| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5938709677419355, | |
| "eval_loss": 0.19997140765190125, | |
| "eval_runtime": 5.4307, | |
| "eval_samples_per_second": 570.829, | |
| "eval_steps_per_second": 11.969, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5329213738441467, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.3229, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8238709677419355, | |
| "eval_loss": 0.09559512138366699, | |
| "eval_runtime": 5.9449, | |
| "eval_samples_per_second": 521.456, | |
| "eval_steps_per_second": 10.934, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8812903225806452, | |
| "eval_loss": 0.06311405450105667, | |
| "eval_runtime": 5.4649, | |
| "eval_samples_per_second": 567.253, | |
| "eval_steps_per_second": 11.894, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.44504690170288086, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1111, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9006451612903226, | |
| "eval_loss": 0.048286207020282745, | |
| "eval_runtime": 5.6179, | |
| "eval_samples_per_second": 551.809, | |
| "eval_steps_per_second": 11.57, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3021944761276245, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0699, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9103225806451613, | |
| "eval_loss": 0.03919174522161484, | |
| "eval_runtime": 5.7364, | |
| "eval_samples_per_second": 540.408, | |
| "eval_steps_per_second": 11.331, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9132258064516129, | |
| "eval_loss": 0.03407873958349228, | |
| "eval_runtime": 5.4521, | |
| "eval_samples_per_second": 568.585, | |
| "eval_steps_per_second": 11.922, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.2565140724182129, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0542, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 520991326672152.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.16854985287286628, | |
| "num_train_epochs": 10, | |
| "temperature": 10 | |
| } | |
| } | |