1{"step": 0, "loss": 0.4286305606365204, "lr": 1e-05, "grad_norm": 0.288160115480423, "tokens": 139, "tok_per_s": 94.72162002545912, "t": 1.96}2{"step": 50, "loss": 0.31785115599632263, "lr": 0.00051, "grad_norm": 0.24622443318367004, "tokens": 8357, "tok_per_s": 145.87070353022366, "t": 57.78}3{"step": 100, "loss": 0.32056716084480286, "lr": 0.001, "grad_norm": 0.08183470368385315, "tokens": 16659, "tok_per_s": 147.7045859928238, "t": 113.29}4{"step": 150, "loss": 0.45618218183517456, "lr": 0.0009993400362330058, "grad_norm": 0.08231659978628159, "tokens": 25169, "tok_per_s": 141.92870730321235, "t": 177.83}5{"step": 200, "loss": 0.48482072353363037, "lr": 0.0009973620807194618, "grad_norm": 0.13018696010112762, "tokens": 33572, "tok_per_s": 133.6781303734516, "t": 251.63}6{"step": 250, "loss": 0.4200207591056824, "lr": 0.000994071935143687, "grad_norm": 0.0899510532617569, "tokens": 41743, "tok_per_s": 126.13482512194747, "t": 331.43}7{"step": 300, "loss": 0.39745840430259705, "lr": 0.000989479250069539, "grad_norm": 0.0366491973400116, "tokens": 49922, "tok_per_s": 123.41107103115928, "t": 405.01}8{"step": 350, "loss": 0.3859606087207794, "lr": 0.0009835974966336503, "grad_norm": 0.061705101281404495, "tokens": 58442, "tok_per_s": 129.52506133103648, "t": 451.7}9{"step": 400, "loss": 0.34935474395751953, "lr": 0.0009764439270322612, "grad_norm": 0.12967334687709808, "tokens": 66780, "tok_per_s": 130.8715547955623, "t": 510.78}10{"step": 450, "loss": 0.4856657087802887, "lr": 0.0009680395239175562, "grad_norm": 0.09350696951150894, "tokens": 74983, "tok_per_s": 132.014199510683, "t": 568.49}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45991415246594064, "val_loss_b2": 0.4582137582101159, "val_loss_b4": 0.4560440452654647, "val_loss_b8": 0.45594245192451155, "t": 631.15}12{"step": 500, "loss": 0.3307649791240692, "lr": 0.0009584089388519307, "grad_norm": 0.0917133241891861, "tokens": 83177, "tok_per_s": 131.58867037186593, "t": 632.59}13{"step": 550, "loss": 0.4043864607810974, "lr": 0.0009475804200007103, "grad_norm": 0.14115306735038757, "tokens": 91423, "tok_per_s": 77.83365324046106, "t": 1175.09}14{"step": 600, "loss": 0.3790096640586853, "lr": 0.0009355857292754152, "grad_norm": 0.08601750433444977, "tokens": 99986, "tok_per_s": 82.69865304528108, "t": 1209.53}15{"step": 650, "loss": 0.35874462127685547, "lr": 0.0009224600491706008, "grad_norm": 0.12288975715637207, "tokens": 108206, "tok_per_s": 86.82476758426246, "t": 1246.75}16{"step": 700, "loss": 0.4448688328266144, "lr": 0.0009082418795675396, "grad_norm": 0.3965795040130615, "tokens": 116415, "tok_per_s": 90.9495282729773, "t": 1280.49}17{"step": 750, "loss": 0.33566004037857056, "lr": 0.0008929729248074364, "grad_norm": 0.06856173276901245, "tokens": 124709, "tok_per_s": 94.54588942503985, "t": 1319.52}18{"step": 800, "loss": 0.4205099642276764, "lr": 0.0008766979713654089, "grad_norm": 0.10444746911525726, "tokens": 132874, "tok_per_s": 98.03745837164875, "t": 1355.83}19{"step": 850, "loss": 0.36364489793777466, "lr": 0.0008594647564840406, "grad_norm": 0.07041320949792862, "tokens": 141506, "tok_per_s": 101.87042160111977, "t": 1389.57}20{"step": 900, "loss": 0.29098373651504517, "lr": 0.0008413238281518224, "grad_norm": 0.09163156151771545, "tokens": 149769, "tok_per_s": 104.3310769081378, "t": 1436.01}21{"step": 950, "loss": 0.35907742381095886, "lr": 0.0008223283968371944, "grad_norm": 0.10033963620662689, "tokens": 158176, "tok_per_s": 107.18496891965025, "t": 1476.22}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4581595543513103, "val_loss_b2": 0.45552558015284716, "val_loss_b4": 0.4524796186496966, "val_loss_b8": 0.4547659468333553, "t": 1560.04}23{"step": 1000, "loss": 0.4160661995410919, "lr": 0.000802534179413072, "grad_norm": 0.0619177520275116, "tokens": 166573, "tok_per_s": 106.65960043220042, "t": 1562.25}24{"step": 1050, "loss": 0.35879725217819214, "lr": 0.0007819992357296598, "grad_norm": 0.10772410780191422, "tokens": 174773, "tok_per_s": 108.70137714271503, "t": 1608.32}25{"step": 1100, "loss": 0.4414120018482208, "lr": 0.0007607837983149056, "grad_norm": 0.1300262063741684, "tokens": 183200, "tok_per_s": 110.14006252885721, "t": 1663.83}26{"step": 1150, "loss": 0.46340739727020264, "lr": 0.0007389500957021191, "grad_norm": 0.17452389001846313, "tokens": 191474, "tok_per_s": 111.96154281751001, "t": 1710.67}27{"step": 1200, "loss": 0.3961964547634125, "lr": 0.0007165621699029615, "grad_norm": 0.10654351860284805, "tokens": 200061, "tok_per_s": 114.59792492604815, "t": 1746.26}28{"step": 1250, "loss": 0.3900180160999298, "lr": 0.000693685688561191, "grad_norm": 0.14469176530838013, "tokens": 208455, "tok_per_s": 117.34871554379043, "t": 1776.87}29{"step": 1300, "loss": 0.37116897106170654, "lr": 0.0006703877523381495, "grad_norm": 0.24014164507389069, "tokens": 216814, "tok_per_s": 117.2966254575157, "t": 1848.92}30{"step": 1350, "loss": 0.33863919973373413, "lr": 0.0006467366980949609, "grad_norm": 0.15197820961475372, "tokens": 225136, "tok_per_s": 119.95532526067531, "t": 1877.33}31{"step": 1400, "loss": 0.3476954698562622, "lr": 0.0006228018984487442, "grad_norm": 0.11041784286499023, "tokens": 233471, "tok_per_s": 122.26656740405103, "t": 1910.02}32{"step": 1450, "loss": 0.30810168385505676, "lr": 0.0005986535582907738, "grad_norm": 0.10357106477022171, "tokens": 241795, "tok_per_s": 124.73678749090243, "t": 1938.94}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45620867229553796, "val_loss_b2": 0.4522371954845054, "val_loss_b4": 0.4492087667360865, "val_loss_b8": 0.4566832965909113, "t": 1975.63}34{"step": 1500, "loss": 0.3126918375492096, "lr": 0.000574362508863438, "grad_norm": 0.0881507620215416, "tokens": 250085, "tok_per_s": 126.5485310347852, "t": 1976.69}35{"step": 1550, "loss": 0.3849959969520569, "lr": 0.00055, "grad_norm": 0.10843630135059357, "tokens": 258187, "tok_per_s": 128.6062844801132, "t": 2008.07}36{"step": 1600, "loss": 0.35646525025367737, "lr": 0.0005256374911365621, "grad_norm": 0.2156679779291153, "tokens": 266247, "tok_per_s": 130.00128151232977, "t": 2048.53}37{"step": 1650, "loss": 0.3089267313480377, "lr": 0.0005013464417092262, "grad_norm": 0.1579325646162033, "tokens": 274738, "tok_per_s": 131.49304788213192, "t": 2089.87}38{"step": 1700, "loss": 0.3701803982257843, "lr": 0.00047719810155125587, "grad_norm": 0.10877073556184769, "tokens": 283093, "tok_per_s": 132.78395707843976, "t": 2132.48}39{"step": 1750, "loss": 0.38600650429725647, "lr": 0.0004532633019050392, "grad_norm": 0.10775183141231537, "tokens": 291261, "tok_per_s": 133.86943745029913, "t": 2176.2}40{"step": 1800, "loss": 0.40248408913612366, "lr": 0.0004296122476618507, "grad_norm": 0.12438409775495529, "tokens": 299555, "tok_per_s": 135.00679515398178, "t": 2219.31}41{"step": 1850, "loss": 0.41069966554641724, "lr": 0.00040631431143880916, "grad_norm": 0.13009047508239746, "tokens": 307521, "tok_per_s": 135.96851989197847, "t": 2262.2}42{"step": 1900, "loss": 0.3870050013065338, "lr": 0.00038343783009703845, "grad_norm": 0.08344994485378265, "tokens": 315892, "tok_per_s": 137.00142867290808, "t": 2306.26}43{"step": 1950, "loss": 0.3489689826965332, "lr": 0.000361049904297881, "grad_norm": 0.059593744575977325, "tokens": 324298, "tok_per_s": 137.81950227580688, "t": 2353.56}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45593262564789316, "val_loss_b2": 0.4525621638855011, "val_loss_b4": 0.4510052832786293, "val_loss_b8": 0.4583740911206452, "t": 2405.07}45{"step": 2000, "loss": 0.2902865707874298, "lr": 0.0003392162016850945, "grad_norm": 0.18714694678783417, "tokens": 332486, "tok_per_s": 138.23133257303078, "t": 2405.78}46{"step": 2050, "loss": 0.4528408944606781, "lr": 0.0003180007642703402, "grad_norm": 0.11507721245288849, "tokens": 340738, "tok_per_s": 139.04673282641986, "t": 2451.03}47{"step": 2100, "loss": 0.36746320128440857, "lr": 0.000297465820586928, "grad_norm": 0.1850506067276001, "tokens": 349148, "tok_per_s": 139.97205769759609, "t": 2494.91}48{"step": 2150, "loss": 0.3373045325279236, "lr": 0.0002776716031628058, "grad_norm": 0.14643891155719757, "tokens": 357576, "tok_per_s": 140.74599520819706, "t": 2541.07}49{"step": 2200, "loss": 0.34844645857810974, "lr": 0.00025867617184817756, "grad_norm": 0.27222228050231934, "tokens": 366097, "tok_per_s": 141.66013454604274, "t": 2584.83}50{"step": 2250, "loss": 0.2790728509426117, "lr": 0.00024053524351595949, "grad_norm": 0.19579842686653137, "tokens": 374285, "tok_per_s": 142.27873411348546, "t": 2631.14}51{"step": 2300, "loss": 0.372869610786438, "lr": 0.0002233020286345913, "grad_norm": 0.11416583508253098, "tokens": 382864, "tok_per_s": 143.292404247305, "t": 2672.4}52{"step": 2350, "loss": 0.45111897587776184, "lr": 0.00020702707519256364, "grad_norm": 0.17036190629005432, "tokens": 391427, "tok_per_s": 144.24364799541803, "t": 2714.15}53{"step": 2400, "loss": 0.6300070285797119, "lr": 0.00019175812043246033, "grad_norm": 0.21115869283676147, "tokens": 399797, "tok_per_s": 145.0544630595018, "t": 2756.68}54{"step": 2450, "loss": 0.343777060508728, "lr": 0.00017753995082939931, "grad_norm": 0.15217843651771545, "tokens": 407948, "tok_per_s": 145.7852316261967, "t": 2798.77}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45470205112417794, "val_loss_b2": 0.4504194010531979, "val_loss_b4": 0.44686487403745545, "val_loss_b8": 0.4499050138500772, "t": 2848.57}56{"step": 2500, "loss": 0.4002193212509155, "lr": 0.00016441427072458494, "grad_norm": 0.15669584274291992, "tokens": 416019, "tok_per_s": 146.0267752641302, "t": 2849.42}57{"step": 2550, "loss": 0.32234424352645874, "lr": 0.0001524195799992897, "grad_norm": 0.13005861639976501, "tokens": 424449, "tok_per_s": 146.80700892986331, "t": 2891.7}58{"step": 2600, "loss": 0.3704676330089569, "lr": 0.0001415910611480694, "grad_norm": 0.14601382613182068, "tokens": 432494, "tok_per_s": 147.31553931201495, "t": 2936.33}59{"step": 2650, "loss": 0.30996987223625183, "lr": 0.00013196047608244387, "grad_norm": 0.22016775608062744, "tokens": 440693, "tok_per_s": 147.91851704674994, "t": 2979.79}60{"step": 2700, "loss": 0.3974631130695343, "lr": 0.00012355607296773896, "grad_norm": 0.21548251807689667, "tokens": 449079, "tok_per_s": 148.62311122219265, "t": 3022.09}61{"step": 2750, "loss": 0.30715763568878174, "lr": 0.00011640250336634969, "grad_norm": 0.09812279045581818, "tokens": 457686, "tok_per_s": 149.6214398130572, "t": 3059.45}62{"step": 2800, "loss": 0.42147737741470337, "lr": 0.00011052074993046101, "grad_norm": 0.17805013060569763, "tokens": 466220, "tok_per_s": 151.22004554229915, "t": 3083.55}63{"step": 2850, "loss": 0.463235080242157, "lr": 0.00010592806485631325, "grad_norm": 0.21063832938671112, "tokens": 474198, "tok_per_s": 152.6009057917254, "t": 3107.93}64{"step": 2900, "loss": 0.4436056911945343, "lr": 0.00010263791928053818, "grad_norm": 0.2252541184425354, "tokens": 482297, "tok_per_s": 153.94081004899508, "t": 3133.5}65{"step": 2950, "loss": 0.40293774008750916, "lr": 0.0001006599637669943, "grad_norm": 0.17939738929271698, "tokens": 490516, "tok_per_s": 155.33269201120885, "t": 3158.33}66{"step": 2999, "loss": 0.5168623328208923, "lr": 0.0001000002640500325, "grad_norm": 0.2487969845533371, "tokens": 498543, "tok_per_s": 156.69806786531154, "t": 3182.05}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4532265388184195, "val_loss_b2": 0.44923476928440204, "val_loss_b4": 0.447293094016461, "val_loss_b8": 0.45502060560712443, "t": 3186.53}68{"step": 2999, "checkpoint_sha256": "9fbfe11629631d608c632ed91f243f12c89be7bfcad52271fb427691636b0f3e", "t": 3186.58}69