| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6267741935483871, | |
| "eval_loss": 0.25039198994636536, | |
| "eval_runtime": 5.6635, | |
| "eval_samples_per_second": 547.363, | |
| "eval_steps_per_second": 11.477, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.6028931736946106, | |
| "learning_rate": 1.606918238993711e-05, | |
| "loss": 0.3994, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.84, | |
| "eval_loss": 0.1111660972237587, | |
| "eval_runtime": 6.4316, | |
| "eval_samples_per_second": 481.998, | |
| "eval_steps_per_second": 10.106, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8896774193548387, | |
| "eval_loss": 0.06922899931669235, | |
| "eval_runtime": 5.7532, | |
| "eval_samples_per_second": 538.827, | |
| "eval_steps_per_second": 11.298, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.5288750529289246, | |
| "learning_rate": 1.2138364779874214e-05, | |
| "loss": 0.13, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9009677419354839, | |
| "eval_loss": 0.05165809765458107, | |
| "eval_runtime": 5.7316, | |
| "eval_samples_per_second": 540.863, | |
| "eval_steps_per_second": 11.341, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.35449567437171936, | |
| "learning_rate": 8.207547169811321e-06, | |
| "loss": 0.0785, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9103225806451613, | |
| "eval_loss": 0.04206559807062149, | |
| "eval_runtime": 5.9441, | |
| "eval_samples_per_second": 521.529, | |
| "eval_steps_per_second": 10.935, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9187096774193548, | |
| "eval_loss": 0.0372014120221138, | |
| "eval_runtime": 5.5883, | |
| "eval_samples_per_second": 554.731, | |
| "eval_steps_per_second": 11.631, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.3057487905025482, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0613, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9241935483870968, | |
| "eval_loss": 0.034800298511981964, | |
| "eval_runtime": 5.9008, | |
| "eval_samples_per_second": 525.35, | |
| "eval_steps_per_second": 11.015, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.3128741979598999, | |
| "learning_rate": 3.459119496855346e-07, | |
| "loss": 0.0547, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2544, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.19595942772864172, | |
| "num_train_epochs": 8, | |
| "temperature": 4 | |
| } | |
| } | |