| { | |
| "best_metric": 0.6896741696836141, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-7940", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 7940, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.4442338943481445, | |
| "learning_rate": 4.032885899938537e-05, | |
| "loss": 0.6141, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6636234849677318, | |
| "eval_f1": 0.3630402384500745, | |
| "eval_loss": 0.6011130809783936, | |
| "eval_mcc": 0.16570379533034763, | |
| "eval_precision": 0.4921212121212121, | |
| "eval_recall": 0.28760330578512394, | |
| "eval_runtime": 23.3169, | |
| "eval_samples_per_second": 544.926, | |
| "eval_steps_per_second": 17.069, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.0027925968170166, | |
| "learning_rate": 3.360738249948781e-05, | |
| "loss": 0.5959, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6799937037620022, | |
| "eval_f1": 0.26447178002894356, | |
| "eval_loss": 0.5925843119621277, | |
| "eval_mcc": 0.1656860671477833, | |
| "eval_precision": 0.5653518948182521, | |
| "eval_recall": 0.17260920897284535, | |
| "eval_runtime": 23.4673, | |
| "eval_samples_per_second": 541.433, | |
| "eval_steps_per_second": 16.96, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.0462489128112793, | |
| "learning_rate": 2.688590599959025e-05, | |
| "loss": 0.5867, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6858177239099638, | |
| "eval_f1": 0.31877133105802047, | |
| "eval_loss": 0.578091561794281, | |
| "eval_mcc": 0.19615574783247955, | |
| "eval_precision": 0.5747692307692308, | |
| "eval_recall": 0.220543093270366, | |
| "eval_runtime": 23.3603, | |
| "eval_samples_per_second": 543.914, | |
| "eval_steps_per_second": 17.037, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.00968861579895, | |
| "learning_rate": 2.0164429499692687e-05, | |
| "loss": 0.5802, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6873917834094129, | |
| "eval_f1": 0.40036231884057977, | |
| "eval_loss": 0.5741773843765259, | |
| "eval_mcc": 0.22635184605022737, | |
| "eval_precision": 0.5550439514441189, | |
| "eval_recall": 0.3131050767414404, | |
| "eval_runtime": 23.4393, | |
| "eval_samples_per_second": 542.082, | |
| "eval_steps_per_second": 16.98, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.98496413230896, | |
| "learning_rate": 1.3442952999795124e-05, | |
| "loss": 0.577, | |
| "step": 7940 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6896741696836141, | |
| "eval_f1": 0.338311797281423, | |
| "eval_loss": 0.5701454281806946, | |
| "eval_mcc": 0.21128704823195582, | |
| "eval_precision": 0.5846867749419954, | |
| "eval_recall": 0.23801652892561984, | |
| "eval_runtime": 23.5382, | |
| "eval_samples_per_second": 539.803, | |
| "eval_steps_per_second": 16.909, | |
| "step": 7940 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 19544470381500.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9830537825786697, | |
| "learning_rate": 4.7050335499282935e-05, | |
| "num_train_epochs": 7, | |
| "temperature": 17 | |
| } | |
| } | |