| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 4046, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.8580246913580246e-05, | |
| "loss": 1.7863, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5.883034610630408e-05, | |
| "loss": 1.2957, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.917336217552534e-05, | |
| "loss": 0.8962, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.9516378244746606e-05, | |
| "loss": 0.6393, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.48078134655952454, | |
| "eval_runtime": 459.2193, | |
| "eval_samples_per_second": 7.778, | |
| "eval_steps_per_second": 0.78, | |
| "step": 2023 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 2.9859394313967864e-05, | |
| "loss": 0.3256, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 2.0202410383189122e-05, | |
| "loss": 0.2953, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.0545426452410384e-05, | |
| "loss": 0.2583, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 8.88442521631644e-07, | |
| "loss": 0.2394, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.32980823516845703, | |
| "eval_runtime": 459.2016, | |
| "eval_samples_per_second": 7.779, | |
| "eval_steps_per_second": 0.78, | |
| "step": 4046 | |
| } | |
| ], | |
| "max_steps": 4046, | |
| "num_train_epochs": 2, | |
| "total_flos": 6.87875414016e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |