{ "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": 1206.5750503627912, "learning_rate": 1.886792452830189e-07, "loss": 8.1277, "step": 10 }, { "epoch": 0.022701475595913734, "grad_norm": 691.7936339563854, "learning_rate": 3.773584905660378e-07, "loss": 7.4544, "step": 20 }, { "epoch": 0.0340522133938706, "grad_norm": 561.9646197537672, "learning_rate": 5.660377358490567e-07, "loss": 5.1912, "step": 30 }, { "epoch": 0.04540295119182747, "grad_norm": 117.90097550719499, "learning_rate": 7.547169811320755e-07, "loss": 3.3559, "step": 40 }, { "epoch": 0.056753688989784334, "grad_norm": 113.9134040946928, "learning_rate": 9.433962264150944e-07, "loss": 3.0027, "step": 50 }, { "epoch": 0.0681044267877412, "grad_norm": 96.8453074635257, "learning_rate": 1.1320754716981133e-06, "loss": 2.8897, "step": 60 }, { "epoch": 0.07945516458569807, "grad_norm": 94.70463457358096, "learning_rate": 1.3207547169811322e-06, "loss": 2.7212, "step": 70 }, { "epoch": 0.09080590238365494, "grad_norm": 90.04798724142616, "learning_rate": 1.509433962264151e-06, "loss": 2.5999, "step": 80 }, { "epoch": 0.1021566401816118, "grad_norm": 88.20951083969834, "learning_rate": 1.6981132075471698e-06, "loss": 2.4028, "step": 90 }, { "epoch": 0.11350737797956867, "grad_norm": 86.62449629667134, "learning_rate": 1.8867924528301889e-06, "loss": 2.1737, "step": 100 }, { "epoch": 0.12485811577752554, "grad_norm": 87.08467233150051, "learning_rate": 2.075471698113208e-06, "loss": 1.9197, "step": 110 }, { "epoch": 0.1362088535754824, "grad_norm": 112.67208228174275, "learning_rate": 2.2641509433962266e-06, "loss": 1.6323, "step": 120 }, { "epoch": 0.14755959137343927, "grad_norm": 83.16517580463494, "learning_rate": 2.4528301886792453e-06, "loss": 1.3457, "step": 130 }, { "epoch": 0.15891032917139614, "grad_norm": 82.34421007538906, "learning_rate": 2.6415094339622644e-06, "loss": 1.0278, "step": 140 }, { "epoch": 0.170261066969353, "grad_norm": 64.32468647980009, "learning_rate": 2.830188679245283e-06, "loss": 0.7877, "step": 150 }, { "epoch": 0.18161180476730987, "grad_norm": 43.14120628314963, "learning_rate": 3.018867924528302e-06, "loss": 0.481, "step": 160 }, { "epoch": 0.19296254256526674, "grad_norm": 22.493201550655076, "learning_rate": 3.207547169811321e-06, "loss": 0.3267, "step": 170 }, { "epoch": 0.2043132803632236, "grad_norm": 22.796019463128786, "learning_rate": 3.3962264150943395e-06, "loss": 0.2832, "step": 180 }, { "epoch": 0.21566401816118047, "grad_norm": 135.38641082925105, "learning_rate": 3.5849056603773586e-06, "loss": 0.2393, "step": 190 }, { "epoch": 0.22701475595913734, "grad_norm": 33.31109072718879, "learning_rate": 3.7735849056603777e-06, "loss": 0.303, "step": 200 }, { "epoch": 0.2383654937570942, "grad_norm": 6.409519637652912, "learning_rate": 3.962264150943396e-06, "loss": 0.2845, "step": 210 }, { "epoch": 0.24971623155505107, "grad_norm": 14.141769254490276, "learning_rate": 4.150943396226416e-06, "loss": 0.244, "step": 220 }, { "epoch": 0.26106696935300794, "grad_norm": 7.346342884094764, "learning_rate": 4.339622641509435e-06, "loss": 0.224, "step": 230 }, { "epoch": 0.2724177071509648, "grad_norm": 10.037671290773133, "learning_rate": 4.528301886792453e-06, "loss": 0.2127, "step": 240 }, { "epoch": 0.28376844494892167, "grad_norm": 16.79677827926362, "learning_rate": 4.716981132075472e-06, "loss": 0.2097, "step": 250 }, { "epoch": 0.29511918274687854, "grad_norm": 20.508572546258026, "learning_rate": 4.905660377358491e-06, "loss": 0.2252, "step": 260 }, { "epoch": 0.3064699205448354, "grad_norm": 12.100708379614426, "learning_rate": 4.999945458825315e-06, "loss": 0.199, "step": 270 }, { "epoch": 0.3178206583427923, "grad_norm": 33.509389666332126, "learning_rate": 4.999509143706472e-06, "loss": 0.2362, "step": 280 }, { "epoch": 0.32917139614074914, "grad_norm": 16.62713499564692, "learning_rate": 4.998636589618803e-06, "loss": 0.2601, "step": 290 }, { "epoch": 0.340522133938706, "grad_norm": 6.557530867846841, "learning_rate": 4.997327948849044e-06, "loss": 0.2045, "step": 300 }, { "epoch": 0.3518728717366629, "grad_norm": 7.646957394872313, "learning_rate": 4.995583449794081e-06, "loss": 0.2263, "step": 310 }, { "epoch": 0.36322360953461974, "grad_norm": 1.7194951424552094, "learning_rate": 4.993403396921082e-06, "loss": 0.2025, "step": 320 }, { "epoch": 0.3745743473325766, "grad_norm": 9.408546143672616, "learning_rate": 4.990788170714355e-06, "loss": 0.2013, "step": 330 }, { "epoch": 0.3859250851305335, "grad_norm": 17.565954648277422, "learning_rate": 4.987738227608954e-06, "loss": 0.1779, "step": 340 }, { "epoch": 0.39727582292849034, "grad_norm": 24.70364767980463, "learning_rate": 4.984254099911009e-06, "loss": 0.2493, "step": 350 }, { "epoch": 0.4086265607264472, "grad_norm": 14.825946643733875, "learning_rate": 4.98033639570482e-06, "loss": 0.1914, "step": 360 }, { "epoch": 0.4199772985244041, "grad_norm": 456.2521878214449, "learning_rate": 4.975985798746736e-06, "loss": 0.3128, "step": 370 }, { "epoch": 0.43132803632236094, "grad_norm": 31.570606997776437, "learning_rate": 4.971203068345811e-06, "loss": 0.2858, "step": 380 }, { "epoch": 0.4426787741203178, "grad_norm": 13.644904420168745, "learning_rate": 4.9659890392312935e-06, "loss": 0.2134, "step": 390 }, { "epoch": 0.4540295119182747, "grad_norm": 12.655671350124598, "learning_rate": 4.960344621406927e-06, "loss": 0.1966, "step": 400 }, { "epoch": 0.46538024971623154, "grad_norm": 56.66320437007611, "learning_rate": 4.954270799992138e-06, "loss": 0.2213, "step": 410 }, { "epoch": 0.4767309875141884, "grad_norm": 77.94997980698344, "learning_rate": 4.947768635050098e-06, "loss": 0.2185, "step": 420 }, { "epoch": 0.4880817253121453, "grad_norm": 14.481013841816727, "learning_rate": 4.940839261402711e-06, "loss": 0.2413, "step": 430 }, { "epoch": 0.49943246311010214, "grad_norm": 11.283767794061362, "learning_rate": 4.933483888432558e-06, "loss": 0.1924, "step": 440 }, { "epoch": 0.5107832009080591, "grad_norm": 13.709826166799393, "learning_rate": 4.925703799871818e-06, "loss": 0.2175, "step": 450 }, { "epoch": 0.5221339387060159, "grad_norm": 10.195236902217145, "learning_rate": 4.91750035357822e-06, "loss": 0.2087, "step": 460 }, { "epoch": 0.5334846765039728, "grad_norm": 10.837742762582247, "learning_rate": 4.908874981298058e-06, "loss": 0.1712, "step": 470 }, { "epoch": 0.5448354143019296, "grad_norm": 9.699425356919754, "learning_rate": 4.899829188416306e-06, "loss": 0.2621, "step": 480 }, { "epoch": 0.5561861520998865, "grad_norm": 25.6288556147825, "learning_rate": 4.890364553693886e-06, "loss": 0.1902, "step": 490 }, { "epoch": 0.5675368898978433, "grad_norm": 6.545065430025562, "learning_rate": 4.880482728992126e-06, "loss": 0.1811, "step": 500 }, { "epoch": 0.5788876276958003, "grad_norm": 15.44343240042883, "learning_rate": 4.870185438984458e-06, "loss": 0.1763, "step": 510 }, { "epoch": 0.5902383654937571, "grad_norm": 35.42109306396802, "learning_rate": 4.859474480855417e-06, "loss": 0.1885, "step": 520 }, { "epoch": 0.601589103291714, "grad_norm": 30.969819891608815, "learning_rate": 4.848351723986974e-06, "loss": 0.9334, "step": 530 }, { "epoch": 0.6129398410896708, "grad_norm": 17.382701612484492, "learning_rate": 4.8368191096322734e-06, "loss": 0.2164, "step": 540 }, { "epoch": 0.6242905788876277, "grad_norm": 17.32090201752753, "learning_rate": 4.824878650576829e-06, "loss": 0.2572, "step": 550 }, { "epoch": 0.6356413166855845, "grad_norm": 8.215255357204917, "learning_rate": 4.81253243078723e-06, "loss": 0.2123, "step": 560 }, { "epoch": 0.6469920544835415, "grad_norm": 6.178919363014515, "learning_rate": 4.7997826050474284e-06, "loss": 0.1805, "step": 570 }, { "epoch": 0.6583427922814983, "grad_norm": 7.791368954636398, "learning_rate": 4.786631398582663e-06, "loss": 0.1721, "step": 580 }, { "epoch": 0.6696935300794552, "grad_norm": 19.90892549898479, "learning_rate": 4.773081106671094e-06, "loss": 0.1435, "step": 590 }, { "epoch": 0.681044267877412, "grad_norm": 27.275860702478145, "learning_rate": 4.759134094243206e-06, "loss": 0.2349, "step": 600 }, { "epoch": 0.6923950056753689, "grad_norm": 5.587915653916369, "learning_rate": 4.744792795469058e-06, "loss": 0.1922, "step": 610 }, { "epoch": 0.7037457434733257, "grad_norm": 16.11537142254851, "learning_rate": 4.730059713333448e-06, "loss": 0.1854, "step": 620 }, { "epoch": 0.7150964812712827, "grad_norm": 6.039833790511542, "learning_rate": 4.714937419199067e-06, "loss": 0.1522, "step": 630 }, { "epoch": 0.7264472190692395, "grad_norm": 12.169201736172996, "learning_rate": 4.69942855235772e-06, "loss": 0.1795, "step": 640 }, { "epoch": 0.7377979568671964, "grad_norm": 3.9460653907979784, "learning_rate": 4.683535819569691e-06, "loss": 0.1831, "step": 650 }, { "epoch": 0.7491486946651532, "grad_norm": 19.30178699341767, "learning_rate": 4.667261994591331e-06, "loss": 0.3443, "step": 660 }, { "epoch": 0.7604994324631101, "grad_norm": 7.5342522952336, "learning_rate": 4.650609917690957e-06, "loss": 0.1429, "step": 670 }, { "epoch": 0.771850170261067, "grad_norm": 798.4757143064382, "learning_rate": 4.633582495153137e-06, "loss": 2.3393, "step": 680 }, { "epoch": 0.7832009080590239, "grad_norm": 6.033877423311216, "learning_rate": 4.616182698771463e-06, "loss": 0.1729, "step": 690 }, { "epoch": 0.7945516458569807, "grad_norm": 8.939389577382123, "learning_rate": 4.598413565329876e-06, "loss": 0.2985, "step": 700 }, { "epoch": 0.8059023836549376, "grad_norm": 35.453083635730046, "learning_rate": 4.58027819607266e-06, "loss": 0.2236, "step": 710 }, { "epoch": 0.8172531214528944, "grad_norm": 24.007884979985725, "learning_rate": 4.5617797561631885e-06, "loss": 0.1813, "step": 720 }, { "epoch": 0.8286038592508513, "grad_norm": 19.42733160571041, "learning_rate": 4.542921474131497e-06, "loss": 0.1495, "step": 730 }, { "epoch": 0.8399545970488081, "grad_norm": 21.16952085041393, "learning_rate": 4.523706641310817e-06, "loss": 0.2174, "step": 740 }, { "epoch": 0.8513053348467651, "grad_norm": 16.637482205281813, "learning_rate": 4.5041386112631394e-06, "loss": 0.1834, "step": 750 }, { "epoch": 0.8626560726447219, "grad_norm": 4.138326546073282, "learning_rate": 4.484220799193913e-06, "loss": 0.2246, "step": 760 }, { "epoch": 0.8740068104426788, "grad_norm": 14.092603065545227, "learning_rate": 4.463956681355993e-06, "loss": 0.2121, "step": 770 }, { "epoch": 0.8853575482406356, "grad_norm": 3.3613021191067602, "learning_rate": 4.4433497944429325e-06, "loss": 0.1351, "step": 780 }, { "epoch": 0.8967082860385925, "grad_norm": 9.167043670344585, "learning_rate": 4.422403734971718e-06, "loss": 0.1644, "step": 790 }, { "epoch": 0.9080590238365494, "grad_norm": 7.767571925426298, "learning_rate": 4.401122158655076e-06, "loss": 0.164, "step": 800 }, { "epoch": 0.9194097616345063, "grad_norm": 8.944958649165523, "learning_rate": 4.379508779763438e-06, "loss": 0.2398, "step": 810 }, { "epoch": 0.9307604994324631, "grad_norm": 4.687538443646885, "learning_rate": 4.357567370476693e-06, "loss": 0.1872, "step": 820 }, { "epoch": 0.94211123723042, "grad_norm": 7.092409386060633, "learning_rate": 4.335301760225824e-06, "loss": 0.1972, "step": 830 }, { "epoch": 0.9534619750283768, "grad_norm": 29.37530222421289, "learning_rate": 4.312715835024565e-06, "loss": 0.15, "step": 840 }, { "epoch": 0.9648127128263337, "grad_norm": 24.07301965110236, "learning_rate": 4.289813536791168e-06, "loss": 0.1524, "step": 850 }, { "epoch": 0.9761634506242906, "grad_norm": 15.619595486142124, "learning_rate": 4.2665988626604285e-06, "loss": 0.1681, "step": 860 }, { "epoch": 0.9875141884222475, "grad_norm": 9.556583485953379, "learning_rate": 4.243075864286059e-06, "loss": 0.1598, "step": 870 }, { "epoch": 0.9988649262202043, "grad_norm": 37.15306214641496, "learning_rate": 4.219248647133559e-06, "loss": 0.1934, "step": 880 }, { "epoch": 1.0102156640181612, "grad_norm": 12.695364862989436, "learning_rate": 4.195121369763687e-06, "loss": 0.1516, "step": 890 }, { "epoch": 1.0215664018161181, "grad_norm": 8.141281055040773, "learning_rate": 4.170698243106668e-06, "loss": 0.1743, "step": 900 }, { "epoch": 1.0329171396140748, "grad_norm": 5.341821537111159, "learning_rate": 4.14598352972726e-06, "loss": 0.1908, "step": 910 }, { "epoch": 1.0442678774120318, "grad_norm": 5.896136908215574, "learning_rate": 4.12098154308081e-06, "loss": 0.1991, "step": 920 }, { "epoch": 1.0556186152099887, "grad_norm": 6.91858003869208, "learning_rate": 4.095696646760425e-06, "loss": 0.1802, "step": 930 }, { "epoch": 1.0669693530079456, "grad_norm": 7.176737968426805, "learning_rate": 4.070133253735399e-06, "loss": 0.1884, "step": 940 }, { "epoch": 1.0783200908059023, "grad_norm": 1.8424829812880597, "learning_rate": 4.044295825581013e-06, "loss": 0.1706, "step": 950 }, { "epoch": 1.0896708286038592, "grad_norm": 10.347346400199541, "learning_rate": 4.018188871699861e-06, "loss": 0.1915, "step": 960 }, { "epoch": 1.1010215664018161, "grad_norm": 7.929806028974846, "learning_rate": 3.991816948534823e-06, "loss": 0.1676, "step": 970 }, { "epoch": 1.112372304199773, "grad_norm": 12.33944323097359, "learning_rate": 3.965184658773828e-06, "loss": 0.1088, "step": 980 }, { "epoch": 1.1237230419977298, "grad_norm": 4.29998830732956, "learning_rate": 3.938296650546552e-06, "loss": 0.2106, "step": 990 }, { "epoch": 1.1350737797956867, "grad_norm": 3.9136962770744503, "learning_rate": 3.911157616613176e-06, "loss": 0.2096, "step": 1000 }, { "epoch": 1.1464245175936436, "grad_norm": 25.089883793755547, "learning_rate": 3.8837722935453615e-06, "loss": 0.1532, "step": 1010 }, { "epoch": 1.1577752553916005, "grad_norm": 7.6668125323142196, "learning_rate": 3.8561454608995765e-06, "loss": 0.1877, "step": 1020 }, { "epoch": 1.1691259931895572, "grad_norm": 3.783220620665327, "learning_rate": 3.828281940382923e-06, "loss": 0.1365, "step": 1030 }, { "epoch": 1.1804767309875142, "grad_norm": 18.977590384658843, "learning_rate": 3.800186595011599e-06, "loss": 0.1746, "step": 1040 }, { "epoch": 1.191827468785471, "grad_norm": 21.06989124815396, "learning_rate": 3.7718643282621648e-06, "loss": 0.2425, "step": 1050 }, { "epoch": 1.203178206583428, "grad_norm": 8.2418184201393, "learning_rate": 3.7433200832157333e-06, "loss": 0.279, "step": 1060 }, { "epoch": 1.2145289443813847, "grad_norm": 10.944042695469072, "learning_rate": 3.714558841695258e-06, "loss": 0.2044, "step": 1070 }, { "epoch": 1.2258796821793416, "grad_norm": 9.16937196979649, "learning_rate": 3.685585623396055e-06, "loss": 0.1746, "step": 1080 }, { "epoch": 1.2372304199772985, "grad_norm": 21.052618197003945, "learning_rate": 3.656405485009719e-06, "loss": 0.1904, "step": 1090 }, { "epoch": 1.2485811577752555, "grad_norm": 2.6432100832166356, "learning_rate": 3.6270235193415754e-06, "loss": 0.1525, "step": 1100 }, { "epoch": 1.2599318955732124, "grad_norm": 24.430632068002872, "learning_rate": 3.597444854421837e-06, "loss": 0.1903, "step": 1110 }, { "epoch": 1.271282633371169, "grad_norm": 6.6021417466472805, "learning_rate": 3.5676746526106084e-06, "loss": 0.183, "step": 1120 }, { "epoch": 1.282633371169126, "grad_norm": 13.476867141459701, "learning_rate": 3.537718109696899e-06, "loss": 0.1044, "step": 1130 }, { "epoch": 1.293984108967083, "grad_norm": 4.574499890384411, "learning_rate": 3.5075804539918047e-06, "loss": 0.2196, "step": 1140 }, { "epoch": 1.3053348467650396, "grad_norm": 20.611309374888794, "learning_rate": 3.4772669454160113e-06, "loss": 0.1418, "step": 1150 }, { "epoch": 1.3166855845629966, "grad_norm": 18.478588878765, "learning_rate": 3.44678287458178e-06, "loss": 0.148, "step": 1160 }, { "epoch": 1.3280363223609535, "grad_norm": 2.993204864262699, "learning_rate": 3.4161335618695774e-06, "loss": 0.1705, "step": 1170 }, { "epoch": 1.3393870601589104, "grad_norm": 7.996301923614626, "learning_rate": 3.385324356499513e-06, "loss": 0.1703, "step": 1180 }, { "epoch": 1.3507377979568673, "grad_norm": 17.828733483707087, "learning_rate": 3.3543606355977377e-06, "loss": 0.151, "step": 1190 }, { "epoch": 1.362088535754824, "grad_norm": 4.113522852445312, "learning_rate": 3.3232478032579746e-06, "loss": 0.1753, "step": 1200 }, { "epoch": 1.373439273552781, "grad_norm": 11.281301025002682, "learning_rate": 3.291991289598347e-06, "loss": 0.1364, "step": 1210 }, { "epoch": 1.3847900113507379, "grad_norm": 13.056732263471224, "learning_rate": 3.2605965498136554e-06, "loss": 0.1682, "step": 1220 }, { "epoch": 1.3961407491486946, "grad_norm": 2.836380932680451, "learning_rate": 3.229069063223289e-06, "loss": 0.1931, "step": 1230 }, { "epoch": 1.4074914869466515, "grad_norm": 2.877323934873046, "learning_rate": 3.197414332314914e-06, "loss": 0.153, "step": 1240 }, { "epoch": 1.4188422247446084, "grad_norm": 4.334605850723007, "learning_rate": 3.165637881784133e-06, "loss": 0.163, "step": 1250 }, { "epoch": 1.4301929625425653, "grad_norm": 83.09153616433825, "learning_rate": 3.133745257570253e-06, "loss": 0.1421, "step": 1260 }, { "epoch": 1.4415437003405223, "grad_norm": 5.713915595696203, "learning_rate": 3.1017420258883607e-06, "loss": 0.1158, "step": 1270 }, { "epoch": 1.452894438138479, "grad_norm": 5.929871097769754, "learning_rate": 3.0696337722578444e-06, "loss": 0.1692, "step": 1280 }, { "epoch": 1.4642451759364359, "grad_norm": 11.204539337010555, "learning_rate": 3.0374261005275606e-06, "loss": 0.1574, "step": 1290 }, { "epoch": 1.4755959137343928, "grad_norm": 13.797809799460202, "learning_rate": 3.0051246318977913e-06, "loss": 0.1471, "step": 1300 }, { "epoch": 1.4869466515323495, "grad_norm": 5.935875481621429, "learning_rate": 2.9727350039391782e-06, "loss": 0.1889, "step": 1310 }, { "epoch": 1.4982973893303064, "grad_norm": 11.674393435478562, "learning_rate": 2.940262869608798e-06, "loss": 0.141, "step": 1320 }, { "epoch": 1.5096481271282634, "grad_norm": 17.94354493890428, "learning_rate": 2.907713896263551e-06, "loss": 0.1494, "step": 1330 }, { "epoch": 1.52099886492622, "grad_norm": 15.590002151583917, "learning_rate": 2.8750937646710416e-06, "loss": 0.1208, "step": 1340 }, { "epoch": 1.5323496027241772, "grad_norm": 9.283675780699014, "learning_rate": 2.842408168018112e-06, "loss": 0.1143, "step": 1350 }, { "epoch": 1.543700340522134, "grad_norm": 13.058681305962347, "learning_rate": 2.8096628109172125e-06, "loss": 0.1824, "step": 1360 }, { "epoch": 1.5550510783200908, "grad_norm": 10.588579716744066, "learning_rate": 2.7768634084107736e-06, "loss": 0.1902, "step": 1370 }, { "epoch": 1.5664018161180477, "grad_norm": 10.854981687331344, "learning_rate": 2.744015684973766e-06, "loss": 0.0786, "step": 1380 }, { "epoch": 1.5777525539160044, "grad_norm": 5.542136745578273, "learning_rate": 2.711125373514602e-06, "loss": 0.1339, "step": 1390 }, { "epoch": 1.5891032917139614, "grad_norm": 1.8736224761480822, "learning_rate": 2.6781982143745776e-06, "loss": 0.1599, "step": 1400 }, { "epoch": 1.6004540295119183, "grad_norm": 2.4966938458178043, "learning_rate": 2.645239954326009e-06, "loss": 0.1708, "step": 1410 }, { "epoch": 1.611804767309875, "grad_norm": 3.1749625912208215, "learning_rate": 2.6122563455692467e-06, "loss": 0.1269, "step": 1420 }, { "epoch": 1.6231555051078321, "grad_norm": 10.558063994203357, "learning_rate": 2.5792531447287515e-06, "loss": 0.1109, "step": 1430 }, { "epoch": 1.6345062429057888, "grad_norm": 19.91578884253852, "learning_rate": 2.546236111848387e-06, "loss": 0.1481, "step": 1440 }, { "epoch": 1.6458569807037458, "grad_norm": 5.672294911256832, "learning_rate": 2.5132110093861174e-06, "loss": 0.1587, "step": 1450 }, { "epoch": 1.6572077185017027, "grad_norm": 13.404776442567327, "learning_rate": 2.480183601208293e-06, "loss": 0.1624, "step": 1460 }, { "epoch": 1.6685584562996594, "grad_norm": 11.245062757507418, "learning_rate": 2.4471596515836797e-06, "loss": 0.1564, "step": 1470 }, { "epoch": 1.6799091940976163, "grad_norm": 6.660666123897264, "learning_rate": 2.414144924177422e-06, "loss": 0.1441, "step": 1480 }, { "epoch": 1.6912599318955732, "grad_norm": 6.1817089256122815, "learning_rate": 2.3811451810451132e-06, "loss": 0.1355, "step": 1490 }, { "epoch": 1.70261066969353, "grad_norm": 5.335461901440856, "learning_rate": 2.3481661816271413e-06, "loss": 0.1217, "step": 1500 }, { "epoch": 1.713961407491487, "grad_norm": 9.665270978016556, "learning_rate": 2.3152136817434983e-06, "loss": 0.1697, "step": 1510 }, { "epoch": 1.7253121452894438, "grad_norm": 6.536790651413538, "learning_rate": 2.2822934325892178e-06, "loss": 0.1041, "step": 1520 }, { "epoch": 1.7366628830874007, "grad_norm": 5.743043115884876, "learning_rate": 2.2494111797306146e-06, "loss": 0.116, "step": 1530 }, { "epoch": 1.7480136208853576, "grad_norm": 12.438707792821603, "learning_rate": 2.216572662102518e-06, "loss": 0.2061, "step": 1540 }, { "epoch": 1.7593643586833143, "grad_norm": 11.33003803470324, "learning_rate": 2.1837836110066544e-06, "loss": 0.1428, "step": 1550 }, { "epoch": 1.7707150964812712, "grad_norm": 28.524010261458162, "learning_rate": 2.151049749111361e-06, "loss": 0.1974, "step": 1560 }, { "epoch": 1.7820658342792282, "grad_norm": 35.140121034961226, "learning_rate": 2.1183767894528135e-06, "loss": 0.1374, "step": 1570 }, { "epoch": 1.7934165720771849, "grad_norm": 17.86516511281602, "learning_rate": 2.085770434437931e-06, "loss": 0.1857, "step": 1580 }, { "epoch": 1.804767309875142, "grad_norm": 8.051592262403368, "learning_rate": 2.053236374849128e-06, "loss": 0.1089, "step": 1590 }, { "epoch": 1.8161180476730987, "grad_norm": 11.298498959086164, "learning_rate": 2.0207802888511155e-06, "loss": 0.1255, "step": 1600 }, { "epoch": 1.8274687854710556, "grad_norm": 10.925138037999181, "learning_rate": 1.988407840999881e-06, "loss": 0.157, "step": 1610 }, { "epoch": 1.8388195232690125, "grad_norm": 2.4764649010942947, "learning_rate": 1.9561246812540572e-06, "loss": 0.1053, "step": 1620 }, { "epoch": 1.8501702610669692, "grad_norm": 8.676600159500298, "learning_rate": 1.923936443988841e-06, "loss": 0.0924, "step": 1630 }, { "epoch": 1.8615209988649262, "grad_norm": 16.999715699108364, "learning_rate": 1.8918487470126207e-06, "loss": 0.1465, "step": 1640 }, { "epoch": 1.872871736662883, "grad_norm": 7.042875352593996, "learning_rate": 1.8598671905865002e-06, "loss": 0.1726, "step": 1650 }, { "epoch": 1.8842224744608398, "grad_norm": 5.970641417323245, "learning_rate": 1.8279973564468906e-06, "loss": 0.1369, "step": 1660 }, { "epoch": 1.895573212258797, "grad_norm": 5.671944998923785, "learning_rate": 1.7962448068313298e-06, "loss": 0.1255, "step": 1670 }, { "epoch": 1.9069239500567536, "grad_norm": 3.9066337636163726, "learning_rate": 1.7646150835077014e-06, "loss": 0.1187, "step": 1680 }, { "epoch": 1.9182746878547106, "grad_norm": 6.460765425460155, "learning_rate": 1.733113706807038e-06, "loss": 0.1361, "step": 1690 }, { "epoch": 1.9296254256526675, "grad_norm": 0.8400731882985526, "learning_rate": 1.7017461746600506e-06, "loss": 0.12, "step": 1700 }, { "epoch": 1.9409761634506242, "grad_norm": 18.28664729910389, "learning_rate": 1.670517961637582e-06, "loss": 0.0937, "step": 1710 }, { "epoch": 1.952326901248581, "grad_norm": 26.28199247922179, "learning_rate": 1.6394345179951293e-06, "loss": 0.1354, "step": 1720 }, { "epoch": 1.963677639046538, "grad_norm": 14.67107084793762, "learning_rate": 1.6085012687216078e-06, "loss": 0.1708, "step": 1730 }, { "epoch": 1.9750283768444947, "grad_norm": 8.71277355438378, "learning_rate": 1.5777236125925333e-06, "loss": 0.1407, "step": 1740 }, { "epoch": 1.9863791146424519, "grad_norm": 4.823515814909873, "learning_rate": 1.5471069212277729e-06, "loss": 0.1613, "step": 1750 }, { "epoch": 1.9977298524404086, "grad_norm": 22.289396989558327, "learning_rate": 1.516656538154035e-06, "loss": 0.1323, "step": 1760 }, { "epoch": 2.0090805902383657, "grad_norm": 1.8865020217516166, "learning_rate": 1.4863777778722682e-06, "loss": 0.1294, "step": 1770 }, { "epoch": 2.0204313280363224, "grad_norm": 11.526367438209128, "learning_rate": 1.4562759249301185e-06, "loss": 0.1071, "step": 1780 }, { "epoch": 2.031782065834279, "grad_norm": 3.125341776592479, "learning_rate": 1.4263562329996194e-06, "loss": 0.0924, "step": 1790 }, { "epoch": 2.0431328036322363, "grad_norm": 7.371278957442239, "learning_rate": 1.396623923960271e-06, "loss": 0.1033, "step": 1800 }, { "epoch": 2.054483541430193, "grad_norm": 6.010115447949314, "learning_rate": 1.367084186987663e-06, "loss": 0.1113, "step": 1810 }, { "epoch": 2.0658342792281497, "grad_norm": 3.4781895115807924, "learning_rate": 1.3377421776478111e-06, "loss": 0.079, "step": 1820 }, { "epoch": 2.077185017026107, "grad_norm": 6.698928451746329, "learning_rate": 1.3086030169973552e-06, "loss": 0.1066, "step": 1830 }, { "epoch": 2.0885357548240635, "grad_norm": 3.1335534069995385, "learning_rate": 1.2796717906897831e-06, "loss": 0.1232, "step": 1840 }, { "epoch": 2.09988649262202, "grad_norm": 7.397065091278335, "learning_rate": 1.2509535480878325e-06, "loss": 0.0748, "step": 1850 }, { "epoch": 2.1112372304199774, "grad_norm": 6.274831936983125, "learning_rate": 1.2224533013822237e-06, "loss": 0.1115, "step": 1860 }, { "epoch": 2.122587968217934, "grad_norm": 12.388118261278406, "learning_rate": 1.1941760247168893e-06, "loss": 0.1797, "step": 1870 }, { "epoch": 2.133938706015891, "grad_norm": 18.585021126186835, "learning_rate": 1.1661266533208274e-06, "loss": 0.1888, "step": 1880 }, { "epoch": 2.145289443813848, "grad_norm": 20.351284177772737, "learning_rate": 1.1383100826467653e-06, "loss": 0.1605, "step": 1890 }, { "epoch": 2.1566401816118046, "grad_norm": 2.9497707731250613, "learning_rate": 1.1107311675167558e-06, "loss": 0.0829, "step": 1900 }, { "epoch": 2.1679909194097617, "grad_norm": 4.686169077934406, "learning_rate": 1.0833947212748597e-06, "loss": 0.1085, "step": 1910 }, { "epoch": 2.1793416572077184, "grad_norm": 5.775630537386335, "learning_rate": 1.056305514947082e-06, "loss": 0.0622, "step": 1920 }, { "epoch": 2.1906923950056756, "grad_norm": 12.385588936279845, "learning_rate": 1.0294682764086794e-06, "loss": 0.0784, "step": 1930 }, { "epoch": 2.2020431328036323, "grad_norm": 14.77375128209118, "learning_rate": 1.002887689559005e-06, "loss": 0.1239, "step": 1940 }, { "epoch": 2.213393870601589, "grad_norm": 6.973161925628441, "learning_rate": 9.765683935040343e-07, "loss": 0.1938, "step": 1950 }, { "epoch": 2.224744608399546, "grad_norm": 7.398991996816738, "learning_rate": 9.505149817466952e-07, "loss": 0.1192, "step": 1960 }, { "epoch": 2.236095346197503, "grad_norm": 9.966016190369428, "learning_rate": 9.247320013851663e-07, "loss": 0.1274, "step": 1970 }, { "epoch": 2.2474460839954595, "grad_norm": 5.664265248921905, "learning_rate": 8.992239523192778e-07, "loss": 0.1399, "step": 1980 }, { "epoch": 2.2587968217934167, "grad_norm": 7.605051348993899, "learning_rate": 8.739952864651379e-07, "loss": 0.0925, "step": 1990 }, { "epoch": 2.2701475595913734, "grad_norm": 15.863965925108607, "learning_rate": 8.490504069781444e-07, "loss": 0.1425, "step": 2000 }, { "epoch": 2.28149829738933, "grad_norm": 1.9703704084725924, "learning_rate": 8.243936674845016e-07, "loss": 0.1084, "step": 2010 }, { "epoch": 2.292849035187287, "grad_norm": 5.22937458859332, "learning_rate": 8.000293713213798e-07, "loss": 0.1231, "step": 2020 }, { "epoch": 2.304199772985244, "grad_norm": 1.3462942894346217, "learning_rate": 7.759617707858619e-07, "loss": 0.093, "step": 2030 }, { "epoch": 2.315550510783201, "grad_norm": 1.8801103919301494, "learning_rate": 7.521950663927824e-07, "loss": 0.1058, "step": 2040 }, { "epoch": 2.3269012485811578, "grad_norm": 4.758734079959568, "learning_rate": 7.287334061416174e-07, "loss": 0.1403, "step": 2050 }, { "epoch": 2.3382519863791145, "grad_norm": 13.650821593172457, "learning_rate": 7.055808847925316e-07, "loss": 0.0879, "step": 2060 }, { "epoch": 2.3496027241770716, "grad_norm": 4.585547885645605, "learning_rate": 6.827415431517232e-07, "loss": 0.0879, "step": 2070 }, { "epoch": 2.3609534619750283, "grad_norm": 14.245392944871602, "learning_rate": 6.602193673661805e-07, "loss": 0.1179, "step": 2080 }, { "epoch": 2.3723041997729855, "grad_norm": 1.6887849231542673, "learning_rate": 6.380182882279817e-07, "loss": 0.0811, "step": 2090 }, { "epoch": 2.383654937570942, "grad_norm": 10.220289279150027, "learning_rate": 6.161421804882528e-07, "loss": 0.0933, "step": 2100 }, { "epoch": 2.395005675368899, "grad_norm": 11.088314030618392, "learning_rate": 5.945948621809092e-07, "loss": 0.0963, "step": 2110 }, { "epoch": 2.406356413166856, "grad_norm": 11.409606317504695, "learning_rate": 5.733800939562941e-07, "loss": 0.1128, "step": 2120 }, { "epoch": 2.4177071509648127, "grad_norm": 7.015734563795156, "learning_rate": 5.525015784248336e-07, "loss": 0.0765, "step": 2130 }, { "epoch": 2.4290578887627694, "grad_norm": 8.898118807774802, "learning_rate": 5.319629595108189e-07, "loss": 0.1023, "step": 2140 }, { "epoch": 2.4404086265607265, "grad_norm": 3.7871407557577825, "learning_rate": 5.117678218164337e-07, "loss": 0.0632, "step": 2150 }, { "epoch": 2.4517593643586832, "grad_norm": 18.723321348211165, "learning_rate": 4.919196899961367e-07, "loss": 0.1472, "step": 2160 }, { "epoch": 2.46311010215664, "grad_norm": 0.701990772027342, "learning_rate": 4.724220281414979e-07, "loss": 0.0849, "step": 2170 }, { "epoch": 2.474460839954597, "grad_norm": 12.198138261167522, "learning_rate": 4.532782391766169e-07, "loss": 0.1082, "step": 2180 }, { "epoch": 2.485811577752554, "grad_norm": 0.4585306231750833, "learning_rate": 4.344916642642108e-07, "loss": 0.0796, "step": 2190 }, { "epoch": 2.497162315550511, "grad_norm": 8.574884374820567, "learning_rate": 4.1606558222247543e-07, "loss": 0.2072, "step": 2200 }, { "epoch": 2.5085130533484676, "grad_norm": 1.041407085041633, "learning_rate": 3.9800320895284264e-07, "loss": 0.0826, "step": 2210 }, { "epoch": 2.519863791146425, "grad_norm": 3.991127061652976, "learning_rate": 3.803076968787034e-07, "loss": 0.0756, "step": 2220 }, { "epoch": 2.5312145289443815, "grad_norm": 1.614628978447142, "learning_rate": 3.629821343952147e-07, "loss": 0.0919, "step": 2230 }, { "epoch": 2.542565266742338, "grad_norm": 1.1422925700016022, "learning_rate": 3.4602954533028946e-07, "loss": 0.0577, "step": 2240 }, { "epoch": 2.5539160045402953, "grad_norm": 4.138694622144, "learning_rate": 3.2945288841684103e-07, "loss": 0.0868, "step": 2250 }, { "epoch": 2.565266742338252, "grad_norm": 18.677134096279577, "learning_rate": 3.1325505677639977e-07, "loss": 0.0986, "step": 2260 }, { "epoch": 2.5766174801362087, "grad_norm": 5.767949591915205, "learning_rate": 2.974388774141759e-07, "loss": 0.0504, "step": 2270 }, { "epoch": 2.587968217934166, "grad_norm": 5.723871453377219, "learning_rate": 2.820071107256625e-07, "loss": 0.1096, "step": 2280 }, { "epoch": 2.5993189557321226, "grad_norm": 0.6617120592844709, "learning_rate": 2.669624500148646e-07, "loss": 0.0957, "step": 2290 }, { "epoch": 2.6106696935300793, "grad_norm": 12.869937707971685, "learning_rate": 2.523075210242376e-07, "loss": 0.1234, "step": 2300 }, { "epoch": 2.6220204313280364, "grad_norm": 11.715296190832456, "learning_rate": 2.3804488147641443e-07, "loss": 0.1098, "step": 2310 }, { "epoch": 2.633371169125993, "grad_norm": 4.819435893099225, "learning_rate": 2.241770206278107e-07, "loss": 0.0629, "step": 2320 }, { "epoch": 2.64472190692395, "grad_norm": 6.748457458976083, "learning_rate": 2.1070635883417124e-07, "loss": 0.1066, "step": 2330 }, { "epoch": 2.656072644721907, "grad_norm": 2.4556518043813864, "learning_rate": 1.9763524712814812e-07, "loss": 0.0801, "step": 2340 }, { "epoch": 2.6674233825198637, "grad_norm": 3.6886493224926564, "learning_rate": 1.8496596680897398e-07, "loss": 0.0698, "step": 2350 }, { "epoch": 2.678774120317821, "grad_norm": 1.4396829744056499, "learning_rate": 1.7270072904430806e-07, "loss": 0.0989, "step": 2360 }, { "epoch": 2.6901248581157775, "grad_norm": 11.527713138221026, "learning_rate": 1.608416744843208e-07, "loss": 0.0936, "step": 2370 }, { "epoch": 2.7014755959137347, "grad_norm": 6.782278885940143, "learning_rate": 1.4939087288808718e-07, "loss": 0.1064, "step": 2380 }, { "epoch": 2.7128263337116914, "grad_norm": 24.520689567279586, "learning_rate": 1.3835032276235088e-07, "loss": 0.1185, "step": 2390 }, { "epoch": 2.724177071509648, "grad_norm": 14.114636610459446, "learning_rate": 1.2772195101272645e-07, "loss": 0.1697, "step": 2400 }, { "epoch": 2.735527809307605, "grad_norm": 10.0462208950971, "learning_rate": 1.1750761260739596e-07, "loss": 0.1074, "step": 2410 }, { "epoch": 2.746878547105562, "grad_norm": 6.125593052727082, "learning_rate": 1.0770909025336202e-07, "loss": 0.072, "step": 2420 }, { "epoch": 2.7582292849035186, "grad_norm": 0.3138265869261904, "learning_rate": 9.83280940853118e-08, "loss": 0.0908, "step": 2430 }, { "epoch": 2.7695800227014757, "grad_norm": 2.8812395503982557, "learning_rate": 8.936626136714754e-08, "loss": 0.1301, "step": 2440 }, { "epoch": 2.7809307604994324, "grad_norm": 1.5115718441522554, "learning_rate": 8.082515620623593e-08, "loss": 0.0461, "step": 2450 }, { "epoch": 2.792281498297389, "grad_norm": 0.40891115954566937, "learning_rate": 7.27062692804234e-08, "loss": 0.1166, "step": 2460 }, { "epoch": 2.8036322360953463, "grad_norm": 2.65696194654047, "learning_rate": 6.501101757786898e-08, "loss": 0.0738, "step": 2470 }, { "epoch": 2.814982973893303, "grad_norm": 7.213135089295003, "learning_rate": 5.7740744149738516e-08, "loss": 0.1448, "step": 2480 }, { "epoch": 2.8263337116912597, "grad_norm": 2.8642186082476857, "learning_rate": 5.089671787579936e-08, "loss": 0.0842, "step": 2490 }, { "epoch": 2.837684449489217, "grad_norm": 11.646216769186815, "learning_rate": 4.448013324296646e-08, "loss": 0.0955, "step": 2500 }, { "epoch": 2.8490351872871735, "grad_norm": 10.176833634286226, "learning_rate": 3.849211013682658e-08, "loss": 0.0803, "step": 2510 }, { "epoch": 2.8603859250851307, "grad_norm": 17.617813943810784, "learning_rate": 3.293369364618465e-08, "loss": 0.117, "step": 2520 }, { "epoch": 2.8717366628830874, "grad_norm": 3.7336376357377596, "learning_rate": 2.780585388066581e-08, "loss": 0.0954, "step": 2530 }, { "epoch": 2.8830874006810445, "grad_norm": 7.835412669053526, "learning_rate": 2.3109485801402208e-08, "loss": 0.064, "step": 2540 }, { "epoch": 2.894438138479001, "grad_norm": 2.2882383365385235, "learning_rate": 1.8845409064833787e-08, "loss": 0.0813, "step": 2550 }, { "epoch": 2.905788876276958, "grad_norm": 6.984695704690644, "learning_rate": 1.5014367879655765e-08, "loss": 0.0643, "step": 2560 }, { "epoch": 2.917139614074915, "grad_norm": 15.127785211570155, "learning_rate": 1.1617030876932001e-08, "loss": 0.1061, "step": 2570 }, { "epoch": 2.9284903518728718, "grad_norm": 2.163747180147469, "learning_rate": 8.653990993396921e-09, "loss": 0.1572, "step": 2580 }, { "epoch": 2.9398410896708285, "grad_norm": 12.65041493232648, "learning_rate": 6.125765367972492e-09, "loss": 0.1233, "step": 2590 }, { "epoch": 2.9511918274687856, "grad_norm": 1.803960606784732, "learning_rate": 4.032795251510957e-09, "loss": 0.1235, "step": 2600 }, { "epoch": 2.9625425652667423, "grad_norm": 0.35198543997535436, "learning_rate": 2.3754459297833932e-09, "loss": 0.0928, "step": 2610 }, { "epoch": 2.973893303064699, "grad_norm": 6.760769411930297, "learning_rate": 1.1540066597270916e-09, "loss": 0.1342, "step": 2620 }, { "epoch": 2.985244040862656, "grad_norm": 9.482706999829817, "learning_rate": 3.6869061896066536e-10, "loss": 0.0706, "step": 2630 }, { "epoch": 2.996594778660613, "grad_norm": 2.550219332049932, "learning_rate": 1.9634868579365696e-11, "loss": 0.0663, "step": 2640 }, { "epoch": 3.0, "step": 2643, "total_flos": 130173844045824.0, "train_loss": 0.3333010443560624, "train_runtime": 10581.2247, "train_samples_per_second": 1.998, "train_steps_per_second": 0.25 } ], "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 }