| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 46.728971962616825, |
| "eval_steps": 2000, |
| "global_step": 10000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 9.345794392523365, |
| "grad_norm": 1.9697070121765137, |
| "learning_rate": 9.09927797833935e-06, |
| "loss": 0.5104, |
| "step": 2000 |
| }, |
| { |
| "epoch": 9.345794392523365, |
| "eval_runtime": 0.0528, |
| "eval_samples_per_second": 18.928, |
| "eval_steps_per_second": 18.928, |
| "step": 2000 |
| }, |
| { |
| "epoch": 18.69158878504673, |
| "grad_norm": 1.371793508529663, |
| "learning_rate": 7.895908543922985e-06, |
| "loss": 0.4437, |
| "step": 4000 |
| }, |
| { |
| "epoch": 18.69158878504673, |
| "eval_runtime": 0.0451, |
| "eval_samples_per_second": 22.17, |
| "eval_steps_per_second": 22.17, |
| "step": 4000 |
| }, |
| { |
| "epoch": 28.037383177570092, |
| "grad_norm": 1.708723783493042, |
| "learning_rate": 6.693140794223827e-06, |
| "loss": 0.4317, |
| "step": 6000 |
| }, |
| { |
| "epoch": 28.037383177570092, |
| "eval_runtime": 0.0454, |
| "eval_samples_per_second": 22.047, |
| "eval_steps_per_second": 22.047, |
| "step": 6000 |
| }, |
| { |
| "epoch": 37.38317757009346, |
| "grad_norm": 1.4688475131988525, |
| "learning_rate": 5.489771359807461e-06, |
| "loss": 0.425, |
| "step": 8000 |
| }, |
| { |
| "epoch": 37.38317757009346, |
| "eval_runtime": 0.0442, |
| "eval_samples_per_second": 22.642, |
| "eval_steps_per_second": 22.642, |
| "step": 8000 |
| }, |
| { |
| "epoch": 46.728971962616825, |
| "grad_norm": 1.5830835103988647, |
| "learning_rate": 4.28820697954272e-06, |
| "loss": 0.4202, |
| "step": 10000 |
| }, |
| { |
| "epoch": 46.728971962616825, |
| "eval_runtime": 0.0442, |
| "eval_samples_per_second": 22.627, |
| "eval_steps_per_second": 22.627, |
| "step": 10000 |
| } |
| ], |
| "logging_steps": 2000, |
| "max_steps": 17120, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 80, |
| "save_steps": 10000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.788114875766555e+16, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
