| { | |
| "best_metric": 0.3853198145814999, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-2/checkpoint-2138", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 2138, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 4.28505277633667, | |
| "learning_rate": 3.0702893894484785e-06, | |
| "loss": 0.6069, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 9.482794761657715, | |
| "learning_rate": 2.9196373094951675e-06, | |
| "loss": 0.5628, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 22.521339416503906, | |
| "learning_rate": 2.7689852295418565e-06, | |
| "loss": 0.5565, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 26.7753849029541, | |
| "learning_rate": 2.6183331495885454e-06, | |
| "loss": 0.5184, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5730993747711182, | |
| "eval_matthews_correlation": 0.3853198145814999, | |
| "eval_runtime": 0.7612, | |
| "eval_samples_per_second": 1370.225, | |
| "eval_steps_per_second": 86.706, | |
| "step": 2138 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10690, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 32530060484016.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 3.2209414694017896e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 16 | |
| } | |
| } | |