| { | |
| "best_metric": 0.8896859247770453, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-645", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 2580, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.358044296503067, | |
| "learning_rate": 2.909167383222674e-05, | |
| "loss": 0.1708, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8896859247770453, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.15664567053318024, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 10.0662, | |
| "eval_samples_per_second": 512.406, | |
| "eval_steps_per_second": 16.093, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2336525022983551, | |
| "learning_rate": 1.9394449221484495e-05, | |
| "loss": 0.1576, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8896859247770453, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.15651842951774597, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 9.954, | |
| "eval_samples_per_second": 518.185, | |
| "eval_steps_per_second": 16.275, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4220493733882904, | |
| "learning_rate": 9.697224610742248e-06, | |
| "loss": 0.1568, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8896859247770453, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.1553562432527542, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 9.9391, | |
| "eval_samples_per_second": 518.962, | |
| "eval_steps_per_second": 16.299, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2776136100292206, | |
| "learning_rate": 0.0, | |
| "loss": 0.1564, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8896859247770453, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.15526321530342102, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 9.934, | |
| "eval_samples_per_second": 519.229, | |
| "eval_steps_per_second": 16.308, | |
| "step": 2580 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2580, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 6347770130400.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.39895457481552465, | |
| "learning_rate": 3.878889844296899e-05, | |
| "num_train_epochs": 4, | |
| "temperature": 23 | |
| } | |
| } | |