{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 2643, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.011350737797956867, "grad_norm": 1698.2979187914962, "learning_rate": 1.886792452830189e-07, "loss": 11.4349, "step": 10 }, { "epoch": 0.022701475595913734, "grad_norm": 1715.9478636120436, "learning_rate": 3.773584905660378e-07, "loss": 10.7287, "step": 20 }, { "epoch": 0.0340522133938706, "grad_norm": 357.11923483514187, "learning_rate": 5.660377358490567e-07, "loss": 7.9114, "step": 30 }, { "epoch": 0.04540295119182747, "grad_norm": 204.4272724165726, "learning_rate": 7.547169811320755e-07, "loss": 5.3773, "step": 40 }, { "epoch": 0.056753688989784334, "grad_norm": 114.04670191645393, "learning_rate": 9.433962264150944e-07, "loss": 4.1808, "step": 50 }, { "epoch": 0.0681044267877412, "grad_norm": 97.74863975137248, "learning_rate": 1.1320754716981133e-06, "loss": 3.391, "step": 60 }, { "epoch": 0.07945516458569807, "grad_norm": 83.85826671008647, "learning_rate": 1.3207547169811322e-06, "loss": 2.9224, "step": 70 }, { "epoch": 0.09080590238365494, "grad_norm": 81.81468518019015, "learning_rate": 1.509433962264151e-06, "loss": 2.7385, "step": 80 }, { "epoch": 0.1021566401816118, "grad_norm": 81.12682810087568, "learning_rate": 1.6981132075471698e-06, "loss": 2.5178, "step": 90 }, { "epoch": 0.11350737797956867, "grad_norm": 80.85347421686487, "learning_rate": 1.8867924528301889e-06, "loss": 2.2834, "step": 100 }, { "epoch": 0.12485811577752554, "grad_norm": 82.47614053239027, "learning_rate": 2.075471698113208e-06, "loss": 2.0212, "step": 110 }, { "epoch": 0.1362088535754824, "grad_norm": 86.62520026408484, "learning_rate": 2.2641509433962266e-06, "loss": 1.7334, "step": 120 }, { "epoch": 0.14755959137343927, "grad_norm": 80.45183334276052, "learning_rate": 2.4528301886792453e-06, "loss": 1.4777, "step": 130 }, { "epoch": 0.15891032917139614, "grad_norm": 79.21574368328108, "learning_rate": 2.6415094339622644e-06, "loss": 1.1486, "step": 140 }, { "epoch": 0.170261066969353, "grad_norm": 74.01719011960272, "learning_rate": 2.830188679245283e-06, "loss": 0.8411, "step": 150 }, { "epoch": 0.18161180476730987, "grad_norm": 51.89253929462267, "learning_rate": 3.018867924528302e-06, "loss": 0.5991, "step": 160 }, { "epoch": 0.19296254256526674, "grad_norm": 32.21614195805101, "learning_rate": 3.207547169811321e-06, "loss": 0.3447, "step": 170 }, { "epoch": 0.2043132803632236, "grad_norm": 19.001109449561184, "learning_rate": 3.3962264150943395e-06, "loss": 0.2428, "step": 180 }, { "epoch": 0.21566401816118047, "grad_norm": 13.591403736514234, "learning_rate": 3.5849056603773586e-06, "loss": 0.2239, "step": 190 }, { "epoch": 0.22701475595913734, "grad_norm": 13.448898680419939, "learning_rate": 3.7735849056603777e-06, "loss": 0.1635, "step": 200 }, { "epoch": 0.2383654937570942, "grad_norm": 36.227059302392355, "learning_rate": 3.962264150943396e-06, "loss": 0.1634, "step": 210 }, { "epoch": 0.24971623155505107, "grad_norm": 7.663015541152662, "learning_rate": 4.150943396226416e-06, "loss": 0.2969, "step": 220 }, { "epoch": 0.26106696935300794, "grad_norm": 14.54749266726622, "learning_rate": 4.339622641509435e-06, "loss": 0.1923, "step": 230 }, { "epoch": 0.2724177071509648, "grad_norm": 2.313868847454448, "learning_rate": 4.528301886792453e-06, "loss": 0.173, "step": 240 }, { "epoch": 0.28376844494892167, "grad_norm": 6.460084440691116, "learning_rate": 4.716981132075472e-06, "loss": 0.1752, "step": 250 }, { "epoch": 0.29511918274687854, "grad_norm": 2.0859035921114373, "learning_rate": 4.905660377358491e-06, "loss": 0.1466, "step": 260 }, { "epoch": 0.3064699205448354, "grad_norm": 6.562559706748138, "learning_rate": 4.999945458825315e-06, "loss": 0.2425, "step": 270 }, { "epoch": 0.3178206583427923, "grad_norm": 12.775706641336116, "learning_rate": 4.999509143706472e-06, "loss": 0.159, "step": 280 }, { "epoch": 0.32917139614074914, "grad_norm": 6.317651713345659, "learning_rate": 4.998636589618803e-06, "loss": 0.1815, "step": 290 }, { "epoch": 0.340522133938706, "grad_norm": 1.5036062720791235, "learning_rate": 4.997327948849044e-06, "loss": 0.1312, "step": 300 }, { "epoch": 0.3518728717366629, "grad_norm": 6.030878230884589, "learning_rate": 4.995583449794081e-06, "loss": 0.1921, "step": 310 }, { "epoch": 0.36322360953461974, "grad_norm": 10.37100502274357, "learning_rate": 4.993403396921082e-06, "loss": 0.1768, "step": 320 }, { "epoch": 0.3745743473325766, "grad_norm": 3.8501803420046756, "learning_rate": 4.990788170714355e-06, "loss": 0.1714, "step": 330 }, { "epoch": 0.3859250851305335, "grad_norm": 11.372431671658788, "learning_rate": 4.987738227608954e-06, "loss": 0.1445, "step": 340 }, { "epoch": 0.39727582292849034, "grad_norm": 14.259075499950056, "learning_rate": 4.984254099911009e-06, "loss": 0.1337, "step": 350 }, { "epoch": 0.4086265607264472, "grad_norm": 9.445055852659305, "learning_rate": 4.98033639570482e-06, "loss": 0.1908, "step": 360 }, { "epoch": 0.4199772985244041, "grad_norm": 10.699934111645568, "learning_rate": 4.975985798746736e-06, "loss": 0.1582, "step": 370 }, { "epoch": 0.43132803632236094, "grad_norm": 2.0231358005945377, "learning_rate": 4.971203068345811e-06, "loss": 0.1693, "step": 380 }, { "epoch": 0.4426787741203178, "grad_norm": 2.862769306726289, "learning_rate": 4.9659890392312935e-06, "loss": 0.1872, "step": 390 }, { "epoch": 0.4540295119182747, "grad_norm": 12.285642072800846, "learning_rate": 4.960344621406927e-06, "loss": 0.1687, "step": 400 }, { "epoch": 0.46538024971623154, "grad_norm": 3.344948970261047, "learning_rate": 4.954270799992138e-06, "loss": 0.1842, "step": 410 }, { "epoch": 0.4767309875141884, "grad_norm": 10.259497150782368, "learning_rate": 4.947768635050098e-06, "loss": 0.13, "step": 420 }, { "epoch": 0.4880817253121453, "grad_norm": 3.790249696004914, "learning_rate": 4.940839261402711e-06, "loss": 0.2288, "step": 430 }, { "epoch": 0.49943246311010214, "grad_norm": 5.501844538152769, "learning_rate": 4.933483888432558e-06, "loss": 0.0884, "step": 440 }, { "epoch": 0.5107832009080591, "grad_norm": 4.577226609737793, "learning_rate": 4.925703799871818e-06, "loss": 0.1534, "step": 450 }, { "epoch": 0.5221339387060159, "grad_norm": 6.504007138408929, "learning_rate": 4.91750035357822e-06, "loss": 0.1884, "step": 460 }, { "epoch": 0.5334846765039728, "grad_norm": 9.703964220834072, "learning_rate": 4.908874981298058e-06, "loss": 0.1373, "step": 470 }, { "epoch": 0.5448354143019296, "grad_norm": 6.729985856749859, "learning_rate": 4.899829188416306e-06, "loss": 0.2131, "step": 480 }, { "epoch": 0.5561861520998865, "grad_norm": 10.837103551470959, "learning_rate": 4.890364553693886e-06, "loss": 0.1385, "step": 490 }, { "epoch": 0.5675368898978433, "grad_norm": 5.681784964714727, "learning_rate": 4.880482728992126e-06, "loss": 0.1326, "step": 500 }, { "epoch": 0.5788876276958003, "grad_norm": 7.269735955872282, "learning_rate": 4.870185438984458e-06, "loss": 0.1697, "step": 510 }, { "epoch": 0.5902383654937571, "grad_norm": 2.714819037626597, "learning_rate": 4.859474480855417e-06, "loss": 0.1196, "step": 520 }, { "epoch": 0.601589103291714, "grad_norm": 4.859213826030436, "learning_rate": 4.848351723986974e-06, "loss": 0.1278, "step": 530 }, { "epoch": 0.6129398410896708, "grad_norm": 5.451460876916098, "learning_rate": 4.8368191096322734e-06, "loss": 0.1383, "step": 540 }, { "epoch": 0.6242905788876277, "grad_norm": 6.638147694909045, "learning_rate": 4.824878650576829e-06, "loss": 0.1235, "step": 550 }, { "epoch": 0.6356413166855845, "grad_norm": 1.7580573391181251, "learning_rate": 4.81253243078723e-06, "loss": 0.1495, "step": 560 }, { "epoch": 0.6469920544835415, "grad_norm": 4.496989393019484, "learning_rate": 4.7997826050474284e-06, "loss": 0.1127, "step": 570 }, { "epoch": 0.6583427922814983, "grad_norm": 2.446667502094507, "learning_rate": 4.786631398582663e-06, "loss": 0.1749, "step": 580 }, { "epoch": 0.6696935300794552, "grad_norm": 6.832338256010423, "learning_rate": 4.773081106671094e-06, "loss": 0.1344, "step": 590 }, { "epoch": 0.681044267877412, "grad_norm": 7.1481219400148435, "learning_rate": 4.759134094243206e-06, "loss": 0.1188, "step": 600 }, { "epoch": 0.6923950056753689, "grad_norm": 2.2164696749091353, "learning_rate": 4.744792795469058e-06, "loss": 0.1461, "step": 610 }, { "epoch": 0.7037457434733257, "grad_norm": 7.191845232523166, "learning_rate": 4.730059713333448e-06, "loss": 0.1846, "step": 620 }, { "epoch": 0.7150964812712827, "grad_norm": 1.4136399900098233, "learning_rate": 4.714937419199067e-06, "loss": 0.1325, "step": 630 }, { "epoch": 0.7264472190692395, "grad_norm": 4.299537462900032, "learning_rate": 4.69942855235772e-06, "loss": 0.1352, "step": 640 }, { "epoch": 0.7377979568671964, "grad_norm": 1.0384507478186595, "learning_rate": 4.683535819569691e-06, "loss": 0.1362, "step": 650 }, { "epoch": 0.7491486946651532, "grad_norm": 13.982613118228628, "learning_rate": 4.667261994591331e-06, "loss": 0.1706, "step": 660 }, { "epoch": 0.7604994324631101, "grad_norm": 3.227347083041028, "learning_rate": 4.650609917690957e-06, "loss": 0.1208, "step": 670 }, { "epoch": 0.771850170261067, "grad_norm": 4.080502292732569, "learning_rate": 4.633582495153137e-06, "loss": 0.1516, "step": 680 }, { "epoch": 0.7832009080590239, "grad_norm": 8.485813586268012, "learning_rate": 4.616182698771463e-06, "loss": 0.1419, "step": 690 }, { "epoch": 0.7945516458569807, "grad_norm": 1.4110543221413474, "learning_rate": 4.598413565329876e-06, "loss": 0.1327, "step": 700 }, { "epoch": 0.8059023836549376, "grad_norm": 5.188520387076118, "learning_rate": 4.58027819607266e-06, "loss": 0.1438, "step": 710 }, { "epoch": 0.8172531214528944, "grad_norm": 14.402414909123417, "learning_rate": 4.5617797561631885e-06, "loss": 0.1214, "step": 720 }, { "epoch": 0.8286038592508513, "grad_norm": 3.2887042762187364, "learning_rate": 4.542921474131497e-06, "loss": 0.1516, "step": 730 }, { "epoch": 0.8399545970488081, "grad_norm": 8.950465768826493, "learning_rate": 4.523706641310817e-06, "loss": 0.1339, "step": 740 }, { "epoch": 0.8513053348467651, "grad_norm": 7.132814447263203, "learning_rate": 4.5041386112631394e-06, "loss": 0.1061, "step": 750 }, { "epoch": 0.8626560726447219, "grad_norm": 9.762861748224212, "learning_rate": 4.484220799193913e-06, "loss": 0.1011, "step": 760 }, { "epoch": 0.8740068104426788, "grad_norm": 14.288034605598874, "learning_rate": 4.463956681355993e-06, "loss": 0.1298, "step": 770 }, { "epoch": 0.8853575482406356, "grad_norm": 2.8446847525051524, "learning_rate": 4.4433497944429325e-06, "loss": 0.1203, "step": 780 }, { "epoch": 0.8967082860385925, "grad_norm": 1.479832068691713, "learning_rate": 4.422403734971718e-06, "loss": 0.1289, "step": 790 }, { "epoch": 0.9080590238365494, "grad_norm": 9.186524889671574, "learning_rate": 4.401122158655076e-06, "loss": 0.1039, "step": 800 }, { "epoch": 0.9194097616345063, "grad_norm": 1.7652174707129338, "learning_rate": 4.379508779763438e-06, "loss": 0.142, "step": 810 }, { "epoch": 0.9307604994324631, "grad_norm": 2.3999930262675524, "learning_rate": 4.357567370476693e-06, "loss": 0.1483, "step": 820 }, { "epoch": 0.94211123723042, "grad_norm": 2.0203809635081216, "learning_rate": 4.335301760225824e-06, "loss": 0.1388, "step": 830 }, { "epoch": 0.9534619750283768, "grad_norm": 6.570252196832083, "learning_rate": 4.312715835024565e-06, "loss": 0.1132, "step": 840 }, { "epoch": 0.9648127128263337, "grad_norm": 8.694169049942309, "learning_rate": 4.289813536791168e-06, "loss": 0.0886, "step": 850 }, { "epoch": 0.9761634506242906, "grad_norm": 0.665758848556138, "learning_rate": 4.2665988626604285e-06, "loss": 0.1119, "step": 860 }, { "epoch": 0.9875141884222475, "grad_norm": 3.408206675708319, "learning_rate": 4.243075864286059e-06, "loss": 0.1049, "step": 870 }, { "epoch": 0.9988649262202043, "grad_norm": 6.496904281473155, "learning_rate": 4.219248647133559e-06, "loss": 0.1067, "step": 880 }, { "epoch": 1.0102156640181612, "grad_norm": 6.318482209546192, "learning_rate": 4.195121369763687e-06, "loss": 0.1328, "step": 890 }, { "epoch": 1.0215664018161181, "grad_norm": 0.12366856429278299, "learning_rate": 4.170698243106668e-06, "loss": 0.0789, "step": 900 }, { "epoch": 1.0329171396140748, "grad_norm": 2.892704125187497, "learning_rate": 4.14598352972726e-06, "loss": 0.1287, "step": 910 }, { "epoch": 1.0442678774120318, "grad_norm": 10.213754089778735, "learning_rate": 4.12098154308081e-06, "loss": 0.1924, "step": 920 }, { "epoch": 1.0556186152099887, "grad_norm": 7.247963791700357, "learning_rate": 4.095696646760425e-06, "loss": 0.0995, "step": 930 }, { "epoch": 1.0669693530079456, "grad_norm": 4.929548257635512, "learning_rate": 4.070133253735399e-06, "loss": 0.0985, "step": 940 }, { "epoch": 1.0783200908059023, "grad_norm": 6.123766237774935, "learning_rate": 4.044295825581013e-06, "loss": 0.0467, "step": 950 }, { "epoch": 1.0896708286038592, "grad_norm": 4.7904026758844855, "learning_rate": 4.018188871699861e-06, "loss": 0.2061, "step": 960 }, { "epoch": 1.1010215664018161, "grad_norm": 0.6786343511201274, "learning_rate": 3.991816948534823e-06, "loss": 0.1263, "step": 970 }, { "epoch": 1.112372304199773, "grad_norm": 4.293100093143904, "learning_rate": 3.965184658773828e-06, "loss": 0.1175, "step": 980 }, { "epoch": 1.1237230419977298, "grad_norm": 3.375509232222427, "learning_rate": 3.938296650546552e-06, "loss": 0.1199, "step": 990 }, { "epoch": 1.1350737797956867, "grad_norm": 3.6936463749871487, "learning_rate": 3.911157616613176e-06, "loss": 0.1169, "step": 1000 }, { "epoch": 1.1464245175936436, "grad_norm": 5.033595510215138, "learning_rate": 3.8837722935453615e-06, "loss": 0.1326, "step": 1010 }, { "epoch": 1.1577752553916005, "grad_norm": 1.1806777716425099, "learning_rate": 3.8561454608995765e-06, "loss": 0.1049, "step": 1020 }, { "epoch": 1.1691259931895572, "grad_norm": 5.613945679887687, "learning_rate": 3.828281940382923e-06, "loss": 0.0978, "step": 1030 }, { "epoch": 1.1804767309875142, "grad_norm": 11.074114312002287, "learning_rate": 3.800186595011599e-06, "loss": 0.19, "step": 1040 }, { "epoch": 1.191827468785471, "grad_norm": 6.543875702026283, "learning_rate": 3.7718643282621648e-06, "loss": 0.1121, "step": 1050 }, { "epoch": 1.203178206583428, "grad_norm": 4.830576085274986, "learning_rate": 3.7433200832157333e-06, "loss": 0.1125, "step": 1060 }, { "epoch": 1.2145289443813847, "grad_norm": 1.1695778203493483, "learning_rate": 3.714558841695258e-06, "loss": 0.144, "step": 1070 }, { "epoch": 1.2258796821793416, "grad_norm": 3.4535216961458275, "learning_rate": 3.685585623396055e-06, "loss": 0.159, "step": 1080 }, { "epoch": 1.2372304199772985, "grad_norm": 5.495394022771164, "learning_rate": 3.656405485009719e-06, "loss": 0.1044, "step": 1090 }, { "epoch": 1.2485811577752555, "grad_norm": 1.632127452923236, "learning_rate": 3.6270235193415754e-06, "loss": 0.0954, "step": 1100 }, { "epoch": 1.2599318955732124, "grad_norm": 22.50519126324992, "learning_rate": 3.597444854421837e-06, "loss": 0.1576, "step": 1110 }, { "epoch": 1.271282633371169, "grad_norm": 1.3800511219133857, "learning_rate": 3.5676746526106084e-06, "loss": 0.1081, "step": 1120 }, { "epoch": 1.282633371169126, "grad_norm": 9.384054185672928, "learning_rate": 3.537718109696899e-06, "loss": 0.0688, "step": 1130 }, { "epoch": 1.293984108967083, "grad_norm": 8.609201641981887, "learning_rate": 3.5075804539918047e-06, "loss": 0.1562, "step": 1140 }, { "epoch": 1.3053348467650396, "grad_norm": 10.72094281108933, "learning_rate": 3.4772669454160113e-06, "loss": 0.0616, "step": 1150 }, { "epoch": 1.3166855845629966, "grad_norm": 1.033026440388663, "learning_rate": 3.44678287458178e-06, "loss": 0.1002, "step": 1160 }, { "epoch": 1.3280363223609535, "grad_norm": 6.3993440529706, "learning_rate": 3.4161335618695774e-06, "loss": 0.0992, "step": 1170 }, { "epoch": 1.3393870601589104, "grad_norm": 5.7196887859906695, "learning_rate": 3.385324356499513e-06, "loss": 0.1226, "step": 1180 }, { "epoch": 1.3507377979568673, "grad_norm": 3.558047516816503, "learning_rate": 3.3543606355977377e-06, "loss": 0.0733, "step": 1190 }, { "epoch": 1.362088535754824, "grad_norm": 3.417141923587784, "learning_rate": 3.3232478032579746e-06, "loss": 0.1127, "step": 1200 }, { "epoch": 1.373439273552781, "grad_norm": 1.4667598966411892, "learning_rate": 3.291991289598347e-06, "loss": 0.1209, "step": 1210 }, { "epoch": 1.3847900113507379, "grad_norm": 10.429323033457198, "learning_rate": 3.2605965498136554e-06, "loss": 0.086, "step": 1220 }, { "epoch": 1.3961407491486946, "grad_norm": 0.6593997341635577, "learning_rate": 3.229069063223289e-06, "loss": 0.1225, "step": 1230 }, { "epoch": 1.4074914869466515, "grad_norm": 5.189796586998311, "learning_rate": 3.197414332314914e-06, "loss": 0.0741, "step": 1240 }, { "epoch": 1.4188422247446084, "grad_norm": 5.315717279596233, "learning_rate": 3.165637881784133e-06, "loss": 0.1664, "step": 1250 }, { "epoch": 1.4301929625425653, "grad_norm": 2.5922568372048396, "learning_rate": 3.133745257570253e-06, "loss": 0.1355, "step": 1260 }, { "epoch": 1.4415437003405223, "grad_norm": 4.807463414524541, "learning_rate": 3.1017420258883607e-06, "loss": 0.0945, "step": 1270 }, { "epoch": 1.452894438138479, "grad_norm": 6.3118612924547755, "learning_rate": 3.0696337722578444e-06, "loss": 0.1288, "step": 1280 }, { "epoch": 1.4642451759364359, "grad_norm": 4.9487309557769334, "learning_rate": 3.0374261005275606e-06, "loss": 0.0707, "step": 1290 }, { "epoch": 1.4755959137343928, "grad_norm": 2.742348930372723, "learning_rate": 3.0051246318977913e-06, "loss": 0.1143, "step": 1300 }, { "epoch": 1.4869466515323495, "grad_norm": 4.751757205070613, "learning_rate": 2.9727350039391782e-06, "loss": 0.0805, "step": 1310 }, { "epoch": 1.4982973893303064, "grad_norm": 4.196706305756519, "learning_rate": 2.940262869608798e-06, "loss": 0.0943, "step": 1320 }, { "epoch": 1.5096481271282634, "grad_norm": 3.2942511414526443, "learning_rate": 2.907713896263551e-06, "loss": 0.0564, "step": 1330 }, { "epoch": 1.52099886492622, "grad_norm": 14.11917089724621, "learning_rate": 2.8750937646710416e-06, "loss": 0.0471, "step": 1340 }, { "epoch": 1.5323496027241772, "grad_norm": 0.5572781856983822, "learning_rate": 2.842408168018112e-06, "loss": 0.1026, "step": 1350 }, { "epoch": 1.543700340522134, "grad_norm": 4.8778881968669525, "learning_rate": 2.8096628109172125e-06, "loss": 0.0858, "step": 1360 }, { "epoch": 1.5550510783200908, "grad_norm": 3.462366587313288, "learning_rate": 2.7768634084107736e-06, "loss": 0.1614, "step": 1370 }, { "epoch": 1.5664018161180477, "grad_norm": 1.741346685901154, "learning_rate": 2.744015684973766e-06, "loss": 0.0271, "step": 1380 }, { "epoch": 1.5777525539160044, "grad_norm": 0.3396882513904551, "learning_rate": 2.711125373514602e-06, "loss": 0.1021, "step": 1390 }, { "epoch": 1.5891032917139614, "grad_norm": 0.8760532067910843, "learning_rate": 2.6781982143745776e-06, "loss": 0.1074, "step": 1400 }, { "epoch": 1.6004540295119183, "grad_norm": 5.835360058589467, "learning_rate": 2.645239954326009e-06, "loss": 0.1118, "step": 1410 }, { "epoch": 1.611804767309875, "grad_norm": 1.5184172732893502, "learning_rate": 2.6122563455692467e-06, "loss": 0.0898, "step": 1420 }, { "epoch": 1.6231555051078321, "grad_norm": 0.8829766157224805, "learning_rate": 2.5792531447287515e-06, "loss": 0.0668, "step": 1430 }, { "epoch": 1.6345062429057888, "grad_norm": 4.567945936076311, "learning_rate": 2.546236111848387e-06, "loss": 0.1002, "step": 1440 }, { "epoch": 1.6458569807037458, "grad_norm": 5.924471519919804, "learning_rate": 2.5132110093861174e-06, "loss": 0.0915, "step": 1450 }, { "epoch": 1.6572077185017027, "grad_norm": 5.241948497812272, "learning_rate": 2.480183601208293e-06, "loss": 0.0728, "step": 1460 }, { "epoch": 1.6685584562996594, "grad_norm": 0.9103309238230624, "learning_rate": 2.4471596515836797e-06, "loss": 0.0832, "step": 1470 }, { "epoch": 1.6799091940976163, "grad_norm": 5.902192630082476, "learning_rate": 2.414144924177422e-06, "loss": 0.0992, "step": 1480 }, { "epoch": 1.6912599318955732, "grad_norm": 5.542753149797352, "learning_rate": 2.3811451810451132e-06, "loss": 0.1376, "step": 1490 }, { "epoch": 1.70261066969353, "grad_norm": 2.079910130147066, "learning_rate": 2.3481661816271413e-06, "loss": 0.1048, "step": 1500 }, { "epoch": 1.713961407491487, "grad_norm": 9.988363254853992, "learning_rate": 2.3152136817434983e-06, "loss": 0.1075, "step": 1510 }, { "epoch": 1.7253121452894438, "grad_norm": 4.264471331176181, "learning_rate": 2.2822934325892178e-06, "loss": 0.1121, "step": 1520 }, { "epoch": 1.7366628830874007, "grad_norm": 0.41819988751325843, "learning_rate": 2.2494111797306146e-06, "loss": 0.1009, "step": 1530 }, { "epoch": 1.7480136208853576, "grad_norm": 3.129032071006884, "learning_rate": 2.216572662102518e-06, "loss": 0.1405, "step": 1540 }, { "epoch": 1.7593643586833143, "grad_norm": 6.840381293999739, "learning_rate": 2.1837836110066544e-06, "loss": 0.1172, "step": 1550 }, { "epoch": 1.7707150964812712, "grad_norm": 10.965691247426783, "learning_rate": 2.151049749111361e-06, "loss": 0.1048, "step": 1560 }, { "epoch": 1.7820658342792282, "grad_norm": 3.350923692736231, "learning_rate": 2.1183767894528135e-06, "loss": 0.0631, "step": 1570 }, { "epoch": 1.7934165720771849, "grad_norm": 2.386724332698608, "learning_rate": 2.085770434437931e-06, "loss": 0.0908, "step": 1580 }, { "epoch": 1.804767309875142, "grad_norm": 2.7765578684736725, "learning_rate": 2.053236374849128e-06, "loss": 0.1064, "step": 1590 }, { "epoch": 1.8161180476730987, "grad_norm": 2.1935113605962395, "learning_rate": 2.0207802888511155e-06, "loss": 0.0929, "step": 1600 }, { "epoch": 1.8274687854710556, "grad_norm": 1.6758715379011846, "learning_rate": 1.988407840999881e-06, "loss": 0.0941, "step": 1610 }, { "epoch": 1.8388195232690125, "grad_norm": 1.248060350449922, "learning_rate": 1.9561246812540572e-06, "loss": 0.0907, "step": 1620 }, { "epoch": 1.8501702610669692, "grad_norm": 1.7569330315190796, "learning_rate": 1.923936443988841e-06, "loss": 0.0793, "step": 1630 }, { "epoch": 1.8615209988649262, "grad_norm": 1.037210865216885, "learning_rate": 1.8918487470126207e-06, "loss": 0.0758, "step": 1640 }, { "epoch": 1.872871736662883, "grad_norm": 1.6171549381169084, "learning_rate": 1.8598671905865002e-06, "loss": 0.0648, "step": 1650 }, { "epoch": 1.8842224744608398, "grad_norm": 1.0342795380634235, "learning_rate": 1.8279973564468906e-06, "loss": 0.1024, "step": 1660 }, { "epoch": 1.895573212258797, "grad_norm": 2.8676197128323366, "learning_rate": 1.7962448068313298e-06, "loss": 0.0836, "step": 1670 }, { "epoch": 1.9069239500567536, "grad_norm": 4.038378184168931, "learning_rate": 1.7646150835077014e-06, "loss": 0.0969, "step": 1680 }, { "epoch": 1.9182746878547106, "grad_norm": 3.9864637976794057, "learning_rate": 1.733113706807038e-06, "loss": 0.0756, "step": 1690 }, { "epoch": 1.9296254256526675, "grad_norm": 0.20549930730007024, "learning_rate": 1.7017461746600506e-06, "loss": 0.124, "step": 1700 }, { "epoch": 1.9409761634506242, "grad_norm": 9.07668892500117, "learning_rate": 1.670517961637582e-06, "loss": 0.1292, "step": 1710 }, { "epoch": 1.952326901248581, "grad_norm": 7.75514295416776, "learning_rate": 1.6394345179951293e-06, "loss": 0.0718, "step": 1720 }, { "epoch": 1.963677639046538, "grad_norm": 3.9663573159802037, "learning_rate": 1.6085012687216078e-06, "loss": 0.0897, "step": 1730 }, { "epoch": 1.9750283768444947, "grad_norm": 7.281368671013816, "learning_rate": 1.5777236125925333e-06, "loss": 0.1087, "step": 1740 }, { "epoch": 1.9863791146424519, "grad_norm": 3.682988149304066, "learning_rate": 1.5471069212277729e-06, "loss": 0.0973, "step": 1750 }, { "epoch": 1.9977298524404086, "grad_norm": 10.757901893388249, "learning_rate": 1.516656538154035e-06, "loss": 0.1185, "step": 1760 }, { "epoch": 2.0090805902383657, "grad_norm": 3.9977547029166924, "learning_rate": 1.4863777778722682e-06, "loss": 0.0295, "step": 1770 }, { "epoch": 2.0204313280363224, "grad_norm": 0.3839917363760503, "learning_rate": 1.4562759249301185e-06, "loss": 0.0535, "step": 1780 }, { "epoch": 2.031782065834279, "grad_norm": 0.04005398561849314, "learning_rate": 1.4263562329996194e-06, "loss": 0.0435, "step": 1790 }, { "epoch": 2.0431328036322363, "grad_norm": 5.3004949945081, "learning_rate": 1.396623923960271e-06, "loss": 0.0444, "step": 1800 }, { "epoch": 2.054483541430193, "grad_norm": 0.693308833242502, "learning_rate": 1.367084186987663e-06, "loss": 0.0738, "step": 1810 }, { "epoch": 2.0658342792281497, "grad_norm": 2.38026116146638, "learning_rate": 1.3377421776478111e-06, "loss": 0.0413, "step": 1820 }, { "epoch": 2.077185017026107, "grad_norm": 0.8248424564976408, "learning_rate": 1.3086030169973552e-06, "loss": 0.0777, "step": 1830 }, { "epoch": 2.0885357548240635, "grad_norm": 0.7034274141736396, "learning_rate": 1.2796717906897831e-06, "loss": 0.0753, "step": 1840 }, { "epoch": 2.09988649262202, "grad_norm": 5.594790712339865, "learning_rate": 1.2509535480878325e-06, "loss": 0.0217, "step": 1850 }, { "epoch": 2.1112372304199774, "grad_norm": 2.267180272136545, "learning_rate": 1.2224533013822237e-06, "loss": 0.0787, "step": 1860 }, { "epoch": 2.122587968217934, "grad_norm": 6.989338497656842, "learning_rate": 1.1941760247168893e-06, "loss": 0.0918, "step": 1870 }, { "epoch": 2.133938706015891, "grad_norm": 8.671147847692886, "learning_rate": 1.1661266533208274e-06, "loss": 0.1123, "step": 1880 }, { "epoch": 2.145289443813848, "grad_norm": 5.4698413382008395, "learning_rate": 1.1383100826467653e-06, "loss": 0.1148, "step": 1890 }, { "epoch": 2.1566401816118046, "grad_norm": 0.22754634659047315, "learning_rate": 1.1107311675167558e-06, "loss": 0.0902, "step": 1900 }, { "epoch": 2.1679909194097617, "grad_norm": 3.149990990261543, "learning_rate": 1.0833947212748597e-06, "loss": 0.0576, "step": 1910 }, { "epoch": 2.1793416572077184, "grad_norm": 1.4213881249353393, "learning_rate": 1.056305514947082e-06, "loss": 0.0209, "step": 1920 }, { "epoch": 2.1906923950056756, "grad_norm": 3.731385777873592, "learning_rate": 1.0294682764086794e-06, "loss": 0.0414, "step": 1930 }, { "epoch": 2.2020431328036323, "grad_norm": 1.4005631532419822, "learning_rate": 1.002887689559005e-06, "loss": 0.0688, "step": 1940 }, { "epoch": 2.213393870601589, "grad_norm": 5.277740757576594, "learning_rate": 9.765683935040343e-07, "loss": 0.0725, "step": 1950 }, { "epoch": 2.224744608399546, "grad_norm": 0.23197604199738106, "learning_rate": 9.505149817466952e-07, "loss": 0.0867, "step": 1960 }, { "epoch": 2.236095346197503, "grad_norm": 5.653129670824799, "learning_rate": 9.247320013851663e-07, "loss": 0.0995, "step": 1970 }, { "epoch": 2.2474460839954595, "grad_norm": 0.38494105515318466, "learning_rate": 8.992239523192778e-07, "loss": 0.0742, "step": 1980 }, { "epoch": 2.2587968217934167, "grad_norm": 5.78569357166171, "learning_rate": 8.739952864651379e-07, "loss": 0.048, "step": 1990 }, { "epoch": 2.2701475595913734, "grad_norm": 0.8109906329530279, "learning_rate": 8.490504069781444e-07, "loss": 0.0612, "step": 2000 }, { "epoch": 2.28149829738933, "grad_norm": 3.1195154688900244, "learning_rate": 8.243936674845016e-07, "loss": 0.0671, "step": 2010 }, { "epoch": 2.292849035187287, "grad_norm": 0.20743083169700335, "learning_rate": 8.000293713213798e-07, "loss": 0.0607, "step": 2020 }, { "epoch": 2.304199772985244, "grad_norm": 0.2315994749909462, "learning_rate": 7.759617707858619e-07, "loss": 0.0585, "step": 2030 }, { "epoch": 2.315550510783201, "grad_norm": 0.048616209525463916, "learning_rate": 7.521950663927824e-07, "loss": 0.0358, "step": 2040 }, { "epoch": 2.3269012485811578, "grad_norm": 4.235127505493877, "learning_rate": 7.287334061416174e-07, "loss": 0.0685, "step": 2050 }, { "epoch": 2.3382519863791145, "grad_norm": 7.130061647365395, "learning_rate": 7.055808847925316e-07, "loss": 0.0798, "step": 2060 }, { "epoch": 2.3496027241770716, "grad_norm": 5.280177307812628, "learning_rate": 6.827415431517232e-07, "loss": 0.0863, "step": 2070 }, { "epoch": 2.3609534619750283, "grad_norm": 4.524586817903732, "learning_rate": 6.602193673661805e-07, "loss": 0.0741, "step": 2080 }, { "epoch": 2.3723041997729855, "grad_norm": 6.615120826799002, "learning_rate": 6.380182882279817e-07, "loss": 0.0403, "step": 2090 }, { "epoch": 2.383654937570942, "grad_norm": 7.204782892023176, "learning_rate": 6.161421804882528e-07, "loss": 0.0736, "step": 2100 }, { "epoch": 2.395005675368899, "grad_norm": 4.652271416196355, "learning_rate": 5.945948621809092e-07, "loss": 0.0671, "step": 2110 }, { "epoch": 2.406356413166856, "grad_norm": 7.364539748390817, "learning_rate": 5.733800939562941e-07, "loss": 0.0527, "step": 2120 }, { "epoch": 2.4177071509648127, "grad_norm": 0.25707463242318057, "learning_rate": 5.525015784248336e-07, "loss": 0.0212, "step": 2130 }, { "epoch": 2.4290578887627694, "grad_norm": 5.875818604365401, "learning_rate": 5.319629595108189e-07, "loss": 0.0411, "step": 2140 }, { "epoch": 2.4404086265607265, "grad_norm": 0.7877520921047158, "learning_rate": 5.117678218164337e-07, "loss": 0.0386, "step": 2150 }, { "epoch": 2.4517593643586832, "grad_norm": 10.800221983599814, "learning_rate": 4.919196899961367e-07, "loss": 0.0663, "step": 2160 }, { "epoch": 2.46311010215664, "grad_norm": 1.3134487933959456, "learning_rate": 4.724220281414979e-07, "loss": 0.0314, "step": 2170 }, { "epoch": 2.474460839954597, "grad_norm": 0.9855800151128657, "learning_rate": 4.532782391766169e-07, "loss": 0.0482, "step": 2180 }, { "epoch": 2.485811577752554, "grad_norm": 0.0737715113639799, "learning_rate": 4.344916642642108e-07, "loss": 0.046, "step": 2190 }, { "epoch": 2.497162315550511, "grad_norm": 7.1629316214213485, "learning_rate": 4.1606558222247543e-07, "loss": 0.1535, "step": 2200 }, { "epoch": 2.5085130533484676, "grad_norm": 0.08329760250651277, "learning_rate": 3.9800320895284264e-07, "loss": 0.0618, "step": 2210 }, { "epoch": 2.519863791146425, "grad_norm": 2.5343289285776853, "learning_rate": 3.803076968787034e-07, "loss": 0.0313, "step": 2220 }, { "epoch": 2.5312145289443815, "grad_norm": 5.919205383656731, "learning_rate": 3.629821343952147e-07, "loss": 0.0379, "step": 2230 }, { "epoch": 2.542565266742338, "grad_norm": 0.03881848686168996, "learning_rate": 3.4602954533028946e-07, "loss": 0.0483, "step": 2240 }, { "epoch": 2.5539160045402953, "grad_norm": 0.18949910276010634, "learning_rate": 3.2945288841684103e-07, "loss": 0.0354, "step": 2250 }, { "epoch": 2.565266742338252, "grad_norm": 4.662116015922482, "learning_rate": 3.1325505677639977e-07, "loss": 0.0608, "step": 2260 }, { "epoch": 2.5766174801362087, "grad_norm": 10.323403528348715, "learning_rate": 2.974388774141759e-07, "loss": 0.053, "step": 2270 }, { "epoch": 2.587968217934166, "grad_norm": 5.1188402568799845, "learning_rate": 2.820071107256625e-07, "loss": 0.0267, "step": 2280 }, { "epoch": 2.5993189557321226, "grad_norm": 0.10111701455499049, "learning_rate": 2.669624500148646e-07, "loss": 0.0478, "step": 2290 }, { "epoch": 2.6106696935300793, "grad_norm": 0.9800333714837519, "learning_rate": 2.523075210242376e-07, "loss": 0.0298, "step": 2300 }, { "epoch": 2.6220204313280364, "grad_norm": 1.4782390039786926, "learning_rate": 2.3804488147641443e-07, "loss": 0.0663, "step": 2310 }, { "epoch": 2.633371169125993, "grad_norm": 2.416909835102068, "learning_rate": 2.241770206278107e-07, "loss": 0.0189, "step": 2320 }, { "epoch": 2.64472190692395, "grad_norm": 0.17696821901805793, "learning_rate": 2.1070635883417124e-07, "loss": 0.0407, "step": 2330 }, { "epoch": 2.656072644721907, "grad_norm": 0.8013523700822813, "learning_rate": 1.9763524712814812e-07, "loss": 0.0389, "step": 2340 }, { "epoch": 2.6674233825198637, "grad_norm": 0.13612222053062578, "learning_rate": 1.8496596680897398e-07, "loss": 0.0649, "step": 2350 }, { "epoch": 2.678774120317821, "grad_norm": 0.8239843030304835, "learning_rate": 1.7270072904430806e-07, "loss": 0.0404, "step": 2360 }, { "epoch": 2.6901248581157775, "grad_norm": 2.0704079536541617, "learning_rate": 1.608416744843208e-07, "loss": 0.0567, "step": 2370 }, { "epoch": 2.7014755959137347, "grad_norm": 3.2577304351460445, "learning_rate": 1.4939087288808718e-07, "loss": 0.046, "step": 2380 }, { "epoch": 2.7128263337116914, "grad_norm": 9.981537154516658, "learning_rate": 1.3835032276235088e-07, "loss": 0.0576, "step": 2390 }, { "epoch": 2.724177071509648, "grad_norm": 5.704050324905945, "learning_rate": 1.2772195101272645e-07, "loss": 0.0925, "step": 2400 }, { "epoch": 2.735527809307605, "grad_norm": 1.3756958845173786, "learning_rate": 1.1750761260739596e-07, "loss": 0.0627, "step": 2410 }, { "epoch": 2.746878547105562, "grad_norm": 8.13235929955815, "learning_rate": 1.0770909025336202e-07, "loss": 0.0325, "step": 2420 }, { "epoch": 2.7582292849035186, "grad_norm": 0.17235125110595084, "learning_rate": 9.83280940853118e-08, "loss": 0.0616, "step": 2430 }, { "epoch": 2.7695800227014757, "grad_norm": 5.873955549616704, "learning_rate": 8.936626136714754e-08, "loss": 0.0424, "step": 2440 }, { "epoch": 2.7809307604994324, "grad_norm": 0.17753402782466365, "learning_rate": 8.082515620623593e-08, "loss": 0.0253, "step": 2450 }, { "epoch": 2.792281498297389, "grad_norm": 0.06207306412666676, "learning_rate": 7.27062692804234e-08, "loss": 0.0585, "step": 2460 }, { "epoch": 2.8036322360953463, "grad_norm": 0.6499405088971257, "learning_rate": 6.501101757786898e-08, "loss": 0.0536, "step": 2470 }, { "epoch": 2.814982973893303, "grad_norm": 4.920856595800931, "learning_rate": 5.7740744149738516e-08, "loss": 0.0795, "step": 2480 }, { "epoch": 2.8263337116912597, "grad_norm": 0.08802887390877849, "learning_rate": 5.089671787579936e-08, "loss": 0.09, "step": 2490 }, { "epoch": 2.837684449489217, "grad_norm": 0.9119840575805679, "learning_rate": 4.448013324296646e-08, "loss": 0.0511, "step": 2500 }, { "epoch": 2.8490351872871735, "grad_norm": 0.3318639596122148, "learning_rate": 3.849211013682658e-08, "loss": 0.0278, "step": 2510 }, { "epoch": 2.8603859250851307, "grad_norm": 9.771929232558433, "learning_rate": 3.293369364618465e-08, "loss": 0.0411, "step": 2520 }, { "epoch": 2.8717366628830874, "grad_norm": 4.634245009884485, "learning_rate": 2.780585388066581e-08, "loss": 0.0527, "step": 2530 }, { "epoch": 2.8830874006810445, "grad_norm": 5.509503329690719, "learning_rate": 2.3109485801402208e-08, "loss": 0.0313, "step": 2540 }, { "epoch": 2.894438138479001, "grad_norm": 0.6312570349344234, "learning_rate": 1.8845409064833787e-08, "loss": 0.0344, "step": 2550 }, { "epoch": 2.905788876276958, "grad_norm": 7.612882204362058, "learning_rate": 1.5014367879655765e-08, "loss": 0.052, "step": 2560 }, { "epoch": 2.917139614074915, "grad_norm": 0.5134906139680172, "learning_rate": 1.1617030876932001e-08, "loss": 0.0398, "step": 2570 }, { "epoch": 2.9284903518728718, "grad_norm": 0.2021032827858405, "learning_rate": 8.653990993396921e-09, "loss": 0.0711, "step": 2580 }, { "epoch": 2.9398410896708285, "grad_norm": 0.5469399923890682, "learning_rate": 6.125765367972492e-09, "loss": 0.0641, "step": 2590 }, { "epoch": 2.9511918274687856, "grad_norm": 0.11492556820368983, "learning_rate": 4.032795251510957e-09, "loss": 0.0095, "step": 2600 }, { "epoch": 2.9625425652667423, "grad_norm": 0.02879337383162824, "learning_rate": 2.3754459297833932e-09, "loss": 0.0419, "step": 2610 }, { "epoch": 2.973893303064699, "grad_norm": 2.144583462175888, "learning_rate": 1.1540066597270916e-09, "loss": 0.1211, "step": 2620 }, { "epoch": 2.985244040862656, "grad_norm": 0.5856229499773629, "learning_rate": 3.6869061896066536e-10, "loss": 0.0493, "step": 2630 }, { "epoch": 2.996594778660613, "grad_norm": 0.1718989447743379, "learning_rate": 1.9634868579365696e-11, "loss": 0.0414, "step": 2640 }, { "epoch": 3.0, "step": 2643, "total_flos": 130173844045824.0, "train_loss": 0.32796121670620854, "train_runtime": 10598.1253, "train_samples_per_second": 1.995, "train_steps_per_second": 0.249 } ], "logging_steps": 10, "max_steps": 2643, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 1000000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 130173844045824.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }