SPB Git

spb/wp31_uqo

Public
1{"step": 0, "loss": 10.542369842529297, "lr": 2e-06, "grad_norm": 9.467050552368164, "tokens": 16384, "tok_per_s": 12518.17567258382, "t": 1.72}2{"step": 20, "loss": 9.081543922424316, "lr": 4.2e-05, "grad_norm": 2.370810031890869, "tokens": 344064, "tok_per_s": 14756.962660925023, "t": 23.73}3{"step": 40, "loss": 8.07042121887207, "lr": 8.199999999999999e-05, "grad_norm": 1.763466477394104, "tokens": 671744, "tok_per_s": 14817.808890631019, "t": 45.74}4{"step": 60, "loss": 7.185491561889648, "lr": 0.00012199999999999998, "grad_norm": 1.0985190868377686, "tokens": 999424, "tok_per_s": 14839.553963122858, "t": 67.76}5{"step": 80, "loss": 6.902057647705078, "lr": 0.00016199999999999998, "grad_norm": 0.8512132167816162, "tokens": 1327104, "tok_per_s": 14849.7726567203, "t": 89.78}6{"step": 100, "loss": 6.584644317626953, "lr": 0.00020199999999999998, "grad_norm": 0.9913674592971802, "tokens": 1654784, "tok_per_s": 14854.402984141629, "t": 111.81}7{"step": 120, "loss": 6.514517307281494, "lr": 0.000242, "grad_norm": 1.3175252676010132, "tokens": 1982464, "tok_per_s": 14856.284552708043, "t": 133.85}8{"step": 140, "loss": 6.289740562438965, "lr": 0.00028199999999999997, "grad_norm": 0.9145916700363159, "tokens": 2310144, "tok_per_s": 14857.504485863466, "t": 155.9}9{"step": 160, "loss": 6.1614789962768555, "lr": 0.00032199999999999997, "grad_norm": 1.1043585538864136, "tokens": 2637824, "tok_per_s": 14858.175647187973, "t": 177.94}10{"step": 180, "loss": 6.154912948608398, "lr": 0.00036199999999999996, "grad_norm": 0.971438467502594, "tokens": 2965504, "tok_per_s": 14858.3693930149, "t": 200.0}11{"step": 200, "loss": 6.065349578857422, "lr": 0.00040199999999999996, "grad_norm": 0.9000188112258911, "tokens": 3293184, "tok_per_s": 14858.81220114511, "t": 222.04}12{"step": 220, "loss": 5.964507102966309, "lr": 0.000442, "grad_norm": 0.7994388937950134, "tokens": 3620864, "tok_per_s": 14858.817231556903, "t": 244.1}13{"step": 240, "loss": 5.9960126876831055, "lr": 0.00048199999999999995, "grad_norm": 1.1917451620101929, "tokens": 3948544, "tok_per_s": 14859.295182508207, "t": 266.14}14{"step": 260, "loss": 5.876160621643066, "lr": 0.000522, "grad_norm": 0.845103919506073, "tokens": 4276224, "tok_per_s": 14859.866591265742, "t": 288.18}15{"step": 280, "loss": 5.791243553161621, "lr": 0.0005619999999999999, "grad_norm": 0.8462221622467041, "tokens": 4603904, "tok_per_s": 14859.888877096422, "t": 310.23}16{"step": 300, "loss": 5.73251485824585, "lr": 0.0006, "grad_norm": 0.7725811004638672, "tokens": 4931584, "tok_per_s": 14860.177291502676, "t": 332.28}17{"step": 320, "loss": 5.770495414733887, "lr": 0.0005999835963914726, "grad_norm": 2.257737159729004, "tokens": 5259264, "tok_per_s": 14860.281383900628, "t": 354.33}18{"step": 340, "loss": 5.756847381591797, "lr": 0.0005999343875590635, "grad_norm": 0.6835245490074158, "tokens": 5586944, "tok_per_s": 14860.270049295179, "t": 376.38}19{"step": 360, "loss": 5.568480491638184, "lr": 0.00059985237948205, "grad_norm": 0.6978513598442078, "tokens": 5914624, "tok_per_s": 14860.510090654503, "t": 398.42}20{"step": 380, "loss": 5.537266254425049, "lr": 0.0005997375821250868, "grad_norm": 0.7629163861274719, "tokens": 6242304, "tok_per_s": 14860.578873273127, "t": 420.47}21{"step": 400, "loss": 5.538397312164307, "lr": 0.0005995900094369953, "grad_norm": 1.092633605003357, "tokens": 6569984, "tok_per_s": 14860.857465748079, "t": 442.51}22{"step": 420, "loss": 5.525542259216309, "lr": 0.000599409679349069, "grad_norm": 0.6111159324645996, "tokens": 6897664, "tok_per_s": 14861.061886563317, "t": 464.55}23{"step": 440, "loss": 5.389138221740723, "lr": 0.0005991966137728945, "grad_norm": 0.7193952798843384, "tokens": 7225344, "tok_per_s": 14861.07333369556, "t": 486.6}24{"step": 460, "loss": 5.280256748199463, "lr": 0.0005989508385976885, "grad_norm": 0.616712749004364, "tokens": 7553024, "tok_per_s": 14861.357357465233, "t": 508.64}25{"step": 480, "loss": 5.338939666748047, "lr": 0.000598672383687153, "grad_norm": 0.6282134652137756, "tokens": 7880704, "tok_per_s": 14861.520408374818, "t": 530.69}26{"step": 499, "val_loss": 5.384630274772644, "val_ppl": 218.02947820527342, "tokens": 8192000, "t": 557.5}27{"step": 500, "loss": 5.339119911193848, "lr": 0.0005983612828758463, "grad_norm": 0.6352164149284363, "tokens": 8208384, "tok_per_s": 14705.160066883498, "t": 558.61}28{"step": 520, "loss": 5.392245292663574, "lr": 0.0005980175739650714, "grad_norm": 0.6777904629707336, "tokens": 8536064, "tok_per_s": 14711.270923250451, "t": 580.65}29{"step": 540, "loss": 5.308452606201172, "lr": 0.0005976412987182833, "grad_norm": 0.6433130502700806, "tokens": 8863744, "tok_per_s": 14716.732355532162, "t": 602.7}30{"step": 560, "loss": 5.355132579803467, "lr": 0.000597232502856014, "grad_norm": 0.6792540550231934, "tokens": 9191424, "tok_per_s": 14721.982681231935, "t": 624.74}31{"step": 580, "loss": 5.169951438903809, "lr": 0.0005967912360503179, "grad_norm": 0.5354589223861694, "tokens": 9519104, "tok_per_s": 14726.892969702027, "t": 646.79}32{"step": 600, "loss": 5.377242088317871, "lr": 0.000596317551918735, "grad_norm": 0.5585838556289673, "tokens": 9846784, "tok_per_s": 14731.31330089456, "t": 668.84}33{"step": 620, "loss": 5.250103950500488, "lr": 0.0005958115080177768, "grad_norm": 0.5638961791992188, "tokens": 10174464, "tok_per_s": 14735.5985884746, "t": 690.88}34{"step": 640, "loss": 5.222980976104736, "lr": 0.0005952731658359328, "grad_norm": 0.6217764019966125, "tokens": 10502144, "tok_per_s": 14739.543879023495, "t": 712.93}35{"step": 660, "loss": 5.158720016479492, "lr": 0.0005947025907861984, "grad_norm": 0.6102428436279297, "tokens": 10829824, "tok_per_s": 14743.329343185296, "t": 734.97}36{"step": 680, "loss": 5.209547996520996, "lr": 0.0005940998521981274, "grad_norm": 0.6341832280158997, "tokens": 11157504, "tok_per_s": 14746.497313342586, "t": 757.03}37{"step": 700, "loss": 5.143675804138184, "lr": 0.0005934650233094076, "grad_norm": 0.5422422885894775, "tokens": 11485184, "tok_per_s": 14749.784421759095, "t": 779.08}38{"step": 720, "loss": 5.103882789611816, "lr": 0.0005927981812569615, "grad_norm": 0.5489329099655151, "tokens": 11812864, "tok_per_s": 14752.91295069364, "t": 801.13}39{"step": 740, "loss": 5.018136978149414, "lr": 0.0005920994070675743, "grad_norm": 0.5919033885002136, "tokens": 12140544, "tok_per_s": 14755.859709459248, "t": 823.17}40{"step": 760, "loss": 4.959678649902344, "lr": 0.000591368785648048, "grad_norm": 0.5504949688911438, "tokens": 12468224, "tok_per_s": 14758.678968518878, "t": 845.22}41{"step": 780, "loss": 5.08703088760376, "lr": 0.0005906064057748841, "grad_norm": 0.5254867672920227, "tokens": 12795904, "tok_per_s": 14761.443030495608, "t": 867.26}42{"step": 800, "loss": 5.257801532745361, "lr": 0.0005898123600834975, "grad_norm": 2.3072996139526367, "tokens": 13123584, "tok_per_s": 14763.990226726408, "t": 889.3}43{"step": 820, "loss": 5.0983734130859375, "lr": 0.00058898674505696, "grad_norm": 0.6171005368232727, "tokens": 13451264, "tok_per_s": 14766.380258070778, "t": 911.35}44{"step": 840, "loss": 4.960577487945557, "lr": 0.0005881296610142765, "grad_norm": 0.5132812857627869, "tokens": 13778944, "tok_per_s": 14768.630232003112, "t": 933.4}45{"step": 860, "loss": 4.998333930969238, "lr": 0.0005872412120981961, "grad_norm": 0.559632420539856, "tokens": 14106624, "tok_per_s": 14770.709640457204, "t": 955.45}46{"step": 880, "loss": 4.7763671875, "lr": 0.0005863215062625574, "grad_norm": 0.5542692542076111, "tokens": 14434304, "tok_per_s": 14772.791555069703, "t": 977.5}47{"step": 900, "loss": 5.003454208374023, "lr": 0.0005853706552591712, "grad_norm": 0.5555323958396912, "tokens": 14761984, "tok_per_s": 14774.674409206718, "t": 999.55}48{"step": 920, "loss": 4.8749589920043945, "lr": 0.0005843887746242426, "grad_norm": 0.5380773544311523, "tokens": 15089664, "tok_per_s": 14776.584409694142, "t": 1021.6}49{"step": 940, "loss": 4.8660478591918945, "lr": 0.000583375983664331, "grad_norm": 0.5357463359832764, "tokens": 15417344, "tok_per_s": 14778.141683205138, "t": 1043.66}50{"step": 960, "loss": 4.947446823120117, "lr": 0.0005823324054418544, "grad_norm": 0.5279881358146667, "tokens": 15745024, "tok_per_s": 14779.922187023618, "t": 1065.71}51{"step": 980, "loss": 5.009394645690918, "lr": 0.0005812581667601359, "grad_norm": 1.623464822769165, "tokens": 16072704, "tok_per_s": 14781.592226630288, "t": 1087.76}52{"step": 999, "val_loss": 4.899298691749573, "val_ppl": 134.19563417086596, "tokens": 16384000, "t": 1114.5}53{"step": 999, "checkpoint_sha256": "f479d3fef1bc2500f36bb651ad640552133e352a95dd0352395afc7fe3779f34", "t": 1114.83}54{"step": 1000, "loss": 4.766724586486816, "lr": 0.0005801533981479964, "grad_norm": 0.5584408640861511, "tokens": 16400384, "tok_per_s": 14701.803764462587, "t": 1115.95}55{"step": 1020, "loss": 4.989593029022217, "lr": 0.0005790182338438938, "grad_norm": 0.6099668145179749, "tokens": 16728064, "tok_per_s": 14705.046079350384, "t": 1137.98}56{"step": 1040, "loss": 4.925856590270996, "lr": 0.0005778528117796123, "grad_norm": 0.4994802474975586, "tokens": 17055744, "tok_per_s": 14708.185159614715, "t": 1160.02}57{"step": 1060, "loss": 4.9076433181762695, "lr": 0.0005766572735635022, "grad_norm": 0.7315026521682739, "tokens": 17383424, "tok_per_s": 14711.083121870011, "t": 1182.07}58{"step": 1080, "loss": 5.0363054275512695, "lr": 0.0005754317644632738, "grad_norm": 0.5618871450424194, "tokens": 17711104, "tok_per_s": 14713.991330044195, "t": 1204.1}59{"step": 1100, "loss": 4.861031532287598, "lr": 0.000574176433388346, "grad_norm": 0.5686145424842834, "tokens": 18038784, "tok_per_s": 14716.74485417585, "t": 1226.14}60{"step": 1120, "loss": 4.877140045166016, "lr": 0.0005728914328717517, "grad_norm": 0.5235225558280945, "tokens": 18366464, "tok_per_s": 14719.333661142755, "t": 1248.19}61{"step": 1140, "loss": 4.685320854187012, "lr": 0.000571576919051606, "grad_norm": 0.5747607350349426, "tokens": 18694144, "tok_per_s": 14721.951564398301, "t": 1270.23}62{"step": 1160, "loss": 4.799270153045654, "lr": 0.0005702330516521316, "grad_norm": 0.5035359859466553, "tokens": 19021824, "tok_per_s": 14724.419744412917, "t": 1292.27}63{"step": 1180, "loss": 4.705811500549316, "lr": 0.0005688599939642524, "grad_norm": 0.5228400230407715, "tokens": 19349504, "tok_per_s": 14726.804454243698, "t": 1314.31}64{"step": 1200, "loss": 4.795140743255615, "lr": 0.000567457912825752, "grad_norm": 0.5400631427764893, "tokens": 19677184, "tok_per_s": 14729.200805270635, "t": 1336.34}65{"step": 1220, "loss": 4.857175350189209, "lr": 0.000566026978601002, "grad_norm": 0.5500919222831726, "tokens": 20004864, "tok_per_s": 14731.506994540023, "t": 1358.38}66{"step": 1240, "loss": 4.6607747077941895, "lr": 0.00056456736516026, "grad_norm": 0.6804667115211487, "tokens": 20332544, "tok_per_s": 14733.554321868092, "t": 1380.43}67{"step": 1260, "loss": 4.759555816650391, "lr": 0.000563079249858545, "grad_norm": 0.5324414372444153, "tokens": 20660224, "tok_per_s": 14663.607964964052, "t": 1409.36}68{"step": 1280, "loss": 4.592136383056641, "lr": 0.000561562813514085, "grad_norm": 0.5207154750823975, "tokens": 20987904, "tok_per_s": 14608.073268336082, "t": 1437.14}69{"step": 1300, "loss": 4.728357315063477, "lr": 0.0005600182403863482, "grad_norm": 0.5382718443870544, "tokens": 21315584, "tok_per_s": 14612.016585409512, "t": 1459.18}70{"step": 1320, "loss": 4.681814193725586, "lr": 0.0005584457181536525, "grad_norm": 0.539245069026947, "tokens": 21643264, "tok_per_s": 14615.794759381371, "t": 1481.22}71{"step": 1340, "loss": 4.577691555023193, "lr": 0.0005568454378903616, "grad_norm": 0.5490024089813232, "tokens": 21970944, "tok_per_s": 14619.47457468547, "t": 1503.27}72{"step": 1360, "loss": 4.614508628845215, "lr": 0.0005552175940436679, "grad_norm": 0.5317873954772949, "tokens": 22298624, "tok_per_s": 14623.020864363956, "t": 1525.31}73{"step": 1380, "loss": 4.631039619445801, "lr": 0.0005535623844099653, "grad_norm": 0.5490886569023132, "tokens": 22626304, "tok_per_s": 14626.499792426104, "t": 1547.35}74{"step": 1400, "loss": 4.680744647979736, "lr": 0.0005518800101108161, "grad_norm": 0.5449748635292053, "tokens": 22953984, "tok_per_s": 14629.83098998142, "t": 1569.4}75{"step": 1420, "loss": 4.541924476623535, "lr": 0.000550170675568512, "grad_norm": 0.5723252892494202, "tokens": 23281664, "tok_per_s": 14633.133304535308, "t": 1591.44}76{"step": 1440, "loss": 4.568117141723633, "lr": 0.0005484345884812357, "grad_norm": 0.5333711504936218, "tokens": 23609344, "tok_per_s": 14636.241945008496, "t": 1613.49}77{"step": 1460, "loss": 4.571066379547119, "lr": 0.000546671959797824, "grad_norm": 0.5370157957077026, "tokens": 23937024, "tok_per_s": 14639.262550571902, "t": 1635.54}78{"step": 1480, "loss": 4.515002727508545, "lr": 0.0005448830036921353, "grad_norm": 0.5326010584831238, "tokens": 24264704, "tok_per_s": 14642.267269777754, "t": 1657.58}79{"step": 1499, "val_loss": 4.569349575042724, "val_ppl": 96.48133549156312, "tokens": 24576000, "t": 1684.34}80{"step": 1500, "loss": 4.470322608947754, "lr": 0.0005430679375370263, "grad_norm": 0.5189173221588135, "tokens": 24592384, "tok_per_s": 14594.574947469444, "t": 1685.45}81{"step": 1520, "loss": 4.531128883361816, "lr": 0.0005412269818779387, "grad_norm": 0.539858341217041, "tokens": 24920064, "tok_per_s": 14598.070935873757, "t": 1707.49}82{"step": 1540, "loss": 4.4841108322143555, "lr": 0.0005393603604061018, "grad_norm": 0.5440859198570251, "tokens": 25247744, "tok_per_s": 14601.533837714102, "t": 1729.53}83{"step": 1560, "loss": 4.579304218292236, "lr": 0.0005374682999313519, "grad_norm": 2.1464147567749023, "tokens": 25575424, "tok_per_s": 14604.848583291969, "t": 1751.57}84{"step": 1580, "loss": 4.410983562469482, "lr": 0.000535551030354573, "grad_norm": 0.5440191030502319, "tokens": 25903104, "tok_per_s": 14608.0907553718, "t": 1773.61}85{"step": 1600, "loss": 4.463959693908691, "lr": 0.000533608784639762, "grad_norm": 0.5575087070465088, "tokens": 26230784, "tok_per_s": 14611.274003108054, "t": 1795.65}86{"step": 1620, "loss": 4.40781307220459, "lr": 0.0005316417987857215, "grad_norm": 0.5336335897445679, "tokens": 26558464, "tok_per_s": 14614.2827619718, "t": 1817.71}87{"step": 1640, "loss": 4.461612701416016, "lr": 0.0005296503117973845, "grad_norm": 0.525187075138092, "tokens": 26886144, "tok_per_s": 14617.273939479368, "t": 1839.75}88{"step": 1660, "loss": 4.4662370681762695, "lr": 0.0005276345656567725, "grad_norm": 0.555442750453949, "tokens": 27213824, "tok_per_s": 14620.198378827954, "t": 1861.8}89{"step": 1680, "loss": 4.632867813110352, "lr": 0.000525594805293594, "grad_norm": 0.5569077730178833, "tokens": 27541504, "tok_per_s": 14623.024012261158, "t": 1883.85}90{"step": 1700, "loss": 4.416189193725586, "lr": 0.0005235312785554818, "grad_norm": 0.507185161113739, "tokens": 27869184, "tok_per_s": 14625.849704185159, "t": 1905.89}91{"step": 1720, "loss": 4.504244804382324, "lr": 0.000521444236177879, "grad_norm": 0.5436880588531494, "tokens": 28196864, "tok_per_s": 14628.581916212159, "t": 1927.93}92{"step": 1740, "loss": 4.464587211608887, "lr": 0.0005193339317535718, "grad_norm": 0.5713441967964172, "tokens": 28524544, "tok_per_s": 14631.28682268555, "t": 1949.97}93{"step": 1760, "loss": 4.414894104003906, "lr": 0.0005172006217018763, "grad_norm": 0.5452090501785278, "tokens": 28852224, "tok_per_s": 14633.92213136254, "t": 1972.01}94{"step": 1780, "loss": 4.44191837310791, "lr": 0.0005150445652374809, "grad_norm": 0.5674588680267334, "tokens": 29179904, "tok_per_s": 14636.421850027471, "t": 1994.06}95{"step": 1800, "loss": 4.338931083679199, "lr": 0.0005128660243389501, "grad_norm": 0.5222944021224976, "tokens": 29507584, "tok_per_s": 14638.899424842659, "t": 2016.11}96{"step": 1820, "loss": 4.488123893737793, "lr": 0.0005106652637168917, "grad_norm": 0.5679968595504761, "tokens": 29835264, "tok_per_s": 14641.29489568718, "t": 2038.16}97{"step": 1840, "loss": 4.525457859039307, "lr": 0.0005084425507817925, "grad_norm": 0.5422910451889038, "tokens": 30162944, "tok_per_s": 14643.726405647612, "t": 2060.2}98{"step": 1860, "loss": 4.366527557373047, "lr": 0.0005061981556115255, "grad_norm": 0.5538021326065063, "tokens": 30490624, "tok_per_s": 14646.03697888566, "t": 2082.25}99{"step": 1880, "loss": 4.353416919708252, "lr": 0.0005039323509185333, "grad_norm": 0.5394396185874939, "tokens": 30818304, "tok_per_s": 14648.311639637379, "t": 2104.29}100{"step": 1900, "loss": 4.390727519989014, "lr": 0.0005016454120166914, "grad_norm": 0.5662208199501038, "tokens": 31145984, "tok_per_s": 14650.557015104174, "t": 2126.34}101{"step": 1920, "loss": 4.439477920532227, "lr": 0.0004993376167878549, "grad_norm": 0.5511561036109924, "tokens": 31473664, "tok_per_s": 14652.72967738711, "t": 2148.38}102{"step": 1940, "loss": 4.207109451293945, "lr": 0.0004970092456480945, "grad_norm": 0.5798001885414124, "tokens": 31801344, "tok_per_s": 14654.85450152307, "t": 2170.43}103{"step": 1960, "loss": 4.291384696960449, "lr": 0.0004946605815136225, "grad_norm": 0.5260293483734131, "tokens": 32129024, "tok_per_s": 14656.99246154965, "t": 2192.47}104{"step": 1980, "loss": 4.324018955230713, "lr": 0.0004922919097664167, "grad_norm": 0.5504523515701294, "tokens": 32456704, "tok_per_s": 14658.99723996013, "t": 2214.53}105{"step": 1999, "val_loss": 4.3897505640983585, "val_ppl": 80.62030687330494, "tokens": 32768000, "t": 2246.69}106{"step": 1999, "checkpoint_sha256": "cb967783bf811eb963dade3688b1cdfa188412f788c8095e4221da1feed064da", "t": 2247.09}107{"step": 2000, "loss": 4.260997295379639, "lr": 0.0004899035182195446, "grad_norm": 1.1323354244232178, "tokens": 32784384, "tok_per_s": 14581.471484153668, "t": 2248.77}108{"step": 2020, "loss": 4.394476413726807, "lr": 0.00048749569708219075, "grad_norm": 0.5696185231208801, "tokens": 33112064, "tok_per_s": 14553.638540512724, "t": 2275.59}109{"step": 2040, "loss": 4.135895252227783, "lr": 0.0004850687389243948, "grad_norm": 0.5677452087402344, "tokens": 33439744, "tok_per_s": 14520.895916178853, "t": 2303.28}110{"step": 2060, "loss": 4.396990776062012, "lr": 0.00048262293864150137, "grad_norm": 0.5605251789093018, "tokens": 33767424, "tok_per_s": 14498.895308151124, "t": 2329.38}111{"step": 2080, "loss": 4.291296005249023, "lr": 0.0004801585934183281, "grad_norm": 0.5201043486595154, "tokens": 34095104, "tok_per_s": 14463.951905940006, "t": 2357.66}112{"step": 2100, "loss": 4.347602844238281, "lr": 0.0004776760026930552, "grad_norm": 0.5589205026626587, "tokens": 34422784, "tok_per_s": 14435.854222527982, "t": 2384.95}113{"step": 2120, "loss": 4.387077331542969, "lr": 0.0004751754681208411, "grad_norm": 0.6422203183174133, "tokens": 34750464, "tok_per_s": 14408.91086610576, "t": 2412.15}114{"step": 2140, "loss": 4.296839714050293, "lr": 0.0004726572935371689, "grad_norm": 0.6081157922744751, "tokens": 35078144, "tok_per_s": 14380.022427355065, "t": 2439.78}115{"step": 2160, "loss": 4.319023132324219, "lr": 0.00047012178492092805, "grad_norm": 0.5665426254272461, "tokens": 35405824, "tok_per_s": 14349.25796299595, "t": 2467.84}116{"step": 2180, "loss": 4.389346599578857, "lr": 0.00046756925035723536, "grad_norm": 0.5185275673866272, "tokens": 35733504, "tok_per_s": 14320.333388262674, "t": 2495.71}117{"step": 2200, "loss": 4.380955696105957, "lr": 0.00046499999999999997, "grad_norm": 0.5352362990379333, "tokens": 36061184, "tok_per_s": 14302.079645978125, "t": 2521.81}118{"step": 2220, "loss": 4.433354377746582, "lr": 0.0004624143460342372, "grad_norm": 0.5563386678695679, "tokens": 36388864, "tok_per_s": 14272.893534600196, "t": 2549.92}119{"step": 2240, "loss": 4.406496524810791, "lr": 0.00045981260263813534, "grad_norm": 0.5777581334114075, "tokens": 36716544, "tok_per_s": 14249.042970080029, "t": 2577.18}120{"step": 2260, "loss": 4.153965950012207, "lr": 0.0004571950859448811, "grad_norm": 0.5618652105331421, "tokens": 37044224, "tok_per_s": 14225.867840229586, "t": 2604.42}121{"step": 2280, "loss": 4.321313381195068, "lr": 0.0004545621140042457, "grad_norm": 0.5788436532020569, "tokens": 37371904, "tok_per_s": 14205.233328074508, "t": 2631.27}122{"step": 2300, "loss": 4.281400680541992, "lr": 0.00045191400674394013, "grad_norm": 0.5440475940704346, "tokens": 37699584, "tok_per_s": 14178.960317657977, "t": 2659.25}123{"step": 2320, "loss": 4.224806785583496, "lr": 0.000449251085930741, "grad_norm": 0.5599042177200317, "tokens": 38027264, "tok_per_s": 14153.349602275077, "t": 2687.21}124{"step": 2340, "loss": 4.257534980773926, "lr": 0.00044657367513139303, "grad_norm": 0.5681010484695435, "tokens": 38354944, "tok_per_s": 14129.954697366495, "t": 2714.85}125{"step": 2360, "loss": 4.303826332092285, "lr": 0.00044388209967329364, "grad_norm": 0.5215063095092773, "tokens": 38682624, "tok_per_s": 14109.980932435801, "t": 2741.92}126{"step": 2380, "loss": 4.3118896484375, "lr": 0.00044117668660496233, "grad_norm": 0.5519024133682251, "tokens": 39010304, "tok_per_s": 14084.774881963896, "t": 2770.09}127{"step": 2400, "loss": 4.191535949707031, "lr": 0.00043845776465630177, "grad_norm": 0.5769554972648621, "tokens": 39337984, "tok_per_s": 14061.3505248443, "t": 2798.01}128{"step": 2420, "loss": 4.170306205749512, "lr": 0.00043572566419865466, "grad_norm": 0.5348464250564575, "tokens": 39665664, "tok_per_s": 14044.187505449492, "t": 2824.76}129{"step": 2440, "loss": 4.297189712524414, "lr": 0.0004329807172046608, "grad_norm": 0.5764148831367493, "tokens": 39993344, "tok_per_s": 14017.007871389693, "t": 2853.61}130{"step": 2460, "loss": 4.179870128631592, "lr": 0.00043022325720791954, "grad_norm": 0.5492531061172485, "tokens": 40321024, "tok_per_s": 13994.327824120344, "t": 2881.65}131{"step": 2480, "loss": 4.268969535827637, "lr": 0.000427453619262463, "grad_norm": 0.5527361035346985, "tokens": 40648704, "tok_per_s": 13974.012638576763, "t": 2909.29}132{"step": 2499, "val_loss": 4.259690976142883, "val_ppl": 70.78810486074492, "tokens": 40960000, "t": 2940.88}133{"step": 2500, "loss": 4.286783695220947, "lr": 0.00042467213990204363, "grad_norm": 0.5369290113449097, "tokens": 40976384, "tok_per_s": 13929.153702973366, "t": 2942.18}134{"step": 2520, "loss": 4.216705799102783, "lr": 0.0004218791570992436, "grad_norm": 0.5588297843933105, "tokens": 41304064, "tok_per_s": 13903.369027190794, "t": 2971.21}135{"step": 2540, "loss": 4.263204097747803, "lr": 0.00041907501022440735, "grad_norm": 1.1087560653686523, "tokens": 41631744, "tok_per_s": 13885.864933029952, "t": 2998.55}136{"step": 2560, "loss": 4.152766227722168, "lr": 0.00041626004000440593, "grad_norm": 0.579130232334137, "tokens": 41959424, "tok_per_s": 13869.14550116566, "t": 3025.79}137{"step": 2580, "loss": 4.249244689941406, "lr": 0.00041343458848123576, "grad_norm": 0.5325732827186584, "tokens": 42287104, "tok_per_s": 13852.602986292439, "t": 3053.06}138{"step": 2600, "loss": 4.2042036056518555, "lr": 0.0004105989989704575, "grad_norm": 0.5551273822784424, "tokens": 42614784, "tok_per_s": 13837.29165437411, "t": 3080.12}139{"step": 2620, "loss": 4.278999328613281, "lr": 0.0004077536160194806, "grad_norm": 0.5325561165809631, "tokens": 42942464, "tok_per_s": 13821.087744291774, "t": 3107.44}140{"step": 2640, "loss": 4.1617231369018555, "lr": 0.000404898785365698, "grad_norm": 0.5404963493347168, "tokens": 43270144, "tok_per_s": 13802.473083731951, "t": 3135.37}141{"step": 2660, "loss": 4.120279788970947, "lr": 0.000402034853894476, "grad_norm": 0.5360659956932068, "tokens": 43597824, "tok_per_s": 13792.592848030334, "t": 3161.37}142{"step": 2680, "loss": 4.048496246337891, "lr": 0.000399162169597005, "grad_norm": 0.5562931895256042, "tokens": 43925504, "tok_per_s": 13772.530405417207, "t": 3189.77}143{"step": 2700, "loss": 4.193229675292969, "lr": 0.00039628108152801574, "grad_norm": 0.5349346995353699, "tokens": 44253184, "tok_per_s": 13757.633206291655, "t": 3217.04}144{"step": 2720, "loss": 4.052156925201416, "lr": 0.000393391939763366, "grad_norm": 0.609096884727478, "tokens": 44580864, "tok_per_s": 13742.953874553468, "t": 3244.32}145{"step": 2740, "loss": 4.08882999420166, "lr": 0.00039049509535750383, "grad_norm": 0.5478166341781616, "tokens": 44908544, "tok_per_s": 13726.398704891522, "t": 3272.1}146{"step": 2760, "loss": 4.067444324493408, "lr": 0.0003875909003008116, "grad_norm": 0.540757954120636, "tokens": 45236224, "tok_per_s": 13712.552937935176, "t": 3299.3}147{"step": 2780, "loss": 4.157739162445068, "lr": 0.0003846797074768358, "grad_norm": 0.5378408432006836, "tokens": 45563904, "tok_per_s": 13699.672743484783, "t": 3326.32}148{"step": 2800, "loss": 4.173252105712891, "lr": 0.00038176187061940935, "grad_norm": 0.5366408824920654, "tokens": 45891584, "tok_per_s": 13683.169976206786, "t": 3354.28}149{"step": 2820, "loss": 4.161990165710449, "lr": 0.0003788377442696697, "grad_norm": 0.5339188575744629, "tokens": 46219264, "tok_per_s": 13669.013062098484, "t": 3381.73}150{"step": 2840, "loss": 4.140442848205566, "lr": 0.00037590768373297907, "grad_norm": 0.524897038936615, "tokens": 46546944, "tok_per_s": 13656.541010075069, "t": 3408.81}151{"step": 2860, "loss": 4.134394645690918, "lr": 0.0003729720450357526, "grad_norm": 0.5448781847953796, "tokens": 46874624, "tok_per_s": 13640.24174846482, "t": 3436.91}152{"step": 2880, "loss": 4.063270568847656, "lr": 0.0003700311848821973, "grad_norm": 0.5637195706367493, "tokens": 47202304, "tok_per_s": 13631.233473181625, "t": 3463.22}153{"step": 2900, "loss": 4.064417362213135, "lr": 0.00036708546061097017, "grad_norm": 0.5436610579490662, "tokens": 47529984, "tok_per_s": 13612.026342480698, "t": 3492.18}154{"step": 2920, "loss": 4.143337726593018, "lr": 0.00036413523015175846, "grad_norm": 0.5450679659843445, "tokens": 47857664, "tok_per_s": 13596.9191180996, "t": 3520.15}155{"step": 2940, "loss": 4.137192726135254, "lr": 0.0003611808519817883, "grad_norm": 0.559813916683197, "tokens": 48185344, "tok_per_s": 13583.946513667212, "t": 3547.64}156{"step": 2960, "loss": 4.054220199584961, "lr": 0.0003582226850822664, "grad_norm": 0.5458910465240479, "tokens": 48513024, "tok_per_s": 13576.655028416264, "t": 3573.68}157{"step": 2980, "loss": 4.1725873947143555, "lr": 0.0003552610888947618, "grad_norm": 0.5455508828163147, "tokens": 48840704, "tok_per_s": 13558.08693878946, "t": 3602.74}158{"step": 2999, "val_loss": 4.1647012591362, "val_ppl": 64.3734485909987, "tokens": 49152000, "t": 3635.12}159{"step": 2999, "checkpoint_sha256": "872a6b071e988f1a211c5af8da2569103e7ccf2be177e6a26e4218b8491f4045", "t": 3635.54}160{"step": 3000, "loss": 4.1911396980285645, "lr": 0.0003522964232775297, "grad_norm": 0.5839354395866394, "tokens": 49168384, "tok_per_s": 13521.072460528125, "t": 3636.84}161{"step": 3020, "loss": 4.099298477172852, "lr": 0.0003493290484617865, "grad_norm": 0.522006094455719, "tokens": 49496064, "tok_per_s": 13509.853726901843, "t": 3664.11}162{"step": 3040, "loss": 4.073601722717285, "lr": 0.0003463593250079383, "grad_norm": 0.5594094395637512, "tokens": 49823744, "tok_per_s": 13492.35579632998, "t": 3693.15}163{"step": 3060, "loss": 4.090610980987549, "lr": 0.00034338761376177, "grad_norm": 0.6630221009254456, "tokens": 50151424, "tok_per_s": 13482.320861969347, "t": 3720.2}164{"step": 3080, "loss": 4.148919105529785, "lr": 0.00034041427581059977, "grad_norm": 0.5785588622093201, "tokens": 50479104, "tok_per_s": 13472.833895169682, "t": 3747.14}165{"step": 3100, "loss": 4.058664798736572, "lr": 0.0003374396724394037, "grad_norm": 0.5250288248062134, "tokens": 50806784, "tok_per_s": 13459.681482147627, "t": 3775.15}166{"step": 3120, "loss": 4.0434160232543945, "lr": 0.0003344641650869168, "grad_norm": 0.5373675227165222, "tokens": 51134464, "tok_per_s": 13453.318027719777, "t": 3801.29}167{"step": 3140, "loss": 4.0473408699035645, "lr": 0.0003314881153017151, "grad_norm": 0.5367396473884583, "tokens": 51462144, "tok_per_s": 13435.196693427943, "t": 3830.81}168{"step": 3160, "loss": 4.105318069458008, "lr": 0.00032851188469828487, "grad_norm": 0.592815637588501, "tokens": 51789824, "tok_per_s": 13425.162579890346, "t": 3858.08}169{"step": 3180, "loss": 3.961264133453369, "lr": 0.0003255358349130832, "grad_norm": 0.5289068222045898, "tokens": 52117504, "tok_per_s": 13414.92196400839, "t": 3885.45}170{"step": 3200, "loss": 4.204805374145508, "lr": 0.00032256032756059627, "grad_norm": 0.5490139126777649, "tokens": 52445184, "tok_per_s": 13402.6920702635, "t": 3913.45}171{"step": 3220, "loss": 4.086887836456299, "lr": 0.00031958572418940007, "grad_norm": 0.5901429653167725, "tokens": 52772864, "tok_per_s": 13395.211384181153, "t": 3940.09}172{"step": 3240, "loss": 3.890124797821045, "lr": 0.00031661238623822995, "grad_norm": 0.5645272135734558, "tokens": 53100544, "tok_per_s": 13385.888604163558, "t": 3967.32}173{"step": 3260, "loss": 4.120551109313965, "lr": 0.00031364067499206177, "grad_norm": 0.543982744216919, "tokens": 53428224, "tok_per_s": 13374.151137731094, "t": 3995.3}174{"step": 3280, "loss": 4.0475993156433105, "lr": 0.00031067095153821354, "grad_norm": 0.5379183292388916, "tokens": 53755904, "tok_per_s": 13369.333554256866, "t": 4021.25}175{"step": 3300, "loss": 4.0359320640563965, "lr": 0.0003077035767224703, "grad_norm": 0.5553006529808044, "tokens": 54083584, "tok_per_s": 13356.484695835723, "t": 4049.65}176{"step": 3320, "loss": 4.045057773590088, "lr": 0.0003047389111052382, "grad_norm": 0.5813668966293335, "tokens": 54411264, "tok_per_s": 13347.337380932393, "t": 4076.97}177{"step": 3340, "loss": 3.9518637657165527, "lr": 0.0003017773149177336, "grad_norm": 0.5595154166221619, "tokens": 54738944, "tok_per_s": 13337.800086515115, "t": 4104.46}178{"step": 3360, "loss": 3.9641404151916504, "lr": 0.0002988191480182117, "grad_norm": 0.5525248646736145, "tokens": 55066624, "tok_per_s": 13326.42901880923, "t": 4132.55}179{"step": 3380, "loss": 4.014085292816162, "lr": 0.0002958647698482415, "grad_norm": 0.5643851161003113, "tokens": 55394304, "tok_per_s": 13317.299467612864, "t": 4159.99}180{"step": 3400, "loss": 3.9723434448242188, "lr": 0.0002929145393890298, "grad_norm": 0.5459938049316406, "tokens": 55721984, "tok_per_s": 13309.413220246935, "t": 4187.07}181{"step": 3420, "loss": 4.1072540283203125, "lr": 0.0002899688151178026, "grad_norm": 0.556307315826416, "tokens": 56049664, "tok_per_s": 13300.189584723837, "t": 4214.61}182{"step": 3440, "loss": 4.065242767333984, "lr": 0.00028702795496424736, "grad_norm": 0.544929563999176, "tokens": 56377344, "tok_per_s": 13294.878839934185, "t": 4240.94}183{"step": 3460, "loss": 3.9689781665802, "lr": 0.00028409231626702093, "grad_norm": 0.5853210687637329, "tokens": 56705024, "tok_per_s": 13283.969410219997, "t": 4269.09}184{"step": 3480, "loss": 3.992543935775757, "lr": 0.00028116225573033036, "grad_norm": 0.5510725975036621, "tokens": 57032704, "tok_per_s": 13275.58090874075, "t": 4296.47}185{"step": 3499, "val_loss": 4.082894301414489, "val_ppl": 59.316902636333495, "tokens": 57344000, "t": 4329.26}186{"step": 3500, "loss": 4.107724666595459, "lr": 0.00027823812938059065, "grad_norm": 0.5537106394767761, "tokens": 57360384, "tok_per_s": 13246.787853915583, "t": 4330.55}187{"step": 3520, "loss": 3.9655847549438477, "lr": 0.0002753202925231642, "grad_norm": 0.563191831111908, "tokens": 57688064, "tok_per_s": 13233.686251971329, "t": 4359.59}188{"step": 3540, "loss": 3.997711181640625, "lr": 0.0002724090996991884, "grad_norm": 0.5504301190376282, "tokens": 58015744, "tok_per_s": 13225.13390843726, "t": 4387.19}189{"step": 3560, "loss": 4.018125534057617, "lr": 0.00026950490464249606, "grad_norm": 0.5749231576919556, "tokens": 58343424, "tok_per_s": 13218.28618305372, "t": 4414.25}190{"step": 3580, "loss": 4.136887550354004, "lr": 0.0002666080602366339, "grad_norm": 0.5497174263000488, "tokens": 58671104, "tok_per_s": 13209.749127494764, "t": 4441.91}191{"step": 3600, "loss": 4.043684959411621, "lr": 0.00026371891847198426, "grad_norm": 0.5669506192207336, "tokens": 58998784, "tok_per_s": 13199.496622231001, "t": 4470.19}192{"step": 3620, "loss": 4.185938835144043, "lr": 0.00026083783040299483, "grad_norm": 0.5813359618186951, "tokens": 59326464, "tok_per_s": 13195.612898205272, "t": 4496.33}193{"step": 3640, "loss": 3.9637696743011475, "lr": 0.0002579651461055239, "grad_norm": 0.5515457987785339, "tokens": 59654144, "tok_per_s": 13182.852448762173, "t": 4525.54}194{"step": 3660, "loss": 3.7645978927612305, "lr": 0.000255101214634302, "grad_norm": 0.5694318413734436, "tokens": 59981824, "tok_per_s": 13174.971121473987, "t": 4553.12}195{"step": 3680, "loss": 4.017639636993408, "lr": 0.0002522463839805194, "grad_norm": 0.5420831441879272, "tokens": 60309504, "tok_per_s": 13164.942891109951, "t": 4581.48}196{"step": 3700, "loss": 3.94439959526062, "lr": 0.00024940100102954256, "grad_norm": 0.5583531260490417, "tokens": 60637184, "tok_per_s": 13158.29313527402, "t": 4608.7}197{"step": 3720, "loss": 3.909278392791748, "lr": 0.00024656541151876424, "grad_norm": 0.5652709007263184, "tokens": 60964864, "tok_per_s": 13152.489939120509, "t": 4635.64}198{"step": 3740, "loss": 4.000510215759277, "lr": 0.00024373995999559398, "grad_norm": 0.5588670372962952, "tokens": 61292544, "tok_per_s": 13149.087778936913, "t": 4661.76}199{"step": 3760, "loss": 3.9347519874572754, "lr": 0.0002409249897755926, "grad_norm": 0.5662419199943542, "tokens": 61620224, "tok_per_s": 13155.096244117349, "t": 4684.54}200{"step": 3780, "loss": 3.8995184898376465, "lr": 0.00023812084290075633, "grad_norm": 0.5653657913208008, "tokens": 61947904, "tok_per_s": 13141.412092741251, "t": 4714.36}201{"step": 3800, "loss": 4.045075416564941, "lr": 0.00023532786009795634, "grad_norm": 0.5492683053016663, "tokens": 62275584, "tok_per_s": 13128.293922315086, "t": 4744.03}202{"step": 3820, "loss": 4.045406341552734, "lr": 0.0002325463807375371, "grad_norm": 0.5755758881568909, "tokens": 62603264, "tok_per_s": 13116.777791866094, "t": 4773.17}203{"step": 3840, "loss": 3.9931726455688477, "lr": 0.00022977674279208034, "grad_norm": 0.5454151630401611, "tokens": 62930944, "tok_per_s": 13105.197141555022, "t": 4802.4}204{"step": 3860, "loss": 4.0127339363098145, "lr": 0.00022701928279533916, "grad_norm": 0.5690537691116333, "tokens": 63258624, "tok_per_s": 13094.299665342298, "t": 4831.42}205{"step": 3880, "loss": 4.082742691040039, "lr": 0.00022427433580134536, "grad_norm": 0.756371021270752, "tokens": 63586304, "tok_per_s": 13082.936615298882, "t": 4860.66}206{"step": 3900, "loss": 3.987462282180786, "lr": 0.0002215422353436983, "grad_norm": 0.557097852230072, "tokens": 63913984, "tok_per_s": 13074.749968177266, "t": 4888.76}207{"step": 3920, "loss": 3.88071346282959, "lr": 0.00021882331339503767, "grad_norm": 0.5510769486427307, "tokens": 64241664, "tok_per_s": 13057.698164268755, "t": 4920.24}208{"step": 3940, "loss": 3.9755029678344727, "lr": 0.00021611790032670633, "grad_norm": 0.5644064545631409, "tokens": 64569344, "tok_per_s": 13037.32301569676, "t": 4953.07}209{"step": 3960, "loss": 4.004444122314453, "lr": 0.0002134263248686069, "grad_norm": 0.5798656344413757, "tokens": 64897024, "tok_per_s": 13018.130992872395, "t": 4985.54}210{"step": 3980, "loss": 3.9098968505859375, "lr": 0.00021074891406925903, "grad_norm": 0.5548845529556274, "tokens": 65224704, "tok_per_s": 13000.977884101969, "t": 5017.32}211{"step": 3999, "val_loss": 4.018327379226685, "val_ppl": 55.608016873054595, "tokens": 65536000, "t": 5055.69}212{"step": 3999, "checkpoint_sha256": "05ec2d9bfdec9aa126836d14cdff1db381edacb93ba574ef5c7d0eca588ada03", "t": 5056.11}213{"step": 4000, "loss": 3.973355770111084, "lr": 0.00020808599325605992, "grad_norm": 0.5739210247993469, "tokens": 65552384, "tok_per_s": 12958.783404110945, "t": 5058.94}214{"step": 4020, "loss": 3.9233062267303467, "lr": 0.0002054378859957543, "grad_norm": 0.5633960366249084, "tokens": 65880064, "tok_per_s": 12941.843541404362, "t": 5090.88}215{"step": 4040, "loss": 3.9292502403259277, "lr": 0.00020280491405511884, "grad_norm": 0.5749434232711792, "tokens": 66207744, "tok_per_s": 12925.65400594016, "t": 5122.61}216{"step": 4060, "loss": 4.071193695068359, "lr": 0.00020018739736186444, "grad_norm": 0.5879931449890137, "tokens": 66535424, "tok_per_s": 12907.132046564762, "t": 5155.35}217{"step": 4080, "loss": 4.0840840339660645, "lr": 0.00019758565396576288, "grad_norm": 0.5677524209022522, "tokens": 66863104, "tok_per_s": 12889.240324649842, "t": 5187.92}218{"step": 4100, "loss": 3.9564380645751953, "lr": 0.00019500000000000005, "grad_norm": 0.589139461517334, "tokens": 67190784, "tok_per_s": 12872.436168905251, "t": 5220.16}219{"step": 4120, "loss": 3.8219876289367676, "lr": 0.0001924307496427646, "grad_norm": 0.5673111081123352, "tokens": 67518464, "tok_per_s": 12853.826172096204, "t": 5253.2}220{"step": 4140, "loss": 3.7888712882995605, "lr": 0.000189878215079072, "grad_norm": 0.5642023086547852, "tokens": 67846144, "tok_per_s": 12834.080800570478, "t": 5286.82}221{"step": 4160, "loss": 3.9583377838134766, "lr": 0.00018734270646283112, "grad_norm": 0.5670055747032166, "tokens": 68173824, "tok_per_s": 12819.193180141128, "t": 5318.52}222{"step": 4180, "loss": 3.8905670642852783, "lr": 0.0001848245318791589, "grad_norm": 0.557534396648407, "tokens": 68501504, "tok_per_s": 12804.44663861656, "t": 5350.23}223{"step": 4200, "loss": 3.8374507427215576, "lr": 0.00018232399730694481, "grad_norm": 0.5740233659744263, "tokens": 68829184, "tok_per_s": 12788.969941777226, "t": 5382.33}224{"step": 4220, "loss": 3.8952245712280273, "lr": 0.00017984140658167184, "grad_norm": 0.5662471652030945, "tokens": 69156864, "tok_per_s": 12768.44340394825, "t": 5416.65}225{"step": 4240, "loss": 3.8184738159179688, "lr": 0.00017737706135849857, "grad_norm": 0.5658909678459167, "tokens": 69484544, "tok_per_s": 12747.367654178182, "t": 5451.3}226{"step": 4260, "loss": 3.9444470405578613, "lr": 0.00017493126107560525, "grad_norm": 0.5692965388298035, "tokens": 69812224, "tok_per_s": 12728.21790182194, "t": 5485.25}227{"step": 4280, "loss": 3.960324287414551, "lr": 0.00017250430291780933, "grad_norm": 0.5702835321426392, "tokens": 70139904, "tok_per_s": 12711.167750496988, "t": 5518.39}228{"step": 4300, "loss": 4.07672643661499, "lr": 0.0001700964817804554, "grad_norm": 0.5966963171958923, "tokens": 70467584, "tok_per_s": 12699.332494061415, "t": 5549.33}229{"step": 4320, "loss": 3.9432711601257324, "lr": 0.00016770809023358322, "grad_norm": 0.5869051218032837, "tokens": 70795264, "tok_per_s": 12676.971750373574, "t": 5584.97}230{"step": 4340, "loss": 3.777398109436035, "lr": 0.00016533941848637757, "grad_norm": 0.5860775113105774, "tokens": 71122944, "tok_per_s": 12660.469199288213, "t": 5618.13}231{"step": 4360, "loss": 3.8525967597961426, "lr": 0.00016299075435190547, "grad_norm": 0.5613743662834167, "tokens": 71450624, "tok_per_s": 12639.787154055311, "t": 5653.25}232{"step": 4380, "loss": 3.9193055629730225, "lr": 0.000160662383212145, "grad_norm": 0.5697399973869324, "tokens": 71778304, "tok_per_s": 12624.829016232057, "t": 5685.9}233{"step": 4400, "loss": 3.9022228717803955, "lr": 0.00015835458798330867, "grad_norm": 0.5720086693763733, "tokens": 72105984, "tok_per_s": 12606.723607071865, "t": 5720.06}234{"step": 4420, "loss": 3.841949939727783, "lr": 0.00015606764908146668, "grad_norm": 0.5949207544326782, "tokens": 72433664, "tok_per_s": 12588.525778780986, "t": 5754.35}235{"step": 4440, "loss": 3.946150302886963, "lr": 0.00015380184438847444, "grad_norm": 0.5758242011070251, "tokens": 72761344, "tok_per_s": 12579.60068581591, "t": 5784.49}236{"step": 4460, "loss": 3.911789894104004, "lr": 0.0001515574492182075, "grad_norm": 0.5771458148956299, "tokens": 73089024, "tok_per_s": 12553.392908510106, "t": 5822.66}237{"step": 4480, "loss": 3.854954719543457, "lr": 0.00014933473628310837, "grad_norm": 0.5866302251815796, "tokens": 73416704, "tok_per_s": 12537.584398590208, "t": 5856.15}238{"step": 4499, "val_loss": 3.961595559120178, "val_ppl": 52.54109152262877, "tokens": 73728000, "t": 5895.67}239{"step": 4500, "loss": 3.911322593688965, "lr": 0.00014713397566104992, "grad_norm": 0.5657970309257507, "tokens": 73744384, "tok_per_s": 12506.69785094281, "t": 5896.8}240{"step": 4520, "loss": 3.9600625038146973, "lr": 0.0001449554347625191, "grad_norm": 0.5794343948364258, "tokens": 74072064, "tok_per_s": 12489.917514082079, "t": 5930.96}241{"step": 4540, "loss": 3.8118693828582764, "lr": 0.00014279937829812366, "grad_norm": 0.5654749274253845, "tokens": 74399744, "tok_per_s": 12477.405766202255, "t": 5963.17}242{"step": 4560, "loss": 3.7892842292785645, "lr": 0.00014066606824642808, "grad_norm": 0.5747179985046387, "tokens": 74727424, "tok_per_s": 12457.112910130205, "t": 5999.19}243{"step": 4580, "loss": 3.7530131340026855, "lr": 0.0001385557638221209, "grad_norm": 0.5684683322906494, "tokens": 75055104, "tok_per_s": 12440.844972916586, "t": 6033.37}244{"step": 4600, "loss": 3.8015036582946777, "lr": 0.0001364687214445182, "grad_norm": 0.5776253938674927, "tokens": 75382784, "tok_per_s": 12424.976151949311, "t": 6067.45}245{"step": 4620, "loss": 3.8649938106536865, "lr": 0.00013440519470640598, "grad_norm": 0.6453270316123962, "tokens": 75710464, "tok_per_s": 12408.981404244509, "t": 6101.67}246{"step": 4640, "loss": 3.9505836963653564, "lr": 0.00013236543434322733, "grad_norm": 0.5761963129043579, "tokens": 76038144, "tok_per_s": 12401.600488207972, "t": 6131.73}247{"step": 4660, "loss": 3.7686171531677246, "lr": 0.00013034968820261554, "grad_norm": 0.5987674593925476, "tokens": 76365824, "tok_per_s": 12379.189755057387, "t": 6169.3}248{"step": 4680, "loss": 3.838618755340576, "lr": 0.00012835820121427852, "grad_norm": 0.5700400471687317, "tokens": 76693504, "tok_per_s": 12365.426836816456, "t": 6202.67}249{"step": 4700, "loss": 3.849212408065796, "lr": 0.00012639121536023803, "grad_norm": 0.5732380747795105, "tokens": 77021184, "tok_per_s": 12347.070532904712, "t": 6238.43}250{"step": 4720, "loss": 3.8740367889404297, "lr": 0.00012444896964542695, "grad_norm": 0.5744330286979675, "tokens": 77348864, "tok_per_s": 12339.19973161873, "t": 6268.96}251{"step": 4740, "loss": 3.8502140045166016, "lr": 0.00012253170006864812, "grad_norm": 0.5890172123908997, "tokens": 77676544, "tok_per_s": 12324.102933659433, "t": 6303.23}252{"step": 4760, "loss": 3.96372127532959, "lr": 0.00012063963959389822, "grad_norm": 0.5833339095115662, "tokens": 78004224, "tok_per_s": 12309.280169935184, "t": 6337.44}253{"step": 4780, "loss": 3.836987018585205, "lr": 0.00011877301812206132, "grad_norm": 0.5671938061714172, "tokens": 78331904, "tok_per_s": 12296.813407633987, "t": 6370.51}254{"step": 4800, "loss": 3.869051933288574, "lr": 0.00011693206246297376, "grad_norm": 0.6167951822280884, "tokens": 78659584, "tok_per_s": 12280.351886255958, "t": 6405.73}255{"step": 4820, "loss": 3.7591121196746826, "lr": 0.00011511699630786462, "grad_norm": 0.5984916090965271, "tokens": 78987264, "tok_per_s": 12266.12661873517, "t": 6439.87}256{"step": 4840, "loss": 4.013945579528809, "lr": 0.00011332804020217591, "grad_norm": 0.5905060172080994, "tokens": 79314944, "tok_per_s": 12252.224374920499, "t": 6473.93}257{"step": 4860, "loss": 3.8013148307800293, "lr": 0.00011156541151876422, "grad_norm": 0.5933244228363037, "tokens": 79642624, "tok_per_s": 12253.364918810903, "t": 6500.06}258{"step": 4880, "loss": 3.915266990661621, "lr": 0.00010982932443148792, "grad_norm": 0.5768188834190369, "tokens": 79970304, "tok_per_s": 12249.093268150475, "t": 6529.09}259{"step": 4900, "loss": 3.825070381164551, "lr": 0.00010811998988918382, "grad_norm": 0.5960748195648193, "tokens": 80297984, "tok_per_s": 12247.08443436279, "t": 6556.91}260{"step": 4920, "loss": 3.8447461128234863, "lr": 0.00010643761559003459, "grad_norm": 0.5644418001174927, "tokens": 80625664, "tok_per_s": 12245.990867816028, "t": 6584.25}261{"step": 4940, "loss": 3.9009509086608887, "lr": 0.00010478240595633218, "grad_norm": 0.595284640789032, "tokens": 80953344, "tok_per_s": 12245.015043889613, "t": 6611.54}262{"step": 4960, "loss": 3.9704666137695312, "lr": 0.00010315456210963839, "grad_norm": 0.5907114148139954, "tokens": 81281024, "tok_per_s": 12243.3772981031, "t": 6639.19}263{"step": 4980, "loss": 3.7940030097961426, "lr": 0.0001015542818463474, "grad_norm": 0.5724697113037109, "tokens": 81608704, "tok_per_s": 12233.866709835484, "t": 6671.13}264{"step": 4999, "val_loss": 3.922847056388855, "val_ppl": 50.5441421465632, "tokens": 81920000, "t": 6702.01}265{"step": 4999, "checkpoint_sha256": "eb65d9fae42e1c8e95b655cb9013e9a7d0970d313acff882d01ca234835cf0a1", "t": 6702.43}266{"step": 5000, "loss": 3.9287710189819336, "lr": 9.998175961365173e-05, "grad_norm": 0.5912283062934875, "tokens": 81936384, "tok_per_s": 12223.2610543437, "t": 6703.73}267{"step": 5020, "loss": 3.8594186305999756, "lr": 9.843718648591489e-05, "grad_norm": 0.5823270082473755, "tokens": 82264064, "tok_per_s": 12202.025313319069, "t": 6742.25}268{"step": 5040, "loss": 3.8604159355163574, "lr": 9.692075014145502e-05, "grad_norm": 0.5919735431671143, "tokens": 82591744, "tok_per_s": 12198.515443239141, "t": 6771.05}269{"step": 5060, "loss": 3.7909655570983887, "lr": 9.543263483973994e-05, "grad_norm": 0.5849449634552002, "tokens": 82919424, "tok_per_s": 12196.053540330715, "t": 6799.28}270{"step": 5080, "loss": 3.895537853240967, "lr": 9.39730213989981e-05, "grad_norm": 0.5843760967254639, "tokens": 83247104, "tok_per_s": 12193.297946151717, "t": 6827.7}271{"step": 5100, "loss": 3.8515191078186035, "lr": 9.254208717424797e-05, "grad_norm": 0.5830948948860168, "tokens": 83574784, "tok_per_s": 12187.090711980496, "t": 6858.06}272{"step": 5120, "loss": 3.8083672523498535, "lr": 9.114000603574761e-05, "grad_norm": 0.5769238471984863, "tokens": 83902464, "tok_per_s": 12177.222164042445, "t": 6890.54}273{"step": 5140, "loss": 3.90797758102417, "lr": 8.976694834786837e-05, "grad_norm": 0.583653450012207, "tokens": 84230144, "tok_per_s": 12163.647252606203, "t": 6925.16}274{"step": 5160, "loss": 3.7567996978759766, "lr": 8.842308094839385e-05, "grad_norm": 0.5679790377616882, "tokens": 84557824, "tok_per_s": 12166.62827085087, "t": 6950.39}275{"step": 5180, "loss": 3.854881525039673, "lr": 8.710856712824807e-05, "grad_norm": 0.5798988938331604, "tokens": 84885504, "tok_per_s": 12136.371141146152, "t": 6994.72}276{"step": 5200, "loss": 3.7935805320739746, "lr": 8.582356661165409e-05, "grad_norm": 0.5701839327812195, "tokens": 85213184, "tok_per_s": 12127.712675333227, "t": 7026.73}277{"step": 5220, "loss": 3.7927613258361816, "lr": 8.456823553672612e-05, "grad_norm": 0.5661200881004333, "tokens": 85540864, "tok_per_s": 12121.086958792577, "t": 7057.61}278{"step": 5240, "loss": 3.864945650100708, "lr": 8.334272643649774e-05, "grad_norm": 0.5982127785682678, "tokens": 85868544, "tok_per_s": 12115.382471587436, "t": 7087.98}279{"step": 5260, "loss": 3.865525245666504, "lr": 8.214718822038769e-05, "grad_norm": 0.5944429039955139, "tokens": 86196224, "tok_per_s": 12109.637481238084, "t": 7118.4}280{"step": 5280, "loss": 3.8212974071502686, "lr": 8.098176615610612e-05, "grad_norm": 0.5823934674263, "tokens": 86523904, "tok_per_s": 12103.922038982491, "t": 7148.83}281{"step": 5300, "loss": 3.7305333614349365, "lr": 7.984660185200352e-05, "grad_norm": 0.5853962302207947, "tokens": 86851584, "tok_per_s": 12085.507074162282, "t": 7186.84}282{"step": 5320, "loss": 3.8025922775268555, "lr": 7.874183323986404e-05, "grad_norm": 0.583455502986908, "tokens": 87179264, "tok_per_s": 12048.374568561961, "t": 7236.18}283{"step": 5340, "loss": 3.857605218887329, "lr": 7.76675945581456e-05, "grad_norm": 0.5699940919876099, "tokens": 87506944, "tok_per_s": 12051.377082173132, "t": 7261.57}284{"step": 5360, "loss": 3.8012077808380127, "lr": 7.662401633566895e-05, "grad_norm": 0.5761530995368958, "tokens": 87834624, "tok_per_s": 12055.76563418795, "t": 7286.11}285{"step": 5380, "loss": 3.7067923545837402, "lr": 7.561122537575734e-05, "grad_norm": 0.5923271775245667, "tokens": 88162304, "tok_per_s": 12050.600273641838, "t": 7316.42}286{"step": 5400, "loss": 3.89043927192688, "lr": 7.462934474082865e-05, "grad_norm": 0.5811970233917236, "tokens": 88489984, "tok_per_s": 12049.62182726618, "t": 7344.21}287{"step": 5420, "loss": 4.051399230957031, "lr": 7.36784937374426e-05, "grad_norm": 0.6072420477867126, "tokens": 88817664, "tok_per_s": 12049.501394707426, "t": 7371.48}288{"step": 5440, "loss": 3.7444400787353516, "lr": 7.27587879018039e-05, "grad_norm": 0.5796211957931519, "tokens": 89145344, "tok_per_s": 12049.396128990667, "t": 7398.74}289{"step": 5460, "loss": 3.8443613052368164, "lr": 7.187033898572343e-05, "grad_norm": 0.591823935508728, "tokens": 89473024, "tok_per_s": 12049.038200606525, "t": 7426.15}290{"step": 5480, "loss": 3.7369208335876465, "lr": 7.101325494303994e-05, "grad_norm": 0.5897295475006104, "tokens": 89800704, "tok_per_s": 12040.80971778417, "t": 7458.44}291{"step": 5499, "val_loss": 3.893156719207764, "val_ppl": 49.065528410782115, "tokens": 90112000, "t": 7489.68}292{"step": 5500, "loss": 3.9190309047698975, "lr": 7.018763991650238e-05, "grad_norm": 0.5958195924758911, "tokens": 90128384, "tok_per_s": 12032.205143404432, "t": 7491.01}293{"step": 5520, "loss": 3.8355796337127686, "lr": 6.939359422511588e-05, "grad_norm": 0.5889571309089661, "tokens": 90456064, "tok_per_s": 12014.133164934934, "t": 7529.55}294{"step": 5540, "loss": 3.8890061378479004, "lr": 6.863121435195203e-05, "grad_norm": 0.5916159749031067, "tokens": 90783744, "tok_per_s": 12011.624127456229, "t": 7558.4}295{"step": 5560, "loss": 3.7868661880493164, "lr": 6.790059293242561e-05, "grad_norm": 0.5939796566963196, "tokens": 91111424, "tok_per_s": 12009.953057038161, "t": 7586.74}296{"step": 5580, "loss": 3.835353374481201, "lr": 6.720181874303845e-05, "grad_norm": 0.5831606984138489, "tokens": 91439104, "tok_per_s": 12008.322118897318, "t": 7615.06}297{"step": 5600, "loss": 3.811063289642334, "lr": 6.653497669059239e-05, "grad_norm": 0.5864984393119812, "tokens": 91766784, "tok_per_s": 12006.668733599558, "t": 7643.4}298{"step": 5620, "loss": 3.7606468200683594, "lr": 6.590014780187247e-05, "grad_norm": 0.5943107604980469, "tokens": 92094464, "tok_per_s": 12004.626521134052, "t": 7671.99}299{"step": 5640, "loss": 3.8167576789855957, "lr": 6.529740921380147e-05, "grad_norm": 0.5821703672409058, "tokens": 92422144, "tok_per_s": 11983.975662202585, "t": 7712.56}300{"step": 5660, "loss": 3.7937381267547607, "lr": 6.47268341640671e-05, "grad_norm": 0.5896317958831787, "tokens": 92749824, "tok_per_s": 11985.578898566824, "t": 7738.86}301{"step": 5680, "loss": 3.880380630493164, "lr": 6.418849198222306e-05, "grad_norm": 0.5858281254768372, "tokens": 93077504, "tok_per_s": 11976.679939894784, "t": 7771.97}302{"step": 5700, "loss": 3.8115720748901367, "lr": 6.368244808126497e-05, "grad_norm": 0.5889918208122253, "tokens": 93405184, "tok_per_s": 11952.473775285534, "t": 7815.13}303{"step": 5720, "loss": 3.790461301803589, "lr": 6.320876394968206e-05, "grad_norm": 0.5702701807022095, "tokens": 93732864, "tok_per_s": 11947.083448842632, "t": 7846.08}304{"step": 5740, "loss": 3.838794708251953, "lr": 6.276749714398588e-05, "grad_norm": 0.5970463752746582, "tokens": 94060544, "tok_per_s": 11942.545589916199, "t": 7876.5}305{"step": 5760, "loss": 3.82938289642334, "lr": 6.23587012817167e-05, "grad_norm": 0.6128417253494263, "tokens": 94388224, "tok_per_s": 11937.885297886773, "t": 7907.02}306{"step": 5780, "loss": 3.7248711585998535, "lr": 6.198242603492857e-05, "grad_norm": 0.59451824426651, "tokens": 94715904, "tok_per_s": 11933.298697891281, "t": 7937.52}307{"step": 5800, "loss": 3.885892868041992, "lr": 6.163871712415363e-05, "grad_norm": 0.6152052283287048, "tokens": 95043584, "tok_per_s": 11921.70173000662, "t": 7972.73}308{"step": 5820, "loss": 3.8548407554626465, "lr": 6.132761631284691e-05, "grad_norm": 0.5863385796546936, "tokens": 95371264, "tok_per_s": 11887.83372451629, "t": 8023.01}309{"step": 5840, "loss": 3.8154215812683105, "lr": 6.10491614023115e-05, "grad_norm": 0.5859852433204651, "tokens": 95698944, "tok_per_s": 11887.85623216603, "t": 8050.55}310{"step": 5860, "loss": 3.763838052749634, "lr": 6.08033862271055e-05, "grad_norm": 0.5976582169532776, "tokens": 96026624, "tok_per_s": 11891.649030150418, "t": 8075.54}311{"step": 5880, "loss": 3.920656204223633, "lr": 6.059032065093087e-05, "grad_norm": 0.5962430834770203, "tokens": 96354304, "tok_per_s": 11887.394847712649, "t": 8106.0}312{"step": 5900, "loss": 3.67857027053833, "lr": 6.040999056300463e-05, "grad_norm": 0.5943389534950256, "tokens": 96681984, "tok_per_s": 11887.10286751046, "t": 8133.76}313{"step": 5920, "loss": 3.8130691051483154, "lr": 6.026241787491321e-05, "grad_norm": 0.5948111414909363, "tokens": 97009664, "tok_per_s": 11887.472005241454, "t": 8161.08}314{"step": 5940, "loss": 3.8854312896728516, "lr": 6.014762051794997e-05, "grad_norm": 0.5870198011398315, "tokens": 97337344, "tok_per_s": 11887.825064797185, "t": 8188.4}315{"step": 5960, "loss": 3.81254243850708, "lr": 6.0065612440936474e-05, "grad_norm": 0.6020442843437195, "tokens": 97665024, "tok_per_s": 11887.77343844748, "t": 8216.0}316{"step": 5980, "loss": 3.7208523750305176, "lr": 6.001640360852739e-05, "grad_norm": 0.5870249271392822, "tokens": 97992704, "tok_per_s": 11882.545600859054, "t": 8247.19}317{"step": 5999, "loss": 3.8835692405700684, "lr": 6.000004100943553e-05, "grad_norm": 0.5731510519981384, "tokens": 98304000, "tok_per_s": 11883.078236132482, "t": 8273.02}318{"step": 5999, "val_loss": 3.874629867076874, "val_ppl": 48.164867588204885, "tokens": 98304000, "t": 8279.44}319{"step": 5999, "checkpoint_sha256": "1f05af70bb9223b63dacb765c1816212ec828ff6e81bf8c0d6d46f33ebc77db2", "t": 8279.79}320
Files · main
Menu

Theme