1{"step": 0, "loss": 0.4435354769229889, "lr": 1e-05, "grad_norm": 0.19450289011001587, "tokens": 146, "tok_per_s": 100.71043400149264, "t": 2.12}2{"step": 50, "loss": 0.3300611674785614, "lr": 0.00051, "grad_norm": 0.11406811326742172, "tokens": 8382, "tok_per_s": 145.67824974156227, "t": 58.2}3{"step": 100, "loss": 0.45315179228782654, "lr": 0.001, "grad_norm": 0.09778544306755066, "tokens": 16548, "tok_per_s": 146.77531101438035, "t": 113.4}4{"step": 150, "loss": 0.3139650523662567, "lr": 0.0009993400362330058, "grad_norm": 0.07414384931325912, "tokens": 24812, "tok_per_s": 147.9820018152166, "t": 168.33}5{"step": 200, "loss": 0.4258403480052948, "lr": 0.0009973620807194618, "grad_norm": 0.07282572239637375, "tokens": 33400, "tok_per_s": 150.43501541413957, "t": 222.68}6{"step": 250, "loss": 0.4037013351917267, "lr": 0.000994071935143687, "grad_norm": 0.12080949544906616, "tokens": 41583, "tok_per_s": 150.93831894680926, "t": 276.15}7{"step": 300, "loss": 0.3474419116973877, "lr": 0.000989479250069539, "grad_norm": 0.048010967671871185, "tokens": 49989, "tok_per_s": 152.2585012543925, "t": 328.97}8{"step": 350, "loss": 0.4580291509628296, "lr": 0.0009835974966336503, "grad_norm": 0.09400822222232819, "tokens": 58214, "tok_per_s": 152.01029630151285, "t": 383.62}9{"step": 400, "loss": 0.4057029187679291, "lr": 0.0009764439270322612, "grad_norm": 0.1198025569319725, "tokens": 66391, "tok_per_s": 151.5622178472651, "t": 438.7}10{"step": 450, "loss": 0.2996162176132202, "lr": 0.0009680395239175562, "grad_norm": 0.030202234163880348, "tokens": 74699, "tok_per_s": 151.5270523681939, "t": 493.63}11{"step": 499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4604895983264034, "val_loss_b2": 0.45880243330532955, "val_loss_b4": 0.45674417852004096, "val_loss_b8": 0.4559993001261975, "t": 559.6}12{"step": 500, "loss": 0.5803603529930115, "lr": 0.0009584089388519307, "grad_norm": 0.05278917774558067, "tokens": 83038, "tok_per_s": 148.2668689996185, "t": 560.71}13{"step": 550, "loss": 0.3950493633747101, "lr": 0.0009475804200007103, "grad_norm": 0.03404582664370537, "tokens": 91441, "tok_per_s": 148.9147894097089, "t": 614.71}14{"step": 600, "loss": 0.475232869386673, "lr": 0.0009355857292754152, "grad_norm": 0.1487923264503479, "tokens": 99553, "tok_per_s": 148.95779821102258, "t": 668.98}15{"step": 650, "loss": 0.4158070981502533, "lr": 0.0009224600491706008, "grad_norm": 0.17912742495536804, "tokens": 107666, "tok_per_s": 149.06493139283754, "t": 722.94}16{"step": 700, "loss": 0.4407401382923126, "lr": 0.0009082418795675396, "grad_norm": 0.08581017702817917, "tokens": 115959, "tok_per_s": 149.3301792351384, "t": 777.19}17{"step": 750, "loss": 0.3968702256679535, "lr": 0.0008929729248074364, "grad_norm": 0.07096485048532486, "tokens": 124302, "tok_per_s": 149.3960236711823, "t": 832.69}18{"step": 800, "loss": 0.4942583739757538, "lr": 0.0008766979713654089, "grad_norm": 0.10351559519767761, "tokens": 132734, "tok_per_s": 149.66710859355942, "t": 887.52}19{"step": 850, "loss": 0.49691328406333923, "lr": 0.0008594647564840406, "grad_norm": 0.07771594822406769, "tokens": 141025, "tok_per_s": 149.85867209263233, "t": 941.71}20{"step": 900, "loss": 0.46251967549324036, "lr": 0.0008413238281518224, "grad_norm": 0.13502085208892822, "tokens": 149035, "tok_per_s": 149.75331982128642, "t": 995.86}21{"step": 950, "loss": 0.40481188893318176, "lr": 0.0008223283968371944, "grad_norm": 0.08101046830415726, "tokens": 157135, "tok_per_s": 149.6001639629448, "t": 1051.02}22{"step": 999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45762825482228875, "val_loss_b2": 0.45453002783040686, "val_loss_b4": 0.4508339424565721, "val_loss_b8": 0.45002922746217444, "t": 1117.13}23{"step": 1000, "loss": 0.4246249198913574, "lr": 0.000802534179413072, "grad_norm": 0.11484304815530777, "tokens": 165602, "tok_per_s": 148.1819017564466, "t": 1118.22}24{"step": 1050, "loss": 0.37108105421066284, "lr": 0.0007819992357296598, "grad_norm": 0.11007954180240631, "tokens": 173674, "tok_per_s": 147.91907525389004, "t": 1174.77}25{"step": 1100, "loss": 0.3745608627796173, "lr": 0.0007607837983149056, "grad_norm": 0.052189599722623825, "tokens": 181834, "tok_per_s": 147.94023683114756, "t": 1229.77}26{"step": 1150, "loss": 0.41452881693840027, "lr": 0.0007389500957021191, "grad_norm": 0.13155852258205414, "tokens": 190158, "tok_per_s": 148.11433369100746, "t": 1284.52}27{"step": 1200, "loss": 0.3948362171649933, "lr": 0.0007165621699029615, "grad_norm": 0.1459951251745224, "tokens": 198573, "tok_per_s": 148.37811120814473, "t": 1338.95}28{"step": 1250, "loss": 0.470586359500885, "lr": 0.000693685688561191, "grad_norm": 0.08344906568527222, "tokens": 206880, "tok_per_s": 148.36688722965613, "t": 1395.04}29{"step": 1300, "loss": 0.30484193563461304, "lr": 0.0006703877523381495, "grad_norm": 0.11007700860500336, "tokens": 215095, "tok_per_s": 148.42659144619103, "t": 1449.82}30{"step": 1350, "loss": 0.4042162597179413, "lr": 0.0006467366980949609, "grad_norm": 0.1435810774564743, "tokens": 223349, "tok_per_s": 148.46603778218147, "t": 1505.05}31{"step": 1400, "loss": 0.43683746457099915, "lr": 0.0006228018984487442, "grad_norm": 0.07633882761001587, "tokens": 231716, "tok_per_s": 148.5661768647728, "t": 1560.34}32{"step": 1450, "loss": 0.5070833563804626, "lr": 0.0005986535582907738, "grad_norm": 0.13211405277252197, "tokens": 239960, "tok_per_s": 148.59401882915603, "t": 1615.52}33{"step": 1499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4575978143277457, "val_loss_b2": 0.45599388923475814, "val_loss_b4": 0.4559580735704357, "val_loss_b8": 0.4651619547866964, "t": 1680.86}34{"step": 1500, "loss": 0.4394952654838562, "lr": 0.000574362508863438, "grad_norm": 0.1050797700881958, "tokens": 248357, "tok_per_s": 147.71261013842792, "t": 1682.01}35{"step": 1550, "loss": 0.567806601524353, "lr": 0.00055, "grad_norm": 0.26982736587524414, "tokens": 256793, "tok_per_s": 147.84676693689931, "t": 1737.55}36{"step": 1600, "loss": 0.479268878698349, "lr": 0.0005256374911365621, "grad_norm": 0.15289431810379028, "tokens": 265023, "tok_per_s": 147.90558313190598, "t": 1792.5}37{"step": 1650, "loss": 0.5186282396316528, "lr": 0.0005013464417092262, "grad_norm": 0.25726112723350525, "tokens": 273283, "tok_per_s": 148.116319686818, "t": 1845.71}38{"step": 1700, "loss": 0.47060999274253845, "lr": 0.00047719810155125587, "grad_norm": 0.15101228654384613, "tokens": 281671, "tok_per_s": 148.88270639797125, "t": 1892.55}39{"step": 1750, "loss": 0.4045570492744446, "lr": 0.0004532633019050392, "grad_norm": 0.0777166336774826, "tokens": 289799, "tok_per_s": 149.46368782455437, "t": 1939.59}40{"step": 1800, "loss": 0.3628152906894684, "lr": 0.0004296122476618507, "grad_norm": 0.14523981511592865, "tokens": 298140, "tok_per_s": 150.97690078066762, "t": 1975.39}41{"step": 1850, "loss": 0.46262747049331665, "lr": 0.00040631431143880916, "grad_norm": 0.15387821197509766, "tokens": 306518, "tok_per_s": 152.48236629216353, "t": 2010.84}42{"step": 1900, "loss": 0.34310415387153625, "lr": 0.00038343783009703845, "grad_norm": 0.09388072043657303, "tokens": 314885, "tok_per_s": 153.90000861254953, "t": 2046.69}43{"step": 1950, "loss": 0.3163958787918091, "lr": 0.000361049904297881, "grad_norm": 0.13515035808086395, "tokens": 322988, "tok_per_s": 155.12653016634013, "t": 2082.75}44{"step": 1999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45598918619621204, "val_loss_b2": 0.4536574708234506, "val_loss_b4": 0.4532863660185369, "val_loss_b8": 0.4615064557988753, "t": 2122.89}45{"step": 2000, "loss": 0.4839879870414734, "lr": 0.0003392162016850945, "grad_norm": 0.1424606591463089, "tokens": 331190, "tok_per_s": 156.00337075213213, "t": 2123.62}46{"step": 2050, "loss": 0.3992325961589813, "lr": 0.0003180007642703402, "grad_norm": 0.1377670168876648, "tokens": 339365, "tok_per_s": 157.19359408135952, "t": 2159.55}47{"step": 2100, "loss": 0.34755998849868774, "lr": 0.000297465820586928, "grad_norm": 0.14228233695030212, "tokens": 347821, "tok_per_s": 158.47889191191004, "t": 2195.4}48{"step": 2150, "loss": 0.4872566759586334, "lr": 0.0002776716031628058, "grad_norm": 0.09842777997255325, "tokens": 355802, "tok_per_s": 159.52032497555248, "t": 2231.1}49{"step": 2200, "loss": 0.34238770604133606, "lr": 0.00025867617184817756, "grad_norm": 0.15344107151031494, "tokens": 363932, "tok_per_s": 160.57914678095014, "t": 2267.03}50{"step": 2250, "loss": 0.3011631369590759, "lr": 0.00024053524351595949, "grad_norm": 0.11105113476514816, "tokens": 372235, "tok_per_s": 161.67821574766884, "t": 2302.97}51{"step": 2300, "loss": 0.3706061840057373, "lr": 0.0002233020286345913, "grad_norm": 0.11747881025075912, "tokens": 380617, "tok_per_s": 162.94730428102258, "t": 2336.48}52{"step": 2350, "loss": 0.34727150201797485, "lr": 0.00020702707519256364, "grad_norm": 0.17075566947460175, "tokens": 388820, "tok_per_s": 164.45657167371513, "t": 2364.93}53{"step": 2400, "loss": 0.36295080184936523, "lr": 0.00019175812043246033, "grad_norm": 0.24125869572162628, "tokens": 397013, "tok_per_s": 165.93431490699876, "t": 2393.25}54{"step": 2450, "loss": 0.396942138671875, "lr": 0.00017753995082939931, "grad_norm": 0.16931453347206116, "tokens": 405444, "tok_per_s": 167.47320295936163, "t": 2421.6}55{"step": 2499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45293897663565796, "val_loss_b2": 0.44958042804686055, "val_loss_b4": 0.4480721317176574, "val_loss_b8": 0.45787124765395176, "t": 2451.23}56{"step": 2500, "loss": 0.46242478489875793, "lr": 0.00016441427072458494, "grad_norm": 0.17871211469173431, "tokens": 413920, "tok_per_s": 168.86966932444523, "t": 2451.78}57{"step": 2550, "loss": 0.32758665084838867, "lr": 0.0001524195799992897, "grad_norm": 0.12175080925226212, "tokens": 422230, "tok_per_s": 170.34278998443114, "t": 2479.36}58{"step": 2600, "loss": 0.4169160723686218, "lr": 0.0001415910611480694, "grad_norm": 0.19377794861793518, "tokens": 430404, "tok_per_s": 171.78538890720333, "t": 2506.13}59{"step": 2650, "loss": 0.3888188600540161, "lr": 0.00013196047608244387, "grad_norm": 0.14564195275306702, "tokens": 438498, "tok_per_s": 173.09827908763896, "t": 2533.89}60{"step": 2700, "loss": 0.3320978283882141, "lr": 0.00012355607296773896, "grad_norm": 0.12690095603466034, "tokens": 446658, "tok_per_s": 174.45579925960985, "t": 2560.95}61{"step": 2750, "loss": 0.3505524694919586, "lr": 0.00011640250336634969, "grad_norm": 0.1341555267572403, "tokens": 455155, "tok_per_s": 175.8759097410928, "t": 2588.59}62{"step": 2800, "loss": 0.43095409870147705, "lr": 0.00011052074993046101, "grad_norm": 0.17616839706897736, "tokens": 463083, "tok_per_s": 177.11846151201095, "t": 2615.19}63{"step": 2850, "loss": 0.3735902011394501, "lr": 0.00010592806485631325, "grad_norm": 0.1756911277770996, "tokens": 471665, "tok_per_s": 178.60829311081648, "t": 2641.43}64{"step": 2900, "loss": 0.42717376351356506, "lr": 0.00010263791928053818, "grad_norm": 0.1406625360250473, "tokens": 480193, "tok_per_s": 180.03573796293364, "t": 2667.86}65{"step": 2950, "loss": 0.32064107060432434, "lr": 0.0001006599637669943, "grad_norm": 0.17328202724456787, "tokens": 488284, "tok_per_s": 181.24731618074574, "t": 2694.67}66{"step": 2999, "loss": 0.4342266023159027, "lr": 0.0001000002640500325, "grad_norm": 0.18881288170814514, "tokens": 496298, "tok_per_s": 182.43782657215047, "t": 2721.02}67{"step": 2999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45249956449656253, "val_loss_b2": 0.4493421927762772, "val_loss_b4": 0.4485216267652968, "val_loss_b8": 0.45878077939909157, "t": 2724.0}68{"step": 2999, "checkpoint_sha256": "ab38f43f354ccb6655bef06a46a625c5971c008ca3299b7110135ddc307080b7", "t": 2724.05}69