| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.98461538461538, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.675e-06, | |
| "loss": 16.3071, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 7.3875e-06, | |
| "loss": 10.3912, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 4.1432, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 3.3835, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 3.1483, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "eval_loss": 3.0498125553131104, | |
| "eval_runtime": 17.1221, | |
| "eval_samples_per_second": 29.728, | |
| "eval_steps_per_second": 3.738, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 18.74, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 3.032, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 2.9869, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 24.98, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 2.9399, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 2.9046, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 2.8449, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "eval_loss": 2.7789549827575684, | |
| "eval_runtime": 16.9653, | |
| "eval_samples_per_second": 30.002, | |
| "eval_steps_per_second": 3.772, | |
| "eval_wer": 0.9493263034563562, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 34.37, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 2.8219, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 37.49, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 2.761, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 40.62, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 2.5482, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 43.74, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 2.1627, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 46.86, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 1.8683, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 46.86, | |
| "eval_loss": 1.233871340751648, | |
| "eval_runtime": 17.2711, | |
| "eval_samples_per_second": 29.471, | |
| "eval_steps_per_second": 3.706, | |
| "eval_wer": 0.8160515524311658, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 49.98, | |
| "learning_rate": 5.988749999999999e-05, | |
| "loss": 1.6679, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 49.98, | |
| "step": 1600, | |
| "total_flos": 6.038839616760506e+18, | |
| "train_loss": 4.119469966888428, | |
| "train_runtime": 2676.0092, | |
| "train_samples_per_second": 19.338, | |
| "train_steps_per_second": 0.598 | |
| } | |
| ], | |
| "max_steps": 1600, | |
| "num_train_epochs": 50, | |
| "total_flos": 6.038839616760506e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |