| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.416856492027335, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 1.5356756448745728, | |
| "learning_rate": 1.7722095671981778e-05, | |
| "loss": 0.1506, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9634930973676268, | |
| "eval_f1": 0.8442987330517893, | |
| "eval_loss": 0.13140565156936646, | |
| "eval_precision": 0.8387987192227007, | |
| "eval_recall": 0.849871350262893, | |
| "eval_runtime": 3.0691, | |
| "eval_samples_per_second": 1058.938, | |
| "eval_steps_per_second": 66.469, | |
| "step": 878 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 3.3906660079956055, | |
| "learning_rate": 1.5444191343963555e-05, | |
| "loss": 0.12, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "grad_norm": 6.772578239440918, | |
| "learning_rate": 1.3166287015945332e-05, | |
| "loss": 0.1046, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9679412839372805, | |
| "eval_f1": 0.866677846721449, | |
| "eval_loss": 0.12156210839748383, | |
| "eval_precision": 0.866145251396648, | |
| "eval_recall": 0.8672110974381922, | |
| "eval_runtime": 3.0579, | |
| "eval_samples_per_second": 1062.822, | |
| "eval_steps_per_second": 66.712, | |
| "step": 1756 | |
| }, | |
| { | |
| "epoch": 2.277904328018223, | |
| "grad_norm": 8.046541213989258, | |
| "learning_rate": 1.0888382687927108e-05, | |
| "loss": 0.098, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.847380410022779, | |
| "grad_norm": 3.613861322402954, | |
| "learning_rate": 8.610478359908885e-06, | |
| "loss": 0.0905, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9694981492366594, | |
| "eval_f1": 0.8722589241103648, | |
| "eval_loss": 0.11584340035915375, | |
| "eval_precision": 0.8657703328190435, | |
| "eval_recall": 0.878845508446135, | |
| "eval_runtime": 3.8011, | |
| "eval_samples_per_second": 855.007, | |
| "eval_steps_per_second": 53.668, | |
| "step": 2634 | |
| }, | |
| { | |
| "epoch": 3.416856492027335, | |
| "grad_norm": 3.454561471939087, | |
| "learning_rate": 6.3325740318906616e-06, | |
| "loss": 0.0841, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4390, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 62399928651750.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |