| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.435754189944134, |
| "eval_steps": 500, |
| "global_step": 900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.7872453421354294, |
| "epoch": 0.01596169193934557, |
| "grad_norm": 3.080909252166748, |
| "learning_rate": 1.1062285714285713e-05, |
| "loss": 0.4824, |
| "mean_token_accuracy": 0.8958534836769104, |
| "num_tokens": 622336.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.9108860075473786, |
| "epoch": 0.03192338387869114, |
| "grad_norm": 1.1728941202163696, |
| "learning_rate": 2.3353714285714284e-05, |
| "loss": 0.324, |
| "mean_token_accuracy": 0.9094921946525574, |
| "num_tokens": 1237848.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.955551016330719, |
| "epoch": 0.047885075818036714, |
| "grad_norm": 1.079156517982483, |
| "learning_rate": 3.564514285714286e-05, |
| "loss": 0.309, |
| "mean_token_accuracy": 0.9126040428876877, |
| "num_tokens": 1863643.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8611825913190841, |
| "epoch": 0.06384676775738228, |
| "grad_norm": 0.7616499066352844, |
| "learning_rate": 4.3019142803974194e-05, |
| "loss": 0.1433, |
| "mean_token_accuracy": 0.9484753370285034, |
| "num_tokens": 2469341.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.8195502609014511, |
| "epoch": 0.07980845969672785, |
| "grad_norm": 0.5501054525375366, |
| "learning_rate": 4.3009500394808225e-05, |
| "loss": 0.1646, |
| "mean_token_accuracy": 0.94924955368042, |
| "num_tokens": 3094629.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8249957531690597, |
| "epoch": 0.09577015163607343, |
| "grad_norm": 0.6578405499458313, |
| "learning_rate": 4.2989150506831564e-05, |
| "loss": 0.1438, |
| "mean_token_accuracy": 0.9527049332857132, |
| "num_tokens": 3692600.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8487251400947571, |
| "epoch": 0.11173184357541899, |
| "grad_norm": 0.7742117047309875, |
| "learning_rate": 4.2958106655488265e-05, |
| "loss": 0.1157, |
| "mean_token_accuracy": 0.9633159458637237, |
| "num_tokens": 4305846.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8406275361776352, |
| "epoch": 0.12769353551476456, |
| "grad_norm": 0.6870937347412109, |
| "learning_rate": 4.291638945865279e-05, |
| "loss": 0.0891, |
| "mean_token_accuracy": 0.9654581695795059, |
| "num_tokens": 4913310.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8179305374622345, |
| "epoch": 0.14365522745411013, |
| "grad_norm": 0.7981634736061096, |
| "learning_rate": 4.286402662293663e-05, |
| "loss": 0.133, |
| "mean_token_accuracy": 0.9567308723926544, |
| "num_tokens": 5533837.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.7972609519958496, |
| "epoch": 0.1596169193934557, |
| "grad_norm": 0.7271296977996826, |
| "learning_rate": 4.280105292528682e-05, |
| "loss": 0.0748, |
| "mean_token_accuracy": 0.9692588925361634, |
| "num_tokens": 6139238.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.7826209843158722, |
| "epoch": 0.17557861133280128, |
| "grad_norm": 0.9311319589614868, |
| "learning_rate": 4.2727510189888735e-05, |
| "loss": 0.0948, |
| "mean_token_accuracy": 0.9660487741231918, |
| "num_tokens": 6750942.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.792391562461853, |
| "epoch": 0.19154030327214686, |
| "grad_norm": 1.1329065561294556, |
| "learning_rate": 4.264344726038844e-05, |
| "loss": 0.1215, |
| "mean_token_accuracy": 0.9598832100629806, |
| "num_tokens": 7354162.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7968044668436051, |
| "epoch": 0.20750199521149243, |
| "grad_norm": 0.7501869797706604, |
| "learning_rate": 4.2548919967452986e-05, |
| "loss": 0.0926, |
| "mean_token_accuracy": 0.9672768324613571, |
| "num_tokens": 7972424.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7914539545774459, |
| "epoch": 0.22346368715083798, |
| "grad_norm": 0.4441527724266052, |
| "learning_rate": 4.2443991091690325e-05, |
| "loss": 0.0764, |
| "mean_token_accuracy": 0.9698785275220871, |
| "num_tokens": 8585919.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.7819067716598511, |
| "epoch": 0.23942537909018355, |
| "grad_norm": 0.3861187994480133, |
| "learning_rate": 4.232873032195334e-05, |
| "loss": 0.0846, |
| "mean_token_accuracy": 0.9694553077220917, |
| "num_tokens": 9187518.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7868628352880478, |
| "epoch": 0.2553870710295291, |
| "grad_norm": 0.7797110676765442, |
| "learning_rate": 4.2203214209055817e-05, |
| "loss": 0.1039, |
| "mean_token_accuracy": 0.9645863115787506, |
| "num_tokens": 9797114.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.7686551660299301, |
| "epoch": 0.2713487629688747, |
| "grad_norm": 0.4940663278102875, |
| "learning_rate": 4.2067526114930886e-05, |
| "loss": 0.0712, |
| "mean_token_accuracy": 0.972442090511322, |
| "num_tokens": 10405377.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7765703171491622, |
| "epoch": 0.28731045490822027, |
| "grad_norm": 0.23396383225917816, |
| "learning_rate": 4.192175615726602e-05, |
| "loss": 0.0852, |
| "mean_token_accuracy": 0.9674565583467484, |
| "num_tokens": 11004149.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7805891185998917, |
| "epoch": 0.3032721468475658, |
| "grad_norm": 0.4413723349571228, |
| "learning_rate": 4.1766001149651005e-05, |
| "loss": 0.0708, |
| "mean_token_accuracy": 0.9729034811258316, |
| "num_tokens": 11614790.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7965189158916474, |
| "epoch": 0.3192338387869114, |
| "grad_norm": 0.19197236001491547, |
| "learning_rate": 4.160036453727891e-05, |
| "loss": 0.1092, |
| "mean_token_accuracy": 0.9651999145746231, |
| "num_tokens": 12226653.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.824384006857872, |
| "epoch": 0.33519553072625696, |
| "grad_norm": 0.7662508487701416, |
| "learning_rate": 4.142495632824267e-05, |
| "loss": 0.0961, |
| "mean_token_accuracy": 0.9674228817224503, |
| "num_tokens": 12862678.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.8039055347442627, |
| "epoch": 0.35115722266560256, |
| "grad_norm": 0.9724727272987366, |
| "learning_rate": 4.123989302047276e-05, |
| "loss": 0.0934, |
| "mean_token_accuracy": 0.9662326157093049, |
| "num_tokens": 13478526.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.775772100687027, |
| "epoch": 0.3671189146049481, |
| "grad_norm": 0.25729894638061523, |
| "learning_rate": 4.1045297524364796e-05, |
| "loss": 0.0843, |
| "mean_token_accuracy": 0.9702257007360459, |
| "num_tokens": 14092792.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7595041334629059, |
| "epoch": 0.3830806065442937, |
| "grad_norm": 0.1401986926794052, |
| "learning_rate": 4.084129908114817e-05, |
| "loss": 0.1248, |
| "mean_token_accuracy": 0.9598525494337082, |
| "num_tokens": 14717341.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.786889699101448, |
| "epoch": 0.39904229848363926, |
| "grad_norm": 0.27574002742767334, |
| "learning_rate": 4.0628033177050074e-05, |
| "loss": 0.0745, |
| "mean_token_accuracy": 0.9735713273286819, |
| "num_tokens": 15340078.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7923183053731918, |
| "epoch": 0.41500399042298486, |
| "grad_norm": 0.47537916898727417, |
| "learning_rate": 4.040564145331186e-05, |
| "loss": 0.0717, |
| "mean_token_accuracy": 0.9725858896970749, |
| "num_tokens": 15960679.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.773842254281044, |
| "epoch": 0.4309656823623304, |
| "grad_norm": 0.2411717027425766, |
| "learning_rate": 4.0174271612117575e-05, |
| "loss": 0.0916, |
| "mean_token_accuracy": 0.9678393840789795, |
| "num_tokens": 16580367.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7822292029857636, |
| "epoch": 0.44692737430167595, |
| "grad_norm": 0.6152486205101013, |
| "learning_rate": 3.9934077318497085e-05, |
| "loss": 0.1304, |
| "mean_token_accuracy": 0.9576453745365143, |
| "num_tokens": 17188540.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.763131606578827, |
| "epoch": 0.46288906624102155, |
| "grad_norm": 0.8452078700065613, |
| "learning_rate": 3.968521809826893e-05, |
| "loss": 0.1053, |
| "mean_token_accuracy": 0.9649925410747529, |
| "num_tokens": 17805354.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7586877375841141, |
| "epoch": 0.4788507581803671, |
| "grad_norm": 0.7443876266479492, |
| "learning_rate": 3.942785923209072e-05, |
| "loss": 0.0856, |
| "mean_token_accuracy": 0.9694090068340302, |
| "num_tokens": 18406366.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7440236926078796, |
| "epoch": 0.4948124501197127, |
| "grad_norm": 0.15965542197227478, |
| "learning_rate": 3.9162171645687486e-05, |
| "loss": 0.0667, |
| "mean_token_accuracy": 0.973630714416504, |
| "num_tokens": 19006152.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.7528551459312439, |
| "epoch": 0.5107741420590582, |
| "grad_norm": 0.24820564687252045, |
| "learning_rate": 3.8888331796330785e-05, |
| "loss": 0.0634, |
| "mean_token_accuracy": 0.9752680540084839, |
| "num_tokens": 19625815.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.758511808514595, |
| "epoch": 0.5267358339984038, |
| "grad_norm": 0.1678536832332611, |
| "learning_rate": 3.8606521555644026e-05, |
| "loss": 0.0765, |
| "mean_token_accuracy": 0.9737292170524597, |
| "num_tokens": 20251706.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.8099140286445617, |
| "epoch": 0.5426975259377494, |
| "grad_norm": 0.12425480037927628, |
| "learning_rate": 3.831692808881186e-05, |
| "loss": 0.08, |
| "mean_token_accuracy": 0.9717760980129242, |
| "num_tokens": 20869928.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.8449289321899414, |
| "epoch": 0.5586592178770949, |
| "grad_norm": 0.3884945213794708, |
| "learning_rate": 3.801974373027381e-05, |
| "loss": 0.0655, |
| "mean_token_accuracy": 0.9746631622314453, |
| "num_tokens": 21487674.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.8119749128818512, |
| "epoch": 0.5746209098164405, |
| "grad_norm": 0.5517822504043579, |
| "learning_rate": 3.7715165855984715e-05, |
| "loss": 0.0889, |
| "mean_token_accuracy": 0.9673811107873916, |
| "num_tokens": 22101768.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.8082006275653839, |
| "epoch": 0.5905826017557861, |
| "grad_norm": 0.12656064331531525, |
| "learning_rate": 3.740339675232686e-05, |
| "loss": 0.1114, |
| "mean_token_accuracy": 0.9613101214170456, |
| "num_tokens": 22714228.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.7816001594066619, |
| "epoch": 0.6065442936951316, |
| "grad_norm": 0.18565888702869415, |
| "learning_rate": 3.7084643481760836e-05, |
| "loss": 0.0822, |
| "mean_token_accuracy": 0.9714775443077087, |
| "num_tokens": 23316960.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.7636521548032761, |
| "epoch": 0.6225059856344772, |
| "grad_norm": 0.13019737601280212, |
| "learning_rate": 3.67591177453043e-05, |
| "loss": 0.0896, |
| "mean_token_accuracy": 0.9703256696462631, |
| "num_tokens": 23923546.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.7770091861486434, |
| "epoch": 0.6384676775738228, |
| "grad_norm": 0.16999919712543488, |
| "learning_rate": 3.642703574193011e-05, |
| "loss": 0.0594, |
| "mean_token_accuracy": 0.9768979340791702, |
| "num_tokens": 24535110.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.7821112215518952, |
| "epoch": 0.6544293695131684, |
| "grad_norm": 0.16182664036750793, |
| "learning_rate": 3.6088618024977e-05, |
| "loss": 0.1196, |
| "mean_token_accuracy": 0.962785828113556, |
| "num_tokens": 25136436.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.784448316693306, |
| "epoch": 0.6703910614525139, |
| "grad_norm": 0.6256292462348938, |
| "learning_rate": 3.574408935566845e-05, |
| "loss": 0.1308, |
| "mean_token_accuracy": 0.9604980528354645, |
| "num_tokens": 25760373.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.7595613598823547, |
| "epoch": 0.6863527533918595, |
| "grad_norm": 0.11680064350366592, |
| "learning_rate": 3.539367855383673e-05, |
| "loss": 0.1065, |
| "mean_token_accuracy": 0.965737396478653, |
| "num_tokens": 26370923.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.785027676820755, |
| "epoch": 0.7023144453312051, |
| "grad_norm": 0.13679790496826172, |
| "learning_rate": 3.503761834595149e-05, |
| "loss": 0.0659, |
| "mean_token_accuracy": 0.9742470771074295, |
| "num_tokens": 26984485.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.7741981953382492, |
| "epoch": 0.7182761372705507, |
| "grad_norm": 0.13371747732162476, |
| "learning_rate": 3.4676145210553744e-05, |
| "loss": 0.1137, |
| "mean_token_accuracy": 0.9615134477615357, |
| "num_tokens": 27602613.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.7374123007059097, |
| "epoch": 0.7342378292098962, |
| "grad_norm": 0.4468773901462555, |
| "learning_rate": 3.4309499221197814e-05, |
| "loss": 0.0647, |
| "mean_token_accuracy": 0.9737134844064712, |
| "num_tokens": 28213445.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.7456966727972031, |
| "epoch": 0.7501995211492418, |
| "grad_norm": 0.12663891911506653, |
| "learning_rate": 3.3937923887005776e-05, |
| "loss": 0.0902, |
| "mean_token_accuracy": 0.9683217167854309, |
| "num_tokens": 28829251.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.759719917178154, |
| "epoch": 0.7661612130885874, |
| "grad_norm": 0.1270993947982788, |
| "learning_rate": 3.356166599094002e-05, |
| "loss": 0.064, |
| "mean_token_accuracy": 0.9761691361665725, |
| "num_tokens": 29447464.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.7696426391601563, |
| "epoch": 0.7821229050279329, |
| "grad_norm": 0.1482442021369934, |
| "learning_rate": 3.3180975425901635e-05, |
| "loss": 0.0936, |
| "mean_token_accuracy": 0.9671059608459472, |
| "num_tokens": 30059263.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.7510176807641983, |
| "epoch": 0.7980845969672785, |
| "grad_norm": 0.6420421004295349, |
| "learning_rate": 3.279610502876312e-05, |
| "loss": 0.0855, |
| "mean_token_accuracy": 0.9702295064926147, |
| "num_tokens": 30679582.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.7388219237327576, |
| "epoch": 0.8140462889066241, |
| "grad_norm": 0.1534568965435028, |
| "learning_rate": 3.2407310412446035e-05, |
| "loss": 0.0701, |
| "mean_token_accuracy": 0.9742257595062256, |
| "num_tokens": 31294619.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.7282670885324478, |
| "epoch": 0.8300079808459697, |
| "grad_norm": 0.5732345581054688, |
| "learning_rate": 3.201484979615482e-05, |
| "loss": 0.0936, |
| "mean_token_accuracy": 0.9658971399068832, |
| "num_tokens": 31910482.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.7382837921380997, |
| "epoch": 0.8459696727853152, |
| "grad_norm": 0.1135987788438797, |
| "learning_rate": 3.1618983833879674e-05, |
| "loss": 0.112, |
| "mean_token_accuracy": 0.9637796878814697, |
| "num_tokens": 32519113.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.7493024051189423, |
| "epoch": 0.8619313647246608, |
| "grad_norm": 0.13224604725837708, |
| "learning_rate": 3.121997544128238e-05, |
| "loss": 0.0822, |
| "mean_token_accuracy": 0.9707857817411423, |
| "num_tokens": 33129187.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.7391972035169602, |
| "epoch": 0.8778930566640064, |
| "grad_norm": 0.33319348096847534, |
| "learning_rate": 3.081808962107999e-05, |
| "loss": 0.0819, |
| "mean_token_accuracy": 0.9717306554317474, |
| "num_tokens": 33743757.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.7449708580970764, |
| "epoch": 0.8938547486033519, |
| "grad_norm": 0.13626126945018768, |
| "learning_rate": 3.0413593287042427e-05, |
| "loss": 0.0975, |
| "mean_token_accuracy": 0.9667524933815003, |
| "num_tokens": 34356378.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.7597370415925979, |
| "epoch": 0.9098164405426975, |
| "grad_norm": 0.14178575575351715, |
| "learning_rate": 3.0006755086720845e-05, |
| "loss": 0.0712, |
| "mean_token_accuracy": 0.9726598799228668, |
| "num_tokens": 34963300.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.746942538022995, |
| "epoch": 0.9257781324820431, |
| "grad_norm": 0.17223705351352692, |
| "learning_rate": 2.9597845223024483e-05, |
| "loss": 0.0768, |
| "mean_token_accuracy": 0.9728372275829316, |
| "num_tokens": 35570505.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.7469703108072281, |
| "epoch": 0.9417398244213887, |
| "grad_norm": 0.1207246407866478, |
| "learning_rate": 2.9187135274764504e-05, |
| "loss": 0.0801, |
| "mean_token_accuracy": 0.9725636392831802, |
| "num_tokens": 36186885.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.7389861464500427, |
| "epoch": 0.9577015163607342, |
| "grad_norm": 0.37582626938819885, |
| "learning_rate": 2.87748980162841e-05, |
| "loss": 0.0685, |
| "mean_token_accuracy": 0.9724197059869766, |
| "num_tokens": 36802324.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.7571559935808182, |
| "epoch": 0.9736632083000798, |
| "grad_norm": 0.10633117705583572, |
| "learning_rate": 2.836140723629446e-05, |
| "loss": 0.092, |
| "mean_token_accuracy": 0.9680209428071975, |
| "num_tokens": 37403435.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.7509357005357742, |
| "epoch": 0.9896249002394254, |
| "grad_norm": 0.4718145728111267, |
| "learning_rate": 2.7946937556037163e-05, |
| "loss": 0.1033, |
| "mean_token_accuracy": 0.9662729442119599, |
| "num_tokens": 38010569.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.7574707489264639, |
| "epoch": 1.0047885075818037, |
| "grad_norm": 0.11187744140625, |
| "learning_rate": 2.7531764246893585e-05, |
| "loss": 0.0671, |
| "mean_token_accuracy": 0.9744049498909398, |
| "num_tokens": 38577624.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.7245491325855256, |
| "epoch": 1.0207501995211492, |
| "grad_norm": 0.12900446355342865, |
| "learning_rate": 2.7116163047562487e-05, |
| "loss": 0.0545, |
| "mean_token_accuracy": 0.9777880311012268, |
| "num_tokens": 39188484.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6937095284461975, |
| "epoch": 1.036711891460495, |
| "grad_norm": 0.589491605758667, |
| "learning_rate": 2.6700409980927298e-05, |
| "loss": 0.0619, |
| "mean_token_accuracy": 0.9753036707639694, |
| "num_tokens": 39782239.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.6824752867221833, |
| "epoch": 1.0526735833998404, |
| "grad_norm": 0.13635259866714478, |
| "learning_rate": 2.62847811707346e-05, |
| "loss": 0.053, |
| "mean_token_accuracy": 0.9780564397573471, |
| "num_tokens": 40398532.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.6895065426826477, |
| "epoch": 1.068635275339186, |
| "grad_norm": 0.18689432740211487, |
| "learning_rate": 2.586955265820566e-05, |
| "loss": 0.0561, |
| "mean_token_accuracy": 0.9772377789020539, |
| "num_tokens": 41009440.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6825103610754013, |
| "epoch": 1.0845969672785316, |
| "grad_norm": 0.15807431936264038, |
| "learning_rate": 2.5455000218702754e-05, |
| "loss": 0.0579, |
| "mean_token_accuracy": 0.976485601067543, |
| "num_tokens": 41612304.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6815812021493912, |
| "epoch": 1.100558659217877, |
| "grad_norm": 0.34784135222435, |
| "learning_rate": 2.5041399178572077e-05, |
| "loss": 0.0544, |
| "mean_token_accuracy": 0.9776277542114258, |
| "num_tokens": 42222964.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6782410234212876, |
| "epoch": 1.1165203511572226, |
| "grad_norm": 0.48382070660591125, |
| "learning_rate": 2.4629024232284813e-05, |
| "loss": 0.0636, |
| "mean_token_accuracy": 0.975392273068428, |
| "num_tokens": 42842402.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.6576968163251877, |
| "epoch": 1.1324820430965683, |
| "grad_norm": 0.1805514246225357, |
| "learning_rate": 2.4218149259998005e-05, |
| "loss": 0.0732, |
| "mean_token_accuracy": 0.9726167500019074, |
| "num_tokens": 43440453.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6495650440454483, |
| "epoch": 1.1484437350359138, |
| "grad_norm": 0.5852122902870178, |
| "learning_rate": 2.380904714565604e-05, |
| "loss": 0.049, |
| "mean_token_accuracy": 0.9791654586791992, |
| "num_tokens": 44062800.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6422303318977356, |
| "epoch": 1.1644054269752593, |
| "grad_norm": 0.17376938462257385, |
| "learning_rate": 2.3401989595754004e-05, |
| "loss": 0.0687, |
| "mean_token_accuracy": 0.973504614830017, |
| "num_tokens": 44687102.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.665783628821373, |
| "epoch": 1.180367118914605, |
| "grad_norm": 0.1703038364648819, |
| "learning_rate": 2.2997246958882838e-05, |
| "loss": 0.0567, |
| "mean_token_accuracy": 0.9763033747673034, |
| "num_tokens": 45284640.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6595148354768753, |
| "epoch": 1.1963288108539505, |
| "grad_norm": 0.5793496370315552, |
| "learning_rate": 2.2595088046176557e-05, |
| "loss": 0.0676, |
| "mean_token_accuracy": 0.972760459780693, |
| "num_tokens": 45898740.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6939929097890853, |
| "epoch": 1.2122905027932962, |
| "grad_norm": 0.5360252857208252, |
| "learning_rate": 2.219577995278033e-05, |
| "loss": 0.0579, |
| "mean_token_accuracy": 0.9773185312747955, |
| "num_tokens": 46510438.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6901521146297455, |
| "epoch": 1.2282521947326417, |
| "grad_norm": 0.4658600389957428, |
| "learning_rate": 2.179958788045847e-05, |
| "loss": 0.0658, |
| "mean_token_accuracy": 0.9745920717716217, |
| "num_tokens": 47127246.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.6779890835285187, |
| "epoch": 1.2442138866719872, |
| "grad_norm": 0.44346046447753906, |
| "learning_rate": 2.1406774961459813e-05, |
| "loss": 0.0656, |
| "mean_token_accuracy": 0.9746302455663681, |
| "num_tokens": 47747789.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.6651463955640793, |
| "epoch": 1.260175578611333, |
| "grad_norm": 0.9126094579696655, |
| "learning_rate": 2.1017602083757534e-05, |
| "loss": 0.0819, |
| "mean_token_accuracy": 0.9712705045938492, |
| "num_tokens": 48352629.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6546548277139663, |
| "epoch": 1.2761372705506784, |
| "grad_norm": 0.5616291761398315, |
| "learning_rate": 2.0632327717779602e-05, |
| "loss": 0.083, |
| "mean_token_accuracy": 0.9693287074565887, |
| "num_tokens": 48980214.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.6701517790555954, |
| "epoch": 1.2920989624900239, |
| "grad_norm": 0.11611852794885635, |
| "learning_rate": 2.0251207744744794e-05, |
| "loss": 0.0692, |
| "mean_token_accuracy": 0.9744849443435669, |
| "num_tokens": 49598634.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6785144120454788, |
| "epoch": 1.3080606544293696, |
| "grad_norm": 0.49090075492858887, |
| "learning_rate": 1.9874495286718322e-05, |
| "loss": 0.0703, |
| "mean_token_accuracy": 0.9746557891368866, |
| "num_tokens": 50207920.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.67068150639534, |
| "epoch": 1.324022346368715, |
| "grad_norm": 0.194056898355484, |
| "learning_rate": 1.950244053850007e-05, |
| "loss": 0.075, |
| "mean_token_accuracy": 0.9720426052808762, |
| "num_tokens": 50819553.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.675808984041214, |
| "epoch": 1.3399840383080606, |
| "grad_norm": 0.20785000920295715, |
| "learning_rate": 1.9135290601456797e-05, |
| "loss": 0.0749, |
| "mean_token_accuracy": 0.971061834692955, |
| "num_tokens": 51438532.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6731903970241546, |
| "epoch": 1.3559457302474063, |
| "grad_norm": 0.19757410883903503, |
| "learning_rate": 1.8773289319409076e-05, |
| "loss": 0.0783, |
| "mean_token_accuracy": 0.9707212895154953, |
| "num_tokens": 52041716.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6670376569032669, |
| "epoch": 1.3719074221867518, |
| "grad_norm": 0.4560127258300781, |
| "learning_rate": 1.841667711668156e-05, |
| "loss": 0.0668, |
| "mean_token_accuracy": 0.9737331926822662, |
| "num_tokens": 52650555.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.6547212153673172, |
| "epoch": 1.3878691141260973, |
| "grad_norm": 0.13788355886936188, |
| "learning_rate": 1.8065690838424397e-05, |
| "loss": 0.0519, |
| "mean_token_accuracy": 0.9778258532285691, |
| "num_tokens": 53275669.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6567257404327392, |
| "epoch": 1.403830806065443, |
| "grad_norm": 0.5234628319740295, |
| "learning_rate": 1.7720563593311783e-05, |
| "loss": 0.0509, |
| "mean_token_accuracy": 0.979963681101799, |
| "num_tokens": 53889427.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6597132682800293, |
| "epoch": 1.4197924980047885, |
| "grad_norm": 0.1748003363609314, |
| "learning_rate": 1.7381524598722046e-05, |
| "loss": 0.0668, |
| "mean_token_accuracy": 0.9738949596881866, |
| "num_tokens": 54498203.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.6624057710170745, |
| "epoch": 1.435754189944134, |
| "grad_norm": 0.145976260304451, |
| "learning_rate": 1.704879902850221e-05, |
| "loss": 0.0743, |
| "mean_token_accuracy": 0.972572660446167, |
| "num_tokens": 55107757.0, |
| "step": 900 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 1254, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.310366809558876e+18, |
| "train_batch_size": 6, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
