01-21 11:34:56.902357: step: 1184/529, loss: 0.001458454062230885 2023-01-21 11:34:58.061795: step: 1188/529, loss: 0.00858993548899889 2023-01-21 11:34:59.183174: step: 1192/529, loss: 0.011379241943359375 2023-01-21 11:35:00.307442: step: 1196/529, loss: 0.004339694976806641 2023-01-21 11:35:01.426396: step: 1200/529, loss: 0.004261398687958717 2023-01-21 11:35:02.576586: step: 1204/529, loss: 0.001772975898347795 2023-01-21 11:35:03.706390: step: 1208/529, loss: 0.0007111548911780119 2023-01-21 11:35:04.834217: step: 1212/529, loss: 0.009340333752334118 2023-01-21 11:35:05.969782: step: 1216/529, loss: 0.01994762197136879 2023-01-21 11:35:07.110741: step: 1220/529, loss: 0.011604642495512962 2023-01-21 11:35:08.274560: step: 1224/529, loss: 0.031477831304073334 2023-01-21 11:35:09.388582: step: 1228/529, loss: 0.0039580343291163445 2023-01-21 11:35:10.551995: step: 1232/529, loss: 0.010865307413041592 2023-01-21 11:35:11.673082: step: 1236/529, loss: 0.04174976423382759 2023-01-21 11:35:12.798936: step: 1240/529, loss: 0.005619239993393421 2023-01-21 11:35:13.944948: step: 1244/529, loss: 0.001994228456169367 2023-01-21 11:35:15.046862: step: 1248/529, loss: 0.00259323138743639 2023-01-21 11:35:16.183577: step: 1252/529, loss: 0.047060441225767136 2023-01-21 11:35:17.356528: step: 1256/529, loss: 0.037805747240781784 2023-01-21 11:35:18.471624: step: 1260/529, loss: 0.02530183643102646 2023-01-21 11:35:19.607181: step: 1264/529, loss: 0.01933145523071289 2023-01-21 11:35:20.757099: step: 1268/529, loss: 0.1589651107788086 2023-01-21 11:35:21.935484: step: 1272/529, loss: 0.021904945373535156 2023-01-21 11:35:23.065075: step: 1276/529, loss: 0.034422971308231354 2023-01-21 11:35:24.205471: step: 1280/529, loss: 0.015123940072953701 2023-01-21 11:35:25.361627: step: 1284/529, loss: 0.055373094975948334 2023-01-21 11:35:26.482574: step: 1288/529, loss: 0.0031748772598803043 2023-01-21 11:35:27.602312: step: 1292/529, loss: 0.002681779908016324 2023-01-21 11:35:28.758207: step: 1296/529, loss: 0.013304234482347965 2023-01-21 11:35:29.880249: step: 1300/529, loss: 0.001902771065942943 2023-01-21 11:35:31.009126: step: 1304/529, loss: 0.0015788078308105469 2023-01-21 11:35:32.218755: step: 1308/529, loss: 0.5124379396438599 2023-01-21 11:35:33.368857: step: 1312/529, loss: 0.0031995773315429688 2023-01-21 11:35:34.534607: step: 1316/529, loss: 0.004222059156745672 2023-01-21 11:35:35.651396: step: 1320/529, loss: 0.0001715660182526335 2023-01-21 11:35:36.788313: step: 1324/529, loss: 0.04402942955493927 2023-01-21 11:35:37.926373: step: 1328/529, loss: 0.02411935292184353 2023-01-21 11:35:39.049651: step: 1332/529, loss: 0.0003262519894633442 2023-01-21 11:35:40.204110: step: 1336/529, loss: 0.058582305908203125 2023-01-21 11:35:41.358823: step: 1340/529, loss: 0.0057294368743896484 2023-01-21 11:35:42.510910: step: 1344/529, loss: 0.044020652770996094 2023-01-21 11:35:43.658180: step: 1348/529, loss: 0.0038466453552246094 2023-01-21 11:35:44.800370: step: 1352/529, loss: 6.456374831032008e-05 2023-01-21 11:35:45.925664: step: 1356/529, loss: 0.0012562752235680819 2023-01-21 11:35:47.072010: step: 1360/529, loss: 0.07114791870117188 2023-01-21 11:35:48.214246: step: 1364/529, loss: 0.03284282982349396 2023-01-21 11:35:49.347592: step: 1368/529, loss: 0.06797914206981659 2023-01-21 11:35:50.469359: step: 1372/529, loss: 0.0006488800281658769 2023-01-21 11:35:51.577032: step: 1376/529, loss: 0.045902520418167114 2023-01-21 11:35:52.716051: step: 1380/529, loss: 0.008991813287138939 2023-01-21 11:35:53.876802: step: 1384/529, loss: 0.00012581348710227758 2023-01-21 11:35:55.033511: step: 1388/529, loss: 0.048009779304265976 2023-01-21 11:35:56.216108: step: 1392/529, loss: 0.0002168655482819304 2023-01-21 11:35:57.344846: step: 1396/529, loss: 0.05149250105023384 2023-01-21 11:35:58.484104: step: 1400/529, loss: 0.03655543178319931 2023-01-21 11:35:59.608117: step: 1404/529, loss: 0.011797333136200905 2023-01-21 11:36:00.749848: step: 1408/529, loss: 0.0029134750366210938 2023-01-21 11:36:01.919455: step: 1412/529, loss: 0.024343110620975494 2023-01-21 11:36:03.091096: step: 1416/529, loss: 0.004498195834457874 2023-01-21 11:36:04.227541: step: 1420/529, loss: 0.00016536115435883403 2023-01-21 11:36:05.364956: step: 1424/529, loss: 0.001154828118160367 2023-01-21 11:36:06.497320: step: 1428/529, loss: 0.0027225492522120476 2023-01-21 11:36:07.647169: step: 1432/529, loss: 0.03768548741936684 2023-01-21 11:36:08.793646: step: 1436/529, loss: 0.0017038345104083419 2023-01-21 11:36:09.918436: step: 1440/529, loss: 0.15786190330982208 2023-01-21 11:36:11.052111: step: 1444/529, loss: 0.0014646649360656738 2023-01-21 11:36:12.204847: step: 1448/529, loss: 0.048471398651599884 2023-01-21 11:36:13.345042: step: 1452/529, loss: 0.018179655075073242 2023-01-21 11:36:14.507704: step: 1456/529, loss: 0.0007168054580688477 2023-01-21 11:36:15.630245: step: 1460/529, loss: 0.0009873866802081466 2023-01-21 11:36:16.769448: step: 1464/529, loss: 0.020849037915468216 2023-01-21 11:36:17.948399: step: 1468/529, loss: 0.00379009242169559 2023-01-21 11:36:19.081790: step: 1472/529, loss: 0.007500267121940851 2023-01-21 11:36:20.220352: step: 1476/529, loss: 0.013592339120805264 2023-01-21 11:36:21.366556: step: 1480/529, loss: 0.018526267260313034 2023-01-21 11:36:22.495980: step: 1484/529, loss: 0.001071929931640625 2023-01-21 11:36:23.640599: step: 1488/529, loss: 0.04881887510418892 2023-01-21 11:36:24.763135: step: 1492/529, loss: 0.001932763960212469 2023-01-21 11:36:25.868279: step: 1496/529, loss: 0.00630607595667243 2023-01-21 11:36:27.013000: step: 1500/529, loss: 0.0009374618530273438 2023-01-21 11:36:28.142742: step: 1504/529, loss: 0.02255268208682537 2023-01-21 11:36:29.244586: step: 1508/529, loss: 0.0019861222244799137 2023-01-21 11:36:30.379961: step: 1512/529, loss: 0.028542710468173027 2023-01-21 11:36:31.504491: step: 1516/529, loss: 0.0031071186531335115 2023-01-21 11:36:32.662274: step: 1520/529, loss: 0.008000754751265049 2023-01-21 11:36:33.789525: step: 1524/529, loss: 0.053278446197509766 2023-01-21 11:36:34.925944: step: 1528/529, loss: 0.004903316497802734 2023-01-21 11:36:36.079250: step: 1532/529, loss: 0.03112621419131756 2023-01-21 11:36:37.240901: step: 1536/529, loss: 0.14855670928955078 2023-01-21 11:36:38.379437: step: 1540/529, loss: 0.004803466610610485 2023-01-21 11:36:39.508241: step: 1544/529, loss: 0.001744222710840404 2023-01-21 11:36:40.647857: step: 1548/529, loss: 0.0007801532628946006 2023-01-21 11:36:41.805896: step: 1552/529, loss: 0.0017047881847247481 2023-01-21 11:36:42.935380: step: 1556/529, loss: 0.020837783813476562 2023-01-21 11:36:44.089243: step: 1560/529, loss: 0.0008647918584756553 2023-01-21 11:36:45.234011: step: 1564/529, loss: 0.012898826971650124 2023-01-21 11:36:46.357802: step: 1568/529, loss: 0.005711174104362726 2023-01-21 11:36:47.522518: step: 1572/529, loss: 0.006619644351303577 2023-01-21 11:36:48.674257: step: 1576/529, loss: 0.011004924774169922 2023-01-21 11:36:49.813546: step: 1580/529, loss: 0.07417793571949005 2023-01-21 11:36:50.975294: step: 1584/529, loss: 0.009404182434082031 2023-01-21 11:36:52.154742: step: 1588/529, loss: 0.01302499882876873 2023-01-21 11:36:53.289818: step: 1592/529, loss: 0.009563923813402653 2023-01-21 11:36:54.411511: step: 1596/529, loss: 0.06909974664449692 2023-01-21 11:36:55.546001: step: 1600/529, loss: 0.006190919782966375 2023-01-21 11:36:56.701723: step: 1604/529, loss: 0.00837564468383789 2023-01-21 11:36:57.880212: step: 1608/529, loss: 0.0008972167852334678 2023-01-21 11:36:59.009274: step: 1612/529, loss: 0.008368909358978271 2023-01-21 11:37:00.204104: step: 1616/529, loss: 0.0001574277994222939 2023-01-21 11:37:01.329456: step: 1620/529, loss: 0.0008803367963992059 2023-01-21 11:37:02.469392: step: 1624/529, loss: 0.013638115487992764 2023-01-21 11:37:03.580429: step: 1628/529, loss: 0.007931900210678577 2023-01-21 11:37:04.727948: step: 1632/529, loss: 0.032462313771247864 2023-01-21 11:37:05.890011: step: 1636/529, loss: 0.01355667132884264 2023-01-21 11:37:07.024084: step: 1640/529, loss: 0.0021865845192223787 2023-01-21 11:37:08.160588: step: 1644/529, loss: 0.031923867762088776 2023-01-21 11:37:09.308808: step: 1648/529, loss: 0.0026908875443041325 2023-01-21 11:37:10.452105: step: 1652/529, loss: 3.15189354296308e-05 2023-01-21 11:37:11.595871: step: 1656/529, loss: 0.009342193603515625 2023-01-21 11:37:12.742807: step: 1660/529, loss: 0.06308746337890625 2023-01-21 11:37:13.879846: step: 1664/529, loss: 0.004986322019249201 2023-01-21 11:37:15.036759: step: 1668/529, loss: 0.03224611654877663 2023-01-21 11:37:16.175743: step: 1672/529, loss: 0.014443683438003063 2023-01-21 11:37:17.314000: step: 1676/529, loss: 2.8705597287626006e-05 2023-01-21 11:37:18.475196: step: 1680/529, loss: 0.007191276643425226 2023-01-21 11:37:19.604404: step: 1684/529, loss: 0.02531290054321289 2023-01-21 11:37:20.725636: step: 1688/529, loss: 0.0535762794315815 2023-01-21 11:37:21.898506: step: 1692/529, loss: 0.018215704709291458 2023-01-21 11:37:23.013254: step: 1696/529, loss: 0.03373556211590767 2023-01-21 11:37:24.156978: step: 1700/529, loss: 0.059294700622558594 2023-01-21 11:37:25.291654: step: 1704/529, loss: 0.0002349853457417339 2023-01-21 11:37:26.425997: step: 1708/529, loss: 0.053058624267578125 2023-01-21 11:37:27.565340: step: 1712/529, loss: 0.07703609764575958 2023-01-21 11:37:28.707783: step: 1716/529, loss: 0.00021595954603981227 2023-01-21 11:37:29.820489: step: 1720/529, loss: 0.00013513564772438258 2023-01-21 11:37:30.946767: step: 1724/529, loss: 0.2744404077529907 2023-01-21 11:37:32.087428: step: 1728/529, loss: 0.038552094250917435 2023-01-21 11:37:33.231130: step: 1732/529, loss: 0.00014696121797896922 2023-01-21 11:37:34.369619: step: 1736/529, loss: 0.00348587054759264 2023-01-21 11:37:35.532167: step: 1740/529, loss: 0.021732330322265625 2023-01-21 11:37:36.662985: step: 1744/529, loss: 0.034837912768125534 2023-01-21 11:37:37.823463: step: 1748/529, loss: 0.004411506932228804 2023-01-21 11:37:38.961306: step: 1752/529, loss: 0.029863454401493073 2023-01-21 11:37:40.105165: step: 1756/529, loss: 0.013660240918397903 2023-01-21 11:37:41.228666: step: 1760/529, loss: 0.0025158883072435856 2023-01-21 11:37:42.400237: step: 1764/529, loss: 0.022594833746552467 2023-01-21 11:37:43.545928: step: 1768/529, loss: 0.006878662388771772 2023-01-21 11:37:44.652590: step: 1772/529, loss: 0.06906967610120773 2023-01-21 11:37:45.791977: step: 1776/529, loss: 0.08014822006225586 2023-01-21 11:37:46.945609: step: 1780/529, loss: 0.017841434106230736 2023-01-21 11:37:48.072106: step: 1784/529, loss: 0.005183410830795765 2023-01-21 11:37:49.203309: step: 1788/529, loss: 0.006915092468261719 2023-01-21 11:37:50.339608: step: 1792/529, loss: 0.0016990662552416325 2023-01-21 11:37:51.498924: step: 1796/529, loss: 0.001139736152254045 2023-01-21 11:37:52.649403: step: 1800/529, loss: 0.03249001502990723 2023-01-21 11:37:53.809274: step: 1804/529, loss: 0.04183673858642578 2023-01-21 11:37:54.957668: step: 1808/529, loss: 0.023811722174286842 2023-01-21 11:37:56.069026: step: 1812/529, loss: 0.01132826879620552 2023-01-21 11:37:57.197283: step: 1816/529, loss: 0.020188618451356888 2023-01-21 11:37:58.380474: step: 1820/529, loss: 0.01529989205300808 2023-01-21 11:37:59.530403: step: 1824/529, loss: 0.0002399444638285786 2023-01-21 11:38:00.659315: step: 1828/529, loss: 0.0646999403834343 2023-01-21 11:38:01.787930: step: 1832/529, loss: 0.0010093450546264648 2023-01-21 11:38:02.909186: step: 1836/529, loss: 0.005635166075080633 2023-01-21 11:38:04.069277: step: 1840/529, loss: 0.04038562998175621 2023-01-21 11:38:05.221752: step: 1844/529, loss: 0.0016530990833416581 2023-01-21 11:38:06.328660: step: 1848/529, loss: 0.08726216107606888 2023-01-21 11:38:07.492581: step: 1852/529, loss: 0.008583259768784046 2023-01-21 11:38:08.646884: step: 1856/529, loss: 0.029050350189208984 2023-01-21 11:38:09.788911: step: 1860/529, loss: 0.00017480850510764867 2023-01-21 11:38:10.910348: step: 1864/529, loss: 0.0016067505348473787 2023-01-21 11:38:12.048653: step: 1868/529, loss: 0.02288360521197319 2023-01-21 11:38:13.178081: step: 1872/529, loss: 0.004376411437988281 2023-01-21 11:38:14.305830: step: 1876/529, loss: 0.0039520263671875 2023-01-21 11:38:15.447626: step: 1880/529, loss: 0.005800056271255016 2023-01-21 11:38:16.571625: step: 1884/529, loss: 0.0042896270751953125 2023-01-21 11:38:17.679936: step: 1888/529, loss: 0.007916641421616077 2023-01-21 11:38:18.815037: step: 1892/529, loss: 0.027043821290135384 2023-01-21 11:38:19.955012: step: 1896/529, loss: 0.033438730984926224 2023-01-21 11:38:21.075136: step: 1900/529, loss: 0.02346782758831978 2023-01-21 11:38:22.197796: step: 1904/529, loss: 0.0004980087396688759 2023-01-21 11:38:23.334301: step: 1908/529, loss: 0.027830936014652252 2023-01-21 11:38:24.467794: step: 1912/529, loss: 0.003227424807846546 2023-01-21 11:38:25.616368: step: 1916/529, loss: 0.1706605702638626 2023-01-21 11:38:26.734787: step: 1920/529, loss: 0.004867172334343195 2023-01-21 11:38:27.861031: step: 1924/529, loss: 0.04011125490069389 2023-01-21 11:38:29.021922: step: 1928/529, loss: 0.032246481627225876 2023-01-21 11:38:30.157361: step: 1932/529, loss: 0.002513313200324774 2023-01-21 11:38:31.282323: step: 1936/529, loss: 0.0019401074387133121 2023-01-21 11:38:32.421466: step: 1940/529, loss: 0.009066772647202015 2023-01-21 11:38:33.560970: step: 1944/529, loss: 0.8269193768501282 2023-01-21 11:38:34.704777: step: 1948/529, loss: 0.04860830307006836 2023-01-21 11:38:35.832966: step: 1952/529, loss: 0.001714420272037387 2023-01-21 11:38:36.972611: step: 1956/529, loss: 0.10605239868164062 2023-01-21 11:38:38.113085: step: 1960/529, loss: 0.045519448816776276 2023-01-21 11:38:39.251117: step: 1964/529, loss: 0.008381461724638939 2023-01-21 11:38:40.372458: step: 1968/529, loss: 0.0006935119745321572 2023-01-21 11:38:41.537440: step: 1972/529, loss: 0.03565178066492081 2023-01-21 11:38:42.671417: step: 1976/529, loss: 0.01486358605325222 2023-01-21 11:38:43.800395: step: 1980/529, loss: 0.019589710980653763 2023-01-21 11:38:44.921588: step: 1984/529, loss: 0.05921641364693642 2023-01-21 11:38:46.041928: step: 1988/529, loss: 0.012999439612030983 2023-01-21 11:38:47.177200: step: 1992/529, loss: 0.004659461788833141 2023-01-21 11:38:48.307851: step: 1996/529, loss: 0.000274658203125 2023-01-21 11:38:49.445672: step: 2000/529, loss: 0.021384811028838158 2023-01-21 11:38:50.593699: step: 2004/529, loss: 0.012132071889936924 2023-01-21 11:38:51.741705: step: 2008/529, loss: 0.004119968507438898 2023-01-21 11:38:52.886977: step: 2012/529, loss: 0.014011621475219727 2023-01-21 11:38:54.033934: step: 2016/529, loss: 0.008139420300722122 2023-01-21 11:38:55.195528: step: 2020/529, loss: 0.004028892610222101 2023-01-21 11:38:56.352223: step: 2024/529, loss: 0.01865110546350479 2023-01-21 11:38:57.472963: step: 2028/529, loss: 0.008967304602265358 2023-01-21 11:38:58.648404: step: 2032/529, loss: 0.00047512055607512593 2023-01-21 11:38:59.803851: step: 2036/529, loss: 0.0005553245428018272 2023-01-21 11:39:00.941428: step: 2040/529, loss: 0.015929032117128372 2023-01-21 11:39:02.100582: step: 2044/529, loss: 0.00013046264939475805 2023-01-21 11:39:03.241800: step: 2048/529, loss: 0.004607581999152899 2023-01-21 11:39:04.379633: step: 2052/529, loss: 0.026881981641054153 2023-01-21 11:39:05.529189: step: 2056/529, loss: 0.014725113287568092 2023-01-21 11:39:06.658590: step: 2060/529, loss: 0.04262485355138779 2023-01-21 11:39:07.835392: step: 2064/529, loss: 0.019208144396543503 2023-01-21 11:39:08.974695: step: 2068/529, loss: 0.0010837555164471269 2023-01-21 11:39:10.115527: step: 2072/529, loss: 0.027039146050810814 2023-01-21 11:39:11.280756: step: 2076/529, loss: 0.00471839914098382 2023-01-21 11:39:12.455097: step: 2080/529, loss: 0.00048236845759674907 2023-01-21 11:39:13.608933: step: 2084/529, loss: 0.003652381943538785 2023-01-21 11:39:14.763944: step: 2088/529, loss: 0.039153292775154114 2023-01-21 11:39:15.917500: step: 2092/529, loss: 0.025563431903719902 2023-01-21 11:39:17.029922: step: 2096/529, loss: 0.005202865693718195 2023-01-21 11:39:18.166767: step: 2100/529, loss: 0.0768117904663086 2023-01-21 11:39:19.314160: step: 2104/529, loss: 0.036226749420166016 2023-01-21 11:39:20.461325: step: 2108/529, loss: 0.050469204783439636 2023-01-21 11:39:21.630157: step: 2112/529, loss: 0.017507268115878105 2023-01-21 11:39:22.750902: step: 2116/529, loss: 0.0002964019658975303 ================================================== Loss: 0.027 -------------------- Dev: {'event': {'p': 0.6127659574468085, 'r': 0.7669773635153129, 'f1': 0.6812536960378475}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Test: {'event': {'p': 0.6105165193113076, 'r': 0.7601390498261877, 'f1': 0.6771612903225807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Chinese: {'event': {'p': 0.6, 'r': 0.8888888888888888, 'f1': 0.7164179104477612}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Korean: {'event': {'p': 0.6153846153846154, 'r': 0.5079365079365079, 'f1': 0.5565217391304348}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Russian: {'event': {'p': 0.4358974358974359, 'r': 0.4722222222222222, 'f1': 0.45333333333333337}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.5733722060252673, 'r': 0.7856191744340879, 'f1': 0.6629213483146068}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Russian: {'event': {'p': 0.5876334519572953, 'r': 0.7653534183082271, 'f1': 0.6648213387015602}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} 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': 6} ****************************** Epoch: 18 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:40:04.744018: step: 4/529, loss: 0.006241417024284601 2023-01-21 11:40:05.870664: step: 8/529, loss: 0.0061032772064208984 2023-01-21 11:40:06.995337: step: 12/529, loss: 0.03459692373871803 2023-01-21 11:40:08.103608: step: 16/529, loss: 0.003876555012539029 2023-01-21 11:40:09.256657: step: 20/529, loss: 0.0016886235680431128 2023-01-21 11:40:10.403310: step: 24/529, loss: 0.0014074802165850997 2023-01-21 11:40:11.535833: step: 28/529, loss: 0.005291938781738281 2023-01-21 11:40:12.663136: step: 32/529, loss: 0.00221672048792243 2023-01-21 11:40:13.803319: step: 36/529, loss: 0.009729958139359951 2023-01-21 11:40:14.954732: step: 40/529, loss: 0.005330181214958429 2023-01-21 11:40:16.090569: step: 44/529, loss: 0.01841764524579048 2023-01-21 11:40:17.213146: step: 48/529, loss: 0.006722688674926758 2023-01-21 11:40:18.391450: step: 52/529, loss: 0.06309223175048828 2023-01-21 11:40:19.536185: step: 56/529, loss: 0.001050758408382535 2023-01-21 11:40:20.677008: step: 60/529, loss: 0.00201759347692132 2023-01-21 11:40:21.832173: step: 64/529, loss: 0.0302263256162405 2023-01-21 11:40:22.981830: step: 68/529, loss: 0.002188778016716242 2023-01-21 11:40:24.126290: step: 72/529, loss: 0.0032983780838549137 2023-01-21 11:40:25.277524: step: 76/529, loss: 0.00020647048950195312 2023-01-21 11:40:26.404029: step: 80/529, loss: 0.00017786026000976562 2023-01-21 11:40:27.547533: step: 84/529, loss: 0.008201408199965954 2023-01-21 11:40:28.666203: step: 88/529, loss: 0.030338671058416367 2023-01-21 11:40:29.766424: step: 92/529, loss: 0.03889918327331543 2023-01-21 11:40:30.887603: step: 96/529, loss: 0.01790752448141575 2023-01-21 11:40:32.051457: step: 100/529, loss: 0.005992889404296875 2023-01-21 11:40:33.228913: step: 104/529, loss: 0.010418891906738281 2023-01-21 11:40:34.329600: step: 108/529, loss: 0.007765770424157381 2023-01-21 11:40:35.493441: step: 112/529, loss: 0.008222723379731178 2023-01-21 11:40:36.635413: step: 116/529, loss: 0.0582612045109272 2023-01-21 11:40:37.784189: step: 120/529, loss: 0.3211834132671356 2023-01-21 11:40:38.925061: step: 124/529, loss: 0.019330216571688652 2023-01-21 11:40:40.080476: step: 128/529, loss: 0.00024390222097281367 2023-01-21 11:40:41.241097: step: 132/529, loss: 0.026429081335663795 2023-01-21 11:40:42.385412: step: 136/529, loss: 0.010874176397919655 2023-01-21 11:40:43.542706: step: 140/529, loss: 0.0034130574204027653 2023-01-21 11:40:44.700536: step: 144/529, loss: 0.0002536773681640625 2023-01-21 11:40:45.827886: step: 148/529, loss: 0.00011529921903274953 2023-01-21 11:40:46.998323: step: 152/529, loss: 0.015891456976532936 2023-01-21 11:40:48.145450: step: 156/529, loss: 1.773834264895413e-05 2023-01-21 11:40:49.277819: step: 160/529, loss: 0.0013849259121343493 2023-01-21 11:40:50.392342: step: 164/529, loss: 0.0010476828319951892 2023-01-21 11:40:51.554570: step: 168/529, loss: 0.0007966518169268966 2023-01-21 11:40:52.728081: step: 172/529, loss: 0.18595227599143982 2023-01-21 11:40:53.864273: step: 176/529, loss: 0.009113120846450329 2023-01-21 11:40:55.001917: step: 180/529, loss: 0.013454628176987171 2023-01-21 11:40:56.129188: step: 184/529, loss: 0.04678931459784508 2023-01-21 11:40:57.264956: step: 188/529, loss: 0.0010718345874920487 2023-01-21 11:40:58.440818: step: 192/529, loss: 0.02949066273868084 2023-01-21 11:40:59.581023: step: 196/529, loss: 0.11929173767566681 2023-01-21 11:41:00.703491: step: 200/529, loss: 3.147125244140625e-05 2023-01-21 11:41:01.828168: step: 204/529, loss: 0.02879619598388672 2023-01-21 11:41:02.965081: step: 208/529, loss: 0.000569248222745955 2023-01-21 11:41:04.072709: step: 212/529, loss: 0.008686209097504616 2023-01-21 11:41:05.209680: step: 216/529, loss: 0.0018167495727539062 2023-01-21 11:41:06.349140: step: 220/529, loss: 0.008172071538865566 2023-01-21 11:41:07.487722: step: 224/529, loss: 0.1416267305612564 2023-01-21 11:41:08.608858: step: 228/529, loss: 0.018035221844911575 2023-01-21 11:41:09.750657: step: 232/529, loss: 0.006991481874138117 2023-01-21 11:41:10.857348: step: 236/529, loss: 0.004014301113784313 2023-01-21 11:41:11.987217: step: 240/529, loss: 0.0028782845474779606 2023-01-21 11:41:13.114935: step: 244/529, loss: 0.035027600824832916 2023-01-21 11:41:14.250371: step: 248/529, loss: 0.0024260522332042456 2023-01-21 11:41:15.372505: step: 252/529, loss: 0.001992130186408758 2023-01-21 11:41:16.509149: step: 256/529, loss: 0.002309418050572276 2023-01-21 11:41:17.640235: step: 260/529, loss: 0.00019178391085006297 2023-01-21 11:41:18.779637: step: 264/529, loss: 0.001349115394987166 2023-01-21 11:41:19.906528: step: 268/529, loss: 0.013648604974150658 2023-01-21 11:41:21.068181: step: 272/529, loss: 0.018121814355254173 2023-01-21 11:41:22.244587: step: 276/529, loss: 0.02869129180908203 2023-01-21 11:41:23.414471: step: 280/529, loss: 0.0020410537254065275 2023-01-21 11:41:24.532795: step: 284/529, loss: 0.005939770024269819 2023-01-21 11:41:25.765698: step: 288/529, loss: 0.03592319414019585 2023-01-21 11:41:26.945365: step: 292/529, loss: 0.023594284430146217 2023-01-21 11:41:28.081842: step: 296/529, loss: 0.014547920785844326 2023-01-21 11:41:29.230110: step: 300/529, loss: 0.006482887081801891 2023-01-21 11:41:30.371284: step: 304/529, loss: 0.00036687852116301656 2023-01-21 11:41:31.516774: step: 308/529, loss: 0.00682144146412611 2023-01-21 11:41:32.646078: step: 312/529, loss: 0.009296512231230736 2023-01-21 11:41:33.791515: step: 316/529, loss: 0.031868934631347656 2023-01-21 11:41:34.948391: step: 320/529, loss: 0.005333900451660156 2023-01-21 11:41:36.053228: step: 324/529, loss: 0.0007100105285644531 2023-01-21 11:41:37.205995: step: 328/529, loss: 0.010809993371367455 2023-01-21 11:41:38.337760: step: 332/529, loss: 0.0008709907415322959 2023-01-21 11:41:39.492047: step: 336/529, loss: 0.04710130766034126 2023-01-21 11:41:40.675010: step: 340/529, loss: 0.0024326324928551912 2023-01-21 11:41:41.859248: step: 344/529, loss: 0.003944873809814453 2023-01-21 11:41:43.033202: step: 348/529, loss: 0.04175719991326332 2023-01-21 11:41:44.175943: step: 352/529, loss: 0.019383490085601807 2023-01-21 11:41:45.321353: step: 356/529, loss: 0.005240630824118853 2023-01-21 11:41:46.460861: step: 360/529, loss: 0.006787300109863281 2023-01-21 11:41:47.562243: step: 364/529, loss: 0.005274200811982155 2023-01-21 11:41:48.689881: step: 368/529, loss: 0.0012506485218182206 2023-01-21 11:41:49.861816: step: 372/529, loss: 0.024988461285829544 2023-01-21 11:41:51.007089: step: 376/529, loss: 0.006541252136230469 2023-01-21 11:41:52.133238: step: 380/529, loss: 0.008476734161376953 2023-01-21 11:41:53.255116: step: 384/529, loss: 0.0029168606270104647 2023-01-21 11:41:54.364539: step: 388/529, loss: 0.010167121887207031 2023-01-21 11:41:55.519493: step: 392/529, loss: 0.05649566650390625 2023-01-21 11:41:56.663650: step: 396/529, loss: 0.0014757156604900956 2023-01-21 11:41:57.806822: step: 400/529, loss: 2.212524486822076e-05 2023-01-21 11:41:58.936361: step: 404/529, loss: 0.026470374315977097 2023-01-21 11:42:00.096857: step: 408/529, loss: 0.001718807267025113 2023-01-21 11:42:01.238268: step: 412/529, loss: 0.02577686496078968 2023-01-21 11:42:02.384004: step: 416/529, loss: 0.011780737899243832 2023-01-21 11:42:03.535648: step: 420/529, loss: 0.004677534103393555 2023-01-21 11:42:04.678894: step: 424/529, loss: 0.4459524154663086 2023-01-21 11:42:05.805313: step: 428/529, loss: 0.01160652656108141 2023-01-21 11:42:06.966579: step: 432/529, loss: 0.004642319865524769 2023-01-21 11:42:08.130300: step: 436/529, loss: 0.037483979016542435 2023-01-21 11:42:09.267386: step: 440/529, loss: 0.012015819549560547 2023-01-21 11:42:10.394379: step: 444/529, loss: 0.018959617242217064 2023-01-21 11:42:11.541753: step: 448/529, loss: 0.0028689384926110506 2023-01-21 11:42:12.699131: step: 452/529, loss: 0.00026264190091751516 2023-01-21 11:42:13.821304: step: 456/529, loss: 0.01378555316478014 2023-01-21 11:42:14.954285: step: 460/529, loss: 0.02205219306051731 2023-01-21 11:42:16.145661: step: 464/529, loss: 0.5884515643119812 2023-01-21 11:42:17.283959: step: 468/529, loss: 0.07545318454504013 2023-01-21 11:42:18.454033: step: 472/529, loss: 0.007654380984604359 2023-01-21 11:42:19.593905: step: 476/529, loss: 0.0004050426068715751 2023-01-21 11:42:20.767165: step: 480/529, loss: 0.04496908187866211 2023-01-21 11:42:21.899961: step: 484/529, loss: 0.07120780646800995 2023-01-21 11:42:23.014206: step: 488/529, loss: 0.0014478684170171618 2023-01-21 11:42:24.142213: step: 492/529, loss: 0.01540532149374485 2023-01-21 11:42:25.263723: step: 496/529, loss: 0.03237638622522354 2023-01-21 11:42:26.418587: step: 500/529, loss: 0.027593422681093216 2023-01-21 11:42:27.539469: step: 504/529, loss: 0.06308670341968536 2023-01-21 11:42:28.673273: step: 508/529, loss: 0.02149190939962864 2023-01-21 11:42:29.835591: step: 512/529, loss: 0.02747969701886177 2023-01-21 11:42:30.966949: step: 516/529, loss: 0.008552360348403454 2023-01-21 11:42:32.098234: step: 520/529, loss: 0.0007078170892782509 2023-01-21 11:42:33.258259: step: 524/529, loss: 0.0004834175342693925 2023-01-21 11:42:34.395491: step: 528/529, loss: 0.000329864036757499 2023-01-21 11:42:35.511225: step: 532/529, loss: 0.0024454593658447266 2023-01-21 11:42:36.658562: step: 536/529, loss: 0.02578124962747097 2023-01-21 11:42:37.822502: step: 540/529, loss: 0.08916592597961426 2023-01-21 11:42:38.947170: step: 544/529, loss: 0.003316497663035989 2023-01-21 11:42:40.066942: step: 548/529, loss: 0.037149809300899506 2023-01-21 11:42:41.196999: step: 552/529, loss: 0.016155825927853584 2023-01-21 11:42:42.298202: step: 556/529, loss: 0.003591919085010886 2023-01-21 11:42:43.470520: step: 560/529, loss: 0.010695266537368298 2023-01-21 11:42:44.617946: step: 564/529, loss: 0.0005551338545046747 2023-01-21 11:42:45.759607: step: 568/529, loss: 0.06317882984876633 2023-01-21 11:42:46.876939: step: 572/529, loss: 0.06723079830408096 2023-01-21 11:42:48.016329: step: 576/529, loss: 0.04876995086669922 2023-01-21 11:42:49.193396: step: 580/529, loss: 0.001155090401880443 2023-01-21 11:42:50.321258: step: 584/529, loss: 0.005017089657485485 2023-01-21 11:42:51.452613: step: 588/529, loss: 0.026406003162264824 2023-01-21 11:42:52.572976: step: 592/529, loss: 0.023432351648807526 2023-01-21 11:42:53.660624: step: 596/529, loss: 0.00027790068998001516 2023-01-21 11:42:54.805817: step: 600/529, loss: 0.011398602277040482 2023-01-21 11:42:55.932807: step: 604/529, loss: 0.0027114988770335913 2023-01-21 11:42:57.054134: step: 608/529, loss: 0.030648041516542435 2023-01-21 11:42:58.180907: step: 612/529, loss: 0.0653233528137207 2023-01-21 11:42:59.322958: step: 616/529, loss: 0.0022360801231116056 2023-01-21 11:43:00.441302: step: 620/529, loss: 0.06946296244859695 2023-01-21 11:43:01.588957: step: 624/529, loss: 0.0036852839402854443 2023-01-21 11:43:02.746476: step: 628/529, loss: 0.030469607561826706 2023-01-21 11:43:03.895608: step: 632/529, loss: 0.00042304996168240905 2023-01-21 11:43:05.038998: step: 636/529, loss: 0.010606765747070312 2023-01-21 11:43:06.178817: step: 640/529, loss: 0.005016613285988569 2023-01-21 11:43:07.302466: step: 644/529, loss: 0.0085289953276515 2023-01-21 11:43:08.435428: step: 648/529, loss: 0.04410257190465927 2023-01-21 11:43:09.594832: step: 652/529, loss: 0.018637657165527344 2023-01-21 11:43:10.757742: step: 656/529, loss: 0.028317928314208984 2023-01-21 11:43:11.929342: step: 660/529, loss: 0.0003578185860533267 2023-01-21 11:43:13.094164: step: 664/529, loss: 0.03711872175335884 2023-01-21 11:43:14.225565: step: 668/529, loss: 0.00867323949933052 2023-01-21 11:43:15.383652: step: 672/529, loss: 0.0015787124866619706 2023-01-21 11:43:16.520034: step: 676/529, loss: 0.0031977654434740543 2023-01-21 11:43:17.674080: step: 680/529, loss: 0.04297066107392311 2023-01-21 11:43:18.813626: step: 684/529, loss: 7.22885160939768e-05 2023-01-21 11:43:19.953415: step: 688/529, loss: 0.0002624511835165322 2023-01-21 11:43:21.083160: step: 692/529, loss: 0.024991370737552643 2023-01-21 11:43:22.222493: step: 696/529, loss: 0.07466640323400497 2023-01-21 11:43:23.356172: step: 700/529, loss: 0.0002079010009765625 2023-01-21 11:43:24.480826: step: 704/529, loss: 0.000599384366068989 2023-01-21 11:43:25.636094: step: 708/529, loss: 0.0005156517145223916 2023-01-21 11:43:26.799125: step: 712/529, loss: 0.03506955876946449 2023-01-21 11:43:27.957939: step: 716/529, loss: -1.602172778802924e-05 2023-01-21 11:43:29.103862: step: 720/529, loss: 5.3215029765851796e-05 2023-01-21 11:43:30.225634: step: 724/529, loss: 0.019298363476991653 2023-01-21 11:43:31.365599: step: 728/529, loss: 0.016275787726044655 2023-01-21 11:43:32.491223: step: 732/529, loss: 0.006672763731330633 2023-01-21 11:43:33.625672: step: 736/529, loss: 0.0016570091247558594 2023-01-21 11:43:34.753209: step: 740/529, loss: 0.018860816955566406 2023-01-21 11:43:35.899347: step: 744/529, loss: 0.011258983984589577 2023-01-21 11:43:37.047098: step: 748/529, loss: 0.004781436640769243 2023-01-21 11:43:38.202829: step: 752/529, loss: 0.00012254714965820312 2023-01-21 11:43:39.328685: step: 756/529, loss: 0.019025038927793503 2023-01-21 11:43:40.463115: step: 760/529, loss: 0.03364739567041397 2023-01-21 11:43:41.589184: step: 764/529, loss: 0.00016317368135787547 2023-01-21 11:43:42.740303: step: 768/529, loss: 0.015932131558656693 2023-01-21 11:43:43.863910: step: 772/529, loss: 0.03281116485595703 2023-01-21 11:43:45.011967: step: 776/529, loss: 0.014200402423739433 2023-01-21 11:43:46.135586: step: 780/529, loss: 0.013457298278808594 2023-01-21 11:43:47.258460: step: 784/529, loss: 0.0076507567428052425 2023-01-21 11:43:48.396767: step: 788/529, loss: 0.08912134170532227 2023-01-21 11:43:49.554230: step: 792/529, loss: 0.04129061847925186 2023-01-21 11:43:50.724723: step: 796/529, loss: 0.005247974302619696 2023-01-21 11:43:51.854681: step: 800/529, loss: 0.0007427216041833162 2023-01-21 11:43:52.992696: step: 804/529, loss: 0.031065750867128372 2023-01-21 11:43:54.134671: step: 808/529, loss: 0.00010566711716819555 2023-01-21 11:43:55.247143: step: 812/529, loss: 0.005059433169662952 2023-01-21 11:43:56.386598: step: 816/529, loss: 0.7163230776786804 2023-01-21 11:43:57.536864: step: 820/529, loss: 0.09655914455652237 2023-01-21 11:43:58.662470: step: 824/529, loss: 0.016176700592041016 2023-01-21 11:43:59.760202: step: 828/529, loss: 0.011141777038574219 2023-01-21 11:44:00.908525: step: 832/529, loss: 0.004449653439223766 2023-01-21 11:44:02.047275: step: 836/529, loss: 0.007259320933371782 2023-01-21 11:44:03.204114: step: 840/529, loss: 0.026501012966036797 2023-01-21 11:44:04.355710: step: 844/529, loss: 0.00048131943913176656 2023-01-21 11:44:05.469795: step: 848/529, loss: 0.24736252427101135 2023-01-21 11:44:06.611779: step: 852/529, loss: 0.0014541626442223787 2023-01-21 11:44:07.756902: step: 856/529, loss: 0.0007814407581463456 2023-01-21 11:44:08.902846: step: 860/529, loss: 0.031156539916992188 2023-01-21 11:44:10.058016: step: 864/529, loss: 0.03480873256921768 2023-01-21 11:44:11.177973: step: 868/529, loss: 0.1090371161699295 2023-01-21 11:44:12.328540: step: 872/529, loss: 0.0026908875443041325 2023-01-21 11:44:13.481420: step: 876/529, loss: 0.05614016205072403 2023-01-21 11:44:14.582409: step: 880/529, loss: 0.012283611111342907 2023-01-21 11:44:15.722376: step: 884/529, loss: 0.003698110580444336 2023-01-21 11:44:16.845980: step: 888/529, loss: 0.009189414791762829 2023-01-21 11:44:17.985046: step: 892/529, loss: 0.06579875946044922 2023-01-21 11:44:19.133310: step: 896/529, loss: 0.042984578758478165 2023-01-21 11:44:20.254158: step: 900/529, loss: 0.009263229556381702 2023-01-21 11:44:21.402237: step: 904/529, loss: 0.06771078705787659 2023-01-21 11:44:22.571406: step: 908/529, loss: 0.019165420904755592 2023-01-21 11:44:23.718995: step: 912/529, loss: 0.08352375030517578 2023-01-21 11:44:24.822050: step: 916/529, loss: 0.0021211623679846525 2023-01-21 11:44:25.984760: step: 920/529, loss: 0.009473037905991077 2023-01-21 11:44:27.125041: step: 924/529, loss: 0.007779979612678289 2023-01-21 11:44:28.308866: step: 928/529, loss: 0.021114541217684746 2023-01-21 11:44:29.452557: step: 932/529, loss: 0.04826173931360245 2023-01-21 11:44:30.596747: step: 936/529, loss: 0.011199855245649815 2023-01-21 11:44:31.726549: step: 940/529, loss: 0.05464959144592285 2023-01-21 11:44:32.852049: step: 944/529, loss: 0.42306995391845703 2023-01-21 11:44:34.009857: step: 948/529, loss: 0.006498241331428289 2023-01-21 11:44:35.130339: step: 952/529, loss: 0.0058394428342580795 2023-01-21 11:44:36.266437: step: 956/529, loss: 0.0027633667923510075 2023-01-21 11:44:37.407392: step: 960/529, loss: 0.00011596679541980848 2023-01-21 11:44:38.561596: step: 964/529, loss: 0.009939479641616344 2023-01-21 11:44:39.692944: step: 968/529, loss: 0.7047474980354309 2023-01-21 11:44:40.843257: step: 972/529, loss: 0.0026189805939793587 2023-01-21 11:44:41.955297: step: 976/529, loss: 0.05056390538811684 2023-01-21 11:44:43.099664: step: 980/529, loss: 0.007739257998764515 2023-01-21 11:44:44.241323: step: 984/529, loss: 0.024247217923402786 2023-01-21 11:44:45.355157: step: 988/529, loss: 0.0012052536476403475 2023-01-21 11:44:46.527031: step: 992/529, loss: 0.0018032074440270662 2023-01-21 11:44:47.668058: step: 996/529, loss: 0.00206336984410882 2023-01-21 11:44:48.798627: step: 1000/529, loss: 0.001304531004279852 2023-01-21 11:44:49.951108: step: 1004/529, loss: 0.03158550336956978 2023-01-21 11:44:51.083317: step: 1008/529, loss: 0.00036458970862440765 2023-01-21 11:44:52.218016: step: 1012/529, loss: 0.010820102877914906 2023-01-21 11:44:53.345252: step: 1016/529, loss: 0.0028075575828552246 2023-01-21 11:44:54.491145: step: 1020/529, loss: 0.02905893325805664 2023-01-21 11:44:55.658704: step: 1024/529, loss: 0.0280793197453022 2023-01-21 11:44:56.795779: step: 1028/529, loss: 0.03313121944665909 2023-01-21 11:44:57.955885: step: 1032/529, loss: 0.07287998497486115 2023-01-21 11:44:59.094303: step: 1036/529, loss: 0.043039895594120026 2023-01-21 11:45:00.261400: step: 1040/529, loss: 0.019349049776792526 2023-01-21 11:45:01.400621: step: 1044/529, loss: 0.005201817024499178 2023-01-21 11:45:02.535373: step: 1048/529, loss: 0.024669744074344635 2023-01-21 11:45:03.641545: step: 1052/529, loss: 0.01853400468826294 2023-01-21 11:45:04.760675: step: 1056/529, loss: 0.07856402546167374 2023-01-21 11:45:05.902230: step: 1060/529, loss: 0.34706515073776245 2023-01-21 11:45:07.022941: step: 1064/529, loss: 0.02746281772851944 2023-01-21 11:45:08.163595: step: 1068/529, loss: 0.015646077692508698 2023-01-21 11:45:09.287034: step: 1072/529, loss: 0.005643940065056086 2023-01-21 11:45:10.407605: step: 1076/529, loss: 0.0016124725807458162 2023-01-21 11:45:11.531122: step: 1080/529, loss: 0.00021800995455123484 2023-01-21 11:45:12.727207: step: 1084/529, loss: 0.012498665601015091 2023-01-21 11:45:13.872031: step: 1088/529, loss: 0.008881950750946999 2023-01-21 11:45:14.997662: step: 1092/529, loss: 0.002604580018669367 2023-01-21 11:45:16.155208: step: 1096/529, loss: 0.07128524780273438 2023-01-21 11:45:17.309007: step: 1100/529, loss: 0.016475200653076172 2023-01-21 11:45:18.426328: step: 1104/529, loss: 0.009147453121840954 2023-01-21 11:45:19.512771: step: 1108/529, loss: 0.03231959417462349 2023-01-21 11:45:20.611872: step: 1112/529, loss: 0.00045847895671613514 2023-01-21 11:45:21.776512: step: 1116/529, loss: 0.007363033480942249 2023-01-21 11:45:22.921106: step: 1120/529, loss: 0.01332855224609375 2023-01-21 11:45:24.067147: step: 1124/529, loss: 0.019433021545410156 2023-01-21 11:45:25.188778: step: 1128/529, loss: 0.0306535717099905 2023-01-21 11:45:26.340964: step: 1132/529, loss: 0.0385868065059185 2023-01-21 11:45:27.460979: step: 1136/529, loss: 0.013625050894916058 2023-01-21 11:45:28.602760: step: 1140/529, loss: 0.1016455665230751 2023-01-21 11:45:29.719779: step: 1144/529, loss: 0.00783233717083931 2023-01-21 11:45:30.868177: step: 1148/529, loss: 0.012146568857133389 2023-01-21 11:45:32.006481: step: 1152/529, loss: 0.0011615753173828125 2023-01-21 11:45:33.155905: step: 1156/529, loss: 0.01069488562643528 2023-01-21 11:45:34.286400: step: 1160/529, loss: 0.003141212509945035 2023-01-21 11:45:35.394182: step: 1164/529, loss: 0.00011200904555153102 2023-01-21 11:45:36.536009: step: 1168/529, loss: 0.013946629129350185 2023-01-21 11:45:37.662164: step: 1172/529, loss: 0.011359596624970436 2023-01-21 11:45:38.808048: step: 1176/529, loss: 0.06890201568603516 2023-01-21 11:45:39.937302: step: 1180/529, loss: 0.00794219970703125 2023-01-21 11:45:41.068338: step: 1184/529, loss: 0.015289593487977982 2023-01-21 11:45:42.222044: step: 1188/529, loss: 0.000607347465120256 2023-01-21 11:45:43.354654: step: 1192/529, loss: 0.0034267425071448088 2023-01-21 11:45:44.493068: step: 1196/529, loss: 8.468628220725805e-05 2023-01-21 11:45:45.626602: step: 1200/529, loss: 1.7029756307601929 2023-01-21 11:45:46.748909: step: 1204/529, loss: 0.6506485939025879 2023-01-21 11:45:47.884129: step: 1208/529, loss: 0.004114913754165173 2023-01-21 11:45:49.013617: step: 1212/529, loss: 0.03726482391357422 2023-01-21 11:45:50.136650: step: 1216/529, loss: 0.034063052386045456 2023-01-21 11:45:51.269613: step: 1220/529, loss: 0.011377430520951748 2023-01-21 11:45:52.411283: step: 1224/529, loss: 0.002095699543133378 2023-01-21 11:45:53.543174: step: 1228/529, loss: 0.0007552147144451737 2023-01-21 11:45:54.681945: step: 1232/529, loss: 0.002673411276191473 2023-01-21 11:45:55.840610: step: 1236/529, loss: 0.00013017654418945312 2023-01-21 11:45:56.960308: step: 1240/529, loss: 0.010288620367646217 2023-01-21 11:45:58.121623: step: 1244/529, loss: 0.00025959016056731343 2023-01-21 11:45:59.245473: step: 1248/529, loss: 0.0037863284815102816 2023-01-21 11:46:00.367844: step: 1252/529, loss: 0.016154861077666283 2023-01-21 11:46:01.516468: step: 1256/529, loss: 0.006078529637306929 2023-01-21 11:46:02.651278: step: 1260/529, loss: 0.007026052568107843 2023-01-21 11:46:03.806928: step: 1264/529, loss: 1.4209747860149946e-05 2023-01-21 11:46:04.949502: step: 1268/529, loss: 0.0015303611289709806 2023-01-21 11:46:06.051363: step: 1272/529, loss: 0.0051552774384617805 2023-01-21 11:46:07.184095: step: 1276/529, loss: 0.012936593033373356 2023-01-21 11:46:08.326443: step: 1280/529, loss: 0.011408329010009766 2023-01-21 11:46:09.502673: step: 1284/529, loss: 0.031004048883914948 2023-01-21 11:46:10.637115: step: 1288/529, loss: 0.0012442589504644275 2023-01-21 11:46:11.778144: step: 1292/529, loss: 0.007152080535888672 2023-01-21 11:46:12.947107: step: 1296/529, loss: 0.1203855499625206 2023-01-21 11:46:14.065667: step: 1300/529, loss: 0.00018138886662200093 2023-01-21 11:46:15.221240: step: 1304/529, loss: 0.0509122833609581 2023-01-21 11:46:16.375062: step: 1308/529, loss: 0.04141216352581978 2023-01-21 11:46:17.501156: step: 1312/529, loss: 0.037805940955877304 2023-01-21 11:46:18.635400: step: 1316/529, loss: 0.024106694385409355 2023-01-21 11:46:19.821248: step: 1320/529, loss: 1.8596649169921875e-05 2023-01-21 11:46:20.937203: step: 1324/529, loss: 0.05854291841387749 2023-01-21 11:46:22.060720: step: 1328/529, loss: 0.03648886829614639 2023-01-21 11:46:23.208150: step: 1332/529, loss: 0.0053764344193041325 2023-01-21 11:46:24.344793: step: 1336/529, loss: 0.004661750514060259 2023-01-21 11:46:25.479980: step: 1340/529, loss: 0.019359208643436432 2023-01-21 11:46:26.627110: step: 1344/529, loss: 0.005584716796875 2023-01-21 11:46:27.762214: step: 1348/529, loss: 0.0011871815659105778 2023-01-21 11:46:28.895348: step: 1352/529, loss: 1.5544890629826114e-05 2023-01-21 11:46:30.024394: step: 1356/529, loss: 0.014545058831572533 2023-01-21 11:46:31.168280: step: 1360/529, loss: 0.0009807586902752519 2023-01-21 11:46:32.316905: step: 1364/529, loss: 0.0021584511268883944 2023-01-21 11:46:33.459463: step: 1368/529, loss: 0.004087543580681086 2023-01-21 11:46:34.570917: step: 1372/529, loss: 0.0004984855768270791 2023-01-21 11:46:35.709742: step: 1376/529, loss: 0.004062843509018421 2023-01-21 11:46:36.877001: step: 1380/529, loss: 0.004050254821777344 2023-01-21 11:46:38.009476: step: 1384/529, loss: 0.02704143524169922 2023-01-21 11:46:39.125833: step: 1388/529, loss: 0.032195284962654114 2023-01-21 11:46:40.238501: step: 1392/529, loss: 0.06419945508241653 2023-01-21 11:46:41.351106: step: 1396/529, loss: 0.0022810937371104956 2023-01-21 11:46:42.482750: step: 1400/529, loss: 0.0012201309436932206 2023-01-21 11:46:43.610049: step: 1404/529, loss: 0.0005521774291992188 2023-01-21 11:46:44.770915: step: 1408/529, loss: 0.017676925286650658 2023-01-21 11:46:45.912271: step: 1412/529, loss: 0.06240854412317276 2023-01-21 11:46:47.078866: step: 1416/529, loss: 0.0006165981176309288 2023-01-21 11:46:48.238996: step: 1420/529, loss: 0.017447376623749733 2023-01-21 11:46:49.349895: step: 1424/529, loss: 0.00034370424691587687 2023-01-21 11:46:50.487551: step: 1428/529, loss: 0.0044151307083666325 2023-01-21 11:46:51.620703: step: 1432/529, loss: 0.027493858709931374 2023-01-21 11:46:52.756191: step: 1436/529, loss: 0.09043464809656143 2023-01-21 11:46:53.903451: step: 1440/529, loss: 0.04465799406170845 2023-01-21 11:46:55.027884: step: 1444/529, loss: 0.007709980010986328 2023-01-21 11:46:56.223126: step: 1448/529, loss: 0.00026607513427734375 2023-01-21 11:46:57.362641: step: 1452/529, loss: 0.00015840530977584422 2023-01-21 11:46:58.480444: step: 1456/529, loss: 0.0013560295337811112 2023-01-21 11:46:59.620564: step: 1460/529, loss: 0.0003067016659770161 2023-01-21 11:47:00.736229: step: 1464/529, loss: 0.004874229431152344 2023-01-21 11:47:01.868029: step: 1468/529, loss: 0.00027561187744140625 2023-01-21 11:47:03.005054: step: 1472/529, loss: 0.004246044438332319 2023-01-21 11:47:04.140772: step: 1476/529, loss: 0.004613971803337336 2023-01-21 11:47:05.263611: step: 1480/529, loss: 0.024268915876746178 2023-01-21 11:47:06.423940: step: 1484/529, loss: 0.00043144228402525187 2023-01-21 11:47:07.554918: step: 1488/529, loss: 0.115624338388443 2023-01-21 11:47:08.663506: step: 1492/529, loss: 0.0001656532404012978 2023-01-21 11:47:09.807643: step: 1496/529, loss: 0.0018731595482677221 2023-01-21 11:47:10.945826: step: 1500/529, loss: 0.005102920811623335 2023-01-21 11:47:12.078766: step: 1504/529, loss: 0.014412689954042435 2023-01-21 11:47:13.240932: step: 1508/529, loss: 7.486343383789062e-05 2023-01-21 11:47:14.366913: step: 1512/529, loss: 0.039426807314157486 2023-01-21 11:47:15.498177: step: 1516/529, loss: 0.0012891768710687757 2023-01-21 11:47:16.630639: step: 1520/529, loss: 0.013417769223451614 2023-01-21 11:47:17.786825: step: 1524/529, loss: 0.00032291412935592234 2023-01-21 11:47:18.901542: step: 1528/529, loss: 0.04516134411096573 2023-01-21 11:47:20.067438: step: 1532/529, loss: 0.028725624084472656 2023-01-21 11:47:21.213348: step: 1536/529, loss: 0.01963825337588787 2023-01-21 11:47:22.332409: step: 1540/529, loss: 0.0003147602255921811 2023-01-21 11:47:23.480712: step: 1544/529, loss: 0.39326995611190796 2023-01-21 11:47:24.618555: step: 1548/529, loss: 0.09417486190795898 2023-01-21 11:47:25.765524: step: 1552/529, loss: 0.012935166247189045 2023-01-21 11:47:26.938807: step: 1556/529, loss: 0.046253204345703125 2023-01-21 11:47:28.054355: step: 1560/529, loss: 0.025860263034701347 2023-01-21 11:47:29.172616: step: 1564/529, loss: 0.04771614074707031 2023-01-21 11:47:30.300018: step: 1568/529, loss: 1.1444091796875e-05 2023-01-21 11:47:31.456136: step: 1572/529, loss: 0.05708504095673561 2023-01-21 11:47:32.614968: step: 1576/529, loss: 0.00832824781537056 2023-01-21 11:47:33.740063: step: 1580/529, loss: 0.009956836700439453 2023-01-21 11:47:34.879817: step: 1584/529, loss: 0.013566303998231888 2023-01-21 11:47:36.029094: step: 1588/529, loss: 0.05560264736413956 2023-01-21 11:47:37.196094: step: 1592/529, loss: 0.051306914538145065 2023-01-21 11:47:38.350594: step: 1596/529, loss: 0.020345641300082207 2023-01-21 11:47:39.501378: step: 1600/529, loss: 0.00832357443869114 2023-01-21 11:47:40.611763: step: 1604/529, loss: 0.008730125613510609 2023-01-21 11:47:41.783543: step: 1608/529, loss: 0.0159741397947073 2023-01-21 11:47:42.931462: step: 1612/529, loss: 0.02404160611331463 2023-01-21 11:47:44.062800: step: 1616/529, loss: 0.24515514075756073 2023-01-21 11:47:45.206426: step: 1620/529, loss: 0.00027379990206100047 2023-01-21 11:47:46.317897: step: 1624/529, loss: 0.002193927764892578 2023-01-21 11:47:47.456571: step: 1628/529, loss: 0.0055561065673828125 2023-01-21 11:47:48.603836: step: 1632/529, loss: 0.0007213592762127519 2023-01-21 11:47:49.759272: step: 1636/529, loss: 0.018894100561738014 2023-01-21 11:47:50.929561: step: 1640/529, loss: 0.009850788861513138 2023-01-21 11:47:52.065500: step: 1644/529, loss: 0.0034782409202307463 2023-01-21 11:47:53.193329: step: 1648/529, loss: 0.0008298158645629883 2023-01-21 11:47:54.342809: step: 1652/529, loss: 0.0031703950371593237 2023-01-21 11:47:55.471126: step: 1656/529, loss: 0.03525600582361221 2023-01-21 11:47:56.593500: step: 1660/529, loss: 0.027184296399354935 2023-01-21 11:47:57.755088: step: 1664/529, loss: 0.06109027937054634 2023-01-21 11:47:58.878045: step: 1668/529, loss: 0.04244265705347061 2023-01-21 11:48:00.014270: step: 1672/529, loss: 0.0003202170191798359 2023-01-21 11:48:01.133227: step: 1676/529, loss: 0.00039196014404296875 2023-01-21 11:48:02.239617: step: 1680/529, loss: 9.90867629298009e-05 2023-01-21 11:48:03.375825: step: 1684/529, loss: 0.036699675023555756 2023-01-21 11:48:04.516465: step: 1688/529, loss: 0.0035342215560376644 2023-01-21 11:48:05.658056: step: 1692/529, loss: 0.02523527294397354 2023-01-21 11:48:06.804186: step: 1696/529, loss: 0.14827804267406464 2023-01-21 11:48:07.940809: step: 1700/529, loss: 0.016812516376376152 2023-01-21 11:48:09.070404: step: 1704/529, loss: 0.007227111142128706 2023-01-21 11:48:10.182432: step: 1708/529, loss: 0.06744912266731262 2023-01-21 11:48:11.306206: step: 1712/529, loss: 0.0082855224609375 2023-01-21 11:48:12.457937: step: 1716/529, loss: 0.05597548559308052 2023-01-21 11:48:13.587870: step: 1720/529, loss: 0.0076354979537427425 2023-01-21 11:48:14.738797: step: 1724/529, loss: 0.013385677710175514 2023-01-21 11:48:15.858016: step: 1728/529, loss: 0.14796094596385956 2023-01-21 11:48:16.988617: step: 1732/529, loss: 0.03515832498669624 2023-01-21 11:48:18.123511: step: 1736/529, loss: 0.01655750349164009 2023-01-21 11:48:19.297882: step: 1740/529, loss: 0.0011521339183673263 2023-01-21 11:48:20.415607: step: 1744/529, loss: 0.06231732666492462 2023-01-21 11:48:21.549082: step: 1748/529, loss: 0.002077388809993863 2023-01-21 11:48:22.696628: step: 1752/529, loss: 0.030359555035829544 2023-01-21 11:48:23.803795: step: 1756/529, loss: 0.016997702419757843 2023-01-21 11:48:24.933010: step: 1760/529, loss: 0.054388999938964844 2023-01-21 11:48:26.069375: step: 1764/529, loss: 0.03872261196374893 2023-01-21 11:48:27.188444: step: 1768/529, loss: 0.00091552734375 2023-01-21 11:48:28.331443: step: 1772/529, loss: 3.80516066798009e-05 2023-01-21 11:48:29.501027: step: 1776/529, loss: 0.008639144711196423 2023-01-21 11:48:30.644015: step: 1780/529, loss: 0.0003904342884197831 2023-01-21 11:48:31.775457: step: 1784/529, loss: 0.012933206744492054 2023-01-21 11:48:32.895787: step: 1788/529, loss: 0.03966650739312172 2023-01-21 11:48:34.042683: step: 1792/529, loss: 0.02050151862204075 2023-01-21 11:48:35.151395: step: 1796/529, loss: 0.0016991615993902087 2023-01-21 11:48:36.298575: step: 1800/529, loss: 0.05046787112951279 2023-01-21 11:48:37.443366: step: 1804/529, loss: 0.058178067207336426 2023-01-21 11:48:38.567407: step: 1808/529, loss: 0.004039192106574774 2023-01-21 11:48:39.703758: step: 1812/529, loss: 0.04495878145098686 2023-01-21 11:48:40.810226: step: 1816/529, loss: 3.2711028325138614e-05 2023-01-21 11:48:41.935586: step: 1820/529, loss: 0.0004555195337161422 2023-01-21 11:48:43.074174: step: 1824/529, loss: 0.014361239038407803 2023-01-21 11:48:44.219830: step: 1828/529, loss: 0.006221866235136986 2023-01-21 11:48:45.330069: step: 1832/529, loss: 0.0017052650218829513 2023-01-21 11:48:46.446114: step: 1836/529, loss: 0.011165808886289597 2023-01-21 11:48:47.584945: step: 1840/529, loss: 0.07832060009241104 2023-01-21 11:48:48.744659: step: 1844/529, loss: 0.010797119699418545 2023-01-21 11:48:49.907605: step: 1848/529, loss: 0.03523902967572212 2023-01-21 11:48:51.051888: step: 1852/529, loss: 0.04129810631275177 2023-01-21 11:48:52.196320: step: 1856/529, loss: 0.016043759882450104 2023-01-21 11:48:53.332992: step: 1860/529, loss: 0.020298006013035774 2023-01-21 11:48:54.482988: step: 1864/529, loss: 0.007849693298339844 2023-01-21 11:48:55.605640: step: 1868/529, loss: 0.07497768849134445 2023-01-21 11:48:56.758101: step: 1872/529, loss: 0.01386184711009264 2023-01-21 11:48:57.866668: step: 1876/529, loss: 0.0025489807594567537 2023-01-21 11:48:59.010647: step: 1880/529, loss: 0.034453582018613815 2023-01-21 11:49:00.128273: step: 1884/529, loss: 0.017970848828554153 2023-01-21 11:49:01.265825: step: 1888/529, loss: 0.04123726114630699 2023-01-21 11:49:02.397166: step: 1892/529, loss: 0.004308212082833052 2023-01-21 11:49:03.512629: step: 1896/529, loss: 0.004065132234245539 2023-01-21 11:49:04.669374: step: 1900/529, loss: 0.2575257122516632 2023-01-21 11:49:05.815510: step: 1904/529, loss: 0.034539032727479935 2023-01-21 11:49:06.918757: step: 1908/529, loss: 0.0095246322453022 2023-01-21 11:49:08.057574: step: 1912/529, loss: 0.10999203473329544 2023-01-21 11:49:09.217267: step: 1916/529, loss: 0.004568958189338446 2023-01-21 11:49:10.346450: step: 1920/529, loss: 0.006252670660614967 2023-01-21 11:49:11.482681: step: 1924/529, loss: 0.0017265320057049394 2023-01-21 11:49:12.668998: step: 1928/529, loss: 0.015929508954286575 2023-01-21 11:49:13.793930: step: 1932/529, loss: 0.013738633133471012 2023-01-21 11:49:14.948639: step: 1936/529, loss: 0.04503937065601349 2023-01-21 11:49:16.079925: step: 1940/529, loss: 0.16811342537403107 2023-01-21 11:49:17.200191: step: 1944/529, loss: 0.01735210418701172 2023-01-21 11:49:18.303493: step: 1948/529, loss: 0.017446376383304596 2023-01-21 11:49:19.426251: step: 1952/529, loss: 0.0003025055048055947 2023-01-21 11:49:20.557121: step: 1956/529, loss: 0.06208610534667969 2023-01-21 11:49:21.682439: step: 1960/529, loss: 0.039640095084905624 2023-01-21 11:49:22.805459: step: 1964/529, loss: 0.21908530592918396 2023-01-21 11:49:23.952388: step: 1968/529, loss: 0.011944960802793503 2023-01-21 11:49:25.062508: step: 1972/529, loss: 0.02257833629846573 2023-01-21 11:49:26.190729: step: 1976/529, loss: 0.0003666877746582031 2023-01-21 11:49:27.343668: step: 1980/529, loss: 0.05057945102453232 2023-01-21 11:49:28.477100: step: 1984/529, loss: 0.007708358578383923 2023-01-21 11:49:29.630633: step: 1988/529, loss: 0.0019529343117028475 2023-01-21 11:49:30.777479: step: 1992/529, loss: 0.02591877058148384 2023-01-21 11:49:31.914989: step: 1996/529, loss: 0.0037382126320153475 2023-01-21 11:49:33.042965: step: 2000/529, loss: 0.01462850533425808 2023-01-21 11:49:34.170132: step: 2004/529, loss: 0.05969696491956711 2023-01-21 11:49:35.347832: step: 2008/529, loss: 0.009320450015366077 2023-01-21 11:49:36.488887: step: 2012/529, loss: 0.008641052059829235 2023-01-21 11:49:37.639335: step: 2016/529, loss: 0.016835499554872513 2023-01-21 11:49:38.765769: step: 2020/529, loss: 0.01511244848370552 2023-01-21 11:49:39.917917: step: 2024/529, loss: 0.03141536936163902 2023-01-21 11:49:41.041824: step: 2028/529, loss: 0.006036281585693359 2023-01-21 11:49:42.205040: step: 2032/529, loss: 0.0009674073080532253 2023-01-21 11:49:43.312077: step: 2036/529, loss: 0.001789951347745955 2023-01-21 11:49:44.464552: step: 2040/529, loss: 0.02393512800335884 2023-01-21 11:49:45.610405: step: 2044/529, loss: 0.0005800247308798134 2023-01-21 11:49:46.734516: step: 2048/529, loss: 0.007439422886818647 2023-01-21 11:49:47.848803: step: 2052/529, loss: 0.00880889967083931 2023-01-21 11:49:48.994635: step: 2056/529, loss: 0.0017215729458257556 2023-01-21 11:49:50.116775: step: 2060/529, loss: 0.0005592346424236894 2023-01-21 11:49:51.248085: step: 2064/529, loss: 0.028040220960974693 2023-01-21 11:49:52.386610: step: 2068/529, loss: 0.04548969492316246 2023-01-21 11:49:53.512182: step: 2072/529, loss: 0.00485496548935771 2023-01-21 11:49:54.649711: step: 2076/529, loss: 0.007137966342270374 2023-01-21 11:49:55.779827: step: 2080/529, loss: 0.08839602023363113 2023-01-21 11:49:56.909503: step: 2084/529, loss: 0.12057837843894958 2023-01-21 11:49:58.048952: step: 2088/529, loss: 0.05204325169324875 2023-01-21 11:49:59.201869: step: 2092/529, loss: 0.0008857727516442537 2023-01-21 11:50:00.324804: step: 2096/529, loss: 0.04516306146979332 2023-01-21 11:50:01.512340: step: 2100/529, loss: 0.07876624912023544 2023-01-21 11:50:02.648517: step: 2104/529, loss: 0.020334243774414062 2023-01-21 11:50:03.790266: step: 2108/529, loss: -3.910064606316155e-06 2023-01-21 11:50:04.915916: step: 2112/529, loss: 0.022933006286621094 2023-01-21 11:50:06.087621: step: 2116/529, loss: 0.006758618168532848 ================================================== Loss: 0.034 -------------------- Dev: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Test: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Chinese: {'event': {'p': 0.6071428571428571, 'r': 0.9444444444444444, 'f1': 0.7391304347826088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} 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': 18} Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 19 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:50:55.229841: step: 4/529, loss: 0.0017229558434337378 2023-01-21 11:50:56.365477: step: 8/529, loss: 0.00012750625319313258 2023-01-21 11:50:57.474505: step: 12/529, loss: 0.0063738347962498665 2023-01-21 11:50:58.607293: step: 16/529, loss: 0.016020966693758965 2023-01-21 11:50:59.758739: step: 20/529, loss: 0.004394107963889837 2023-01-21 11:51:00.907368: step: 24/529, loss: 0.2875579595565796 2023-01-21 11:51:02.032597: step: 28/529, loss: 0.0014423370594158769 2023-01-21 11:51:03.209593: step: 32/529, loss: 0.026410864666104317 2023-01-21 11:51:04.364890: step: 36/529, loss: 0.058995820581912994 2023-01-21 11:51:05.520431: step: 40/529, loss: 0.022314071655273438 2023-01-21 11:51:06.686515: step: 44/529, loss: 0.011196709237992764 2023-01-21 11:51:07.806397: step: 48/529, loss: 0.0017934800125658512 2023-01-21 11:51:08.950870: step: 52/529, loss: 0.0030163764022290707 2023-01-21 11:51:10.108469: step: 56/529, loss: 0.014001942239701748 2023-01-21 11:51:11.206867: step: 60/529, loss: 0.0009846300818026066 2023-01-21 11:51:12.311035: step: 64/529, loss: 0.00209465017542243 2023-01-21 11:51:13.444551: step: 68/529, loss: 0.03479714319109917 2023-01-21 11:51:14.574283: step: 72/529, loss: 0.00035190582275390625 2023-01-21 11:51:15.712075: step: 76/529, loss: 0.0031116483733057976 2023-01-21 11:51:16.847839: step: 80/529, loss: 0.00028715134249068797 2023-01-21 11:51:17.996225: step: 84/529, loss: 0.00045452118501998484 2023-01-21 11:51:19.139173: step: 88/529, loss: 0.03635597229003906 2023-01-21 11:51:20.269475: step: 92/529, loss: 0.03024139441549778 2023-01-21 11:51:21.411721: step: 96/529, loss: 0.013460826128721237 2023-01-21 11:51:22.536767: step: 100/529, loss: 0.0026773454155772924 2023-01-21 11:51:23.680751: step: 104/529, loss: 0.007898807525634766 2023-01-21 11:51:24.830193: step: 108/529, loss: 0.0016506194369867444 2023-01-21 11:51:25.978392: step: 112/529, loss: 0.010470437817275524 2023-01-21 11:51:27.140016: step: 116/529, loss: 0.00010795594425871968 2023-01-21 11:51:28.257736: step: 120/529, loss: 8.926391456043348e-05 2023-01-21 11:51:29.371477: step: 124/529, loss: 0.04072628170251846 2023-01-21 11:51:30.523009: step: 128/529, loss: 0.029494667425751686 2023-01-21 11:51:31.643793: step: 132/529, loss: 0.0008565902826376259 2023-01-21 11:51:32.789343: step: 136/529, loss: 3.4046173823298886e-05 2023-01-21 11:51:33.907360: step: 140/529, loss: 0.016927719116210938 2023-01-21 11:51:35.065290: step: 144/529, loss: 0.10821304470300674 2023-01-21 11:51:36.192189: step: 148/529, loss: 0.035622384399175644 2023-01-21 11:51:37.359362: step: 152/529, loss: 0.04031829908490181 2023-01-21 11:51:38.492044: step: 156/529, loss: 0.017365645617246628 2023-01-21 11:51:39.646220: step: 160/529, loss: 0.035645198076963425 2023-01-21 11:51:40.790806: step: 164/529, loss: 0.00818481482565403 2023-01-21 11:51:41.935553: step: 168/529, loss: 0.010926151648163795 2023-01-21 11:51:43.085771: step: 172/529, loss: 0.06105957180261612 2023-01-21 11:51:44.233217: step: 176/529, loss: 0.01154251117259264 2023-01-21 11:51:45.379090: step: 180/529, loss: 0.00013365744962356985 2023-01-21 11:51:46.497423: step: 184/529, loss: 0.00179967877920717 2023-01-21 11:51:47.607984: step: 188/529, loss: 3.337860107421875e-05 2023-01-21 11:51:48.751274: step: 192/529, loss: 0.045339204370975494 2023-01-21 11:51:49.883566: step: 196/529, loss: 0.012264168821275234 2023-01-21 11:51:51.001136: step: 200/529, loss: 0.026839256286621094 2023-01-21 11:51:52.124184: step: 204/529, loss: 0.00029506682767532766 2023-01-21 11:51:53.230924: step: 208/529, loss: 0.011596537195146084 2023-01-21 11:51:54.380605: step: 212/529, loss: 0.004087972454726696 2023-01-21 11:51:55.526901: step: 216/529, loss: 0.01124496478587389 2023-01-21 11:51:56.696560: step: 220/529, loss: 0.06577281653881073 2023-01-21 11:51:57.824830: step: 224/529, loss: 0.0023153305519372225 2023-01-21 11:51:58.965509: step: 228/529, loss: 0.008156394585967064 2023-01-21 11:52:00.101627: step: 232/529, loss: 0.01878051646053791 2023-01-21 11:52:01.218485: step: 236/529, loss: 0.012361167930066586 2023-01-21 11:52:02.370921: step: 240/529, loss: 0.010026931762695312 2023-01-21 11:52:03.535450: step: 244/529, loss: 0.0016252041095867753 2023-01-21 11:52:04.669945: step: 248/529, loss: 0.02277221716940403 2023-01-21 11:52:05.846446: step: 252/529, loss: 0.013946819119155407 2023-01-21 11:52:07.005025: step: 256/529, loss: 0.021763229742646217 2023-01-21 11:52:08.124807: step: 260/529, loss: 0.020238209515810013 2023-01-21 11:52:09.254157: step: 264/529, loss: 0.002921962644904852 2023-01-21 11:52:10.387170: step: 268/529, loss: 0.0015006065368652344 2023-01-21 11:52:11.533809: step: 272/529, loss: 0.004385757260024548 2023-01-21 11:52:12.659167: step: 276/529, loss: 0.00016307830810546875 2023-01-21 11:52:13.770857: step: 280/529, loss: 0.010633468627929688 2023-01-21 11:52:14.909976: step: 284/529, loss: 1.3256072634248994e-05 2023-01-21 11:52:16.068454: step: 288/529, loss: 0.01273269671946764 2023-01-21 11:52:17.177784: step: 292/529, loss: 0.0026335238944739103 2023-01-21 11:52:18.296598: step: 296/529, loss: 0.019414758309721947 2023-01-21 11:52:19.402848: step: 300/529, loss: 0.00013175011554267257 2023-01-21 11:52:20.537750: step: 304/529, loss: 0.018715476617217064 2023-01-21 11:52:21.651160: step: 308/529, loss: 0.028333188965916634 2023-01-21 11:52:22.785569: step: 312/529, loss: 0.0027208328247070312 2023-01-21 11:52:23.922116: step: 316/529, loss: 0.005942726042121649 2023-01-21 11:52:25.067544: step: 320/529, loss: 0.028026580810546875 2023-01-21 11:52:26.211848: step: 324/529, loss: 0.00010776520502986386 2023-01-21 11:52:27.355405: step: 328/529, loss: 0.00025968553381972015 2023-01-21 11:52:28.488683: step: 332/529, loss: 5.645752389682457e-05 2023-01-21 11:52:29.621142: step: 336/529, loss: 0.005469226744025946 2023-01-21 11:52:30.767604: step: 340/529, loss: 0.00493278494104743 2023-01-21 11:52:31.891041: step: 344/529, loss: 0.0223814956843853 2023-01-21 11:52:33.010501: step: 348/529, loss: 0.04321708530187607 2023-01-21 11:52:34.128979: step: 352/529, loss: 0.007588577456772327 2023-01-21 11:52:35.280592: step: 356/529, loss: 0.0003215789911337197 2023-01-21 11:52:36.418990: step: 360/529, loss: 0.00026835798053070903 2023-01-21 11:52:37.528415: step: 364/529, loss: 0.029219914227724075 2023-01-21 11:52:38.636009: step: 368/529, loss: 0.0019974708557128906 2023-01-21 11:52:39.735452: step: 372/529, loss: 0.0020823481027036905 2023-01-21 11:52:40.900316: step: 376/529, loss: 0.0012748718727380037 2023-01-21 11:52:42.045560: step: 380/529, loss: 0.010187816806137562 2023-01-21 11:52:43.163751: step: 384/529, loss: 0.011214351281523705 2023-01-21 11:52:44.283414: step: 388/529, loss: 0.007705736439675093 2023-01-21 11:52:45.408429: step: 392/529, loss: 0.11236324161291122 2023-01-21 11:52:46.581004: step: 396/529, loss: 0.020964432507753372 2023-01-21 11:52:47.731196: step: 400/529, loss: 0.029291262850165367 2023-01-21 11:52:48.861706: step: 404/529, loss: 0.042046405375003815 2023-01-21 11:52:49.972906: step: 408/529, loss: 0.001918792724609375 2023-01-21 11:52:51.094247: step: 412/529, loss: 0.044573210179805756 2023-01-21 11:52:52.206808: step: 416/529, loss: 0.00019841194443870336 2023-01-21 11:52:53.310667: step: 420/529, loss: 0.015165758319199085 2023-01-21 11:52:54.396989: step: 424/529, loss: 3.0326844353112392e-05 2023-01-21 11:52:55.541575: step: 428/529, loss: 0.003480339189991355 2023-01-21 11:52:56.649167: step: 432/529, loss: 0.015481377020478249 2023-01-21 11:52:57.786446: step: 436/529, loss: 0.052968598902225494 2023-01-21 11:52:58.929217: step: 440/529, loss: 0.0530790314078331 2023-01-21 11:53:00.067680: step: 444/529, loss: 0.017420196905732155 2023-01-21 11:53:01.209088: step: 448/529, loss: 0.0022007941734045744 2023-01-21 11:53:02.365006: step: 452/529, loss: 0.06569195538759232 2023-01-21 11:53:03.495068: step: 456/529, loss: 0.0030723572708666325 2023-01-21 11:53:04.647928: step: 460/529, loss: 0.0022013664711266756 2023-01-21 11:53:05.769169: step: 464/529, loss: 0.0004341602325439453 2023-01-21 11:53:06.891458: step: 468/529, loss: 0.015452575869858265 2023-01-21 11:53:07.980150: step: 472/529, loss: 0.00043087004451081157 2023-01-21 11:53:09.101854: step: 476/529, loss: 0.00425376882776618 2023-01-21 11:53:10.225114: step: 480/529, loss: 0.02633051946759224 2023-01-21 11:53:11.338637: step: 484/529, loss: 0.0038394927978515625 2023-01-21 11:53:12.475178: step: 488/529, loss: 0.00047283171443268657 2023-01-21 11:53:13.601559: step: 492/529, loss: 0.12710438668727875 2023-01-21 11:53:14.730042: step: 496/529, loss: 0.0012802124256268144 2023-01-21 11:53:15.851997: step: 500/529, loss: 0.06940305233001709 2023-01-21 11:53:16.982661: step: 504/529, loss: 0.010061264038085938 2023-01-21 11:53:18.098379: step: 508/529, loss: 0.020330429077148438 2023-01-21 11:53:19.235227: step: 512/529, loss: 0.006386184599250555 2023-01-21 11:53:20.328806: step: 516/529, loss: 2.384185791015625e-05 2023-01-21 11:53:21.451706: step: 520/529, loss: 0.002096462296321988 2023-01-21 11:53:22.593723: step: 524/529, loss: 0.013880539685487747 2023-01-21 11:53:23.690960: step: 528/529, loss: 0.0001996994105866179 2023-01-21 11:53:24.814682: step: 532/529, loss: 0.005055236630141735 2023-01-21 11:53:25.962733: step: 536/529, loss: 0.04180869832634926 2023-01-21 11:53:27.084816: step: 540/529, loss: 6.442070298362523e-05 2023-01-21 11:53:28.290899: step: 544/529, loss: 0.039975740015506744 2023-01-21 11:53:29.430440: step: 548/529, loss: 0.0005557060358114541 2023-01-21 11:53:30.544139: step: 552/529, loss: 0.01630993001163006 2023-01-21 11:53:31.691278: step: 556/529, loss: 0.0002898216189350933 2023-01-21 11:53:32.829242: step: 560/529, loss: 0.0024632455315440893 2023-01-21 11:53:33.959529: step: 564/529, loss: 0.028363991528749466 2023-01-21 11:53:35.089233: step: 568/529, loss: 0.0022933962754905224 2023-01-21 11:53:36.216606: step: 572/529, loss: 0.0013566971756517887 2023-01-21 11:53:37.371082: step: 576/529, loss: 0.00023632051306776702 2023-01-21 11:53:38.502753: step: 580/529, loss: 0.0027883530128747225 2023-01-21 11:53:39.633873: step: 584/529, loss: 0.00439796457067132 2023-01-21 11:53:40.767203: step: 588/529, loss: 0.0018667221302166581 2023-01-21 11:53:41.883892: step: 592/529, loss: 0.0052741048857569695 2023-01-21 11:53:43.056984: step: 596/529, loss: 0.005711555480957031 2023-01-21 11:53:44.188140: step: 600/529, loss: 0.0007303237798623741 2023-01-21 11:53:45.307823: step: 604/529, loss: 0.0024530410300940275 2023-01-21 11:53:46.442265: step: 608/529, loss: 0.0002964019950013608 2023-01-21 11:53:47.568708: step: 612/529, loss: 0.0032460212241858244 2023-01-21 11:53:48.689898: step: 616/529, loss: 0.0027873993385583162 2023-01-21 11:53:49.815297: step: 620/529, loss: 0.0004849433898925781 2023-01-21 11:53:50.952870: step: 624/529, loss: 0.13007669150829315 2023-01-21 11:53:52.074958: step: 628/529, loss: 0.0013151407474651933 2023-01-21 11:53:53.209924: step: 632/529, loss: 0.0029778482858091593 2023-01-21 11:53:54.324659: step: 636/529, loss: 0.0012725830310955644 2023-01-21 11:53:55.456310: step: 640/529, loss: 0.0347936637699604 2023-01-21 11:53:56.636612: step: 644/529, loss: 0.018931293860077858 2023-01-21 11:53:57.775755: step: 648/529, loss: 9.536756806483027e-08 2023-01-21 11:53:58.917384: step: 652/529, loss: 0.027465151622891426 2023-01-21 11:54:00.043691: step: 656/529, loss: 0.0004594326310325414 2023-01-21 11:54:01.161189: step: 660/529, loss: 0.005196523852646351 2023-01-21 11:54:02.306438: step: 664/529, loss: 0.048226453363895416 2023-01-21 11:54:03.418840: step: 668/529, loss: 0.027511024847626686 2023-01-21 11:54:04.568211: step: 672/529, loss: 0.02921733818948269 2023-01-21 11:54:05.715312: step: 676/529, loss: 0.00010108948481502011 2023-01-21 11:54:06.844347: step: 680/529, loss: 0.03141040727496147 2023-01-21 11:54:07.976354: step: 684/529, loss: 0.03327608108520508 2023-01-21 11:54:09.110168: step: 688/529, loss: 0.004957771860063076 2023-01-21 11:54:10.221087: step: 692/529, loss: 0.0009335517534054816 2023-01-21 11:54:11.360851: step: 696/529, loss: 0.0051246643997728825 2023-01-21 11:54:12.525019: step: 700/529, loss: 0.047982312738895416 2023-01-21 11:54:13.644513: step: 704/529, loss: 0.01125488243997097 2023-01-21 11:54:14.779560: step: 708/529, loss: 0.00027751922607421875 2023-01-21 11:54:15.872846: step: 712/529, loss: 0.14670322835445404 2023-01-21 11:54:17.001104: step: 716/529, loss: 0.0011331557761877775 2023-01-21 11:54:18.116219: step: 720/529, loss: 0.0857355147600174 2023-01-21 11:54:19.241135: step: 724/529, loss: 0.0027174949645996094 2023-01-21 11:54:20.373998: step: 728/529, loss: 0.4464663565158844 2023-01-21 11:54:21.521486: step: 732/529, loss: 0.016971588134765625 2023-01-21 11:54:22.663802: step: 736/529, loss: 0.09289821982383728 2023-01-21 11:54:23.795627: step: 740/529, loss: 0.015959572046995163 2023-01-21 11:54:24.963613: step: 744/529, loss: 0.043720148503780365 2023-01-21 11:54:26.087421: step: 748/529, loss: 2.822876012942288e-05 2023-01-21 11:54:27.236253: step: 752/529, loss: 0.0005828857538290322 2023-01-21 11:54:28.373354: step: 756/529, loss: 0.07199706882238388 2023-01-21 11:54:29.500943: step: 760/529, loss: 0.03749069944024086 2023-01-21 11:54:30.643791: step: 764/529, loss: 0.09189119189977646 2023-01-21 11:54:31.746227: step: 768/529, loss: 1.544952465337701e-05 2023-01-21 11:54:32.880837: step: 772/529, loss: 0.03544311970472336 2023-01-21 11:54:34.054907: step: 776/529, loss: 0.004003333859145641 2023-01-21 11:54:35.165195: step: 780/529, loss: 0.007344531826674938 2023-01-21 11:54:36.279731: step: 784/529, loss: 0.0013620376121252775 2023-01-21 11:54:37.450977: step: 788/529, loss: 0.06679697334766388 2023-01-21 11:54:38.613871: step: 792/529, loss: 0.006496143061667681 2023-01-21 11:54:39.738513: step: 796/529, loss: 0.001853132271207869 2023-01-21 11:54:40.874924: step: 800/529, loss: 0.0025047301314771175 2023-01-21 11:54:42.010188: step: 804/529, loss: 0.01312112808227539 2023-01-21 11:54:43.141302: step: 808/529, loss: 0.0004287719784770161 2023-01-21 11:54:44.279696: step: 812/529, loss: 0.03422670438885689 2023-01-21 11:54:45.405569: step: 816/529, loss: 0.0035051347222179174 2023-01-21 11:54:46.525863: step: 820/529, loss: 0.005028247833251953 2023-01-21 11:54:47.627252: step: 824/529, loss: 5.817413330078125e-05 2023-01-21 11:54:48.739197: step: 828/529, loss: 0.021057654172182083 2023-01-21 11:54:49.859336: step: 832/529, loss: 0.878865122795105 2023-01-21 11:54:50.990040: step: 836/529, loss: 0.00186500558629632 2023-01-21 11:54:52.102248: step: 840/529, loss: 0.12342415004968643 2023-01-21 11:54:53.223949: step: 844/529, loss: 0.020163346081972122 2023-01-21 11:54:54.337852: step: 848/529, loss: 0.0003006935294251889 2023-01-21 11:54:55.464878: step: 852/529, loss: 0.01709880866110325 2023-01-21 11:54:56.607449: step: 856/529, loss: 0.00023312569828704 2023-01-21 11:54:57.756512: step: 860/529, loss: 0.00040693284245207906 2023-01-21 11:54:58.866332: step: 864/529, loss: 0.015584086999297142 2023-01-21 11:55:00.028048: step: 868/529, loss: 0.0018819809192791581 2023-01-21 11:55:01.160604: step: 872/529, loss: 0.038526538759469986 2023-01-21 11:55:02.286854: step: 876/529, loss: 0.00029811859712935984 2023-01-21 11:55:03.406675: step: 880/529, loss: 0.008828258141875267 2023-01-21 11:55:04.522616: step: 884/529, loss: 7.171630568336695e-05 2023-01-21 11:55:05.642277: step: 888/529, loss: 0.0012014388339594007 2023-01-21 11:55:06.788054: step: 892/529, loss: 0.0314946174621582 2023-01-21 11:55:07.918050: step: 896/529, loss: 0.0072608948685228825 2023-01-21 11:55:09.052943: step: 900/529, loss: 0.004921531770378351 2023-01-21 11:55:10.195490: step: 904/529, loss: 0.0023704529739916325 2023-01-21 11:55:11.340799: step: 908/529, loss: 0.00011987685866188258 2023-01-21 11:55:12.461023: step: 912/529, loss: 0.005391979124397039 2023-01-21 11:55:13.559649: step: 916/529, loss: 0.015199804678559303 2023-01-21 11:55:14.644853: step: 920/529, loss: 4.705586910247803 2023-01-21 11:55:15.759894: step: 924/529, loss: 0.00301532750017941 2023-01-21 11:55:16.869048: step: 928/529, loss: 0.06875982880592346 2023-01-21 11:55:17.989134: step: 932/529, loss: 0.2977858781814575 2023-01-21 11:55:19.109249: step: 936/529, loss: 0.002957821125164628 2023-01-21 11:55:20.233671: step: 940/529, loss: 0.005190849304199219 2023-01-21 11:55:21.343794: step: 944/529, loss: 0.005279255099594593 2023-01-21 11:55:22.475959: step: 948/529, loss: 0.01789245568215847 2023-01-21 11:55:23.603196: step: 952/529, loss: 0.00038733481778763235 2023-01-21 11:55:24.735407: step: 956/529, loss: 0.00030231475830078125 2023-01-21 11:55:25.845346: step: 960/529, loss: 0.005360794253647327 2023-01-21 11:55:26.967669: step: 964/529, loss: 0.03072347678244114 2023-01-21 11:55:28.100655: step: 968/529, loss: 0.029320908710360527 2023-01-21 11:55:29.246933: step: 972/529, loss: 0.17129193246364594 2023-01-21 11:55:30.352375: step: 976/529, loss: 0.0044345855712890625 2023-01-21 11:55:31.515948: step: 980/529, loss: 0.05413408204913139 2023-01-21 11:55:32.639777: step: 984/529, loss: 0.021362684667110443 2023-01-21 11:55:33.774402: step: 988/529, loss: 0.0007761955494061112 2023-01-21 11:55:34.888288: step: 992/529, loss: 0.0004703521844930947 2023-01-21 11:55:36.003170: step: 996/529, loss: 0.006201458163559437 2023-01-21 11:55:37.116128: step: 1000/529, loss: 0.002151393797248602 2023-01-21 11:55:38.266668: step: 1004/529, loss: 0.02366781234741211 2023-01-21 11:55:39.403236: step: 1008/529, loss: 8.029938180698082e-05 2023-01-21 11:55:40.529743: step: 1012/529, loss: 0.0004131317255087197 2023-01-21 11:55:41.673666: step: 1016/529, loss: 0.019981957972049713 2023-01-21 11:55:42.812420: step: 1020/529, loss: 0.10239439457654953 2023-01-21 11:55:43.922237: step: 1024/529, loss: 1.316070574830519e-05 2023-01-21 11:55:45.053436: step: 1028/529, loss: 8.249282836914062e-05 2023-01-21 11:55:46.204883: step: 1032/529, loss: 0.020923709496855736 2023-01-21 11:55:47.332588: step: 1036/529, loss: 0.00030660632182843983 2023-01-21 11:55:48.469757: step: 1040/529, loss: 0.006591606419533491 2023-01-21 11:55:49.608700: step: 1044/529, loss: 0.04016093909740448 2023-01-21 11:55:50.744950: step: 1048/529, loss: 0.032352447509765625 2023-01-21 11:55:51.869933: step: 1052/529, loss: 0.07407794147729874 2023-01-21 11:55:52.985202: step: 1056/529, loss: 0.00048093797522597015 2023-01-21 11:55:54.125427: step: 1060/529, loss: 0.012898826040327549 2023-01-21 11:55:55.307927: step: 1064/529, loss: 0.006937217898666859 2023-01-21 11:55:56.468490: step: 1068/529, loss: 0.0030711174476891756 2023-01-21 11:55:57.594691: step: 1072/529, loss: 0.010633469559252262 2023-01-21 11:55:58.730746: step: 1076/529, loss: 0.08028993755578995 2023-01-21 11:55:59.873317: step: 1080/529, loss: 0.025843143463134766 2023-01-21 11:56:01.013523: step: 1084/529, loss: 0.011153602972626686 2023-01-21 11:56:02.143784: step: 1088/529, loss: 0.9641178846359253 2023-01-21 11:56:03.294556: step: 1092/529, loss: 0.0005270004039630294 2023-01-21 11:56:04.445404: step: 1096/529, loss: 0.009481240063905716 2023-01-21 11:56:05.566583: step: 1100/529, loss: 0.0016740323044359684 2023-01-21 11:56:06.685015: step: 1104/529, loss: 0.0005876540672034025 2023-01-21 11:56:07.813991: step: 1108/529, loss: 0.0004922866937704384 2023-01-21 11:56:08.924196: step: 1112/529, loss: 0.0013348580105230212 2023-01-21 11:56:10.057741: step: 1116/529, loss: 0.09525832533836365 2023-01-21 11:56:11.201082: step: 1120/529, loss: 0.0009435653919354081 2023-01-21 11:56:12.351259: step: 1124/529, loss: 0.00011196135892532766 2023-01-21 11:56:13.517132: step: 1128/529, loss: 0.014861107803881168 2023-01-21 11:56:14.639989: step: 1132/529, loss: 0.013185501098632812 2023-01-21 11:56:15.769440: step: 1136/529, loss: 0.018764592707157135 2023-01-21 11:56:16.887096: step: 1140/529, loss: 0.05502796545624733 2023-01-21 11:56:18.026819: step: 1144/529, loss: 0.03473053127527237 2023-01-21 11:56:19.140126: step: 1148/529, loss: 0.05977487564086914 2023-01-21 11:56:20.330801: step: 1152/529, loss: 0.0037065506912767887 2023-01-21 11:56:21.461503: step: 1156/529, loss: 0.00016288757615257055 2023-01-21 11:56:22.601684: step: 1160/529, loss: 3.0517578125e-05 2023-01-21 11:56:23.736735: step: 1164/529, loss: 0.014849090948700905 2023-01-21 11:56:24.866666: step: 1168/529, loss: 0.00044116974459029734 2023-01-21 11:56:26.028702: step: 1172/529, loss: 0.016167735680937767 2023-01-21 11:56:27.169522: step: 1176/529, loss: 0.004170751664787531 2023-01-21 11:56:28.344825: step: 1180/529, loss: 0.0013672829372808337 2023-01-21 11:56:29.466230: step: 1184/529, loss: 0.00926055945456028 2023-01-21 11:56:30.599421: step: 1188/529, loss: 0.0036658288445323706 2023-01-21 11:56:31.744776: step: 1192/529, loss: 0.15600451827049255 2023-01-21 11:56:32.872256: step: 1196/529, loss: 0.04693803936243057 2023-01-21 11:56:33.988765: step: 1200/529, loss: 0.31173840165138245 2023-01-21 11:56:35.127937: step: 1204/529, loss: 0.007909155450761318 2023-01-21 11:56:36.250391: step: 1208/529, loss: 0.00023279190645553172 2023-01-21 11:56:37.351604: step: 1212/529, loss: 0.0025196552742272615 2023-01-21 11:56:38.466615: step: 1216/529, loss: 0.01823749579489231 2023-01-21 11:56:39.593425: step: 1220/529, loss: 0.005439949221909046 2023-01-21 11:56:40.698187: step: 1224/529, loss: 0.0027477266266942024 2023-01-21 11:56:41.859408: step: 1228/529, loss: 0.026699256151914597 2023-01-21 11:56:42.997091: step: 1232/529, loss: 0.0008554458618164062 2023-01-21 11:56:44.121696: step: 1236/529, loss: 0.0007020949851721525 2023-01-21 11:56:45.236528: step: 1240/529, loss: 0.0002577781560830772 2023-01-21 11:56:46.340411: step: 1244/529, loss: 0.00034332275390625 2023-01-21 11:56:47.467146: step: 1248/529, loss: 0.02109546773135662 2023-01-21 11:56:48.586957: step: 1252/529, loss: 0.009605121798813343 2023-01-21 11:56:49.754145: step: 1256/529, loss: 0.2840808928012848 2023-01-21 11:56:50.904095: step: 1260/529, loss: 0.002441597171127796 2023-01-21 11:56:52.045200: step: 1264/529, loss: 0.02648944780230522 2023-01-21 11:56:53.176926: step: 1268/529, loss: 9.441375732421875e-05 2023-01-21 11:56:54.332601: step: 1272/529, loss: 0.00016117095947265625 2023-01-21 11:56:55.495495: step: 1276/529, loss: 0.0028770449571311474 2023-01-21 11:56:56.635193: step: 1280/529, loss: 0.002125263214111328 2023-01-21 11:56:57.765777: step: 1284/529, loss: 0.00380706787109375 2023-01-21 11:56:58.895690: step: 1288/529, loss: 0.0012969970703125 2023-01-21 11:57:00.031672: step: 1292/529, loss: 0.0019219398964196444 2023-01-21 11:57:01.161505: step: 1296/529, loss: 0.030411623418331146 2023-01-21 11:57:02.321814: step: 1300/529, loss: 0.024067021906375885 2023-01-21 11:57:03.453691: step: 1304/529, loss: 0.009385108947753906 2023-01-21 11:57:04.564196: step: 1308/529, loss: 0.00565834017470479 2023-01-21 11:57:05.673689: step: 1312/529, loss: 0.006525897886604071 2023-01-21 11:57:06.791421: step: 1316/529, loss: 0.0001958847133209929 2023-01-21 11:57:07.945951: step: 1320/529, loss: 0.11978740990161896 2023-01-21 11:57:09.073562: step: 1324/529, loss: 0.044416047632694244 2023-01-21 11:57:10.203928: step: 1328/529, loss: 0.06220836937427521 2023-01-21 11:57:11.332575: step: 1332/529, loss: 0.0038527490105479956 2023-01-21 11:57:12.450162: step: 1336/529, loss: 0.001304245088249445 2023-01-21 11:57:13.580730: step: 1340/529, loss: 0.0003205776447430253 2023-01-21 11:57:14.724121: step: 1344/529, loss: 0.00019922255887649953 2023-01-21 11:57:15.854658: step: 1348/529, loss: 4.4441225327318534e-05 2023-01-21 11:57:16.972402: step: 1352/529, loss: 0.06800708919763565 2023-01-21 11:57:18.113533: step: 1356/529, loss: 0.028739355504512787 2023-01-21 11:57:19.255921: step: 1360/529, loss: 0.024689961224794388 2023-01-21 11:57:20.385020: step: 1364/529, loss: 0.012892341241240501 2023-01-21 11:57:21.496808: step: 1368/529, loss: 0.021869372576475143 2023-01-21 11:57:22.669294: step: 1372/529, loss: 0.0013822675682604313 2023-01-21 11:57:23.799367: step: 1376/529, loss: 0.006617831997573376 2023-01-21 11:57:24.950040: step: 1380/529, loss: 0.057739924639463425 2023-01-21 11:57:26.067644: step: 1384/529, loss: 0.02259845845401287 2023-01-21 11:57:27.192776: step: 1388/529, loss: 0.023250199854373932 2023-01-21 11:57:28.317870: step: 1392/529, loss: 0.006877136416733265 2023-01-21 11:57:29.441174: step: 1396/529, loss: 0.02518177032470703 2023-01-21 11:57:30.565954: step: 1400/529, loss: 0.0035586003214120865 2023-01-21 11:57:31.696545: step: 1404/529, loss: 0.04266529157757759 2023-01-21 11:57:32.850291: step: 1408/529, loss: 0.002956867218017578 2023-01-21 11:57:33.986844: step: 1412/529, loss: 0.015162426978349686 2023-01-21 11:57:35.074215: step: 1416/529, loss: 0.0006565094226971269 2023-01-21 11:57:36.218283: step: 1420/529, loss: 0.0005176544073037803 2023-01-21 11:57:37.348521: step: 1424/529, loss: 0.006370067596435547 2023-01-21 11:57:38.470059: step: 1428/529, loss: 0.04891815409064293 2023-01-21 11:57:39.581371: step: 1432/529, loss: 0.024137306958436966 2023-01-21 11:57:40.739649: step: 1436/529, loss: 0.0005458355299197137 2023-01-21 11:57:41.883072: step: 1440/529, loss: 0.004716396797448397 2023-01-21 11:57:43.026476: step: 1444/529, loss: 0.0007860183832235634 2023-01-21 11:57:44.171133: step: 1448/529, loss: 0.034719087183475494 2023-01-21 11:57:45.293952: step: 1452/529, loss: 0.01760726049542427 2023-01-21 11:57:46.438966: step: 1456/529, loss: 0.003089523408561945 2023-01-21 11:57:47.566376: step: 1460/529, loss: 0.003487443784251809 2023-01-21 11:57:48.726507: step: 1464/529, loss: 0.012828350998461246 2023-01-21 11:57:49.861952: step: 1468/529, loss: 0.007272911258041859 2023-01-21 11:57:50.989287: step: 1472/529, loss: 0.0628393143415451 2023-01-21 11:57:52.203759: step: 1476/529, loss: 0.0005155563121661544 2023-01-21 11:57:53.316352: step: 1480/529, loss: 0.004618645180016756 2023-01-21 11:57:54.467106: step: 1484/529, loss: 0.0005880832322873175 2023-01-21 11:57:55.613085: step: 1488/529, loss: 0.0013024808140471578 2023-01-21 11:57:56.719910: step: 1492/529, loss: 0.04302950203418732 2023-01-21 11:57:57.853144: step: 1496/529, loss: 0.0036716461181640625 2023-01-21 11:57:58.994768: step: 1500/529, loss: 0.00414619455114007 2023-01-21 11:58:00.128187: step: 1504/529, loss: 0.00033607485238462687 2023-01-21 11:58:01.249641: step: 1508/529, loss: 0.06471367180347443 2023-01-21 11:58:02.378886: step: 1512/529, loss: 0.0039650918915867805 2023-01-21 11:58:03.486998: step: 1516/529, loss: 0.008624029345810413 2023-01-21 11:58:04.610960: step: 1520/529, loss: 0.0003334045468363911 2023-01-21 11:58:05.732420: step: 1524/529, loss: 0.04535217210650444 2023-01-21 11:58:06.897678: step: 1528/529, loss: 0.0048965453170239925 2023-01-21 11:58:08.029143: step: 1532/529, loss: 0.0308137908577919 2023-01-21 11:58:09.149793: step: 1536/529, loss: 0.03395118936896324 2023-01-21 11:58:10.306387: step: 1540/529, loss: 0.03233480453491211 2023-01-21 11:58:11.433631: step: 1544/529, loss: 0.009851264767348766 2023-01-21 11:58:12.570169: step: 1548/529, loss: 0.005098151974380016 2023-01-21 11:58:13.694432: step: 1552/529, loss: 0.009525108151137829 2023-01-21 11:58:14.835916: step: 1556/529, loss: 0.013803482055664062 2023-01-21 11:58:16.002241: step: 1560/529, loss: 0.005675602238625288 2023-01-21 11:58:17.107244: step: 1564/529, loss: 0.00043172837467864156 2023-01-21 11:58:18.222772: step: 1568/529, loss: 0.009839248843491077 2023-01-21 11:58:19.393759: step: 1572/529, loss: 0.09858636558055878 2023-01-21 11:58:20.537339: step: 1576/529, loss: 0.0009611130226403475 2023-01-21 11:58:21.724702: step: 1580/529, loss: 0.4569973051548004 2023-01-21 11:58:22.844103: step: 1584/529, loss: 0.022182513028383255 2023-01-21 11:58:23.954097: step: 1588/529, loss: 0.0006231308216229081 2023-01-21 11:58:25.076228: step: 1592/529, loss: 0.001380109810270369 2023-01-21 11:58:26.194694: step: 1596/529, loss: 0.002073192736133933 2023-01-21 11:58:27.310979: step: 1600/529, loss: 0.0012027740012854338 2023-01-21 11:58:28.417293: step: 1604/529, loss: 0.0012117385631427169 2023-01-21 11:58:29.567731: step: 1608/529, loss: 0.021396446973085403 2023-01-21 11:58:30.698188: step: 1612/529, loss: 0.0024911880027502775 2023-01-21 11:58:31.818831: step: 1616/529, loss: 0.0030375956557691097 2023-01-21 11:58:32.984228: step: 1620/529, loss: 0.01107335090637207 2023-01-21 11:58:34.145113: step: 1624/529, loss: 0.0019267082680016756 2023-01-21 11:58:35.273575: step: 1628/529, loss: 0.0006121635669842362 2023-01-21 11:58:36.405149: step: 1632/529, loss: 0.05602245032787323 2023-01-21 11:58:37.526080: step: 1636/529, loss: 0.007361126132309437 2023-01-21 11:58:38.671426: step: 1640/529, loss: 0.006140708923339844 2023-01-21 11:58:39.824606: step: 1644/529, loss: 0.011189651675522327 2023-01-21 11:58:40.956782: step: 1648/529, loss: 0.0018308639992028475 2023-01-21 11:58:42.118454: step: 1652/529, loss: 0.045818231999874115 2023-01-21 11:58:43.247365: step: 1656/529, loss: 0.00015935898409225047 2023-01-21 11:58:44.372085: step: 1660/529, loss: 0.11245336383581161 2023-01-21 11:58:45.496170: step: 1664/529, loss: 0.0027372362092137337 2023-01-21 11:58:46.593558: step: 1668/529, loss: 0.010306263342499733 2023-01-21 11:58:47.718266: step: 1672/529, loss: 0.0013591767055913806 2023-01-21 11:58:48.852920: step: 1676/529, loss: 0.0006219864590093493 2023-01-21 11:58:50.003349: step: 1680/529, loss: 0.0009169578552246094 2023-01-21 11:58:51.147837: step: 1684/529, loss: 0.04617347940802574 2023-01-21 11:58:52.268266: step: 1688/529, loss: 0.29647645354270935 2023-01-21 11:58:53.376419: step: 1692/529, loss: 0.03921337425708771 2023-01-21 11:58:54.501765: step: 1696/529, loss: 0.12983036041259766 2023-01-21 11:58:55.663276: step: 1700/529, loss: 0.00022926331439521164 2023-01-21 11:58:56.815423: step: 1704/529, loss: 0.020938683301210403 2023-01-21 11:58:57.964132: step: 1708/529, loss: 0.03709878772497177 2023-01-21 11:58:59.099292: step: 1712/529, loss: 0.011890649795532227 2023-01-21 11:59:00.252373: step: 1716/529, loss: 0.0007921218639239669 2023-01-21 11:59:01.389821: step: 1720/529, loss: 0.004993248265236616 2023-01-21 11:59:02.515148: step: 1724/529, loss: 0.1916790008544922 2023-01-21 11:59:03.645248: step: 1728/529, loss: 0.0020202635787427425 2023-01-21 11:59:04.775371: step: 1732/529, loss: 0.0537593849003315 2023-01-21 11:59:05.910739: step: 1736/529, loss: 0.014293670654296875 2023-01-21 11:59:07.054530: step: 1740/529, loss: 0.04201831668615341 2023-01-21 11:59:08.207092: step: 1744/529, loss: 0.03522996976971626 2023-01-21 11:59:09.322598: step: 1748/529, loss: 0.005183362867683172 2023-01-21 11:59:10.453662: step: 1752/529, loss: 0.003144502639770508 2023-01-21 11:59:11.564482: step: 1756/529, loss: 0.00644760113209486 2023-01-21 11:59:12.698077: step: 1760/529, loss: 0.012038993649184704 2023-01-21 11:59:13.853314: step: 1764/529, loss: 0.01338181458413601 2023-01-21 11:59:14.981084: step: 1768/529, loss: 0.010101891122758389 2023-01-21 11:59:16.113628: step: 1772/529, loss: 0.02355017699301243 2023-01-21 11:59:17.238558: step: 1776/529, loss: 0.06967373192310333 2023-01-21 11:59:18.372643: step: 1780/529, loss: 0.004602814093232155 2023-01-21 11:59:19.488533: step: 1784/529, loss: 0.00010795592970680445 2023-01-21 11:59:20.641566: step: 1788/529, loss: 0.0015532494289800525 2023-01-21 11:59:21.757402: step: 1792/529, loss: 0.02895355224609375 2023-01-21 11:59:22.904595: step: 1796/529, loss: 0.0024499893188476562 2023-01-21 11:59:24.044114: step: 1800/529, loss: 0.0002652168332133442 2023-01-21 11:59:25.165212: step: 1804/529, loss: 0.00024118424335028976 2023-01-21 11:59:26.291338: step: 1808/529, loss: 0.0008556366083212197 2023-01-21 11:59:27.465808: step: 1812/529, loss: 0.040833473205566406 2023-01-21 11:59:28.583527: step: 1816/529, loss: 0.00027952194795943797 2023-01-21 11:59:29.739237: step: 1820/529, loss: 0.023070333525538445 2023-01-21 11:59:30.857159: step: 1824/529, loss: 0.0003982544003520161 2023-01-21 11:59:32.005297: step: 1828/529, loss: 8.640289161121473e-05 2023-01-21 11:59:33.185735: step: 1832/529, loss: 0.027905656024813652 2023-01-21 11:59:34.322424: step: 1836/529, loss: 4.763603283208795e-05 2023-01-21 11:59:35.475602: step: 1840/529, loss: 0.008761310018599033 2023-01-21 11:59:36.629425: step: 1844/529, loss: 0.00010185241990257055 2023-01-21 11:59:37.761396: step: 1848/529, loss: 0.0023704529739916325 2023-01-21 11:59:38.871516: step: 1852/529, loss: 0.0010433197021484375 2023-01-21 11:59:40.004049: step: 1856/529, loss: 0.03183021396398544 2023-01-21 11:59:41.129933: step: 1860/529, loss: 0.00014452934556175023 2023-01-21 11:59:42.254894: step: 1864/529, loss: 0.003361320588737726 2023-01-21 11:59:43.384128: step: 1868/529, loss: 0.037209652364254 2023-01-21 11:59:44.517919: step: 1872/529, loss: 0.006120109464973211 2023-01-21 11:59:45.656285: step: 1876/529, loss: 0.10689258575439453 2023-01-21 11:59:46.781311: step: 1880/529, loss: 0.010372161865234375 2023-01-21 11:59:47.916552: step: 1884/529, loss: 0.004285621456801891 2023-01-21 11:59:49.047798: step: 1888/529, loss: 0.03818473964929581 2023-01-21 11:59:50.164941: step: 1892/529, loss: 0.0001446724054403603 2023-01-21 11:59:51.330868: step: 1896/529, loss: 0.03526516258716583 2023-01-21 11:59:52.478116: step: 1900/529, loss: 0.0015977859729900956 2023-01-21 11:59:53.596310: step: 1904/529, loss: 0.007211733143776655 2023-01-21 11:59:54.733044: step: 1908/529, loss: 0.049468230456113815 2023-01-21 11:59:55.877839: step: 1912/529, loss: 0.015325259417295456 2023-01-21 11:59:57.023022: step: 1916/529, loss: 0.02558603323996067 2023-01-21 11:59:58.144270: step: 1920/529, loss: 0.004660940263420343 2023-01-21 11:59:59.257930: step: 1924/529, loss: 0.03502912446856499 2023-01-21 12:00:00.345046: step: 1928/529, loss: 0.00011463165719760582 2023-01-21 12:00:01.474353: step: 1932/529, loss: 0.0035083771217614412 2023-01-21 12:00:02.586867: step: 1936/529, loss: 0.044994208961725235 2023-01-21 12:00:03.750372: step: 1940/529, loss: 0.013011455535888672 2023-01-21 12:00:04.914384: step: 1944/529, loss: 0.0690038651227951 2023-01-21 12:00:06.062585: step: 1948/529, loss: 0.021395491436123848 2023-01-21 12:00:07.204868: step: 1952/529, loss: 0.1346874237060547 2023-01-21 12:00:08.346993: step: 1956/529, loss: 0.0006036758422851562 2023-01-21 12:00:09.490490: step: 1960/529, loss: 0.002509689424186945 2023-01-21 12:00:10.618022: step: 1964/529, loss: 0.0053873066790401936 2023-01-21 12:00:11.743918: step: 1968/529, loss: 0.00026035308837890625 2023-01-21 12:00:12.939570: step: 1972/529, loss: 0.0001183509812108241 2023-01-21 12:00:14.069512: step: 1976/529, loss: 0.009867191314697266 2023-01-21 12:00:15.235798: step: 1980/529, loss: 0.0022149085998535156 2023-01-21 12:00:16.382455: step: 1984/529, loss: 8.39233416627394e-06 2023-01-21 12:00:17.493575: step: 1988/529, loss: 0.019773198291659355 2023-01-21 12:00:18.644432: step: 1992/529, loss: 0.03413581848144531 2023-01-21 12:00:19.782093: step: 1996/529, loss: 0.0015562057960778475 2023-01-21 12:00:20.931494: step: 2000/529, loss: 0.06823521107435226 2023-01-21 12:00:22.101250: step: 2004/529, loss: 0.016434289515018463 2023-01-21 12:00:23.224271: step: 2008/529, loss: 0.0008293151622638106 2023-01-21 12:00:24.353567: step: 2012/529, loss: 0.0006480216979980469 2023-01-21 12:00:25.467482: step: 2016/529, loss: 0.010793304070830345 2023-01-21 12:00:26.611329: step: 2020/529, loss: 0.006809521000832319 2023-01-21 12:00:27.756376: step: 2024/529, loss: 0.0006639480707235634 2023-01-21 12:00:28.892548: step: 2028/529, loss: 0.0005534171941690147 2023-01-21 12:00:30.042546: step: 2032/529, loss: 0.023509599268436432 2023-01-21 12:00:31.180577: step: 2036/529, loss: 0.0009523392072878778 2023-01-21 12:00:32.334918: step: 2040/529, loss: 0.03637246787548065 2023-01-21 12:00:33.493107: step: 2044/529, loss: 0.02288341522216797 2023-01-21 12:00:34.612421: step: 2048/529, loss: 0.056105807423591614 2023-01-21 12:00:35.754730: step: 2052/529, loss: 0.014925575815141201 2023-01-21 12:00:36.879230: step: 2056/529, loss: 0.011881733313202858 2023-01-21 12:00:38.015992: step: 2060/529, loss: 0.003666400909423828 2023-01-21 12:00:39.166109: step: 2064/529, loss: 0.0021726610139012337 2023-01-21 12:00:40.314611: step: 2068/529, loss: 0.035338688641786575 2023-01-21 12:00:41.471560: step: 2072/529, loss: 0.0012248515849933028 2023-01-21 12:00:42.626345: step: 2076/529, loss: 0.0005747318500652909 2023-01-21 12:00:43.769502: step: 2080/529, loss: 0.0023475647903978825 2023-01-21 12:00:44.931098: step: 2084/529, loss: 0.00099945068359375 2023-01-21 12:00:46.045012: step: 2088/529, loss: 0.0014196396805346012 2023-01-21 12:00:47.176011: step: 2092/529, loss: 0.010527086444199085 2023-01-21 12:00:48.311748: step: 2096/529, loss: 0.009007644839584827 2023-01-21 12:00:49.426096: step: 2100/529, loss: 0.06700649112462997 2023-01-21 12:00:50.543440: step: 2104/529, loss: 0.021753691136837006 2023-01-21 12:00:51.672473: step: 2108/529, loss: 0.019083648920059204 2023-01-21 12:00:52.804690: step: 2112/529, loss: 0.023957347497344017 2023-01-21 12:00:53.923117: step: 2116/529, loss: 2.6226043701171875e-06 ================================================== Loss: 0.034 -------------------- Dev: {'event': {'p': 0.581532416502947, 'r': 0.7882822902796272, 'f1': 0.6693046919163369}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Test: {'event': {'p': 0.6024150268336315, 'r': 0.7804171494785631, 'f1': 0.6799596163553759}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Chinese: {'event': {'p': 0.5444444444444444, 'r': 0.9074074074074074, 'f1': 0.6805555555555556}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Korean: {'event': {'p': 0.5967741935483871, 'r': 0.5873015873015873, 'f1': 0.592}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Russian: {'event': {'p': 0.4444444444444444, 'r': 0.5555555555555556, 'f1': 0.49382716049382713}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 20 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:01:35.274386: step: 4/529, loss: 0.014572334475815296 2023-01-21 12:01:36.398185: step: 8/529, loss: 0.00015687942504882812 2023-01-21 12:01:37.561688: step: 12/529, loss: 0.031372833997011185 2023-01-21 12:01:38.690092: step: 16/529, loss: 0.04549608379602432 2023-01-21 12:01:39.822241: step: 20/529, loss: 0.005827045533806086 2023-01-21 12:01:40.944105: step: 24/529, loss: 0.0014043807750567794 2023-01-21 12:01:42.094887: step: 28/529, loss: 0.008192635141313076 2023-01-21 12:01:43.216016: step: 32/529, loss: 0.005195140838623047 2023-01-21 12:01:44.339745: step: 36/529, loss: 0.00022268295288085938 2023-01-21 12:01:45.508978: step: 40/529, loss: 0.008117484860122204 2023-01-21 12:01:46.650554: step: 44/529, loss: 0.0619657076895237 2023-01-21 12:01:47.804189: step: 48/529, loss: 0.0008929252508096397 2023-01-21 12:01:48.933460: step: 52/529, loss: 0.004566192626953125 2023-01-21 12:01:50.086842: step: 56/529, loss: 0.0005871772882528603 2023-01-21 12:01:51.213951: step: 60/529, loss: 0.010891533456742764 2023-01-21 12:01:52.360181: step: 64/529, loss: 0.0003082275507040322 2023-01-21 12:01:53.510282: step: 68/529, loss: 0.006340503692626953 2023-01-21 12:01:54.696724: step: 72/529, loss: 6.532669067382812e-05 2023-01-21 12:01:55.852773: step: 76/529, loss: 0.044126369059085846 2023-01-21 12:01:57.029330: step: 80/529, loss: 0.03161201626062393 2023-01-21 12:01:58.170318: step: 84/529, loss: 0.04349403455853462 2023-01-21 12:01:59.301947: step: 88/529, loss: 0.004472637083381414 2023-01-21 12:02:00.422166: step: 92/529, loss: 0.0007043838268145919 2023-01-21 12:02:01.556200: step: 96/529, loss: 0.00051965715829283 2023-01-21 12:02:02.720095: step: 100/529, loss: 0.018740464001893997 2023-01-21 12:02:03.861089: step: 104/529, loss: 0.005173110868781805 2023-01-21 12:02:05.015325: step: 108/529, loss: 0.003543853759765625 2023-01-21 12:02:06.141136: step: 112/529, loss: 0.005423164460808039 2023-01-21 12:02:07.256498: step: 116/529, loss: 0.0014303206698969007 2023-01-21 12:02:08.397605: step: 120/529, loss: 1.1219558715820312 2023-01-21 12:02:09.511405: step: 124/529, loss: 0.00016241073899436742 2023-01-21 12:02:10.666921: step: 128/529, loss: 0.0008438110817223787 2023-01-21 12:02:11.782892: step: 132/529, loss: 0.002221775008365512 2023-01-21 12:02:12.907828: step: 136/529, loss: 0.025299929082393646 2023-01-21 12:02:14.075436: step: 140/529, loss: 0.024934101849794388 2023-01-21 12:02:15.214164: step: 144/529, loss: 0.00021629335242323577 2023-01-21 12:02:16.331600: step: 148/529, loss: 0.05854148790240288 2023-01-21 12:02:17.462719: step: 152/529, loss: 0.0031544207595288754 2023-01-21 12:02:18.575685: step: 156/529, loss: 0.006803989876061678 2023-01-21 12:02:19.696294: step: 160/529, loss: 0.008624744601547718 2023-01-21 12:02:20.813537: step: 164/529, loss: 0.0027171135880053043 2023-01-21 12:02:21.926972: step: 168/529, loss: 0.011455345898866653 2023-01-21 12:02:23.055825: step: 172/529, loss: 0.001856040908023715 2023-01-21 12:02:24.192111: step: 176/529, loss: 0.06771411746740341 2023-01-21 12:02:25.350400: step: 180/529, loss: 0.009473037905991077 2023-01-21 12:02:26.494470: step: 184/529, loss: 0.24229221045970917 2023-01-21 12:02:27.620672: step: 188/529, loss: 0.004801750183105469 2023-01-21 12:02:28.809137: step: 192/529, loss: 0.005550098605453968 2023-01-21 12:02:29.937094: step: 196/529, loss: 4.024505687993951e-05 2023-01-21 12:02:31.076917: step: 200/529, loss: 0.024713803082704544 2023-01-21 12:02:32.206291: step: 204/529, loss: 0.03115091286599636 2023-01-21 12:02:33.372029: step: 208/529, loss: 0.006118202582001686 2023-01-21 12:02:34.503159: step: 212/529, loss: 0.023369980975985527 2023-01-21 12:02:35.611269: step: 216/529, loss: 0.000720977783203125 2023-01-21 12:02:36.764981: step: 220/529, loss: 0.003168296767398715 2023-01-21 12:02:37.895473: step: 224/529, loss: 0.00013551712618209422 2023-01-21 12:02:39.019721: step: 228/529, loss: 0.0326593779027462 2023-01-21 12:02:40.171329: step: 232/529, loss: 0.02245769463479519 2023-01-21 12:02:41.288543: step: 236/529, loss: 0.046436309814453125 2023-01-21 12:02:42.445122: step: 240/529, loss: 0.0004540443478617817 2023-01-21 12:02:43.569930: step: 244/529, loss: 0.048932984471321106 2023-01-21 12:02:44.684568: step: 248/529, loss: 0.00019569398136809468 2023-01-21 12:02:45.848563: step: 252/529, loss: 0.04384374991059303 2023-01-21 12:02:46.983346: step: 256/529, loss: 0.0011034011840820312 2023-01-21 12:02:48.122097: step: 260/529, loss: 0.27853965759277344 2023-01-21 12:02:49.289910: step: 264/529, loss: 0.012395954690873623 2023-01-21 12:02:50.441762: step: 268/529, loss: 0.028108788654208183 2023-01-21 12:02:51.565611: step: 272/529, loss: 0.0011301040649414062 2023-01-21 12:02:52.716505: step: 276/529, loss: 0.0002620697196107358 2023-01-21 12:02:53.884275: step: 280/529, loss: 0.0031561851501464844 2023-01-21 12:02:55.047352: step: 284/529, loss: 0.016859054565429688 2023-01-21 12:02:56.192072: step: 288/529, loss: 0.0010635375510901213 2023-01-21 12:02:57.351000: step: 292/529, loss: 0.028599359095096588 2023-01-21 12:02:58.471421: step: 296/529, loss: 0.008155345916748047 2023-01-21 12:02:59.619607: step: 300/529, loss: 0.004108619876205921 2023-01-21 12:03:00.724960: step: 304/529, loss: 4.5585635234601796e-05 2023-01-21 12:03:01.892642: step: 308/529, loss: 0.0002601623709779233 2023-01-21 12:03:03.021158: step: 312/529, loss: 0.012130547314882278 2023-01-21 12:03:04.177733: step: 316/529, loss: 0.0016498565673828125 2023-01-21 12:03:05.311173: step: 320/529, loss: 5.073547436040826e-05 2023-01-21 12:03:06.455627: step: 324/529, loss: 8.888245065463707e-05 2023-01-21 12:03:07.578971: step: 328/529, loss: 0.018139267340302467 2023-01-21 12:03:08.729402: step: 332/529, loss: 0.01211319025605917 2023-01-21 12:03:09.886748: step: 336/529, loss: 0.020360421389341354 2023-01-21 12:03:11.022658: step: 340/529, loss: 0.005267334170639515 2023-01-21 12:03:12.144844: step: 344/529, loss: 0.0009886741172522306 2023-01-21 12:03:13.286708: step: 348/529, loss: 0.0011377334594726562 2023-01-21 12:03:14.408092: step: 352/529, loss: 0.003307342529296875 2023-01-21 12:03:15.546016: step: 356/529, loss: 0.06897401809692383 2023-01-21 12:03:16.642279: step: 360/529, loss: 0.0003910541709046811 2023-01-21 12:03:17.774851: step: 364/529, loss: 0.013899422250688076 2023-01-21 12:03:18.901043: step: 368/529, loss: 0.011314105242490768 2023-01-21 12:03:20.016270: step: 372/529, loss: 0.0009882927406579256 2023-01-21 12:03:21.142962: step: 376/529, loss: 0.010402584448456764 2023-01-21 12:03:22.302329: step: 380/529, loss: 0.00013828277587890625 2023-01-21 12:03:23.391071: step: 384/529, loss: 0.0003765106375794858 2023-01-21 12:03:24.559818: step: 388/529, loss: 0.01226730365306139 2023-01-21 12:03:25.680570: step: 392/529, loss: 0.005211162846535444 2023-01-21 12:03:26.815220: step: 396/529, loss: 0.03441572189331055 2023-01-21 12:03:27.973303: step: 400/529, loss: 6.732941255904734e-05 2023-01-21 12:03:29.100863: step: 404/529, loss: 2.1457672119140625e-05 2023-01-21 12:03:30.240658: step: 408/529, loss: 0.024457359686493874 2023-01-21 12:03:31.366360: step: 412/529, loss: 0.013788414187729359 2023-01-21 12:03:32.497782: step: 416/529, loss: 0.011552381329238415 2023-01-21 12:03:33.640439: step: 420/529, loss: 0.0007276535034179688 2023-01-21 12:03:34.764769: step: 424/529, loss: 0.0019419670570641756 2023-01-21 12:03:35.890396: step: 428/529, loss: 0.0018526078201830387 2023-01-21 12:03:37.029366: step: 432/529, loss: 0.012841415591537952 2023-01-21 12:03:38.164267: step: 436/529, loss: 0.0014563561417162418 2023-01-21 12:03:39.299091: step: 440/529, loss: 0.0023183822631835938 2023-01-21 12:03:40.449857: step: 444/529, loss: 1.392364538332913e-05 2023-01-21 12:03:41.588144: step: 448/529, loss: 0.001147461007349193 2023-01-21 12:03:42.732897: step: 452/529, loss: 0.0017309188842773438 2023-01-21 12:03:43.858809: step: 456/529, loss: 0.00011940002877963707 2023-01-21 12:03:45.006998: step: 460/529, loss: 0.000778579676989466 2023-01-21 12:03:46.120940: step: 464/529, loss: 0.0028966425452381372 2023-01-21 12:03:47.255755: step: 468/529, loss: 0.006875229068100452 2023-01-21 12:03:48.413408: step: 472/529, loss: 0.00010414123244117945 2023-01-21 12:03:49.525969: step: 476/529, loss: 0.00557289132848382 2023-01-21 12:03:50.658320: step: 480/529, loss: 0.024851465597748756 2023-01-21 12:03:51.809325: step: 484/529, loss: 0.00038242340087890625 2023-01-21 12:03:52.943980: step: 488/529, loss: 0.0017158507835119963 2023-01-21 12:03:54.094501: step: 492/529, loss: 1.3351443612918956e-06 2023-01-21 12:03:55.235880: step: 496/529, loss: 0.007054328918457031 2023-01-21 12:03:56.353834: step: 500/529, loss: 1.5163421267061494e-05 2023-01-21 12:03:57.491854: step: 504/529, loss: 0.032987404614686966 2023-01-21 12:03:58.595894: step: 508/529, loss: 0.005805682856589556 2023-01-21 12:03:59.740148: step: 512/529, loss: 4.081726001459174e-05 2023-01-21 12:04:00.879733: step: 516/529, loss: 0.008034229278564453 2023-01-21 12:04:02.016516: step: 520/529, loss: 0.0009360313415527344 2023-01-21 12:04:03.124515: step: 524/529, loss: 0.00223121652379632 2023-01-21 12:04:04.238115: step: 528/529, loss: 0.0037462234031409025 2023-01-21 12:04:05.396025: step: 532/529, loss: 0.01670379750430584 2023-01-21 12:04:06.524585: step: 536/529, loss: 0.07366657257080078 2023-01-21 12:04:07.656797: step: 540/529, loss: 0.02952442318201065 2023-01-21 12:04:08.794500: step: 544/529, loss: 0.020560646429657936 2023-01-21 12:04:09.950679: step: 548/529, loss: 0.0035467147827148438 2023-01-21 12:04:11.098527: step: 552/529, loss: 0.01634988747537136 2023-01-21 12:04:12.223494: step: 556/529, loss: 0.004773521330207586 2023-01-21 12:04:13.361282: step: 560/529, loss: 0.0002556085819378495 2023-01-21 12:04:14.501203: step: 564/529, loss: 0.05723247677087784 2023-01-21 12:04:15.646022: step: 568/529, loss: 0.002836227649822831 2023-01-21 12:04:16.800257: step: 572/529, loss: 0.0002426147402729839 2023-01-21 12:04:17.934627: step: 576/529, loss: 0.020611191168427467 2023-01-21 12:04:19.067196: step: 580/529, loss: 0.08501420170068741 2023-01-21 12:04:20.206174: step: 584/529, loss: 0.00020275116548873484 2023-01-21 12:04:21.298112: step: 588/529, loss: 0.004576015751808882 2023-01-21 12:04:22.438818: step: 592/529, loss: 0.0013484954833984375 2023-01-21 12:04:23.559371: step: 596/529, loss: 0.0008884907001629472 2023-01-21 12:04:24.691632: step: 600/529, loss: 0.0037179947830736637 2023-01-21 12:04:25.797924: step: 604/529, loss: 0.0022679329849779606 2023-01-21 12:04:26.926638: step: 608/529, loss: 0.00551948556676507 2023-01-21 12:04:28.075910: step: 612/529, loss: 0.029462814331054688 2023-01-21 12:04:29.201267: step: 616/529, loss: 0.0009654999012127519 2023-01-21 12:04:30.325362: step: 620/529, loss: 0.00010318757267668843 2023-01-21 12:04:31.444257: step: 624/529, loss: 0.003912544343620539 2023-01-21 12:04:32.567852: step: 628/529, loss: 0.0024280548095703125 2023-01-21 12:04:33.692224: step: 632/529, loss: 0.0032265663612633944 2023-01-21 12:04:34.818685: step: 636/529, loss: 0.010932350531220436 2023-01-21 12:04:35.945934: step: 640/529, loss: 0.01750502735376358 2023-01-21 12:04:37.088164: step: 644/529, loss: 0.036095716059207916 2023-01-21 12:04:38.231625: step: 648/529, loss: 0.009757328778505325 2023-01-21 12:04:39.382093: step: 652/529, loss: 5.7411190937273204e-05 2023-01-21 12:04:40.549374: step: 656/529, loss: 0.012451362796127796 2023-01-21 12:04:41.685450: step: 660/529, loss: 0.0005082368734292686 2023-01-21 12:04:42.804401: step: 664/529, loss: 0.0015271187294274569 2023-01-21 12:04:43.944322: step: 668/529, loss: 0.014600562863051891 2023-01-21 12:04:45.084883: step: 672/529, loss: 0.028201866894960403 2023-01-21 12:04:46.209508: step: 676/529, loss: 0.0005444526905193925 2023-01-21 12:04:47.345313: step: 680/529, loss: 0.0003059387090615928 2023-01-21 12:04:48.487185: step: 684/529, loss: 0.047283269464969635 2023-01-21 12:04:49.618543: step: 688/529, loss: 0.0043516638688743114 2023-01-21 12:04:50.793533: step: 692/529, loss: 0.0015348434681072831 2023-01-21 12:04:51.933041: step: 696/529, loss: 0.0014476776123046875 2023-01-21 12:04:53.053628: step: 700/529, loss: 0.0035474777687340975 2023-01-21 12:04:54.185079: step: 704/529, loss: 0.002964592073112726 2023-01-21 12:04:55.325286: step: 708/529, loss: 0.0007718086126260459 2023-01-21 12:04:56.461912: step: 712/529, loss: 0.009532356634736061 2023-01-21 12:04:57.616146: step: 716/529, loss: 0.00044918060302734375 2023-01-21 12:04:58.761793: step: 720/529, loss: 0.0024585723876953125 2023-01-21 12:04:59.896689: step: 724/529, loss: 0.0013605117565020919 2023-01-21 12:05:01.052051: step: 728/529, loss: 0.00160818116273731 2023-01-21 12:05:02.183130: step: 732/529, loss: 0.02752971649169922 2023-01-21 12:05:03.317186: step: 736/529, loss: 0.011965370737016201 2023-01-21 12:05:04.472054: step: 740/529, loss: 0.0008462906116619706 2023-01-21 12:05:05.601797: step: 744/529, loss: 3.156661841785535e-05 2023-01-21 12:05:06.739160: step: 748/529, loss: 0.0010281563736498356 2023-01-21 12:05:07.870477: step: 752/529, loss: 0.6975657939910889 2023-01-21 12:05:09.007435: step: 756/529, loss: 0.0001426696835551411 2023-01-21 12:05:10.111721: step: 760/529, loss: 0.00016217232041526586 2023-01-21 12:05:11.247261: step: 764/529, loss: 0.0007642746204510331 2023-01-21 12:05:12.379945: step: 768/529, loss: 0.0023048401344567537 2023-01-21 12:05:13.502752: step: 772/529, loss: 0.12246724218130112 2023-01-21 12:05:14.637656: step: 776/529, loss: 0.009594726376235485 2023-01-21 12:05:15.757276: step: 780/529, loss: 0.0011806489201262593 2023-01-21 12:05:16.885667: step: 784/529, loss: 9.51766996877268e-05 2023-01-21 12:05:18.026916: step: 788/529, loss: 0.0018289090367034078 2023-01-21 12:05:19.167014: step: 792/529, loss: 0.029907381162047386 2023-01-21 12:05:20.315746: step: 796/529, loss: 0.030140208080410957 2023-01-21 12:05:21.457289: step: 800/529, loss: 0.00010604858107399195 2023-01-21 12:05:22.589265: step: 804/529, loss: 0.001104068709537387 2023-01-21 12:05:23.716408: step: 808/529, loss: 3.919601658708416e-05 2023-01-21 12:05:24.826508: step: 812/529, loss: 0.00020470618619583547 2023-01-21 12:05:25.955859: step: 816/529, loss: 0.07751578837633133 2023-01-21 12:05:27.086492: step: 820/529, loss: 0.055478475987911224 2023-01-21 12:05:28.239371: step: 824/529, loss: 0.0002076149103231728 2023-01-21 12:05:29.361495: step: 828/529, loss: 0.0015285492409020662 2023-01-21 12:05:30.491795: step: 832/529, loss: 0.0005379676586017013 2023-01-21 12:05:31.626288: step: 836/529, loss: 5.185840606689453 2023-01-21 12:05:32.739678: step: 840/529, loss: 0.06187605857849121 2023-01-21 12:05:33.880032: step: 844/529, loss: 0.01474685687571764 2023-01-21 12:05:35.019159: step: 848/529, loss: 0.00047321320744231343 2023-01-21 12:05:36.143657: step: 852/529, loss: 0.0045833587646484375 2023-01-21 12:05:37.271968: step: 856/529, loss: 0.04401881620287895 2023-01-21 12:05:38.408025: step: 860/529, loss: 0.0010614395141601562 2023-01-21 12:05:39.533621: step: 864/529, loss: 0.0009538650629110634 2023-01-21 12:05:40.649432: step: 868/529, loss: 0.020210839807987213 2023-01-21 12:05:41.769704: step: 872/529, loss: 0.046051979064941406 2023-01-21 12:05:42.966686: step: 876/529, loss: 0.017238711938261986 2023-01-21 12:05:44.111176: step: 880/529, loss: 0.03224611282348633 2023-01-21 12:05:45.260154: step: 884/529, loss: 0.026822710409760475 2023-01-21 12:05:46.386449: step: 888/529, loss: 0.021601391956210136 2023-01-21 12:05:47.504300: step: 892/529, loss: 0.002486324403434992 2023-01-21 12:05:48.653836: step: 896/529, loss: 0.0012258529895916581 2023-01-21 12:05:49.784869: step: 900/529, loss: 0.03464663028717041 2023-01-21 12:05:50.918688: step: 904/529, loss: 0.025440216064453125 2023-01-21 12:05:52.068645: step: 908/529, loss: 0.00488514918833971 2023-01-21 12:05:53.211820: step: 912/529, loss: 0.05924968793988228 2023-01-21 12:05:54.341087: step: 916/529, loss: 0.011789608746767044 2023-01-21 12:05:55.508378: step: 920/529, loss: 0.039760712534189224 2023-01-21 12:05:56.648299: step: 924/529, loss: 0.005775547120720148 2023-01-21 12:05:57.785344: step: 928/529, loss: 0.0044307708740234375 2023-01-21 12:05:58.910337: step: 932/529, loss: 0.2582445740699768 2023-01-21 12:06:00.024171: step: 936/529, loss: 0.010762214660644531 2023-01-21 12:06:01.168084: step: 940/529, loss: 0.05814332887530327 2023-01-21 12:06:02.293625: step: 944/529, loss: 0.4702529311180115 2023-01-21 12:06:03.407607: step: 948/529, loss: 0.14858825504779816 2023-01-21 12:06:04.545429: step: 952/529, loss: 0.004035663791000843 2023-01-21 12:06:05.710481: step: 956/529, loss: 0.0032676695846021175 2023-01-21 12:06:06.817890: step: 960/529, loss: 0.03535151854157448 2023-01-21 12:06:07.913941: step: 964/529, loss: 0.0012454986572265625 2023-01-21 12:06:09.039271: step: 968/529, loss: 0.005442905239760876 2023-01-21 12:06:10.187117: step: 972/529, loss: 0.002786731580272317 2023-01-21 12:06:11.354046: step: 976/529, loss: 0.005497360602021217 2023-01-21 12:06:12.487303: step: 980/529, loss: 0.02868823893368244 2023-01-21 12:06:13.618267: step: 984/529, loss: 0.001906967256218195 2023-01-21 12:06:14.763445: step: 988/529, loss: 0.00047397613525390625 2023-01-21 12:06:15.909408: step: 992/529, loss: 0.005586051847785711 2023-01-21 12:06:17.046827: step: 996/529, loss: 0.0027933118399232626 2023-01-21 12:06:18.192588: step: 1000/529, loss: 0.009168493561446667 2023-01-21 12:06:19.337184: step: 1004/529, loss: 0.01244125422090292 2023-01-21 12:06:20.509886: step: 1008/529, loss: 0.00011668205115711316 2023-01-21 12:06:21.651285: step: 1012/529, loss: 0.00028285980806685984 2023-01-21 12:06:22.785193: step: 1016/529, loss: 0.05203742906451225 2023-01-21 12:06:23.925502: step: 1020/529, loss: 0.014027118682861328 2023-01-21 12:06:25.046367: step: 1024/529, loss: 0.013417815789580345 2023-01-21 12:06:26.181867: step: 1028/529, loss: 0.014138412661850452 2023-01-21 12:06:27.312132: step: 1032/529, loss: 0.0002900123654399067 2023-01-21 12:06:28.465772: step: 1036/529, loss: 0.013271331787109375 2023-01-21 12:06:29.607793: step: 1040/529, loss: 0.0005670070531778038 2023-01-21 12:06:30.733629: step: 1044/529, loss: 0.0016752242809161544 2023-01-21 12:06:31.880522: step: 1048/529, loss: 0.02094888687133789 2023-01-21 12:06:33.018566: step: 1052/529, loss: 0.00692825298756361 2023-01-21 12:06:34.155302: step: 1056/529, loss: 0.015047455206513405 2023-01-21 12:06:35.290144: step: 1060/529, loss: 0.016508102416992188 2023-01-21 12:06:36.425216: step: 1064/529, loss: 0.015811825171113014 2023-01-21 12:06:37.583611: step: 1068/529, loss: 0.009197711944580078 2023-01-21 12:06:38.739014: step: 1072/529, loss: 0.0005840301746502519 2023-01-21 12:06:39.884825: step: 1076/529, loss: 0.05636577680706978 2023-01-21 12:06:41.051997: step: 1080/529, loss: 0.0032368660904467106 2023-01-21 12:06:42.206074: step: 1084/529, loss: 0.04805965721607208 2023-01-21 12:06:43.304952: step: 1088/529, loss: 0.008366644382476807 2023-01-21 12:06:44.471705: step: 1092/529, loss: 0.0026272772811353207 2023-01-21 12:06:45.616567: step: 1096/529, loss: 0.004409313201904297 2023-01-21 12:06:46.725620: step: 1100/529, loss: 0.0015930174849927425 2023-01-21 12:06:47.857890: step: 1104/529, loss: 0.004221916198730469 2023-01-21 12:06:49.007679: step: 1108/529, loss: 0.0007401466718874872 2023-01-21 12:06:50.139995: step: 1112/529, loss: 0.007759476080536842 2023-01-21 12:06:51.254241: step: 1116/529, loss: 0.010179519653320312 2023-01-21 12:06:52.409003: step: 1120/529, loss: 0.011844921857118607 2023-01-21 12:06:53.533390: step: 1124/529, loss: 0.004370689392089844 2023-01-21 12:06:54.679256: step: 1128/529, loss: 0.010753441601991653 2023-01-21 12:06:55.800165: step: 1132/529, loss: 0.002340126084163785 2023-01-21 12:06:56.955742: step: 1136/529, loss: 0.023742008954286575 2023-01-21 12:06:58.080246: step: 1140/529, loss: 0.0017427444690838456 2023-01-21 12:06:59.205324: step: 1144/529, loss: 0.0019750595092773438 2023-01-21 12:07:00.314590: step: 1148/529, loss: 0.02421693690121174 2023-01-21 12:07:01.467476: step: 1152/529, loss: 0.0011692047119140625 2023-01-21 12:07:02.636744: step: 1156/529, loss: 0.017490768805146217 2023-01-21 12:07:03.760985: step: 1160/529, loss: 0.00044231413630768657 2023-01-21 12:07:04.879194: step: 1164/529, loss: 0.004469490610063076 2023-01-21 12:07:06.011512: step: 1168/529, loss: 0.010682106018066406 2023-01-21 12:07:07.150733: step: 1172/529, loss: 0.04354462772607803 2023-01-21 12:07:08.299272: step: 1176/529, loss: 0.0017967225285246968 2023-01-21 12:07:09.424271: step: 1180/529, loss: 0.0011360167991369963 2023-01-21 12:07:10.552765: step: 1184/529, loss: 0.00029926298884674907 2023-01-21 12:07:11.702631: step: 1188/529, loss: 0.005479144863784313 2023-01-21 12:07:12.834173: step: 1192/529, loss: 0.06189250946044922 2023-01-21 12:07:13.962138: step: 1196/529, loss: 0.0006612300639972091 2023-01-21 12:07:15.086967: step: 1200/529, loss: 0.03426055982708931 2023-01-21 12:07:16.243714: step: 1204/529, loss: 0.0020305158104747534 2023-01-21 12:07:17.353694: step: 1208/529, loss: 0.0006325722206383944 2023-01-21 12:07:18.523607: step: 1212/529, loss: 0.022826479747891426 2023-01-21 12:07:19.658519: step: 1216/529, loss: 0.004124545957893133 2023-01-21 12:07:20.827952: step: 1220/529, loss: 0.09534435719251633 2023-01-21 12:07:21.970508: step: 1224/529, loss: 0.0014231681125238538 2023-01-21 12:07:23.128585: step: 1228/529, loss: 3.900528099620715e-05 2023-01-21 12:07:24.257477: step: 1232/529, loss: 0.03799267113208771 2023-01-21 12:07:25.399084: step: 1236/529, loss: 0.0035696029663085938 2023-01-21 12:07:26.523623: step: 1240/529, loss: 0.00010929107520496473 2023-01-21 12:07:27.660315: step: 1244/529, loss: 0.052858833223581314 2023-01-21 12:07:28.785187: step: 1248/529, loss: 0.008006763644516468 2023-01-21 12:07:29.946223: step: 1252/529, loss: 0.032919932156801224 2023-01-21 12:07:31.070337: step: 1256/529, loss: 0.02332305908203125 2023-01-21 12:07:32.224298: step: 1260/529, loss: 0.06275387108325958 2023-01-21 12:07:33.344042: step: 1264/529, loss: 0.033109571784734726 2023-01-21 12:07:34.489583: step: 1268/529, loss: 0.04139823839068413 2023-01-21 12:07:35.613884: step: 1272/529, loss: 0.10718798637390137 2023-01-21 12:07:36.752554: step: 1276/529, loss: 0.07257866859436035 2023-01-21 12:07:37.895539: step: 1280/529, loss: 0.0015145301586017013 2023-01-21 12:07:39.028480: step: 1284/529, loss: 0.004060077480971813 2023-01-21 12:07:40.164855: step: 1288/529, loss: 0.006896257400512695 2023-01-21 12:07:41.342564: step: 1292/529, loss: 0.00709037808701396 2023-01-21 12:07:42.480380: step: 1296/529, loss: 0.044680021703243256 2023-01-21 12:07:43.605399: step: 1300/529, loss: 0.05229535326361656 2023-01-21 12:07:44.758534: step: 1304/529, loss: 0.03190222010016441 2023-01-21 12:07:45.902915: step: 1308/529, loss: 4.00543212890625e-05 2023-01-21 12:07:47.077263: step: 1312/529, loss: 0.017056657001376152 2023-01-21 12:07:48.196667: step: 1316/529, loss: 0.020643234252929688 2023-01-21 12:07:49.322061: step: 1320/529, loss: -1.5258790426742053e-06 2023-01-21 12:07:50.431362: step: 1324/529, loss: -8.01086389401462e-06 2023-01-21 12:07:51.583599: step: 1328/529, loss: 0.003177070524543524 2023-01-21 12:07:52.729910: step: 1332/529, loss: 0.01180648896843195 2023-01-21 12:07:53.880283: step: 1336/529, loss: 0.013173295184969902 2023-01-21 12:07:55.002671: step: 1340/529, loss: 0.02430706098675728 2023-01-21 12:07:56.137391: step: 1344/529, loss: 0.03292083740234375 2023-01-21 12:07:57.250301: step: 1348/529, loss: 0.0043885232880711555 2023-01-21 12:07:58.391215: step: 1352/529, loss: 0.04280147701501846 2023-01-21 12:07:59.491687: step: 1356/529, loss: 0.011481190100312233 2023-01-21 12:08:00.609006: step: 1360/529, loss: 0.00034751894418150187 2023-01-21 12:08:01.728258: step: 1364/529, loss: 0.006107655353844166 2023-01-21 12:08:02.896642: step: 1368/529, loss: 0.009689903818070889 2023-01-21 12:08:04.014889: step: 1372/529, loss: 1.8596649169921875e-05 2023-01-21 12:08:05.157268: step: 1376/529, loss: 0.03760051727294922 2023-01-21 12:08:06.282698: step: 1380/529, loss: 0.010795784182846546 2023-01-21 12:08:07.423268: step: 1384/529, loss: 0.012355995364487171 2023-01-21 12:08:08.574412: step: 1388/529, loss: 0.002890014788135886 2023-01-21 12:08:09.716482: step: 1392/529, loss: 0.004197883419692516 2023-01-21 12:08:10.861054: step: 1396/529, loss: 0.003769779345020652 2023-01-21 12:08:11.977522: step: 1400/529, loss: 9.193420555675402e-05 2023-01-21 12:08:13.110928: step: 1404/529, loss: 0.0003452301025390625 2023-01-21 12:08:14.245505: step: 1408/529, loss: 0.00027046201284974813 2023-01-21 12:08:15.369393: step: 1412/529, loss: 0.022012829780578613 2023-01-21 12:08:16.471923: step: 1416/529, loss: 0.00020923613919876516 2023-01-21 12:08:17.601613: step: 1420/529, loss: 0.0612335205078125 2023-01-21 12:08:18.768742: step: 1424/529, loss: 0.0005936623201705515 2023-01-21 12:08:19.922906: step: 1428/529, loss: 0.0163332000374794 2023-01-21 12:08:21.062797: step: 1432/529, loss: 0.009732437320053577 2023-01-21 12:08:22.195543: step: 1436/529, loss: 0.0014553071232512593 2023-01-21 12:08:23.328998: step: 1440/529, loss: 0.04503779485821724 2023-01-21 12:08:24.475156: step: 1444/529, loss: 0.00019503833027556539 2023-01-21 12:08:25.617390: step: 1448/529, loss: 0.0036336900666356087 2023-01-21 12:08:26.756573: step: 1452/529, loss: 0.005965232849121094 2023-01-21 12:08:27.885259: step: 1456/529, loss: 0.002640151884406805 2023-01-21 12:08:29.021202: step: 1460/529, loss: 0.12068986892700195 2023-01-21 12:08:30.158043: step: 1464/529, loss: 0.009997272863984108 2023-01-21 12:08:31.299012: step: 1468/529, loss: 0.06069488823413849 2023-01-21 12:08:32.433863: step: 1472/529, loss: 6.351471529342234e-05 2023-01-21 12:08:33.601505: step: 1476/529, loss: 0.0010111809242516756 2023-01-21 12:08:34.734286: step: 1480/529, loss: 0.07010859996080399 2023-01-21 12:08:35.858302: step: 1484/529, loss: 0.007163238245993853 2023-01-21 12:08:36.988936: step: 1488/529, loss: 0.00947561301290989 2023-01-21 12:08:38.123768: step: 1492/529, loss: 0.038003161549568176 2023-01-21 12:08:39.271385: step: 1496/529, loss: 0.013702679425477982 2023-01-21 12:08:40.410600: step: 1500/529, loss: 0.03276081383228302 2023-01-21 12:08:41.546891: step: 1504/529, loss: 0.42165660858154297 2023-01-21 12:08:42.693144: step: 1508/529, loss: 0.0001623153657419607 2023-01-21 12:08:43.823467: step: 1512/529, loss: 0.06620216369628906 2023-01-21 12:08:44.959964: step: 1516/529, loss: 0.00433802604675293 2023-01-21 12:08:46.120887: step: 1520/529, loss: 0.0019710541237145662 2023-01-21 12:08:47.275626: step: 1524/529, loss: 0.09248819202184677 2023-01-21 12:08:48.411366: step: 1528/529, loss: 0.0026967048179358244 2023-01-21 12:08:49.531518: step: 1532/529, loss: 0.02864818647503853 2023-01-21 12:08:50.652952: step: 1536/529, loss: 0.0017695426940917969 2023-01-21 12:08:51.775178: step: 1540/529, loss: 0.011477851308882236 2023-01-21 12:08:52.908249: step: 1544/529, loss: 0.0023189543280750513 2023-01-21 12:08:54.047465: step: 1548/529, loss: 0.009379958733916283 2023-01-21 12:08:55.176593: step: 1552/529, loss: 0.011292267590761185 2023-01-21 12:08:56.290083: step: 1556/529, loss: 0.021019697189331055 2023-01-21 12:08:57.401402: step: 1560/529, loss: 0.004131889436393976 2023-01-21 12:08:58.551410: step: 1564/529, loss: 0.0018106460338458419 2023-01-21 12:08:59.676385: step: 1568/529, loss: 0.001531314803287387 2023-01-21 12:09:00.823696: step: 1572/529, loss: 0.05347738042473793 2023-01-21 12:09:01.961748: step: 1576/529, loss: 0.0026765824295580387 2023-01-21 12:09:03.136011: step: 1580/529, loss: 0.02463665045797825 2023-01-21 12:09:04.298567: step: 1584/529, loss: 0.022698307409882545 2023-01-21 12:09:05.433952: step: 1588/529, loss: 0.0006662369123660028 2023-01-21 12:09:06.590359: step: 1592/529, loss: 0.00033886433811858296 2023-01-21 12:09:07.744623: step: 1596/529, loss: 0.04249248653650284 2023-01-21 12:09:08.870922: step: 1600/529, loss: 0.010273169726133347 2023-01-21 12:09:10.011724: step: 1604/529, loss: 0.003914594650268555 2023-01-21 12:09:11.147214: step: 1608/529, loss: 0.010500717908143997 2023-01-21 12:09:12.280638: step: 1612/529, loss: 0.03784504160284996 2023-01-21 12:09:13.433882: step: 1616/529, loss: 0.029215574264526367 2023-01-21 12:09:14.558959: step: 1620/529, loss: 0.007312011905014515 2023-01-21 12:09:15.678783: step: 1624/529, loss: 0.001105880830436945 2023-01-21 12:09:16.827519: step: 1628/529, loss: 0.0091393468901515 2023-01-21 12:09:17.966539: step: 1632/529, loss: 0.0004163742414675653 2023-01-21 12:09:19.123700: step: 1636/529, loss: 0.025879859924316406 2023-01-21 12:09:20.267873: step: 1640/529, loss: 0.007389450445771217 2023-01-21 12:09:21.413045: step: 1644/529, loss: 0.05246887356042862 2023-01-21 12:09:22.541830: step: 1648/529, loss: 0.00019731521024368703 2023-01-21 12:09:23.676474: step: 1652/529, loss: 0.03374490886926651 2023-01-21 12:09:24.820468: step: 1656/529, loss: 0.008224106393754482 2023-01-21 12:09:25.929143: step: 1660/529, loss: 0.004824733827263117 2023-01-21 12:09:27.079079: step: 1664/529, loss: 0.0060592652298510075 2023-01-21 12:09:28.188001: step: 1668/529, loss: 0.8607851266860962 2023-01-21 12:09:29.305276: step: 1672/529, loss: 0.006881999783217907 2023-01-21 12:09:30.379486: step: 1676/529, loss: 0.0006945609929971397 2023-01-21 12:09:31.480642: step: 1680/529, loss: 0.025945473462343216 2023-01-21 12:09:32.621024: step: 1684/529, loss: 0.008211707696318626 2023-01-21 12:09:33.754433: step: 1688/529, loss: 0.022715091705322266 2023-01-21 12:09:34.894278: step: 1692/529, loss: 0.011531829833984375 2023-01-21 12:09:36.003795: step: 1696/529, loss: 0.009366989135742188 2023-01-21 12:09:37.114298: step: 1700/529, loss: 0.033548545092344284 2023-01-21 12:09:38.251816: step: 1704/529, loss: 0.0002498626708984375 2023-01-21 12:09:39.375328: step: 1708/529, loss: 0.006572151090949774 2023-01-21 12:09:40.501699: step: 1712/529, loss: 0.009911156259477139 2023-01-21 12:09:41.625025: step: 1716/529, loss: 0.028431225568056107 2023-01-21 12:09:42.770809: step: 1720/529, loss: 0.11921444535255432 2023-01-21 12:09:43.921310: step: 1724/529, loss: 0.0007851601112633944 2023-01-21 12:09:45.043988: step: 1728/529, loss: 0.002255010651424527 2023-01-21 12:09:46.176886: step: 1732/529, loss: 0.0001012802094919607 2023-01-21 12:09:47.294943: step: 1736/529, loss: 0.011163998395204544 2023-01-21 12:09:48.412375: step: 1740/529, loss: 0.00021929740614723414 2023-01-21 12:09:49.514280: step: 1744/529, loss: 0.0036584853660315275 2023-01-21 12:09:50.650312: step: 1748/529, loss: 9.746551950229332e-05 2023-01-21 12:09:51.829334: step: 1752/529, loss: 0.0025769234634935856 2023-01-21 12:09:52.953361: step: 1756/529, loss: 0.010653305798768997 2023-01-21 12:09:54.061899: step: 1760/529, loss: 0.0016676426166668534 2023-01-21 12:09:55.211939: step: 1764/529, loss: 0.014390897937119007 2023-01-21 12:09:56.347133: step: 1768/529, loss: 0.016907215118408203 2023-01-21 12:09:57.477099: step: 1772/529, loss: 0.00181837088894099 2023-01-21 12:09:58.617209: step: 1776/529, loss: 0.02114715427160263 2023-01-21 12:09:59.761045: step: 1780/529, loss: 0.0076562403701245785 2023-01-21 12:10:00.904966: step: 1784/529, loss: 0.013175916858017445 2023-01-21 12:10:02.034731: step: 1788/529, loss: 0.008378982543945312 2023-01-21 12:10:03.182689: step: 1792/529, loss: 0.0009254455799236894 2023-01-21 12:10:04.290903: step: 1796/529, loss: 3.719329833984375e-05 2023-01-21 12:10:05.418677: step: 1800/529, loss: 0.005161189939826727 2023-01-21 12:10:06.529922: step: 1804/529, loss: 0.005507952068001032 2023-01-21 12:10:07.643656: step: 1808/529, loss: 6.432532973121852e-05 2023-01-21 12:10:08.775715: step: 1812/529, loss: 0.013687323778867722 2023-01-21 12:10:09.916061: step: 1816/529, loss: 0.01939830742776394 2023-01-21 12:10:11.040051: step: 1820/529, loss: 0.012773562222719193 2023-01-21 12:10:12.186597: step: 1824/529, loss: 0.014494610950350761 2023-01-21 12:10:13.326065: step: 1828/529, loss: 0.013984155841171741 2023-01-21 12:10:14.451812: step: 1832/529, loss: 0.0008624077308923006 2023-01-21 12:10:15.556769: step: 1836/529, loss: 0.0001469850685680285 2023-01-21 12:10:16.718095: step: 1840/529, loss: 0.0025678633246570826 2023-01-21 12:10:17.879780: step: 1844/529, loss: 0.0008081913110800087 2023-01-21 12:10:19.033737: step: 1848/529, loss: 0.0008213043329305947 2023-01-21 12:10:20.181602: step: 1852/529, loss: 0.0001049041748046875 2023-01-21 12:10:21.310244: step: 1856/529, loss: 0.026462554931640625 2023-01-21 12:10:22.478737: step: 1860/529, loss: 0.0002541542053222656 2023-01-21 12:10:23.645119: step: 1864/529, loss: 0.0042692190036177635 2023-01-21 12:10:24.767527: step: 1868/529, loss: 0.0003243446699343622 2023-01-21 12:10:25.898191: step: 1872/529, loss: 0.050055406987667084 2023-01-21 12:10:27.043500: step: 1876/529, loss: 0.004984998609870672 2023-01-21 12:10:28.186940: step: 1880/529, loss: 0.009220505133271217 2023-01-21 12:10:29.338405: step: 1884/529, loss: 0.008108139038085938 2023-01-21 12:10:30.462175: step: 1888/529, loss: 0.013308144174516201 2023-01-21 12:10:31.590407: step: 1892/529, loss: 0.0006589888944290578 2023-01-21 12:10:32.743285: step: 1896/529, loss: 0.06773176044225693 2023-01-21 12:10:33.875069: step: 1900/529, loss: 1.068115216185106e-05 2023-01-21 12:10:35.014736: step: 1904/529, loss: 0.04861483350396156 2023-01-21 12:10:36.154390: step: 1908/529, loss: 0.040887072682380676 2023-01-21 12:10:37.298732: step: 1912/529, loss: 0.00913095474243164 2023-01-21 12:10:38.433484: step: 1916/529, loss: -1.4972686585679185e-05 2023-01-21 12:10:39.558004: step: 1920/529, loss: 0.00932455062866211 2023-01-21 12:10:40.694641: step: 1924/529, loss: 0.002048921538516879 2023-01-21 12:10:41.881607: step: 1928/529, loss: 7.524490501964465e-05 2023-01-21 12:10:43.012163: step: 1932/529, loss: 0.004367828369140625 2023-01-21 12:10:44.159817: step: 1936/529, loss: 0.019753647968173027 2023-01-21 12:10:45.337514: step: 1940/529, loss: 0.17170429229736328 2023-01-21 12:10:46.477276: step: 1944/529, loss: 0.0036743164528161287 2023-01-21 12:10:47.635725: step: 1948/529, loss: 0.014732170850038528 2023-01-21 12:10:48.853447: step: 1952/529, loss: 0.0004589080926962197 2023-01-21 12:10:50.001642: step: 1956/529, loss: 0.0001373291015625 2023-01-21 12:10:51.165871: step: 1960/529, loss: 2.222557783126831 2023-01-21 12:10:52.288647: step: 1964/529, loss: 0.0262908935546875 2023-01-21 12:10:53.419095: step: 1968/529, loss: 0.09304027259349823 2023-01-21 12:10:54.613386: step: 1972/529, loss: 0.0005405425908975303 2023-01-21 12:10:55.741461: step: 1976/529, loss: 3.4904482163256034e-05 2023-01-21 12:10:56.843081: step: 1980/529, loss: 0.0441831573843956 2023-01-21 12:10:57.964139: step: 1984/529, loss: 0.0041325571946799755 2023-01-21 12:10:59.101352: step: 1988/529, loss: 0.055832672864198685 2023-01-21 12:11:00.250068: step: 1992/529, loss: 0.01119308453053236 2023-01-21 12:11:01.387410: step: 1996/529, loss: 0.002911376766860485 2023-01-21 12:11:02.502202: step: 2000/529, loss: 0.034895945340394974 2023-01-21 12:11:03.631599: step: 2004/529, loss: 0.00010619163367664441 2023-01-21 12:11:04.732031: step: 2008/529, loss: 0.0012832642532885075 2023-01-21 12:11:05.829756: step: 2012/529, loss: 0.06861686706542969 2023-01-21 12:11:06.981305: step: 2016/529, loss: 2.8896332878503017e-05 2023-01-21 12:11:08.132279: step: 2020/529, loss: 0.007028484251350164 2023-01-21 12:11:09.254089: step: 2024/529, loss: 2.5081633793888614e-05 2023-01-21 12:11:10.368681: step: 2028/529, loss: 0.00206584925763309 2023-01-21 12:11:11.508514: step: 2032/529, loss: 0.00129871373064816 2023-01-21 12:11:12.632757: step: 2036/529, loss: 0.010124778375029564 2023-01-21 12:11:13.758874: step: 2040/529, loss: 0.0001924514799611643 2023-01-21 12:11:14.872070: step: 2044/529, loss: 0.006847286596894264 2023-01-21 12:11:16.020973: step: 2048/529, loss: 0.03153085708618164 2023-01-21 12:11:17.172289: step: 2052/529, loss: 0.000244140625 2023-01-21 12:11:18.301243: step: 2056/529, loss: 0.005618286319077015 2023-01-21 12:11:19.440701: step: 2060/529, loss: 0.004194832406938076 2023-01-21 12:11:20.580128: step: 2064/529, loss: 0.0020910261664539576 2023-01-21 12:11:21.721195: step: 2068/529, loss: 0.002396106719970703 2023-01-21 12:11:22.867651: step: 2072/529, loss: 0.0002865791320800781 2023-01-21 12:11:24.011926: step: 2076/529, loss: 0.002443695208057761 2023-01-21 12:11:25.174076: step: 2080/529, loss: 0.00046138762263581157 2023-01-21 12:11:26.288990: step: 2084/529, loss: 0.0004123687685932964 2023-01-21 12:11:27.408101: step: 2088/529, loss: 0.0017341614002361894 2023-01-21 12:11:28.542421: step: 2092/529, loss: 0.0003719329833984375 2023-01-21 12:11:29.686477: step: 2096/529, loss: 0.004864406771957874 2023-01-21 12:11:30.839590: step: 2100/529, loss: 0.0026886940468102694 2023-01-21 12:11:31.958276: step: 2104/529, loss: 0.033385373651981354 2023-01-21 12:11:33.080879: step: 2108/529, loss: 0.018949126824736595 2023-01-21 12:11:34.227964: step: 2112/529, loss: 5.569458153331652e-05 2023-01-21 12:11:35.364657: step: 2116/529, loss: 0.0035600662231445312 ================================================== Loss: 0.036 -------------------- Dev: {'event': {'p': 0.6273022751895991, 'r': 0.7709720372836218, 'f1': 0.6917562724014338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Test: {'event': {'p': 0.6206896551724138, 'r': 0.7717265353418308, 'f1': 0.6880165289256198}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Chinese: {'event': {'p': 0.5930232558139535, 'r': 0.9444444444444444, 'f1': 0.7285714285714286}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Korean: {'event': {'p': 0.625, 'r': 0.5555555555555556, 'f1': 0.5882352941176471}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Russian: {'event': {'p': 0.47368421052631576, 'r': 0.5, 'f1': 0.4864864864864865}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 21 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:12:16.228785: step: 4/529, loss: 0.003976917359977961 2023-01-21 12:12:17.368422: step: 8/529, loss: 0.0007116317865438759 2023-01-21 12:12:18.499038: step: 12/529, loss: 9.403229341842234e-05 2023-01-21 12:12:19.628301: step: 16/529, loss: 0.0007730484358035028 2023-01-21 12:12:20.767554: step: 20/529, loss: 0.01691007800400257 2023-01-21 12:12:21.912793: step: 24/529, loss: 0.00024003982252907008 2023-01-21 12:12:23.051052: step: 28/529, loss: 0.001754379365593195 2023-01-21 12:12:24.163882: step: 32/529, loss: 7.877349707996473e-05 2023-01-21 12:12:25.277833: step: 36/529, loss: 0.0045642852783203125 2023-01-21 12:12:26.423144: step: 40/529, loss: 0.0006595611339434981 2023-01-21 12:12:27.563398: step: 44/529, loss: 0.0001396715670125559 2023-01-21 12:12:28.710614: step: 48/529, loss: 0.03726153448224068 2023-01-21 12:12:29.858336: step: 52/529, loss: 0.004734421148896217 2023-01-21 12:12:30.964299: step: 56/529, loss: 3.910065061063506e-06 2023-01-21 12:12:32.143383: step: 60/529, loss: 0.0041914465837180614 2023-01-21 12:12:33.316018: step: 64/529, loss: 0.042818259447813034 2023-01-21 12:12:34.484514: step: 68/529, loss: 0.0004276275576557964 2023-01-21 12:12:35.593831: step: 72/529, loss: 0.0005197286955080926 2023-01-21 12:12:36.721152: step: 76/529, loss: 0.0005234241834841669 2023-01-21 12:12:37.847570: step: 80/529, loss: 3.4046173823298886e-05 2023-01-21 12:12:38.969817: step: 84/529, loss: 0.06422004848718643 2023-01-21 12:12:40.127110: step: 88/529, loss: -6.961822691664565e-06 2023-01-21 12:12:41.281409: step: 92/529, loss: 0.6131331920623779 2023-01-21 12:12:42.409676: step: 96/529, loss: 0.0006634712335653603 2023-01-21 12:12:43.572980: step: 100/529, loss: 0.020165253430604935 2023-01-21 12:12:44.732235: step: 104/529, loss: 0.008333302102982998 2023-01-21 12:12:45.907295: step: 108/529, loss: 0.0014547348255291581 2023-01-21 12:12:47.035594: step: 112/529, loss: 0.0008433342445641756 2023-01-21 12:12:48.138865: step: 116/529, loss: 0.0036572455428540707 2023-01-21 12:12:49.296862: step: 120/529, loss: 0.016436384990811348 2023-01-21 12:12:50.441292: step: 124/529, loss: 0.0006477356655523181 2023-01-21 12:12:51.567432: step: 128/529, loss: 0.008415688760578632 2023-01-21 12:12:52.670859: step: 132/529, loss: 0.06058826297521591 2023-01-21 12:12:53.809696: step: 136/529, loss: 0.1271534115076065 2023-01-21 12:12:54.925805: step: 140/529, loss: 1.8882754375226796e-05 2023-01-21 12:12:56.094988: step: 144/529, loss: 0.01803712733089924 2023-01-21 12:12:57.232280: step: 148/529, loss: 0.0022476196754723787 2023-01-21 12:12:58.379124: step: 152/529, loss: 0.025887491181492805 2023-01-21 12:12:59.550028: step: 156/529, loss: 0.003461647080257535 2023-01-21 12:13:00.679433: step: 160/529, loss: 4.692077709478326e-05 2023-01-21 12:13:01.814074: step: 164/529, loss: 0.0012077331775799394 2023-01-21 12:13:02.961851: step: 168/529, loss: 0.0024536133278161287 2023-01-21 12:13:04.086790: step: 172/529, loss: 0.0021148682571947575 2023-01-21 12:13:05.210032: step: 176/529, loss: 0.024805927649140358 2023-01-21 12:13:06.366392: step: 180/529, loss: 0.0050389766693115234 2023-01-21 12:13:07.487407: step: 184/529, loss: 0.008747505955398083 2023-01-21 12:13:08.608060: step: 188/529, loss: 0.010238075628876686 2023-01-21 12:13:09.751621: step: 192/529, loss: 0.00833969097584486 2023-01-21 12:13:10.874981: step: 196/529, loss: 0.0011187553172931075 2023-01-21 12:13:12.006206: step: 200/529, loss: 0.0023651123046875 2023-01-21 12:13:13.133755: step: 204/529, loss: 0.0006448745843954384 2023-01-21 12:13:14.261200: step: 208/529, loss: 0.007282924838364124 2023-01-21 12:13:15.375666: step: 212/529, loss: 0.01189346332103014 2023-01-21 12:13:16.523555: step: 216/529, loss: 0.00533142127096653 2023-01-21 12:13:17.655521: step: 220/529, loss: 3.0517576306010596e-06 2023-01-21 12:13:18.789916: step: 224/529, loss: 0.02095642127096653 2023-01-21 12:13:19.913855: step: 228/529, loss: 0.015183067880570889 2023-01-21 12:13:21.079621: step: 232/529, loss: 0.0008382797823287547 2023-01-21 12:13:22.195399: step: 236/529, loss: 0.0733637809753418 2023-01-21 12:13:23.330345: step: 240/529, loss: 3.5977365769213066e-05 2023-01-21 12:13:24.478847: step: 244/529, loss: 0.0012701033847406507 2023-01-21 12:13:25.607674: step: 248/529, loss: 0.06110992282629013 2023-01-21 12:13:26.731931: step: 252/529, loss: 0.02470588870346546 2023-01-21 12:13:27.869143: step: 256/529, loss: 0.00014648436626885086 2023-01-21 12:13:29.019556: step: 260/529, loss: 0.0009017467382363975 2023-01-21 12:13:30.176774: step: 264/529, loss: 0.00015544891357421875 2023-01-21 12:13:31.303001: step: 268/529, loss: 0.023493194952607155 2023-01-21 12:13:32.458297: step: 272/529, loss: 0.00033969880314543843 2023-01-21 12:13:33.606430: step: 276/529, loss: 0.05209064856171608 2023-01-21 12:13:34.789947: step: 280/529, loss: 0.2062522917985916 2023-01-21 12:13:35.951092: step: 284/529, loss: 0.014591837301850319 2023-01-21 12:13:37.094522: step: 288/529, loss: 0.02383279800415039 2023-01-21 12:13:38.267451: step: 292/529, loss: 0.008657360449433327 2023-01-21 12:13:39.406101: step: 296/529, loss: 8.873939805198461e-05 2023-01-21 12:13:40.533767: step: 300/529, loss: 0.014673185534775257 2023-01-21 12:13:41.688270: step: 304/529, loss: 0.010455131530761719 2023-01-21 12:13:42.812194: step: 308/529, loss: 0.011625290848314762 2023-01-21 12:13:43.941240: step: 312/529, loss: 0.029275894165039062 2023-01-21 12:13:45.108127: step: 316/529, loss: 0.010558987967669964 2023-01-21 12:13:46.249716: step: 320/529, loss: 0.07185535132884979 2023-01-21 12:13:47.404722: step: 324/529, loss: 0.06779918074607849 2023-01-21 12:13:48.521594: step: 328/529, loss: 0.0048233033157885075 2023-01-21 12:13:49.653529: step: 332/529, loss: 0.0001733779936330393 2023-01-21 12:13:50.781997: step: 336/529, loss: 0.000358390825567767 2023-01-21 12:13:51.928950: step: 340/529, loss: 0.00010032653517555445 2023-01-21 12:13:53.079352: step: 344/529, loss: 0.0033226015511900187 2023-01-21 12:13:54.195135: step: 348/529, loss: 0.0001926422119140625 2023-01-21 12:13:55.332803: step: 352/529, loss: 0.0001847267267294228 2023-01-21 12:13:56.477903: step: 356/529, loss: 0.002921485807746649 2023-01-21 12:13:57.607686: step: 360/529, loss: 0.0011449814774096012 2023-01-21 12:13:58.753898: step: 364/529, loss: 1.831054760259576e-05 2023-01-21 12:13:59.946437: step: 368/529, loss: 0.0016298294067382812 2023-01-21 12:14:01.081768: step: 372/529, loss: 7.228850881801918e-05 2023-01-21 12:14:02.205018: step: 376/529, loss: 1.373290979245212e-05 2023-01-21 12:14:03.375991: step: 380/529, loss: 0.0043497090227901936 2023-01-21 12:14:04.533589: step: 384/529, loss: 0.026889802888035774 2023-01-21 12:14:05.672318: step: 388/529, loss: 0.0010818481678143144 2023-01-21 12:14:06.811184: step: 392/529, loss: 0.004097366705536842 2023-01-21 12:14:07.954596: step: 396/529, loss: 0.022572899237275124 2023-01-21 12:14:09.105492: step: 400/529, loss: 0.002540969755500555 2023-01-21 12:14:10.238440: step: 404/529, loss: 0.006014490034431219 2023-01-21 12:14:11.391626: step: 408/529, loss: 0.0003115653817076236 2023-01-21 12:14:12.501394: step: 412/529, loss: 0.002440929412841797 2023-01-21 12:14:13.645693: step: 416/529, loss: 0.0684998482465744 2023-01-21 12:14:14.781775: step: 420/529, loss: 0.061145883053541183 2023-01-21 12:14:15.923883: step: 424/529, loss: 0.003925133030861616 2023-01-21 12:14:17.074089: step: 428/529, loss: 0.0005898475646972656 2023-01-21 12:14:18.208096: step: 432/529, loss: 0.000568389892578125 2023-01-21 12:14:19.328450: step: 436/529, loss: 0.0006813049549236894 2023-01-21 12:14:20.438348: step: 440/529, loss: 0.00014905929856467992 2023-01-21 12:14:21.595119: step: 444/529, loss: 0.00042982102604582906 2023-01-21 12:14:22.711364: step: 448/529, loss: 0.0012778282398357987 2023-01-21 12:14:23.859955: step: 452/529, loss: 0.03290600702166557 2023-01-21 12:14:25.004628: step: 456/529, loss: 0.008971787057816982 2023-01-21 12:14:26.143614: step: 460/529, loss: 0.0010578155051916838 2023-01-21 12:14:27.278437: step: 464/529, loss: 0.00021486282639671117 2023-01-21 12:14:28.426149: step: 468/529, loss: 0.01743030734360218 2023-01-21 12:14:29.570546: step: 472/529, loss: 0.03139457851648331 2023-01-21 12:14:30.692272: step: 476/529, loss: 0.00033626556978560984 2023-01-21 12:14:31.835413: step: 480/529, loss: 0.0017955780494958162 2023-01-21 12:14:32.951793: step: 484/529, loss: 0.0002829551522154361 2023-01-21 12:14:34.079639: step: 488/529, loss: 0.01783580705523491 2023-01-21 12:14:35.215418: step: 492/529, loss: 0.01042957417666912 2023-01-21 12:14:36.307904: step: 496/529, loss: 0.00020550489716697484 2023-01-21 12:14:37.429812: step: 500/529, loss: 0.0010252000065520406 2023-01-21 12:14:38.575666: step: 504/529, loss: 0.02346658706665039 2023-01-21 12:14:39.712802: step: 508/529, loss: 0.000286275171674788 2023-01-21 12:14:40.851355: step: 512/529, loss: 0.005281639285385609 2023-01-21 12:14:42.005856: step: 516/529, loss: 0.010306740179657936 2023-01-21 12:14:43.101664: step: 520/529, loss: 0.00015897752018645406 2023-01-21 12:14:44.261823: step: 524/529, loss: 0.02437725104391575 2023-01-21 12:14:45.372928: step: 528/529, loss: 0.0004380226309876889 2023-01-21 12:14:46.513990: step: 532/529, loss: 0.00303914537653327 2023-01-21 12:14:47.621260: step: 536/529, loss: 0.0004276275576557964 2023-01-21 12:14:48.757704: step: 540/529, loss: 0.00116815569344908 2023-01-21 12:14:49.894888: step: 544/529, loss: 0.0029674528632313013 2023-01-21 12:14:51.046099: step: 548/529, loss: 0.05657081678509712 2023-01-21 12:14:52.190587: step: 552/529, loss: 0.0002506732998881489 2023-01-21 12:14:53.315915: step: 556/529, loss: 0.004369258880615234 2023-01-21 12:14:54.439089: step: 560/529, loss: 0.03371744230389595 2023-01-21 12:14:55.569906: step: 564/529, loss: 0.013393736444413662 2023-01-21 12:14:56.690139: step: 568/529, loss: 0.022926712408661842 2023-01-21 12:14:57.825705: step: 572/529, loss: 0.021451378241181374 2023-01-21 12:14:58.963020: step: 576/529, loss: 0.040910910815000534 2023-01-21 12:15:00.107227: step: 580/529, loss: 0.0003177642938680947 2023-01-21 12:15:01.247704: step: 584/529, loss: 0.0004455089510884136 2023-01-21 12:15:02.373429: step: 588/529, loss: 0.06816544383764267 2023-01-21 12:15:03.541149: step: 592/529, loss: 0.018665695562958717 2023-01-21 12:15:04.714638: step: 596/529, loss: 0.007628059946000576 2023-01-21 12:15:05.849346: step: 600/529, loss: 0.026865579187870026 2023-01-21 12:15:06.992856: step: 604/529, loss: 0.03370781987905502 2023-01-21 12:15:08.135058: step: 608/529, loss: 0.6259747743606567 2023-01-21 12:15:09.271448: step: 612/529, loss: 0.028895188122987747 2023-01-21 12:15:10.414790: step: 616/529, loss: 6.103515625e-05 2023-01-21 12:15:11.535105: step: 620/529, loss: 0.027737855911254883 2023-01-21 12:15:12.661836: step: 624/529, loss: 0.013202572241425514 2023-01-21 12:15:13.792819: step: 628/529, loss: 1.4925002687959932e-05 2023-01-21 12:15:14.938690: step: 632/529, loss: 0.019480038434267044 2023-01-21 12:15:16.073021: step: 636/529, loss: 0.002047896385192871 2023-01-21 12:15:17.207619: step: 640/529, loss: 3.80516066798009e-05 2023-01-21 12:15:18.367579: step: 644/529, loss: 0.0040798187255859375 2023-01-21 12:15:19.489778: step: 648/529, loss: 0.06549052894115448 2023-01-21 12:15:20.632300: step: 652/529, loss: 0.0027616501320153475 2023-01-21 12:15:21.775244: step: 656/529, loss: 0.001287317369133234 2023-01-21 12:15:22.889396: step: 660/529, loss: 0.0009366989252157509 2023-01-21 12:15:24.085402: step: 664/529, loss: 6.027221752447076e-05 2023-01-21 12:15:25.200265: step: 668/529, loss: 0.005042982287704945 2023-01-21 12:15:26.328395: step: 672/529, loss: 0.0068329814821481705 2023-01-21 12:15:27.465036: step: 676/529, loss: 0.007976150140166283 2023-01-21 12:15:28.577886: step: 680/529, loss: 3.471374657237902e-05 2023-01-21 12:15:29.725705: step: 684/529, loss: 0.0006537437438964844 2023-01-21 12:15:30.886943: step: 688/529, loss: 0.004265117924660444 2023-01-21 12:15:32.036400: step: 692/529, loss: 0.009064865298569202 2023-01-21 12:15:33.159536: step: 696/529, loss: 0.0008216142887249589 2023-01-21 12:15:34.286367: step: 700/529, loss: 0.0013271331554278731 2023-01-21 12:15:35.431249: step: 704/529, loss: 6.461143493652344e-05 2023-01-21 12:15:36.586115: step: 708/529, loss: 2.47955322265625e-05 2023-01-21 12:15:37.725587: step: 712/529, loss: 0.06059322506189346 2023-01-21 12:15:38.892151: step: 716/529, loss: 0.0009393692016601562 2023-01-21 12:15:40.037010: step: 720/529, loss: 0.013641834259033203 2023-01-21 12:15:41.168655: step: 724/529, loss: 0.005165958311408758 2023-01-21 12:15:42.325273: step: 728/529, loss: 0.021885670721530914 2023-01-21 12:15:43.482245: step: 732/529, loss: 0.0003195345343556255 2023-01-21 12:15:44.643880: step: 736/529, loss: 2.5653840566519648e-05 2023-01-21 12:15:45.778152: step: 740/529, loss: 5.154609971214086e-05 2023-01-21 12:15:46.931798: step: 744/529, loss: 0.013858318328857422 2023-01-21 12:15:48.049368: step: 748/529, loss: 0.04328355938196182 2023-01-21 12:15:49.190951: step: 752/529, loss: 0.00010395050048828125 2023-01-21 12:15:50.328897: step: 756/529, loss: 0.055783748626708984 2023-01-21 12:15:51.463550: step: 760/529, loss: 0.005009269807487726 2023-01-21 12:15:52.587802: step: 764/529, loss: 0.02104063145816326 2023-01-21 12:15:53.724352: step: 768/529, loss: 0.009302712045609951 2023-01-21 12:15:54.860283: step: 772/529, loss: 0.00038337710429914296 2023-01-21 12:15:55.993063: step: 776/529, loss: 0.0026230812072753906 2023-01-21 12:15:57.133905: step: 780/529, loss: 0.00015726088895462453 2023-01-21 12:15:58.274747: step: 784/529, loss: 0.0016744614113122225 2023-01-21 12:15:59.390709: step: 788/529, loss: 0.0002925872977357358 2023-01-21 12:16:00.526142: step: 792/529, loss: 0.0008006095886230469 2023-01-21 12:16:01.649771: step: 796/529, loss: 0.00017790794663596898 2023-01-21 12:16:02.793485: step: 800/529, loss: 0.00040311814518645406 2023-01-21 12:16:03.935375: step: 804/529, loss: 1.7353135347366333 2023-01-21 12:16:05.036870: step: 808/529, loss: 0.0259794220328331 2023-01-21 12:16:06.195247: step: 812/529, loss: 0.016849517822265625 2023-01-21 12:16:07.304080: step: 816/529, loss: 0.00044155120849609375 2023-01-21 12:16:08.436106: step: 820/529, loss: 0.08364319801330566 2023-01-21 12:16:09.600740: step: 824/529, loss: 0.00048542022705078125 2023-01-21 12:16:10.730000: step: 828/529, loss: 0.0005571842193603516 2023-01-21 12:16:11.878226: step: 832/529, loss: 0.015031195245683193 2023-01-21 12:16:12.999778: step: 836/529, loss: 0.026267431676387787 2023-01-21 12:16:14.133895: step: 840/529, loss: 1.0108947208209429e-05 2023-01-21 12:16:15.228546: step: 844/529, loss: 0.00038313865661621094 2023-01-21 12:16:16.377398: step: 848/529, loss: 0.006134796421974897 2023-01-21 12:16:17.516879: step: 852/529, loss: 0.0001253128139069304 2023-01-21 12:16:18.626138: step: 856/529, loss: 0.0006963729974813759 2023-01-21 12:16:19.741764: step: 860/529, loss: 0.0036382675170898438 2023-01-21 12:16:20.855649: step: 864/529, loss: 4.386901309771929e-06 2023-01-21 12:16:22.024228: step: 868/529, loss: 0.006336594000458717 2023-01-21 12:16:23.133666: step: 872/529, loss: 0.0009226799593307078 2023-01-21 12:16:24.272386: step: 876/529, loss: 0.0299332607537508 2023-01-21 12:16:25.409961: step: 880/529, loss: 0.04130411148071289 2023-01-21 12:16:26.539910: step: 884/529, loss: 0.0034194947220385075 2023-01-21 12:16:27.691836: step: 888/529, loss: 0.0030481338035315275 2023-01-21 12:16:28.811653: step: 892/529, loss: 0.0015429496997967362 2023-01-21 12:16:29.967449: step: 896/529, loss: 0.02643413655459881 2023-01-21 12:16:31.085687: step: 900/529, loss: 0.0004451751592569053 2023-01-21 12:16:32.231493: step: 904/529, loss: 0.07229682058095932 2023-01-21 12:16:33.364823: step: 908/529, loss: 0.007631492801010609 2023-01-21 12:16:34.485549: step: 912/529, loss: 0.005826759152114391 2023-01-21 12:16:35.644437: step: 916/529, loss: 0.00020294189744163305 2023-01-21 12:16:36.801992: step: 920/529, loss: 0.04877859726548195 2023-01-21 12:16:37.935878: step: 924/529, loss: 0.0048805237747728825 2023-01-21 12:16:39.055638: step: 928/529, loss: 0.00014190674119163305 2023-01-21 12:16:40.169739: step: 932/529, loss: 0.0009296417119912803 2023-01-21 12:16:41.307689: step: 936/529, loss: 0.011898231692612171 2023-01-21 12:16:42.485787: step: 940/529, loss: 0.0026494981721043587 2023-01-21 12:16:43.633406: step: 944/529, loss: 0.0008689879905432463 2023-01-21 12:16:44.782922: step: 948/529, loss: 0.00023279190645553172 2023-01-21 12:16:45.913551: step: 952/529, loss: 0.0005958557012490928 2023-01-21 12:16:47.049839: step: 956/529, loss: 0.0007064819219522178 2023-01-21 12:16:48.193525: step: 960/529, loss: 0.00023002624220680445 2023-01-21 12:16:49.308552: step: 964/529, loss: 5.807876732433215e-05 2023-01-21 12:16:50.473046: step: 968/529, loss: 0.5852420926094055 2023-01-21 12:16:51.583900: step: 972/529, loss: 1.3828278042637976e-06 2023-01-21 12:16:52.747651: step: 976/529, loss: 0.010821724310517311 2023-01-21 12:16:53.865529: step: 980/529, loss: 0.0022260667756199837 2023-01-21 12:16:55.002422: step: 984/529, loss: 0.0006770610925741494 2023-01-21 12:16:56.123881: step: 988/529, loss: 0.04202480614185333 2023-01-21 12:16:57.263715: step: 992/529, loss: 0.0030164720956236124 2023-01-21 12:16:58.433183: step: 996/529, loss: 0.0007476806640625 2023-01-21 12:16:59.570532: step: 1000/529, loss: 0.03749847412109375 2023-01-21 12:17:00.689708: step: 1004/529, loss: 0.0014341354835778475 2023-01-21 12:17:01.847417: step: 1008/529, loss: 0.030923843383789062 2023-01-21 12:17:02.948806: step: 1012/529, loss: 0.0007534027099609375 2023-01-21 12:17:04.081155: step: 1016/529, loss: 0.03112630732357502 2023-01-21 12:17:05.261024: step: 1020/529, loss: 0.030054474249482155 2023-01-21 12:17:06.420887: step: 1024/529, loss: 0.0100892074406147 2023-01-21 12:17:07.538340: step: 1028/529, loss: 0.00301532750017941 2023-01-21 12:17:08.689952: step: 1032/529, loss: 0.007128524594008923 2023-01-21 12:17:09.835179: step: 1036/529, loss: 0.00010027885582530871 2023-01-21 12:17:10.997051: step: 1040/529, loss: 0.06332512199878693 2023-01-21 12:17:12.112377: step: 1044/529, loss: 0.0002983570157084614 2023-01-21 12:17:13.270114: step: 1048/529, loss: 2.441406286379788e-05 2023-01-21 12:17:14.409765: step: 1052/529, loss: 0.00030736924964003265 2023-01-21 12:17:15.546236: step: 1056/529, loss: 0.00013723372831009328 2023-01-21 12:17:16.698102: step: 1060/529, loss: 0.009287262335419655 2023-01-21 12:17:17.822435: step: 1064/529, loss: 4.0721897676121444e-05 2023-01-21 12:17:18.975913: step: 1068/529, loss: 0.0005447387811727822 2023-01-21 12:17:20.111945: step: 1072/529, loss: 0.01667480543255806 2023-01-21 12:17:21.248987: step: 1076/529, loss: 0.00048708918620832264 2023-01-21 12:17:22.363003: step: 1080/529, loss: 0.006722068879753351 2023-01-21 12:17:23.561790: step: 1084/529, loss: 0.0016569137806072831 2023-01-21 12:17:24.695618: step: 1088/529, loss: 1.296997106692288e-05 2023-01-21 12:17:25.822970: step: 1092/529, loss: 0.007367610931396484 2023-01-21 12:17:26.954224: step: 1096/529, loss: 0.0014804840320721269 2023-01-21 12:17:28.082266: step: 1100/529, loss: 0.0068183899857103825 2023-01-21 12:17:29.233246: step: 1104/529, loss: 0.00018205643573310226 2023-01-21 12:17:30.356842: step: 1108/529, loss: 0.009785271249711514 2023-01-21 12:17:31.490148: step: 1112/529, loss: 0.0004044056113343686 2023-01-21 12:17:32.612927: step: 1116/529, loss: 0.008419417776167393 2023-01-21 12:17:33.743233: step: 1120/529, loss: 0.002857733052223921 2023-01-21 12:17:34.885245: step: 1124/529, loss: 0.0010179519886150956 2023-01-21 12:17:36.039097: step: 1128/529, loss: 0.018828105181455612 2023-01-21 12:17:37.204447: step: 1132/529, loss: 0.00017976760864257812 2023-01-21 12:17:38.355472: step: 1136/529, loss: 0.032177068293094635 2023-01-21 12:17:39.497720: step: 1140/529, loss: 0.019377898424863815 2023-01-21 12:17:40.644646: step: 1144/529, loss: 0.05404262617230415 2023-01-21 12:17:41.808736: step: 1148/529, loss: 0.0038116932846605778 2023-01-21 12:17:42.926530: step: 1152/529, loss: 0.030964471399784088 2023-01-21 12:17:44.064601: step: 1156/529, loss: 0.011748981662094593 2023-01-21 12:17:45.224149: step: 1160/529, loss: 0.0012697220081463456 2023-01-21 12:17:46.408195: step: 1164/529, loss: 0.005949640180915594 2023-01-21 12:17:47.524067: step: 1168/529, loss: 0.001901340438053012 2023-01-21 12:17:48.650144: step: 1172/529, loss: 0.012233162298798561 2023-01-21 12:17:49.784320: step: 1176/529, loss: 0.012092018499970436 2023-01-21 12:17:50.910718: step: 1180/529, loss: 1.635551598155871e-05 2023-01-21 12:17:51.996922: step: 1184/529, loss: 0.01755847968161106 2023-01-21 12:17:53.124377: step: 1188/529, loss: 0.00023651123046875 2023-01-21 12:17:54.264262: step: 1192/529, loss: 0.0073566436767578125 2023-01-21 12:17:55.368397: step: 1196/529, loss: 0.006438589189201593 2023-01-21 12:17:56.484299: step: 1200/529, loss: 0.0003258705255575478 2023-01-21 12:17:57.620011: step: 1204/529, loss: 0.03811035305261612 2023-01-21 12:17:58.752005: step: 1208/529, loss: 0.013130568899214268 2023-01-21 12:17:59.857861: step: 1212/529, loss: 0.021236037835478783 2023-01-21 12:18:00.988704: step: 1216/529, loss: 0.0031888962257653475 2023-01-21 12:18:02.134837: step: 1220/529, loss: 0.10588055104017258 2023-01-21 12:18:03.279939: step: 1224/529, loss: 0.006593131925910711 2023-01-21 12:18:04.395501: step: 1228/529, loss: 0.030215073376893997 2023-01-21 12:18:05.520229: step: 1232/529, loss: 0.00016098022751975805 2023-01-21 12:18:06.651929: step: 1236/529, loss: 0.051476430147886276 2023-01-21 12:18:07.741551: step: 1240/529, loss: 0.0008342743385583162 2023-01-21 12:18:08.849464: step: 1244/529, loss: 0.03817396238446236 2023-01-21 12:18:09.988311: step: 1248/529, loss: 0.003032875247299671 2023-01-21 12:18:11.118205: step: 1252/529, loss: 0.0049485210329294205 2023-01-21 12:18:12.261502: step: 1256/529, loss: 7.05719003235572e-06 2023-01-21 12:18:13.394908: step: 1260/529, loss: 0.03828296810388565 2023-01-21 12:18:14.549498: step: 1264/529, loss: 0.00037994387093931437 2023-01-21 12:18:15.659339: step: 1268/529, loss: 0.0001733779936330393 2023-01-21 12:18:16.807539: step: 1272/529, loss: 0.014648723416030407 2023-01-21 12:18:17.919124: step: 1276/529, loss: 0.0009513854747638106 2023-01-21 12:18:19.027808: step: 1280/529, loss: 5.9521196817513555e-05 2023-01-21 12:18:20.165411: step: 1284/529, loss: 0.0001517772616352886 2023-01-21 12:18:21.288426: step: 1288/529, loss: 0.02330961264669895 2023-01-21 12:18:22.418401: step: 1292/529, loss: 0.025935126468539238 2023-01-21 12:18:23.556584: step: 1296/529, loss: 0.07380924373865128 2023-01-21 12:18:24.667238: step: 1300/529, loss: 0.000289726274786517 2023-01-21 12:18:25.800406: step: 1304/529, loss: 0.010420035570859909 2023-01-21 12:18:26.936353: step: 1308/529, loss: 0.0023464204277843237 2023-01-21 12:18:28.065419: step: 1312/529, loss: 0.06412182003259659 2023-01-21 12:18:29.218153: step: 1316/529, loss: 0.0025056840386241674 2023-01-21 12:18:30.330490: step: 1320/529, loss: 0.0003246903361286968 2023-01-21 12:18:31.476136: step: 1324/529, loss: 0.0019264222355559468 2023-01-21 12:18:32.589491: step: 1328/529, loss: 1.0251998901367188e-05 2023-01-21 12:18:33.728898: step: 1332/529, loss: 0.004190540406852961 2023-01-21 12:18:34.866451: step: 1336/529, loss: 0.0263349786400795 2023-01-21 12:18:36.002944: step: 1340/529, loss: 0.00012922286987304688 2023-01-21 12:18:37.144978: step: 1344/529, loss: 1.2588501704158261e-05 2023-01-21 12:18:38.274752: step: 1348/529, loss: 0.0010848998790606856 2023-01-21 12:18:39.425583: step: 1352/529, loss: 0.06205024942755699 2023-01-21 12:18:40.565504: step: 1356/529, loss: 0.0017721176845952868 2023-01-21 12:18:41.688312: step: 1360/529, loss: 0.00556449918076396 2023-01-21 12:18:42.802419: step: 1364/529, loss: 0.016636276617646217 2023-01-21 12:18:43.966782: step: 1368/529, loss: 0.00018901826115325093 2023-01-21 12:18:45.117184: step: 1372/529, loss: 0.004513740539550781 2023-01-21 12:18:46.247512: step: 1376/529, loss: 0.00034399033756926656 2023-01-21 12:18:47.434436: step: 1380/529, loss: 0.00011663437180686742 2023-01-21 12:18:48.536598: step: 1384/529, loss: 0.0010886192321777344 2023-01-21 12:18:49.657403: step: 1388/529, loss: 0.0024131773971021175 2023-01-21 12:18:50.813481: step: 1392/529, loss: 0.008314704522490501 2023-01-21 12:18:51.954632: step: 1396/529, loss: 0.0014326096279546618 2023-01-21 12:18:53.109328: step: 1400/529, loss: 0.002627086825668812 2023-01-21 12:18:54.260155: step: 1404/529, loss: 0.0030498504638671875 2023-01-21 12:18:55.391897: step: 1408/529, loss: 0.0010994315380230546 2023-01-21 12:18:56.509287: step: 1412/529, loss: 0.026863574981689453 2023-01-21 12:18:57.645572: step: 1416/529, loss: 0.006167411804199219 2023-01-21 12:18:58.778486: step: 1420/529, loss: 0.004493999294936657 2023-01-21 12:18:59.929746: step: 1424/529, loss: 3.826618012681138e-06 2023-01-21 12:19:01.095411: step: 1428/529, loss: 0.0033059597481042147 2023-01-21 12:19:02.238597: step: 1432/529, loss: 0.02643613889813423 2023-01-21 12:19:03.383495: step: 1436/529, loss: 0.014071083627641201 2023-01-21 12:19:04.519761: step: 1440/529, loss: 0.009051084518432617 2023-01-21 12:19:05.633393: step: 1444/529, loss: 0.04303565248847008 2023-01-21 12:19:06.760556: step: 1448/529, loss: 0.08850689232349396 2023-01-21 12:19:07.889268: step: 1452/529, loss: 0.0008841037633828819 2023-01-21 12:19:08.990782: step: 1456/529, loss: 6.69479341013357e-05 2023-01-21 12:19:10.143592: step: 1460/529, loss: 0.0007533550960943103 2023-01-21 12:19:11.285723: step: 1464/529, loss: 0.002031183335930109 2023-01-21 12:19:12.457201: step: 1468/529, loss: 0.04259643703699112 2023-01-21 12:19:13.602598: step: 1472/529, loss: 0.0010677337413653731 2023-01-21 12:19:14.763336: step: 1476/529, loss: 0.004445266909897327 2023-01-21 12:19:15.897368: step: 1480/529, loss: 0.0008053779602050781 2023-01-21 12:19:17.017619: step: 1484/529, loss: 0.008470630273222923 2023-01-21 12:19:18.136653: step: 1488/529, loss: 0.005029773339629173 2023-01-21 12:19:19.263143: step: 1492/529, loss: 0.026469040662050247 2023-01-21 12:19:20.397670: step: 1496/529, loss: 0.004339504521340132 2023-01-21 12:19:21.549895: step: 1500/529, loss: 0.04278764873743057 2023-01-21 12:19:22.702697: step: 1504/529, loss: 0.09206171333789825 2023-01-21 12:19:23.831177: step: 1508/529, loss: 0.049345873296260834 2023-01-21 12:19:24.997709: step: 1512/529, loss: 0.05683489143848419 2023-01-21 12:19:26.143250: step: 1516/529, loss: 0.009336471557617188 2023-01-21 12:19:27.263211: step: 1520/529, loss: 0.01987314224243164 2023-01-21 12:19:28.377004: step: 1524/529, loss: 0.00692405691370368 2023-01-21 12:19:29.522696: step: 1528/529, loss: 0.059420399367809296 2023-01-21 12:19:30.654084: step: 1532/529, loss: 0.009808255359530449 2023-01-21 12:19:31.800729: step: 1536/529, loss: 0.0006288528093136847 2023-01-21 12:19:32.902330: step: 1540/529, loss: 0.0033903121948242188 2023-01-21 12:19:34.030275: step: 1544/529, loss: 0.004057502839714289 2023-01-21 12:19:35.209904: step: 1548/529, loss: 0.0006958961603231728 2023-01-21 12:19:36.328565: step: 1552/529, loss: 0.0012894630199298263 2023-01-21 12:19:37.445124: step: 1556/529, loss: 0.0383269302546978 2023-01-21 12:19:38.584573: step: 1560/529, loss: 0.00046415330143645406 2023-01-21 12:19:39.723248: step: 1564/529, loss: 0.015267753973603249 2023-01-21 12:19:40.858096: step: 1568/529, loss: 0.024472331628203392 2023-01-21 12:19:42.033337: step: 1572/529, loss: 0.003404522081837058 2023-01-21 12:19:43.167086: step: 1576/529, loss: 0.021689655259251595 2023-01-21 12:19:44.303347: step: 1580/529, loss: 0.0005347728729248047 2023-01-21 12:19:45.421880: step: 1584/529, loss: 0.03808346018195152 2023-01-21 12:19:46.525381: step: 1588/529, loss: 0.0006064891931600869 2023-01-21 12:19:47.633900: step: 1592/529, loss: 0.00553164491429925 2023-01-21 12:19:48.761938: step: 1596/529, loss: 0.007785034365952015 2023-01-21 12:19:49.881632: step: 1600/529, loss: 0.04555225372314453 2023-01-21 12:19:51.015959: step: 1604/529, loss: 0.0019316673278808594 2023-01-21 12:19:52.111637: step: 1608/529, loss: 0.04235520213842392 2023-01-21 12:19:53.249332: step: 1612/529, loss: 0.0020699501037597656 2023-01-21 12:19:54.382160: step: 1616/529, loss: 0.0003566741943359375 2023-01-21 12:19:55.507628: step: 1620/529, loss: 0.06817102432250977 2023-01-21 12:19:56.685991: step: 1624/529, loss: 0.0009162425994873047 2023-01-21 12:19:57.828131: step: 1628/529, loss: 0.002708053681999445 2023-01-21 12:19:58.952192: step: 1632/529, loss: 0.001903533935546875 2023-01-21 12:20:00.077668: step: 1636/529, loss: 0.0016307830810546875 2023-01-21 12:20:01.176783: step: 1640/529, loss: 0.026835203170776367 2023-01-21 12:20:02.296718: step: 1644/529, loss: 0.000567579292692244 2023-01-21 12:20:03.444749: step: 1648/529, loss: 0.0003002166631631553 2023-01-21 12:20:04.591508: step: 1652/529, loss: 0.003636646317318082 2023-01-21 12:20:05.716688: step: 1656/529, loss: 0.049230292439460754 2023-01-21 12:20:06.807347: step: 1660/529, loss: 0.05169343948364258 2023-01-21 12:20:07.955079: step: 1664/529, loss: 0.002552127931267023 2023-01-21 12:20:09.088418: step: 1668/529, loss: 0.007974052801728249 2023-01-21 12:20:10.222407: step: 1672/529, loss: 0.000599765800870955 2023-01-21 12:20:11.359248: step: 1676/529, loss: 0.6159076690673828 2023-01-21 12:20:12.510769: step: 1680/529, loss: 0.009006691165268421 2023-01-21 12:20:13.627643: step: 1684/529, loss: 0.001956081483513117 2023-01-21 12:20:14.787117: step: 1688/529, loss: 0.1901434063911438 2023-01-21 12:20:15.907310: step: 1692/529, loss: 0.001997661776840687 2023-01-21 12:20:17.033550: step: 1696/529, loss: 0.0021135329734534025 2023-01-21 12:20:18.205870: step: 1700/529, loss: 0.03945045545697212 2023-01-21 12:20:19.356590: step: 1704/529, loss: 0.0002437591610942036 2023-01-21 12:20:20.518357: step: 1708/529, loss: 0.0026149749755859375 2023-01-21 12:20:21.677173: step: 1712/529, loss: 0.04519524797797203 2023-01-21 12:20:22.832908: step: 1716/529, loss: 0.005949783604592085 2023-01-21 12:20:23.955311: step: 1720/529, loss: 0.0793725997209549 2023-01-21 12:20:25.095683: step: 1724/529, loss: 0.02308826334774494 2023-01-21 12:20:26.228166: step: 1728/529, loss: 0.0018892288208007812 2023-01-21 12:20:27.372919: step: 1732/529, loss: 0.019918251782655716 2023-01-21 12:20:28.550822: step: 1736/529, loss: 0.00215834379196167 2023-01-21 12:20:29.682349: step: 1740/529, loss: 4.025680065155029 2023-01-21 12:20:30.800145: step: 1744/529, loss: 0.014436721801757812 2023-01-21 12:20:31.930989: step: 1748/529, loss: 0.01016922015696764 2023-01-21 12:20:33.086111: step: 1752/529, loss: 4.2343137465650216e-05 2023-01-21 12:20:34.239177: step: 1756/529, loss: 0.014159775339066982 2023-01-21 12:20:35.381432: step: 1760/529, loss: 0.018416594713926315 2023-01-21 12:20:36.518609: step: 1764/529, loss: 3.929138256353326e-05 2023-01-21 12:20:37.684323: step: 1768/529, loss: 0.33783960342407227 2023-01-21 12:20:38.818131: step: 1772/529, loss: 0.0008186817285604775 2023-01-21 12:20:39.941167: step: 1776/529, loss: 0.015990637242794037 2023-01-21 12:20:41.075454: step: 1780/529, loss: 0.014895057305693626 2023-01-21 12:20:42.195673: step: 1784/529, loss: 0.00017471313185524195 2023-01-21 12:20:43.332724: step: 1788/529, loss: 0.00138597481418401 2023-01-21 12:20:44.487879: step: 1792/529, loss: 0.20976239442825317 2023-01-21 12:20:45.621494: step: 1796/529, loss: 0.00028362273587845266 2023-01-21 12:20:46.755470: step: 1800/529, loss: 3.395080420887098e-05 2023-01-21 12:20:47.892312: step: 1804/529, loss: 0.0013628959422931075 2023-01-21 12:20:49.018598: step: 1808/529, loss: 0.00021014214144088328 2023-01-21 12:20:50.120171: step: 1812/529, loss: 0.00016026497178245336 2023-01-21 12:20:51.257979: step: 1816/529, loss: 0.02072782628238201 2023-01-21 12:20:52.383170: step: 1820/529, loss: 0.001722145127132535 2023-01-21 12:20:53.516909: step: 1824/529, loss: 0.01542973518371582 2023-01-21 12:20:54.711749: step: 1828/529, loss: 0.0018113136757165194 2023-01-21 12:20:55.833291: step: 1832/529, loss: 0.0002037048398051411 2023-01-21 12:20:56.979614: step: 1836/529, loss: 0.004458618350327015 2023-01-21 12:20:58.130330: step: 1840/529, loss: 0.005287551786750555 2023-01-21 12:20:59.274085: step: 1844/529, loss: 0.0004940032958984375 2023-01-21 12:21:00.424676: step: 1848/529, loss: 0.00474472064524889 2023-01-21 12:21:01.540298: step: 1852/529, loss: 1.773834264895413e-05 2023-01-21 12:21:02.705536: step: 1856/529, loss: 0.065185546875 2023-01-21 12:21:03.839617: step: 1860/529, loss: 0.05085029453039169 2023-01-21 12:21:04.972704: step: 1864/529, loss: 0.004641437437385321 2023-01-21 12:21:06.092395: step: 1868/529, loss: 0.0006619930500164628 2023-01-21 12:21:07.216870: step: 1872/529, loss: 2.4795535864541307e-06 2023-01-21 12:21:08.383662: step: 1876/529, loss: 0.060254864394664764 2023-01-21 12:21:09.512279: step: 1880/529, loss: 0.0017715455032885075 2023-01-21 12:21:10.646629: step: 1884/529, loss: 0.0027696609031409025 2023-01-21 12:21:11.776391: step: 1888/529, loss: 0.00237789168022573 2023-01-21 12:21:12.913700: step: 1892/529, loss: 0.01651782914996147 2023-01-21 12:21:14.035770: step: 1896/529, loss: 6.4849853515625e-05 2023-01-21 12:21:15.184507: step: 1900/529, loss: 0.002273178193718195 2023-01-21 12:21:16.330020: step: 1904/529, loss: 0.003034830093383789 2023-01-21 12:21:17.450955: step: 1908/529, loss: 0.0002570629003457725 2023-01-21 12:21:18.610427: step: 1912/529, loss: 0.0035238265991210938 2023-01-21 12:21:19.751813: step: 1916/529, loss: 0.046530820429325104 2023-01-21 12:21:20.895169: step: 1920/529, loss: 0.08855066448450089 2023-01-21 12:21:22.017180: step: 1924/529, loss: 0.00024657248286530375 2023-01-21 12:21:23.167112: step: 1928/529, loss: 0.03462934494018555 2023-01-21 12:21:24.294022: step: 1932/529, loss: 0.018462801352143288 2023-01-21 12:21:25.426857: step: 1936/529, loss: 0.0017685890197753906 2023-01-21 12:21:26.581214: step: 1940/529, loss: 0.0233930591493845 2023-01-21 12:21:27.718160: step: 1944/529, loss: 0.008492564782500267 2023-01-21 12:21:28.872603: step: 1948/529, loss: 0.00037288665771484375 2023-01-21 12:21:30.002789: step: 1952/529, loss: 0.012282943353056908 2023-01-21 12:21:31.133404: step: 1956/529, loss: 0.0009721756214275956 2023-01-21 12:21:32.246169: step: 1960/529, loss: 0.00583572406321764 2023-01-21 12:21:33.367847: step: 1964/529, loss: 0.08711566776037216 2023-01-21 12:21:34.544040: step: 1968/529, loss: 0.01244430523365736 2023-01-21 12:21:35.677647: step: 1972/529, loss: 0.03322754055261612 2023-01-21 12:21:36.789503: step: 1976/529, loss: 0.03882112726569176 2023-01-21 12:21:37.930608: step: 1980/529, loss: 0.0075327870436012745 2023-01-21 12:21:39.053712: step: 1984/529, loss: 0.0005918502574786544 2023-01-21 12:21:40.169029: step: 1988/529, loss: 0.0001316070556640625 2023-01-21 12:21:41.291620: step: 1992/529, loss: 0.006556892301887274 2023-01-21 12:21:42.460734: step: 1996/529, loss: 0.003131103701889515 2023-01-21 12:21:43.639201: step: 2000/529, loss: 0.005212068557739258 2023-01-21 12:21:44.796142: step: 2004/529, loss: 0.008368587121367455 2023-01-21 12:21:45.936300: step: 2008/529, loss: 0.0014554978115484118 2023-01-21 12:21:47.077298: step: 2012/529, loss: 0.0007446288946084678 2023-01-21 12:21:48.204414: step: 2016/529, loss: 0.015977095812559128 2023-01-21 12:21:49.312690: step: 2020/529, loss: 0.04450549930334091 2023-01-21 12:21:50.447349: step: 2024/529, loss: 0.00022735596576239914 2023-01-21 12:21:51.571970: step: 2028/529, loss: 0.00783395767211914 2023-01-21 12:21:52.742408: step: 2032/529, loss: 0.003346347715705633 2023-01-21 12:21:53.867395: step: 2036/529, loss: 0.00032558440580032766 2023-01-21 12:21:54.991045: step: 2040/529, loss: 0.0001697540283203125 2023-01-21 12:21:56.139358: step: 2044/529, loss: 0.06764717400074005 2023-01-21 12:21:57.278751: step: 2048/529, loss: 0.1772165298461914 2023-01-21 12:21:58.402412: step: 2052/529, loss: 0.0019796371925622225 2023-01-21 12:21:59.521242: step: 2056/529, loss: 0.004189491271972656 2023-01-21 12:22:00.633431: step: 2060/529, loss: 0.0007213592762127519 2023-01-21 12:22:01.784547: step: 2064/529, loss: 0.5605897903442383 2023-01-21 12:22:02.902227: step: 2068/529, loss: 0.016448400914669037 2023-01-21 12:22:04.015893: step: 2072/529, loss: 0.00016870499530341476 2023-01-21 12:22:05.165809: step: 2076/529, loss: 0.011147118173539639 2023-01-21 12:22:06.360255: step: 2080/529, loss: 0.0062665934674441814 2023-01-21 12:22:07.491861: step: 2084/529, loss: 0.004565429873764515 2023-01-21 12:22:08.642954: step: 2088/529, loss: 0.48561668395996094 2023-01-21 12:22:09.770129: step: 2092/529, loss: 0.011810111813247204 2023-01-21 12:22:10.929339: step: 2096/529, loss: 0.0024503706954419613 2023-01-21 12:22:12.093240: step: 2100/529, loss: 0.0017020226223394275 2023-01-21 12:22:13.225833: step: 2104/529, loss: 0.0039160726591944695 2023-01-21 12:22:14.367121: step: 2108/529, loss: 0.0016578674549236894 2023-01-21 12:22:15.488881: step: 2112/529, loss: 0.0037282942794263363 2023-01-21 12:22:16.624790: step: 2116/529, loss: 0.08283558487892151 ================================================== Loss: 0.032 -------------------- Dev: {'event': {'p': 0.6409691629955947, 'r': 0.7749667110519307, 'f1': 0.701627486437613}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Test: {'event': {'p': 0.6427525622254758, 'r': 0.7630359212050984, 'f1': 0.6977483443708609}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} 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': 21} Korean: {'event': {'p': 0.6727272727272727, 'r': 0.5873015873015873, 'f1': 0.6271186440677966}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Russian: {'event': {'p': 0.5, 'r': 0.5277777777777778, 'f1': 0.5135135135135136}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 22 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:22:57.229319: step: 4/529, loss: 0.012122249230742455 2023-01-21 12:22:58.312037: step: 8/529, loss: 0.0017339707119390368 2023-01-21 12:22:59.438899: step: 12/529, loss: 0.018571853637695312 2023-01-21 12:23:00.575623: step: 16/529, loss: 0.0002926826709881425 2023-01-21 12:23:01.707783: step: 20/529, loss: 0.0009027480846270919 2023-01-21 12:23:02.808895: step: 24/529, loss: 0.007097435183823109 2023-01-21 12:23:03.930082: step: 28/529, loss: 0.37966862320899963 2023-01-21 12:23:05.031113: step: 32/529, loss: 0.009923267178237438 2023-01-21 12:23:06.155867: step: 36/529, loss: 0.018733788281679153 2023-01-21 12:23:07.273515: step: 40/529, loss: 0.013580131344497204 2023-01-21 12:23:08.420237: step: 44/529, loss: 0.017580604180693626 2023-01-21 12:23:09.540394: step: 48/529, loss: 0.011322021484375 2023-01-21 12:23:10.677998: step: 52/529, loss: 0.0021246911492198706 2023-01-21 12:23:11.805551: step: 56/529, loss: 0.00020723343186546117 2023-01-21 12:23:12.920927: step: 60/529, loss: 0.00550422677770257 2023-01-21 12:23:14.069173: step: 64/529, loss: 0.07198028266429901 2023-01-21 12:23:15.173542: step: 68/529, loss: 0.004307484719902277 2023-01-21 12:23:16.328063: step: 72/529, loss: 0.003279400058090687 2023-01-21 12:23:17.484375: step: 76/529, loss: 0.0037776471581310034 2023-01-21 12:23:18.601359: step: 80/529, loss: 0.0004974365583620965 2023-01-21 12:23:19.730855: step: 84/529, loss: 0.0011445998679846525 2023-01-21 12:23:20.880221: step: 88/529, loss: 0.009108353406190872 2023-01-21 12:23:22.001706: step: 92/529, loss: 0.00018730164447333664 2023-01-21 12:23:23.172786: step: 96/529, loss: 0.0016743660671636462 2023-01-21 12:23:24.303848: step: 100/529, loss: 0.0030121803283691406 2023-01-21 12:23:25.450430: step: 104/529, loss: 0.012143135070800781 2023-01-21 12:23:26.587186: step: 108/529, loss: 0.0001581192045705393 2023-01-21 12:23:27.745107: step: 112/529, loss: 0.0008585453033447266 2023-01-21 12:23:28.888158: step: 116/529, loss: 0.0999293327331543 2023-01-21 12:23:30.035983: step: 120/529, loss: 0.0001371383696096018 2023-01-21 12:23:31.162026: step: 124/529, loss: 0.0008690833928994834 2023-01-21 12:23:32.293954: step: 128/529, loss: 0.008140373043715954 2023-01-21 12:23:33.421596: step: 132/529, loss: 0.013624383136630058 2023-01-21 12:23:34.568426: step: 136/529, loss: 4.1675568354548886e-05 2023-01-21 12:23:35.680275: step: 140/529, loss: 0.00470733642578125 2023-01-21 12:23:36.802075: step: 144/529, loss: 0.004238796420395374 2023-01-21 12:23:37.956522: step: 148/529, loss: 0.029905272647738457 2023-01-21 12:23:39.066211: step: 152/529, loss: 0.00018134116544388235 2023-01-21 12:23:40.183269: step: 156/529, loss: 0.004186582285910845 2023-01-21 12:23:41.307489: step: 160/529, loss: 0.012182379141449928 2023-01-21 12:23:42.439286: step: 164/529, loss: 0.00018320084200240672 2023-01-21 12:23:43.585271: step: 168/529, loss: 0.006047630216926336 2023-01-21 12:23:44.733105: step: 172/529, loss: 0.01682748831808567 2023-01-21 12:23:45.887560: step: 176/529, loss: 0.09782848507165909 2023-01-21 12:23:47.000452: step: 180/529, loss: 0.003018188290297985 2023-01-21 12:23:48.158168: step: 184/529, loss: 0.00026841164799407125 2023-01-21 12:23:49.281450: step: 188/529, loss: 0.0029460906516760588 2023-01-21 12:23:50.440657: step: 192/529, loss: 0.013586235232651234 2023-01-21 12:23:51.597799: step: 196/529, loss: 0.00010070800635730848 2023-01-21 12:23:52.771751: step: 200/529, loss: 0.00012636184692382812 2023-01-21 12:23:53.942306: step: 204/529, loss: 0.01437072828412056 2023-01-21 12:23:55.084413: step: 208/529, loss: 0.0004572868347167969 2023-01-21 12:23:56.193298: step: 212/529, loss: 0.0013071059947833419 2023-01-21 12:23:57.311176: step: 216/529, loss: 3.3903121220646426e-05 2023-01-21 12:23:58.453742: step: 220/529, loss: 0.005053901579231024 2023-01-21 12:23:59.572900: step: 224/529, loss: 8.535384949936997e-06 2023-01-21 12:24:00.714697: step: 228/529, loss: 0.005337000358849764 2023-01-21 12:24:01.852657: step: 232/529, loss: 0.0026384354569017887 2023-01-21 12:24:03.006092: step: 236/529, loss: 6.828307959949598e-05 2023-01-21 12:24:04.155967: step: 240/529, loss: 0.003358841175213456 2023-01-21 12:24:05.278536: step: 244/529, loss: 0.0003646850527729839 2023-01-21 12:24:06.425735: step: 248/529, loss: 0.0006046295166015625 2023-01-21 12:24:07.576076: step: 252/529, loss: -1.735687328618951e-05 2023-01-21 12:24:08.702261: step: 256/529, loss: 0.01948099210858345 2023-01-21 12:24:09.856078: step: 260/529, loss: 0.022429944947361946 2023-01-21 12:24:10.985705: step: 264/529, loss: 2.913475145760458e-05 2023-01-21 12:24:12.124037: step: 268/529, loss: 8.678436643094756e-06 2023-01-21 12:24:13.226270: step: 272/529, loss: 0.0376923568546772 2023-01-21 12:24:14.367574: step: 276/529, loss: 0.0718500167131424 2023-01-21 12:24:15.488356: step: 280/529, loss: 0.050713252276182175 2023-01-21 12:24:16.654176: step: 284/529, loss: 0.0017803192604333162 2023-01-21 12:24:17.819484: step: 288/529, loss: 0.042156413197517395 2023-01-21 12:24:18.935890: step: 292/529, loss: 0.00011577606346691027 2023-01-21 12:24:20.066711: step: 296/529, loss: 0.00672836322337389 2023-01-21 12:24:21.208963: step: 300/529, loss: -1.6689300537109375e-06 2023-01-21 12:24:22.352826: step: 304/529, loss: 8.94546537892893e-05 2023-01-21 12:24:23.470090: step: 308/529, loss: 0.15695782005786896 2023-01-21 12:24:24.605166: step: 312/529, loss: 0.00039477349491789937 2023-01-21 12:24:25.744472: step: 316/529, loss: 0.6103473901748657 2023-01-21 12:24:26.915024: step: 320/529, loss: 0.0002446174621582031 2023-01-21 12:24:28.055479: step: 324/529, loss: 0.000648879969958216 2023-01-21 12:24:29.178414: step: 328/529, loss: 0.013531113043427467 2023-01-21 12:24:30.341144: step: 332/529, loss: 0.006122589111328125 2023-01-21 12:24:31.489255: step: 336/529, loss: 0.004767417907714844 2023-01-21 12:24:32.606042: step: 340/529, loss: 0.006417084019631147 2023-01-21 12:24:33.730030: step: 344/529, loss: 0.09183750301599503 2023-01-21 12:24:34.875554: step: 348/529, loss: 0.0003725052229128778 2023-01-21 12:24:36.035983: step: 352/529, loss: 0.0028888701926916838 2023-01-21 12:24:37.193948: step: 356/529, loss: 0.003117561573162675 2023-01-21 12:24:38.319819: step: 360/529, loss: 0.0002739906485658139 2023-01-21 12:24:39.453569: step: 364/529, loss: 0.03527984768152237 2023-01-21 12:24:40.569825: step: 368/529, loss: 7.724761962890625e-05 2023-01-21 12:24:41.677656: step: 372/529, loss: 0.014986039139330387 2023-01-21 12:24:42.814806: step: 376/529, loss: 0.003082370851188898 2023-01-21 12:24:43.962193: step: 380/529, loss: 4.5776364459015895e-06 2023-01-21 12:24:45.093782: step: 384/529, loss: 3.2091142202261835e-05 2023-01-21 12:24:46.223927: step: 388/529, loss: 2.422332727292087e-05 2023-01-21 12:24:47.374989: step: 392/529, loss: 0.05735492706298828 2023-01-21 12:24:48.467129: step: 396/529, loss: 0.1433638632297516 2023-01-21 12:24:49.583771: step: 400/529, loss: 0.0004175186331849545 2023-01-21 12:24:50.732652: step: 404/529, loss: 0.0009712219471111894 2023-01-21 12:24:51.855886: step: 408/529, loss: 9.298920485889539e-05 2023-01-21 12:24:52.973343: step: 412/529, loss: 0.03572253882884979 2023-01-21 12:24:54.095874: step: 416/529, loss: 1.9931794668082148e-05 2023-01-21 12:24:55.271882: step: 420/529, loss: 0.005035782232880592 2023-01-21 12:24:56.453247: step: 424/529, loss: 0.00011863708641612902 2023-01-21 12:24:57.596269: step: 428/529, loss: 0.0044644358567893505 2023-01-21 12:24:58.736528: step: 432/529, loss: 0.00063748360844329 2023-01-21 12:24:59.880249: step: 436/529, loss: 4.9304962885798886e-05 2023-01-21 12:25:01.033141: step: 440/529, loss: 0.0005977631080895662 2023-01-21 12:25:02.161018: step: 444/529, loss: 0.0001125335693359375 2023-01-21 12:25:03.289551: step: 448/529, loss: 0.0006055831909179688 2023-01-21 12:25:04.406161: step: 452/529, loss: 7.381439354503527e-05 2023-01-21 12:25:05.548533: step: 456/529, loss: 0.011651611886918545 2023-01-21 12:25:06.658544: step: 460/529, loss: 0.03778381273150444 2023-01-21 12:25:07.790250: step: 464/529, loss: 0.006230163853615522 2023-01-21 12:25:08.934052: step: 468/529, loss: 5.91278057981981e-06 2023-01-21 12:25:10.062773: step: 472/529, loss: 0.0003252982860431075 2023-01-21 12:25:11.243962: step: 476/529, loss: 9.298324584960938e-05 2023-01-21 12:25:12.373603: step: 480/529, loss: 0.0015970231033861637 2023-01-21 12:25:13.495504: step: 484/529, loss: 0.007323837373405695 2023-01-21 12:25:14.622459: step: 488/529, loss: 7.43865966796875e-05 2023-01-21 12:25:15.737587: step: 492/529, loss: 0.028985215350985527 2023-01-21 12:25:16.871049: step: 496/529, loss: 0.05895958095788956 2023-01-21 12:25:18.017854: step: 500/529, loss: 0.00012083054025424644 2023-01-21 12:25:19.155808: step: 504/529, loss: 0.000110626220703125 2023-01-21 12:25:20.265141: step: 508/529, loss: 0.002862358232960105 2023-01-21 12:25:21.382113: step: 512/529, loss: 0.00020904542179778218 2023-01-21 12:25:22.521463: step: 516/529, loss: 0.026233388110995293 2023-01-21 12:25:23.675180: step: 520/529, loss: 0.0008932113414630294 2023-01-21 12:25:24.829465: step: 524/529, loss: 0.00119609828107059 2023-01-21 12:25:25.990646: step: 528/529, loss: 0.0008561611175537109 2023-01-21 12:25:27.132355: step: 532/529, loss: 0.046364977955818176 2023-01-21 12:25:28.273964: step: 536/529, loss: 0.0003383159637451172 2023-01-21 12:25:29.430058: step: 540/529, loss: 0.006731224246323109 2023-01-21 12:25:30.536901: step: 544/529, loss: 8.735657320357859e-05 2023-01-21 12:25:31.647998: step: 548/529, loss: 0.002952385228127241 2023-01-21 12:25:32.788539: step: 552/529, loss: 0.0008687973022460938 2023-01-21 12:25:33.944023: step: 556/529, loss: 0.0007780074956826866 2023-01-21 12:25:35.080282: step: 560/529, loss: 0.00423774728551507 2023-01-21 12:25:36.217557: step: 564/529, loss: 0.000526809657458216 2023-01-21 12:25:37.332088: step: 568/529, loss: 0.00021219253540039062 2023-01-21 12:25:38.498876: step: 572/529, loss: 0.02101736143231392 2023-01-21 12:25:39.639953: step: 576/529, loss: 0.00578761100769043 2023-01-21 12:25:40.782032: step: 580/529, loss: 0.04334073141217232 2023-01-21 12:25:41.935148: step: 584/529, loss: 0.00902633648365736 2023-01-21 12:25:43.073006: step: 588/529, loss: 0.00021696090698242188 2023-01-21 12:25:44.201786: step: 592/529, loss: 0.006221580319106579 2023-01-21 12:25:45.321853: step: 596/529, loss: 0.0004119873046875 2023-01-21 12:25:46.445964: step: 600/529, loss: 1.838056206703186 2023-01-21 12:25:47.595161: step: 604/529, loss: 6.69479341013357e-05 2023-01-21 12:25:48.717213: step: 608/529, loss: 0.000965976738370955 2023-01-21 12:25:49.870646: step: 612/529, loss: 0.10055360943078995 2023-01-21 12:25:50.997341: step: 616/529, loss: 0.000614166259765625 2023-01-21 12:25:52.170217: step: 620/529, loss: 0.01166625041514635 2023-01-21 12:25:53.284340: step: 624/529, loss: 0.0032042504753917456 2023-01-21 12:25:54.421638: step: 628/529, loss: 0.0014080047840252519 2023-01-21 12:25:55.563834: step: 632/529, loss: 0.08251648396253586 2023-01-21 12:25:56.660140: step: 636/529, loss: 8.20159948489163e-06 2023-01-21 12:25:57.795176: step: 640/529, loss: 0.011662102304399014 2023-01-21 12:25:58.932521: step: 644/529, loss: 0.00038566591683775187 2023-01-21 12:26:00.079617: step: 648/529, loss: 0.010095500387251377 2023-01-21 12:26:01.244110: step: 652/529, loss: 0.0009060860029421747 2023-01-21 12:26:02.372099: step: 656/529, loss: 0.002723598387092352 2023-01-21 12:26:03.504825: step: 660/529, loss: 3.352165367687121e-05 2023-01-21 12:26:04.623689: step: 664/529, loss: 0.13838224112987518 2023-01-21 12:26:05.764959: step: 668/529, loss: 0.0010177611839026213 2023-01-21 12:26:06.902846: step: 672/529, loss: 0.006105995271354914 2023-01-21 12:26:08.045644: step: 676/529, loss: 7.114410982467234e-05 2023-01-21 12:26:09.220226: step: 680/529, loss: 0.028000831604003906 2023-01-21 12:26:10.353406: step: 684/529, loss: 0.011997699737548828 2023-01-21 12:26:11.484155: step: 688/529, loss: 0.05275878682732582 2023-01-21 12:26:12.631652: step: 692/529, loss: 0.009086799807846546 2023-01-21 12:26:13.761056: step: 696/529, loss: 0.0004677772521972656 2023-01-21 12:26:14.884661: step: 700/529, loss: 0.00010404586646473035 2023-01-21 12:26:16.022129: step: 704/529, loss: 0.03290243446826935 2023-01-21 12:26:17.166546: step: 708/529, loss: 0.0022081376519054174 2023-01-21 12:26:18.314485: step: 712/529, loss: 1.1730194273695815e-05 2023-01-21 12:26:19.460945: step: 716/529, loss: 0.02408619038760662 2023-01-21 12:26:20.590127: step: 720/529, loss: 0.0021425248123705387 2023-01-21 12:26:21.731582: step: 724/529, loss: 0.00099439627956599 2023-01-21 12:26:22.854516: step: 728/529, loss: 0.03503723070025444 2023-01-21 12:26:23.968268: step: 732/529, loss: 0.0015258790226653218 2023-01-21 12:26:25.128376: step: 736/529, loss: 0.005784893408417702 2023-01-21 12:26:26.270316: step: 740/529, loss: 0.0034919262398034334 2023-01-21 12:26:27.428246: step: 744/529, loss: 0.011863135732710361 2023-01-21 12:26:28.580067: step: 748/529, loss: 0.01698136329650879 2023-01-21 12:26:29.713237: step: 752/529, loss: 0.0012704849941655993 2023-01-21 12:26:30.840323: step: 756/529, loss: 0.00010528563871048391 2023-01-21 12:26:31.963542: step: 760/529, loss: 2.899169885495212e-05 2023-01-21 12:26:33.105791: step: 764/529, loss: 0.0042091370560228825 2023-01-21 12:26:34.255508: step: 768/529, loss: 0.0036523817107081413 2023-01-21 12:26:35.390780: step: 772/529, loss: 0.02281961590051651 2023-01-21 12:26:36.509774: step: 776/529, loss: 0.0006132125854492188 2023-01-21 12:26:37.662863: step: 780/529, loss: -8.20159948489163e-06 2023-01-21 12:26:38.798265: step: 784/529, loss: 0.0001070022553903982 2023-01-21 12:26:39.915075: step: 788/529, loss: 0.0035590173210948706 2023-01-21 12:26:41.054330: step: 792/529, loss: 0.013995361514389515 2023-01-21 12:26:42.202494: step: 796/529, loss: 0.0015935420524328947 2023-01-21 12:26:43.352255: step: 800/529, loss: 0.03227977827191353 2023-01-21 12:26:44.485497: step: 804/529, loss: 0.09013023972511292 2023-01-21 12:26:45.596112: step: 808/529, loss: 0.0013823509216308594 2023-01-21 12:26:46.725794: step: 812/529, loss: 0.027702046558260918 2023-01-21 12:26:47.872753: step: 816/529, loss: 0.03227744251489639 2023-01-21 12:26:49.021471: step: 820/529, loss: 0.001745033310726285 2023-01-21 12:26:50.160439: step: 824/529, loss: 0.007443809416145086 2023-01-21 12:26:51.289779: step: 828/529, loss: 0.006693935487419367 2023-01-21 12:26:52.426138: step: 832/529, loss: 0.006846142001450062 2023-01-21 12:26:53.555469: step: 836/529, loss: 0.0009369850158691406 2023-01-21 12:26:54.679878: step: 840/529, loss: 0.01755070686340332 2023-01-21 12:26:55.820219: step: 844/529, loss: 0.00886230543255806 2023-01-21 12:26:56.961318: step: 848/529, loss: 0.02380330115556717 2023-01-21 12:26:58.111343: step: 852/529, loss: 7.381439354503527e-05 2023-01-21 12:26:59.254415: step: 856/529, loss: 2.098083541568485e-06 2023-01-21 12:27:00.385852: step: 860/529, loss: 0.005394172854721546 2023-01-21 12:27:01.511819: step: 864/529, loss: 6.022453453624621e-05 2023-01-21 12:27:02.656791: step: 868/529, loss: 0.003189659211784601 2023-01-21 12:27:03.778656: step: 872/529, loss: 3.185272362316027e-05 2023-01-21 12:27:04.902035: step: 876/529, loss: -4.7683716530855236e-08 2023-01-21 12:27:06.052592: step: 880/529, loss: 1.23977656585339e-06 2023-01-21 12:27:07.194117: step: 884/529, loss: 4.749298022943549e-05 2023-01-21 12:27:08.338043: step: 888/529, loss: 0.0707613006234169 2023-01-21 12:27:09.492543: step: 892/529, loss: 0.014056872576475143 2023-01-21 12:27:10.607670: step: 896/529, loss: 0.0010799408191815019 2023-01-21 12:27:11.759528: step: 900/529, loss: 0.003029441926628351 2023-01-21 12:27:12.907799: step: 904/529, loss: 0.0001663207949604839 2023-01-21 12:27:14.042793: step: 908/529, loss: 2.403259350103326e-05 2023-01-21 12:27:15.200373: step: 912/529, loss: 0.01768665201961994 2023-01-21 12:27:16.336911: step: 916/529, loss: 0.0015914917457848787 2023-01-21 12:27:17.482116: step: 920/529, loss: 0.00318145751953125 2023-01-21 12:27:18.600578: step: 924/529, loss: 4.014969090349041e-05 2023-01-21 12:27:19.750226: step: 928/529, loss: 0.010680390521883965 2023-01-21 12:27:20.906232: step: 932/529, loss: 6.580352783203125e-05 2023-01-21 12:27:22.050983: step: 936/529, loss: 0.0017225266201421618 2023-01-21 12:27:23.172267: step: 940/529, loss: 0.00531086977571249 2023-01-21 12:27:24.332197: step: 944/529, loss: 0.00135889055673033 2023-01-21 12:27:25.424619: step: 948/529, loss: 0.0004180431133136153 2023-01-21 12:27:26.538316: step: 952/529, loss: 0.006670856382697821 2023-01-21 12:27:27.657759: step: 956/529, loss: 0.012012672610580921 2023-01-21 12:27:28.785614: step: 960/529, loss: 0.00032844545785337687 2023-01-21 12:27:29.891774: step: 964/529, loss: 0.0008128166082315147 2023-01-21 12:27:31.018628: step: 968/529, loss: 0.0012675761245191097 2023-01-21 12:27:32.169669: step: 972/529, loss: 0.0002422332763671875 2023-01-21 12:27:33.284846: step: 976/529, loss: 0.0014566422905772924 2023-01-21 12:27:34.423115: step: 980/529, loss: 0.0017336368327960372 2023-01-21 12:27:35.563588: step: 984/529, loss: 0.00020017623319290578 2023-01-21 12:27:36.695506: step: 988/529, loss: 3.886222839355469e-05 2023-01-21 12:27:37.813437: step: 992/529, loss: 0.0005237102741375566 2023-01-21 12:27:38.951961: step: 996/529, loss: 0.009807873517274857 2023-01-21 12:27:40.069961: step: 1000/529, loss: 3.24249267578125e-05 2023-01-21 12:27:41.198257: step: 1004/529, loss: 0.00832066498696804 2023-01-21 12:27:42.376027: step: 1008/529, loss: 0.0005116462707519531 2023-01-21 12:27:43.495800: step: 1012/529, loss: 2.0408631826285273e-05 2023-01-21 12:27:44.648082: step: 1016/529, loss: 0.033586978912353516 2023-01-21 12:27:45.773275: step: 1020/529, loss: 0.0008363723754882812 2023-01-21 12:27:46.884076: step: 1024/529, loss: 0.0017265318892896175 2023-01-21 12:27:47.995317: step: 1028/529, loss: 0.0013116359477862716 2023-01-21 12:27:49.087155: step: 1032/529, loss: 0.0001892089785542339 2023-01-21 12:27:50.249604: step: 1036/529, loss: 0.0021267891861498356 2023-01-21 12:27:51.359677: step: 1040/529, loss: 0.07324304431676865 2023-01-21 12:27:52.513732: step: 1044/529, loss: 0.08907298743724823 2023-01-21 12:27:53.630048: step: 1048/529, loss: 0.00023002624220680445 2023-01-21 12:27:54.777842: step: 1052/529, loss: 0.07630138099193573 2023-01-21 12:27:55.937830: step: 1056/529, loss: 0.010723686777055264 2023-01-21 12:27:57.054945: step: 1060/529, loss: 0.017180634662508965 2023-01-21 12:27:58.179825: step: 1064/529, loss: 0.06581058353185654 2023-01-21 12:27:59.327020: step: 1068/529, loss: 0.00010347366333007812 2023-01-21 12:28:00.460065: step: 1072/529, loss: 0.021439744159579277 2023-01-21 12:28:01.605050: step: 1076/529, loss: 0.01027059555053711 2023-01-21 12:28:02.738022: step: 1080/529, loss: 0.001177978585474193 2023-01-21 12:28:03.869998: step: 1084/529, loss: 0.0034956932067871094 2023-01-21 12:28:05.009923: step: 1088/529, loss: 0.00027532578678801656 2023-01-21 12:28:06.156974: step: 1092/529, loss: 0.003433704376220703 2023-01-21 12:28:07.279413: step: 1096/529, loss: 0.09546375274658203 2023-01-21 12:28:08.431369: step: 1100/529, loss: 8.649825758766383e-05 2023-01-21 12:28:09.564637: step: 1104/529, loss: 0.0017650604713708162 2023-01-21 12:28:10.698590: step: 1108/529, loss: 0.006481361575424671 2023-01-21 12:28:11.829707: step: 1112/529, loss: 0.05636923015117645 2023-01-21 12:28:13.012691: step: 1116/529, loss: 0.002726268721744418 2023-01-21 12:28:14.139168: step: 1120/529, loss: 0.0010936737526208162 2023-01-21 12:28:15.240504: step: 1124/529, loss: 0.001168537070043385 2023-01-21 12:28:16.352130: step: 1128/529, loss: 7.739067223155871e-05 2023-01-21 12:28:17.495073: step: 1132/529, loss: 0.0013621330726891756 2023-01-21 12:28:18.625716: step: 1136/529, loss: 0.2411360740661621 2023-01-21 12:28:19.780293: step: 1140/529, loss: 1.888275073724799e-05 2023-01-21 12:28:20.944870: step: 1144/529, loss: 0.05358944088220596 2023-01-21 12:28:22.090731: step: 1148/529, loss: 0.014494705013930798 2023-01-21 12:28:23.227400: step: 1152/529, loss: 0.0008991241920739412 2023-01-21 12:28:24.361949: step: 1156/529, loss: 7.114410254871473e-05 2023-01-21 12:28:25.502359: step: 1160/529, loss: 0.00054168701171875 2023-01-21 12:28:26.652299: step: 1164/529, loss: 0.00783319491893053 2023-01-21 12:28:27.775921: step: 1168/529, loss: 0.02441844902932644 2023-01-21 12:28:28.930080: step: 1172/529, loss: 0.002945852465927601 2023-01-21 12:28:30.054750: step: 1176/529, loss: 0.019594382494688034 2023-01-21 12:28:31.192542: step: 1180/529, loss: 0.02733621746301651 2023-01-21 12:28:32.318820: step: 1184/529, loss: 4.892349534202367e-05 2023-01-21 12:28:33.462897: step: 1188/529, loss: 0.03534355014562607 2023-01-21 12:28:34.573437: step: 1192/529, loss: 0.0004415988805703819 2023-01-21 12:28:35.696202: step: 1196/529, loss: 0.05653982236981392 2023-01-21 12:28:36.842145: step: 1200/529, loss: 0.0002679348108358681 2023-01-21 12:28:37.971932: step: 1204/529, loss: 0.009637976065278053 2023-01-21 12:28:39.107606: step: 1208/529, loss: 0.009605693630874157 2023-01-21 12:28:40.232649: step: 1212/529, loss: 0.00013618469529319555 2023-01-21 12:28:41.392954: step: 1216/529, loss: 5.292892819852568e-05 2023-01-21 12:28:42.534606: step: 1220/529, loss: 0.03351764753460884 2023-01-21 12:28:43.670771: step: 1224/529, loss: 0.0004139900556765497 2023-01-21 12:28:44.806401: step: 1228/529, loss: 1.0013580322265625e-05 2023-01-21 12:28:45.951576: step: 1232/529, loss: 0.000308990478515625 2023-01-21 12:28:47.111307: step: 1236/529, loss: 0.005173587705940008 2023-01-21 12:28:48.256153: step: 1240/529, loss: 0.006445026490837336 2023-01-21 12:28:49.379332: step: 1244/529, loss: 0.0062697408720850945 2023-01-21 12:28:50.541358: step: 1248/529, loss: 0.0017696380382403731 2023-01-21 12:28:51.700517: step: 1252/529, loss: 0.0006448745843954384 2023-01-21 12:28:52.811253: step: 1256/529, loss: 0.04432172700762749 2023-01-21 12:28:53.937660: step: 1260/529, loss: 0.0018756865756586194 2023-01-21 12:28:55.047598: step: 1264/529, loss: 0.0027548314537853003 2023-01-21 12:28:56.233605: step: 1268/529, loss: 0.01964740827679634 2023-01-21 12:28:57.403580: step: 1272/529, loss: 0.00234298687428236 2023-01-21 12:28:58.521477: step: 1276/529, loss: 0.0043801311403512955 2023-01-21 12:28:59.687645: step: 1280/529, loss: 0.00016851426335051656 2023-01-21 12:29:00.850221: step: 1284/529, loss: 0.005230331793427467 2023-01-21 12:29:01.976459: step: 1288/529, loss: 0.0008947372552938759 2023-01-21 12:29:03.106707: step: 1292/529, loss: 0.0004623413260560483 2023-01-21 12:29:04.269439: step: 1296/529, loss: 0.022509528324007988 2023-01-21 12:29:05.387309: step: 1300/529, loss: 0.00602989224717021 2023-01-21 12:29:06.534606: step: 1304/529, loss: 0.0015788078308105469 2023-01-21 12:29:07.708690: step: 1308/529, loss: 0.02596607245504856 2023-01-21 12:29:08.845800: step: 1312/529, loss: 0.0008823395473882556 2023-01-21 12:29:09.999996: step: 1316/529, loss: 0.008840465918183327 2023-01-21 12:29:11.139591: step: 1320/529, loss: 6.361008126987144e-05 2023-01-21 12:29:12.313862: step: 1324/529, loss: 0.0007719993591308594 2023-01-21 12:29:13.444918: step: 1328/529, loss: 0.03787994384765625 2023-01-21 12:29:14.605651: step: 1332/529, loss: 0.00713882502168417 2023-01-21 12:29:15.754109: step: 1336/529, loss: 0.008744621649384499 2023-01-21 12:29:16.874015: step: 1340/529, loss: 0.0284592155367136 2023-01-21 12:29:17.977116: step: 1344/529, loss: 0.0049797059036791325 2023-01-21 12:29:19.109803: step: 1348/529, loss: 0.000581932021304965 2023-01-21 12:29:20.243351: step: 1352/529, loss: 0.038088131695985794 2023-01-21 12:29:21.412003: step: 1356/529, loss: 0.0025080679915845394 2023-01-21 12:29:22.590377: step: 1360/529, loss: 0.008681297302246094 2023-01-21 12:29:23.730526: step: 1364/529, loss: 0.00630421657115221 2023-01-21 12:29:24.892636: step: 1368/529, loss: 0.04978370666503906 2023-01-21 12:29:26.003976: step: 1372/529, loss: 0.0004442215140443295 2023-01-21 12:29:27.145181: step: 1376/529, loss: 0.0031043528579175472 2023-01-21 12:29:28.311891: step: 1380/529, loss: 0.09928111732006073 2023-01-21 12:29:29.459408: step: 1384/529, loss: 0.006635615136474371 2023-01-21 12:29:30.629717: step: 1388/529, loss: 0.0027212142013013363 2023-01-21 12:29:31.749384: step: 1392/529, loss: 4.224777512717992e-05 2023-01-21 12:29:32.893625: step: 1396/529, loss: 0.016762161627411842 2023-01-21 12:29:34.025337: step: 1400/529, loss: 0.06715116649866104 2023-01-21 12:29:35.137659: step: 1404/529, loss: 0.002013635588809848 2023-01-21 12:29:36.282601: step: 1408/529, loss: 0.035502053797245026 2023-01-21 12:29:37.410431: step: 1412/529, loss: 0.002112960908561945 2023-01-21 12:29:38.549487: step: 1416/529, loss: 0.0013929844135418534 2023-01-21 12:29:39.654013: step: 1420/529, loss: 0.00018458366685081273 2023-01-21 12:29:40.782396: step: 1424/529, loss: 0.004747653380036354 2023-01-21 12:29:41.928667: step: 1428/529, loss: 0.006683158688247204 2023-01-21 12:29:43.092825: step: 1432/529, loss: 0.0011385440593585372 2023-01-21 12:29:44.227702: step: 1436/529, loss: 0.017820073291659355 2023-01-21 12:29:45.345820: step: 1440/529, loss: 0.01777510531246662 2023-01-21 12:29:46.512998: step: 1444/529, loss: 0.0003845214960165322 2023-01-21 12:29:47.648983: step: 1448/529, loss: 0.010808277875185013 2023-01-21 12:29:48.811416: step: 1452/529, loss: 1.821518162614666e-05 2023-01-21 12:29:49.924752: step: 1456/529, loss: 0.008607101626694202 2023-01-21 12:29:51.073639: step: 1460/529, loss: 0.00011177062697242945 2023-01-21 12:29:52.198039: step: 1464/529, loss: 0.012480735778808594 2023-01-21 12:29:53.338469: step: 1468/529, loss: 0.0021741867531090975 2023-01-21 12:29:54.488424: step: 1472/529, loss: 2.918243444582913e-05 2023-01-21 12:29:55.615668: step: 1476/529, loss: 0.01189580000936985 2023-01-21 12:29:56.730757: step: 1480/529, loss: 0.0013535498874261975 2023-01-21 12:29:57.836379: step: 1484/529, loss: 0.0054073333740234375 2023-01-21 12:29:58.975280: step: 1488/529, loss: 0.011902237311005592 2023-01-21 12:30:00.099494: step: 1492/529, loss: 0.0004708290216512978 2023-01-21 12:30:01.223242: step: 1496/529, loss: 7.171630568336695e-05 2023-01-21 12:30:02.364656: step: 1500/529, loss: 0.021419046446681023 2023-01-21 12:30:03.507261: step: 1504/529, loss: 0.01244669035077095 2023-01-21 12:30:04.629722: step: 1508/529, loss: 0.023098565638065338 2023-01-21 12:30:05.754125: step: 1512/529, loss: 0.2991792559623718 2023-01-21 12:30:06.874874: step: 1516/529, loss: 6.27517729299143e-05 2023-01-21 12:30:07.997231: step: 1520/529, loss: 0.005256080999970436 2023-01-21 12:30:09.123746: step: 1524/529, loss: 0.026103878393769264 2023-01-21 12:30:10.282024: step: 1528/529, loss: 0.009510422125458717 2023-01-21 12:30:11.402680: step: 1532/529, loss: 0.0006998926401138306 2023-01-21 12:30:12.570344: step: 1536/529, loss: 2.489090184099041e-05 2023-01-21 12:30:13.744678: step: 1540/529, loss: 0.00039548875065520406 2023-01-21 12:30:14.892089: step: 1544/529, loss: 0.2413882166147232 2023-01-21 12:30:16.055569: step: 1548/529, loss: 0.0030515671242028475 2023-01-21 12:30:17.191533: step: 1552/529, loss: 0.0025078775361180305 2023-01-21 12:30:18.313618: step: 1556/529, loss: 7.152557827794226e-06 2023-01-21 12:30:19.448847: step: 1560/529, loss: 0.026714039966464043 2023-01-21 12:30:20.566619: step: 1564/529, loss: 6.942749314475805e-05 2023-01-21 12:30:21.671639: step: 1568/529, loss: 0.0024944304022938013 2023-01-21 12:30:22.829577: step: 1572/529, loss: 0.0001884460507426411 2023-01-21 12:30:23.949776: step: 1576/529, loss: 0.004269408993422985 2023-01-21 12:30:25.111623: step: 1580/529, loss: 0.048248291015625 2023-01-21 12:30:26.265358: step: 1584/529, loss: 0.0060249329544603825 2023-01-21 12:30:27.412864: step: 1588/529, loss: 0.0058532715775072575 2023-01-21 12:30:28.577340: step: 1592/529, loss: 0.0030053136870265007 2023-01-21 12:30:29.717265: step: 1596/529, loss: 0.001884555909782648 2023-01-21 12:30:30.845429: step: 1600/529, loss: 0.015384435653686523 2023-01-21 12:30:32.019051: step: 1604/529, loss: 0.006488037295639515 2023-01-21 12:30:33.172470: step: 1608/529, loss: 7.2479248046875e-05 2023-01-21 12:30:34.308290: step: 1612/529, loss: 2.3937225705594756e-05 2023-01-21 12:30:35.449755: step: 1616/529, loss: 0.0009454727405682206 2023-01-21 12:30:36.597641: step: 1620/529, loss: 4.596710277837701e-05 2023-01-21 12:30:37.738144: step: 1624/529, loss: 0.00010433197167003527 2023-01-21 12:30:38.889023: step: 1628/529, loss: 0.01026377733796835 2023-01-21 12:30:40.041984: step: 1632/529, loss: 0.01127481460571289 2023-01-21 12:30:41.186593: step: 1636/529, loss: 0.013681793585419655 2023-01-21 12:30:42.341575: step: 1640/529, loss: 0.005292415618896484 2023-01-21 12:30:43.458082: step: 1644/529, loss: 0.04929952695965767 2023-01-21 12:30:44.602659: step: 1648/529, loss: -8.678436643094756e-06 2023-01-21 12:30:45.739204: step: 1652/529, loss: 0.00047740936861373484 2023-01-21 12:30:46.850739: step: 1656/529, loss: 0.00239906320348382 2023-01-21 12:30:48.005964: step: 1660/529, loss: 0.046666525304317474 2023-01-21 12:30:49.140791: step: 1664/529, loss: 0.013225173577666283 2023-01-21 12:30:50.288501: step: 1668/529, loss: 0.004532432183623314 2023-01-21 12:30:51.388661: step: 1672/529, loss: 5.834102921653539e-05 2023-01-21 12:30:52.504986: step: 1676/529, loss: 5.264282299322076e-05 2023-01-21 12:30:53.651658: step: 1680/529, loss: 0.02164936065673828 2023-01-21 12:30:54.789100: step: 1684/529, loss: 0.0037299632094800472 2023-01-21 12:30:55.928624: step: 1688/529, loss: 0.00605010986328125 2023-01-21 12:30:57.043193: step: 1692/529, loss: 0.007861089892685413 2023-01-21 12:30:58.159745: step: 1696/529, loss: 0.0046967025846242905 2023-01-21 12:30:59.287330: step: 1700/529, loss: 4.711151632363908e-05 2023-01-21 12:31:00.406171: step: 1704/529, loss: 0.1127108559012413 2023-01-21 12:31:01.526891: step: 1708/529, loss: 0.08870165050029755 2023-01-21 12:31:02.675653: step: 1712/529, loss: 0.0024352073669433594 2023-01-21 12:31:03.810861: step: 1716/529, loss: 0.006780385971069336 2023-01-21 12:31:04.928185: step: 1720/529, loss: 0.00016927719116210938 2023-01-21 12:31:06.058461: step: 1724/529, loss: 0.0030871392227709293 2023-01-21 12:31:07.226855: step: 1728/529, loss: 8.611679368186742e-05 2023-01-21 12:31:08.360502: step: 1732/529, loss: 0.028707217425107956 2023-01-21 12:31:09.523907: step: 1736/529, loss: 0.01104354951530695 2023-01-21 12:31:10.659018: step: 1740/529, loss: 0.13848638534545898 2023-01-21 12:31:11.773554: step: 1744/529, loss: 0.014813710004091263 2023-01-21 12:31:12.927478: step: 1748/529, loss: 2.1648407710017636e-05 2023-01-21 12:31:14.064551: step: 1752/529, loss: 0.020134640857577324 2023-01-21 12:31:15.196456: step: 1756/529, loss: 0.0006126403459347785 2023-01-21 12:31:16.328241: step: 1760/529, loss: 0.008276558481156826 2023-01-21 12:31:17.440814: step: 1764/529, loss: 0.0077707297168672085 2023-01-21 12:31:18.568682: step: 1768/529, loss: 0.03609352186322212 2023-01-21 12:31:19.698878: step: 1772/529, loss: 0.0002729416301008314 2023-01-21 12:31:20.838644: step: 1776/529, loss: 0.00041415690793655813 2023-01-21 12:31:21.973547: step: 1780/529, loss: 0.01828022114932537 2023-01-21 12:31:23.103400: step: 1784/529, loss: 0.0008789062849245965 2023-01-21 12:31:24.258756: step: 1788/529, loss: 0.0016079902416095138 2023-01-21 12:31:25.388319: step: 1792/529, loss: 0.033232927322387695 2023-01-21 12:31:26.493381: step: 1796/529, loss: 9.679794311523438e-05 2023-01-21 12:31:27.623395: step: 1800/529, loss: 2.593994213384576e-05 2023-01-21 12:31:28.748048: step: 1804/529, loss: 9.31739850784652e-05 2023-01-21 12:31:29.872845: step: 1808/529, loss: 0.007866382598876953 2023-01-21 12:31:30.982817: step: 1812/529, loss: 0.006460762117058039 2023-01-21 12:31:32.115555: step: 1816/529, loss: 0.00035295484121888876 2023-01-21 12:31:33.262413: step: 1820/529, loss: 0.0003211021248716861 2023-01-21 12:31:34.414264: step: 1824/529, loss: 0.022278403863310814 2023-01-21 12:31:35.554493: step: 1828/529, loss: 0.02132449299097061 2023-01-21 12:31:36.683472: step: 1832/529, loss: 0.0002819538349285722 2023-01-21 12:31:37.823091: step: 1836/529, loss: 7.390975952148438e-05 2023-01-21 12:31:38.973570: step: 1840/529, loss: 0.006366729736328125 2023-01-21 12:31:40.125245: step: 1844/529, loss: 0.0022705078590661287 2023-01-21 12:31:41.251774: step: 1848/529, loss: 0.01170973852276802 2023-01-21 12:31:42.382879: step: 1852/529, loss: 0.000561618828214705 2023-01-21 12:31:43.550441: step: 1856/529, loss: 0.0003860473516397178 2023-01-21 12:31:44.690449: step: 1860/529, loss: 0.009035635739564896 2023-01-21 12:31:45.801090: step: 1864/529, loss: 0.0010900550987571478 2023-01-21 12:31:46.921419: step: 1868/529, loss: 1.711845470708795e-05 2023-01-21 12:31:48.061465: step: 1872/529, loss: 0.05508995056152344 2023-01-21 12:31:49.200428: step: 1876/529, loss: 2.3746491933707148e-05 2023-01-21 12:31:50.337607: step: 1880/529, loss: 0.0018678664928302169 2023-01-21 12:31:51.462653: step: 1884/529, loss: 0.0012666701804846525 2023-01-21 12:31:52.605722: step: 1888/529, loss: 0.00551528949290514 2023-01-21 12:31:53.722653: step: 1892/529, loss: 0.0016518592601642013 2023-01-21 12:31:54.845058: step: 1896/529, loss: 0.00233974470756948 2023-01-21 12:31:55.983882: step: 1900/529, loss: 0.00015149115642998368 2023-01-21 12:31:57.146551: step: 1904/529, loss: 0.014877939596772194 2023-01-21 12:31:58.280501: step: 1908/529, loss: 0.00535507220774889 2023-01-21 12:31:59.422499: step: 1912/529, loss: 4.7492983867414296e-05 2023-01-21 12:32:00.556757: step: 1916/529, loss: 0.0003095626598224044 2023-01-21 12:32:01.740626: step: 1920/529, loss: 0.06734533607959747 2023-01-21 12:32:02.894454: step: 1924/529, loss: 0.00812997855246067 2023-01-21 12:32:04.028672: step: 1928/529, loss: 0.0054565430618822575 2023-01-21 12:32:05.168793: step: 1932/529, loss: 0.010118294507265091 2023-01-21 12:32:06.294668: step: 1936/529, loss: 0.01666870154440403 2023-01-21 12:32:07.417558: step: 1940/529, loss: 0.0016904831863939762 2023-01-21 12:32:08.559305: step: 1944/529, loss: 0.0338054895401001 2023-01-21 12:32:09.692883: step: 1948/529, loss: 0.007516384124755859 2023-01-21 12:32:10.807051: step: 1952/529, loss: 0.028334714472293854 2023-01-21 12:32:11.954768: step: 1956/529, loss: 0.023342037573456764 2023-01-21 12:32:13.115782: step: 1960/529, loss: 0.008712577633559704 2023-01-21 12:32:14.227477: step: 1964/529, loss: 0.0016159057850018144 2023-01-21 12:32:15.357826: step: 1968/529, loss: 0.009116077795624733 2023-01-21 12:32:16.493385: step: 1972/529, loss: 0.06287412345409393 2023-01-21 12:32:17.634005: step: 1976/529, loss: 0.021939467638731003 2023-01-21 12:32:18.793464: step: 1980/529, loss: 0.016839217394590378 2023-01-21 12:32:19.934077: step: 1984/529, loss: 0.16899852454662323 2023-01-21 12:32:21.054289: step: 1988/529, loss: 0.00038166047306731343 2023-01-21 12:32:22.196146: step: 1992/529, loss: 6.09397902735509e-05 2023-01-21 12:32:23.343709: step: 1996/529, loss: 0.003913116175681353 2023-01-21 12:32:24.533673: step: 2000/529, loss: 0.0010168076260015368 2023-01-21 12:32:25.673170: step: 2004/529, loss: 0.001932048937305808 2023-01-21 12:32:26.816566: step: 2008/529, loss: 0.01073470152914524 2023-01-21 12:32:27.946965: step: 2012/529, loss: 3.62396240234375e-05 2023-01-21 12:32:29.078183: step: 2016/529, loss: 0.0009984970092773438 2023-01-21 12:32:30.219183: step: 2020/529, loss: 0.014743566513061523 2023-01-21 12:32:31.374937: step: 2024/529, loss: 0.01982126198709011 2023-01-21 12:32:32.501939: step: 2028/529, loss: 0.0001123428373830393 2023-01-21 12:32:33.657580: step: 2032/529, loss: 5.6076052715070546e-05 2023-01-21 12:32:34.776955: step: 2036/529, loss: 1.3256072634248994e-05 2023-01-21 12:32:35.891592: step: 2040/529, loss: 0.0021954537369310856 2023-01-21 12:32:37.064682: step: 2044/529, loss: 0.001407623291015625 2023-01-21 12:32:38.180171: step: 2048/529, loss: 0.006967353634536266 2023-01-21 12:32:39.314732: step: 2052/529, loss: 0.020756341516971588 2023-01-21 12:32:40.446652: step: 2056/529, loss: 0.01763630099594593 2023-01-21 12:32:41.567043: step: 2060/529, loss: 0.012850750237703323 2023-01-21 12:32:42.715035: step: 2064/529, loss: 0.0027101514860987663 2023-01-21 12:32:43.837909: step: 2068/529, loss: 0.012981010600924492 2023-01-21 12:32:45.007298: step: 2072/529, loss: 0.004489326383918524 2023-01-21 12:32:46.138297: step: 2076/529, loss: 0.011792373843491077 2023-01-21 12:32:47.249752: step: 2080/529, loss: 0.002345276065170765 2023-01-21 12:32:48.384448: step: 2084/529, loss: 0.003320694202557206 2023-01-21 12:32:49.554333: step: 2088/529, loss: 0.04024448245763779 2023-01-21 12:32:50.726064: step: 2092/529, loss: 0.016018344089388847 2023-01-21 12:32:51.865809: step: 2096/529, loss: 3.0803679692326114e-05 2023-01-21 12:32:52.999507: step: 2100/529, loss: 7.095336331985891e-05 2023-01-21 12:32:54.140289: step: 2104/529, loss: 0.0070673939771950245 2023-01-21 12:32:55.277858: step: 2108/529, loss: 0.08051382005214691 2023-01-21 12:32:56.426602: step: 2112/529, loss: 0.0003147602255921811 2023-01-21 12:32:57.557286: step: 2116/529, loss: 0.005423259921371937 ================================================== Loss: 0.019 -------------------- Dev: {'event': {'p': 0.6186895810955961, 'r': 0.7669773635153129, 'f1': 0.6848989298454221}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Test: {'event': {'p': 0.6382775119617224, 'r': 0.7728852838933952, 'f1': 0.69916142557652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Chinese: {'event': {'p': 0.6153846153846154, 'r': 0.8888888888888888, 'f1': 0.7272727272727274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Korean: {'event': {'p': 0.6037735849056604, 'r': 0.5079365079365079, 'f1': 0.5517241379310345}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Russian: {'event': {'p': 0.37777777777777777, 'r': 0.4722222222222222, 'f1': 0.4197530864197531}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 23 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:33:38.650988: step: 4/529, loss: 0.012882137671113014 2023-01-21 12:33:39.780872: step: 8/529, loss: 0.048334553837776184 2023-01-21 12:33:40.908055: step: 12/529, loss: 0.006993246264755726 2023-01-21 12:33:42.050612: step: 16/529, loss: 0.006013012025505304 2023-01-21 12:33:43.160795: step: 20/529, loss: 6.737709190929309e-05 2023-01-21 12:33:44.289555: step: 24/529, loss: 0.0003803253348451108 2023-01-21 12:33:45.400749: step: 28/529, loss: 3.185272362316027e-05 2023-01-21 12:33:46.557955: step: 32/529, loss: 0.015850448980927467 2023-01-21 12:33:47.703179: step: 36/529, loss: 0.0028118134941905737 2023-01-21 12:33:48.864800: step: 40/529, loss: 0.008745861239731312 2023-01-21 12:33:50.027065: step: 44/529, loss: 0.003170585725456476 2023-01-21 12:33:51.255796: step: 48/529, loss: 0.017382431775331497 2023-01-21 12:33:52.396670: step: 52/529, loss: 0.00034341809805482626 2023-01-21 12:33:53.559007: step: 56/529, loss: 0.009341764263808727 2023-01-21 12:33:54.674046: step: 60/529, loss: 0.0009114264976233244 2023-01-21 12:33:55.786473: step: 64/529, loss: 0.026303337886929512 2023-01-21 12:33:56.913484: step: 68/529, loss: 0.000392723100958392 2023-01-21 12:33:58.095582: step: 72/529, loss: 0.021758748218417168 2023-01-21 12:33:59.227511: step: 76/529, loss: 0.0533016175031662 2023-01-21 12:34:00.362887: step: 80/529, loss: 0.0008165359613485634 2023-01-21 12:34:01.521645: step: 84/529, loss: 0.025231456384062767 2023-01-21 12:34:02.640464: step: 88/529, loss: 0.003562021302059293 2023-01-21 12:34:03.775720: step: 92/529, loss: 0.008714770898222923 2023-01-21 12:34:04.928837: step: 96/529, loss: 7.791518874000758e-05 2023-01-21 12:34:06.038929: step: 100/529, loss: 0.0015239715576171875 2023-01-21 12:34:07.184348: step: 104/529, loss: 0.007922744378447533 2023-01-21 12:34:08.317054: step: 108/529, loss: 0.024698639288544655 2023-01-21 12:34:09.455950: step: 112/529, loss: 0.011738014407455921 2023-01-21 12:34:10.584622: step: 116/529, loss: 0.0028238536324352026 2023-01-21 12:34:11.718672: step: 120/529, loss: 0.008063125424087048 2023-01-21 12:34:12.892917: step: 124/529, loss: 0.0003464698966126889 2023-01-21 12:34:14.068844: step: 128/529, loss: 0.0007472991710528731 2023-01-21 12:34:15.199507: step: 132/529, loss: 0.030215872451663017 2023-01-21 12:34:16.332632: step: 136/529, loss: 0.028722573071718216 2023-01-21 12:34:17.449635: step: 140/529, loss: 0.0021591188851743937 2023-01-21 12:34:18.604801: step: 144/529, loss: 0.05266590043902397 2023-01-21 12:34:19.763315: step: 148/529, loss: 0.0004047393740620464 2023-01-21 12:34:20.912145: step: 152/529, loss: 0.06136741489171982 2023-01-21 12:34:22.049038: step: 156/529, loss: 0.009961318224668503 2023-01-21 12:34:23.172473: step: 160/529, loss: 2.079009937006049e-05 2023-01-21 12:34:24.309395: step: 164/529, loss: 0.0032651901710778475 2023-01-21 12:34:25.463791: step: 168/529, loss: 0.014388656243681908 2023-01-21 12:34:26.585223: step: 172/529, loss: 0.0002536773681640625 2023-01-21 12:34:27.711899: step: 176/529, loss: 0.002376270480453968 2023-01-21 12:34:28.851495: step: 180/529, loss: 0.00013895034499000758 2023-01-21 12:34:29.986259: step: 184/529, loss: 0.054497767239809036 2023-01-21 12:34:31.106578: step: 188/529, loss: 0.0149962417781353 2023-01-21 12:34:32.216195: step: 192/529, loss: 0.0007214546203613281 2023-01-21 12:34:33.362284: step: 196/529, loss: 0.0009597778553143144 2023-01-21 12:34:34.496363: step: 200/529, loss: 0.013955306261777878 2023-01-21 12:34:35.623314: step: 204/529, loss: 0.010005999356508255 2023-01-21 12:34:36.771807: step: 208/529, loss: 0.002341651823371649 2023-01-21 12:34:37.900204: step: 212/529, loss: 0.00018601419287733734 2023-01-21 12:34:39.030977: step: 216/529, loss: 0.011185312643647194 2023-01-21 12:34:40.138520: step: 220/529, loss: 2.956390380859375e-05 2023-01-21 12:34:41.271250: step: 224/529, loss: 0.00118427281267941 2023-01-21 12:34:42.429146: step: 228/529, loss: 0.01641826704144478 2023-01-21 12:34:43.553889: step: 232/529, loss: 0.08445654064416885 2023-01-21 12:34:44.700418: step: 236/529, loss: 0.0006719589582644403 2023-01-21 12:34:45.828343: step: 240/529, loss: 7.247925168485381e-06 2023-01-21 12:34:46.980859: step: 244/529, loss: 0.0007128715515136719 2023-01-21 12:34:48.110461: step: 248/529, loss: 6.28471389063634e-05 2023-01-21 12:34:49.233567: step: 252/529, loss: 0.004500961396843195 2023-01-21 12:34:50.345696: step: 256/529, loss: 0.00037956240703351796 2023-01-21 12:34:51.507380: step: 260/529, loss: 2.269744800287299e-05 2023-01-21 12:34:52.635347: step: 264/529, loss: 0.0006556510925292969 2023-01-21 12:34:53.761133: step: 268/529, loss: 0.03521137312054634 2023-01-21 12:34:54.885459: step: 272/529, loss: 0.02560262754559517 2023-01-21 12:34:56.030565: step: 276/529, loss: 0.003953170962631702 2023-01-21 12:34:57.188676: step: 280/529, loss: 0.008655357174575329 2023-01-21 12:34:58.316170: step: 284/529, loss: 0.024987315759062767 2023-01-21 12:34:59.457218: step: 288/529, loss: 0.009451293386518955 2023-01-21 12:35:00.580451: step: 292/529, loss: 0.0004525184631347656 2023-01-21 12:35:01.719186: step: 296/529, loss: 0.011457539163529873 2023-01-21 12:35:02.860253: step: 300/529, loss: 0.009892558678984642 2023-01-21 12:35:03.986302: step: 304/529, loss: 0.00036964414175599813 2023-01-21 12:35:05.128489: step: 308/529, loss: 0.0017229079967364669 2023-01-21 12:35:06.229569: step: 312/529, loss: 1.7976761228055693e-05 2023-01-21 12:35:07.369544: step: 316/529, loss: 0.0034877776633948088 2023-01-21 12:35:08.515755: step: 320/529, loss: 0.25906485319137573 2023-01-21 12:35:09.630600: step: 324/529, loss: 0.006595802493393421 2023-01-21 12:35:10.766713: step: 328/529, loss: 0.06577229499816895 2023-01-21 12:35:11.930194: step: 332/529, loss: 0.01117486972361803 2023-01-21 12:35:13.064306: step: 336/529, loss: 7.247925168485381e-06 2023-01-21 12:35:14.207527: step: 340/529, loss: 0.32857057452201843 2023-01-21 12:35:15.327932: step: 344/529, loss: 0.0018787384033203125 2023-01-21 12:35:16.461358: step: 348/529, loss: 0.0004990339512005448 2023-01-21 12:35:17.602184: step: 352/529, loss: 2.193450927734375e-05 2023-01-21 12:35:18.743712: step: 356/529, loss: 0.002851295517757535 2023-01-21 12:35:19.857304: step: 360/529, loss: 0.00017371178546454757 2023-01-21 12:35:21.007740: step: 364/529, loss: 0.05306658893823624 2023-01-21 12:35:22.146810: step: 368/529, loss: 0.0007446289528161287 2023-01-21 12:35:23.249949: step: 372/529, loss: 0.0009305953863076866 2023-01-21 12:35:24.364382: step: 376/529, loss: 0.0002480983966961503 2023-01-21 12:35:25.552845: step: 380/529, loss: 0.001983690308406949 2023-01-21 12:35:26.722552: step: 384/529, loss: 8.37326078908518e-05 2023-01-21 12:35:27.846439: step: 388/529, loss: 0.0003231048467569053 2023-01-21 12:35:28.959999: step: 392/529, loss: 1.602172778802924e-05 2023-01-21 12:35:30.103359: step: 396/529, loss: 0.06917381286621094 2023-01-21 12:35:31.239383: step: 400/529, loss: 0.0008682251209393144 2023-01-21 12:35:32.353850: step: 404/529, loss: 0.04975871741771698 2023-01-21 12:35:33.490457: step: 408/529, loss: 0.0004084586980752647 2023-01-21 12:35:34.627081: step: 412/529, loss: 0.005503273103386164 2023-01-21 12:35:35.737774: step: 416/529, loss: 0.001552581787109375 2023-01-21 12:35:36.867177: step: 420/529, loss: 0.024096298962831497 2023-01-21 12:35:37.987410: step: 424/529, loss: 0.0010208130115643144 2023-01-21 12:35:39.116774: step: 428/529, loss: 0.020763827487826347 2023-01-21 12:35:40.239734: step: 432/529, loss: 0.004187298007309437 2023-01-21 12:35:41.370866: step: 436/529, loss: 0.004733562469482422 2023-01-21 12:35:42.476099: step: 440/529, loss: 8.96453857421875e-05 2023-01-21 12:35:43.610859: step: 444/529, loss: 0.020523834973573685 2023-01-21 12:35:44.736756: step: 448/529, loss: 0.014217520132660866 2023-01-21 12:35:45.910093: step: 452/529, loss: 0.0004267692565917969 2023-01-21 12:35:47.037032: step: 456/529, loss: 0.00015325545973610133 2023-01-21 12:35:48.174712: step: 460/529, loss: 0.00027151702670380473 2023-01-21 12:35:49.339351: step: 464/529, loss: 1.4470517271547578e-05 2023-01-21 12:35:50.453923: step: 468/529, loss: 0.06356659531593323 2023-01-21 12:35:51.604495: step: 472/529, loss: 0.010550403967499733 2023-01-21 12:35:52.723596: step: 476/529, loss: 0.0013629914028570056 2023-01-21 12:35:53.830351: step: 480/529, loss: 0.041469767689704895 2023-01-21 12:35:54.970964: step: 484/529, loss: 0.0012273788452148438 2023-01-21 12:35:56.094358: step: 488/529, loss: 0.006653976161032915 2023-01-21 12:35:57.219955: step: 492/529, loss: 0.004865646827965975 2023-01-21 12:35:58.359910: step: 496/529, loss: 0.21770639717578888 2023-01-21 12:35:59.517951: step: 500/529, loss: 0.2050682157278061 2023-01-21 12:36:00.661151: step: 504/529, loss: 8.926392183639109e-05 2023-01-21 12:36:01.827078: step: 508/529, loss: 0.004333019256591797 2023-01-21 12:36:02.987610: step: 512/529, loss: 0.0018676758045330644 2023-01-21 12:36:04.119569: step: 516/529, loss: 0.010727119632065296 2023-01-21 12:36:05.270166: step: 520/529, loss: 0.0009628295665606856 2023-01-21 12:36:06.394629: step: 524/529, loss: 8.296967280330136e-06 2023-01-21 12:36:07.518547: step: 528/529, loss: 0.0008838653448037803 2023-01-21 12:36:08.642848: step: 532/529, loss: 0.013854789547622204 2023-01-21 12:36:09.759947: step: 536/529, loss: 4.310607982915826e-05 2023-01-21 12:36:10.898550: step: 540/529, loss: 9.15527380129788e-06 2023-01-21 12:36:12.039846: step: 544/529, loss: 0.00039014816866256297 2023-01-21 12:36:13.198562: step: 548/529, loss: 0.02130408212542534 2023-01-21 12:36:14.335455: step: 552/529, loss: 0.000251579302130267 2023-01-21 12:36:15.528476: step: 556/529, loss: 0.0018236159812659025 2023-01-21 12:36:16.649521: step: 560/529, loss: 0.0002956390380859375 2023-01-21 12:36:17.754572: step: 564/529, loss: 0.025072097778320312 2023-01-21 12:36:18.897707: step: 568/529, loss: 0.015985680744051933 2023-01-21 12:36:20.016669: step: 572/529, loss: 0.00033359526423737407 2023-01-21 12:36:21.131290: step: 576/529, loss: -5.722044988942798e-07 2023-01-21 12:36:22.255095: step: 580/529, loss: 0.0002017974911723286 2023-01-21 12:36:23.389853: step: 584/529, loss: 1.487731969973538e-05 2023-01-21 12:36:24.504633: step: 588/529, loss: 0.0006965161301195621 2023-01-21 12:36:25.639384: step: 592/529, loss: 0.00042514799861237407 2023-01-21 12:36:26.765260: step: 596/529, loss: 0.00014047621516510844 2023-01-21 12:36:27.889081: step: 600/529, loss: 1.4019013178767636e-05 2023-01-21 12:36:29.038769: step: 604/529, loss: 0.0005537032848224044 2023-01-21 12:36:30.193556: step: 608/529, loss: 0.00769310025498271 2023-01-21 12:36:31.316881: step: 612/529, loss: 0.0006879806751385331 2023-01-21 12:36:32.440284: step: 616/529, loss: 0.001835298608057201 2023-01-21 12:36:33.580261: step: 620/529, loss: 0.005445671267807484 2023-01-21 12:36:34.732162: step: 624/529, loss: 0.0006382942083291709 2023-01-21 12:36:35.859499: step: 628/529, loss: 0.00039935109089128673 2023-01-21 12:36:37.004269: step: 632/529, loss: 3.719329833984375e-05 2023-01-21 12:36:38.156120: step: 636/529, loss: 0.0002956390380859375 2023-01-21 12:36:39.288341: step: 640/529, loss: 0.0010980607476085424 2023-01-21 12:36:40.405978: step: 644/529, loss: 0.002021980471909046 2023-01-21 12:36:41.567516: step: 648/529, loss: 9.307861910201609e-05 2023-01-21 12:36:42.727729: step: 652/529, loss: 0.00014057158841751516 2023-01-21 12:36:43.857929: step: 656/529, loss: 0.0010408401722088456 2023-01-21 12:36:44.988768: step: 660/529, loss: 1.792907642084174e-05 2023-01-21 12:36:46.123437: step: 664/529, loss: 0.0005223274347372353 2023-01-21 12:36:47.262279: step: 668/529, loss: 0.002431964734569192 2023-01-21 12:36:48.391685: step: 672/529, loss: 0.046492960304021835 2023-01-21 12:36:49.565051: step: 676/529, loss: 0.002403831575065851 2023-01-21 12:36:50.726250: step: 680/529, loss: 0.010167312808334827 2023-01-21 12:36:51.855054: step: 684/529, loss: 0.0009099007002077997 2023-01-21 12:36:53.020643: step: 688/529, loss: 0.004207420628517866 2023-01-21 12:36:54.133115: step: 692/529, loss: 0.04148273542523384 2023-01-21 12:36:55.262116: step: 696/529, loss: 0.0002237796870758757 2023-01-21 12:36:56.383993: step: 700/529, loss: 0.0004596710205078125 2023-01-21 12:36:57.547048: step: 704/529, loss: 0.015514945611357689 2023-01-21 12:36:58.653631: step: 708/529, loss: 0.006739520933479071 2023-01-21 12:36:59.760713: step: 712/529, loss: 0.006988429930061102 2023-01-21 12:37:00.933810: step: 716/529, loss: 0.06101221963763237 2023-01-21 12:37:02.096346: step: 720/529, loss: 0.0013698578113690019 2023-01-21 12:37:03.224045: step: 724/529, loss: 7.934570749057457e-05 2023-01-21 12:37:04.361957: step: 728/529, loss: 2.8038026357535273e-05 2023-01-21 12:37:05.522403: step: 732/529, loss: 0.0002007007715292275 2023-01-21 12:37:06.630193: step: 736/529, loss: 0.003244781633839011 2023-01-21 12:37:07.753395: step: 740/529, loss: 0.0007438659085892141 2023-01-21 12:37:08.900024: step: 744/529, loss: 0.06237611547112465 2023-01-21 12:37:10.030074: step: 748/529, loss: 0.0011137009132653475 2023-01-21 12:37:11.171438: step: 752/529, loss: 0.1900525987148285 2023-01-21 12:37:12.315480: step: 756/529, loss: 0.0007118224748410285 2023-01-21 12:37:13.447800: step: 760/529, loss: 3.070831371587701e-05 2023-01-21 12:37:14.573826: step: 764/529, loss: 0.00010261535499012098 2023-01-21 12:37:15.738835: step: 768/529, loss: 0.008781243115663528 2023-01-21 12:37:16.888339: step: 772/529, loss: 0.0004192352353129536 2023-01-21 12:37:18.046024: step: 776/529, loss: 0.005642509553581476 2023-01-21 12:37:19.185771: step: 780/529, loss: 3.585815284168348e-05 2023-01-21 12:37:20.339594: step: 784/529, loss: 0.05169963836669922 2023-01-21 12:37:21.474948: step: 788/529, loss: 0.00028996466426178813 2023-01-21 12:37:22.576372: step: 792/529, loss: 0.00020322800264693797 2023-01-21 12:37:23.732854: step: 796/529, loss: 3.1256677175406367e-05 2023-01-21 12:37:24.878928: step: 800/529, loss: 0.008978747762739658 2023-01-21 12:37:25.999187: step: 804/529, loss: 0.012032079510390759 2023-01-21 12:37:27.104280: step: 808/529, loss: 0.0001898527261801064 2023-01-21 12:37:28.218626: step: 812/529, loss: 0.0015240252250805497 2023-01-21 12:37:29.316791: step: 816/529, loss: 0.01431417465209961 2023-01-21 12:37:30.462717: step: 820/529, loss: 0.015032864175736904 2023-01-21 12:37:31.571835: step: 824/529, loss: 0.005633163265883923 2023-01-21 12:37:32.691025: step: 828/529, loss: 6.151199340820312e-05 2023-01-21 12:37:33.837840: step: 832/529, loss: 0.00212268834002316 2023-01-21 12:37:34.948295: step: 836/529, loss: 0.00031828880310058594 2023-01-21 12:37:36.065803: step: 840/529, loss: 0.0003203391970600933 2023-01-21 12:37:37.205154: step: 844/529, loss: 0.0004020690976176411 2023-01-21 12:37:38.339659: step: 848/529, loss: 0.0007855416042730212 2023-01-21 12:37:39.467072: step: 852/529, loss: 1.5258789289873675e-06 2023-01-21 12:37:40.595334: step: 856/529, loss: 0.0020596503745764494 2023-01-21 12:37:41.757810: step: 860/529, loss: 0.007476329803466797 2023-01-21 12:37:42.895483: step: 864/529, loss: 7.286071922862902e-05 2023-01-21 12:37:44.020528: step: 868/529, loss: 0.008346081711351871 2023-01-21 12:37:45.159762: step: 872/529, loss: 0.004449081607162952 2023-01-21 12:37:46.287873: step: 876/529, loss: 0.0017236232524737716 2023-01-21 12:37:47.428863: step: 880/529, loss: 0.010668945498764515 2023-01-21 12:37:48.572519: step: 884/529, loss: 0.0004935223842039704 2023-01-21 12:37:49.693974: step: 888/529, loss: 0.0005986213800497353 2023-01-21 12:37:50.836133: step: 892/529, loss: 0.0002609252987895161 2023-01-21 12:37:51.994122: step: 896/529, loss: 0.006785011384636164 2023-01-21 12:37:53.126661: step: 900/529, loss: 0.0016184330452233553 2023-01-21 12:37:54.242283: step: 904/529, loss: 6.880760338390246e-05 2023-01-21 12:37:55.389198: step: 908/529, loss: 0.00557708740234375 2023-01-21 12:37:56.564083: step: 912/529, loss: 0.0028951645363122225 2023-01-21 12:37:57.694003: step: 916/529, loss: 0.006602478213608265 2023-01-21 12:37:58.810189: step: 920/529, loss: 6.0558319091796875e-05 2023-01-21 12:37:59.938608: step: 924/529, loss: 0.06734314560890198 2023-01-21 12:38:01.064188: step: 928/529, loss: 0.00024752618628554046 2023-01-21 12:38:02.204602: step: 932/529, loss: 0.014574432745575905 2023-01-21 12:38:03.332387: step: 936/529, loss: 0.021332740783691406 2023-01-21 12:38:04.478168: step: 940/529, loss: 0.0013212204212322831 2023-01-21 12:38:05.612010: step: 944/529, loss: 0.07049636542797089 2023-01-21 12:38:06.725230: step: 948/529, loss: 0.0006916046841070056 2023-01-21 12:38:07.856703: step: 952/529, loss: 3.948211815441027e-05 2023-01-21 12:38:08.981646: step: 956/529, loss: 4.806519064004533e-05 2023-01-21 12:38:10.119948: step: 960/529, loss: 0.00018720627122092992 2023-01-21 12:38:11.263137: step: 964/529, loss: 0.003830146975815296 2023-01-21 12:38:12.381845: step: 968/529, loss: 0.0006706237909384072 2023-01-21 12:38:13.501911: step: 972/529, loss: 0.03444795683026314 2023-01-21 12:38:14.635528: step: 976/529, loss: 0.0001558303920319304 2023-01-21 12:38:15.781479: step: 980/529, loss: 0.02129383198916912 2023-01-21 12:38:16.956955: step: 984/529, loss: 0.020460795611143112 2023-01-21 12:38:18.097440: step: 988/529, loss: 0.0044920919463038445 2023-01-21 12:38:19.211846: step: 992/529, loss: 3.3664702641544864e-05 2023-01-21 12:38:20.340543: step: 996/529, loss: 0.0010754584800451994 2023-01-21 12:38:21.476373: step: 1000/529, loss: 0.039188578724861145 2023-01-21 12:38:22.633409: step: 1004/529, loss: 2.6702882678364404e-05 2023-01-21 12:38:23.758497: step: 1008/529, loss: 0.006126213353127241 2023-01-21 12:38:24.895504: step: 1012/529, loss: 0.0001625060976948589 2023-01-21 12:38:26.060978: step: 1016/529, loss: 0.0007300376892089844 2023-01-21 12:38:27.192589: step: 1020/529, loss: 0.0005289077525958419 2023-01-21 12:38:28.321456: step: 1024/529, loss: 5.683899507857859e-05 2023-01-21 12:38:29.441722: step: 1028/529, loss: 0.021750641986727715 2023-01-21 12:38:30.566421: step: 1032/529, loss: 0.0009449959034100175 2023-01-21 12:38:31.695215: step: 1036/529, loss: 0.0005115509266033769 2023-01-21 12:38:32.830309: step: 1040/529, loss: 0.006106376647949219 2023-01-21 12:38:33.972078: step: 1044/529, loss: 0.0032206536270678043 2023-01-21 12:38:35.082378: step: 1048/529, loss: 0.0008114815573208034 2023-01-21 12:38:36.167358: step: 1052/529, loss: 0.6567395329475403 2023-01-21 12:38:37.312775: step: 1056/529, loss: 0.020140837877988815 2023-01-21 12:38:38.449754: step: 1060/529, loss: 0.0006835937965661287 2023-01-21 12:38:39.569571: step: 1064/529, loss: 0.012934398837387562 2023-01-21 12:38:40.717764: step: 1068/529, loss: 0.19415606558322906 2023-01-21 12:38:41.875114: step: 1072/529, loss: 0.013765716925263405 2023-01-21 12:38:42.991218: step: 1076/529, loss: 0.004931545350700617 2023-01-21 12:38:44.124586: step: 1080/529, loss: 0.0013827324146404862 2023-01-21 12:38:45.248358: step: 1084/529, loss: 0.00039777756319381297 2023-01-21 12:38:46.388505: step: 1088/529, loss: 0.001670456025749445 2023-01-21 12:38:47.553904: step: 1092/529, loss: 0.025157928466796875 2023-01-21 12:38:48.670197: step: 1096/529, loss: 0.0002342224179301411 2023-01-21 12:38:49.822289: step: 1100/529, loss: 0.007592892739921808 2023-01-21 12:38:50.966091: step: 1104/529, loss: 0.014551926404237747 2023-01-21 12:38:52.083050: step: 1108/529, loss: 1.634371042251587 2023-01-21 12:38:53.272611: step: 1112/529, loss: 1.8739699953584932e-05 2023-01-21 12:38:54.404977: step: 1116/529, loss: 0.0014375686878338456 2023-01-21 12:38:55.512840: step: 1120/529, loss: 0.029003240168094635 2023-01-21 12:38:56.664589: step: 1124/529, loss: 0.00012493133544921875 2023-01-21 12:38:57.802737: step: 1128/529, loss: 0.00037021638127043843 2023-01-21 12:38:58.928660: step: 1132/529, loss: 0.0034265518188476562 2023-01-21 12:39:00.063072: step: 1136/529, loss: 0.09349896013736725 2023-01-21 12:39:01.193991: step: 1140/529, loss: 0.009424782358109951 2023-01-21 12:39:02.361597: step: 1144/529, loss: 0.0001985549897653982 2023-01-21 12:39:03.491268: step: 1148/529, loss: 0.07612569630146027 2023-01-21 12:39:04.658457: step: 1152/529, loss: 0.030254008248448372 2023-01-21 12:39:05.801140: step: 1156/529, loss: 0.0068933009169995785 2023-01-21 12:39:06.939398: step: 1160/529, loss: 0.0003364086151123047 2023-01-21 12:39:08.070032: step: 1164/529, loss: 0.08749179542064667 2023-01-21 12:39:09.189165: step: 1168/529, loss: 0.0007830619579181075 2023-01-21 12:39:10.347407: step: 1172/529, loss: 0.0009251594892702997 2023-01-21 12:39:11.457675: step: 1176/529, loss: 0.00047969818115234375 2023-01-21 12:39:12.636526: step: 1180/529, loss: 0.003381919814273715 2023-01-21 12:39:13.802014: step: 1184/529, loss: 0.01605110429227352 2023-01-21 12:39:14.968171: step: 1188/529, loss: 0.009189128875732422 2023-01-21 12:39:16.107385: step: 1192/529, loss: 0.00719528179615736 2023-01-21 12:39:17.252147: step: 1196/529, loss: 0.09588488936424255 2023-01-21 12:39:18.391010: step: 1200/529, loss: 6.599425978492945e-05 2023-01-21 12:39:19.494683: step: 1204/529, loss: 0.0011001586681231856 2023-01-21 12:39:20.614738: step: 1208/529, loss: 0.0010382651817053556 2023-01-21 12:39:21.769161: step: 1212/529, loss: 0.00047712327796034515 2023-01-21 12:39:22.908432: step: 1216/529, loss: 0.0001516819029347971 2023-01-21 12:39:24.056840: step: 1220/529, loss: 0.002011299366131425 2023-01-21 12:39:25.184397: step: 1224/529, loss: 0.0009478569263592362 2023-01-21 12:39:26.318633: step: 1228/529, loss: 0.004867363255470991 2023-01-21 12:39:27.442260: step: 1232/529, loss: 0.0034030915703624487 2023-01-21 12:39:28.578838: step: 1236/529, loss: 0.001762819243595004 2023-01-21 12:39:29.745375: step: 1240/529, loss: 0.002093506045639515 2023-01-21 12:39:30.893796: step: 1244/529, loss: 0.00010175704665016383 2023-01-21 12:39:32.026063: step: 1248/529, loss: 3.6811830796068534e-05 2023-01-21 12:39:33.137880: step: 1252/529, loss: 4.329681542003527e-05 2023-01-21 12:39:34.283343: step: 1256/529, loss: 0.00011024475679732859 2023-01-21 12:39:35.457082: step: 1260/529, loss: 0.15667057037353516 2023-01-21 12:39:36.600426: step: 1264/529, loss: -1.5640260244254023e-05 2023-01-21 12:39:37.747542: step: 1268/529, loss: 0.07108354568481445 2023-01-21 12:39:38.875148: step: 1272/529, loss: 0.0019440652104094625 2023-01-21 12:39:40.026942: step: 1276/529, loss: 0.0006957054138183594 2023-01-21 12:39:41.162307: step: 1280/529, loss: 1.859665098891128e-05 2023-01-21 12:39:42.294885: step: 1284/529, loss: 0.02098236046731472 2023-01-21 12:39:43.433208: step: 1288/529, loss: 0.10217314213514328 2023-01-21 12:39:44.543539: step: 1292/529, loss: 2.6130677724722773e-05 2023-01-21 12:39:45.657946: step: 1296/529, loss: 0.21503810584545135 2023-01-21 12:39:46.835227: step: 1300/529, loss: 0.4905174672603607 2023-01-21 12:39:47.965346: step: 1304/529, loss: 0.014315987005829811 2023-01-21 12:39:49.077309: step: 1308/529, loss: 0.035672761499881744 2023-01-21 12:39:50.175067: step: 1312/529, loss: 0.010269737802445889 2023-01-21 12:39:51.326365: step: 1316/529, loss: 0.0024518966674804688 2023-01-21 12:39:52.449083: step: 1320/529, loss: 0.002910614013671875 2023-01-21 12:39:53.584549: step: 1324/529, loss: 0.0022567748092114925 2023-01-21 12:39:54.709393: step: 1328/529, loss: 5.340576171875e-05 2023-01-21 12:39:55.844994: step: 1332/529, loss: 4.024505687993951e-05 2023-01-21 12:39:56.976485: step: 1336/529, loss: 0.005511855706572533 2023-01-21 12:39:58.099325: step: 1340/529, loss: 0.0011119365226477385 2023-01-21 12:39:59.256850: step: 1344/529, loss: 0.15271367132663727 2023-01-21 12:40:00.446261: step: 1348/529, loss: 0.0032224655151367188 2023-01-21 12:40:01.572376: step: 1352/529, loss: 0.0020291327964514494 2023-01-21 12:40:02.702521: step: 1356/529, loss: 0.0032241821754723787 2023-01-21 12:40:03.838213: step: 1360/529, loss: 1.0585785275907256e-05 2023-01-21 12:40:04.983390: step: 1364/529, loss: 0.0847620964050293 2023-01-21 12:40:06.167388: step: 1368/529, loss: 0.00219554896466434 2023-01-21 12:40:07.297114: step: 1372/529, loss: 1.0251998901367188e-05 2023-01-21 12:40:08.457373: step: 1376/529, loss: 0.0014752388233318925 2023-01-21 12:40:09.640557: step: 1380/529, loss: 0.003334236331284046 2023-01-21 12:40:10.762216: step: 1384/529, loss: 0.030622007325291634 2023-01-21 12:40:11.935859: step: 1388/529, loss: 0.00020895004854537547 2023-01-21 12:40:13.083882: step: 1392/529, loss: 0.027738571166992188 2023-01-21 12:40:14.246411: step: 1396/529, loss: 0.0002605438057798892 2023-01-21 12:40:15.385318: step: 1400/529, loss: 0.015262985602021217 2023-01-21 12:40:16.532880: step: 1404/529, loss: 0.009144115261733532 2023-01-21 12:40:17.679138: step: 1408/529, loss: 8.678436643094756e-06 2023-01-21 12:40:18.833111: step: 1412/529, loss: 0.0011796951293945312 2023-01-21 12:40:19.990616: step: 1416/529, loss: 4.1484832763671875e-05 2023-01-21 12:40:21.141325: step: 1420/529, loss: 0.03665952757000923 2023-01-21 12:40:22.309504: step: 1424/529, loss: 0.017798423767089844 2023-01-21 12:40:23.458518: step: 1428/529, loss: 0.2316930741071701 2023-01-21 12:40:24.575930: step: 1432/529, loss: 0.0008451461908407509 2023-01-21 12:40:25.707060: step: 1436/529, loss: 0.00020847321138717234 2023-01-21 12:40:26.827384: step: 1440/529, loss: 0.01667856052517891 2023-01-21 12:40:27.950595: step: 1444/529, loss: 8.077621896518394e-05 2023-01-21 12:40:29.089057: step: 1448/529, loss: 3.912448664777912e-05 2023-01-21 12:40:30.260147: step: 1452/529, loss: 0.020384978502988815 2023-01-21 12:40:31.403010: step: 1456/529, loss: 0.002882003551349044 2023-01-21 12:40:32.523444: step: 1460/529, loss: 0.022124910727143288 2023-01-21 12:40:33.638833: step: 1464/529, loss: 0.5624160170555115 2023-01-21 12:40:34.765219: step: 1468/529, loss: 0.007760906592011452 2023-01-21 12:40:35.877745: step: 1472/529, loss: 0.000457572954474017 2023-01-21 12:40:37.012688: step: 1476/529, loss: 0.0004424095095600933 2023-01-21 12:40:38.142627: step: 1480/529, loss: 2.574920654296875e-05 2023-01-21 12:40:39.305801: step: 1484/529, loss: 0.019994543865323067 2023-01-21 12:40:40.473318: step: 1488/529, loss: 0.004147720523178577 2023-01-21 12:40:41.586433: step: 1492/529, loss: 0.01243824977427721 2023-01-21 12:40:42.739543: step: 1496/529, loss: 0.12420777976512909 2023-01-21 12:40:43.874300: step: 1500/529, loss: 0.006187248043715954 2023-01-21 12:40:45.035732: step: 1504/529, loss: 0.00045223237248137593 2023-01-21 12:40:46.203659: step: 1508/529, loss: 0.021202564239501953 2023-01-21 12:40:47.323078: step: 1512/529, loss: 0.06444673985242844 2023-01-21 12:40:48.497975: step: 1516/529, loss: 0.002732944441959262 2023-01-21 12:40:49.611147: step: 1520/529, loss: 0.007241630461066961 2023-01-21 12:40:50.722097: step: 1524/529, loss: 0.0011219263542443514 2023-01-21 12:40:51.845769: step: 1528/529, loss: 5.8746340073412284e-05 2023-01-21 12:40:53.002988: step: 1532/529, loss: 2.4890898203011602e-05 2023-01-21 12:40:54.138640: step: 1536/529, loss: 4.6062468754826114e-05 2023-01-21 12:40:55.257799: step: 1540/529, loss: 0.00015048979548737407 2023-01-21 12:40:56.414225: step: 1544/529, loss: 0.002295875456184149 2023-01-21 12:40:57.532082: step: 1548/529, loss: 0.00024538038996979594 2023-01-21 12:40:58.652659: step: 1552/529, loss: 0.01145162619650364 2023-01-21 12:40:59.805891: step: 1556/529, loss: 4.6348573960131034e-05 2023-01-21 12:41:00.919540: step: 1560/529, loss: 0.005608749110251665 2023-01-21 12:41:02.043289: step: 1564/529, loss: 0.0028388977516442537 2023-01-21 12:41:03.154730: step: 1568/529, loss: 0.03195914998650551 2023-01-21 12:41:04.259685: step: 1572/529, loss: 0.007009029388427734 2023-01-21 12:41:05.378332: step: 1576/529, loss: 0.0004108428838662803 2023-01-21 12:41:06.526817: step: 1580/529, loss: 0.005553627386689186 2023-01-21 12:41:07.681243: step: 1584/529, loss: 0.0016283990116789937 2023-01-21 12:41:08.800562: step: 1588/529, loss: 0.010003376752138138 2023-01-21 12:41:09.927748: step: 1592/529, loss: 0.00010910033597610891 2023-01-21 12:41:11.076212: step: 1596/529, loss: 0.012080764397978783 2023-01-21 12:41:12.205751: step: 1600/529, loss: 0.00033059119596146047 2023-01-21 12:41:13.344940: step: 1604/529, loss: 0.002101039979606867 2023-01-21 12:41:14.468208: step: 1608/529, loss: 0.006660080049186945 2023-01-21 12:41:15.626498: step: 1612/529, loss: 0.0007124424446374178 2023-01-21 12:41:16.751602: step: 1616/529, loss: 0.00019922257342841476 2023-01-21 12:41:17.909597: step: 1620/529, loss: 0.007243060972541571 2023-01-21 12:41:19.060654: step: 1624/529, loss: 0.0008311271667480469 2023-01-21 12:41:20.182983: step: 1628/529, loss: 0.00033082961454056203 2023-01-21 12:41:21.294447: step: 1632/529, loss: 0.01825103722512722 2023-01-21 12:41:22.412120: step: 1636/529, loss: 0.010642719455063343 2023-01-21 12:41:23.532615: step: 1640/529, loss: 0.003831291338428855 2023-01-21 12:41:24.692171: step: 1644/529, loss: 0.009654616937041283 2023-01-21 12:41:25.822330: step: 1648/529, loss: 0.0029598234687000513 2023-01-21 12:41:26.979092: step: 1652/529, loss: 0.03948516771197319 2023-01-21 12:41:28.120629: step: 1656/529, loss: 0.00042591095552779734 2023-01-21 12:41:29.248296: step: 1660/529, loss: 0.002352428622543812 2023-01-21 12:41:30.408389: step: 1664/529, loss: 0.006963538937270641 2023-01-21 12:41:31.549640: step: 1668/529, loss: 0.018339253962039948 2023-01-21 12:41:32.711483: step: 1672/529, loss: 0.007504081819206476 2023-01-21 12:41:33.858035: step: 1676/529, loss: 0.0008610725635662675 2023-01-21 12:41:35.002332: step: 1680/529, loss: 0.005444335751235485 2023-01-21 12:41:36.125166: step: 1684/529, loss: 4.186630030744709e-05 2023-01-21 12:41:37.257612: step: 1688/529, loss: 0.026473140344023705 2023-01-21 12:41:38.402879: step: 1692/529, loss: 9.632110595703125e-05 2023-01-21 12:41:39.521049: step: 1696/529, loss: 0.020908165723085403 2023-01-21 12:41:40.628419: step: 1700/529, loss: 0.0003749847528524697 2023-01-21 12:41:41.765483: step: 1704/529, loss: 0.010650825686752796 2023-01-21 12:41:42.895806: step: 1708/529, loss: -1.239776111106039e-06 2023-01-21 12:41:44.068136: step: 1712/529, loss: 0.007590102963149548 2023-01-21 12:41:45.191954: step: 1716/529, loss: 0.03956222906708717 2023-01-21 12:41:46.314262: step: 1720/529, loss: 0.0017209530342370272 2023-01-21 12:41:47.478459: step: 1724/529, loss: 0.00017280578322242945 2023-01-21 12:41:48.626439: step: 1728/529, loss: -1.792907642084174e-05 2023-01-21 12:41:49.760394: step: 1732/529, loss: 0.00018224716768600047 2023-01-21 12:41:50.892010: step: 1736/529, loss: 0.0002305984526174143 2023-01-21 12:41:52.040453: step: 1740/529, loss: 0.001506853150203824 2023-01-21 12:41:53.177823: step: 1744/529, loss: 0.014883900061249733 2023-01-21 12:41:54.303683: step: 1748/529, loss: 0.025953389704227448 2023-01-21 12:41:55.450778: step: 1752/529, loss: 0.0035105706192553043 2023-01-21 12:41:56.597883: step: 1756/529, loss: 0.006053662393242121 2023-01-21 12:41:57.763537: step: 1760/529, loss: 0.0061754705384373665 2023-01-21 12:41:58.915331: step: 1764/529, loss: 0.09113350510597229 2023-01-21 12:42:00.026681: step: 1768/529, loss: 0.024425029754638672 2023-01-21 12:42:01.169421: step: 1772/529, loss: 0.010309219360351562 2023-01-21 12:42:02.291019: step: 1776/529, loss: 0.00047364237252622843 2023-01-21 12:42:03.393538: step: 1780/529, loss: 0.0006474495166912675 2023-01-21 12:42:04.537901: step: 1784/529, loss: 0.004006385803222656 2023-01-21 12:42:05.664027: step: 1788/529, loss: 0.0003407478507142514 2023-01-21 12:42:06.791240: step: 1792/529, loss: 0.011897278018295765 2023-01-21 12:42:07.915248: step: 1796/529, loss: 0.022905731573700905 2023-01-21 12:42:09.028345: step: 1800/529, loss: 0.0003079414600506425 2023-01-21 12:42:10.186513: step: 1804/529, loss: 0.0005990982172079384 2023-01-21 12:42:11.343362: step: 1808/529, loss: 0.0005940438131801784 2023-01-21 12:42:12.488353: step: 1812/529, loss: 0.00010051726712845266 2023-01-21 12:42:13.598894: step: 1816/529, loss: 0.000255393999395892 2023-01-21 12:42:14.712233: step: 1820/529, loss: 0.0005625725025311112 2023-01-21 12:42:15.865647: step: 1824/529, loss: 0.013180064968764782 2023-01-21 12:42:16.996465: step: 1828/529, loss: 3.566294981283136e-05 2023-01-21 12:42:18.155536: step: 1832/529, loss: 0.017393684014678 2023-01-21 12:42:19.283389: step: 1836/529, loss: 0.019994545727968216 2023-01-21 12:42:20.407584: step: 1840/529, loss: -1.106262243411038e-05 2023-01-21 12:42:21.527513: step: 1844/529, loss: 0.00133857736364007 2023-01-21 12:42:22.662336: step: 1848/529, loss: 0.0010074615711346269 2023-01-21 12:42:23.798276: step: 1852/529, loss: 0.0013295174576342106 2023-01-21 12:42:24.979001: step: 1856/529, loss: 0.000559949898160994 2023-01-21 12:42:26.131357: step: 1860/529, loss: 0.0001224517764057964 2023-01-21 12:42:27.239940: step: 1864/529, loss: 0.024665452539920807 2023-01-21 12:42:28.364176: step: 1868/529, loss: 0.0014485359424725175 2023-01-21 12:42:29.501505: step: 1872/529, loss: 0.00034694670466706157 2023-01-21 12:42:30.615145: step: 1876/529, loss: 0.013160753063857555 2023-01-21 12:42:31.740483: step: 1880/529, loss: 3.814697265625e-06 2023-01-21 12:42:32.875029: step: 1884/529, loss: 0.018774986267089844 2023-01-21 12:42:33.994012: step: 1888/529, loss: 0.012254858389496803 2023-01-21 12:42:35.117370: step: 1892/529, loss: 0.007096671964973211 2023-01-21 12:42:36.223958: step: 1896/529, loss: 0.002977705094963312 2023-01-21 12:42:37.353357: step: 1900/529, loss: 0.0007457733154296875 2023-01-21 12:42:38.517504: step: 1904/529, loss: 0.000501442002132535 2023-01-21 12:42:39.656484: step: 1908/529, loss: 0.011918258853256702 2023-01-21 12:42:40.758745: step: 1912/529, loss: 0.0003729343297891319 2023-01-21 12:42:41.885630: step: 1916/529, loss: 3.24249267578125e-05 2023-01-21 12:42:43.011041: step: 1920/529, loss: -1.1229514711885713e-05 2023-01-21 12:42:44.143068: step: 1924/529, loss: 0.0013481139903888106 2023-01-21 12:42:45.301752: step: 1928/529, loss: 0.005128288641571999 2023-01-21 12:42:46.437670: step: 1932/529, loss: 0.00969762820750475 2023-01-21 12:42:47.569787: step: 1936/529, loss: 0.0006916046258993447 2023-01-21 12:42:48.705583: step: 1940/529, loss: 0.005804634187370539 2023-01-21 12:42:49.839667: step: 1944/529, loss: 0.0015323639381676912 2023-01-21 12:42:50.978999: step: 1948/529, loss: 0.0004495620960369706 2023-01-21 12:42:52.122122: step: 1952/529, loss: 0.06397495418787003 2023-01-21 12:42:53.268645: step: 1956/529, loss: 0.00016584395780228078 2023-01-21 12:42:54.386574: step: 1960/529, loss: 0.010841942392289639 2023-01-21 12:42:55.516445: step: 1964/529, loss: 0.0727274939417839 2023-01-21 12:42:56.671352: step: 1968/529, loss: 0.002363014267757535 2023-01-21 12:42:57.807454: step: 1972/529, loss: 0.005018997006118298 2023-01-21 12:42:58.948652: step: 1976/529, loss: 0.0017253876430913806 2023-01-21 12:43:00.088665: step: 1980/529, loss: 0.001307678292505443 2023-01-21 12:43:01.207515: step: 1984/529, loss: 0.0013345718616619706 2023-01-21 12:43:02.353304: step: 1988/529, loss: 0.007514762692153454 2023-01-21 12:43:03.468050: step: 1992/529, loss: 0.00115289690438658 2023-01-21 12:43:04.615521: step: 1996/529, loss: 0.01179647445678711 2023-01-21 12:43:05.781250: step: 2000/529, loss: 0.10675173252820969 2023-01-21 12:43:06.915434: step: 2004/529, loss: 0.0003572463756427169 2023-01-21 12:43:08.040857: step: 2008/529, loss: 0.006406307686120272 2023-01-21 12:43:09.160440: step: 2012/529, loss: 0.013830756768584251 2023-01-21 12:43:10.363903: step: 2016/529, loss: 0.0017137527465820312 2023-01-21 12:43:11.495502: step: 2020/529, loss: 0.005376243963837624 2023-01-21 12:43:12.660181: step: 2024/529, loss: 0.030878830701112747 2023-01-21 12:43:13.810281: step: 2028/529, loss: 0.25017786026000977 2023-01-21 12:43:14.958692: step: 2032/529, loss: 0.0047931671142578125 2023-01-21 12:43:16.082144: step: 2036/529, loss: 4.863739013671875e-05 2023-01-21 12:43:17.202071: step: 2040/529, loss: 0.018825722858309746 2023-01-21 12:43:18.309454: step: 2044/529, loss: 0.022179795429110527 2023-01-21 12:43:19.462497: step: 2048/529, loss: 0.005130005069077015 2023-01-21 12:43:20.598252: step: 2052/529, loss: 0.0025697709061205387 2023-01-21 12:43:21.755161: step: 2056/529, loss: 0.014226055704057217 2023-01-21 12:43:22.880591: step: 2060/529, loss: 0.005357170011848211 2023-01-21 12:43:24.002608: step: 2064/529, loss: 0.05276503413915634 2023-01-21 12:43:25.130989: step: 2068/529, loss: 0.0840001180768013 2023-01-21 12:43:26.253929: step: 2072/529, loss: 0.004636287689208984 2023-01-21 12:43:27.365187: step: 2076/529, loss: 0.0019740702118724585 2023-01-21 12:43:28.500306: step: 2080/529, loss: 0.000949859619140625 2023-01-21 12:43:29.636623: step: 2084/529, loss: 0.0011232375400140882 2023-01-21 12:43:30.816826: step: 2088/529, loss: 0.004808330442756414 2023-01-21 12:43:31.956642: step: 2092/529, loss: 0.0001665592280915007 2023-01-21 12:43:33.080540: step: 2096/529, loss: 0.02196826972067356 2023-01-21 12:43:34.209808: step: 2100/529, loss: 4.229545811540447e-05 2023-01-21 12:43:35.344290: step: 2104/529, loss: 0.02030634880065918 2023-01-21 12:43:36.493004: step: 2108/529, loss: 0.0010421753395348787 2023-01-21 12:43:37.608006: step: 2112/529, loss: 0.011526776477694511 2023-01-21 12:43:38.725079: step: 2116/529, loss: 0.00022633076878264546 ================================================== Loss: 0.021 -------------------- Dev: {'event': {'p': 0.617740232312566, 'r': 0.7789613848202397, 'f1': 0.6890459363957597}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Test: {'event': {'p': 0.6130374479889042, 'r': 0.7682502896871379, 'f1': 0.6819233736178967}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Chinese: {'event': {'p': 0.625, 'r': 0.9259259259259259, 'f1': 0.7462686567164178}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Korean: {'event': {'p': 0.5517241379310345, 'r': 0.5079365079365079, 'f1': 0.5289256198347106}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Russian: {'event': {'p': 0.425, 'r': 0.4722222222222222, 'f1': 0.4473684210526316}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 24 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:44:19.740960: step: 4/529, loss: 0.016735076904296875 2023-01-21 12:44:20.861843: step: 8/529, loss: 0.0003046035999432206 2023-01-21 12:44:21.978297: step: 12/529, loss: 0.00016117095947265625 2023-01-21 12:44:23.108155: step: 16/529, loss: 0.0005403042305260897 2023-01-21 12:44:24.244509: step: 20/529, loss: 8.468628220725805e-05 2023-01-21 12:44:25.367632: step: 24/529, loss: 0.0004939079517498612 2023-01-21 12:44:26.534725: step: 28/529, loss: 0.0001625060976948589 2023-01-21 12:44:27.630309: step: 32/529, loss: 0.0026155950035899878 2023-01-21 12:44:28.740466: step: 36/529, loss: 4.882812572759576e-05 2023-01-21 12:44:29.871632: step: 40/529, loss: 8.535384949936997e-06 2023-01-21 12:44:30.990398: step: 44/529, loss: 1.9073513612966053e-07 2023-01-21 12:44:32.088555: step: 48/529, loss: 1.5258789289873675e-06 2023-01-21 12:44:33.199853: step: 52/529, loss: 0.001263427664525807 2023-01-21 12:44:34.336859: step: 56/529, loss: 0.009744835086166859 2023-01-21 12:44:35.477431: step: 60/529, loss: 0.005537509918212891 2023-01-21 12:44:36.623609: step: 64/529, loss: 0.03282728046178818 2023-01-21 12:44:37.751048: step: 68/529, loss: 0.007790184114128351 2023-01-21 12:44:38.891962: step: 72/529, loss: 0.03408775478601456 2023-01-21 12:44:40.025643: step: 76/529, loss: 0.04925575107336044 2023-01-21 12:44:41.169945: step: 80/529, loss: 0.0003067493671551347 2023-01-21 12:44:42.366582: step: 84/529, loss: 0.005382537841796875 2023-01-21 12:44:43.494210: step: 88/529, loss: 0.002236104104667902 2023-01-21 12:44:44.602825: step: 92/529, loss: 0.004925775341689587 2023-01-21 12:44:45.705797: step: 96/529, loss: 0.003086090087890625 2023-01-21 12:44:46.857475: step: 100/529, loss: 0.004764556884765625 2023-01-21 12:44:48.011804: step: 104/529, loss: 2.670288495210116e-06 2023-01-21 12:44:49.134611: step: 108/529, loss: 0.0032699585426598787 2023-01-21 12:44:50.241750: step: 112/529, loss: 0.013358878903090954 2023-01-21 12:44:51.373977: step: 116/529, loss: 0.0001672744838288054 2023-01-21 12:44:52.545463: step: 120/529, loss: 5.531311398954131e-05 2023-01-21 12:44:53.673319: step: 124/529, loss: 0.02824885956943035 2023-01-21 12:44:54.800619: step: 128/529, loss: 1.182556115963962e-05 2023-01-21 12:44:55.936223: step: 132/529, loss: 0.004144859034568071 2023-01-21 12:44:57.057663: step: 136/529, loss: 0.0004817962762899697 2023-01-21 12:44:58.209558: step: 140/529, loss: 0.0010352134704589844 2023-01-21 12:44:59.325475: step: 144/529, loss: 5.5694577895337716e-05 2023-01-21 12:45:00.493337: step: 148/529, loss: 0.007573366165161133 2023-01-21 12:45:01.642859: step: 152/529, loss: 3.43322744811303e-06 2023-01-21 12:45:02.787847: step: 156/529, loss: 1.8119810647476697e-06 2023-01-21 12:45:03.924582: step: 160/529, loss: 0.003052663756534457 2023-01-21 12:45:05.065871: step: 164/529, loss: 0.06635075062513351 2023-01-21 12:45:06.210697: step: 168/529, loss: 0.00048704148503020406 2023-01-21 12:45:07.353263: step: 172/529, loss: 0.6253318786621094 2023-01-21 12:45:08.491226: step: 176/529, loss: 0.001586103462614119 2023-01-21 12:45:09.631918: step: 180/529, loss: 0.013029813766479492 2023-01-21 12:45:10.752910: step: 184/529, loss: 0.03591776266694069 2023-01-21 12:45:11.927156: step: 188/529, loss: 0.0006441115983761847 2023-01-21 12:45:13.050707: step: 192/529, loss: 0.0005155563121661544 2023-01-21 12:45:14.191453: step: 196/529, loss: 0.01294403150677681 2023-01-21 12:45:15.335203: step: 200/529, loss: 0.0015408515464514494 2023-01-21 12:45:16.473386: step: 204/529, loss: 0.00023283959308173507 2023-01-21 12:45:17.620779: step: 208/529, loss: 0.02212676964700222 2023-01-21 12:45:18.736710: step: 212/529, loss: 0.008818721398711205 2023-01-21 12:45:19.874190: step: 216/529, loss: 0.004449843894690275 2023-01-21 12:45:20.997704: step: 220/529, loss: 0.000255393999395892 2023-01-21 12:45:22.184775: step: 224/529, loss: 0.0005142689333297312 2023-01-21 12:45:23.293612: step: 228/529, loss: 0.00025081634521484375 2023-01-21 12:45:24.437532: step: 232/529, loss: 0.00059595110360533 2023-01-21 12:45:25.559904: step: 236/529, loss: 0.010797691531479359 2023-01-21 12:45:26.673988: step: 240/529, loss: 0.0006122589111328125 2023-01-21 12:45:27.805934: step: 244/529, loss: 0.004810523707419634 2023-01-21 12:45:28.958855: step: 248/529, loss: 0.00010070800635730848 2023-01-21 12:45:30.067568: step: 252/529, loss: 3.051757857974735e-06 2023-01-21 12:45:31.218242: step: 256/529, loss: 0.0003528594970703125 2023-01-21 12:45:32.327744: step: 260/529, loss: 0.0048877401277422905 2023-01-21 12:45:33.452886: step: 264/529, loss: 0.005888843908905983 2023-01-21 12:45:34.578154: step: 268/529, loss: 0.0011059760581701994 2023-01-21 12:45:35.726834: step: 272/529, loss: 0.0018854141235351562 2023-01-21 12:45:36.845125: step: 276/529, loss: 0.001800537109375 2023-01-21 12:45:37.964671: step: 280/529, loss: 0.014523983001708984 2023-01-21 12:45:39.106830: step: 284/529, loss: 0.006717491429299116 2023-01-21 12:45:40.242793: step: 288/529, loss: 0.0014739991165697575 2023-01-21 12:45:41.384948: step: 292/529, loss: 3.24249267578125e-05 2023-01-21 12:45:42.515768: step: 296/529, loss: 1.908540798467584e-05 2023-01-21 12:45:43.680664: step: 300/529, loss: 0.02439298667013645 2023-01-21 12:45:44.795056: step: 304/529, loss: 0.019907474517822266 2023-01-21 12:45:45.966632: step: 308/529, loss: 0.003631591796875 2023-01-21 12:45:47.086426: step: 312/529, loss: 0.0022954940795898438 2023-01-21 12:45:48.177009: step: 316/529, loss: 0.0010959625942632556 2023-01-21 12:45:49.300237: step: 320/529, loss: 0.00017099380784202367 2023-01-21 12:45:50.451296: step: 324/529, loss: 0.005465173628181219 2023-01-21 12:45:51.551965: step: 328/529, loss: 3.938675217796117e-05 2023-01-21 12:45:52.672236: step: 332/529, loss: 0.0010619163513183594 2023-01-21 12:45:53.801741: step: 336/529, loss: 0.00025305748567916453 2023-01-21 12:45:54.952281: step: 340/529, loss: 0.04783010482788086 2023-01-21 12:45:56.074697: step: 344/529, loss: -9.34600848268019e-06 2023-01-21 12:45:57.176671: step: 348/529, loss: 0.00017452241445425898 2023-01-21 12:45:58.317873: step: 352/529, loss: 0.00254058837890625 2023-01-21 12:45:59.450569: step: 356/529, loss: 0.00018424988957121968 2023-01-21 12:46:00.581597: step: 360/529, loss: 7.286071922862902e-05 2023-01-21 12:46:01.718684: step: 364/529, loss: 0.06492900848388672 2023-01-21 12:46:02.842682: step: 368/529, loss: 0.0008711814880371094 2023-01-21 12:46:03.965518: step: 372/529, loss: 0.024489879608154297 2023-01-21 12:46:05.109112: step: 376/529, loss: 8.850097947288305e-05 2023-01-21 12:46:06.245212: step: 380/529, loss: 0.01964578591287136 2023-01-21 12:46:07.348771: step: 384/529, loss: -1.697540210443549e-05 2023-01-21 12:46:08.492615: step: 388/529, loss: 9.384155418956652e-05 2023-01-21 12:46:09.651003: step: 392/529, loss: 0.005912399385124445 2023-01-21 12:46:10.804651: step: 396/529, loss: 0.030942726880311966 2023-01-21 12:46:11.940847: step: 400/529, loss: 0.0001846313534770161 2023-01-21 12:46:13.096690: step: 404/529, loss: 0.47950154542922974 2023-01-21 12:46:14.278360: step: 408/529, loss: 0.002935123397037387 2023-01-21 12:46:15.403544: step: 412/529, loss: 0.0002761363866738975 2023-01-21 12:46:16.514582: step: 416/529, loss: 0.019716359674930573 2023-01-21 12:46:17.664626: step: 420/529, loss: 0.025677969679236412 2023-01-21 12:46:18.819012: step: 424/529, loss: 0.16553755104541779 2023-01-21 12:46:19.960817: step: 428/529, loss: 0.003288078587502241 2023-01-21 12:46:21.073239: step: 432/529, loss: 0.0010836601722985506 2023-01-21 12:46:22.207654: step: 436/529, loss: 0.005914879031479359 2023-01-21 12:46:23.324812: step: 440/529, loss: 0.0001415252627339214 2023-01-21 12:46:24.465495: step: 444/529, loss: 0.018170548602938652 2023-01-21 12:46:25.604352: step: 448/529, loss: 0.0002191543608205393 2023-01-21 12:46:26.741859: step: 452/529, loss: 0.8585755228996277 2023-01-21 12:46:27.865208: step: 456/529, loss: 0.00054426200222224 2023-01-21 12:46:28.989143: step: 460/529, loss: 0.0030008316971361637 2023-01-21 12:46:30.111101: step: 464/529, loss: 1.411438006471144e-05 2023-01-21 12:46:31.229075: step: 468/529, loss: 5.257129487290513e-06 2023-01-21 12:46:32.379402: step: 472/529, loss: 0.014207884669303894 2023-01-21 12:46:33.549161: step: 476/529, loss: 0.016267204657197 2023-01-21 12:46:34.669808: step: 480/529, loss: 0.00076718331547454 2023-01-21 12:46:35.834451: step: 484/529, loss: 0.018243027850985527 2023-01-21 12:46:36.975595: step: 488/529, loss: 2.6702882678364404e-05 2023-01-21 12:46:38.136749: step: 492/529, loss: 0.4671728014945984 2023-01-21 12:46:39.253307: step: 496/529, loss: 0.0017086028819903731 2023-01-21 12:46:40.359400: step: 500/529, loss: 0.006110477726906538 2023-01-21 12:46:41.500587: step: 504/529, loss: 0.06197643280029297 2023-01-21 12:46:42.682494: step: 508/529, loss: 0.002379798796027899 2023-01-21 12:46:43.797107: step: 512/529, loss: 0.00022730827913619578 2023-01-21 12:46:44.966687: step: 516/529, loss: 0.0006975174183025956 2023-01-21 12:46:46.076290: step: 520/529, loss: 0.0005310773849487305 2023-01-21 12:46:47.184790: step: 524/529, loss: 0.005242586135864258 2023-01-21 12:46:48.310192: step: 528/529, loss: 0.02395658567547798 2023-01-21 12:46:49.407837: step: 532/529, loss: 0.0016062736976891756 2023-01-21 12:46:50.529687: step: 536/529, loss: 0.00571098318323493 2023-01-21 12:46:51.654055: step: 540/529, loss: 0.0012387275928631425 2023-01-21 12:46:52.802143: step: 544/529, loss: 0.11403026431798935 2023-01-21 12:46:53.941368: step: 548/529, loss: 0.0005660057067871094 2023-01-21 12:46:55.081815: step: 552/529, loss: 0.010829925537109375 2023-01-21 12:46:56.199732: step: 556/529, loss: 0.0021503448951989412 2023-01-21 12:46:57.351792: step: 560/529, loss: 0.002021694090217352 2023-01-21 12:46:58.490307: step: 564/529, loss: 8.471011824440211e-05 2023-01-21 12:46:59.662753: step: 568/529, loss: 0.008575248531997204 2023-01-21 12:47:00.777703: step: 572/529, loss: 0.0007753372774459422 2023-01-21 12:47:01.915583: step: 576/529, loss: 0.0674201026558876 2023-01-21 12:47:03.056583: step: 580/529, loss: 0.005821323487907648 2023-01-21 12:47:04.197491: step: 584/529, loss: 0.03056478500366211 2023-01-21 12:47:05.312038: step: 588/529, loss: 0.0003113746643066406 2023-01-21 12:47:06.453369: step: 592/529, loss: 0.0001746654452290386 2023-01-21 12:47:07.591283: step: 596/529, loss: 2.5320052372990176e-05 2023-01-21 12:47:08.737716: step: 600/529, loss: 0.04817180708050728 2023-01-21 12:47:09.881432: step: 604/529, loss: 0.021023273468017578 2023-01-21 12:47:11.011191: step: 608/529, loss: 0.008202552795410156 2023-01-21 12:47:12.140850: step: 612/529, loss: 0.018994808197021484 2023-01-21 12:47:13.274605: step: 616/529, loss: 0.004295063205063343 2023-01-21 12:47:14.412357: step: 620/529, loss: 0.017871476709842682 2023-01-21 12:47:15.553762: step: 624/529, loss: 0.018616868183016777 2023-01-21 12:47:16.691331: step: 628/529, loss: 0.0038326263893395662 2023-01-21 12:47:17.836278: step: 632/529, loss: 0.00198612199164927 2023-01-21 12:47:18.996151: step: 636/529, loss: 0.009025477804243565 2023-01-21 12:47:20.132696: step: 640/529, loss: 0.0007748603820800781 2023-01-21 12:47:21.273010: step: 644/529, loss: 0.005166816990822554 2023-01-21 12:47:22.431223: step: 648/529, loss: 0.001348209334537387 2023-01-21 12:47:23.574999: step: 652/529, loss: 6.160736666060984e-05 2023-01-21 12:47:24.713640: step: 656/529, loss: 0.0026733397971838713 2023-01-21 12:47:25.847757: step: 660/529, loss: 1.1634827387752011e-05 2023-01-21 12:47:26.986418: step: 664/529, loss: 5.2928924560546875e-05 2023-01-21 12:47:28.137885: step: 668/529, loss: -4.76837158203125e-06 2023-01-21 12:47:29.266768: step: 672/529, loss: 0.15119494497776031 2023-01-21 12:47:30.414852: step: 676/529, loss: 0.009927129372954369 2023-01-21 12:47:31.566514: step: 680/529, loss: 0.005011940374970436 2023-01-21 12:47:32.721042: step: 684/529, loss: 0.00017652512178756297 2023-01-21 12:47:33.841332: step: 688/529, loss: 0.002528286073356867 2023-01-21 12:47:34.971924: step: 692/529, loss: 0.0009696959750726819 2023-01-21 12:47:36.099551: step: 696/529, loss: 3.800392369157635e-05 2023-01-21 12:47:37.266161: step: 700/529, loss: 0.0009294509654864669 2023-01-21 12:47:38.402195: step: 704/529, loss: 0.01226959191262722 2023-01-21 12:47:39.560742: step: 708/529, loss: 0.006658363621681929 2023-01-21 12:47:40.678242: step: 712/529, loss: 0.057465553283691406 2023-01-21 12:47:41.802051: step: 716/529, loss: 0.0003197670157533139 2023-01-21 12:47:42.961073: step: 720/529, loss: 0.02584371529519558 2023-01-21 12:47:44.094846: step: 724/529, loss: 0.00026450157747603953 2023-01-21 12:47:45.244870: step: 728/529, loss: 0.00086383824236691 2023-01-21 12:47:46.369586: step: 732/529, loss: 0.00013346671767067164 2023-01-21 12:47:47.504053: step: 736/529, loss: 0.0002896309015341103 2023-01-21 12:47:48.664807: step: 740/529, loss: 0.03700866922736168 2023-01-21 12:47:49.773199: step: 744/529, loss: 8.02993745310232e-05 2023-01-21 12:47:50.935309: step: 748/529, loss: 0.02893218956887722 2023-01-21 12:47:52.058936: step: 752/529, loss: 0.00023608207993675023 2023-01-21 12:47:53.189035: step: 756/529, loss: 0.003627300262451172 2023-01-21 12:47:54.307418: step: 760/529, loss: 0.00026617050752975047 2023-01-21 12:47:55.421472: step: 764/529, loss: 0.000696372939273715 2023-01-21 12:47:56.550642: step: 768/529, loss: -6.628036317124497e-06 2023-01-21 12:47:57.710090: step: 772/529, loss: 0.00017399787611793727 2023-01-21 12:47:58.833522: step: 776/529, loss: 0.0005844116094522178 2023-01-21 12:47:59.948907: step: 780/529, loss: 0.5288364291191101 2023-01-21 12:48:01.068469: step: 784/529, loss: 0.001960658933967352 2023-01-21 12:48:02.187275: step: 788/529, loss: 0.00046401022700592875 2023-01-21 12:48:03.316971: step: 792/529, loss: 0.00016288757615257055 2023-01-21 12:48:04.453784: step: 796/529, loss: 0.0033720016945153475 2023-01-21 12:48:05.662269: step: 800/529, loss: 0.0031495571602135897 2023-01-21 12:48:06.786349: step: 804/529, loss: 0.012894725427031517 2023-01-21 12:48:07.931775: step: 808/529, loss: 0.004593467339873314 2023-01-21 12:48:09.049833: step: 812/529, loss: 1.5544892448815517e-05 2023-01-21 12:48:10.167013: step: 816/529, loss: 0.002072238828986883 2023-01-21 12:48:11.280885: step: 820/529, loss: 0.0011301040649414062 2023-01-21 12:48:12.407027: step: 824/529, loss: 0.0159822478890419 2023-01-21 12:48:13.517117: step: 828/529, loss: 0.02233438566327095 2023-01-21 12:48:14.638364: step: 832/529, loss: 0.005690193269401789 2023-01-21 12:48:15.773784: step: 836/529, loss: 0.0003026008780580014 2023-01-21 12:48:16.919659: step: 840/529, loss: 0.00027122499886900187 2023-01-21 12:48:18.055088: step: 844/529, loss: 0.0023713114205747843 2023-01-21 12:48:19.172028: step: 848/529, loss: 0.0025087357498705387 2023-01-21 12:48:20.295335: step: 852/529, loss: 0.01297693233937025 2023-01-21 12:48:21.406037: step: 856/529, loss: 0.0004116535419598222 2023-01-21 12:48:22.561767: step: 860/529, loss: 0.005075645633041859 2023-01-21 12:48:23.694781: step: 864/529, loss: 0.0013816833961755037 2023-01-21 12:48:24.837763: step: 868/529, loss: 0.009405974298715591 2023-01-21 12:48:25.981848: step: 872/529, loss: 0.06452655792236328 2023-01-21 12:48:27.119012: step: 876/529, loss: 6.723403657815652e-06 2023-01-21 12:48:28.269244: step: 880/529, loss: 0.00450897216796875 2023-01-21 12:48:29.391783: step: 884/529, loss: 0.02373971976339817 2023-01-21 12:48:30.501185: step: 888/529, loss: 0.04092784225940704 2023-01-21 12:48:31.625520: step: 892/529, loss: 0.008195591159164906 2023-01-21 12:48:32.742853: step: 896/529, loss: 0.3558819890022278 2023-01-21 12:48:33.903656: step: 900/529, loss: 0.00039167405338957906 2023-01-21 12:48:35.050195: step: 904/529, loss: 0.00016107558622024953 2023-01-21 12:48:36.196931: step: 908/529, loss: 0.0002172470121877268 2023-01-21 12:48:37.306515: step: 912/529, loss: 6.580352874152595e-06 2023-01-21 12:48:38.457469: step: 916/529, loss: 0.00269145960919559 2023-01-21 12:48:39.572183: step: 920/529, loss: 0.0008691787952557206 2023-01-21 12:48:40.678878: step: 924/529, loss: 0.00017051697068382055 2023-01-21 12:48:41.822660: step: 928/529, loss: 0.6793000102043152 2023-01-21 12:48:42.947914: step: 932/529, loss: 0.0004522323433775455 2023-01-21 12:48:44.113880: step: 936/529, loss: 0.000759124755859375 2023-01-21 12:48:45.299226: step: 940/529, loss: 0.0019077302422374487 2023-01-21 12:48:46.423207: step: 944/529, loss: 4.317760613048449e-05 2023-01-21 12:48:47.556532: step: 948/529, loss: 0.002985382219776511 2023-01-21 12:48:48.672751: step: 952/529, loss: 0.00032787321833893657 2023-01-21 12:48:49.809674: step: 956/529, loss: 0.011626770719885826 2023-01-21 12:48:50.946765: step: 960/529, loss: 5.1307681133039296e-05 2023-01-21 12:48:52.102594: step: 964/529, loss: 0.03523445129394531 2023-01-21 12:48:53.224390: step: 968/529, loss: 0.009836768731474876 2023-01-21 12:48:54.343064: step: 972/529, loss: 1.9073485191256623e-07 2023-01-21 12:48:55.475390: step: 976/529, loss: 0.00022172927856445312 2023-01-21 12:48:56.611606: step: 980/529, loss: 0.00026364324730820954 2023-01-21 12:48:57.788471: step: 984/529, loss: 0.000995635986328125 2023-01-21 12:48:58.952583: step: 988/529, loss: 0.0017072678310796618 2023-01-21 12:49:00.091682: step: 992/529, loss: 0.00033626556978560984 2023-01-21 12:49:01.220812: step: 996/529, loss: 0.00811758078634739 2023-01-21 12:49:02.363918: step: 1000/529, loss: 0.0019409179221838713 2023-01-21 12:49:03.523484: step: 1004/529, loss: 0.05268058925867081 2023-01-21 12:49:04.749834: step: 1008/529, loss: 0.0003059387090615928 2023-01-21 12:49:05.901765: step: 1012/529, loss: 0.047658536583185196 2023-01-21 12:49:07.028717: step: 1016/529, loss: 0.008768654428422451 2023-01-21 12:49:08.182536: step: 1020/529, loss: 0.05646724998950958 2023-01-21 12:49:09.333280: step: 1024/529, loss: -8.01086389401462e-06 2023-01-21 12:49:10.466650: step: 1028/529, loss: 0.00038971900357864797 2023-01-21 12:49:11.604527: step: 1032/529, loss: 0.009548378176987171 2023-01-21 12:49:12.724062: step: 1036/529, loss: 0.06464119255542755 2023-01-21 12:49:13.847520: step: 1040/529, loss: 0.000858783780131489 2023-01-21 12:49:14.986466: step: 1044/529, loss: 5.626678466796875e-05 2023-01-21 12:49:16.114550: step: 1048/529, loss: 0.02667827531695366 2023-01-21 12:49:17.230481: step: 1052/529, loss: 3.156662205583416e-05 2023-01-21 12:49:18.346897: step: 1056/529, loss: 0.020381735637784004 2023-01-21 12:49:19.481730: step: 1060/529, loss: 0.004177475348114967 2023-01-21 12:49:20.650733: step: 1064/529, loss: 5.817413330078125e-05 2023-01-21 12:49:21.811465: step: 1068/529, loss: 0.0014999390114098787 2023-01-21 12:49:22.939002: step: 1072/529, loss: 0.01082911528646946 2023-01-21 12:49:24.080478: step: 1076/529, loss: 1.220703143189894e-05 2023-01-21 12:49:25.208917: step: 1080/529, loss: 0.01040344312787056 2023-01-21 12:49:26.340338: step: 1084/529, loss: 0.0007085800170898438 2023-01-21 12:49:27.445191: step: 1088/529, loss: 0.0016246795421466231 2023-01-21 12:49:28.571765: step: 1092/529, loss: 0.00043020248995162547 2023-01-21 12:49:29.744067: step: 1096/529, loss: 0.0033636093139648438 2023-01-21 12:49:30.883763: step: 1100/529, loss: 0.0005567550542764366 2023-01-21 12:49:31.996859: step: 1104/529, loss: 0.11901941150426865 2023-01-21 12:49:33.123840: step: 1108/529, loss: 0.008436488918960094 2023-01-21 12:49:34.257734: step: 1112/529, loss: 9.822845640883315e-06 2023-01-21 12:49:35.409260: step: 1116/529, loss: 0.0013326645130291581 2023-01-21 12:49:36.565945: step: 1120/529, loss: 7.400512549793348e-05 2023-01-21 12:49:37.717453: step: 1124/529, loss: 7.114410982467234e-05 2023-01-21 12:49:38.827990: step: 1128/529, loss: 0.0004428386746440083 2023-01-21 12:49:39.962541: step: 1132/529, loss: 0.11768370121717453 2023-01-21 12:49:41.073243: step: 1136/529, loss: 0.0034254074562340975 2023-01-21 12:49:42.184666: step: 1140/529, loss: 0.04099302366375923 2023-01-21 12:49:43.315138: step: 1144/529, loss: 4.100799560546875e-05 2023-01-21 12:49:44.451412: step: 1148/529, loss: 0.009046555496752262 2023-01-21 12:49:45.580772: step: 1152/529, loss: 0.026387309655547142 2023-01-21 12:49:46.724081: step: 1156/529, loss: 7.324219041038305e-05 2023-01-21 12:49:47.848114: step: 1160/529, loss: 0.015204811468720436 2023-01-21 12:49:48.997029: step: 1164/529, loss: 0.07096929848194122 2023-01-21 12:49:50.120444: step: 1168/529, loss: 0.002702140947803855 2023-01-21 12:49:51.264664: step: 1172/529, loss: 5.283355858409777e-05 2023-01-21 12:49:52.423407: step: 1176/529, loss: 0.0001697540283203125 2023-01-21 12:49:53.563047: step: 1180/529, loss: 0.005737018771469593 2023-01-21 12:49:54.672870: step: 1184/529, loss: 0.0004400253528729081 2023-01-21 12:49:55.775951: step: 1188/529, loss: 0.00013427734666038305 2023-01-21 12:49:56.936229: step: 1192/529, loss: 0.0003814697265625 2023-01-21 12:49:58.066133: step: 1196/529, loss: 0.015169811435043812 2023-01-21 12:49:59.190643: step: 1200/529, loss: 0.0001317977876169607 2023-01-21 12:50:00.316317: step: 1204/529, loss: 0.0026090622413903475 2023-01-21 12:50:01.436308: step: 1208/529, loss: 0.001383018447086215 2023-01-21 12:50:02.562441: step: 1212/529, loss: 5.722044988942798e-07 2023-01-21 12:50:03.703255: step: 1216/529, loss: 0.01858210563659668 2023-01-21 12:50:04.819680: step: 1220/529, loss: 0.0005140304565429688 2023-01-21 12:50:05.952305: step: 1224/529, loss: -1.71661376953125e-05 2023-01-21 12:50:07.064160: step: 1228/529, loss: 0.0018571853870525956 2023-01-21 12:50:08.179392: step: 1232/529, loss: 0.0011067390441894531 2023-01-21 12:50:09.332148: step: 1236/529, loss: 0.0008175373659469187 2023-01-21 12:50:10.462885: step: 1240/529, loss: 0.01364278793334961 2023-01-21 12:50:11.612817: step: 1244/529, loss: 0.04728398472070694 2023-01-21 12:50:12.763151: step: 1248/529, loss: 0.005242061801254749 2023-01-21 12:50:13.904291: step: 1252/529, loss: 0.22273731231689453 2023-01-21 12:50:15.027204: step: 1256/529, loss: 0.00027217864408157766 2023-01-21 12:50:16.136924: step: 1260/529, loss: 2.2983551389188506e-05 2023-01-21 12:50:17.291779: step: 1264/529, loss: 0.0019635679200291634 2023-01-21 12:50:18.434620: step: 1268/529, loss: 0.006431865505874157 2023-01-21 12:50:19.596850: step: 1272/529, loss: 0.014790821820497513 2023-01-21 12:50:20.737818: step: 1276/529, loss: 0.0005578040727414191 2023-01-21 12:50:21.846253: step: 1280/529, loss: 0.004365062806755304 2023-01-21 12:50:22.974227: step: 1284/529, loss: 0.0003487587091512978 2023-01-21 12:50:24.113860: step: 1288/529, loss: 0.008034419268369675 2023-01-21 12:50:25.237928: step: 1292/529, loss: 0.018192673102021217 2023-01-21 12:50:26.366078: step: 1296/529, loss: 0.0025106428656727076 2023-01-21 12:50:27.519261: step: 1300/529, loss: 0.02113189734518528 2023-01-21 12:50:28.659382: step: 1304/529, loss: 0.0002846717834472656 2023-01-21 12:50:29.793678: step: 1308/529, loss: 0.17111073434352875 2023-01-21 12:50:30.919322: step: 1312/529, loss: 0.0045678140595555305 2023-01-21 12:50:32.035526: step: 1316/529, loss: 0.0003182411310262978 2023-01-21 12:50:33.161569: step: 1320/529, loss: 0.020606422796845436 2023-01-21 12:50:34.330394: step: 1324/529, loss: 0.0359928123652935 2023-01-21 12:50:35.440998: step: 1328/529, loss: 1.6069412595243193e-05 2023-01-21 12:50:36.617126: step: 1332/529, loss: 0.0014671325916424394 2023-01-21 12:50:37.769800: step: 1336/529, loss: 0.007130575366318226 2023-01-21 12:50:38.879745: step: 1340/529, loss: 2.632141149661038e-05 2023-01-21 12:50:40.035060: step: 1344/529, loss: 0.09547348320484161 2023-01-21 12:50:41.168168: step: 1348/529, loss: 0.0024971961975097656 2023-01-21 12:50:42.338421: step: 1352/529, loss: 1.697540210443549e-05 2023-01-21 12:50:43.452903: step: 1356/529, loss: 0.00036420824471861124 2023-01-21 12:50:44.599311: step: 1360/529, loss: 0.03777885437011719 2023-01-21 12:50:45.752576: step: 1364/529, loss: 0.016987131908535957 2023-01-21 12:50:46.881710: step: 1368/529, loss: 0.004500960931181908 2023-01-21 12:50:48.002470: step: 1372/529, loss: 0.011851596646010876 2023-01-21 12:50:49.130195: step: 1376/529, loss: 0.00048122406587935984 2023-01-21 12:50:50.266600: step: 1380/529, loss: 0.0009137153974734247 2023-01-21 12:50:51.396371: step: 1384/529, loss: 0.005162620916962624 2023-01-21 12:50:52.526995: step: 1388/529, loss: 9.15527380129788e-06 2023-01-21 12:50:53.660402: step: 1392/529, loss: 0.00013872981071472168 2023-01-21 12:50:54.807777: step: 1396/529, loss: 0.003924655728042126 2023-01-21 12:50:55.948860: step: 1400/529, loss: 0.0004564285045489669 2023-01-21 12:50:57.084524: step: 1404/529, loss: 0.009838689118623734 2023-01-21 12:50:58.207279: step: 1408/529, loss: 0.0010721206199377775 2023-01-21 12:50:59.327185: step: 1412/529, loss: 0.039681438356637955 2023-01-21 12:51:00.479240: step: 1416/529, loss: 1.258850079466356e-05 2023-01-21 12:51:01.580397: step: 1420/529, loss: 0.0007408141973428428 2023-01-21 12:51:02.742424: step: 1424/529, loss: 0.0016587257850915194 2023-01-21 12:51:03.872139: step: 1428/529, loss: 0.005351162049919367 2023-01-21 12:51:05.052967: step: 1432/529, loss: 1.2006261348724365 2023-01-21 12:51:06.187168: step: 1436/529, loss: 0.03083324432373047 2023-01-21 12:51:07.303553: step: 1440/529, loss: 0.014517402276396751 2023-01-21 12:51:08.422934: step: 1444/529, loss: 2.9659271604032256e-05 2023-01-21 12:51:09.583185: step: 1448/529, loss: 0.011948967352509499 2023-01-21 12:51:10.698929: step: 1452/529, loss: 0.0023942948319017887 2023-01-21 12:51:11.869117: step: 1456/529, loss: 0.00017585755267646164 2023-01-21 12:51:12.982958: step: 1460/529, loss: 0.0011234760750085115 2023-01-21 12:51:14.092517: step: 1464/529, loss: 0.00020284653874114156 2023-01-21 12:51:15.255198: step: 1468/529, loss: 5.549192428588867e-05 2023-01-21 12:51:16.405872: step: 1472/529, loss: -2.2411345526052173e-06 2023-01-21 12:51:17.529789: step: 1476/529, loss: 0.010947560891509056 2023-01-21 12:51:18.641533: step: 1480/529, loss: 0.02134866639971733 2023-01-21 12:51:19.786041: step: 1484/529, loss: 0.04446563497185707 2023-01-21 12:51:20.911814: step: 1488/529, loss: 6.542205665027723e-05 2023-01-21 12:51:22.056941: step: 1492/529, loss: 0.015706205740571022 2023-01-21 12:51:23.209252: step: 1496/529, loss: 0.2147197723388672 2023-01-21 12:51:24.339124: step: 1500/529, loss: 0.005993532948195934 2023-01-21 12:51:25.473824: step: 1504/529, loss: 0.0005027771112509072 2023-01-21 12:51:26.607056: step: 1508/529, loss: 5.4168704082258046e-05 2023-01-21 12:51:27.723016: step: 1512/529, loss: 7.581710815429688e-05 2023-01-21 12:51:28.893662: step: 1516/529, loss: 2.174377368646674e-05 2023-01-21 12:51:30.014806: step: 1520/529, loss: 0.0012263774406164885 2023-01-21 12:51:31.163899: step: 1524/529, loss: 0.00014729499525856227 2023-01-21 12:51:32.301442: step: 1528/529, loss: 0.00019006729417014867 2023-01-21 12:51:33.465636: step: 1532/529, loss: 0.013119888491928577 2023-01-21 12:51:34.573862: step: 1536/529, loss: 0.0004646301385946572 2023-01-21 12:51:35.719918: step: 1540/529, loss: 0.009221267886459827 2023-01-21 12:51:36.849719: step: 1544/529, loss: 0.0016561507945880294 2023-01-21 12:51:37.971009: step: 1548/529, loss: 0.007258796598762274 2023-01-21 12:51:39.145418: step: 1552/529, loss: 0.006394767668098211 2023-01-21 12:51:40.311598: step: 1556/529, loss: 0.0008476257207803428 2023-01-21 12:51:41.443841: step: 1560/529, loss: 0.6624734997749329 2023-01-21 12:51:42.560079: step: 1564/529, loss: 0.014646196737885475 2023-01-21 12:51:43.709799: step: 1568/529, loss: 0.6264400482177734 2023-01-21 12:51:44.851616: step: 1572/529, loss: 0.004799270536750555 2023-01-21 12:51:45.995358: step: 1576/529, loss: 0.03553476184606552 2023-01-21 12:51:47.171689: step: 1580/529, loss: 0.0050566671416163445 2023-01-21 12:51:48.281199: step: 1584/529, loss: 0.007789802737534046 2023-01-21 12:51:49.404113: step: 1588/529, loss: 0.0031312943901866674 2023-01-21 12:51:50.561462: step: 1592/529, loss: 0.027535725384950638 2023-01-21 12:51:51.701755: step: 1596/529, loss: 0.00347480783239007 2023-01-21 12:51:52.861067: step: 1600/529, loss: 0.01957111246883869 2023-01-21 12:51:54.000755: step: 1604/529, loss: 0.005470586009323597 2023-01-21 12:51:55.115717: step: 1608/529, loss: 0.0021026614122092724 2023-01-21 12:51:56.267254: step: 1612/529, loss: 0.01912364922463894 2023-01-21 12:51:57.396731: step: 1616/529, loss: 0.0017991543281823397 2023-01-21 12:51:58.545470: step: 1620/529, loss: 0.0031447410583496094 2023-01-21 12:51:59.672539: step: 1624/529, loss: 5.459785461425781e-05 2023-01-21 12:52:00.819039: step: 1628/529, loss: 0.00010395050048828125 2023-01-21 12:52:01.950360: step: 1632/529, loss: 0.0017313957214355469 2023-01-21 12:52:03.077439: step: 1636/529, loss: 1.2389501333236694 2023-01-21 12:52:04.239385: step: 1640/529, loss: 0.00022821426682639867 2023-01-21 12:52:05.361496: step: 1644/529, loss: 0.000545501708984375 2023-01-21 12:52:06.484911: step: 1648/529, loss: 0.04394121095538139 2023-01-21 12:52:07.654709: step: 1652/529, loss: 0.002732658525928855 2023-01-21 12:52:08.775242: step: 1656/529, loss: 0.0035509110894054174 2023-01-21 12:52:09.900911: step: 1660/529, loss: 2.765655608527595e-06 2023-01-21 12:52:11.035341: step: 1664/529, loss: 0.0003696441708598286 2023-01-21 12:52:12.187434: step: 1668/529, loss: 0.0013166428543627262 2023-01-21 12:52:13.309143: step: 1672/529, loss: 6.189346458995715e-05 2023-01-21 12:52:14.429588: step: 1676/529, loss: 0.00555343646556139 2023-01-21 12:52:15.574422: step: 1680/529, loss: 4.634857032215223e-05 2023-01-21 12:52:16.718738: step: 1684/529, loss: 0.00014719963655807078 2023-01-21 12:52:17.855249: step: 1688/529, loss: 0.024477005004882812 2023-01-21 12:52:18.976086: step: 1692/529, loss: 0.03810019791126251 2023-01-21 12:52:20.106786: step: 1696/529, loss: 0.0006124496576376259 2023-01-21 12:52:21.212334: step: 1700/529, loss: 0.0006933212280273438 2023-01-21 12:52:22.333793: step: 1704/529, loss: 0.06569556891918182 2023-01-21 12:52:23.474995: step: 1708/529, loss: 0.00969705544412136 2023-01-21 12:52:24.581193: step: 1712/529, loss: 0.00046415330143645406 2023-01-21 12:52:25.708855: step: 1716/529, loss: 0.024320030584931374 2023-01-21 12:52:26.863754: step: 1720/529, loss: 0.01116476021707058 2023-01-21 12:52:27.993353: step: 1724/529, loss: 0.027220536023378372 2023-01-21 12:52:29.127586: step: 1728/529, loss: 0.0036932947114109993 2023-01-21 12:52:30.264620: step: 1732/529, loss: 2.913475145760458e-05 2023-01-21 12:52:31.368062: step: 1736/529, loss: 0.0012719155056402087 2023-01-21 12:52:32.491105: step: 1740/529, loss: 0.022940635681152344 2023-01-21 12:52:33.602947: step: 1744/529, loss: 0.00013027191744185984 2023-01-21 12:52:34.759068: step: 1748/529, loss: 0.025810621678829193 2023-01-21 12:52:35.875172: step: 1752/529, loss: 0.0021206140518188477 2023-01-21 12:52:37.026143: step: 1756/529, loss: 0.0001560211239848286 2023-01-21 12:52:38.127053: step: 1760/529, loss: 9.613037400413305e-05 2023-01-21 12:52:39.271638: step: 1764/529, loss: 0.0020975114312022924 2023-01-21 12:52:40.402467: step: 1768/529, loss: 0.0005016326904296875 2023-01-21 12:52:41.545940: step: 1772/529, loss: 0.00112152099609375 2023-01-21 12:52:42.675562: step: 1776/529, loss: 0.00016994477482512593 2023-01-21 12:52:43.849695: step: 1780/529, loss: 0.024969100952148438 2023-01-21 12:52:44.993181: step: 1784/529, loss: 0.0065469746477901936 2023-01-21 12:52:46.156461: step: 1788/529, loss: 0.04405250772833824 2023-01-21 12:52:47.261987: step: 1792/529, loss: 0.0012364387512207031 2023-01-21 12:52:48.413350: step: 1796/529, loss: 7.514953904319555e-05 2023-01-21 12:52:49.526893: step: 1800/529, loss: 6.246566772460938e-05 2023-01-21 12:52:50.685848: step: 1804/529, loss: 0.05991210788488388 2023-01-21 12:52:51.831235: step: 1808/529, loss: 0.00048513413639739156 2023-01-21 12:52:52.962274: step: 1812/529, loss: 0.07357731461524963 2023-01-21 12:52:54.097511: step: 1816/529, loss: 0.013456725515425205 2023-01-21 12:52:55.234815: step: 1820/529, loss: 1.52587890625e-05 2023-01-21 12:52:56.391296: step: 1824/529, loss: 0.00155725481454283 2023-01-21 12:52:57.517697: step: 1828/529, loss: 0.02734222449362278 2023-01-21 12:52:58.656732: step: 1832/529, loss: 0.02555208094418049 2023-01-21 12:52:59.801735: step: 1836/529, loss: 0.00018491744413040578 2023-01-21 12:53:00.932150: step: 1840/529, loss: 0.00638313265517354 2023-01-21 12:53:02.073208: step: 1844/529, loss: 0.00017032623873092234 2023-01-21 12:53:03.204946: step: 1848/529, loss: 0.0005761146894656122 2023-01-21 12:53:04.366479: step: 1852/529, loss: 0.0024094581604003906 2023-01-21 12:53:05.505349: step: 1856/529, loss: 0.001040983246639371 2023-01-21 12:53:06.655412: step: 1860/529, loss: 0.00175647740252316 2023-01-21 12:53:07.807746: step: 1864/529, loss: 0.005059623625129461 2023-01-21 12:53:08.970462: step: 1868/529, loss: 0.04588184133172035 2023-01-21 12:53:10.106582: step: 1872/529, loss: 0.004602909088134766 2023-01-21 12:53:11.220446: step: 1876/529, loss: 0.012368394061923027 2023-01-21 12:53:12.370999: step: 1880/529, loss: 0.007129574194550514 2023-01-21 12:53:13.489131: step: 1884/529, loss: 0.02068462409079075 2023-01-21 12:53:14.600514: step: 1888/529, loss: 0.005769253242760897 2023-01-21 12:53:15.749253: step: 1892/529, loss: 0.0006010055658407509 2023-01-21 12:53:16.854044: step: 1896/529, loss: 0.016878414899110794 2023-01-21 12:53:18.053997: step: 1900/529, loss: 0.018114661797881126 2023-01-21 12:53:19.199357: step: 1904/529, loss: 0.0034267425071448088 2023-01-21 12:53:20.343571: step: 1908/529, loss: 0.0016595841152593493 2023-01-21 12:53:21.451044: step: 1912/529, loss: 0.00011882781836902723 2023-01-21 12:53:22.588575: step: 1916/529, loss: 0.0007312774541787803 2023-01-21 12:53:23.704158: step: 1920/529, loss: 0.00044541360693983734 2023-01-21 12:53:24.840327: step: 1924/529, loss: 5.197525297262473e-06 2023-01-21 12:53:26.003747: step: 1928/529, loss: -1.1920928955078125e-05 2023-01-21 12:53:27.143391: step: 1932/529, loss: 0.031696319580078125 2023-01-21 12:53:28.301274: step: 1936/529, loss: 0.00727801350876689 2023-01-21 12:53:29.441868: step: 1940/529, loss: 0.0265673641115427 2023-01-21 12:53:30.598735: step: 1944/529, loss: 0.0005628585931845009 2023-01-21 12:53:31.744151: step: 1948/529, loss: 1.4626979464082979e-05 2023-01-21 12:53:32.892676: step: 1952/529, loss: 0.0024222375359386206 2023-01-21 12:53:34.029856: step: 1956/529, loss: 0.0003109932004008442 2023-01-21 12:53:35.178140: step: 1960/529, loss: 0.05009441450238228 2023-01-21 12:53:36.298734: step: 1964/529, loss: 0.0002948761102743447 2023-01-21 12:53:37.415358: step: 1968/529, loss: 0.00027446745662018657 2023-01-21 12:53:38.574932: step: 1972/529, loss: 0.0025302888825535774 2023-01-21 12:53:39.723030: step: 1976/529, loss: 0.0003734588681254536 2023-01-21 12:53:40.868724: step: 1980/529, loss: 0.0026082515250891447 2023-01-21 12:53:41.995317: step: 1984/529, loss: 1.506805438111769e-05 2023-01-21 12:53:43.102740: step: 1988/529, loss: 0.004876136779785156 2023-01-21 12:53:44.239338: step: 1992/529, loss: 0.05193672329187393 2023-01-21 12:53:45.415447: step: 1996/529, loss: 4.901886131847277e-05 2023-01-21 12:53:46.529715: step: 2000/529, loss: 0.004590416327118874 2023-01-21 12:53:47.667256: step: 2004/529, loss: 0.0017933845520019531 2023-01-21 12:53:48.784354: step: 2008/529, loss: 1.3351440202313825e-06 2023-01-21 12:53:49.909097: step: 2012/529, loss: 0.00017604828462935984 2023-01-21 12:53:51.058700: step: 2016/529, loss: 0.00034208298893645406 2023-01-21 12:53:52.231151: step: 2020/529, loss: 3.147125244140625e-05 2023-01-21 12:53:53.351453: step: 2024/529, loss: 0.000743865966796875 2023-01-21 12:53:54.484583: step: 2028/529, loss: 0.09983968734741211 2023-01-21 12:53:55.626852: step: 2032/529, loss: 0.006250858306884766 2023-01-21 12:53:56.803315: step: 2036/529, loss: 0.0001996994105866179 2023-01-21 12:53:57.952110: step: 2040/529, loss: -1.5497207641601562e-05 2023-01-21 12:53:59.092227: step: 2044/529, loss: 0.01919403113424778 2023-01-21 12:54:00.222221: step: 2048/529, loss: 0.00012168884859420359 2023-01-21 12:54:01.371644: step: 2052/529, loss: 0.00235824566334486 2023-01-21 12:54:02.512593: step: 2056/529, loss: 5.4359438763640355e-06 2023-01-21 12:54:03.655869: step: 2060/529, loss: 0.0008721351623535156 2023-01-21 12:54:04.779489: step: 2064/529, loss: 0.00018768310837913305 2023-01-21 12:54:05.892028: step: 2068/529, loss: 0.0955018624663353 2023-01-21 12:54:07.026126: step: 2072/529, loss: 0.006647014524787664 2023-01-21 12:54:08.135260: step: 2076/529, loss: 0.003270912216976285 2023-01-21 12:54:09.233524: step: 2080/529, loss: 0.008613205514848232 2023-01-21 12:54:10.380016: step: 2084/529, loss: 0.040334321558475494 2023-01-21 12:54:11.496185: step: 2088/529, loss: 0.02104778215289116 2023-01-21 12:54:12.682670: step: 2092/529, loss: 0.03226776421070099 2023-01-21 12:54:13.808570: step: 2096/529, loss: 0.03032503090798855 2023-01-21 12:54:14.993513: step: 2100/529, loss: 0.0033143998589366674 2023-01-21 12:54:16.112080: step: 2104/529, loss: 0.0003887176571879536 2023-01-21 12:54:17.270364: step: 2108/529, loss: 0.0028007030487060547 2023-01-21 12:54:18.422067: step: 2112/529, loss: 3.2138825190486386e-05 2023-01-21 12:54:19.590787: step: 2116/529, loss: 0.0011140824062749743 ================================================== Loss: 0.025 -------------------- Dev: {'event': {'p': 0.6008230452674898, 'r': 0.7776298268974701, 'f1': 0.677887405687754}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Test: {'event': {'p': 0.614115490375802, 'r': 0.776361529548088, 'f1': 0.6857727737973388}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Chinese: {'event': {'p': 0.5647058823529412, 'r': 0.8888888888888888, 'f1': 0.6906474820143885}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Korean: {'event': {'p': 0.6290322580645161, 'r': 0.6190476190476191, 'f1': 0.6240000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Russian: {'event': {'p': 0.40425531914893614, 'r': 0.5277777777777778, 'f1': 0.45783132530120485}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 25 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:55:00.580486: step: 4/529, loss: 0.001013946603052318 2023-01-21 12:55:01.778148: step: 8/529, loss: 0.0014806747203692794 2023-01-21 12:55:02.911920: step: 12/529, loss: 1.0967254638671875e-05 2023-01-21 12:55:04.063702: step: 16/529, loss: 0.04911251366138458 2023-01-21 12:55:05.186619: step: 20/529, loss: 0.010462189093232155 2023-01-21 12:55:06.338098: step: 24/529, loss: 0.0006329536554403603 2023-01-21 12:55:07.451889: step: 28/529, loss: 0.0020308971870690584 2023-01-21 12:55:08.585201: step: 32/529, loss: 0.0007181167602539062 2023-01-21 12:55:09.770798: step: 36/529, loss: 0.000355434458469972 2023-01-21 12:55:10.913643: step: 40/529, loss: 0.01822490617632866 2023-01-21 12:55:12.040930: step: 44/529, loss: 0.004099845886230469 2023-01-21 12:55:13.160843: step: 48/529, loss: 5.05924217577558e-05 2023-01-21 12:55:14.296872: step: 52/529, loss: 0.0006593704456463456 2023-01-21 12:55:15.423418: step: 56/529, loss: 0.0023607255425304174 2023-01-21 12:55:16.581387: step: 60/529, loss: 0.00033893584623001516 2023-01-21 12:55:17.685786: step: 64/529, loss: 0.0007385254139080644 2023-01-21 12:55:18.832869: step: 68/529, loss: 0.00047130585880950093 2023-01-21 12:55:19.949863: step: 72/529, loss: 0.00011682510375976562 2023-01-21 12:55:21.106617: step: 76/529, loss: 4.9352649512002245e-05 2023-01-21 12:55:22.251787: step: 80/529, loss: 0.035765647888183594 2023-01-21 12:55:23.396621: step: 84/529, loss: 0.0018444061279296875 2023-01-21 12:55:24.533017: step: 88/529, loss: 0.0009490966913290322 2023-01-21 12:55:25.666387: step: 92/529, loss: 0.17182083427906036 2023-01-21 12:55:26.810013: step: 96/529, loss: 0.0010110378498211503 2023-01-21 12:55:27.942774: step: 100/529, loss: 0.00028457643929868937 2023-01-21 12:55:29.077938: step: 104/529, loss: 0.05334491655230522 2023-01-21 12:55:30.209872: step: 108/529, loss: 0.0011515617370605469 2023-01-21 12:55:31.324202: step: 112/529, loss: 0.023218251764774323 2023-01-21 12:55:32.469720: step: 116/529, loss: 0.03040294721722603 2023-01-21 12:55:33.597699: step: 120/529, loss: 0.0005878448719158769 2023-01-21 12:55:34.717274: step: 124/529, loss: 0.0004543304385151714 2023-01-21 12:55:35.836017: step: 128/529, loss: 0.001139402505941689 2023-01-21 12:55:36.984835: step: 132/529, loss: 0.0005869865417480469 2023-01-21 12:55:38.105793: step: 136/529, loss: 0.0009716033819131553 2023-01-21 12:55:39.219083: step: 140/529, loss: 0.01589231565594673 2023-01-21 12:55:40.353416: step: 144/529, loss: 0.00355033902451396 2023-01-21 12:55:41.502490: step: 148/529, loss: 0.005140400025993586 2023-01-21 12:55:42.636235: step: 152/529, loss: 3.776550147449598e-05 2023-01-21 12:55:43.763254: step: 156/529, loss: 0.0007276535616256297 2023-01-21 12:55:44.916738: step: 160/529, loss: 0.025597669184207916 2023-01-21 12:55:46.038548: step: 164/529, loss: 0.0008609771612100303 2023-01-21 12:55:47.156124: step: 168/529, loss: -1.5258789289873675e-06 2023-01-21 12:55:48.275055: step: 172/529, loss: 0.0009891510708257556 2023-01-21 12:55:49.424706: step: 176/529, loss: 3.376007225597277e-05 2023-01-21 12:55:50.566271: step: 180/529, loss: 0.00047426222590729594 2023-01-21 12:55:51.697943: step: 184/529, loss: 0.0013746261829510331 2023-01-21 12:55:52.833617: step: 188/529, loss: 1.296997106692288e-05 2023-01-21 12:55:54.003979: step: 192/529, loss: 0.019568586722016335 2023-01-21 12:55:55.180386: step: 196/529, loss: 0.01928405836224556 2023-01-21 12:55:56.297415: step: 200/529, loss: 5.6743621826171875e-05 2023-01-21 12:55:57.455484: step: 204/529, loss: 1.2991780042648315 2023-01-21 12:55:58.619560: step: 208/529, loss: 0.02201652340590954 2023-01-21 12:55:59.796931: step: 212/529, loss: 0.01048126257956028 2023-01-21 12:56:00.936199: step: 216/529, loss: 0.0033668517135083675 2023-01-21 12:56:02.018922: step: 220/529, loss: 0.0005156517145223916 2023-01-21 12:56:03.161149: step: 224/529, loss: 0.0027120590675622225 2023-01-21 12:56:04.304997: step: 228/529, loss: 0.07571563869714737 2023-01-21 12:56:05.441603: step: 232/529, loss: 0.001874733017757535 2023-01-21 12:56:06.612952: step: 236/529, loss: 0.0017540991539135575 2023-01-21 12:56:07.756287: step: 240/529, loss: 0.1379198133945465 2023-01-21 12:56:08.903088: step: 244/529, loss: 0.0015982151962816715 2023-01-21 12:56:10.031953: step: 248/529, loss: 0.0008983612060546875 2023-01-21 12:56:11.151941: step: 252/529, loss: 0.012747693806886673 2023-01-21 12:56:12.311398: step: 256/529, loss: 0.06093959882855415 2023-01-21 12:56:13.427506: step: 260/529, loss: 1.6450882185381488e-06 2023-01-21 12:56:14.579221: step: 264/529, loss: 0.0004726409970317036 2023-01-21 12:56:15.719009: step: 268/529, loss: 0.04040488973259926 2023-01-21 12:56:16.858726: step: 272/529, loss: 0.0008518219110555947 2023-01-21 12:56:17.992233: step: 276/529, loss: 0.06798019260168076 2023-01-21 12:56:19.125685: step: 280/529, loss: 4.382133556646295e-05 2023-01-21 12:56:20.217813: step: 284/529, loss: 0.0037672999314963818 2023-01-21 12:56:21.369199: step: 288/529, loss: 0.0003046989440917969 2023-01-21 12:56:22.480297: step: 292/529, loss: 0.001876449678093195 2023-01-21 12:56:23.658168: step: 296/529, loss: 0.001920413924381137 2023-01-21 12:56:24.806772: step: 300/529, loss: 0.00024356841458939016 2023-01-21 12:56:25.922537: step: 304/529, loss: 3.051757857974735e-06 2023-01-21 12:56:27.034464: step: 308/529, loss: 0.014220619574189186 2023-01-21 12:56:28.170763: step: 312/529, loss: 0.026253510266542435 2023-01-21 12:56:29.304316: step: 316/529, loss: 0.0007222175481729209 2023-01-21 12:56:30.437315: step: 320/529, loss: 0.017185306176543236 2023-01-21 12:56:31.562397: step: 324/529, loss: 1.8596649169921875e-05 2023-01-21 12:56:32.671837: step: 328/529, loss: 4.415512375999242e-05 2023-01-21 12:56:33.800098: step: 332/529, loss: 0.00037369728670455515 2023-01-21 12:56:34.935544: step: 336/529, loss: 0.000602674437686801 2023-01-21 12:56:36.078562: step: 340/529, loss: 0.00016145706467796117 2023-01-21 12:56:37.178799: step: 344/529, loss: 0.009807157330214977 2023-01-21 12:56:38.314233: step: 348/529, loss: 0.0002236366126453504 2023-01-21 12:56:39.454782: step: 352/529, loss: 7.152557373046875e-05 2023-01-21 12:56:40.611945: step: 356/529, loss: 0.0069252969697117805 2023-01-21 12:56:41.743301: step: 360/529, loss: 0.003307914827018976 2023-01-21 12:56:42.904420: step: 364/529, loss: 0.0007120132795535028 2023-01-21 12:56:44.035944: step: 368/529, loss: 0.0014183997409418225 2023-01-21 12:56:45.175039: step: 372/529, loss: 0.01787404902279377 2023-01-21 12:56:46.320982: step: 376/529, loss: 0.0011308669345453382 2023-01-21 12:56:47.459664: step: 380/529, loss: 0.01737823523581028 2023-01-21 12:56:48.600071: step: 384/529, loss: 0.00014543533325195312 2023-01-21 12:56:49.745704: step: 388/529, loss: 0.015139603987336159 2023-01-21 12:56:50.884074: step: 392/529, loss: 0.0008463859558105469 2023-01-21 12:56:52.020169: step: 396/529, loss: 0.0147552490234375 2023-01-21 12:56:53.155882: step: 400/529, loss: 0.00144872663076967 2023-01-21 12:56:54.292810: step: 404/529, loss: 6.0081479205109645e-06 2023-01-21 12:56:55.439871: step: 408/529, loss: 5.8841709687840194e-05 2023-01-21 12:56:56.562966: step: 412/529, loss: 0.023936079815030098 2023-01-21 12:56:57.660850: step: 416/529, loss: 0.00039892198401503265 2023-01-21 12:56:58.794835: step: 420/529, loss: 0.00295085902325809 2023-01-21 12:56:59.970067: step: 424/529, loss: 0.002486991696059704 2023-01-21 12:57:01.138079: step: 428/529, loss: 0.0019152641762048006 2023-01-21 12:57:02.280571: step: 432/529, loss: 0.0036919116973876953 2023-01-21 12:57:03.421931: step: 436/529, loss: 0.0011075973743572831 2023-01-21 12:57:04.569676: step: 440/529, loss: 0.00022335053654387593 2023-01-21 12:57:05.712579: step: 444/529, loss: 0.028205109760165215 2023-01-21 12:57:06.860596: step: 448/529, loss: 0.0006905078771524131 2023-01-21 12:57:08.019684: step: 452/529, loss: 0.002415561582893133 2023-01-21 12:57:09.143595: step: 456/529, loss: 0.000102996826171875 2023-01-21 12:57:10.274537: step: 460/529, loss: 3.62396240234375e-05 2023-01-21 12:57:11.416589: step: 464/529, loss: 0.004718208219856024 2023-01-21 12:57:12.557338: step: 468/529, loss: 0.004766655154526234 2023-01-21 12:57:13.675428: step: 472/529, loss: 0.00042591095552779734 2023-01-21 12:57:14.843551: step: 476/529, loss: 0.016556359827518463 2023-01-21 12:57:15.994297: step: 480/529, loss: 0.021502207964658737 2023-01-21 12:57:17.136900: step: 484/529, loss: 0.008498096838593483 2023-01-21 12:57:18.270523: step: 488/529, loss: 0.031981468200683594 2023-01-21 12:57:19.390119: step: 492/529, loss: 0.0004113197501283139 2023-01-21 12:57:20.509004: step: 496/529, loss: 0.004234504885971546 2023-01-21 12:57:21.629353: step: 500/529, loss: 7.905960228526965e-05 2023-01-21 12:57:22.765579: step: 504/529, loss: 0.002686119172722101 2023-01-21 12:57:23.890807: step: 508/529, loss: 0.0034885406494140625 2023-01-21 12:57:25.018939: step: 512/529, loss: 0.20905399322509766 2023-01-21 12:57:26.133360: step: 516/529, loss: 0.006168842315673828 2023-01-21 12:57:27.284525: step: 520/529, loss: 0.0006644249078817666 2023-01-21 12:57:28.412897: step: 524/529, loss: 0.019455863162875175 2023-01-21 12:57:29.555570: step: 528/529, loss: 0.014922523871064186 2023-01-21 12:57:30.677274: step: 532/529, loss: 0.002437591552734375 2023-01-21 12:57:31.788214: step: 536/529, loss: 0.01637725718319416 2023-01-21 12:57:32.916390: step: 540/529, loss: 9.946823411155492e-05 2023-01-21 12:57:34.048168: step: 544/529, loss: 4.5239925384521484e-05 2023-01-21 12:57:35.212894: step: 548/529, loss: 1.23977656585339e-06 2023-01-21 12:57:36.405732: step: 552/529, loss: 6.752014451194555e-05 2023-01-21 12:57:37.559011: step: 556/529, loss: 0.003042602678760886 2023-01-21 12:57:38.710086: step: 560/529, loss: 0.0006097793811932206 2023-01-21 12:57:39.848197: step: 564/529, loss: 0.004331398289650679 2023-01-21 12:57:40.987294: step: 568/529, loss: 0.0007318496936932206 2023-01-21 12:57:42.137272: step: 572/529, loss: 0.00024309157743118703 2023-01-21 12:57:43.275020: step: 576/529, loss: 0.05574455112218857 2023-01-21 12:57:44.403443: step: 580/529, loss: 8.726119995117188e-05 2023-01-21 12:57:45.533434: step: 584/529, loss: 9.241104271495715e-05 2023-01-21 12:57:46.670762: step: 588/529, loss: 2.0599367417162284e-05 2023-01-21 12:57:47.808607: step: 592/529, loss: 0.0001791000395314768 2023-01-21 12:57:48.989386: step: 596/529, loss: 0.00029177666874602437 2023-01-21 12:57:50.118736: step: 600/529, loss: 0.007574653718620539 2023-01-21 12:57:51.268758: step: 604/529, loss: 0.00065784459002316 2023-01-21 12:57:52.381249: step: 608/529, loss: 0.00047807692317292094 2023-01-21 12:57:53.541227: step: 612/529, loss: 0.0003779649850912392 2023-01-21 12:57:54.661114: step: 616/529, loss: 6.599426706088707e-05 2023-01-21 12:57:55.796105: step: 620/529, loss: 0.02099933661520481 2023-01-21 12:57:56.932518: step: 624/529, loss: 0.00035381317138671875 2023-01-21 12:57:58.064132: step: 628/529, loss: 0.0005969524499960244 2023-01-21 12:57:59.183929: step: 632/529, loss: 0.0008316040621139109 2023-01-21 12:58:00.333646: step: 636/529, loss: 0.0018774032359942794 2023-01-21 12:58:01.469068: step: 640/529, loss: 0.029640579596161842 2023-01-21 12:58:02.591924: step: 644/529, loss: 0.5012916326522827 2023-01-21 12:58:03.719376: step: 648/529, loss: 3.509521411615424e-05 2023-01-21 12:58:04.810228: step: 652/529, loss: 0.003305673599243164 2023-01-21 12:58:05.964877: step: 656/529, loss: 0.026328660547733307 2023-01-21 12:58:07.080715: step: 660/529, loss: 0.026063919067382812 2023-01-21 12:58:08.233230: step: 664/529, loss: 0.00020313262939453125 2023-01-21 12:58:09.384324: step: 668/529, loss: 0.0005033016204833984 2023-01-21 12:58:10.522089: step: 672/529, loss: 0.03033750131726265 2023-01-21 12:58:11.668841: step: 676/529, loss: 0.022629834711551666 2023-01-21 12:58:12.813962: step: 680/529, loss: 0.0004142284451518208 2023-01-21 12:58:13.954590: step: 684/529, loss: 0.002850341610610485 2023-01-21 12:58:15.116694: step: 688/529, loss: 0.0006895065307617188 2023-01-21 12:58:16.269717: step: 692/529, loss: 0.004275131039321423 2023-01-21 12:58:17.386625: step: 696/529, loss: 5.53131121705519e-06 2023-01-21 12:58:18.559464: step: 700/529, loss: 0.015840720385313034 2023-01-21 12:58:19.686194: step: 704/529, loss: 0.0015626907115802169 2023-01-21 12:58:20.817916: step: 708/529, loss: 4.4059750507585704e-05 2023-01-21 12:58:21.967852: step: 712/529, loss: 0.0018787384033203125 2023-01-21 12:58:23.097638: step: 716/529, loss: 0.014497852884232998 2023-01-21 12:58:24.226284: step: 720/529, loss: 0.00018768310837913305 2023-01-21 12:58:25.361258: step: 724/529, loss: 2.86102294921875e-06 2023-01-21 12:58:26.500821: step: 728/529, loss: 0.0003380775451660156 2023-01-21 12:58:27.663290: step: 732/529, loss: -4.873275611316785e-05 2023-01-21 12:58:28.798201: step: 736/529, loss: 0.012883090414106846 2023-01-21 12:58:29.928801: step: 740/529, loss: 0.029575729742646217 2023-01-21 12:58:31.080883: step: 744/529, loss: 0.009055329486727715 2023-01-21 12:58:32.191077: step: 748/529, loss: 0.003642844967544079 2023-01-21 12:58:33.358672: step: 752/529, loss: 0.0011959075927734375 2023-01-21 12:58:34.504417: step: 756/529, loss: 0.0010707855690270662 2023-01-21 12:58:35.628784: step: 760/529, loss: 0.011237716302275658 2023-01-21 12:58:36.808629: step: 764/529, loss: 0.023442650213837624 2023-01-21 12:58:37.942805: step: 768/529, loss: 0.0003484726184979081 2023-01-21 12:58:39.066918: step: 772/529, loss: 0.00029983522836118937 2023-01-21 12:58:40.227062: step: 776/529, loss: 2.7847288947668858e-05 2023-01-21 12:58:41.370409: step: 780/529, loss: 0.002143478486686945 2023-01-21 12:58:42.474907: step: 784/529, loss: 0.029391765594482422 2023-01-21 12:58:43.627596: step: 788/529, loss: 2.5606155759305693e-05 2023-01-21 12:58:44.781564: step: 792/529, loss: 0.0021522522438317537 2023-01-21 12:58:45.914784: step: 796/529, loss: 0.00026187897310592234 2023-01-21 12:58:47.056301: step: 800/529, loss: 0.0777493491768837 2023-01-21 12:58:48.206028: step: 804/529, loss: 0.010334014892578125 2023-01-21 12:58:49.324521: step: 808/529, loss: 0.0003005981561727822 2023-01-21 12:58:50.464989: step: 812/529, loss: 0.005246162414550781 2023-01-21 12:58:51.605816: step: 816/529, loss: 6.923676119185984e-05 2023-01-21 12:58:52.736770: step: 820/529, loss: 0.030755234882235527 2023-01-21 12:58:53.868380: step: 824/529, loss: 0.025723746046423912 2023-01-21 12:58:55.010975: step: 828/529, loss: 0.4868951737880707 2023-01-21 12:58:56.132453: step: 832/529, loss: 0.07975530624389648 2023-01-21 12:58:57.277954: step: 836/529, loss: 5.7697296142578125e-05 2023-01-21 12:58:58.418767: step: 840/529, loss: 0.00030241013155318797 2023-01-21 12:58:59.551878: step: 844/529, loss: 0.0010480880737304688 2023-01-21 12:59:00.720850: step: 848/529, loss: 0.00161323556676507 2023-01-21 12:59:01.875772: step: 852/529, loss: 0.0004894256708212197 2023-01-21 12:59:02.991548: step: 856/529, loss: 0.0001163482666015625 2023-01-21 12:59:04.130829: step: 860/529, loss: 2.0408631826285273e-05 2023-01-21 12:59:05.276119: step: 864/529, loss: 2.28881845032447e-06 2023-01-21 12:59:06.396668: step: 868/529, loss: 1.0967254638671875e-05 2023-01-21 12:59:07.557330: step: 872/529, loss: 0.02040095254778862 2023-01-21 12:59:08.685685: step: 876/529, loss: 0.0025188445579260588 2023-01-21 12:59:09.819017: step: 880/529, loss: 0.00023384093947242945 2023-01-21 12:59:10.973578: step: 884/529, loss: 0.17876902222633362 2023-01-21 12:59:12.127910: step: 888/529, loss: 0.0006556510925292969 2023-01-21 12:59:13.247608: step: 892/529, loss: 0.00018453598022460938 2023-01-21 12:59:14.380780: step: 896/529, loss: 0.02855224721133709 2023-01-21 12:59:15.491897: step: 900/529, loss: 0.0027256966568529606 2023-01-21 12:59:16.651479: step: 904/529, loss: 0.009267712011933327 2023-01-21 12:59:17.821619: step: 908/529, loss: 0.01761026307940483 2023-01-21 12:59:18.978607: step: 912/529, loss: 5.7649613154353574e-05 2023-01-21 12:59:20.138102: step: 916/529, loss: 0.00031566619873046875 2023-01-21 12:59:21.254563: step: 920/529, loss: 0.015160655602812767 2023-01-21 12:59:22.414363: step: 924/529, loss: 0.07174129784107208 2023-01-21 12:59:23.530078: step: 928/529, loss: 4.3201445805607364e-05 2023-01-21 12:59:24.680371: step: 932/529, loss: 0.00011310577247058973 2023-01-21 12:59:25.809168: step: 936/529, loss: 0.00039958953857421875 2023-01-21 12:59:26.939969: step: 940/529, loss: 0.08551912754774094 2023-01-21 12:59:28.090119: step: 944/529, loss: 0.1024555191397667 2023-01-21 12:59:29.230663: step: 948/529, loss: 0.008458327502012253 2023-01-21 12:59:30.369886: step: 952/529, loss: 7.209777686512098e-05 2023-01-21 12:59:31.540570: step: 956/529, loss: 0.025521088391542435 2023-01-21 12:59:32.655885: step: 960/529, loss: 4.520416405284777e-05 2023-01-21 12:59:33.780160: step: 964/529, loss: 0.0006912231328897178 2023-01-21 12:59:34.901834: step: 968/529, loss: 0.0019927024841308594 2023-01-21 12:59:36.031096: step: 972/529, loss: 0.005639934912323952 2023-01-21 12:59:37.187028: step: 976/529, loss: 0.004732894711196423 2023-01-21 12:59:38.316959: step: 980/529, loss: 0.00022478103346657008 2023-01-21 12:59:39.445294: step: 984/529, loss: 0.0009107113000936806 2023-01-21 12:59:40.594585: step: 988/529, loss: 0.01344447210431099 2023-01-21 12:59:41.728647: step: 992/529, loss: 0.005950737278908491 2023-01-21 12:59:42.878164: step: 996/529, loss: 1.5735627130197827e-06 2023-01-21 12:59:44.020783: step: 1000/529, loss: 3.080368333030492e-05 2023-01-21 12:59:45.132089: step: 1004/529, loss: 0.022426128387451172 2023-01-21 12:59:46.293495: step: 1008/529, loss: 5.817413239128655e-06 2023-01-21 12:59:47.420817: step: 1012/529, loss: 0.004836845677345991 2023-01-21 12:59:48.574038: step: 1016/529, loss: 0.01833801344037056 2023-01-21 12:59:49.705855: step: 1020/529, loss: 0.0005708694225177169 2023-01-21 12:59:50.822719: step: 1024/529, loss: 0.0004096984921488911 2023-01-21 12:59:51.939551: step: 1028/529, loss: 2.956390289909905e-06 2023-01-21 12:59:53.036935: step: 1032/529, loss: 0.000637912773527205 2023-01-21 12:59:54.167412: step: 1036/529, loss: 0.00017151833162643015 2023-01-21 12:59:55.312151: step: 1040/529, loss: 0.013858413323760033 2023-01-21 12:59:56.436080: step: 1044/529, loss: 2.193450927734375e-05 2023-01-21 12:59:57.596394: step: 1048/529, loss: -3.6239621294953395e-06 2023-01-21 12:59:58.711421: step: 1052/529, loss: 0.0003209591086488217 2023-01-21 12:59:59.845732: step: 1056/529, loss: 1.277923547604587e-05 2023-01-21 13:00:01.009098: step: 1060/529, loss: 7.123947580112144e-05 2023-01-21 13:00:02.154461: step: 1064/529, loss: 0.029862498864531517 2023-01-21 13:00:03.325910: step: 1068/529, loss: 0.00033655168954283 2023-01-21 13:00:04.481069: step: 1072/529, loss: 0.000215435036807321 2023-01-21 13:00:05.654379: step: 1076/529, loss: 0.00070872314972803 2023-01-21 13:00:06.761485: step: 1080/529, loss: 0.043189048767089844 2023-01-21 13:00:07.888562: step: 1084/529, loss: 0.0008001327514648438 2023-01-21 13:00:09.041352: step: 1088/529, loss: 0.011207198724150658 2023-01-21 13:00:10.171894: step: 1092/529, loss: 0.00025463104248046875 2023-01-21 13:00:11.338572: step: 1096/529, loss: 0.014928244985640049 2023-01-21 13:00:12.499975: step: 1100/529, loss: 0.0008836746565066278 2023-01-21 13:00:13.676258: step: 1104/529, loss: 0.0008476257207803428 2023-01-21 13:00:14.826919: step: 1108/529, loss: 0.0005447387811727822 2023-01-21 13:00:15.957182: step: 1112/529, loss: 0.012865448370575905 2023-01-21 13:00:17.099684: step: 1116/529, loss: 0.006090450566262007 2023-01-21 13:00:18.247308: step: 1120/529, loss: 3.452301098150201e-05 2023-01-21 13:00:19.375817: step: 1124/529, loss: 0.0001719355641398579 2023-01-21 13:00:20.503691: step: 1128/529, loss: 0.001524257706478238 2023-01-21 13:00:21.623628: step: 1132/529, loss: 0.0004589080926962197 2023-01-21 13:00:22.779497: step: 1136/529, loss: 0.02356262318789959 2023-01-21 13:00:23.932288: step: 1140/529, loss: 0.0019364356994628906 2023-01-21 13:00:25.081843: step: 1144/529, loss: 7.343292054429185e-06 2023-01-21 13:00:26.212138: step: 1148/529, loss: 6.265639967750758e-05 2023-01-21 13:00:27.351361: step: 1152/529, loss: 0.000568008457776159 2023-01-21 13:00:28.518775: step: 1156/529, loss: 4.520416405284777e-05 2023-01-21 13:00:29.678820: step: 1160/529, loss: 0.006527900695800781 2023-01-21 13:00:30.820521: step: 1164/529, loss: 0.0015972137916833162 2023-01-21 13:00:31.950978: step: 1168/529, loss: 0.0838109478354454 2023-01-21 13:00:33.087495: step: 1172/529, loss: 0.010888099670410156 2023-01-21 13:00:34.217568: step: 1176/529, loss: 0.018135881051421165 2023-01-21 13:00:35.366367: step: 1180/529, loss: 0.015790175646543503 2023-01-21 13:00:36.489677: step: 1184/529, loss: 2.3651124138268642e-05 2023-01-21 13:00:37.606538: step: 1188/529, loss: 0.00017004013352561742 2023-01-21 13:00:38.704505: step: 1192/529, loss: 0.0024940490256994963 2023-01-21 13:00:39.874356: step: 1196/529, loss: 0.027503585442900658 2023-01-21 13:00:41.027018: step: 1200/529, loss: 0.022104360163211823 2023-01-21 13:00:42.202202: step: 1204/529, loss: 0.005454540252685547 2023-01-21 13:00:43.366354: step: 1208/529, loss: 9.536745437799254e-08 2023-01-21 13:00:44.513947: step: 1212/529, loss: 0.003224658779799938 2023-01-21 13:00:45.644690: step: 1216/529, loss: 0.011770820245146751 2023-01-21 13:00:46.774637: step: 1220/529, loss: 0.003178215119987726 2023-01-21 13:00:47.898757: step: 1224/529, loss: 0.0016875267028808594 2023-01-21 13:00:49.015313: step: 1228/529, loss: 0.010014343075454235 2023-01-21 13:00:50.142196: step: 1232/529, loss: 0.00030155183048918843 2023-01-21 13:00:51.299741: step: 1236/529, loss: 2.9850005375919864e-05 2023-01-21 13:00:52.414417: step: 1240/529, loss: 0.05257682874798775 2023-01-21 13:00:53.520150: step: 1244/529, loss: 7.286071922862902e-05 2023-01-21 13:00:54.647836: step: 1248/529, loss: 0.024857331067323685 2023-01-21 13:00:55.759441: step: 1252/529, loss: 5.407333082985133e-05 2023-01-21 13:00:56.887725: step: 1256/529, loss: 0.03173065185546875 2023-01-21 13:00:58.061543: step: 1260/529, loss: 0.00028362273587845266 2023-01-21 13:00:59.180641: step: 1264/529, loss: 7.82012939453125e-05 2023-01-21 13:01:00.300167: step: 1268/529, loss: 0.007254886440932751 2023-01-21 13:01:01.443926: step: 1272/529, loss: 0.02992229536175728 2023-01-21 13:01:02.587926: step: 1276/529, loss: 0.012044524773955345 2023-01-21 13:01:03.728755: step: 1280/529, loss: 8.583069757150952e-06 2023-01-21 13:01:04.858518: step: 1284/529, loss: 0.006409454625099897 2023-01-21 13:01:06.011616: step: 1288/529, loss: 0.0036017419770359993 2023-01-21 13:01:07.139802: step: 1292/529, loss: 0.060391616076231 2023-01-21 13:01:08.259148: step: 1296/529, loss: 0.023407461121678352 2023-01-21 13:01:09.393455: step: 1300/529, loss: 0.0007427216041833162 2023-01-21 13:01:10.540818: step: 1304/529, loss: 4.787445141118951e-05 2023-01-21 13:01:11.696424: step: 1308/529, loss: 9.250641596736386e-06 2023-01-21 13:01:12.862147: step: 1312/529, loss: 0.01085739117115736 2023-01-21 13:01:13.975924: step: 1316/529, loss: 0.0005583763704635203 2023-01-21 13:01:15.074245: step: 1320/529, loss: 0.0003733634948730469 2023-01-21 13:01:16.208292: step: 1324/529, loss: 0.00016384123591706157 2023-01-21 13:01:17.333479: step: 1328/529, loss: -5.078316007711692e-06 2023-01-21 13:01:18.437046: step: 1332/529, loss: 0.00017313957505393773 2023-01-21 13:01:19.542547: step: 1336/529, loss: 0.0031850815284997225 2023-01-21 13:01:20.666417: step: 1340/529, loss: 5.016326758777723e-05 2023-01-21 13:01:21.805244: step: 1344/529, loss: 0.00014157296391204 2023-01-21 13:01:22.943812: step: 1348/529, loss: 0.00375537876971066 2023-01-21 13:01:24.072613: step: 1352/529, loss: 0.00039844511775299907 2023-01-21 13:01:25.203383: step: 1356/529, loss: 3.843307786155492e-05 2023-01-21 13:01:26.348797: step: 1360/529, loss: 0.0011050999164581299 2023-01-21 13:01:27.457357: step: 1364/529, loss: 0.005911063868552446 2023-01-21 13:01:28.611621: step: 1368/529, loss: 7.781982276355848e-05 2023-01-21 13:01:29.753198: step: 1372/529, loss: 0.04933986812829971 2023-01-21 13:01:30.882453: step: 1376/529, loss: 0.0006162644131109118 2023-01-21 13:01:32.003152: step: 1380/529, loss: 0.013153172098100185 2023-01-21 13:01:33.092223: step: 1384/529, loss: 0.00022001266188453883 2023-01-21 13:01:34.211385: step: 1388/529, loss: 0.0002662658807821572 2023-01-21 13:01:35.342336: step: 1392/529, loss: 3.8814545405330136e-05 2023-01-21 13:01:36.470500: step: 1396/529, loss: 0.00038909912109375 2023-01-21 13:01:37.616590: step: 1400/529, loss: 0.0007755279657430947 2023-01-21 13:01:38.736780: step: 1404/529, loss: -9.5367431640625e-07 2023-01-21 13:01:39.845537: step: 1408/529, loss: 0.00013246535672806203 2023-01-21 13:01:40.990227: step: 1412/529, loss: 5.149841854290571e-06 2023-01-21 13:01:42.121264: step: 1416/529, loss: 0.0003727913135662675 2023-01-21 13:01:43.244115: step: 1420/529, loss: 0.031937409192323685 2023-01-21 13:01:44.354880: step: 1424/529, loss: 0.00014982224092818797 2023-01-21 13:01:45.503751: step: 1428/529, loss: 0.0013782501919195056 2023-01-21 13:01:46.625086: step: 1432/529, loss: 0.01560354232788086 2023-01-21 13:01:47.769886: step: 1436/529, loss: 0.010646057315170765 2023-01-21 13:01:48.869751: step: 1440/529, loss: 0.0003925800265278667 2023-01-21 13:01:49.969296: step: 1444/529, loss: 0.0003275871276855469 2023-01-21 13:01:51.098735: step: 1448/529, loss: 0.0013589859008789062 2023-01-21 13:01:52.225527: step: 1452/529, loss: 0.0010542869567871094 2023-01-21 13:01:53.337273: step: 1456/529, loss: 0.0009743691189214587 2023-01-21 13:01:54.460630: step: 1460/529, loss: 0.0018149376846849918 2023-01-21 13:01:55.589912: step: 1464/529, loss: 4.234314110362902e-05 2023-01-21 13:01:56.739040: step: 1468/529, loss: 0.013597773388028145 2023-01-21 13:01:57.870627: step: 1472/529, loss: 0.001113987062126398 2023-01-21 13:01:59.000213: step: 1476/529, loss: 0.018045520409941673 2023-01-21 13:02:00.150421: step: 1480/529, loss: 0.00022363664174918085 2023-01-21 13:02:01.286363: step: 1484/529, loss: 0.0009395600063726306 2023-01-21 13:02:02.406496: step: 1488/529, loss: 2.0694731574621983e-05 2023-01-21 13:02:03.535668: step: 1492/529, loss: 0.005797672085464001 2023-01-21 13:02:04.662092: step: 1496/529, loss: 0.0010786056518554688 2023-01-21 13:02:05.785017: step: 1500/529, loss: 2.8198955988045782e-05 2023-01-21 13:02:06.885916: step: 1504/529, loss: 0.00015583037748001516 2023-01-21 13:02:08.042152: step: 1508/529, loss: 0.05582170560956001 2023-01-21 13:02:09.166032: step: 1512/529, loss: 0.00010457039024913684 2023-01-21 13:02:10.293150: step: 1516/529, loss: 0.008766270242631435 2023-01-21 13:02:11.417768: step: 1520/529, loss: 0.00030136111308820546 2023-01-21 13:02:12.571304: step: 1524/529, loss: 0.005076408386230469 2023-01-21 13:02:13.711734: step: 1528/529, loss: 0.023174572736024857 2023-01-21 13:02:14.822554: step: 1532/529, loss: 0.0015774727798998356 2023-01-21 13:02:15.951267: step: 1536/529, loss: 0.0013002395862713456 2023-01-21 13:02:17.081621: step: 1540/529, loss: 3.967285010730848e-05 2023-01-21 13:02:18.215426: step: 1544/529, loss: 0.006561290938407183 2023-01-21 13:02:19.342010: step: 1548/529, loss: 0.00968628004193306 2023-01-21 13:02:20.467316: step: 1552/529, loss: 0.00025920866755768657 2023-01-21 13:02:21.599782: step: 1556/529, loss: 9.222030348610133e-05 2023-01-21 13:02:22.731517: step: 1560/529, loss: 0.0006708145374432206 2023-01-21 13:02:23.863621: step: 1564/529, loss: 0.017557907849550247 2023-01-21 13:02:25.006554: step: 1568/529, loss: 4.510879807639867e-05 2023-01-21 13:02:26.200271: step: 1572/529, loss: 0.0346345417201519 2023-01-21 13:02:27.331285: step: 1576/529, loss: 0.0022831440437585115 2023-01-21 13:02:28.500130: step: 1580/529, loss: 5.474090721691027e-05 2023-01-21 13:02:29.645051: step: 1584/529, loss: 0.023574447259306908 2023-01-21 13:02:30.779008: step: 1588/529, loss: 0.0004336357524152845 2023-01-21 13:02:31.892479: step: 1592/529, loss: 6.361008126987144e-05 2023-01-21 13:02:33.047325: step: 1596/529, loss: 0.00021452904911711812 2023-01-21 13:02:34.156782: step: 1600/529, loss: 6.365776062011719e-05 2023-01-21 13:02:35.292842: step: 1604/529, loss: 0.000492095947265625 2023-01-21 13:02:36.418726: step: 1608/529, loss: 0.0033493044320493937 2023-01-21 13:02:37.536638: step: 1612/529, loss: 0.0005512237548828125 2023-01-21 13:02:38.666937: step: 1616/529, loss: 0.0008112907526083291 2023-01-21 13:02:39.775635: step: 1620/529, loss: 5.187988426769152e-05 2023-01-21 13:02:40.875126: step: 1624/529, loss: 0.0016328812343999743 2023-01-21 13:02:42.001643: step: 1628/529, loss: 0.014824867248535156 2023-01-21 13:02:43.160532: step: 1632/529, loss: 0.0015411376953125 2023-01-21 13:02:44.277294: step: 1636/529, loss: 0.00016584395780228078 2023-01-21 13:02:45.408772: step: 1640/529, loss: 0.020351506769657135 2023-01-21 13:02:46.532395: step: 1644/529, loss: 5.340576535672881e-06 2023-01-21 13:02:47.666623: step: 1648/529, loss: 0.030076026916503906 2023-01-21 13:02:48.791042: step: 1652/529, loss: 1.6880036127986386e-05 2023-01-21 13:02:49.933431: step: 1656/529, loss: 0.0028074264992028475 2023-01-21 13:02:51.087715: step: 1660/529, loss: 0.0017692565452307463 2023-01-21 13:02:52.224829: step: 1664/529, loss: 0.0008330822456628084 2023-01-21 13:02:53.377703: step: 1668/529, loss: 4.520416405284777e-05 2023-01-21 13:02:54.511558: step: 1672/529, loss: 0.0011369704734534025 2023-01-21 13:02:55.640364: step: 1676/529, loss: 0.0015088021755218506 2023-01-21 13:02:56.762910: step: 1680/529, loss: 0.003866863204166293 2023-01-21 13:02:57.878370: step: 1684/529, loss: 0.0014294624561443925 2023-01-21 13:02:58.990553: step: 1688/529, loss: 0.00011692046973621473 2023-01-21 13:03:00.166333: step: 1692/529, loss: 0.04233064502477646 2023-01-21 13:03:01.322155: step: 1696/529, loss: 7.019639451755211e-05 2023-01-21 13:03:02.428364: step: 1700/529, loss: 2.47955313170678e-06 2023-01-21 13:03:03.539349: step: 1704/529, loss: 0.003833484835922718 2023-01-21 13:03:04.666410: step: 1708/529, loss: 0.0008071899646893144 2023-01-21 13:03:05.845739: step: 1712/529, loss: 0.031946755945682526 2023-01-21 13:03:06.998636: step: 1716/529, loss: 0.00388755789026618 2023-01-21 13:03:08.133627: step: 1720/529, loss: 0.000125885009765625 2023-01-21 13:03:09.249268: step: 1724/529, loss: 0.017482567578554153 2023-01-21 13:03:10.393343: step: 1728/529, loss: 3.309249586891383e-05 2023-01-21 13:03:11.590089: step: 1732/529, loss: 0.021950339898467064 2023-01-21 13:03:12.709731: step: 1736/529, loss: 0.020756147801876068 2023-01-21 13:03:13.839246: step: 1740/529, loss: 0.007392740808427334 2023-01-21 13:03:14.952413: step: 1744/529, loss: 0.0001312255917582661 2023-01-21 13:03:16.090215: step: 1748/529, loss: 0.0014501571422442794 2023-01-21 13:03:17.230953: step: 1752/529, loss: 0.0005636215792037547 2023-01-21 13:03:18.358265: step: 1756/529, loss: 1.630783299333416e-05 2023-01-21 13:03:19.512186: step: 1760/529, loss: 0.0012319564120844007 2023-01-21 13:03:20.638027: step: 1764/529, loss: 2.5272369384765625e-05 2023-01-21 13:03:21.773820: step: 1768/529, loss: 0.0031745911110192537 2023-01-21 13:03:22.897460: step: 1772/529, loss: 0.05386314541101456 2023-01-21 13:03:24.018933: step: 1776/529, loss: 0.00989675521850586 2023-01-21 13:03:25.171032: step: 1780/529, loss: 7.057189577608369e-06 2023-01-21 13:03:26.285363: step: 1784/529, loss: 1.913309097290039e-05 2023-01-21 13:03:27.400248: step: 1788/529, loss: 0.0403226837515831 2023-01-21 13:03:28.518504: step: 1792/529, loss: 0.0128027917817235 2023-01-21 13:03:29.621732: step: 1796/529, loss: 0.00035600661067292094 2023-01-21 13:03:30.720563: step: 1800/529, loss: 4.76837158203125e-07 2023-01-21 13:03:31.827038: step: 1804/529, loss: 3.6239625842426904e-06 2023-01-21 13:03:32.949867: step: 1808/529, loss: 0.0009952544933184981 2023-01-21 13:03:34.049119: step: 1812/529, loss: 1.0204315913142636e-05 2023-01-21 13:03:35.209449: step: 1816/529, loss: 0.0016179084777832031 2023-01-21 13:03:36.365633: step: 1820/529, loss: 0.00019531250291038305 2023-01-21 13:03:37.464210: step: 1824/529, loss: 0.0003185272216796875 2023-01-21 13:03:38.589127: step: 1828/529, loss: 0.04625721275806427 2023-01-21 13:03:39.709821: step: 1832/529, loss: 0.001628112862817943 2023-01-21 13:03:40.835944: step: 1836/529, loss: 0.0002382278471486643 2023-01-21 13:03:41.959554: step: 1840/529, loss: 0.0005735397571697831 2023-01-21 13:03:43.077145: step: 1844/529, loss: 0.004591417498886585 2023-01-21 13:03:44.168858: step: 1848/529, loss: 0.0018312096362933517 2023-01-21 13:03:45.318944: step: 1852/529, loss: 0.011274718679487705 2023-01-21 13:03:46.455967: step: 1856/529, loss: 0.20404338836669922 2023-01-21 13:03:47.558093: step: 1860/529, loss: 0.00748634384945035 2023-01-21 13:03:48.699244: step: 1864/529, loss: 0.003515625139698386 2023-01-21 13:03:49.807981: step: 1868/529, loss: 0.004873943515121937 2023-01-21 13:03:50.917791: step: 1872/529, loss: 0.09463195502758026 2023-01-21 13:03:52.059916: step: 1876/529, loss: 0.00016899108595680445 2023-01-21 13:03:53.190142: step: 1880/529, loss: 0.005844330880790949 2023-01-21 13:03:54.298919: step: 1884/529, loss: 0.012272643856704235 2023-01-21 13:03:55.444562: step: 1888/529, loss: 0.0001358032168354839 2023-01-21 13:03:56.565241: step: 1892/529, loss: 0.000583267246838659 2023-01-21 13:03:57.671002: step: 1896/529, loss: 0.00044279100256972015 2023-01-21 13:03:58.817903: step: 1900/529, loss: 0.005678749177604914 2023-01-21 13:03:59.975942: step: 1904/529, loss: 0.06716375052928925 2023-01-21 13:04:01.096636: step: 1908/529, loss: 0.0012897491687908769 2023-01-21 13:04:02.232491: step: 1912/529, loss: 7.629394644936838e-07 2023-01-21 13:04:03.385189: step: 1916/529, loss: 0.003146457951515913 2023-01-21 13:04:04.520640: step: 1920/529, loss: 0.017220117151737213 2023-01-21 13:04:05.660373: step: 1924/529, loss: 0.025179291144013405 2023-01-21 13:04:06.790785: step: 1928/529, loss: 0.01859769970178604 2023-01-21 13:04:07.932887: step: 1932/529, loss: 0.00045080186100676656 2023-01-21 13:04:09.067960: step: 1936/529, loss: 0.00013189316086936742 2023-01-21 13:04:10.162627: step: 1940/529, loss: 0.0013317108387127519 2023-01-21 13:04:11.307540: step: 1944/529, loss: 8.01086389401462e-06 2023-01-21 13:04:12.471674: step: 1948/529, loss: 0.003793192096054554 2023-01-21 13:04:13.606681: step: 1952/529, loss: 0.0007216691737994552 2023-01-21 13:04:14.765684: step: 1956/529, loss: 0.0025382996536791325 2023-01-21 13:04:15.896625: step: 1960/529, loss: 0.011472892947494984 2023-01-21 13:04:16.990878: step: 1964/529, loss: 0.00029506682767532766 2023-01-21 13:04:18.126690: step: 1968/529, loss: 0.02381540648639202 2023-01-21 13:04:19.251091: step: 1972/529, loss: 0.004935312084853649 2023-01-21 13:04:20.385623: step: 1976/529, loss: 0.03358879312872887 2023-01-21 13:04:21.508458: step: 1980/529, loss: 7.653236389160156e-05 2023-01-21 13:04:22.618029: step: 1984/529, loss: 0.00019989012798760086 2023-01-21 13:04:23.756111: step: 1988/529, loss: 0.03326358646154404 2023-01-21 13:04:24.891108: step: 1992/529, loss: 7.495880709029734e-05 2023-01-21 13:04:26.042604: step: 1996/529, loss: 0.01268157921731472 2023-01-21 13:04:27.173501: step: 2000/529, loss: 0.0002911090850830078 2023-01-21 13:04:28.315634: step: 2004/529, loss: 3.099441755693988e-06 2023-01-21 13:04:29.450526: step: 2008/529, loss: 0.04755115509033203 2023-01-21 13:04:30.558975: step: 2012/529, loss: 0.00027958155260421336 2023-01-21 13:04:31.692757: step: 2016/529, loss: 0.00018339157395530492 2023-01-21 13:04:32.790162: step: 2020/529, loss: 0.004827404394745827 2023-01-21 13:04:33.949935: step: 2024/529, loss: 0.010423803701996803 2023-01-21 13:04:35.075455: step: 2028/529, loss: 0.004334449768066406 2023-01-21 13:04:36.198500: step: 2032/529, loss: 0.0029649734497070312 2023-01-21 13:04:37.316249: step: 2036/529, loss: 0.000263214111328125 2023-01-21 13:04:38.440251: step: 2040/529, loss: 0.0004093170282430947 2023-01-21 13:04:39.609801: step: 2044/529, loss: 0.011854076758027077 2023-01-21 13:04:40.722404: step: 2048/529, loss: 0.011608600616455078 2023-01-21 13:04:41.858797: step: 2052/529, loss: 0.1497991681098938 2023-01-21 13:04:43.011144: step: 2056/529, loss: 0.006811714265495539 2023-01-21 13:04:44.141549: step: 2060/529, loss: 0.0003196716425009072 2023-01-21 13:04:45.288131: step: 2064/529, loss: 0.08308602124452591 2023-01-21 13:04:46.420953: step: 2068/529, loss: 0.012317419052124023 2023-01-21 13:04:47.579496: step: 2072/529, loss: 0.00596199044957757 2023-01-21 13:04:48.724172: step: 2076/529, loss: 0.00810022372752428 2023-01-21 13:04:49.839350: step: 2080/529, loss: 0.0073060994036495686 2023-01-21 13:04:50.952740: step: 2084/529, loss: 5.6982044043252245e-05 2023-01-21 13:04:52.075463: step: 2088/529, loss: 0.0009246825939044356 2023-01-21 13:04:53.192792: step: 2092/529, loss: 2.632141149661038e-05 2023-01-21 13:04:54.369917: step: 2096/529, loss: 0.002387619111686945 2023-01-21 13:04:55.513154: step: 2100/529, loss: 0.42215338349342346 2023-01-21 13:04:56.626046: step: 2104/529, loss: 2.822876012942288e-05 2023-01-21 13:04:57.739303: step: 2108/529, loss: 0.00105705251917243 2023-01-21 13:04:58.862226: step: 2112/529, loss: 0.00037355421227402985 2023-01-21 13:04:59.980766: step: 2116/529, loss: 2.288818359375e-05 ================================================== Loss: 0.015 -------------------- Dev: {'event': {'p': 0.6153846153846154, 'r': 0.7776298268974701, 'f1': 0.6870588235294117}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Test: {'event': {'p': 0.6281622911694511, 'r': 0.7624565469293163, 'f1': 0.688824914943732}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Chinese: {'event': {'p': 0.6071428571428571, 'r': 0.9444444444444444, 'f1': 0.7391304347826088}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Korean: {'event': {'p': 0.6346153846153846, 'r': 0.5238095238095238, 'f1': 0.5739130434782609}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Russian: {'event': {'p': 0.45, 'r': 0.5, 'f1': 0.4736842105263158}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6493506493506493, 'r': 0.9259259259259259, 'f1': 0.7633587786259541}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 26 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:05:41.095598: step: 4/529, loss: 8.96453821042087e-06 2023-01-21 13:05:42.219891: step: 8/529, loss: 4.4488908315543085e-05 2023-01-21 13:05:43.365983: step: 12/529, loss: 0.002907657530158758 2023-01-21 13:05:44.474640: step: 16/529, loss: 0.00037097930908203125 2023-01-21 13:05:45.619625: step: 20/529, loss: 0.004302501678466797 2023-01-21 13:05:46.722283: step: 24/529, loss: 0.30834951996803284 2023-01-21 13:05:47.855999: step: 28/529, loss: 0.00033321380033157766 2023-01-21 13:05:48.963829: step: 32/529, loss: 3.261566234868951e-05 2023-01-21 13:05:50.074337: step: 36/529, loss: 0.09074363857507706 2023-01-21 13:05:51.181832: step: 40/529, loss: 0.0003279685915913433 2023-01-21 13:05:52.289659: step: 44/529, loss: 0.0006367683527059853 2023-01-21 13:05:53.410801: step: 48/529, loss: 0.00043802260188385844 2023-01-21 13:05:54.535330: step: 52/529, loss: 3.1471254260395654e-06 2023-01-21 13:05:55.679824: step: 56/529, loss: 0.006996059790253639 2023-01-21 13:05:56.813200: step: 60/529, loss: 0.019316673278808594 2023-01-21 13:05:57.924110: step: 64/529, loss: 3.24249276673072e-06 2023-01-21 13:05:59.079440: step: 68/529, loss: 0.0003741264226846397 2023-01-21 13:06:00.236451: step: 72/529, loss: 0.0005276203155517578 2023-01-21 13:06:01.345663: step: 76/529, loss: 0.055098723620176315 2023-01-21 13:06:02.457690: step: 80/529, loss: 0.002068138215690851 2023-01-21 13:06:03.587099: step: 84/529, loss: 0.027022553607821465 2023-01-21 13:06:04.706936: step: 88/529, loss: 0.0007644862052984536 2023-01-21 13:06:05.824242: step: 92/529, loss: 0.0041280267760157585 2023-01-21 13:06:06.993830: step: 96/529, loss: 0.0018192291026934981 2023-01-21 13:06:08.125914: step: 100/529, loss: 0.00018129349336959422 2023-01-21 13:06:09.260825: step: 104/529, loss: 0.01891937293112278 2023-01-21 13:06:10.402369: step: 108/529, loss: 1.2207032341393642e-05 2023-01-21 13:06:11.553265: step: 112/529, loss: 0.03567390888929367 2023-01-21 13:06:12.670823: step: 116/529, loss: 0.007838440127670765 2023-01-21 13:06:13.809532: step: 120/529, loss: 0.004745387937873602 2023-01-21 13:06:14.934772: step: 124/529, loss: 0.08916740864515305 2023-01-21 13:06:16.087050: step: 128/529, loss: 0.00011096000525867566 2023-01-21 13:06:17.205947: step: 132/529, loss: 0.003695583436638117 2023-01-21 13:06:18.325841: step: 136/529, loss: 0.0042276387102901936 2023-01-21 13:06:19.485957: step: 140/529, loss: 0.00316448207013309 2023-01-21 13:06:20.625705: step: 144/529, loss: 1.33514404296875e-05 2023-01-21 13:06:21.767613: step: 148/529, loss: -5.53131121705519e-06 2023-01-21 13:06:22.878124: step: 152/529, loss: 0.012916945852339268 2023-01-21 13:06:24.031083: step: 156/529, loss: 0.01326827984303236 2023-01-21 13:06:25.144428: step: 160/529, loss: 0.005604743957519531 2023-01-21 13:06:26.285361: step: 164/529, loss: 8.535384949936997e-06 2023-01-21 13:06:27.395103: step: 168/529, loss: 0.00024299621873069555 2023-01-21 13:06:28.493526: step: 172/529, loss: 0.1924852430820465 2023-01-21 13:06:29.600944: step: 176/529, loss: 0.010990954004228115 2023-01-21 13:06:30.722192: step: 180/529, loss: 0.0005296707386150956 2023-01-21 13:06:31.876406: step: 184/529, loss: 0.0011724472278729081 2023-01-21 13:06:32.984516: step: 188/529, loss: 0.06467151641845703 2023-01-21 13:06:34.124463: step: 192/529, loss: 0.016022492200136185 2023-01-21 13:06:35.252743: step: 196/529, loss: 9.536745437799254e-08 2023-01-21 13:06:36.396411: step: 200/529, loss: 0.004081535153090954 2023-01-21 13:06:37.507421: step: 204/529, loss: -8.583068620282575e-07 2023-01-21 13:06:38.622305: step: 208/529, loss: 1.792907642084174e-05 2023-01-21 13:06:39.765472: step: 212/529, loss: 0.0718531683087349 2023-01-21 13:06:40.883074: step: 216/529, loss: 0.00038776398287154734 2023-01-21 13:06:42.002723: step: 220/529, loss: 2.174377368646674e-05 2023-01-21 13:06:43.141047: step: 224/529, loss: 0.044367216527462006 2023-01-21 13:06:44.266461: step: 228/529, loss: 2.250671423098538e-05 2023-01-21 13:06:45.380583: step: 232/529, loss: 4.024505687993951e-05 2023-01-21 13:06:46.493534: step: 236/529, loss: -3.24249276673072e-06 2023-01-21 13:06:47.623734: step: 240/529, loss: 0.006203079596161842 2023-01-21 13:06:48.732623: step: 244/529, loss: 0.0977473333477974 2023-01-21 13:06:49.898326: step: 248/529, loss: 0.021058129146695137 2023-01-21 13:06:51.027764: step: 252/529, loss: 5.340576171875e-05 2023-01-21 13:06:52.192110: step: 256/529, loss: 0.0009552002302370965 2023-01-21 13:06:53.324508: step: 260/529, loss: 5.3739546274300665e-05 2023-01-21 13:06:54.473231: step: 264/529, loss: 0.0010913849109783769 2023-01-21 13:06:55.586467: step: 268/529, loss: 0.012735843658447266 2023-01-21 13:06:56.720412: step: 272/529, loss: 0.00018100737361237407 2023-01-21 13:06:57.843197: step: 276/529, loss: 0.030130673199892044 2023-01-21 13:06:58.973145: step: 280/529, loss: 0.0010991096496582031 2023-01-21 13:07:00.102270: step: 284/529, loss: 0.001411533448845148 2023-01-21 13:07:01.232548: step: 288/529, loss: 0.001535916468128562 2023-01-21 13:07:02.364954: step: 292/529, loss: 0.0002761840878520161 2023-01-21 13:07:03.506615: step: 296/529, loss: 0.00039272307185456157 2023-01-21 13:07:04.633462: step: 300/529, loss: 0.0019807815551757812 2023-01-21 13:07:05.792845: step: 304/529, loss: 0.009366036392748356 2023-01-21 13:07:06.927522: step: 308/529, loss: 0.010482979007065296 2023-01-21 13:07:08.057617: step: 312/529, loss: -1.239776611328125e-05 2023-01-21 13:07:09.196836: step: 316/529, loss: 3.051757857974735e-06 2023-01-21 13:07:10.343839: step: 320/529, loss: 0.00038471221341751516 2023-01-21 13:07:11.455324: step: 324/529, loss: 0.09281168133020401 2023-01-21 13:07:12.589028: step: 328/529, loss: 0.0006923199398443103 2023-01-21 13:07:13.735212: step: 332/529, loss: 0.040256500244140625 2023-01-21 13:07:14.846019: step: 336/529, loss: 0.0006785392761230469 2023-01-21 13:07:15.970883: step: 340/529, loss: 4.291534423828125e-06 2023-01-21 13:07:17.109397: step: 344/529, loss: 0.00109691615216434 2023-01-21 13:07:18.259550: step: 348/529, loss: 0.006427288055419922 2023-01-21 13:07:19.394886: step: 352/529, loss: 3.24249276673072e-06 2023-01-21 13:07:20.523732: step: 356/529, loss: 0.03780064731836319 2023-01-21 13:07:21.676843: step: 360/529, loss: 0.5025267004966736 2023-01-21 13:07:22.807554: step: 364/529, loss: 9.059906005859375e-06 2023-01-21 13:07:23.936437: step: 368/529, loss: 0.00015249251737259328 2023-01-21 13:07:25.053748: step: 372/529, loss: 0.00033359526423737407 2023-01-21 13:07:26.174288: step: 376/529, loss: 0.00014324189396575093 2023-01-21 13:07:27.349414: step: 380/529, loss: 0.003743934677913785 2023-01-21 13:07:28.473756: step: 384/529, loss: 2.77042381640058e-05 2023-01-21 13:07:29.616566: step: 388/529, loss: 0.0007335662958212197 2023-01-21 13:07:30.755690: step: 392/529, loss: 1.296997106692288e-05 2023-01-21 13:07:31.922869: step: 396/529, loss: 0.0031459808815270662 2023-01-21 13:07:33.057856: step: 400/529, loss: 0.05556111037731171 2023-01-21 13:07:34.172830: step: 404/529, loss: 1.2351150512695312 2023-01-21 13:07:35.301812: step: 408/529, loss: 9.098052396439016e-05 2023-01-21 13:07:36.428733: step: 412/529, loss: 0.00017623901658225805 2023-01-21 13:07:37.595595: step: 416/529, loss: 0.00036535263643600047 2023-01-21 13:07:38.726666: step: 420/529, loss: 0.00022583008103538305 2023-01-21 13:07:39.869136: step: 424/529, loss: 3.910065061063506e-05 2023-01-21 13:07:41.006364: step: 428/529, loss: 0.000933742499910295 2023-01-21 13:07:42.120290: step: 432/529, loss: 3.0265749956015497e-05 2023-01-21 13:07:43.248883: step: 436/529, loss: 0.0075724124908447266 2023-01-21 13:07:44.381474: step: 440/529, loss: 0.7691519856452942 2023-01-21 13:07:45.537486: step: 444/529, loss: 0.001775455428287387 2023-01-21 13:07:46.692019: step: 448/529, loss: 0.034665536135435104 2023-01-21 13:07:47.832934: step: 452/529, loss: 1.3256072634248994e-05 2023-01-21 13:07:48.981868: step: 456/529, loss: 0.024933243170380592 2023-01-21 13:07:50.130120: step: 460/529, loss: 0.04186420515179634 2023-01-21 13:07:51.240228: step: 464/529, loss: 0.009877776727080345 2023-01-21 13:07:52.383523: step: 468/529, loss: 6.237030902411789e-05 2023-01-21 13:07:53.518981: step: 472/529, loss: 0.001939964247867465 2023-01-21 13:07:54.658207: step: 476/529, loss: 0.0004507065168581903 2023-01-21 13:07:55.789810: step: 480/529, loss: 3.1185150874080136e-05 2023-01-21 13:07:56.901082: step: 484/529, loss: 0.00043935776920989156 2023-01-21 13:07:58.021274: step: 488/529, loss: 0.004915905185043812 2023-01-21 13:07:59.182734: step: 492/529, loss: 0.0007184982532635331 2023-01-21 13:08:00.325442: step: 496/529, loss: 0.00047335628187283874 2023-01-21 13:08:01.474757: step: 500/529, loss: 0.08391036838293076 2023-01-21 13:08:02.631732: step: 504/529, loss: 5.9127810345671605e-06 2023-01-21 13:08:03.755548: step: 508/529, loss: 0.010232162661850452 2023-01-21 13:08:04.884293: step: 512/529, loss: 0.056586649268865585 2023-01-21 13:08:06.037468: step: 516/529, loss: 0.0015436173416674137 2023-01-21 13:08:07.176551: step: 520/529, loss: 0.0031437876168638468 2023-01-21 13:08:08.304373: step: 524/529, loss: 0.06874551624059677 2023-01-21 13:08:09.435657: step: 528/529, loss: 4.6348573960131034e-05 2023-01-21 13:08:10.565000: step: 532/529, loss: 6.146430678199977e-05 2023-01-21 13:08:11.711265: step: 536/529, loss: 0.04005012661218643 2023-01-21 13:08:12.912692: step: 540/529, loss: 0.0006504058837890625 2023-01-21 13:08:14.049955: step: 544/529, loss: 8.42094377730973e-05 2023-01-21 13:08:15.181373: step: 548/529, loss: 0.0010725021129474044 2023-01-21 13:08:16.335374: step: 552/529, loss: 0.021811293438076973 2023-01-21 13:08:17.453545: step: 556/529, loss: 3.4618376957951114e-05 2023-01-21 13:08:18.602007: step: 560/529, loss: 0.02456836774945259 2023-01-21 13:08:19.731369: step: 564/529, loss: 0.0021480561699718237 2023-01-21 13:08:20.846533: step: 568/529, loss: 0.025208855047822 2023-01-21 13:08:21.996264: step: 572/529, loss: 2.346038854739163e-05 2023-01-21 13:08:23.131081: step: 576/529, loss: 0.056058503687381744 2023-01-21 13:08:24.263392: step: 580/529, loss: 0.00014934540376998484 2023-01-21 13:08:25.389458: step: 584/529, loss: 0.006783866789191961 2023-01-21 13:08:26.504488: step: 588/529, loss: 0.022428322583436966 2023-01-21 13:08:27.632681: step: 592/529, loss: 0.0047227381728589535 2023-01-21 13:08:28.759299: step: 596/529, loss: 2.0313262211857364e-05 2023-01-21 13:08:29.899746: step: 600/529, loss: 0.013123512268066406 2023-01-21 13:08:31.056830: step: 604/529, loss: 0.0038618086837232113 2023-01-21 13:08:32.200708: step: 608/529, loss: 1.1873245057358872e-05 2023-01-21 13:08:33.320453: step: 612/529, loss: 0.0003950119426008314 2023-01-21 13:08:34.474658: step: 616/529, loss: 0.0004839897155761719 2023-01-21 13:08:35.649858: step: 620/529, loss: 0.0001333236723439768 2023-01-21 13:08:36.789156: step: 624/529, loss: 0.0266297347843647 2023-01-21 13:08:37.921640: step: 628/529, loss: 0.0013212204212322831 2023-01-21 13:08:39.055705: step: 632/529, loss: 3.24249276673072e-06 2023-01-21 13:08:40.196725: step: 636/529, loss: 7.886887033237144e-05 2023-01-21 13:08:41.347456: step: 640/529, loss: 0.0001672267826506868 2023-01-21 13:08:42.490019: step: 644/529, loss: 0.09417933970689774 2023-01-21 13:08:43.604341: step: 648/529, loss: 0.4512098431587219 2023-01-21 13:08:44.729973: step: 652/529, loss: 1.1157989320054185e-05 2023-01-21 13:08:45.853873: step: 656/529, loss: 0.0016958237392827868 2023-01-21 13:08:46.977965: step: 660/529, loss: 0.009069633670151234 2023-01-21 13:08:48.118665: step: 664/529, loss: 0.0366973876953125 2023-01-21 13:08:49.255239: step: 668/529, loss: 0.01766986958682537 2023-01-21 13:08:50.396933: step: 672/529, loss: 0.17697934806346893 2023-01-21 13:08:51.535882: step: 676/529, loss: 0.0019632340408861637 2023-01-21 13:08:52.692229: step: 680/529, loss: 0.00139617919921875 2023-01-21 13:08:53.810973: step: 684/529, loss: 0.0015018463600426912 2023-01-21 13:08:54.948870: step: 688/529, loss: 0.0141334543004632 2023-01-21 13:08:56.065814: step: 692/529, loss: 0.0002841949462890625 2023-01-21 13:08:57.191844: step: 696/529, loss: 0.24535417556762695 2023-01-21 13:08:58.288526: step: 700/529, loss: 0.000568008457776159 2023-01-21 13:08:59.422190: step: 704/529, loss: 0.004303932189941406 2023-01-21 13:09:00.545906: step: 708/529, loss: 2.727508581301663e-05 2023-01-21 13:09:01.664822: step: 712/529, loss: 0.0004912376753054559 2023-01-21 13:09:02.792261: step: 716/529, loss: 0.016112612560391426 2023-01-21 13:09:03.924681: step: 720/529, loss: 0.00021057129197288305 2023-01-21 13:09:05.043110: step: 724/529, loss: 1.735687328618951e-05 2023-01-21 13:09:06.139224: step: 728/529, loss: 0.03154563903808594 2023-01-21 13:09:07.285182: step: 732/529, loss: 1.0681153071345761e-05 2023-01-21 13:09:08.432564: step: 736/529, loss: 3.166198803228326e-05 2023-01-21 13:09:09.562147: step: 740/529, loss: 0.01360998209565878 2023-01-21 13:09:10.710419: step: 744/529, loss: 0.00011034011549782008 2023-01-21 13:09:11.883307: step: 748/529, loss: 8.769035048317164e-05 2023-01-21 13:09:13.024026: step: 752/529, loss: 0.0005148887867107987 2023-01-21 13:09:14.174684: step: 756/529, loss: 0.02043437957763672 2023-01-21 13:09:15.343110: step: 760/529, loss: 0.00034484866773709655 2023-01-21 13:09:16.498256: step: 764/529, loss: 0.0001930236758198589 2023-01-21 13:09:17.642494: step: 768/529, loss: 0.0018659591441974044 2023-01-21 13:09:18.742302: step: 772/529, loss: 0.00015707017155364156 2023-01-21 13:09:19.881067: step: 776/529, loss: 0.0005457877996377647 2023-01-21 13:09:21.009268: step: 780/529, loss: 0.015224170871078968 2023-01-21 13:09:22.132735: step: 784/529, loss: 0.00010557175119174644 2023-01-21 13:09:23.261031: step: 788/529, loss: 0.0006026267656125128 2023-01-21 13:09:24.415982: step: 792/529, loss: 0.1750757247209549 2023-01-21 13:09:25.550321: step: 796/529, loss: 0.012186002917587757 2023-01-21 13:09:26.712829: step: 800/529, loss: 0.0015291214222088456 2023-01-21 13:09:27.878528: step: 804/529, loss: 0.00023746490478515625 2023-01-21 13:09:29.009618: step: 808/529, loss: 0.001124477363191545 2023-01-21 13:09:30.112014: step: 812/529, loss: 0.006601524073630571 2023-01-21 13:09:31.227900: step: 816/529, loss: 0.01242895144969225 2023-01-21 13:09:32.384649: step: 820/529, loss: 0.009865950793027878 2023-01-21 13:09:33.504110: step: 824/529, loss: 0.0004544258408714086 2023-01-21 13:09:34.626987: step: 828/529, loss: 0.0067078592255711555 2023-01-21 13:09:35.802731: step: 832/529, loss: 0.00017529279284644872 2023-01-21 13:09:36.897880: step: 836/529, loss: 6.256103370105848e-05 2023-01-21 13:09:38.053561: step: 840/529, loss: 0.0012102127075195312 2023-01-21 13:09:39.188965: step: 844/529, loss: 0.0049002645537257195 2023-01-21 13:09:40.319855: step: 848/529, loss: 0.009509658440947533 2023-01-21 13:09:41.444078: step: 852/529, loss: 1.1920928955078125e-05 2023-01-21 13:09:42.577696: step: 856/529, loss: 0.00020799637422896922 2023-01-21 13:09:43.707565: step: 860/529, loss: 0.00026636122493073344 2023-01-21 13:09:44.828617: step: 864/529, loss: 0.0007151603931561112 2023-01-21 13:09:45.977365: step: 868/529, loss: 3.156662205583416e-05 2023-01-21 13:09:47.132136: step: 872/529, loss: 0.0005590915679931641 2023-01-21 13:09:48.275279: step: 876/529, loss: 0.005429076962172985 2023-01-21 13:09:49.418027: step: 880/529, loss: 0.005028152372688055 2023-01-21 13:09:50.539191: step: 884/529, loss: 0.007692909333854914 2023-01-21 13:09:51.674479: step: 888/529, loss: 0.0010052680736407638 2023-01-21 13:09:52.812293: step: 892/529, loss: 7.32898770365864e-05 2023-01-21 13:09:53.945353: step: 896/529, loss: 0.0015866756439208984 2023-01-21 13:09:55.082612: step: 900/529, loss: 0.0001235961972270161 2023-01-21 13:09:56.192634: step: 904/529, loss: 0.00016899110050871968 2023-01-21 13:09:57.307107: step: 908/529, loss: 0.034955691546201706 2023-01-21 13:09:58.454648: step: 912/529, loss: 5.9127810345671605e-06 2023-01-21 13:09:59.615573: step: 916/529, loss: 0.0010168076260015368 2023-01-21 13:10:00.744059: step: 920/529, loss: 0.010859775356948376 2023-01-21 13:10:01.877853: step: 924/529, loss: 6.532669431180693e-06 2023-01-21 13:10:02.996654: step: 928/529, loss: 0.0006397724500857294 2023-01-21 13:10:04.100521: step: 932/529, loss: 0.0002849578741006553 2023-01-21 13:10:05.235250: step: 936/529, loss: 0.0031124113593250513 2023-01-21 13:10:06.370660: step: 940/529, loss: 0.0004236697859596461 2023-01-21 13:10:07.528610: step: 944/529, loss: 0.0029403686057776213 2023-01-21 13:10:08.688155: step: 948/529, loss: 4.0626528061693534e-05 2023-01-21 13:10:09.832847: step: 952/529, loss: 0.0004571914905682206 2023-01-21 13:10:10.952354: step: 956/529, loss: 4.539489600574598e-05 2023-01-21 13:10:12.106887: step: 960/529, loss: 0.017323875799775124 2023-01-21 13:10:13.230283: step: 964/529, loss: 0.004857063293457031 2023-01-21 13:10:14.369553: step: 968/529, loss: 0.018172835931181908 2023-01-21 13:10:15.501974: step: 972/529, loss: 0.00021157265291549265 2023-01-21 13:10:16.624488: step: 976/529, loss: 3.070831371587701e-05 2023-01-21 13:10:17.729270: step: 980/529, loss: 0.00037059784517623484 2023-01-21 13:10:18.850662: step: 984/529, loss: 0.04704561084508896 2023-01-21 13:10:19.953377: step: 988/529, loss: 0.00142078404314816 2023-01-21 13:10:21.085286: step: 992/529, loss: 0.0006403445731848478 2023-01-21 13:10:22.233342: step: 996/529, loss: 0.04136943817138672 2023-01-21 13:10:23.372323: step: 1000/529, loss: 0.08497877418994904 2023-01-21 13:10:24.530320: step: 1004/529, loss: 0.021245384588837624 2023-01-21 13:10:25.700526: step: 1008/529, loss: 0.0018030166393145919 2023-01-21 13:10:26.815884: step: 1012/529, loss: 0.0008520125993527472 2023-01-21 13:10:27.969278: step: 1016/529, loss: 0.0025169372092932463 2023-01-21 13:10:29.103101: step: 1020/529, loss: 0.0009592533460818231 2023-01-21 13:10:30.208178: step: 1024/529, loss: 0.0011817931663244963 2023-01-21 13:10:31.363952: step: 1028/529, loss: 0.06252622604370117 2023-01-21 13:10:32.515697: step: 1032/529, loss: 0.012900067493319511 2023-01-21 13:10:33.663781: step: 1036/529, loss: 4.8160552978515625e-05 2023-01-21 13:10:34.770864: step: 1040/529, loss: 0.00027866364689543843 2023-01-21 13:10:35.930224: step: 1044/529, loss: 0.000592136406339705 2023-01-21 13:10:37.136961: step: 1048/529, loss: 0.001291465712711215 2023-01-21 13:10:38.280217: step: 1052/529, loss: 0.0004100799560546875 2023-01-21 13:10:39.431293: step: 1056/529, loss: 7.152557373046875e-07 2023-01-21 13:10:40.578858: step: 1060/529, loss: 0.008410071954131126 2023-01-21 13:10:41.725025: step: 1064/529, loss: 0.04589061811566353 2023-01-21 13:10:42.884340: step: 1068/529, loss: 0.010752486996352673 2023-01-21 13:10:44.005787: step: 1072/529, loss: 0.0020864487159997225 2023-01-21 13:10:45.136604: step: 1076/529, loss: 0.002227210905402899 2023-01-21 13:10:46.258375: step: 1080/529, loss: 0.00018053055100608617 2023-01-21 13:10:47.387081: step: 1084/529, loss: 0.005826139822602272 2023-01-21 13:10:48.536847: step: 1088/529, loss: 0.005850601010024548 2023-01-21 13:10:49.671501: step: 1092/529, loss: 2.784729076665826e-05 2023-01-21 13:10:50.801782: step: 1096/529, loss: 0.0032237053383141756 2023-01-21 13:10:51.935120: step: 1100/529, loss: 0.03524189069867134 2023-01-21 13:10:53.061358: step: 1104/529, loss: 0.0002799004432745278 2023-01-21 13:10:54.194732: step: 1108/529, loss: 0.021262170746922493 2023-01-21 13:10:55.332573: step: 1112/529, loss: 0.00638999929651618 2023-01-21 13:10:56.469168: step: 1116/529, loss: 0.002037906553596258 2023-01-21 13:10:57.583003: step: 1120/529, loss: 0.0007957458728924394 2023-01-21 13:10:58.712393: step: 1124/529, loss: 0.43960848450660706 2023-01-21 13:10:59.833732: step: 1128/529, loss: 1.258850079466356e-05 2023-01-21 13:11:00.997616: step: 1132/529, loss: 0.004829979036003351 2023-01-21 13:11:02.135036: step: 1136/529, loss: 0.016794968396425247 2023-01-21 13:11:03.263531: step: 1140/529, loss: 0.00046176911564543843 2023-01-21 13:11:04.452514: step: 1144/529, loss: 0.022540664300322533 2023-01-21 13:11:05.592405: step: 1148/529, loss: 0.0004364013730082661 2023-01-21 13:11:06.728814: step: 1152/529, loss: 0.0005045891157351434 2023-01-21 13:11:07.843899: step: 1156/529, loss: 0.002950024791061878 2023-01-21 13:11:08.994514: step: 1160/529, loss: 0.0016188144218176603 2023-01-21 13:11:10.138596: step: 1164/529, loss: 0.03747672960162163 2023-01-21 13:11:11.268962: step: 1168/529, loss: 0.015600014477968216 2023-01-21 13:11:12.387582: step: 1172/529, loss: 0.0008846282726153731 2023-01-21 13:11:13.506354: step: 1176/529, loss: 0.009301901794970036 2023-01-21 13:11:14.638958: step: 1180/529, loss: 0.002321624895557761 2023-01-21 13:11:15.764356: step: 1184/529, loss: 0.00018367767916060984 2023-01-21 13:11:16.910680: step: 1188/529, loss: 0.031671904027462006 2023-01-21 13:11:18.062298: step: 1192/529, loss: 0.004562854766845703 2023-01-21 13:11:19.176793: step: 1196/529, loss: 0.0003017425478901714 2023-01-21 13:11:20.337872: step: 1200/529, loss: 0.09037017822265625 2023-01-21 13:11:21.497174: step: 1204/529, loss: 0.00662574777379632 2023-01-21 13:11:22.628594: step: 1208/529, loss: 8.98361176950857e-05 2023-01-21 13:11:23.747721: step: 1212/529, loss: 0.0009596825111657381 2023-01-21 13:11:24.883214: step: 1216/529, loss: 0.0002099037083098665 2023-01-21 13:11:26.017241: step: 1220/529, loss: 0.2793978750705719 2023-01-21 13:11:27.154138: step: 1224/529, loss: 3.14712519866589e-06 2023-01-21 13:11:28.285850: step: 1228/529, loss: 0.001050710678100586 2023-01-21 13:11:29.426667: step: 1232/529, loss: 0.03034496307373047 2023-01-21 13:11:30.592066: step: 1236/529, loss: 0.2675352096557617 2023-01-21 13:11:31.750890: step: 1240/529, loss: 0.00013446807861328125 2023-01-21 13:11:32.901697: step: 1244/529, loss: 0.0014706611400470138 2023-01-21 13:11:34.022618: step: 1248/529, loss: 3.6907196772517636e-05 2023-01-21 13:11:35.145486: step: 1252/529, loss: 0.0008793831220827997 2023-01-21 13:11:36.275238: step: 1256/529, loss: 0.0881032943725586 2023-01-21 13:11:37.372945: step: 1260/529, loss: 2.4938583010225557e-05 2023-01-21 13:11:38.553944: step: 1264/529, loss: 0.011941909790039062 2023-01-21 13:11:39.700439: step: 1268/529, loss: 0.010002613067626953 2023-01-21 13:11:40.833338: step: 1272/529, loss: 0.0001237869291799143 2023-01-21 13:11:41.968546: step: 1276/529, loss: 0.0019558907952159643 2023-01-21 13:11:43.079389: step: 1280/529, loss: 0.031830836087465286 2023-01-21 13:11:44.171308: step: 1284/529, loss: 9.002684964798391e-05 2023-01-21 13:11:45.310068: step: 1288/529, loss: 4.76837158203125e-07 2023-01-21 13:11:46.451894: step: 1292/529, loss: 0.00022544861712958664 2023-01-21 13:11:47.574664: step: 1296/529, loss: 1.7833710444392636e-05 2023-01-21 13:11:48.706670: step: 1300/529, loss: 0.03179354593157768 2023-01-21 13:11:49.818485: step: 1304/529, loss: 0.004338455386459827 2023-01-21 13:11:50.952693: step: 1308/529, loss: 0.1165735274553299 2023-01-21 13:11:52.094556: step: 1312/529, loss: 0.005232191178947687 2023-01-21 13:11:53.234623: step: 1316/529, loss: 0.001739120576530695 2023-01-21 13:11:54.367015: step: 1320/529, loss: 0.046267319470644 2023-01-21 13:11:55.512261: step: 1324/529, loss: 0.0024361610412597656 2023-01-21 13:11:56.636918: step: 1328/529, loss: 0.003507423447445035 2023-01-21 13:11:57.776311: step: 1332/529, loss: 0.014626026153564453 2023-01-21 13:11:58.936429: step: 1336/529, loss: 0.004086542408913374 2023-01-21 13:12:00.101823: step: 1340/529, loss: 9.727478754939511e-06 2023-01-21 13:12:01.259162: step: 1344/529, loss: 6.65664701955393e-05 2023-01-21 13:12:02.393363: step: 1348/529, loss: 0.00417251605540514 2023-01-21 13:12:03.536848: step: 1352/529, loss: 0.0012601851485669613 2023-01-21 13:12:04.667758: step: 1356/529, loss: 7.3909759521484375e-06 2023-01-21 13:12:05.821095: step: 1360/529, loss: 0.016218949109315872 2023-01-21 13:12:06.961513: step: 1364/529, loss: 0.0018317223293706775 2023-01-21 13:12:08.086304: step: 1368/529, loss: 0.00023431777663063258 2023-01-21 13:12:09.220643: step: 1372/529, loss: 0.0007516384357586503 2023-01-21 13:12:10.326162: step: 1376/529, loss: 0.0014029026497155428 2023-01-21 13:12:11.438127: step: 1380/529, loss: 0.0012340785469859838 2023-01-21 13:12:12.545453: step: 1384/529, loss: 7.98225446487777e-05 2023-01-21 13:12:13.669471: step: 1388/529, loss: 0.000264072441495955 2023-01-21 13:12:14.808048: step: 1392/529, loss: 0.00384101876989007 2023-01-21 13:12:15.945042: step: 1396/529, loss: 8.77380352903856e-06 2023-01-21 13:12:17.072918: step: 1400/529, loss: 0.031851768493652344 2023-01-21 13:12:18.234596: step: 1404/529, loss: 0.0003501892206259072 2023-01-21 13:12:19.406245: step: 1408/529, loss: 5.035400317865424e-05 2023-01-21 13:12:20.550158: step: 1412/529, loss: 0.024449730291962624 2023-01-21 13:12:21.684643: step: 1416/529, loss: 6.675720669591101e-07 2023-01-21 13:12:22.806418: step: 1420/529, loss: 0.0012067795032635331 2023-01-21 13:12:23.943673: step: 1424/529, loss: 0.024570656940340996 2023-01-21 13:12:25.121420: step: 1428/529, loss: 0.05239295959472656 2023-01-21 13:12:26.259673: step: 1432/529, loss: 0.014387321658432484 2023-01-21 13:12:27.399376: step: 1436/529, loss: 0.03099660947918892 2023-01-21 13:12:28.542172: step: 1440/529, loss: 0.02800159528851509 2023-01-21 13:12:29.652094: step: 1444/529, loss: 0.001612854073755443 2023-01-21 13:12:30.773204: step: 1448/529, loss: 0.2234325408935547 2023-01-21 13:12:31.926510: step: 1452/529, loss: 0.02595958672463894 2023-01-21 13:12:33.073488: step: 1456/529, loss: 0.00014762878709007055 2023-01-21 13:12:34.205191: step: 1460/529, loss: 0.006358528509736061 2023-01-21 13:12:35.364288: step: 1464/529, loss: 0.0019478797912597656 2023-01-21 13:12:36.501407: step: 1468/529, loss: 8.60214204294607e-05 2023-01-21 13:12:37.618039: step: 1472/529, loss: 0.042075347155332565 2023-01-21 13:12:38.762397: step: 1476/529, loss: 5.726814197259955e-05 2023-01-21 13:12:39.900451: step: 1480/529, loss: 0.0007891893037594855 2023-01-21 13:12:41.021367: step: 1484/529, loss: 0.004864311311393976 2023-01-21 13:12:42.189043: step: 1488/529, loss: 0.6411482691764832 2023-01-21 13:12:43.305098: step: 1492/529, loss: 0.0012891769874840975 2023-01-21 13:12:44.465684: step: 1496/529, loss: 0.004120063968002796 2023-01-21 13:12:45.621201: step: 1500/529, loss: 0.026310063898563385 2023-01-21 13:12:46.743586: step: 1504/529, loss: 0.00031447410583496094 2023-01-21 13:12:47.870495: step: 1508/529, loss: 0.000286960625089705 2023-01-21 13:12:49.033625: step: 1512/529, loss: 0.0010220528347417712 2023-01-21 13:12:50.186435: step: 1516/529, loss: 0.005054282955825329 2023-01-21 13:12:51.338006: step: 1520/529, loss: 0.038840580731630325 2023-01-21 13:12:52.487420: step: 1524/529, loss: 0.0013293266529217362 2023-01-21 13:12:53.613750: step: 1528/529, loss: 6.86645489622606e-06 2023-01-21 13:12:54.753457: step: 1532/529, loss: 0.020098304376006126 2023-01-21 13:12:55.870281: step: 1536/529, loss: 4.57763690064894e-06 2023-01-21 13:12:57.010492: step: 1540/529, loss: 9.70840483205393e-05 2023-01-21 13:12:58.142246: step: 1544/529, loss: 0.008495330810546875 2023-01-21 13:12:59.325587: step: 1548/529, loss: 0.0585138313472271 2023-01-21 13:13:00.487859: step: 1552/529, loss: 8.993149094749242e-05 2023-01-21 13:13:01.619002: step: 1556/529, loss: 0.0025121690705418587 2023-01-21 13:13:02.808291: step: 1560/529, loss: 0.008127069100737572 2023-01-21 13:13:03.960187: step: 1564/529, loss: 0.005356073845177889 2023-01-21 13:13:05.096673: step: 1568/529, loss: 3.409385681152344e-05 2023-01-21 13:13:06.206737: step: 1572/529, loss: 0.00020461082749534398 2023-01-21 13:13:07.350968: step: 1576/529, loss: 0.001741170883178711 2023-01-21 13:13:08.449739: step: 1580/529, loss: 0.0009002685546875 2023-01-21 13:13:09.559673: step: 1584/529, loss: 0.00589942978695035 2023-01-21 13:13:10.713770: step: 1588/529, loss: 0.015538979321718216 2023-01-21 13:13:11.878577: step: 1592/529, loss: 0.0016294480301439762 2023-01-21 13:13:12.983492: step: 1596/529, loss: 1.883506956801284e-05 2023-01-21 13:13:14.094091: step: 1600/529, loss: 0.0008074759971350431 2023-01-21 13:13:15.230713: step: 1604/529, loss: 0.0193954948335886 2023-01-21 13:13:16.359474: step: 1608/529, loss: 0.00011940002877963707 2023-01-21 13:13:17.537777: step: 1612/529, loss: 8.39233416627394e-06 2023-01-21 13:13:18.660929: step: 1616/529, loss: 0.007128715515136719 2023-01-21 13:13:19.800016: step: 1620/529, loss: 0.0029812813736498356 2023-01-21 13:13:20.924844: step: 1624/529, loss: 0.0038179398979991674 2023-01-21 13:13:22.054104: step: 1628/529, loss: 0.3166733682155609 2023-01-21 13:13:23.185186: step: 1632/529, loss: 0.00024018288240768015 2023-01-21 13:13:24.336407: step: 1636/529, loss: 0.00017385483079124242 2023-01-21 13:13:25.458426: step: 1640/529, loss: 9.412765211891383e-05 2023-01-21 13:13:26.593353: step: 1644/529, loss: 0.07429618388414383 2023-01-21 13:13:27.725227: step: 1648/529, loss: 0.14327794313430786 2023-01-21 13:13:28.856669: step: 1652/529, loss: 0.007124996278434992 2023-01-21 13:13:29.984842: step: 1656/529, loss: 0.01361846923828125 2023-01-21 13:13:31.134207: step: 1660/529, loss: 0.0011077881790697575 2023-01-21 13:13:32.286504: step: 1664/529, loss: 0.0008556365501135588 2023-01-21 13:13:33.420938: step: 1668/529, loss: 0.0033148766960948706 2023-01-21 13:13:34.583384: step: 1672/529, loss: 0.001680564833804965 2023-01-21 13:13:35.684812: step: 1676/529, loss: 0.00018877982802223414 2023-01-21 13:13:36.828887: step: 1680/529, loss: 0.0029356004670262337 2023-01-21 13:13:37.941625: step: 1684/529, loss: 0.021885110065340996 2023-01-21 13:13:39.071086: step: 1688/529, loss: 0.0006837845430709422 2023-01-21 13:13:40.198025: step: 1692/529, loss: 0.057054709643125534 2023-01-21 13:13:41.317838: step: 1696/529, loss: 0.0024429322220385075 2023-01-21 13:13:42.479879: step: 1700/529, loss: 4.234314110362902e-05 2023-01-21 13:13:43.625499: step: 1704/529, loss: 0.021677494049072266 2023-01-21 13:13:44.718400: step: 1708/529, loss: 0.0029172897338867188 2023-01-21 13:13:45.847304: step: 1712/529, loss: 6.942749314475805e-05 2023-01-21 13:13:47.017839: step: 1716/529, loss: 5.168914867681451e-05 2023-01-21 13:13:48.172717: step: 1720/529, loss: 0.001136875245720148 2023-01-21 13:13:49.301469: step: 1724/529, loss: 0.06153149902820587 2023-01-21 13:13:50.419668: step: 1728/529, loss: 0.052689552307128906 2023-01-21 13:13:51.565169: step: 1732/529, loss: 0.00362396240234375 2023-01-21 13:13:52.698498: step: 1736/529, loss: 0.017075633630156517 2023-01-21 13:13:53.834857: step: 1740/529, loss: 6.27517729299143e-05 2023-01-21 13:13:54.970686: step: 1744/529, loss: 0.00016860960749909282 2023-01-21 13:13:56.123688: step: 1748/529, loss: 0.0006542205810546875 2023-01-21 13:13:57.285328: step: 1752/529, loss: 3.738403393072076e-05 2023-01-21 13:13:58.413368: step: 1756/529, loss: 0.00019721986609511077 2023-01-21 13:13:59.514085: step: 1760/529, loss: 0.00017056464275810868 2023-01-21 13:14:00.652228: step: 1764/529, loss: 1.4781950994802173e-05 2023-01-21 13:14:01.812340: step: 1768/529, loss: 0.0031566619873046875 2023-01-21 13:14:02.929536: step: 1772/529, loss: 2.3746490114717744e-05 2023-01-21 13:14:04.057500: step: 1776/529, loss: 0.0035467150155454874 2023-01-21 13:14:05.214317: step: 1780/529, loss: 0.005482482723891735 2023-01-21 13:14:06.326449: step: 1784/529, loss: 0.0015802383422851562 2023-01-21 13:14:07.484187: step: 1788/529, loss: 0.05314197763800621 2023-01-21 13:14:08.637191: step: 1792/529, loss: 0.00035266875056549907 2023-01-21 13:14:09.764111: step: 1796/529, loss: 0.0001321792515227571 2023-01-21 13:14:10.880569: step: 1800/529, loss: 0.06937695294618607 2023-01-21 13:14:11.974086: step: 1804/529, loss: 0.0022573471069335938 2023-01-21 13:14:13.121061: step: 1808/529, loss: 0.0017583847511559725 2023-01-21 13:14:14.237563: step: 1812/529, loss: 0.00016269684419967234 2023-01-21 13:14:15.384147: step: 1816/529, loss: 0.0035915374755859375 2023-01-21 13:14:16.507744: step: 1820/529, loss: 0.04482555389404297 2023-01-21 13:14:17.628409: step: 1824/529, loss: 1.2588501704158261e-05 2023-01-21 13:14:18.749590: step: 1828/529, loss: 7.200241270766128e-06 2023-01-21 13:14:19.878858: step: 1832/529, loss: 1.010894811770413e-05 2023-01-21 13:14:21.000940: step: 1836/529, loss: 0.006625365931540728 2023-01-21 13:14:22.119525: step: 1840/529, loss: 0.0014390945434570312 2023-01-21 13:14:23.246887: step: 1844/529, loss: 7.905960228526965e-05 2023-01-21 13:14:24.385723: step: 1848/529, loss: 0.022931288927793503 2023-01-21 13:14:25.538504: step: 1852/529, loss: 6.29425048828125e-05 2023-01-21 13:14:26.669430: step: 1856/529, loss: 4.539489600574598e-05 2023-01-21 13:14:27.807155: step: 1860/529, loss: 0.002043819520622492 2023-01-21 13:14:28.971957: step: 1864/529, loss: 0.020247364416718483 2023-01-21 13:14:30.093436: step: 1868/529, loss: 0.0005067826132290065 2023-01-21 13:14:31.237199: step: 1872/529, loss: 0.0029332160484045744 2023-01-21 13:14:32.350854: step: 1876/529, loss: 0.0002668410597834736 2023-01-21 13:14:33.469665: step: 1880/529, loss: 0.00011348724365234375 2023-01-21 13:14:34.621199: step: 1884/529, loss: 0.01920933648943901 2023-01-21 13:14:35.772302: step: 1888/529, loss: 0.0016807080246508121 2023-01-21 13:14:36.916305: step: 1892/529, loss: 0.00021810531325172633 2023-01-21 13:14:38.073712: step: 1896/529, loss: 0.014886093325912952 2023-01-21 13:14:39.235462: step: 1900/529, loss: 0.01766815222799778 2023-01-21 13:14:40.385415: step: 1904/529, loss: 0.00021224022202659398 2023-01-21 13:14:41.495246: step: 1908/529, loss: 0.011406040750443935 2023-01-21 13:14:42.665926: step: 1912/529, loss: -1.2302398317842744e-05 2023-01-21 13:14:43.794596: step: 1916/529, loss: 0.0007192611810751259 2023-01-21 13:14:44.918362: step: 1920/529, loss: 0.011962317861616611 2023-01-21 13:14:46.054175: step: 1924/529, loss: 0.0008192062377929688 2023-01-21 13:14:47.163633: step: 1928/529, loss: 0.0005619526491500437 2023-01-21 13:14:48.311009: step: 1932/529, loss: 0.0021730423904955387 2023-01-21 13:14:49.457078: step: 1936/529, loss: 0.0028985978569835424 2023-01-21 13:14:50.595362: step: 1940/529, loss: 7.328986976062879e-05 2023-01-21 13:14:51.775929: step: 1944/529, loss: 0.0033998489379882812 2023-01-21 13:14:52.904240: step: 1948/529, loss: 0.0002611160452943295 2023-01-21 13:14:54.059453: step: 1952/529, loss: 0.0026807785034179688 2023-01-21 13:14:55.189686: step: 1956/529, loss: 2.0027162008773303e-06 2023-01-21 13:14:56.336416: step: 1960/529, loss: 5.702972339349799e-05 2023-01-21 13:14:57.481250: step: 1964/529, loss: 0.0003987312666140497 2023-01-21 13:14:58.641543: step: 1968/529, loss: 0.0005405425908975303 2023-01-21 13:14:59.767710: step: 1972/529, loss: 0.056925203651189804 2023-01-21 13:15:00.915443: step: 1976/529, loss: 0.003406715579330921 2023-01-21 13:15:02.035813: step: 1980/529, loss: 7.543563697254285e-05 2023-01-21 13:15:03.150570: step: 1984/529, loss: 0.004996395204216242 2023-01-21 13:15:04.277298: step: 1988/529, loss: 0.005165576934814453 2023-01-21 13:15:05.444267: step: 1992/529, loss: 0.12199898064136505 2023-01-21 13:15:06.537846: step: 1996/529, loss: 0.0004701971774920821 2023-01-21 13:15:07.675561: step: 2000/529, loss: 0.042629048228263855 2023-01-21 13:15:08.794238: step: 2004/529, loss: 0.07139568030834198 2023-01-21 13:15:09.911771: step: 2008/529, loss: 0.0044276234693825245 2023-01-21 13:15:11.069348: step: 2012/529, loss: 0.0021921158768236637 2023-01-21 13:15:12.197883: step: 2016/529, loss: 0.0017807483673095703 2023-01-21 13:15:13.311559: step: 2020/529, loss: 0.013555336743593216 2023-01-21 13:15:14.441336: step: 2024/529, loss: 0.001087188720703125 2023-01-21 13:15:15.592501: step: 2028/529, loss: 0.01615452766418457 2023-01-21 13:15:16.706648: step: 2032/529, loss: 0.0007308006170205772 2023-01-21 13:15:17.805630: step: 2036/529, loss: 0.0006583214271813631 2023-01-21 13:15:18.960684: step: 2040/529, loss: 6.237030174816027e-05 2023-01-21 13:15:20.074641: step: 2044/529, loss: 0.04879894480109215 2023-01-21 13:15:21.184140: step: 2048/529, loss: 2.746581958490424e-05 2023-01-21 13:15:22.320612: step: 2052/529, loss: 0.0022540094796568155 2023-01-21 13:15:23.483441: step: 2056/529, loss: 0.001861619995906949 2023-01-21 13:15:24.605718: step: 2060/529, loss: 0.01276102103292942 2023-01-21 13:15:25.744733: step: 2064/529, loss: 0.00034198761568404734 2023-01-21 13:15:26.867818: step: 2068/529, loss: 6.675720669591101e-07 2023-01-21 13:15:27.976006: step: 2072/529, loss: 0.00024704934912733734 2023-01-21 13:15:29.108000: step: 2076/529, loss: 3.9863589336164296e-05 2023-01-21 13:15:30.225436: step: 2080/529, loss: 0.001080417656339705 2023-01-21 13:15:31.350997: step: 2084/529, loss: 0.02632465399801731 2023-01-21 13:15:32.457898: step: 2088/529, loss: 0.0032678607385605574 2023-01-21 13:15:33.580908: step: 2092/529, loss: 0.017656518146395683 2023-01-21 13:15:34.720090: step: 2096/529, loss: 2.0134873390197754 2023-01-21 13:15:35.839358: step: 2100/529, loss: 0.004620742984116077 2023-01-21 13:15:36.989935: step: 2104/529, loss: 0.0014112471835687757 2023-01-21 13:15:38.128609: step: 2108/529, loss: 3.290176755399443e-05 2023-01-21 13:15:39.258153: step: 2112/529, loss: 0.0011428833240643144 2023-01-21 13:15:40.402716: step: 2116/529, loss: 0.0002211570827057585 ================================================== Loss: 0.025 -------------------- Dev: {'event': {'p': 0.6096033402922756, 'r': 0.7776298268974701, 'f1': 0.6834406085430076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Test: {'event': {'p': 0.6149954001839927, 'r': 0.7746234067207416, 'f1': 0.6856410256410257}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Chinese: {'event': {'p': 0.6375, 'r': 0.9444444444444444, 'f1': 0.7611940298507462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Korean: {'event': {'p': 0.6078431372549019, 'r': 0.49206349206349204, 'f1': 0.543859649122807}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Russian: {'event': {'p': 0.3953488372093023, 'r': 0.4722222222222222, 'f1': 0.43037974683544306}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6096033402922756, 'r': 0.7776298268974701, 'f1': 0.6834406085430076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Eng Test for Chinese: {'event': {'p': 0.6149954001839927, 'r': 0.7746234067207416, 'f1': 0.6856410256410257}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Sample Chinese: {'event': {'p': 0.6375, 'r': 0.9444444444444444, 'f1': 0.7611940298507462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} -------------------- Eng Dev for Korean: {'event': {'p': 0.6123535676251332, 'r': 0.7656458055925432, 'f1': 0.6804733727810652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Korean: {'event': {'p': 0.6227824463118581, 'r': 0.7728852838933952, 'f1': 0.6897621509824199}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Korean: {'event': {'p': 0.74, 'r': 0.5873015873015873, 'f1': 0.6548672566371682}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 27 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:16:29.261118: step: 4/529, loss: 0.01877613179385662 2023-01-21 13:16:30.392627: step: 8/529, loss: 0.0045996191911399364 2023-01-21 13:16:31.520293: step: 12/529, loss: 0.024889184162020683 2023-01-21 13:16:32.620390: step: 16/529, loss: 0.0029392242431640625 2023-01-21 13:16:33.776624: step: 20/529, loss: 0.005602359771728516 2023-01-21 13:16:34.928881: step: 24/529, loss: 0.0003555297735147178 2023-01-21 13:16:36.041118: step: 28/529, loss: 0.005156993865966797 2023-01-21 13:16:37.202744: step: 32/529, loss: 0.006790542975068092 2023-01-21 13:16:38.317552: step: 36/529, loss: 0.00028667450533248484 2023-01-21 13:16:39.434467: step: 40/529, loss: 0.010752296075224876 2023-01-21 13:16:40.540532: step: 44/529, loss: 0.0002865791320800781 2023-01-21 13:16:41.685340: step: 48/529, loss: 1.1444091796875e-05 2023-01-21 13:16:42.817060: step: 52/529, loss: 5.1975250244140625e-05 2023-01-21 13:16:43.919853: step: 56/529, loss: 0.002250194549560547 2023-01-21 13:16:45.043079: step: 60/529, loss: 0.014471721835434437 2023-01-21 13:16:46.169903: step: 64/529, loss: 3.533363633323461e-05 2023-01-21 13:16:47.308485: step: 68/529, loss: 0.00017604828462935984 2023-01-21 13:16:48.441770: step: 72/529, loss: 0.0006144524086266756 2023-01-21 13:16:49.552665: step: 76/529, loss: 0.002903461456298828 2023-01-21 13:16:50.678173: step: 80/529, loss: 0.029028989374637604 2023-01-21 13:16:51.802432: step: 84/529, loss: 0.007819557562470436 2023-01-21 13:16:52.932934: step: 88/529, loss: 0.0008594513055868447 2023-01-21 13:16:54.088755: step: 92/529, loss: 0.0024612427223473787 2023-01-21 13:16:55.205267: step: 96/529, loss: 0.0023249625228345394 2023-01-21 13:16:56.318861: step: 100/529, loss: 0.0005080223199911416 2023-01-21 13:16:57.454566: step: 104/529, loss: 0.00034351350041106343 2023-01-21 13:16:58.616077: step: 108/529, loss: 0.009500122629106045 2023-01-21 13:16:59.735038: step: 112/529, loss: 1.621246337890625e-05 2023-01-21 13:17:00.865765: step: 116/529, loss: 0.03311539068818092 2023-01-21 13:17:02.034253: step: 120/529, loss: 3.5715103877009824e-05 2023-01-21 13:17:03.173145: step: 124/529, loss: 0.0008008956792764366 2023-01-21 13:17:04.292204: step: 128/529, loss: 0.03768310323357582 2023-01-21 13:17:05.448604: step: 132/529, loss: 4.310607982915826e-05 2023-01-21 13:17:06.578569: step: 136/529, loss: 0.015679167583584785 2023-01-21 13:17:07.682758: step: 140/529, loss: 0.00017795563326217234 2023-01-21 13:17:08.826247: step: 144/529, loss: 0.00023899078951217234 2023-01-21 13:17:09.956137: step: 148/529, loss: 0.00026454924955032766 2023-01-21 13:17:11.076520: step: 152/529, loss: 0.014461660757660866 2023-01-21 13:17:12.250116: step: 156/529, loss: 0.02699088864028454 2023-01-21 13:17:13.353418: step: 160/529, loss: 0.0017379760975018144 2023-01-21 13:17:14.494942: step: 164/529, loss: 0.006054020021110773 2023-01-21 13:17:15.632501: step: 168/529, loss: 0.023838043212890625 2023-01-21 13:17:16.791349: step: 172/529, loss: 0.0016600609524175525 2023-01-21 13:17:17.908895: step: 176/529, loss: 0.0006956100114621222 2023-01-21 13:17:19.080993: step: 180/529, loss: 0.008653211407363415 2023-01-21 13:17:20.215986: step: 184/529, loss: 0.0007164000999182463 2023-01-21 13:17:21.366699: step: 188/529, loss: 0.021378088742494583 2023-01-21 13:17:22.530301: step: 192/529, loss: 0.009367180988192558 2023-01-21 13:17:23.641847: step: 196/529, loss: 0.004074192140251398 2023-01-21 13:17:24.795125: step: 200/529, loss: 0.05244855582714081 2023-01-21 13:17:25.929158: step: 204/529, loss: 0.0002850055752787739 2023-01-21 13:17:27.065742: step: 208/529, loss: 0.0032117844093590975 2023-01-21 13:17:28.184452: step: 212/529, loss: 0.0041141510009765625 2023-01-21 13:17:29.299969: step: 216/529, loss: 2.7990341550321318e-05 2023-01-21 13:17:30.446572: step: 220/529, loss: 0.01878166012465954 2023-01-21 13:17:31.546974: step: 224/529, loss: 0.0006101608742028475 2023-01-21 13:17:32.667795: step: 228/529, loss: 7.247925168485381e-06 2023-01-21 13:17:33.807236: step: 232/529, loss: 5.273819260764867e-05 2023-01-21 13:17:34.934777: step: 236/529, loss: 0.0005838394281454384 2023-01-21 13:17:36.038653: step: 240/529, loss: 0.007197141647338867 2023-01-21 13:17:37.151941: step: 244/529, loss: 0.03648977354168892 2023-01-21 13:17:38.266019: step: 248/529, loss: -2.0456314814509824e-05 2023-01-21 13:17:39.419087: step: 252/529, loss: 0.008053970523178577 2023-01-21 13:17:40.551254: step: 256/529, loss: 3.814697265625e-05 2023-01-21 13:17:41.694812: step: 260/529, loss: 0.10630855709314346 2023-01-21 13:17:42.848240: step: 264/529, loss: 0.00013952255540061742 2023-01-21 13:17:43.987963: step: 268/529, loss: 0.0020471573807299137 2023-01-21 13:17:45.127512: step: 272/529, loss: 1.52587890625e-05 2023-01-21 13:17:46.264256: step: 276/529, loss: 0.05373058468103409 2023-01-21 13:17:47.425551: step: 280/529, loss: 3.461838059592992e-05 2023-01-21 13:17:48.563153: step: 284/529, loss: 0.009411811828613281 2023-01-21 13:17:49.712595: step: 288/529, loss: 0.0004093170282430947 2023-01-21 13:17:50.827461: step: 292/529, loss: -4.291534423828125e-06 2023-01-21 13:17:51.981207: step: 296/529, loss: 0.010420894250273705 2023-01-21 13:17:53.107257: step: 300/529, loss: 0.0005058288807049394 2023-01-21 13:17:54.231789: step: 304/529, loss: 0.0002392292080912739 2023-01-21 13:17:55.365669: step: 308/529, loss: 0.004777717404067516 2023-01-21 13:17:56.488219: step: 312/529, loss: 4.9114227294921875e-05 2023-01-21 13:17:57.629424: step: 316/529, loss: 0.0011608123313635588 2023-01-21 13:17:58.767839: step: 320/529, loss: 0.0007036208990029991 2023-01-21 13:17:59.906649: step: 324/529, loss: 0.014149094000458717 2023-01-21 13:18:01.013388: step: 328/529, loss: 0.00208702078089118 2023-01-21 13:18:02.147443: step: 332/529, loss: 0.0010094642639160156 2023-01-21 13:18:03.296899: step: 336/529, loss: 0.003392982529476285 2023-01-21 13:18:04.447585: step: 340/529, loss: 0.0889488235116005 2023-01-21 13:18:05.578173: step: 344/529, loss: 0.0003709793381858617 2023-01-21 13:18:06.733227: step: 348/529, loss: 0.00039014816866256297 2023-01-21 13:18:07.892113: step: 352/529, loss: 0.018527794629335403 2023-01-21 13:18:09.026542: step: 356/529, loss: 7.433891732944176e-05 2023-01-21 13:18:10.149178: step: 360/529, loss: 0.0016128539573401213 2023-01-21 13:18:11.267674: step: 364/529, loss: 0.0012690543662756681 2023-01-21 13:18:12.367483: step: 368/529, loss: 0.0034369470085948706 2023-01-21 13:18:13.518963: step: 372/529, loss: 0.019535206258296967 2023-01-21 13:18:14.648535: step: 376/529, loss: 0.0009048461797647178 2023-01-21 13:18:15.779018: step: 380/529, loss: 9.91821252682712e-06 2023-01-21 13:18:16.921948: step: 384/529, loss: 0.12987442314624786 2023-01-21 13:18:18.052043: step: 388/529, loss: 3.886222839355469e-05 2023-01-21 13:18:19.184946: step: 392/529, loss: 0.004604244604706764 2023-01-21 13:18:20.298289: step: 396/529, loss: 0.0010246277088299394 2023-01-21 13:18:21.441406: step: 400/529, loss: 0.009026146493852139 2023-01-21 13:18:22.578567: step: 404/529, loss: 0.002623939421027899 2023-01-21 13:18:23.727999: step: 408/529, loss: 0.0002058029203908518 2023-01-21 13:18:24.857551: step: 412/529, loss: 0.01665782928466797 2023-01-21 13:18:25.974974: step: 416/529, loss: 5.578995114774443e-05 2023-01-21 13:18:27.137915: step: 420/529, loss: 0.0038452150765806437 2023-01-21 13:18:28.320627: step: 424/529, loss: 0.000984668848104775 2023-01-21 13:18:29.455026: step: 428/529, loss: 0.00011081695993198082 2023-01-21 13:18:30.565503: step: 432/529, loss: -1.1777878171415068e-05 2023-01-21 13:18:31.712323: step: 436/529, loss: 0.0004066467226948589 2023-01-21 13:18:32.871824: step: 440/529, loss: 0.029067421332001686 2023-01-21 13:18:34.016144: step: 444/529, loss: 5.119443085277453e-05 2023-01-21 13:18:35.137516: step: 448/529, loss: 4.5776364459015895e-06 2023-01-21 13:18:36.291402: step: 452/529, loss: 0.017982101067900658 2023-01-21 13:18:37.403235: step: 456/529, loss: 0.0021360397804528475 2023-01-21 13:18:38.538058: step: 460/529, loss: 0.04451503977179527 2023-01-21 13:18:39.658685: step: 464/529, loss: 0.03155536949634552 2023-01-21 13:18:40.820950: step: 468/529, loss: 0.025257302448153496 2023-01-21 13:18:41.967820: step: 472/529, loss: 0.0005167007329873741 2023-01-21 13:18:43.074938: step: 476/529, loss: 0.32609355449676514 2023-01-21 13:18:44.202166: step: 480/529, loss: 0.00048446658183820546 2023-01-21 13:18:45.309178: step: 484/529, loss: 0.0002944946463685483 2023-01-21 13:18:46.427544: step: 488/529, loss: 0.00026035308837890625 2023-01-21 13:18:47.576515: step: 492/529, loss: 3.600120544433594e-05 2023-01-21 13:18:48.718298: step: 496/529, loss: 0.0007471084827557206 2023-01-21 13:18:49.875524: step: 500/529, loss: 0.07924485206604004 2023-01-21 13:18:51.021192: step: 504/529, loss: 0.05313320457935333 2023-01-21 13:18:52.145945: step: 508/529, loss: 4.580021050060168e-05 2023-01-21 13:18:53.270205: step: 512/529, loss: 0.0015383244026452303 2023-01-21 13:18:54.419488: step: 516/529, loss: 0.039701513946056366 2023-01-21 13:18:55.557429: step: 520/529, loss: 0.02309436909854412 2023-01-21 13:18:56.694232: step: 524/529, loss: 0.012848759070038795 2023-01-21 13:18:57.808058: step: 528/529, loss: 7.953643944347277e-05 2023-01-21 13:18:58.944135: step: 532/529, loss: 0.00240497593767941 2023-01-21 13:19:00.094497: step: 536/529, loss: 0.04354157671332359 2023-01-21 13:19:01.232315: step: 540/529, loss: 0.024097824469208717 2023-01-21 13:19:02.373612: step: 544/529, loss: 0.11643514782190323 2023-01-21 13:19:03.510861: step: 548/529, loss: 1.354217511106981e-05 2023-01-21 13:19:04.644688: step: 552/529, loss: 0.0005895614740438759 2023-01-21 13:19:05.771826: step: 556/529, loss: 0.0022148131392896175 2023-01-21 13:19:06.882751: step: 560/529, loss: 1.1634827387752011e-05 2023-01-21 13:19:08.003092: step: 564/529, loss: 0.00010156631469726562 2023-01-21 13:19:09.162397: step: 568/529, loss: 0.0015079498989507556 2023-01-21 13:19:10.321523: step: 572/529, loss: 0.027283288538455963 2023-01-21 13:19:11.463818: step: 576/529, loss: 0.004084968473762274 2023-01-21 13:19:12.635864: step: 580/529, loss: 0.0007440135232172906 2023-01-21 13:19:13.773148: step: 584/529, loss: 0.001910400460474193 2023-01-21 13:19:14.877967: step: 588/529, loss: 0.0008209228399209678 2023-01-21 13:19:15.991200: step: 592/529, loss: 0.011160850524902344 2023-01-21 13:19:17.119155: step: 596/529, loss: 2.86102294921875e-05 2023-01-21 13:19:18.279071: step: 600/529, loss: 0.0006711006280966103 2023-01-21 13:19:19.433750: step: 604/529, loss: 0.0004781723255291581 2023-01-21 13:19:20.539509: step: 608/529, loss: 0.02392597310245037 2023-01-21 13:19:21.674719: step: 612/529, loss: 0.004221725277602673 2023-01-21 13:19:22.826698: step: 616/529, loss: 0.006194591522216797 2023-01-21 13:19:23.948796: step: 620/529, loss: 4.243850526108872e-06 2023-01-21 13:19:25.085546: step: 624/529, loss: 0.06521406769752502 2023-01-21 13:19:26.229284: step: 628/529, loss: 0.00603065500035882 2023-01-21 13:19:27.328560: step: 632/529, loss: 0.00013961792865302414 2023-01-21 13:19:28.486004: step: 636/529, loss: 0.007125282660126686 2023-01-21 13:19:29.613814: step: 640/529, loss: 1.811981201171875e-05 2023-01-21 13:19:30.752017: step: 644/529, loss: -5.435942512121983e-06 2023-01-21 13:19:31.850601: step: 648/529, loss: 2.86102294921875e-05 2023-01-21 13:19:32.988582: step: 652/529, loss: 3.261566234868951e-05 2023-01-21 13:19:34.129265: step: 656/529, loss: 3.147125607938506e-05 2023-01-21 13:19:35.249308: step: 660/529, loss: 0.0734824389219284 2023-01-21 13:19:36.391028: step: 664/529, loss: 0.0017293930286541581 2023-01-21 13:19:37.528875: step: 668/529, loss: 0.0008948325994424522 2023-01-21 13:19:38.649691: step: 672/529, loss: 0.0018323898548260331 2023-01-21 13:19:39.770475: step: 676/529, loss: 3.7765505112474784e-05 2023-01-21 13:19:40.921379: step: 680/529, loss: 5.5789947509765625e-05 2023-01-21 13:19:42.045114: step: 684/529, loss: 9.493828110862523e-05 2023-01-21 13:19:43.207377: step: 688/529, loss: 0.0031824109610170126 2023-01-21 13:19:44.341601: step: 692/529, loss: 0.0001234054652741179 2023-01-21 13:19:45.486605: step: 696/529, loss: 0.01739940606057644 2023-01-21 13:19:46.612479: step: 700/529, loss: 5.064010838395916e-05 2023-01-21 13:19:47.746638: step: 704/529, loss: 0.01585989072918892 2023-01-21 13:19:48.900461: step: 708/529, loss: 0.0009117603185586631 2023-01-21 13:19:50.052146: step: 712/529, loss: 0.01773970201611519 2023-01-21 13:19:51.186999: step: 716/529, loss: 0.0016792297828942537 2023-01-21 13:19:52.320326: step: 720/529, loss: 0.002284812740981579 2023-01-21 13:19:53.463834: step: 724/529, loss: 0.0020591735374182463 2023-01-21 13:19:54.602175: step: 728/529, loss: 0.03585081174969673 2023-01-21 13:19:55.717953: step: 732/529, loss: -6.198883056640625e-06 2023-01-21 13:19:56.830703: step: 736/529, loss: 0.03481331095099449 2023-01-21 13:19:57.974211: step: 740/529, loss: 0.3417752981185913 2023-01-21 13:19:59.101722: step: 744/529, loss: 0.00028705596923828125 2023-01-21 13:20:00.238654: step: 748/529, loss: 1.76429750808893e-06 2023-01-21 13:20:01.347190: step: 752/529, loss: 0.00616798410192132 2023-01-21 13:20:02.463566: step: 756/529, loss: 0.028513241559267044 2023-01-21 13:20:03.576276: step: 760/529, loss: 1.1539460501808207e-05 2023-01-21 13:20:04.713983: step: 764/529, loss: 0.004114341922104359 2023-01-21 13:20:05.865676: step: 768/529, loss: 0.008676767349243164 2023-01-21 13:20:07.012865: step: 772/529, loss: 0.0022975921165198088 2023-01-21 13:20:08.160009: step: 776/529, loss: 0.05476045608520508 2023-01-21 13:20:09.308405: step: 780/529, loss: 0.00019664765568450093 2023-01-21 13:20:10.430807: step: 784/529, loss: 1.449584942747606e-05 2023-01-21 13:20:11.583888: step: 788/529, loss: 6.103516352595761e-05 2023-01-21 13:20:12.692688: step: 792/529, loss: 0.00016412737022619694 2023-01-21 13:20:13.855217: step: 796/529, loss: 0.0017936707008630037 2023-01-21 13:20:15.010190: step: 800/529, loss: 0.0012067795032635331 2023-01-21 13:20:16.176872: step: 804/529, loss: 0.00033178331796079874 2023-01-21 13:20:17.278363: step: 808/529, loss: 4.959106809110381e-05 2023-01-21 13:20:18.398749: step: 812/529, loss: 0.0001415252627339214 2023-01-21 13:20:19.540103: step: 816/529, loss: 0.0259381290525198 2023-01-21 13:20:20.682765: step: 820/529, loss: 0.00010719299461925402 2023-01-21 13:20:21.814138: step: 824/529, loss: 0.00014090538024902344 2023-01-21 13:20:22.956171: step: 828/529, loss: 0.03328676521778107 2023-01-21 13:20:24.079051: step: 832/529, loss: 0.0008338928455486894 2023-01-21 13:20:25.223098: step: 836/529, loss: 0.4863090515136719 2023-01-21 13:20:26.373566: step: 840/529, loss: 0.008635140024125576 2023-01-21 13:20:27.506160: step: 844/529, loss: 0.01300191879272461 2023-01-21 13:20:28.638125: step: 848/529, loss: 4.1770937968976796e-05 2023-01-21 13:20:29.748486: step: 852/529, loss: 3.5285952435515355e-06 2023-01-21 13:20:30.888453: step: 856/529, loss: 0.004270744509994984 2023-01-21 13:20:32.030700: step: 860/529, loss: 2.918243444582913e-05 2023-01-21 13:20:33.177162: step: 864/529, loss: 0.00195140833966434 2023-01-21 13:20:34.334338: step: 868/529, loss: 0.00493545550853014 2023-01-21 13:20:35.437604: step: 872/529, loss: 2.5272369384765625e-05 2023-01-21 13:20:36.549853: step: 876/529, loss: 0.0014650345547124743 2023-01-21 13:20:37.675163: step: 880/529, loss: 0.0003293991321697831 2023-01-21 13:20:38.796650: step: 884/529, loss: 0.0555514320731163 2023-01-21 13:20:39.936911: step: 888/529, loss: 8.1062319168268e-07 2023-01-21 13:20:41.060086: step: 892/529, loss: 0.020026778802275658 2023-01-21 13:20:42.208624: step: 896/529, loss: 0.0020479203667491674 2023-01-21 13:20:43.343145: step: 900/529, loss: 0.0001936435728566721 2023-01-21 13:20:44.469614: step: 904/529, loss: 0.10913264006376266 2023-01-21 13:20:45.636156: step: 908/529, loss: 0.0008628845680505037 2023-01-21 13:20:46.747896: step: 912/529, loss: 1.125335711549269e-05 2023-01-21 13:20:47.910258: step: 916/529, loss: 7.371902756858617e-05 2023-01-21 13:20:49.037000: step: 920/529, loss: 0.0010141372913494706 2023-01-21 13:20:50.208065: step: 924/529, loss: 0.0001165389985544607 2023-01-21 13:20:51.346498: step: 928/529, loss: 0.03232503682374954 2023-01-21 13:20:52.508758: step: 932/529, loss: 0.0846710205078125 2023-01-21 13:20:53.638938: step: 936/529, loss: 0.00018634795560501516 2023-01-21 13:20:54.748513: step: 940/529, loss: 5.455017526401207e-05 2023-01-21 13:20:55.912237: step: 944/529, loss: 0.00016307830810546875 2023-01-21 13:20:57.054977: step: 948/529, loss: 0.0007653236389160156 2023-01-21 13:20:58.183307: step: 952/529, loss: 0.0009449005592614412 2023-01-21 13:20:59.302547: step: 956/529, loss: 3.80516066798009e-05 2023-01-21 13:21:00.435399: step: 960/529, loss: 0.000457763671875 2023-01-21 13:21:01.584833: step: 964/529, loss: 5.340575626178179e-06 2023-01-21 13:21:02.731999: step: 968/529, loss: 0.0009399413829669356 2023-01-21 13:21:03.885522: step: 972/529, loss: 1.926422191900201e-05 2023-01-21 13:21:05.007638: step: 976/529, loss: 0.21249523758888245 2023-01-21 13:21:06.125343: step: 980/529, loss: 0.00044312476529739797 2023-01-21 13:21:07.266229: step: 984/529, loss: 0.12654857337474823 2023-01-21 13:21:08.409292: step: 988/529, loss: 0.0014652252430096269 2023-01-21 13:21:09.539201: step: 992/529, loss: 0.0030826570000499487 2023-01-21 13:21:10.688479: step: 996/529, loss: 1.0108947208209429e-05 2023-01-21 13:21:11.824989: step: 1000/529, loss: 0.0001900673087220639 2023-01-21 13:21:12.954595: step: 1004/529, loss: 2.117157055181451e-05 2023-01-21 13:21:14.104794: step: 1008/529, loss: 0.00432434119284153 2023-01-21 13:21:15.259441: step: 1012/529, loss: 0.02536649815738201 2023-01-21 13:21:16.384533: step: 1016/529, loss: 8.583068620282575e-07 2023-01-21 13:21:17.523406: step: 1020/529, loss: 0.007507228758186102 2023-01-21 13:21:18.651582: step: 1024/529, loss: 0.003591442247852683 2023-01-21 13:21:19.775940: step: 1028/529, loss: 0.0001751899835653603 2023-01-21 13:21:20.908605: step: 1032/529, loss: 0.00323486328125 2023-01-21 13:21:22.034383: step: 1036/529, loss: -1.0490416570974048e-06 2023-01-21 13:21:23.136599: step: 1040/529, loss: 8.630753654870205e-06 2023-01-21 13:21:24.287156: step: 1044/529, loss: 7.743835158180445e-05 2023-01-21 13:21:25.437770: step: 1048/529, loss: 0.004591560456901789 2023-01-21 13:21:26.607068: step: 1052/529, loss: -2.365112231927924e-05 2023-01-21 13:21:27.749960: step: 1056/529, loss: 4.348754737293348e-05 2023-01-21 13:21:28.910851: step: 1060/529, loss: 0.0002006530703511089 2023-01-21 13:21:30.099250: step: 1064/529, loss: 0.00022516251192428172 2023-01-21 13:21:31.236293: step: 1068/529, loss: 0.037786293774843216 2023-01-21 13:21:32.340105: step: 1072/529, loss: 0.0012212276924401522 2023-01-21 13:21:33.467961: step: 1076/529, loss: 0.00016450881958007812 2023-01-21 13:21:34.584274: step: 1080/529, loss: 0.0004562377871479839 2023-01-21 13:21:35.769353: step: 1084/529, loss: 7.82012921263231e-06 2023-01-21 13:21:36.903571: step: 1088/529, loss: 3.5095217754133046e-05 2023-01-21 13:21:38.039269: step: 1092/529, loss: 0.017914295196533203 2023-01-21 13:21:39.167429: step: 1096/529, loss: -8.58306884765625e-06 2023-01-21 13:21:40.297173: step: 1100/529, loss: 9.32693510549143e-05 2023-01-21 13:21:41.422954: step: 1104/529, loss: 0.0004447936953511089 2023-01-21 13:21:42.571488: step: 1108/529, loss: 0.00035190582275390625 2023-01-21 13:21:43.724573: step: 1112/529, loss: -1.7499924069852568e-05 2023-01-21 13:21:44.875178: step: 1116/529, loss: -3.814697322468419e-07 2023-01-21 13:21:46.032858: step: 1120/529, loss: 0.00222187046892941 2023-01-21 13:21:47.164591: step: 1124/529, loss: 0.014529800973832607 2023-01-21 13:21:48.303054: step: 1128/529, loss: -9.91821252682712e-06 2023-01-21 13:21:49.417088: step: 1132/529, loss: 1.4972687495173886e-05 2023-01-21 13:21:50.570319: step: 1136/529, loss: 0.01158981304615736 2023-01-21 13:21:51.689618: step: 1140/529, loss: 0.012137985788285732 2023-01-21 13:21:52.852082: step: 1144/529, loss: 0.00037975312443450093 2023-01-21 13:21:54.019389: step: 1148/529, loss: 0.0024137496948242188 2023-01-21 13:21:55.167529: step: 1152/529, loss: 0.00871572457253933 2023-01-21 13:21:56.320252: step: 1156/529, loss: 0.0002229690580861643 2023-01-21 13:21:57.447956: step: 1160/529, loss: 0.0304107666015625 2023-01-21 13:21:58.589023: step: 1164/529, loss: 1.3637542906508315e-05 2023-01-21 13:21:59.712420: step: 1168/529, loss: 5.14984139954322e-06 2023-01-21 13:22:00.867685: step: 1172/529, loss: 8.611679368186742e-05 2023-01-21 13:22:02.006765: step: 1176/529, loss: 1.4591216313419864e-05 2023-01-21 13:22:03.176503: step: 1180/529, loss: 2.6226043701171875e-05 2023-01-21 13:22:04.321385: step: 1184/529, loss: 0.0064296722412109375 2023-01-21 13:22:05.466002: step: 1188/529, loss: 0.009478569030761719 2023-01-21 13:22:06.603049: step: 1192/529, loss: 7.667541649425402e-05 2023-01-21 13:22:07.728768: step: 1196/529, loss: 0.03364238515496254 2023-01-21 13:22:08.903302: step: 1200/529, loss: 4.014969090349041e-05 2023-01-21 13:22:10.046829: step: 1204/529, loss: 0.0008069992181845009 2023-01-21 13:22:11.185852: step: 1208/529, loss: 0.057192422449588776 2023-01-21 13:22:12.309792: step: 1212/529, loss: 0.05859978124499321 2023-01-21 13:22:13.434378: step: 1216/529, loss: 0.0009019851568154991 2023-01-21 13:22:14.566067: step: 1220/529, loss: 0.0013179780216887593 2023-01-21 13:22:15.707029: step: 1224/529, loss: 3.814697265625e-06 2023-01-21 13:22:16.827763: step: 1228/529, loss: 0.01581592671573162 2023-01-21 13:22:17.936041: step: 1232/529, loss: -2.8610230629055877e-07 2023-01-21 13:22:19.090218: step: 1236/529, loss: 0.13412819802761078 2023-01-21 13:22:20.229278: step: 1240/529, loss: 0.016009999439120293 2023-01-21 13:22:21.332433: step: 1244/529, loss: -3.910064697265625e-05 2023-01-21 13:22:22.465030: step: 1248/529, loss: 0.0004861831839662045 2023-01-21 13:22:23.613144: step: 1252/529, loss: 0.00756072998046875 2023-01-21 13:22:24.732944: step: 1256/529, loss: 0.1195463165640831 2023-01-21 13:22:25.864559: step: 1260/529, loss: 0.013532351702451706 2023-01-21 13:22:27.002312: step: 1264/529, loss: 3.814697265625e-06 2023-01-21 13:22:28.129265: step: 1268/529, loss: 0.00022134781465865672 2023-01-21 13:22:29.268679: step: 1272/529, loss: 0.002338123507797718 2023-01-21 13:22:30.402668: step: 1276/529, loss: 0.0007865905645303428 2023-01-21 13:22:31.534746: step: 1280/529, loss: 0.006599903106689453 2023-01-21 13:22:32.663239: step: 1284/529, loss: 0.00030765534029342234 2023-01-21 13:22:33.806364: step: 1288/529, loss: 0.09362554550170898 2023-01-21 13:22:34.932652: step: 1292/529, loss: 0.011049842461943626 2023-01-21 13:22:36.079790: step: 1296/529, loss: 0.003641319228336215 2023-01-21 13:22:37.200268: step: 1300/529, loss: 2.1934511096333154e-06 2023-01-21 13:22:38.337237: step: 1304/529, loss: 0.002183532575145364 2023-01-21 13:22:39.459872: step: 1308/529, loss: 0.00498809851706028 2023-01-21 13:22:40.617825: step: 1312/529, loss: 6.65664701955393e-05 2023-01-21 13:22:41.741304: step: 1316/529, loss: 5.626678557746345e-06 2023-01-21 13:22:42.872960: step: 1320/529, loss: 0.042349621653556824 2023-01-21 13:22:44.008585: step: 1324/529, loss: 0.0400175079703331 2023-01-21 13:22:45.166686: step: 1328/529, loss: 0.0001619339018361643 2023-01-21 13:22:46.330551: step: 1332/529, loss: 0.00034494398278184235 2023-01-21 13:22:47.438013: step: 1336/529, loss: 0.0008558273548260331 2023-01-21 13:22:48.556877: step: 1340/529, loss: 0.00022430421086028218 2023-01-21 13:22:49.685202: step: 1344/529, loss: 7.514953904319555e-05 2023-01-21 13:22:50.816951: step: 1348/529, loss: 0.01218109205365181 2023-01-21 13:22:51.950654: step: 1352/529, loss: 0.003837204072624445 2023-01-21 13:22:53.086683: step: 1356/529, loss: 7.314682443393394e-05 2023-01-21 13:22:54.231968: step: 1360/529, loss: 0.00014553070650435984 2023-01-21 13:22:55.358657: step: 1364/529, loss: 0.0011056900257244706 2023-01-21 13:22:56.529247: step: 1368/529, loss: 0.007794762030243874 2023-01-21 13:22:57.678793: step: 1372/529, loss: 0.033864784985780716 2023-01-21 13:22:58.827437: step: 1376/529, loss: 0.0004436493036337197 2023-01-21 13:22:59.945397: step: 1380/529, loss: 0.0004638671816792339 2023-01-21 13:23:01.088314: step: 1384/529, loss: 0.0018686294788494706 2023-01-21 13:23:02.214591: step: 1388/529, loss: 0.029398633167147636 2023-01-21 13:23:03.330371: step: 1392/529, loss: 3.509521411615424e-05 2023-01-21 13:23:04.438056: step: 1396/529, loss: 0.009296417236328125 2023-01-21 13:23:05.539958: step: 1400/529, loss: 0.00023374556622002274 2023-01-21 13:23:06.685538: step: 1404/529, loss: 0.0009126663790084422 2023-01-21 13:23:07.808357: step: 1408/529, loss: 0.0002902984560932964 2023-01-21 13:23:08.955739: step: 1412/529, loss: 0.0008914947393350303 2023-01-21 13:23:10.088559: step: 1416/529, loss: 0.36139461398124695 2023-01-21 13:23:11.187570: step: 1420/529, loss: 0.00016822815814521164 2023-01-21 13:23:12.325788: step: 1424/529, loss: 0.2363910675048828 2023-01-21 13:23:13.447383: step: 1428/529, loss: 0.000152587890625 2023-01-21 13:23:14.594574: step: 1432/529, loss: 0.01467132568359375 2023-01-21 13:23:15.706301: step: 1436/529, loss: 0.0033197880256921053 2023-01-21 13:23:16.821758: step: 1440/529, loss: 0.0002608299255371094 2023-01-21 13:23:17.964465: step: 1444/529, loss: 0.001408481621183455 2023-01-21 13:23:19.100713: step: 1448/529, loss: 0.0007710456848144531 2023-01-21 13:23:20.246400: step: 1452/529, loss: 0.00029544829158112407 2023-01-21 13:23:21.362682: step: 1456/529, loss: 3.3950807846849784e-05 2023-01-21 13:23:22.494997: step: 1460/529, loss: 6.723403930664062e-05 2023-01-21 13:23:23.640938: step: 1464/529, loss: 1.144409225162235e-06 2023-01-21 13:23:24.770887: step: 1468/529, loss: 0.012156200595200062 2023-01-21 13:23:25.918990: step: 1472/529, loss: 0.0005641937023028731 2023-01-21 13:23:27.053174: step: 1476/529, loss: 4.1675568354548886e-05 2023-01-21 13:23:28.186475: step: 1480/529, loss: 0.00102148053701967 2023-01-21 13:23:29.347748: step: 1484/529, loss: 0.004835319705307484 2023-01-21 13:23:30.460925: step: 1488/529, loss: 0.003568506333976984 2023-01-21 13:23:31.595962: step: 1492/529, loss: 0.04547462239861488 2023-01-21 13:23:32.713411: step: 1496/529, loss: 5.130767749506049e-05 2023-01-21 13:23:33.828782: step: 1500/529, loss: 5.054473876953125e-05 2023-01-21 13:23:34.947481: step: 1504/529, loss: 0.00032062531681731343 2023-01-21 13:23:36.113560: step: 1508/529, loss: 2.7704239982995205e-05 2023-01-21 13:23:37.244208: step: 1512/529, loss: 0.0053046224638819695 2023-01-21 13:23:38.407468: step: 1516/529, loss: 0.00011329651169944555 2023-01-21 13:23:39.539997: step: 1520/529, loss: 0.003033256623893976 2023-01-21 13:23:40.706972: step: 1524/529, loss: 0.002505683805793524 2023-01-21 13:23:41.873766: step: 1528/529, loss: 0.0015397071838378906 2023-01-21 13:23:43.016947: step: 1532/529, loss: 0.04596518352627754 2023-01-21 13:23:44.133818: step: 1536/529, loss: 0.0005479812389239669 2023-01-21 13:23:45.292719: step: 1540/529, loss: 0.0005599975702352822 2023-01-21 13:23:46.420810: step: 1544/529, loss: 0.010690880008041859 2023-01-21 13:23:47.564694: step: 1548/529, loss: 0.00918655376881361 2023-01-21 13:23:48.713103: step: 1552/529, loss: 0.004045295994728804 2023-01-21 13:23:49.867095: step: 1556/529, loss: 0.00359001150354743 2023-01-21 13:23:51.008959: step: 1560/529, loss: 0.00015907287888694555 2023-01-21 13:23:52.137428: step: 1564/529, loss: 0.001544094062410295 2023-01-21 13:23:53.298583: step: 1568/529, loss: 0.0002754211309365928 2023-01-21 13:23:54.433508: step: 1572/529, loss: 0.012155341915786266 2023-01-21 13:23:55.599560: step: 1576/529, loss: 0.008628081530332565 2023-01-21 13:23:56.758025: step: 1580/529, loss: 0.0001415252627339214 2023-01-21 13:23:57.881366: step: 1584/529, loss: 7.057189577608369e-06 2023-01-21 13:23:59.048873: step: 1588/529, loss: 9.117127046920359e-05 2023-01-21 13:24:00.178412: step: 1592/529, loss: 0.0012561797630041838 2023-01-21 13:24:01.324077: step: 1596/529, loss: 0.004355430603027344 2023-01-21 13:24:02.467029: step: 1600/529, loss: 0.014101123437285423 2023-01-21 13:24:03.595661: step: 1604/529, loss: 2.269744800287299e-05 2023-01-21 13:24:04.745857: step: 1608/529, loss: 0.001446199486963451 2023-01-21 13:24:05.880430: step: 1612/529, loss: 0.0010208607418462634 2023-01-21 13:24:07.031094: step: 1616/529, loss: 8.39233416627394e-06 2023-01-21 13:24:08.150800: step: 1620/529, loss: 0.00016055107698775828 2023-01-21 13:24:09.281434: step: 1624/529, loss: 0.00603408832103014 2023-01-21 13:24:10.403824: step: 1628/529, loss: -1.3351441339182202e-06 2023-01-21 13:24:11.544726: step: 1632/529, loss: 0.010101891122758389 2023-01-21 13:24:12.686724: step: 1636/529, loss: 4.8255922592943534e-05 2023-01-21 13:24:13.817036: step: 1640/529, loss: 0.09000034630298615 2023-01-21 13:24:14.954552: step: 1644/529, loss: 1.2683868590102065e-05 2023-01-21 13:24:16.079253: step: 1648/529, loss: 0.020532798022031784 2023-01-21 13:24:17.179471: step: 1652/529, loss: 0.00018777848163153976 2023-01-21 13:24:18.317772: step: 1656/529, loss: 0.018397904932498932 2023-01-21 13:24:19.435424: step: 1660/529, loss: 0.00014038087101653218 2023-01-21 13:24:20.560103: step: 1664/529, loss: 0.00021247864060569555 2023-01-21 13:24:21.740115: step: 1668/529, loss: 4.596710277837701e-05 2023-01-21 13:24:22.898263: step: 1672/529, loss: 0.0013607979053631425 2023-01-21 13:24:24.026425: step: 1676/529, loss: 0.004532814025878906 2023-01-21 13:24:25.165428: step: 1680/529, loss: 0.04919777065515518 2023-01-21 13:24:26.310014: step: 1684/529, loss: 5.950927879894152e-05 2023-01-21 13:24:27.430204: step: 1688/529, loss: 0.005524826236069202 2023-01-21 13:24:28.570659: step: 1692/529, loss: 0.00240249652415514 2023-01-21 13:24:29.697067: step: 1696/529, loss: 0.05209184065461159 2023-01-21 13:24:30.833749: step: 1700/529, loss: 1.0395049685030244e-05 2023-01-21 13:24:31.997674: step: 1704/529, loss: 0.00014963149442337453 2023-01-21 13:24:33.113502: step: 1708/529, loss: 0.005444812588393688 2023-01-21 13:24:34.265840: step: 1712/529, loss: 2.517700158932712e-05 2023-01-21 13:24:35.403373: step: 1716/529, loss: 0.0003444671747274697 2023-01-21 13:24:36.520844: step: 1720/529, loss: 0.0006761550903320312 2023-01-21 13:24:37.661587: step: 1724/529, loss: 0.00900116004049778 2023-01-21 13:24:38.798248: step: 1728/529, loss: 4.291534423828125e-06 2023-01-21 13:24:39.941279: step: 1732/529, loss: 0.01634521596133709 2023-01-21 13:24:41.068943: step: 1736/529, loss: 0.00071125035174191 2023-01-21 13:24:42.242415: step: 1740/529, loss: 0.04398870840668678 2023-01-21 13:24:43.372960: step: 1744/529, loss: 0.030103780329227448 2023-01-21 13:24:44.489772: step: 1748/529, loss: 0.14822901785373688 2023-01-21 13:24:45.602016: step: 1752/529, loss: 5.722046353184851e-06 2023-01-21 13:24:46.765465: step: 1756/529, loss: 0.00106048583984375 2023-01-21 13:24:47.908144: step: 1760/529, loss: 0.0007371903047896922 2023-01-21 13:24:49.027703: step: 1764/529, loss: 0.0029416086617857218 2023-01-21 13:24:50.159643: step: 1768/529, loss: 0.00013074875460006297 2023-01-21 13:24:51.322972: step: 1772/529, loss: 0.0026731491088867188 2023-01-21 13:24:52.464493: step: 1776/529, loss: 4.00543194700731e-06 2023-01-21 13:24:53.588780: step: 1780/529, loss: 0.0009098052978515625 2023-01-21 13:24:54.709968: step: 1784/529, loss: 0.011314011178910732 2023-01-21 13:24:55.814088: step: 1788/529, loss: 0.0007757187122479081 2023-01-21 13:24:56.955677: step: 1792/529, loss: 0.00011749268014682457 2023-01-21 13:24:58.077777: step: 1796/529, loss: 0.0004980087396688759 2023-01-21 13:24:59.212696: step: 1800/529, loss: 0.00014314652071334422 2023-01-21 13:25:00.368108: step: 1804/529, loss: 2.250671423098538e-05 2023-01-21 13:25:01.518286: step: 1808/529, loss: 7.047652616165578e-05 2023-01-21 13:25:02.676182: step: 1812/529, loss: 0.0013103485107421875 2023-01-21 13:25:03.792316: step: 1816/529, loss: 1.773834264895413e-05 2023-01-21 13:25:04.916224: step: 1820/529, loss: -3.7670135952794226e-06 2023-01-21 13:25:06.061052: step: 1824/529, loss: 0.08901538699865341 2023-01-21 13:25:07.188246: step: 1828/529, loss: 0.00022029878164175898 2023-01-21 13:25:08.331424: step: 1832/529, loss: 0.00161914830096066 2023-01-21 13:25:09.487084: step: 1836/529, loss: 7.648468454135582e-05 2023-01-21 13:25:10.631406: step: 1840/529, loss: 0.024843288585543633 2023-01-21 13:25:11.787101: step: 1844/529, loss: 3.2997133530443534e-05 2023-01-21 13:25:12.907421: step: 1848/529, loss: 0.0012079238658770919 2023-01-21 13:25:14.024538: step: 1852/529, loss: 0.00011940002150367945 2023-01-21 13:25:15.177301: step: 1856/529, loss: 0.0010656357044354081 2023-01-21 13:25:16.343160: step: 1860/529, loss: 1.3949804306030273 2023-01-21 13:25:17.478916: step: 1864/529, loss: 0.000669288681820035 2023-01-21 13:25:18.647292: step: 1868/529, loss: 0.0031375885009765625 2023-01-21 13:25:19.785474: step: 1872/529, loss: 0.031613539904356 2023-01-21 13:25:20.910824: step: 1876/529, loss: 3.623962356869015e-06 2023-01-21 13:25:22.042440: step: 1880/529, loss: 0.0003383636649232358 2023-01-21 13:25:23.162605: step: 1884/529, loss: 0.00025238993111997843 2023-01-21 13:25:24.301920: step: 1888/529, loss: 0.0007088661659508944 2023-01-21 13:25:25.442979: step: 1892/529, loss: 0.0006690025329589844 2023-01-21 13:25:26.600074: step: 1896/529, loss: 0.0005206108326092362 2023-01-21 13:25:27.711334: step: 1900/529, loss: 5.064010474598035e-05 2023-01-21 13:25:28.828330: step: 1904/529, loss: 1.0165572348341811e-05 2023-01-21 13:25:29.979780: step: 1908/529, loss: 0.0011874199844896793 2023-01-21 13:25:31.123265: step: 1912/529, loss: 0.0009941102471202612 2023-01-21 13:25:32.264682: step: 1916/529, loss: 0.00010929107520496473 2023-01-21 13:25:33.393541: step: 1920/529, loss: 0.00044016839819960296 2023-01-21 13:25:34.540490: step: 1924/529, loss: 0.0017185211181640625 2023-01-21 13:25:35.677091: step: 1928/529, loss: 0.0020595071837306023 2023-01-21 13:25:36.824431: step: 1932/529, loss: 3.728866431629285e-05 2023-01-21 13:25:37.956564: step: 1936/529, loss: 1.8453598386258818e-05 2023-01-21 13:25:39.081911: step: 1940/529, loss: 0.0010535240871831775 2023-01-21 13:25:40.209849: step: 1944/529, loss: 0.011309958063066006 2023-01-21 13:25:41.328974: step: 1948/529, loss: 0.004026985261589289 2023-01-21 13:25:42.481524: step: 1952/529, loss: 0.00011882782564498484 2023-01-21 13:25:43.649147: step: 1956/529, loss: 6.31332368357107e-05 2023-01-21 13:25:44.780505: step: 1960/529, loss: 8.0108642578125e-05 2023-01-21 13:25:45.909792: step: 1964/529, loss: 0.001093912054784596 2023-01-21 13:25:47.047747: step: 1968/529, loss: 0.14452219009399414 2023-01-21 13:25:48.192736: step: 1972/529, loss: 0.02083406411111355 2023-01-21 13:25:49.349658: step: 1976/529, loss: 0.016637230291962624 2023-01-21 13:25:50.476742: step: 1980/529, loss: 0.032593440264463425 2023-01-21 13:25:51.616579: step: 1984/529, loss: 0.0014526366721838713 2023-01-21 13:25:52.747082: step: 1988/529, loss: 0.009725459851324558 2023-01-21 13:25:53.895859: step: 1992/529, loss: 0.00320606236346066 2023-01-21 13:25:55.017467: step: 1996/529, loss: 0.010757159441709518 2023-01-21 13:25:56.178672: step: 2000/529, loss: 0.03268565982580185 2023-01-21 13:25:57.320558: step: 2004/529, loss: 0.0001977920619538054 2023-01-21 13:25:58.460762: step: 2008/529, loss: 0.04256686940789223 2023-01-21 13:25:59.609437: step: 2012/529, loss: 0.039610959589481354 2023-01-21 13:26:00.728053: step: 2016/529, loss: 0.003604888916015625 2023-01-21 13:26:01.886072: step: 2020/529, loss: 1.621246337890625e-05 2023-01-21 13:26:03.055383: step: 2024/529, loss: 0.0339779369533062 2023-01-21 13:26:04.210934: step: 2028/529, loss: 7.390976679744199e-05 2023-01-21 13:26:05.334665: step: 2032/529, loss: 0.010361862368881702 2023-01-21 13:26:06.457552: step: 2036/529, loss: 0.11191482841968536 2023-01-21 13:26:07.558783: step: 2040/529, loss: 0.00010452271089889109 2023-01-21 13:26:08.685369: step: 2044/529, loss: 0.013841485604643822 2023-01-21 13:26:09.811796: step: 2048/529, loss: 0.0001349449303233996 2023-01-21 13:26:10.962062: step: 2052/529, loss: 0.0004261970752850175 2023-01-21 13:26:12.104925: step: 2056/529, loss: 0.06896381080150604 2023-01-21 13:26:13.260050: step: 2060/529, loss: 5.0926206313306466e-05 2023-01-21 13:26:14.386298: step: 2064/529, loss: 0.0013132095336914062 2023-01-21 13:26:15.498562: step: 2068/529, loss: 6.29425048828125e-05 2023-01-21 13:26:16.640235: step: 2072/529, loss: 0.012125778011977673 2023-01-21 13:26:17.777823: step: 2076/529, loss: 0.012499809265136719 2023-01-21 13:26:18.928430: step: 2080/529, loss: 0.00036907196044921875 2023-01-21 13:26:20.096103: step: 2084/529, loss: 0.00013608932204078883 2023-01-21 13:26:21.226750: step: 2088/529, loss: 0.03581704944372177 2023-01-21 13:26:22.375303: step: 2092/529, loss: 0.018338393419981003 2023-01-21 13:26:23.508957: step: 2096/529, loss: 0.00014157296391204 2023-01-21 13:26:24.637091: step: 2100/529, loss: 0.0076236724853515625 2023-01-21 13:26:25.770689: step: 2104/529, loss: 8.0108642578125e-05 2023-01-21 13:26:26.896133: step: 2108/529, loss: 0.00025405886117368937 2023-01-21 13:26:28.000969: step: 2112/529, loss: 2.47955313170678e-06 2023-01-21 13:26:29.122033: step: 2116/529, loss: 0.00010061264038085938 ================================================== Loss: 0.016 -------------------- Dev: {'event': {'p': 0.6104944500504541, 'r': 0.8055925432756325, 'f1': 0.6946039035591274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Test: {'event': {'p': 0.61996336996337, 'r': 0.7844727694090382, 'f1': 0.6925831202046036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Chinese: {'event': {'p': 0.611764705882353, 'r': 0.9629629629629629, 'f1': 0.748201438848921}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Korean: {'event': {'p': 0.6721311475409836, 'r': 0.6507936507936508, 'f1': 0.6612903225806451}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6096033402922756, 'r': 0.7776298268974701, 'f1': 0.6834406085430076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Eng Test for Chinese: {'event': {'p': 0.6149954001839927, 'r': 0.7746234067207416, 'f1': 0.6856410256410257}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Sample Chinese: {'event': {'p': 0.6375, 'r': 0.9444444444444444, 'f1': 0.7611940298507462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} -------------------- Eng Dev for Korean: {'event': {'p': 0.6104944500504541, 'r': 0.8055925432756325, 'f1': 0.6946039035591274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Korean: {'event': {'p': 0.61996336996337, 'r': 0.7844727694090382, 'f1': 0.6925831202046036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Sample Korean: {'event': {'p': 0.6721311475409836, 'r': 0.6507936507936508, 'f1': 0.6612903225806451}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 28 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:27:17.136175: step: 4/529, loss: 0.0021514892578125 2023-01-21 13:27:18.280317: step: 8/529, loss: 0.009921645745635033 2023-01-21 13:27:19.421284: step: 12/529, loss: 0.00017242431931663305 2023-01-21 13:27:20.525143: step: 16/529, loss: 0.0006539822206832469 2023-01-21 13:27:21.624981: step: 20/529, loss: 0.005362081807106733 2023-01-21 13:27:22.784477: step: 24/529, loss: -4.76837158203125e-06 2023-01-21 13:27:23.890328: step: 28/529, loss: 4.1961669921875e-05 2023-01-21 13:27:25.015184: step: 32/529, loss: 0.00027360915555618703 2023-01-21 13:27:26.173005: step: 36/529, loss: 0.0034100532066076994 2023-01-21 13:27:27.299532: step: 40/529, loss: 8.964539119915571e-06 2023-01-21 13:27:28.418842: step: 44/529, loss: 0.0014062882401049137 2023-01-21 13:27:29.598001: step: 48/529, loss: 6.67572021484375e-06 2023-01-21 13:27:30.734154: step: 52/529, loss: 3.547668529790826e-05 2023-01-21 13:27:31.866757: step: 56/529, loss: 0.01944456249475479 2023-01-21 13:27:32.986564: step: 60/529, loss: 0.6814148426055908 2023-01-21 13:27:34.117276: step: 64/529, loss: 0.00012073517427779734 2023-01-21 13:27:35.246063: step: 68/529, loss: 0.0006240844959393144 2023-01-21 13:27:36.380477: step: 72/529, loss: 0.0010915755992755294 2023-01-21 13:27:37.486148: step: 76/529, loss: 0.003563118167221546 2023-01-21 13:27:38.619774: step: 80/529, loss: 6.0844424297101796e-05 2023-01-21 13:27:39.749496: step: 84/529, loss: 4.472732689464465e-05 2023-01-21 13:27:40.882668: step: 88/529, loss: 5.34057608092553e-06 2023-01-21 13:27:42.028419: step: 92/529, loss: 0.0015334129566326737 2023-01-21 13:27:43.198203: step: 96/529, loss: 0.07135754078626633 2023-01-21 13:27:44.313240: step: 100/529, loss: 8.96453821042087e-06 2023-01-21 13:27:45.468264: step: 104/529, loss: 1.8501283193472773e-05 2023-01-21 13:27:46.593020: step: 108/529, loss: 0.00018301009549759328 2023-01-21 13:27:47.722295: step: 112/529, loss: 1.33514404296875e-05 2023-01-21 13:27:48.895457: step: 116/529, loss: 4.138946678722277e-05 2023-01-21 13:27:50.065309: step: 120/529, loss: 0.003964233677834272 2023-01-21 13:27:51.205678: step: 124/529, loss: 0.0011302947532385588 2023-01-21 13:27:52.340259: step: 128/529, loss: 0.0023619653657078743 2023-01-21 13:27:53.464473: step: 132/529, loss: 0.0006250381702557206 2023-01-21 13:27:54.593811: step: 136/529, loss: -1.0776519957289565e-05 2023-01-21 13:27:55.762543: step: 140/529, loss: 0.0015142441261559725 2023-01-21 13:27:56.875849: step: 144/529, loss: 0.00010986327833961695 2023-01-21 13:27:58.025252: step: 148/529, loss: 0.0003440856817178428 2023-01-21 13:27:59.134417: step: 152/529, loss: 4.186630030744709e-05 2023-01-21 13:28:00.255254: step: 156/529, loss: 0.006328677758574486 2023-01-21 13:28:01.404168: step: 160/529, loss: 0.00047912599984556437 2023-01-21 13:28:02.543795: step: 164/529, loss: 3.0517578125e-05 2023-01-21 13:28:03.669577: step: 168/529, loss: 0.20622768998146057 2023-01-21 13:28:04.817223: step: 172/529, loss: 0.0017513275379315019 2023-01-21 13:28:05.983749: step: 176/529, loss: 3.0946732294978574e-05 2023-01-21 13:28:07.140135: step: 180/529, loss: 0.0010950089199468493 2023-01-21 13:28:08.292628: step: 184/529, loss: 0.009531163610517979 2023-01-21 13:28:09.425693: step: 188/529, loss: 9.822845458984375e-05 2023-01-21 13:28:10.568832: step: 192/529, loss: 0.0005132675287313759 2023-01-21 13:28:11.699571: step: 196/529, loss: 3.4809112548828125e-05 2023-01-21 13:28:12.832535: step: 200/529, loss: 0.024631166830658913 2023-01-21 13:28:13.968668: step: 204/529, loss: 0.0006175994640216231 2023-01-21 13:28:15.136978: step: 208/529, loss: 0.0007814407581463456 2023-01-21 13:28:16.266144: step: 212/529, loss: 0.013393021188676357 2023-01-21 13:28:17.387025: step: 216/529, loss: 0.0008969306945800781 2023-01-21 13:28:18.535685: step: 220/529, loss: 8.106231689453125e-05 2023-01-21 13:28:19.683354: step: 224/529, loss: 0.000736236572265625 2023-01-21 13:28:20.815191: step: 228/529, loss: 9.04560147318989e-05 2023-01-21 13:28:21.927594: step: 232/529, loss: 0.003286171006038785 2023-01-21 13:28:23.060888: step: 236/529, loss: 0.0001070022553903982 2023-01-21 13:28:24.184191: step: 240/529, loss: 0.0008007049327716231 2023-01-21 13:28:25.319918: step: 244/529, loss: 0.00014510155597236007 2023-01-21 13:28:26.434025: step: 248/529, loss: 0.0048881531693041325 2023-01-21 13:28:27.574730: step: 252/529, loss: 0.014361190609633923 2023-01-21 13:28:28.709578: step: 256/529, loss: 1.125335711549269e-05 2023-01-21 13:28:29.877331: step: 260/529, loss: 0.568461000919342 2023-01-21 13:28:31.013366: step: 264/529, loss: 1.3160706657799892e-05 2023-01-21 13:28:32.146653: step: 268/529, loss: 0.0057701109908521175 2023-01-21 13:28:33.275276: step: 272/529, loss: 9.250640869140625e-05 2023-01-21 13:28:34.398494: step: 276/529, loss: 0.0020244598854333162 2023-01-21 13:28:35.570825: step: 280/529, loss: 0.012895154766738415 2023-01-21 13:28:36.703131: step: 284/529, loss: 0.03386859968304634 2023-01-21 13:28:37.828774: step: 288/529, loss: 0.01651744917035103 2023-01-21 13:28:38.950146: step: 292/529, loss: 0.04547891393303871 2023-01-21 13:28:40.083001: step: 296/529, loss: 4.215240551275201e-05 2023-01-21 13:28:41.228225: step: 300/529, loss: 0.010183095932006836 2023-01-21 13:28:42.386278: step: 304/529, loss: 0.001212692353874445 2023-01-21 13:28:43.522725: step: 308/529, loss: 0.032172299921512604 2023-01-21 13:28:44.667856: step: 312/529, loss: 0.0002727508544921875 2023-01-21 13:28:45.816437: step: 316/529, loss: 0.07113972306251526 2023-01-21 13:28:47.008696: step: 320/529, loss: 0.00016460419283248484 2023-01-21 13:28:48.153600: step: 324/529, loss: 0.005909538362175226 2023-01-21 13:28:49.270846: step: 328/529, loss: 0.2711924612522125 2023-01-21 13:28:50.388962: step: 332/529, loss: 3.8337704609148204e-05 2023-01-21 13:28:51.565976: step: 336/529, loss: 0.004338264465332031 2023-01-21 13:28:52.709175: step: 340/529, loss: -4.00543194700731e-06 2023-01-21 13:28:53.836799: step: 344/529, loss: -3.14712519866589e-06 2023-01-21 13:28:54.938709: step: 348/529, loss: 0.007489395327866077 2023-01-21 13:28:56.049295: step: 352/529, loss: -1.5258789289873675e-06 2023-01-21 13:28:57.159167: step: 356/529, loss: 0.0009853363735601306 2023-01-21 13:28:58.297661: step: 360/529, loss: 0.0005168914794921875 2023-01-21 13:28:59.426517: step: 364/529, loss: 0.009820557199418545 2023-01-21 13:29:00.564720: step: 368/529, loss: 0.005489349365234375 2023-01-21 13:29:01.704799: step: 372/529, loss: 0.002441024873405695 2023-01-21 13:29:02.815027: step: 376/529, loss: 0.0020363808143883944 2023-01-21 13:29:03.988876: step: 380/529, loss: 0.00018877982802223414 2023-01-21 13:29:05.113385: step: 384/529, loss: 0.0001125335693359375 2023-01-21 13:29:06.265298: step: 388/529, loss: 0.0017709732055664062 2023-01-21 13:29:07.390880: step: 392/529, loss: 0.000159788120072335 2023-01-21 13:29:08.507435: step: 396/529, loss: 0.0001983642578125 2023-01-21 13:29:09.622668: step: 400/529, loss: 1.9464597702026367 2023-01-21 13:29:10.754319: step: 404/529, loss: 0.0001008987455861643 2023-01-21 13:29:11.883127: step: 408/529, loss: -8.20159948489163e-06 2023-01-21 13:29:12.986575: step: 412/529, loss: 0.0005866051069460809 2023-01-21 13:29:14.092131: step: 416/529, loss: 0.005843830294907093 2023-01-21 13:29:15.250805: step: 420/529, loss: 0.017465878278017044 2023-01-21 13:29:16.373456: step: 424/529, loss: 2.250671423098538e-05 2023-01-21 13:29:17.508524: step: 428/529, loss: 0.0001617431698832661 2023-01-21 13:29:18.614396: step: 432/529, loss: 3.43322744811303e-06 2023-01-21 13:29:19.717738: step: 436/529, loss: 0.0490659698843956 2023-01-21 13:29:20.867735: step: 440/529, loss: 4.000663830083795e-05 2023-01-21 13:29:21.991012: step: 444/529, loss: 0.015586186200380325 2023-01-21 13:29:23.110365: step: 448/529, loss: 0.004702854435890913 2023-01-21 13:29:24.273241: step: 452/529, loss: 0.0007944107055664062 2023-01-21 13:29:25.405182: step: 456/529, loss: 0.00414962787181139 2023-01-21 13:29:26.600789: step: 460/529, loss: 0.0032614709343761206 2023-01-21 13:29:27.723680: step: 464/529, loss: 0.007906246930360794 2023-01-21 13:29:28.860212: step: 468/529, loss: 0.0007512092124670744 2023-01-21 13:29:29.959837: step: 472/529, loss: 0.0030537606216967106 2023-01-21 13:29:31.096385: step: 476/529, loss: 0.008762264624238014 2023-01-21 13:29:32.229031: step: 480/529, loss: 0.0009031295776367188 2023-01-21 13:29:33.359987: step: 484/529, loss: 0.0018049241043627262 2023-01-21 13:29:34.507988: step: 488/529, loss: 0.020003559067845345 2023-01-21 13:29:35.633290: step: 492/529, loss: 0.02207164652645588 2023-01-21 13:29:36.806116: step: 496/529, loss: 0.00011959076073253527 2023-01-21 13:29:37.944618: step: 500/529, loss: 5.14984139954322e-06 2023-01-21 13:29:39.055408: step: 504/529, loss: 0.06974411010742188 2023-01-21 13:29:40.191043: step: 508/529, loss: 0.004493141081184149 2023-01-21 13:29:41.325620: step: 512/529, loss: 0.00010161400132346898 2023-01-21 13:29:42.479167: step: 516/529, loss: 0.005316448397934437 2023-01-21 13:29:43.598102: step: 520/529, loss: 2.0313262211857364e-05 2023-01-21 13:29:44.744437: step: 524/529, loss: 1.392364538332913e-05 2023-01-21 13:29:45.901502: step: 528/529, loss: 4.100799742445815e-06 2023-01-21 13:29:47.061767: step: 532/529, loss: 0.018885230645537376 2023-01-21 13:29:48.191301: step: 536/529, loss: 0.0001813888520700857 2023-01-21 13:29:49.317074: step: 540/529, loss: 0.00017070770263671875 2023-01-21 13:29:50.466898: step: 544/529, loss: 0.0009386062738485634 2023-01-21 13:29:51.599577: step: 548/529, loss: 0.00023727417283225805 2023-01-21 13:29:52.751115: step: 552/529, loss: 0.0032910346053540707 2023-01-21 13:29:53.892929: step: 556/529, loss: 0.06110534816980362 2023-01-21 13:29:55.051303: step: 560/529, loss: 0.0006925583002157509 2023-01-21 13:29:56.174846: step: 564/529, loss: 3.910064606316155e-06 2023-01-21 13:29:57.315630: step: 568/529, loss: 0.004631233401596546 2023-01-21 13:29:58.452898: step: 572/529, loss: 0.018642233684659004 2023-01-21 13:29:59.595713: step: 576/529, loss: 7.62939453125e-05 2023-01-21 13:30:00.713809: step: 580/529, loss: 0.00036258698673918843 2023-01-21 13:30:01.864362: step: 584/529, loss: 0.00038814544677734375 2023-01-21 13:30:03.040844: step: 588/529, loss: 0.00036115647526457906 2023-01-21 13:30:04.191545: step: 592/529, loss: 0.0002307891845703125 2023-01-21 13:30:05.330654: step: 596/529, loss: 0.004060172941535711 2023-01-21 13:30:06.464319: step: 600/529, loss: 0.5602739453315735 2023-01-21 13:30:07.593006: step: 604/529, loss: -3.52859501617786e-06 2023-01-21 13:30:08.747972: step: 608/529, loss: 0.0010477066971361637 2023-01-21 13:30:09.875806: step: 612/529, loss: 0.0002510070917196572 2023-01-21 13:30:11.009413: step: 616/529, loss: 0.005847168155014515 2023-01-21 13:30:12.145874: step: 620/529, loss: 3.814697265625e-06 2023-01-21 13:30:13.281256: step: 624/529, loss: 0.00115623464807868 2023-01-21 13:30:14.407024: step: 628/529, loss: 0.0001163482666015625 2023-01-21 13:30:15.547815: step: 632/529, loss: 0.0206132885068655 2023-01-21 13:30:16.664179: step: 636/529, loss: 5.092620995128527e-05 2023-01-21 13:30:17.819578: step: 640/529, loss: 0.006295967381447554 2023-01-21 13:30:18.980067: step: 644/529, loss: 0.00027112962561659515 2023-01-21 13:30:20.125219: step: 648/529, loss: 0.0006616592290811241 2023-01-21 13:30:21.247681: step: 652/529, loss: 0.030700301751494408 2023-01-21 13:30:22.409166: step: 656/529, loss: 0.0003421783621888608 2023-01-21 13:30:23.541304: step: 660/529, loss: -3.2901763916015625e-05 2023-01-21 13:30:24.688035: step: 664/529, loss: 0.0029228210914880037 2023-01-21 13:30:25.856890: step: 668/529, loss: 0.015261746011674404 2023-01-21 13:30:26.979888: step: 672/529, loss: 7.62939453125e-05 2023-01-21 13:30:28.113617: step: 676/529, loss: 0.04491777718067169 2023-01-21 13:30:29.257543: step: 680/529, loss: 0.0007167339790612459 2023-01-21 13:30:30.394174: step: 684/529, loss: 4.6157834731275216e-05 2023-01-21 13:30:31.525143: step: 688/529, loss: 9.331703768111765e-05 2023-01-21 13:30:32.667114: step: 692/529, loss: 0.0006875991821289062 2023-01-21 13:30:33.835787: step: 696/529, loss: 0.0033746720291674137 2023-01-21 13:30:34.978932: step: 700/529, loss: 0.00663800211623311 2023-01-21 13:30:36.077448: step: 704/529, loss: 0.0011798858176916838 2023-01-21 13:30:37.203888: step: 708/529, loss: 0.008669758215546608 2023-01-21 13:30:38.372762: step: 712/529, loss: 0.0008595467079430819 2023-01-21 13:30:39.526954: step: 716/529, loss: 0.05906028673052788 2023-01-21 13:30:40.650825: step: 720/529, loss: 1.5354156857938506e-05 2023-01-21 13:30:41.798549: step: 724/529, loss: 0.06484727561473846 2023-01-21 13:30:42.921030: step: 728/529, loss: 0.0010814666748046875 2023-01-21 13:30:44.052863: step: 732/529, loss: 7.433891005348414e-05 2023-01-21 13:30:45.179338: step: 736/529, loss: 0.005944633856415749 2023-01-21 13:30:46.336298: step: 740/529, loss: 0.0698789581656456 2023-01-21 13:30:47.457291: step: 744/529, loss: 5.779266211902723e-05 2023-01-21 13:30:48.563234: step: 748/529, loss: 2.1934511096333154e-06 2023-01-21 13:30:49.696425: step: 752/529, loss: 0.008106708526611328 2023-01-21 13:30:50.822774: step: 756/529, loss: 0.006346511654555798 2023-01-21 13:30:51.947089: step: 760/529, loss: 0.0013070106506347656 2023-01-21 13:30:53.081767: step: 764/529, loss: 0.0001321792515227571 2023-01-21 13:30:54.228454: step: 768/529, loss: 0.03649485111236572 2023-01-21 13:30:55.349720: step: 772/529, loss: 3.623962356869015e-06 2023-01-21 13:30:56.511114: step: 776/529, loss: 0.006076431833207607 2023-01-21 13:30:57.673744: step: 780/529, loss: 0.00015039443678688258 2023-01-21 13:30:58.833305: step: 784/529, loss: 2.708435022213962e-05 2023-01-21 13:30:59.973972: step: 788/529, loss: 0.00036296845064498484 2023-01-21 13:31:01.109212: step: 792/529, loss: 1.1207162970094942e-05 2023-01-21 13:31:02.205702: step: 796/529, loss: 0.01554346177726984 2023-01-21 13:31:03.332703: step: 800/529, loss: 0.0014945983421057463 2023-01-21 13:31:04.488978: step: 804/529, loss: 1.1920928955078125e-05 2023-01-21 13:31:05.652339: step: 808/529, loss: 0.0004927635309286416 2023-01-21 13:31:06.801436: step: 812/529, loss: 0.012382506392896175 2023-01-21 13:31:07.910827: step: 816/529, loss: 4.863739377469756e-06 2023-01-21 13:31:09.057289: step: 820/529, loss: 0.0037783621810376644 2023-01-21 13:31:10.194147: step: 824/529, loss: 0.0007633209461346269 2023-01-21 13:31:11.332344: step: 828/529, loss: 0.00011577607074286789 2023-01-21 13:31:12.469736: step: 832/529, loss: 0.027956390753388405 2023-01-21 13:31:13.588440: step: 836/529, loss: 0.010231208987534046 2023-01-21 13:31:14.723177: step: 840/529, loss: 0.0005560875288210809 2023-01-21 13:31:15.859065: step: 844/529, loss: 0.014875221066176891 2023-01-21 13:31:17.022427: step: 848/529, loss: 0.02590932883322239 2023-01-21 13:31:18.148102: step: 852/529, loss: 0.02759408950805664 2023-01-21 13:31:19.274201: step: 856/529, loss: 0.0012287141289561987 2023-01-21 13:31:20.417717: step: 860/529, loss: 0.0002777099725790322 2023-01-21 13:31:21.543279: step: 864/529, loss: 0.002498817630112171 2023-01-21 13:31:22.693219: step: 868/529, loss: 0.006489372346550226 2023-01-21 13:31:23.857451: step: 872/529, loss: 0.0006650924915447831 2023-01-21 13:31:25.023351: step: 876/529, loss: 0.015350342728197575 2023-01-21 13:31:26.168154: step: 880/529, loss: 0.0015981674659997225 2023-01-21 13:31:27.291461: step: 884/529, loss: 0.0019067764515057206 2023-01-21 13:31:28.446291: step: 888/529, loss: 0.0013967513805255294 2023-01-21 13:31:29.573965: step: 892/529, loss: 0.00019989012798760086 2023-01-21 13:31:30.708877: step: 896/529, loss: 0.001756954239681363 2023-01-21 13:31:31.823041: step: 900/529, loss: 6.179809861350805e-05 2023-01-21 13:31:32.975403: step: 904/529, loss: 0.0001277923583984375 2023-01-21 13:31:34.117777: step: 908/529, loss: 1.52587890625e-05 2023-01-21 13:31:35.249860: step: 912/529, loss: 0.003928565885871649 2023-01-21 13:31:36.403066: step: 916/529, loss: 7.629394644936838e-07 2023-01-21 13:31:37.526814: step: 920/529, loss: 0.017749978229403496 2023-01-21 13:31:38.649466: step: 924/529, loss: 2.3698807126493193e-05 2023-01-21 13:31:39.799772: step: 928/529, loss: 0.0006034850957803428 2023-01-21 13:31:40.959991: step: 932/529, loss: 0.019618988037109375 2023-01-21 13:31:42.125531: step: 936/529, loss: 0.029285335913300514 2023-01-21 13:31:43.257931: step: 940/529, loss: 0.0013185502029955387 2023-01-21 13:31:44.418659: step: 944/529, loss: 0.0022099497728049755 2023-01-21 13:31:45.566302: step: 948/529, loss: 2.8753280275850557e-05 2023-01-21 13:31:46.673684: step: 952/529, loss: 4.682540748035535e-05 2023-01-21 13:31:47.789125: step: 956/529, loss: 0.02216968685388565 2023-01-21 13:31:48.947577: step: 960/529, loss: 0.013094902038574219 2023-01-21 13:31:50.092559: step: 964/529, loss: 0.0041100503876805305 2023-01-21 13:31:51.229161: step: 968/529, loss: 0.0008859634399414062 2023-01-21 13:31:52.348808: step: 972/529, loss: 0.0006857395055703819 2023-01-21 13:31:53.498583: step: 976/529, loss: 0.0002868652227334678 2023-01-21 13:31:54.627405: step: 980/529, loss: 0.0003831863577943295 2023-01-21 13:31:55.788239: step: 984/529, loss: 0.0009023667080327868 2023-01-21 13:31:56.897295: step: 988/529, loss: 0.00024120807938743383 2023-01-21 13:31:58.035806: step: 992/529, loss: 0.0002884388086386025 2023-01-21 13:31:59.146378: step: 996/529, loss: 1.2254715329618193e-05 2023-01-21 13:32:00.284091: step: 1000/529, loss: 1.2683868590102065e-05 2023-01-21 13:32:01.420324: step: 1004/529, loss: 0.0004634857177734375 2023-01-21 13:32:02.557225: step: 1008/529, loss: 0.025662995874881744 2023-01-21 13:32:03.725055: step: 1012/529, loss: 0.008013153448700905 2023-01-21 13:32:04.865176: step: 1016/529, loss: 0.0005756377940997481 2023-01-21 13:32:05.991587: step: 1020/529, loss: 2.4890900022001006e-05 2023-01-21 13:32:07.111639: step: 1024/529, loss: 0.0002533912775106728 2023-01-21 13:32:08.230942: step: 1028/529, loss: 0.00010900497727561742 2023-01-21 13:32:09.334168: step: 1032/529, loss: 0.041082192212343216 2023-01-21 13:32:10.500935: step: 1036/529, loss: 0.1614711731672287 2023-01-21 13:32:11.628492: step: 1040/529, loss: 0.03937234729528427 2023-01-21 13:32:12.787350: step: 1044/529, loss: 0.029336929321289062 2023-01-21 13:32:13.914236: step: 1048/529, loss: 3.376007225597277e-05 2023-01-21 13:32:15.040332: step: 1052/529, loss: 0.0004026412789244205 2023-01-21 13:32:16.190332: step: 1056/529, loss: 3.9386748539982364e-05 2023-01-21 13:32:17.332154: step: 1060/529, loss: 0.048660848289728165 2023-01-21 13:32:18.463671: step: 1064/529, loss: 0.0009765148861333728 2023-01-21 13:32:19.581539: step: 1068/529, loss: 3.166198803228326e-05 2023-01-21 13:32:20.732835: step: 1072/529, loss: 4.997253563487902e-05 2023-01-21 13:32:21.863567: step: 1076/529, loss: 0.0034602168016135693 2023-01-21 13:32:22.999679: step: 1080/529, loss: 8.068085298873484e-05 2023-01-21 13:32:24.137332: step: 1084/529, loss: 9.088516526389867e-05 2023-01-21 13:32:25.264404: step: 1088/529, loss: 0.004467296879738569 2023-01-21 13:32:26.396790: step: 1092/529, loss: 2.1171568732825108e-05 2023-01-21 13:32:27.523601: step: 1096/529, loss: 0.00025653839111328125 2023-01-21 13:32:28.668104: step: 1100/529, loss: 0.05918290466070175 2023-01-21 13:32:29.784654: step: 1104/529, loss: 0.0006971359252929688 2023-01-21 13:32:30.908611: step: 1108/529, loss: 0.00013809204392600805 2023-01-21 13:32:32.038805: step: 1112/529, loss: 0.002896976424381137 2023-01-21 13:32:33.164753: step: 1116/529, loss: 4.806518700206652e-05 2023-01-21 13:32:34.264288: step: 1120/529, loss: 0.00015306472778320312 2023-01-21 13:32:35.391993: step: 1124/529, loss: 8.20159912109375e-05 2023-01-21 13:32:36.555500: step: 1128/529, loss: 0.18203915655612946 2023-01-21 13:32:37.703144: step: 1132/529, loss: 0.016922760754823685 2023-01-21 13:32:38.861676: step: 1136/529, loss: 0.00048408505972474813 2023-01-21 13:32:40.014929: step: 1140/529, loss: 0.010960960760712624 2023-01-21 13:32:41.193324: step: 1144/529, loss: 1.3685226804227568e-05 2023-01-21 13:32:42.340707: step: 1148/529, loss: 0.0016571045853197575 2023-01-21 13:32:43.451414: step: 1152/529, loss: 1.392364538332913e-05 2023-01-21 13:32:44.580897: step: 1156/529, loss: 7.786750938976184e-05 2023-01-21 13:32:45.721876: step: 1160/529, loss: 0.021187974140048027 2023-01-21 13:32:46.884936: step: 1164/529, loss: 4.806518700206652e-05 2023-01-21 13:32:48.017293: step: 1168/529, loss: 0.00065698625985533 2023-01-21 13:32:49.151561: step: 1172/529, loss: 0.0008242607582360506 2023-01-21 13:32:50.279336: step: 1176/529, loss: 0.0015277862548828125 2023-01-21 13:32:51.423432: step: 1180/529, loss: 0.006106281187385321 2023-01-21 13:32:52.616538: step: 1184/529, loss: 0.0004264831659384072 2023-01-21 13:32:53.789735: step: 1188/529, loss: 0.0009979248279705644 2023-01-21 13:32:54.972981: step: 1192/529, loss: 3.5667417250806466e-05 2023-01-21 13:32:56.084944: step: 1196/529, loss: 0.0024368285667151213 2023-01-21 13:32:57.214588: step: 1200/529, loss: 0.1896514892578125 2023-01-21 13:32:58.322028: step: 1204/529, loss: 1.544952465337701e-05 2023-01-21 13:32:59.449154: step: 1208/529, loss: 0.024172021076083183 2023-01-21 13:33:00.587205: step: 1212/529, loss: 0.07178258895874023 2023-01-21 13:33:01.732719: step: 1216/529, loss: 0.020547199994325638 2023-01-21 13:33:02.889311: step: 1220/529, loss: 0.0056362152099609375 2023-01-21 13:33:04.025951: step: 1224/529, loss: 0.022477244958281517 2023-01-21 13:33:05.163356: step: 1228/529, loss: 0.0005513191572390497 2023-01-21 13:33:06.337988: step: 1232/529, loss: 0.0011640548473224044 2023-01-21 13:33:07.447044: step: 1236/529, loss: 0.00029783250647597015 2023-01-21 13:33:08.596705: step: 1240/529, loss: 0.0050369263626635075 2023-01-21 13:33:09.769416: step: 1244/529, loss: 0.00013179780216887593 2023-01-21 13:33:10.912560: step: 1248/529, loss: 0.0027900696732103825 2023-01-21 13:33:12.047898: step: 1252/529, loss: 0.00015907287888694555 2023-01-21 13:33:13.172834: step: 1256/529, loss: 3.5858156479662284e-05 2023-01-21 13:33:14.291495: step: 1260/529, loss: 0.00011272430856479332 2023-01-21 13:33:15.406449: step: 1264/529, loss: 0.0013935089809820056 2023-01-21 13:33:16.535786: step: 1268/529, loss: 0.0035043240059167147 2023-01-21 13:33:17.659209: step: 1272/529, loss: 0.006024742033332586 2023-01-21 13:33:18.783050: step: 1276/529, loss: 3.051757857974735e-06 2023-01-21 13:33:19.930343: step: 1280/529, loss: 6.570816185558215e-05 2023-01-21 13:33:21.071998: step: 1284/529, loss: 0.01336746197193861 2023-01-21 13:33:22.218744: step: 1288/529, loss: 0.00023536682419944555 2023-01-21 13:33:23.371670: step: 1292/529, loss: 0.00289497384801507 2023-01-21 13:33:24.500660: step: 1296/529, loss: 4.9400332500226796e-05 2023-01-21 13:33:25.634884: step: 1300/529, loss: 8.77380352903856e-06 2023-01-21 13:33:26.771833: step: 1304/529, loss: 2.975463939947076e-05 2023-01-21 13:33:27.876049: step: 1308/529, loss: 3.4618376957951114e-05 2023-01-21 13:33:29.017173: step: 1312/529, loss: 2.727508581301663e-05 2023-01-21 13:33:30.155878: step: 1316/529, loss: 5.445480201160535e-05 2023-01-21 13:33:31.288348: step: 1320/529, loss: 1.0490421118447557e-06 2023-01-21 13:33:32.422050: step: 1324/529, loss: 0.006299972999840975 2023-01-21 13:33:33.597995: step: 1328/529, loss: 0.0010353088146075606 2023-01-21 13:33:34.719643: step: 1332/529, loss: 6.532669431180693e-06 2023-01-21 13:33:35.871606: step: 1336/529, loss: 7.80105619924143e-05 2023-01-21 13:33:36.997221: step: 1340/529, loss: 0.0023738862946629524 2023-01-21 13:33:38.127160: step: 1344/529, loss: 0.002029991243034601 2023-01-21 13:33:39.234676: step: 1348/529, loss: 3.4809113458322827e-06 2023-01-21 13:33:40.381982: step: 1352/529, loss: 3.4713742934400216e-05 2023-01-21 13:33:41.508998: step: 1356/529, loss: 0.0023852349258959293 2023-01-21 13:33:42.674234: step: 1360/529, loss: 0.017384052276611328 2023-01-21 13:33:43.799983: step: 1364/529, loss: 7.143020775401965e-05 2023-01-21 13:33:44.913041: step: 1368/529, loss: 0.041935063898563385 2023-01-21 13:33:46.036102: step: 1372/529, loss: 0.00025157927302643657 2023-01-21 13:33:47.173094: step: 1376/529, loss: 0.005435562692582607 2023-01-21 13:33:48.300487: step: 1380/529, loss: 8.268356759799644e-05 2023-01-21 13:33:49.417396: step: 1384/529, loss: 5.340576535672881e-06 2023-01-21 13:33:50.546005: step: 1388/529, loss: 0.011392593383789062 2023-01-21 13:33:51.666023: step: 1392/529, loss: 0.01087112445384264 2023-01-21 13:33:52.784198: step: 1396/529, loss: 0.0008609771612100303 2023-01-21 13:33:53.933526: step: 1400/529, loss: -1.754760705807712e-05 2023-01-21 13:33:55.055667: step: 1404/529, loss: 1.773834264895413e-05 2023-01-21 13:33:56.208471: step: 1408/529, loss: 0.04037752002477646 2023-01-21 13:33:57.334704: step: 1412/529, loss: 0.1447887420654297 2023-01-21 13:33:58.464197: step: 1416/529, loss: 0.0012201310601085424 2023-01-21 13:33:59.606334: step: 1420/529, loss: 0.0497930534183979 2023-01-21 13:34:00.743866: step: 1424/529, loss: 0.0015901564620435238 2023-01-21 13:34:01.832487: step: 1428/529, loss: 0.0003097534354310483 2023-01-21 13:34:02.957335: step: 1432/529, loss: 0.02193136140704155 2023-01-21 13:34:04.080775: step: 1436/529, loss: 0.018889809027314186 2023-01-21 13:34:05.209095: step: 1440/529, loss: 0.00017223358736373484 2023-01-21 13:34:06.343268: step: 1444/529, loss: 1.125335711549269e-05 2023-01-21 13:34:07.464445: step: 1448/529, loss: 0.008352279663085938 2023-01-21 13:34:08.587627: step: 1452/529, loss: 0.0004635810910258442 2023-01-21 13:34:09.734415: step: 1456/529, loss: 0.000211620339541696 2023-01-21 13:34:10.882019: step: 1460/529, loss: 0.005231666378676891 2023-01-21 13:34:12.011692: step: 1464/529, loss: 0.0028430940583348274 2023-01-21 13:34:13.136609: step: 1468/529, loss: 0.014341259375214577 2023-01-21 13:34:14.253752: step: 1472/529, loss: 0.01955385133624077 2023-01-21 13:34:15.372654: step: 1476/529, loss: 6.041527012712322e-05 2023-01-21 13:34:16.494739: step: 1480/529, loss: 1.3351436791708693e-06 2023-01-21 13:34:17.600507: step: 1484/529, loss: 0.0025349617935717106 2023-01-21 13:34:18.699671: step: 1488/529, loss: 0.0002658367156982422 2023-01-21 13:34:19.841037: step: 1492/529, loss: 0.002275657607242465 2023-01-21 13:34:20.993064: step: 1496/529, loss: 0.0009624481317587197 2023-01-21 13:34:22.116892: step: 1500/529, loss: 0.0010394096607342362 2023-01-21 13:34:23.301496: step: 1504/529, loss: 0.00041666030301712453 2023-01-21 13:34:24.436995: step: 1508/529, loss: 0.0063919066451489925 2023-01-21 13:34:25.554581: step: 1512/529, loss: -1.697540210443549e-05 2023-01-21 13:34:26.698594: step: 1516/529, loss: 0.0012765884166583419 2023-01-21 13:34:27.827116: step: 1520/529, loss: 0.041562844067811966 2023-01-21 13:34:28.944580: step: 1524/529, loss: 7.82012921263231e-06 2023-01-21 13:34:30.070764: step: 1528/529, loss: 0.000408172607421875 2023-01-21 13:34:31.206996: step: 1532/529, loss: -2.479552676959429e-06 2023-01-21 13:34:32.322408: step: 1536/529, loss: 0.0003198623890057206 2023-01-21 13:34:33.439923: step: 1540/529, loss: 0.003070882288739085 2023-01-21 13:34:34.606102: step: 1544/529, loss: 0.0013587952125817537 2023-01-21 13:34:35.716132: step: 1548/529, loss: 6.637573824264109e-05 2023-01-21 13:34:36.868476: step: 1552/529, loss: 0.04584341123700142 2023-01-21 13:34:38.026052: step: 1556/529, loss: 0.004402923863381147 2023-01-21 13:34:39.134483: step: 1560/529, loss: 0.018163301050662994 2023-01-21 13:34:40.295293: step: 1564/529, loss: 0.005459404084831476 2023-01-21 13:34:41.430863: step: 1568/529, loss: 0.0016266345046460629 2023-01-21 13:34:42.524794: step: 1572/529, loss: 0.000804042792879045 2023-01-21 13:34:43.644314: step: 1576/529, loss: 0.00014543533325195312 2023-01-21 13:34:44.773246: step: 1580/529, loss: 0.001674652099609375 2023-01-21 13:34:45.920459: step: 1584/529, loss: 0.006045246031135321 2023-01-21 13:34:47.114509: step: 1588/529, loss: 0.000324058550177142 2023-01-21 13:34:48.251541: step: 1592/529, loss: 0.000911712646484375 2023-01-21 13:34:49.391851: step: 1596/529, loss: 0.002735328860580921 2023-01-21 13:34:50.511490: step: 1600/529, loss: 4.291534423828125e-06 2023-01-21 13:34:51.673761: step: 1604/529, loss: 0.0002971649228129536 2023-01-21 13:34:52.837889: step: 1608/529, loss: 0.00019931793212890625 2023-01-21 13:34:53.970590: step: 1612/529, loss: 0.0013868332607671618 2023-01-21 13:34:55.073336: step: 1616/529, loss: 0.0015884817112237215 2023-01-21 13:34:56.236548: step: 1620/529, loss: 0.0009127617231570184 2023-01-21 13:34:57.351192: step: 1624/529, loss: 0.00016679764667060226 2023-01-21 13:34:58.484864: step: 1628/529, loss: 0.0013290405040606856 2023-01-21 13:34:59.616547: step: 1632/529, loss: 0.06161098554730415 2023-01-21 13:35:00.752507: step: 1636/529, loss: 3.356933666509576e-05 2023-01-21 13:35:01.884563: step: 1640/529, loss: 5.8746340073412284e-05 2023-01-21 13:35:03.019286: step: 1644/529, loss: 3.43322744811303e-06 2023-01-21 13:35:04.116447: step: 1648/529, loss: 0.0005939721595495939 2023-01-21 13:35:05.232931: step: 1652/529, loss: 6.32286028121598e-05 2023-01-21 13:35:06.361295: step: 1656/529, loss: 0.00811462476849556 2023-01-21 13:35:07.482138: step: 1660/529, loss: 3.046989513677545e-05 2023-01-21 13:35:08.617723: step: 1664/529, loss: 0.0068115233443677425 2023-01-21 13:35:09.750550: step: 1668/529, loss: 7.677078247070312e-05 2023-01-21 13:35:10.903464: step: 1672/529, loss: 0.00017452239990234375 2023-01-21 13:35:12.028302: step: 1676/529, loss: -1.144409225162235e-06 2023-01-21 13:35:13.179447: step: 1680/529, loss: 4.601478576660156e-05 2023-01-21 13:35:14.300019: step: 1684/529, loss: 0.00011558533151401207 2023-01-21 13:35:15.415344: step: 1688/529, loss: 0.010588455945253372 2023-01-21 13:35:16.577578: step: 1692/529, loss: 0.02328033559024334 2023-01-21 13:35:17.727969: step: 1696/529, loss: 0.00018939972505904734 2023-01-21 13:35:18.865615: step: 1700/529, loss: 5.3882598876953125e-05 2023-01-21 13:35:19.993691: step: 1704/529, loss: 4.696846008300781e-05 2023-01-21 13:35:21.141840: step: 1708/529, loss: 0.0001966476411325857 2023-01-21 13:35:22.301990: step: 1712/529, loss: 0.024821758270263672 2023-01-21 13:35:23.428049: step: 1716/529, loss: 0.00035004617529921234 2023-01-21 13:35:24.561572: step: 1720/529, loss: 0.0013475418090820312 2023-01-21 13:35:25.694137: step: 1724/529, loss: 0.0004724502796307206 2023-01-21 13:35:26.814259: step: 1728/529, loss: 0.0009954451816156507 2023-01-21 13:35:27.969446: step: 1732/529, loss: 0.0014881135430186987 2023-01-21 13:35:29.103791: step: 1736/529, loss: 0.014658165164291859 2023-01-21 13:35:30.219804: step: 1740/529, loss: 0.43324583768844604 2023-01-21 13:35:31.357348: step: 1744/529, loss: 0.0011410177685320377 2023-01-21 13:35:32.525696: step: 1748/529, loss: 0.0013059616321697831 2023-01-21 13:35:33.667735: step: 1752/529, loss: -3.814697322468419e-07 2023-01-21 13:35:34.830274: step: 1756/529, loss: 0.025440694764256477 2023-01-21 13:35:35.967673: step: 1760/529, loss: 0.0010429382091388106 2023-01-21 13:35:37.094080: step: 1764/529, loss: 0.011001873761415482 2023-01-21 13:35:38.245324: step: 1768/529, loss: 0.00017170906357932836 2023-01-21 13:35:39.390233: step: 1772/529, loss: 0.0002925872686319053 2023-01-21 13:35:40.510619: step: 1776/529, loss: 0.005158424377441406 2023-01-21 13:35:41.685710: step: 1780/529, loss: 2.4223329091910273e-05 2023-01-21 13:35:42.802430: step: 1784/529, loss: 0.00024366378784179688 2023-01-21 13:35:43.976378: step: 1788/529, loss: 0.006735706701874733 2023-01-21 13:35:45.105746: step: 1792/529, loss: 0.00011568069749046117 2023-01-21 13:35:46.233001: step: 1796/529, loss: 0.005925369448959827 2023-01-21 13:35:47.358868: step: 1800/529, loss: 0.00017452239990234375 2023-01-21 13:35:48.503287: step: 1804/529, loss: 0.002086067106574774 2023-01-21 13:35:49.638210: step: 1808/529, loss: 0.03179149702191353 2023-01-21 13:35:50.775874: step: 1812/529, loss: 0.03338279947638512 2023-01-21 13:35:51.906688: step: 1816/529, loss: 0.0031040192116051912 2023-01-21 13:35:53.024753: step: 1820/529, loss: 0.005063438322395086 2023-01-21 13:35:54.122180: step: 1824/529, loss: -0.0 2023-01-21 13:35:55.232990: step: 1828/529, loss: 2.231597864010837e-05 2023-01-21 13:35:56.374074: step: 1832/529, loss: -1.7166134966828395e-06 2023-01-21 13:35:57.550906: step: 1836/529, loss: 2.6464462280273438e-05 2023-01-21 13:35:58.709012: step: 1840/529, loss: 0.00030012131901457906 2023-01-21 13:35:59.869697: step: 1844/529, loss: 0.00018959045701194555 2023-01-21 13:36:00.996452: step: 1848/529, loss: 3.1948089599609375e-05 2023-01-21 13:36:02.122267: step: 1852/529, loss: 0.00029621124849654734 2023-01-21 13:36:03.293325: step: 1856/529, loss: 0.023758413270115852 2023-01-21 13:36:04.403886: step: 1860/529, loss: 0.0006875992403365672 2023-01-21 13:36:05.516874: step: 1864/529, loss: 0.0006736755603924394 2023-01-21 13:36:06.654733: step: 1868/529, loss: 0.00376987480558455 2023-01-21 13:36:07.747483: step: 1872/529, loss: 1.5735626220703125e-05 2023-01-21 13:36:08.863387: step: 1876/529, loss: 6.732941255904734e-05 2023-01-21 13:36:09.995473: step: 1880/529, loss: 0.09929961711168289 2023-01-21 13:36:11.102843: step: 1884/529, loss: 0.003735494799911976 2023-01-21 13:36:12.216430: step: 1888/529, loss: 0.00010643005225574598 2023-01-21 13:36:13.354988: step: 1892/529, loss: 0.013337898068130016 2023-01-21 13:36:14.482421: step: 1896/529, loss: 0.00020556450181175023 2023-01-21 13:36:15.651088: step: 1900/529, loss: 2.956390380859375e-05 2023-01-21 13:36:16.781211: step: 1904/529, loss: 0.06222815439105034 2023-01-21 13:36:17.934447: step: 1908/529, loss: 0.007288932800292969 2023-01-21 13:36:19.058549: step: 1912/529, loss: 0.00010604858107399195 2023-01-21 13:36:20.247055: step: 1916/529, loss: -1.544952465337701e-05 2023-01-21 13:36:21.409244: step: 1920/529, loss: 0.00024127960205078125 2023-01-21 13:36:22.529085: step: 1924/529, loss: 0.03608379513025284 2023-01-21 13:36:23.641944: step: 1928/529, loss: 0.3224422335624695 2023-01-21 13:36:24.771817: step: 1932/529, loss: 0.0006774902576580644 2023-01-21 13:36:25.916823: step: 1936/529, loss: 0.0008180618169717491 2023-01-21 13:36:27.057380: step: 1940/529, loss: 0.0015459060668945312 2023-01-21 13:36:28.185303: step: 1944/529, loss: 0.0005205154302529991 2023-01-21 13:36:29.344290: step: 1948/529, loss: 0.010601806454360485 2023-01-21 13:36:30.480174: step: 1952/529, loss: 0.034684088081121445 2023-01-21 13:36:31.606724: step: 1956/529, loss: 0.0008123397710733116 2023-01-21 13:36:32.726771: step: 1960/529, loss: 0.002321911044418812 2023-01-21 13:36:33.858361: step: 1964/529, loss: 0.0009657859918661416 2023-01-21 13:36:35.013051: step: 1968/529, loss: 0.00589942978695035 2023-01-21 13:36:36.138654: step: 1972/529, loss: 0.008754539303481579 2023-01-21 13:36:37.275749: step: 1976/529, loss: 0.002149295760318637 2023-01-21 13:36:38.404321: step: 1980/529, loss: 0.0008223534096032381 2023-01-21 13:36:39.536706: step: 1984/529, loss: 0.09383898228406906 2023-01-21 13:36:40.684796: step: 1988/529, loss: 0.1353701502084732 2023-01-21 13:36:41.806898: step: 1992/529, loss: 0.0041592600755393505 2023-01-21 13:36:42.976883: step: 1996/529, loss: 0.004723262973129749 2023-01-21 13:36:44.089649: step: 2000/529, loss: 0.00025300978450104594 2023-01-21 13:36:45.224540: step: 2004/529, loss: 0.0002091407950501889 2023-01-21 13:36:46.391645: step: 2008/529, loss: 9.250640869140625e-05 2023-01-21 13:36:47.496811: step: 2012/529, loss: 4.6443939936580136e-05 2023-01-21 13:36:48.600243: step: 2016/529, loss: 0.0003479004080872983 2023-01-21 13:36:49.704726: step: 2020/529, loss: 9.341239638160914e-05 2023-01-21 13:36:50.847185: step: 2024/529, loss: 0.0008009910816326737 2023-01-21 13:36:51.998273: step: 2028/529, loss: 0.0006376266246661544 2023-01-21 13:36:53.123714: step: 2032/529, loss: 0.005684280302375555 2023-01-21 13:36:54.237258: step: 2036/529, loss: 0.0003005981561727822 2023-01-21 13:36:55.326502: step: 2040/529, loss: 0.00034513475839048624 2023-01-21 13:36:56.456161: step: 2044/529, loss: 9.15527380129788e-06 2023-01-21 13:36:57.582301: step: 2048/529, loss: 0.0018419264815747738 2023-01-21 13:36:58.713293: step: 2052/529, loss: 0.0013955116737633944 2023-01-21 13:36:59.861414: step: 2056/529, loss: 1.3780594599666074e-05 2023-01-21 13:37:00.984606: step: 2060/529, loss: 0.004334306810051203 2023-01-21 13:37:02.128273: step: 2064/529, loss: 0.00029973985510878265 2023-01-21 13:37:03.273011: step: 2068/529, loss: 0.000598526035901159 2023-01-21 13:37:04.408331: step: 2072/529, loss: 0.00211677560582757 2023-01-21 13:37:05.533156: step: 2076/529, loss: 3.5858156479662284e-05 2023-01-21 13:37:06.678597: step: 2080/529, loss: 0.0006576538435183465 2023-01-21 13:37:07.829454: step: 2084/529, loss: 4.863738922722405e-06 2023-01-21 13:37:08.961415: step: 2088/529, loss: 0.9498537182807922 2023-01-21 13:37:10.089107: step: 2092/529, loss: 0.046903230249881744 2023-01-21 13:37:11.221073: step: 2096/529, loss: 0.0009557724115438759 2023-01-21 13:37:12.368137: step: 2100/529, loss: 0.0024808882735669613 2023-01-21 13:37:13.544918: step: 2104/529, loss: 2.918243444582913e-05 2023-01-21 13:37:14.664861: step: 2108/529, loss: 0.007530784700065851 2023-01-21 13:37:15.781301: step: 2112/529, loss: 0.00777969416230917 2023-01-21 13:37:16.906327: step: 2116/529, loss: 0.007008171174675226 ================================================== Loss: 0.019 -------------------- Dev: {'event': {'p': 0.5850956696878147, 'r': 0.7736351531291611, 'f1': 0.6662844036697246}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Test: {'event': {'p': 0.6100316312697696, 'r': 0.7821552723059096, 'f1': 0.6854531607006854}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Chinese: {'event': {'p': 0.6049382716049383, 'r': 0.9074074074074074, 'f1': 0.725925925925926}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Korean: {'event': {'p': 0.546875, 'r': 0.5555555555555556, 'f1': 0.5511811023622047}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Russian: {'event': {'p': 0.3695652173913043, 'r': 0.4722222222222222, 'f1': 0.41463414634146345}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6096033402922756, 'r': 0.7776298268974701, 'f1': 0.6834406085430076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Eng Test for Chinese: {'event': {'p': 0.6149954001839927, 'r': 0.7746234067207416, 'f1': 0.6856410256410257}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Sample Chinese: {'event': {'p': 0.6375, 'r': 0.9444444444444444, 'f1': 0.7611940298507462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} -------------------- Eng Dev for Korean: {'event': {'p': 0.6104944500504541, 'r': 0.8055925432756325, 'f1': 0.6946039035591274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Korean: {'event': {'p': 0.61996336996337, 'r': 0.7844727694090382, 'f1': 0.6925831202046036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Sample Korean: {'event': {'p': 0.6721311475409836, 'r': 0.6507936507936508, 'f1': 0.6612903225806451}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ****************************** Epoch: 29 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:37:58.294658: step: 4/529, loss: 2.3460386728402227e-05 2023-01-21 13:37:59.429133: step: 8/529, loss: 0.00029935839120298624 2023-01-21 13:38:00.579348: step: 12/529, loss: 0.014672184363007545 2023-01-21 13:38:01.706457: step: 16/529, loss: 5.7220458984375e-05 2023-01-21 13:38:02.816687: step: 20/529, loss: 9.5367431640625e-07 2023-01-21 13:38:03.948712: step: 24/529, loss: 0.004786873236298561 2023-01-21 13:38:05.067501: step: 28/529, loss: 0.00012731552124023438 2023-01-21 13:38:06.230930: step: 32/529, loss: 0.0003940582391805947 2023-01-21 13:38:07.367603: step: 36/529, loss: -2.47955313170678e-06 2023-01-21 13:38:08.491555: step: 40/529, loss: 0.0008508205064572394 2023-01-21 13:38:09.640932: step: 44/529, loss: 0.006895923521369696 2023-01-21 13:38:10.768685: step: 48/529, loss: 2.0503998712229077e-06 2023-01-21 13:38:11.891864: step: 52/529, loss: 0.0001802921324269846 2023-01-21 13:38:13.013403: step: 56/529, loss: 0.00015592575073242188 2023-01-21 13:38:14.145049: step: 60/529, loss: 0.00083751673810184 2023-01-21 13:38:15.269368: step: 64/529, loss: 0.5032185316085815 2023-01-21 13:38:16.413156: step: 68/529, loss: 0.0005693435668945312 2023-01-21 13:38:17.535101: step: 72/529, loss: 0.0004905700916424394 2023-01-21 13:38:18.679015: step: 76/529, loss: 0.2699142396450043 2023-01-21 13:38:19.822560: step: 80/529, loss: 9.584427061781753e-06 2023-01-21 13:38:20.952221: step: 84/529, loss: 0.0004817962762899697 2023-01-21 13:38:22.108711: step: 88/529, loss: 0.0006841659778729081 2023-01-21 13:38:23.215646: step: 92/529, loss: 0.007121467962861061 2023-01-21 13:38:24.358453: step: 96/529, loss: 4.9877166020451114e-05 2023-01-21 13:38:25.487270: step: 100/529, loss: 8.659363084007055e-05 2023-01-21 13:38:26.634779: step: 104/529, loss: 0.0076386453583836555 2023-01-21 13:38:27.760492: step: 108/529, loss: 0.009651850908994675 2023-01-21 13:38:28.881767: step: 112/529, loss: 0.0008951186900958419 2023-01-21 13:38:30.008364: step: 116/529, loss: 0.0006156921153888106 2023-01-21 13:38:31.147442: step: 120/529, loss: 0.14990416169166565 2023-01-21 13:38:32.294379: step: 124/529, loss: 0.0014657974243164062 2023-01-21 13:38:33.434812: step: 128/529, loss: 0.00019092560978606343 2023-01-21 13:38:34.545589: step: 132/529, loss: 0.00017852784367278218 2023-01-21 13:38:35.673240: step: 136/529, loss: 0.014473057352006435 2023-01-21 13:38:36.785756: step: 140/529, loss: 2.365112231927924e-05 2023-01-21 13:38:37.930206: step: 144/529, loss: 3.700256274896674e-05 2023-01-21 13:38:39.046731: step: 148/529, loss: 0.0011466980213299394 2023-01-21 13:38:40.189277: step: 152/529, loss: 0.00042872430640272796 2023-01-21 13:38:41.302436: step: 156/529, loss: 0.0023859022185206413 2023-01-21 13:38:42.409618: step: 160/529, loss: 0.0006986618391238153 2023-01-21 13:38:43.559735: step: 164/529, loss: 0.003813839051872492 2023-01-21 13:38:44.679781: step: 168/529, loss: 5.53131121705519e-06 2023-01-21 13:38:45.806936: step: 172/529, loss: 7.247925168485381e-06 2023-01-21 13:38:46.943356: step: 176/529, loss: 0.00010890961129916832 2023-01-21 13:38:48.078402: step: 180/529, loss: 0.0056404112838208675 2023-01-21 13:38:49.228089: step: 184/529, loss: 0.0022989274002611637 2023-01-21 13:38:50.386604: step: 188/529, loss: -1.754760705807712e-05 2023-01-21 13:38:51.481152: step: 192/529, loss: 0.002397251082584262 2023-01-21 13:38:52.591982: step: 196/529, loss: 0.009088468737900257 2023-01-21 13:38:53.715178: step: 200/529, loss: 0.00033540723961777985 2023-01-21 13:38:54.855307: step: 204/529, loss: 0.0004325866757426411 2023-01-21 13:38:55.990610: step: 208/529, loss: 0.04682159423828125 2023-01-21 13:38:57.137036: step: 212/529, loss: 1.888275073724799e-05 2023-01-21 13:38:58.246151: step: 216/529, loss: 0.0016709327464923263 2023-01-21 13:38:59.381381: step: 220/529, loss: 0.053383637219667435 2023-01-21 13:39:00.512991: step: 224/529, loss: 0.00042934418888762593 2023-01-21 13:39:01.662544: step: 228/529, loss: 0.0007314682006835938 2023-01-21 13:39:02.780618: step: 232/529, loss: 0.0005847931606695056 2023-01-21 13:39:03.894055: step: 236/529, loss: 4.596710277837701e-05 2023-01-21 13:39:05.001739: step: 240/529, loss: 5.073547436040826e-05 2023-01-21 13:39:06.122338: step: 244/529, loss: 0.004123473074287176 2023-01-21 13:39:07.270489: step: 248/529, loss: 0.010294127278029919 2023-01-21 13:39:08.421277: step: 252/529, loss: -1.9073486328125e-06 2023-01-21 13:39:09.559482: step: 256/529, loss: 0.00010976791236316785 2023-01-21 13:39:10.683478: step: 260/529, loss: 3.24249267578125e-05 2023-01-21 13:39:11.789111: step: 264/529, loss: 0.01105346716940403 2023-01-21 13:39:12.934644: step: 268/529, loss: 7.343292963923886e-05 2023-01-21 13:39:14.116034: step: 272/529, loss: 0.07409544289112091 2023-01-21 13:39:15.250578: step: 276/529, loss: 0.00010738372657215223 2023-01-21 13:39:16.374275: step: 280/529, loss: 0.0003066062927246094 2023-01-21 13:39:17.513167: step: 284/529, loss: -1.5258810890372843e-06 2023-01-21 13:39:18.651122: step: 288/529, loss: 0.014914512634277344 2023-01-21 13:39:19.771943: step: 292/529, loss: -4.768378403241513e-08 2023-01-21 13:39:20.898267: step: 296/529, loss: 0.014096450991928577 2023-01-21 13:39:22.043693: step: 300/529, loss: 0.0008718490717001259 2023-01-21 13:39:23.165465: step: 304/529, loss: 1.8596649169921875e-05 2023-01-21 13:39:24.301425: step: 308/529, loss: 0.002388572786003351 2023-01-21 13:39:25.445527: step: 312/529, loss: 0.005264568608254194 2023-01-21 13:39:26.564805: step: 316/529, loss: 3.6048892070539296e-05 2023-01-21 13:39:27.685385: step: 320/529, loss: 0.020990561693906784 2023-01-21 13:39:28.838998: step: 324/529, loss: 0.0017747879028320312 2023-01-21 13:39:29.962829: step: 328/529, loss: 0.00013570785813499242 2023-01-21 13:39:31.095784: step: 332/529, loss: 9.307861182605848e-05 2023-01-21 13:39:32.198073: step: 336/529, loss: 0.28476542234420776 2023-01-21 13:39:33.339079: step: 340/529, loss: 0.00011329650442348793 2023-01-21 13:39:34.499433: step: 344/529, loss: 0.0011203765170648694 2023-01-21 13:39:35.663851: step: 348/529, loss: 0.019318770617246628 2023-01-21 13:39:36.775068: step: 352/529, loss: 0.01470184326171875 2023-01-21 13:39:37.880733: step: 356/529, loss: 0.0005246639484539628 2023-01-21 13:39:39.031862: step: 360/529, loss: 0.002130126813426614 2023-01-21 13:39:40.199787: step: 364/529, loss: 8.296967280330136e-05 2023-01-21 13:39:41.340833: step: 368/529, loss: 0.0001904487726278603 2023-01-21 13:39:42.464130: step: 372/529, loss: 0.0015802383422851562 2023-01-21 13:39:43.625906: step: 376/529, loss: 0.03137655556201935 2023-01-21 13:39:44.754652: step: 380/529, loss: 0.00015354156494140625 2023-01-21 13:39:45.888389: step: 384/529, loss: 7.81059279688634e-05 2023-01-21 13:39:46.988954: step: 388/529, loss: 3.1280520488508046e-05 2023-01-21 13:39:48.132925: step: 392/529, loss: 0.005615520756691694 2023-01-21 13:39:49.281366: step: 396/529, loss: 0.02689342573285103 2023-01-21 13:39:50.396981: step: 400/529, loss: 0.0017635822296142578 2023-01-21 13:39:51.523022: step: 404/529, loss: 1.3732910701946821e-05 2023-01-21 13:39:52.651938: step: 408/529, loss: 6.990432302700356e-05 2023-01-21 13:39:53.768709: step: 412/529, loss: 5.684341886080802e-14 2023-01-21 13:39:54.923001: step: 416/529, loss: 0.004117202945053577 2023-01-21 13:39:56.017484: step: 420/529, loss: 0.04629566892981529 2023-01-21 13:39:57.140752: step: 424/529, loss: 0.01478795986622572 2023-01-21 13:39:58.272675: step: 428/529, loss: 0.0011396408081054688 2023-01-21 13:39:59.404067: step: 432/529, loss: 0.00019121170043945312 2023-01-21 13:40:00.536112: step: 436/529, loss: 0.002372837159782648 2023-01-21 13:40:01.678275: step: 440/529, loss: 0.0002368926943745464 2023-01-21 13:40:02.841222: step: 444/529, loss: 0.02082247845828533 2023-01-21 13:40:03.954627: step: 448/529, loss: 0.004999017808586359 2023-01-21 13:40:05.087544: step: 452/529, loss: 3.261566234868951e-05 2023-01-21 13:40:06.211259: step: 456/529, loss: 0.0018061638111248612 2023-01-21 13:40:07.355186: step: 460/529, loss: 0.0003569602849893272 2023-01-21 13:40:08.485596: step: 464/529, loss: 0.0007804393535479903 2023-01-21 13:40:09.604083: step: 468/529, loss: 0.0010192871559411287 2023-01-21 13:40:10.739628: step: 472/529, loss: 1.9073486328125e-05 2023-01-21 13:40:11.873052: step: 476/529, loss: 0.0007602691766805947 2023-01-21 13:40:13.016003: step: 480/529, loss: 0.0001371383696096018 2023-01-21 13:40:14.141636: step: 484/529, loss: 0.0013807296054437757 2023-01-21 13:40:15.246943: step: 488/529, loss: 0.00026416778564453125 2023-01-21 13:40:16.363049: step: 492/529, loss: 0.0008924484718590975 2023-01-21 13:40:17.478604: step: 496/529, loss: 5.1021575927734375e-05 2023-01-21 13:40:18.586504: step: 500/529, loss: 0.021207714453339577 2023-01-21 13:40:19.725812: step: 504/529, loss: 0.005426979158073664 2023-01-21 13:40:20.842419: step: 508/529, loss: 0.00489354133605957 2023-01-21 13:40:21.974333: step: 512/529, loss: 0.00035858154296875 2023-01-21 13:40:23.096894: step: 516/529, loss: 0.0008200645679607987 2023-01-21 13:40:24.235474: step: 520/529, loss: 1.316070574830519e-05 2023-01-21 13:40:25.352076: step: 524/529, loss: 0.002777671907097101 2023-01-21 13:40:26.491632: step: 528/529, loss: 4.596710277837701e-05 2023-01-21 13:40:27.610132: step: 532/529, loss: 0.0003758430539164692 2023-01-21 13:40:28.752603: step: 536/529, loss: 0.0016123296227306128 2023-01-21 13:40:29.907130: step: 540/529, loss: 2.6607513063936494e-05 2023-01-21 13:40:31.004554: step: 544/529, loss: 3.890991138177924e-05 2023-01-21 13:40:32.151685: step: 548/529, loss: 0.00042486193706281483 2023-01-21 13:40:33.293406: step: 552/529, loss: 0.004232645500451326 2023-01-21 13:40:34.427319: step: 556/529, loss: -4.9591067181609105e-06 2023-01-21 13:40:35.555339: step: 560/529, loss: 2.059936559817288e-05 2023-01-21 13:40:36.676761: step: 564/529, loss: 0.053559113293886185 2023-01-21 13:40:37.806271: step: 568/529, loss: 0.008004474453628063 2023-01-21 13:40:38.956173: step: 572/529, loss: -3.862380708596902e-06 2023-01-21 13:40:40.073137: step: 576/529, loss: 0.00014247893705032766 2023-01-21 13:40:41.205291: step: 580/529, loss: 0.06905770301818848 2023-01-21 13:40:42.369185: step: 584/529, loss: 0.0037467957008630037 2023-01-21 13:40:43.515203: step: 588/529, loss: 0.00011959076800849289 2023-01-21 13:40:44.659344: step: 592/529, loss: 8.773804438533261e-06 2023-01-21 13:40:45.787973: step: 596/529, loss: 0.00252361292950809 2023-01-21 13:40:46.921498: step: 600/529, loss: 0.004902648739516735 2023-01-21 13:40:48.067590: step: 604/529, loss: 1.3542176020564511e-05 2023-01-21 13:40:49.225361: step: 608/529, loss: 0.0007461548084393144 2023-01-21 13:40:50.390551: step: 612/529, loss: 0.0007884979131631553 2023-01-21 13:40:51.525665: step: 616/529, loss: 2.2268293832894415e-05 2023-01-21 13:40:52.659633: step: 620/529, loss: 0.0004235267697367817 2023-01-21 13:40:53.776927: step: 624/529, loss: 0.00041961669921875 2023-01-21 13:40:54.900758: step: 628/529, loss: 5.817413693876006e-05 2023-01-21 13:40:56.046527: step: 632/529, loss: -3.814693627646193e-07 2023-01-21 13:40:57.200566: step: 636/529, loss: 0.000263214111328125 2023-01-21 13:40:58.340381: step: 640/529, loss: 0.0021175385918468237 2023-01-21 13:40:59.480036: step: 644/529, loss: 0.028862619772553444 2023-01-21 13:41:00.620471: step: 648/529, loss: 0.00018520356388762593 2023-01-21 13:41:01.739936: step: 652/529, loss: 6.361007399391383e-05 2023-01-21 13:41:02.884369: step: 656/529, loss: 0.0072679524309933186 2023-01-21 13:41:04.020863: step: 660/529, loss: 5.683898780262098e-05 2023-01-21 13:41:05.158786: step: 664/529, loss: 0.0001659393310546875 2023-01-21 13:41:06.266232: step: 668/529, loss: 1.8739699953584932e-05 2023-01-21 13:41:07.364152: step: 672/529, loss: 0.0007623672718182206 2023-01-21 13:41:08.498427: step: 676/529, loss: 7.457733590854332e-05 2023-01-21 13:41:09.630846: step: 680/529, loss: 0.00019531250291038305 2023-01-21 13:41:10.753561: step: 684/529, loss: 0.0016321182483807206 2023-01-21 13:41:11.880594: step: 688/529, loss: 0.0650150328874588 2023-01-21 13:41:13.032919: step: 692/529, loss: 0.056092265993356705 2023-01-21 13:41:14.164882: step: 696/529, loss: 0.004901600535959005 2023-01-21 13:41:15.307882: step: 700/529, loss: 0.006706810090690851 2023-01-21 13:41:16.445795: step: 704/529, loss: 2.384185791015625e-05 2023-01-21 13:41:17.562233: step: 708/529, loss: 0.0006958008161745965 2023-01-21 13:41:18.681219: step: 712/529, loss: 0.004782963078469038 2023-01-21 13:41:19.812394: step: 716/529, loss: 0.011318016797304153 2023-01-21 13:41:20.953027: step: 720/529, loss: 0.03778582066297531 2023-01-21 13:41:22.062586: step: 724/529, loss: 0.00013236999802757055 2023-01-21 13:41:23.195242: step: 728/529, loss: 3.337860471219756e-05 2023-01-21 13:41:24.333873: step: 732/529, loss: 0.014777183532714844 2023-01-21 13:41:25.465493: step: 736/529, loss: 0.0031271218322217464 2023-01-21 13:41:26.598029: step: 740/529, loss: 9.202957153320312e-05 2023-01-21 13:41:27.769878: step: 744/529, loss: 0.0004380226309876889 2023-01-21 13:41:28.902201: step: 748/529, loss: 0.0006619453779421747 2023-01-21 13:41:30.059701: step: 752/529, loss: -1.8119812921213452e-06 2023-01-21 13:41:31.223940: step: 756/529, loss: 5.245208740234375e-06 2023-01-21 13:41:32.359997: step: 760/529, loss: 9.698867506813258e-05 2023-01-21 13:41:33.504974: step: 764/529, loss: 1.926422191900201e-05 2023-01-21 13:41:34.673478: step: 768/529, loss: 0.0018943786853924394 2023-01-21 13:41:35.836786: step: 772/529, loss: 0.005262661259621382 2023-01-21 13:41:36.978592: step: 776/529, loss: -3.814697322468419e-07 2023-01-21 13:41:38.103829: step: 780/529, loss: 0.000484466552734375 2023-01-21 13:41:39.211953: step: 784/529, loss: -4.19616708313697e-06 2023-01-21 13:41:40.352096: step: 788/529, loss: 0.0003603935183491558 2023-01-21 13:41:41.485772: step: 792/529, loss: 0.023030854761600494 2023-01-21 13:41:42.635158: step: 796/529, loss: 0.0053886412642896175 2023-01-21 13:41:43.785240: step: 800/529, loss: 7.24792471373803e-06 2023-01-21 13:41:44.903062: step: 804/529, loss: 0.033295489847660065 2023-01-21 13:41:46.030436: step: 808/529, loss: 1.869201696536038e-05 2023-01-21 13:41:47.146061: step: 812/529, loss: 0.0011773109436035156 2023-01-21 13:41:48.258950: step: 816/529, loss: 0.003726184368133545 2023-01-21 13:41:49.386616: step: 820/529, loss: 2.8467178708524443e-05 2023-01-21 13:41:50.508765: step: 824/529, loss: 0.00016708373732399195 2023-01-21 13:41:51.641391: step: 828/529, loss: 0.0010879039764404297 2023-01-21 13:41:52.781420: step: 832/529, loss: 3.795623706537299e-05 2023-01-21 13:41:53.887186: step: 836/529, loss: 4.668235487770289e-05 2023-01-21 13:41:55.028341: step: 840/529, loss: 0.033077385276556015 2023-01-21 13:41:56.168174: step: 844/529, loss: 0.007237625308334827 2023-01-21 13:41:57.322839: step: 848/529, loss: 0.0013727188343182206 2023-01-21 13:41:58.448145: step: 852/529, loss: 0.0005105972522869706 2023-01-21 13:41:59.599028: step: 856/529, loss: 0.03170204162597656 2023-01-21 13:42:00.746282: step: 860/529, loss: 0.00256690988317132 2023-01-21 13:42:01.890468: step: 864/529, loss: 0.0013068198459222913 2023-01-21 13:42:03.042395: step: 868/529, loss: 0.00027236941969022155 2023-01-21 13:42:04.173304: step: 872/529, loss: 0.05469665676355362 2023-01-21 13:42:05.277437: step: 876/529, loss: 1.9454957509879023e-05 2023-01-21 13:42:06.375847: step: 880/529, loss: 1.964569128176663e-05 2023-01-21 13:42:07.517844: step: 884/529, loss: 0.00011425017873989418 2023-01-21 13:42:08.654815: step: 888/529, loss: 0.006271743681281805 2023-01-21 13:42:09.763651: step: 892/529, loss: 1.621246337890625e-05 2023-01-21 13:42:10.975236: step: 896/529, loss: 0.00096044538076967 2023-01-21 13:42:12.113206: step: 900/529, loss: 0.005579191725701094 2023-01-21 13:42:13.231784: step: 904/529, loss: 0.0014452934265136719 2023-01-21 13:42:14.380249: step: 908/529, loss: 0.023462487384676933 2023-01-21 13:42:15.513273: step: 912/529, loss: 0.0772615447640419 2023-01-21 13:42:16.661543: step: 916/529, loss: 0.00016660690016578883 2023-01-21 13:42:17.794082: step: 920/529, loss: 0.006028270814567804 2023-01-21 13:42:18.917311: step: 924/529, loss: 0.0023314475547522306 2023-01-21 13:42:20.069516: step: 928/529, loss: 2.193450927734375e-05 2023-01-21 13:42:21.177792: step: 932/529, loss: 5.14984139954322e-06 2023-01-21 13:42:22.336313: step: 936/529, loss: 0.00025196076603606343 2023-01-21 13:42:23.471070: step: 940/529, loss: 0.05692453682422638 2023-01-21 13:42:24.616626: step: 944/529, loss: 0.000346946733770892 2023-01-21 13:42:25.770904: step: 948/529, loss: 0.00018634795560501516 2023-01-21 13:42:26.919466: step: 952/529, loss: 0.00021378994279075414 2023-01-21 13:42:28.066961: step: 956/529, loss: 3.4809113458322827e-06 2023-01-21 13:42:29.212145: step: 960/529, loss: 1.926422191900201e-05 2023-01-21 13:42:30.365055: step: 964/529, loss: 0.0012403487926349044 2023-01-21 13:42:31.508494: step: 968/529, loss: 0.023621369153261185 2023-01-21 13:42:32.632409: step: 972/529, loss: 0.0001446724054403603 2023-01-21 13:42:33.768594: step: 976/529, loss: 0.00833206158131361 2023-01-21 13:42:34.905041: step: 980/529, loss: 0.025003530085086823 2023-01-21 13:42:36.049068: step: 984/529, loss: 0.04435539245605469 2023-01-21 13:42:37.201631: step: 988/529, loss: 7.686615572310984e-05 2023-01-21 13:42:38.347549: step: 992/529, loss: 0.00627479562535882 2023-01-21 13:42:39.496203: step: 996/529, loss: 0.0008912086486816406 2023-01-21 13:42:40.641183: step: 1000/529, loss: 0.0004402160702738911 2023-01-21 13:42:41.760175: step: 1004/529, loss: 0.0003693580802064389 2023-01-21 13:42:42.937671: step: 1008/529, loss: 0.0011402606032788754 2023-01-21 13:42:44.067779: step: 1012/529, loss: 0.12761488556861877 2023-01-21 13:42:45.190492: step: 1016/529, loss: 0.0008741378551349044 2023-01-21 13:42:46.351320: step: 1020/529, loss: 1.4781952813791577e-05 2023-01-21 13:42:47.476311: step: 1024/529, loss: 0.0065240864641964436 2023-01-21 13:42:48.639586: step: 1028/529, loss: 0.02981739118695259 2023-01-21 13:42:49.765294: step: 1032/529, loss: 0.00018424987501930445 2023-01-21 13:42:50.872056: step: 1036/529, loss: 0.017676925286650658 2023-01-21 13:42:52.021769: step: 1040/529, loss: 0.00027637480525299907 2023-01-21 13:42:53.144643: step: 1044/529, loss: 3.538131932145916e-05 2023-01-21 13:42:54.290045: step: 1048/529, loss: 0.00035533905611373484 2023-01-21 13:42:55.420834: step: 1052/529, loss: 0.0002127170591847971 2023-01-21 13:42:56.544938: step: 1056/529, loss: 7.591248140670359e-05 2023-01-21 13:42:57.678330: step: 1060/529, loss: 3.7622452509822324e-05 2023-01-21 13:42:58.817593: step: 1064/529, loss: 0.002129149390384555 2023-01-21 13:42:59.956188: step: 1068/529, loss: 7.43865984986769e-06 2023-01-21 13:43:01.085965: step: 1072/529, loss: 0.00011397599882911891 2023-01-21 13:43:02.217391: step: 1076/529, loss: 0.00026998520479537547 2023-01-21 13:43:03.347009: step: 1080/529, loss: 5.33103957423009e-05 2023-01-21 13:43:04.490753: step: 1084/529, loss: 0.01102142408490181 2023-01-21 13:43:05.653032: step: 1088/529, loss: 0.012543297372758389 2023-01-21 13:43:06.792676: step: 1092/529, loss: 0.0003061771276406944 2023-01-21 13:43:07.997915: step: 1096/529, loss: 0.0028985501267015934 2023-01-21 13:43:09.115511: step: 1100/529, loss: 0.00029082296532578766 2023-01-21 13:43:10.250566: step: 1104/529, loss: 0.00033473968505859375 2023-01-21 13:43:11.382168: step: 1108/529, loss: 0.0009959221351891756 2023-01-21 13:43:12.532902: step: 1112/529, loss: 3.910064606316155e-06 2023-01-21 13:43:13.683611: step: 1116/529, loss: 0.001818847726099193 2023-01-21 13:43:14.805785: step: 1120/529, loss: 7.915497008070815e-06 2023-01-21 13:43:15.931566: step: 1124/529, loss: 0.0004977226490154862 2023-01-21 13:43:17.043699: step: 1128/529, loss: 0.0020755769219249487 2023-01-21 13:43:18.177927: step: 1132/529, loss: 0.0018720626831054688 2023-01-21 13:43:19.319983: step: 1136/529, loss: 1.4257430848374497e-05 2023-01-21 13:43:20.463827: step: 1140/529, loss: 0.003218937199562788 2023-01-21 13:43:21.604791: step: 1144/529, loss: 6.508827027573716e-06 2023-01-21 13:43:22.729136: step: 1148/529, loss: 0.0028348921332508326 2023-01-21 13:43:23.858821: step: 1152/529, loss: 0.0027710916474461555 2023-01-21 13:43:24.995023: step: 1156/529, loss: 0.0001966476411325857 2023-01-21 13:43:26.086693: step: 1160/529, loss: 0.00021901130094192922 2023-01-21 13:43:27.237547: step: 1164/529, loss: 4.863738922722405e-06 2023-01-21 13:43:28.355676: step: 1168/529, loss: 0.0004108429129701108 2023-01-21 13:43:29.513120: step: 1172/529, loss: 0.001256656600162387 2023-01-21 13:43:30.639805: step: 1176/529, loss: 0.00730056781321764 2023-01-21 13:43:31.737759: step: 1180/529, loss: 0.003490447998046875 2023-01-21 13:43:32.914291: step: 1184/529, loss: 0.025895357131958008 2023-01-21 13:43:34.040198: step: 1188/529, loss: 0.0071398732252418995 2023-01-21 13:43:35.155537: step: 1192/529, loss: 0.024666596204042435 2023-01-21 13:43:36.319199: step: 1196/529, loss: 0.00036582950269803405 2023-01-21 13:43:37.446186: step: 1200/529, loss: 0.00165901193395257 2023-01-21 13:43:38.589169: step: 1204/529, loss: 3.62396240234375e-05 2023-01-21 13:43:39.730415: step: 1208/529, loss: 0.5964565873146057 2023-01-21 13:43:40.823702: step: 1212/529, loss: 0.00023236274137161672 2023-01-21 13:43:41.949994: step: 1216/529, loss: 0.00144109723623842 2023-01-21 13:43:43.049987: step: 1220/529, loss: 3.662109520519152e-05 2023-01-21 13:43:44.165787: step: 1224/529, loss: 0.005763912573456764 2023-01-21 13:43:45.295035: step: 1228/529, loss: 0.07522864639759064 2023-01-21 13:43:46.463996: step: 1232/529, loss: 0.0003174305020365864 2023-01-21 13:43:47.606596: step: 1236/529, loss: 3.070831371587701e-05 2023-01-21 13:43:48.723522: step: 1240/529, loss: 7.410049147438258e-05 2023-01-21 13:43:49.857251: step: 1244/529, loss: 7.591248140670359e-05 2023-01-21 13:43:50.989087: step: 1248/529, loss: 6.389617738022935e-06 2023-01-21 13:43:52.166746: step: 1252/529, loss: 1.9073486612342094e-07 2023-01-21 13:43:53.316320: step: 1256/529, loss: 2.0827352273045108e-05 2023-01-21 13:43:54.450780: step: 1260/529, loss: 0.09041042625904083 2023-01-21 13:43:55.586215: step: 1264/529, loss: 0.004899788182228804 2023-01-21 13:43:56.731102: step: 1268/529, loss: 0.0001458644837839529 2023-01-21 13:43:57.878847: step: 1272/529, loss: 0.7516994476318359 2023-01-21 13:43:59.018462: step: 1276/529, loss: 0.023574162274599075 2023-01-21 13:44:00.179283: step: 1280/529, loss: 0.004985761363059282 2023-01-21 13:44:01.337677: step: 1284/529, loss: 0.012679862789809704 2023-01-21 13:44:02.460623: step: 1288/529, loss: 0.028561020269989967 2023-01-21 13:44:03.613229: step: 1292/529, loss: 0.0013056754833087325 2023-01-21 13:44:04.731376: step: 1296/529, loss: 9.784699068404734e-05 2023-01-21 13:44:05.872579: step: 1300/529, loss: 0.02716655842959881 2023-01-21 13:44:07.005695: step: 1304/529, loss: 0.0332736037671566 2023-01-21 13:44:08.161026: step: 1308/529, loss: 0.00038585663423873484 2023-01-21 13:44:09.307633: step: 1312/529, loss: 0.03784656524658203 2023-01-21 13:44:10.450991: step: 1316/529, loss: 0.0308977123349905 2023-01-21 13:44:11.585489: step: 1320/529, loss: 0.019786834716796875 2023-01-21 13:44:12.699012: step: 1324/529, loss: 0.0004259109846316278 2023-01-21 13:44:13.859266: step: 1328/529, loss: 0.036594390869140625 2023-01-21 13:44:15.015171: step: 1332/529, loss: 7.057189577608369e-06 2023-01-21 13:44:16.133028: step: 1336/529, loss: 0.0019504546653479338 2023-01-21 13:44:17.267914: step: 1340/529, loss: 0.017937278375029564 2023-01-21 13:44:18.405928: step: 1344/529, loss: 0.008871365338563919 2023-01-21 13:44:19.546958: step: 1348/529, loss: 0.08070297539234161 2023-01-21 13:44:20.692508: step: 1352/529, loss: 0.0002893448108807206 2023-01-21 13:44:21.829775: step: 1356/529, loss: 8.02993745310232e-05 2023-01-21 13:44:22.976676: step: 1360/529, loss: 0.003805828047916293 2023-01-21 13:44:24.115033: step: 1364/529, loss: 8.811950829112902e-05 2023-01-21 13:44:25.243818: step: 1368/529, loss: 0.024486638605594635 2023-01-21 13:44:26.392985: step: 1372/529, loss: 6.46591215627268e-05 2023-01-21 13:44:27.501000: step: 1376/529, loss: 0.013908577151596546 2023-01-21 13:44:28.655223: step: 1380/529, loss: 0.0063065532594919205 2023-01-21 13:44:29.780930: step: 1384/529, loss: 0.00014247893705032766 2023-01-21 13:44:30.920164: step: 1388/529, loss: 0.0030319690704345703 2023-01-21 13:44:32.056313: step: 1392/529, loss: 0.013313484378159046 2023-01-21 13:44:33.183331: step: 1396/529, loss: 0.006332111544907093 2023-01-21 13:44:34.290832: step: 1400/529, loss: 0.003944015596061945 2023-01-21 13:44:35.463654: step: 1404/529, loss: 3.843307786155492e-05 2023-01-21 13:44:36.591260: step: 1408/529, loss: 0.0009023666498251259 2023-01-21 13:44:37.734383: step: 1412/529, loss: 0.015327262692153454 2023-01-21 13:44:38.875280: step: 1416/529, loss: 0.00046062469482421875 2023-01-21 13:44:40.014613: step: 1420/529, loss: 0.0008195877308025956 2023-01-21 13:44:41.151626: step: 1424/529, loss: 0.03784484788775444 2023-01-21 13:44:42.250871: step: 1428/529, loss: 2.2888182229507947e-06 2023-01-21 13:44:43.391091: step: 1432/529, loss: 0.0006546020740643144 2023-01-21 13:44:44.588612: step: 1436/529, loss: 0.0011878968216478825 2023-01-21 13:44:45.721105: step: 1440/529, loss: 0.031148720532655716 2023-01-21 13:44:46.861016: step: 1444/529, loss: 0.001228809473104775 2023-01-21 13:44:47.983852: step: 1448/529, loss: 0.000751876796130091 2023-01-21 13:44:49.143678: step: 1452/529, loss: 0.03620429337024689 2023-01-21 13:44:50.293707: step: 1456/529, loss: 0.002294826554134488 2023-01-21 13:44:51.444399: step: 1460/529, loss: 0.10650186240673065 2023-01-21 13:44:52.579907: step: 1464/529, loss: 0.06921463459730148 2023-01-21 13:44:53.703377: step: 1468/529, loss: 0.0002173423854401335 2023-01-21 13:44:54.864859: step: 1472/529, loss: 0.0038563727866858244 2023-01-21 13:44:56.019171: step: 1476/529, loss: 2.09808349609375e-05 2023-01-21 13:44:57.157115: step: 1480/529, loss: 0.0002063751162495464 2023-01-21 13:44:58.303260: step: 1484/529, loss: 0.00020675660925917327 2023-01-21 13:44:59.414302: step: 1488/529, loss: 0.0016550064319744706 2023-01-21 13:45:00.561231: step: 1492/529, loss: 3.2901763916015625e-05 2023-01-21 13:45:01.698546: step: 1496/529, loss: 4.119873119634576e-05 2023-01-21 13:45:02.845296: step: 1500/529, loss: 0.0003685951232910156 2023-01-21 13:45:04.006053: step: 1504/529, loss: 3.06129441014491e-05 2023-01-21 13:45:05.129897: step: 1508/529, loss: 1.0967251000693068e-06 2023-01-21 13:45:06.311065: step: 1512/529, loss: 0.00045871734619140625 2023-01-21 13:45:07.449528: step: 1516/529, loss: 0.024495983496308327 2023-01-21 13:45:08.612198: step: 1520/529, loss: 8.60214204294607e-05 2023-01-21 13:45:09.746459: step: 1524/529, loss: 0.051072120666503906 2023-01-21 13:45:10.868174: step: 1528/529, loss: 9.765625145519152e-05 2023-01-21 13:45:11.997383: step: 1532/529, loss: 0.024402141571044922 2023-01-21 13:45:13.156174: step: 1536/529, loss: 0.23023644089698792 2023-01-21 13:45:14.269709: step: 1540/529, loss: 0.007288933265954256 2023-01-21 13:45:15.408334: step: 1544/529, loss: 0.00013408661470748484 2023-01-21 13:45:16.539629: step: 1548/529, loss: 0.0003101348993368447 2023-01-21 13:45:17.677526: step: 1552/529, loss: 0.019306089729070663 2023-01-21 13:45:18.779857: step: 1556/529, loss: 0.00035858154296875 2023-01-21 13:45:19.910580: step: 1560/529, loss: 2.346038854739163e-05 2023-01-21 13:45:21.035794: step: 1564/529, loss: 0.0006791114574298263 2023-01-21 13:45:22.185846: step: 1568/529, loss: 0.04008903726935387 2023-01-21 13:45:23.311821: step: 1572/529, loss: 0.0009317398071289062 2023-01-21 13:45:24.456198: step: 1576/529, loss: 1.2683869499596767e-05 2023-01-21 13:45:25.594109: step: 1580/529, loss: 0.04442129284143448 2023-01-21 13:45:26.746046: step: 1584/529, loss: 0.0014568328624591231 2023-01-21 13:45:27.855028: step: 1588/529, loss: 0.0013950347201898694 2023-01-21 13:45:28.976322: step: 1592/529, loss: 0.015591430477797985 2023-01-21 13:45:30.075589: step: 1596/529, loss: 4.7683710135970614e-07 2023-01-21 13:45:31.205074: step: 1600/529, loss: 0.0007187843439169228 2023-01-21 13:45:32.332525: step: 1604/529, loss: 4.291534423828125e-06 2023-01-21 13:45:33.453005: step: 1608/529, loss: 0.0032068253494799137 2023-01-21 13:45:34.586597: step: 1612/529, loss: 1.2588501704158261e-05 2023-01-21 13:45:35.732072: step: 1616/529, loss: 1.1396408808650449e-05 2023-01-21 13:45:36.871391: step: 1620/529, loss: 0.00022563934908248484 2023-01-21 13:45:38.027577: step: 1624/529, loss: 0.0004671096976380795 2023-01-21 13:45:39.180157: step: 1628/529, loss: 0.00029478073702193797 2023-01-21 13:45:40.317756: step: 1632/529, loss: 4.901886131847277e-05 2023-01-21 13:45:41.433272: step: 1636/529, loss: 0.0856361836194992 2023-01-21 13:45:42.634942: step: 1640/529, loss: 0.0035624506417661905 2023-01-21 13:45:43.772649: step: 1644/529, loss: 0.00023083687119651586 2023-01-21 13:45:44.888627: step: 1648/529, loss: 3.318786548334174e-05 2023-01-21 13:45:46.015614: step: 1652/529, loss: 9.441375368623994e-06 2023-01-21 13:45:47.137900: step: 1656/529, loss: 1.659393274167087e-05 2023-01-21 13:45:48.274120: step: 1660/529, loss: 0.013043785467743874 2023-01-21 13:45:49.396820: step: 1664/529, loss: 0.00016469955153297633 2023-01-21 13:45:50.524271: step: 1668/529, loss: 0.04447536543011665 2023-01-21 13:45:51.654783: step: 1672/529, loss: 0.0002231597900390625 2023-01-21 13:45:52.803507: step: 1676/529, loss: -3.719329924933845e-06 2023-01-21 13:45:53.947194: step: 1680/529, loss: 0.0025533675216138363 2023-01-21 13:45:55.067613: step: 1684/529, loss: 9.975433931685984e-05 2023-01-21 13:45:56.198207: step: 1688/529, loss: 0.0004222869756631553 2023-01-21 13:45:57.334584: step: 1692/529, loss: 0.011151027865707874 2023-01-21 13:45:58.469772: step: 1696/529, loss: 0.00291023263707757 2023-01-21 13:45:59.586724: step: 1700/529, loss: 0.0002573966921772808 2023-01-21 13:46:00.719433: step: 1704/529, loss: 0.0005517005920410156 2023-01-21 13:46:01.858531: step: 1708/529, loss: 1.2874604180979077e-05 2023-01-21 13:46:02.975608: step: 1712/529, loss: 2.28881845032447e-06 2023-01-21 13:46:04.146736: step: 1716/529, loss: 0.02035699039697647 2023-01-21 13:46:05.250934: step: 1720/529, loss: -3.8146970382513246e-07 2023-01-21 13:46:06.374238: step: 1724/529, loss: 0.00012292862811591476 2023-01-21 13:46:07.500218: step: 1728/529, loss: 3.14712519866589e-06 2023-01-21 13:46:08.617497: step: 1732/529, loss: 6.351470801746473e-05 2023-01-21 13:46:09.743790: step: 1736/529, loss: 8.7738037109375e-05 2023-01-21 13:46:10.822581: step: 1740/529, loss: 1.8882752556237392e-05 2023-01-21 13:46:11.958930: step: 1744/529, loss: 8.239746239269152e-05 2023-01-21 13:46:13.092161: step: 1748/529, loss: 0.00829000398516655 2023-01-21 13:46:14.253729: step: 1752/529, loss: 0.0005329132545739412 2023-01-21 13:46:15.373628: step: 1756/529, loss: 0.08768520504236221 2023-01-21 13:46:16.517073: step: 1760/529, loss: 0.0005876541254110634 2023-01-21 13:46:17.657202: step: 1764/529, loss: 2.6321413315599784e-05 2023-01-21 13:46:18.801291: step: 1768/529, loss: 0.020531844347715378 2023-01-21 13:46:19.932342: step: 1772/529, loss: 0.00024600030155852437 2023-01-21 13:46:21.056148: step: 1776/529, loss: -6.580352874152595e-06 2023-01-21 13:46:22.184072: step: 1780/529, loss: 0.008731079287827015 2023-01-21 13:46:23.298083: step: 1784/529, loss: 0.0001260757417185232 2023-01-21 13:46:24.412104: step: 1788/529, loss: 1.3828278497385327e-05 2023-01-21 13:46:25.553845: step: 1792/529, loss: 0.0009692192543298006 2023-01-21 13:46:26.686870: step: 1796/529, loss: 0.0007339477888308465 2023-01-21 13:46:27.820025: step: 1800/529, loss: 0.006910133641213179 2023-01-21 13:46:28.972675: step: 1804/529, loss: 0.0007410050020553172 2023-01-21 13:46:30.100686: step: 1808/529, loss: 1.621246337890625e-05 2023-01-21 13:46:31.243681: step: 1812/529, loss: 6.2227250055002514e-06 2023-01-21 13:46:32.384288: step: 1816/529, loss: 0.000681972480379045 2023-01-21 13:46:33.551084: step: 1820/529, loss: 3.604888843256049e-05 2023-01-21 13:46:34.712387: step: 1824/529, loss: 0.00016946792311500758 2023-01-21 13:46:35.846695: step: 1828/529, loss: 2.632141149661038e-05 2023-01-21 13:46:36.986744: step: 1832/529, loss: 0.0004347801150288433 2023-01-21 13:46:38.119540: step: 1836/529, loss: 1.0633469173626509e-05 2023-01-21 13:46:39.259767: step: 1840/529, loss: 7.3432925091765355e-06 2023-01-21 13:46:40.372086: step: 1844/529, loss: 8.487701961712446e-06 2023-01-21 13:46:41.509119: step: 1848/529, loss: 1.6880036127986386e-05 2023-01-21 13:46:42.651166: step: 1852/529, loss: 0.013188933953642845 2023-01-21 13:46:43.813288: step: 1856/529, loss: 0.0029180527199059725 2023-01-21 13:46:44.947328: step: 1860/529, loss: 0.0033987045753747225 2023-01-21 13:46:46.066840: step: 1864/529, loss: 0.0032112121116369963 2023-01-21 13:46:47.217179: step: 1868/529, loss: 0.00013217926607467234 2023-01-21 13:46:48.321846: step: 1872/529, loss: 8.869170414982364e-06 2023-01-21 13:46:49.455074: step: 1876/529, loss: 0.04085225984454155 2023-01-21 13:46:50.605933: step: 1880/529, loss: 0.0005708694225177169 2023-01-21 13:46:51.734218: step: 1884/529, loss: 3.8623809814453125e-05 2023-01-21 13:46:52.870876: step: 1888/529, loss: 0.00015745163545943797 2023-01-21 13:46:53.986361: step: 1892/529, loss: 1.0490417707842425e-06 2023-01-21 13:46:55.121836: step: 1896/529, loss: 0.012462044134736061 2023-01-21 13:46:56.248497: step: 1900/529, loss: 0.0053730015642941 2023-01-21 13:46:57.381496: step: 1904/529, loss: 0.0001237869291799143 2023-01-21 13:46:58.507980: step: 1908/529, loss: 0.011725139804184437 2023-01-21 13:46:59.625753: step: 1912/529, loss: 0.01125650480389595 2023-01-21 13:47:00.733806: step: 1916/529, loss: 6.723403930664062e-05 2023-01-21 13:47:01.899555: step: 1920/529, loss: 0.04301281273365021 2023-01-21 13:47:03.046569: step: 1924/529, loss: 0.01636066474020481 2023-01-21 13:47:04.191844: step: 1928/529, loss: 0.000640058540739119 2023-01-21 13:47:05.397081: step: 1932/529, loss: 0.00029277801513671875 2023-01-21 13:47:06.556951: step: 1936/529, loss: 0.0030913352966308594 2023-01-21 13:47:07.740272: step: 1940/529, loss: 2.2602082026423886e-05 2023-01-21 13:47:08.883621: step: 1944/529, loss: 1.4591216313419864e-05 2023-01-21 13:47:10.038097: step: 1948/529, loss: 5.2738188969669864e-05 2023-01-21 13:47:11.176416: step: 1952/529, loss: 0.00028896331787109375 2023-01-21 13:47:12.314951: step: 1956/529, loss: 0.00011730194091796875 2023-01-21 13:47:13.435974: step: 1960/529, loss: 0.00019721985154319555 2023-01-21 13:47:14.579173: step: 1964/529, loss: 0.0011264801723882556 2023-01-21 13:47:15.707643: step: 1968/529, loss: 3.719329924933845e-06 2023-01-21 13:47:16.842703: step: 1972/529, loss: 0.00028676987858489156 2023-01-21 13:47:17.974386: step: 1976/529, loss: 8.487701052217744e-06 2023-01-21 13:47:19.123080: step: 1980/529, loss: 8.77380352903856e-06 2023-01-21 13:47:20.267244: step: 1984/529, loss: 0.00012969972158316523 2023-01-21 13:47:21.402838: step: 1988/529, loss: -1.716613724056515e-06 2023-01-21 13:47:22.520098: step: 1992/529, loss: 0.005074310582131147 2023-01-21 13:47:23.629486: step: 1996/529, loss: 1.6498564946232364e-05 2023-01-21 13:47:24.759436: step: 2000/529, loss: 0.00019345282635185868 2023-01-21 13:47:25.871468: step: 2004/529, loss: 1.850128137448337e-05 2023-01-21 13:47:27.034063: step: 2008/529, loss: 0.004309940617531538 2023-01-21 13:47:28.164642: step: 2012/529, loss: 0.0011140822898596525 2023-01-21 13:47:29.280173: step: 2016/529, loss: 0.00025830266531556845 2023-01-21 13:47:30.431875: step: 2020/529, loss: 0.02773265913128853 2023-01-21 13:47:31.565652: step: 2024/529, loss: 0.011422920972108841 2023-01-21 13:47:32.687777: step: 2028/529, loss: 5.435943603515625e-05 2023-01-21 13:47:33.829261: step: 2032/529, loss: 9.15527380129788e-06 2023-01-21 13:47:34.962503: step: 2036/529, loss: 0.0007202148553915322 2023-01-21 13:47:36.104817: step: 2040/529, loss: 0.0003040313604287803 2023-01-21 13:47:37.240210: step: 2044/529, loss: 0.003810310270637274 2023-01-21 13:47:38.362783: step: 2048/529, loss: 0.0016807556385174394 2023-01-21 13:47:39.503582: step: 2052/529, loss: 0.00011568069749046117 2023-01-21 13:47:40.658258: step: 2056/529, loss: 4.625320343620842e-06 2023-01-21 13:47:41.812840: step: 2060/529, loss: 0.007999802008271217 2023-01-21 13:47:42.963521: step: 2064/529, loss: 0.0006359100807458162 2023-01-21 13:47:44.105010: step: 2068/529, loss: 0.001352024031803012 2023-01-21 13:47:45.235913: step: 2072/529, loss: 0.08775339275598526 2023-01-21 13:47:46.400383: step: 2076/529, loss: 5.05924217577558e-05 2023-01-21 13:47:47.565799: step: 2080/529, loss: 0.0003522873157635331 2023-01-21 13:47:48.709472: step: 2084/529, loss: 0.01676921918988228 2023-01-21 13:47:49.824087: step: 2088/529, loss: 9.90867629298009e-05 2023-01-21 13:47:50.948283: step: 2092/529, loss: 0.00012645722017623484 2023-01-21 13:47:52.053410: step: 2096/529, loss: 0.06145744025707245 2023-01-21 13:47:53.192278: step: 2100/529, loss: 0.0030261995270848274 2023-01-21 13:47:54.357125: step: 2104/529, loss: 0.001667976495809853 2023-01-21 13:47:55.500054: step: 2108/529, loss: 0.00508427619934082 2023-01-21 13:47:56.583379: step: 2112/529, loss: 6.67572021484375e-06 2023-01-21 13:47:57.765337: step: 2116/529, loss: 0.0004968643188476562 ================================================== Loss: 0.012 -------------------- Dev: {'event': {'p': 0.6272824919441461, 'r': 0.7776298268974701, 'f1': 0.6944114149821642}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Test: {'event': {'p': 0.6351028216164515, 'r': 0.7694090382387022, 'f1': 0.6958344249410533}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Chinese: {'event': {'p': 0.6153846153846154, 'r': 0.8888888888888888, 'f1': 0.7272727272727274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Korean: {'event': {'p': 0.7115384615384616, 'r': 0.5873015873015873, 'f1': 0.6434782608695652}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Russian: {'event': {'p': 0.46511627906976744, 'r': 0.5555555555555556, 'f1': 0.5063291139240507}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6096033402922756, 'r': 0.7776298268974701, 'f1': 0.6834406085430076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Eng Test for Chinese: {'event': {'p': 0.6149954001839927, 'r': 0.7746234067207416, 'f1': 0.6856410256410257}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Sample Chinese: {'event': {'p': 0.6375, 'r': 0.9444444444444444, 'f1': 0.7611940298507462}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} -------------------- Eng Dev for Korean: {'event': {'p': 0.6104944500504541, 'r': 0.8055925432756325, 'f1': 0.6946039035591274}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Eng Test for Korean: {'event': {'p': 0.61996336996337, 'r': 0.7844727694090382, 'f1': 0.6925831202046036}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Sample Korean: {'event': {'p': 0.6721311475409836, 'r': 0.6507936507936508, 'f1': 0.6612903225806451}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} -------------------- Eng Dev for Russian: {'event': {'p': 0.6103092783505155, 'r': 0.7882822902796272, 'f1': 0.6879721092388147}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Eng Test for Russian: {'event': {'p': 0.6224677716390423, 'r': 0.7833140208574739, 'f1': 0.693689071318625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Sample Russian: {'event': {'p': 0.5405405405405406, 'r': 0.5555555555555556, 'f1': 0.547945205479452}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18}