| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.7272727272727275, |
| "eval_steps": 500, |
| "global_step": 30, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.45454545454545453, |
| "grad_norm": 2.859375, |
| "learning_rate": 2.610496780537124e-05, |
| "loss": 1.6328, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.9090909090909091, |
| "grad_norm": 2.734375, |
| "learning_rate": 5.873617756208528e-05, |
| "loss": 1.4944, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.9090909090909091, |
| "eval_loss": 1.4644855260849, |
| "eval_runtime": 5.2708, |
| "eval_samples_per_second": 26.751, |
| "eval_steps_per_second": 26.751, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.3636363636363638, |
| "grad_norm": 2.640625, |
| "learning_rate": 9.136738731879934e-05, |
| "loss": 1.3533, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.8181818181818183, |
| "grad_norm": 2.109375, |
| "learning_rate": 0.00012399859707551338, |
| "loss": 1.2788, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.8181818181818183, |
| "eval_loss": 1.3469083309173584, |
| "eval_runtime": 5.1673, |
| "eval_samples_per_second": 27.287, |
| "eval_steps_per_second": 27.287, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.2727272727272725, |
| "grad_norm": 2.265625, |
| "learning_rate": 0.00015662980683222745, |
| "loss": 1.1351, |
| "step": 25 |
| }, |
| { |
| "epoch": 2.7272727272727275, |
| "grad_norm": 2.25, |
| "learning_rate": 0.0001892610165889415, |
| "loss": 1.041, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.7272727272727275, |
| "eval_loss": 1.3093161582946777, |
| "eval_runtime": 5.1555, |
| "eval_samples_per_second": 27.349, |
| "eval_steps_per_second": 27.349, |
| "step": 30 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 33, |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.5295163362443264e+16, |
| "train_batch_size": 140, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
