| { | |
| "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": 1.2838, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 4.3630573248407646e-05, | |
| "loss": 0.2252, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 4.044585987261147e-05, | |
| "loss": 0.1119, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 3.7261146496815283e-05, | |
| "loss": 0.066, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 3.407643312101911e-05, | |
| "loss": 0.0493, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.11, | |
| "learning_rate": 3.089171974522293e-05, | |
| "loss": 0.041, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.29, | |
| "learning_rate": 2.7707006369426753e-05, | |
| "loss": 0.0492, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 25.48, | |
| "learning_rate": 2.4522292993630575e-05, | |
| "loss": 0.0654, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 28.66, | |
| "learning_rate": 2.1337579617834397e-05, | |
| "loss": 0.0816, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.85, | |
| "learning_rate": 1.8152866242038215e-05, | |
| "loss": 0.1484, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 35.03, | |
| "learning_rate": 1.4968152866242039e-05, | |
| "loss": 0.3258, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.22, | |
| "learning_rate": 1.178343949044586e-05, | |
| "loss": 0.6468, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.4, | |
| "learning_rate": 8.598726114649681e-06, | |
| "loss": 0.7716, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.59, | |
| "learning_rate": 5.414012738853504e-06, | |
| "loss": 1.0718, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 47.77, | |
| "learning_rate": 2.229299363057325e-06, | |
| "loss": 1.2439, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 7850, | |
| "total_flos": 4.8993141888e+16, | |
| "train_loss": 0.4468785848435323, | |
| "train_runtime": 2136.4763, | |
| "train_samples_per_second": 117.015, | |
| "train_steps_per_second": 3.674 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7850, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 4.8993141888e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |