1{"step": 0, "loss": 0.44630467891693115, "lr": 1e-05, "grad_norm": 0.36323049664497375, "tokens": 139, "tok_per_s": 185.66089438242037, "t": 1.39}2{"step": 50, "loss": 0.31444671750068665, "lr": 0.00051, "grad_norm": 0.39926421642303467, "tokens": 8357, "tok_per_s": 343.70225611940765, "t": 24.96}3{"step": 100, "loss": 0.319304883480072, "lr": 0.001, "grad_norm": 0.12341362982988358, "tokens": 16659, "tok_per_s": 345.2637420054847, "t": 48.89}4{"step": 150, "loss": 0.45460402965545654, "lr": 0.0009993400362330058, "grad_norm": 0.0630602017045021, "tokens": 25169, "tok_per_s": 348.2216041361454, "t": 72.92}5{"step": 200, "loss": 0.48729655146598816, "lr": 0.0009973620807194618, "grad_norm": 0.1538952887058258, "tokens": 33572, "tok_per_s": 347.75170159581654, "t": 97.18}6{"step": 250, "loss": 0.41836345195770264, "lr": 0.000994071935143687, "grad_norm": 0.06097071245312691, "tokens": 41743, "tok_per_s": 346.06972369084247, "t": 121.26}7{"step": 300, "loss": 0.3973003923892975, "lr": 0.000989479250069539, "grad_norm": 0.0183993149548769, "tokens": 49922, "tok_per_s": 343.27066959854045, "t": 146.07}8{"step": 350, "loss": 0.38906919956207275, "lr": 0.0009835974966336503, "grad_norm": 0.06859804689884186, "tokens": 58442, "tok_per_s": 344.77789853786237, "t": 170.15}9{"step": 400, "loss": 0.35240718722343445, "lr": 0.0009764439270322612, "grad_norm": 0.09619385004043579, "tokens": 66780, "tok_per_s": 343.9132954432775, "t": 194.82}10{"step": 450, "loss": 0.4899499714374542, "lr": 0.0009680395239175562, "grad_norm": 0.11018010228872299, "tokens": 74983, "tok_per_s": 342.7655599824271, "t": 219.4}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4595996721323163, "val_loss_b2": 0.45795790276073833, "val_loss_b4": 0.4561124710451505, "val_loss_b8": 0.4606572457168301, "t": 249.36}12{"step": 500, "loss": 0.32592374086380005, "lr": 0.0009584089388519307, "grad_norm": 0.09187829494476318, "tokens": 83177, "tok_per_s": 333.72881406667693, "t": 249.88}13{"step": 550, "loss": 0.4120193421840668, "lr": 0.0009475804200007103, "grad_norm": 0.14943233132362366, "tokens": 91423, "tok_per_s": 334.2975453056424, "t": 274.12}14{"step": 600, "loss": 0.377208411693573, "lr": 0.0009355857292754152, "grad_norm": 0.11115574091672897, "tokens": 99986, "tok_per_s": 335.38774364185116, "t": 298.77}15{"step": 650, "loss": 0.35030150413513184, "lr": 0.0009224600491706008, "grad_norm": 0.053661055862903595, "tokens": 108206, "tok_per_s": 335.4076050258134, "t": 323.26}16{"step": 700, "loss": 0.436791330575943, "lr": 0.0009082418795675396, "grad_norm": 0.075992152094841, "tokens": 116415, "tok_per_s": 335.4938322633065, "t": 347.64}17{"step": 750, "loss": 0.3400858938694, "lr": 0.0008929729248074364, "grad_norm": 0.04048042371869087, "tokens": 124709, "tok_per_s": 335.5258687542365, "t": 372.32}18{"step": 800, "loss": 0.43084487318992615, "lr": 0.0008766979713654089, "grad_norm": 0.11450743675231934, "tokens": 132874, "tok_per_s": 335.58095432573805, "t": 396.59}19{"step": 850, "loss": 0.38120555877685547, "lr": 0.0008594647564840406, "grad_norm": 0.3344448208808899, "tokens": 141506, "tok_per_s": 336.58339961761004, "t": 421.06}20{"step": 900, "loss": 0.29077932238578796, "lr": 0.0008413238281518224, "grad_norm": 0.04789956659078598, "tokens": 149769, "tok_per_s": 336.7548221949956, "t": 445.39}21{"step": 950, "loss": 0.3581276834011078, "lr": 0.0008223283968371944, "grad_norm": 0.08371356874704361, "tokens": 158176, "tok_per_s": 337.19570698280273, "t": 469.73}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4583282527105746, "val_loss_b2": 0.455843098707655, "val_loss_b4": 0.45254595753240606, "val_loss_b8": 0.4538473963091208, "t": 499.5}23{"step": 1000, "loss": 0.41414856910705566, "lr": 0.000802534179413072, "grad_norm": 0.08701185882091522, "tokens": 166573, "tok_per_s": 333.59698980411207, "t": 499.96}24{"step": 1050, "loss": 0.3617335557937622, "lr": 0.0007819992357296598, "grad_norm": 0.0783894881606102, "tokens": 174773, "tok_per_s": 334.7232131790798, "t": 522.78}25{"step": 1100, "loss": 0.4375030994415283, "lr": 0.0007607837983149056, "grad_norm": 0.05900983512401581, "tokens": 183200, "tok_per_s": 335.1450270439675, "t": 547.27}26{"step": 1150, "loss": 0.47239094972610474, "lr": 0.0007389500957021191, "grad_norm": 0.1730019450187683, "tokens": 191474, "tok_per_s": 335.28958438879323, "t": 571.71}27{"step": 1200, "loss": 0.3963163495063782, "lr": 0.0007165621699029615, "grad_norm": 0.16108326613903046, "tokens": 200061, "tok_per_s": 336.0413287886918, "t": 595.99}28{"step": 1250, "loss": 0.3846502900123596, "lr": 0.000693685688561191, "grad_norm": 0.10626472532749176, "tokens": 208455, "tok_per_s": 336.4991584086481, "t": 620.12}29{"step": 1300, "loss": 0.3769327700138092, "lr": 0.0006703877523381495, "grad_norm": 0.22268860042095184, "tokens": 216814, "tok_per_s": 336.8966169979037, "t": 644.21}30{"step": 1350, "loss": 0.3357068598270416, "lr": 0.0006467366980949609, "grad_norm": 0.16063249111175537, "tokens": 225136, "tok_per_s": 337.25707927415783, "t": 668.19}31{"step": 1400, "loss": 0.3461225628852844, "lr": 0.0006228018984487442, "grad_norm": 0.08123651146888733, "tokens": 233471, "tok_per_s": 337.34156993318976, "t": 692.74}32{"step": 1450, "loss": 0.3089819550514221, "lr": 0.0005986535582907738, "grad_norm": 0.09042122215032578, "tokens": 241795, "tok_per_s": 337.4237323573589, "t": 717.23}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45599394657742276, "val_loss_b2": 0.45265138155136747, "val_loss_b4": 0.45284782812006663, "val_loss_b8": 0.46695765725140975, "t": 747.09}34{"step": 1500, "loss": 0.3163526952266693, "lr": 0.000574362508863438, "grad_norm": 0.06742726266384125, "tokens": 250085, "tok_per_s": 334.7886787190004, "t": 747.64}35{"step": 1550, "loss": 0.3836338222026825, "lr": 0.00055, "grad_norm": 0.10181848704814911, "tokens": 258187, "tok_per_s": 334.7975443564139, "t": 771.82}36{"step": 1600, "loss": 0.34999901056289673, "lr": 0.0005256374911365621, "grad_norm": 0.19403280317783356, "tokens": 266247, "tok_per_s": 334.74264491550485, "t": 796.02}37{"step": 1650, "loss": 0.31186237931251526, "lr": 0.0005013464417092262, "grad_norm": 0.21614901721477509, "tokens": 274738, "tok_per_s": 335.2623276977972, "t": 820.11}38{"step": 1700, "loss": 0.37356457114219666, "lr": 0.00047719810155125587, "grad_norm": 0.09632813185453415, "tokens": 283093, "tok_per_s": 335.5401729006363, "t": 844.34}39{"step": 1750, "loss": 0.38320884108543396, "lr": 0.0004532633019050392, "grad_norm": 0.12168017774820328, "tokens": 291261, "tok_per_s": 335.5637029482605, "t": 868.62}40{"step": 1800, "loss": 0.3961431682109833, "lr": 0.0004296122476618507, "grad_norm": 0.11367359012365341, "tokens": 299555, "tok_per_s": 335.7631532638505, "t": 892.81}41{"step": 1850, "loss": 0.40813907980918884, "lr": 0.00040631431143880916, "grad_norm": 0.10751962661743164, "tokens": 307521, "tok_per_s": 335.6346189525162, "t": 916.88}42{"step": 1900, "loss": 0.3903748095035553, "lr": 0.00038343783009703845, "grad_norm": 0.08989487588405609, "tokens": 315892, "tok_per_s": 335.9041805504241, "t": 941.06}43{"step": 1950, "loss": 0.34593990445137024, "lr": 0.000361049904297881, "grad_norm": 0.07539328187704086, "tokens": 324298, "tok_per_s": 336.1352635755803, "t": 965.43}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4544454022779319, "val_loss_b2": 0.45132772860237974, "val_loss_b4": 0.45280485120296715, "val_loss_b8": 0.46560857048053467, "t": 995.08}45{"step": 2000, "loss": 0.2857680022716522, "lr": 0.0003392162016850945, "grad_norm": 0.18842291831970215, "tokens": 332486, "tok_per_s": 334.1745674285471, "t": 995.59}46{"step": 2050, "loss": 0.4558248221874237, "lr": 0.0003180007642703402, "grad_norm": 0.1957319676876068, "tokens": 340738, "tok_per_s": 335.46848253012683, "t": 1016.35}47{"step": 2100, "loss": 0.3753258287906647, "lr": 0.000297465820586928, "grad_norm": 0.20918971300125122, "tokens": 349148, "tok_per_s": 338.077603609073, "t": 1033.39}48{"step": 2150, "loss": 0.34026336669921875, "lr": 0.0002776716031628058, "grad_norm": 0.14431822299957275, "tokens": 357576, "tok_per_s": 340.53742908318986, "t": 1050.68}49{"step": 2200, "loss": 0.345895379781723, "lr": 0.00025867617184817756, "grad_norm": 0.2337612509727478, "tokens": 366097, "tok_per_s": 342.9367370202564, "t": 1068.18}50{"step": 2250, "loss": 0.2733730673789978, "lr": 0.00024053524351595949, "grad_norm": 0.13830573856830597, "tokens": 374285, "tok_per_s": 344.982674032632, "t": 1085.58}51{"step": 2300, "loss": 0.3699128329753876, "lr": 0.0002233020286345913, "grad_norm": 0.11756070703268051, "tokens": 382864, "tok_per_s": 347.27315043434515, "t": 1103.13}52{"step": 2350, "loss": 0.45119956135749817, "lr": 0.00020702707519256364, "grad_norm": 0.1400342434644699, "tokens": 391427, "tok_per_s": 349.5675149790633, "t": 1120.39}53{"step": 2400, "loss": 0.6333733797073364, "lr": 0.00019175812043246033, "grad_norm": 0.23754920065402985, "tokens": 399797, "tok_per_s": 351.5667859444344, "t": 1137.83}54{"step": 2450, "loss": 0.3419897258281708, "lr": 0.00017753995082939931, "grad_norm": 0.14708295464515686, "tokens": 407948, "tok_per_s": 353.28015341566737, "t": 1155.38}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4532845667749276, "val_loss_b2": 0.44889283121484086, "val_loss_b4": 0.4461287916323538, "val_loss_b8": 0.4507280531159876, "t": 1176.69}56{"step": 2500, "loss": 0.40073591470718384, "lr": 0.00016441427072458494, "grad_norm": 0.1634984165430069, "tokens": 416019, "tok_per_s": 353.6390003120886, "t": 1177.04}57{"step": 2550, "loss": 0.3194027841091156, "lr": 0.0001524195799992897, "grad_norm": 0.11849913001060486, "tokens": 424449, "tok_per_s": 355.5497642618238, "t": 1194.42}58{"step": 2600, "loss": 0.3681815266609192, "lr": 0.0001415910611480694, "grad_norm": 0.19154568016529083, "tokens": 432494, "tok_per_s": 357.1147464628771, "t": 1211.72}59{"step": 2650, "loss": 0.2969713509082794, "lr": 0.00013196047608244387, "grad_norm": 0.19740034639835358, "tokens": 440693, "tok_per_s": 358.7349880774228, "t": 1229.1}60{"step": 2700, "loss": 0.3999093770980835, "lr": 0.00012355607296773896, "grad_norm": 0.19648607075214386, "tokens": 449079, "tok_per_s": 360.489044202197, "t": 1246.39}61{"step": 2750, "loss": 0.3013218343257904, "lr": 0.00011640250336634969, "grad_norm": 0.09931579977273941, "tokens": 457686, "tok_per_s": 362.3942796096135, "t": 1263.59}62{"step": 2800, "loss": 0.411276251077652, "lr": 0.00011052074993046101, "grad_norm": 0.1908463090658188, "tokens": 466220, "tok_per_s": 364.0730488754968, "t": 1281.21}63{"step": 2850, "loss": 0.4745626449584961, "lr": 0.00010592806485631325, "grad_norm": 0.2395177036523819, "tokens": 474198, "tok_per_s": 365.3824359648253, "t": 1298.45}64{"step": 2900, "loss": 0.44445785880088806, "lr": 0.00010263791928053818, "grad_norm": 0.2229040116071701, "tokens": 482297, "tok_per_s": 366.96709877213766, "t": 1314.92}65{"step": 2950, "loss": 0.39826861023902893, "lr": 0.0001006599637669943, "grad_norm": 0.2285844385623932, "tokens": 490516, "tok_per_s": 368.6378912368141, "t": 1331.26}66{"step": 2999, "loss": 0.5058788657188416, "lr": 0.0001000002640500325, "grad_norm": 0.24879617989063263, "tokens": 498543, "tok_per_s": 370.3409946719339, "t": 1346.81}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4518907141838243, "val_loss_b2": 0.44901827628416285, "val_loss_b4": 0.4504248251052488, "val_loss_b8": 0.46164766258063017, "t": 1350.74}68{"step": 2999, "checkpoint_sha256": "7cd030fa94e0c6c96a92c5ed6b1f634f004eac8a461bb6c99d0ff8fcfbedebb7", "t": 1350.79}69