| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25316455696202533, | |
| "grad_norm": 4.975057833206688, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.6219, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5063291139240507, | |
| "grad_norm": 3.329824512344662, | |
| "learning_rate": 9.896703108827758e-06, | |
| "loss": 1.3156, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.759493670886076, | |
| "grad_norm": 3.1011128116963076, | |
| "learning_rate": 9.401006955900555e-06, | |
| "loss": 1.2738, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.286114229086214, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 1.2392, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2531645569620253, | |
| "grad_norm": 3.715139891693918, | |
| "learning_rate": 7.373001848738203e-06, | |
| "loss": 0.7197, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5063291139240507, | |
| "grad_norm": 2.8850206230172413, | |
| "learning_rate": 6.011087861660191e-06, | |
| "loss": 0.6507, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.759493670886076, | |
| "grad_norm": 3.4047755587817092, | |
| "learning_rate": 4.564221286261709e-06, | |
| "loss": 0.6525, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.553288498796671, | |
| "learning_rate": 3.1539692634365788e-06, | |
| "loss": 0.6397, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.2531645569620253, | |
| "grad_norm": 2.79377523557134, | |
| "learning_rate": 1.8988225436587005e-06, | |
| "loss": 0.3171, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5063291139240507, | |
| "grad_norm": 3.0196884442909346, | |
| "learning_rate": 9.042397785550405e-07, | |
| "loss": 0.2635, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.759493670886076, | |
| "grad_norm": 3.106499382571576, | |
| "learning_rate": 2.5378678213483057e-07, | |
| "loss": 0.2825, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.8625553748005914, | |
| "learning_rate": 2.1152495889970035e-09, | |
| "loss": 0.2523, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 120, | |
| "total_flos": 4908562513920.0, | |
| "train_loss": 0.7690611402193706, | |
| "train_runtime": 205.6825, | |
| "train_samples_per_second": 72.928, | |
| "train_steps_per_second": 0.583 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4908562513920.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |