| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 7850, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.681528662420383e-05, | |
| "loss": 1.5828, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 4.3630573248407646e-05, | |
| "loss": 0.2954, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 4.044585987261147e-05, | |
| "loss": 0.123, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 3.7261146496815283e-05, | |
| "loss": 0.0777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 3.407643312101911e-05, | |
| "loss": 0.0501, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.11, | |
| "learning_rate": 3.089171974522293e-05, | |
| "loss": 0.0339, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.29, | |
| "learning_rate": 2.7707006369426753e-05, | |
| "loss": 0.0237, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 25.48, | |
| "learning_rate": 2.4522292993630575e-05, | |
| "loss": 0.0156, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 28.66, | |
| "learning_rate": 2.1337579617834397e-05, | |
| "loss": 0.0123, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.85, | |
| "learning_rate": 1.8152866242038215e-05, | |
| "loss": 0.0092, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 35.03, | |
| "learning_rate": 1.4968152866242039e-05, | |
| "loss": 0.0065, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.22, | |
| "learning_rate": 1.178343949044586e-05, | |
| "loss": 0.0057, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.4, | |
| "learning_rate": 8.598726114649681e-06, | |
| "loss": 0.0035, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.59, | |
| "learning_rate": 5.414012738853504e-06, | |
| "loss": 0.0034, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 47.77, | |
| "learning_rate": 2.229299363057325e-06, | |
| "loss": 0.0024, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 7850, | |
| "total_flos": 2.4497456256e+16, | |
| "train_loss": 0.14310269137856307, | |
| "train_runtime": 1143.4603, | |
| "train_samples_per_second": 218.635, | |
| "train_steps_per_second": 6.865 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7850, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 2.4497456256e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |