| { | |
| "best_metric": 0.8146666666666667, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-4/checkpoint-1204", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 1376, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.065812110900879, | |
| "learning_rate": 8.050072712118835e-05, | |
| "loss": 0.5092, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.8145119999999999, | |
| "eval_BAC": 0.73, | |
| "eval_MCC": 0.4608302423279951, | |
| "eval_PR_AUC": 0.7678112276484446, | |
| "eval_accuracy": 0.73, | |
| "eval_loss": 0.4519721269607544, | |
| "eval_runtime": 6.3105, | |
| "eval_samples_per_second": 237.697, | |
| "eval_steps_per_second": 7.448, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.7611809968948364, | |
| "learning_rate": 6.900062324673287e-05, | |
| "loss": 0.4743, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.8335324444444443, | |
| "eval_BAC": 0.7613333333333334, | |
| "eval_MCC": 0.553275105402295, | |
| "eval_PR_AUC": 0.7822581410349926, | |
| "eval_accuracy": 0.7613333333333333, | |
| "eval_loss": 0.4407980144023895, | |
| "eval_runtime": 6.2949, | |
| "eval_samples_per_second": 238.289, | |
| "eval_steps_per_second": 7.466, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.037668466567993, | |
| "learning_rate": 5.750051937227739e-05, | |
| "loss": 0.4621, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.8552204444444444, | |
| "eval_BAC": 0.7933333333333333, | |
| "eval_MCC": 0.5994394641665934, | |
| "eval_PR_AUC": 0.8137656581601591, | |
| "eval_accuracy": 0.7933333333333333, | |
| "eval_loss": 0.4244216978549957, | |
| "eval_runtime": 6.1875, | |
| "eval_samples_per_second": 242.424, | |
| "eval_steps_per_second": 7.596, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.76123309135437, | |
| "learning_rate": 4.600041549782191e-05, | |
| "loss": 0.4529, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8632462222222221, | |
| "eval_BAC": 0.7953333333333333, | |
| "eval_MCC": 0.596541705530537, | |
| "eval_PR_AUC": 0.8260242182570992, | |
| "eval_accuracy": 0.7953333333333333, | |
| "eval_loss": 0.4147915542125702, | |
| "eval_runtime": 6.222, | |
| "eval_samples_per_second": 241.079, | |
| "eval_steps_per_second": 7.554, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.963057041168213, | |
| "learning_rate": 3.450031162336643e-05, | |
| "loss": 0.4457, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_AUC": 0.8692337777777777, | |
| "eval_BAC": 0.81, | |
| "eval_MCC": 0.6276847092784394, | |
| "eval_PR_AUC": 0.8299132636472669, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.40833592414855957, | |
| "eval_runtime": 6.2503, | |
| "eval_samples_per_second": 239.989, | |
| "eval_steps_per_second": 7.52, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.1222100257873535, | |
| "learning_rate": 2.3000207748910956e-05, | |
| "loss": 0.4422, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_AUC": 0.8663128888888889, | |
| "eval_BAC": 0.81, | |
| "eval_MCC": 0.633318117479026, | |
| "eval_PR_AUC": 0.8122133069255465, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.4122408926486969, | |
| "eval_runtime": 6.2825, | |
| "eval_samples_per_second": 238.758, | |
| "eval_steps_per_second": 7.481, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.754328966140747, | |
| "learning_rate": 1.1500103874455478e-05, | |
| "loss": 0.4406, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_AUC": 0.8696568888888889, | |
| "eval_BAC": 0.8146666666666667, | |
| "eval_MCC": 0.6396251179192641, | |
| "eval_PR_AUC": 0.8147222095880156, | |
| "eval_accuracy": 0.8146666666666667, | |
| "eval_loss": 0.4100054204463959, | |
| "eval_runtime": 6.247, | |
| "eval_samples_per_second": 240.113, | |
| "eval_steps_per_second": 7.524, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.56061589717865, | |
| "learning_rate": 0.0, | |
| "loss": 0.4369, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_AUC": 0.8715377777777777, | |
| "eval_BAC": 0.8146666666666667, | |
| "eval_MCC": 0.6375468401321781, | |
| "eval_PR_AUC": 0.8228418225735421, | |
| "eval_accuracy": 0.8146666666666667, | |
| "eval_loss": 0.4085315763950348, | |
| "eval_runtime": 6.2249, | |
| "eval_samples_per_second": 240.968, | |
| "eval_steps_per_second": 7.55, | |
| "step": 1376 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1376, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3603021840000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7709874476999774, | |
| "learning_rate": 9.200083099564382e-05, | |
| "num_train_epochs": 8, | |
| "temperature": 3 | |
| } | |
| } | |