| { | |
| "best_metric": 0.5128712871287129, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-594", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 594, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.6258772015571594, | |
| "learning_rate": 1.171336966673164e-05, | |
| "loss": 0.303, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4910891089108911, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.28316354751586914, | |
| "eval_mcc": -0.09980217586956908, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.9281, | |
| "eval_samples_per_second": 544.113, | |
| "eval_steps_per_second": 17.239, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.44756296277046204, | |
| "learning_rate": 1.0249198458390185e-05, | |
| "loss": 0.2807, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5128712871287129, | |
| "eval_f1": 0.13986013986013984, | |
| "eval_loss": 0.2766987085342407, | |
| "eval_mcc": 0.047945551909639166, | |
| "eval_precision": 0.5882352941176471, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 0.9353, | |
| "eval_samples_per_second": 539.951, | |
| "eval_steps_per_second": 17.107, | |
| "step": 594 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2673, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 1461402531960.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.33782488262757904, | |
| "learning_rate": 1.3177540875073095e-05, | |
| "num_train_epochs": 9, | |
| "temperature": 26 | |
| } | |
| } | |