| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 4167, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0007199424046076314, |
| "grad_norm": 10.14134407043457, |
| "learning_rate": 1.9995200383969286e-05, |
| "loss": 1.8741, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.3599712023038157, |
| "grad_norm": 10.300615310668945, |
| "learning_rate": 1.7600191984641232e-05, |
| "loss": 0.8262, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.7199424046076314, |
| "grad_norm": 6.374919414520264, |
| "learning_rate": 1.520038396928246e-05, |
| "loss": 0.6731, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7772375292634612, |
| "eval_loss": 0.6286355257034302, |
| "eval_runtime": 9.0605, |
| "eval_samples_per_second": 612.877, |
| "eval_steps_per_second": 19.204, |
| "step": 1389 |
| }, |
| { |
| "epoch": 1.079913606911447, |
| "grad_norm": 8.19632339477539, |
| "learning_rate": 1.2800575953923686e-05, |
| "loss": 0.5969, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.4398848092152627, |
| "grad_norm": 9.30715560913086, |
| "learning_rate": 1.0400767938564915e-05, |
| "loss": 0.473, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.7998560115190785, |
| "grad_norm": 12.246039390563965, |
| "learning_rate": 8.000959923206144e-06, |
| "loss": 0.4456, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7797586889969386, |
| "eval_loss": 0.6453832387924194, |
| "eval_runtime": 8.9254, |
| "eval_samples_per_second": 622.156, |
| "eval_steps_per_second": 19.495, |
| "step": 2778 |
| }, |
| { |
| "epoch": 2.159827213822894, |
| "grad_norm": 7.0848212242126465, |
| "learning_rate": 5.6011519078473734e-06, |
| "loss": 0.3738, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.5197984161267097, |
| "grad_norm": 10.242433547973633, |
| "learning_rate": 3.2013438924886014e-06, |
| "loss": 0.2944, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8797696184305255, |
| "grad_norm": 9.204344749450684, |
| "learning_rate": 8.015358771298297e-07, |
| "loss": 0.2798, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7801188546731497, |
| "eval_loss": 0.7414410710334778, |
| "eval_runtime": 9.0231, |
| "eval_samples_per_second": 615.419, |
| "eval_steps_per_second": 19.284, |
| "step": 4167 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 4167, |
| "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": 8765464937720832.0, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
