1{"step": 0, "loss": 0.4137183129787445, "lr": 1e-05, "grad_norm": 0.10044993460178375, "tokens": 157, "tok_per_s": 110.57646325308255, "t": 2.0}2{"step": 50, "loss": 0.3858620524406433, "lr": 0.00051, "grad_norm": 0.06520561873912811, "tokens": 8616, "tok_per_s": 352.8851899791728, "t": 24.99}3{"step": 100, "loss": 0.40426504611968994, "lr": 0.001, "grad_norm": 0.07917547971010208, "tokens": 16635, "tok_per_s": 353.1679023365741, "t": 47.68}4{"step": 150, "loss": 0.533927321434021, "lr": 0.0009993400362330058, "grad_norm": 0.0400385819375515, "tokens": 25042, "tok_per_s": 358.545548711767, "t": 70.42}5{"step": 200, "loss": 0.4574252665042877, "lr": 0.0009973620807194618, "grad_norm": 0.049605514854192734, "tokens": 33661, "tok_per_s": 365.8310516252515, "t": 92.6}6{"step": 250, "loss": 0.4378852844238281, "lr": 0.000994071935143687, "grad_norm": 0.08066040277481079, "tokens": 42410, "tok_per_s": 370.7679679607564, "t": 114.96}7{"step": 300, "loss": 0.3958844840526581, "lr": 0.000989479250069539, "grad_norm": 0.06712526828050613, "tokens": 50771, "tok_per_s": 372.22939584197604, "t": 136.98}8{"step": 350, "loss": 0.42079076170921326, "lr": 0.0009835974966336503, "grad_norm": 0.11212937533855438, "tokens": 59133, "tok_per_s": 373.0011554500961, "t": 159.11}9{"step": 400, "loss": 0.5918116569519043, "lr": 0.0009764439270322612, "grad_norm": 0.06798093020915985, "tokens": 67299, "tok_per_s": 372.15348383780355, "t": 181.41}10{"step": 450, "loss": 0.46382343769073486, "lr": 0.0009680395239175562, "grad_norm": 0.015121468342840672, "tokens": 75575, "tok_per_s": 372.3986252025674, "t": 203.52}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4597370012338788, "val_loss_b2": 0.45776695450754107, "val_loss_b4": 0.4561186696931962, "val_loss_b8": 0.4560271084984281, "t": 232.66}12{"step": 500, "loss": 0.40864333510398865, "lr": 0.0009584089388519307, "grad_norm": 0.05807528272271156, "tokens": 84036, "tok_per_s": 361.3287513946813, "t": 233.15}13{"step": 550, "loss": 0.28721174597740173, "lr": 0.0009475804200007103, "grad_norm": 0.08433183282613754, "tokens": 92359, "tok_per_s": 361.9514280629721, "t": 255.75}14{"step": 600, "loss": 0.3134419322013855, "lr": 0.0009355857292754152, "grad_norm": 0.045008111745119095, "tokens": 100627, "tok_per_s": 363.01403234170647, "t": 277.77}15{"step": 650, "loss": 0.5152975916862488, "lr": 0.0009224600491706008, "grad_norm": 0.08435051143169403, "tokens": 108981, "tok_per_s": 363.9216188979352, "t": 300.04}16{"step": 700, "loss": 0.38695329427719116, "lr": 0.0009082418795675396, "grad_norm": 0.087889164686203, "tokens": 117352, "tok_per_s": 364.63599134174325, "t": 322.41}17{"step": 750, "loss": 0.5036046504974365, "lr": 0.0008929729248074364, "grad_norm": 0.07477554678916931, "tokens": 125499, "tok_per_s": 364.7053974294801, "t": 344.69}18{"step": 800, "loss": 0.37904876470565796, "lr": 0.0008766979713654089, "grad_norm": 0.06618175655603409, "tokens": 133880, "tok_per_s": 365.5902073349358, "t": 366.78}19{"step": 850, "loss": 0.468196302652359, "lr": 0.0008594647564840406, "grad_norm": 0.05971226096153259, "tokens": 142063, "tok_per_s": 365.9379469981493, "t": 388.79}20{"step": 900, "loss": 0.34824201464653015, "lr": 0.0008413238281518224, "grad_norm": 0.06103726848959923, "tokens": 150436, "tok_per_s": 366.9154569343321, "t": 410.58}21{"step": 950, "loss": 0.40944772958755493, "lr": 0.0008223283968371944, "grad_norm": 0.09832854568958282, "tokens": 158527, "tok_per_s": 366.8356609075256, "t": 432.72}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4592474875866104, "val_loss_b2": 0.4574946041116578, "val_loss_b4": 0.4579144116397442, "val_loss_b8": 0.4660264791203814, "t": 461.54}23{"step": 1000, "loss": 0.39744919538497925, "lr": 0.000802534179413072, "grad_norm": 0.08527916669845581, "tokens": 166917, "tok_per_s": 361.7136840816551, "t": 462.04}24{"step": 1050, "loss": 0.4436836838722229, "lr": 0.0007819992357296598, "grad_norm": 0.07505074143409729, "tokens": 175091, "tok_per_s": 362.20084414952635, "t": 483.98}25{"step": 1100, "loss": 0.35451316833496094, "lr": 0.0007607837983149056, "grad_norm": 0.08908887952566147, "tokens": 183159, "tok_per_s": 362.05806049915236, "t": 506.46}26{"step": 1150, "loss": 0.2943210303783417, "lr": 0.0007389500957021191, "grad_norm": 0.07543206214904785, "tokens": 191581, "tok_per_s": 362.6823305412923, "t": 528.81}27{"step": 1200, "loss": 0.5098519921302795, "lr": 0.0007165621699029615, "grad_norm": 0.1077631488442421, "tokens": 200185, "tok_per_s": 363.6277754986882, "t": 551.1}28{"step": 1250, "loss": 0.5962792634963989, "lr": 0.000693685688561191, "grad_norm": 0.09632866084575653, "tokens": 208401, "tok_per_s": 364.1356649772418, "t": 572.9}29{"step": 1300, "loss": 0.43365633487701416, "lr": 0.0006703877523381495, "grad_norm": 0.08209537714719772, "tokens": 216624, "tok_per_s": 364.19793540759633, "t": 595.37}30{"step": 1350, "loss": 0.5365902781486511, "lr": 0.0006467366980949609, "grad_norm": 0.1703610122203827, "tokens": 225093, "tok_per_s": 364.8580509675281, "t": 617.51}31{"step": 1400, "loss": 0.40109652280807495, "lr": 0.0006228018984487442, "grad_norm": 0.14173512160778046, "tokens": 233304, "tok_per_s": 364.98585826276576, "t": 639.79}32{"step": 1450, "loss": 0.41093844175338745, "lr": 0.0005986535582907738, "grad_norm": 0.18193194270133972, "tokens": 241497, "tok_per_s": 364.77421081456043, "t": 662.62}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45820023474169225, "val_loss_b2": 0.45515030155398095, "val_loss_b4": 0.4523657944003839, "val_loss_b8": 0.45371860186415625, "t": 691.59}34{"step": 1500, "loss": 0.44612929224967957, "lr": 0.000574362508863438, "grad_norm": 0.09360631555318832, "tokens": 249582, "tok_per_s": 360.93843936916954, "t": 692.06}35{"step": 1550, "loss": 0.3238929510116577, "lr": 0.00055, "grad_norm": 0.07228416204452515, "tokens": 257750, "tok_per_s": 361.0698478284881, "t": 714.43}36{"step": 1600, "loss": 0.43549948930740356, "lr": 0.0005256374911365621, "grad_norm": 0.14837487041950226, "tokens": 266068, "tok_per_s": 361.2441316545166, "t": 737.11}37{"step": 1650, "loss": 0.46337950229644775, "lr": 0.0005013464417092262, "grad_norm": 0.160983607172966, "tokens": 274463, "tok_per_s": 361.3457212075176, "t": 760.13}38{"step": 1700, "loss": 0.30904147028923035, "lr": 0.00047719810155125587, "grad_norm": 0.1331629753112793, "tokens": 282527, "tok_per_s": 361.3411182169491, "t": 782.46}39{"step": 1750, "loss": 0.46572744846343994, "lr": 0.0004532633019050392, "grad_norm": 0.11443528532981873, "tokens": 290793, "tok_per_s": 361.3097991582336, "t": 805.41}40{"step": 1800, "loss": 0.4176121652126312, "lr": 0.0004296122476618507, "grad_norm": 0.08894649893045425, "tokens": 299230, "tok_per_s": 361.6833318911378, "t": 827.9}41{"step": 1850, "loss": 0.38357260823249817, "lr": 0.00040631431143880916, "grad_norm": 0.12010200321674347, "tokens": 307566, "tok_per_s": 361.7583467454993, "t": 850.77}42{"step": 1900, "loss": 0.3929412066936493, "lr": 0.00038343783009703845, "grad_norm": 0.10914478451013565, "tokens": 315780, "tok_per_s": 361.6524413214449, "t": 873.73}43{"step": 1950, "loss": 0.41690322756767273, "lr": 0.000361049904297881, "grad_norm": 0.12565456330776215, "tokens": 324229, "tok_per_s": 361.84309772545487, "t": 896.63}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45526016633930083, "val_loss_b2": 0.4523257071775659, "val_loss_b4": 0.4511483363061776, "val_loss_b8": 0.4613353452875204, "t": 925.71}45{"step": 2000, "loss": 0.4632568955421448, "lr": 0.0003392162016850945, "grad_norm": 0.1264740228652954, "tokens": 332550, "tok_per_s": 359.281743046281, "t": 926.17}46{"step": 2050, "loss": 0.5164177417755127, "lr": 0.0003180007642703402, "grad_norm": 0.14528663456439972, "tokens": 340880, "tok_per_s": 359.39192069344836, "t": 949.07}47{"step": 2100, "loss": 0.3050396144390106, "lr": 0.000297465820586928, "grad_norm": 0.22422313690185547, "tokens": 349421, "tok_per_s": 359.6148130380448, "t": 972.23}48{"step": 2150, "loss": 0.40407732129096985, "lr": 0.0002776716031628058, "grad_norm": 0.13354219496250153, "tokens": 357735, "tok_per_s": 359.6221965946385, "t": 995.33}49{"step": 2200, "loss": 0.4561748504638672, "lr": 0.00025867617184817756, "grad_norm": 0.15694083273410797, "tokens": 366168, "tok_per_s": 359.7171820514619, "t": 1018.51}50{"step": 2250, "loss": 0.38216352462768555, "lr": 0.00024053524351595949, "grad_norm": 0.15827631950378418, "tokens": 374489, "tok_per_s": 359.8972627719169, "t": 1041.12}51{"step": 2300, "loss": 0.3161202073097229, "lr": 0.0002233020286345913, "grad_norm": 0.1013115718960762, "tokens": 382832, "tok_per_s": 360.0900573368917, "t": 1063.73}52{"step": 2350, "loss": 0.5332074761390686, "lr": 0.00020702707519256364, "grad_norm": 0.1345881074666977, "tokens": 391097, "tok_per_s": 360.20406128830825, "t": 1086.34}53{"step": 2400, "loss": 0.2796306312084198, "lr": 0.00019175812043246033, "grad_norm": 0.14743152260780334, "tokens": 399648, "tok_per_s": 360.4025356652881, "t": 1109.47}54{"step": 2450, "loss": 0.3587510883808136, "lr": 0.00017753995082939931, "grad_norm": 0.10878352075815201, "tokens": 407914, "tok_per_s": 360.68772790593596, "t": 1131.51}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45391109366908106, "val_loss_b2": 0.4510277663033957, "val_loss_b4": 0.4492735789878548, "val_loss_b8": 0.45716799439439165, "t": 1161.04}56{"step": 2500, "loss": 0.25248464941978455, "lr": 0.00016441427072458494, "grad_norm": 0.12934383749961853, "tokens": 416524, "tok_per_s": 358.7910106442589, "t": 1161.49}57{"step": 2550, "loss": 0.3198677599430084, "lr": 0.0001524195799992897, "grad_norm": 0.08468490839004517, "tokens": 424563, "tok_per_s": 358.73006003080604, "t": 1184.09}58{"step": 2600, "loss": 0.28548017144203186, "lr": 0.0001415910611480694, "grad_norm": 0.1642276644706726, "tokens": 432991, "tok_per_s": 358.99486533289587, "t": 1206.69}59{"step": 2650, "loss": 0.3853985667228699, "lr": 0.00013196047608244387, "grad_norm": 0.15403875708580017, "tokens": 441165, "tok_per_s": 359.0488478238739, "t": 1229.28}60{"step": 2700, "loss": 0.34854891896247864, "lr": 0.00012355607296773896, "grad_norm": 0.07540357112884521, "tokens": 449578, "tok_per_s": 359.225430391451, "t": 1252.1}61{"step": 2750, "loss": 0.3658033013343811, "lr": 0.00011640250336634969, "grad_norm": 0.15412695705890656, "tokens": 457759, "tok_per_s": 359.21405724097565, "t": 1274.91}62{"step": 2800, "loss": 0.3679767847061157, "lr": 0.00011052074993046101, "grad_norm": 0.11753562837839127, "tokens": 465905, "tok_per_s": 359.18129457119943, "t": 1297.71}63{"step": 2850, "loss": 0.3965524733066559, "lr": 0.00010592806485631325, "grad_norm": 0.23466067016124725, "tokens": 474059, "tok_per_s": 359.1115845180539, "t": 1320.66}64{"step": 2900, "loss": 0.5315443277359009, "lr": 0.00010263791928053818, "grad_norm": 0.12089382112026215, "tokens": 482466, "tok_per_s": 359.35251398290194, "t": 1343.17}65{"step": 2950, "loss": 0.37633103132247925, "lr": 0.0001006599637669943, "grad_norm": 0.19408325850963593, "tokens": 490971, "tok_per_s": 359.59745300445, "t": 1365.91}66{"step": 2999, "loss": 0.36032557487487793, "lr": 0.0001000002640500325, "grad_norm": 0.16732542216777802, "tokens": 499017, "tok_per_s": 359.5962858037261, "t": 1388.29}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4517723805449642, "val_loss_b2": 0.44892461785193793, "val_loss_b4": 0.44919403155135545, "val_loss_b8": 0.46115965958356037, "t": 1395.07}68{"step": 2999, "checkpoint_sha256": "1ce74b9e2c601b4b95c3f7bae72ce8d6803b3f86233e0d73e1a8e4e1d84fe27a", "t": 1395.19}69