| { | |
| "best_metric": 7.215361500971087, | |
| "best_model_checkpoint": "./checkpoint-9000", | |
| "epoch": 5.048, | |
| "eval_steps": 500, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025, | |
| "grad_norm": 6.131621360778809, | |
| "learning_rate": 2.1875e-07, | |
| "loss": 0.9345, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 6.021520137786865, | |
| "learning_rate": 4.375e-07, | |
| "loss": 0.8231, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0075, | |
| "grad_norm": 5.526496410369873, | |
| "learning_rate": 6.5625e-07, | |
| "loss": 0.5623, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.9277825355529785, | |
| "learning_rate": 8.75e-07, | |
| "loss": 0.4173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 4.292990684509277, | |
| "learning_rate": 1.09375e-06, | |
| "loss": 0.385, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 5.749295234680176, | |
| "learning_rate": 1.3125e-06, | |
| "loss": 0.3931, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0175, | |
| "grad_norm": 3.8306965827941895, | |
| "learning_rate": 1.5312499999999997e-06, | |
| "loss": 0.3516, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 4.687748908996582, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.3235, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0225, | |
| "grad_norm": 4.232759952545166, | |
| "learning_rate": 1.96875e-06, | |
| "loss": 0.3314, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 4.185751914978027, | |
| "learning_rate": 2.1875e-06, | |
| "loss": 0.309, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0275, | |
| "grad_norm": 4.818612098693848, | |
| "learning_rate": 2.40625e-06, | |
| "loss": 0.2991, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 4.171736717224121, | |
| "learning_rate": 2.625e-06, | |
| "loss": 0.2832, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.0325, | |
| "grad_norm": 5.217376708984375, | |
| "learning_rate": 2.8437499999999997e-06, | |
| "loss": 0.2873, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 4.671106815338135, | |
| "learning_rate": 3.0624999999999995e-06, | |
| "loss": 0.2957, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 3.9175262451171875, | |
| "learning_rate": 3.2812499999999997e-06, | |
| "loss": 0.2634, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 4.647582054138184, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.2541, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.0425, | |
| "grad_norm": 3.25675368309021, | |
| "learning_rate": 3.7187499999999998e-06, | |
| "loss": 0.2244, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 4.597206115722656, | |
| "learning_rate": 3.9375e-06, | |
| "loss": 0.2492, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.0475, | |
| "grad_norm": 4.602332592010498, | |
| "learning_rate": 4.156249999999999e-06, | |
| "loss": 0.246, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 3.6419622898101807, | |
| "learning_rate": 4.375e-06, | |
| "loss": 0.2208, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.2592349350452423, | |
| "eval_runtime": 4116.5906, | |
| "eval_samples_per_second": 3.311, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 20.691487412510533, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0525, | |
| "grad_norm": 3.6599488258361816, | |
| "learning_rate": 4.363486842105263e-06, | |
| "loss": 0.2539, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 3.6934616565704346, | |
| "learning_rate": 4.351973684210526e-06, | |
| "loss": 0.2313, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.0575, | |
| "grad_norm": 3.7546138763427734, | |
| "learning_rate": 4.340460526315789e-06, | |
| "loss": 0.2272, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 3.096877098083496, | |
| "learning_rate": 4.3289473684210525e-06, | |
| "loss": 0.2373, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 3.572812795639038, | |
| "learning_rate": 4.3174342105263155e-06, | |
| "loss": 0.2285, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 3.3494396209716797, | |
| "learning_rate": 4.3059210526315785e-06, | |
| "loss": 0.2293, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.0675, | |
| "grad_norm": 3.5156869888305664, | |
| "learning_rate": 4.2944078947368415e-06, | |
| "loss": 0.2063, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.698807716369629, | |
| "learning_rate": 4.282894736842105e-06, | |
| "loss": 0.2113, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.0725, | |
| "grad_norm": 3.716585636138916, | |
| "learning_rate": 4.271381578947368e-06, | |
| "loss": 0.2055, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 4.204227924346924, | |
| "learning_rate": 4.2598684210526314e-06, | |
| "loss": 0.2114, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.0775, | |
| "grad_norm": 3.479562282562256, | |
| "learning_rate": 4.2483552631578944e-06, | |
| "loss": 0.2224, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 4.5203094482421875, | |
| "learning_rate": 4.2368421052631575e-06, | |
| "loss": 0.2523, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.0825, | |
| "grad_norm": 3.6081738471984863, | |
| "learning_rate": 4.2253289473684205e-06, | |
| "loss": 0.2383, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 3.2602758407592773, | |
| "learning_rate": 4.2138157894736835e-06, | |
| "loss": 0.1808, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 3.6786868572235107, | |
| "learning_rate": 4.202302631578947e-06, | |
| "loss": 0.1747, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 3.1120803356170654, | |
| "learning_rate": 4.19078947368421e-06, | |
| "loss": 0.1662, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.0925, | |
| "grad_norm": 3.1962203979492188, | |
| "learning_rate": 4.179276315789473e-06, | |
| "loss": 0.1771, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 3.172363758087158, | |
| "learning_rate": 4.167763157894736e-06, | |
| "loss": 0.1751, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.0975, | |
| "grad_norm": 2.4304590225219727, | |
| "learning_rate": 4.156249999999999e-06, | |
| "loss": 0.1701, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.193345308303833, | |
| "learning_rate": 4.144736842105262e-06, | |
| "loss": 0.1489, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.1971057653427124, | |
| "eval_runtime": 4130.6867, | |
| "eval_samples_per_second": 3.3, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 14.68265601524424, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1025, | |
| "grad_norm": 3.322065591812134, | |
| "learning_rate": 4.133223684210526e-06, | |
| "loss": 0.1701, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 3.5462722778320312, | |
| "learning_rate": 4.121710526315789e-06, | |
| "loss": 0.1875, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.1075, | |
| "grad_norm": 3.39326810836792, | |
| "learning_rate": 4.110197368421052e-06, | |
| "loss": 0.1506, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.9165821075439453, | |
| "learning_rate": 4.098684210526315e-06, | |
| "loss": 0.1525, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 3.262007236480713, | |
| "learning_rate": 4.087171052631578e-06, | |
| "loss": 0.157, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 2.4523119926452637, | |
| "learning_rate": 4.075657894736842e-06, | |
| "loss": 0.1416, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.1175, | |
| "grad_norm": 2.7651798725128174, | |
| "learning_rate": 4.064144736842105e-06, | |
| "loss": 0.1527, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 3.609523296356201, | |
| "learning_rate": 4.052631578947368e-06, | |
| "loss": 0.1822, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.1225, | |
| "grad_norm": 3.8101985454559326, | |
| "learning_rate": 4.041118421052631e-06, | |
| "loss": 0.1703, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 3.8921287059783936, | |
| "learning_rate": 4.029605263157894e-06, | |
| "loss": 0.1924, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.1275, | |
| "grad_norm": 4.463279724121094, | |
| "learning_rate": 4.018092105263157e-06, | |
| "loss": 0.1818, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 3.6556308269500732, | |
| "learning_rate": 4.00657894736842e-06, | |
| "loss": 0.1726, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.1325, | |
| "grad_norm": 2.98067569732666, | |
| "learning_rate": 3.995065789473683e-06, | |
| "loss": 0.174, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 2.8287429809570312, | |
| "learning_rate": 3.983552631578947e-06, | |
| "loss": 0.1631, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.1375, | |
| "grad_norm": 2.6438794136047363, | |
| "learning_rate": 3.97203947368421e-06, | |
| "loss": 0.1475, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.513123035430908, | |
| "learning_rate": 3.960526315789473e-06, | |
| "loss": 0.1457, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.1425, | |
| "grad_norm": 2.4688916206359863, | |
| "learning_rate": 3.949013157894737e-06, | |
| "loss": 0.1375, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 4.005943775177002, | |
| "learning_rate": 3.9375e-06, | |
| "loss": 0.1623, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.1475, | |
| "grad_norm": 2.91786789894104, | |
| "learning_rate": 3.925986842105263e-06, | |
| "loss": 0.1701, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.5332415103912354, | |
| "learning_rate": 3.914473684210526e-06, | |
| "loss": 0.1973, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.17469166219234467, | |
| "eval_runtime": 4132.0041, | |
| "eval_samples_per_second": 3.299, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 12.377697973542453, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1525, | |
| "grad_norm": 4.05070686340332, | |
| "learning_rate": 3.902960526315789e-06, | |
| "loss": 0.1796, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 2.989821195602417, | |
| "learning_rate": 3.891447368421052e-06, | |
| "loss": 0.1561, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.1575, | |
| "grad_norm": 2.9603219032287598, | |
| "learning_rate": 3.879934210526315e-06, | |
| "loss": 0.1609, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.2663583755493164, | |
| "learning_rate": 3.868421052631579e-06, | |
| "loss": 0.1833, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.1625, | |
| "grad_norm": 3.459775686264038, | |
| "learning_rate": 3.856907894736842e-06, | |
| "loss": 0.1727, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 3.427720069885254, | |
| "learning_rate": 3.845394736842105e-06, | |
| "loss": 0.181, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.1675, | |
| "grad_norm": 4.471118450164795, | |
| "learning_rate": 3.833881578947368e-06, | |
| "loss": 0.1536, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 3.1428306102752686, | |
| "learning_rate": 3.822368421052632e-06, | |
| "loss": 0.1372, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0021, | |
| "grad_norm": 2.8270132541656494, | |
| "learning_rate": 3.8108552631578944e-06, | |
| "loss": 0.1454, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 1.0046, | |
| "grad_norm": 3.0873589515686035, | |
| "learning_rate": 3.799342105263158e-06, | |
| "loss": 0.1303, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.0071, | |
| "grad_norm": 3.187711000442505, | |
| "learning_rate": 3.787828947368421e-06, | |
| "loss": 0.1383, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 1.0096, | |
| "grad_norm": 3.1710643768310547, | |
| "learning_rate": 3.776315789473684e-06, | |
| "loss": 0.1626, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.0121, | |
| "grad_norm": 3.4516818523406982, | |
| "learning_rate": 3.7648026315789473e-06, | |
| "loss": 0.1405, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.0146, | |
| "grad_norm": 2.930408000946045, | |
| "learning_rate": 3.7532894736842103e-06, | |
| "loss": 0.143, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.0171, | |
| "grad_norm": 3.066941261291504, | |
| "learning_rate": 3.7417763157894733e-06, | |
| "loss": 0.1437, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.0196, | |
| "grad_norm": 3.389916181564331, | |
| "learning_rate": 3.7302631578947363e-06, | |
| "loss": 0.1289, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.0221, | |
| "grad_norm": 3.048574209213257, | |
| "learning_rate": 3.7187499999999998e-06, | |
| "loss": 0.1415, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.0246, | |
| "grad_norm": 2.5267295837402344, | |
| "learning_rate": 3.7072368421052628e-06, | |
| "loss": 0.1386, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.0271, | |
| "grad_norm": 3.151757001876831, | |
| "learning_rate": 3.6957236842105258e-06, | |
| "loss": 0.1436, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.0296, | |
| "grad_norm": 3.629039764404297, | |
| "learning_rate": 3.684210526315789e-06, | |
| "loss": 0.1353, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0296, | |
| "eval_loss": 0.1527385264635086, | |
| "eval_runtime": 4116.9756, | |
| "eval_samples_per_second": 3.311, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 10.719520685990693, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0321, | |
| "grad_norm": 1.8788173198699951, | |
| "learning_rate": 3.6726973684210522e-06, | |
| "loss": 0.1322, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.0346, | |
| "grad_norm": 2.587233066558838, | |
| "learning_rate": 3.6611842105263157e-06, | |
| "loss": 0.1176, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.0371, | |
| "grad_norm": 4.001532077789307, | |
| "learning_rate": 3.6496710526315787e-06, | |
| "loss": 0.1233, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.0396, | |
| "grad_norm": 3.3947739601135254, | |
| "learning_rate": 3.638157894736842e-06, | |
| "loss": 0.1188, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.0421, | |
| "grad_norm": 3.4743120670318604, | |
| "learning_rate": 3.626644736842105e-06, | |
| "loss": 0.1318, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.0446, | |
| "grad_norm": 2.9288718700408936, | |
| "learning_rate": 3.615131578947368e-06, | |
| "loss": 0.1224, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.0471, | |
| "grad_norm": 2.6081368923187256, | |
| "learning_rate": 3.603618421052631e-06, | |
| "loss": 0.1232, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.0496, | |
| "grad_norm": 2.4068429470062256, | |
| "learning_rate": 3.5921052631578946e-06, | |
| "loss": 0.1073, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.0521, | |
| "grad_norm": 3.049074411392212, | |
| "learning_rate": 3.5805921052631576e-06, | |
| "loss": 0.1071, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.0546, | |
| "grad_norm": 2.0809032917022705, | |
| "learning_rate": 3.5690789473684206e-06, | |
| "loss": 0.1217, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.0571, | |
| "grad_norm": 3.0854332447052, | |
| "learning_rate": 3.5575657894736836e-06, | |
| "loss": 0.1332, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.0596, | |
| "grad_norm": 3.580145835876465, | |
| "learning_rate": 3.546052631578947e-06, | |
| "loss": 0.131, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.0621, | |
| "grad_norm": 3.8924479484558105, | |
| "learning_rate": 3.53453947368421e-06, | |
| "loss": 0.136, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.0646, | |
| "grad_norm": 2.8398871421813965, | |
| "learning_rate": 3.523026315789473e-06, | |
| "loss": 0.1081, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.0671, | |
| "grad_norm": 3.007026195526123, | |
| "learning_rate": 3.511513157894737e-06, | |
| "loss": 0.1115, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.0695999999999999, | |
| "grad_norm": 1.5712552070617676, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.1183, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.0721, | |
| "grad_norm": 3.844963312149048, | |
| "learning_rate": 3.488486842105263e-06, | |
| "loss": 0.113, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.0746, | |
| "grad_norm": 2.8939759731292725, | |
| "learning_rate": 3.476973684210526e-06, | |
| "loss": 0.1115, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.0771, | |
| "grad_norm": 1.8150537014007568, | |
| "learning_rate": 3.4654605263157894e-06, | |
| "loss": 0.1117, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.0796000000000001, | |
| "grad_norm": 2.839418649673462, | |
| "learning_rate": 3.4539473684210524e-06, | |
| "loss": 0.1065, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0796000000000001, | |
| "eval_loss": 0.1456422209739685, | |
| "eval_runtime": 4133.4016, | |
| "eval_samples_per_second": 3.298, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 9.869361281102277, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0821, | |
| "grad_norm": 3.4274985790252686, | |
| "learning_rate": 3.4424342105263154e-06, | |
| "loss": 0.1067, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.0846, | |
| "grad_norm": 2.2946057319641113, | |
| "learning_rate": 3.4309210526315784e-06, | |
| "loss": 0.1038, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.0871, | |
| "grad_norm": 2.5364551544189453, | |
| "learning_rate": 3.419407894736842e-06, | |
| "loss": 0.1073, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.0896, | |
| "grad_norm": 2.9779515266418457, | |
| "learning_rate": 3.4083552631578944e-06, | |
| "loss": 0.1067, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.0921, | |
| "grad_norm": 2.502685308456421, | |
| "learning_rate": 3.3968421052631574e-06, | |
| "loss": 0.1229, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.0946, | |
| "grad_norm": 2.181756019592285, | |
| "learning_rate": 3.3853289473684205e-06, | |
| "loss": 0.1071, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.0971, | |
| "grad_norm": 2.428738594055176, | |
| "learning_rate": 3.3738157894736843e-06, | |
| "loss": 0.101, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 1.0996, | |
| "grad_norm": 3.797952651977539, | |
| "learning_rate": 3.3623026315789473e-06, | |
| "loss": 0.1198, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.1021, | |
| "grad_norm": 2.9902758598327637, | |
| "learning_rate": 3.3507894736842103e-06, | |
| "loss": 0.1013, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 1.1046, | |
| "grad_norm": 3.0514307022094727, | |
| "learning_rate": 3.3392763157894734e-06, | |
| "loss": 0.1075, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.1071, | |
| "grad_norm": 3.2877554893493652, | |
| "learning_rate": 3.327763157894737e-06, | |
| "loss": 0.1059, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 1.1096, | |
| "grad_norm": 2.3952691555023193, | |
| "learning_rate": 3.31625e-06, | |
| "loss": 0.0926, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1121, | |
| "grad_norm": 2.2840464115142822, | |
| "learning_rate": 3.304736842105263e-06, | |
| "loss": 0.1048, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 1.1146, | |
| "grad_norm": 2.7062416076660156, | |
| "learning_rate": 3.293223684210526e-06, | |
| "loss": 0.1049, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.1171, | |
| "grad_norm": 2.971315860748291, | |
| "learning_rate": 3.2817105263157893e-06, | |
| "loss": 0.1073, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.1196, | |
| "grad_norm": 2.8689844608306885, | |
| "learning_rate": 3.2701973684210523e-06, | |
| "loss": 0.1141, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.1221, | |
| "grad_norm": 3.6150734424591064, | |
| "learning_rate": 3.2586842105263153e-06, | |
| "loss": 0.1066, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 1.1246, | |
| "grad_norm": 2.3004024028778076, | |
| "learning_rate": 3.2471710526315783e-06, | |
| "loss": 0.1248, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.1271, | |
| "grad_norm": 2.5995240211486816, | |
| "learning_rate": 3.2356578947368417e-06, | |
| "loss": 0.0972, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 1.1296, | |
| "grad_norm": 2.957960367202759, | |
| "learning_rate": 3.224144736842105e-06, | |
| "loss": 0.106, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1296, | |
| "eval_loss": 0.13624447584152222, | |
| "eval_runtime": 4123.4662, | |
| "eval_samples_per_second": 3.305, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 9.09249148008355, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1320999999999999, | |
| "grad_norm": 2.653007984161377, | |
| "learning_rate": 3.212631578947368e-06, | |
| "loss": 0.1083, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 1.1346, | |
| "grad_norm": 2.6895744800567627, | |
| "learning_rate": 3.2011184210526316e-06, | |
| "loss": 0.1119, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.1371, | |
| "grad_norm": 2.1507463455200195, | |
| "learning_rate": 3.1896052631578946e-06, | |
| "loss": 0.0944, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 1.1396, | |
| "grad_norm": 3.61063289642334, | |
| "learning_rate": 3.1780921052631576e-06, | |
| "loss": 0.095, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.1421000000000001, | |
| "grad_norm": 2.570584774017334, | |
| "learning_rate": 3.1665789473684206e-06, | |
| "loss": 0.1076, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 1.1446, | |
| "grad_norm": 3.05507230758667, | |
| "learning_rate": 3.155065789473684e-06, | |
| "loss": 0.1175, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.1471, | |
| "grad_norm": 2.82817006111145, | |
| "learning_rate": 3.143552631578947e-06, | |
| "loss": 0.0965, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 1.1496, | |
| "grad_norm": 2.336517572402954, | |
| "learning_rate": 3.13203947368421e-06, | |
| "loss": 0.0955, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.1521, | |
| "grad_norm": 3.8640036582946777, | |
| "learning_rate": 3.120526315789473e-06, | |
| "loss": 0.1044, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 1.1546, | |
| "grad_norm": 3.7205588817596436, | |
| "learning_rate": 3.1090131578947366e-06, | |
| "loss": 0.1013, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.1571, | |
| "grad_norm": 2.1962900161743164, | |
| "learning_rate": 3.0974999999999996e-06, | |
| "loss": 0.0978, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 1.1596, | |
| "grad_norm": 3.3310599327087402, | |
| "learning_rate": 3.0859868421052626e-06, | |
| "loss": 0.1089, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.1621, | |
| "grad_norm": 2.699566602706909, | |
| "learning_rate": 3.074473684210526e-06, | |
| "loss": 0.1078, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 1.1646, | |
| "grad_norm": 3.79370379447937, | |
| "learning_rate": 3.0629605263157894e-06, | |
| "loss": 0.1118, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.1671, | |
| "grad_norm": 1.9741384983062744, | |
| "learning_rate": 3.0514473684210525e-06, | |
| "loss": 0.1119, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 1.1696, | |
| "grad_norm": 2.29034686088562, | |
| "learning_rate": 3.0399342105263155e-06, | |
| "loss": 0.1015, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.0017, | |
| "grad_norm": 2.011443853378296, | |
| "learning_rate": 3.028421052631579e-06, | |
| "loss": 0.0708, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 2.0042, | |
| "grad_norm": 1.2196134328842163, | |
| "learning_rate": 3.016907894736842e-06, | |
| "loss": 0.0668, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.0067, | |
| "grad_norm": 2.863933563232422, | |
| "learning_rate": 3.005394736842105e-06, | |
| "loss": 0.0673, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 2.0092, | |
| "grad_norm": 1.9341013431549072, | |
| "learning_rate": 2.9938815789473684e-06, | |
| "loss": 0.0718, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0092, | |
| "eval_loss": 0.13255682587623596, | |
| "eval_runtime": 4133.4892, | |
| "eval_samples_per_second": 3.297, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 8.542819451060867, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0117, | |
| "grad_norm": 2.795734405517578, | |
| "learning_rate": 2.9823684210526314e-06, | |
| "loss": 0.071, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 2.0142, | |
| "grad_norm": 1.982479214668274, | |
| "learning_rate": 2.9708552631578944e-06, | |
| "loss": 0.0629, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.0167, | |
| "grad_norm": 3.168161630630493, | |
| "learning_rate": 2.9593421052631574e-06, | |
| "loss": 0.0593, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 2.0192, | |
| "grad_norm": 2.259500741958618, | |
| "learning_rate": 2.947828947368421e-06, | |
| "loss": 0.0696, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.0217, | |
| "grad_norm": 2.1626062393188477, | |
| "learning_rate": 2.936315789473684e-06, | |
| "loss": 0.0687, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 2.0242, | |
| "grad_norm": 2.4419946670532227, | |
| "learning_rate": 2.924802631578947e-06, | |
| "loss": 0.0686, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.0267, | |
| "grad_norm": 2.445758819580078, | |
| "learning_rate": 2.9132894736842103e-06, | |
| "loss": 0.0631, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 2.0292, | |
| "grad_norm": 2.614476442337036, | |
| "learning_rate": 2.9017763157894737e-06, | |
| "loss": 0.0647, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.0317, | |
| "grad_norm": 1.4166672229766846, | |
| "learning_rate": 2.8902631578947367e-06, | |
| "loss": 0.0653, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 2.0342, | |
| "grad_norm": 1.8435245752334595, | |
| "learning_rate": 2.8787499999999998e-06, | |
| "loss": 0.0567, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.0367, | |
| "grad_norm": 1.8179950714111328, | |
| "learning_rate": 2.867236842105263e-06, | |
| "loss": 0.0636, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 2.0392, | |
| "grad_norm": 1.487122893333435, | |
| "learning_rate": 2.855723684210526e-06, | |
| "loss": 0.0598, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.0417, | |
| "grad_norm": 2.9211690425872803, | |
| "learning_rate": 2.8442105263157892e-06, | |
| "loss": 0.0599, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 2.0442, | |
| "grad_norm": 2.5018093585968018, | |
| "learning_rate": 2.8326973684210522e-06, | |
| "loss": 0.055, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.0467, | |
| "grad_norm": 2.186502456665039, | |
| "learning_rate": 2.8211842105263157e-06, | |
| "loss": 0.0533, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 2.0492, | |
| "grad_norm": 1.039233922958374, | |
| "learning_rate": 2.8096710526315787e-06, | |
| "loss": 0.0514, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.0517, | |
| "grad_norm": 1.871267557144165, | |
| "learning_rate": 2.7981578947368417e-06, | |
| "loss": 0.0512, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 2.0542, | |
| "grad_norm": 2.0849483013153076, | |
| "learning_rate": 2.7866447368421047e-06, | |
| "loss": 0.0579, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.0567, | |
| "grad_norm": 1.6887531280517578, | |
| "learning_rate": 2.775131578947368e-06, | |
| "loss": 0.0575, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 2.0592, | |
| "grad_norm": 1.88097083568573, | |
| "learning_rate": 2.763618421052631e-06, | |
| "loss": 0.0683, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0592, | |
| "eval_loss": 0.1342601627111435, | |
| "eval_runtime": 4125.8373, | |
| "eval_samples_per_second": 3.304, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 8.485103888013485, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0617, | |
| "grad_norm": 2.1877427101135254, | |
| "learning_rate": 2.7521052631578946e-06, | |
| "loss": 0.0614, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 2.0642, | |
| "grad_norm": 1.4176368713378906, | |
| "learning_rate": 2.740592105263158e-06, | |
| "loss": 0.0559, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 2.0667, | |
| "grad_norm": 2.4362101554870605, | |
| "learning_rate": 2.729078947368421e-06, | |
| "loss": 0.0593, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 2.0692, | |
| "grad_norm": 1.8663033246994019, | |
| "learning_rate": 2.717565789473684e-06, | |
| "loss": 0.0591, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.0717, | |
| "grad_norm": 1.627626657485962, | |
| "learning_rate": 2.706052631578947e-06, | |
| "loss": 0.0637, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 2.0742, | |
| "grad_norm": 2.2072463035583496, | |
| "learning_rate": 2.6945394736842105e-06, | |
| "loss": 0.0571, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 2.0767, | |
| "grad_norm": 1.7411611080169678, | |
| "learning_rate": 2.6830263157894735e-06, | |
| "loss": 0.0588, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 2.0792, | |
| "grad_norm": 1.324000358581543, | |
| "learning_rate": 2.6715131578947365e-06, | |
| "loss": 0.0482, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.0817, | |
| "grad_norm": 1.4138795137405396, | |
| "learning_rate": 2.6599999999999995e-06, | |
| "loss": 0.0477, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 2.0842, | |
| "grad_norm": 2.403547763824463, | |
| "learning_rate": 2.648486842105263e-06, | |
| "loss": 0.0558, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.0867, | |
| "grad_norm": 1.3718703985214233, | |
| "learning_rate": 2.636973684210526e-06, | |
| "loss": 0.0546, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 2.0892, | |
| "grad_norm": 2.296445369720459, | |
| "learning_rate": 2.625460526315789e-06, | |
| "loss": 0.0554, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.0917, | |
| "grad_norm": 2.3471312522888184, | |
| "learning_rate": 2.613947368421052e-06, | |
| "loss": 0.051, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 2.0942, | |
| "grad_norm": 1.6061975955963135, | |
| "learning_rate": 2.602434210526316e-06, | |
| "loss": 0.0548, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 2.0967, | |
| "grad_norm": 2.979126453399658, | |
| "learning_rate": 2.590921052631579e-06, | |
| "loss": 0.0492, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 2.0992, | |
| "grad_norm": 1.7963169813156128, | |
| "learning_rate": 2.579407894736842e-06, | |
| "loss": 0.0514, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.1017, | |
| "grad_norm": 2.4996039867401123, | |
| "learning_rate": 2.5678947368421053e-06, | |
| "loss": 0.0399, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 2.1042, | |
| "grad_norm": 1.7498191595077515, | |
| "learning_rate": 2.5563815789473683e-06, | |
| "loss": 0.0522, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 2.1067, | |
| "grad_norm": 1.413889765739441, | |
| "learning_rate": 2.5448684210526313e-06, | |
| "loss": 0.0517, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 2.1092, | |
| "grad_norm": 2.0956978797912598, | |
| "learning_rate": 2.5333552631578943e-06, | |
| "loss": 0.0482, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.1092, | |
| "eval_loss": 0.1336347758769989, | |
| "eval_runtime": 4119.9162, | |
| "eval_samples_per_second": 3.308, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 8.104914067939463, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.1117, | |
| "grad_norm": 3.138298749923706, | |
| "learning_rate": 2.5218421052631578e-06, | |
| "loss": 0.0568, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 2.1142, | |
| "grad_norm": 1.4262772798538208, | |
| "learning_rate": 2.510328947368421e-06, | |
| "loss": 0.0475, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 2.1167, | |
| "grad_norm": 3.3500139713287354, | |
| "learning_rate": 2.498815789473684e-06, | |
| "loss": 0.0474, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 2.1192, | |
| "grad_norm": 4.509912014007568, | |
| "learning_rate": 2.4873026315789472e-06, | |
| "loss": 0.0586, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.1217, | |
| "grad_norm": 2.1386468410491943, | |
| "learning_rate": 2.4757894736842102e-06, | |
| "loss": 0.062, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 2.1242, | |
| "grad_norm": 1.1121129989624023, | |
| "learning_rate": 2.4642763157894733e-06, | |
| "loss": 0.0563, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 2.1267, | |
| "grad_norm": 1.677538514137268, | |
| "learning_rate": 2.4527631578947363e-06, | |
| "loss": 0.0519, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 2.1292, | |
| "grad_norm": 1.579513430595398, | |
| "learning_rate": 2.44125e-06, | |
| "loss": 0.0544, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.1317, | |
| "grad_norm": 2.1100914478302, | |
| "learning_rate": 2.429736842105263e-06, | |
| "loss": 0.0578, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 2.1342, | |
| "grad_norm": 1.779682993888855, | |
| "learning_rate": 2.418223684210526e-06, | |
| "loss": 0.0486, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.1367, | |
| "grad_norm": 1.7443439960479736, | |
| "learning_rate": 2.4067105263157896e-06, | |
| "loss": 0.0534, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 2.1391999999999998, | |
| "grad_norm": 1.9388935565948486, | |
| "learning_rate": 2.3951973684210526e-06, | |
| "loss": 0.0516, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.1417, | |
| "grad_norm": 1.82517409324646, | |
| "learning_rate": 2.3836842105263156e-06, | |
| "loss": 0.0451, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 2.1442, | |
| "grad_norm": 1.9101967811584473, | |
| "learning_rate": 2.3721710526315786e-06, | |
| "loss": 0.0546, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 2.1467, | |
| "grad_norm": 1.7242915630340576, | |
| "learning_rate": 2.360657894736842e-06, | |
| "loss": 0.0495, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 2.1492, | |
| "grad_norm": 1.9127079248428345, | |
| "learning_rate": 2.349144736842105e-06, | |
| "loss": 0.0465, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.1517, | |
| "grad_norm": 2.7716519832611084, | |
| "learning_rate": 2.337631578947368e-06, | |
| "loss": 0.0493, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 2.1542, | |
| "grad_norm": 3.141706705093384, | |
| "learning_rate": 2.326118421052631e-06, | |
| "loss": 0.046, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.1567, | |
| "grad_norm": 2.2624270915985107, | |
| "learning_rate": 2.3146052631578945e-06, | |
| "loss": 0.0522, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 2.1592000000000002, | |
| "grad_norm": 1.2777652740478516, | |
| "learning_rate": 2.3030921052631575e-06, | |
| "loss": 0.0548, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1592000000000002, | |
| "eval_loss": 0.13162237405776978, | |
| "eval_runtime": 4127.2085, | |
| "eval_samples_per_second": 3.302, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 7.9244384184103485, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1617, | |
| "grad_norm": 2.106818675994873, | |
| "learning_rate": 2.2915789473684206e-06, | |
| "loss": 0.0527, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 2.1642, | |
| "grad_norm": 2.2705554962158203, | |
| "learning_rate": 2.2800657894736844e-06, | |
| "loss": 0.0483, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 2.1667, | |
| "grad_norm": 1.5468271970748901, | |
| "learning_rate": 2.2685526315789474e-06, | |
| "loss": 0.0516, | |
| "step": 5075 | |
| }, | |
| { | |
| "epoch": 2.1692, | |
| "grad_norm": 2.0331270694732666, | |
| "learning_rate": 2.2570394736842104e-06, | |
| "loss": 0.0551, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.0013, | |
| "grad_norm": 1.107423186302185, | |
| "learning_rate": 2.2455263157894734e-06, | |
| "loss": 0.0434, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 3.0038, | |
| "grad_norm": 3.9103100299835205, | |
| "learning_rate": 2.234013157894737e-06, | |
| "loss": 0.0362, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 3.0063, | |
| "grad_norm": 1.193088173866272, | |
| "learning_rate": 2.2225e-06, | |
| "loss": 0.0327, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 3.0088, | |
| "grad_norm": 1.0432852506637573, | |
| "learning_rate": 2.210986842105263e-06, | |
| "loss": 0.0326, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.0113, | |
| "grad_norm": 0.7116020917892456, | |
| "learning_rate": 2.199473684210526e-06, | |
| "loss": 0.0296, | |
| "step": 5225 | |
| }, | |
| { | |
| "epoch": 3.0138, | |
| "grad_norm": 2.009617805480957, | |
| "learning_rate": 2.1879605263157894e-06, | |
| "loss": 0.0367, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 3.0163, | |
| "grad_norm": 1.9047244787216187, | |
| "learning_rate": 2.1764473684210524e-06, | |
| "loss": 0.0347, | |
| "step": 5275 | |
| }, | |
| { | |
| "epoch": 3.0188, | |
| "grad_norm": 1.630439043045044, | |
| "learning_rate": 2.164934210526316e-06, | |
| "loss": 0.0291, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.0213, | |
| "grad_norm": 1.4158824682235718, | |
| "learning_rate": 2.153421052631579e-06, | |
| "loss": 0.0321, | |
| "step": 5325 | |
| }, | |
| { | |
| "epoch": 3.0238, | |
| "grad_norm": 1.2792794704437256, | |
| "learning_rate": 2.141907894736842e-06, | |
| "loss": 0.0338, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 3.0263, | |
| "grad_norm": 1.6505346298217773, | |
| "learning_rate": 2.1303947368421053e-06, | |
| "loss": 0.0348, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 3.0288, | |
| "grad_norm": 1.5343618392944336, | |
| "learning_rate": 2.1188815789473683e-06, | |
| "loss": 0.0318, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.0313, | |
| "grad_norm": 1.8325493335723877, | |
| "learning_rate": 2.1073684210526313e-06, | |
| "loss": 0.0333, | |
| "step": 5425 | |
| }, | |
| { | |
| "epoch": 3.0338, | |
| "grad_norm": 1.7224900722503662, | |
| "learning_rate": 2.0958552631578943e-06, | |
| "loss": 0.0322, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 3.0362999999999998, | |
| "grad_norm": 1.3443737030029297, | |
| "learning_rate": 2.0843421052631577e-06, | |
| "loss": 0.0304, | |
| "step": 5475 | |
| }, | |
| { | |
| "epoch": 3.0388, | |
| "grad_norm": 1.3260679244995117, | |
| "learning_rate": 2.0728289473684207e-06, | |
| "loss": 0.0282, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0388, | |
| "eval_loss": 0.13909843564033508, | |
| "eval_runtime": 4135.2147, | |
| "eval_samples_per_second": 3.296, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 7.8181684927992965, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0413, | |
| "grad_norm": 1.0075204372406006, | |
| "learning_rate": 2.061315789473684e-06, | |
| "loss": 0.0308, | |
| "step": 5525 | |
| }, | |
| { | |
| "epoch": 3.0438, | |
| "grad_norm": 1.0206842422485352, | |
| "learning_rate": 2.049802631578947e-06, | |
| "loss": 0.0306, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 3.0463, | |
| "grad_norm": 1.411301851272583, | |
| "learning_rate": 2.03828947368421e-06, | |
| "loss": 0.0243, | |
| "step": 5575 | |
| }, | |
| { | |
| "epoch": 3.0488, | |
| "grad_norm": 0.959862470626831, | |
| "learning_rate": 2.0267763157894732e-06, | |
| "loss": 0.0272, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.0513, | |
| "grad_norm": 2.2999842166900635, | |
| "learning_rate": 2.0152631578947367e-06, | |
| "loss": 0.0246, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 3.0538, | |
| "grad_norm": 2.890066146850586, | |
| "learning_rate": 2.00375e-06, | |
| "loss": 0.0299, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 3.0563, | |
| "grad_norm": 1.7101376056671143, | |
| "learning_rate": 1.992236842105263e-06, | |
| "loss": 0.0322, | |
| "step": 5675 | |
| }, | |
| { | |
| "epoch": 3.0588, | |
| "grad_norm": 1.531943917274475, | |
| "learning_rate": 1.980723684210526e-06, | |
| "loss": 0.0345, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.0613, | |
| "grad_norm": 1.6334413290023804, | |
| "learning_rate": 1.969210526315789e-06, | |
| "loss": 0.032, | |
| "step": 5725 | |
| }, | |
| { | |
| "epoch": 3.0638, | |
| "grad_norm": 2.112278461456299, | |
| "learning_rate": 1.9576973684210526e-06, | |
| "loss": 0.0304, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 3.0663, | |
| "grad_norm": 1.7582517862319946, | |
| "learning_rate": 1.9461842105263156e-06, | |
| "loss": 0.0254, | |
| "step": 5775 | |
| }, | |
| { | |
| "epoch": 3.0688, | |
| "grad_norm": 1.3391777276992798, | |
| "learning_rate": 1.934671052631579e-06, | |
| "loss": 0.0316, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.0713, | |
| "grad_norm": 0.8350562453269958, | |
| "learning_rate": 1.923157894736842e-06, | |
| "loss": 0.0329, | |
| "step": 5825 | |
| }, | |
| { | |
| "epoch": 3.0738, | |
| "grad_norm": 0.7084619402885437, | |
| "learning_rate": 1.911644736842105e-06, | |
| "loss": 0.0325, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 3.0763, | |
| "grad_norm": 1.2961277961730957, | |
| "learning_rate": 1.9001315789473683e-06, | |
| "loss": 0.0313, | |
| "step": 5875 | |
| }, | |
| { | |
| "epoch": 3.0788, | |
| "grad_norm": 1.032840371131897, | |
| "learning_rate": 1.8886184210526315e-06, | |
| "loss": 0.0224, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.0813, | |
| "grad_norm": 1.2073044776916504, | |
| "learning_rate": 1.8771052631578945e-06, | |
| "loss": 0.0215, | |
| "step": 5925 | |
| }, | |
| { | |
| "epoch": 3.0838, | |
| "grad_norm": 0.8210967779159546, | |
| "learning_rate": 1.8655921052631577e-06, | |
| "loss": 0.0258, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 3.0863, | |
| "grad_norm": 1.5273653268814087, | |
| "learning_rate": 1.854078947368421e-06, | |
| "loss": 0.0254, | |
| "step": 5975 | |
| }, | |
| { | |
| "epoch": 3.0888, | |
| "grad_norm": 3.194197177886963, | |
| "learning_rate": 1.8425657894736842e-06, | |
| "loss": 0.025, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0888, | |
| "eval_loss": 0.14247554540634155, | |
| "eval_runtime": 4123.5746, | |
| "eval_samples_per_second": 3.305, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 7.940928579281029, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0913, | |
| "grad_norm": 2.1373400688171387, | |
| "learning_rate": 1.8310526315789472e-06, | |
| "loss": 0.031, | |
| "step": 6025 | |
| }, | |
| { | |
| "epoch": 3.0938, | |
| "grad_norm": 1.0779415369033813, | |
| "learning_rate": 1.8195394736842104e-06, | |
| "loss": 0.024, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 3.0963, | |
| "grad_norm": 0.9637121558189392, | |
| "learning_rate": 1.8080263157894734e-06, | |
| "loss": 0.0282, | |
| "step": 6075 | |
| }, | |
| { | |
| "epoch": 3.0987999999999998, | |
| "grad_norm": 1.1645703315734863, | |
| "learning_rate": 1.7965131578947366e-06, | |
| "loss": 0.0278, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.1013, | |
| "grad_norm": 1.2814173698425293, | |
| "learning_rate": 1.7849999999999996e-06, | |
| "loss": 0.0199, | |
| "step": 6125 | |
| }, | |
| { | |
| "epoch": 3.1038, | |
| "grad_norm": 1.458809494972229, | |
| "learning_rate": 1.773486842105263e-06, | |
| "loss": 0.0264, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 3.1063, | |
| "grad_norm": 1.6669671535491943, | |
| "learning_rate": 1.7619736842105263e-06, | |
| "loss": 0.0272, | |
| "step": 6175 | |
| }, | |
| { | |
| "epoch": 3.1088, | |
| "grad_norm": 1.5049173831939697, | |
| "learning_rate": 1.7504605263157893e-06, | |
| "loss": 0.0243, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.1113, | |
| "grad_norm": 0.861107587814331, | |
| "learning_rate": 1.7389473684210525e-06, | |
| "loss": 0.0274, | |
| "step": 6225 | |
| }, | |
| { | |
| "epoch": 3.1138, | |
| "grad_norm": 1.0454998016357422, | |
| "learning_rate": 1.7274342105263155e-06, | |
| "loss": 0.0258, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 3.1163, | |
| "grad_norm": 1.7108014822006226, | |
| "learning_rate": 1.7159210526315788e-06, | |
| "loss": 0.0259, | |
| "step": 6275 | |
| }, | |
| { | |
| "epoch": 3.1188, | |
| "grad_norm": 0.8804712295532227, | |
| "learning_rate": 1.704407894736842e-06, | |
| "loss": 0.0255, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.1213, | |
| "grad_norm": 2.0050883293151855, | |
| "learning_rate": 1.6928947368421052e-06, | |
| "loss": 0.0304, | |
| "step": 6325 | |
| }, | |
| { | |
| "epoch": 3.1238, | |
| "grad_norm": 1.4400875568389893, | |
| "learning_rate": 1.6813815789473682e-06, | |
| "loss": 0.0333, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 3.1263, | |
| "grad_norm": 1.4423948526382446, | |
| "learning_rate": 1.6698684210526315e-06, | |
| "loss": 0.0279, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 3.1288, | |
| "grad_norm": 1.3972327709197998, | |
| "learning_rate": 1.6583552631578947e-06, | |
| "loss": 0.0255, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.1313, | |
| "grad_norm": 1.6908966302871704, | |
| "learning_rate": 1.6468421052631577e-06, | |
| "loss": 0.0267, | |
| "step": 6425 | |
| }, | |
| { | |
| "epoch": 3.1338, | |
| "grad_norm": 0.9540082216262817, | |
| "learning_rate": 1.635328947368421e-06, | |
| "loss": 0.0265, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 3.1363, | |
| "grad_norm": 1.41488778591156, | |
| "learning_rate": 1.6238157894736841e-06, | |
| "loss": 0.0224, | |
| "step": 6475 | |
| }, | |
| { | |
| "epoch": 3.1388, | |
| "grad_norm": 0.4790860116481781, | |
| "learning_rate": 1.6123026315789474e-06, | |
| "loss": 0.0274, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.1388, | |
| "eval_loss": 0.13914132118225098, | |
| "eval_runtime": 4133.8202, | |
| "eval_samples_per_second": 3.297, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 7.731137088204039, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.1413, | |
| "grad_norm": 2.5638585090637207, | |
| "learning_rate": 1.6007894736842104e-06, | |
| "loss": 0.025, | |
| "step": 6525 | |
| }, | |
| { | |
| "epoch": 3.1438, | |
| "grad_norm": 1.8847306966781616, | |
| "learning_rate": 1.5892763157894736e-06, | |
| "loss": 0.0294, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 3.1463, | |
| "grad_norm": 1.0196236371994019, | |
| "learning_rate": 1.5777631578947366e-06, | |
| "loss": 0.0255, | |
| "step": 6575 | |
| }, | |
| { | |
| "epoch": 3.1488, | |
| "grad_norm": 1.0703202486038208, | |
| "learning_rate": 1.5662499999999998e-06, | |
| "loss": 0.0246, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.1513, | |
| "grad_norm": 2.646519422531128, | |
| "learning_rate": 1.5547368421052628e-06, | |
| "loss": 0.0213, | |
| "step": 6625 | |
| }, | |
| { | |
| "epoch": 3.1538, | |
| "grad_norm": 1.7430530786514282, | |
| "learning_rate": 1.5432236842105263e-06, | |
| "loss": 0.0267, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 3.1563, | |
| "grad_norm": 1.0606240034103394, | |
| "learning_rate": 1.5317105263157895e-06, | |
| "loss": 0.0269, | |
| "step": 6675 | |
| }, | |
| { | |
| "epoch": 3.1588, | |
| "grad_norm": 1.4670476913452148, | |
| "learning_rate": 1.5201973684210525e-06, | |
| "loss": 0.0271, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 3.1612999999999998, | |
| "grad_norm": 2.345014810562134, | |
| "learning_rate": 1.5086842105263157e-06, | |
| "loss": 0.0252, | |
| "step": 6725 | |
| }, | |
| { | |
| "epoch": 3.1638, | |
| "grad_norm": 2.9098987579345703, | |
| "learning_rate": 1.4971710526315787e-06, | |
| "loss": 0.0272, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 3.1663, | |
| "grad_norm": 0.5682694911956787, | |
| "learning_rate": 1.485657894736842e-06, | |
| "loss": 0.0237, | |
| "step": 6775 | |
| }, | |
| { | |
| "epoch": 3.1688, | |
| "grad_norm": 1.4645904302597046, | |
| "learning_rate": 1.4746052631578947e-06, | |
| "loss": 0.0303, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.0009, | |
| "grad_norm": 1.3764489889144897, | |
| "learning_rate": 1.4630921052631578e-06, | |
| "loss": 0.0242, | |
| "step": 6825 | |
| }, | |
| { | |
| "epoch": 4.0034, | |
| "grad_norm": 0.8848748803138733, | |
| "learning_rate": 1.451578947368421e-06, | |
| "loss": 0.0163, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 4.0059, | |
| "grad_norm": 0.619125485420227, | |
| "learning_rate": 1.440065789473684e-06, | |
| "loss": 0.0188, | |
| "step": 6875 | |
| }, | |
| { | |
| "epoch": 4.0084, | |
| "grad_norm": 0.9328649044036865, | |
| "learning_rate": 1.4285526315789472e-06, | |
| "loss": 0.0173, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 4.0109, | |
| "grad_norm": 1.77474045753479, | |
| "learning_rate": 1.4170394736842104e-06, | |
| "loss": 0.0146, | |
| "step": 6925 | |
| }, | |
| { | |
| "epoch": 4.0134, | |
| "grad_norm": 1.3934537172317505, | |
| "learning_rate": 1.4055263157894737e-06, | |
| "loss": 0.0156, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 4.0159, | |
| "grad_norm": 1.2856354713439941, | |
| "learning_rate": 1.3940131578947367e-06, | |
| "loss": 0.0173, | |
| "step": 6975 | |
| }, | |
| { | |
| "epoch": 4.0184, | |
| "grad_norm": 2.1229758262634277, | |
| "learning_rate": 1.3824999999999999e-06, | |
| "loss": 0.0155, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0184, | |
| "eval_loss": 0.14916160702705383, | |
| "eval_runtime": 4128.7355, | |
| "eval_samples_per_second": 3.301, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 7.697240646414307, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0209, | |
| "grad_norm": 0.44512999057769775, | |
| "learning_rate": 1.3709868421052631e-06, | |
| "loss": 0.0153, | |
| "step": 7025 | |
| }, | |
| { | |
| "epoch": 4.0234, | |
| "grad_norm": 1.8791674375534058, | |
| "learning_rate": 1.3594736842105261e-06, | |
| "loss": 0.0165, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 4.0259, | |
| "grad_norm": 5.244405746459961, | |
| "learning_rate": 1.3479605263157894e-06, | |
| "loss": 0.0179, | |
| "step": 7075 | |
| }, | |
| { | |
| "epoch": 4.0284, | |
| "grad_norm": 1.1926153898239136, | |
| "learning_rate": 1.3364473684210526e-06, | |
| "loss": 0.0161, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 4.0309, | |
| "grad_norm": 1.1147819757461548, | |
| "learning_rate": 1.3249342105263158e-06, | |
| "loss": 0.015, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 4.0334, | |
| "grad_norm": 1.9370721578598022, | |
| "learning_rate": 1.3134210526315788e-06, | |
| "loss": 0.0142, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 4.0359, | |
| "grad_norm": 0.49344903230667114, | |
| "learning_rate": 1.301907894736842e-06, | |
| "loss": 0.0134, | |
| "step": 7175 | |
| }, | |
| { | |
| "epoch": 4.0384, | |
| "grad_norm": 1.8190902471542358, | |
| "learning_rate": 1.290394736842105e-06, | |
| "loss": 0.0168, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.0409, | |
| "grad_norm": 0.7560425400733948, | |
| "learning_rate": 1.2788815789473683e-06, | |
| "loss": 0.0143, | |
| "step": 7225 | |
| }, | |
| { | |
| "epoch": 4.0434, | |
| "grad_norm": 1.0451087951660156, | |
| "learning_rate": 1.2673684210526313e-06, | |
| "loss": 0.0149, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 4.0459, | |
| "grad_norm": 1.0334726572036743, | |
| "learning_rate": 1.2558552631578947e-06, | |
| "loss": 0.0136, | |
| "step": 7275 | |
| }, | |
| { | |
| "epoch": 4.0484, | |
| "grad_norm": 0.6531663537025452, | |
| "learning_rate": 1.244342105263158e-06, | |
| "loss": 0.0137, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 4.0509, | |
| "grad_norm": 0.8954887986183167, | |
| "learning_rate": 1.232828947368421e-06, | |
| "loss": 0.0118, | |
| "step": 7325 | |
| }, | |
| { | |
| "epoch": 4.0534, | |
| "grad_norm": 1.0640511512756348, | |
| "learning_rate": 1.2213157894736842e-06, | |
| "loss": 0.0126, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 4.0559, | |
| "grad_norm": 0.2824617922306061, | |
| "learning_rate": 1.2098026315789472e-06, | |
| "loss": 0.0139, | |
| "step": 7375 | |
| }, | |
| { | |
| "epoch": 4.0584, | |
| "grad_norm": 1.0095443725585938, | |
| "learning_rate": 1.1982894736842104e-06, | |
| "loss": 0.018, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.0609, | |
| "grad_norm": 1.1475225687026978, | |
| "learning_rate": 1.1867763157894734e-06, | |
| "loss": 0.0133, | |
| "step": 7425 | |
| }, | |
| { | |
| "epoch": 4.0634, | |
| "grad_norm": 1.5951991081237793, | |
| "learning_rate": 1.1752631578947369e-06, | |
| "loss": 0.013, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 4.0659, | |
| "grad_norm": 0.3482917249202728, | |
| "learning_rate": 1.1637499999999999e-06, | |
| "loss": 0.0154, | |
| "step": 7475 | |
| }, | |
| { | |
| "epoch": 4.0684, | |
| "grad_norm": 1.1572391986846924, | |
| "learning_rate": 1.152236842105263e-06, | |
| "loss": 0.0189, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0684, | |
| "eval_loss": 0.15172211825847626, | |
| "eval_runtime": 4117.5679, | |
| "eval_samples_per_second": 3.31, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 7.656931364285977, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0709, | |
| "grad_norm": 1.3942557573318481, | |
| "learning_rate": 1.140723684210526e-06, | |
| "loss": 0.0143, | |
| "step": 7525 | |
| }, | |
| { | |
| "epoch": 4.0734, | |
| "grad_norm": 0.8097572326660156, | |
| "learning_rate": 1.1292105263157893e-06, | |
| "loss": 0.0127, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 4.0759, | |
| "grad_norm": 0.740375816822052, | |
| "learning_rate": 1.1176973684210526e-06, | |
| "loss": 0.0124, | |
| "step": 7575 | |
| }, | |
| { | |
| "epoch": 4.0784, | |
| "grad_norm": 0.8702480792999268, | |
| "learning_rate": 1.1061842105263156e-06, | |
| "loss": 0.0137, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.0809, | |
| "grad_norm": 1.223105788230896, | |
| "learning_rate": 1.094671052631579e-06, | |
| "loss": 0.0137, | |
| "step": 7625 | |
| }, | |
| { | |
| "epoch": 4.0834, | |
| "grad_norm": 0.43614983558654785, | |
| "learning_rate": 1.083157894736842e-06, | |
| "loss": 0.0109, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 4.0859, | |
| "grad_norm": 1.0974986553192139, | |
| "learning_rate": 1.0716447368421052e-06, | |
| "loss": 0.0118, | |
| "step": 7675 | |
| }, | |
| { | |
| "epoch": 4.0884, | |
| "grad_norm": 0.7234652042388916, | |
| "learning_rate": 1.0601315789473682e-06, | |
| "loss": 0.0125, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 4.0909, | |
| "grad_norm": 0.7752431035041809, | |
| "learning_rate": 1.0486184210526315e-06, | |
| "loss": 0.0135, | |
| "step": 7725 | |
| }, | |
| { | |
| "epoch": 4.0934, | |
| "grad_norm": 0.8796952366828918, | |
| "learning_rate": 1.0371052631578947e-06, | |
| "loss": 0.0158, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 4.0959, | |
| "grad_norm": 3.9135661125183105, | |
| "learning_rate": 1.0255921052631577e-06, | |
| "loss": 0.0139, | |
| "step": 7775 | |
| }, | |
| { | |
| "epoch": 4.0984, | |
| "grad_norm": 0.4837290942668915, | |
| "learning_rate": 1.014078947368421e-06, | |
| "loss": 0.0103, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 4.1009, | |
| "grad_norm": 1.1155998706817627, | |
| "learning_rate": 1.0025657894736842e-06, | |
| "loss": 0.0106, | |
| "step": 7825 | |
| }, | |
| { | |
| "epoch": 4.1034, | |
| "grad_norm": 2.628676652908325, | |
| "learning_rate": 9.910526315789474e-07, | |
| "loss": 0.0089, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 4.1059, | |
| "grad_norm": 1.716665506362915, | |
| "learning_rate": 9.795394736842104e-07, | |
| "loss": 0.0132, | |
| "step": 7875 | |
| }, | |
| { | |
| "epoch": 4.1084, | |
| "grad_norm": 1.6751716136932373, | |
| "learning_rate": 9.680263157894736e-07, | |
| "loss": 0.0137, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 4.1109, | |
| "grad_norm": 0.9773244261741638, | |
| "learning_rate": 9.565131578947368e-07, | |
| "loss": 0.0111, | |
| "step": 7925 | |
| }, | |
| { | |
| "epoch": 4.1134, | |
| "grad_norm": 1.44219172000885, | |
| "learning_rate": 9.45e-07, | |
| "loss": 0.0139, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 4.1159, | |
| "grad_norm": 0.8723123073577881, | |
| "learning_rate": 9.334868421052631e-07, | |
| "loss": 0.0117, | |
| "step": 7975 | |
| }, | |
| { | |
| "epoch": 4.1184, | |
| "grad_norm": 0.6484673023223877, | |
| "learning_rate": 9.219736842105263e-07, | |
| "loss": 0.0139, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.1184, | |
| "eval_loss": 0.15393850207328796, | |
| "eval_runtime": 4128.9341, | |
| "eval_samples_per_second": 3.301, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 7.626699402689728, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.1209, | |
| "grad_norm": 1.3702197074890137, | |
| "learning_rate": 9.104605263157894e-07, | |
| "loss": 0.0158, | |
| "step": 8025 | |
| }, | |
| { | |
| "epoch": 4.1234, | |
| "grad_norm": 1.425645351409912, | |
| "learning_rate": 8.989473684210525e-07, | |
| "loss": 0.0117, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 4.1259, | |
| "grad_norm": 1.4255399703979492, | |
| "learning_rate": 8.874342105263158e-07, | |
| "loss": 0.015, | |
| "step": 8075 | |
| }, | |
| { | |
| "epoch": 4.1284, | |
| "grad_norm": 0.6988621950149536, | |
| "learning_rate": 8.759210526315789e-07, | |
| "loss": 0.0141, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 4.1309, | |
| "grad_norm": 1.1563546657562256, | |
| "learning_rate": 8.64407894736842e-07, | |
| "loss": 0.0122, | |
| "step": 8125 | |
| }, | |
| { | |
| "epoch": 4.1334, | |
| "grad_norm": 1.2023714780807495, | |
| "learning_rate": 8.528947368421051e-07, | |
| "loss": 0.013, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 4.1359, | |
| "grad_norm": 0.9450110197067261, | |
| "learning_rate": 8.413815789473683e-07, | |
| "loss": 0.0123, | |
| "step": 8175 | |
| }, | |
| { | |
| "epoch": 4.1384, | |
| "grad_norm": 0.9265995621681213, | |
| "learning_rate": 8.298684210526316e-07, | |
| "loss": 0.0114, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 4.1409, | |
| "grad_norm": 0.4234980046749115, | |
| "learning_rate": 8.183552631578947e-07, | |
| "loss": 0.0085, | |
| "step": 8225 | |
| }, | |
| { | |
| "epoch": 4.1434, | |
| "grad_norm": 1.3323073387145996, | |
| "learning_rate": 8.068421052631579e-07, | |
| "loss": 0.014, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 4.1459, | |
| "grad_norm": 1.2050007581710815, | |
| "learning_rate": 7.95328947368421e-07, | |
| "loss": 0.0106, | |
| "step": 8275 | |
| }, | |
| { | |
| "epoch": 4.1484, | |
| "grad_norm": 1.261042594909668, | |
| "learning_rate": 7.838157894736841e-07, | |
| "loss": 0.0107, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 4.1509, | |
| "grad_norm": 1.2892303466796875, | |
| "learning_rate": 7.723026315789474e-07, | |
| "loss": 0.0145, | |
| "step": 8325 | |
| }, | |
| { | |
| "epoch": 4.1534, | |
| "grad_norm": 1.1626112461090088, | |
| "learning_rate": 7.607894736842105e-07, | |
| "loss": 0.0139, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 4.1559, | |
| "grad_norm": 1.0547322034835815, | |
| "learning_rate": 7.492763157894736e-07, | |
| "loss": 0.0154, | |
| "step": 8375 | |
| }, | |
| { | |
| "epoch": 4.1584, | |
| "grad_norm": 0.44805532693862915, | |
| "learning_rate": 7.377631578947367e-07, | |
| "loss": 0.0109, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.1609, | |
| "grad_norm": 0.7095866203308105, | |
| "learning_rate": 7.262499999999999e-07, | |
| "loss": 0.0114, | |
| "step": 8425 | |
| }, | |
| { | |
| "epoch": 4.1634, | |
| "grad_norm": 1.4220194816589355, | |
| "learning_rate": 7.14736842105263e-07, | |
| "loss": 0.0134, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 4.1659, | |
| "grad_norm": 1.0814168453216553, | |
| "learning_rate": 7.032236842105263e-07, | |
| "loss": 0.0142, | |
| "step": 8475 | |
| }, | |
| { | |
| "epoch": 4.1684, | |
| "grad_norm": 0.7026916146278381, | |
| "learning_rate": 6.917105263157895e-07, | |
| "loss": 0.0141, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.1684, | |
| "eval_loss": 0.15496784448623657, | |
| "eval_runtime": 4124.1829, | |
| "eval_samples_per_second": 3.305, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 7.542416358239584, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.0005, | |
| "grad_norm": 4.648550033569336, | |
| "learning_rate": 6.801973684210526e-07, | |
| "loss": 0.0285, | |
| "step": 8525 | |
| }, | |
| { | |
| "epoch": 5.003, | |
| "grad_norm": 1.9204503297805786, | |
| "learning_rate": 6.691447368421053e-07, | |
| "loss": 0.0761, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 5.0055, | |
| "grad_norm": 1.7285746335983276, | |
| "learning_rate": 6.576315789473684e-07, | |
| "loss": 0.0602, | |
| "step": 8575 | |
| }, | |
| { | |
| "epoch": 5.008, | |
| "grad_norm": 1.1516830921173096, | |
| "learning_rate": 6.461184210526315e-07, | |
| "loss": 0.0585, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.0105, | |
| "grad_norm": 3.3867828845977783, | |
| "learning_rate": 6.346052631578947e-07, | |
| "loss": 0.0656, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 5.013, | |
| "grad_norm": 4.064920902252197, | |
| "learning_rate": 6.230921052631579e-07, | |
| "loss": 0.0683, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 5.0155, | |
| "grad_norm": 3.695047378540039, | |
| "learning_rate": 6.11578947368421e-07, | |
| "loss": 0.0659, | |
| "step": 8675 | |
| }, | |
| { | |
| "epoch": 5.018, | |
| "grad_norm": 2.9087939262390137, | |
| "learning_rate": 6.000657894736842e-07, | |
| "loss": 0.0611, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 5.0205, | |
| "grad_norm": 3.368290424346924, | |
| "learning_rate": 5.885526315789473e-07, | |
| "loss": 0.0603, | |
| "step": 8725 | |
| }, | |
| { | |
| "epoch": 5.023, | |
| "grad_norm": 3.7565319538116455, | |
| "learning_rate": 5.770394736842104e-07, | |
| "loss": 0.0614, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 5.0255, | |
| "grad_norm": 2.4887771606445312, | |
| "learning_rate": 5.655263157894735e-07, | |
| "loss": 0.0497, | |
| "step": 8775 | |
| }, | |
| { | |
| "epoch": 5.028, | |
| "grad_norm": 2.1670076847076416, | |
| "learning_rate": 5.540131578947369e-07, | |
| "loss": 0.0662, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.0305, | |
| "grad_norm": 1.3746148347854614, | |
| "learning_rate": 5.425e-07, | |
| "loss": 0.0507, | |
| "step": 8825 | |
| }, | |
| { | |
| "epoch": 5.033, | |
| "grad_norm": 1.8274154663085938, | |
| "learning_rate": 5.309868421052631e-07, | |
| "loss": 0.0449, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 5.0355, | |
| "grad_norm": 2.9424078464508057, | |
| "learning_rate": 5.194736842105262e-07, | |
| "loss": 0.0529, | |
| "step": 8875 | |
| }, | |
| { | |
| "epoch": 5.038, | |
| "grad_norm": 2.457754611968994, | |
| "learning_rate": 5.079605263157895e-07, | |
| "loss": 0.042, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 5.0405, | |
| "grad_norm": 2.208768606185913, | |
| "learning_rate": 4.964473684210526e-07, | |
| "loss": 0.0407, | |
| "step": 8925 | |
| }, | |
| { | |
| "epoch": 5.043, | |
| "grad_norm": 1.9554438591003418, | |
| "learning_rate": 4.849342105263158e-07, | |
| "loss": 0.0465, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 5.0455, | |
| "grad_norm": 1.1464567184448242, | |
| "learning_rate": 4.734210526315789e-07, | |
| "loss": 0.0537, | |
| "step": 8975 | |
| }, | |
| { | |
| "epoch": 5.048, | |
| "grad_norm": 3.1216509342193604, | |
| "learning_rate": 4.6190789473684203e-07, | |
| "loss": 0.0368, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.048, | |
| "eval_loss": 0.12588092684745789, | |
| "eval_runtime": 4149.257, | |
| "eval_samples_per_second": 3.285, | |
| "eval_steps_per_second": 0.411, | |
| "eval_wer": 7.215361500971087, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.048, | |
| "step": 9000, | |
| "total_flos": 4.891718061785088e+20, | |
| "train_loss": 0.0, | |
| "train_runtime": 289.8068, | |
| "train_samples_per_second": 552.092, | |
| "train_steps_per_second": 34.506 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.891718061785088e+20, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |