| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 98176, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.168681083504881e-06, | |
| "loss": 0.3631, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9129903209373408, | |
| "eval_loss": 0.31294044852256775, | |
| "eval_runtime": 78.2833, | |
| "eval_samples_per_second": 125.378, | |
| "eval_steps_per_second": 15.674, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2.334304873326132e-06, | |
| "loss": 0.2901, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.999286631473819e-07, | |
| "loss": 0.2267, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.915333672949567, | |
| "eval_loss": 0.41572055220603943, | |
| "eval_runtime": 78.1399, | |
| "eval_samples_per_second": 125.608, | |
| "eval_steps_per_second": 15.703, | |
| "step": 98176 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 98176, | |
| "total_flos": 3.65975017449044e+17, | |
| "train_loss": 0.2868618026395198, | |
| "train_runtime": 21298.9707, | |
| "train_samples_per_second": 36.875, | |
| "train_steps_per_second": 4.609 | |
| } | |
| ], | |
| "max_steps": 98176, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.65975017449044e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |