| { | |
| "best_metric": 0.81, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-24/checkpoint-688", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 688, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.6114831566810608, | |
| "learning_rate": 0.00012438553278352026, | |
| "loss": 0.4032, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.812864, | |
| "eval_BAC": 0.736, | |
| "eval_MCC": 0.47260700716368276, | |
| "eval_PR_AUC": 0.7744742146180674, | |
| "eval_accuracy": 0.736, | |
| "eval_loss": 0.3765888214111328, | |
| "eval_runtime": 6.2583, | |
| "eval_samples_per_second": 239.681, | |
| "eval_steps_per_second": 7.51, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8825994729995728, | |
| "learning_rate": 0.00011056491802979578, | |
| "loss": 0.3836, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.834104888888889, | |
| "eval_BAC": 0.7606666666666666, | |
| "eval_MCC": 0.5584922194756476, | |
| "eval_PR_AUC": 0.7802314251570694, | |
| "eval_accuracy": 0.7606666666666667, | |
| "eval_loss": 0.36870983242988586, | |
| "eval_runtime": 6.277, | |
| "eval_samples_per_second": 238.968, | |
| "eval_steps_per_second": 7.488, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.0203838348388672, | |
| "learning_rate": 9.67443032760713e-05, | |
| "loss": 0.3769, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.8587253333333335, | |
| "eval_BAC": 0.804, | |
| "eval_MCC": 0.6132578536355836, | |
| "eval_PR_AUC": 0.8153312043691034, | |
| "eval_accuracy": 0.804, | |
| "eval_loss": 0.35289254784584045, | |
| "eval_runtime": 6.2324, | |
| "eval_samples_per_second": 240.677, | |
| "eval_steps_per_second": 7.541, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.1677234172821045, | |
| "learning_rate": 8.292368852234684e-05, | |
| "loss": 0.3677, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8672444444444444, | |
| "eval_BAC": 0.81, | |
| "eval_MCC": 0.6290847145741534, | |
| "eval_PR_AUC": 0.8293327638808494, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.3481285870075226, | |
| "eval_runtime": 6.2093, | |
| "eval_samples_per_second": 241.573, | |
| "eval_steps_per_second": 7.569, | |
| "step": 688 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1720, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1801510920000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.605164320881552, | |
| "learning_rate": 0.00013820614753724472, | |
| "num_train_epochs": 10, | |
| "temperature": 9 | |
| } | |
| } | |