| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 9459, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001331923890063425, | |
| "loss": 1.4868, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019262304710442854, | |
| "loss": 1.3444, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00017782215435216728, | |
| "loss": 1.2245, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00016302126159990603, | |
| "loss": 1.3047, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0001482203688476448, | |
| "loss": 1.2314, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00013341947609538354, | |
| "loss": 1.1001, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00011864207682368143, | |
| "loss": 1.1696, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00010384118407142019, | |
| "loss": 1.2371, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 8.904029131915894e-05, | |
| "loss": 1.1162, | |
| "step": 5670 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 7.426289204745683e-05, | |
| "loss": 1.0911, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.946199929519558e-05, | |
| "loss": 1.29, | |
| "step": 6930 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.466110654293434e-05, | |
| "loss": 1.1845, | |
| "step": 7560 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.9883707271232237e-05, | |
| "loss": 1.1131, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.5082814518970987e-05, | |
| "loss": 1.3343, | |
| "step": 8820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.819217667097381e-07, | |
| "loss": 1.2919, | |
| "step": 9450 | |
| } | |
| ], | |
| "logging_steps": 630, | |
| "max_steps": 9459, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.9268305152049152e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |