| { | |
| "best_metric": 0.4491418956579208, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/checkpoint-1605", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1605, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 5.041173934936523, | |
| "learning_rate": 4.466779044509044e-06, | |
| "loss": 0.5565, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5430340766906738, | |
| "eval_matthews_correlation": 0.3453182575293756, | |
| "eval_runtime": 0.8193, | |
| "eval_samples_per_second": 1273.053, | |
| "eval_steps_per_second": 80.557, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 9.476218223571777, | |
| "learning_rate": 3.4399332871506427e-06, | |
| "loss": 0.4477, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5347909927368164, | |
| "eval_matthews_correlation": 0.4329619258860067, | |
| "eval_runtime": 1.3398, | |
| "eval_samples_per_second": 778.476, | |
| "eval_steps_per_second": 49.261, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 13.160967826843262, | |
| "learning_rate": 2.413087529792242e-06, | |
| "loss": 0.3814, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.518277108669281, | |
| "eval_matthews_correlation": 0.4491418956579208, | |
| "eval_runtime": 0.952, | |
| "eval_samples_per_second": 1095.631, | |
| "eval_steps_per_second": 69.33, | |
| "step": 1605 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2675, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 128366859981588.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 5.493624801867445e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 16, | |
| "seed": 31 | |
| } | |
| } | |