| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3732, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.330117899249732e-05, | |
| "loss": 0.6382, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.660235798499465e-05, | |
| "loss": 0.4071, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 2.9903536977491963e-05, | |
| "loss": 0.4028, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.3204715969989283e-05, | |
| "loss": 0.3586, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.6505894962486605e-05, | |
| "loss": 0.346, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.807073954983924e-06, | |
| "loss": 0.3211, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.1082529474812435e-06, | |
| "loss": 0.3348, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 3732, | |
| "total_flos": 88369737498624.0, | |
| "train_loss": 0.3977263656076511, | |
| "train_runtime": 1126.9469, | |
| "train_samples_per_second": 13.241, | |
| "train_steps_per_second": 3.312 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3732, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 88369737498624.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |