| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008721881019006433, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00014536468365010721, | |
| "grad_norm": 11.293152809143066, | |
| "learning_rate": 2e-05, | |
| "loss": 5.4024, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00014536468365010721, | |
| "eval_loss": 4.953571796417236, | |
| "eval_runtime": 1311.2116, | |
| "eval_samples_per_second": 2.209, | |
| "eval_steps_per_second": 1.105, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00029072936730021443, | |
| "grad_norm": 10.99425983428955, | |
| "learning_rate": 4e-05, | |
| "loss": 4.8248, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00043609405095032164, | |
| "grad_norm": 11.894266128540039, | |
| "learning_rate": 6e-05, | |
| "loss": 3.6072, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00043609405095032164, | |
| "eval_loss": 4.856856822967529, | |
| "eval_runtime": 1306.8243, | |
| "eval_samples_per_second": 2.217, | |
| "eval_steps_per_second": 1.109, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005814587346004289, | |
| "grad_norm": 9.180171012878418, | |
| "learning_rate": 8e-05, | |
| "loss": 3.6817, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.000726823418250536, | |
| "grad_norm": 12.513172149658203, | |
| "learning_rate": 0.0001, | |
| "loss": 5.2037, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008721881019006433, | |
| "grad_norm": 6.623383045196533, | |
| "learning_rate": 0.00012, | |
| "loss": 2.7651, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008721881019006433, | |
| "eval_loss": 3.960355758666992, | |
| "eval_runtime": 1307.1519, | |
| "eval_samples_per_second": 2.216, | |
| "eval_steps_per_second": 1.109, | |
| "step": 6 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 372738871001088.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |