| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.99830220713073, | |
| "global_step": 14700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 20.4331, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 7.425e-06, | |
| "loss": 11.9236, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.1174999999999999e-05, | |
| "loss": 8.0522, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.4925e-05, | |
| "loss": 6.5594, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 1.8675e-05, | |
| "loss": 5.3156, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 4.558254241943359, | |
| "eval_runtime": 60.0958, | |
| "eval_samples_per_second": 34.861, | |
| "eval_steps_per_second": 4.36, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 2.2424999999999996e-05, | |
| "loss": 4.3699, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 2.6174999999999996e-05, | |
| "loss": 3.7573, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 2.9925e-05, | |
| "loss": 3.5165, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.3675e-05, | |
| "loss": 3.4516, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 3.7424999999999995e-05, | |
| "loss": 3.3329, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 3.427361011505127, | |
| "eval_runtime": 59.6683, | |
| "eval_samples_per_second": 35.111, | |
| "eval_steps_per_second": 4.391, | |
| "eval_wer": 1.0001231148045553, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 4.1175e-05, | |
| "loss": 3.2764, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 4.4924999999999994e-05, | |
| "loss": 3.2324, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 4.8675e-05, | |
| "loss": 3.0751, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 2.6121, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 2.1275, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "eval_loss": 1.7220778465270996, | |
| "eval_runtime": 60.7788, | |
| "eval_samples_per_second": 34.469, | |
| "eval_steps_per_second": 4.311, | |
| "eval_wer": 0.876269621421976, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 1.9104, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 6.367499999999999e-05, | |
| "loss": 1.7621, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 1.687, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 7.1175e-05, | |
| "loss": 1.6241, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 7.492499999999999e-05, | |
| "loss": 1.5737, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "eval_loss": 1.418828010559082, | |
| "eval_runtime": 60.9775, | |
| "eval_samples_per_second": 34.357, | |
| "eval_steps_per_second": 4.297, | |
| "eval_wer": 0.8143428747306863, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 7.442125984251968e-05, | |
| "loss": 1.5115, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 7.383070866141732e-05, | |
| "loss": 1.4744, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 7.324015748031495e-05, | |
| "loss": 1.4613, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 16.33, | |
| "learning_rate": 7.264960629921259e-05, | |
| "loss": 1.4378, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 7.206496062992125e-05, | |
| "loss": 1.3835, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "eval_loss": 1.2250615358352661, | |
| "eval_runtime": 59.6142, | |
| "eval_samples_per_second": 35.143, | |
| "eval_steps_per_second": 4.395, | |
| "eval_wer": 0.7447214527546937, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 7.14744094488189e-05, | |
| "loss": 1.3824, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.37, | |
| "learning_rate": 7.088385826771653e-05, | |
| "loss": 1.3698, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 7.029330708661417e-05, | |
| "loss": 1.3394, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 6.970866141732283e-05, | |
| "loss": 1.3291, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "learning_rate": 6.911811023622046e-05, | |
| "loss": 1.3247, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "eval_loss": 1.2827171087265015, | |
| "eval_runtime": 60.3104, | |
| "eval_samples_per_second": 34.737, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.7394275161588181, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.09, | |
| "learning_rate": 6.85275590551181e-05, | |
| "loss": 1.2949, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "learning_rate": 6.793700787401574e-05, | |
| "loss": 1.2651, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 22.45, | |
| "learning_rate": 6.734645669291339e-05, | |
| "loss": 1.2827, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 23.13, | |
| "learning_rate": 6.675590551181102e-05, | |
| "loss": 1.2618, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "learning_rate": 6.616535433070866e-05, | |
| "loss": 1.231, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "eval_loss": 1.2216109037399292, | |
| "eval_runtime": 59.8545, | |
| "eval_samples_per_second": 35.002, | |
| "eval_steps_per_second": 4.377, | |
| "eval_wer": 0.7073561095721761, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.49, | |
| "learning_rate": 6.55748031496063e-05, | |
| "loss": 1.1867, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 25.17, | |
| "learning_rate": 6.498425196850392e-05, | |
| "loss": 1.25, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 6.439370078740157e-05, | |
| "loss": 1.2151, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 26.53, | |
| "learning_rate": 6.38031496062992e-05, | |
| "loss": 1.1807, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 27.21, | |
| "learning_rate": 6.321259842519685e-05, | |
| "loss": 1.1819, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.21, | |
| "eval_loss": 1.2209904193878174, | |
| "eval_runtime": 60.9728, | |
| "eval_samples_per_second": 34.36, | |
| "eval_steps_per_second": 4.297, | |
| "eval_wer": 0.6863034779932287, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 27.89, | |
| "learning_rate": 6.262204724409448e-05, | |
| "loss": 1.1837, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 6.203149606299212e-05, | |
| "loss": 1.1666, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 6.144094488188977e-05, | |
| "loss": 1.1337, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 29.93, | |
| "learning_rate": 6.08503937007874e-05, | |
| "loss": 1.1603, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 30.61, | |
| "learning_rate": 6.0259842519685034e-05, | |
| "loss": 1.1546, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 30.61, | |
| "eval_loss": 1.323254108428955, | |
| "eval_runtime": 59.8005, | |
| "eval_samples_per_second": 35.033, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.7308094798399507, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.29, | |
| "learning_rate": 5.966929133858267e-05, | |
| "loss": 1.1288, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 31.97, | |
| "learning_rate": 5.907874015748031e-05, | |
| "loss": 1.111, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 32.65, | |
| "learning_rate": 5.848818897637795e-05, | |
| "loss": 1.1357, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 5.789763779527559e-05, | |
| "loss": 1.1217, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 5.731299212598425e-05, | |
| "loss": 1.0902, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "eval_loss": 1.3251430988311768, | |
| "eval_runtime": 59.9088, | |
| "eval_samples_per_second": 34.97, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.7009541397353032, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.69, | |
| "learning_rate": 5.672244094488189e-05, | |
| "loss": 1.0936, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 35.37, | |
| "learning_rate": 5.613188976377952e-05, | |
| "loss": 1.0918, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 36.05, | |
| "learning_rate": 5.5541338582677156e-05, | |
| "loss": 1.0831, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 36.73, | |
| "learning_rate": 5.49507874015748e-05, | |
| "loss": 1.0702, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "learning_rate": 5.436023622047244e-05, | |
| "loss": 1.0749, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "eval_loss": 1.3274425268173218, | |
| "eval_runtime": 59.5105, | |
| "eval_samples_per_second": 35.204, | |
| "eval_steps_per_second": 4.403, | |
| "eval_wer": 0.7234841489689136, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "learning_rate": 5.3775590551181096e-05, | |
| "loss": 1.0629, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 38.77, | |
| "learning_rate": 5.318503937007874e-05, | |
| "loss": 1.0426, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 39.46, | |
| "learning_rate": 5.259448818897638e-05, | |
| "loss": 1.0421, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 40.14, | |
| "learning_rate": 5.200393700787401e-05, | |
| "loss": 1.0551, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 40.81, | |
| "learning_rate": 5.1413385826771645e-05, | |
| "loss": 1.0412, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 40.81, | |
| "eval_loss": 1.29422128200531, | |
| "eval_runtime": 60.0962, | |
| "eval_samples_per_second": 34.861, | |
| "eval_steps_per_second": 4.36, | |
| "eval_wer": 0.6856263465681748, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.5, | |
| "learning_rate": 5.0822834645669286e-05, | |
| "loss": 1.0112, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 42.18, | |
| "learning_rate": 5.0232283464566926e-05, | |
| "loss": 1.0332, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "learning_rate": 4.964173228346457e-05, | |
| "loss": 1.0179, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 43.54, | |
| "learning_rate": 4.90511811023622e-05, | |
| "loss": 0.9993, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 44.22, | |
| "learning_rate": 4.846062992125984e-05, | |
| "loss": 1.0064, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.22, | |
| "eval_loss": 1.258121371269226, | |
| "eval_runtime": 59.6202, | |
| "eval_samples_per_second": 35.139, | |
| "eval_steps_per_second": 4.394, | |
| "eval_wer": 0.6731917513080948, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.9, | |
| "learning_rate": 4.7870078740157475e-05, | |
| "loss": 1.0096, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 45.58, | |
| "learning_rate": 4.727952755905511e-05, | |
| "loss": 1.0142, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 46.26, | |
| "learning_rate": 4.668897637795275e-05, | |
| "loss": 0.9758, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 46.94, | |
| "learning_rate": 4.609842519685039e-05, | |
| "loss": 0.9863, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "learning_rate": 4.550787401574803e-05, | |
| "loss": 1.0006, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "eval_loss": 1.2766607999801636, | |
| "eval_runtime": 60.9279, | |
| "eval_samples_per_second": 34.385, | |
| "eval_steps_per_second": 4.3, | |
| "eval_wer": 0.6884579870729455, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 48.3, | |
| "learning_rate": 4.4917322834645664e-05, | |
| "loss": 0.9802, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 48.98, | |
| "learning_rate": 4.4326771653543305e-05, | |
| "loss": 0.9793, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 49.66, | |
| "learning_rate": 4.3736220472440945e-05, | |
| "loss": 0.9972, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 50.34, | |
| "learning_rate": 4.314566929133857e-05, | |
| "loss": 0.9695, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 51.02, | |
| "learning_rate": 4.255511811023621e-05, | |
| "loss": 0.9518, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 51.02, | |
| "eval_loss": 1.2966111898422241, | |
| "eval_runtime": 60.0454, | |
| "eval_samples_per_second": 34.89, | |
| "eval_steps_per_second": 4.363, | |
| "eval_wer": 0.6925207756232687, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 51.7, | |
| "learning_rate": 4.1964566929133854e-05, | |
| "loss": 0.9623, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 52.38, | |
| "learning_rate": 4.1374015748031494e-05, | |
| "loss": 0.9595, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 53.06, | |
| "learning_rate": 4.078346456692913e-05, | |
| "loss": 0.9442, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 53.74, | |
| "learning_rate": 4.019291338582677e-05, | |
| "loss": 0.9311, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 54.42, | |
| "learning_rate": 3.960236220472441e-05, | |
| "loss": 0.9514, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 54.42, | |
| "eval_loss": 1.2980769872665405, | |
| "eval_runtime": 59.9476, | |
| "eval_samples_per_second": 34.947, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.7066789781471222, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 55.1, | |
| "learning_rate": 3.901181102362205e-05, | |
| "loss": 0.9373, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 55.78, | |
| "learning_rate": 3.842125984251968e-05, | |
| "loss": 0.923, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 56.46, | |
| "learning_rate": 3.783070866141732e-05, | |
| "loss": 0.9034, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "learning_rate": 3.724015748031496e-05, | |
| "loss": 0.9385, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 57.82, | |
| "learning_rate": 3.664960629921259e-05, | |
| "loss": 0.9241, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 57.82, | |
| "eval_loss": 1.3834576606750488, | |
| "eval_runtime": 62.5486, | |
| "eval_samples_per_second": 33.494, | |
| "eval_steps_per_second": 4.189, | |
| "eval_wer": 0.7124038165589412, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 58.5, | |
| "learning_rate": 3.605905511811023e-05, | |
| "loss": 0.9117, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 59.18, | |
| "learning_rate": 3.546850393700787e-05, | |
| "loss": 0.9159, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 59.86, | |
| "learning_rate": 3.4877952755905506e-05, | |
| "loss": 0.9087, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 60.54, | |
| "learning_rate": 3.428740157480315e-05, | |
| "loss": 0.8926, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 61.22, | |
| "learning_rate": 3.369685039370079e-05, | |
| "loss": 0.9059, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 61.22, | |
| "eval_loss": 1.3317581415176392, | |
| "eval_runtime": 60.468, | |
| "eval_samples_per_second": 34.646, | |
| "eval_steps_per_second": 4.333, | |
| "eval_wer": 0.7082794706063404, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 61.9, | |
| "learning_rate": 3.310629921259842e-05, | |
| "loss": 0.8938, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 62.58, | |
| "learning_rate": 3.2515748031496055e-05, | |
| "loss": 0.8958, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 63.26, | |
| "learning_rate": 3.1925196850393696e-05, | |
| "loss": 0.8745, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 63.94, | |
| "learning_rate": 3.1334645669291336e-05, | |
| "loss": 0.8912, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 64.62, | |
| "learning_rate": 3.074409448818898e-05, | |
| "loss": 0.8906, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 64.62, | |
| "eval_loss": 1.3640189170837402, | |
| "eval_runtime": 60.5502, | |
| "eval_samples_per_second": 34.599, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.6961526623576485, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 65.31, | |
| "learning_rate": 3.015354330708661e-05, | |
| "loss": 0.8696, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 65.98, | |
| "learning_rate": 2.9562992125984248e-05, | |
| "loss": 0.8547, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 2.8978346456692914e-05, | |
| "loss": 0.8759, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 67.35, | |
| "learning_rate": 2.8387795275590547e-05, | |
| "loss": 0.8627, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 68.03, | |
| "learning_rate": 2.780314960629921e-05, | |
| "loss": 0.8468, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 68.03, | |
| "eval_loss": 1.4726500511169434, | |
| "eval_runtime": 61.219, | |
| "eval_samples_per_second": 34.221, | |
| "eval_steps_per_second": 4.28, | |
| "eval_wer": 0.6981840566328101, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 68.71, | |
| "learning_rate": 2.721259842519685e-05, | |
| "loss": 0.8642, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 69.39, | |
| "learning_rate": 2.6622047244094484e-05, | |
| "loss": 0.8622, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 70.07, | |
| "learning_rate": 2.603149606299212e-05, | |
| "loss": 0.8442, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "learning_rate": 2.5440944881889762e-05, | |
| "loss": 0.8295, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "learning_rate": 2.4850393700787402e-05, | |
| "loss": 0.8631, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "eval_loss": 1.3400819301605225, | |
| "eval_runtime": 61.2593, | |
| "eval_samples_per_second": 34.199, | |
| "eval_steps_per_second": 4.277, | |
| "eval_wer": 0.6809479839950754, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 72.11, | |
| "learning_rate": 2.4259842519685036e-05, | |
| "loss": 0.844, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 72.79, | |
| "learning_rate": 2.3669291338582673e-05, | |
| "loss": 0.8245, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 73.47, | |
| "learning_rate": 2.3078740157480314e-05, | |
| "loss": 0.8285, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 74.15, | |
| "learning_rate": 2.248818897637795e-05, | |
| "loss": 0.8661, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 74.83, | |
| "learning_rate": 2.1897637795275588e-05, | |
| "loss": 0.8154, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 74.83, | |
| "eval_loss": 1.4123561382293701, | |
| "eval_runtime": 60.3325, | |
| "eval_samples_per_second": 34.724, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.6955370883348723, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 75.51, | |
| "learning_rate": 2.1307086614173225e-05, | |
| "loss": 0.8111, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 76.19, | |
| "learning_rate": 2.0716535433070866e-05, | |
| "loss": 0.8382, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 76.87, | |
| "learning_rate": 2.0125984251968503e-05, | |
| "loss": 0.8245, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 77.55, | |
| "learning_rate": 1.9535433070866137e-05, | |
| "loss": 0.815, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 78.23, | |
| "learning_rate": 1.8944881889763778e-05, | |
| "loss": 0.7953, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 78.23, | |
| "eval_loss": 1.424477219581604, | |
| "eval_runtime": 61.148, | |
| "eval_samples_per_second": 34.261, | |
| "eval_steps_per_second": 4.285, | |
| "eval_wer": 0.6949830717143737, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 78.91, | |
| "learning_rate": 1.8354330708661415e-05, | |
| "loss": 0.8118, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 79.59, | |
| "learning_rate": 1.7763779527559052e-05, | |
| "loss": 0.8287, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 80.27, | |
| "learning_rate": 1.7173228346456693e-05, | |
| "loss": 0.7935, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 80.95, | |
| "learning_rate": 1.658267716535433e-05, | |
| "loss": 0.805, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 81.63, | |
| "learning_rate": 1.5992125984251967e-05, | |
| "loss": 0.818, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 81.63, | |
| "eval_loss": 1.394355297088623, | |
| "eval_runtime": 62.5457, | |
| "eval_samples_per_second": 33.495, | |
| "eval_steps_per_second": 4.189, | |
| "eval_wer": 0.6994767620806402, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 82.31, | |
| "learning_rate": 1.5401574803149604e-05, | |
| "loss": 0.8009, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 82.99, | |
| "learning_rate": 1.4811023622047243e-05, | |
| "loss": 0.788, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 83.67, | |
| "learning_rate": 1.422047244094488e-05, | |
| "loss": 0.8125, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 84.35, | |
| "learning_rate": 1.3629921259842519e-05, | |
| "loss": 0.7974, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 85.03, | |
| "learning_rate": 1.3039370078740156e-05, | |
| "loss": 0.7772, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 85.03, | |
| "eval_loss": 1.373496651649475, | |
| "eval_runtime": 60.3662, | |
| "eval_samples_per_second": 34.705, | |
| "eval_steps_per_second": 4.34, | |
| "eval_wer": 0.6784856879039705, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 85.71, | |
| "learning_rate": 1.2448818897637795e-05, | |
| "loss": 0.7884, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 86.39, | |
| "learning_rate": 1.1858267716535432e-05, | |
| "loss": 0.8055, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 87.07, | |
| "learning_rate": 1.1273622047244093e-05, | |
| "loss": 0.782, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 87.75, | |
| "learning_rate": 1.0683070866141732e-05, | |
| "loss": 0.769, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 88.43, | |
| "learning_rate": 1.0092519685039369e-05, | |
| "loss": 0.7857, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 88.43, | |
| "eval_loss": 1.3695785999298096, | |
| "eval_runtime": 61.0297, | |
| "eval_samples_per_second": 34.328, | |
| "eval_steps_per_second": 4.293, | |
| "eval_wer": 0.6808248691905202, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 89.12, | |
| "learning_rate": 9.501968503937008e-06, | |
| "loss": 0.7796, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 89.79, | |
| "learning_rate": 8.911417322834645e-06, | |
| "loss": 0.7692, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 90.48, | |
| "learning_rate": 8.320866141732284e-06, | |
| "loss": 0.7752, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 91.16, | |
| "learning_rate": 7.730314960629921e-06, | |
| "loss": 0.7827, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 91.84, | |
| "learning_rate": 7.139763779527559e-06, | |
| "loss": 0.7705, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 91.84, | |
| "eval_loss": 1.4100677967071533, | |
| "eval_runtime": 62.2619, | |
| "eval_samples_per_second": 33.648, | |
| "eval_steps_per_second": 4.208, | |
| "eval_wer": 0.6870421668205602, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 92.52, | |
| "learning_rate": 6.549212598425197e-06, | |
| "loss": 0.7586, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 93.2, | |
| "learning_rate": 5.958661417322834e-06, | |
| "loss": 0.7721, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 93.88, | |
| "learning_rate": 5.368110236220472e-06, | |
| "loss": 0.7739, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 94.56, | |
| "learning_rate": 4.77755905511811e-06, | |
| "loss": 0.7644, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 95.24, | |
| "learning_rate": 4.1870078740157475e-06, | |
| "loss": 0.7537, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 95.24, | |
| "eval_loss": 1.4177812337875366, | |
| "eval_runtime": 60.4279, | |
| "eval_samples_per_second": 34.669, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.6832256078793475, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 95.92, | |
| "learning_rate": 3.5964566929133855e-06, | |
| "loss": 0.7864, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 96.6, | |
| "learning_rate": 3.0059055118110235e-06, | |
| "loss": 0.756, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 97.28, | |
| "learning_rate": 2.415354330708661e-06, | |
| "loss": 0.757, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 97.96, | |
| "learning_rate": 1.8248031496062992e-06, | |
| "loss": 0.7544, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 98.64, | |
| "learning_rate": 1.234251968503937e-06, | |
| "loss": 0.7734, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 98.64, | |
| "eval_loss": 1.4026765823364258, | |
| "eval_runtime": 60.9928, | |
| "eval_samples_per_second": 34.348, | |
| "eval_steps_per_second": 4.296, | |
| "eval_wer": 0.6831024930747922, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 99.32, | |
| "learning_rate": 6.437007874015748e-07, | |
| "loss": 0.7542, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 5.31496062992126e-08, | |
| "loss": 0.7478, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 14700, | |
| "total_flos": 5.846461401666596e+19, | |
| "train_loss": 1.4769830259984853, | |
| "train_runtime": 27582.5281, | |
| "train_samples_per_second": 17.08, | |
| "train_steps_per_second": 0.533 | |
| } | |
| ], | |
| "max_steps": 14700, | |
| "num_train_epochs": 100, | |
| "total_flos": 5.846461401666596e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |