| { | |
| "best_metric": 0.9417647782287369, | |
| "best_model_checkpoint": "model_saves/deberta-large_spell_10k_2_p3/checkpoint-268", | |
| "epoch": 7.0, | |
| "global_step": 1876, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9417647782287369, | |
| "eval_loss": 0.42341962456703186, | |
| "eval_runtime": 4.7165, | |
| "eval_samples_per_second": 929.086, | |
| "eval_steps_per_second": 7.421, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9413510884513467, | |
| "eval_loss": 0.4421840310096741, | |
| "eval_runtime": 4.7038, | |
| "eval_samples_per_second": 931.593, | |
| "eval_steps_per_second": 7.441, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9410156643075168, | |
| "eval_loss": 0.4872588813304901, | |
| "eval_runtime": 4.7049, | |
| "eval_samples_per_second": 931.375, | |
| "eval_steps_per_second": 7.439, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2514, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9403727680318429, | |
| "eval_loss": 0.5361831784248352, | |
| "eval_runtime": 4.6987, | |
| "eval_samples_per_second": 932.608, | |
| "eval_steps_per_second": 7.449, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.94000380147363, | |
| "eval_loss": 0.5959360003471375, | |
| "eval_runtime": 4.6891, | |
| "eval_samples_per_second": 934.505, | |
| "eval_steps_per_second": 7.464, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9387124185198851, | |
| "eval_loss": 0.6461729407310486, | |
| "eval_runtime": 4.6774, | |
| "eval_samples_per_second": 936.845, | |
| "eval_steps_per_second": 7.483, | |
| "step": 1608 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9388857209941972, | |
| "eval_loss": 0.7177152633666992, | |
| "eval_runtime": 4.6912, | |
| "eval_samples_per_second": 934.082, | |
| "eval_steps_per_second": 7.461, | |
| "step": 1876 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 1876, | |
| "total_flos": 2.8568658875777024e+16, | |
| "train_loss": 0.03849986062121036, | |
| "train_runtime": 360.0465, | |
| "train_samples_per_second": 1429.149, | |
| "train_steps_per_second": 11.165 | |
| } | |
| ], | |
| "max_steps": 4020, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.8568658875777024e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |