| { | |
| "best_metric": 52.66343825665859, | |
| "best_model_checkpoint": "./checkpoint-200", | |
| "epoch": 4.752941176470588, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 4.3134, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 6e-06, | |
| "loss": 2.9329, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 2.0584, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 9.829787234042554e-06, | |
| "loss": 1.566, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 9.617021276595745e-06, | |
| "loss": 1.3777, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 9.404255319148937e-06, | |
| "loss": 1.1469, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 9.191489361702128e-06, | |
| "loss": 1.0638, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 8.97872340425532e-06, | |
| "loss": 0.9974, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 8.765957446808512e-06, | |
| "loss": 0.9615, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 8.553191489361703e-06, | |
| "loss": 0.8262, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_cer": 149.33157314260887, | |
| "eval_loss": 0.8188337683677673, | |
| "eval_runtime": 459.9395, | |
| "eval_samples_per_second": 1.113, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 146.85230024213075, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 8.340425531914894e-06, | |
| "loss": 0.7986, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 8.127659574468085e-06, | |
| "loss": 0.7707, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.914893617021278e-06, | |
| "loss": 0.7623, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 7.702127659574469e-06, | |
| "loss": 0.6451, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 7.48936170212766e-06, | |
| "loss": 0.6205, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 7.2765957446808524e-06, | |
| "loss": 0.6097, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.0638297872340434e-06, | |
| "loss": 0.6217, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.8510638297872344e-06, | |
| "loss": 0.4584, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 6.6382978723404254e-06, | |
| "loss": 0.4929, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 6.425531914893618e-06, | |
| "loss": 0.4843, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_cer": 22.099862972494236, | |
| "eval_loss": 0.6699215173721313, | |
| "eval_runtime": 1062.3991, | |
| "eval_samples_per_second": 0.482, | |
| "eval_steps_per_second": 0.06, | |
| "eval_wer": 52.66343825665859, | |
| "step": 200 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 12, | |
| "total_flos": 3.72044298092544e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |