| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 7320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.25e-05, | |
| "loss": 6.1167, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.39470717462922345, | |
| "eval_loss": 3.535817861557007, | |
| "eval_runtime": 3.8077, | |
| "eval_samples_per_second": 1179.442, | |
| "eval_steps_per_second": 2.364, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.000125, | |
| "loss": 3.4801, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.43199785397566787, | |
| "eval_loss": 3.1201677322387695, | |
| "eval_runtime": 3.9073, | |
| "eval_samples_per_second": 1149.375, | |
| "eval_steps_per_second": 2.303, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0001875, | |
| "loss": 3.1329, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00025, | |
| "loss": 2.9429, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.44977794609341165, | |
| "eval_loss": 2.915814161300659, | |
| "eval_runtime": 4.1629, | |
| "eval_samples_per_second": 1078.826, | |
| "eval_steps_per_second": 2.162, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.0003125, | |
| "loss": 2.8071, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.460516834193321, | |
| "eval_loss": 2.790581464767456, | |
| "eval_runtime": 4.0203, | |
| "eval_samples_per_second": 1117.085, | |
| "eval_steps_per_second": 2.239, | |
| "step": 2928 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 0.000375, | |
| "loss": 2.7064, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.00043750000000000006, | |
| "loss": 2.6197, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4701406312186929, | |
| "eval_loss": 2.6998250484466553, | |
| "eval_runtime": 4.0911, | |
| "eval_samples_per_second": 1097.737, | |
| "eval_steps_per_second": 2.2, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 0.0005, | |
| "loss": 2.5459, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4759738199057783, | |
| "eval_loss": 2.641901969909668, | |
| "eval_runtime": 4.078, | |
| "eval_samples_per_second": 1101.267, | |
| "eval_steps_per_second": 2.207, | |
| "step": 4392 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 2.5015, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.000625, | |
| "loss": 2.4492, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4802465824036524, | |
| "eval_loss": 2.6036274433135986, | |
| "eval_runtime": 4.1331, | |
| "eval_samples_per_second": 1086.595, | |
| "eval_steps_per_second": 2.178, | |
| "step": 5124 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.0006875, | |
| "loss": 2.4065, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4823891001600752, | |
| "eval_loss": 2.5769591331481934, | |
| "eval_runtime": 3.9166, | |
| "eval_samples_per_second": 1146.648, | |
| "eval_steps_per_second": 2.298, | |
| "step": 5856 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 0.00075, | |
| "loss": 2.3867, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.0008125, | |
| "loss": 2.3626, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4863024386480748, | |
| "eval_loss": 2.5622453689575195, | |
| "eval_runtime": 3.9696, | |
| "eval_samples_per_second": 1131.334, | |
| "eval_steps_per_second": 2.267, | |
| "step": 6588 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 0.0008750000000000001, | |
| "loss": 2.3276, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4861165901356519, | |
| "eval_loss": 2.557232141494751, | |
| "eval_runtime": 4.198, | |
| "eval_samples_per_second": 1069.803, | |
| "eval_steps_per_second": 2.144, | |
| "step": 7320 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 7320, | |
| "total_flos": 2273237316403200.0, | |
| "train_loss": 2.8881424867390284, | |
| "train_runtime": 618.3653, | |
| "train_samples_per_second": 757.562, | |
| "train_steps_per_second": 11.838 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7320, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "total_flos": 2273237316403200.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |