| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 8088, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.8145400593471808e-05, | |
| "loss": 1.3617, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.6290801186943622e-05, | |
| "loss": 1.256, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.443620178041543e-05, | |
| "loss": 1.2066, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 2.258160237388724e-05, | |
| "loss": 1.135, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 2.072700296735905e-05, | |
| "loss": 1.106, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.887240356083086e-05, | |
| "loss": 0.9174, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.7017804154302673e-05, | |
| "loss": 0.7988, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.516320474777448e-05, | |
| "loss": 0.7843, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.3308605341246291e-05, | |
| "loss": 0.7841, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.14540059347181e-05, | |
| "loss": 0.7567, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.599406528189911e-06, | |
| "loss": 0.6834, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 7.74480712166172e-06, | |
| "loss": 0.5171, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 5.890207715133531e-06, | |
| "loss": 0.5309, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.0356083086053414e-06, | |
| "loss": 0.5064, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2.1810089020771514e-06, | |
| "loss": 0.4908, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 3.264094955489614e-07, | |
| "loss": 0.5029, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 8088, | |
| "total_flos": 1.9020656992075776e+16, | |
| "train_loss": 0.8305775371668482, | |
| "train_runtime": 62910.5204, | |
| "train_samples_per_second": 1.543, | |
| "train_steps_per_second": 0.129 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8088, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.9020656992075776e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |