| { | |
| "best_metric": 0.819954128440367, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-527", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 527, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00013462097079767568, | |
| "loss": 1.19, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.819954128440367, | |
| "eval_loss": 1.0689420700073242, | |
| "eval_runtime": 10.3665, | |
| "eval_samples_per_second": 84.117, | |
| "eval_steps_per_second": 0.675, | |
| "step": 527 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3689, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 8069051778960.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6204520709084631, | |
| "learning_rate": 0.00015705779926395499, | |
| "num_train_epochs": 7, | |
| "temperature": 16 | |
| } | |
| } | |