| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 1000, |
| "global_step": 3933, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.000508646998982706, |
| "grad_norm": 5.145923137664795, |
| "learning_rate": 0.0, |
| "loss": 3.4988, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.508646998982706, |
| "grad_norm": 3.175929546356201, |
| "learning_rate": 2.7688394875659382e-05, |
| "loss": 2.5805, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.508646998982706, |
| "eval_cosine_accuracy": 0.9643495678901672, |
| "eval_loss": 1.3221157789230347, |
| "eval_runtime": 22.697, |
| "eval_samples_per_second": 418.953, |
| "eval_steps_per_second": 1.674, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0172852058973056, |
| "grad_norm": 3.1664931774139404, |
| "learning_rate": 2.203654860587792e-05, |
| "loss": 1.7853, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.0172852058973056, |
| "eval_cosine_accuracy": 0.9605636596679688, |
| "eval_loss": 1.2634038925170898, |
| "eval_runtime": 22.9567, |
| "eval_samples_per_second": 414.214, |
| "eval_steps_per_second": 1.655, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.5256736146415861, |
| "grad_norm": 3.256283760070801, |
| "learning_rate": 1.6384702336096458e-05, |
| "loss": 1.6637, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.5256736146415861, |
| "eval_cosine_accuracy": 0.9682406187057495, |
| "eval_loss": 1.1969646215438843, |
| "eval_runtime": 22.9019, |
| "eval_samples_per_second": 415.205, |
| "eval_steps_per_second": 1.659, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 5898, |
| "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": 0.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
