| { | |
| "best_metric": 2.182462215423584, | |
| "best_model_checkpoint": "output/skillet/checkpoint-56", | |
| "epoch": 2.0, | |
| "global_step": 56, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00012668528006706028, | |
| "loss": 2.8505, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.836442450346448e-05, | |
| "loss": 2.6309, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 6.0919236939313083e-05, | |
| "loss": 2.4816, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.5828599592490882e-05, | |
| "loss": 2.5761, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.849603540845984e-06, | |
| "loss": 2.3489, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.3767073154449463, | |
| "eval_runtime": 1.7074, | |
| "eval_samples_per_second": 22.256, | |
| "eval_steps_per_second": 2.928, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.7199452243268996e-06, | |
| "loss": 2.2694, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.009247481060283e-05, | |
| "loss": 2.3714, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5.333506393059682e-05, | |
| "loss": 2.3801, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.125714365012444e-05, | |
| "loss": 2.3492, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00012223363969730684, | |
| "loss": 2.3538, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00013676865759867644, | |
| "loss": 2.1417, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.182462215423584, | |
| "eval_runtime": 1.7149, | |
| "eval_samples_per_second": 22.159, | |
| "eval_steps_per_second": 2.916, | |
| "step": 56 | |
| } | |
| ], | |
| "max_steps": 56, | |
| "num_train_epochs": 2, | |
| "total_flos": 58398769152000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |