| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 86, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.11678832116788321, |
| "grad_norm": 1.9757797718048096, |
| "learning_rate": 4.935806850983034e-06, |
| "loss": 1.5171465873718262, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.23357664233576642, |
| "grad_norm": 0.815299391746521, |
| "learning_rate": 4.555390029237026e-06, |
| "loss": 1.1862421989440919, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.35036496350364965, |
| "grad_norm": 0.5920405387878418, |
| "learning_rate": 3.884004270673711e-06, |
| "loss": 1.0512928009033202, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.46715328467153283, |
| "grad_norm": 0.5378215312957764, |
| "learning_rate": 3.0166935317123824e-06, |
| "loss": 0.9863924026489258, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.583941605839416, |
| "grad_norm": 0.4719153046607971, |
| "learning_rate": 2.0762376643597586e-06, |
| "loss": 0.9521537780761719, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.7007299270072993, |
| "grad_norm": 0.5109142661094666, |
| "learning_rate": 1.195771226998789e-06, |
| "loss": 0.9442000389099121, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.8175182481751825, |
| "grad_norm": 0.45715343952178955, |
| "learning_rate": 4.999364430216639e-07, |
| "loss": 0.9409050941467285, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.9343065693430657, |
| "grad_norm": 0.4816999137401581, |
| "learning_rate": 8.723836714516681e-08, |
| "loss": 0.9243898391723633, |
| "step": 80 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 86, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "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": 8.288309829907251e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
