| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.022298456260720412, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005717552887364208, | |
| "eval_loss": 8.388371467590332, | |
| "eval_runtime": 101.2069, | |
| "eval_samples_per_second": 7.282, | |
| "eval_steps_per_second": 3.646, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002858776443682104, | |
| "grad_norm": 7.614388942718506, | |
| "learning_rate": 5e-05, | |
| "loss": 7.9936, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.005717552887364208, | |
| "grad_norm": 10.528021812438965, | |
| "learning_rate": 0.0001, | |
| "loss": 6.6528, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005717552887364208, | |
| "eval_loss": 5.3022379875183105, | |
| "eval_runtime": 107.0555, | |
| "eval_samples_per_second": 6.884, | |
| "eval_steps_per_second": 3.447, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008576329331046312, | |
| "grad_norm": 17.963598251342773, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.4307, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011435105774728416, | |
| "grad_norm": 7.288106918334961, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.3588, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011435105774728416, | |
| "eval_loss": 1.4997512102127075, | |
| "eval_runtime": 107.1778, | |
| "eval_samples_per_second": 6.876, | |
| "eval_steps_per_second": 3.443, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01429388221841052, | |
| "grad_norm": 3.5524449348449707, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.1797, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.017152658662092625, | |
| "grad_norm": 5.827010631561279, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9964, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017152658662092625, | |
| "eval_loss": 1.1370601654052734, | |
| "eval_runtime": 107.098, | |
| "eval_samples_per_second": 6.882, | |
| "eval_steps_per_second": 3.445, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.020011435105774727, | |
| "grad_norm": 6.282993316650391, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.0478, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2631125277016064e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |