| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 111, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 1.2176, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00019595959595959596, | |
| "loss": 0.9247, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.6798, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00016767676767676768, | |
| "loss": 0.5238, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00015353535353535353, | |
| "loss": 0.4068, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0001393939393939394, | |
| "loss": 0.3102, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00012525252525252527, | |
| "loss": 0.2554, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.1935, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 9.696969696969698e-05, | |
| "loss": 0.1378, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 8.282828282828283e-05, | |
| "loss": 0.1465, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.86868686868687e-05, | |
| "loss": 0.1331, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.1259, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.1218, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.6262626262626268e-05, | |
| "loss": 0.127, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.1137, | |
| "step": 105 | |
| } | |
| ], | |
| "max_steps": 111, | |
| "num_train_epochs": 3, | |
| "total_flos": 9089713377902592.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |