| { |
| "best_global_step": 1980, |
| "best_metric": 3.208970546722412, |
| "best_model_checkpoint": null, |
| "epoch": 4.040404040404041, |
| "eval_steps": 500, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00202020202020202, |
| "grad_norm": 29.40569305419922, |
| "learning_rate": 0.0, |
| "loss": 10.3751, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.1550122499465942, |
| "learning_rate": 9.88e-05, |
| "loss": 4.808, |
| "step": 495 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.7857861518859863, |
| "eval_runtime": 19.6646, |
| "eval_samples_per_second": 1616.608, |
| "eval_steps_per_second": 6.357, |
| "step": 495 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.0673727989196777, |
| "learning_rate": 9.478678038379531e-05, |
| "loss": 3.5376, |
| "step": 990 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.434614658355713, |
| "eval_runtime": 19.6542, |
| "eval_samples_per_second": 1617.466, |
| "eval_steps_per_second": 6.36, |
| "step": 990 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.007203459739685, |
| "learning_rate": 8.950959488272922e-05, |
| "loss": 3.2817, |
| "step": 1485 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 3.2914164066314697, |
| "eval_runtime": 19.557, |
| "eval_samples_per_second": 1625.503, |
| "eval_steps_per_second": 6.392, |
| "step": 1485 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.060210108757019, |
| "learning_rate": 8.42324093816631e-05, |
| "loss": 3.1383, |
| "step": 1980 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 3.208970546722412, |
| "eval_runtime": 19.9027, |
| "eval_samples_per_second": 1597.274, |
| "eval_steps_per_second": 6.281, |
| "step": 1980 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9900, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 20, |
| "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": 3.3406970167296e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
