SPB Git

spb/wp31_uqo

Public
11.3 KB · 69 lines
1{"step": 0, "loss": 0.8017767071723938, "lr": 1e-05, "grad_norm": 4.934864044189453, "tokens": 163, "tok_per_s": 142.65310171253444, "t": 3.81}2{"step": 50, "loss": 0.4542183578014374, "lr": 0.00051, "grad_norm": 1.8498035669326782, "tokens": 8255, "tok_per_s": 1562.7960579276628, "t": 7.95}3{"step": 100, "loss": 0.388010174036026, "lr": 0.001, "grad_norm": 0.9171138405799866, "tokens": 16571, "tok_per_s": 1985.9098872659356, "t": 11.02}4{"step": 150, "loss": 0.5282694101333618, "lr": 0.0009993400362330058, "grad_norm": 0.6141986846923828, "tokens": 25277, "tok_per_s": 2191.589505317393, "t": 14.2}5{"step": 200, "loss": 0.5822941064834595, "lr": 0.0009973620807194618, "grad_norm": 0.6423128843307495, "tokens": 33752, "tok_per_s": 2302.530308577057, "t": 17.33}6{"step": 250, "loss": 0.44418156147003174, "lr": 0.000994071935143687, "grad_norm": 0.7697323560714722, "tokens": 41944, "tok_per_s": 2349.107500893104, "t": 20.53}7{"step": 300, "loss": 0.5605698227882385, "lr": 0.000989479250069539, "grad_norm": 0.6770164370536804, "tokens": 49880, "tok_per_s": 2345.219391275488, "t": 23.94}8{"step": 350, "loss": 0.35941705107688904, "lr": 0.0009835974966336503, "grad_norm": 0.4187207520008087, "tokens": 58163, "tok_per_s": 2362.5913956952104, "t": 27.29}9{"step": 400, "loss": 0.5421345233917236, "lr": 0.0009764439270322612, "grad_norm": 0.6944790482521057, "tokens": 66164, "tok_per_s": 2376.60745516817, "t": 30.51}10{"step": 450, "loss": 0.39259129762649536, "lr": 0.0009680395239175562, "grad_norm": 0.5315750241279602, "tokens": 74558, "tok_per_s": 2393.3893185095753, "t": 33.82}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4560640945171358, "val_loss_b2": 0.46063048643334975, "val_loss_b4": 0.4723873279434287, "val_loss_b8": 0.490189107196445, "t": 38.77}12{"step": 500, "loss": 0.48681995272636414, "lr": 0.0009584089388519307, "grad_norm": 0.44911009073257446, "tokens": 82715, "tok_per_s": 2287.739448226678, "t": 38.83}13{"step": 550, "loss": 0.46766170859336853, "lr": 0.0009475804200007103, "grad_norm": 0.3917340040206909, "tokens": 91267, "tok_per_s": 2316.192057459725, "t": 42.08}14{"step": 600, "loss": 0.3250301480293274, "lr": 0.0009355857292754152, "grad_norm": 0.3431518077850342, "tokens": 99524, "tok_per_s": 2333.4008401081055, "t": 45.32}15{"step": 650, "loss": 0.4490772783756256, "lr": 0.0009224600491706008, "grad_norm": 0.3514217436313629, "tokens": 107970, "tok_per_s": 2353.542450279552, "t": 48.55}16{"step": 700, "loss": 0.485782653093338, "lr": 0.0009082418795675396, "grad_norm": 0.497722864151001, "tokens": 116272, "tok_per_s": 2367.529225313142, "t": 51.78}17{"step": 750, "loss": 0.42726218700408936, "lr": 0.0008929729248074364, "grad_norm": 0.2790203392505646, "tokens": 124586, "tok_per_s": 2379.303483277164, "t": 55.03}18{"step": 800, "loss": 0.3589288592338562, "lr": 0.0008766979713654089, "grad_norm": 0.3077276051044464, "tokens": 132471, "tok_per_s": 2387.0173759006457, "t": 58.17}19{"step": 850, "loss": 0.4341985285282135, "lr": 0.0008594647564840406, "grad_norm": 0.24051214754581451, "tokens": 140749, "tok_per_s": 2396.7107457693733, "t": 61.4}20{"step": 900, "loss": 0.2865341901779175, "lr": 0.0008413238281518224, "grad_norm": 0.2615520656108856, "tokens": 149107, "tok_per_s": 2404.1130891411603, "t": 64.69}21{"step": 950, "loss": 0.3770018219947815, "lr": 0.0008223283968371944, "grad_norm": 0.3301536738872528, "tokens": 157427, "tok_per_s": 2412.320676627515, "t": 67.93}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4550681605369602, "val_loss_b2": 0.4573730395426593, "val_loss_b4": 0.46683786560479734, "val_loss_b8": 0.48176490949139095, "t": 72.65}23{"step": 1000, "loss": 0.4199550449848175, "lr": 0.000802534179413072, "grad_norm": 0.252043217420578, "tokens": 165723, "tok_per_s": 2365.8670376614177, "t": 72.72}24{"step": 1050, "loss": 0.4242469072341919, "lr": 0.0007819992357296598, "grad_norm": 0.4165693521499634, "tokens": 173820, "tok_per_s": 2372.909287204212, "t": 75.92}25{"step": 1100, "loss": 0.3397441506385803, "lr": 0.0007607837983149056, "grad_norm": 0.2649449110031128, "tokens": 182174, "tok_per_s": 2383.267209460103, "t": 79.11}26{"step": 1150, "loss": 0.42873257398605347, "lr": 0.0007389500957021191, "grad_norm": 0.3752756416797638, "tokens": 190590, "tok_per_s": 2392.1075314088143, "t": 82.35}27{"step": 1200, "loss": 0.3328815996646881, "lr": 0.0007165621699029615, "grad_norm": 0.2853265404701233, "tokens": 198577, "tok_per_s": 2394.104433639815, "t": 85.62}28{"step": 1250, "loss": 0.3614024519920349, "lr": 0.000693685688561191, "grad_norm": 0.2816864848136902, "tokens": 206821, "tok_per_s": 2400.505050392292, "t": 88.83}29{"step": 1300, "loss": 0.4437832534313202, "lr": 0.0006703877523381495, "grad_norm": 0.23911873996257782, "tokens": 214819, "tok_per_s": 2403.7890477335372, "t": 92.04}30{"step": 1350, "loss": 0.41722896695137024, "lr": 0.0006467366980949609, "grad_norm": 0.29949334263801575, "tokens": 222971, "tok_per_s": 2408.9286035851183, "t": 95.23}31{"step": 1400, "loss": 0.3843200206756592, "lr": 0.0006228018984487442, "grad_norm": 0.20001114904880524, "tokens": 231317, "tok_per_s": 2414.503674613457, "t": 98.47}32{"step": 1450, "loss": 0.5306939482688904, "lr": 0.0005986535582907738, "grad_norm": 0.46018195152282715, "tokens": 239557, "tok_per_s": 2418.121034069561, "t": 101.74}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.44941918141387144, "val_loss_b2": 0.44756747020648113, "val_loss_b4": 0.45192503705877807, "val_loss_b8": 0.4625963932073075, "t": 106.52}34{"step": 1500, "loss": 0.4694669246673584, "lr": 0.000574362508863438, "grad_norm": 0.2524014711380005, "tokens": 247607, "tok_per_s": 2382.631072953821, "t": 106.59}35{"step": 1550, "loss": 0.42443329095840454, "lr": 0.00055, "grad_norm": 0.3638123869895935, "tokens": 256272, "tok_per_s": 2391.3805706116004, "t": 109.84}36{"step": 1600, "loss": 0.3917281925678253, "lr": 0.0005256374911365621, "grad_norm": 0.2598702013492584, "tokens": 264490, "tok_per_s": 2395.740371599433, "t": 113.07}37{"step": 1650, "loss": 0.33224740624427795, "lr": 0.0005013464417092262, "grad_norm": 0.2798418700695038, "tokens": 272714, "tok_per_s": 2399.696095857251, "t": 116.32}38{"step": 1700, "loss": 0.523506760597229, "lr": 0.00047719810155125587, "grad_norm": 0.30509859323501587, "tokens": 280990, "tok_per_s": 2403.5979209828793, "t": 119.58}39{"step": 1750, "loss": 0.3866342604160309, "lr": 0.0004532633019050392, "grad_norm": 0.3371191620826721, "tokens": 289345, "tok_per_s": 2407.6552948478156, "t": 122.85}40{"step": 1800, "loss": 0.26081669330596924, "lr": 0.0004296122476618507, "grad_norm": 0.24196313321590424, "tokens": 297639, "tok_per_s": 2412.3292161651307, "t": 126.05}41{"step": 1850, "loss": 0.39694738388061523, "lr": 0.00040631431143880916, "grad_norm": 0.4532412588596344, "tokens": 305990, "tok_per_s": 2416.1701114078355, "t": 129.31}42{"step": 1900, "loss": 0.29010137915611267, "lr": 0.00038343783009703845, "grad_norm": 0.23989279568195343, "tokens": 313976, "tok_per_s": 2417.922415592106, "t": 132.52}43{"step": 1950, "loss": 0.34788772463798523, "lr": 0.000361049904297881, "grad_norm": 0.26917609572410583, "tokens": 322226, "tok_per_s": 2420.849201376273, "t": 135.78}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4478304605874127, "val_loss_b2": 0.44517554973604173, "val_loss_b4": 0.4477500830923412, "val_loss_b8": 0.4528308471709769, "t": 140.47}45{"step": 2000, "loss": 0.38147029280662537, "lr": 0.0003392162016850945, "grad_norm": 0.30243563652038574, "tokens": 330697, "tok_per_s": 2398.657296977859, "t": 140.54}46{"step": 2050, "loss": 0.43788599967956543, "lr": 0.0003180007642703402, "grad_norm": 0.27747324109077454, "tokens": 338821, "tok_per_s": 2401.9769839815717, "t": 143.73}47{"step": 2100, "loss": 0.44873154163360596, "lr": 0.000297465820586928, "grad_norm": 0.3451139032840729, "tokens": 347181, "tok_per_s": 2405.806584992432, "t": 146.98}48{"step": 2150, "loss": 0.42883092164993286, "lr": 0.0002776716031628058, "grad_norm": 0.32427701354026794, "tokens": 355319, "tok_per_s": 2408.221474986868, "t": 150.22}49{"step": 2200, "loss": 0.3522076904773712, "lr": 0.00025867617184817756, "grad_norm": 0.24677053093910217, "tokens": 363537, "tok_per_s": 2411.9855391220253, "t": 153.39}50{"step": 2250, "loss": 0.4798969626426697, "lr": 0.00024053524351595949, "grad_norm": 0.5011794567108154, "tokens": 371638, "tok_per_s": 2412.9948922869485, "t": 156.69}51{"step": 2300, "loss": 0.3174619674682617, "lr": 0.0002233020286345913, "grad_norm": 0.3664951026439667, "tokens": 379727, "tok_per_s": 2415.181340226791, "t": 159.9}52{"step": 2350, "loss": 0.3939259350299835, "lr": 0.00020702707519256364, "grad_norm": 0.3263911008834839, "tokens": 388116, "tok_per_s": 2418.6368217202507, "t": 163.14}53{"step": 2400, "loss": 0.2784528434276581, "lr": 0.00019175812043246033, "grad_norm": 0.30013370513916016, "tokens": 396308, "tok_per_s": 2421.4478570210695, "t": 166.34}54{"step": 2450, "loss": 0.3584173321723938, "lr": 0.00017753995082939931, "grad_norm": 0.3617996275424957, "tokens": 404717, "tok_per_s": 2424.557681388625, "t": 169.6}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4459256011902728, "val_loss_b2": 0.443198884162132, "val_loss_b4": 0.4446484240009727, "val_loss_b8": 0.44663645564305365, "t": 174.33}56{"step": 2500, "loss": 0.4710994362831116, "lr": 0.00016441427072458494, "grad_norm": 0.39778396487236023, "tokens": 413027, "tok_per_s": 2405.2832119542627, "t": 174.39}57{"step": 2550, "loss": 0.43619775772094727, "lr": 0.0001524195799992897, "grad_norm": 0.32504820823669434, "tokens": 421798, "tok_per_s": 2410.6739763444807, "t": 177.64}58{"step": 2600, "loss": 0.2805783748626709, "lr": 0.0001415910611480694, "grad_norm": 0.34408047795295715, "tokens": 430295, "tok_per_s": 2414.882197230238, "t": 180.86}59{"step": 2650, "loss": 0.49194765090942383, "lr": 0.00013196047608244387, "grad_norm": 0.4103563129901886, "tokens": 438698, "tok_per_s": 2418.627332551002, "t": 184.05}60{"step": 2700, "loss": 0.4329032301902771, "lr": 0.00012355607296773896, "grad_norm": 0.40356290340423584, "tokens": 447187, "tok_per_s": 2423.53404609926, "t": 187.19}61{"step": 2750, "loss": 0.38922011852264404, "lr": 0.00011640250336634969, "grad_norm": 0.3549686670303345, "tokens": 455379, "tok_per_s": 2431.4413530678526, "t": 189.96}62{"step": 2800, "loss": 0.3951728940010071, "lr": 0.00011052074993046101, "grad_norm": 0.3550599217414856, "tokens": 463791, "tok_per_s": 2436.2774769070716, "t": 193.04}63{"step": 2850, "loss": 0.452644407749176, "lr": 0.00010592806485631325, "grad_norm": 0.46200940012931824, "tokens": 472187, "tok_per_s": 2439.1026555877615, "t": 196.26}64{"step": 2900, "loss": 0.42934152483940125, "lr": 0.00010263791928053818, "grad_norm": 0.38615965843200684, "tokens": 480621, "tok_per_s": 2442.913499356355, "t": 199.41}65{"step": 2950, "loss": 0.3274889886379242, "lr": 0.0001006599637669943, "grad_norm": 0.29778072237968445, "tokens": 488878, "tok_per_s": 2444.5964933995356, "t": 202.65}66{"step": 2999, "loss": 0.37491413950920105, "lr": 0.0001000002640500325, "grad_norm": 0.4036896526813507, "tokens": 497078, "tok_per_s": 2447.2519891388392, "t": 205.79}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4432142472960433, "val_loss_b2": 0.4394001046561913, "val_loss_b4": 0.4398238368666542, "val_loss_b8": 0.4413375271665104, "t": 207.42}68{"step": 2999, "checkpoint_sha256": "bc836ad6cec6ba3c7ff8d1a3d168a311540e823ca58ff4480fc035132d2f369e", "t": 207.46}69
Files · f94c71f5c2e9d93426ef4886b309b4e0a6711296
Menu

Theme