1{"step": 0, "loss": 0.42700591683387756, "lr": 1e-05, "grad_norm": 0.19014699757099152, "tokens": 139, "tok_per_s": 141.08927509211813, "t": 1.55}2{"step": 50, "loss": 0.3162084221839905, "lr": 0.00051, "grad_norm": 0.20671513676643372, "tokens": 8357, "tok_per_s": 204.13513898420368, "t": 41.5}3{"step": 100, "loss": 0.3213880658149719, "lr": 0.001, "grad_norm": 0.09552402794361115, "tokens": 16659, "tok_per_s": 208.167314384547, "t": 80.59}4{"step": 150, "loss": 0.45502933859825134, "lr": 0.0009993400362330058, "grad_norm": 0.06884507089853287, "tokens": 25169, "tok_per_s": 211.81369905231588, "t": 119.39}5{"step": 200, "loss": 0.4861842095851898, "lr": 0.0009973620807194618, "grad_norm": 0.10142555087804794, "tokens": 33572, "tok_per_s": 212.66389846759517, "t": 158.43}6{"step": 250, "loss": 0.4161961078643799, "lr": 0.000994071935143687, "grad_norm": 0.04243913292884827, "tokens": 41743, "tok_per_s": 212.08094123051873, "t": 197.39}7{"step": 300, "loss": 0.3963637351989746, "lr": 0.000989479250069539, "grad_norm": 0.023883970454335213, "tokens": 49922, "tok_per_s": 211.56933000428543, "t": 236.52}8{"step": 350, "loss": 0.3898469805717468, "lr": 0.0009835974966336503, "grad_norm": 0.06805329769849777, "tokens": 58442, "tok_per_s": 212.02469995944455, "t": 276.21}9{"step": 400, "loss": 0.3457052707672119, "lr": 0.0009764439270322612, "grad_norm": 0.08834843337535858, "tokens": 66780, "tok_per_s": 211.9288089449011, "t": 315.67}10{"step": 450, "loss": 0.48768770694732666, "lr": 0.0009680395239175562, "grad_norm": 0.06388355046510696, "tokens": 74983, "tok_per_s": 211.21237059204725, "t": 355.58}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4599487357329946, "val_loss_b2": 0.45813315792469156, "val_loss_b4": 0.4558413962885451, "val_loss_b8": 0.45609057742777337, "t": 404.81}12{"step": 500, "loss": 0.32736486196517944, "lr": 0.0009584089388519307, "grad_norm": 0.06014663353562355, "tokens": 83177, "tok_per_s": 205.3316740037942, "t": 405.65}13{"step": 550, "loss": 0.4096522927284241, "lr": 0.0009475804200007103, "grad_norm": 0.14391857385635376, "tokens": 91423, "tok_per_s": 205.43538916338025, "t": 445.59}14{"step": 600, "loss": 0.3794155418872833, "lr": 0.0009355857292754152, "grad_norm": 0.0798565074801445, "tokens": 99986, "tok_per_s": 206.34374166779835, "t": 485.12}15{"step": 650, "loss": 0.3514331579208374, "lr": 0.0009224600491706008, "grad_norm": 0.07307248562574387, "tokens": 108206, "tok_per_s": 206.10986238521465, "t": 525.56}16{"step": 700, "loss": 0.43887510895729065, "lr": 0.0009082418795675396, "grad_norm": 0.16070057451725006, "tokens": 116415, "tok_per_s": 206.3725409739687, "t": 564.66}17{"step": 750, "loss": 0.3402610719203949, "lr": 0.0008929729248074364, "grad_norm": 0.04802549630403519, "tokens": 124709, "tok_per_s": 206.50187067040375, "t": 604.48}18{"step": 800, "loss": 0.4268815517425537, "lr": 0.0008766979713654089, "grad_norm": 0.09154785424470901, "tokens": 132874, "tok_per_s": 206.5733235837465, "t": 643.79}19{"step": 850, "loss": 0.36020877957344055, "lr": 0.0008594647564840406, "grad_norm": 0.06821068376302719, "tokens": 141506, "tok_per_s": 207.19723778611174, "t": 683.52}20{"step": 900, "loss": 0.2954847514629364, "lr": 0.0008413238281518224, "grad_norm": 0.05820298567414284, "tokens": 149769, "tok_per_s": 207.5034497153204, "t": 722.33}21{"step": 950, "loss": 0.3571053445339203, "lr": 0.0008223283968371944, "grad_norm": 0.057510823011398315, "tokens": 158176, "tok_per_s": 207.69982560049277, "t": 762.12}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45754753515131663, "val_loss_b2": 0.4549290105707837, "val_loss_b4": 0.4524524636078257, "val_loss_b8": 0.4563225604678211, "t": 811.16}23{"step": 1000, "loss": 0.41362935304641724, "lr": 0.000802534179413072, "grad_norm": 0.0771273598074913, "tokens": 166573, "tok_per_s": 205.30446787739646, "t": 811.91}24{"step": 1050, "loss": 0.35846754908561707, "lr": 0.0007819992357296598, "grad_norm": 0.08417769521474838, "tokens": 174773, "tok_per_s": 205.41238180171163, "t": 851.4}25{"step": 1100, "loss": 0.4409148395061493, "lr": 0.0007607837983149056, "grad_norm": 0.09280560165643692, "tokens": 183200, "tok_per_s": 205.7743560020583, "t": 890.87}26{"step": 1150, "loss": 0.4675360918045044, "lr": 0.0007389500957021191, "grad_norm": 0.18736550211906433, "tokens": 191474, "tok_per_s": 205.9418632465054, "t": 930.31}27{"step": 1200, "loss": 0.39667487144470215, "lr": 0.0007165621699029615, "grad_norm": 0.1355508267879486, "tokens": 200061, "tok_per_s": 206.40740884137188, "t": 969.82}28{"step": 1250, "loss": 0.38612574338912964, "lr": 0.000693685688561191, "grad_norm": 0.10137220472097397, "tokens": 208455, "tok_per_s": 206.77078066125807, "t": 1008.71}29{"step": 1300, "loss": 0.3725875914096832, "lr": 0.0006703877523381495, "grad_norm": 0.18081337213516235, "tokens": 216814, "tok_per_s": 207.03418130852123, "t": 1047.8}30{"step": 1350, "loss": 0.3409733474254608, "lr": 0.0006467366980949609, "grad_norm": 0.11957166343927383, "tokens": 225136, "tok_per_s": 207.26141344165737, "t": 1086.81}31{"step": 1400, "loss": 0.3498188555240631, "lr": 0.0006228018984487442, "grad_norm": 0.06516694277524948, "tokens": 233471, "tok_per_s": 207.48507352015395, "t": 1125.81}32{"step": 1450, "loss": 0.30884382128715515, "lr": 0.0005986535582907738, "grad_norm": 0.07461652904748917, "tokens": 241795, "tok_per_s": 207.47327003185686, "t": 1165.99}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4564531701371375, "val_loss_b2": 0.4539141121493472, "val_loss_b4": 0.453822034870597, "val_loss_b8": 0.46919941044370783, "t": 1214.84}34{"step": 1500, "loss": 0.30781200528144836, "lr": 0.000574362508863438, "grad_norm": 0.08000863343477249, "tokens": 250085, "tok_per_s": 205.80751933561302, "t": 1215.7}35{"step": 1550, "loss": 0.3868008553981781, "lr": 0.00055, "grad_norm": 0.14513389766216278, "tokens": 258187, "tok_per_s": 205.80026858566174, "t": 1255.12}36{"step": 1600, "loss": 0.35159066319465637, "lr": 0.0005256374911365621, "grad_norm": 0.16477453708648682, "tokens": 266247, "tok_per_s": 205.69683851311868, "t": 1294.93}37{"step": 1650, "loss": 0.3101235628128052, "lr": 0.0005013464417092262, "grad_norm": 0.14392562210559845, "tokens": 274738, "tok_per_s": 206.00209059744554, "t": 1334.23}38{"step": 1700, "loss": 0.36903131008148193, "lr": 0.00047719810155125587, "grad_norm": 0.0779525563120842, "tokens": 283093, "tok_per_s": 206.1192170025354, "t": 1374.01}39{"step": 1750, "loss": 0.38403069972991943, "lr": 0.0004532633019050392, "grad_norm": 0.09295156598091125, "tokens": 291261, "tok_per_s": 206.84679504779712, "t": 1408.66}40{"step": 1800, "loss": 0.39940544962882996, "lr": 0.0004296122476618507, "grad_norm": 0.10168357938528061, "tokens": 299555, "tok_per_s": 208.32095984471098, "t": 1438.51}41{"step": 1850, "loss": 0.40904366970062256, "lr": 0.00040631431143880916, "grad_norm": 0.10414779186248779, "tokens": 307521, "tok_per_s": 209.4970241919546, "t": 1468.46}42{"step": 1900, "loss": 0.38660928606987, "lr": 0.00038343783009703845, "grad_norm": 0.08787985146045685, "tokens": 315892, "tok_per_s": 210.8574451086165, "t": 1498.7}43{"step": 1950, "loss": 0.3478291928768158, "lr": 0.000361049904297881, "grad_norm": 0.05510476231575012, "tokens": 324298, "tok_per_s": 212.11921235470095, "t": 1529.41}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.455852878922133, "val_loss_b2": 0.4536832769026364, "val_loss_b4": 0.45171031040061926, "val_loss_b8": 0.4598283875805219, "t": 1562.79}45{"step": 2000, "loss": 0.2844361960887909, "lr": 0.0003392162016850945, "grad_norm": 0.1366422474384308, "tokens": 332486, "tok_per_s": 212.76413658267825, "t": 1563.26}46{"step": 2050, "loss": 0.4516213834285736, "lr": 0.0003180007642703402, "grad_norm": 0.12291610985994339, "tokens": 340738, "tok_per_s": 214.9047807684758, "t": 1586.09}47{"step": 2100, "loss": 0.3686562478542328, "lr": 0.000297465820586928, "grad_norm": 0.12205690145492554, "tokens": 349148, "tok_per_s": 217.17580095127036, "t": 1608.24}48{"step": 2150, "loss": 0.34120872616767883, "lr": 0.0002776716031628058, "grad_norm": 0.1477447748184204, "tokens": 357576, "tok_per_s": 219.29036809350615, "t": 1631.17}49{"step": 2200, "loss": 0.3476885259151459, "lr": 0.00025867617184817756, "grad_norm": 0.20538027584552765, "tokens": 366097, "tok_per_s": 221.41487807942522, "t": 1654.01}50{"step": 2250, "loss": 0.27146413922309875, "lr": 0.00024053524351595949, "grad_norm": 0.11159098893404007, "tokens": 374285, "tok_per_s": 223.25918732799454, "t": 1677.02}51{"step": 2300, "loss": 0.3733380436897278, "lr": 0.0002233020286345913, "grad_norm": 0.10940872132778168, "tokens": 382864, "tok_per_s": 225.2656137953855, "t": 1700.17}52{"step": 2350, "loss": 0.449852854013443, "lr": 0.00020702707519256364, "grad_norm": 0.12116213142871857, "tokens": 391427, "tok_per_s": 227.2407725559828, "t": 1723.08}53{"step": 2400, "loss": 0.6381807923316956, "lr": 0.00019175812043246033, "grad_norm": 0.1937704086303711, "tokens": 399797, "tok_per_s": 228.98645908247713, "t": 1746.5}54{"step": 2450, "loss": 0.34651926159858704, "lr": 0.00017753995082939931, "grad_norm": 0.14424139261245728, "tokens": 407948, "tok_per_s": 230.79735387487898, "t": 1768.12}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4543318271401952, "val_loss_b2": 0.45045524021861233, "val_loss_b4": 0.4462455629186127, "val_loss_b8": 0.4503415362948911, "t": 1794.41}56{"step": 2500, "loss": 0.4002971351146698, "lr": 0.00016441427072458494, "grad_norm": 0.1266235113143921, "tokens": 416019, "tok_per_s": 231.86084681395099, "t": 1794.82}57{"step": 2550, "loss": 0.32214826345443726, "lr": 0.0001524195799992897, "grad_norm": 0.1003807932138443, "tokens": 424449, "tok_per_s": 233.796983474693, "t": 1816.02}58{"step": 2600, "loss": 0.3698647916316986, "lr": 0.0001415910611480694, "grad_norm": 0.15956762433052063, "tokens": 432494, "tok_per_s": 235.48537872789294, "t": 1837.17}59{"step": 2650, "loss": 0.2979693114757538, "lr": 0.00013196047608244387, "grad_norm": 0.13832715153694153, "tokens": 440693, "tok_per_s": 237.17346792383154, "t": 1858.67}60{"step": 2700, "loss": 0.4025195837020874, "lr": 0.00012355607296773896, "grad_norm": 0.1842956691980362, "tokens": 449079, "tok_per_s": 238.96218072680782, "t": 1879.85}61{"step": 2750, "loss": 0.3038519024848938, "lr": 0.00011640250336634969, "grad_norm": 0.09114228934049606, "tokens": 457686, "tok_per_s": 240.78215603394227, "t": 1901.39}62{"step": 2800, "loss": 0.42162710428237915, "lr": 0.00011052074993046101, "grad_norm": 0.13537965714931488, "tokens": 466220, "tok_per_s": 242.5626722305205, "t": 1922.62}63{"step": 2850, "loss": 0.46892648935317993, "lr": 0.00010592806485631325, "grad_norm": 0.21180543303489685, "tokens": 474198, "tok_per_s": 244.21748487511235, "t": 1942.27}64{"step": 2900, "loss": 0.4419267475605011, "lr": 0.00010263791928053818, "grad_norm": 0.21288521587848663, "tokens": 482297, "tok_per_s": 245.85104327637006, "t": 1962.31}65{"step": 2950, "loss": 0.4069972634315491, "lr": 0.0001006599637669943, "grad_norm": 0.16384905576705933, "tokens": 490516, "tok_per_s": 247.51407611214606, "t": 1982.33}66{"step": 2999, "loss": 0.51930832862854, "lr": 0.0001000002640500325, "grad_norm": 0.2191876620054245, "tokens": 498543, "tok_per_s": 249.16098865562157, "t": 2001.45}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4530690843216476, "val_loss_b2": 0.4499849119228703, "val_loss_b4": 0.4483420483404454, "val_loss_b8": 0.4592861722642533, "t": 2006.56}68{"step": 2999, "checkpoint_sha256": "72ce074de8f72417f5b824c0aee5ed437b7e6ec162c8c973b3aa382f6447bd1f", "t": 2006.59}69