| { | |
| "best_metric": 0.7580550673696543, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-32/checkpoint-1070", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1070, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.148311614990234, | |
| "learning_rate": 0.00021207436837213607, | |
| "loss": 0.5606, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7129466900995899, | |
| "eval_f1": 0.5040485829959513, | |
| "eval_loss": 0.5154260396957397, | |
| "eval_mcc": 0.3157140452463625, | |
| "eval_precision": 0.594272076372315, | |
| "eval_recall": 0.437609841827768, | |
| "eval_runtime": 3.126, | |
| "eval_samples_per_second": 546.06, | |
| "eval_steps_per_second": 17.274, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.1550638675689697, | |
| "learning_rate": 0.0001817780300332595, | |
| "loss": 0.4995, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7393087287639133, | |
| "eval_f1": 0.5071982281284606, | |
| "eval_loss": 0.4903852641582489, | |
| "eval_mcc": 0.36859649869532995, | |
| "eval_precision": 0.6856287425149701, | |
| "eval_recall": 0.4024604569420035, | |
| "eval_runtime": 3.1521, | |
| "eval_samples_per_second": 541.536, | |
| "eval_steps_per_second": 17.131, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.5803661346435547, | |
| "learning_rate": 0.00015148169169438292, | |
| "loss": 0.479, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7574692442882249, | |
| "eval_f1": 0.541019955654102, | |
| "eval_loss": 0.48610907793045044, | |
| "eval_mcc": 0.41710213681200975, | |
| "eval_precision": 0.7327327327327328, | |
| "eval_recall": 0.4288224956063269, | |
| "eval_runtime": 3.1361, | |
| "eval_samples_per_second": 544.307, | |
| "eval_steps_per_second": 17.219, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.8908817768096924, | |
| "learning_rate": 0.00012118535335550633, | |
| "loss": 0.4702, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7527826596367897, | |
| "eval_f1": 0.5193621867881549, | |
| "eval_loss": 0.4752483069896698, | |
| "eval_mcc": 0.4034441704043775, | |
| "eval_precision": 0.7378640776699029, | |
| "eval_recall": 0.4007029876977153, | |
| "eval_runtime": 3.1566, | |
| "eval_samples_per_second": 540.775, | |
| "eval_steps_per_second": 17.107, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 6.063438415527344, | |
| "learning_rate": 9.088901501662975e-05, | |
| "loss": 0.4592, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7580550673696543, | |
| "eval_f1": 0.565720294426919, | |
| "eval_loss": 0.4738541543483734, | |
| "eval_mcc": 0.42241011732814765, | |
| "eval_precision": 0.7041884816753927, | |
| "eval_recall": 0.4727592267135325, | |
| "eval_runtime": 3.1379, | |
| "eval_samples_per_second": 544.002, | |
| "eval_steps_per_second": 17.209, | |
| "step": 1070 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1712, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 2623878322200.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9251938490023766, | |
| "learning_rate": 0.00024237070671101265, | |
| "num_train_epochs": 8, | |
| "temperature": 22 | |
| } | |
| } | |