| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7100513015390462, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.936664766609665e-05, | |
| "loss": 0.9425, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.87332953321933e-05, | |
| "loss": 0.8195, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.809994299828995e-05, | |
| "loss": 0.65, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.7466590664386602e-05, | |
| "loss": 0.6137, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.683323833048325e-05, | |
| "loss": 0.5306, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.6199885996579897e-05, | |
| "loss": 0.5254, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.556653366267655e-05, | |
| "loss": 0.4436, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.4933181328773198e-05, | |
| "loss": 0.3741, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4299828994869846e-05, | |
| "loss": 0.3795, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.3666476660966497e-05, | |
| "loss": 0.3805, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1342121362686157, | |
| "eval_runtime": 354.7905, | |
| "eval_samples_per_second": 11.88, | |
| "eval_steps_per_second": 5.942, | |
| "step": 5263 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.3033124327063148e-05, | |
| "loss": 0.4431, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.2399771993159795e-05, | |
| "loss": 0.2834, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.1766419659256444e-05, | |
| "loss": 0.3121, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.1133067325353095e-05, | |
| "loss": 0.2991, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.0499714991449746e-05, | |
| "loss": 0.3684, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 9.866362657546393e-06, | |
| "loss": 0.2117, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.233010323643044e-06, | |
| "loss": 0.2777, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.599657989739692e-06, | |
| "loss": 0.249, | |
| "step": 9000 | |
| } | |
| ], | |
| "max_steps": 15789, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.2537544393728e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |