| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 32028, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.997814548860443e-05, | |
| "loss": 2.2558, | |
| "step": 2135 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.889158716392021e-05, | |
| "loss": 1.8307, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.6671639202081526e-05, | |
| "loss": 1.6597, | |
| "step": 6405 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.445273200346921e-05, | |
| "loss": 1.5658, | |
| "step": 8540 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.2230702515177798e-05, | |
| "loss": 1.5067, | |
| "step": 10675 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.000971379011275e-05, | |
| "loss": 1.454, | |
| "step": 12810 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.7787684301821338e-05, | |
| "loss": 1.4184, | |
| "step": 14945 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.5565654813529924e-05, | |
| "loss": 1.3922, | |
| "step": 17080 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3343625325238507e-05, | |
| "loss": 1.3727, | |
| "step": 19215 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.112263660017346e-05, | |
| "loss": 1.3499, | |
| "step": 21350 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.900607111882046e-06, | |
| "loss": 1.3314, | |
| "step": 23485 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.680659150043365e-06, | |
| "loss": 1.3167, | |
| "step": 25620 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.458629661751952e-06, | |
| "loss": 1.3074, | |
| "step": 27755 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.237640936686904e-06, | |
| "loss": 1.304, | |
| "step": 29890 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.5611448395490027e-08, | |
| "loss": 1.2955, | |
| "step": 32025 | |
| } | |
| ], | |
| "logging_steps": 2135, | |
| "max_steps": 32028, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.801023057794171e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |