| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9997758637251448, | |
| "global_step": 7806, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.74046197767973e-05, | |
| "loss": 1.6733, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.4160394497793926e-05, | |
| "loss": 1.2466, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.091616921879056e-05, | |
| "loss": 1.1359, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.767194393978718e-05, | |
| "loss": 1.0703, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.442771866078381e-05, | |
| "loss": 1.0288, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.1183493381780434e-05, | |
| "loss": 0.984, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.793926810277706e-05, | |
| "loss": 0.9522, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.4695042823773685e-05, | |
| "loss": 0.9249, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.1450817544770312e-05, | |
| "loss": 0.8816, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.8206592265766935e-05, | |
| "loss": 0.8639, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4962366986763562e-05, | |
| "loss": 0.8497, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.1718141707760188e-05, | |
| "loss": 0.8336, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.473916428756813e-06, | |
| "loss": 0.8292, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.229691149753439e-06, | |
| "loss": 0.8207, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.985465870750065e-06, | |
| "loss": 0.8128, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 7806, | |
| "total_flos": 2.5178769031430144e+17, | |
| "train_loss": 0.9868680038667171, | |
| "train_runtime": 7400.7434, | |
| "train_samples_per_second": 405.116, | |
| "train_steps_per_second": 1.055 | |
| } | |
| ], | |
| "max_steps": 7806, | |
| "num_train_epochs": 2, | |
| "total_flos": 2.5178769031430144e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |