| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "global_step": 1650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8713, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 74.73982970671712, | |
| "eval_f1": 83.85927956580818, | |
| "eval_runtime": 4.7071, | |
| "eval_samples_per_second": 2257.456, | |
| "eval_steps_per_second": 7.223, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0257, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 78.6092715231788, | |
| "eval_f1": 86.65693824950102, | |
| "eval_runtime": 3.83, | |
| "eval_samples_per_second": 2774.405, | |
| "eval_steps_per_second": 8.877, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8274, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 79.36613055818354, | |
| "eval_f1": 87.32332344288875, | |
| "eval_runtime": 3.8102, | |
| "eval_samples_per_second": 2788.796, | |
| "eval_steps_per_second": 8.923, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6861, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 79.39451277199622, | |
| "eval_f1": 87.21948865558768, | |
| "eval_runtime": 3.8507, | |
| "eval_samples_per_second": 2759.476, | |
| "eval_steps_per_second": 8.829, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5711, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 79.05392620624409, | |
| "eval_f1": 87.26064183346908, | |
| "eval_runtime": 3.8281, | |
| "eval_samples_per_second": 2775.791, | |
| "eval_steps_per_second": 8.882, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4746, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 79.16745506149479, | |
| "eval_f1": 87.46509960322021, | |
| "eval_runtime": 3.8602, | |
| "eval_samples_per_second": 2752.71, | |
| "eval_steps_per_second": 8.808, | |
| "step": 1650 | |
| } | |
| ], | |
| "max_steps": 5500, | |
| "num_train_epochs": 20, | |
| "total_flos": 198561583071232.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |