| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.6623376623376624, |
| "eval_steps": 500, |
| "global_step": 40, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2077922077922078, |
| "grad_norm": 0.1298828125, |
| "learning_rate": 0.0001, |
| "loss": 0.627, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.4155844155844156, |
| "grad_norm": 0.059814453125, |
| "learning_rate": 0.0001, |
| "loss": 0.5575, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.6233766233766234, |
| "grad_norm": 0.0498046875, |
| "learning_rate": 0.0001, |
| "loss": 0.4482, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.8311688311688312, |
| "grad_norm": 0.032958984375, |
| "learning_rate": 0.0001, |
| "loss": 0.3657, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.0389610389610389, |
| "grad_norm": 0.038818359375, |
| "learning_rate": 0.0001, |
| "loss": 0.2997, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.2467532467532467, |
| "grad_norm": 0.11279296875, |
| "learning_rate": 0.0001, |
| "loss": 0.2444, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.4545454545454546, |
| "grad_norm": 0.057861328125, |
| "learning_rate": 0.0001, |
| "loss": 0.1786, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.6623376623376624, |
| "grad_norm": 0.049560546875, |
| "learning_rate": 0.0001, |
| "loss": 0.1838, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.6623376623376624, |
| "step": 40, |
| "total_flos": 8.20631894326272e+16, |
| "train_loss": 0.36311745792627337, |
| "train_runtime": 2711.4929, |
| "train_samples_per_second": 0.944, |
| "train_steps_per_second": 0.015 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 40, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 90, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.20631894326272e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
