1{"step": 0, "loss": 0.8037000298500061, "lr": 1e-05, "grad_norm": 6.844403266906738, "tokens": 157, "tok_per_s": 113.42990230816216, "t": 2.41}2{"step": 50, "loss": 0.42063286900520325, "lr": 0.00051, "grad_norm": 0.9522086977958679, "tokens": 8616, "tok_per_s": 576.7258213814504, "t": 16.07}3{"step": 100, "loss": 0.4459417164325714, "lr": 0.001, "grad_norm": 1.2287894487380981, "tokens": 16635, "tok_per_s": 656.5870056008258, "t": 26.36}4{"step": 150, "loss": 0.5174918174743652, "lr": 0.0009993400362330058, "grad_norm": 0.6301496624946594, "tokens": 25042, "tok_per_s": 766.5616342932171, "t": 33.7}5{"step": 200, "loss": 0.5022031664848328, "lr": 0.0009973620807194618, "grad_norm": 0.7607206702232361, "tokens": 33661, "tok_per_s": 840.1927279960261, "t": 41.09}6{"step": 250, "loss": 0.4696323871612549, "lr": 0.000994071935143687, "grad_norm": 0.6131431460380554, "tokens": 42410, "tok_per_s": 895.2385005087517, "t": 48.4}7{"step": 300, "loss": 0.4114721417427063, "lr": 0.000989479250069539, "grad_norm": 0.6016066670417786, "tokens": 50771, "tok_per_s": 927.1228233959765, "t": 55.8}8{"step": 350, "loss": 0.42751550674438477, "lr": 0.0009835974966336503, "grad_norm": 0.5024498105049133, "tokens": 59133, "tok_per_s": 952.1144545001423, "t": 63.13}9{"step": 400, "loss": 0.5906541347503662, "lr": 0.0009764439270322612, "grad_norm": 0.5494341850280762, "tokens": 67299, "tok_per_s": 969.4613011442525, "t": 70.44}10{"step": 450, "loss": 0.4852248728275299, "lr": 0.0009680395239175562, "grad_norm": 0.3938865661621094, "tokens": 75575, "tok_per_s": 986.8625642358254, "t": 77.61}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45602453089864925, "val_loss_b2": 0.46150722193447696, "val_loss_b4": 0.47519532302848927, "val_loss_b8": 0.49500827779906204, "t": 88.31}12{"step": 500, "loss": 0.4366694986820221, "lr": 0.0009584089388519307, "grad_norm": 0.45590856671333313, "tokens": 84036, "tok_per_s": 960.8362839139693, "t": 88.49}13{"step": 550, "loss": 0.29039642214775085, "lr": 0.0009475804200007103, "grad_norm": 0.4643513560295105, "tokens": 92359, "tok_per_s": 974.5541445589798, "t": 95.79}14{"step": 600, "loss": 0.32419073581695557, "lr": 0.0009355857292754152, "grad_norm": 0.2859732508659363, "tokens": 100627, "tok_per_s": 985.2654907451823, "t": 103.16}15{"step": 650, "loss": 0.5258869528770447, "lr": 0.0009224600491706008, "grad_norm": 0.483050137758255, "tokens": 108981, "tok_per_s": 995.7396108691922, "t": 110.47}16{"step": 700, "loss": 0.40977731347084045, "lr": 0.0009082418795675396, "grad_norm": 0.4632984697818756, "tokens": 117352, "tok_per_s": 1006.4020494232938, "t": 117.63}17{"step": 750, "loss": 0.49572449922561646, "lr": 0.0008929729248074364, "grad_norm": 0.300184041261673, "tokens": 125499, "tok_per_s": 1013.0382385577657, "t": 124.91}18{"step": 800, "loss": 0.38430145382881165, "lr": 0.0008766979713654089, "grad_norm": 0.3628772795200348, "tokens": 133880, "tok_per_s": 1020.4753572903207, "t": 132.23}19{"step": 850, "loss": 0.46930575370788574, "lr": 0.0008594647564840406, "grad_norm": 0.3109002113342285, "tokens": 142063, "tok_per_s": 1026.3411048740304, "t": 139.44}20{"step": 900, "loss": 0.3519968092441559, "lr": 0.0008413238281518224, "grad_norm": 0.24886322021484375, "tokens": 150436, "tok_per_s": 1032.629347959949, "t": 146.71}21{"step": 950, "loss": 0.41887497901916504, "lr": 0.0008223283968371944, "grad_norm": 0.33314406871795654, "tokens": 158527, "tok_per_s": 1036.1990695661163, "t": 154.01}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4543949078314994, "val_loss_b2": 0.456732710927152, "val_loss_b4": 0.46626014389857806, "val_loss_b8": 0.4808533115302359, "t": 164.84}23{"step": 1000, "loss": 0.4150533378124237, "lr": 0.000802534179413072, "grad_norm": 0.3819475471973419, "tokens": 166917, "tok_per_s": 1018.0268214297902, "t": 164.99}24{"step": 1050, "loss": 0.4375992715358734, "lr": 0.0007819992357296598, "grad_norm": 0.28463947772979736, "tokens": 175091, "tok_per_s": 1021.0970851829112, "t": 172.5}25{"step": 1100, "loss": 0.36315029859542847, "lr": 0.0007607837983149056, "grad_norm": 0.37634792923927307, "tokens": 183159, "tok_per_s": 1024.898569247119, "t": 179.75}26{"step": 1150, "loss": 0.2975662648677826, "lr": 0.0007389500957021191, "grad_norm": 0.2428784966468811, "tokens": 191581, "tok_per_s": 1028.657217136464, "t": 187.27}27{"step": 1200, "loss": 0.5097301602363586, "lr": 0.0007165621699029615, "grad_norm": 0.31668105721473694, "tokens": 200185, "tok_per_s": 1034.0241395695932, "t": 194.63}28{"step": 1250, "loss": 0.598894476890564, "lr": 0.000693685688561191, "grad_norm": 0.455051988363266, "tokens": 208401, "tok_per_s": 1035.4910054210613, "t": 202.28}29{"step": 1300, "loss": 0.4287705421447754, "lr": 0.0006703877523381495, "grad_norm": 0.25633248686790466, "tokens": 216624, "tok_per_s": 1038.1547520128609, "t": 209.69}30{"step": 1350, "loss": 0.5249320864677429, "lr": 0.0006467366980949609, "grad_norm": 0.4181002080440521, "tokens": 225093, "tok_per_s": 1041.7413050221696, "t": 217.1}31{"step": 1400, "loss": 0.3932661712169647, "lr": 0.0006228018984487442, "grad_norm": 0.3809398412704468, "tokens": 233304, "tok_per_s": 1044.0471193884905, "t": 224.49}32{"step": 1450, "loss": 0.4037530720233917, "lr": 0.0005986535582907738, "grad_norm": 0.3494563102722168, "tokens": 241497, "tok_per_s": 1046.4387155346137, "t": 231.8}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45060894300104537, "val_loss_b2": 0.4475770793330088, "val_loss_b4": 0.44943427851542045, "val_loss_b8": 0.45889760896805887, "t": 242.68}34{"step": 1500, "loss": 0.44686347246170044, "lr": 0.000574362508863438, "grad_norm": 0.28552675247192383, "tokens": 249582, "tok_per_s": 1032.0551034210926, "t": 242.86}35{"step": 1550, "loss": 0.30785953998565674, "lr": 0.00055, "grad_norm": 0.24650314450263977, "tokens": 257750, "tok_per_s": 1034.9591082621985, "t": 250.07}36{"step": 1600, "loss": 0.42331069707870483, "lr": 0.0005256374911365621, "grad_norm": 0.3392142653465271, "tokens": 266068, "tok_per_s": 1037.9404158885227, "t": 257.36}37{"step": 1650, "loss": 0.4791591465473175, "lr": 0.0005013464417092262, "grad_norm": 0.3857705295085907, "tokens": 274463, "tok_per_s": 1040.3811274831244, "t": 264.84}38{"step": 1700, "loss": 0.2984299957752228, "lr": 0.00047719810155125587, "grad_norm": 0.3463330864906311, "tokens": 282527, "tok_per_s": 1041.689216522204, "t": 272.24}39{"step": 1750, "loss": 0.4619162380695343, "lr": 0.0004532633019050392, "grad_norm": 0.3091307580471039, "tokens": 290793, "tok_per_s": 1043.3458197402385, "t": 279.74}40{"step": 1800, "loss": 0.4184896945953369, "lr": 0.0004296122476618507, "grad_norm": 0.316865473985672, "tokens": 299230, "tok_per_s": 1045.6612014914963, "t": 287.19}41{"step": 1850, "loss": 0.369024395942688, "lr": 0.00040631431143880916, "grad_norm": 0.34649598598480225, "tokens": 307566, "tok_per_s": 1048.3282156083312, "t": 294.41}42{"step": 1900, "loss": 0.38607558608055115, "lr": 0.00038343783009703845, "grad_norm": 0.27774542570114136, "tokens": 315780, "tok_per_s": 1050.175065968497, "t": 301.72}43{"step": 1950, "loss": 0.4044061601161957, "lr": 0.000361049904297881, "grad_norm": 0.2859710156917572, "tokens": 324229, "tok_per_s": 1052.189927223667, "t": 309.17}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.44869982050580304, "val_loss_b2": 0.44706041816421893, "val_loss_b4": 0.44987122397989165, "val_loss_b8": 0.45629776493559454, "t": 319.95}45{"step": 2000, "loss": 0.45515358448028564, "lr": 0.0003392162016850945, "grad_norm": 0.32580721378326416, "tokens": 332550, "tok_per_s": 1042.2083506452743, "t": 320.11}46{"step": 2050, "loss": 0.5230012536048889, "lr": 0.0003180007642703402, "grad_norm": 0.3839978277683258, "tokens": 340880, "tok_per_s": 1043.898405410261, "t": 327.57}47{"step": 2100, "loss": 0.3024647533893585, "lr": 0.000297465820586928, "grad_norm": 0.35114040970802307, "tokens": 349421, "tok_per_s": 1046.410622445776, "t": 334.95}48{"step": 2150, "loss": 0.40944814682006836, "lr": 0.0002776716031628058, "grad_norm": 0.37512972950935364, "tokens": 357735, "tok_per_s": 1048.0825104224796, "t": 342.35}49{"step": 2200, "loss": 0.45188620686531067, "lr": 0.00025867617184817756, "grad_norm": 0.4663980305194855, "tokens": 366168, "tok_per_s": 1049.934410420727, "t": 349.78}50{"step": 2250, "loss": 0.3599855303764343, "lr": 0.00024053524351595949, "grad_norm": 0.259933203458786, "tokens": 374489, "tok_per_s": 1051.8360777135051, "t": 357.06}51{"step": 2300, "loss": 0.32005634903907776, "lr": 0.0002233020286345913, "grad_norm": 0.29534101486206055, "tokens": 382832, "tok_per_s": 1053.7181897406365, "t": 364.34}52{"step": 2350, "loss": 0.5054188966751099, "lr": 0.00020702707519256364, "grad_norm": 0.3756987452507019, "tokens": 391097, "tok_per_s": 1055.1558788329398, "t": 371.68}53{"step": 2400, "loss": 0.24998830258846283, "lr": 0.00019175812043246033, "grad_norm": 0.23193702101707458, "tokens": 399648, "tok_per_s": 1057.1422044939397, "t": 379.07}54{"step": 2450, "loss": 0.35542044043540955, "lr": 0.00017753995082939931, "grad_norm": 0.3055308163166046, "tokens": 407914, "tok_per_s": 1058.0476814190697, "t": 386.56}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4480612760932061, "val_loss_b2": 0.44662669516949305, "val_loss_b4": 0.44857055697904075, "val_loss_b8": 0.4507717472864173, "t": 397.12}56{"step": 2500, "loss": 0.26540321111679077, "lr": 0.00016441427072458494, "grad_norm": 0.24998833239078522, "tokens": 416524, "tok_per_s": 1051.1790741584086, "t": 397.27}57{"step": 2550, "loss": 0.3136857748031616, "lr": 0.0001524195799992897, "grad_norm": 0.31415534019470215, "tokens": 424563, "tok_per_s": 1051.7894590446622, "t": 404.68}58{"step": 2600, "loss": 0.2800491154193878, "lr": 0.0001415910611480694, "grad_norm": 0.3327774405479431, "tokens": 432991, "tok_per_s": 1053.3795110204705, "t": 412.07}59{"step": 2650, "loss": 0.37426966428756714, "lr": 0.00013196047608244387, "grad_norm": 0.45562273263931274, "tokens": 441165, "tok_per_s": 1054.0526813099998, "t": 419.57}60{"step": 2700, "loss": 0.3309275805950165, "lr": 0.00012355607296773896, "grad_norm": 0.21226947009563446, "tokens": 449578, "tok_per_s": 1055.6517357665955, "t": 426.9}61{"step": 2750, "loss": 0.3827971816062927, "lr": 0.00011640250336634969, "grad_norm": 0.393187940120697, "tokens": 457759, "tok_per_s": 1056.5969486944014, "t": 434.26}62{"step": 2800, "loss": 0.3721027076244354, "lr": 0.00011052074993046101, "grad_norm": 0.34396201372146606, "tokens": 465905, "tok_per_s": 1057.264291505355, "t": 441.7}63{"step": 2850, "loss": 0.39056888222694397, "lr": 0.00010592806485631325, "grad_norm": 0.4981256425380707, "tokens": 474059, "tok_per_s": 1059.7168733076867, "t": 448.37}64{"step": 2900, "loss": 0.517935037612915, "lr": 0.00010263791928053818, "grad_norm": 0.3624151349067688, "tokens": 482466, "tok_per_s": 1062.5151534671477, "t": 455.1}65{"step": 2950, "loss": 0.3701931834220886, "lr": 0.0001006599637669943, "grad_norm": 0.3012923300266266, "tokens": 490971, "tok_per_s": 1064.200164750797, "t": 462.38}66{"step": 2999, "loss": 0.35989832878112793, "lr": 0.0001000002640500325, "grad_norm": 0.43859222531318665, "tokens": 499017, "tok_per_s": 1065.1031673452076, "t": 469.54}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4463276588135367, "val_loss_b2": 0.44537581288209394, "val_loss_b4": 0.448317376893985, "val_loss_b8": 0.45043363237451484, "t": 473.17}68{"step": 2999, "checkpoint_sha256": "29c37958b4eefbbcba2b5fb04b3f1ee570c53678afafc28776ec659d5cbace17", "t": 473.3}69