1{"step": 0, "loss": 0.42600929737091064, "lr": 1e-05, "grad_norm": 0.09819529205560684, "tokens": 139, "tok_per_s": 79.7060319112183, "t": 2.38}2{"step": 50, "loss": 0.31310924887657166, "lr": 0.00051, "grad_norm": 0.10251859575510025, "tokens": 8357, "tok_per_s": 246.26434594839742, "t": 34.58}3{"step": 100, "loss": 0.32046571373939514, "lr": 0.001, "grad_norm": 0.06212421879172325, "tokens": 16659, "tok_per_s": 253.18751172057708, "t": 66.44}4{"step": 150, "loss": 0.45319104194641113, "lr": 0.0009993400362330058, "grad_norm": 0.0695132166147232, "tokens": 25169, "tok_per_s": 260.7716218650578, "t": 97.17}5{"step": 200, "loss": 0.4841672480106354, "lr": 0.0009973620807194618, "grad_norm": 0.08476220071315765, "tokens": 33572, "tok_per_s": 264.1846091123452, "t": 127.72}6{"step": 250, "loss": 0.41961753368377686, "lr": 0.000994071935143687, "grad_norm": 0.055930305272340775, "tokens": 41743, "tok_per_s": 264.0656520179998, "t": 158.72}7{"step": 300, "loss": 0.39448386430740356, "lr": 0.000989479250069539, "grad_norm": 0.037942465394735336, "tokens": 49922, "tok_per_s": 262.21443579151406, "t": 191.03}8{"step": 350, "loss": 0.3877321481704712, "lr": 0.0009835974966336503, "grad_norm": 0.05625578761100769, "tokens": 58442, "tok_per_s": 262.4308507571978, "t": 223.34}9{"step": 400, "loss": 0.3457121253013611, "lr": 0.0009764439270322612, "grad_norm": 0.08046363294124603, "tokens": 66780, "tok_per_s": 261.71835935488946, "t": 255.82}10{"step": 450, "loss": 0.4830707609653473, "lr": 0.0009680395239175562, "grad_norm": 0.06497915834188461, "tokens": 74983, "tok_per_s": 261.23929436872913, "t": 287.67}11{"step": 499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45937798351064113, "val_loss_b2": 0.4570861372705632, "val_loss_b4": 0.4544256792214188, "val_loss_b8": 0.4547588579639418, "t": 328.11}12{"step": 500, "loss": 0.32655662298202515, "lr": 0.0009584089388519307, "grad_norm": 0.05491314083337784, "tokens": 83177, "tok_per_s": 253.50666250881477, "t": 328.75}13{"step": 550, "loss": 0.4081708490848541, "lr": 0.0009475804200007103, "grad_norm": 0.14327313005924225, "tokens": 91423, "tok_per_s": 254.5563796546934, "t": 359.78}14{"step": 600, "loss": 0.3794846534729004, "lr": 0.0009355857292754152, "grad_norm": 0.0578434020280838, "tokens": 99986, "tok_per_s": 255.62167843719413, "t": 391.79}15{"step": 650, "loss": 0.3533087372779846, "lr": 0.0009224600491706008, "grad_norm": 0.08094552159309387, "tokens": 108206, "tok_per_s": 255.19348191170613, "t": 424.66}16{"step": 700, "loss": 0.4506843388080597, "lr": 0.0009082418795675396, "grad_norm": 0.24620503187179565, "tokens": 116415, "tok_per_s": 255.98515056826773, "t": 455.42}17{"step": 750, "loss": 0.33725234866142273, "lr": 0.0008929729248074364, "grad_norm": 0.052444323897361755, "tokens": 124709, "tok_per_s": 256.2899363331144, "t": 487.23}18{"step": 800, "loss": 0.424780935049057, "lr": 0.0008766979713654089, "grad_norm": 0.09239649772644043, "tokens": 132874, "tok_per_s": 256.3333561835178, "t": 519.0}19{"step": 850, "loss": 0.3684316575527191, "lr": 0.0008594647564840406, "grad_norm": 0.07766187936067581, "tokens": 141506, "tok_per_s": 257.4207648338504, "t": 550.35}20{"step": 900, "loss": 0.29275602102279663, "lr": 0.0008413238281518224, "grad_norm": 0.07673187553882599, "tokens": 149769, "tok_per_s": 257.8366133458832, "t": 581.51}21{"step": 950, "loss": 0.3583630919456482, "lr": 0.0008223283968371944, "grad_norm": 0.06168017163872719, "tokens": 158176, "tok_per_s": 258.07221619742216, "t": 613.56}22{"step": 999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4572246621076904, "val_loss_b2": 0.45431125992445476, "val_loss_b4": 0.45148454668957827, "val_loss_b8": 0.45394115344593244, "t": 654.02}23{"step": 1000, "loss": 0.41413575410842896, "lr": 0.000802534179413072, "grad_norm": 0.06757870316505432, "tokens": 166573, "tok_per_s": 254.68138137106783, "t": 654.68}24{"step": 1050, "loss": 0.3562120497226715, "lr": 0.0007819992357296598, "grad_norm": 0.06940312683582306, "tokens": 174773, "tok_per_s": 254.6977232026141, "t": 686.85}25{"step": 1100, "loss": 0.43775850534439087, "lr": 0.0007607837983149056, "grad_norm": 0.08178038150072098, "tokens": 183200, "tok_per_s": 255.11529028163676, "t": 718.75}26{"step": 1150, "loss": 0.46260759234428406, "lr": 0.0007389500957021191, "grad_norm": 0.1526055485010147, "tokens": 191474, "tok_per_s": 255.2527667740732, "t": 750.77}27{"step": 1200, "loss": 0.39114105701446533, "lr": 0.0007165621699029615, "grad_norm": 0.14781397581100464, "tokens": 200061, "tok_per_s": 255.61852356086183, "t": 783.3}28{"step": 1250, "loss": 0.38562870025634766, "lr": 0.000693685688561191, "grad_norm": 0.10610388219356537, "tokens": 208455, "tok_per_s": 255.99772727246176, "t": 814.93}29{"step": 1300, "loss": 0.36638128757476807, "lr": 0.0006703877523381495, "grad_norm": 0.16020601987838745, "tokens": 216814, "tok_per_s": 256.24097531384416, "t": 846.77}30{"step": 1350, "loss": 0.3341001868247986, "lr": 0.0006467366980949609, "grad_norm": 0.10322532802820206, "tokens": 225136, "tok_per_s": 256.5302183667267, "t": 878.26}31{"step": 1400, "loss": 0.3434232175350189, "lr": 0.0006228018984487442, "grad_norm": 0.0803132876753807, "tokens": 233471, "tok_per_s": 256.6490270356868, "t": 910.33}32{"step": 1450, "loss": 0.3097214996814728, "lr": 0.0005986535582907738, "grad_norm": 0.07866869866847992, "tokens": 241795, "tok_per_s": 256.81827698463593, "t": 942.14}33{"step": 1499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4558672521900461, "val_loss_b2": 0.452787159580853, "val_loss_b4": 0.4519247766666779, "val_loss_b8": 0.4633053797932903, "t": 981.03}34{"step": 1500, "loss": 0.3169599771499634, "lr": 0.000574362508863438, "grad_norm": 0.07063798606395721, "tokens": 250085, "tok_per_s": 254.9221820512249, "t": 981.66}35{"step": 1550, "loss": 0.38266003131866455, "lr": 0.00055, "grad_norm": 0.12140710651874542, "tokens": 258187, "tok_per_s": 254.91181981321466, "t": 1013.5}36{"step": 1600, "loss": 0.35804569721221924, "lr": 0.0005256374911365621, "grad_norm": 0.16125799715518951, "tokens": 266247, "tok_per_s": 254.8457754610941, "t": 1045.38}37{"step": 1650, "loss": 0.3065197169780731, "lr": 0.0005013464417092262, "grad_norm": 0.17282940447330475, "tokens": 274738, "tok_per_s": 255.29078320111887, "t": 1076.81}38{"step": 1700, "loss": 0.36491021513938904, "lr": 0.00047719810155125587, "grad_norm": 0.0831088125705719, "tokens": 283093, "tok_per_s": 255.6103373261805, "t": 1108.16}39{"step": 1750, "loss": 0.3896067440509796, "lr": 0.0004532633019050392, "grad_norm": 0.09102942794561386, "tokens": 291261, "tok_per_s": 255.8231402199511, "t": 1139.16}40{"step": 1800, "loss": 0.39703449606895447, "lr": 0.0004296122476618507, "grad_norm": 0.1041037067770958, "tokens": 299555, "tok_per_s": 255.8620145741759, "t": 1171.41}41{"step": 1850, "loss": 0.40914106369018555, "lr": 0.00040631431143880916, "grad_norm": 0.0925544798374176, "tokens": 307521, "tok_per_s": 255.72536078081464, "t": 1203.19}42{"step": 1900, "loss": 0.3876003623008728, "lr": 0.00038343783009703845, "grad_norm": 0.05495703965425491, "tokens": 315892, "tok_per_s": 255.97748760986968, "t": 1234.71}43{"step": 1950, "loss": 0.3475407361984253, "lr": 0.000361049904297881, "grad_norm": 0.05317451432347298, "tokens": 324298, "tok_per_s": 256.26236218123745, "t": 1266.13}44{"step": 1999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4557142036780164, "val_loss_b2": 0.453089340382929, "val_loss_b4": 0.45314410168763863, "val_loss_b8": 0.4629644409901641, "t": 1305.89}45{"step": 2000, "loss": 0.28176215291023254, "lr": 0.0003392162016850945, "grad_norm": 0.1228204146027565, "tokens": 332486, "tok_per_s": 254.5943576729419, "t": 1306.58}46{"step": 2050, "loss": 0.4493873715400696, "lr": 0.0003180007642703402, "grad_norm": 0.14940015971660614, "tokens": 340738, "tok_per_s": 254.84835192201527, "t": 1337.67}47{"step": 2100, "loss": 0.37307602167129517, "lr": 0.000297465820586928, "grad_norm": 0.1644899696111679, "tokens": 349148, "tok_per_s": 255.10349297053202, "t": 1369.3}48{"step": 2150, "loss": 0.3442464768886566, "lr": 0.0002776716031628058, "grad_norm": 0.1238342821598053, "tokens": 357576, "tok_per_s": 255.26519459532304, "t": 1401.44}49{"step": 2200, "loss": 0.3451700806617737, "lr": 0.00025867617184817756, "grad_norm": 0.16494151949882507, "tokens": 366097, "tok_per_s": 255.60067649378738, "t": 1432.94}50{"step": 2250, "loss": 0.2721230089664459, "lr": 0.00024053524351595949, "grad_norm": 0.1108018159866333, "tokens": 374285, "tok_per_s": 255.52634512818412, "t": 1465.4}51{"step": 2300, "loss": 0.3701423406600952, "lr": 0.0002233020286345913, "grad_norm": 0.09862100332975388, "tokens": 382864, "tok_per_s": 255.72215473018485, "t": 1497.83}52{"step": 2350, "loss": 0.447528213262558, "lr": 0.00020702707519256364, "grad_norm": 0.12663058936595917, "tokens": 391427, "tok_per_s": 256.0176132920307, "t": 1529.55}53{"step": 2400, "loss": 0.6319020390510559, "lr": 0.00019175812043246033, "grad_norm": 0.18015898764133453, "tokens": 399797, "tok_per_s": 256.08424181814854, "t": 1561.83}54{"step": 2450, "loss": 0.34492528438568115, "lr": 0.00017753995082939931, "grad_norm": 0.1323632001876831, "tokens": 407948, "tok_per_s": 256.1596790055352, "t": 1593.19}55{"step": 2499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45335943937419143, "val_loss_b2": 0.44949900240308255, "val_loss_b4": 0.4477254266865798, "val_loss_b8": 0.4550224104909958, "t": 1631.86}56{"step": 2500, "loss": 0.3921298086643219, "lr": 0.00016441427072458494, "grad_norm": 0.12986211478710175, "tokens": 416019, "tok_per_s": 254.94583374977543, "t": 1632.43}57{"step": 2550, "loss": 0.3198011815547943, "lr": 0.0001524195799992897, "grad_norm": 0.0904293954372406, "tokens": 424449, "tok_per_s": 255.17975005781074, "t": 1663.97}58{"step": 2600, "loss": 0.3731447458267212, "lr": 0.0001415910611480694, "grad_norm": 0.16146691143512726, "tokens": 432494, "tok_per_s": 255.28724920087942, "t": 1694.79}59{"step": 2650, "loss": 0.2958233952522278, "lr": 0.00013196047608244387, "grad_norm": 0.12624478340148926, "tokens": 440693, "tok_per_s": 255.30217142133347, "t": 1726.8}60{"step": 2700, "loss": 0.3928040862083435, "lr": 0.00012355607296773896, "grad_norm": 0.19339953362941742, "tokens": 449079, "tok_per_s": 255.5791929947166, "t": 1757.74}61{"step": 2750, "loss": 0.3011181652545929, "lr": 0.00011640250336634969, "grad_norm": 0.08886498957872391, "tokens": 457686, "tok_per_s": 255.84505891289226, "t": 1789.56}62{"step": 2800, "loss": 0.41742876172065735, "lr": 0.00011052074993046101, "grad_norm": 0.14468859136104584, "tokens": 466220, "tok_per_s": 255.99489803077134, "t": 1821.86}63{"step": 2850, "loss": 0.4699750542640686, "lr": 0.00010592806485631325, "grad_norm": 0.20657917857170105, "tokens": 474198, "tok_per_s": 256.23372637910563, "t": 1851.29}64{"step": 2900, "loss": 0.43470683693885803, "lr": 0.00010263791928053818, "grad_norm": 0.20605768263339996, "tokens": 482297, "tok_per_s": 256.7521826701799, "t": 1879.1}65{"step": 2950, "loss": 0.4044400751590729, "lr": 0.0001006599637669943, "grad_norm": 0.12754181027412415, "tokens": 490516, "tok_per_s": 257.3401621951249, "t": 1906.75}66{"step": 2999, "loss": 0.5092548727989197, "lr": 0.0001000002640500325, "grad_norm": 0.20245473086833954, "tokens": 498543, "tok_per_s": 257.98041468321844, "t": 1933.13}67{"step": 2999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.45174925453031867, "val_loss_b2": 0.44875170903701744, "val_loss_b4": 0.44885935437920177, "val_loss_b8": 0.45871206972621353, "t": 1940.15}68{"step": 2999, "checkpoint_sha256": "828aa743ae7541657750d9df6a08150d35dde3bffbac8ba6d719d644644af6cf", "t": 1940.29}69