| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 198.91205211726384, | |
| "global_step": 91500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 5.988000000000001e-05, | |
| "loss": 6.0553, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_cer": 0.7651548792050215, | |
| "eval_loss": 3.0141708850860596, | |
| "eval_runtime": 204.505, | |
| "eval_samples_per_second": 21.858, | |
| "eval_steps_per_second": 1.369, | |
| "eval_wer": 0.9928897917724734, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 7.911466666666667e-05, | |
| "loss": 2.3839, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_cer": 0.19668038836815765, | |
| "eval_loss": 0.5387711524963379, | |
| "eval_runtime": 202.7107, | |
| "eval_samples_per_second": 22.051, | |
| "eval_steps_per_second": 1.381, | |
| "eval_wer": 0.45625045345715737, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 7.778133333333334e-05, | |
| "loss": 1.3597, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "eval_cer": 0.15966779880705206, | |
| "eval_loss": 0.4122415781021118, | |
| "eval_runtime": 202.9557, | |
| "eval_samples_per_second": 22.025, | |
| "eval_steps_per_second": 1.38, | |
| "eval_wer": 0.36247551331350214, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.644800000000001e-05, | |
| "loss": 1.2294, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "eval_cer": 0.15757353911885075, | |
| "eval_loss": 0.36426475644111633, | |
| "eval_runtime": 204.7716, | |
| "eval_samples_per_second": 21.829, | |
| "eval_steps_per_second": 1.367, | |
| "eval_wer": 0.3416164840745846, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 7.511644444444445e-05, | |
| "loss": 1.1672, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "eval_cer": 0.1442218835135557, | |
| "eval_loss": 0.34255123138427734, | |
| "eval_runtime": 202.492, | |
| "eval_samples_per_second": 22.075, | |
| "eval_steps_per_second": 1.383, | |
| "eval_wer": 0.3202314445331205, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "learning_rate": 7.378311111111112e-05, | |
| "loss": 1.1288, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "eval_cer": 0.1378970992403058, | |
| "eval_loss": 0.31974485516548157, | |
| "eval_runtime": 204.9353, | |
| "eval_samples_per_second": 21.812, | |
| "eval_steps_per_second": 1.366, | |
| "eval_wer": 0.3065551766669085, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 22.83, | |
| "learning_rate": 7.245155555555557e-05, | |
| "loss": 1.0864, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.83, | |
| "eval_cer": 0.13394260888350154, | |
| "eval_loss": 0.30638357996940613, | |
| "eval_runtime": 205.7125, | |
| "eval_samples_per_second": 21.729, | |
| "eval_steps_per_second": 1.361, | |
| "eval_wer": 0.29559965174490316, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 7.111822222222222e-05, | |
| "loss": 1.0493, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "eval_cer": 0.1299701162944205, | |
| "eval_loss": 0.29015836119651794, | |
| "eval_runtime": 205.4201, | |
| "eval_samples_per_second": 21.76, | |
| "eval_steps_per_second": 1.363, | |
| "eval_wer": 0.2849162011173184, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "learning_rate": 6.97848888888889e-05, | |
| "loss": 1.0193, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "eval_cer": 0.12554156715432713, | |
| "eval_loss": 0.27857705950737, | |
| "eval_runtime": 204.6144, | |
| "eval_samples_per_second": 21.846, | |
| "eval_steps_per_second": 1.368, | |
| "eval_wer": 0.2710222738155699, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "learning_rate": 6.845333333333334e-05, | |
| "loss": 0.991, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "eval_cer": 0.12242118022634807, | |
| "eval_loss": 0.27255550026893616, | |
| "eval_runtime": 205.3956, | |
| "eval_samples_per_second": 21.763, | |
| "eval_steps_per_second": 1.363, | |
| "eval_wer": 0.2632046724225495, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 35.87, | |
| "learning_rate": 6.712000000000001e-05, | |
| "loss": 0.965, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 35.87, | |
| "eval_cer": 0.1227452204073305, | |
| "eval_loss": 0.2663341462612152, | |
| "eval_runtime": 207.401, | |
| "eval_samples_per_second": 21.552, | |
| "eval_steps_per_second": 1.35, | |
| "eval_wer": 0.2638395124428644, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 39.13, | |
| "learning_rate": 6.578755555555556e-05, | |
| "loss": 0.9443, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 39.13, | |
| "eval_cer": 0.11871272037732679, | |
| "eval_loss": 0.2498483508825302, | |
| "eval_runtime": 206.2679, | |
| "eval_samples_per_second": 21.671, | |
| "eval_steps_per_second": 1.357, | |
| "eval_wer": 0.2540629761300152, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 42.39, | |
| "learning_rate": 6.445422222222224e-05, | |
| "loss": 0.9258, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 42.39, | |
| "eval_cer": 0.11627641827586621, | |
| "eval_loss": 0.24673064053058624, | |
| "eval_runtime": 211.3131, | |
| "eval_samples_per_second": 21.153, | |
| "eval_steps_per_second": 1.325, | |
| "eval_wer": 0.24822244794311835, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 45.65, | |
| "learning_rate": 6.312088888888889e-05, | |
| "loss": 0.9062, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 45.65, | |
| "eval_cer": 0.11399013477671231, | |
| "eval_loss": 0.23752263188362122, | |
| "eval_runtime": 210.4679, | |
| "eval_samples_per_second": 21.238, | |
| "eval_steps_per_second": 1.33, | |
| "eval_wer": 0.24243633461510555, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 48.91, | |
| "learning_rate": 6.178844444444445e-05, | |
| "loss": 0.8797, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 48.91, | |
| "eval_cer": 0.11319803655653302, | |
| "eval_loss": 0.23775836825370789, | |
| "eval_runtime": 200.5119, | |
| "eval_samples_per_second": 22.293, | |
| "eval_steps_per_second": 1.396, | |
| "eval_wer": 0.24310745120800986, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 52.17, | |
| "learning_rate": 6.0456000000000005e-05, | |
| "loss": 0.8691, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 52.17, | |
| "eval_cer": 0.11229792494269289, | |
| "eval_loss": 0.23617656528949738, | |
| "eval_runtime": 200.8473, | |
| "eval_samples_per_second": 22.256, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.23674091271856634, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 55.43, | |
| "learning_rate": 5.912266666666667e-05, | |
| "loss": 0.8554, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 55.43, | |
| "eval_cer": 0.11031767939224464, | |
| "eval_loss": 0.23144488036632538, | |
| "eval_runtime": 200.5411, | |
| "eval_samples_per_second": 22.29, | |
| "eval_steps_per_second": 1.396, | |
| "eval_wer": 0.23295001088297176, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 58.69, | |
| "learning_rate": 5.778933333333334e-05, | |
| "loss": 0.8393, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 58.69, | |
| "eval_cer": 0.10892550676283859, | |
| "eval_loss": 0.22345688939094543, | |
| "eval_runtime": 199.926, | |
| "eval_samples_per_second": 22.358, | |
| "eval_steps_per_second": 1.401, | |
| "eval_wer": 0.22779873757527389, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 61.96, | |
| "learning_rate": 5.645688888888889e-05, | |
| "loss": 0.8268, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 61.96, | |
| "eval_cer": 0.10809140333401342, | |
| "eval_loss": 0.21886605024337769, | |
| "eval_runtime": 200.5151, | |
| "eval_samples_per_second": 22.293, | |
| "eval_steps_per_second": 1.396, | |
| "eval_wer": 0.2251324094899514, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "learning_rate": 5.512355555555556e-05, | |
| "loss": 0.8116, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "eval_cer": 0.10730530585792639, | |
| "eval_loss": 0.22185355424880981, | |
| "eval_runtime": 199.8759, | |
| "eval_samples_per_second": 22.364, | |
| "eval_steps_per_second": 1.401, | |
| "eval_wer": 0.2221395922513241, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 68.48, | |
| "learning_rate": 5.3791111111111116e-05, | |
| "loss": 0.7981, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 68.48, | |
| "eval_cer": 0.10677724037780685, | |
| "eval_loss": 0.21808896958827972, | |
| "eval_runtime": 201.4847, | |
| "eval_samples_per_second": 22.185, | |
| "eval_steps_per_second": 1.39, | |
| "eval_wer": 0.221831241384314, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 71.74, | |
| "learning_rate": 5.245777777777778e-05, | |
| "loss": 0.7839, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 71.74, | |
| "eval_cer": 0.10549308114206161, | |
| "eval_loss": 0.21510420739650726, | |
| "eval_runtime": 200.7972, | |
| "eval_samples_per_second": 22.261, | |
| "eval_steps_per_second": 1.394, | |
| "eval_wer": 0.2175143292461728, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 5.112533333333334e-05, | |
| "loss": 0.7714, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_cer": 0.10464097548095964, | |
| "eval_loss": 0.21727131307125092, | |
| "eval_runtime": 200.0575, | |
| "eval_samples_per_second": 22.344, | |
| "eval_steps_per_second": 1.4, | |
| "eval_wer": 0.21642603206849018, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 78.26, | |
| "learning_rate": 4.9792e-05, | |
| "loss": 0.7582, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 78.26, | |
| "eval_cer": 0.10387288023714941, | |
| "eval_loss": 0.21982263028621674, | |
| "eval_runtime": 201.7359, | |
| "eval_samples_per_second": 22.158, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 0.2148298628745556, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 81.52, | |
| "learning_rate": 4.845955555555556e-05, | |
| "loss": 0.7486, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 81.52, | |
| "eval_cer": 0.10329080806019947, | |
| "eval_loss": 0.21195770800113678, | |
| "eval_runtime": 199.1645, | |
| "eval_samples_per_second": 22.444, | |
| "eval_steps_per_second": 1.406, | |
| "eval_wer": 0.21250816222883262, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 84.78, | |
| "learning_rate": 4.7127111111111114e-05, | |
| "loss": 0.7326, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 84.78, | |
| "eval_cer": 0.10266072993051138, | |
| "eval_loss": 0.21248655021190643, | |
| "eval_runtime": 199.3582, | |
| "eval_samples_per_second": 22.422, | |
| "eval_steps_per_second": 1.405, | |
| "eval_wer": 0.2124537473699485, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 88.04, | |
| "learning_rate": 4.5793777777777784e-05, | |
| "loss": 0.7263, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 88.04, | |
| "eval_cer": 0.1012565558129208, | |
| "eval_loss": 0.20737077295780182, | |
| "eval_runtime": 206.4127, | |
| "eval_samples_per_second": 21.656, | |
| "eval_steps_per_second": 1.357, | |
| "eval_wer": 0.2079735906551549, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 91.3, | |
| "learning_rate": 4.446044444444445e-05, | |
| "loss": 0.7142, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 91.3, | |
| "eval_cer": 0.10114254167516772, | |
| "eval_loss": 0.21454070508480072, | |
| "eval_runtime": 204.8009, | |
| "eval_samples_per_second": 21.826, | |
| "eval_steps_per_second": 1.367, | |
| "eval_wer": 0.2069397083363564, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 94.56, | |
| "learning_rate": 4.312711111111111e-05, | |
| "loss": 0.7044, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 94.56, | |
| "eval_cer": 0.10105253051378371, | |
| "eval_loss": 0.20794141292572021, | |
| "eval_runtime": 205.8959, | |
| "eval_samples_per_second": 21.71, | |
| "eval_steps_per_second": 1.36, | |
| "eval_wer": 0.2077922077922078, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 97.83, | |
| "learning_rate": 4.179555555555556e-05, | |
| "loss": 0.6975, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 97.83, | |
| "eval_cer": 0.0996543571402854, | |
| "eval_loss": 0.20573750138282776, | |
| "eval_runtime": 205.1157, | |
| "eval_samples_per_second": 21.793, | |
| "eval_steps_per_second": 1.365, | |
| "eval_wer": 0.2029855619241094, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 101.09, | |
| "learning_rate": 4.0462222222222225e-05, | |
| "loss": 0.683, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 101.09, | |
| "eval_cer": 0.09906628421924318, | |
| "eval_loss": 0.20465880632400513, | |
| "eval_runtime": 205.9334, | |
| "eval_samples_per_second": 21.706, | |
| "eval_steps_per_second": 1.36, | |
| "eval_wer": 0.20209678589566857, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 104.35, | |
| "learning_rate": 3.913066666666667e-05, | |
| "loss": 0.6763, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 104.35, | |
| "eval_cer": 0.0991922998451808, | |
| "eval_loss": 0.20821230113506317, | |
| "eval_runtime": 207.1439, | |
| "eval_samples_per_second": 21.579, | |
| "eval_steps_per_second": 1.352, | |
| "eval_wer": 0.20189726474642675, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 107.61, | |
| "learning_rate": 3.779733333333333e-05, | |
| "loss": 0.6687, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 107.61, | |
| "eval_cer": 0.09791414135352784, | |
| "eval_loss": 0.20512840151786804, | |
| "eval_runtime": 206.0617, | |
| "eval_samples_per_second": 21.693, | |
| "eval_steps_per_second": 1.359, | |
| "eval_wer": 0.19807008633824277, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 110.87, | |
| "learning_rate": 3.6464e-05, | |
| "loss": 0.6573, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 110.87, | |
| "eval_cer": 0.09767411158983715, | |
| "eval_loss": 0.20361328125, | |
| "eval_runtime": 205.3176, | |
| "eval_samples_per_second": 21.771, | |
| "eval_steps_per_second": 1.364, | |
| "eval_wer": 0.197616629180875, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 114.13, | |
| "learning_rate": 3.513155555555556e-05, | |
| "loss": 0.6497, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 114.13, | |
| "eval_cer": 0.09658197616504446, | |
| "eval_loss": 0.2067527323961258, | |
| "eval_runtime": 206.5547, | |
| "eval_samples_per_second": 21.641, | |
| "eval_steps_per_second": 1.356, | |
| "eval_wer": 0.19573024740622505, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 117.39, | |
| "learning_rate": 3.379822222222222e-05, | |
| "loss": 0.6406, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 117.39, | |
| "eval_cer": 0.09563985934255848, | |
| "eval_loss": 0.20097151398658752, | |
| "eval_runtime": 205.4638, | |
| "eval_samples_per_second": 21.756, | |
| "eval_steps_per_second": 1.363, | |
| "eval_wer": 0.1919212072843358, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 120.65, | |
| "learning_rate": 3.2466666666666665e-05, | |
| "loss": 0.6313, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 120.65, | |
| "eval_cer": 0.09662398137369034, | |
| "eval_loss": 0.20263761281967163, | |
| "eval_runtime": 210.159, | |
| "eval_samples_per_second": 21.27, | |
| "eval_steps_per_second": 1.332, | |
| "eval_wer": 0.19520423710367846, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 123.91, | |
| "learning_rate": 3.1133333333333336e-05, | |
| "loss": 0.6261, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 123.91, | |
| "eval_cer": 0.09623993375178522, | |
| "eval_loss": 0.20339860022068024, | |
| "eval_runtime": 208.9616, | |
| "eval_samples_per_second": 21.391, | |
| "eval_steps_per_second": 1.34, | |
| "eval_wer": 0.1936080679097439, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 127.17, | |
| "learning_rate": 2.9800000000000003e-05, | |
| "loss": 0.6179, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 127.17, | |
| "eval_cer": 0.09569986678348115, | |
| "eval_loss": 0.20739464461803436, | |
| "eval_runtime": 215.3433, | |
| "eval_samples_per_second": 20.758, | |
| "eval_steps_per_second": 1.3, | |
| "eval_wer": 0.1929550896031343, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 130.43, | |
| "learning_rate": 2.846755555555556e-05, | |
| "loss": 0.6082, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 130.43, | |
| "eval_cer": 0.09512979609471574, | |
| "eval_loss": 0.20194286108016968, | |
| "eval_runtime": 213.4283, | |
| "eval_samples_per_second": 20.944, | |
| "eval_steps_per_second": 1.312, | |
| "eval_wer": 0.1902887615178118, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 133.69, | |
| "learning_rate": 2.7134222222222226e-05, | |
| "loss": 0.6039, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 133.69, | |
| "eval_cer": 0.09520780576791522, | |
| "eval_loss": 0.20227235555648804, | |
| "eval_runtime": 209.2948, | |
| "eval_samples_per_second": 21.357, | |
| "eval_steps_per_second": 1.338, | |
| "eval_wer": 0.1912682289777262, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 136.96, | |
| "learning_rate": 2.580177777777778e-05, | |
| "loss": 0.5948, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 136.96, | |
| "eval_cer": 0.09544183478751365, | |
| "eval_loss": 0.20598873496055603, | |
| "eval_runtime": 207.4517, | |
| "eval_samples_per_second": 21.547, | |
| "eval_steps_per_second": 1.35, | |
| "eval_wer": 0.19108684611477908, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 140.22, | |
| "learning_rate": 2.4468444444444447e-05, | |
| "loss": 0.5867, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 140.22, | |
| "eval_cer": 0.09470374326416475, | |
| "eval_loss": 0.20384302735328674, | |
| "eval_runtime": 209.3468, | |
| "eval_samples_per_second": 21.352, | |
| "eval_steps_per_second": 1.337, | |
| "eval_wer": 0.18905535804977147, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 143.48, | |
| "learning_rate": 2.3136000000000003e-05, | |
| "loss": 0.5804, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 143.48, | |
| "eval_cer": 0.09508779088606988, | |
| "eval_loss": 0.2093733698129654, | |
| "eval_runtime": 212.8045, | |
| "eval_samples_per_second": 21.005, | |
| "eval_steps_per_second": 1.316, | |
| "eval_wer": 0.19039759123558006, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 146.74, | |
| "learning_rate": 2.1803555555555557e-05, | |
| "loss": 0.5757, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 146.74, | |
| "eval_cer": 0.09395365025263133, | |
| "eval_loss": 0.20662076771259308, | |
| "eval_runtime": 210.0316, | |
| "eval_samples_per_second": 21.283, | |
| "eval_steps_per_second": 1.333, | |
| "eval_wer": 0.18718711456141623, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 2.0470222222222224e-05, | |
| "loss": 0.5684, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "eval_cer": 0.09401965843764627, | |
| "eval_loss": 0.20758455991744995, | |
| "eval_runtime": 210.1895, | |
| "eval_samples_per_second": 21.267, | |
| "eval_steps_per_second": 1.332, | |
| "eval_wer": 0.1878038162954364, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 153.26, | |
| "learning_rate": 1.913777777777778e-05, | |
| "loss": 0.5622, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 153.26, | |
| "eval_cer": 0.09470974400825702, | |
| "eval_loss": 0.21354267001152039, | |
| "eval_runtime": 210.2099, | |
| "eval_samples_per_second": 21.264, | |
| "eval_steps_per_second": 1.332, | |
| "eval_wer": 0.1891641877675397, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 156.52, | |
| "learning_rate": 1.7805333333333334e-05, | |
| "loss": 0.5572, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 156.52, | |
| "eval_cer": 0.09412167108721481, | |
| "eval_loss": 0.21093089878559113, | |
| "eval_runtime": 208.6793, | |
| "eval_samples_per_second": 21.42, | |
| "eval_steps_per_second": 1.342, | |
| "eval_wer": 0.18854748603351956, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 159.78, | |
| "learning_rate": 1.647288888888889e-05, | |
| "loss": 0.5513, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 159.78, | |
| "eval_cer": 0.09389964355580092, | |
| "eval_loss": 0.20995952188968658, | |
| "eval_runtime": 211.9593, | |
| "eval_samples_per_second": 21.089, | |
| "eval_steps_per_second": 1.321, | |
| "eval_wer": 0.1872959442791845, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 163.04, | |
| "learning_rate": 1.5139555555555556e-05, | |
| "loss": 0.5463, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 163.04, | |
| "eval_cer": 0.09368961751257156, | |
| "eval_loss": 0.2099747508764267, | |
| "eval_runtime": 206.1662, | |
| "eval_samples_per_second": 21.682, | |
| "eval_steps_per_second": 1.358, | |
| "eval_wer": 0.18666110425886961, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 166.3, | |
| "learning_rate": 1.3807111111111111e-05, | |
| "loss": 0.5402, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 166.3, | |
| "eval_cer": 0.0937376234653097, | |
| "eval_loss": 0.2125527560710907, | |
| "eval_runtime": 206.4719, | |
| "eval_samples_per_second": 21.649, | |
| "eval_steps_per_second": 1.356, | |
| "eval_wer": 0.18675179569034317, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 169.56, | |
| "learning_rate": 1.2473777777777778e-05, | |
| "loss": 0.5361, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 169.56, | |
| "eval_cer": 0.09316155203245202, | |
| "eval_loss": 0.20593136548995972, | |
| "eval_runtime": 204.7532, | |
| "eval_samples_per_second": 21.831, | |
| "eval_steps_per_second": 1.368, | |
| "eval_wer": 0.18530073278676631, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 172.83, | |
| "learning_rate": 1.1140444444444445e-05, | |
| "loss": 0.5322, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 172.83, | |
| "eval_cer": 0.09320955798519016, | |
| "eval_loss": 0.20699192583560944, | |
| "eval_runtime": 206.6944, | |
| "eval_samples_per_second": 21.626, | |
| "eval_steps_per_second": 1.355, | |
| "eval_wer": 0.1852825945004716, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 176.09, | |
| "learning_rate": 9.807111111111112e-06, | |
| "loss": 0.5274, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 176.09, | |
| "eval_cer": 0.0925194724145794, | |
| "eval_loss": 0.2075292468070984, | |
| "eval_runtime": 205.8591, | |
| "eval_samples_per_second": 21.714, | |
| "eval_steps_per_second": 1.36, | |
| "eval_wer": 0.18423057389537836, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 179.35, | |
| "learning_rate": 8.474666666666667e-06, | |
| "loss": 0.5229, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 179.35, | |
| "eval_cer": 0.09307154087106802, | |
| "eval_loss": 0.2078840136528015, | |
| "eval_runtime": 207.0411, | |
| "eval_samples_per_second": 21.59, | |
| "eval_steps_per_second": 1.352, | |
| "eval_wer": 0.18519190306899805, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 182.61, | |
| "learning_rate": 7.141333333333333e-06, | |
| "loss": 0.5167, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 182.61, | |
| "eval_cer": 0.09254347539094848, | |
| "eval_loss": 0.2052079439163208, | |
| "eval_runtime": 205.153, | |
| "eval_samples_per_second": 21.789, | |
| "eval_steps_per_second": 1.365, | |
| "eval_wer": 0.1834869041572952, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 185.87, | |
| "learning_rate": 5.80888888888889e-06, | |
| "loss": 0.515, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 185.87, | |
| "eval_cer": 0.0924474634854722, | |
| "eval_loss": 0.20651493966579437, | |
| "eval_runtime": 206.3346, | |
| "eval_samples_per_second": 21.664, | |
| "eval_steps_per_second": 1.357, | |
| "eval_wer": 0.1834869041572952, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 189.13, | |
| "learning_rate": 4.476444444444445e-06, | |
| "loss": 0.5116, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 189.13, | |
| "eval_cer": 0.09247746720593354, | |
| "eval_loss": 0.20912431180477142, | |
| "eval_runtime": 206.6532, | |
| "eval_samples_per_second": 21.63, | |
| "eval_steps_per_second": 1.355, | |
| "eval_wer": 0.18323296814916926, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 192.39, | |
| "learning_rate": 3.143111111111111e-06, | |
| "loss": 0.5097, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 192.39, | |
| "eval_cer": 0.0925194724145794, | |
| "eval_loss": 0.2088840752840042, | |
| "eval_runtime": 206.9892, | |
| "eval_samples_per_second": 21.595, | |
| "eval_steps_per_second": 1.353, | |
| "eval_wer": 0.18355945730247406, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 195.65, | |
| "learning_rate": 1.8106666666666667e-06, | |
| "loss": 0.5056, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 195.65, | |
| "eval_cer": 0.09236345306818046, | |
| "eval_loss": 0.2069932371377945, | |
| "eval_runtime": 205.214, | |
| "eval_samples_per_second": 21.782, | |
| "eval_steps_per_second": 1.364, | |
| "eval_wer": 0.1829608938547486, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 198.91, | |
| "learning_rate": 4.782222222222222e-07, | |
| "loss": 0.5068, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 198.91, | |
| "eval_cer": 0.09240545827682632, | |
| "eval_loss": 0.20784138143062592, | |
| "eval_runtime": 205.9346, | |
| "eval_samples_per_second": 21.706, | |
| "eval_steps_per_second": 1.36, | |
| "eval_wer": 0.183124138431401, | |
| "step": 91500 | |
| } | |
| ], | |
| "max_steps": 92000, | |
| "num_train_epochs": 200, | |
| "total_flos": 1.0091178676342205e+21, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |