{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.2, "eval_steps": 500, "global_step": 3000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "grad_norm": 0.20317015051841736, "learning_rate": 1.2000000000000002e-06, "loss": 1.2026, "step": 10 }, { "grad_norm": 0.17602789402008057, "learning_rate": 2.5333333333333334e-06, "loss": 1.212, "step": 20 }, { "grad_norm": 0.15509557723999023, "learning_rate": 3.866666666666667e-06, "loss": 1.2101, "step": 30 }, { "grad_norm": 0.2354014366865158, "learning_rate": 5.2e-06, "loss": 1.1932, "step": 40 }, { "grad_norm": 0.279198557138443, "learning_rate": 6.533333333333333e-06, "loss": 1.1606, "step": 50 }, { "grad_norm": 0.33185648918151855, "learning_rate": 7.866666666666667e-06, "loss": 1.1288, "step": 60 }, { "grad_norm": 0.38778021931648254, "learning_rate": 9.2e-06, "loss": 1.1153, "step": 70 }, { "grad_norm": 0.46419402956962585, "learning_rate": 1.0533333333333335e-05, "loss": 1.114, "step": 80 }, { "grad_norm": 0.33136653900146484, "learning_rate": 1.1866666666666668e-05, "loss": 1.1032, "step": 90 }, { "grad_norm": 0.5122213363647461, "learning_rate": 1.32e-05, "loss": 1.0945, "step": 100 }, { "grad_norm": 0.2915997803211212, "learning_rate": 1.4533333333333335e-05, "loss": 1.0947, "step": 110 }, { "grad_norm": 0.2920270264148712, "learning_rate": 1.586666666666667e-05, "loss": 1.0932, "step": 120 }, { "grad_norm": 0.487945020198822, "learning_rate": 1.7199999999999998e-05, "loss": 1.0891, "step": 130 }, { "grad_norm": 0.32595834136009216, "learning_rate": 1.8533333333333334e-05, "loss": 1.0794, "step": 140 }, { "grad_norm": 0.5054919719696045, "learning_rate": 1.9866666666666667e-05, "loss": 1.0857, "step": 150 }, { "grad_norm": 0.4286811053752899, "learning_rate": 2.12e-05, "loss": 1.0692, "step": 160 }, { "grad_norm": 0.3598857522010803, "learning_rate": 2.2533333333333333e-05, "loss": 1.0568, "step": 170 }, { "grad_norm": 0.6221235394477844, "learning_rate": 2.3866666666666666e-05, "loss": 1.0221, "step": 180 }, { "grad_norm": 0.5267907977104187, "learning_rate": 2.5200000000000003e-05, "loss": 0.9844, "step": 190 }, { "grad_norm": 0.5575628876686096, "learning_rate": 2.6533333333333332e-05, "loss": 0.9463, "step": 200 }, { "grad_norm": 0.9670434594154358, "learning_rate": 2.786666666666667e-05, "loss": 0.926, "step": 210 }, { "grad_norm": 0.6401852965354919, "learning_rate": 2.9199999999999998e-05, "loss": 0.8947, "step": 220 }, { "grad_norm": 1.0171445608139038, "learning_rate": 3.0533333333333335e-05, "loss": 0.8689, "step": 230 }, { "grad_norm": 0.8912598490715027, "learning_rate": 3.1866666666666664e-05, "loss": 0.8356, "step": 240 }, { "grad_norm": 0.8143109083175659, "learning_rate": 3.32e-05, "loss": 0.8007, "step": 250 }, { "grad_norm": 1.0494921207427979, "learning_rate": 3.453333333333334e-05, "loss": 0.7777, "step": 260 }, { "grad_norm": 0.8461057543754578, "learning_rate": 3.586666666666667e-05, "loss": 0.7676, "step": 270 }, { "grad_norm": 1.216251015663147, "learning_rate": 3.72e-05, "loss": 0.7257, "step": 280 }, { "grad_norm": 0.9196227192878723, "learning_rate": 3.853333333333334e-05, "loss": 0.7105, "step": 290 }, { "grad_norm": 0.9573485851287842, "learning_rate": 3.986666666666667e-05, "loss": 0.6779, "step": 300 }, { "grad_norm": 0.9070004224777222, "learning_rate": 4.12e-05, "loss": 0.6641, "step": 310 }, { "grad_norm": 1.0238310098648071, "learning_rate": 4.2533333333333335e-05, "loss": 0.6377, "step": 320 }, { "grad_norm": 0.9185635447502136, "learning_rate": 4.3866666666666665e-05, "loss": 0.6237, "step": 330 }, { "grad_norm": 1.0374126434326172, "learning_rate": 4.52e-05, "loss": 0.6077, "step": 340 }, { "grad_norm": 1.21366286277771, "learning_rate": 4.653333333333334e-05, "loss": 0.5892, "step": 350 }, { "grad_norm": 1.0136992931365967, "learning_rate": 4.7866666666666674e-05, "loss": 0.5755, "step": 360 }, { "grad_norm": 0.9329037070274353, "learning_rate": 4.92e-05, "loss": 0.5773, "step": 370 }, { "grad_norm": 0.8955318927764893, "learning_rate": 5.053333333333333e-05, "loss": 0.5684, "step": 380 }, { "grad_norm": 1.2243609428405762, "learning_rate": 5.1866666666666676e-05, "loss": 0.565, "step": 390 }, { "grad_norm": 1.0137560367584229, "learning_rate": 5.3200000000000006e-05, "loss": 0.5542, "step": 400 }, { "grad_norm": 1.0478407144546509, "learning_rate": 5.4533333333333335e-05, "loss": 0.5511, "step": 410 }, { "grad_norm": 0.9874824285507202, "learning_rate": 5.5866666666666665e-05, "loss": 0.5434, "step": 420 }, { "grad_norm": 1.015325665473938, "learning_rate": 5.72e-05, "loss": 0.5326, "step": 430 }, { "grad_norm": 1.4563086032867432, "learning_rate": 5.853333333333334e-05, "loss": 0.52, "step": 440 }, { "grad_norm": 1.0814019441604614, "learning_rate": 5.9866666666666674e-05, "loss": 0.5138, "step": 450 }, { "grad_norm": 1.3471565246582031, "learning_rate": 6.12e-05, "loss": 0.501, "step": 460 }, { "grad_norm": 1.1411954164505005, "learning_rate": 6.253333333333333e-05, "loss": 0.4894, "step": 470 }, { "grad_norm": 1.0365638732910156, "learning_rate": 6.386666666666667e-05, "loss": 0.4751, "step": 480 }, { "grad_norm": 1.1257281303405762, "learning_rate": 6.52e-05, "loss": 0.4655, "step": 490 }, { "grad_norm": 0.9818600416183472, "learning_rate": 6.653333333333334e-05, "loss": 0.4481, "step": 500 }, { "grad_norm": 1.2712481021881104, "learning_rate": 6.786666666666667e-05, "loss": 0.445, "step": 510 }, { "grad_norm": 1.0390636920928955, "learning_rate": 6.92e-05, "loss": 0.4333, "step": 520 }, { "grad_norm": 0.9593179821968079, "learning_rate": 7.053333333333334e-05, "loss": 0.4231, "step": 530 }, { "grad_norm": 0.9991907477378845, "learning_rate": 7.186666666666667e-05, "loss": 0.4107, "step": 540 }, { "grad_norm": 1.1595443487167358, "learning_rate": 7.32e-05, "loss": 0.4125, "step": 550 }, { "grad_norm": 1.430122971534729, "learning_rate": 7.453333333333333e-05, "loss": 0.3922, "step": 560 }, { "grad_norm": 1.0540282726287842, "learning_rate": 7.586666666666668e-05, "loss": 0.3865, "step": 570 }, { "grad_norm": 1.3749494552612305, "learning_rate": 7.72e-05, "loss": 0.3765, "step": 580 }, { "grad_norm": 1.1181503534317017, "learning_rate": 7.853333333333334e-05, "loss": 0.3777, "step": 590 }, { "grad_norm": 1.1761486530303955, "learning_rate": 7.986666666666667e-05, "loss": 0.3755, "step": 600 }, { "grad_norm": 1.1582212448120117, "learning_rate": 8.120000000000001e-05, "loss": 0.3737, "step": 610 }, { "grad_norm": 1.0623677968978882, "learning_rate": 8.253333333333334e-05, "loss": 0.3422, "step": 620 }, { "grad_norm": 1.0630512237548828, "learning_rate": 8.386666666666667e-05, "loss": 0.3304, "step": 630 }, { "grad_norm": 1.501352310180664, "learning_rate": 8.52e-05, "loss": 0.33, "step": 640 }, { "grad_norm": 1.2959400415420532, "learning_rate": 8.653333333333333e-05, "loss": 0.3319, "step": 650 }, { "grad_norm": 1.3737542629241943, "learning_rate": 8.786666666666667e-05, "loss": 0.3159, "step": 660 }, { "grad_norm": 1.3062279224395752, "learning_rate": 8.92e-05, "loss": 0.322, "step": 670 }, { "grad_norm": 1.303208827972412, "learning_rate": 9.053333333333334e-05, "loss": 0.303, "step": 680 }, { "grad_norm": 1.2014473676681519, "learning_rate": 9.186666666666667e-05, "loss": 0.2871, "step": 690 }, { "grad_norm": 1.2867250442504883, "learning_rate": 9.320000000000002e-05, "loss": 0.2809, "step": 700 }, { "grad_norm": 1.4270846843719482, "learning_rate": 9.453333333333335e-05, "loss": 0.2908, "step": 710 }, { "grad_norm": 1.4763070344924927, "learning_rate": 9.586666666666667e-05, "loss": 0.2645, "step": 720 }, { "grad_norm": 1.3187637329101562, "learning_rate": 9.72e-05, "loss": 0.2533, "step": 730 }, { "grad_norm": 1.326865553855896, "learning_rate": 9.853333333333333e-05, "loss": 0.2303, "step": 740 }, { "grad_norm": 1.082550048828125, "learning_rate": 9.986666666666668e-05, "loss": 0.2364, "step": 750 }, { "grad_norm": 1.3675532341003418, "learning_rate": 9.999990157738453e-05, "loss": 0.2219, "step": 760 }, { "grad_norm": 1.141180396080017, "learning_rate": 9.999956135155687e-05, "loss": 0.2205, "step": 770 }, { "grad_norm": 1.4188575744628906, "learning_rate": 9.99989781090763e-05, "loss": 0.2108, "step": 780 }, { "grad_norm": 1.0454844236373901, "learning_rate": 9.999815185277755e-05, "loss": 0.1953, "step": 790 }, { "grad_norm": 1.4326947927474976, "learning_rate": 9.999708258667652e-05, "loss": 0.1939, "step": 800 }, { "grad_norm": 1.13813316822052, "learning_rate": 9.999577031597029e-05, "loss": 0.1968, "step": 810 }, { "grad_norm": 1.29367995262146, "learning_rate": 9.999421504703696e-05, "loss": 0.1827, "step": 820 }, { "grad_norm": 1.23212468624115, "learning_rate": 9.999241678743574e-05, "loss": 0.1839, "step": 830 }, { "grad_norm": 1.0282104015350342, "learning_rate": 9.999037554590683e-05, "loss": 0.1666, "step": 840 }, { "grad_norm": 1.1773988008499146, "learning_rate": 9.998809133237143e-05, "loss": 0.1752, "step": 850 }, { "grad_norm": 1.2939362525939941, "learning_rate": 9.998556415793169e-05, "loss": 0.1605, "step": 860 }, { "grad_norm": 1.160797357559204, "learning_rate": 9.998279403487062e-05, "loss": 0.1543, "step": 870 }, { "grad_norm": 1.4621920585632324, "learning_rate": 9.997978097665205e-05, "loss": 0.148, "step": 880 }, { "grad_norm": 1.086423397064209, "learning_rate": 9.99765249979206e-05, "loss": 0.1361, "step": 890 }, { "grad_norm": 1.481440782546997, "learning_rate": 9.997302611450154e-05, "loss": 0.1574, "step": 900 }, { "grad_norm": 1.2555721998214722, "learning_rate": 9.996928434340073e-05, "loss": 0.1333, "step": 910 }, { "grad_norm": 1.2162728309631348, "learning_rate": 9.996529970280462e-05, "loss": 0.149, "step": 920 }, { "grad_norm": 1.0630161762237549, "learning_rate": 9.996107221208004e-05, "loss": 0.1377, "step": 930 }, { "grad_norm": 1.0551754236221313, "learning_rate": 9.995660189177419e-05, "loss": 0.1314, "step": 940 }, { "grad_norm": 1.0824174880981445, "learning_rate": 9.995188876361451e-05, "loss": 0.1478, "step": 950 }, { "grad_norm": 1.032842993736267, "learning_rate": 9.994693285050857e-05, "loss": 0.126, "step": 960 }, { "grad_norm": 0.9250110983848572, "learning_rate": 9.994173417654395e-05, "loss": 0.1284, "step": 970 }, { "grad_norm": 0.9521851539611816, "learning_rate": 9.993629276698821e-05, "loss": 0.1372, "step": 980 }, { "grad_norm": 1.08961820602417, "learning_rate": 9.993060864828858e-05, "loss": 0.1385, "step": 990 }, { "grad_norm": 1.079972505569458, "learning_rate": 9.992468184807206e-05, "loss": 0.1333, "step": 1000 }, { "grad_norm": 0.9935213923454285, "learning_rate": 9.991851239514511e-05, "loss": 0.1228, "step": 1010 }, { "grad_norm": 0.9603984355926514, "learning_rate": 9.991210031949359e-05, "loss": 0.1266, "step": 1020 }, { "grad_norm": 1.0508372783660889, "learning_rate": 9.990544565228259e-05, "loss": 0.1244, "step": 1030 }, { "grad_norm": 0.9042491316795349, "learning_rate": 9.989854842585631e-05, "loss": 0.1323, "step": 1040 }, { "grad_norm": 1.0979080200195312, "learning_rate": 9.989140867373783e-05, "loss": 0.1196, "step": 1050 }, { "grad_norm": 0.9999712705612183, "learning_rate": 9.988402643062907e-05, "loss": 0.1334, "step": 1060 }, { "grad_norm": 0.9937282800674438, "learning_rate": 9.987640173241046e-05, "loss": 0.1299, "step": 1070 }, { "grad_norm": 1.0657336711883545, "learning_rate": 9.986853461614093e-05, "loss": 0.1278, "step": 1080 }, { "grad_norm": 1.0081781148910522, "learning_rate": 9.986042512005763e-05, "loss": 0.106, "step": 1090 }, { "grad_norm": 1.0005905628204346, "learning_rate": 9.985207328357573e-05, "loss": 0.1169, "step": 1100 }, { "grad_norm": 0.8901131749153137, "learning_rate": 9.984347914728829e-05, "loss": 0.1267, "step": 1110 }, { "grad_norm": 1.0940711498260498, "learning_rate": 9.983464275296605e-05, "loss": 0.1302, "step": 1120 }, { "grad_norm": 0.9827264547348022, "learning_rate": 9.982556414355724e-05, "loss": 0.1324, "step": 1130 }, { "grad_norm": 1.1492973566055298, "learning_rate": 9.981624336318726e-05, "loss": 0.1337, "step": 1140 }, { "grad_norm": 1.1605128049850464, "learning_rate": 9.980668045715864e-05, "loss": 0.1267, "step": 1150 }, { "grad_norm": 0.9685631990432739, "learning_rate": 9.979687547195066e-05, "loss": 0.126, "step": 1160 }, { "grad_norm": 1.1568074226379395, "learning_rate": 9.978682845521927e-05, "loss": 0.1238, "step": 1170 }, { "grad_norm": 1.0317803621292114, "learning_rate": 9.977653945579673e-05, "loss": 0.1254, "step": 1180 }, { "grad_norm": 0.914109468460083, "learning_rate": 9.976600852369144e-05, "loss": 0.1236, "step": 1190 }, { "grad_norm": 1.127572774887085, "learning_rate": 9.975523571008769e-05, "loss": 0.1271, "step": 1200 }, { "grad_norm": 1.0008586645126343, "learning_rate": 9.97442210673454e-05, "loss": 0.1232, "step": 1210 }, { "grad_norm": 0.7532076239585876, "learning_rate": 9.973296464899988e-05, "loss": 0.1149, "step": 1220 }, { "grad_norm": 1.0849032402038574, "learning_rate": 9.972146650976154e-05, "loss": 0.1238, "step": 1230 }, { "grad_norm": 0.9845348596572876, "learning_rate": 9.970972670551566e-05, "loss": 0.1166, "step": 1240 }, { "grad_norm": 1.0177475214004517, "learning_rate": 9.969774529332212e-05, "loss": 0.1251, "step": 1250 }, { "grad_norm": 1.1123126745224, "learning_rate": 9.968552233141504e-05, "loss": 0.1142, "step": 1260 }, { "grad_norm": 0.8898728489875793, "learning_rate": 9.967305787920264e-05, "loss": 0.1273, "step": 1270 }, { "grad_norm": 0.9093737602233887, "learning_rate": 9.966035199726684e-05, "loss": 0.125, "step": 1280 }, { "grad_norm": 0.9875252842903137, "learning_rate": 9.9647404747363e-05, "loss": 0.1256, "step": 1290 }, { "grad_norm": 0.9476165771484375, "learning_rate": 9.96342161924196e-05, "loss": 0.1136, "step": 1300 }, { "grad_norm": 1.0349255800247192, "learning_rate": 9.962078639653797e-05, "loss": 0.114, "step": 1310 }, { "grad_norm": 0.9363329410552979, "learning_rate": 9.960711542499202e-05, "loss": 0.1228, "step": 1320 }, { "grad_norm": 0.7981815934181213, "learning_rate": 9.959320334422772e-05, "loss": 0.1092, "step": 1330 }, { "grad_norm": 0.9523260593414307, "learning_rate": 9.957905022186309e-05, "loss": 0.12, "step": 1340 }, { "grad_norm": 1.0835494995117188, "learning_rate": 9.956465612668757e-05, "loss": 0.1093, "step": 1350 }, { "grad_norm": 1.0732860565185547, "learning_rate": 9.95500211286619e-05, "loss": 0.1203, "step": 1360 }, { "grad_norm": 0.9584275484085083, "learning_rate": 9.953514529891763e-05, "loss": 0.1008, "step": 1370 }, { "grad_norm": 0.8816396594047546, "learning_rate": 9.952002870975693e-05, "loss": 0.1155, "step": 1380 }, { "grad_norm": 0.9119324684143066, "learning_rate": 9.950467143465207e-05, "loss": 0.1081, "step": 1390 }, { "grad_norm": 0.9153920412063599, "learning_rate": 9.94890735482452e-05, "loss": 0.1174, "step": 1400 }, { "grad_norm": 0.9040220975875854, "learning_rate": 9.947323512634788e-05, "loss": 0.1035, "step": 1410 }, { "grad_norm": 1.02975332736969, "learning_rate": 9.945715624594081e-05, "loss": 0.1168, "step": 1420 }, { "grad_norm": 0.9844525456428528, "learning_rate": 9.944083698517339e-05, "loss": 0.1127, "step": 1430 }, { "grad_norm": 0.7640677094459534, "learning_rate": 9.942427742336334e-05, "loss": 0.1149, "step": 1440 }, { "grad_norm": 0.9592134356498718, "learning_rate": 9.940747764099638e-05, "loss": 0.1193, "step": 1450 }, { "grad_norm": 0.9228342175483704, "learning_rate": 9.939043771972574e-05, "loss": 0.112, "step": 1460 }, { "grad_norm": 0.7710512280464172, "learning_rate": 9.937315774237186e-05, "loss": 0.1089, "step": 1470 }, { "grad_norm": 0.759921669960022, "learning_rate": 9.93556377929219e-05, "loss": 0.1154, "step": 1480 }, { "grad_norm": 1.0939874649047852, "learning_rate": 9.933787795652942e-05, "loss": 0.1077, "step": 1490 }, { "grad_norm": 0.9302868843078613, "learning_rate": 9.931987831951386e-05, "loss": 0.1248, "step": 1500 }, { "grad_norm": 0.9951125979423523, "learning_rate": 9.930163896936027e-05, "loss": 0.1109, "step": 1510 }, { "grad_norm": 0.9155976176261902, "learning_rate": 9.92831599947187e-05, "loss": 0.1123, "step": 1520 }, { "grad_norm": 0.822991669178009, "learning_rate": 9.926444148540393e-05, "loss": 0.108, "step": 1530 }, { "grad_norm": 0.7966244220733643, "learning_rate": 9.924548353239495e-05, "loss": 0.1185, "step": 1540 }, { "grad_norm": 0.7267993688583374, "learning_rate": 9.922628622783451e-05, "loss": 0.1129, "step": 1550 }, { "grad_norm": 0.834526777267456, "learning_rate": 9.920684966502878e-05, "loss": 0.1064, "step": 1560 }, { "grad_norm": 0.7787484526634216, "learning_rate": 9.918717393844669e-05, "loss": 0.0978, "step": 1570 }, { "grad_norm": 0.9177464842796326, "learning_rate": 9.916725914371969e-05, "loss": 0.1, "step": 1580 }, { "grad_norm": 0.9456977844238281, "learning_rate": 9.914710537764117e-05, "loss": 0.1092, "step": 1590 }, { "grad_norm": 0.7702245116233826, "learning_rate": 9.912671273816601e-05, "loss": 0.1105, "step": 1600 }, { "grad_norm": 0.7368134260177612, "learning_rate": 9.910608132441008e-05, "loss": 0.0979, "step": 1610 }, { "grad_norm": 0.9000690579414368, "learning_rate": 9.908521123664981e-05, "loss": 0.0934, "step": 1620 }, { "grad_norm": 0.8163759112358093, "learning_rate": 9.906410257632168e-05, "loss": 0.0994, "step": 1630 }, { "grad_norm": 0.7865023612976074, "learning_rate": 9.904275544602169e-05, "loss": 0.1008, "step": 1640 }, { "grad_norm": 0.8050644993782043, "learning_rate": 9.902116994950493e-05, "loss": 0.111, "step": 1650 }, { "grad_norm": 0.9535992741584778, "learning_rate": 9.899934619168501e-05, "loss": 0.09, "step": 1660 }, { "grad_norm": 0.7359194755554199, "learning_rate": 9.89772842786336e-05, "loss": 0.1015, "step": 1670 }, { "grad_norm": 0.9039879441261292, "learning_rate": 9.895498431757989e-05, "loss": 0.1075, "step": 1680 }, { "grad_norm": 0.8281698822975159, "learning_rate": 9.893244641691006e-05, "loss": 0.1007, "step": 1690 }, { "grad_norm": 0.7881840467453003, "learning_rate": 9.890967068616677e-05, "loss": 0.0998, "step": 1700 }, { "grad_norm": 0.845009982585907, "learning_rate": 9.888665723604864e-05, "loss": 0.0993, "step": 1710 }, { "grad_norm": 0.7727966904640198, "learning_rate": 9.886340617840968e-05, "loss": 0.1102, "step": 1720 }, { "grad_norm": 0.7697362303733826, "learning_rate": 9.883991762625876e-05, "loss": 0.11, "step": 1730 }, { "grad_norm": 0.9101146459579468, "learning_rate": 9.881619169375908e-05, "loss": 0.1039, "step": 1740 }, { "grad_norm": 0.7709200382232666, "learning_rate": 9.879222849622758e-05, "loss": 0.1048, "step": 1750 }, { "grad_norm": 0.9604688286781311, "learning_rate": 9.876802815013439e-05, "loss": 0.1049, "step": 1760 }, { "grad_norm": 0.7826374173164368, "learning_rate": 9.87435907731023e-05, "loss": 0.0997, "step": 1770 }, { "grad_norm": 0.7154682874679565, "learning_rate": 9.871891648390614e-05, "loss": 0.0983, "step": 1780 }, { "grad_norm": 0.7230797410011292, "learning_rate": 9.869400540247223e-05, "loss": 0.1047, "step": 1790 }, { "grad_norm": 1.0190340280532837, "learning_rate": 9.866885764987776e-05, "loss": 0.1018, "step": 1800 }, { "grad_norm": 0.9111382961273193, "learning_rate": 9.86434733483503e-05, "loss": 0.0991, "step": 1810 }, { "grad_norm": 0.8749095797538757, "learning_rate": 9.861785262126705e-05, "loss": 0.0959, "step": 1820 }, { "grad_norm": 0.8093475699424744, "learning_rate": 9.85919955931544e-05, "loss": 0.098, "step": 1830 }, { "grad_norm": 0.8200317025184631, "learning_rate": 9.856590238968721e-05, "loss": 0.089, "step": 1840 }, { "grad_norm": 0.8023530840873718, "learning_rate": 9.853957313768824e-05, "loss": 0.0936, "step": 1850 }, { "grad_norm": 1.0508698225021362, "learning_rate": 9.851300796512755e-05, "loss": 0.107, "step": 1860 }, { "grad_norm": 0.7715315222740173, "learning_rate": 9.848620700112188e-05, "loss": 0.1098, "step": 1870 }, { "grad_norm": 0.7154637575149536, "learning_rate": 9.845917037593396e-05, "loss": 0.0902, "step": 1880 }, { "grad_norm": 0.7053757905960083, "learning_rate": 9.843189822097196e-05, "loss": 0.087, "step": 1890 }, { "grad_norm": 0.7190327048301697, "learning_rate": 9.84043906687888e-05, "loss": 0.1095, "step": 1900 }, { "grad_norm": 0.72620689868927, "learning_rate": 9.837664785308149e-05, "loss": 0.1024, "step": 1910 }, { "grad_norm": 0.8315520286560059, "learning_rate": 9.834866990869059e-05, "loss": 0.0963, "step": 1920 }, { "grad_norm": 0.9104861617088318, "learning_rate": 9.832045697159938e-05, "loss": 0.1051, "step": 1930 }, { "grad_norm": 0.7454295754432678, "learning_rate": 9.829200917893334e-05, "loss": 0.101, "step": 1940 }, { "grad_norm": 0.7716201543807983, "learning_rate": 9.826332666895944e-05, "loss": 0.1022, "step": 1950 }, { "grad_norm": 1.0068689584732056, "learning_rate": 9.823440958108545e-05, "loss": 0.0938, "step": 1960 }, { "grad_norm": 0.813112199306488, "learning_rate": 9.820525805585927e-05, "loss": 0.1067, "step": 1970 }, { "grad_norm": 0.8323954939842224, "learning_rate": 9.81758722349683e-05, "loss": 0.1124, "step": 1980 }, { "grad_norm": 0.6970189213752747, "learning_rate": 9.814625226123862e-05, "loss": 0.0982, "step": 1990 }, { "grad_norm": 0.7715316414833069, "learning_rate": 9.811639827863449e-05, "loss": 0.0924, "step": 2000 }, { "grad_norm": 0.9239809513092041, "learning_rate": 9.808631043225741e-05, "loss": 0.09, "step": 2010 }, { "grad_norm": 0.7948368191719055, "learning_rate": 9.805598886834567e-05, "loss": 0.0971, "step": 2020 }, { "grad_norm": 0.7786247134208679, "learning_rate": 9.802543373427344e-05, "loss": 0.101, "step": 2030 }, { "grad_norm": 0.6877073645591736, "learning_rate": 9.799464517855018e-05, "loss": 0.1028, "step": 2040 }, { "grad_norm": 0.7866289615631104, "learning_rate": 9.79636233508198e-05, "loss": 0.0941, "step": 2050 }, { "grad_norm": 0.7173218727111816, "learning_rate": 9.793236840186005e-05, "loss": 0.0889, "step": 2060 }, { "grad_norm": 0.6833612322807312, "learning_rate": 9.790088048358175e-05, "loss": 0.088, "step": 2070 }, { "grad_norm": 0.6063569784164429, "learning_rate": 9.786915974902798e-05, "loss": 0.0874, "step": 2080 }, { "grad_norm": 0.7039399743080139, "learning_rate": 9.783720635237343e-05, "loss": 0.0894, "step": 2090 }, { "grad_norm": 0.861274242401123, "learning_rate": 9.780502044892362e-05, "loss": 0.0906, "step": 2100 }, { "grad_norm": 0.9218710660934448, "learning_rate": 9.777260219511415e-05, "loss": 0.0934, "step": 2110 }, { "grad_norm": 0.8045639991760254, "learning_rate": 9.773995174850989e-05, "loss": 0.0876, "step": 2120 }, { "grad_norm": 0.7548984289169312, "learning_rate": 9.770706926780428e-05, "loss": 0.0923, "step": 2130 }, { "grad_norm": 0.7165963053703308, "learning_rate": 9.767395491281855e-05, "loss": 0.0925, "step": 2140 }, { "grad_norm": 0.7390260100364685, "learning_rate": 9.764060884450086e-05, "loss": 0.0991, "step": 2150 }, { "grad_norm": 0.8069873452186584, "learning_rate": 9.76070312249257e-05, "loss": 0.0936, "step": 2160 }, { "grad_norm": 0.583582878112793, "learning_rate": 9.757322221729283e-05, "loss": 0.0849, "step": 2170 }, { "grad_norm": 0.7859311103820801, "learning_rate": 9.753918198592682e-05, "loss": 0.1005, "step": 2180 }, { "grad_norm": 0.7041631937026978, "learning_rate": 9.750491069627593e-05, "loss": 0.0848, "step": 2190 }, { "grad_norm": 0.687656044960022, "learning_rate": 9.747040851491149e-05, "loss": 0.0879, "step": 2200 }, { "grad_norm": 0.7902995944023132, "learning_rate": 9.743567560952711e-05, "loss": 0.0868, "step": 2210 }, { "grad_norm": 0.858651340007782, "learning_rate": 9.740071214893773e-05, "loss": 0.0851, "step": 2220 }, { "grad_norm": 0.7264502644538879, "learning_rate": 9.736551830307892e-05, "loss": 0.0962, "step": 2230 }, { "grad_norm": 0.7296851873397827, "learning_rate": 9.733009424300597e-05, "loss": 0.0815, "step": 2240 }, { "grad_norm": 0.7284076809883118, "learning_rate": 9.729444014089314e-05, "loss": 0.092, "step": 2250 }, { "grad_norm": 0.7667093873023987, "learning_rate": 9.725855617003275e-05, "loss": 0.0928, "step": 2260 }, { "grad_norm": 0.7198819518089294, "learning_rate": 9.72224425048344e-05, "loss": 0.0849, "step": 2270 }, { "grad_norm": 0.6823953986167908, "learning_rate": 9.718609932082405e-05, "loss": 0.0977, "step": 2280 }, { "grad_norm": 0.8046625852584839, "learning_rate": 9.714952679464323e-05, "loss": 0.0983, "step": 2290 }, { "grad_norm": 0.7668691873550415, "learning_rate": 9.711272510404816e-05, "loss": 0.0884, "step": 2300 }, { "grad_norm": 0.8391445875167847, "learning_rate": 9.70756944279089e-05, "loss": 0.105, "step": 2310 }, { "grad_norm": 0.8058165907859802, "learning_rate": 9.70384349462084e-05, "loss": 0.0916, "step": 2320 }, { "grad_norm": 0.812201201915741, "learning_rate": 9.700094684004182e-05, "loss": 0.1023, "step": 2330 }, { "grad_norm": 0.7463688254356384, "learning_rate": 9.696323029161535e-05, "loss": 0.0909, "step": 2340 }, { "grad_norm": 0.6171747446060181, "learning_rate": 9.692528548424567e-05, "loss": 0.0944, "step": 2350 }, { "grad_norm": 0.8733457326889038, "learning_rate": 9.688711260235872e-05, "loss": 0.0896, "step": 2360 }, { "grad_norm": 0.6366527676582336, "learning_rate": 9.684871183148912e-05, "loss": 0.0874, "step": 2370 }, { "grad_norm": 0.9249413013458252, "learning_rate": 9.681008335827898e-05, "loss": 0.088, "step": 2380 }, { "grad_norm": 0.8163881897926331, "learning_rate": 9.677122737047724e-05, "loss": 0.101, "step": 2390 }, { "grad_norm": 0.8050516247749329, "learning_rate": 9.673214405693857e-05, "loss": 0.099, "step": 2400 }, { "grad_norm": 0.6908136010169983, "learning_rate": 9.669283360762258e-05, "loss": 0.0869, "step": 2410 }, { "grad_norm": 0.6969667077064514, "learning_rate": 9.66532962135928e-05, "loss": 0.0897, "step": 2420 }, { "grad_norm": 0.8575372695922852, "learning_rate": 9.661353206701582e-05, "loss": 0.0854, "step": 2430 }, { "grad_norm": 0.8028751611709595, "learning_rate": 9.657354136116035e-05, "loss": 0.0941, "step": 2440 }, { "grad_norm": 0.7840549349784851, "learning_rate": 9.653332429039625e-05, "loss": 0.0916, "step": 2450 }, { "grad_norm": 0.5778301954269409, "learning_rate": 9.649288105019356e-05, "loss": 0.0983, "step": 2460 }, { "grad_norm": 0.7935901284217834, "learning_rate": 9.645221183712165e-05, "loss": 0.0911, "step": 2470 }, { "grad_norm": 0.7805438041687012, "learning_rate": 9.641131684884817e-05, "loss": 0.096, "step": 2480 }, { "grad_norm": 0.8421733975410461, "learning_rate": 9.637019628413813e-05, "loss": 0.0803, "step": 2490 }, { "grad_norm": 0.602931559085846, "learning_rate": 9.632885034285291e-05, "loss": 0.092, "step": 2500 }, { "grad_norm": 0.765745222568512, "learning_rate": 9.628727922594931e-05, "loss": 0.092, "step": 2510 }, { "grad_norm": 0.6613947749137878, "learning_rate": 9.624548313547862e-05, "loss": 0.0828, "step": 2520 }, { "grad_norm": 0.6143419742584229, "learning_rate": 9.620346227458547e-05, "loss": 0.0838, "step": 2530 }, { "grad_norm": 0.6555906534194946, "learning_rate": 9.616121684750712e-05, "loss": 0.0907, "step": 2540 }, { "grad_norm": 0.7640756964683533, "learning_rate": 9.611874705957215e-05, "loss": 0.1019, "step": 2550 }, { "grad_norm": 0.7113750576972961, "learning_rate": 9.607605311719972e-05, "loss": 0.0929, "step": 2560 }, { "grad_norm": 0.7020587921142578, "learning_rate": 9.603313522789841e-05, "loss": 0.0862, "step": 2570 }, { "grad_norm": 0.7489782571792603, "learning_rate": 9.598999360026529e-05, "loss": 0.0862, "step": 2580 }, { "grad_norm": 0.6860311627388, "learning_rate": 9.59466284439849e-05, "loss": 0.0918, "step": 2590 }, { "grad_norm": 0.639656126499176, "learning_rate": 9.590303996982815e-05, "loss": 0.0834, "step": 2600 }, { "grad_norm": 0.6633687019348145, "learning_rate": 9.585922838965145e-05, "loss": 0.0829, "step": 2610 }, { "grad_norm": 0.8332882523536682, "learning_rate": 9.581519391639549e-05, "loss": 0.0843, "step": 2620 }, { "grad_norm": 0.8247968554496765, "learning_rate": 9.577093676408439e-05, "loss": 0.0806, "step": 2630 }, { "grad_norm": 0.6226173639297485, "learning_rate": 9.572645714782453e-05, "loss": 0.0754, "step": 2640 }, { "grad_norm": 0.7613680958747864, "learning_rate": 9.568175528380354e-05, "loss": 0.0876, "step": 2650 }, { "grad_norm": 0.7251188158988953, "learning_rate": 9.56368313892893e-05, "loss": 0.0965, "step": 2660 }, { "grad_norm": 0.6383359432220459, "learning_rate": 9.55916856826288e-05, "loss": 0.0869, "step": 2670 }, { "grad_norm": 0.6451210379600525, "learning_rate": 9.554631838324713e-05, "loss": 0.0884, "step": 2680 }, { "grad_norm": 0.629679799079895, "learning_rate": 9.55007297116464e-05, "loss": 0.0833, "step": 2690 }, { "grad_norm": 0.7963083386421204, "learning_rate": 9.545491988940472e-05, "loss": 0.0919, "step": 2700 }, { "grad_norm": 0.6921216249465942, "learning_rate": 9.540888913917501e-05, "loss": 0.0884, "step": 2710 }, { "grad_norm": 0.6248165965080261, "learning_rate": 9.536263768468401e-05, "loss": 0.0835, "step": 2720 }, { "grad_norm": 0.5854783058166504, "learning_rate": 9.531616575073117e-05, "loss": 0.0821, "step": 2730 }, { "grad_norm": 0.7334324717521667, "learning_rate": 9.526947356318754e-05, "loss": 0.0856, "step": 2740 }, { "grad_norm": 0.7493296265602112, "learning_rate": 9.52225613489947e-05, "loss": 0.0772, "step": 2750 }, { "grad_norm": 0.8404133319854736, "learning_rate": 9.517542933616365e-05, "loss": 0.0819, "step": 2760 }, { "grad_norm": 0.7154518961906433, "learning_rate": 9.512807775377366e-05, "loss": 0.0791, "step": 2770 }, { "grad_norm": 0.7117947340011597, "learning_rate": 9.508050683197121e-05, "loss": 0.0836, "step": 2780 }, { "grad_norm": 0.6990528106689453, "learning_rate": 9.503271680196888e-05, "loss": 0.0769, "step": 2790 }, { "grad_norm": 0.9077825546264648, "learning_rate": 9.498470789604413e-05, "loss": 0.0871, "step": 2800 }, { "grad_norm": 0.6880998611450195, "learning_rate": 9.49364803475383e-05, "loss": 0.0885, "step": 2810 }, { "grad_norm": 0.5601368546485901, "learning_rate": 9.48880343908554e-05, "loss": 0.0745, "step": 2820 }, { "grad_norm": 0.9818617105484009, "learning_rate": 9.4839370261461e-05, "loss": 0.084, "step": 2830 }, { "grad_norm": 0.6174105405807495, "learning_rate": 9.479048819588098e-05, "loss": 0.0885, "step": 2840 }, { "grad_norm": 0.6585171818733215, "learning_rate": 9.474138843170063e-05, "loss": 0.0889, "step": 2850 }, { "grad_norm": 0.6053249835968018, "learning_rate": 9.46920712075632e-05, "loss": 0.0814, "step": 2860 }, { "grad_norm": 0.6319931149482727, "learning_rate": 9.464253676316893e-05, "loss": 0.0793, "step": 2870 }, { "grad_norm": 0.7367843985557556, "learning_rate": 9.459278533927384e-05, "loss": 0.0832, "step": 2880 }, { "grad_norm": 0.8618568181991577, "learning_rate": 9.454281717768854e-05, "loss": 0.0853, "step": 2890 }, { "grad_norm": 0.66495281457901, "learning_rate": 9.449263252127708e-05, "loss": 0.0889, "step": 2900 }, { "grad_norm": 0.7153729200363159, "learning_rate": 9.444223161395573e-05, "loss": 0.1072, "step": 2910 }, { "grad_norm": 0.9363037943840027, "learning_rate": 9.439161470069184e-05, "loss": 0.0845, "step": 2920 }, { "grad_norm": 0.6011478304862976, "learning_rate": 9.43407820275026e-05, "loss": 0.087, "step": 2930 }, { "grad_norm": 0.6171697974205017, "learning_rate": 9.428973384145396e-05, "loss": 0.0825, "step": 2940 }, { "grad_norm": 0.6081671714782715, "learning_rate": 9.423847039065922e-05, "loss": 0.079, "step": 2950 }, { "grad_norm": 0.760906994342804, "learning_rate": 9.418699192427805e-05, "loss": 0.0957, "step": 2960 }, { "grad_norm": 0.61189204454422, "learning_rate": 9.41352986925151e-05, "loss": 0.0839, "step": 2970 }, { "grad_norm": 0.66178297996521, "learning_rate": 9.408339094661895e-05, "loss": 0.0793, "step": 2980 }, { "grad_norm": 0.9399906396865845, "learning_rate": 9.40312689388807e-05, "loss": 0.0784, "step": 2990 }, { "grad_norm": 0.5865365862846375, "learning_rate": 9.397893292263292e-05, "loss": 0.0958, "step": 3000 } ], "logging_steps": 10, "max_steps": 15000, "num_input_tokens_seen": 0, "num_train_epochs": 9223372036854775807, "save_steps": 3000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 32, "trial_name": null, "trial_params": null }