| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.8, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.25, | |
| "learning_rate": 2.9968542393565676e-06, | |
| "loss": 1.7053, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.0625, | |
| "learning_rate": 2.9616157869703894e-06, | |
| "loss": 1.5958, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 2.8881318444640566e-06, | |
| "loss": 1.6746, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 2.778325235483954e-06, | |
| "loss": 1.5386, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.921875, | |
| "learning_rate": 2.6350692237265428e-06, | |
| "loss": 1.5512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.453125, | |
| "learning_rate": 2.4621123294467098e-06, | |
| "loss": 1.5223, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 2.25, | |
| "learning_rate": 2.2639802434931445e-06, | |
| "loss": 1.4408, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 2.0458574054452316e-06, | |
| "loss": 1.551, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 1.813451344546913e-06, | |
| "loss": 1.4931, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 1.5728433331716726e-06, | |
| "loss": 1.4152, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 2.0, | |
| "learning_rate": 1.3303292607070737e-06, | |
| "loss": 1.436, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 1.953125, | |
| "learning_rate": 1.0922548916454855e-06, | |
| "loss": 1.4471, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 20, | |
| "total_flos": 7398534869286912.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |