| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 28.571428571428573, | |
| "eval_steps": 1000, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.991371750831604, | |
| "learning_rate": 9.8e-05, | |
| "loss": 3.541, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.3066103756427765, | |
| "learning_rate": 8.96842105263158e-05, | |
| "loss": 2.8609, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_loss": 3.1199309825897217, | |
| "eval_runtime": 29.6116, | |
| "eval_samples_per_second": 23.639, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.7332219481468201, | |
| "learning_rate": 7.915789473684211e-05, | |
| "loss": 2.8625, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 0.29804107546806335, | |
| "learning_rate": 6.863157894736843e-05, | |
| "loss": 2.8589, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "eval_loss": 3.051466941833496, | |
| "eval_runtime": 29.5006, | |
| "eval_samples_per_second": 23.728, | |
| "eval_steps_per_second": 2.983, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 1.4042483568191528, | |
| "learning_rate": 5.810526315789474e-05, | |
| "loss": 2.8584, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 0.9517878890037537, | |
| "learning_rate": 4.7578947368421054e-05, | |
| "loss": 2.8587, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "eval_loss": 2.99222993850708, | |
| "eval_runtime": 29.1002, | |
| "eval_samples_per_second": 24.055, | |
| "eval_steps_per_second": 3.024, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.560330867767334, | |
| "learning_rate": 3.7052631578947375e-05, | |
| "loss": 2.8581, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 0.43670666217803955, | |
| "learning_rate": 2.652631578947369e-05, | |
| "loss": 2.856, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_loss": 3.0176596641540527, | |
| "eval_runtime": 30.2333, | |
| "eval_samples_per_second": 23.153, | |
| "eval_steps_per_second": 2.911, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 1.1510722637176514, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.8548, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 0.9515011310577393, | |
| "learning_rate": 5.4736842105263165e-06, | |
| "loss": 2.8531, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_loss": 3.0313847064971924, | |
| "eval_runtime": 29.8329, | |
| "eval_samples_per_second": 23.464, | |
| "eval_steps_per_second": 2.95, | |
| "eval_wer": 1.0, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 6.487789626604962e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |