SPB Git

spb/wp31_uqo

Public
1{"step": 0, "loss": 0.4486299157142639, "lr": 1e-05, "grad_norm": 0.17592467367649078, "tokens": 166, "tok_per_s": 106.3858552412938, "t": 2.02}2{"step": 50, "loss": 0.37719470262527466, "lr": 0.00051, "grad_norm": 0.15952259302139282, "tokens": 8464, "tok_per_s": 239.97767160410393, "t": 35.73}3{"step": 100, "loss": 0.3221386969089508, "lr": 0.001, "grad_norm": 0.07408096641302109, "tokens": 16901, "tok_per_s": 269.19441778160103, "t": 63.24}4{"step": 150, "loss": 0.2747827172279358, "lr": 0.0009993400362330058, "grad_norm": 0.04085555672645569, "tokens": 25274, "tok_per_s": 287.2149244456461, "t": 88.45}5{"step": 200, "loss": 0.379184752702713, "lr": 0.0009973620807194618, "grad_norm": 0.030557939782738686, "tokens": 33315, "tok_per_s": 297.0203420953969, "t": 112.62}6{"step": 250, "loss": 0.4596056044101715, "lr": 0.000994071935143687, "grad_norm": 0.07464102655649185, "tokens": 41251, "tok_per_s": 302.1524723552098, "t": 136.98}7{"step": 300, "loss": 0.33422812819480896, "lr": 0.000989479250069539, "grad_norm": 0.05552876368165016, "tokens": 49559, "tok_per_s": 307.742798681905, "t": 161.5}8{"step": 350, "loss": 0.35483047366142273, "lr": 0.0009835974966336503, "grad_norm": 0.045748304575681686, "tokens": 57986, "tok_per_s": 312.26108291488515, "t": 186.16}9{"step": 400, "loss": 0.4323274791240692, "lr": 0.0009764439270322612, "grad_norm": 0.10098536312580109, "tokens": 66476, "tok_per_s": 320.04046563332184, "t": 208.17}10{"step": 450, "loss": 0.30157196521759033, "lr": 0.0009680395239175562, "grad_norm": 0.06968528777360916, "tokens": 74868, "tok_per_s": 322.49127534017447, "t": 232.61}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45953497644596464, "val_loss_b2": 0.457609494370507, "val_loss_b4": 0.45578995333803646, "val_loss_b8": 0.4582821886900201, "t": 261.93}12{"step": 500, "loss": 0.44416940212249756, "lr": 0.0009584089388519307, "grad_norm": 0.15549325942993164, "tokens": 83020, "tok_per_s": 316.9159460280944, "t": 262.42}13{"step": 550, "loss": 0.5532170534133911, "lr": 0.0009475804200007103, "grad_norm": 0.033402711153030396, "tokens": 91309, "tok_per_s": 320.91278306571115, "t": 284.99}14{"step": 600, "loss": 0.37355974316596985, "lr": 0.0009355857292754152, "grad_norm": 0.03156343847513199, "tokens": 99553, "tok_per_s": 329.13437655676603, "t": 302.93}15{"step": 650, "loss": 0.39205750823020935, "lr": 0.0009224600491706008, "grad_norm": 0.09555712342262268, "tokens": 107933, "tok_per_s": 336.74225922353975, "t": 320.98}16{"step": 700, "loss": 0.3293212056159973, "lr": 0.0009082418795675396, "grad_norm": 0.11402836441993713, "tokens": 115964, "tok_per_s": 344.22885970195915, "t": 337.35}17{"step": 750, "loss": 0.46303293108940125, "lr": 0.0008929729248074364, "grad_norm": 0.07426144182682037, "tokens": 124286, "tok_per_s": 350.8631988909037, "t": 354.69}18{"step": 800, "loss": 0.44278421998023987, "lr": 0.0008766979713654089, "grad_norm": 0.07942043244838715, "tokens": 132702, "tok_per_s": 357.5254552904496, "t": 371.63}19{"step": 850, "loss": 0.46774259209632874, "lr": 0.0008594647564840406, "grad_norm": 0.0651860460639, "tokens": 141243, "tok_per_s": 364.3835353020137, "t": 388.08}20{"step": 900, "loss": 0.4555820822715759, "lr": 0.0008413238281518224, "grad_norm": 0.030174003913998604, "tokens": 149475, "tok_per_s": 369.85197431727767, "t": 404.61}21{"step": 950, "loss": 0.32870858907699585, "lr": 0.0008223283968371944, "grad_norm": 0.07059863209724426, "tokens": 157833, "tok_per_s": 374.46490210582186, "t": 421.95}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45945677985223776, "val_loss_b2": 0.45834000137652475, "val_loss_b4": 0.460141320602244, "val_loss_b8": 0.4710176043935807, "t": 440.91}23{"step": 1000, "loss": 0.5771011114120483, "lr": 0.000802534179413072, "grad_norm": 0.13217830657958984, "tokens": 166058, "tok_per_s": 376.60221350977366, "t": 441.4}24{"step": 1050, "loss": 0.39108771085739136, "lr": 0.0007819992357296598, "grad_norm": 0.06527083367109299, "tokens": 174370, "tok_per_s": 382.14431269600595, "t": 456.75}25{"step": 1100, "loss": 0.36131393909454346, "lr": 0.0007607837983149056, "grad_norm": 0.07991710305213928, "tokens": 182838, "tok_per_s": 387.0664967905951, "t": 472.83}26{"step": 1150, "loss": 0.43831026554107666, "lr": 0.0007389500957021191, "grad_norm": 0.15939369797706604, "tokens": 190769, "tok_per_s": 390.62090249954383, "t": 488.83}27{"step": 1200, "loss": 0.5175586342811584, "lr": 0.0007165621699029615, "grad_norm": 0.1856045126914978, "tokens": 199061, "tok_per_s": 395.13579275683645, "t": 504.24}28{"step": 1250, "loss": 0.3390142619609833, "lr": 0.000693685688561191, "grad_norm": 0.12978513538837433, "tokens": 207183, "tok_per_s": 398.39456509186255, "t": 520.5}29{"step": 1300, "loss": 0.5125949382781982, "lr": 0.0006703877523381495, "grad_norm": 0.09692317247390747, "tokens": 215576, "tok_per_s": 402.24649901158267, "t": 536.39}30{"step": 1350, "loss": 0.43845367431640625, "lr": 0.0006467366980949609, "grad_norm": 0.08677507936954498, "tokens": 223982, "tok_per_s": 406.1692861625648, "t": 551.91}31{"step": 1400, "loss": 0.3496563136577606, "lr": 0.0006228018984487442, "grad_norm": 0.17799852788448334, "tokens": 232005, "tok_per_s": 408.7535880259082, "t": 568.05}32{"step": 1450, "loss": 0.35586103796958923, "lr": 0.0005986535582907738, "grad_norm": 0.16138681769371033, "tokens": 239958, "tok_per_s": 411.73498766489666, "t": 583.26}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4566013887047239, "val_loss_b2": 0.4550019259990995, "val_loss_b4": 0.45443114651547917, "val_loss_b8": 0.4615690175860665, "t": 601.02}34{"step": 1500, "loss": 0.5495741963386536, "lr": 0.000574362508863438, "grad_norm": 0.2881101965904236, "tokens": 248410, "tok_per_s": 413.4240631708515, "t": 601.32}35{"step": 1550, "loss": 0.3491958975791931, "lr": 0.00055, "grad_norm": 0.10841384530067444, "tokens": 256596, "tok_per_s": 416.45360216178034, "t": 616.62}36{"step": 1600, "loss": 0.36749354004859924, "lr": 0.0005256374911365621, "grad_norm": 0.07231283187866211, "tokens": 264965, "tok_per_s": 419.52978215858195, "t": 632.03}37{"step": 1650, "loss": 0.45016416907310486, "lr": 0.0005013464417092262, "grad_norm": 0.10523931682109833, "tokens": 273349, "tok_per_s": 422.7300446960677, "t": 647.09}38{"step": 1700, "loss": 0.33430466055870056, "lr": 0.00047719810155125587, "grad_norm": 0.12313695251941681, "tokens": 281608, "tok_per_s": 425.13852770804283, "t": 662.85}39{"step": 1750, "loss": 0.4626242220401764, "lr": 0.0004532633019050392, "grad_norm": 0.1501327008008957, "tokens": 290020, "tok_per_s": 427.5290568914447, "t": 678.82}40{"step": 1800, "loss": 0.28245288133621216, "lr": 0.0004296122476618507, "grad_norm": 0.07538749277591705, "tokens": 298457, "tok_per_s": 430.2696747241201, "t": 694.11}41{"step": 1850, "loss": 0.3799683451652527, "lr": 0.00040631431143880916, "grad_norm": 0.09456677734851837, "tokens": 306802, "tok_per_s": 432.819384712973, "t": 709.3}42{"step": 1900, "loss": 0.3121398389339447, "lr": 0.00038343783009703845, "grad_norm": 0.08691007643938065, "tokens": 314844, "tok_per_s": 434.97351580115725, "t": 724.28}43{"step": 1950, "loss": 0.3585811257362366, "lr": 0.000361049904297881, "grad_norm": 0.13431037962436676, "tokens": 323167, "tok_per_s": 437.32580955801467, "t": 739.42}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45512917877242387, "val_loss_b2": 0.45211709456352134, "val_loss_b4": 0.45086460480142426, "val_loss_b8": 0.4578209910385711, "t": 758.67}45{"step": 2000, "loss": 0.45094215869903564, "lr": 0.0003392162016850945, "grad_norm": 0.14535140991210938, "tokens": 331553, "tok_per_s": 437.10003617233207, "t": 758.99}46{"step": 2050, "loss": 0.43527212738990784, "lr": 0.0003180007642703402, "grad_norm": 0.10725580155849457, "tokens": 339799, "tok_per_s": 439.17192012613424, "t": 774.19}47{"step": 2100, "loss": 0.33179789781570435, "lr": 0.000297465820586928, "grad_norm": 0.08665356040000916, "tokens": 348148, "tok_per_s": 441.2237225657991, "t": 789.51}48{"step": 2150, "loss": 0.3704075217247009, "lr": 0.0002776716031628058, "grad_norm": 0.1106739342212677, "tokens": 356667, "tok_per_s": 443.3363079600697, "t": 804.96}49{"step": 2200, "loss": 0.28455647826194763, "lr": 0.00025867617184817756, "grad_norm": 0.1338651180267334, "tokens": 365197, "tok_per_s": 445.4397745206811, "t": 820.31}50{"step": 2250, "loss": 0.47649896144866943, "lr": 0.00024053524351595949, "grad_norm": 0.2042640745639801, "tokens": 373572, "tok_per_s": 447.2489438151145, "t": 835.72}51{"step": 2300, "loss": 0.3501816391944885, "lr": 0.0002233020286345913, "grad_norm": 0.23293881118297577, "tokens": 381835, "tok_per_s": 448.6506508212014, "t": 851.53}52{"step": 2350, "loss": 0.47184765338897705, "lr": 0.00020702707519256364, "grad_norm": 0.14280860126018524, "tokens": 390126, "tok_per_s": 450.1907304357378, "t": 867.04}53{"step": 2400, "loss": 0.3817217946052551, "lr": 0.00019175812043246033, "grad_norm": 0.13446959853172302, "tokens": 398650, "tok_per_s": 452.1822295245753, "t": 882.07}54{"step": 2450, "loss": 0.3624129593372345, "lr": 0.00017753995082939931, "grad_norm": 0.1451910436153412, "tokens": 406902, "tok_per_s": 453.416606319051, "t": 897.87}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45296013701896237, "val_loss_b2": 0.4492219574290464, "val_loss_b4": 0.4472375740964992, "val_loss_b8": 0.45367186759245576, "t": 916.48}56{"step": 2500, "loss": 0.4591003954410553, "lr": 0.00016441427072458494, "grad_norm": 0.12424816936254501, "tokens": 415163, "tok_per_s": 448.60437903682345, "t": 925.91}57{"step": 2550, "loss": 0.3386523127555847, "lr": 0.0001524195799992897, "grad_norm": 0.14101435244083405, "tokens": 423133, "tok_per_s": 449.8828689358959, "t": 941.0}58{"step": 2600, "loss": 0.3013549745082855, "lr": 0.0001415910611480694, "grad_norm": 0.1469569355249405, "tokens": 431396, "tok_per_s": 451.6122835135839, "t": 955.69}59{"step": 2650, "loss": 0.41308698058128357, "lr": 0.00013196047608244387, "grad_norm": 0.19445662200450897, "tokens": 439671, "tok_per_s": 453.29600636995156, "t": 970.4}60{"step": 2700, "loss": 0.37488898634910583, "lr": 0.00012355607296773896, "grad_norm": 0.0984608456492424, "tokens": 448101, "tok_per_s": 454.66135603176326, "t": 986.03}61{"step": 2750, "loss": 0.502743124961853, "lr": 0.00011640250336634969, "grad_norm": 0.10401865094900131, "tokens": 456477, "tok_per_s": 456.4250762152752, "t": 1000.57}62{"step": 2800, "loss": 0.39014238119125366, "lr": 0.00011052074993046101, "grad_norm": 0.10965035855770111, "tokens": 464767, "tok_per_s": 457.52776916745677, "t": 1016.28}63{"step": 2850, "loss": 0.3458496332168579, "lr": 0.00010592806485631325, "grad_norm": 0.1411612331867218, "tokens": 473172, "tok_per_s": 459.32526684300507, "t": 1030.6}64{"step": 2900, "loss": 0.36798033118247986, "lr": 0.00010263791928053818, "grad_norm": 0.17196622490882874, "tokens": 481142, "tok_per_s": 460.5459922078011, "t": 1045.18}65{"step": 2950, "loss": 0.4091894030570984, "lr": 0.0001006599637669943, "grad_norm": 0.12208164483308792, "tokens": 489885, "tok_per_s": 462.1181634210887, "t": 1060.54}66{"step": 2999, "loss": 0.3905685544013977, "lr": 0.0001000002640500325, "grad_norm": 0.1331799030303955, "tokens": 497821, "tok_per_s": 463.08148035857977, "t": 1075.48}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45222347366686466, "val_loss_b2": 0.44876532932994045, "val_loss_b4": 0.44669421098689366, "val_loss_b8": 0.4537831960257901, "t": 1078.28}68{"step": 2999, "checkpoint_sha256": "65045ef16316b83087e70da42e8febdbff86fe691e7547db981b3b3e65988cba", "t": 1078.38}69
Files · main
Menu

Theme