| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 21.278782894736842, | |
| "global_step": 103500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.588973502853666e-06, | |
| "loss": 0.6499, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.177782529317917e-06, | |
| "loss": 0.5129, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 8.766591555782168e-06, | |
| "loss": 0.4726, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 8.355400582246419e-06, | |
| "loss": 0.4026, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.94420960871067e-06, | |
| "loss": 0.3312, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 7.533018635174921e-06, | |
| "loss": 0.2952, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 7.121827661639172e-06, | |
| "loss": 0.229, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 6.710636688103423e-06, | |
| "loss": 0.1724, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 6.2994457145676745e-06, | |
| "loss": 0.1658, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 5.888254741031926e-06, | |
| "loss": 0.1506, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 5.477063767496177e-06, | |
| "loss": 0.1034, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 5.065872793960427e-06, | |
| "loss": 0.0735, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "learning_rate": 4.654681820424679e-06, | |
| "loss": 0.0548, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 4.24349084688893e-06, | |
| "loss": 0.0437, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 15.42, | |
| "learning_rate": 3.83229987335318e-06, | |
| "loss": 0.0721, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 16.45, | |
| "learning_rate": 3.421108899817431e-06, | |
| "loss": 0.0555, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 3.0099179262816823e-06, | |
| "loss": 0.041, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 2.5987269527459334e-06, | |
| "loss": 0.034, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 2.187535979210185e-06, | |
| "loss": 0.0392, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "learning_rate": 1.7763450056744355e-06, | |
| "loss": 0.035, | |
| "step": 100000 | |
| } | |
| ], | |
| "max_steps": 121600, | |
| "num_train_epochs": 25, | |
| "total_flos": 8.999206543215821e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |