| { | |
| "best_metric": 0.8256880733944955, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-3162", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 3689, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.095917701721191, | |
| "learning_rate": 0.0006378820903022044, | |
| "loss": 1.3367, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8245412844036697, | |
| "eval_loss": 1.8473267555236816, | |
| "eval_runtime": 2.651, | |
| "eval_samples_per_second": 328.932, | |
| "eval_steps_per_second": 2.641, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.007329940795898, | |
| "learning_rate": 0.0005315684085851703, | |
| "loss": 0.5859, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8130733944954128, | |
| "eval_loss": 1.815151333808899, | |
| "eval_runtime": 2.5701, | |
| "eval_samples_per_second": 339.281, | |
| "eval_steps_per_second": 2.724, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 5.04394006729126, | |
| "learning_rate": 0.00042525472686813623, | |
| "loss": 0.379, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8142201834862385, | |
| "eval_loss": 1.8933576345443726, | |
| "eval_runtime": 2.5998, | |
| "eval_samples_per_second": 335.415, | |
| "eval_steps_per_second": 2.693, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 11.948037147521973, | |
| "learning_rate": 0.0003189410451511022, | |
| "loss": 0.2807, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8130733944954128, | |
| "eval_loss": 1.9181652069091797, | |
| "eval_runtime": 2.6566, | |
| "eval_samples_per_second": 328.243, | |
| "eval_steps_per_second": 2.635, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 6.663651466369629, | |
| "learning_rate": 0.0002128290971754667, | |
| "loss": 0.2134, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8119266055045872, | |
| "eval_loss": 1.9277732372283936, | |
| "eval_runtime": 2.6204, | |
| "eval_samples_per_second": 332.778, | |
| "eval_steps_per_second": 2.671, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.037877082824707, | |
| "learning_rate": 0.00010651541545843262, | |
| "loss": 0.1741, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8256880733944955, | |
| "eval_loss": 1.8173047304153442, | |
| "eval_runtime": 2.6704, | |
| "eval_samples_per_second": 326.541, | |
| "eval_steps_per_second": 2.621, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.2663307189941406, | |
| "learning_rate": 2.017337413985466e-07, | |
| "loss": 0.1397, | |
| "step": 3689 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8222477064220184, | |
| "eval_loss": 1.8090132474899292, | |
| "eval_runtime": 2.6108, | |
| "eval_samples_per_second": 333.996, | |
| "eval_steps_per_second": 2.681, | |
| "step": 3689 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3689, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 56628612372420.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.2549821714207776, | |
| "learning_rate": 0.0007441957720192385, | |
| "num_train_epochs": 7, | |
| "temperature": 19 | |
| } | |
| } | |