| { | |
| "best_metric": 13.88186822698473, | |
| "best_model_checkpoint": "/speechbrain/data/whis/whisper-medium-ar-aug24-cont2/checkpoint-3000", | |
| "epoch": 0.7232084155161078, | |
| "eval_steps": 300, | |
| "global_step": 3300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005478851632697786, | |
| "grad_norm": Infinity, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 0.9139, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.010957703265395573, | |
| "grad_norm": 3.586649179458618, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 0.8206, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01643655489809336, | |
| "grad_norm": 2.649549961090088, | |
| "learning_rate": 1.46e-06, | |
| "loss": 0.6306, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.021915406530791146, | |
| "grad_norm": 1.3442453145980835, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.4395, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.027394258163488932, | |
| "grad_norm": 1.2841758728027344, | |
| "learning_rate": 2.46e-06, | |
| "loss": 0.3333, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03287310979618672, | |
| "grad_norm": 1.4466370344161987, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.2598, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03835196142888451, | |
| "grad_norm": 1.1411246061325073, | |
| "learning_rate": 3.46e-06, | |
| "loss": 0.2296, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04383081306158229, | |
| "grad_norm": 0.7538527846336365, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.2049, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04930966469428008, | |
| "grad_norm": 0.5670434236526489, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 0.1957, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.054788516326977864, | |
| "grad_norm": 0.5980058312416077, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.189, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.060267367959675654, | |
| "grad_norm": 0.5877336263656616, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.177, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06574621959237344, | |
| "grad_norm": 0.5727600455284119, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.1771, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.06574621959237344, | |
| "eval_loss": 0.12760603427886963, | |
| "eval_runtime": 1379.0902, | |
| "eval_samples_per_second": 2.415, | |
| "eval_steps_per_second": 0.051, | |
| "eval_wer": 15.454304146614502, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07122507122507123, | |
| "grad_norm": 0.6107267141342163, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.1789, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07670392285776902, | |
| "grad_norm": 0.6271827220916748, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.1771, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.08218277449046679, | |
| "grad_norm": 0.5440804362297058, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.1705, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08766162612316458, | |
| "grad_norm": 0.6523880958557129, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.1673, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09314047775586237, | |
| "grad_norm": 0.6077700257301331, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.166, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.09861932938856016, | |
| "grad_norm": 0.5785207748413086, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.167, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.10409818102125794, | |
| "grad_norm": 0.6054052710533142, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.1653, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.10957703265395573, | |
| "grad_norm": 0.5897749662399292, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.1604, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11505588428665352, | |
| "grad_norm": 0.6138154864311218, | |
| "learning_rate": 9.997465843984135e-06, | |
| "loss": 0.1633, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.12053473591935131, | |
| "grad_norm": 0.5625219941139221, | |
| "learning_rate": 9.994711326575585e-06, | |
| "loss": 0.1615, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12601358755204908, | |
| "grad_norm": 0.5982335805892944, | |
| "learning_rate": 9.991956809167035e-06, | |
| "loss": 0.1528, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.13149243918474687, | |
| "grad_norm": 0.5922495722770691, | |
| "learning_rate": 9.989202291758486e-06, | |
| "loss": 0.1612, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13149243918474687, | |
| "eval_loss": 0.11784320324659348, | |
| "eval_runtime": 1336.8758, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.94362312317511, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13697129081744466, | |
| "grad_norm": 0.5509978532791138, | |
| "learning_rate": 9.986447774349934e-06, | |
| "loss": 0.1664, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.14245014245014245, | |
| "grad_norm": 0.636675238609314, | |
| "learning_rate": 9.983693256941385e-06, | |
| "loss": 0.1572, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14792899408284024, | |
| "grad_norm": 0.5626579523086548, | |
| "learning_rate": 9.980938739532835e-06, | |
| "loss": 0.1563, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.15340784571553803, | |
| "grad_norm": 0.587232232093811, | |
| "learning_rate": 9.978184222124284e-06, | |
| "loss": 0.1603, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.15888669734823582, | |
| "grad_norm": 0.5765535235404968, | |
| "learning_rate": 9.975429704715734e-06, | |
| "loss": 0.1552, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.16436554898093358, | |
| "grad_norm": 0.5985578298568726, | |
| "learning_rate": 9.972675187307184e-06, | |
| "loss": 0.1567, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.16984440061363137, | |
| "grad_norm": 0.5767161250114441, | |
| "learning_rate": 9.969920669898635e-06, | |
| "loss": 0.1559, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.17532325224632916, | |
| "grad_norm": 0.549917459487915, | |
| "learning_rate": 9.967166152490085e-06, | |
| "loss": 0.1553, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.18080210387902695, | |
| "grad_norm": 0.5054299831390381, | |
| "learning_rate": 9.964411635081535e-06, | |
| "loss": 0.1569, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.18628095551172474, | |
| "grad_norm": 0.6237090826034546, | |
| "learning_rate": 9.961657117672986e-06, | |
| "loss": 0.1516, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.19175980714442253, | |
| "grad_norm": 0.6381577253341675, | |
| "learning_rate": 9.958902600264434e-06, | |
| "loss": 0.1548, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.19723865877712032, | |
| "grad_norm": 0.5359848737716675, | |
| "learning_rate": 9.956148082855884e-06, | |
| "loss": 0.1517, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19723865877712032, | |
| "eval_loss": 0.11450996994972229, | |
| "eval_runtime": 1336.2266, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.663758381516232, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.2027175104098181, | |
| "grad_norm": 0.5749765038490295, | |
| "learning_rate": 9.953393565447335e-06, | |
| "loss": 0.1526, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.20819636204251588, | |
| "grad_norm": 0.5857889652252197, | |
| "learning_rate": 9.950639048038785e-06, | |
| "loss": 0.1505, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.21367521367521367, | |
| "grad_norm": 0.5977457165718079, | |
| "learning_rate": 9.947884530630234e-06, | |
| "loss": 0.1526, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.21915406530791146, | |
| "grad_norm": 0.5572954416275024, | |
| "learning_rate": 9.945130013221684e-06, | |
| "loss": 0.1413, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22463291694060925, | |
| "grad_norm": 0.6073783040046692, | |
| "learning_rate": 9.942375495813134e-06, | |
| "loss": 0.1512, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.23011176857330704, | |
| "grad_norm": 0.5608137845993042, | |
| "learning_rate": 9.939620978404585e-06, | |
| "loss": 0.1523, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.23559062020600482, | |
| "grad_norm": 0.6059215068817139, | |
| "learning_rate": 9.936866460996035e-06, | |
| "loss": 0.1461, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.24106947183870261, | |
| "grad_norm": 0.6084004640579224, | |
| "learning_rate": 9.934111943587485e-06, | |
| "loss": 0.1499, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.2465483234714004, | |
| "grad_norm": 0.5527833104133606, | |
| "learning_rate": 9.931357426178936e-06, | |
| "loss": 0.1475, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.25202717510409817, | |
| "grad_norm": 0.5086527466773987, | |
| "learning_rate": 9.928602908770384e-06, | |
| "loss": 0.1487, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.25750602673679596, | |
| "grad_norm": 0.6129872798919678, | |
| "learning_rate": 9.925848391361834e-06, | |
| "loss": 0.1496, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.26298487836949375, | |
| "grad_norm": 0.6175655126571655, | |
| "learning_rate": 9.923093873953285e-06, | |
| "loss": 0.149, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.26298487836949375, | |
| "eval_loss": 0.11248071491718292, | |
| "eval_runtime": 1336.7392, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.334268139274545, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.26846373000219154, | |
| "grad_norm": 0.5802096128463745, | |
| "learning_rate": 9.920339356544733e-06, | |
| "loss": 0.145, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.2739425816348893, | |
| "grad_norm": 0.5316102504730225, | |
| "learning_rate": 9.917584839136184e-06, | |
| "loss": 0.1474, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.2794214332675871, | |
| "grad_norm": 0.5899500846862793, | |
| "learning_rate": 9.914830321727634e-06, | |
| "loss": 0.1431, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.2849002849002849, | |
| "grad_norm": 0.5635184049606323, | |
| "learning_rate": 9.912075804319084e-06, | |
| "loss": 0.1418, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.2903791365329827, | |
| "grad_norm": 0.5536847114562988, | |
| "learning_rate": 9.909321286910533e-06, | |
| "loss": 0.1461, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.2958579881656805, | |
| "grad_norm": 0.5551804900169373, | |
| "learning_rate": 9.906566769501983e-06, | |
| "loss": 0.1437, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.3013368397983783, | |
| "grad_norm": 0.5187482833862305, | |
| "learning_rate": 9.903812252093434e-06, | |
| "loss": 0.1432, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.30681569143107607, | |
| "grad_norm": 0.6163837313652039, | |
| "learning_rate": 9.901057734684884e-06, | |
| "loss": 0.1497, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.31229454306377386, | |
| "grad_norm": 0.5211063027381897, | |
| "learning_rate": 9.898303217276334e-06, | |
| "loss": 0.1368, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.31777339469647164, | |
| "grad_norm": 0.5287565588951111, | |
| "learning_rate": 9.895548699867785e-06, | |
| "loss": 0.1393, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.3232522463291694, | |
| "grad_norm": 0.5906191468238831, | |
| "learning_rate": 9.892794182459235e-06, | |
| "loss": 0.1511, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.32873109796186717, | |
| "grad_norm": 0.5539998412132263, | |
| "learning_rate": 9.890039665050683e-06, | |
| "loss": 0.1405, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.32873109796186717, | |
| "eval_loss": 0.11105980724096298, | |
| "eval_runtime": 1335.343, | |
| "eval_samples_per_second": 2.494, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.397165576059734, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.33420994959456496, | |
| "grad_norm": 0.5551910400390625, | |
| "learning_rate": 9.887285147642134e-06, | |
| "loss": 0.1449, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.33968880122726275, | |
| "grad_norm": 0.5776943564414978, | |
| "learning_rate": 9.884530630233584e-06, | |
| "loss": 0.1438, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.34516765285996054, | |
| "grad_norm": 0.5213550925254822, | |
| "learning_rate": 9.881776112825033e-06, | |
| "loss": 0.1423, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.35064650449265833, | |
| "grad_norm": 0.5484446883201599, | |
| "learning_rate": 9.879021595416483e-06, | |
| "loss": 0.147, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.3561253561253561, | |
| "grad_norm": 0.5987962484359741, | |
| "learning_rate": 9.876267078007933e-06, | |
| "loss": 0.1374, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.3616042077580539, | |
| "grad_norm": 0.5952353477478027, | |
| "learning_rate": 9.873512560599384e-06, | |
| "loss": 0.1439, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.3670830593907517, | |
| "grad_norm": 0.6191999316215515, | |
| "learning_rate": 9.870758043190834e-06, | |
| "loss": 0.1396, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.3725619110234495, | |
| "grad_norm": 0.5286071300506592, | |
| "learning_rate": 9.868003525782284e-06, | |
| "loss": 0.1446, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.3780407626561473, | |
| "grad_norm": 0.5466740131378174, | |
| "learning_rate": 9.865249008373735e-06, | |
| "loss": 0.1447, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.38351961428884507, | |
| "grad_norm": 0.5799579620361328, | |
| "learning_rate": 9.862494490965183e-06, | |
| "loss": 0.1426, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.38899846592154286, | |
| "grad_norm": 0.6488582491874695, | |
| "learning_rate": 9.859739973556633e-06, | |
| "loss": 0.1382, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.39447731755424065, | |
| "grad_norm": 0.5399666428565979, | |
| "learning_rate": 9.856985456148084e-06, | |
| "loss": 0.1422, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.39447731755424065, | |
| "eval_loss": 0.11009500920772552, | |
| "eval_runtime": 1344.7575, | |
| "eval_samples_per_second": 2.476, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.303684981938625, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.39995616918693844, | |
| "grad_norm": 0.508437991142273, | |
| "learning_rate": 9.854230938739534e-06, | |
| "loss": 0.1382, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.4054350208196362, | |
| "grad_norm": 0.5800564289093018, | |
| "learning_rate": 9.851476421330983e-06, | |
| "loss": 0.1446, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.410913872452334, | |
| "grad_norm": 0.5412048101425171, | |
| "learning_rate": 9.848721903922433e-06, | |
| "loss": 0.1404, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.41639272408503175, | |
| "grad_norm": 0.5806304216384888, | |
| "learning_rate": 9.845967386513883e-06, | |
| "loss": 0.1383, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.42187157571772954, | |
| "grad_norm": 0.5252566933631897, | |
| "learning_rate": 9.843212869105334e-06, | |
| "loss": 0.143, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.42735042735042733, | |
| "grad_norm": 0.5622444748878479, | |
| "learning_rate": 9.840458351696784e-06, | |
| "loss": 0.1416, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.4328292789831251, | |
| "grad_norm": 0.5336723923683167, | |
| "learning_rate": 9.837703834288234e-06, | |
| "loss": 0.1373, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.4383081306158229, | |
| "grad_norm": 0.5631132125854492, | |
| "learning_rate": 9.834949316879685e-06, | |
| "loss": 0.1442, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4437869822485207, | |
| "grad_norm": 0.5697339177131653, | |
| "learning_rate": 9.832194799471133e-06, | |
| "loss": 0.1392, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.4492658338812185, | |
| "grad_norm": 0.5948196649551392, | |
| "learning_rate": 9.829440282062584e-06, | |
| "loss": 0.1418, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.4547446855139163, | |
| "grad_norm": 0.5385356545448303, | |
| "learning_rate": 9.826685764654034e-06, | |
| "loss": 0.1378, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 0.46022353714661407, | |
| "grad_norm": 0.5994429588317871, | |
| "learning_rate": 9.823931247245482e-06, | |
| "loss": 0.1407, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.46022353714661407, | |
| "eval_loss": 0.10945474356412888, | |
| "eval_runtime": 1339.6539, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 13.913605465729553, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.46570238877931186, | |
| "grad_norm": 0.5439214110374451, | |
| "learning_rate": 9.821176729836933e-06, | |
| "loss": 0.1368, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.47118124041200965, | |
| "grad_norm": 0.5190943479537964, | |
| "learning_rate": 9.818422212428383e-06, | |
| "loss": 0.1382, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.47666009204470744, | |
| "grad_norm": 0.49516212940216064, | |
| "learning_rate": 9.815667695019833e-06, | |
| "loss": 0.1361, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 0.48213894367740523, | |
| "grad_norm": 0.5503794550895691, | |
| "learning_rate": 9.812913177611284e-06, | |
| "loss": 0.1359, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.487617795310103, | |
| "grad_norm": 0.555046796798706, | |
| "learning_rate": 9.810158660202734e-06, | |
| "loss": 0.1358, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "grad_norm": 0.5206555724143982, | |
| "learning_rate": 9.807404142794184e-06, | |
| "loss": 0.1357, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.4985754985754986, | |
| "grad_norm": 0.6200060844421387, | |
| "learning_rate": 9.804649625385633e-06, | |
| "loss": 0.1353, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 0.5040543502081963, | |
| "grad_norm": 0.5105571150779724, | |
| "learning_rate": 9.801895107977083e-06, | |
| "loss": 0.1309, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.5095332018408941, | |
| "grad_norm": 0.5761134028434753, | |
| "learning_rate": 9.799140590568534e-06, | |
| "loss": 0.1378, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 0.5150120534735919, | |
| "grad_norm": 0.5118030309677124, | |
| "learning_rate": 9.796386073159984e-06, | |
| "loss": 0.1352, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.5204909051062897, | |
| "grad_norm": 0.4720148742198944, | |
| "learning_rate": 9.793631555751433e-06, | |
| "loss": 0.1342, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.5259697567389875, | |
| "grad_norm": 0.5468887686729431, | |
| "learning_rate": 9.790877038342883e-06, | |
| "loss": 0.129, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5259697567389875, | |
| "eval_loss": 0.10870281606912613, | |
| "eval_runtime": 1334.3649, | |
| "eval_samples_per_second": 2.496, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.091911043405, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5314486083716853, | |
| "grad_norm": 0.5378099083900452, | |
| "learning_rate": 9.788122520934333e-06, | |
| "loss": 0.1331, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 0.5369274600043831, | |
| "grad_norm": 0.5661837458610535, | |
| "learning_rate": 9.785368003525782e-06, | |
| "loss": 0.1373, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.5424063116370809, | |
| "grad_norm": 0.5715429782867432, | |
| "learning_rate": 9.782613486117232e-06, | |
| "loss": 0.1394, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 0.5478851632697787, | |
| "grad_norm": 0.5901737213134766, | |
| "learning_rate": 9.779858968708682e-06, | |
| "loss": 0.135, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5533640149024764, | |
| "grad_norm": 0.5925373435020447, | |
| "learning_rate": 9.777104451300133e-06, | |
| "loss": 0.1354, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 0.5588428665351742, | |
| "grad_norm": 0.5552576780319214, | |
| "learning_rate": 9.774349933891583e-06, | |
| "loss": 0.1332, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.564321718167872, | |
| "grad_norm": 0.5988050103187561, | |
| "learning_rate": 9.771595416483033e-06, | |
| "loss": 0.1345, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 0.5698005698005698, | |
| "grad_norm": 0.535481333732605, | |
| "learning_rate": 9.768840899074484e-06, | |
| "loss": 0.1361, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5752794214332676, | |
| "grad_norm": 0.48405903577804565, | |
| "learning_rate": 9.766086381665934e-06, | |
| "loss": 0.1334, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.5807582730659654, | |
| "grad_norm": 0.5512611269950867, | |
| "learning_rate": 9.763331864257383e-06, | |
| "loss": 0.1317, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.5862371246986632, | |
| "grad_norm": 0.509512722492218, | |
| "learning_rate": 9.760577346848833e-06, | |
| "loss": 0.1332, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "grad_norm": 0.5346803665161133, | |
| "learning_rate": 9.757822829440283e-06, | |
| "loss": 0.1289, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "eval_loss": 0.1080569475889206, | |
| "eval_runtime": 1334.6136, | |
| "eval_samples_per_second": 2.495, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 13.955152396450046, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5971948279640588, | |
| "grad_norm": 0.5665607452392578, | |
| "learning_rate": 9.755068312031732e-06, | |
| "loss": 0.1278, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 0.6026736795967566, | |
| "grad_norm": 0.5593467950820923, | |
| "learning_rate": 9.752313794623182e-06, | |
| "loss": 0.1311, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.6081525312294543, | |
| "grad_norm": 0.5182927250862122, | |
| "learning_rate": 9.749559277214632e-06, | |
| "loss": 0.1304, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 0.6136313828621521, | |
| "grad_norm": 0.5251736044883728, | |
| "learning_rate": 9.746804759806083e-06, | |
| "loss": 0.1264, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.6191102344948499, | |
| "grad_norm": 0.5915343761444092, | |
| "learning_rate": 9.744050242397533e-06, | |
| "loss": 0.1303, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 0.6245890861275477, | |
| "grad_norm": 0.5850628018379211, | |
| "learning_rate": 9.741295724988983e-06, | |
| "loss": 0.1324, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.6300679377602455, | |
| "grad_norm": 0.5724135637283325, | |
| "learning_rate": 9.738541207580434e-06, | |
| "loss": 0.1291, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.6355467893929433, | |
| "grad_norm": 0.4958917200565338, | |
| "learning_rate": 9.735786690171882e-06, | |
| "loss": 0.1297, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 0.582492470741272, | |
| "learning_rate": 9.733032172763333e-06, | |
| "loss": 0.1312, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 0.6465044926583388, | |
| "grad_norm": 0.5806357264518738, | |
| "learning_rate": 9.730277655354783e-06, | |
| "loss": 0.1309, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.6519833442910365, | |
| "grad_norm": 0.5375156998634338, | |
| "learning_rate": 9.727523137946233e-06, | |
| "loss": 0.1294, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 0.6574621959237343, | |
| "grad_norm": 0.5173301696777344, | |
| "learning_rate": 9.724768620537682e-06, | |
| "loss": 0.1299, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6574621959237343, | |
| "eval_loss": 0.10775619745254517, | |
| "eval_runtime": 1347.5045, | |
| "eval_samples_per_second": 2.471, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 13.88186822698473, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6629410475564321, | |
| "grad_norm": 0.5214000940322876, | |
| "learning_rate": 9.722014103129132e-06, | |
| "loss": 0.1323, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 0.6684198991891299, | |
| "grad_norm": 0.5085532069206238, | |
| "learning_rate": 9.719259585720582e-06, | |
| "loss": 0.1305, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.6738987508218277, | |
| "grad_norm": 0.5308182239532471, | |
| "learning_rate": 9.716505068312033e-06, | |
| "loss": 0.1303, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 0.6793776024545255, | |
| "grad_norm": 0.5280865430831909, | |
| "learning_rate": 9.713750550903483e-06, | |
| "loss": 0.1255, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6848564540872233, | |
| "grad_norm": 0.6718000769615173, | |
| "learning_rate": 9.710996033494933e-06, | |
| "loss": 0.1281, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.6903353057199211, | |
| "grad_norm": 0.5682989954948425, | |
| "learning_rate": 9.708241516086384e-06, | |
| "loss": 0.1268, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.6958141573526189, | |
| "grad_norm": 0.6062389016151428, | |
| "learning_rate": 9.705486998677832e-06, | |
| "loss": 0.1286, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 0.7012930089853167, | |
| "grad_norm": 0.536504864692688, | |
| "learning_rate": 9.702732481269283e-06, | |
| "loss": 0.1304, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.7067718606180144, | |
| "grad_norm": 0.5489104986190796, | |
| "learning_rate": 9.699977963860733e-06, | |
| "loss": 0.1264, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 0.7122507122507122, | |
| "grad_norm": 0.6555972695350647, | |
| "learning_rate": 9.697223446452182e-06, | |
| "loss": 0.1271, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.71772956388341, | |
| "grad_norm": 0.6393175721168518, | |
| "learning_rate": 9.694468929043632e-06, | |
| "loss": 0.1297, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 0.7232084155161078, | |
| "grad_norm": 0.4772554934024811, | |
| "learning_rate": 9.691714411635082e-06, | |
| "loss": 0.1264, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.7232084155161078, | |
| "eval_loss": 0.10748741775751114, | |
| "eval_runtime": 1333.6749, | |
| "eval_samples_per_second": 2.497, | |
| "eval_steps_per_second": 0.052, | |
| "eval_wer": 14.096527369040611, | |
| "step": 3300 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 91260, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.233276471284146e+20, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |