| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "global_step": 9648, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.16, |
| "learning_rate": 4.740878938640133e-05, |
| "loss": 0.0449, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 4.4817578772802655e-05, |
| "loss": 0.0385, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.31, |
| "eval_dev_RMSE": 0.18822340609718077, |
| "eval_dev_loss": 0.03542805090546608, |
| "eval_dev_runtime": 133.0034, |
| "eval_dev_samples_per_second": 92.178, |
| "eval_dev_steps_per_second": 11.526, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 4.222636815920398e-05, |
| "loss": 0.0369, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 3.9635157545605314e-05, |
| "loss": 0.0359, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.62, |
| "eval_dev_RMSE": 0.18694981664838944, |
| "eval_dev_loss": 0.03495023399591446, |
| "eval_dev_runtime": 133.0499, |
| "eval_dev_samples_per_second": 92.146, |
| "eval_dev_steps_per_second": 11.522, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 3.7043946932006634e-05, |
| "loss": 0.0343, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 3.445273631840796e-05, |
| "loss": 0.0329, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.93, |
| "eval_dev_RMSE": 0.17415666711451466, |
| "eval_dev_loss": 0.030330544337630272, |
| "eval_dev_runtime": 133.0816, |
| "eval_dev_samples_per_second": 92.124, |
| "eval_dev_steps_per_second": 11.519, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 3.1861525704809286e-05, |
| "loss": 0.0263, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.24, |
| "learning_rate": 2.9270315091210616e-05, |
| "loss": 0.0214, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_dev_RMSE": 0.18201157943729365, |
| "eval_dev_loss": 0.03312821686267853, |
| "eval_dev_runtime": 133.0186, |
| "eval_dev_samples_per_second": 92.168, |
| "eval_dev_steps_per_second": 11.525, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.4, |
| "learning_rate": 2.6679104477611942e-05, |
| "loss": 0.0203, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.55, |
| "learning_rate": 2.408789386401327e-05, |
| "loss": 0.0211, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.55, |
| "eval_dev_RMSE": 0.17764016881454997, |
| "eval_dev_loss": 0.03155602887272835, |
| "eval_dev_runtime": 132.9944, |
| "eval_dev_samples_per_second": 92.184, |
| "eval_dev_steps_per_second": 11.527, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 2.1496683250414595e-05, |
| "loss": 0.0202, |
| "step": 5500 |
| }, |
| { |
| "epoch": 1.87, |
| "learning_rate": 1.890547263681592e-05, |
| "loss": 0.0211, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.87, |
| "eval_dev_RMSE": 0.17477992169342263, |
| "eval_dev_loss": 0.03054802119731903, |
| "eval_dev_runtime": 132.9682, |
| "eval_dev_samples_per_second": 92.202, |
| "eval_dev_steps_per_second": 11.529, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 1.6314262023217248e-05, |
| "loss": 0.0201, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.18, |
| "learning_rate": 1.3723051409618576e-05, |
| "loss": 0.0114, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.18, |
| "eval_dev_RMSE": 0.18422651848103616, |
| "eval_dev_loss": 0.033939410001039505, |
| "eval_dev_runtime": 133.0526, |
| "eval_dev_samples_per_second": 92.144, |
| "eval_dev_steps_per_second": 11.522, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.33, |
| "learning_rate": 1.1131840796019902e-05, |
| "loss": 0.0113, |
| "step": 7500 |
| }, |
| { |
| "epoch": 2.49, |
| "learning_rate": 8.540630182421228e-06, |
| "loss": 0.0114, |
| "step": 8000 |
| }, |
| { |
| "epoch": 2.49, |
| "eval_dev_RMSE": 0.18135978128058372, |
| "eval_dev_loss": 0.032891370356082916, |
| "eval_dev_runtime": 133.0402, |
| "eval_dev_samples_per_second": 92.153, |
| "eval_dev_steps_per_second": 11.523, |
| "step": 8000 |
| }, |
| { |
| "epoch": 2.64, |
| "learning_rate": 5.9494195688225545e-06, |
| "loss": 0.0108, |
| "step": 8500 |
| }, |
| { |
| "epoch": 2.8, |
| "learning_rate": 3.358208955223881e-06, |
| "loss": 0.0107, |
| "step": 9000 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_dev_RMSE": 0.17957866610604725, |
| "eval_dev_loss": 0.03224850073456764, |
| "eval_dev_runtime": 133.0207, |
| "eval_dev_samples_per_second": 92.166, |
| "eval_dev_steps_per_second": 11.525, |
| "step": 9000 |
| }, |
| { |
| "epoch": 2.95, |
| "learning_rate": 7.669983416252073e-07, |
| "loss": 0.0105, |
| "step": 9500 |
| } |
| ], |
| "max_steps": 9648, |
| "num_train_epochs": 3, |
| "total_flos": 8.121060296152474e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
