| { | |
| "best_metric": 0.8698768019676208, | |
| "best_model_checkpoint": "./electra-large/fine_tuned_models/checkpoint-9944", | |
| "epoch": 5.0, | |
| "global_step": 12430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.7777777777777777e-05, | |
| "loss": 0.482, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8477043509483337, | |
| "eval_loss": 0.3800182342529297, | |
| "eval_runtime": 124.5247, | |
| "eval_samples_per_second": 35.856, | |
| "eval_steps_per_second": 4.489, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.3208, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.850615918636322, | |
| "eval_loss": 0.3961940407752991, | |
| "eval_runtime": 124.3491, | |
| "eval_samples_per_second": 35.907, | |
| "eval_steps_per_second": 4.495, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.239, | |
| "step": 7458 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8685330152511597, | |
| "eval_loss": 0.44575047492980957, | |
| "eval_runtime": 124.3295, | |
| "eval_samples_per_second": 35.913, | |
| "eval_steps_per_second": 4.496, | |
| "step": 7458 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.1725, | |
| "step": 9944 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8698768019676208, | |
| "eval_loss": 0.4581369161605835, | |
| "eval_runtime": 124.3253, | |
| "eval_samples_per_second": 35.914, | |
| "eval_steps_per_second": 4.496, | |
| "step": 9944 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1233, | |
| "step": 12430 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8687570095062256, | |
| "eval_loss": 0.5259796380996704, | |
| "eval_runtime": 124.3289, | |
| "eval_samples_per_second": 35.913, | |
| "eval_steps_per_second": 4.496, | |
| "step": 12430 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 12430, | |
| "total_flos": 4.632747298523904e+16, | |
| "train_loss": 0.2675398982348024, | |
| "train_runtime": 16286.6127, | |
| "train_samples_per_second": 12.209, | |
| "train_steps_per_second": 0.763 | |
| } | |
| ], | |
| "max_steps": 12430, | |
| "num_train_epochs": 5, | |
| "total_flos": 4.632747298523904e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |