| { | |
| "best_metric": 0.5312319613383731, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-6\\checkpoint-4276", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 8552, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 20.960588455200195, | |
| "learning_rate": 1.1786229789654309e-05, | |
| "loss": 0.5916, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 8.482439994812012, | |
| "learning_rate": 1.1207906443743891e-05, | |
| "loss": 0.5669, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 5.808671951293945, | |
| "learning_rate": 1.0629583097833473e-05, | |
| "loss": 0.5521, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 45.093868255615234, | |
| "learning_rate": 1.0051259751923056e-05, | |
| "loss": 0.5405, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6268027424812317, | |
| "eval_matthews_correlation": 0.41222715814310473, | |
| "eval_runtime": 0.8446, | |
| "eval_samples_per_second": 1234.939, | |
| "eval_steps_per_second": 78.146, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.9396941065788269, | |
| "learning_rate": 9.472936406012638e-06, | |
| "loss": 0.4693, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3982331454753876, | |
| "learning_rate": 8.89461306010222e-06, | |
| "loss": 0.4566, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 30.939773559570312, | |
| "learning_rate": 8.316289714191805e-06, | |
| "loss": 0.5094, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 21.7435302734375, | |
| "learning_rate": 7.737966368281385e-06, | |
| "loss": 0.485, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7867382764816284, | |
| "eval_matthews_correlation": 0.5312319613383731, | |
| "eval_runtime": 0.8357, | |
| "eval_samples_per_second": 1247.998, | |
| "eval_steps_per_second": 78.972, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.48719292879104614, | |
| "learning_rate": 7.159643022370969e-06, | |
| "loss": 0.3606, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 8.735502243041992, | |
| "learning_rate": 6.58131967646055e-06, | |
| "loss": 0.3344, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 7.660526275634766, | |
| "learning_rate": 6.0029963305501335e-06, | |
| "loss": 0.2929, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 0.15256187319755554, | |
| "learning_rate": 5.424672984639716e-06, | |
| "loss": 0.3086, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.9165342450141907, | |
| "eval_matthews_correlation": 0.5238027074526898, | |
| "eval_runtime": 0.8417, | |
| "eval_samples_per_second": 1239.165, | |
| "eval_steps_per_second": 78.413, | |
| "step": 6414 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 138.51580810546875, | |
| "learning_rate": 4.846349638729298e-06, | |
| "loss": 0.3321, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 0.0786728635430336, | |
| "learning_rate": 4.268026292818881e-06, | |
| "loss": 0.2311, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 14.107312202453613, | |
| "learning_rate": 3.6897029469084635e-06, | |
| "loss": 0.2378, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 5.254330635070801, | |
| "learning_rate": 3.1113796009980463e-06, | |
| "loss": 0.266, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 0.3652295768260956, | |
| "learning_rate": 2.5330562550876287e-06, | |
| "loss": 0.2457, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.043048620223999, | |
| "eval_matthews_correlation": 0.5135566841640817, | |
| "eval_runtime": 0.861, | |
| "eval_samples_per_second": 1211.328, | |
| "eval_steps_per_second": 76.652, | |
| "step": 8552 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10690, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 138507342838860.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.2364553135564726e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 5 | |
| } | |
| } | |