| { | |
| "best_metric": 0.14963118731975555, | |
| "best_model_checkpoint": "subtaskA/0/checkpoint-8621", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 8621, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.903336813208058e-06, | |
| "loss": 0.3323, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.806673626416116e-06, | |
| "loss": 0.1967, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.710010439624174e-06, | |
| "loss": 0.1751, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.613347252832231e-06, | |
| "loss": 0.1455, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.516684066040289e-06, | |
| "loss": 0.1402, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.420020879248348e-06, | |
| "loss": 0.1204, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.323357692456405e-06, | |
| "loss": 0.121, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.226694505664463e-06, | |
| "loss": 0.1137, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.130031318872521e-06, | |
| "loss": 0.1053, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.033368132080579e-06, | |
| "loss": 0.1108, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.936704945288636e-06, | |
| "loss": 0.1065, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.840041758496694e-06, | |
| "loss": 0.1126, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.743378571704752e-06, | |
| "loss": 0.0865, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3.64671538491281e-06, | |
| "loss": 0.092, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3.550052198120868e-06, | |
| "loss": 0.0961, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.453389011328926e-06, | |
| "loss": 0.0967, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.356725824536984e-06, | |
| "loss": 0.0922, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9706820554460039, | |
| "eval_loss": 0.14963118731975555, | |
| "eval_runtime": 893.6565, | |
| "eval_samples_per_second": 38.588, | |
| "eval_steps_per_second": 2.413, | |
| "step": 8621 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 25863, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.2852245045657875e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |