1{"step": 0, "loss": 0.43779659271240234, "lr": 1e-05, "grad_norm": 0.3840331435203552, "tokens": 146, "tok_per_s": 274.5289456046875, "t": 1.14}2{"step": 50, "loss": 0.33513498306274414, "lr": 0.00051, "grad_norm": 0.1813535839319229, "tokens": 8382, "tok_per_s": 334.59064962784765, "t": 25.66}3{"step": 100, "loss": 0.45694223046302795, "lr": 0.001, "grad_norm": 0.17014357447624207, "tokens": 16548, "tok_per_s": 326.22448775835807, "t": 51.34}4{"step": 150, "loss": 0.31460925936698914, "lr": 0.0009993400362330058, "grad_norm": 0.06550630927085876, "tokens": 24812, "tok_per_s": 323.3110140948488, "t": 77.36}5{"step": 200, "loss": 0.4250691533088684, "lr": 0.0009973620807194618, "grad_norm": 0.07026832550764084, "tokens": 33400, "tok_per_s": 324.13440915296286, "t": 103.66}6{"step": 250, "loss": 0.4035921096801758, "lr": 0.000994071935143687, "grad_norm": 0.10469461232423782, "tokens": 41583, "tok_per_s": 327.11656137967424, "t": 127.73}7{"step": 300, "loss": 0.34740641713142395, "lr": 0.000989479250069539, "grad_norm": 0.044001031666994095, "tokens": 49989, "tok_per_s": 337.8221260125827, "t": 148.59}8{"step": 350, "loss": 0.45567741990089417, "lr": 0.0009835974966336503, "grad_norm": 0.09509779512882233, "tokens": 58214, "tok_per_s": 349.9748060356412, "t": 166.95}9{"step": 400, "loss": 0.4068395793437958, "lr": 0.0009764439270322612, "grad_norm": 0.08176733553409576, "tokens": 66391, "tok_per_s": 358.71227799991755, "t": 185.69}10{"step": 450, "loss": 0.2987080216407776, "lr": 0.0009680395239175562, "grad_norm": 0.03786255419254303, "tokens": 74699, "tok_per_s": 366.68359234265995, "t": 204.33}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4608102462866319, "val_loss_b2": 0.459550321719046, "val_loss_b4": 0.4578858587529167, "val_loss_b8": 0.4581340575464962, "t": 226.56}12{"step": 500, "loss": 0.5788761973381042, "lr": 0.0009584089388519307, "grad_norm": 0.049350131303071976, "tokens": 83038, "tok_per_s": 367.0297363091709, "t": 226.86}13{"step": 550, "loss": 0.3949809968471527, "lr": 0.0009475804200007103, "grad_norm": 0.046304769814014435, "tokens": 91441, "tok_per_s": 377.2196461763008, "t": 243.02}14{"step": 600, "loss": 0.4798583984375, "lr": 0.0009355857292754152, "grad_norm": 0.05259697511792183, "tokens": 99553, "tok_per_s": 383.1011970578553, "t": 260.47}15{"step": 650, "loss": 0.42109909653663635, "lr": 0.0009224600491706008, "grad_norm": 0.22771432995796204, "tokens": 107666, "tok_per_s": 386.1848923552643, "t": 279.41}16{"step": 700, "loss": 0.4334016740322113, "lr": 0.0009082418795675396, "grad_norm": 0.05693872645497322, "tokens": 115959, "tok_per_s": 388.94430294631206, "t": 298.75}17{"step": 750, "loss": 0.4061746895313263, "lr": 0.0008929729248074364, "grad_norm": 0.07976570725440979, "tokens": 124302, "tok_per_s": 392.02558118165956, "t": 317.69}18{"step": 800, "loss": 0.48092207312583923, "lr": 0.0008766979713654089, "grad_norm": 0.07909360527992249, "tokens": 132734, "tok_per_s": 398.20234700145204, "t": 333.95}19{"step": 850, "loss": 0.49930500984191895, "lr": 0.0008594647564840406, "grad_norm": 0.11126729100942612, "tokens": 141025, "tok_per_s": 402.6385212826339, "t": 350.86}20{"step": 900, "loss": 0.46097785234451294, "lr": 0.0008413238281518224, "grad_norm": 0.13809849321842194, "tokens": 149035, "tok_per_s": 407.04164249631145, "t": 366.75}21{"step": 950, "loss": 0.40798768401145935, "lr": 0.0008223283968371944, "grad_norm": 0.09210410714149475, "tokens": 157135, "tok_per_s": 411.43902686393767, "t": 382.53}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45761174107490943, "val_loss_b2": 0.4550720438574147, "val_loss_b4": 0.45343953579620166, "val_loss_b8": 0.45511318110902654, "t": 403.19}23{"step": 1000, "loss": 0.4275533854961395, "lr": 0.000802534179413072, "grad_norm": 0.09284070134162903, "tokens": 165602, "tok_per_s": 410.90871253799554, "t": 403.63}24{"step": 1050, "loss": 0.36501625180244446, "lr": 0.0007819992357296598, "grad_norm": 0.14007098972797394, "tokens": 173674, "tok_per_s": 413.07880263259204, "t": 421.05}25{"step": 1100, "loss": 0.37572839856147766, "lr": 0.0007607837983149056, "grad_norm": 0.05231399089097977, "tokens": 181834, "tok_per_s": 415.2825693269543, "t": 438.47}26{"step": 1150, "loss": 0.4159546494483948, "lr": 0.0007389500957021191, "grad_norm": 0.08069592714309692, "tokens": 190158, "tok_per_s": 418.48279718261034, "t": 455.01}27{"step": 1200, "loss": 0.399723082780838, "lr": 0.0007165621699029615, "grad_norm": 0.10650932788848877, "tokens": 198573, "tok_per_s": 421.1865625634561, "t": 472.08}28{"step": 1250, "loss": 0.4725813865661621, "lr": 0.000693685688561191, "grad_norm": 0.11986038088798523, "tokens": 206880, "tok_per_s": 424.0328645589352, "t": 488.5}29{"step": 1300, "loss": 0.32328495383262634, "lr": 0.0006703877523381495, "grad_norm": 0.16857141256332397, "tokens": 215095, "tok_per_s": 426.09098345444903, "t": 505.42}30{"step": 1350, "loss": 0.39778172969818115, "lr": 0.0006467366980949609, "grad_norm": 0.13652725517749786, "tokens": 223349, "tok_per_s": 428.21529284087615, "t": 522.19}31{"step": 1400, "loss": 0.4366261661052704, "lr": 0.0006228018984487442, "grad_norm": 0.1033504456281662, "tokens": 231716, "tok_per_s": 430.79146747840684, "t": 538.5}32{"step": 1450, "loss": 0.4944872558116913, "lr": 0.0005986535582907738, "grad_norm": 0.12253924459218979, "tokens": 239960, "tok_per_s": 432.9219803428239, "t": 554.89}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45505173139351235, "val_loss_b2": 0.45201983294385945, "val_loss_b4": 0.4509287806436192, "val_loss_b8": 0.45961170469145873, "t": 575.04}34{"step": 1500, "loss": 0.44044923782348633, "lr": 0.000574362508863438, "grad_norm": 0.15195706486701965, "tokens": 248357, "tok_per_s": 432.14555004516905, "t": 575.32}35{"step": 1550, "loss": 0.5652903914451599, "lr": 0.00055, "grad_norm": 0.24265924096107483, "tokens": 256793, "tok_per_s": 433.84788495448146, "t": 592.51}36{"step": 1600, "loss": 0.4927304685115814, "lr": 0.0005256374911365621, "grad_norm": 0.17728808522224426, "tokens": 265023, "tok_per_s": 435.1714576708198, "t": 609.62}37{"step": 1650, "loss": 0.519521176815033, "lr": 0.0005013464417092262, "grad_norm": 0.16842544078826904, "tokens": 273283, "tok_per_s": 437.32557721268887, "t": 625.51}38{"step": 1700, "loss": 0.47538238763809204, "lr": 0.00047719810155125587, "grad_norm": 0.2132110595703125, "tokens": 281671, "tok_per_s": 439.9578789322084, "t": 640.84}39{"step": 1750, "loss": 0.407649964094162, "lr": 0.0004532633019050392, "grad_norm": 0.11514782160520554, "tokens": 289799, "tok_per_s": 440.9577781457429, "t": 657.82}40{"step": 1800, "loss": 0.3596852719783783, "lr": 0.0004296122476618507, "grad_norm": 0.11727385967969894, "tokens": 298140, "tok_per_s": 442.1674917021307, "t": 674.88}41{"step": 1850, "loss": 0.4693586528301239, "lr": 0.00040631431143880916, "grad_norm": 0.2084578573703766, "tokens": 306518, "tok_per_s": 444.2693058894364, "t": 690.55}42{"step": 1900, "loss": 0.3544922173023224, "lr": 0.00038343783009703845, "grad_norm": 0.09360159188508987, "tokens": 314885, "tok_per_s": 445.0234871259929, "t": 708.18}43{"step": 1950, "loss": 0.31849583983421326, "lr": 0.000361049904297881, "grad_norm": 0.1614905148744583, "tokens": 322988, "tok_per_s": 442.21149915246247, "t": 731.0}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4547543166129092, "val_loss_b2": 0.45134746763963546, "val_loss_b4": 0.45024871920176246, "val_loss_b8": 0.45466716422301434, "t": 752.78}45{"step": 2000, "loss": 0.4919474124908447, "lr": 0.0003392162016850945, "grad_norm": 0.19673824310302734, "tokens": 331190, "tok_per_s": 439.98075860497977, "t": 753.35}46{"step": 2050, "loss": 0.39737120270729065, "lr": 0.0003180007642703402, "grad_norm": 0.14760281145572662, "tokens": 339365, "tok_per_s": 440.11716961871787, "t": 771.69}47{"step": 2100, "loss": 0.3537965416908264, "lr": 0.000297465820586928, "grad_norm": 0.13771742582321167, "tokens": 347821, "tok_per_s": 440.0906499797637, "t": 790.95}48{"step": 2150, "loss": 0.4854559004306793, "lr": 0.0002776716031628058, "grad_norm": 0.11919410526752472, "tokens": 355802, "tok_per_s": 440.1833271967434, "t": 808.92}49{"step": 2200, "loss": 0.3459104299545288, "lr": 0.00025867617184817756, "grad_norm": 0.16386087238788605, "tokens": 363932, "tok_per_s": 441.02331138402127, "t": 825.81}50{"step": 2250, "loss": 0.29812291264533997, "lr": 0.00024053524351595949, "grad_norm": 0.10713409632444382, "tokens": 372235, "tok_per_s": 440.6885164820033, "t": 845.28}51{"step": 2300, "loss": 0.3709394335746765, "lr": 0.0002233020286345913, "grad_norm": 0.13289542496204376, "tokens": 380617, "tok_per_s": 441.2056338526665, "t": 863.29}52{"step": 2350, "loss": 0.35798755288124084, "lr": 0.00020702707519256364, "grad_norm": 0.1917424499988556, "tokens": 388820, "tok_per_s": 441.3021273484676, "t": 881.69}53{"step": 2400, "loss": 0.3608905076980591, "lr": 0.00019175812043246033, "grad_norm": 0.28338897228240967, "tokens": 397013, "tok_per_s": 441.2851998842173, "t": 900.29}54{"step": 2450, "loss": 0.404866486787796, "lr": 0.00017753995082939931, "grad_norm": 0.20190341770648956, "tokens": 405444, "tok_per_s": 440.90438602141285, "t": 920.19}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4525752135719085, "val_loss_b2": 0.4489051185258214, "val_loss_b4": 0.44804231541212, "val_loss_b8": 0.45578558965525995, "t": 942.78}56{"step": 2500, "loss": 0.4608103036880493, "lr": 0.00016441427072458494, "grad_norm": 0.18037889897823334, "tokens": 413920, "tok_per_s": 439.1103415630656, "t": 943.25}57{"step": 2550, "loss": 0.32784488797187805, "lr": 0.0001524195799992897, "grad_norm": 0.1584175080060959, "tokens": 422230, "tok_per_s": 439.4567247974956, "t": 961.41}58{"step": 2600, "loss": 0.40597325563430786, "lr": 0.0001415910611480694, "grad_norm": 0.17738477885723114, "tokens": 430404, "tok_per_s": 439.6258036630286, "t": 979.64}59{"step": 2650, "loss": 0.3840937912464142, "lr": 0.00013196047608244387, "grad_norm": 0.20046763122081757, "tokens": 438498, "tok_per_s": 439.44202172949645, "t": 998.46}60{"step": 2700, "loss": 0.33217403292655945, "lr": 0.00012355607296773896, "grad_norm": 0.1853371560573578, "tokens": 446658, "tok_per_s": 439.8398126013153, "t": 1016.11}61{"step": 2750, "loss": 0.35564398765563965, "lr": 0.00011640250336634969, "grad_norm": 0.17339706420898438, "tokens": 455155, "tok_per_s": 440.03475630642447, "t": 1034.98}62{"step": 2800, "loss": 0.43018004298210144, "lr": 0.00011052074993046101, "grad_norm": 0.20705685019493103, "tokens": 463083, "tok_per_s": 439.96410169273696, "t": 1053.16}63{"step": 2850, "loss": 0.3808097839355469, "lr": 0.00010592806485631325, "grad_norm": 0.1915445774793625, "tokens": 471665, "tok_per_s": 440.45188232844157, "t": 1071.48}64{"step": 2900, "loss": 0.4274345338344574, "lr": 0.00010263791928053818, "grad_norm": 0.20706531405448914, "tokens": 480193, "tok_per_s": 440.5729602058125, "t": 1090.54}65{"step": 2950, "loss": 0.3146987855434418, "lr": 0.0001006599637669943, "grad_norm": 0.18104468286037445, "tokens": 488284, "tok_per_s": 440.3070418147245, "t": 1109.58}66{"step": 2999, "loss": 0.44395455718040466, "lr": 0.0001000002640500325, "grad_norm": 0.2267700880765915, "tokens": 496298, "tok_per_s": 440.26020693796875, "t": 1127.9}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45191332599457523, "val_loss_b2": 0.4481894040471994, "val_loss_b4": 0.4480794067157202, "val_loss_b8": 0.457967686735391, "t": 1132.1}68{"step": 2999, "checkpoint_sha256": "074a3c42cc84c5ed2754c1ff7f7e13d4808de2d7e2d8bd7094fe8fb0f816236d", "t": 1132.32}69