| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 576, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00013103448275862068, | |
| "loss": 0.7427, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019305019305019305, | |
| "loss": 0.3556, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00017837837837837839, | |
| "loss": 0.1922, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00016370656370656372, | |
| "loss": 0.1829, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00014903474903474903, | |
| "loss": 0.1719, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00013436293436293437, | |
| "loss": 0.1595, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0001196911196911197, | |
| "loss": 0.1608, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00010501930501930501, | |
| "loss": 0.1588, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 9.034749034749035e-05, | |
| "loss": 0.1547, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 7.567567567567568e-05, | |
| "loss": 0.1539, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 6.100386100386101e-05, | |
| "loss": 0.1433, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.6332046332046336e-05, | |
| "loss": 0.1439, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 3.166023166023166e-05, | |
| "loss": 0.1371, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.698841698841699e-05, | |
| "loss": 0.1431, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.3166023166023166e-06, | |
| "loss": 0.1422, | |
| "step": 570 | |
| } | |
| ], | |
| "logging_steps": 38, | |
| "max_steps": 576, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 9.340699891728384e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |