| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 44850, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.44e-06, | |
| "loss": 13.7906, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 3.6307, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.44e-06, | |
| "loss": 2.9673, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.940000000000001e-06, | |
| "loss": 2.2254, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 7.44e-06, | |
| "loss": 1.3092, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 8.939999999999999e-06, | |
| "loss": 0.9856, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.044e-05, | |
| "loss": 0.9015, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.1940000000000001e-05, | |
| "loss": 0.8429, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.344e-05, | |
| "loss": 0.8406, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.4940000000000001e-05, | |
| "loss": 0.7523, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.6440000000000002e-05, | |
| "loss": 0.8257, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.794e-05, | |
| "loss": 0.8072, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.944e-05, | |
| "loss": 0.7369, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 2.094e-05, | |
| "loss": 0.7961, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.2440000000000002e-05, | |
| "loss": 0.7844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_cer": 0.09403695288721224, | |
| "eval_loss": 0.3411804735660553, | |
| "eval_runtime": 134.4976, | |
| "eval_samples_per_second": 23.584, | |
| "eval_steps_per_second": 2.952, | |
| "eval_wer": 0.8600252206809584, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.394e-05, | |
| "loss": 0.7554, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.544e-05, | |
| "loss": 0.7882, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.6940000000000003e-05, | |
| "loss": 0.7351, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.844e-05, | |
| "loss": 0.7208, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.994e-05, | |
| "loss": 0.7801, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2.9932788798133023e-05, | |
| "loss": 0.801, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.9862777129521585e-05, | |
| "loss": 0.7913, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.9792765460910154e-05, | |
| "loss": 0.7623, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.972275379229872e-05, | |
| "loss": 0.7715, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.965274212368728e-05, | |
| "loss": 0.7206, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 2.9582730455075846e-05, | |
| "loss": 0.7429, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 2.951271878646441e-05, | |
| "loss": 0.7427, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 2.9442707117852977e-05, | |
| "loss": 0.6917, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 2.9372695449241542e-05, | |
| "loss": 0.733, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 2.9302683780630104e-05, | |
| "loss": 0.7272, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_cer": 0.05267715522784756, | |
| "eval_loss": 0.19258654117584229, | |
| "eval_runtime": 135.378, | |
| "eval_samples_per_second": 23.431, | |
| "eval_steps_per_second": 2.933, | |
| "eval_wer": 0.6409205548549811, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 2.923267211201867e-05, | |
| "loss": 0.7339, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.9162660443407235e-05, | |
| "loss": 0.7333, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 2.90926487747958e-05, | |
| "loss": 0.7111, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 2.9022637106184366e-05, | |
| "loss": 0.7085, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 2.8952625437572927e-05, | |
| "loss": 0.7105, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 2.8882613768961496e-05, | |
| "loss": 0.6775, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 2.8812602100350058e-05, | |
| "loss": 0.6894, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 2.8742590431738623e-05, | |
| "loss": 0.6949, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 2.867257876312719e-05, | |
| "loss": 0.6945, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 2.8602567094515754e-05, | |
| "loss": 0.6708, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 2.853255542590432e-05, | |
| "loss": 0.6635, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 2.846254375729288e-05, | |
| "loss": 0.6715, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 2.8392532088681446e-05, | |
| "loss": 0.7409, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 2.8322520420070015e-05, | |
| "loss": 0.6738, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 2.8252508751458577e-05, | |
| "loss": 0.6924, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "eval_cer": 0.040116603077978305, | |
| "eval_loss": 0.14127615094184875, | |
| "eval_runtime": 136.4272, | |
| "eval_samples_per_second": 23.25, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.5721941992433796, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 2.8183197199533258e-05, | |
| "loss": 0.6429, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 2.811318553092182e-05, | |
| "loss": 0.6965, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 2.8043173862310385e-05, | |
| "loss": 0.667, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.797316219369895e-05, | |
| "loss": 0.646, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 2.7903150525087515e-05, | |
| "loss": 0.6626, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 2.783313885647608e-05, | |
| "loss": 0.6515, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 2.7763127187864643e-05, | |
| "loss": 0.6297, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 2.7693115519253208e-05, | |
| "loss": 0.6643, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 2.7623103850641777e-05, | |
| "loss": 0.6692, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 2.755309218203034e-05, | |
| "loss": 0.6321, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 2.7483080513418904e-05, | |
| "loss": 0.6707, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 2.7413068844807466e-05, | |
| "loss": 0.6479, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 2.7343057176196034e-05, | |
| "loss": 0.6332, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 2.72730455075846e-05, | |
| "loss": 0.6519, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 2.720303383897316e-05, | |
| "loss": 0.6327, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "eval_cer": 0.03712436232691731, | |
| "eval_loss": 0.12522774934768677, | |
| "eval_runtime": 136.1329, | |
| "eval_samples_per_second": 23.301, | |
| "eval_steps_per_second": 2.916, | |
| "eval_wer": 0.5365699873896596, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 2.7133022170361727e-05, | |
| "loss": 0.6117, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 2.7063010501750292e-05, | |
| "loss": 0.6419, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 2.6992998833138858e-05, | |
| "loss": 0.6638, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 2.6922987164527423e-05, | |
| "loss": 0.6646, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 2.6852975495915985e-05, | |
| "loss": 0.6726, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 2.6782963827304553e-05, | |
| "loss": 0.6302, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 2.6712952158693115e-05, | |
| "loss": 0.625, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 2.664294049008168e-05, | |
| "loss": 0.6296, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 2.6572928821470246e-05, | |
| "loss": 0.6193, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 2.650291715285881e-05, | |
| "loss": 0.6311, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 2.6432905484247377e-05, | |
| "loss": 0.655, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 2.636289381563594e-05, | |
| "loss": 0.6316, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 2.6292882147024504e-05, | |
| "loss": 0.6176, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 2.622287047841307e-05, | |
| "loss": 0.5987, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 2.6153558926487746e-05, | |
| "loss": 0.6363, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "eval_cer": 0.038916277275260425, | |
| "eval_loss": 0.12347792088985443, | |
| "eval_runtime": 135.2557, | |
| "eval_samples_per_second": 23.452, | |
| "eval_steps_per_second": 2.935, | |
| "eval_wer": 0.5740857503152585, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 2.6083547257876315e-05, | |
| "loss": 0.6513, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 2.601353558926488e-05, | |
| "loss": 0.6384, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 2.5943523920653442e-05, | |
| "loss": 0.6068, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 2.5873512252042007e-05, | |
| "loss": 0.6214, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "learning_rate": 2.5803500583430573e-05, | |
| "loss": 0.6386, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.5733488914819138e-05, | |
| "loss": 0.6183, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 2.56634772462077e-05, | |
| "loss": 0.6456, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 2.5593465577596265e-05, | |
| "loss": 0.6169, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 2.5524154025670946e-05, | |
| "loss": 0.6284, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.545414235705951e-05, | |
| "loss": 0.5984, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 2.5384130688448076e-05, | |
| "loss": 0.5826, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 2.531411901983664e-05, | |
| "loss": 0.6275, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 2.5244107351225204e-05, | |
| "loss": 0.6378, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "learning_rate": 2.517409568261377e-05, | |
| "loss": 0.6146, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.5104084014002334e-05, | |
| "loss": 0.6238, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "eval_cer": 0.03622411797487889, | |
| "eval_loss": 0.11803685128688812, | |
| "eval_runtime": 133.9685, | |
| "eval_samples_per_second": 23.677, | |
| "eval_steps_per_second": 2.963, | |
| "eval_wer": 0.5542244640605296, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 2.50340723453909e-05, | |
| "loss": 0.6072, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 2.4964060676779465e-05, | |
| "loss": 0.6303, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 10.37, | |
| "learning_rate": 2.4894049008168027e-05, | |
| "loss": 0.6006, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 2.4824037339556595e-05, | |
| "loss": 0.5905, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 2.4754025670945157e-05, | |
| "loss": 0.6223, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 2.4684014002333723e-05, | |
| "loss": 0.5953, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 2.4614002333722288e-05, | |
| "loss": 0.6121, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 2.4543990665110853e-05, | |
| "loss": 0.603, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 2.447397899649942e-05, | |
| "loss": 0.5822, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 11.15, | |
| "learning_rate": 2.440396732788798e-05, | |
| "loss": 0.6054, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 2.4333955659276546e-05, | |
| "loss": 0.5959, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 11.37, | |
| "learning_rate": 2.4263943990665114e-05, | |
| "loss": 0.5825, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 2.4193932322053676e-05, | |
| "loss": 0.5921, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 2.412392065344224e-05, | |
| "loss": 0.613, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 2.4053908984830804e-05, | |
| "loss": 0.6018, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "eval_cer": 0.03688429716637373, | |
| "eval_loss": 0.11921114474534988, | |
| "eval_runtime": 135.0278, | |
| "eval_samples_per_second": 23.491, | |
| "eval_steps_per_second": 2.94, | |
| "eval_wer": 0.5693568726355611, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 2.3983897316219372e-05, | |
| "loss": 0.5977, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 2.3913885647607938e-05, | |
| "loss": 0.6051, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 2.38438739789965e-05, | |
| "loss": 0.5845, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.3773862310385065e-05, | |
| "loss": 0.5986, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 2.370385064177363e-05, | |
| "loss": 0.5905, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 2.3633838973162195e-05, | |
| "loss": 0.6006, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 2.356382730455076e-05, | |
| "loss": 0.5672, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 2.3493815635939323e-05, | |
| "loss": 0.5808, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 12.71, | |
| "learning_rate": 2.3423803967327888e-05, | |
| "loss": 0.5849, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 2.3353792298716453e-05, | |
| "loss": 0.5994, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 12.93, | |
| "learning_rate": 2.328378063010502e-05, | |
| "loss": 0.574, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.3213768961493584e-05, | |
| "loss": 0.6094, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 2.3143757292882146e-05, | |
| "loss": 0.5734, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 13.27, | |
| "learning_rate": 2.3073745624270714e-05, | |
| "loss": 0.5673, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 2.3003733955659276e-05, | |
| "loss": 0.583, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "eval_cer": 0.03845329446564067, | |
| "eval_loss": 0.12164086103439331, | |
| "eval_runtime": 134.8964, | |
| "eval_samples_per_second": 23.514, | |
| "eval_steps_per_second": 2.943, | |
| "eval_wer": 0.5772383354350568, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 2.293372228704784e-05, | |
| "loss": 0.5649, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 2.2863710618436407e-05, | |
| "loss": 0.5526, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 2.2793698949824972e-05, | |
| "loss": 0.5737, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 2.2724387397899653e-05, | |
| "loss": 0.5797, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 2.2654375729288215e-05, | |
| "loss": 0.5465, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 2.258436406067678e-05, | |
| "loss": 0.5685, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 2.2514352392065345e-05, | |
| "loss": 0.5651, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 14.27, | |
| "learning_rate": 2.244434072345391e-05, | |
| "loss": 0.5824, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 2.2374329054842476e-05, | |
| "loss": 0.5812, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 2.2304317386231038e-05, | |
| "loss": 0.5595, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 2.2234305717619603e-05, | |
| "loss": 0.5395, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 2.2164294049008172e-05, | |
| "loss": 0.552, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 2.2094282380396734e-05, | |
| "loss": 0.5692, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 14.94, | |
| "learning_rate": 2.20242707117853e-05, | |
| "loss": 0.5396, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 2.195425904317386e-05, | |
| "loss": 0.5643, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "eval_cer": 0.03712436232691731, | |
| "eval_loss": 0.11954670399427414, | |
| "eval_runtime": 135.1674, | |
| "eval_samples_per_second": 23.467, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 0.5419293820933165, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 2.1884247374562426e-05, | |
| "loss": 0.5546, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 2.1814235705950995e-05, | |
| "loss": 0.554, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 2.1744224037339557e-05, | |
| "loss": 0.5403, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 2.1674212368728122e-05, | |
| "loss": 0.5404, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 2.1604200700116684e-05, | |
| "loss": 0.5787, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 2.1534189031505253e-05, | |
| "loss": 0.5475, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "learning_rate": 2.1464177362893818e-05, | |
| "loss": 0.5505, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 2.139416569428238e-05, | |
| "loss": 0.5477, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "learning_rate": 2.1324154025670945e-05, | |
| "loss": 0.5841, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 16.16, | |
| "learning_rate": 2.125414235705951e-05, | |
| "loss": 0.5188, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.1184130688448076e-05, | |
| "loss": 0.5343, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 16.39, | |
| "learning_rate": 2.1114819136522756e-05, | |
| "loss": 0.5564, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "learning_rate": 2.1044807467911318e-05, | |
| "loss": 0.5275, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 16.61, | |
| "learning_rate": 2.0974795799299884e-05, | |
| "loss": 0.5191, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 2.090478413068845e-05, | |
| "loss": 0.5399, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "eval_cer": 0.03700432974664552, | |
| "eval_loss": 0.12402583658695221, | |
| "eval_runtime": 134.3497, | |
| "eval_samples_per_second": 23.61, | |
| "eval_steps_per_second": 2.955, | |
| "eval_wer": 0.5223833543505675, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 2.0835472578763126e-05, | |
| "loss": 0.5308, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 2.0765460910151695e-05, | |
| "loss": 0.5418, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 17.06, | |
| "learning_rate": 2.0695449241540257e-05, | |
| "loss": 0.548, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 2.0625437572928822e-05, | |
| "loss": 0.508, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "learning_rate": 2.0555425904317384e-05, | |
| "loss": 0.5362, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 2.0485414235705952e-05, | |
| "loss": 0.5151, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 2.0415402567094518e-05, | |
| "loss": 0.5116, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 17.61, | |
| "learning_rate": 2.034539089848308e-05, | |
| "loss": 0.5526, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "learning_rate": 2.0275379229871645e-05, | |
| "loss": 0.5171, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 17.84, | |
| "learning_rate": 2.020536756126021e-05, | |
| "loss": 0.539, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 17.95, | |
| "learning_rate": 2.0135355892648776e-05, | |
| "loss": 0.5332, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.006534422403734e-05, | |
| "loss": 0.5373, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 18.17, | |
| "learning_rate": 1.9995332555425903e-05, | |
| "loss": 0.518, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 1.992532088681447e-05, | |
| "loss": 0.5344, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 1.9855309218203033e-05, | |
| "loss": 0.5529, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "eval_cer": 0.03666137951729755, | |
| "eval_loss": 0.117447629570961, | |
| "eval_runtime": 135.0568, | |
| "eval_samples_per_second": 23.486, | |
| "eval_steps_per_second": 2.94, | |
| "eval_wer": 0.5554854981084489, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 18.51, | |
| "learning_rate": 1.97852975495916e-05, | |
| "loss": 0.5231, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 1.9715285880980164e-05, | |
| "loss": 0.5468, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 18.73, | |
| "learning_rate": 1.964527421236873e-05, | |
| "loss": 0.5243, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 1.9575262543757295e-05, | |
| "loss": 0.5114, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "learning_rate": 1.9505250875145857e-05, | |
| "loss": 0.5193, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 1.9435239206534422e-05, | |
| "loss": 0.5149, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 1.9366627771295218e-05, | |
| "loss": 0.5127, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 1.929661610268378e-05, | |
| "loss": 0.5344, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 19.4, | |
| "learning_rate": 1.9226604434072345e-05, | |
| "loss": 0.5143, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "learning_rate": 1.9156592765460913e-05, | |
| "loss": 0.5393, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 19.62, | |
| "learning_rate": 1.9086581096849475e-05, | |
| "loss": 0.5179, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 1.901656942823804e-05, | |
| "loss": 0.5195, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 1.8946557759626603e-05, | |
| "loss": 0.5504, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 19.96, | |
| "learning_rate": 1.887654609101517e-05, | |
| "loss": 0.5033, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 20.07, | |
| "learning_rate": 1.8806534422403737e-05, | |
| "loss": 0.5246, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 20.07, | |
| "eval_cer": 0.03394349894971492, | |
| "eval_loss": 0.10971692949533463, | |
| "eval_runtime": 135.691, | |
| "eval_samples_per_second": 23.377, | |
| "eval_steps_per_second": 2.926, | |
| "eval_wer": 0.5047288776796973, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 20.18, | |
| "learning_rate": 1.87365227537923e-05, | |
| "loss": 0.502, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 1.8666511085180864e-05, | |
| "loss": 0.4994, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 20.4, | |
| "learning_rate": 1.859649941656943e-05, | |
| "loss": 0.518, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 20.51, | |
| "learning_rate": 1.8526487747957994e-05, | |
| "loss": 0.5148, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 1.845647607934656e-05, | |
| "loss": 0.497, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 20.74, | |
| "learning_rate": 1.838646441073512e-05, | |
| "loss": 0.5259, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 20.85, | |
| "learning_rate": 1.831645274212369e-05, | |
| "loss": 0.494, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "learning_rate": 1.8246441073512252e-05, | |
| "loss": 0.5343, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 21.07, | |
| "learning_rate": 1.8176429404900818e-05, | |
| "loss": 0.5048, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 21.18, | |
| "learning_rate": 1.8106417736289383e-05, | |
| "loss": 0.4934, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 1.8036406067677945e-05, | |
| "loss": 0.5035, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 21.4, | |
| "learning_rate": 1.7966394399066513e-05, | |
| "loss": 0.5009, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 21.52, | |
| "learning_rate": 1.7896382730455075e-05, | |
| "loss": 0.4869, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 21.63, | |
| "learning_rate": 1.782637106184364e-05, | |
| "loss": 0.5099, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 1.7756359393232206e-05, | |
| "loss": 0.4936, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "eval_cer": 0.038196081793629696, | |
| "eval_loss": 0.12249617278575897, | |
| "eval_runtime": 134.7749, | |
| "eval_samples_per_second": 23.536, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 0.5189155107187894, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 21.85, | |
| "learning_rate": 1.768634772462077e-05, | |
| "loss": 0.4913, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 21.96, | |
| "learning_rate": 1.7616336056009337e-05, | |
| "loss": 0.4884, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 22.07, | |
| "learning_rate": 1.75463243873979e-05, | |
| "loss": 0.4866, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 22.19, | |
| "learning_rate": 1.7476312718786464e-05, | |
| "loss": 0.4995, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 22.3, | |
| "learning_rate": 1.7406301050175032e-05, | |
| "loss": 0.4839, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 22.41, | |
| "learning_rate": 1.7336289381563594e-05, | |
| "loss": 0.4913, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 22.52, | |
| "learning_rate": 1.726627771295216e-05, | |
| "loss": 0.4866, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 1.719626604434072e-05, | |
| "loss": 0.4816, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 22.74, | |
| "learning_rate": 1.7126954492415402e-05, | |
| "loss": 0.481, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 22.85, | |
| "learning_rate": 1.705694282380397e-05, | |
| "loss": 0.4851, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "learning_rate": 1.6986931155192533e-05, | |
| "loss": 0.4711, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 1.6916919486581098e-05, | |
| "loss": 0.4686, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 23.19, | |
| "learning_rate": 1.684690781796966e-05, | |
| "loss": 0.4714, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 23.3, | |
| "learning_rate": 1.6776896149358225e-05, | |
| "loss": 0.4822, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 23.41, | |
| "learning_rate": 1.6706884480746794e-05, | |
| "loss": 0.4629, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 23.41, | |
| "eval_cer": 0.03436361298066618, | |
| "eval_loss": 0.11421633511781693, | |
| "eval_runtime": 135.1212, | |
| "eval_samples_per_second": 23.475, | |
| "eval_steps_per_second": 2.938, | |
| "eval_wer": 0.5047288776796973, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 23.52, | |
| "learning_rate": 1.6636872812135356e-05, | |
| "loss": 0.4744, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 23.63, | |
| "learning_rate": 1.656686114352392e-05, | |
| "loss": 0.476, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 1.6496849474912483e-05, | |
| "loss": 0.4981, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 1.6426837806301052e-05, | |
| "loss": 0.4689, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 23.97, | |
| "learning_rate": 1.6356826137689617e-05, | |
| "loss": 0.4678, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 24.08, | |
| "learning_rate": 1.628681446907818e-05, | |
| "loss": 0.4987, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 1.6216802800466744e-05, | |
| "loss": 0.4857, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 24.3, | |
| "learning_rate": 1.614679113185531e-05, | |
| "loss": 0.4808, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "learning_rate": 1.6076779463243875e-05, | |
| "loss": 0.4735, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 1.600676779463244e-05, | |
| "loss": 0.4499, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 24.64, | |
| "learning_rate": 1.5936756126021002e-05, | |
| "loss": 0.4963, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "learning_rate": 1.586674445740957e-05, | |
| "loss": 0.4802, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 24.86, | |
| "learning_rate": 1.5796732788798133e-05, | |
| "loss": 0.4517, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 24.97, | |
| "learning_rate": 1.5726721120186698e-05, | |
| "loss": 0.4942, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 1.5656709451575263e-05, | |
| "loss": 0.4463, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "eval_cer": 0.03392635143824752, | |
| "eval_loss": 0.11677591502666473, | |
| "eval_runtime": 134.8589, | |
| "eval_samples_per_second": 23.521, | |
| "eval_steps_per_second": 2.944, | |
| "eval_wer": 0.48865069356872637, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 25.2, | |
| "learning_rate": 1.558669778296383e-05, | |
| "loss": 0.4649, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 25.31, | |
| "learning_rate": 1.5516686114352394e-05, | |
| "loss": 0.4559, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 1.544737456242707e-05, | |
| "loss": 0.4828, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 25.53, | |
| "learning_rate": 1.5377362893815636e-05, | |
| "loss": 0.4674, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 25.64, | |
| "learning_rate": 1.53073512252042e-05, | |
| "loss": 0.4706, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 1.5237339556592764e-05, | |
| "loss": 0.487, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 25.86, | |
| "learning_rate": 1.5167327887981332e-05, | |
| "loss": 0.4783, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "learning_rate": 1.5097316219369896e-05, | |
| "loss": 0.4779, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 1.502730455075846e-05, | |
| "loss": 0.4587, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 26.2, | |
| "learning_rate": 1.4957292882147025e-05, | |
| "loss": 0.4619, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 1.4887281213535588e-05, | |
| "loss": 0.4457, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 1.4817269544924155e-05, | |
| "loss": 0.4561, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 26.53, | |
| "learning_rate": 1.4747257876312719e-05, | |
| "loss": 0.4696, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 26.64, | |
| "learning_rate": 1.4677246207701284e-05, | |
| "loss": 0.4562, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 26.76, | |
| "learning_rate": 1.4607234539089848e-05, | |
| "loss": 0.4671, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 26.76, | |
| "eval_cer": 0.03379774510224204, | |
| "eval_loss": 0.11189751327037811, | |
| "eval_runtime": 135.618, | |
| "eval_samples_per_second": 23.389, | |
| "eval_steps_per_second": 2.927, | |
| "eval_wer": 0.507250945775536, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 26.87, | |
| "learning_rate": 1.4537222870478413e-05, | |
| "loss": 0.4331, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 26.98, | |
| "learning_rate": 1.4467211201866978e-05, | |
| "loss": 0.4647, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 27.09, | |
| "learning_rate": 1.4397199533255544e-05, | |
| "loss": 0.4136, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 1.4327187864644107e-05, | |
| "loss": 0.4365, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 27.31, | |
| "learning_rate": 1.4257176196032673e-05, | |
| "loss": 0.4593, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "learning_rate": 1.4187164527421236e-05, | |
| "loss": 0.4484, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 27.54, | |
| "learning_rate": 1.4117152858809803e-05, | |
| "loss": 0.4693, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 27.65, | |
| "learning_rate": 1.4047141190198367e-05, | |
| "loss": 0.4582, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 27.76, | |
| "learning_rate": 1.3977129521586932e-05, | |
| "loss": 0.4206, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.3907117852975496e-05, | |
| "loss": 0.4356, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 27.98, | |
| "learning_rate": 1.3837106184364061e-05, | |
| "loss": 0.4426, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 28.09, | |
| "learning_rate": 1.3767094515752626e-05, | |
| "loss": 0.4161, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 28.21, | |
| "learning_rate": 1.3697082847141192e-05, | |
| "loss": 0.4368, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 28.32, | |
| "learning_rate": 1.3627071178529755e-05, | |
| "loss": 0.4613, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 28.43, | |
| "learning_rate": 1.355705950991832e-05, | |
| "loss": 0.4359, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 28.43, | |
| "eval_cer": 0.03627556050928109, | |
| "eval_loss": 0.12062488496303558, | |
| "eval_runtime": 135.3296, | |
| "eval_samples_per_second": 23.439, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.5479192938209332, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 28.54, | |
| "learning_rate": 1.3487047841306884e-05, | |
| "loss": 0.4364, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 28.65, | |
| "learning_rate": 1.341703617269545e-05, | |
| "loss": 0.4751, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 28.76, | |
| "learning_rate": 1.3347024504084015e-05, | |
| "loss": 0.4574, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 28.87, | |
| "learning_rate": 1.3277012835472578e-05, | |
| "loss": 0.4432, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "learning_rate": 1.3207001166861144e-05, | |
| "loss": 0.4309, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 29.1, | |
| "learning_rate": 1.3136989498249707e-05, | |
| "loss": 0.4458, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 29.21, | |
| "learning_rate": 1.3066977829638274e-05, | |
| "loss": 0.4236, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 29.32, | |
| "learning_rate": 1.2996966161026838e-05, | |
| "loss": 0.4266, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 29.43, | |
| "learning_rate": 1.2926954492415403e-05, | |
| "loss": 0.4565, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 29.54, | |
| "learning_rate": 1.2856942823803967e-05, | |
| "loss": 0.4376, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "learning_rate": 1.2786931155192532e-05, | |
| "loss": 0.4144, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 29.77, | |
| "learning_rate": 1.2716919486581098e-05, | |
| "loss": 0.4527, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 29.88, | |
| "learning_rate": 1.2646907817969663e-05, | |
| "loss": 0.4446, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 29.99, | |
| "learning_rate": 1.2576896149358226e-05, | |
| "loss": 0.4229, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 30.1, | |
| "learning_rate": 1.2507584597432905e-05, | |
| "loss": 0.4225, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 30.1, | |
| "eval_cer": 0.03446649804947057, | |
| "eval_loss": 0.11224740743637085, | |
| "eval_runtime": 135.3661, | |
| "eval_samples_per_second": 23.433, | |
| "eval_steps_per_second": 2.933, | |
| "eval_wer": 0.5170239596469105, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 30.21, | |
| "learning_rate": 1.243757292882147e-05, | |
| "loss": 0.4344, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 1.2367561260210036e-05, | |
| "loss": 0.4477, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 30.43, | |
| "learning_rate": 1.2297549591598601e-05, | |
| "loss": 0.4037, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 30.55, | |
| "learning_rate": 1.2227537922987165e-05, | |
| "loss": 0.4201, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 30.66, | |
| "learning_rate": 1.215752625437573e-05, | |
| "loss": 0.4318, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 1.2087514585764294e-05, | |
| "loss": 0.4264, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 30.88, | |
| "learning_rate": 1.2017502917152859e-05, | |
| "loss": 0.4141, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 30.99, | |
| "learning_rate": 1.1947491248541424e-05, | |
| "loss": 0.4099, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 31.1, | |
| "learning_rate": 1.1877479579929988e-05, | |
| "loss": 0.4215, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 31.22, | |
| "learning_rate": 1.1807467911318553e-05, | |
| "loss": 0.4162, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 31.33, | |
| "learning_rate": 1.1737456242707117e-05, | |
| "loss": 0.4349, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 31.44, | |
| "learning_rate": 1.1667444574095684e-05, | |
| "loss": 0.4063, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 31.55, | |
| "learning_rate": 1.1597432905484247e-05, | |
| "loss": 0.4288, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 31.66, | |
| "learning_rate": 1.1527421236872813e-05, | |
| "loss": 0.4224, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 31.77, | |
| "learning_rate": 1.1457409568261376e-05, | |
| "loss": 0.4038, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 31.77, | |
| "eval_cer": 0.03431217044626399, | |
| "eval_loss": 0.11591120064258575, | |
| "eval_runtime": 135.0079, | |
| "eval_samples_per_second": 23.495, | |
| "eval_steps_per_second": 2.941, | |
| "eval_wer": 0.5031525851197982, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 31.88, | |
| "learning_rate": 1.1387397899649942e-05, | |
| "loss": 0.435, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 1.1317386231038507e-05, | |
| "loss": 0.4307, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 32.11, | |
| "learning_rate": 1.1248074679113186e-05, | |
| "loss": 0.4127, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 32.22, | |
| "learning_rate": 1.1178063010501751e-05, | |
| "loss": 0.419, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 32.33, | |
| "learning_rate": 1.1108051341890315e-05, | |
| "loss": 0.4103, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 32.44, | |
| "learning_rate": 1.103803967327888e-05, | |
| "loss": 0.4211, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 32.55, | |
| "learning_rate": 1.0968028004667445e-05, | |
| "loss": 0.4065, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 32.66, | |
| "learning_rate": 1.089801633605601e-05, | |
| "loss": 0.4161, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 32.78, | |
| "learning_rate": 1.0828004667444574e-05, | |
| "loss": 0.4006, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 32.89, | |
| "learning_rate": 1.075799299883314e-05, | |
| "loss": 0.4208, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 1.0687981330221703e-05, | |
| "loss": 0.4003, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 33.11, | |
| "learning_rate": 1.0617969661610268e-05, | |
| "loss": 0.3983, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 33.22, | |
| "learning_rate": 1.0547957992998834e-05, | |
| "loss": 0.4042, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 1.0477946324387397e-05, | |
| "loss": 0.3925, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 33.44, | |
| "learning_rate": 1.0407934655775963e-05, | |
| "loss": 0.4271, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 33.44, | |
| "eval_cer": 0.03387490890384533, | |
| "eval_loss": 0.11164171248674393, | |
| "eval_runtime": 136.1618, | |
| "eval_samples_per_second": 23.296, | |
| "eval_steps_per_second": 2.916, | |
| "eval_wer": 0.512610340479193, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 33.56, | |
| "learning_rate": 1.0337922987164526e-05, | |
| "loss": 0.4145, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 33.67, | |
| "learning_rate": 1.0267911318553093e-05, | |
| "loss": 0.436, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "learning_rate": 1.0197899649941657e-05, | |
| "loss": 0.3896, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 33.89, | |
| "learning_rate": 1.0127887981330222e-05, | |
| "loss": 0.3886, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 1.0057876312718786e-05, | |
| "loss": 0.4135, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 34.11, | |
| "learning_rate": 9.987864644107351e-06, | |
| "loss": 0.3844, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 34.23, | |
| "learning_rate": 9.917852975495916e-06, | |
| "loss": 0.3802, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 34.34, | |
| "learning_rate": 9.847841306884482e-06, | |
| "loss": 0.3988, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 34.45, | |
| "learning_rate": 9.777829638273045e-06, | |
| "loss": 0.3751, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 34.56, | |
| "learning_rate": 9.70781796966161e-06, | |
| "loss": 0.406, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "learning_rate": 9.637806301050174e-06, | |
| "loss": 0.4041, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 9.567794632438741e-06, | |
| "loss": 0.3848, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 34.89, | |
| "learning_rate": 9.497782963827305e-06, | |
| "loss": 0.3839, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "learning_rate": 9.42777129521587e-06, | |
| "loss": 0.4006, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 35.12, | |
| "learning_rate": 9.357759626604434e-06, | |
| "loss": 0.3867, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 35.12, | |
| "eval_cer": 0.03265743558966005, | |
| "eval_loss": 0.11009135842323303, | |
| "eval_runtime": 135.5282, | |
| "eval_samples_per_second": 23.405, | |
| "eval_steps_per_second": 2.929, | |
| "eval_wer": 0.4936948297604035, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 35.23, | |
| "learning_rate": 9.287747957992999e-06, | |
| "loss": 0.3943, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 35.34, | |
| "learning_rate": 9.217736289381564e-06, | |
| "loss": 0.3736, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 35.45, | |
| "learning_rate": 9.14772462077013e-06, | |
| "loss": 0.3964, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "learning_rate": 9.077712952158693e-06, | |
| "loss": 0.3847, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 35.67, | |
| "learning_rate": 9.007701283547259e-06, | |
| "loss": 0.3879, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 35.79, | |
| "learning_rate": 8.937689614935822e-06, | |
| "loss": 0.4009, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 35.9, | |
| "learning_rate": 8.867677946324389e-06, | |
| "loss": 0.3658, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 36.01, | |
| "learning_rate": 8.797666277712953e-06, | |
| "loss": 0.3778, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 36.12, | |
| "learning_rate": 8.727654609101516e-06, | |
| "loss": 0.3814, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 36.23, | |
| "learning_rate": 8.657642940490082e-06, | |
| "loss": 0.3357, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 36.34, | |
| "learning_rate": 8.587631271878645e-06, | |
| "loss": 0.3977, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 36.45, | |
| "learning_rate": 8.517619603267212e-06, | |
| "loss": 0.3925, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 36.57, | |
| "learning_rate": 8.447607934655776e-06, | |
| "loss": 0.3721, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 36.68, | |
| "learning_rate": 8.377596266044341e-06, | |
| "loss": 0.4022, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 8.30828471411902e-06, | |
| "loss": 0.3674, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "eval_cer": 0.032983238307540616, | |
| "eval_loss": 0.11423581093549728, | |
| "eval_runtime": 134.7435, | |
| "eval_samples_per_second": 23.541, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 0.49401008827238335, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "learning_rate": 8.238273045507585e-06, | |
| "loss": 0.3753, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 8.168961493582264e-06, | |
| "loss": 0.4126, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 37.12, | |
| "learning_rate": 8.09894982497083e-06, | |
| "loss": 0.3656, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 37.24, | |
| "learning_rate": 8.028938156359393e-06, | |
| "loss": 0.3834, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 37.35, | |
| "learning_rate": 7.958926487747958e-06, | |
| "loss": 0.3759, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 37.46, | |
| "learning_rate": 7.888914819136524e-06, | |
| "loss": 0.3487, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 37.57, | |
| "learning_rate": 7.818903150525089e-06, | |
| "loss": 0.3792, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 37.68, | |
| "learning_rate": 7.748891481913652e-06, | |
| "loss": 0.3553, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 37.79, | |
| "learning_rate": 7.678879813302216e-06, | |
| "loss": 0.3542, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "learning_rate": 7.608868144690782e-06, | |
| "loss": 0.3827, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "learning_rate": 7.539556592765461e-06, | |
| "loss": 0.3921, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 38.13, | |
| "learning_rate": 7.469544924154026e-06, | |
| "loss": 0.3859, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 38.24, | |
| "learning_rate": 7.399533255542591e-06, | |
| "loss": 0.3651, | |
| "step": 34300 | |
| }, | |
| { | |
| "epoch": 38.35, | |
| "learning_rate": 7.329521586931156e-06, | |
| "loss": 0.3503, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 7.260210035005834e-06, | |
| "loss": 0.3607, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_cer": 0.032717451879795946, | |
| "eval_loss": 0.11055753380060196, | |
| "eval_runtime": 138.3806, | |
| "eval_samples_per_second": 22.922, | |
| "eval_steps_per_second": 2.869, | |
| "eval_wer": 0.5145018915510718, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 38.57, | |
| "learning_rate": 7.190198366394399e-06, | |
| "loss": 0.3566, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 7.120186697782964e-06, | |
| "loss": 0.3847, | |
| "step": 34700 | |
| }, | |
| { | |
| "epoch": 38.8, | |
| "learning_rate": 7.050175029171528e-06, | |
| "loss": 0.3647, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 38.91, | |
| "learning_rate": 6.9801633605600935e-06, | |
| "loss": 0.3895, | |
| "step": 34900 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "learning_rate": 6.910151691948658e-06, | |
| "loss": 0.3707, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 39.13, | |
| "learning_rate": 6.840140023337223e-06, | |
| "loss": 0.3544, | |
| "step": 35100 | |
| }, | |
| { | |
| "epoch": 39.24, | |
| "learning_rate": 6.770128354725788e-06, | |
| "loss": 0.3592, | |
| "step": 35200 | |
| }, | |
| { | |
| "epoch": 39.35, | |
| "learning_rate": 6.700116686114352e-06, | |
| "loss": 0.3536, | |
| "step": 35300 | |
| }, | |
| { | |
| "epoch": 39.46, | |
| "learning_rate": 6.6301050175029175e-06, | |
| "loss": 0.3778, | |
| "step": 35400 | |
| }, | |
| { | |
| "epoch": 39.58, | |
| "learning_rate": 6.560093348891482e-06, | |
| "loss": 0.3561, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 39.69, | |
| "learning_rate": 6.490081680280047e-06, | |
| "loss": 0.3639, | |
| "step": 35600 | |
| }, | |
| { | |
| "epoch": 39.8, | |
| "learning_rate": 6.420070011668612e-06, | |
| "loss": 0.3356, | |
| "step": 35700 | |
| }, | |
| { | |
| "epoch": 39.91, | |
| "learning_rate": 6.350058343057176e-06, | |
| "loss": 0.3705, | |
| "step": 35800 | |
| }, | |
| { | |
| "epoch": 40.02, | |
| "learning_rate": 6.2800466744457415e-06, | |
| "loss": 0.361, | |
| "step": 35900 | |
| }, | |
| { | |
| "epoch": 40.13, | |
| "learning_rate": 6.210035005834306e-06, | |
| "loss": 0.3651, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 40.13, | |
| "eval_cer": 0.03165430616881725, | |
| "eval_loss": 0.11723905056715012, | |
| "eval_runtime": 136.1596, | |
| "eval_samples_per_second": 23.296, | |
| "eval_steps_per_second": 2.916, | |
| "eval_wer": 0.4921185372005044, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 40.25, | |
| "learning_rate": 6.140023337222871e-06, | |
| "loss": 0.3448, | |
| "step": 36100 | |
| }, | |
| { | |
| "epoch": 40.36, | |
| "learning_rate": 6.070011668611436e-06, | |
| "loss": 0.3615, | |
| "step": 36200 | |
| }, | |
| { | |
| "epoch": 40.47, | |
| "learning_rate": 6e-06, | |
| "loss": 0.3516, | |
| "step": 36300 | |
| }, | |
| { | |
| "epoch": 40.58, | |
| "learning_rate": 5.9299883313885655e-06, | |
| "loss": 0.3803, | |
| "step": 36400 | |
| }, | |
| { | |
| "epoch": 40.69, | |
| "learning_rate": 5.85997666277713e-06, | |
| "loss": 0.3686, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 40.8, | |
| "learning_rate": 5.789964994165694e-06, | |
| "loss": 0.3421, | |
| "step": 36600 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 5.719953325554259e-06, | |
| "loss": 0.3824, | |
| "step": 36700 | |
| }, | |
| { | |
| "epoch": 41.03, | |
| "learning_rate": 5.649941656942823e-06, | |
| "loss": 0.3497, | |
| "step": 36800 | |
| }, | |
| { | |
| "epoch": 41.14, | |
| "learning_rate": 5.579929988331389e-06, | |
| "loss": 0.3391, | |
| "step": 36900 | |
| }, | |
| { | |
| "epoch": 41.25, | |
| "learning_rate": 5.509918319719953e-06, | |
| "loss": 0.3645, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 41.36, | |
| "learning_rate": 5.439906651108518e-06, | |
| "loss": 0.3543, | |
| "step": 37100 | |
| }, | |
| { | |
| "epoch": 41.47, | |
| "learning_rate": 5.369894982497083e-06, | |
| "loss": 0.3476, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 41.58, | |
| "learning_rate": 5.299883313885647e-06, | |
| "loss": 0.3536, | |
| "step": 37300 | |
| }, | |
| { | |
| "epoch": 41.69, | |
| "learning_rate": 5.2298716452742126e-06, | |
| "loss": 0.3592, | |
| "step": 37400 | |
| }, | |
| { | |
| "epoch": 41.81, | |
| "learning_rate": 5.159859976662777e-06, | |
| "loss": 0.3268, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 41.81, | |
| "eval_cer": 0.03096840571012132, | |
| "eval_loss": 0.10934926569461823, | |
| "eval_runtime": 136.4324, | |
| "eval_samples_per_second": 23.25, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.48297604035308955, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "learning_rate": 5.089848308051342e-06, | |
| "loss": 0.3537, | |
| "step": 37600 | |
| }, | |
| { | |
| "epoch": 42.03, | |
| "learning_rate": 5.019836639439907e-06, | |
| "loss": 0.3285, | |
| "step": 37700 | |
| }, | |
| { | |
| "epoch": 42.14, | |
| "learning_rate": 4.949824970828471e-06, | |
| "loss": 0.352, | |
| "step": 37800 | |
| }, | |
| { | |
| "epoch": 42.25, | |
| "learning_rate": 4.8798133022170366e-06, | |
| "loss": 0.3556, | |
| "step": 37900 | |
| }, | |
| { | |
| "epoch": 42.36, | |
| "learning_rate": 4.809801633605601e-06, | |
| "loss": 0.363, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 42.47, | |
| "learning_rate": 4.7397899649941655e-06, | |
| "loss": 0.347, | |
| "step": 38100 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "learning_rate": 4.669778296382731e-06, | |
| "loss": 0.3555, | |
| "step": 38200 | |
| }, | |
| { | |
| "epoch": 42.7, | |
| "learning_rate": 4.599766627771295e-06, | |
| "loss": 0.3511, | |
| "step": 38300 | |
| }, | |
| { | |
| "epoch": 42.81, | |
| "learning_rate": 4.5297549591598605e-06, | |
| "loss": 0.3473, | |
| "step": 38400 | |
| }, | |
| { | |
| "epoch": 42.92, | |
| "learning_rate": 4.459743290548425e-06, | |
| "loss": 0.357, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 43.03, | |
| "learning_rate": 4.390431738623105e-06, | |
| "loss": 0.3444, | |
| "step": 38600 | |
| }, | |
| { | |
| "epoch": 43.14, | |
| "learning_rate": 4.320420070011668e-06, | |
| "loss": 0.3428, | |
| "step": 38700 | |
| }, | |
| { | |
| "epoch": 43.26, | |
| "learning_rate": 4.251108518086348e-06, | |
| "loss": 0.3321, | |
| "step": 38800 | |
| }, | |
| { | |
| "epoch": 43.37, | |
| "learning_rate": 4.181096849474912e-06, | |
| "loss": 0.3516, | |
| "step": 38900 | |
| }, | |
| { | |
| "epoch": 43.48, | |
| "learning_rate": 4.111085180863478e-06, | |
| "loss": 0.3345, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 43.48, | |
| "eval_cer": 0.03137137222960518, | |
| "eval_loss": 0.11307822167873383, | |
| "eval_runtime": 136.8015, | |
| "eval_samples_per_second": 23.187, | |
| "eval_steps_per_second": 2.902, | |
| "eval_wer": 0.4760403530895334, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 43.59, | |
| "learning_rate": 4.041073512252042e-06, | |
| "loss": 0.3352, | |
| "step": 39100 | |
| }, | |
| { | |
| "epoch": 43.7, | |
| "learning_rate": 3.9710618436406065e-06, | |
| "loss": 0.3211, | |
| "step": 39200 | |
| }, | |
| { | |
| "epoch": 43.81, | |
| "learning_rate": 3.901050175029172e-06, | |
| "loss": 0.3512, | |
| "step": 39300 | |
| }, | |
| { | |
| "epoch": 43.92, | |
| "learning_rate": 3.831038506417736e-06, | |
| "loss": 0.3104, | |
| "step": 39400 | |
| }, | |
| { | |
| "epoch": 44.04, | |
| "learning_rate": 3.761026837806301e-06, | |
| "loss": 0.3397, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 44.15, | |
| "learning_rate": 3.691015169194866e-06, | |
| "loss": 0.3331, | |
| "step": 39600 | |
| }, | |
| { | |
| "epoch": 44.26, | |
| "learning_rate": 3.621003500583431e-06, | |
| "loss": 0.3353, | |
| "step": 39700 | |
| }, | |
| { | |
| "epoch": 44.37, | |
| "learning_rate": 3.5509918319719954e-06, | |
| "loss": 0.3344, | |
| "step": 39800 | |
| }, | |
| { | |
| "epoch": 44.48, | |
| "learning_rate": 3.4809801633605603e-06, | |
| "loss": 0.324, | |
| "step": 39900 | |
| }, | |
| { | |
| "epoch": 44.59, | |
| "learning_rate": 3.410968494749125e-06, | |
| "loss": 0.3322, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 44.7, | |
| "learning_rate": 3.3409568261376896e-06, | |
| "loss": 0.3476, | |
| "step": 40100 | |
| }, | |
| { | |
| "epoch": 44.82, | |
| "learning_rate": 3.271645274212369e-06, | |
| "loss": 0.3481, | |
| "step": 40200 | |
| }, | |
| { | |
| "epoch": 44.93, | |
| "learning_rate": 3.2016336056009337e-06, | |
| "loss": 0.3347, | |
| "step": 40300 | |
| }, | |
| { | |
| "epoch": 45.04, | |
| "learning_rate": 3.1316219369894986e-06, | |
| "loss": 0.3146, | |
| "step": 40400 | |
| }, | |
| { | |
| "epoch": 45.15, | |
| "learning_rate": 3.061610268378063e-06, | |
| "loss": 0.3236, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 45.15, | |
| "eval_cer": 0.03166287992455095, | |
| "eval_loss": 0.11318423599004745, | |
| "eval_runtime": 136.6817, | |
| "eval_samples_per_second": 23.207, | |
| "eval_steps_per_second": 2.905, | |
| "eval_wer": 0.4864438839848676, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 45.26, | |
| "learning_rate": 2.991598599766628e-06, | |
| "loss": 0.3063, | |
| "step": 40600 | |
| }, | |
| { | |
| "epoch": 45.37, | |
| "learning_rate": 2.9215869311551928e-06, | |
| "loss": 0.3653, | |
| "step": 40700 | |
| }, | |
| { | |
| "epoch": 45.48, | |
| "learning_rate": 2.8515752625437572e-06, | |
| "loss": 0.3366, | |
| "step": 40800 | |
| }, | |
| { | |
| "epoch": 45.6, | |
| "learning_rate": 2.781563593932322e-06, | |
| "loss": 0.3487, | |
| "step": 40900 | |
| }, | |
| { | |
| "epoch": 45.71, | |
| "learning_rate": 2.7115519253208866e-06, | |
| "loss": 0.3334, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 45.82, | |
| "learning_rate": 2.6415402567094515e-06, | |
| "loss": 0.368, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 45.93, | |
| "learning_rate": 2.5715285880980163e-06, | |
| "loss": 0.339, | |
| "step": 41200 | |
| }, | |
| { | |
| "epoch": 46.04, | |
| "learning_rate": 2.5015169194865812e-06, | |
| "loss": 0.3168, | |
| "step": 41300 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 2.431505250875146e-06, | |
| "loss": 0.3242, | |
| "step": 41400 | |
| }, | |
| { | |
| "epoch": 46.27, | |
| "learning_rate": 2.3614935822637106e-06, | |
| "loss": 0.3119, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 46.38, | |
| "learning_rate": 2.2914819136522754e-06, | |
| "loss": 0.3275, | |
| "step": 41600 | |
| }, | |
| { | |
| "epoch": 46.49, | |
| "learning_rate": 2.2214702450408403e-06, | |
| "loss": 0.3187, | |
| "step": 41700 | |
| }, | |
| { | |
| "epoch": 46.6, | |
| "learning_rate": 2.151458576429405e-06, | |
| "loss": 0.3162, | |
| "step": 41800 | |
| }, | |
| { | |
| "epoch": 46.71, | |
| "learning_rate": 2.08144690781797e-06, | |
| "loss": 0.3277, | |
| "step": 41900 | |
| }, | |
| { | |
| "epoch": 46.82, | |
| "learning_rate": 2.0114352392065345e-06, | |
| "loss": 0.312, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 46.82, | |
| "eval_cer": 0.031474257298409566, | |
| "eval_loss": 0.11236841976642609, | |
| "eval_runtime": 139.3204, | |
| "eval_samples_per_second": 22.768, | |
| "eval_steps_per_second": 2.85, | |
| "eval_wer": 0.48612862547288777, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 46.93, | |
| "learning_rate": 1.941423570595099e-06, | |
| "loss": 0.3065, | |
| "step": 42100 | |
| }, | |
| { | |
| "epoch": 47.05, | |
| "learning_rate": 1.871411901983664e-06, | |
| "loss": 0.3186, | |
| "step": 42200 | |
| }, | |
| { | |
| "epoch": 47.16, | |
| "learning_rate": 1.8014002333722288e-06, | |
| "loss": 0.3077, | |
| "step": 42300 | |
| }, | |
| { | |
| "epoch": 47.27, | |
| "learning_rate": 1.7313885647607934e-06, | |
| "loss": 0.3199, | |
| "step": 42400 | |
| }, | |
| { | |
| "epoch": 47.38, | |
| "learning_rate": 1.661376896149358e-06, | |
| "loss": 0.3238, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 47.49, | |
| "learning_rate": 1.591365227537923e-06, | |
| "loss": 0.3261, | |
| "step": 42600 | |
| }, | |
| { | |
| "epoch": 47.6, | |
| "learning_rate": 1.5213535589264879e-06, | |
| "loss": 0.3271, | |
| "step": 42700 | |
| }, | |
| { | |
| "epoch": 47.71, | |
| "learning_rate": 1.4520420070011668e-06, | |
| "loss": 0.3041, | |
| "step": 42800 | |
| }, | |
| { | |
| "epoch": 47.83, | |
| "learning_rate": 1.3820303383897317e-06, | |
| "loss": 0.3355, | |
| "step": 42900 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "learning_rate": 1.3120186697782964e-06, | |
| "loss": 0.3324, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 48.05, | |
| "learning_rate": 1.2420070011668613e-06, | |
| "loss": 0.3219, | |
| "step": 43100 | |
| }, | |
| { | |
| "epoch": 48.16, | |
| "learning_rate": 1.171995332555426e-06, | |
| "loss": 0.3002, | |
| "step": 43200 | |
| }, | |
| { | |
| "epoch": 48.27, | |
| "learning_rate": 1.1019836639439906e-06, | |
| "loss": 0.3221, | |
| "step": 43300 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "learning_rate": 1.0319719953325555e-06, | |
| "loss": 0.3147, | |
| "step": 43400 | |
| }, | |
| { | |
| "epoch": 48.49, | |
| "learning_rate": 9.619603267211202e-07, | |
| "loss": 0.3106, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 48.49, | |
| "eval_cer": 0.030599734213572254, | |
| "eval_loss": 0.11161273717880249, | |
| "eval_runtime": 137.8833, | |
| "eval_samples_per_second": 23.005, | |
| "eval_steps_per_second": 2.879, | |
| "eval_wer": 0.4744640605296343, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 48.61, | |
| "learning_rate": 8.91948658109685e-07, | |
| "loss": 0.341, | |
| "step": 43600 | |
| }, | |
| { | |
| "epoch": 48.72, | |
| "learning_rate": 8.219369894982497e-07, | |
| "loss": 0.3272, | |
| "step": 43700 | |
| }, | |
| { | |
| "epoch": 48.83, | |
| "learning_rate": 7.519253208868145e-07, | |
| "loss": 0.293, | |
| "step": 43800 | |
| }, | |
| { | |
| "epoch": 48.94, | |
| "learning_rate": 6.819136522753793e-07, | |
| "loss": 0.2981, | |
| "step": 43900 | |
| }, | |
| { | |
| "epoch": 49.05, | |
| "learning_rate": 6.119019836639439e-07, | |
| "loss": 0.3162, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 49.16, | |
| "learning_rate": 5.418903150525087e-07, | |
| "loss": 0.32, | |
| "step": 44100 | |
| }, | |
| { | |
| "epoch": 49.28, | |
| "learning_rate": 4.7187864644107353e-07, | |
| "loss": 0.3359, | |
| "step": 44200 | |
| }, | |
| { | |
| "epoch": 49.39, | |
| "learning_rate": 4.018669778296383e-07, | |
| "loss": 0.3144, | |
| "step": 44300 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 3.3185530921820303e-07, | |
| "loss": 0.3311, | |
| "step": 44400 | |
| }, | |
| { | |
| "epoch": 49.61, | |
| "learning_rate": 2.618436406067678e-07, | |
| "loss": 0.2944, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 49.72, | |
| "learning_rate": 1.9183197199533256e-07, | |
| "loss": 0.325, | |
| "step": 44600 | |
| }, | |
| { | |
| "epoch": 49.83, | |
| "learning_rate": 1.2182030338389733e-07, | |
| "loss": 0.3112, | |
| "step": 44700 | |
| }, | |
| { | |
| "epoch": 49.94, | |
| "learning_rate": 5.180863477246208e-08, | |
| "loss": 0.3129, | |
| "step": 44800 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 44850, | |
| "total_flos": 1.2547723559972522e+20, | |
| "train_loss": 0.5348544037647737, | |
| "train_runtime": 44549.5681, | |
| "train_samples_per_second": 8.052, | |
| "train_steps_per_second": 1.007 | |
| } | |
| ], | |
| "max_steps": 44850, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.2547723559972522e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |