| { | |
| "best_metric": 0.6143654584884644, | |
| "best_model_checkpoint": "bert_base_lda_50_v1_cola/checkpoint-136", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 306, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.0722399950027466, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6197, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6198203563690186, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6997, | |
| "eval_samples_per_second": 1490.583, | |
| "eval_steps_per_second": 7.146, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8135132789611816, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6125, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.617854118347168, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6932, | |
| "eval_samples_per_second": 1504.688, | |
| "eval_steps_per_second": 7.213, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.1037389039993286, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6104, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6171332001686096, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.693, | |
| "eval_samples_per_second": 1505.018, | |
| "eval_steps_per_second": 7.215, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4333645105361938, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6101, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6143654584884644, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6884, | |
| "eval_samples_per_second": 1514.999, | |
| "eval_steps_per_second": 7.263, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.863571286201477, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5982, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6144019961357117, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6975, | |
| "eval_samples_per_second": 1495.254, | |
| "eval_steps_per_second": 7.168, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.4660353660583496, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5682, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6874400973320007, | |
| "eval_loss": 0.6281394362449646, | |
| "eval_matthews_correlation": 0.05263079744232175, | |
| "eval_runtime": 0.6831, | |
| "eval_samples_per_second": 1526.915, | |
| "eval_steps_per_second": 7.32, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.6993749141693115, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5152, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6567593216896057, | |
| "eval_loss": 0.6455426812171936, | |
| "eval_matthews_correlation": 0.10027467906851985, | |
| "eval_runtime": 0.6842, | |
| "eval_samples_per_second": 1524.455, | |
| "eval_steps_per_second": 7.308, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.5550780296325684, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.4552, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6634707450866699, | |
| "eval_loss": 0.6827359795570374, | |
| "eval_matthews_correlation": 0.09702614755535717, | |
| "eval_runtime": 0.6854, | |
| "eval_samples_per_second": 1521.657, | |
| "eval_steps_per_second": 7.295, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.5828475952148438, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.4108, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6529242396354675, | |
| "eval_loss": 0.675087034702301, | |
| "eval_matthews_correlation": 0.10114728369225008, | |
| "eval_runtime": 0.6887, | |
| "eval_samples_per_second": 1514.398, | |
| "eval_steps_per_second": 7.26, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 306, | |
| "total_flos": 1.012438185472512e+16, | |
| "train_loss": 0.5555953512004778, | |
| "train_runtime": 156.1294, | |
| "train_samples_per_second": 2738.434, | |
| "train_steps_per_second": 10.888 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.012438185472512e+16, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |