| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 3180, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6691791415214539, | |
| "learning_rate": 1.8006289308176103e-05, | |
| "loss": 0.4586, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6229032258064516, | |
| "eval_loss": 0.22628666460514069, | |
| "eval_runtime": 2.6808, | |
| "eval_samples_per_second": 1156.356, | |
| "eval_steps_per_second": 24.246, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.5440394282341003, | |
| "learning_rate": 1.6012578616352204e-05, | |
| "loss": 0.186, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.10251960903406143, | |
| "eval_runtime": 2.7874, | |
| "eval_samples_per_second": 1112.152, | |
| "eval_steps_per_second": 23.319, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.5485900044441223, | |
| "learning_rate": 1.4018867924528304e-05, | |
| "loss": 0.1103, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8877419354838709, | |
| "eval_loss": 0.06421811133623123, | |
| "eval_runtime": 2.7299, | |
| "eval_samples_per_second": 1135.562, | |
| "eval_steps_per_second": 23.81, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.39198654890060425, | |
| "learning_rate": 1.2025157232704403e-05, | |
| "loss": 0.0795, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9019354838709678, | |
| "eval_loss": 0.047090593725442886, | |
| "eval_runtime": 2.8587, | |
| "eval_samples_per_second": 1084.393, | |
| "eval_steps_per_second": 22.737, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.2942151129245758, | |
| "learning_rate": 1.0031446540880504e-05, | |
| "loss": 0.0642, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9148387096774193, | |
| "eval_loss": 0.038098402321338654, | |
| "eval_runtime": 3.089, | |
| "eval_samples_per_second": 1003.57, | |
| "eval_steps_per_second": 21.043, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.31143322587013245, | |
| "learning_rate": 8.037735849056606e-06, | |
| "loss": 0.0554, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9216129032258065, | |
| "eval_loss": 0.03310990333557129, | |
| "eval_runtime": 3.2922, | |
| "eval_samples_per_second": 941.618, | |
| "eval_steps_per_second": 19.744, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.35357871651649475, | |
| "learning_rate": 6.044025157232704e-06, | |
| "loss": 0.0498, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.925483870967742, | |
| "eval_loss": 0.03013771027326584, | |
| "eval_runtime": 3.905, | |
| "eval_samples_per_second": 793.849, | |
| "eval_steps_per_second": 16.645, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.31530439853668213, | |
| "learning_rate": 4.0503144654088055e-06, | |
| "loss": 0.0464, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.927741935483871, | |
| "eval_loss": 0.028143620118498802, | |
| "eval_runtime": 3.2035, | |
| "eval_samples_per_second": 967.697, | |
| "eval_steps_per_second": 20.29, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.4920739531517029, | |
| "learning_rate": 2.056603773584906e-06, | |
| "loss": 0.0443, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9303225806451613, | |
| "eval_loss": 0.02711585909128189, | |
| "eval_runtime": 3.0854, | |
| "eval_samples_per_second": 1004.729, | |
| "eval_steps_per_second": 21.067, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.28209543228149414, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0433, | |
| "step": 3170 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 825254092458012.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.3431028024543458, | |
| "num_train_epochs": 10, | |
| "temperature": 5 | |
| } | |
| } | |