|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.5161290322580645, |
|
"eval_steps": 1000, |
|
"global_step": 2000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 2.5e-08, |
|
"loss": 0.7748, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 5e-08, |
|
"loss": 1.2926, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 7.5e-08, |
|
"loss": 0.7723, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 1e-07, |
|
"loss": 1.3901, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 1.25e-07, |
|
"loss": 0.7494, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 1.5e-07, |
|
"loss": 1.1551, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 1.75e-07, |
|
"loss": 0.5737, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 2e-07, |
|
"loss": 0.7115, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 2.25e-07, |
|
"loss": 0.2719, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 2.5e-07, |
|
"loss": 0.4535, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 2.75e-07, |
|
"loss": 0.2216, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 3e-07, |
|
"loss": 0.3218, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 3.25e-07, |
|
"loss": 0.1867, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 3.5e-07, |
|
"loss": 0.34, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 3.75e-07, |
|
"loss": 0.1971, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 4e-07, |
|
"loss": 0.2871, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 4.2499999999999995e-07, |
|
"loss": 0.1857, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 4.5e-07, |
|
"loss": 0.247, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 4.7499999999999995e-07, |
|
"loss": 0.167, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 5e-07, |
|
"loss": 0.2332, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 5.25e-07, |
|
"loss": 0.1908, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 5.5e-07, |
|
"loss": 0.2329, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 5.749999999999999e-07, |
|
"loss": 0.1589, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 6e-07, |
|
"loss": 0.2232, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 6.249999999999999e-07, |
|
"loss": 0.166, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 6.5e-07, |
|
"loss": 0.2115, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 6.75e-07, |
|
"loss": 0.1378, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7e-07, |
|
"loss": 0.1946, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.249999999999999e-07, |
|
"loss": 0.1553, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.5e-07, |
|
"loss": 0.2011, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 7.75e-07, |
|
"loss": 0.1491, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 8e-07, |
|
"loss": 0.1517, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 8.249999999999999e-07, |
|
"loss": 0.1266, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 8.499999999999999e-07, |
|
"loss": 0.1926, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 8.75e-07, |
|
"loss": 0.1404, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 9e-07, |
|
"loss": 0.1729, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 9.25e-07, |
|
"loss": 0.137, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 9.499999999999999e-07, |
|
"loss": 0.1753, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 9.75e-07, |
|
"loss": 0.1408, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 1e-06, |
|
"loss": 0.1819, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"eval_loss": 0.14837130904197693, |
|
"eval_runtime": 179172.9142, |
|
"eval_samples_per_second": 0.052, |
|
"eval_steps_per_second": 0.007, |
|
"eval_wer": 0.10759534583063995, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 9.9375e-07, |
|
"loss": 0.144, |
|
"step": 1025 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 9.875e-07, |
|
"loss": 0.2088, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 9.8125e-07, |
|
"loss": 0.1592, |
|
"step": 1075 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 9.75e-07, |
|
"loss": 0.1775, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 9.6875e-07, |
|
"loss": 0.1345, |
|
"step": 1125 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 9.624999999999999e-07, |
|
"loss": 0.1857, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 9.5625e-07, |
|
"loss": 0.1481, |
|
"step": 1175 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 9.499999999999999e-07, |
|
"loss": 0.1926, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 9.4375e-07, |
|
"loss": 0.1298, |
|
"step": 1225 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 9.374999999999999e-07, |
|
"loss": 0.193, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 9.3125e-07, |
|
"loss": 0.1336, |
|
"step": 1275 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 9.25e-07, |
|
"loss": 0.1697, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 9.187499999999999e-07, |
|
"loss": 0.1332, |
|
"step": 1325 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 9.124999999999999e-07, |
|
"loss": 0.1668, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 9.0625e-07, |
|
"loss": 0.1425, |
|
"step": 1375 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 9e-07, |
|
"loss": 0.1815, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 8.9375e-07, |
|
"loss": 0.1579, |
|
"step": 1425 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 8.874999999999999e-07, |
|
"loss": 0.1855, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 8.812499999999999e-07, |
|
"loss": 0.1359, |
|
"step": 1475 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 8.75e-07, |
|
"loss": 0.1923, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 8.687499999999999e-07, |
|
"loss": 0.126, |
|
"step": 1525 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 8.625e-07, |
|
"loss": 0.1824, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 8.5625e-07, |
|
"loss": 0.1286, |
|
"step": 1575 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 8.499999999999999e-07, |
|
"loss": 0.1776, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 8.4375e-07, |
|
"loss": 0.1408, |
|
"step": 1625 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 8.375e-07, |
|
"loss": 0.1684, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 8.3125e-07, |
|
"loss": 0.1511, |
|
"step": 1675 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 8.249999999999999e-07, |
|
"loss": 0.1579, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 8.187499999999999e-07, |
|
"loss": 0.1303, |
|
"step": 1725 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 8.125e-07, |
|
"loss": 0.1663, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 8.0625e-07, |
|
"loss": 0.1466, |
|
"step": 1775 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 8e-07, |
|
"loss": 0.1738, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 7.937499999999999e-07, |
|
"loss": 0.1324, |
|
"step": 1825 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 7.875e-07, |
|
"loss": 0.1775, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 7.812499999999999e-07, |
|
"loss": 0.0969, |
|
"step": 1875 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 7.75e-07, |
|
"loss": 0.172, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 7.6875e-07, |
|
"loss": 0.1633, |
|
"step": 1925 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 7.624999999999999e-07, |
|
"loss": 0.1695, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 7.5625e-07, |
|
"loss": 0.1374, |
|
"step": 1975 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 7.5e-07, |
|
"loss": 0.1703, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"eval_loss": 0.13918790221214294, |
|
"eval_runtime": 176892.8221, |
|
"eval_samples_per_second": 0.053, |
|
"eval_steps_per_second": 0.007, |
|
"eval_wer": 0.10190691661279896, |
|
"step": 2000 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 5000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 1000, |
|
"total_flos": 5.435997290496e+19, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|