1{"step": 0, "loss": 0.5386869311332703, "lr": 1e-05, "grad_norm": 2.047222137451172, "tokens": 157, "tok_per_s": 108.53798837755332, "t": 2.06}2{"step": 50, "loss": 0.4027070701122284, "lr": 0.00051, "grad_norm": 0.6417726278305054, "tokens": 8616, "tok_per_s": 151.0746099762931, "t": 57.64}3{"step": 100, "loss": 0.388218492269516, "lr": 0.001, "grad_norm": 0.24445490539073944, "tokens": 16635, "tok_per_s": 149.97555068847748, "t": 111.53}4{"step": 150, "loss": 0.5297861695289612, "lr": 0.0009993400362330058, "grad_norm": 0.09588441252708435, "tokens": 25042, "tok_per_s": 147.24815661876391, "t": 170.68}5{"step": 200, "loss": 0.4561012089252472, "lr": 0.0009973620807194618, "grad_norm": 0.026288924738764763, "tokens": 33661, "tok_per_s": 147.32162247246347, "t": 229.1}6{"step": 250, "loss": 0.4421783685684204, "lr": 0.000994071935143687, "grad_norm": 0.12178074568510056, "tokens": 42410, "tok_per_s": 147.74345903120184, "t": 287.66}7{"step": 300, "loss": 0.39179977774620056, "lr": 0.000989479250069539, "grad_norm": 0.14763128757476807, "tokens": 50771, "tok_per_s": 149.17345450422212, "t": 340.96}8{"step": 350, "loss": 0.4274325370788574, "lr": 0.0009835974966336503, "grad_norm": 0.0800609216094017, "tokens": 59133, "tok_per_s": 149.8500495720553, "t": 395.23}9{"step": 400, "loss": 0.5898694396018982, "lr": 0.0009764439270322612, "grad_norm": 0.036028724163770676, "tokens": 67299, "tok_per_s": 154.79297467731337, "t": 435.38}10{"step": 450, "loss": 0.46369680762290955, "lr": 0.0009680395239175562, "grad_norm": 0.012092310935258865, "tokens": 75575, "tok_per_s": 159.65829860416582, "t": 473.97}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4600998317742947, "val_loss_b2": 0.4591627877706845, "val_loss_b4": 0.458192854458737, "val_loss_b8": 0.4576896274542209, "t": 520.74}12{"step": 500, "loss": 0.410146027803421, "lr": 0.0009584089388519307, "grad_norm": 0.04791094362735748, "tokens": 84036, "tok_per_s": 161.30938491244189, "t": 521.57}13{"step": 550, "loss": 0.2804412245750427, "lr": 0.0009475804200007103, "grad_norm": 0.14613698422908783, "tokens": 92359, "tok_per_s": 164.95889405610785, "t": 560.5}14{"step": 600, "loss": 0.3099477291107178, "lr": 0.0009355857292754152, "grad_norm": 0.021462541073560715, "tokens": 100627, "tok_per_s": 168.19695181770106, "t": 598.88}15{"step": 650, "loss": 0.5158343315124512, "lr": 0.0009224600491706008, "grad_norm": 0.04912424087524414, "tokens": 108981, "tok_per_s": 171.17123718585367, "t": 637.29}16{"step": 700, "loss": 0.39171019196510315, "lr": 0.0009082418795675396, "grad_norm": 0.08785135298967361, "tokens": 117352, "tok_per_s": 173.79025825030234, "t": 675.86}17{"step": 750, "loss": 0.5009042024612427, "lr": 0.0008929729248074364, "grad_norm": 0.03156060725450516, "tokens": 125499, "tok_per_s": 175.84512158047488, "t": 714.3}18{"step": 800, "loss": 0.38524213433265686, "lr": 0.0008766979713654089, "grad_norm": 0.060281507670879364, "tokens": 133880, "tok_per_s": 177.98306300120157, "t": 752.82}19{"step": 850, "loss": 0.4725104570388794, "lr": 0.0008594647564840406, "grad_norm": 0.12807083129882812, "tokens": 142063, "tok_per_s": 179.88573821773767, "t": 790.35}20{"step": 900, "loss": 0.35125645995140076, "lr": 0.0008413238281518224, "grad_norm": 0.12373542785644531, "tokens": 150436, "tok_per_s": 181.58515282575408, "t": 829.07}21{"step": 950, "loss": 0.42932042479515076, "lr": 0.0008223283968371944, "grad_norm": 0.24489270150661469, "tokens": 158527, "tok_per_s": 182.88139827410885, "t": 867.44}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45945562641863874, "val_loss_b2": 0.4588168977574799, "val_loss_b4": 0.4600678317472816, "val_loss_b8": 0.4664086745606672, "t": 914.51}23{"step": 1000, "loss": 0.3966899812221527, "lr": 0.000802534179413072, "grad_norm": 0.12175209075212479, "tokens": 166917, "tok_per_s": 182.48083048658242, "t": 915.32}24{"step": 1050, "loss": 0.44860029220581055, "lr": 0.0007819992357296598, "grad_norm": 0.09363958984613419, "tokens": 175091, "tok_per_s": 183.69255600230042, "t": 953.79}25{"step": 1100, "loss": 0.36443936824798584, "lr": 0.0007607837983149056, "grad_norm": 0.11300823837518692, "tokens": 183159, "tok_per_s": 184.63379566930055, "t": 992.62}26{"step": 1150, "loss": 0.29148757457733154, "lr": 0.0007389500957021191, "grad_norm": 0.05884606018662453, "tokens": 191581, "tok_per_s": 185.84910469856806, "t": 1031.45}27{"step": 1200, "loss": 0.5251035690307617, "lr": 0.0007165621699029615, "grad_norm": 0.11791376024484634, "tokens": 200185, "tok_per_s": 187.07470896176733, "t": 1070.69}28{"step": 1250, "loss": 0.5974413156509399, "lr": 0.000693685688561191, "grad_norm": 0.20647424459457397, "tokens": 208401, "tok_per_s": 187.86359063644076, "t": 1109.93}29{"step": 1300, "loss": 0.4335373640060425, "lr": 0.0006703877523381495, "grad_norm": 0.09647855907678604, "tokens": 216624, "tok_per_s": 188.69433493404804, "t": 1148.63}30{"step": 1350, "loss": 0.5368632674217224, "lr": 0.0006467366980949609, "grad_norm": 0.24670147895812988, "tokens": 225093, "tok_per_s": 189.69969130356372, "t": 1187.19}31{"step": 1400, "loss": 0.39917272329330444, "lr": 0.0006228018984487442, "grad_norm": 0.11133939772844315, "tokens": 233304, "tok_per_s": 190.4038452890033, "t": 1225.92}32{"step": 1450, "loss": 0.40073394775390625, "lr": 0.0005986535582907738, "grad_norm": 0.08613388985395432, "tokens": 241497, "tok_per_s": 190.98942743509858, "t": 1265.06}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45951020629491757, "val_loss_b2": 0.457997818895606, "val_loss_b4": 0.4560729121268824, "val_loss_b8": 0.45441868153624254, "t": 1311.78}34{"step": 1500, "loss": 0.4567154347896576, "lr": 0.000574362508863438, "grad_norm": 0.0857902467250824, "tokens": 249582, "tok_per_s": 190.24127127445163, "t": 1312.53}35{"step": 1550, "loss": 0.3248290717601776, "lr": 0.00055, "grad_norm": 0.11136104166507721, "tokens": 257750, "tok_per_s": 190.88391306838335, "t": 1350.91}36{"step": 1600, "loss": 0.44797515869140625, "lr": 0.0005256374911365621, "grad_norm": 0.11583836376667023, "tokens": 266068, "tok_per_s": 191.62138082383635, "t": 1389.12}37{"step": 1650, "loss": 0.4636852443218231, "lr": 0.0005013464417092262, "grad_norm": 0.09402605146169662, "tokens": 274463, "tok_per_s": 192.21849767955476, "t": 1428.48}38{"step": 1700, "loss": 0.31061282753944397, "lr": 0.00047719810155125587, "grad_norm": 0.14376108348369598, "tokens": 282527, "tok_per_s": 192.70873722319683, "t": 1466.69}39{"step": 1750, "loss": 0.4656156301498413, "lr": 0.0004532633019050392, "grad_norm": 0.08692286908626556, "tokens": 290793, "tok_per_s": 193.20096390628433, "t": 1505.74}40{"step": 1800, "loss": 0.4125155508518219, "lr": 0.0004296122476618507, "grad_norm": 0.12374787777662277, "tokens": 299230, "tok_per_s": 193.89576936706504, "t": 1543.86}41{"step": 1850, "loss": 0.38021793961524963, "lr": 0.00040631431143880916, "grad_norm": 0.13919970393180847, "tokens": 307566, "tok_per_s": 194.43673426771232, "t": 1582.44}42{"step": 1900, "loss": 0.4004985988140106, "lr": 0.00038343783009703845, "grad_norm": 0.1314348429441452, "tokens": 315780, "tok_per_s": 194.82937678032528, "t": 1621.41}43{"step": 1950, "loss": 0.41394156217575073, "lr": 0.000361049904297881, "grad_norm": 0.10713706165552139, "tokens": 324229, "tok_per_s": 195.33619878950483, "t": 1660.46}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45604945333677305, "val_loss_b2": 0.4542124509928907, "val_loss_b4": 0.4544667600704097, "val_loss_b8": 0.45909418620282993, "t": 1707.16}45{"step": 2000, "loss": 0.4587438404560089, "lr": 0.0003392162016850945, "grad_norm": 0.1573607474565506, "tokens": 332550, "tok_per_s": 194.7768476269539, "t": 1707.95}46{"step": 2050, "loss": 0.520692765712738, "lr": 0.0003180007642703402, "grad_norm": 0.16873250901699066, "tokens": 340880, "tok_per_s": 195.20546020534974, "t": 1746.87}47{"step": 2100, "loss": 0.314407616853714, "lr": 0.000297465820586928, "grad_norm": 0.3143044710159302, "tokens": 349421, "tok_per_s": 195.74576625810982, "t": 1785.69}48{"step": 2150, "loss": 0.41941237449645996, "lr": 0.0002776716031628058, "grad_norm": 0.22679312527179718, "tokens": 357735, "tok_per_s": 196.15056410437285, "t": 1824.39}49{"step": 2200, "loss": 0.47400176525115967, "lr": 0.00025867617184817756, "grad_norm": 0.16610753536224365, "tokens": 366168, "tok_per_s": 196.52537461689533, "t": 1863.82}50{"step": 2250, "loss": 0.3774552643299103, "lr": 0.00024053524351595949, "grad_norm": 0.12167929857969284, "tokens": 374489, "tok_per_s": 196.9154064832074, "t": 1902.39}51{"step": 2300, "loss": 0.32181909680366516, "lr": 0.0002233020286345913, "grad_norm": 0.12349731475114822, "tokens": 382832, "tok_per_s": 197.3058478881978, "t": 1940.91}52{"step": 2350, "loss": 0.543024480342865, "lr": 0.00020702707519256364, "grad_norm": 0.1322794407606125, "tokens": 391097, "tok_per_s": 197.65496367235286, "t": 1979.3}53{"step": 2400, "loss": 0.2784748375415802, "lr": 0.00019175812043246033, "grad_norm": 0.25010865926742554, "tokens": 399648, "tok_per_s": 198.1177079667548, "t": 2017.84}54{"step": 2450, "loss": 0.3597331941127777, "lr": 0.00017753995082939931, "grad_norm": 0.13647907972335815, "tokens": 407914, "tok_per_s": 198.35343780099873, "t": 2057.11}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45384118732059453, "val_loss_b2": 0.451490054405517, "val_loss_b4": 0.4518919766624906, "val_loss_b8": 0.45715422557456675, "t": 2103.7}56{"step": 2500, "loss": 0.2583939731121063, "lr": 0.00016441427072458494, "grad_norm": 0.1782824546098709, "tokens": 416524, "tok_per_s": 197.98035661882295, "t": 2104.48}57{"step": 2550, "loss": 0.32008928060531616, "lr": 0.0001524195799992897, "grad_norm": 0.1557321846485138, "tokens": 424563, "tok_per_s": 198.1957744507611, "t": 2142.75}58{"step": 2600, "loss": 0.28543955087661743, "lr": 0.0001415910611480694, "grad_norm": 0.1554701328277588, "tokens": 432991, "tok_per_s": 198.58361934401847, "t": 2181.01}59{"step": 2650, "loss": 0.40035825967788696, "lr": 0.00013196047608244387, "grad_norm": 0.30165985226631165, "tokens": 441165, "tok_per_s": 198.81174794946463, "t": 2219.62}60{"step": 2700, "loss": 0.3471825122833252, "lr": 0.00012355607296773896, "grad_norm": 0.08197468519210815, "tokens": 449578, "tok_per_s": 199.13072878037008, "t": 2258.31}61{"step": 2750, "loss": 0.39044150710105896, "lr": 0.00011640250336634969, "grad_norm": 0.2220086306333542, "tokens": 457759, "tok_per_s": 199.32410620854412, "t": 2297.17}62{"step": 2800, "loss": 0.3730180263519287, "lr": 0.00011052074993046101, "grad_norm": 0.18050849437713623, "tokens": 465905, "tok_per_s": 199.57806640553716, "t": 2335.06}63{"step": 2850, "loss": 0.4111465811729431, "lr": 0.00010592806485631325, "grad_norm": 0.20909717679023743, "tokens": 474059, "tok_per_s": 199.78478109549698, "t": 2373.46}64{"step": 2900, "loss": 0.5412601232528687, "lr": 0.00010263791928053818, "grad_norm": 0.11476024985313416, "tokens": 482466, "tok_per_s": 200.10664464180218, "t": 2411.66}65{"step": 2950, "loss": 0.3797526955604553, "lr": 0.0001006599637669943, "grad_norm": 0.2450113296508789, "tokens": 490971, "tok_per_s": 200.46420714904679, "t": 2449.78}66{"step": 2999, "loss": 0.37283211946487427, "lr": 0.0001000002640500325, "grad_norm": 0.2713042199611664, "tokens": 499017, "tok_per_s": 200.65562202756848, "t": 2487.54}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4513307423824275, "val_loss_b2": 0.4491817329598507, "val_loss_b4": 0.45153857852038987, "val_loss_b8": 0.4599452180895328, "t": 2496.35}68{"step": 2999, "checkpoint_sha256": "ad0aff87189cbcd49c815e59f0adc835f1b576d111a7b0d60b7332142b4382da", "t": 2496.38}69