| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 2226, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6747176051139832, | |
| "learning_rate": 1.7151841868823e-05, | |
| "loss": 0.4394, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5922580645161291, | |
| "eval_loss": 0.2238195538520813, | |
| "eval_runtime": 1.3658, | |
| "eval_samples_per_second": 2269.765, | |
| "eval_steps_per_second": 47.592, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.5481359362602234, | |
| "learning_rate": 1.4303683737646003e-05, | |
| "loss": 0.1854, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.83, | |
| "eval_loss": 0.10747804492712021, | |
| "eval_runtime": 1.3823, | |
| "eval_samples_per_second": 2242.709, | |
| "eval_steps_per_second": 47.025, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.48537126183509827, | |
| "learning_rate": 1.1455525606469004e-05, | |
| "loss": 0.1148, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8774193548387097, | |
| "eval_loss": 0.07071655243635178, | |
| "eval_runtime": 1.5824, | |
| "eval_samples_per_second": 1959.095, | |
| "eval_steps_per_second": 41.078, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.453905314207077, | |
| "learning_rate": 8.607367475292004e-06, | |
| "loss": 0.0866, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8964516129032258, | |
| "eval_loss": 0.05417931452393532, | |
| "eval_runtime": 1.4031, | |
| "eval_samples_per_second": 2209.342, | |
| "eval_steps_per_second": 46.325, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.3496621549129486, | |
| "learning_rate": 5.759209344115006e-06, | |
| "loss": 0.0723, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9032258064516129, | |
| "eval_loss": 0.04600543528795242, | |
| "eval_runtime": 1.4121, | |
| "eval_samples_per_second": 2195.311, | |
| "eval_steps_per_second": 46.031, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.40788304805755615, | |
| "learning_rate": 2.911051212938006e-06, | |
| "loss": 0.0647, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9074193548387097, | |
| "eval_loss": 0.04155374690890312, | |
| "eval_runtime": 1.4178, | |
| "eval_samples_per_second": 2186.495, | |
| "eval_steps_per_second": 45.846, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.26997336745262146, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0612, | |
| "step": 2219 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 578219881979544.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.8262279517048804, | |
| "num_train_epochs": 7, | |
| "temperature": 6 | |
| } | |
| } | |