| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "global_step": 8452, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.9059157595835306e-05, | |
| "loss": 0.0116, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.8118315191670613e-05, | |
| "loss": 0.0759, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.717557974443919e-05, | |
| "loss": 0.2127, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.6229058211074304e-05, | |
| "loss": 0.0926, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.5282536677709418e-05, | |
| "loss": 0.1933, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.4336015144344535e-05, | |
| "loss": 0.0232, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.3389493610979651e-05, | |
| "loss": 0.0176, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.2442972077614768e-05, | |
| "loss": 0.0192, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.1496450544249882e-05, | |
| "loss": 0.0148, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.0551822053951728e-05, | |
| "loss": 0.013, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 9.605300520586845e-06, | |
| "loss": 0.019, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 8.65877898722196e-06, | |
| "loss": 0.0361, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.714150496923806e-06, | |
| "loss": 0.021, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 6.767628963558922e-06, | |
| "loss": 0.0146, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 5.8211074301940376e-06, | |
| "loss": 0.0071, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 4.874585896829153e-06, | |
| "loss": 0.0093, | |
| "step": 8000 | |
| } | |
| ], | |
| "max_steps": 10565, | |
| "num_train_epochs": 5, | |
| "total_flos": 4.709658898479514e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |