| { | |
| "best_metric": 0.5115958993985192, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-1340", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1340, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.46541309356689453, | |
| "eval_matthews_correlation": 0.4611631351812362, | |
| "eval_runtime": 0.7349, | |
| "eval_samples_per_second": 1419.226, | |
| "eval_steps_per_second": 89.807, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 5.322132587432861, | |
| "learning_rate": 2.8926577590888014e-05, | |
| "loss": 0.415, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5235322117805481, | |
| "eval_matthews_correlation": 0.48584699342231225, | |
| "eval_runtime": 0.8286, | |
| "eval_samples_per_second": 1258.804, | |
| "eval_steps_per_second": 79.656, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6811769604682922, | |
| "eval_matthews_correlation": 0.4886634303966225, | |
| "eval_runtime": 1.1682, | |
| "eval_samples_per_second": 892.827, | |
| "eval_steps_per_second": 56.497, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 5.797658920288086, | |
| "learning_rate": 1.1708376643930863e-05, | |
| "loss": 0.1523, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8018879294395447, | |
| "eval_matthews_correlation": 0.5029119946692125, | |
| "eval_runtime": 0.7895, | |
| "eval_samples_per_second": 1321.02, | |
| "eval_steps_per_second": 83.593, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.930472731590271, | |
| "eval_matthews_correlation": 0.5115958993985192, | |
| "eval_runtime": 1.1184, | |
| "eval_samples_per_second": 932.607, | |
| "eval_steps_per_second": 59.014, | |
| "step": 1340 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 191380204405152.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.6144778537845165e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 32, | |
| "seed": 3 | |
| } | |
| } | |