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 08:25:37.854827: step: 4/529, loss: 3.181039333343506 2023-01-21 08:25:38.984426: step: 8/529, loss: 4.148381233215332 2023-01-21 08:25:40.068900: step: 12/529, loss: 16.890193939208984 2023-01-21 08:25:41.195221: step: 16/529, loss: 14.784475326538086 2023-01-21 08:25:42.293219: step: 20/529, loss: 11.164883613586426 2023-01-21 08:25:43.429850: step: 24/529, loss: 5.907369613647461 2023-01-21 08:25:44.524210: step: 28/529, loss: 4.50492525100708 2023-01-21 08:25:45.632581: step: 32/529, loss: 11.313998222351074 2023-01-21 08:25:46.765386: step: 36/529, loss: 10.420550346374512 2023-01-21 08:25:47.867829: step: 40/529, loss: 3.6236844062805176 2023-01-21 08:25:48.979721: step: 44/529, loss: 3.544320821762085 2023-01-21 08:25:50.089240: step: 48/529, loss: 11.984138488769531 2023-01-21 08:25:51.220254: step: 52/529, loss: 8.288003921508789 2023-01-21 08:25:52.354362: step: 56/529, loss: 8.057869911193848 2023-01-21 08:25:53.439058: step: 60/529, loss: 21.183391571044922 2023-01-21 08:25:54.579333: step: 64/529, loss: 17.67329216003418 2023-01-21 08:25:55.680483: step: 68/529, loss: 4.543078899383545 2023-01-21 08:25:56.769804: step: 72/529, loss: 9.818075180053711 2023-01-21 08:25:57.852680: step: 76/529, loss: 3.9991519451141357 2023-01-21 08:25:58.947266: step: 80/529, loss: 6.863047122955322 2023-01-21 08:26:00.055166: step: 84/529, loss: 10.582906723022461 2023-01-21 08:26:01.147714: step: 88/529, loss: 3.1144561767578125 2023-01-21 08:26:02.267665: step: 92/529, loss: 4.24136209487915 2023-01-21 08:26:03.399908: step: 96/529, loss: 12.366209030151367 2023-01-21 08:26:04.504646: step: 100/529, loss: 24.492359161376953 2023-01-21 08:26:05.611012: step: 104/529, loss: 6.899333477020264 2023-01-21 08:26:06.764034: step: 108/529, loss: 5.953688621520996 2023-01-21 08:26:07.876293: step: 112/529, loss: 23.85785675048828 2023-01-21 08:26:08.980816: step: 116/529, loss: 2.31353497505188 2023-01-21 08:26:10.111658: step: 120/529, loss: 3.1844780445098877 2023-01-21 08:26:11.234203: step: 124/529, loss: 5.4410400390625 2023-01-21 08:26:12.352034: step: 128/529, loss: 3.788682699203491 2023-01-21 08:26:13.475265: step: 132/529, loss: 11.890104293823242 2023-01-21 08:26:14.622963: step: 136/529, loss: 27.87828826904297 2023-01-21 08:26:15.738432: step: 140/529, loss: 13.653682708740234 2023-01-21 08:26:16.860858: step: 144/529, loss: 4.0965728759765625 2023-01-21 08:26:18.018645: step: 148/529, loss: 3.878430128097534 2023-01-21 08:26:19.156535: step: 152/529, loss: 24.20764923095703 2023-01-21 08:26:20.257019: step: 156/529, loss: 3.142374038696289 2023-01-21 08:26:21.347243: step: 160/529, loss: 15.011707305908203 2023-01-21 08:26:22.474404: step: 164/529, loss: 13.921144485473633 2023-01-21 08:26:23.586673: step: 168/529, loss: 10.176070213317871 2023-01-21 08:26:24.692889: step: 172/529, loss: 3.584871530532837 2023-01-21 08:26:25.801661: step: 176/529, loss: 14.051507949829102 2023-01-21 08:26:26.909700: step: 180/529, loss: 3.3138160705566406 2023-01-21 08:26:28.028399: step: 184/529, loss: 16.57437515258789 2023-01-21 08:26:29.113842: step: 188/529, loss: 10.407448768615723 2023-01-21 08:26:30.223786: step: 192/529, loss: 3.171351432800293 2023-01-21 08:26:31.345219: step: 196/529, loss: 35.400482177734375 2023-01-21 08:26:32.492790: step: 200/529, loss: 2.9554643630981445 2023-01-21 08:26:33.620614: step: 204/529, loss: 7.949367523193359 2023-01-21 08:26:34.737388: step: 208/529, loss: 16.064252853393555 2023-01-21 08:26:35.855064: step: 212/529, loss: 20.374942779541016 2023-01-21 08:26:36.958771: step: 216/529, loss: 12.60849666595459 2023-01-21 08:26:38.075853: step: 220/529, loss: 9.891040802001953 2023-01-21 08:26:39.172079: step: 224/529, loss: 19.38750457763672 2023-01-21 08:26:40.309059: step: 228/529, loss: 11.243318557739258 2023-01-21 08:26:41.434264: step: 232/529, loss: 11.916040420532227 2023-01-21 08:26:42.586934: step: 236/529, loss: 4.941216468811035 2023-01-21 08:26:43.717704: step: 240/529, loss: 22.27077293395996 2023-01-21 08:26:44.828586: step: 244/529, loss: 8.686281204223633 2023-01-21 08:26:45.951470: step: 248/529, loss: 2.625258207321167 2023-01-21 08:26:47.088617: step: 252/529, loss: 5.587532997131348 2023-01-21 08:26:48.214049: step: 256/529, loss: 22.95246124267578 2023-01-21 08:26:49.342483: step: 260/529, loss: 3.4424848556518555 2023-01-21 08:26:50.428812: step: 264/529, loss: 23.686487197875977 2023-01-21 08:26:51.563152: step: 268/529, loss: 2.5376386642456055 2023-01-21 08:26:52.759078: step: 272/529, loss: 2.2287583351135254 2023-01-21 08:26:53.858465: step: 276/529, loss: 18.74136734008789 2023-01-21 08:26:54.970417: step: 280/529, loss: 2.306668996810913 2023-01-21 08:26:56.109528: step: 284/529, loss: 40.0789680480957 2023-01-21 08:26:57.237649: step: 288/529, loss: 16.543773651123047 2023-01-21 08:26:58.337340: step: 292/529, loss: 11.939416885375977 2023-01-21 08:26:59.446732: step: 296/529, loss: 3.78615140914917 2023-01-21 08:27:00.586683: step: 300/529, loss: 23.49304962158203 2023-01-21 08:27:01.723272: step: 304/529, loss: 18.0043888092041 2023-01-21 08:27:02.833198: step: 308/529, loss: 14.882118225097656 2023-01-21 08:27:04.011101: step: 312/529, loss: 3.264894485473633 2023-01-21 08:27:05.132735: step: 316/529, loss: 3.409369468688965 2023-01-21 08:27:06.252101: step: 320/529, loss: 6.533860206604004 2023-01-21 08:27:07.358877: step: 324/529, loss: 8.993955612182617 2023-01-21 08:27:08.476539: step: 328/529, loss: 2.536569595336914 2023-01-21 08:27:09.580422: step: 332/529, loss: 8.379501342773438 2023-01-21 08:27:10.706244: step: 336/529, loss: 9.593441009521484 2023-01-21 08:27:11.825168: step: 340/529, loss: 17.786354064941406 2023-01-21 08:27:12.948625: step: 344/529, loss: 3.9665145874023438 2023-01-21 08:27:14.063492: step: 348/529, loss: 3.009249687194824 2023-01-21 08:27:15.184954: step: 352/529, loss: 8.091318130493164 2023-01-21 08:27:16.307602: step: 356/529, loss: 15.7746000289917 2023-01-21 08:27:17.430261: step: 360/529, loss: 3.286609172821045 2023-01-21 08:27:18.536583: step: 364/529, loss: 1.8642345666885376 2023-01-21 08:27:19.660849: step: 368/529, loss: 3.2300446033477783 2023-01-21 08:27:20.791692: step: 372/529, loss: 3.2253260612487793 2023-01-21 08:27:21.913773: step: 376/529, loss: 3.3433282375335693 2023-01-21 08:27:23.039395: step: 380/529, loss: 19.346525192260742 2023-01-21 08:27:24.176155: step: 384/529, loss: 7.14265251159668 2023-01-21 08:27:25.287796: step: 388/529, loss: 3.98744797706604 2023-01-21 08:27:26.425523: step: 392/529, loss: 7.2256550788879395 2023-01-21 08:27:27.512560: step: 396/529, loss: 6.609988212585449 2023-01-21 08:27:28.653113: step: 400/529, loss: 4.187349319458008 2023-01-21 08:27:29.776516: step: 404/529, loss: 6.621717929840088 2023-01-21 08:27:30.948945: step: 408/529, loss: 4.242406368255615 2023-01-21 08:27:32.047320: step: 412/529, loss: 4.040754795074463 2023-01-21 08:27:33.161480: step: 416/529, loss: 10.107298851013184 2023-01-21 08:27:34.292838: step: 420/529, loss: 2.737809181213379 2023-01-21 08:27:35.426126: step: 424/529, loss: 21.25225067138672 2023-01-21 08:27:36.558210: step: 428/529, loss: 13.049256324768066 2023-01-21 08:27:37.657106: step: 432/529, loss: 16.234838485717773 2023-01-21 08:27:38.788167: step: 436/529, loss: 19.74777603149414 2023-01-21 08:27:39.897158: step: 440/529, loss: 3.085660934448242 2023-01-21 08:27:41.071338: step: 444/529, loss: 22.53052520751953 2023-01-21 08:27:42.181052: step: 448/529, loss: 2.378647565841675 2023-01-21 08:27:43.289538: step: 452/529, loss: 12.57361888885498 2023-01-21 08:27:44.408764: step: 456/529, loss: 16.89199447631836 2023-01-21 08:27:45.521548: step: 460/529, loss: 17.527381896972656 2023-01-21 08:27:46.626778: step: 464/529, loss: 3.964460611343384 2023-01-21 08:27:47.744413: step: 468/529, loss: 6.356296539306641 2023-01-21 08:27:48.872868: step: 472/529, loss: 5.459007263183594 2023-01-21 08:27:49.991537: step: 476/529, loss: 6.411293983459473 2023-01-21 08:27:51.100883: step: 480/529, loss: 1.7834112644195557 2023-01-21 08:27:52.212951: step: 484/529, loss: 2.5891640186309814 2023-01-21 08:27:53.311891: step: 488/529, loss: 2.1815664768218994 2023-01-21 08:27:54.410502: step: 492/529, loss: 1.5952057838439941 2023-01-21 08:27:55.536220: step: 496/529, loss: 6.655587196350098 2023-01-21 08:27:56.647179: step: 500/529, loss: 1.0166592597961426 2023-01-21 08:27:57.787735: step: 504/529, loss: 3.089906692504883 2023-01-21 08:27:58.932373: step: 508/529, loss: 5.251044750213623 2023-01-21 08:28:00.040569: step: 512/529, loss: 1.4529364109039307 2023-01-21 08:28:01.169822: step: 516/529, loss: 9.458945274353027 2023-01-21 08:28:02.299521: step: 520/529, loss: 1.1500593423843384 2023-01-21 08:28:03.434716: step: 524/529, loss: 3.7878050804138184 2023-01-21 08:28:04.523093: step: 528/529, loss: 5.998301029205322 2023-01-21 08:28:05.644883: step: 532/529, loss: 0.9841156005859375 2023-01-21 08:28:06.738741: step: 536/529, loss: 3.292997121810913 2023-01-21 08:28:07.874699: step: 540/529, loss: 2.9554991722106934 2023-01-21 08:28:08.999736: step: 544/529, loss: 3.5481033325195312 2023-01-21 08:28:10.140943: step: 548/529, loss: 4.928516864776611 2023-01-21 08:28:11.269889: step: 552/529, loss: 2.248809576034546 2023-01-21 08:28:12.386040: step: 556/529, loss: 0.8243222832679749 2023-01-21 08:28:13.477939: step: 560/529, loss: 0.8489829897880554 2023-01-21 08:28:14.565322: step: 564/529, loss: 0.9535242319107056 2023-01-21 08:28:15.681460: step: 568/529, loss: 0.9830405712127686 2023-01-21 08:28:16.811795: step: 572/529, loss: 1.7841838598251343 2023-01-21 08:28:17.912945: step: 576/529, loss: 0.6517889499664307 2023-01-21 08:28:19.021833: step: 580/529, loss: 0.5556949377059937 2023-01-21 08:28:20.169327: step: 584/529, loss: 4.726264476776123 2023-01-21 08:28:21.289171: step: 588/529, loss: 2.586247444152832 2023-01-21 08:28:22.404031: step: 592/529, loss: 2.321176528930664 2023-01-21 08:28:23.523281: step: 596/529, loss: 3.589031457901001 2023-01-21 08:28:24.637520: step: 600/529, loss: 3.104370355606079 2023-01-21 08:28:25.758186: step: 604/529, loss: 0.33468902111053467 2023-01-21 08:28:26.861831: step: 608/529, loss: 0.5355984568595886 2023-01-21 08:28:27.989310: step: 612/529, loss: 0.7732580900192261 2023-01-21 08:28:29.077319: step: 616/529, loss: 0.4325505197048187 2023-01-21 08:28:30.188981: step: 620/529, loss: 1.6645264625549316 2023-01-21 08:28:31.318448: step: 624/529, loss: 0.6770639419555664 2023-01-21 08:28:32.422897: step: 628/529, loss: 3.181455135345459 2023-01-21 08:28:33.541798: step: 632/529, loss: 2.588618516921997 2023-01-21 08:28:34.667693: step: 636/529, loss: 2.1700592041015625 2023-01-21 08:28:35.774242: step: 640/529, loss: 5.2609734535217285 2023-01-21 08:28:36.900686: step: 644/529, loss: 0.5777246356010437 2023-01-21 08:28:38.018931: step: 648/529, loss: 0.8448666334152222 2023-01-21 08:28:39.127388: step: 652/529, loss: 4.5122222900390625 2023-01-21 08:28:40.257667: step: 656/529, loss: 0.8130577206611633 2023-01-21 08:28:41.345046: step: 660/529, loss: 2.5856754779815674 2023-01-21 08:28:42.448375: step: 664/529, loss: 1.1001721620559692 2023-01-21 08:28:43.560513: step: 668/529, loss: 2.394239902496338 2023-01-21 08:28:44.648867: step: 672/529, loss: 7.212648391723633 2023-01-21 08:28:45.757070: step: 676/529, loss: 0.9773256778717041 2023-01-21 08:28:46.881394: step: 680/529, loss: 2.4082207679748535 2023-01-21 08:28:47.989841: step: 684/529, loss: 0.5685349702835083 2023-01-21 08:28:49.109110: step: 688/529, loss: 0.5630002021789551 2023-01-21 08:28:50.235859: step: 692/529, loss: 1.9442408084869385 2023-01-21 08:28:51.324939: step: 696/529, loss: 1.033196210861206 2023-01-21 08:28:52.440499: step: 700/529, loss: 0.6768302321434021 2023-01-21 08:28:53.540137: step: 704/529, loss: 0.6083481907844543 2023-01-21 08:28:54.694158: step: 708/529, loss: 2.609952211380005 2023-01-21 08:28:55.797650: step: 712/529, loss: 1.4654203653335571 2023-01-21 08:28:56.888354: step: 716/529, loss: 0.6262234449386597 2023-01-21 08:28:58.012327: step: 720/529, loss: 0.6272705793380737 2023-01-21 08:28:59.137857: step: 724/529, loss: 2.0398786067962646 2023-01-21 08:29:00.256298: step: 728/529, loss: 2.28105092048645 2023-01-21 08:29:01.362480: step: 732/529, loss: 1.607466697692871 2023-01-21 08:29:02.459127: step: 736/529, loss: 0.6245580911636353 2023-01-21 08:29:03.576656: step: 740/529, loss: 3.9305334091186523 2023-01-21 08:29:04.675272: step: 744/529, loss: 3.0112438201904297 2023-01-21 08:29:05.767274: step: 748/529, loss: 0.27578938007354736 2023-01-21 08:29:06.901037: step: 752/529, loss: 1.5843188762664795 2023-01-21 08:29:08.054716: step: 756/529, loss: 1.4928663969039917 2023-01-21 08:29:09.171407: step: 760/529, loss: 0.6400212049484253 2023-01-21 08:29:10.323598: step: 764/529, loss: 1.4070143699645996 2023-01-21 08:29:11.479307: step: 768/529, loss: 1.2450342178344727 2023-01-21 08:29:12.593840: step: 772/529, loss: 1.6456449031829834 2023-01-21 08:29:13.698041: step: 776/529, loss: 0.6609829664230347 2023-01-21 08:29:14.823820: step: 780/529, loss: 4.106722354888916 2023-01-21 08:29:15.931849: step: 784/529, loss: 6.329195022583008 2023-01-21 08:29:17.058001: step: 788/529, loss: 5.429108619689941 2023-01-21 08:29:18.173420: step: 792/529, loss: 0.5938752889633179 2023-01-21 08:29:19.315940: step: 796/529, loss: 0.5918770432472229 2023-01-21 08:29:20.459310: step: 800/529, loss: 0.5234934091567993 2023-01-21 08:29:21.583033: step: 804/529, loss: 2.6645586490631104 2023-01-21 08:29:22.677501: step: 808/529, loss: 0.789321780204773 2023-01-21 08:29:23.807965: step: 812/529, loss: 3.8546249866485596 2023-01-21 08:29:24.926261: step: 816/529, loss: 1.5031120777130127 2023-01-21 08:29:26.013396: step: 820/529, loss: 2.1692745685577393 2023-01-21 08:29:27.124175: step: 824/529, loss: 0.46271783113479614 2023-01-21 08:29:28.247517: step: 828/529, loss: 1.9248688220977783 2023-01-21 08:29:29.384631: step: 832/529, loss: 0.9935060739517212 2023-01-21 08:29:30.487701: step: 836/529, loss: 0.5535085201263428 2023-01-21 08:29:31.592567: step: 840/529, loss: 4.084255695343018 2023-01-21 08:29:32.725253: step: 844/529, loss: 1.3110045194625854 2023-01-21 08:29:33.825267: step: 848/529, loss: 0.6956640481948853 2023-01-21 08:29:34.932112: step: 852/529, loss: 1.6176519393920898 2023-01-21 08:29:36.053002: step: 856/529, loss: 2.4390833377838135 2023-01-21 08:29:37.157427: step: 860/529, loss: 2.0054521560668945 2023-01-21 08:29:38.250283: step: 864/529, loss: 0.6310192942619324 2023-01-21 08:29:39.360645: step: 868/529, loss: 2.099611759185791 2023-01-21 08:29:40.461721: step: 872/529, loss: 0.6212886571884155 2023-01-21 08:29:41.572924: step: 876/529, loss: 2.1017985343933105 2023-01-21 08:29:42.668926: step: 880/529, loss: 0.3207300305366516 2023-01-21 08:29:43.776244: step: 884/529, loss: 1.1053235530853271 2023-01-21 08:29:44.863626: step: 888/529, loss: 3.5223312377929688 2023-01-21 08:29:45.980971: step: 892/529, loss: 1.734467625617981 2023-01-21 08:29:47.092931: step: 896/529, loss: 1.2115098237991333 2023-01-21 08:29:48.206394: step: 900/529, loss: 2.927518367767334 2023-01-21 08:29:49.307862: step: 904/529, loss: 0.40983572602272034 2023-01-21 08:29:50.456087: step: 908/529, loss: 0.6246107816696167 2023-01-21 08:29:51.596776: step: 912/529, loss: 0.25027745962142944 2023-01-21 08:29:52.695301: step: 916/529, loss: 0.7158151865005493 2023-01-21 08:29:53.784518: step: 920/529, loss: 0.37871497869491577 2023-01-21 08:29:54.893735: step: 924/529, loss: 2.64094877243042 2023-01-21 08:29:55.996037: step: 928/529, loss: 1.651828646659851 2023-01-21 08:29:57.095862: step: 932/529, loss: 0.42743366956710815 2023-01-21 08:29:58.190943: step: 936/529, loss: 0.5861097574234009 2023-01-21 08:29:59.284431: step: 940/529, loss: 0.5607304573059082 2023-01-21 08:30:00.379027: step: 944/529, loss: 0.8535863757133484 2023-01-21 08:30:01.487490: step: 948/529, loss: 1.1883795261383057 2023-01-21 08:30:02.598781: step: 952/529, loss: 0.43676966428756714 2023-01-21 08:30:03.711623: step: 956/529, loss: 0.616195023059845 2023-01-21 08:30:04.842228: step: 960/529, loss: 2.350834846496582 2023-01-21 08:30:05.955735: step: 964/529, loss: 1.7469561100006104 2023-01-21 08:30:07.050277: step: 968/529, loss: 1.112709879875183 2023-01-21 08:30:08.168887: step: 972/529, loss: 1.9984407424926758 2023-01-21 08:30:09.281685: step: 976/529, loss: 2.033287525177002 2023-01-21 08:30:10.447292: step: 980/529, loss: 7.328566551208496 2023-01-21 08:30:11.604889: step: 984/529, loss: 0.5868288278579712 2023-01-21 08:30:12.714656: step: 988/529, loss: 1.683711051940918 2023-01-21 08:30:13.813135: step: 992/529, loss: 0.4353392720222473 2023-01-21 08:30:14.917599: step: 996/529, loss: 0.43061304092407227 2023-01-21 08:30:16.026209: step: 1000/529, loss: 0.4987958073616028 2023-01-21 08:30:17.140673: step: 1004/529, loss: 2.024533271789551 2023-01-21 08:30:18.295733: step: 1008/529, loss: 1.1182819604873657 2023-01-21 08:30:19.464080: step: 1012/529, loss: 2.887819528579712 2023-01-21 08:30:20.569125: step: 1016/529, loss: 1.126031756401062 2023-01-21 08:30:21.678151: step: 1020/529, loss: 0.7851737141609192 2023-01-21 08:30:22.786552: step: 1024/529, loss: 1.4584814310073853 2023-01-21 08:30:23.920875: step: 1028/529, loss: 0.9805344343185425 2023-01-21 08:30:25.021489: step: 1032/529, loss: 0.5942932367324829 2023-01-21 08:30:26.147449: step: 1036/529, loss: 3.926556348800659 2023-01-21 08:30:27.287026: step: 1040/529, loss: 1.3170838356018066 2023-01-21 08:30:28.379272: step: 1044/529, loss: 1.9300658702850342 2023-01-21 08:30:29.517811: step: 1048/529, loss: 0.864709734916687 2023-01-21 08:30:30.637551: step: 1052/529, loss: 6.743732929229736 2023-01-21 08:30:31.750462: step: 1056/529, loss: 0.4914347529411316 2023-01-21 08:30:32.865577: step: 1060/529, loss: 0.24263663589954376 2023-01-21 08:30:33.982831: step: 1064/529, loss: 0.2656261920928955 2023-01-21 08:30:35.064408: step: 1068/529, loss: 0.5748631954193115 2023-01-21 08:30:36.149206: step: 1072/529, loss: 0.4373299479484558 2023-01-21 08:30:37.260441: step: 1076/529, loss: 1.7935054302215576 2023-01-21 08:30:38.405946: step: 1080/529, loss: 5.035520553588867 2023-01-21 08:30:39.510202: step: 1084/529, loss: 0.4074358344078064 2023-01-21 08:30:40.638906: step: 1088/529, loss: 0.6925681233406067 2023-01-21 08:30:41.744362: step: 1092/529, loss: 1.405806541442871 2023-01-21 08:30:42.865440: step: 1096/529, loss: 1.9536575078964233 2023-01-21 08:30:43.962795: step: 1100/529, loss: 0.8262926936149597 2023-01-21 08:30:45.043167: step: 1104/529, loss: 0.6741202473640442 2023-01-21 08:30:46.184861: step: 1108/529, loss: 0.6606097221374512 2023-01-21 08:30:47.279351: step: 1112/529, loss: 0.7221142649650574 2023-01-21 08:30:48.377810: step: 1116/529, loss: 0.3612290024757385 2023-01-21 08:30:49.475689: step: 1120/529, loss: 0.8333176970481873 2023-01-21 08:30:50.585205: step: 1124/529, loss: 2.135603666305542 2023-01-21 08:30:51.667668: step: 1128/529, loss: 0.4876803457736969 2023-01-21 08:30:52.759854: step: 1132/529, loss: 0.4415576159954071 2023-01-21 08:30:53.878366: step: 1136/529, loss: 0.35199204087257385 2023-01-21 08:30:54.973864: step: 1140/529, loss: 0.773186981678009 2023-01-21 08:30:56.074886: step: 1144/529, loss: 0.3825199007987976 2023-01-21 08:30:57.196654: step: 1148/529, loss: 4.911041736602783 2023-01-21 08:30:58.320424: step: 1152/529, loss: 0.28321534395217896 2023-01-21 08:30:59.413076: step: 1156/529, loss: 1.4916503429412842 2023-01-21 08:31:00.520248: step: 1160/529, loss: 0.3870924711227417 2023-01-21 08:31:01.673711: step: 1164/529, loss: 0.5239297747612 2023-01-21 08:31:02.784782: step: 1168/529, loss: 0.601505696773529 2023-01-21 08:31:03.898910: step: 1172/529, loss: 2.5727901458740234 2023-01-21 08:31:04.982755: step: 1176/529, loss: 0.18190708756446838 2023-01-21 08:31:06.102097: step: 1180/529, loss: 0.9664733409881592 2023-01-21 08:31:07.221070: step: 1184/529, loss: 2.6611785888671875 2023-01-21 08:31:08.326319: step: 1188/529, loss: 0.5117823481559753 2023-01-21 08:31:09.446169: step: 1192/529, loss: 0.6579575538635254 2023-01-21 08:31:10.575377: step: 1196/529, loss: 0.19508495926856995 2023-01-21 08:31:11.712519: step: 1200/529, loss: 1.1678178310394287 2023-01-21 08:31:12.850781: step: 1204/529, loss: 1.455307960510254 2023-01-21 08:31:13.964175: step: 1208/529, loss: 2.3669700622558594 2023-01-21 08:31:15.117049: step: 1212/529, loss: 1.2312983274459839 2023-01-21 08:31:16.218792: step: 1216/529, loss: 1.9613423347473145 2023-01-21 08:31:17.339942: step: 1220/529, loss: 0.9335511922836304 2023-01-21 08:31:18.434028: step: 1224/529, loss: 0.9120486378669739 2023-01-21 08:31:19.522368: step: 1228/529, loss: 4.602816581726074 2023-01-21 08:31:20.684614: step: 1232/529, loss: 1.3636798858642578 2023-01-21 08:31:21.784595: step: 1236/529, loss: 2.050300359725952 2023-01-21 08:31:22.914459: step: 1240/529, loss: 1.53571355342865 2023-01-21 08:31:24.036413: step: 1244/529, loss: 1.6129343509674072 2023-01-21 08:31:25.167546: step: 1248/529, loss: 0.9573297500610352 2023-01-21 08:31:26.250868: step: 1252/529, loss: 1.0015984773635864 2023-01-21 08:31:27.366932: step: 1256/529, loss: 1.0051642656326294 2023-01-21 08:31:28.473339: step: 1260/529, loss: 1.1919217109680176 2023-01-21 08:31:29.571317: step: 1264/529, loss: 1.0392515659332275 2023-01-21 08:31:30.667579: step: 1268/529, loss: 0.1835152804851532 2023-01-21 08:31:31.762466: step: 1272/529, loss: 1.2002732753753662 2023-01-21 08:31:32.898317: step: 1276/529, loss: 1.7138750553131104 2023-01-21 08:31:34.000262: step: 1280/529, loss: 0.7357588410377502 2023-01-21 08:31:35.094600: step: 1284/529, loss: 1.2817983627319336 2023-01-21 08:31:36.177381: step: 1288/529, loss: 8.869969367980957 2023-01-21 08:31:37.292475: step: 1292/529, loss: 0.7919747829437256 2023-01-21 08:31:38.392732: step: 1296/529, loss: 0.2611018121242523 2023-01-21 08:31:39.508187: step: 1300/529, loss: 0.5784379243850708 2023-01-21 08:31:40.642071: step: 1304/529, loss: 3.6391208171844482 2023-01-21 08:31:41.770271: step: 1308/529, loss: 0.409001886844635 2023-01-21 08:31:42.900546: step: 1312/529, loss: 0.5249953269958496 2023-01-21 08:31:43.991202: step: 1316/529, loss: 0.35564833879470825 2023-01-21 08:31:45.141483: step: 1320/529, loss: 0.5548809170722961 2023-01-21 08:31:46.254544: step: 1324/529, loss: 0.7999005913734436 2023-01-21 08:31:47.380954: step: 1328/529, loss: 1.296330213546753 2023-01-21 08:31:48.500653: step: 1332/529, loss: 0.31105977296829224 2023-01-21 08:31:49.612869: step: 1336/529, loss: 1.0918596982955933 2023-01-21 08:31:50.724638: step: 1340/529, loss: 0.5353461503982544 2023-01-21 08:31:51.839475: step: 1344/529, loss: 1.0331121683120728 2023-01-21 08:31:52.983700: step: 1348/529, loss: 0.296822726726532 2023-01-21 08:31:54.101584: step: 1352/529, loss: 2.8359856605529785 2023-01-21 08:31:55.228039: step: 1356/529, loss: 0.17553797364234924 2023-01-21 08:31:56.310248: step: 1360/529, loss: 0.9902456402778625 2023-01-21 08:31:57.439406: step: 1364/529, loss: 0.969699501991272 2023-01-21 08:31:58.558158: step: 1368/529, loss: 1.8549939393997192 2023-01-21 08:31:59.662927: step: 1372/529, loss: 0.30430424213409424 2023-01-21 08:32:00.759098: step: 1376/529, loss: 1.735209345817566 2023-01-21 08:32:01.870243: step: 1380/529, loss: 0.3211958110332489 2023-01-21 08:32:02.979016: step: 1384/529, loss: 0.3108389973640442 2023-01-21 08:32:04.096148: step: 1388/529, loss: 0.7214325070381165 2023-01-21 08:32:05.187115: step: 1392/529, loss: 0.39114007353782654 2023-01-21 08:32:06.310693: step: 1396/529, loss: 0.4138685166835785 2023-01-21 08:32:07.438421: step: 1400/529, loss: 0.2378009408712387 2023-01-21 08:32:08.571506: step: 1404/529, loss: 0.1629185676574707 2023-01-21 08:32:09.699203: step: 1408/529, loss: 0.6681604385375977 2023-01-21 08:32:10.818648: step: 1412/529, loss: 1.066933274269104 2023-01-21 08:32:11.907615: step: 1416/529, loss: 0.5689336657524109 2023-01-21 08:32:13.021881: step: 1420/529, loss: 1.0357985496520996 2023-01-21 08:32:14.162594: step: 1424/529, loss: 9.961202621459961 2023-01-21 08:32:15.278191: step: 1428/529, loss: 1.0123765468597412 2023-01-21 08:32:16.376787: step: 1432/529, loss: 0.2440262883901596 2023-01-21 08:32:17.517713: step: 1436/529, loss: 1.2150473594665527 2023-01-21 08:32:18.638557: step: 1440/529, loss: 0.529533326625824 2023-01-21 08:32:19.760296: step: 1444/529, loss: 0.575178861618042 2023-01-21 08:32:20.860153: step: 1448/529, loss: 0.3072776794433594 2023-01-21 08:32:21.957495: step: 1452/529, loss: 0.33888113498687744 2023-01-21 08:32:23.051896: step: 1456/529, loss: 1.2264699935913086 2023-01-21 08:32:24.181371: step: 1460/529, loss: 1.055977463722229 2023-01-21 08:32:25.348120: step: 1464/529, loss: 0.43419021368026733 2023-01-21 08:32:26.442184: step: 1468/529, loss: 1.4148719310760498 2023-01-21 08:32:27.582013: step: 1472/529, loss: 0.9533913135528564 2023-01-21 08:32:28.693251: step: 1476/529, loss: 0.4055001735687256 2023-01-21 08:32:29.819257: step: 1480/529, loss: 3.253361225128174 2023-01-21 08:32:30.972349: step: 1484/529, loss: 0.47833073139190674 2023-01-21 08:32:32.107410: step: 1488/529, loss: 2.780667781829834 2023-01-21 08:32:33.229424: step: 1492/529, loss: 0.5709744691848755 2023-01-21 08:32:34.332109: step: 1496/529, loss: 0.3533855676651001 2023-01-21 08:32:35.468609: step: 1500/529, loss: 0.5642304420471191 2023-01-21 08:32:36.584588: step: 1504/529, loss: 2.8406996726989746 2023-01-21 08:32:37.705740: step: 1508/529, loss: 2.093322277069092 2023-01-21 08:32:38.815966: step: 1512/529, loss: 1.0322731733322144 2023-01-21 08:32:39.906949: step: 1516/529, loss: 1.583207130432129 2023-01-21 08:32:41.034187: step: 1520/529, loss: 0.9847851991653442 2023-01-21 08:32:42.176013: step: 1524/529, loss: 0.9704713821411133 2023-01-21 08:32:43.292736: step: 1528/529, loss: 0.7138993740081787 2023-01-21 08:32:44.404174: step: 1532/529, loss: 0.4014458656311035 2023-01-21 08:32:45.510411: step: 1536/529, loss: 0.2671958804130554 2023-01-21 08:32:46.612896: step: 1540/529, loss: 0.3832261860370636 2023-01-21 08:32:47.747240: step: 1544/529, loss: 2.7566158771514893 2023-01-21 08:32:48.839083: step: 1548/529, loss: 8.708740234375 2023-01-21 08:32:49.954420: step: 1552/529, loss: 0.5072838068008423 2023-01-21 08:32:51.069662: step: 1556/529, loss: 0.7055354118347168 2023-01-21 08:32:52.198981: step: 1560/529, loss: 1.6156117916107178 2023-01-21 08:32:53.305030: step: 1564/529, loss: 0.3917275369167328 2023-01-21 08:32:54.395451: step: 1568/529, loss: 1.8289592266082764 2023-01-21 08:32:55.511797: step: 1572/529, loss: 0.966462254524231 2023-01-21 08:32:56.608398: step: 1576/529, loss: 1.2601213455200195 2023-01-21 08:32:57.687636: step: 1580/529, loss: 1.5898438692092896 2023-01-21 08:32:58.810474: step: 1584/529, loss: 1.1666059494018555 2023-01-21 08:32:59.922133: step: 1588/529, loss: 1.4859836101531982 2023-01-21 08:33:01.021030: step: 1592/529, loss: 1.085548996925354 2023-01-21 08:33:02.112138: step: 1596/529, loss: 0.14204354584217072 2023-01-21 08:33:03.228123: step: 1600/529, loss: 0.7242662906646729 2023-01-21 08:33:04.340723: step: 1604/529, loss: 2.4299025535583496 2023-01-21 08:33:05.450560: step: 1608/529, loss: 3.8800811767578125 2023-01-21 08:33:06.567068: step: 1612/529, loss: 1.9202399253845215 2023-01-21 08:33:07.684068: step: 1616/529, loss: 1.2000303268432617 2023-01-21 08:33:08.779507: step: 1620/529, loss: 1.6702711582183838 2023-01-21 08:33:09.928648: step: 1624/529, loss: 2.940463066101074 2023-01-21 08:33:11.038736: step: 1628/529, loss: 1.8647282123565674 2023-01-21 08:33:12.167337: step: 1632/529, loss: 1.2855159044265747 2023-01-21 08:33:13.260386: step: 1636/529, loss: 0.27251410484313965 2023-01-21 08:33:14.361687: step: 1640/529, loss: 0.6215863823890686 2023-01-21 08:33:15.477187: step: 1644/529, loss: 0.7701816558837891 2023-01-21 08:33:16.581805: step: 1648/529, loss: 0.5594610571861267 2023-01-21 08:33:17.691468: step: 1652/529, loss: 1.321620225906372 2023-01-21 08:33:18.827663: step: 1656/529, loss: 0.7079131603240967 2023-01-21 08:33:19.943744: step: 1660/529, loss: 0.294409841299057 2023-01-21 08:33:21.044209: step: 1664/529, loss: 0.601701021194458 2023-01-21 08:33:22.148459: step: 1668/529, loss: 0.24158337712287903 2023-01-21 08:33:23.257773: step: 1672/529, loss: 0.8594356179237366 2023-01-21 08:33:24.366026: step: 1676/529, loss: 0.16382399201393127 2023-01-21 08:33:25.502309: step: 1680/529, loss: 0.4015040397644043 2023-01-21 08:33:26.599389: step: 1684/529, loss: 0.24749022722244263 2023-01-21 08:33:27.731707: step: 1688/529, loss: 1.1471235752105713 2023-01-21 08:33:28.825310: step: 1692/529, loss: 0.17843137681484222 2023-01-21 08:33:29.963996: step: 1696/529, loss: 1.7409522533416748 2023-01-21 08:33:31.074537: step: 1700/529, loss: 0.35028308629989624 2023-01-21 08:33:32.185775: step: 1704/529, loss: 5.486498832702637 2023-01-21 08:33:33.326367: step: 1708/529, loss: 3.1866891384124756 2023-01-21 08:33:34.455164: step: 1712/529, loss: 3.423689126968384 2023-01-21 08:33:35.573353: step: 1716/529, loss: 2.252361297607422 2023-01-21 08:33:36.686288: step: 1720/529, loss: 0.9043692350387573 2023-01-21 08:33:37.772326: step: 1724/529, loss: 0.31061768531799316 2023-01-21 08:33:38.892705: step: 1728/529, loss: 0.7108418941497803 2023-01-21 08:33:40.020249: step: 1732/529, loss: 0.41635262966156006 2023-01-21 08:33:41.141405: step: 1736/529, loss: 0.9218595623970032 2023-01-21 08:33:42.249211: step: 1740/529, loss: 0.2296573668718338 2023-01-21 08:33:43.376502: step: 1744/529, loss: 1.355069875717163 2023-01-21 08:33:44.489870: step: 1748/529, loss: 0.502210259437561 2023-01-21 08:33:45.589233: step: 1752/529, loss: 1.055375576019287 2023-01-21 08:33:46.698907: step: 1756/529, loss: 0.25308629870414734 2023-01-21 08:33:47.843646: step: 1760/529, loss: 1.177011489868164 2023-01-21 08:33:48.970436: step: 1764/529, loss: 1.0240637063980103 2023-01-21 08:33:50.126772: step: 1768/529, loss: 3.3754591941833496 2023-01-21 08:33:51.239679: step: 1772/529, loss: 0.40824973583221436 2023-01-21 08:33:52.361915: step: 1776/529, loss: 1.3220211267471313 2023-01-21 08:33:53.475232: step: 1780/529, loss: 0.5259435772895813 2023-01-21 08:33:54.589083: step: 1784/529, loss: 0.2256631851196289 2023-01-21 08:33:55.695551: step: 1788/529, loss: 1.046749234199524 2023-01-21 08:33:56.833920: step: 1792/529, loss: 1.698093056678772 2023-01-21 08:33:57.949629: step: 1796/529, loss: 3.0170767307281494 2023-01-21 08:33:59.077737: step: 1800/529, loss: 2.597223997116089 2023-01-21 08:34:00.184992: step: 1804/529, loss: 2.6770811080932617 2023-01-21 08:34:01.301042: step: 1808/529, loss: 1.8501085042953491 2023-01-21 08:34:02.422100: step: 1812/529, loss: 2.499187469482422 2023-01-21 08:34:03.502376: step: 1816/529, loss: 1.0655276775360107 2023-01-21 08:34:04.646495: step: 1820/529, loss: 1.0780234336853027 2023-01-21 08:34:05.744074: step: 1824/529, loss: 1.364769458770752 2023-01-21 08:34:06.836546: step: 1828/529, loss: 1.3978493213653564 2023-01-21 08:34:07.947616: step: 1832/529, loss: 1.6892147064208984 2023-01-21 08:34:09.030497: step: 1836/529, loss: 0.48332297801971436 2023-01-21 08:34:10.107167: step: 1840/529, loss: 0.4020690321922302 2023-01-21 08:34:11.205422: step: 1844/529, loss: 0.3157860040664673 2023-01-21 08:34:12.358688: step: 1848/529, loss: 3.5969014167785645 2023-01-21 08:34:13.492823: step: 1852/529, loss: 0.256857305765152 2023-01-21 08:34:14.608710: step: 1856/529, loss: 0.2124207466840744 2023-01-21 08:34:15.726981: step: 1860/529, loss: 0.5977795720100403 2023-01-21 08:34:16.836242: step: 1864/529, loss: 1.1359336376190186 2023-01-21 08:34:17.960674: step: 1868/529, loss: 0.8792713284492493 2023-01-21 08:34:19.110523: step: 1872/529, loss: 4.399348735809326 2023-01-21 08:34:20.226038: step: 1876/529, loss: 3.5783493518829346 2023-01-21 08:34:21.336276: step: 1880/529, loss: 0.8334358334541321 2023-01-21 08:34:22.486088: step: 1884/529, loss: 2.101957082748413 2023-01-21 08:34:23.591381: step: 1888/529, loss: 0.38834822177886963 2023-01-21 08:34:24.677887: step: 1892/529, loss: 0.7221038937568665 2023-01-21 08:34:25.809852: step: 1896/529, loss: 1.9736963510513306 2023-01-21 08:34:26.899457: step: 1900/529, loss: 0.29701775312423706 2023-01-21 08:34:27.986163: step: 1904/529, loss: 0.307219535112381 2023-01-21 08:34:29.094080: step: 1908/529, loss: 0.48136910796165466 2023-01-21 08:34:30.236227: step: 1912/529, loss: 2.253765821456909 2023-01-21 08:34:31.378366: step: 1916/529, loss: 2.1686315536499023 2023-01-21 08:34:32.486954: step: 1920/529, loss: 2.344276189804077 2023-01-21 08:34:33.597516: step: 1924/529, loss: 0.6985849142074585 2023-01-21 08:34:34.724657: step: 1928/529, loss: 1.2473480701446533 2023-01-21 08:34:35.822714: step: 1932/529, loss: 1.4237642288208008 2023-01-21 08:34:36.942451: step: 1936/529, loss: 1.9876503944396973 2023-01-21 08:34:38.037546: step: 1940/529, loss: 1.2747325897216797 2023-01-21 08:34:39.149942: step: 1944/529, loss: 0.4919694662094116 2023-01-21 08:34:40.276957: step: 1948/529, loss: 0.2038504183292389 2023-01-21 08:34:41.414616: step: 1952/529, loss: 0.7296062707901001 2023-01-21 08:34:42.506988: step: 1956/529, loss: 1.0391838550567627 2023-01-21 08:34:43.633494: step: 1960/529, loss: 1.391580581665039 2023-01-21 08:34:44.727512: step: 1964/529, loss: 1.7046737670898438 2023-01-21 08:34:45.824504: step: 1968/529, loss: 2.649137020111084 2023-01-21 08:34:46.949830: step: 1972/529, loss: 0.374473512172699 2023-01-21 08:34:48.108732: step: 1976/529, loss: 0.9903206825256348 2023-01-21 08:34:49.227992: step: 1980/529, loss: 0.4089466333389282 2023-01-21 08:34:50.342412: step: 1984/529, loss: 0.5839372873306274 2023-01-21 08:34:51.464801: step: 1988/529, loss: 0.6511991620063782 2023-01-21 08:34:52.586078: step: 1992/529, loss: 7.43670129776001 2023-01-21 08:34:53.704873: step: 1996/529, loss: 2.7200427055358887 2023-01-21 08:34:54.804572: step: 2000/529, loss: 1.0286198854446411 2023-01-21 08:34:55.939193: step: 2004/529, loss: 0.8617674112319946 2023-01-21 08:34:57.060000: step: 2008/529, loss: 1.0878889560699463 2023-01-21 08:34:58.177483: step: 2012/529, loss: 0.11046610027551651 2023-01-21 08:34:59.293309: step: 2016/529, loss: 1.6425484418869019 2023-01-21 08:35:00.410971: step: 2020/529, loss: 0.32265591621398926 2023-01-21 08:35:01.530825: step: 2024/529, loss: 1.2610677480697632 2023-01-21 08:35:02.657860: step: 2028/529, loss: 0.9181358218193054 2023-01-21 08:35:03.766862: step: 2032/529, loss: 1.7233859300613403 2023-01-21 08:35:04.877940: step: 2036/529, loss: 1.2776845693588257 2023-01-21 08:35:05.972807: step: 2040/529, loss: 1.0303120613098145 2023-01-21 08:35:07.076545: step: 2044/529, loss: 1.9361991882324219 2023-01-21 08:35:08.185800: step: 2048/529, loss: 1.1504749059677124 2023-01-21 08:35:09.294385: step: 2052/529, loss: 0.4072703421115875 2023-01-21 08:35:10.412906: step: 2056/529, loss: 0.865858256816864 2023-01-21 08:35:11.571145: step: 2060/529, loss: 0.8232263326644897 2023-01-21 08:35:12.690370: step: 2064/529, loss: 0.28695160150527954 2023-01-21 08:35:13.820439: step: 2068/529, loss: 0.3389263153076172 2023-01-21 08:35:14.924264: step: 2072/529, loss: 1.2257319688796997 2023-01-21 08:35:16.013362: step: 2076/529, loss: 0.27962803840637207 2023-01-21 08:35:17.122222: step: 2080/529, loss: 0.3544461131095886 2023-01-21 08:35:18.208905: step: 2084/529, loss: 1.265763521194458 2023-01-21 08:35:19.313340: step: 2088/529, loss: 0.5422605872154236 2023-01-21 08:35:20.464821: step: 2092/529, loss: 0.6045862436294556 2023-01-21 08:35:21.569443: step: 2096/529, loss: 0.2181510478258133 2023-01-21 08:35:22.676988: step: 2100/529, loss: 1.9928064346313477 2023-01-21 08:35:23.829814: step: 2104/529, loss: 0.5899723768234253 2023-01-21 08:35:24.924838: step: 2108/529, loss: 0.6856964230537415 2023-01-21 08:35:26.038665: step: 2112/529, loss: 0.21868202090263367 2023-01-21 08:35:27.147897: step: 2116/529, loss: 0.9663521647453308 ================================================== Loss: 3.450 -------------------- Dev: {'event': {'p': 0.5475578406169666, 'r': 0.5672436750998668, 'f1': 0.5572269457161544}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Test: {'event': {'p': 0.43515470704410797, 'r': 0.37302483069977427, 'f1': 0.4017016104527499}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Chinese: {'event': {'p': 0.5, 'r': 0.42592592592592593, 'f1': 0.46}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Korean: {'event': {'p': 0.8095238095238095, 'r': 0.2698412698412698, 'f1': 0.40476190476190477}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Russian: {'event': {'p': 0.5, 'r': 0.19444444444444445, 'f1': 0.28}, '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': 0.5475578406169666, 'r': 0.5672436750998668, 'f1': 0.5572269457161544}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Chinese: {'event': {'p': 0.43515470704410797, 'r': 0.37302483069977427, 'f1': 0.4017016104527499}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Chinese: {'event': {'p': 0.5, 'r': 0.42592592592592593, 'f1': 0.46}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Korean: {'event': {'p': 0.5475578406169666, 'r': 0.5672436750998668, 'f1': 0.5572269457161544}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Korean: {'event': {'p': 0.43515470704410797, 'r': 0.37302483069977427, 'f1': 0.4017016104527499}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Korean: {'event': {'p': 0.8095238095238095, 'r': 0.2698412698412698, 'f1': 0.40476190476190477}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Russian: {'event': {'p': 0.5475578406169666, 'r': 0.5672436750998668, 'f1': 0.5572269457161544}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Russian: {'event': {'p': 0.43515470704410797, 'r': 0.37302483069977427, 'f1': 0.4017016104527499}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Russian: {'event': {'p': 0.5, 'r': 0.19444444444444445, 'f1': 0.28}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} ****************************** Epoch: 1 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:36:25.541536: step: 4/529, loss: 0.7036115527153015 2023-01-21 08:36:26.680740: step: 8/529, loss: 0.5496631264686584 2023-01-21 08:36:27.806442: step: 12/529, loss: 0.4532729983329773 2023-01-21 08:36:28.924639: step: 16/529, loss: 0.9465816020965576 2023-01-21 08:36:30.057032: step: 20/529, loss: 0.7065269947052002 2023-01-21 08:36:31.165767: step: 24/529, loss: 0.21801280975341797 2023-01-21 08:36:32.274244: step: 28/529, loss: 2.4378793239593506 2023-01-21 08:36:33.382045: step: 32/529, loss: 0.3654881715774536 2023-01-21 08:36:34.488498: step: 36/529, loss: 7.119620323181152 2023-01-21 08:36:35.623373: step: 40/529, loss: 0.2349395900964737 2023-01-21 08:36:36.731093: step: 44/529, loss: 0.13511833548545837 2023-01-21 08:36:37.861199: step: 48/529, loss: 0.38923513889312744 2023-01-21 08:36:38.993751: step: 52/529, loss: 1.208553433418274 2023-01-21 08:36:40.111535: step: 56/529, loss: 0.284493625164032 2023-01-21 08:36:41.254311: step: 60/529, loss: 0.9339026808738708 2023-01-21 08:36:42.411249: step: 64/529, loss: 1.5366356372833252 2023-01-21 08:36:43.515614: step: 68/529, loss: 0.10496456921100616 2023-01-21 08:36:44.653889: step: 72/529, loss: 0.2413443922996521 2023-01-21 08:36:45.762532: step: 76/529, loss: 0.31122028827667236 2023-01-21 08:36:46.914753: step: 80/529, loss: 1.0785201787948608 2023-01-21 08:36:48.018796: step: 84/529, loss: 0.25327664613723755 2023-01-21 08:36:49.143795: step: 88/529, loss: 0.7175557613372803 2023-01-21 08:36:50.236371: step: 92/529, loss: 0.7942209243774414 2023-01-21 08:36:51.416805: step: 96/529, loss: 0.44551795721054077 2023-01-21 08:36:52.539514: step: 100/529, loss: 0.18662285804748535 2023-01-21 08:36:53.639913: step: 104/529, loss: 1.2460848093032837 2023-01-21 08:36:54.748398: step: 108/529, loss: 0.6176908612251282 2023-01-21 08:36:55.867409: step: 112/529, loss: 2.0322630405426025 2023-01-21 08:36:56.989944: step: 116/529, loss: 0.09562650322914124 2023-01-21 08:36:58.109319: step: 120/529, loss: 0.888964056968689 2023-01-21 08:36:59.204158: step: 124/529, loss: 0.4442671537399292 2023-01-21 08:37:00.300099: step: 128/529, loss: 1.3291382789611816 2023-01-21 08:37:01.402691: step: 132/529, loss: 0.26445913314819336 2023-01-21 08:37:02.504079: step: 136/529, loss: 0.9219101071357727 2023-01-21 08:37:03.624481: step: 140/529, loss: 2.302140474319458 2023-01-21 08:37:04.748965: step: 144/529, loss: 1.288550615310669 2023-01-21 08:37:05.863443: step: 148/529, loss: 0.6995092630386353 2023-01-21 08:37:06.977302: step: 152/529, loss: 0.5388339757919312 2023-01-21 08:37:08.117678: step: 156/529, loss: 0.5267397165298462 2023-01-21 08:37:09.234222: step: 160/529, loss: 0.30286896228790283 2023-01-21 08:37:10.342215: step: 164/529, loss: 0.6007466316223145 2023-01-21 08:37:11.465598: step: 168/529, loss: 0.4984992742538452 2023-01-21 08:37:12.560531: step: 172/529, loss: 0.7836005687713623 2023-01-21 08:37:13.655358: step: 176/529, loss: 3.4949352741241455 2023-01-21 08:37:14.835366: step: 180/529, loss: 0.56257164478302 2023-01-21 08:37:15.965782: step: 184/529, loss: 0.6769910454750061 2023-01-21 08:37:17.095332: step: 188/529, loss: 1.000937819480896 2023-01-21 08:37:18.211747: step: 192/529, loss: 1.4219428300857544 2023-01-21 08:37:19.356813: step: 196/529, loss: 0.5274946093559265 2023-01-21 08:37:20.493269: step: 200/529, loss: 0.49316632747650146 2023-01-21 08:37:21.581686: step: 204/529, loss: 0.39516887068748474 2023-01-21 08:37:22.694514: step: 208/529, loss: 0.16148653626441956 2023-01-21 08:37:23.806273: step: 212/529, loss: 0.33876028656959534 2023-01-21 08:37:24.909176: step: 216/529, loss: 0.39098015427589417 2023-01-21 08:37:26.040462: step: 220/529, loss: 0.14921438694000244 2023-01-21 08:37:27.175804: step: 224/529, loss: 1.6781399250030518 2023-01-21 08:37:28.323748: step: 228/529, loss: 0.11910490691661835 2023-01-21 08:37:29.456879: step: 232/529, loss: 0.7250655293464661 2023-01-21 08:37:30.618402: step: 236/529, loss: 1.2986040115356445 2023-01-21 08:37:31.719800: step: 240/529, loss: 0.24357634782791138 2023-01-21 08:37:32.838992: step: 244/529, loss: 0.35963886976242065 2023-01-21 08:37:33.958921: step: 248/529, loss: 1.0588583946228027 2023-01-21 08:37:35.105735: step: 252/529, loss: 0.2864035665988922 2023-01-21 08:37:36.250807: step: 256/529, loss: 0.6596001386642456 2023-01-21 08:37:37.380104: step: 260/529, loss: 0.4405243992805481 2023-01-21 08:37:38.497667: step: 264/529, loss: 0.3350624144077301 2023-01-21 08:37:39.601568: step: 268/529, loss: 0.3889426589012146 2023-01-21 08:37:40.739739: step: 272/529, loss: 0.231179341673851 2023-01-21 08:37:41.867615: step: 276/529, loss: 0.5403510928153992 2023-01-21 08:37:43.032345: step: 280/529, loss: 1.8436577320098877 2023-01-21 08:37:44.138487: step: 284/529, loss: 0.27053770422935486 2023-01-21 08:37:45.242936: step: 288/529, loss: 0.6908745765686035 2023-01-21 08:37:46.383766: step: 292/529, loss: 0.6526495218276978 2023-01-21 08:37:47.486325: step: 296/529, loss: 0.08405972272157669 2023-01-21 08:37:48.585760: step: 300/529, loss: 0.34178784489631653 2023-01-21 08:37:49.704565: step: 304/529, loss: 1.9574577808380127 2023-01-21 08:37:50.807196: step: 308/529, loss: 0.43951866030693054 2023-01-21 08:37:51.941445: step: 312/529, loss: 1.7129429578781128 2023-01-21 08:37:53.043005: step: 316/529, loss: 0.3309894800186157 2023-01-21 08:37:54.167792: step: 320/529, loss: 1.1224126815795898 2023-01-21 08:37:55.251100: step: 324/529, loss: 0.23305730521678925 2023-01-21 08:37:56.373732: step: 328/529, loss: 0.07745695114135742 2023-01-21 08:37:57.470142: step: 332/529, loss: 6.454632759094238 2023-01-21 08:37:58.563475: step: 336/529, loss: 0.21550145745277405 2023-01-21 08:37:59.669810: step: 340/529, loss: 0.21445322036743164 2023-01-21 08:38:00.793448: step: 344/529, loss: 0.7850074768066406 2023-01-21 08:38:01.884396: step: 348/529, loss: 1.282077431678772 2023-01-21 08:38:02.986149: step: 352/529, loss: 0.6246235370635986 2023-01-21 08:38:04.136277: step: 356/529, loss: 1.216802716255188 2023-01-21 08:38:05.263053: step: 360/529, loss: 0.09026036411523819 2023-01-21 08:38:06.408731: step: 364/529, loss: 0.8711719512939453 2023-01-21 08:38:07.514914: step: 368/529, loss: 0.9840619564056396 2023-01-21 08:38:08.635609: step: 372/529, loss: 0.34870004653930664 2023-01-21 08:38:09.741530: step: 376/529, loss: 1.8122676610946655 2023-01-21 08:38:10.842795: step: 380/529, loss: 0.5293132662773132 2023-01-21 08:38:11.990810: step: 384/529, loss: 1.0776166915893555 2023-01-21 08:38:13.107945: step: 388/529, loss: 1.0728297233581543 2023-01-21 08:38:14.243159: step: 392/529, loss: 0.9263383746147156 2023-01-21 08:38:15.321635: step: 396/529, loss: 6.870106220245361 2023-01-21 08:38:16.422349: step: 400/529, loss: 1.1725728511810303 2023-01-21 08:38:17.558004: step: 404/529, loss: 0.29519081115722656 2023-01-21 08:38:18.673300: step: 408/529, loss: 0.34393128752708435 2023-01-21 08:38:19.779998: step: 412/529, loss: 0.1468048095703125 2023-01-21 08:38:20.954718: step: 416/529, loss: 0.3922603726387024 2023-01-21 08:38:22.077288: step: 420/529, loss: 0.5837979316711426 2023-01-21 08:38:23.203782: step: 424/529, loss: 0.7148324847221375 2023-01-21 08:38:24.300325: step: 428/529, loss: 0.5638807415962219 2023-01-21 08:38:25.422882: step: 432/529, loss: 0.8276436924934387 2023-01-21 08:38:26.535855: step: 436/529, loss: 0.7725765705108643 2023-01-21 08:38:27.675160: step: 440/529, loss: 0.450872004032135 2023-01-21 08:38:28.805563: step: 444/529, loss: 1.2682194709777832 2023-01-21 08:38:29.899983: step: 448/529, loss: 0.31782713532447815 2023-01-21 08:38:31.006210: step: 452/529, loss: 2.5355019569396973 2023-01-21 08:38:32.113216: step: 456/529, loss: 0.14280109107494354 2023-01-21 08:38:33.244151: step: 460/529, loss: 0.26157569885253906 2023-01-21 08:38:34.361795: step: 464/529, loss: 0.9144749641418457 2023-01-21 08:38:35.488652: step: 468/529, loss: 1.2637255191802979 2023-01-21 08:38:36.610233: step: 472/529, loss: 0.7113516330718994 2023-01-21 08:38:37.723924: step: 476/529, loss: 0.3593212366104126 2023-01-21 08:38:38.838937: step: 480/529, loss: 1.1298960447311401 2023-01-21 08:38:39.934226: step: 484/529, loss: 0.4189690947532654 2023-01-21 08:38:41.079134: step: 488/529, loss: 0.3121662139892578 2023-01-21 08:38:42.206271: step: 492/529, loss: 0.3331439197063446 2023-01-21 08:38:43.326443: step: 496/529, loss: 0.16605758666992188 2023-01-21 08:38:44.429472: step: 500/529, loss: 1.9005495309829712 2023-01-21 08:38:45.522854: step: 504/529, loss: 0.245101660490036 2023-01-21 08:38:46.665747: step: 508/529, loss: 0.7208782434463501 2023-01-21 08:38:47.771282: step: 512/529, loss: 0.1765473484992981 2023-01-21 08:38:48.867035: step: 516/529, loss: 0.1999608874320984 2023-01-21 08:38:49.965362: step: 520/529, loss: 0.2706255316734314 2023-01-21 08:38:51.076927: step: 524/529, loss: 0.7563664317131042 2023-01-21 08:38:52.188549: step: 528/529, loss: 0.5105339288711548 2023-01-21 08:38:53.322029: step: 532/529, loss: 0.3250455856323242 2023-01-21 08:38:54.427673: step: 536/529, loss: 0.6623229384422302 2023-01-21 08:38:55.525649: step: 540/529, loss: 0.5985125303268433 2023-01-21 08:38:56.639220: step: 544/529, loss: 0.941774308681488 2023-01-21 08:38:57.745973: step: 548/529, loss: 0.7502085566520691 2023-01-21 08:38:58.876677: step: 552/529, loss: 0.2502526342868805 2023-01-21 08:38:59.973747: step: 556/529, loss: 0.5988870859146118 2023-01-21 08:39:01.094768: step: 560/529, loss: 0.643242597579956 2023-01-21 08:39:02.256157: step: 564/529, loss: 0.32086771726608276 2023-01-21 08:39:03.360741: step: 568/529, loss: 0.28589364886283875 2023-01-21 08:39:04.456550: step: 572/529, loss: 0.3787912130355835 2023-01-21 08:39:05.546364: step: 576/529, loss: 0.2798100709915161 2023-01-21 08:39:06.644917: step: 580/529, loss: 0.6718266010284424 2023-01-21 08:39:07.753797: step: 584/529, loss: 0.6852802634239197 2023-01-21 08:39:08.915881: step: 588/529, loss: 0.17157721519470215 2023-01-21 08:39:10.055597: step: 592/529, loss: 0.26092034578323364 2023-01-21 08:39:11.151183: step: 596/529, loss: 0.21380558609962463 2023-01-21 08:39:12.267306: step: 600/529, loss: 0.24265041947364807 2023-01-21 08:39:13.397528: step: 604/529, loss: 0.41870707273483276 2023-01-21 08:39:14.498045: step: 608/529, loss: 0.29454079270362854 2023-01-21 08:39:15.611364: step: 612/529, loss: 0.19271717965602875 2023-01-21 08:39:16.737414: step: 616/529, loss: 0.480964332818985 2023-01-21 08:39:17.851081: step: 620/529, loss: 0.2303260862827301 2023-01-21 08:39:18.961449: step: 624/529, loss: 0.3906545341014862 2023-01-21 08:39:20.070870: step: 628/529, loss: 0.48213696479797363 2023-01-21 08:39:21.183557: step: 632/529, loss: 0.7085826992988586 2023-01-21 08:39:22.317245: step: 636/529, loss: 0.8184240460395813 2023-01-21 08:39:23.434824: step: 640/529, loss: 0.6251978874206543 2023-01-21 08:39:24.533675: step: 644/529, loss: 0.17595921456813812 2023-01-21 08:39:25.661953: step: 648/529, loss: 0.08084774017333984 2023-01-21 08:39:26.782289: step: 652/529, loss: 1.0904420614242554 2023-01-21 08:39:27.906749: step: 656/529, loss: 1.6529580354690552 2023-01-21 08:39:29.025049: step: 660/529, loss: 1.0905689001083374 2023-01-21 08:39:30.147708: step: 664/529, loss: 0.6265753507614136 2023-01-21 08:39:31.262538: step: 668/529, loss: 0.6836234927177429 2023-01-21 08:39:32.367188: step: 672/529, loss: 0.3016183078289032 2023-01-21 08:39:33.482485: step: 676/529, loss: 6.876376152038574 2023-01-21 08:39:34.587469: step: 680/529, loss: 0.21166175603866577 2023-01-21 08:39:35.712037: step: 684/529, loss: 0.30103379487991333 2023-01-21 08:39:36.839646: step: 688/529, loss: 0.19192494451999664 2023-01-21 08:39:37.959623: step: 692/529, loss: 0.15551204979419708 2023-01-21 08:39:39.129369: step: 696/529, loss: 1.1883652210235596 2023-01-21 08:39:40.244455: step: 700/529, loss: 0.3992368280887604 2023-01-21 08:39:41.351983: step: 704/529, loss: 2.868074655532837 2023-01-21 08:39:42.474941: step: 708/529, loss: 0.9678558111190796 2023-01-21 08:39:43.607589: step: 712/529, loss: 0.5711044073104858 2023-01-21 08:39:44.698518: step: 716/529, loss: 0.1846696436405182 2023-01-21 08:39:45.807042: step: 720/529, loss: 1.7841217517852783 2023-01-21 08:39:46.919900: step: 724/529, loss: 0.3993869721889496 2023-01-21 08:39:48.082149: step: 728/529, loss: 0.601866602897644 2023-01-21 08:39:49.198494: step: 732/529, loss: 0.5240635871887207 2023-01-21 08:39:50.310970: step: 736/529, loss: 0.44053757190704346 2023-01-21 08:39:51.444112: step: 740/529, loss: 0.28165900707244873 2023-01-21 08:39:52.550390: step: 744/529, loss: 0.5654706954956055 2023-01-21 08:39:53.652286: step: 748/529, loss: 0.3745911717414856 2023-01-21 08:39:54.772029: step: 752/529, loss: 0.24808111786842346 2023-01-21 08:39:55.867119: step: 756/529, loss: 0.35858869552612305 2023-01-21 08:39:56.968897: step: 760/529, loss: 0.223505899310112 2023-01-21 08:39:58.094587: step: 764/529, loss: 0.7637803554534912 2023-01-21 08:39:59.229367: step: 768/529, loss: 1.00881028175354 2023-01-21 08:40:00.324486: step: 772/529, loss: 0.33720290660858154 2023-01-21 08:40:01.443966: step: 776/529, loss: 0.26835161447525024 2023-01-21 08:40:02.550405: step: 780/529, loss: 0.20082814991474152 2023-01-21 08:40:03.639583: step: 784/529, loss: 0.5037037134170532 2023-01-21 08:40:04.745814: step: 788/529, loss: 0.08456037193536758 2023-01-21 08:40:05.833563: step: 792/529, loss: 0.7846124172210693 2023-01-21 08:40:06.936178: step: 796/529, loss: 0.17308759689331055 2023-01-21 08:40:08.031529: step: 800/529, loss: 0.20586219429969788 2023-01-21 08:40:09.134933: step: 804/529, loss: 0.5375609397888184 2023-01-21 08:40:10.255976: step: 808/529, loss: 0.14410077035427094 2023-01-21 08:40:11.375852: step: 812/529, loss: 2.275195598602295 2023-01-21 08:40:12.474283: step: 816/529, loss: 0.358944296836853 2023-01-21 08:40:13.562370: step: 820/529, loss: 1.0653698444366455 2023-01-21 08:40:14.718137: step: 824/529, loss: 0.3719783425331116 2023-01-21 08:40:15.846929: step: 828/529, loss: 0.17783474922180176 2023-01-21 08:40:16.951400: step: 832/529, loss: 0.2557685971260071 2023-01-21 08:40:18.085696: step: 836/529, loss: 3.7622501850128174 2023-01-21 08:40:19.200058: step: 840/529, loss: 0.24286863207817078 2023-01-21 08:40:20.297701: step: 844/529, loss: 0.7735523581504822 2023-01-21 08:40:21.410821: step: 848/529, loss: 2.3334944248199463 2023-01-21 08:40:22.519983: step: 852/529, loss: 0.25273099541664124 2023-01-21 08:40:23.638056: step: 856/529, loss: 1.5338503122329712 2023-01-21 08:40:24.753162: step: 860/529, loss: 0.5666224956512451 2023-01-21 08:40:25.868930: step: 864/529, loss: 0.3180598020553589 2023-01-21 08:40:27.040453: step: 868/529, loss: 0.6632059812545776 2023-01-21 08:40:28.138171: step: 872/529, loss: 1.041609764099121 2023-01-21 08:40:29.276776: step: 876/529, loss: 0.8791897296905518 2023-01-21 08:40:30.368427: step: 880/529, loss: 2.2251648902893066 2023-01-21 08:40:31.506065: step: 884/529, loss: 0.2375658005475998 2023-01-21 08:40:32.618853: step: 888/529, loss: 0.3670843243598938 2023-01-21 08:40:33.759958: step: 892/529, loss: 0.44228315353393555 2023-01-21 08:40:34.873950: step: 896/529, loss: 0.6931800246238708 2023-01-21 08:40:35.991135: step: 900/529, loss: 0.6793521642684937 2023-01-21 08:40:37.116075: step: 904/529, loss: 0.8143658638000488 2023-01-21 08:40:38.244084: step: 908/529, loss: 0.1742684543132782 2023-01-21 08:40:39.367140: step: 912/529, loss: 1.215588092803955 2023-01-21 08:40:40.455321: step: 916/529, loss: 0.49752432107925415 2023-01-21 08:40:41.618188: step: 920/529, loss: 0.37126654386520386 2023-01-21 08:40:42.749838: step: 924/529, loss: 0.32738104462623596 2023-01-21 08:40:43.885058: step: 928/529, loss: 1.0940759181976318 2023-01-21 08:40:45.015391: step: 932/529, loss: 2.2437081336975098 2023-01-21 08:40:46.127929: step: 936/529, loss: 0.49070829153060913 2023-01-21 08:40:47.231949: step: 940/529, loss: 0.29194411635398865 2023-01-21 08:40:48.360984: step: 944/529, loss: 0.8337479829788208 2023-01-21 08:40:49.472549: step: 948/529, loss: 2.0440027713775635 2023-01-21 08:40:50.638977: step: 952/529, loss: 0.9109687805175781 2023-01-21 08:40:51.780010: step: 956/529, loss: 0.5357109308242798 2023-01-21 08:40:52.895995: step: 960/529, loss: 0.4600534439086914 2023-01-21 08:40:54.044136: step: 964/529, loss: 7.5580549240112305 2023-01-21 08:40:55.179705: step: 968/529, loss: 0.6537569761276245 2023-01-21 08:40:56.313751: step: 972/529, loss: 1.4608421325683594 2023-01-21 08:40:57.458160: step: 976/529, loss: 0.3898892402648926 2023-01-21 08:40:58.564861: step: 980/529, loss: 0.9173405170440674 2023-01-21 08:40:59.669709: step: 984/529, loss: 2.3218090534210205 2023-01-21 08:41:00.815736: step: 988/529, loss: 1.6060621738433838 2023-01-21 08:41:01.909836: step: 992/529, loss: 0.3135437071323395 2023-01-21 08:41:03.024151: step: 996/529, loss: 0.4413120448589325 2023-01-21 08:41:04.156878: step: 1000/529, loss: 0.23012995719909668 2023-01-21 08:41:05.297505: step: 1004/529, loss: 1.0736677646636963 2023-01-21 08:41:06.418046: step: 1008/529, loss: 0.35119175910949707 2023-01-21 08:41:07.576900: step: 1012/529, loss: 0.3621875047683716 2023-01-21 08:41:08.668620: step: 1016/529, loss: 1.4045734405517578 2023-01-21 08:41:09.778400: step: 1020/529, loss: 0.14289188385009766 2023-01-21 08:41:10.913460: step: 1024/529, loss: 1.5238171815872192 2023-01-21 08:41:12.034701: step: 1028/529, loss: 0.255527138710022 2023-01-21 08:41:13.170466: step: 1032/529, loss: 2.3973302841186523 2023-01-21 08:41:14.281478: step: 1036/529, loss: 0.24948197603225708 2023-01-21 08:41:15.418658: step: 1040/529, loss: 0.7357456088066101 2023-01-21 08:41:16.516910: step: 1044/529, loss: 0.5525869131088257 2023-01-21 08:41:17.633564: step: 1048/529, loss: 0.28170856833457947 2023-01-21 08:41:18.769397: step: 1052/529, loss: 0.14208835363388062 2023-01-21 08:41:19.887355: step: 1056/529, loss: 1.1579396724700928 2023-01-21 08:41:21.009962: step: 1060/529, loss: 6.846298694610596 2023-01-21 08:41:22.120475: step: 1064/529, loss: 0.5858241319656372 2023-01-21 08:41:23.230800: step: 1068/529, loss: 1.2356059551239014 2023-01-21 08:41:24.322686: step: 1072/529, loss: 1.5922828912734985 2023-01-21 08:41:25.441738: step: 1076/529, loss: 0.1442646086215973 2023-01-21 08:41:26.566495: step: 1080/529, loss: 0.5200090408325195 2023-01-21 08:41:27.720496: step: 1084/529, loss: 2.0759453773498535 2023-01-21 08:41:28.860907: step: 1088/529, loss: 0.16801553964614868 2023-01-21 08:41:29.974234: step: 1092/529, loss: 1.6849634647369385 2023-01-21 08:41:31.081505: step: 1096/529, loss: 0.699837863445282 2023-01-21 08:41:32.181008: step: 1100/529, loss: 0.3389669358730316 2023-01-21 08:41:33.316637: step: 1104/529, loss: 2.048841953277588 2023-01-21 08:41:34.422007: step: 1108/529, loss: 0.4343034625053406 2023-01-21 08:41:35.544055: step: 1112/529, loss: 0.14017248153686523 2023-01-21 08:41:36.692619: step: 1116/529, loss: 0.22366638481616974 2023-01-21 08:41:37.802144: step: 1120/529, loss: 0.15395641326904297 2023-01-21 08:41:38.886508: step: 1124/529, loss: 0.24705466628074646 2023-01-21 08:41:40.022148: step: 1128/529, loss: 0.250865638256073 2023-01-21 08:41:41.117576: step: 1132/529, loss: 0.22677364945411682 2023-01-21 08:41:42.229743: step: 1136/529, loss: 0.32051271200180054 2023-01-21 08:41:43.317953: step: 1140/529, loss: 0.2160608172416687 2023-01-21 08:41:44.454965: step: 1144/529, loss: 1.5062017440795898 2023-01-21 08:41:45.588722: step: 1148/529, loss: 2.317227363586426 2023-01-21 08:41:46.729990: step: 1152/529, loss: 0.08393897861242294 2023-01-21 08:41:47.845966: step: 1156/529, loss: 0.3460056483745575 2023-01-21 08:41:48.989880: step: 1160/529, loss: 0.5972245931625366 2023-01-21 08:41:50.120761: step: 1164/529, loss: 0.35770758986473083 2023-01-21 08:41:51.226361: step: 1168/529, loss: 0.3233143985271454 2023-01-21 08:41:52.366659: step: 1172/529, loss: 0.1382053792476654 2023-01-21 08:41:53.471257: step: 1176/529, loss: 1.7616643905639648 2023-01-21 08:41:54.568715: step: 1180/529, loss: 1.4795652627944946 2023-01-21 08:41:55.668868: step: 1184/529, loss: 0.4514956474304199 2023-01-21 08:41:56.787351: step: 1188/529, loss: 0.11908464133739471 2023-01-21 08:41:57.886276: step: 1192/529, loss: 1.882380723953247 2023-01-21 08:41:59.005710: step: 1196/529, loss: 0.7030121088027954 2023-01-21 08:42:00.125603: step: 1200/529, loss: 0.26276880502700806 2023-01-21 08:42:01.236750: step: 1204/529, loss: 0.4116804599761963 2023-01-21 08:42:02.331812: step: 1208/529, loss: 0.8785880208015442 2023-01-21 08:42:03.425974: step: 1212/529, loss: 1.0726191997528076 2023-01-21 08:42:04.533018: step: 1216/529, loss: 0.46557560563087463 2023-01-21 08:42:05.659323: step: 1220/529, loss: 1.3430287837982178 2023-01-21 08:42:06.743968: step: 1224/529, loss: 0.3480450510978699 2023-01-21 08:42:07.835302: step: 1228/529, loss: 0.1810993254184723 2023-01-21 08:42:08.943224: step: 1232/529, loss: 1.1802563667297363 2023-01-21 08:42:10.054221: step: 1236/529, loss: 1.2568211555480957 2023-01-21 08:42:11.174371: step: 1240/529, loss: 0.7923814058303833 2023-01-21 08:42:12.282658: step: 1244/529, loss: 0.1576269268989563 2023-01-21 08:42:13.373243: step: 1248/529, loss: 0.13066306710243225 2023-01-21 08:42:14.483595: step: 1252/529, loss: 0.3931480646133423 2023-01-21 08:42:15.606895: step: 1256/529, loss: 0.09242627769708633 2023-01-21 08:42:16.681797: step: 1260/529, loss: 0.16336041688919067 2023-01-21 08:42:17.793053: step: 1264/529, loss: 0.13587962090969086 2023-01-21 08:42:18.906126: step: 1268/529, loss: 0.27442023158073425 2023-01-21 08:42:20.007262: step: 1272/529, loss: 0.3076659142971039 2023-01-21 08:42:21.120572: step: 1276/529, loss: 0.22706851363182068 2023-01-21 08:42:22.231232: step: 1280/529, loss: 0.2371736615896225 2023-01-21 08:42:23.335432: step: 1284/529, loss: 0.15129134058952332 2023-01-21 08:42:24.483622: step: 1288/529, loss: 0.9143304824829102 2023-01-21 08:42:25.596562: step: 1292/529, loss: 0.3363853693008423 2023-01-21 08:42:26.735850: step: 1296/529, loss: 0.5157318115234375 2023-01-21 08:42:27.825596: step: 1300/529, loss: 0.1957314908504486 2023-01-21 08:42:28.922271: step: 1304/529, loss: 0.29106923937797546 2023-01-21 08:42:30.036761: step: 1308/529, loss: 1.2171783447265625 2023-01-21 08:42:31.153621: step: 1312/529, loss: 0.37879467010498047 2023-01-21 08:42:32.248454: step: 1316/529, loss: 0.15320749580860138 2023-01-21 08:42:33.390419: step: 1320/529, loss: 0.16589584946632385 2023-01-21 08:42:34.521496: step: 1324/529, loss: 1.137056589126587 2023-01-21 08:42:35.644290: step: 1328/529, loss: 0.20302076637744904 2023-01-21 08:42:36.758985: step: 1332/529, loss: 0.21741986274719238 2023-01-21 08:42:37.850808: step: 1336/529, loss: 0.25976860523223877 2023-01-21 08:42:39.003565: step: 1340/529, loss: 2.7457759380340576 2023-01-21 08:42:40.133294: step: 1344/529, loss: 1.6406248807907104 2023-01-21 08:42:41.265018: step: 1348/529, loss: 0.2748958170413971 2023-01-21 08:42:42.379859: step: 1352/529, loss: 1.4270340204238892 2023-01-21 08:42:43.529147: step: 1356/529, loss: 1.1498936414718628 2023-01-21 08:42:44.668019: step: 1360/529, loss: 0.19536876678466797 2023-01-21 08:42:45.791031: step: 1364/529, loss: 0.668630838394165 2023-01-21 08:42:46.892926: step: 1368/529, loss: 0.14152947068214417 2023-01-21 08:42:47.989393: step: 1372/529, loss: 0.7408544421195984 2023-01-21 08:42:49.087819: step: 1376/529, loss: 0.27183055877685547 2023-01-21 08:42:50.226268: step: 1380/529, loss: 0.3502856492996216 2023-01-21 08:42:51.343701: step: 1384/529, loss: 0.27408719062805176 2023-01-21 08:42:52.449220: step: 1388/529, loss: 0.31254032254219055 2023-01-21 08:42:53.559891: step: 1392/529, loss: 0.11689014732837677 2023-01-21 08:42:54.671614: step: 1396/529, loss: 0.15305423736572266 2023-01-21 08:42:55.802265: step: 1400/529, loss: 0.8012265563011169 2023-01-21 08:42:56.903744: step: 1404/529, loss: 2.31156325340271 2023-01-21 08:42:58.056815: step: 1408/529, loss: 0.374700129032135 2023-01-21 08:42:59.157736: step: 1412/529, loss: 0.23223847150802612 2023-01-21 08:43:00.330901: step: 1416/529, loss: 0.11860805004835129 2023-01-21 08:43:01.434858: step: 1420/529, loss: 0.4748643636703491 2023-01-21 08:43:02.548722: step: 1424/529, loss: 0.1042964980006218 2023-01-21 08:43:03.692266: step: 1428/529, loss: 0.7844324707984924 2023-01-21 08:43:04.802394: step: 1432/529, loss: 0.13150271773338318 2023-01-21 08:43:05.942705: step: 1436/529, loss: 0.25126132369041443 2023-01-21 08:43:07.089515: step: 1440/529, loss: 0.2835647463798523 2023-01-21 08:43:08.189822: step: 1444/529, loss: 0.05659355968236923 2023-01-21 08:43:09.305534: step: 1448/529, loss: 0.9214946627616882 2023-01-21 08:43:10.474373: step: 1452/529, loss: 0.12622341513633728 2023-01-21 08:43:11.573480: step: 1456/529, loss: 0.7260076999664307 2023-01-21 08:43:12.651446: step: 1460/529, loss: 0.08598079532384872 2023-01-21 08:43:13.789790: step: 1464/529, loss: 0.5306049585342407 2023-01-21 08:43:14.917601: step: 1468/529, loss: 0.19852668046951294 2023-01-21 08:43:16.024361: step: 1472/529, loss: 1.3862760066986084 2023-01-21 08:43:17.164293: step: 1476/529, loss: 0.8602093458175659 2023-01-21 08:43:18.241546: step: 1480/529, loss: 0.1600692868232727 2023-01-21 08:43:19.336463: step: 1484/529, loss: 0.8565475344657898 2023-01-21 08:43:20.437586: step: 1488/529, loss: 0.11720190942287445 2023-01-21 08:43:21.559287: step: 1492/529, loss: 0.45756617188453674 2023-01-21 08:43:22.675592: step: 1496/529, loss: 6.647948265075684 2023-01-21 08:43:23.804665: step: 1500/529, loss: 0.2115098088979721 2023-01-21 08:43:24.891732: step: 1504/529, loss: 0.16186895966529846 2023-01-21 08:43:26.003962: step: 1508/529, loss: 0.6816369891166687 2023-01-21 08:43:27.106041: step: 1512/529, loss: 0.4905688166618347 2023-01-21 08:43:28.227184: step: 1516/529, loss: 0.8361889123916626 2023-01-21 08:43:29.368201: step: 1520/529, loss: 0.10857601463794708 2023-01-21 08:43:30.476182: step: 1524/529, loss: 0.6395866870880127 2023-01-21 08:43:31.579391: step: 1528/529, loss: 0.6413912773132324 2023-01-21 08:43:32.712451: step: 1532/529, loss: 0.12036089599132538 2023-01-21 08:43:33.825442: step: 1536/529, loss: 0.7675918936729431 2023-01-21 08:43:34.943485: step: 1540/529, loss: 0.33137089014053345 2023-01-21 08:43:36.076592: step: 1544/529, loss: 0.19094133377075195 2023-01-21 08:43:37.198695: step: 1548/529, loss: 0.2136303335428238 2023-01-21 08:43:38.311102: step: 1552/529, loss: 0.12485656142234802 2023-01-21 08:43:39.436274: step: 1556/529, loss: 0.8323894143104553 2023-01-21 08:43:40.535934: step: 1560/529, loss: 0.35746800899505615 2023-01-21 08:43:41.661712: step: 1564/529, loss: 2.2854628562927246 2023-01-21 08:43:42.767966: step: 1568/529, loss: 1.7288563251495361 2023-01-21 08:43:43.901579: step: 1572/529, loss: 0.09137392044067383 2023-01-21 08:43:44.998474: step: 1576/529, loss: 0.2721441388130188 2023-01-21 08:43:46.153057: step: 1580/529, loss: 1.2794252634048462 2023-01-21 08:43:47.274661: step: 1584/529, loss: 2.586719274520874 2023-01-21 08:43:48.386063: step: 1588/529, loss: 0.2548113167285919 2023-01-21 08:43:49.494722: step: 1592/529, loss: 1.0186208486557007 2023-01-21 08:43:50.642624: step: 1596/529, loss: 1.733971357345581 2023-01-21 08:43:51.752070: step: 1600/529, loss: 0.9125651121139526 2023-01-21 08:43:52.858465: step: 1604/529, loss: 0.6683480739593506 2023-01-21 08:43:54.013471: step: 1608/529, loss: 0.15578250586986542 2023-01-21 08:43:55.147322: step: 1612/529, loss: 0.1358446180820465 2023-01-21 08:43:56.279825: step: 1616/529, loss: 0.3594614863395691 2023-01-21 08:43:57.414351: step: 1620/529, loss: 0.1455286145210266 2023-01-21 08:43:58.537506: step: 1624/529, loss: 0.8110268712043762 2023-01-21 08:43:59.678495: step: 1628/529, loss: 0.08677692711353302 2023-01-21 08:44:00.804446: step: 1632/529, loss: 0.521518886089325 2023-01-21 08:44:01.910306: step: 1636/529, loss: 0.1192840114235878 2023-01-21 08:44:03.008653: step: 1640/529, loss: 0.1738756150007248 2023-01-21 08:44:04.147675: step: 1644/529, loss: 0.8496429324150085 2023-01-21 08:44:05.255916: step: 1648/529, loss: 0.2477641999721527 2023-01-21 08:44:06.382030: step: 1652/529, loss: 0.3932759463787079 2023-01-21 08:44:07.512486: step: 1656/529, loss: 0.21739532053470612 2023-01-21 08:44:08.609753: step: 1660/529, loss: 0.6948230862617493 2023-01-21 08:44:09.734691: step: 1664/529, loss: 0.48446816205978394 2023-01-21 08:44:10.837858: step: 1668/529, loss: 1.647700309753418 2023-01-21 08:44:11.969680: step: 1672/529, loss: 0.3154051899909973 2023-01-21 08:44:13.093527: step: 1676/529, loss: 0.22243304550647736 2023-01-21 08:44:14.191652: step: 1680/529, loss: 0.14397311210632324 2023-01-21 08:44:15.282777: step: 1684/529, loss: 0.10298324376344681 2023-01-21 08:44:16.379992: step: 1688/529, loss: 0.3207573890686035 2023-01-21 08:44:17.498155: step: 1692/529, loss: 0.3615146279335022 2023-01-21 08:44:18.617619: step: 1696/529, loss: 0.44014739990234375 2023-01-21 08:44:19.758192: step: 1700/529, loss: 0.19119468331336975 2023-01-21 08:44:20.882946: step: 1704/529, loss: 0.6909996271133423 2023-01-21 08:44:22.000114: step: 1708/529, loss: 0.5721532106399536 2023-01-21 08:44:23.128907: step: 1712/529, loss: 0.3849565386772156 2023-01-21 08:44:24.242156: step: 1716/529, loss: 2.519810914993286 2023-01-21 08:44:25.337397: step: 1720/529, loss: 0.1966181844472885 2023-01-21 08:44:26.440388: step: 1724/529, loss: 0.22541581094264984 2023-01-21 08:44:27.558011: step: 1728/529, loss: 0.2048512101173401 2023-01-21 08:44:28.695666: step: 1732/529, loss: 0.2168704867362976 2023-01-21 08:44:29.823309: step: 1736/529, loss: 0.5078675150871277 2023-01-21 08:44:30.983311: step: 1740/529, loss: 0.21213750541210175 2023-01-21 08:44:32.098606: step: 1744/529, loss: 0.3765409588813782 2023-01-21 08:44:33.259559: step: 1748/529, loss: 0.9130523800849915 2023-01-21 08:44:34.371778: step: 1752/529, loss: 0.5442807078361511 2023-01-21 08:44:35.475031: step: 1756/529, loss: 0.4957461953163147 2023-01-21 08:44:36.575194: step: 1760/529, loss: 0.13172730803489685 2023-01-21 08:44:37.697189: step: 1764/529, loss: 0.7788804173469543 2023-01-21 08:44:38.797741: step: 1768/529, loss: 1.0385152101516724 2023-01-21 08:44:39.923649: step: 1772/529, loss: 0.15449944138526917 2023-01-21 08:44:41.036118: step: 1776/529, loss: 0.6412561535835266 2023-01-21 08:44:42.156745: step: 1780/529, loss: 0.39307743310928345 2023-01-21 08:44:43.250332: step: 1784/529, loss: 0.166100412607193 2023-01-21 08:44:44.370760: step: 1788/529, loss: 0.7248702645301819 2023-01-21 08:44:45.497854: step: 1792/529, loss: 0.5662744641304016 2023-01-21 08:44:46.598578: step: 1796/529, loss: 0.44758567214012146 2023-01-21 08:44:47.712264: step: 1800/529, loss: 0.33152151107788086 2023-01-21 08:44:48.846846: step: 1804/529, loss: 0.7885726690292358 2023-01-21 08:44:49.951631: step: 1808/529, loss: 0.6790496706962585 2023-01-21 08:44:51.081828: step: 1812/529, loss: 1.0043998956680298 2023-01-21 08:44:52.225136: step: 1816/529, loss: 0.3004891872406006 2023-01-21 08:44:53.347038: step: 1820/529, loss: 0.1599387228488922 2023-01-21 08:44:54.454121: step: 1824/529, loss: 0.2533267140388489 2023-01-21 08:44:55.581638: step: 1828/529, loss: 0.7207271456718445 2023-01-21 08:44:56.687706: step: 1832/529, loss: 0.24773702025413513 2023-01-21 08:44:57.814120: step: 1836/529, loss: 0.7244818806648254 2023-01-21 08:44:58.911122: step: 1840/529, loss: 0.6834341883659363 2023-01-21 08:45:00.036107: step: 1844/529, loss: 1.0085415840148926 2023-01-21 08:45:01.133451: step: 1848/529, loss: 0.2685631215572357 2023-01-21 08:45:02.255933: step: 1852/529, loss: 1.247779369354248 2023-01-21 08:45:03.354684: step: 1856/529, loss: 0.6834778189659119 2023-01-21 08:45:04.453286: step: 1860/529, loss: 0.1104244738817215 2023-01-21 08:45:05.562618: step: 1864/529, loss: 0.5080947875976562 2023-01-21 08:45:06.642065: step: 1868/529, loss: 0.23532609641551971 2023-01-21 08:45:07.754981: step: 1872/529, loss: 0.1740579605102539 2023-01-21 08:45:08.870690: step: 1876/529, loss: 0.046334266662597656 2023-01-21 08:45:09.992684: step: 1880/529, loss: 0.26787012815475464 2023-01-21 08:45:11.100656: step: 1884/529, loss: 0.4222313165664673 2023-01-21 08:45:12.207854: step: 1888/529, loss: 0.39106273651123047 2023-01-21 08:45:13.309441: step: 1892/529, loss: 0.09448876976966858 2023-01-21 08:45:14.440605: step: 1896/529, loss: 0.19445809721946716 2023-01-21 08:45:15.542353: step: 1900/529, loss: 0.5161245465278625 2023-01-21 08:45:16.658307: step: 1904/529, loss: 0.3839361369609833 2023-01-21 08:45:17.791213: step: 1908/529, loss: 2.7287416458129883 2023-01-21 08:45:18.942967: step: 1912/529, loss: 2.105461359024048 2023-01-21 08:45:20.047234: step: 1916/529, loss: 0.1062900573015213 2023-01-21 08:45:21.183108: step: 1920/529, loss: 0.1700780987739563 2023-01-21 08:45:22.299236: step: 1924/529, loss: 0.550543487071991 2023-01-21 08:45:23.396787: step: 1928/529, loss: 0.5219593048095703 2023-01-21 08:45:24.508042: step: 1932/529, loss: 0.08461695164442062 2023-01-21 08:45:25.624341: step: 1936/529, loss: 0.3319927453994751 2023-01-21 08:45:26.712809: step: 1940/529, loss: 1.2415722608566284 2023-01-21 08:45:27.815542: step: 1944/529, loss: 0.19582301378250122 2023-01-21 08:45:28.961733: step: 1948/529, loss: 0.16134147346019745 2023-01-21 08:45:30.062588: step: 1952/529, loss: 0.16087284684181213 2023-01-21 08:45:31.185261: step: 1956/529, loss: 2.597561836242676 2023-01-21 08:45:32.308894: step: 1960/529, loss: 0.12044987082481384 2023-01-21 08:45:33.426164: step: 1964/529, loss: 0.15179438889026642 2023-01-21 08:45:34.536143: step: 1968/529, loss: 0.5907953381538391 2023-01-21 08:45:35.641163: step: 1972/529, loss: 0.589765727519989 2023-01-21 08:45:36.770030: step: 1976/529, loss: 0.40967217087745667 2023-01-21 08:45:37.893861: step: 1980/529, loss: 0.5136450529098511 2023-01-21 08:45:39.014269: step: 1984/529, loss: 0.23131780326366425 2023-01-21 08:45:40.150203: step: 1988/529, loss: 0.19981002807617188 2023-01-21 08:45:41.297727: step: 1992/529, loss: 0.1134800910949707 2023-01-21 08:45:42.421302: step: 1996/529, loss: 0.14816227555274963 2023-01-21 08:45:43.522501: step: 2000/529, loss: 0.17059297859668732 2023-01-21 08:45:44.618993: step: 2004/529, loss: 0.06281504780054092 2023-01-21 08:45:45.714019: step: 2008/529, loss: 0.3616578280925751 2023-01-21 08:45:46.825557: step: 2012/529, loss: 0.6039113998413086 2023-01-21 08:45:47.916759: step: 2016/529, loss: 0.433920294046402 2023-01-21 08:45:49.047787: step: 2020/529, loss: 0.17409348487854004 2023-01-21 08:45:50.190057: step: 2024/529, loss: 0.19531483948230743 2023-01-21 08:45:51.300431: step: 2028/529, loss: 0.4293663501739502 2023-01-21 08:45:52.394147: step: 2032/529, loss: 0.16214847564697266 2023-01-21 08:45:53.503657: step: 2036/529, loss: 0.21864166855812073 2023-01-21 08:45:54.601985: step: 2040/529, loss: 0.6431942582130432 2023-01-21 08:45:55.711257: step: 2044/529, loss: 0.15865576267242432 2023-01-21 08:45:56.805989: step: 2048/529, loss: 0.4180523753166199 2023-01-21 08:45:57.888986: step: 2052/529, loss: 0.220478817820549 2023-01-21 08:45:58.982717: step: 2056/529, loss: 0.21052700281143188 2023-01-21 08:46:00.073784: step: 2060/529, loss: 0.16934219002723694 2023-01-21 08:46:01.189623: step: 2064/529, loss: 0.0997040718793869 2023-01-21 08:46:02.277812: step: 2068/529, loss: 0.8931032419204712 2023-01-21 08:46:03.389599: step: 2072/529, loss: 0.029203986749053 2023-01-21 08:46:04.562741: step: 2076/529, loss: 0.30535459518432617 2023-01-21 08:46:05.671528: step: 2080/529, loss: 0.22159448266029358 2023-01-21 08:46:06.771974: step: 2084/529, loss: 0.4067089557647705 2023-01-21 08:46:07.899772: step: 2088/529, loss: 0.3600439429283142 2023-01-21 08:46:08.979836: step: 2092/529, loss: 0.06934278458356857 2023-01-21 08:46:10.082746: step: 2096/529, loss: 0.12507152557373047 2023-01-21 08:46:11.191205: step: 2100/529, loss: 0.745621919631958 2023-01-21 08:46:12.325749: step: 2104/529, loss: 0.8621286749839783 2023-01-21 08:46:13.454497: step: 2108/529, loss: 0.26914387941360474 2023-01-21 08:46:14.575149: step: 2112/529, loss: 0.10843344032764435 2023-01-21 08:46:15.703348: step: 2116/529, loss: 0.274789035320282 ================================================== Loss: 0.702 -------------------- Dev: {'event': {'p': 0.5240040858018387, 'r': 0.6830892143808256, 'f1': 0.593063583815029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Test: {'event': {'p': 0.5527544351073763, 'r': 0.6681715575620768, 'f1': 0.6050076647930506}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Chinese: {'event': {'p': 0.5254237288135594, 'r': 0.5740740740740741, 'f1': 0.5486725663716815}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Korean: {'event': {'p': 0.5714285714285714, 'r': 0.31746031746031744, 'f1': 0.4081632653061224}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Russian: {'event': {'p': 0.40625, 'r': 0.3611111111111111, 'f1': 0.38235294117647056}, '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': 0.5240040858018387, 'r': 0.6830892143808256, 'f1': 0.593063583815029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Chinese: {'event': {'p': 0.5527544351073763, 'r': 0.6681715575620768, 'f1': 0.6050076647930506}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Chinese: {'event': {'p': 0.5254237288135594, 'r': 0.5740740740740741, 'f1': 0.5486725663716815}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Korean: {'event': {'p': 0.5240040858018387, 'r': 0.6830892143808256, 'f1': 0.593063583815029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5527544351073763, 'r': 0.6681715575620768, 'f1': 0.6050076647930506}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.5714285714285714, 'r': 0.31746031746031744, 'f1': 0.4081632653061224}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.5240040858018387, 'r': 0.6830892143808256, 'f1': 0.593063583815029}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5527544351073763, 'r': 0.6681715575620768, 'f1': 0.6050076647930506}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.40625, 'r': 0.3611111111111111, 'f1': 0.38235294117647056}, '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 08:47:11.905433: step: 4/529, loss: 0.2860356569290161 2023-01-21 08:47:13.019587: step: 8/529, loss: 0.27518218755722046 2023-01-21 08:47:14.115393: step: 12/529, loss: 0.03611459955573082 2023-01-21 08:47:15.243670: step: 16/529, loss: 0.3172217607498169 2023-01-21 08:47:16.385178: step: 20/529, loss: 0.5037998557090759 2023-01-21 08:47:17.502616: step: 24/529, loss: 0.05699215084314346 2023-01-21 08:47:18.609034: step: 28/529, loss: 0.21777325868606567 2023-01-21 08:47:19.721796: step: 32/529, loss: 0.23721735179424286 2023-01-21 08:47:20.812502: step: 36/529, loss: 0.2301737368106842 2023-01-21 08:47:21.972292: step: 40/529, loss: 0.16542445123195648 2023-01-21 08:47:23.048627: step: 44/529, loss: 0.09972095489501953 2023-01-21 08:47:24.165412: step: 48/529, loss: 0.12891975045204163 2023-01-21 08:47:25.263903: step: 52/529, loss: 0.3611389994621277 2023-01-21 08:47:26.387244: step: 56/529, loss: 0.13424763083457947 2023-01-21 08:47:27.497762: step: 60/529, loss: 0.9503765106201172 2023-01-21 08:47:28.628250: step: 64/529, loss: 0.3756004869937897 2023-01-21 08:47:29.732888: step: 68/529, loss: 0.3070555329322815 2023-01-21 08:47:30.833600: step: 72/529, loss: 0.629376232624054 2023-01-21 08:47:31.955054: step: 76/529, loss: 0.3099052309989929 2023-01-21 08:47:33.095261: step: 80/529, loss: 0.41937607526779175 2023-01-21 08:47:34.190609: step: 84/529, loss: 0.15592436492443085 2023-01-21 08:47:35.337683: step: 88/529, loss: 0.3427921533584595 2023-01-21 08:47:36.466264: step: 92/529, loss: 0.11011648178100586 2023-01-21 08:47:37.563076: step: 96/529, loss: 0.1171741932630539 2023-01-21 08:47:38.671768: step: 100/529, loss: 0.9733048677444458 2023-01-21 08:47:39.791465: step: 104/529, loss: 0.17517490684986115 2023-01-21 08:47:40.922600: step: 108/529, loss: 0.07490038871765137 2023-01-21 08:47:42.061533: step: 112/529, loss: 0.13443231582641602 2023-01-21 08:47:43.171630: step: 116/529, loss: 0.2305595874786377 2023-01-21 08:47:44.279724: step: 120/529, loss: 0.3005775511264801 2023-01-21 08:47:45.398226: step: 124/529, loss: 0.23565511405467987 2023-01-21 08:47:46.515627: step: 128/529, loss: 0.5500003695487976 2023-01-21 08:47:47.656357: step: 132/529, loss: 0.22139397263526917 2023-01-21 08:47:48.754729: step: 136/529, loss: 0.19197499752044678 2023-01-21 08:47:49.882112: step: 140/529, loss: 1.0892413854599 2023-01-21 08:47:51.014626: step: 144/529, loss: 0.8368270993232727 2023-01-21 08:47:52.189188: step: 148/529, loss: 0.11113100498914719 2023-01-21 08:47:53.352907: step: 152/529, loss: 0.34302181005477905 2023-01-21 08:47:54.528038: step: 156/529, loss: 0.21152593195438385 2023-01-21 08:47:55.644422: step: 160/529, loss: 0.9145278930664062 2023-01-21 08:47:56.757588: step: 164/529, loss: 0.11634044349193573 2023-01-21 08:47:57.895150: step: 168/529, loss: 0.6937212944030762 2023-01-21 08:47:59.029896: step: 172/529, loss: 0.28506529331207275 2023-01-21 08:48:00.161030: step: 176/529, loss: 0.47687777876853943 2023-01-21 08:48:01.285300: step: 180/529, loss: 0.2957010269165039 2023-01-21 08:48:02.417996: step: 184/529, loss: 0.22779053449630737 2023-01-21 08:48:03.530299: step: 188/529, loss: 0.14600515365600586 2023-01-21 08:48:04.672002: step: 192/529, loss: 1.1644660234451294 2023-01-21 08:48:05.787923: step: 196/529, loss: 0.15589895844459534 2023-01-21 08:48:06.915666: step: 200/529, loss: 0.8292381167411804 2023-01-21 08:48:08.035703: step: 204/529, loss: 1.2156381607055664 2023-01-21 08:48:09.165037: step: 208/529, loss: 0.25848084688186646 2023-01-21 08:48:10.295834: step: 212/529, loss: 0.13020116090774536 2023-01-21 08:48:11.409706: step: 216/529, loss: 0.1744847297668457 2023-01-21 08:48:12.533364: step: 220/529, loss: 0.8398873209953308 2023-01-21 08:48:13.641604: step: 224/529, loss: 6.926356315612793 2023-01-21 08:48:14.763593: step: 228/529, loss: 7.660332202911377 2023-01-21 08:48:15.898563: step: 232/529, loss: 0.30642032623291016 2023-01-21 08:48:17.037600: step: 236/529, loss: 0.12316484749317169 2023-01-21 08:48:18.141121: step: 240/529, loss: 0.22747182846069336 2023-01-21 08:48:19.254106: step: 244/529, loss: 0.09230899810791016 2023-01-21 08:48:20.383970: step: 248/529, loss: 0.1784875988960266 2023-01-21 08:48:21.466580: step: 252/529, loss: 0.08645868301391602 2023-01-21 08:48:22.570717: step: 256/529, loss: 1.8523404598236084 2023-01-21 08:48:23.671571: step: 260/529, loss: 1.0354951620101929 2023-01-21 08:48:24.769490: step: 264/529, loss: 0.707019031047821 2023-01-21 08:48:25.874677: step: 268/529, loss: 0.37256985902786255 2023-01-21 08:48:26.965890: step: 272/529, loss: 0.1137721985578537 2023-01-21 08:48:28.084987: step: 276/529, loss: 0.37722882628440857 2023-01-21 08:48:29.199957: step: 280/529, loss: 1.10685133934021 2023-01-21 08:48:30.303558: step: 284/529, loss: 0.10990338027477264 2023-01-21 08:48:31.394437: step: 288/529, loss: 0.22471565008163452 2023-01-21 08:48:32.532293: step: 292/529, loss: 0.5236320495605469 2023-01-21 08:48:33.633552: step: 296/529, loss: 0.2033737748861313 2023-01-21 08:48:34.731481: step: 300/529, loss: 0.3361794650554657 2023-01-21 08:48:35.883122: step: 304/529, loss: 0.2288854569196701 2023-01-21 08:48:37.011009: step: 308/529, loss: 1.3480192422866821 2023-01-21 08:48:38.118686: step: 312/529, loss: 0.39426031708717346 2023-01-21 08:48:39.232441: step: 316/529, loss: 0.10438446700572968 2023-01-21 08:48:40.320938: step: 320/529, loss: 1.3860032558441162 2023-01-21 08:48:41.429521: step: 324/529, loss: 0.31454306840896606 2023-01-21 08:48:42.575404: step: 328/529, loss: 0.7631552815437317 2023-01-21 08:48:43.691351: step: 332/529, loss: 0.28372231125831604 2023-01-21 08:48:44.769826: step: 336/529, loss: 0.7222623825073242 2023-01-21 08:48:45.898841: step: 340/529, loss: 0.9577038884162903 2023-01-21 08:48:47.003575: step: 344/529, loss: 0.6774893999099731 2023-01-21 08:48:48.140640: step: 348/529, loss: 1.6684404611587524 2023-01-21 08:48:49.242495: step: 352/529, loss: 0.09131982922554016 2023-01-21 08:48:50.396667: step: 356/529, loss: 2.005427122116089 2023-01-21 08:48:51.546299: step: 360/529, loss: 0.2461986541748047 2023-01-21 08:48:52.686385: step: 364/529, loss: 0.14637012779712677 2023-01-21 08:48:53.792533: step: 368/529, loss: 0.48065081238746643 2023-01-21 08:48:54.921657: step: 372/529, loss: 0.2120942622423172 2023-01-21 08:48:56.024812: step: 376/529, loss: 0.0920385867357254 2023-01-21 08:48:57.115222: step: 380/529, loss: 0.7578844428062439 2023-01-21 08:48:58.258907: step: 384/529, loss: 1.5524718761444092 2023-01-21 08:48:59.382278: step: 388/529, loss: 0.08506367355585098 2023-01-21 08:49:00.493971: step: 392/529, loss: 0.2760906517505646 2023-01-21 08:49:01.607058: step: 396/529, loss: 0.10079348087310791 2023-01-21 08:49:02.695562: step: 400/529, loss: 0.15668383240699768 2023-01-21 08:49:03.845715: step: 404/529, loss: 0.13622145354747772 2023-01-21 08:49:04.962798: step: 408/529, loss: 0.10001173615455627 2023-01-21 08:49:06.071800: step: 412/529, loss: 0.13937222957611084 2023-01-21 08:49:07.202203: step: 416/529, loss: 0.12169227749109268 2023-01-21 08:49:08.329865: step: 420/529, loss: 0.13117718696594238 2023-01-21 08:49:09.463324: step: 424/529, loss: 0.053714849054813385 2023-01-21 08:49:10.574232: step: 428/529, loss: 0.11165256798267365 2023-01-21 08:49:11.687763: step: 432/529, loss: 0.4186984896659851 2023-01-21 08:49:12.797516: step: 436/529, loss: 0.18234197795391083 2023-01-21 08:49:13.925454: step: 440/529, loss: 0.16542989015579224 2023-01-21 08:49:15.037570: step: 444/529, loss: 0.20645684003829956 2023-01-21 08:49:16.162144: step: 448/529, loss: 0.8766270875930786 2023-01-21 08:49:17.278625: step: 452/529, loss: 0.6662218570709229 2023-01-21 08:49:18.388640: step: 456/529, loss: 1.0815006494522095 2023-01-21 08:49:19.507280: step: 460/529, loss: 0.16531048715114594 2023-01-21 08:49:20.646282: step: 464/529, loss: 0.8627262115478516 2023-01-21 08:49:21.771249: step: 468/529, loss: 0.31351956725120544 2023-01-21 08:49:22.875063: step: 472/529, loss: 0.21563774347305298 2023-01-21 08:49:24.031115: step: 476/529, loss: 0.2534770369529724 2023-01-21 08:49:25.129996: step: 480/529, loss: 0.3346182703971863 2023-01-21 08:49:26.237301: step: 484/529, loss: 0.1588408499956131 2023-01-21 08:49:27.327884: step: 488/529, loss: 0.8445334434509277 2023-01-21 08:49:28.416346: step: 492/529, loss: 0.09498071670532227 2023-01-21 08:49:29.515700: step: 496/529, loss: 0.4935251474380493 2023-01-21 08:49:30.630506: step: 500/529, loss: 0.12832719087600708 2023-01-21 08:49:31.752370: step: 504/529, loss: 0.7315385341644287 2023-01-21 08:49:32.874924: step: 508/529, loss: 0.5774371027946472 2023-01-21 08:49:33.997022: step: 512/529, loss: 0.7118686437606812 2023-01-21 08:49:35.143066: step: 516/529, loss: 0.03820633888244629 2023-01-21 08:49:36.259603: step: 520/529, loss: 0.34821614623069763 2023-01-21 08:49:37.384136: step: 524/529, loss: 0.16866126656532288 2023-01-21 08:49:38.502779: step: 528/529, loss: 0.19749489426612854 2023-01-21 08:49:39.617013: step: 532/529, loss: 0.4907470643520355 2023-01-21 08:49:40.731107: step: 536/529, loss: 1.2128127813339233 2023-01-21 08:49:41.841667: step: 540/529, loss: 0.7097907662391663 2023-01-21 08:49:42.971903: step: 544/529, loss: 0.08406487107276917 2023-01-21 08:49:44.088718: step: 548/529, loss: 0.27390700578689575 2023-01-21 08:49:45.192236: step: 552/529, loss: 0.0980248898267746 2023-01-21 08:49:46.351914: step: 556/529, loss: 0.2770588994026184 2023-01-21 08:49:47.518874: step: 560/529, loss: 0.676925003528595 2023-01-21 08:49:48.640968: step: 564/529, loss: 0.2937759757041931 2023-01-21 08:49:49.734187: step: 568/529, loss: 0.7336345314979553 2023-01-21 08:49:50.895326: step: 572/529, loss: 0.0925624817609787 2023-01-21 08:49:52.016842: step: 576/529, loss: 1.334926724433899 2023-01-21 08:49:53.134896: step: 580/529, loss: 0.03041706047952175 2023-01-21 08:49:54.229119: step: 584/529, loss: 6.906869888305664 2023-01-21 08:49:55.344800: step: 588/529, loss: 0.19706524908542633 2023-01-21 08:49:56.464122: step: 592/529, loss: 0.8816401958465576 2023-01-21 08:49:57.564699: step: 596/529, loss: 0.036309242248535156 2023-01-21 08:49:58.675670: step: 600/529, loss: 0.278812974691391 2023-01-21 08:49:59.791496: step: 604/529, loss: 0.8872995376586914 2023-01-21 08:50:00.912725: step: 608/529, loss: 1.1183286905288696 2023-01-21 08:50:02.012782: step: 612/529, loss: 0.789084255695343 2023-01-21 08:50:03.119405: step: 616/529, loss: 0.09627886116504669 2023-01-21 08:50:04.221417: step: 620/529, loss: 0.16537365317344666 2023-01-21 08:50:05.321507: step: 624/529, loss: 0.37847715616226196 2023-01-21 08:50:06.417560: step: 628/529, loss: 0.37421226501464844 2023-01-21 08:50:07.540041: step: 632/529, loss: 0.31230393052101135 2023-01-21 08:50:08.645588: step: 636/529, loss: 2.34479022026062 2023-01-21 08:50:09.752762: step: 640/529, loss: 6.490022659301758 2023-01-21 08:50:10.875669: step: 644/529, loss: 2.0917792320251465 2023-01-21 08:50:11.964355: step: 648/529, loss: 0.05058545991778374 2023-01-21 08:50:13.079958: step: 652/529, loss: 1.8338184356689453 2023-01-21 08:50:14.222506: step: 656/529, loss: 0.16708441078662872 2023-01-21 08:50:15.368488: step: 660/529, loss: 0.4179391860961914 2023-01-21 08:50:16.492074: step: 664/529, loss: 0.1016329824924469 2023-01-21 08:50:17.629330: step: 668/529, loss: 1.067753553390503 2023-01-21 08:50:18.769536: step: 672/529, loss: 0.14349260926246643 2023-01-21 08:50:19.877297: step: 676/529, loss: 0.0998539924621582 2023-01-21 08:50:20.982269: step: 680/529, loss: 0.18899011611938477 2023-01-21 08:50:22.116854: step: 684/529, loss: 0.1560390591621399 2023-01-21 08:50:23.234297: step: 688/529, loss: 0.7194616794586182 2023-01-21 08:50:24.333896: step: 692/529, loss: 0.7706704139709473 2023-01-21 08:50:25.442726: step: 696/529, loss: 0.19845056533813477 2023-01-21 08:50:26.570074: step: 700/529, loss: 0.8047633171081543 2023-01-21 08:50:27.684198: step: 704/529, loss: 0.6636126041412354 2023-01-21 08:50:28.772413: step: 708/529, loss: 1.625649094581604 2023-01-21 08:50:29.877267: step: 712/529, loss: 0.15570569038391113 2023-01-21 08:50:30.978776: step: 716/529, loss: 0.24859140813350677 2023-01-21 08:50:32.094903: step: 720/529, loss: 0.20342475175857544 2023-01-21 08:50:33.196146: step: 724/529, loss: 0.1019449234008789 2023-01-21 08:50:34.316175: step: 728/529, loss: 0.2307766079902649 2023-01-21 08:50:35.430737: step: 732/529, loss: 0.2157575637102127 2023-01-21 08:50:36.545515: step: 736/529, loss: 0.15788784623146057 2023-01-21 08:50:37.659362: step: 740/529, loss: 6.91624641418457 2023-01-21 08:50:38.762063: step: 744/529, loss: 0.11491823196411133 2023-01-21 08:50:39.897269: step: 748/529, loss: 0.14090195298194885 2023-01-21 08:50:40.997653: step: 752/529, loss: 0.5507321357727051 2023-01-21 08:50:42.128961: step: 756/529, loss: 1.4055075645446777 2023-01-21 08:50:43.225909: step: 760/529, loss: 0.9117105603218079 2023-01-21 08:50:44.306559: step: 764/529, loss: 0.4151371121406555 2023-01-21 08:50:45.429184: step: 768/529, loss: 0.15255776047706604 2023-01-21 08:50:46.538626: step: 772/529, loss: 0.18844282627105713 2023-01-21 08:50:47.662421: step: 776/529, loss: 2.801220417022705 2023-01-21 08:50:48.754505: step: 780/529, loss: 0.26680856943130493 2023-01-21 08:50:49.851650: step: 784/529, loss: 0.9256723523139954 2023-01-21 08:50:50.972365: step: 788/529, loss: 0.830964207649231 2023-01-21 08:50:52.081244: step: 792/529, loss: 0.06066305935382843 2023-01-21 08:50:53.201970: step: 796/529, loss: 0.09732022881507874 2023-01-21 08:50:54.305896: step: 800/529, loss: 0.1584729254245758 2023-01-21 08:50:55.427453: step: 804/529, loss: 0.7567061185836792 2023-01-21 08:50:56.560609: step: 808/529, loss: 0.06499285995960236 2023-01-21 08:50:57.669247: step: 812/529, loss: 0.11879683285951614 2023-01-21 08:50:58.767182: step: 816/529, loss: 0.09615898132324219 2023-01-21 08:50:59.879629: step: 820/529, loss: 0.7568000555038452 2023-01-21 08:51:00.964968: step: 824/529, loss: 0.49963921308517456 2023-01-21 08:51:02.060012: step: 828/529, loss: 0.40783947706222534 2023-01-21 08:51:03.187281: step: 832/529, loss: 0.9488718509674072 2023-01-21 08:51:04.294947: step: 836/529, loss: 0.2055523842573166 2023-01-21 08:51:05.391135: step: 840/529, loss: 0.732813835144043 2023-01-21 08:51:06.556840: step: 844/529, loss: 1.3017147779464722 2023-01-21 08:51:07.668342: step: 848/529, loss: 0.2309926152229309 2023-01-21 08:51:08.756414: step: 852/529, loss: 0.0938730239868164 2023-01-21 08:51:09.855491: step: 856/529, loss: 0.19092193245887756 2023-01-21 08:51:10.958229: step: 860/529, loss: 0.09991435706615448 2023-01-21 08:51:12.072907: step: 864/529, loss: 0.742562472820282 2023-01-21 08:51:13.206696: step: 868/529, loss: 0.07854809612035751 2023-01-21 08:51:14.330524: step: 872/529, loss: 0.7029688954353333 2023-01-21 08:51:15.441389: step: 876/529, loss: 1.4036539793014526 2023-01-21 08:51:16.564109: step: 880/529, loss: 0.350328266620636 2023-01-21 08:51:17.688536: step: 884/529, loss: 0.04996442794799805 2023-01-21 08:51:18.823171: step: 888/529, loss: 8.201461791992188 2023-01-21 08:51:19.960220: step: 892/529, loss: 0.0975864976644516 2023-01-21 08:51:21.079988: step: 896/529, loss: 0.12895183265209198 2023-01-21 08:51:22.165209: step: 900/529, loss: 0.34391850233078003 2023-01-21 08:51:23.283700: step: 904/529, loss: 0.10912913829088211 2023-01-21 08:51:24.364081: step: 908/529, loss: 0.23584072291851044 2023-01-21 08:51:25.481773: step: 912/529, loss: 1.426085114479065 2023-01-21 08:51:26.597555: step: 916/529, loss: 0.5282402038574219 2023-01-21 08:51:27.708025: step: 920/529, loss: 0.17803211510181427 2023-01-21 08:51:28.814586: step: 924/529, loss: 1.6859261989593506 2023-01-21 08:51:29.927845: step: 928/529, loss: 0.727862536907196 2023-01-21 08:51:31.081581: step: 932/529, loss: 0.09496030956506729 2023-01-21 08:51:32.194190: step: 936/529, loss: 0.9600954055786133 2023-01-21 08:51:33.306916: step: 940/529, loss: 0.15700650215148926 2023-01-21 08:51:34.419726: step: 944/529, loss: 0.2523081600666046 2023-01-21 08:51:35.535878: step: 948/529, loss: 0.06626439094543457 2023-01-21 08:51:36.644761: step: 952/529, loss: 0.5524584054946899 2023-01-21 08:51:37.764954: step: 956/529, loss: 0.16383175551891327 2023-01-21 08:51:38.878158: step: 960/529, loss: 0.3369196653366089 2023-01-21 08:51:40.017000: step: 964/529, loss: 0.05577874183654785 2023-01-21 08:51:41.120632: step: 968/529, loss: 0.21209755539894104 2023-01-21 08:51:42.206775: step: 972/529, loss: 0.21974115073680878 2023-01-21 08:51:43.338999: step: 976/529, loss: 0.6180623173713684 2023-01-21 08:51:44.465466: step: 980/529, loss: 0.07699732482433319 2023-01-21 08:51:45.575624: step: 984/529, loss: 0.13494902849197388 2023-01-21 08:51:46.723678: step: 988/529, loss: 0.08483333885669708 2023-01-21 08:51:47.815800: step: 992/529, loss: 0.16446347534656525 2023-01-21 08:51:48.931999: step: 996/529, loss: 0.4612805247306824 2023-01-21 08:51:50.059562: step: 1000/529, loss: 0.6165770292282104 2023-01-21 08:51:51.177034: step: 1004/529, loss: 0.9742149710655212 2023-01-21 08:51:52.281841: step: 1008/529, loss: 0.20880833268165588 2023-01-21 08:51:53.393756: step: 1012/529, loss: 0.4361006021499634 2023-01-21 08:51:54.515234: step: 1016/529, loss: 0.3200344741344452 2023-01-21 08:51:55.614053: step: 1020/529, loss: 0.6977413892745972 2023-01-21 08:51:56.719672: step: 1024/529, loss: 0.5455732941627502 2023-01-21 08:51:57.857744: step: 1028/529, loss: 0.9375991821289062 2023-01-21 08:51:58.955328: step: 1032/529, loss: 0.16531938314437866 2023-01-21 08:52:00.066714: step: 1036/529, loss: 0.5652868747711182 2023-01-21 08:52:01.150721: step: 1040/529, loss: 0.8211908936500549 2023-01-21 08:52:02.252950: step: 1044/529, loss: 0.756533682346344 2023-01-21 08:52:03.370005: step: 1048/529, loss: 0.8813340067863464 2023-01-21 08:52:04.512856: step: 1052/529, loss: 0.7167045474052429 2023-01-21 08:52:05.628377: step: 1056/529, loss: 0.20362785458564758 2023-01-21 08:52:06.749933: step: 1060/529, loss: 0.1289883702993393 2023-01-21 08:52:07.851266: step: 1064/529, loss: 1.2439157962799072 2023-01-21 08:52:08.957344: step: 1068/529, loss: 1.7101339101791382 2023-01-21 08:52:10.064253: step: 1072/529, loss: 0.06328277289867401 2023-01-21 08:52:11.164410: step: 1076/529, loss: 0.4203033745288849 2023-01-21 08:52:12.288476: step: 1080/529, loss: 0.6678031086921692 2023-01-21 08:52:13.426610: step: 1084/529, loss: 0.9080505967140198 2023-01-21 08:52:14.525798: step: 1088/529, loss: 0.08743324130773544 2023-01-21 08:52:15.653828: step: 1092/529, loss: 0.2029522955417633 2023-01-21 08:52:16.828181: step: 1096/529, loss: 1.1038970947265625 2023-01-21 08:52:17.917104: step: 1100/529, loss: 0.45893487334251404 2023-01-21 08:52:19.037964: step: 1104/529, loss: 0.3377528190612793 2023-01-21 08:52:20.184266: step: 1108/529, loss: 1.651160717010498 2023-01-21 08:52:21.257171: step: 1112/529, loss: 0.13979625701904297 2023-01-21 08:52:22.372213: step: 1116/529, loss: 0.23218198120594025 2023-01-21 08:52:23.476946: step: 1120/529, loss: 1.791353464126587 2023-01-21 08:52:24.600949: step: 1124/529, loss: 0.17537803947925568 2023-01-21 08:52:25.716560: step: 1128/529, loss: 0.5524899959564209 2023-01-21 08:52:26.832864: step: 1132/529, loss: 0.12660765647888184 2023-01-21 08:52:27.955958: step: 1136/529, loss: 0.16297034919261932 2023-01-21 08:52:29.043299: step: 1140/529, loss: 0.8825723528862 2023-01-21 08:52:30.171921: step: 1144/529, loss: 0.10069847106933594 2023-01-21 08:52:31.305542: step: 1148/529, loss: 0.7003568410873413 2023-01-21 08:52:32.454272: step: 1152/529, loss: 0.18493251502513885 2023-01-21 08:52:33.576820: step: 1156/529, loss: 0.8470341563224792 2023-01-21 08:52:34.673368: step: 1160/529, loss: 0.27277079224586487 2023-01-21 08:52:35.799734: step: 1164/529, loss: 0.059940434992313385 2023-01-21 08:52:36.909588: step: 1168/529, loss: 0.2493133544921875 2023-01-21 08:52:38.007755: step: 1172/529, loss: 0.3488086760044098 2023-01-21 08:52:39.132311: step: 1176/529, loss: 0.15316271781921387 2023-01-21 08:52:40.263970: step: 1180/529, loss: 0.5007015466690063 2023-01-21 08:52:41.374683: step: 1184/529, loss: 0.2625235617160797 2023-01-21 08:52:42.489028: step: 1188/529, loss: 1.4159905910491943 2023-01-21 08:52:43.624863: step: 1192/529, loss: 0.06470618396997452 2023-01-21 08:52:44.729504: step: 1196/529, loss: 0.9081791043281555 2023-01-21 08:52:45.844414: step: 1200/529, loss: 0.7985243797302246 2023-01-21 08:52:46.972035: step: 1204/529, loss: 0.5703268051147461 2023-01-21 08:52:48.064808: step: 1208/529, loss: 0.06196432188153267 2023-01-21 08:52:49.154526: step: 1212/529, loss: 0.996475338935852 2023-01-21 08:52:50.270873: step: 1216/529, loss: 0.07075901329517365 2023-01-21 08:52:51.358993: step: 1220/529, loss: 0.2182464599609375 2023-01-21 08:52:52.482513: step: 1224/529, loss: 0.11064356565475464 2023-01-21 08:52:53.581424: step: 1228/529, loss: 0.1474200338125229 2023-01-21 08:52:54.688362: step: 1232/529, loss: 0.5438947677612305 2023-01-21 08:52:55.844286: step: 1236/529, loss: 0.08816280961036682 2023-01-21 08:52:56.954729: step: 1240/529, loss: 0.3141501545906067 2023-01-21 08:52:58.100163: step: 1244/529, loss: 1.2704408168792725 2023-01-21 08:52:59.228933: step: 1248/529, loss: 0.7938334941864014 2023-01-21 08:53:00.366316: step: 1252/529, loss: 0.13236045837402344 2023-01-21 08:53:01.522323: step: 1256/529, loss: 0.41035032272338867 2023-01-21 08:53:02.623467: step: 1260/529, loss: 0.30850380659103394 2023-01-21 08:53:03.741545: step: 1264/529, loss: 0.22101956605911255 2023-01-21 08:53:04.882440: step: 1268/529, loss: 0.7396289706230164 2023-01-21 08:53:05.956660: step: 1272/529, loss: 0.8715603947639465 2023-01-21 08:53:07.048491: step: 1276/529, loss: 0.16690178215503693 2023-01-21 08:53:08.132910: step: 1280/529, loss: 0.2530982196331024 2023-01-21 08:53:09.309014: step: 1284/529, loss: 0.5470666885375977 2023-01-21 08:53:10.418524: step: 1288/529, loss: 0.17506083846092224 2023-01-21 08:53:11.525283: step: 1292/529, loss: 1.537326693534851 2023-01-21 08:53:12.670126: step: 1296/529, loss: 0.27675867080688477 2023-01-21 08:53:13.785104: step: 1300/529, loss: 0.43148624897003174 2023-01-21 08:53:14.921786: step: 1304/529, loss: 0.794842541217804 2023-01-21 08:53:16.061867: step: 1308/529, loss: 0.6822460889816284 2023-01-21 08:53:17.163258: step: 1312/529, loss: 0.3071845769882202 2023-01-21 08:53:18.284718: step: 1316/529, loss: 0.1444409340620041 2023-01-21 08:53:19.406897: step: 1320/529, loss: 0.6636388897895813 2023-01-21 08:53:20.513194: step: 1324/529, loss: 0.44452428817749023 2023-01-21 08:53:21.621804: step: 1328/529, loss: 0.11891746520996094 2023-01-21 08:53:22.740491: step: 1332/529, loss: 0.4584599733352661 2023-01-21 08:53:23.838112: step: 1336/529, loss: 0.3394605815410614 2023-01-21 08:53:25.000632: step: 1340/529, loss: 0.5514322519302368 2023-01-21 08:53:26.128021: step: 1344/529, loss: 0.4926857650279999 2023-01-21 08:53:27.266493: step: 1348/529, loss: 0.8418720364570618 2023-01-21 08:53:28.403145: step: 1352/529, loss: 0.16090410947799683 2023-01-21 08:53:29.500809: step: 1356/529, loss: 0.09347229450941086 2023-01-21 08:53:30.574667: step: 1360/529, loss: 0.7828988432884216 2023-01-21 08:53:31.714352: step: 1364/529, loss: 0.39452803134918213 2023-01-21 08:53:32.823348: step: 1368/529, loss: 0.15116086602210999 2023-01-21 08:53:33.927991: step: 1372/529, loss: 0.31756696105003357 2023-01-21 08:53:35.041503: step: 1376/529, loss: 0.24744626879692078 2023-01-21 08:53:36.185729: step: 1380/529, loss: 0.1375924050807953 2023-01-21 08:53:37.317325: step: 1384/529, loss: 0.18175840377807617 2023-01-21 08:53:38.410891: step: 1388/529, loss: 0.3222396969795227 2023-01-21 08:53:39.504891: step: 1392/529, loss: 0.10834214836359024 2023-01-21 08:53:40.586551: step: 1396/529, loss: 0.13487406075000763 2023-01-21 08:53:41.696780: step: 1400/529, loss: 0.6108031868934631 2023-01-21 08:53:42.816683: step: 1404/529, loss: 4.350836277008057 2023-01-21 08:53:43.954110: step: 1408/529, loss: 0.1640222668647766 2023-01-21 08:53:45.047695: step: 1412/529, loss: 0.1067267432808876 2023-01-21 08:53:46.142830: step: 1416/529, loss: 0.1348743438720703 2023-01-21 08:53:47.240138: step: 1420/529, loss: 0.4730883538722992 2023-01-21 08:53:48.352867: step: 1424/529, loss: 0.24327431619167328 2023-01-21 08:53:49.523252: step: 1428/529, loss: 0.22100697457790375 2023-01-21 08:53:50.647967: step: 1432/529, loss: 0.6290817856788635 2023-01-21 08:53:51.748580: step: 1436/529, loss: 0.30470961332321167 2023-01-21 08:53:52.858156: step: 1440/529, loss: 0.12193021178245544 2023-01-21 08:53:53.980134: step: 1444/529, loss: 0.1623374968767166 2023-01-21 08:53:55.099155: step: 1448/529, loss: 0.14038699865341187 2023-01-21 08:53:56.203427: step: 1452/529, loss: 0.9382328987121582 2023-01-21 08:53:57.322514: step: 1456/529, loss: 0.43177807331085205 2023-01-21 08:53:58.450265: step: 1460/529, loss: 0.4245401620864868 2023-01-21 08:53:59.599816: step: 1464/529, loss: 0.8357953429222107 2023-01-21 08:54:00.715007: step: 1468/529, loss: 0.7871928811073303 2023-01-21 08:54:01.829607: step: 1472/529, loss: 0.9252025485038757 2023-01-21 08:54:02.950741: step: 1476/529, loss: 0.15233880281448364 2023-01-21 08:54:04.047387: step: 1480/529, loss: 0.05792422592639923 2023-01-21 08:54:05.153332: step: 1484/529, loss: 1.088201880455017 2023-01-21 08:54:06.263703: step: 1488/529, loss: 0.106000617146492 2023-01-21 08:54:07.387487: step: 1492/529, loss: 0.06665768474340439 2023-01-21 08:54:08.484552: step: 1496/529, loss: 0.05442757532000542 2023-01-21 08:54:09.606120: step: 1500/529, loss: 0.23096761107444763 2023-01-21 08:54:10.739814: step: 1504/529, loss: 0.9938667416572571 2023-01-21 08:54:11.858472: step: 1508/529, loss: 0.5638448596000671 2023-01-21 08:54:12.989224: step: 1512/529, loss: 0.1069970577955246 2023-01-21 08:54:14.131262: step: 1516/529, loss: 0.16256123781204224 2023-01-21 08:54:15.249012: step: 1520/529, loss: 0.14033308625221252 2023-01-21 08:54:16.357611: step: 1524/529, loss: 0.7638886570930481 2023-01-21 08:54:17.460227: step: 1528/529, loss: 0.32965677976608276 2023-01-21 08:54:18.558140: step: 1532/529, loss: 0.2397201508283615 2023-01-21 08:54:19.697863: step: 1536/529, loss: 0.7115920782089233 2023-01-21 08:54:20.828422: step: 1540/529, loss: 0.20499220490455627 2023-01-21 08:54:21.914368: step: 1544/529, loss: 0.3668995797634125 2023-01-21 08:54:23.043890: step: 1548/529, loss: 0.1908825933933258 2023-01-21 08:54:24.138865: step: 1552/529, loss: 0.1789708137512207 2023-01-21 08:54:25.271001: step: 1556/529, loss: 0.11772161722183228 2023-01-21 08:54:26.381449: step: 1560/529, loss: 0.1586398184299469 2023-01-21 08:54:27.469985: step: 1564/529, loss: 0.256094366312027 2023-01-21 08:54:28.566795: step: 1568/529, loss: 0.48130911588668823 2023-01-21 08:54:29.680021: step: 1572/529, loss: 0.16873493790626526 2023-01-21 08:54:30.801437: step: 1576/529, loss: 0.0388309508562088 2023-01-21 08:54:31.920183: step: 1580/529, loss: 0.2380487024784088 2023-01-21 08:54:33.037533: step: 1584/529, loss: 0.2640243172645569 2023-01-21 08:54:34.155042: step: 1588/529, loss: 0.12211154401302338 2023-01-21 08:54:35.271733: step: 1592/529, loss: 0.10584276169538498 2023-01-21 08:54:36.396765: step: 1596/529, loss: 0.144094318151474 2023-01-21 08:54:37.523338: step: 1600/529, loss: 0.14474931359291077 2023-01-21 08:54:38.631705: step: 1604/529, loss: 1.1855412721633911 2023-01-21 08:54:39.758624: step: 1608/529, loss: 0.2967948913574219 2023-01-21 08:54:40.876995: step: 1612/529, loss: 0.20573052763938904 2023-01-21 08:54:41.995269: step: 1616/529, loss: 0.22052517533302307 2023-01-21 08:54:43.110036: step: 1620/529, loss: 0.31615716218948364 2023-01-21 08:54:44.215070: step: 1624/529, loss: 1.3725531101226807 2023-01-21 08:54:45.339275: step: 1628/529, loss: 0.44345206022262573 2023-01-21 08:54:46.454252: step: 1632/529, loss: 0.043688204139471054 2023-01-21 08:54:47.587926: step: 1636/529, loss: 0.4386120140552521 2023-01-21 08:54:48.728839: step: 1640/529, loss: 0.11705875396728516 2023-01-21 08:54:49.845301: step: 1644/529, loss: 6.528853893280029 2023-01-21 08:54:50.965733: step: 1648/529, loss: 0.12868361175060272 2023-01-21 08:54:52.078641: step: 1652/529, loss: 0.13987331092357635 2023-01-21 08:54:53.215984: step: 1656/529, loss: 1.27565336227417 2023-01-21 08:54:54.314781: step: 1660/529, loss: 0.28140562772750854 2023-01-21 08:54:55.445021: step: 1664/529, loss: 0.07724638283252716 2023-01-21 08:54:56.560449: step: 1668/529, loss: 0.6375335454940796 2023-01-21 08:54:57.667110: step: 1672/529, loss: 0.10636444389820099 2023-01-21 08:54:58.791108: step: 1676/529, loss: 0.16035687923431396 2023-01-21 08:54:59.910384: step: 1680/529, loss: 0.2522631585597992 2023-01-21 08:55:01.019956: step: 1684/529, loss: 1.538737416267395 2023-01-21 08:55:02.131267: step: 1688/529, loss: 1.489425778388977 2023-01-21 08:55:03.220630: step: 1692/529, loss: 0.19276371598243713 2023-01-21 08:55:04.335992: step: 1696/529, loss: 0.21864581108093262 2023-01-21 08:55:05.469802: step: 1700/529, loss: 0.2302083969116211 2023-01-21 08:55:06.562562: step: 1704/529, loss: 0.08679866790771484 2023-01-21 08:55:07.634823: step: 1708/529, loss: 0.12485584616661072 2023-01-21 08:55:08.754747: step: 1712/529, loss: 6.14894962310791 2023-01-21 08:55:09.878411: step: 1716/529, loss: 0.794439435005188 2023-01-21 08:55:11.036346: step: 1720/529, loss: 0.7427870035171509 2023-01-21 08:55:12.142683: step: 1724/529, loss: 0.4588386118412018 2023-01-21 08:55:13.269230: step: 1728/529, loss: 0.2772024869918823 2023-01-21 08:55:14.373110: step: 1732/529, loss: 2.1954152584075928 2023-01-21 08:55:15.494024: step: 1736/529, loss: 0.9224813580513 2023-01-21 08:55:16.597957: step: 1740/529, loss: 0.222464457154274 2023-01-21 08:55:17.718534: step: 1744/529, loss: 0.2673835754394531 2023-01-21 08:55:18.862863: step: 1748/529, loss: 0.7512953281402588 2023-01-21 08:55:19.988148: step: 1752/529, loss: 0.36012017726898193 2023-01-21 08:55:21.102417: step: 1756/529, loss: 0.13930052518844604 2023-01-21 08:55:22.249832: step: 1760/529, loss: 0.8526419401168823 2023-01-21 08:55:23.363629: step: 1764/529, loss: 0.05703430250287056 2023-01-21 08:55:24.469408: step: 1768/529, loss: 0.21558809280395508 2023-01-21 08:55:25.605405: step: 1772/529, loss: 0.2808411717414856 2023-01-21 08:55:26.717372: step: 1776/529, loss: 0.3419952392578125 2023-01-21 08:55:27.823680: step: 1780/529, loss: 0.12058372795581818 2023-01-21 08:55:28.914299: step: 1784/529, loss: 0.058970026671886444 2023-01-21 08:55:30.029755: step: 1788/529, loss: 1.2495381832122803 2023-01-21 08:55:31.154468: step: 1792/529, loss: 2.34295654296875 2023-01-21 08:55:32.251431: step: 1796/529, loss: 0.255861759185791 2023-01-21 08:55:33.349712: step: 1800/529, loss: 0.022750426083803177 2023-01-21 08:55:34.475029: step: 1804/529, loss: 0.1370266079902649 2023-01-21 08:55:35.602963: step: 1808/529, loss: 0.1721268892288208 2023-01-21 08:55:36.696636: step: 1812/529, loss: 1.962512493133545 2023-01-21 08:55:37.803974: step: 1816/529, loss: 0.9100700616836548 2023-01-21 08:55:38.914632: step: 1820/529, loss: 0.5272812843322754 2023-01-21 08:55:40.066729: step: 1824/529, loss: 0.24549484252929688 2023-01-21 08:55:41.151607: step: 1828/529, loss: 0.2327841818332672 2023-01-21 08:55:42.275474: step: 1832/529, loss: 0.5335144400596619 2023-01-21 08:55:43.366727: step: 1836/529, loss: 1.3104196786880493 2023-01-21 08:55:44.490677: step: 1840/529, loss: 0.9715365767478943 2023-01-21 08:55:45.609999: step: 1844/529, loss: 0.5726670622825623 2023-01-21 08:55:46.736533: step: 1848/529, loss: 0.14137744903564453 2023-01-21 08:55:47.870062: step: 1852/529, loss: 6.426032066345215 2023-01-21 08:55:49.004177: step: 1856/529, loss: 0.2132217437028885 2023-01-21 08:55:50.112563: step: 1860/529, loss: 0.4421789348125458 2023-01-21 08:55:51.239571: step: 1864/529, loss: 0.21401147544384003 2023-01-21 08:55:52.376411: step: 1868/529, loss: 0.3631221652030945 2023-01-21 08:55:53.514749: step: 1872/529, loss: 1.142690658569336 2023-01-21 08:55:54.707217: step: 1876/529, loss: 0.1468295156955719 2023-01-21 08:55:55.847505: step: 1880/529, loss: 0.1995573192834854 2023-01-21 08:55:56.956700: step: 1884/529, loss: 0.13073283433914185 2023-01-21 08:55:58.066277: step: 1888/529, loss: 0.230586439371109 2023-01-21 08:55:59.168253: step: 1892/529, loss: 0.9619349241256714 2023-01-21 08:56:00.252829: step: 1896/529, loss: 0.11495314538478851 2023-01-21 08:56:01.372630: step: 1900/529, loss: 0.18295860290527344 2023-01-21 08:56:02.540846: step: 1904/529, loss: 0.25179794430732727 2023-01-21 08:56:03.648351: step: 1908/529, loss: 0.3422781229019165 2023-01-21 08:56:04.775125: step: 1912/529, loss: 0.09139509499073029 2023-01-21 08:56:05.849522: step: 1916/529, loss: 0.44558393955230713 2023-01-21 08:56:06.949326: step: 1920/529, loss: 0.7917593717575073 2023-01-21 08:56:08.075602: step: 1924/529, loss: 0.18368090689182281 2023-01-21 08:56:09.203648: step: 1928/529, loss: 0.717082142829895 2023-01-21 08:56:10.349561: step: 1932/529, loss: 0.8882566690444946 2023-01-21 08:56:11.468021: step: 1936/529, loss: 0.1455463469028473 2023-01-21 08:56:12.569927: step: 1940/529, loss: 0.039681900292634964 2023-01-21 08:56:13.686231: step: 1944/529, loss: 0.2294284701347351 2023-01-21 08:56:14.827519: step: 1948/529, loss: 0.13052520155906677 2023-01-21 08:56:15.947196: step: 1952/529, loss: 0.3179095387458801 2023-01-21 08:56:17.052200: step: 1956/529, loss: 0.10047879070043564 2023-01-21 08:56:18.198059: step: 1960/529, loss: 0.18036779761314392 2023-01-21 08:56:19.304880: step: 1964/529, loss: 0.07747345417737961 2023-01-21 08:56:20.431769: step: 1968/529, loss: 0.7979085445404053 2023-01-21 08:56:21.555557: step: 1972/529, loss: 0.2452598512172699 2023-01-21 08:56:22.645702: step: 1976/529, loss: 0.7543932199478149 2023-01-21 08:56:23.744360: step: 1980/529, loss: 0.12381067126989365 2023-01-21 08:56:24.846878: step: 1984/529, loss: 0.16027717292308807 2023-01-21 08:56:25.987520: step: 1988/529, loss: 0.13522708415985107 2023-01-21 08:56:27.078278: step: 1992/529, loss: 0.8572558164596558 2023-01-21 08:56:28.187569: step: 1996/529, loss: 0.6600879430770874 2023-01-21 08:56:29.271501: step: 2000/529, loss: 0.12676039338111877 2023-01-21 08:56:30.385932: step: 2004/529, loss: 0.84746915102005 2023-01-21 08:56:31.531028: step: 2008/529, loss: 0.07441587746143341 2023-01-21 08:56:32.637754: step: 2012/529, loss: 0.7854389548301697 2023-01-21 08:56:33.731557: step: 2016/529, loss: 1.5562529563903809 2023-01-21 08:56:34.826249: step: 2020/529, loss: 0.4920474588871002 2023-01-21 08:56:35.929049: step: 2024/529, loss: 0.40680745244026184 2023-01-21 08:56:37.035591: step: 2028/529, loss: 0.12970677018165588 2023-01-21 08:56:38.157916: step: 2032/529, loss: 0.9063763618469238 2023-01-21 08:56:39.261441: step: 2036/529, loss: 0.5326225757598877 2023-01-21 08:56:40.353585: step: 2040/529, loss: 0.12516899406909943 2023-01-21 08:56:41.466537: step: 2044/529, loss: 0.0875358134508133 2023-01-21 08:56:42.588463: step: 2048/529, loss: 1.0739879608154297 2023-01-21 08:56:43.690470: step: 2052/529, loss: 1.4095219373703003 2023-01-21 08:56:44.800700: step: 2056/529, loss: 0.22320863604545593 2023-01-21 08:56:45.914856: step: 2060/529, loss: 0.4662877023220062 2023-01-21 08:56:47.025551: step: 2064/529, loss: 0.4730173349380493 2023-01-21 08:56:48.126127: step: 2068/529, loss: 0.12959042191505432 2023-01-21 08:56:49.216669: step: 2072/529, loss: 1.5932559967041016 2023-01-21 08:56:50.348570: step: 2076/529, loss: 0.2425878494977951 2023-01-21 08:56:51.438496: step: 2080/529, loss: 0.13698916137218475 2023-01-21 08:56:52.556809: step: 2084/529, loss: 0.7115722894668579 2023-01-21 08:56:53.675966: step: 2088/529, loss: 0.14050406217575073 2023-01-21 08:56:54.840468: step: 2092/529, loss: 0.10151372104883194 2023-01-21 08:56:55.931590: step: 2096/529, loss: 0.7830831408500671 2023-01-21 08:56:57.059234: step: 2100/529, loss: 1.3276368379592896 2023-01-21 08:56:58.175966: step: 2104/529, loss: 0.1844952553510666 2023-01-21 08:56:59.304400: step: 2108/529, loss: 0.25663870573043823 2023-01-21 08:57:00.437100: step: 2112/529, loss: 0.23623895645141602 2023-01-21 08:57:01.564085: step: 2116/529, loss: 0.8579446077346802 ================================================== Loss: 0.574 -------------------- Dev: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Test: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Korean: {'event': {'p': 0.7575757575757576, 'r': 0.3968253968253968, 'f1': 0.5208333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Russian: {'event': {'p': 0.6176470588235294, 'r': 0.5833333333333334, 'f1': 0.6}, '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': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.7575757575757576, 'r': 0.3968253968253968, 'f1': 0.5208333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Russian: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Russian: {'event': {'p': 0.6176470588235294, 'r': 0.5833333333333334, 'f1': 0.6}, '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 08:58:04.313899: step: 4/529, loss: 0.11128297448158264 2023-01-21 08:58:05.409467: step: 8/529, loss: 0.13652391731739044 2023-01-21 08:58:06.510227: step: 12/529, loss: 0.18167619407176971 2023-01-21 08:58:07.617441: step: 16/529, loss: 0.12678176164627075 2023-01-21 08:58:08.773166: step: 20/529, loss: 0.29143857955932617 2023-01-21 08:58:09.881040: step: 24/529, loss: 0.12004499137401581 2023-01-21 08:58:11.005672: step: 28/529, loss: 0.07078561931848526 2023-01-21 08:58:12.108798: step: 32/529, loss: 0.17433582246303558 2023-01-21 08:58:13.234962: step: 36/529, loss: 0.37478169798851013 2023-01-21 08:58:14.328788: step: 40/529, loss: 0.1412227749824524 2023-01-21 08:58:15.448592: step: 44/529, loss: 0.1689353883266449 2023-01-21 08:58:16.564004: step: 48/529, loss: 0.12702178955078125 2023-01-21 08:58:17.681386: step: 52/529, loss: 0.12902812659740448 2023-01-21 08:58:18.812980: step: 56/529, loss: 0.18746919929981232 2023-01-21 08:58:19.932785: step: 60/529, loss: 0.13098326325416565 2023-01-21 08:58:21.036155: step: 64/529, loss: 0.7679271697998047 2023-01-21 08:58:22.128759: step: 68/529, loss: 0.2812924385070801 2023-01-21 08:58:23.269428: step: 72/529, loss: 0.04974498599767685 2023-01-21 08:58:24.370904: step: 76/529, loss: 0.16167688369750977 2023-01-21 08:58:25.481862: step: 80/529, loss: 0.3687445819377899 2023-01-21 08:58:26.586067: step: 84/529, loss: 0.2611977159976959 2023-01-21 08:58:27.703831: step: 88/529, loss: 0.754612922668457 2023-01-21 08:58:28.824650: step: 92/529, loss: 0.35346242785453796 2023-01-21 08:58:29.947648: step: 96/529, loss: 0.09254489094018936 2023-01-21 08:58:31.066365: step: 100/529, loss: 0.5874930024147034 2023-01-21 08:58:32.153829: step: 104/529, loss: 0.4549519419670105 2023-01-21 08:58:33.246856: step: 108/529, loss: 0.7409536242485046 2023-01-21 08:58:34.360455: step: 112/529, loss: 0.2267622947692871 2023-01-21 08:58:35.478765: step: 116/529, loss: 0.24147367477416992 2023-01-21 08:58:36.576684: step: 120/529, loss: 0.05178098380565643 2023-01-21 08:58:37.688503: step: 124/529, loss: 0.05651436001062393 2023-01-21 08:58:38.793951: step: 128/529, loss: 0.6378711462020874 2023-01-21 08:58:39.884325: step: 132/529, loss: 0.11588358879089355 2023-01-21 08:58:40.993746: step: 136/529, loss: 0.14608259499073029 2023-01-21 08:58:42.093579: step: 140/529, loss: 0.2938465178012848 2023-01-21 08:58:43.209412: step: 144/529, loss: 0.14757758378982544 2023-01-21 08:58:44.304485: step: 148/529, loss: 0.06289301067590714 2023-01-21 08:58:45.430816: step: 152/529, loss: 0.5249178409576416 2023-01-21 08:58:46.561754: step: 156/529, loss: 0.13243542611598969 2023-01-21 08:58:47.665379: step: 160/529, loss: 0.13089700043201447 2023-01-21 08:58:48.780721: step: 164/529, loss: 1.393819808959961 2023-01-21 08:58:49.916692: step: 168/529, loss: 0.5352674722671509 2023-01-21 08:58:51.032490: step: 172/529, loss: 1.7434730529785156 2023-01-21 08:58:52.124625: step: 176/529, loss: 0.33729565143585205 2023-01-21 08:58:53.269073: step: 180/529, loss: 0.13706666231155396 2023-01-21 08:58:54.374457: step: 184/529, loss: 0.12028179317712784 2023-01-21 08:58:55.504169: step: 188/529, loss: 0.3744351863861084 2023-01-21 08:58:56.642983: step: 192/529, loss: 0.6514228582382202 2023-01-21 08:58:57.763433: step: 196/529, loss: 0.14292487502098083 2023-01-21 08:58:58.870059: step: 200/529, loss: 0.1727512925863266 2023-01-21 08:59:00.007224: step: 204/529, loss: 0.8078638315200806 2023-01-21 08:59:01.098132: step: 208/529, loss: 0.07706775516271591 2023-01-21 08:59:02.196706: step: 212/529, loss: 0.14386291801929474 2023-01-21 08:59:03.285975: step: 216/529, loss: 0.11720123887062073 2023-01-21 08:59:04.412616: step: 220/529, loss: 0.12753859162330627 2023-01-21 08:59:05.516297: step: 224/529, loss: 0.6784946918487549 2023-01-21 08:59:06.613269: step: 228/529, loss: 0.36837512254714966 2023-01-21 08:59:07.724187: step: 232/529, loss: 0.2314557135105133 2023-01-21 08:59:08.870635: step: 236/529, loss: 0.6996122598648071 2023-01-21 08:59:09.992455: step: 240/529, loss: 0.12325844168663025 2023-01-21 08:59:11.097609: step: 244/529, loss: 0.1912892758846283 2023-01-21 08:59:12.193911: step: 248/529, loss: 0.155488058924675 2023-01-21 08:59:13.308503: step: 252/529, loss: 0.2988097071647644 2023-01-21 08:59:14.423441: step: 256/529, loss: 0.2391349822282791 2023-01-21 08:59:15.564506: step: 260/529, loss: 0.21240109205245972 2023-01-21 08:59:16.675298: step: 264/529, loss: 0.14369621872901917 2023-01-21 08:59:17.806368: step: 268/529, loss: 0.073394775390625 2023-01-21 08:59:18.911812: step: 272/529, loss: 0.7761455178260803 2023-01-21 08:59:20.101878: step: 276/529, loss: 0.17235898971557617 2023-01-21 08:59:21.187068: step: 280/529, loss: 0.05230455473065376 2023-01-21 08:59:22.340510: step: 284/529, loss: 0.5639230608940125 2023-01-21 08:59:23.443734: step: 288/529, loss: 0.0593293160200119 2023-01-21 08:59:24.574333: step: 292/529, loss: 0.09990637749433517 2023-01-21 08:59:25.714247: step: 296/529, loss: 0.09749503433704376 2023-01-21 08:59:26.832910: step: 300/529, loss: 0.11232343316078186 2023-01-21 08:59:27.933297: step: 304/529, loss: 0.25702133774757385 2023-01-21 08:59:29.070600: step: 308/529, loss: 0.7633522152900696 2023-01-21 08:59:30.180530: step: 312/529, loss: 0.09290137141942978 2023-01-21 08:59:31.356319: step: 316/529, loss: 0.12361364811658859 2023-01-21 08:59:32.491152: step: 320/529, loss: 0.3535715639591217 2023-01-21 08:59:33.600359: step: 324/529, loss: 0.08157294243574142 2023-01-21 08:59:34.722634: step: 328/529, loss: 0.2751094400882721 2023-01-21 08:59:35.827989: step: 332/529, loss: 0.3667985200881958 2023-01-21 08:59:36.940036: step: 336/529, loss: 0.17447009682655334 2023-01-21 08:59:38.048526: step: 340/529, loss: 0.4401130676269531 2023-01-21 08:59:39.173762: step: 344/529, loss: 0.3819130063056946 2023-01-21 08:59:40.301549: step: 348/529, loss: 0.28337109088897705 2023-01-21 08:59:41.443404: step: 352/529, loss: 0.7851705551147461 2023-01-21 08:59:42.574587: step: 356/529, loss: 0.3282890319824219 2023-01-21 08:59:43.702460: step: 360/529, loss: 0.29331180453300476 2023-01-21 08:59:44.819220: step: 364/529, loss: 1.098751187324524 2023-01-21 08:59:45.910048: step: 368/529, loss: 0.7009627819061279 2023-01-21 08:59:47.010353: step: 372/529, loss: 0.02841215208172798 2023-01-21 08:59:48.127330: step: 376/529, loss: 0.21262173354625702 2023-01-21 08:59:49.255523: step: 380/529, loss: 0.1965780258178711 2023-01-21 08:59:50.395045: step: 384/529, loss: 0.6844325065612793 2023-01-21 08:59:51.531458: step: 388/529, loss: 0.2385258674621582 2023-01-21 08:59:52.650258: step: 392/529, loss: 0.03980207443237305 2023-01-21 08:59:53.753739: step: 396/529, loss: 0.859296441078186 2023-01-21 08:59:54.857780: step: 400/529, loss: 0.1316230744123459 2023-01-21 08:59:55.983235: step: 404/529, loss: 0.135273739695549 2023-01-21 08:59:57.110788: step: 408/529, loss: 1.2298753261566162 2023-01-21 08:59:58.222915: step: 412/529, loss: 0.25858479738235474 2023-01-21 08:59:59.338346: step: 416/529, loss: 0.05019988864660263 2023-01-21 09:00:00.460191: step: 420/529, loss: 0.8769996762275696 2023-01-21 09:00:01.563609: step: 424/529, loss: 0.1840103268623352 2023-01-21 09:00:02.708300: step: 428/529, loss: 1.0613447427749634 2023-01-21 09:00:03.821404: step: 432/529, loss: 0.09065236896276474 2023-01-21 09:00:04.943805: step: 436/529, loss: 0.12538452446460724 2023-01-21 09:00:06.054690: step: 440/529, loss: 0.22005310654640198 2023-01-21 09:00:07.169778: step: 444/529, loss: 0.1148044615983963 2023-01-21 09:00:08.302963: step: 448/529, loss: 0.07437839359045029 2023-01-21 09:00:09.382882: step: 452/529, loss: 0.06118325889110565 2023-01-21 09:00:10.502087: step: 456/529, loss: 0.13268804550170898 2023-01-21 09:00:11.627096: step: 460/529, loss: 0.34643229842185974 2023-01-21 09:00:12.762892: step: 464/529, loss: 0.3244630992412567 2023-01-21 09:00:13.878103: step: 468/529, loss: 0.30398598313331604 2023-01-21 09:00:15.013320: step: 472/529, loss: 0.22082296013832092 2023-01-21 09:00:16.145184: step: 476/529, loss: 0.1869596391916275 2023-01-21 09:00:17.257148: step: 480/529, loss: 0.5201295614242554 2023-01-21 09:00:18.361879: step: 484/529, loss: 6.798293590545654 2023-01-21 09:00:19.476805: step: 488/529, loss: 0.12290272861719131 2023-01-21 09:00:20.574805: step: 492/529, loss: 0.10110712051391602 2023-01-21 09:00:21.664467: step: 496/529, loss: 0.32220375537872314 2023-01-21 09:00:22.766740: step: 500/529, loss: 0.7954610586166382 2023-01-21 09:00:23.860194: step: 504/529, loss: 0.8145643472671509 2023-01-21 09:00:24.968471: step: 508/529, loss: 0.2596490979194641 2023-01-21 09:00:26.099160: step: 512/529, loss: 0.09804592281579971 2023-01-21 09:00:27.226834: step: 516/529, loss: 0.68377685546875 2023-01-21 09:00:28.320916: step: 520/529, loss: 0.08331127464771271 2023-01-21 09:00:29.422247: step: 524/529, loss: 0.7304783463478088 2023-01-21 09:00:30.527196: step: 528/529, loss: 0.029674889519810677 2023-01-21 09:00:31.647158: step: 532/529, loss: 1.4471884965896606 2023-01-21 09:00:32.777808: step: 536/529, loss: 0.07054352760314941 2023-01-21 09:00:33.876258: step: 540/529, loss: 0.17877325415611267 2023-01-21 09:00:34.975638: step: 544/529, loss: 0.46924152970314026 2023-01-21 09:00:36.089203: step: 548/529, loss: 0.09788914024829865 2023-01-21 09:00:37.222457: step: 552/529, loss: 0.5880101323127747 2023-01-21 09:00:38.343065: step: 556/529, loss: 1.1348435878753662 2023-01-21 09:00:39.466622: step: 560/529, loss: 0.24176493287086487 2023-01-21 09:00:40.596946: step: 564/529, loss: 0.12000855803489685 2023-01-21 09:00:41.761586: step: 568/529, loss: 0.49378010630607605 2023-01-21 09:00:42.879503: step: 572/529, loss: 0.09984531998634338 2023-01-21 09:00:44.021298: step: 576/529, loss: 0.3143824636936188 2023-01-21 09:00:45.130261: step: 580/529, loss: 0.21825580298900604 2023-01-21 09:00:46.261828: step: 584/529, loss: 0.26523810625076294 2023-01-21 09:00:47.375551: step: 588/529, loss: 0.26183652877807617 2023-01-21 09:00:48.503557: step: 592/529, loss: 0.5874617099761963 2023-01-21 09:00:49.667037: step: 596/529, loss: 1.2181098461151123 2023-01-21 09:00:50.798077: step: 600/529, loss: 0.6602262258529663 2023-01-21 09:00:51.936907: step: 604/529, loss: 0.4645019471645355 2023-01-21 09:00:53.091519: step: 608/529, loss: 0.2207527458667755 2023-01-21 09:00:54.216816: step: 612/529, loss: 0.09679269790649414 2023-01-21 09:00:55.318709: step: 616/529, loss: 1.0556899309158325 2023-01-21 09:00:56.444809: step: 620/529, loss: 1.3556993007659912 2023-01-21 09:00:57.554453: step: 624/529, loss: 0.06871695816516876 2023-01-21 09:00:58.683109: step: 628/529, loss: 0.4564691483974457 2023-01-21 09:00:59.803803: step: 632/529, loss: 0.2663634419441223 2023-01-21 09:01:00.933730: step: 636/529, loss: 1.0426877737045288 2023-01-21 09:01:02.058082: step: 640/529, loss: 0.040747880935668945 2023-01-21 09:01:03.153101: step: 644/529, loss: 0.3934265375137329 2023-01-21 09:01:04.286471: step: 648/529, loss: 0.20571193099021912 2023-01-21 09:01:05.417551: step: 652/529, loss: 0.24363946914672852 2023-01-21 09:01:06.529190: step: 656/529, loss: 0.503278374671936 2023-01-21 09:01:07.645726: step: 660/529, loss: 0.07683592289686203 2023-01-21 09:01:08.767121: step: 664/529, loss: 0.22995014488697052 2023-01-21 09:01:09.873442: step: 668/529, loss: 0.1132812574505806 2023-01-21 09:01:11.017796: step: 672/529, loss: 0.29321157932281494 2023-01-21 09:01:12.144153: step: 676/529, loss: 6.404324054718018 2023-01-21 09:01:13.251849: step: 680/529, loss: 0.7435594797134399 2023-01-21 09:01:14.370433: step: 684/529, loss: 0.1779492348432541 2023-01-21 09:01:15.474296: step: 688/529, loss: 0.19195112586021423 2023-01-21 09:01:16.580722: step: 692/529, loss: 0.04702947288751602 2023-01-21 09:01:17.691690: step: 696/529, loss: 0.03895268589258194 2023-01-21 09:01:18.843485: step: 700/529, loss: 0.1714916229248047 2023-01-21 09:01:19.928002: step: 704/529, loss: 0.11572809517383575 2023-01-21 09:01:21.038840: step: 708/529, loss: 0.7540234923362732 2023-01-21 09:01:22.162030: step: 712/529, loss: 0.5779246091842651 2023-01-21 09:01:23.280572: step: 716/529, loss: 0.18387430906295776 2023-01-21 09:01:24.382700: step: 720/529, loss: 0.19179849326610565 2023-01-21 09:01:25.493075: step: 724/529, loss: 0.9198240041732788 2023-01-21 09:01:26.620974: step: 728/529, loss: 0.9045203924179077 2023-01-21 09:01:27.729872: step: 732/529, loss: 0.038695622235536575 2023-01-21 09:01:28.851148: step: 736/529, loss: 0.6395885944366455 2023-01-21 09:01:29.964544: step: 740/529, loss: 0.13464191555976868 2023-01-21 09:01:31.080013: step: 744/529, loss: 0.3594089448451996 2023-01-21 09:01:32.188454: step: 748/529, loss: 0.3688087463378906 2023-01-21 09:01:33.283772: step: 752/529, loss: 0.06678777188062668 2023-01-21 09:01:34.383762: step: 756/529, loss: 0.40065494179725647 2023-01-21 09:01:35.524179: step: 760/529, loss: 0.17610502243041992 2023-01-21 09:01:36.631353: step: 764/529, loss: 0.7588016390800476 2023-01-21 09:01:37.735130: step: 768/529, loss: 0.8652192950248718 2023-01-21 09:01:38.894075: step: 772/529, loss: 0.5788482427597046 2023-01-21 09:01:39.980026: step: 776/529, loss: 0.06973810493946075 2023-01-21 09:01:41.117541: step: 780/529, loss: 0.1131652370095253 2023-01-21 09:01:42.245574: step: 784/529, loss: 0.7334602475166321 2023-01-21 09:01:43.334837: step: 788/529, loss: 0.20820684731006622 2023-01-21 09:01:44.524115: step: 792/529, loss: 0.2806049585342407 2023-01-21 09:01:45.647404: step: 796/529, loss: 0.12473927438259125 2023-01-21 09:01:46.752171: step: 800/529, loss: 0.6042113304138184 2023-01-21 09:01:47.861475: step: 804/529, loss: 0.34010282158851624 2023-01-21 09:01:48.998467: step: 808/529, loss: 0.1561412364244461 2023-01-21 09:01:50.115778: step: 812/529, loss: 0.09789478778839111 2023-01-21 09:01:51.235308: step: 816/529, loss: 0.9361017942428589 2023-01-21 09:01:52.367415: step: 820/529, loss: 0.1968742311000824 2023-01-21 09:01:53.471424: step: 824/529, loss: 1.2033723592758179 2023-01-21 09:01:54.594763: step: 828/529, loss: 0.13836780190467834 2023-01-21 09:01:55.726157: step: 832/529, loss: 0.1800197809934616 2023-01-21 09:01:56.836701: step: 836/529, loss: 0.35231488943099976 2023-01-21 09:01:57.960500: step: 840/529, loss: 0.7589906454086304 2023-01-21 09:01:59.082999: step: 844/529, loss: 0.12929654121398926 2023-01-21 09:02:00.195022: step: 848/529, loss: 0.15113696455955505 2023-01-21 09:02:01.306537: step: 852/529, loss: 0.7858595252037048 2023-01-21 09:02:02.472342: step: 856/529, loss: 0.7015398740768433 2023-01-21 09:02:03.571314: step: 860/529, loss: 0.2875896394252777 2023-01-21 09:02:04.679012: step: 864/529, loss: 0.5598541498184204 2023-01-21 09:02:05.826282: step: 868/529, loss: 0.20690393447875977 2023-01-21 09:02:06.929251: step: 872/529, loss: 0.6398031711578369 2023-01-21 09:02:08.026943: step: 876/529, loss: 0.6948320269584656 2023-01-21 09:02:09.140602: step: 880/529, loss: 0.7434417009353638 2023-01-21 09:02:10.277767: step: 884/529, loss: 0.31769075989723206 2023-01-21 09:02:11.395529: step: 888/529, loss: 0.04603876918554306 2023-01-21 09:02:12.544117: step: 892/529, loss: 0.68703693151474 2023-01-21 09:02:13.648060: step: 896/529, loss: 1.257704496383667 2023-01-21 09:02:14.774978: step: 900/529, loss: 0.04991288483142853 2023-01-21 09:02:15.903118: step: 904/529, loss: 0.033030033111572266 2023-01-21 09:02:17.015322: step: 908/529, loss: 0.18021836876869202 2023-01-21 09:02:18.144593: step: 912/529, loss: 0.31129178404808044 2023-01-21 09:02:19.222171: step: 916/529, loss: 0.08821897953748703 2023-01-21 09:02:20.351992: step: 920/529, loss: 0.3284940719604492 2023-01-21 09:02:21.498537: step: 924/529, loss: 0.44003069400787354 2023-01-21 09:02:22.601677: step: 928/529, loss: 0.7072644233703613 2023-01-21 09:02:23.720221: step: 932/529, loss: 0.04998297989368439 2023-01-21 09:02:24.843271: step: 936/529, loss: 1.1905598640441895 2023-01-21 09:02:25.948685: step: 940/529, loss: 0.04226398468017578 2023-01-21 09:02:27.065767: step: 944/529, loss: 0.4978475570678711 2023-01-21 09:02:28.197627: step: 948/529, loss: 0.1130976676940918 2023-01-21 09:02:29.320595: step: 952/529, loss: 0.7615178227424622 2023-01-21 09:02:30.414854: step: 956/529, loss: 0.09622316062450409 2023-01-21 09:02:31.539898: step: 960/529, loss: 0.05787640064954758 2023-01-21 09:02:32.667879: step: 964/529, loss: 0.21396437287330627 2023-01-21 09:02:33.794087: step: 968/529, loss: 0.1683407723903656 2023-01-21 09:02:34.903922: step: 972/529, loss: 0.13061919808387756 2023-01-21 09:02:36.019341: step: 976/529, loss: 0.08372955769300461 2023-01-21 09:02:37.136078: step: 980/529, loss: 0.09518122673034668 2023-01-21 09:02:38.248067: step: 984/529, loss: 0.7406069040298462 2023-01-21 09:02:39.357933: step: 988/529, loss: 0.08976936340332031 2023-01-21 09:02:40.449489: step: 992/529, loss: 0.33814573287963867 2023-01-21 09:02:41.572039: step: 996/529, loss: 0.14185667037963867 2023-01-21 09:02:42.679152: step: 1000/529, loss: 0.038360595703125 2023-01-21 09:02:43.786367: step: 1004/529, loss: 0.12126073986291885 2023-01-21 09:02:44.928442: step: 1008/529, loss: 0.3978820741176605 2023-01-21 09:02:46.064880: step: 1012/529, loss: 0.07474441826343536 2023-01-21 09:02:47.182015: step: 1016/529, loss: 0.1304573118686676 2023-01-21 09:02:48.289884: step: 1020/529, loss: 0.23377451300621033 2023-01-21 09:02:49.364861: step: 1024/529, loss: 7.644143581390381 2023-01-21 09:02:50.499515: step: 1028/529, loss: 0.0811273604631424 2023-01-21 09:02:51.619550: step: 1032/529, loss: 6.218911647796631 2023-01-21 09:02:52.748389: step: 1036/529, loss: 0.1395483911037445 2023-01-21 09:02:53.864511: step: 1040/529, loss: 0.8954647779464722 2023-01-21 09:02:54.975141: step: 1044/529, loss: 0.15395589172840118 2023-01-21 09:02:56.101295: step: 1048/529, loss: 0.7681409120559692 2023-01-21 09:02:57.218459: step: 1052/529, loss: 0.11389818787574768 2023-01-21 09:02:58.341723: step: 1056/529, loss: 0.2399909943342209 2023-01-21 09:02:59.449595: step: 1060/529, loss: 0.05504007637500763 2023-01-21 09:03:00.606029: step: 1064/529, loss: 0.4736661911010742 2023-01-21 09:03:01.709965: step: 1068/529, loss: 0.3544514775276184 2023-01-21 09:03:02.824019: step: 1072/529, loss: 0.2522240877151489 2023-01-21 09:03:03.941217: step: 1076/529, loss: 0.08705344796180725 2023-01-21 09:03:05.078728: step: 1080/529, loss: 0.12338896095752716 2023-01-21 09:03:06.191074: step: 1084/529, loss: 0.7841986417770386 2023-01-21 09:03:07.297934: step: 1088/529, loss: 0.80646151304245 2023-01-21 09:03:08.428341: step: 1092/529, loss: 0.7311602830886841 2023-01-21 09:03:09.533316: step: 1096/529, loss: 0.21718692779541016 2023-01-21 09:03:10.628519: step: 1100/529, loss: 0.17857493460178375 2023-01-21 09:03:11.739551: step: 1104/529, loss: 0.05556363984942436 2023-01-21 09:03:12.841689: step: 1108/529, loss: 0.7564243674278259 2023-01-21 09:03:13.978083: step: 1112/529, loss: 0.33097362518310547 2023-01-21 09:03:15.086493: step: 1116/529, loss: 1.0567814111709595 2023-01-21 09:03:16.197670: step: 1120/529, loss: 1.129134178161621 2023-01-21 09:03:17.284172: step: 1124/529, loss: 0.3711276054382324 2023-01-21 09:03:18.422480: step: 1128/529, loss: 0.09012921154499054 2023-01-21 09:03:19.558193: step: 1132/529, loss: 0.12297835946083069 2023-01-21 09:03:20.698685: step: 1136/529, loss: 0.10164375603199005 2023-01-21 09:03:21.822790: step: 1140/529, loss: 0.13223138451576233 2023-01-21 09:03:22.927604: step: 1144/529, loss: 0.05243721231818199 2023-01-21 09:03:24.032084: step: 1148/529, loss: 0.1398342251777649 2023-01-21 09:03:25.162764: step: 1152/529, loss: 0.09825744479894638 2023-01-21 09:03:26.280811: step: 1156/529, loss: 0.24497276544570923 2023-01-21 09:03:27.424860: step: 1160/529, loss: 0.12554606795310974 2023-01-21 09:03:28.535452: step: 1164/529, loss: 0.3235568404197693 2023-01-21 09:03:29.649476: step: 1168/529, loss: 0.2562273144721985 2023-01-21 09:03:30.774631: step: 1172/529, loss: 0.19707414507865906 2023-01-21 09:03:31.892404: step: 1176/529, loss: 0.29449576139450073 2023-01-21 09:03:33.014822: step: 1180/529, loss: 0.5240601301193237 2023-01-21 09:03:34.133663: step: 1184/529, loss: 0.5446715354919434 2023-01-21 09:03:35.252125: step: 1188/529, loss: 0.16990713775157928 2023-01-21 09:03:36.391333: step: 1192/529, loss: 0.25677481293678284 2023-01-21 09:03:37.523696: step: 1196/529, loss: 0.11575216054916382 2023-01-21 09:03:38.636710: step: 1200/529, loss: 0.05895288288593292 2023-01-21 09:03:39.746187: step: 1204/529, loss: 0.29789239168167114 2023-01-21 09:03:40.849824: step: 1208/529, loss: 0.24778369069099426 2023-01-21 09:03:41.951491: step: 1212/529, loss: 0.09482560306787491 2023-01-21 09:03:43.070065: step: 1216/529, loss: 0.38792508840560913 2023-01-21 09:03:44.176383: step: 1220/529, loss: 0.1624559462070465 2023-01-21 09:03:45.354799: step: 1224/529, loss: 0.06335077434778214 2023-01-21 09:03:46.467980: step: 1228/529, loss: 0.13133525848388672 2023-01-21 09:03:47.607546: step: 1232/529, loss: 0.36569005250930786 2023-01-21 09:03:48.730846: step: 1236/529, loss: 0.11706504225730896 2023-01-21 09:03:49.842452: step: 1240/529, loss: 0.2033659964799881 2023-01-21 09:03:50.997341: step: 1244/529, loss: 0.2285289764404297 2023-01-21 09:03:52.143119: step: 1248/529, loss: 0.3451489210128784 2023-01-21 09:03:53.245985: step: 1252/529, loss: 6.266887664794922 2023-01-21 09:03:54.372455: step: 1256/529, loss: 0.2964404225349426 2023-01-21 09:03:55.513057: step: 1260/529, loss: 0.26768603920936584 2023-01-21 09:03:56.641997: step: 1264/529, loss: 0.14428062736988068 2023-01-21 09:03:57.772718: step: 1268/529, loss: 0.07190468907356262 2023-01-21 09:03:58.925074: step: 1272/529, loss: 0.2975412607192993 2023-01-21 09:04:00.076031: step: 1276/529, loss: 0.1005764976143837 2023-01-21 09:04:01.199784: step: 1280/529, loss: 0.15953359007835388 2023-01-21 09:04:02.336475: step: 1284/529, loss: 0.24019962549209595 2023-01-21 09:04:03.473772: step: 1288/529, loss: 0.1978778839111328 2023-01-21 09:04:04.590230: step: 1292/529, loss: 0.39905303716659546 2023-01-21 09:04:05.728532: step: 1296/529, loss: 0.44769737124443054 2023-01-21 09:04:06.880463: step: 1300/529, loss: 0.3252217471599579 2023-01-21 09:04:08.032449: step: 1304/529, loss: 7.657644271850586 2023-01-21 09:04:09.146208: step: 1308/529, loss: 0.6352218985557556 2023-01-21 09:04:10.279767: step: 1312/529, loss: 0.5822814702987671 2023-01-21 09:04:11.412805: step: 1316/529, loss: 0.20116862654685974 2023-01-21 09:04:12.512378: step: 1320/529, loss: 0.04348144680261612 2023-01-21 09:04:13.614464: step: 1324/529, loss: 0.10757599025964737 2023-01-21 09:04:14.745258: step: 1328/529, loss: 0.24369993805885315 2023-01-21 09:04:15.844265: step: 1332/529, loss: 0.02398652955889702 2023-01-21 09:04:16.984484: step: 1336/529, loss: 1.0330395698547363 2023-01-21 09:04:18.084987: step: 1340/529, loss: 0.09930892288684845 2023-01-21 09:04:19.271546: step: 1344/529, loss: 1.3859328031539917 2023-01-21 09:04:20.389720: step: 1348/529, loss: 0.14805221557617188 2023-01-21 09:04:21.484655: step: 1352/529, loss: 0.6569976806640625 2023-01-21 09:04:22.599134: step: 1356/529, loss: 0.14053136110305786 2023-01-21 09:04:23.742521: step: 1360/529, loss: 0.11296391487121582 2023-01-21 09:04:24.859323: step: 1364/529, loss: 0.20357605814933777 2023-01-21 09:04:25.948044: step: 1368/529, loss: 0.6747820973396301 2023-01-21 09:04:27.054842: step: 1372/529, loss: 0.7554526925086975 2023-01-21 09:04:28.142182: step: 1376/529, loss: 0.27689898014068604 2023-01-21 09:04:29.249044: step: 1380/529, loss: 0.16605934500694275 2023-01-21 09:04:30.348177: step: 1384/529, loss: 0.18032407760620117 2023-01-21 09:04:31.457023: step: 1388/529, loss: 0.1738419085741043 2023-01-21 09:04:32.584808: step: 1392/529, loss: 6.825258731842041 2023-01-21 09:04:33.697205: step: 1396/529, loss: 0.10907812416553497 2023-01-21 09:04:34.826678: step: 1400/529, loss: 0.09888801723718643 2023-01-21 09:04:35.931989: step: 1404/529, loss: 0.9205310940742493 2023-01-21 09:04:37.040440: step: 1408/529, loss: 0.3518640398979187 2023-01-21 09:04:38.157026: step: 1412/529, loss: 0.11014070361852646 2023-01-21 09:04:39.249322: step: 1416/529, loss: 2.866546154022217 2023-01-21 09:04:40.363553: step: 1420/529, loss: 0.08448782563209534 2023-01-21 09:04:41.494831: step: 1424/529, loss: 0.17541609704494476 2023-01-21 09:04:42.613437: step: 1428/529, loss: 0.0688701644539833 2023-01-21 09:04:43.716248: step: 1432/529, loss: 0.22905360162258148 2023-01-21 09:04:44.865121: step: 1436/529, loss: 0.05497455969452858 2023-01-21 09:04:45.963188: step: 1440/529, loss: 0.6626628041267395 2023-01-21 09:04:47.088183: step: 1444/529, loss: 0.2110680639743805 2023-01-21 09:04:48.191596: step: 1448/529, loss: 0.2953304350376129 2023-01-21 09:04:49.307931: step: 1452/529, loss: 0.08944135159254074 2023-01-21 09:04:50.422053: step: 1456/529, loss: 0.5831474661827087 2023-01-21 09:04:51.533927: step: 1460/529, loss: 0.51947420835495 2023-01-21 09:04:52.661624: step: 1464/529, loss: 0.2941949665546417 2023-01-21 09:04:53.787210: step: 1468/529, loss: 0.12500719726085663 2023-01-21 09:04:54.923326: step: 1472/529, loss: 2.186870574951172 2023-01-21 09:04:56.030628: step: 1476/529, loss: 0.17179012298583984 2023-01-21 09:04:57.147072: step: 1480/529, loss: 1.3619778156280518 2023-01-21 09:04:58.260356: step: 1484/529, loss: 0.26388871669769287 2023-01-21 09:04:59.348013: step: 1488/529, loss: 0.8584086894989014 2023-01-21 09:05:00.466136: step: 1492/529, loss: 0.1718585044145584 2023-01-21 09:05:01.608808: step: 1496/529, loss: 0.8099654912948608 2023-01-21 09:05:02.779147: step: 1500/529, loss: 0.7845700979232788 2023-01-21 09:05:03.893887: step: 1504/529, loss: 0.10382194072008133 2023-01-21 09:05:04.998669: step: 1508/529, loss: 0.08117265999317169 2023-01-21 09:05:06.093639: step: 1512/529, loss: 0.22360281646251678 2023-01-21 09:05:07.212646: step: 1516/529, loss: 1.9387227296829224 2023-01-21 09:05:08.316805: step: 1520/529, loss: 0.13123612105846405 2023-01-21 09:05:09.443237: step: 1524/529, loss: 0.1132390946149826 2023-01-21 09:05:10.579677: step: 1528/529, loss: 0.7765964269638062 2023-01-21 09:05:11.695581: step: 1532/529, loss: 0.25017109513282776 2023-01-21 09:05:12.828639: step: 1536/529, loss: 0.3031921982765198 2023-01-21 09:05:13.939113: step: 1540/529, loss: 0.04496727138757706 2023-01-21 09:05:15.078273: step: 1544/529, loss: 0.8941881060600281 2023-01-21 09:05:16.188512: step: 1548/529, loss: 0.5266496539115906 2023-01-21 09:05:17.331517: step: 1552/529, loss: 0.3112978935241699 2023-01-21 09:05:18.465983: step: 1556/529, loss: 0.2610532343387604 2023-01-21 09:05:19.597562: step: 1560/529, loss: 0.13099098205566406 2023-01-21 09:05:20.726547: step: 1564/529, loss: 0.1538313329219818 2023-01-21 09:05:21.847543: step: 1568/529, loss: 0.5081146955490112 2023-01-21 09:05:22.988685: step: 1572/529, loss: 0.3228822946548462 2023-01-21 09:05:24.088449: step: 1576/529, loss: 0.7244106531143188 2023-01-21 09:05:25.227099: step: 1580/529, loss: 0.07521490752696991 2023-01-21 09:05:26.336655: step: 1584/529, loss: 0.1976000815629959 2023-01-21 09:05:27.506007: step: 1588/529, loss: 1.2680259943008423 2023-01-21 09:05:28.608575: step: 1592/529, loss: 0.4451565742492676 2023-01-21 09:05:29.719684: step: 1596/529, loss: 0.12641993165016174 2023-01-21 09:05:30.842108: step: 1600/529, loss: 0.06996989250183105 2023-01-21 09:05:31.949729: step: 1604/529, loss: 0.4087117314338684 2023-01-21 09:05:33.105259: step: 1608/529, loss: 0.12463803589344025 2023-01-21 09:05:34.210481: step: 1612/529, loss: 7.072520732879639 2023-01-21 09:05:35.347031: step: 1616/529, loss: 0.9215995669364929 2023-01-21 09:05:36.456654: step: 1620/529, loss: 0.11515302956104279 2023-01-21 09:05:37.560728: step: 1624/529, loss: 0.03171682357788086 2023-01-21 09:05:38.696674: step: 1628/529, loss: 0.604544997215271 2023-01-21 09:05:39.824625: step: 1632/529, loss: 0.11626625806093216 2023-01-21 09:05:40.909930: step: 1636/529, loss: 0.09962864220142365 2023-01-21 09:05:42.031609: step: 1640/529, loss: 0.17779016494750977 2023-01-21 09:05:43.153836: step: 1644/529, loss: 0.11542224884033203 2023-01-21 09:05:44.259881: step: 1648/529, loss: 0.6304776668548584 2023-01-21 09:05:45.377547: step: 1652/529, loss: 0.150037482380867 2023-01-21 09:05:46.494869: step: 1656/529, loss: 0.10930319130420685 2023-01-21 09:05:47.611335: step: 1660/529, loss: 0.054425764828920364 2023-01-21 09:05:48.713142: step: 1664/529, loss: 0.8273213505744934 2023-01-21 09:05:49.852983: step: 1668/529, loss: 0.11428241431713104 2023-01-21 09:05:50.991669: step: 1672/529, loss: 0.5153927803039551 2023-01-21 09:05:52.087786: step: 1676/529, loss: 0.07320769131183624 2023-01-21 09:05:53.237689: step: 1680/529, loss: 1.597503900527954 2023-01-21 09:05:54.371486: step: 1684/529, loss: 0.15810072422027588 2023-01-21 09:05:55.489736: step: 1688/529, loss: 0.9738703370094299 2023-01-21 09:05:56.596243: step: 1692/529, loss: 0.14176030457019806 2023-01-21 09:05:57.724172: step: 1696/529, loss: 0.15938439965248108 2023-01-21 09:05:58.842701: step: 1700/529, loss: 0.12168703973293304 2023-01-21 09:05:59.984517: step: 1704/529, loss: 0.24064913392066956 2023-01-21 09:06:01.091663: step: 1708/529, loss: 0.3153999447822571 2023-01-21 09:06:02.215324: step: 1712/529, loss: 6.021102428436279 2023-01-21 09:06:03.305836: step: 1716/529, loss: 0.29667505621910095 2023-01-21 09:06:04.405777: step: 1720/529, loss: 0.1244380921125412 2023-01-21 09:06:05.509532: step: 1724/529, loss: 0.11710548400878906 2023-01-21 09:06:06.623108: step: 1728/529, loss: 0.10147824138402939 2023-01-21 09:06:07.736797: step: 1732/529, loss: 0.16996492445468903 2023-01-21 09:06:08.822759: step: 1736/529, loss: 0.6971297860145569 2023-01-21 09:06:09.968779: step: 1740/529, loss: 0.1374097466468811 2023-01-21 09:06:11.074953: step: 1744/529, loss: 6.310600757598877 2023-01-21 09:06:12.199760: step: 1748/529, loss: 1.4986711740493774 2023-01-21 09:06:13.310391: step: 1752/529, loss: 0.6713878512382507 2023-01-21 09:06:14.423033: step: 1756/529, loss: 0.1214049905538559 2023-01-21 09:06:15.572669: step: 1760/529, loss: 0.24521350860595703 2023-01-21 09:06:16.678222: step: 1764/529, loss: 0.2385249137878418 2023-01-21 09:06:17.785324: step: 1768/529, loss: 0.3238133192062378 2023-01-21 09:06:18.899955: step: 1772/529, loss: 0.4617188572883606 2023-01-21 09:06:20.043097: step: 1776/529, loss: 0.13507910072803497 2023-01-21 09:06:21.181464: step: 1780/529, loss: 0.27470436692237854 2023-01-21 09:06:22.288331: step: 1784/529, loss: 0.3224661946296692 2023-01-21 09:06:23.434373: step: 1788/529, loss: 0.17379580438137054 2023-01-21 09:06:24.551580: step: 1792/529, loss: 0.21926411986351013 2023-01-21 09:06:25.649153: step: 1796/529, loss: 0.10533003509044647 2023-01-21 09:06:26.799596: step: 1800/529, loss: 4.866824150085449 2023-01-21 09:06:27.885575: step: 1804/529, loss: 0.46708834171295166 2023-01-21 09:06:28.986936: step: 1808/529, loss: 0.2679416537284851 2023-01-21 09:06:30.095489: step: 1812/529, loss: 0.0899043083190918 2023-01-21 09:06:31.211728: step: 1816/529, loss: 0.09973478317260742 2023-01-21 09:06:32.337437: step: 1820/529, loss: 0.1612849235534668 2023-01-21 09:06:33.437731: step: 1824/529, loss: 1.3038971424102783 2023-01-21 09:06:34.543097: step: 1828/529, loss: 0.2686694860458374 2023-01-21 09:06:35.671440: step: 1832/529, loss: 0.2632485628128052 2023-01-21 09:06:36.792105: step: 1836/529, loss: 0.12849335372447968 2023-01-21 09:06:37.897052: step: 1840/529, loss: 0.2791590690612793 2023-01-21 09:06:39.004180: step: 1844/529, loss: 0.42025184631347656 2023-01-21 09:06:40.108084: step: 1848/529, loss: 0.24211034178733826 2023-01-21 09:06:41.197203: step: 1852/529, loss: 0.15408745408058167 2023-01-21 09:06:42.323692: step: 1856/529, loss: 0.1954731047153473 2023-01-21 09:06:43.428810: step: 1860/529, loss: 0.07411280274391174 2023-01-21 09:06:44.539818: step: 1864/529, loss: 0.28218376636505127 2023-01-21 09:06:45.629779: step: 1868/529, loss: 0.06740951538085938 2023-01-21 09:06:46.747404: step: 1872/529, loss: 0.42508286237716675 2023-01-21 09:06:47.867931: step: 1876/529, loss: 0.2036547213792801 2023-01-21 09:06:48.969941: step: 1880/529, loss: 0.1151064932346344 2023-01-21 09:06:50.104296: step: 1884/529, loss: 0.5739890336990356 2023-01-21 09:06:51.228497: step: 1888/529, loss: 0.42548856139183044 2023-01-21 09:06:52.324832: step: 1892/529, loss: 0.08126717060804367 2023-01-21 09:06:53.429634: step: 1896/529, loss: 0.2387586236000061 2023-01-21 09:06:54.517749: step: 1900/529, loss: 0.638191819190979 2023-01-21 09:06:55.644279: step: 1904/529, loss: 0.668250322341919 2023-01-21 09:06:56.778104: step: 1908/529, loss: 0.1001528799533844 2023-01-21 09:06:57.882282: step: 1912/529, loss: 0.0707489550113678 2023-01-21 09:06:59.020697: step: 1916/529, loss: 0.2851121127605438 2023-01-21 09:07:00.146053: step: 1920/529, loss: 0.2972749173641205 2023-01-21 09:07:01.284475: step: 1924/529, loss: 0.06979580223560333 2023-01-21 09:07:02.390095: step: 1928/529, loss: 0.10663776844739914 2023-01-21 09:07:03.490752: step: 1932/529, loss: 0.11231927573680878 2023-01-21 09:07:04.618164: step: 1936/529, loss: 0.904582679271698 2023-01-21 09:07:05.723416: step: 1940/529, loss: 0.683761715888977 2023-01-21 09:07:06.831732: step: 1944/529, loss: 0.5466469526290894 2023-01-21 09:07:07.931723: step: 1948/529, loss: 0.40903064608573914 2023-01-21 09:07:09.079256: step: 1952/529, loss: 0.21860074996948242 2023-01-21 09:07:10.186370: step: 1956/529, loss: 0.4554615020751953 2023-01-21 09:07:11.321815: step: 1960/529, loss: 0.14795179665088654 2023-01-21 09:07:12.455495: step: 1964/529, loss: 0.16426968574523926 2023-01-21 09:07:13.564337: step: 1968/529, loss: 0.19438907504081726 2023-01-21 09:07:14.650667: step: 1972/529, loss: 0.6508668661117554 2023-01-21 09:07:15.753754: step: 1976/529, loss: 0.05450606346130371 2023-01-21 09:07:16.860173: step: 1980/529, loss: 0.5855188965797424 2023-01-21 09:07:17.977567: step: 1984/529, loss: 0.14482136070728302 2023-01-21 09:07:19.077915: step: 1988/529, loss: 0.9255144596099854 2023-01-21 09:07:20.236713: step: 1992/529, loss: 0.17067299783229828 2023-01-21 09:07:21.354962: step: 1996/529, loss: 0.09915242344141006 2023-01-21 09:07:22.457963: step: 2000/529, loss: 0.08624716103076935 2023-01-21 09:07:23.553525: step: 2004/529, loss: 0.06186475604772568 2023-01-21 09:07:24.642412: step: 2008/529, loss: 0.2465021163225174 2023-01-21 09:07:25.769054: step: 2012/529, loss: 0.08015727996826172 2023-01-21 09:07:26.852513: step: 2016/529, loss: 0.137742280960083 2023-01-21 09:07:27.985014: step: 2020/529, loss: 0.7874522805213928 2023-01-21 09:07:29.117570: step: 2024/529, loss: 0.7217190265655518 2023-01-21 09:07:30.268670: step: 2028/529, loss: 1.4282710552215576 2023-01-21 09:07:31.369488: step: 2032/529, loss: 0.07230110466480255 2023-01-21 09:07:32.494663: step: 2036/529, loss: 0.21527576446533203 2023-01-21 09:07:33.612755: step: 2040/529, loss: 0.18370608985424042 2023-01-21 09:07:34.738519: step: 2044/529, loss: 1.398308515548706 2023-01-21 09:07:35.841589: step: 2048/529, loss: 0.126674085855484 2023-01-21 09:07:36.954832: step: 2052/529, loss: 0.30660954117774963 2023-01-21 09:07:38.050160: step: 2056/529, loss: 0.08510196208953857 2023-01-21 09:07:39.193182: step: 2060/529, loss: 0.21761254966259003 2023-01-21 09:07:40.336441: step: 2064/529, loss: 2.3670084476470947 2023-01-21 09:07:41.451789: step: 2068/529, loss: 0.43338441848754883 2023-01-21 09:07:42.569336: step: 2072/529, loss: 0.1459934264421463 2023-01-21 09:07:43.692771: step: 2076/529, loss: 0.3969666659832001 2023-01-21 09:07:44.804781: step: 2080/529, loss: 0.669745683670044 2023-01-21 09:07:45.933723: step: 2084/529, loss: 0.7680432796478271 2023-01-21 09:07:47.029079: step: 2088/529, loss: 0.3431036174297333 2023-01-21 09:07:48.167273: step: 2092/529, loss: 0.20547810196876526 2023-01-21 09:07:49.278613: step: 2096/529, loss: 0.6426596641540527 2023-01-21 09:07:50.427129: step: 2100/529, loss: 0.9980220198631287 2023-01-21 09:07:51.537088: step: 2104/529, loss: 6.7970499992370605 2023-01-21 09:07:52.647604: step: 2108/529, loss: 0.9798686504364014 2023-01-21 09:07:53.758142: step: 2112/529, loss: 0.06655607372522354 2023-01-21 09:07:54.917362: step: 2116/529, loss: 0.3421632647514343 ================================================== Loss: 0.501 -------------------- Dev: {'event': {'p': 0.5572815533980583, 'r': 0.7643142476697736, 'f1': 0.6445816956765862}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Test: {'event': {'p': 0.5485519591141397, 'r': 0.7268623024830699, 'f1': 0.6252427184466018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Chinese: {'event': {'p': 0.5632183908045977, 'r': 0.9074074074074074, 'f1': 0.6950354609929078}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Korean: {'event': {'p': 0.7111111111111111, 'r': 0.5079365079365079, 'f1': 0.5925925925925924}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Russian: {'event': {'p': 0.5945945945945946, 'r': 0.6111111111111112, 'f1': 0.6027397260273972}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5572815533980583, 'r': 0.7643142476697736, 'f1': 0.6445816956765862}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Korean: {'event': {'p': 0.5485519591141397, 'r': 0.7268623024830699, 'f1': 0.6252427184466018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Korean: {'event': {'p': 0.7111111111111111, 'r': 0.5079365079365079, 'f1': 0.5925925925925924}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Russian: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Russian: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Russian: {'event': {'p': 0.6176470588235294, 'r': 0.5833333333333334, 'f1': 0.6}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} ****************************** Epoch: 4 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:08:44.191555: step: 4/529, loss: 0.6615797281265259 2023-01-21 09:08:45.278476: step: 8/529, loss: 0.05763940513134003 2023-01-21 09:08:46.421832: step: 12/529, loss: 0.19935540854930878 2023-01-21 09:08:47.534799: step: 16/529, loss: 0.0393078587949276 2023-01-21 09:08:48.675713: step: 20/529, loss: 0.8522058725357056 2023-01-21 09:08:49.788150: step: 24/529, loss: 0.12867431342601776 2023-01-21 09:08:50.929214: step: 28/529, loss: 0.18457098305225372 2023-01-21 09:08:52.075269: step: 32/529, loss: 0.08480529487133026 2023-01-21 09:08:53.189065: step: 36/529, loss: 0.04493327438831329 2023-01-21 09:08:54.302400: step: 40/529, loss: 0.1839262843132019 2023-01-21 09:08:55.400898: step: 44/529, loss: 0.5116775631904602 2023-01-21 09:08:56.502127: step: 48/529, loss: 0.05727691948413849 2023-01-21 09:08:57.626418: step: 52/529, loss: 0.1392023265361786 2023-01-21 09:08:58.727639: step: 56/529, loss: 0.21133175492286682 2023-01-21 09:08:59.839773: step: 60/529, loss: 1.5186076164245605 2023-01-21 09:09:00.945442: step: 64/529, loss: 0.1221562922000885 2023-01-21 09:09:02.146835: step: 68/529, loss: 0.1041574478149414 2023-01-21 09:09:03.248738: step: 72/529, loss: 0.1703396737575531 2023-01-21 09:09:04.362381: step: 76/529, loss: 0.17172327637672424 2023-01-21 09:09:05.483543: step: 80/529, loss: 1.1180455684661865 2023-01-21 09:09:06.598780: step: 84/529, loss: 0.1603957712650299 2023-01-21 09:09:07.694708: step: 88/529, loss: 0.46226492524147034 2023-01-21 09:09:08.821436: step: 92/529, loss: 0.5222301483154297 2023-01-21 09:09:09.924031: step: 96/529, loss: 0.07915011048316956 2023-01-21 09:09:11.052049: step: 100/529, loss: 0.22018995881080627 2023-01-21 09:09:12.168060: step: 104/529, loss: 0.1150629073381424 2023-01-21 09:09:13.280966: step: 108/529, loss: 0.6044923663139343 2023-01-21 09:09:14.418461: step: 112/529, loss: 0.12086892873048782 2023-01-21 09:09:15.553227: step: 116/529, loss: 0.15235671401023865 2023-01-21 09:09:16.677759: step: 120/529, loss: 0.3981592357158661 2023-01-21 09:09:17.793349: step: 124/529, loss: 0.9084380865097046 2023-01-21 09:09:18.923648: step: 128/529, loss: 1.6567373275756836 2023-01-21 09:09:20.049015: step: 132/529, loss: 0.8471486568450928 2023-01-21 09:09:21.168991: step: 136/529, loss: 0.15174369513988495 2023-01-21 09:09:22.291262: step: 140/529, loss: 0.049643948674201965 2023-01-21 09:09:23.385544: step: 144/529, loss: 0.49346303939819336 2023-01-21 09:09:24.457096: step: 148/529, loss: 0.9318476915359497 2023-01-21 09:09:25.575120: step: 152/529, loss: 0.17108678817749023 2023-01-21 09:09:26.657501: step: 156/529, loss: 0.08710145950317383 2023-01-21 09:09:27.795506: step: 160/529, loss: 0.3642563223838806 2023-01-21 09:09:28.890732: step: 164/529, loss: 0.12289419770240784 2023-01-21 09:09:30.018009: step: 168/529, loss: 0.256794273853302 2023-01-21 09:09:31.147772: step: 172/529, loss: 0.10855893790721893 2023-01-21 09:09:32.253555: step: 176/529, loss: 0.33685681223869324 2023-01-21 09:09:33.341660: step: 180/529, loss: 0.0270296111702919 2023-01-21 09:09:34.452323: step: 184/529, loss: 0.030338477343320847 2023-01-21 09:09:35.554774: step: 188/529, loss: 0.13597774505615234 2023-01-21 09:09:36.645530: step: 192/529, loss: 0.11101846396923065 2023-01-21 09:09:37.753869: step: 196/529, loss: 0.20447483658790588 2023-01-21 09:09:38.878192: step: 200/529, loss: 0.053606078028678894 2023-01-21 09:09:39.997552: step: 204/529, loss: 0.10037890076637268 2023-01-21 09:09:41.086784: step: 208/529, loss: 0.2880972921848297 2023-01-21 09:09:42.203432: step: 212/529, loss: 0.7638998627662659 2023-01-21 09:09:43.318604: step: 216/529, loss: 0.9142026305198669 2023-01-21 09:09:44.419182: step: 220/529, loss: 1.1600990295410156 2023-01-21 09:09:45.552702: step: 224/529, loss: 0.7649625539779663 2023-01-21 09:09:46.675560: step: 228/529, loss: 0.06582774966955185 2023-01-21 09:09:47.816082: step: 232/529, loss: 1.1401495933532715 2023-01-21 09:09:48.922137: step: 236/529, loss: 0.05163174122571945 2023-01-21 09:09:50.017689: step: 240/529, loss: 0.2893497943878174 2023-01-21 09:09:51.185841: step: 244/529, loss: 0.7470611333847046 2023-01-21 09:09:52.316769: step: 248/529, loss: 0.12130794674158096 2023-01-21 09:09:53.419482: step: 252/529, loss: 0.13695335388183594 2023-01-21 09:09:54.545646: step: 256/529, loss: 0.09405794739723206 2023-01-21 09:09:55.669438: step: 260/529, loss: 1.3281975984573364 2023-01-21 09:09:56.827211: step: 264/529, loss: 0.3186616897583008 2023-01-21 09:09:57.932415: step: 268/529, loss: 0.13274574279785156 2023-01-21 09:09:59.024713: step: 272/529, loss: 0.18047982454299927 2023-01-21 09:10:00.143116: step: 276/529, loss: 0.7832397818565369 2023-01-21 09:10:01.275195: step: 280/529, loss: 0.28849613666534424 2023-01-21 09:10:02.399980: step: 284/529, loss: 0.1275920867919922 2023-01-21 09:10:03.530906: step: 288/529, loss: 0.18065795302391052 2023-01-21 09:10:04.699659: step: 292/529, loss: 0.5579242706298828 2023-01-21 09:10:05.803612: step: 296/529, loss: 0.09338913857936859 2023-01-21 09:10:06.918982: step: 300/529, loss: 0.17106589674949646 2023-01-21 09:10:08.026626: step: 304/529, loss: 0.14436836540699005 2023-01-21 09:10:09.120279: step: 308/529, loss: 0.14182138442993164 2023-01-21 09:10:10.254284: step: 312/529, loss: 0.19877047836780548 2023-01-21 09:10:11.369730: step: 316/529, loss: 6.462676048278809 2023-01-21 09:10:12.526733: step: 320/529, loss: 0.18959026038646698 2023-01-21 09:10:13.639284: step: 324/529, loss: 0.4373766779899597 2023-01-21 09:10:14.788714: step: 328/529, loss: 0.25225162506103516 2023-01-21 09:10:15.926581: step: 332/529, loss: 0.1100100576877594 2023-01-21 09:10:17.050730: step: 336/529, loss: 0.1648460328578949 2023-01-21 09:10:18.163240: step: 340/529, loss: 0.04197349399328232 2023-01-21 09:10:19.261049: step: 344/529, loss: 0.18330413103103638 2023-01-21 09:10:20.410877: step: 348/529, loss: 0.4961535334587097 2023-01-21 09:10:21.554100: step: 352/529, loss: 0.0831550657749176 2023-01-21 09:10:22.662933: step: 356/529, loss: 0.7677115797996521 2023-01-21 09:10:23.752334: step: 360/529, loss: 0.06961259990930557 2023-01-21 09:10:24.878781: step: 364/529, loss: 0.0457647331058979 2023-01-21 09:10:25.983793: step: 368/529, loss: 0.05683698505163193 2023-01-21 09:10:27.126983: step: 372/529, loss: 0.14422345161437988 2023-01-21 09:10:28.235707: step: 376/529, loss: 0.2522476315498352 2023-01-21 09:10:29.351601: step: 380/529, loss: 1.3967328071594238 2023-01-21 09:10:30.449804: step: 384/529, loss: 0.15042105317115784 2023-01-21 09:10:31.575060: step: 388/529, loss: 1.2744789123535156 2023-01-21 09:10:32.675840: step: 392/529, loss: 0.05517597496509552 2023-01-21 09:10:33.799485: step: 396/529, loss: 0.1627863049507141 2023-01-21 09:10:34.914223: step: 400/529, loss: 0.23674874007701874 2023-01-21 09:10:36.066129: step: 404/529, loss: 0.4522775113582611 2023-01-21 09:10:37.191242: step: 408/529, loss: 0.07138939201831818 2023-01-21 09:10:38.319487: step: 412/529, loss: 0.18453297019004822 2023-01-21 09:10:39.437857: step: 416/529, loss: 0.10821361839771271 2023-01-21 09:10:40.573662: step: 420/529, loss: 0.07641144096851349 2023-01-21 09:10:41.691233: step: 424/529, loss: 0.839661717414856 2023-01-21 09:10:42.801087: step: 428/529, loss: 0.09891395270824432 2023-01-21 09:10:43.926514: step: 432/529, loss: 0.8882951736450195 2023-01-21 09:10:45.051100: step: 436/529, loss: 0.11361751705408096 2023-01-21 09:10:46.179360: step: 440/529, loss: 0.6129663586616516 2023-01-21 09:10:47.317443: step: 444/529, loss: 0.16441236436367035 2023-01-21 09:10:48.444127: step: 448/529, loss: 0.4480552077293396 2023-01-21 09:10:49.553405: step: 452/529, loss: 0.07353896647691727 2023-01-21 09:10:50.678379: step: 456/529, loss: 0.17931196093559265 2023-01-21 09:10:51.791703: step: 460/529, loss: 0.11137089878320694 2023-01-21 09:10:52.892384: step: 464/529, loss: 0.10110397636890411 2023-01-21 09:10:54.026681: step: 468/529, loss: 0.17838773131370544 2023-01-21 09:10:55.142135: step: 472/529, loss: 0.855185866355896 2023-01-21 09:10:56.254835: step: 476/529, loss: 0.11198330670595169 2023-01-21 09:10:57.414546: step: 480/529, loss: 0.15248270332813263 2023-01-21 09:10:58.549574: step: 484/529, loss: 0.194318488240242 2023-01-21 09:10:59.650898: step: 488/529, loss: 1.2352555990219116 2023-01-21 09:11:00.775512: step: 492/529, loss: 0.1180235892534256 2023-01-21 09:11:01.908516: step: 496/529, loss: 0.31113529205322266 2023-01-21 09:11:03.006036: step: 500/529, loss: 0.09890633076429367 2023-01-21 09:11:04.134237: step: 504/529, loss: 0.20155149698257446 2023-01-21 09:11:05.284869: step: 508/529, loss: 0.6794360876083374 2023-01-21 09:11:06.410339: step: 512/529, loss: 0.3900454044342041 2023-01-21 09:11:07.513937: step: 516/529, loss: 0.03242664411664009 2023-01-21 09:11:08.641222: step: 520/529, loss: 0.8571157455444336 2023-01-21 09:11:09.791843: step: 524/529, loss: 0.08405561000108719 2023-01-21 09:11:10.909692: step: 528/529, loss: 0.3595353066921234 2023-01-21 09:11:12.047750: step: 532/529, loss: 0.1927490234375 2023-01-21 09:11:13.141514: step: 536/529, loss: 0.44470512866973877 2023-01-21 09:11:14.209930: step: 540/529, loss: 0.12757444381713867 2023-01-21 09:11:15.324416: step: 544/529, loss: 0.18529057502746582 2023-01-21 09:11:16.457977: step: 548/529, loss: 0.3882457911968231 2023-01-21 09:11:17.553961: step: 552/529, loss: 0.4076805114746094 2023-01-21 09:11:18.688100: step: 556/529, loss: 0.03375823795795441 2023-01-21 09:11:19.832206: step: 560/529, loss: 0.5030654668807983 2023-01-21 09:11:20.911022: step: 564/529, loss: 0.03556113317608833 2023-01-21 09:11:22.021190: step: 568/529, loss: 0.15654611587524414 2023-01-21 09:11:23.132887: step: 572/529, loss: 0.22579175233840942 2023-01-21 09:11:24.263789: step: 576/529, loss: 0.07832713425159454 2023-01-21 09:11:25.376063: step: 580/529, loss: 0.04010143503546715 2023-01-21 09:11:26.496559: step: 584/529, loss: 0.1238139197230339 2023-01-21 09:11:27.596278: step: 588/529, loss: 0.048009444028139114 2023-01-21 09:11:28.720991: step: 592/529, loss: 0.31844595074653625 2023-01-21 09:11:29.817662: step: 596/529, loss: 0.20688945055007935 2023-01-21 09:11:30.909816: step: 600/529, loss: 0.6314584612846375 2023-01-21 09:11:32.023786: step: 604/529, loss: 0.050681307911872864 2023-01-21 09:11:33.135337: step: 608/529, loss: 0.2556617856025696 2023-01-21 09:11:34.252550: step: 612/529, loss: 0.08270960301160812 2023-01-21 09:11:35.362378: step: 616/529, loss: 0.11264973133802414 2023-01-21 09:11:36.490383: step: 620/529, loss: 0.06644923985004425 2023-01-21 09:11:37.599350: step: 624/529, loss: 0.10791674256324768 2023-01-21 09:11:38.721086: step: 628/529, loss: 0.5957445502281189 2023-01-21 09:11:39.859043: step: 632/529, loss: 0.1686287820339203 2023-01-21 09:11:40.978690: step: 636/529, loss: 0.09545373916625977 2023-01-21 09:11:42.070987: step: 640/529, loss: 1.1102495193481445 2023-01-21 09:11:43.167584: step: 644/529, loss: 0.04795694351196289 2023-01-21 09:11:44.304783: step: 648/529, loss: 0.4312848746776581 2023-01-21 09:11:45.442160: step: 652/529, loss: 0.8475534319877625 2023-01-21 09:11:46.540161: step: 656/529, loss: 0.1673000603914261 2023-01-21 09:11:47.670560: step: 660/529, loss: 0.07831769436597824 2023-01-21 09:11:48.783836: step: 664/529, loss: 0.30501502752304077 2023-01-21 09:11:49.906204: step: 668/529, loss: 0.11304493248462677 2023-01-21 09:11:51.039021: step: 672/529, loss: 0.3964780867099762 2023-01-21 09:11:52.179666: step: 676/529, loss: 0.5287326574325562 2023-01-21 09:11:53.274552: step: 680/529, loss: 0.5996501445770264 2023-01-21 09:11:54.364693: step: 684/529, loss: 0.6651312112808228 2023-01-21 09:11:55.491582: step: 688/529, loss: 0.3873750567436218 2023-01-21 09:11:56.586253: step: 692/529, loss: 5.405714988708496 2023-01-21 09:11:57.687638: step: 696/529, loss: 0.0684031993150711 2023-01-21 09:11:58.813029: step: 700/529, loss: 0.03897981718182564 2023-01-21 09:11:59.951291: step: 704/529, loss: 0.24262170493602753 2023-01-21 09:12:01.056990: step: 708/529, loss: 0.26134711503982544 2023-01-21 09:12:02.170326: step: 712/529, loss: 0.3341098725795746 2023-01-21 09:12:03.276171: step: 716/529, loss: 0.013199329376220703 2023-01-21 09:12:04.372461: step: 720/529, loss: 5.43837833404541 2023-01-21 09:12:05.505687: step: 724/529, loss: 0.5513275861740112 2023-01-21 09:12:06.609682: step: 728/529, loss: 0.0895567387342453 2023-01-21 09:12:07.721822: step: 732/529, loss: 0.6694862246513367 2023-01-21 09:12:08.828443: step: 736/529, loss: 0.11934996396303177 2023-01-21 09:12:09.940937: step: 740/529, loss: 0.252363383769989 2023-01-21 09:12:11.045569: step: 744/529, loss: 0.064427949488163 2023-01-21 09:12:12.205935: step: 748/529, loss: 0.2742616534233093 2023-01-21 09:12:13.310926: step: 752/529, loss: 0.05188789591193199 2023-01-21 09:12:14.422575: step: 756/529, loss: 0.1607518196105957 2023-01-21 09:12:15.515671: step: 760/529, loss: 0.4873901307582855 2023-01-21 09:12:16.639878: step: 764/529, loss: 0.9535151720046997 2023-01-21 09:12:17.784897: step: 768/529, loss: 0.07002559304237366 2023-01-21 09:12:18.885980: step: 772/529, loss: 0.09257479012012482 2023-01-21 09:12:19.979584: step: 776/529, loss: 0.04662149026989937 2023-01-21 09:12:21.110077: step: 780/529, loss: 0.7043101191520691 2023-01-21 09:12:22.254941: step: 784/529, loss: 0.889208972454071 2023-01-21 09:12:23.364767: step: 788/529, loss: 0.1505270004272461 2023-01-21 09:12:24.506638: step: 792/529, loss: 0.1305987387895584 2023-01-21 09:12:25.632070: step: 796/529, loss: 0.6375781893730164 2023-01-21 09:12:26.742538: step: 800/529, loss: 0.07734694331884384 2023-01-21 09:12:27.831510: step: 804/529, loss: 0.08246222138404846 2023-01-21 09:12:28.920240: step: 808/529, loss: 0.11478619277477264 2023-01-21 09:12:30.089258: step: 812/529, loss: 1.2779672145843506 2023-01-21 09:12:31.218557: step: 816/529, loss: 0.12480269372463226 2023-01-21 09:12:32.318196: step: 820/529, loss: 0.037714291363954544 2023-01-21 09:12:33.446062: step: 824/529, loss: 0.45819342136383057 2023-01-21 09:12:34.570002: step: 828/529, loss: 0.2332388013601303 2023-01-21 09:12:35.716266: step: 832/529, loss: 0.10851164162158966 2023-01-21 09:12:36.842366: step: 836/529, loss: 0.07501611858606339 2023-01-21 09:12:37.968318: step: 840/529, loss: 0.05262966454029083 2023-01-21 09:12:39.059517: step: 844/529, loss: 0.10124874114990234 2023-01-21 09:12:40.192424: step: 848/529, loss: 0.9039085507392883 2023-01-21 09:12:41.337709: step: 852/529, loss: 0.05500993877649307 2023-01-21 09:12:42.506630: step: 856/529, loss: 0.17180556058883667 2023-01-21 09:12:43.596867: step: 860/529, loss: 0.09094668179750443 2023-01-21 09:12:44.692840: step: 864/529, loss: 0.06945426762104034 2023-01-21 09:12:45.793231: step: 868/529, loss: 0.48413917422294617 2023-01-21 09:12:46.917022: step: 872/529, loss: 0.1979755461215973 2023-01-21 09:12:48.026298: step: 876/529, loss: 0.5440086126327515 2023-01-21 09:12:49.115506: step: 880/529, loss: 0.12148857116699219 2023-01-21 09:12:50.254608: step: 884/529, loss: 0.0992768332362175 2023-01-21 09:12:51.360363: step: 888/529, loss: 0.3339608311653137 2023-01-21 09:12:52.485035: step: 892/529, loss: 0.15464048087596893 2023-01-21 09:12:53.613083: step: 896/529, loss: 0.3399643003940582 2023-01-21 09:12:54.728643: step: 900/529, loss: 0.06553025543689728 2023-01-21 09:12:55.838198: step: 904/529, loss: 0.21869993209838867 2023-01-21 09:12:56.960927: step: 908/529, loss: 0.12087535858154297 2023-01-21 09:12:58.056697: step: 912/529, loss: 0.3500957489013672 2023-01-21 09:12:59.157795: step: 916/529, loss: 0.24492283165454865 2023-01-21 09:13:00.272225: step: 920/529, loss: 0.10903721302747726 2023-01-21 09:13:01.397860: step: 924/529, loss: 0.37089115381240845 2023-01-21 09:13:02.525890: step: 928/529, loss: 0.10839272290468216 2023-01-21 09:13:03.662505: step: 932/529, loss: 0.09038126468658447 2023-01-21 09:13:04.796844: step: 936/529, loss: 0.4788447320461273 2023-01-21 09:13:05.920781: step: 940/529, loss: 0.9450740814208984 2023-01-21 09:13:07.035113: step: 944/529, loss: 0.028471995145082474 2023-01-21 09:13:08.117138: step: 948/529, loss: 0.5146323442459106 2023-01-21 09:13:09.208531: step: 952/529, loss: 0.1454252302646637 2023-01-21 09:13:10.338559: step: 956/529, loss: 1.1756420135498047 2023-01-21 09:13:11.479240: step: 960/529, loss: 0.3385176658630371 2023-01-21 09:13:12.601784: step: 964/529, loss: 1.9688810110092163 2023-01-21 09:13:13.720199: step: 968/529, loss: 0.33385169506073 2023-01-21 09:13:14.833045: step: 972/529, loss: 0.07641222327947617 2023-01-21 09:13:15.959719: step: 976/529, loss: 0.1423380821943283 2023-01-21 09:13:17.060698: step: 980/529, loss: 0.18539465963840485 2023-01-21 09:13:18.189217: step: 984/529, loss: 0.039935946464538574 2023-01-21 09:13:19.298318: step: 988/529, loss: 0.11682549118995667 2023-01-21 09:13:20.428642: step: 992/529, loss: 5.441535949707031 2023-01-21 09:13:21.548535: step: 996/529, loss: 0.39364928007125854 2023-01-21 09:13:22.658618: step: 1000/529, loss: 0.09668254852294922 2023-01-21 09:13:23.758252: step: 1004/529, loss: 0.20870447158813477 2023-01-21 09:13:24.897831: step: 1008/529, loss: 0.8664548397064209 2023-01-21 09:13:26.028250: step: 1012/529, loss: 0.025395464152097702 2023-01-21 09:13:27.173530: step: 1016/529, loss: 0.14459332823753357 2023-01-21 09:13:28.292454: step: 1020/529, loss: 0.14206562936306 2023-01-21 09:13:29.389017: step: 1024/529, loss: 0.4411858916282654 2023-01-21 09:13:30.498885: step: 1028/529, loss: 0.09147129207849503 2023-01-21 09:13:31.607949: step: 1032/529, loss: 0.13718290627002716 2023-01-21 09:13:32.735074: step: 1036/529, loss: 1.230292797088623 2023-01-21 09:13:33.837306: step: 1040/529, loss: 0.1264922171831131 2023-01-21 09:13:34.927864: step: 1044/529, loss: 0.742927074432373 2023-01-21 09:13:36.043753: step: 1048/529, loss: 0.3204580545425415 2023-01-21 09:13:37.184481: step: 1052/529, loss: 0.4529357850551605 2023-01-21 09:13:38.303779: step: 1056/529, loss: 0.17225661873817444 2023-01-21 09:13:39.430082: step: 1060/529, loss: 0.1273764669895172 2023-01-21 09:13:40.503303: step: 1064/529, loss: 0.22609463334083557 2023-01-21 09:13:41.626469: step: 1068/529, loss: 0.08944205939769745 2023-01-21 09:13:42.765593: step: 1072/529, loss: 0.19543838500976562 2023-01-21 09:13:43.865763: step: 1076/529, loss: 0.13746052980422974 2023-01-21 09:13:44.965396: step: 1080/529, loss: 1.2111434936523438 2023-01-21 09:13:46.052549: step: 1084/529, loss: 0.9062343239784241 2023-01-21 09:13:47.165531: step: 1088/529, loss: 0.7073706388473511 2023-01-21 09:13:48.279616: step: 1092/529, loss: 0.07628098130226135 2023-01-21 09:13:49.384714: step: 1096/529, loss: 0.14415112137794495 2023-01-21 09:13:50.510397: step: 1100/529, loss: 0.17419585585594177 2023-01-21 09:13:51.608199: step: 1104/529, loss: 0.643078088760376 2023-01-21 09:13:52.731796: step: 1108/529, loss: 0.07637377083301544 2023-01-21 09:13:53.868639: step: 1112/529, loss: 0.3820033073425293 2023-01-21 09:13:54.986136: step: 1116/529, loss: 0.1541009396314621 2023-01-21 09:13:56.107258: step: 1120/529, loss: 0.18002644181251526 2023-01-21 09:13:57.208608: step: 1124/529, loss: 0.2700064778327942 2023-01-21 09:13:58.305072: step: 1128/529, loss: 0.13846750557422638 2023-01-21 09:13:59.402624: step: 1132/529, loss: 0.019024468958377838 2023-01-21 09:14:00.510520: step: 1136/529, loss: 0.1351548284292221 2023-01-21 09:14:01.630000: step: 1140/529, loss: 0.16659009456634521 2023-01-21 09:14:02.725182: step: 1144/529, loss: 0.66960608959198 2023-01-21 09:14:03.856532: step: 1148/529, loss: 0.3959817886352539 2023-01-21 09:14:04.953768: step: 1152/529, loss: 0.6599615216255188 2023-01-21 09:14:06.064717: step: 1156/529, loss: 0.24681758880615234 2023-01-21 09:14:07.198511: step: 1160/529, loss: 0.6472986936569214 2023-01-21 09:14:08.296459: step: 1164/529, loss: 0.4726060926914215 2023-01-21 09:14:09.400277: step: 1168/529, loss: 0.13578566908836365 2023-01-21 09:14:10.530229: step: 1172/529, loss: 0.384170264005661 2023-01-21 09:14:11.649100: step: 1176/529, loss: 0.05696434900164604 2023-01-21 09:14:12.754655: step: 1180/529, loss: 0.2683303952217102 2023-01-21 09:14:13.873480: step: 1184/529, loss: 1.3950036764144897 2023-01-21 09:14:15.008788: step: 1188/529, loss: 0.07179585099220276 2023-01-21 09:14:16.129589: step: 1192/529, loss: 0.0888860747218132 2023-01-21 09:14:17.263618: step: 1196/529, loss: 0.08678951859474182 2023-01-21 09:14:18.378560: step: 1200/529, loss: 1.2237968444824219 2023-01-21 09:14:19.510571: step: 1204/529, loss: 0.19454210996627808 2023-01-21 09:14:20.625786: step: 1208/529, loss: 0.06480474770069122 2023-01-21 09:14:21.736189: step: 1212/529, loss: 0.06839413940906525 2023-01-21 09:14:22.846696: step: 1216/529, loss: 3.6366331577301025 2023-01-21 09:14:23.969642: step: 1220/529, loss: 0.7353880405426025 2023-01-21 09:14:25.099005: step: 1224/529, loss: 0.1569223403930664 2023-01-21 09:14:26.204626: step: 1228/529, loss: 0.1309196949005127 2023-01-21 09:14:27.334772: step: 1232/529, loss: 0.945237934589386 2023-01-21 09:14:28.461991: step: 1236/529, loss: 0.13430005311965942 2023-01-21 09:14:29.555041: step: 1240/529, loss: 0.6541158556938171 2023-01-21 09:14:30.672327: step: 1244/529, loss: 0.17606186866760254 2023-01-21 09:14:31.793682: step: 1248/529, loss: 0.552053689956665 2023-01-21 09:14:32.890580: step: 1252/529, loss: 0.29235193133354187 2023-01-21 09:14:34.030694: step: 1256/529, loss: 0.440141499042511 2023-01-21 09:14:35.150698: step: 1260/529, loss: 0.3931196331977844 2023-01-21 09:14:36.254609: step: 1264/529, loss: 0.6595352292060852 2023-01-21 09:14:37.386646: step: 1268/529, loss: 0.11719191074371338 2023-01-21 09:14:38.461616: step: 1272/529, loss: 0.3888036012649536 2023-01-21 09:14:39.585554: step: 1276/529, loss: 2.185023069381714 2023-01-21 09:14:40.695614: step: 1280/529, loss: 0.29192036390304565 2023-01-21 09:14:41.839621: step: 1284/529, loss: 0.4574712812900543 2023-01-21 09:14:42.951054: step: 1288/529, loss: 0.16599641740322113 2023-01-21 09:14:44.042006: step: 1292/529, loss: 0.7036542892456055 2023-01-21 09:14:45.163039: step: 1296/529, loss: 0.16504187881946564 2023-01-21 09:14:46.303027: step: 1300/529, loss: 0.48771950602531433 2023-01-21 09:14:47.439608: step: 1304/529, loss: 0.14329034090042114 2023-01-21 09:14:48.571965: step: 1308/529, loss: 0.16670912504196167 2023-01-21 09:14:49.688498: step: 1312/529, loss: 0.17818352580070496 2023-01-21 09:14:50.809901: step: 1316/529, loss: 0.4472460150718689 2023-01-21 09:14:51.913133: step: 1320/529, loss: 0.25523626804351807 2023-01-21 09:14:53.023607: step: 1324/529, loss: 0.8591079711914062 2023-01-21 09:14:54.175551: step: 1328/529, loss: 0.538292646408081 2023-01-21 09:14:55.294277: step: 1332/529, loss: 5.669210433959961 2023-01-21 09:14:56.416626: step: 1336/529, loss: 0.4780655801296234 2023-01-21 09:14:57.521191: step: 1340/529, loss: 0.4573246240615845 2023-01-21 09:14:58.618884: step: 1344/529, loss: 0.1562003642320633 2023-01-21 09:14:59.739842: step: 1348/529, loss: 0.08025407791137695 2023-01-21 09:15:00.861873: step: 1352/529, loss: 0.268251895904541 2023-01-21 09:15:02.000640: step: 1356/529, loss: 0.6470426321029663 2023-01-21 09:15:03.114809: step: 1360/529, loss: 0.12363128364086151 2023-01-21 09:15:04.238651: step: 1364/529, loss: 0.1786234974861145 2023-01-21 09:15:05.351197: step: 1368/529, loss: 0.49543020129203796 2023-01-21 09:15:06.436373: step: 1372/529, loss: 0.03644266352057457 2023-01-21 09:15:07.570584: step: 1376/529, loss: 0.38713881373405457 2023-01-21 09:15:08.701316: step: 1380/529, loss: 0.27469301223754883 2023-01-21 09:15:09.793078: step: 1384/529, loss: 0.12839722633361816 2023-01-21 09:15:10.896323: step: 1388/529, loss: 1.3029556274414062 2023-01-21 09:15:12.013031: step: 1392/529, loss: 0.06806459277868271 2023-01-21 09:15:13.116697: step: 1396/529, loss: 0.11641321331262589 2023-01-21 09:15:14.234727: step: 1400/529, loss: 0.07987819612026215 2023-01-21 09:15:15.335761: step: 1404/529, loss: 0.11395521461963654 2023-01-21 09:15:16.453355: step: 1408/529, loss: 0.47873154282569885 2023-01-21 09:15:17.571854: step: 1412/529, loss: 0.21033935248851776 2023-01-21 09:15:18.688461: step: 1416/529, loss: 0.17056827247142792 2023-01-21 09:15:19.791859: step: 1420/529, loss: 0.10613755881786346 2023-01-21 09:15:20.917396: step: 1424/529, loss: 0.15171536803245544 2023-01-21 09:15:22.007463: step: 1428/529, loss: 0.22751502692699432 2023-01-21 09:15:23.135255: step: 1432/529, loss: 0.06238298490643501 2023-01-21 09:15:24.246486: step: 1436/529, loss: 0.6377886533737183 2023-01-21 09:15:25.349697: step: 1440/529, loss: 0.1844189167022705 2023-01-21 09:15:26.443905: step: 1444/529, loss: 0.3766494691371918 2023-01-21 09:15:27.559592: step: 1448/529, loss: 0.16443492472171783 2023-01-21 09:15:28.701481: step: 1452/529, loss: 0.28271445631980896 2023-01-21 09:15:29.806401: step: 1456/529, loss: 0.1826373189687729 2023-01-21 09:15:30.933202: step: 1460/529, loss: 0.03917064517736435 2023-01-21 09:15:32.095284: step: 1464/529, loss: 0.11006021499633789 2023-01-21 09:15:33.232395: step: 1468/529, loss: 0.10037532448768616 2023-01-21 09:15:34.336132: step: 1472/529, loss: 0.7557169795036316 2023-01-21 09:15:35.450132: step: 1476/529, loss: 0.2257649004459381 2023-01-21 09:15:36.590787: step: 1480/529, loss: 0.10338573902845383 2023-01-21 09:15:37.679395: step: 1484/529, loss: 0.3532348573207855 2023-01-21 09:15:38.781285: step: 1488/529, loss: 0.05444040149450302 2023-01-21 09:15:39.911214: step: 1492/529, loss: 0.4606480598449707 2023-01-21 09:15:41.029901: step: 1496/529, loss: 0.11883287131786346 2023-01-21 09:15:42.166229: step: 1500/529, loss: 0.7818244099617004 2023-01-21 09:15:43.243492: step: 1504/529, loss: 0.6513429284095764 2023-01-21 09:15:44.364031: step: 1508/529, loss: 0.13442793488502502 2023-01-21 09:15:45.480467: step: 1512/529, loss: 0.889224648475647 2023-01-21 09:15:46.588192: step: 1516/529, loss: 0.08269243687391281 2023-01-21 09:15:47.689286: step: 1520/529, loss: 0.1586645245552063 2023-01-21 09:15:48.815727: step: 1524/529, loss: 0.2881953716278076 2023-01-21 09:15:49.938159: step: 1528/529, loss: 0.32926616072654724 2023-01-21 09:15:51.067916: step: 1532/529, loss: 0.09252652525901794 2023-01-21 09:15:52.195543: step: 1536/529, loss: 0.5710971355438232 2023-01-21 09:15:53.305355: step: 1540/529, loss: 0.04394941404461861 2023-01-21 09:15:54.436886: step: 1544/529, loss: 0.5223848819732666 2023-01-21 09:15:55.539589: step: 1548/529, loss: 1.9456416368484497 2023-01-21 09:15:56.651374: step: 1552/529, loss: 0.15376925468444824 2023-01-21 09:15:57.795856: step: 1556/529, loss: 0.6080523133277893 2023-01-21 09:15:58.925789: step: 1560/529, loss: 0.12001819163560867 2023-01-21 09:16:00.040184: step: 1564/529, loss: 0.051761530339717865 2023-01-21 09:16:01.156714: step: 1568/529, loss: 0.1882733553647995 2023-01-21 09:16:02.249870: step: 1572/529, loss: 0.5648950338363647 2023-01-21 09:16:03.371490: step: 1576/529, loss: 0.22642555832862854 2023-01-21 09:16:04.493420: step: 1580/529, loss: 0.046761300414800644 2023-01-21 09:16:05.598765: step: 1584/529, loss: 0.23881873488426208 2023-01-21 09:16:06.692966: step: 1588/529, loss: 0.4894735813140869 2023-01-21 09:16:07.800267: step: 1592/529, loss: 0.18804016709327698 2023-01-21 09:16:08.898076: step: 1596/529, loss: 0.6195987462997437 2023-01-21 09:16:10.044506: step: 1600/529, loss: 0.22057685256004333 2023-01-21 09:16:11.164533: step: 1604/529, loss: 0.08101234585046768 2023-01-21 09:16:12.290646: step: 1608/529, loss: 0.626783013343811 2023-01-21 09:16:13.414632: step: 1612/529, loss: 0.03388338163495064 2023-01-21 09:16:14.572692: step: 1616/529, loss: 0.2831869125366211 2023-01-21 09:16:15.678458: step: 1620/529, loss: 0.07124157249927521 2023-01-21 09:16:16.790571: step: 1624/529, loss: 0.6699535250663757 2023-01-21 09:16:17.918742: step: 1628/529, loss: 0.022092128172516823 2023-01-21 09:16:19.014768: step: 1632/529, loss: 0.09064579010009766 2023-01-21 09:16:20.163639: step: 1636/529, loss: 0.0951623022556305 2023-01-21 09:16:21.273345: step: 1640/529, loss: 0.2127547264099121 2023-01-21 09:16:22.394396: step: 1644/529, loss: 0.16088123619556427 2023-01-21 09:16:23.513273: step: 1648/529, loss: 0.6161696910858154 2023-01-21 09:16:24.616328: step: 1652/529, loss: 0.12120532989501953 2023-01-21 09:16:25.752519: step: 1656/529, loss: 1.0246459245681763 2023-01-21 09:16:26.873128: step: 1660/529, loss: 0.05651235580444336 2023-01-21 09:16:27.985666: step: 1664/529, loss: 0.9207844734191895 2023-01-21 09:16:29.075916: step: 1668/529, loss: 0.1448526382446289 2023-01-21 09:16:30.208239: step: 1672/529, loss: 0.08858537673950195 2023-01-21 09:16:31.353522: step: 1676/529, loss: 0.386136919260025 2023-01-21 09:16:32.444545: step: 1680/529, loss: 0.18111905455589294 2023-01-21 09:16:33.546390: step: 1684/529, loss: 0.17898598313331604 2023-01-21 09:16:34.673008: step: 1688/529, loss: 0.21989670395851135 2023-01-21 09:16:35.791539: step: 1692/529, loss: 0.031917523592710495 2023-01-21 09:16:36.878818: step: 1696/529, loss: 0.061630867421627045 2023-01-21 09:16:38.000504: step: 1700/529, loss: 0.17774944007396698 2023-01-21 09:16:39.117126: step: 1704/529, loss: 0.13025054335594177 2023-01-21 09:16:40.218563: step: 1708/529, loss: 0.10247235000133514 2023-01-21 09:16:41.342974: step: 1712/529, loss: 0.21328100562095642 2023-01-21 09:16:42.459720: step: 1716/529, loss: 0.11350855231285095 2023-01-21 09:16:43.594256: step: 1720/529, loss: 0.18481570482254028 2023-01-21 09:16:44.720795: step: 1724/529, loss: 0.7798261642456055 2023-01-21 09:16:45.827497: step: 1728/529, loss: 0.05474729463458061 2023-01-21 09:16:46.919082: step: 1732/529, loss: 0.2823188304901123 2023-01-21 09:16:48.025224: step: 1736/529, loss: 0.5540623664855957 2023-01-21 09:16:49.161558: step: 1740/529, loss: 0.34217938780784607 2023-01-21 09:16:50.266712: step: 1744/529, loss: 4.49744176864624 2023-01-21 09:16:51.378483: step: 1748/529, loss: 0.1192934513092041 2023-01-21 09:16:52.487172: step: 1752/529, loss: 0.1877593994140625 2023-01-21 09:16:53.617941: step: 1756/529, loss: 0.17658326029777527 2023-01-21 09:16:54.713540: step: 1760/529, loss: 0.2475995123386383 2023-01-21 09:16:55.802978: step: 1764/529, loss: 0.41121023893356323 2023-01-21 09:16:56.906112: step: 1768/529, loss: 2.9980592727661133 2023-01-21 09:16:57.995690: step: 1772/529, loss: 0.286930650472641 2023-01-21 09:16:59.100818: step: 1776/529, loss: 0.16938361525535583 2023-01-21 09:17:00.213194: step: 1780/529, loss: 0.05066043138504028 2023-01-21 09:17:01.327128: step: 1784/529, loss: 0.33080074191093445 2023-01-21 09:17:02.429346: step: 1788/529, loss: 0.23414728045463562 2023-01-21 09:17:03.527348: step: 1792/529, loss: 0.2902383804321289 2023-01-21 09:17:04.639634: step: 1796/529, loss: 0.2578304409980774 2023-01-21 09:17:05.751395: step: 1800/529, loss: 0.41495877504348755 2023-01-21 09:17:06.898515: step: 1804/529, loss: 0.24060077965259552 2023-01-21 09:17:08.028368: step: 1808/529, loss: 0.147386834025383 2023-01-21 09:17:09.160184: step: 1812/529, loss: 0.054194070398807526 2023-01-21 09:17:10.279327: step: 1816/529, loss: 0.054935719817876816 2023-01-21 09:17:11.391469: step: 1820/529, loss: 0.45009851455688477 2023-01-21 09:17:12.530328: step: 1824/529, loss: 0.186998188495636 2023-01-21 09:17:13.644094: step: 1828/529, loss: 0.48846301436424255 2023-01-21 09:17:14.734244: step: 1832/529, loss: 0.11444559693336487 2023-01-21 09:17:15.849176: step: 1836/529, loss: 0.1369856894016266 2023-01-21 09:17:16.971960: step: 1840/529, loss: 0.22920675575733185 2023-01-21 09:17:18.081119: step: 1844/529, loss: 0.12228593975305557 2023-01-21 09:17:19.179312: step: 1848/529, loss: 0.12914076447486877 2023-01-21 09:17:20.282086: step: 1852/529, loss: 0.06727103888988495 2023-01-21 09:17:21.399192: step: 1856/529, loss: 0.25102242827415466 2023-01-21 09:17:22.527071: step: 1860/529, loss: 0.07208700478076935 2023-01-21 09:17:23.626403: step: 1864/529, loss: 0.3147136867046356 2023-01-21 09:17:24.727007: step: 1868/529, loss: 0.11604194343090057 2023-01-21 09:17:25.835000: step: 1872/529, loss: 0.16825714707374573 2023-01-21 09:17:26.981058: step: 1876/529, loss: 0.5729438066482544 2023-01-21 09:17:28.086547: step: 1880/529, loss: 0.02032341994345188 2023-01-21 09:17:29.186072: step: 1884/529, loss: 1.2662216424942017 2023-01-21 09:17:30.288692: step: 1888/529, loss: 0.09326568245887756 2023-01-21 09:17:31.411650: step: 1892/529, loss: 0.16760015487670898 2023-01-21 09:17:32.514717: step: 1896/529, loss: 0.07247710227966309 2023-01-21 09:17:33.646087: step: 1900/529, loss: 0.06102919578552246 2023-01-21 09:17:34.795677: step: 1904/529, loss: 0.1514964997768402 2023-01-21 09:17:35.888547: step: 1908/529, loss: 0.04268374666571617 2023-01-21 09:17:37.014608: step: 1912/529, loss: 0.09840288758277893 2023-01-21 09:17:38.119075: step: 1916/529, loss: 0.3508193790912628 2023-01-21 09:17:39.224861: step: 1920/529, loss: 0.06758122891187668 2023-01-21 09:17:40.338027: step: 1924/529, loss: 0.21594581007957458 2023-01-21 09:17:41.451526: step: 1928/529, loss: 0.14001736044883728 2023-01-21 09:17:42.560123: step: 1932/529, loss: 0.06528887897729874 2023-01-21 09:17:43.680933: step: 1936/529, loss: 0.22349201142787933 2023-01-21 09:17:44.771168: step: 1940/529, loss: 0.10970593243837357 2023-01-21 09:17:45.942124: step: 1944/529, loss: 0.7762815952301025 2023-01-21 09:17:47.039702: step: 1948/529, loss: 0.15965993702411652 2023-01-21 09:17:48.139454: step: 1952/529, loss: 0.3056747317314148 2023-01-21 09:17:49.254838: step: 1956/529, loss: 0.5440365672111511 2023-01-21 09:17:50.384262: step: 1960/529, loss: 0.11164084076881409 2023-01-21 09:17:51.489957: step: 1964/529, loss: 0.2548164427280426 2023-01-21 09:17:52.627604: step: 1968/529, loss: 0.14545422792434692 2023-01-21 09:17:53.731905: step: 1972/529, loss: 4.151213645935059 2023-01-21 09:17:54.837959: step: 1976/529, loss: 0.49384671449661255 2023-01-21 09:17:55.969872: step: 1980/529, loss: 0.13706551492214203 2023-01-21 09:17:57.079495: step: 1984/529, loss: 0.2585916519165039 2023-01-21 09:17:58.176218: step: 1988/529, loss: 0.12723436951637268 2023-01-21 09:17:59.281856: step: 1992/529, loss: 0.19709739089012146 2023-01-21 09:18:00.384389: step: 1996/529, loss: 0.4232083559036255 2023-01-21 09:18:01.501023: step: 2000/529, loss: 0.20833492279052734 2023-01-21 09:18:02.660472: step: 2004/529, loss: 0.6804494261741638 2023-01-21 09:18:03.794661: step: 2008/529, loss: 0.30604103207588196 2023-01-21 09:18:04.896692: step: 2012/529, loss: 4.064002990722656 2023-01-21 09:18:06.042175: step: 2016/529, loss: 0.3089672327041626 2023-01-21 09:18:07.147313: step: 2020/529, loss: 0.07604637742042542 2023-01-21 09:18:08.277894: step: 2024/529, loss: 0.1061973124742508 2023-01-21 09:18:09.406914: step: 2028/529, loss: 0.22420406341552734 2023-01-21 09:18:10.535600: step: 2032/529, loss: 0.2828930914402008 2023-01-21 09:18:11.696055: step: 2036/529, loss: 0.055702365934848785 2023-01-21 09:18:12.800724: step: 2040/529, loss: 0.1882980465888977 2023-01-21 09:18:13.926967: step: 2044/529, loss: 0.3381475806236267 2023-01-21 09:18:15.049526: step: 2048/529, loss: 0.34776657819747925 2023-01-21 09:18:16.172737: step: 2052/529, loss: 0.16548840701580048 2023-01-21 09:18:17.283980: step: 2056/529, loss: 0.27984774112701416 2023-01-21 09:18:18.426058: step: 2060/529, loss: 0.21640339493751526 2023-01-21 09:18:19.540021: step: 2064/529, loss: 0.03279542922973633 2023-01-21 09:18:20.667546: step: 2068/529, loss: 0.4831317067146301 2023-01-21 09:18:21.795912: step: 2072/529, loss: 0.13899001479148865 2023-01-21 09:18:22.912973: step: 2076/529, loss: 0.08098659664392471 2023-01-21 09:18:24.016441: step: 2080/529, loss: 0.23954276740550995 2023-01-21 09:18:25.142577: step: 2084/529, loss: 4.100245952606201 2023-01-21 09:18:26.263388: step: 2088/529, loss: 0.153880313038826 2023-01-21 09:18:27.408006: step: 2092/529, loss: 0.20263110101222992 2023-01-21 09:18:28.522127: step: 2096/529, loss: 0.5009497404098511 2023-01-21 09:18:29.655263: step: 2100/529, loss: 0.07189646363258362 2023-01-21 09:18:30.789188: step: 2104/529, loss: 0.2859005928039551 2023-01-21 09:18:31.951184: step: 2108/529, loss: 0.5420616269111633 2023-01-21 09:18:33.076094: step: 2112/529, loss: 0.10264568030834198 2023-01-21 09:18:34.221688: step: 2116/529, loss: 2.240382194519043 ================================================== Loss: 0.406 -------------------- Dev: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Test: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Chinese: {'event': {'p': 0.5, 'r': 0.8703703703703703, 'f1': 0.6351351351351352}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Korean: {'event': {'p': 0.453125, 'r': 0.4603174603174603, 'f1': 0.4566929133858268}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5572815533980583, 'r': 0.7643142476697736, 'f1': 0.6445816956765862}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Korean: {'event': {'p': 0.5485519591141397, 'r': 0.7268623024830699, 'f1': 0.6252427184466018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Korean: {'event': {'p': 0.7111111111111111, 'r': 0.5079365079365079, 'f1': 0.5925925925925924}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, '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 09:19:23.330892: step: 4/529, loss: 0.07176800072193146 2023-01-21 09:19:24.508799: step: 8/529, loss: 0.48882177472114563 2023-01-21 09:19:25.622721: step: 12/529, loss: 0.08888855576515198 2023-01-21 09:19:26.731421: step: 16/529, loss: 0.12938714027404785 2023-01-21 09:19:27.835447: step: 20/529, loss: 0.46757346391677856 2023-01-21 09:19:28.931062: step: 24/529, loss: 0.34735241532325745 2023-01-21 09:19:30.028780: step: 28/529, loss: 4.39654541015625 2023-01-21 09:19:31.133855: step: 32/529, loss: 0.07400093227624893 2023-01-21 09:19:32.225597: step: 36/529, loss: 0.2866958677768707 2023-01-21 09:19:33.321769: step: 40/529, loss: 0.11702385544776917 2023-01-21 09:19:34.456954: step: 44/529, loss: 0.6336420774459839 2023-01-21 09:19:35.562088: step: 48/529, loss: 0.3022029995918274 2023-01-21 09:19:36.728234: step: 52/529, loss: 0.2538001239299774 2023-01-21 09:19:37.830860: step: 56/529, loss: 0.09459337592124939 2023-01-21 09:19:38.941033: step: 60/529, loss: 0.3749246597290039 2023-01-21 09:19:40.057344: step: 64/529, loss: 0.09169311821460724 2023-01-21 09:19:41.171867: step: 68/529, loss: 0.26982012391090393 2023-01-21 09:19:42.308997: step: 72/529, loss: 2.723605155944824 2023-01-21 09:19:43.461675: step: 76/529, loss: 0.26373690366744995 2023-01-21 09:19:44.581567: step: 80/529, loss: 0.16915054619312286 2023-01-21 09:19:45.696432: step: 84/529, loss: 0.13807296752929688 2023-01-21 09:19:46.819963: step: 88/529, loss: 0.1197231262922287 2023-01-21 09:19:47.938765: step: 92/529, loss: 0.12036142498254776 2023-01-21 09:19:49.041032: step: 96/529, loss: 0.09978790581226349 2023-01-21 09:19:50.180010: step: 100/529, loss: 0.05086259916424751 2023-01-21 09:19:51.255822: step: 104/529, loss: 0.11878304183483124 2023-01-21 09:19:52.357571: step: 108/529, loss: 0.30495673418045044 2023-01-21 09:19:53.469622: step: 112/529, loss: 0.15193353593349457 2023-01-21 09:19:54.565282: step: 116/529, loss: 0.032694898545742035 2023-01-21 09:19:55.655363: step: 120/529, loss: 0.07092676311731339 2023-01-21 09:19:56.771665: step: 124/529, loss: 0.36928021907806396 2023-01-21 09:19:57.883584: step: 128/529, loss: 0.24337291717529297 2023-01-21 09:19:58.987489: step: 132/529, loss: 0.15207695960998535 2023-01-21 09:20:00.125424: step: 136/529, loss: 0.04180867597460747 2023-01-21 09:20:01.251397: step: 140/529, loss: 0.33201128244400024 2023-01-21 09:20:02.346406: step: 144/529, loss: 0.11735763400793076 2023-01-21 09:20:03.441601: step: 148/529, loss: 0.038109827786684036 2023-01-21 09:20:04.556727: step: 152/529, loss: 0.15805473923683167 2023-01-21 09:20:05.676719: step: 156/529, loss: 0.10979817807674408 2023-01-21 09:20:06.802796: step: 160/529, loss: 0.1887872815132141 2023-01-21 09:20:07.927300: step: 164/529, loss: 0.06425094604492188 2023-01-21 09:20:09.033470: step: 168/529, loss: 0.10059285163879395 2023-01-21 09:20:10.176114: step: 172/529, loss: 0.25150051712989807 2023-01-21 09:20:11.286693: step: 176/529, loss: 0.08196024596691132 2023-01-21 09:20:12.422217: step: 180/529, loss: 0.5017951130867004 2023-01-21 09:20:13.535723: step: 184/529, loss: 0.3754850924015045 2023-01-21 09:20:14.649761: step: 188/529, loss: 0.49523216485977173 2023-01-21 09:20:15.750258: step: 192/529, loss: 0.052175380289554596 2023-01-21 09:20:16.860116: step: 196/529, loss: 0.310914546251297 2023-01-21 09:20:17.977050: step: 200/529, loss: 0.11522388458251953 2023-01-21 09:20:19.097295: step: 204/529, loss: 0.048056744039058685 2023-01-21 09:20:20.186714: step: 208/529, loss: 0.06516657024621964 2023-01-21 09:20:21.278277: step: 212/529, loss: 0.1264423429965973 2023-01-21 09:20:22.403200: step: 216/529, loss: 0.08873319625854492 2023-01-21 09:20:23.504564: step: 220/529, loss: 0.06373219192028046 2023-01-21 09:20:24.636459: step: 224/529, loss: 0.1512037217617035 2023-01-21 09:20:25.751708: step: 228/529, loss: 0.20380745828151703 2023-01-21 09:20:26.851501: step: 232/529, loss: 0.08971276879310608 2023-01-21 09:20:27.986990: step: 236/529, loss: 1.130300760269165 2023-01-21 09:20:29.096690: step: 240/529, loss: 0.2372211068868637 2023-01-21 09:20:30.195592: step: 244/529, loss: 0.38963738083839417 2023-01-21 09:20:31.309525: step: 248/529, loss: 0.11961832642555237 2023-01-21 09:20:32.437355: step: 252/529, loss: 0.11456408351659775 2023-01-21 09:20:33.613078: step: 256/529, loss: 0.4340522885322571 2023-01-21 09:20:34.716976: step: 260/529, loss: 0.4952842593193054 2023-01-21 09:20:35.821613: step: 264/529, loss: 1.0062696933746338 2023-01-21 09:20:36.922753: step: 268/529, loss: 0.31698256731033325 2023-01-21 09:20:38.021396: step: 272/529, loss: 0.033536337316036224 2023-01-21 09:20:39.121976: step: 276/529, loss: 0.14295777678489685 2023-01-21 09:20:40.253170: step: 280/529, loss: 0.06467914581298828 2023-01-21 09:20:41.360918: step: 284/529, loss: 0.2953764498233795 2023-01-21 09:20:42.487103: step: 288/529, loss: 0.17570915818214417 2023-01-21 09:20:43.586906: step: 292/529, loss: 0.04101882129907608 2023-01-21 09:20:44.717034: step: 296/529, loss: 0.23953105509281158 2023-01-21 09:20:45.853611: step: 300/529, loss: 0.27949273586273193 2023-01-21 09:20:46.948593: step: 304/529, loss: 0.008367824368178844 2023-01-21 09:20:48.076148: step: 308/529, loss: 0.23857679963111877 2023-01-21 09:20:49.192924: step: 312/529, loss: 0.24934254586696625 2023-01-21 09:20:50.297736: step: 316/529, loss: 0.6869691610336304 2023-01-21 09:20:51.443240: step: 320/529, loss: 0.09876394271850586 2023-01-21 09:20:52.565521: step: 324/529, loss: 0.2421676218509674 2023-01-21 09:20:53.681012: step: 328/529, loss: 0.37847620248794556 2023-01-21 09:20:54.801584: step: 332/529, loss: 0.15747416019439697 2023-01-21 09:20:55.916621: step: 336/529, loss: 0.1132832020521164 2023-01-21 09:20:57.022508: step: 340/529, loss: 0.5885408520698547 2023-01-21 09:20:58.126480: step: 344/529, loss: 0.2205224484205246 2023-01-21 09:20:59.217697: step: 348/529, loss: 0.366557776927948 2023-01-21 09:21:00.339788: step: 352/529, loss: 0.1642545759677887 2023-01-21 09:21:01.460061: step: 356/529, loss: 0.2124018669128418 2023-01-21 09:21:02.595752: step: 360/529, loss: 0.148661807179451 2023-01-21 09:21:03.715985: step: 364/529, loss: 1.1682111024856567 2023-01-21 09:21:04.823704: step: 368/529, loss: 0.12335072457790375 2023-01-21 09:21:05.941475: step: 372/529, loss: 0.025620698928833008 2023-01-21 09:21:07.057059: step: 376/529, loss: 0.05413227155804634 2023-01-21 09:21:08.170434: step: 380/529, loss: 0.16415753960609436 2023-01-21 09:21:09.275730: step: 384/529, loss: 0.18194541335105896 2023-01-21 09:21:10.395218: step: 388/529, loss: 0.43905311822891235 2023-01-21 09:21:11.506518: step: 392/529, loss: 0.0664089173078537 2023-01-21 09:21:12.627741: step: 396/529, loss: 0.5605072975158691 2023-01-21 09:21:13.724734: step: 400/529, loss: 0.4242580533027649 2023-01-21 09:21:14.827369: step: 404/529, loss: 0.246148020029068 2023-01-21 09:21:15.986385: step: 408/529, loss: 0.4521218240261078 2023-01-21 09:21:17.103363: step: 412/529, loss: 0.1906055510044098 2023-01-21 09:21:18.228042: step: 416/529, loss: 0.21712461113929749 2023-01-21 09:21:19.361062: step: 420/529, loss: 1.8412694931030273 2023-01-21 09:21:20.493732: step: 424/529, loss: 1.2602217197418213 2023-01-21 09:21:21.618496: step: 428/529, loss: 0.495075523853302 2023-01-21 09:21:22.744706: step: 432/529, loss: 0.2181301862001419 2023-01-21 09:21:23.870670: step: 436/529, loss: 0.10174522548913956 2023-01-21 09:21:24.976770: step: 440/529, loss: 0.36892640590667725 2023-01-21 09:21:26.115474: step: 444/529, loss: 0.7357617616653442 2023-01-21 09:21:27.292574: step: 448/529, loss: 0.19989554584026337 2023-01-21 09:21:28.429639: step: 452/529, loss: 0.38063403964042664 2023-01-21 09:21:29.544310: step: 456/529, loss: 0.8308342099189758 2023-01-21 09:21:30.662847: step: 460/529, loss: 0.2380828857421875 2023-01-21 09:21:31.794389: step: 464/529, loss: 0.04858918488025665 2023-01-21 09:21:32.933457: step: 468/529, loss: 0.1345197707414627 2023-01-21 09:21:34.107421: step: 472/529, loss: 0.24324941635131836 2023-01-21 09:21:35.215797: step: 476/529, loss: 0.17164115607738495 2023-01-21 09:21:36.364933: step: 480/529, loss: 0.31392115354537964 2023-01-21 09:21:37.517654: step: 484/529, loss: 0.3716011643409729 2023-01-21 09:21:38.613281: step: 488/529, loss: 0.07454583793878555 2023-01-21 09:21:39.733520: step: 492/529, loss: 0.3151795268058777 2023-01-21 09:21:40.806689: step: 496/529, loss: 0.126824289560318 2023-01-21 09:21:41.922447: step: 500/529, loss: 1.211808443069458 2023-01-21 09:21:43.029808: step: 504/529, loss: 0.16941985487937927 2023-01-21 09:21:44.133448: step: 508/529, loss: 0.05509186163544655 2023-01-21 09:21:45.249577: step: 512/529, loss: 0.1899217665195465 2023-01-21 09:21:46.362218: step: 516/529, loss: 0.10406293720006943 2023-01-21 09:21:47.461802: step: 520/529, loss: 0.20110595226287842 2023-01-21 09:21:48.597966: step: 524/529, loss: 0.32373762130737305 2023-01-21 09:21:49.710345: step: 528/529, loss: 0.963670015335083 2023-01-21 09:21:50.850169: step: 532/529, loss: 0.1061980202794075 2023-01-21 09:21:52.005699: step: 536/529, loss: 0.25688838958740234 2023-01-21 09:21:53.111559: step: 540/529, loss: 0.08908119052648544 2023-01-21 09:21:54.244071: step: 544/529, loss: 0.09998014569282532 2023-01-21 09:21:55.351832: step: 548/529, loss: 0.05474138259887695 2023-01-21 09:21:56.468722: step: 552/529, loss: 0.16262155771255493 2023-01-21 09:21:57.570232: step: 556/529, loss: 0.04006953164935112 2023-01-21 09:21:58.668885: step: 560/529, loss: 0.15252971649169922 2023-01-21 09:21:59.807462: step: 564/529, loss: 0.16855831444263458 2023-01-21 09:22:00.928421: step: 568/529, loss: 0.23425331711769104 2023-01-21 09:22:02.043546: step: 572/529, loss: 0.12185021489858627 2023-01-21 09:22:03.186105: step: 576/529, loss: 0.09067702293395996 2023-01-21 09:22:04.312155: step: 580/529, loss: 0.23523512482643127 2023-01-21 09:22:05.427372: step: 584/529, loss: 0.19013480842113495 2023-01-21 09:22:06.561363: step: 588/529, loss: 0.030865196138620377 2023-01-21 09:22:07.694505: step: 592/529, loss: 0.014112567529082298 2023-01-21 09:22:08.806735: step: 596/529, loss: 0.2811559736728668 2023-01-21 09:22:09.939810: step: 600/529, loss: 0.13876524567604065 2023-01-21 09:22:11.054149: step: 604/529, loss: 0.11205069720745087 2023-01-21 09:22:12.171598: step: 608/529, loss: 0.45882484316825867 2023-01-21 09:22:13.292644: step: 612/529, loss: 0.46615585684776306 2023-01-21 09:22:14.433035: step: 616/529, loss: 0.1883348524570465 2023-01-21 09:22:15.548180: step: 620/529, loss: 0.15784187614917755 2023-01-21 09:22:16.679648: step: 624/529, loss: 0.1897747963666916 2023-01-21 09:22:17.819394: step: 628/529, loss: 0.05434370040893555 2023-01-21 09:22:18.960900: step: 632/529, loss: 0.7836352586746216 2023-01-21 09:22:20.073778: step: 636/529, loss: 0.039287567138671875 2023-01-21 09:22:21.229130: step: 640/529, loss: 0.06502513587474823 2023-01-21 09:22:22.378636: step: 644/529, loss: 0.9671511054039001 2023-01-21 09:22:23.521766: step: 648/529, loss: 0.10725164413452148 2023-01-21 09:22:24.649357: step: 652/529, loss: 0.26869791746139526 2023-01-21 09:22:25.811364: step: 656/529, loss: 0.050656892359256744 2023-01-21 09:22:26.966418: step: 660/529, loss: 0.19444340467453003 2023-01-21 09:22:28.069382: step: 664/529, loss: 0.09351330250501633 2023-01-21 09:22:29.168138: step: 668/529, loss: 0.6219432950019836 2023-01-21 09:22:30.305594: step: 672/529, loss: 0.16720303893089294 2023-01-21 09:22:31.418448: step: 676/529, loss: 0.628582239151001 2023-01-21 09:22:32.547885: step: 680/529, loss: 4.476847171783447 2023-01-21 09:22:33.655659: step: 684/529, loss: 0.14616942405700684 2023-01-21 09:22:34.748778: step: 688/529, loss: 0.07479887455701828 2023-01-21 09:22:35.852623: step: 692/529, loss: 0.1100873053073883 2023-01-21 09:22:36.945191: step: 696/529, loss: 0.107672318816185 2023-01-21 09:22:38.046374: step: 700/529, loss: 0.13694247603416443 2023-01-21 09:22:39.144725: step: 704/529, loss: 1.302291750907898 2023-01-21 09:22:40.302009: step: 708/529, loss: 0.23167285323143005 2023-01-21 09:22:41.403290: step: 712/529, loss: 0.38994550704956055 2023-01-21 09:22:42.529788: step: 716/529, loss: 0.003508281661197543 2023-01-21 09:22:43.634464: step: 720/529, loss: 2.912047863006592 2023-01-21 09:22:44.751096: step: 724/529, loss: 0.22274373471736908 2023-01-21 09:22:45.883222: step: 728/529, loss: 0.8981214761734009 2023-01-21 09:22:47.022211: step: 732/529, loss: 0.16637632250785828 2023-01-21 09:22:48.162074: step: 736/529, loss: 0.8429765701293945 2023-01-21 09:22:49.274858: step: 740/529, loss: 0.14332342147827148 2023-01-21 09:22:50.436179: step: 744/529, loss: 0.9117512106895447 2023-01-21 09:22:51.574309: step: 748/529, loss: 0.8392467498779297 2023-01-21 09:22:52.724426: step: 752/529, loss: 0.287754625082016 2023-01-21 09:22:53.843860: step: 756/529, loss: 0.17760629951953888 2023-01-21 09:22:54.944588: step: 760/529, loss: 0.6325515508651733 2023-01-21 09:22:56.036512: step: 764/529, loss: 0.008400964550673962 2023-01-21 09:22:57.160579: step: 768/529, loss: 0.4714752435684204 2023-01-21 09:22:58.311527: step: 772/529, loss: 0.14944672584533691 2023-01-21 09:22:59.464419: step: 776/529, loss: 0.2023017406463623 2023-01-21 09:23:00.586413: step: 780/529, loss: 0.059515856206417084 2023-01-21 09:23:01.692952: step: 784/529, loss: 0.4514063000679016 2023-01-21 09:23:02.822072: step: 788/529, loss: 3.194371461868286 2023-01-21 09:23:03.932990: step: 792/529, loss: 0.10299444198608398 2023-01-21 09:23:05.059479: step: 796/529, loss: 0.1376463919878006 2023-01-21 09:23:06.181680: step: 800/529, loss: 0.11490288376808167 2023-01-21 09:23:07.319908: step: 804/529, loss: 0.0795622318983078 2023-01-21 09:23:08.433993: step: 808/529, loss: 0.3224641978740692 2023-01-21 09:23:09.537703: step: 812/529, loss: 0.21379882097244263 2023-01-21 09:23:10.646667: step: 816/529, loss: 0.35221731662750244 2023-01-21 09:23:11.737042: step: 820/529, loss: 0.03069133684039116 2023-01-21 09:23:12.852081: step: 824/529, loss: 0.11250162124633789 2023-01-21 09:23:13.980926: step: 828/529, loss: 0.14740753173828125 2023-01-21 09:23:15.103095: step: 832/529, loss: 0.2560974359512329 2023-01-21 09:23:16.202658: step: 836/529, loss: 0.5588887929916382 2023-01-21 09:23:17.316225: step: 840/529, loss: 0.8817388415336609 2023-01-21 09:23:18.447517: step: 844/529, loss: 0.28461599349975586 2023-01-21 09:23:19.537045: step: 848/529, loss: 0.3847476840019226 2023-01-21 09:23:20.668163: step: 852/529, loss: 0.03926238790154457 2023-01-21 09:23:21.788679: step: 856/529, loss: 0.9807325601577759 2023-01-21 09:23:22.902613: step: 860/529, loss: 0.07923221588134766 2023-01-21 09:23:24.014350: step: 864/529, loss: 0.09899359196424484 2023-01-21 09:23:25.142541: step: 868/529, loss: 0.16891708970069885 2023-01-21 09:23:26.252262: step: 872/529, loss: 0.15810136497020721 2023-01-21 09:23:27.355388: step: 876/529, loss: 0.042960025370121 2023-01-21 09:23:28.479939: step: 880/529, loss: 0.05048312991857529 2023-01-21 09:23:29.614379: step: 884/529, loss: 0.21107368171215057 2023-01-21 09:23:30.737792: step: 888/529, loss: 0.07904662936925888 2023-01-21 09:23:31.894935: step: 892/529, loss: 0.2755317687988281 2023-01-21 09:23:32.994646: step: 896/529, loss: 0.14549875259399414 2023-01-21 09:23:34.107831: step: 900/529, loss: 0.20205456018447876 2023-01-21 09:23:35.236523: step: 904/529, loss: 0.5239003896713257 2023-01-21 09:23:36.353208: step: 908/529, loss: 0.17224177718162537 2023-01-21 09:23:37.504861: step: 912/529, loss: 0.06187457963824272 2023-01-21 09:23:38.582503: step: 916/529, loss: 0.12379169464111328 2023-01-21 09:23:39.673514: step: 920/529, loss: 0.11094065010547638 2023-01-21 09:23:40.790939: step: 924/529, loss: 0.2921140789985657 2023-01-21 09:23:41.883702: step: 928/529, loss: 0.7407182455062866 2023-01-21 09:23:42.995693: step: 932/529, loss: 0.336601585149765 2023-01-21 09:23:44.137290: step: 936/529, loss: 0.07369747757911682 2023-01-21 09:23:45.258647: step: 940/529, loss: 0.13514146208763123 2023-01-21 09:23:46.367639: step: 944/529, loss: 0.16474317014217377 2023-01-21 09:23:47.456100: step: 948/529, loss: 0.024876069277524948 2023-01-21 09:23:48.555190: step: 952/529, loss: 0.9385433197021484 2023-01-21 09:23:49.665724: step: 956/529, loss: 0.02317180670797825 2023-01-21 09:23:50.783655: step: 960/529, loss: 1.1053519248962402 2023-01-21 09:23:51.882856: step: 964/529, loss: 0.011385440826416016 2023-01-21 09:23:53.013151: step: 968/529, loss: 0.11097273975610733 2023-01-21 09:23:54.126936: step: 972/529, loss: 0.4098798334598541 2023-01-21 09:23:55.243001: step: 976/529, loss: 0.31534457206726074 2023-01-21 09:23:56.350503: step: 980/529, loss: 0.021639680489897728 2023-01-21 09:23:57.457601: step: 984/529, loss: 0.252638578414917 2023-01-21 09:23:58.584201: step: 988/529, loss: 0.14069929718971252 2023-01-21 09:23:59.712632: step: 992/529, loss: 0.18577317893505096 2023-01-21 09:24:00.810319: step: 996/529, loss: 0.7731400728225708 2023-01-21 09:24:01.911807: step: 1000/529, loss: 0.979948103427887 2023-01-21 09:24:03.049235: step: 1004/529, loss: 0.2003905326128006 2023-01-21 09:24:04.190580: step: 1008/529, loss: 0.5764471292495728 2023-01-21 09:24:05.305956: step: 1012/529, loss: 0.35344016551971436 2023-01-21 09:24:06.433544: step: 1016/529, loss: 0.10928221046924591 2023-01-21 09:24:07.535257: step: 1020/529, loss: 0.09975261986255646 2023-01-21 09:24:08.633435: step: 1024/529, loss: 0.10154419392347336 2023-01-21 09:24:09.742676: step: 1028/529, loss: 0.36940690875053406 2023-01-21 09:24:10.859746: step: 1032/529, loss: 0.2673810124397278 2023-01-21 09:24:12.003057: step: 1036/529, loss: 0.04351988062262535 2023-01-21 09:24:13.100980: step: 1040/529, loss: 0.8431259393692017 2023-01-21 09:24:14.195867: step: 1044/529, loss: 0.1008274108171463 2023-01-21 09:24:15.330026: step: 1048/529, loss: 0.10504446178674698 2023-01-21 09:24:16.439620: step: 1052/529, loss: 0.3396528363227844 2023-01-21 09:24:17.574478: step: 1056/529, loss: 0.32092055678367615 2023-01-21 09:24:18.690606: step: 1060/529, loss: 2.079563617706299 2023-01-21 09:24:19.787242: step: 1064/529, loss: 0.0264403335750103 2023-01-21 09:24:20.929287: step: 1068/529, loss: 0.08323893696069717 2023-01-21 09:24:22.021677: step: 1072/529, loss: 0.13182954490184784 2023-01-21 09:24:23.136323: step: 1076/529, loss: 0.4027717411518097 2023-01-21 09:24:24.268876: step: 1080/529, loss: 0.1484191119670868 2023-01-21 09:24:25.400608: step: 1084/529, loss: 0.08234620094299316 2023-01-21 09:24:26.503078: step: 1088/529, loss: 1.2295171022415161 2023-01-21 09:24:27.624948: step: 1092/529, loss: 0.17482653260231018 2023-01-21 09:24:28.764962: step: 1096/529, loss: 0.6987869739532471 2023-01-21 09:24:29.897367: step: 1100/529, loss: 0.12622275948524475 2023-01-21 09:24:31.020910: step: 1104/529, loss: 0.1770588457584381 2023-01-21 09:24:32.131856: step: 1108/529, loss: 0.2589738965034485 2023-01-21 09:24:33.268722: step: 1112/529, loss: 0.49493399262428284 2023-01-21 09:24:34.361730: step: 1116/529, loss: 0.14513494074344635 2023-01-21 09:24:35.500483: step: 1120/529, loss: 0.16341623663902283 2023-01-21 09:24:36.587959: step: 1124/529, loss: 0.10604343563318253 2023-01-21 09:24:37.722972: step: 1128/529, loss: 1.788381576538086 2023-01-21 09:24:38.825036: step: 1132/529, loss: 0.11593261361122131 2023-01-21 09:24:39.920983: step: 1136/529, loss: 0.051560401916503906 2023-01-21 09:24:41.030982: step: 1140/529, loss: 0.06905083358287811 2023-01-21 09:24:42.148905: step: 1144/529, loss: 0.2862909436225891 2023-01-21 09:24:43.258861: step: 1148/529, loss: 0.0763545036315918 2023-01-21 09:24:44.356663: step: 1152/529, loss: 0.1487351506948471 2023-01-21 09:24:45.474038: step: 1156/529, loss: 0.07588143646717072 2023-01-21 09:24:46.616206: step: 1160/529, loss: 0.28959304094314575 2023-01-21 09:24:47.733390: step: 1164/529, loss: 0.027448464184999466 2023-01-21 09:24:48.832233: step: 1168/529, loss: 0.1361057311296463 2023-01-21 09:24:49.923501: step: 1172/529, loss: 1.1816768646240234 2023-01-21 09:24:51.056906: step: 1176/529, loss: 0.1459817886352539 2023-01-21 09:24:52.177457: step: 1180/529, loss: 0.4176911413669586 2023-01-21 09:24:53.273363: step: 1184/529, loss: 0.05482196807861328 2023-01-21 09:24:54.358504: step: 1188/529, loss: 0.4296841025352478 2023-01-21 09:24:55.474964: step: 1192/529, loss: 0.08291494846343994 2023-01-21 09:24:56.601709: step: 1196/529, loss: 0.6060510873794556 2023-01-21 09:24:57.712546: step: 1200/529, loss: 0.22685785591602325 2023-01-21 09:24:58.829018: step: 1204/529, loss: 0.25125741958618164 2023-01-21 09:24:59.917301: step: 1208/529, loss: 0.10662183910608292 2023-01-21 09:25:01.056876: step: 1212/529, loss: 0.11419658362865448 2023-01-21 09:25:02.183213: step: 1216/529, loss: 0.11586398631334305 2023-01-21 09:25:03.303511: step: 1220/529, loss: 0.2619855999946594 2023-01-21 09:25:04.476806: step: 1224/529, loss: 0.1439422219991684 2023-01-21 09:25:05.625975: step: 1228/529, loss: 0.7763123512268066 2023-01-21 09:25:06.750242: step: 1232/529, loss: 0.10478439182043076 2023-01-21 09:25:07.862257: step: 1236/529, loss: 0.15502873063087463 2023-01-21 09:25:08.993379: step: 1240/529, loss: 0.21867504715919495 2023-01-21 09:25:10.112195: step: 1244/529, loss: 0.4688391387462616 2023-01-21 09:25:11.199974: step: 1248/529, loss: 0.06149768829345703 2023-01-21 09:25:12.295197: step: 1252/529, loss: 0.12986259162425995 2023-01-21 09:25:13.407103: step: 1256/529, loss: 0.24377644062042236 2023-01-21 09:25:14.572351: step: 1260/529, loss: 0.23835992813110352 2023-01-21 09:25:15.647122: step: 1264/529, loss: 0.03838854283094406 2023-01-21 09:25:16.776174: step: 1268/529, loss: 0.10492667555809021 2023-01-21 09:25:17.895419: step: 1272/529, loss: 0.1702306866645813 2023-01-21 09:25:18.994501: step: 1276/529, loss: 0.3218880891799927 2023-01-21 09:25:20.105092: step: 1280/529, loss: 0.1520165503025055 2023-01-21 09:25:21.238486: step: 1284/529, loss: 0.14958299696445465 2023-01-21 09:25:22.341489: step: 1288/529, loss: 0.08503346145153046 2023-01-21 09:25:23.438601: step: 1292/529, loss: 0.40545588731765747 2023-01-21 09:25:24.549120: step: 1296/529, loss: 0.12282752990722656 2023-01-21 09:25:25.641771: step: 1300/529, loss: 0.02224903181195259 2023-01-21 09:25:26.747673: step: 1304/529, loss: 0.18410158157348633 2023-01-21 09:25:27.863339: step: 1308/529, loss: 0.06564927101135254 2023-01-21 09:25:28.975203: step: 1312/529, loss: 0.16934624314308167 2023-01-21 09:25:30.052637: step: 1316/529, loss: 0.04940595477819443 2023-01-21 09:25:31.176323: step: 1320/529, loss: 0.1700696051120758 2023-01-21 09:25:32.294520: step: 1324/529, loss: 0.05648774653673172 2023-01-21 09:25:33.388650: step: 1328/529, loss: 0.21957778930664062 2023-01-21 09:25:34.484556: step: 1332/529, loss: 0.15166538953781128 2023-01-21 09:25:35.665253: step: 1336/529, loss: 0.18586693704128265 2023-01-21 09:25:36.779632: step: 1340/529, loss: 0.3524380624294281 2023-01-21 09:25:37.903341: step: 1344/529, loss: 0.52020663022995 2023-01-21 09:25:39.016059: step: 1348/529, loss: 0.03911638259887695 2023-01-21 09:25:40.120869: step: 1352/529, loss: 0.7306599617004395 2023-01-21 09:25:41.232628: step: 1356/529, loss: 1.309922456741333 2023-01-21 09:25:42.329878: step: 1360/529, loss: 0.661201000213623 2023-01-21 09:25:43.439898: step: 1364/529, loss: 0.2064886838197708 2023-01-21 09:25:44.554756: step: 1368/529, loss: 0.22166918218135834 2023-01-21 09:25:45.692356: step: 1372/529, loss: 0.291122168302536 2023-01-21 09:25:46.825421: step: 1376/529, loss: 1.22531259059906 2023-01-21 09:25:47.931310: step: 1380/529, loss: 0.11571875214576721 2023-01-21 09:25:49.040268: step: 1384/529, loss: 0.18571476638317108 2023-01-21 09:25:50.176204: step: 1388/529, loss: 0.09639438986778259 2023-01-21 09:25:51.252170: step: 1392/529, loss: 0.14718157052993774 2023-01-21 09:25:52.368140: step: 1396/529, loss: 0.320192813873291 2023-01-21 09:25:53.493134: step: 1400/529, loss: 0.40457314252853394 2023-01-21 09:25:54.590764: step: 1404/529, loss: 0.151570126414299 2023-01-21 09:25:55.683648: step: 1408/529, loss: 0.32254552841186523 2023-01-21 09:25:56.798514: step: 1412/529, loss: 0.1903323233127594 2023-01-21 09:25:57.946610: step: 1416/529, loss: 1.7809343338012695 2023-01-21 09:25:59.093837: step: 1420/529, loss: 0.1599511206150055 2023-01-21 09:26:00.223096: step: 1424/529, loss: 0.07046294212341309 2023-01-21 09:26:01.347189: step: 1428/529, loss: 0.3870340883731842 2023-01-21 09:26:02.484896: step: 1432/529, loss: 0.0396091490983963 2023-01-21 09:26:03.586042: step: 1436/529, loss: 0.09298677742481232 2023-01-21 09:26:04.709504: step: 1440/529, loss: 1.5005346536636353 2023-01-21 09:26:05.811266: step: 1444/529, loss: 0.10666771233081818 2023-01-21 09:26:06.918873: step: 1448/529, loss: 0.3341483175754547 2023-01-21 09:26:08.064506: step: 1452/529, loss: 0.2629149556159973 2023-01-21 09:26:09.167227: step: 1456/529, loss: 0.591636061668396 2023-01-21 09:26:10.277027: step: 1460/529, loss: 0.10159340500831604 2023-01-21 09:26:11.394611: step: 1464/529, loss: 0.0817832499742508 2023-01-21 09:26:12.490578: step: 1468/529, loss: 0.12911973893642426 2023-01-21 09:26:13.601964: step: 1472/529, loss: 0.0567837730050087 2023-01-21 09:26:14.729331: step: 1476/529, loss: 0.6025118827819824 2023-01-21 09:26:15.826045: step: 1480/529, loss: 0.13568849861621857 2023-01-21 09:26:16.941354: step: 1484/529, loss: 0.007483577821403742 2023-01-21 09:26:18.048958: step: 1488/529, loss: 0.03460747003555298 2023-01-21 09:26:19.193109: step: 1492/529, loss: 0.0917363166809082 2023-01-21 09:26:20.325448: step: 1496/529, loss: 0.27723801136016846 2023-01-21 09:26:21.419567: step: 1500/529, loss: 0.194269061088562 2023-01-21 09:26:22.532481: step: 1504/529, loss: 0.14470605552196503 2023-01-21 09:26:23.628963: step: 1508/529, loss: 0.04845142737030983 2023-01-21 09:26:24.825763: step: 1512/529, loss: 0.803070068359375 2023-01-21 09:26:25.936617: step: 1516/529, loss: 0.18038225173950195 2023-01-21 09:26:27.056224: step: 1520/529, loss: 0.46802178025245667 2023-01-21 09:26:28.163274: step: 1524/529, loss: 0.1540185958147049 2023-01-21 09:26:29.274622: step: 1528/529, loss: 0.06892995536327362 2023-01-21 09:26:30.369984: step: 1532/529, loss: 0.1509508192539215 2023-01-21 09:26:31.511567: step: 1536/529, loss: 0.08187460899353027 2023-01-21 09:26:32.639021: step: 1540/529, loss: 0.2218642681837082 2023-01-21 09:26:33.743559: step: 1544/529, loss: 0.3623002767562866 2023-01-21 09:26:34.846235: step: 1548/529, loss: 0.3448387384414673 2023-01-21 09:26:35.940418: step: 1552/529, loss: 0.1906498670578003 2023-01-21 09:26:37.059862: step: 1556/529, loss: 0.054180387407541275 2023-01-21 09:26:38.156711: step: 1560/529, loss: 0.06531481444835663 2023-01-21 09:26:39.273950: step: 1564/529, loss: 0.061733536422252655 2023-01-21 09:26:40.362832: step: 1568/529, loss: 0.08037319779396057 2023-01-21 09:26:41.456261: step: 1572/529, loss: 0.16298404335975647 2023-01-21 09:26:42.569428: step: 1576/529, loss: 0.09026866406202316 2023-01-21 09:26:43.691557: step: 1580/529, loss: 0.22508811950683594 2023-01-21 09:26:44.777077: step: 1584/529, loss: 0.3090890049934387 2023-01-21 09:26:45.891038: step: 1588/529, loss: 0.04922962188720703 2023-01-21 09:26:47.026900: step: 1592/529, loss: 0.5558130145072937 2023-01-21 09:26:48.160200: step: 1596/529, loss: 0.1086030900478363 2023-01-21 09:26:49.280412: step: 1600/529, loss: 0.13965165615081787 2023-01-21 09:26:50.418999: step: 1604/529, loss: 0.13384190201759338 2023-01-21 09:26:51.589917: step: 1608/529, loss: 0.6677474975585938 2023-01-21 09:26:52.737093: step: 1612/529, loss: 0.08470864593982697 2023-01-21 09:26:53.872978: step: 1616/529, loss: 0.3861212730407715 2023-01-21 09:26:54.959291: step: 1620/529, loss: 0.1051628589630127 2023-01-21 09:26:56.096663: step: 1624/529, loss: 0.30070409178733826 2023-01-21 09:26:57.243152: step: 1628/529, loss: 0.877350389957428 2023-01-21 09:26:58.385879: step: 1632/529, loss: 2.248197555541992 2023-01-21 09:26:59.499980: step: 1636/529, loss: 0.19142599403858185 2023-01-21 09:27:00.590272: step: 1640/529, loss: 0.15653257071971893 2023-01-21 09:27:01.718685: step: 1644/529, loss: 0.20536357164382935 2023-01-21 09:27:02.839632: step: 1648/529, loss: 0.17358675599098206 2023-01-21 09:27:03.948863: step: 1652/529, loss: 0.17834492027759552 2023-01-21 09:27:05.070046: step: 1656/529, loss: 0.2236747294664383 2023-01-21 09:27:06.181548: step: 1660/529, loss: 0.28338009119033813 2023-01-21 09:27:07.323049: step: 1664/529, loss: 0.19938039779663086 2023-01-21 09:27:08.438623: step: 1668/529, loss: 0.6517542004585266 2023-01-21 09:27:09.570952: step: 1672/529, loss: 0.6286314129829407 2023-01-21 09:27:10.675208: step: 1676/529, loss: 0.10910497605800629 2023-01-21 09:27:11.803441: step: 1680/529, loss: 0.6725395321846008 2023-01-21 09:27:12.924293: step: 1684/529, loss: 0.07103876769542694 2023-01-21 09:27:14.060591: step: 1688/529, loss: 0.8393896818161011 2023-01-21 09:27:15.163165: step: 1692/529, loss: 0.035956382751464844 2023-01-21 09:27:16.269638: step: 1696/529, loss: 0.18945342302322388 2023-01-21 09:27:17.403850: step: 1700/529, loss: 0.040915776044130325 2023-01-21 09:27:18.510492: step: 1704/529, loss: 0.28783971071243286 2023-01-21 09:27:19.638902: step: 1708/529, loss: 0.21551647782325745 2023-01-21 09:27:20.787220: step: 1712/529, loss: 0.3832671046257019 2023-01-21 09:27:21.900054: step: 1716/529, loss: 0.12304279208183289 2023-01-21 09:27:23.043478: step: 1720/529, loss: 0.14177627861499786 2023-01-21 09:27:24.189918: step: 1724/529, loss: 0.16448087990283966 2023-01-21 09:27:25.297961: step: 1728/529, loss: 0.15942887961864471 2023-01-21 09:27:26.382157: step: 1732/529, loss: 0.5898841619491577 2023-01-21 09:27:27.497734: step: 1736/529, loss: 0.33383965492248535 2023-01-21 09:27:28.642934: step: 1740/529, loss: 0.24272727966308594 2023-01-21 09:27:29.747227: step: 1744/529, loss: 0.08787364512681961 2023-01-21 09:27:30.863309: step: 1748/529, loss: 0.6018533706665039 2023-01-21 09:27:31.981807: step: 1752/529, loss: 0.3568808436393738 2023-01-21 09:27:33.060866: step: 1756/529, loss: 0.05798931419849396 2023-01-21 09:27:34.149875: step: 1760/529, loss: 0.24187889695167542 2023-01-21 09:27:35.281904: step: 1764/529, loss: 0.16320466995239258 2023-01-21 09:27:36.421840: step: 1768/529, loss: 0.04443786293268204 2023-01-21 09:27:37.530249: step: 1772/529, loss: 0.0804021805524826 2023-01-21 09:27:38.647732: step: 1776/529, loss: 0.435655415058136 2023-01-21 09:27:39.785098: step: 1780/529, loss: 0.23417937755584717 2023-01-21 09:27:40.885288: step: 1784/529, loss: 0.10017052292823792 2023-01-21 09:27:41.992984: step: 1788/529, loss: 0.3650458753108978 2023-01-21 09:27:43.144453: step: 1792/529, loss: 0.20216694474220276 2023-01-21 09:27:44.257758: step: 1796/529, loss: 0.15957202017307281 2023-01-21 09:27:45.358728: step: 1800/529, loss: 0.44204607605934143 2023-01-21 09:27:46.467017: step: 1804/529, loss: 0.26193952560424805 2023-01-21 09:27:47.579652: step: 1808/529, loss: 0.38003477454185486 2023-01-21 09:27:48.701798: step: 1812/529, loss: 0.4287465214729309 2023-01-21 09:27:49.796591: step: 1816/529, loss: 0.3446895480155945 2023-01-21 09:27:50.930659: step: 1820/529, loss: 0.053977206349372864 2023-01-21 09:27:52.062707: step: 1824/529, loss: 0.5122342705726624 2023-01-21 09:27:53.164637: step: 1828/529, loss: 0.0634603500366211 2023-01-21 09:27:54.265651: step: 1832/529, loss: 0.10035190731287003 2023-01-21 09:27:55.400431: step: 1836/529, loss: 0.032689858227968216 2023-01-21 09:27:56.520766: step: 1840/529, loss: 1.1659190654754639 2023-01-21 09:27:57.620104: step: 1844/529, loss: 0.16175565123558044 2023-01-21 09:27:58.740198: step: 1848/529, loss: 0.006393528077751398 2023-01-21 09:27:59.860802: step: 1852/529, loss: 0.9888699054718018 2023-01-21 09:28:00.992524: step: 1856/529, loss: 0.13269320130348206 2023-01-21 09:28:02.117989: step: 1860/529, loss: 0.45847225189208984 2023-01-21 09:28:03.205822: step: 1864/529, loss: 0.5527905821800232 2023-01-21 09:28:04.328542: step: 1868/529, loss: 0.2132083922624588 2023-01-21 09:28:05.422180: step: 1872/529, loss: 0.049288153648376465 2023-01-21 09:28:06.543631: step: 1876/529, loss: 0.14698143303394318 2023-01-21 09:28:07.661678: step: 1880/529, loss: 0.23518601059913635 2023-01-21 09:28:08.759626: step: 1884/529, loss: 1.0319111347198486 2023-01-21 09:28:09.868587: step: 1888/529, loss: 0.029146291315555573 2023-01-21 09:28:10.996880: step: 1892/529, loss: 0.5506752729415894 2023-01-21 09:28:12.095546: step: 1896/529, loss: 0.3898093104362488 2023-01-21 09:28:13.204767: step: 1900/529, loss: 0.059124141931533813 2023-01-21 09:28:14.344596: step: 1904/529, loss: 0.04013986885547638 2023-01-21 09:28:15.472957: step: 1908/529, loss: 0.06953268498182297 2023-01-21 09:28:16.574719: step: 1912/529, loss: 0.1196565181016922 2023-01-21 09:28:17.730256: step: 1916/529, loss: 0.7917718887329102 2023-01-21 09:28:18.836098: step: 1920/529, loss: 0.1666768193244934 2023-01-21 09:28:19.915899: step: 1924/529, loss: 0.13419543206691742 2023-01-21 09:28:21.034084: step: 1928/529, loss: 0.3345470130443573 2023-01-21 09:28:22.157340: step: 1932/529, loss: 0.25586605072021484 2023-01-21 09:28:23.276304: step: 1936/529, loss: 0.12137804180383682 2023-01-21 09:28:24.418289: step: 1940/529, loss: 1.7105233669281006 2023-01-21 09:28:25.546883: step: 1944/529, loss: 0.25304508209228516 2023-01-21 09:28:26.660704: step: 1948/529, loss: 0.2101493626832962 2023-01-21 09:28:27.818796: step: 1952/529, loss: 0.22466516494750977 2023-01-21 09:28:28.920447: step: 1956/529, loss: 0.1067010909318924 2023-01-21 09:28:30.045039: step: 1960/529, loss: 0.08519849926233292 2023-01-21 09:28:31.143592: step: 1964/529, loss: 0.13402409851551056 2023-01-21 09:28:32.235851: step: 1968/529, loss: 0.06219963729381561 2023-01-21 09:28:33.383817: step: 1972/529, loss: 1.2101415395736694 2023-01-21 09:28:34.493314: step: 1976/529, loss: 0.0624537467956543 2023-01-21 09:28:35.609504: step: 1980/529, loss: 0.05861020088195801 2023-01-21 09:28:36.736148: step: 1984/529, loss: 0.573178768157959 2023-01-21 09:28:37.845133: step: 1988/529, loss: 0.5627970099449158 2023-01-21 09:28:38.986625: step: 1992/529, loss: 0.09205227345228195 2023-01-21 09:28:40.104869: step: 1996/529, loss: 0.27253830432891846 2023-01-21 09:28:41.219422: step: 2000/529, loss: 0.1250534951686859 2023-01-21 09:28:42.346459: step: 2004/529, loss: 0.09731898456811905 2023-01-21 09:28:43.461737: step: 2008/529, loss: 0.2533365488052368 2023-01-21 09:28:44.588679: step: 2012/529, loss: 0.13066311180591583 2023-01-21 09:28:45.736482: step: 2016/529, loss: 0.3655029535293579 2023-01-21 09:28:46.886883: step: 2020/529, loss: 0.1358109414577484 2023-01-21 09:28:48.012966: step: 2024/529, loss: 0.11143863201141357 2023-01-21 09:28:49.133501: step: 2028/529, loss: 0.14439940452575684 2023-01-21 09:28:50.252698: step: 2032/529, loss: 0.08167238533496857 2023-01-21 09:28:51.342992: step: 2036/529, loss: 0.504469096660614 2023-01-21 09:28:52.475656: step: 2040/529, loss: 0.1421714872121811 2023-01-21 09:28:53.594464: step: 2044/529, loss: 0.3318385183811188 2023-01-21 09:28:54.694249: step: 2048/529, loss: 0.0830760970711708 2023-01-21 09:28:55.802603: step: 2052/529, loss: 0.10016363114118576 2023-01-21 09:28:56.963063: step: 2056/529, loss: 0.27056217193603516 2023-01-21 09:28:58.058910: step: 2060/529, loss: 0.06310959160327911 2023-01-21 09:28:59.173987: step: 2064/529, loss: 0.05799704045057297 2023-01-21 09:29:00.301058: step: 2068/529, loss: 0.2633543014526367 2023-01-21 09:29:01.428842: step: 2072/529, loss: 0.13678884506225586 2023-01-21 09:29:02.520665: step: 2076/529, loss: 0.15997877717018127 2023-01-21 09:29:03.621968: step: 2080/529, loss: 0.10812397301197052 2023-01-21 09:29:04.708554: step: 2084/529, loss: 0.1123078316450119 2023-01-21 09:29:05.836800: step: 2088/529, loss: 0.1491670310497284 2023-01-21 09:29:06.928432: step: 2092/529, loss: 0.07064428180456161 2023-01-21 09:29:08.041618: step: 2096/529, loss: 0.5811355710029602 2023-01-21 09:29:09.147863: step: 2100/529, loss: 0.5360218286514282 2023-01-21 09:29:10.279181: step: 2104/529, loss: 0.09982369095087051 2023-01-21 09:29:11.397318: step: 2108/529, loss: 0.1943781077861786 2023-01-21 09:29:12.493503: step: 2112/529, loss: 0.7589541077613831 2023-01-21 09:29:13.581947: step: 2116/529, loss: 0.042855359613895416 ================================================== Loss: 0.314 -------------------- Dev: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Test: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Russian: {'event': {'p': 0.4222222222222222, 'r': 0.5277777777777778, 'f1': 0.46913580246913583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, '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 09:30:00.997459: step: 4/529, loss: 0.41854560375213623 2023-01-21 09:30:02.120432: step: 8/529, loss: 0.2615930736064911 2023-01-21 09:30:03.219265: step: 12/529, loss: 0.12992648780345917 2023-01-21 09:30:04.341756: step: 16/529, loss: 0.1943288892507553 2023-01-21 09:30:05.466517: step: 20/529, loss: 0.14407122135162354 2023-01-21 09:30:06.586068: step: 24/529, loss: 0.20248670876026154 2023-01-21 09:30:07.698980: step: 28/529, loss: 0.08645305782556534 2023-01-21 09:30:08.788971: step: 32/529, loss: 0.12083292752504349 2023-01-21 09:30:09.905701: step: 36/529, loss: 0.17055034637451172 2023-01-21 09:30:11.008961: step: 40/529, loss: 0.11218681931495667 2023-01-21 09:30:12.104787: step: 44/529, loss: 0.040493011474609375 2023-01-21 09:30:13.249984: step: 48/529, loss: 0.09289960563182831 2023-01-21 09:30:14.395363: step: 52/529, loss: 0.349210262298584 2023-01-21 09:30:15.510904: step: 56/529, loss: 0.10196900367736816 2023-01-21 09:30:16.659623: step: 60/529, loss: 0.11311708390712738 2023-01-21 09:30:17.773281: step: 64/529, loss: 0.12583903968334198 2023-01-21 09:30:18.858692: step: 68/529, loss: 0.1280498504638672 2023-01-21 09:30:19.998921: step: 72/529, loss: 0.2826705873012543 2023-01-21 09:30:21.101135: step: 76/529, loss: 0.2821914553642273 2023-01-21 09:30:22.188516: step: 80/529, loss: 0.21954460442066193 2023-01-21 09:30:23.323257: step: 84/529, loss: 0.3892592191696167 2023-01-21 09:30:24.414474: step: 88/529, loss: 0.418751984834671 2023-01-21 09:30:25.526893: step: 92/529, loss: 0.21727648377418518 2023-01-21 09:30:26.632383: step: 96/529, loss: 0.0854133665561676 2023-01-21 09:30:27.727586: step: 100/529, loss: 0.15490931272506714 2023-01-21 09:30:28.823009: step: 104/529, loss: 0.1582166999578476 2023-01-21 09:30:29.925609: step: 108/529, loss: 0.35520878434181213 2023-01-21 09:30:31.061645: step: 112/529, loss: 0.10291080921888351 2023-01-21 09:30:32.172248: step: 116/529, loss: 0.09500789642333984 2023-01-21 09:30:33.273335: step: 120/529, loss: 0.08937511593103409 2023-01-21 09:30:34.416334: step: 124/529, loss: 0.24723386764526367 2023-01-21 09:30:35.512288: step: 128/529, loss: 0.07378844916820526 2023-01-21 09:30:36.621503: step: 132/529, loss: 0.06312823295593262 2023-01-21 09:30:37.739848: step: 136/529, loss: 0.11058712750673294 2023-01-21 09:30:38.851577: step: 140/529, loss: 0.9841111302375793 2023-01-21 09:30:40.000752: step: 144/529, loss: 0.025448037311434746 2023-01-21 09:30:41.092684: step: 148/529, loss: 0.14135445654392242 2023-01-21 09:30:42.217768: step: 152/529, loss: 0.041330672800540924 2023-01-21 09:30:43.306076: step: 156/529, loss: 0.05406588315963745 2023-01-21 09:30:44.416888: step: 160/529, loss: 0.2682167887687683 2023-01-21 09:30:45.547295: step: 164/529, loss: 0.12867975234985352 2023-01-21 09:30:46.643522: step: 168/529, loss: 0.11226129531860352 2023-01-21 09:30:47.770960: step: 172/529, loss: 0.03821101039648056 2023-01-21 09:30:48.891737: step: 176/529, loss: 0.0940001979470253 2023-01-21 09:30:50.032892: step: 180/529, loss: 0.22178764641284943 2023-01-21 09:30:51.155691: step: 184/529, loss: 0.2003268301486969 2023-01-21 09:30:52.303655: step: 188/529, loss: 0.1507258415222168 2023-01-21 09:30:53.444257: step: 192/529, loss: 0.10898199677467346 2023-01-21 09:30:54.555587: step: 196/529, loss: 0.07951474189758301 2023-01-21 09:30:55.664743: step: 200/529, loss: 0.10532769560813904 2023-01-21 09:30:56.779759: step: 204/529, loss: 0.17203585803508759 2023-01-21 09:30:57.890989: step: 208/529, loss: 0.15487423539161682 2023-01-21 09:30:59.015965: step: 212/529, loss: 0.11974792182445526 2023-01-21 09:31:00.120505: step: 216/529, loss: 0.12406959384679794 2023-01-21 09:31:01.242401: step: 220/529, loss: 0.7428269386291504 2023-01-21 09:31:02.364942: step: 224/529, loss: 0.056182861328125 2023-01-21 09:31:03.476254: step: 228/529, loss: 0.01631774939596653 2023-01-21 09:31:04.602873: step: 232/529, loss: 0.05640731006860733 2023-01-21 09:31:05.688706: step: 236/529, loss: 0.230677992105484 2023-01-21 09:31:06.792839: step: 240/529, loss: 0.3648664355278015 2023-01-21 09:31:07.902128: step: 244/529, loss: 0.08810434490442276 2023-01-21 09:31:09.003839: step: 248/529, loss: 0.3475028872489929 2023-01-21 09:31:10.117824: step: 252/529, loss: 0.056288961321115494 2023-01-21 09:31:11.221433: step: 256/529, loss: 0.01311262883245945 2023-01-21 09:31:12.309140: step: 260/529, loss: 0.38443297147750854 2023-01-21 09:31:13.413903: step: 264/529, loss: 0.335693359375 2023-01-21 09:31:14.543695: step: 268/529, loss: 0.16695089638233185 2023-01-21 09:31:15.641622: step: 272/529, loss: 0.06345872581005096 2023-01-21 09:31:16.778832: step: 276/529, loss: 0.2842293381690979 2023-01-21 09:31:17.862862: step: 280/529, loss: 0.06773070991039276 2023-01-21 09:31:18.989601: step: 284/529, loss: 0.135711669921875 2023-01-21 09:31:20.124336: step: 288/529, loss: 0.14520874619483948 2023-01-21 09:31:21.279683: step: 292/529, loss: 1.0627326965332031 2023-01-21 09:31:22.389502: step: 296/529, loss: 0.12591552734375 2023-01-21 09:31:23.490277: step: 300/529, loss: 0.16250251233577728 2023-01-21 09:31:24.638960: step: 304/529, loss: 0.22057685256004333 2023-01-21 09:31:25.741250: step: 308/529, loss: 0.029260635375976562 2023-01-21 09:31:26.866072: step: 312/529, loss: 0.019703388214111328 2023-01-21 09:31:27.997038: step: 316/529, loss: 0.4171700179576874 2023-01-21 09:31:29.164490: step: 320/529, loss: 0.2709653079509735 2023-01-21 09:31:30.290395: step: 324/529, loss: 0.1707896590232849 2023-01-21 09:31:31.390709: step: 328/529, loss: 0.10557766258716583 2023-01-21 09:31:32.513623: step: 332/529, loss: 0.1140020340681076 2023-01-21 09:31:33.617046: step: 336/529, loss: 0.23821982741355896 2023-01-21 09:31:34.735196: step: 340/529, loss: 0.05409717559814453 2023-01-21 09:31:35.844659: step: 344/529, loss: 0.07998485863208771 2023-01-21 09:31:36.971647: step: 348/529, loss: 0.030089855194091797 2023-01-21 09:31:38.072921: step: 352/529, loss: 0.02464304119348526 2023-01-21 09:31:39.201569: step: 356/529, loss: 0.13897676765918732 2023-01-21 09:31:40.314769: step: 360/529, loss: 0.0669856071472168 2023-01-21 09:31:41.415947: step: 364/529, loss: 0.08883456885814667 2023-01-21 09:31:42.510024: step: 368/529, loss: 0.11503658443689346 2023-01-21 09:31:43.605537: step: 372/529, loss: 0.2833155393600464 2023-01-21 09:31:44.770599: step: 376/529, loss: 0.09108105301856995 2023-01-21 09:31:45.903573: step: 380/529, loss: 0.14699287712574005 2023-01-21 09:31:47.031007: step: 384/529, loss: 0.5391696691513062 2023-01-21 09:31:48.146531: step: 388/529, loss: 0.09249172359704971 2023-01-21 09:31:49.281574: step: 392/529, loss: 0.09266749024391174 2023-01-21 09:31:50.410720: step: 396/529, loss: 0.06567039340734482 2023-01-21 09:31:51.531500: step: 400/529, loss: 0.1745540201663971 2023-01-21 09:31:52.656835: step: 404/529, loss: 0.15886621177196503 2023-01-21 09:31:53.789461: step: 408/529, loss: 0.1589856594800949 2023-01-21 09:31:54.905232: step: 412/529, loss: 0.3134172558784485 2023-01-21 09:31:56.004699: step: 416/529, loss: 0.125766322016716 2023-01-21 09:31:57.137152: step: 420/529, loss: 0.18280763924121857 2023-01-21 09:31:58.247732: step: 424/529, loss: 0.8104706406593323 2023-01-21 09:31:59.353093: step: 428/529, loss: 0.06938963383436203 2023-01-21 09:32:00.476634: step: 432/529, loss: 0.6461702585220337 2023-01-21 09:32:01.583703: step: 436/529, loss: 0.15197500586509705 2023-01-21 09:32:02.688215: step: 440/529, loss: 0.0949891060590744 2023-01-21 09:32:03.805743: step: 444/529, loss: 0.23109427094459534 2023-01-21 09:32:04.912314: step: 448/529, loss: 0.13077601790428162 2023-01-21 09:32:06.001291: step: 452/529, loss: 0.09038834273815155 2023-01-21 09:32:07.117879: step: 456/529, loss: 0.2053825557231903 2023-01-21 09:32:08.224084: step: 460/529, loss: 0.17105592787265778 2023-01-21 09:32:09.328441: step: 464/529, loss: 0.01946844905614853 2023-01-21 09:32:10.433527: step: 468/529, loss: 0.04283332824707031 2023-01-21 09:32:11.530353: step: 472/529, loss: 0.2230476438999176 2023-01-21 09:32:12.692569: step: 476/529, loss: 0.5186672210693359 2023-01-21 09:32:13.810120: step: 480/529, loss: 0.0735451728105545 2023-01-21 09:32:14.919085: step: 484/529, loss: 0.11196231842041016 2023-01-21 09:32:16.061197: step: 488/529, loss: 0.5283088684082031 2023-01-21 09:32:17.185125: step: 492/529, loss: 0.7329642176628113 2023-01-21 09:32:18.297953: step: 496/529, loss: 0.1420719176530838 2023-01-21 09:32:19.393805: step: 500/529, loss: 0.03245124965906143 2023-01-21 09:32:20.486022: step: 504/529, loss: 0.908942461013794 2023-01-21 09:32:21.598486: step: 508/529, loss: 0.14457498490810394 2023-01-21 09:32:22.715046: step: 512/529, loss: 0.3416524827480316 2023-01-21 09:32:23.845546: step: 516/529, loss: 0.16285952925682068 2023-01-21 09:32:24.937756: step: 520/529, loss: 0.05690765380859375 2023-01-21 09:32:26.066771: step: 524/529, loss: 0.7925837635993958 2023-01-21 09:32:27.191273: step: 528/529, loss: 0.21937322616577148 2023-01-21 09:32:28.279243: step: 532/529, loss: 0.2915459871292114 2023-01-21 09:32:29.362128: step: 536/529, loss: 0.12484331429004669 2023-01-21 09:32:30.465747: step: 540/529, loss: 0.15269441902637482 2023-01-21 09:32:31.562464: step: 544/529, loss: 0.13786816596984863 2023-01-21 09:32:32.659364: step: 548/529, loss: 0.054793838411569595 2023-01-21 09:32:33.750884: step: 552/529, loss: 1.2599841356277466 2023-01-21 09:32:34.869410: step: 556/529, loss: 0.18873223662376404 2023-01-21 09:32:35.958961: step: 560/529, loss: 0.0888739600777626 2023-01-21 09:32:37.093946: step: 564/529, loss: 0.17296858131885529 2023-01-21 09:32:38.189944: step: 568/529, loss: 0.021279525011777878 2023-01-21 09:32:39.278396: step: 572/529, loss: 0.07156114280223846 2023-01-21 09:32:40.408137: step: 576/529, loss: 0.29135942459106445 2023-01-21 09:32:41.540570: step: 580/529, loss: 0.18069811165332794 2023-01-21 09:32:42.640847: step: 584/529, loss: 0.08618822693824768 2023-01-21 09:32:43.793336: step: 588/529, loss: 0.08463239669799805 2023-01-21 09:32:44.914063: step: 592/529, loss: 0.175147145986557 2023-01-21 09:32:46.024856: step: 596/529, loss: 0.1767711490392685 2023-01-21 09:32:47.156353: step: 600/529, loss: 0.1781170815229416 2023-01-21 09:32:48.281071: step: 604/529, loss: 0.07512360066175461 2023-01-21 09:32:49.402055: step: 608/529, loss: 0.173912912607193 2023-01-21 09:32:50.513998: step: 612/529, loss: 0.1476341187953949 2023-01-21 09:32:51.642888: step: 616/529, loss: 0.08104848861694336 2023-01-21 09:32:52.777134: step: 620/529, loss: 0.040274716913700104 2023-01-21 09:32:53.896337: step: 624/529, loss: 0.201176717877388 2023-01-21 09:32:55.067266: step: 628/529, loss: 0.03966636583209038 2023-01-21 09:32:56.158288: step: 632/529, loss: 0.11002178490161896 2023-01-21 09:32:57.291711: step: 636/529, loss: 0.23965173959732056 2023-01-21 09:32:58.429345: step: 640/529, loss: 0.5823952555656433 2023-01-21 09:32:59.531392: step: 644/529, loss: 0.2902754843235016 2023-01-21 09:33:00.632560: step: 648/529, loss: 0.024506473913788795 2023-01-21 09:33:01.762859: step: 652/529, loss: 0.1066705659031868 2023-01-21 09:33:02.898665: step: 656/529, loss: 0.15058794617652893 2023-01-21 09:33:04.030679: step: 660/529, loss: 0.08300383388996124 2023-01-21 09:33:05.162853: step: 664/529, loss: 0.0694490522146225 2023-01-21 09:33:06.279432: step: 668/529, loss: 0.21036848425865173 2023-01-21 09:33:07.397672: step: 672/529, loss: 0.06634902954101562 2023-01-21 09:33:08.534634: step: 676/529, loss: 0.15171471238136292 2023-01-21 09:33:09.683269: step: 680/529, loss: 0.08180160820484161 2023-01-21 09:33:10.836808: step: 684/529, loss: 0.009794425219297409 2023-01-21 09:33:11.949016: step: 688/529, loss: 0.15289267897605896 2023-01-21 09:33:13.041926: step: 692/529, loss: 0.05382108688354492 2023-01-21 09:33:14.157869: step: 696/529, loss: 0.09971089661121368 2023-01-21 09:33:15.272849: step: 700/529, loss: 0.037062834948301315 2023-01-21 09:33:16.353872: step: 704/529, loss: 0.3365079164505005 2023-01-21 09:33:17.437168: step: 708/529, loss: 0.5037580728530884 2023-01-21 09:33:18.540861: step: 712/529, loss: 0.20521163940429688 2023-01-21 09:33:19.681633: step: 716/529, loss: 0.10319998115301132 2023-01-21 09:33:20.823785: step: 720/529, loss: 0.0753898173570633 2023-01-21 09:33:21.923792: step: 724/529, loss: 0.060896776616573334 2023-01-21 09:33:23.072343: step: 728/529, loss: 0.15772362053394318 2023-01-21 09:33:24.221032: step: 732/529, loss: 0.10997290164232254 2023-01-21 09:33:25.371517: step: 736/529, loss: 0.10512867569923401 2023-01-21 09:33:26.512196: step: 740/529, loss: 0.07967662811279297 2023-01-21 09:33:27.643047: step: 744/529, loss: 0.10420303046703339 2023-01-21 09:33:28.766210: step: 748/529, loss: 0.4614826440811157 2023-01-21 09:33:29.923033: step: 752/529, loss: 0.1640513390302658 2023-01-21 09:33:31.033190: step: 756/529, loss: 0.12356098741292953 2023-01-21 09:33:32.158836: step: 760/529, loss: 0.09436388313770294 2023-01-21 09:33:33.285621: step: 764/529, loss: 0.1559332013130188 2023-01-21 09:33:34.397931: step: 768/529, loss: 0.02114286459982395 2023-01-21 09:33:35.538641: step: 772/529, loss: 0.268502801656723 2023-01-21 09:33:36.660175: step: 776/529, loss: 0.030556965619325638 2023-01-21 09:33:37.760168: step: 780/529, loss: 0.0412907600402832 2023-01-21 09:33:38.884822: step: 784/529, loss: 0.1025749146938324 2023-01-21 09:33:40.032505: step: 788/529, loss: 0.5738879442214966 2023-01-21 09:33:41.148864: step: 792/529, loss: 0.1171075850725174 2023-01-21 09:33:42.289886: step: 796/529, loss: 0.1833280622959137 2023-01-21 09:33:43.375575: step: 800/529, loss: 0.04897637665271759 2023-01-21 09:33:44.501748: step: 804/529, loss: 0.1427517831325531 2023-01-21 09:33:45.591923: step: 808/529, loss: 0.12133254110813141 2023-01-21 09:33:46.714483: step: 812/529, loss: 0.08046194165945053 2023-01-21 09:33:47.819015: step: 816/529, loss: 0.33535730838775635 2023-01-21 09:33:48.928365: step: 820/529, loss: 0.8368805646896362 2023-01-21 09:33:50.069234: step: 824/529, loss: 0.4099559783935547 2023-01-21 09:33:51.202747: step: 828/529, loss: 0.34186702966690063 2023-01-21 09:33:52.318056: step: 832/529, loss: 0.08430376648902893 2023-01-21 09:33:53.421805: step: 836/529, loss: 0.33013916015625 2023-01-21 09:33:54.537434: step: 840/529, loss: 0.03210580348968506 2023-01-21 09:33:55.711560: step: 844/529, loss: 0.19541558623313904 2023-01-21 09:33:56.837324: step: 848/529, loss: 0.04549083858728409 2023-01-21 09:33:57.948808: step: 852/529, loss: 0.17762994766235352 2023-01-21 09:33:59.067232: step: 856/529, loss: 0.09288106113672256 2023-01-21 09:34:00.181487: step: 860/529, loss: 0.050241902470588684 2023-01-21 09:34:01.288711: step: 864/529, loss: 0.10067228972911835 2023-01-21 09:34:02.411362: step: 868/529, loss: 0.10773982852697372 2023-01-21 09:34:03.499179: step: 872/529, loss: 0.15044693648815155 2023-01-21 09:34:04.629575: step: 876/529, loss: 0.1522907167673111 2023-01-21 09:34:05.752938: step: 880/529, loss: 0.23786316812038422 2023-01-21 09:34:06.885278: step: 884/529, loss: 0.2061445713043213 2023-01-21 09:34:08.036083: step: 888/529, loss: 0.273600697517395 2023-01-21 09:34:09.148097: step: 892/529, loss: 0.052793025970458984 2023-01-21 09:34:10.251421: step: 896/529, loss: 0.04912302643060684 2023-01-21 09:34:11.378499: step: 900/529, loss: 0.11058846116065979 2023-01-21 09:34:12.475556: step: 904/529, loss: 0.0588347464799881 2023-01-21 09:34:13.595978: step: 908/529, loss: 0.05803394317626953 2023-01-21 09:34:14.718818: step: 912/529, loss: 0.060607049614191055 2023-01-21 09:34:15.819084: step: 916/529, loss: 0.12683998048305511 2023-01-21 09:34:16.929696: step: 920/529, loss: 0.42692652344703674 2023-01-21 09:34:18.039800: step: 924/529, loss: 0.13693276047706604 2023-01-21 09:34:19.167003: step: 928/529, loss: 0.20188485085964203 2023-01-21 09:34:20.289958: step: 932/529, loss: 0.865402102470398 2023-01-21 09:34:21.438793: step: 936/529, loss: 0.03904347866773605 2023-01-21 09:34:22.560997: step: 940/529, loss: 0.11379461735486984 2023-01-21 09:34:23.775682: step: 944/529, loss: 0.34854668378829956 2023-01-21 09:34:24.897179: step: 948/529, loss: 0.194636732339859 2023-01-21 09:34:26.007594: step: 952/529, loss: 0.05015420913696289 2023-01-21 09:34:27.125370: step: 956/529, loss: 0.2486804872751236 2023-01-21 09:34:28.241340: step: 960/529, loss: 0.0297896396368742 2023-01-21 09:34:29.347117: step: 964/529, loss: 0.03972931206226349 2023-01-21 09:34:30.480103: step: 968/529, loss: 0.6017562747001648 2023-01-21 09:34:31.600439: step: 972/529, loss: 0.06138205528259277 2023-01-21 09:34:32.734210: step: 976/529, loss: 0.12083926051855087 2023-01-21 09:34:33.865419: step: 980/529, loss: 0.13510942459106445 2023-01-21 09:34:35.002659: step: 984/529, loss: 0.11949028819799423 2023-01-21 09:34:36.131734: step: 988/529, loss: 0.3286727964878082 2023-01-21 09:34:37.246876: step: 992/529, loss: 0.048412345349788666 2023-01-21 09:34:38.384477: step: 996/529, loss: 0.15781421959400177 2023-01-21 09:34:39.473824: step: 1000/529, loss: 0.13084203004837036 2023-01-21 09:34:40.589966: step: 1004/529, loss: 0.1309451162815094 2023-01-21 09:34:41.727966: step: 1008/529, loss: 0.18249750137329102 2023-01-21 09:34:42.852665: step: 1012/529, loss: 0.011563658714294434 2023-01-21 09:34:43.969251: step: 1016/529, loss: 0.4749208688735962 2023-01-21 09:34:45.071870: step: 1020/529, loss: 0.1711321771144867 2023-01-21 09:34:46.189481: step: 1024/529, loss: 0.07321977615356445 2023-01-21 09:34:47.282120: step: 1028/529, loss: 0.1281047761440277 2023-01-21 09:34:48.398953: step: 1032/529, loss: 0.07690820842981339 2023-01-21 09:34:49.500021: step: 1036/529, loss: 0.29134905338287354 2023-01-21 09:34:50.665719: step: 1040/529, loss: 0.06452398002147675 2023-01-21 09:34:51.776719: step: 1044/529, loss: 0.34968137741088867 2023-01-21 09:34:52.923291: step: 1048/529, loss: 0.059043314307928085 2023-01-21 09:34:54.015233: step: 1052/529, loss: 1.9237512350082397 2023-01-21 09:34:55.129408: step: 1056/529, loss: 0.2849012315273285 2023-01-21 09:34:56.232283: step: 1060/529, loss: 0.031221581622958183 2023-01-21 09:34:57.360665: step: 1064/529, loss: 0.11973848193883896 2023-01-21 09:34:58.461460: step: 1068/529, loss: 0.1436946988105774 2023-01-21 09:34:59.587999: step: 1072/529, loss: 0.13366976380348206 2023-01-21 09:35:00.696393: step: 1076/529, loss: 0.11427507549524307 2023-01-21 09:35:01.838081: step: 1080/529, loss: 0.08487396687269211 2023-01-21 09:35:02.994390: step: 1084/529, loss: 0.06237798184156418 2023-01-21 09:35:04.105083: step: 1088/529, loss: 0.3203534185886383 2023-01-21 09:35:05.177332: step: 1092/529, loss: 0.10885315388441086 2023-01-21 09:35:06.307097: step: 1096/529, loss: 0.11889419704675674 2023-01-21 09:35:07.406355: step: 1100/529, loss: 0.10720300674438477 2023-01-21 09:35:08.508011: step: 1104/529, loss: 0.07960281521081924 2023-01-21 09:35:09.612595: step: 1108/529, loss: 0.12990784645080566 2023-01-21 09:35:10.721642: step: 1112/529, loss: 0.08828286826610565 2023-01-21 09:35:11.811065: step: 1116/529, loss: 0.45694124698638916 2023-01-21 09:35:12.924324: step: 1120/529, loss: 0.3834283947944641 2023-01-21 09:35:14.042443: step: 1124/529, loss: 0.38184791803359985 2023-01-21 09:35:15.159775: step: 1128/529, loss: 0.4068889617919922 2023-01-21 09:35:16.277880: step: 1132/529, loss: 0.126484677195549 2023-01-21 09:35:17.397279: step: 1136/529, loss: 0.043662022799253464 2023-01-21 09:35:18.519979: step: 1140/529, loss: 0.5155819654464722 2023-01-21 09:35:19.625971: step: 1144/529, loss: 0.6236297488212585 2023-01-21 09:35:20.753312: step: 1148/529, loss: 0.4206400215625763 2023-01-21 09:35:21.874104: step: 1152/529, loss: 0.45251503586769104 2023-01-21 09:35:23.005511: step: 1156/529, loss: 0.09222922474145889 2023-01-21 09:35:24.148105: step: 1160/529, loss: 0.15893355011940002 2023-01-21 09:35:25.284898: step: 1164/529, loss: 0.2277422845363617 2023-01-21 09:35:26.392335: step: 1168/529, loss: 0.0982811450958252 2023-01-21 09:35:27.490760: step: 1172/529, loss: 0.5409174561500549 2023-01-21 09:35:28.612616: step: 1176/529, loss: 0.14365874230861664 2023-01-21 09:35:29.727126: step: 1180/529, loss: 0.092638298869133 2023-01-21 09:35:30.855530: step: 1184/529, loss: 0.18114447593688965 2023-01-21 09:35:31.967164: step: 1188/529, loss: 0.34872645139694214 2023-01-21 09:35:33.073402: step: 1192/529, loss: 0.08287941664457321 2023-01-21 09:35:34.184085: step: 1196/529, loss: 0.06466036289930344 2023-01-21 09:35:35.312233: step: 1200/529, loss: 0.11960086971521378 2023-01-21 09:35:36.413833: step: 1204/529, loss: 0.07670927047729492 2023-01-21 09:35:37.530246: step: 1208/529, loss: 0.10437555611133575 2023-01-21 09:35:38.644507: step: 1212/529, loss: 0.08500556647777557 2023-01-21 09:35:39.745777: step: 1216/529, loss: 4.565764427185059 2023-01-21 09:35:40.866896: step: 1220/529, loss: 0.11555786430835724 2023-01-21 09:35:42.022599: step: 1224/529, loss: 0.22940626740455627 2023-01-21 09:35:43.152058: step: 1228/529, loss: 0.16571807861328125 2023-01-21 09:35:44.265270: step: 1232/529, loss: 0.24964351952075958 2023-01-21 09:35:45.367984: step: 1236/529, loss: 0.027376480400562286 2023-01-21 09:35:46.536826: step: 1240/529, loss: 0.04574167728424072 2023-01-21 09:35:47.645382: step: 1244/529, loss: 0.2441386878490448 2023-01-21 09:35:48.762668: step: 1248/529, loss: 0.19641664624214172 2023-01-21 09:35:49.892744: step: 1252/529, loss: 0.13524214923381805 2023-01-21 09:35:51.021119: step: 1256/529, loss: 0.17239660024642944 2023-01-21 09:35:52.105494: step: 1260/529, loss: 0.06970057636499405 2023-01-21 09:35:53.251841: step: 1264/529, loss: 0.07716956734657288 2023-01-21 09:35:54.383262: step: 1268/529, loss: 0.0870828628540039 2023-01-21 09:35:55.503686: step: 1272/529, loss: 0.028278542682528496 2023-01-21 09:35:56.639892: step: 1276/529, loss: 0.1458393633365631 2023-01-21 09:35:57.718395: step: 1280/529, loss: 0.15465813875198364 2023-01-21 09:35:58.847963: step: 1284/529, loss: 0.1331310272216797 2023-01-21 09:35:59.949425: step: 1288/529, loss: 0.06034698709845543 2023-01-21 09:36:01.119738: step: 1292/529, loss: 0.10232067108154297 2023-01-21 09:36:02.222633: step: 1296/529, loss: 0.06815081089735031 2023-01-21 09:36:03.330125: step: 1300/529, loss: 0.19847507774829865 2023-01-21 09:36:04.476442: step: 1304/529, loss: 0.8876577615737915 2023-01-21 09:36:05.595053: step: 1308/529, loss: 0.02369384840130806 2023-01-21 09:36:06.695314: step: 1312/529, loss: 0.33368349075317383 2023-01-21 09:36:07.814474: step: 1316/529, loss: 0.14370593428611755 2023-01-21 09:36:08.926982: step: 1320/529, loss: 0.498554527759552 2023-01-21 09:36:10.040732: step: 1324/529, loss: 0.0928216204047203 2023-01-21 09:36:11.143149: step: 1328/529, loss: 0.31891125440597534 2023-01-21 09:36:12.255814: step: 1332/529, loss: 0.6354032158851624 2023-01-21 09:36:13.370645: step: 1336/529, loss: 0.12121353298425674 2023-01-21 09:36:14.522006: step: 1340/529, loss: 0.501778244972229 2023-01-21 09:36:15.603737: step: 1344/529, loss: 0.08527665585279465 2023-01-21 09:36:16.715170: step: 1348/529, loss: 0.04209490120410919 2023-01-21 09:36:17.826794: step: 1352/529, loss: 0.055812835693359375 2023-01-21 09:36:18.922076: step: 1356/529, loss: 0.47979336977005005 2023-01-21 09:36:20.071598: step: 1360/529, loss: 0.08332987129688263 2023-01-21 09:36:21.211408: step: 1364/529, loss: 0.08934378623962402 2023-01-21 09:36:22.315633: step: 1368/529, loss: 0.0654059424996376 2023-01-21 09:36:23.420109: step: 1372/529, loss: 0.09324436634778976 2023-01-21 09:36:24.573742: step: 1376/529, loss: 0.5453853607177734 2023-01-21 09:36:25.705500: step: 1380/529, loss: 0.06243286281824112 2023-01-21 09:36:26.825551: step: 1384/529, loss: 0.09188661724328995 2023-01-21 09:36:27.956342: step: 1388/529, loss: 0.05345487594604492 2023-01-21 09:36:29.085365: step: 1392/529, loss: 0.13730306923389435 2023-01-21 09:36:30.195169: step: 1396/529, loss: 0.11564353108406067 2023-01-21 09:36:31.297416: step: 1400/529, loss: 0.07212846726179123 2023-01-21 09:36:32.392981: step: 1404/529, loss: 0.22219763696193695 2023-01-21 09:36:33.478993: step: 1408/529, loss: 0.01991863176226616 2023-01-21 09:36:34.589675: step: 1412/529, loss: 0.04755754768848419 2023-01-21 09:36:35.698263: step: 1416/529, loss: 0.19270697236061096 2023-01-21 09:36:36.875467: step: 1420/529, loss: 0.07280278205871582 2023-01-21 09:36:37.968010: step: 1424/529, loss: 0.1721513420343399 2023-01-21 09:36:39.078662: step: 1428/529, loss: 0.2575359344482422 2023-01-21 09:36:40.185911: step: 1432/529, loss: 0.05882635340094566 2023-01-21 09:36:41.293495: step: 1436/529, loss: 0.06694427877664566 2023-01-21 09:36:42.407995: step: 1440/529, loss: 0.06245095655322075 2023-01-21 09:36:43.499019: step: 1444/529, loss: 0.07255986332893372 2023-01-21 09:36:44.623375: step: 1448/529, loss: 0.3289529085159302 2023-01-21 09:36:45.748587: step: 1452/529, loss: 0.18068045377731323 2023-01-21 09:36:46.864799: step: 1456/529, loss: 0.0072196004912257195 2023-01-21 09:36:48.012677: step: 1460/529, loss: 0.09402026981115341 2023-01-21 09:36:49.127375: step: 1464/529, loss: 0.08594217896461487 2023-01-21 09:36:50.258413: step: 1468/529, loss: 0.13697758316993713 2023-01-21 09:36:51.355769: step: 1472/529, loss: 0.06967172771692276 2023-01-21 09:36:52.504985: step: 1476/529, loss: 0.06850939244031906 2023-01-21 09:36:53.603064: step: 1480/529, loss: 0.702797532081604 2023-01-21 09:36:54.721982: step: 1484/529, loss: 1.2072794437408447 2023-01-21 09:36:55.839303: step: 1488/529, loss: 0.0856056734919548 2023-01-21 09:36:56.953938: step: 1492/529, loss: 0.11645841598510742 2023-01-21 09:36:58.068933: step: 1496/529, loss: 0.12628121674060822 2023-01-21 09:36:59.179168: step: 1500/529, loss: 0.25913891196250916 2023-01-21 09:37:00.312573: step: 1504/529, loss: 0.16423968970775604 2023-01-21 09:37:01.428954: step: 1508/529, loss: 0.40236568450927734 2023-01-21 09:37:02.523328: step: 1512/529, loss: 0.10830002278089523 2023-01-21 09:37:03.618608: step: 1516/529, loss: 0.008442115969955921 2023-01-21 09:37:04.733532: step: 1520/529, loss: 0.4802786111831665 2023-01-21 09:37:05.883853: step: 1524/529, loss: 0.11132478713989258 2023-01-21 09:37:07.012808: step: 1528/529, loss: 0.12264872342348099 2023-01-21 09:37:08.117900: step: 1532/529, loss: 0.07006797939538956 2023-01-21 09:37:09.247621: step: 1536/529, loss: 0.08137015998363495 2023-01-21 09:37:10.337773: step: 1540/529, loss: 0.07594804465770721 2023-01-21 09:37:11.468478: step: 1544/529, loss: 0.08455229550600052 2023-01-21 09:37:12.579777: step: 1548/529, loss: 0.15987785160541534 2023-01-21 09:37:13.678923: step: 1552/529, loss: 0.1525147259235382 2023-01-21 09:37:14.809850: step: 1556/529, loss: 0.13496610522270203 2023-01-21 09:37:15.926840: step: 1560/529, loss: 0.04631509631872177 2023-01-21 09:37:17.023621: step: 1564/529, loss: 0.0944204330444336 2023-01-21 09:37:18.138134: step: 1568/529, loss: 0.04984412342309952 2023-01-21 09:37:19.261416: step: 1572/529, loss: 0.165119469165802 2023-01-21 09:37:20.403736: step: 1576/529, loss: 0.3083455264568329 2023-01-21 09:37:21.516178: step: 1580/529, loss: 0.08931250125169754 2023-01-21 09:37:22.618672: step: 1584/529, loss: 0.10379371792078018 2023-01-21 09:37:23.740637: step: 1588/529, loss: 1.0492594242095947 2023-01-21 09:37:24.832404: step: 1592/529, loss: 0.0825003832578659 2023-01-21 09:37:25.958624: step: 1596/529, loss: 0.33698076009750366 2023-01-21 09:37:27.100351: step: 1600/529, loss: 0.01838088035583496 2023-01-21 09:37:28.198352: step: 1604/529, loss: 0.06181211397051811 2023-01-21 09:37:29.304128: step: 1608/529, loss: 0.7557794451713562 2023-01-21 09:37:30.425436: step: 1612/529, loss: 0.23803357779979706 2023-01-21 09:37:31.529133: step: 1616/529, loss: 0.1964745968580246 2023-01-21 09:37:32.632082: step: 1620/529, loss: 0.15219077467918396 2023-01-21 09:37:33.761426: step: 1624/529, loss: 0.09088575839996338 2023-01-21 09:37:34.900964: step: 1628/529, loss: 0.08471640944480896 2023-01-21 09:37:36.034504: step: 1632/529, loss: 0.23300760984420776 2023-01-21 09:37:37.141079: step: 1636/529, loss: 0.29435643553733826 2023-01-21 09:37:38.246230: step: 1640/529, loss: 0.09191122651100159 2023-01-21 09:37:39.360658: step: 1644/529, loss: 0.30674517154693604 2023-01-21 09:37:40.477203: step: 1648/529, loss: 0.1423693597316742 2023-01-21 09:37:41.582555: step: 1652/529, loss: 0.07620749622583389 2023-01-21 09:37:42.711632: step: 1656/529, loss: 0.10474634170532227 2023-01-21 09:37:43.834296: step: 1660/529, loss: 0.22713498771190643 2023-01-21 09:37:44.939452: step: 1664/529, loss: 0.23579463362693787 2023-01-21 09:37:46.053163: step: 1668/529, loss: 0.2250600904226303 2023-01-21 09:37:47.165614: step: 1672/529, loss: 0.05981481075286865 2023-01-21 09:37:48.284484: step: 1676/529, loss: 0.10235671699047089 2023-01-21 09:37:49.398588: step: 1680/529, loss: 0.1429756134748459 2023-01-21 09:37:50.501026: step: 1684/529, loss: 0.04197216033935547 2023-01-21 09:37:51.613696: step: 1688/529, loss: 0.26908689737319946 2023-01-21 09:37:52.714052: step: 1692/529, loss: 0.04483480751514435 2023-01-21 09:37:53.844317: step: 1696/529, loss: 0.011022090911865234 2023-01-21 09:37:54.944353: step: 1700/529, loss: 0.12480011582374573 2023-01-21 09:37:56.045374: step: 1704/529, loss: 0.1403886377811432 2023-01-21 09:37:57.170923: step: 1708/529, loss: 0.06724562495946884 2023-01-21 09:37:58.281429: step: 1712/529, loss: 0.3738456964492798 2023-01-21 09:37:59.407524: step: 1716/529, loss: 0.0936737060546875 2023-01-21 09:38:00.551256: step: 1720/529, loss: 0.16844210028648376 2023-01-21 09:38:01.691867: step: 1724/529, loss: 6.272885322570801 2023-01-21 09:38:02.864889: step: 1728/529, loss: 0.034799911081790924 2023-01-21 09:38:04.009026: step: 1732/529, loss: 0.23540247976779938 2023-01-21 09:38:05.154964: step: 1736/529, loss: 0.3427356481552124 2023-01-21 09:38:06.243484: step: 1740/529, loss: 0.24107341468334198 2023-01-21 09:38:07.370411: step: 1744/529, loss: 0.08122486621141434 2023-01-21 09:38:08.544108: step: 1748/529, loss: 0.16982164978981018 2023-01-21 09:38:09.647236: step: 1752/529, loss: 0.09848213195800781 2023-01-21 09:38:10.754286: step: 1756/529, loss: 0.1173684149980545 2023-01-21 09:38:11.867827: step: 1760/529, loss: 0.03551540523767471 2023-01-21 09:38:12.979395: step: 1764/529, loss: 0.6210320591926575 2023-01-21 09:38:14.095917: step: 1768/529, loss: 0.1533488780260086 2023-01-21 09:38:15.227344: step: 1772/529, loss: 0.11132596433162689 2023-01-21 09:38:16.332876: step: 1776/529, loss: 2.3619251251220703 2023-01-21 09:38:17.434811: step: 1780/529, loss: 0.09598284214735031 2023-01-21 09:38:18.566329: step: 1784/529, loss: 0.136785089969635 2023-01-21 09:38:19.679855: step: 1788/529, loss: 0.08897629380226135 2023-01-21 09:38:20.829604: step: 1792/529, loss: 0.2811811566352844 2023-01-21 09:38:21.929588: step: 1796/529, loss: 3.5141055583953857 2023-01-21 09:38:23.067883: step: 1800/529, loss: 0.7922897338867188 2023-01-21 09:38:24.187177: step: 1804/529, loss: 0.06357469409704208 2023-01-21 09:38:25.300317: step: 1808/529, loss: 0.1857217699289322 2023-01-21 09:38:26.411552: step: 1812/529, loss: 0.10023783892393112 2023-01-21 09:38:27.526643: step: 1816/529, loss: 2.0442261695861816 2023-01-21 09:38:28.647916: step: 1820/529, loss: 0.11358185112476349 2023-01-21 09:38:29.781173: step: 1824/529, loss: 0.05978505685925484 2023-01-21 09:38:30.921830: step: 1828/529, loss: 0.13109531998634338 2023-01-21 09:38:32.056426: step: 1832/529, loss: 0.05982523411512375 2023-01-21 09:38:33.178158: step: 1836/529, loss: 0.0687781348824501 2023-01-21 09:38:34.348389: step: 1840/529, loss: 0.6758712530136108 2023-01-21 09:38:35.469427: step: 1844/529, loss: 0.8594818115234375 2023-01-21 09:38:36.595354: step: 1848/529, loss: 0.09919323772192001 2023-01-21 09:38:37.704108: step: 1852/529, loss: 0.12762147188186646 2023-01-21 09:38:38.815767: step: 1856/529, loss: 0.024706363677978516 2023-01-21 09:38:39.947889: step: 1860/529, loss: 0.14579400420188904 2023-01-21 09:38:41.049171: step: 1864/529, loss: 0.07150142639875412 2023-01-21 09:38:42.169891: step: 1868/529, loss: 0.09360170364379883 2023-01-21 09:38:43.280232: step: 1872/529, loss: 0.2526983618736267 2023-01-21 09:38:44.412469: step: 1876/529, loss: 1.3848556280136108 2023-01-21 09:38:45.535844: step: 1880/529, loss: 0.8493753671646118 2023-01-21 09:38:46.654793: step: 1884/529, loss: 0.14791786670684814 2023-01-21 09:38:47.782919: step: 1888/529, loss: 0.11459656059741974 2023-01-21 09:38:48.933541: step: 1892/529, loss: 1.0388320684432983 2023-01-21 09:38:50.052939: step: 1896/529, loss: 0.13002696633338928 2023-01-21 09:38:51.200652: step: 1900/529, loss: 0.5713960528373718 2023-01-21 09:38:52.313886: step: 1904/529, loss: 0.2998400330543518 2023-01-21 09:38:53.454037: step: 1908/529, loss: 0.16257724165916443 2023-01-21 09:38:54.577029: step: 1912/529, loss: 0.14772339165210724 2023-01-21 09:38:55.693560: step: 1916/529, loss: 0.4524449408054352 2023-01-21 09:38:56.794407: step: 1920/529, loss: 0.038282107561826706 2023-01-21 09:38:57.870288: step: 1924/529, loss: 0.015547753311693668 2023-01-21 09:38:58.989592: step: 1928/529, loss: 0.2802380621433258 2023-01-21 09:39:00.127619: step: 1932/529, loss: 0.08170171082019806 2023-01-21 09:39:01.256877: step: 1936/529, loss: 0.10421142727136612 2023-01-21 09:39:02.361788: step: 1940/529, loss: 0.03324277326464653 2023-01-21 09:39:03.498886: step: 1944/529, loss: 0.15926380455493927 2023-01-21 09:39:04.631467: step: 1948/529, loss: 0.19992399215698242 2023-01-21 09:39:05.725491: step: 1952/529, loss: 0.16030746698379517 2023-01-21 09:39:06.827216: step: 1956/529, loss: 0.376537024974823 2023-01-21 09:39:07.950974: step: 1960/529, loss: 0.07121982425451279 2023-01-21 09:39:09.055540: step: 1964/529, loss: 0.03374643251299858 2023-01-21 09:39:10.161902: step: 1968/529, loss: 0.005858802702277899 2023-01-21 09:39:11.319163: step: 1972/529, loss: 0.12500429153442383 2023-01-21 09:39:12.436301: step: 1976/529, loss: 0.397409051656723 2023-01-21 09:39:13.541263: step: 1980/529, loss: 0.06790447235107422 2023-01-21 09:39:14.658397: step: 1984/529, loss: 0.27846452593803406 2023-01-21 09:39:15.756853: step: 1988/529, loss: 0.07734747231006622 2023-01-21 09:39:16.862095: step: 1992/529, loss: 0.05856599658727646 2023-01-21 09:39:18.036545: step: 1996/529, loss: 0.22134943306446075 2023-01-21 09:39:19.126368: step: 2000/529, loss: 0.2235075980424881 2023-01-21 09:39:20.254949: step: 2004/529, loss: 0.054174572229385376 2023-01-21 09:39:21.358970: step: 2008/529, loss: 0.03818502649664879 2023-01-21 09:39:22.444386: step: 2012/529, loss: 0.20997053384780884 2023-01-21 09:39:23.561616: step: 2016/529, loss: 0.01845831796526909 2023-01-21 09:39:24.654577: step: 2020/529, loss: 0.0970921516418457 2023-01-21 09:39:25.745226: step: 2024/529, loss: 0.10060720145702362 2023-01-21 09:39:26.870635: step: 2028/529, loss: 0.07642068713903427 2023-01-21 09:39:27.965167: step: 2032/529, loss: 0.7391047477722168 2023-01-21 09:39:29.070294: step: 2036/529, loss: 0.036525726318359375 2023-01-21 09:39:30.182855: step: 2040/529, loss: 0.977632999420166 2023-01-21 09:39:31.297273: step: 2044/529, loss: 0.15666624903678894 2023-01-21 09:39:32.423738: step: 2048/529, loss: 0.05006761848926544 2023-01-21 09:39:33.517218: step: 2052/529, loss: 0.03367052227258682 2023-01-21 09:39:34.620470: step: 2056/529, loss: 0.0880374014377594 2023-01-21 09:39:35.752998: step: 2060/529, loss: 0.3515944480895996 2023-01-21 09:39:36.866140: step: 2064/529, loss: 0.860822856426239 2023-01-21 09:39:38.027944: step: 2068/529, loss: 0.08030586689710617 2023-01-21 09:39:39.163487: step: 2072/529, loss: 0.03577813878655434 2023-01-21 09:39:40.311857: step: 2076/529, loss: 0.07949074357748032 2023-01-21 09:39:41.430960: step: 2080/529, loss: 0.25775212049484253 2023-01-21 09:39:42.540278: step: 2084/529, loss: 0.7873144149780273 2023-01-21 09:39:43.621461: step: 2088/529, loss: 0.09113284200429916 2023-01-21 09:39:44.738454: step: 2092/529, loss: 0.21279078722000122 2023-01-21 09:39:45.882185: step: 2096/529, loss: 0.6283648610115051 2023-01-21 09:39:46.996403: step: 2100/529, loss: 0.06660356372594833 2023-01-21 09:39:48.108162: step: 2104/529, loss: 0.13064375519752502 2023-01-21 09:39:49.220880: step: 2108/529, loss: 0.05028362572193146 2023-01-21 09:39:50.330348: step: 2112/529, loss: 0.14067357778549194 2023-01-21 09:39:51.438625: step: 2116/529, loss: 0.7217963337898254 ================================================== Loss: 0.233 -------------------- Dev: {'event': {'p': 0.66388557806913, 'r': 0.7416777629826897, 'f1': 0.7006289308176101}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Test: {'event': {'p': 0.6845425867507886, 'r': 0.7347629796839729, 'f1': 0.708764289602613}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Chinese: {'event': {'p': 0.5652173913043478, 'r': 0.7222222222222222, 'f1': 0.6341463414634146}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Korean: {'event': {'p': 0.6170212765957447, 'r': 0.4603174603174603, 'f1': 0.5272727272727272}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Russian: {'event': {'p': 0.3939393939393939, 'r': 0.3611111111111111, 'f1': 0.37681159420289856}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 7 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:40:31.379477: step: 4/529, loss: 0.08535995334386826 2023-01-21 09:40:32.521633: step: 8/529, loss: 0.016001129522919655 2023-01-21 09:40:33.690673: step: 12/529, loss: 0.10105905681848526 2023-01-21 09:40:34.819276: step: 16/529, loss: 0.1079353615641594 2023-01-21 09:40:35.945422: step: 20/529, loss: 0.05083636939525604 2023-01-21 09:40:37.058181: step: 24/529, loss: 0.03182806819677353 2023-01-21 09:40:38.154330: step: 28/529, loss: 0.0721050500869751 2023-01-21 09:40:39.291180: step: 32/529, loss: 0.13523554801940918 2023-01-21 09:40:40.403951: step: 36/529, loss: 0.04191866144537926 2023-01-21 09:40:41.515816: step: 40/529, loss: 0.051964666694402695 2023-01-21 09:40:42.668963: step: 44/529, loss: 0.17937658727169037 2023-01-21 09:40:43.760201: step: 48/529, loss: 0.03992772102355957 2023-01-21 09:40:44.894078: step: 52/529, loss: 0.11854753643274307 2023-01-21 09:40:46.007311: step: 56/529, loss: 0.0499691516160965 2023-01-21 09:40:47.133095: step: 60/529, loss: 0.05157136917114258 2023-01-21 09:40:48.267493: step: 64/529, loss: 0.20484094321727753 2023-01-21 09:40:49.400328: step: 68/529, loss: 0.2316499650478363 2023-01-21 09:40:50.506824: step: 72/529, loss: 0.0871744155883789 2023-01-21 09:40:51.619892: step: 76/529, loss: 0.07359835505485535 2023-01-21 09:40:52.711794: step: 80/529, loss: 0.3921722173690796 2023-01-21 09:40:53.813081: step: 84/529, loss: 0.1913851648569107 2023-01-21 09:40:54.957778: step: 88/529, loss: 0.06026472896337509 2023-01-21 09:40:56.058798: step: 92/529, loss: 0.09796439111232758 2023-01-21 09:40:57.141743: step: 96/529, loss: 0.03011322021484375 2023-01-21 09:40:58.299340: step: 100/529, loss: 0.32214635610580444 2023-01-21 09:40:59.400854: step: 104/529, loss: 0.11700396984815598 2023-01-21 09:41:00.530117: step: 108/529, loss: 0.04085185378789902 2023-01-21 09:41:01.677659: step: 112/529, loss: 0.0914732962846756 2023-01-21 09:41:02.792004: step: 116/529, loss: 0.04465823248028755 2023-01-21 09:41:03.918987: step: 120/529, loss: 0.13958540558815002 2023-01-21 09:41:05.057036: step: 124/529, loss: 0.07259884476661682 2023-01-21 09:41:06.184555: step: 128/529, loss: 0.08847961574792862 2023-01-21 09:41:07.308287: step: 132/529, loss: 0.08840484917163849 2023-01-21 09:41:08.427239: step: 136/529, loss: 0.1915014237165451 2023-01-21 09:41:09.545990: step: 140/529, loss: 0.12636291980743408 2023-01-21 09:41:10.663279: step: 144/529, loss: 0.08276939392089844 2023-01-21 09:41:11.762268: step: 148/529, loss: 0.03695201873779297 2023-01-21 09:41:12.885286: step: 152/529, loss: 0.07858090847730637 2023-01-21 09:41:14.030964: step: 156/529, loss: 0.06089286878705025 2023-01-21 09:41:15.164136: step: 160/529, loss: 0.12468412518501282 2023-01-21 09:41:16.274229: step: 164/529, loss: 0.04841804504394531 2023-01-21 09:41:17.407823: step: 168/529, loss: 0.19520455598831177 2023-01-21 09:41:18.502154: step: 172/529, loss: 0.042046450078487396 2023-01-21 09:41:19.643414: step: 176/529, loss: 0.10202236473560333 2023-01-21 09:41:20.764770: step: 180/529, loss: 0.026439856737852097 2023-01-21 09:41:21.917682: step: 184/529, loss: 0.013095999136567116 2023-01-21 09:41:23.040923: step: 188/529, loss: 0.0721866562962532 2023-01-21 09:41:24.154940: step: 192/529, loss: 0.13443756103515625 2023-01-21 09:41:25.265675: step: 196/529, loss: 0.005830669775605202 2023-01-21 09:41:26.354867: step: 200/529, loss: 0.03621787950396538 2023-01-21 09:41:27.482642: step: 204/529, loss: 0.019101334735751152 2023-01-21 09:41:28.582150: step: 208/529, loss: 0.18747597932815552 2023-01-21 09:41:29.681871: step: 212/529, loss: 0.0082693574950099 2023-01-21 09:41:30.802479: step: 216/529, loss: 0.059357643127441406 2023-01-21 09:41:31.963802: step: 220/529, loss: 0.1062961146235466 2023-01-21 09:41:33.067035: step: 224/529, loss: 0.13642655313014984 2023-01-21 09:41:34.198869: step: 228/529, loss: 0.12942571938037872 2023-01-21 09:41:35.325418: step: 232/529, loss: 0.010764027014374733 2023-01-21 09:41:36.437254: step: 236/529, loss: 0.022208143025636673 2023-01-21 09:41:37.522907: step: 240/529, loss: 0.19263982772827148 2023-01-21 09:41:38.654211: step: 244/529, loss: 0.02089548110961914 2023-01-21 09:41:39.755454: step: 248/529, loss: 0.0337187759578228 2023-01-21 09:41:40.853496: step: 252/529, loss: 1.025361180305481 2023-01-21 09:41:41.958249: step: 256/529, loss: 0.0693356990814209 2023-01-21 09:41:43.091176: step: 260/529, loss: 0.12926062941551208 2023-01-21 09:41:44.203842: step: 264/529, loss: 0.10652542114257812 2023-01-21 09:41:45.325926: step: 268/529, loss: 0.13642553985118866 2023-01-21 09:41:46.448929: step: 272/529, loss: 0.1928671896457672 2023-01-21 09:41:47.555883: step: 276/529, loss: 0.0912637710571289 2023-01-21 09:41:48.686786: step: 280/529, loss: 0.0898033156991005 2023-01-21 09:41:49.793597: step: 284/529, loss: 0.07509009540081024 2023-01-21 09:41:50.939711: step: 288/529, loss: 0.04311618581414223 2023-01-21 09:41:52.029734: step: 292/529, loss: 0.17369318008422852 2023-01-21 09:41:53.138711: step: 296/529, loss: 0.09234333783388138 2023-01-21 09:41:54.238196: step: 300/529, loss: 0.19594478607177734 2023-01-21 09:41:55.350920: step: 304/529, loss: 0.04409150779247284 2023-01-21 09:41:56.487128: step: 308/529, loss: 0.047379374504089355 2023-01-21 09:41:57.615853: step: 312/529, loss: 0.17639216780662537 2023-01-21 09:41:58.759756: step: 316/529, loss: 0.05121898651123047 2023-01-21 09:41:59.867738: step: 320/529, loss: 0.7693721652030945 2023-01-21 09:42:00.982472: step: 324/529, loss: 0.35669606924057007 2023-01-21 09:42:02.124474: step: 328/529, loss: 0.38664036989212036 2023-01-21 09:42:03.214909: step: 332/529, loss: 0.0558900348842144 2023-01-21 09:42:04.332533: step: 336/529, loss: 0.1913757175207138 2023-01-21 09:42:05.417909: step: 340/529, loss: 0.11070023477077484 2023-01-21 09:42:06.529797: step: 344/529, loss: 0.039197541773319244 2023-01-21 09:42:07.620198: step: 348/529, loss: 0.1517617255449295 2023-01-21 09:42:08.765424: step: 352/529, loss: 0.13259035348892212 2023-01-21 09:42:09.860266: step: 356/529, loss: 0.052670955657958984 2023-01-21 09:42:10.960785: step: 360/529, loss: 0.2514592111110687 2023-01-21 09:42:12.084283: step: 364/529, loss: 0.16890506446361542 2023-01-21 09:42:13.202632: step: 368/529, loss: 0.06347131729125977 2023-01-21 09:42:14.320686: step: 372/529, loss: 0.4140257239341736 2023-01-21 09:42:15.454590: step: 376/529, loss: 0.3851228952407837 2023-01-21 09:42:16.557983: step: 380/529, loss: 0.03889770805835724 2023-01-21 09:42:17.653455: step: 384/529, loss: 0.16113929450511932 2023-01-21 09:42:18.778359: step: 388/529, loss: 0.06517477333545685 2023-01-21 09:42:19.882857: step: 392/529, loss: 0.035893917083740234 2023-01-21 09:42:21.015209: step: 396/529, loss: 0.04008650779724121 2023-01-21 09:42:22.129107: step: 400/529, loss: 0.21062499284744263 2023-01-21 09:42:23.233689: step: 404/529, loss: 0.0751979798078537 2023-01-21 09:42:24.364904: step: 408/529, loss: 0.17072072625160217 2023-01-21 09:42:25.469295: step: 412/529, loss: 0.13009461760520935 2023-01-21 09:42:26.605535: step: 416/529, loss: 0.16932764649391174 2023-01-21 09:42:27.735461: step: 420/529, loss: 0.23037000000476837 2023-01-21 09:42:28.858844: step: 424/529, loss: 0.0965181365609169 2023-01-21 09:42:29.954971: step: 428/529, loss: 0.07707658410072327 2023-01-21 09:42:31.054527: step: 432/529, loss: 0.08312736451625824 2023-01-21 09:42:32.144465: step: 436/529, loss: 0.07003231346607208 2023-01-21 09:42:33.296219: step: 440/529, loss: 0.025251388549804688 2023-01-21 09:42:34.446537: step: 444/529, loss: 0.05140228196978569 2023-01-21 09:42:35.548992: step: 448/529, loss: 0.11442232131958008 2023-01-21 09:42:36.678813: step: 452/529, loss: 0.10901746898889542 2023-01-21 09:42:37.823042: step: 456/529, loss: 0.1073119193315506 2023-01-21 09:42:38.970805: step: 460/529, loss: 0.25079044699668884 2023-01-21 09:42:40.085043: step: 464/529, loss: 0.31729984283447266 2023-01-21 09:42:41.202303: step: 468/529, loss: 0.27446651458740234 2023-01-21 09:42:42.307616: step: 472/529, loss: 0.22311820089817047 2023-01-21 09:42:43.428149: step: 476/529, loss: 0.1324368268251419 2023-01-21 09:42:44.527682: step: 480/529, loss: 0.02860879898071289 2023-01-21 09:42:45.617534: step: 484/529, loss: 0.10911176353693008 2023-01-21 09:42:46.731377: step: 488/529, loss: 0.6487019062042236 2023-01-21 09:42:47.874661: step: 492/529, loss: 0.10456905514001846 2023-01-21 09:42:48.988061: step: 496/529, loss: 0.10264454036951065 2023-01-21 09:42:50.094393: step: 500/529, loss: 0.17587712407112122 2023-01-21 09:42:51.234002: step: 504/529, loss: 0.38415825366973877 2023-01-21 09:42:52.345618: step: 508/529, loss: 0.14278526604175568 2023-01-21 09:42:53.434096: step: 512/529, loss: 0.876387357711792 2023-01-21 09:42:54.552912: step: 516/529, loss: 0.02136249653995037 2023-01-21 09:42:55.670921: step: 520/529, loss: 0.30383986234664917 2023-01-21 09:42:56.799973: step: 524/529, loss: 0.06644754856824875 2023-01-21 09:42:57.920869: step: 528/529, loss: 0.05408506467938423 2023-01-21 09:42:59.030758: step: 532/529, loss: 0.18358105421066284 2023-01-21 09:43:00.126539: step: 536/529, loss: 0.43189141154289246 2023-01-21 09:43:01.240885: step: 540/529, loss: 0.1558370590209961 2023-01-21 09:43:02.367519: step: 544/529, loss: 0.02861003950238228 2023-01-21 09:43:03.453163: step: 548/529, loss: 0.1691078096628189 2023-01-21 09:43:04.565371: step: 552/529, loss: 0.08777222037315369 2023-01-21 09:43:05.701906: step: 556/529, loss: 0.18570689857006073 2023-01-21 09:43:06.807787: step: 560/529, loss: 0.1552501618862152 2023-01-21 09:43:07.920983: step: 564/529, loss: 0.12489528208971024 2023-01-21 09:43:09.032673: step: 568/529, loss: 0.06810909509658813 2023-01-21 09:43:10.118114: step: 572/529, loss: 0.08333795517683029 2023-01-21 09:43:11.232975: step: 576/529, loss: 0.0747753158211708 2023-01-21 09:43:12.331052: step: 580/529, loss: 0.1213603988289833 2023-01-21 09:43:13.455869: step: 584/529, loss: 0.10235204547643661 2023-01-21 09:43:14.540824: step: 588/529, loss: 0.21795593202114105 2023-01-21 09:43:15.659242: step: 592/529, loss: 0.2904343605041504 2023-01-21 09:43:16.772797: step: 596/529, loss: 0.02508638985455036 2023-01-21 09:43:17.853660: step: 600/529, loss: 0.03851909935474396 2023-01-21 09:43:18.968573: step: 604/529, loss: 0.19919902086257935 2023-01-21 09:43:20.114282: step: 608/529, loss: 0.5803912281990051 2023-01-21 09:43:21.237066: step: 612/529, loss: 0.16329871118068695 2023-01-21 09:43:22.395752: step: 616/529, loss: 0.16156300902366638 2023-01-21 09:43:23.501312: step: 620/529, loss: 0.4201776683330536 2023-01-21 09:43:24.630632: step: 624/529, loss: 0.2616937756538391 2023-01-21 09:43:25.759563: step: 628/529, loss: 0.08690319210290909 2023-01-21 09:43:26.884241: step: 632/529, loss: 0.09721431881189346 2023-01-21 09:43:28.030793: step: 636/529, loss: 0.0719393715262413 2023-01-21 09:43:29.139261: step: 640/529, loss: 0.07727108150720596 2023-01-21 09:43:30.234170: step: 644/529, loss: 0.02064523659646511 2023-01-21 09:43:31.368988: step: 648/529, loss: 0.06515903770923615 2023-01-21 09:43:32.476201: step: 652/529, loss: 0.032294370234012604 2023-01-21 09:43:33.621449: step: 656/529, loss: 0.3659224510192871 2023-01-21 09:43:34.715118: step: 660/529, loss: 0.06363248825073242 2023-01-21 09:43:35.821160: step: 664/529, loss: 0.1395689994096756 2023-01-21 09:43:36.953442: step: 668/529, loss: 0.1669834703207016 2023-01-21 09:43:38.066375: step: 672/529, loss: 0.07678094506263733 2023-01-21 09:43:39.204021: step: 676/529, loss: 0.18050357699394226 2023-01-21 09:43:40.309237: step: 680/529, loss: 0.03705973923206329 2023-01-21 09:43:41.454595: step: 684/529, loss: 0.20744390785694122 2023-01-21 09:43:42.570219: step: 688/529, loss: 0.11302336305379868 2023-01-21 09:43:43.705115: step: 692/529, loss: 0.0552794486284256 2023-01-21 09:43:44.855425: step: 696/529, loss: 0.10831566154956818 2023-01-21 09:43:45.978110: step: 700/529, loss: 0.14847736060619354 2023-01-21 09:43:47.107285: step: 704/529, loss: 0.10636377334594727 2023-01-21 09:43:48.221628: step: 708/529, loss: 0.10097765922546387 2023-01-21 09:43:49.364876: step: 712/529, loss: 0.14710339903831482 2023-01-21 09:43:50.501755: step: 716/529, loss: 0.04197302088141441 2023-01-21 09:43:51.610954: step: 720/529, loss: 0.35984688997268677 2023-01-21 09:43:52.712906: step: 724/529, loss: 1.2314952611923218 2023-01-21 09:43:53.836607: step: 728/529, loss: 0.08138785511255264 2023-01-21 09:43:54.944896: step: 732/529, loss: 0.13064813613891602 2023-01-21 09:43:56.062515: step: 736/529, loss: 0.0573970302939415 2023-01-21 09:43:57.178888: step: 740/529, loss: 0.08001823723316193 2023-01-21 09:43:58.319897: step: 744/529, loss: 0.14377526938915253 2023-01-21 09:43:59.462583: step: 748/529, loss: 0.12418365478515625 2023-01-21 09:44:00.625811: step: 752/529, loss: 0.037191201001405716 2023-01-21 09:44:01.714685: step: 756/529, loss: 0.07091245800256729 2023-01-21 09:44:02.824612: step: 760/529, loss: 0.1054040938615799 2023-01-21 09:44:03.961140: step: 764/529, loss: 0.0629824697971344 2023-01-21 09:44:05.058396: step: 768/529, loss: 0.09661710262298584 2023-01-21 09:44:06.187221: step: 772/529, loss: 0.10616204887628555 2023-01-21 09:44:07.277712: step: 776/529, loss: 0.05436534807085991 2023-01-21 09:44:08.412240: step: 780/529, loss: 0.08986649662256241 2023-01-21 09:44:09.523195: step: 784/529, loss: 0.15237101912498474 2023-01-21 09:44:10.661114: step: 788/529, loss: 0.03706107288599014 2023-01-21 09:44:11.752859: step: 792/529, loss: 0.06959257274866104 2023-01-21 09:44:12.872102: step: 796/529, loss: 0.041742704808712006 2023-01-21 09:44:13.992303: step: 800/529, loss: 0.08936090767383575 2023-01-21 09:44:15.142290: step: 804/529, loss: 0.16828516125679016 2023-01-21 09:44:16.279002: step: 808/529, loss: 0.09065123647451401 2023-01-21 09:44:17.392827: step: 812/529, loss: 0.12946276366710663 2023-01-21 09:44:18.485156: step: 816/529, loss: 0.05584364011883736 2023-01-21 09:44:19.589135: step: 820/529, loss: 0.49276256561279297 2023-01-21 09:44:20.685962: step: 824/529, loss: 0.11095300316810608 2023-01-21 09:44:21.812141: step: 828/529, loss: 0.06222191080451012 2023-01-21 09:44:22.926482: step: 832/529, loss: 0.07886314392089844 2023-01-21 09:44:24.032679: step: 836/529, loss: 0.1289811134338379 2023-01-21 09:44:25.111958: step: 840/529, loss: 0.05126838758587837 2023-01-21 09:44:26.213587: step: 844/529, loss: 0.12138824164867401 2023-01-21 09:44:27.309591: step: 848/529, loss: 0.0704202651977539 2023-01-21 09:44:28.418559: step: 852/529, loss: 0.03680281713604927 2023-01-21 09:44:29.501737: step: 856/529, loss: 0.007145881652832031 2023-01-21 09:44:30.631241: step: 860/529, loss: 0.06799888610839844 2023-01-21 09:44:31.748782: step: 864/529, loss: 0.0625530257821083 2023-01-21 09:44:32.839006: step: 868/529, loss: 0.2909412384033203 2023-01-21 09:44:33.961814: step: 872/529, loss: 0.1304069459438324 2023-01-21 09:44:35.073987: step: 876/529, loss: 0.14396364986896515 2023-01-21 09:44:36.174357: step: 880/529, loss: 0.38703393936157227 2023-01-21 09:44:37.291926: step: 884/529, loss: 0.12164366245269775 2023-01-21 09:44:38.426065: step: 888/529, loss: 0.07425308227539062 2023-01-21 09:44:39.546596: step: 892/529, loss: 0.35207778215408325 2023-01-21 09:44:40.680299: step: 896/529, loss: 0.16876216232776642 2023-01-21 09:44:41.796517: step: 900/529, loss: 0.08585414290428162 2023-01-21 09:44:42.940623: step: 904/529, loss: 0.08714637905359268 2023-01-21 09:44:44.052169: step: 908/529, loss: 0.13681745529174805 2023-01-21 09:44:45.168969: step: 912/529, loss: 0.06272868812084198 2023-01-21 09:44:46.349915: step: 916/529, loss: 0.0487370528280735 2023-01-21 09:44:47.516006: step: 920/529, loss: 0.17080570757389069 2023-01-21 09:44:48.677065: step: 924/529, loss: 0.06933536380529404 2023-01-21 09:44:49.789466: step: 928/529, loss: 0.0676342025399208 2023-01-21 09:44:50.922402: step: 932/529, loss: 0.27520352602005005 2023-01-21 09:44:52.045985: step: 936/529, loss: 0.1539454460144043 2023-01-21 09:44:53.158036: step: 940/529, loss: 1.416111707687378 2023-01-21 09:44:54.291641: step: 944/529, loss: 0.1712416708469391 2023-01-21 09:44:55.400639: step: 948/529, loss: 0.1625732034444809 2023-01-21 09:44:56.523360: step: 952/529, loss: 0.08752937614917755 2023-01-21 09:44:57.632124: step: 956/529, loss: 0.23820781707763672 2023-01-21 09:44:58.765447: step: 960/529, loss: 0.06130237877368927 2023-01-21 09:44:59.868111: step: 964/529, loss: 0.22669734060764313 2023-01-21 09:45:01.017038: step: 968/529, loss: 0.09415054321289062 2023-01-21 09:45:02.124937: step: 972/529, loss: 0.02753582037985325 2023-01-21 09:45:03.253023: step: 976/529, loss: 0.22043377161026 2023-01-21 09:45:04.374083: step: 980/529, loss: 0.06653241813182831 2023-01-21 09:45:05.495761: step: 984/529, loss: 0.15861931443214417 2023-01-21 09:45:06.612570: step: 988/529, loss: 0.1564960777759552 2023-01-21 09:45:07.768314: step: 992/529, loss: 0.060117244720458984 2023-01-21 09:45:08.891789: step: 996/529, loss: 0.21202115714550018 2023-01-21 09:45:10.023272: step: 1000/529, loss: 0.1770804524421692 2023-01-21 09:45:11.158343: step: 1004/529, loss: 0.17375174164772034 2023-01-21 09:45:12.312061: step: 1008/529, loss: 0.38376694917678833 2023-01-21 09:45:13.450484: step: 1012/529, loss: 0.04024467617273331 2023-01-21 09:45:14.544971: step: 1016/529, loss: 0.15649671852588654 2023-01-21 09:45:15.634970: step: 1020/529, loss: 0.07410085201263428 2023-01-21 09:45:16.768499: step: 1024/529, loss: 0.08371081948280334 2023-01-21 09:45:17.874967: step: 1028/529, loss: 0.43277493119239807 2023-01-21 09:45:18.981723: step: 1032/529, loss: 0.11992263793945312 2023-01-21 09:45:20.087366: step: 1036/529, loss: 0.12016751617193222 2023-01-21 09:45:21.228842: step: 1040/529, loss: 0.12992163002490997 2023-01-21 09:45:22.347977: step: 1044/529, loss: 0.025912238284945488 2023-01-21 09:45:23.445054: step: 1048/529, loss: 0.5583709478378296 2023-01-21 09:45:24.573520: step: 1052/529, loss: 0.14327797293663025 2023-01-21 09:45:25.685409: step: 1056/529, loss: 0.1275242418050766 2023-01-21 09:45:26.795713: step: 1060/529, loss: 0.15893106162548065 2023-01-21 09:45:27.905341: step: 1064/529, loss: 0.1193297952413559 2023-01-21 09:45:29.027117: step: 1068/529, loss: 0.054459381848573685 2023-01-21 09:45:30.133283: step: 1072/529, loss: 0.17819052934646606 2023-01-21 09:45:31.247961: step: 1076/529, loss: 0.15902462601661682 2023-01-21 09:45:32.384656: step: 1080/529, loss: 0.14294835925102234 2023-01-21 09:45:33.493265: step: 1084/529, loss: 0.05848880112171173 2023-01-21 09:45:34.614568: step: 1088/529, loss: 0.10031929612159729 2023-01-21 09:45:35.708838: step: 1092/529, loss: 0.31642967462539673 2023-01-21 09:45:36.809650: step: 1096/529, loss: 0.2695261538028717 2023-01-21 09:45:37.903653: step: 1100/529, loss: 0.2811482548713684 2023-01-21 09:45:39.024937: step: 1104/529, loss: 0.05957643687725067 2023-01-21 09:45:40.159508: step: 1108/529, loss: 0.062278151512145996 2023-01-21 09:45:41.303662: step: 1112/529, loss: 0.021506119519472122 2023-01-21 09:45:42.409971: step: 1116/529, loss: 0.7041778564453125 2023-01-21 09:45:43.519035: step: 1120/529, loss: 0.09198732674121857 2023-01-21 09:45:44.631774: step: 1124/529, loss: 0.08571124076843262 2023-01-21 09:45:45.719971: step: 1128/529, loss: 0.06545896828174591 2023-01-21 09:45:46.839490: step: 1132/529, loss: 0.0499020554125309 2023-01-21 09:45:47.944472: step: 1136/529, loss: 0.04824574291706085 2023-01-21 09:45:49.078854: step: 1140/529, loss: 0.04275684431195259 2023-01-21 09:45:50.194265: step: 1144/529, loss: 0.07564029842615128 2023-01-21 09:45:51.286715: step: 1148/529, loss: 0.05088425055146217 2023-01-21 09:45:52.409721: step: 1152/529, loss: 0.08161468803882599 2023-01-21 09:45:53.533343: step: 1156/529, loss: 0.16641655564308167 2023-01-21 09:45:54.656348: step: 1160/529, loss: 0.08168898522853851 2023-01-21 09:45:55.774320: step: 1164/529, loss: 0.19990329444408417 2023-01-21 09:45:56.889591: step: 1168/529, loss: 0.07986479252576828 2023-01-21 09:45:57.995946: step: 1172/529, loss: 0.15243932604789734 2023-01-21 09:45:59.110655: step: 1176/529, loss: 0.055414773523807526 2023-01-21 09:46:00.225564: step: 1180/529, loss: 0.10719247162342072 2023-01-21 09:46:01.349513: step: 1184/529, loss: 0.011735511012375355 2023-01-21 09:46:02.488810: step: 1188/529, loss: 0.2419900894165039 2023-01-21 09:46:03.594261: step: 1192/529, loss: 0.06695299595594406 2023-01-21 09:46:04.702769: step: 1196/529, loss: 0.6303110718727112 2023-01-21 09:46:05.822285: step: 1200/529, loss: 0.11464748531579971 2023-01-21 09:46:06.975103: step: 1204/529, loss: 0.6920454502105713 2023-01-21 09:46:08.087029: step: 1208/529, loss: 0.05030231550335884 2023-01-21 09:46:09.203890: step: 1212/529, loss: 0.12416372448205948 2023-01-21 09:46:10.329336: step: 1216/529, loss: 0.03824073076248169 2023-01-21 09:46:11.441195: step: 1220/529, loss: 0.3069017231464386 2023-01-21 09:46:12.542163: step: 1224/529, loss: 0.18959569931030273 2023-01-21 09:46:13.682363: step: 1228/529, loss: 0.08258872479200363 2023-01-21 09:46:14.794377: step: 1232/529, loss: 0.3074410557746887 2023-01-21 09:46:15.906791: step: 1236/529, loss: 0.2622811198234558 2023-01-21 09:46:17.052463: step: 1240/529, loss: 0.12124772369861603 2023-01-21 09:46:18.156146: step: 1244/529, loss: 0.0762840211391449 2023-01-21 09:46:19.240222: step: 1248/529, loss: 0.042694661766290665 2023-01-21 09:46:20.336987: step: 1252/529, loss: 0.12141668796539307 2023-01-21 09:46:21.506037: step: 1256/529, loss: 0.07203512638807297 2023-01-21 09:46:22.626457: step: 1260/529, loss: 0.9234479069709778 2023-01-21 09:46:23.728239: step: 1264/529, loss: 0.03592194616794586 2023-01-21 09:46:24.836106: step: 1268/529, loss: 0.016762543469667435 2023-01-21 09:46:25.949068: step: 1272/529, loss: 0.11272773891687393 2023-01-21 09:46:27.065399: step: 1276/529, loss: 0.10014477372169495 2023-01-21 09:46:28.192669: step: 1280/529, loss: 0.07927493751049042 2023-01-21 09:46:29.313233: step: 1284/529, loss: 0.18116407096385956 2023-01-21 09:46:30.425157: step: 1288/529, loss: 0.07261667400598526 2023-01-21 09:46:31.565573: step: 1292/529, loss: 0.12926602363586426 2023-01-21 09:46:32.698676: step: 1296/529, loss: 0.08017349243164062 2023-01-21 09:46:33.838885: step: 1300/529, loss: 0.5795788168907166 2023-01-21 09:46:34.954096: step: 1304/529, loss: 0.11471805721521378 2023-01-21 09:46:36.073367: step: 1308/529, loss: 0.6441377997398376 2023-01-21 09:46:37.193860: step: 1312/529, loss: 0.9785250425338745 2023-01-21 09:46:38.323867: step: 1316/529, loss: 0.15522976219654083 2023-01-21 09:46:39.422800: step: 1320/529, loss: 0.1446654349565506 2023-01-21 09:46:40.524731: step: 1324/529, loss: 0.07780542224645615 2023-01-21 09:46:41.633968: step: 1328/529, loss: 0.04367389902472496 2023-01-21 09:46:42.742005: step: 1332/529, loss: 0.1036238744854927 2023-01-21 09:46:43.894616: step: 1336/529, loss: 0.14732956886291504 2023-01-21 09:46:45.031767: step: 1340/529, loss: 0.10989589989185333 2023-01-21 09:46:46.166159: step: 1344/529, loss: 0.06113891676068306 2023-01-21 09:46:47.255056: step: 1348/529, loss: 0.316057026386261 2023-01-21 09:46:48.335622: step: 1352/529, loss: 0.11345696449279785 2023-01-21 09:46:49.468195: step: 1356/529, loss: 0.03488645702600479 2023-01-21 09:46:50.589202: step: 1360/529, loss: 0.14009495079517365 2023-01-21 09:46:51.707420: step: 1364/529, loss: 0.20713704824447632 2023-01-21 09:46:52.842988: step: 1368/529, loss: 0.03848094865679741 2023-01-21 09:46:53.965179: step: 1372/529, loss: 0.09145593643188477 2023-01-21 09:46:55.113003: step: 1376/529, loss: 0.42008477449417114 2023-01-21 09:46:56.247476: step: 1380/529, loss: 0.04385938495397568 2023-01-21 09:46:57.386132: step: 1384/529, loss: 0.21850094199180603 2023-01-21 09:46:58.501156: step: 1388/529, loss: 0.14133116602897644 2023-01-21 09:46:59.622185: step: 1392/529, loss: 0.14615249633789062 2023-01-21 09:47:00.714479: step: 1396/529, loss: 0.09533253312110901 2023-01-21 09:47:01.836234: step: 1400/529, loss: 0.0530826561152935 2023-01-21 09:47:02.940488: step: 1404/529, loss: 0.050835803151130676 2023-01-21 09:47:04.051442: step: 1408/529, loss: 0.8031697869300842 2023-01-21 09:47:05.152619: step: 1412/529, loss: 0.0764227882027626 2023-01-21 09:47:06.282265: step: 1416/529, loss: 0.17257362604141235 2023-01-21 09:47:07.412059: step: 1420/529, loss: 0.34137916564941406 2023-01-21 09:47:08.507402: step: 1424/529, loss: 0.02362356148660183 2023-01-21 09:47:09.675464: step: 1428/529, loss: 0.07728901505470276 2023-01-21 09:47:10.775271: step: 1432/529, loss: 1.083054542541504 2023-01-21 09:47:11.921056: step: 1436/529, loss: 0.3218139708042145 2023-01-21 09:47:13.019853: step: 1440/529, loss: 0.08511407673358917 2023-01-21 09:47:14.129513: step: 1444/529, loss: 0.029111862182617188 2023-01-21 09:47:15.223382: step: 1448/529, loss: 0.06409845501184464 2023-01-21 09:47:16.348046: step: 1452/529, loss: 0.028685474768280983 2023-01-21 09:47:17.492893: step: 1456/529, loss: 0.25656455755233765 2023-01-21 09:47:18.626442: step: 1460/529, loss: 0.1137796938419342 2023-01-21 09:47:19.733382: step: 1464/529, loss: 0.02136397361755371 2023-01-21 09:47:20.848543: step: 1468/529, loss: 0.04566369205713272 2023-01-21 09:47:21.950639: step: 1472/529, loss: 0.08377714455127716 2023-01-21 09:47:23.061035: step: 1476/529, loss: 0.10929107666015625 2023-01-21 09:47:24.181337: step: 1480/529, loss: 0.034664154052734375 2023-01-21 09:47:25.326010: step: 1484/529, loss: 0.06072532385587692 2023-01-21 09:47:26.442106: step: 1488/529, loss: 0.15987873077392578 2023-01-21 09:47:27.564360: step: 1492/529, loss: 0.07636241614818573 2023-01-21 09:47:28.696006: step: 1496/529, loss: 0.22534483671188354 2023-01-21 09:47:29.822643: step: 1500/529, loss: 0.05913911014795303 2023-01-21 09:47:30.919394: step: 1504/529, loss: 0.040613315999507904 2023-01-21 09:47:32.041737: step: 1508/529, loss: 0.005683040712028742 2023-01-21 09:47:33.167956: step: 1512/529, loss: 0.04234590381383896 2023-01-21 09:47:34.285126: step: 1516/529, loss: 0.08957324177026749 2023-01-21 09:47:35.368206: step: 1520/529, loss: 0.10978803783655167 2023-01-21 09:47:36.492551: step: 1524/529, loss: 0.13055287301540375 2023-01-21 09:47:37.619260: step: 1528/529, loss: 0.7986448407173157 2023-01-21 09:47:38.768143: step: 1532/529, loss: 0.06465015560388565 2023-01-21 09:47:39.872896: step: 1536/529, loss: 0.1254071295261383 2023-01-21 09:47:40.989596: step: 1540/529, loss: 0.15743064880371094 2023-01-21 09:47:42.104199: step: 1544/529, loss: 0.02288849465548992 2023-01-21 09:47:43.227427: step: 1548/529, loss: 0.27756041288375854 2023-01-21 09:47:44.357967: step: 1552/529, loss: 1.0142234563827515 2023-01-21 09:47:45.453650: step: 1556/529, loss: 0.24193666875362396 2023-01-21 09:47:46.564169: step: 1560/529, loss: 0.05651231110095978 2023-01-21 09:47:47.703943: step: 1564/529, loss: 0.2549936771392822 2023-01-21 09:47:48.837111: step: 1568/529, loss: 0.1243053525686264 2023-01-21 09:47:49.951549: step: 1572/529, loss: 0.3023591935634613 2023-01-21 09:47:51.102598: step: 1576/529, loss: 0.13557367026805878 2023-01-21 09:47:52.239911: step: 1580/529, loss: 0.10327358543872833 2023-01-21 09:47:53.374318: step: 1584/529, loss: 0.042285822331905365 2023-01-21 09:47:54.484266: step: 1588/529, loss: 0.05124945566058159 2023-01-21 09:47:55.643601: step: 1592/529, loss: 0.24578005075454712 2023-01-21 09:47:56.748585: step: 1596/529, loss: 0.002261877292767167 2023-01-21 09:47:57.886544: step: 1600/529, loss: 0.12609444558620453 2023-01-21 09:47:59.042150: step: 1604/529, loss: 0.032897043973207474 2023-01-21 09:48:00.150165: step: 1608/529, loss: 0.23663026094436646 2023-01-21 09:48:01.246652: step: 1612/529, loss: 0.03747249022126198 2023-01-21 09:48:02.371428: step: 1616/529, loss: 0.326956570148468 2023-01-21 09:48:03.516754: step: 1620/529, loss: 0.24182011187076569 2023-01-21 09:48:04.657111: step: 1624/529, loss: 0.11921420693397522 2023-01-21 09:48:05.769412: step: 1628/529, loss: 0.23733974993228912 2023-01-21 09:48:06.891965: step: 1632/529, loss: 0.1739656925201416 2023-01-21 09:48:08.006556: step: 1636/529, loss: 0.1930617392063141 2023-01-21 09:48:09.127755: step: 1640/529, loss: 0.13622340559959412 2023-01-21 09:48:10.247122: step: 1644/529, loss: 0.09134988486766815 2023-01-21 09:48:11.318967: step: 1648/529, loss: 0.11957743763923645 2023-01-21 09:48:12.429646: step: 1652/529, loss: 0.17525739967823029 2023-01-21 09:48:13.547857: step: 1656/529, loss: 0.14496764540672302 2023-01-21 09:48:14.659475: step: 1660/529, loss: 0.07576890289783478 2023-01-21 09:48:15.763926: step: 1664/529, loss: 0.039725493639707565 2023-01-21 09:48:16.879557: step: 1668/529, loss: 0.25583988428115845 2023-01-21 09:48:18.019828: step: 1672/529, loss: 0.10584848374128342 2023-01-21 09:48:19.169187: step: 1676/529, loss: 0.08671937137842178 2023-01-21 09:48:20.303423: step: 1680/529, loss: 0.24428725242614746 2023-01-21 09:48:21.425568: step: 1684/529, loss: 0.02268686331808567 2023-01-21 09:48:22.572630: step: 1688/529, loss: 0.1452745497226715 2023-01-21 09:48:23.691104: step: 1692/529, loss: 0.13715052604675293 2023-01-21 09:48:24.822801: step: 1696/529, loss: 0.24294573068618774 2023-01-21 09:48:25.950423: step: 1700/529, loss: 0.046733953058719635 2023-01-21 09:48:27.077757: step: 1704/529, loss: 0.12026557326316833 2023-01-21 09:48:28.198099: step: 1708/529, loss: 0.5860031843185425 2023-01-21 09:48:29.304009: step: 1712/529, loss: 0.14337225258350372 2023-01-21 09:48:30.410706: step: 1716/529, loss: 0.015405749902129173 2023-01-21 09:48:31.510869: step: 1720/529, loss: 0.1179957389831543 2023-01-21 09:48:32.606240: step: 1724/529, loss: 0.07472304999828339 2023-01-21 09:48:33.716573: step: 1728/529, loss: 0.04063577577471733 2023-01-21 09:48:34.826357: step: 1732/529, loss: 0.2767406404018402 2023-01-21 09:48:35.968834: step: 1736/529, loss: 0.14368943870067596 2023-01-21 09:48:37.080249: step: 1740/529, loss: 0.03707614168524742 2023-01-21 09:48:38.187687: step: 1744/529, loss: 0.08122568577528 2023-01-21 09:48:39.315477: step: 1748/529, loss: 0.3294723331928253 2023-01-21 09:48:40.411782: step: 1752/529, loss: 0.2733578681945801 2023-01-21 09:48:41.539490: step: 1756/529, loss: 0.10837097465991974 2023-01-21 09:48:42.659418: step: 1760/529, loss: 0.13828545808792114 2023-01-21 09:48:43.740861: step: 1764/529, loss: 0.025430480018258095 2023-01-21 09:48:44.898648: step: 1768/529, loss: 0.2437111884355545 2023-01-21 09:48:46.014567: step: 1772/529, loss: 0.846227765083313 2023-01-21 09:48:47.123526: step: 1776/529, loss: 0.19628086686134338 2023-01-21 09:48:48.226269: step: 1780/529, loss: 0.02281503565609455 2023-01-21 09:48:49.344887: step: 1784/529, loss: 0.11977167427539825 2023-01-21 09:48:50.450692: step: 1788/529, loss: 0.049698732793331146 2023-01-21 09:48:51.562745: step: 1792/529, loss: 0.09053479135036469 2023-01-21 09:48:52.673064: step: 1796/529, loss: 0.010919428430497646 2023-01-21 09:48:53.788471: step: 1800/529, loss: 0.0467991828918457 2023-01-21 09:48:54.903492: step: 1804/529, loss: 0.1239691749215126 2023-01-21 09:48:56.038927: step: 1808/529, loss: 0.07301340252161026 2023-01-21 09:48:57.161563: step: 1812/529, loss: 1.4295861721038818 2023-01-21 09:48:58.261470: step: 1816/529, loss: 0.12106428295373917 2023-01-21 09:48:59.354138: step: 1820/529, loss: 0.10724583268165588 2023-01-21 09:49:00.487609: step: 1824/529, loss: 0.12029847502708435 2023-01-21 09:49:01.586882: step: 1828/529, loss: 0.09509435296058655 2023-01-21 09:49:02.691890: step: 1832/529, loss: 0.3433665335178375 2023-01-21 09:49:03.819075: step: 1836/529, loss: 0.3622245788574219 2023-01-21 09:49:04.910961: step: 1840/529, loss: 0.2004317343235016 2023-01-21 09:49:06.067382: step: 1844/529, loss: 0.07593774795532227 2023-01-21 09:49:07.203014: step: 1848/529, loss: 0.11956129968166351 2023-01-21 09:49:08.344743: step: 1852/529, loss: 0.13061046600341797 2023-01-21 09:49:09.453607: step: 1856/529, loss: 0.011540794745087624 2023-01-21 09:49:10.564021: step: 1860/529, loss: 0.15258969366550446 2023-01-21 09:49:11.696271: step: 1864/529, loss: 0.3293248116970062 2023-01-21 09:49:12.796219: step: 1868/529, loss: 0.1792701780796051 2023-01-21 09:49:13.915927: step: 1872/529, loss: 0.05549650266766548 2023-01-21 09:49:15.000162: step: 1876/529, loss: 0.04407348483800888 2023-01-21 09:49:16.113425: step: 1880/529, loss: 0.18867263197898865 2023-01-21 09:49:17.237981: step: 1884/529, loss: 0.1611851304769516 2023-01-21 09:49:18.351744: step: 1888/529, loss: 2.4172065258026123 2023-01-21 09:49:19.468508: step: 1892/529, loss: 0.02128305472433567 2023-01-21 09:49:20.568422: step: 1896/529, loss: 0.09441566467285156 2023-01-21 09:49:21.688452: step: 1900/529, loss: 0.09435348212718964 2023-01-21 09:49:22.794045: step: 1904/529, loss: 0.12584267556667328 2023-01-21 09:49:23.941997: step: 1908/529, loss: 0.32921791076660156 2023-01-21 09:49:25.042692: step: 1912/529, loss: 0.11085724830627441 2023-01-21 09:49:26.136912: step: 1916/529, loss: 0.05103917047381401 2023-01-21 09:49:27.270940: step: 1920/529, loss: 0.11318647861480713 2023-01-21 09:49:28.364648: step: 1924/529, loss: 0.048955440521240234 2023-01-21 09:49:29.467378: step: 1928/529, loss: 0.24831648170948029 2023-01-21 09:49:30.589383: step: 1932/529, loss: 0.6102821230888367 2023-01-21 09:49:31.731015: step: 1936/529, loss: 0.22656604647636414 2023-01-21 09:49:32.859903: step: 1940/529, loss: 0.08460502326488495 2023-01-21 09:49:33.987235: step: 1944/529, loss: 0.24833299219608307 2023-01-21 09:49:35.095105: step: 1948/529, loss: 0.022953206673264503 2023-01-21 09:49:36.209515: step: 1952/529, loss: 0.13200822472572327 2023-01-21 09:49:37.337445: step: 1956/529, loss: 0.49229127168655396 2023-01-21 09:49:38.439500: step: 1960/529, loss: 0.018326759338378906 2023-01-21 09:49:39.560771: step: 1964/529, loss: 0.07689686119556427 2023-01-21 09:49:40.687119: step: 1968/529, loss: 0.1271245926618576 2023-01-21 09:49:41.809663: step: 1972/529, loss: 0.08844394981861115 2023-01-21 09:49:42.951643: step: 1976/529, loss: 0.1441512107849121 2023-01-21 09:49:44.058741: step: 1980/529, loss: 0.013108921237289906 2023-01-21 09:49:45.189342: step: 1984/529, loss: 0.06107616424560547 2023-01-21 09:49:46.292433: step: 1988/529, loss: 0.057968832552433014 2023-01-21 09:49:47.406085: step: 1992/529, loss: 0.19316329061985016 2023-01-21 09:49:48.521645: step: 1996/529, loss: 0.10691499710083008 2023-01-21 09:49:49.617250: step: 2000/529, loss: 0.1370590329170227 2023-01-21 09:49:50.753742: step: 2004/529, loss: 0.3125910758972168 2023-01-21 09:49:51.907640: step: 2008/529, loss: 0.09253168106079102 2023-01-21 09:49:53.047517: step: 2012/529, loss: 0.12707695364952087 2023-01-21 09:49:54.181804: step: 2016/529, loss: 0.04439949989318848 2023-01-21 09:49:55.294705: step: 2020/529, loss: 0.09081140160560608 2023-01-21 09:49:56.417359: step: 2024/529, loss: 0.040528394281864166 2023-01-21 09:49:57.559167: step: 2028/529, loss: 0.11726704239845276 2023-01-21 09:49:58.682429: step: 2032/529, loss: 0.05750226974487305 2023-01-21 09:49:59.783475: step: 2036/529, loss: 0.1197289451956749 2023-01-21 09:50:00.862000: step: 2040/529, loss: 0.11227035522460938 2023-01-21 09:50:01.963543: step: 2044/529, loss: 0.08200503885746002 2023-01-21 09:50:03.064114: step: 2048/529, loss: 0.0762428343296051 2023-01-21 09:50:04.174878: step: 2052/529, loss: 0.140770822763443 2023-01-21 09:50:05.294703: step: 2056/529, loss: 0.19765663146972656 2023-01-21 09:50:06.415694: step: 2060/529, loss: 0.07162541896104813 2023-01-21 09:50:07.534756: step: 2064/529, loss: 0.05690574645996094 2023-01-21 09:50:08.618240: step: 2068/529, loss: 0.06004057079553604 2023-01-21 09:50:09.726441: step: 2072/529, loss: 0.04730527475476265 2023-01-21 09:50:10.851427: step: 2076/529, loss: 0.983487069606781 2023-01-21 09:50:11.985516: step: 2080/529, loss: 0.07275024056434631 2023-01-21 09:50:13.103245: step: 2084/529, loss: 0.11760297417640686 2023-01-21 09:50:14.221288: step: 2088/529, loss: 0.028754521161317825 2023-01-21 09:50:15.314268: step: 2092/529, loss: 0.23850007355213165 2023-01-21 09:50:16.409056: step: 2096/529, loss: 0.08069562911987305 2023-01-21 09:50:17.507883: step: 2100/529, loss: 0.04391040652990341 2023-01-21 09:50:18.614379: step: 2104/529, loss: 0.019821597263216972 2023-01-21 09:50:19.718983: step: 2108/529, loss: 0.46703338623046875 2023-01-21 09:50:20.817727: step: 2112/529, loss: 0.12980279326438904 2023-01-21 09:50:21.949305: step: 2116/529, loss: 0.07758083939552307 ================================================== Loss: 0.162 -------------------- Dev: {'event': {'p': 0.58, 'r': 0.7723035952063915, 'f1': 0.6624785836664763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Test: {'event': {'p': 0.6209532374100719, 'r': 0.7793453724604966, 'f1': 0.6911911911911912}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Chinese: {'event': {'p': 0.5476190476190477, 'r': 0.8518518518518519, 'f1': 0.6666666666666667}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Korean: {'event': {'p': 0.603448275862069, 'r': 0.5555555555555556, 'f1': 0.5785123966942148}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6170442286947141, 'r': 0.7616511318242344, 'f1': 0.6817640047675805}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6101614434947769, 'r': 0.7251693002257337, 'f1': 0.662712738525013}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6052631578947368, 'r': 0.8518518518518519, 'f1': 0.7076923076923076}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 8 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:51:02.521696: step: 4/529, loss: 0.03989434242248535 2023-01-21 09:51:03.679980: step: 8/529, loss: 0.2188451737165451 2023-01-21 09:51:04.819227: step: 12/529, loss: 0.10504093766212463 2023-01-21 09:51:05.936310: step: 16/529, loss: 0.309755802154541 2023-01-21 09:51:07.049393: step: 20/529, loss: 0.11937002837657928 2023-01-21 09:51:08.140931: step: 24/529, loss: 0.06205105409026146 2023-01-21 09:51:09.268499: step: 28/529, loss: 0.09447021782398224 2023-01-21 09:51:10.395999: step: 32/529, loss: 0.09353575855493546 2023-01-21 09:51:11.528926: step: 36/529, loss: 0.1375660002231598 2023-01-21 09:51:12.620803: step: 40/529, loss: 0.17229272425174713 2023-01-21 09:51:13.715311: step: 44/529, loss: 0.06236105039715767 2023-01-21 09:51:14.842737: step: 48/529, loss: 0.06906719505786896 2023-01-21 09:51:15.964129: step: 52/529, loss: 0.05072145536541939 2023-01-21 09:51:17.074278: step: 56/529, loss: 0.2119751125574112 2023-01-21 09:51:18.221473: step: 60/529, loss: 0.14456796646118164 2023-01-21 09:51:19.312399: step: 64/529, loss: 0.015509462915360928 2023-01-21 09:51:20.429137: step: 68/529, loss: 0.16052651405334473 2023-01-21 09:51:21.527338: step: 72/529, loss: 0.10224280506372452 2023-01-21 09:51:22.643832: step: 76/529, loss: 0.04146871715784073 2023-01-21 09:51:23.787594: step: 80/529, loss: 0.16864213347434998 2023-01-21 09:51:24.926157: step: 84/529, loss: 0.020550251007080078 2023-01-21 09:51:26.024575: step: 88/529, loss: 0.08963613212108612 2023-01-21 09:51:27.137122: step: 92/529, loss: 0.061861325055360794 2023-01-21 09:51:28.265087: step: 96/529, loss: 0.07162876427173615 2023-01-21 09:51:29.396918: step: 100/529, loss: 0.055115606635808945 2023-01-21 09:51:30.511246: step: 104/529, loss: 0.03399200364947319 2023-01-21 09:51:31.608695: step: 108/529, loss: 0.08677015453577042 2023-01-21 09:51:32.727101: step: 112/529, loss: 0.25093039870262146 2023-01-21 09:51:33.835022: step: 116/529, loss: 0.006470757536590099 2023-01-21 09:51:34.971830: step: 120/529, loss: 0.1085021048784256 2023-01-21 09:51:36.066973: step: 124/529, loss: 0.09303541481494904 2023-01-21 09:51:37.225295: step: 128/529, loss: 0.03469095379114151 2023-01-21 09:51:38.344271: step: 132/529, loss: 0.01894378662109375 2023-01-21 09:51:39.454666: step: 136/529, loss: 0.6561349630355835 2023-01-21 09:51:40.562009: step: 140/529, loss: 0.4928744435310364 2023-01-21 09:51:41.671579: step: 144/529, loss: 0.05222949758172035 2023-01-21 09:51:42.773557: step: 148/529, loss: 0.07344484329223633 2023-01-21 09:51:43.897854: step: 152/529, loss: 0.6262034177780151 2023-01-21 09:51:45.005085: step: 156/529, loss: 0.02418375015258789 2023-01-21 09:51:46.185124: step: 160/529, loss: 0.13597603142261505 2023-01-21 09:51:47.303182: step: 164/529, loss: 0.0391719825565815 2023-01-21 09:51:48.442572: step: 168/529, loss: 0.09543323516845703 2023-01-21 09:51:49.542515: step: 172/529, loss: 0.022873498499393463 2023-01-21 09:51:50.669728: step: 176/529, loss: 0.09264907985925674 2023-01-21 09:51:51.809975: step: 180/529, loss: 0.10887741297483444 2023-01-21 09:51:52.897320: step: 184/529, loss: 0.1670815497636795 2023-01-21 09:51:54.041702: step: 188/529, loss: 0.48168981075286865 2023-01-21 09:51:55.139940: step: 192/529, loss: 0.04652233421802521 2023-01-21 09:51:56.259443: step: 196/529, loss: 0.11765947937965393 2023-01-21 09:51:57.360925: step: 200/529, loss: 0.10074634850025177 2023-01-21 09:51:58.491587: step: 204/529, loss: 0.030885985121130943 2023-01-21 09:51:59.583226: step: 208/529, loss: 0.14120788872241974 2023-01-21 09:52:00.705456: step: 212/529, loss: 0.007666588295251131 2023-01-21 09:52:01.844053: step: 216/529, loss: 0.36684083938598633 2023-01-21 09:52:02.940368: step: 220/529, loss: 0.04493532329797745 2023-01-21 09:52:04.056735: step: 224/529, loss: 0.1575014889240265 2023-01-21 09:52:05.194041: step: 228/529, loss: 0.04303760826587677 2023-01-21 09:52:06.333342: step: 232/529, loss: 0.09628082066774368 2023-01-21 09:52:07.458468: step: 236/529, loss: 0.07134309411048889 2023-01-21 09:52:08.584705: step: 240/529, loss: 0.05675921589136124 2023-01-21 09:52:09.706810: step: 244/529, loss: 0.08725909888744354 2023-01-21 09:52:10.810138: step: 248/529, loss: 0.14610710740089417 2023-01-21 09:52:11.921553: step: 252/529, loss: 0.05349111557006836 2023-01-21 09:52:13.056229: step: 256/529, loss: 0.500694751739502 2023-01-21 09:52:14.200002: step: 260/529, loss: 0.40201592445373535 2023-01-21 09:52:15.325083: step: 264/529, loss: 0.10057735443115234 2023-01-21 09:52:16.457417: step: 268/529, loss: 0.01609516143798828 2023-01-21 09:52:17.585363: step: 272/529, loss: 0.06261692196130753 2023-01-21 09:52:18.701797: step: 276/529, loss: 0.1567176878452301 2023-01-21 09:52:19.793806: step: 280/529, loss: 0.03514843061566353 2023-01-21 09:52:20.928578: step: 284/529, loss: 0.02055797539651394 2023-01-21 09:52:22.039668: step: 288/529, loss: 0.02409052848815918 2023-01-21 09:52:23.148597: step: 292/529, loss: 0.04031085968017578 2023-01-21 09:52:24.260734: step: 296/529, loss: 0.07768788933753967 2023-01-21 09:52:25.382805: step: 300/529, loss: 0.09934453666210175 2023-01-21 09:52:26.504702: step: 304/529, loss: 0.10431576520204544 2023-01-21 09:52:27.597326: step: 308/529, loss: 0.12383947521448135 2023-01-21 09:52:28.701638: step: 312/529, loss: 0.02905302122235298 2023-01-21 09:52:29.828735: step: 316/529, loss: 0.13825073838233948 2023-01-21 09:52:30.936457: step: 320/529, loss: 0.05561008304357529 2023-01-21 09:52:32.059954: step: 324/529, loss: 0.04135637357831001 2023-01-21 09:52:33.165451: step: 328/529, loss: 0.021561909466981888 2023-01-21 09:52:34.280230: step: 332/529, loss: 0.06837920844554901 2023-01-21 09:52:35.404337: step: 336/529, loss: 0.04222039878368378 2023-01-21 09:52:36.511709: step: 340/529, loss: 0.02071533352136612 2023-01-21 09:52:37.625213: step: 344/529, loss: 0.6496995687484741 2023-01-21 09:52:38.727845: step: 348/529, loss: 0.05602569505572319 2023-01-21 09:52:39.833766: step: 352/529, loss: 0.1458621472120285 2023-01-21 09:52:40.939533: step: 356/529, loss: 0.08259530365467072 2023-01-21 09:52:42.050358: step: 360/529, loss: 0.10850724577903748 2023-01-21 09:52:43.203222: step: 364/529, loss: 0.2262999564409256 2023-01-21 09:52:44.329509: step: 368/529, loss: 0.10388288646936417 2023-01-21 09:52:45.444488: step: 372/529, loss: 0.11695671081542969 2023-01-21 09:52:46.569658: step: 376/529, loss: 0.09209509193897247 2023-01-21 09:52:47.658056: step: 380/529, loss: 0.06842823326587677 2023-01-21 09:52:48.748286: step: 384/529, loss: 0.07121582329273224 2023-01-21 09:52:49.857185: step: 388/529, loss: 0.11692357063293457 2023-01-21 09:52:50.973235: step: 392/529, loss: 0.06335633993148804 2023-01-21 09:52:52.106251: step: 396/529, loss: 0.193498894572258 2023-01-21 09:52:53.215861: step: 400/529, loss: 0.2050722986459732 2023-01-21 09:52:54.333150: step: 404/529, loss: 0.05390138179063797 2023-01-21 09:52:55.433705: step: 408/529, loss: 0.1637982428073883 2023-01-21 09:52:56.541277: step: 412/529, loss: 0.042731188237667084 2023-01-21 09:52:57.664918: step: 416/529, loss: 0.022153139114379883 2023-01-21 09:52:58.765256: step: 420/529, loss: 0.0784984603524208 2023-01-21 09:52:59.890294: step: 424/529, loss: 0.02855692058801651 2023-01-21 09:53:01.014666: step: 428/529, loss: 0.07432766258716583 2023-01-21 09:53:02.167692: step: 432/529, loss: 0.22140216827392578 2023-01-21 09:53:03.241339: step: 436/529, loss: 0.12895917892456055 2023-01-21 09:53:04.368280: step: 440/529, loss: 0.2757503390312195 2023-01-21 09:53:05.467814: step: 444/529, loss: 0.12068644165992737 2023-01-21 09:53:06.593474: step: 448/529, loss: 0.054241567850112915 2023-01-21 09:53:07.715770: step: 452/529, loss: 0.045977212488651276 2023-01-21 09:53:08.833945: step: 456/529, loss: 0.09137392044067383 2023-01-21 09:53:09.936619: step: 460/529, loss: 0.053114842623472214 2023-01-21 09:53:11.074581: step: 464/529, loss: 0.11413630843162537 2023-01-21 09:53:12.170242: step: 468/529, loss: 0.02526998519897461 2023-01-21 09:53:13.307798: step: 472/529, loss: 0.20771875977516174 2023-01-21 09:53:14.414777: step: 476/529, loss: 0.10484285652637482 2023-01-21 09:53:15.538606: step: 480/529, loss: 0.08314388245344162 2023-01-21 09:53:16.653929: step: 484/529, loss: 0.1397075653076172 2023-01-21 09:53:17.776992: step: 488/529, loss: 0.26071417331695557 2023-01-21 09:53:18.888701: step: 492/529, loss: 0.1529836654663086 2023-01-21 09:53:20.047801: step: 496/529, loss: 0.022241592407226562 2023-01-21 09:53:21.225459: step: 500/529, loss: 0.09012861549854279 2023-01-21 09:53:22.349149: step: 504/529, loss: 0.019520092755556107 2023-01-21 09:53:23.454349: step: 508/529, loss: 0.12705931067466736 2023-01-21 09:53:24.599412: step: 512/529, loss: 0.16276608407497406 2023-01-21 09:53:25.711048: step: 516/529, loss: 0.08586812764406204 2023-01-21 09:53:26.878038: step: 520/529, loss: 0.013790750876069069 2023-01-21 09:53:27.988036: step: 524/529, loss: 0.06156792491674423 2023-01-21 09:53:29.091511: step: 528/529, loss: 0.050742436200380325 2023-01-21 09:53:30.198108: step: 532/529, loss: 0.029255010187625885 2023-01-21 09:53:31.318417: step: 536/529, loss: 0.47186341881752014 2023-01-21 09:53:32.420274: step: 540/529, loss: 0.005035304930061102 2023-01-21 09:53:33.527220: step: 544/529, loss: 0.031760167330503464 2023-01-21 09:53:34.624205: step: 548/529, loss: 0.030845189467072487 2023-01-21 09:53:35.727804: step: 552/529, loss: 0.18544712662696838 2023-01-21 09:53:36.866554: step: 556/529, loss: 0.10837945342063904 2023-01-21 09:53:37.944928: step: 560/529, loss: 0.035142041742801666 2023-01-21 09:53:39.064299: step: 564/529, loss: 0.07538872212171555 2023-01-21 09:53:40.190684: step: 568/529, loss: 0.03586931526660919 2023-01-21 09:53:41.306510: step: 572/529, loss: 0.017791176214814186 2023-01-21 09:53:42.404144: step: 576/529, loss: 0.12268924713134766 2023-01-21 09:53:43.523031: step: 580/529, loss: 0.10870666801929474 2023-01-21 09:53:44.647676: step: 584/529, loss: 0.024970388039946556 2023-01-21 09:53:45.763441: step: 588/529, loss: 0.14508552849292755 2023-01-21 09:53:46.863064: step: 592/529, loss: 0.11120348423719406 2023-01-21 09:53:47.963045: step: 596/529, loss: 0.1060466319322586 2023-01-21 09:53:49.057107: step: 600/529, loss: 0.034670162945985794 2023-01-21 09:53:50.184503: step: 604/529, loss: 0.020639659836888313 2023-01-21 09:53:51.321302: step: 608/529, loss: 0.14827147126197815 2023-01-21 09:53:52.456180: step: 612/529, loss: 0.4108858108520508 2023-01-21 09:53:53.565556: step: 616/529, loss: 0.04742574691772461 2023-01-21 09:53:54.644408: step: 620/529, loss: 0.02974834479391575 2023-01-21 09:53:55.763500: step: 624/529, loss: 0.194295734167099 2023-01-21 09:53:56.881886: step: 628/529, loss: 0.1625930368900299 2023-01-21 09:53:58.011007: step: 632/529, loss: 0.10084104537963867 2023-01-21 09:53:59.116986: step: 636/529, loss: 0.15282422304153442 2023-01-21 09:54:00.230912: step: 640/529, loss: 0.07411523908376694 2023-01-21 09:54:01.342431: step: 644/529, loss: 0.06980600953102112 2023-01-21 09:54:02.451277: step: 648/529, loss: 0.3058091104030609 2023-01-21 09:54:03.555550: step: 652/529, loss: 0.19508886337280273 2023-01-21 09:54:04.694703: step: 656/529, loss: 0.14279861748218536 2023-01-21 09:54:05.833571: step: 660/529, loss: 0.07856860011816025 2023-01-21 09:54:06.984854: step: 664/529, loss: 0.047208692878484726 2023-01-21 09:54:08.133579: step: 668/529, loss: 0.06210184097290039 2023-01-21 09:54:09.274729: step: 672/529, loss: 0.009226322174072266 2023-01-21 09:54:10.391727: step: 676/529, loss: 0.3376133441925049 2023-01-21 09:54:11.504453: step: 680/529, loss: 0.2074717879295349 2023-01-21 09:54:12.608534: step: 684/529, loss: 0.10775518417358398 2023-01-21 09:54:13.743739: step: 688/529, loss: 0.1584509313106537 2023-01-21 09:54:14.896564: step: 692/529, loss: 0.1252274513244629 2023-01-21 09:54:16.033803: step: 696/529, loss: 0.2734691798686981 2023-01-21 09:54:17.145137: step: 700/529, loss: 0.08306089043617249 2023-01-21 09:54:18.227220: step: 704/529, loss: 0.0260987039655447 2023-01-21 09:54:19.327136: step: 708/529, loss: 0.1043708324432373 2023-01-21 09:54:20.457486: step: 712/529, loss: 0.10401935875415802 2023-01-21 09:54:21.560373: step: 716/529, loss: 0.13772697746753693 2023-01-21 09:54:22.688275: step: 720/529, loss: 0.15383297204971313 2023-01-21 09:54:23.822947: step: 724/529, loss: 0.19827446341514587 2023-01-21 09:54:24.915145: step: 728/529, loss: 0.059778306633234024 2023-01-21 09:54:26.011369: step: 732/529, loss: 0.026334714144468307 2023-01-21 09:54:27.122780: step: 736/529, loss: 0.11739101260900497 2023-01-21 09:54:28.228913: step: 740/529, loss: 0.06597013771533966 2023-01-21 09:54:29.332931: step: 744/529, loss: 0.17049741744995117 2023-01-21 09:54:30.429010: step: 748/529, loss: 0.13470125198364258 2023-01-21 09:54:31.538242: step: 752/529, loss: 0.1963629424571991 2023-01-21 09:54:32.658845: step: 756/529, loss: 0.1465105265378952 2023-01-21 09:54:33.789781: step: 760/529, loss: 0.24067707359790802 2023-01-21 09:54:34.901802: step: 764/529, loss: 0.0548405647277832 2023-01-21 09:54:36.021898: step: 768/529, loss: 0.05560598522424698 2023-01-21 09:54:37.143774: step: 772/529, loss: 1.8638938665390015 2023-01-21 09:54:38.265007: step: 776/529, loss: 0.12758255004882812 2023-01-21 09:54:39.351363: step: 780/529, loss: 0.05670108646154404 2023-01-21 09:54:40.460713: step: 784/529, loss: 0.06468353420495987 2023-01-21 09:54:41.568071: step: 788/529, loss: 0.048354532569646835 2023-01-21 09:54:42.655556: step: 792/529, loss: 0.0731218233704567 2023-01-21 09:54:43.787875: step: 796/529, loss: 0.2670901417732239 2023-01-21 09:54:44.933448: step: 800/529, loss: 0.16406115889549255 2023-01-21 09:54:46.040661: step: 804/529, loss: 0.11984129250049591 2023-01-21 09:54:47.171037: step: 808/529, loss: 0.14334678649902344 2023-01-21 09:54:48.291453: step: 812/529, loss: 0.24785137176513672 2023-01-21 09:54:49.412896: step: 816/529, loss: 0.007072162814438343 2023-01-21 09:54:50.509015: step: 820/529, loss: 0.09852228313684464 2023-01-21 09:54:51.609435: step: 824/529, loss: 0.3261339068412781 2023-01-21 09:54:52.726679: step: 828/529, loss: 0.02823066897690296 2023-01-21 09:54:53.868138: step: 832/529, loss: 0.062287621200084686 2023-01-21 09:54:55.004961: step: 836/529, loss: 0.048912905156612396 2023-01-21 09:54:56.132542: step: 840/529, loss: 0.05576535686850548 2023-01-21 09:54:57.269341: step: 844/529, loss: 0.20508727431297302 2023-01-21 09:54:58.376088: step: 848/529, loss: 0.04440021514892578 2023-01-21 09:54:59.471471: step: 852/529, loss: 0.03855152428150177 2023-01-21 09:55:00.614414: step: 856/529, loss: 0.3242805600166321 2023-01-21 09:55:01.753413: step: 860/529, loss: 0.5788240432739258 2023-01-21 09:55:02.869287: step: 864/529, loss: 0.033098507672548294 2023-01-21 09:55:03.998144: step: 868/529, loss: 0.12986326217651367 2023-01-21 09:55:05.115576: step: 872/529, loss: 0.2866189479827881 2023-01-21 09:55:06.208602: step: 876/529, loss: 0.0950084701180458 2023-01-21 09:55:07.306541: step: 880/529, loss: 0.18983830511569977 2023-01-21 09:55:08.425540: step: 884/529, loss: 0.08172015845775604 2023-01-21 09:55:09.557430: step: 888/529, loss: 0.15899963676929474 2023-01-21 09:55:10.667328: step: 892/529, loss: 0.4648832082748413 2023-01-21 09:55:11.783236: step: 896/529, loss: 0.17418155074119568 2023-01-21 09:55:12.891540: step: 900/529, loss: 0.05792875587940216 2023-01-21 09:55:14.035602: step: 904/529, loss: 0.04514046013355255 2023-01-21 09:55:15.134734: step: 908/529, loss: 0.038633015006780624 2023-01-21 09:55:16.262073: step: 912/529, loss: 0.16504965722560883 2023-01-21 09:55:17.358803: step: 916/529, loss: 0.03185730054974556 2023-01-21 09:55:18.448260: step: 920/529, loss: 0.029438089579343796 2023-01-21 09:55:19.567019: step: 924/529, loss: 0.14387011528015137 2023-01-21 09:55:20.665408: step: 928/529, loss: 0.13822278380393982 2023-01-21 09:55:21.754125: step: 932/529, loss: 0.043334536254405975 2023-01-21 09:55:22.881574: step: 936/529, loss: 0.13769817352294922 2023-01-21 09:55:23.986932: step: 940/529, loss: 0.10432787984609604 2023-01-21 09:55:25.162702: step: 944/529, loss: 0.3595340847969055 2023-01-21 09:55:26.296414: step: 948/529, loss: 0.16920499503612518 2023-01-21 09:55:27.417175: step: 952/529, loss: 0.055108167231082916 2023-01-21 09:55:28.523495: step: 956/529, loss: 0.12649258971214294 2023-01-21 09:55:29.729850: step: 960/529, loss: 0.13529044389724731 2023-01-21 09:55:30.847431: step: 964/529, loss: 0.06275320053100586 2023-01-21 09:55:31.959007: step: 968/529, loss: 0.3659912943840027 2023-01-21 09:55:33.079912: step: 972/529, loss: 0.13667917251586914 2023-01-21 09:55:34.216822: step: 976/529, loss: 0.13459502160549164 2023-01-21 09:55:35.307976: step: 980/529, loss: 0.012503433041274548 2023-01-21 09:55:36.449904: step: 984/529, loss: 0.11710643768310547 2023-01-21 09:55:37.568057: step: 988/529, loss: 0.3439061939716339 2023-01-21 09:55:38.657573: step: 992/529, loss: 0.10192890465259552 2023-01-21 09:55:39.786359: step: 996/529, loss: 0.3993247449398041 2023-01-21 09:55:40.893813: step: 1000/529, loss: 0.023722808808088303 2023-01-21 09:55:42.010024: step: 1004/529, loss: 0.02011260949075222 2023-01-21 09:55:43.131739: step: 1008/529, loss: 0.08597803115844727 2023-01-21 09:55:44.231013: step: 1012/529, loss: 0.03914952278137207 2023-01-21 09:55:45.356150: step: 1016/529, loss: 0.02920856513082981 2023-01-21 09:55:46.476321: step: 1020/529, loss: 0.013190162368118763 2023-01-21 09:55:47.562253: step: 1024/529, loss: 0.11433763802051544 2023-01-21 09:55:48.682072: step: 1028/529, loss: 0.045697782188653946 2023-01-21 09:55:49.797256: step: 1032/529, loss: 0.1532432585954666 2023-01-21 09:55:50.940497: step: 1036/529, loss: 0.1075565367937088 2023-01-21 09:55:52.050723: step: 1040/529, loss: 0.03354320675134659 2023-01-21 09:55:53.163806: step: 1044/529, loss: 0.023433685302734375 2023-01-21 09:55:54.293713: step: 1048/529, loss: 0.12937398254871368 2023-01-21 09:55:55.426901: step: 1052/529, loss: 0.0020016669295728207 2023-01-21 09:55:56.561893: step: 1056/529, loss: 0.15442132949829102 2023-01-21 09:55:57.689093: step: 1060/529, loss: 0.02352127991616726 2023-01-21 09:55:58.817563: step: 1064/529, loss: 0.0686376541852951 2023-01-21 09:55:59.936944: step: 1068/529, loss: 0.39713162183761597 2023-01-21 09:56:01.067696: step: 1072/529, loss: 0.10481038689613342 2023-01-21 09:56:02.182118: step: 1076/529, loss: 0.030427539721131325 2023-01-21 09:56:03.289161: step: 1080/529, loss: 0.04467516019940376 2023-01-21 09:56:04.407400: step: 1084/529, loss: 0.1046157106757164 2023-01-21 09:56:05.549042: step: 1088/529, loss: 0.05927839130163193 2023-01-21 09:56:06.670127: step: 1092/529, loss: 0.11200771480798721 2023-01-21 09:56:07.800157: step: 1096/529, loss: 0.0482630729675293 2023-01-21 09:56:08.946379: step: 1100/529, loss: 0.0799582451581955 2023-01-21 09:56:10.054546: step: 1104/529, loss: 0.057820986956357956 2023-01-21 09:56:11.179146: step: 1108/529, loss: 0.05879940837621689 2023-01-21 09:56:12.302650: step: 1112/529, loss: 0.11061849445104599 2023-01-21 09:56:13.440207: step: 1116/529, loss: 0.07461271435022354 2023-01-21 09:56:14.566676: step: 1120/529, loss: 0.09170246124267578 2023-01-21 09:56:15.721358: step: 1124/529, loss: 0.07950973510742188 2023-01-21 09:56:16.828411: step: 1128/529, loss: 0.0448024719953537 2023-01-21 09:56:17.914207: step: 1132/529, loss: 0.22226496040821075 2023-01-21 09:56:19.045367: step: 1136/529, loss: 0.05509338527917862 2023-01-21 09:56:20.214330: step: 1140/529, loss: 0.3249807357788086 2023-01-21 09:56:21.335444: step: 1144/529, loss: 0.02433748170733452 2023-01-21 09:56:22.444950: step: 1148/529, loss: 0.34807834029197693 2023-01-21 09:56:23.556621: step: 1152/529, loss: 0.18976755440235138 2023-01-21 09:56:24.655170: step: 1156/529, loss: 0.04614315181970596 2023-01-21 09:56:25.750444: step: 1160/529, loss: 0.02836017496883869 2023-01-21 09:56:26.878933: step: 1164/529, loss: 0.09402541816234589 2023-01-21 09:56:27.998040: step: 1168/529, loss: 0.05163278430700302 2023-01-21 09:56:29.117024: step: 1172/529, loss: 0.246564581990242 2023-01-21 09:56:30.214438: step: 1176/529, loss: 0.19903278350830078 2023-01-21 09:56:31.347511: step: 1180/529, loss: 0.0760578140616417 2023-01-21 09:56:32.460523: step: 1184/529, loss: 0.13230033218860626 2023-01-21 09:56:33.564190: step: 1188/529, loss: 0.48669958114624023 2023-01-21 09:56:34.699881: step: 1192/529, loss: 0.014771080575883389 2023-01-21 09:56:35.813513: step: 1196/529, loss: 0.02422809787094593 2023-01-21 09:56:36.934985: step: 1200/529, loss: 0.009517192840576172 2023-01-21 09:56:38.054652: step: 1204/529, loss: 0.11889752745628357 2023-01-21 09:56:39.195544: step: 1208/529, loss: 0.12151375412940979 2023-01-21 09:56:40.320345: step: 1212/529, loss: 0.10582761466503143 2023-01-21 09:56:41.450987: step: 1216/529, loss: 0.07957177609205246 2023-01-21 09:56:42.558149: step: 1220/529, loss: 0.07901802659034729 2023-01-21 09:56:43.701646: step: 1224/529, loss: 0.17792721092700958 2023-01-21 09:56:44.789959: step: 1228/529, loss: 0.06355170905590057 2023-01-21 09:56:45.933240: step: 1232/529, loss: 0.1157953292131424 2023-01-21 09:56:47.051613: step: 1236/529, loss: 0.08996716141700745 2023-01-21 09:56:48.152547: step: 1240/529, loss: 0.04892396926879883 2023-01-21 09:56:49.266624: step: 1244/529, loss: 0.014222336001694202 2023-01-21 09:56:50.403210: step: 1248/529, loss: 0.5267826318740845 2023-01-21 09:56:51.516789: step: 1252/529, loss: 0.07460909336805344 2023-01-21 09:56:52.612483: step: 1256/529, loss: 0.10695400834083557 2023-01-21 09:56:53.714297: step: 1260/529, loss: 0.05616407468914986 2023-01-21 09:56:54.839214: step: 1264/529, loss: 0.03346419334411621 2023-01-21 09:56:55.950120: step: 1268/529, loss: 0.06809739768505096 2023-01-21 09:56:57.044777: step: 1272/529, loss: 0.07186833024024963 2023-01-21 09:56:58.173073: step: 1276/529, loss: 0.17860373854637146 2023-01-21 09:56:59.276352: step: 1280/529, loss: 0.029743101447820663 2023-01-21 09:57:00.435767: step: 1284/529, loss: 1.512211799621582 2023-01-21 09:57:01.549683: step: 1288/529, loss: 0.005821418948471546 2023-01-21 09:57:02.668353: step: 1292/529, loss: 0.08688574284315109 2023-01-21 09:57:03.796331: step: 1296/529, loss: 0.04068155586719513 2023-01-21 09:57:04.881402: step: 1300/529, loss: 0.08950367569923401 2023-01-21 09:57:06.013864: step: 1304/529, loss: 0.03899574279785156 2023-01-21 09:57:07.120617: step: 1308/529, loss: 0.0820469930768013 2023-01-21 09:57:08.199724: step: 1312/529, loss: 0.05772824212908745 2023-01-21 09:57:09.367688: step: 1316/529, loss: 0.23979711532592773 2023-01-21 09:57:10.480321: step: 1320/529, loss: 0.06210651248693466 2023-01-21 09:57:11.619491: step: 1324/529, loss: 0.14506320655345917 2023-01-21 09:57:12.724926: step: 1328/529, loss: 0.01283798273652792 2023-01-21 09:57:13.826707: step: 1332/529, loss: 0.13258647918701172 2023-01-21 09:57:14.946476: step: 1336/529, loss: 0.022126102820038795 2023-01-21 09:57:16.044619: step: 1340/529, loss: 0.06437671184539795 2023-01-21 09:57:17.156885: step: 1344/529, loss: 0.13645707070827484 2023-01-21 09:57:18.245527: step: 1348/529, loss: 0.09878106415271759 2023-01-21 09:57:19.352305: step: 1352/529, loss: 0.08160868287086487 2023-01-21 09:57:20.451772: step: 1356/529, loss: 0.07624302059412003 2023-01-21 09:57:21.542484: step: 1360/529, loss: 0.07845602184534073 2023-01-21 09:57:22.672223: step: 1364/529, loss: 0.06410370022058487 2023-01-21 09:57:23.763504: step: 1368/529, loss: 0.02327604405581951 2023-01-21 09:57:24.884605: step: 1372/529, loss: 0.2503858208656311 2023-01-21 09:57:26.014070: step: 1376/529, loss: 0.12697726488113403 2023-01-21 09:57:27.144453: step: 1380/529, loss: 0.05052796006202698 2023-01-21 09:57:28.300112: step: 1384/529, loss: 0.10965590178966522 2023-01-21 09:57:29.420062: step: 1388/529, loss: 0.04164714738726616 2023-01-21 09:57:30.508886: step: 1392/529, loss: 0.021717073395848274 2023-01-21 09:57:31.655262: step: 1396/529, loss: 0.07528341561555862 2023-01-21 09:57:32.757714: step: 1400/529, loss: 0.13904786109924316 2023-01-21 09:57:33.895552: step: 1404/529, loss: 0.03939530998468399 2023-01-21 09:57:35.026814: step: 1408/529, loss: 0.14536018669605255 2023-01-21 09:57:36.146546: step: 1412/529, loss: 0.11746422201395035 2023-01-21 09:57:37.256053: step: 1416/529, loss: 0.10972604900598526 2023-01-21 09:57:38.394609: step: 1420/529, loss: 0.030263518914580345 2023-01-21 09:57:39.488104: step: 1424/529, loss: 0.15680274367332458 2023-01-21 09:57:40.613853: step: 1428/529, loss: 0.4611307382583618 2023-01-21 09:57:41.752527: step: 1432/529, loss: 0.19050291180610657 2023-01-21 09:57:42.867531: step: 1436/529, loss: 0.2391345500946045 2023-01-21 09:57:43.963945: step: 1440/529, loss: 0.12605568766593933 2023-01-21 09:57:45.108281: step: 1444/529, loss: 0.17011089622974396 2023-01-21 09:57:46.248001: step: 1448/529, loss: 0.45945197343826294 2023-01-21 09:57:47.406026: step: 1452/529, loss: 0.008951187133789062 2023-01-21 09:57:48.516954: step: 1456/529, loss: 0.019978713244199753 2023-01-21 09:57:49.640338: step: 1460/529, loss: 0.39568251371383667 2023-01-21 09:57:50.741271: step: 1464/529, loss: 0.0516299270093441 2023-01-21 09:57:51.842267: step: 1468/529, loss: 0.01072320993989706 2023-01-21 09:57:52.958010: step: 1472/529, loss: 0.13910111784934998 2023-01-21 09:57:54.068801: step: 1476/529, loss: 0.03418789058923721 2023-01-21 09:57:55.245170: step: 1480/529, loss: 0.09762802720069885 2023-01-21 09:57:56.341841: step: 1484/529, loss: 0.09149055182933807 2023-01-21 09:57:57.479311: step: 1488/529, loss: 0.12796077132225037 2023-01-21 09:57:58.637508: step: 1492/529, loss: 0.1952836960554123 2023-01-21 09:57:59.741395: step: 1496/529, loss: 0.09624385833740234 2023-01-21 09:58:00.850588: step: 1500/529, loss: 0.015612030401825905 2023-01-21 09:58:01.973904: step: 1504/529, loss: 0.01025619637221098 2023-01-21 09:58:03.104476: step: 1508/529, loss: 0.11805696785449982 2023-01-21 09:58:04.202079: step: 1512/529, loss: 0.06687374413013458 2023-01-21 09:58:05.316137: step: 1516/529, loss: 0.021505165845155716 2023-01-21 09:58:06.427342: step: 1520/529, loss: 0.5881984233856201 2023-01-21 09:58:07.526846: step: 1524/529, loss: 0.18597888946533203 2023-01-21 09:58:08.620386: step: 1528/529, loss: 0.2339155673980713 2023-01-21 09:58:09.736778: step: 1532/529, loss: 0.7397167086601257 2023-01-21 09:58:10.869045: step: 1536/529, loss: 0.8801301121711731 2023-01-21 09:58:11.975602: step: 1540/529, loss: 0.05980673059821129 2023-01-21 09:58:13.089002: step: 1544/529, loss: 0.1172122061252594 2023-01-21 09:58:14.254770: step: 1548/529, loss: 0.007224726490676403 2023-01-21 09:58:15.336867: step: 1552/529, loss: 0.33185186982154846 2023-01-21 09:58:16.446549: step: 1556/529, loss: 0.11558599770069122 2023-01-21 09:58:17.554314: step: 1560/529, loss: 0.0766727477312088 2023-01-21 09:58:18.707740: step: 1564/529, loss: 0.1181788444519043 2023-01-21 09:58:19.808316: step: 1568/529, loss: 0.016602516174316406 2023-01-21 09:58:20.917443: step: 1572/529, loss: 0.11860628426074982 2023-01-21 09:58:22.058891: step: 1576/529, loss: 0.16875658929347992 2023-01-21 09:58:23.191152: step: 1580/529, loss: 0.05875139683485031 2023-01-21 09:58:24.345652: step: 1584/529, loss: 0.7112193703651428 2023-01-21 09:58:25.454053: step: 1588/529, loss: 0.17542333900928497 2023-01-21 09:58:26.571948: step: 1592/529, loss: 0.11918716132640839 2023-01-21 09:58:27.704875: step: 1596/529, loss: 0.05150041729211807 2023-01-21 09:58:28.834511: step: 1600/529, loss: 0.13940078020095825 2023-01-21 09:58:29.949941: step: 1604/529, loss: 0.13837090134620667 2023-01-21 09:58:31.073577: step: 1608/529, loss: 0.08137612789869308 2023-01-21 09:58:32.194218: step: 1612/529, loss: 0.04746093600988388 2023-01-21 09:58:33.334491: step: 1616/529, loss: 0.3600454330444336 2023-01-21 09:58:34.444962: step: 1620/529, loss: 0.035196926444768906 2023-01-21 09:58:35.555795: step: 1624/529, loss: 0.0704149454832077 2023-01-21 09:58:36.662349: step: 1628/529, loss: 0.13225889205932617 2023-01-21 09:58:37.798522: step: 1632/529, loss: 0.11158795654773712 2023-01-21 09:58:38.894958: step: 1636/529, loss: 0.5553426742553711 2023-01-21 09:58:39.993900: step: 1640/529, loss: 0.26906871795654297 2023-01-21 09:58:41.127098: step: 1644/529, loss: 0.06360411643981934 2023-01-21 09:58:42.246560: step: 1648/529, loss: 0.13051024079322815 2023-01-21 09:58:43.366166: step: 1652/529, loss: 0.08371467888355255 2023-01-21 09:58:44.494366: step: 1656/529, loss: 0.17047472298145294 2023-01-21 09:58:45.591277: step: 1660/529, loss: 0.1417037546634674 2023-01-21 09:58:46.701605: step: 1664/529, loss: 0.2107383757829666 2023-01-21 09:58:47.812498: step: 1668/529, loss: 0.10277318954467773 2023-01-21 09:58:48.960746: step: 1672/529, loss: 0.009971666149795055 2023-01-21 09:58:50.082212: step: 1676/529, loss: 0.11623702943325043 2023-01-21 09:58:51.186257: step: 1680/529, loss: 0.8221315145492554 2023-01-21 09:58:52.400350: step: 1684/529, loss: 1.698832631111145 2023-01-21 09:58:53.513581: step: 1688/529, loss: 0.10908474773168564 2023-01-21 09:58:54.621420: step: 1692/529, loss: 0.04670019447803497 2023-01-21 09:58:55.760015: step: 1696/529, loss: 0.10548000037670135 2023-01-21 09:58:56.857928: step: 1700/529, loss: 0.06395301967859268 2023-01-21 09:58:57.939874: step: 1704/529, loss: 0.36515045166015625 2023-01-21 09:58:59.090179: step: 1708/529, loss: 0.04566478729248047 2023-01-21 09:59:00.213196: step: 1712/529, loss: 0.12969131767749786 2023-01-21 09:59:01.314217: step: 1716/529, loss: 0.09466209262609482 2023-01-21 09:59:02.437114: step: 1720/529, loss: 0.4289804697036743 2023-01-21 09:59:03.538059: step: 1724/529, loss: 0.07257632911205292 2023-01-21 09:59:04.650121: step: 1728/529, loss: 0.09229574352502823 2023-01-21 09:59:05.754805: step: 1732/529, loss: 0.834445059299469 2023-01-21 09:59:06.904941: step: 1736/529, loss: 0.39292111992836 2023-01-21 09:59:07.989957: step: 1740/529, loss: 0.5804386138916016 2023-01-21 09:59:09.104829: step: 1744/529, loss: 0.0479232557117939 2023-01-21 09:59:10.242638: step: 1748/529, loss: 1.06978178024292 2023-01-21 09:59:11.364002: step: 1752/529, loss: 1.2441238164901733 2023-01-21 09:59:12.463867: step: 1756/529, loss: 0.07084637135267258 2023-01-21 09:59:13.564063: step: 1760/529, loss: 0.03839244693517685 2023-01-21 09:59:14.682933: step: 1764/529, loss: 0.10491366684436798 2023-01-21 09:59:15.792665: step: 1768/529, loss: 0.19416770339012146 2023-01-21 09:59:16.907453: step: 1772/529, loss: 0.2712913453578949 2023-01-21 09:59:18.067225: step: 1776/529, loss: 0.04234728962182999 2023-01-21 09:59:19.181028: step: 1780/529, loss: 0.06250448524951935 2023-01-21 09:59:20.302953: step: 1784/529, loss: 0.01987132988870144 2023-01-21 09:59:21.410195: step: 1788/529, loss: 0.04949207603931427 2023-01-21 09:59:22.504240: step: 1792/529, loss: 0.0982908308506012 2023-01-21 09:59:23.615713: step: 1796/529, loss: 0.04162416607141495 2023-01-21 09:59:24.751498: step: 1800/529, loss: 0.016437197104096413 2023-01-21 09:59:25.870557: step: 1804/529, loss: 0.06721341609954834 2023-01-21 09:59:26.977129: step: 1808/529, loss: 0.05924713611602783 2023-01-21 09:59:28.093572: step: 1812/529, loss: 0.031153012067079544 2023-01-21 09:59:29.186871: step: 1816/529, loss: 0.010325432755053043 2023-01-21 09:59:30.281559: step: 1820/529, loss: 0.057450439780950546 2023-01-21 09:59:31.410444: step: 1824/529, loss: 0.0797567367553711 2023-01-21 09:59:32.552443: step: 1828/529, loss: 0.21246853470802307 2023-01-21 09:59:33.662376: step: 1832/529, loss: 1.2157001495361328 2023-01-21 09:59:34.811345: step: 1836/529, loss: 0.09803581237792969 2023-01-21 09:59:35.948997: step: 1840/529, loss: 0.13262850046157837 2023-01-21 09:59:37.060886: step: 1844/529, loss: 0.06733512878417969 2023-01-21 09:59:38.202123: step: 1848/529, loss: 0.05996561422944069 2023-01-21 09:59:39.331450: step: 1852/529, loss: 0.025759601965546608 2023-01-21 09:59:40.484841: step: 1856/529, loss: 0.15843135118484497 2023-01-21 09:59:41.600790: step: 1860/529, loss: 0.03394956886768341 2023-01-21 09:59:42.717862: step: 1864/529, loss: 0.4541475176811218 2023-01-21 09:59:43.829061: step: 1868/529, loss: 0.053975533694028854 2023-01-21 09:59:44.947257: step: 1872/529, loss: 0.055335454642772675 2023-01-21 09:59:46.051986: step: 1876/529, loss: 0.09587828814983368 2023-01-21 09:59:47.178187: step: 1880/529, loss: 0.27150774002075195 2023-01-21 09:59:48.262600: step: 1884/529, loss: 0.055780794471502304 2023-01-21 09:59:49.388727: step: 1888/529, loss: 0.21617546677589417 2023-01-21 09:59:50.504235: step: 1892/529, loss: 0.17622947692871094 2023-01-21 09:59:51.609525: step: 1896/529, loss: 0.4544470012187958 2023-01-21 09:59:52.758039: step: 1900/529, loss: 0.16246432065963745 2023-01-21 09:59:53.880793: step: 1904/529, loss: 0.041576579213142395 2023-01-21 09:59:54.981009: step: 1908/529, loss: 0.42554759979248047 2023-01-21 09:59:56.091217: step: 1912/529, loss: 0.16898003220558167 2023-01-21 09:59:57.237645: step: 1916/529, loss: 0.018013715744018555 2023-01-21 09:59:58.371299: step: 1920/529, loss: 0.048802945762872696 2023-01-21 09:59:59.477772: step: 1924/529, loss: 0.08774089813232422 2023-01-21 10:00:00.654778: step: 1928/529, loss: 0.0564337782561779 2023-01-21 10:00:01.767979: step: 1932/529, loss: 0.018157958984375 2023-01-21 10:00:02.850545: step: 1936/529, loss: 0.00856781005859375 2023-01-21 10:00:03.969357: step: 1940/529, loss: 0.0033669949043542147 2023-01-21 10:00:05.084122: step: 1944/529, loss: 0.39807718992233276 2023-01-21 10:00:06.188234: step: 1948/529, loss: 0.08601990342140198 2023-01-21 10:00:07.299881: step: 1952/529, loss: 0.03397779539227486 2023-01-21 10:00:08.416456: step: 1956/529, loss: 0.09619808197021484 2023-01-21 10:00:09.525200: step: 1960/529, loss: 0.07587385177612305 2023-01-21 10:00:10.628976: step: 1964/529, loss: 0.15980473160743713 2023-01-21 10:00:11.733517: step: 1968/529, loss: 0.06064872816205025 2023-01-21 10:00:12.870951: step: 1972/529, loss: 0.08335505425930023 2023-01-21 10:00:13.984058: step: 1976/529, loss: 0.05104637145996094 2023-01-21 10:00:15.088878: step: 1980/529, loss: 0.06667838245630264 2023-01-21 10:00:16.229620: step: 1984/529, loss: 0.06676197052001953 2023-01-21 10:00:17.338148: step: 1988/529, loss: 0.04147772863507271 2023-01-21 10:00:18.464463: step: 1992/529, loss: 0.1335565149784088 2023-01-21 10:00:19.577733: step: 1996/529, loss: 0.09469977021217346 2023-01-21 10:00:20.704777: step: 2000/529, loss: 0.10614728927612305 2023-01-21 10:00:21.809266: step: 2004/529, loss: 0.03339986875653267 2023-01-21 10:00:22.987129: step: 2008/529, loss: 0.06371164321899414 2023-01-21 10:00:24.122655: step: 2012/529, loss: 0.10844211280345917 2023-01-21 10:00:25.256227: step: 2016/529, loss: 0.046425629407167435 2023-01-21 10:00:26.373437: step: 2020/529, loss: 0.4454374313354492 2023-01-21 10:00:27.495837: step: 2024/529, loss: 0.028484439477324486 2023-01-21 10:00:28.595320: step: 2028/529, loss: 0.05515031889081001 2023-01-21 10:00:29.707852: step: 2032/529, loss: 0.15749740600585938 2023-01-21 10:00:30.814763: step: 2036/529, loss: 0.13970308005809784 2023-01-21 10:00:31.924871: step: 2040/529, loss: 0.04967961460351944 2023-01-21 10:00:33.014378: step: 2044/529, loss: 0.06730881333351135 2023-01-21 10:00:34.131201: step: 2048/529, loss: 0.10771723091602325 2023-01-21 10:00:35.238012: step: 2052/529, loss: 0.02240614965558052 2023-01-21 10:00:36.387177: step: 2056/529, loss: 0.8355466723442078 2023-01-21 10:00:37.544564: step: 2060/529, loss: 0.29652151465415955 2023-01-21 10:00:38.651076: step: 2064/529, loss: 0.07766053080558777 2023-01-21 10:00:39.801978: step: 2068/529, loss: 0.022115757688879967 2023-01-21 10:00:40.936457: step: 2072/529, loss: 0.05893435329198837 2023-01-21 10:00:42.060828: step: 2076/529, loss: 0.04326352849602699 2023-01-21 10:00:43.193957: step: 2080/529, loss: 0.07256489992141724 2023-01-21 10:00:44.301172: step: 2084/529, loss: 0.07202653586864471 2023-01-21 10:00:45.382252: step: 2088/529, loss: 0.0957399383187294 2023-01-21 10:00:46.500763: step: 2092/529, loss: 0.402616024017334 2023-01-21 10:00:47.599577: step: 2096/529, loss: 0.03434129059314728 2023-01-21 10:00:48.717437: step: 2100/529, loss: 0.18043117225170135 2023-01-21 10:00:49.850416: step: 2104/529, loss: 0.19229133427143097 2023-01-21 10:00:50.972683: step: 2108/529, loss: 0.009256839752197266 2023-01-21 10:00:52.116823: step: 2112/529, loss: 0.3673548996448517 2023-01-21 10:00:53.255029: step: 2116/529, loss: 0.1312166303396225 ================================================== Loss: 0.144 -------------------- Dev: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Test: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Korean: {'event': {'p': 0.5, 'r': 0.5396825396825397, 'f1': 0.5190839694656489}, '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... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 9 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:01:40.782070: step: 4/529, loss: 0.10318270325660706 2023-01-21 10:01:41.874590: step: 8/529, loss: 0.105072021484375 2023-01-21 10:01:43.003359: step: 12/529, loss: 0.015979385003447533 2023-01-21 10:01:44.139584: step: 16/529, loss: 0.01690196990966797 2023-01-21 10:01:45.221485: step: 20/529, loss: 0.13673830032348633 2023-01-21 10:01:46.344825: step: 24/529, loss: 0.09302611649036407 2023-01-21 10:01:47.467427: step: 28/529, loss: 0.07394295185804367 2023-01-21 10:01:48.575624: step: 32/529, loss: 0.04304695129394531 2023-01-21 10:01:49.709174: step: 36/529, loss: 0.003732919692993164 2023-01-21 10:01:50.802734: step: 40/529, loss: 0.049425218254327774 2023-01-21 10:01:51.947628: step: 44/529, loss: 0.12254457175731659 2023-01-21 10:01:53.054297: step: 48/529, loss: 0.03959474712610245 2023-01-21 10:01:54.155379: step: 52/529, loss: 0.004907703492790461 2023-01-21 10:01:55.274206: step: 56/529, loss: 0.02334756776690483 2023-01-21 10:01:56.387739: step: 60/529, loss: 0.06160583347082138 2023-01-21 10:01:57.507740: step: 64/529, loss: 0.001178693724796176 2023-01-21 10:01:58.647740: step: 68/529, loss: 0.09667682647705078 2023-01-21 10:01:59.773210: step: 72/529, loss: 0.1715146005153656 2023-01-21 10:02:00.883444: step: 76/529, loss: 0.12750092148780823 2023-01-21 10:02:02.013473: step: 80/529, loss: 0.22564248740673065 2023-01-21 10:02:03.107522: step: 84/529, loss: 0.3637579083442688 2023-01-21 10:02:04.216348: step: 88/529, loss: 0.1013256087899208 2023-01-21 10:02:05.324452: step: 92/529, loss: 0.17825783789157867 2023-01-21 10:02:06.443797: step: 96/529, loss: 0.03938794136047363 2023-01-21 10:02:07.541886: step: 100/529, loss: 0.09575791656970978 2023-01-21 10:02:08.672170: step: 104/529, loss: 0.01864643022418022 2023-01-21 10:02:09.810034: step: 108/529, loss: 0.043246082961559296 2023-01-21 10:02:10.909164: step: 112/529, loss: 0.04642505943775177 2023-01-21 10:02:12.004176: step: 116/529, loss: 0.09158239513635635 2023-01-21 10:02:13.109206: step: 120/529, loss: 0.3673229217529297 2023-01-21 10:02:14.223761: step: 124/529, loss: 0.06282272934913635 2023-01-21 10:02:15.342653: step: 128/529, loss: 0.030046749860048294 2023-01-21 10:02:16.455218: step: 132/529, loss: 0.913371741771698 2023-01-21 10:02:17.542005: step: 136/529, loss: 0.014583492651581764 2023-01-21 10:02:18.680695: step: 140/529, loss: 0.1392134726047516 2023-01-21 10:02:19.823377: step: 144/529, loss: 0.012332153506577015 2023-01-21 10:02:20.935482: step: 148/529, loss: 0.24209661781787872 2023-01-21 10:02:22.073575: step: 152/529, loss: 0.03439502790570259 2023-01-21 10:02:23.172888: step: 156/529, loss: 0.042597465217113495 2023-01-21 10:02:24.291878: step: 160/529, loss: 0.0428801067173481 2023-01-21 10:02:25.392548: step: 164/529, loss: 0.04165458679199219 2023-01-21 10:02:26.532062: step: 168/529, loss: 0.06124467775225639 2023-01-21 10:02:27.670803: step: 172/529, loss: 0.11269131302833557 2023-01-21 10:02:28.778413: step: 176/529, loss: 0.03401822969317436 2023-01-21 10:02:29.915559: step: 180/529, loss: 0.04655499383807182 2023-01-21 10:02:31.012056: step: 184/529, loss: 0.12600970268249512 2023-01-21 10:02:32.129596: step: 188/529, loss: 0.34927740693092346 2023-01-21 10:02:33.249234: step: 192/529, loss: 0.08241882175207138 2023-01-21 10:02:34.383442: step: 196/529, loss: 0.0403929203748703 2023-01-21 10:02:35.484546: step: 200/529, loss: 0.13260889053344727 2023-01-21 10:02:36.611011: step: 204/529, loss: 0.09866514056921005 2023-01-21 10:02:37.740071: step: 208/529, loss: 0.01948518678545952 2023-01-21 10:02:38.887845: step: 212/529, loss: 0.06461925804615021 2023-01-21 10:02:40.012957: step: 216/529, loss: 0.031583406031131744 2023-01-21 10:02:41.143715: step: 220/529, loss: 0.27349501848220825 2023-01-21 10:02:42.256181: step: 224/529, loss: 0.22106106579303741 2023-01-21 10:02:43.380627: step: 228/529, loss: 0.027596376836299896 2023-01-21 10:02:44.511663: step: 232/529, loss: 0.11111310124397278 2023-01-21 10:02:45.623236: step: 236/529, loss: 0.021947573870420456 2023-01-21 10:02:46.735686: step: 240/529, loss: 0.2623458802700043 2023-01-21 10:02:47.835893: step: 244/529, loss: 0.01020063180476427 2023-01-21 10:02:48.965481: step: 248/529, loss: 0.3105677664279938 2023-01-21 10:02:50.092398: step: 252/529, loss: 0.08165402710437775 2023-01-21 10:02:51.185957: step: 256/529, loss: 0.013071322813630104 2023-01-21 10:02:52.292295: step: 260/529, loss: 0.01860819011926651 2023-01-21 10:02:53.385578: step: 264/529, loss: 0.09093275666236877 2023-01-21 10:02:54.496444: step: 268/529, loss: 0.06408461928367615 2023-01-21 10:02:55.601842: step: 272/529, loss: 0.1365509033203125 2023-01-21 10:02:56.697420: step: 276/529, loss: 0.051250554621219635 2023-01-21 10:02:57.803359: step: 280/529, loss: 0.08675403892993927 2023-01-21 10:02:58.910600: step: 284/529, loss: 0.02930011786520481 2023-01-21 10:03:00.027033: step: 288/529, loss: 0.06257658451795578 2023-01-21 10:03:01.142547: step: 292/529, loss: 0.0641520544886589 2023-01-21 10:03:02.261118: step: 296/529, loss: 0.017392493784427643 2023-01-21 10:03:03.396141: step: 300/529, loss: 0.03448858484625816 2023-01-21 10:03:04.506847: step: 304/529, loss: 0.045565128326416016 2023-01-21 10:03:05.664302: step: 308/529, loss: 0.09231019020080566 2023-01-21 10:03:06.775595: step: 312/529, loss: 0.1087920218706131 2023-01-21 10:03:07.951614: step: 316/529, loss: 0.04938516765832901 2023-01-21 10:03:09.089292: step: 320/529, loss: 0.08481334149837494 2023-01-21 10:03:10.218398: step: 324/529, loss: 0.03702735900878906 2023-01-21 10:03:11.316053: step: 328/529, loss: 0.13977545499801636 2023-01-21 10:03:12.434606: step: 332/529, loss: 0.20704421401023865 2023-01-21 10:03:13.543068: step: 336/529, loss: 0.30477452278137207 2023-01-21 10:03:14.686212: step: 340/529, loss: 8.35095500946045 2023-01-21 10:03:15.817122: step: 344/529, loss: 0.0928533598780632 2023-01-21 10:03:16.963356: step: 348/529, loss: 0.09858093410730362 2023-01-21 10:03:18.068293: step: 352/529, loss: 0.04527135193347931 2023-01-21 10:03:19.183026: step: 356/529, loss: 0.0664377212524414 2023-01-21 10:03:20.320271: step: 360/529, loss: 0.06432053446769714 2023-01-21 10:03:21.445474: step: 364/529, loss: 0.060967445373535156 2023-01-21 10:03:22.545966: step: 368/529, loss: 0.017093658447265625 2023-01-21 10:03:23.675443: step: 372/529, loss: 0.35670509934425354 2023-01-21 10:03:24.811546: step: 376/529, loss: 0.05980215221643448 2023-01-21 10:03:25.935138: step: 380/529, loss: 0.08419743180274963 2023-01-21 10:03:27.043618: step: 384/529, loss: 0.02702946774661541 2023-01-21 10:03:28.163085: step: 388/529, loss: 0.0651729553937912 2023-01-21 10:03:29.328301: step: 392/529, loss: 0.0750269889831543 2023-01-21 10:03:30.472494: step: 396/529, loss: 0.031039047986268997 2023-01-21 10:03:31.571832: step: 400/529, loss: 0.13804897665977478 2023-01-21 10:03:32.746891: step: 404/529, loss: 0.016185570508241653 2023-01-21 10:03:33.859366: step: 408/529, loss: 0.01443338394165039 2023-01-21 10:03:34.958552: step: 412/529, loss: 0.14113016426563263 2023-01-21 10:03:36.052963: step: 416/529, loss: 0.23866643011569977 2023-01-21 10:03:37.169859: step: 420/529, loss: 0.19302988052368164 2023-01-21 10:03:38.278490: step: 424/529, loss: 0.3451392948627472 2023-01-21 10:03:39.379394: step: 428/529, loss: 0.04785561561584473 2023-01-21 10:03:40.518568: step: 432/529, loss: 0.07870607078075409 2023-01-21 10:03:41.610609: step: 436/529, loss: 0.08337584137916565 2023-01-21 10:03:42.763814: step: 440/529, loss: 0.024120450019836426 2023-01-21 10:03:43.887371: step: 444/529, loss: 0.04644451290369034 2023-01-21 10:03:45.019509: step: 448/529, loss: 0.07648640125989914 2023-01-21 10:03:46.174718: step: 452/529, loss: 0.10845880955457687 2023-01-21 10:03:47.276668: step: 456/529, loss: 0.1912718415260315 2023-01-21 10:03:48.391181: step: 460/529, loss: 0.11780153214931488 2023-01-21 10:03:49.503845: step: 464/529, loss: 0.1476554423570633 2023-01-21 10:03:50.638940: step: 468/529, loss: 0.0731138214468956 2023-01-21 10:03:51.763882: step: 472/529, loss: 0.015170956030488014 2023-01-21 10:03:52.839455: step: 476/529, loss: 0.11919479817152023 2023-01-21 10:03:53.960789: step: 480/529, loss: 0.16263265907764435 2023-01-21 10:03:55.085565: step: 484/529, loss: 0.1139630377292633 2023-01-21 10:03:56.183648: step: 488/529, loss: 0.18790999054908752 2023-01-21 10:03:57.302756: step: 492/529, loss: 0.0621342696249485 2023-01-21 10:03:58.430304: step: 496/529, loss: 0.0645180195569992 2023-01-21 10:03:59.534242: step: 500/529, loss: 0.05574359744787216 2023-01-21 10:04:00.652285: step: 504/529, loss: 0.14265528321266174 2023-01-21 10:04:01.758666: step: 508/529, loss: 0.07000589370727539 2023-01-21 10:04:02.849278: step: 512/529, loss: 0.09280486404895782 2023-01-21 10:04:03.993436: step: 516/529, loss: 0.14939813315868378 2023-01-21 10:04:05.112166: step: 520/529, loss: 0.05603532865643501 2023-01-21 10:04:06.220769: step: 524/529, loss: 0.0800023078918457 2023-01-21 10:04:07.327220: step: 528/529, loss: 0.05761460214853287 2023-01-21 10:04:08.415896: step: 532/529, loss: 0.040703631937503815 2023-01-21 10:04:09.554420: step: 536/529, loss: 0.15799598395824432 2023-01-21 10:04:10.663959: step: 540/529, loss: 0.01688542403280735 2023-01-21 10:04:11.764040: step: 544/529, loss: 0.10103998333215714 2023-01-21 10:04:12.902320: step: 548/529, loss: 0.028824234381318092 2023-01-21 10:04:14.028017: step: 552/529, loss: 0.03666649013757706 2023-01-21 10:04:15.140103: step: 556/529, loss: 0.1671028435230255 2023-01-21 10:04:16.270345: step: 560/529, loss: 0.1756400167942047 2023-01-21 10:04:17.397062: step: 564/529, loss: 0.015729619190096855 2023-01-21 10:04:18.504195: step: 568/529, loss: 0.10626716911792755 2023-01-21 10:04:19.592752: step: 572/529, loss: 0.011851310729980469 2023-01-21 10:04:20.691617: step: 576/529, loss: 0.05698714405298233 2023-01-21 10:04:21.808969: step: 580/529, loss: 0.027351761236786842 2023-01-21 10:04:22.936993: step: 584/529, loss: 0.04248428717255592 2023-01-21 10:04:24.024183: step: 588/529, loss: 0.02740459330379963 2023-01-21 10:04:25.110506: step: 592/529, loss: 0.06386041641235352 2023-01-21 10:04:26.240867: step: 596/529, loss: 0.0068108560517430305 2023-01-21 10:04:27.360459: step: 600/529, loss: 0.05725877359509468 2023-01-21 10:04:28.483421: step: 604/529, loss: 0.08643150329589844 2023-01-21 10:04:29.595664: step: 608/529, loss: 0.6223519444465637 2023-01-21 10:04:30.711944: step: 612/529, loss: 0.04120521619915962 2023-01-21 10:04:31.821065: step: 616/529, loss: 0.007879400625824928 2023-01-21 10:04:32.948220: step: 620/529, loss: 0.04057574272155762 2023-01-21 10:04:34.083263: step: 624/529, loss: 0.10579729080200195 2023-01-21 10:04:35.187118: step: 628/529, loss: 0.057329509407281876 2023-01-21 10:04:36.307030: step: 632/529, loss: 0.06602416187524796 2023-01-21 10:04:37.413175: step: 636/529, loss: 0.09648457169532776 2023-01-21 10:04:38.519148: step: 640/529, loss: 0.05993414297699928 2023-01-21 10:04:39.644239: step: 644/529, loss: 0.016345595940947533 2023-01-21 10:04:40.767457: step: 648/529, loss: 0.12050972133874893 2023-01-21 10:04:41.906735: step: 652/529, loss: 0.13320191204547882 2023-01-21 10:04:43.027176: step: 656/529, loss: 0.07330746948719025 2023-01-21 10:04:44.125939: step: 660/529, loss: 0.0422452948987484 2023-01-21 10:04:45.248424: step: 664/529, loss: 0.06081104278564453 2023-01-21 10:04:46.382499: step: 668/529, loss: 0.02058258093893528 2023-01-21 10:04:47.520879: step: 672/529, loss: 0.49855002760887146 2023-01-21 10:04:48.620367: step: 676/529, loss: 0.18403233587741852 2023-01-21 10:04:49.742973: step: 680/529, loss: 0.31557580828666687 2023-01-21 10:04:50.860769: step: 684/529, loss: 0.012333488091826439 2023-01-21 10:04:51.952166: step: 688/529, loss: 0.02431764453649521 2023-01-21 10:04:53.096532: step: 692/529, loss: 0.28456413745880127 2023-01-21 10:04:54.216804: step: 696/529, loss: 0.17587552964687347 2023-01-21 10:04:55.316680: step: 700/529, loss: 0.10035070031881332 2023-01-21 10:04:56.407682: step: 704/529, loss: 0.04790849983692169 2023-01-21 10:04:57.538829: step: 708/529, loss: 0.06891946494579315 2023-01-21 10:04:58.655859: step: 712/529, loss: 0.028696918860077858 2023-01-21 10:04:59.811625: step: 716/529, loss: 0.01744098775088787 2023-01-21 10:05:00.913553: step: 720/529, loss: 0.05084972456097603 2023-01-21 10:05:02.028161: step: 724/529, loss: 0.7905058264732361 2023-01-21 10:05:03.134762: step: 728/529, loss: 0.023940706625580788 2023-01-21 10:05:04.249733: step: 732/529, loss: 0.048513222485780716 2023-01-21 10:05:05.342307: step: 736/529, loss: 0.04477844387292862 2023-01-21 10:05:06.447261: step: 740/529, loss: 0.042609453201293945 2023-01-21 10:05:07.542731: step: 744/529, loss: 0.13574886322021484 2023-01-21 10:05:08.656814: step: 748/529, loss: 0.057868484407663345 2023-01-21 10:05:09.748516: step: 752/529, loss: 0.077783964574337 2023-01-21 10:05:10.872023: step: 756/529, loss: 0.08376830816268921 2023-01-21 10:05:11.991546: step: 760/529, loss: 0.6848520636558533 2023-01-21 10:05:13.107545: step: 764/529, loss: 0.06643247604370117 2023-01-21 10:05:14.241703: step: 768/529, loss: 0.014272689819335938 2023-01-21 10:05:15.370423: step: 772/529, loss: 0.044089894741773605 2023-01-21 10:05:16.472882: step: 776/529, loss: 0.001986376941204071 2023-01-21 10:05:17.580299: step: 780/529, loss: 0.026595210656523705 2023-01-21 10:05:18.697735: step: 784/529, loss: 0.10692854225635529 2023-01-21 10:05:19.786936: step: 788/529, loss: 0.028864096850156784 2023-01-21 10:05:20.946560: step: 792/529, loss: 0.02127533033490181 2023-01-21 10:05:22.086947: step: 796/529, loss: 0.3346277177333832 2023-01-21 10:05:23.215852: step: 800/529, loss: 0.12227492034435272 2023-01-21 10:05:24.357900: step: 804/529, loss: 0.006898021791130304 2023-01-21 10:05:25.476749: step: 808/529, loss: 0.11880576610565186 2023-01-21 10:05:26.587283: step: 812/529, loss: 0.014857769012451172 2023-01-21 10:05:27.739060: step: 816/529, loss: 0.1547391712665558 2023-01-21 10:05:28.851962: step: 820/529, loss: 0.016665935516357422 2023-01-21 10:05:29.967648: step: 824/529, loss: 0.044320181012153625 2023-01-21 10:05:31.049751: step: 828/529, loss: 0.14291496574878693 2023-01-21 10:05:32.181964: step: 832/529, loss: 0.17214222252368927 2023-01-21 10:05:33.289665: step: 836/529, loss: 0.06896086037158966 2023-01-21 10:05:34.393486: step: 840/529, loss: 0.04852494224905968 2023-01-21 10:05:35.496375: step: 844/529, loss: 0.06520214676856995 2023-01-21 10:05:36.605186: step: 848/529, loss: 0.07643423229455948 2023-01-21 10:05:37.734076: step: 852/529, loss: 0.07438163459300995 2023-01-21 10:05:38.860340: step: 856/529, loss: 0.05007372051477432 2023-01-21 10:05:39.971560: step: 860/529, loss: 0.112081378698349 2023-01-21 10:05:41.093890: step: 864/529, loss: 0.06313200294971466 2023-01-21 10:05:42.193333: step: 868/529, loss: 0.05984840914607048 2023-01-21 10:05:43.310260: step: 872/529, loss: 0.03737363964319229 2023-01-21 10:05:44.430411: step: 876/529, loss: 0.04383154213428497 2023-01-21 10:05:45.523507: step: 880/529, loss: 0.18145990371704102 2023-01-21 10:05:46.636180: step: 884/529, loss: 0.07630787789821625 2023-01-21 10:05:47.752200: step: 888/529, loss: 0.03633446618914604 2023-01-21 10:05:48.840868: step: 892/529, loss: 0.03711829334497452 2023-01-21 10:05:49.937949: step: 896/529, loss: 0.04978237301111221 2023-01-21 10:05:51.037365: step: 900/529, loss: 0.09621293842792511 2023-01-21 10:05:52.182989: step: 904/529, loss: 0.1381158083677292 2023-01-21 10:05:53.300790: step: 908/529, loss: 0.07454118132591248 2023-01-21 10:05:54.420164: step: 912/529, loss: 0.6612171530723572 2023-01-21 10:05:55.557210: step: 916/529, loss: 0.09623517841100693 2023-01-21 10:05:56.677821: step: 920/529, loss: 0.04454955831170082 2023-01-21 10:05:57.813855: step: 924/529, loss: 0.06601305305957794 2023-01-21 10:05:58.939787: step: 928/529, loss: 0.060262106359004974 2023-01-21 10:06:00.032022: step: 932/529, loss: 0.0812913030385971 2023-01-21 10:06:01.148403: step: 936/529, loss: 0.05629901960492134 2023-01-21 10:06:02.253522: step: 940/529, loss: 0.1083926185965538 2023-01-21 10:06:03.378625: step: 944/529, loss: 0.044487953186035156 2023-01-21 10:06:04.482256: step: 948/529, loss: 0.09143247455358505 2023-01-21 10:06:05.610564: step: 952/529, loss: 0.13394412398338318 2023-01-21 10:06:06.724550: step: 956/529, loss: 0.06732816994190216 2023-01-21 10:06:07.857428: step: 960/529, loss: 0.08902082592248917 2023-01-21 10:06:08.973976: step: 964/529, loss: 0.08154439926147461 2023-01-21 10:06:10.101404: step: 968/529, loss: 0.032379150390625 2023-01-21 10:06:11.224703: step: 972/529, loss: 0.10394573211669922 2023-01-21 10:06:12.354849: step: 976/529, loss: 0.33629971742630005 2023-01-21 10:06:13.493799: step: 980/529, loss: 0.5062699317932129 2023-01-21 10:06:14.603452: step: 984/529, loss: 0.023540304973721504 2023-01-21 10:06:15.717510: step: 988/529, loss: 0.0951499491930008 2023-01-21 10:06:16.858550: step: 992/529, loss: 0.01879253424704075 2023-01-21 10:06:17.998708: step: 996/529, loss: 0.012501239776611328 2023-01-21 10:06:19.101243: step: 1000/529, loss: 0.10540914535522461 2023-01-21 10:06:20.227447: step: 1004/529, loss: 0.2552165985107422 2023-01-21 10:06:21.345276: step: 1008/529, loss: 0.05841560289263725 2023-01-21 10:06:22.457525: step: 1012/529, loss: 0.1975843459367752 2023-01-21 10:06:23.623975: step: 1016/529, loss: 0.060215093195438385 2023-01-21 10:06:24.765228: step: 1020/529, loss: 0.0384943000972271 2023-01-21 10:06:25.918923: step: 1024/529, loss: 0.06896467506885529 2023-01-21 10:06:27.056412: step: 1028/529, loss: 0.018214941024780273 2023-01-21 10:06:28.177531: step: 1032/529, loss: 0.11634024977684021 2023-01-21 10:06:29.301462: step: 1036/529, loss: 0.3821887969970703 2023-01-21 10:06:30.405776: step: 1040/529, loss: 0.2518870234489441 2023-01-21 10:06:31.512991: step: 1044/529, loss: 0.11563281714916229 2023-01-21 10:06:32.622673: step: 1048/529, loss: 0.12849339842796326 2023-01-21 10:06:33.715945: step: 1052/529, loss: 0.1095949113368988 2023-01-21 10:06:34.810763: step: 1056/529, loss: 0.32417404651641846 2023-01-21 10:06:35.921929: step: 1060/529, loss: 0.008027458563446999 2023-01-21 10:06:37.032896: step: 1064/529, loss: 0.06524324417114258 2023-01-21 10:06:38.175185: step: 1068/529, loss: 0.06365480273962021 2023-01-21 10:06:39.284531: step: 1072/529, loss: 0.031239653006196022 2023-01-21 10:06:40.396877: step: 1076/529, loss: 0.06233043596148491 2023-01-21 10:06:41.507358: step: 1080/529, loss: 0.10023470222949982 2023-01-21 10:06:42.616243: step: 1084/529, loss: 0.19548740983009338 2023-01-21 10:06:43.754760: step: 1088/529, loss: 0.08747239410877228 2023-01-21 10:06:44.881838: step: 1092/529, loss: 0.026029586791992188 2023-01-21 10:06:46.016384: step: 1096/529, loss: 0.05856723710894585 2023-01-21 10:06:47.204999: step: 1100/529, loss: 0.0699123814702034 2023-01-21 10:06:48.318544: step: 1104/529, loss: 0.060286711901426315 2023-01-21 10:06:49.418457: step: 1108/529, loss: 0.030222607776522636 2023-01-21 10:06:50.559457: step: 1112/529, loss: 0.09417829662561417 2023-01-21 10:06:51.658393: step: 1116/529, loss: 0.01377573050558567 2023-01-21 10:06:52.788530: step: 1120/529, loss: 0.1039462611079216 2023-01-21 10:06:53.882642: step: 1124/529, loss: 0.06650018692016602 2023-01-21 10:06:54.982833: step: 1128/529, loss: 0.045432060956954956 2023-01-21 10:06:56.074024: step: 1132/529, loss: 0.6863029599189758 2023-01-21 10:06:57.177979: step: 1136/529, loss: 0.18507003784179688 2023-01-21 10:06:58.322561: step: 1140/529, loss: 0.03508634865283966 2023-01-21 10:06:59.442707: step: 1144/529, loss: 0.040506839752197266 2023-01-21 10:07:00.562989: step: 1148/529, loss: 0.18322515487670898 2023-01-21 10:07:01.686432: step: 1152/529, loss: 0.1081615537405014 2023-01-21 10:07:02.800908: step: 1156/529, loss: 0.044408418238162994 2023-01-21 10:07:03.937868: step: 1160/529, loss: 0.04965457692742348 2023-01-21 10:07:05.075565: step: 1164/529, loss: 0.21872863173484802 2023-01-21 10:07:06.192774: step: 1168/529, loss: 0.08637867122888565 2023-01-21 10:07:07.364499: step: 1172/529, loss: 0.07635793834924698 2023-01-21 10:07:08.466055: step: 1176/529, loss: 0.07406573742628098 2023-01-21 10:07:09.559046: step: 1180/529, loss: 0.0973435491323471 2023-01-21 10:07:10.701624: step: 1184/529, loss: 0.04751301184296608 2023-01-21 10:07:11.803959: step: 1188/529, loss: 0.02748269960284233 2023-01-21 10:07:12.936587: step: 1192/529, loss: 0.2465149611234665 2023-01-21 10:07:14.071246: step: 1196/529, loss: 0.041822243481874466 2023-01-21 10:07:15.203332: step: 1200/529, loss: 0.08025749027729034 2023-01-21 10:07:16.340563: step: 1204/529, loss: 0.09643816947937012 2023-01-21 10:07:17.486634: step: 1208/529, loss: 0.0659547820687294 2023-01-21 10:07:18.591950: step: 1212/529, loss: 0.18151094019412994 2023-01-21 10:07:19.699034: step: 1216/529, loss: 0.07799167931079865 2023-01-21 10:07:20.828032: step: 1220/529, loss: 0.09307403862476349 2023-01-21 10:07:21.955949: step: 1224/529, loss: 0.006068134680390358 2023-01-21 10:07:23.075368: step: 1228/529, loss: 0.13911017775535583 2023-01-21 10:07:24.203902: step: 1232/529, loss: 0.06770257651805878 2023-01-21 10:07:25.320663: step: 1236/529, loss: 0.1125824972987175 2023-01-21 10:07:26.437677: step: 1240/529, loss: 0.06380510330200195 2023-01-21 10:07:27.565114: step: 1244/529, loss: 0.1834530383348465 2023-01-21 10:07:28.649894: step: 1248/529, loss: 0.03923363983631134 2023-01-21 10:07:29.745164: step: 1252/529, loss: 0.11145992577075958 2023-01-21 10:07:30.888424: step: 1256/529, loss: 0.08441410213708878 2023-01-21 10:07:31.984730: step: 1260/529, loss: 0.07938718795776367 2023-01-21 10:07:33.121654: step: 1264/529, loss: 0.02035675011575222 2023-01-21 10:07:34.252671: step: 1268/529, loss: 0.10938110947608948 2023-01-21 10:07:35.361853: step: 1272/529, loss: 0.19580736756324768 2023-01-21 10:07:36.484229: step: 1276/529, loss: 0.14145508408546448 2023-01-21 10:07:37.636710: step: 1280/529, loss: 0.04451394081115723 2023-01-21 10:07:38.763633: step: 1284/529, loss: 0.003300571581348777 2023-01-21 10:07:39.877912: step: 1288/529, loss: 0.3471525311470032 2023-01-21 10:07:41.015005: step: 1292/529, loss: 0.21434137225151062 2023-01-21 10:07:42.134706: step: 1296/529, loss: 0.0013439178001135588 2023-01-21 10:07:43.239126: step: 1300/529, loss: 0.13756266236305237 2023-01-21 10:07:44.335410: step: 1304/529, loss: 0.04944667965173721 2023-01-21 10:07:45.424883: step: 1308/529, loss: 0.1475081443786621 2023-01-21 10:07:46.531988: step: 1312/529, loss: 0.009082937613129616 2023-01-21 10:07:47.655002: step: 1316/529, loss: 0.04438390955328941 2023-01-21 10:07:48.808255: step: 1320/529, loss: 0.015398358926177025 2023-01-21 10:07:49.916002: step: 1324/529, loss: 0.08248873054981232 2023-01-21 10:07:51.025019: step: 1328/529, loss: 0.0918426513671875 2023-01-21 10:07:52.146859: step: 1332/529, loss: 0.04811515659093857 2023-01-21 10:07:53.268658: step: 1336/529, loss: 0.12027950584888458 2023-01-21 10:07:54.373882: step: 1340/529, loss: 0.04040975868701935 2023-01-21 10:07:55.520587: step: 1344/529, loss: 0.06928618252277374 2023-01-21 10:07:56.643920: step: 1348/529, loss: 0.07258549332618713 2023-01-21 10:07:57.745276: step: 1352/529, loss: 0.10316307842731476 2023-01-21 10:07:58.879286: step: 1356/529, loss: 0.14421701431274414 2023-01-21 10:07:59.968363: step: 1360/529, loss: 0.04028075933456421 2023-01-21 10:08:01.078067: step: 1364/529, loss: 0.025164222344756126 2023-01-21 10:08:02.205183: step: 1368/529, loss: 0.1758926957845688 2023-01-21 10:08:03.305093: step: 1372/529, loss: 0.031922150403261185 2023-01-21 10:08:04.414351: step: 1376/529, loss: 0.14625835418701172 2023-01-21 10:08:05.539044: step: 1380/529, loss: 0.2937980592250824 2023-01-21 10:08:06.650390: step: 1384/529, loss: 0.17285165190696716 2023-01-21 10:08:07.752077: step: 1388/529, loss: 0.007736682891845703 2023-01-21 10:08:08.860447: step: 1392/529, loss: 0.1878342628479004 2023-01-21 10:08:09.960150: step: 1396/529, loss: 0.13419370353221893 2023-01-21 10:08:11.089516: step: 1400/529, loss: 0.02635507471859455 2023-01-21 10:08:12.220924: step: 1404/529, loss: 0.05696258693933487 2023-01-21 10:08:13.343276: step: 1408/529, loss: 0.009402275085449219 2023-01-21 10:08:14.495337: step: 1412/529, loss: 0.12353183329105377 2023-01-21 10:08:15.601795: step: 1416/529, loss: 0.08068886399269104 2023-01-21 10:08:16.748322: step: 1420/529, loss: 0.02467794343829155 2023-01-21 10:08:17.846989: step: 1424/529, loss: 0.060294248163700104 2023-01-21 10:08:18.949706: step: 1428/529, loss: 0.08148124814033508 2023-01-21 10:08:20.076639: step: 1432/529, loss: 0.10907106101512909 2023-01-21 10:08:21.240651: step: 1436/529, loss: 0.14531207084655762 2023-01-21 10:08:22.338961: step: 1440/529, loss: 0.00984878558665514 2023-01-21 10:08:23.471787: step: 1444/529, loss: 0.014052271842956543 2023-01-21 10:08:24.603540: step: 1448/529, loss: 0.5679744482040405 2023-01-21 10:08:25.736016: step: 1452/529, loss: 0.09771182388067245 2023-01-21 10:08:26.850179: step: 1456/529, loss: 0.0191174503415823 2023-01-21 10:08:27.959556: step: 1460/529, loss: 0.0334656722843647 2023-01-21 10:08:29.061144: step: 1464/529, loss: 0.2043953835964203 2023-01-21 10:08:30.191693: step: 1468/529, loss: 0.009652710519731045 2023-01-21 10:08:31.303217: step: 1472/529, loss: 0.052018169313669205 2023-01-21 10:08:32.422239: step: 1476/529, loss: 0.11485262215137482 2023-01-21 10:08:33.564647: step: 1480/529, loss: 0.07693605870008469 2023-01-21 10:08:34.686913: step: 1484/529, loss: 0.13758641481399536 2023-01-21 10:08:35.801135: step: 1488/529, loss: 0.03982071578502655 2023-01-21 10:08:36.916201: step: 1492/529, loss: 0.0795324370265007 2023-01-21 10:08:38.057308: step: 1496/529, loss: 0.08048317581415176 2023-01-21 10:08:39.177218: step: 1500/529, loss: 0.03495984151959419 2023-01-21 10:08:40.302890: step: 1504/529, loss: 0.05106806755065918 2023-01-21 10:08:41.441777: step: 1508/529, loss: 0.1111084371805191 2023-01-21 10:08:42.545551: step: 1512/529, loss: 0.026851128786802292 2023-01-21 10:08:43.652052: step: 1516/529, loss: 0.2244413197040558 2023-01-21 10:08:44.783625: step: 1520/529, loss: 0.29151174426078796 2023-01-21 10:08:45.876729: step: 1524/529, loss: 0.13247866928577423 2023-01-21 10:08:47.001551: step: 1528/529, loss: 0.05381937325000763 2023-01-21 10:08:48.094101: step: 1532/529, loss: 0.013446998782455921 2023-01-21 10:08:49.197264: step: 1536/529, loss: 0.11355744302272797 2023-01-21 10:08:50.310087: step: 1540/529, loss: 0.5881436467170715 2023-01-21 10:08:51.434237: step: 1544/529, loss: 0.16452428698539734 2023-01-21 10:08:52.552871: step: 1548/529, loss: 0.018399905413389206 2023-01-21 10:08:53.692234: step: 1552/529, loss: 0.1797565519809723 2023-01-21 10:08:54.830891: step: 1556/529, loss: 0.10114951431751251 2023-01-21 10:08:55.936553: step: 1560/529, loss: 0.3463018536567688 2023-01-21 10:08:57.033734: step: 1564/529, loss: 0.0870586410164833 2023-01-21 10:08:58.121486: step: 1568/529, loss: 0.08252926170825958 2023-01-21 10:08:59.243717: step: 1572/529, loss: 0.17114019393920898 2023-01-21 10:09:00.364638: step: 1576/529, loss: 0.10457129776477814 2023-01-21 10:09:01.483347: step: 1580/529, loss: 0.07745208591222763 2023-01-21 10:09:02.596878: step: 1584/529, loss: 0.03918357193470001 2023-01-21 10:09:03.737216: step: 1588/529, loss: 0.12574857473373413 2023-01-21 10:09:04.822127: step: 1592/529, loss: 0.1182066947221756 2023-01-21 10:09:05.969464: step: 1596/529, loss: 0.010142517276108265 2023-01-21 10:09:07.084271: step: 1600/529, loss: 0.08751335740089417 2023-01-21 10:09:08.202758: step: 1604/529, loss: 0.11351537704467773 2023-01-21 10:09:09.312211: step: 1608/529, loss: 0.9204500317573547 2023-01-21 10:09:10.423677: step: 1612/529, loss: 0.05030689388513565 2023-01-21 10:09:11.562814: step: 1616/529, loss: 0.1784447729587555 2023-01-21 10:09:12.717509: step: 1620/529, loss: 0.05158596113324165 2023-01-21 10:09:13.817709: step: 1624/529, loss: 0.05183391645550728 2023-01-21 10:09:14.926068: step: 1628/529, loss: 0.60284423828125 2023-01-21 10:09:16.054850: step: 1632/529, loss: 0.003727531526237726 2023-01-21 10:09:17.193466: step: 1636/529, loss: 0.08788242936134338 2023-01-21 10:09:18.330314: step: 1640/529, loss: 0.025651073083281517 2023-01-21 10:09:19.409929: step: 1644/529, loss: 0.017536640167236328 2023-01-21 10:09:20.519253: step: 1648/529, loss: 0.08216686546802521 2023-01-21 10:09:21.630712: step: 1652/529, loss: 0.017567062750458717 2023-01-21 10:09:22.773274: step: 1656/529, loss: 0.10507622361183167 2023-01-21 10:09:23.891592: step: 1660/529, loss: 0.1343797743320465 2023-01-21 10:09:25.013275: step: 1664/529, loss: 0.0984993502497673 2023-01-21 10:09:26.103073: step: 1668/529, loss: 0.03480367735028267 2023-01-21 10:09:27.217796: step: 1672/529, loss: 0.07474584877490997 2023-01-21 10:09:28.354438: step: 1676/529, loss: 0.029497720301151276 2023-01-21 10:09:29.487877: step: 1680/529, loss: 0.02427997626364231 2023-01-21 10:09:30.611135: step: 1684/529, loss: 0.04129553213715553 2023-01-21 10:09:31.765323: step: 1688/529, loss: 0.0964839905500412 2023-01-21 10:09:32.866431: step: 1692/529, loss: 0.1205843985080719 2023-01-21 10:09:33.999283: step: 1696/529, loss: 0.026320649310946465 2023-01-21 10:09:35.141119: step: 1700/529, loss: 0.18613243103027344 2023-01-21 10:09:36.234680: step: 1704/529, loss: 0.010318756103515625 2023-01-21 10:09:37.318733: step: 1708/529, loss: 0.015778400003910065 2023-01-21 10:09:38.447000: step: 1712/529, loss: 0.3122726380825043 2023-01-21 10:09:39.548335: step: 1716/529, loss: 0.021357107907533646 2023-01-21 10:09:40.638330: step: 1720/529, loss: 0.05111951753497124 2023-01-21 10:09:41.726313: step: 1724/529, loss: 0.005657053552567959 2023-01-21 10:09:42.871546: step: 1728/529, loss: 0.11666889488697052 2023-01-21 10:09:43.985228: step: 1732/529, loss: 0.41078507900238037 2023-01-21 10:09:45.092062: step: 1736/529, loss: 0.36201146245002747 2023-01-21 10:09:46.245412: step: 1740/529, loss: 0.02566089667379856 2023-01-21 10:09:47.327376: step: 1744/529, loss: 0.22240696847438812 2023-01-21 10:09:48.433794: step: 1748/529, loss: 0.04610276222229004 2023-01-21 10:09:49.562057: step: 1752/529, loss: 0.04786386340856552 2023-01-21 10:09:50.669990: step: 1756/529, loss: 0.039374664425849915 2023-01-21 10:09:51.754282: step: 1760/529, loss: 0.08335819840431213 2023-01-21 10:09:52.877547: step: 1764/529, loss: 0.07185249775648117 2023-01-21 10:09:54.010560: step: 1768/529, loss: 0.15856905281543732 2023-01-21 10:09:55.133937: step: 1772/529, loss: 0.03192448616027832 2023-01-21 10:09:56.257792: step: 1776/529, loss: 0.12803632020950317 2023-01-21 10:09:57.380275: step: 1780/529, loss: 0.19158053398132324 2023-01-21 10:09:58.495576: step: 1784/529, loss: 0.630646824836731 2023-01-21 10:09:59.627136: step: 1788/529, loss: 0.021214868873357773 2023-01-21 10:10:00.734888: step: 1792/529, loss: 0.07143059372901917 2023-01-21 10:10:01.869552: step: 1796/529, loss: 0.1357056051492691 2023-01-21 10:10:03.003869: step: 1800/529, loss: 0.04163026809692383 2023-01-21 10:10:04.089020: step: 1804/529, loss: 0.05554194375872612 2023-01-21 10:10:05.179775: step: 1808/529, loss: 0.001104164170101285 2023-01-21 10:10:06.301428: step: 1812/529, loss: 0.1678251326084137 2023-01-21 10:10:07.434143: step: 1816/529, loss: 0.09261083602905273 2023-01-21 10:10:08.590405: step: 1820/529, loss: 0.07316761463880539 2023-01-21 10:10:09.708008: step: 1824/529, loss: 0.16719341278076172 2023-01-21 10:10:10.817443: step: 1828/529, loss: 0.25669464468955994 2023-01-21 10:10:11.917880: step: 1832/529, loss: 0.02870168723165989 2023-01-21 10:10:13.044423: step: 1836/529, loss: 0.05384483560919762 2023-01-21 10:10:14.159890: step: 1840/529, loss: 0.023204613476991653 2023-01-21 10:10:15.304987: step: 1844/529, loss: 0.3067391514778137 2023-01-21 10:10:16.439900: step: 1848/529, loss: 0.1798103302717209 2023-01-21 10:10:17.538678: step: 1852/529, loss: 0.024246692657470703 2023-01-21 10:10:18.623283: step: 1856/529, loss: 0.19331133365631104 2023-01-21 10:10:19.735894: step: 1860/529, loss: 0.1499234288930893 2023-01-21 10:10:20.856846: step: 1864/529, loss: 0.08906231075525284 2023-01-21 10:10:21.928791: step: 1868/529, loss: 0.013643741607666016 2023-01-21 10:10:23.023706: step: 1872/529, loss: 0.10728979110717773 2023-01-21 10:10:24.124127: step: 1876/529, loss: 0.2997283935546875 2023-01-21 10:10:25.221547: step: 1880/529, loss: 0.0766817107796669 2023-01-21 10:10:26.325618: step: 1884/529, loss: 0.0796487033367157 2023-01-21 10:10:27.448489: step: 1888/529, loss: 0.02410421520471573 2023-01-21 10:10:28.562911: step: 1892/529, loss: 0.26116690039634705 2023-01-21 10:10:29.679869: step: 1896/529, loss: 0.12138329446315765 2023-01-21 10:10:30.793552: step: 1900/529, loss: 0.2350689023733139 2023-01-21 10:10:31.893643: step: 1904/529, loss: 0.11842866241931915 2023-01-21 10:10:33.068653: step: 1908/529, loss: 0.05647840350866318 2023-01-21 10:10:34.197057: step: 1912/529, loss: 0.06652762740850449 2023-01-21 10:10:35.326366: step: 1916/529, loss: 0.19929274916648865 2023-01-21 10:10:36.440821: step: 1920/529, loss: 0.18102556467056274 2023-01-21 10:10:37.560502: step: 1924/529, loss: 0.16160912811756134 2023-01-21 10:10:38.668944: step: 1928/529, loss: 0.017178857699036598 2023-01-21 10:10:39.788820: step: 1932/529, loss: 0.07090263068675995 2023-01-21 10:10:40.874279: step: 1936/529, loss: 0.05810260772705078 2023-01-21 10:10:42.027713: step: 1940/529, loss: 0.1487790048122406 2023-01-21 10:10:43.134610: step: 1944/529, loss: 0.08140314370393753 2023-01-21 10:10:44.262121: step: 1948/529, loss: 0.17723149061203003 2023-01-21 10:10:45.368323: step: 1952/529, loss: 0.028296852484345436 2023-01-21 10:10:46.498296: step: 1956/529, loss: 0.0789031982421875 2023-01-21 10:10:47.600067: step: 1960/529, loss: 0.004878330510109663 2023-01-21 10:10:48.728808: step: 1964/529, loss: 0.08981533348560333 2023-01-21 10:10:49.817242: step: 1968/529, loss: 0.051788900047540665 2023-01-21 10:10:50.947921: step: 1972/529, loss: 0.18572577834129333 2023-01-21 10:10:52.071746: step: 1976/529, loss: 0.5622392892837524 2023-01-21 10:10:53.193244: step: 1980/529, loss: 0.05681190639734268 2023-01-21 10:10:54.306234: step: 1984/529, loss: 0.056949131190776825 2023-01-21 10:10:55.452022: step: 1988/529, loss: 0.11759801208972931 2023-01-21 10:10:56.576007: step: 1992/529, loss: 0.09613952785730362 2023-01-21 10:10:57.685638: step: 1996/529, loss: 0.04975314438343048 2023-01-21 10:10:58.834996: step: 2000/529, loss: 0.045046284794807434 2023-01-21 10:10:59.931844: step: 2004/529, loss: 0.09842129051685333 2023-01-21 10:11:01.023335: step: 2008/529, loss: 0.14135675132274628 2023-01-21 10:11:02.150320: step: 2012/529, loss: 0.1797056347131729 2023-01-21 10:11:03.266863: step: 2016/529, loss: 0.05608811229467392 2023-01-21 10:11:04.375386: step: 2020/529, loss: 0.04219484701752663 2023-01-21 10:11:05.493830: step: 2024/529, loss: 0.03295278549194336 2023-01-21 10:11:06.613139: step: 2028/529, loss: 0.020505620166659355 2023-01-21 10:11:07.753559: step: 2032/529, loss: 0.03445548936724663 2023-01-21 10:11:08.858573: step: 2036/529, loss: 0.003640031674876809 2023-01-21 10:11:09.950309: step: 2040/529, loss: 0.08202600479125977 2023-01-21 10:11:11.062408: step: 2044/529, loss: 0.008945586159825325 2023-01-21 10:11:12.191002: step: 2048/529, loss: 0.026433609426021576 2023-01-21 10:11:13.340536: step: 2052/529, loss: 0.09628425538539886 2023-01-21 10:11:14.468743: step: 2056/529, loss: 0.11806364357471466 2023-01-21 10:11:15.589750: step: 2060/529, loss: 0.08045964688062668 2023-01-21 10:11:16.719130: step: 2064/529, loss: 0.04945630952715874 2023-01-21 10:11:17.831919: step: 2068/529, loss: 0.06084764003753662 2023-01-21 10:11:18.941761: step: 2072/529, loss: 0.2920656204223633 2023-01-21 10:11:20.069667: step: 2076/529, loss: 0.05531301349401474 2023-01-21 10:11:21.176200: step: 2080/529, loss: 0.31405580043792725 2023-01-21 10:11:22.329733: step: 2084/529, loss: 0.03344249725341797 2023-01-21 10:11:23.448896: step: 2088/529, loss: 0.1520463079214096 2023-01-21 10:11:24.550382: step: 2092/529, loss: 0.20689153671264648 2023-01-21 10:11:25.657168: step: 2096/529, loss: 0.21345052123069763 2023-01-21 10:11:26.765372: step: 2100/529, loss: 0.11501679569482803 2023-01-21 10:11:27.876376: step: 2104/529, loss: 0.029697764664888382 2023-01-21 10:11:29.004053: step: 2108/529, loss: 0.2440405786037445 2023-01-21 10:11:30.119753: step: 2112/529, loss: 0.10995569825172424 2023-01-21 10:11:31.241613: step: 2116/529, loss: 0.04227665811777115 ================================================== Loss: 0.124 -------------------- Dev: {'event': {'p': 0.5798816568047337, 'r': 0.7829560585885486, 'f1': 0.6662889518413597}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Test: {'event': {'p': 0.611233967271119, 'r': 0.7799097065462754, 'f1': 0.6853458963550707}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Chinese: {'event': {'p': 0.5581395348837209, 'r': 0.8888888888888888, 'f1': 0.6857142857142857}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Korean: {'event': {'p': 0.4788732394366197, 'r': 0.5396825396825397, 'f1': 0.5074626865671641}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Russian: {'event': {'p': 0.4418604651162791, 'r': 0.5277777777777778, 'f1': 0.4810126582278481}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** 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 10:12:12.144347: step: 4/529, loss: 0.06950245052576065 2023-01-21 10:12:13.238799: step: 8/529, loss: 0.04021148383617401 2023-01-21 10:12:14.357595: step: 12/529, loss: 0.14570626616477966 2023-01-21 10:12:15.460892: step: 16/529, loss: 0.0861416831612587 2023-01-21 10:12:16.580879: step: 20/529, loss: 0.1217346265912056 2023-01-21 10:12:17.667694: step: 24/529, loss: 0.02738165855407715 2023-01-21 10:12:18.776392: step: 28/529, loss: 0.012563848868012428 2023-01-21 10:12:19.881907: step: 32/529, loss: 0.021525001153349876 2023-01-21 10:12:21.001953: step: 36/529, loss: 0.025179320946335793 2023-01-21 10:12:22.085382: step: 40/529, loss: 0.016197968274354935 2023-01-21 10:12:23.192588: step: 44/529, loss: 0.028243256732821465 2023-01-21 10:12:24.323305: step: 48/529, loss: 0.06085710972547531 2023-01-21 10:12:25.443713: step: 52/529, loss: 0.009467840194702148 2023-01-21 10:12:26.545730: step: 56/529, loss: 0.03009519726037979 2023-01-21 10:12:27.653550: step: 60/529, loss: 0.1907138228416443 2023-01-21 10:12:28.774646: step: 64/529, loss: 0.1298326551914215 2023-01-21 10:12:29.868672: step: 68/529, loss: 0.04455399513244629 2023-01-21 10:12:31.004453: step: 72/529, loss: 0.027497030794620514 2023-01-21 10:12:32.121385: step: 76/529, loss: 0.0569947250187397 2023-01-21 10:12:33.213130: step: 80/529, loss: 0.044991157948970795 2023-01-21 10:12:34.312807: step: 84/529, loss: 0.10817378014326096 2023-01-21 10:12:35.409646: step: 88/529, loss: 0.017240095883607864 2023-01-21 10:12:36.542785: step: 92/529, loss: 0.01642007753252983 2023-01-21 10:12:37.664324: step: 96/529, loss: 0.019191836938261986 2023-01-21 10:12:38.778664: step: 100/529, loss: 0.0050757406279444695 2023-01-21 10:12:39.878393: step: 104/529, loss: 0.0720667839050293 2023-01-21 10:12:41.019840: step: 108/529, loss: 0.043734073638916016 2023-01-21 10:12:42.136013: step: 112/529, loss: 0.01588411256670952 2023-01-21 10:12:43.262251: step: 116/529, loss: 0.0852283462882042 2023-01-21 10:12:44.389675: step: 120/529, loss: 0.10721588134765625 2023-01-21 10:12:45.539069: step: 124/529, loss: 0.05493059381842613 2023-01-21 10:12:46.660062: step: 128/529, loss: 0.04765377193689346 2023-01-21 10:12:47.804720: step: 132/529, loss: 0.044267941266298294 2023-01-21 10:12:48.953583: step: 136/529, loss: 0.040538981556892395 2023-01-21 10:12:50.102766: step: 140/529, loss: 0.038683030754327774 2023-01-21 10:12:51.211818: step: 144/529, loss: 0.11157217621803284 2023-01-21 10:12:52.303333: step: 148/529, loss: 0.1508493423461914 2023-01-21 10:12:53.424746: step: 152/529, loss: 0.10179463028907776 2023-01-21 10:12:54.550319: step: 156/529, loss: 0.001695013022981584 2023-01-21 10:12:55.667041: step: 160/529, loss: 0.046556759625673294 2023-01-21 10:12:56.763737: step: 164/529, loss: 0.006415701471269131 2023-01-21 10:12:57.881262: step: 168/529, loss: 0.012030458077788353 2023-01-21 10:12:58.993052: step: 172/529, loss: 0.17725610733032227 2023-01-21 10:13:00.099469: step: 176/529, loss: 0.026179790496826172 2023-01-21 10:13:01.218571: step: 180/529, loss: 0.11263325810432434 2023-01-21 10:13:02.317198: step: 184/529, loss: 0.05074343830347061 2023-01-21 10:13:03.414765: step: 188/529, loss: 0.07608699798583984 2023-01-21 10:13:04.526398: step: 192/529, loss: 0.07712230831384659 2023-01-21 10:13:05.612479: step: 196/529, loss: 0.06077613681554794 2023-01-21 10:13:06.713969: step: 200/529, loss: 0.09429769217967987 2023-01-21 10:13:07.799473: step: 204/529, loss: 0.01578202284872532 2023-01-21 10:13:08.918522: step: 208/529, loss: 0.03738556057214737 2023-01-21 10:13:10.033669: step: 212/529, loss: 0.0626899003982544 2023-01-21 10:13:11.172390: step: 216/529, loss: 0.06241665035486221 2023-01-21 10:13:12.292206: step: 220/529, loss: 0.01506347581744194 2023-01-21 10:13:13.413672: step: 224/529, loss: 0.04915342107415199 2023-01-21 10:13:14.516427: step: 228/529, loss: 0.062230776995420456 2023-01-21 10:13:15.646041: step: 232/529, loss: 0.014750766567885876 2023-01-21 10:13:16.787052: step: 236/529, loss: 0.1026659607887268 2023-01-21 10:13:17.927111: step: 240/529, loss: 0.05019540712237358 2023-01-21 10:13:19.020911: step: 244/529, loss: 0.005922412965446711 2023-01-21 10:13:20.134574: step: 248/529, loss: 0.024925900623202324 2023-01-21 10:13:21.273595: step: 252/529, loss: 0.004024219699203968 2023-01-21 10:13:22.369526: step: 256/529, loss: 0.02569289319217205 2023-01-21 10:13:23.511845: step: 260/529, loss: 0.008908748626708984 2023-01-21 10:13:24.605175: step: 264/529, loss: 0.01979266107082367 2023-01-21 10:13:25.717588: step: 268/529, loss: 0.035695791244506836 2023-01-21 10:13:26.819278: step: 272/529, loss: 0.13730759918689728 2023-01-21 10:13:27.932682: step: 276/529, loss: 0.07338166236877441 2023-01-21 10:13:29.058735: step: 280/529, loss: 0.10899315029382706 2023-01-21 10:13:30.147654: step: 284/529, loss: 0.07925891876220703 2023-01-21 10:13:31.257406: step: 288/529, loss: 0.04474029690027237 2023-01-21 10:13:32.371045: step: 292/529, loss: 0.21635162830352783 2023-01-21 10:13:33.507365: step: 296/529, loss: 0.03347168117761612 2023-01-21 10:13:34.606591: step: 300/529, loss: 0.139997199177742 2023-01-21 10:13:35.714447: step: 304/529, loss: 0.013127708807587624 2023-01-21 10:13:36.871625: step: 308/529, loss: 0.0664484053850174 2023-01-21 10:13:37.974000: step: 312/529, loss: 0.04754181206226349 2023-01-21 10:13:39.079639: step: 316/529, loss: 0.02710428275167942 2023-01-21 10:13:40.181265: step: 320/529, loss: 0.08882904052734375 2023-01-21 10:13:41.295158: step: 324/529, loss: 0.06909863650798798 2023-01-21 10:13:42.401008: step: 328/529, loss: 0.05803317576646805 2023-01-21 10:13:43.541805: step: 332/529, loss: 0.24504026770591736 2023-01-21 10:13:44.638261: step: 336/529, loss: 0.08480215072631836 2023-01-21 10:13:45.765981: step: 340/529, loss: 0.10151185840368271 2023-01-21 10:13:46.869117: step: 344/529, loss: 0.06432318687438965 2023-01-21 10:13:47.990919: step: 348/529, loss: 0.15553614497184753 2023-01-21 10:13:49.110518: step: 352/529, loss: 0.10246536880731583 2023-01-21 10:13:50.237271: step: 356/529, loss: 0.034815311431884766 2023-01-21 10:13:51.341559: step: 360/529, loss: 0.07155991345643997 2023-01-21 10:13:52.473148: step: 364/529, loss: 0.10293684154748917 2023-01-21 10:13:53.573297: step: 368/529, loss: 0.04079904407262802 2023-01-21 10:13:54.709603: step: 372/529, loss: 0.12297402322292328 2023-01-21 10:13:55.860526: step: 376/529, loss: 0.0068219187669456005 2023-01-21 10:13:56.978429: step: 380/529, loss: 0.06119251251220703 2023-01-21 10:13:58.093001: step: 384/529, loss: 0.024859953671693802 2023-01-21 10:13:59.194202: step: 388/529, loss: 0.1250099539756775 2023-01-21 10:14:00.301396: step: 392/529, loss: 0.013640498742461205 2023-01-21 10:14:01.407068: step: 396/529, loss: 0.011499738320708275 2023-01-21 10:14:02.531971: step: 400/529, loss: 0.013403559103608131 2023-01-21 10:14:03.644407: step: 404/529, loss: 0.19005174934864044 2023-01-21 10:14:04.791080: step: 408/529, loss: 0.06413193047046661 2023-01-21 10:14:05.927506: step: 412/529, loss: 0.09859924018383026 2023-01-21 10:14:07.037000: step: 416/529, loss: 0.07810349762439728 2023-01-21 10:14:08.137526: step: 420/529, loss: 0.052854157984256744 2023-01-21 10:14:09.241867: step: 424/529, loss: 0.11092114448547363 2023-01-21 10:14:10.334293: step: 428/529, loss: 0.06058554723858833 2023-01-21 10:14:11.420682: step: 432/529, loss: 0.011052798479795456 2023-01-21 10:14:12.531205: step: 436/529, loss: 0.08526740968227386 2023-01-21 10:14:13.629715: step: 440/529, loss: 0.08960095047950745 2023-01-21 10:14:14.736873: step: 444/529, loss: 0.022571945562958717 2023-01-21 10:14:15.870283: step: 448/529, loss: 0.3154134154319763 2023-01-21 10:14:16.979433: step: 452/529, loss: 0.05916395038366318 2023-01-21 10:14:18.071748: step: 456/529, loss: 0.05604463070631027 2023-01-21 10:14:19.174776: step: 460/529, loss: 0.062203485518693924 2023-01-21 10:14:20.320046: step: 464/529, loss: 0.0969790518283844 2023-01-21 10:14:21.409099: step: 468/529, loss: 0.06417083740234375 2023-01-21 10:14:22.516923: step: 472/529, loss: 0.11251182109117508 2023-01-21 10:14:23.627335: step: 476/529, loss: 0.031088639050722122 2023-01-21 10:14:24.779187: step: 480/529, loss: 0.08974189311265945 2023-01-21 10:14:25.886891: step: 484/529, loss: 0.047586895525455475 2023-01-21 10:14:27.010850: step: 488/529, loss: 0.5900993943214417 2023-01-21 10:14:28.131682: step: 492/529, loss: 0.04580201953649521 2023-01-21 10:14:29.283052: step: 496/529, loss: 0.08017158508300781 2023-01-21 10:14:30.397452: step: 500/529, loss: 0.033750295639038086 2023-01-21 10:14:31.521987: step: 504/529, loss: 0.04029817506670952 2023-01-21 10:14:32.630764: step: 508/529, loss: 0.06806888431310654 2023-01-21 10:14:33.732168: step: 512/529, loss: 0.0344812385737896 2023-01-21 10:14:34.834783: step: 516/529, loss: 0.0668981596827507 2023-01-21 10:14:35.958368: step: 520/529, loss: 0.02187640778720379 2023-01-21 10:14:37.086630: step: 524/529, loss: 0.03815527260303497 2023-01-21 10:14:38.171529: step: 528/529, loss: 0.6970036625862122 2023-01-21 10:14:39.286444: step: 532/529, loss: 0.17776814103126526 2023-01-21 10:14:40.412464: step: 536/529, loss: 0.047922708094120026 2023-01-21 10:14:41.546704: step: 540/529, loss: 0.06548643112182617 2023-01-21 10:14:42.661887: step: 544/529, loss: 0.239887997508049 2023-01-21 10:14:43.776577: step: 548/529, loss: 0.0249770637601614 2023-01-21 10:14:44.867923: step: 552/529, loss: 0.025175761431455612 2023-01-21 10:14:45.964230: step: 556/529, loss: 0.04618034511804581 2023-01-21 10:14:47.079687: step: 560/529, loss: 0.0830078125 2023-01-21 10:14:48.175114: step: 564/529, loss: 0.11975231766700745 2023-01-21 10:14:49.280760: step: 568/529, loss: 0.6255862712860107 2023-01-21 10:14:50.418165: step: 572/529, loss: 0.0041533950716257095 2023-01-21 10:14:51.513032: step: 576/529, loss: 0.06521320343017578 2023-01-21 10:14:52.602838: step: 580/529, loss: 0.027253033593297005 2023-01-21 10:14:53.702443: step: 584/529, loss: 0.043028928339481354 2023-01-21 10:14:54.894098: step: 588/529, loss: 0.08211460709571838 2023-01-21 10:14:55.997320: step: 592/529, loss: 0.08129291236400604 2023-01-21 10:14:57.124320: step: 596/529, loss: 0.040040016174316406 2023-01-21 10:14:58.247529: step: 600/529, loss: 0.05761575698852539 2023-01-21 10:14:59.357539: step: 604/529, loss: 0.1538984328508377 2023-01-21 10:15:00.473527: step: 608/529, loss: 0.10057392716407776 2023-01-21 10:15:01.596035: step: 612/529, loss: 0.10902485251426697 2023-01-21 10:15:02.704840: step: 616/529, loss: 0.09169511497020721 2023-01-21 10:15:03.831230: step: 620/529, loss: 0.18789586424827576 2023-01-21 10:15:04.981338: step: 624/529, loss: 0.02539382129907608 2023-01-21 10:15:06.094162: step: 628/529, loss: 0.17691142857074738 2023-01-21 10:15:07.215977: step: 632/529, loss: 0.10114289075136185 2023-01-21 10:15:08.319032: step: 636/529, loss: 0.08607912063598633 2023-01-21 10:15:09.429311: step: 640/529, loss: 0.03666725382208824 2023-01-21 10:15:10.560048: step: 644/529, loss: 0.11375327408313751 2023-01-21 10:15:11.688146: step: 648/529, loss: 0.031490758061409 2023-01-21 10:15:12.810620: step: 652/529, loss: 0.0033008574973791838 2023-01-21 10:15:13.923295: step: 656/529, loss: 0.0477236770093441 2023-01-21 10:15:15.037135: step: 660/529, loss: 0.004269409459084272 2023-01-21 10:15:16.156465: step: 664/529, loss: 0.06674763560295105 2023-01-21 10:15:17.309573: step: 668/529, loss: 0.030095862224698067 2023-01-21 10:15:18.436695: step: 672/529, loss: 0.02253112755715847 2023-01-21 10:15:19.560846: step: 676/529, loss: 0.01704731024801731 2023-01-21 10:15:20.668967: step: 680/529, loss: 0.05232486501336098 2023-01-21 10:15:21.777237: step: 684/529, loss: 0.07303166389465332 2023-01-21 10:15:22.872631: step: 688/529, loss: 0.22654886543750763 2023-01-21 10:15:24.018139: step: 692/529, loss: 0.17165042459964752 2023-01-21 10:15:25.130116: step: 696/529, loss: 0.08732864260673523 2023-01-21 10:15:26.213076: step: 700/529, loss: 0.0269335750490427 2023-01-21 10:15:27.339516: step: 704/529, loss: 0.12308493256568909 2023-01-21 10:15:28.463139: step: 708/529, loss: 0.059477806091308594 2023-01-21 10:15:29.589408: step: 712/529, loss: 0.01904277689754963 2023-01-21 10:15:30.712534: step: 716/529, loss: 0.03123355098068714 2023-01-21 10:15:31.825465: step: 720/529, loss: 0.0851263552904129 2023-01-21 10:15:32.937547: step: 724/529, loss: 0.2921674847602844 2023-01-21 10:15:34.048164: step: 728/529, loss: 0.15794268250465393 2023-01-21 10:15:35.182764: step: 732/529, loss: 0.04390287399291992 2023-01-21 10:15:36.325249: step: 736/529, loss: 0.09830375015735626 2023-01-21 10:15:37.458293: step: 740/529, loss: 0.028116608038544655 2023-01-21 10:15:38.562788: step: 744/529, loss: 0.025487517938017845 2023-01-21 10:15:39.683818: step: 748/529, loss: 0.07403984665870667 2023-01-21 10:15:40.776651: step: 752/529, loss: 0.0343218669295311 2023-01-21 10:15:41.897910: step: 756/529, loss: 0.1437728852033615 2023-01-21 10:15:43.018264: step: 760/529, loss: 0.13192437589168549 2023-01-21 10:15:44.140448: step: 764/529, loss: 0.026557542383670807 2023-01-21 10:15:45.254022: step: 768/529, loss: 0.05518627539277077 2023-01-21 10:15:46.358098: step: 772/529, loss: 0.006531811319291592 2023-01-21 10:15:47.480156: step: 776/529, loss: 0.05170631781220436 2023-01-21 10:15:48.597093: step: 780/529, loss: 0.0362226739525795 2023-01-21 10:15:49.761606: step: 784/529, loss: 0.18278665840625763 2023-01-21 10:15:50.916988: step: 788/529, loss: 0.0196931604295969 2023-01-21 10:15:52.033332: step: 792/529, loss: 0.06992997974157333 2023-01-21 10:15:53.121760: step: 796/529, loss: 0.2194765955209732 2023-01-21 10:15:54.233999: step: 800/529, loss: 0.03704891353845596 2023-01-21 10:15:55.375721: step: 804/529, loss: 0.06582746654748917 2023-01-21 10:15:56.478455: step: 808/529, loss: 0.05394287407398224 2023-01-21 10:15:57.598696: step: 812/529, loss: 0.03265209123492241 2023-01-21 10:15:58.709705: step: 816/529, loss: 0.10381412506103516 2023-01-21 10:15:59.778659: step: 820/529, loss: 0.02874586544930935 2023-01-21 10:16:00.913930: step: 824/529, loss: 0.005927944090217352 2023-01-21 10:16:02.033270: step: 828/529, loss: 0.049616239964962006 2023-01-21 10:16:03.140480: step: 832/529, loss: 0.32866764068603516 2023-01-21 10:16:04.259262: step: 836/529, loss: 0.013270759955048561 2023-01-21 10:16:05.386307: step: 840/529, loss: 0.1976299285888672 2023-01-21 10:16:06.494373: step: 844/529, loss: 0.5636752843856812 2023-01-21 10:16:07.622261: step: 848/529, loss: 0.08105993270874023 2023-01-21 10:16:08.759997: step: 852/529, loss: 0.1500086784362793 2023-01-21 10:16:09.865479: step: 856/529, loss: 0.03187904506921768 2023-01-21 10:16:11.010162: step: 860/529, loss: 0.03752412647008896 2023-01-21 10:16:12.139077: step: 864/529, loss: 0.07712717354297638 2023-01-21 10:16:13.238587: step: 868/529, loss: 0.009344959631562233 2023-01-21 10:16:14.361395: step: 872/529, loss: 0.03660454973578453 2023-01-21 10:16:15.471926: step: 876/529, loss: 0.09308962523937225 2023-01-21 10:16:16.558837: step: 880/529, loss: 0.08583150804042816 2023-01-21 10:16:17.663226: step: 884/529, loss: 0.05918636545538902 2023-01-21 10:16:18.805215: step: 888/529, loss: 0.04658002778887749 2023-01-21 10:16:19.904973: step: 892/529, loss: 0.09445350617170334 2023-01-21 10:16:21.038751: step: 896/529, loss: 0.0929691344499588 2023-01-21 10:16:22.140927: step: 900/529, loss: 0.014909219928085804 2023-01-21 10:16:23.287083: step: 904/529, loss: 0.03253130987286568 2023-01-21 10:16:24.385478: step: 908/529, loss: 0.07860460877418518 2023-01-21 10:16:25.501830: step: 912/529, loss: 0.030065346509218216 2023-01-21 10:16:26.641677: step: 916/529, loss: 0.16117145121097565 2023-01-21 10:16:27.745376: step: 920/529, loss: 0.011951446533203125 2023-01-21 10:16:28.885421: step: 924/529, loss: 0.0917079895734787 2023-01-21 10:16:29.968536: step: 928/529, loss: 0.005024814512580633 2023-01-21 10:16:31.070196: step: 932/529, loss: 0.10253190994262695 2023-01-21 10:16:32.189648: step: 936/529, loss: 0.17319193482398987 2023-01-21 10:16:33.329371: step: 940/529, loss: 0.04013633728027344 2023-01-21 10:16:34.437407: step: 944/529, loss: 0.07327938079833984 2023-01-21 10:16:35.533721: step: 948/529, loss: 0.007530689239501953 2023-01-21 10:16:36.671382: step: 952/529, loss: 0.07090263813734055 2023-01-21 10:16:37.776814: step: 956/529, loss: 0.07901877909898758 2023-01-21 10:16:38.916726: step: 960/529, loss: 0.0318666473031044 2023-01-21 10:16:40.032060: step: 964/529, loss: 0.05167827755212784 2023-01-21 10:16:41.161913: step: 968/529, loss: 0.04520568996667862 2023-01-21 10:16:42.302905: step: 972/529, loss: 0.16680584847927094 2023-01-21 10:16:43.410535: step: 976/529, loss: 0.17241564393043518 2023-01-21 10:16:44.545631: step: 980/529, loss: 0.060108184814453125 2023-01-21 10:16:45.649501: step: 984/529, loss: 0.21506793797016144 2023-01-21 10:16:46.778843: step: 988/529, loss: 0.05670928955078125 2023-01-21 10:16:47.903545: step: 992/529, loss: 0.06589365005493164 2023-01-21 10:16:49.004064: step: 996/529, loss: 0.029415704309940338 2023-01-21 10:16:50.096781: step: 1000/529, loss: 0.06434173882007599 2023-01-21 10:16:51.252607: step: 1004/529, loss: 0.07568206638097763 2023-01-21 10:16:52.402882: step: 1008/529, loss: 0.028164483606815338 2023-01-21 10:16:53.555539: step: 1012/529, loss: 0.13938426971435547 2023-01-21 10:16:54.667715: step: 1016/529, loss: 0.02475166507065296 2023-01-21 10:16:55.817106: step: 1020/529, loss: 0.4329979121685028 2023-01-21 10:16:56.929033: step: 1024/529, loss: 0.08415546268224716 2023-01-21 10:16:58.048189: step: 1028/529, loss: 0.006848430726677179 2023-01-21 10:16:59.178633: step: 1032/529, loss: 0.002384758088737726 2023-01-21 10:17:00.295394: step: 1036/529, loss: 0.2246207296848297 2023-01-21 10:17:01.406512: step: 1040/529, loss: 0.19224748015403748 2023-01-21 10:17:02.521333: step: 1044/529, loss: 0.020032882690429688 2023-01-21 10:17:03.645242: step: 1048/529, loss: 0.055196452885866165 2023-01-21 10:17:04.758069: step: 1052/529, loss: 0.04472285509109497 2023-01-21 10:17:05.869713: step: 1056/529, loss: 0.1340726912021637 2023-01-21 10:17:06.986618: step: 1060/529, loss: 0.39980563521385193 2023-01-21 10:17:08.099540: step: 1064/529, loss: 0.026726534590125084 2023-01-21 10:17:09.241942: step: 1068/529, loss: 0.01442565955221653 2023-01-21 10:17:10.350635: step: 1072/529, loss: 0.0331222303211689 2023-01-21 10:17:11.463434: step: 1076/529, loss: 1.912477731704712 2023-01-21 10:17:12.589715: step: 1080/529, loss: 0.020315194502472878 2023-01-21 10:17:13.714798: step: 1084/529, loss: 0.1686621606349945 2023-01-21 10:17:14.843919: step: 1088/529, loss: 0.13259124755859375 2023-01-21 10:17:15.980174: step: 1092/529, loss: 0.034163523465394974 2023-01-21 10:17:17.115499: step: 1096/529, loss: 0.5313602685928345 2023-01-21 10:17:18.240279: step: 1100/529, loss: 0.05535001680254936 2023-01-21 10:17:19.355449: step: 1104/529, loss: 0.0796608030796051 2023-01-21 10:17:20.495599: step: 1108/529, loss: 0.017916584387421608 2023-01-21 10:17:21.621004: step: 1112/529, loss: 0.0645868331193924 2023-01-21 10:17:22.749263: step: 1116/529, loss: 0.07953600585460663 2023-01-21 10:17:23.894923: step: 1120/529, loss: 0.030167676508426666 2023-01-21 10:17:25.042401: step: 1124/529, loss: 0.06073722988367081 2023-01-21 10:17:26.143596: step: 1128/529, loss: 0.02266998402774334 2023-01-21 10:17:27.285022: step: 1132/529, loss: 0.09259691834449768 2023-01-21 10:17:28.392292: step: 1136/529, loss: 0.05960436165332794 2023-01-21 10:17:29.535545: step: 1140/529, loss: 0.06338825076818466 2023-01-21 10:17:30.645800: step: 1144/529, loss: 0.0370662696659565 2023-01-21 10:17:31.753838: step: 1148/529, loss: 0.07592706382274628 2023-01-21 10:17:32.896339: step: 1152/529, loss: 0.16204296052455902 2023-01-21 10:17:34.027344: step: 1156/529, loss: 0.044230081140995026 2023-01-21 10:17:35.125290: step: 1160/529, loss: 0.3214394450187683 2023-01-21 10:17:36.248723: step: 1164/529, loss: 0.10289172828197479 2023-01-21 10:17:37.403257: step: 1168/529, loss: 0.4607110023498535 2023-01-21 10:17:38.513046: step: 1172/529, loss: 0.007484054192900658 2023-01-21 10:17:39.643513: step: 1176/529, loss: 0.10458743572235107 2023-01-21 10:17:40.750071: step: 1180/529, loss: 0.07534690201282501 2023-01-21 10:17:41.861650: step: 1184/529, loss: 0.1800021231174469 2023-01-21 10:17:42.956920: step: 1188/529, loss: 0.0186602883040905 2023-01-21 10:17:44.099823: step: 1192/529, loss: 0.10658188164234161 2023-01-21 10:17:45.222512: step: 1196/529, loss: 0.13283920288085938 2023-01-21 10:17:46.335801: step: 1200/529, loss: 0.029497386887669563 2023-01-21 10:17:47.450356: step: 1204/529, loss: 0.13918933272361755 2023-01-21 10:17:48.573562: step: 1208/529, loss: 0.15209394693374634 2023-01-21 10:17:49.712020: step: 1212/529, loss: 0.0947432592511177 2023-01-21 10:17:50.854131: step: 1216/529, loss: 0.007777786813676357 2023-01-21 10:17:51.996998: step: 1220/529, loss: 0.045752912759780884 2023-01-21 10:17:53.116063: step: 1224/529, loss: 0.1949934959411621 2023-01-21 10:17:54.239610: step: 1228/529, loss: 0.42869171500205994 2023-01-21 10:17:55.328364: step: 1232/529, loss: 0.052896786481142044 2023-01-21 10:17:56.446855: step: 1236/529, loss: 0.044683218002319336 2023-01-21 10:17:57.591936: step: 1240/529, loss: 0.3001171052455902 2023-01-21 10:17:58.740395: step: 1244/529, loss: 0.025469493120908737 2023-01-21 10:17:59.829941: step: 1248/529, loss: 0.055413633584976196 2023-01-21 10:18:00.931561: step: 1252/529, loss: 0.002893017139285803 2023-01-21 10:18:02.043568: step: 1256/529, loss: 0.02019209787249565 2023-01-21 10:18:03.155065: step: 1260/529, loss: 0.5539092421531677 2023-01-21 10:18:04.290769: step: 1264/529, loss: 0.07252664864063263 2023-01-21 10:18:05.480701: step: 1268/529, loss: 0.3468475341796875 2023-01-21 10:18:06.568325: step: 1272/529, loss: 0.031056594103574753 2023-01-21 10:18:07.663475: step: 1276/529, loss: 0.015266155824065208 2023-01-21 10:18:08.777050: step: 1280/529, loss: 0.0981055274605751 2023-01-21 10:18:09.863916: step: 1284/529, loss: 0.0528842955827713 2023-01-21 10:18:10.965209: step: 1288/529, loss: 0.06403903663158417 2023-01-21 10:18:12.074021: step: 1292/529, loss: 0.3747352659702301 2023-01-21 10:18:13.202669: step: 1296/529, loss: 0.01789417304098606 2023-01-21 10:18:14.308720: step: 1300/529, loss: 0.06086044758558273 2023-01-21 10:18:15.403969: step: 1304/529, loss: 0.008039974607527256 2023-01-21 10:18:16.506161: step: 1308/529, loss: 0.8727419376373291 2023-01-21 10:18:17.609765: step: 1312/529, loss: 0.05934109911322594 2023-01-21 10:18:18.712919: step: 1316/529, loss: 0.03868570551276207 2023-01-21 10:18:19.815251: step: 1320/529, loss: 0.0005094528314657509 2023-01-21 10:18:20.944926: step: 1324/529, loss: 0.0983087569475174 2023-01-21 10:18:22.086943: step: 1328/529, loss: 0.04227724298834801 2023-01-21 10:18:23.202937: step: 1332/529, loss: 0.0652991309762001 2023-01-21 10:18:24.300935: step: 1336/529, loss: 0.0022724152076989412 2023-01-21 10:18:25.393575: step: 1340/529, loss: 0.19092664122581482 2023-01-21 10:18:26.557579: step: 1344/529, loss: 0.05441854149103165 2023-01-21 10:18:27.680512: step: 1348/529, loss: 0.030962657183408737 2023-01-21 10:18:28.786220: step: 1352/529, loss: 0.06297724694013596 2023-01-21 10:18:29.906604: step: 1356/529, loss: 0.0033655166625976562 2023-01-21 10:18:31.025676: step: 1360/529, loss: 0.02559366449713707 2023-01-21 10:18:32.146021: step: 1364/529, loss: 1.1154946088790894 2023-01-21 10:18:33.258870: step: 1368/529, loss: 0.01449041347950697 2023-01-21 10:18:34.402051: step: 1372/529, loss: 0.20906372368335724 2023-01-21 10:18:35.499985: step: 1376/529, loss: 0.0015120506286621094 2023-01-21 10:18:36.604334: step: 1380/529, loss: 0.15007105469703674 2023-01-21 10:18:37.709821: step: 1384/529, loss: 0.06564774364233017 2023-01-21 10:18:38.803278: step: 1388/529, loss: 0.13308311998844147 2023-01-21 10:18:39.895895: step: 1392/529, loss: 0.10358686745166779 2023-01-21 10:18:40.998359: step: 1396/529, loss: 0.03520488739013672 2023-01-21 10:18:42.144185: step: 1400/529, loss: 0.1521722823381424 2023-01-21 10:18:43.276294: step: 1404/529, loss: 0.08083543926477432 2023-01-21 10:18:44.407400: step: 1408/529, loss: 0.11886010318994522 2023-01-21 10:18:45.511533: step: 1412/529, loss: 0.0628652572631836 2023-01-21 10:18:46.639836: step: 1416/529, loss: 0.15924377739429474 2023-01-21 10:18:47.771851: step: 1420/529, loss: 0.17097461223602295 2023-01-21 10:18:48.880092: step: 1424/529, loss: 0.10538473725318909 2023-01-21 10:18:50.000672: step: 1428/529, loss: 0.045882657170295715 2023-01-21 10:18:51.163204: step: 1432/529, loss: 0.10997632145881653 2023-01-21 10:18:52.292635: step: 1436/529, loss: 0.14286451041698456 2023-01-21 10:18:53.413284: step: 1440/529, loss: 0.03170280531048775 2023-01-21 10:18:54.528438: step: 1444/529, loss: 0.11582193523645401 2023-01-21 10:18:55.664762: step: 1448/529, loss: 0.08530484139919281 2023-01-21 10:18:56.781252: step: 1452/529, loss: 0.035477638244628906 2023-01-21 10:18:57.890562: step: 1456/529, loss: 0.09922733157873154 2023-01-21 10:18:58.997745: step: 1460/529, loss: 0.05142870172858238 2023-01-21 10:19:00.123576: step: 1464/529, loss: 0.11757383495569229 2023-01-21 10:19:01.223877: step: 1468/529, loss: 0.059836193919181824 2023-01-21 10:19:02.350166: step: 1472/529, loss: 0.04173774644732475 2023-01-21 10:19:03.459793: step: 1476/529, loss: 0.06623878329992294 2023-01-21 10:19:04.546808: step: 1480/529, loss: 0.08290930092334747 2023-01-21 10:19:05.696247: step: 1484/529, loss: 0.06116848066449165 2023-01-21 10:19:06.794839: step: 1488/529, loss: 0.05420083925127983 2023-01-21 10:19:07.894484: step: 1492/529, loss: 0.07713265717029572 2023-01-21 10:19:09.007489: step: 1496/529, loss: 0.0663883239030838 2023-01-21 10:19:10.146375: step: 1500/529, loss: 0.09286212921142578 2023-01-21 10:19:11.256496: step: 1504/529, loss: 0.08248787373304367 2023-01-21 10:19:12.404303: step: 1508/529, loss: 0.006128883920609951 2023-01-21 10:19:13.553851: step: 1512/529, loss: 0.03740748390555382 2023-01-21 10:19:14.694698: step: 1516/529, loss: 0.04056992754340172 2023-01-21 10:19:15.836402: step: 1520/529, loss: 0.042075350880622864 2023-01-21 10:19:16.941360: step: 1524/529, loss: 0.09767808765172958 2023-01-21 10:19:18.068314: step: 1528/529, loss: 0.07884716987609863 2023-01-21 10:19:19.182526: step: 1532/529, loss: 0.07051058113574982 2023-01-21 10:19:20.340427: step: 1536/529, loss: 0.05629577860236168 2023-01-21 10:19:21.447487: step: 1540/529, loss: 0.06162138283252716 2023-01-21 10:19:22.562504: step: 1544/529, loss: 0.11712093651294708 2023-01-21 10:19:23.666185: step: 1548/529, loss: 0.05829601362347603 2023-01-21 10:19:24.772154: step: 1552/529, loss: 0.044644832611083984 2023-01-21 10:19:25.885093: step: 1556/529, loss: 0.0030996324494481087 2023-01-21 10:19:27.023063: step: 1560/529, loss: 0.04493984207510948 2023-01-21 10:19:28.159433: step: 1564/529, loss: 0.13707047700881958 2023-01-21 10:19:29.260168: step: 1568/529, loss: 0.060991764068603516 2023-01-21 10:19:30.370694: step: 1572/529, loss: 0.12425756454467773 2023-01-21 10:19:31.471272: step: 1576/529, loss: 0.06178493797779083 2023-01-21 10:19:32.625795: step: 1580/529, loss: 0.06205492466688156 2023-01-21 10:19:33.724573: step: 1584/529, loss: 0.11130929738283157 2023-01-21 10:19:34.860435: step: 1588/529, loss: 0.11500425636768341 2023-01-21 10:19:35.991411: step: 1592/529, loss: 0.16467972099781036 2023-01-21 10:19:37.126621: step: 1596/529, loss: 0.0663367286324501 2023-01-21 10:19:38.249534: step: 1600/529, loss: 0.07432594895362854 2023-01-21 10:19:39.398440: step: 1604/529, loss: 0.006584644317626953 2023-01-21 10:19:40.526299: step: 1608/529, loss: 0.0016658783424645662 2023-01-21 10:19:41.653264: step: 1612/529, loss: 0.3817630708217621 2023-01-21 10:19:42.749890: step: 1616/529, loss: 0.03890247642993927 2023-01-21 10:19:43.851644: step: 1620/529, loss: 0.08566303551197052 2023-01-21 10:19:44.966817: step: 1624/529, loss: 0.41197946667671204 2023-01-21 10:19:46.092098: step: 1628/529, loss: 0.1474742889404297 2023-01-21 10:19:47.226191: step: 1632/529, loss: 0.018197059631347656 2023-01-21 10:19:48.365034: step: 1636/529, loss: 0.11582642048597336 2023-01-21 10:19:49.468422: step: 1640/529, loss: 0.01915464550256729 2023-01-21 10:19:50.626674: step: 1644/529, loss: 0.01825237274169922 2023-01-21 10:19:51.732467: step: 1648/529, loss: 0.07225757092237473 2023-01-21 10:19:52.844030: step: 1652/529, loss: 0.10269680619239807 2023-01-21 10:19:53.961379: step: 1656/529, loss: 0.06576109677553177 2023-01-21 10:19:55.091537: step: 1660/529, loss: 0.15911951661109924 2023-01-21 10:19:56.250935: step: 1664/529, loss: 0.0031041146721690893 2023-01-21 10:19:57.392694: step: 1668/529, loss: 0.13001003861427307 2023-01-21 10:19:58.502799: step: 1672/529, loss: 0.02845921367406845 2023-01-21 10:19:59.617946: step: 1676/529, loss: 0.015304755419492722 2023-01-21 10:20:00.760359: step: 1680/529, loss: 0.17093610763549805 2023-01-21 10:20:01.875976: step: 1684/529, loss: 0.36709803342819214 2023-01-21 10:20:03.008204: step: 1688/529, loss: 0.00982294138520956 2023-01-21 10:20:04.160056: step: 1692/529, loss: 0.08794493973255157 2023-01-21 10:20:05.275727: step: 1696/529, loss: 0.043036460876464844 2023-01-21 10:20:06.385814: step: 1700/529, loss: 0.0265686996281147 2023-01-21 10:20:07.499008: step: 1704/529, loss: 0.296053409576416 2023-01-21 10:20:08.619072: step: 1708/529, loss: 0.037857912480831146 2023-01-21 10:20:09.732154: step: 1712/529, loss: 0.012712288647890091 2023-01-21 10:20:10.848469: step: 1716/529, loss: 0.06006031110882759 2023-01-21 10:20:11.959460: step: 1720/529, loss: 0.15958309173583984 2023-01-21 10:20:13.038144: step: 1724/529, loss: 0.03719472885131836 2023-01-21 10:20:14.159174: step: 1728/529, loss: 0.11248188465833664 2023-01-21 10:20:15.305526: step: 1732/529, loss: 0.04556594043970108 2023-01-21 10:20:16.404743: step: 1736/529, loss: 0.16809554398059845 2023-01-21 10:20:17.486815: step: 1740/529, loss: 0.11940333247184753 2023-01-21 10:20:18.606921: step: 1744/529, loss: 0.008860588073730469 2023-01-21 10:20:19.717056: step: 1748/529, loss: 0.005829811096191406 2023-01-21 10:20:20.834613: step: 1752/529, loss: 0.010744738392531872 2023-01-21 10:20:21.970188: step: 1756/529, loss: 0.15486498177051544 2023-01-21 10:20:23.074934: step: 1760/529, loss: 0.044983863830566406 2023-01-21 10:20:24.187868: step: 1764/529, loss: 0.12944364547729492 2023-01-21 10:20:25.309885: step: 1768/529, loss: 0.010328715667128563 2023-01-21 10:20:26.430151: step: 1772/529, loss: 0.08861346542835236 2023-01-21 10:20:27.561842: step: 1776/529, loss: 0.10198240727186203 2023-01-21 10:20:28.687459: step: 1780/529, loss: 0.10300026088953018 2023-01-21 10:20:29.829518: step: 1784/529, loss: 0.018926048651337624 2023-01-21 10:20:30.950723: step: 1788/529, loss: 0.0031180144287645817 2023-01-21 10:20:32.057250: step: 1792/529, loss: 0.09198741614818573 2023-01-21 10:20:33.168147: step: 1796/529, loss: 0.500396728515625 2023-01-21 10:20:34.286101: step: 1800/529, loss: 0.015450287610292435 2023-01-21 10:20:35.406247: step: 1804/529, loss: 0.12958678603172302 2023-01-21 10:20:36.504923: step: 1808/529, loss: 0.06736544519662857 2023-01-21 10:20:37.641543: step: 1812/529, loss: 0.06629429012537003 2023-01-21 10:20:38.737484: step: 1816/529, loss: 0.13922032713890076 2023-01-21 10:20:39.885559: step: 1820/529, loss: 0.2159046232700348 2023-01-21 10:20:41.005693: step: 1824/529, loss: 0.06676407158374786 2023-01-21 10:20:42.098468: step: 1828/529, loss: 0.5868772864341736 2023-01-21 10:20:43.194971: step: 1832/529, loss: 0.02198047749698162 2023-01-21 10:20:44.308482: step: 1836/529, loss: 0.11488499492406845 2023-01-21 10:20:45.417756: step: 1840/529, loss: 0.06417850404977798 2023-01-21 10:20:46.540428: step: 1844/529, loss: 0.017812252044677734 2023-01-21 10:20:47.641363: step: 1848/529, loss: 0.19078870117664337 2023-01-21 10:20:48.781073: step: 1852/529, loss: 0.20754380524158478 2023-01-21 10:20:49.868571: step: 1856/529, loss: 0.03929634019732475 2023-01-21 10:20:50.990299: step: 1860/529, loss: 0.022656727582216263 2023-01-21 10:20:52.127983: step: 1864/529, loss: 0.06775032728910446 2023-01-21 10:20:53.236742: step: 1868/529, loss: 0.05289635807275772 2023-01-21 10:20:54.351864: step: 1872/529, loss: 0.06041469797492027 2023-01-21 10:20:55.464790: step: 1876/529, loss: 0.05957479774951935 2023-01-21 10:20:56.560564: step: 1880/529, loss: 0.0886511355638504 2023-01-21 10:20:57.669825: step: 1884/529, loss: 0.184962198138237 2023-01-21 10:20:58.778238: step: 1888/529, loss: 0.07012825459241867 2023-01-21 10:20:59.924085: step: 1892/529, loss: 0.0929986983537674 2023-01-21 10:21:01.042357: step: 1896/529, loss: 0.04324252903461456 2023-01-21 10:21:02.153501: step: 1900/529, loss: 0.15844345092773438 2023-01-21 10:21:03.262606: step: 1904/529, loss: 0.10504860430955887 2023-01-21 10:21:04.364184: step: 1908/529, loss: 0.03284740447998047 2023-01-21 10:21:05.499818: step: 1912/529, loss: 0.14668092131614685 2023-01-21 10:21:06.621409: step: 1916/529, loss: 0.010412168689072132 2023-01-21 10:21:07.768754: step: 1920/529, loss: 0.0695522204041481 2023-01-21 10:21:08.892242: step: 1924/529, loss: 0.12053041905164719 2023-01-21 10:21:10.012753: step: 1928/529, loss: 0.04767570272088051 2023-01-21 10:21:11.117606: step: 1932/529, loss: 0.044094182550907135 2023-01-21 10:21:12.246965: step: 1936/529, loss: 0.31926095485687256 2023-01-21 10:21:13.371780: step: 1940/529, loss: 0.05498848110437393 2023-01-21 10:21:14.482899: step: 1944/529, loss: 0.02191934548318386 2023-01-21 10:21:15.584687: step: 1948/529, loss: 1.19880211353302 2023-01-21 10:21:16.708239: step: 1952/529, loss: 0.02490520477294922 2023-01-21 10:21:17.841362: step: 1956/529, loss: 0.04433279111981392 2023-01-21 10:21:18.970006: step: 1960/529, loss: 0.45662468671798706 2023-01-21 10:21:20.065598: step: 1964/529, loss: 0.04059944301843643 2023-01-21 10:21:21.216662: step: 1968/529, loss: 0.027599716559052467 2023-01-21 10:21:22.338974: step: 1972/529, loss: 0.04951610416173935 2023-01-21 10:21:23.438644: step: 1976/529, loss: 0.13329724967479706 2023-01-21 10:21:24.545085: step: 1980/529, loss: 0.02415151707828045 2023-01-21 10:21:25.669293: step: 1984/529, loss: 0.03460407257080078 2023-01-21 10:21:26.794291: step: 1988/529, loss: 0.09172935783863068 2023-01-21 10:21:27.899549: step: 1992/529, loss: 0.0057312012650072575 2023-01-21 10:21:29.029027: step: 1996/529, loss: 0.10153666138648987 2023-01-21 10:21:30.138454: step: 2000/529, loss: 0.04121112823486328 2023-01-21 10:21:31.254968: step: 2004/529, loss: 0.01641373708844185 2023-01-21 10:21:32.376308: step: 2008/529, loss: 0.027023935690522194 2023-01-21 10:21:33.493849: step: 2012/529, loss: 0.0180834773927927 2023-01-21 10:21:34.567703: step: 2016/529, loss: 0.05376157909631729 2023-01-21 10:21:35.730139: step: 2020/529, loss: 0.3516879081726074 2023-01-21 10:21:36.827223: step: 2024/529, loss: 0.11295566707849503 2023-01-21 10:21:37.938344: step: 2028/529, loss: 0.07551927864551544 2023-01-21 10:21:39.071964: step: 2032/529, loss: 0.586672306060791 2023-01-21 10:21:40.192480: step: 2036/529, loss: 0.042920973151922226 2023-01-21 10:21:41.326218: step: 2040/529, loss: 0.03974013403058052 2023-01-21 10:21:42.430006: step: 2044/529, loss: 0.0026085854042321444 2023-01-21 10:21:43.565214: step: 2048/529, loss: 0.05784473568201065 2023-01-21 10:21:44.669104: step: 2052/529, loss: 0.04035158455371857 2023-01-21 10:21:45.787318: step: 2056/529, loss: 0.16087280213832855 2023-01-21 10:21:46.915346: step: 2060/529, loss: 0.05341053009033203 2023-01-21 10:21:48.011760: step: 2064/529, loss: 0.024414371699094772 2023-01-21 10:21:49.115262: step: 2068/529, loss: 0.09351043403148651 2023-01-21 10:21:50.239835: step: 2072/529, loss: 0.017180848866701126 2023-01-21 10:21:51.372449: step: 2076/529, loss: 0.06118660047650337 2023-01-21 10:21:52.530026: step: 2080/529, loss: 0.03235488012433052 2023-01-21 10:21:53.647339: step: 2084/529, loss: 0.1415139138698578 2023-01-21 10:21:54.758358: step: 2088/529, loss: 0.010071372613310814 2023-01-21 10:21:55.900308: step: 2092/529, loss: 0.13087940216064453 2023-01-21 10:21:57.005855: step: 2096/529, loss: 0.0761045515537262 2023-01-21 10:21:58.124184: step: 2100/529, loss: 0.09669361263513565 2023-01-21 10:21:59.251012: step: 2104/529, loss: 0.023715972900390625 2023-01-21 10:22:00.350070: step: 2108/529, loss: 0.03133287653326988 2023-01-21 10:22:01.480687: step: 2112/529, loss: 0.053655434399843216 2023-01-21 10:22:02.609784: step: 2116/529, loss: 0.051026277244091034 ================================================== Loss: 0.097 -------------------- Dev: {'event': {'p': 0.6004119464469619, 'r': 0.7762982689747004, 'f1': 0.6771196283391405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Test: {'event': {'p': 0.6153512575888985, 'r': 0.8007900677200903, 'f1': 0.6959293771456597}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Chinese: {'event': {'p': 0.5454545454545454, 'r': 0.8888888888888888, 'f1': 0.676056338028169}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Korean: {'event': {'p': 0.5454545454545454, 'r': 0.5714285714285714, 'f1': 0.5581395348837208}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Russian: {'event': {'p': 0.46153846153846156, 'r': 0.5, 'f1': 0.48000000000000004}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 11 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:22:43.634416: step: 4/529, loss: 0.044892121106386185 2023-01-21 10:22:44.734260: step: 8/529, loss: 0.0625268965959549 2023-01-21 10:22:45.834637: step: 12/529, loss: 0.03142843395471573 2023-01-21 10:22:46.963521: step: 16/529, loss: 0.18187494575977325 2023-01-21 10:22:48.070046: step: 20/529, loss: 0.11568813025951385 2023-01-21 10:22:49.188352: step: 24/529, loss: 0.04502449184656143 2023-01-21 10:22:50.301380: step: 28/529, loss: 0.0923745185136795 2023-01-21 10:22:51.402897: step: 32/529, loss: 0.03228185325860977 2023-01-21 10:22:52.504897: step: 36/529, loss: 0.0037368773482739925 2023-01-21 10:22:53.656179: step: 40/529, loss: 0.0487639456987381 2023-01-21 10:22:54.785514: step: 44/529, loss: 0.01682748831808567 2023-01-21 10:22:55.882120: step: 48/529, loss: 0.01993861235678196 2023-01-21 10:22:57.032758: step: 52/529, loss: 0.10721054673194885 2023-01-21 10:22:58.157525: step: 56/529, loss: 0.03785209730267525 2023-01-21 10:22:59.269662: step: 60/529, loss: 0.04608885943889618 2023-01-21 10:23:00.381616: step: 64/529, loss: 0.0338287353515625 2023-01-21 10:23:01.489170: step: 68/529, loss: 0.012132120318710804 2023-01-21 10:23:02.573841: step: 72/529, loss: 0.004388713743537664 2023-01-21 10:23:03.721339: step: 76/529, loss: 0.0578186996281147 2023-01-21 10:23:04.827219: step: 80/529, loss: 0.03146343305706978 2023-01-21 10:23:05.924433: step: 84/529, loss: 0.007626248057931662 2023-01-21 10:23:07.033536: step: 88/529, loss: 0.034311581403017044 2023-01-21 10:23:08.160552: step: 92/529, loss: 0.1295042484998703 2023-01-21 10:23:09.279978: step: 96/529, loss: 0.017191410064697266 2023-01-21 10:23:10.423619: step: 100/529, loss: 0.05328216403722763 2023-01-21 10:23:11.554553: step: 104/529, loss: 0.2050791233778 2023-01-21 10:23:12.708804: step: 108/529, loss: 0.015751982107758522 2023-01-21 10:23:13.833925: step: 112/529, loss: 0.18110105395317078 2023-01-21 10:23:14.950429: step: 116/529, loss: 0.08174113929271698 2023-01-21 10:23:16.074907: step: 120/529, loss: 0.12513476610183716 2023-01-21 10:23:17.196179: step: 124/529, loss: 0.022868728265166283 2023-01-21 10:23:18.331473: step: 128/529, loss: 0.052204132080078125 2023-01-21 10:23:19.419441: step: 132/529, loss: 0.008044051937758923 2023-01-21 10:23:20.542947: step: 136/529, loss: 0.024648284539580345 2023-01-21 10:23:21.666135: step: 140/529, loss: 0.01880931854248047 2023-01-21 10:23:22.804939: step: 144/529, loss: 0.5584419965744019 2023-01-21 10:23:23.927775: step: 148/529, loss: 0.04732847213745117 2023-01-21 10:23:25.036699: step: 152/529, loss: 0.02764282189309597 2023-01-21 10:23:26.151757: step: 156/529, loss: 0.05077533796429634 2023-01-21 10:23:27.290655: step: 160/529, loss: 0.025649739429354668 2023-01-21 10:23:28.411382: step: 164/529, loss: 0.09128904342651367 2023-01-21 10:23:29.524279: step: 168/529, loss: 0.020452309399843216 2023-01-21 10:23:30.636563: step: 172/529, loss: 0.0022405623458325863 2023-01-21 10:23:31.763255: step: 176/529, loss: 0.003482246305793524 2023-01-21 10:23:32.868296: step: 180/529, loss: 0.026781463995575905 2023-01-21 10:23:33.990831: step: 184/529, loss: 0.017145443707704544 2023-01-21 10:23:35.105490: step: 188/529, loss: 0.050605013966560364 2023-01-21 10:23:36.233359: step: 192/529, loss: 0.056218527257442474 2023-01-21 10:23:37.344110: step: 196/529, loss: 0.08923760056495667 2023-01-21 10:23:38.496963: step: 200/529, loss: 0.22632159292697906 2023-01-21 10:23:39.600690: step: 204/529, loss: 0.013107681646943092 2023-01-21 10:23:40.711688: step: 208/529, loss: 0.030262090265750885 2023-01-21 10:23:41.835959: step: 212/529, loss: 0.010067367926239967 2023-01-21 10:23:42.953612: step: 216/529, loss: 0.040558528155088425 2023-01-21 10:23:44.068746: step: 220/529, loss: 0.09151677787303925 2023-01-21 10:23:45.177382: step: 224/529, loss: 0.016289377585053444 2023-01-21 10:23:46.303712: step: 228/529, loss: 0.07078304141759872 2023-01-21 10:23:47.412152: step: 232/529, loss: 0.032830189913511276 2023-01-21 10:23:48.512487: step: 236/529, loss: 0.1679474413394928 2023-01-21 10:23:49.634796: step: 240/529, loss: 0.01542511023581028 2023-01-21 10:23:50.758621: step: 244/529, loss: 0.05822563171386719 2023-01-21 10:23:51.853954: step: 248/529, loss: 0.0029464722611010075 2023-01-21 10:23:52.996894: step: 252/529, loss: 0.07984809577465057 2023-01-21 10:23:54.134920: step: 256/529, loss: 0.04578285291790962 2023-01-21 10:23:55.245654: step: 260/529, loss: 0.08597288280725479 2023-01-21 10:23:56.382678: step: 264/529, loss: 0.014625740237534046 2023-01-21 10:23:57.515496: step: 268/529, loss: 0.08937425911426544 2023-01-21 10:23:58.635771: step: 272/529, loss: 0.046263791620731354 2023-01-21 10:23:59.738410: step: 276/529, loss: 0.005494880955666304 2023-01-21 10:24:00.827773: step: 280/529, loss: 0.030833005905151367 2023-01-21 10:24:01.945639: step: 284/529, loss: 0.1821443736553192 2023-01-21 10:24:03.062475: step: 288/529, loss: 0.07623839378356934 2023-01-21 10:24:04.183986: step: 292/529, loss: 0.09481838345527649 2023-01-21 10:24:05.283594: step: 296/529, loss: 0.013539028353989124 2023-01-21 10:24:06.419773: step: 300/529, loss: 0.028644908219575882 2023-01-21 10:24:07.512732: step: 304/529, loss: 0.014699173159897327 2023-01-21 10:24:08.616342: step: 308/529, loss: 0.021160125732421875 2023-01-21 10:24:09.724964: step: 312/529, loss: 0.4187335968017578 2023-01-21 10:24:10.845754: step: 316/529, loss: 0.02988738939166069 2023-01-21 10:24:11.968035: step: 320/529, loss: 0.02997131273150444 2023-01-21 10:24:13.087158: step: 324/529, loss: 0.1314006745815277 2023-01-21 10:24:14.216906: step: 328/529, loss: 0.020139073953032494 2023-01-21 10:24:15.350161: step: 332/529, loss: 0.3240695595741272 2023-01-21 10:24:16.471477: step: 336/529, loss: 0.015273381024599075 2023-01-21 10:24:17.592095: step: 340/529, loss: 0.22114935517311096 2023-01-21 10:24:18.714991: step: 344/529, loss: 0.0349545031785965 2023-01-21 10:24:19.830780: step: 348/529, loss: 0.019772624596953392 2023-01-21 10:24:20.921231: step: 352/529, loss: 0.09770698845386505 2023-01-21 10:24:22.016375: step: 356/529, loss: 0.11316695809364319 2023-01-21 10:24:23.113414: step: 360/529, loss: 0.05522461235523224 2023-01-21 10:24:24.234579: step: 364/529, loss: 0.044603198766708374 2023-01-21 10:24:25.326930: step: 368/529, loss: 0.020693683996796608 2023-01-21 10:24:26.505282: step: 372/529, loss: 0.03904552757740021 2023-01-21 10:24:27.623837: step: 376/529, loss: 0.03238171339035034 2023-01-21 10:24:28.731054: step: 380/529, loss: 0.07206100970506668 2023-01-21 10:24:29.851614: step: 384/529, loss: 0.03099827840924263 2023-01-21 10:24:30.978536: step: 388/529, loss: 0.1448509693145752 2023-01-21 10:24:32.083687: step: 392/529, loss: 0.0327390693128109 2023-01-21 10:24:33.188964: step: 396/529, loss: 0.0008177757263183594 2023-01-21 10:24:34.313869: step: 400/529, loss: 0.003954553510993719 2023-01-21 10:24:35.432185: step: 404/529, loss: 0.1691402643918991 2023-01-21 10:24:36.542224: step: 408/529, loss: 0.014127827249467373 2023-01-21 10:24:37.652575: step: 412/529, loss: 0.03034992329776287 2023-01-21 10:24:38.803709: step: 416/529, loss: 0.04700450971722603 2023-01-21 10:24:39.923436: step: 420/529, loss: 0.04196205362677574 2023-01-21 10:24:41.062686: step: 424/529, loss: 0.07173824310302734 2023-01-21 10:24:42.195671: step: 428/529, loss: 0.051961325109004974 2023-01-21 10:24:43.316875: step: 432/529, loss: 0.032865334302186966 2023-01-21 10:24:44.477746: step: 436/529, loss: 0.09171953052282333 2023-01-21 10:24:45.601938: step: 440/529, loss: 0.09884472191333771 2023-01-21 10:24:46.762487: step: 444/529, loss: 0.062410928308963776 2023-01-21 10:24:47.866399: step: 448/529, loss: 0.009027529507875443 2023-01-21 10:24:48.972504: step: 452/529, loss: 0.056571390479803085 2023-01-21 10:24:50.085423: step: 456/529, loss: 0.023335743695497513 2023-01-21 10:24:51.227021: step: 460/529, loss: 0.03396587446331978 2023-01-21 10:24:52.353107: step: 464/529, loss: 0.022298144176602364 2023-01-21 10:24:53.460364: step: 468/529, loss: 0.12664107978343964 2023-01-21 10:24:54.587975: step: 472/529, loss: 0.047626398503780365 2023-01-21 10:24:55.697224: step: 476/529, loss: 0.025447560474276543 2023-01-21 10:24:56.810498: step: 480/529, loss: 0.10914306342601776 2023-01-21 10:24:57.920388: step: 484/529, loss: 0.05278158187866211 2023-01-21 10:24:59.032119: step: 488/529, loss: 0.0972658172249794 2023-01-21 10:25:00.124814: step: 492/529, loss: 0.031143857166171074 2023-01-21 10:25:01.195545: step: 496/529, loss: 0.10382290184497833 2023-01-21 10:25:02.305416: step: 500/529, loss: 0.004127121064811945 2023-01-21 10:25:03.430720: step: 504/529, loss: 0.04012127220630646 2023-01-21 10:25:04.543539: step: 508/529, loss: 0.03874502331018448 2023-01-21 10:25:05.716000: step: 512/529, loss: 0.03890252113342285 2023-01-21 10:25:06.821121: step: 516/529, loss: 0.09594850242137909 2023-01-21 10:25:07.937579: step: 520/529, loss: 0.049819089472293854 2023-01-21 10:25:09.111056: step: 524/529, loss: 0.01969614066183567 2023-01-21 10:25:10.215347: step: 528/529, loss: 0.027335453778505325 2023-01-21 10:25:11.328128: step: 532/529, loss: 0.021639443933963776 2023-01-21 10:25:12.440960: step: 536/529, loss: 0.050621893256902695 2023-01-21 10:25:13.577692: step: 540/529, loss: 0.050080299377441406 2023-01-21 10:25:14.702716: step: 544/529, loss: 0.0535762794315815 2023-01-21 10:25:15.830188: step: 548/529, loss: 0.11981220543384552 2023-01-21 10:25:16.980727: step: 552/529, loss: 0.09423381090164185 2023-01-21 10:25:18.101922: step: 556/529, loss: 0.0670996680855751 2023-01-21 10:25:19.210694: step: 560/529, loss: 0.10288429260253906 2023-01-21 10:25:20.326586: step: 564/529, loss: 0.06350155174732208 2023-01-21 10:25:21.444530: step: 568/529, loss: 0.10281587392091751 2023-01-21 10:25:22.547724: step: 572/529, loss: 0.12779581546783447 2023-01-21 10:25:23.679892: step: 576/529, loss: 0.08085943758487701 2023-01-21 10:25:24.800415: step: 580/529, loss: 0.09658604115247726 2023-01-21 10:25:25.909792: step: 584/529, loss: 0.008464241400361061 2023-01-21 10:25:27.014583: step: 588/529, loss: 0.21583539247512817 2023-01-21 10:25:28.150995: step: 592/529, loss: 0.039789870381355286 2023-01-21 10:25:29.285828: step: 596/529, loss: 0.07090745121240616 2023-01-21 10:25:30.403542: step: 600/529, loss: 0.09214477241039276 2023-01-21 10:25:31.524542: step: 604/529, loss: 0.005883693695068359 2023-01-21 10:25:32.666208: step: 608/529, loss: 0.03277955204248428 2023-01-21 10:25:33.796449: step: 612/529, loss: 0.04826831817626953 2023-01-21 10:25:34.942477: step: 616/529, loss: 0.028658676892518997 2023-01-21 10:25:36.057170: step: 620/529, loss: 0.030988717451691628 2023-01-21 10:25:37.163660: step: 624/529, loss: 0.011056710034608841 2023-01-21 10:25:38.292486: step: 628/529, loss: 0.014819574542343616 2023-01-21 10:25:39.438736: step: 632/529, loss: 0.047968581318855286 2023-01-21 10:25:40.562047: step: 636/529, loss: 0.01238107681274414 2023-01-21 10:25:41.688792: step: 640/529, loss: 0.06460742652416229 2023-01-21 10:25:42.822893: step: 644/529, loss: 0.0994729995727539 2023-01-21 10:25:43.935162: step: 648/529, loss: 0.04190712049603462 2023-01-21 10:25:45.067307: step: 652/529, loss: 0.02900867536664009 2023-01-21 10:25:46.174666: step: 656/529, loss: 0.07848282158374786 2023-01-21 10:25:47.308062: step: 660/529, loss: 0.009172726422548294 2023-01-21 10:25:48.416481: step: 664/529, loss: 0.04128170385956764 2023-01-21 10:25:49.573460: step: 668/529, loss: 0.019165707752108574 2023-01-21 10:25:50.693872: step: 672/529, loss: 0.11756525933742523 2023-01-21 10:25:51.816570: step: 676/529, loss: 0.06627778708934784 2023-01-21 10:25:52.923707: step: 680/529, loss: 0.24567127227783203 2023-01-21 10:25:54.029361: step: 684/529, loss: 0.05248670652508736 2023-01-21 10:25:55.151098: step: 688/529, loss: 0.07553596794605255 2023-01-21 10:25:56.272611: step: 692/529, loss: 0.08473311364650726 2023-01-21 10:25:57.360652: step: 696/529, loss: 0.04260959476232529 2023-01-21 10:25:58.446161: step: 700/529, loss: 0.07594089210033417 2023-01-21 10:25:59.549757: step: 704/529, loss: 0.01995544508099556 2023-01-21 10:26:00.680313: step: 708/529, loss: 0.11991729587316513 2023-01-21 10:26:01.774228: step: 712/529, loss: 0.03400077670812607 2023-01-21 10:26:02.895463: step: 716/529, loss: 0.03663177415728569 2023-01-21 10:26:04.011119: step: 720/529, loss: 0.09200859069824219 2023-01-21 10:26:05.130907: step: 724/529, loss: 0.10756722092628479 2023-01-21 10:26:06.237229: step: 728/529, loss: 0.06871548295021057 2023-01-21 10:26:07.345721: step: 732/529, loss: 0.16469335556030273 2023-01-21 10:26:08.441651: step: 736/529, loss: 0.08558855205774307 2023-01-21 10:26:09.581506: step: 740/529, loss: 0.09734473377466202 2023-01-21 10:26:10.698985: step: 744/529, loss: 0.0015523910988122225 2023-01-21 10:26:11.828182: step: 748/529, loss: 0.1339588165283203 2023-01-21 10:26:12.929635: step: 752/529, loss: 0.01891326904296875 2023-01-21 10:26:14.046006: step: 756/529, loss: 0.22935573756694794 2023-01-21 10:26:15.153029: step: 760/529, loss: 0.03181047737598419 2023-01-21 10:26:16.278411: step: 764/529, loss: 0.06916847825050354 2023-01-21 10:26:17.429366: step: 768/529, loss: 0.15863971412181854 2023-01-21 10:26:18.569515: step: 772/529, loss: 0.019536782056093216 2023-01-21 10:26:19.685656: step: 776/529, loss: 0.04048867151141167 2023-01-21 10:26:20.793267: step: 780/529, loss: 0.017746353521943092 2023-01-21 10:26:21.913351: step: 784/529, loss: 0.0579533576965332 2023-01-21 10:26:23.018528: step: 788/529, loss: 0.006070256233215332 2023-01-21 10:26:24.143938: step: 792/529, loss: 0.003806042717769742 2023-01-21 10:26:25.229615: step: 796/529, loss: 0.044641975313425064 2023-01-21 10:26:26.341297: step: 800/529, loss: 0.062311507761478424 2023-01-21 10:26:27.440611: step: 804/529, loss: 0.004421234130859375 2023-01-21 10:26:28.539288: step: 808/529, loss: 0.10612402111291885 2023-01-21 10:26:29.664990: step: 812/529, loss: 0.010794401168823242 2023-01-21 10:26:30.751175: step: 816/529, loss: 0.32740622758865356 2023-01-21 10:26:31.857005: step: 820/529, loss: 0.015049362555146217 2023-01-21 10:26:32.990217: step: 824/529, loss: 0.060502246022224426 2023-01-21 10:26:34.137023: step: 828/529, loss: 0.013302040286362171 2023-01-21 10:26:35.277500: step: 832/529, loss: 0.010504627600312233 2023-01-21 10:26:36.406316: step: 836/529, loss: 0.049408912658691406 2023-01-21 10:26:37.535963: step: 840/529, loss: 0.06428012996912003 2023-01-21 10:26:38.614529: step: 844/529, loss: 0.11536665260791779 2023-01-21 10:26:39.720970: step: 848/529, loss: 0.04080810397863388 2023-01-21 10:26:40.812350: step: 852/529, loss: 0.06284475326538086 2023-01-21 10:26:41.906966: step: 856/529, loss: 0.03513374179601669 2023-01-21 10:26:43.018010: step: 860/529, loss: 0.014374161139130592 2023-01-21 10:26:44.157235: step: 864/529, loss: 0.00845341756939888 2023-01-21 10:26:45.248086: step: 868/529, loss: 0.011118650436401367 2023-01-21 10:26:46.390682: step: 872/529, loss: 0.08096780627965927 2023-01-21 10:26:47.507282: step: 876/529, loss: 0.09415154904127121 2023-01-21 10:26:48.615744: step: 880/529, loss: 0.0038011549040675163 2023-01-21 10:26:49.713553: step: 884/529, loss: 0.043622732162475586 2023-01-21 10:26:50.804070: step: 888/529, loss: 0.02371211163699627 2023-01-21 10:26:51.965677: step: 892/529, loss: 0.11620187759399414 2023-01-21 10:26:53.074269: step: 896/529, loss: 0.010068322531878948 2023-01-21 10:26:54.163454: step: 900/529, loss: 0.0750301331281662 2023-01-21 10:26:55.282840: step: 904/529, loss: 0.14115962386131287 2023-01-21 10:26:56.433349: step: 908/529, loss: 0.027680015191435814 2023-01-21 10:26:57.564312: step: 912/529, loss: 0.055596206337213516 2023-01-21 10:26:58.657585: step: 916/529, loss: 0.059714317321777344 2023-01-21 10:26:59.778468: step: 920/529, loss: 0.046524617820978165 2023-01-21 10:27:00.906836: step: 924/529, loss: 0.07550687342882156 2023-01-21 10:27:02.057468: step: 928/529, loss: 0.05579419061541557 2023-01-21 10:27:03.161612: step: 932/529, loss: 0.024288274347782135 2023-01-21 10:27:04.298816: step: 936/529, loss: 0.035298872739076614 2023-01-21 10:27:05.424010: step: 940/529, loss: 0.01730179786682129 2023-01-21 10:27:06.562001: step: 944/529, loss: 0.018743038177490234 2023-01-21 10:27:07.709705: step: 948/529, loss: 0.06931991875171661 2023-01-21 10:27:08.823806: step: 952/529, loss: 0.10813484340906143 2023-01-21 10:27:09.933372: step: 956/529, loss: 0.2916943430900574 2023-01-21 10:27:11.040606: step: 960/529, loss: 0.0444183349609375 2023-01-21 10:27:12.179990: step: 964/529, loss: 0.005892562679946423 2023-01-21 10:27:13.269772: step: 968/529, loss: 0.0025506019592285156 2023-01-21 10:27:14.384653: step: 972/529, loss: 0.011970138177275658 2023-01-21 10:27:15.491518: step: 976/529, loss: 0.08323879539966583 2023-01-21 10:27:16.583804: step: 980/529, loss: 0.02632913738489151 2023-01-21 10:27:17.713156: step: 984/529, loss: 0.0189119353890419 2023-01-21 10:27:18.827129: step: 988/529, loss: 0.04235429689288139 2023-01-21 10:27:19.953728: step: 992/529, loss: 0.0024054530076682568 2023-01-21 10:27:21.053502: step: 996/529, loss: 0.04512377083301544 2023-01-21 10:27:22.177637: step: 1000/529, loss: 0.11528539657592773 2023-01-21 10:27:23.282977: step: 1004/529, loss: 0.009076309390366077 2023-01-21 10:27:24.384097: step: 1008/529, loss: 0.05923214182257652 2023-01-21 10:27:25.503653: step: 1012/529, loss: 0.04387540742754936 2023-01-21 10:27:26.607388: step: 1016/529, loss: 0.06292214244604111 2023-01-21 10:27:27.728648: step: 1020/529, loss: 0.037033937871456146 2023-01-21 10:27:28.857757: step: 1024/529, loss: 0.006282615941017866 2023-01-21 10:27:29.983734: step: 1028/529, loss: 0.10495968163013458 2023-01-21 10:27:31.104868: step: 1032/529, loss: 0.020461272448301315 2023-01-21 10:27:32.219871: step: 1036/529, loss: 0.02928142622113228 2023-01-21 10:27:33.309545: step: 1040/529, loss: 0.039076901972293854 2023-01-21 10:27:34.449966: step: 1044/529, loss: 1.8212745189666748 2023-01-21 10:27:35.598032: step: 1048/529, loss: 0.04300975799560547 2023-01-21 10:27:36.750478: step: 1052/529, loss: 0.08516388386487961 2023-01-21 10:27:37.885870: step: 1056/529, loss: 0.07014532387256622 2023-01-21 10:27:39.042888: step: 1060/529, loss: 0.02226734161376953 2023-01-21 10:27:40.200121: step: 1064/529, loss: 0.4326225817203522 2023-01-21 10:27:41.350637: step: 1068/529, loss: 0.4570654332637787 2023-01-21 10:27:42.498485: step: 1072/529, loss: 0.010837078094482422 2023-01-21 10:27:43.588068: step: 1076/529, loss: 0.03327608108520508 2023-01-21 10:27:44.743673: step: 1080/529, loss: 0.4362497329711914 2023-01-21 10:27:45.857575: step: 1084/529, loss: 0.016222380101680756 2023-01-21 10:27:46.984770: step: 1088/529, loss: 0.11577854305505753 2023-01-21 10:27:48.116008: step: 1092/529, loss: 0.08445930480957031 2023-01-21 10:27:49.239700: step: 1096/529, loss: 0.023132704198360443 2023-01-21 10:27:50.380671: step: 1100/529, loss: 0.012119198217988014 2023-01-21 10:27:51.534120: step: 1104/529, loss: 0.04031524434685707 2023-01-21 10:27:52.683788: step: 1108/529, loss: 0.07496533542871475 2023-01-21 10:27:53.825290: step: 1112/529, loss: 0.016175080090761185 2023-01-21 10:27:54.960898: step: 1116/529, loss: 0.007507038302719593 2023-01-21 10:27:56.086194: step: 1120/529, loss: 0.057738494127988815 2023-01-21 10:27:57.211155: step: 1124/529, loss: 0.07852659374475479 2023-01-21 10:27:58.330063: step: 1128/529, loss: 0.06139850616455078 2023-01-21 10:27:59.436195: step: 1132/529, loss: 0.08248867839574814 2023-01-21 10:28:00.533879: step: 1136/529, loss: 0.02164936065673828 2023-01-21 10:28:01.623960: step: 1140/529, loss: 0.008149909786880016 2023-01-21 10:28:02.764327: step: 1144/529, loss: 0.04141101986169815 2023-01-21 10:28:03.879800: step: 1148/529, loss: 0.044330596923828125 2023-01-21 10:28:05.000145: step: 1152/529, loss: 0.08643980324268341 2023-01-21 10:28:06.137620: step: 1156/529, loss: 0.05093735456466675 2023-01-21 10:28:07.250600: step: 1160/529, loss: 0.08762378990650177 2023-01-21 10:28:08.348172: step: 1164/529, loss: 0.029524803161621094 2023-01-21 10:28:09.511077: step: 1168/529, loss: 0.0019824982155114412 2023-01-21 10:28:10.613038: step: 1172/529, loss: 0.1028904914855957 2023-01-21 10:28:11.750291: step: 1176/529, loss: 0.034407127648591995 2023-01-21 10:28:12.839844: step: 1180/529, loss: 0.013445568270981312 2023-01-21 10:28:13.968519: step: 1184/529, loss: 0.059815600514411926 2023-01-21 10:28:15.068729: step: 1188/529, loss: 0.13818645477294922 2023-01-21 10:28:16.220858: step: 1192/529, loss: 0.02448730543255806 2023-01-21 10:28:17.342274: step: 1196/529, loss: 0.04031181335449219 2023-01-21 10:28:18.444120: step: 1200/529, loss: 0.06374092400074005 2023-01-21 10:28:19.538018: step: 1204/529, loss: 0.07150955498218536 2023-01-21 10:28:20.699397: step: 1208/529, loss: 0.21542797982692719 2023-01-21 10:28:21.798872: step: 1212/529, loss: 0.04838629066944122 2023-01-21 10:28:22.895695: step: 1216/529, loss: 0.0038162232376635075 2023-01-21 10:28:24.025934: step: 1220/529, loss: 0.07872195541858673 2023-01-21 10:28:25.110094: step: 1224/529, loss: 0.10093734413385391 2023-01-21 10:28:26.223789: step: 1228/529, loss: 0.23815956711769104 2023-01-21 10:28:27.329500: step: 1232/529, loss: 0.05550441890954971 2023-01-21 10:28:28.461112: step: 1236/529, loss: 0.0231951717287302 2023-01-21 10:28:29.561993: step: 1240/529, loss: 0.013909196481108665 2023-01-21 10:28:30.673577: step: 1244/529, loss: 0.15298815071582794 2023-01-21 10:28:31.782550: step: 1248/529, loss: 0.00274238595739007 2023-01-21 10:28:32.873095: step: 1252/529, loss: 0.04117918014526367 2023-01-21 10:28:33.984072: step: 1256/529, loss: 0.07713261246681213 2023-01-21 10:28:35.102064: step: 1260/529, loss: 0.027042007073760033 2023-01-21 10:28:36.196212: step: 1264/529, loss: 0.01716022565960884 2023-01-21 10:28:37.304146: step: 1268/529, loss: 0.14862041175365448 2023-01-21 10:28:38.395678: step: 1272/529, loss: 0.08967561274766922 2023-01-21 10:28:39.497693: step: 1276/529, loss: 0.013333129696547985 2023-01-21 10:28:40.625341: step: 1280/529, loss: 0.06216287612915039 2023-01-21 10:28:41.730908: step: 1284/529, loss: 0.025332070887088776 2023-01-21 10:28:42.821470: step: 1288/529, loss: 0.07996253669261932 2023-01-21 10:28:43.948531: step: 1292/529, loss: 0.19331598281860352 2023-01-21 10:28:45.064215: step: 1296/529, loss: 0.020469380542635918 2023-01-21 10:28:46.169540: step: 1300/529, loss: 0.08120517432689667 2023-01-21 10:28:47.278503: step: 1304/529, loss: 0.0030630589462816715 2023-01-21 10:28:48.388720: step: 1308/529, loss: 0.06431379169225693 2023-01-21 10:28:49.490533: step: 1312/529, loss: 0.06495921313762665 2023-01-21 10:28:50.593853: step: 1316/529, loss: 0.006024623289704323 2023-01-21 10:28:51.705431: step: 1320/529, loss: 0.01922912523150444 2023-01-21 10:28:52.825867: step: 1324/529, loss: 0.06610298156738281 2023-01-21 10:28:53.930931: step: 1328/529, loss: 0.10196170955896378 2023-01-21 10:28:55.059834: step: 1332/529, loss: 0.043562889099121094 2023-01-21 10:28:56.175271: step: 1336/529, loss: 0.002493953797966242 2023-01-21 10:28:57.299765: step: 1340/529, loss: 0.11579771339893341 2023-01-21 10:28:58.440717: step: 1344/529, loss: 0.023792171850800514 2023-01-21 10:28:59.568535: step: 1348/529, loss: 0.11767597496509552 2023-01-21 10:29:00.701491: step: 1352/529, loss: 0.07517480850219727 2023-01-21 10:29:01.804406: step: 1356/529, loss: 0.009095883928239346 2023-01-21 10:29:02.910151: step: 1360/529, loss: 0.03336753696203232 2023-01-21 10:29:04.030704: step: 1364/529, loss: 0.026984401047229767 2023-01-21 10:29:05.144612: step: 1368/529, loss: 0.192336767911911 2023-01-21 10:29:06.281802: step: 1372/529, loss: 0.3153672218322754 2023-01-21 10:29:07.379459: step: 1376/529, loss: 0.008458280935883522 2023-01-21 10:29:08.500652: step: 1380/529, loss: 0.04469604790210724 2023-01-21 10:29:09.618162: step: 1384/529, loss: 0.04680805280804634 2023-01-21 10:29:10.702923: step: 1388/529, loss: 0.05678558349609375 2023-01-21 10:29:11.804647: step: 1392/529, loss: 0.009009790606796741 2023-01-21 10:29:12.921663: step: 1396/529, loss: 0.09715954959392548 2023-01-21 10:29:14.033313: step: 1400/529, loss: 0.018822574988007545 2023-01-21 10:29:15.147498: step: 1404/529, loss: 0.01781749166548252 2023-01-21 10:29:16.292167: step: 1408/529, loss: 0.1727115660905838 2023-01-21 10:29:17.437093: step: 1412/529, loss: 0.013215900398790836 2023-01-21 10:29:18.562353: step: 1416/529, loss: 0.00029096603975631297 2023-01-21 10:29:19.674419: step: 1420/529, loss: 0.01745138131082058 2023-01-21 10:29:20.789639: step: 1424/529, loss: 0.10967454314231873 2023-01-21 10:29:21.918755: step: 1428/529, loss: 0.1127542108297348 2023-01-21 10:29:23.030644: step: 1432/529, loss: 0.4352186322212219 2023-01-21 10:29:24.169475: step: 1436/529, loss: 0.030798912048339844 2023-01-21 10:29:25.317101: step: 1440/529, loss: 0.33824315667152405 2023-01-21 10:29:26.459914: step: 1444/529, loss: 0.08715954422950745 2023-01-21 10:29:27.556480: step: 1448/529, loss: 0.11694198101758957 2023-01-21 10:29:28.694600: step: 1452/529, loss: 0.08019194006919861 2023-01-21 10:29:29.809795: step: 1456/529, loss: 0.041665174067020416 2023-01-21 10:29:30.947757: step: 1460/529, loss: 0.04990215227007866 2023-01-21 10:29:32.078193: step: 1464/529, loss: 0.04919024184346199 2023-01-21 10:29:33.181457: step: 1468/529, loss: 0.018529510125517845 2023-01-21 10:29:34.296427: step: 1472/529, loss: 0.10785657912492752 2023-01-21 10:29:35.422250: step: 1476/529, loss: 0.032964326441287994 2023-01-21 10:29:36.549749: step: 1480/529, loss: 0.03557319566607475 2023-01-21 10:29:37.649311: step: 1484/529, loss: 0.015588140115141869 2023-01-21 10:29:38.770221: step: 1488/529, loss: 0.11451463401317596 2023-01-21 10:29:39.896764: step: 1492/529, loss: 0.024896003305912018 2023-01-21 10:29:41.059638: step: 1496/529, loss: 0.03855838626623154 2023-01-21 10:29:42.178279: step: 1500/529, loss: 0.0448671355843544 2023-01-21 10:29:43.311331: step: 1504/529, loss: 0.016811847686767578 2023-01-21 10:29:44.403051: step: 1508/529, loss: 0.1740700751543045 2023-01-21 10:29:45.557476: step: 1512/529, loss: 0.06055469810962677 2023-01-21 10:29:46.691047: step: 1516/529, loss: 0.04968509450554848 2023-01-21 10:29:47.828757: step: 1520/529, loss: 0.21747250854969025 2023-01-21 10:29:48.987277: step: 1524/529, loss: 0.17350006103515625 2023-01-21 10:29:50.133537: step: 1528/529, loss: 0.011863518506288528 2023-01-21 10:29:51.293390: step: 1532/529, loss: 0.014476967044174671 2023-01-21 10:29:52.395252: step: 1536/529, loss: 0.033213138580322266 2023-01-21 10:29:53.571354: step: 1540/529, loss: 0.054854393005371094 2023-01-21 10:29:54.682783: step: 1544/529, loss: 0.06466779857873917 2023-01-21 10:29:55.837404: step: 1548/529, loss: 0.13704600930213928 2023-01-21 10:29:56.942891: step: 1552/529, loss: 0.007265186402946711 2023-01-21 10:29:58.038252: step: 1556/529, loss: 0.011124134063720703 2023-01-21 10:29:59.135515: step: 1560/529, loss: 0.21895579993724823 2023-01-21 10:30:00.266645: step: 1564/529, loss: 0.02361755259335041 2023-01-21 10:30:01.386248: step: 1568/529, loss: 0.051512762904167175 2023-01-21 10:30:02.475244: step: 1572/529, loss: 0.09439558535814285 2023-01-21 10:30:03.578455: step: 1576/529, loss: 0.03271055221557617 2023-01-21 10:30:04.711136: step: 1580/529, loss: 0.06723175197839737 2023-01-21 10:30:05.842592: step: 1584/529, loss: 0.11662255227565765 2023-01-21 10:30:06.929783: step: 1588/529, loss: 0.04502303525805473 2023-01-21 10:30:08.043943: step: 1592/529, loss: 0.07824917137622833 2023-01-21 10:30:09.138244: step: 1596/529, loss: 0.15707798302173615 2023-01-21 10:30:10.250604: step: 1600/529, loss: 0.13245125114917755 2023-01-21 10:30:11.384249: step: 1604/529, loss: 0.03590588644146919 2023-01-21 10:30:12.489957: step: 1608/529, loss: 0.03267412260174751 2023-01-21 10:30:13.610322: step: 1612/529, loss: 0.008365154266357422 2023-01-21 10:30:14.733774: step: 1616/529, loss: 0.061268400400877 2023-01-21 10:30:15.846260: step: 1620/529, loss: 0.4694023132324219 2023-01-21 10:30:16.970475: step: 1624/529, loss: 0.01960287243127823 2023-01-21 10:30:18.088528: step: 1628/529, loss: 0.07314815372228622 2023-01-21 10:30:19.302122: step: 1632/529, loss: 0.06247849389910698 2023-01-21 10:30:20.417455: step: 1636/529, loss: 0.0679868683218956 2023-01-21 10:30:21.506980: step: 1640/529, loss: 0.025119304656982422 2023-01-21 10:30:22.621062: step: 1644/529, loss: 0.08025984466075897 2023-01-21 10:30:23.706911: step: 1648/529, loss: 0.2910061776638031 2023-01-21 10:30:24.825957: step: 1652/529, loss: 0.020085670053958893 2023-01-21 10:30:25.939060: step: 1656/529, loss: 0.10923662781715393 2023-01-21 10:30:27.047076: step: 1660/529, loss: 0.6070134043693542 2023-01-21 10:30:28.188302: step: 1664/529, loss: 0.030062103644013405 2023-01-21 10:30:29.298395: step: 1668/529, loss: 0.016527796164155006 2023-01-21 10:30:30.399683: step: 1672/529, loss: 0.021309684962034225 2023-01-21 10:30:31.528542: step: 1676/529, loss: 0.11420612037181854 2023-01-21 10:30:32.670105: step: 1680/529, loss: 0.05534858629107475 2023-01-21 10:30:33.823947: step: 1684/529, loss: 0.07394818961620331 2023-01-21 10:30:34.931970: step: 1688/529, loss: 0.08401527255773544 2023-01-21 10:30:36.027873: step: 1692/529, loss: 0.050826214253902435 2023-01-21 10:30:37.147901: step: 1696/529, loss: 0.03811311721801758 2023-01-21 10:30:38.262865: step: 1700/529, loss: 0.04200448840856552 2023-01-21 10:30:39.348760: step: 1704/529, loss: 0.05162642151117325 2023-01-21 10:30:40.455496: step: 1708/529, loss: 0.041212987154722214 2023-01-21 10:30:41.561750: step: 1712/529, loss: 0.04088649898767471 2023-01-21 10:30:42.661821: step: 1716/529, loss: 0.08517150580883026 2023-01-21 10:30:43.778060: step: 1720/529, loss: 0.1025327742099762 2023-01-21 10:30:44.886769: step: 1724/529, loss: 0.033370211720466614 2023-01-21 10:30:46.022353: step: 1728/529, loss: 0.16073723137378693 2023-01-21 10:30:47.133249: step: 1732/529, loss: 0.008471203036606312 2023-01-21 10:30:48.277795: step: 1736/529, loss: 0.1142568588256836 2023-01-21 10:30:49.421648: step: 1740/529, loss: 0.026398276910185814 2023-01-21 10:30:50.564111: step: 1744/529, loss: 0.029660701751708984 2023-01-21 10:30:51.647463: step: 1748/529, loss: 0.07304278016090393 2023-01-21 10:30:52.744346: step: 1752/529, loss: 0.14309760928153992 2023-01-21 10:30:53.851976: step: 1756/529, loss: 0.054224301129579544 2023-01-21 10:30:54.974065: step: 1760/529, loss: 0.0358828529715538 2023-01-21 10:30:56.077854: step: 1764/529, loss: 0.0690569132566452 2023-01-21 10:30:57.201705: step: 1768/529, loss: 0.026938725262880325 2023-01-21 10:30:58.309422: step: 1772/529, loss: 0.014733219519257545 2023-01-21 10:30:59.440859: step: 1776/529, loss: 0.005133247468620539 2023-01-21 10:31:00.587485: step: 1780/529, loss: 0.02994842454791069 2023-01-21 10:31:01.689529: step: 1784/529, loss: 0.023548031225800514 2023-01-21 10:31:02.797375: step: 1788/529, loss: 0.10314564406871796 2023-01-21 10:31:03.921722: step: 1792/529, loss: 0.058193400502204895 2023-01-21 10:31:05.039763: step: 1796/529, loss: 0.09013862162828445 2023-01-21 10:31:06.146220: step: 1800/529, loss: 0.007958268746733665 2023-01-21 10:31:07.232802: step: 1804/529, loss: 0.0593988411128521 2023-01-21 10:31:08.342552: step: 1808/529, loss: 0.1553330421447754 2023-01-21 10:31:09.465916: step: 1812/529, loss: 0.09362685680389404 2023-01-21 10:31:10.570500: step: 1816/529, loss: 0.030263518914580345 2023-01-21 10:31:11.710714: step: 1820/529, loss: 0.041013337671756744 2023-01-21 10:31:12.869397: step: 1824/529, loss: 0.08637180924415588 2023-01-21 10:31:13.967511: step: 1828/529, loss: 0.13820229470729828 2023-01-21 10:31:15.073665: step: 1832/529, loss: 0.029839742928743362 2023-01-21 10:31:16.180691: step: 1836/529, loss: 0.026963043957948685 2023-01-21 10:31:17.321201: step: 1840/529, loss: 0.0016651629703119397 2023-01-21 10:31:18.435526: step: 1844/529, loss: 0.06535835564136505 2023-01-21 10:31:19.525891: step: 1848/529, loss: 0.02187070995569229 2023-01-21 10:31:20.647811: step: 1852/529, loss: 0.04506850242614746 2023-01-21 10:31:21.766881: step: 1856/529, loss: 0.06857109069824219 2023-01-21 10:31:22.871545: step: 1860/529, loss: 0.4883388578891754 2023-01-21 10:31:23.976939: step: 1864/529, loss: 0.025562096387147903 2023-01-21 10:31:25.085506: step: 1868/529, loss: 0.27632060647010803 2023-01-21 10:31:26.194686: step: 1872/529, loss: 0.17139454185962677 2023-01-21 10:31:27.310367: step: 1876/529, loss: 0.019004035741090775 2023-01-21 10:31:28.394632: step: 1880/529, loss: 0.03706197813153267 2023-01-21 10:31:29.528778: step: 1884/529, loss: 0.04961347579956055 2023-01-21 10:31:30.623737: step: 1888/529, loss: 0.023571301251649857 2023-01-21 10:31:31.723248: step: 1892/529, loss: 0.0631624236702919 2023-01-21 10:31:32.853046: step: 1896/529, loss: 0.019174670800566673 2023-01-21 10:31:33.960222: step: 1900/529, loss: 0.005490779876708984 2023-01-21 10:31:35.079747: step: 1904/529, loss: 0.11076603084802628 2023-01-21 10:31:36.201480: step: 1908/529, loss: 0.06614828109741211 2023-01-21 10:31:37.286737: step: 1912/529, loss: 0.3008156716823578 2023-01-21 10:31:38.390068: step: 1916/529, loss: 0.20936965942382812 2023-01-21 10:31:39.485504: step: 1920/529, loss: 0.03309326246380806 2023-01-21 10:31:40.618989: step: 1924/529, loss: 0.08620509505271912 2023-01-21 10:31:41.730401: step: 1928/529, loss: 0.019383668899536133 2023-01-21 10:31:42.854457: step: 1932/529, loss: 0.06761030852794647 2023-01-21 10:31:43.988781: step: 1936/529, loss: 0.1002282127737999 2023-01-21 10:31:45.123130: step: 1940/529, loss: 0.05093355104327202 2023-01-21 10:31:46.266773: step: 1944/529, loss: 0.3376179039478302 2023-01-21 10:31:47.413025: step: 1948/529, loss: 0.12597909569740295 2023-01-21 10:31:48.517445: step: 1952/529, loss: 0.15820197761058807 2023-01-21 10:31:49.611467: step: 1956/529, loss: 0.09152374416589737 2023-01-21 10:31:50.739448: step: 1960/529, loss: 0.0688050240278244 2023-01-21 10:31:51.877247: step: 1964/529, loss: 0.03563213720917702 2023-01-21 10:31:52.999908: step: 1968/529, loss: 0.12054653465747833 2023-01-21 10:31:54.142264: step: 1972/529, loss: 0.18245716392993927 2023-01-21 10:31:55.255587: step: 1976/529, loss: 0.1645512580871582 2023-01-21 10:31:56.345886: step: 1980/529, loss: 0.15209999680519104 2023-01-21 10:31:57.458253: step: 1984/529, loss: 0.014332962222397327 2023-01-21 10:31:58.570958: step: 1988/529, loss: 0.05136473476886749 2023-01-21 10:31:59.718767: step: 1992/529, loss: 0.3699323534965515 2023-01-21 10:32:00.828426: step: 1996/529, loss: 0.10661458969116211 2023-01-21 10:32:01.958888: step: 2000/529, loss: 0.12488795071840286 2023-01-21 10:32:03.069488: step: 2004/529, loss: 0.053247831761837006 2023-01-21 10:32:04.188698: step: 2008/529, loss: 0.029947664588689804 2023-01-21 10:32:05.298446: step: 2012/529, loss: 0.05026264116168022 2023-01-21 10:32:06.393536: step: 2016/529, loss: 0.11242732405662537 2023-01-21 10:32:07.487524: step: 2020/529, loss: 0.027987100183963776 2023-01-21 10:32:08.628969: step: 2024/529, loss: 0.16999435424804688 2023-01-21 10:32:09.735809: step: 2028/529, loss: 0.079311802983284 2023-01-21 10:32:10.848584: step: 2032/529, loss: 0.02260131761431694 2023-01-21 10:32:11.994041: step: 2036/529, loss: 0.006755828857421875 2023-01-21 10:32:13.164346: step: 2040/529, loss: 0.001632904983125627 2023-01-21 10:32:14.275961: step: 2044/529, loss: 0.1355283260345459 2023-01-21 10:32:15.396032: step: 2048/529, loss: 0.13581351935863495 2023-01-21 10:32:16.544218: step: 2052/529, loss: 0.11528797447681427 2023-01-21 10:32:17.681202: step: 2056/529, loss: 0.026104355230927467 2023-01-21 10:32:18.782428: step: 2060/529, loss: 0.48078620433807373 2023-01-21 10:32:19.891996: step: 2064/529, loss: 0.09292316436767578 2023-01-21 10:32:21.018682: step: 2068/529, loss: 0.16723380982875824 2023-01-21 10:32:22.157781: step: 2072/529, loss: 0.014786816202104092 2023-01-21 10:32:23.276533: step: 2076/529, loss: 0.0852775126695633 2023-01-21 10:32:24.374359: step: 2080/529, loss: 0.008470773696899414 2023-01-21 10:32:25.467238: step: 2084/529, loss: 0.05221007019281387 2023-01-21 10:32:26.597269: step: 2088/529, loss: 0.05150194466114044 2023-01-21 10:32:27.715302: step: 2092/529, loss: 0.026974773034453392 2023-01-21 10:32:28.804057: step: 2096/529, loss: 0.09263592213392258 2023-01-21 10:32:29.945263: step: 2100/529, loss: 0.027650833129882812 2023-01-21 10:32:31.026412: step: 2104/529, loss: 0.08481350541114807 2023-01-21 10:32:32.111099: step: 2108/529, loss: 0.002028989838436246 2023-01-21 10:32:33.226423: step: 2112/529, loss: 0.02868366241455078 2023-01-21 10:32:34.345138: step: 2116/529, loss: 0.019691206514835358 ================================================== Loss: 0.076 -------------------- Dev: {'event': {'p': 0.5742574257425742, 'r': 0.7723035952063915, 'f1': 0.6587166382737082}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Test: {'event': {'p': 0.6344232515894641, 'r': 0.7883747178329571, 'f1': 0.7030699547055864}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Chinese: {'event': {'p': 0.573170731707317, 'r': 0.8703703703703703, 'f1': 0.6911764705882353}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Korean: {'event': {'p': 0.6, 'r': 0.6190476190476191, 'f1': 0.609375}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Russian: {'event': {'p': 0.41025641025641024, 'r': 0.4444444444444444, 'f1': 0.42666666666666664}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** 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 10:33:15.389444: step: 4/529, loss: 0.015634585171937943 2023-01-21 10:33:16.500518: step: 8/529, loss: 0.006419563200324774 2023-01-21 10:33:17.593726: step: 12/529, loss: 0.04357729107141495 2023-01-21 10:33:18.675453: step: 16/529, loss: 0.014276808127760887 2023-01-21 10:33:19.819559: step: 20/529, loss: 0.08221979439258575 2023-01-21 10:33:20.967269: step: 24/529, loss: 0.015496348962187767 2023-01-21 10:33:22.057344: step: 28/529, loss: 0.0071776630356907845 2023-01-21 10:33:23.174638: step: 32/529, loss: 0.06823565810918808 2023-01-21 10:33:24.287835: step: 36/529, loss: 0.08248968422412872 2023-01-21 10:33:25.397473: step: 40/529, loss: 0.004479527473449707 2023-01-21 10:33:26.551939: step: 44/529, loss: 0.13717421889305115 2023-01-21 10:33:27.635630: step: 48/529, loss: 0.007056808564811945 2023-01-21 10:33:28.766430: step: 52/529, loss: 0.08719882369041443 2023-01-21 10:33:29.900841: step: 56/529, loss: 0.6353353261947632 2023-01-21 10:33:30.999627: step: 60/529, loss: 0.05805234983563423 2023-01-21 10:33:32.108683: step: 64/529, loss: 0.04892003536224365 2023-01-21 10:33:33.233536: step: 68/529, loss: 0.1041591688990593 2023-01-21 10:33:34.328479: step: 72/529, loss: 0.04996924474835396 2023-01-21 10:33:35.438595: step: 76/529, loss: 0.3311885893344879 2023-01-21 10:33:36.552120: step: 80/529, loss: 0.6494429111480713 2023-01-21 10:33:37.645357: step: 84/529, loss: 0.15629205107688904 2023-01-21 10:33:38.761965: step: 88/529, loss: 0.035772036761045456 2023-01-21 10:33:39.911332: step: 92/529, loss: 0.05627737194299698 2023-01-21 10:33:41.027944: step: 96/529, loss: 0.011458969675004482 2023-01-21 10:33:42.145200: step: 100/529, loss: 0.15678353607654572 2023-01-21 10:33:43.276431: step: 104/529, loss: 0.03597602993249893 2023-01-21 10:33:44.386705: step: 108/529, loss: 0.058695316314697266 2023-01-21 10:33:45.490584: step: 112/529, loss: 0.0338197723031044 2023-01-21 10:33:46.629107: step: 116/529, loss: 0.14727360010147095 2023-01-21 10:33:47.740279: step: 120/529, loss: 0.03142547607421875 2023-01-21 10:33:48.850620: step: 124/529, loss: 0.033518124371767044 2023-01-21 10:33:49.952965: step: 128/529, loss: 0.11283960938453674 2023-01-21 10:33:51.077752: step: 132/529, loss: 0.017546653747558594 2023-01-21 10:33:52.200768: step: 136/529, loss: 0.11903420090675354 2023-01-21 10:33:53.282603: step: 140/529, loss: 0.3157869279384613 2023-01-21 10:33:54.383575: step: 144/529, loss: 0.1855243742465973 2023-01-21 10:33:55.474590: step: 148/529, loss: 0.021838141605257988 2023-01-21 10:33:56.582366: step: 152/529, loss: 0.007050132844597101 2023-01-21 10:33:57.684217: step: 156/529, loss: 0.05045595392584801 2023-01-21 10:33:58.821163: step: 160/529, loss: 0.006897354498505592 2023-01-21 10:33:59.920626: step: 164/529, loss: 0.07123079895973206 2023-01-21 10:34:01.061360: step: 168/529, loss: 0.020231151953339577 2023-01-21 10:34:02.177101: step: 172/529, loss: 0.06526605039834976 2023-01-21 10:34:03.285789: step: 176/529, loss: 0.042412757873535156 2023-01-21 10:34:04.394617: step: 180/529, loss: 0.02250983752310276 2023-01-21 10:34:05.468285: step: 184/529, loss: 0.04285914823412895 2023-01-21 10:34:06.582880: step: 188/529, loss: 0.048415567725896835 2023-01-21 10:34:07.691973: step: 192/529, loss: 0.08613376319408417 2023-01-21 10:34:08.797270: step: 196/529, loss: 0.011941147036850452 2023-01-21 10:34:09.940904: step: 200/529, loss: 0.021020984277129173 2023-01-21 10:34:11.037418: step: 204/529, loss: 0.12843847274780273 2023-01-21 10:34:12.154530: step: 208/529, loss: 0.0032606124877929688 2023-01-21 10:34:13.256062: step: 212/529, loss: 0.3989928364753723 2023-01-21 10:34:14.364436: step: 216/529, loss: 0.020298242568969727 2023-01-21 10:34:15.460577: step: 220/529, loss: 0.05939054489135742 2023-01-21 10:34:16.618084: step: 224/529, loss: 0.1745346188545227 2023-01-21 10:34:17.712834: step: 228/529, loss: 0.10474815964698792 2023-01-21 10:34:18.829019: step: 232/529, loss: 0.0008929253090173006 2023-01-21 10:34:20.010189: step: 236/529, loss: 0.13309641182422638 2023-01-21 10:34:21.137454: step: 240/529, loss: 0.01626415364444256 2023-01-21 10:34:22.274892: step: 244/529, loss: 0.012050438672304153 2023-01-21 10:34:23.378812: step: 248/529, loss: 0.034877873957157135 2023-01-21 10:34:24.517257: step: 252/529, loss: 0.0013515471946448088 2023-01-21 10:34:25.631933: step: 256/529, loss: 0.00017495155043434352 2023-01-21 10:34:26.754073: step: 260/529, loss: 0.059261322021484375 2023-01-21 10:34:27.870549: step: 264/529, loss: 0.008837556466460228 2023-01-21 10:34:28.973900: step: 268/529, loss: 0.03994917869567871 2023-01-21 10:34:30.088519: step: 272/529, loss: 0.029516959562897682 2023-01-21 10:34:31.186066: step: 276/529, loss: 0.01847228966653347 2023-01-21 10:34:32.337611: step: 280/529, loss: 0.05752287060022354 2023-01-21 10:34:33.443391: step: 284/529, loss: 0.41338494420051575 2023-01-21 10:34:34.547025: step: 288/529, loss: 0.006676006130874157 2023-01-21 10:34:35.673835: step: 292/529, loss: 0.02326345629990101 2023-01-21 10:34:36.800487: step: 296/529, loss: 0.23070678114891052 2023-01-21 10:34:37.902767: step: 300/529, loss: 0.05969208478927612 2023-01-21 10:34:39.017363: step: 304/529, loss: 0.17352323234081268 2023-01-21 10:34:40.163053: step: 308/529, loss: 0.10289621353149414 2023-01-21 10:34:41.271989: step: 312/529, loss: 0.0342593677341938 2023-01-21 10:34:42.384903: step: 316/529, loss: 0.03202342987060547 2023-01-21 10:34:43.505592: step: 320/529, loss: 0.01991640403866768 2023-01-21 10:34:44.612452: step: 324/529, loss: 0.10441437363624573 2023-01-21 10:34:45.731019: step: 328/529, loss: 0.022480487823486328 2023-01-21 10:34:46.833232: step: 332/529, loss: 0.01696782186627388 2023-01-21 10:34:47.931679: step: 336/529, loss: 0.13196228444576263 2023-01-21 10:34:49.074302: step: 340/529, loss: 0.001255083130672574 2023-01-21 10:34:50.237567: step: 344/529, loss: 0.09300652146339417 2023-01-21 10:34:51.327471: step: 348/529, loss: 0.071904756128788 2023-01-21 10:34:52.464495: step: 352/529, loss: 0.49297693371772766 2023-01-21 10:34:53.584578: step: 356/529, loss: 0.1843484491109848 2023-01-21 10:34:54.708943: step: 360/529, loss: 0.018433570861816406 2023-01-21 10:34:55.828749: step: 364/529, loss: 0.09107494354248047 2023-01-21 10:34:56.945472: step: 368/529, loss: 0.07947796583175659 2023-01-21 10:34:58.067441: step: 372/529, loss: 0.04172487556934357 2023-01-21 10:34:59.203539: step: 376/529, loss: 0.01647162437438965 2023-01-21 10:35:00.318870: step: 380/529, loss: 0.08153554052114487 2023-01-21 10:35:01.399956: step: 384/529, loss: 0.03464052826166153 2023-01-21 10:35:02.505338: step: 388/529, loss: 0.03863248974084854 2023-01-21 10:35:03.639572: step: 392/529, loss: 0.007461499888449907 2023-01-21 10:35:04.778780: step: 396/529, loss: 0.054502058774232864 2023-01-21 10:35:05.894589: step: 400/529, loss: 0.06054706871509552 2023-01-21 10:35:07.024879: step: 404/529, loss: 0.23481817543506622 2023-01-21 10:35:08.125890: step: 408/529, loss: 0.06370801478624344 2023-01-21 10:35:09.227044: step: 412/529, loss: 0.01590433157980442 2023-01-21 10:35:10.373380: step: 416/529, loss: 0.05279908329248428 2023-01-21 10:35:11.490619: step: 420/529, loss: 0.17367516458034515 2023-01-21 10:35:12.592621: step: 424/529, loss: 0.002672863192856312 2023-01-21 10:35:13.720355: step: 428/529, loss: 0.003030681749805808 2023-01-21 10:35:14.822219: step: 432/529, loss: 0.1320597231388092 2023-01-21 10:35:15.930352: step: 436/529, loss: 0.010323334485292435 2023-01-21 10:35:17.024238: step: 440/529, loss: 0.02219104766845703 2023-01-21 10:35:18.130869: step: 444/529, loss: 0.016097258776426315 2023-01-21 10:35:19.255031: step: 448/529, loss: 0.04870261996984482 2023-01-21 10:35:20.440148: step: 452/529, loss: 0.02037172205746174 2023-01-21 10:35:21.570854: step: 456/529, loss: 0.11630412936210632 2023-01-21 10:35:22.706956: step: 460/529, loss: 0.03807549551129341 2023-01-21 10:35:23.809172: step: 464/529, loss: 0.0593935027718544 2023-01-21 10:35:24.906838: step: 468/529, loss: 0.006680297665297985 2023-01-21 10:35:26.025095: step: 472/529, loss: 0.002512168837711215 2023-01-21 10:35:27.110535: step: 476/529, loss: 0.03684987872838974 2023-01-21 10:35:28.218316: step: 480/529, loss: 0.012775707989931107 2023-01-21 10:35:29.314148: step: 484/529, loss: 0.03353925049304962 2023-01-21 10:35:30.422882: step: 488/529, loss: 0.03995390236377716 2023-01-21 10:35:31.540252: step: 492/529, loss: 0.2139018028974533 2023-01-21 10:35:32.631295: step: 496/529, loss: 0.1484355479478836 2023-01-21 10:35:33.780499: step: 500/529, loss: 0.04455547407269478 2023-01-21 10:35:34.925450: step: 504/529, loss: 0.3537158966064453 2023-01-21 10:35:36.036321: step: 508/529, loss: 0.3285512924194336 2023-01-21 10:35:37.159836: step: 512/529, loss: 0.1303553283214569 2023-01-21 10:35:38.278422: step: 516/529, loss: 0.06636085361242294 2023-01-21 10:35:39.431002: step: 520/529, loss: 0.0025569917634129524 2023-01-21 10:35:40.524400: step: 524/529, loss: 0.0895816832780838 2023-01-21 10:35:41.654661: step: 528/529, loss: 0.018782544881105423 2023-01-21 10:35:42.768981: step: 532/529, loss: 0.03824892267584801 2023-01-21 10:35:43.868569: step: 536/529, loss: 0.00927422009408474 2023-01-21 10:35:44.992666: step: 540/529, loss: 0.10300073772668839 2023-01-21 10:35:46.122051: step: 544/529, loss: 0.03220396116375923 2023-01-21 10:35:47.246362: step: 548/529, loss: 0.0069176675751805305 2023-01-21 10:35:48.368373: step: 552/529, loss: 0.08452148735523224 2023-01-21 10:35:49.473086: step: 556/529, loss: 0.4841233491897583 2023-01-21 10:35:50.572451: step: 560/529, loss: 0.017428064718842506 2023-01-21 10:35:51.675513: step: 564/529, loss: 0.05555610731244087 2023-01-21 10:35:52.787784: step: 568/529, loss: 0.07069268822669983 2023-01-21 10:35:53.932520: step: 572/529, loss: 0.006107664201408625 2023-01-21 10:35:55.066872: step: 576/529, loss: 0.2340092658996582 2023-01-21 10:35:56.185922: step: 580/529, loss: 0.08533401787281036 2023-01-21 10:35:57.299271: step: 584/529, loss: 0.013085913844406605 2023-01-21 10:35:58.423822: step: 588/529, loss: 0.026302529498934746 2023-01-21 10:35:59.541776: step: 592/529, loss: 0.009231185540556908 2023-01-21 10:36:00.665922: step: 596/529, loss: 0.06901588290929794 2023-01-21 10:36:01.792010: step: 600/529, loss: 0.0263992790132761 2023-01-21 10:36:02.907950: step: 604/529, loss: 0.021418094635009766 2023-01-21 10:36:04.035317: step: 608/529, loss: 0.05278473347425461 2023-01-21 10:36:05.156906: step: 612/529, loss: 0.013976049609482288 2023-01-21 10:36:06.277538: step: 616/529, loss: 0.07042255997657776 2023-01-21 10:36:07.384982: step: 620/529, loss: 0.012071705423295498 2023-01-21 10:36:08.503883: step: 624/529, loss: 0.03619384765625 2023-01-21 10:36:09.632717: step: 628/529, loss: 0.010102653875946999 2023-01-21 10:36:10.752888: step: 632/529, loss: 0.0027556419372558594 2023-01-21 10:36:11.873043: step: 636/529, loss: 0.04009075090289116 2023-01-21 10:36:13.053752: step: 640/529, loss: 0.054708197712898254 2023-01-21 10:36:14.161661: step: 644/529, loss: 0.027746200561523438 2023-01-21 10:36:15.264677: step: 648/529, loss: 0.015670109540224075 2023-01-21 10:36:16.393939: step: 652/529, loss: 0.0326201468706131 2023-01-21 10:36:17.498295: step: 656/529, loss: 0.04464688524603844 2023-01-21 10:36:18.592601: step: 660/529, loss: 0.052407365292310715 2023-01-21 10:36:19.709793: step: 664/529, loss: 0.5318622589111328 2023-01-21 10:36:20.809692: step: 668/529, loss: 0.015525818802416325 2023-01-21 10:36:21.911768: step: 672/529, loss: 0.019034672528505325 2023-01-21 10:36:23.023768: step: 676/529, loss: 0.01824321784079075 2023-01-21 10:36:24.101880: step: 680/529, loss: 0.03645677492022514 2023-01-21 10:36:25.225682: step: 684/529, loss: 0.019097138196229935 2023-01-21 10:36:26.339591: step: 688/529, loss: 0.04615459591150284 2023-01-21 10:36:27.437612: step: 692/529, loss: 0.011239337734878063 2023-01-21 10:36:28.574332: step: 696/529, loss: 0.006378937046974897 2023-01-21 10:36:29.668246: step: 700/529, loss: 0.054081253707408905 2023-01-21 10:36:30.771997: step: 704/529, loss: 0.1335126906633377 2023-01-21 10:36:31.869519: step: 708/529, loss: 0.002607727190479636 2023-01-21 10:36:33.006238: step: 712/529, loss: 0.11718215048313141 2023-01-21 10:36:34.154140: step: 716/529, loss: 0.02962026558816433 2023-01-21 10:36:35.264157: step: 720/529, loss: 0.061727799475193024 2023-01-21 10:36:36.353306: step: 724/529, loss: 0.022665690630674362 2023-01-21 10:36:37.481945: step: 728/529, loss: 0.0004139900556765497 2023-01-21 10:36:38.600209: step: 732/529, loss: 0.019195938482880592 2023-01-21 10:36:39.703384: step: 736/529, loss: 0.031145954504609108 2023-01-21 10:36:40.824556: step: 740/529, loss: 0.05455417558550835 2023-01-21 10:36:41.913926: step: 744/529, loss: 0.008442306891083717 2023-01-21 10:36:43.046285: step: 748/529, loss: 0.0033290863502770662 2023-01-21 10:36:44.148073: step: 752/529, loss: 0.07226744294166565 2023-01-21 10:36:45.257754: step: 756/529, loss: 0.07950744777917862 2023-01-21 10:36:46.360422: step: 760/529, loss: 0.0033757209312170744 2023-01-21 10:36:47.467239: step: 764/529, loss: 0.04112539440393448 2023-01-21 10:36:48.568086: step: 768/529, loss: 0.25970783829689026 2023-01-21 10:36:49.681835: step: 772/529, loss: 0.005614042282104492 2023-01-21 10:36:50.786452: step: 776/529, loss: 0.06463642418384552 2023-01-21 10:36:51.931816: step: 780/529, loss: 0.031077099964022636 2023-01-21 10:36:53.009324: step: 784/529, loss: 0.03211555629968643 2023-01-21 10:36:54.118345: step: 788/529, loss: 0.05314221233129501 2023-01-21 10:36:55.230529: step: 792/529, loss: 0.0020237923599779606 2023-01-21 10:36:56.343353: step: 796/529, loss: 0.07007293403148651 2023-01-21 10:36:57.460226: step: 800/529, loss: 0.025855064392089844 2023-01-21 10:36:58.564910: step: 804/529, loss: 0.008034516125917435 2023-01-21 10:36:59.697074: step: 808/529, loss: 0.03595848008990288 2023-01-21 10:37:00.816412: step: 812/529, loss: 0.0032608031760901213 2023-01-21 10:37:01.915108: step: 816/529, loss: 0.013169003650546074 2023-01-21 10:37:03.030309: step: 820/529, loss: 0.024141501635313034 2023-01-21 10:37:04.191374: step: 824/529, loss: 0.1024479866027832 2023-01-21 10:37:05.295124: step: 828/529, loss: 0.019344709813594818 2023-01-21 10:37:06.377397: step: 832/529, loss: 0.0014323710929602385 2023-01-21 10:37:07.465550: step: 836/529, loss: 0.029245663434267044 2023-01-21 10:37:08.569146: step: 840/529, loss: 0.013254070654511452 2023-01-21 10:37:09.675257: step: 844/529, loss: 0.020444203168153763 2023-01-21 10:37:10.830768: step: 848/529, loss: 0.07971153408288956 2023-01-21 10:37:11.978601: step: 852/529, loss: 0.18301840126514435 2023-01-21 10:37:13.090977: step: 856/529, loss: 0.02782120741903782 2023-01-21 10:37:14.231793: step: 860/529, loss: 0.015463829040527344 2023-01-21 10:37:15.352400: step: 864/529, loss: 0.03686103969812393 2023-01-21 10:37:16.459392: step: 868/529, loss: 0.023042678833007812 2023-01-21 10:37:17.549126: step: 872/529, loss: 0.046110060065984726 2023-01-21 10:37:18.660825: step: 876/529, loss: 0.032063912600278854 2023-01-21 10:37:19.773064: step: 880/529, loss: 0.06866931915283203 2023-01-21 10:37:20.896492: step: 884/529, loss: 0.02893519401550293 2023-01-21 10:37:22.025258: step: 888/529, loss: 0.5382149815559387 2023-01-21 10:37:23.160122: step: 892/529, loss: 0.05076196417212486 2023-01-21 10:37:24.261167: step: 896/529, loss: 0.06138572841882706 2023-01-21 10:37:25.377060: step: 900/529, loss: 0.037667322903871536 2023-01-21 10:37:26.497102: step: 904/529, loss: 0.062125399708747864 2023-01-21 10:37:27.614557: step: 908/529, loss: 0.14272108674049377 2023-01-21 10:37:28.743783: step: 912/529, loss: 0.031221581622958183 2023-01-21 10:37:29.871007: step: 916/529, loss: 0.03425560146570206 2023-01-21 10:37:31.023854: step: 920/529, loss: 0.03369269147515297 2023-01-21 10:37:32.119997: step: 924/529, loss: 0.0652259811758995 2023-01-21 10:37:33.232518: step: 928/529, loss: 0.06813813000917435 2023-01-21 10:37:34.389778: step: 932/529, loss: 0.013027573004364967 2023-01-21 10:37:35.516265: step: 936/529, loss: 0.025594711303710938 2023-01-21 10:37:36.630643: step: 940/529, loss: 0.016342926770448685 2023-01-21 10:37:37.739102: step: 944/529, loss: 0.012179946526885033 2023-01-21 10:37:38.870069: step: 948/529, loss: 1.0101159811019897 2023-01-21 10:37:39.975610: step: 952/529, loss: 0.11328279972076416 2023-01-21 10:37:41.098211: step: 956/529, loss: 0.017920399084687233 2023-01-21 10:37:42.218489: step: 960/529, loss: 0.037798501551151276 2023-01-21 10:37:43.341337: step: 964/529, loss: 0.12885704636573792 2023-01-21 10:37:44.478235: step: 968/529, loss: 0.001981496810913086 2023-01-21 10:37:45.585881: step: 972/529, loss: 0.030063629150390625 2023-01-21 10:37:46.698921: step: 976/529, loss: 0.03332536295056343 2023-01-21 10:37:47.816474: step: 980/529, loss: 0.04087886959314346 2023-01-21 10:37:48.959965: step: 984/529, loss: 0.039319656789302826 2023-01-21 10:37:50.077722: step: 988/529, loss: 0.10448842495679855 2023-01-21 10:37:51.215741: step: 992/529, loss: 0.04496314749121666 2023-01-21 10:37:52.352327: step: 996/529, loss: 0.03794918209314346 2023-01-21 10:37:53.464151: step: 1000/529, loss: 0.018438149243593216 2023-01-21 10:37:54.577192: step: 1004/529, loss: 0.01054925937205553 2023-01-21 10:37:55.698669: step: 1008/529, loss: 0.1984979659318924 2023-01-21 10:37:56.828754: step: 1012/529, loss: 0.012137318029999733 2023-01-21 10:37:57.956678: step: 1016/529, loss: 0.016981076449155807 2023-01-21 10:37:59.051263: step: 1020/529, loss: 0.054285429418087006 2023-01-21 10:38:00.173893: step: 1024/529, loss: 0.025885486975312233 2023-01-21 10:38:01.289670: step: 1028/529, loss: 0.031004810705780983 2023-01-21 10:38:02.398733: step: 1032/529, loss: 0.009832668118178844 2023-01-21 10:38:03.497770: step: 1036/529, loss: 0.020828628912568092 2023-01-21 10:38:04.612225: step: 1040/529, loss: 0.07629795372486115 2023-01-21 10:38:05.737987: step: 1044/529, loss: 0.04315061494708061 2023-01-21 10:38:06.867740: step: 1048/529, loss: 0.009748315438628197 2023-01-21 10:38:07.989700: step: 1052/529, loss: 0.06804456561803818 2023-01-21 10:38:09.074110: step: 1056/529, loss: 0.05234070122241974 2023-01-21 10:38:10.175679: step: 1060/529, loss: 0.03478412330150604 2023-01-21 10:38:11.271334: step: 1064/529, loss: 0.022547150030732155 2023-01-21 10:38:12.407178: step: 1068/529, loss: 0.0698062926530838 2023-01-21 10:38:13.508527: step: 1072/529, loss: 0.07025470584630966 2023-01-21 10:38:14.639145: step: 1076/529, loss: 0.06398516148328781 2023-01-21 10:38:15.760820: step: 1080/529, loss: 0.1375909149646759 2023-01-21 10:38:16.909974: step: 1084/529, loss: 0.040517523884773254 2023-01-21 10:38:17.997398: step: 1088/529, loss: 0.14841842651367188 2023-01-21 10:38:19.132318: step: 1092/529, loss: 0.086273193359375 2023-01-21 10:38:20.236699: step: 1096/529, loss: 0.06956568360328674 2023-01-21 10:38:21.339933: step: 1100/529, loss: 0.08271627128124237 2023-01-21 10:38:22.454645: step: 1104/529, loss: 0.08006992191076279 2023-01-21 10:38:23.556749: step: 1108/529, loss: 0.2676321864128113 2023-01-21 10:38:24.667294: step: 1112/529, loss: 0.08240614086389542 2023-01-21 10:38:25.799130: step: 1116/529, loss: 0.10963144898414612 2023-01-21 10:38:26.893933: step: 1120/529, loss: 0.015755273401737213 2023-01-21 10:38:28.042997: step: 1124/529, loss: 0.0639006644487381 2023-01-21 10:38:29.153052: step: 1128/529, loss: 0.008260917849838734 2023-01-21 10:38:30.294626: step: 1132/529, loss: 0.00912246759980917 2023-01-21 10:38:31.471062: step: 1136/529, loss: 0.05913090705871582 2023-01-21 10:38:32.583763: step: 1140/529, loss: 0.06082306057214737 2023-01-21 10:38:33.701172: step: 1144/529, loss: 0.047199077904224396 2023-01-21 10:38:34.808376: step: 1148/529, loss: 0.00542340287938714 2023-01-21 10:38:35.909131: step: 1152/529, loss: 0.00975647009909153 2023-01-21 10:38:37.006266: step: 1156/529, loss: 0.056063272058963776 2023-01-21 10:38:38.125674: step: 1160/529, loss: 0.018897246569395065 2023-01-21 10:38:39.239473: step: 1164/529, loss: 0.05194282531738281 2023-01-21 10:38:40.357370: step: 1168/529, loss: 0.029150735586881638 2023-01-21 10:38:41.468978: step: 1172/529, loss: 0.018677283078432083 2023-01-21 10:38:42.572365: step: 1176/529, loss: 0.07526731491088867 2023-01-21 10:38:43.673295: step: 1180/529, loss: 0.03196830675005913 2023-01-21 10:38:44.792741: step: 1184/529, loss: 0.055865198373794556 2023-01-21 10:38:45.907093: step: 1188/529, loss: 0.1328238546848297 2023-01-21 10:38:47.019236: step: 1192/529, loss: 0.04624681547284126 2023-01-21 10:38:48.148709: step: 1196/529, loss: 0.08423338085412979 2023-01-21 10:38:49.264638: step: 1200/529, loss: 0.026012137532234192 2023-01-21 10:38:50.394494: step: 1204/529, loss: 0.001549196313135326 2023-01-21 10:38:51.504851: step: 1208/529, loss: 0.023470783606171608 2023-01-21 10:38:52.651131: step: 1212/529, loss: 0.049822043627500534 2023-01-21 10:38:53.819486: step: 1216/529, loss: 0.10277938842773438 2023-01-21 10:38:54.965107: step: 1220/529, loss: 0.009597779251635075 2023-01-21 10:38:56.056931: step: 1224/529, loss: 0.01658325269818306 2023-01-21 10:38:57.198214: step: 1228/529, loss: 0.061242491006851196 2023-01-21 10:38:58.343982: step: 1232/529, loss: 0.36316853761672974 2023-01-21 10:38:59.456596: step: 1236/529, loss: 0.09373664855957031 2023-01-21 10:39:00.582020: step: 1240/529, loss: 0.00899038277566433 2023-01-21 10:39:01.669595: step: 1244/529, loss: 0.0021594048012048006 2023-01-21 10:39:02.779226: step: 1248/529, loss: 0.12025928497314453 2023-01-21 10:39:03.892989: step: 1252/529, loss: 0.03190011903643608 2023-01-21 10:39:04.999044: step: 1256/529, loss: 0.05552215874195099 2023-01-21 10:39:06.120174: step: 1260/529, loss: 0.031192127615213394 2023-01-21 10:39:07.263331: step: 1264/529, loss: 0.006292581558227539 2023-01-21 10:39:08.386339: step: 1268/529, loss: 0.0015749931335449219 2023-01-21 10:39:09.493361: step: 1272/529, loss: 0.06892938911914825 2023-01-21 10:39:10.601953: step: 1276/529, loss: 0.14832572638988495 2023-01-21 10:39:11.728043: step: 1280/529, loss: 0.028995418921113014 2023-01-21 10:39:12.833100: step: 1284/529, loss: 0.10269031673669815 2023-01-21 10:39:13.946111: step: 1288/529, loss: 0.01741924323141575 2023-01-21 10:39:15.097913: step: 1292/529, loss: 0.02010469324886799 2023-01-21 10:39:16.225426: step: 1296/529, loss: 0.02647099643945694 2023-01-21 10:39:17.332274: step: 1300/529, loss: 0.01633281819522381 2023-01-21 10:39:18.459551: step: 1304/529, loss: 0.014727115631103516 2023-01-21 10:39:19.570412: step: 1308/529, loss: 0.011113643646240234 2023-01-21 10:39:20.703356: step: 1312/529, loss: 0.016550255939364433 2023-01-21 10:39:21.815079: step: 1316/529, loss: 0.02597637102007866 2023-01-21 10:39:22.937932: step: 1320/529, loss: 0.012275981716811657 2023-01-21 10:39:24.064578: step: 1324/529, loss: 0.006790637969970703 2023-01-21 10:39:25.201294: step: 1328/529, loss: 0.047884371131658554 2023-01-21 10:39:26.357002: step: 1332/529, loss: 0.08185453712940216 2023-01-21 10:39:27.446991: step: 1336/529, loss: 0.041016869246959686 2023-01-21 10:39:28.582426: step: 1340/529, loss: 0.042281534522771835 2023-01-21 10:39:29.711612: step: 1344/529, loss: 0.05097236484289169 2023-01-21 10:39:30.826653: step: 1348/529, loss: 0.03887634351849556 2023-01-21 10:39:31.925445: step: 1352/529, loss: 0.08085336536169052 2023-01-21 10:39:33.012698: step: 1356/529, loss: 0.06968836486339569 2023-01-21 10:39:34.125013: step: 1360/529, loss: 0.018033599480986595 2023-01-21 10:39:35.240909: step: 1364/529, loss: 0.1166471540927887 2023-01-21 10:39:36.374864: step: 1368/529, loss: 0.022437667474150658 2023-01-21 10:39:37.502104: step: 1372/529, loss: 0.043592311441898346 2023-01-21 10:39:38.624939: step: 1376/529, loss: 0.044651225209236145 2023-01-21 10:39:39.747021: step: 1380/529, loss: 0.1370231658220291 2023-01-21 10:39:40.867612: step: 1384/529, loss: 0.03458767011761665 2023-01-21 10:39:41.977495: step: 1388/529, loss: 0.017298508435487747 2023-01-21 10:39:43.089814: step: 1392/529, loss: 0.00023469925508834422 2023-01-21 10:39:44.187466: step: 1396/529, loss: 0.0022085190284997225 2023-01-21 10:39:45.284991: step: 1400/529, loss: 0.06543579697608948 2023-01-21 10:39:46.390292: step: 1404/529, loss: 0.15252408385276794 2023-01-21 10:39:47.497724: step: 1408/529, loss: 0.02981758303940296 2023-01-21 10:39:48.610296: step: 1412/529, loss: 0.018331432715058327 2023-01-21 10:39:49.724743: step: 1416/529, loss: 0.05113248899579048 2023-01-21 10:39:50.844031: step: 1420/529, loss: 0.0463414192199707 2023-01-21 10:39:51.951425: step: 1424/529, loss: 0.02421570010483265 2023-01-21 10:39:53.085343: step: 1428/529, loss: 0.06854267418384552 2023-01-21 10:39:54.196255: step: 1432/529, loss: 0.07120855152606964 2023-01-21 10:39:55.325030: step: 1436/529, loss: 0.015619087032973766 2023-01-21 10:39:56.448875: step: 1440/529, loss: 0.07546329498291016 2023-01-21 10:39:57.585348: step: 1444/529, loss: 0.009140062145888805 2023-01-21 10:39:58.691483: step: 1448/529, loss: 0.01577281951904297 2023-01-21 10:39:59.774054: step: 1452/529, loss: 0.011563492007553577 2023-01-21 10:40:00.877431: step: 1456/529, loss: 0.033582307398319244 2023-01-21 10:40:01.973133: step: 1460/529, loss: 0.011561584658920765 2023-01-21 10:40:03.109811: step: 1464/529, loss: 0.035955097526311874 2023-01-21 10:40:04.236397: step: 1468/529, loss: 0.0008926392183639109 2023-01-21 10:40:05.381066: step: 1472/529, loss: 0.010932732373476028 2023-01-21 10:40:06.487237: step: 1476/529, loss: 0.026738930493593216 2023-01-21 10:40:07.645682: step: 1480/529, loss: 0.005289840511977673 2023-01-21 10:40:08.746262: step: 1484/529, loss: 1.2817879915237427 2023-01-21 10:40:09.903289: step: 1488/529, loss: 0.18614941835403442 2023-01-21 10:40:10.999048: step: 1492/529, loss: 0.020973587408661842 2023-01-21 10:40:12.070800: step: 1496/529, loss: 0.021105099469423294 2023-01-21 10:40:13.231054: step: 1500/529, loss: 0.04409465938806534 2023-01-21 10:40:14.341664: step: 1504/529, loss: 0.02840442582964897 2023-01-21 10:40:15.462834: step: 1508/529, loss: 0.4077225625514984 2023-01-21 10:40:16.580935: step: 1512/529, loss: 0.12635794281959534 2023-01-21 10:40:17.684326: step: 1516/529, loss: 0.028376102447509766 2023-01-21 10:40:18.818350: step: 1520/529, loss: 0.03752918168902397 2023-01-21 10:40:19.942493: step: 1524/529, loss: 0.03224983066320419 2023-01-21 10:40:21.052630: step: 1528/529, loss: 0.023200606927275658 2023-01-21 10:40:22.156752: step: 1532/529, loss: 0.0007997512584552169 2023-01-21 10:40:23.268994: step: 1536/529, loss: 0.0032644271850585938 2023-01-21 10:40:24.389815: step: 1540/529, loss: 0.04025211185216904 2023-01-21 10:40:25.493705: step: 1544/529, loss: 0.16581708192825317 2023-01-21 10:40:26.656555: step: 1548/529, loss: 0.04532432556152344 2023-01-21 10:40:27.791682: step: 1552/529, loss: 0.08706779778003693 2023-01-21 10:40:28.911958: step: 1556/529, loss: 0.04276624694466591 2023-01-21 10:40:30.005690: step: 1560/529, loss: 0.03359370306134224 2023-01-21 10:40:31.170471: step: 1564/529, loss: 0.028411865234375 2023-01-21 10:40:32.272911: step: 1568/529, loss: 0.03997201845049858 2023-01-21 10:40:33.390172: step: 1572/529, loss: 0.03255004808306694 2023-01-21 10:40:34.484801: step: 1576/529, loss: 0.0849708542227745 2023-01-21 10:40:35.624409: step: 1580/529, loss: 0.03191395103931427 2023-01-21 10:40:36.786900: step: 1584/529, loss: 0.08040819317102432 2023-01-21 10:40:37.876185: step: 1588/529, loss: 0.01630554348230362 2023-01-21 10:40:38.998738: step: 1592/529, loss: 0.02791604958474636 2023-01-21 10:40:40.129150: step: 1596/529, loss: 0.09858722239732742 2023-01-21 10:40:41.243807: step: 1600/529, loss: 0.06216542795300484 2023-01-21 10:40:42.398444: step: 1604/529, loss: 0.1791582703590393 2023-01-21 10:40:43.522277: step: 1608/529, loss: 0.09997282177209854 2023-01-21 10:40:44.616499: step: 1612/529, loss: 0.0009524345514364541 2023-01-21 10:40:45.717279: step: 1616/529, loss: 0.08954592049121857 2023-01-21 10:40:46.829961: step: 1620/529, loss: 0.012497663497924805 2023-01-21 10:40:47.945500: step: 1624/529, loss: 0.009469079785048962 2023-01-21 10:40:49.039200: step: 1628/529, loss: 0.018001150339841843 2023-01-21 10:40:50.184409: step: 1632/529, loss: 0.028762245550751686 2023-01-21 10:40:51.284705: step: 1636/529, loss: 0.08118200302124023 2023-01-21 10:40:52.415556: step: 1640/529, loss: 0.10868954658508301 2023-01-21 10:40:53.538932: step: 1644/529, loss: 0.0661388412117958 2023-01-21 10:40:54.652551: step: 1648/529, loss: 0.024476435035467148 2023-01-21 10:40:55.793598: step: 1652/529, loss: 0.02030305750668049 2023-01-21 10:40:56.896209: step: 1656/529, loss: 0.050966549664735794 2023-01-21 10:40:58.023956: step: 1660/529, loss: 0.2167210578918457 2023-01-21 10:40:59.133813: step: 1664/529, loss: 0.032126620411872864 2023-01-21 10:41:00.256035: step: 1668/529, loss: 0.01632622443139553 2023-01-21 10:41:01.410055: step: 1672/529, loss: 0.45482900738716125 2023-01-21 10:41:02.499950: step: 1676/529, loss: 0.016525840386748314 2023-01-21 10:41:03.663298: step: 1680/529, loss: 0.18198880553245544 2023-01-21 10:41:04.810521: step: 1684/529, loss: 0.02922077104449272 2023-01-21 10:41:05.931043: step: 1688/529, loss: 0.33754825592041016 2023-01-21 10:41:07.048307: step: 1692/529, loss: 0.01591520383954048 2023-01-21 10:41:08.161429: step: 1696/529, loss: 0.1771833449602127 2023-01-21 10:41:09.255551: step: 1700/529, loss: 0.0012199401389807463 2023-01-21 10:41:10.388046: step: 1704/529, loss: 0.030598830431699753 2023-01-21 10:41:11.480203: step: 1708/529, loss: 0.018317176029086113 2023-01-21 10:41:12.587714: step: 1712/529, loss: 0.008237361907958984 2023-01-21 10:41:13.693994: step: 1716/529, loss: 0.07449178397655487 2023-01-21 10:41:14.820590: step: 1720/529, loss: 0.13098879158496857 2023-01-21 10:41:15.927611: step: 1724/529, loss: 0.012220572680234909 2023-01-21 10:41:17.027817: step: 1728/529, loss: 0.061811067163944244 2023-01-21 10:41:18.153772: step: 1732/529, loss: 0.009442901238799095 2023-01-21 10:41:19.275171: step: 1736/529, loss: 0.003170651150867343 2023-01-21 10:41:20.419247: step: 1740/529, loss: 0.018052101135253906 2023-01-21 10:41:21.524467: step: 1744/529, loss: 0.043543148785829544 2023-01-21 10:41:22.635383: step: 1748/529, loss: 0.6302715539932251 2023-01-21 10:41:23.778086: step: 1752/529, loss: 0.29814016819000244 2023-01-21 10:41:24.898902: step: 1756/529, loss: 0.018412113189697266 2023-01-21 10:41:26.020564: step: 1760/529, loss: 0.04240760952234268 2023-01-21 10:41:27.184134: step: 1764/529, loss: 0.06515979766845703 2023-01-21 10:41:28.301576: step: 1768/529, loss: 0.14517177641391754 2023-01-21 10:41:29.425242: step: 1772/529, loss: 0.07652683556079865 2023-01-21 10:41:30.548620: step: 1776/529, loss: 0.05062122642993927 2023-01-21 10:41:31.660487: step: 1780/529, loss: 0.017912961542606354 2023-01-21 10:41:32.760480: step: 1784/529, loss: 0.03451428562402725 2023-01-21 10:41:33.903971: step: 1788/529, loss: 0.0849205031991005 2023-01-21 10:41:35.046335: step: 1792/529, loss: 0.00943670328706503 2023-01-21 10:41:36.170085: step: 1796/529, loss: 0.048021744936704636 2023-01-21 10:41:37.278049: step: 1800/529, loss: 0.07249584794044495 2023-01-21 10:41:38.412807: step: 1804/529, loss: 0.022319890558719635 2023-01-21 10:41:39.542444: step: 1808/529, loss: 0.027266694232821465 2023-01-21 10:41:40.626972: step: 1812/529, loss: 0.03726387396454811 2023-01-21 10:41:41.768950: step: 1816/529, loss: 0.1139443963766098 2023-01-21 10:41:42.993857: step: 1820/529, loss: 0.08703994750976562 2023-01-21 10:41:44.116978: step: 1824/529, loss: 0.015526198782026768 2023-01-21 10:41:45.234073: step: 1828/529, loss: 0.03237032890319824 2023-01-21 10:41:46.364348: step: 1832/529, loss: 0.009862994775176048 2023-01-21 10:41:47.495852: step: 1836/529, loss: 0.04492530971765518 2023-01-21 10:41:48.658049: step: 1840/529, loss: 0.027873801067471504 2023-01-21 10:41:49.784140: step: 1844/529, loss: 0.19703903794288635 2023-01-21 10:41:50.917920: step: 1848/529, loss: 0.013106727972626686 2023-01-21 10:41:52.050096: step: 1852/529, loss: 0.010895348154008389 2023-01-21 10:41:53.170513: step: 1856/529, loss: 0.01015615463256836 2023-01-21 10:41:54.304884: step: 1860/529, loss: 0.10346546024084091 2023-01-21 10:41:55.423109: step: 1864/529, loss: 0.43522942066192627 2023-01-21 10:41:56.553817: step: 1868/529, loss: 0.035448648035526276 2023-01-21 10:41:57.678996: step: 1872/529, loss: 0.007220649626106024 2023-01-21 10:41:58.787452: step: 1876/529, loss: 0.01151819247752428 2023-01-21 10:41:59.925617: step: 1880/529, loss: 0.07154957950115204 2023-01-21 10:42:01.041268: step: 1884/529, loss: 0.0020329237449914217 2023-01-21 10:42:02.186999: step: 1888/529, loss: 0.015955829992890358 2023-01-21 10:42:03.296920: step: 1892/529, loss: 0.1267586201429367 2023-01-21 10:42:04.387375: step: 1896/529, loss: 0.04730387032032013 2023-01-21 10:42:05.498165: step: 1900/529, loss: 0.1389884054660797 2023-01-21 10:42:06.623812: step: 1904/529, loss: 0.2638680338859558 2023-01-21 10:42:07.751961: step: 1908/529, loss: 0.0234375 2023-01-21 10:42:08.852359: step: 1912/529, loss: 0.6985609531402588 2023-01-21 10:42:09.944905: step: 1916/529, loss: 0.35749319195747375 2023-01-21 10:42:11.068792: step: 1920/529, loss: 0.0287201888859272 2023-01-21 10:42:12.179196: step: 1924/529, loss: 0.03194260597229004 2023-01-21 10:42:13.299672: step: 1928/529, loss: 0.02880726009607315 2023-01-21 10:42:14.404121: step: 1932/529, loss: 0.14732703566551208 2023-01-21 10:42:15.529639: step: 1936/529, loss: 0.033373452723026276 2023-01-21 10:42:16.652137: step: 1940/529, loss: 0.030338667333126068 2023-01-21 10:42:17.785569: step: 1944/529, loss: 0.07796420902013779 2023-01-21 10:42:18.878891: step: 1948/529, loss: 0.07340216636657715 2023-01-21 10:42:19.982451: step: 1952/529, loss: 6.369919300079346 2023-01-21 10:42:21.116620: step: 1956/529, loss: 0.1040363758802414 2023-01-21 10:42:22.238388: step: 1960/529, loss: 0.12936702370643616 2023-01-21 10:42:23.383732: step: 1964/529, loss: 0.037236977368593216 2023-01-21 10:42:24.500354: step: 1968/529, loss: 0.04321565479040146 2023-01-21 10:42:25.633796: step: 1972/529, loss: 0.009425735101103783 2023-01-21 10:42:26.778360: step: 1976/529, loss: 0.007780981250107288 2023-01-21 10:42:27.917410: step: 1980/529, loss: 0.06220731511712074 2023-01-21 10:42:29.043732: step: 1984/529, loss: 0.11651439964771271 2023-01-21 10:42:30.183749: step: 1988/529, loss: 0.013585568405687809 2023-01-21 10:42:31.291777: step: 1992/529, loss: 0.016615772619843483 2023-01-21 10:42:32.398707: step: 1996/529, loss: 0.011364936828613281 2023-01-21 10:42:33.509748: step: 2000/529, loss: 0.3767964243888855 2023-01-21 10:42:34.601916: step: 2004/529, loss: 0.0048768045380711555 2023-01-21 10:42:35.705156: step: 2008/529, loss: 0.011681205593049526 2023-01-21 10:42:36.786805: step: 2012/529, loss: 0.0638306587934494 2023-01-21 10:42:37.907056: step: 2016/529, loss: 0.02537841722369194 2023-01-21 10:42:39.016424: step: 2020/529, loss: 0.09910902380943298 2023-01-21 10:42:40.174921: step: 2024/529, loss: 0.0655437484383583 2023-01-21 10:42:41.276797: step: 2028/529, loss: 0.0441129207611084 2023-01-21 10:42:42.400007: step: 2032/529, loss: 0.031630899757146835 2023-01-21 10:42:43.517079: step: 2036/529, loss: 0.008608913980424404 2023-01-21 10:42:44.635062: step: 2040/529, loss: 0.00925140455365181 2023-01-21 10:42:45.776351: step: 2044/529, loss: 0.05186887085437775 2023-01-21 10:42:46.890770: step: 2048/529, loss: 0.0050865174271166325 2023-01-21 10:42:48.007324: step: 2052/529, loss: 0.04473453015089035 2023-01-21 10:42:49.104251: step: 2056/529, loss: 0.04050302505493164 2023-01-21 10:42:50.226000: step: 2060/529, loss: 0.023241043090820312 2023-01-21 10:42:51.340178: step: 2064/529, loss: 0.10674099624156952 2023-01-21 10:42:52.438054: step: 2068/529, loss: 0.04414501041173935 2023-01-21 10:42:53.554648: step: 2072/529, loss: 0.013210296630859375 2023-01-21 10:42:54.643684: step: 2076/529, loss: 0.018911074846982956 2023-01-21 10:42:55.732304: step: 2080/529, loss: 0.030109308660030365 2023-01-21 10:42:56.867053: step: 2084/529, loss: 0.015364646911621094 2023-01-21 10:42:57.985107: step: 2088/529, loss: 0.007328510750085115 2023-01-21 10:42:59.105084: step: 2092/529, loss: 0.040717221796512604 2023-01-21 10:43:00.231380: step: 2096/529, loss: 0.1877738982439041 2023-01-21 10:43:01.342201: step: 2100/529, loss: 0.04347119480371475 2023-01-21 10:43:02.434834: step: 2104/529, loss: 0.03549699857831001 2023-01-21 10:43:03.564265: step: 2108/529, loss: 0.16354884207248688 2023-01-21 10:43:04.685136: step: 2112/529, loss: 0.008365154266357422 2023-01-21 10:43:05.787503: step: 2116/529, loss: 0.06444015353918076 ================================================== Loss: 0.083 -------------------- Dev: {'event': {'p': 0.6060606060606061, 'r': 0.7723035952063915, 'f1': 0.6791569086651054}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Test: {'event': {'p': 0.6296123306865951, 'r': 0.7607223476297968, 'f1': 0.6889854331714796}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Chinese: {'event': {'p': 0.5434782608695652, 'r': 0.9259259259259259, 'f1': 0.6849315068493151}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Korean: {'event': {'p': 0.532258064516129, 'r': 0.5238095238095238, 'f1': 0.5280000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Russian: {'event': {'p': 0.5348837209302325, 'r': 0.6388888888888888, 'f1': 0.5822784810126581}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 13 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:43:46.869989: step: 4/529, loss: 0.07293281704187393 2023-01-21 10:43:47.994122: step: 8/529, loss: 0.05306410789489746 2023-01-21 10:43:49.127073: step: 12/529, loss: 0.0043087005615234375 2023-01-21 10:43:50.259276: step: 16/529, loss: 0.02919778972864151 2023-01-21 10:43:51.350191: step: 20/529, loss: 0.06782207638025284 2023-01-21 10:43:52.454072: step: 24/529, loss: 0.05834408104419708 2023-01-21 10:43:53.573343: step: 28/529, loss: 0.02410259284079075 2023-01-21 10:43:54.686750: step: 32/529, loss: 0.03040762059390545 2023-01-21 10:43:55.855515: step: 36/529, loss: 0.04593805968761444 2023-01-21 10:43:56.966975: step: 40/529, loss: 0.0033075332175940275 2023-01-21 10:43:58.104284: step: 44/529, loss: 0.1294124573469162 2023-01-21 10:43:59.236383: step: 48/529, loss: 0.014727211557328701 2023-01-21 10:44:00.355228: step: 52/529, loss: 0.03246564790606499 2023-01-21 10:44:01.462891: step: 56/529, loss: 0.04186735302209854 2023-01-21 10:44:02.601986: step: 60/529, loss: 0.11083802580833435 2023-01-21 10:44:03.723300: step: 64/529, loss: 0.1341545134782791 2023-01-21 10:44:04.823229: step: 68/529, loss: 0.03129563480615616 2023-01-21 10:44:05.926910: step: 72/529, loss: 0.026824189350008965 2023-01-21 10:44:07.043818: step: 76/529, loss: 0.004168987274169922 2023-01-21 10:44:08.149315: step: 80/529, loss: 0.010345744900405407 2023-01-21 10:44:09.254512: step: 84/529, loss: 0.08374261856079102 2023-01-21 10:44:10.375982: step: 88/529, loss: 0.0028049468528479338 2023-01-21 10:44:11.491895: step: 92/529, loss: 0.013827514834702015 2023-01-21 10:44:12.597196: step: 96/529, loss: 0.10413875430822372 2023-01-21 10:44:13.675999: step: 100/529, loss: 0.0017693996196612716 2023-01-21 10:44:14.791480: step: 104/529, loss: 0.030693912878632545 2023-01-21 10:44:15.926621: step: 108/529, loss: 0.09507226943969727 2023-01-21 10:44:17.042380: step: 112/529, loss: 0.003758668899536133 2023-01-21 10:44:18.162425: step: 116/529, loss: 0.04741577059030533 2023-01-21 10:44:19.276729: step: 120/529, loss: 0.0698552206158638 2023-01-21 10:44:20.410635: step: 124/529, loss: 0.019583702087402344 2023-01-21 10:44:21.549240: step: 128/529, loss: 0.0826694443821907 2023-01-21 10:44:22.671584: step: 132/529, loss: 0.0446566604077816 2023-01-21 10:44:23.772318: step: 136/529, loss: 0.02641153521835804 2023-01-21 10:44:24.875830: step: 140/529, loss: 0.015978194773197174 2023-01-21 10:44:26.030681: step: 144/529, loss: 0.016172980889678 2023-01-21 10:44:27.139730: step: 148/529, loss: 0.04405966028571129 2023-01-21 10:44:28.273519: step: 152/529, loss: 0.012769126333296299 2023-01-21 10:44:29.400385: step: 156/529, loss: 0.06906633824110031 2023-01-21 10:44:30.545002: step: 160/529, loss: 0.02988309971988201 2023-01-21 10:44:31.647528: step: 164/529, loss: 0.026522064581513405 2023-01-21 10:44:32.783554: step: 168/529, loss: 0.10043449699878693 2023-01-21 10:44:33.901783: step: 172/529, loss: 0.04777669906616211 2023-01-21 10:44:35.025704: step: 176/529, loss: 0.08264217525720596 2023-01-21 10:44:36.159179: step: 180/529, loss: 0.04310645908117294 2023-01-21 10:44:37.282893: step: 184/529, loss: 0.004085731692612171 2023-01-21 10:44:38.381869: step: 188/529, loss: 0.02329711988568306 2023-01-21 10:44:39.519263: step: 192/529, loss: 0.04551362991333008 2023-01-21 10:44:40.620375: step: 196/529, loss: 0.012001561932265759 2023-01-21 10:44:41.740780: step: 200/529, loss: 0.004966163542121649 2023-01-21 10:44:42.859137: step: 204/529, loss: 0.04300098493695259 2023-01-21 10:44:43.946174: step: 208/529, loss: 0.042791180312633514 2023-01-21 10:44:45.064378: step: 212/529, loss: 0.030836915597319603 2023-01-21 10:44:46.174095: step: 216/529, loss: 0.0011765003437176347 2023-01-21 10:44:47.260354: step: 220/529, loss: 0.007782173343002796 2023-01-21 10:44:48.381196: step: 224/529, loss: 0.0169556625187397 2023-01-21 10:44:49.502729: step: 228/529, loss: 0.006486034020781517 2023-01-21 10:44:50.597258: step: 232/529, loss: 0.053126052021980286 2023-01-21 10:44:51.730813: step: 236/529, loss: 0.08858838677406311 2023-01-21 10:44:52.839538: step: 240/529, loss: 0.03427934646606445 2023-01-21 10:44:53.964420: step: 244/529, loss: 0.11107286810874939 2023-01-21 10:44:55.087560: step: 248/529, loss: 0.012371874414384365 2023-01-21 10:44:56.202505: step: 252/529, loss: 0.15168094635009766 2023-01-21 10:44:57.307156: step: 256/529, loss: 0.002318096114322543 2023-01-21 10:44:58.392075: step: 260/529, loss: 0.0011807441478595138 2023-01-21 10:44:59.510034: step: 264/529, loss: 0.05001430585980415 2023-01-21 10:45:00.606953: step: 268/529, loss: 0.027407310903072357 2023-01-21 10:45:01.722589: step: 272/529, loss: 0.05107688903808594 2023-01-21 10:45:02.812532: step: 276/529, loss: 0.012657929211854935 2023-01-21 10:45:03.954077: step: 280/529, loss: 0.0623016357421875 2023-01-21 10:45:05.109727: step: 284/529, loss: 0.42091426253318787 2023-01-21 10:45:06.239229: step: 288/529, loss: 0.024303628131747246 2023-01-21 10:45:07.332628: step: 292/529, loss: 0.00238971714861691 2023-01-21 10:45:08.442370: step: 296/529, loss: 0.06544456630945206 2023-01-21 10:45:09.569161: step: 300/529, loss: 0.022367000579833984 2023-01-21 10:45:10.723302: step: 304/529, loss: 0.038046568632125854 2023-01-21 10:45:11.893158: step: 308/529, loss: 0.030861474573612213 2023-01-21 10:45:13.020626: step: 312/529, loss: 0.07363948971033096 2023-01-21 10:45:14.136177: step: 316/529, loss: 0.09577986598014832 2023-01-21 10:45:15.262654: step: 320/529, loss: 0.003931427374482155 2023-01-21 10:45:16.371917: step: 324/529, loss: 0.007610988803207874 2023-01-21 10:45:17.499343: step: 328/529, loss: 0.018694495782256126 2023-01-21 10:45:18.621500: step: 332/529, loss: 0.1604211926460266 2023-01-21 10:45:19.734723: step: 336/529, loss: 0.03756294399499893 2023-01-21 10:45:20.832160: step: 340/529, loss: 0.0027392387855798006 2023-01-21 10:45:21.963011: step: 344/529, loss: 0.022695541381835938 2023-01-21 10:45:23.073358: step: 348/529, loss: 0.05300109460949898 2023-01-21 10:45:24.171451: step: 352/529, loss: 0.0515655055642128 2023-01-21 10:45:25.288235: step: 356/529, loss: 0.06699886173009872 2023-01-21 10:45:26.407966: step: 360/529, loss: 0.02273263968527317 2023-01-21 10:45:27.503557: step: 364/529, loss: 0.009206676855683327 2023-01-21 10:45:28.620678: step: 368/529, loss: 0.022063683718442917 2023-01-21 10:45:29.757794: step: 372/529, loss: 0.021879959851503372 2023-01-21 10:45:30.860150: step: 376/529, loss: 0.006477165035903454 2023-01-21 10:45:31.998893: step: 380/529, loss: 0.029498625546693802 2023-01-21 10:45:33.102730: step: 384/529, loss: 0.04163990169763565 2023-01-21 10:45:34.213653: step: 388/529, loss: 0.05253582447767258 2023-01-21 10:45:35.360454: step: 392/529, loss: 0.03861827775835991 2023-01-21 10:45:36.472331: step: 396/529, loss: 0.012377738952636719 2023-01-21 10:45:37.564487: step: 400/529, loss: 0.03954572603106499 2023-01-21 10:45:38.665404: step: 404/529, loss: 0.02371964603662491 2023-01-21 10:45:39.804789: step: 408/529, loss: 0.0023258209694176912 2023-01-21 10:45:40.906008: step: 412/529, loss: 0.022819137200713158 2023-01-21 10:45:42.026438: step: 416/529, loss: 0.19572706520557404 2023-01-21 10:45:43.154585: step: 420/529, loss: 0.03498144447803497 2023-01-21 10:45:44.272989: step: 424/529, loss: 0.09448914974927902 2023-01-21 10:45:45.389072: step: 428/529, loss: 0.07198276370763779 2023-01-21 10:45:46.510360: step: 432/529, loss: 0.07265033572912216 2023-01-21 10:45:47.598668: step: 436/529, loss: 0.010719014331698418 2023-01-21 10:45:48.752249: step: 440/529, loss: 0.002009487245231867 2023-01-21 10:45:49.882005: step: 444/529, loss: 0.04184990003705025 2023-01-21 10:45:51.007588: step: 448/529, loss: 0.04704742506146431 2023-01-21 10:45:52.127827: step: 452/529, loss: 0.014574432745575905 2023-01-21 10:45:53.257431: step: 456/529, loss: 0.02465210109949112 2023-01-21 10:45:54.395831: step: 460/529, loss: 0.00980224646627903 2023-01-21 10:45:55.507910: step: 464/529, loss: 0.0031983377411961555 2023-01-21 10:45:56.652394: step: 468/529, loss: 0.0863095298409462 2023-01-21 10:45:57.767463: step: 472/529, loss: 0.025953197851777077 2023-01-21 10:45:58.890653: step: 476/529, loss: 0.008178329095244408 2023-01-21 10:46:00.031815: step: 480/529, loss: 0.021496010944247246 2023-01-21 10:46:01.148369: step: 484/529, loss: 0.06223573535680771 2023-01-21 10:46:02.246536: step: 488/529, loss: 0.15695570409297943 2023-01-21 10:46:03.370140: step: 492/529, loss: 0.024889133870601654 2023-01-21 10:46:04.474069: step: 496/529, loss: 0.01629314385354519 2023-01-21 10:46:05.590855: step: 500/529, loss: 0.029558181762695312 2023-01-21 10:46:06.728061: step: 504/529, loss: 0.009670638479292393 2023-01-21 10:46:07.878109: step: 508/529, loss: 0.031327057629823685 2023-01-21 10:46:08.984515: step: 512/529, loss: 0.31777894496917725 2023-01-21 10:46:10.093585: step: 516/529, loss: 0.01933746412396431 2023-01-21 10:46:11.220254: step: 520/529, loss: 0.03571300581097603 2023-01-21 10:46:12.338281: step: 524/529, loss: 0.05715503916144371 2023-01-21 10:46:13.429076: step: 528/529, loss: 0.007799291983246803 2023-01-21 10:46:14.581069: step: 532/529, loss: 0.1013161689043045 2023-01-21 10:46:15.686916: step: 536/529, loss: 0.02707824856042862 2023-01-21 10:46:16.816145: step: 540/529, loss: 0.04957713931798935 2023-01-21 10:46:17.910742: step: 544/529, loss: 0.0015176773304119706 2023-01-21 10:46:19.005592: step: 548/529, loss: 0.11670036613941193 2023-01-21 10:46:20.112005: step: 552/529, loss: 0.053476713597774506 2023-01-21 10:46:21.220792: step: 556/529, loss: 0.002906894776970148 2023-01-21 10:46:22.320644: step: 560/529, loss: 0.017011260613799095 2023-01-21 10:46:23.430956: step: 564/529, loss: 0.023801662027835846 2023-01-21 10:46:24.524906: step: 568/529, loss: 0.04368725046515465 2023-01-21 10:46:25.635025: step: 572/529, loss: 0.015525627881288528 2023-01-21 10:46:26.773521: step: 576/529, loss: 0.030176306143403053 2023-01-21 10:46:27.892733: step: 580/529, loss: 0.030388785526156425 2023-01-21 10:46:28.985967: step: 584/529, loss: 0.03607215732336044 2023-01-21 10:46:30.093623: step: 588/529, loss: 0.05016031116247177 2023-01-21 10:46:31.197516: step: 592/529, loss: 0.00012159347534179688 2023-01-21 10:46:32.292760: step: 596/529, loss: 0.020132923498749733 2023-01-21 10:46:33.403919: step: 600/529, loss: 0.017664004117250443 2023-01-21 10:46:34.539268: step: 604/529, loss: 0.0875476822257042 2023-01-21 10:46:35.641431: step: 608/529, loss: 0.008771324530243874 2023-01-21 10:46:36.757920: step: 612/529, loss: 0.018830489367246628 2023-01-21 10:46:37.896191: step: 616/529, loss: 0.07054519653320312 2023-01-21 10:46:39.048955: step: 620/529, loss: 0.01978626288473606 2023-01-21 10:46:40.172128: step: 624/529, loss: 0.011927127838134766 2023-01-21 10:46:41.332965: step: 628/529, loss: 0.06222324073314667 2023-01-21 10:46:42.483671: step: 632/529, loss: 0.022385025396943092 2023-01-21 10:46:43.592390: step: 636/529, loss: 0.011860180646181107 2023-01-21 10:46:44.729397: step: 640/529, loss: 0.05222740024328232 2023-01-21 10:46:45.837244: step: 644/529, loss: 0.03897275775671005 2023-01-21 10:46:46.984010: step: 648/529, loss: 0.11284789443016052 2023-01-21 10:46:48.087993: step: 652/529, loss: 0.006290435791015625 2023-01-21 10:46:49.214819: step: 656/529, loss: 0.02301778830587864 2023-01-21 10:46:50.395547: step: 660/529, loss: 0.07369986176490784 2023-01-21 10:46:51.500706: step: 664/529, loss: 0.013823938556015491 2023-01-21 10:46:52.617638: step: 668/529, loss: 0.03308677673339844 2023-01-21 10:46:53.737809: step: 672/529, loss: 0.09483185410499573 2023-01-21 10:46:54.858825: step: 676/529, loss: 0.0015774727798998356 2023-01-21 10:46:56.016078: step: 680/529, loss: 0.02214374579489231 2023-01-21 10:46:57.156294: step: 684/529, loss: 0.035392455756664276 2023-01-21 10:46:58.266363: step: 688/529, loss: 0.030090808868408203 2023-01-21 10:46:59.401723: step: 692/529, loss: 0.005709934514015913 2023-01-21 10:47:00.533345: step: 696/529, loss: 0.06359271705150604 2023-01-21 10:47:01.652283: step: 700/529, loss: 0.043273165822029114 2023-01-21 10:47:02.768720: step: 704/529, loss: 0.07162122428417206 2023-01-21 10:47:03.882544: step: 708/529, loss: 0.033919334411621094 2023-01-21 10:47:05.021468: step: 712/529, loss: 0.024324918165802956 2023-01-21 10:47:06.165408: step: 716/529, loss: 0.011709785088896751 2023-01-21 10:47:07.297820: step: 720/529, loss: 0.05855837091803551 2023-01-21 10:47:08.442029: step: 724/529, loss: 0.04289679229259491 2023-01-21 10:47:09.542810: step: 728/529, loss: 0.01994934119284153 2023-01-21 10:47:10.659917: step: 732/529, loss: 0.02031421661376953 2023-01-21 10:47:11.777237: step: 736/529, loss: 0.0349552184343338 2023-01-21 10:47:12.881047: step: 740/529, loss: 0.01612529717385769 2023-01-21 10:47:13.978658: step: 744/529, loss: 0.019127940759062767 2023-01-21 10:47:15.163460: step: 748/529, loss: 0.02261028252542019 2023-01-21 10:47:16.269891: step: 752/529, loss: 0.011346054263412952 2023-01-21 10:47:17.394335: step: 756/529, loss: 0.004826259799301624 2023-01-21 10:47:18.517374: step: 760/529, loss: 0.03502840921282768 2023-01-21 10:47:19.642117: step: 764/529, loss: 0.01723918877542019 2023-01-21 10:47:20.764542: step: 768/529, loss: 0.10384960472583771 2023-01-21 10:47:21.902225: step: 772/529, loss: 0.050301648676395416 2023-01-21 10:47:23.015617: step: 776/529, loss: 0.020436907187104225 2023-01-21 10:47:24.122160: step: 780/529, loss: 0.012467408552765846 2023-01-21 10:47:25.249183: step: 784/529, loss: 0.08142328262329102 2023-01-21 10:47:26.372343: step: 788/529, loss: 0.09496984630823135 2023-01-21 10:47:27.486648: step: 792/529, loss: 0.02488865703344345 2023-01-21 10:47:28.600814: step: 796/529, loss: 0.07901573181152344 2023-01-21 10:47:29.713378: step: 800/529, loss: 0.014664888381958008 2023-01-21 10:47:30.835020: step: 804/529, loss: 0.045256901532411575 2023-01-21 10:47:31.967988: step: 808/529, loss: 0.023755645379424095 2023-01-21 10:47:33.094317: step: 812/529, loss: 0.016444873064756393 2023-01-21 10:47:34.199011: step: 816/529, loss: 1.7624906301498413 2023-01-21 10:47:35.291502: step: 820/529, loss: 0.002922630403190851 2023-01-21 10:47:36.391225: step: 824/529, loss: 0.010241365991532803 2023-01-21 10:47:37.542314: step: 828/529, loss: 0.034173011779785156 2023-01-21 10:47:38.651239: step: 832/529, loss: 0.008757353760302067 2023-01-21 10:47:39.793046: step: 836/529, loss: 0.07103338092565536 2023-01-21 10:47:40.906701: step: 840/529, loss: 0.05490932613611221 2023-01-21 10:47:41.998215: step: 844/529, loss: 0.022204017266631126 2023-01-21 10:47:43.096915: step: 848/529, loss: 0.011773490346968174 2023-01-21 10:47:44.221120: step: 852/529, loss: 0.013375520706176758 2023-01-21 10:47:45.346586: step: 856/529, loss: 0.002599048661068082 2023-01-21 10:47:46.445702: step: 860/529, loss: 0.03408966213464737 2023-01-21 10:47:47.552142: step: 864/529, loss: 0.12001418322324753 2023-01-21 10:47:48.650515: step: 868/529, loss: 0.14384213089942932 2023-01-21 10:47:49.776572: step: 872/529, loss: 0.014993762597441673 2023-01-21 10:47:50.914876: step: 876/529, loss: 0.0626152977347374 2023-01-21 10:47:52.023517: step: 880/529, loss: 0.0652843490242958 2023-01-21 10:47:53.146635: step: 884/529, loss: 0.10830879211425781 2023-01-21 10:47:54.232506: step: 888/529, loss: 0.13697996735572815 2023-01-21 10:47:55.364085: step: 892/529, loss: 0.3752841353416443 2023-01-21 10:47:56.506404: step: 896/529, loss: 0.05899696797132492 2023-01-21 10:47:57.637107: step: 900/529, loss: 0.03325328975915909 2023-01-21 10:47:58.771425: step: 904/529, loss: 0.11066999286413193 2023-01-21 10:47:59.915738: step: 908/529, loss: 0.060028839856386185 2023-01-21 10:48:01.013960: step: 912/529, loss: 0.01938605308532715 2023-01-21 10:48:02.128696: step: 916/529, loss: 0.0058522941544651985 2023-01-21 10:48:03.243308: step: 920/529, loss: 0.014265060424804688 2023-01-21 10:48:04.347530: step: 924/529, loss: 0.017250824719667435 2023-01-21 10:48:05.464744: step: 928/529, loss: 0.032796718180179596 2023-01-21 10:48:06.556213: step: 932/529, loss: 0.04005365073680878 2023-01-21 10:48:07.705811: step: 936/529, loss: 0.04614553228020668 2023-01-21 10:48:08.792002: step: 940/529, loss: 0.14798007905483246 2023-01-21 10:48:09.876130: step: 944/529, loss: 0.036196134984493256 2023-01-21 10:48:10.990727: step: 948/529, loss: 0.029706383123993874 2023-01-21 10:48:12.096467: step: 952/529, loss: 0.04193758964538574 2023-01-21 10:48:13.222062: step: 956/529, loss: 0.021543120965361595 2023-01-21 10:48:14.338932: step: 960/529, loss: 0.01782970502972603 2023-01-21 10:48:15.504819: step: 964/529, loss: 0.03339047729969025 2023-01-21 10:48:16.597611: step: 968/529, loss: 0.341751366853714 2023-01-21 10:48:17.710020: step: 972/529, loss: 0.02943885326385498 2023-01-21 10:48:18.827454: step: 976/529, loss: 0.006870842073112726 2023-01-21 10:48:19.953443: step: 980/529, loss: 0.0309968963265419 2023-01-21 10:48:21.064574: step: 984/529, loss: 0.25331854820251465 2023-01-21 10:48:22.187905: step: 988/529, loss: 0.03831310197710991 2023-01-21 10:48:23.285692: step: 992/529, loss: 0.0008615494007244706 2023-01-21 10:48:24.371423: step: 996/529, loss: 0.006113243289291859 2023-01-21 10:48:25.509512: step: 1000/529, loss: 0.7010480165481567 2023-01-21 10:48:26.638618: step: 1004/529, loss: 0.11332845687866211 2023-01-21 10:48:27.726465: step: 1008/529, loss: 0.05447997897863388 2023-01-21 10:48:28.848825: step: 1012/529, loss: 0.007079887203872204 2023-01-21 10:48:29.946612: step: 1016/529, loss: 0.009257650002837181 2023-01-21 10:48:31.113850: step: 1020/529, loss: 0.19070139527320862 2023-01-21 10:48:32.220411: step: 1024/529, loss: 0.01338118314743042 2023-01-21 10:48:33.323119: step: 1028/529, loss: 0.001272869179956615 2023-01-21 10:48:34.442452: step: 1032/529, loss: 0.06684227287769318 2023-01-21 10:48:35.573738: step: 1036/529, loss: 0.01168808899819851 2023-01-21 10:48:36.689866: step: 1040/529, loss: 0.32215794920921326 2023-01-21 10:48:37.806218: step: 1044/529, loss: 0.0520719513297081 2023-01-21 10:48:38.916915: step: 1048/529, loss: 0.12193412333726883 2023-01-21 10:48:40.032292: step: 1052/529, loss: 0.06630926579236984 2023-01-21 10:48:41.183544: step: 1056/529, loss: 0.011602640151977539 2023-01-21 10:48:42.300019: step: 1060/529, loss: 0.05215277522802353 2023-01-21 10:48:43.418552: step: 1064/529, loss: 0.018575571477413177 2023-01-21 10:48:44.552088: step: 1068/529, loss: 0.004266071133315563 2023-01-21 10:48:45.642106: step: 1072/529, loss: 0.08493126928806305 2023-01-21 10:48:46.776984: step: 1076/529, loss: 0.035504911094903946 2023-01-21 10:48:47.886746: step: 1080/529, loss: 0.03413458168506622 2023-01-21 10:48:48.994334: step: 1084/529, loss: 0.055048421025276184 2023-01-21 10:48:50.102559: step: 1088/529, loss: 0.06059369817376137 2023-01-21 10:48:51.224652: step: 1092/529, loss: 0.004771423526108265 2023-01-21 10:48:52.330472: step: 1096/529, loss: 0.0011592864757403731 2023-01-21 10:48:53.433100: step: 1100/529, loss: 0.49570825695991516 2023-01-21 10:48:54.563855: step: 1104/529, loss: 0.09929446876049042 2023-01-21 10:48:55.705116: step: 1108/529, loss: 0.022623255848884583 2023-01-21 10:48:56.848448: step: 1112/529, loss: 0.2396383285522461 2023-01-21 10:48:57.996408: step: 1116/529, loss: 0.0005959034315310419 2023-01-21 10:48:59.073245: step: 1120/529, loss: 0.013457298278808594 2023-01-21 10:49:00.193751: step: 1124/529, loss: 0.06752882152795792 2023-01-21 10:49:01.312073: step: 1128/529, loss: 0.006843948271125555 2023-01-21 10:49:02.448510: step: 1132/529, loss: 0.10009326785802841 2023-01-21 10:49:03.567642: step: 1136/529, loss: 0.02182168886065483 2023-01-21 10:49:04.675876: step: 1140/529, loss: 0.00824661273509264 2023-01-21 10:49:05.829318: step: 1144/529, loss: 0.08958091586828232 2023-01-21 10:49:06.943341: step: 1148/529, loss: 0.09390156716108322 2023-01-21 10:49:08.069897: step: 1152/529, loss: 0.06819095462560654 2023-01-21 10:49:09.177601: step: 1156/529, loss: 0.2882021963596344 2023-01-21 10:49:10.329048: step: 1160/529, loss: 0.04466400295495987 2023-01-21 10:49:11.429680: step: 1164/529, loss: 0.1249622330069542 2023-01-21 10:49:12.553493: step: 1168/529, loss: 0.01870594173669815 2023-01-21 10:49:13.666485: step: 1172/529, loss: 0.10197696834802628 2023-01-21 10:49:14.773732: step: 1176/529, loss: 0.07632827758789062 2023-01-21 10:49:15.887704: step: 1180/529, loss: 0.013493537902832031 2023-01-21 10:49:16.980300: step: 1184/529, loss: 0.04766673967242241 2023-01-21 10:49:18.097370: step: 1188/529, loss: 0.041199587285518646 2023-01-21 10:49:19.206356: step: 1192/529, loss: 0.06494684517383575 2023-01-21 10:49:20.352811: step: 1196/529, loss: 0.0155525216832757 2023-01-21 10:49:21.466462: step: 1200/529, loss: 0.012423325330018997 2023-01-21 10:49:22.572485: step: 1204/529, loss: 0.0001695632963674143 2023-01-21 10:49:23.688886: step: 1208/529, loss: 0.07731194794178009 2023-01-21 10:49:24.817541: step: 1212/529, loss: 0.017555713653564453 2023-01-21 10:49:25.915053: step: 1216/529, loss: 0.00043268201989121735 2023-01-21 10:49:27.040431: step: 1220/529, loss: 0.030835531651973724 2023-01-21 10:49:28.171843: step: 1224/529, loss: 0.2707158923149109 2023-01-21 10:49:29.299978: step: 1228/529, loss: 0.025734711438417435 2023-01-21 10:49:30.409376: step: 1232/529, loss: 0.0034225464332848787 2023-01-21 10:49:31.531026: step: 1236/529, loss: 0.004409885499626398 2023-01-21 10:49:32.630642: step: 1240/529, loss: 0.1639002412557602 2023-01-21 10:49:33.760095: step: 1244/529, loss: 0.11163682490587234 2023-01-21 10:49:34.888033: step: 1248/529, loss: 0.004207706544548273 2023-01-21 10:49:35.988184: step: 1252/529, loss: 0.0468423031270504 2023-01-21 10:49:37.100427: step: 1256/529, loss: 0.026732921600341797 2023-01-21 10:49:38.210477: step: 1260/529, loss: 0.0011005401611328125 2023-01-21 10:49:39.340607: step: 1264/529, loss: 0.0655902847647667 2023-01-21 10:49:40.435638: step: 1268/529, loss: 0.04583730548620224 2023-01-21 10:49:41.525658: step: 1272/529, loss: 0.008058547973632812 2023-01-21 10:49:42.607273: step: 1276/529, loss: 0.001617431640625 2023-01-21 10:49:43.737996: step: 1280/529, loss: 0.1286485642194748 2023-01-21 10:49:44.850726: step: 1284/529, loss: 0.37244758009910583 2023-01-21 10:49:45.983348: step: 1288/529, loss: 0.058251574635505676 2023-01-21 10:49:47.126013: step: 1292/529, loss: 0.012123584747314453 2023-01-21 10:49:48.233617: step: 1296/529, loss: 0.041721489280462265 2023-01-21 10:49:49.358200: step: 1300/529, loss: 0.0674106627702713 2023-01-21 10:49:50.503116: step: 1304/529, loss: 0.056020356714725494 2023-01-21 10:49:51.624857: step: 1308/529, loss: 0.029988478869199753 2023-01-21 10:49:52.738362: step: 1312/529, loss: 0.013033008202910423 2023-01-21 10:49:53.871373: step: 1316/529, loss: 0.049913980066776276 2023-01-21 10:49:54.973777: step: 1320/529, loss: 0.03180895000696182 2023-01-21 10:49:56.059476: step: 1324/529, loss: 0.032746508717536926 2023-01-21 10:49:57.188467: step: 1328/529, loss: 0.06756601482629776 2023-01-21 10:49:58.295218: step: 1332/529, loss: 0.028934193775057793 2023-01-21 10:49:59.390014: step: 1336/529, loss: 0.0183321014046669 2023-01-21 10:50:00.526391: step: 1340/529, loss: 0.025823401287198067 2023-01-21 10:50:01.669412: step: 1344/529, loss: 0.013457298278808594 2023-01-21 10:50:02.784728: step: 1348/529, loss: 0.11250286549329758 2023-01-21 10:50:03.906775: step: 1352/529, loss: 0.031952571123838425 2023-01-21 10:50:05.037079: step: 1356/529, loss: 0.00920867919921875 2023-01-21 10:50:06.158976: step: 1360/529, loss: 0.01450414676219225 2023-01-21 10:50:07.267941: step: 1364/529, loss: 0.11879067122936249 2023-01-21 10:50:08.399462: step: 1368/529, loss: 0.020023442804813385 2023-01-21 10:50:09.520165: step: 1372/529, loss: 0.273714542388916 2023-01-21 10:50:10.628458: step: 1376/529, loss: 0.04705781862139702 2023-01-21 10:50:11.745706: step: 1380/529, loss: 0.03726167976856232 2023-01-21 10:50:12.838688: step: 1384/529, loss: 0.024841928854584694 2023-01-21 10:50:13.974032: step: 1388/529, loss: 0.008778381161391735 2023-01-21 10:50:15.083721: step: 1392/529, loss: 0.028741836547851562 2023-01-21 10:50:16.221401: step: 1396/529, loss: 0.04222536087036133 2023-01-21 10:50:17.321034: step: 1400/529, loss: 0.003799915313720703 2023-01-21 10:50:18.434181: step: 1404/529, loss: 0.08891496807336807 2023-01-21 10:50:19.529044: step: 1408/529, loss: 0.01604478433728218 2023-01-21 10:50:20.679338: step: 1412/529, loss: 0.14737826585769653 2023-01-21 10:50:21.810407: step: 1416/529, loss: 0.118444062769413 2023-01-21 10:50:22.907486: step: 1420/529, loss: 0.07689027488231659 2023-01-21 10:50:24.038683: step: 1424/529, loss: 0.08351021260023117 2023-01-21 10:50:25.162524: step: 1428/529, loss: 0.7681671380996704 2023-01-21 10:50:26.282867: step: 1432/529, loss: 0.05300083011388779 2023-01-21 10:50:27.379156: step: 1436/529, loss: 0.015417384915053844 2023-01-21 10:50:28.476093: step: 1440/529, loss: 0.08357041329145432 2023-01-21 10:50:29.592140: step: 1444/529, loss: 0.060283709317445755 2023-01-21 10:50:30.703110: step: 1448/529, loss: 0.07480278611183167 2023-01-21 10:50:31.816743: step: 1452/529, loss: 0.0731332004070282 2023-01-21 10:50:32.910286: step: 1456/529, loss: 0.05331382900476456 2023-01-21 10:50:34.038011: step: 1460/529, loss: 0.6107146739959717 2023-01-21 10:50:35.151900: step: 1464/529, loss: 0.02226409874856472 2023-01-21 10:50:36.259226: step: 1468/529, loss: 0.0247986800968647 2023-01-21 10:50:37.406980: step: 1472/529, loss: 0.03485984727740288 2023-01-21 10:50:38.530755: step: 1476/529, loss: 0.07304602861404419 2023-01-21 10:50:39.625073: step: 1480/529, loss: 0.08118553459644318 2023-01-21 10:50:40.749861: step: 1484/529, loss: 0.050348568707704544 2023-01-21 10:50:41.854459: step: 1488/529, loss: 0.03297467157244682 2023-01-21 10:50:42.971364: step: 1492/529, loss: 0.045783109962940216 2023-01-21 10:50:44.061896: step: 1496/529, loss: 0.047002602368593216 2023-01-21 10:50:45.182849: step: 1500/529, loss: 0.006312465760856867 2023-01-21 10:50:46.300716: step: 1504/529, loss: 0.010353517718613148 2023-01-21 10:50:47.411858: step: 1508/529, loss: 0.0301087386906147 2023-01-21 10:50:48.526461: step: 1512/529, loss: 0.011759853921830654 2023-01-21 10:50:49.677784: step: 1516/529, loss: 0.015729617327451706 2023-01-21 10:50:50.828261: step: 1520/529, loss: 0.013049221597611904 2023-01-21 10:50:51.952217: step: 1524/529, loss: 0.034577276557683945 2023-01-21 10:50:53.112566: step: 1528/529, loss: 0.01103891059756279 2023-01-21 10:50:54.223146: step: 1532/529, loss: 0.014093780890107155 2023-01-21 10:50:55.340541: step: 1536/529, loss: 0.020003415644168854 2023-01-21 10:50:56.437007: step: 1540/529, loss: 0.02118401601910591 2023-01-21 10:50:57.554376: step: 1544/529, loss: 0.024834156036376953 2023-01-21 10:50:58.709717: step: 1548/529, loss: 0.02395954169332981 2023-01-21 10:50:59.810654: step: 1552/529, loss: 0.03550539165735245 2023-01-21 10:51:00.952917: step: 1556/529, loss: 0.019602488726377487 2023-01-21 10:51:02.085000: step: 1560/529, loss: 0.0151703841984272 2023-01-21 10:51:03.212640: step: 1564/529, loss: 0.03718709945678711 2023-01-21 10:51:04.337341: step: 1568/529, loss: 0.04355449602007866 2023-01-21 10:51:05.465248: step: 1572/529, loss: 0.0421999916434288 2023-01-21 10:51:06.600604: step: 1576/529, loss: 0.03545551374554634 2023-01-21 10:51:07.718040: step: 1580/529, loss: 0.030100345611572266 2023-01-21 10:51:08.854821: step: 1584/529, loss: 0.01910381391644478 2023-01-21 10:51:09.968933: step: 1588/529, loss: 0.02902050130069256 2023-01-21 10:51:11.092421: step: 1592/529, loss: 0.06081266701221466 2023-01-21 10:51:12.200886: step: 1596/529, loss: 0.004420375917106867 2023-01-21 10:51:13.301984: step: 1600/529, loss: 0.014257130213081837 2023-01-21 10:51:14.408378: step: 1604/529, loss: 0.04803962633013725 2023-01-21 10:51:15.530933: step: 1608/529, loss: 0.04708156734704971 2023-01-21 10:51:16.693846: step: 1612/529, loss: 0.01819610595703125 2023-01-21 10:51:17.816866: step: 1616/529, loss: 0.0681610107421875 2023-01-21 10:51:18.944650: step: 1620/529, loss: 0.04678454250097275 2023-01-21 10:51:20.069502: step: 1624/529, loss: 0.06651964783668518 2023-01-21 10:51:21.201803: step: 1628/529, loss: 0.020511532202363014 2023-01-21 10:51:22.307016: step: 1632/529, loss: 0.08219242095947266 2023-01-21 10:51:23.427959: step: 1636/529, loss: 0.03554592281579971 2023-01-21 10:51:24.519723: step: 1640/529, loss: 0.01932506635785103 2023-01-21 10:51:25.650071: step: 1644/529, loss: 0.008379315957427025 2023-01-21 10:51:26.770736: step: 1648/529, loss: 0.010648544877767563 2023-01-21 10:51:27.860979: step: 1652/529, loss: 0.03567094728350639 2023-01-21 10:51:28.993213: step: 1656/529, loss: 0.043998800218105316 2023-01-21 10:51:30.153703: step: 1660/529, loss: 0.07056055217981339 2023-01-21 10:51:31.274845: step: 1664/529, loss: 0.010371970944106579 2023-01-21 10:51:32.425595: step: 1668/529, loss: 0.03254912048578262 2023-01-21 10:51:33.510806: step: 1672/529, loss: 0.028117896988987923 2023-01-21 10:51:34.608699: step: 1676/529, loss: 0.11690196394920349 2023-01-21 10:51:35.767241: step: 1680/529, loss: 0.0931786522269249 2023-01-21 10:51:36.856033: step: 1684/529, loss: 0.006766319274902344 2023-01-21 10:51:37.976881: step: 1688/529, loss: 0.07551456242799759 2023-01-21 10:51:39.076957: step: 1692/529, loss: 0.17780761420726776 2023-01-21 10:51:40.214270: step: 1696/529, loss: 0.060100436210632324 2023-01-21 10:51:41.354523: step: 1700/529, loss: 0.024545002728700638 2023-01-21 10:51:42.469627: step: 1704/529, loss: 0.0704471617937088 2023-01-21 10:51:43.582621: step: 1708/529, loss: 0.023421861231327057 2023-01-21 10:51:44.674806: step: 1712/529, loss: 0.013410281389951706 2023-01-21 10:51:45.765641: step: 1716/529, loss: 0.05588135868310928 2023-01-21 10:51:46.875591: step: 1720/529, loss: 0.007395029067993164 2023-01-21 10:51:48.005851: step: 1724/529, loss: 0.1294885128736496 2023-01-21 10:51:49.147401: step: 1728/529, loss: 0.04471225664019585 2023-01-21 10:51:50.282393: step: 1732/529, loss: 0.05116519704461098 2023-01-21 10:51:51.423360: step: 1736/529, loss: 0.08819885551929474 2023-01-21 10:51:52.528314: step: 1740/529, loss: 0.17374782264232635 2023-01-21 10:51:53.615879: step: 1744/529, loss: 0.0027803422417491674 2023-01-21 10:51:54.738563: step: 1748/529, loss: 0.002521610353142023 2023-01-21 10:51:55.860081: step: 1752/529, loss: 0.0056362152099609375 2023-01-21 10:51:56.984203: step: 1756/529, loss: 0.01562805101275444 2023-01-21 10:51:58.087410: step: 1760/529, loss: 0.04769149050116539 2023-01-21 10:51:59.208412: step: 1764/529, loss: 0.012091398239135742 2023-01-21 10:52:00.307458: step: 1768/529, loss: 0.015734482556581497 2023-01-21 10:52:01.446431: step: 1772/529, loss: 0.07509279996156693 2023-01-21 10:52:02.563721: step: 1776/529, loss: 0.1594143956899643 2023-01-21 10:52:03.720812: step: 1780/529, loss: 0.013957023620605469 2023-01-21 10:52:04.824735: step: 1784/529, loss: 5.817413330078125e-05 2023-01-21 10:52:05.934927: step: 1788/529, loss: 0.1091470792889595 2023-01-21 10:52:07.067439: step: 1792/529, loss: 0.1412241905927658 2023-01-21 10:52:08.204904: step: 1796/529, loss: 0.015224265865981579 2023-01-21 10:52:09.339150: step: 1800/529, loss: 0.14745500683784485 2023-01-21 10:52:10.447653: step: 1804/529, loss: 0.003253340721130371 2023-01-21 10:52:11.545964: step: 1808/529, loss: 0.01631612703204155 2023-01-21 10:52:12.688801: step: 1812/529, loss: 0.011667251586914062 2023-01-21 10:52:13.828025: step: 1816/529, loss: 0.021270085126161575 2023-01-21 10:52:15.058035: step: 1820/529, loss: 0.3226621448993683 2023-01-21 10:52:16.182764: step: 1824/529, loss: 0.11391735076904297 2023-01-21 10:52:17.290959: step: 1828/529, loss: 0.028792666271328926 2023-01-21 10:52:18.403213: step: 1832/529, loss: 0.018990231677889824 2023-01-21 10:52:19.530939: step: 1836/529, loss: 0.003859901335090399 2023-01-21 10:52:20.629747: step: 1840/529, loss: 0.054879650473594666 2023-01-21 10:52:21.754947: step: 1844/529, loss: 0.08546887338161469 2023-01-21 10:52:22.861576: step: 1848/529, loss: 0.020268870517611504 2023-01-21 10:52:23.972195: step: 1852/529, loss: 0.13133297860622406 2023-01-21 10:52:25.061850: step: 1856/529, loss: 0.004431438632309437 2023-01-21 10:52:26.192160: step: 1860/529, loss: 0.04313831031322479 2023-01-21 10:52:27.305666: step: 1864/529, loss: 0.11732502281665802 2023-01-21 10:52:28.407769: step: 1868/529, loss: 0.0823608934879303 2023-01-21 10:52:29.501552: step: 1872/529, loss: 0.06921844929456711 2023-01-21 10:52:30.594993: step: 1876/529, loss: 0.18772143125534058 2023-01-21 10:52:31.713436: step: 1880/529, loss: 0.023085784167051315 2023-01-21 10:52:32.824244: step: 1884/529, loss: 0.0010964394314214587 2023-01-21 10:52:33.947604: step: 1888/529, loss: 0.04357624053955078 2023-01-21 10:52:35.051389: step: 1892/529, loss: 0.09856834262609482 2023-01-21 10:52:36.209858: step: 1896/529, loss: 0.037571147084236145 2023-01-21 10:52:37.376542: step: 1900/529, loss: 0.03520956262946129 2023-01-21 10:52:38.471653: step: 1904/529, loss: 0.08882074803113937 2023-01-21 10:52:39.600551: step: 1908/529, loss: 0.02763180620968342 2023-01-21 10:52:40.700040: step: 1912/529, loss: 0.15327243506908417 2023-01-21 10:52:41.826953: step: 1916/529, loss: 0.11791391670703888 2023-01-21 10:52:42.946825: step: 1920/529, loss: 0.018830109387636185 2023-01-21 10:52:44.064097: step: 1924/529, loss: 0.08174372464418411 2023-01-21 10:52:45.201730: step: 1928/529, loss: 0.06939716637134552 2023-01-21 10:52:46.305477: step: 1932/529, loss: 0.05230093002319336 2023-01-21 10:52:47.406965: step: 1936/529, loss: 0.17533046007156372 2023-01-21 10:52:48.504973: step: 1940/529, loss: 0.007310295477509499 2023-01-21 10:52:49.618647: step: 1944/529, loss: 0.014315605163574219 2023-01-21 10:52:50.747353: step: 1948/529, loss: 0.023603271692991257 2023-01-21 10:52:51.873181: step: 1952/529, loss: 0.002918052952736616 2023-01-21 10:52:52.963390: step: 1956/529, loss: 0.003342056181281805 2023-01-21 10:52:54.079038: step: 1960/529, loss: 0.08848772197961807 2023-01-21 10:52:55.183919: step: 1964/529, loss: 0.06822852790355682 2023-01-21 10:52:56.300320: step: 1968/529, loss: 0.09141135215759277 2023-01-21 10:52:57.401003: step: 1972/529, loss: 0.012217319570481777 2023-01-21 10:52:58.546690: step: 1976/529, loss: 0.06513891369104385 2023-01-21 10:52:59.641694: step: 1980/529, loss: 0.0016072273720055819 2023-01-21 10:53:00.759207: step: 1984/529, loss: 0.0009511947864666581 2023-01-21 10:53:01.839057: step: 1988/529, loss: 0.000521802925504744 2023-01-21 10:53:02.939971: step: 1992/529, loss: 0.06409807503223419 2023-01-21 10:53:04.040070: step: 1996/529, loss: 0.024903012439608574 2023-01-21 10:53:05.153169: step: 2000/529, loss: 0.006233835127204657 2023-01-21 10:53:06.266678: step: 2004/529, loss: 0.00327816023491323 2023-01-21 10:53:07.369053: step: 2008/529, loss: 0.000973129237536341 2023-01-21 10:53:08.503685: step: 2012/529, loss: 0.15951919555664062 2023-01-21 10:53:09.622856: step: 2016/529, loss: 0.02843017503619194 2023-01-21 10:53:10.756119: step: 2020/529, loss: 0.07817402482032776 2023-01-21 10:53:11.900080: step: 2024/529, loss: 0.19659671187400818 2023-01-21 10:53:12.996883: step: 2028/529, loss: 0.007664203643798828 2023-01-21 10:53:14.087949: step: 2032/529, loss: 0.04596920311450958 2023-01-21 10:53:15.163750: step: 2036/529, loss: 0.02553539350628853 2023-01-21 10:53:16.280319: step: 2040/529, loss: 0.0249815471470356 2023-01-21 10:53:17.383858: step: 2044/529, loss: 0.007485067937523127 2023-01-21 10:53:18.492785: step: 2048/529, loss: 0.04357624053955078 2023-01-21 10:53:19.596716: step: 2052/529, loss: 0.07949686050415039 2023-01-21 10:53:20.755736: step: 2056/529, loss: 0.00826110877096653 2023-01-21 10:53:21.874717: step: 2060/529, loss: 0.0363314151763916 2023-01-21 10:53:22.992004: step: 2064/529, loss: 0.018836308270692825 2023-01-21 10:53:24.142791: step: 2068/529, loss: 0.012301159091293812 2023-01-21 10:53:25.273144: step: 2072/529, loss: 0.1309131681919098 2023-01-21 10:53:26.418780: step: 2076/529, loss: 0.03051462210714817 2023-01-21 10:53:27.539837: step: 2080/529, loss: 0.0825420394539833 2023-01-21 10:53:28.648358: step: 2084/529, loss: 0.05312299728393555 2023-01-21 10:53:29.742323: step: 2088/529, loss: 0.017402267083525658 2023-01-21 10:53:30.835631: step: 2092/529, loss: 0.011228561401367188 2023-01-21 10:53:31.959629: step: 2096/529, loss: 0.011606263928115368 2023-01-21 10:53:33.041962: step: 2100/529, loss: 0.005156659986823797 2023-01-21 10:53:34.150573: step: 2104/529, loss: 0.021921921521425247 2023-01-21 10:53:35.260895: step: 2108/529, loss: 0.09455004334449768 2023-01-21 10:53:36.334734: step: 2112/529, loss: 0.0348484069108963 2023-01-21 10:53:37.447047: step: 2116/529, loss: 0.0025954246520996094 ================================================== Loss: 0.057 -------------------- Dev: {'event': {'p': 0.5737211634904714, 'r': 0.7616511318242344, 'f1': 0.6544622425629291}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Test: {'event': {'p': 0.6309523809523809, 'r': 0.7776523702031602, 'f1': 0.6966632962588473}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Chinese: {'event': {'p': 0.5853658536585366, 'r': 0.8888888888888888, 'f1': 0.7058823529411764}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Korean: {'event': {'p': 0.5694444444444444, 'r': 0.6507936507936508, 'f1': 0.6074074074074073}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Russian: {'event': {'p': 0.4090909090909091, 'r': 0.5, 'f1': 0.45}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 14 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:54:18.113541: step: 4/529, loss: 0.28542229533195496 2023-01-21 10:54:19.274875: step: 8/529, loss: 0.0042168619111180305 2023-01-21 10:54:20.420756: step: 12/529, loss: 0.04806122928857803 2023-01-21 10:54:21.545012: step: 16/529, loss: 0.017220783978700638 2023-01-21 10:54:22.646690: step: 20/529, loss: 0.0016108989948406816 2023-01-21 10:54:23.767460: step: 24/529, loss: 0.0064242840744555 2023-01-21 10:54:24.899221: step: 28/529, loss: 0.06656856834888458 2023-01-21 10:54:26.013270: step: 32/529, loss: 0.005133247002959251 2023-01-21 10:54:27.116955: step: 36/529, loss: 0.038977812975645065 2023-01-21 10:54:28.246847: step: 40/529, loss: 0.0028447629883885384 2023-01-21 10:54:29.368549: step: 44/529, loss: 0.06085243076086044 2023-01-21 10:54:30.489930: step: 48/529, loss: 0.026661396026611328 2023-01-21 10:54:31.592679: step: 52/529, loss: 0.10043134540319443 2023-01-21 10:54:32.696431: step: 56/529, loss: 0.014155769720673561 2023-01-21 10:54:33.793896: step: 60/529, loss: 0.009986400604248047 2023-01-21 10:54:34.882656: step: 64/529, loss: 0.009379482828080654 2023-01-21 10:54:35.992320: step: 68/529, loss: 0.02329711988568306 2023-01-21 10:54:37.106081: step: 72/529, loss: 0.017309190705418587 2023-01-21 10:54:38.205949: step: 76/529, loss: 0.03655261918902397 2023-01-21 10:54:39.325774: step: 80/529, loss: 0.0006664872053079307 2023-01-21 10:54:40.459039: step: 84/529, loss: 0.0023785592056810856 2023-01-21 10:54:41.581211: step: 88/529, loss: 0.044165994971990585 2023-01-21 10:54:42.689597: step: 92/529, loss: 0.010099363513290882 2023-01-21 10:54:43.826844: step: 96/529, loss: 0.007143402472138405 2023-01-21 10:54:44.941806: step: 100/529, loss: 0.009181499481201172 2023-01-21 10:54:46.088132: step: 104/529, loss: 0.036187078803777695 2023-01-21 10:54:47.204075: step: 108/529, loss: 0.014006709679961205 2023-01-21 10:54:48.343825: step: 112/529, loss: 0.0037601948715746403 2023-01-21 10:54:49.459979: step: 116/529, loss: 0.002389908069744706 2023-01-21 10:54:50.562483: step: 120/529, loss: 0.00984582956880331 2023-01-21 10:54:51.691574: step: 124/529, loss: 0.06060028076171875 2023-01-21 10:54:52.826659: step: 128/529, loss: 0.021651554852724075 2023-01-21 10:54:53.948195: step: 132/529, loss: 0.02246398851275444 2023-01-21 10:54:55.049840: step: 136/529, loss: 0.027455855160951614 2023-01-21 10:54:56.139658: step: 140/529, loss: 0.03280293941497803 2023-01-21 10:54:57.280483: step: 144/529, loss: 0.04183788597583771 2023-01-21 10:54:58.380831: step: 148/529, loss: 0.47634103894233704 2023-01-21 10:54:59.476574: step: 152/529, loss: 0.007531070616096258 2023-01-21 10:55:00.590340: step: 156/529, loss: 0.016284942626953125 2023-01-21 10:55:01.701900: step: 160/529, loss: 0.0007408142555505037 2023-01-21 10:55:02.834065: step: 164/529, loss: 0.15402469038963318 2023-01-21 10:55:03.984132: step: 168/529, loss: 0.04643745720386505 2023-01-21 10:55:05.105629: step: 172/529, loss: 0.030351685360074043 2023-01-21 10:55:06.208983: step: 176/529, loss: 0.000744724296964705 2023-01-21 10:55:07.320925: step: 180/529, loss: 0.018285728991031647 2023-01-21 10:55:08.422418: step: 184/529, loss: 0.027756405994296074 2023-01-21 10:55:09.544421: step: 188/529, loss: 0.07891988754272461 2023-01-21 10:55:10.680672: step: 192/529, loss: 0.032425977289676666 2023-01-21 10:55:11.780349: step: 196/529, loss: 0.019214630126953125 2023-01-21 10:55:12.896153: step: 200/529, loss: 0.005147123243659735 2023-01-21 10:55:14.069870: step: 204/529, loss: 0.07959537208080292 2023-01-21 10:55:15.190177: step: 208/529, loss: 0.029631519690155983 2023-01-21 10:55:16.291983: step: 212/529, loss: 0.023068904876708984 2023-01-21 10:55:17.422219: step: 216/529, loss: 0.3601033091545105 2023-01-21 10:55:18.548349: step: 220/529, loss: 0.15707498788833618 2023-01-21 10:55:19.649568: step: 224/529, loss: 0.0167996883392334 2023-01-21 10:55:20.745671: step: 228/529, loss: 0.027569960802793503 2023-01-21 10:55:21.837309: step: 232/529, loss: 0.008847236633300781 2023-01-21 10:55:22.938972: step: 236/529, loss: 0.021175097674131393 2023-01-21 10:55:24.065713: step: 240/529, loss: 0.0478336364030838 2023-01-21 10:55:25.175368: step: 244/529, loss: 0.00977716501802206 2023-01-21 10:55:26.302608: step: 248/529, loss: 0.027523290365934372 2023-01-21 10:55:27.397974: step: 252/529, loss: 0.015262030996382236 2023-01-21 10:55:28.521171: step: 256/529, loss: 0.12833844125270844 2023-01-21 10:55:29.637982: step: 260/529, loss: 0.08741166442632675 2023-01-21 10:55:30.760380: step: 264/529, loss: 3.805160304182209e-05 2023-01-21 10:55:31.894696: step: 268/529, loss: 0.24247604608535767 2023-01-21 10:55:33.023428: step: 272/529, loss: 0.012257575988769531 2023-01-21 10:55:34.166978: step: 276/529, loss: 0.07098159939050674 2023-01-21 10:55:35.298770: step: 280/529, loss: 0.011781120672821999 2023-01-21 10:55:36.391741: step: 284/529, loss: 0.016565512865781784 2023-01-21 10:55:37.486240: step: 288/529, loss: 0.0062236785888671875 2023-01-21 10:55:38.577380: step: 292/529, loss: 0.05268917232751846 2023-01-21 10:55:39.704628: step: 296/529, loss: 0.05069713667035103 2023-01-21 10:55:40.852938: step: 300/529, loss: 0.11061763763427734 2023-01-21 10:55:41.968725: step: 304/529, loss: 0.03950538486242294 2023-01-21 10:55:43.086100: step: 308/529, loss: 0.009746265597641468 2023-01-21 10:55:44.215329: step: 312/529, loss: 0.0030401230324059725 2023-01-21 10:55:45.318024: step: 316/529, loss: 0.1662786602973938 2023-01-21 10:55:46.433301: step: 320/529, loss: 0.011009598150849342 2023-01-21 10:55:47.546729: step: 324/529, loss: 0.010093117132782936 2023-01-21 10:55:48.677627: step: 328/529, loss: 0.031298160552978516 2023-01-21 10:55:49.781101: step: 332/529, loss: 0.02296752855181694 2023-01-21 10:55:50.915248: step: 336/529, loss: 0.02767777256667614 2023-01-21 10:55:52.050063: step: 340/529, loss: 0.01911487616598606 2023-01-21 10:55:53.188749: step: 344/529, loss: 0.015407824888825417 2023-01-21 10:55:54.289239: step: 348/529, loss: 0.005767822265625 2023-01-21 10:55:55.403306: step: 352/529, loss: 0.06564221531152725 2023-01-21 10:55:56.515883: step: 356/529, loss: 0.05016353353857994 2023-01-21 10:55:57.639558: step: 360/529, loss: 0.11410064995288849 2023-01-21 10:55:58.712670: step: 364/529, loss: 0.0035696029663085938 2023-01-21 10:55:59.835743: step: 368/529, loss: 0.011361980810761452 2023-01-21 10:56:00.929731: step: 372/529, loss: 0.5807350277900696 2023-01-21 10:56:02.104083: step: 376/529, loss: 0.09486951678991318 2023-01-21 10:56:03.192490: step: 380/529, loss: 0.04363078996539116 2023-01-21 10:56:04.304673: step: 384/529, loss: 0.024254370480775833 2023-01-21 10:56:05.468917: step: 388/529, loss: 0.007377815432846546 2023-01-21 10:56:06.571994: step: 392/529, loss: 0.09204283356666565 2023-01-21 10:56:07.674396: step: 396/529, loss: 0.01895122602581978 2023-01-21 10:56:08.797095: step: 400/529, loss: 0.0023217203561216593 2023-01-21 10:56:09.926095: step: 404/529, loss: 0.19429558515548706 2023-01-21 10:56:11.018392: step: 408/529, loss: 0.04059314727783203 2023-01-21 10:56:12.140244: step: 412/529, loss: 0.009821319952607155 2023-01-21 10:56:13.214752: step: 416/529, loss: 0.003291606903076172 2023-01-21 10:56:14.332901: step: 420/529, loss: 0.027608491480350494 2023-01-21 10:56:15.467490: step: 424/529, loss: 0.04036521911621094 2023-01-21 10:56:16.593431: step: 428/529, loss: 0.00449371337890625 2023-01-21 10:56:17.729139: step: 432/529, loss: 0.051935769617557526 2023-01-21 10:56:18.848506: step: 436/529, loss: 0.044599153101444244 2023-01-21 10:56:19.976161: step: 440/529, loss: 0.018408585339784622 2023-01-21 10:56:21.077197: step: 444/529, loss: 0.1137228012084961 2023-01-21 10:56:22.179344: step: 448/529, loss: 0.047884751111269 2023-01-21 10:56:23.292320: step: 452/529, loss: 0.001063489937223494 2023-01-21 10:56:24.440887: step: 456/529, loss: 0.12674351036548615 2023-01-21 10:56:25.531349: step: 460/529, loss: 0.029877090826630592 2023-01-21 10:56:26.679044: step: 464/529, loss: 0.04743204265832901 2023-01-21 10:56:27.812345: step: 468/529, loss: 0.003746605012565851 2023-01-21 10:56:28.913234: step: 472/529, loss: 0.00803365744650364 2023-01-21 10:56:30.061259: step: 476/529, loss: 0.06054859235882759 2023-01-21 10:56:31.172731: step: 480/529, loss: 0.009893608279526234 2023-01-21 10:56:32.279019: step: 484/529, loss: 0.016243744641542435 2023-01-21 10:56:33.404809: step: 488/529, loss: 0.02563939243555069 2023-01-21 10:56:34.546703: step: 492/529, loss: 0.00029244425240904093 2023-01-21 10:56:35.656858: step: 496/529, loss: 0.08795707672834396 2023-01-21 10:56:36.807915: step: 500/529, loss: 0.04089775308966637 2023-01-21 10:56:37.925442: step: 504/529, loss: 0.06499533355236053 2023-01-21 10:56:39.028280: step: 508/529, loss: 0.006019783206284046 2023-01-21 10:56:40.133902: step: 512/529, loss: 0.02006855048239231 2023-01-21 10:56:41.250125: step: 516/529, loss: 0.08030052483081818 2023-01-21 10:56:42.385016: step: 520/529, loss: 0.022870827466249466 2023-01-21 10:56:43.500782: step: 524/529, loss: 0.031397152692079544 2023-01-21 10:56:44.612239: step: 528/529, loss: 0.1236150711774826 2023-01-21 10:56:45.726279: step: 532/529, loss: 0.06669120490550995 2023-01-21 10:56:46.845361: step: 536/529, loss: 0.0012908936478197575 2023-01-21 10:56:47.934358: step: 540/529, loss: 0.020603036507964134 2023-01-21 10:56:49.042020: step: 544/529, loss: 0.00563020771369338 2023-01-21 10:56:50.236567: step: 548/529, loss: 0.027902984991669655 2023-01-21 10:56:51.362995: step: 552/529, loss: 0.0007241249550133944 2023-01-21 10:56:52.516238: step: 556/529, loss: 0.04835481941699982 2023-01-21 10:56:53.625537: step: 560/529, loss: 0.01345295924693346 2023-01-21 10:56:54.735951: step: 564/529, loss: 0.010922098532319069 2023-01-21 10:56:55.875713: step: 568/529, loss: 0.20667631924152374 2023-01-21 10:56:57.019893: step: 572/529, loss: 0.04800424724817276 2023-01-21 10:56:58.145915: step: 576/529, loss: 0.005822932813316584 2023-01-21 10:56:59.270509: step: 580/529, loss: 0.1518479287624359 2023-01-21 10:57:00.402286: step: 584/529, loss: 0.015522956848144531 2023-01-21 10:57:01.497425: step: 588/529, loss: 0.04946718364953995 2023-01-21 10:57:02.613999: step: 592/529, loss: 0.019103432074189186 2023-01-21 10:57:03.719726: step: 596/529, loss: 0.15405121445655823 2023-01-21 10:57:04.829755: step: 600/529, loss: 0.04489975422620773 2023-01-21 10:57:05.911466: step: 604/529, loss: 0.08661253750324249 2023-01-21 10:57:07.019845: step: 608/529, loss: 0.03749570995569229 2023-01-21 10:57:08.110446: step: 612/529, loss: 0.002836799481883645 2023-01-21 10:57:09.231381: step: 616/529, loss: 0.0020859241485595703 2023-01-21 10:57:10.347814: step: 620/529, loss: 0.000773525214754045 2023-01-21 10:57:11.469118: step: 624/529, loss: 0.012419032864272594 2023-01-21 10:57:12.600548: step: 628/529, loss: 0.02784571796655655 2023-01-21 10:57:13.723644: step: 632/529, loss: 0.04108161851763725 2023-01-21 10:57:14.835483: step: 636/529, loss: 0.03397693857550621 2023-01-21 10:57:15.953914: step: 640/529, loss: 0.007110023871064186 2023-01-21 10:57:17.059491: step: 644/529, loss: 0.00018262863159179688 2023-01-21 10:57:18.180784: step: 648/529, loss: 0.029517151415348053 2023-01-21 10:57:19.302142: step: 652/529, loss: 0.007136869709938765 2023-01-21 10:57:20.398116: step: 656/529, loss: 0.0023105621803551912 2023-01-21 10:57:21.515045: step: 660/529, loss: 0.0077842059545218945 2023-01-21 10:57:22.609875: step: 664/529, loss: 0.013601399026811123 2023-01-21 10:57:23.734881: step: 668/529, loss: 0.15256109833717346 2023-01-21 10:57:24.848974: step: 672/529, loss: 0.017472077161073685 2023-01-21 10:57:25.995265: step: 676/529, loss: 0.028391553089022636 2023-01-21 10:57:27.120961: step: 680/529, loss: 0.031828153878450394 2023-01-21 10:57:28.223718: step: 684/529, loss: 0.010595321655273438 2023-01-21 10:57:29.316036: step: 688/529, loss: 0.005619335453957319 2023-01-21 10:57:30.435946: step: 692/529, loss: 0.036168668419122696 2023-01-21 10:57:31.565660: step: 696/529, loss: 0.1293676346540451 2023-01-21 10:57:32.687081: step: 700/529, loss: 0.0038583758287131786 2023-01-21 10:57:33.823328: step: 704/529, loss: 0.006727505009621382 2023-01-21 10:57:34.938898: step: 708/529, loss: 0.017378998920321465 2023-01-21 10:57:36.049995: step: 712/529, loss: 0.18939504027366638 2023-01-21 10:57:37.157968: step: 716/529, loss: 0.0037367106415331364 2023-01-21 10:57:38.285749: step: 720/529, loss: 0.009022330865263939 2023-01-21 10:57:39.428493: step: 724/529, loss: 0.08170413970947266 2023-01-21 10:57:40.537896: step: 728/529, loss: 0.143260195851326 2023-01-21 10:57:41.632781: step: 732/529, loss: 0.002220106078311801 2023-01-21 10:57:42.752224: step: 736/529, loss: 0.00667648296803236 2023-01-21 10:57:43.852931: step: 740/529, loss: 0.0002624511835165322 2023-01-21 10:57:44.973852: step: 744/529, loss: 0.0014350891578942537 2023-01-21 10:57:46.100628: step: 748/529, loss: 0.055256180465221405 2023-01-21 10:57:47.243315: step: 752/529, loss: 0.009876633062958717 2023-01-21 10:57:48.350025: step: 756/529, loss: 0.006892490200698376 2023-01-21 10:57:49.482170: step: 760/529, loss: 0.04370613023638725 2023-01-21 10:57:50.572399: step: 764/529, loss: 0.04192943871021271 2023-01-21 10:57:51.699809: step: 768/529, loss: 0.012794646434485912 2023-01-21 10:57:52.810895: step: 772/529, loss: 0.004298686981201172 2023-01-21 10:57:53.919798: step: 776/529, loss: 0.009894943796098232 2023-01-21 10:57:55.041629: step: 780/529, loss: 0.06111784279346466 2023-01-21 10:57:56.177288: step: 784/529, loss: 0.004887962713837624 2023-01-21 10:57:57.286225: step: 788/529, loss: 0.05174970626831055 2023-01-21 10:57:58.392678: step: 792/529, loss: 0.012651443481445312 2023-01-21 10:57:59.538155: step: 796/529, loss: 0.014636039733886719 2023-01-21 10:58:00.675440: step: 800/529, loss: 0.007879066281020641 2023-01-21 10:58:01.796543: step: 804/529, loss: 0.011621571145951748 2023-01-21 10:58:02.916482: step: 808/529, loss: 0.005214977078139782 2023-01-21 10:58:04.033556: step: 812/529, loss: 0.11691112816333771 2023-01-21 10:58:05.137563: step: 816/529, loss: 0.015779685229063034 2023-01-21 10:58:06.246416: step: 820/529, loss: 0.017511941492557526 2023-01-21 10:58:07.384197: step: 824/529, loss: 0.045705415308475494 2023-01-21 10:58:08.501825: step: 828/529, loss: 0.012109566479921341 2023-01-21 10:58:09.604496: step: 832/529, loss: 0.000602293002884835 2023-01-21 10:58:10.756208: step: 836/529, loss: 0.06588821858167648 2023-01-21 10:58:11.868405: step: 840/529, loss: 0.14721809327602386 2023-01-21 10:58:12.967177: step: 844/529, loss: 0.01297235582023859 2023-01-21 10:58:14.096197: step: 848/529, loss: 0.04844837263226509 2023-01-21 10:58:15.223707: step: 852/529, loss: 0.07427187263965607 2023-01-21 10:58:16.347958: step: 856/529, loss: 0.021271325647830963 2023-01-21 10:58:17.474014: step: 860/529, loss: 0.03399200737476349 2023-01-21 10:58:18.600526: step: 864/529, loss: 0.3483470380306244 2023-01-21 10:58:19.702101: step: 868/529, loss: 0.011655235663056374 2023-01-21 10:58:20.861078: step: 872/529, loss: 0.09243965148925781 2023-01-21 10:58:21.972993: step: 876/529, loss: 0.014841079711914062 2023-01-21 10:58:23.063313: step: 880/529, loss: 0.00798502005636692 2023-01-21 10:58:24.197098: step: 884/529, loss: 0.02210206910967827 2023-01-21 10:58:25.312478: step: 888/529, loss: 0.004823732189834118 2023-01-21 10:58:26.454061: step: 892/529, loss: 0.008584022521972656 2023-01-21 10:58:27.561607: step: 896/529, loss: 0.06557025760412216 2023-01-21 10:58:28.667677: step: 900/529, loss: 0.006735515780746937 2023-01-21 10:58:29.761042: step: 904/529, loss: 0.07895927131175995 2023-01-21 10:58:30.853972: step: 908/529, loss: 0.0009802818531170487 2023-01-21 10:58:31.940397: step: 912/529, loss: 0.004580020904541016 2023-01-21 10:58:33.057110: step: 916/529, loss: 0.015382385812699795 2023-01-21 10:58:34.166247: step: 920/529, loss: 0.06200065463781357 2023-01-21 10:58:35.296587: step: 924/529, loss: 0.0052734375931322575 2023-01-21 10:58:36.440511: step: 928/529, loss: 0.010949802584946156 2023-01-21 10:58:37.557048: step: 932/529, loss: 0.0022441863548010588 2023-01-21 10:58:38.649979: step: 936/529, loss: 0.0019631385803222656 2023-01-21 10:58:39.786356: step: 940/529, loss: 0.03144259378314018 2023-01-21 10:58:40.922271: step: 944/529, loss: 0.020041657611727715 2023-01-21 10:58:42.052033: step: 948/529, loss: 0.04442944750189781 2023-01-21 10:58:43.160319: step: 952/529, loss: 0.04252944141626358 2023-01-21 10:58:44.265388: step: 956/529, loss: 0.01144332904368639 2023-01-21 10:58:45.386293: step: 960/529, loss: 0.004132079891860485 2023-01-21 10:58:46.484823: step: 964/529, loss: 0.004291629884392023 2023-01-21 10:58:47.609039: step: 968/529, loss: 0.05982246622443199 2023-01-21 10:58:48.728555: step: 972/529, loss: 0.0009747505537234247 2023-01-21 10:58:49.846882: step: 976/529, loss: 0.12167644500732422 2023-01-21 10:58:50.936592: step: 980/529, loss: 0.007058620452880859 2023-01-21 10:58:52.055717: step: 984/529, loss: 0.03338613733649254 2023-01-21 10:58:53.171025: step: 988/529, loss: 0.05592937394976616 2023-01-21 10:58:54.275544: step: 992/529, loss: 0.03216876834630966 2023-01-21 10:58:55.357245: step: 996/529, loss: 0.13147468864917755 2023-01-21 10:58:56.475395: step: 1000/529, loss: 0.884032130241394 2023-01-21 10:58:57.574661: step: 1004/529, loss: 0.05263042449951172 2023-01-21 10:58:58.677110: step: 1008/529, loss: 0.060459233820438385 2023-01-21 10:58:59.815230: step: 1012/529, loss: 0.05985517427325249 2023-01-21 10:59:00.941218: step: 1016/529, loss: 0.010261249728500843 2023-01-21 10:59:02.047115: step: 1020/529, loss: 0.046550750732421875 2023-01-21 10:59:03.136969: step: 1024/529, loss: 0.025100301951169968 2023-01-21 10:59:04.255349: step: 1028/529, loss: 0.004088974092155695 2023-01-21 10:59:05.340843: step: 1032/529, loss: 0.1093822717666626 2023-01-21 10:59:06.477879: step: 1036/529, loss: 0.0045642852783203125 2023-01-21 10:59:07.605329: step: 1040/529, loss: 0.0670870840549469 2023-01-21 10:59:08.718900: step: 1044/529, loss: 0.006023979280143976 2023-01-21 10:59:09.809525: step: 1048/529, loss: 0.0063266754150390625 2023-01-21 10:59:10.914777: step: 1052/529, loss: 0.02001981809735298 2023-01-21 10:59:12.042562: step: 1056/529, loss: 0.044989779591560364 2023-01-21 10:59:13.155782: step: 1060/529, loss: 0.11159010231494904 2023-01-21 10:59:14.274542: step: 1064/529, loss: 0.006058120634406805 2023-01-21 10:59:15.383717: step: 1068/529, loss: 0.01607370376586914 2023-01-21 10:59:16.491852: step: 1072/529, loss: 0.0021827698219567537 2023-01-21 10:59:17.601051: step: 1076/529, loss: 0.3098045289516449 2023-01-21 10:59:18.739668: step: 1080/529, loss: 0.0009037017589434981 2023-01-21 10:59:19.876231: step: 1084/529, loss: 0.012361908331513405 2023-01-21 10:59:20.992674: step: 1088/529, loss: 0.004008507821708918 2023-01-21 10:59:22.090825: step: 1092/529, loss: 0.0027400015387684107 2023-01-21 10:59:23.216236: step: 1096/529, loss: 0.04791250452399254 2023-01-21 10:59:24.311497: step: 1100/529, loss: 0.038619138300418854 2023-01-21 10:59:25.447710: step: 1104/529, loss: 0.00642161350697279 2023-01-21 10:59:26.592283: step: 1108/529, loss: 0.0019177437061443925 2023-01-21 10:59:27.748695: step: 1112/529, loss: 0.06991234421730042 2023-01-21 10:59:28.840983: step: 1116/529, loss: 0.04513683170080185 2023-01-21 10:59:29.967380: step: 1120/529, loss: 0.04821434244513512 2023-01-21 10:59:31.087651: step: 1124/529, loss: 0.014318719506263733 2023-01-21 10:59:32.198164: step: 1128/529, loss: 0.045820239931344986 2023-01-21 10:59:33.323527: step: 1132/529, loss: 0.00822591781616211 2023-01-21 10:59:34.456721: step: 1136/529, loss: 0.10409088432788849 2023-01-21 10:59:35.595621: step: 1140/529, loss: 0.019231416285037994 2023-01-21 10:59:36.694161: step: 1144/529, loss: 0.0037736238446086645 2023-01-21 10:59:37.843971: step: 1148/529, loss: 0.009040355682373047 2023-01-21 10:59:38.957035: step: 1152/529, loss: 0.03665924072265625 2023-01-21 10:59:40.050491: step: 1156/529, loss: 0.007178878877311945 2023-01-21 10:59:41.169867: step: 1160/529, loss: 0.06026649475097656 2023-01-21 10:59:42.273621: step: 1164/529, loss: 0.36482754349708557 2023-01-21 10:59:43.385597: step: 1168/529, loss: 0.004725265316665173 2023-01-21 10:59:44.521841: step: 1172/529, loss: 0.04809093475341797 2023-01-21 10:59:45.627282: step: 1176/529, loss: 0.013447762466967106 2023-01-21 10:59:46.757943: step: 1180/529, loss: 0.00018405914306640625 2023-01-21 10:59:47.882024: step: 1184/529, loss: 0.0003297329240012914 2023-01-21 10:59:49.013349: step: 1188/529, loss: 0.026902198791503906 2023-01-21 10:59:50.162892: step: 1192/529, loss: 0.020311832427978516 2023-01-21 10:59:51.271153: step: 1196/529, loss: 0.0022874355781823397 2023-01-21 10:59:52.397178: step: 1200/529, loss: 0.007556915283203125 2023-01-21 10:59:53.520967: step: 1204/529, loss: 0.004610705655068159 2023-01-21 10:59:54.626852: step: 1208/529, loss: 0.06693993508815765 2023-01-21 10:59:55.751368: step: 1212/529, loss: 0.026430893689393997 2023-01-21 10:59:56.885470: step: 1216/529, loss: 0.01758584938943386 2023-01-21 10:59:57.987628: step: 1220/529, loss: 0.017342805862426758 2023-01-21 10:59:59.133136: step: 1224/529, loss: 0.011812401004135609 2023-01-21 11:00:00.260925: step: 1228/529, loss: 0.02156543731689453 2023-01-21 11:00:01.374274: step: 1232/529, loss: 0.024366378784179688 2023-01-21 11:00:02.496213: step: 1236/529, loss: 0.0028502463828772306 2023-01-21 11:00:03.633048: step: 1240/529, loss: 0.012836921028792858 2023-01-21 11:00:04.739643: step: 1244/529, loss: 0.002596950624138117 2023-01-21 11:00:05.857000: step: 1248/529, loss: 0.013252640143036842 2023-01-21 11:00:06.985417: step: 1252/529, loss: 0.021851157769560814 2023-01-21 11:00:08.135113: step: 1256/529, loss: 0.04858551174402237 2023-01-21 11:00:09.234587: step: 1260/529, loss: 0.056967735290527344 2023-01-21 11:00:10.362438: step: 1264/529, loss: 0.045302774757146835 2023-01-21 11:00:11.517895: step: 1268/529, loss: 0.01122751273214817 2023-01-21 11:00:12.649744: step: 1272/529, loss: 0.009167671203613281 2023-01-21 11:00:13.758621: step: 1276/529, loss: 0.04904327541589737 2023-01-21 11:00:14.852432: step: 1280/529, loss: 0.038149259984493256 2023-01-21 11:00:15.960529: step: 1284/529, loss: 0.08703117817640305 2023-01-21 11:00:17.066185: step: 1288/529, loss: 0.034574221819639206 2023-01-21 11:00:18.175661: step: 1292/529, loss: 0.043476298451423645 2023-01-21 11:00:19.310585: step: 1296/529, loss: 0.07742729038000107 2023-01-21 11:00:20.410633: step: 1300/529, loss: 0.024439431726932526 2023-01-21 11:00:21.562324: step: 1304/529, loss: 0.03694910928606987 2023-01-21 11:00:22.685413: step: 1308/529, loss: 0.028636742383241653 2023-01-21 11:00:23.779790: step: 1312/529, loss: 0.048247016966342926 2023-01-21 11:00:24.883905: step: 1316/529, loss: 0.0014256477588787675 2023-01-21 11:00:25.997177: step: 1320/529, loss: 0.012219620868563652 2023-01-21 11:00:27.096533: step: 1324/529, loss: 0.0881534144282341 2023-01-21 11:00:28.194124: step: 1328/529, loss: 0.06409988552331924 2023-01-21 11:00:29.301328: step: 1332/529, loss: 0.034006405621767044 2023-01-21 11:00:30.375820: step: 1336/529, loss: 0.008554316125810146 2023-01-21 11:00:31.517870: step: 1340/529, loss: 0.019129181280732155 2023-01-21 11:00:32.619883: step: 1344/529, loss: 0.006757402792572975 2023-01-21 11:00:33.700860: step: 1348/529, loss: 0.08117235451936722 2023-01-21 11:00:34.829976: step: 1352/529, loss: 0.015814591199159622 2023-01-21 11:00:35.928450: step: 1356/529, loss: 0.06495046615600586 2023-01-21 11:00:37.020299: step: 1360/529, loss: 0.011937332339584827 2023-01-21 11:00:38.125622: step: 1364/529, loss: 0.013608360663056374 2023-01-21 11:00:39.222170: step: 1368/529, loss: 0.02305459976196289 2023-01-21 11:00:40.318208: step: 1372/529, loss: 0.03589329868555069 2023-01-21 11:00:41.423896: step: 1376/529, loss: 0.016468429937958717 2023-01-21 11:00:42.564554: step: 1380/529, loss: 0.0040308949537575245 2023-01-21 11:00:43.705333: step: 1384/529, loss: 0.0046943663619458675 2023-01-21 11:00:44.819363: step: 1388/529, loss: 0.0018876552348956466 2023-01-21 11:00:45.954488: step: 1392/529, loss: 0.008018684573471546 2023-01-21 11:00:47.070603: step: 1396/529, loss: 0.029947759583592415 2023-01-21 11:00:48.178423: step: 1400/529, loss: 0.010931015014648438 2023-01-21 11:00:49.341572: step: 1404/529, loss: 0.00782022438943386 2023-01-21 11:00:50.497381: step: 1408/529, loss: 0.05877294763922691 2023-01-21 11:00:51.627552: step: 1412/529, loss: 0.015551663003861904 2023-01-21 11:00:52.795587: step: 1416/529, loss: 0.005804348271340132 2023-01-21 11:00:53.900026: step: 1420/529, loss: 0.001958656357601285 2023-01-21 11:00:55.008187: step: 1424/529, loss: 0.003894853638485074 2023-01-21 11:00:56.110495: step: 1428/529, loss: 0.006210899446159601 2023-01-21 11:00:57.242178: step: 1432/529, loss: 0.010835647583007812 2023-01-21 11:00:58.389632: step: 1436/529, loss: 0.023590851575136185 2023-01-21 11:00:59.496738: step: 1440/529, loss: 0.0008525848388671875 2023-01-21 11:01:00.608476: step: 1444/529, loss: 0.002076101489365101 2023-01-21 11:01:01.734520: step: 1448/529, loss: 0.051577188074588776 2023-01-21 11:01:02.844867: step: 1452/529, loss: 0.0019813061226159334 2023-01-21 11:01:03.965167: step: 1456/529, loss: 0.05126991495490074 2023-01-21 11:01:05.052647: step: 1460/529, loss: 0.02657637745141983 2023-01-21 11:01:06.149847: step: 1464/529, loss: 0.08320770412683487 2023-01-21 11:01:07.303475: step: 1468/529, loss: 0.06510219722986221 2023-01-21 11:01:08.413115: step: 1472/529, loss: 0.013584066182374954 2023-01-21 11:01:09.523720: step: 1476/529, loss: 0.2623741030693054 2023-01-21 11:01:10.662157: step: 1480/529, loss: 0.023077774792909622 2023-01-21 11:01:11.830521: step: 1484/529, loss: 0.053269293159246445 2023-01-21 11:01:12.965857: step: 1488/529, loss: 0.004222107119858265 2023-01-21 11:01:14.073239: step: 1492/529, loss: 0.023256875574588776 2023-01-21 11:01:15.182042: step: 1496/529, loss: 0.01752915419638157 2023-01-21 11:01:16.268350: step: 1500/529, loss: 0.005416584201157093 2023-01-21 11:01:17.364916: step: 1504/529, loss: 0.002806043718010187 2023-01-21 11:01:18.460688: step: 1508/529, loss: 0.030350064858794212 2023-01-21 11:01:19.577148: step: 1512/529, loss: 0.014558697119355202 2023-01-21 11:01:20.705517: step: 1516/529, loss: 0.08598404377698898 2023-01-21 11:01:21.814875: step: 1520/529, loss: 0.0382019504904747 2023-01-21 11:01:22.909873: step: 1524/529, loss: 0.007064259145408869 2023-01-21 11:01:24.012503: step: 1528/529, loss: 0.05376832187175751 2023-01-21 11:01:25.118239: step: 1532/529, loss: 0.03178863599896431 2023-01-21 11:01:26.237619: step: 1536/529, loss: 0.009189700707793236 2023-01-21 11:01:27.350186: step: 1540/529, loss: 0.051366519182920456 2023-01-21 11:01:28.451228: step: 1544/529, loss: 0.034120846539735794 2023-01-21 11:01:29.593252: step: 1548/529, loss: 0.023156166076660156 2023-01-21 11:01:30.725580: step: 1552/529, loss: 0.013827420771121979 2023-01-21 11:01:31.851785: step: 1556/529, loss: 0.03251180797815323 2023-01-21 11:01:32.979614: step: 1560/529, loss: 0.028513146564364433 2023-01-21 11:01:34.093351: step: 1564/529, loss: 0.5190154910087585 2023-01-21 11:01:35.224759: step: 1568/529, loss: 0.029238274320960045 2023-01-21 11:01:36.321724: step: 1572/529, loss: 0.002899265382438898 2023-01-21 11:01:37.458417: step: 1576/529, loss: 0.035485681146383286 2023-01-21 11:01:38.556960: step: 1580/529, loss: 0.01726846769452095 2023-01-21 11:01:39.701150: step: 1584/529, loss: 0.013656712137162685 2023-01-21 11:01:40.815752: step: 1588/529, loss: 0.03733348846435547 2023-01-21 11:01:41.953643: step: 1592/529, loss: 0.06979713588953018 2023-01-21 11:01:43.071504: step: 1596/529, loss: 0.049913790076971054 2023-01-21 11:01:44.190208: step: 1600/529, loss: 0.006816482171416283 2023-01-21 11:01:45.318214: step: 1604/529, loss: 0.010957884602248669 2023-01-21 11:01:46.423653: step: 1608/529, loss: 0.011243057437241077 2023-01-21 11:01:47.546055: step: 1612/529, loss: 0.10663265734910965 2023-01-21 11:01:48.633885: step: 1616/529, loss: 0.022330619394779205 2023-01-21 11:01:49.726474: step: 1620/529, loss: 0.004148292355239391 2023-01-21 11:01:50.882126: step: 1624/529, loss: 0.078783318400383 2023-01-21 11:01:52.008860: step: 1628/529, loss: 0.019055012613534927 2023-01-21 11:01:53.154779: step: 1632/529, loss: 0.00965871848165989 2023-01-21 11:01:54.297534: step: 1636/529, loss: 0.021849345415830612 2023-01-21 11:01:55.401221: step: 1640/529, loss: 0.07309045642614365 2023-01-21 11:01:56.545966: step: 1644/529, loss: 0.27570265531539917 2023-01-21 11:01:57.697849: step: 1648/529, loss: 0.021379852667450905 2023-01-21 11:01:58.813186: step: 1652/529, loss: 0.00640716589987278 2023-01-21 11:01:59.930923: step: 1656/529, loss: 0.010583113878965378 2023-01-21 11:02:01.052058: step: 1660/529, loss: 0.022100163623690605 2023-01-21 11:02:02.172064: step: 1664/529, loss: 0.0035991668701171875 2023-01-21 11:02:03.278211: step: 1668/529, loss: 0.018027067184448242 2023-01-21 11:02:04.389193: step: 1672/529, loss: 0.022297287359833717 2023-01-21 11:02:05.511997: step: 1676/529, loss: 0.1069829985499382 2023-01-21 11:02:06.614047: step: 1680/529, loss: 0.015033340081572533 2023-01-21 11:02:07.750193: step: 1684/529, loss: 0.7259511351585388 2023-01-21 11:02:08.880407: step: 1688/529, loss: 0.013080645352602005 2023-01-21 11:02:10.020841: step: 1692/529, loss: 0.011502360925078392 2023-01-21 11:02:11.149303: step: 1696/529, loss: 0.04706869274377823 2023-01-21 11:02:12.250703: step: 1700/529, loss: 0.028891896829009056 2023-01-21 11:02:13.360355: step: 1704/529, loss: 0.016765978187322617 2023-01-21 11:02:14.482777: step: 1708/529, loss: 0.008686160668730736 2023-01-21 11:02:15.579752: step: 1712/529, loss: 0.04540519788861275 2023-01-21 11:02:16.667378: step: 1716/529, loss: 0.05992479249835014 2023-01-21 11:02:17.808699: step: 1720/529, loss: 0.309914767742157 2023-01-21 11:02:18.932264: step: 1724/529, loss: 0.2603258192539215 2023-01-21 11:02:20.081643: step: 1728/529, loss: 0.019812941551208496 2023-01-21 11:02:21.201051: step: 1732/529, loss: 0.05162016302347183 2023-01-21 11:02:22.280447: step: 1736/529, loss: 0.0030622482299804688 2023-01-21 11:02:23.405168: step: 1740/529, loss: 0.0012599944602698088 2023-01-21 11:02:24.555284: step: 1744/529, loss: 0.31702733039855957 2023-01-21 11:02:25.672284: step: 1748/529, loss: 0.0022668838500976562 2023-01-21 11:02:26.786714: step: 1752/529, loss: 0.03912544250488281 2023-01-21 11:02:27.910689: step: 1756/529, loss: 0.026839256286621094 2023-01-21 11:02:29.034470: step: 1760/529, loss: 0.07360468059778214 2023-01-21 11:02:30.180086: step: 1764/529, loss: 0.042208481580019 2023-01-21 11:02:31.325051: step: 1768/529, loss: 0.010390853509306908 2023-01-21 11:02:32.445968: step: 1772/529, loss: 0.08274458348751068 2023-01-21 11:02:33.559919: step: 1776/529, loss: 0.038687705993652344 2023-01-21 11:02:34.656870: step: 1780/529, loss: 0.03593158721923828 2023-01-21 11:02:35.759453: step: 1784/529, loss: 0.004655647091567516 2023-01-21 11:02:36.843170: step: 1788/529, loss: 0.2558261752128601 2023-01-21 11:02:37.967618: step: 1792/529, loss: 0.05115795135498047 2023-01-21 11:02:39.082419: step: 1796/529, loss: 0.08115635067224503 2023-01-21 11:02:40.212643: step: 1800/529, loss: 0.07354607433080673 2023-01-21 11:02:41.312585: step: 1804/529, loss: 0.009752940386533737 2023-01-21 11:02:42.411907: step: 1808/529, loss: 0.0009815931553021073 2023-01-21 11:02:43.545789: step: 1812/529, loss: 0.010867690667510033 2023-01-21 11:02:44.660445: step: 1816/529, loss: 0.016701603308320045 2023-01-21 11:02:45.809497: step: 1820/529, loss: 0.05021953582763672 2023-01-21 11:02:46.949896: step: 1824/529, loss: 0.02888660319149494 2023-01-21 11:02:48.064846: step: 1828/529, loss: 0.014526177197694778 2023-01-21 11:02:49.186510: step: 1832/529, loss: 0.025641441345214844 2023-01-21 11:02:50.287818: step: 1836/529, loss: 0.014953423291444778 2023-01-21 11:02:51.422790: step: 1840/529, loss: 0.031143762171268463 2023-01-21 11:02:52.524122: step: 1844/529, loss: 0.02171764336526394 2023-01-21 11:02:53.621017: step: 1848/529, loss: 0.01120681781321764 2023-01-21 11:02:54.734280: step: 1852/529, loss: 0.04787120968103409 2023-01-21 11:02:55.861982: step: 1856/529, loss: 0.08074355870485306 2023-01-21 11:02:56.981033: step: 1860/529, loss: 0.005659008398652077 2023-01-21 11:02:58.123394: step: 1864/529, loss: 0.0758235901594162 2023-01-21 11:02:59.284797: step: 1868/529, loss: 0.10522308945655823 2023-01-21 11:03:00.397167: step: 1872/529, loss: 0.014465617947280407 2023-01-21 11:03:01.531914: step: 1876/529, loss: 0.0687403678894043 2023-01-21 11:03:02.664298: step: 1880/529, loss: 0.022282984107732773 2023-01-21 11:03:03.780421: step: 1884/529, loss: 0.015118788927793503 2023-01-21 11:03:04.898360: step: 1888/529, loss: 0.03575773537158966 2023-01-21 11:03:06.007012: step: 1892/529, loss: 0.007437491789460182 2023-01-21 11:03:07.111824: step: 1896/529, loss: 0.08016910403966904 2023-01-21 11:03:08.281104: step: 1900/529, loss: 0.019894791767001152 2023-01-21 11:03:09.428280: step: 1904/529, loss: 0.05418711155653 2023-01-21 11:03:10.532781: step: 1908/529, loss: 0.022116661071777344 2023-01-21 11:03:11.652928: step: 1912/529, loss: 0.016554927453398705 2023-01-21 11:03:12.802870: step: 1916/529, loss: 0.004768180660903454 2023-01-21 11:03:13.916238: step: 1920/529, loss: 0.010572051629424095 2023-01-21 11:03:15.045625: step: 1924/529, loss: 0.028961801901459694 2023-01-21 11:03:16.150455: step: 1928/529, loss: 0.07067890465259552 2023-01-21 11:03:17.247910: step: 1932/529, loss: 0.0065080164931714535 2023-01-21 11:03:18.400202: step: 1936/529, loss: 0.002762413118034601 2023-01-21 11:03:19.508168: step: 1940/529, loss: 0.05084487050771713 2023-01-21 11:03:20.648046: step: 1944/529, loss: 0.016050148755311966 2023-01-21 11:03:21.757667: step: 1948/529, loss: 0.035315800458192825 2023-01-21 11:03:22.913566: step: 1952/529, loss: 0.03607607260346413 2023-01-21 11:03:24.050319: step: 1956/529, loss: 0.04850387945771217 2023-01-21 11:03:25.149742: step: 1960/529, loss: 0.027379514649510384 2023-01-21 11:03:26.266271: step: 1964/529, loss: 0.032022763043642044 2023-01-21 11:03:27.376713: step: 1968/529, loss: 0.013166236691176891 2023-01-21 11:03:28.490448: step: 1972/529, loss: 0.03219404071569443 2023-01-21 11:03:29.585235: step: 1976/529, loss: 0.011235046200454235 2023-01-21 11:03:30.697469: step: 1980/529, loss: 0.03635311499238014 2023-01-21 11:03:31.806034: step: 1984/529, loss: 0.011315440759062767 2023-01-21 11:03:32.928888: step: 1988/529, loss: 0.4759610891342163 2023-01-21 11:03:34.014707: step: 1992/529, loss: 0.011649226769804955 2023-01-21 11:03:35.122587: step: 1996/529, loss: 0.010376167483627796 2023-01-21 11:03:36.205623: step: 2000/529, loss: 0.03820042684674263 2023-01-21 11:03:37.327541: step: 2004/529, loss: 0.016927147284150124 2023-01-21 11:03:38.459815: step: 2008/529, loss: 0.05818786472082138 2023-01-21 11:03:39.568750: step: 2012/529, loss: 0.0522676482796669 2023-01-21 11:03:40.670247: step: 2016/529, loss: 0.03870201110839844 2023-01-21 11:03:41.768459: step: 2020/529, loss: 0.020067311823368073 2023-01-21 11:03:42.869394: step: 2024/529, loss: 0.0005195617559365928 2023-01-21 11:03:44.012176: step: 2028/529, loss: 0.10371380299329758 2023-01-21 11:03:45.118323: step: 2032/529, loss: 0.010229396633803844 2023-01-21 11:03:46.227711: step: 2036/529, loss: 0.0035326003562659025 2023-01-21 11:03:47.358307: step: 2040/529, loss: 0.03127174451947212 2023-01-21 11:03:48.485816: step: 2044/529, loss: 0.02008695714175701 2023-01-21 11:03:49.603830: step: 2048/529, loss: 0.01170883234590292 2023-01-21 11:03:50.727670: step: 2052/529, loss: 0.01786642149090767 2023-01-21 11:03:51.852072: step: 2056/529, loss: 0.04937922954559326 2023-01-21 11:03:52.950825: step: 2060/529, loss: 0.10546989738941193 2023-01-21 11:03:54.100713: step: 2064/529, loss: 0.06906566768884659 2023-01-21 11:03:55.221015: step: 2068/529, loss: 0.06697960197925568 2023-01-21 11:03:56.330306: step: 2072/529, loss: 0.05685315281152725 2023-01-21 11:03:57.450400: step: 2076/529, loss: 0.04433488845825195 2023-01-21 11:03:58.566555: step: 2080/529, loss: 0.011316823773086071 2023-01-21 11:03:59.673726: step: 2084/529, loss: 0.0305952075868845 2023-01-21 11:04:00.775664: step: 2088/529, loss: 0.0007498741615563631 2023-01-21 11:04:01.911036: step: 2092/529, loss: 0.00426592817530036 2023-01-21 11:04:03.010068: step: 2096/529, loss: 0.13869649171829224 2023-01-21 11:04:04.158975: step: 2100/529, loss: 0.008698845282196999 2023-01-21 11:04:05.254016: step: 2104/529, loss: 0.004303073976188898 2023-01-21 11:04:06.381152: step: 2108/529, loss: 0.024271393194794655 2023-01-21 11:04:07.540002: step: 2112/529, loss: 0.02950286865234375 2023-01-21 11:04:08.641780: step: 2116/529, loss: 0.009170723147690296 ================================================== Loss: 0.045 -------------------- Dev: {'event': {'p': 0.5793650793650794, 'r': 0.7776298268974701, 'f1': 0.6640136441159751}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Test: {'event': {'p': 0.6310592459605027, 'r': 0.7934537246049661, 'f1': 0.703}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Chinese: {'event': {'p': 0.5333333333333333, 'r': 0.8888888888888888, 'f1': 0.6666666666666667}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} 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': 14} 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': 14} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5842696629213483, 'r': 0.7616511318242344, 'f1': 0.661271676300578}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Eng Test for Korean: {'event': {'p': 0.6081741448245225, 'r': 0.7725733634311512, 'f1': 0.6805866268953518}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Sample Korean: {'event': {'p': 0.6190476190476191, 'r': 0.6190476190476191, 'f1': 0.6190476190476191}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 15 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:04:49.716626: step: 4/529, loss: 0.02822570875287056 2023-01-21 11:04:50.831015: step: 8/529, loss: 0.00940475519746542 2023-01-21 11:04:51.941905: step: 12/529, loss: 0.006194019690155983 2023-01-21 11:04:53.050062: step: 16/529, loss: 0.0021185874938964844 2023-01-21 11:04:54.178303: step: 20/529, loss: 0.0009635925525799394 2023-01-21 11:04:55.263936: step: 24/529, loss: 0.003908062353730202 2023-01-21 11:04:56.365730: step: 28/529, loss: 0.10393648594617844 2023-01-21 11:04:57.485250: step: 32/529, loss: 0.0164628978818655 2023-01-21 11:04:58.591838: step: 36/529, loss: 0.23524247109889984 2023-01-21 11:04:59.689392: step: 40/529, loss: 0.0393267385661602 2023-01-21 11:05:00.833638: step: 44/529, loss: 0.010589790530502796 2023-01-21 11:05:01.947814: step: 48/529, loss: 0.003737163729965687 2023-01-21 11:05:03.088974: step: 52/529, loss: 0.05091991648077965 2023-01-21 11:05:04.195127: step: 56/529, loss: 0.004349804017692804 2023-01-21 11:05:05.355672: step: 60/529, loss: 0.014333629980683327 2023-01-21 11:05:06.456288: step: 64/529, loss: 0.014070892706513405 2023-01-21 11:05:07.570013: step: 68/529, loss: 0.004953670781105757 2023-01-21 11:05:08.685316: step: 72/529, loss: 0.011232947930693626 2023-01-21 11:05:09.783878: step: 76/529, loss: 0.027726365253329277 2023-01-21 11:05:10.883014: step: 80/529, loss: 0.04711728170514107 2023-01-21 11:05:12.007848: step: 84/529, loss: 0.006350231356918812 2023-01-21 11:05:13.126521: step: 88/529, loss: 0.029451562091708183 2023-01-21 11:05:14.257488: step: 92/529, loss: 0.0012616157764568925 2023-01-21 11:05:15.354107: step: 96/529, loss: 0.01969289965927601 2023-01-21 11:05:16.465055: step: 100/529, loss: 0.008139610290527344 2023-01-21 11:05:17.582008: step: 104/529, loss: 0.05815467983484268 2023-01-21 11:05:18.715010: step: 108/529, loss: 0.016068648546934128 2023-01-21 11:05:19.834859: step: 112/529, loss: 0.031229544430971146 2023-01-21 11:05:20.928358: step: 116/529, loss: 0.0016626358265057206 2023-01-21 11:05:22.080624: step: 120/529, loss: 0.001676845597103238 2023-01-21 11:05:23.193062: step: 124/529, loss: 0.036770012229681015 2023-01-21 11:05:24.303830: step: 128/529, loss: 0.004294777289032936 2023-01-21 11:05:25.401635: step: 132/529, loss: 0.02440032921731472 2023-01-21 11:05:26.484233: step: 136/529, loss: 0.09974256157875061 2023-01-21 11:05:27.611418: step: 140/529, loss: 0.004360604099929333 2023-01-21 11:05:28.707684: step: 144/529, loss: 0.001702594687230885 2023-01-21 11:05:29.809457: step: 148/529, loss: 0.05367422103881836 2023-01-21 11:05:30.926367: step: 152/529, loss: 0.327165424823761 2023-01-21 11:05:32.050093: step: 156/529, loss: 0.06527883559465408 2023-01-21 11:05:33.175585: step: 160/529, loss: 0.010801506228744984 2023-01-21 11:05:34.284627: step: 164/529, loss: 0.13365012407302856 2023-01-21 11:05:35.385995: step: 168/529, loss: 0.010923480615019798 2023-01-21 11:05:36.514410: step: 172/529, loss: 0.07018747180700302 2023-01-21 11:05:37.654447: step: 176/529, loss: 0.03233347088098526 2023-01-21 11:05:38.751842: step: 180/529, loss: 0.01683983951807022 2023-01-21 11:05:39.837017: step: 184/529, loss: 0.04851646348834038 2023-01-21 11:05:40.924021: step: 188/529, loss: 0.00893859937787056 2023-01-21 11:05:42.006353: step: 192/529, loss: 0.013376235030591488 2023-01-21 11:05:43.099590: step: 196/529, loss: 0.007291603367775679 2023-01-21 11:05:44.204938: step: 200/529, loss: 0.00826406478881836 2023-01-21 11:05:45.302028: step: 204/529, loss: 0.005395126529037952 2023-01-21 11:05:46.387167: step: 208/529, loss: 0.016063833609223366 2023-01-21 11:05:47.516194: step: 212/529, loss: 0.047331809997558594 2023-01-21 11:05:48.597368: step: 216/529, loss: 0.02811565436422825 2023-01-21 11:05:49.721818: step: 220/529, loss: 0.0023702620528638363 2023-01-21 11:05:50.838107: step: 224/529, loss: 0.06053737550973892 2023-01-21 11:05:51.955631: step: 228/529, loss: 0.006602573674172163 2023-01-21 11:05:53.052045: step: 232/529, loss: 0.027394484728574753 2023-01-21 11:05:54.172472: step: 236/529, loss: 0.06755141913890839 2023-01-21 11:05:55.311757: step: 240/529, loss: 0.0016830444801598787 2023-01-21 11:05:56.452437: step: 244/529, loss: 0.3285873532295227 2023-01-21 11:05:57.582303: step: 248/529, loss: 0.004387283232063055 2023-01-21 11:05:58.678009: step: 252/529, loss: 0.0550416000187397 2023-01-21 11:05:59.801685: step: 256/529, loss: 7.834434654796496e-05 2023-01-21 11:06:00.915890: step: 260/529, loss: 0.007579613011330366 2023-01-21 11:06:01.996072: step: 264/529, loss: 0.027159595862030983 2023-01-21 11:06:03.129109: step: 268/529, loss: 0.11134834587574005 2023-01-21 11:06:04.230729: step: 272/529, loss: 0.017879962921142578 2023-01-21 11:06:05.355376: step: 276/529, loss: 0.017342282459139824 2023-01-21 11:06:06.471244: step: 280/529, loss: 0.0011494637001305819 2023-01-21 11:06:07.600378: step: 284/529, loss: 0.0005737781757488847 2023-01-21 11:06:08.711225: step: 288/529, loss: 0.027031267061829567 2023-01-21 11:06:09.789945: step: 292/529, loss: 0.006271743681281805 2023-01-21 11:06:10.883266: step: 296/529, loss: 0.07226734608411789 2023-01-21 11:06:11.985895: step: 300/529, loss: 0.011046409606933594 2023-01-21 11:06:13.137397: step: 304/529, loss: 0.02284526824951172 2023-01-21 11:06:14.262763: step: 308/529, loss: 0.028666973114013672 2023-01-21 11:06:15.413633: step: 312/529, loss: 0.0015861510764807463 2023-01-21 11:06:16.591247: step: 316/529, loss: 0.005906105041503906 2023-01-21 11:06:17.742797: step: 320/529, loss: 0.04558046534657478 2023-01-21 11:06:18.862828: step: 324/529, loss: 0.2842459976673126 2023-01-21 11:06:19.987954: step: 328/529, loss: 0.03322868421673775 2023-01-21 11:06:21.108528: step: 332/529, loss: 0.05849600210785866 2023-01-21 11:06:22.198297: step: 336/529, loss: 0.000492095947265625 2023-01-21 11:06:23.325614: step: 340/529, loss: 0.035881806164979935 2023-01-21 11:06:24.425141: step: 344/529, loss: 0.008594751358032227 2023-01-21 11:06:25.549099: step: 348/529, loss: 0.05172138288617134 2023-01-21 11:06:26.712067: step: 352/529, loss: 0.020600175485014915 2023-01-21 11:06:27.837938: step: 356/529, loss: 0.0032741546165198088 2023-01-21 11:06:28.932822: step: 360/529, loss: 0.03638134151697159 2023-01-21 11:06:30.023240: step: 364/529, loss: 0.0177778247743845 2023-01-21 11:06:31.138355: step: 368/529, loss: 0.02365584298968315 2023-01-21 11:06:32.236985: step: 372/529, loss: 0.0453500971198082 2023-01-21 11:06:33.359526: step: 376/529, loss: 0.10588875412940979 2023-01-21 11:06:34.476934: step: 380/529, loss: 0.012697696685791016 2023-01-21 11:06:35.586664: step: 384/529, loss: 0.01796755939722061 2023-01-21 11:06:36.712610: step: 388/529, loss: 0.025812674313783646 2023-01-21 11:06:37.852344: step: 392/529, loss: 0.01332387886941433 2023-01-21 11:06:38.981389: step: 396/529, loss: 0.021006394177675247 2023-01-21 11:06:40.097458: step: 400/529, loss: 0.015304852277040482 2023-01-21 11:06:41.213761: step: 404/529, loss: 0.004199981689453125 2023-01-21 11:06:42.341898: step: 408/529, loss: 0.02092723734676838 2023-01-21 11:06:43.449514: step: 412/529, loss: 0.21907749772071838 2023-01-21 11:06:44.563142: step: 416/529, loss: 0.0034286021254956722 2023-01-21 11:06:45.689581: step: 420/529, loss: 0.025684071704745293 2023-01-21 11:06:46.816799: step: 424/529, loss: 0.02913360670208931 2023-01-21 11:06:47.946890: step: 428/529, loss: 0.09381237626075745 2023-01-21 11:06:49.087653: step: 432/529, loss: 0.002601051237434149 2023-01-21 11:06:50.238163: step: 436/529, loss: 0.015467643737792969 2023-01-21 11:06:51.366860: step: 440/529, loss: 0.015519714914262295 2023-01-21 11:06:52.476971: step: 444/529, loss: 0.11587519943714142 2023-01-21 11:06:53.631303: step: 448/529, loss: 0.06438083946704865 2023-01-21 11:06:54.776689: step: 452/529, loss: 0.04938783496618271 2023-01-21 11:06:55.894564: step: 456/529, loss: 0.018175793811678886 2023-01-21 11:06:57.054112: step: 460/529, loss: 0.002646827604621649 2023-01-21 11:06:58.206606: step: 464/529, loss: 0.04045305401086807 2023-01-21 11:06:59.303780: step: 468/529, loss: 0.023006536066532135 2023-01-21 11:07:00.417096: step: 472/529, loss: 0.023910902440547943 2023-01-21 11:07:01.522063: step: 476/529, loss: 0.0004017830069642514 2023-01-21 11:07:02.643152: step: 480/529, loss: 0.0016859055031090975 2023-01-21 11:07:03.760100: step: 484/529, loss: 0.3180274963378906 2023-01-21 11:07:04.853312: step: 488/529, loss: 0.012497425079345703 2023-01-21 11:07:05.974811: step: 492/529, loss: 0.01527872122824192 2023-01-21 11:07:07.097680: step: 496/529, loss: 0.036000825464725494 2023-01-21 11:07:08.249098: step: 500/529, loss: 0.03184108808636665 2023-01-21 11:07:09.369456: step: 504/529, loss: 0.014605618081986904 2023-01-21 11:07:10.483915: step: 508/529, loss: 0.0010390281677246094 2023-01-21 11:07:11.592858: step: 512/529, loss: 0.005808734800666571 2023-01-21 11:07:12.704377: step: 516/529, loss: 0.015707969665527344 2023-01-21 11:07:13.837963: step: 520/529, loss: 0.0053192139603197575 2023-01-21 11:07:14.938324: step: 524/529, loss: 0.00015954971604514867 2023-01-21 11:07:16.037049: step: 528/529, loss: 0.04520053789019585 2023-01-21 11:07:17.198125: step: 532/529, loss: 0.04092540591955185 2023-01-21 11:07:18.324752: step: 536/529, loss: 0.012979078106582165 2023-01-21 11:07:19.429479: step: 540/529, loss: 0.12213268131017685 2023-01-21 11:07:20.605321: step: 544/529, loss: 0.0021313666366040707 2023-01-21 11:07:21.706351: step: 548/529, loss: 0.007439231965690851 2023-01-21 11:07:22.786400: step: 552/529, loss: 0.006300831213593483 2023-01-21 11:07:23.902732: step: 556/529, loss: 0.0004946708795614541 2023-01-21 11:07:25.016151: step: 560/529, loss: 0.03266744688153267 2023-01-21 11:07:26.143708: step: 564/529, loss: 0.006650161929428577 2023-01-21 11:07:27.263968: step: 568/529, loss: 0.003451061435043812 2023-01-21 11:07:28.368874: step: 572/529, loss: 0.030447769910097122 2023-01-21 11:07:29.475710: step: 576/529, loss: 0.01534972246736288 2023-01-21 11:07:30.589579: step: 580/529, loss: 0.06121664121747017 2023-01-21 11:07:31.719271: step: 584/529, loss: 0.07532940059900284 2023-01-21 11:07:32.840752: step: 588/529, loss: 0.00793371256440878 2023-01-21 11:07:33.947559: step: 592/529, loss: 0.036535073071718216 2023-01-21 11:07:35.058410: step: 596/529, loss: 0.034652043133974075 2023-01-21 11:07:36.179952: step: 600/529, loss: 0.06478134542703629 2023-01-21 11:07:37.271310: step: 604/529, loss: 0.014225340448319912 2023-01-21 11:07:38.370228: step: 608/529, loss: 0.01539382990449667 2023-01-21 11:07:39.473842: step: 612/529, loss: 0.008771324530243874 2023-01-21 11:07:40.587359: step: 616/529, loss: 0.05536933243274689 2023-01-21 11:07:41.705524: step: 620/529, loss: 0.036444857716560364 2023-01-21 11:07:42.828735: step: 624/529, loss: 0.003640127135440707 2023-01-21 11:07:43.947234: step: 628/529, loss: 0.36066606640815735 2023-01-21 11:07:45.059854: step: 632/529, loss: 0.16692844033241272 2023-01-21 11:07:46.164569: step: 636/529, loss: 0.03918934240937233 2023-01-21 11:07:47.275272: step: 640/529, loss: 0.003963756840676069 2023-01-21 11:07:48.400474: step: 644/529, loss: 0.03221387788653374 2023-01-21 11:07:49.521882: step: 648/529, loss: 0.0691455826163292 2023-01-21 11:07:50.600346: step: 652/529, loss: 0.024606704711914062 2023-01-21 11:07:51.699511: step: 656/529, loss: 0.02569131925702095 2023-01-21 11:07:52.811237: step: 660/529, loss: 0.050624653697013855 2023-01-21 11:07:53.896503: step: 664/529, loss: 0.0037870886735618114 2023-01-21 11:07:55.013650: step: 668/529, loss: 0.006491470150649548 2023-01-21 11:07:56.130377: step: 672/529, loss: 0.1585306078195572 2023-01-21 11:07:57.234561: step: 676/529, loss: 0.02358388900756836 2023-01-21 11:07:58.332912: step: 680/529, loss: 0.5519237518310547 2023-01-21 11:07:59.461582: step: 684/529, loss: 0.0049301148392260075 2023-01-21 11:08:00.556191: step: 688/529, loss: 0.287584513425827 2023-01-21 11:08:01.650305: step: 692/529, loss: 0.024935245513916016 2023-01-21 11:08:02.756647: step: 696/529, loss: 0.004320716951042414 2023-01-21 11:08:03.890802: step: 700/529, loss: 0.0030488967895507812 2023-01-21 11:08:05.020831: step: 704/529, loss: 0.03374214097857475 2023-01-21 11:08:06.135673: step: 708/529, loss: 0.03261518478393555 2023-01-21 11:08:07.262454: step: 712/529, loss: 0.043166350573301315 2023-01-21 11:08:08.365364: step: 716/529, loss: 0.009520197287201881 2023-01-21 11:08:09.503153: step: 720/529, loss: 0.002094268798828125 2023-01-21 11:08:10.656732: step: 724/529, loss: 0.07955612987279892 2023-01-21 11:08:11.786048: step: 728/529, loss: 0.002158737275749445 2023-01-21 11:08:12.895286: step: 732/529, loss: 0.0661383643746376 2023-01-21 11:08:13.995523: step: 736/529, loss: 0.028620148077607155 2023-01-21 11:08:15.100563: step: 740/529, loss: 0.004321098327636719 2023-01-21 11:08:16.213175: step: 744/529, loss: 0.06198091432452202 2023-01-21 11:08:17.312424: step: 748/529, loss: 0.01991443708539009 2023-01-21 11:08:18.404599: step: 752/529, loss: 0.021515274420380592 2023-01-21 11:08:19.500941: step: 756/529, loss: 0.005078697111457586 2023-01-21 11:08:20.596201: step: 760/529, loss: 0.046824172139167786 2023-01-21 11:08:21.706818: step: 764/529, loss: 0.027736950665712357 2023-01-21 11:08:22.822132: step: 768/529, loss: 0.0024580953177064657 2023-01-21 11:08:23.966911: step: 772/529, loss: 0.019012451171875 2023-01-21 11:08:25.082590: step: 776/529, loss: 0.017098141834139824 2023-01-21 11:08:26.173602: step: 780/529, loss: 0.026880167424678802 2023-01-21 11:08:27.334834: step: 784/529, loss: 0.00014579296112060547 2023-01-21 11:08:28.467301: step: 788/529, loss: 0.02515239641070366 2023-01-21 11:08:29.648010: step: 792/529, loss: 0.024906732141971588 2023-01-21 11:08:30.755415: step: 796/529, loss: 0.0066581727005541325 2023-01-21 11:08:31.873021: step: 800/529, loss: 0.017708491533994675 2023-01-21 11:08:32.997255: step: 804/529, loss: 0.06911754608154297 2023-01-21 11:08:34.103944: step: 808/529, loss: 0.0449383519589901 2023-01-21 11:08:35.218180: step: 812/529, loss: 0.15337209403514862 2023-01-21 11:08:36.315490: step: 816/529, loss: 0.0415617935359478 2023-01-21 11:08:37.417930: step: 820/529, loss: 0.04999237135052681 2023-01-21 11:08:38.539902: step: 824/529, loss: 0.09302043914794922 2023-01-21 11:08:39.678742: step: 828/529, loss: 0.004516792483627796 2023-01-21 11:08:40.774549: step: 832/529, loss: 0.048300981521606445 2023-01-21 11:08:41.900593: step: 836/529, loss: 0.005957460962235928 2023-01-21 11:08:43.008005: step: 840/529, loss: 0.006057166960090399 2023-01-21 11:08:44.138707: step: 844/529, loss: 0.004037285223603249 2023-01-21 11:08:45.250034: step: 848/529, loss: 0.06434802711009979 2023-01-21 11:08:46.386189: step: 852/529, loss: 0.002489280654117465 2023-01-21 11:08:47.490169: step: 856/529, loss: 0.23927058279514313 2023-01-21 11:08:48.604982: step: 860/529, loss: 0.005560684017837048 2023-01-21 11:08:49.722281: step: 864/529, loss: 0.043326377868652344 2023-01-21 11:08:50.843331: step: 868/529, loss: 0.007733345031738281 2023-01-21 11:08:51.950970: step: 872/529, loss: 0.07311849296092987 2023-01-21 11:08:53.072981: step: 876/529, loss: 3.06129441014491e-05 2023-01-21 11:08:54.197127: step: 880/529, loss: 0.00482940673828125 2023-01-21 11:08:55.341606: step: 884/529, loss: 0.08264847099781036 2023-01-21 11:08:56.450990: step: 888/529, loss: 0.0009354114881716669 2023-01-21 11:08:57.603881: step: 892/529, loss: 0.15960197150707245 2023-01-21 11:08:58.723041: step: 896/529, loss: 0.004975604824721813 2023-01-21 11:08:59.832729: step: 900/529, loss: 0.054918862879276276 2023-01-21 11:09:00.908167: step: 904/529, loss: 0.01359262503683567 2023-01-21 11:09:02.045149: step: 908/529, loss: 0.05474882200360298 2023-01-21 11:09:03.153155: step: 912/529, loss: 0.003283691592514515 2023-01-21 11:09:04.272890: step: 916/529, loss: 0.00036773679312318563 2023-01-21 11:09:05.398021: step: 920/529, loss: 0.0027414322830736637 2023-01-21 11:09:06.520160: step: 924/529, loss: 0.024626540020108223 2023-01-21 11:09:07.635169: step: 928/529, loss: 0.0008567810291424394 2023-01-21 11:09:08.741214: step: 932/529, loss: 0.060804370790719986 2023-01-21 11:09:09.849688: step: 936/529, loss: 0.006113338749855757 2023-01-21 11:09:10.933665: step: 940/529, loss: 0.04157848656177521 2023-01-21 11:09:12.055569: step: 944/529, loss: 0.01965959183871746 2023-01-21 11:09:13.189348: step: 948/529, loss: 0.27173954248428345 2023-01-21 11:09:14.310766: step: 952/529, loss: 0.20857134461402893 2023-01-21 11:09:15.434998: step: 956/529, loss: 0.008108711801469326 2023-01-21 11:09:16.561560: step: 960/529, loss: 0.11366715282201767 2023-01-21 11:09:17.692029: step: 964/529, loss: 0.02399921603500843 2023-01-21 11:09:18.819002: step: 968/529, loss: 0.028089476749300957 2023-01-21 11:09:19.925352: step: 972/529, loss: 0.11822175979614258 2023-01-21 11:09:21.002765: step: 976/529, loss: 0.01480865478515625 2023-01-21 11:09:22.128698: step: 980/529, loss: 0.017039109021425247 2023-01-21 11:09:23.238863: step: 984/529, loss: 0.036610983312129974 2023-01-21 11:09:24.336368: step: 988/529, loss: 0.006156873889267445 2023-01-21 11:09:25.456537: step: 992/529, loss: 0.008353233337402344 2023-01-21 11:09:26.571897: step: 996/529, loss: 0.15275651216506958 2023-01-21 11:09:27.719765: step: 1000/529, loss: 0.022661972790956497 2023-01-21 11:09:28.803577: step: 1004/529, loss: 0.005509567447006702 2023-01-21 11:09:29.942916: step: 1008/529, loss: 0.0036905291490256786 2023-01-21 11:09:31.051439: step: 1012/529, loss: 0.01190652884542942 2023-01-21 11:09:32.141739: step: 1016/529, loss: 0.0016812323592603207 2023-01-21 11:09:33.243668: step: 1020/529, loss: 0.0014493941562250257 2023-01-21 11:09:34.368532: step: 1024/529, loss: 0.01684284210205078 2023-01-21 11:09:35.515240: step: 1028/529, loss: 0.009927082806825638 2023-01-21 11:09:36.633943: step: 1032/529, loss: 0.15844497084617615 2023-01-21 11:09:37.776339: step: 1036/529, loss: 1.012674331665039 2023-01-21 11:09:38.866700: step: 1040/529, loss: 0.05566806718707085 2023-01-21 11:09:39.968075: step: 1044/529, loss: 0.04399833828210831 2023-01-21 11:09:41.101528: step: 1048/529, loss: 0.19140681624412537 2023-01-21 11:09:42.202549: step: 1052/529, loss: 0.010280419141054153 2023-01-21 11:09:43.346673: step: 1056/529, loss: 0.13566331565380096 2023-01-21 11:09:44.472899: step: 1060/529, loss: 0.0197919849306345 2023-01-21 11:09:45.596855: step: 1064/529, loss: 0.12768897414207458 2023-01-21 11:09:46.712127: step: 1068/529, loss: 0.023880768567323685 2023-01-21 11:09:47.820607: step: 1072/529, loss: 2.784109592437744 2023-01-21 11:09:48.921983: step: 1076/529, loss: 0.014256644994020462 2023-01-21 11:09:50.027032: step: 1080/529, loss: 0.03602752462029457 2023-01-21 11:09:51.152344: step: 1084/529, loss: 0.024809647351503372 2023-01-21 11:09:52.261169: step: 1088/529, loss: 0.017095183953642845 2023-01-21 11:09:53.389869: step: 1092/529, loss: 0.061055686324834824 2023-01-21 11:09:54.494780: step: 1096/529, loss: 0.07091826945543289 2023-01-21 11:09:55.595937: step: 1100/529, loss: 0.013196468353271484 2023-01-21 11:09:56.703595: step: 1104/529, loss: 0.0017759323818609118 2023-01-21 11:09:57.789121: step: 1108/529, loss: 0.0013233184581622481 2023-01-21 11:09:58.912389: step: 1112/529, loss: 0.012111281976103783 2023-01-21 11:10:00.021912: step: 1116/529, loss: 0.044538021087646484 2023-01-21 11:10:01.130635: step: 1120/529, loss: 0.03583526611328125 2023-01-21 11:10:02.268136: step: 1124/529, loss: 0.006453681271523237 2023-01-21 11:10:03.377305: step: 1128/529, loss: 0.05612201616168022 2023-01-21 11:10:04.544714: step: 1132/529, loss: 0.0349210724234581 2023-01-21 11:10:05.682766: step: 1136/529, loss: 0.015649283304810524 2023-01-21 11:10:06.796892: step: 1140/529, loss: 0.07226867973804474 2023-01-21 11:10:07.911888: step: 1144/529, loss: 0.8925444483757019 2023-01-21 11:10:09.048464: step: 1148/529, loss: 0.00380878453142941 2023-01-21 11:10:10.190362: step: 1152/529, loss: 0.03990011289715767 2023-01-21 11:10:11.324476: step: 1156/529, loss: 0.005954170133918524 2023-01-21 11:10:12.442696: step: 1160/529, loss: 0.001020240830257535 2023-01-21 11:10:13.560669: step: 1164/529, loss: 0.33216121792793274 2023-01-21 11:10:14.679687: step: 1168/529, loss: 0.017474651336669922 2023-01-21 11:10:15.827979: step: 1172/529, loss: 0.04607431963086128 2023-01-21 11:10:16.979212: step: 1176/529, loss: 0.006637764163315296 2023-01-21 11:10:18.099333: step: 1180/529, loss: 0.060863494873046875 2023-01-21 11:10:19.209546: step: 1184/529, loss: 0.0466187484562397 2023-01-21 11:10:20.331266: step: 1188/529, loss: 0.014934349805116653 2023-01-21 11:10:21.453021: step: 1192/529, loss: 0.021178342401981354 2023-01-21 11:10:22.574306: step: 1196/529, loss: 0.009847735986113548 2023-01-21 11:10:23.683877: step: 1200/529, loss: 0.012802696786820889 2023-01-21 11:10:24.788926: step: 1204/529, loss: 0.0938815101981163 2023-01-21 11:10:25.887894: step: 1208/529, loss: 0.1923181563615799 2023-01-21 11:10:27.031498: step: 1212/529, loss: 0.03403129428625107 2023-01-21 11:10:28.135950: step: 1216/529, loss: 0.0035559178795665503 2023-01-21 11:10:29.255206: step: 1220/529, loss: 0.13441739976406097 2023-01-21 11:10:30.370808: step: 1224/529, loss: 0.014824485406279564 2023-01-21 11:10:31.474592: step: 1228/529, loss: 0.0183365810662508 2023-01-21 11:10:32.594122: step: 1232/529, loss: 0.028984833508729935 2023-01-21 11:10:33.741741: step: 1236/529, loss: 0.04728889465332031 2023-01-21 11:10:34.878295: step: 1240/529, loss: 0.013409232720732689 2023-01-21 11:10:35.970432: step: 1244/529, loss: 0.21374927461147308 2023-01-21 11:10:37.110901: step: 1248/529, loss: 0.06234750896692276 2023-01-21 11:10:38.204616: step: 1252/529, loss: 0.003662490751594305 2023-01-21 11:10:39.338847: step: 1256/529, loss: 0.05922403559088707 2023-01-21 11:10:40.424233: step: 1260/529, loss: 0.06785106658935547 2023-01-21 11:10:41.525051: step: 1264/529, loss: 0.004977608099579811 2023-01-21 11:10:42.625472: step: 1268/529, loss: 0.003946209326386452 2023-01-21 11:10:43.732283: step: 1272/529, loss: 0.013188743032515049 2023-01-21 11:10:44.853497: step: 1276/529, loss: 0.027324475347995758 2023-01-21 11:10:45.991633: step: 1280/529, loss: 0.012353707104921341 2023-01-21 11:10:47.111703: step: 1284/529, loss: 0.08095379173755646 2023-01-21 11:10:48.228530: step: 1288/529, loss: 0.010207748971879482 2023-01-21 11:10:49.334071: step: 1292/529, loss: 0.008919095620512962 2023-01-21 11:10:50.437948: step: 1296/529, loss: 0.001168060232885182 2023-01-21 11:10:51.556413: step: 1300/529, loss: 0.015664292499423027 2023-01-21 11:10:52.679403: step: 1304/529, loss: 0.03494663164019585 2023-01-21 11:10:53.791849: step: 1308/529, loss: 0.01756284199655056 2023-01-21 11:10:54.950079: step: 1312/529, loss: 0.011372661218047142 2023-01-21 11:10:56.044342: step: 1316/529, loss: 0.01863689348101616 2023-01-21 11:10:57.159045: step: 1320/529, loss: 0.005563593469560146 2023-01-21 11:10:58.325266: step: 1324/529, loss: 0.025613784790039062 2023-01-21 11:10:59.447498: step: 1328/529, loss: 0.05599260330200195 2023-01-21 11:11:00.591442: step: 1332/529, loss: 0.012350654229521751 2023-01-21 11:11:01.708765: step: 1336/529, loss: 0.10752163082361221 2023-01-21 11:11:02.825064: step: 1340/529, loss: 0.044048357754945755 2023-01-21 11:11:03.984435: step: 1344/529, loss: 0.04717312008142471 2023-01-21 11:11:05.079591: step: 1348/529, loss: 0.025150252506136894 2023-01-21 11:11:06.210193: step: 1352/529, loss: 0.01689434051513672 2023-01-21 11:11:07.321074: step: 1356/529, loss: 0.5690412521362305 2023-01-21 11:11:08.486215: step: 1360/529, loss: 0.11586761474609375 2023-01-21 11:11:09.618999: step: 1364/529, loss: 0.05314836651086807 2023-01-21 11:11:10.767456: step: 1368/529, loss: 0.04475117102265358 2023-01-21 11:11:11.910247: step: 1372/529, loss: 0.004604530055075884 2023-01-21 11:11:13.028316: step: 1376/529, loss: 0.01128406636416912 2023-01-21 11:11:14.135474: step: 1380/529, loss: 0.031459711492061615 2023-01-21 11:11:15.233283: step: 1384/529, loss: 0.030724240466952324 2023-01-21 11:11:16.333986: step: 1388/529, loss: 0.02933340147137642 2023-01-21 11:11:17.447583: step: 1392/529, loss: 0.04226074367761612 2023-01-21 11:11:18.602115: step: 1396/529, loss: 0.0802769660949707 2023-01-21 11:11:19.725140: step: 1400/529, loss: 0.029257584363222122 2023-01-21 11:11:20.848749: step: 1404/529, loss: 0.03714311122894287 2023-01-21 11:11:21.976254: step: 1408/529, loss: 0.003967666532844305 2023-01-21 11:11:23.090174: step: 1412/529, loss: 0.008970069698989391 2023-01-21 11:11:24.196122: step: 1416/529, loss: 0.00018949508375953883 2023-01-21 11:11:25.281805: step: 1420/529, loss: 0.0017972945934161544 2023-01-21 11:11:26.423040: step: 1424/529, loss: 0.05047006532549858 2023-01-21 11:11:27.569314: step: 1428/529, loss: 0.01947307586669922 2023-01-21 11:11:28.762346: step: 1432/529, loss: 0.05660073459148407 2023-01-21 11:11:29.889327: step: 1436/529, loss: 0.0013745308388024569 2023-01-21 11:11:31.004258: step: 1440/529, loss: 0.060073185712099075 2023-01-21 11:11:32.128269: step: 1444/529, loss: 0.010673332959413528 2023-01-21 11:11:33.263236: step: 1448/529, loss: 0.034828949719667435 2023-01-21 11:11:34.392007: step: 1452/529, loss: 0.07121066749095917 2023-01-21 11:11:35.542208: step: 1456/529, loss: 0.038569070398807526 2023-01-21 11:11:36.661542: step: 1460/529, loss: 0.04711113125085831 2023-01-21 11:11:37.780616: step: 1464/529, loss: 0.02956237830221653 2023-01-21 11:11:38.950069: step: 1468/529, loss: 0.0025766375474631786 2023-01-21 11:11:40.060033: step: 1472/529, loss: 0.012780284509062767 2023-01-21 11:11:41.152037: step: 1476/529, loss: 0.03177938610315323 2023-01-21 11:11:42.288052: step: 1480/529, loss: 0.05872516706585884 2023-01-21 11:11:43.393895: step: 1484/529, loss: 0.0188769344240427 2023-01-21 11:11:44.511097: step: 1488/529, loss: 0.022259997203946114 2023-01-21 11:11:45.619262: step: 1492/529, loss: 0.00033202170743606985 2023-01-21 11:11:46.744639: step: 1496/529, loss: 0.029712343588471413 2023-01-21 11:11:47.867118: step: 1500/529, loss: 0.04746971279382706 2023-01-21 11:11:49.003212: step: 1504/529, loss: 0.02252502366900444 2023-01-21 11:11:50.128987: step: 1508/529, loss: 0.04196300730109215 2023-01-21 11:11:51.306246: step: 1512/529, loss: 0.034624673426151276 2023-01-21 11:11:52.434180: step: 1516/529, loss: 0.007928848266601562 2023-01-21 11:11:53.557936: step: 1520/529, loss: 0.0027473450172692537 2023-01-21 11:11:54.654218: step: 1524/529, loss: 0.020561981946229935 2023-01-21 11:11:55.755062: step: 1528/529, loss: 0.005965805146843195 2023-01-21 11:11:56.876556: step: 1532/529, loss: 0.08079500496387482 2023-01-21 11:11:58.016922: step: 1536/529, loss: 0.11873187869787216 2023-01-21 11:11:59.148032: step: 1540/529, loss: 0.28856000304222107 2023-01-21 11:12:00.293740: step: 1544/529, loss: 0.0011587142944335938 2023-01-21 11:12:01.403221: step: 1548/529, loss: 0.0003411293146200478 2023-01-21 11:12:02.505856: step: 1552/529, loss: 0.0868481695652008 2023-01-21 11:12:03.592128: step: 1556/529, loss: 0.01118860300630331 2023-01-21 11:12:04.708144: step: 1560/529, loss: 0.0008830070728436112 2023-01-21 11:12:05.835962: step: 1564/529, loss: 0.009162998758256435 2023-01-21 11:12:06.936880: step: 1568/529, loss: 0.005147743504494429 2023-01-21 11:12:08.078605: step: 1572/529, loss: 0.0026901247911155224 2023-01-21 11:12:09.211267: step: 1576/529, loss: 0.010972786694765091 2023-01-21 11:12:10.321080: step: 1580/529, loss: 0.05199165269732475 2023-01-21 11:12:11.457588: step: 1584/529, loss: 0.14903374016284943 2023-01-21 11:12:12.601356: step: 1588/529, loss: 0.06888733059167862 2023-01-21 11:12:13.721999: step: 1592/529, loss: 0.026501081883907318 2023-01-21 11:12:14.826475: step: 1596/529, loss: 0.02939748764038086 2023-01-21 11:12:15.951257: step: 1600/529, loss: 0.023587321862578392 2023-01-21 11:12:17.094262: step: 1604/529, loss: 0.02911682054400444 2023-01-21 11:12:18.200995: step: 1608/529, loss: 0.004750251770019531 2023-01-21 11:12:19.343249: step: 1612/529, loss: 0.022888565436005592 2023-01-21 11:12:20.491264: step: 1616/529, loss: 0.05539379268884659 2023-01-21 11:12:21.609627: step: 1620/529, loss: 0.019959401339292526 2023-01-21 11:12:22.686069: step: 1624/529, loss: 0.04987926408648491 2023-01-21 11:12:23.790295: step: 1628/529, loss: 0.00900125503540039 2023-01-21 11:12:24.914930: step: 1632/529, loss: 0.0690908432006836 2023-01-21 11:12:26.051403: step: 1636/529, loss: 0.012089347466826439 2023-01-21 11:12:27.174556: step: 1640/529, loss: 0.013158036395907402 2023-01-21 11:12:28.253956: step: 1644/529, loss: 0.007267570123076439 2023-01-21 11:12:29.373268: step: 1648/529, loss: 0.009405136108398438 2023-01-21 11:12:30.503684: step: 1652/529, loss: 0.038481712341308594 2023-01-21 11:12:31.637656: step: 1656/529, loss: 0.05638699233531952 2023-01-21 11:12:32.711642: step: 1660/529, loss: 0.04726886749267578 2023-01-21 11:12:33.837620: step: 1664/529, loss: 0.13326320052146912 2023-01-21 11:12:34.976052: step: 1668/529, loss: 0.04291302338242531 2023-01-21 11:12:36.073938: step: 1672/529, loss: 0.08542804419994354 2023-01-21 11:12:37.214075: step: 1676/529, loss: 0.006975841708481312 2023-01-21 11:12:38.306072: step: 1680/529, loss: 4.4417382014216855e-05 2023-01-21 11:12:39.422174: step: 1684/529, loss: 0.05392418056726456 2023-01-21 11:12:40.532150: step: 1688/529, loss: 0.00897512398660183 2023-01-21 11:12:41.646923: step: 1692/529, loss: 0.008361482992768288 2023-01-21 11:12:42.774662: step: 1696/529, loss: 0.3334774971008301 2023-01-21 11:12:43.870150: step: 1700/529, loss: 0.09300556033849716 2023-01-21 11:12:44.994837: step: 1704/529, loss: 0.028688624501228333 2023-01-21 11:12:46.084887: step: 1708/529, loss: 0.0024240135680884123 2023-01-21 11:12:47.198714: step: 1712/529, loss: 0.07397708296775818 2023-01-21 11:12:48.338866: step: 1716/529, loss: 0.2645646035671234 2023-01-21 11:12:49.461320: step: 1720/529, loss: 0.02503833919763565 2023-01-21 11:12:50.567145: step: 1724/529, loss: 0.004874467849731445 2023-01-21 11:12:51.676523: step: 1728/529, loss: 7.82012939453125e-05 2023-01-21 11:12:52.773988: step: 1732/529, loss: 0.8507784008979797 2023-01-21 11:12:53.907775: step: 1736/529, loss: 0.1318719983100891 2023-01-21 11:12:55.041757: step: 1740/529, loss: 0.021417807787656784 2023-01-21 11:12:56.147821: step: 1744/529, loss: 0.0011737822787836194 2023-01-21 11:12:57.257190: step: 1748/529, loss: 0.018645858392119408 2023-01-21 11:12:58.396701: step: 1752/529, loss: 0.02441110834479332 2023-01-21 11:12:59.537451: step: 1756/529, loss: 0.07416267693042755 2023-01-21 11:13:00.643873: step: 1760/529, loss: 0.016349315643310547 2023-01-21 11:13:01.770544: step: 1764/529, loss: 0.03471078723669052 2023-01-21 11:13:02.891117: step: 1768/529, loss: 0.03573598712682724 2023-01-21 11:13:04.012133: step: 1772/529, loss: 0.009135628119111061 2023-01-21 11:13:05.131184: step: 1776/529, loss: 0.030147649347782135 2023-01-21 11:13:06.279311: step: 1780/529, loss: 0.0015956878196448088 2023-01-21 11:13:07.380331: step: 1784/529, loss: 0.3427744507789612 2023-01-21 11:13:08.492793: step: 1788/529, loss: 0.02135854959487915 2023-01-21 11:13:09.581249: step: 1792/529, loss: 0.027435969561338425 2023-01-21 11:13:10.690128: step: 1796/529, loss: 0.039005398750305176 2023-01-21 11:13:11.816195: step: 1800/529, loss: 0.022168828174471855 2023-01-21 11:13:12.929241: step: 1804/529, loss: 0.04677620157599449 2023-01-21 11:13:14.016001: step: 1808/529, loss: 0.03503427654504776 2023-01-21 11:13:15.114338: step: 1812/529, loss: 0.010628987103700638 2023-01-21 11:13:16.220901: step: 1816/529, loss: 0.019567154347896576 2023-01-21 11:13:17.326790: step: 1820/529, loss: 0.01189203280955553 2023-01-21 11:13:18.477615: step: 1824/529, loss: 0.0355776771903038 2023-01-21 11:13:19.608005: step: 1828/529, loss: 0.013114642351865768 2023-01-21 11:13:20.728823: step: 1832/529, loss: 0.027872467413544655 2023-01-21 11:13:21.860028: step: 1836/529, loss: 0.13127270340919495 2023-01-21 11:13:23.001671: step: 1840/529, loss: 0.007633971981704235 2023-01-21 11:13:24.109377: step: 1844/529, loss: 0.03973674774169922 2023-01-21 11:13:25.200562: step: 1848/529, loss: 0.014048004522919655 2023-01-21 11:13:26.337761: step: 1852/529, loss: 0.011321032419800758 2023-01-21 11:13:27.472011: step: 1856/529, loss: 0.006206703372299671 2023-01-21 11:13:28.583001: step: 1860/529, loss: 0.13167135417461395 2023-01-21 11:13:29.692531: step: 1864/529, loss: 0.06124553829431534 2023-01-21 11:13:30.803585: step: 1868/529, loss: 0.01782989501953125 2023-01-21 11:13:31.899645: step: 1872/529, loss: 0.04866466671228409 2023-01-21 11:13:33.004749: step: 1876/529, loss: 0.006202983669936657 2023-01-21 11:13:34.118749: step: 1880/529, loss: 0.011512374505400658 2023-01-21 11:13:35.211396: step: 1884/529, loss: 0.004651069641113281 2023-01-21 11:13:36.309700: step: 1888/529, loss: 0.013984394259750843 2023-01-21 11:13:37.411206: step: 1892/529, loss: 0.05618901178240776 2023-01-21 11:13:38.512516: step: 1896/529, loss: 0.014040661044418812 2023-01-21 11:13:39.607649: step: 1900/529, loss: 0.03840827941894531 2023-01-21 11:13:40.712629: step: 1904/529, loss: 0.040229037404060364 2023-01-21 11:13:41.838374: step: 1908/529, loss: 0.019501496106386185 2023-01-21 11:13:42.951071: step: 1912/529, loss: 0.23917751014232635 2023-01-21 11:13:44.090784: step: 1916/529, loss: 0.11080218106508255 2023-01-21 11:13:45.205237: step: 1920/529, loss: 0.08093462139368057 2023-01-21 11:13:46.352349: step: 1924/529, loss: 0.061968617141246796 2023-01-21 11:13:47.467638: step: 1928/529, loss: 0.012190151028335094 2023-01-21 11:13:48.560135: step: 1932/529, loss: 0.04579916223883629 2023-01-21 11:13:49.701819: step: 1936/529, loss: 0.05415744706988335 2023-01-21 11:13:50.835092: step: 1940/529, loss: 0.024793434888124466 2023-01-21 11:13:51.973023: step: 1944/529, loss: 0.036486051976680756 2023-01-21 11:13:53.104795: step: 1948/529, loss: 0.021683646366000175 2023-01-21 11:13:54.195705: step: 1952/529, loss: 0.10531672835350037 2023-01-21 11:13:55.348826: step: 1956/529, loss: 0.04025259241461754 2023-01-21 11:13:56.488481: step: 1960/529, loss: 0.005909156985580921 2023-01-21 11:13:57.580039: step: 1964/529, loss: 0.029218867421150208 2023-01-21 11:13:58.733383: step: 1968/529, loss: 0.023702239617705345 2023-01-21 11:13:59.861610: step: 1972/529, loss: 0.012431717477738857 2023-01-21 11:14:00.979934: step: 1976/529, loss: 0.012362003326416016 2023-01-21 11:14:02.112154: step: 1980/529, loss: 0.10852088779211044 2023-01-21 11:14:03.221060: step: 1984/529, loss: 0.03397426754236221 2023-01-21 11:14:04.325099: step: 1988/529, loss: 0.017452050000429153 2023-01-21 11:14:05.443903: step: 1992/529, loss: 0.005427340045571327 2023-01-21 11:14:06.573161: step: 1996/529, loss: 0.05447263643145561 2023-01-21 11:14:07.695144: step: 2000/529, loss: 0.05303196981549263 2023-01-21 11:14:08.811313: step: 2004/529, loss: 0.029610633850097656 2023-01-21 11:14:09.903452: step: 2008/529, loss: 0.01581287384033203 2023-01-21 11:14:11.018791: step: 2012/529, loss: 0.01895303837954998 2023-01-21 11:14:12.138652: step: 2016/529, loss: 0.012185860425233841 2023-01-21 11:14:13.231301: step: 2020/529, loss: 0.03466777876019478 2023-01-21 11:14:14.365060: step: 2024/529, loss: 0.05949373543262482 2023-01-21 11:14:15.498691: step: 2028/529, loss: 0.000171661376953125 2023-01-21 11:14:16.623608: step: 2032/529, loss: 0.005026054568588734 2023-01-21 11:14:17.724374: step: 2036/529, loss: 0.0004886210081167519 2023-01-21 11:14:18.827672: step: 2040/529, loss: 0.01560983620584011 2023-01-21 11:14:19.941644: step: 2044/529, loss: 0.04823331907391548 2023-01-21 11:14:21.075983: step: 2048/529, loss: 0.013470268808305264 2023-01-21 11:14:22.191825: step: 2052/529, loss: 0.05151491239666939 2023-01-21 11:14:23.346829: step: 2056/529, loss: 0.024869920685887337 2023-01-21 11:14:24.462795: step: 2060/529, loss: 0.020335961133241653 2023-01-21 11:14:25.596230: step: 2064/529, loss: 0.01641559600830078 2023-01-21 11:14:26.708468: step: 2068/529, loss: 0.012571334838867188 2023-01-21 11:14:27.825461: step: 2072/529, loss: 0.01003799494355917 2023-01-21 11:14:28.961909: step: 2076/529, loss: 0.044074155390262604 2023-01-21 11:14:30.061759: step: 2080/529, loss: 0.023874282836914062 2023-01-21 11:14:31.189867: step: 2084/529, loss: 0.04250383377075195 2023-01-21 11:14:32.307019: step: 2088/529, loss: 0.036631397902965546 2023-01-21 11:14:33.423303: step: 2092/529, loss: 0.06407790631055832 2023-01-21 11:14:34.543756: step: 2096/529, loss: 0.001379680703394115 2023-01-21 11:14:35.691998: step: 2100/529, loss: 0.05193347856402397 2023-01-21 11:14:36.824437: step: 2104/529, loss: 0.03535137325525284 2023-01-21 11:14:37.970298: step: 2108/529, loss: 0.031992726027965546 2023-01-21 11:14:39.073180: step: 2112/529, loss: 0.002292537596076727 2023-01-21 11:14:40.208788: step: 2116/529, loss: 0.024569178000092506 ================================================== Loss: 0.053 -------------------- Dev: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Test: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Chinese: {'event': {'p': 0.5465116279069767, 'r': 0.8703703703703703, 'f1': 0.6714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} 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': 15} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 16 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:15:27.393072: step: 4/529, loss: 0.01759509928524494 2023-01-21 11:15:28.513335: step: 8/529, loss: 0.005558204837143421 2023-01-21 11:15:29.613495: step: 12/529, loss: 0.002117633819580078 2023-01-21 11:15:30.737951: step: 16/529, loss: 0.008744430728256702 2023-01-21 11:15:31.836212: step: 20/529, loss: 0.008201027289032936 2023-01-21 11:15:32.954368: step: 24/529, loss: 0.0010945319663733244 2023-01-21 11:15:34.070787: step: 28/529, loss: 0.0016349792713299394 2023-01-21 11:15:35.161782: step: 32/529, loss: 0.013190556317567825 2023-01-21 11:15:36.251601: step: 36/529, loss: 0.01243896596133709 2023-01-21 11:15:37.366155: step: 40/529, loss: 0.005727005191147327 2023-01-21 11:15:38.461854: step: 44/529, loss: 0.007983779534697533 2023-01-21 11:15:39.591897: step: 48/529, loss: 0.0060284677892923355 2023-01-21 11:15:40.708337: step: 52/529, loss: 0.0008068084716796875 2023-01-21 11:15:41.790387: step: 56/529, loss: 0.010333872400224209 2023-01-21 11:15:42.867225: step: 60/529, loss: 0.005064924713224173 2023-01-21 11:15:43.975566: step: 64/529, loss: 0.019730854779481888 2023-01-21 11:15:45.072598: step: 68/529, loss: 0.0005346297984942794 2023-01-21 11:15:46.216917: step: 72/529, loss: 0.010563087649643421 2023-01-21 11:15:47.318321: step: 76/529, loss: 0.043862439692020416 2023-01-21 11:15:48.437130: step: 80/529, loss: 0.04054107517004013 2023-01-21 11:15:49.530479: step: 84/529, loss: 0.025189924985170364 2023-01-21 11:15:50.657961: step: 88/529, loss: 0.0018634796142578125 2023-01-21 11:15:51.789170: step: 92/529, loss: 0.002568149706348777 2023-01-21 11:15:52.910457: step: 96/529, loss: 0.04394207149744034 2023-01-21 11:15:54.063595: step: 100/529, loss: 0.059703826904296875 2023-01-21 11:15:55.186308: step: 104/529, loss: 0.006520175840705633 2023-01-21 11:15:56.312758: step: 108/529, loss: 0.006911659147590399 2023-01-21 11:15:57.413772: step: 112/529, loss: 0.4324420392513275 2023-01-21 11:15:58.530739: step: 116/529, loss: 0.00030512810917571187 2023-01-21 11:15:59.616975: step: 120/529, loss: 0.04342012479901314 2023-01-21 11:16:00.731580: step: 124/529, loss: 0.007519531529396772 2023-01-21 11:16:01.855373: step: 128/529, loss: 0.006271171383559704 2023-01-21 11:16:02.979861: step: 132/529, loss: 0.07997055351734161 2023-01-21 11:16:04.088199: step: 136/529, loss: 0.03751564025878906 2023-01-21 11:16:05.182409: step: 140/529, loss: 0.011118745431303978 2023-01-21 11:16:06.293716: step: 144/529, loss: 0.019529057666659355 2023-01-21 11:16:07.413353: step: 148/529, loss: 0.004872131161391735 2023-01-21 11:16:08.544856: step: 152/529, loss: 0.030228614807128906 2023-01-21 11:16:09.669501: step: 156/529, loss: 0.01074142474681139 2023-01-21 11:16:10.781961: step: 160/529, loss: 0.0026817324105650187 2023-01-21 11:16:11.892238: step: 164/529, loss: 0.06483488529920578 2023-01-21 11:16:13.006687: step: 168/529, loss: 0.0009596824529580772 2023-01-21 11:16:14.124530: step: 172/529, loss: 0.01357660349458456 2023-01-21 11:16:15.230010: step: 176/529, loss: 0.012223434634506702 2023-01-21 11:16:16.373906: step: 180/529, loss: 0.0004608154413290322 2023-01-21 11:16:17.528518: step: 184/529, loss: 0.005642223171889782 2023-01-21 11:16:18.633709: step: 188/529, loss: 0.027587413787841797 2023-01-21 11:16:19.742176: step: 192/529, loss: 0.0011214256519451737 2023-01-21 11:16:20.867251: step: 196/529, loss: 0.04644327238202095 2023-01-21 11:16:21.963809: step: 200/529, loss: 0.020015234127640724 2023-01-21 11:16:23.087367: step: 204/529, loss: 0.005170631222426891 2023-01-21 11:16:24.214131: step: 208/529, loss: 0.08247242122888565 2023-01-21 11:16:25.354153: step: 212/529, loss: 0.0524735189974308 2023-01-21 11:16:26.496329: step: 216/529, loss: 0.005427646916359663 2023-01-21 11:16:27.602108: step: 220/529, loss: 0.07797518372535706 2023-01-21 11:16:28.711677: step: 224/529, loss: 0.015720367431640625 2023-01-21 11:16:29.825183: step: 228/529, loss: 0.0032287598587572575 2023-01-21 11:16:30.919272: step: 232/529, loss: 0.01717534102499485 2023-01-21 11:16:32.033995: step: 236/529, loss: 0.07521457970142365 2023-01-21 11:16:33.161957: step: 240/529, loss: 0.030816268175840378 2023-01-21 11:16:34.250316: step: 244/529, loss: 0.0008546352619305253 2023-01-21 11:16:35.405787: step: 248/529, loss: 0.02214374579489231 2023-01-21 11:16:36.533523: step: 252/529, loss: 0.0053382874466478825 2023-01-21 11:16:37.664278: step: 256/529, loss: 0.01530303992331028 2023-01-21 11:16:38.786927: step: 260/529, loss: 0.07069454342126846 2023-01-21 11:16:39.906398: step: 264/529, loss: 0.010964298620820045 2023-01-21 11:16:40.995673: step: 268/529, loss: 0.0720447525382042 2023-01-21 11:16:42.087665: step: 272/529, loss: 0.07317724078893661 2023-01-21 11:16:43.187220: step: 276/529, loss: 0.01610383950173855 2023-01-21 11:16:44.325188: step: 280/529, loss: 7.224082946777344e-05 2023-01-21 11:16:45.432721: step: 284/529, loss: 0.047832489013671875 2023-01-21 11:16:46.540743: step: 288/529, loss: 9.632110595703125e-05 2023-01-21 11:16:47.639037: step: 292/529, loss: 6.48498553346144e-06 2023-01-21 11:16:48.737637: step: 296/529, loss: 0.018159102648496628 2023-01-21 11:16:49.864532: step: 300/529, loss: 0.00666003255173564 2023-01-21 11:16:50.978607: step: 304/529, loss: 0.00021076203847769648 2023-01-21 11:16:52.091904: step: 308/529, loss: 0.04769439622759819 2023-01-21 11:16:53.182506: step: 312/529, loss: 0.0011162757873535156 2023-01-21 11:16:54.278030: step: 316/529, loss: 0.007105064578354359 2023-01-21 11:16:55.404370: step: 320/529, loss: 0.011291789822280407 2023-01-21 11:16:56.524163: step: 324/529, loss: 0.025879669934511185 2023-01-21 11:16:57.658144: step: 328/529, loss: 0.1660911589860916 2023-01-21 11:16:58.778374: step: 332/529, loss: 0.06511249393224716 2023-01-21 11:16:59.880968: step: 336/529, loss: 0.04030895233154297 2023-01-21 11:17:01.005558: step: 340/529, loss: 0.03682670742273331 2023-01-21 11:17:02.120306: step: 344/529, loss: 0.023293208330869675 2023-01-21 11:17:03.238152: step: 348/529, loss: 0.017755890265107155 2023-01-21 11:17:04.350310: step: 352/529, loss: 0.008890628814697266 2023-01-21 11:17:05.459593: step: 356/529, loss: 0.01848297193646431 2023-01-21 11:17:06.555905: step: 360/529, loss: 0.04334626346826553 2023-01-21 11:17:07.679570: step: 364/529, loss: 0.0025909424293786287 2023-01-21 11:17:08.803179: step: 368/529, loss: 0.10297536849975586 2023-01-21 11:17:09.943395: step: 372/529, loss: 0.2696473002433777 2023-01-21 11:17:11.075708: step: 376/529, loss: 7.23838820704259e-05 2023-01-21 11:17:12.185121: step: 380/529, loss: 0.05467405170202255 2023-01-21 11:17:13.288515: step: 384/529, loss: 0.03773241117596626 2023-01-21 11:17:14.379109: step: 388/529, loss: 0.005128288175910711 2023-01-21 11:17:15.492537: step: 392/529, loss: 0.0326143279671669 2023-01-21 11:17:16.615075: step: 396/529, loss: 0.002808475401252508 2023-01-21 11:17:17.739973: step: 400/529, loss: 0.0018304826226085424 2023-01-21 11:17:18.850793: step: 404/529, loss: 0.016602396965026855 2023-01-21 11:17:19.953541: step: 408/529, loss: 0.03432111814618111 2023-01-21 11:17:21.108949: step: 412/529, loss: 0.030054284259676933 2023-01-21 11:17:22.206988: step: 416/529, loss: 0.029904749244451523 2023-01-21 11:17:23.322789: step: 420/529, loss: 0.011365986429154873 2023-01-21 11:17:24.433284: step: 424/529, loss: 0.036942388862371445 2023-01-21 11:17:25.560278: step: 428/529, loss: 0.1254151463508606 2023-01-21 11:17:26.687472: step: 432/529, loss: 0.05761528015136719 2023-01-21 11:17:27.797920: step: 436/529, loss: 0.0015936612617224455 2023-01-21 11:17:28.887064: step: 440/529, loss: 0.007861328311264515 2023-01-21 11:17:30.022848: step: 444/529, loss: 0.04824542999267578 2023-01-21 11:17:31.122581: step: 448/529, loss: 0.02397623099386692 2023-01-21 11:17:32.226837: step: 452/529, loss: 0.017048239707946777 2023-01-21 11:17:33.321856: step: 456/529, loss: 0.12359494715929031 2023-01-21 11:17:34.417499: step: 460/529, loss: 0.00046868325443938375 2023-01-21 11:17:35.535605: step: 464/529, loss: 0.04047088697552681 2023-01-21 11:17:36.663147: step: 468/529, loss: 0.02680187113583088 2023-01-21 11:17:37.775861: step: 472/529, loss: 0.008870887570083141 2023-01-21 11:17:38.903325: step: 476/529, loss: 0.047852709889411926 2023-01-21 11:17:40.006180: step: 480/529, loss: 0.006445503327995539 2023-01-21 11:17:41.132195: step: 484/529, loss: 0.01652688905596733 2023-01-21 11:17:42.242795: step: 488/529, loss: 0.0021993399132043123 2023-01-21 11:17:43.394702: step: 492/529, loss: 0.002772808074951172 2023-01-21 11:17:44.561849: step: 496/529, loss: 0.07501278072595596 2023-01-21 11:17:45.664140: step: 500/529, loss: 0.055669210851192474 2023-01-21 11:17:46.791757: step: 504/529, loss: 0.04873981326818466 2023-01-21 11:17:47.913390: step: 508/529, loss: 0.05307312309741974 2023-01-21 11:17:49.007197: step: 512/529, loss: 0.0007361412281170487 2023-01-21 11:17:50.120205: step: 516/529, loss: 0.005601024720817804 2023-01-21 11:17:51.221198: step: 520/529, loss: 0.011008357629179955 2023-01-21 11:17:52.320396: step: 524/529, loss: 0.02171478420495987 2023-01-21 11:17:53.433542: step: 528/529, loss: 0.020067309960722923 2023-01-21 11:17:54.563054: step: 532/529, loss: 0.05024278163909912 2023-01-21 11:17:55.683798: step: 536/529, loss: 0.06246328726410866 2023-01-21 11:17:56.801713: step: 540/529, loss: 0.009054947644472122 2023-01-21 11:17:57.925881: step: 544/529, loss: 0.046475209295749664 2023-01-21 11:17:59.034400: step: 548/529, loss: 0.007139307446777821 2023-01-21 11:18:00.140033: step: 552/529, loss: 0.021619129925966263 2023-01-21 11:18:01.264410: step: 556/529, loss: 0.008919572457671165 2023-01-21 11:18:02.368765: step: 560/529, loss: 0.06467042118310928 2023-01-21 11:18:03.498661: step: 564/529, loss: 7.982253737282008e-05 2023-01-21 11:18:04.670961: step: 568/529, loss: 0.0006339072715491056 2023-01-21 11:18:05.823108: step: 572/529, loss: 0.010019492357969284 2023-01-21 11:18:06.983451: step: 576/529, loss: 0.014912128448486328 2023-01-21 11:18:08.086456: step: 580/529, loss: 0.021863173693418503 2023-01-21 11:18:09.182816: step: 584/529, loss: 0.0007422447670251131 2023-01-21 11:18:10.310911: step: 588/529, loss: 0.015879059210419655 2023-01-21 11:18:11.400554: step: 592/529, loss: 0.0011568069458007812 2023-01-21 11:18:12.549421: step: 596/529, loss: 0.020446205511689186 2023-01-21 11:18:13.686697: step: 600/529, loss: 0.05336475372314453 2023-01-21 11:18:14.813156: step: 604/529, loss: 6.895065598655492e-05 2023-01-21 11:18:15.956536: step: 608/529, loss: 0.46739864349365234 2023-01-21 11:18:17.075331: step: 612/529, loss: 0.010795939713716507 2023-01-21 11:18:18.195055: step: 616/529, loss: 0.027636241167783737 2023-01-21 11:18:19.304483: step: 620/529, loss: 0.09429407119750977 2023-01-21 11:18:20.420869: step: 624/529, loss: 0.07420787960290909 2023-01-21 11:18:21.570109: step: 628/529, loss: 0.0014320374466478825 2023-01-21 11:18:22.701608: step: 632/529, loss: 0.10110418498516083 2023-01-21 11:18:23.815677: step: 636/529, loss: 0.006761932745575905 2023-01-21 11:18:24.922629: step: 640/529, loss: 0.12080192565917969 2023-01-21 11:18:26.011112: step: 644/529, loss: 0.0023211955558508635 2023-01-21 11:18:27.164663: step: 648/529, loss: 0.09358477592468262 2023-01-21 11:18:28.310657: step: 652/529, loss: 0.004716969095170498 2023-01-21 11:18:29.470743: step: 656/529, loss: 0.020423317328095436 2023-01-21 11:18:30.549939: step: 660/529, loss: 3.0040740966796875e-05 2023-01-21 11:18:31.698489: step: 664/529, loss: 0.0008813910535536706 2023-01-21 11:18:32.831204: step: 668/529, loss: 0.007923126220703125 2023-01-21 11:18:33.972533: step: 672/529, loss: 0.037394046783447266 2023-01-21 11:18:35.071910: step: 676/529, loss: 0.0008081436390057206 2023-01-21 11:18:36.192681: step: 680/529, loss: 0.05073799192905426 2023-01-21 11:18:37.319970: step: 684/529, loss: 0.021272849291563034 2023-01-21 11:18:38.419745: step: 688/529, loss: 0.026344871148467064 2023-01-21 11:18:39.551351: step: 692/529, loss: 0.011310672387480736 2023-01-21 11:18:40.665765: step: 696/529, loss: 0.12726135551929474 2023-01-21 11:18:41.807942: step: 700/529, loss: 0.04306221008300781 2023-01-21 11:18:42.914823: step: 704/529, loss: 0.007143735885620117 2023-01-21 11:18:44.025164: step: 708/529, loss: 0.03567161411046982 2023-01-21 11:18:45.123637: step: 712/529, loss: 0.0015853882068768144 2023-01-21 11:18:46.244530: step: 716/529, loss: 0.004586124327033758 2023-01-21 11:18:47.369935: step: 720/529, loss: 0.012326335534453392 2023-01-21 11:18:48.496029: step: 724/529, loss: 0.0005102157592773438 2023-01-21 11:18:49.615039: step: 728/529, loss: 0.0015952109824866056 2023-01-21 11:18:50.746019: step: 732/529, loss: 0.014678955078125 2023-01-21 11:18:51.854988: step: 736/529, loss: 0.06628882884979248 2023-01-21 11:18:52.977080: step: 740/529, loss: 0.000903868698514998 2023-01-21 11:18:54.092636: step: 744/529, loss: 0.052529335021972656 2023-01-21 11:18:55.205129: step: 748/529, loss: 0.021105384454131126 2023-01-21 11:18:56.315363: step: 752/529, loss: 0.0865146666765213 2023-01-21 11:18:57.447840: step: 756/529, loss: 0.0017707825172692537 2023-01-21 11:18:58.560145: step: 760/529, loss: 0.00037088393582962453 2023-01-21 11:18:59.658935: step: 764/529, loss: 0.012800311669707298 2023-01-21 11:19:00.781487: step: 768/529, loss: 0.0025392533279955387 2023-01-21 11:19:01.880803: step: 772/529, loss: 0.04057617112994194 2023-01-21 11:19:03.008700: step: 776/529, loss: 0.004737472627311945 2023-01-21 11:19:04.122849: step: 780/529, loss: 0.056742168962955475 2023-01-21 11:19:05.244742: step: 784/529, loss: 0.0018266676925122738 2023-01-21 11:19:06.353581: step: 788/529, loss: 0.010502815246582031 2023-01-21 11:19:07.460207: step: 792/529, loss: 0.012565422803163528 2023-01-21 11:19:08.563033: step: 796/529, loss: 0.019710158929228783 2023-01-21 11:19:09.668212: step: 800/529, loss: 0.05536976084113121 2023-01-21 11:19:10.804861: step: 804/529, loss: 0.004331779666244984 2023-01-21 11:19:11.904693: step: 808/529, loss: 0.011479330249130726 2023-01-21 11:19:13.028132: step: 812/529, loss: 0.05956754833459854 2023-01-21 11:19:14.165379: step: 816/529, loss: 0.055561065673828125 2023-01-21 11:19:15.298555: step: 820/529, loss: 0.005342293065041304 2023-01-21 11:19:16.403601: step: 824/529, loss: 0.003073072526603937 2023-01-21 11:19:17.502375: step: 828/529, loss: 0.0037328721955418587 2023-01-21 11:19:18.641691: step: 832/529, loss: 0.020804787054657936 2023-01-21 11:19:19.757621: step: 836/529, loss: 0.014915943145751953 2023-01-21 11:19:20.881702: step: 840/529, loss: 0.028075218200683594 2023-01-21 11:19:22.057096: step: 844/529, loss: 0.03220729902386665 2023-01-21 11:19:23.179817: step: 848/529, loss: 0.0013166428543627262 2023-01-21 11:19:24.292151: step: 852/529, loss: 0.03249330446124077 2023-01-21 11:19:25.397857: step: 856/529, loss: 0.03174781799316406 2023-01-21 11:19:26.526556: step: 860/529, loss: 0.03603344410657883 2023-01-21 11:19:27.671424: step: 864/529, loss: 0.044113826006650925 2023-01-21 11:19:28.810805: step: 868/529, loss: 0.011435795575380325 2023-01-21 11:19:29.948930: step: 872/529, loss: 0.006736183539032936 2023-01-21 11:19:31.059757: step: 876/529, loss: 0.027289342135190964 2023-01-21 11:19:32.161582: step: 880/529, loss: 0.0060266610234975815 2023-01-21 11:19:33.254742: step: 884/529, loss: 0.006854248698800802 2023-01-21 11:19:34.373742: step: 888/529, loss: 0.003201723098754883 2023-01-21 11:19:35.493672: step: 892/529, loss: 0.0032588958274573088 2023-01-21 11:19:36.616699: step: 896/529, loss: 0.0595417246222496 2023-01-21 11:19:37.728738: step: 900/529, loss: 0.0007473469013348222 2023-01-21 11:19:38.848168: step: 904/529, loss: 0.058427829295396805 2023-01-21 11:19:40.010540: step: 908/529, loss: 0.03661804273724556 2023-01-21 11:19:41.136046: step: 912/529, loss: 0.022238921374082565 2023-01-21 11:19:42.247961: step: 916/529, loss: 0.03210010752081871 2023-01-21 11:19:43.327722: step: 920/529, loss: 2.6655196052161045e-05 2023-01-21 11:19:44.431410: step: 924/529, loss: 0.029155446216464043 2023-01-21 11:19:45.546656: step: 928/529, loss: 0.00039548875065520406 2023-01-21 11:19:46.667127: step: 932/529, loss: 0.0001445770321879536 2023-01-21 11:19:47.800291: step: 936/529, loss: 0.0034053800627589226 2023-01-21 11:19:48.904797: step: 940/529, loss: 0.009549761191010475 2023-01-21 11:19:50.020237: step: 944/529, loss: 0.02778167836368084 2023-01-21 11:19:51.161233: step: 948/529, loss: 0.07834930717945099 2023-01-21 11:19:52.282855: step: 952/529, loss: 0.0006052016979083419 2023-01-21 11:19:53.375840: step: 956/529, loss: 0.05377340316772461 2023-01-21 11:19:54.463080: step: 960/529, loss: 0.0004207611200399697 2023-01-21 11:19:55.572942: step: 964/529, loss: 0.03605365753173828 2023-01-21 11:19:56.702138: step: 968/529, loss: 0.004719352815300226 2023-01-21 11:19:57.835007: step: 972/529, loss: 0.006659412756562233 2023-01-21 11:19:58.969713: step: 976/529, loss: 0.06506204605102539 2023-01-21 11:20:00.065958: step: 980/529, loss: 0.0002754211309365928 2023-01-21 11:20:01.188047: step: 984/529, loss: 0.0007025718805380166 2023-01-21 11:20:02.302220: step: 988/529, loss: 0.3186468183994293 2023-01-21 11:20:03.396729: step: 992/529, loss: 0.10364627838134766 2023-01-21 11:20:04.521523: step: 996/529, loss: 0.028425216674804688 2023-01-21 11:20:05.648596: step: 1000/529, loss: 0.019411087036132812 2023-01-21 11:20:06.770136: step: 1004/529, loss: 0.010079288855195045 2023-01-21 11:20:07.897625: step: 1008/529, loss: 0.012311267666518688 2023-01-21 11:20:09.042312: step: 1012/529, loss: 0.0022874833084642887 2023-01-21 11:20:10.152597: step: 1016/529, loss: 0.03241335600614548 2023-01-21 11:20:11.255160: step: 1020/529, loss: 0.03270426020026207 2023-01-21 11:20:12.352398: step: 1024/529, loss: 0.0014565945602953434 2023-01-21 11:20:13.452769: step: 1028/529, loss: 0.013020897284150124 2023-01-21 11:20:14.596428: step: 1032/529, loss: 0.12631407380104065 2023-01-21 11:20:15.746220: step: 1036/529, loss: 0.019868278875947 2023-01-21 11:20:16.882894: step: 1040/529, loss: 0.03553180769085884 2023-01-21 11:20:17.980718: step: 1044/529, loss: 0.0031086921226233244 2023-01-21 11:20:19.099923: step: 1048/529, loss: 0.016033364459872246 2023-01-21 11:20:20.236414: step: 1052/529, loss: 0.0021656989119946957 2023-01-21 11:20:21.366345: step: 1056/529, loss: 0.005913925357162952 2023-01-21 11:20:22.487812: step: 1060/529, loss: 0.019308757036924362 2023-01-21 11:20:23.591069: step: 1064/529, loss: 0.15139923989772797 2023-01-21 11:20:24.705983: step: 1068/529, loss: 0.029529571533203125 2023-01-21 11:20:25.824312: step: 1072/529, loss: 0.02379711903631687 2023-01-21 11:20:26.964563: step: 1076/529, loss: 0.0012794971698895097 2023-01-21 11:20:28.074382: step: 1080/529, loss: 0.028619766235351562 2023-01-21 11:20:29.203611: step: 1084/529, loss: 0.009420335292816162 2023-01-21 11:20:30.306378: step: 1088/529, loss: 0.023762226104736328 2023-01-21 11:20:31.411032: step: 1092/529, loss: 0.05547494813799858 2023-01-21 11:20:32.512310: step: 1096/529, loss: 0.0640069991350174 2023-01-21 11:20:33.674207: step: 1100/529, loss: 0.044533442705869675 2023-01-21 11:20:34.841102: step: 1104/529, loss: 1.912116931634955e-05 2023-01-21 11:20:35.956049: step: 1108/529, loss: 0.02153453789651394 2023-01-21 11:20:37.075113: step: 1112/529, loss: 0.08407821506261826 2023-01-21 11:20:38.191393: step: 1116/529, loss: 0.0314946174621582 2023-01-21 11:20:39.304162: step: 1120/529, loss: 0.015937425196170807 2023-01-21 11:20:40.416210: step: 1124/529, loss: 0.03895702585577965 2023-01-21 11:20:41.546296: step: 1128/529, loss: 0.043290603905916214 2023-01-21 11:20:42.655857: step: 1132/529, loss: 0.04781074821949005 2023-01-21 11:20:43.770420: step: 1136/529, loss: 0.006644487846642733 2023-01-21 11:20:44.881347: step: 1140/529, loss: 0.02222309075295925 2023-01-21 11:20:45.989208: step: 1144/529, loss: 0.00012764931307174265 2023-01-21 11:20:47.088444: step: 1148/529, loss: 0.09591178596019745 2023-01-21 11:20:48.197255: step: 1152/529, loss: 0.03915290907025337 2023-01-21 11:20:49.332501: step: 1156/529, loss: 0.0045146942138671875 2023-01-21 11:20:50.441934: step: 1160/529, loss: 0.003057861467823386 2023-01-21 11:20:51.536464: step: 1164/529, loss: 0.013817215338349342 2023-01-21 11:20:52.664547: step: 1168/529, loss: 0.007404518313705921 2023-01-21 11:20:53.770816: step: 1172/529, loss: 0.0034379959106445312 2023-01-21 11:20:54.905590: step: 1176/529, loss: 0.02954845502972603 2023-01-21 11:20:56.081891: step: 1180/529, loss: 0.04490194469690323 2023-01-21 11:20:57.222890: step: 1184/529, loss: 0.008612059988081455 2023-01-21 11:20:58.368077: step: 1188/529, loss: 0.038852788507938385 2023-01-21 11:20:59.491861: step: 1192/529, loss: 0.02552204206585884 2023-01-21 11:21:00.599876: step: 1196/529, loss: 0.00870294589549303 2023-01-21 11:21:01.731243: step: 1200/529, loss: 0.01828165166079998 2023-01-21 11:21:02.827601: step: 1204/529, loss: 0.0019250869518145919 2023-01-21 11:21:03.952918: step: 1208/529, loss: 0.062151338905096054 2023-01-21 11:21:05.066306: step: 1212/529, loss: 0.00042018890962935984 2023-01-21 11:21:06.216284: step: 1216/529, loss: 0.0059455870650708675 2023-01-21 11:21:07.351043: step: 1220/529, loss: 0.011283207684755325 2023-01-21 11:21:08.496756: step: 1224/529, loss: 0.012472724542021751 2023-01-21 11:21:09.619104: step: 1228/529, loss: 0.0035551071632653475 2023-01-21 11:21:10.739307: step: 1232/529, loss: 0.01893634721636772 2023-01-21 11:21:11.865248: step: 1236/529, loss: 0.00046453476534225047 2023-01-21 11:21:12.965764: step: 1240/529, loss: 0.029596518725156784 2023-01-21 11:21:14.080368: step: 1244/529, loss: 0.002021408174186945 2023-01-21 11:21:15.214539: step: 1248/529, loss: 0.04575634002685547 2023-01-21 11:21:16.329094: step: 1252/529, loss: 0.06333780288696289 2023-01-21 11:21:17.451675: step: 1256/529, loss: 0.0009008407359942794 2023-01-21 11:21:18.574905: step: 1260/529, loss: 0.006089329719543457 2023-01-21 11:21:19.678889: step: 1264/529, loss: 0.005976581480354071 2023-01-21 11:21:20.800566: step: 1268/529, loss: 0.0009419440757483244 2023-01-21 11:21:21.918686: step: 1272/529, loss: 0.0240189079195261 2023-01-21 11:21:23.030248: step: 1276/529, loss: 0.01443700771778822 2023-01-21 11:21:24.180701: step: 1280/529, loss: 0.0021149159874767065 2023-01-21 11:21:25.274614: step: 1284/529, loss: 0.00947418250143528 2023-01-21 11:21:26.386543: step: 1288/529, loss: 0.17442846298217773 2023-01-21 11:21:27.488445: step: 1292/529, loss: 0.0038775443099439144 2023-01-21 11:21:28.593662: step: 1296/529, loss: 0.012499738484621048 2023-01-21 11:21:29.727511: step: 1300/529, loss: 0.028967667371034622 2023-01-21 11:21:30.843123: step: 1304/529, loss: 0.05090503767132759 2023-01-21 11:21:31.938550: step: 1308/529, loss: 0.015236472710967064 2023-01-21 11:21:33.078425: step: 1312/529, loss: 0.014673185534775257 2023-01-21 11:21:34.183054: step: 1316/529, loss: 0.00047292711678892374 2023-01-21 11:21:35.291882: step: 1320/529, loss: 0.0027740478981286287 2023-01-21 11:21:36.398232: step: 1324/529, loss: 0.0005285263177938759 2023-01-21 11:21:37.531658: step: 1328/529, loss: 0.005422926042228937 2023-01-21 11:21:38.625288: step: 1332/529, loss: 0.026267338544130325 2023-01-21 11:21:39.735403: step: 1336/529, loss: 0.08325314521789551 2023-01-21 11:21:40.834124: step: 1340/529, loss: 0.0005667209625244141 2023-01-21 11:21:41.941376: step: 1344/529, loss: 0.05620746687054634 2023-01-21 11:21:43.062901: step: 1348/529, loss: 0.008568763732910156 2023-01-21 11:21:44.209845: step: 1352/529, loss: 0.0054128170013427734 2023-01-21 11:21:45.319016: step: 1356/529, loss: 9.50813337112777e-05 2023-01-21 11:21:46.450985: step: 1360/529, loss: 0.04579639434814453 2023-01-21 11:21:47.560050: step: 1364/529, loss: 0.025598812848329544 2023-01-21 11:21:48.685484: step: 1368/529, loss: 0.01798391342163086 2023-01-21 11:21:49.808908: step: 1372/529, loss: 0.04349198192358017 2023-01-21 11:21:50.952054: step: 1376/529, loss: 0.00796742457896471 2023-01-21 11:21:52.090399: step: 1380/529, loss: 0.32121312618255615 2023-01-21 11:21:53.199619: step: 1384/529, loss: 0.12058448791503906 2023-01-21 11:21:54.323790: step: 1388/529, loss: 0.0397859588265419 2023-01-21 11:21:55.448168: step: 1392/529, loss: 0.04373965412378311 2023-01-21 11:21:56.563104: step: 1396/529, loss: 0.03370971605181694 2023-01-21 11:21:57.686519: step: 1400/529, loss: 0.00021610260591842234 2023-01-21 11:21:58.810095: step: 1404/529, loss: 0.004474925808608532 2023-01-21 11:21:59.934195: step: 1408/529, loss: 0.001635646796785295 2023-01-21 11:22:01.065653: step: 1412/529, loss: 0.0037488937377929688 2023-01-21 11:22:02.168026: step: 1416/529, loss: 0.0010494232410565019 2023-01-21 11:22:03.298064: step: 1420/529, loss: 0.0026595115195959806 2023-01-21 11:22:04.424482: step: 1424/529, loss: 0.01549067534506321 2023-01-21 11:22:05.533243: step: 1428/529, loss: 0.013635540381073952 2023-01-21 11:22:06.650121: step: 1432/529, loss: 0.03781929239630699 2023-01-21 11:22:07.788278: step: 1436/529, loss: 0.003968143370002508 2023-01-21 11:22:08.884267: step: 1440/529, loss: 0.048525430262088776 2023-01-21 11:22:10.006235: step: 1444/529, loss: 0.005127429962158203 2023-01-21 11:22:11.136300: step: 1448/529, loss: 0.0021885870955884457 2023-01-21 11:22:12.236282: step: 1452/529, loss: 0.0070400238037109375 2023-01-21 11:22:13.343058: step: 1456/529, loss: 0.027751445770263672 2023-01-21 11:22:14.466562: step: 1460/529, loss: 0.010463809594511986 2023-01-21 11:22:15.568181: step: 1464/529, loss: 0.006424332037568092 2023-01-21 11:22:16.671747: step: 1468/529, loss: 0.010878181084990501 2023-01-21 11:22:17.768007: step: 1472/529, loss: 0.006523704621940851 2023-01-21 11:22:18.844128: step: 1476/529, loss: 0.004263115115463734 2023-01-21 11:22:19.993256: step: 1480/529, loss: 0.05791836231946945 2023-01-21 11:22:21.090567: step: 1484/529, loss: 0.028817366808652878 2023-01-21 11:22:22.192093: step: 1488/529, loss: 0.04910030588507652 2023-01-21 11:22:23.306076: step: 1492/529, loss: 0.0054565430618822575 2023-01-21 11:22:24.430597: step: 1496/529, loss: 0.01650269702076912 2023-01-21 11:22:25.543225: step: 1500/529, loss: 0.0033977509010583162 2023-01-21 11:22:26.659406: step: 1504/529, loss: 0.014163351617753506 2023-01-21 11:22:27.740802: step: 1508/529, loss: 0.007196045480668545 2023-01-21 11:22:28.924599: step: 1512/529, loss: 0.09505720436573029 2023-01-21 11:22:30.052148: step: 1516/529, loss: 0.21728669106960297 2023-01-21 11:22:31.174021: step: 1520/529, loss: 0.022905994206666946 2023-01-21 11:22:32.297497: step: 1524/529, loss: 0.029943037778139114 2023-01-21 11:22:33.432930: step: 1528/529, loss: 0.03110961988568306 2023-01-21 11:22:34.563489: step: 1532/529, loss: 0.004485511686652899 2023-01-21 11:22:35.693669: step: 1536/529, loss: 0.16147682070732117 2023-01-21 11:22:36.809087: step: 1540/529, loss: 0.007404995150864124 2023-01-21 11:22:37.952895: step: 1544/529, loss: 0.00901050679385662 2023-01-21 11:22:39.075929: step: 1548/529, loss: 0.05372605100274086 2023-01-21 11:22:40.187197: step: 1552/529, loss: 0.022143268957734108 2023-01-21 11:22:41.301944: step: 1556/529, loss: 0.04845810309052467 2023-01-21 11:22:42.436340: step: 1560/529, loss: 0.0153045654296875 2023-01-21 11:22:43.561940: step: 1564/529, loss: 0.00569496164098382 2023-01-21 11:22:44.661440: step: 1568/529, loss: 0.09224672615528107 2023-01-21 11:22:45.804411: step: 1572/529, loss: 0.07145166397094727 2023-01-21 11:22:46.959480: step: 1576/529, loss: 0.006652355194091797 2023-01-21 11:22:48.073098: step: 1580/529, loss: 0.030744075775146484 2023-01-21 11:22:49.182729: step: 1584/529, loss: 0.0035976411309093237 2023-01-21 11:22:50.293228: step: 1588/529, loss: 0.00803766306489706 2023-01-21 11:22:51.369897: step: 1592/529, loss: 0.010059547610580921 2023-01-21 11:22:52.505307: step: 1596/529, loss: 0.002731514163315296 2023-01-21 11:22:53.608365: step: 1600/529, loss: 0.07111578434705734 2023-01-21 11:22:54.735149: step: 1604/529, loss: 0.12307175993919373 2023-01-21 11:22:55.869476: step: 1608/529, loss: 0.021628571674227715 2023-01-21 11:22:57.024144: step: 1612/529, loss: 0.03154271841049194 2023-01-21 11:22:58.154936: step: 1616/529, loss: 0.00012645722017623484 2023-01-21 11:22:59.290924: step: 1620/529, loss: 0.017206192016601562 2023-01-21 11:23:00.393182: step: 1624/529, loss: 0.002230310346931219 2023-01-21 11:23:01.509893: step: 1628/529, loss: 0.0238389503210783 2023-01-21 11:23:02.639891: step: 1632/529, loss: 0.06463997066020966 2023-01-21 11:23:03.741831: step: 1636/529, loss: 0.03412942960858345 2023-01-21 11:23:04.837670: step: 1640/529, loss: 0.0066480631940066814 2023-01-21 11:23:05.991338: step: 1644/529, loss: 0.0455290786921978 2023-01-21 11:23:07.094980: step: 1648/529, loss: 0.18373900651931763 2023-01-21 11:23:08.213015: step: 1652/529, loss: 0.010761070996522903 2023-01-21 11:23:09.332288: step: 1656/529, loss: 0.002349758055061102 2023-01-21 11:23:10.452598: step: 1660/529, loss: 0.04829740524291992 2023-01-21 11:23:11.544132: step: 1664/529, loss: 0.16388721764087677 2023-01-21 11:23:12.663158: step: 1668/529, loss: 0.03143024444580078 2023-01-21 11:23:13.762096: step: 1672/529, loss: 0.05122409015893936 2023-01-21 11:23:14.910690: step: 1676/529, loss: 0.0005262374761514366 2023-01-21 11:23:16.033179: step: 1680/529, loss: 0.003352260682731867 2023-01-21 11:23:17.141355: step: 1684/529, loss: 0.03806591033935547 2023-01-21 11:23:18.271591: step: 1688/529, loss: 0.006078911013901234 2023-01-21 11:23:19.379682: step: 1692/529, loss: 0.028930090367794037 2023-01-21 11:23:20.485540: step: 1696/529, loss: 0.054664041846990585 2023-01-21 11:23:21.609134: step: 1700/529, loss: 0.027724266052246094 2023-01-21 11:23:22.699240: step: 1704/529, loss: 0.005173015408217907 2023-01-21 11:23:23.795681: step: 1708/529, loss: 0.026000645011663437 2023-01-21 11:23:24.953176: step: 1712/529, loss: 0.043677520006895065 2023-01-21 11:23:26.076398: step: 1716/529, loss: 0.056861989200115204 2023-01-21 11:23:27.183291: step: 1720/529, loss: 0.021589089184999466 2023-01-21 11:23:28.293598: step: 1724/529, loss: 0.0004339933511801064 2023-01-21 11:23:29.438320: step: 1728/529, loss: 0.004943370819091797 2023-01-21 11:23:30.556772: step: 1732/529, loss: 0.009901904501020908 2023-01-21 11:23:31.642750: step: 1736/529, loss: 0.015550781041383743 2023-01-21 11:23:32.725503: step: 1740/529, loss: 0.0015162468189373612 2023-01-21 11:23:33.842126: step: 1744/529, loss: 0.0029075623024255037 2023-01-21 11:23:34.955410: step: 1748/529, loss: 0.4520382285118103 2023-01-21 11:23:36.043504: step: 1752/529, loss: 0.00653419503942132 2023-01-21 11:23:37.172470: step: 1756/529, loss: 0.0431545227766037 2023-01-21 11:23:38.287949: step: 1760/529, loss: 0.007971953600645065 2023-01-21 11:23:39.396771: step: 1764/529, loss: 0.0013872147537767887 2023-01-21 11:23:40.509401: step: 1768/529, loss: 0.29605618119239807 2023-01-21 11:23:41.601065: step: 1772/529, loss: 0.020236873999238014 2023-01-21 11:23:42.728105: step: 1776/529, loss: 0.014721298590302467 2023-01-21 11:23:43.832799: step: 1780/529, loss: 0.00017032623873092234 2023-01-21 11:23:44.929804: step: 1784/529, loss: 0.07722122967243195 2023-01-21 11:23:46.066666: step: 1788/529, loss: 0.0676647201180458 2023-01-21 11:23:47.195442: step: 1792/529, loss: 0.012405872344970703 2023-01-21 11:23:48.279614: step: 1796/529, loss: 4.57763671875e-05 2023-01-21 11:23:49.389612: step: 1800/529, loss: 0.004603528883308172 2023-01-21 11:23:50.488965: step: 1804/529, loss: 0.001621246337890625 2023-01-21 11:23:51.592166: step: 1808/529, loss: 0.0024099349975585938 2023-01-21 11:23:52.718341: step: 1812/529, loss: 0.0013746261829510331 2023-01-21 11:23:53.837436: step: 1816/529, loss: 0.08621510863304138 2023-01-21 11:23:54.950487: step: 1820/529, loss: 0.06648464500904083 2023-01-21 11:23:56.064728: step: 1824/529, loss: 0.0012690543662756681 2023-01-21 11:23:57.191045: step: 1828/529, loss: 0.0007116794586181641 2023-01-21 11:23:58.322141: step: 1832/529, loss: 0.0034635544288903475 2023-01-21 11:23:59.450629: step: 1836/529, loss: 0.00012102127220714465 2023-01-21 11:24:00.543779: step: 1840/529, loss: 0.036646127700805664 2023-01-21 11:24:01.658555: step: 1844/529, loss: 0.01183543261140585 2023-01-21 11:24:02.810619: step: 1848/529, loss: 0.0014579773414880037 2023-01-21 11:24:03.929821: step: 1852/529, loss: 0.0003961563343182206 2023-01-21 11:24:05.041863: step: 1856/529, loss: 0.04968290403485298 2023-01-21 11:24:06.157492: step: 1860/529, loss: 0.00946798361837864 2023-01-21 11:24:07.285789: step: 1864/529, loss: 0.05361900478601456 2023-01-21 11:24:08.369102: step: 1868/529, loss: 0.021692801266908646 2023-01-21 11:24:09.511363: step: 1872/529, loss: 0.03295927122235298 2023-01-21 11:24:10.626086: step: 1876/529, loss: 0.0465368777513504 2023-01-21 11:24:11.751527: step: 1880/529, loss: 0.02541074901819229 2023-01-21 11:24:12.881670: step: 1884/529, loss: 0.055675603449344635 2023-01-21 11:24:13.986206: step: 1888/529, loss: 0.0022763251326978207 2023-01-21 11:24:15.090480: step: 1892/529, loss: 0.01654643937945366 2023-01-21 11:24:16.199936: step: 1896/529, loss: 0.03304043039679527 2023-01-21 11:24:17.313885: step: 1900/529, loss: 0.0023160933051258326 2023-01-21 11:24:18.415433: step: 1904/529, loss: 0.017093371599912643 2023-01-21 11:24:19.536067: step: 1908/529, loss: 0.01160345133394003 2023-01-21 11:24:20.678823: step: 1912/529, loss: 0.016899585723876953 2023-01-21 11:24:21.766827: step: 1916/529, loss: 0.04355192184448242 2023-01-21 11:24:22.909710: step: 1920/529, loss: 0.01504297275096178 2023-01-21 11:24:24.034993: step: 1924/529, loss: 0.028369521722197533 2023-01-21 11:24:25.131466: step: 1928/529, loss: 0.0086212158203125 2023-01-21 11:24:26.234302: step: 1932/529, loss: 0.00090789794921875 2023-01-21 11:24:27.372502: step: 1936/529, loss: 0.007388782687485218 2023-01-21 11:24:28.497496: step: 1940/529, loss: 2.7275083994027227e-05 2023-01-21 11:24:29.590870: step: 1944/529, loss: 0.0023542402777820826 2023-01-21 11:24:30.699708: step: 1948/529, loss: 0.0009572983253747225 2023-01-21 11:24:31.810983: step: 1952/529, loss: 0.07070598751306534 2023-01-21 11:24:32.902695: step: 1956/529, loss: 0.027022838592529297 2023-01-21 11:24:34.018705: step: 1960/529, loss: 0.0007778167491778731 2023-01-21 11:24:35.146045: step: 1964/529, loss: 0.007288360968232155 2023-01-21 11:24:36.248268: step: 1968/529, loss: 0.004833221435546875 2023-01-21 11:24:37.362991: step: 1972/529, loss: 0.0564996711909771 2023-01-21 11:24:38.520269: step: 1976/529, loss: 0.00883016549050808 2023-01-21 11:24:39.641439: step: 1980/529, loss: 0.004055881407111883 2023-01-21 11:24:40.757978: step: 1984/529, loss: 0.009660912677645683 2023-01-21 11:24:41.880548: step: 1988/529, loss: 0.015068341046571732 2023-01-21 11:24:43.054973: step: 1992/529, loss: 0.006891059689223766 2023-01-21 11:24:44.164785: step: 1996/529, loss: 0.0006494522094726562 2023-01-21 11:24:45.284396: step: 2000/529, loss: 0.056334689259529114 2023-01-21 11:24:46.373872: step: 2004/529, loss: 0.04470730200409889 2023-01-21 11:24:47.470724: step: 2008/529, loss: 0.006189537234604359 2023-01-21 11:24:48.563953: step: 2012/529, loss: 0.03673258051276207 2023-01-21 11:24:49.669645: step: 2016/529, loss: 0.00304241175763309 2023-01-21 11:24:50.778263: step: 2020/529, loss: 0.005485296715050936 2023-01-21 11:24:51.895843: step: 2024/529, loss: 0.021397780627012253 2023-01-21 11:24:53.030044: step: 2028/529, loss: 0.04088955000042915 2023-01-21 11:24:54.193405: step: 2032/529, loss: 0.04782600700855255 2023-01-21 11:24:55.282108: step: 2036/529, loss: 0.0009453773964196444 2023-01-21 11:24:56.411800: step: 2040/529, loss: 0.04519510269165039 2023-01-21 11:24:57.519710: step: 2044/529, loss: 0.010474396869540215 2023-01-21 11:24:58.649761: step: 2048/529, loss: 0.009509850293397903 2023-01-21 11:24:59.778116: step: 2052/529, loss: 0.07295895367860794 2023-01-21 11:25:00.890719: step: 2056/529, loss: 0.005055427551269531 2023-01-21 11:25:02.006516: step: 2060/529, loss: 0.014079665765166283 2023-01-21 11:25:03.123115: step: 2064/529, loss: 0.16413459181785583 2023-01-21 11:25:04.259442: step: 2068/529, loss: 0.014315415173768997 2023-01-21 11:25:05.334305: step: 2072/529, loss: 0.0027179718017578125 2023-01-21 11:25:06.437023: step: 2076/529, loss: 0.00542449951171875 2023-01-21 11:25:07.571329: step: 2080/529, loss: 0.007073068991303444 2023-01-21 11:25:08.684232: step: 2084/529, loss: 0.006602764129638672 2023-01-21 11:25:09.833367: step: 2088/529, loss: 0.01106405258178711 2023-01-21 11:25:10.948633: step: 2092/529, loss: 0.02700824663043022 2023-01-21 11:25:12.068718: step: 2096/529, loss: 0.04171320050954819 2023-01-21 11:25:13.194215: step: 2100/529, loss: 0.06613288819789886 2023-01-21 11:25:14.331285: step: 2104/529, loss: 0.007037639617919922 2023-01-21 11:25:15.444274: step: 2108/529, loss: 0.08476867526769638 2023-01-21 11:25:16.617273: step: 2112/529, loss: 0.011888790875673294 2023-01-21 11:25:17.741425: step: 2116/529, loss: 0.027255915105342865 ================================================== Loss: 0.030 -------------------- Dev: {'event': {'p': 0.5814889336016097, 'r': 0.7696404793608522, 'f1': 0.6624641833810889}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Test: {'event': {'p': 0.6259440248778321, 'r': 0.7951467268623025, 'f1': 0.7004722843649017}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Chinese: {'event': {'p': 0.5795454545454546, 'r': 0.9444444444444444, 'f1': 0.7183098591549296}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Korean: {'event': {'p': 0.5806451612903226, 'r': 0.5714285714285714, 'f1': 0.576}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} 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': 16} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** 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 11:25:59.234892: step: 4/529, loss: 0.007236194796860218 2023-01-21 11:26:00.326805: step: 8/529, loss: 0.012864941731095314 2023-01-21 11:26:01.417841: step: 12/529, loss: 0.02682342566549778 2023-01-21 11:26:02.553759: step: 16/529, loss: 0.021349716931581497 2023-01-21 11:26:03.700426: step: 20/529, loss: 0.02792081981897354 2023-01-21 11:26:04.812634: step: 24/529, loss: 0.0148963937535882 2023-01-21 11:26:05.942699: step: 28/529, loss: 0.007674217224121094 2023-01-21 11:26:07.045214: step: 32/529, loss: 0.052739571779966354 2023-01-21 11:26:08.170048: step: 36/529, loss: 0.0015630245907232165 2023-01-21 11:26:09.280073: step: 40/529, loss: 0.021659279242157936 2023-01-21 11:26:10.400884: step: 44/529, loss: 0.0002853393671102822 2023-01-21 11:26:11.499372: step: 48/529, loss: 0.04237651824951172 2023-01-21 11:26:12.670461: step: 52/529, loss: 0.006783866789191961 2023-01-21 11:26:13.779200: step: 56/529, loss: 0.015005683526396751 2023-01-21 11:26:14.878763: step: 60/529, loss: 0.030582809820771217 2023-01-21 11:26:15.989701: step: 64/529, loss: 0.010562468320131302 2023-01-21 11:26:17.111928: step: 68/529, loss: 0.08216266334056854 2023-01-21 11:26:18.207059: step: 72/529, loss: 0.010211657732725143 2023-01-21 11:26:19.326068: step: 76/529, loss: 0.008458519354462624 2023-01-21 11:26:20.435122: step: 80/529, loss: 0.05699892342090607 2023-01-21 11:26:21.537899: step: 84/529, loss: 0.017818927764892578 2023-01-21 11:26:22.632367: step: 88/529, loss: 0.00033926963806152344 2023-01-21 11:26:23.714403: step: 92/529, loss: 0.010666560381650925 2023-01-21 11:26:24.811525: step: 96/529, loss: 0.0004039764462504536 2023-01-21 11:26:25.919529: step: 100/529, loss: 0.009726524353027344 2023-01-21 11:26:27.016601: step: 104/529, loss: 0.002474975772202015 2023-01-21 11:26:28.122416: step: 108/529, loss: 0.035889625549316406 2023-01-21 11:26:29.233912: step: 112/529, loss: 0.0011489868629723787 2023-01-21 11:26:30.347117: step: 116/529, loss: 0.0012229920830577612 2023-01-21 11:26:31.487743: step: 120/529, loss: 0.0005732536083087325 2023-01-21 11:26:32.606879: step: 124/529, loss: 0.08603204041719437 2023-01-21 11:26:33.717592: step: 128/529, loss: 0.003493881318718195 2023-01-21 11:26:34.819379: step: 132/529, loss: 0.031853388994932175 2023-01-21 11:26:35.922716: step: 136/529, loss: 0.03113851696252823 2023-01-21 11:26:37.023680: step: 140/529, loss: 0.0057045938447117805 2023-01-21 11:26:38.160184: step: 144/529, loss: 0.03841762617230415 2023-01-21 11:26:39.286646: step: 148/529, loss: 0.005462074652314186 2023-01-21 11:26:40.385998: step: 152/529, loss: 0.0007432461134158075 2023-01-21 11:26:41.507969: step: 156/529, loss: 0.004455757327377796 2023-01-21 11:26:42.673388: step: 160/529, loss: 0.0036363601684570312 2023-01-21 11:26:43.776788: step: 164/529, loss: 0.009899521246552467 2023-01-21 11:26:44.888239: step: 168/529, loss: 0.006796318106353283 2023-01-21 11:26:45.984685: step: 172/529, loss: 0.3571142256259918 2023-01-21 11:26:47.107464: step: 176/529, loss: 0.010512541979551315 2023-01-21 11:26:48.239391: step: 180/529, loss: 0.0033124922774732113 2023-01-21 11:26:49.338183: step: 184/529, loss: 8.678436643094756e-06 2023-01-21 11:26:50.500530: step: 188/529, loss: 0.003802585881203413 2023-01-21 11:26:51.609946: step: 192/529, loss: 0.012919998727738857 2023-01-21 11:26:52.734583: step: 196/529, loss: 0.013766909018158913 2023-01-21 11:26:53.838991: step: 200/529, loss: 0.0013263702858239412 2023-01-21 11:26:54.940147: step: 204/529, loss: 0.007224178407341242 2023-01-21 11:26:56.062429: step: 208/529, loss: 0.0013332366943359375 2023-01-21 11:26:57.203743: step: 212/529, loss: 0.007906913757324219 2023-01-21 11:26:58.324802: step: 216/529, loss: 0.01584453694522381 2023-01-21 11:26:59.451821: step: 220/529, loss: 0.041970063000917435 2023-01-21 11:27:00.573832: step: 224/529, loss: 0.013851357623934746 2023-01-21 11:27:01.694390: step: 228/529, loss: 0.022777557373046875 2023-01-21 11:27:02.795571: step: 232/529, loss: 0.0034509659744799137 2023-01-21 11:27:03.900108: step: 236/529, loss: 0.0014532088534906507 2023-01-21 11:27:04.996991: step: 240/529, loss: 0.00862200278788805 2023-01-21 11:27:06.116867: step: 244/529, loss: 9.884834435069934e-05 2023-01-21 11:27:07.209425: step: 248/529, loss: 0.05267906188964844 2023-01-21 11:27:08.347660: step: 252/529, loss: 0.0010055541060864925 2023-01-21 11:27:09.460462: step: 256/529, loss: 0.022330857813358307 2023-01-21 11:27:10.596995: step: 260/529, loss: 0.15139131247997284 2023-01-21 11:27:11.734679: step: 264/529, loss: 0.09378699958324432 2023-01-21 11:27:12.843018: step: 268/529, loss: 0.007386207580566406 2023-01-21 11:27:13.968861: step: 272/529, loss: 0.02782278135418892 2023-01-21 11:27:15.097253: step: 276/529, loss: 0.0014570235507562757 2023-01-21 11:27:16.230864: step: 280/529, loss: 0.0122238639742136 2023-01-21 11:27:17.360902: step: 284/529, loss: 0.0002176284760935232 2023-01-21 11:27:18.478548: step: 288/529, loss: 0.00545611372217536 2023-01-21 11:27:19.569317: step: 292/529, loss: 0.0002180576411774382 2023-01-21 11:27:20.681597: step: 296/529, loss: 6.837844557594508e-05 2023-01-21 11:27:21.801725: step: 300/529, loss: 0.023854253813624382 2023-01-21 11:27:22.891686: step: 304/529, loss: 0.024345112964510918 2023-01-21 11:27:24.009616: step: 308/529, loss: 0.11616702377796173 2023-01-21 11:27:25.101907: step: 312/529, loss: 0.05671720579266548 2023-01-21 11:27:26.200284: step: 316/529, loss: 0.005731392186135054 2023-01-21 11:27:27.287816: step: 320/529, loss: 0.0009192362776957452 2023-01-21 11:27:28.413751: step: 324/529, loss: 0.019611548632383347 2023-01-21 11:27:29.551419: step: 328/529, loss: 0.015974808484315872 2023-01-21 11:27:30.680377: step: 332/529, loss: 0.0137062082067132 2023-01-21 11:27:31.785456: step: 336/529, loss: 0.004550695884972811 2023-01-21 11:27:32.922198: step: 340/529, loss: 0.11097335815429688 2023-01-21 11:27:34.041826: step: 344/529, loss: 0.022510338574647903 2023-01-21 11:27:35.144152: step: 348/529, loss: 0.01965322531759739 2023-01-21 11:27:36.272678: step: 352/529, loss: 0.031092356890439987 2023-01-21 11:27:37.337575: step: 356/529, loss: 0.0006049156654626131 2023-01-21 11:27:38.461877: step: 360/529, loss: 0.004075813572853804 2023-01-21 11:27:39.606947: step: 364/529, loss: 0.08412008732557297 2023-01-21 11:27:40.723078: step: 368/529, loss: 6.628036499023438e-05 2023-01-21 11:27:41.811970: step: 372/529, loss: 0.011430168524384499 2023-01-21 11:27:42.942432: step: 376/529, loss: 0.02530393749475479 2023-01-21 11:27:44.062601: step: 380/529, loss: 0.03544940799474716 2023-01-21 11:27:45.207812: step: 384/529, loss: 0.04354820027947426 2023-01-21 11:27:46.316983: step: 388/529, loss: 0.009791946969926357 2023-01-21 11:27:47.424789: step: 392/529, loss: 0.001482200575992465 2023-01-21 11:27:48.523751: step: 396/529, loss: 0.03681983798742294 2023-01-21 11:27:49.634635: step: 400/529, loss: 0.005219554994255304 2023-01-21 11:27:50.767508: step: 404/529, loss: 0.10754509270191193 2023-01-21 11:27:51.897796: step: 408/529, loss: 0.013379668816924095 2023-01-21 11:27:53.008803: step: 412/529, loss: 0.038869570940732956 2023-01-21 11:27:54.133929: step: 416/529, loss: 0.006100082769989967 2023-01-21 11:27:55.288934: step: 420/529, loss: 0.023983241990208626 2023-01-21 11:27:56.424068: step: 424/529, loss: 8.678436643094756e-06 2023-01-21 11:27:57.538210: step: 428/529, loss: 0.00715904263779521 2023-01-21 11:27:58.634286: step: 432/529, loss: 0.015358162112534046 2023-01-21 11:27:59.758024: step: 436/529, loss: 0.009467792697250843 2023-01-21 11:28:00.859337: step: 440/529, loss: 0.03373704105615616 2023-01-21 11:28:01.972565: step: 444/529, loss: 0.00965423509478569 2023-01-21 11:28:03.071477: step: 448/529, loss: 0.0037589548155665398 2023-01-21 11:28:04.193351: step: 452/529, loss: 0.10768795758485794 2023-01-21 11:28:05.279479: step: 456/529, loss: 0.05358562618494034 2023-01-21 11:28:06.375621: step: 460/529, loss: 0.0011633873218670487 2023-01-21 11:28:07.474780: step: 464/529, loss: 0.02670455165207386 2023-01-21 11:28:08.561290: step: 468/529, loss: 0.014192391186952591 2023-01-21 11:28:09.671086: step: 472/529, loss: 0.006118393037468195 2023-01-21 11:28:10.775792: step: 476/529, loss: 0.002104187151417136 2023-01-21 11:28:11.887420: step: 480/529, loss: 0.0248170867562294 2023-01-21 11:28:12.972035: step: 484/529, loss: 0.002963257022202015 2023-01-21 11:28:14.063036: step: 488/529, loss: 0.000711977481842041 2023-01-21 11:28:15.203559: step: 492/529, loss: 0.0024950981605798006 2023-01-21 11:28:16.334983: step: 496/529, loss: 0.006621265783905983 2023-01-21 11:28:17.445601: step: 500/529, loss: 0.014836407266557217 2023-01-21 11:28:18.531784: step: 504/529, loss: 0.011418056674301624 2023-01-21 11:28:19.640385: step: 508/529, loss: 0.0636986717581749 2023-01-21 11:28:20.780942: step: 512/529, loss: 0.039900828152894974 2023-01-21 11:28:21.889774: step: 516/529, loss: -1.3828277587890625e-05 2023-01-21 11:28:22.998651: step: 520/529, loss: 0.002452850341796875 2023-01-21 11:28:24.135775: step: 524/529, loss: 0.0038934708572924137 2023-01-21 11:28:25.249950: step: 528/529, loss: 0.010477161034941673 2023-01-21 11:28:26.369944: step: 532/529, loss: 0.0071919444017112255 2023-01-21 11:28:27.507153: step: 536/529, loss: 0.00015864371380303055 2023-01-21 11:28:28.617919: step: 540/529, loss: 0.018250465393066406 2023-01-21 11:28:29.751536: step: 544/529, loss: 0.0032693862449377775 2023-01-21 11:28:30.844405: step: 548/529, loss: 0.00015745163545943797 2023-01-21 11:28:31.948384: step: 552/529, loss: 0.06083851307630539 2023-01-21 11:28:33.054036: step: 556/529, loss: 0.0032431127037853003 2023-01-21 11:28:34.169156: step: 560/529, loss: 0.02098102495074272 2023-01-21 11:28:35.262139: step: 564/529, loss: 0.010703039355576038 2023-01-21 11:28:36.395854: step: 568/529, loss: 0.05978946387767792 2023-01-21 11:28:37.488291: step: 572/529, loss: 0.002813053084537387 2023-01-21 11:28:38.599080: step: 576/529, loss: 0.04347124323248863 2023-01-21 11:28:39.728319: step: 580/529, loss: 0.01481637917459011 2023-01-21 11:28:40.821396: step: 584/529, loss: 0.02793254889547825 2023-01-21 11:28:41.951667: step: 588/529, loss: 0.0008490562322549522 2023-01-21 11:28:43.049366: step: 592/529, loss: 0.00684547470882535 2023-01-21 11:28:44.156799: step: 596/529, loss: 0.0012743950355798006 2023-01-21 11:28:45.285148: step: 600/529, loss: 0.003914475440979004 2023-01-21 11:28:46.390686: step: 604/529, loss: 0.0004927635309286416 2023-01-21 11:28:47.502543: step: 608/529, loss: 0.010393905453383923 2023-01-21 11:28:48.640146: step: 612/529, loss: 0.01200017984956503 2023-01-21 11:28:49.806312: step: 616/529, loss: 0.11614914238452911 2023-01-21 11:28:50.924348: step: 620/529, loss: 0.0013360977172851562 2023-01-21 11:28:52.043791: step: 624/529, loss: 0.0043464661575853825 2023-01-21 11:28:53.165716: step: 628/529, loss: 0.0014190673828125 2023-01-21 11:28:54.266287: step: 632/529, loss: 0.07492131739854813 2023-01-21 11:28:55.375336: step: 636/529, loss: 0.012208842672407627 2023-01-21 11:28:56.495976: step: 640/529, loss: 0.001449680421501398 2023-01-21 11:28:57.638895: step: 644/529, loss: 0.021663475781679153 2023-01-21 11:28:58.715741: step: 648/529, loss: 1.3542176020564511e-05 2023-01-21 11:28:59.811855: step: 652/529, loss: 0.027807235717773438 2023-01-21 11:29:00.928896: step: 656/529, loss: 0.0062777516432106495 2023-01-21 11:29:02.070792: step: 660/529, loss: 0.009154224768280983 2023-01-21 11:29:03.199228: step: 664/529, loss: 0.03375568240880966 2023-01-21 11:29:04.330822: step: 668/529, loss: 0.0016595839988440275 2023-01-21 11:29:05.458716: step: 672/529, loss: 0.005206489935517311 2023-01-21 11:29:06.552798: step: 676/529, loss: 0.004783108830451965 2023-01-21 11:29:07.690013: step: 680/529, loss: 0.13351936638355255 2023-01-21 11:29:08.852251: step: 684/529, loss: 0.03379116207361221 2023-01-21 11:29:09.969164: step: 688/529, loss: 0.002034950302913785 2023-01-21 11:29:11.077725: step: 692/529, loss: 0.0030596733558923006 2023-01-21 11:29:12.183632: step: 696/529, loss: 0.01688232459127903 2023-01-21 11:29:13.277504: step: 700/529, loss: 0.005562114994972944 2023-01-21 11:29:14.367130: step: 704/529, loss: 0.024741649627685547 2023-01-21 11:29:15.467646: step: 708/529, loss: 0.000754928623791784 2023-01-21 11:29:16.574331: step: 712/529, loss: 0.0010656357044354081 2023-01-21 11:29:17.687770: step: 716/529, loss: 0.07712183147668839 2023-01-21 11:29:18.797880: step: 720/529, loss: 0.0018567085498943925 2023-01-21 11:29:19.938525: step: 724/529, loss: 0.025477027520537376 2023-01-21 11:29:21.054520: step: 728/529, loss: 0.07873684912919998 2023-01-21 11:29:22.136166: step: 732/529, loss: 0.032294657081365585 2023-01-21 11:29:23.278113: step: 736/529, loss: 0.008506583981215954 2023-01-21 11:29:24.431657: step: 740/529, loss: 0.007420158479362726 2023-01-21 11:29:25.581028: step: 744/529, loss: 0.005598259158432484 2023-01-21 11:29:26.700731: step: 748/529, loss: 6.632805161643773e-05 2023-01-21 11:29:27.810917: step: 752/529, loss: 0.005912971682846546 2023-01-21 11:29:28.939876: step: 756/529, loss: 0.016251374036073685 2023-01-21 11:29:30.052956: step: 760/529, loss: 0.5293411016464233 2023-01-21 11:29:31.165757: step: 764/529, loss: 0.0023424148093909025 2023-01-21 11:29:32.285024: step: 768/529, loss: 0.002216339111328125 2023-01-21 11:29:33.428046: step: 772/529, loss: 8.735656592762098e-05 2023-01-21 11:29:34.523862: step: 776/529, loss: 0.0282000545412302 2023-01-21 11:29:35.644936: step: 780/529, loss: 0.0027191161643713713 2023-01-21 11:29:36.755103: step: 784/529, loss: 0.042397309094667435 2023-01-21 11:29:37.849812: step: 788/529, loss: 0.010380363091826439 2023-01-21 11:29:38.981643: step: 792/529, loss: 0.0034318447578698397 2023-01-21 11:29:40.110812: step: 796/529, loss: 0.006031608674675226 2023-01-21 11:29:41.234141: step: 800/529, loss: 0.036104727536439896 2023-01-21 11:29:42.359645: step: 804/529, loss: 0.027247143909335136 2023-01-21 11:29:43.488885: step: 808/529, loss: 0.0003262520185671747 2023-01-21 11:29:44.586331: step: 812/529, loss: 0.017936134710907936 2023-01-21 11:29:45.726545: step: 816/529, loss: 0.02042541652917862 2023-01-21 11:29:46.820524: step: 820/529, loss: 0.022796964272856712 2023-01-21 11:29:47.948569: step: 824/529, loss: 0.1252817064523697 2023-01-21 11:29:49.081965: step: 828/529, loss: 0.01477670669555664 2023-01-21 11:29:50.225472: step: 832/529, loss: 0.08678007125854492 2023-01-21 11:29:51.374682: step: 836/529, loss: 0.017061900347471237 2023-01-21 11:29:52.486006: step: 840/529, loss: 0.019028853625059128 2023-01-21 11:29:53.615511: step: 844/529, loss: 0.0069172861985862255 2023-01-21 11:29:54.726860: step: 848/529, loss: 0.010430335998535156 2023-01-21 11:29:55.852814: step: 852/529, loss: 0.02069452591240406 2023-01-21 11:29:57.022997: step: 856/529, loss: 0.020403005182743073 2023-01-21 11:29:58.195538: step: 860/529, loss: 0.0027370452880859375 2023-01-21 11:29:59.302631: step: 864/529, loss: 0.004788017366081476 2023-01-21 11:30:00.403489: step: 868/529, loss: 0.03317966312170029 2023-01-21 11:30:01.520463: step: 872/529, loss: 0.054764557629823685 2023-01-21 11:30:02.646139: step: 876/529, loss: 0.0829157829284668 2023-01-21 11:30:03.767384: step: 880/529, loss: 0.027825355529785156 2023-01-21 11:30:04.875845: step: 884/529, loss: 0.05098400264978409 2023-01-21 11:30:05.985379: step: 888/529, loss: 0.0057725911028683186 2023-01-21 11:30:07.082303: step: 892/529, loss: 0.03614835813641548 2023-01-21 11:30:08.207234: step: 896/529, loss: 0.0006322860717773438 2023-01-21 11:30:09.342549: step: 900/529, loss: 0.002818298526108265 2023-01-21 11:30:10.469622: step: 904/529, loss: 0.26575013995170593 2023-01-21 11:30:11.622618: step: 908/529, loss: 0.007422256283462048 2023-01-21 11:30:12.717212: step: 912/529, loss: 0.01878061331808567 2023-01-21 11:30:13.860513: step: 916/529, loss: 0.007219791878014803 2023-01-21 11:30:14.966957: step: 920/529, loss: 0.0006025314796715975 2023-01-21 11:30:16.091225: step: 924/529, loss: 0.0012438774574548006 2023-01-21 11:30:17.193676: step: 928/529, loss: 0.020307209342718124 2023-01-21 11:30:18.294717: step: 932/529, loss: 0.0311342254281044 2023-01-21 11:30:19.410063: step: 936/529, loss: 0.0036264420486986637 2023-01-21 11:30:20.488772: step: 940/529, loss: 0.003145313123241067 2023-01-21 11:30:21.606125: step: 944/529, loss: 0.017205860465765 2023-01-21 11:30:22.763454: step: 948/529, loss: 0.041124965995550156 2023-01-21 11:30:23.872005: step: 952/529, loss: 0.04513072967529297 2023-01-21 11:30:25.012542: step: 956/529, loss: 0.007253456395119429 2023-01-21 11:30:26.083849: step: 960/529, loss: 0.029857492074370384 2023-01-21 11:30:27.193374: step: 964/529, loss: 0.07467031478881836 2023-01-21 11:30:28.287961: step: 968/529, loss: 0.04811658710241318 2023-01-21 11:30:29.417722: step: 972/529, loss: 0.011242199689149857 2023-01-21 11:30:30.600325: step: 976/529, loss: 0.0834631472826004 2023-01-21 11:30:31.725629: step: 980/529, loss: 0.007007026579231024 2023-01-21 11:30:32.849447: step: 984/529, loss: 0.00588116655126214 2023-01-21 11:30:33.998575: step: 988/529, loss: 0.027263546362519264 2023-01-21 11:30:35.093709: step: 992/529, loss: 0.005964470095932484 2023-01-21 11:30:36.206248: step: 996/529, loss: 0.008327484130859375 2023-01-21 11:30:37.356629: step: 1000/529, loss: 0.0031923295464366674 2023-01-21 11:30:38.465618: step: 1004/529, loss: 0.03464236110448837 2023-01-21 11:30:39.560623: step: 1008/529, loss: 0.041654013097286224 2023-01-21 11:30:40.679011: step: 1012/529, loss: 0.011207103729248047 2023-01-21 11:30:41.826251: step: 1016/529, loss: 0.00887765921652317 2023-01-21 11:30:42.941506: step: 1020/529, loss: 0.007322216406464577 2023-01-21 11:30:44.023567: step: 1024/529, loss: 0.0009346485021524131 2023-01-21 11:30:45.111550: step: 1028/529, loss: 0.0038628578186035156 2023-01-21 11:30:46.214412: step: 1032/529, loss: 0.05310516431927681 2023-01-21 11:30:47.344063: step: 1036/529, loss: 0.007705688942223787 2023-01-21 11:30:48.437462: step: 1040/529, loss: 0.005209731869399548 2023-01-21 11:30:49.542141: step: 1044/529, loss: 0.029564954340457916 2023-01-21 11:30:50.679348: step: 1048/529, loss: 0.6632478833198547 2023-01-21 11:30:51.805087: step: 1052/529, loss: 0.045206259936094284 2023-01-21 11:30:52.920513: step: 1056/529, loss: 0.036804962903261185 2023-01-21 11:30:54.047093: step: 1060/529, loss: 0.06002082675695419 2023-01-21 11:30:55.142247: step: 1064/529, loss: 0.030590534210205078 2023-01-21 11:30:56.249049: step: 1068/529, loss: 0.02960033528506756 2023-01-21 11:30:57.366797: step: 1072/529, loss: 0.0005201339954510331 2023-01-21 11:30:58.527698: step: 1076/529, loss: 0.07917747646570206 2023-01-21 11:30:59.644263: step: 1080/529, loss: 0.012373924255371094 2023-01-21 11:31:00.790044: step: 1084/529, loss: 0.03654327616095543 2023-01-21 11:31:01.902033: step: 1088/529, loss: 0.025353502482175827 2023-01-21 11:31:03.026323: step: 1092/529, loss: 0.015633774921298027 2023-01-21 11:31:04.133323: step: 1096/529, loss: 0.0013218403328210115 2023-01-21 11:31:05.229360: step: 1100/529, loss: 0.0006042480817995965 2023-01-21 11:31:06.338030: step: 1104/529, loss: 0.0016802788013592362 2023-01-21 11:31:07.455577: step: 1108/529, loss: 0.0032924653496593237 2023-01-21 11:31:08.647587: step: 1112/529, loss: 0.00364513392560184 2023-01-21 11:31:09.808769: step: 1116/529, loss: 0.001890039537101984 2023-01-21 11:31:10.977144: step: 1120/529, loss: 0.27777138352394104 2023-01-21 11:31:12.101997: step: 1124/529, loss: 0.006247472949326038 2023-01-21 11:31:13.246507: step: 1128/529, loss: 0.007719802670180798 2023-01-21 11:31:14.352060: step: 1132/529, loss: 0.026265526190400124 2023-01-21 11:31:15.468560: step: 1136/529, loss: 0.0666351318359375 2023-01-21 11:31:16.614276: step: 1140/529, loss: 0.048487283289432526 2023-01-21 11:31:17.718210: step: 1144/529, loss: 0.09375075995922089 2023-01-21 11:31:18.808756: step: 1148/529, loss: 0.0002989768981933594 2023-01-21 11:31:19.894039: step: 1152/529, loss: 0.00042276381282135844 2023-01-21 11:31:21.000686: step: 1156/529, loss: 0.042872242629528046 2023-01-21 11:31:22.117018: step: 1160/529, loss: 0.0010498047340661287 2023-01-21 11:31:23.275932: step: 1164/529, loss: 0.011212730780243874 2023-01-21 11:31:24.366780: step: 1168/529, loss: 0.0013476371532306075 2023-01-21 11:31:25.471945: step: 1172/529, loss: 0.11217289417982101 2023-01-21 11:31:26.599603: step: 1176/529, loss: 0.06064195930957794 2023-01-21 11:31:27.725295: step: 1180/529, loss: 0.0345364585518837 2023-01-21 11:31:28.852760: step: 1184/529, loss: 0.08045224845409393 2023-01-21 11:31:29.947129: step: 1188/529, loss: 0.003103935858234763 2023-01-21 11:31:31.086238: step: 1192/529, loss: 0.0014510632026940584 2023-01-21 11:31:32.173921: step: 1196/529, loss: 0.009851646609604359 2023-01-21 11:31:33.294053: step: 1200/529, loss: 0.013622951693832874 2023-01-21 11:31:34.400822: step: 1204/529, loss: 0.20609799027442932 2023-01-21 11:31:35.483830: step: 1208/529, loss: 0.005534553900361061 2023-01-21 11:31:36.597166: step: 1212/529, loss: 0.05313606560230255 2023-01-21 11:31:37.709263: step: 1216/529, loss: 0.034075211733579636 2023-01-21 11:31:38.832208: step: 1220/529, loss: 0.00260162353515625 2023-01-21 11:31:39.968306: step: 1224/529, loss: 5.455017162603326e-05 2023-01-21 11:31:41.073611: step: 1228/529, loss: 0.0028947831597179174 2023-01-21 11:31:42.183087: step: 1232/529, loss: 0.0009201050270348787 2023-01-21 11:31:43.327380: step: 1236/529, loss: 0.01868276670575142 2023-01-21 11:31:44.493011: step: 1240/529, loss: 0.02008838765323162 2023-01-21 11:31:45.608788: step: 1244/529, loss: 0.0003276825009379536 2023-01-21 11:31:46.776240: step: 1248/529, loss: 0.8614374995231628 2023-01-21 11:31:47.893029: step: 1252/529, loss: 0.0036573412362486124 2023-01-21 11:31:49.017678: step: 1256/529, loss: 0.0009625435341149569 2023-01-21 11:31:50.171911: step: 1260/529, loss: 0.03019714541733265 2023-01-21 11:31:51.295986: step: 1264/529, loss: 0.005444717593491077 2023-01-21 11:31:52.435948: step: 1268/529, loss: 0.0035126684233546257 2023-01-21 11:31:53.604853: step: 1272/529, loss: 0.011497306637465954 2023-01-21 11:31:54.737401: step: 1276/529, loss: 0.0055480957962572575 2023-01-21 11:31:55.851894: step: 1280/529, loss: 0.006733989808708429 2023-01-21 11:31:56.959412: step: 1284/529, loss: 0.04899817705154419 2023-01-21 11:31:58.082387: step: 1288/529, loss: 0.011456918902695179 2023-01-21 11:31:59.224274: step: 1292/529, loss: 0.027710085734725 2023-01-21 11:32:00.316888: step: 1296/529, loss: 0.01874075084924698 2023-01-21 11:32:01.460806: step: 1300/529, loss: 0.012926960363984108 2023-01-21 11:32:02.552742: step: 1304/529, loss: 0.0043496135622262955 2023-01-21 11:32:03.645675: step: 1308/529, loss: 0.013659668155014515 2023-01-21 11:32:04.726377: step: 1312/529, loss: 0.0006468773353844881 2023-01-21 11:32:05.812303: step: 1316/529, loss: 0.040293123573064804 2023-01-21 11:32:06.907960: step: 1320/529, loss: 0.00016374587721657008 2023-01-21 11:32:08.042483: step: 1324/529, loss: 0.018106840550899506 2023-01-21 11:32:09.150406: step: 1328/529, loss: 0.09573593735694885 2023-01-21 11:32:10.243263: step: 1332/529, loss: 0.009221458807587624 2023-01-21 11:32:11.335948: step: 1336/529, loss: 0.005654716864228249 2023-01-21 11:32:12.464062: step: 1340/529, loss: 0.007609749212861061 2023-01-21 11:32:13.580009: step: 1344/529, loss: 0.017772484570741653 2023-01-21 11:32:14.712789: step: 1348/529, loss: 0.004604435060173273 2023-01-21 11:32:15.815361: step: 1352/529, loss: 0.044849492609500885 2023-01-21 11:32:16.945378: step: 1356/529, loss: 0.019938278943300247 2023-01-21 11:32:18.064959: step: 1360/529, loss: 0.053911399096250534 2023-01-21 11:32:19.167656: step: 1364/529, loss: 0.0022809982765465975 2023-01-21 11:32:20.271136: step: 1368/529, loss: 0.0145721435546875 2023-01-21 11:32:21.387640: step: 1372/529, loss: 0.002082443330436945 2023-01-21 11:32:22.500282: step: 1376/529, loss: 0.07693834602832794 2023-01-21 11:32:23.606948: step: 1380/529, loss: 0.11413994431495667 2023-01-21 11:32:24.757271: step: 1384/529, loss: 0.06834125518798828 2023-01-21 11:32:25.861751: step: 1388/529, loss: 0.057936858385801315 2023-01-21 11:32:26.984384: step: 1392/529, loss: 0.05052356794476509 2023-01-21 11:32:28.092027: step: 1396/529, loss: 0.03631248697638512 2023-01-21 11:32:29.182446: step: 1400/529, loss: 0.023670196533203125 2023-01-21 11:32:30.296921: step: 1404/529, loss: 0.04947681725025177 2023-01-21 11:32:31.410031: step: 1408/529, loss: 0.10579337924718857 2023-01-21 11:32:32.541593: step: 1412/529, loss: 0.0837366133928299 2023-01-21 11:32:33.654222: step: 1416/529, loss: 0.00044727325439453125 2023-01-21 11:32:34.757043: step: 1420/529, loss: 0.01910400576889515 2023-01-21 11:32:35.864404: step: 1424/529, loss: 0.003924655728042126 2023-01-21 11:32:36.962283: step: 1428/529, loss: 0.0013859033351764083 2023-01-21 11:32:38.091496: step: 1432/529, loss: 0.002593517303466797 2023-01-21 11:32:39.233946: step: 1436/529, loss: 0.000583648681640625 2023-01-21 11:32:40.307031: step: 1440/529, loss: 0.00019402503676246852 2023-01-21 11:32:41.378937: step: 1444/529, loss: 0.042620182037353516 2023-01-21 11:32:42.503190: step: 1448/529, loss: 0.016722679138183594 2023-01-21 11:32:43.620095: step: 1452/529, loss: 0.00067052838858217 2023-01-21 11:32:44.760871: step: 1456/529, loss: 0.0017459869850426912 2023-01-21 11:32:45.865652: step: 1460/529, loss: 0.08023948967456818 2023-01-21 11:32:46.986699: step: 1464/529, loss: 0.009878730401396751 2023-01-21 11:32:48.094364: step: 1468/529, loss: 0.023539256304502487 2023-01-21 11:32:49.211928: step: 1472/529, loss: 0.019981766119599342 2023-01-21 11:32:50.309314: step: 1476/529, loss: 0.010811900720000267 2023-01-21 11:32:51.450213: step: 1480/529, loss: 0.062145281583070755 2023-01-21 11:32:52.579175: step: 1484/529, loss: 0.003956461325287819 2023-01-21 11:32:53.717858: step: 1488/529, loss: 0.009970379061996937 2023-01-21 11:32:54.828871: step: 1492/529, loss: 0.020316505804657936 2023-01-21 11:32:55.953338: step: 1496/529, loss: 0.0733499526977539 2023-01-21 11:32:57.081627: step: 1500/529, loss: 0.010421562008559704 2023-01-21 11:32:58.196834: step: 1504/529, loss: 0.010651921853423119 2023-01-21 11:32:59.291013: step: 1508/529, loss: 0.05582462251186371 2023-01-21 11:33:00.411437: step: 1512/529, loss: 0.004432773683220148 2023-01-21 11:33:01.535984: step: 1516/529, loss: 0.026908492669463158 2023-01-21 11:33:02.646512: step: 1520/529, loss: 0.001308441162109375 2023-01-21 11:33:03.781415: step: 1524/529, loss: 0.00920639093965292 2023-01-21 11:33:04.895725: step: 1528/529, loss: 0.07841330021619797 2023-01-21 11:33:06.046348: step: 1532/529, loss: 0.011953807435929775 2023-01-21 11:33:07.164299: step: 1536/529, loss: 0.05583386868238449 2023-01-21 11:33:08.280225: step: 1540/529, loss: 0.027892112731933594 2023-01-21 11:33:09.381937: step: 1544/529, loss: 0.02685546875 2023-01-21 11:33:10.487181: step: 1548/529, loss: 0.04306450113654137 2023-01-21 11:33:11.612838: step: 1552/529, loss: 0.0007810592651367188 2023-01-21 11:33:12.726063: step: 1556/529, loss: 0.0005102157592773438 2023-01-21 11:33:13.851316: step: 1560/529, loss: 0.027725793421268463 2023-01-21 11:33:14.959979: step: 1564/529, loss: 0.008467674255371094 2023-01-21 11:33:16.058113: step: 1568/529, loss: 0.0001546144630992785 2023-01-21 11:33:17.180097: step: 1572/529, loss: 0.014438152313232422 2023-01-21 11:33:18.315217: step: 1576/529, loss: 0.042174626141786575 2023-01-21 11:33:19.461312: step: 1580/529, loss: 0.02753734588623047 2023-01-21 11:33:20.606534: step: 1584/529, loss: 0.00016841889009810984 2023-01-21 11:33:21.741236: step: 1588/529, loss: 0.0010137557983398438 2023-01-21 11:33:22.856392: step: 1592/529, loss: 0.017825031653046608 2023-01-21 11:33:23.945099: step: 1596/529, loss: 0.0008176803821697831 2023-01-21 11:33:25.074366: step: 1600/529, loss: 0.018732547760009766 2023-01-21 11:33:26.207924: step: 1604/529, loss: 0.0032693862449377775 2023-01-21 11:33:27.351902: step: 1608/529, loss: 0.04006028175354004 2023-01-21 11:33:28.470073: step: 1612/529, loss: 0.002649164292961359 2023-01-21 11:33:29.580383: step: 1616/529, loss: 0.0437227264046669 2023-01-21 11:33:30.701693: step: 1620/529, loss: 0.02005309984087944 2023-01-21 11:33:31.828001: step: 1624/529, loss: 0.0009645462268963456 2023-01-21 11:33:32.977219: step: 1628/529, loss: 0.007365926168859005 2023-01-21 11:33:34.067929: step: 1632/529, loss: 0.09861570596694946 2023-01-21 11:33:35.187360: step: 1636/529, loss: 0.060585781931877136 2023-01-21 11:33:36.323794: step: 1640/529, loss: 0.03949327394366264 2023-01-21 11:33:37.491238: step: 1644/529, loss: 0.00044736862764693797 2023-01-21 11:33:38.616551: step: 1648/529, loss: 0.012586403638124466 2023-01-21 11:33:39.734330: step: 1652/529, loss: 0.005134296603500843 2023-01-21 11:33:40.863946: step: 1656/529, loss: 0.015186309814453125 2023-01-21 11:33:41.950009: step: 1660/529, loss: 0.02126135863363743 2023-01-21 11:33:43.056296: step: 1664/529, loss: 0.006412029266357422 2023-01-21 11:33:44.153678: step: 1668/529, loss: 0.025492003187537193 2023-01-21 11:33:45.241262: step: 1672/529, loss: 0.005620527546852827 2023-01-21 11:33:46.330807: step: 1676/529, loss: 0.009557342156767845 2023-01-21 11:33:47.460945: step: 1680/529, loss: 0.010870552621781826 2023-01-21 11:33:48.603362: step: 1684/529, loss: 0.03641529381275177 2023-01-21 11:33:49.716393: step: 1688/529, loss: 0.00741653423756361 2023-01-21 11:33:50.838875: step: 1692/529, loss: 0.013895892538130283 2023-01-21 11:33:51.947179: step: 1696/529, loss: 0.03253583982586861 2023-01-21 11:33:53.078011: step: 1700/529, loss: 0.016987323760986328 2023-01-21 11:33:54.183542: step: 1704/529, loss: 0.002588987350463867 2023-01-21 11:33:55.301073: step: 1708/529, loss: 0.0005561828729696572 2023-01-21 11:33:56.409461: step: 1712/529, loss: 0.008816909044981003 2023-01-21 11:33:57.550238: step: 1716/529, loss: 0.019388962537050247 2023-01-21 11:33:58.676589: step: 1720/529, loss: 0.015825558453798294 2023-01-21 11:33:59.794524: step: 1724/529, loss: 0.016736602410674095 2023-01-21 11:34:00.894173: step: 1728/529, loss: 0.0016808033687993884 2023-01-21 11:34:01.997039: step: 1732/529, loss: 0.007079220376908779 2023-01-21 11:34:03.093410: step: 1736/529, loss: 0.008332825265824795 2023-01-21 11:34:04.191161: step: 1740/529, loss: 0.014214038848876953 2023-01-21 11:34:05.334906: step: 1744/529, loss: 0.04362468793988228 2023-01-21 11:34:06.448142: step: 1748/529, loss: 0.016611767932772636 2023-01-21 11:34:07.588537: step: 1752/529, loss: 0.0069817546755075455 2023-01-21 11:34:08.708727: step: 1756/529, loss: 0.0020694732666015625 2023-01-21 11:34:09.835928: step: 1760/529, loss: 0.010686875320971012 2023-01-21 11:34:10.945506: step: 1764/529, loss: 0.01836833916604519 2023-01-21 11:34:12.070990: step: 1768/529, loss: 0.0010773659450933337 2023-01-21 11:34:13.205772: step: 1772/529, loss: 0.011740017682313919 2023-01-21 11:34:14.319526: step: 1776/529, loss: 0.0022144317626953125 2023-01-21 11:34:15.427503: step: 1780/529, loss: 0.04025249555706978 2023-01-21 11:34:16.549860: step: 1784/529, loss: 0.0018579483730718493 2023-01-21 11:34:17.676455: step: 1788/529, loss: 0.05710163339972496 2023-01-21 11:34:18.791741: step: 1792/529, loss: 0.0002894401550292969 2023-01-21 11:34:19.922485: step: 1796/529, loss: 0.017987919971346855 2023-01-21 11:34:21.041485: step: 1800/529, loss: 0.0022237778175622225 2023-01-21 11:34:22.142971: step: 1804/529, loss: 0.061766814440488815 2023-01-21 11:34:23.241408: step: 1808/529, loss: 0.05389413982629776 2023-01-21 11:34:24.358855: step: 1812/529, loss: 0.031571388244628906 2023-01-21 11:34:25.489547: step: 1816/529, loss: 0.01135721243917942 2023-01-21 11:34:26.652905: step: 1820/529, loss: 0.04927196353673935 2023-01-21 11:34:27.811672: step: 1824/529, loss: 0.036455631256103516 2023-01-21 11:34:28.965180: step: 1828/529, loss: 0.023421669378876686 2023-01-21 11:34:30.047345: step: 1832/529, loss: 0.046378135681152344 2023-01-21 11:34:31.154900: step: 1836/529, loss: 0.013359260745346546 2023-01-21 11:34:32.285768: step: 1840/529, loss: 0.9466407895088196 2023-01-21 11:34:33.364645: step: 1844/529, loss: 0.007301521487534046 2023-01-21 11:34:34.519317: step: 1848/529, loss: 0.007399082183837891 2023-01-21 11:34:35.611075: step: 1852/529, loss: 5.991458965581842e-05 2023-01-21 11:34:36.741546: step: 1856/529, loss: 0.01962299272418022 2023-01-21 11:34:37.873815: step: 1860/529, loss: 0.03183336555957794 2023-01-21 11:34:39.002169: step: 1864/529, loss: 0.003655338194221258 2023-01-21 11:34:40.093865: step: 1868/529, loss: 0.023988496512174606 2023-01-21 11:34:41.176745: step: 1872/529, loss: 0.015140724368393421 2023-01-21 11:34:42.302635: step: 1876/529, loss: 0.04513759911060333 2023-01-21 11:34:43.419184: step: 1880/529, loss: 0.025017166510224342 2023-01-21 11:34:44.535910: step: 1884/529, loss: 0.013048839755356312 2023-01-21 11:34:45.651054: step: 1888/529, loss: 0.0011128425830975175 2023-01-21 11:34:46.783788: step: 1892/529, loss: 5.059242539573461e-05 2023-01-21 11:34:47.924932: step: 1896/529, loss: 0.013761520385742188 2023-01-21 11:34:49.069130: step: 1900/529, loss: 0.0029476166237145662 2023-01-21 11:34:50.184615: step: 1904/529, loss: 0.000491285347379744 2023-01-21 11:34:51.309381: step: 1908/529, loss: 0.10273055732250214 2023-01-21 11:34:52.407561: step: 1912/529, loss: 1.6620161533355713 2023-01-21 11:34:53.486759: step: 1916/529, loss: 0.0021393776405602694 2023-01-21 11:34:54.602454: step: 1920/529, loss: 0.0120093347504735 2023-01-21 11:34:55.695215: step: 1924/529, loss: 0.04906215891242027 2023-01-21 11:34:56.795011: step: 1928/529, loss: 0.0005199432489462197 2023-01-21 11:34:57.911164: step: 1932/529, loss: 0.01281890831887722 2023-01-21 11:34:59.039017: step: 1936/529, loss: 0.03534068912267685 2023-01-21 11:35:00.143563: step: 1940/529, loss: 0.04288816452026367 2023-01-21 11:35:01.278225: step: 1944/529, loss: 0.013931083492934704 2023-01-21 11:35:02.431260: step: 1948/529, loss: 0.00337033299729228 2023-01-21 11:35:03.569877: step: 1952/529, loss: 0.0031180856749415398 2023-01-21 11:35:04.658935: step: 1956/529, loss: 0.004653358366340399 2023-01-21 11:35:05.781872: step: 1960/529, loss: 0.02243976667523384 2023-01-21 11:35:06.882398: step: 1964/529, loss: 0.032325174659490585 2023-01-21 11:35:08.007584: step: 1968/529, loss: 0.016070544719696045 2023-01-21 11:35:09.120633: step: 1972/529, loss: 0.043839454650878906 2023-01-21 11:35:10.233381: step: 1976/529, loss: 0.009467696771025658 2023-01-21 11:35:11.328306: step: 1980/529, loss: 0.05063324049115181 2023-01-21 11:35:12.456762: step: 1984/529, loss: 0.0005445480346679688 2023-01-21 11:35:13.571576: step: 1988/529, loss: 0.013491821475327015 2023-01-21 11:35:14.694046: step: 1992/529, loss: 0.004249191377311945 2023-01-21 11:35:15.836795: step: 1996/529, loss: 0.028296470642089844 2023-01-21 11:35:16.948818: step: 2000/529, loss: 0.0012382508721202612 2023-01-21 11:35:18.055384: step: 2004/529, loss: 0.00345611572265625 2023-01-21 11:35:19.201613: step: 2008/529, loss: 0.003939628601074219 2023-01-21 11:35:20.332372: step: 2012/529, loss: 0.0013811112148687243 2023-01-21 11:35:21.449254: step: 2016/529, loss: 0.004080319777131081 2023-01-21 11:35:22.589974: step: 2020/529, loss: 0.004145240876823664 2023-01-21 11:35:23.687068: step: 2024/529, loss: 0.0001321792515227571 2023-01-21 11:35:24.784775: step: 2028/529, loss: 0.002094364259392023 2023-01-21 11:35:25.902176: step: 2032/529, loss: 0.009102821350097656 2023-01-21 11:35:27.017617: step: 2036/529, loss: 0.0030755999032408 2023-01-21 11:35:28.117229: step: 2040/529, loss: 0.008584881201386452 2023-01-21 11:35:29.224452: step: 2044/529, loss: 0.0020897865761071444 2023-01-21 11:35:30.341961: step: 2048/529, loss: 0.0003311157342977822 2023-01-21 11:35:31.451531: step: 2052/529, loss: 0.022060489282011986 2023-01-21 11:35:32.587597: step: 2056/529, loss: 0.0026691914536058903 2023-01-21 11:35:33.728104: step: 2060/529, loss: 0.004038190934807062 2023-01-21 11:35:34.869675: step: 2064/529, loss: 0.0020453452598303556 2023-01-21 11:35:35.997978: step: 2068/529, loss: 0.010776758193969727 2023-01-21 11:35:37.136244: step: 2072/529, loss: 0.08061805367469788 2023-01-21 11:35:38.264201: step: 2076/529, loss: 0.09675603359937668 2023-01-21 11:35:39.386059: step: 2080/529, loss: 0.0015539169544354081 2023-01-21 11:35:40.530588: step: 2084/529, loss: 0.06428580731153488 2023-01-21 11:35:41.664618: step: 2088/529, loss: 0.013783454895019531 2023-01-21 11:35:42.779520: step: 2092/529, loss: 0.024349594488739967 2023-01-21 11:35:43.875989: step: 2096/529, loss: 0.030339624732732773 2023-01-21 11:35:44.988726: step: 2100/529, loss: 0.0008707046508789062 2023-01-21 11:35:46.123497: step: 2104/529, loss: 0.3763510584831238 2023-01-21 11:35:47.250541: step: 2108/529, loss: 0.01416854839771986 2023-01-21 11:35:48.358108: step: 2112/529, loss: 0.04438743740320206 2023-01-21 11:35:49.451572: step: 2116/529, loss: 0.0005743026849813759 ================================================== Loss: 0.032 -------------------- Dev: {'event': {'p': 0.5784114052953157, 'r': 0.7563249001331558, 'f1': 0.6555106751298327}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Test: {'event': {'p': 0.6442486085343229, 'r': 0.7838600451467269, 'f1': 0.7072301425661915}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Chinese: {'event': {'p': 0.5977011494252874, 'r': 0.9629629629629629, 'f1': 0.7375886524822696}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Korean: {'event': {'p': 0.5714285714285714, 'r': 0.5714285714285714, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} 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': 17} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 18 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:36:30.186535: step: 4/529, loss: 0.002998876851052046 2023-01-21 11:36:31.327801: step: 8/529, loss: 0.017006922513246536 2023-01-21 11:36:32.432212: step: 12/529, loss: 0.003844833467155695 2023-01-21 11:36:33.562307: step: 16/529, loss: 0.0018792152404785156 2023-01-21 11:36:34.716163: step: 20/529, loss: 0.031246567144989967 2023-01-21 11:36:35.837975: step: 24/529, loss: 0.009868621826171875 2023-01-21 11:36:36.941162: step: 28/529, loss: 0.009680366143584251 2023-01-21 11:36:38.089063: step: 32/529, loss: 0.03929562494158745 2023-01-21 11:36:39.200508: step: 36/529, loss: 0.00036087038461118937 2023-01-21 11:36:40.309761: step: 40/529, loss: 0.023014290258288383 2023-01-21 11:36:41.442136: step: 44/529, loss: 0.029668522998690605 2023-01-21 11:36:42.578826: step: 48/529, loss: 0.02438831329345703 2023-01-21 11:36:43.748336: step: 52/529, loss: 0.04918208345770836 2023-01-21 11:36:44.869373: step: 56/529, loss: 0.0012639046180993319 2023-01-21 11:36:45.954099: step: 60/529, loss: 0.003215980716049671 2023-01-21 11:36:47.043598: step: 64/529, loss: 0.001334381173364818 2023-01-21 11:36:48.160507: step: 68/529, loss: 0.03056182898581028 2023-01-21 11:36:49.274619: step: 72/529, loss: 0.030424689874053 2023-01-21 11:36:50.429978: step: 76/529, loss: 5.264282663119957e-05 2023-01-21 11:36:51.555729: step: 80/529, loss: 0.005367946811020374 2023-01-21 11:36:52.684510: step: 84/529, loss: 0.001497554825618863 2023-01-21 11:36:53.811074: step: 88/529, loss: 0.04599914699792862 2023-01-21 11:36:54.930024: step: 92/529, loss: 0.024938583374023438 2023-01-21 11:36:56.052373: step: 96/529, loss: 0.02764587476849556 2023-01-21 11:36:57.218118: step: 100/529, loss: 0.004647445864975452 2023-01-21 11:36:58.348423: step: 104/529, loss: 0.04710127040743828 2023-01-21 11:36:59.494816: step: 108/529, loss: 0.00010104477405548096 2023-01-21 11:37:00.616484: step: 112/529, loss: 0.007490253541618586 2023-01-21 11:37:01.716513: step: 116/529, loss: 0.06377296149730682 2023-01-21 11:37:02.827047: step: 120/529, loss: 0.0008453368791379035 2023-01-21 11:37:03.900012: step: 124/529, loss: 0.00010070800635730848 2023-01-21 11:37:05.046487: step: 128/529, loss: 0.03658290207386017 2023-01-21 11:37:06.172474: step: 132/529, loss: 0.07246951758861542 2023-01-21 11:37:07.305321: step: 136/529, loss: 0.06288440525531769 2023-01-21 11:37:08.408201: step: 140/529, loss: 0.028320789337158203 2023-01-21 11:37:09.508222: step: 144/529, loss: 0.007649326231330633 2023-01-21 11:37:10.613651: step: 148/529, loss: 0.005911732092499733 2023-01-21 11:37:11.744937: step: 152/529, loss: 0.017657184973359108 2023-01-21 11:37:12.860307: step: 156/529, loss: 0.012812996283173561 2023-01-21 11:37:13.952582: step: 160/529, loss: 0.0008520126575604081 2023-01-21 11:37:15.076732: step: 164/529, loss: 0.007799720857292414 2023-01-21 11:37:16.214239: step: 168/529, loss: 0.7407258749008179 2023-01-21 11:37:17.322302: step: 172/529, loss: 0.0006013870588503778 2023-01-21 11:37:18.436194: step: 176/529, loss: 4.6253204345703125e-05 2023-01-21 11:37:19.544635: step: 180/529, loss: 0.00033855438232421875 2023-01-21 11:37:20.685411: step: 184/529, loss: 0.035860441625118256 2023-01-21 11:37:21.770478: step: 188/529, loss: 0.005360031500458717 2023-01-21 11:37:22.881178: step: 192/529, loss: 0.0020776509772986174 2023-01-21 11:37:23.983880: step: 196/529, loss: 0.008920478634536266 2023-01-21 11:37:25.082526: step: 200/529, loss: 0.00017471313185524195 2023-01-21 11:37:26.220519: step: 204/529, loss: 0.002953433897346258 2023-01-21 11:37:27.325550: step: 208/529, loss: 0.0018602609634399414 2023-01-21 11:37:28.414198: step: 212/529, loss: 0.022137831896543503 2023-01-21 11:37:29.541602: step: 216/529, loss: 0.059075355529785156 2023-01-21 11:37:30.647552: step: 220/529, loss: 0.0029333115089684725 2023-01-21 11:37:31.748307: step: 224/529, loss: 0.05030112341046333 2023-01-21 11:37:32.911610: step: 228/529, loss: 0.034888651221990585 2023-01-21 11:37:34.030877: step: 232/529, loss: 0.01607859879732132 2023-01-21 11:37:35.144044: step: 236/529, loss: 0.018686486408114433 2023-01-21 11:37:36.275095: step: 240/529, loss: 0.0004830360412597656 2023-01-21 11:37:37.387874: step: 244/529, loss: 0.7120140790939331 2023-01-21 11:37:38.496573: step: 248/529, loss: 0.011679649353027344 2023-01-21 11:37:39.641293: step: 252/529, loss: 0.00034103391226381063 2023-01-21 11:37:40.792555: step: 256/529, loss: 0.021803760901093483 2023-01-21 11:37:41.903743: step: 260/529, loss: 0.00021166801161598414 2023-01-21 11:37:43.006061: step: 264/529, loss: 0.0002639770391397178 2023-01-21 11:37:44.129927: step: 268/529, loss: 0.043738484382629395 2023-01-21 11:37:45.223564: step: 272/529, loss: 0.005989647004753351 2023-01-21 11:37:46.340965: step: 276/529, loss: 0.01483612135052681 2023-01-21 11:37:47.450076: step: 280/529, loss: 0.051297903060913086 2023-01-21 11:37:48.566574: step: 284/529, loss: 0.026013851165771484 2023-01-21 11:37:49.693281: step: 288/529, loss: 0.010733413510024548 2023-01-21 11:37:50.788519: step: 292/529, loss: 0.0016160011291503906 2023-01-21 11:37:51.906329: step: 296/529, loss: 0.00427665701135993 2023-01-21 11:37:53.011834: step: 300/529, loss: 0.0008289337274618447 2023-01-21 11:37:54.107368: step: 304/529, loss: 0.03157844394445419 2023-01-21 11:37:55.206684: step: 308/529, loss: 0.013669395819306374 2023-01-21 11:37:56.334173: step: 312/529, loss: 0.2017984390258789 2023-01-21 11:37:57.459349: step: 316/529, loss: 0.048645973205566406 2023-01-21 11:37:58.546224: step: 320/529, loss: 0.0008381843799725175 2023-01-21 11:37:59.675331: step: 324/529, loss: 0.0011798859341070056 2023-01-21 11:38:00.788845: step: 328/529, loss: 0.0009162903297692537 2023-01-21 11:38:01.899310: step: 332/529, loss: 0.2096746563911438 2023-01-21 11:38:03.004827: step: 336/529, loss: 5.1212311518611386e-05 2023-01-21 11:38:04.102200: step: 340/529, loss: -2.026558377110632e-06 2023-01-21 11:38:05.189754: step: 344/529, loss: 0.0006766319856978953 2023-01-21 11:38:06.294611: step: 348/529, loss: 0.05172481760382652 2023-01-21 11:38:07.413680: step: 352/529, loss: 0.00019484758377075195 2023-01-21 11:38:08.541315: step: 356/529, loss: 0.0008844852563925087 2023-01-21 11:38:09.640272: step: 360/529, loss: 0.039675142616033554 2023-01-21 11:38:10.747405: step: 364/529, loss: 0.003983497619628906 2023-01-21 11:38:11.852650: step: 368/529, loss: 0.003175782971084118 2023-01-21 11:38:12.989971: step: 372/529, loss: 0.03502597659826279 2023-01-21 11:38:14.107717: step: 376/529, loss: 0.0018219947814941406 2023-01-21 11:38:15.213639: step: 380/529, loss: 0.012453271076083183 2023-01-21 11:38:16.313671: step: 384/529, loss: 0.0008155822870321572 2023-01-21 11:38:17.420244: step: 388/529, loss: 0.0072119953110814095 2023-01-21 11:38:18.560619: step: 392/529, loss: 0.00017900466627907008 2023-01-21 11:38:19.687166: step: 396/529, loss: 0.0008088112226687372 2023-01-21 11:38:20.832865: step: 400/529, loss: 0.013781547546386719 2023-01-21 11:38:21.939976: step: 404/529, loss: 0.006226074881851673 2023-01-21 11:38:23.096183: step: 408/529, loss: 0.0018296241760253906 2023-01-21 11:38:24.212870: step: 412/529, loss: 0.0027431489434093237 2023-01-21 11:38:25.331958: step: 416/529, loss: 0.031016064807772636 2023-01-21 11:38:26.472866: step: 420/529, loss: 0.04904906824231148 2023-01-21 11:38:27.578180: step: 424/529, loss: 0.000911331211682409 2023-01-21 11:38:28.690115: step: 428/529, loss: 0.01816740073263645 2023-01-21 11:38:29.808019: step: 432/529, loss: 0.06186337396502495 2023-01-21 11:38:30.929630: step: 436/529, loss: 0.02241954766213894 2023-01-21 11:38:32.038654: step: 440/529, loss: 0.00855712965130806 2023-01-21 11:38:33.157281: step: 444/529, loss: 0.06964576244354248 2023-01-21 11:38:34.244629: step: 448/529, loss: 0.01756115071475506 2023-01-21 11:38:35.373501: step: 452/529, loss: 0.10006542503833771 2023-01-21 11:38:36.517819: step: 456/529, loss: 0.005373287480324507 2023-01-21 11:38:37.623000: step: 460/529, loss: 0.0046129231341183186 2023-01-21 11:38:38.739833: step: 464/529, loss: 0.18387280404567719 2023-01-21 11:38:39.856760: step: 468/529, loss: -8.01086389401462e-06 2023-01-21 11:38:40.999420: step: 472/529, loss: 0.0007985115516930819 2023-01-21 11:38:42.116277: step: 476/529, loss: 0.007560539525002241 2023-01-21 11:38:43.232019: step: 480/529, loss: 0.007399940397590399 2023-01-21 11:38:44.332723: step: 484/529, loss: 0.01924762688577175 2023-01-21 11:38:45.458536: step: 488/529, loss: 0.18841572105884552 2023-01-21 11:38:46.581908: step: 492/529, loss: 0.0055946349166333675 2023-01-21 11:38:47.677476: step: 496/529, loss: 0.004186153411865234 2023-01-21 11:38:48.772508: step: 500/529, loss: 0.004673385992646217 2023-01-21 11:38:49.886144: step: 504/529, loss: 0.0016197204822674394 2023-01-21 11:38:51.019937: step: 508/529, loss: 0.00099945068359375 2023-01-21 11:38:52.137795: step: 512/529, loss: 0.009970665909349918 2023-01-21 11:38:53.265818: step: 516/529, loss: 0.0025717734824866056 2023-01-21 11:38:54.354109: step: 520/529, loss: 0.001628828002139926 2023-01-21 11:38:55.467326: step: 524/529, loss: 0.04026193544268608 2023-01-21 11:38:56.585935: step: 528/529, loss: 0.024524927139282227 2023-01-21 11:38:57.687964: step: 532/529, loss: 0.059690192341804504 2023-01-21 11:38:58.799237: step: 536/529, loss: 0.004295348655432463 2023-01-21 11:38:59.890037: step: 540/529, loss: 0.0007504463428631425 2023-01-21 11:39:01.008800: step: 544/529, loss: 0.0011619568103924394 2023-01-21 11:39:02.134325: step: 548/529, loss: 0.002483653835952282 2023-01-21 11:39:03.222824: step: 552/529, loss: 0.016633223742246628 2023-01-21 11:39:04.362359: step: 556/529, loss: 0.007696914952248335 2023-01-21 11:39:05.504137: step: 560/529, loss: 0.01152792014181614 2023-01-21 11:39:06.591560: step: 564/529, loss: 0.0009445667965337634 2023-01-21 11:39:07.695715: step: 568/529, loss: 0.06436033546924591 2023-01-21 11:39:08.798558: step: 572/529, loss: 0.007954549975693226 2023-01-21 11:39:09.920763: step: 576/529, loss: 0.00019187928410246968 2023-01-21 11:39:11.029689: step: 580/529, loss: 0.0014989853370934725 2023-01-21 11:39:12.155320: step: 584/529, loss: 0.027545928955078125 2023-01-21 11:39:13.257239: step: 588/529, loss: 0.03548269718885422 2023-01-21 11:39:14.381896: step: 592/529, loss: 0.039650533348321915 2023-01-21 11:39:15.508008: step: 596/529, loss: 0.10010643303394318 2023-01-21 11:39:16.646017: step: 600/529, loss: 0.04125375673174858 2023-01-21 11:39:17.766254: step: 604/529, loss: 0.0014337539905682206 2023-01-21 11:39:18.863044: step: 608/529, loss: 0.0053543089888989925 2023-01-21 11:39:19.987699: step: 612/529, loss: 0.015803862363100052 2023-01-21 11:39:21.108789: step: 616/529, loss: 0.006803130730986595 2023-01-21 11:39:22.224078: step: 620/529, loss: 0.0362127311527729 2023-01-21 11:39:23.304335: step: 624/529, loss: 0.01996450498700142 2023-01-21 11:39:24.436935: step: 628/529, loss: 0.044396210461854935 2023-01-21 11:39:25.551294: step: 632/529, loss: 0.01491236686706543 2023-01-21 11:39:26.652179: step: 636/529, loss: 0.001185512519441545 2023-01-21 11:39:27.743850: step: 640/529, loss: 0.013698006048798561 2023-01-21 11:39:28.937866: step: 644/529, loss: 0.10506153106689453 2023-01-21 11:39:30.065872: step: 648/529, loss: 0.01607685163617134 2023-01-21 11:39:31.172472: step: 652/529, loss: 0.0027788160368800163 2023-01-21 11:39:32.313521: step: 656/529, loss: 0.00062475202139467 2023-01-21 11:39:33.426096: step: 660/529, loss: 0.00652580289170146 2023-01-21 11:39:34.552130: step: 664/529, loss: 0.09088211506605148 2023-01-21 11:39:35.656486: step: 668/529, loss: 0.051072120666503906 2023-01-21 11:39:36.764606: step: 672/529, loss: 0.009678077884018421 2023-01-21 11:39:37.866318: step: 676/529, loss: 0.39746949076652527 2023-01-21 11:39:38.971631: step: 680/529, loss: 0.006929969880729914 2023-01-21 11:39:40.091289: step: 684/529, loss: 0.0039031982887536287 2023-01-21 11:39:41.219124: step: 688/529, loss: 0.025171469897031784 2023-01-21 11:39:42.399948: step: 692/529, loss: 0.11790676414966583 2023-01-21 11:39:43.492852: step: 696/529, loss: 0.011981487274169922 2023-01-21 11:39:44.634412: step: 700/529, loss: 0.00302543630823493 2023-01-21 11:39:45.745495: step: 704/529, loss: 0.0005494594806805253 2023-01-21 11:39:46.872452: step: 708/529, loss: 0.0020995140075683594 2023-01-21 11:39:47.989402: step: 712/529, loss: 0.009449435397982597 2023-01-21 11:39:49.101867: step: 716/529, loss: 0.001592922257259488 2023-01-21 11:39:50.195356: step: 720/529, loss: 0.012718009762465954 2023-01-21 11:39:51.290207: step: 724/529, loss: 2.8133392333984375e-05 2023-01-21 11:39:52.425937: step: 728/529, loss: 0.003930950071662664 2023-01-21 11:39:53.560795: step: 732/529, loss: 0.03534838184714317 2023-01-21 11:39:54.685556: step: 736/529, loss: 0.009404946118593216 2023-01-21 11:39:55.808969: step: 740/529, loss: 0.029752518981695175 2023-01-21 11:39:56.938562: step: 744/529, loss: 0.0020939826499670744 2023-01-21 11:39:58.060977: step: 748/529, loss: 0.009065628051757812 2023-01-21 11:39:59.180070: step: 752/529, loss: 0.008211707696318626 2023-01-21 11:40:00.303646: step: 756/529, loss: 0.003811454866081476 2023-01-21 11:40:01.454428: step: 760/529, loss: 0.056560613214969635 2023-01-21 11:40:02.565539: step: 764/529, loss: 0.0009943009354174137 2023-01-21 11:40:03.708787: step: 768/529, loss: 0.014568901620805264 2023-01-21 11:40:04.871890: step: 772/529, loss: 0.009718704037368298 2023-01-21 11:40:06.021723: step: 776/529, loss: 0.002511978382244706 2023-01-21 11:40:07.147885: step: 780/529, loss: 0.0017391204601153731 2023-01-21 11:40:08.261173: step: 784/529, loss: 0.0023900987580418587 2023-01-21 11:40:09.371604: step: 788/529, loss: 0.062331557273864746 2023-01-21 11:40:10.515726: step: 792/529, loss: 0.007258367724716663 2023-01-21 11:40:11.643666: step: 796/529, loss: 0.005029201507568359 2023-01-21 11:40:12.778622: step: 800/529, loss: 0.025862693786621094 2023-01-21 11:40:13.916296: step: 804/529, loss: 0.006263923831284046 2023-01-21 11:40:15.047022: step: 808/529, loss: 0.009171104989945889 2023-01-21 11:40:16.185291: step: 812/529, loss: 0.02699914015829563 2023-01-21 11:40:17.278394: step: 816/529, loss: 0.010309792123734951 2023-01-21 11:40:18.388676: step: 820/529, loss: 0.0490327812731266 2023-01-21 11:40:19.488945: step: 824/529, loss: 0.004374122712761164 2023-01-21 11:40:20.640483: step: 828/529, loss: 0.0011720657348632812 2023-01-21 11:40:21.737917: step: 832/529, loss: 0.005498313810676336 2023-01-21 11:40:22.855893: step: 836/529, loss: 0.03520713001489639 2023-01-21 11:40:23.962223: step: 840/529, loss: 0.08220825344324112 2023-01-21 11:40:25.077606: step: 844/529, loss: 0.005397772882133722 2023-01-21 11:40:26.202053: step: 848/529, loss: 0.021099282428622246 2023-01-21 11:40:27.347408: step: 852/529, loss: 0.006635904312133789 2023-01-21 11:40:28.471871: step: 856/529, loss: 0.003965330310165882 2023-01-21 11:40:29.552081: step: 860/529, loss: 0.00016231538029387593 2023-01-21 11:40:30.728208: step: 864/529, loss: 0.01697998121380806 2023-01-21 11:40:31.849206: step: 868/529, loss: 0.005830002017319202 2023-01-21 11:40:32.969192: step: 872/529, loss: 0.006504678633064032 2023-01-21 11:40:34.081805: step: 876/529, loss: 0.015044712461531162 2023-01-21 11:40:35.189916: step: 880/529, loss: 0.024628639221191406 2023-01-21 11:40:36.322195: step: 884/529, loss: 0.02321024052798748 2023-01-21 11:40:37.426789: step: 888/529, loss: 0.05142345651984215 2023-01-21 11:40:38.562730: step: 892/529, loss: 0.015437603928148746 2023-01-21 11:40:39.678826: step: 896/529, loss: 0.4070507884025574 2023-01-21 11:40:40.801887: step: 900/529, loss: 0.005261897575110197 2023-01-21 11:40:41.950306: step: 904/529, loss: 0.09527730941772461 2023-01-21 11:40:43.110349: step: 908/529, loss: 0.06323669105768204 2023-01-21 11:40:44.238219: step: 912/529, loss: 0.01142120361328125 2023-01-21 11:40:45.351782: step: 916/529, loss: 0.00011005402484443039 2023-01-21 11:40:46.463936: step: 920/529, loss: 0.0073678018525242805 2023-01-21 11:40:47.569753: step: 924/529, loss: 0.008548641577363014 2023-01-21 11:40:48.716876: step: 928/529, loss: 0.06399421393871307 2023-01-21 11:40:49.820981: step: 932/529, loss: 0.0050106048583984375 2023-01-21 11:40:50.952123: step: 936/529, loss: 0.03731298819184303 2023-01-21 11:40:52.036487: step: 940/529, loss: 0.006247329525649548 2023-01-21 11:40:53.147814: step: 944/529, loss: 0.0015016555553302169 2023-01-21 11:40:54.256422: step: 948/529, loss: 0.06272812187671661 2023-01-21 11:40:55.362805: step: 952/529, loss: 0.014216327108442783 2023-01-21 11:40:56.473656: step: 956/529, loss: 0.0019206047290936112 2023-01-21 11:40:57.634486: step: 960/529, loss: 0.01927342638373375 2023-01-21 11:40:58.754725: step: 964/529, loss: 0.03705215826630592 2023-01-21 11:40:59.865021: step: 968/529, loss: 0.00043144228402525187 2023-01-21 11:41:00.957346: step: 972/529, loss: 0.013107729144394398 2023-01-21 11:41:02.095706: step: 976/529, loss: 0.0005803108215332031 2023-01-21 11:41:03.197721: step: 980/529, loss: 0.0036330223083496094 2023-01-21 11:41:04.335197: step: 984/529, loss: 0.04333171993494034 2023-01-21 11:41:05.445137: step: 988/529, loss: 0.0007926941034384072 2023-01-21 11:41:06.534341: step: 992/529, loss: 0.016463281586766243 2023-01-21 11:41:07.646544: step: 996/529, loss: 0.014256191439926624 2023-01-21 11:41:08.802720: step: 1000/529, loss: 0.018715476617217064 2023-01-21 11:41:09.928821: step: 1004/529, loss: 0.005838441662490368 2023-01-21 11:41:11.065951: step: 1008/529, loss: 0.02546234242618084 2023-01-21 11:41:12.207970: step: 1012/529, loss: 0.008594894781708717 2023-01-21 11:41:13.362433: step: 1016/529, loss: 0.003566741943359375 2023-01-21 11:41:14.505605: step: 1020/529, loss: 0.013203669339418411 2023-01-21 11:41:15.610301: step: 1024/529, loss: 0.00263557443395257 2023-01-21 11:41:16.725884: step: 1028/529, loss: 0.018791770562529564 2023-01-21 11:41:17.828910: step: 1032/529, loss: 0.009150695987045765 2023-01-21 11:41:18.957647: step: 1036/529, loss: 0.11048545688390732 2023-01-21 11:41:20.060179: step: 1040/529, loss: 0.005415535066276789 2023-01-21 11:41:21.190968: step: 1044/529, loss: 0.00793304480612278 2023-01-21 11:41:22.292221: step: 1048/529, loss: 0.00057134625967592 2023-01-21 11:41:23.395893: step: 1052/529, loss: 0.006138897035270929 2023-01-21 11:41:24.480053: step: 1056/529, loss: 0.00030097959097474813 2023-01-21 11:41:25.586487: step: 1060/529, loss: 0.0046217916533350945 2023-01-21 11:41:26.714086: step: 1064/529, loss: 9.164810762740672e-05 2023-01-21 11:41:27.833827: step: 1068/529, loss: 0.03677258640527725 2023-01-21 11:41:28.952294: step: 1072/529, loss: 6.246566772460938e-05 2023-01-21 11:41:30.076981: step: 1076/529, loss: 0.010938645340502262 2023-01-21 11:41:31.196097: step: 1080/529, loss: 0.0027471540961414576 2023-01-21 11:41:32.328843: step: 1084/529, loss: 0.10885143280029297 2023-01-21 11:41:33.462011: step: 1088/529, loss: 0.051083944737911224 2023-01-21 11:41:34.575291: step: 1092/529, loss: 0.008761787787079811 2023-01-21 11:41:35.696977: step: 1096/529, loss: 0.013092994689941406 2023-01-21 11:41:36.793197: step: 1100/529, loss: 0.011665535159409046 2023-01-21 11:41:37.902067: step: 1104/529, loss: 0.007028103340417147 2023-01-21 11:41:39.047075: step: 1108/529, loss: 0.04167022928595543 2023-01-21 11:41:40.127268: step: 1112/529, loss: 0.0443972572684288 2023-01-21 11:41:41.257558: step: 1116/529, loss: 2.002716064453125e-05 2023-01-21 11:41:42.383496: step: 1120/529, loss: 0.03889331966638565 2023-01-21 11:41:43.484347: step: 1124/529, loss: 0.03410196304321289 2023-01-21 11:41:44.589283: step: 1128/529, loss: 0.006856727413833141 2023-01-21 11:41:45.702478: step: 1132/529, loss: 0.004176878836005926 2023-01-21 11:41:46.828954: step: 1136/529, loss: 0.010348701849579811 2023-01-21 11:41:47.950859: step: 1140/529, loss: 0.0006244659889489412 2023-01-21 11:41:49.071094: step: 1144/529, loss: 0.00046367646427825093 2023-01-21 11:41:50.176293: step: 1148/529, loss: 0.0033706664107739925 2023-01-21 11:41:51.307366: step: 1152/529, loss: 0.1157769188284874 2023-01-21 11:41:52.429605: step: 1156/529, loss: 0.002324152272194624 2023-01-21 11:41:53.538473: step: 1160/529, loss: 0.06894226372241974 2023-01-21 11:41:54.651566: step: 1164/529, loss: 0.00021476746769621968 2023-01-21 11:41:55.749889: step: 1168/529, loss: 0.028710652142763138 2023-01-21 11:41:56.910904: step: 1172/529, loss: 3.633499363786541e-05 2023-01-21 11:41:58.011836: step: 1176/529, loss: 0.006111431401222944 2023-01-21 11:41:59.118032: step: 1180/529, loss: 0.0037243845872581005 2023-01-21 11:42:00.267562: step: 1184/529, loss: 0.002398157026618719 2023-01-21 11:42:01.386141: step: 1188/529, loss: 0.0007041931385174394 2023-01-21 11:42:02.462231: step: 1192/529, loss: 0.007344198413193226 2023-01-21 11:42:03.607844: step: 1196/529, loss: 0.05882740020751953 2023-01-21 11:42:04.694075: step: 1200/529, loss: 0.0030315399635583162 2023-01-21 11:42:05.821129: step: 1204/529, loss: 0.0015590668190270662 2023-01-21 11:42:06.907823: step: 1208/529, loss: 0.015018177218735218 2023-01-21 11:42:08.022795: step: 1212/529, loss: 0.02497711218893528 2023-01-21 11:42:09.163463: step: 1216/529, loss: 0.001429700874723494 2023-01-21 11:42:10.284652: step: 1220/529, loss: 0.2614055573940277 2023-01-21 11:42:11.378453: step: 1224/529, loss: 0.04572753980755806 2023-01-21 11:42:12.478783: step: 1228/529, loss: 0.08436601608991623 2023-01-21 11:42:13.627100: step: 1232/529, loss: 0.00014324189396575093 2023-01-21 11:42:14.766780: step: 1236/529, loss: 0.022670842707157135 2023-01-21 11:42:15.877249: step: 1240/529, loss: 0.016945458948612213 2023-01-21 11:42:16.953963: step: 1244/529, loss: 0.014529895968735218 2023-01-21 11:42:18.074155: step: 1248/529, loss: 0.010189247317612171 2023-01-21 11:42:19.172832: step: 1252/529, loss: 0.0005743026849813759 2023-01-21 11:42:20.251449: step: 1256/529, loss: 0.0006975174183025956 2023-01-21 11:42:21.368552: step: 1260/529, loss: 0.024681951850652695 2023-01-21 11:42:22.466640: step: 1264/529, loss: 0.060361672192811966 2023-01-21 11:42:23.575526: step: 1268/529, loss: 0.024263381958007812 2023-01-21 11:42:24.680954: step: 1272/529, loss: 0.007101631257683039 2023-01-21 11:42:25.803914: step: 1276/529, loss: 0.006656646728515625 2023-01-21 11:42:26.929281: step: 1280/529, loss: 0.06424445658922195 2023-01-21 11:42:28.053215: step: 1284/529, loss: 6.828308687545359e-05 2023-01-21 11:42:29.201710: step: 1288/529, loss: 0.03032694011926651 2023-01-21 11:42:30.316153: step: 1292/529, loss: 0.0008079528925009072 2023-01-21 11:42:31.423609: step: 1296/529, loss: 0.0007113456958904862 2023-01-21 11:42:32.560205: step: 1300/529, loss: 0.0009732246398925781 2023-01-21 11:42:33.704573: step: 1304/529, loss: 0.010020256042480469 2023-01-21 11:42:34.774828: step: 1308/529, loss: 0.006526088807731867 2023-01-21 11:42:35.908149: step: 1312/529, loss: 0.008561898022890091 2023-01-21 11:42:37.040745: step: 1316/529, loss: 0.005761242005974054 2023-01-21 11:42:38.194103: step: 1320/529, loss: 0.00023384093947242945 2023-01-21 11:42:39.307174: step: 1324/529, loss: 0.00028743743314407766 2023-01-21 11:42:40.411416: step: 1328/529, loss: 0.00019149782019667327 2023-01-21 11:42:41.520123: step: 1332/529, loss: 0.030652904883027077 2023-01-21 11:42:42.643053: step: 1336/529, loss: 0.018249893561005592 2023-01-21 11:42:43.782002: step: 1340/529, loss: 0.011918735690414906 2023-01-21 11:42:44.905235: step: 1344/529, loss: 0.017681121826171875 2023-01-21 11:42:46.027412: step: 1348/529, loss: 0.0380517952144146 2023-01-21 11:42:47.128681: step: 1352/529, loss: 0.0001126289353123866 2023-01-21 11:42:48.242477: step: 1356/529, loss: 0.044347286224365234 2023-01-21 11:42:49.354307: step: 1360/529, loss: 0.03411588817834854 2023-01-21 11:42:50.504383: step: 1364/529, loss: 0.022443199530243874 2023-01-21 11:42:51.642009: step: 1368/529, loss: 0.10914954543113708 2023-01-21 11:42:52.801349: step: 1372/529, loss: 0.8595108389854431 2023-01-21 11:42:53.915812: step: 1376/529, loss: 0.11565246433019638 2023-01-21 11:42:55.042474: step: 1380/529, loss: 0.0047623636201024055 2023-01-21 11:42:56.180640: step: 1384/529, loss: 0.03679618611931801 2023-01-21 11:42:57.315604: step: 1388/529, loss: 0.0006746292347088456 2023-01-21 11:42:58.433806: step: 1392/529, loss: 0.03144397586584091 2023-01-21 11:42:59.553640: step: 1396/529, loss: 0.0007534027099609375 2023-01-21 11:43:00.664304: step: 1400/529, loss: 0.008688926696777344 2023-01-21 11:43:01.786838: step: 1404/529, loss: 0.014102840796113014 2023-01-21 11:43:02.915756: step: 1408/529, loss: 0.06872701644897461 2023-01-21 11:43:04.038716: step: 1412/529, loss: 0.008532905951142311 2023-01-21 11:43:05.168193: step: 1416/529, loss: 0.12858524918556213 2023-01-21 11:43:06.274479: step: 1420/529, loss: 0.001873624394647777 2023-01-21 11:43:07.404013: step: 1424/529, loss: 0.0013468742836266756 2023-01-21 11:43:08.525896: step: 1428/529, loss: 0.013448620215058327 2023-01-21 11:43:09.661554: step: 1432/529, loss: 0.0015245438553392887 2023-01-21 11:43:10.776184: step: 1436/529, loss: 0.0005301475175656378 2023-01-21 11:43:11.908964: step: 1440/529, loss: 0.010562896728515625 2023-01-21 11:43:13.042278: step: 1444/529, loss: 0.03284778818488121 2023-01-21 11:43:14.181111: step: 1448/529, loss: 0.029086114838719368 2023-01-21 11:43:15.301004: step: 1452/529, loss: 0.048919677734375 2023-01-21 11:43:16.417852: step: 1456/529, loss: 0.004399681463837624 2023-01-21 11:43:17.515800: step: 1460/529, loss: 0.01125488243997097 2023-01-21 11:43:18.612232: step: 1464/529, loss: 0.0551794059574604 2023-01-21 11:43:19.755878: step: 1468/529, loss: 0.015229607000946999 2023-01-21 11:43:20.868257: step: 1472/529, loss: 0.18658971786499023 2023-01-21 11:43:21.994215: step: 1476/529, loss: 0.04222278669476509 2023-01-21 11:43:23.107307: step: 1480/529, loss: 0.027546502649784088 2023-01-21 11:43:24.235219: step: 1484/529, loss: 0.0049821375869214535 2023-01-21 11:43:25.381904: step: 1488/529, loss: 0.0003444671747274697 2023-01-21 11:43:26.527585: step: 1492/529, loss: 0.0006232261657714844 2023-01-21 11:43:27.650074: step: 1496/529, loss: 0.013267231173813343 2023-01-21 11:43:28.770559: step: 1500/529, loss: 0.00242443080060184 2023-01-21 11:43:29.880800: step: 1504/529, loss: 0.00020465851412154734 2023-01-21 11:43:30.994730: step: 1508/529, loss: 0.0023739340249449015 2023-01-21 11:43:32.123929: step: 1512/529, loss: 0.0065437317825853825 2023-01-21 11:43:33.228523: step: 1516/529, loss: 0.011449528858065605 2023-01-21 11:43:34.319162: step: 1520/529, loss: 0.0242049228399992 2023-01-21 11:43:35.428199: step: 1524/529, loss: 0.018066883087158203 2023-01-21 11:43:36.557734: step: 1528/529, loss: 0.02419414557516575 2023-01-21 11:43:37.665886: step: 1532/529, loss: 0.014324760064482689 2023-01-21 11:43:38.789389: step: 1536/529, loss: 0.011695289984345436 2023-01-21 11:43:39.906687: step: 1540/529, loss: 0.006884956732392311 2023-01-21 11:43:41.033338: step: 1544/529, loss: 0.09910842031240463 2023-01-21 11:43:42.180531: step: 1548/529, loss: 0.0006841659778729081 2023-01-21 11:43:43.407951: step: 1552/529, loss: 0.03175327181816101 2023-01-21 11:43:44.551388: step: 1556/529, loss: 0.037883758544921875 2023-01-21 11:43:45.697951: step: 1560/529, loss: 0.13884659111499786 2023-01-21 11:43:46.811735: step: 1564/529, loss: 0.001976394560188055 2023-01-21 11:43:47.933385: step: 1568/529, loss: 0.00010504722740733996 2023-01-21 11:43:49.047111: step: 1572/529, loss: 0.019742773845791817 2023-01-21 11:43:50.184245: step: 1576/529, loss: 0.008479118347167969 2023-01-21 11:43:51.286154: step: 1580/529, loss: 0.15628795325756073 2023-01-21 11:43:52.457966: step: 1584/529, loss: 0.11336851119995117 2023-01-21 11:43:53.574431: step: 1588/529, loss: 0.0001903533993754536 2023-01-21 11:43:54.684874: step: 1592/529, loss: 0.03635072708129883 2023-01-21 11:43:55.797828: step: 1596/529, loss: 0.006968498229980469 2023-01-21 11:43:56.904782: step: 1600/529, loss: 0.03251037746667862 2023-01-21 11:43:58.021077: step: 1604/529, loss: 0.22056026756763458 2023-01-21 11:43:59.149291: step: 1608/529, loss: 0.013068771921098232 2023-01-21 11:44:00.275246: step: 1612/529, loss: 0.07117576897144318 2023-01-21 11:44:01.380211: step: 1616/529, loss: 0.016994666308164597 2023-01-21 11:44:02.484624: step: 1620/529, loss: 0.07627410441637039 2023-01-21 11:44:03.597812: step: 1624/529, loss: 0.0030422210693359375 2023-01-21 11:44:04.728596: step: 1628/529, loss: 0.020182132720947266 2023-01-21 11:44:05.854205: step: 1632/529, loss: 0.015240669250488281 2023-01-21 11:44:06.959613: step: 1636/529, loss: 0.004169655032455921 2023-01-21 11:44:08.091175: step: 1640/529, loss: 0.004290199372917414 2023-01-21 11:44:09.216360: step: 1644/529, loss: 0.0020856857299804688 2023-01-21 11:44:10.334467: step: 1648/529, loss: 0.0017930985195562243 2023-01-21 11:44:11.481047: step: 1652/529, loss: 0.05909614637494087 2023-01-21 11:44:12.599483: step: 1656/529, loss: 4.997253563487902e-05 2023-01-21 11:44:13.750817: step: 1660/529, loss: 0.002527427626773715 2023-01-21 11:44:14.866577: step: 1664/529, loss: 0.004800987429916859 2023-01-21 11:44:16.010660: step: 1668/529, loss: 0.01271915528923273 2023-01-21 11:44:17.142547: step: 1672/529, loss: 0.0016716003883630037 2023-01-21 11:44:18.273942: step: 1676/529, loss: 0.003542328020557761 2023-01-21 11:44:19.403192: step: 1680/529, loss: 0.006430245004594326 2023-01-21 11:44:20.534867: step: 1684/529, loss: 0.5775493383407593 2023-01-21 11:44:21.669802: step: 1688/529, loss: 0.009461593814194202 2023-01-21 11:44:22.789899: step: 1692/529, loss: 0.1243845522403717 2023-01-21 11:44:23.892124: step: 1696/529, loss: 0.020012855529785156 2023-01-21 11:44:24.979851: step: 1700/529, loss: 0.0006521225441247225 2023-01-21 11:44:26.111940: step: 1704/529, loss: 0.0007424354553222656 2023-01-21 11:44:27.222926: step: 1708/529, loss: 0.00028476715669967234 2023-01-21 11:44:28.322523: step: 1712/529, loss: 0.0010662078857421875 2023-01-21 11:44:29.445977: step: 1716/529, loss: 0.13426867127418518 2023-01-21 11:44:30.558975: step: 1720/529, loss: 1.6117095583467744e-05 2023-01-21 11:44:31.685598: step: 1724/529, loss: 0.016417598351836205 2023-01-21 11:44:32.806341: step: 1728/529, loss: 0.012674522586166859 2023-01-21 11:44:33.920106: step: 1732/529, loss: 0.1762961447238922 2023-01-21 11:44:35.006794: step: 1736/529, loss: 0.056645773351192474 2023-01-21 11:44:36.140751: step: 1740/529, loss: 0.01091537531465292 2023-01-21 11:44:37.249343: step: 1744/529, loss: 0.0075928689911961555 2023-01-21 11:44:38.359208: step: 1748/529, loss: 0.08950477838516235 2023-01-21 11:44:39.470861: step: 1752/529, loss: 0.04054737091064453 2023-01-21 11:44:40.556325: step: 1756/529, loss: 0.005307960789650679 2023-01-21 11:44:41.687942: step: 1760/529, loss: 0.0028938292525708675 2023-01-21 11:44:42.778126: step: 1764/529, loss: 0.024899005889892578 2023-01-21 11:44:43.884633: step: 1768/529, loss: 0.02730732038617134 2023-01-21 11:44:44.972831: step: 1772/529, loss: 0.0139128677546978 2023-01-21 11:44:46.089511: step: 1776/529, loss: 0.021582795307040215 2023-01-21 11:44:47.236482: step: 1780/529, loss: 0.020476914942264557 2023-01-21 11:44:48.352042: step: 1784/529, loss: 0.00108509068377316 2023-01-21 11:44:49.463080: step: 1788/529, loss: 0.05165863037109375 2023-01-21 11:44:50.591487: step: 1792/529, loss: 0.03100724332034588 2023-01-21 11:44:51.679701: step: 1796/529, loss: 0.002113723661750555 2023-01-21 11:44:52.798710: step: 1800/529, loss: 0.017702486366033554 2023-01-21 11:44:53.953228: step: 1804/529, loss: 0.04667053371667862 2023-01-21 11:44:55.060025: step: 1808/529, loss: 0.009372711181640625 2023-01-21 11:44:56.199204: step: 1812/529, loss: 0.002443504286929965 2023-01-21 11:44:57.387231: step: 1816/529, loss: 0.001627588295377791 2023-01-21 11:44:58.545000: step: 1820/529, loss: 0.010867882519960403 2023-01-21 11:44:59.656892: step: 1824/529, loss: 0.0048233033157885075 2023-01-21 11:45:00.807577: step: 1828/529, loss: 0.03398914262652397 2023-01-21 11:45:01.944346: step: 1832/529, loss: 0.0035132409539073706 2023-01-21 11:45:03.074799: step: 1836/529, loss: 2.593994213384576e-05 2023-01-21 11:45:04.175755: step: 1840/529, loss: 0.012753963470458984 2023-01-21 11:45:05.292706: step: 1844/529, loss: 0.0012276172637939453 2023-01-21 11:45:06.416805: step: 1848/529, loss: 0.0008684158092364669 2023-01-21 11:45:07.542384: step: 1852/529, loss: 0.0020522119011729956 2023-01-21 11:45:08.674506: step: 1856/529, loss: 0.0026235580444335938 2023-01-21 11:45:09.782220: step: 1860/529, loss: 0.0723545104265213 2023-01-21 11:45:10.923421: step: 1864/529, loss: 0.000373649614630267 2023-01-21 11:45:12.007973: step: 1868/529, loss: 0.0012364387512207031 2023-01-21 11:45:13.129430: step: 1872/529, loss: 0.0015027045737951994 2023-01-21 11:45:14.257819: step: 1876/529, loss: 0.07378172874450684 2023-01-21 11:45:15.344832: step: 1880/529, loss: 0.0022869110107421875 2023-01-21 11:45:16.477562: step: 1884/529, loss: 0.008147239685058594 2023-01-21 11:45:17.589805: step: 1888/529, loss: 0.009593391790986061 2023-01-21 11:45:18.702735: step: 1892/529, loss: 0.014007187448441982 2023-01-21 11:45:19.820955: step: 1896/529, loss: 0.01309671439230442 2023-01-21 11:45:20.949971: step: 1900/529, loss: 0.05403494834899902 2023-01-21 11:45:22.040289: step: 1904/529, loss: 0.015779972076416016 2023-01-21 11:45:23.142764: step: 1908/529, loss: 0.004146766848862171 2023-01-21 11:45:24.224514: step: 1912/529, loss: 2.0599363779183477e-05 2023-01-21 11:45:25.326852: step: 1916/529, loss: 0.010004043579101562 2023-01-21 11:45:26.428654: step: 1920/529, loss: 0.00024127960205078125 2023-01-21 11:45:27.533577: step: 1924/529, loss: 0.008876895532011986 2023-01-21 11:45:28.635705: step: 1928/529, loss: 0.042061615735292435 2023-01-21 11:45:29.756711: step: 1932/529, loss: 0.0005893707275390625 2023-01-21 11:45:30.890827: step: 1936/529, loss: 0.007040500640869141 2023-01-21 11:45:31.992048: step: 1940/529, loss: 0.0038908959832042456 2023-01-21 11:45:33.096118: step: 1944/529, loss: 0.002587700029835105 2023-01-21 11:45:34.183592: step: 1948/529, loss: 0.04027128219604492 2023-01-21 11:45:35.291582: step: 1952/529, loss: 0.6564247608184814 2023-01-21 11:45:36.387197: step: 1956/529, loss: 0.0012781142722815275 2023-01-21 11:45:37.483532: step: 1960/529, loss: 0.0016482830978929996 2023-01-21 11:45:38.585359: step: 1964/529, loss: 0.0017163276206701994 2023-01-21 11:45:39.703629: step: 1968/529, loss: 0.025080585852265358 2023-01-21 11:45:40.822451: step: 1972/529, loss: 0.0020043374970555305 2023-01-21 11:45:41.965752: step: 1976/529, loss: 0.002396488096565008 2023-01-21 11:45:43.093699: step: 1980/529, loss: 0.030241012573242188 2023-01-21 11:45:44.204871: step: 1984/529, loss: 0.0022886276710778475 2023-01-21 11:45:45.330698: step: 1988/529, loss: 0.02503395266830921 2023-01-21 11:45:46.461406: step: 1992/529, loss: 0.010606956668198109 2023-01-21 11:45:47.567990: step: 1996/529, loss: 0.006884432397782803 2023-01-21 11:45:48.708906: step: 2000/529, loss: 0.001625823904760182 2023-01-21 11:45:49.812849: step: 2004/529, loss: 0.053549814969301224 2023-01-21 11:45:50.956674: step: 2008/529, loss: 0.5644946098327637 2023-01-21 11:45:52.104528: step: 2012/529, loss: 0.05977001041173935 2023-01-21 11:45:53.223556: step: 2016/529, loss: 0.014062881469726562 2023-01-21 11:45:54.311107: step: 2020/529, loss: 0.16801948845386505 2023-01-21 11:45:55.414047: step: 2024/529, loss: 0.0013593673938885331 2023-01-21 11:45:56.519037: step: 2028/529, loss: 0.0016456603771075606 2023-01-21 11:45:57.637343: step: 2032/529, loss: 0.0001735687255859375 2023-01-21 11:45:58.752659: step: 2036/529, loss: 0.28335294127464294 2023-01-21 11:45:59.847506: step: 2040/529, loss: 0.0058845519088208675 2023-01-21 11:46:00.976930: step: 2044/529, loss: 0.04492301866412163 2023-01-21 11:46:02.092499: step: 2048/529, loss: 0.030686378479003906 2023-01-21 11:46:03.216882: step: 2052/529, loss: 0.034900762140750885 2023-01-21 11:46:04.298889: step: 2056/529, loss: 0.013790893368422985 2023-01-21 11:46:05.433692: step: 2060/529, loss: 0.0001466751127736643 2023-01-21 11:46:06.547123: step: 2064/529, loss: 0.0045375823974609375 2023-01-21 11:46:07.652590: step: 2068/529, loss: 0.06097235903143883 2023-01-21 11:46:08.813225: step: 2072/529, loss: 0.044893838465213776 2023-01-21 11:46:09.923017: step: 2076/529, loss: 0.005265617277473211 2023-01-21 11:46:11.030435: step: 2080/529, loss: 0.046027567237615585 2023-01-21 11:46:12.132189: step: 2084/529, loss: 0.002574348356574774 2023-01-21 11:46:13.221318: step: 2088/529, loss: 0.028423458337783813 2023-01-21 11:46:14.336464: step: 2092/529, loss: 0.012430763803422451 2023-01-21 11:46:15.450198: step: 2096/529, loss: 0.015805387869477272 2023-01-21 11:46:16.567629: step: 2100/529, loss: 0.0267868060618639 2023-01-21 11:46:17.672811: step: 2104/529, loss: 0.04761314392089844 2023-01-21 11:46:18.783562: step: 2108/529, loss: 0.0019538879860192537 2023-01-21 11:46:19.878752: step: 2112/529, loss: 0.0946991965174675 2023-01-21 11:46:20.997330: step: 2116/529, loss: 0.009499598294496536 ================================================== Loss: 0.032 -------------------- Dev: {'event': {'p': 0.5637583892617449, 'r': 0.7829560585885486, 'f1': 0.6555183946488293}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Test: {'event': {'p': 0.6213251426064064, 'r': 0.7990970654627539, 'f1': 0.6990866452727721}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Chinese: {'event': {'p': 0.5416666666666666, 'r': 0.9629629629629629, 'f1': 0.6933333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Korean: {'event': {'p': 0.5806451612903226, 'r': 0.5714285714285714, 'f1': 0.576}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} 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': 18} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 19 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:47:01.837787: step: 4/529, loss: 0.014355468563735485 2023-01-21 11:47:02.945732: step: 8/529, loss: 0.0006323337438516319 2023-01-21 11:47:04.042406: step: 12/529, loss: 0.0033627511002123356 2023-01-21 11:47:05.148397: step: 16/529, loss: 0.000671434449031949 2023-01-21 11:47:06.243861: step: 20/529, loss: 0.017496107146143913 2023-01-21 11:47:07.367240: step: 24/529, loss: 0.010253524407744408 2023-01-21 11:47:08.475915: step: 28/529, loss: 0.0040130615234375 2023-01-21 11:47:09.569537: step: 32/529, loss: 0.04756765440106392 2023-01-21 11:47:10.703037: step: 36/529, loss: 0.0031833648681640625 2023-01-21 11:47:11.810780: step: 40/529, loss: 0.0004937172052450478 2023-01-21 11:47:12.927296: step: 44/529, loss: 0.024747466668486595 2023-01-21 11:47:14.055662: step: 48/529, loss: 0.0034503936767578125 2023-01-21 11:47:15.153338: step: 52/529, loss: 0.21310386061668396 2023-01-21 11:47:16.278502: step: 56/529, loss: 0.016933441162109375 2023-01-21 11:47:17.384020: step: 60/529, loss: 0.28695639967918396 2023-01-21 11:47:18.477553: step: 64/529, loss: 0.0012784004211425781 2023-01-21 11:47:19.596082: step: 68/529, loss: 0.02222595177590847 2023-01-21 11:47:20.738516: step: 72/529, loss: 6.380081322276965e-05 2023-01-21 11:47:21.808430: step: 76/529, loss: 0.01010971050709486 2023-01-21 11:47:22.953183: step: 80/529, loss: 0.0035266876220703125 2023-01-21 11:47:24.073676: step: 84/529, loss: 0.004527282901108265 2023-01-21 11:47:25.163503: step: 88/529, loss: 0.013827419839799404 2023-01-21 11:47:26.286851: step: 92/529, loss: 0.02768847905099392 2023-01-21 11:47:27.395264: step: 96/529, loss: 0.04118843004107475 2023-01-21 11:47:28.512101: step: 100/529, loss: 0.0007052421569824219 2023-01-21 11:47:29.625880: step: 104/529, loss: 0.03379478678107262 2023-01-21 11:47:30.732626: step: 108/529, loss: 0.0008284568903036416 2023-01-21 11:47:31.833640: step: 112/529, loss: 4.711151268566027e-05 2023-01-21 11:47:32.946616: step: 116/529, loss: 0.002983665792271495 2023-01-21 11:47:34.041896: step: 120/529, loss: 0.012008953839540482 2023-01-21 11:47:35.155807: step: 124/529, loss: 0.0012169122928753495 2023-01-21 11:47:36.259409: step: 128/529, loss: 0.00524826068431139 2023-01-21 11:47:37.348060: step: 132/529, loss: 0.04662275314331055 2023-01-21 11:47:38.457014: step: 136/529, loss: 0.004094064235687256 2023-01-21 11:47:39.570834: step: 140/529, loss: 0.0022203444968909025 2023-01-21 11:47:40.710288: step: 144/529, loss: 0.004037189297378063 2023-01-21 11:47:41.815684: step: 148/529, loss: 0.004108428955078125 2023-01-21 11:47:42.941412: step: 152/529, loss: 0.009538555517792702 2023-01-21 11:47:44.077027: step: 156/529, loss: 0.0075501445680856705 2023-01-21 11:47:45.146940: step: 160/529, loss: 0.0029109478928148746 2023-01-21 11:47:46.242517: step: 164/529, loss: 0.051416970789432526 2023-01-21 11:47:47.385086: step: 168/529, loss: 0.000705718994140625 2023-01-21 11:47:48.510345: step: 172/529, loss: 0.00024313927860930562 2023-01-21 11:47:49.651301: step: 176/529, loss: 0.002595901722088456 2023-01-21 11:47:50.767162: step: 180/529, loss: 0.002785822842270136 2023-01-21 11:47:51.866730: step: 184/529, loss: 0.0002342224179301411 2023-01-21 11:47:52.974342: step: 188/529, loss: 0.021434593945741653 2023-01-21 11:47:54.069918: step: 192/529, loss: 0.0009616852039471269 2023-01-21 11:47:55.214684: step: 196/529, loss: 0.04459095001220703 2023-01-21 11:47:56.324617: step: 200/529, loss: 0.014522934332489967 2023-01-21 11:47:57.469670: step: 204/529, loss: 0.03135986253619194 2023-01-21 11:47:58.574846: step: 208/529, loss: 0.03364114463329315 2023-01-21 11:47:59.705320: step: 212/529, loss: 0.00024359226517844945 2023-01-21 11:48:00.823684: step: 216/529, loss: 0.03321495279669762 2023-01-21 11:48:01.936638: step: 220/529, loss: 0.001216888427734375 2023-01-21 11:48:03.040500: step: 224/529, loss: 0.004745102021843195 2023-01-21 11:48:04.169875: step: 228/529, loss: 0.019183922559022903 2023-01-21 11:48:05.279638: step: 232/529, loss: 0.006996536161750555 2023-01-21 11:48:06.426438: step: 236/529, loss: 0.0012030602665618062 2023-01-21 11:48:07.526330: step: 240/529, loss: 0.007956123910844326 2023-01-21 11:48:08.635218: step: 244/529, loss: 0.00935440044850111 2023-01-21 11:48:09.781298: step: 248/529, loss: 0.05777540057897568 2023-01-21 11:48:10.889961: step: 252/529, loss: 0.0004600524844136089 2023-01-21 11:48:11.999786: step: 256/529, loss: 0.012191391550004482 2023-01-21 11:48:13.136381: step: 260/529, loss: 0.0032075883354991674 2023-01-21 11:48:14.246264: step: 264/529, loss: 0.0007696151733398438 2023-01-21 11:48:15.355570: step: 268/529, loss: 0.24476762115955353 2023-01-21 11:48:16.477514: step: 272/529, loss: 0.01579446718096733 2023-01-21 11:48:17.590961: step: 276/529, loss: 0.00043735504732467234 2023-01-21 11:48:18.707833: step: 280/529, loss: 0.000748109829146415 2023-01-21 11:48:19.845304: step: 284/529, loss: 0.0019321441650390625 2023-01-21 11:48:20.978481: step: 288/529, loss: 0.0882728099822998 2023-01-21 11:48:22.071556: step: 292/529, loss: 0.010863685980439186 2023-01-21 11:48:23.194839: step: 296/529, loss: 0.0010265350574627519 2023-01-21 11:48:24.336736: step: 300/529, loss: 0.0011432648170739412 2023-01-21 11:48:25.428255: step: 304/529, loss: 0.0004352569521870464 2023-01-21 11:48:26.525726: step: 308/529, loss: 0.001831150148063898 2023-01-21 11:48:27.642537: step: 312/529, loss: 0.6326780319213867 2023-01-21 11:48:28.739027: step: 316/529, loss: 0.0367220900952816 2023-01-21 11:48:29.841199: step: 320/529, loss: 0.018698502331972122 2023-01-21 11:48:30.923200: step: 324/529, loss: 0.009912515059113503 2023-01-21 11:48:32.020871: step: 328/529, loss: 0.046420954167842865 2023-01-21 11:48:33.088166: step: 332/529, loss: 0.05252427980303764 2023-01-21 11:48:34.222006: step: 336/529, loss: 0.008003043942153454 2023-01-21 11:48:35.333672: step: 340/529, loss: 0.0013790130615234375 2023-01-21 11:48:36.442148: step: 344/529, loss: 0.0034000396262854338 2023-01-21 11:48:37.561724: step: 348/529, loss: 9.694099571788684e-05 2023-01-21 11:48:38.685641: step: 352/529, loss: 0.007418155670166016 2023-01-21 11:48:39.808432: step: 356/529, loss: 0.0044692992232739925 2023-01-21 11:48:40.910157: step: 360/529, loss: 0.0019312858348712325 2023-01-21 11:48:42.047584: step: 364/529, loss: 0.0312647819519043 2023-01-21 11:48:43.161076: step: 368/529, loss: 0.0022410391829907894 2023-01-21 11:48:44.259543: step: 372/529, loss: 0.0030117034912109375 2023-01-21 11:48:45.409002: step: 376/529, loss: 0.000590419746004045 2023-01-21 11:48:46.512461: step: 380/529, loss: 0.039423368871212006 2023-01-21 11:48:47.618350: step: 384/529, loss: 0.00378589634783566 2023-01-21 11:48:48.741726: step: 388/529, loss: 0.02882976643741131 2023-01-21 11:48:49.863544: step: 392/529, loss: 0.010480308905243874 2023-01-21 11:48:50.953113: step: 396/529, loss: 0.009607887826859951 2023-01-21 11:48:52.005824: step: 400/529, loss: 0.012318229302763939 2023-01-21 11:48:53.122012: step: 404/529, loss: 0.0019186973804607987 2023-01-21 11:48:54.240203: step: 408/529, loss: 0.013365937396883965 2023-01-21 11:48:55.370822: step: 412/529, loss: 0.04006042331457138 2023-01-21 11:48:56.480345: step: 416/529, loss: 0.005264186765998602 2023-01-21 11:48:57.566381: step: 420/529, loss: 0.0007431865087710321 2023-01-21 11:48:58.670727: step: 424/529, loss: 0.0017409325810149312 2023-01-21 11:48:59.764859: step: 428/529, loss: 0.020402144640684128 2023-01-21 11:49:00.900806: step: 432/529, loss: 0.0006545067299157381 2023-01-21 11:49:01.990407: step: 436/529, loss: 0.00023484230041503906 2023-01-21 11:49:03.081558: step: 440/529, loss: 0.0034800528082996607 2023-01-21 11:49:04.170313: step: 444/529, loss: 0.01729297637939453 2023-01-21 11:49:05.292304: step: 448/529, loss: 0.25457268953323364 2023-01-21 11:49:06.415814: step: 452/529, loss: 0.01759650744497776 2023-01-21 11:49:07.578966: step: 456/529, loss: 0.0005899429088458419 2023-01-21 11:49:08.726066: step: 460/529, loss: 0.03666868060827255 2023-01-21 11:49:09.841441: step: 464/529, loss: 0.06310119479894638 2023-01-21 11:49:10.964824: step: 468/529, loss: 0.09546433389186859 2023-01-21 11:49:12.060918: step: 472/529, loss: 0.038512710481882095 2023-01-21 11:49:13.167000: step: 476/529, loss: 0.022290658205747604 2023-01-21 11:49:14.255292: step: 480/529, loss: 0.03739328309893608 2023-01-21 11:49:15.348741: step: 484/529, loss: 2.193451337006991e-06 2023-01-21 11:49:16.443655: step: 488/529, loss: 0.028816889971494675 2023-01-21 11:49:17.538674: step: 492/529, loss: 0.001351213431917131 2023-01-21 11:49:18.620644: step: 496/529, loss: 0.005806589499115944 2023-01-21 11:49:19.723770: step: 500/529, loss: 0.00861969031393528 2023-01-21 11:49:20.826537: step: 504/529, loss: 0.028588009998202324 2023-01-21 11:49:21.939763: step: 508/529, loss: 0.0270233154296875 2023-01-21 11:49:23.061951: step: 512/529, loss: 0.0028108598198741674 2023-01-21 11:49:24.259118: step: 516/529, loss: 0.00611114501953125 2023-01-21 11:49:25.402588: step: 520/529, loss: 0.02396526373922825 2023-01-21 11:49:26.533919: step: 524/529, loss: 0.015333366580307484 2023-01-21 11:49:27.617143: step: 528/529, loss: 0.0013791085220873356 2023-01-21 11:49:28.715265: step: 532/529, loss: 0.0027270319405943155 2023-01-21 11:49:29.824266: step: 536/529, loss: 0.020116616040468216 2023-01-21 11:49:30.915013: step: 540/529, loss: 0.0017522812122479081 2023-01-21 11:49:32.059044: step: 544/529, loss: 0.11413393169641495 2023-01-21 11:49:33.177584: step: 548/529, loss: 0.0027193070854991674 2023-01-21 11:49:34.276284: step: 552/529, loss: 0.002698803087696433 2023-01-21 11:49:35.379086: step: 556/529, loss: 0.05554747208952904 2023-01-21 11:49:36.489008: step: 560/529, loss: 0.0005871772882528603 2023-01-21 11:49:37.658550: step: 564/529, loss: 0.039246559143066406 2023-01-21 11:49:38.758458: step: 568/529, loss: 1.4781951904296875e-05 2023-01-21 11:49:39.848253: step: 572/529, loss: 6.0749058320652694e-05 2023-01-21 11:49:40.967664: step: 576/529, loss: 0.00945911556482315 2023-01-21 11:49:42.042662: step: 580/529, loss: 1.8119812921213452e-06 2023-01-21 11:49:43.146889: step: 584/529, loss: 0.0006213188171386719 2023-01-21 11:49:44.262899: step: 588/529, loss: 0.04135427623987198 2023-01-21 11:49:45.377904: step: 592/529, loss: 0.0008311271667480469 2023-01-21 11:49:46.484340: step: 596/529, loss: 0.0040066721849143505 2023-01-21 11:49:47.581468: step: 600/529, loss: 0.0004817962762899697 2023-01-21 11:49:48.724022: step: 604/529, loss: 0.007086372934281826 2023-01-21 11:49:49.803534: step: 608/529, loss: 0.0003685951232910156 2023-01-21 11:49:50.934231: step: 612/529, loss: 0.2023063600063324 2023-01-21 11:49:52.066719: step: 616/529, loss: 0.0038887024857103825 2023-01-21 11:49:53.205241: step: 620/529, loss: 0.014190674759447575 2023-01-21 11:49:54.306349: step: 624/529, loss: 0.0025453567504882812 2023-01-21 11:49:55.432962: step: 628/529, loss: 0.00730133056640625 2023-01-21 11:49:56.537526: step: 632/529, loss: 0.004276657477021217 2023-01-21 11:49:57.637648: step: 636/529, loss: 0.04006042331457138 2023-01-21 11:49:58.750269: step: 640/529, loss: 2.021789623540826e-05 2023-01-21 11:49:59.849704: step: 644/529, loss: 0.08157215267419815 2023-01-21 11:50:00.954476: step: 648/529, loss: 0.0002880096435546875 2023-01-21 11:50:02.105718: step: 652/529, loss: 0.014393031597137451 2023-01-21 11:50:03.240155: step: 656/529, loss: 0.007098769769072533 2023-01-21 11:50:04.343122: step: 660/529, loss: 0.0006195545429363847 2023-01-21 11:50:05.453448: step: 664/529, loss: 0.012794685550034046 2023-01-21 11:50:06.572602: step: 668/529, loss: 5.569458153331652e-05 2023-01-21 11:50:07.712773: step: 672/529, loss: 0.06984329223632812 2023-01-21 11:50:08.811735: step: 676/529, loss: 0.002380180172622204 2023-01-21 11:50:09.929481: step: 680/529, loss: 0.0007987499702721834 2023-01-21 11:50:11.063740: step: 684/529, loss: 0.16498775780200958 2023-01-21 11:50:12.194661: step: 688/529, loss: 0.03566398844122887 2023-01-21 11:50:13.290158: step: 692/529, loss: 0.0005817413912154734 2023-01-21 11:50:14.411527: step: 696/529, loss: 0.0017530441982671618 2023-01-21 11:50:15.540499: step: 700/529, loss: 0.010729407891631126 2023-01-21 11:50:16.669299: step: 704/529, loss: 0.016094589605927467 2023-01-21 11:50:17.780106: step: 708/529, loss: 0.4038047790527344 2023-01-21 11:50:18.911618: step: 712/529, loss: 0.001567077706567943 2023-01-21 11:50:20.041010: step: 716/529, loss: 0.005270099733024836 2023-01-21 11:50:21.189230: step: 720/529, loss: 0.0032434463500976562 2023-01-21 11:50:22.270934: step: 724/529, loss: 0.022444821894168854 2023-01-21 11:50:23.396778: step: 728/529, loss: 0.0007795333513058722 2023-01-21 11:50:24.531000: step: 732/529, loss: 0.00010957718041026965 2023-01-21 11:50:25.638272: step: 736/529, loss: 0.006377029232680798 2023-01-21 11:50:26.781431: step: 740/529, loss: 0.05097942426800728 2023-01-21 11:50:27.878281: step: 744/529, loss: 0.044149018824100494 2023-01-21 11:50:29.021452: step: 748/529, loss: 0.042908765375614166 2023-01-21 11:50:30.110147: step: 752/529, loss: 0.005693626590073109 2023-01-21 11:50:31.210894: step: 756/529, loss: 0.005429267883300781 2023-01-21 11:50:32.308687: step: 760/529, loss: 0.01851024106144905 2023-01-21 11:50:33.413904: step: 764/529, loss: 0.0230986587703228 2023-01-21 11:50:34.516072: step: 768/529, loss: 0.016632651910185814 2023-01-21 11:50:35.630944: step: 772/529, loss: 0.00023937225341796875 2023-01-21 11:50:36.749277: step: 776/529, loss: 0.02695140801370144 2023-01-21 11:50:37.870496: step: 780/529, loss: 0.008199882693588734 2023-01-21 11:50:39.000970: step: 784/529, loss: 0.013084793463349342 2023-01-21 11:50:40.131948: step: 788/529, loss: 0.019443130120635033 2023-01-21 11:50:41.271102: step: 792/529, loss: 0.0005764007801190019 2023-01-21 11:50:42.383113: step: 796/529, loss: 0.025236893445253372 2023-01-21 11:50:43.478289: step: 800/529, loss: 0.004144476726651192 2023-01-21 11:50:44.581093: step: 804/529, loss: 0.03200054168701172 2023-01-21 11:50:45.683673: step: 808/529, loss: 0.00026149750920012593 2023-01-21 11:50:46.791016: step: 812/529, loss: 0.013726524077355862 2023-01-21 11:50:47.907580: step: 816/529, loss: 0.014700889587402344 2023-01-21 11:50:49.011298: step: 820/529, loss: 0.05683021992444992 2023-01-21 11:50:50.103631: step: 824/529, loss: 0.03358612209558487 2023-01-21 11:50:51.239633: step: 828/529, loss: 0.016904450953006744 2023-01-21 11:50:52.363404: step: 832/529, loss: 0.02762165106832981 2023-01-21 11:50:53.485698: step: 836/529, loss: 0.0017857551574707031 2023-01-21 11:50:54.596297: step: 840/529, loss: 0.010833931155502796 2023-01-21 11:50:55.697262: step: 844/529, loss: 0.0014444352127611637 2023-01-21 11:50:56.803763: step: 848/529, loss: 0.027597809210419655 2023-01-21 11:50:57.954883: step: 852/529, loss: 0.05364866554737091 2023-01-21 11:50:59.076823: step: 856/529, loss: 0.002945518586784601 2023-01-21 11:51:00.194803: step: 860/529, loss: 0.008834457956254482 2023-01-21 11:51:01.333491: step: 864/529, loss: 0.00010619164095260203 2023-01-21 11:51:02.426709: step: 868/529, loss: 0.008788681589066982 2023-01-21 11:51:03.554388: step: 872/529, loss: 0.014927864074707031 2023-01-21 11:51:04.646700: step: 876/529, loss: 0.003241920378059149 2023-01-21 11:51:05.741301: step: 880/529, loss: 0.0011375427711755037 2023-01-21 11:51:06.858995: step: 884/529, loss: 0.06047344207763672 2023-01-21 11:51:07.985009: step: 888/529, loss: 0.02774982526898384 2023-01-21 11:51:09.125426: step: 892/529, loss: 0.00427322369068861 2023-01-21 11:51:10.220528: step: 896/529, loss: 0.004375743679702282 2023-01-21 11:51:11.302001: step: 900/529, loss: 0.0007824897766113281 2023-01-21 11:51:12.421536: step: 904/529, loss: 0.00011625289334915578 2023-01-21 11:51:13.513331: step: 908/529, loss: 0.031965069472789764 2023-01-21 11:51:14.657779: step: 912/529, loss: 0.019513322040438652 2023-01-21 11:51:15.778367: step: 916/529, loss: 0.012214088812470436 2023-01-21 11:51:16.911444: step: 920/529, loss: 0.04048614948987961 2023-01-21 11:51:18.017670: step: 924/529, loss: 0.00038022996159270406 2023-01-21 11:51:19.145944: step: 928/529, loss: 0.003941345028579235 2023-01-21 11:51:20.266634: step: 932/529, loss: 7.476807513739914e-05 2023-01-21 11:51:21.393117: step: 936/529, loss: 0.0008944034343585372 2023-01-21 11:51:22.565485: step: 940/529, loss: 0.0842016190290451 2023-01-21 11:51:23.681170: step: 944/529, loss: 0.02452554740011692 2023-01-21 11:51:24.793364: step: 948/529, loss: 0.035330578684806824 2023-01-21 11:51:25.903259: step: 952/529, loss: 0.003975686617195606 2023-01-21 11:51:27.024225: step: 956/529, loss: 0.00010614395432639867 2023-01-21 11:51:28.124032: step: 960/529, loss: 0.01897735707461834 2023-01-21 11:51:29.287425: step: 964/529, loss: 0.005387497134506702 2023-01-21 11:51:30.404480: step: 968/529, loss: 0.040128327906131744 2023-01-21 11:51:31.514577: step: 972/529, loss: 0.07488436996936798 2023-01-21 11:51:32.661505: step: 976/529, loss: 0.08108091354370117 2023-01-21 11:51:33.784705: step: 980/529, loss: 0.03888225555419922 2023-01-21 11:51:34.886053: step: 984/529, loss: 0.024538803845643997 2023-01-21 11:51:35.989522: step: 988/529, loss: 0.004850483033806086 2023-01-21 11:51:37.082144: step: 992/529, loss: 0.0006504058837890625 2023-01-21 11:51:38.207216: step: 996/529, loss: 0.0026613236404955387 2023-01-21 11:51:39.330321: step: 1000/529, loss: 9.908677020575851e-05 2023-01-21 11:51:40.434216: step: 1004/529, loss: 0.006530189886689186 2023-01-21 11:51:41.547289: step: 1008/529, loss: 0.001556015107780695 2023-01-21 11:51:42.666203: step: 1012/529, loss: 0.006156063172966242 2023-01-21 11:51:43.770020: step: 1016/529, loss: 0.013196755200624466 2023-01-21 11:51:44.890683: step: 1020/529, loss: 0.00469474820420146 2023-01-21 11:51:45.953552: step: 1024/529, loss: 0.009793663397431374 2023-01-21 11:51:47.053185: step: 1028/529, loss: 0.03442397341132164 2023-01-21 11:51:48.180232: step: 1032/529, loss: 0.004127549938857555 2023-01-21 11:51:49.270616: step: 1036/529, loss: 0.017177294939756393 2023-01-21 11:51:50.417518: step: 1040/529, loss: 0.0050782207399606705 2023-01-21 11:51:51.521343: step: 1044/529, loss: 0.0014782905345782638 2023-01-21 11:51:52.658409: step: 1048/529, loss: 0.003768539521843195 2023-01-21 11:51:53.775699: step: 1052/529, loss: 0.6389240026473999 2023-01-21 11:51:54.874938: step: 1056/529, loss: 0.001468658447265625 2023-01-21 11:51:55.976912: step: 1060/529, loss: 0.00111303327139467 2023-01-21 11:51:57.096116: step: 1064/529, loss: 0.24889594316482544 2023-01-21 11:51:58.200554: step: 1068/529, loss: 0.002685165498405695 2023-01-21 11:51:59.290754: step: 1072/529, loss: 0.0006547927623614669 2023-01-21 11:52:00.408126: step: 1076/529, loss: 0.00028820039005950093 2023-01-21 11:52:01.525074: step: 1080/529, loss: 0.04075603559613228 2023-01-21 11:52:02.624380: step: 1084/529, loss: 0.07119331508874893 2023-01-21 11:52:03.742636: step: 1088/529, loss: 0.01147375162690878 2023-01-21 11:52:04.879088: step: 1092/529, loss: 0.19418220221996307 2023-01-21 11:52:06.015602: step: 1096/529, loss: 0.007575607392936945 2023-01-21 11:52:07.141901: step: 1100/529, loss: 0.049197006970644 2023-01-21 11:52:08.276871: step: 1104/529, loss: 0.00016412735567428172 2023-01-21 11:52:09.382157: step: 1108/529, loss: 0.0010312080848962069 2023-01-21 11:52:10.496714: step: 1112/529, loss: 0.014920998364686966 2023-01-21 11:52:11.605543: step: 1116/529, loss: 0.007286357693374157 2023-01-21 11:52:12.735093: step: 1120/529, loss: 0.002814102452248335 2023-01-21 11:52:13.822933: step: 1124/529, loss: 0.0026006698608398438 2023-01-21 11:52:14.917814: step: 1128/529, loss: 0.021419048309326172 2023-01-21 11:52:16.028021: step: 1132/529, loss: 0.05027348920702934 2023-01-21 11:52:17.136970: step: 1136/529, loss: 0.048186734318733215 2023-01-21 11:52:18.250350: step: 1140/529, loss: 0.003751134965568781 2023-01-21 11:52:19.347679: step: 1144/529, loss: 0.0060787200927734375 2023-01-21 11:52:20.474987: step: 1148/529, loss: 0.03135070949792862 2023-01-21 11:52:21.621375: step: 1152/529, loss: 0.023996246978640556 2023-01-21 11:52:22.737758: step: 1156/529, loss: 0.004766654688864946 2023-01-21 11:52:23.881567: step: 1160/529, loss: 0.002567768096923828 2023-01-21 11:52:25.005503: step: 1164/529, loss: 0.00115032191388309 2023-01-21 11:52:26.157389: step: 1168/529, loss: 0.0008221626630984247 2023-01-21 11:52:27.260571: step: 1172/529, loss: 0.0032484056428074837 2023-01-21 11:52:28.361567: step: 1176/529, loss: 0.010023307986557484 2023-01-21 11:52:29.466648: step: 1180/529, loss: 0.22237291932106018 2023-01-21 11:52:30.579474: step: 1184/529, loss: 0.0008107662433758378 2023-01-21 11:52:31.687081: step: 1188/529, loss: 0.003206253284588456 2023-01-21 11:52:32.779780: step: 1192/529, loss: 0.00039157868013717234 2023-01-21 11:52:33.889188: step: 1196/529, loss: 0.002326059387996793 2023-01-21 11:52:34.983834: step: 1200/529, loss: 8.869171142578125e-05 2023-01-21 11:52:36.125949: step: 1204/529, loss: 0.00040845872717909515 2023-01-21 11:52:37.241583: step: 1208/529, loss: 0.013240433298051357 2023-01-21 11:52:38.353275: step: 1212/529, loss: 0.02336854860186577 2023-01-21 11:52:39.468411: step: 1216/529, loss: 0.016819000244140625 2023-01-21 11:52:40.619353: step: 1220/529, loss: 0.0018589496612548828 2023-01-21 11:52:41.743330: step: 1224/529, loss: 0.2546026110649109 2023-01-21 11:52:42.854834: step: 1228/529, loss: 0.014909124001860619 2023-01-21 11:52:43.948421: step: 1232/529, loss: 0.005594539921730757 2023-01-21 11:52:45.111571: step: 1236/529, loss: 0.0008432388422079384 2023-01-21 11:52:46.217491: step: 1240/529, loss: 0.019699478521943092 2023-01-21 11:52:47.345905: step: 1244/529, loss: 0.0480710007250309 2023-01-21 11:52:48.437157: step: 1248/529, loss: 0.027770565822720528 2023-01-21 11:52:49.527211: step: 1252/529, loss: 0.005346587393432856 2023-01-21 11:52:50.671018: step: 1256/529, loss: 0.005622673314064741 2023-01-21 11:52:51.817125: step: 1260/529, loss: 0.03550710901618004 2023-01-21 11:52:52.936842: step: 1264/529, loss: 0.00888280849903822 2023-01-21 11:52:54.065735: step: 1268/529, loss: 0.016545962542295456 2023-01-21 11:52:55.202068: step: 1272/529, loss: 0.1267203390598297 2023-01-21 11:52:56.309830: step: 1276/529, loss: 0.002758884336799383 2023-01-21 11:52:57.373302: step: 1280/529, loss: 0.0006591796991415322 2023-01-21 11:52:58.509905: step: 1284/529, loss: 0.07975693047046661 2023-01-21 11:52:59.638607: step: 1288/529, loss: 0.009151888079941273 2023-01-21 11:53:00.753414: step: 1292/529, loss: 0.00054168701171875 2023-01-21 11:53:01.867548: step: 1296/529, loss: 0.014697838574647903 2023-01-21 11:53:02.939275: step: 1300/529, loss: 0.00033702849759720266 2023-01-21 11:53:04.053704: step: 1304/529, loss: 0.008200549520552158 2023-01-21 11:53:05.152752: step: 1308/529, loss: 0.0017564772861078382 2023-01-21 11:53:06.273327: step: 1312/529, loss: 0.009581851772964 2023-01-21 11:53:07.402864: step: 1316/529, loss: 0.0881785899400711 2023-01-21 11:53:08.491057: step: 1320/529, loss: 0.001798439072445035 2023-01-21 11:53:09.579300: step: 1324/529, loss: 0.0024221420753747225 2023-01-21 11:53:10.694024: step: 1328/529, loss: 0.37553003430366516 2023-01-21 11:53:11.804644: step: 1332/529, loss: 0.008842849172651768 2023-01-21 11:53:12.911485: step: 1336/529, loss: 0.009419179521501064 2023-01-21 11:53:13.987973: step: 1340/529, loss: 0.06424083560705185 2023-01-21 11:53:15.087979: step: 1344/529, loss: 0.0006669044378213584 2023-01-21 11:53:16.203386: step: 1348/529, loss: 0.010067272931337357 2023-01-21 11:53:17.335467: step: 1352/529, loss: 0.024624444544315338 2023-01-21 11:53:18.426113: step: 1356/529, loss: 0.00037059784517623484 2023-01-21 11:53:19.491213: step: 1360/529, loss: 0.01033774670213461 2023-01-21 11:53:20.593300: step: 1364/529, loss: 0.06833630055189133 2023-01-21 11:53:21.695378: step: 1368/529, loss: 0.003238201141357422 2023-01-21 11:53:22.781320: step: 1372/529, loss: 0.002369022462517023 2023-01-21 11:53:23.901938: step: 1376/529, loss: 0.12176161259412766 2023-01-21 11:53:25.051562: step: 1380/529, loss: 0.000920248101465404 2023-01-21 11:53:26.172703: step: 1384/529, loss: 0.0020751953125 2023-01-21 11:53:27.330675: step: 1388/529, loss: 0.0157470703125 2023-01-21 11:53:28.468288: step: 1392/529, loss: 0.0021440505515784025 2023-01-21 11:53:29.583799: step: 1396/529, loss: 0.024956895038485527 2023-01-21 11:53:30.708395: step: 1400/529, loss: 0.007620644755661488 2023-01-21 11:53:31.790765: step: 1404/529, loss: 0.037957288324832916 2023-01-21 11:53:32.878343: step: 1408/529, loss: 0.03819417953491211 2023-01-21 11:53:34.020525: step: 1412/529, loss: 0.028168868273496628 2023-01-21 11:53:35.113139: step: 1416/529, loss: 0.3601018786430359 2023-01-21 11:53:36.204472: step: 1420/529, loss: 0.01972637139260769 2023-01-21 11:53:37.320433: step: 1424/529, loss: 0.0028842927422374487 2023-01-21 11:53:38.463454: step: 1428/529, loss: 0.0618322379887104 2023-01-21 11:53:39.591776: step: 1432/529, loss: 7.629387255292386e-07 2023-01-21 11:53:40.726121: step: 1436/529, loss: 0.08798065036535263 2023-01-21 11:53:41.831460: step: 1440/529, loss: 0.00034427642822265625 2023-01-21 11:53:42.973119: step: 1444/529, loss: 0.00987997092306614 2023-01-21 11:53:44.105167: step: 1448/529, loss: 0.0006999970064498484 2023-01-21 11:53:45.210303: step: 1452/529, loss: 0.010639239102602005 2023-01-21 11:53:46.331557: step: 1456/529, loss: 0.021783066913485527 2023-01-21 11:53:47.432447: step: 1460/529, loss: 0.0015151978004723787 2023-01-21 11:53:48.569758: step: 1464/529, loss: 0.07769856601953506 2023-01-21 11:53:49.678489: step: 1468/529, loss: 0.0008311271667480469 2023-01-21 11:53:50.804979: step: 1472/529, loss: 0.0011723518837243319 2023-01-21 11:53:51.925034: step: 1476/529, loss: 0.0002541542053222656 2023-01-21 11:53:53.041277: step: 1480/529, loss: 0.0057230000384151936 2023-01-21 11:53:54.142967: step: 1484/529, loss: 0.02397918701171875 2023-01-21 11:53:55.274560: step: 1488/529, loss: 0.034673500806093216 2023-01-21 11:53:56.399169: step: 1492/529, loss: 0.2048366516828537 2023-01-21 11:53:57.513808: step: 1496/529, loss: 0.0008666992071084678 2023-01-21 11:53:58.650956: step: 1500/529, loss: 0.031070329248905182 2023-01-21 11:53:59.786067: step: 1504/529, loss: 0.0027331113815307617 2023-01-21 11:54:00.924464: step: 1508/529, loss: 0.013514221645891666 2023-01-21 11:54:02.051407: step: 1512/529, loss: 0.01528463326394558 2023-01-21 11:54:03.143828: step: 1516/529, loss: 0.005129528231918812 2023-01-21 11:54:04.247808: step: 1520/529, loss: 7.848739915061742e-05 2023-01-21 11:54:05.364030: step: 1524/529, loss: 0.0653371810913086 2023-01-21 11:54:06.475767: step: 1528/529, loss: 5.1689145038835704e-05 2023-01-21 11:54:07.619447: step: 1532/529, loss: 0.0012197494506835938 2023-01-21 11:54:08.722035: step: 1536/529, loss: 0.0005622863536700606 2023-01-21 11:54:09.854407: step: 1540/529, loss: 0.011204147711396217 2023-01-21 11:54:11.013866: step: 1544/529, loss: 0.008355235680937767 2023-01-21 11:54:12.117849: step: 1548/529, loss: 0.050809670239686966 2023-01-21 11:54:13.240720: step: 1552/529, loss: 0.023400498554110527 2023-01-21 11:54:14.350206: step: 1556/529, loss: 0.007494163233786821 2023-01-21 11:54:15.485023: step: 1560/529, loss: 0.03584270551800728 2023-01-21 11:54:16.597826: step: 1564/529, loss: 0.0004709243949037045 2023-01-21 11:54:17.696910: step: 1568/529, loss: 0.10162229835987091 2023-01-21 11:54:18.812980: step: 1572/529, loss: 0.005187034606933594 2023-01-21 11:54:19.921580: step: 1576/529, loss: 0.0015937327407300472 2023-01-21 11:54:21.044171: step: 1580/529, loss: 0.0359896644949913 2023-01-21 11:54:22.192681: step: 1584/529, loss: 0.06848374009132385 2023-01-21 11:54:23.282543: step: 1588/529, loss: 0.014183426275849342 2023-01-21 11:54:24.443208: step: 1592/529, loss: 0.0007497787592001259 2023-01-21 11:54:25.538478: step: 1596/529, loss: 0.04402303695678711 2023-01-21 11:54:26.627723: step: 1600/529, loss: 0.09942970424890518 2023-01-21 11:54:27.778360: step: 1604/529, loss: 0.012021827511489391 2023-01-21 11:54:28.891027: step: 1608/529, loss: 0.000706577324308455 2023-01-21 11:54:29.996659: step: 1612/529, loss: 0.0056008342653512955 2023-01-21 11:54:31.104620: step: 1616/529, loss: 0.0052515375427901745 2023-01-21 11:54:32.213463: step: 1620/529, loss: 0.0035362245980650187 2023-01-21 11:54:33.311344: step: 1624/529, loss: 0.00013875961303710938 2023-01-21 11:54:34.441860: step: 1628/529, loss: 0.01101598795503378 2023-01-21 11:54:35.563549: step: 1632/529, loss: 0.004846763331443071 2023-01-21 11:54:36.686157: step: 1636/529, loss: 0.0014327527023851871 2023-01-21 11:54:37.786668: step: 1640/529, loss: 0.004573249723762274 2023-01-21 11:54:38.897916: step: 1644/529, loss: 0.011237716302275658 2023-01-21 11:54:39.993719: step: 1648/529, loss: 0.01625981368124485 2023-01-21 11:54:41.100362: step: 1652/529, loss: 0.004238605499267578 2023-01-21 11:54:42.215555: step: 1656/529, loss: 0.02864856831729412 2023-01-21 11:54:43.310882: step: 1660/529, loss: 0.17483024299144745 2023-01-21 11:54:44.453184: step: 1664/529, loss: 0.6686692237854004 2023-01-21 11:54:45.554906: step: 1668/529, loss: 0.06777181476354599 2023-01-21 11:54:46.663289: step: 1672/529, loss: 0.04550151899456978 2023-01-21 11:54:47.750031: step: 1676/529, loss: 0.0031262398697435856 2023-01-21 11:54:48.852605: step: 1680/529, loss: 0.013519239611923695 2023-01-21 11:54:49.981054: step: 1684/529, loss: 0.0050182342529296875 2023-01-21 11:54:51.100760: step: 1688/529, loss: 0.028247740119695663 2023-01-21 11:54:52.220636: step: 1692/529, loss: 0.00013875961303710938 2023-01-21 11:54:53.321747: step: 1696/529, loss: 0.022792436182498932 2023-01-21 11:54:54.452242: step: 1700/529, loss: 0.003612041473388672 2023-01-21 11:54:55.572045: step: 1704/529, loss: 0.0006196022150106728 2023-01-21 11:54:56.688236: step: 1708/529, loss: 0.01440343912690878 2023-01-21 11:54:57.811094: step: 1712/529, loss: 0.03115013986825943 2023-01-21 11:54:58.938085: step: 1716/529, loss: 0.024906540289521217 2023-01-21 11:55:00.053104: step: 1720/529, loss: 0.028415298089385033 2023-01-21 11:55:01.159531: step: 1724/529, loss: 0.018455125391483307 2023-01-21 11:55:02.230733: step: 1728/529, loss: 8.039474778342992e-05 2023-01-21 11:55:03.380535: step: 1732/529, loss: 0.023265648633241653 2023-01-21 11:55:04.495586: step: 1736/529, loss: 0.014386728405952454 2023-01-21 11:55:05.599878: step: 1740/529, loss: 0.008734608069062233 2023-01-21 11:55:06.711769: step: 1744/529, loss: 0.017172623425722122 2023-01-21 11:55:07.862220: step: 1748/529, loss: 0.00036087038461118937 2023-01-21 11:55:08.975045: step: 1752/529, loss: 0.028025055304169655 2023-01-21 11:55:10.082267: step: 1756/529, loss: 0.068903349339962 2023-01-21 11:55:11.203839: step: 1760/529, loss: 0.07005921006202698 2023-01-21 11:55:12.302519: step: 1764/529, loss: 0.0003143310605082661 2023-01-21 11:55:13.432465: step: 1768/529, loss: 0.00530853308737278 2023-01-21 11:55:14.528667: step: 1772/529, loss: 0.0020563125144690275 2023-01-21 11:55:15.640569: step: 1776/529, loss: 0.00047931671724654734 2023-01-21 11:55:16.759442: step: 1780/529, loss: 0.00572280865162611 2023-01-21 11:55:17.872152: step: 1784/529, loss: 0.0036665915977209806 2023-01-21 11:55:18.972734: step: 1788/529, loss: 0.024647904559969902 2023-01-21 11:55:20.105876: step: 1792/529, loss: 0.15339794754981995 2023-01-21 11:55:21.224601: step: 1796/529, loss: 0.0010932922596111894 2023-01-21 11:55:22.364836: step: 1800/529, loss: 0.00855483952909708 2023-01-21 11:55:23.458673: step: 1804/529, loss: 0.045563891530036926 2023-01-21 11:55:24.585107: step: 1808/529, loss: 0.0002680778852663934 2023-01-21 11:55:25.761970: step: 1812/529, loss: 0.01278610248118639 2023-01-21 11:55:26.865014: step: 1816/529, loss: 0.09324774891138077 2023-01-21 11:55:27.966478: step: 1820/529, loss: 4.158019874012098e-05 2023-01-21 11:55:29.063460: step: 1824/529, loss: 0.0025367499329149723 2023-01-21 11:55:30.176646: step: 1828/529, loss: 0.0010714053642004728 2023-01-21 11:55:31.301624: step: 1832/529, loss: 0.08440761268138885 2023-01-21 11:55:32.446781: step: 1836/529, loss: 0.009079933166503906 2023-01-21 11:55:33.574860: step: 1840/529, loss: 0.004768466576933861 2023-01-21 11:55:34.692480: step: 1844/529, loss: 0.000629425048828125 2023-01-21 11:55:35.778277: step: 1848/529, loss: 0.056734468787908554 2023-01-21 11:55:36.866145: step: 1852/529, loss: 0.000171661376953125 2023-01-21 11:55:37.989767: step: 1856/529, loss: 0.015237808227539062 2023-01-21 11:55:39.109086: step: 1860/529, loss: 0.08406729251146317 2023-01-21 11:55:40.241736: step: 1864/529, loss: 0.012540149502456188 2023-01-21 11:55:41.387581: step: 1868/529, loss: 0.0014110564952716231 2023-01-21 11:55:42.508694: step: 1872/529, loss: 0.0038299560546875 2023-01-21 11:55:43.631637: step: 1876/529, loss: 0.0012363434070721269 2023-01-21 11:55:44.769604: step: 1880/529, loss: 0.004500627517700195 2023-01-21 11:55:45.875301: step: 1884/529, loss: 0.02730264887213707 2023-01-21 11:55:46.989911: step: 1888/529, loss: 0.0019573213066905737 2023-01-21 11:55:48.106055: step: 1892/529, loss: 0.0054178242571651936 2023-01-21 11:55:49.234909: step: 1896/529, loss: 0.00030956268892623484 2023-01-21 11:55:50.385713: step: 1900/529, loss: 0.00115966796875 2023-01-21 11:55:51.522359: step: 1904/529, loss: 0.014822816476225853 2023-01-21 11:55:52.650408: step: 1908/529, loss: 0.020738791674375534 2023-01-21 11:55:53.756041: step: 1912/529, loss: 0.028678035363554955 2023-01-21 11:55:54.883230: step: 1916/529, loss: 0.012335110455751419 2023-01-21 11:55:55.978189: step: 1920/529, loss: 0.32428044080734253 2023-01-21 11:55:57.074995: step: 1924/529, loss: 0.00010275840759277344 2023-01-21 11:55:58.176722: step: 1928/529, loss: 0.001251316163688898 2023-01-21 11:55:59.294875: step: 1932/529, loss: 0.0030506611801683903 2023-01-21 11:56:00.388924: step: 1936/529, loss: 0.0007130622980184853 2023-01-21 11:56:01.489941: step: 1940/529, loss: 0.00028476715669967234 2023-01-21 11:56:02.618342: step: 1944/529, loss: 0.6257377862930298 2023-01-21 11:56:03.760675: step: 1948/529, loss: 0.00018177033052779734 2023-01-21 11:56:04.886513: step: 1952/529, loss: 0.01791834831237793 2023-01-21 11:56:05.993557: step: 1956/529, loss: 0.00022716523380950093 2023-01-21 11:56:07.105387: step: 1960/529, loss: 0.002274131868034601 2023-01-21 11:56:08.203993: step: 1964/529, loss: 0.05136413499712944 2023-01-21 11:56:09.296242: step: 1968/529, loss: 0.012688255868852139 2023-01-21 11:56:10.429243: step: 1972/529, loss: 0.009505749680101871 2023-01-21 11:56:11.530891: step: 1976/529, loss: 0.0001369476376567036 2023-01-21 11:56:12.618144: step: 1980/529, loss: 0.00018324852862861007 2023-01-21 11:56:13.736345: step: 1984/529, loss: 0.03185243532061577 2023-01-21 11:56:14.850921: step: 1988/529, loss: 0.0008454322814941406 2023-01-21 11:56:15.957647: step: 1992/529, loss: 0.02744889259338379 2023-01-21 11:56:17.085462: step: 1996/529, loss: 0.027533387765288353 2023-01-21 11:56:18.204702: step: 2000/529, loss: 0.005228138063102961 2023-01-21 11:56:19.316208: step: 2004/529, loss: 0.01917858049273491 2023-01-21 11:56:20.454239: step: 2008/529, loss: 0.023989295586943626 2023-01-21 11:56:21.599796: step: 2012/529, loss: 0.0012901783920824528 2023-01-21 11:56:22.750430: step: 2016/529, loss: 0.00036678314791060984 2023-01-21 11:56:23.873477: step: 2020/529, loss: 0.01152567844837904 2023-01-21 11:56:24.968916: step: 2024/529, loss: 0.005275917239487171 2023-01-21 11:56:26.098441: step: 2028/529, loss: 0.02753319777548313 2023-01-21 11:56:27.209244: step: 2032/529, loss: 0.02874736860394478 2023-01-21 11:56:28.336071: step: 2036/529, loss: 0.004689216613769531 2023-01-21 11:56:29.429894: step: 2040/529, loss: 0.006198406219482422 2023-01-21 11:56:30.547828: step: 2044/529, loss: 0.013045703992247581 2023-01-21 11:56:31.663116: step: 2048/529, loss: 0.0075130462646484375 2023-01-21 11:56:32.751472: step: 2052/529, loss: 0.0035832407884299755 2023-01-21 11:56:33.910013: step: 2056/529, loss: 0.0739908218383789 2023-01-21 11:56:35.020645: step: 2060/529, loss: 0.000676059746183455 2023-01-21 11:56:36.154910: step: 2064/529, loss: 0.0018342972034588456 2023-01-21 11:56:37.252681: step: 2068/529, loss: 0.0003898620780091733 2023-01-21 11:56:38.397585: step: 2072/529, loss: 0.01401367224752903 2023-01-21 11:56:39.520839: step: 2076/529, loss: 0.009934616275131702 2023-01-21 11:56:40.669851: step: 2080/529, loss: 0.0006170272827148438 2023-01-21 11:56:41.768481: step: 2084/529, loss: 0.0003917693975381553 2023-01-21 11:56:42.923173: step: 2088/529, loss: 0.0016091347206383944 2023-01-21 11:56:44.040957: step: 2092/529, loss: 0.05111531913280487 2023-01-21 11:56:45.142013: step: 2096/529, loss: 0.04278087615966797 2023-01-21 11:56:46.272861: step: 2100/529, loss: 0.7997318506240845 2023-01-21 11:56:47.400959: step: 2104/529, loss: 0.00029544829158112407 2023-01-21 11:56:48.503900: step: 2108/529, loss: 0.010148381814360619 2023-01-21 11:56:49.607350: step: 2112/529, loss: 0.04116103798151016 2023-01-21 11:56:50.729010: step: 2116/529, loss: 0.0017887115245684981 ================================================== Loss: 0.030 -------------------- Dev: {'event': {'p': 0.5898989898989899, 'r': 0.7776298268974701, 'f1': 0.6708788052843194}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Test: {'event': {'p': 0.6178458765209554, 'r': 0.7737020316027088, 'f1': 0.6870458531696316}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Chinese: {'event': {'p': 0.5494505494505495, 'r': 0.9259259259259259, 'f1': 0.6896551724137931}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Korean: {'event': {'p': 0.65, 'r': 0.6190476190476191, 'f1': 0.6341463414634146}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Russian: {'event': {'p': 0.4318181818181818, 'r': 0.5277777777777778, 'f1': 0.47500000000000003}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 20 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:57:31.496514: step: 4/529, loss: 0.020984530448913574 2023-01-21 11:57:32.657758: step: 8/529, loss: 0.024711037054657936 2023-01-21 11:57:33.764963: step: 12/529, loss: 0.01419610995799303 2023-01-21 11:57:34.867177: step: 16/529, loss: 0.0010246275924146175 2023-01-21 11:57:35.985180: step: 20/529, loss: 0.006266022101044655 2023-01-21 11:57:37.096139: step: 24/529, loss: 0.007914543151855469 2023-01-21 11:57:38.223446: step: 28/529, loss: 0.014003658667206764 2023-01-21 11:57:39.343619: step: 32/529, loss: 0.0009417533874511719 2023-01-21 11:57:40.445231: step: 36/529, loss: 0.007920265197753906 2023-01-21 11:57:41.565752: step: 40/529, loss: 0.021269990131258965 2023-01-21 11:57:42.675469: step: 44/529, loss: 0.0017608166672289371 2023-01-21 11:57:43.813846: step: 48/529, loss: 0.0015537261497229338 2023-01-21 11:57:44.915299: step: 52/529, loss: 0.0004150390741415322 2023-01-21 11:57:46.015022: step: 56/529, loss: 3.361701965332031e-05 2023-01-21 11:57:47.138097: step: 60/529, loss: 0.007055950351059437 2023-01-21 11:57:48.272601: step: 64/529, loss: 0.004631614778190851 2023-01-21 11:57:49.397389: step: 68/529, loss: 0.0010828971862792969 2023-01-21 11:57:50.499987: step: 72/529, loss: 0.002620199229568243 2023-01-21 11:57:51.614338: step: 76/529, loss: 0.055018093436956406 2023-01-21 11:57:52.715595: step: 80/529, loss: 0.0021310807205736637 2023-01-21 11:57:53.822623: step: 84/529, loss: 0.017327500507235527 2023-01-21 11:57:54.899617: step: 88/529, loss: 0.05563826858997345 2023-01-21 11:57:56.004009: step: 92/529, loss: 0.013677025213837624 2023-01-21 11:57:57.103184: step: 96/529, loss: 0.0010696410899981856 2023-01-21 11:57:58.178952: step: 100/529, loss: 0.008022118359804153 2023-01-21 11:57:59.305006: step: 104/529, loss: 0.009745026007294655 2023-01-21 11:58:00.419329: step: 108/529, loss: 0.0053882603533566 2023-01-21 11:58:01.554292: step: 112/529, loss: 0.016594314947724342 2023-01-21 11:58:02.664234: step: 116/529, loss: 0.011065680533647537 2023-01-21 11:58:03.789053: step: 120/529, loss: 2.689361645025201e-05 2023-01-21 11:58:04.896609: step: 124/529, loss: 0.04512729495763779 2023-01-21 11:58:05.984344: step: 128/529, loss: 0.012955261394381523 2023-01-21 11:58:07.103036: step: 132/529, loss: 0.03737945854663849 2023-01-21 11:58:08.205609: step: 136/529, loss: 0.039859771728515625 2023-01-21 11:58:09.295948: step: 140/529, loss: 0.04422473907470703 2023-01-21 11:58:10.425999: step: 144/529, loss: 0.3335091471672058 2023-01-21 11:58:11.553973: step: 148/529, loss: 0.008353996090590954 2023-01-21 11:58:12.657349: step: 152/529, loss: 0.060912229120731354 2023-01-21 11:58:13.754446: step: 156/529, loss: 0.02604498900473118 2023-01-21 11:58:14.872767: step: 160/529, loss: 0.0006613731384277344 2023-01-21 11:58:15.986478: step: 164/529, loss: 0.026853563264012337 2023-01-21 11:58:17.080390: step: 168/529, loss: 0.00016641616821289062 2023-01-21 11:58:18.221855: step: 172/529, loss: 0.005078315734863281 2023-01-21 11:58:19.315811: step: 176/529, loss: 0.00032548903254792094 2023-01-21 11:58:20.415222: step: 180/529, loss: 0.0005249023670330644 2023-01-21 11:58:21.528993: step: 184/529, loss: 8.93592878128402e-05 2023-01-21 11:58:22.621058: step: 188/529, loss: 0.0009150505647994578 2023-01-21 11:58:23.717420: step: 192/529, loss: 0.004787635989487171 2023-01-21 11:58:24.823350: step: 196/529, loss: 0.007172918412834406 2023-01-21 11:58:25.908525: step: 200/529, loss: 0.0004311561933718622 2023-01-21 11:58:27.003790: step: 204/529, loss: 0.0011467933654785156 2023-01-21 11:58:28.147102: step: 208/529, loss: 0.01928262785077095 2023-01-21 11:58:29.275298: step: 212/529, loss: 0.0008554458618164062 2023-01-21 11:58:30.383576: step: 216/529, loss: 0.0002612113894429058 2023-01-21 11:58:31.470829: step: 220/529, loss: 0.01692953146994114 2023-01-21 11:58:32.585725: step: 224/529, loss: 0.0007033348083496094 2023-01-21 11:58:33.700630: step: 228/529, loss: 0.002319431398063898 2023-01-21 11:58:34.797976: step: 232/529, loss: 0.18481312692165375 2023-01-21 11:58:35.907016: step: 236/529, loss: 0.7903178930282593 2023-01-21 11:58:37.023217: step: 240/529, loss: 0.002063655760139227 2023-01-21 11:58:38.139075: step: 244/529, loss: 0.011079215444624424 2023-01-21 11:58:39.258095: step: 248/529, loss: 0.0017589569324627519 2023-01-21 11:58:40.388894: step: 252/529, loss: 0.006633568089455366 2023-01-21 11:58:41.527602: step: 256/529, loss: 0.00037751198397018015 2023-01-21 11:58:42.623541: step: 260/529, loss: 0.005647039506584406 2023-01-21 11:58:43.762824: step: 264/529, loss: 0.0020733834244310856 2023-01-21 11:58:44.898202: step: 268/529, loss: 0.0057232859544456005 2023-01-21 11:58:46.014395: step: 272/529, loss: 0.0011430741287767887 2023-01-21 11:58:47.144497: step: 276/529, loss: 0.0009099960443563759 2023-01-21 11:58:48.288339: step: 280/529, loss: 0.00025367739726789296 2023-01-21 11:58:49.432626: step: 284/529, loss: 0.0008880614768713713 2023-01-21 11:58:50.551949: step: 288/529, loss: 0.003036880400031805 2023-01-21 11:58:51.667914: step: 292/529, loss: 0.0017561913700774312 2023-01-21 11:58:52.783135: step: 296/529, loss: 0.007814884185791016 2023-01-21 11:58:53.880700: step: 300/529, loss: 0.02033405192196369 2023-01-21 11:58:55.008635: step: 304/529, loss: 0.004198742099106312 2023-01-21 11:58:56.124906: step: 308/529, loss: 0.0005224705091677606 2023-01-21 11:58:57.233633: step: 312/529, loss: 0.0005334377055987716 2023-01-21 11:58:58.369624: step: 316/529, loss: 0.00020122528076171875 2023-01-21 11:58:59.476284: step: 320/529, loss: 5.445480201160535e-05 2023-01-21 11:59:00.577870: step: 324/529, loss: 9.13620024221018e-05 2023-01-21 11:59:01.691238: step: 328/529, loss: 0.0016367912758141756 2023-01-21 11:59:02.795787: step: 332/529, loss: 0.00014352798461914062 2023-01-21 11:59:03.904218: step: 336/529, loss: 0.0016219615936279297 2023-01-21 11:59:05.050919: step: 340/529, loss: 0.000988960382528603 2023-01-21 11:59:06.186018: step: 344/529, loss: 0.00015392302884720266 2023-01-21 11:59:07.261195: step: 348/529, loss: 0.0006551742553710938 2023-01-21 11:59:08.374680: step: 352/529, loss: 0.17161349952220917 2023-01-21 11:59:09.473862: step: 356/529, loss: 5.817413693876006e-06 2023-01-21 11:59:10.629902: step: 360/529, loss: 0.0066123963333666325 2023-01-21 11:59:11.758366: step: 364/529, loss: 7.686615572310984e-05 2023-01-21 11:59:12.861955: step: 368/529, loss: 6.952285912120715e-05 2023-01-21 11:59:13.975678: step: 372/529, loss: 0.05263710021972656 2023-01-21 11:59:15.092854: step: 376/529, loss: 0.00021228790865279734 2023-01-21 11:59:16.210505: step: 380/529, loss: 0.0030956268310546875 2023-01-21 11:59:17.335702: step: 384/529, loss: 0.008430957794189453 2023-01-21 11:59:18.423900: step: 388/529, loss: 0.0005131721263751388 2023-01-21 11:59:19.521631: step: 392/529, loss: 0.0002020835963776335 2023-01-21 11:59:20.638494: step: 396/529, loss: 0.005168342962861061 2023-01-21 11:59:21.758647: step: 400/529, loss: 0.030016519129276276 2023-01-21 11:59:22.892127: step: 404/529, loss: 0.02531719207763672 2023-01-21 11:59:23.982977: step: 408/529, loss: 0.0022316696122288704 2023-01-21 11:59:25.123998: step: 412/529, loss: 0.02414998970925808 2023-01-21 11:59:26.219688: step: 416/529, loss: 0.00096893310546875 2023-01-21 11:59:27.306248: step: 420/529, loss: 0.002853012178093195 2023-01-21 11:59:28.423303: step: 424/529, loss: 0.004511356353759766 2023-01-21 11:59:29.542932: step: 428/529, loss: 0.0008190632215701044 2023-01-21 11:59:30.673404: step: 432/529, loss: 0.015511131845414639 2023-01-21 11:59:31.762549: step: 436/529, loss: 0.0012537003494799137 2023-01-21 11:59:32.918801: step: 440/529, loss: 0.006739330478012562 2023-01-21 11:59:34.047315: step: 444/529, loss: 0.002609825227409601 2023-01-21 11:59:35.171267: step: 448/529, loss: 0.0006969452369958162 2023-01-21 11:59:36.291074: step: 452/529, loss: 0.0026191710494458675 2023-01-21 11:59:37.403822: step: 456/529, loss: 0.007627010345458984 2023-01-21 11:59:38.495515: step: 460/529, loss: 0.00930566806346178 2023-01-21 11:59:39.599200: step: 464/529, loss: 0.014107896015048027 2023-01-21 11:59:40.696844: step: 468/529, loss: 0.00014209747314453125 2023-01-21 11:59:41.787717: step: 472/529, loss: 0.0002609252987895161 2023-01-21 11:59:42.908825: step: 476/529, loss: 0.011373329907655716 2023-01-21 11:59:44.017908: step: 480/529, loss: 0.00033054352388717234 2023-01-21 11:59:45.130229: step: 484/529, loss: 0.0007009506225585938 2023-01-21 11:59:46.245506: step: 488/529, loss: 8.726120540814009e-06 2023-01-21 11:59:47.360050: step: 492/529, loss: 0.026540184393525124 2023-01-21 11:59:48.494128: step: 496/529, loss: 0.00015287399583030492 2023-01-21 11:59:49.625171: step: 500/529, loss: 0.017219258472323418 2023-01-21 11:59:50.774371: step: 504/529, loss: 0.05229397118091583 2023-01-21 11:59:51.913020: step: 508/529, loss: 0.0006369590992107987 2023-01-21 11:59:53.029547: step: 512/529, loss: 0.01734142377972603 2023-01-21 11:59:54.149612: step: 516/529, loss: 0.00670967111364007 2023-01-21 11:59:55.255034: step: 520/529, loss: 0.00349521660245955 2023-01-21 11:59:56.358908: step: 524/529, loss: 0.00948266964405775 2023-01-21 11:59:57.462714: step: 528/529, loss: 0.0013858318561688066 2023-01-21 11:59:58.544849: step: 532/529, loss: 5.8507917856331915e-05 2023-01-21 11:59:59.632457: step: 536/529, loss: 0.0011786938412114978 2023-01-21 12:00:00.779698: step: 540/529, loss: 0.0003211975272279233 2023-01-21 12:00:01.869359: step: 544/529, loss: 0.006136131007224321 2023-01-21 12:00:02.998741: step: 548/529, loss: 0.002990245819091797 2023-01-21 12:00:04.119094: step: 552/529, loss: 0.022269917652010918 2023-01-21 12:00:05.238050: step: 556/529, loss: 0.0010646820301190019 2023-01-21 12:00:06.367827: step: 560/529, loss: 0.00048022271948866546 2023-01-21 12:00:07.470799: step: 564/529, loss: 0.0010446547530591488 2023-01-21 12:00:08.588115: step: 568/529, loss: 0.05416097864508629 2023-01-21 12:00:09.740000: step: 572/529, loss: 0.00020618439884856343 2023-01-21 12:00:10.845094: step: 576/529, loss: 0.08019447326660156 2023-01-21 12:00:11.970009: step: 580/529, loss: 0.0031221387907862663 2023-01-21 12:00:13.054991: step: 584/529, loss: 0.004239177796989679 2023-01-21 12:00:14.155669: step: 588/529, loss: 0.00028438569279387593 2023-01-21 12:00:15.272156: step: 592/529, loss: 0.0190290454775095 2023-01-21 12:00:16.395608: step: 596/529, loss: 0.003799867583438754 2023-01-21 12:00:17.543064: step: 600/529, loss: 0.009310055524110794 2023-01-21 12:00:18.648489: step: 604/529, loss: 0.00041160586988553405 2023-01-21 12:00:19.769869: step: 608/529, loss: 0.0012542724143713713 2023-01-21 12:00:20.863811: step: 612/529, loss: 0.0010925770038738847 2023-01-21 12:00:21.967233: step: 616/529, loss: 1.9073486328125e-05 2023-01-21 12:00:23.075208: step: 620/529, loss: 0.02089872397482395 2023-01-21 12:00:24.181526: step: 624/529, loss: 0.00042786600533872843 2023-01-21 12:00:25.314097: step: 628/529, loss: 0.0002280235494254157 2023-01-21 12:00:26.419793: step: 632/529, loss: 0.00016098022751975805 2023-01-21 12:00:27.519908: step: 636/529, loss: 0.00039877890958450735 2023-01-21 12:00:28.629410: step: 640/529, loss: 0.02602844312787056 2023-01-21 12:00:29.728510: step: 644/529, loss: 3.833770824712701e-05 2023-01-21 12:00:30.828116: step: 648/529, loss: 0.005907726474106312 2023-01-21 12:00:31.917774: step: 652/529, loss: 0.0008953094366006553 2023-01-21 12:00:33.034450: step: 656/529, loss: 0.00023565292940475047 2023-01-21 12:00:34.155913: step: 660/529, loss: 0.0008718490717001259 2023-01-21 12:00:35.257245: step: 664/529, loss: 0.00237693777307868 2023-01-21 12:00:36.387020: step: 668/529, loss: 0.012457657605409622 2023-01-21 12:00:37.497763: step: 672/529, loss: 0.023097991943359375 2023-01-21 12:00:38.608886: step: 676/529, loss: 0.09297113120555878 2023-01-21 12:00:39.720850: step: 680/529, loss: 0.007905769161880016 2023-01-21 12:00:40.841547: step: 684/529, loss: 0.03787994384765625 2023-01-21 12:00:41.944441: step: 688/529, loss: 4.920959327137098e-05 2023-01-21 12:00:43.084720: step: 692/529, loss: 0.02614612877368927 2023-01-21 12:00:44.211806: step: 696/529, loss: 0.10093307495117188 2023-01-21 12:00:45.327413: step: 700/529, loss: 0.0029104233253747225 2023-01-21 12:00:46.441880: step: 704/529, loss: 0.0014276504516601562 2023-01-21 12:00:47.559854: step: 708/529, loss: 0.09107530117034912 2023-01-21 12:00:48.660173: step: 712/529, loss: 0.0002925872977357358 2023-01-21 12:00:49.750701: step: 716/529, loss: 0.017434168606996536 2023-01-21 12:00:50.850684: step: 720/529, loss: 0.00061626429669559 2023-01-21 12:00:51.950379: step: 724/529, loss: 0.03289041668176651 2023-01-21 12:00:53.073794: step: 728/529, loss: 0.0009641647338867188 2023-01-21 12:00:54.229792: step: 732/529, loss: 0.026391983032226562 2023-01-21 12:00:55.328880: step: 736/529, loss: 0.006184863857924938 2023-01-21 12:00:56.480713: step: 740/529, loss: 0.006008434109389782 2023-01-21 12:00:57.586352: step: 744/529, loss: 0.046254731714725494 2023-01-21 12:00:58.711720: step: 748/529, loss: 8.625983900856227e-05 2023-01-21 12:00:59.849159: step: 752/529, loss: 0.02526693418622017 2023-01-21 12:01:00.953451: step: 756/529, loss: 0.07945775985717773 2023-01-21 12:01:02.036829: step: 760/529, loss: 0.0048278807662427425 2023-01-21 12:01:03.179648: step: 764/529, loss: 0.04965219274163246 2023-01-21 12:01:04.284591: step: 768/529, loss: 0.0017281054751947522 2023-01-21 12:01:05.401292: step: 772/529, loss: 0.0003504753112792969 2023-01-21 12:01:06.523054: step: 776/529, loss: 0.016553306952118874 2023-01-21 12:01:07.644705: step: 780/529, loss: 0.0009123802301473916 2023-01-21 12:01:08.774116: step: 784/529, loss: 0.008713293820619583 2023-01-21 12:01:09.893389: step: 788/529, loss: 0.02239713817834854 2023-01-21 12:01:11.043545: step: 792/529, loss: 0.0030422210693359375 2023-01-21 12:01:12.180925: step: 796/529, loss: 0.04105987772345543 2023-01-21 12:01:13.314560: step: 800/529, loss: 0.05293693393468857 2023-01-21 12:01:14.427181: step: 804/529, loss: 0.0005140304565429688 2023-01-21 12:01:15.534462: step: 808/529, loss: 0.0014772415161132812 2023-01-21 12:01:16.650379: step: 812/529, loss: 0.0171419158577919 2023-01-21 12:01:17.765379: step: 816/529, loss: 0.02715468406677246 2023-01-21 12:01:18.890993: step: 820/529, loss: 0.0012449264759197831 2023-01-21 12:01:20.002670: step: 824/529, loss: 0.005867648404091597 2023-01-21 12:01:21.123581: step: 828/529, loss: 0.0019479751354083419 2023-01-21 12:01:22.225357: step: 832/529, loss: 0.0030104639008641243 2023-01-21 12:01:23.342466: step: 836/529, loss: 0.0190995205193758 2023-01-21 12:01:24.478891: step: 840/529, loss: 0.25192564725875854 2023-01-21 12:01:25.603562: step: 844/529, loss: 0.059175681322813034 2023-01-21 12:01:26.706564: step: 848/529, loss: 0.013447665609419346 2023-01-21 12:01:27.816878: step: 852/529, loss: 0.016181660816073418 2023-01-21 12:01:28.901862: step: 856/529, loss: 0.011273693293333054 2023-01-21 12:01:30.047874: step: 860/529, loss: 0.007645034696906805 2023-01-21 12:01:31.208460: step: 864/529, loss: 0.00015401840209960938 2023-01-21 12:01:32.326085: step: 868/529, loss: 0.00196418771520257 2023-01-21 12:01:33.455510: step: 872/529, loss: 0.011222553439438343 2023-01-21 12:01:34.551863: step: 876/529, loss: 0.07223377376794815 2023-01-21 12:01:35.681627: step: 880/529, loss: 0.021004199981689453 2023-01-21 12:01:36.832801: step: 884/529, loss: 0.9754303097724915 2023-01-21 12:01:37.929253: step: 888/529, loss: 0.000204181662411429 2023-01-21 12:01:39.026737: step: 892/529, loss: 0.03430728614330292 2023-01-21 12:01:40.110849: step: 896/529, loss: 0.0022115707397460938 2023-01-21 12:01:41.235474: step: 900/529, loss: 0.027570199221372604 2023-01-21 12:01:42.343451: step: 904/529, loss: 0.01960773579776287 2023-01-21 12:01:43.479807: step: 908/529, loss: 0.06493254005908966 2023-01-21 12:01:44.677337: step: 912/529, loss: 0.007010173983871937 2023-01-21 12:01:45.795044: step: 916/529, loss: 0.0026778222527354956 2023-01-21 12:01:46.916345: step: 920/529, loss: 0.024400806054472923 2023-01-21 12:01:48.005617: step: 924/529, loss: 0.0006547927623614669 2023-01-21 12:01:49.125670: step: 928/529, loss: 0.011786842718720436 2023-01-21 12:01:50.237191: step: 932/529, loss: 0.00019025802612304688 2023-01-21 12:01:51.358515: step: 936/529, loss: 0.00651474017649889 2023-01-21 12:01:52.488667: step: 940/529, loss: 0.00028705596923828125 2023-01-21 12:01:53.589257: step: 944/529, loss: 0.0037679672241210938 2023-01-21 12:01:54.684820: step: 948/529, loss: 0.005447006318718195 2023-01-21 12:01:55.828422: step: 952/529, loss: 1.4877318790240679e-05 2023-01-21 12:01:56.969380: step: 956/529, loss: 0.0004461407952476293 2023-01-21 12:01:58.078010: step: 960/529, loss: 0.00044097902718931437 2023-01-21 12:01:59.183818: step: 964/529, loss: 0.00014171600923873484 2023-01-21 12:02:00.322796: step: 968/529, loss: 0.00296535505913198 2023-01-21 12:02:01.425578: step: 972/529, loss: 0.008984755724668503 2023-01-21 12:02:02.533891: step: 976/529, loss: 0.009597063064575195 2023-01-21 12:02:03.641706: step: 980/529, loss: 0.0025451183319091797 2023-01-21 12:02:04.784085: step: 984/529, loss: 0.0005354404565878212 2023-01-21 12:02:05.886029: step: 988/529, loss: 0.09144258499145508 2023-01-21 12:02:06.971786: step: 992/529, loss: 0.002868366427719593 2023-01-21 12:02:08.076582: step: 996/529, loss: 0.00020256043353583664 2023-01-21 12:02:09.183800: step: 1000/529, loss: 0.0018606185913085938 2023-01-21 12:02:10.284695: step: 1004/529, loss: 0.0270251277834177 2023-01-21 12:02:11.404371: step: 1008/529, loss: 0.0188356414437294 2023-01-21 12:02:12.511899: step: 1012/529, loss: 0.0010814189445227385 2023-01-21 12:02:13.632538: step: 1016/529, loss: 4.434585662238533e-06 2023-01-21 12:02:14.792375: step: 1020/529, loss: 0.06488993018865585 2023-01-21 12:02:15.913441: step: 1024/529, loss: 0.0231507308781147 2023-01-21 12:02:17.049417: step: 1028/529, loss: 0.014320564456284046 2023-01-21 12:02:18.186257: step: 1032/529, loss: 0.0012389183975756168 2023-01-21 12:02:19.277707: step: 1036/529, loss: 0.00915222242474556 2023-01-21 12:02:20.413299: step: 1040/529, loss: 0.011677075177431107 2023-01-21 12:02:21.537887: step: 1044/529, loss: 0.005236148834228516 2023-01-21 12:02:22.661482: step: 1048/529, loss: 0.07618493586778641 2023-01-21 12:02:23.807656: step: 1052/529, loss: 0.019169950857758522 2023-01-21 12:02:24.924864: step: 1056/529, loss: 0.015366650186479092 2023-01-21 12:02:26.078390: step: 1060/529, loss: 0.0028061866760253906 2023-01-21 12:02:27.178355: step: 1064/529, loss: 0.0021330355666577816 2023-01-21 12:02:28.312446: step: 1068/529, loss: 0.00016889571270439774 2023-01-21 12:02:29.430911: step: 1072/529, loss: 5.483627683133818e-05 2023-01-21 12:02:30.538339: step: 1076/529, loss: 0.026884840801358223 2023-01-21 12:02:31.662746: step: 1080/529, loss: 6.523132469737902e-05 2023-01-21 12:02:32.799431: step: 1084/529, loss: 0.019875813275575638 2023-01-21 12:02:33.943945: step: 1088/529, loss: 0.004985237028449774 2023-01-21 12:02:35.040927: step: 1092/529, loss: 0.0018360137473791838 2023-01-21 12:02:36.171907: step: 1096/529, loss: 0.016112040728330612 2023-01-21 12:02:37.302654: step: 1100/529, loss: 0.0006847381591796875 2023-01-21 12:02:38.427537: step: 1104/529, loss: 0.03131461143493652 2023-01-21 12:02:39.537647: step: 1108/529, loss: 0.005988788791000843 2023-01-21 12:02:40.665618: step: 1112/529, loss: 0.0032947540748864412 2023-01-21 12:02:41.748943: step: 1116/529, loss: 0.03661594167351723 2023-01-21 12:02:42.842021: step: 1120/529, loss: 0.0005634307744912803 2023-01-21 12:02:43.969114: step: 1124/529, loss: 0.022788239642977715 2023-01-21 12:02:45.082951: step: 1128/529, loss: 0.027476264163851738 2023-01-21 12:02:46.178625: step: 1132/529, loss: 0.0023464204277843237 2023-01-21 12:02:47.281226: step: 1136/529, loss: 1.8978118532686494e-05 2023-01-21 12:02:48.381796: step: 1140/529, loss: 0.01290369126945734 2023-01-21 12:02:49.509507: step: 1144/529, loss: 0.0519598051905632 2023-01-21 12:02:50.677067: step: 1148/529, loss: 0.03210010379552841 2023-01-21 12:02:51.836049: step: 1152/529, loss: 0.0386316291987896 2023-01-21 12:02:52.957301: step: 1156/529, loss: 0.002997493837028742 2023-01-21 12:02:54.070632: step: 1160/529, loss: 0.002531051868572831 2023-01-21 12:02:55.183062: step: 1164/529, loss: 0.043094635009765625 2023-01-21 12:02:56.309610: step: 1168/529, loss: 0.0020805359818041325 2023-01-21 12:02:57.426453: step: 1172/529, loss: 0.005076789762824774 2023-01-21 12:02:58.533808: step: 1176/529, loss: 0.007420539855957031 2023-01-21 12:02:59.649097: step: 1180/529, loss: 0.007191467564553022 2023-01-21 12:03:00.777255: step: 1184/529, loss: 0.03194174915552139 2023-01-21 12:03:01.905564: step: 1188/529, loss: 0.0026786804664880037 2023-01-21 12:03:03.021952: step: 1192/529, loss: 0.0077419281005859375 2023-01-21 12:03:04.138865: step: 1196/529, loss: 0.02488403208553791 2023-01-21 12:03:05.241868: step: 1200/529, loss: 0.019365692511200905 2023-01-21 12:03:06.360083: step: 1204/529, loss: 0.06426334381103516 2023-01-21 12:03:07.501309: step: 1208/529, loss: 0.19323483109474182 2023-01-21 12:03:08.656660: step: 1212/529, loss: 0.0002995491086039692 2023-01-21 12:03:09.790722: step: 1216/529, loss: 0.006313800811767578 2023-01-21 12:03:10.900615: step: 1220/529, loss: 0.01786499097943306 2023-01-21 12:03:12.034185: step: 1224/529, loss: 0.0005496978992596269 2023-01-21 12:03:13.148260: step: 1228/529, loss: 0.00012912749662064016 2023-01-21 12:03:14.272565: step: 1232/529, loss: 0.04875793680548668 2023-01-21 12:03:15.379695: step: 1236/529, loss: 0.00183353410102427 2023-01-21 12:03:16.528408: step: 1240/529, loss: 0.0017376424511894584 2023-01-21 12:03:17.622068: step: 1244/529, loss: 0.0004755973641294986 2023-01-21 12:03:18.727313: step: 1248/529, loss: 0.04385976865887642 2023-01-21 12:03:19.849313: step: 1252/529, loss: 5.664825584972277e-05 2023-01-21 12:03:20.959383: step: 1256/529, loss: 0.0031322957947850227 2023-01-21 12:03:22.074109: step: 1260/529, loss: 0.02064957655966282 2023-01-21 12:03:23.172635: step: 1264/529, loss: 0.004171943757683039 2023-01-21 12:03:24.280946: step: 1268/529, loss: 0.022301102057099342 2023-01-21 12:03:25.381356: step: 1272/529, loss: 0.025876808911561966 2023-01-21 12:03:26.496141: step: 1276/529, loss: 0.007781696040183306 2023-01-21 12:03:27.602696: step: 1280/529, loss: 0.01775054819881916 2023-01-21 12:03:28.695413: step: 1284/529, loss: 0.0023900510277599096 2023-01-21 12:03:29.796272: step: 1288/529, loss: 0.027754688635468483 2023-01-21 12:03:30.905261: step: 1292/529, loss: 0.00011917352094314992 2023-01-21 12:03:31.997922: step: 1296/529, loss: 0.0007245064480230212 2023-01-21 12:03:33.112414: step: 1300/529, loss: 0.06863269954919815 2023-01-21 12:03:34.238639: step: 1304/529, loss: 0.013102531433105469 2023-01-21 12:03:35.372770: step: 1308/529, loss: 0.024417974054813385 2023-01-21 12:03:36.473158: step: 1312/529, loss: 7.629396350239404e-07 2023-01-21 12:03:37.579853: step: 1316/529, loss: 3.457069760770537e-05 2023-01-21 12:03:38.698807: step: 1320/529, loss: 0.006758022587746382 2023-01-21 12:03:39.824008: step: 1324/529, loss: 0.041713714599609375 2023-01-21 12:03:40.929374: step: 1328/529, loss: 0.08664874732494354 2023-01-21 12:03:42.016606: step: 1332/529, loss: 0.0007704734453000128 2023-01-21 12:03:43.139115: step: 1336/529, loss: 0.03634205088019371 2023-01-21 12:03:44.257494: step: 1340/529, loss: 0.00029077532235533 2023-01-21 12:03:45.379570: step: 1344/529, loss: 0.0003410339413676411 2023-01-21 12:03:46.470923: step: 1348/529, loss: 0.00039544107858091593 2023-01-21 12:03:47.613131: step: 1352/529, loss: 0.007435822859406471 2023-01-21 12:03:48.757614: step: 1356/529, loss: 0.03292837366461754 2023-01-21 12:03:49.849244: step: 1360/529, loss: 0.021404074504971504 2023-01-21 12:03:50.992792: step: 1364/529, loss: 0.03869705647230148 2023-01-21 12:03:52.120457: step: 1368/529, loss: 0.0014341354835778475 2023-01-21 12:03:53.229480: step: 1372/529, loss: 3.681182715808973e-05 2023-01-21 12:03:54.330769: step: 1376/529, loss: 0.0034673691261559725 2023-01-21 12:03:55.469231: step: 1380/529, loss: 0.0009338856325484812 2023-01-21 12:03:56.593501: step: 1384/529, loss: 0.00010204315185546875 2023-01-21 12:03:57.708379: step: 1388/529, loss: 0.0025687217712402344 2023-01-21 12:03:58.795873: step: 1392/529, loss: 0.000102996826171875 2023-01-21 12:03:59.925295: step: 1396/529, loss: 0.024662017822265625 2023-01-21 12:04:01.066168: step: 1400/529, loss: 0.011120045557618141 2023-01-21 12:04:02.220411: step: 1404/529, loss: 0.0006376743549481034 2023-01-21 12:04:03.340388: step: 1408/529, loss: 0.010180091485381126 2023-01-21 12:04:04.463642: step: 1412/529, loss: 0.0077002523466944695 2023-01-21 12:04:05.574931: step: 1416/529, loss: 0.00026502611581236124 2023-01-21 12:04:06.716791: step: 1420/529, loss: 0.06222133710980415 2023-01-21 12:04:07.838480: step: 1424/529, loss: 0.011392593383789062 2023-01-21 12:04:08.977067: step: 1428/529, loss: 0.041582297533750534 2023-01-21 12:04:10.097982: step: 1432/529, loss: 0.0016268730396404862 2023-01-21 12:04:11.188336: step: 1436/529, loss: 6.399154517566785e-05 2023-01-21 12:04:12.303453: step: 1440/529, loss: 0.00040950774564407766 2023-01-21 12:04:13.413706: step: 1444/529, loss: 0.01871499978005886 2023-01-21 12:04:14.503761: step: 1448/529, loss: 0.01595153845846653 2023-01-21 12:04:15.602159: step: 1452/529, loss: 5.7220458984375e-05 2023-01-21 12:04:16.710586: step: 1456/529, loss: 0.09844265133142471 2023-01-21 12:04:17.790689: step: 1460/529, loss: 0.006976318545639515 2023-01-21 12:04:18.885465: step: 1464/529, loss: 0.011003208346664906 2023-01-21 12:04:19.979355: step: 1468/529, loss: 0.018002891913056374 2023-01-21 12:04:21.125442: step: 1472/529, loss: 0.002467250684276223 2023-01-21 12:04:22.285010: step: 1476/529, loss: 0.02062053792178631 2023-01-21 12:04:23.405536: step: 1480/529, loss: 0.00406570453196764 2023-01-21 12:04:24.530072: step: 1484/529, loss: 6.446838233387098e-05 2023-01-21 12:04:25.649139: step: 1488/529, loss: 9.398460679221898e-05 2023-01-21 12:04:26.791250: step: 1492/529, loss: 0.17100505530834198 2023-01-21 12:04:27.918941: step: 1496/529, loss: 4.563331822282635e-05 2023-01-21 12:04:28.989682: step: 1500/529, loss: 0.0005127906915731728 2023-01-21 12:04:30.091034: step: 1504/529, loss: 0.000396537798224017 2023-01-21 12:04:31.202854: step: 1508/529, loss: 0.0030138969887048006 2023-01-21 12:04:32.311166: step: 1512/529, loss: 0.001436424208804965 2023-01-21 12:04:33.412377: step: 1516/529, loss: 0.00069427490234375 2023-01-21 12:04:34.571610: step: 1520/529, loss: 0.0021524429321289062 2023-01-21 12:04:35.698482: step: 1524/529, loss: 0.0012069703079760075 2023-01-21 12:04:36.810168: step: 1528/529, loss: 0.010901331901550293 2023-01-21 12:04:37.916938: step: 1532/529, loss: 0.002715492155402899 2023-01-21 12:04:39.025408: step: 1536/529, loss: 0.010830498300492764 2023-01-21 12:04:40.128710: step: 1540/529, loss: 0.002168846083804965 2023-01-21 12:04:41.256064: step: 1544/529, loss: 0.0012269974686205387 2023-01-21 12:04:42.394540: step: 1548/529, loss: 0.08148574829101562 2023-01-21 12:04:43.475299: step: 1552/529, loss: 0.0006669998401775956 2023-01-21 12:04:44.603460: step: 1556/529, loss: 0.0008137703407555819 2023-01-21 12:04:45.716628: step: 1560/529, loss: 0.07223119586706161 2023-01-21 12:04:46.824957: step: 1564/529, loss: 0.09631986916065216 2023-01-21 12:04:47.958855: step: 1568/529, loss: 0.009471584111452103 2023-01-21 12:04:49.097348: step: 1572/529, loss: 0.02781658247113228 2023-01-21 12:04:50.236765: step: 1576/529, loss: 1.8787384760798886e-05 2023-01-21 12:04:51.341445: step: 1580/529, loss: 0.0013381957542151213 2023-01-21 12:04:52.440989: step: 1584/529, loss: 0.005512905307114124 2023-01-21 12:04:53.565645: step: 1588/529, loss: 0.003703498747199774 2023-01-21 12:04:54.667333: step: 1592/529, loss: 0.002418899443000555 2023-01-21 12:04:55.839763: step: 1596/529, loss: 0.017078112810850143 2023-01-21 12:04:56.983814: step: 1600/529, loss: 0.006324720103293657 2023-01-21 12:04:58.126077: step: 1604/529, loss: 0.06168804317712784 2023-01-21 12:04:59.246078: step: 1608/529, loss: 0.00040493012056685984 2023-01-21 12:05:00.345753: step: 1612/529, loss: 0.0030885697342455387 2023-01-21 12:05:01.442016: step: 1616/529, loss: 0.005735206883400679 2023-01-21 12:05:02.562904: step: 1620/529, loss: 0.03360319137573242 2023-01-21 12:05:03.662595: step: 1624/529, loss: 0.002379798796027899 2023-01-21 12:05:04.784706: step: 1628/529, loss: 0.006606435868889093 2023-01-21 12:05:05.897482: step: 1632/529, loss: 0.0019009590614587069 2023-01-21 12:05:07.047334: step: 1636/529, loss: 0.0011101722484454513 2023-01-21 12:05:08.144789: step: 1640/529, loss: 0.0010530471336096525 2023-01-21 12:05:09.258317: step: 1644/529, loss: 0.0002922058047261089 2023-01-21 12:05:10.376825: step: 1648/529, loss: 0.0316736213862896 2023-01-21 12:05:11.484084: step: 1652/529, loss: 0.005758475977927446 2023-01-21 12:05:12.612211: step: 1656/529, loss: 0.02818622626364231 2023-01-21 12:05:13.747317: step: 1660/529, loss: 1.716613724056515e-06 2023-01-21 12:05:14.864999: step: 1664/529, loss: 0.00474205007776618 2023-01-21 12:05:15.989995: step: 1668/529, loss: 0.0318845771253109 2023-01-21 12:05:17.109450: step: 1672/529, loss: 0.014821624383330345 2023-01-21 12:05:18.203641: step: 1676/529, loss: 0.0004411697154864669 2023-01-21 12:05:19.340836: step: 1680/529, loss: 0.0005587577470578253 2023-01-21 12:05:20.491704: step: 1684/529, loss: 0.0539977103471756 2023-01-21 12:05:21.611960: step: 1688/529, loss: 0.021155642345547676 2023-01-21 12:05:22.712982: step: 1692/529, loss: 0.027691461145877838 2023-01-21 12:05:23.823404: step: 1696/529, loss: 0.03290477767586708 2023-01-21 12:05:24.935182: step: 1700/529, loss: 3.4332275390625e-05 2023-01-21 12:05:26.053654: step: 1704/529, loss: 0.04723091423511505 2023-01-21 12:05:27.153767: step: 1708/529, loss: -1.0013580322265625e-05 2023-01-21 12:05:28.281211: step: 1712/529, loss: 0.00556185282766819 2023-01-21 12:05:29.415243: step: 1716/529, loss: 0.005481719970703125 2023-01-21 12:05:30.543743: step: 1720/529, loss: 0.0052058217115700245 2023-01-21 12:05:31.716792: step: 1724/529, loss: 0.009868049062788486 2023-01-21 12:05:32.822350: step: 1728/529, loss: 0.000530719815287739 2023-01-21 12:05:33.928217: step: 1732/529, loss: 0.05551014095544815 2023-01-21 12:05:35.032438: step: 1736/529, loss: 0.007211876101791859 2023-01-21 12:05:36.141050: step: 1740/529, loss: 0.028414295986294746 2023-01-21 12:05:37.288768: step: 1744/529, loss: 0.007656860165297985 2023-01-21 12:05:38.396680: step: 1748/529, loss: 0.000996494316495955 2023-01-21 12:05:39.512019: step: 1752/529, loss: 0.006771469488739967 2023-01-21 12:05:40.615566: step: 1756/529, loss: 0.0017566680908203125 2023-01-21 12:05:41.706408: step: 1760/529, loss: 0.0005850791931152344 2023-01-21 12:05:42.832404: step: 1764/529, loss: 0.0010857820743694901 2023-01-21 12:05:43.955483: step: 1768/529, loss: 0.2705098092556 2023-01-21 12:05:45.096515: step: 1772/529, loss: 0.07381203770637512 2023-01-21 12:05:46.231930: step: 1776/529, loss: 0.03545703738927841 2023-01-21 12:05:47.350548: step: 1780/529, loss: 0.10171470791101456 2023-01-21 12:05:48.451533: step: 1784/529, loss: 8.087605237960815e-05 2023-01-21 12:05:49.562555: step: 1788/529, loss: 0.02450408972799778 2023-01-21 12:05:50.666468: step: 1792/529, loss: 0.00019836427236441523 2023-01-21 12:05:51.827622: step: 1796/529, loss: 0.0007505417452193797 2023-01-21 12:05:52.962442: step: 1800/529, loss: 0.004984474275261164 2023-01-21 12:05:54.059020: step: 1804/529, loss: 0.0047702789306640625 2023-01-21 12:05:55.182379: step: 1808/529, loss: 0.014599801041185856 2023-01-21 12:05:56.303570: step: 1812/529, loss: 0.05559387058019638 2023-01-21 12:05:57.411367: step: 1816/529, loss: 0.03633527830243111 2023-01-21 12:05:58.530035: step: 1820/529, loss: 0.004245663061738014 2023-01-21 12:05:59.651059: step: 1824/529, loss: 0.05724906921386719 2023-01-21 12:06:00.772520: step: 1828/529, loss: 0.004823446273803711 2023-01-21 12:06:01.898721: step: 1832/529, loss: 0.026070214807987213 2023-01-21 12:06:03.002522: step: 1836/529, loss: 0.01709003560245037 2023-01-21 12:06:04.149874: step: 1840/529, loss: 0.00035419463529251516 2023-01-21 12:06:05.289078: step: 1844/529, loss: 0.0033585550263524055 2023-01-21 12:06:06.398928: step: 1848/529, loss: 0.03517098352313042 2023-01-21 12:06:07.518063: step: 1852/529, loss: 2.1266938347253017e-05 2023-01-21 12:06:08.652454: step: 1856/529, loss: 0.18844357132911682 2023-01-21 12:06:09.753360: step: 1860/529, loss: 0.010949421674013138 2023-01-21 12:06:10.897860: step: 1864/529, loss: 0.0004792422114405781 2023-01-21 12:06:12.064584: step: 1868/529, loss: 0.07842407375574112 2023-01-21 12:06:13.152384: step: 1872/529, loss: 0.005452823359519243 2023-01-21 12:06:14.249460: step: 1876/529, loss: 0.0003111839178018272 2023-01-21 12:06:15.350054: step: 1880/529, loss: 4.65869925392326e-05 2023-01-21 12:06:16.438672: step: 1884/529, loss: 0.02787027321755886 2023-01-21 12:06:17.562508: step: 1888/529, loss: 0.01788453944027424 2023-01-21 12:06:18.672335: step: 1892/529, loss: 0.4779691696166992 2023-01-21 12:06:19.796727: step: 1896/529, loss: 0.3006353974342346 2023-01-21 12:06:20.909342: step: 1900/529, loss: 0.011092566885054111 2023-01-21 12:06:21.983207: step: 1904/529, loss: 0.0001417681633029133 2023-01-21 12:06:23.105693: step: 1908/529, loss: 0.10302677005529404 2023-01-21 12:06:24.212628: step: 1912/529, loss: 4.94956984766759e-05 2023-01-21 12:06:25.311879: step: 1916/529, loss: 0.039360810071229935 2023-01-21 12:06:26.386300: step: 1920/529, loss: 4.491806248552166e-05 2023-01-21 12:06:27.532184: step: 1924/529, loss: 0.0074977874755859375 2023-01-21 12:06:28.674761: step: 1928/529, loss: 0.010804463177919388 2023-01-21 12:06:29.811306: step: 1932/529, loss: 0.00016790107474662364 2023-01-21 12:06:30.925077: step: 1936/529, loss: 0.012201881036162376 2023-01-21 12:06:32.014919: step: 1940/529, loss: 0.004649448674172163 2023-01-21 12:06:33.104742: step: 1944/529, loss: 0.004570770543068647 2023-01-21 12:06:34.201448: step: 1948/529, loss: 0.0002151489316020161 2023-01-21 12:06:35.288547: step: 1952/529, loss: 0.000701522862073034 2023-01-21 12:06:36.416799: step: 1956/529, loss: 0.015621663071215153 2023-01-21 12:06:37.537826: step: 1960/529, loss: 0.035024262964725494 2023-01-21 12:06:38.658096: step: 1964/529, loss: 0.0008228301885537803 2023-01-21 12:06:39.781450: step: 1968/529, loss: 0.038230132311582565 2023-01-21 12:06:40.891395: step: 1972/529, loss: 0.003978633787482977 2023-01-21 12:06:42.036291: step: 1976/529, loss: 0.004439544398337603 2023-01-21 12:06:43.146919: step: 1980/529, loss: 0.05241961404681206 2023-01-21 12:06:44.273772: step: 1984/529, loss: 0.03841552883386612 2023-01-21 12:06:45.365062: step: 1988/529, loss: 0.0017431258456781507 2023-01-21 12:06:46.486917: step: 1992/529, loss: 0.0002325058012502268 2023-01-21 12:06:47.607861: step: 1996/529, loss: 0.008177090436220169 2023-01-21 12:06:48.725936: step: 2000/529, loss: 9.965896606445312e-05 2023-01-21 12:06:49.840490: step: 2004/529, loss: 0.0190550796687603 2023-01-21 12:06:50.936473: step: 2008/529, loss: 0.022170066833496094 2023-01-21 12:06:52.091829: step: 2012/529, loss: 0.015341853722929955 2023-01-21 12:06:53.228275: step: 2016/529, loss: 0.11924886703491211 2023-01-21 12:06:54.364181: step: 2020/529, loss: 0.013314819894731045 2023-01-21 12:06:55.507268: step: 2024/529, loss: 0.0006289482116699219 2023-01-21 12:06:56.646046: step: 2028/529, loss: 0.0013704299926757812 2023-01-21 12:06:57.774521: step: 2032/529, loss: 0.02000293880701065 2023-01-21 12:06:58.878234: step: 2036/529, loss: 0.07647235691547394 2023-01-21 12:06:59.992556: step: 2040/529, loss: 0.0012081146705895662 2023-01-21 12:07:01.092727: step: 2044/529, loss: 0.006828213110566139 2023-01-21 12:07:02.200197: step: 2048/529, loss: 0.0034330368507653475 2023-01-21 12:07:03.319380: step: 2052/529, loss: 0.003417587373405695 2023-01-21 12:07:04.447709: step: 2056/529, loss: 0.020191097632050514 2023-01-21 12:07:05.569901: step: 2060/529, loss: 0.03416576609015465 2023-01-21 12:07:06.698346: step: 2064/529, loss: 0.02537689171731472 2023-01-21 12:07:07.777601: step: 2068/529, loss: 0.0013364792102947831 2023-01-21 12:07:08.885282: step: 2072/529, loss: 0.0010828971862792969 2023-01-21 12:07:10.030795: step: 2076/529, loss: 0.0036756517365574837 2023-01-21 12:07:11.157487: step: 2080/529, loss: 0.007564115803688765 2023-01-21 12:07:12.254085: step: 2084/529, loss: 0.06037912145256996 2023-01-21 12:07:13.358393: step: 2088/529, loss: 0.009715653024613857 2023-01-21 12:07:14.482356: step: 2092/529, loss: 0.006321525666862726 2023-01-21 12:07:15.573938: step: 2096/529, loss: 0.044431500136852264 2023-01-21 12:07:16.689143: step: 2100/529, loss: 0.037435341626405716 2023-01-21 12:07:17.778819: step: 2104/529, loss: 0.00018548965454101562 2023-01-21 12:07:18.873133: step: 2108/529, loss: 0.011276436038315296 2023-01-21 12:07:19.974737: step: 2112/529, loss: 0.004372215364128351 2023-01-21 12:07:21.080582: step: 2116/529, loss: 0.05443563312292099 ================================================== Loss: 0.022 -------------------- Dev: {'event': {'p': 0.6067653276955602, 'r': 0.7643142476697736, 'f1': 0.6764879198585739}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Test: {'event': {'p': 0.6389806512505899, 'r': 0.7641083521444695, 'f1': 0.695965047545618}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} 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': 20} Korean: {'event': {'p': 0.6666666666666666, 'r': 0.5714285714285714, 'f1': 0.6153846153846153}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Russian: {'event': {'p': 0.425, 'r': 0.4722222222222222, 'f1': 0.4473684210526316}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 21 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:08:01.294527: step: 4/529, loss: 2.613067590573337e-05 2023-01-21 12:08:02.397485: step: 8/529, loss: 0.021919824182987213 2023-01-21 12:08:03.520757: step: 12/529, loss: 0.08860154449939728 2023-01-21 12:08:04.623481: step: 16/529, loss: 0.015587473288178444 2023-01-21 12:08:05.806124: step: 20/529, loss: 0.00036115647526457906 2023-01-21 12:08:06.931265: step: 24/529, loss: 0.025742148980498314 2023-01-21 12:08:08.055038: step: 28/529, loss: 0.00487442035228014 2023-01-21 12:08:09.179615: step: 32/529, loss: 0.002731320448219776 2023-01-21 12:08:10.305454: step: 36/529, loss: 0.01446390151977539 2023-01-21 12:08:11.406690: step: 40/529, loss: 0.004962826147675514 2023-01-21 12:08:12.527427: step: 44/529, loss: 0.00042304996168240905 2023-01-21 12:08:13.642357: step: 48/529, loss: 0.0037470818497240543 2023-01-21 12:08:14.774541: step: 52/529, loss: 0.012492799200117588 2023-01-21 12:08:15.910957: step: 56/529, loss: 0.0003494262637104839 2023-01-21 12:08:17.016906: step: 60/529, loss: 0.00018911361985374242 2023-01-21 12:08:18.132919: step: 64/529, loss: 0.00023968219466041774 2023-01-21 12:08:19.243306: step: 68/529, loss: 0.9430027604103088 2023-01-21 12:08:20.392839: step: 72/529, loss: 0.00154962542001158 2023-01-21 12:08:21.545094: step: 76/529, loss: 0.0009286046260967851 2023-01-21 12:08:22.677809: step: 80/529, loss: 0.0041918279603123665 2023-01-21 12:08:23.786299: step: 84/529, loss: 0.0048690796829760075 2023-01-21 12:08:24.884967: step: 88/529, loss: 0.00898208562284708 2023-01-21 12:08:26.010142: step: 92/529, loss: 0.0012594700092449784 2023-01-21 12:08:27.096381: step: 96/529, loss: 0.0014060973189771175 2023-01-21 12:08:28.222184: step: 100/529, loss: 0.0011108398903161287 2023-01-21 12:08:29.325822: step: 104/529, loss: 0.001168966293334961 2023-01-21 12:08:30.463729: step: 108/529, loss: 0.005302238743752241 2023-01-21 12:08:31.565279: step: 112/529, loss: 0.020262623205780983 2023-01-21 12:08:32.656847: step: 116/529, loss: 0.015787124633789062 2023-01-21 12:08:33.779075: step: 120/529, loss: 0.019242573529481888 2023-01-21 12:08:34.915872: step: 124/529, loss: 0.03191409260034561 2023-01-21 12:08:36.015546: step: 128/529, loss: 0.0023286822251975536 2023-01-21 12:08:37.084364: step: 132/529, loss: 3.528594970703125e-05 2023-01-21 12:08:38.169684: step: 136/529, loss: 0.00017943383136298507 2023-01-21 12:08:39.292273: step: 140/529, loss: 0.0031764984596520662 2023-01-21 12:08:40.396084: step: 144/529, loss: 6.12258882028982e-05 2023-01-21 12:08:41.531661: step: 148/529, loss: 0.0012099266750738025 2023-01-21 12:08:42.655051: step: 152/529, loss: 0.0002994537353515625 2023-01-21 12:08:43.783845: step: 156/529, loss: 0.005751896183937788 2023-01-21 12:08:44.869416: step: 160/529, loss: 0.0032262802124023438 2023-01-21 12:08:46.005272: step: 164/529, loss: 0.08008003234863281 2023-01-21 12:08:47.107691: step: 168/529, loss: 0.00014925003051757812 2023-01-21 12:08:48.249734: step: 172/529, loss: 0.0010047913528978825 2023-01-21 12:08:49.354230: step: 176/529, loss: 0.044236376881599426 2023-01-21 12:08:50.467647: step: 180/529, loss: 0.00016288757615257055 2023-01-21 12:08:51.594969: step: 184/529, loss: 0.0011455535423010588 2023-01-21 12:08:52.682258: step: 188/529, loss: 0.0034526349045336246 2023-01-21 12:08:53.781233: step: 192/529, loss: 0.0437747985124588 2023-01-21 12:08:54.929512: step: 196/529, loss: 0.002347278641536832 2023-01-21 12:08:56.080909: step: 200/529, loss: 0.022580433636903763 2023-01-21 12:08:57.202567: step: 204/529, loss: 0.007929230108857155 2023-01-21 12:08:58.321067: step: 208/529, loss: 0.06851902604103088 2023-01-21 12:08:59.446577: step: 212/529, loss: 0.005290794186294079 2023-01-21 12:09:00.549361: step: 216/529, loss: 0.0003197670157533139 2023-01-21 12:09:01.678150: step: 220/529, loss: 0.002115535782650113 2023-01-21 12:09:02.792909: step: 224/529, loss: 0.0036491393111646175 2023-01-21 12:09:03.932207: step: 228/529, loss: 0.0006009340286254883 2023-01-21 12:09:05.040895: step: 232/529, loss: 0.2991218566894531 2023-01-21 12:09:06.166430: step: 236/529, loss: 0.02958526648581028 2023-01-21 12:09:07.269503: step: 240/529, loss: 0.00078496930655092 2023-01-21 12:09:08.400030: step: 244/529, loss: 0.14212216436862946 2023-01-21 12:09:09.538442: step: 248/529, loss: 0.007867050357162952 2023-01-21 12:09:10.647980: step: 252/529, loss: 0.0003217697376385331 2023-01-21 12:09:11.746763: step: 256/529, loss: 0.014172124676406384 2023-01-21 12:09:12.869760: step: 260/529, loss: 9.493828110862523e-05 2023-01-21 12:09:13.996136: step: 264/529, loss: 0.0013633727794513106 2023-01-21 12:09:15.084481: step: 268/529, loss: 0.006245422642678022 2023-01-21 12:09:16.185849: step: 272/529, loss: 0.0030531883239746094 2023-01-21 12:09:17.316888: step: 276/529, loss: 0.11727866530418396 2023-01-21 12:09:18.443054: step: 280/529, loss: 0.0007067203987389803 2023-01-21 12:09:19.536777: step: 284/529, loss: 0.09139136970043182 2023-01-21 12:09:20.676658: step: 288/529, loss: 0.019987013190984726 2023-01-21 12:09:21.782197: step: 292/529, loss: 0.1840529441833496 2023-01-21 12:09:22.899637: step: 296/529, loss: 0.004201030824333429 2023-01-21 12:09:24.013892: step: 300/529, loss: 0.00919113215059042 2023-01-21 12:09:25.133891: step: 304/529, loss: 0.0030369760934263468 2023-01-21 12:09:26.248941: step: 308/529, loss: 0.0007394790882244706 2023-01-21 12:09:27.386135: step: 312/529, loss: 0.0018394470680505037 2023-01-21 12:09:28.547623: step: 316/529, loss: 0.03560009226202965 2023-01-21 12:09:29.652573: step: 320/529, loss: 0.0011362076038494706 2023-01-21 12:09:30.786734: step: 324/529, loss: 0.029270555824041367 2023-01-21 12:09:31.887369: step: 328/529, loss: 0.022177983075380325 2023-01-21 12:09:32.997111: step: 332/529, loss: 0.02502279356122017 2023-01-21 12:09:34.130838: step: 336/529, loss: 0.004890823271125555 2023-01-21 12:09:35.233388: step: 340/529, loss: 0.0002095222589559853 2023-01-21 12:09:36.345994: step: 344/529, loss: 0.003365421202033758 2023-01-21 12:09:37.497537: step: 348/529, loss: 0.011507987976074219 2023-01-21 12:09:38.629041: step: 352/529, loss: 0.0010006905067712069 2023-01-21 12:09:39.752620: step: 356/529, loss: 0.0006176948663778603 2023-01-21 12:09:40.864773: step: 360/529, loss: 0.0005784988170489669 2023-01-21 12:09:41.964484: step: 364/529, loss: 0.004149246029555798 2023-01-21 12:09:43.071016: step: 368/529, loss: 0.0002172470121877268 2023-01-21 12:09:44.187222: step: 372/529, loss: 0.01653308980166912 2023-01-21 12:09:45.303247: step: 376/529, loss: 4.919171624351293e-05 2023-01-21 12:09:46.424493: step: 380/529, loss: 0.09673252701759338 2023-01-21 12:09:47.523668: step: 384/529, loss: 0.0005697250016964972 2023-01-21 12:09:48.608386: step: 388/529, loss: 0.035703565925359726 2023-01-21 12:09:49.740559: step: 392/529, loss: 0.00445103645324707 2023-01-21 12:09:50.851133: step: 396/529, loss: 0.0008612633100710809 2023-01-21 12:09:51.949806: step: 400/529, loss: 0.0566289909183979 2023-01-21 12:09:53.075742: step: 404/529, loss: 0.004525851923972368 2023-01-21 12:09:54.163882: step: 408/529, loss: 6.67572021484375e-06 2023-01-21 12:09:55.265056: step: 412/529, loss: 0.0006334304925985634 2023-01-21 12:09:56.368747: step: 416/529, loss: 0.00047788620577193797 2023-01-21 12:09:57.471616: step: 420/529, loss: 0.0042795659974217415 2023-01-21 12:09:58.582634: step: 424/529, loss: 0.005630016326904297 2023-01-21 12:09:59.669384: step: 428/529, loss: 0.016614211723208427 2023-01-21 12:10:00.801429: step: 432/529, loss: 0.005085754673928022 2023-01-21 12:10:01.913759: step: 436/529, loss: 0.004018020816147327 2023-01-21 12:10:03.033029: step: 440/529, loss: 0.0323827750980854 2023-01-21 12:10:04.135571: step: 444/529, loss: 0.00857534445822239 2023-01-21 12:10:05.251440: step: 448/529, loss: 3.929138256353326e-05 2023-01-21 12:10:06.376443: step: 452/529, loss: 0.03530597314238548 2023-01-21 12:10:07.490084: step: 456/529, loss: 0.01349721010774374 2023-01-21 12:10:08.616101: step: 460/529, loss: 0.0017461777897551656 2023-01-21 12:10:09.752964: step: 464/529, loss: 0.0006258010980673134 2023-01-21 12:10:10.874076: step: 468/529, loss: 0.025965308770537376 2023-01-21 12:10:11.977582: step: 472/529, loss: 0.0011900425888597965 2023-01-21 12:10:13.036834: step: 476/529, loss: 5.846023850608617e-05 2023-01-21 12:10:14.138130: step: 480/529, loss: 0.002334499265998602 2023-01-21 12:10:15.256262: step: 484/529, loss: 0.001425069523975253 2023-01-21 12:10:16.385417: step: 488/529, loss: 0.010773325338959694 2023-01-21 12:10:17.503125: step: 492/529, loss: 0.004265308380126953 2023-01-21 12:10:18.618256: step: 496/529, loss: 0.0007693290826864541 2023-01-21 12:10:19.703602: step: 500/529, loss: 0.002647972200065851 2023-01-21 12:10:20.813854: step: 504/529, loss: 0.002762031741440296 2023-01-21 12:10:21.921590: step: 508/529, loss: 0.001010894775390625 2023-01-21 12:10:23.025237: step: 512/529, loss: 0.028596973046660423 2023-01-21 12:10:24.174359: step: 516/529, loss: 0.0014122009743005037 2023-01-21 12:10:25.283578: step: 520/529, loss: 0.00185222621075809 2023-01-21 12:10:26.397225: step: 524/529, loss: 0.0004059791681356728 2023-01-21 12:10:27.520410: step: 528/529, loss: 0.0020330429542809725 2023-01-21 12:10:28.626188: step: 532/529, loss: 0.000370025634765625 2023-01-21 12:10:29.725152: step: 536/529, loss: 0.012719536200165749 2023-01-21 12:10:30.866989: step: 540/529, loss: 0.03291015699505806 2023-01-21 12:10:31.986707: step: 544/529, loss: 0.049634553492069244 2023-01-21 12:10:33.129946: step: 548/529, loss: 0.012201976962387562 2023-01-21 12:10:34.234756: step: 552/529, loss: 0.017602063715457916 2023-01-21 12:10:35.324151: step: 556/529, loss: 0.02855362929403782 2023-01-21 12:10:36.441528: step: 560/529, loss: 0.0002857208310160786 2023-01-21 12:10:37.530733: step: 564/529, loss: 0.003656578017398715 2023-01-21 12:10:38.652429: step: 568/529, loss: 0.017104625701904297 2023-01-21 12:10:39.764919: step: 572/529, loss: 6.027221752447076e-05 2023-01-21 12:10:40.912701: step: 576/529, loss: 0.007122325710952282 2023-01-21 12:10:42.005298: step: 580/529, loss: 0.001768398331478238 2023-01-21 12:10:43.106843: step: 584/529, loss: 0.003937148954719305 2023-01-21 12:10:44.230416: step: 588/529, loss: 0.005389594938606024 2023-01-21 12:10:45.335837: step: 592/529, loss: 0.012478924356400967 2023-01-21 12:10:46.460816: step: 596/529, loss: 0.02545471116900444 2023-01-21 12:10:47.566403: step: 600/529, loss: 0.06955185532569885 2023-01-21 12:10:48.706611: step: 604/529, loss: 0.011044503189623356 2023-01-21 12:10:49.854814: step: 608/529, loss: 0.007721519563347101 2023-01-21 12:10:50.988716: step: 612/529, loss: 0.0008298874017782509 2023-01-21 12:10:52.115389: step: 616/529, loss: 0.020178042352199554 2023-01-21 12:10:53.224469: step: 620/529, loss: 0.011399651877582073 2023-01-21 12:10:54.351856: step: 624/529, loss: 0.046204518526792526 2023-01-21 12:10:55.445459: step: 628/529, loss: 0.0010167121654376388 2023-01-21 12:10:56.574693: step: 632/529, loss: 0.00149116525426507 2023-01-21 12:10:57.721567: step: 636/529, loss: 0.026746368035674095 2023-01-21 12:10:58.823881: step: 640/529, loss: 0.020307207480072975 2023-01-21 12:10:59.964712: step: 644/529, loss: 0.020549917593598366 2023-01-21 12:11:01.078740: step: 648/529, loss: 1.316070574830519e-05 2023-01-21 12:11:02.182231: step: 652/529, loss: 0.0017923355335369706 2023-01-21 12:11:03.291962: step: 656/529, loss: 0.009404277428984642 2023-01-21 12:11:04.383726: step: 660/529, loss: 0.00814600009471178 2023-01-21 12:11:05.480055: step: 664/529, loss: 0.0017849921714514494 2023-01-21 12:11:06.590460: step: 668/529, loss: 0.005227899178862572 2023-01-21 12:11:07.721959: step: 672/529, loss: 0.024492979049682617 2023-01-21 12:11:08.842308: step: 676/529, loss: 1.773834264895413e-05 2023-01-21 12:11:09.960285: step: 680/529, loss: 0.007250881288200617 2023-01-21 12:11:11.056250: step: 684/529, loss: 0.0010100365616381168 2023-01-21 12:11:12.187477: step: 688/529, loss: 0.11091961711645126 2023-01-21 12:11:13.302109: step: 692/529, loss: 0.0006041526794433594 2023-01-21 12:11:14.395352: step: 696/529, loss: 0.003327846759930253 2023-01-21 12:11:15.494250: step: 700/529, loss: 0.0003040313604287803 2023-01-21 12:11:16.616078: step: 704/529, loss: 0.0030141829047352076 2023-01-21 12:11:17.740895: step: 708/529, loss: 0.000522232090588659 2023-01-21 12:11:18.865731: step: 712/529, loss: 0.0003856658877339214 2023-01-21 12:11:19.971405: step: 716/529, loss: 0.0007453918224200606 2023-01-21 12:11:21.101779: step: 720/529, loss: 0.06306882202625275 2023-01-21 12:11:22.208662: step: 724/529, loss: 0.006275367923080921 2023-01-21 12:11:23.335503: step: 728/529, loss: 0.00041105749551206827 2023-01-21 12:11:24.438532: step: 732/529, loss: 0.07748498767614365 2023-01-21 12:11:25.531707: step: 736/529, loss: 0.026653816923499107 2023-01-21 12:11:26.666382: step: 740/529, loss: 0.0443546287715435 2023-01-21 12:11:27.839053: step: 744/529, loss: 0.004307555966079235 2023-01-21 12:11:28.941647: step: 748/529, loss: 0.00023136139498092234 2023-01-21 12:11:30.060467: step: 752/529, loss: 0.001127052353695035 2023-01-21 12:11:31.175240: step: 756/529, loss: 0.0012309551239013672 2023-01-21 12:11:32.293813: step: 760/529, loss: 0.0002533912775106728 2023-01-21 12:11:33.402633: step: 764/529, loss: 0.0018871307838708162 2023-01-21 12:11:34.517730: step: 768/529, loss: 0.0025529861450195312 2023-01-21 12:11:35.634426: step: 772/529, loss: 0.0025209428276866674 2023-01-21 12:11:36.750215: step: 776/529, loss: 0.0029299496673047543 2023-01-21 12:11:37.887727: step: 780/529, loss: 1.2588501704158261e-05 2023-01-21 12:11:39.025405: step: 784/529, loss: 0.013435936532914639 2023-01-21 12:11:40.156402: step: 788/529, loss: 0.029555892571806908 2023-01-21 12:11:41.253207: step: 792/529, loss: 0.002090024994686246 2023-01-21 12:11:42.393941: step: 796/529, loss: 0.0005940914270468056 2023-01-21 12:11:43.525543: step: 800/529, loss: 0.004150867462158203 2023-01-21 12:11:44.652763: step: 804/529, loss: 0.008687401190400124 2023-01-21 12:11:45.783097: step: 808/529, loss: 0.007641410920768976 2023-01-21 12:11:46.890786: step: 812/529, loss: 0.0060357097536325455 2023-01-21 12:11:47.990618: step: 816/529, loss: 0.00730476388707757 2023-01-21 12:11:49.145273: step: 820/529, loss: 0.000347137451171875 2023-01-21 12:11:50.296654: step: 824/529, loss: 0.04311399161815643 2023-01-21 12:11:51.418713: step: 828/529, loss: 0.009614849463105202 2023-01-21 12:11:52.519258: step: 832/529, loss: 0.0257905013859272 2023-01-21 12:11:53.664784: step: 836/529, loss: 0.011482619680464268 2023-01-21 12:11:54.769360: step: 840/529, loss: 0.0001369476376567036 2023-01-21 12:11:55.931201: step: 844/529, loss: 0.0016026496887207031 2023-01-21 12:11:57.041589: step: 848/529, loss: 9.312629845226184e-05 2023-01-21 12:11:58.128100: step: 852/529, loss: 0.02016162872314453 2023-01-21 12:11:59.254618: step: 856/529, loss: 0.05476274713873863 2023-01-21 12:12:00.392768: step: 860/529, loss: 2.937316821771674e-05 2023-01-21 12:12:01.487483: step: 864/529, loss: 0.00026493071345612407 2023-01-21 12:12:02.641537: step: 868/529, loss: 0.010068893432617188 2023-01-21 12:12:03.755385: step: 872/529, loss: 0.00020074844360351562 2023-01-21 12:12:04.881197: step: 876/529, loss: 0.00027675629826262593 2023-01-21 12:12:05.975932: step: 880/529, loss: 0.014531517401337624 2023-01-21 12:12:07.109069: step: 884/529, loss: 0.0014039992820471525 2023-01-21 12:12:08.241464: step: 888/529, loss: 0.011753464117646217 2023-01-21 12:12:09.345257: step: 892/529, loss: 0.0047266483306884766 2023-01-21 12:12:10.439409: step: 896/529, loss: 0.0032814026344567537 2023-01-21 12:12:11.563846: step: 900/529, loss: 0.08225230872631073 2023-01-21 12:12:12.656972: step: 904/529, loss: 0.00025501250638626516 2023-01-21 12:12:13.750964: step: 908/529, loss: 0.00024185179790947586 2023-01-21 12:12:14.884595: step: 912/529, loss: 0.10555868595838547 2023-01-21 12:12:15.981783: step: 916/529, loss: 4.682540748035535e-05 2023-01-21 12:12:17.084796: step: 920/529, loss: 0.006677818484604359 2023-01-21 12:12:18.190434: step: 924/529, loss: 0.00028543471125885844 2023-01-21 12:12:19.315741: step: 928/529, loss: 0.0066127777099609375 2023-01-21 12:12:20.445150: step: 932/529, loss: 0.0003414154052734375 2023-01-21 12:12:21.552999: step: 936/529, loss: 0.00147504813503474 2023-01-21 12:12:22.625852: step: 940/529, loss: 0.0008194923866540194 2023-01-21 12:12:23.730126: step: 944/529, loss: 0.006486701779067516 2023-01-21 12:12:24.845938: step: 948/529, loss: 0.016590215265750885 2023-01-21 12:12:25.964774: step: 952/529, loss: 1.016620397567749 2023-01-21 12:12:27.138840: step: 956/529, loss: 0.006287908647209406 2023-01-21 12:12:28.266161: step: 960/529, loss: 0.060610584914684296 2023-01-21 12:12:29.378600: step: 964/529, loss: 0.0005306244129315019 2023-01-21 12:12:30.531985: step: 968/529, loss: 0.04360991716384888 2023-01-21 12:12:31.642877: step: 972/529, loss: 0.009817409329116344 2023-01-21 12:12:32.768635: step: 976/529, loss: 3.9768219721736386e-05 2023-01-21 12:12:33.915407: step: 980/529, loss: 0.003338718321174383 2023-01-21 12:12:35.023062: step: 984/529, loss: 0.010203361511230469 2023-01-21 12:12:36.142333: step: 988/529, loss: 0.0001313209650106728 2023-01-21 12:12:37.310655: step: 992/529, loss: 0.0002304077206645161 2023-01-21 12:12:38.447117: step: 996/529, loss: 0.06900434195995331 2023-01-21 12:12:39.609803: step: 1000/529, loss: 0.027210809290409088 2023-01-21 12:12:40.735751: step: 1004/529, loss: 0.12073516845703125 2023-01-21 12:12:41.863793: step: 1008/529, loss: 0.013430643826723099 2023-01-21 12:12:42.988627: step: 1012/529, loss: 0.0010364532936364412 2023-01-21 12:12:44.091113: step: 1016/529, loss: 0.006598282139748335 2023-01-21 12:12:45.221198: step: 1020/529, loss: 0.005657768342643976 2023-01-21 12:12:46.365939: step: 1024/529, loss: 0.0011806488037109375 2023-01-21 12:12:47.479970: step: 1028/529, loss: 0.0003438472922425717 2023-01-21 12:12:48.613552: step: 1032/529, loss: 0.004993343725800514 2023-01-21 12:12:49.729921: step: 1036/529, loss: 0.003002262208610773 2023-01-21 12:12:50.892006: step: 1040/529, loss: 0.014216137118637562 2023-01-21 12:12:52.003224: step: 1044/529, loss: 0.08601570129394531 2023-01-21 12:12:53.121531: step: 1048/529, loss: 0.0009085655328817666 2023-01-21 12:12:54.239810: step: 1052/529, loss: 0.00025286676827818155 2023-01-21 12:12:55.394301: step: 1056/529, loss: 0.0013441084884107113 2023-01-21 12:12:56.539520: step: 1060/529, loss: 0.008284187875688076 2023-01-21 12:12:57.674122: step: 1064/529, loss: 0.008956098929047585 2023-01-21 12:12:58.767824: step: 1068/529, loss: 0.0033546448685228825 2023-01-21 12:12:59.858689: step: 1072/529, loss: 0.0009879589779302478 2023-01-21 12:13:00.962308: step: 1076/529, loss: 0.0033112524542957544 2023-01-21 12:13:02.065466: step: 1080/529, loss: 0.02704768255352974 2023-01-21 12:13:03.165183: step: 1084/529, loss: 0.01084604300558567 2023-01-21 12:13:04.273493: step: 1088/529, loss: 0.0033391951583325863 2023-01-21 12:13:05.364834: step: 1092/529, loss: 0.005550289060920477 2023-01-21 12:13:06.511804: step: 1096/529, loss: 0.00013618469529319555 2023-01-21 12:13:07.641633: step: 1100/529, loss: 0.00012168884859420359 2023-01-21 12:13:08.772440: step: 1104/529, loss: 0.01589522324502468 2023-01-21 12:13:09.914551: step: 1108/529, loss: 0.044370412826538086 2023-01-21 12:13:11.013676: step: 1112/529, loss: 0.008757210336625576 2023-01-21 12:13:12.137736: step: 1116/529, loss: 0.00046672820462845266 2023-01-21 12:13:13.268897: step: 1120/529, loss: 0.013955128379166126 2023-01-21 12:13:14.382173: step: 1124/529, loss: 0.004243373870849609 2023-01-21 12:13:15.509273: step: 1128/529, loss: 0.04430122300982475 2023-01-21 12:13:16.659450: step: 1132/529, loss: 0.008907509967684746 2023-01-21 12:13:17.771695: step: 1136/529, loss: 0.0011700630420818925 2023-01-21 12:13:18.882074: step: 1140/529, loss: 6.999969627941027e-05 2023-01-21 12:13:20.002448: step: 1144/529, loss: 0.009433365426957607 2023-01-21 12:13:21.116443: step: 1148/529, loss: 0.04876670613884926 2023-01-21 12:13:22.255607: step: 1152/529, loss: 0.03381934389472008 2023-01-21 12:13:23.384266: step: 1156/529, loss: 3.5858156479662284e-05 2023-01-21 12:13:24.481236: step: 1160/529, loss: 0.008650779724121094 2023-01-21 12:13:25.589422: step: 1164/529, loss: 0.0090477941557765 2023-01-21 12:13:26.707646: step: 1168/529, loss: 0.009417342953383923 2023-01-21 12:13:27.819264: step: 1172/529, loss: 0.0002548217889852822 2023-01-21 12:13:28.936935: step: 1176/529, loss: 0.0003794670046772808 2023-01-21 12:13:30.057014: step: 1180/529, loss: 0.005774688441306353 2023-01-21 12:13:31.156773: step: 1184/529, loss: 8.430481102550402e-05 2023-01-21 12:13:32.258193: step: 1188/529, loss: 0.02354869805276394 2023-01-21 12:13:33.363916: step: 1192/529, loss: 0.0048355101607739925 2023-01-21 12:13:34.506196: step: 1196/529, loss: 0.0001371383696096018 2023-01-21 12:13:35.645136: step: 1200/529, loss: 0.027585793286561966 2023-01-21 12:13:36.750720: step: 1204/529, loss: 0.05016765743494034 2023-01-21 12:13:37.881135: step: 1208/529, loss: 0.0015378951793536544 2023-01-21 12:13:38.984077: step: 1212/529, loss: 0.0613313689827919 2023-01-21 12:13:40.096483: step: 1216/529, loss: 0.0005920410621911287 2023-01-21 12:13:41.213049: step: 1220/529, loss: 0.00033845900907181203 2023-01-21 12:13:42.306248: step: 1224/529, loss: 0.001554298447445035 2023-01-21 12:13:43.431383: step: 1228/529, loss: 0.0038946152199059725 2023-01-21 12:13:44.545847: step: 1232/529, loss: 0.030986785888671875 2023-01-21 12:13:45.709349: step: 1236/529, loss: 0.011160088703036308 2023-01-21 12:13:46.823309: step: 1240/529, loss: 0.04942665249109268 2023-01-21 12:13:47.952421: step: 1244/529, loss: 0.0011415063636377454 2023-01-21 12:13:49.093097: step: 1248/529, loss: 0.007232284639030695 2023-01-21 12:13:50.184269: step: 1252/529, loss: 0.02969837188720703 2023-01-21 12:13:51.288177: step: 1256/529, loss: 0.03896084055304527 2023-01-21 12:13:52.397160: step: 1260/529, loss: 0.001159071922302246 2023-01-21 12:13:53.530623: step: 1264/529, loss: 0.011816549114882946 2023-01-21 12:13:54.624794: step: 1268/529, loss: 0.001774597098119557 2023-01-21 12:13:55.765773: step: 1272/529, loss: 0.00081806187517941 2023-01-21 12:13:56.883056: step: 1276/529, loss: 0.0014037609798833728 2023-01-21 12:13:58.035835: step: 1280/529, loss: 0.06323757022619247 2023-01-21 12:13:59.169714: step: 1284/529, loss: 0.05183591693639755 2023-01-21 12:14:00.250478: step: 1288/529, loss: 0.0024730684235692024 2023-01-21 12:14:01.357676: step: 1292/529, loss: 0.004104614723473787 2023-01-21 12:14:02.465092: step: 1296/529, loss: 0.014987755566835403 2023-01-21 12:14:03.587753: step: 1300/529, loss: 0.0004473685985431075 2023-01-21 12:14:04.730586: step: 1304/529, loss: 0.007988072000443935 2023-01-21 12:14:05.836769: step: 1308/529, loss: 0.013030433095991611 2023-01-21 12:14:06.935943: step: 1312/529, loss: 0.00043201446533203125 2023-01-21 12:14:08.014505: step: 1316/529, loss: 0.0005203246837481856 2023-01-21 12:14:09.189449: step: 1320/529, loss: 0.0009806633461266756 2023-01-21 12:14:10.290311: step: 1324/529, loss: 0.0018163680797442794 2023-01-21 12:14:11.391193: step: 1328/529, loss: 0.0016283035511150956 2023-01-21 12:14:12.515123: step: 1332/529, loss: 0.0352325439453125 2023-01-21 12:14:13.629881: step: 1336/529, loss: 0.005730247590690851 2023-01-21 12:14:14.720970: step: 1340/529, loss: 0.020990945398807526 2023-01-21 12:14:15.864061: step: 1344/529, loss: 0.013221168890595436 2023-01-21 12:14:16.970713: step: 1348/529, loss: 0.011582947336137295 2023-01-21 12:14:18.056182: step: 1352/529, loss: 3.147125244140625e-05 2023-01-21 12:14:19.144626: step: 1356/529, loss: 0.0009323119884356856 2023-01-21 12:14:20.268094: step: 1360/529, loss: 0.0051641943864524364 2023-01-21 12:14:21.389638: step: 1364/529, loss: 0.004590797703713179 2023-01-21 12:14:22.507086: step: 1368/529, loss: 0.019164467230439186 2023-01-21 12:14:23.610787: step: 1372/529, loss: 3.621578071033582e-05 2023-01-21 12:14:24.722387: step: 1376/529, loss: 0.007919264025986195 2023-01-21 12:14:25.812301: step: 1380/529, loss: 0.004098796751350164 2023-01-21 12:14:26.935711: step: 1384/529, loss: 0.0010932921431958675 2023-01-21 12:14:28.049593: step: 1388/529, loss: 0.002154398011043668 2023-01-21 12:14:29.191848: step: 1392/529, loss: 0.0005236626020632684 2023-01-21 12:14:30.291312: step: 1396/529, loss: 0.43070220947265625 2023-01-21 12:14:31.413637: step: 1400/529, loss: 0.022328950464725494 2023-01-21 12:14:32.536080: step: 1404/529, loss: 0.014774608425796032 2023-01-21 12:14:33.664908: step: 1408/529, loss: 0.00046024323091842234 2023-01-21 12:14:34.781143: step: 1412/529, loss: 0.00028676987858489156 2023-01-21 12:14:35.899402: step: 1416/529, loss: 0.005784797482192516 2023-01-21 12:14:37.000412: step: 1420/529, loss: 0.009596442803740501 2023-01-21 12:14:38.125882: step: 1424/529, loss: 0.05635375902056694 2023-01-21 12:14:39.221295: step: 1428/529, loss: 4.01496872655116e-05 2023-01-21 12:14:40.305450: step: 1432/529, loss: 0.004881381988525391 2023-01-21 12:14:41.413889: step: 1436/529, loss: 3.223419116693549e-05 2023-01-21 12:14:42.515355: step: 1440/529, loss: 0.0004536628839559853 2023-01-21 12:14:43.652777: step: 1444/529, loss: 0.21957054734230042 2023-01-21 12:14:44.760774: step: 1448/529, loss: 0.05635872110724449 2023-01-21 12:14:45.882452: step: 1452/529, loss: 0.001562404679134488 2023-01-21 12:14:47.010118: step: 1456/529, loss: 0.010155296884477139 2023-01-21 12:14:48.129135: step: 1460/529, loss: 0.00622406043112278 2023-01-21 12:14:49.237475: step: 1464/529, loss: 0.017214488238096237 2023-01-21 12:14:50.378554: step: 1468/529, loss: 0.010006332769989967 2023-01-21 12:14:51.505065: step: 1472/529, loss: 0.0018203735817223787 2023-01-21 12:14:52.614814: step: 1476/529, loss: 0.040147777646780014 2023-01-21 12:14:53.749937: step: 1480/529, loss: 0.000255584716796875 2023-01-21 12:14:54.833962: step: 1484/529, loss: 0.0011684418423101306 2023-01-21 12:14:55.945131: step: 1488/529, loss: 0.005466938018798828 2023-01-21 12:14:57.060616: step: 1492/529, loss: 0.0014110564952716231 2023-01-21 12:14:58.190563: step: 1496/529, loss: 0.0007377624860964715 2023-01-21 12:14:59.291256: step: 1500/529, loss: 0.009814644232392311 2023-01-21 12:15:00.392610: step: 1504/529, loss: 4.677772813010961e-05 2023-01-21 12:15:01.506846: step: 1508/529, loss: 0.004952812101691961 2023-01-21 12:15:02.666801: step: 1512/529, loss: 0.2715301513671875 2023-01-21 12:15:03.793133: step: 1516/529, loss: 0.0002231597900390625 2023-01-21 12:15:04.901800: step: 1520/529, loss: 0.02539835125207901 2023-01-21 12:15:06.030523: step: 1524/529, loss: 0.0042705535888671875 2023-01-21 12:15:07.153841: step: 1528/529, loss: 0.7415567636489868 2023-01-21 12:15:08.267070: step: 1532/529, loss: 0.07374286651611328 2023-01-21 12:15:09.399404: step: 1536/529, loss: 0.02031269110739231 2023-01-21 12:15:10.526968: step: 1540/529, loss: 0.00041675567626953125 2023-01-21 12:15:11.647862: step: 1544/529, loss: 0.0003694534534588456 2023-01-21 12:15:12.753944: step: 1548/529, loss: 0.05913887172937393 2023-01-21 12:15:13.881654: step: 1552/529, loss: 0.013163280673325062 2023-01-21 12:15:15.016937: step: 1556/529, loss: 0.029240036383271217 2023-01-21 12:15:16.142387: step: 1560/529, loss: 0.0071218013763427734 2023-01-21 12:15:17.253874: step: 1564/529, loss: 0.0009699404472485185 2023-01-21 12:15:18.391541: step: 1568/529, loss: 0.000539588974788785 2023-01-21 12:15:19.522181: step: 1572/529, loss: 0.005153656005859375 2023-01-21 12:15:20.711714: step: 1576/529, loss: 0.02655940130352974 2023-01-21 12:15:21.843044: step: 1580/529, loss: 0.04200410842895508 2023-01-21 12:15:22.937026: step: 1584/529, loss: 0.0011692047119140625 2023-01-21 12:15:24.085071: step: 1588/529, loss: 0.04342499002814293 2023-01-21 12:15:25.207960: step: 1592/529, loss: 0.04300084337592125 2023-01-21 12:15:26.341242: step: 1596/529, loss: 0.011670398525893688 2023-01-21 12:15:27.456425: step: 1600/529, loss: 0.022550202906131744 2023-01-21 12:15:28.590071: step: 1604/529, loss: 0.0027776716742664576 2023-01-21 12:15:29.684486: step: 1608/529, loss: 0.004341507330536842 2023-01-21 12:15:30.799960: step: 1612/529, loss: 0.0013125420082360506 2023-01-21 12:15:31.941888: step: 1616/529, loss: 0.0021593093406409025 2023-01-21 12:15:33.056262: step: 1620/529, loss: 0.00011749267287086695 2023-01-21 12:15:34.168086: step: 1624/529, loss: 0.02150888554751873 2023-01-21 12:15:35.272704: step: 1628/529, loss: 0.045619964599609375 2023-01-21 12:15:36.400000: step: 1632/529, loss: 0.004310798831284046 2023-01-21 12:15:37.493119: step: 1636/529, loss: 2.2983551389188506e-05 2023-01-21 12:15:38.594919: step: 1640/529, loss: 0.028566457331180573 2023-01-21 12:15:39.727274: step: 1644/529, loss: 0.000789642333984375 2023-01-21 12:15:40.840893: step: 1648/529, loss: 0.004481792449951172 2023-01-21 12:15:41.942250: step: 1652/529, loss: 0.0011348724365234375 2023-01-21 12:15:43.057741: step: 1656/529, loss: 0.00030126573983579874 2023-01-21 12:15:44.152214: step: 1660/529, loss: 0.00104694371111691 2023-01-21 12:15:45.265295: step: 1664/529, loss: 0.0154876708984375 2023-01-21 12:15:46.371801: step: 1668/529, loss: 0.00035572052001953125 2023-01-21 12:15:47.512736: step: 1672/529, loss: 9.260178194381297e-05 2023-01-21 12:15:48.615898: step: 1676/529, loss: 0.007158947177231312 2023-01-21 12:15:49.721618: step: 1680/529, loss: 0.025133611634373665 2023-01-21 12:15:50.814983: step: 1684/529, loss: 0.006118583492934704 2023-01-21 12:15:51.919661: step: 1688/529, loss: 0.013369262218475342 2023-01-21 12:15:53.051453: step: 1692/529, loss: 0.009079551324248314 2023-01-21 12:15:54.176926: step: 1696/529, loss: 0.0014762879582121968 2023-01-21 12:15:55.270851: step: 1700/529, loss: 0.01657400280237198 2023-01-21 12:15:56.368627: step: 1704/529, loss: 0.043244171887636185 2023-01-21 12:15:57.477059: step: 1708/529, loss: 1.354217511106981e-05 2023-01-21 12:15:58.615338: step: 1712/529, loss: 0.002960681915283203 2023-01-21 12:15:59.726730: step: 1716/529, loss: 0.0009980201721191406 2023-01-21 12:16:00.849941: step: 1720/529, loss: 4.863738922722405e-06 2023-01-21 12:16:01.966215: step: 1724/529, loss: 0.0004366874636616558 2023-01-21 12:16:03.066688: step: 1728/529, loss: 0.006194877438247204 2023-01-21 12:16:04.186022: step: 1732/529, loss: 0.020427893847227097 2023-01-21 12:16:05.339326: step: 1736/529, loss: 0.013141060248017311 2023-01-21 12:16:06.459556: step: 1740/529, loss: 0.002741909120231867 2023-01-21 12:16:07.598358: step: 1744/529, loss: 0.016582680866122246 2023-01-21 12:16:08.707863: step: 1748/529, loss: 0.007382869720458984 2023-01-21 12:16:09.834832: step: 1752/529, loss: 0.00027446745662018657 2023-01-21 12:16:10.967933: step: 1756/529, loss: 0.010853147134184837 2023-01-21 12:16:12.072908: step: 1760/529, loss: 0.07315180450677872 2023-01-21 12:16:13.187320: step: 1764/529, loss: 0.004629040136933327 2023-01-21 12:16:14.300239: step: 1768/529, loss: 0.0005635261768475175 2023-01-21 12:16:15.400815: step: 1772/529, loss: 0.00105199811514467 2023-01-21 12:16:16.535115: step: 1776/529, loss: 0.0004665374872274697 2023-01-21 12:16:17.661814: step: 1780/529, loss: 0.0010008811950683594 2023-01-21 12:16:18.768858: step: 1784/529, loss: 0.0030444145668298006 2023-01-21 12:16:19.885041: step: 1788/529, loss: 0.021874619647860527 2023-01-21 12:16:21.036442: step: 1792/529, loss: 0.0029835226014256477 2023-01-21 12:16:22.140140: step: 1796/529, loss: 0.008694648742675781 2023-01-21 12:16:23.249586: step: 1800/529, loss: 0.00028266909066587687 2023-01-21 12:16:24.369781: step: 1804/529, loss: 0.0015039443969726562 2023-01-21 12:16:25.532146: step: 1808/529, loss: 0.00198268867097795 2023-01-21 12:16:26.631792: step: 1812/529, loss: 0.0007439613691531122 2023-01-21 12:16:27.714842: step: 1816/529, loss: 0.0008378028869628906 2023-01-21 12:16:28.812171: step: 1820/529, loss: 0.00011162758164573461 2023-01-21 12:16:29.910277: step: 1824/529, loss: 0.0003180503845214844 2023-01-21 12:16:31.026155: step: 1828/529, loss: 0.051531314849853516 2023-01-21 12:16:32.143097: step: 1832/529, loss: 0.019864464178681374 2023-01-21 12:16:33.283411: step: 1836/529, loss: 0.008056068792939186 2023-01-21 12:16:34.406759: step: 1840/529, loss: 0.11192789673805237 2023-01-21 12:16:35.517158: step: 1844/529, loss: 0.0037076950538903475 2023-01-21 12:16:36.647545: step: 1848/529, loss: 0.04940738528966904 2023-01-21 12:16:37.758329: step: 1852/529, loss: 0.006371498107910156 2023-01-21 12:16:38.901336: step: 1856/529, loss: 0.3699088990688324 2023-01-21 12:16:40.023640: step: 1860/529, loss: 3.0517576306010596e-06 2023-01-21 12:16:41.119883: step: 1864/529, loss: 4.0483475459041074e-05 2023-01-21 12:16:42.232283: step: 1868/529, loss: 0.00414695730432868 2023-01-21 12:16:43.338171: step: 1872/529, loss: 0.00820924062281847 2023-01-21 12:16:44.450441: step: 1876/529, loss: 0.2512439787387848 2023-01-21 12:16:45.569238: step: 1880/529, loss: 0.02620558813214302 2023-01-21 12:16:46.714254: step: 1884/529, loss: 0.05309629440307617 2023-01-21 12:16:47.817249: step: 1888/529, loss: 0.10921745747327805 2023-01-21 12:16:48.929065: step: 1892/529, loss: 0.0004646301094908267 2023-01-21 12:16:50.049451: step: 1896/529, loss: 0.11730308830738068 2023-01-21 12:16:51.177141: step: 1900/529, loss: 0.010869168676435947 2023-01-21 12:16:52.306447: step: 1904/529, loss: 0.04337196424603462 2023-01-21 12:16:53.438985: step: 1908/529, loss: 0.003235435578972101 2023-01-21 12:16:54.555812: step: 1912/529, loss: 0.00013227463932707906 2023-01-21 12:16:55.683158: step: 1916/529, loss: 0.018679238855838776 2023-01-21 12:16:56.805676: step: 1920/529, loss: 6.599426706088707e-05 2023-01-21 12:16:57.916035: step: 1924/529, loss: 0.3135354816913605 2023-01-21 12:16:59.025431: step: 1928/529, loss: 0.0017744064098224044 2023-01-21 12:17:00.151992: step: 1932/529, loss: 0.044943057000637054 2023-01-21 12:17:01.283487: step: 1936/529, loss: 0.0017619132995605469 2023-01-21 12:17:02.410349: step: 1940/529, loss: 0.06216239929199219 2023-01-21 12:17:03.537267: step: 1944/529, loss: 0.0032334327697753906 2023-01-21 12:17:04.648098: step: 1948/529, loss: 0.003442192217335105 2023-01-21 12:17:05.737900: step: 1952/529, loss: 0.006398964207619429 2023-01-21 12:17:06.863741: step: 1956/529, loss: 0.011774063110351562 2023-01-21 12:17:07.972436: step: 1960/529, loss: 0.012065314687788486 2023-01-21 12:17:09.089842: step: 1964/529, loss: 0.7654058337211609 2023-01-21 12:17:10.205340: step: 1968/529, loss: 0.0014257431030273438 2023-01-21 12:17:11.340420: step: 1972/529, loss: 0.0005074501386843622 2023-01-21 12:17:12.457730: step: 1976/529, loss: 0.0012655259342864156 2023-01-21 12:17:13.561056: step: 1980/529, loss: 0.0024993896950036287 2023-01-21 12:17:14.661830: step: 1984/529, loss: 0.008470725268125534 2023-01-21 12:17:15.777818: step: 1988/529, loss: 0.0005449295276775956 2023-01-21 12:17:16.918443: step: 1992/529, loss: 0.002991199493408203 2023-01-21 12:17:18.047925: step: 1996/529, loss: 1.0204315003647935e-05 2023-01-21 12:17:19.161885: step: 2000/529, loss: 0.025909423828125 2023-01-21 12:17:20.253301: step: 2004/529, loss: 0.0010359763400629163 2023-01-21 12:17:21.359159: step: 2008/529, loss: 0.01930256001651287 2023-01-21 12:17:22.478854: step: 2012/529, loss: 0.013964653015136719 2023-01-21 12:17:23.570487: step: 2016/529, loss: 0.01030054036527872 2023-01-21 12:17:24.653501: step: 2020/529, loss: 0.00588755588978529 2023-01-21 12:17:25.765175: step: 2024/529, loss: 5.5122378398664296e-05 2023-01-21 12:17:26.870191: step: 2028/529, loss: 0.010770846158266068 2023-01-21 12:17:27.974377: step: 2032/529, loss: 0.0028553009033203125 2023-01-21 12:17:29.096238: step: 2036/529, loss: 0.010937022976577282 2023-01-21 12:17:30.187669: step: 2040/529, loss: 0.27414292097091675 2023-01-21 12:17:31.282673: step: 2044/529, loss: 3.9482114516431466e-05 2023-01-21 12:17:32.391607: step: 2048/529, loss: 0.38262826204299927 2023-01-21 12:17:33.504275: step: 2052/529, loss: 0.0010560035007074475 2023-01-21 12:17:34.617648: step: 2056/529, loss: 0.0007604121929034591 2023-01-21 12:17:35.748336: step: 2060/529, loss: 0.004108143039047718 2023-01-21 12:17:36.881852: step: 2064/529, loss: 0.016538048163056374 2023-01-21 12:17:38.008533: step: 2068/529, loss: 0.005854463670402765 2023-01-21 12:17:39.107836: step: 2072/529, loss: 0.013224029913544655 2023-01-21 12:17:40.207463: step: 2076/529, loss: 0.03184051439166069 2023-01-21 12:17:41.327196: step: 2080/529, loss: 0.009176159277558327 2023-01-21 12:17:42.458408: step: 2084/529, loss: 0.0009187221294268966 2023-01-21 12:17:43.569366: step: 2088/529, loss: 0.00248298654332757 2023-01-21 12:17:44.674302: step: 2092/529, loss: 0.007567978464066982 2023-01-21 12:17:45.785504: step: 2096/529, loss: 0.004356956575065851 2023-01-21 12:17:46.889965: step: 2100/529, loss: 0.07182197272777557 2023-01-21 12:17:48.002575: step: 2104/529, loss: 0.021478915587067604 2023-01-21 12:17:49.107537: step: 2108/529, loss: 0.006385231390595436 2023-01-21 12:17:50.251189: step: 2112/529, loss: 1.024275779724121 2023-01-21 12:17:51.372352: step: 2116/529, loss: 0.026096520945429802 ================================================== Loss: 0.027 -------------------- Dev: {'event': {'p': 0.6094929881337648, 'r': 0.7523302263648469, 'f1': 0.6734207389749702}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Test: {'event': {'p': 0.6587263004375303, 'r': 0.7646726862302483, 'f1': 0.7077565944110733}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Chinese: {'event': {'p': 0.5568181818181818, 'r': 0.9074074074074074, 'f1': 0.6901408450704225}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Korean: {'event': {'p': 0.6153846153846154, 'r': 0.5079365079365079, 'f1': 0.5565217391304348}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Russian: {'event': {'p': 0.38461538461538464, 'r': 0.4166666666666667, 'f1': 0.4}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 22 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:18:31.472194: step: 4/529, loss: 4.901886131847277e-05 2023-01-21 12:18:32.570135: step: 8/529, loss: 0.0004726409970317036 2023-01-21 12:18:33.680084: step: 12/529, loss: 0.0005843162653036416 2023-01-21 12:18:34.786379: step: 16/529, loss: 0.049048613756895065 2023-01-21 12:18:35.909707: step: 20/529, loss: 0.0026485442649573088 2023-01-21 12:18:37.063038: step: 24/529, loss: 0.0010603904956951737 2023-01-21 12:18:38.177784: step: 28/529, loss: 0.048999905586242676 2023-01-21 12:18:39.299407: step: 32/529, loss: 0.03976612165570259 2023-01-21 12:18:40.424188: step: 36/529, loss: 1.52587890625e-05 2023-01-21 12:18:41.519260: step: 40/529, loss: 8.912086923373863e-05 2023-01-21 12:18:42.656406: step: 44/529, loss: 0.0007578849908895791 2023-01-21 12:18:43.800813: step: 48/529, loss: 0.028232479467988014 2023-01-21 12:18:44.927221: step: 52/529, loss: 0.006174850277602673 2023-01-21 12:18:46.033674: step: 56/529, loss: 0.0016797066200524569 2023-01-21 12:18:47.174475: step: 60/529, loss: 0.002244377275928855 2023-01-21 12:18:48.294669: step: 64/529, loss: 0.005736160557717085 2023-01-21 12:18:49.391847: step: 68/529, loss: 0.0019905567169189453 2023-01-21 12:18:50.495075: step: 72/529, loss: 0.003225994063541293 2023-01-21 12:18:51.600051: step: 76/529, loss: 0.0010509968269616365 2023-01-21 12:18:52.724781: step: 80/529, loss: 0.003352546598762274 2023-01-21 12:18:53.834974: step: 84/529, loss: 0.004018831066787243 2023-01-21 12:18:54.928992: step: 88/529, loss: 0.023929011076688766 2023-01-21 12:18:56.031003: step: 92/529, loss: 0.005522346589714289 2023-01-21 12:18:57.163921: step: 96/529, loss: 0.0014234542613849044 2023-01-21 12:18:58.285734: step: 100/529, loss: 0.027295876294374466 2023-01-21 12:18:59.401112: step: 104/529, loss: 0.012465191073715687 2023-01-21 12:19:00.556487: step: 108/529, loss: 0.005788040347397327 2023-01-21 12:19:01.672813: step: 112/529, loss: 0.03737945482134819 2023-01-21 12:19:02.813420: step: 116/529, loss: 0.019512368366122246 2023-01-21 12:19:03.914972: step: 120/529, loss: 0.0006292343605309725 2023-01-21 12:19:05.047541: step: 124/529, loss: 0.003832245012745261 2023-01-21 12:19:06.158751: step: 128/529, loss: 0.0006239890935830772 2023-01-21 12:19:07.296146: step: 132/529, loss: 0.00016345977201126516 2023-01-21 12:19:08.408878: step: 136/529, loss: 0.00896678026765585 2023-01-21 12:19:09.506566: step: 140/529, loss: 2.0694733393611386e-05 2023-01-21 12:19:10.617388: step: 144/529, loss: 0.0008705139043740928 2023-01-21 12:19:11.742700: step: 148/529, loss: 0.003926277160644531 2023-01-21 12:19:12.880630: step: 152/529, loss: 0.0008455276256427169 2023-01-21 12:19:13.996780: step: 156/529, loss: 0.0031423568725585938 2023-01-21 12:19:15.123284: step: 160/529, loss: 0.553687334060669 2023-01-21 12:19:16.229735: step: 164/529, loss: 0.0011714935535565019 2023-01-21 12:19:17.375900: step: 168/529, loss: 0.5581789016723633 2023-01-21 12:19:18.472264: step: 172/529, loss: 0.005222129635512829 2023-01-21 12:19:19.591104: step: 176/529, loss: 0.009992409497499466 2023-01-21 12:19:20.713465: step: 180/529, loss: 0.004883193876594305 2023-01-21 12:19:21.852916: step: 184/529, loss: 0.0013727188343182206 2023-01-21 12:19:22.962871: step: 188/529, loss: 0.009607887826859951 2023-01-21 12:19:24.046120: step: 192/529, loss: 0.00033283233642578125 2023-01-21 12:19:25.195374: step: 196/529, loss: 0.017383098602294922 2023-01-21 12:19:26.335690: step: 200/529, loss: 0.0001848220854299143 2023-01-21 12:19:27.458029: step: 204/529, loss: 9.93728608591482e-05 2023-01-21 12:19:28.561801: step: 208/529, loss: 0.0021793365012854338 2023-01-21 12:19:29.665780: step: 212/529, loss: 0.00042705534724518657 2023-01-21 12:19:30.790256: step: 216/529, loss: 0.3826066851615906 2023-01-21 12:19:31.924054: step: 220/529, loss: 0.26318082213401794 2023-01-21 12:19:33.032245: step: 224/529, loss: 9.231566946255043e-05 2023-01-21 12:19:34.149681: step: 228/529, loss: 0.0006345749134197831 2023-01-21 12:19:35.269989: step: 232/529, loss: 0.0035393715370446444 2023-01-21 12:19:36.410656: step: 236/529, loss: 0.07823028415441513 2023-01-21 12:19:37.525825: step: 240/529, loss: 0.0004249572812113911 2023-01-21 12:19:38.694915: step: 244/529, loss: 0.00012073517427779734 2023-01-21 12:19:39.856969: step: 248/529, loss: 3.82423386326991e-05 2023-01-21 12:19:40.962459: step: 252/529, loss: 0.00955276470631361 2023-01-21 12:19:42.051589: step: 256/529, loss: 0.00027799609233625233 2023-01-21 12:19:43.161422: step: 260/529, loss: 0.011612558737397194 2023-01-21 12:19:44.254850: step: 264/529, loss: 0.040853217244148254 2023-01-21 12:19:45.370706: step: 268/529, loss: 0.01077041681855917 2023-01-21 12:19:46.493564: step: 272/529, loss: 0.006834125611931086 2023-01-21 12:19:47.599760: step: 276/529, loss: 0.0014696121215820312 2023-01-21 12:19:48.725912: step: 280/529, loss: 0.0016334534157067537 2023-01-21 12:19:49.842712: step: 284/529, loss: 0.0001579284726176411 2023-01-21 12:19:50.984900: step: 288/529, loss: 0.00015726090350653976 2023-01-21 12:19:52.113484: step: 292/529, loss: 0.004276657477021217 2023-01-21 12:19:53.213178: step: 296/529, loss: 3.5762786865234375e-06 2023-01-21 12:19:54.327544: step: 300/529, loss: 0.014804649166762829 2023-01-21 12:19:55.420312: step: 304/529, loss: 0.024490738287568092 2023-01-21 12:19:56.555205: step: 308/529, loss: 0.0019912063144147396 2023-01-21 12:19:57.676296: step: 312/529, loss: 0.1300937682390213 2023-01-21 12:19:58.811192: step: 316/529, loss: 0.002980518387630582 2023-01-21 12:19:59.905474: step: 320/529, loss: 0.023563861846923828 2023-01-21 12:20:01.021797: step: 324/529, loss: 4.3106076191179454e-05 2023-01-21 12:20:02.161215: step: 328/529, loss: 0.0004907608381472528 2023-01-21 12:20:03.262662: step: 332/529, loss: 0.010013771243393421 2023-01-21 12:20:04.384955: step: 336/529, loss: 0.0009902477031573653 2023-01-21 12:20:05.483641: step: 340/529, loss: 0.41364210844039917 2023-01-21 12:20:06.601822: step: 344/529, loss: 3.1948089599609375e-05 2023-01-21 12:20:07.699268: step: 348/529, loss: 0.002775049302726984 2023-01-21 12:20:08.791196: step: 352/529, loss: 0.0022239687386900187 2023-01-21 12:20:09.895552: step: 356/529, loss: 0.015602302737534046 2023-01-21 12:20:11.028245: step: 360/529, loss: 0.08141651004552841 2023-01-21 12:20:12.154206: step: 364/529, loss: 0.0009212494478560984 2023-01-21 12:20:13.258743: step: 368/529, loss: 0.1698707640171051 2023-01-21 12:20:14.337939: step: 372/529, loss: 0.013544559478759766 2023-01-21 12:20:15.430857: step: 376/529, loss: 0.0018564224010333419 2023-01-21 12:20:16.580418: step: 380/529, loss: 1.583099401614163e-05 2023-01-21 12:20:17.698403: step: 384/529, loss: 0.010606384836137295 2023-01-21 12:20:18.795570: step: 388/529, loss: 0.0024221420753747225 2023-01-21 12:20:19.905837: step: 392/529, loss: 0.0005651474348269403 2023-01-21 12:20:21.018511: step: 396/529, loss: 0.0028150558937340975 2023-01-21 12:20:22.128149: step: 400/529, loss: 0.001256752060726285 2023-01-21 12:20:23.251650: step: 404/529, loss: 0.002836036728695035 2023-01-21 12:20:24.367617: step: 408/529, loss: 2.326965295651462e-05 2023-01-21 12:20:25.481372: step: 412/529, loss: 0.005748748779296875 2023-01-21 12:20:26.553225: step: 416/529, loss: 4.9400332500226796e-05 2023-01-21 12:20:27.702354: step: 420/529, loss: 0.00025959016056731343 2023-01-21 12:20:28.832404: step: 424/529, loss: 0.006981086917221546 2023-01-21 12:20:29.918627: step: 428/529, loss: 0.00010318756540073082 2023-01-21 12:20:31.004338: step: 432/529, loss: 0.0007403373601846397 2023-01-21 12:20:32.155889: step: 436/529, loss: 0.03095264546573162 2023-01-21 12:20:33.247870: step: 440/529, loss: 0.04470587149262428 2023-01-21 12:20:34.335874: step: 444/529, loss: 0.0004709243949037045 2023-01-21 12:20:35.441801: step: 448/529, loss: 0.007936859503388405 2023-01-21 12:20:36.524397: step: 452/529, loss: 8.668900409247726e-05 2023-01-21 12:20:37.623538: step: 456/529, loss: 0.0007596969953738153 2023-01-21 12:20:38.732606: step: 460/529, loss: 0.003589821048080921 2023-01-21 12:20:39.869468: step: 464/529, loss: 0.0002636909484863281 2023-01-21 12:20:40.971281: step: 468/529, loss: 0.04603540897369385 2023-01-21 12:20:42.088913: step: 472/529, loss: 0.024915315210819244 2023-01-21 12:20:43.191540: step: 476/529, loss: 0.012034988962113857 2023-01-21 12:20:44.302729: step: 480/529, loss: 0.02733173407614231 2023-01-21 12:20:45.390670: step: 484/529, loss: 0.03851928561925888 2023-01-21 12:20:46.514263: step: 488/529, loss: 0.015702057629823685 2023-01-21 12:20:47.634494: step: 492/529, loss: 0.0033882143907248974 2023-01-21 12:20:48.732285: step: 496/529, loss: 0.0029401779174804688 2023-01-21 12:20:49.843703: step: 500/529, loss: 0.009630394168198109 2023-01-21 12:20:50.963311: step: 504/529, loss: 0.026005173102021217 2023-01-21 12:20:52.051147: step: 508/529, loss: 0.00022544861712958664 2023-01-21 12:20:53.167585: step: 512/529, loss: 0.0009356499067507684 2023-01-21 12:20:54.305567: step: 516/529, loss: 0.0028443813789635897 2023-01-21 12:20:55.406486: step: 520/529, loss: 0.008927345275878906 2023-01-21 12:20:56.503873: step: 524/529, loss: 0.0003406524483580142 2023-01-21 12:20:57.610501: step: 528/529, loss: 0.00324764265678823 2023-01-21 12:20:58.745565: step: 532/529, loss: 0.007606315892189741 2023-01-21 12:20:59.853932: step: 536/529, loss: 0.19162806868553162 2023-01-21 12:21:00.981037: step: 540/529, loss: 0.004835891537368298 2023-01-21 12:21:02.083469: step: 544/529, loss: 0.0001619339018361643 2023-01-21 12:21:03.198852: step: 548/529, loss: 0.6177568435668945 2023-01-21 12:21:04.315639: step: 552/529, loss: 0.012438583187758923 2023-01-21 12:21:05.460767: step: 556/529, loss: 0.2524687349796295 2023-01-21 12:21:06.585543: step: 560/529, loss: 0.0002919197140727192 2023-01-21 12:21:07.671096: step: 564/529, loss: 0.0002232074912171811 2023-01-21 12:21:08.761590: step: 568/529, loss: 0.0011180877918377519 2023-01-21 12:21:09.884436: step: 572/529, loss: 0.009082472883164883 2023-01-21 12:21:10.980223: step: 576/529, loss: 0.0005714416620321572 2023-01-21 12:21:12.157302: step: 580/529, loss: 2.155303991457913e-05 2023-01-21 12:21:13.267151: step: 584/529, loss: 8.897780935512856e-05 2023-01-21 12:21:14.395391: step: 588/529, loss: 0.0023399353958666325 2023-01-21 12:21:15.551583: step: 592/529, loss: 0.24765463173389435 2023-01-21 12:21:16.668122: step: 596/529, loss: 0.22901973128318787 2023-01-21 12:21:17.781531: step: 600/529, loss: 0.01999054104089737 2023-01-21 12:21:18.880242: step: 604/529, loss: 2.460479663568549e-05 2023-01-21 12:21:20.015440: step: 608/529, loss: 0.004704666323959827 2023-01-21 12:21:21.127527: step: 612/529, loss: 0.007064437959343195 2023-01-21 12:21:22.233812: step: 616/529, loss: 0.0025806427001953125 2023-01-21 12:21:23.351673: step: 620/529, loss: 0.00015382767014671117 2023-01-21 12:21:24.461830: step: 624/529, loss: 0.007519149221479893 2023-01-21 12:21:25.560809: step: 628/529, loss: 0.0013467789394780993 2023-01-21 12:21:26.648236: step: 632/529, loss: 0.0016680718399584293 2023-01-21 12:21:27.736806: step: 636/529, loss: 2.784729076665826e-05 2023-01-21 12:21:28.830265: step: 640/529, loss: 0.002814328530803323 2023-01-21 12:21:29.982422: step: 644/529, loss: 0.005088615696877241 2023-01-21 12:21:31.136313: step: 648/529, loss: 1.8596649169921875e-05 2023-01-21 12:21:32.266304: step: 652/529, loss: 0.01365804672241211 2023-01-21 12:21:33.381299: step: 656/529, loss: 0.015225409530103207 2023-01-21 12:21:34.478855: step: 660/529, loss: 0.0418458953499794 2023-01-21 12:21:35.579704: step: 664/529, loss: 0.0024587633088231087 2023-01-21 12:21:36.741266: step: 668/529, loss: 0.08581829071044922 2023-01-21 12:21:37.869927: step: 672/529, loss: 0.26394766569137573 2023-01-21 12:21:39.002664: step: 676/529, loss: 0.0002870559983421117 2023-01-21 12:21:40.155136: step: 680/529, loss: 0.04272060468792915 2023-01-21 12:21:41.280778: step: 684/529, loss: 0.01952838897705078 2023-01-21 12:21:42.386071: step: 688/529, loss: 0.00015697479830123484 2023-01-21 12:21:43.503048: step: 692/529, loss: 0.004638290498405695 2023-01-21 12:21:44.611420: step: 696/529, loss: 0.0224138256162405 2023-01-21 12:21:45.724996: step: 700/529, loss: 4.6205521357478574e-05 2023-01-21 12:21:46.825888: step: 704/529, loss: 0.02264728583395481 2023-01-21 12:21:48.011396: step: 708/529, loss: 6.980895705055445e-05 2023-01-21 12:21:49.123082: step: 712/529, loss: 0.02875976637005806 2023-01-21 12:21:50.249739: step: 716/529, loss: 0.01690845564007759 2023-01-21 12:21:51.334489: step: 720/529, loss: 0.0027151107788085938 2023-01-21 12:21:52.445784: step: 724/529, loss: 5.617141869151965e-05 2023-01-21 12:21:53.551083: step: 728/529, loss: 0.0006729125743731856 2023-01-21 12:21:54.684626: step: 732/529, loss: 4.4298172724666074e-05 2023-01-21 12:21:55.799919: step: 736/529, loss: 0.007973956875503063 2023-01-21 12:21:56.945912: step: 740/529, loss: 0.00034084319486282766 2023-01-21 12:21:58.057307: step: 744/529, loss: 0.00021448136249091476 2023-01-21 12:21:59.170958: step: 748/529, loss: 0.020554019138216972 2023-01-21 12:22:00.280799: step: 752/529, loss: 3.814697265625e-05 2023-01-21 12:22:01.402010: step: 756/529, loss: 2.174377368646674e-05 2023-01-21 12:22:02.513023: step: 760/529, loss: 0.01640625111758709 2023-01-21 12:22:03.651516: step: 764/529, loss: 0.0030371665488928556 2023-01-21 12:22:04.798589: step: 768/529, loss: 0.006185627076774836 2023-01-21 12:22:05.910503: step: 772/529, loss: 0.005714702419936657 2023-01-21 12:22:06.991628: step: 776/529, loss: 3.5762786865234375e-05 2023-01-21 12:22:08.083524: step: 780/529, loss: 0.003929233644157648 2023-01-21 12:22:09.165529: step: 784/529, loss: 0.015489769168198109 2023-01-21 12:22:10.265209: step: 788/529, loss: 0.018300725147128105 2023-01-21 12:22:11.364262: step: 792/529, loss: 0.09302692860364914 2023-01-21 12:22:12.460159: step: 796/529, loss: 0.014369487762451172 2023-01-21 12:22:13.605882: step: 800/529, loss: 0.03820314630866051 2023-01-21 12:22:14.714289: step: 804/529, loss: 0.01919526979327202 2023-01-21 12:22:15.806552: step: 808/529, loss: 0.008376121520996094 2023-01-21 12:22:16.918606: step: 812/529, loss: 0.0006199836498126388 2023-01-21 12:22:18.055615: step: 816/529, loss: 0.008919382467865944 2023-01-21 12:22:19.202953: step: 820/529, loss: 0.0005117416149005294 2023-01-21 12:22:20.341701: step: 824/529, loss: 0.007665443699806929 2023-01-21 12:22:21.444397: step: 828/529, loss: 0.00035896303597837687 2023-01-21 12:22:22.590072: step: 832/529, loss: 0.015477752313017845 2023-01-21 12:22:23.690696: step: 836/529, loss: 0.003006553743034601 2023-01-21 12:22:24.831991: step: 840/529, loss: 0.0019371986854821444 2023-01-21 12:22:25.924505: step: 844/529, loss: 0.04035945236682892 2023-01-21 12:22:27.018674: step: 848/529, loss: 0.00023889541625976562 2023-01-21 12:22:28.145441: step: 852/529, loss: 7.877349707996473e-05 2023-01-21 12:22:29.278699: step: 856/529, loss: 0.14644727110862732 2023-01-21 12:22:30.416728: step: 860/529, loss: 3.147125244140625e-05 2023-01-21 12:22:31.572126: step: 864/529, loss: 0.005475949961692095 2023-01-21 12:22:32.679607: step: 868/529, loss: 0.004447365179657936 2023-01-21 12:22:33.801043: step: 872/529, loss: 0.019777726382017136 2023-01-21 12:22:34.904908: step: 876/529, loss: 0.00755462609231472 2023-01-21 12:22:36.018702: step: 880/529, loss: 0.0007345199701376259 2023-01-21 12:22:37.159142: step: 884/529, loss: 0.03260960429906845 2023-01-21 12:22:38.273554: step: 888/529, loss: 0.00593643169850111 2023-01-21 12:22:39.373526: step: 892/529, loss: 0.041945651173591614 2023-01-21 12:22:40.510469: step: 896/529, loss: 0.0018585205543786287 2023-01-21 12:22:41.624603: step: 900/529, loss: 0.018059730529785156 2023-01-21 12:22:42.702962: step: 904/529, loss: 0.004139136988669634 2023-01-21 12:22:43.816647: step: 908/529, loss: 0.0861610397696495 2023-01-21 12:22:44.925645: step: 912/529, loss: 0.0001224517764057964 2023-01-21 12:22:46.005501: step: 916/529, loss: 0.02920396439731121 2023-01-21 12:22:47.082577: step: 920/529, loss: 0.0016676902305334806 2023-01-21 12:22:48.189383: step: 924/529, loss: 0.00087738037109375 2023-01-21 12:22:49.289355: step: 928/529, loss: -2.3245811462402344e-06 2023-01-21 12:22:50.409769: step: 932/529, loss: 0.009790134616196156 2023-01-21 12:22:51.516494: step: 936/529, loss: 5.3405768994707614e-05 2023-01-21 12:22:52.612038: step: 940/529, loss: 0.0036839961539953947 2023-01-21 12:22:53.760929: step: 944/529, loss: 0.018985319882631302 2023-01-21 12:22:54.902343: step: 948/529, loss: 0.006720351986587048 2023-01-21 12:22:56.048549: step: 952/529, loss: 0.015435028821229935 2023-01-21 12:22:57.200292: step: 956/529, loss: 0.007334804628044367 2023-01-21 12:22:58.297339: step: 960/529, loss: 0.004056167788803577 2023-01-21 12:22:59.435230: step: 964/529, loss: 0.0017294883728027344 2023-01-21 12:23:00.541915: step: 968/529, loss: 0.008531475439667702 2023-01-21 12:23:01.658649: step: 972/529, loss: 0.0009712219471111894 2023-01-21 12:23:02.771976: step: 976/529, loss: 0.020585060119628906 2023-01-21 12:23:03.895153: step: 980/529, loss: 0.004735756199806929 2023-01-21 12:23:04.980969: step: 984/529, loss: 0.0003392219659872353 2023-01-21 12:23:06.100905: step: 988/529, loss: 0.11765418201684952 2023-01-21 12:23:07.198572: step: 992/529, loss: 0.0001665592280915007 2023-01-21 12:23:08.315971: step: 996/529, loss: 0.0006993294227868319 2023-01-21 12:23:09.440124: step: 1000/529, loss: 0.00245761894620955 2023-01-21 12:23:10.552315: step: 1004/529, loss: 0.002206802600994706 2023-01-21 12:23:11.658542: step: 1008/529, loss: 0.011037826538085938 2023-01-21 12:23:12.792155: step: 1012/529, loss: 0.0883277952671051 2023-01-21 12:23:13.889894: step: 1016/529, loss: 0.02365279383957386 2023-01-21 12:23:15.024059: step: 1020/529, loss: 0.0008777619223110378 2023-01-21 12:23:16.130245: step: 1024/529, loss: 0.0206924919039011 2023-01-21 12:23:17.277164: step: 1028/529, loss: 0.0024992942344397306 2023-01-21 12:23:18.398383: step: 1032/529, loss: 0.03853616863489151 2023-01-21 12:23:19.476244: step: 1036/529, loss: 0.021762752905488014 2023-01-21 12:23:20.598480: step: 1040/529, loss: 0.0027247429825365543 2023-01-21 12:23:21.726404: step: 1044/529, loss: 0.047234728932380676 2023-01-21 12:23:22.850714: step: 1048/529, loss: 0.0011034011840820312 2023-01-21 12:23:23.951665: step: 1052/529, loss: 0.00025615692720748484 2023-01-21 12:23:25.099716: step: 1056/529, loss: 0.0006631851429119706 2023-01-21 12:23:26.206708: step: 1060/529, loss: 0.019662827253341675 2023-01-21 12:23:27.337140: step: 1064/529, loss: 0.03680616617202759 2023-01-21 12:23:28.463559: step: 1068/529, loss: 0.0013397217262536287 2023-01-21 12:23:29.575596: step: 1072/529, loss: 5.130767749506049e-05 2023-01-21 12:23:30.721967: step: 1076/529, loss: 0.00908365286886692 2023-01-21 12:23:31.843031: step: 1080/529, loss: 0.026836395263671875 2023-01-21 12:23:32.950536: step: 1084/529, loss: 0.018147969618439674 2023-01-21 12:23:34.084549: step: 1088/529, loss: 0.0041119991801679134 2023-01-21 12:23:35.226249: step: 1092/529, loss: 0.00146064767614007 2023-01-21 12:23:36.335674: step: 1096/529, loss: 0.011078167706727982 2023-01-21 12:23:37.450192: step: 1100/529, loss: 0.007022762671113014 2023-01-21 12:23:38.573345: step: 1104/529, loss: 0.004471206571906805 2023-01-21 12:23:39.659433: step: 1108/529, loss: 0.004838466644287109 2023-01-21 12:23:40.789979: step: 1112/529, loss: 0.001230430556461215 2023-01-21 12:23:41.931740: step: 1116/529, loss: 0.00297126779332757 2023-01-21 12:23:43.050631: step: 1120/529, loss: 0.0006961346371099353 2023-01-21 12:23:44.160958: step: 1124/529, loss: 0.050437163561582565 2023-01-21 12:23:45.283529: step: 1128/529, loss: 0.014729499816894531 2023-01-21 12:23:46.421633: step: 1132/529, loss: 0.00025882720365189016 2023-01-21 12:23:47.506000: step: 1136/529, loss: 0.0005743026849813759 2023-01-21 12:23:48.641915: step: 1140/529, loss: 0.04766368865966797 2023-01-21 12:23:49.735853: step: 1144/529, loss: 0.004402542021125555 2023-01-21 12:23:50.842692: step: 1148/529, loss: 0.029987763613462448 2023-01-21 12:23:51.979559: step: 1152/529, loss: 4.38690185546875e-05 2023-01-21 12:23:53.112969: step: 1156/529, loss: 0.00021882056898903102 2023-01-21 12:23:54.236451: step: 1160/529, loss: 0.0006704330444335938 2023-01-21 12:23:55.354230: step: 1164/529, loss: 0.0028752328362315893 2023-01-21 12:23:56.459666: step: 1168/529, loss: 0.00045638083247467875 2023-01-21 12:23:57.558794: step: 1172/529, loss: 0.054761506617069244 2023-01-21 12:23:58.676698: step: 1176/529, loss: 0.02462015114724636 2023-01-21 12:23:59.807364: step: 1180/529, loss: 0.0006656170007772744 2023-01-21 12:24:00.926972: step: 1184/529, loss: 0.002207088517025113 2023-01-21 12:24:02.047992: step: 1188/529, loss: 0.02447948418557644 2023-01-21 12:24:03.204079: step: 1192/529, loss: 0.3125797212123871 2023-01-21 12:24:04.316881: step: 1196/529, loss: 0.0005607605562545359 2023-01-21 12:24:05.406060: step: 1200/529, loss: 0.0037145614624023438 2023-01-21 12:24:06.526549: step: 1204/529, loss: 0.0024746896233409643 2023-01-21 12:24:07.660655: step: 1208/529, loss: 0.029154395684599876 2023-01-21 12:24:08.770649: step: 1212/529, loss: 0.017415069043636322 2023-01-21 12:24:09.922245: step: 1216/529, loss: 0.0004466056707315147 2023-01-21 12:24:11.038609: step: 1220/529, loss: 0.003207397647202015 2023-01-21 12:24:12.186634: step: 1224/529, loss: 0.0042174337431788445 2023-01-21 12:24:13.321929: step: 1228/529, loss: 0.017203141003847122 2023-01-21 12:24:14.413596: step: 1232/529, loss: 0.00029935839120298624 2023-01-21 12:24:15.548690: step: 1236/529, loss: 0.037312984466552734 2023-01-21 12:24:16.647769: step: 1240/529, loss: 0.006142330355942249 2023-01-21 12:24:17.752226: step: 1244/529, loss: 0.004630709066987038 2023-01-21 12:24:18.850184: step: 1248/529, loss: 0.004734516143798828 2023-01-21 12:24:19.953839: step: 1252/529, loss: 0.013993168249726295 2023-01-21 12:24:21.059392: step: 1256/529, loss: 0.01038427371531725 2023-01-21 12:24:22.188752: step: 1260/529, loss: 0.0028164866380393505 2023-01-21 12:24:23.321600: step: 1264/529, loss: 0.005387067794799805 2023-01-21 12:24:24.415479: step: 1268/529, loss: 0.011713218875229359 2023-01-21 12:24:25.519694: step: 1272/529, loss: 0.00017104149446822703 2023-01-21 12:24:26.616743: step: 1276/529, loss: 0.009637641720473766 2023-01-21 12:24:27.752487: step: 1280/529, loss: 3.643035961431451e-05 2023-01-21 12:24:28.834171: step: 1284/529, loss: 0.00011596679541980848 2023-01-21 12:24:29.937415: step: 1288/529, loss: 0.0003276825009379536 2023-01-21 12:24:31.066562: step: 1292/529, loss: 0.009432983584702015 2023-01-21 12:24:32.158196: step: 1296/529, loss: 0.0001576423819642514 2023-01-21 12:24:33.264130: step: 1300/529, loss: 0.0010855287546291947 2023-01-21 12:24:34.414400: step: 1304/529, loss: 0.02320423163473606 2023-01-21 12:24:35.551016: step: 1308/529, loss: 0.008127403445541859 2023-01-21 12:24:36.654940: step: 1312/529, loss: 0.0034690857864916325 2023-01-21 12:24:37.763396: step: 1316/529, loss: 0.09195423126220703 2023-01-21 12:24:38.853616: step: 1320/529, loss: 0.0010787963401526213 2023-01-21 12:24:39.976865: step: 1324/529, loss: 6.580352783203125e-05 2023-01-21 12:24:41.077558: step: 1328/529, loss: 3.223419116693549e-05 2023-01-21 12:24:42.200015: step: 1332/529, loss: 0.01861104927957058 2023-01-21 12:24:43.320754: step: 1336/529, loss: 0.03626365587115288 2023-01-21 12:24:44.432981: step: 1340/529, loss: 0.0009206772083416581 2023-01-21 12:24:45.557815: step: 1344/529, loss: 0.003135064383968711 2023-01-21 12:24:46.674738: step: 1348/529, loss: 0.00027303697424940765 2023-01-21 12:24:47.769490: step: 1352/529, loss: 0.027225017547607422 2023-01-21 12:24:48.861360: step: 1356/529, loss: 0.002821683883666992 2023-01-21 12:24:50.000762: step: 1360/529, loss: 0.00284061417914927 2023-01-21 12:24:51.115818: step: 1364/529, loss: 0.003860569093376398 2023-01-21 12:24:52.229764: step: 1368/529, loss: 0.026370812207460403 2023-01-21 12:24:53.362885: step: 1372/529, loss: 0.009623098187148571 2023-01-21 12:24:54.450556: step: 1376/529, loss: 0.02466106414794922 2023-01-21 12:24:55.567408: step: 1380/529, loss: 0.0011364937527105212 2023-01-21 12:24:56.656585: step: 1384/529, loss: 0.01477899495512247 2023-01-21 12:24:57.745666: step: 1388/529, loss: 0.005523490719497204 2023-01-21 12:24:58.829557: step: 1392/529, loss: 0.0035267830826342106 2023-01-21 12:24:59.903597: step: 1396/529, loss: 1.583099401614163e-05 2023-01-21 12:25:01.011097: step: 1400/529, loss: 0.006800174713134766 2023-01-21 12:25:02.120278: step: 1404/529, loss: 0.48850059509277344 2023-01-21 12:25:03.263918: step: 1408/529, loss: 0.0010370254749432206 2023-01-21 12:25:04.368775: step: 1412/529, loss: 7.247924258990679e-06 2023-01-21 12:25:05.504516: step: 1416/529, loss: 0.00026645659818314016 2023-01-21 12:25:06.655287: step: 1420/529, loss: 0.009211158379912376 2023-01-21 12:25:07.755655: step: 1424/529, loss: 0.0007445812225341797 2023-01-21 12:25:08.889086: step: 1428/529, loss: 6.86645489622606e-06 2023-01-21 12:25:10.011365: step: 1432/529, loss: 0.028753947466611862 2023-01-21 12:25:11.145244: step: 1436/529, loss: 0.00615768413990736 2023-01-21 12:25:12.245073: step: 1440/529, loss: 0.004310321994125843 2023-01-21 12:25:13.334014: step: 1444/529, loss: 0.0011844635009765625 2023-01-21 12:25:14.442469: step: 1448/529, loss: 0.009363938122987747 2023-01-21 12:25:15.555358: step: 1452/529, loss: 0.0027895928360521793 2023-01-21 12:25:16.683778: step: 1456/529, loss: 0.007210922427475452 2023-01-21 12:25:17.762416: step: 1460/529, loss: 0.004180908203125 2023-01-21 12:25:18.874631: step: 1464/529, loss: 0.0015493392711505294 2023-01-21 12:25:19.999123: step: 1468/529, loss: 0.0043792724609375 2023-01-21 12:25:21.109461: step: 1472/529, loss: 0.0038200379349291325 2023-01-21 12:25:22.194194: step: 1476/529, loss: 0.014241695404052734 2023-01-21 12:25:23.301051: step: 1480/529, loss: 0.008434867486357689 2023-01-21 12:25:24.415949: step: 1484/529, loss: 0.00013966560072731227 2023-01-21 12:25:25.555721: step: 1488/529, loss: 7.324219041038305e-05 2023-01-21 12:25:26.673751: step: 1492/529, loss: 0.008545302785933018 2023-01-21 12:25:27.774046: step: 1496/529, loss: 0.12751750648021698 2023-01-21 12:25:28.894637: step: 1500/529, loss: 0.029554177075624466 2023-01-21 12:25:30.029109: step: 1504/529, loss: 0.009050941094756126 2023-01-21 12:25:31.146986: step: 1508/529, loss: 0.027750778943300247 2023-01-21 12:25:32.259108: step: 1512/529, loss: 0.0012173652648925781 2023-01-21 12:25:33.397520: step: 1516/529, loss: 0.014473533257842064 2023-01-21 12:25:34.509408: step: 1520/529, loss: 0.0008525848388671875 2023-01-21 12:25:35.626546: step: 1524/529, loss: 0.0025352477096021175 2023-01-21 12:25:36.754645: step: 1528/529, loss: 0.2877357602119446 2023-01-21 12:25:37.864312: step: 1532/529, loss: 0.0002552032528910786 2023-01-21 12:25:38.997204: step: 1536/529, loss: 0.011604405008256435 2023-01-21 12:25:40.084437: step: 1540/529, loss: 0.0034202574752271175 2023-01-21 12:25:41.203241: step: 1544/529, loss: 0.063776396214962 2023-01-21 12:25:42.345126: step: 1548/529, loss: 0.01563110388815403 2023-01-21 12:25:43.492886: step: 1552/529, loss: 0.03626823425292969 2023-01-21 12:25:44.608160: step: 1556/529, loss: 0.10788936913013458 2023-01-21 12:25:45.730449: step: 1560/529, loss: 0.0006166458479128778 2023-01-21 12:25:46.872428: step: 1564/529, loss: 0.00020904542179778218 2023-01-21 12:25:47.973074: step: 1568/529, loss: 0.0024717331398278475 2023-01-21 12:25:49.110377: step: 1572/529, loss: 0.00015592575073242188 2023-01-21 12:25:50.267901: step: 1576/529, loss: 0.0013476371532306075 2023-01-21 12:25:51.365744: step: 1580/529, loss: 0.0695217102766037 2023-01-21 12:25:52.479911: step: 1584/529, loss: 0.04180781915783882 2023-01-21 12:25:53.586727: step: 1588/529, loss: 0.0023132325150072575 2023-01-21 12:25:54.708872: step: 1592/529, loss: 0.0003120899200439453 2023-01-21 12:25:55.844255: step: 1596/529, loss: 0.006851578131318092 2023-01-21 12:25:56.953282: step: 1600/529, loss: 0.010571003891527653 2023-01-21 12:25:58.074086: step: 1604/529, loss: 0.07600346207618713 2023-01-21 12:25:59.226032: step: 1608/529, loss: 0.05953731760382652 2023-01-21 12:26:00.330369: step: 1612/529, loss: 0.0007984161493368447 2023-01-21 12:26:01.430267: step: 1616/529, loss: 0.0001522690145066008 2023-01-21 12:26:02.529857: step: 1620/529, loss: 2.040863000729587e-05 2023-01-21 12:26:03.657282: step: 1624/529, loss: 0.0011659623123705387 2023-01-21 12:26:04.742681: step: 1628/529, loss: 0.00942907389253378 2023-01-21 12:26:05.843003: step: 1632/529, loss: 0.0026019096840173006 2023-01-21 12:26:06.968767: step: 1636/529, loss: 0.014038706198334694 2023-01-21 12:26:08.078637: step: 1640/529, loss: 0.0005350112915039062 2023-01-21 12:26:09.173169: step: 1644/529, loss: 4.720688230008818e-05 2023-01-21 12:26:10.294573: step: 1648/529, loss: 0.00210914621129632 2023-01-21 12:26:11.409137: step: 1652/529, loss: 0.005322504322975874 2023-01-21 12:26:12.488621: step: 1656/529, loss: 0.02371368370950222 2023-01-21 12:26:13.598365: step: 1660/529, loss: 0.00985736958682537 2023-01-21 12:26:14.693925: step: 1664/529, loss: 0.026445960626006126 2023-01-21 12:26:15.834124: step: 1668/529, loss: 0.0037639616057276726 2023-01-21 12:26:16.940208: step: 1672/529, loss: 0.017408180981874466 2023-01-21 12:26:18.073417: step: 1676/529, loss: 0.0005366325494833291 2023-01-21 12:26:19.217098: step: 1680/529, loss: 0.030685901641845703 2023-01-21 12:26:20.382975: step: 1684/529, loss: 0.034714438021183014 2023-01-21 12:26:21.501381: step: 1688/529, loss: 0.00029416085453704 2023-01-21 12:26:22.596670: step: 1692/529, loss: 0.0002092361537506804 2023-01-21 12:26:23.694499: step: 1696/529, loss: 0.0003733634948730469 2023-01-21 12:26:24.820501: step: 1700/529, loss: 0.044469740241765976 2023-01-21 12:26:25.931906: step: 1704/529, loss: 0.022730302065610886 2023-01-21 12:26:27.082928: step: 1708/529, loss: 0.00975952111184597 2023-01-21 12:26:28.197840: step: 1712/529, loss: 0.004509353544563055 2023-01-21 12:26:29.324210: step: 1716/529, loss: 0.022693252190947533 2023-01-21 12:26:30.423533: step: 1720/529, loss: 0.0021711348090320826 2023-01-21 12:26:31.555428: step: 1724/529, loss: 0.05210094898939133 2023-01-21 12:26:32.681748: step: 1728/529, loss: 0.03552999347448349 2023-01-21 12:26:33.776758: step: 1732/529, loss: 0.0009191512945108116 2023-01-21 12:26:34.908574: step: 1736/529, loss: 0.013871765695512295 2023-01-21 12:26:36.030003: step: 1740/529, loss: 1.697540210443549e-05 2023-01-21 12:26:37.129960: step: 1744/529, loss: 0.0002285003720317036 2023-01-21 12:26:38.233889: step: 1748/529, loss: 0.04652118682861328 2023-01-21 12:26:39.339444: step: 1752/529, loss: 0.002030658768489957 2023-01-21 12:26:40.471960: step: 1756/529, loss: 1.36113703250885 2023-01-21 12:26:41.589527: step: 1760/529, loss: 0.007953738793730736 2023-01-21 12:26:42.705377: step: 1764/529, loss: 2.8419495720299892e-05 2023-01-21 12:26:43.845069: step: 1768/529, loss: 0.023436356335878372 2023-01-21 12:26:45.001842: step: 1772/529, loss: 0.034502413123846054 2023-01-21 12:26:46.139575: step: 1776/529, loss: 0.04574594646692276 2023-01-21 12:26:47.253662: step: 1780/529, loss: 0.002615261124446988 2023-01-21 12:26:48.382555: step: 1784/529, loss: 0.004746842198073864 2023-01-21 12:26:49.506816: step: 1788/529, loss: 0.01699981838464737 2023-01-21 12:26:50.673677: step: 1792/529, loss: 0.050270941108465195 2023-01-21 12:26:51.785174: step: 1796/529, loss: 0.0001127243012888357 2023-01-21 12:26:52.895373: step: 1800/529, loss: 0.016205977648496628 2023-01-21 12:26:54.020566: step: 1804/529, loss: 0.0032377243041992188 2023-01-21 12:26:55.140349: step: 1808/529, loss: 0.005022430792450905 2023-01-21 12:26:56.247688: step: 1812/529, loss: 0.01888890378177166 2023-01-21 12:26:57.345942: step: 1816/529, loss: 0.015808487311005592 2023-01-21 12:26:58.429141: step: 1820/529, loss: 9.91821252682712e-06 2023-01-21 12:26:59.593633: step: 1824/529, loss: 0.0036518098786473274 2023-01-21 12:27:00.704152: step: 1828/529, loss: 0.016133643686771393 2023-01-21 12:27:01.834762: step: 1832/529, loss: 0.00464673014357686 2023-01-21 12:27:02.940935: step: 1836/529, loss: 0.004633140750229359 2023-01-21 12:27:04.040247: step: 1840/529, loss: 0.000640869140625 2023-01-21 12:27:05.161649: step: 1844/529, loss: 0.023274803534150124 2023-01-21 12:27:06.263543: step: 1848/529, loss: 0.009393502026796341 2023-01-21 12:27:07.360185: step: 1852/529, loss: 0.02821226231753826 2023-01-21 12:27:08.489101: step: 1856/529, loss: 0.001249122666195035 2023-01-21 12:27:09.647650: step: 1860/529, loss: 0.020393753424286842 2023-01-21 12:27:10.809414: step: 1864/529, loss: 0.046022988855838776 2023-01-21 12:27:11.916401: step: 1868/529, loss: 0.016793441027402878 2023-01-21 12:27:13.001445: step: 1872/529, loss: 0.001484871027059853 2023-01-21 12:27:14.124071: step: 1876/529, loss: 0.002045535948127508 2023-01-21 12:27:15.239979: step: 1880/529, loss: 0.009863662533462048 2023-01-21 12:27:16.382027: step: 1884/529, loss: 0.04506435617804527 2023-01-21 12:27:17.510692: step: 1888/529, loss: 0.011983584612607956 2023-01-21 12:27:18.621006: step: 1892/529, loss: 0.07662144303321838 2023-01-21 12:27:19.727863: step: 1896/529, loss: 0.0010465621016919613 2023-01-21 12:27:20.803682: step: 1900/529, loss: 0.0026866437401622534 2023-01-21 12:27:21.901160: step: 1904/529, loss: 0.01636953465640545 2023-01-21 12:27:23.003991: step: 1908/529, loss: 0.0008162498706951737 2023-01-21 12:27:24.103813: step: 1912/529, loss: 0.13790854811668396 2023-01-21 12:27:25.190057: step: 1916/529, loss: 0.00020818710618186742 2023-01-21 12:27:26.303773: step: 1920/529, loss: 0.0005504608270712197 2023-01-21 12:27:27.410404: step: 1924/529, loss: 0.008896350860595703 2023-01-21 12:27:28.512937: step: 1928/529, loss: 0.0002675056457519531 2023-01-21 12:27:29.633086: step: 1932/529, loss: 0.0005533218500204384 2023-01-21 12:27:30.732126: step: 1936/529, loss: 0.00042896269587799907 2023-01-21 12:27:31.898761: step: 1940/529, loss: 0.014613008126616478 2023-01-21 12:27:33.026609: step: 1944/529, loss: 0.004792308900505304 2023-01-21 12:27:34.125993: step: 1948/529, loss: 0.005675888154655695 2023-01-21 12:27:35.235157: step: 1952/529, loss: 0.013755322434008121 2023-01-21 12:27:36.341125: step: 1956/529, loss: 0.00043594837188720703 2023-01-21 12:27:37.461909: step: 1960/529, loss: 0.012760925106704235 2023-01-21 12:27:38.562101: step: 1964/529, loss: 9.32693510549143e-05 2023-01-21 12:27:39.669795: step: 1968/529, loss: 0.001598787377588451 2023-01-21 12:27:40.829997: step: 1972/529, loss: 0.01162738911807537 2023-01-21 12:27:41.968591: step: 1976/529, loss: 0.0074347020126879215 2023-01-21 12:27:43.070570: step: 1980/529, loss: 0.00021009446936659515 2023-01-21 12:27:44.216575: step: 1984/529, loss: 0.0006614684825763106 2023-01-21 12:27:45.302132: step: 1988/529, loss: 0.0036611557006835938 2023-01-21 12:27:46.428071: step: 1992/529, loss: 0.009977913461625576 2023-01-21 12:27:47.527084: step: 1996/529, loss: 0.022965241223573685 2023-01-21 12:27:48.645992: step: 2000/529, loss: 4.76837158203125e-05 2023-01-21 12:27:49.767927: step: 2004/529, loss: 0.036416102200746536 2023-01-21 12:27:50.886393: step: 2008/529, loss: 0.012509727850556374 2023-01-21 12:27:52.004348: step: 2012/529, loss: 0.00040836335392668843 2023-01-21 12:27:53.096981: step: 2016/529, loss: 2.861023403966101e-06 2023-01-21 12:27:54.253314: step: 2020/529, loss: 0.0002965927415061742 2023-01-21 12:27:55.377662: step: 2024/529, loss: 0.08842658996582031 2023-01-21 12:27:56.527823: step: 2028/529, loss: 0.0011669158702716231 2023-01-21 12:27:57.618804: step: 2032/529, loss: 0.00984115619212389 2023-01-21 12:27:58.751361: step: 2036/529, loss: 0.00340442662127316 2023-01-21 12:27:59.847615: step: 2040/529, loss: 0.0038094522897154093 2023-01-21 12:28:00.998899: step: 2044/529, loss: 0.0007734298706054688 2023-01-21 12:28:02.110541: step: 2048/529, loss: 0.0006843566661700606 2023-01-21 12:28:03.216006: step: 2052/529, loss: 0.00023689270892646164 2023-01-21 12:28:04.312927: step: 2056/529, loss: 0.00010099411883857101 2023-01-21 12:28:05.430603: step: 2060/529, loss: 2.708435022213962e-05 2023-01-21 12:28:06.541681: step: 2064/529, loss: 0.0001605987490620464 2023-01-21 12:28:07.655751: step: 2068/529, loss: 0.006698036100715399 2023-01-21 12:28:08.798562: step: 2072/529, loss: 0.005943012423813343 2023-01-21 12:28:09.950136: step: 2076/529, loss: 0.02304706536233425 2023-01-21 12:28:11.072740: step: 2080/529, loss: 0.09497890621423721 2023-01-21 12:28:12.224326: step: 2084/529, loss: 0.0004995345952920616 2023-01-21 12:28:13.343392: step: 2088/529, loss: 0.012242508120834827 2023-01-21 12:28:14.436151: step: 2092/529, loss: 0.018155433237552643 2023-01-21 12:28:15.519353: step: 2096/529, loss: 0.0022264004219323397 2023-01-21 12:28:16.623220: step: 2100/529, loss: 0.03890056908130646 2023-01-21 12:28:17.740907: step: 2104/529, loss: 0.014135170727968216 2023-01-21 12:28:18.839458: step: 2108/529, loss: 7.99179106252268e-05 2023-01-21 12:28:19.941511: step: 2112/529, loss: 1.6307831174344756e-05 2023-01-21 12:28:21.043199: step: 2116/529, loss: 0.004066276364028454 ================================================== Loss: 0.024 -------------------- Dev: {'event': {'p': 0.5803212851405622, 'r': 0.7696404793608522, 'f1': 0.661705781339439}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Test: {'event': {'p': 0.6334106728538283, 'r': 0.7703160270880361, 'f1': 0.6951871657754011}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Chinese: {'event': {'p': 0.550561797752809, 'r': 0.9074074074074074, 'f1': 0.6853146853146853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Korean: {'event': {'p': 0.5714285714285714, 'r': 0.5079365079365079, 'f1': 0.5378151260504201}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Russian: {'event': {'p': 0.4523809523809524, 'r': 0.5277777777777778, 'f1': 0.4871794871794871}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 23 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:29:01.711309: step: 4/529, loss: 0.0002883911074604839 2023-01-21 12:29:02.838519: step: 8/529, loss: 0.03681345283985138 2023-01-21 12:29:03.953434: step: 12/529, loss: 0.0006464004400186241 2023-01-21 12:29:05.067870: step: 16/529, loss: 0.006453132722526789 2023-01-21 12:29:06.193868: step: 20/529, loss: 0.0021348954178392887 2023-01-21 12:29:07.346285: step: 24/529, loss: 0.0010612488258630037 2023-01-21 12:29:08.435987: step: 28/529, loss: 0.013253974728286266 2023-01-21 12:29:09.537413: step: 32/529, loss: 9.374618821311742e-05 2023-01-21 12:29:10.632406: step: 36/529, loss: 0.013882351107895374 2023-01-21 12:29:11.773356: step: 40/529, loss: 0.00047893525334075093 2023-01-21 12:29:12.888803: step: 44/529, loss: 0.0007560729864053428 2023-01-21 12:29:14.004350: step: 48/529, loss: 0.0011245727073401213 2023-01-21 12:29:15.089712: step: 52/529, loss: 0.004950141999870539 2023-01-21 12:29:16.234996: step: 56/529, loss: 3.4332279028603807e-06 2023-01-21 12:29:17.414196: step: 60/529, loss: 1.0315948724746704 2023-01-21 12:29:18.544296: step: 64/529, loss: 0.050814438611269 2023-01-21 12:29:19.675512: step: 68/529, loss: 0.0006988048553466797 2023-01-21 12:29:20.784336: step: 72/529, loss: 0.0038265229668468237 2023-01-21 12:29:21.884371: step: 76/529, loss: 0.00022995471954345703 2023-01-21 12:29:23.019040: step: 80/529, loss: 0.0007982254028320312 2023-01-21 12:29:24.147633: step: 84/529, loss: 6.103516170696821e-06 2023-01-21 12:29:25.233572: step: 88/529, loss: 7.581710815429688e-05 2023-01-21 12:29:26.378836: step: 92/529, loss: 0.0014209747314453125 2023-01-21 12:29:27.497354: step: 96/529, loss: 0.00044593808706849813 2023-01-21 12:29:28.607728: step: 100/529, loss: 0.0033233645372092724 2023-01-21 12:29:29.741664: step: 104/529, loss: 0.00080108642578125 2023-01-21 12:29:30.844863: step: 108/529, loss: 0.03795995935797691 2023-01-21 12:29:31.963100: step: 112/529, loss: 0.020905114710330963 2023-01-21 12:29:33.058130: step: 116/529, loss: 0.0023916244972497225 2023-01-21 12:29:34.171324: step: 120/529, loss: 0.00012936592975165695 2023-01-21 12:29:35.292654: step: 124/529, loss: 0.0007274627569131553 2023-01-21 12:29:36.416351: step: 128/529, loss: 0.012909985147416592 2023-01-21 12:29:37.515259: step: 132/529, loss: 0.00016303063603118062 2023-01-21 12:29:38.641733: step: 136/529, loss: 0.03207588195800781 2023-01-21 12:29:39.742679: step: 140/529, loss: 0.00023555755615234375 2023-01-21 12:29:40.864527: step: 144/529, loss: 0.00034775736276060343 2023-01-21 12:29:41.990718: step: 148/529, loss: 0.0015834808582440019 2023-01-21 12:29:43.115575: step: 152/529, loss: 0.0014513017376884818 2023-01-21 12:29:44.253962: step: 156/529, loss: 0.0654090866446495 2023-01-21 12:29:45.369280: step: 160/529, loss: 0.0019367217319086194 2023-01-21 12:29:46.493093: step: 164/529, loss: 0.03619079664349556 2023-01-21 12:29:47.638798: step: 168/529, loss: 0.03627815470099449 2023-01-21 12:29:48.757433: step: 172/529, loss: 0.00168952951207757 2023-01-21 12:29:49.872984: step: 176/529, loss: 0.00041542056715115905 2023-01-21 12:29:50.984449: step: 180/529, loss: 0.0015731812454760075 2023-01-21 12:29:52.097088: step: 184/529, loss: 0.010916901752352715 2023-01-21 12:29:53.178428: step: 188/529, loss: 0.054041292518377304 2023-01-21 12:29:54.281155: step: 192/529, loss: 0.018085859715938568 2023-01-21 12:29:55.424312: step: 196/529, loss: 0.035671282559633255 2023-01-21 12:29:56.527521: step: 200/529, loss: 0.05806489288806915 2023-01-21 12:29:57.656405: step: 204/529, loss: 0.0025993348099291325 2023-01-21 12:29:58.736880: step: 208/529, loss: 3.6239625842426904e-06 2023-01-21 12:29:59.848580: step: 212/529, loss: 0.0005254745483398438 2023-01-21 12:30:00.979289: step: 216/529, loss: 0.004301833920180798 2023-01-21 12:30:02.119922: step: 220/529, loss: 0.0010030746925622225 2023-01-21 12:30:03.217940: step: 224/529, loss: 1.9073486328125e-06 2023-01-21 12:30:04.344905: step: 228/529, loss: 0.06467180699110031 2023-01-21 12:30:05.462048: step: 232/529, loss: 0.004024696536362171 2023-01-21 12:30:06.574887: step: 236/529, loss: 0.006682396400719881 2023-01-21 12:30:07.660053: step: 240/529, loss: 0.0002454012574162334 2023-01-21 12:30:08.745829: step: 244/529, loss: 0.00267715472728014 2023-01-21 12:30:09.880808: step: 248/529, loss: 0.012624263763427734 2023-01-21 12:30:11.042021: step: 252/529, loss: 0.00028333664522506297 2023-01-21 12:30:12.132847: step: 256/529, loss: 0.034265995025634766 2023-01-21 12:30:13.249158: step: 260/529, loss: 6.12258882028982e-05 2023-01-21 12:30:14.343968: step: 264/529, loss: 8.745193190407008e-05 2023-01-21 12:30:15.472134: step: 268/529, loss: 0.03480052947998047 2023-01-21 12:30:16.573798: step: 272/529, loss: 0.003998184110969305 2023-01-21 12:30:17.666842: step: 276/529, loss: 0.012942028231918812 2023-01-21 12:30:18.794319: step: 280/529, loss: 0.009005356580018997 2023-01-21 12:30:19.908825: step: 284/529, loss: 0.02213153801858425 2023-01-21 12:30:21.027033: step: 288/529, loss: 0.000880813633557409 2023-01-21 12:30:22.123013: step: 292/529, loss: 0.0037752152420580387 2023-01-21 12:30:23.258198: step: 296/529, loss: 0.0006952285766601562 2023-01-21 12:30:24.382812: step: 300/529, loss: 4.129410081077367e-05 2023-01-21 12:30:25.501209: step: 304/529, loss: 0.0005508422618731856 2023-01-21 12:30:26.640423: step: 308/529, loss: 0.058629799634218216 2023-01-21 12:30:27.775041: step: 312/529, loss: 0.017781639471650124 2023-01-21 12:30:28.860192: step: 316/529, loss: 0.0012475013500079513 2023-01-21 12:30:30.001095: step: 320/529, loss: 2.2459029423771426e-05 2023-01-21 12:30:31.094217: step: 324/529, loss: 0.02574281580746174 2023-01-21 12:30:32.212119: step: 328/529, loss: 0.0009464264148846269 2023-01-21 12:30:33.335041: step: 332/529, loss: 0.01752842217683792 2023-01-21 12:30:34.444231: step: 336/529, loss: 0.10592098534107208 2023-01-21 12:30:35.554059: step: 340/529, loss: 0.006802177522331476 2023-01-21 12:30:36.701744: step: 344/529, loss: 0.00028142929659225047 2023-01-21 12:30:37.823007: step: 348/529, loss: 0.03823566436767578 2023-01-21 12:30:38.930969: step: 352/529, loss: 0.0006147384410724044 2023-01-21 12:30:40.067112: step: 356/529, loss: 0.00023174287343863398 2023-01-21 12:30:41.190707: step: 360/529, loss: 0.00016736984252929688 2023-01-21 12:30:42.304390: step: 364/529, loss: 0.010930824093520641 2023-01-21 12:30:43.435906: step: 368/529, loss: 0.002102422760799527 2023-01-21 12:30:44.576678: step: 372/529, loss: 0.0004864692746195942 2023-01-21 12:30:45.704157: step: 376/529, loss: 0.006289339158684015 2023-01-21 12:30:46.886105: step: 380/529, loss: 0.003525686217471957 2023-01-21 12:30:48.009435: step: 384/529, loss: 0.003340149065479636 2023-01-21 12:30:49.115141: step: 388/529, loss: 0.0002739906485658139 2023-01-21 12:30:50.259744: step: 392/529, loss: 0.023203223943710327 2023-01-21 12:30:51.373290: step: 396/529, loss: 0.0090522775426507 2023-01-21 12:30:52.451863: step: 400/529, loss: 0.000854110752698034 2023-01-21 12:30:53.561371: step: 404/529, loss: 0.00061883928719908 2023-01-21 12:30:54.668561: step: 408/529, loss: 0.0019355774857103825 2023-01-21 12:30:55.795902: step: 412/529, loss: 0.005366802215576172 2023-01-21 12:30:56.889630: step: 416/529, loss: 0.028545141220092773 2023-01-21 12:30:58.047483: step: 420/529, loss: 0.003251934191212058 2023-01-21 12:30:59.124351: step: 424/529, loss: 0.00029354094294831157 2023-01-21 12:31:00.252781: step: 428/529, loss: 0.005344772711396217 2023-01-21 12:31:01.364075: step: 432/529, loss: 0.0008024215931072831 2023-01-21 12:31:02.453893: step: 436/529, loss: 0.03718223795294762 2023-01-21 12:31:03.614284: step: 440/529, loss: 0.43041613698005676 2023-01-21 12:31:04.724951: step: 444/529, loss: 0.001266944338567555 2023-01-21 12:31:05.855908: step: 448/529, loss: 0.004234886262565851 2023-01-21 12:31:06.991661: step: 452/529, loss: 0.003745937254279852 2023-01-21 12:31:08.128819: step: 456/529, loss: 0.008444547653198242 2023-01-21 12:31:09.240087: step: 460/529, loss: 0.00043907167855650187 2023-01-21 12:31:10.359099: step: 464/529, loss: 0.00045180320739746094 2023-01-21 12:31:11.469376: step: 468/529, loss: 7.42912307032384e-05 2023-01-21 12:31:12.596928: step: 472/529, loss: 0.0066436766646802425 2023-01-21 12:31:13.701281: step: 476/529, loss: 0.02650442160665989 2023-01-21 12:31:14.833035: step: 480/529, loss: 0.00038185121957212687 2023-01-21 12:31:15.924675: step: 484/529, loss: 0.0009094238630495965 2023-01-21 12:31:17.064399: step: 488/529, loss: 0.0017925500869750977 2023-01-21 12:31:18.178526: step: 492/529, loss: 0.0003252029709983617 2023-01-21 12:31:19.281783: step: 496/529, loss: 0.006175899412482977 2023-01-21 12:31:20.423614: step: 500/529, loss: 0.001824235892854631 2023-01-21 12:31:21.510605: step: 504/529, loss: 0.00011224746413063258 2023-01-21 12:31:22.569330: step: 508/529, loss: 0.00032939910306595266 2023-01-21 12:31:23.715495: step: 512/529, loss: 0.0002157211274607107 2023-01-21 12:31:24.816737: step: 516/529, loss: 0.0014985561138018966 2023-01-21 12:31:25.948034: step: 520/529, loss: 0.00955963134765625 2023-01-21 12:31:27.051043: step: 524/529, loss: 0.006502533331513405 2023-01-21 12:31:28.161051: step: 528/529, loss: 0.07747097313404083 2023-01-21 12:31:29.296628: step: 532/529, loss: 2.5224686396541074e-05 2023-01-21 12:31:30.406557: step: 536/529, loss: 0.008847285062074661 2023-01-21 12:31:31.534575: step: 540/529, loss: 0.011590957641601562 2023-01-21 12:31:32.649798: step: 544/529, loss: 0.00023403167142532766 2023-01-21 12:31:33.775847: step: 548/529, loss: 0.00014133454533293843 2023-01-21 12:31:34.869622: step: 552/529, loss: 0.0001375198335153982 2023-01-21 12:31:35.976948: step: 556/529, loss: 0.0002899169921875 2023-01-21 12:31:37.083161: step: 560/529, loss: 0.012487697415053844 2023-01-21 12:31:38.205021: step: 564/529, loss: 0.010420609265565872 2023-01-21 12:31:39.320531: step: 568/529, loss: 7.028579420875758e-05 2023-01-21 12:31:40.418601: step: 572/529, loss: 0.049776457250118256 2023-01-21 12:31:41.536528: step: 576/529, loss: 0.00022802352032158524 2023-01-21 12:31:42.629116: step: 580/529, loss: 9.822845640883315e-06 2023-01-21 12:31:43.715006: step: 584/529, loss: 0.0054267882369458675 2023-01-21 12:31:44.844530: step: 588/529, loss: 0.04016227647662163 2023-01-21 12:31:45.985981: step: 592/529, loss: 5.340576535672881e-05 2023-01-21 12:31:47.090924: step: 596/529, loss: 0.003691291669383645 2023-01-21 12:31:48.188392: step: 600/529, loss: 0.0008340835920535028 2023-01-21 12:31:49.294926: step: 604/529, loss: 0.00025072097196243703 2023-01-21 12:31:50.400224: step: 608/529, loss: 0.03420591354370117 2023-01-21 12:31:51.522195: step: 612/529, loss: 0.011543131433427334 2023-01-21 12:31:52.640437: step: 616/529, loss: 0.04497117921710014 2023-01-21 12:31:53.739535: step: 620/529, loss: 0.000164031982421875 2023-01-21 12:31:54.837963: step: 624/529, loss: 0.010802173987030983 2023-01-21 12:31:55.970222: step: 628/529, loss: 0.012547683902084827 2023-01-21 12:31:57.073301: step: 632/529, loss: 0.00019340515427757055 2023-01-21 12:31:58.217065: step: 636/529, loss: 0.0006591796991415322 2023-01-21 12:31:59.324084: step: 640/529, loss: 0.012115002609789371 2023-01-21 12:32:00.446217: step: 644/529, loss: 0.004157543648034334 2023-01-21 12:32:01.580128: step: 648/529, loss: 2.7179717108083423e-06 2023-01-21 12:32:02.771738: step: 652/529, loss: 0.03401994705200195 2023-01-21 12:32:03.886115: step: 656/529, loss: 8.58306884765625e-05 2023-01-21 12:32:05.036371: step: 660/529, loss: 0.003197288606315851 2023-01-21 12:32:06.165126: step: 664/529, loss: 0.055520057678222656 2023-01-21 12:32:07.259182: step: 668/529, loss: 0.01597623899579048 2023-01-21 12:32:08.350453: step: 672/529, loss: 0.06323447078466415 2023-01-21 12:32:09.456737: step: 676/529, loss: 2.0408631826285273e-05 2023-01-21 12:32:10.590199: step: 680/529, loss: 3.0517578125e-05 2023-01-21 12:32:11.707773: step: 684/529, loss: 0.03477153927087784 2023-01-21 12:32:12.821332: step: 688/529, loss: 8.378028724109754e-05 2023-01-21 12:32:13.986998: step: 692/529, loss: 0.0023954391945153475 2023-01-21 12:32:15.096065: step: 696/529, loss: 2.593994213384576e-05 2023-01-21 12:32:16.204796: step: 700/529, loss: 0.3161248564720154 2023-01-21 12:32:17.378112: step: 704/529, loss: 1.296997106692288e-05 2023-01-21 12:32:18.498136: step: 708/529, loss: 0.0002456665097270161 2023-01-21 12:32:19.572874: step: 712/529, loss: 0.0005173683166503906 2023-01-21 12:32:20.680415: step: 716/529, loss: 0.0027935027610510588 2023-01-21 12:32:21.773862: step: 720/529, loss: 0.009577560238540173 2023-01-21 12:32:22.914714: step: 724/529, loss: 0.0449712760746479 2023-01-21 12:32:24.015878: step: 728/529, loss: 0.006515884771943092 2023-01-21 12:32:25.101094: step: 732/529, loss: 0.00012111664545955136 2023-01-21 12:32:26.195603: step: 736/529, loss: 0.0007436752202920616 2023-01-21 12:32:27.283605: step: 740/529, loss: 0.005751228425651789 2023-01-21 12:32:28.415110: step: 744/529, loss: 0.012774658389389515 2023-01-21 12:32:29.547645: step: 748/529, loss: 0.015220833010971546 2023-01-21 12:32:30.637670: step: 752/529, loss: 0.005015754606574774 2023-01-21 12:32:31.775494: step: 756/529, loss: 0.0040826802141964436 2023-01-21 12:32:32.876438: step: 760/529, loss: 0.0005961417919024825 2023-01-21 12:32:33.980115: step: 764/529, loss: 0.0018270493019372225 2023-01-21 12:32:35.110250: step: 768/529, loss: 0.00012331009202171117 2023-01-21 12:32:36.202072: step: 772/529, loss: 0.018587494269013405 2023-01-21 12:32:37.304043: step: 776/529, loss: 0.00492439279332757 2023-01-21 12:32:38.442643: step: 780/529, loss: 0.00092992780264467 2023-01-21 12:32:39.565919: step: 784/529, loss: 0.0021121979225426912 2023-01-21 12:32:40.672048: step: 788/529, loss: 1.850128137448337e-05 2023-01-21 12:32:41.823118: step: 792/529, loss: -2.2411345526052173e-06 2023-01-21 12:32:42.936756: step: 796/529, loss: 0.00023880005755927414 2023-01-21 12:32:44.046171: step: 800/529, loss: 0.03480348363518715 2023-01-21 12:32:45.186429: step: 804/529, loss: -2.47955322265625e-05 2023-01-21 12:32:46.298167: step: 808/529, loss: 0.0057960511185228825 2023-01-21 12:32:47.426039: step: 812/529, loss: 0.0013547898270189762 2023-01-21 12:32:48.554696: step: 816/529, loss: 0.014643669128417969 2023-01-21 12:32:49.656988: step: 820/529, loss: 0.03941555321216583 2023-01-21 12:32:50.803909: step: 824/529, loss: 0.035974979400634766 2023-01-21 12:32:51.899339: step: 828/529, loss: 0.0013183593982830644 2023-01-21 12:32:53.026579: step: 832/529, loss: 0.0019962310325354338 2023-01-21 12:32:54.147891: step: 836/529, loss: 0.0009966850047931075 2023-01-21 12:32:55.256539: step: 840/529, loss: 0.0027654648292809725 2023-01-21 12:32:56.370147: step: 844/529, loss: 2.584457615739666e-05 2023-01-21 12:32:57.482608: step: 848/529, loss: 0.00047864916268736124 2023-01-21 12:32:58.582908: step: 852/529, loss: 0.0004040718195028603 2023-01-21 12:32:59.688614: step: 856/529, loss: 0.025358200073242188 2023-01-21 12:33:00.816288: step: 860/529, loss: 1.640319896978326e-05 2023-01-21 12:33:01.943836: step: 864/529, loss: 0.001890850136987865 2023-01-21 12:33:03.061980: step: 868/529, loss: 0.008761894889175892 2023-01-21 12:33:04.150937: step: 872/529, loss: 0.005912589840590954 2023-01-21 12:33:05.308909: step: 876/529, loss: 0.1778927743434906 2023-01-21 12:33:06.424513: step: 880/529, loss: 2.140998731192667e-05 2023-01-21 12:33:07.527607: step: 884/529, loss: 0.004208373837172985 2023-01-21 12:33:08.636392: step: 888/529, loss: 0.04290733486413956 2023-01-21 12:33:09.784583: step: 892/529, loss: 0.0015522956382483244 2023-01-21 12:33:10.906841: step: 896/529, loss: 0.0017154216766357422 2023-01-21 12:33:12.022557: step: 900/529, loss: 0.005001259036362171 2023-01-21 12:33:13.148351: step: 904/529, loss: 8.58306884765625e-05 2023-01-21 12:33:14.289795: step: 908/529, loss: 0.02731633186340332 2023-01-21 12:33:15.444562: step: 912/529, loss: 0.0008234024280682206 2023-01-21 12:33:16.544744: step: 916/529, loss: 0.0015952109824866056 2023-01-21 12:33:17.688678: step: 920/529, loss: 0.006677437108010054 2023-01-21 12:33:18.786336: step: 924/529, loss: 9.059906005859375e-06 2023-01-21 12:33:19.891789: step: 928/529, loss: 0.0006336212391033769 2023-01-21 12:33:21.066483: step: 932/529, loss: 0.007230567745864391 2023-01-21 12:33:22.197510: step: 936/529, loss: 0.0015255927573889494 2023-01-21 12:33:23.290747: step: 940/529, loss: 0.06697139889001846 2023-01-21 12:33:24.400876: step: 944/529, loss: 0.004878711886703968 2023-01-21 12:33:25.502530: step: 948/529, loss: 2.365112231927924e-05 2023-01-21 12:33:26.609697: step: 952/529, loss: 0.004432773683220148 2023-01-21 12:33:27.761760: step: 956/529, loss: 0.0011806488037109375 2023-01-21 12:33:28.863033: step: 960/529, loss: 0.011307192035019398 2023-01-21 12:33:29.994317: step: 964/529, loss: 0.003016758244484663 2023-01-21 12:33:31.110736: step: 968/529, loss: 0.05561990663409233 2023-01-21 12:33:32.218889: step: 972/529, loss: 0.053038597106933594 2023-01-21 12:33:33.325334: step: 976/529, loss: 0.0059722899459302425 2023-01-21 12:33:34.435285: step: 980/529, loss: 0.0032155991066247225 2023-01-21 12:33:35.533536: step: 984/529, loss: 0.0002642631880007684 2023-01-21 12:33:36.645018: step: 988/529, loss: 0.000899887119885534 2023-01-21 12:33:37.796248: step: 992/529, loss: 0.0380154624581337 2023-01-21 12:33:38.934335: step: 996/529, loss: 0.0031909942626953125 2023-01-21 12:33:40.070671: step: 1000/529, loss: 0.018430233001708984 2023-01-21 12:33:41.157584: step: 1004/529, loss: 0.004919052589684725 2023-01-21 12:33:42.298588: step: 1008/529, loss: 0.00019683838763739914 2023-01-21 12:33:43.392166: step: 1012/529, loss: 0.01740889437496662 2023-01-21 12:33:44.512275: step: 1016/529, loss: 0.008913611993193626 2023-01-21 12:33:45.624683: step: 1020/529, loss: 0.09894199669361115 2023-01-21 12:33:46.728151: step: 1024/529, loss: 0.03086087480187416 2023-01-21 12:33:47.851761: step: 1028/529, loss: 0.0005334853776730597 2023-01-21 12:33:48.931554: step: 1032/529, loss: 0.0055595398880541325 2023-01-21 12:33:50.058244: step: 1036/529, loss: 4.243851071805693e-05 2023-01-21 12:33:51.203287: step: 1040/529, loss: 0.008710193447768688 2023-01-21 12:33:52.327867: step: 1044/529, loss: 0.012061690911650658 2023-01-21 12:33:53.441621: step: 1048/529, loss: 0.00426902761682868 2023-01-21 12:33:54.556171: step: 1052/529, loss: 0.002290916396304965 2023-01-21 12:33:55.682353: step: 1056/529, loss: 0.0005729675758630037 2023-01-21 12:33:56.813028: step: 1060/529, loss: 0.005400658119469881 2023-01-21 12:33:57.896452: step: 1064/529, loss: 0.0010809899540618062 2023-01-21 12:33:59.000328: step: 1068/529, loss: 0.0001702308509266004 2023-01-21 12:34:00.109013: step: 1072/529, loss: 0.00017502308764960617 2023-01-21 12:34:01.229308: step: 1076/529, loss: 0.0005151748773641884 2023-01-21 12:34:02.360146: step: 1080/529, loss: 0.0039052963256835938 2023-01-21 12:34:03.497487: step: 1084/529, loss: 0.00012598038301803172 2023-01-21 12:34:04.601939: step: 1088/529, loss: 0.027127837762236595 2023-01-21 12:34:05.773133: step: 1092/529, loss: 0.029183007776737213 2023-01-21 12:34:06.885618: step: 1096/529, loss: 0.004023742862045765 2023-01-21 12:34:08.027351: step: 1100/529, loss: 0.0034817694686353207 2023-01-21 12:34:09.131775: step: 1104/529, loss: 0.0076423645950853825 2023-01-21 12:34:10.240856: step: 1108/529, loss: 0.00019702911959029734 2023-01-21 12:34:11.355429: step: 1112/529, loss: -2.441406286379788e-05 2023-01-21 12:34:12.494535: step: 1116/529, loss: 0.0019440650939941406 2023-01-21 12:34:13.589501: step: 1120/529, loss: 0.0003028869687113911 2023-01-21 12:34:14.691830: step: 1124/529, loss: 0.10020647197961807 2023-01-21 12:34:15.795877: step: 1128/529, loss: 0.04185142740607262 2023-01-21 12:34:16.891413: step: 1132/529, loss: 0.005414009094238281 2023-01-21 12:34:17.993631: step: 1136/529, loss: 0.0011066437000408769 2023-01-21 12:34:19.129563: step: 1140/529, loss: 0.000976133334916085 2023-01-21 12:34:20.257907: step: 1144/529, loss: 0.0117934700101614 2023-01-21 12:34:21.361869: step: 1148/529, loss: 0.013056183233857155 2023-01-21 12:34:22.476573: step: 1152/529, loss: 0.0009434462408535182 2023-01-21 12:34:23.589719: step: 1156/529, loss: 0.00021486282639671117 2023-01-21 12:34:24.717599: step: 1160/529, loss: 0.0484987273812294 2023-01-21 12:34:25.859749: step: 1164/529, loss: 0.035813476890325546 2023-01-21 12:34:26.958033: step: 1168/529, loss: 0.03295135498046875 2023-01-21 12:34:28.112207: step: 1172/529, loss: 0.0008213043329305947 2023-01-21 12:34:29.224907: step: 1176/529, loss: 0.0531865693628788 2023-01-21 12:34:30.324178: step: 1180/529, loss: 0.002627754118293524 2023-01-21 12:34:31.466180: step: 1184/529, loss: 0.12368451058864594 2023-01-21 12:34:32.574926: step: 1188/529, loss: 0.00022735596576239914 2023-01-21 12:34:33.703007: step: 1192/529, loss: 0.004833448212593794 2023-01-21 12:34:34.846743: step: 1196/529, loss: 0.029396915808320045 2023-01-21 12:34:35.980553: step: 1200/529, loss: 0.003652954241260886 2023-01-21 12:34:37.105949: step: 1204/529, loss: 3.894728660583496 2023-01-21 12:34:38.233179: step: 1208/529, loss: 0.0001337051362497732 2023-01-21 12:34:39.360329: step: 1212/529, loss: 0.0005608559004031122 2023-01-21 12:34:40.479667: step: 1216/529, loss: 0.0067729949951171875 2023-01-21 12:34:41.595040: step: 1220/529, loss: 0.044348813593387604 2023-01-21 12:34:42.692064: step: 1224/529, loss: 0.0035758973099291325 2023-01-21 12:34:43.793574: step: 1228/529, loss: 0.0009157180902548134 2023-01-21 12:34:44.917209: step: 1232/529, loss: 0.0061286925338208675 2023-01-21 12:34:46.000380: step: 1236/529, loss: 0.0019105911487713456 2023-01-21 12:34:47.099754: step: 1240/529, loss: 0.0036661147605627775 2023-01-21 12:34:48.198906: step: 1244/529, loss: 0.0006690978771075606 2023-01-21 12:34:49.339989: step: 1248/529, loss: 0.05568714439868927 2023-01-21 12:34:50.514554: step: 1252/529, loss: 0.021386336535215378 2023-01-21 12:34:51.627886: step: 1256/529, loss: 0.003376817563548684 2023-01-21 12:34:52.737055: step: 1260/529, loss: 0.0014566421741619706 2023-01-21 12:34:53.832518: step: 1264/529, loss: 0.00011391640873625875 2023-01-21 12:34:54.948585: step: 1268/529, loss: 0.001253414200618863 2023-01-21 12:34:56.063710: step: 1272/529, loss: 0.0007602691766805947 2023-01-21 12:34:57.172490: step: 1276/529, loss: 0.0004299640713725239 2023-01-21 12:34:58.287724: step: 1280/529, loss: 0.009141732007265091 2023-01-21 12:34:59.429495: step: 1284/529, loss: 0.09636344760656357 2023-01-21 12:35:00.548914: step: 1288/529, loss: 0.06481647491455078 2023-01-21 12:35:01.682542: step: 1292/529, loss: 0.002770710038021207 2023-01-21 12:35:02.823706: step: 1296/529, loss: 9.069443331100047e-05 2023-01-21 12:35:03.933484: step: 1300/529, loss: 0.018890762701630592 2023-01-21 12:35:05.018329: step: 1304/529, loss: 0.00021400452533271164 2023-01-21 12:35:06.154042: step: 1308/529, loss: 0.002646541688591242 2023-01-21 12:35:07.281433: step: 1312/529, loss: 0.0012115478748455644 2023-01-21 12:35:08.382456: step: 1316/529, loss: 0.0699225440621376 2023-01-21 12:35:09.494017: step: 1320/529, loss: 7.286072650458664e-05 2023-01-21 12:35:10.580108: step: 1324/529, loss: 0.05436668545007706 2023-01-21 12:35:11.668121: step: 1328/529, loss: 0.0004395008145365864 2023-01-21 12:35:12.785428: step: 1332/529, loss: 0.0006276130443438888 2023-01-21 12:35:13.886707: step: 1336/529, loss: 0.013247775845229626 2023-01-21 12:35:14.983654: step: 1340/529, loss: 0.0005140304565429688 2023-01-21 12:35:16.089334: step: 1344/529, loss: 0.0002480507246218622 2023-01-21 12:35:17.210769: step: 1348/529, loss: 0.0017969131004065275 2023-01-21 12:35:18.339180: step: 1352/529, loss: 0.0001752853422658518 2023-01-21 12:35:19.464837: step: 1356/529, loss: 0.0019431114196777344 2023-01-21 12:35:20.590553: step: 1360/529, loss: 0.008020590990781784 2023-01-21 12:35:21.710180: step: 1364/529, loss: 0.0009081840398721397 2023-01-21 12:35:22.839920: step: 1368/529, loss: 0.022505568340420723 2023-01-21 12:35:23.943742: step: 1372/529, loss: 2.365112231927924e-05 2023-01-21 12:35:25.058441: step: 1376/529, loss: 0.0006483554607257247 2023-01-21 12:35:26.164573: step: 1380/529, loss: 0.0015235661994665861 2023-01-21 12:35:27.306339: step: 1384/529, loss: 0.0006504058837890625 2023-01-21 12:35:28.421286: step: 1388/529, loss: 0.0016690255142748356 2023-01-21 12:35:29.538335: step: 1392/529, loss: 0.020243167877197266 2023-01-21 12:35:30.634477: step: 1396/529, loss: 0.0031223297119140625 2023-01-21 12:35:31.741928: step: 1400/529, loss: 0.0007585525745525956 2023-01-21 12:35:32.872487: step: 1404/529, loss: 0.0008188247447833419 2023-01-21 12:35:33.987113: step: 1408/529, loss: 1.52587890625e-05 2023-01-21 12:35:35.159633: step: 1412/529, loss: 0.000181102761416696 2023-01-21 12:35:36.274691: step: 1416/529, loss: 0.029534244909882545 2023-01-21 12:35:37.358544: step: 1420/529, loss: 0.0017442703247070312 2023-01-21 12:35:38.478516: step: 1424/529, loss: 0.027648448944091797 2023-01-21 12:35:39.613902: step: 1428/529, loss: 0.002980804303660989 2023-01-21 12:35:40.691149: step: 1432/529, loss: 1.8596649169921875e-05 2023-01-21 12:35:41.787453: step: 1436/529, loss: 0.0008102416759356856 2023-01-21 12:35:42.917132: step: 1440/529, loss: 0.00788583792746067 2023-01-21 12:35:44.033888: step: 1444/529, loss: 2.574920654296875e-05 2023-01-21 12:35:45.130179: step: 1448/529, loss: 0.00309278373606503 2023-01-21 12:35:46.257562: step: 1452/529, loss: 0.003055000212043524 2023-01-21 12:35:47.391832: step: 1456/529, loss: 0.0030759810470044613 2023-01-21 12:35:48.502045: step: 1460/529, loss: 0.009456634521484375 2023-01-21 12:35:49.618250: step: 1464/529, loss: 0.06446877121925354 2023-01-21 12:35:50.741406: step: 1468/529, loss: 0.00015506744966842234 2023-01-21 12:35:51.869810: step: 1472/529, loss: 0.026921462267637253 2023-01-21 12:35:52.990792: step: 1476/529, loss: 0.3829685151576996 2023-01-21 12:35:54.093579: step: 1480/529, loss: 7.04765334376134e-05 2023-01-21 12:35:55.218681: step: 1484/529, loss: 7.925033423816785e-05 2023-01-21 12:35:56.324698: step: 1488/529, loss: 0.03731689602136612 2023-01-21 12:35:57.428824: step: 1492/529, loss: 0.0006093978881835938 2023-01-21 12:35:58.531740: step: 1496/529, loss: 0.07322235405445099 2023-01-21 12:35:59.647696: step: 1500/529, loss: 0.065831758081913 2023-01-21 12:36:00.766511: step: 1504/529, loss: 0.0005354881286621094 2023-01-21 12:36:01.865189: step: 1508/529, loss: 0.0026178359985351562 2023-01-21 12:36:02.994813: step: 1512/529, loss: 0.11140112578868866 2023-01-21 12:36:04.112766: step: 1516/529, loss: 0.001960277557373047 2023-01-21 12:36:05.247787: step: 1520/529, loss: 0.700862467288971 2023-01-21 12:36:06.364411: step: 1524/529, loss: 0.0034385682083666325 2023-01-21 12:36:07.471181: step: 1528/529, loss: 0.022772978991270065 2023-01-21 12:36:08.601763: step: 1532/529, loss: 0.006360912229865789 2023-01-21 12:36:09.728539: step: 1536/529, loss: 1.8978118532686494e-05 2023-01-21 12:36:10.834532: step: 1540/529, loss: 0.00010452271089889109 2023-01-21 12:36:11.952735: step: 1544/529, loss: 0.013959693722426891 2023-01-21 12:36:13.081981: step: 1548/529, loss: 0.006504154298454523 2023-01-21 12:36:14.196553: step: 1552/529, loss: 0.045899201184511185 2023-01-21 12:36:15.317101: step: 1556/529, loss: 0.016573715955018997 2023-01-21 12:36:16.432530: step: 1560/529, loss: 0.002582299755886197 2023-01-21 12:36:17.538492: step: 1564/529, loss: 0.004678439814597368 2023-01-21 12:36:18.654589: step: 1568/529, loss: -4.00543194700731e-06 2023-01-21 12:36:19.784339: step: 1572/529, loss: 0.00033626556978560984 2023-01-21 12:36:20.899275: step: 1576/529, loss: 0.002205467317253351 2023-01-21 12:36:22.021166: step: 1580/529, loss: 0.0002117156982421875 2023-01-21 12:36:23.125578: step: 1584/529, loss: 1.52587890625e-05 2023-01-21 12:36:24.231898: step: 1588/529, loss: 0.0009866714244708419 2023-01-21 12:36:25.336431: step: 1592/529, loss: 0.023676397278904915 2023-01-21 12:36:26.448328: step: 1596/529, loss: 0.0067746639251708984 2023-01-21 12:36:27.562659: step: 1600/529, loss: 0.02019519731402397 2023-01-21 12:36:28.682833: step: 1604/529, loss: 0.04520416632294655 2023-01-21 12:36:29.805844: step: 1608/529, loss: 9.91821252682712e-06 2023-01-21 12:36:30.919694: step: 1612/529, loss: 0.0018724919063970447 2023-01-21 12:36:32.054381: step: 1616/529, loss: 0.01027293223887682 2023-01-21 12:36:33.221735: step: 1620/529, loss: 0.034696388989686966 2023-01-21 12:36:34.350991: step: 1624/529, loss: 0.022735213860869408 2023-01-21 12:36:35.483774: step: 1628/529, loss: 0.0034936906304210424 2023-01-21 12:36:36.604066: step: 1632/529, loss: 0.0008212804677896202 2023-01-21 12:36:37.716253: step: 1636/529, loss: 0.00149879464879632 2023-01-21 12:36:38.832641: step: 1640/529, loss: 0.09865227341651917 2023-01-21 12:36:39.965498: step: 1644/529, loss: 0.05509652942419052 2023-01-21 12:36:41.071531: step: 1648/529, loss: 0.0005855560302734375 2023-01-21 12:36:42.192132: step: 1652/529, loss: 0.0034090043045580387 2023-01-21 12:36:43.326775: step: 1656/529, loss: 0.0008191108936443925 2023-01-21 12:36:44.450829: step: 1660/529, loss: 0.0007898330804891884 2023-01-21 12:36:45.553995: step: 1664/529, loss: 0.007561827078461647 2023-01-21 12:36:46.680189: step: 1668/529, loss: 9.355545626021922e-05 2023-01-21 12:36:47.762419: step: 1672/529, loss: 0.0020486831199377775 2023-01-21 12:36:48.882373: step: 1676/529, loss: 0.0004220008850097656 2023-01-21 12:36:50.008861: step: 1680/529, loss: 0.0004980087978765368 2023-01-21 12:36:51.130199: step: 1684/529, loss: 2.47955322265625e-05 2023-01-21 12:36:52.231786: step: 1688/529, loss: 5.52177443751134e-05 2023-01-21 12:36:53.368167: step: 1692/529, loss: 0.000846219074446708 2023-01-21 12:36:54.510048: step: 1696/529, loss: 0.006716919131577015 2023-01-21 12:36:55.608726: step: 1700/529, loss: 0.000965118408203125 2023-01-21 12:36:56.706926: step: 1704/529, loss: 0.0015100479358807206 2023-01-21 12:36:57.831181: step: 1708/529, loss: 0.052350856363773346 2023-01-21 12:36:58.913093: step: 1712/529, loss: 0.0038344860076904297 2023-01-21 12:37:00.000302: step: 1716/529, loss: 0.0007698058616369963 2023-01-21 12:37:01.111145: step: 1720/529, loss: 0.005507374182343483 2023-01-21 12:37:02.184932: step: 1724/529, loss: 0.003798961639404297 2023-01-21 12:37:03.338566: step: 1728/529, loss: 0.0007230758783407509 2023-01-21 12:37:04.448719: step: 1732/529, loss: 0.0028829574584960938 2023-01-21 12:37:05.580695: step: 1736/529, loss: 4.224777512717992e-05 2023-01-21 12:37:06.700485: step: 1740/529, loss: 0.016884375363588333 2023-01-21 12:37:07.801118: step: 1744/529, loss: 0.01977691613137722 2023-01-21 12:37:08.935934: step: 1748/529, loss: 4.158019874012098e-05 2023-01-21 12:37:10.054254: step: 1752/529, loss: 0.00023193359083961695 2023-01-21 12:37:11.150034: step: 1756/529, loss: 0.00020713807316496968 2023-01-21 12:37:12.268289: step: 1760/529, loss: 0.013233184814453125 2023-01-21 12:37:13.395269: step: 1764/529, loss: 0.001970577286556363 2023-01-21 12:37:14.507035: step: 1768/529, loss: 0.01619892194867134 2023-01-21 12:37:15.626931: step: 1772/529, loss: 0.0009502411121502519 2023-01-21 12:37:16.748357: step: 1776/529, loss: 0.0692434310913086 2023-01-21 12:37:17.900383: step: 1780/529, loss: 0.008482170291244984 2023-01-21 12:37:19.011746: step: 1784/529, loss: 3.924369957530871e-05 2023-01-21 12:37:20.131082: step: 1788/529, loss: 0.00017886162095237523 2023-01-21 12:37:21.244663: step: 1792/529, loss: 0.0010334014659747481 2023-01-21 12:37:22.355310: step: 1796/529, loss: 9.34600848268019e-06 2023-01-21 12:37:23.453016: step: 1800/529, loss: 0.0004550933954305947 2023-01-21 12:37:24.561923: step: 1804/529, loss: 9.679795039119199e-05 2023-01-21 12:37:25.665275: step: 1808/529, loss: 0.019238758832216263 2023-01-21 12:37:26.776041: step: 1812/529, loss: 0.00070362095721066 2023-01-21 12:37:27.892406: step: 1816/529, loss: 0.0054340362548828125 2023-01-21 12:37:29.046574: step: 1820/529, loss: 0.00012798310490325093 2023-01-21 12:37:30.151564: step: 1824/529, loss: 3.566742088878527e-05 2023-01-21 12:37:31.249213: step: 1828/529, loss: 0.006403446197509766 2023-01-21 12:37:32.364524: step: 1832/529, loss: 0.08049946278333664 2023-01-21 12:37:33.514364: step: 1836/529, loss: 0.0001386404037475586 2023-01-21 12:37:34.627251: step: 1840/529, loss: 0.034163858741521835 2023-01-21 12:37:35.723567: step: 1844/529, loss: 0.01553115900605917 2023-01-21 12:37:36.831375: step: 1848/529, loss: 0.0001234054652741179 2023-01-21 12:37:37.951994: step: 1852/529, loss: 8.20159948489163e-06 2023-01-21 12:37:39.064593: step: 1856/529, loss: 0.0010892868740484118 2023-01-21 12:37:40.164067: step: 1860/529, loss: 0.0004696846008300781 2023-01-21 12:37:41.265933: step: 1864/529, loss: 0.00040454865666106343 2023-01-21 12:37:42.372240: step: 1868/529, loss: 0.0007097244379110634 2023-01-21 12:37:43.487033: step: 1872/529, loss: 0.0008018494118005037 2023-01-21 12:37:44.573442: step: 1876/529, loss: 0.007692813873291016 2023-01-21 12:37:45.677499: step: 1880/529, loss: 0.00024509429931640625 2023-01-21 12:37:46.773381: step: 1884/529, loss: 0.002017498016357422 2023-01-21 12:37:47.863457: step: 1888/529, loss: 0.002883243840187788 2023-01-21 12:37:48.986267: step: 1892/529, loss: 0.004039764404296875 2023-01-21 12:37:50.076686: step: 1896/529, loss: 0.04199400171637535 2023-01-21 12:37:51.214788: step: 1900/529, loss: 0.01093978900462389 2023-01-21 12:37:52.350517: step: 1904/529, loss: 0.0019046784145757556 2023-01-21 12:37:53.444435: step: 1908/529, loss: 0.006050693802535534 2023-01-21 12:37:54.525122: step: 1912/529, loss: 8.354186866199598e-05 2023-01-21 12:37:55.633039: step: 1916/529, loss: 0.0001407623349223286 2023-01-21 12:37:56.726172: step: 1920/529, loss: 1.182556115963962e-05 2023-01-21 12:37:57.866780: step: 1924/529, loss: 0.04809394106268883 2023-01-21 12:37:58.984333: step: 1928/529, loss: 0.02637319825589657 2023-01-21 12:38:00.104191: step: 1932/529, loss: 0.05322151258587837 2023-01-21 12:38:01.223892: step: 1936/529, loss: 0.12012310326099396 2023-01-21 12:38:02.328805: step: 1940/529, loss: 0.0001468658447265625 2023-01-21 12:38:03.453473: step: 1944/529, loss: 0.0003570556582417339 2023-01-21 12:38:04.552077: step: 1948/529, loss: 0.0027442933060228825 2023-01-21 12:38:05.643020: step: 1952/529, loss: 0.0007123947143554688 2023-01-21 12:38:06.750004: step: 1956/529, loss: 0.0002639770391397178 2023-01-21 12:38:07.864985: step: 1960/529, loss: 0.033715344965457916 2023-01-21 12:38:08.970534: step: 1964/529, loss: 0.00021018982806708664 2023-01-21 12:38:10.104457: step: 1968/529, loss: 0.0033559321891516447 2023-01-21 12:38:11.202787: step: 1972/529, loss: 0.0077040670439600945 2023-01-21 12:38:12.309270: step: 1976/529, loss: 0.006897187326103449 2023-01-21 12:38:13.398630: step: 1980/529, loss: 0.001946258475072682 2023-01-21 12:38:14.503955: step: 1984/529, loss: 0.022505760192871094 2023-01-21 12:38:15.630861: step: 1988/529, loss: 0.0010709763737395406 2023-01-21 12:38:16.805614: step: 1992/529, loss: 0.00034427642822265625 2023-01-21 12:38:17.940145: step: 1996/529, loss: 0.00024776457576081157 2023-01-21 12:38:19.058464: step: 2000/529, loss: 0.0005731582641601562 2023-01-21 12:38:20.184422: step: 2004/529, loss: 0.028496552258729935 2023-01-21 12:38:21.282194: step: 2008/529, loss: 9.040832810569555e-05 2023-01-21 12:38:22.413773: step: 2012/529, loss: 0.01640176773071289 2023-01-21 12:38:23.530235: step: 2016/529, loss: 0.01987914927303791 2023-01-21 12:38:24.631830: step: 2020/529, loss: 0.007909774780273438 2023-01-21 12:38:25.766620: step: 2024/529, loss: 0.006548881530761719 2023-01-21 12:38:26.874144: step: 2028/529, loss: 0.037400055676698685 2023-01-21 12:38:27.967178: step: 2032/529, loss: 6.252527782635298e-06 2023-01-21 12:38:29.085125: step: 2036/529, loss: 0.001966858049854636 2023-01-21 12:38:30.190671: step: 2040/529, loss: 0.061445046216249466 2023-01-21 12:38:31.293045: step: 2044/529, loss: 0.00042428969754837453 2023-01-21 12:38:32.413454: step: 2048/529, loss: 0.01059055421501398 2023-01-21 12:38:33.524662: step: 2052/529, loss: 0.002580070635303855 2023-01-21 12:38:34.642480: step: 2056/529, loss: 1.9216537111788057e-05 2023-01-21 12:38:35.770081: step: 2060/529, loss: 0.03037882037460804 2023-01-21 12:38:36.915541: step: 2064/529, loss: 0.035574913024902344 2023-01-21 12:38:38.026863: step: 2068/529, loss: 0.004059553146362305 2023-01-21 12:38:39.185795: step: 2072/529, loss: 0.39640700817108154 2023-01-21 12:38:40.329815: step: 2076/529, loss: 0.011905480176210403 2023-01-21 12:38:41.442678: step: 2080/529, loss: 0.001651763916015625 2023-01-21 12:38:42.528422: step: 2084/529, loss: 0.0009592056157998741 2023-01-21 12:38:43.641953: step: 2088/529, loss: 0.012651586905121803 2023-01-21 12:38:44.752766: step: 2092/529, loss: 0.0075893402099609375 2023-01-21 12:38:45.863501: step: 2096/529, loss: 0.00016288757615257055 2023-01-21 12:38:46.958612: step: 2100/529, loss: 0.004981613252311945 2023-01-21 12:38:48.069438: step: 2104/529, loss: 0.020656870678067207 2023-01-21 12:38:49.188313: step: 2108/529, loss: 1.945495569088962e-05 2023-01-21 12:38:50.296180: step: 2112/529, loss: 0.0013467788230627775 2023-01-21 12:38:51.381494: step: 2116/529, loss: 0.0005268097156658769 ================================================== Loss: 0.025 -------------------- Dev: {'event': {'p': 0.5873015873015873, 'r': 0.7882822902796272, 'f1': 0.6731097214326321}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Test: {'event': {'p': 0.6295793758480326, 'r': 0.7855530474040632, 'f1': 0.6989706251569168}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} 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': 23} Korean: {'event': {'p': 0.5774647887323944, 'r': 0.6507936507936508, 'f1': 0.6119402985074627}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} 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': 23} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 24 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:39:31.904676: step: 4/529, loss: 0.0003925323544535786 2023-01-21 12:39:33.033929: step: 8/529, loss: 0.0002473831409588456 2023-01-21 12:39:34.148812: step: 12/529, loss: 0.008356285281479359 2023-01-21 12:39:35.244493: step: 16/529, loss: 0.21779365837574005 2023-01-21 12:39:36.347836: step: 20/529, loss: 0.005253219977021217 2023-01-21 12:39:37.480684: step: 24/529, loss: 0.0007912635919637978 2023-01-21 12:39:38.618487: step: 28/529, loss: 0.003102731890976429 2023-01-21 12:39:39.745839: step: 32/529, loss: 0.00017538070096634328 2023-01-21 12:39:40.861424: step: 36/529, loss: 6.132126145530492e-05 2023-01-21 12:39:41.981077: step: 40/529, loss: 0.00975132081657648 2023-01-21 12:39:43.082385: step: 44/529, loss: 0.019554710015654564 2023-01-21 12:39:44.233986: step: 48/529, loss: 0.0008663177723065019 2023-01-21 12:39:45.407749: step: 52/529, loss: 0.001243639038875699 2023-01-21 12:39:46.542650: step: 56/529, loss: 8.525847806595266e-05 2023-01-21 12:39:47.646205: step: 60/529, loss: 0.00014123917208053172 2023-01-21 12:39:48.767962: step: 64/529, loss: 0.002201271243393421 2023-01-21 12:39:49.867900: step: 68/529, loss: 0.006466865539550781 2023-01-21 12:39:51.015581: step: 72/529, loss: 0.0014100074768066406 2023-01-21 12:39:52.129861: step: 76/529, loss: 0.0026243210304528475 2023-01-21 12:39:53.251462: step: 80/529, loss: 0.003959274850785732 2023-01-21 12:39:54.362971: step: 84/529, loss: 0.018009185791015625 2023-01-21 12:39:55.487893: step: 88/529, loss: 0.008799361996352673 2023-01-21 12:39:56.600969: step: 92/529, loss: 0.034955788403749466 2023-01-21 12:39:57.698998: step: 96/529, loss: 2.326965295651462e-05 2023-01-21 12:39:58.815078: step: 100/529, loss: 0.033087920397520065 2023-01-21 12:39:59.935396: step: 104/529, loss: 0.002468871884047985 2023-01-21 12:40:01.029352: step: 108/529, loss: 0.007968520745635033 2023-01-21 12:40:02.165440: step: 112/529, loss: 0.010781098157167435 2023-01-21 12:40:03.263182: step: 116/529, loss: 0.0010690211784094572 2023-01-21 12:40:04.374809: step: 120/529, loss: 0.02180042304098606 2023-01-21 12:40:05.470383: step: 124/529, loss: 0.002035522600635886 2023-01-21 12:40:06.599152: step: 128/529, loss: 0.08149462193250656 2023-01-21 12:40:07.725707: step: 132/529, loss: 0.0009987950325012207 2023-01-21 12:40:08.895641: step: 136/529, loss: 0.002506256103515625 2023-01-21 12:40:10.021653: step: 140/529, loss: 4.38690185546875e-05 2023-01-21 12:40:11.173388: step: 144/529, loss: 0.005530166439712048 2023-01-21 12:40:12.264391: step: 148/529, loss: 0.05460701137781143 2023-01-21 12:40:13.363981: step: 152/529, loss: 0.030742358416318893 2023-01-21 12:40:14.471591: step: 156/529, loss: 0.00010547637793933973 2023-01-21 12:40:15.602926: step: 160/529, loss: 0.0019791603554040194 2023-01-21 12:40:16.711316: step: 164/529, loss: 4.253388033248484e-05 2023-01-21 12:40:17.833854: step: 168/529, loss: 0.0008592605590820312 2023-01-21 12:40:18.964728: step: 172/529, loss: 0.000217437744140625 2023-01-21 12:40:20.069721: step: 176/529, loss: 0.0017484663985669613 2023-01-21 12:40:21.183663: step: 180/529, loss: 0.0002155303955078125 2023-01-21 12:40:22.279711: step: 184/529, loss: 0.006379890255630016 2023-01-21 12:40:23.377044: step: 188/529, loss: 0.0006915092235431075 2023-01-21 12:40:24.541148: step: 192/529, loss: 0.012439919635653496 2023-01-21 12:40:25.624354: step: 196/529, loss: 0.0020318033639341593 2023-01-21 12:40:26.751029: step: 200/529, loss: 0.014362812042236328 2023-01-21 12:40:27.891714: step: 204/529, loss: 0.000274658203125 2023-01-21 12:40:29.003379: step: 208/529, loss: 0.582240104675293 2023-01-21 12:40:30.118466: step: 212/529, loss: 0.008256340399384499 2023-01-21 12:40:31.226845: step: 216/529, loss: 4.920959327137098e-05 2023-01-21 12:40:32.356297: step: 220/529, loss: 0.0012525558704510331 2023-01-21 12:40:33.466956: step: 224/529, loss: 1.33514404296875e-05 2023-01-21 12:40:34.572866: step: 228/529, loss: 0.026694010943174362 2023-01-21 12:40:35.703170: step: 232/529, loss: 0.01851978339254856 2023-01-21 12:40:36.785597: step: 236/529, loss: 0.0005932808271609247 2023-01-21 12:40:37.925851: step: 240/529, loss: 0.012573814950883389 2023-01-21 12:40:39.038150: step: 244/529, loss: 0.004208183381706476 2023-01-21 12:40:40.140008: step: 248/529, loss: 0.008863353170454502 2023-01-21 12:40:41.251363: step: 252/529, loss: 0.0008830070728436112 2023-01-21 12:40:42.368317: step: 256/529, loss: 0.00016670227341819555 2023-01-21 12:40:43.475725: step: 260/529, loss: 5.588531348621473e-05 2023-01-21 12:40:44.570730: step: 264/529, loss: 0.013968467712402344 2023-01-21 12:40:45.719093: step: 268/529, loss: 0.04032020643353462 2023-01-21 12:40:46.824505: step: 272/529, loss: 0.0003326416190247983 2023-01-21 12:40:47.945640: step: 276/529, loss: 0.0013057708274573088 2023-01-21 12:40:49.033495: step: 280/529, loss: 9.050369408214465e-05 2023-01-21 12:40:50.184400: step: 284/529, loss: 0.00013809204392600805 2023-01-21 12:40:51.294654: step: 288/529, loss: 0.01706409640610218 2023-01-21 12:40:52.401509: step: 292/529, loss: 0.0028888704255223274 2023-01-21 12:40:53.507241: step: 296/529, loss: 0.00011711120896507055 2023-01-21 12:40:54.626892: step: 300/529, loss: 0.0006399155245162547 2023-01-21 12:40:55.717472: step: 304/529, loss: 0.00108251569326967 2023-01-21 12:40:56.833166: step: 308/529, loss: 0.0013418197631835938 2023-01-21 12:40:58.018149: step: 312/529, loss: 0.003175163408741355 2023-01-21 12:40:59.149544: step: 316/529, loss: 0.00037212372990325093 2023-01-21 12:41:00.248765: step: 320/529, loss: 0.0007799148443154991 2023-01-21 12:41:01.356330: step: 324/529, loss: 0.006014466285705566 2023-01-21 12:41:02.490753: step: 328/529, loss: 0.0095977783203125 2023-01-21 12:41:03.603057: step: 332/529, loss: 0.0011815071338787675 2023-01-21 12:41:04.742880: step: 336/529, loss: 0.0010402679909020662 2023-01-21 12:41:05.870014: step: 340/529, loss: 0.06329450756311417 2023-01-21 12:41:07.028283: step: 344/529, loss: 0.006624317727982998 2023-01-21 12:41:08.133592: step: 348/529, loss: 0.04829845577478409 2023-01-21 12:41:09.257200: step: 352/529, loss: 0.0002283096400788054 2023-01-21 12:41:10.398949: step: 356/529, loss: 0.000133514404296875 2023-01-21 12:41:11.522400: step: 360/529, loss: 4.19616708313697e-06 2023-01-21 12:41:12.620784: step: 364/529, loss: 0.0002418518124613911 2023-01-21 12:41:13.755166: step: 368/529, loss: 7.765292684780434e-05 2023-01-21 12:41:14.871877: step: 372/529, loss: 0.0004209518665447831 2023-01-21 12:41:15.976452: step: 376/529, loss: 0.012350846081972122 2023-01-21 12:41:17.057712: step: 380/529, loss: 0.008633232675492764 2023-01-21 12:41:18.198357: step: 384/529, loss: 0.0006875991821289062 2023-01-21 12:41:19.323469: step: 388/529, loss: 0.0002342224179301411 2023-01-21 12:41:20.417256: step: 392/529, loss: 9.5367431640625e-06 2023-01-21 12:41:21.543309: step: 396/529, loss: 0.007999230176210403 2023-01-21 12:41:22.645078: step: 400/529, loss: 0.0005586147890426219 2023-01-21 12:41:23.769635: step: 404/529, loss: 0.0002388000430073589 2023-01-21 12:41:24.908885: step: 408/529, loss: 0.000357079494278878 2023-01-21 12:41:26.014618: step: 412/529, loss: 0.004529762081801891 2023-01-21 12:41:27.117964: step: 416/529, loss: 0.0011648177169263363 2023-01-21 12:41:28.232433: step: 420/529, loss: 0.0017258645966649055 2023-01-21 12:41:29.372841: step: 424/529, loss: 0.0800677016377449 2023-01-21 12:41:30.477531: step: 428/529, loss: 0.0007628441089764237 2023-01-21 12:41:31.599728: step: 432/529, loss: 0.0019208431476727128 2023-01-21 12:41:32.713918: step: 436/529, loss: 0.00410385150462389 2023-01-21 12:41:33.869608: step: 440/529, loss: 8.36372419144027e-05 2023-01-21 12:41:35.010715: step: 444/529, loss: 0.013046455569565296 2023-01-21 12:41:36.128476: step: 448/529, loss: 0.02074127271771431 2023-01-21 12:41:37.252699: step: 452/529, loss: 0.0021046637557446957 2023-01-21 12:41:38.382555: step: 456/529, loss: 0.0009842872386798263 2023-01-21 12:41:39.477979: step: 460/529, loss: 0.009906863793730736 2023-01-21 12:41:40.599582: step: 464/529, loss: 0.02371845208108425 2023-01-21 12:41:41.708357: step: 468/529, loss: 2.8705597287626006e-05 2023-01-21 12:41:42.816274: step: 472/529, loss: 0.0011688233353197575 2023-01-21 12:41:43.913759: step: 476/529, loss: 6.532669067382812e-05 2023-01-21 12:41:45.007621: step: 480/529, loss: 0.006134224124252796 2023-01-21 12:41:46.093880: step: 484/529, loss: 9.202957699017134e-06 2023-01-21 12:41:47.186623: step: 488/529, loss: 0.007199001498520374 2023-01-21 12:41:48.322528: step: 492/529, loss: 0.2831043303012848 2023-01-21 12:41:49.422305: step: 496/529, loss: 0.009466147050261497 2023-01-21 12:41:50.507144: step: 500/529, loss: 0.0004096984921488911 2023-01-21 12:41:51.578370: step: 504/529, loss: 0.019950199872255325 2023-01-21 12:41:52.679466: step: 508/529, loss: 0.002665710635483265 2023-01-21 12:41:53.779384: step: 512/529, loss: 0.007513618562370539 2023-01-21 12:41:54.907567: step: 516/529, loss: 2.517700158932712e-05 2023-01-21 12:41:56.046750: step: 520/529, loss: -8.01086389401462e-06 2023-01-21 12:41:57.145152: step: 524/529, loss: 0.00019149782019667327 2023-01-21 12:41:58.267865: step: 528/529, loss: 0.0006608963012695312 2023-01-21 12:41:59.375237: step: 532/529, loss: 0.010239506140351295 2023-01-21 12:42:00.478153: step: 536/529, loss: 0.01025466900318861 2023-01-21 12:42:01.611302: step: 540/529, loss: 0.019496344029903412 2023-01-21 12:42:02.746874: step: 544/529, loss: 0.0022645951248705387 2023-01-21 12:42:03.861601: step: 548/529, loss: 0.0009126186487264931 2023-01-21 12:42:04.980571: step: 552/529, loss: 0.00038337710429914296 2023-01-21 12:42:06.097255: step: 556/529, loss: 0.00153522496111691 2023-01-21 12:42:07.182554: step: 560/529, loss: 0.0026688575744628906 2023-01-21 12:42:08.280363: step: 564/529, loss: 0.0023283003829419613 2023-01-21 12:42:09.423685: step: 568/529, loss: 5.5122378398664296e-05 2023-01-21 12:42:10.550583: step: 572/529, loss: 0.01564769819378853 2023-01-21 12:42:11.672572: step: 576/529, loss: 0.00650634802877903 2023-01-21 12:42:12.794736: step: 580/529, loss: 0.00034718512324616313 2023-01-21 12:42:13.902058: step: 584/529, loss: 2.4414064682787284e-05 2023-01-21 12:42:15.013758: step: 588/529, loss: 1.087188684323337e-05 2023-01-21 12:42:16.116537: step: 592/529, loss: 0.021163273602724075 2023-01-21 12:42:17.250875: step: 596/529, loss: 0.023978807032108307 2023-01-21 12:42:18.338952: step: 600/529, loss: 0.0012805939186364412 2023-01-21 12:42:19.489540: step: 604/529, loss: 0.0027101996820420027 2023-01-21 12:42:20.583579: step: 608/529, loss: 0.0006782531854696572 2023-01-21 12:42:21.699369: step: 612/529, loss: 0.011052703484892845 2023-01-21 12:42:22.797861: step: 616/529, loss: -1.087188684323337e-05 2023-01-21 12:42:23.923021: step: 620/529, loss: 0.061460401862859726 2023-01-21 12:42:25.037696: step: 624/529, loss: 0.0017684936756268144 2023-01-21 12:42:26.131501: step: 628/529, loss: 0.0009705544216558337 2023-01-21 12:42:27.271418: step: 632/529, loss: 0.006282901391386986 2023-01-21 12:42:28.398678: step: 636/529, loss: 8.58306884765625e-06 2023-01-21 12:42:29.531908: step: 640/529, loss: 0.03083820268511772 2023-01-21 12:42:30.640085: step: 644/529, loss: 0.0019453049171715975 2023-01-21 12:42:31.757013: step: 648/529, loss: 0.00011291503324173391 2023-01-21 12:42:32.878886: step: 652/529, loss: 0.0010592460166662931 2023-01-21 12:42:33.983192: step: 656/529, loss: 0.024510765448212624 2023-01-21 12:42:35.077023: step: 660/529, loss: 0.0017028809525072575 2023-01-21 12:42:36.192143: step: 664/529, loss: 0.007291984278708696 2023-01-21 12:42:37.312176: step: 668/529, loss: 0.0016824722988530993 2023-01-21 12:42:38.405322: step: 672/529, loss: 0.0003899097500834614 2023-01-21 12:42:39.551928: step: 676/529, loss: 0.02898998372256756 2023-01-21 12:42:40.714997: step: 680/529, loss: -2.86102294921875e-06 2023-01-21 12:42:41.818318: step: 684/529, loss: 0.0019882202614098787 2023-01-21 12:42:42.935762: step: 688/529, loss: 0.0005960463895462453 2023-01-21 12:42:44.075137: step: 692/529, loss: 0.00981826800853014 2023-01-21 12:42:45.203547: step: 696/529, loss: 0.014781760983169079 2023-01-21 12:42:46.321324: step: 700/529, loss: 0.003985595889389515 2023-01-21 12:42:47.441120: step: 704/529, loss: 0.0008586883777752519 2023-01-21 12:42:48.517715: step: 708/529, loss: 2.2697449821862392e-05 2023-01-21 12:42:49.618850: step: 712/529, loss: 0.00965585745871067 2023-01-21 12:42:50.713916: step: 716/529, loss: 0.0016796470154076815 2023-01-21 12:42:51.813024: step: 720/529, loss: 0.01489119604229927 2023-01-21 12:42:52.926461: step: 724/529, loss: 0.015282725915312767 2023-01-21 12:42:54.015853: step: 728/529, loss: 0.10183233767747879 2023-01-21 12:42:55.152173: step: 732/529, loss: 0.028457069769501686 2023-01-21 12:42:56.282486: step: 736/529, loss: 0.0004585266287904233 2023-01-21 12:42:57.383936: step: 740/529, loss: 0.0028816224075853825 2023-01-21 12:42:58.484534: step: 744/529, loss: 0.0035650255158543587 2023-01-21 12:42:59.586882: step: 748/529, loss: 0.0007166862487792969 2023-01-21 12:43:00.714633: step: 752/529, loss: 0.0003093719424214214 2023-01-21 12:43:01.817099: step: 756/529, loss: 0.02471332438290119 2023-01-21 12:43:02.975355: step: 760/529, loss: 0.0008200645097531378 2023-01-21 12:43:04.069828: step: 764/529, loss: 0.0005647182697430253 2023-01-21 12:43:05.196111: step: 768/529, loss: 0.0003486633358988911 2023-01-21 12:43:06.273014: step: 772/529, loss: 0.007267951965332031 2023-01-21 12:43:07.379717: step: 776/529, loss: 0.0013552665477618575 2023-01-21 12:43:08.480671: step: 780/529, loss: 0.0013607979053631425 2023-01-21 12:43:09.592595: step: 784/529, loss: 0.0005233764532022178 2023-01-21 12:43:10.688741: step: 788/529, loss: 6.046295311534777e-05 2023-01-21 12:43:11.825769: step: 792/529, loss: 6.723403930664062e-05 2023-01-21 12:43:12.957461: step: 796/529, loss: 0.0008447408908978105 2023-01-21 12:43:14.071079: step: 800/529, loss: 0.02805805206298828 2023-01-21 12:43:15.167261: step: 804/529, loss: 0.0001445770321879536 2023-01-21 12:43:16.302557: step: 808/529, loss: 2.3174286980065517e-05 2023-01-21 12:43:17.432634: step: 812/529, loss: 0.00216846470721066 2023-01-21 12:43:18.531035: step: 816/529, loss: 0.0009506225469522178 2023-01-21 12:43:19.674708: step: 820/529, loss: 0.0017351150745525956 2023-01-21 12:43:20.826153: step: 824/529, loss: 0.03768196329474449 2023-01-21 12:43:21.931079: step: 828/529, loss: 0.03584575653076172 2023-01-21 12:43:23.053874: step: 832/529, loss: 0.02200775034725666 2023-01-21 12:43:24.184912: step: 836/529, loss: 0.00029315947904251516 2023-01-21 12:43:25.304641: step: 840/529, loss: 0.0001504898100392893 2023-01-21 12:43:26.433168: step: 844/529, loss: 0.02613077126443386 2023-01-21 12:43:27.542446: step: 848/529, loss: 0.05107174068689346 2023-01-21 12:43:28.643652: step: 852/529, loss: 0.004071045201271772 2023-01-21 12:43:29.779376: step: 856/529, loss: 0.007562350947409868 2023-01-21 12:43:30.896348: step: 860/529, loss: 0.030723953619599342 2023-01-21 12:43:31.998919: step: 864/529, loss: 8.230209641624242e-05 2023-01-21 12:43:33.125600: step: 868/529, loss: 0.040189363062381744 2023-01-21 12:43:34.263481: step: 872/529, loss: 4.100799742445815e-06 2023-01-21 12:43:35.357086: step: 876/529, loss: 0.03584747016429901 2023-01-21 12:43:36.486780: step: 880/529, loss: 0.0792156234383583 2023-01-21 12:43:37.607637: step: 884/529, loss: 0.00014514924259856343 2023-01-21 12:43:38.727152: step: 888/529, loss: 0.0010093689197674394 2023-01-21 12:43:39.829900: step: 892/529, loss: 0.0003459930594544858 2023-01-21 12:43:40.960760: step: 896/529, loss: 0.0009670734289102256 2023-01-21 12:43:42.087516: step: 900/529, loss: 0.00011773109872592613 2023-01-21 12:43:43.201749: step: 904/529, loss: 0.038805581629276276 2023-01-21 12:43:44.344560: step: 908/529, loss: 0.00028705596923828125 2023-01-21 12:43:45.459456: step: 912/529, loss: 0.03404046222567558 2023-01-21 12:43:46.578141: step: 916/529, loss: 0.001767158624716103 2023-01-21 12:43:47.721392: step: 920/529, loss: 0.008560466580092907 2023-01-21 12:43:48.860921: step: 924/529, loss: 6.294250852079131e-06 2023-01-21 12:43:49.985048: step: 928/529, loss: 0.0023124695289880037 2023-01-21 12:43:51.162947: step: 932/529, loss: 0.006863737478852272 2023-01-21 12:43:52.265718: step: 936/529, loss: 7.63893112889491e-05 2023-01-21 12:43:53.397214: step: 940/529, loss: 0.0763576477766037 2023-01-21 12:43:54.538741: step: 944/529, loss: 1.0965994596481323 2023-01-21 12:43:55.679129: step: 948/529, loss: 0.01883668825030327 2023-01-21 12:43:56.776665: step: 952/529, loss: 0.016630172729492188 2023-01-21 12:43:57.894304: step: 956/529, loss: 0.00018901826115325093 2023-01-21 12:43:58.987245: step: 960/529, loss: 0.004084587097167969 2023-01-21 12:44:00.090714: step: 964/529, loss: 0.005934644024819136 2023-01-21 12:44:01.239337: step: 968/529, loss: 0.0012413025833666325 2023-01-21 12:44:02.368944: step: 972/529, loss: 0.0002578735293354839 2023-01-21 12:44:03.495198: step: 976/529, loss: 0.0015085219638422132 2023-01-21 12:44:04.662809: step: 980/529, loss: 0.0023611069191247225 2023-01-21 12:44:05.762471: step: 984/529, loss: 0.057363320142030716 2023-01-21 12:44:06.868396: step: 988/529, loss: 0.0008186340564861894 2023-01-21 12:44:07.970334: step: 992/529, loss: 0.01078262459486723 2023-01-21 12:44:09.092294: step: 996/529, loss: 0.0016262053977698088 2023-01-21 12:44:10.189692: step: 1000/529, loss: 0.005451202858239412 2023-01-21 12:44:11.303114: step: 1004/529, loss: 0.0014255524147301912 2023-01-21 12:44:12.470938: step: 1008/529, loss: 0.015783309936523438 2023-01-21 12:44:13.587056: step: 1012/529, loss: 0.0006582260248251259 2023-01-21 12:44:14.702433: step: 1016/529, loss: 0.0008030892349779606 2023-01-21 12:44:15.817322: step: 1020/529, loss: 0.004277610685676336 2023-01-21 12:44:16.909152: step: 1024/529, loss: 0.0001354217529296875 2023-01-21 12:44:18.032846: step: 1028/529, loss: 0.02090606652200222 2023-01-21 12:44:19.134556: step: 1032/529, loss: 0.0012190818088129163 2023-01-21 12:44:20.246510: step: 1036/529, loss: 0.01520385779440403 2023-01-21 12:44:21.357070: step: 1040/529, loss: 0.009545708075165749 2023-01-21 12:44:22.487159: step: 1044/529, loss: 0.009309959597885609 2023-01-21 12:44:23.609214: step: 1048/529, loss: 0.01163473166525364 2023-01-21 12:44:24.717668: step: 1052/529, loss: 0.00399703998118639 2023-01-21 12:44:25.836150: step: 1056/529, loss: 0.00531768798828125 2023-01-21 12:44:26.952133: step: 1060/529, loss: 0.054778195917606354 2023-01-21 12:44:28.094967: step: 1064/529, loss: 0.004431057255715132 2023-01-21 12:44:29.238702: step: 1068/529, loss: 0.006871271412819624 2023-01-21 12:44:30.364096: step: 1072/529, loss: 0.007038688287138939 2023-01-21 12:44:31.491407: step: 1076/529, loss: 0.0028528214897960424 2023-01-21 12:44:32.583627: step: 1080/529, loss: 8.21113571873866e-05 2023-01-21 12:44:33.707491: step: 1084/529, loss: 0.001213455107063055 2023-01-21 12:44:34.824694: step: 1088/529, loss: 0.002114200731739402 2023-01-21 12:44:35.946140: step: 1092/529, loss: 0.021538924425840378 2023-01-21 12:44:37.060659: step: 1096/529, loss: 0.01445617713034153 2023-01-21 12:44:38.150911: step: 1100/529, loss: 0.00035409926204010844 2023-01-21 12:44:39.246856: step: 1104/529, loss: 0.0006986617809161544 2023-01-21 12:44:40.380488: step: 1108/529, loss: 0.0022852900438010693 2023-01-21 12:44:41.522041: step: 1112/529, loss: 0.011485004797577858 2023-01-21 12:44:42.623717: step: 1116/529, loss: 0.0010349274380132556 2023-01-21 12:44:43.706782: step: 1120/529, loss: 6.444454629672691e-05 2023-01-21 12:44:44.848280: step: 1124/529, loss: 7.381439354503527e-05 2023-01-21 12:44:45.963251: step: 1128/529, loss: 0.00018558502779342234 2023-01-21 12:44:47.069569: step: 1132/529, loss: 0.001514530275017023 2023-01-21 12:44:48.227558: step: 1136/529, loss: 0.006368446629494429 2023-01-21 12:44:49.359805: step: 1140/529, loss: 0.0006160736083984375 2023-01-21 12:44:50.484863: step: 1144/529, loss: 0.007418299093842506 2023-01-21 12:44:51.596923: step: 1148/529, loss: 0.021350478753447533 2023-01-21 12:44:52.762657: step: 1152/529, loss: 0.00022666454606223851 2023-01-21 12:44:53.875141: step: 1156/529, loss: 0.0007070541614666581 2023-01-21 12:44:54.987983: step: 1160/529, loss: 0.0007173537742346525 2023-01-21 12:44:56.096624: step: 1164/529, loss: 3.2901763916015625e-05 2023-01-21 12:44:57.218161: step: 1168/529, loss: 0.027675151824951172 2023-01-21 12:44:58.327598: step: 1172/529, loss: 0.0005023002740927041 2023-01-21 12:44:59.451042: step: 1176/529, loss: 3.948211815441027e-05 2023-01-21 12:45:00.574810: step: 1180/529, loss: 0.004982948303222656 2023-01-21 12:45:01.710210: step: 1184/529, loss: 0.0015054703690111637 2023-01-21 12:45:02.825662: step: 1188/529, loss: 6.723403930664062e-05 2023-01-21 12:45:03.923254: step: 1192/529, loss: 0.00017843247042037547 2023-01-21 12:45:05.028934: step: 1196/529, loss: 0.0019850730895996094 2023-01-21 12:45:06.163707: step: 1200/529, loss: 0.39782485365867615 2023-01-21 12:45:07.287958: step: 1204/529, loss: 0.007019424811005592 2023-01-21 12:45:08.388200: step: 1208/529, loss: 0.00014715194993186742 2023-01-21 12:45:09.496470: step: 1212/529, loss: 0.07494864612817764 2023-01-21 12:45:10.633905: step: 1216/529, loss: 1.9602477550506592e-05 2023-01-21 12:45:11.741683: step: 1220/529, loss: 1.792907642084174e-05 2023-01-21 12:45:12.867674: step: 1224/529, loss: 0.0011323929065838456 2023-01-21 12:45:13.970429: step: 1228/529, loss: 0.0034996033646166325 2023-01-21 12:45:15.094517: step: 1232/529, loss: 0.021793365478515625 2023-01-21 12:45:16.183612: step: 1236/529, loss: 0.0005487442249432206 2023-01-21 12:45:17.337342: step: 1240/529, loss: 0.022512435913085938 2023-01-21 12:45:18.444398: step: 1244/529, loss: 0.0017073630588129163 2023-01-21 12:45:19.566243: step: 1248/529, loss: 0.0031631470192223787 2023-01-21 12:45:20.696392: step: 1252/529, loss: 0.0064199455082416534 2023-01-21 12:45:21.784404: step: 1256/529, loss: 0.0006179810152389109 2023-01-21 12:45:22.939802: step: 1260/529, loss: 0.00012683868408203125 2023-01-21 12:45:24.112288: step: 1264/529, loss: 0.0013432980049401522 2023-01-21 12:45:25.241340: step: 1268/529, loss: 0.0003296852228231728 2023-01-21 12:45:26.337836: step: 1272/529, loss: 0.009262013249099255 2023-01-21 12:45:27.437807: step: 1276/529, loss: 0.0056625367142260075 2023-01-21 12:45:28.563728: step: 1280/529, loss: 0.00039805175038054585 2023-01-21 12:45:29.662992: step: 1284/529, loss: 0.0005619049188680947 2023-01-21 12:45:30.806124: step: 1288/529, loss: 0.007315456867218018 2023-01-21 12:45:31.907174: step: 1292/529, loss: 0.00041909218998625875 2023-01-21 12:45:33.010182: step: 1296/529, loss: 0.6855027079582214 2023-01-21 12:45:34.089542: step: 1300/529, loss: 0.01907634735107422 2023-01-21 12:45:35.177083: step: 1304/529, loss: 0.0005462646367959678 2023-01-21 12:45:36.298935: step: 1308/529, loss: 0.0066363574005663395 2023-01-21 12:45:37.428877: step: 1312/529, loss: 0.00030727387638762593 2023-01-21 12:45:38.548602: step: 1316/529, loss: 0.0001274108944926411 2023-01-21 12:45:39.652269: step: 1320/529, loss: 4.94956984766759e-05 2023-01-21 12:45:40.791676: step: 1324/529, loss: 8.1062319168268e-07 2023-01-21 12:45:41.865635: step: 1328/529, loss: 0.01365447137504816 2023-01-21 12:45:42.974396: step: 1332/529, loss: 0.06259842216968536 2023-01-21 12:45:44.084606: step: 1336/529, loss: 0.00010309219942428172 2023-01-21 12:45:45.190238: step: 1340/529, loss: 0.00035572052001953125 2023-01-21 12:45:46.320794: step: 1344/529, loss: 0.015739820897579193 2023-01-21 12:45:47.446332: step: 1348/529, loss: 0.0008097171667031944 2023-01-21 12:45:48.549418: step: 1352/529, loss: 0.0005952835199423134 2023-01-21 12:45:49.671177: step: 1356/529, loss: 0.0034111500717699528 2023-01-21 12:45:50.813115: step: 1360/529, loss: 0.04425692558288574 2023-01-21 12:45:51.936589: step: 1364/529, loss: 0.0025315284729003906 2023-01-21 12:45:53.057681: step: 1368/529, loss: 0.016381217166781425 2023-01-21 12:45:54.180168: step: 1372/529, loss: 0.00038700102595612407 2023-01-21 12:45:55.266568: step: 1376/529, loss: 0.00019340515427757055 2023-01-21 12:45:56.391533: step: 1380/529, loss: 0.0010005474323406816 2023-01-21 12:45:57.493381: step: 1384/529, loss: 0.00034012793912552297 2023-01-21 12:45:58.604459: step: 1388/529, loss: 0.004056835547089577 2023-01-21 12:45:59.720056: step: 1392/529, loss: 0.00090198521502316 2023-01-21 12:46:00.811373: step: 1396/529, loss: 2.0790101189049892e-05 2023-01-21 12:46:01.944581: step: 1400/529, loss: 0.0012370110489428043 2023-01-21 12:46:03.062500: step: 1404/529, loss: 0.02625093422830105 2023-01-21 12:46:04.172579: step: 1408/529, loss: 0.03659839928150177 2023-01-21 12:46:05.303370: step: 1412/529, loss: 0.005652999971061945 2023-01-21 12:46:06.383480: step: 1416/529, loss: -1.23977656585339e-06 2023-01-21 12:46:07.477557: step: 1420/529, loss: 0.00016813278489280492 2023-01-21 12:46:08.614234: step: 1424/529, loss: 0.0003372192441020161 2023-01-21 12:46:09.739337: step: 1428/529, loss: 0.003093529026955366 2023-01-21 12:46:10.881959: step: 1432/529, loss: 0.014352607540786266 2023-01-21 12:46:11.995173: step: 1436/529, loss: 0.00524139404296875 2023-01-21 12:46:13.113303: step: 1440/529, loss: 0.02582702599465847 2023-01-21 12:46:14.252414: step: 1444/529, loss: 2.4509430659236386e-05 2023-01-21 12:46:15.391064: step: 1448/529, loss: 0.044438984245061874 2023-01-21 12:46:16.488250: step: 1452/529, loss: 5.6743621826171875e-05 2023-01-21 12:46:17.626780: step: 1456/529, loss: 3.814697265625e-05 2023-01-21 12:46:18.746242: step: 1460/529, loss: 0.003301334334537387 2023-01-21 12:46:19.874056: step: 1464/529, loss: 0.02140064351260662 2023-01-21 12:46:20.991385: step: 1468/529, loss: 0.008984756655991077 2023-01-21 12:46:22.089595: step: 1472/529, loss: 0.002600050065666437 2023-01-21 12:46:23.172593: step: 1476/529, loss: 0.3212962746620178 2023-01-21 12:46:24.286415: step: 1480/529, loss: 1.0997517108917236 2023-01-21 12:46:25.392890: step: 1484/529, loss: 1.926422191900201e-05 2023-01-21 12:46:26.488897: step: 1488/529, loss: 0.0004656791570596397 2023-01-21 12:46:27.593001: step: 1492/529, loss: 0.0016533852322027087 2023-01-21 12:46:28.692483: step: 1496/529, loss: 0.0009114265558309853 2023-01-21 12:46:29.779917: step: 1500/529, loss: 0.001280117081478238 2023-01-21 12:46:30.890058: step: 1504/529, loss: 0.07775726914405823 2023-01-21 12:46:31.995021: step: 1508/529, loss: 0.07758484035730362 2023-01-21 12:46:33.081836: step: 1512/529, loss: 0.007211207877844572 2023-01-21 12:46:34.185041: step: 1516/529, loss: 0.0004280090506654233 2023-01-21 12:46:35.346129: step: 1520/529, loss: 0.0003231048758607358 2023-01-21 12:46:36.497166: step: 1524/529, loss: 5.598067946266383e-05 2023-01-21 12:46:37.592622: step: 1528/529, loss: 0.17844629287719727 2023-01-21 12:46:38.719135: step: 1532/529, loss: 0.0026567461900413036 2023-01-21 12:46:39.830095: step: 1536/529, loss: 0.0003501892206259072 2023-01-21 12:46:40.925178: step: 1540/529, loss: 0.007067298982292414 2023-01-21 12:46:42.037345: step: 1544/529, loss: 0.000979518867097795 2023-01-21 12:46:43.157278: step: 1548/529, loss: 0.008297729305922985 2023-01-21 12:46:44.283945: step: 1552/529, loss: 0.004073619842529297 2023-01-21 12:46:45.376179: step: 1556/529, loss: 0.0011669874656945467 2023-01-21 12:46:46.515054: step: 1560/529, loss: 0.00564079312607646 2023-01-21 12:46:47.643725: step: 1564/529, loss: 0.0003078937588725239 2023-01-21 12:46:48.746784: step: 1568/529, loss: 0.002226901240646839 2023-01-21 12:46:49.880889: step: 1572/529, loss: 0.009495544247329235 2023-01-21 12:46:51.002949: step: 1576/529, loss: 0.045978449285030365 2023-01-21 12:46:52.102625: step: 1580/529, loss: 0.00076465611346066 2023-01-21 12:46:53.222790: step: 1584/529, loss: 0.0009368896717205644 2023-01-21 12:46:54.345996: step: 1588/529, loss: 0.00045070648775435984 2023-01-21 12:46:55.450450: step: 1592/529, loss: 0.0002929687616415322 2023-01-21 12:46:56.594357: step: 1596/529, loss: 0.005933189298957586 2023-01-21 12:46:57.711739: step: 1600/529, loss: 2.269744800287299e-05 2023-01-21 12:46:58.830954: step: 1604/529, loss: 1.8501283193472773e-05 2023-01-21 12:46:59.930212: step: 1608/529, loss: 0.0010840415488928556 2023-01-21 12:47:01.077730: step: 1612/529, loss: 0.008745002560317516 2023-01-21 12:47:02.198902: step: 1616/529, loss: 0.0017584800953045487 2023-01-21 12:47:03.299435: step: 1620/529, loss: 7.629394644936838e-07 2023-01-21 12:47:04.418377: step: 1624/529, loss: 0.0005194664117880166 2023-01-21 12:47:05.486703: step: 1628/529, loss: 0.000526046787854284 2023-01-21 12:47:06.609795: step: 1632/529, loss: 0.011903000064194202 2023-01-21 12:47:07.714396: step: 1636/529, loss: 0.00925521831959486 2023-01-21 12:47:08.834234: step: 1640/529, loss: 3.814697629422881e-05 2023-01-21 12:47:09.937497: step: 1644/529, loss: 0.0025532725267112255 2023-01-21 12:47:11.041129: step: 1648/529, loss: 0.00042743684025481343 2023-01-21 12:47:12.183532: step: 1652/529, loss: 0.0001200675978907384 2023-01-21 12:47:13.315516: step: 1656/529, loss: 0.0008883416885510087 2023-01-21 12:47:14.424137: step: 1660/529, loss: 2.6226043701171875e-05 2023-01-21 12:47:15.548480: step: 1664/529, loss: 0.0276491641998291 2023-01-21 12:47:16.639870: step: 1668/529, loss: 0.0007463454967364669 2023-01-21 12:47:17.753066: step: 1672/529, loss: 7.190704491222277e-05 2023-01-21 12:47:18.876848: step: 1676/529, loss: 0.017272569239139557 2023-01-21 12:47:19.964664: step: 1680/529, loss: 0.02514667622745037 2023-01-21 12:47:21.065936: step: 1684/529, loss: 0.014276313595473766 2023-01-21 12:47:22.243409: step: 1688/529, loss: 0.012987899594008923 2023-01-21 12:47:23.345494: step: 1692/529, loss: 0.18673306703567505 2023-01-21 12:47:24.458731: step: 1696/529, loss: 0.000686264073010534 2023-01-21 12:47:25.593764: step: 1700/529, loss: 0.0017111777560785413 2023-01-21 12:47:26.714657: step: 1704/529, loss: 0.006614399142563343 2023-01-21 12:47:27.842193: step: 1708/529, loss: 0.0029592514038085938 2023-01-21 12:47:28.967496: step: 1712/529, loss: 0.04951363056898117 2023-01-21 12:47:30.074023: step: 1716/529, loss: 1.52587890625e-05 2023-01-21 12:47:31.168445: step: 1720/529, loss: 0.0018980979220941663 2023-01-21 12:47:32.299993: step: 1724/529, loss: 0.07986793667078018 2023-01-21 12:47:33.407439: step: 1728/529, loss: 0.0013504981761798263 2023-01-21 12:47:34.524444: step: 1732/529, loss: 0.013439178466796875 2023-01-21 12:47:35.660809: step: 1736/529, loss: 0.07295895367860794 2023-01-21 12:47:36.770675: step: 1740/529, loss: 4.634857032215223e-05 2023-01-21 12:47:37.875328: step: 1744/529, loss: 0.015958214178681374 2023-01-21 12:47:38.957490: step: 1748/529, loss: 0.00034971238346770406 2023-01-21 12:47:40.085577: step: 1752/529, loss: 0.021625613793730736 2023-01-21 12:47:41.192860: step: 1756/529, loss: 0.001922607421875 2023-01-21 12:47:42.289433: step: 1760/529, loss: 9.33647170313634e-05 2023-01-21 12:47:43.394813: step: 1764/529, loss: 9.946823411155492e-05 2023-01-21 12:47:44.486955: step: 1768/529, loss: 0.09372882544994354 2023-01-21 12:47:45.624027: step: 1772/529, loss: 0.029911041259765625 2023-01-21 12:47:46.740977: step: 1776/529, loss: 0.00015077591524459422 2023-01-21 12:47:47.848250: step: 1780/529, loss: 0.000301170366583392 2023-01-21 12:47:48.943081: step: 1784/529, loss: 0.0011503220302984118 2023-01-21 12:47:50.085574: step: 1788/529, loss: 4.113429546356201 2023-01-21 12:47:51.200550: step: 1792/529, loss: 0.003711128141731024 2023-01-21 12:47:52.329518: step: 1796/529, loss: 0.004560828674584627 2023-01-21 12:47:53.475599: step: 1800/529, loss: 0.003116512205451727 2023-01-21 12:47:54.615455: step: 1804/529, loss: 0.012748622335493565 2023-01-21 12:47:55.755362: step: 1808/529, loss: 0.00034503935603424907 2023-01-21 12:47:56.866334: step: 1812/529, loss: 0.0001386642543366179 2023-01-21 12:47:57.968795: step: 1816/529, loss: 0.002111673355102539 2023-01-21 12:47:59.059850: step: 1820/529, loss: 0.0020411014556884766 2023-01-21 12:48:00.173181: step: 1824/529, loss: 0.0006233692402020097 2023-01-21 12:48:01.274008: step: 1828/529, loss: 0.02498159557580948 2023-01-21 12:48:02.401012: step: 1832/529, loss: 0.0004571914905682206 2023-01-21 12:48:03.520706: step: 1836/529, loss: 0.019417762756347656 2023-01-21 12:48:04.660652: step: 1840/529, loss: 0.05352463573217392 2023-01-21 12:48:05.779241: step: 1844/529, loss: 0.009871482849121094 2023-01-21 12:48:06.898455: step: 1848/529, loss: 0.0005116462707519531 2023-01-21 12:48:08.021315: step: 1852/529, loss: 0.0006893158424645662 2023-01-21 12:48:09.150019: step: 1856/529, loss: 0.00017662048048805445 2023-01-21 12:48:10.275100: step: 1860/529, loss: 0.03502922132611275 2023-01-21 12:48:11.404945: step: 1864/529, loss: 0.004192352294921875 2023-01-21 12:48:12.505241: step: 1868/529, loss: 3.24249276673072e-06 2023-01-21 12:48:13.642569: step: 1872/529, loss: 0.004363632295280695 2023-01-21 12:48:14.730542: step: 1876/529, loss: 5.1403050747467205e-05 2023-01-21 12:48:15.844679: step: 1880/529, loss: 0.006640816107392311 2023-01-21 12:48:16.928993: step: 1884/529, loss: 0.0002799987851176411 2023-01-21 12:48:18.048096: step: 1888/529, loss: 0.00011196136620128527 2023-01-21 12:48:19.177273: step: 1892/529, loss: 0.00013818741717841476 2023-01-21 12:48:20.299683: step: 1896/529, loss: 0.0016016960144042969 2023-01-21 12:48:21.412921: step: 1900/529, loss: 0.0011799812782555819 2023-01-21 12:48:22.533948: step: 1904/529, loss: 0.0006229400751180947 2023-01-21 12:48:23.674705: step: 1908/529, loss: 0.043372441083192825 2023-01-21 12:48:24.798234: step: 1912/529, loss: 0.00036525726318359375 2023-01-21 12:48:25.901668: step: 1916/529, loss: 4.535161018371582 2023-01-21 12:48:27.022432: step: 1920/529, loss: 0.0005552291986532509 2023-01-21 12:48:28.182284: step: 1924/529, loss: 0.008031273260712624 2023-01-21 12:48:29.277719: step: 1928/529, loss: 0.017209816724061966 2023-01-21 12:48:30.419296: step: 1932/529, loss: 0.030343318358063698 2023-01-21 12:48:31.538882: step: 1936/529, loss: 0.06060619279742241 2023-01-21 12:48:32.663144: step: 1940/529, loss: 0.05756263807415962 2023-01-21 12:48:33.769569: step: 1944/529, loss: 2.460479663568549e-05 2023-01-21 12:48:34.894001: step: 1948/529, loss: 1.2540817806439009e-05 2023-01-21 12:48:35.999317: step: 1952/529, loss: 0.01860685460269451 2023-01-21 12:48:37.103477: step: 1956/529, loss: 0.0010583162074908614 2023-01-21 12:48:38.203721: step: 1960/529, loss: 7.629395213371026e-07 2023-01-21 12:48:39.309586: step: 1964/529, loss: 0.030353952199220657 2023-01-21 12:48:40.430282: step: 1968/529, loss: 0.016438055783510208 2023-01-21 12:48:41.531024: step: 1972/529, loss: 0.007317352574318647 2023-01-21 12:48:42.642272: step: 1976/529, loss: 0.00286617293022573 2023-01-21 12:48:43.779596: step: 1980/529, loss: 0.016619157046079636 2023-01-21 12:48:44.908026: step: 1984/529, loss: 0.00387821183539927 2023-01-21 12:48:46.012077: step: 1988/529, loss: 2.3937225705594756e-05 2023-01-21 12:48:47.130195: step: 1992/529, loss: 0.00010986327833961695 2023-01-21 12:48:48.253599: step: 1996/529, loss: 0.0067840577103197575 2023-01-21 12:48:49.349756: step: 2000/529, loss: 0.00044889451237395406 2023-01-21 12:48:50.482021: step: 2004/529, loss: 0.14281636476516724 2023-01-21 12:48:51.581672: step: 2008/529, loss: 0.00017395020404364914 2023-01-21 12:48:52.719375: step: 2012/529, loss: 1.0299681889591739e-05 2023-01-21 12:48:53.849726: step: 2016/529, loss: 0.021697044372558594 2023-01-21 12:48:54.951684: step: 2020/529, loss: 0.1025310531258583 2023-01-21 12:48:56.074439: step: 2024/529, loss: 0.049604035913944244 2023-01-21 12:48:57.202059: step: 2028/529, loss: 0.00198955531232059 2023-01-21 12:48:58.314030: step: 2032/529, loss: 0.0008503913995809853 2023-01-21 12:48:59.424665: step: 2036/529, loss: 0.03170070797204971 2023-01-21 12:49:00.529212: step: 2040/529, loss: 0.0010135649936273694 2023-01-21 12:49:01.638264: step: 2044/529, loss: 0.01690368726849556 2023-01-21 12:49:02.738411: step: 2048/529, loss: 0.028699208050966263 2023-01-21 12:49:03.873077: step: 2052/529, loss: 0.006765174679458141 2023-01-21 12:49:04.989787: step: 2056/529, loss: 0.0016994476318359375 2023-01-21 12:49:06.110977: step: 2060/529, loss: 0.002978420350700617 2023-01-21 12:49:07.223238: step: 2064/529, loss: 0.00122241978533566 2023-01-21 12:49:08.312791: step: 2068/529, loss: 0.0004337310674600303 2023-01-21 12:49:09.409031: step: 2072/529, loss: 0.001850128173828125 2023-01-21 12:49:10.516487: step: 2076/529, loss: 0.0004557609499897808 2023-01-21 12:49:11.589270: step: 2080/529, loss: 0.00013418197340797633 2023-01-21 12:49:12.719970: step: 2084/529, loss: 0.0005438804510049522 2023-01-21 12:49:13.810687: step: 2088/529, loss: 0.0394718199968338 2023-01-21 12:49:14.933567: step: 2092/529, loss: 0.02898426167666912 2023-01-21 12:49:16.058766: step: 2096/529, loss: 0.036324311047792435 2023-01-21 12:49:17.174763: step: 2100/529, loss: 0.0004966735723428428 2023-01-21 12:49:18.283003: step: 2104/529, loss: 0.022907542064785957 2023-01-21 12:49:19.392489: step: 2108/529, loss: 0.6578750610351562 2023-01-21 12:49:20.531422: step: 2112/529, loss: 0.0012264251708984375 2023-01-21 12:49:21.643280: step: 2116/529, loss: 0.0040412903763353825 ================================================== Loss: 0.037 -------------------- Dev: {'event': {'p': 0.5921325051759835, 'r': 0.7616511318242344, 'f1': 0.6662783925451369}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Test: {'event': {'p': 0.641958041958042, 'r': 0.7770880361173815, 'f1': 0.7030890988001022}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Chinese: {'event': {'p': 0.5375, 'r': 0.7962962962962963, 'f1': 0.6417910447761194}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} 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': 24} Russian: {'event': {'p': 0.38636363636363635, 'r': 0.4722222222222222, 'f1': 0.425}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 25 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:50:02.375946: step: 4/529, loss: 8.239746239269152e-05 2023-01-21 12:50:03.473099: step: 8/529, loss: 0.06492728739976883 2023-01-21 12:50:04.573798: step: 12/529, loss: 0.0010535240871831775 2023-01-21 12:50:05.681000: step: 16/529, loss: 2.6512147087487392e-05 2023-01-21 12:50:06.792744: step: 20/529, loss: 3.61442580469884e-05 2023-01-21 12:50:07.891957: step: 24/529, loss: 0.00010061264038085938 2023-01-21 12:50:09.006622: step: 28/529, loss: 2.574920654296875e-05 2023-01-21 12:50:10.099189: step: 32/529, loss: 0.0016889096004888415 2023-01-21 12:50:11.231507: step: 36/529, loss: 0.0001029014601954259 2023-01-21 12:50:12.343327: step: 40/529, loss: 0.0011649131774902344 2023-01-21 12:50:13.456290: step: 44/529, loss: 0.013121413998305798 2023-01-21 12:50:14.576506: step: 48/529, loss: 4.1675571992527694e-05 2023-01-21 12:50:15.709616: step: 52/529, loss: 0.014288235455751419 2023-01-21 12:50:16.802073: step: 56/529, loss: 6.053448305465281e-05 2023-01-21 12:50:17.892247: step: 60/529, loss: 0.0013905526138842106 2023-01-21 12:50:19.033706: step: 64/529, loss: 0.00577621441334486 2023-01-21 12:50:20.164151: step: 68/529, loss: 6.341934204101562e-05 2023-01-21 12:50:21.274698: step: 72/529, loss: 2.5844574338407256e-05 2023-01-21 12:50:22.381882: step: 76/529, loss: 0.004614830017089844 2023-01-21 12:50:23.474797: step: 80/529, loss: 0.0009738922235555947 2023-01-21 12:50:24.587917: step: 84/529, loss: 0.02246265485882759 2023-01-21 12:50:25.716865: step: 88/529, loss: 3.046989513677545e-05 2023-01-21 12:50:26.847518: step: 92/529, loss: 0.0001928329438669607 2023-01-21 12:50:27.990627: step: 96/529, loss: 0.0019365311600267887 2023-01-21 12:50:29.141662: step: 100/529, loss: 0.0017698288429528475 2023-01-21 12:50:30.262260: step: 104/529, loss: 0.02767029032111168 2023-01-21 12:50:31.375430: step: 108/529, loss: 5.3119660151423886e-05 2023-01-21 12:50:32.470988: step: 112/529, loss: 9.70840483205393e-05 2023-01-21 12:50:33.586266: step: 116/529, loss: 0.0028869628440588713 2023-01-21 12:50:34.686297: step: 120/529, loss: -1.239776611328125e-05 2023-01-21 12:50:35.800576: step: 124/529, loss: 0.00014681815810035914 2023-01-21 12:50:36.932972: step: 128/529, loss: 0.00014324189396575093 2023-01-21 12:50:38.049236: step: 132/529, loss: 0.0001350402890238911 2023-01-21 12:50:39.152478: step: 136/529, loss: 5.91278076171875e-05 2023-01-21 12:50:40.286039: step: 140/529, loss: 0.0089270593598485 2023-01-21 12:50:41.426119: step: 144/529, loss: 0.0009086608770303428 2023-01-21 12:50:42.548551: step: 148/529, loss: 0.0016155242919921875 2023-01-21 12:50:43.637118: step: 152/529, loss: 0.042005158960819244 2023-01-21 12:50:44.785854: step: 156/529, loss: 0.0006219864008016884 2023-01-21 12:50:45.915937: step: 160/529, loss: 0.0007740020519122481 2023-01-21 12:50:47.008138: step: 164/529, loss: 3.814697265625e-05 2023-01-21 12:50:48.139458: step: 168/529, loss: 0.0016374588012695312 2023-01-21 12:50:49.273387: step: 172/529, loss: 0.008846855722367764 2023-01-21 12:50:50.422182: step: 176/529, loss: 0.002421474317088723 2023-01-21 12:50:51.531421: step: 180/529, loss: 2.574920654296875e-05 2023-01-21 12:50:52.637150: step: 184/529, loss: 0.0017307281959801912 2023-01-21 12:50:53.744558: step: 188/529, loss: 0.0006530761602334678 2023-01-21 12:50:54.842127: step: 192/529, loss: 0.002630615374073386 2023-01-21 12:50:55.940558: step: 196/529, loss: 0.0028165339026600122 2023-01-21 12:50:57.046749: step: 200/529, loss: 0.016945648938417435 2023-01-21 12:50:58.174031: step: 204/529, loss: 0.005320549476891756 2023-01-21 12:50:59.282244: step: 208/529, loss: 0.048088837414979935 2023-01-21 12:51:00.357846: step: 212/529, loss: 1.5211106074275449e-05 2023-01-21 12:51:01.497615: step: 216/529, loss: 0.002927398541942239 2023-01-21 12:51:02.615661: step: 220/529, loss: 2.651214526849799e-05 2023-01-21 12:51:03.731510: step: 224/529, loss: 0.00015563964552711695 2023-01-21 12:51:04.847000: step: 228/529, loss: 0.005594635382294655 2023-01-21 12:51:05.973376: step: 232/529, loss: 0.03022041544318199 2023-01-21 12:51:07.074919: step: 236/529, loss: 0.0025920867919921875 2023-01-21 12:51:08.182740: step: 240/529, loss: 0.0674860030412674 2023-01-21 12:51:09.299859: step: 244/529, loss: 0.00022382737370207906 2023-01-21 12:51:10.391041: step: 248/529, loss: 0.1946265697479248 2023-01-21 12:51:11.490302: step: 252/529, loss: 0.0003417968691792339 2023-01-21 12:51:12.596651: step: 256/529, loss: 0.01829090155661106 2023-01-21 12:51:13.697072: step: 260/529, loss: 0.001224517822265625 2023-01-21 12:51:14.839670: step: 264/529, loss: 0.00019187928410246968 2023-01-21 12:51:15.954287: step: 268/529, loss: 0.004771900363266468 2023-01-21 12:51:17.091185: step: 272/529, loss: 0.00033273696317337453 2023-01-21 12:51:18.195387: step: 276/529, loss: 0.00021781922259833664 2023-01-21 12:51:19.320625: step: 280/529, loss: 0.0014924048446118832 2023-01-21 12:51:20.398478: step: 284/529, loss: 0.001709079835563898 2023-01-21 12:51:21.538225: step: 288/529, loss: 0.0004725456237792969 2023-01-21 12:51:22.632891: step: 292/529, loss: 0.009066772647202015 2023-01-21 12:51:23.758198: step: 296/529, loss: 0.010678482241928577 2023-01-21 12:51:24.897162: step: 300/529, loss: 0.0405920036137104 2023-01-21 12:51:25.995857: step: 304/529, loss: 9.91821252682712e-06 2023-01-21 12:51:27.123863: step: 308/529, loss: 0.00014743805513717234 2023-01-21 12:51:28.221117: step: 312/529, loss: 1.9931794668082148e-05 2023-01-21 12:51:29.304591: step: 316/529, loss: 0.01950206607580185 2023-01-21 12:51:30.427065: step: 320/529, loss: 0.07506103068590164 2023-01-21 12:51:31.533043: step: 324/529, loss: 9.965896424546372e-06 2023-01-21 12:51:32.672658: step: 328/529, loss: 0.04887409135699272 2023-01-21 12:51:33.766844: step: 332/529, loss: 0.006680631544440985 2023-01-21 12:51:34.868832: step: 336/529, loss: 0.19106845557689667 2023-01-21 12:51:36.004613: step: 340/529, loss: 0.00011348725092830136 2023-01-21 12:51:37.103781: step: 344/529, loss: 0.0013216972583904862 2023-01-21 12:51:38.227831: step: 348/529, loss: 0.014134597964584827 2023-01-21 12:51:39.322756: step: 352/529, loss: 7.581710815429688e-05 2023-01-21 12:51:40.448547: step: 356/529, loss: 0.0002537727414164692 2023-01-21 12:51:41.550440: step: 360/529, loss: 0.007184028625488281 2023-01-21 12:51:42.650968: step: 364/529, loss: 0.00045909881009720266 2023-01-21 12:51:43.781857: step: 368/529, loss: 0.008728695102036 2023-01-21 12:51:44.917372: step: 372/529, loss: 0.00012292862811591476 2023-01-21 12:51:46.048931: step: 376/529, loss: 0.06993818283081055 2023-01-21 12:51:47.147195: step: 380/529, loss: 0.00019149780564475805 2023-01-21 12:51:48.257193: step: 384/529, loss: 0.002189445775002241 2023-01-21 12:51:49.385560: step: 388/529, loss: 0.0017707825172692537 2023-01-21 12:51:50.491398: step: 392/529, loss: 0.00016074179438874125 2023-01-21 12:51:51.615799: step: 396/529, loss: 0.0007915973546914756 2023-01-21 12:51:52.762877: step: 400/529, loss: 0.006847381591796875 2023-01-21 12:51:53.873844: step: 404/529, loss: 0.0010313987731933594 2023-01-21 12:51:54.983143: step: 408/529, loss: 0.01844930648803711 2023-01-21 12:51:56.137110: step: 412/529, loss: 0.6985594034194946 2023-01-21 12:51:57.265255: step: 416/529, loss: 0.001399993896484375 2023-01-21 12:51:58.373835: step: 420/529, loss: 0.005720996763557196 2023-01-21 12:51:59.481952: step: 424/529, loss: 7.896423630882055e-05 2023-01-21 12:52:00.596950: step: 428/529, loss: 0.026981163769960403 2023-01-21 12:52:01.709962: step: 432/529, loss: 0.002355384873226285 2023-01-21 12:52:02.807776: step: 436/529, loss: 1.640319896978326e-05 2023-01-21 12:52:03.919600: step: 440/529, loss: 8.258819434558973e-05 2023-01-21 12:52:05.004438: step: 444/529, loss: 0.0003450393851380795 2023-01-21 12:52:06.113391: step: 448/529, loss: 0.001911258790642023 2023-01-21 12:52:07.243059: step: 452/529, loss: 6.160735938465223e-05 2023-01-21 12:52:08.339051: step: 456/529, loss: 0.1715896725654602 2023-01-21 12:52:09.469699: step: 460/529, loss: 0.000431919121183455 2023-01-21 12:52:10.587617: step: 464/529, loss: 2.7561187380342744e-05 2023-01-21 12:52:11.695165: step: 468/529, loss: 0.09603271633386612 2023-01-21 12:52:12.819985: step: 472/529, loss: 0.013927746564149857 2023-01-21 12:52:13.909440: step: 476/529, loss: 0.0016885757213458419 2023-01-21 12:52:15.014992: step: 480/529, loss: 0.008866596966981888 2023-01-21 12:52:16.139523: step: 484/529, loss: 0.0033962251618504524 2023-01-21 12:52:17.254409: step: 488/529, loss: 0.0012064933544024825 2023-01-21 12:52:18.339857: step: 492/529, loss: 0.00138940813485533 2023-01-21 12:52:19.463706: step: 496/529, loss: 0.007605123333632946 2023-01-21 12:52:20.641834: step: 500/529, loss: 0.0014417648781090975 2023-01-21 12:52:21.786440: step: 504/529, loss: 0.014421463012695312 2023-01-21 12:52:22.921850: step: 508/529, loss: 0.002451992128044367 2023-01-21 12:52:24.042702: step: 512/529, loss: 0.0010552406311035156 2023-01-21 12:52:25.127844: step: 516/529, loss: 0.022318458184599876 2023-01-21 12:52:26.234532: step: 520/529, loss: 0.004051971714943647 2023-01-21 12:52:27.340108: step: 524/529, loss: 0.007374858483672142 2023-01-21 12:52:28.440970: step: 528/529, loss: 0.0007042884826660156 2023-01-21 12:52:29.581787: step: 532/529, loss: 0.0037859915755689144 2023-01-21 12:52:30.693504: step: 536/529, loss: 0.002202987438067794 2023-01-21 12:52:31.786897: step: 540/529, loss: 9.679794311523438e-05 2023-01-21 12:52:32.896053: step: 544/529, loss: 5.1975250244140625e-05 2023-01-21 12:52:34.040738: step: 548/529, loss: 0.017271708697080612 2023-01-21 12:52:35.174835: step: 552/529, loss: 0.06082115322351456 2023-01-21 12:52:36.296900: step: 556/529, loss: 0.00034880638122558594 2023-01-21 12:52:37.429299: step: 560/529, loss: 0.0004460811906028539 2023-01-21 12:52:38.547356: step: 564/529, loss: 0.0010293960804119706 2023-01-21 12:52:39.683511: step: 568/529, loss: 0.0021224976517260075 2023-01-21 12:52:40.797011: step: 572/529, loss: 6.809234764659777e-05 2023-01-21 12:52:41.913752: step: 576/529, loss: 0.010007381439208984 2023-01-21 12:52:43.011442: step: 580/529, loss: 1.373290979245212e-05 2023-01-21 12:52:44.131038: step: 584/529, loss: 7.152556918299524e-06 2023-01-21 12:52:45.270534: step: 588/529, loss: 0.0261096004396677 2023-01-21 12:52:46.389657: step: 592/529, loss: 0.025314997881650925 2023-01-21 12:52:47.494683: step: 596/529, loss: 0.007330894470214844 2023-01-21 12:52:48.613417: step: 600/529, loss: 0.023163508623838425 2023-01-21 12:52:49.716437: step: 604/529, loss: 0.012703800573945045 2023-01-21 12:52:50.840806: step: 608/529, loss: 0.005654830019921064 2023-01-21 12:52:51.958254: step: 612/529, loss: 0.0005739211919717491 2023-01-21 12:52:53.053626: step: 616/529, loss: 0.0010097504127770662 2023-01-21 12:52:54.166816: step: 620/529, loss: 0.0003084182972088456 2023-01-21 12:52:55.281545: step: 624/529, loss: 0.00014472007751464844 2023-01-21 12:52:56.416996: step: 628/529, loss: 0.00014610291691496968 2023-01-21 12:52:57.505145: step: 632/529, loss: 0.0011796951293945312 2023-01-21 12:52:58.603504: step: 636/529, loss: 0.00035533905611373484 2023-01-21 12:52:59.702738: step: 640/529, loss: 0.000594711338635534 2023-01-21 12:53:00.820817: step: 644/529, loss: 0.006466484162956476 2023-01-21 12:53:01.944775: step: 648/529, loss: 0.00022411346435546875 2023-01-21 12:53:03.062777: step: 652/529, loss: 0.0021045685280114412 2023-01-21 12:53:04.191614: step: 656/529, loss: 0.0012584686046466231 2023-01-21 12:53:05.316349: step: 660/529, loss: 0.04673194885253906 2023-01-21 12:53:06.418398: step: 664/529, loss: 0.0001146316499216482 2023-01-21 12:53:07.544881: step: 668/529, loss: 0.0004225253942422569 2023-01-21 12:53:08.688759: step: 672/529, loss: 0.0008788108825683594 2023-01-21 12:53:09.810835: step: 676/529, loss: 0.047112368047237396 2023-01-21 12:53:10.906691: step: 680/529, loss: 0.40907785296440125 2023-01-21 12:53:12.020126: step: 684/529, loss: 0.0010457038879394531 2023-01-21 12:53:13.136973: step: 688/529, loss: 0.02270498313009739 2023-01-21 12:53:14.237989: step: 692/529, loss: 0.02062349207699299 2023-01-21 12:53:15.356750: step: 696/529, loss: 0.00029754638671875 2023-01-21 12:53:16.482481: step: 700/529, loss: 0.0008483410347253084 2023-01-21 12:53:17.594764: step: 704/529, loss: 0.002695751143619418 2023-01-21 12:53:18.692649: step: 708/529, loss: 0.08103179931640625 2023-01-21 12:53:19.797669: step: 712/529, loss: 0.024988271296024323 2023-01-21 12:53:20.926566: step: 716/529, loss: 0.0007702827570028603 2023-01-21 12:53:22.036851: step: 720/529, loss: 4.425048973644152e-05 2023-01-21 12:53:23.156227: step: 724/529, loss: 0.0004734992980957031 2023-01-21 12:53:24.251941: step: 728/529, loss: 0.00011892318434547633 2023-01-21 12:53:25.378720: step: 732/529, loss: 0.005088520236313343 2023-01-21 12:53:26.506866: step: 736/529, loss: 0.009314918890595436 2023-01-21 12:53:27.625037: step: 740/529, loss: 0.005568313878029585 2023-01-21 12:53:28.729840: step: 744/529, loss: 0.03599729761481285 2023-01-21 12:53:29.823047: step: 748/529, loss: 5.14984139954322e-06 2023-01-21 12:53:30.919613: step: 752/529, loss: 0.02781515195965767 2023-01-21 12:53:32.034541: step: 756/529, loss: 0.1732281595468521 2023-01-21 12:53:33.181338: step: 760/529, loss: 0.009294891729950905 2023-01-21 12:53:34.300537: step: 764/529, loss: 0.026825524866580963 2023-01-21 12:53:35.415272: step: 768/529, loss: 0.00013275146193336695 2023-01-21 12:53:36.541168: step: 772/529, loss: 0.00018062590970657766 2023-01-21 12:53:37.638648: step: 776/529, loss: 0.0003990173281636089 2023-01-21 12:53:38.741084: step: 780/529, loss: 0.01617913320660591 2023-01-21 12:53:39.861425: step: 784/529, loss: 0.008256912231445312 2023-01-21 12:53:40.970058: step: 788/529, loss: 0.02213773876428604 2023-01-21 12:53:42.071255: step: 792/529, loss: 0.0031621933449059725 2023-01-21 12:53:43.182824: step: 796/529, loss: 0.07532644271850586 2023-01-21 12:53:44.303591: step: 800/529, loss: 0.0036155700217932463 2023-01-21 12:53:45.445776: step: 804/529, loss: 0.0035490035079419613 2023-01-21 12:53:46.547653: step: 808/529, loss: 7.724762326688506e-06 2023-01-21 12:53:47.677748: step: 812/529, loss: 0.015631485730409622 2023-01-21 12:53:48.819462: step: 816/529, loss: 0.06131019815802574 2023-01-21 12:53:49.927859: step: 820/529, loss: 0.02664356306195259 2023-01-21 12:53:51.033104: step: 824/529, loss: 0.010900306515395641 2023-01-21 12:53:52.140469: step: 828/529, loss: 0.001399993896484375 2023-01-21 12:53:53.247778: step: 832/529, loss: 0.002303504850715399 2023-01-21 12:53:54.365816: step: 836/529, loss: 0.0019275665981695056 2023-01-21 12:53:55.512025: step: 840/529, loss: 0.0014806748367846012 2023-01-21 12:53:56.600326: step: 844/529, loss: 0.004483222961425781 2023-01-21 12:53:57.711905: step: 848/529, loss: 0.031519509851932526 2023-01-21 12:53:58.821115: step: 852/529, loss: 0.0001905441313283518 2023-01-21 12:53:59.924717: step: 856/529, loss: 0.0016553879249840975 2023-01-21 12:54:01.055288: step: 860/529, loss: 0.028945446014404297 2023-01-21 12:54:02.173363: step: 864/529, loss: 0.0028638362418860197 2023-01-21 12:54:03.269101: step: 868/529, loss: 0.002261257264763117 2023-01-21 12:54:04.443893: step: 872/529, loss: 0.00027875902014784515 2023-01-21 12:54:05.555390: step: 876/529, loss: 3.14235694531817e-05 2023-01-21 12:54:06.661693: step: 880/529, loss: 6.065369234420359e-05 2023-01-21 12:54:07.792868: step: 884/529, loss: 0.015969276428222656 2023-01-21 12:54:08.897060: step: 888/529, loss: 0.00020036697969771922 2023-01-21 12:54:10.009118: step: 892/529, loss: 0.0001226902095368132 2023-01-21 12:54:11.128705: step: 896/529, loss: 0.010181712917983532 2023-01-21 12:54:12.256241: step: 900/529, loss: 0.00031824110192246735 2023-01-21 12:54:13.376804: step: 904/529, loss: 0.02685851976275444 2023-01-21 12:54:14.495483: step: 908/529, loss: -1.068115216185106e-05 2023-01-21 12:54:15.581614: step: 912/529, loss: 0.0007848739624023438 2023-01-21 12:54:16.687374: step: 916/529, loss: 0.017832279205322266 2023-01-21 12:54:17.819276: step: 920/529, loss: 0.017981721088290215 2023-01-21 12:54:18.951531: step: 924/529, loss: 0.8941981792449951 2023-01-21 12:54:20.067876: step: 928/529, loss: 0.0027111053932458162 2023-01-21 12:54:21.207411: step: 932/529, loss: 0.0010915757156908512 2023-01-21 12:54:22.306178: step: 936/529, loss: 0.007136440835893154 2023-01-21 12:54:23.431137: step: 940/529, loss: 6.84738188283518e-05 2023-01-21 12:54:24.556285: step: 944/529, loss: 0.00024788378505036235 2023-01-21 12:54:25.661659: step: 948/529, loss: 3.266334533691406e-05 2023-01-21 12:54:26.759839: step: 952/529, loss: 5.846023850608617e-05 2023-01-21 12:54:27.880717: step: 956/529, loss: 0.0003006935294251889 2023-01-21 12:54:28.999849: step: 960/529, loss: 0.0004750251828227192 2023-01-21 12:54:30.135387: step: 964/529, loss: 0.0004093170282430947 2023-01-21 12:54:31.202433: step: 968/529, loss: 0.006163883488625288 2023-01-21 12:54:32.333281: step: 972/529, loss: 0.0003914833068847656 2023-01-21 12:54:33.461487: step: 976/529, loss: 0.02201995812356472 2023-01-21 12:54:34.596026: step: 980/529, loss: 0.0007093429449014366 2023-01-21 12:54:35.700716: step: 984/529, loss: 0.01027908269315958 2023-01-21 12:54:36.867123: step: 988/529, loss: 0.0004440307675395161 2023-01-21 12:54:37.980132: step: 992/529, loss: 0.05865812301635742 2023-01-21 12:54:39.085699: step: 996/529, loss: 0.024614524096250534 2023-01-21 12:54:40.195019: step: 1000/529, loss: 0.0036050318740308285 2023-01-21 12:54:41.315300: step: 1004/529, loss: 0.0009887695778161287 2023-01-21 12:54:42.438079: step: 1008/529, loss: 0.024555588141083717 2023-01-21 12:54:43.552984: step: 1012/529, loss: 0.00018520356388762593 2023-01-21 12:54:44.682303: step: 1016/529, loss: 0.0006617546314373612 2023-01-21 12:54:45.805191: step: 1020/529, loss: 0.001563263009302318 2023-01-21 12:54:46.946781: step: 1024/529, loss: 0.00018014907254837453 2023-01-21 12:54:48.074868: step: 1028/529, loss: 0.00782079715281725 2023-01-21 12:54:49.212169: step: 1032/529, loss: 0.0005350112915039062 2023-01-21 12:54:50.326458: step: 1036/529, loss: 0.0004913330194540322 2023-01-21 12:54:51.498636: step: 1040/529, loss: 0.0004896164173260331 2023-01-21 12:54:52.606832: step: 1044/529, loss: 0.0010759353172034025 2023-01-21 12:54:53.710789: step: 1048/529, loss: 0.01298608910292387 2023-01-21 12:54:54.820833: step: 1052/529, loss: 0.00133686070330441 2023-01-21 12:54:55.912580: step: 1056/529, loss: 0.008045865222811699 2023-01-21 12:54:57.015557: step: 1060/529, loss: 0.0009768486488610506 2023-01-21 12:54:58.140770: step: 1064/529, loss: 0.0010396003490313888 2023-01-21 12:54:59.224894: step: 1068/529, loss: 0.013844775967299938 2023-01-21 12:55:00.310575: step: 1072/529, loss: 0.04795026779174805 2023-01-21 12:55:01.438407: step: 1076/529, loss: 0.04873504862189293 2023-01-21 12:55:02.534308: step: 1080/529, loss: 0.0020037651993334293 2023-01-21 12:55:03.657301: step: 1084/529, loss: 0.00018558502779342234 2023-01-21 12:55:04.786815: step: 1088/529, loss: 0.029660606756806374 2023-01-21 12:55:05.915861: step: 1092/529, loss: 0.0016973495949059725 2023-01-21 12:55:07.072561: step: 1096/529, loss: 0.0004295349062886089 2023-01-21 12:55:08.172363: step: 1100/529, loss: 0.0003998756583314389 2023-01-21 12:55:09.298529: step: 1104/529, loss: 9.365082223666832e-05 2023-01-21 12:55:10.446203: step: 1108/529, loss: 0.003821754362434149 2023-01-21 12:55:11.532980: step: 1112/529, loss: 0.010478687472641468 2023-01-21 12:55:12.647961: step: 1116/529, loss: 0.00570602435618639 2023-01-21 12:55:13.792038: step: 1120/529, loss: 0.0005778313498012722 2023-01-21 12:55:14.920049: step: 1124/529, loss: 0.0006515979766845703 2023-01-21 12:55:16.036965: step: 1128/529, loss: 0.00044827460078522563 2023-01-21 12:55:17.132286: step: 1132/529, loss: 0.002516245935112238 2023-01-21 12:55:18.281469: step: 1136/529, loss: 0.14684423804283142 2023-01-21 12:55:19.412083: step: 1140/529, loss: 0.0006649970891885459 2023-01-21 12:55:20.527564: step: 1144/529, loss: 2.346038854739163e-05 2023-01-21 12:55:21.642332: step: 1148/529, loss: 0.0003563404025044292 2023-01-21 12:55:22.780790: step: 1152/529, loss: 0.0018716811900958419 2023-01-21 12:55:23.920687: step: 1156/529, loss: 0.03818530961871147 2023-01-21 12:55:25.059463: step: 1160/529, loss: 0.010581206530332565 2023-01-21 12:55:26.183696: step: 1164/529, loss: 0.0004951476585119963 2023-01-21 12:55:27.321767: step: 1168/529, loss: 0.0015140533214434981 2023-01-21 12:55:28.430319: step: 1172/529, loss: 3.528594970703125e-05 2023-01-21 12:55:29.552295: step: 1176/529, loss: 0.2581014633178711 2023-01-21 12:55:30.652924: step: 1180/529, loss: 0.007263565436005592 2023-01-21 12:55:31.750818: step: 1184/529, loss: 0.0006606102106161416 2023-01-21 12:55:32.835334: step: 1188/529, loss: 0.004563236143440008 2023-01-21 12:55:33.957975: step: 1192/529, loss: 0.009277725592255592 2023-01-21 12:55:35.056338: step: 1196/529, loss: 0.00015840530977584422 2023-01-21 12:55:36.165522: step: 1200/529, loss: 0.001851844834163785 2023-01-21 12:55:37.340865: step: 1204/529, loss: 0.009067822247743607 2023-01-21 12:55:38.419805: step: 1208/529, loss: 0.00018033981905318797 2023-01-21 12:55:39.522314: step: 1212/529, loss: 0.001878175069577992 2023-01-21 12:55:40.619171: step: 1216/529, loss: 0.002087879227474332 2023-01-21 12:55:41.739512: step: 1220/529, loss: 0.03204955905675888 2023-01-21 12:55:42.874224: step: 1224/529, loss: 0.00027046201284974813 2023-01-21 12:55:44.005014: step: 1228/529, loss: 0.002836894942447543 2023-01-21 12:55:45.152070: step: 1232/529, loss: 0.0004068374400958419 2023-01-21 12:55:46.264837: step: 1236/529, loss: 2.746581958490424e-05 2023-01-21 12:55:47.409869: step: 1240/529, loss: 0.09197786450386047 2023-01-21 12:55:48.508637: step: 1244/529, loss: 0.056519556790590286 2023-01-21 12:55:49.615460: step: 1248/529, loss: 0.006399346049875021 2023-01-21 12:55:50.711985: step: 1252/529, loss: 0.0005100727430544794 2023-01-21 12:55:51.822180: step: 1256/529, loss: 0.00021066666522528976 2023-01-21 12:55:52.916579: step: 1260/529, loss: 0.007651233579963446 2023-01-21 12:55:54.047940: step: 1264/529, loss: 0.0013234138023108244 2023-01-21 12:55:55.136020: step: 1268/529, loss: 0.06373654305934906 2023-01-21 12:55:56.236076: step: 1272/529, loss: -1.4877318790240679e-05 2023-01-21 12:55:57.330538: step: 1276/529, loss: 1.6212466107390355e-06 2023-01-21 12:55:58.430240: step: 1280/529, loss: 0.0004873276047874242 2023-01-21 12:55:59.558920: step: 1284/529, loss: 0.011513615027070045 2023-01-21 12:56:00.664226: step: 1288/529, loss: 0.0006726265419274569 2023-01-21 12:56:01.765779: step: 1292/529, loss: 0.0005554198869504035 2023-01-21 12:56:02.877642: step: 1296/529, loss: 0.02368316799402237 2023-01-21 12:56:04.000173: step: 1300/529, loss: 0.019910622388124466 2023-01-21 12:56:05.121434: step: 1304/529, loss: 0.03583727404475212 2023-01-21 12:56:06.232639: step: 1308/529, loss: 0.0019070626003667712 2023-01-21 12:56:07.328605: step: 1312/529, loss: 2.8856097742391285e-06 2023-01-21 12:56:08.445816: step: 1316/529, loss: 0.0004150390741415322 2023-01-21 12:56:09.571508: step: 1320/529, loss: 0.00015940220328047872 2023-01-21 12:56:10.718834: step: 1324/529, loss: 0.00286788959056139 2023-01-21 12:56:11.839878: step: 1328/529, loss: 0.0075577739626169205 2023-01-21 12:56:12.950902: step: 1332/529, loss: 0.00010166168067371473 2023-01-21 12:56:14.074233: step: 1336/529, loss: 0.0002239704190287739 2023-01-21 12:56:15.213904: step: 1340/529, loss: 0.4347163736820221 2023-01-21 12:56:16.318793: step: 1344/529, loss: 1.0013580322265625e-05 2023-01-21 12:56:17.440328: step: 1348/529, loss: 0.0404568687081337 2023-01-21 12:56:18.547390: step: 1352/529, loss: 0.007117748726159334 2023-01-21 12:56:19.661478: step: 1356/529, loss: 0.00026063917903229594 2023-01-21 12:56:20.767902: step: 1360/529, loss: 0.016696738079190254 2023-01-21 12:56:21.913246: step: 1364/529, loss: 0.04320878908038139 2023-01-21 12:56:23.057612: step: 1368/529, loss: 0.0007826805231161416 2023-01-21 12:56:24.166565: step: 1372/529, loss: 0.005545139312744141 2023-01-21 12:56:25.243545: step: 1376/529, loss: 0.0026499389205127954 2023-01-21 12:56:26.342618: step: 1380/529, loss: 0.000301361083984375 2023-01-21 12:56:27.443625: step: 1384/529, loss: 0.00021457672119140625 2023-01-21 12:56:28.543595: step: 1388/529, loss: 0.00017261505126953125 2023-01-21 12:56:29.664589: step: 1392/529, loss: 0.001270294189453125 2023-01-21 12:56:30.788796: step: 1396/529, loss: 0.006884270813316107 2023-01-21 12:56:31.896735: step: 1400/529, loss: 0.003435802645981312 2023-01-21 12:56:33.005853: step: 1404/529, loss: 0.0012341499095782638 2023-01-21 12:56:34.119774: step: 1408/529, loss: 0.007880115881562233 2023-01-21 12:56:35.273904: step: 1412/529, loss: 0.003738784696906805 2023-01-21 12:56:36.382092: step: 1416/529, loss: 0.000274658203125 2023-01-21 12:56:37.511321: step: 1420/529, loss: -1.3828277587890625e-05 2023-01-21 12:56:38.639688: step: 1424/529, loss: 0.01891174353659153 2023-01-21 12:56:39.751640: step: 1428/529, loss: 0.04813346639275551 2023-01-21 12:56:40.878196: step: 1432/529, loss: 0.02498798258602619 2023-01-21 12:56:41.994957: step: 1436/529, loss: 0.017072487622499466 2023-01-21 12:56:43.091249: step: 1440/529, loss: 0.005184078589081764 2023-01-21 12:56:44.211989: step: 1444/529, loss: 0.00177173619158566 2023-01-21 12:56:45.315206: step: 1448/529, loss: 0.001831150148063898 2023-01-21 12:56:46.455629: step: 1452/529, loss: 0.002214884851127863 2023-01-21 12:56:47.600006: step: 1456/529, loss: 0.0010780334705486894 2023-01-21 12:56:48.732619: step: 1460/529, loss: 0.0035539628006517887 2023-01-21 12:56:49.856468: step: 1464/529, loss: 0.010195350274443626 2023-01-21 12:56:50.960419: step: 1468/529, loss: 7.381439354503527e-05 2023-01-21 12:56:52.061739: step: 1472/529, loss: 0.00023307801166083664 2023-01-21 12:56:53.176539: step: 1476/529, loss: 7.03811674611643e-05 2023-01-21 12:56:54.269851: step: 1480/529, loss: 0.021119307726621628 2023-01-21 12:56:55.372575: step: 1484/529, loss: 0.00022010804968886077 2023-01-21 12:56:56.464596: step: 1488/529, loss: 0.005672645289450884 2023-01-21 12:56:57.547788: step: 1492/529, loss: 5.5027012422215194e-05 2023-01-21 12:56:58.648281: step: 1496/529, loss: 0.008964920416474342 2023-01-21 12:56:59.805856: step: 1500/529, loss: 0.004028368275612593 2023-01-21 12:57:00.890083: step: 1504/529, loss: 0.015865325927734375 2023-01-21 12:57:02.035224: step: 1508/529, loss: 0.7701981067657471 2023-01-21 12:57:03.133288: step: 1512/529, loss: -1.1920928955078125e-06 2023-01-21 12:57:04.250635: step: 1516/529, loss: 0.014901733957231045 2023-01-21 12:57:05.394909: step: 1520/529, loss: 0.0008633613470010459 2023-01-21 12:57:06.523960: step: 1524/529, loss: 0.002908897353336215 2023-01-21 12:57:07.670188: step: 1528/529, loss: -4.673003786592744e-06 2023-01-21 12:57:08.798162: step: 1532/529, loss: 0.0021669387351721525 2023-01-21 12:57:09.904672: step: 1536/529, loss: 0.5673606991767883 2023-01-21 12:57:10.997902: step: 1540/529, loss: 0.0021881582215428352 2023-01-21 12:57:12.127792: step: 1544/529, loss: 0.05304775387048721 2023-01-21 12:57:13.240526: step: 1548/529, loss: 3.4332275390625e-05 2023-01-21 12:57:14.371115: step: 1552/529, loss: 0.008408928290009499 2023-01-21 12:57:15.471974: step: 1556/529, loss: 2.1743775505456142e-05 2023-01-21 12:57:16.594985: step: 1560/529, loss: 0.05852699652314186 2023-01-21 12:57:17.691603: step: 1564/529, loss: 0.0036611557006835938 2023-01-21 12:57:18.802078: step: 1568/529, loss: 5.550384594243951e-05 2023-01-21 12:57:19.899458: step: 1572/529, loss: 0.00034048559609800577 2023-01-21 12:57:21.000672: step: 1576/529, loss: 0.08137702941894531 2023-01-21 12:57:22.146669: step: 1580/529, loss: 0.0020496367942541838 2023-01-21 12:57:23.260873: step: 1584/529, loss: 0.03141505643725395 2023-01-21 12:57:24.355433: step: 1588/529, loss: 0.0007108688587322831 2023-01-21 12:57:25.492311: step: 1592/529, loss: 0.00016412735567428172 2023-01-21 12:57:26.651510: step: 1596/529, loss: 1.2492332458496094 2023-01-21 12:57:27.753738: step: 1600/529, loss: 0.04213924705982208 2023-01-21 12:57:28.872441: step: 1604/529, loss: 0.0007680415874347091 2023-01-21 12:57:29.973068: step: 1608/529, loss: 1.106262243411038e-05 2023-01-21 12:57:31.095782: step: 1612/529, loss: 0.002388191409409046 2023-01-21 12:57:32.235005: step: 1616/529, loss: 0.015336036682128906 2023-01-21 12:57:33.336967: step: 1620/529, loss: 0.00013313292583916336 2023-01-21 12:57:34.460505: step: 1624/529, loss: 0.004322623834013939 2023-01-21 12:57:35.557527: step: 1628/529, loss: 0.01439743023365736 2023-01-21 12:57:36.703970: step: 1632/529, loss: 0.013621330261230469 2023-01-21 12:57:37.816891: step: 1636/529, loss: 2.7370453608455136e-05 2023-01-21 12:57:38.933941: step: 1640/529, loss: 1.4066696166992188e-05 2023-01-21 12:57:40.072747: step: 1644/529, loss: 0.00033760073711164296 2023-01-21 12:57:41.172062: step: 1648/529, loss: 0.04081978648900986 2023-01-21 12:57:42.305852: step: 1652/529, loss: 0.0015802383422851562 2023-01-21 12:57:43.473570: step: 1656/529, loss: 0.01882457733154297 2023-01-21 12:57:44.574322: step: 1660/529, loss: 0.0020250321831554174 2023-01-21 12:57:45.661040: step: 1664/529, loss: 0.003597450442612171 2023-01-21 12:57:46.814641: step: 1668/529, loss: 0.035089682787656784 2023-01-21 12:57:47.931329: step: 1672/529, loss: 0.003991127014160156 2023-01-21 12:57:49.036078: step: 1676/529, loss: 1.049041748046875e-05 2023-01-21 12:57:50.113647: step: 1680/529, loss: 0.006166267674416304 2023-01-21 12:57:51.245970: step: 1684/529, loss: 0.025664711371064186 2023-01-21 12:57:52.330609: step: 1688/529, loss: 0.3616229295730591 2023-01-21 12:57:53.464554: step: 1692/529, loss: 0.0004459381161723286 2023-01-21 12:57:54.595021: step: 1696/529, loss: 0.010399913415312767 2023-01-21 12:57:55.698576: step: 1700/529, loss: 0.020368624478578568 2023-01-21 12:57:56.818032: step: 1704/529, loss: 0.0010580063099041581 2023-01-21 12:57:57.960432: step: 1708/529, loss: 3.080368333030492e-05 2023-01-21 12:57:59.087354: step: 1712/529, loss: 0.05797433853149414 2023-01-21 12:58:00.198166: step: 1716/529, loss: 0.012995148077607155 2023-01-21 12:58:01.312976: step: 1720/529, loss: 0.00011148452904308215 2023-01-21 12:58:02.418849: step: 1724/529, loss: 0.04312143474817276 2023-01-21 12:58:03.528712: step: 1728/529, loss: 0.04819145053625107 2023-01-21 12:58:04.624338: step: 1732/529, loss: 0.024585723876953125 2023-01-21 12:58:05.723887: step: 1736/529, loss: 0.0012557030422613025 2023-01-21 12:58:06.848615: step: 1740/529, loss: 0.010304260067641735 2023-01-21 12:58:07.924817: step: 1744/529, loss: 0.0007087707635946572 2023-01-21 12:58:09.045682: step: 1748/529, loss: 0.0021219253540039062 2023-01-21 12:58:10.164414: step: 1752/529, loss: 0.0021059990394860506 2023-01-21 12:58:11.272241: step: 1756/529, loss: 4.1770937968976796e-05 2023-01-21 12:58:12.408315: step: 1760/529, loss: 0.0029232024680823088 2023-01-21 12:58:13.520503: step: 1764/529, loss: 0.0004554748593363911 2023-01-21 12:58:14.605033: step: 1768/529, loss: 1.7547608877066523e-05 2023-01-21 12:58:15.724216: step: 1772/529, loss: 0.06798458099365234 2023-01-21 12:58:16.856865: step: 1776/529, loss: 0.03369732201099396 2023-01-21 12:58:17.933901: step: 1780/529, loss: 0.0004027843533549458 2023-01-21 12:58:19.036268: step: 1784/529, loss: 0.014034987427294254 2023-01-21 12:58:20.168711: step: 1788/529, loss: 0.0003894329129252583 2023-01-21 12:58:21.284580: step: 1792/529, loss: 0.0487096793949604 2023-01-21 12:58:22.376220: step: 1796/529, loss: 0.00046291350736282766 2023-01-21 12:58:23.517136: step: 1800/529, loss: 0.0002025604189839214 2023-01-21 12:58:24.619392: step: 1804/529, loss: 0.02630739100277424 2023-01-21 12:58:25.762279: step: 1808/529, loss: 0.0006984710926190019 2023-01-21 12:58:26.873591: step: 1812/529, loss: 0.009124946780502796 2023-01-21 12:58:28.013817: step: 1816/529, loss: 0.05882749333977699 2023-01-21 12:58:29.142551: step: 1820/529, loss: 3.237724376958795e-05 2023-01-21 12:58:30.252787: step: 1824/529, loss: 0.0009759903186932206 2023-01-21 12:58:31.396735: step: 1828/529, loss: 0.07676620781421661 2023-01-21 12:58:32.492547: step: 1832/529, loss: 0.00011663437180686742 2023-01-21 12:58:33.621698: step: 1836/529, loss: 0.021224021911621094 2023-01-21 12:58:34.737660: step: 1840/529, loss: 0.019350243732333183 2023-01-21 12:58:35.833464: step: 1844/529, loss: 0.007243919186294079 2023-01-21 12:58:36.948511: step: 1848/529, loss: 0.0014892577892169356 2023-01-21 12:58:38.069300: step: 1852/529, loss: 0.030296802520751953 2023-01-21 12:58:39.184024: step: 1856/529, loss: 0.008112716488540173 2023-01-21 12:58:40.314643: step: 1860/529, loss: 0.00026006699772551656 2023-01-21 12:58:41.419893: step: 1864/529, loss: 0.007880019955337048 2023-01-21 12:58:42.531212: step: 1868/529, loss: 0.0003829956112895161 2023-01-21 12:58:43.663601: step: 1872/529, loss: 0.07741031050682068 2023-01-21 12:58:44.797127: step: 1876/529, loss: 0.028162576258182526 2023-01-21 12:58:45.930784: step: 1880/529, loss: 0.0015431403880938888 2023-01-21 12:58:47.044618: step: 1884/529, loss: 0.004068565554916859 2023-01-21 12:58:48.140545: step: 1888/529, loss: 0.0016448021633550525 2023-01-21 12:58:49.240277: step: 1892/529, loss: 0.009966659359633923 2023-01-21 12:58:50.328902: step: 1896/529, loss: 0.00016231538029387593 2023-01-21 12:58:51.425401: step: 1900/529, loss: 0.0016377449501305819 2023-01-21 12:58:52.569697: step: 1904/529, loss: 0.0023283003829419613 2023-01-21 12:58:53.704529: step: 1908/529, loss: 0.00015869141498114914 2023-01-21 12:58:54.818276: step: 1912/529, loss: 0.0021070719230920076 2023-01-21 12:58:55.905967: step: 1916/529, loss: 7.781982276355848e-05 2023-01-21 12:58:57.048309: step: 1920/529, loss: 0.0002738475741352886 2023-01-21 12:58:58.222146: step: 1924/529, loss: 0.0158906951546669 2023-01-21 12:58:59.354809: step: 1928/529, loss: 0.0011287688976153731 2023-01-21 12:59:00.442781: step: 1932/529, loss: 0.009584140963852406 2023-01-21 12:59:01.582744: step: 1936/529, loss: 0.0729038268327713 2023-01-21 12:59:02.676586: step: 1940/529, loss: 0.009826993569731712 2023-01-21 12:59:03.800068: step: 1944/529, loss: 0.028536701574921608 2023-01-21 12:59:04.900210: step: 1948/529, loss: 0.00018758774967864156 2023-01-21 12:59:05.994569: step: 1952/529, loss: 0.00035581589327193797 2023-01-21 12:59:07.112388: step: 1956/529, loss: -6.67572021484375e-06 2023-01-21 12:59:08.227461: step: 1960/529, loss: 6.680488877464086e-05 2023-01-21 12:59:09.349341: step: 1964/529, loss: 1.912116931634955e-05 2023-01-21 12:59:10.488373: step: 1968/529, loss: 0.00015220641216728836 2023-01-21 12:59:11.589474: step: 1972/529, loss: 0.0015476227272301912 2023-01-21 12:59:12.699846: step: 1976/529, loss: 0.00914688128978014 2023-01-21 12:59:13.792832: step: 1980/529, loss: 0.00144872663076967 2023-01-21 12:59:14.899605: step: 1984/529, loss: 0.0004364967462606728 2023-01-21 12:59:16.049275: step: 1988/529, loss: 0.0039215087890625 2023-01-21 12:59:17.137928: step: 1992/529, loss: 1.029968279908644e-05 2023-01-21 12:59:18.270849: step: 1996/529, loss: 2.727508581301663e-05 2023-01-21 12:59:19.370700: step: 2000/529, loss: 0.016031980514526367 2023-01-21 12:59:20.489488: step: 2004/529, loss: 0.0006994247669354081 2023-01-21 12:59:21.587737: step: 2008/529, loss: 0.06915827095508575 2023-01-21 12:59:22.726510: step: 2012/529, loss: 0.0018802642589434981 2023-01-21 12:59:23.830901: step: 2016/529, loss: 0.013062858022749424 2023-01-21 12:59:24.939512: step: 2020/529, loss: 0.010177422314882278 2023-01-21 12:59:26.062374: step: 2024/529, loss: 0.0034377097617834806 2023-01-21 12:59:27.169330: step: 2028/529, loss: 0.00869131088256836 2023-01-21 12:59:28.293482: step: 2032/529, loss: 2.3651124138268642e-05 2023-01-21 12:59:29.394456: step: 2036/529, loss: 0.037790872156620026 2023-01-21 12:59:30.497990: step: 2040/529, loss: 0.029758168384432793 2023-01-21 12:59:31.594094: step: 2044/529, loss: 0.03365917131304741 2023-01-21 12:59:32.715688: step: 2048/529, loss: 0.013937568292021751 2023-01-21 12:59:33.821861: step: 2052/529, loss: -4.9591067181609105e-06 2023-01-21 12:59:34.906065: step: 2056/529, loss: 3.643035961431451e-05 2023-01-21 12:59:36.027695: step: 2060/529, loss: 0.01580982096493244 2023-01-21 12:59:37.138801: step: 2064/529, loss: 0.0012409688206389546 2023-01-21 12:59:38.267717: step: 2068/529, loss: 0.0008633614052087069 2023-01-21 12:59:39.370812: step: 2072/529, loss: 0.01094207726418972 2023-01-21 12:59:40.464514: step: 2076/529, loss: 0.015546416863799095 2023-01-21 12:59:41.592359: step: 2080/529, loss: 0.0002918243408203125 2023-01-21 12:59:42.730623: step: 2084/529, loss: 0.001398277236148715 2023-01-21 12:59:43.843379: step: 2088/529, loss: 0.001557350275106728 2023-01-21 12:59:44.949439: step: 2092/529, loss: 3.013610876223538e-05 2023-01-21 12:59:46.063879: step: 2096/529, loss: 0.012003613635897636 2023-01-21 12:59:47.178983: step: 2100/529, loss: 0.01590251922607422 2023-01-21 12:59:48.309392: step: 2104/529, loss: 0.001154136611148715 2023-01-21 12:59:49.449948: step: 2108/529, loss: 0.0012329102028161287 2023-01-21 12:59:50.582871: step: 2112/529, loss: 0.001546669052913785 2023-01-21 12:59:51.676936: step: 2116/529, loss: 0.05431947857141495 ================================================== Loss: 0.022 -------------------- Dev: {'event': {'p': 0.5945674044265593, 'r': 0.7869507323568575, 'f1': 0.6773638968481376}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Test: {'event': {'p': 0.6358950328022493, 'r': 0.7658013544018059, 'f1': 0.6948284690220173}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Chinese: {'event': {'p': 0.5882352941176471, 'r': 0.9259259259259259, 'f1': 0.7194244604316546}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Korean: {'event': {'p': 0.6, 'r': 0.5238095238095238, 'f1': 0.559322033898305}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Russian: {'event': {'p': 0.375, 'r': 0.5, 'f1': 0.42857142857142855}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 26 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:00:32.006148: step: 4/529, loss: 0.013025665655732155 2023-01-21 13:00:33.150354: step: 8/529, loss: 0.0027847292367368937 2023-01-21 13:00:34.250139: step: 12/529, loss: 9.34600848268019e-06 2023-01-21 13:00:35.326376: step: 16/529, loss: 0.0058952332474291325 2023-01-21 13:00:36.429697: step: 20/529, loss: 8.40187058201991e-05 2023-01-21 13:00:37.537197: step: 24/529, loss: 1.0013579867518274e-06 2023-01-21 13:00:38.648961: step: 28/529, loss: 0.00048770903958939016 2023-01-21 13:00:39.734210: step: 32/529, loss: 0.0002868652518372983 2023-01-21 13:00:40.834394: step: 36/529, loss: 0.011477470397949219 2023-01-21 13:00:41.962423: step: 40/529, loss: 0.03354759141802788 2023-01-21 13:00:43.076791: step: 44/529, loss: 0.0010120392544195056 2023-01-21 13:00:44.182100: step: 48/529, loss: 0.00029969215393066406 2023-01-21 13:00:45.275040: step: 52/529, loss: 6.303787813521922e-05 2023-01-21 13:00:46.401408: step: 56/529, loss: 5.588531348621473e-05 2023-01-21 13:00:47.503630: step: 60/529, loss: 0.05170316621661186 2023-01-21 13:00:48.628999: step: 64/529, loss: 0.021767234429717064 2023-01-21 13:00:49.724554: step: 68/529, loss: 0.03834161534905434 2023-01-21 13:00:50.843921: step: 72/529, loss: 0.00029392243595793843 2023-01-21 13:00:51.932141: step: 76/529, loss: 7.276535325217992e-05 2023-01-21 13:00:53.056457: step: 80/529, loss: 3.4332276754867053e-06 2023-01-21 13:00:54.170281: step: 84/529, loss: 0.0021758077200502157 2023-01-21 13:00:55.263462: step: 88/529, loss: 0.0002058029203908518 2023-01-21 13:00:56.360120: step: 92/529, loss: 8.487701052217744e-06 2023-01-21 13:00:57.477854: step: 96/529, loss: 0.0054690358228981495 2023-01-21 13:00:58.583610: step: 100/529, loss: 0.02786245383322239 2023-01-21 13:00:59.671878: step: 104/529, loss: 0.20832061767578125 2023-01-21 13:01:00.790803: step: 108/529, loss: 0.0001484871027059853 2023-01-21 13:01:01.903989: step: 112/529, loss: 0.002085351850837469 2023-01-21 13:01:03.009325: step: 116/529, loss: 0.0005060195690020919 2023-01-21 13:01:04.087772: step: 120/529, loss: -1.23977656585339e-06 2023-01-21 13:01:05.185855: step: 124/529, loss: 3.14712519866589e-06 2023-01-21 13:01:06.292537: step: 128/529, loss: 6.713867333019152e-05 2023-01-21 13:01:07.404080: step: 132/529, loss: 0.0018955230480059981 2023-01-21 13:01:08.576247: step: 136/529, loss: 0.0005881786346435547 2023-01-21 13:01:09.677184: step: 140/529, loss: 5.187988426769152e-05 2023-01-21 13:01:10.774651: step: 144/529, loss: 0.047556400299072266 2023-01-21 13:01:11.904238: step: 148/529, loss: 0.0008852005121298134 2023-01-21 13:01:12.993863: step: 152/529, loss: 1.5401839846163057e-05 2023-01-21 13:01:14.094194: step: 156/529, loss: 6.6784028604161e-05 2023-01-21 13:01:15.205125: step: 160/529, loss: 3.280639793956652e-05 2023-01-21 13:01:16.305885: step: 164/529, loss: 0.005192136857658625 2023-01-21 13:01:17.460720: step: 168/529, loss: 0.01980915106832981 2023-01-21 13:01:18.587895: step: 172/529, loss: 0.0140380859375 2023-01-21 13:01:19.667721: step: 176/529, loss: 0.00016603470430709422 2023-01-21 13:01:20.820723: step: 180/529, loss: 6.427764310501516e-05 2023-01-21 13:01:21.936480: step: 184/529, loss: 4.405975778354332e-05 2023-01-21 13:01:23.067722: step: 188/529, loss: 1.9550323031580774e-06 2023-01-21 13:01:24.172766: step: 192/529, loss: 2.784729076665826e-05 2023-01-21 13:01:25.288991: step: 196/529, loss: -6.48498553346144e-06 2023-01-21 13:01:26.463202: step: 200/529, loss: 0.008663367480039597 2023-01-21 13:01:27.587371: step: 204/529, loss: 0.010529518127441406 2023-01-21 13:01:28.672003: step: 208/529, loss: 0.00012340545072220266 2023-01-21 13:01:29.807253: step: 212/529, loss: 0.005165386013686657 2023-01-21 13:01:30.927233: step: 216/529, loss: 0.002931022783741355 2023-01-21 13:01:32.075523: step: 220/529, loss: 0.00018577575974632055 2023-01-21 13:01:33.175480: step: 224/529, loss: 0.04927084594964981 2023-01-21 13:01:34.276956: step: 228/529, loss: 1.6498566765221767e-05 2023-01-21 13:01:35.418394: step: 232/529, loss: 5.030632019042969e-05 2023-01-21 13:01:36.580220: step: 236/529, loss: 5.340576535672881e-06 2023-01-21 13:01:37.704821: step: 240/529, loss: 0.0010513782035559416 2023-01-21 13:01:38.808868: step: 244/529, loss: 0.0006593704456463456 2023-01-21 13:01:39.928570: step: 248/529, loss: 0.04218254238367081 2023-01-21 13:01:41.038045: step: 252/529, loss: 0.0010068893898278475 2023-01-21 13:01:42.143060: step: 256/529, loss: 0.0006353378412313759 2023-01-21 13:01:43.277812: step: 260/529, loss: 0.0013450622791424394 2023-01-21 13:01:44.371067: step: 264/529, loss: 0.00035037993802689016 2023-01-21 13:01:45.497441: step: 268/529, loss: 0.0051208497025072575 2023-01-21 13:01:46.599320: step: 272/529, loss: 0.03809080272912979 2023-01-21 13:01:47.707358: step: 276/529, loss: 0.05685706064105034 2023-01-21 13:01:48.805092: step: 280/529, loss: 0.009435271844267845 2023-01-21 13:01:49.945305: step: 284/529, loss: 0.1060028150677681 2023-01-21 13:01:51.063113: step: 288/529, loss: 0.06328850239515305 2023-01-21 13:01:52.157011: step: 292/529, loss: 0.008836936205625534 2023-01-21 13:01:53.301301: step: 296/529, loss: 4.1389463149243966e-05 2023-01-21 13:01:54.416228: step: 300/529, loss: 0.08662425726652145 2023-01-21 13:01:55.526177: step: 304/529, loss: 0.06862354278564453 2023-01-21 13:01:56.629531: step: 308/529, loss: 0.002481174422428012 2023-01-21 13:01:57.736430: step: 312/529, loss: 2.0694733393611386e-05 2023-01-21 13:01:58.850110: step: 316/529, loss: 0.00011291504779364914 2023-01-21 13:01:59.930038: step: 320/529, loss: 5.8650970458984375e-05 2023-01-21 13:02:01.043536: step: 324/529, loss: 0.00011100769916083664 2023-01-21 13:02:02.189651: step: 328/529, loss: 0.0003315925714559853 2023-01-21 13:02:03.318401: step: 332/529, loss: 0.0005913734785281122 2023-01-21 13:02:04.424679: step: 336/529, loss: 0.007082748226821423 2023-01-21 13:02:05.492175: step: 340/529, loss: -1.144409225162235e-06 2023-01-21 13:02:06.581267: step: 344/529, loss: 0.0002701759513001889 2023-01-21 13:02:07.677577: step: 348/529, loss: 7.05718994140625e-05 2023-01-21 13:02:08.768049: step: 352/529, loss: 1.296997106692288e-05 2023-01-21 13:02:09.884923: step: 356/529, loss: 1.926422191900201e-05 2023-01-21 13:02:11.010307: step: 360/529, loss: 0.00025048255338333547 2023-01-21 13:02:12.143133: step: 364/529, loss: 0.01147613488137722 2023-01-21 13:02:13.277656: step: 368/529, loss: 0.001139736152254045 2023-01-21 13:02:14.389914: step: 372/529, loss: 0.0008258819580078125 2023-01-21 13:02:15.507603: step: 376/529, loss: 0.0013508796691894531 2023-01-21 13:02:16.613950: step: 380/529, loss: 0.1910679191350937 2023-01-21 13:02:17.718575: step: 384/529, loss: 0.04526319354772568 2023-01-21 13:02:18.823580: step: 388/529, loss: 0.03982257843017578 2023-01-21 13:02:19.956735: step: 392/529, loss: 0.012942124158143997 2023-01-21 13:02:21.061096: step: 396/529, loss: 0.0037052154075354338 2023-01-21 13:02:22.157757: step: 400/529, loss: 0.00827803649008274 2023-01-21 13:02:23.261507: step: 404/529, loss: 8.125305612338707e-05 2023-01-21 13:02:24.362178: step: 408/529, loss: 0.0006135940784588456 2023-01-21 13:02:25.493689: step: 412/529, loss: 1.430511474609375e-05 2023-01-21 13:02:26.601056: step: 416/529, loss: 2.288818359375e-05 2023-01-21 13:02:27.728836: step: 420/529, loss: 0.000404548627557233 2023-01-21 13:02:28.848079: step: 424/529, loss: 0.0029531477484852076 2023-01-21 13:02:29.969638: step: 428/529, loss: 0.0005135536193847656 2023-01-21 13:02:31.083290: step: 432/529, loss: 0.00644607562571764 2023-01-21 13:02:32.221977: step: 436/529, loss: 0.0004478454648051411 2023-01-21 13:02:33.310528: step: 440/529, loss: 7.05719003235572e-06 2023-01-21 13:02:34.405046: step: 444/529, loss: 1.869201696536038e-05 2023-01-21 13:02:35.508300: step: 448/529, loss: 0.004633331671357155 2023-01-21 13:02:36.629369: step: 452/529, loss: 4.00543194700731e-06 2023-01-21 13:02:37.728176: step: 456/529, loss: 0.0007177353254519403 2023-01-21 13:02:38.825456: step: 460/529, loss: 0.1050867810845375 2023-01-21 13:02:39.935166: step: 464/529, loss: 0.00042257309542037547 2023-01-21 13:02:41.065514: step: 468/529, loss: 4.6443939936580136e-05 2023-01-21 13:02:42.162814: step: 472/529, loss: 0.00022220611572265625 2023-01-21 13:02:43.297843: step: 476/529, loss: 0.0016588212456554174 2023-01-21 13:02:44.406856: step: 480/529, loss: 0.006584167946130037 2023-01-21 13:02:45.515326: step: 484/529, loss: 0.00019583702669478953 2023-01-21 13:02:46.655284: step: 488/529, loss: 0.00016369819059036672 2023-01-21 13:02:47.772028: step: 492/529, loss: 0.009474468417465687 2023-01-21 13:02:48.853061: step: 496/529, loss: 2.6226043701171875e-06 2023-01-21 13:02:49.962881: step: 500/529, loss: 0.03279514238238335 2023-01-21 13:02:51.071868: step: 504/529, loss: 0.013543128967285156 2023-01-21 13:02:52.184794: step: 508/529, loss: 0.005602359771728516 2023-01-21 13:02:53.307678: step: 512/529, loss: -6.48498553346144e-06 2023-01-21 13:02:54.414647: step: 516/529, loss: 0.04302806779742241 2023-01-21 13:02:55.506467: step: 520/529, loss: 1.7547608877066523e-05 2023-01-21 13:02:56.622162: step: 524/529, loss: 2.403259350103326e-05 2023-01-21 13:02:57.754975: step: 528/529, loss: 0.0003664970281533897 2023-01-21 13:02:58.884188: step: 532/529, loss: 0.00047162771807052195 2023-01-21 13:03:00.033213: step: 536/529, loss: 4.501343209994957e-05 2023-01-21 13:03:01.162666: step: 540/529, loss: 0.019955063238739967 2023-01-21 13:03:02.278986: step: 544/529, loss: 9.751320612849668e-05 2023-01-21 13:03:03.387923: step: 548/529, loss: 3.852844383800402e-05 2023-01-21 13:03:04.503831: step: 552/529, loss: 0.0004089355352334678 2023-01-21 13:03:05.610486: step: 556/529, loss: 0.014489288441836834 2023-01-21 13:03:06.729628: step: 560/529, loss: 4.482269105210435e-06 2023-01-21 13:03:07.812988: step: 564/529, loss: 0.013504648581147194 2023-01-21 13:03:08.915161: step: 568/529, loss: 0.06941261142492294 2023-01-21 13:03:10.028385: step: 572/529, loss: 0.0018215716117992997 2023-01-21 13:03:11.146942: step: 576/529, loss: 0.02603950724005699 2023-01-21 13:03:12.273641: step: 580/529, loss: 0.009235001169145107 2023-01-21 13:03:13.381438: step: 584/529, loss: 1.716613724056515e-06 2023-01-21 13:03:14.516113: step: 588/529, loss: 0.008207893930375576 2023-01-21 13:03:15.641017: step: 592/529, loss: 3.2336976528167725 2023-01-21 13:03:16.785361: step: 596/529, loss: 0.022122908383607864 2023-01-21 13:03:17.902479: step: 600/529, loss: 5.245208740234375e-05 2023-01-21 13:03:19.038865: step: 604/529, loss: 0.00014190674119163305 2023-01-21 13:03:20.170002: step: 608/529, loss: 0.00774726876989007 2023-01-21 13:03:21.283025: step: 612/529, loss: 0.00158348074182868 2023-01-21 13:03:22.391559: step: 616/529, loss: 0.01694359816610813 2023-01-21 13:03:23.515376: step: 620/529, loss: 0.0017402649391442537 2023-01-21 13:03:24.610695: step: 624/529, loss: 0.04059801250696182 2023-01-21 13:03:25.719032: step: 628/529, loss: 0.03342742845416069 2023-01-21 13:03:26.814851: step: 632/529, loss: 0.0038545611314475536 2023-01-21 13:03:27.895979: step: 636/529, loss: 0.0020542622078210115 2023-01-21 13:03:28.997197: step: 640/529, loss: 0.0344633124768734 2023-01-21 13:03:30.109473: step: 644/529, loss: 0.0003381729475222528 2023-01-21 13:03:31.235352: step: 648/529, loss: 0.0012248993152752519 2023-01-21 13:03:32.351504: step: 652/529, loss: 0.0007341623422689736 2023-01-21 13:03:33.442301: step: 656/529, loss: 0.017383193597197533 2023-01-21 13:03:34.585995: step: 660/529, loss: 0.009919739328324795 2023-01-21 13:03:35.682833: step: 664/529, loss: 0.0011301517952233553 2023-01-21 13:03:36.800697: step: 668/529, loss: 0.029977895319461823 2023-01-21 13:03:37.887864: step: 672/529, loss: 0.0017427444690838456 2023-01-21 13:03:38.989406: step: 676/529, loss: 0.021220874041318893 2023-01-21 13:03:40.075994: step: 680/529, loss: 0.016243362799286842 2023-01-21 13:03:41.208688: step: 684/529, loss: 9.384155418956652e-05 2023-01-21 13:03:42.302869: step: 688/529, loss: 0.00158605573233217 2023-01-21 13:03:43.405988: step: 692/529, loss: 0.1293531358242035 2023-01-21 13:03:44.506649: step: 696/529, loss: 0.006448936183005571 2023-01-21 13:03:45.622725: step: 700/529, loss: 0.0023102760314941406 2023-01-21 13:03:46.717677: step: 704/529, loss: 7.104873020580271e-06 2023-01-21 13:03:47.837033: step: 708/529, loss: 0.12760746479034424 2023-01-21 13:03:48.932983: step: 712/529, loss: 0.0018495559925213456 2023-01-21 13:03:50.013980: step: 716/529, loss: 0.00471839914098382 2023-01-21 13:03:51.113837: step: 720/529, loss: 3.337860107421875e-06 2023-01-21 13:03:52.219129: step: 724/529, loss: 0.04250144958496094 2023-01-21 13:03:53.325362: step: 728/529, loss: 0.053050994873046875 2023-01-21 13:03:54.448375: step: 732/529, loss: 0.0004034042649436742 2023-01-21 13:03:55.552477: step: 736/529, loss: 0.0002035141078522429 2023-01-21 13:03:56.683370: step: 740/529, loss: 0.09148712456226349 2023-01-21 13:03:57.827010: step: 744/529, loss: 0.014636993408203125 2023-01-21 13:03:58.942804: step: 748/529, loss: 0.0017119408585131168 2023-01-21 13:04:00.034380: step: 752/529, loss: 4.310607982915826e-05 2023-01-21 13:04:01.163235: step: 756/529, loss: 0.0004661559942178428 2023-01-21 13:04:02.267212: step: 760/529, loss: 0.006013727281242609 2023-01-21 13:04:03.387264: step: 764/529, loss: 0.00046925543574616313 2023-01-21 13:04:04.497312: step: 768/529, loss: 0.04130706936120987 2023-01-21 13:04:05.644443: step: 772/529, loss: 0.007816696539521217 2023-01-21 13:04:06.772748: step: 776/529, loss: 0.012048817239701748 2023-01-21 13:04:07.867270: step: 780/529, loss: 0.0006542205810546875 2023-01-21 13:04:08.979031: step: 784/529, loss: 0.009632492437958717 2023-01-21 13:04:10.079513: step: 788/529, loss: -8.583067483414197e-07 2023-01-21 13:04:11.207162: step: 792/529, loss: 0.0007371902465820312 2023-01-21 13:04:12.325939: step: 796/529, loss: 0.00015506744966842234 2023-01-21 13:04:13.427568: step: 800/529, loss: 0.00017261505126953125 2023-01-21 13:04:14.521703: step: 804/529, loss: -1.2040136425639503e-06 2023-01-21 13:04:15.633619: step: 808/529, loss: 0.0007036208990029991 2023-01-21 13:04:16.747272: step: 812/529, loss: 0.021425379440188408 2023-01-21 13:04:17.854949: step: 816/529, loss: 0.25662097334861755 2023-01-21 13:04:18.978924: step: 820/529, loss: 0.0047245025634765625 2023-01-21 13:04:20.131371: step: 824/529, loss: 0.0062185293063521385 2023-01-21 13:04:21.263624: step: 828/529, loss: 0.0017829417483881116 2023-01-21 13:04:22.374887: step: 832/529, loss: 4.806518700206652e-05 2023-01-21 13:04:23.502366: step: 836/529, loss: 0.0004596710205078125 2023-01-21 13:04:24.620738: step: 840/529, loss: 2.670288040462765e-06 2023-01-21 13:04:25.753014: step: 844/529, loss: 3.623962356869015e-06 2023-01-21 13:04:26.872992: step: 848/529, loss: 0.04018077999353409 2023-01-21 13:04:27.983477: step: 852/529, loss: 0.22704820334911346 2023-01-21 13:04:29.115910: step: 856/529, loss: 0.0003452301025390625 2023-01-21 13:04:30.220312: step: 860/529, loss: 0.01042270753532648 2023-01-21 13:04:31.331080: step: 864/529, loss: 0.5174850225448608 2023-01-21 13:04:32.440728: step: 868/529, loss: 0.5592904090881348 2023-01-21 13:04:33.548199: step: 872/529, loss: 0.010742664337158203 2023-01-21 13:04:34.651963: step: 876/529, loss: 0.0008894919883459806 2023-01-21 13:04:35.755535: step: 880/529, loss: 0.005408096127212048 2023-01-21 13:04:36.854153: step: 884/529, loss: 0.0006146430969238281 2023-01-21 13:04:37.959361: step: 888/529, loss: 0.04459371790289879 2023-01-21 13:04:39.050711: step: 892/529, loss: 0.0012407302856445312 2023-01-21 13:04:40.154175: step: 896/529, loss: 0.0011411666637286544 2023-01-21 13:04:41.273920: step: 900/529, loss: 0.0003591299173422158 2023-01-21 13:04:42.375717: step: 904/529, loss: 0.0058914185501635075 2023-01-21 13:04:43.471721: step: 908/529, loss: 1.0585785275907256e-05 2023-01-21 13:04:44.592613: step: 912/529, loss: 0.0003442287561483681 2023-01-21 13:04:45.711564: step: 916/529, loss: 0.026396943256258965 2023-01-21 13:04:46.860231: step: 920/529, loss: 0.00126991281285882 2023-01-21 13:04:47.983112: step: 924/529, loss: 0.009387397207319736 2023-01-21 13:04:49.061072: step: 928/529, loss: 0.007410907652229071 2023-01-21 13:04:50.184021: step: 932/529, loss: 0.03458261862397194 2023-01-21 13:04:51.255753: step: 936/529, loss: 0.00617795018479228 2023-01-21 13:04:52.404778: step: 940/529, loss: 0.0156415943056345 2023-01-21 13:04:53.518208: step: 944/529, loss: 0.02698078379034996 2023-01-21 13:04:54.613724: step: 948/529, loss: 0.017787646502256393 2023-01-21 13:04:55.723534: step: 952/529, loss: 0.06816940009593964 2023-01-21 13:04:56.834798: step: 956/529, loss: 0.0009677887428551912 2023-01-21 13:04:57.941204: step: 960/529, loss: 0.006277275271713734 2023-01-21 13:04:59.069443: step: 964/529, loss: 6.160735938465223e-05 2023-01-21 13:05:00.177285: step: 968/529, loss: 0.04080080986022949 2023-01-21 13:05:01.288874: step: 972/529, loss: 0.010431194677948952 2023-01-21 13:05:02.394027: step: 976/529, loss: 0.01596083678305149 2023-01-21 13:05:03.523374: step: 980/529, loss: 0.0007116794586181641 2023-01-21 13:05:04.652459: step: 984/529, loss: 0.001605987548828125 2023-01-21 13:05:05.792417: step: 988/529, loss: 3.337860107421875e-05 2023-01-21 13:05:06.971152: step: 992/529, loss: 0.002616691403090954 2023-01-21 13:05:08.109088: step: 996/529, loss: 0.007730865851044655 2023-01-21 13:05:09.225365: step: 1000/529, loss: 0.0001491546572651714 2023-01-21 13:05:10.349429: step: 1004/529, loss: 0.0003711700264830142 2023-01-21 13:05:11.463626: step: 1008/529, loss: 0.0011664391495287418 2023-01-21 13:05:12.598987: step: 1012/529, loss: -1.583099401614163e-05 2023-01-21 13:05:13.711462: step: 1016/529, loss: 0.007588386535644531 2023-01-21 13:05:14.828874: step: 1020/529, loss: 0.0003841400321107358 2023-01-21 13:05:15.965926: step: 1024/529, loss: 0.0013720989227294922 2023-01-21 13:05:17.077119: step: 1028/529, loss: 0.002115202136337757 2023-01-21 13:05:18.203147: step: 1032/529, loss: 0.012281798757612705 2023-01-21 13:05:19.310459: step: 1036/529, loss: 0.011130142025649548 2023-01-21 13:05:20.419166: step: 1040/529, loss: 6.613731966353953e-05 2023-01-21 13:05:21.556544: step: 1044/529, loss: 0.00023927688016556203 2023-01-21 13:05:22.645687: step: 1048/529, loss: 0.0009638786432333291 2023-01-21 13:05:23.762050: step: 1052/529, loss: 0.02169027365744114 2023-01-21 13:05:24.843414: step: 1056/529, loss: 0.0015693664317950606 2023-01-21 13:05:25.929333: step: 1060/529, loss: 0.0024577141739428043 2023-01-21 13:05:27.045542: step: 1064/529, loss: 0.00015354156494140625 2023-01-21 13:05:28.137235: step: 1068/529, loss: 0.0005867004510946572 2023-01-21 13:05:29.237969: step: 1072/529, loss: 0.005464458838105202 2023-01-21 13:05:30.375250: step: 1076/529, loss: 0.002259588334709406 2023-01-21 13:05:31.497041: step: 1080/529, loss: 0.00015416146197821945 2023-01-21 13:05:32.636919: step: 1084/529, loss: 0.08061598986387253 2023-01-21 13:05:33.713919: step: 1088/529, loss: 0.009635353460907936 2023-01-21 13:05:34.885461: step: 1092/529, loss: 0.03504829481244087 2023-01-21 13:05:36.047060: step: 1096/529, loss: 0.009019183926284313 2023-01-21 13:05:37.185037: step: 1100/529, loss: 0.03563690558075905 2023-01-21 13:05:38.290583: step: 1104/529, loss: 0.001275444170460105 2023-01-21 13:05:39.416212: step: 1108/529, loss: 0.0002767562691587955 2023-01-21 13:05:40.547137: step: 1112/529, loss: 0.00045614244299940765 2023-01-21 13:05:41.656227: step: 1116/529, loss: 0.011727713979780674 2023-01-21 13:05:42.781121: step: 1120/529, loss: 0.0015352844493463635 2023-01-21 13:05:43.908005: step: 1124/529, loss: 0.007339191623032093 2023-01-21 13:05:45.038954: step: 1128/529, loss: 0.00920496042817831 2023-01-21 13:05:46.179847: step: 1132/529, loss: 0.011356162838637829 2023-01-21 13:05:47.287919: step: 1136/529, loss: 4.358291334938258e-05 2023-01-21 13:05:48.368853: step: 1140/529, loss: 0.00033245087251998484 2023-01-21 13:05:49.535912: step: 1144/529, loss: 9.155274165095761e-05 2023-01-21 13:05:50.667193: step: 1148/529, loss: 0.0015883445739746094 2023-01-21 13:05:51.804662: step: 1152/529, loss: 0.0006401062128134072 2023-01-21 13:05:52.915101: step: 1156/529, loss: 0.0001905441313283518 2023-01-21 13:05:54.035262: step: 1160/529, loss: 0.0004949569702148438 2023-01-21 13:05:55.151608: step: 1164/529, loss: 0.0011981964344158769 2023-01-21 13:05:56.261846: step: 1168/529, loss: 7.114410254871473e-05 2023-01-21 13:05:57.387718: step: 1172/529, loss: 0.014460334554314613 2023-01-21 13:05:58.486611: step: 1176/529, loss: 2.574920654296875e-05 2023-01-21 13:05:59.584720: step: 1180/529, loss: 0.0006227493286132812 2023-01-21 13:06:00.700888: step: 1184/529, loss: 0.564540684223175 2023-01-21 13:06:01.789762: step: 1188/529, loss: 8.029938180698082e-05 2023-01-21 13:06:02.906489: step: 1192/529, loss: 0.022452354431152344 2023-01-21 13:06:04.042016: step: 1196/529, loss: 0.004067373927682638 2023-01-21 13:06:05.175288: step: 1200/529, loss: 0.0031614305917173624 2023-01-21 13:06:06.286607: step: 1204/529, loss: 5.254745701677166e-05 2023-01-21 13:06:07.389801: step: 1208/529, loss: 0.00013580323138739914 2023-01-21 13:06:08.517726: step: 1212/529, loss: 0.005173874087631702 2023-01-21 13:06:09.615539: step: 1216/529, loss: 0.000751113926526159 2023-01-21 13:06:10.749668: step: 1220/529, loss: 0.016982652246952057 2023-01-21 13:06:11.850598: step: 1224/529, loss: 0.0006086349603720009 2023-01-21 13:06:12.962189: step: 1228/529, loss: 0.0014345645904541016 2023-01-21 13:06:14.062474: step: 1232/529, loss: 6.86645489622606e-06 2023-01-21 13:06:15.159363: step: 1236/529, loss: 2.09808349609375e-05 2023-01-21 13:06:16.256812: step: 1240/529, loss: 0.00014209747314453125 2023-01-21 13:06:17.371180: step: 1244/529, loss: 0.00017266272334381938 2023-01-21 13:06:18.438807: step: 1248/529, loss: 0.5793344378471375 2023-01-21 13:06:19.567527: step: 1252/529, loss: 0.0024230005219578743 2023-01-21 13:06:20.739196: step: 1256/529, loss: 0.007236480712890625 2023-01-21 13:06:21.844905: step: 1260/529, loss: 0.0005718230968341231 2023-01-21 13:06:22.965793: step: 1264/529, loss: 0.00046710966853424907 2023-01-21 13:06:24.095389: step: 1268/529, loss: 0.01821594312787056 2023-01-21 13:06:25.195842: step: 1272/529, loss: 0.04938144609332085 2023-01-21 13:06:26.314515: step: 1276/529, loss: 0.0003917694266419858 2023-01-21 13:06:27.435243: step: 1280/529, loss: 0.00631141709163785 2023-01-21 13:06:28.549368: step: 1284/529, loss: 0.06543522328138351 2023-01-21 13:06:29.660870: step: 1288/529, loss: 0.5150752663612366 2023-01-21 13:06:30.799391: step: 1292/529, loss: 0.00045375822810456157 2023-01-21 13:06:31.940132: step: 1296/529, loss: 0.000759857939556241 2023-01-21 13:06:33.064521: step: 1300/529, loss: 0.005687904544174671 2023-01-21 13:06:34.167473: step: 1304/529, loss: 0.0656890869140625 2023-01-21 13:06:35.251357: step: 1308/529, loss: 0.0001586914004292339 2023-01-21 13:06:36.365149: step: 1312/529, loss: 3.4904482163256034e-05 2023-01-21 13:06:37.499263: step: 1316/529, loss: 0.000712966953869909 2023-01-21 13:06:38.611222: step: 1320/529, loss: 0.000575637852307409 2023-01-21 13:06:39.726641: step: 1324/529, loss: 0.013082695193588734 2023-01-21 13:06:40.857801: step: 1328/529, loss: 0.0021923065651208162 2023-01-21 13:06:41.966952: step: 1332/529, loss: 0.003923511598259211 2023-01-21 13:06:43.131025: step: 1336/529, loss: 0.0002298072213307023 2023-01-21 13:06:44.257154: step: 1340/529, loss: -4.95910626341356e-06 2023-01-21 13:06:45.370283: step: 1344/529, loss: -1.2302398317842744e-05 2023-01-21 13:06:46.498924: step: 1348/529, loss: 0.009788895025849342 2023-01-21 13:06:47.602737: step: 1352/529, loss: 0.0022041320335119963 2023-01-21 13:06:48.728132: step: 1356/529, loss: 0.0001466751127736643 2023-01-21 13:06:49.839889: step: 1360/529, loss: 0.004870033357292414 2023-01-21 13:06:50.964989: step: 1364/529, loss: 0.001064300537109375 2023-01-21 13:06:52.068662: step: 1368/529, loss: 9.651183790992945e-05 2023-01-21 13:06:53.207416: step: 1372/529, loss: 0.00019216537475585938 2023-01-21 13:06:54.306584: step: 1376/529, loss: 0.0033470706548541784 2023-01-21 13:06:55.445170: step: 1380/529, loss: 5.5599211918888614e-05 2023-01-21 13:06:56.587571: step: 1384/529, loss: 2.269744800287299e-05 2023-01-21 13:06:57.673057: step: 1388/529, loss: 9.822845458984375e-05 2023-01-21 13:06:58.849490: step: 1392/529, loss: 8.39233416627394e-06 2023-01-21 13:06:59.959692: step: 1396/529, loss: 6.771087646484375e-05 2023-01-21 13:07:01.121363: step: 1400/529, loss: 3.2806394301587716e-05 2023-01-21 13:07:02.200034: step: 1404/529, loss: 0.0352204330265522 2023-01-21 13:07:03.323309: step: 1408/529, loss: 0.17665652930736542 2023-01-21 13:07:04.446841: step: 1412/529, loss: 0.10771813243627548 2023-01-21 13:07:05.589448: step: 1416/529, loss: 7.247925168485381e-06 2023-01-21 13:07:06.743280: step: 1420/529, loss: 0.029523469507694244 2023-01-21 13:07:07.878484: step: 1424/529, loss: 0.0025140761863440275 2023-01-21 13:07:08.972752: step: 1428/529, loss: 7.61985793360509e-05 2023-01-21 13:07:10.089718: step: 1432/529, loss: 5.9700014389818534e-05 2023-01-21 13:07:11.203964: step: 1436/529, loss: 0.003639650298282504 2023-01-21 13:07:12.307273: step: 1440/529, loss: 0.0025553943123668432 2023-01-21 13:07:13.430116: step: 1444/529, loss: 5.054473876953125e-05 2023-01-21 13:07:14.549578: step: 1448/529, loss: 0.002869558287784457 2023-01-21 13:07:15.649446: step: 1452/529, loss: 0.00020494461932685226 2023-01-21 13:07:16.769336: step: 1456/529, loss: 0.0295270923525095 2023-01-21 13:07:17.850146: step: 1460/529, loss: 0.000720977783203125 2023-01-21 13:07:18.971682: step: 1464/529, loss: 0.0030329704750329256 2023-01-21 13:07:20.119912: step: 1468/529, loss: 0.0011715888977050781 2023-01-21 13:07:21.225986: step: 1472/529, loss: 0.03882484510540962 2023-01-21 13:07:22.333548: step: 1476/529, loss: 0.0004371643008198589 2023-01-21 13:07:23.437047: step: 1480/529, loss: 0.06216182932257652 2023-01-21 13:07:24.552236: step: 1484/529, loss: 2.670288040462765e-06 2023-01-21 13:07:25.678067: step: 1488/529, loss: 0.004218196962028742 2023-01-21 13:07:26.773543: step: 1492/529, loss: 1.9073486612342094e-07 2023-01-21 13:07:27.865794: step: 1496/529, loss: 0.017743874341249466 2023-01-21 13:07:28.981358: step: 1500/529, loss: 1.907349087559851e-06 2023-01-21 13:07:30.093073: step: 1504/529, loss: 0.0002779960632324219 2023-01-21 13:07:31.188736: step: 1508/529, loss: 4.5299530029296875e-06 2023-01-21 13:07:32.316994: step: 1512/529, loss: 0.0018747331341728568 2023-01-21 13:07:33.436043: step: 1516/529, loss: 0.0005968093755654991 2023-01-21 13:07:34.551869: step: 1520/529, loss: 0.0003841877041850239 2023-01-21 13:07:35.664324: step: 1524/529, loss: 0.06617489457130432 2023-01-21 13:07:36.822242: step: 1528/529, loss: 0.0034699919633567333 2023-01-21 13:07:37.935864: step: 1532/529, loss: 0.1304466277360916 2023-01-21 13:07:39.067146: step: 1536/529, loss: 5.316734313964844e-05 2023-01-21 13:07:40.185169: step: 1540/529, loss: 0.01122360210865736 2023-01-21 13:07:41.303253: step: 1544/529, loss: 0.08428402245044708 2023-01-21 13:07:42.416631: step: 1548/529, loss: 0.00036230089608579874 2023-01-21 13:07:43.545309: step: 1552/529, loss: 0.023296929895877838 2023-01-21 13:07:44.656276: step: 1556/529, loss: 4.482269287109375e-05 2023-01-21 13:07:45.768866: step: 1560/529, loss: 0.0013785362243652344 2023-01-21 13:07:46.871751: step: 1564/529, loss: 0.00013961792865302414 2023-01-21 13:07:47.975583: step: 1568/529, loss: 0.0018196107121184468 2023-01-21 13:07:49.109241: step: 1572/529, loss: 0.0009608268737792969 2023-01-21 13:07:50.234672: step: 1576/529, loss: 0.0014187813503667712 2023-01-21 13:07:51.333319: step: 1580/529, loss: 0.0006870270008221269 2023-01-21 13:07:52.416803: step: 1584/529, loss: 0.028442764654755592 2023-01-21 13:07:53.582525: step: 1588/529, loss: 0.0001369476376567036 2023-01-21 13:07:54.695181: step: 1592/529, loss: 0.45006120204925537 2023-01-21 13:07:55.803279: step: 1596/529, loss: 0.0002739906485658139 2023-01-21 13:07:56.910497: step: 1600/529, loss: 0.2774600088596344 2023-01-21 13:07:58.003523: step: 1604/529, loss: 8.77380352903856e-06 2023-01-21 13:07:59.112782: step: 1608/529, loss: 0.0007786750793457031 2023-01-21 13:08:00.260315: step: 1612/529, loss: 0.0005665779463015497 2023-01-21 13:08:01.375921: step: 1616/529, loss: 0.12063168734312057 2023-01-21 13:08:02.543964: step: 1620/529, loss: -2.555847095209174e-05 2023-01-21 13:08:03.639815: step: 1624/529, loss: 0.0011086941231042147 2023-01-21 13:08:04.767988: step: 1628/529, loss: 0.0011747360695153475 2023-01-21 13:08:05.917843: step: 1632/529, loss: 1.4448165529756807e-05 2023-01-21 13:08:07.039497: step: 1636/529, loss: 0.028077125549316406 2023-01-21 13:08:08.166786: step: 1640/529, loss: 1.4352797734318301e-05 2023-01-21 13:08:09.263645: step: 1644/529, loss: 0.00059423444326967 2023-01-21 13:08:10.395671: step: 1648/529, loss: 0.17636986076831818 2023-01-21 13:08:11.522063: step: 1652/529, loss: 0.0003440857108216733 2023-01-21 13:08:12.647272: step: 1656/529, loss: 0.0004421234189067036 2023-01-21 13:08:13.744893: step: 1660/529, loss: 1.2683868590102065e-05 2023-01-21 13:08:14.855210: step: 1664/529, loss: 0.016375159844756126 2023-01-21 13:08:15.965757: step: 1668/529, loss: 5.178451829124242e-05 2023-01-21 13:08:17.093699: step: 1672/529, loss: 0.006978416349738836 2023-01-21 13:08:18.200071: step: 1676/529, loss: 0.0014961243141442537 2023-01-21 13:08:19.315670: step: 1680/529, loss: 0.012310790829360485 2023-01-21 13:08:20.417199: step: 1684/529, loss: 0.00024929046048782766 2023-01-21 13:08:21.526553: step: 1688/529, loss: 6.790161569369957e-05 2023-01-21 13:08:22.650714: step: 1692/529, loss: 0.03449878841638565 2023-01-21 13:08:23.733885: step: 1696/529, loss: 0.0007984161493368447 2023-01-21 13:08:24.829704: step: 1700/529, loss: 0.00013570785813499242 2023-01-21 13:08:25.980946: step: 1704/529, loss: 4.57763690064894e-06 2023-01-21 13:08:27.100164: step: 1708/529, loss: 0.0020987512543797493 2023-01-21 13:08:28.234181: step: 1712/529, loss: 0.0171248447149992 2023-01-21 13:08:29.359847: step: 1716/529, loss: 6.675720101156912e-07 2023-01-21 13:08:30.478271: step: 1720/529, loss: 3.266334533691406e-05 2023-01-21 13:08:31.625746: step: 1724/529, loss: 0.0007851123809814453 2023-01-21 13:08:32.723443: step: 1728/529, loss: 2.8157235647086054e-05 2023-01-21 13:08:33.842525: step: 1732/529, loss: 0.0003845214669127017 2023-01-21 13:08:34.957425: step: 1736/529, loss: 0.0008475304348394275 2023-01-21 13:08:36.094105: step: 1740/529, loss: 0.007184791844338179 2023-01-21 13:08:37.191820: step: 1744/529, loss: 0.007298660464584827 2023-01-21 13:08:38.334990: step: 1748/529, loss: 0.0032461166847497225 2023-01-21 13:08:39.461280: step: 1752/529, loss: 0.010535622015595436 2023-01-21 13:08:40.563720: step: 1756/529, loss: 0.06780029088258743 2023-01-21 13:08:41.662750: step: 1760/529, loss: 0.0024008750915527344 2023-01-21 13:08:42.750202: step: 1764/529, loss: 0.050618939101696014 2023-01-21 13:08:43.871494: step: 1768/529, loss: 0.0025347708724439144 2023-01-21 13:08:44.969853: step: 1772/529, loss: 0.06196479871869087 2023-01-21 13:08:46.086380: step: 1776/529, loss: 0.00031375885009765625 2023-01-21 13:08:47.248616: step: 1780/529, loss: 0.024052632972598076 2023-01-21 13:08:48.373496: step: 1784/529, loss: 1.0395049685030244e-05 2023-01-21 13:08:49.505226: step: 1788/529, loss: 0.025150012224912643 2023-01-21 13:08:50.606938: step: 1792/529, loss: 0.01808338239789009 2023-01-21 13:08:51.717605: step: 1796/529, loss: 0.009450817480683327 2023-01-21 13:08:52.827168: step: 1800/529, loss: 0.001556396484375 2023-01-21 13:08:53.918935: step: 1804/529, loss: 0.00036430361797101796 2023-01-21 13:08:54.999303: step: 1808/529, loss: 0.002257442567497492 2023-01-21 13:08:56.122304: step: 1812/529, loss: 0.00048284532385878265 2023-01-21 13:08:57.233627: step: 1816/529, loss: 0.26395654678344727 2023-01-21 13:08:58.322480: step: 1820/529, loss: 0.0002574920654296875 2023-01-21 13:08:59.455533: step: 1824/529, loss: 0.00429191580042243 2023-01-21 13:09:00.588557: step: 1828/529, loss: 7.514953904319555e-05 2023-01-21 13:09:01.688615: step: 1832/529, loss: 7.171631295932457e-05 2023-01-21 13:09:02.818988: step: 1836/529, loss: 0.0008050918695516884 2023-01-21 13:09:03.947460: step: 1840/529, loss: 0.028122615069150925 2023-01-21 13:09:05.050809: step: 1844/529, loss: 0.00010633468627929688 2023-01-21 13:09:06.161580: step: 1848/529, loss: 0.015671061351895332 2023-01-21 13:09:07.289699: step: 1852/529, loss: 0.006744480226188898 2023-01-21 13:09:08.394865: step: 1856/529, loss: 0.0002468586026225239 2023-01-21 13:09:09.504941: step: 1860/529, loss: 0.0068874359130859375 2023-01-21 13:09:10.613746: step: 1864/529, loss: 6.370544724632055e-05 2023-01-21 13:09:11.724817: step: 1868/529, loss: 0.0001638412504689768 2023-01-21 13:09:12.825828: step: 1872/529, loss: 0.015439462848007679 2023-01-21 13:09:13.962956: step: 1876/529, loss: 7.457733590854332e-05 2023-01-21 13:09:15.087341: step: 1880/529, loss: 2.536773718020413e-05 2023-01-21 13:09:16.167487: step: 1884/529, loss: 0.04463768005371094 2023-01-21 13:09:17.263542: step: 1888/529, loss: -1.23977656585339e-06 2023-01-21 13:09:18.352709: step: 1892/529, loss: 0.004299736116081476 2023-01-21 13:09:19.449262: step: 1896/529, loss: 0.037055160850286484 2023-01-21 13:09:20.564337: step: 1900/529, loss: 0.0802450180053711 2023-01-21 13:09:21.718582: step: 1904/529, loss: 1.049041748046875e-05 2023-01-21 13:09:22.854448: step: 1908/529, loss: 0.008633041754364967 2023-01-21 13:09:23.960903: step: 1912/529, loss: 0.0007404327625408769 2023-01-21 13:09:25.072268: step: 1916/529, loss: 0.0005960464477539062 2023-01-21 13:09:26.192829: step: 1920/529, loss: 2.098083541568485e-06 2023-01-21 13:09:27.298069: step: 1924/529, loss: 0.053429603576660156 2023-01-21 13:09:28.424891: step: 1928/529, loss: 0.001379299210384488 2023-01-21 13:09:29.586219: step: 1932/529, loss: 0.0007281303405761719 2023-01-21 13:09:30.700407: step: 1936/529, loss: -7.724762326688506e-06 2023-01-21 13:09:31.818830: step: 1940/529, loss: 0.0005960464477539062 2023-01-21 13:09:32.924602: step: 1944/529, loss: 0.001238059950992465 2023-01-21 13:09:34.042137: step: 1948/529, loss: 0.028688622638583183 2023-01-21 13:09:35.166048: step: 1952/529, loss: 0.00026226043701171875 2023-01-21 13:09:36.284917: step: 1956/529, loss: 2.975463939947076e-05 2023-01-21 13:09:37.416676: step: 1960/529, loss: 0.036904145032167435 2023-01-21 13:09:38.533245: step: 1964/529, loss: 0.014790916815400124 2023-01-21 13:09:39.638129: step: 1968/529, loss: 0.001980686094611883 2023-01-21 13:09:40.756863: step: 1972/529, loss: 0.0017474174965173006 2023-01-21 13:09:41.894628: step: 1976/529, loss: 0.00021162032498978078 2023-01-21 13:09:43.004299: step: 1980/529, loss: 0.003277301788330078 2023-01-21 13:09:44.123015: step: 1984/529, loss: 0.018806075677275658 2023-01-21 13:09:45.245796: step: 1988/529, loss: 0.010902022942900658 2023-01-21 13:09:46.329348: step: 1992/529, loss: 0.006938457489013672 2023-01-21 13:09:47.424491: step: 1996/529, loss: 0.004268646705895662 2023-01-21 13:09:48.533553: step: 2000/529, loss: 6.561279587913305e-05 2023-01-21 13:09:49.630459: step: 2004/529, loss: 0.01737232133746147 2023-01-21 13:09:50.775035: step: 2008/529, loss: 0.01622590981423855 2023-01-21 13:09:51.932901: step: 2012/529, loss: 0.003128433134406805 2023-01-21 13:09:53.031010: step: 2016/529, loss: 0.01795358583331108 2023-01-21 13:09:54.145507: step: 2020/529, loss: 0.0007976532215252519 2023-01-21 13:09:55.243946: step: 2024/529, loss: 0.00010929107520496473 2023-01-21 13:09:56.364266: step: 2028/529, loss: 0.0023077011574059725 2023-01-21 13:09:57.494737: step: 2032/529, loss: 0.0004248619079589844 2023-01-21 13:09:58.588912: step: 2036/529, loss: 1.487731969973538e-05 2023-01-21 13:09:59.706060: step: 2040/529, loss: 0.0026039122603833675 2023-01-21 13:10:00.820635: step: 2044/529, loss: 0.00868759211152792 2023-01-21 13:10:01.945143: step: 2048/529, loss: 0.0034478188026696444 2023-01-21 13:10:03.067282: step: 2052/529, loss: 0.0036638739984482527 2023-01-21 13:10:04.183355: step: 2056/529, loss: 0.0008930206531658769 2023-01-21 13:10:05.280055: step: 2060/529, loss: 0.019074583426117897 2023-01-21 13:10:06.410066: step: 2064/529, loss: 0.0005733490106649697 2023-01-21 13:10:07.552478: step: 2068/529, loss: 5.106926255393773e-05 2023-01-21 13:10:08.675615: step: 2072/529, loss: 0.020857907831668854 2023-01-21 13:10:09.803378: step: 2076/529, loss: 0.030022908002138138 2023-01-21 13:10:10.939240: step: 2080/529, loss: 0.0017591476207599044 2023-01-21 13:10:12.031671: step: 2084/529, loss: 0.02714099921286106 2023-01-21 13:10:13.136234: step: 2088/529, loss: 0.33061426877975464 2023-01-21 13:10:14.262851: step: 2092/529, loss: 0.0679231658577919 2023-01-21 13:10:15.379934: step: 2096/529, loss: 0.0006399154663085938 2023-01-21 13:10:16.509080: step: 2100/529, loss: 0.00016908645920921117 2023-01-21 13:10:17.632768: step: 2104/529, loss: 0.0020572661887854338 2023-01-21 13:10:18.759634: step: 2108/529, loss: 0.07673779129981995 2023-01-21 13:10:19.847781: step: 2112/529, loss: 1.0633468264131807e-05 2023-01-21 13:10:20.960066: step: 2116/529, loss: 0.02581310085952282 ================================================== Loss: 0.026 -------------------- Dev: {'event': {'p': 0.5639810426540285, 'r': 0.7922769640479361, 'f1': 0.6589147286821706}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Test: {'event': {'p': 0.6171977302487996, 'r': 0.7979683972911964, 'f1': 0.6960374107802116}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Chinese: {'event': {'p': 0.5208333333333334, 'r': 0.9259259259259259, 'f1': 0.6666666666666667}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Korean: {'event': {'p': 0.625, 'r': 0.6349206349206349, 'f1': 0.6299212598425197}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Russian: {'event': {'p': 0.5111111111111111, 'r': 0.6388888888888888, 'f1': 0.5679012345679012}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 27 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:11:01.696677: step: 4/529, loss: 0.009249688126146793 2023-01-21 13:11:02.843710: step: 8/529, loss: 0.000713348388671875 2023-01-21 13:11:03.956511: step: 12/529, loss: 0.00022373200044967234 2023-01-21 13:11:05.069495: step: 16/529, loss: 0.0004798889276571572 2023-01-21 13:11:06.189866: step: 20/529, loss: 0.020532751455903053 2023-01-21 13:11:07.308173: step: 24/529, loss: 0.0016321182483807206 2023-01-21 13:11:08.448497: step: 28/529, loss: 8.516311936546117e-05 2023-01-21 13:11:09.555208: step: 32/529, loss: 0.0007098674541339278 2023-01-21 13:11:10.678055: step: 36/529, loss: 0.005714798346161842 2023-01-21 13:11:11.805055: step: 40/529, loss: 0.001139783882535994 2023-01-21 13:11:12.929936: step: 44/529, loss: 0.0032330036628991365 2023-01-21 13:11:14.087107: step: 48/529, loss: 0.03191394731402397 2023-01-21 13:11:15.166268: step: 52/529, loss: 0.00016098022751975805 2023-01-21 13:11:16.312766: step: 56/529, loss: 0.007484626490622759 2023-01-21 13:11:17.425861: step: 60/529, loss: 0.007655906490981579 2023-01-21 13:11:18.529636: step: 64/529, loss: 0.0001906395045807585 2023-01-21 13:11:19.651555: step: 68/529, loss: 0.003427696181461215 2023-01-21 13:11:20.764037: step: 72/529, loss: 0.041635800153017044 2023-01-21 13:11:21.859926: step: 76/529, loss: 0.00031147003755904734 2023-01-21 13:11:22.971342: step: 80/529, loss: 6.389617919921875e-05 2023-01-21 13:11:24.112188: step: 84/529, loss: 0.005901527591049671 2023-01-21 13:11:25.194457: step: 88/529, loss: 9.536745437799254e-08 2023-01-21 13:11:26.321938: step: 92/529, loss: 0.0019780159927904606 2023-01-21 13:11:27.402030: step: 96/529, loss: 0.0002933502255473286 2023-01-21 13:11:28.537362: step: 100/529, loss: 0.0012407302856445312 2023-01-21 13:11:29.669516: step: 104/529, loss: 0.0025183677207678556 2023-01-21 13:11:30.778715: step: 108/529, loss: 0.05780811607837677 2023-01-21 13:11:31.893568: step: 112/529, loss: 5.4073334467830136e-05 2023-01-21 13:11:33.010732: step: 116/529, loss: 0.0001324176846537739 2023-01-21 13:11:34.133649: step: 120/529, loss: 0.00012874603271484375 2023-01-21 13:11:35.257596: step: 124/529, loss: 4.5585635234601796e-05 2023-01-21 13:11:36.353566: step: 128/529, loss: 6.999969627941027e-05 2023-01-21 13:11:37.471404: step: 132/529, loss: 0.015172004699707031 2023-01-21 13:11:38.564219: step: 136/529, loss: -2.2888179955771193e-06 2023-01-21 13:11:39.697551: step: 140/529, loss: 0.022600747644901276 2023-01-21 13:11:40.797269: step: 144/529, loss: 2.2888182229507947e-06 2023-01-21 13:11:41.920690: step: 148/529, loss: 0.0007817268487997353 2023-01-21 13:11:43.010435: step: 152/529, loss: 0.0002841949462890625 2023-01-21 13:11:44.129650: step: 156/529, loss: 0.0003963470517192036 2023-01-21 13:11:45.244295: step: 160/529, loss: 8.39233416627394e-06 2023-01-21 13:11:46.355285: step: 164/529, loss: 3.795623706537299e-05 2023-01-21 13:11:47.471105: step: 168/529, loss: 0.00017805100651457906 2023-01-21 13:11:48.619723: step: 172/529, loss: 0.002005005022510886 2023-01-21 13:11:49.738655: step: 176/529, loss: 0.00011520386033225805 2023-01-21 13:11:50.831761: step: 180/529, loss: 4.520416405284777e-05 2023-01-21 13:11:51.922935: step: 184/529, loss: 0.13518676161766052 2023-01-21 13:11:53.039611: step: 188/529, loss: 1.697540210443549e-05 2023-01-21 13:11:54.134677: step: 192/529, loss: 0.0003010749933309853 2023-01-21 13:11:55.274143: step: 196/529, loss: 0.001999282743781805 2023-01-21 13:11:56.392101: step: 200/529, loss: 0.0006336212391033769 2023-01-21 13:11:57.503735: step: 204/529, loss: 0.006953620817512274 2023-01-21 13:11:58.601682: step: 208/529, loss: 1.5068055290612392e-05 2023-01-21 13:11:59.716685: step: 212/529, loss: 3.0517578125e-05 2023-01-21 13:12:00.844287: step: 216/529, loss: 3.9386748539982364e-05 2023-01-21 13:12:01.974257: step: 220/529, loss: 0.0037118911277502775 2023-01-21 13:12:03.082070: step: 224/529, loss: 3.967285010730848e-05 2023-01-21 13:12:04.165951: step: 228/529, loss: 6.079673767089844e-05 2023-01-21 13:12:05.309489: step: 232/529, loss: 0.00022010803513694555 2023-01-21 13:12:06.460604: step: 236/529, loss: 0.04304943233728409 2023-01-21 13:12:07.577707: step: 240/529, loss: 0.00010061264038085938 2023-01-21 13:12:08.669288: step: 244/529, loss: 2.6226043701171875e-06 2023-01-21 13:12:09.756969: step: 248/529, loss: 1.45191490650177 2023-01-21 13:12:10.872842: step: 252/529, loss: 0.0006059646257199347 2023-01-21 13:12:11.996348: step: 256/529, loss: 0.0001846313534770161 2023-01-21 13:12:13.107076: step: 260/529, loss: 0.0012180329067632556 2023-01-21 13:12:14.245043: step: 264/529, loss: 1.029968279908644e-05 2023-01-21 13:12:15.354821: step: 268/529, loss: -7.724762326688506e-06 2023-01-21 13:12:16.475185: step: 272/529, loss: 0.0010332107776775956 2023-01-21 13:12:17.573955: step: 276/529, loss: 9.517669241176918e-05 2023-01-21 13:12:18.684748: step: 280/529, loss: 0.0005318641779012978 2023-01-21 13:12:19.788657: step: 284/529, loss: 0.0512206070125103 2023-01-21 13:12:20.907016: step: 288/529, loss: 0.019914817065000534 2023-01-21 13:12:21.988666: step: 292/529, loss: 2.09808349609375e-05 2023-01-21 13:12:23.105286: step: 296/529, loss: 0.0018430709606036544 2023-01-21 13:12:24.223714: step: 300/529, loss: -8.20159948489163e-06 2023-01-21 13:12:25.340307: step: 304/529, loss: -2.193450927734375e-05 2023-01-21 13:12:26.457664: step: 308/529, loss: 0.0005424499395303428 2023-01-21 13:12:27.530678: step: 312/529, loss: 0.0024175646249204874 2023-01-21 13:12:28.659074: step: 316/529, loss: 0.0012498855357989669 2023-01-21 13:12:29.829284: step: 320/529, loss: 0.00037899016751907766 2023-01-21 13:12:30.927877: step: 324/529, loss: 0.00155048375017941 2023-01-21 13:12:32.036812: step: 328/529, loss: 4.863739013671875e-05 2023-01-21 13:12:33.162184: step: 332/529, loss: 0.0035194396041333675 2023-01-21 13:12:34.271272: step: 336/529, loss: 0.016431475058197975 2023-01-21 13:12:35.396317: step: 340/529, loss: 1.1539459592313506e-05 2023-01-21 13:12:36.529239: step: 344/529, loss: 0.016260242089629173 2023-01-21 13:12:37.675593: step: 348/529, loss: 0.0001227855682373047 2023-01-21 13:12:38.782482: step: 352/529, loss: 0.0006414413801394403 2023-01-21 13:12:39.880851: step: 356/529, loss: 0.029979515820741653 2023-01-21 13:12:41.031077: step: 360/529, loss: 0.00261440291069448 2023-01-21 13:12:42.150320: step: 364/529, loss: 0.00019760131544899195 2023-01-21 13:12:43.261304: step: 368/529, loss: 0.003555011935532093 2023-01-21 13:12:44.366576: step: 372/529, loss: 2.593994213384576e-05 2023-01-21 13:12:45.507413: step: 376/529, loss: 7.381439354503527e-05 2023-01-21 13:12:46.616604: step: 380/529, loss: 0.00032224657479673624 2023-01-21 13:12:47.737121: step: 384/529, loss: 9.412765211891383e-05 2023-01-21 13:12:48.873054: step: 388/529, loss: 0.030803680419921875 2023-01-21 13:12:49.976944: step: 392/529, loss: 0.02362666092813015 2023-01-21 13:12:51.110369: step: 396/529, loss: 0.01728825643658638 2023-01-21 13:12:52.208288: step: 400/529, loss: 0.008720779791474342 2023-01-21 13:12:53.316433: step: 404/529, loss: 0.001058387802913785 2023-01-21 13:12:54.433980: step: 408/529, loss: 0.00369853968732059 2023-01-21 13:12:55.532689: step: 412/529, loss: 0.013082314282655716 2023-01-21 13:12:56.626804: step: 416/529, loss: 0.0007542610983364284 2023-01-21 13:12:57.722396: step: 420/529, loss: 0.001131343888118863 2023-01-21 13:12:58.851110: step: 424/529, loss: 0.0004672050417866558 2023-01-21 13:12:59.983727: step: 428/529, loss: 6.580352783203125e-05 2023-01-21 13:13:01.100404: step: 432/529, loss: 0.0031196593772619963 2023-01-21 13:13:02.209056: step: 436/529, loss: 5.939006950939074e-05 2023-01-21 13:13:03.355911: step: 440/529, loss: 0.013279056176543236 2023-01-21 13:13:04.462891: step: 444/529, loss: 1.2493133908719756e-05 2023-01-21 13:13:05.577645: step: 448/529, loss: 6.294250852079131e-06 2023-01-21 13:13:06.691693: step: 452/529, loss: 4.329681542003527e-05 2023-01-21 13:13:07.789974: step: 456/529, loss: 0.0001659393310546875 2023-01-21 13:13:08.871430: step: 460/529, loss: 8.583069188716763e-07 2023-01-21 13:13:09.974004: step: 464/529, loss: 0.0037823680322617292 2023-01-21 13:13:11.107352: step: 468/529, loss: 0.005554962437599897 2023-01-21 13:13:12.229529: step: 472/529, loss: 0.02116413228213787 2023-01-21 13:13:13.390246: step: 476/529, loss: 1.125335711549269e-05 2023-01-21 13:13:14.527566: step: 480/529, loss: 0.018422506749629974 2023-01-21 13:13:15.646573: step: 484/529, loss: 1.487731969973538e-05 2023-01-21 13:13:16.757112: step: 488/529, loss: 3.61442580469884e-05 2023-01-21 13:13:17.846385: step: 492/529, loss: 9.365082223666832e-05 2023-01-21 13:13:19.016083: step: 496/529, loss: 0.00935821607708931 2023-01-21 13:13:20.163842: step: 500/529, loss: 0.017224693670868874 2023-01-21 13:13:21.280082: step: 504/529, loss: 2.098083541568485e-06 2023-01-21 13:13:22.399452: step: 508/529, loss: -1.716613724056515e-06 2023-01-21 13:13:23.505543: step: 512/529, loss: 0.039348602294921875 2023-01-21 13:13:24.614983: step: 516/529, loss: 0.020688343793153763 2023-01-21 13:13:25.731614: step: 520/529, loss: 3.147125244140625e-05 2023-01-21 13:13:26.819302: step: 524/529, loss: 0.0017095565563067794 2023-01-21 13:13:27.916451: step: 528/529, loss: 1.8692018784349784e-05 2023-01-21 13:13:29.028920: step: 532/529, loss: 0.003843689104542136 2023-01-21 13:13:30.133149: step: 536/529, loss: 0.005428696051239967 2023-01-21 13:13:31.250636: step: 540/529, loss: 0.00017557144747115672 2023-01-21 13:13:32.373906: step: 544/529, loss: 7.629394644936838e-07 2023-01-21 13:13:33.478318: step: 548/529, loss: 0.05489044263958931 2023-01-21 13:13:34.616886: step: 552/529, loss: 0.0044387816451489925 2023-01-21 13:13:35.736168: step: 556/529, loss: -3.814697322468419e-07 2023-01-21 13:13:36.844251: step: 560/529, loss: 0.08018913120031357 2023-01-21 13:13:37.948369: step: 564/529, loss: 0.35270100831985474 2023-01-21 13:13:39.066658: step: 568/529, loss: 0.0008023262489587069 2023-01-21 13:13:40.186978: step: 572/529, loss: 0.00158863072283566 2023-01-21 13:13:41.312624: step: 576/529, loss: 7.925033423816785e-05 2023-01-21 13:13:42.444871: step: 580/529, loss: 0.03548803552985191 2023-01-21 13:13:43.568534: step: 584/529, loss: 9.574890282237902e-05 2023-01-21 13:13:44.689776: step: 588/529, loss: 3.6239625842426904e-06 2023-01-21 13:13:45.767290: step: 592/529, loss: 0.0001885414239950478 2023-01-21 13:13:46.883494: step: 596/529, loss: 0.00014934540376998484 2023-01-21 13:13:48.000357: step: 600/529, loss: 0.0001095295010600239 2023-01-21 13:13:49.140349: step: 604/529, loss: 0.00011510848707985133 2023-01-21 13:13:50.315890: step: 608/529, loss: 0.007095527835190296 2023-01-21 13:13:51.461302: step: 612/529, loss: 0.0008422852260991931 2023-01-21 13:13:52.623966: step: 616/529, loss: 1.430511474609375e-05 2023-01-21 13:13:53.738134: step: 620/529, loss: 0.0004137992800679058 2023-01-21 13:13:54.838529: step: 624/529, loss: 0.0005496978992596269 2023-01-21 13:13:55.927038: step: 628/529, loss: 0.0013875962467864156 2023-01-21 13:13:57.030284: step: 632/529, loss: 3.514289710437879e-05 2023-01-21 13:13:58.151380: step: 636/529, loss: 0.0001331329403910786 2023-01-21 13:13:59.284392: step: 640/529, loss: 0.002977657364681363 2023-01-21 13:14:00.401965: step: 644/529, loss: 0.0007304191822186112 2023-01-21 13:14:01.500864: step: 648/529, loss: 0.0443304069340229 2023-01-21 13:14:02.603103: step: 652/529, loss: 0.0002265930233988911 2023-01-21 13:14:03.713661: step: 656/529, loss: 0.012698746286332607 2023-01-21 13:14:04.808252: step: 660/529, loss: 0.0002633571857586503 2023-01-21 13:14:05.949187: step: 664/529, loss: 0.198590949177742 2023-01-21 13:14:07.050699: step: 668/529, loss: 0.0010515213944017887 2023-01-21 13:14:08.187787: step: 672/529, loss: 0.00012722014798782766 2023-01-21 13:14:09.323128: step: 676/529, loss: 3.662109520519152e-05 2023-01-21 13:14:10.456805: step: 680/529, loss: 0.006915092468261719 2023-01-21 13:14:11.550573: step: 684/529, loss: 0.024335289373993874 2023-01-21 13:14:12.673958: step: 688/529, loss: 3.2615658710710704e-05 2023-01-21 13:14:13.797599: step: 692/529, loss: 0.0022559165954589844 2023-01-21 13:14:14.930786: step: 696/529, loss: 0.0008451461908407509 2023-01-21 13:14:16.131844: step: 700/529, loss: 0.0013437271118164062 2023-01-21 13:14:17.269716: step: 704/529, loss: 0.00838174857199192 2023-01-21 13:14:18.381660: step: 708/529, loss: 0.05835456773638725 2023-01-21 13:14:19.495454: step: 712/529, loss: 0.15174180269241333 2023-01-21 13:14:20.589614: step: 716/529, loss: 0.048986151814460754 2023-01-21 13:14:21.709402: step: 720/529, loss: 0.00033588410587981343 2023-01-21 13:14:22.806105: step: 724/529, loss: 0.002157020615413785 2023-01-21 13:14:23.927023: step: 728/529, loss: 0.009601403027772903 2023-01-21 13:14:25.014528: step: 732/529, loss: 0.000854110752698034 2023-01-21 13:14:26.117025: step: 736/529, loss: 0.006050873082131147 2023-01-21 13:14:27.235436: step: 740/529, loss: 0.001328754355199635 2023-01-21 13:14:28.336647: step: 744/529, loss: 7.486343292839592e-06 2023-01-21 13:14:29.456041: step: 748/529, loss: 0.0005989194032736123 2023-01-21 13:14:30.570907: step: 752/529, loss: 0.03419437259435654 2023-01-21 13:14:31.701837: step: 756/529, loss: 0.0005435943603515625 2023-01-21 13:14:32.824940: step: 760/529, loss: 0.019770454615354538 2023-01-21 13:14:33.907383: step: 764/529, loss: 0.0010349273215979338 2023-01-21 13:14:34.993071: step: 768/529, loss: 0.0013326643966138363 2023-01-21 13:14:36.117678: step: 772/529, loss: 0.00383338937535882 2023-01-21 13:14:37.231898: step: 776/529, loss: 8.98361176950857e-05 2023-01-21 13:14:38.335078: step: 780/529, loss: 0.0009382247808389366 2023-01-21 13:14:39.473492: step: 784/529, loss: 0.004414177034050226 2023-01-21 13:14:40.606823: step: 788/529, loss: 0.0022933960426598787 2023-01-21 13:14:41.752121: step: 792/529, loss: 0.008202457800507545 2023-01-21 13:14:42.859586: step: 796/529, loss: -2.670288040462765e-06 2023-01-21 13:14:43.974579: step: 800/529, loss: 0.001972389407455921 2023-01-21 13:14:45.075729: step: 804/529, loss: 0.017171764746308327 2023-01-21 13:14:46.176783: step: 808/529, loss: 0.004138469696044922 2023-01-21 13:14:47.307594: step: 812/529, loss: 0.01918220706284046 2023-01-21 13:14:48.443214: step: 816/529, loss: 1.8119811784345075e-06 2023-01-21 13:14:49.539353: step: 820/529, loss: 5.135536048328504e-05 2023-01-21 13:14:50.681060: step: 824/529, loss: 0.0001255035458598286 2023-01-21 13:14:51.841791: step: 828/529, loss: 0.001076090382412076 2023-01-21 13:14:52.941214: step: 832/529, loss: 0.0008924484136514366 2023-01-21 13:14:54.053672: step: 836/529, loss: 5.950927879894152e-05 2023-01-21 13:14:55.146129: step: 840/529, loss: 0.0001996994105866179 2023-01-21 13:14:56.254678: step: 844/529, loss: 0.00040988920954987407 2023-01-21 13:14:57.351683: step: 848/529, loss: 1.5926361811580136e-05 2023-01-21 13:14:58.478470: step: 852/529, loss: 0.006637192331254482 2023-01-21 13:14:59.616428: step: 856/529, loss: 0.0005097389221191406 2023-01-21 13:15:00.706720: step: 860/529, loss: 1.9884109860868193e-05 2023-01-21 13:15:01.778636: step: 864/529, loss: 0.0001581192045705393 2023-01-21 13:15:02.893768: step: 868/529, loss: 0.00472679128870368 2023-01-21 13:15:04.021856: step: 872/529, loss: 0.0008339881896972656 2023-01-21 13:15:05.147659: step: 876/529, loss: 0.010357093997299671 2023-01-21 13:15:06.263139: step: 880/529, loss: 9.346007573185489e-06 2023-01-21 13:15:07.367466: step: 884/529, loss: 0.0036376952193677425 2023-01-21 13:15:08.470137: step: 888/529, loss: 3.24249276673072e-06 2023-01-21 13:15:09.574773: step: 892/529, loss: 0.012949180789291859 2023-01-21 13:15:10.674550: step: 896/529, loss: -6.198883397701138e-07 2023-01-21 13:15:11.791111: step: 900/529, loss: 0.004258441738784313 2023-01-21 13:15:12.902005: step: 904/529, loss: 8.869172233971767e-06 2023-01-21 13:15:14.057616: step: 908/529, loss: 0.047396469861269 2023-01-21 13:15:15.196457: step: 912/529, loss: -2.09808349609375e-05 2023-01-21 13:15:16.285299: step: 916/529, loss: 0.004712868016213179 2023-01-21 13:15:17.412896: step: 920/529, loss: 0.02129039727151394 2023-01-21 13:15:18.530511: step: 924/529, loss: 8.544922457076609e-05 2023-01-21 13:15:19.632824: step: 928/529, loss: 2.212524486822076e-05 2023-01-21 13:15:20.760133: step: 932/529, loss: 0.00032005313551053405 2023-01-21 13:15:21.882980: step: 936/529, loss: 0.0009243488311767578 2023-01-21 13:15:22.979955: step: 940/529, loss: 0.014508294872939587 2023-01-21 13:15:24.122207: step: 944/529, loss: 0.0003635406610555947 2023-01-21 13:15:25.242868: step: 948/529, loss: 0.00013170242891646922 2023-01-21 13:15:26.366419: step: 952/529, loss: 3.1471258807869162e-06 2023-01-21 13:15:27.482828: step: 956/529, loss: 2.746581958490424e-05 2023-01-21 13:15:28.575300: step: 960/529, loss: 4.5299530029296875e-05 2023-01-21 13:15:29.683441: step: 964/529, loss: 0.00018787384033203125 2023-01-21 13:15:30.816142: step: 968/529, loss: 0.00263137836009264 2023-01-21 13:15:31.910874: step: 972/529, loss: 0.009904003702104092 2023-01-21 13:15:33.073670: step: 976/529, loss: 0.0003784179862122983 2023-01-21 13:15:34.191748: step: 980/529, loss: 0.0004046917019877583 2023-01-21 13:15:35.289311: step: 984/529, loss: 0.003150367643684149 2023-01-21 13:15:36.407642: step: 988/529, loss: 0.01798992231488228 2023-01-21 13:15:37.530958: step: 992/529, loss: 4.00543194700731e-06 2023-01-21 13:15:38.638049: step: 996/529, loss: 0.024334335699677467 2023-01-21 13:15:39.759847: step: 1000/529, loss: 0.011549186892807484 2023-01-21 13:15:40.876029: step: 1004/529, loss: 3.795623706537299e-05 2023-01-21 13:15:41.977064: step: 1008/529, loss: 8.773804438533261e-05 2023-01-21 13:15:43.062365: step: 1012/529, loss: 0.004868126008659601 2023-01-21 13:15:44.199158: step: 1016/529, loss: 0.005821609869599342 2023-01-21 13:15:45.300616: step: 1020/529, loss: 2.2792815798311494e-05 2023-01-21 13:15:46.402752: step: 1024/529, loss: 0.0018356323707848787 2023-01-21 13:15:47.521996: step: 1028/529, loss: 0.0001569747837493196 2023-01-21 13:15:48.618021: step: 1032/529, loss: 0.0028627398423850536 2023-01-21 13:15:49.732881: step: 1036/529, loss: 0.009147453121840954 2023-01-21 13:15:50.815452: step: 1040/529, loss: 5.7220458984375e-06 2023-01-21 13:15:51.932244: step: 1044/529, loss: 0.004546642303466797 2023-01-21 13:15:53.063721: step: 1048/529, loss: 4.987716965842992e-05 2023-01-21 13:15:54.191684: step: 1052/529, loss: 0.02788076363503933 2023-01-21 13:15:55.309811: step: 1056/529, loss: 6.904602196300402e-05 2023-01-21 13:15:56.370927: step: 1060/529, loss: 0.00038757326547056437 2023-01-21 13:15:57.480745: step: 1064/529, loss: 0.0011676788562908769 2023-01-21 13:15:58.582928: step: 1068/529, loss: 5.769729796156753e-06 2023-01-21 13:15:59.704413: step: 1072/529, loss: 0.11478310078382492 2023-01-21 13:16:00.830725: step: 1076/529, loss: 0.0031817913986742496 2023-01-21 13:16:01.934905: step: 1080/529, loss: 0.0009860516292974353 2023-01-21 13:16:03.081954: step: 1084/529, loss: 0.0174013152718544 2023-01-21 13:16:04.194313: step: 1088/529, loss: 0.0009235382312908769 2023-01-21 13:16:05.291781: step: 1092/529, loss: 0.003840160556137562 2023-01-21 13:16:06.409475: step: 1096/529, loss: 6.971359835006297e-05 2023-01-21 13:16:07.527199: step: 1100/529, loss: 0.03844146803021431 2023-01-21 13:16:08.624462: step: 1104/529, loss: 0.0017689228989183903 2023-01-21 13:16:09.721938: step: 1108/529, loss: 0.002466583391651511 2023-01-21 13:16:10.850725: step: 1112/529, loss: 0.008643818087875843 2023-01-21 13:16:11.974773: step: 1116/529, loss: 6.47544875391759e-05 2023-01-21 13:16:13.075871: step: 1120/529, loss: 0.00039443967398256063 2023-01-21 13:16:14.225445: step: 1124/529, loss: 0.015641022473573685 2023-01-21 13:16:15.330830: step: 1128/529, loss: 0.00446662912145257 2023-01-21 13:16:16.452739: step: 1132/529, loss: 0.045239828526973724 2023-01-21 13:16:17.579165: step: 1136/529, loss: 0.03819665685296059 2023-01-21 13:16:18.691542: step: 1140/529, loss: 2.117157055181451e-05 2023-01-21 13:16:19.774919: step: 1144/529, loss: 0.00012054442777298391 2023-01-21 13:16:20.888931: step: 1148/529, loss: 0.03194008022546768 2023-01-21 13:16:21.999959: step: 1152/529, loss: 1.1931575536727905 2023-01-21 13:16:23.110455: step: 1156/529, loss: 0.05352325364947319 2023-01-21 13:16:24.211190: step: 1160/529, loss: 0.03431911766529083 2023-01-21 13:16:25.303820: step: 1164/529, loss: 0.009469223208725452 2023-01-21 13:16:26.409567: step: 1168/529, loss: 0.004056835547089577 2023-01-21 13:16:27.514380: step: 1172/529, loss: 0.0005332946893759072 2023-01-21 13:16:28.627936: step: 1176/529, loss: 0.004301262088119984 2023-01-21 13:16:29.747903: step: 1180/529, loss: 0.0010605811839923263 2023-01-21 13:16:30.875062: step: 1184/529, loss: 0.0005177497514523566 2023-01-21 13:16:31.992826: step: 1188/529, loss: 0.011030006222426891 2023-01-21 13:16:33.100651: step: 1192/529, loss: 7.057189577608369e-06 2023-01-21 13:16:34.234119: step: 1196/529, loss: 8.983612497104332e-05 2023-01-21 13:16:35.330986: step: 1200/529, loss: 0.00803918857127428 2023-01-21 13:16:36.424357: step: 1204/529, loss: 0.0003170013369526714 2023-01-21 13:16:37.522404: step: 1208/529, loss: 0.0014414788456633687 2023-01-21 13:16:38.644665: step: 1212/529, loss: 6.198883056640625e-06 2023-01-21 13:16:39.759573: step: 1216/529, loss: 0.02641163021326065 2023-01-21 13:16:40.872808: step: 1220/529, loss: 0.00011849403381347656 2023-01-21 13:16:41.987898: step: 1224/529, loss: 0.030809976160526276 2023-01-21 13:16:43.099332: step: 1228/529, loss: 0.00039272307185456157 2023-01-21 13:16:44.206853: step: 1232/529, loss: 0.00017204285541083664 2023-01-21 13:16:45.295425: step: 1236/529, loss: 0.013741064816713333 2023-01-21 13:16:46.406169: step: 1240/529, loss: 4.9591067181609105e-06 2023-01-21 13:16:47.490660: step: 1244/529, loss: 0.07047024369239807 2023-01-21 13:16:48.597484: step: 1248/529, loss: 0.007892417721450329 2023-01-21 13:16:49.702479: step: 1252/529, loss: 0.01241369266062975 2023-01-21 13:16:50.821713: step: 1256/529, loss: 2.532005419197958e-05 2023-01-21 13:16:51.938292: step: 1260/529, loss: 1.0585785275907256e-05 2023-01-21 13:16:53.005600: step: 1264/529, loss: 0.5731164813041687 2023-01-21 13:16:54.124060: step: 1268/529, loss: 1.5258788153005298e-06 2023-01-21 13:16:55.232005: step: 1272/529, loss: 0.0003201484796591103 2023-01-21 13:16:56.322068: step: 1276/529, loss: 9.5367431640625e-06 2023-01-21 13:16:57.452238: step: 1280/529, loss: 1.6975403923424892e-05 2023-01-21 13:16:58.590401: step: 1284/529, loss: 0.0057170395739376545 2023-01-21 13:16:59.669947: step: 1288/529, loss: 0.0005504131549969316 2023-01-21 13:17:00.804191: step: 1292/529, loss: 0.019441604614257812 2023-01-21 13:17:01.921473: step: 1296/529, loss: 0.0054035186767578125 2023-01-21 13:17:03.033378: step: 1300/529, loss: 0.009398058988153934 2023-01-21 13:17:04.135352: step: 1304/529, loss: 0.003941154573112726 2023-01-21 13:17:05.242310: step: 1308/529, loss: 0.0706939697265625 2023-01-21 13:17:06.379306: step: 1312/529, loss: 0.00103168492205441 2023-01-21 13:17:07.480801: step: 1316/529, loss: 0.0002323150692973286 2023-01-21 13:17:08.585015: step: 1320/529, loss: 7.314681715797633e-05 2023-01-21 13:17:09.717447: step: 1324/529, loss: 0.005231034941971302 2023-01-21 13:17:10.799932: step: 1328/529, loss: 0.0006383895524777472 2023-01-21 13:17:11.939803: step: 1332/529, loss: 0.04486904293298721 2023-01-21 13:17:13.065019: step: 1336/529, loss: 0.02622051350772381 2023-01-21 13:17:14.187921: step: 1340/529, loss: 0.0026576996315270662 2023-01-21 13:17:15.278426: step: 1344/529, loss: 1.215934844367439e-05 2023-01-21 13:17:16.406581: step: 1348/529, loss: 0.021055983379483223 2023-01-21 13:17:17.558297: step: 1352/529, loss: 0.001291451626457274 2023-01-21 13:17:18.673978: step: 1356/529, loss: 0.0005256652948446572 2023-01-21 13:17:19.792178: step: 1360/529, loss: -2.403259350103326e-05 2023-01-21 13:17:20.950558: step: 1364/529, loss: 0.016245460137724876 2023-01-21 13:17:22.040122: step: 1368/529, loss: 0.03890140354633331 2023-01-21 13:17:23.143219: step: 1372/529, loss: 0.0002449035819154233 2023-01-21 13:17:24.297186: step: 1376/529, loss: 0.15285950899124146 2023-01-21 13:17:25.406947: step: 1380/529, loss: 0.08986129611730576 2023-01-21 13:17:26.515951: step: 1384/529, loss: 0.012361908331513405 2023-01-21 13:17:27.625057: step: 1388/529, loss: 0.008724594488739967 2023-01-21 13:17:28.735072: step: 1392/529, loss: 0.0007171630859375 2023-01-21 13:17:29.862240: step: 1396/529, loss: 0.0007783889886923134 2023-01-21 13:17:30.970001: step: 1400/529, loss: 0.00027761459932662547 2023-01-21 13:17:32.097230: step: 1404/529, loss: 0.001807403634302318 2023-01-21 13:17:33.267653: step: 1408/529, loss: 0.0015897273551672697 2023-01-21 13:17:34.384552: step: 1412/529, loss: 5.1021575927734375e-05 2023-01-21 13:17:35.502915: step: 1416/529, loss: 1.0681153071345761e-05 2023-01-21 13:17:36.601585: step: 1420/529, loss: 0.016759108752012253 2023-01-21 13:17:37.731035: step: 1424/529, loss: 0.0016318499110639095 2023-01-21 13:17:38.873365: step: 1428/529, loss: 0.03196106106042862 2023-01-21 13:17:39.976797: step: 1432/529, loss: 0.0001388549862895161 2023-01-21 13:17:41.092062: step: 1436/529, loss: 1.678466833254788e-05 2023-01-21 13:17:42.188314: step: 1440/529, loss: 0.0005296707386150956 2023-01-21 13:17:43.311996: step: 1444/529, loss: 0.0009529591188766062 2023-01-21 13:17:44.417286: step: 1448/529, loss: 0.0046860696747899055 2023-01-21 13:17:45.517612: step: 1452/529, loss: 0.032364655286073685 2023-01-21 13:17:46.641039: step: 1456/529, loss: 0.01510009728372097 2023-01-21 13:17:47.756672: step: 1460/529, loss: 0.0028751373756676912 2023-01-21 13:17:48.890816: step: 1464/529, loss: 0.001624679658561945 2023-01-21 13:17:50.017747: step: 1468/529, loss: 0.09992923587560654 2023-01-21 13:17:51.142953: step: 1472/529, loss: 7.562637620139867e-05 2023-01-21 13:17:52.215798: step: 1476/529, loss: 5.626678557746345e-06 2023-01-21 13:17:53.300616: step: 1480/529, loss: 0.00019855500431731343 2023-01-21 13:17:54.471809: step: 1484/529, loss: 0.0008626937633380294 2023-01-21 13:17:55.593769: step: 1488/529, loss: 0.024179935455322266 2023-01-21 13:17:56.711135: step: 1492/529, loss: 0.00036945342435501516 2023-01-21 13:17:57.829511: step: 1496/529, loss: 0.0031189920846372843 2023-01-21 13:17:58.934243: step: 1500/529, loss: -3.14712519866589e-06 2023-01-21 13:18:00.044447: step: 1504/529, loss: 0.007780313491821289 2023-01-21 13:18:01.164050: step: 1508/529, loss: 0.027561282739043236 2023-01-21 13:18:02.296859: step: 1512/529, loss: 0.0009529113885946572 2023-01-21 13:18:03.420282: step: 1516/529, loss: 0.018427658826112747 2023-01-21 13:18:04.536824: step: 1520/529, loss: 0.06845531612634659 2023-01-21 13:18:05.673585: step: 1524/529, loss: 0.029588699340820312 2023-01-21 13:18:06.813401: step: 1528/529, loss: 0.02111683040857315 2023-01-21 13:18:07.905343: step: 1532/529, loss: 0.02276000939309597 2023-01-21 13:18:09.014532: step: 1536/529, loss: 0.02129554934799671 2023-01-21 13:18:10.136040: step: 1540/529, loss: 0.0004553794860839844 2023-01-21 13:18:11.278919: step: 1544/529, loss: 0.0002037048398051411 2023-01-21 13:18:12.384469: step: 1548/529, loss: 0.0003536224539857358 2023-01-21 13:18:13.489019: step: 1552/529, loss: 0.0057548522017896175 2023-01-21 13:18:14.596860: step: 1556/529, loss: 0.5353145599365234 2023-01-21 13:18:15.702307: step: 1560/529, loss: 0.0004351615789346397 2023-01-21 13:18:16.803878: step: 1564/529, loss: 3.3950807846849784e-05 2023-01-21 13:18:17.909424: step: 1568/529, loss: 0.016290094703435898 2023-01-21 13:18:19.033883: step: 1572/529, loss: 0.028105545789003372 2023-01-21 13:18:20.182965: step: 1576/529, loss: 4.37736525782384e-05 2023-01-21 13:18:21.286004: step: 1580/529, loss: 0.001196575234644115 2023-01-21 13:18:22.398250: step: 1584/529, loss: 0.018286513164639473 2023-01-21 13:18:23.503847: step: 1588/529, loss: 0.004174232482910156 2023-01-21 13:18:24.612508: step: 1592/529, loss: 0.003111171768978238 2023-01-21 13:18:25.714545: step: 1596/529, loss: 0.00027942657470703125 2023-01-21 13:18:26.820938: step: 1600/529, loss: 7.848739915061742e-05 2023-01-21 13:18:27.919162: step: 1604/529, loss: 0.017615366727113724 2023-01-21 13:18:29.020052: step: 1608/529, loss: 1.7166138377433526e-06 2023-01-21 13:18:30.134194: step: 1612/529, loss: 4.653930591302924e-05 2023-01-21 13:18:31.255360: step: 1616/529, loss: 0.1845863312482834 2023-01-21 13:18:32.351560: step: 1620/529, loss: 0.00018749237642623484 2023-01-21 13:18:33.466896: step: 1624/529, loss: 0.0024513243697583675 2023-01-21 13:18:34.573559: step: 1628/529, loss: 1.964569128176663e-05 2023-01-21 13:18:35.713874: step: 1632/529, loss: 0.00013608933659270406 2023-01-21 13:18:36.804042: step: 1636/529, loss: 0.002893829485401511 2023-01-21 13:18:37.895471: step: 1640/529, loss: 5.741119457525201e-05 2023-01-21 13:18:39.014903: step: 1644/529, loss: 0.013698482885956764 2023-01-21 13:18:40.136666: step: 1648/529, loss: 0.0008569717756472528 2023-01-21 13:18:41.280498: step: 1652/529, loss: 0.0013421059120446444 2023-01-21 13:18:42.376362: step: 1656/529, loss: 3.7813188100699335e-05 2023-01-21 13:18:43.508939: step: 1660/529, loss: 0.05416569858789444 2023-01-21 13:18:44.645963: step: 1664/529, loss: 0.06966057419776917 2023-01-21 13:18:45.774814: step: 1668/529, loss: 1.2731551578326616e-05 2023-01-21 13:18:46.868064: step: 1672/529, loss: 0.007915640249848366 2023-01-21 13:18:48.028556: step: 1676/529, loss: 0.012709617614746094 2023-01-21 13:18:49.137972: step: 1680/529, loss: 0.006764126010239124 2023-01-21 13:18:50.270584: step: 1684/529, loss: 0.014030933380126953 2023-01-21 13:18:51.378933: step: 1688/529, loss: 0.02049226127564907 2023-01-21 13:18:52.492049: step: 1692/529, loss: 0.00020647048950195312 2023-01-21 13:18:53.606511: step: 1696/529, loss: 0.0025929692201316357 2023-01-21 13:18:54.724531: step: 1700/529, loss: 0.0012098312145099044 2023-01-21 13:18:55.870877: step: 1704/529, loss: 0.011114120483398438 2023-01-21 13:18:56.993737: step: 1708/529, loss: 0.0011238098377361894 2023-01-21 13:18:58.083574: step: 1712/529, loss: 0.007730007171630859 2023-01-21 13:18:59.159431: step: 1716/529, loss: 0.0012298583751544356 2023-01-21 13:19:00.297613: step: 1720/529, loss: 0.006230831146240234 2023-01-21 13:19:01.400501: step: 1724/529, loss: 0.02225780487060547 2023-01-21 13:19:02.489637: step: 1728/529, loss: 0.0054454803466796875 2023-01-21 13:19:03.593059: step: 1732/529, loss: 0.00015673637972213328 2023-01-21 13:19:04.734543: step: 1736/529, loss: 3.280639793956652e-05 2023-01-21 13:19:05.829373: step: 1740/529, loss: 0.07476921379566193 2023-01-21 13:19:06.982475: step: 1744/529, loss: 0.006712436676025391 2023-01-21 13:19:08.106820: step: 1748/529, loss: 3.4141543437726796e-05 2023-01-21 13:19:09.220427: step: 1752/529, loss: 0.08540745079517365 2023-01-21 13:19:10.338557: step: 1756/529, loss: 0.00019397735013626516 2023-01-21 13:19:11.493552: step: 1760/529, loss: 0.13180121779441833 2023-01-21 13:19:12.621080: step: 1764/529, loss: 0.0006969451787881553 2023-01-21 13:19:13.749721: step: 1768/529, loss: 0.004588651936501265 2023-01-21 13:19:14.883464: step: 1772/529, loss: 0.012662315741181374 2023-01-21 13:19:16.000825: step: 1776/529, loss: 2.174377368646674e-05 2023-01-21 13:19:17.138478: step: 1780/529, loss: 0.0035919188521802425 2023-01-21 13:19:18.252326: step: 1784/529, loss: 9.250640869140625e-05 2023-01-21 13:19:19.343265: step: 1788/529, loss: 0.001991844270378351 2023-01-21 13:19:20.488583: step: 1792/529, loss: 0.0007330894004553556 2023-01-21 13:19:21.610016: step: 1796/529, loss: 0.036548711359500885 2023-01-21 13:19:22.699705: step: 1800/529, loss: 0.000274848920525983 2023-01-21 13:19:23.786332: step: 1804/529, loss: 0.0002880096435546875 2023-01-21 13:19:24.909546: step: 1808/529, loss: 0.00017296076111961156 2023-01-21 13:19:26.008450: step: 1812/529, loss: 0.0006086349603720009 2023-01-21 13:19:27.112663: step: 1816/529, loss: 0.00013275146193336695 2023-01-21 13:19:28.224435: step: 1820/529, loss: 0.0446077361702919 2023-01-21 13:19:29.341634: step: 1824/529, loss: 8.735656592762098e-05 2023-01-21 13:19:30.401566: step: 1828/529, loss: 0.030761336907744408 2023-01-21 13:19:31.474077: step: 1832/529, loss: 3.414153979974799e-05 2023-01-21 13:19:32.604458: step: 1836/529, loss: 0.0003948211669921875 2023-01-21 13:19:33.715425: step: 1840/529, loss: 0.0025129318237304688 2023-01-21 13:19:34.846974: step: 1844/529, loss: 0.0031431198585778475 2023-01-21 13:19:35.996655: step: 1848/529, loss: 0.008995628915727139 2023-01-21 13:19:37.126077: step: 1852/529, loss: 7.22885160939768e-05 2023-01-21 13:19:38.254931: step: 1856/529, loss: 7.22885160939768e-05 2023-01-21 13:19:39.352036: step: 1860/529, loss: 2.9850005375919864e-05 2023-01-21 13:19:40.442596: step: 1864/529, loss: 0.001070261001586914 2023-01-21 13:19:41.552554: step: 1868/529, loss: 1.33514404296875e-05 2023-01-21 13:19:42.687367: step: 1872/529, loss: 0.008897877298295498 2023-01-21 13:19:43.791724: step: 1876/529, loss: 7.82012921263231e-06 2023-01-21 13:19:44.920644: step: 1880/529, loss: 6.752014451194555e-05 2023-01-21 13:19:46.079364: step: 1884/529, loss: -6.389617738022935e-06 2023-01-21 13:19:47.245763: step: 1888/529, loss: 0.02295098267495632 2023-01-21 13:19:48.359643: step: 1892/529, loss: 0.01791095733642578 2023-01-21 13:19:49.465915: step: 1896/529, loss: 3.0517578125e-05 2023-01-21 13:19:50.558053: step: 1900/529, loss: 0.0015121459728106856 2023-01-21 13:19:51.667984: step: 1904/529, loss: 0.0013645171420648694 2023-01-21 13:19:52.762456: step: 1908/529, loss: 1.1157989320054185e-05 2023-01-21 13:19:53.870539: step: 1912/529, loss: -3.814697322468419e-07 2023-01-21 13:19:55.013457: step: 1916/529, loss: 0.0026739120949059725 2023-01-21 13:19:56.149578: step: 1920/529, loss: 0.008389092050492764 2023-01-21 13:19:57.281268: step: 1924/529, loss: 0.0005657196161337197 2023-01-21 13:19:58.378143: step: 1928/529, loss: 0.038118936121463776 2023-01-21 13:19:59.490268: step: 1932/529, loss: 3.52382667188067e-05 2023-01-21 13:20:00.641713: step: 1936/529, loss: 6.122589547885582e-05 2023-01-21 13:20:01.759954: step: 1940/529, loss: 1.2302398317842744e-05 2023-01-21 13:20:02.881789: step: 1944/529, loss: 0.00013666153245139867 2023-01-21 13:20:03.997807: step: 1948/529, loss: 0.07349758595228195 2023-01-21 13:20:05.140885: step: 1952/529, loss: -1.811981201171875e-05 2023-01-21 13:20:06.247007: step: 1956/529, loss: 0.0015071391826495528 2023-01-21 13:20:07.347437: step: 1960/529, loss: 0.026781557127833366 2023-01-21 13:20:08.445215: step: 1964/529, loss: 0.07961349934339523 2023-01-21 13:20:09.607555: step: 1968/529, loss: 0.013255120255053043 2023-01-21 13:20:10.678726: step: 1972/529, loss: 4.148483185417717e-06 2023-01-21 13:20:11.798235: step: 1976/529, loss: 0.002250194549560547 2023-01-21 13:20:12.894452: step: 1980/529, loss: 0.00086212158203125 2023-01-21 13:20:14.044923: step: 1984/529, loss: 5.178451829124242e-05 2023-01-21 13:20:15.146926: step: 1988/529, loss: 0.019272232428193092 2023-01-21 13:20:16.273933: step: 1992/529, loss: 0.001979827880859375 2023-01-21 13:20:17.394469: step: 1996/529, loss: 0.00032019615173339844 2023-01-21 13:20:18.506898: step: 2000/529, loss: 1.4972686585679185e-05 2023-01-21 13:20:19.628374: step: 2004/529, loss: 7.562637620139867e-05 2023-01-21 13:20:20.746528: step: 2008/529, loss: 0.0007724761962890625 2023-01-21 13:20:21.840320: step: 2012/529, loss: 1.0728836059570312e-05 2023-01-21 13:20:22.963553: step: 2016/529, loss: 0.03607349470257759 2023-01-21 13:20:24.084420: step: 2020/529, loss: 0.0013422967167571187 2023-01-21 13:20:25.169854: step: 2024/529, loss: 2.0214916730765253e-05 2023-01-21 13:20:26.276947: step: 2028/529, loss: 0.00013995172048453242 2023-01-21 13:20:27.401843: step: 2032/529, loss: 3.6716460272145923e-06 2023-01-21 13:20:28.508530: step: 2036/529, loss: 0.07480011880397797 2023-01-21 13:20:29.633357: step: 2040/529, loss: 6.885529001010582e-05 2023-01-21 13:20:30.762200: step: 2044/529, loss: 0.05830421298742294 2023-01-21 13:20:31.873678: step: 2048/529, loss: 0.012754631228744984 2023-01-21 13:20:32.987366: step: 2052/529, loss: 0.000579833984375 2023-01-21 13:20:34.090505: step: 2056/529, loss: 0.002208614256232977 2023-01-21 13:20:35.195078: step: 2060/529, loss: 0.00030879973201081157 2023-01-21 13:20:36.310552: step: 2064/529, loss: 1.163482647825731e-05 2023-01-21 13:20:37.422594: step: 2068/529, loss: 0.0025478361640125513 2023-01-21 13:20:38.553796: step: 2072/529, loss: 0.025456810370087624 2023-01-21 13:20:39.698715: step: 2076/529, loss: 0.0857234001159668 2023-01-21 13:20:40.869307: step: 2080/529, loss: 0.01265411451458931 2023-01-21 13:20:41.965404: step: 2084/529, loss: 0.017270851880311966 2023-01-21 13:20:43.080016: step: 2088/529, loss: 7.638931856490672e-05 2023-01-21 13:20:44.175066: step: 2092/529, loss: 1.9931794668082148e-05 2023-01-21 13:20:45.303506: step: 2096/529, loss: 0.02219982258975506 2023-01-21 13:20:46.431518: step: 2100/529, loss: 4.19616708313697e-06 2023-01-21 13:20:47.527003: step: 2104/529, loss: 0.0005815505865029991 2023-01-21 13:20:48.702300: step: 2108/529, loss: 0.011740684509277344 2023-01-21 13:20:49.836134: step: 2112/529, loss: 0.002584457630291581 2023-01-21 13:20:50.945260: step: 2116/529, loss: 0.0003216266632080078 ================================================== Loss: 0.018 -------------------- Dev: {'event': {'p': 0.582591493570722, 'r': 0.7842876165113183, 'f1': 0.6685584562996595}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Test: {'event': {'p': 0.6431818181818182, 'r': 0.7985327313769752, 'f1': 0.7124874118831823}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Chinese: {'event': {'p': 0.5909090909090909, 'r': 0.9629629629629629, 'f1': 0.732394366197183}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Korean: {'event': {'p': 0.5789473684210527, 'r': 0.5238095238095238, 'f1': 0.5500000000000002}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} 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': 27} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 28 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:21:31.741822: step: 4/529, loss: 4.1675568354548886e-05 2023-01-21 13:21:32.842374: step: 8/529, loss: 0.0019330024952068925 2023-01-21 13:21:33.951562: step: 12/529, loss: 0.0040760040283203125 2023-01-21 13:21:35.067230: step: 16/529, loss: 0.003063392825424671 2023-01-21 13:21:36.164352: step: 20/529, loss: 7.476806786144152e-05 2023-01-21 13:21:37.262441: step: 24/529, loss: 0.016347313299775124 2023-01-21 13:21:38.379765: step: 28/529, loss: 0.0026133540086448193 2023-01-21 13:21:39.481498: step: 32/529, loss: 0.006859302520751953 2023-01-21 13:21:40.602501: step: 36/529, loss: 0.04303283989429474 2023-01-21 13:21:41.698311: step: 40/529, loss: 1.201629675051663e-05 2023-01-21 13:21:42.788041: step: 44/529, loss: 0.0008183956379070878 2023-01-21 13:21:43.919619: step: 48/529, loss: 6.675719760096399e-06 2023-01-21 13:21:45.024001: step: 52/529, loss: 0.042644403874874115 2023-01-21 13:21:46.117285: step: 56/529, loss: 0.04410209879279137 2023-01-21 13:21:47.230448: step: 60/529, loss: 3.547668529790826e-05 2023-01-21 13:21:48.343751: step: 64/529, loss: 0.0013964653480798006 2023-01-21 13:21:49.463147: step: 68/529, loss: 0.0005510210758075118 2023-01-21 13:21:50.547044: step: 72/529, loss: 0.010648155584931374 2023-01-21 13:21:51.673496: step: 76/529, loss: 0.018959617242217064 2023-01-21 13:21:52.814066: step: 80/529, loss: 0.0011581421131268144 2023-01-21 13:21:53.932419: step: 84/529, loss: 5.645752025884576e-05 2023-01-21 13:21:55.070663: step: 88/529, loss: 0.003786277724429965 2023-01-21 13:21:56.228642: step: 92/529, loss: 0.015619468875229359 2023-01-21 13:21:57.328007: step: 96/529, loss: 0.00026931765023618937 2023-01-21 13:21:58.437468: step: 100/529, loss: 7.534027190558845e-06 2023-01-21 13:21:59.547834: step: 104/529, loss: 3.361701828907826e-06 2023-01-21 13:22:00.671436: step: 108/529, loss: 0.006878566928207874 2023-01-21 13:22:01.780080: step: 112/529, loss: 2.536773718020413e-05 2023-01-21 13:22:02.889548: step: 116/529, loss: 0.00047340395394712687 2023-01-21 13:22:04.002607: step: 120/529, loss: 0.0002574920654296875 2023-01-21 13:22:05.113807: step: 124/529, loss: 1.7166139514301904e-06 2023-01-21 13:22:06.199614: step: 128/529, loss: 0.00016016959853004664 2023-01-21 13:22:07.308846: step: 132/529, loss: 0.06166372075676918 2023-01-21 13:22:08.408083: step: 136/529, loss: 0.0010903357760980725 2023-01-21 13:22:09.500212: step: 140/529, loss: 7.343292054429185e-06 2023-01-21 13:22:10.579253: step: 144/529, loss: 0.04374518245458603 2023-01-21 13:22:11.668649: step: 148/529, loss: 0.017122268676757812 2023-01-21 13:22:12.761813: step: 152/529, loss: 0.021277237683534622 2023-01-21 13:22:13.866098: step: 156/529, loss: 3.4713742934400216e-05 2023-01-21 13:22:14.992364: step: 160/529, loss: 0.00244483957067132 2023-01-21 13:22:16.100990: step: 164/529, loss: 0.0003370285267010331 2023-01-21 13:22:17.215844: step: 168/529, loss: 0.027969172224402428 2023-01-21 13:22:18.341171: step: 172/529, loss: 0.0002017974911723286 2023-01-21 13:22:19.490711: step: 176/529, loss: -7.915497008070815e-06 2023-01-21 13:22:20.670968: step: 180/529, loss: 3.013610876223538e-05 2023-01-21 13:22:21.773022: step: 184/529, loss: 0.016840744763612747 2023-01-21 13:22:22.948087: step: 188/529, loss: 0.0024993896950036287 2023-01-21 13:22:24.084982: step: 192/529, loss: 0.0007230758783407509 2023-01-21 13:22:25.219069: step: 196/529, loss: 0.09292087703943253 2023-01-21 13:22:26.316184: step: 200/529, loss: 9.250641596736386e-06 2023-01-21 13:22:27.419722: step: 204/529, loss: 0.04134989157319069 2023-01-21 13:22:28.527941: step: 208/529, loss: 0.022498415783047676 2023-01-21 13:22:29.638486: step: 212/529, loss: 0.04449932649731636 2023-01-21 13:22:30.756125: step: 216/529, loss: 8.840560622047633e-05 2023-01-21 13:22:31.863549: step: 220/529, loss: 0.002008151961490512 2023-01-21 13:22:32.999717: step: 224/529, loss: 3.643035961431451e-05 2023-01-21 13:22:34.145372: step: 228/529, loss: 0.0123443603515625 2023-01-21 13:22:35.263955: step: 232/529, loss: 4.301071021473035e-05 2023-01-21 13:22:36.373517: step: 236/529, loss: 4.291534423828125e-05 2023-01-21 13:22:37.509431: step: 240/529, loss: 0.0002036094811046496 2023-01-21 13:22:38.612951: step: 244/529, loss: 4.711151268566027e-05 2023-01-21 13:22:39.710771: step: 248/529, loss: 0.001927280449308455 2023-01-21 13:22:40.806688: step: 252/529, loss: 0.00010385513451183215 2023-01-21 13:22:41.887756: step: 256/529, loss: 6.580352783203125e-05 2023-01-21 13:22:43.031780: step: 260/529, loss: 0.0008489609463140368 2023-01-21 13:22:44.149548: step: 264/529, loss: 0.0055624013766646385 2023-01-21 13:22:45.261675: step: 268/529, loss: 0.4556151032447815 2023-01-21 13:22:46.375544: step: 272/529, loss: 0.0001372337428620085 2023-01-21 13:22:47.486668: step: 276/529, loss: 0.02538023144006729 2023-01-21 13:22:48.624744: step: 280/529, loss: 0.0007228851318359375 2023-01-21 13:22:49.727733: step: 284/529, loss: 0.0503961555659771 2023-01-21 13:22:50.865950: step: 288/529, loss: 0.01491613406687975 2023-01-21 13:22:51.977915: step: 292/529, loss: 2.2792815798311494e-05 2023-01-21 13:22:53.096574: step: 296/529, loss: 0.0011026383144780993 2023-01-21 13:22:54.202240: step: 300/529, loss: 0.049337007105350494 2023-01-21 13:22:55.314596: step: 304/529, loss: 0.018014907836914062 2023-01-21 13:22:56.396813: step: 308/529, loss: 6.961822509765625e-05 2023-01-21 13:22:57.506549: step: 312/529, loss: 4.863739377469756e-06 2023-01-21 13:22:58.593424: step: 316/529, loss: 0.00013365744962356985 2023-01-21 13:22:59.728760: step: 320/529, loss: 0.019527243450284004 2023-01-21 13:23:00.819587: step: 324/529, loss: 0.00510063162073493 2023-01-21 13:23:01.901136: step: 328/529, loss: 3.147125244140625e-05 2023-01-21 13:23:03.036593: step: 332/529, loss: 0.04550590366125107 2023-01-21 13:23:04.149837: step: 336/529, loss: 0.0004485130193643272 2023-01-21 13:23:05.279344: step: 340/529, loss: 3.070831371587701e-05 2023-01-21 13:23:06.375612: step: 344/529, loss: -6.055831818230217e-06 2023-01-21 13:23:07.483124: step: 348/529, loss: 2.384185791015625e-05 2023-01-21 13:23:08.589072: step: 352/529, loss: 0.00046577455941587687 2023-01-21 13:23:09.690096: step: 356/529, loss: 0.0888335257768631 2023-01-21 13:23:10.810653: step: 360/529, loss: 0.061147596687078476 2023-01-21 13:23:11.905155: step: 364/529, loss: 0.07501792907714844 2023-01-21 13:23:13.027805: step: 368/529, loss: 0.035291098058223724 2023-01-21 13:23:14.155682: step: 372/529, loss: 0.03167152777314186 2023-01-21 13:23:15.263778: step: 376/529, loss: 3.452301098150201e-05 2023-01-21 13:23:16.395631: step: 380/529, loss: 0.0009988785022869706 2023-01-21 13:23:17.523107: step: 384/529, loss: 0.07250308990478516 2023-01-21 13:23:18.661603: step: 388/529, loss: 8.521079871570691e-05 2023-01-21 13:23:19.746236: step: 392/529, loss: 0.031995583325624466 2023-01-21 13:23:20.853019: step: 396/529, loss: 0.009700489230453968 2023-01-21 13:23:21.970552: step: 400/529, loss: 0.011137199588119984 2023-01-21 13:23:23.102916: step: 404/529, loss: 3.80516066798009e-05 2023-01-21 13:23:24.185769: step: 408/529, loss: 0.0001894474116852507 2023-01-21 13:23:25.300164: step: 412/529, loss: 0.00128173828125 2023-01-21 13:23:26.404755: step: 416/529, loss: 1.125335711549269e-05 2023-01-21 13:23:27.517275: step: 420/529, loss: 0.00017080306133721024 2023-01-21 13:23:28.643101: step: 424/529, loss: 0.00037288665771484375 2023-01-21 13:23:29.781972: step: 428/529, loss: 0.010532189160585403 2023-01-21 13:23:30.906221: step: 432/529, loss: 0.019727326929569244 2023-01-21 13:23:32.013276: step: 436/529, loss: 0.00014085769362282008 2023-01-21 13:23:33.144993: step: 440/529, loss: 0.010293484665453434 2023-01-21 13:23:34.280998: step: 444/529, loss: 1.1539459592313506e-05 2023-01-21 13:23:35.407454: step: 448/529, loss: 6.837844557594508e-05 2023-01-21 13:23:36.537033: step: 452/529, loss: 0.007476424798369408 2023-01-21 13:23:37.662009: step: 456/529, loss: 0.05014381557703018 2023-01-21 13:23:38.763190: step: 460/529, loss: 0.0006255150074139237 2023-01-21 13:23:39.865427: step: 464/529, loss: 0.005975103471428156 2023-01-21 13:23:40.958921: step: 468/529, loss: 0.00031452177790924907 2023-01-21 13:23:42.080892: step: 472/529, loss: 0.0006915092817507684 2023-01-21 13:23:43.177767: step: 476/529, loss: 1.754760705807712e-05 2023-01-21 13:23:44.283010: step: 480/529, loss: 0.027279948815703392 2023-01-21 13:23:45.374031: step: 484/529, loss: 0.00023002624220680445 2023-01-21 13:23:46.521921: step: 488/529, loss: 0.0020013810135424137 2023-01-21 13:23:47.623708: step: 492/529, loss: 0.0008915901416912675 2023-01-21 13:23:48.711931: step: 496/529, loss: 0.0063003539107739925 2023-01-21 13:23:49.848232: step: 500/529, loss: 0.00211086287163198 2023-01-21 13:23:51.009486: step: 504/529, loss: -1.525878360553179e-06 2023-01-21 13:23:52.090704: step: 508/529, loss: 1.5926361811580136e-05 2023-01-21 13:23:53.195672: step: 512/529, loss: 0.00013341903104446828 2023-01-21 13:23:54.318575: step: 516/529, loss: 0.0005537032848224044 2023-01-21 13:23:55.425987: step: 520/529, loss: 0.0002683639759197831 2023-01-21 13:23:56.549899: step: 524/529, loss: 0.0008466661092825234 2023-01-21 13:23:57.704676: step: 528/529, loss: 0.003160190535709262 2023-01-21 13:23:58.830565: step: 532/529, loss: 0.000599288905505091 2023-01-21 13:23:59.950491: step: 536/529, loss: 0.0003606319660320878 2023-01-21 13:24:01.066075: step: 540/529, loss: 0.2402210235595703 2023-01-21 13:24:02.183308: step: 544/529, loss: 0.00023555755615234375 2023-01-21 13:24:03.260748: step: 548/529, loss: 0.00020904540724586695 2023-01-21 13:24:04.381426: step: 552/529, loss: 0.014582633972167969 2023-01-21 13:24:05.482899: step: 556/529, loss: 1.029968279908644e-05 2023-01-21 13:24:06.576768: step: 560/529, loss: 0.0020666124764829874 2023-01-21 13:24:07.694347: step: 564/529, loss: 0.008440589532256126 2023-01-21 13:24:08.802346: step: 568/529, loss: 0.11373557895421982 2023-01-21 13:24:09.917806: step: 572/529, loss: 0.01681804656982422 2023-01-21 13:24:11.008004: step: 576/529, loss: 0.0019423485500738025 2023-01-21 13:24:12.105671: step: 580/529, loss: 0.0007164001581259072 2023-01-21 13:24:13.239776: step: 584/529, loss: 0.0013063430087640882 2023-01-21 13:24:14.320627: step: 588/529, loss: 0.011357021518051624 2023-01-21 13:24:15.440571: step: 592/529, loss: 0.0005343913799151778 2023-01-21 13:24:16.549115: step: 596/529, loss: 9.15527380129788e-06 2023-01-21 13:24:17.639231: step: 600/529, loss: 0.057305146008729935 2023-01-21 13:24:18.724717: step: 604/529, loss: 0.012722969055175781 2023-01-21 13:24:19.845930: step: 608/529, loss: 0.00033626556978560984 2023-01-21 13:24:20.918554: step: 612/529, loss: 0.0031541348434984684 2023-01-21 13:24:22.030694: step: 616/529, loss: 3.757476952159777e-05 2023-01-21 13:24:23.152497: step: 620/529, loss: 0.009058380499482155 2023-01-21 13:24:24.262323: step: 624/529, loss: 0.00424881000071764 2023-01-21 13:24:25.372117: step: 628/529, loss: 0.0002155303955078125 2023-01-21 13:24:26.475368: step: 632/529, loss: 0.0017523766728118062 2023-01-21 13:24:27.574683: step: 636/529, loss: 0.0002902030828408897 2023-01-21 13:24:28.696598: step: 640/529, loss: 0.02160634845495224 2023-01-21 13:24:29.843474: step: 644/529, loss: 0.00945220049470663 2023-01-21 13:24:30.945026: step: 648/529, loss: 0.011438179761171341 2023-01-21 13:24:32.060123: step: 652/529, loss: 0.0010198593372479081 2023-01-21 13:24:33.176811: step: 656/529, loss: 0.00018892288790084422 2023-01-21 13:24:34.303682: step: 660/529, loss: 1.220703143189894e-05 2023-01-21 13:24:35.441163: step: 664/529, loss: 5.3024294174974784e-05 2023-01-21 13:24:36.535420: step: 668/529, loss: 0.00011568069749046117 2023-01-21 13:24:37.623622: step: 672/529, loss: 8.94546537892893e-05 2023-01-21 13:24:38.735324: step: 676/529, loss: 0.0008773327572271228 2023-01-21 13:24:39.858751: step: 680/529, loss: 0.0001535534975118935 2023-01-21 13:24:40.965828: step: 684/529, loss: 3.62396240234375e-05 2023-01-21 13:24:42.080269: step: 688/529, loss: 0.0007985115516930819 2023-01-21 13:24:43.187332: step: 692/529, loss: 9.059906005859375e-05 2023-01-21 13:24:44.279564: step: 696/529, loss: 0.0007797241560183465 2023-01-21 13:24:45.412274: step: 700/529, loss: 0.0002004623383982107 2023-01-21 13:24:46.523633: step: 704/529, loss: 0.0003780364932026714 2023-01-21 13:24:47.649473: step: 708/529, loss: 0.013645363971590996 2023-01-21 13:24:48.750114: step: 712/529, loss: 0.008221722207963467 2023-01-21 13:24:49.885201: step: 716/529, loss: 0.004894256591796875 2023-01-21 13:24:51.016690: step: 720/529, loss: 3.814697606685513e-07 2023-01-21 13:24:52.104017: step: 724/529, loss: 9.393691470904741e-06 2023-01-21 13:24:53.203818: step: 728/529, loss: 1.9454957509879023e-05 2023-01-21 13:24:54.293640: step: 732/529, loss: 0.0007548332214355469 2023-01-21 13:24:55.407759: step: 736/529, loss: 5.817413239128655e-06 2023-01-21 13:24:56.502932: step: 740/529, loss: 1.373290979245212e-05 2023-01-21 13:24:57.589640: step: 744/529, loss: 0.04232082515954971 2023-01-21 13:24:58.747125: step: 748/529, loss: 0.0006826400640420616 2023-01-21 13:24:59.878686: step: 752/529, loss: 0.00011243821063544601 2023-01-21 13:25:00.987288: step: 756/529, loss: 0.00010986327833961695 2023-01-21 13:25:02.088996: step: 760/529, loss: 0.001678466796875 2023-01-21 13:25:03.199669: step: 764/529, loss: 0.0007234573713503778 2023-01-21 13:25:04.315212: step: 768/529, loss: 9.5367431640625e-05 2023-01-21 13:25:05.404086: step: 772/529, loss: 0.00010766983905341476 2023-01-21 13:25:06.534794: step: 776/529, loss: 0.006166649051010609 2023-01-21 13:25:07.632480: step: 780/529, loss: 1.354217511106981e-05 2023-01-21 13:25:08.787680: step: 784/529, loss: 0.01085443515330553 2023-01-21 13:25:09.911327: step: 788/529, loss: 8.678436643094756e-06 2023-01-21 13:25:10.998631: step: 792/529, loss: 0.0002346038818359375 2023-01-21 13:25:12.085855: step: 796/529, loss: 0.0014541626442223787 2023-01-21 13:25:13.187094: step: 800/529, loss: 0.0018613814609125257 2023-01-21 13:25:14.301053: step: 804/529, loss: 0.001175022218376398 2023-01-21 13:25:15.425062: step: 808/529, loss: 0.0007246494060382247 2023-01-21 13:25:16.524415: step: 812/529, loss: 0.0009040356380864978 2023-01-21 13:25:17.637148: step: 816/529, loss: 0.032077983021736145 2023-01-21 13:25:18.749852: step: 820/529, loss: 0.0003749847237486392 2023-01-21 13:25:19.876199: step: 824/529, loss: 0.005182075779885054 2023-01-21 13:25:20.995242: step: 828/529, loss: -1.0871887752728071e-05 2023-01-21 13:25:22.083689: step: 832/529, loss: 0.007000637240707874 2023-01-21 13:25:23.199550: step: 836/529, loss: 0.590520441532135 2023-01-21 13:25:24.332450: step: 840/529, loss: 6.103515625e-05 2023-01-21 13:25:25.435363: step: 844/529, loss: 0.0010389328235760331 2023-01-21 13:25:26.561321: step: 848/529, loss: 5.779266211902723e-05 2023-01-21 13:25:27.654107: step: 852/529, loss: 0.0032056807540357113 2023-01-21 13:25:28.763022: step: 856/529, loss: 0.012347030453383923 2023-01-21 13:25:29.866034: step: 860/529, loss: 4.863739377469756e-06 2023-01-21 13:25:31.007478: step: 864/529, loss: 0.0014041900867596269 2023-01-21 13:25:32.140171: step: 868/529, loss: 0.0015330315800383687 2023-01-21 13:25:33.260540: step: 872/529, loss: 0.003272533416748047 2023-01-21 13:25:34.365989: step: 876/529, loss: 5.53131121705519e-06 2023-01-21 13:25:35.478741: step: 880/529, loss: 0.00414199847728014 2023-01-21 13:25:36.585079: step: 884/529, loss: 0.11855840682983398 2023-01-21 13:25:37.696569: step: 888/529, loss: 1.2683868590102065e-05 2023-01-21 13:25:38.797644: step: 892/529, loss: 0.11132202297449112 2023-01-21 13:25:39.909946: step: 896/529, loss: 1.544952465337701e-05 2023-01-21 13:25:40.997740: step: 900/529, loss: 0.0017951965564861894 2023-01-21 13:25:42.093480: step: 904/529, loss: 2.3746490114717744e-05 2023-01-21 13:25:43.192351: step: 908/529, loss: 0.0007156372303143144 2023-01-21 13:25:44.323879: step: 912/529, loss: 4.701614670921117e-05 2023-01-21 13:25:45.447087: step: 916/529, loss: 9.54627976170741e-05 2023-01-21 13:25:46.547521: step: 920/529, loss: 2.880096508306451e-05 2023-01-21 13:25:47.643704: step: 924/529, loss: 0.00024147033400367945 2023-01-21 13:25:48.740216: step: 928/529, loss: 0.005999756045639515 2023-01-21 13:25:49.838427: step: 932/529, loss: 0.00012702941603492945 2023-01-21 13:25:50.951180: step: 936/529, loss: 0.0007169723394326866 2023-01-21 13:25:52.049677: step: 940/529, loss: 0.0020458223298192024 2023-01-21 13:25:53.173578: step: 944/529, loss: 0.00039119721623137593 2023-01-21 13:25:54.334913: step: 948/529, loss: 0.0009192943689413369 2023-01-21 13:25:55.466833: step: 952/529, loss: 0.08902054280042648 2023-01-21 13:25:56.562239: step: 956/529, loss: -4.768372150465439e-07 2023-01-21 13:25:57.676990: step: 960/529, loss: 9.088515798794106e-05 2023-01-21 13:25:58.777875: step: 964/529, loss: 0.17736896872520447 2023-01-21 13:25:59.912983: step: 968/529, loss: 0.0014201165176928043 2023-01-21 13:26:01.041190: step: 972/529, loss: 0.0027999880257993937 2023-01-21 13:26:02.133224: step: 976/529, loss: 6.294250852079131e-06 2023-01-21 13:26:03.304782: step: 980/529, loss: 0.026045847684144974 2023-01-21 13:26:04.418109: step: 984/529, loss: 0.0005786895635537803 2023-01-21 13:26:05.519683: step: 988/529, loss: 0.00011186600022483617 2023-01-21 13:26:06.665232: step: 992/529, loss: 0.0009332656627520919 2023-01-21 13:26:07.786649: step: 996/529, loss: 3.2138825190486386e-05 2023-01-21 13:26:08.919847: step: 1000/529, loss: 0.007909584790468216 2023-01-21 13:26:10.013450: step: 1004/529, loss: 0.0005084037547931075 2023-01-21 13:26:11.147397: step: 1008/529, loss: 6.103515261202119e-06 2023-01-21 13:26:12.261028: step: 1012/529, loss: 0.0002325058012502268 2023-01-21 13:26:13.391526: step: 1016/529, loss: 9.984970529330894e-05 2023-01-21 13:26:14.503575: step: 1020/529, loss: 0.00112152099609375 2023-01-21 13:26:15.618211: step: 1024/529, loss: 4.673004150390625e-05 2023-01-21 13:26:16.734858: step: 1028/529, loss: 0.00011215209815418348 2023-01-21 13:26:17.853615: step: 1032/529, loss: 0.0017841339576989412 2023-01-21 13:26:18.959337: step: 1036/529, loss: 0.011297762393951416 2023-01-21 13:26:20.093296: step: 1040/529, loss: 0.0012785911094397306 2023-01-21 13:26:21.211349: step: 1044/529, loss: 0.001488971640355885 2023-01-21 13:26:22.332581: step: 1048/529, loss: 3.82423386326991e-05 2023-01-21 13:26:23.424559: step: 1052/529, loss: 0.02686181105673313 2023-01-21 13:26:24.528996: step: 1056/529, loss: 0.002179908799007535 2023-01-21 13:26:25.641700: step: 1060/529, loss: 0.00045108795166015625 2023-01-21 13:26:26.763498: step: 1064/529, loss: 0.018274975940585136 2023-01-21 13:26:27.866201: step: 1068/529, loss: 3.1185150874080136e-05 2023-01-21 13:26:28.968702: step: 1072/529, loss: 0.0016240120166912675 2023-01-21 13:26:30.068497: step: 1076/529, loss: 0.01855173148214817 2023-01-21 13:26:31.218530: step: 1080/529, loss: -3.681182715808973e-05 2023-01-21 13:26:32.344586: step: 1084/529, loss: 3.695487976074219e-05 2023-01-21 13:26:33.470540: step: 1088/529, loss: 0.00036106110201217234 2023-01-21 13:26:34.567834: step: 1092/529, loss: 0.036031533032655716 2023-01-21 13:26:35.662913: step: 1096/529, loss: 0.0006015778053551912 2023-01-21 13:26:36.757339: step: 1100/529, loss: 0.0013172150356695056 2023-01-21 13:26:37.870965: step: 1104/529, loss: 0.01700124889612198 2023-01-21 13:26:38.975372: step: 1108/529, loss: 0.005195140838623047 2023-01-21 13:26:40.075647: step: 1112/529, loss: 0.0018299103248864412 2023-01-21 13:26:41.194060: step: 1116/529, loss: 0.01735048182308674 2023-01-21 13:26:42.335309: step: 1120/529, loss: 0.000347137451171875 2023-01-21 13:26:43.454528: step: 1124/529, loss: 0.01861095428466797 2023-01-21 13:26:44.565834: step: 1128/529, loss: 0.03182811662554741 2023-01-21 13:26:45.657799: step: 1132/529, loss: 0.0003147125244140625 2023-01-21 13:26:46.793916: step: 1136/529, loss: 0.0008786201360635459 2023-01-21 13:26:47.939933: step: 1140/529, loss: 0.0053993226028978825 2023-01-21 13:26:49.031991: step: 1144/529, loss: 0.019384384155273438 2023-01-21 13:26:50.184531: step: 1148/529, loss: 0.02094736136496067 2023-01-21 13:26:51.270865: step: 1152/529, loss: 7.209777686512098e-05 2023-01-21 13:26:52.358400: step: 1156/529, loss: 0.0014679909218102694 2023-01-21 13:26:53.463969: step: 1160/529, loss: 0.0001363277406198904 2023-01-21 13:26:54.558136: step: 1164/529, loss: 8.01086389401462e-06 2023-01-21 13:26:55.687038: step: 1168/529, loss: 0.003501319792121649 2023-01-21 13:26:56.793432: step: 1172/529, loss: 0.0010204315185546875 2023-01-21 13:26:57.926761: step: 1176/529, loss: 0.043367769569158554 2023-01-21 13:26:59.075154: step: 1180/529, loss: 0.0002151489316020161 2023-01-21 13:27:00.169745: step: 1184/529, loss: 0.0004848480166401714 2023-01-21 13:27:01.270599: step: 1188/529, loss: 0.09257078915834427 2023-01-21 13:27:02.357768: step: 1192/529, loss: -8.77380352903856e-06 2023-01-21 13:27:03.517741: step: 1196/529, loss: 0.0006591796991415322 2023-01-21 13:27:04.650972: step: 1200/529, loss: 0.002654838375747204 2023-01-21 13:27:05.781207: step: 1204/529, loss: 0.005988502409309149 2023-01-21 13:27:06.935723: step: 1208/529, loss: 0.0018761635292321444 2023-01-21 13:27:08.073324: step: 1212/529, loss: 0.0011517525417730212 2023-01-21 13:27:09.186964: step: 1216/529, loss: 0.0055709839798510075 2023-01-21 13:27:10.323573: step: 1220/529, loss: 0.06934338063001633 2023-01-21 13:27:11.475299: step: 1224/529, loss: 5.426407005870715e-05 2023-01-21 13:27:12.587075: step: 1228/529, loss: 0.880345344543457 2023-01-21 13:27:13.713684: step: 1232/529, loss: -1.502037139289314e-05 2023-01-21 13:27:14.833639: step: 1236/529, loss: 7.100104994606227e-05 2023-01-21 13:27:15.956592: step: 1240/529, loss: 0.00224912166595459 2023-01-21 13:27:17.058433: step: 1244/529, loss: 0.0007546782144345343 2023-01-21 13:27:18.156188: step: 1248/529, loss: 0.011242485605180264 2023-01-21 13:27:19.267345: step: 1252/529, loss: 0.015579987317323685 2023-01-21 13:27:20.407632: step: 1256/529, loss: 0.06078395992517471 2023-01-21 13:27:21.539772: step: 1260/529, loss: 0.0741397887468338 2023-01-21 13:27:22.656706: step: 1264/529, loss: 0.0013955116737633944 2023-01-21 13:27:23.748827: step: 1268/529, loss: 0.0017473220359534025 2023-01-21 13:27:24.863856: step: 1272/529, loss: 0.0008215904235839844 2023-01-21 13:27:25.983149: step: 1276/529, loss: 7.03811674611643e-05 2023-01-21 13:27:27.110450: step: 1280/529, loss: -1.1777877261920366e-05 2023-01-21 13:27:28.212313: step: 1284/529, loss: 0.45340174436569214 2023-01-21 13:27:29.397702: step: 1288/529, loss: 0.09187754988670349 2023-01-21 13:27:30.523816: step: 1292/529, loss: 0.008336067199707031 2023-01-21 13:27:31.617954: step: 1296/529, loss: 2.689361645025201e-05 2023-01-21 13:27:32.757908: step: 1300/529, loss: 0.00023446083650924265 2023-01-21 13:27:33.900224: step: 1304/529, loss: 0.002177715301513672 2023-01-21 13:27:34.999480: step: 1308/529, loss: 1.411438006471144e-05 2023-01-21 13:27:36.104029: step: 1312/529, loss: 0.0009828567272052169 2023-01-21 13:27:37.215038: step: 1316/529, loss: 0.00015392302884720266 2023-01-21 13:27:38.350289: step: 1320/529, loss: 0.021826744079589844 2023-01-21 13:27:39.476822: step: 1324/529, loss: 0.0007827759254723787 2023-01-21 13:27:40.576531: step: 1328/529, loss: 6.48498553346144e-06 2023-01-21 13:27:41.710165: step: 1332/529, loss: 8.292197890114039e-05 2023-01-21 13:27:42.830726: step: 1336/529, loss: 0.0290861614048481 2023-01-21 13:27:43.937996: step: 1340/529, loss: 0.00013065338134765625 2023-01-21 13:27:45.100459: step: 1344/529, loss: 0.0011649131774902344 2023-01-21 13:27:46.216285: step: 1348/529, loss: 0.020685292780399323 2023-01-21 13:27:47.299904: step: 1352/529, loss: 8.78334030858241e-05 2023-01-21 13:27:48.393860: step: 1356/529, loss: 0.0004375457938294858 2023-01-21 13:27:49.517098: step: 1360/529, loss: 0.007586384192109108 2023-01-21 13:27:50.673058: step: 1364/529, loss: 0.24356335401535034 2023-01-21 13:27:51.813067: step: 1368/529, loss: 8.37326078908518e-05 2023-01-21 13:27:52.935894: step: 1372/529, loss: 7.333755638683215e-05 2023-01-21 13:27:54.062119: step: 1376/529, loss: 6.86645489622606e-06 2023-01-21 13:27:55.183433: step: 1380/529, loss: 6.29425048828125e-05 2023-01-21 13:27:56.306382: step: 1384/529, loss: 0.00018367767916060984 2023-01-21 13:27:57.430575: step: 1388/529, loss: 0.0017717838054522872 2023-01-21 13:27:58.584380: step: 1392/529, loss: 0.0002916336234193295 2023-01-21 13:27:59.692142: step: 1396/529, loss: 0.0019001007312908769 2023-01-21 13:28:00.801067: step: 1400/529, loss: 0.0002880096435546875 2023-01-21 13:28:01.921414: step: 1404/529, loss: 0.05448494106531143 2023-01-21 13:28:03.032782: step: 1408/529, loss: 0.00046625136747024953 2023-01-21 13:28:04.164438: step: 1412/529, loss: 0.00908813439309597 2023-01-21 13:28:05.268700: step: 1416/529, loss: 0.0049460409209132195 2023-01-21 13:28:06.372377: step: 1420/529, loss: 0.0005999565473757684 2023-01-21 13:28:07.475201: step: 1424/529, loss: 0.003921127412468195 2023-01-21 13:28:08.566297: step: 1428/529, loss: 0.0006348371971398592 2023-01-21 13:28:09.678727: step: 1432/529, loss: 0.0014342308277264237 2023-01-21 13:28:10.783865: step: 1436/529, loss: 0.02320881001651287 2023-01-21 13:28:11.882585: step: 1440/529, loss: 0.0009210586431436241 2023-01-21 13:28:13.006224: step: 1444/529, loss: 3.4761429560603574e-05 2023-01-21 13:28:14.098949: step: 1448/529, loss: 0.0005372524028643966 2023-01-21 13:28:15.191890: step: 1452/529, loss: 0.0002694130234885961 2023-01-21 13:28:16.304296: step: 1456/529, loss: 0.00567207345739007 2023-01-21 13:28:17.395850: step: 1460/529, loss: 3.0422210329561494e-05 2023-01-21 13:28:18.509954: step: 1464/529, loss: 0.00018806457228492945 2023-01-21 13:28:19.628733: step: 1468/529, loss: 0.007238960359245539 2023-01-21 13:28:20.761982: step: 1472/529, loss: 7.43865984986769e-06 2023-01-21 13:28:21.895522: step: 1476/529, loss: 0.00917520560324192 2023-01-21 13:28:23.001357: step: 1480/529, loss: 0.0030061721336096525 2023-01-21 13:28:24.113994: step: 1484/529, loss: 0.0005032539484091103 2023-01-21 13:28:25.240584: step: 1488/529, loss: 0.005369377322494984 2023-01-21 13:28:26.323389: step: 1492/529, loss: 6.385446613421664e-05 2023-01-21 13:28:27.464730: step: 1496/529, loss: 0.0016664029099047184 2023-01-21 13:28:28.571196: step: 1500/529, loss: 0.015155792236328125 2023-01-21 13:28:29.708722: step: 1504/529, loss: 0.020765209570527077 2023-01-21 13:28:30.829017: step: 1508/529, loss: 0.000118255615234375 2023-01-21 13:28:31.939557: step: 1512/529, loss: 0.0002330779971089214 2023-01-21 13:28:33.048934: step: 1516/529, loss: 4.463195728021674e-05 2023-01-21 13:28:34.168335: step: 1520/529, loss: 0.0027618408203125 2023-01-21 13:28:35.302703: step: 1524/529, loss: 0.001327419187873602 2023-01-21 13:28:36.426667: step: 1528/529, loss: 0.00013084412785246968 2023-01-21 13:28:37.545734: step: 1532/529, loss: 1.316070574830519e-05 2023-01-21 13:28:38.658958: step: 1536/529, loss: 0.03496236726641655 2023-01-21 13:28:39.781315: step: 1540/529, loss: 6.017684791004285e-05 2023-01-21 13:28:40.902672: step: 1544/529, loss: -0.0 2023-01-21 13:28:42.039634: step: 1548/529, loss: 0.0003089427773375064 2023-01-21 13:28:43.151250: step: 1552/529, loss: 0.0006787300226278603 2023-01-21 13:28:44.251824: step: 1556/529, loss: 0.0032237053383141756 2023-01-21 13:28:45.342720: step: 1560/529, loss: 9.083747863769531e-05 2023-01-21 13:28:46.430699: step: 1564/529, loss: 0.02887878566980362 2023-01-21 13:28:47.532874: step: 1568/529, loss: 0.04644288867712021 2023-01-21 13:28:48.656674: step: 1572/529, loss: 2.28881845032447e-06 2023-01-21 13:28:49.801435: step: 1576/529, loss: -2.28881845032447e-06 2023-01-21 13:28:50.929228: step: 1580/529, loss: 1.621246337890625e-05 2023-01-21 13:28:52.036351: step: 1584/529, loss: 0.0022249221801757812 2023-01-21 13:28:53.130424: step: 1588/529, loss: 2.1648407710017636e-05 2023-01-21 13:28:54.259489: step: 1592/529, loss: -6.10351571594947e-06 2023-01-21 13:28:55.356317: step: 1596/529, loss: 5.626678102998994e-06 2023-01-21 13:28:56.465958: step: 1600/529, loss: 0.0006933212280273438 2023-01-21 13:28:57.584778: step: 1604/529, loss: 0.0010688782203942537 2023-01-21 13:28:58.690829: step: 1608/529, loss: 0.0034814835526049137 2023-01-21 13:28:59.817288: step: 1612/529, loss: 9.155272891803179e-06 2023-01-21 13:29:00.922133: step: 1616/529, loss: 0.0014782429207116365 2023-01-21 13:29:02.092868: step: 1620/529, loss: 1.583099401614163e-05 2023-01-21 13:29:03.228727: step: 1624/529, loss: 0.0001087188720703125 2023-01-21 13:29:04.357561: step: 1628/529, loss: 6.29425039733178e-06 2023-01-21 13:29:05.462765: step: 1632/529, loss: 1.0299681889591739e-05 2023-01-21 13:29:06.563769: step: 1636/529, loss: 0.035803988575935364 2023-01-21 13:29:07.679949: step: 1640/529, loss: 0.004652786068618298 2023-01-21 13:29:08.830792: step: 1644/529, loss: 0.056727029383182526 2023-01-21 13:29:09.954998: step: 1648/529, loss: 2.555847095209174e-05 2023-01-21 13:29:11.027228: step: 1652/529, loss: 1.2683867680607364e-05 2023-01-21 13:29:12.134073: step: 1656/529, loss: 0.0020586014725267887 2023-01-21 13:29:13.255341: step: 1660/529, loss: 0.0009336948860436678 2023-01-21 13:29:14.353521: step: 1664/529, loss: 0.04230918735265732 2023-01-21 13:29:15.504648: step: 1668/529, loss: 0.018568802624940872 2023-01-21 13:29:16.596655: step: 1672/529, loss: -1.9073486328125e-06 2023-01-21 13:29:17.718614: step: 1676/529, loss: 0.01316919457167387 2023-01-21 13:29:18.871101: step: 1680/529, loss: 2.09808349609375e-05 2023-01-21 13:29:19.982739: step: 1684/529, loss: 1.735687328618951e-05 2023-01-21 13:29:21.114627: step: 1688/529, loss: 0.0011392593150958419 2023-01-21 13:29:22.232130: step: 1692/529, loss: 0.00011844635446323082 2023-01-21 13:29:23.346323: step: 1696/529, loss: 2.593994213384576e-05 2023-01-21 13:29:24.468204: step: 1700/529, loss: 0.00018711091252043843 2023-01-21 13:29:25.555323: step: 1704/529, loss: 8.20159948489163e-06 2023-01-21 13:29:26.706547: step: 1708/529, loss: -1.144409225162235e-06 2023-01-21 13:29:27.827595: step: 1712/529, loss: 0.04522743448615074 2023-01-21 13:29:28.945831: step: 1716/529, loss: 3.9672853745287284e-05 2023-01-21 13:29:30.046034: step: 1720/529, loss: 0.03481006622314453 2023-01-21 13:29:31.141220: step: 1724/529, loss: 0.011277103796601295 2023-01-21 13:29:32.243453: step: 1728/529, loss: 0.007256889250129461 2023-01-21 13:29:33.358414: step: 1732/529, loss: 2.517700158932712e-05 2023-01-21 13:29:34.461933: step: 1736/529, loss: 7.314682443393394e-05 2023-01-21 13:29:35.566115: step: 1740/529, loss: 0.05733747407793999 2023-01-21 13:29:36.692821: step: 1744/529, loss: 9.13620024221018e-05 2023-01-21 13:29:37.805359: step: 1748/529, loss: 0.015453720465302467 2023-01-21 13:29:38.953412: step: 1752/529, loss: 0.04392261430621147 2023-01-21 13:29:40.082860: step: 1756/529, loss: 0.021573258563876152 2023-01-21 13:29:41.167929: step: 1760/529, loss: 0.00043702125549316406 2023-01-21 13:29:42.273924: step: 1764/529, loss: 0.0047928811982274055 2023-01-21 13:29:43.411489: step: 1768/529, loss: 0.0014234542613849044 2023-01-21 13:29:44.550706: step: 1772/529, loss: 3.738403393072076e-05 2023-01-21 13:29:45.665004: step: 1776/529, loss: 6.103515625e-05 2023-01-21 13:29:46.792729: step: 1780/529, loss: 0.005189466290175915 2023-01-21 13:29:47.896293: step: 1784/529, loss: 0.00032806396484375 2023-01-21 13:29:48.991893: step: 1788/529, loss: 6.675720669591101e-07 2023-01-21 13:29:50.092264: step: 1792/529, loss: 3.24249276673072e-06 2023-01-21 13:29:51.240713: step: 1796/529, loss: 0.02434062957763672 2023-01-21 13:29:52.386424: step: 1800/529, loss: 0.02499408647418022 2023-01-21 13:29:53.461023: step: 1804/529, loss: 0.011557006277143955 2023-01-21 13:29:54.556780: step: 1808/529, loss: 0.0013374327681958675 2023-01-21 13:29:55.660557: step: 1812/529, loss: 0.009341811761260033 2023-01-21 13:29:56.787284: step: 1816/529, loss: 0.00012111663818359375 2023-01-21 13:29:57.924530: step: 1820/529, loss: 0.000495529209729284 2023-01-21 13:29:59.040912: step: 1824/529, loss: 0.016745759174227715 2023-01-21 13:30:00.121666: step: 1828/529, loss: 0.0001548767031636089 2023-01-21 13:30:01.247390: step: 1832/529, loss: 0.0006437778356485069 2023-01-21 13:30:02.351677: step: 1836/529, loss: 0.03592805936932564 2023-01-21 13:30:03.473858: step: 1840/529, loss: 0.001772117568179965 2023-01-21 13:30:04.610024: step: 1844/529, loss: 0.04893684759736061 2023-01-21 13:30:05.724646: step: 1848/529, loss: 2.86102294921875e-05 2023-01-21 13:30:06.875698: step: 1852/529, loss: 0.003789234207943082 2023-01-21 13:30:07.989601: step: 1856/529, loss: 0.6194444894790649 2023-01-21 13:30:09.092724: step: 1860/529, loss: 0.0020144463051110506 2023-01-21 13:30:10.211928: step: 1864/529, loss: 7.438658940372989e-06 2023-01-21 13:30:11.335863: step: 1868/529, loss: 0.013620566576719284 2023-01-21 13:30:12.419628: step: 1872/529, loss: 0.00036087038461118937 2023-01-21 13:30:13.556510: step: 1876/529, loss: 0.0003662109375 2023-01-21 13:30:14.702584: step: 1880/529, loss: 0.0039619444869458675 2023-01-21 13:30:15.806439: step: 1884/529, loss: 3.42369094141759e-05 2023-01-21 13:30:16.908880: step: 1888/529, loss: 0.0002265930233988911 2023-01-21 13:30:18.045173: step: 1892/529, loss: 0.0004863739013671875 2023-01-21 13:30:19.142929: step: 1896/529, loss: 0.0008038521045818925 2023-01-21 13:30:20.246742: step: 1900/529, loss: 0.010232735425233841 2023-01-21 13:30:21.373562: step: 1904/529, loss: 0.00863275583833456 2023-01-21 13:30:22.519286: step: 1908/529, loss: 0.018400192260742188 2023-01-21 13:30:23.641577: step: 1912/529, loss: 7.696151442360133e-05 2023-01-21 13:30:24.749356: step: 1916/529, loss: 0.0006605148082599044 2023-01-21 13:30:25.851444: step: 1920/529, loss: 0.0014730454422533512 2023-01-21 13:30:26.971487: step: 1924/529, loss: 0.08725529164075851 2023-01-21 13:30:28.090593: step: 1928/529, loss: 0.27439358830451965 2023-01-21 13:30:29.217282: step: 1932/529, loss: 0.00011577607074286789 2023-01-21 13:30:30.345786: step: 1936/529, loss: 0.0632333755493164 2023-01-21 13:30:31.434215: step: 1940/529, loss: 0.00046730044414289296 2023-01-21 13:30:32.548029: step: 1944/529, loss: 0.007697486784309149 2023-01-21 13:30:33.701915: step: 1948/529, loss: 0.21207351982593536 2023-01-21 13:30:34.824365: step: 1952/529, loss: 0.0023055076599121094 2023-01-21 13:30:35.944599: step: 1956/529, loss: 1.3351441339182202e-06 2023-01-21 13:30:37.030595: step: 1960/529, loss: 0.0021831037010997534 2023-01-21 13:30:38.211939: step: 1964/529, loss: 0.00088586809579283 2023-01-21 13:30:39.334117: step: 1968/529, loss: 0.044402312487363815 2023-01-21 13:30:40.438868: step: 1972/529, loss: 0.0005143165471963584 2023-01-21 13:30:41.558510: step: 1976/529, loss: 2.9468536013155244e-05 2023-01-21 13:30:42.695671: step: 1980/529, loss: 0.03940201178193092 2023-01-21 13:30:43.797060: step: 1984/529, loss: 0.00587310828268528 2023-01-21 13:30:44.902430: step: 1988/529, loss: 0.003031921572983265 2023-01-21 13:30:46.011669: step: 1992/529, loss: 0.09160403907299042 2023-01-21 13:30:47.158546: step: 1996/529, loss: 0.0006291389581747353 2023-01-21 13:30:48.334331: step: 2000/529, loss: 0.024505138397216797 2023-01-21 13:30:49.459261: step: 2004/529, loss: 0.00021419525728560984 2023-01-21 13:30:50.583697: step: 2008/529, loss: 0.00011277198791503906 2023-01-21 13:30:51.708318: step: 2012/529, loss: 0.00037384033203125 2023-01-21 13:30:52.842656: step: 2016/529, loss: 0.06208706274628639 2023-01-21 13:30:53.956850: step: 2020/529, loss: 3.261566234868951e-05 2023-01-21 13:30:55.099072: step: 2024/529, loss: 5.5789947509765625e-05 2023-01-21 13:30:56.223972: step: 2028/529, loss: 0.0006503105396404862 2023-01-21 13:30:57.337620: step: 2032/529, loss: 8.60214204294607e-05 2023-01-21 13:30:58.439778: step: 2036/529, loss: 0.0036647438537329435 2023-01-21 13:30:59.579956: step: 2040/529, loss: 0.0003295898495707661 2023-01-21 13:31:00.716665: step: 2044/529, loss: 0.0007380485767498612 2023-01-21 13:31:01.820856: step: 2048/529, loss: 0.0006364822620525956 2023-01-21 13:31:02.953046: step: 2052/529, loss: 0.0015435218811035156 2023-01-21 13:31:04.080814: step: 2056/529, loss: 6.45637555862777e-05 2023-01-21 13:31:05.192959: step: 2060/529, loss: 0.0014858245849609375 2023-01-21 13:31:06.363627: step: 2064/529, loss: 0.0010208130115643144 2023-01-21 13:31:07.484959: step: 2068/529, loss: 0.004155635833740234 2023-01-21 13:31:08.595198: step: 2072/529, loss: 0.00348663330078125 2023-01-21 13:31:09.715252: step: 2076/529, loss: 0.0008179664728231728 2023-01-21 13:31:10.846652: step: 2080/529, loss: 0.0011617183918133378 2023-01-21 13:31:11.961889: step: 2084/529, loss: 6.299019150901586e-05 2023-01-21 13:31:13.098764: step: 2088/529, loss: 0.0008985519525595009 2023-01-21 13:31:14.204140: step: 2092/529, loss: 0.012169456109404564 2023-01-21 13:31:15.312171: step: 2096/529, loss: 0.011667441576719284 2023-01-21 13:31:16.435811: step: 2100/529, loss: 0.0008366584661416709 2023-01-21 13:31:17.549185: step: 2104/529, loss: 0.0024226189125329256 2023-01-21 13:31:18.654293: step: 2108/529, loss: 0.004815768916159868 2023-01-21 13:31:19.813342: step: 2112/529, loss: 0.0008327484247274697 2023-01-21 13:31:20.931529: step: 2116/529, loss: 0.003198051592335105 ================================================== Loss: 0.017 -------------------- Dev: {'event': {'p': 0.5875763747454175, 'r': 0.7683089214380826, 'f1': 0.6658972879399884}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Test: {'event': {'p': 0.6284153005464481, 'r': 0.7787810383747178, 'f1': 0.6955645161290323}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Chinese: {'event': {'p': 0.5632183908045977, 'r': 0.9074074074074074, 'f1': 0.6950354609929078}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Korean: {'event': {'p': 0.5869565217391305, 'r': 0.42857142857142855, 'f1': 0.49541284403669733}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} 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': 28} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 29 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:32:01.446765: step: 4/529, loss: 0.004092597868293524 2023-01-21 13:32:02.554936: step: 8/529, loss: 0.00031185150146484375 2023-01-21 13:32:03.666264: step: 12/529, loss: 1.621246337890625e-05 2023-01-21 13:32:04.798841: step: 16/529, loss: 3.108978125965223e-05 2023-01-21 13:32:05.883505: step: 20/529, loss: 0.01533365249633789 2023-01-21 13:32:06.978707: step: 24/529, loss: 0.0003192901494912803 2023-01-21 13:32:08.093141: step: 28/529, loss: 0.0003437042178120464 2023-01-21 13:32:09.187744: step: 32/529, loss: 0.00042705534724518657 2023-01-21 13:32:10.295185: step: 36/529, loss: 0.0002023696870310232 2023-01-21 13:32:11.408547: step: 40/529, loss: 7.152557373046875e-05 2023-01-21 13:32:12.524733: step: 44/529, loss: 0.010753917507827282 2023-01-21 13:32:13.687725: step: 48/529, loss: 0.019957160577178 2023-01-21 13:32:14.786139: step: 52/529, loss: 3.814697322468419e-07 2023-01-21 13:32:15.915596: step: 56/529, loss: 0.0009521484607830644 2023-01-21 13:32:17.070937: step: 60/529, loss: 0.00011396408081054688 2023-01-21 13:32:18.210178: step: 64/529, loss: 0.050718214362859726 2023-01-21 13:32:19.302064: step: 68/529, loss: 0.009755993261933327 2023-01-21 13:32:20.412174: step: 72/529, loss: 7.629394076502649e-07 2023-01-21 13:32:21.518231: step: 76/529, loss: 4.320144944358617e-05 2023-01-21 13:32:22.626518: step: 80/529, loss: 0.00041990281897597015 2023-01-21 13:32:23.738963: step: 84/529, loss: 0.01795501820743084 2023-01-21 13:32:24.861920: step: 88/529, loss: 6.561279587913305e-05 2023-01-21 13:32:25.987013: step: 92/529, loss: 0.0010890960693359375 2023-01-21 13:32:27.115468: step: 96/529, loss: 0.00036525726318359375 2023-01-21 13:32:28.207315: step: 100/529, loss: -2.098083541568485e-06 2023-01-21 13:32:29.357174: step: 104/529, loss: 0.0006759643438272178 2023-01-21 13:32:30.492370: step: 108/529, loss: 7.629394076502649e-06 2023-01-21 13:32:31.591212: step: 112/529, loss: 2.4127961296471767e-05 2023-01-21 13:32:32.698891: step: 116/529, loss: 0.0004878044128417969 2023-01-21 13:32:33.789453: step: 120/529, loss: 0.00014495849609375 2023-01-21 13:32:34.927036: step: 124/529, loss: 0.0016555309994146228 2023-01-21 13:32:36.116767: step: 128/529, loss: 0.009090328589081764 2023-01-21 13:32:37.212507: step: 132/529, loss: 0.00012474060349632055 2023-01-21 13:32:38.299585: step: 136/529, loss: 1.3542176020564511e-05 2023-01-21 13:32:39.404470: step: 140/529, loss: 0.0008403778192587197 2023-01-21 13:32:40.518506: step: 144/529, loss: 0.008500290103256702 2023-01-21 13:32:41.645267: step: 148/529, loss: 0.0007547378772869706 2023-01-21 13:32:42.774743: step: 152/529, loss: 0.0004295349062886089 2023-01-21 13:32:43.887168: step: 156/529, loss: 0.001307678292505443 2023-01-21 13:32:44.997221: step: 160/529, loss: 0.0062122344970703125 2023-01-21 13:32:46.090254: step: 164/529, loss: 0.004171943757683039 2023-01-21 13:32:47.223536: step: 168/529, loss: 0.001203346299007535 2023-01-21 13:32:48.343852: step: 172/529, loss: 0.0015020370483398438 2023-01-21 13:32:49.450698: step: 176/529, loss: 0.0003014564572367817 2023-01-21 13:32:50.550119: step: 180/529, loss: 0.00488014193251729 2023-01-21 13:32:51.649431: step: 184/529, loss: 0.00313491839915514 2023-01-21 13:32:52.767900: step: 188/529, loss: 0.002962303115054965 2023-01-21 13:32:53.890809: step: 192/529, loss: 0.05255451053380966 2023-01-21 13:32:54.999603: step: 196/529, loss: 3.3855438232421875e-05 2023-01-21 13:32:56.100719: step: 200/529, loss: 7.62939453125e-06 2023-01-21 13:32:57.221782: step: 204/529, loss: 0.00600929232314229 2023-01-21 13:32:58.330951: step: 208/529, loss: 2.8419495720299892e-05 2023-01-21 13:32:59.494303: step: 212/529, loss: 0.0008700728067196906 2023-01-21 13:33:00.621292: step: 216/529, loss: 0.0007225990411825478 2023-01-21 13:33:01.714289: step: 220/529, loss: 0.05460834875702858 2023-01-21 13:33:02.832888: step: 224/529, loss: 0.0004034042649436742 2023-01-21 13:33:03.939297: step: 228/529, loss: 0.010929679498076439 2023-01-21 13:33:05.038711: step: 232/529, loss: 2.765655517578125e-05 2023-01-21 13:33:06.141030: step: 236/529, loss: 0.0027210237458348274 2023-01-21 13:33:07.228877: step: 240/529, loss: -4.00543194700731e-06 2023-01-21 13:33:08.330503: step: 244/529, loss: 0.0014099121326580644 2023-01-21 13:33:09.474084: step: 248/529, loss: -2.098083541568485e-06 2023-01-21 13:33:10.606692: step: 252/529, loss: 0.04202747717499733 2023-01-21 13:33:11.706511: step: 256/529, loss: 0.016085434705018997 2023-01-21 13:33:12.830882: step: 260/529, loss: 0.0030867578461766243 2023-01-21 13:33:13.945205: step: 264/529, loss: 3.051757857974735e-06 2023-01-21 13:33:15.060439: step: 268/529, loss: 0.015006543137133121 2023-01-21 13:33:16.159881: step: 272/529, loss: 0.0004784584161825478 2023-01-21 13:33:17.292242: step: 276/529, loss: 2.765655517578125e-05 2023-01-21 13:33:18.396631: step: 280/529, loss: 3.80516066798009e-05 2023-01-21 13:33:19.503576: step: 284/529, loss: 0.0004985809209756553 2023-01-21 13:33:20.633678: step: 288/529, loss: 0.0572628490626812 2023-01-21 13:33:21.750229: step: 292/529, loss: 0.0008211136446334422 2023-01-21 13:33:22.844041: step: 296/529, loss: 0.002113437745720148 2023-01-21 13:33:23.971733: step: 300/529, loss: 0.5935726165771484 2023-01-21 13:33:25.097309: step: 304/529, loss: 0.002004432724788785 2023-01-21 13:33:26.192548: step: 308/529, loss: 7.61032133596018e-05 2023-01-21 13:33:27.309116: step: 312/529, loss: 0.006684112828224897 2023-01-21 13:33:28.416062: step: 316/529, loss: 0.0004756927373819053 2023-01-21 13:33:29.545590: step: 320/529, loss: 0.0001390457182424143 2023-01-21 13:33:30.683312: step: 324/529, loss: 0.7323215007781982 2023-01-21 13:33:31.796359: step: 328/529, loss: 0.0002038002130575478 2023-01-21 13:33:32.938223: step: 332/529, loss: 1.5258790426742053e-06 2023-01-21 13:33:34.060424: step: 336/529, loss: 0.0012258529895916581 2023-01-21 13:33:35.192784: step: 340/529, loss: 0.00409665098413825 2023-01-21 13:33:36.316844: step: 344/529, loss: 0.0008167267078533769 2023-01-21 13:33:37.456272: step: 348/529, loss: 0.10731334239244461 2023-01-21 13:33:38.571912: step: 352/529, loss: 0.07027512043714523 2023-01-21 13:33:39.695857: step: 356/529, loss: 1.1842942237854004 2023-01-21 13:33:40.816940: step: 360/529, loss: 0.0015718459617346525 2023-01-21 13:33:41.900219: step: 364/529, loss: 0.022486304864287376 2023-01-21 13:33:42.993963: step: 368/529, loss: 0.0006540298345498741 2023-01-21 13:33:44.078953: step: 372/529, loss: 2.7751922971219756e-05 2023-01-21 13:33:45.184222: step: 376/529, loss: 0.0028486251831054688 2023-01-21 13:33:46.309793: step: 380/529, loss: 0.00137500767596066 2023-01-21 13:33:47.402211: step: 384/529, loss: 4.47750098828692e-05 2023-01-21 13:33:48.519946: step: 388/529, loss: 0.00015735626220703125 2023-01-21 13:33:49.620527: step: 392/529, loss: 0.002513885498046875 2023-01-21 13:33:50.739424: step: 396/529, loss: 0.010579205118119717 2023-01-21 13:33:51.824821: step: 400/529, loss: 2.1266938347253017e-05 2023-01-21 13:33:52.909241: step: 404/529, loss: 0.0045219422318041325 2023-01-21 13:33:54.027442: step: 408/529, loss: 5.569458153331652e-05 2023-01-21 13:33:55.122692: step: 412/529, loss: 5.9604644775390625e-05 2023-01-21 13:33:56.235756: step: 416/529, loss: 0.011366844177246094 2023-01-21 13:33:57.347169: step: 420/529, loss: 0.00616378802806139 2023-01-21 13:33:58.439312: step: 424/529, loss: 0.045984648168087006 2023-01-21 13:33:59.554915: step: 428/529, loss: 0.004430866800248623 2023-01-21 13:34:00.671259: step: 432/529, loss: 0.0026763915084302425 2023-01-21 13:34:01.809647: step: 436/529, loss: 0.0008678436279296875 2023-01-21 13:34:02.915765: step: 440/529, loss: 0.07574901729822159 2023-01-21 13:34:04.026803: step: 444/529, loss: 0.0001638412504689768 2023-01-21 13:34:05.166214: step: 448/529, loss: 8.46386028570123e-05 2023-01-21 13:34:06.255928: step: 452/529, loss: 1.621246337890625e-05 2023-01-21 13:34:07.368375: step: 456/529, loss: 0.0003631591971497983 2023-01-21 13:34:08.485040: step: 460/529, loss: 0.000377655029296875 2023-01-21 13:34:09.597073: step: 464/529, loss: 0.0018795013893395662 2023-01-21 13:34:10.711246: step: 468/529, loss: 0.004645919892936945 2023-01-21 13:34:11.801404: step: 472/529, loss: 3.1089784897631034e-05 2023-01-21 13:34:12.902944: step: 476/529, loss: 0.007003116421401501 2023-01-21 13:34:14.011297: step: 480/529, loss: 0.008932781405746937 2023-01-21 13:34:15.132856: step: 484/529, loss: 0.0029113772325217724 2023-01-21 13:34:16.243907: step: 488/529, loss: 0.017908478155732155 2023-01-21 13:34:17.329598: step: 492/529, loss: 0.002963257022202015 2023-01-21 13:34:18.404657: step: 496/529, loss: 0.3411533236503601 2023-01-21 13:34:19.531269: step: 500/529, loss: 0.0005902767297811806 2023-01-21 13:34:20.652126: step: 504/529, loss: 0.0001161575346486643 2023-01-21 13:34:21.779166: step: 508/529, loss: -2.6702882678364404e-06 2023-01-21 13:34:22.862381: step: 512/529, loss: 0.04043960198760033 2023-01-21 13:34:23.978984: step: 516/529, loss: 0.0002285957452841103 2023-01-21 13:34:25.165116: step: 520/529, loss: 0.11449813842773438 2023-01-21 13:34:26.281523: step: 524/529, loss: 0.0010884285438805819 2023-01-21 13:34:27.400507: step: 528/529, loss: 0.0011043548583984375 2023-01-21 13:34:28.508989: step: 532/529, loss: 0.010164165869355202 2023-01-21 13:34:29.630288: step: 536/529, loss: 0.00032272341195493937 2023-01-21 13:34:30.770850: step: 540/529, loss: 3.366470627952367e-05 2023-01-21 13:34:31.891721: step: 544/529, loss: 0.0005010605091229081 2023-01-21 13:34:33.023953: step: 548/529, loss: 0.00010848045349121094 2023-01-21 13:34:34.136446: step: 552/529, loss: 3.0136110581224784e-05 2023-01-21 13:34:35.254579: step: 556/529, loss: 1.201629675051663e-05 2023-01-21 13:34:36.339246: step: 560/529, loss: 1.296997106692288e-05 2023-01-21 13:34:37.446284: step: 564/529, loss: 0.018535614013671875 2023-01-21 13:34:38.549807: step: 568/529, loss: 0.004475736990571022 2023-01-21 13:34:39.686772: step: 572/529, loss: 0.03180637210607529 2023-01-21 13:34:40.836408: step: 576/529, loss: 0.0005739211919717491 2023-01-21 13:34:41.933817: step: 580/529, loss: 0.0015921592712402344 2023-01-21 13:34:43.050482: step: 584/529, loss: 0.011671066284179688 2023-01-21 13:34:44.182173: step: 588/529, loss: 0.00987854041159153 2023-01-21 13:34:45.280977: step: 592/529, loss: 0.003669929690659046 2023-01-21 13:34:46.427779: step: 596/529, loss: 1.5974044799804688e-05 2023-01-21 13:34:47.533810: step: 600/529, loss: 0.024297334253787994 2023-01-21 13:34:48.647278: step: 604/529, loss: 0.003869438311085105 2023-01-21 13:34:49.746556: step: 608/529, loss: 5.855560448253527e-05 2023-01-21 13:34:50.906244: step: 612/529, loss: 0.0010390282841399312 2023-01-21 13:34:52.035252: step: 616/529, loss: 0.0002941131533589214 2023-01-21 13:34:53.163465: step: 620/529, loss: 0.04735575243830681 2023-01-21 13:34:54.240233: step: 624/529, loss: -1.3351439065445447e-06 2023-01-21 13:34:55.351177: step: 628/529, loss: 0.00014495849609375 2023-01-21 13:34:56.451561: step: 632/529, loss: 0.0012647629482671618 2023-01-21 13:34:57.583543: step: 636/529, loss: 0.0002586364862509072 2023-01-21 13:34:58.729876: step: 640/529, loss: 9.5367431640625e-07 2023-01-21 13:34:59.867369: step: 644/529, loss: -2.09808349609375e-05 2023-01-21 13:35:00.985523: step: 648/529, loss: 0.024466706439852715 2023-01-21 13:35:02.125384: step: 652/529, loss: 0.002342700958251953 2023-01-21 13:35:03.246903: step: 656/529, loss: 0.0024061203002929688 2023-01-21 13:35:04.336604: step: 660/529, loss: 1.3732910701946821e-05 2023-01-21 13:35:05.431003: step: 664/529, loss: 0.000324058550177142 2023-01-21 13:35:06.548094: step: 668/529, loss: 4.158019874012098e-05 2023-01-21 13:35:07.715998: step: 672/529, loss: 0.005105209536850452 2023-01-21 13:35:08.858375: step: 676/529, loss: 0.0008537292596884072 2023-01-21 13:35:09.993133: step: 680/529, loss: 3.738403393072076e-05 2023-01-21 13:35:11.109919: step: 684/529, loss: 0.010645199567079544 2023-01-21 13:35:12.194232: step: 688/529, loss: 8.144378807628527e-05 2023-01-21 13:35:13.335265: step: 692/529, loss: 6.895065598655492e-05 2023-01-21 13:35:14.423680: step: 696/529, loss: 0.02693800814449787 2023-01-21 13:35:15.562067: step: 700/529, loss: 0.0011053085327148438 2023-01-21 13:35:16.663297: step: 704/529, loss: 4.100799742445815e-06 2023-01-21 13:35:17.752420: step: 708/529, loss: 2.6226043701171875e-06 2023-01-21 13:35:18.875650: step: 712/529, loss: 9.765624417923391e-05 2023-01-21 13:35:19.986886: step: 716/529, loss: 4.606247239280492e-05 2023-01-21 13:35:21.106604: step: 720/529, loss: 0.014852523803710938 2023-01-21 13:35:22.224949: step: 724/529, loss: 0.0006097317091189325 2023-01-21 13:35:23.307932: step: 728/529, loss: 6.675720101156912e-07 2023-01-21 13:35:24.427892: step: 732/529, loss: 0.014451599679887295 2023-01-21 13:35:25.564857: step: 736/529, loss: 0.022385217249393463 2023-01-21 13:35:26.685637: step: 740/529, loss: 2.6702882678364404e-06 2023-01-21 13:35:27.810102: step: 744/529, loss: 0.00013256072998046875 2023-01-21 13:35:28.900847: step: 748/529, loss: 2.021789623540826e-05 2023-01-21 13:35:30.027554: step: 752/529, loss: 0.0003832101938314736 2023-01-21 13:35:31.137686: step: 756/529, loss: 0.0033186913933604956 2023-01-21 13:35:32.238086: step: 760/529, loss: 0.0003864288446493447 2023-01-21 13:35:33.319384: step: 764/529, loss: 7.667542377021164e-05 2023-01-21 13:35:34.423962: step: 768/529, loss: 8.58306884765625e-06 2023-01-21 13:35:35.587397: step: 772/529, loss: 0.008380889892578125 2023-01-21 13:35:36.677201: step: 776/529, loss: 0.01411972101777792 2023-01-21 13:35:37.812296: step: 780/529, loss: -4.95910626341356e-06 2023-01-21 13:35:38.920020: step: 784/529, loss: 3.919601658708416e-05 2023-01-21 13:35:40.069355: step: 788/529, loss: 7.152557373046875e-06 2023-01-21 13:35:41.196952: step: 792/529, loss: 0.00010452270362293348 2023-01-21 13:35:42.304684: step: 796/529, loss: 0.015958786010742188 2023-01-21 13:35:43.435997: step: 800/529, loss: 2.746581958490424e-05 2023-01-21 13:35:44.559025: step: 804/529, loss: 0.00023328961106017232 2023-01-21 13:35:45.646787: step: 808/529, loss: 5.4788590205134824e-05 2023-01-21 13:35:46.734552: step: 812/529, loss: 2.19345088225964e-06 2023-01-21 13:35:47.850773: step: 816/529, loss: -1.735687328618951e-05 2023-01-21 13:35:48.942379: step: 820/529, loss: 0.00315780658274889 2023-01-21 13:35:50.059360: step: 824/529, loss: 5.636215428239666e-05 2023-01-21 13:35:51.202551: step: 828/529, loss: 0.00043354035005904734 2023-01-21 13:35:52.333941: step: 832/529, loss: 0.01212158240377903 2023-01-21 13:35:53.471247: step: 836/529, loss: 0.008632087148725986 2023-01-21 13:35:54.579482: step: 840/529, loss: 0.00041236879769712687 2023-01-21 13:35:55.692108: step: 844/529, loss: 0.0009363174322061241 2023-01-21 13:35:56.832221: step: 848/529, loss: 0.005694389343261719 2023-01-21 13:35:57.952712: step: 852/529, loss: 0.013215827755630016 2023-01-21 13:35:59.061276: step: 856/529, loss: 0.012385082431137562 2023-01-21 13:36:00.187450: step: 860/529, loss: 0.0038232803344726562 2023-01-21 13:36:01.302203: step: 864/529, loss: 0.029938984662294388 2023-01-21 13:36:02.414658: step: 868/529, loss: 1.831054760259576e-05 2023-01-21 13:36:03.517687: step: 872/529, loss: 1.0633468264131807e-05 2023-01-21 13:36:04.690512: step: 876/529, loss: 0.014688300900161266 2023-01-21 13:36:05.812867: step: 880/529, loss: 0.0007961273076944053 2023-01-21 13:36:06.938418: step: 884/529, loss: 0.0016300200950354338 2023-01-21 13:36:08.099131: step: 888/529, loss: 4.3821339204441756e-05 2023-01-21 13:36:09.215686: step: 892/529, loss: 1.773834264895413e-05 2023-01-21 13:36:10.341200: step: 896/529, loss: 0.00011291504779364914 2023-01-21 13:36:11.455419: step: 900/529, loss: 0.0004173278866801411 2023-01-21 13:36:12.562257: step: 904/529, loss: 0.008144665509462357 2023-01-21 13:36:13.677229: step: 908/529, loss: 0.004697227850556374 2023-01-21 13:36:14.786487: step: 912/529, loss: 0.0015292167663574219 2023-01-21 13:36:15.905729: step: 916/529, loss: 0.01612091064453125 2023-01-21 13:36:17.014643: step: 920/529, loss: 0.007676315493881702 2023-01-21 13:36:18.102421: step: 924/529, loss: 0.00025177001953125 2023-01-21 13:36:19.186600: step: 928/529, loss: 0.03468160703778267 2023-01-21 13:36:20.319988: step: 932/529, loss: 8.01086389401462e-06 2023-01-21 13:36:21.412784: step: 936/529, loss: 0.043299295008182526 2023-01-21 13:36:22.541188: step: 940/529, loss: 0.0011086465092375875 2023-01-21 13:36:23.665743: step: 944/529, loss: 0.005763053894042969 2023-01-21 13:36:24.800799: step: 948/529, loss: 0.0007173538324423134 2023-01-21 13:36:25.898289: step: 952/529, loss: 0.01083984412252903 2023-01-21 13:36:27.031812: step: 956/529, loss: 0.00032453538733534515 2023-01-21 13:36:28.128124: step: 960/529, loss: 0.0036390305031090975 2023-01-21 13:36:29.218590: step: 964/529, loss: 8.535385859431699e-05 2023-01-21 13:36:30.334279: step: 968/529, loss: 0.00025119780912064016 2023-01-21 13:36:31.506755: step: 972/529, loss: 0.0005382538074627519 2023-01-21 13:36:32.641726: step: 976/529, loss: 0.0005075454828329384 2023-01-21 13:36:33.765307: step: 980/529, loss: 5.512237476068549e-05 2023-01-21 13:36:34.860863: step: 984/529, loss: 5.91278057981981e-06 2023-01-21 13:36:35.984016: step: 988/529, loss: 0.00013828277587890625 2023-01-21 13:36:37.103033: step: 992/529, loss: 0.00013628005399368703 2023-01-21 13:36:38.182082: step: 996/529, loss: -8.010864803509321e-06 2023-01-21 13:36:39.294093: step: 1000/529, loss: 3.108978125965223e-05 2023-01-21 13:36:40.444202: step: 1004/529, loss: 0.00890426617115736 2023-01-21 13:36:41.570639: step: 1008/529, loss: -4.00543194700731e-06 2023-01-21 13:36:42.721655: step: 1012/529, loss: 5.9127810345671605e-06 2023-01-21 13:36:43.842997: step: 1016/529, loss: 3.7670135498046875e-05 2023-01-21 13:36:44.954996: step: 1020/529, loss: 0.05157947540283203 2023-01-21 13:36:46.072611: step: 1024/529, loss: 2.1934511096333154e-06 2023-01-21 13:36:47.185679: step: 1028/529, loss: 7.04765334376134e-05 2023-01-21 13:36:48.309258: step: 1032/529, loss: 0.0005767822149209678 2023-01-21 13:36:49.442718: step: 1036/529, loss: 0.01827411726117134 2023-01-21 13:36:50.541817: step: 1040/529, loss: 6.647109694313258e-05 2023-01-21 13:36:51.683701: step: 1044/529, loss: 0.002139949705451727 2023-01-21 13:36:52.794104: step: 1048/529, loss: 0.0008876801002770662 2023-01-21 13:36:53.915705: step: 1052/529, loss: 0.0016753196250647306 2023-01-21 13:36:55.031164: step: 1056/529, loss: 0.017768096178770065 2023-01-21 13:36:56.111833: step: 1060/529, loss: 1.9168854123563506e-05 2023-01-21 13:36:57.220836: step: 1064/529, loss: 0.0011323929065838456 2023-01-21 13:36:58.331459: step: 1068/529, loss: 0.00023193360539153218 2023-01-21 13:36:59.465902: step: 1072/529, loss: 0.007983589544892311 2023-01-21 13:37:00.573688: step: 1076/529, loss: -1.9073486328125e-06 2023-01-21 13:37:01.705189: step: 1080/529, loss: 0.0003346920129843056 2023-01-21 13:37:02.855844: step: 1084/529, loss: 0.006165123078972101 2023-01-21 13:37:03.957135: step: 1088/529, loss: 3.185272362316027e-05 2023-01-21 13:37:05.041191: step: 1092/529, loss: 0.00026836394681595266 2023-01-21 13:37:06.147463: step: 1096/529, loss: 0.010147953405976295 2023-01-21 13:37:07.261242: step: 1100/529, loss: 0.0009092330583371222 2023-01-21 13:37:08.391980: step: 1104/529, loss: 0.04119310528039932 2023-01-21 13:37:09.513169: step: 1108/529, loss: 0.0002062797429971397 2023-01-21 13:37:10.608061: step: 1112/529, loss: 1.8119812921213452e-06 2023-01-21 13:37:11.735681: step: 1116/529, loss: 0.006128692999482155 2023-01-21 13:37:12.839024: step: 1120/529, loss: 0.001466941786929965 2023-01-21 13:37:13.966224: step: 1124/529, loss: 2.47955322265625e-05 2023-01-21 13:37:15.090738: step: 1128/529, loss: 3.471374657237902e-05 2023-01-21 13:37:16.228618: step: 1132/529, loss: 0.0001370430109091103 2023-01-21 13:37:17.348107: step: 1136/529, loss: 0.0009796142112463713 2023-01-21 13:37:18.442932: step: 1140/529, loss: -9.536742027194123e-07 2023-01-21 13:37:19.544519: step: 1144/529, loss: 0.0003751754993572831 2023-01-21 13:37:20.673849: step: 1148/529, loss: 0.053568124771118164 2023-01-21 13:37:21.805490: step: 1152/529, loss: 0.00016880035400390625 2023-01-21 13:37:22.932278: step: 1156/529, loss: 8.392333984375e-05 2023-01-21 13:37:24.027292: step: 1160/529, loss: 0.0073226927779614925 2023-01-21 13:37:25.168374: step: 1164/529, loss: 0.0010486602550372481 2023-01-21 13:37:26.274478: step: 1168/529, loss: 5.054473876953125e-05 2023-01-21 13:37:27.394379: step: 1172/529, loss: 0.0002512931823730469 2023-01-21 13:37:28.512985: step: 1176/529, loss: 4.825591895496473e-05 2023-01-21 13:37:29.597111: step: 1180/529, loss: 0.0023343085777014494 2023-01-21 13:37:30.748928: step: 1184/529, loss: 0.0002460479736328125 2023-01-21 13:37:31.835284: step: 1188/529, loss: 0.000404548627557233 2023-01-21 13:37:32.969488: step: 1192/529, loss: -1.220703143189894e-05 2023-01-21 13:37:34.044983: step: 1196/529, loss: 0.00011239051673328504 2023-01-21 13:37:35.186217: step: 1200/529, loss: 0.0005481719854287803 2023-01-21 13:37:36.321024: step: 1204/529, loss: 5.14984139954322e-06 2023-01-21 13:37:37.457217: step: 1208/529, loss: 0.05054473876953125 2023-01-21 13:37:38.587356: step: 1212/529, loss: 1.811981201171875e-05 2023-01-21 13:37:39.721319: step: 1216/529, loss: 0.005728295538574457 2023-01-21 13:37:40.812024: step: 1220/529, loss: 1.8024446035269648e-05 2023-01-21 13:37:41.958218: step: 1224/529, loss: 0.01786670833826065 2023-01-21 13:37:43.060086: step: 1228/529, loss: -1.382828031637473e-06 2023-01-21 13:37:44.184616: step: 1232/529, loss: 9.860992577159777e-05 2023-01-21 13:37:45.317306: step: 1236/529, loss: 0.0050325398333370686 2023-01-21 13:37:46.446270: step: 1240/529, loss: 1.52587890625e-05 2023-01-21 13:37:47.560730: step: 1244/529, loss: 0.0020725249778479338 2023-01-21 13:37:48.683679: step: 1248/529, loss: 0.007106971461325884 2023-01-21 13:37:49.794675: step: 1252/529, loss: 3.871917579090223e-05 2023-01-21 13:37:50.945229: step: 1256/529, loss: 0.0026372908614575863 2023-01-21 13:37:52.067951: step: 1260/529, loss: 0.0071929930709302425 2023-01-21 13:37:53.182923: step: 1264/529, loss: 1.6880036127986386e-05 2023-01-21 13:37:54.285293: step: 1268/529, loss: 0.00011920928955078125 2023-01-21 13:37:55.375527: step: 1272/529, loss: 9.13620024221018e-05 2023-01-21 13:37:56.508657: step: 1276/529, loss: 0.016804886981844902 2023-01-21 13:37:57.620505: step: 1280/529, loss: 0.0008592605590820312 2023-01-21 13:37:58.718213: step: 1284/529, loss: 0.0005680083995684981 2023-01-21 13:37:59.801051: step: 1288/529, loss: 0.00018558502779342234 2023-01-21 13:38:00.932845: step: 1292/529, loss: 2.2737367544323206e-13 2023-01-21 13:38:02.079277: step: 1296/529, loss: 0.00027647020760923624 2023-01-21 13:38:03.171949: step: 1300/529, loss: 1.3446808225126006e-05 2023-01-21 13:38:04.324561: step: 1304/529, loss: 0.00505905132740736 2023-01-21 13:38:05.462198: step: 1308/529, loss: 0.1079990416765213 2023-01-21 13:38:06.585805: step: 1312/529, loss: 0.06154899671673775 2023-01-21 13:38:07.705493: step: 1316/529, loss: 6.389618647517636e-06 2023-01-21 13:38:08.814356: step: 1320/529, loss: 2.098083541568485e-06 2023-01-21 13:38:09.927281: step: 1324/529, loss: 1.010894811770413e-05 2023-01-21 13:38:11.054317: step: 1328/529, loss: 0.0015863417647778988 2023-01-21 13:38:12.182654: step: 1332/529, loss: 0.004235458094626665 2023-01-21 13:38:13.280527: step: 1336/529, loss: 5.674362546415068e-05 2023-01-21 13:38:14.429585: step: 1340/529, loss: 2.517700158932712e-05 2023-01-21 13:38:15.570436: step: 1344/529, loss: 1.3256072634248994e-05 2023-01-21 13:38:16.648973: step: 1348/529, loss: 0.5052580833435059 2023-01-21 13:38:17.810742: step: 1352/529, loss: 9.117127046920359e-05 2023-01-21 13:38:18.939880: step: 1356/529, loss: 0.0016935348976403475 2023-01-21 13:38:20.035498: step: 1360/529, loss: 0.504020094871521 2023-01-21 13:38:21.180567: step: 1364/529, loss: 0.00012311936006881297 2023-01-21 13:38:22.299267: step: 1368/529, loss: 0.0006533623090945184 2023-01-21 13:38:23.412121: step: 1372/529, loss: 3.0994415283203125e-05 2023-01-21 13:38:24.513457: step: 1376/529, loss: 0.01027364656329155 2023-01-21 13:38:25.616318: step: 1380/529, loss: -2.0503998712229077e-06 2023-01-21 13:38:26.747353: step: 1384/529, loss: 4.415512375999242e-05 2023-01-21 13:38:27.851734: step: 1388/529, loss: 0.00037231447640806437 2023-01-21 13:38:28.982425: step: 1392/529, loss: 0.00029757022275589406 2023-01-21 13:38:30.096861: step: 1396/529, loss: 0.009863471612334251 2023-01-21 13:38:31.198065: step: 1400/529, loss: 0.00014219284639693797 2023-01-21 13:38:32.322146: step: 1404/529, loss: 0.011015797033905983 2023-01-21 13:38:33.430115: step: 1408/529, loss: 0.00020713807316496968 2023-01-21 13:38:34.566508: step: 1412/529, loss: 0.0011715890141204 2023-01-21 13:38:35.669504: step: 1416/529, loss: 0.006308651063591242 2023-01-21 13:38:36.786677: step: 1420/529, loss: 0.000431251508416608 2023-01-21 13:38:37.881884: step: 1424/529, loss: 0.03197136148810387 2023-01-21 13:38:38.989158: step: 1428/529, loss: 0.0047660828568041325 2023-01-21 13:38:40.089855: step: 1432/529, loss: 4.57763690064894e-06 2023-01-21 13:38:41.223165: step: 1436/529, loss: 4.386902219266631e-06 2023-01-21 13:38:42.323229: step: 1440/529, loss: 5.760193016612902e-05 2023-01-21 13:38:43.430723: step: 1444/529, loss: 0.013804435729980469 2023-01-21 13:38:44.538392: step: 1448/529, loss: 0.03641214221715927 2023-01-21 13:38:45.663583: step: 1452/529, loss: 0.019589995965361595 2023-01-21 13:38:46.806871: step: 1456/529, loss: 0.00010566711716819555 2023-01-21 13:38:47.944380: step: 1460/529, loss: 9.670257713878527e-05 2023-01-21 13:38:49.058203: step: 1464/529, loss: -6.675720669591101e-07 2023-01-21 13:38:50.212523: step: 1468/529, loss: 0.0007474899175576866 2023-01-21 13:38:51.331423: step: 1472/529, loss: 0.02275843545794487 2023-01-21 13:38:52.414659: step: 1476/529, loss: 0.0007833481067791581 2023-01-21 13:38:53.526337: step: 1480/529, loss: 0.00023517609224654734 2023-01-21 13:38:54.624317: step: 1484/529, loss: 0.0009361267439089715 2023-01-21 13:38:55.753303: step: 1488/529, loss: 0.007314491551369429 2023-01-21 13:38:56.890998: step: 1492/529, loss: 0.0011762619251385331 2023-01-21 13:38:58.014088: step: 1496/529, loss: 0.00039615630521439016 2023-01-21 13:38:59.144883: step: 1500/529, loss: 0.02231578901410103 2023-01-21 13:39:00.288580: step: 1504/529, loss: 0.02380080334842205 2023-01-21 13:39:01.397790: step: 1508/529, loss: 0.0009593963623046875 2023-01-21 13:39:02.506199: step: 1512/529, loss: 0.00014925003051757812 2023-01-21 13:39:03.610327: step: 1516/529, loss: 7.452964928233996e-05 2023-01-21 13:39:04.729683: step: 1520/529, loss: 0.003545093582943082 2023-01-21 13:39:05.865964: step: 1524/529, loss: 0.0012932777171954513 2023-01-21 13:39:06.985895: step: 1528/529, loss: 0.00015916825213935226 2023-01-21 13:39:08.108787: step: 1532/529, loss: 2.102851794916205e-05 2023-01-21 13:39:09.210278: step: 1536/529, loss: 0.018183136358857155 2023-01-21 13:39:10.310075: step: 1540/529, loss: 4.982948303222656e-05 2023-01-21 13:39:11.415013: step: 1544/529, loss: 0.007026863284409046 2023-01-21 13:39:12.557698: step: 1548/529, loss: 0.00039997100248001516 2023-01-21 13:39:13.664639: step: 1552/529, loss: -1.1444089977885596e-06 2023-01-21 13:39:14.767521: step: 1556/529, loss: 7.209778414107859e-05 2023-01-21 13:39:15.868483: step: 1560/529, loss: 0.0002583026944193989 2023-01-21 13:39:16.984393: step: 1564/529, loss: 0.007990170270204544 2023-01-21 13:39:18.107898: step: 1568/529, loss: 2.803802453854587e-05 2023-01-21 13:39:19.277604: step: 1572/529, loss: 0.0016742706065997481 2023-01-21 13:39:20.414915: step: 1576/529, loss: 0.00016789436631370336 2023-01-21 13:39:21.514984: step: 1580/529, loss: 0.0010194778442382812 2023-01-21 13:39:22.639589: step: 1584/529, loss: 0.0024636269081383944 2023-01-21 13:39:23.756563: step: 1588/529, loss: 0.0002288818359375 2023-01-21 13:39:24.854910: step: 1592/529, loss: 0.0005058288807049394 2023-01-21 13:39:25.999067: step: 1596/529, loss: 0.0007925033569335938 2023-01-21 13:39:27.115209: step: 1600/529, loss: 0.0012141228653490543 2023-01-21 13:39:28.236422: step: 1604/529, loss: 0.002918243408203125 2023-01-21 13:39:29.360654: step: 1608/529, loss: 0.00102748884819448 2023-01-21 13:39:30.500771: step: 1612/529, loss: 0.0005796432378701866 2023-01-21 13:39:31.611051: step: 1616/529, loss: 0.0001220703125 2023-01-21 13:39:32.717340: step: 1620/529, loss: 2.136230432370212e-05 2023-01-21 13:39:33.821561: step: 1624/529, loss: 0.007133865728974342 2023-01-21 13:39:34.937186: step: 1628/529, loss: 0.003608131315559149 2023-01-21 13:39:36.086395: step: 1632/529, loss: 0.011096049100160599 2023-01-21 13:39:37.175930: step: 1636/529, loss: 0.0022530558053404093 2023-01-21 13:39:38.299506: step: 1640/529, loss: 9.860992577159777e-05 2023-01-21 13:39:39.409897: step: 1644/529, loss: 7.05719003235572e-06 2023-01-21 13:39:40.541745: step: 1648/529, loss: 0.003171730088070035 2023-01-21 13:39:41.646331: step: 1652/529, loss: 7.43865966796875e-05 2023-01-21 13:39:42.781703: step: 1656/529, loss: 0.02402954176068306 2023-01-21 13:39:43.901231: step: 1660/529, loss: 0.00013294219388626516 2023-01-21 13:39:45.015734: step: 1664/529, loss: 0.011955262161791325 2023-01-21 13:39:46.141712: step: 1668/529, loss: 0.00034580231294967234 2023-01-21 13:39:47.263417: step: 1672/529, loss: 0.00046443939208984375 2023-01-21 13:39:48.378079: step: 1676/529, loss: 0.0020887376740574837 2023-01-21 13:39:49.490459: step: 1680/529, loss: 0.0001638412504689768 2023-01-21 13:39:50.652279: step: 1684/529, loss: 0.01572876051068306 2023-01-21 13:39:51.756698: step: 1688/529, loss: 5.350113133317791e-05 2023-01-21 13:39:52.882998: step: 1692/529, loss: 0.01955118216574192 2023-01-21 13:39:54.001136: step: 1696/529, loss: 0.00024518967256881297 2023-01-21 13:39:55.099621: step: 1700/529, loss: 8.98361176950857e-05 2023-01-21 13:39:56.190038: step: 1704/529, loss: 0.019835758954286575 2023-01-21 13:39:57.272266: step: 1708/529, loss: 8.01086389401462e-06 2023-01-21 13:39:58.389615: step: 1712/529, loss: 0.0008781433571130037 2023-01-21 13:39:59.519782: step: 1716/529, loss: 7.152557373046875e-06 2023-01-21 13:40:00.643536: step: 1720/529, loss: 0.002100372454151511 2023-01-21 13:40:01.783034: step: 1724/529, loss: 0.06157030910253525 2023-01-21 13:40:02.878625: step: 1728/529, loss: 0.010883521288633347 2023-01-21 13:40:04.000365: step: 1732/529, loss: 0.00036258698673918843 2023-01-21 13:40:05.116500: step: 1736/529, loss: 7.43865984986769e-06 2023-01-21 13:40:06.213059: step: 1740/529, loss: 0.006790733430534601 2023-01-21 13:40:07.342849: step: 1744/529, loss: 0.00015544892812613398 2023-01-21 13:40:08.443253: step: 1748/529, loss: 0.0012887954944744706 2023-01-21 13:40:09.563814: step: 1752/529, loss: 3.204345557605848e-05 2023-01-21 13:40:10.674684: step: 1756/529, loss: 0.03918743133544922 2023-01-21 13:40:11.801511: step: 1760/529, loss: 0.0003551483096089214 2023-01-21 13:40:12.909205: step: 1764/529, loss: 2.1266938347253017e-05 2023-01-21 13:40:14.010434: step: 1768/529, loss: 3.871917579090223e-05 2023-01-21 13:40:15.109406: step: 1772/529, loss: 1.029968279908644e-05 2023-01-21 13:40:16.249957: step: 1776/529, loss: 0.0028073787689208984 2023-01-21 13:40:17.363950: step: 1780/529, loss: 9.136199514614418e-05 2023-01-21 13:40:18.461779: step: 1784/529, loss: 0.0005895138019695878 2023-01-21 13:40:19.577879: step: 1788/529, loss: 1.831054760259576e-05 2023-01-21 13:40:20.731084: step: 1792/529, loss: 4.9400332500226796e-05 2023-01-21 13:40:21.830609: step: 1796/529, loss: 0.014212418347597122 2023-01-21 13:40:22.974754: step: 1800/529, loss: 2.0980833141948096e-06 2023-01-21 13:40:24.097912: step: 1804/529, loss: 0.046585846692323685 2023-01-21 13:40:25.219005: step: 1808/529, loss: -3.52859501617786e-06 2023-01-21 13:40:26.323714: step: 1812/529, loss: 0.0023492814507335424 2023-01-21 13:40:27.449274: step: 1816/529, loss: 0.0005411148304119706 2023-01-21 13:40:28.560108: step: 1820/529, loss: 0.0026319504249840975 2023-01-21 13:40:29.693482: step: 1824/529, loss: 0.008322334848344326 2023-01-21 13:40:30.869689: step: 1828/529, loss: 0.0011900902027264237 2023-01-21 13:40:31.989377: step: 1832/529, loss: 0.00018825530423782766 2023-01-21 13:40:33.093266: step: 1836/529, loss: 9.72747784544481e-06 2023-01-21 13:40:34.212574: step: 1840/529, loss: 2.079009937006049e-05 2023-01-21 13:40:35.303687: step: 1844/529, loss: 0.05358171835541725 2023-01-21 13:40:36.414084: step: 1848/529, loss: 0.0012691498268395662 2023-01-21 13:40:37.538041: step: 1852/529, loss: 0.009786320850253105 2023-01-21 13:40:38.651256: step: 1856/529, loss: 6.980896432651207e-05 2023-01-21 13:40:39.763268: step: 1860/529, loss: 3.051757857974735e-06 2023-01-21 13:40:40.880823: step: 1864/529, loss: 2.880096508306451e-05 2023-01-21 13:40:41.961304: step: 1868/529, loss: 0.0014727592933923006 2023-01-21 13:40:43.057094: step: 1872/529, loss: 0.0025319100823253393 2023-01-21 13:40:44.165974: step: 1876/529, loss: 0.0006879806751385331 2023-01-21 13:40:45.290525: step: 1880/529, loss: 0.03494777902960777 2023-01-21 13:40:46.415851: step: 1884/529, loss: 7.99179106252268e-05 2023-01-21 13:40:47.525011: step: 1888/529, loss: 0.0007623672718182206 2023-01-21 13:40:48.624492: step: 1892/529, loss: 0.014723301865160465 2023-01-21 13:40:49.738374: step: 1896/529, loss: 0.0006964683998376131 2023-01-21 13:40:50.877374: step: 1900/529, loss: 0.0009338379022665322 2023-01-21 13:40:52.011325: step: 1904/529, loss: 0.004417228978127241 2023-01-21 13:40:53.131063: step: 1908/529, loss: 4.596710277837701e-05 2023-01-21 13:40:54.231411: step: 1912/529, loss: 6.69479341013357e-05 2023-01-21 13:40:55.335390: step: 1916/529, loss: 0.0020542144775390625 2023-01-21 13:40:56.441585: step: 1920/529, loss: 0.00020294189744163305 2023-01-21 13:40:57.535908: step: 1924/529, loss: 0.018357563763856888 2023-01-21 13:40:58.655772: step: 1928/529, loss: 6.31332368357107e-05 2023-01-21 13:40:59.752093: step: 1932/529, loss: 0.0002918243408203125 2023-01-21 13:41:00.849504: step: 1936/529, loss: 0.30141276121139526 2023-01-21 13:41:01.940224: step: 1940/529, loss: 0.0002953529474325478 2023-01-21 13:41:03.075370: step: 1944/529, loss: -1.316070574830519e-05 2023-01-21 13:41:04.189049: step: 1948/529, loss: 0.06830983608961105 2023-01-21 13:41:05.338226: step: 1952/529, loss: 0.004886818118393421 2023-01-21 13:41:06.453789: step: 1956/529, loss: 0.023650072515010834 2023-01-21 13:41:07.613480: step: 1960/529, loss: 0.0025791169609874487 2023-01-21 13:41:08.712903: step: 1964/529, loss: 0.008010578341782093 2023-01-21 13:41:09.824354: step: 1968/529, loss: 8.58306884765625e-06 2023-01-21 13:41:10.956593: step: 1972/529, loss: 0.008647155947983265 2023-01-21 13:41:12.083618: step: 1976/529, loss: 0.013877678662538528 2023-01-21 13:41:13.193539: step: 1980/529, loss: 1.239776611328125e-05 2023-01-21 13:41:14.319476: step: 1984/529, loss: 0.033011436462402344 2023-01-21 13:41:15.464465: step: 1988/529, loss: 0.004666232969611883 2023-01-21 13:41:16.599055: step: 1992/529, loss: 0.03419189527630806 2023-01-21 13:41:17.723008: step: 1996/529, loss: 0.0012063980102539062 2023-01-21 13:41:18.855667: step: 2000/529, loss: 0.0019620894454419613 2023-01-21 13:41:19.981437: step: 2004/529, loss: 0.003364181611686945 2023-01-21 13:41:21.087962: step: 2008/529, loss: 0.0037262917030602694 2023-01-21 13:41:22.231449: step: 2012/529, loss: 0.06978455185890198 2023-01-21 13:41:23.340528: step: 2016/529, loss: 0.1108541488647461 2023-01-21 13:41:24.442250: step: 2020/529, loss: 0.0027518272399902344 2023-01-21 13:41:25.572588: step: 2024/529, loss: -2.9802322387695312e-05 2023-01-21 13:41:26.719490: step: 2028/529, loss: 0.29665735363960266 2023-01-21 13:41:27.816366: step: 2032/529, loss: -3.8146970382513246e-07 2023-01-21 13:41:28.908481: step: 2036/529, loss: 0.0005783081287518144 2023-01-21 13:41:30.006291: step: 2040/529, loss: 2.708435022213962e-05 2023-01-21 13:41:31.148062: step: 2044/529, loss: 3.795623706537299e-05 2023-01-21 13:41:32.270005: step: 2048/529, loss: 0.00013809202937409282 2023-01-21 13:41:33.373204: step: 2052/529, loss: -1.182556115963962e-05 2023-01-21 13:41:34.542193: step: 2056/529, loss: 0.0034593583550304174 2023-01-21 13:41:35.642570: step: 2060/529, loss: 2.670288040462765e-06 2023-01-21 13:41:36.751200: step: 2064/529, loss: 1.964569128176663e-05 2023-01-21 13:41:37.879078: step: 2068/529, loss: 0.00012283325486350805 2023-01-21 13:41:39.002244: step: 2072/529, loss: 6.46591215627268e-05 2023-01-21 13:41:40.138752: step: 2076/529, loss: -4.95910626341356e-06 2023-01-21 13:41:41.222101: step: 2080/529, loss: 0.00048465729923918843 2023-01-21 13:41:42.339951: step: 2084/529, loss: 6.580352783203125e-05 2023-01-21 13:41:43.459630: step: 2088/529, loss: 0.0009780883556231856 2023-01-21 13:41:44.549608: step: 2092/529, loss: 0.02933654747903347 2023-01-21 13:41:45.639243: step: 2096/529, loss: 7.057190487103071e-06 2023-01-21 13:41:46.755493: step: 2100/529, loss: 0.0002346038818359375 2023-01-21 13:41:47.861935: step: 2104/529, loss: 0.005618095397949219 2023-01-21 13:41:48.975585: step: 2108/529, loss: 0.018789293244481087 2023-01-21 13:41:50.071551: step: 2112/529, loss: 3.0326846172101796e-05 2023-01-21 13:41:51.208547: step: 2116/529, loss: 0.013628196902573109 ================================================== Loss: 0.015 -------------------- Dev: {'event': {'p': 0.5617342130065975, 'r': 0.7936085219707057, 'f1': 0.6578366445916115}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Test: {'event': {'p': 0.6246042514699232, 'r': 0.7793453724604966, 'f1': 0.6934471503891539}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Chinese: {'event': {'p': 0.5494505494505495, 'r': 0.9259259259259259, 'f1': 0.6896551724137931}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Korean: {'event': {'p': 0.5873015873015873, 'r': 0.5873015873015873, 'f1': 0.5873015873015873}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Russian: {'event': {'p': 0.4782608695652174, 'r': 0.6111111111111112, 'f1': 0.5365853658536586}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.574384236453202, 'r': 0.7762982689747004, 'f1': 0.6602491506228766}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Chinese: {'event': {'p': 0.592640136927685, 'r': 0.7816027088036117, 'f1': 0.6741299586274032}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Chinese: {'event': {'p': 0.6046511627906976, 'r': 0.9629629629629629, 'f1': 0.7428571428571429}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} -------------------- Eng Dev for Korean: {'event': {'p': 0.5833333333333334, 'r': 0.7643142476697736, 'f1': 0.661671469740634}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Eng Test for Korean: {'event': {'p': 0.634571297569922, 'r': 0.781038374717833, 'f1': 0.7002276751834049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Sample Korean: {'event': {'p': 0.6774193548387096, 'r': 0.6666666666666666, 'f1': 0.6719999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} -------------------- Eng Dev for Russian: {'event': {'p': 0.6099893730074389, 'r': 0.7643142476697736, 'f1': 0.6784869976359338}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.6238532110091743, 'r': 0.7291196388261851, 'f1': 0.6723913609159511}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.6046511627906976, 'r': 0.7222222222222222, 'f1': 0.6582278481012659}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4}