| { | |
| "best_metric": 23.978183729793663, | |
| "best_model_checkpoint": "./openai/whisper-small-gom-LDC-v1.non-repeating/checkpoint-2000", | |
| "epoch": 2.73972602739726, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.52e-07, | |
| "loss": 2.5215, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.52e-07, | |
| "loss": 2.0216, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.152e-06, | |
| "loss": 1.5534, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.552e-06, | |
| "loss": 1.1894, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.9519999999999997e-06, | |
| "loss": 0.9816, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.3519999999999997e-06, | |
| "loss": 0.9194, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.7519999999999997e-06, | |
| "loss": 0.8616, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.152e-06, | |
| "loss": 0.7917, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.552e-06, | |
| "loss": 0.7533, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.952e-06, | |
| "loss": 0.6568, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.352e-06, | |
| "loss": 0.6336, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.7519999999999996e-06, | |
| "loss": 0.6031, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.152e-06, | |
| "loss": 0.5941, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.5519999999999995e-06, | |
| "loss": 0.5556, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 5.952e-06, | |
| "loss": 0.4818, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 6.352e-06, | |
| "loss": 0.4068, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 6.751999999999999e-06, | |
| "loss": 0.3764, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.1519999999999995e-06, | |
| "loss": 0.3416, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.551999999999999e-06, | |
| "loss": 0.321, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 7.952e-06, | |
| "loss": 0.2987, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.3088768422603607, | |
| "eval_runtime": 1198.5779, | |
| "eval_samples_per_second": 1.73, | |
| "eval_steps_per_second": 0.217, | |
| "eval_wer": 54.862662636351686, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 7.89585798816568e-06, | |
| "loss": 0.2955, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.777514792899408e-06, | |
| "loss": 0.3082, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 7.659171597633136e-06, | |
| "loss": 0.25, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 7.540828402366864e-06, | |
| "loss": 0.2989, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.422485207100592e-06, | |
| "loss": 0.2643, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 7.30414201183432e-06, | |
| "loss": 0.2576, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 7.185798816568046e-06, | |
| "loss": 0.2646, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 7.0674556213017745e-06, | |
| "loss": 0.2231, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6.949112426035503e-06, | |
| "loss": 0.2104, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.83076923076923e-06, | |
| "loss": 0.1716, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 6.712426035502958e-06, | |
| "loss": 0.1727, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.594082840236686e-06, | |
| "loss": 0.155, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 6.475739644970413e-06, | |
| "loss": 0.2016, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 6.3573964497041415e-06, | |
| "loss": 0.1728, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.23905325443787e-06, | |
| "loss": 0.1752, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 6.120710059171597e-06, | |
| "loss": 0.1408, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 6.002366863905325e-06, | |
| "loss": 0.1729, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 5.884023668639053e-06, | |
| "loss": 0.1163, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 5.76568047337278e-06, | |
| "loss": 0.1382, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 5.6473372781065085e-06, | |
| "loss": 0.1506, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.1685582995414734, | |
| "eval_runtime": 1145.8994, | |
| "eval_samples_per_second": 1.81, | |
| "eval_steps_per_second": 0.227, | |
| "eval_wer": 35.832566697332105, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 5.528994082840237e-06, | |
| "loss": 0.1395, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.410650887573964e-06, | |
| "loss": 0.1616, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.292307692307692e-06, | |
| "loss": 0.1542, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 5.17396449704142e-06, | |
| "loss": 0.1315, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 5.055621301775148e-06, | |
| "loss": 0.114, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.9372781065088755e-06, | |
| "loss": 0.1127, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.818934911242604e-06, | |
| "loss": 0.1258, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 4.700591715976332e-06, | |
| "loss": 0.1195, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.582248520710059e-06, | |
| "loss": 0.1189, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 4.463905325443787e-06, | |
| "loss": 0.1496, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.345562130177515e-06, | |
| "loss": 0.1282, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 4.2272189349112425e-06, | |
| "loss": 0.1212, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.108875739644971e-06, | |
| "loss": 0.1013, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.990532544378698e-06, | |
| "loss": 0.1151, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 3.872189349112426e-06, | |
| "loss": 0.1096, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.7538461538461537e-06, | |
| "loss": 0.1127, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 3.6355029585798814e-06, | |
| "loss": 0.0894, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 3.5171597633136095e-06, | |
| "loss": 0.095, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 3.3988165680473372e-06, | |
| "loss": 0.0865, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.280473372781065e-06, | |
| "loss": 0.0669, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.1228846088051796, | |
| "eval_runtime": 1086.3387, | |
| "eval_samples_per_second": 1.909, | |
| "eval_steps_per_second": 0.239, | |
| "eval_wer": 25.535550006571167, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.162130177514793e-06, | |
| "loss": 0.0664, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.0437869822485203e-06, | |
| "loss": 0.0908, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.9254437869822484e-06, | |
| "loss": 0.0796, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.807100591715976e-06, | |
| "loss": 0.0595, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.688757396449704e-06, | |
| "loss": 0.0717, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 2.570414201183432e-06, | |
| "loss": 0.0519, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 2.4520710059171596e-06, | |
| "loss": 0.0612, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.3337278106508873e-06, | |
| "loss": 0.0756, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.2153846153846154e-06, | |
| "loss": 0.0775, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.097041420118343e-06, | |
| "loss": 0.0676, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 1.978698224852071e-06, | |
| "loss": 0.0764, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.860355029585799e-06, | |
| "loss": 0.0677, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.7420118343195264e-06, | |
| "loss": 0.0576, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 1.6236686390532543e-06, | |
| "loss": 0.0557, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 1.5053254437869822e-06, | |
| "loss": 0.0558, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.38698224852071e-06, | |
| "loss": 0.0644, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.2686390532544378e-06, | |
| "loss": 0.0775, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.1502958579881657e-06, | |
| "loss": 0.0665, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.0319526627218936e-06, | |
| "loss": 0.0657, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 9.136094674556213e-07, | |
| "loss": 0.0613, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 0.10533993691205978, | |
| "eval_runtime": 1099.5879, | |
| "eval_samples_per_second": 1.886, | |
| "eval_steps_per_second": 0.236, | |
| "eval_wer": 23.978183729793663, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 2190, | |
| "num_train_epochs": 3, | |
| "total_flos": 9.23242412703744e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |