{ "best_metric": null, "best_model_checkpoint": null, "epoch": 20.0, "eval_steps": 500, "global_step": 3660, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0546448087431694, "grad_norm": 14.50916862487793, "learning_rate": 5.46448087431694e-06, "loss": 1.7462, "step": 10 }, { "epoch": 0.1092896174863388, "grad_norm": 9.00194263458252, "learning_rate": 1.092896174863388e-05, "loss": 1.397, "step": 20 }, { "epoch": 0.16393442622950818, "grad_norm": 2.302015781402588, "learning_rate": 1.6393442622950818e-05, "loss": 0.5593, "step": 30 }, { "epoch": 0.2185792349726776, "grad_norm": 2.4071569442749023, "learning_rate": 2.185792349726776e-05, "loss": 0.3376, "step": 40 }, { "epoch": 0.273224043715847, "grad_norm": 1.562657117843628, "learning_rate": 2.7322404371584703e-05, "loss": 0.2425, "step": 50 }, { "epoch": 0.32786885245901637, "grad_norm": 1.5590847730636597, "learning_rate": 3.2786885245901635e-05, "loss": 0.2093, "step": 60 }, { "epoch": 0.3825136612021858, "grad_norm": 1.3504713773727417, "learning_rate": 3.825136612021858e-05, "loss": 0.1947, "step": 70 }, { "epoch": 0.4371584699453552, "grad_norm": 1.1426506042480469, "learning_rate": 4.371584699453552e-05, "loss": 0.1745, "step": 80 }, { "epoch": 0.4918032786885246, "grad_norm": 1.2180607318878174, "learning_rate": 4.918032786885246e-05, "loss": 0.1541, "step": 90 }, { "epoch": 0.546448087431694, "grad_norm": 1.4522769451141357, "learning_rate": 5.4644808743169406e-05, "loss": 0.1492, "step": 100 }, { "epoch": 0.6010928961748634, "grad_norm": 1.8644955158233643, "learning_rate": 6.010928961748634e-05, "loss": 0.134, "step": 110 }, { "epoch": 0.6557377049180327, "grad_norm": 0.8796963095664978, "learning_rate": 6.557377049180327e-05, "loss": 0.1363, "step": 120 }, { "epoch": 0.7103825136612022, "grad_norm": 0.7654175758361816, "learning_rate": 7.103825136612023e-05, "loss": 0.1299, "step": 130 }, { "epoch": 0.7650273224043715, "grad_norm": 0.8335488438606262, "learning_rate": 7.650273224043716e-05, "loss": 0.1308, "step": 140 }, { "epoch": 0.819672131147541, "grad_norm": 1.0280165672302246, "learning_rate": 8.19672131147541e-05, "loss": 0.1134, "step": 150 }, { "epoch": 0.8743169398907104, "grad_norm": 1.133649468421936, "learning_rate": 8.743169398907104e-05, "loss": 0.1128, "step": 160 }, { "epoch": 0.9289617486338798, "grad_norm": 0.851076602935791, "learning_rate": 9.289617486338798e-05, "loss": 0.1044, "step": 170 }, { "epoch": 0.9836065573770492, "grad_norm": 0.8838961720466614, "learning_rate": 9.836065573770493e-05, "loss": 0.1092, "step": 180 }, { "epoch": 1.0382513661202186, "grad_norm": 0.7312606573104858, "learning_rate": 9.999899994242224e-05, "loss": 0.1036, "step": 190 }, { "epoch": 1.092896174863388, "grad_norm": 0.835260808467865, "learning_rate": 9.999410179752908e-05, "loss": 0.0993, "step": 200 }, { "epoch": 1.1475409836065573, "grad_norm": 0.8507461547851562, "learning_rate": 9.998512228064598e-05, "loss": 0.0938, "step": 210 }, { "epoch": 1.2021857923497268, "grad_norm": 0.8367434740066528, "learning_rate": 9.99720621248344e-05, "loss": 0.0874, "step": 220 }, { "epoch": 1.2568306010928962, "grad_norm": 0.8626418113708496, "learning_rate": 9.995492239628717e-05, "loss": 0.0921, "step": 230 }, { "epoch": 1.3114754098360657, "grad_norm": 0.94256991147995, "learning_rate": 9.993370449424153e-05, "loss": 0.0874, "step": 240 }, { "epoch": 1.366120218579235, "grad_norm": 0.7513673305511475, "learning_rate": 9.990841015086483e-05, "loss": 0.0856, "step": 250 }, { "epoch": 1.4207650273224044, "grad_norm": 0.6465060114860535, "learning_rate": 9.987904143111314e-05, "loss": 0.0879, "step": 260 }, { "epoch": 1.4754098360655736, "grad_norm": 0.8078054785728455, "learning_rate": 9.984560073256271e-05, "loss": 0.0913, "step": 270 }, { "epoch": 1.530054644808743, "grad_norm": 0.4887768030166626, "learning_rate": 9.980809078521416e-05, "loss": 0.0848, "step": 280 }, { "epoch": 1.5846994535519126, "grad_norm": 0.593806266784668, "learning_rate": 9.976651465126973e-05, "loss": 0.0805, "step": 290 }, { "epoch": 1.639344262295082, "grad_norm": 0.7595932483673096, "learning_rate": 9.972087572488315e-05, "loss": 0.085, "step": 300 }, { "epoch": 1.6939890710382515, "grad_norm": 0.5838643312454224, "learning_rate": 9.967117773188265e-05, "loss": 0.0802, "step": 310 }, { "epoch": 1.748633879781421, "grad_norm": 1.192397117614746, "learning_rate": 9.96174247294668e-05, "loss": 0.0803, "step": 320 }, { "epoch": 1.8032786885245902, "grad_norm": 0.6883787512779236, "learning_rate": 9.955962110587318e-05, "loss": 0.0771, "step": 330 }, { "epoch": 1.8579234972677594, "grad_norm": 0.5919109582901001, "learning_rate": 9.949777158002027e-05, "loss": 0.071, "step": 340 }, { "epoch": 1.9125683060109289, "grad_norm": 0.7835716605186462, "learning_rate": 9.943188120112214e-05, "loss": 0.0865, "step": 350 }, { "epoch": 1.9672131147540983, "grad_norm": 0.6744383573532104, "learning_rate": 9.936195534827629e-05, "loss": 0.081, "step": 360 }, { "epoch": 2.021857923497268, "grad_norm": 0.5261861681938171, "learning_rate": 9.928799973002445e-05, "loss": 0.0726, "step": 370 }, { "epoch": 2.0765027322404372, "grad_norm": 0.7296725511550903, "learning_rate": 9.92100203838866e-05, "loss": 0.0723, "step": 380 }, { "epoch": 2.1311475409836067, "grad_norm": 0.5582314729690552, "learning_rate": 9.91280236758681e-05, "loss": 0.0712, "step": 390 }, { "epoch": 2.185792349726776, "grad_norm": 0.41796624660491943, "learning_rate": 9.904201629993993e-05, "loss": 0.0833, "step": 400 }, { "epoch": 2.240437158469945, "grad_norm": 0.46099790930747986, "learning_rate": 9.895200527749227e-05, "loss": 0.0651, "step": 410 }, { "epoch": 2.2950819672131146, "grad_norm": 0.5075456500053406, "learning_rate": 9.885799795676126e-05, "loss": 0.0675, "step": 420 }, { "epoch": 2.349726775956284, "grad_norm": 0.5505486130714417, "learning_rate": 9.876000201222912e-05, "loss": 0.0672, "step": 430 }, { "epoch": 2.4043715846994536, "grad_norm": 1.0104068517684937, "learning_rate": 9.865802544399767e-05, "loss": 0.0772, "step": 440 }, { "epoch": 2.459016393442623, "grad_norm": 0.9975788593292236, "learning_rate": 9.855207657713511e-05, "loss": 0.0707, "step": 450 }, { "epoch": 2.5136612021857925, "grad_norm": 0.5779078006744385, "learning_rate": 9.844216406099654e-05, "loss": 0.073, "step": 460 }, { "epoch": 2.5683060109289615, "grad_norm": 0.8848289847373962, "learning_rate": 9.83282968685177e-05, "loss": 0.0653, "step": 470 }, { "epoch": 2.6229508196721314, "grad_norm": 0.6772857308387756, "learning_rate": 9.821048429548258e-05, "loss": 0.0659, "step": 480 }, { "epoch": 2.6775956284153004, "grad_norm": 0.5181351900100708, "learning_rate": 9.808873595976447e-05, "loss": 0.0638, "step": 490 }, { "epoch": 2.73224043715847, "grad_norm": 0.7094351053237915, "learning_rate": 9.796306180054076e-05, "loss": 0.0614, "step": 500 }, { "epoch": 2.7868852459016393, "grad_norm": 0.5130027532577515, "learning_rate": 9.783347207748158e-05, "loss": 0.0625, "step": 510 }, { "epoch": 2.841530054644809, "grad_norm": 0.7600369453430176, "learning_rate": 9.769997736991226e-05, "loss": 0.0671, "step": 520 }, { "epoch": 2.8961748633879782, "grad_norm": 0.6283718943595886, "learning_rate": 9.756258857594956e-05, "loss": 0.0631, "step": 530 }, { "epoch": 2.9508196721311473, "grad_norm": 0.5392641425132751, "learning_rate": 9.74213169116121e-05, "loss": 0.0652, "step": 540 }, { "epoch": 3.0054644808743167, "grad_norm": 0.33834314346313477, "learning_rate": 9.72761739099046e-05, "loss": 0.0597, "step": 550 }, { "epoch": 3.060109289617486, "grad_norm": 0.4568038284778595, "learning_rate": 9.712717141987651e-05, "loss": 0.0666, "step": 560 }, { "epoch": 3.1147540983606556, "grad_norm": 0.7953147888183594, "learning_rate": 9.69743216056545e-05, "loss": 0.0631, "step": 570 }, { "epoch": 3.169398907103825, "grad_norm": 0.8522589206695557, "learning_rate": 9.681763694544957e-05, "loss": 0.061, "step": 580 }, { "epoch": 3.2240437158469946, "grad_norm": 0.8012239933013916, "learning_rate": 9.66571302305383e-05, "loss": 0.0626, "step": 590 }, { "epoch": 3.278688524590164, "grad_norm": 0.6893075108528137, "learning_rate": 9.649281456421862e-05, "loss": 0.0546, "step": 600 }, { "epoch": 3.3333333333333335, "grad_norm": 0.5718370079994202, "learning_rate": 9.632470336074009e-05, "loss": 0.0563, "step": 610 }, { "epoch": 3.387978142076503, "grad_norm": 0.8569139242172241, "learning_rate": 9.61528103442088e-05, "loss": 0.0613, "step": 620 }, { "epoch": 3.442622950819672, "grad_norm": 0.685554563999176, "learning_rate": 9.5977149547467e-05, "loss": 0.0615, "step": 630 }, { "epoch": 3.4972677595628414, "grad_norm": 0.7499738335609436, "learning_rate": 9.579773531094742e-05, "loss": 0.0635, "step": 640 }, { "epoch": 3.551912568306011, "grad_norm": 0.9926331043243408, "learning_rate": 9.561458228150266e-05, "loss": 0.0627, "step": 650 }, { "epoch": 3.6065573770491803, "grad_norm": 0.6195826530456543, "learning_rate": 9.542770541120945e-05, "loss": 0.0591, "step": 660 }, { "epoch": 3.66120218579235, "grad_norm": 0.4464719891548157, "learning_rate": 9.523711995614788e-05, "loss": 0.0552, "step": 670 }, { "epoch": 3.7158469945355193, "grad_norm": 1.1997320652008057, "learning_rate": 9.504284147515615e-05, "loss": 0.0593, "step": 680 }, { "epoch": 3.7704918032786887, "grad_norm": 0.5045487284660339, "learning_rate": 9.484488582856018e-05, "loss": 0.0551, "step": 690 }, { "epoch": 3.8251366120218577, "grad_norm": 0.5805765986442566, "learning_rate": 9.464326917687898e-05, "loss": 0.061, "step": 700 }, { "epoch": 3.879781420765027, "grad_norm": 0.7676146030426025, "learning_rate": 9.443800797950523e-05, "loss": 0.055, "step": 710 }, { "epoch": 3.9344262295081966, "grad_norm": 0.600481390953064, "learning_rate": 9.422911899336174e-05, "loss": 0.0566, "step": 720 }, { "epoch": 3.989071038251366, "grad_norm": 0.6507868766784668, "learning_rate": 9.401661927153323e-05, "loss": 0.0598, "step": 730 }, { "epoch": 4.043715846994536, "grad_norm": 0.6208195090293884, "learning_rate": 9.380052616187441e-05, "loss": 0.0622, "step": 740 }, { "epoch": 4.098360655737705, "grad_norm": 0.45560985803604126, "learning_rate": 9.35808573055936e-05, "loss": 0.0572, "step": 750 }, { "epoch": 4.1530054644808745, "grad_norm": 0.5124120116233826, "learning_rate": 9.335763063581262e-05, "loss": 0.0511, "step": 760 }, { "epoch": 4.2076502732240435, "grad_norm": 0.6840647459030151, "learning_rate": 9.313086437610273e-05, "loss": 0.0549, "step": 770 }, { "epoch": 4.262295081967213, "grad_norm": 0.5145549178123474, "learning_rate": 9.290057703899697e-05, "loss": 0.0568, "step": 780 }, { "epoch": 4.316939890710382, "grad_norm": 0.6335499286651611, "learning_rate": 9.266678742447878e-05, "loss": 0.0505, "step": 790 }, { "epoch": 4.371584699453552, "grad_norm": 0.3851253092288971, "learning_rate": 9.242951461844734e-05, "loss": 0.0491, "step": 800 }, { "epoch": 4.426229508196721, "grad_norm": 0.5057039856910706, "learning_rate": 9.218877799115928e-05, "loss": 0.051, "step": 810 }, { "epoch": 4.48087431693989, "grad_norm": 0.6589703559875488, "learning_rate": 9.194459719564755e-05, "loss": 0.0564, "step": 820 }, { "epoch": 4.53551912568306, "grad_norm": 0.5856819152832031, "learning_rate": 9.169699216611683e-05, "loss": 0.0547, "step": 830 }, { "epoch": 4.590163934426229, "grad_norm": 0.4095700681209564, "learning_rate": 9.144598311631626e-05, "loss": 0.0566, "step": 840 }, { "epoch": 4.644808743169399, "grad_norm": 0.4853191673755646, "learning_rate": 9.119159053788922e-05, "loss": 0.0559, "step": 850 }, { "epoch": 4.699453551912568, "grad_norm": 0.2956223487854004, "learning_rate": 9.093383519870047e-05, "loss": 0.055, "step": 860 }, { "epoch": 4.754098360655737, "grad_norm": 0.8387729525566101, "learning_rate": 9.067273814114066e-05, "loss": 0.0483, "step": 870 }, { "epoch": 4.808743169398907, "grad_norm": 0.5920037627220154, "learning_rate": 9.040832068040857e-05, "loss": 0.0483, "step": 880 }, { "epoch": 4.863387978142076, "grad_norm": 0.6177905201911926, "learning_rate": 9.014060440277097e-05, "loss": 0.0506, "step": 890 }, { "epoch": 4.918032786885246, "grad_norm": 0.5852020978927612, "learning_rate": 8.986961116380037e-05, "loss": 0.0494, "step": 900 }, { "epoch": 4.972677595628415, "grad_norm": 0.6510049700737, "learning_rate": 8.95953630865908e-05, "loss": 0.0481, "step": 910 }, { "epoch": 5.027322404371585, "grad_norm": 0.7334024906158447, "learning_rate": 8.931788255995175e-05, "loss": 0.0465, "step": 920 }, { "epoch": 5.081967213114754, "grad_norm": 0.8580073714256287, "learning_rate": 8.903719223658038e-05, "loss": 0.0502, "step": 930 }, { "epoch": 5.136612021857924, "grad_norm": 0.7086485624313354, "learning_rate": 8.875331503121231e-05, "loss": 0.0514, "step": 940 }, { "epoch": 5.191256830601093, "grad_norm": 0.49265387654304504, "learning_rate": 8.846627411875081e-05, "loss": 0.0499, "step": 950 }, { "epoch": 5.245901639344262, "grad_norm": 0.5628988742828369, "learning_rate": 8.817609293237499e-05, "loss": 0.0504, "step": 960 }, { "epoch": 5.300546448087432, "grad_norm": 1.3476946353912354, "learning_rate": 8.788279516162666e-05, "loss": 0.0496, "step": 970 }, { "epoch": 5.355191256830601, "grad_norm": 0.4882979393005371, "learning_rate": 8.758640475047649e-05, "loss": 0.0511, "step": 980 }, { "epoch": 5.409836065573771, "grad_norm": 0.3951756954193115, "learning_rate": 8.72869458953692e-05, "loss": 0.0529, "step": 990 }, { "epoch": 5.46448087431694, "grad_norm": 0.4163540303707123, "learning_rate": 8.698444304324835e-05, "loss": 0.0491, "step": 1000 }, { "epoch": 5.51912568306011, "grad_norm": 0.6048452258110046, "learning_rate": 8.667892088956045e-05, "loss": 0.0503, "step": 1010 }, { "epoch": 5.573770491803279, "grad_norm": 0.5346564054489136, "learning_rate": 8.637040437623897e-05, "loss": 0.0449, "step": 1020 }, { "epoch": 5.628415300546449, "grad_norm": 0.35331907868385315, "learning_rate": 8.60589186896681e-05, "loss": 0.0435, "step": 1030 }, { "epoch": 5.683060109289618, "grad_norm": 0.4379464387893677, "learning_rate": 8.574448925862667e-05, "loss": 0.0527, "step": 1040 }, { "epoch": 5.737704918032787, "grad_norm": 0.45317304134368896, "learning_rate": 8.542714175221216e-05, "loss": 0.0463, "step": 1050 }, { "epoch": 5.7923497267759565, "grad_norm": 0.374124675989151, "learning_rate": 8.510690207774517e-05, "loss": 0.044, "step": 1060 }, { "epoch": 5.8469945355191255, "grad_norm": 0.481268972158432, "learning_rate": 8.47837963786544e-05, "loss": 0.0492, "step": 1070 }, { "epoch": 5.901639344262295, "grad_norm": 0.6521291732788086, "learning_rate": 8.445785103234252e-05, "loss": 0.0535, "step": 1080 }, { "epoch": 5.956284153005464, "grad_norm": 0.41385921835899353, "learning_rate": 8.412909264803252e-05, "loss": 0.0439, "step": 1090 }, { "epoch": 6.0109289617486334, "grad_norm": 0.4521181285381317, "learning_rate": 8.379754806459568e-05, "loss": 0.0509, "step": 1100 }, { "epoch": 6.065573770491803, "grad_norm": 0.295654296875, "learning_rate": 8.346324434836038e-05, "loss": 0.0425, "step": 1110 }, { "epoch": 6.120218579234972, "grad_norm": 0.4982050359249115, "learning_rate": 8.31262087909025e-05, "loss": 0.0459, "step": 1120 }, { "epoch": 6.174863387978142, "grad_norm": 0.9648367762565613, "learning_rate": 8.278646890681746e-05, "loss": 0.0501, "step": 1130 }, { "epoch": 6.229508196721311, "grad_norm": 0.6268149018287659, "learning_rate": 8.244405243147397e-05, "loss": 0.0426, "step": 1140 }, { "epoch": 6.284153005464481, "grad_norm": 1.036202311515808, "learning_rate": 8.209898731874981e-05, "loss": 0.047, "step": 1150 }, { "epoch": 6.33879781420765, "grad_norm": 0.6712212562561035, "learning_rate": 8.175130173874975e-05, "loss": 0.0485, "step": 1160 }, { "epoch": 6.39344262295082, "grad_norm": 0.4752486050128937, "learning_rate": 8.140102407550585e-05, "loss": 0.0494, "step": 1170 }, { "epoch": 6.448087431693989, "grad_norm": 0.7658908367156982, "learning_rate": 8.104818292466022e-05, "loss": 0.0422, "step": 1180 }, { "epoch": 6.502732240437158, "grad_norm": 0.4015105366706848, "learning_rate": 8.06928070911306e-05, "loss": 0.041, "step": 1190 }, { "epoch": 6.557377049180328, "grad_norm": 0.6179288029670715, "learning_rate": 8.033492558675883e-05, "loss": 0.0432, "step": 1200 }, { "epoch": 6.612021857923497, "grad_norm": 0.4464608132839203, "learning_rate": 7.997456762794232e-05, "loss": 0.0415, "step": 1210 }, { "epoch": 6.666666666666667, "grad_norm": 0.6367069482803345, "learning_rate": 7.961176263324901e-05, "loss": 0.045, "step": 1220 }, { "epoch": 6.721311475409836, "grad_norm": 0.2885402739048004, "learning_rate": 7.924654022101565e-05, "loss": 0.0424, "step": 1230 }, { "epoch": 6.775956284153006, "grad_norm": 0.8153219819068909, "learning_rate": 7.887893020692989e-05, "loss": 0.047, "step": 1240 }, { "epoch": 6.830601092896175, "grad_norm": 0.312220960855484, "learning_rate": 7.850896260159621e-05, "loss": 0.0459, "step": 1250 }, { "epoch": 6.885245901639344, "grad_norm": 0.48526033759117126, "learning_rate": 7.81366676080859e-05, "loss": 0.0452, "step": 1260 }, { "epoch": 6.939890710382514, "grad_norm": 0.7081798315048218, "learning_rate": 7.776207561947142e-05, "loss": 0.0474, "step": 1270 }, { "epoch": 6.994535519125683, "grad_norm": 0.5795289874076843, "learning_rate": 7.738521721634522e-05, "loss": 0.0367, "step": 1280 }, { "epoch": 7.049180327868853, "grad_norm": 0.6600906848907471, "learning_rate": 7.700612316432309e-05, "loss": 0.0433, "step": 1290 }, { "epoch": 7.103825136612022, "grad_norm": 0.696719765663147, "learning_rate": 7.662482441153274e-05, "loss": 0.0461, "step": 1300 }, { "epoch": 7.158469945355192, "grad_norm": 0.7991256713867188, "learning_rate": 7.624135208608714e-05, "loss": 0.039, "step": 1310 }, { "epoch": 7.213114754098361, "grad_norm": 0.8350341320037842, "learning_rate": 7.58557374935434e-05, "loss": 0.0411, "step": 1320 }, { "epoch": 7.26775956284153, "grad_norm": 0.8367980122566223, "learning_rate": 7.546801211434702e-05, "loss": 0.0457, "step": 1330 }, { "epoch": 7.3224043715847, "grad_norm": 0.7723026275634766, "learning_rate": 7.507820760126193e-05, "loss": 0.0449, "step": 1340 }, { "epoch": 7.377049180327869, "grad_norm": 0.5425959229469299, "learning_rate": 7.468635577678646e-05, "loss": 0.0389, "step": 1350 }, { "epoch": 7.4316939890710385, "grad_norm": 0.8735256195068359, "learning_rate": 7.429248863055545e-05, "loss": 0.0407, "step": 1360 }, { "epoch": 7.4863387978142075, "grad_norm": 0.5877657532691956, "learning_rate": 7.389663831672868e-05, "loss": 0.0408, "step": 1370 }, { "epoch": 7.540983606557377, "grad_norm": 0.40603581070899963, "learning_rate": 7.3498837151366e-05, "loss": 0.0377, "step": 1380 }, { "epoch": 7.595628415300546, "grad_norm": 0.4147934019565582, "learning_rate": 7.309911760978899e-05, "loss": 0.0455, "step": 1390 }, { "epoch": 7.6502732240437155, "grad_norm": 0.3599822223186493, "learning_rate": 7.269751232392983e-05, "loss": 0.0442, "step": 1400 }, { "epoch": 7.704918032786885, "grad_norm": 0.4773503243923187, "learning_rate": 7.229405407966739e-05, "loss": 0.0386, "step": 1410 }, { "epoch": 7.759562841530054, "grad_norm": 0.46479761600494385, "learning_rate": 7.188877581415059e-05, "loss": 0.0378, "step": 1420 }, { "epoch": 7.814207650273224, "grad_norm": 0.3012316823005676, "learning_rate": 7.148171061310963e-05, "loss": 0.0386, "step": 1430 }, { "epoch": 7.868852459016393, "grad_norm": 0.7883979082107544, "learning_rate": 7.107289170815481e-05, "loss": 0.0345, "step": 1440 }, { "epoch": 7.923497267759563, "grad_norm": 0.4798952341079712, "learning_rate": 7.06623524740637e-05, "loss": 0.0326, "step": 1450 }, { "epoch": 7.978142076502732, "grad_norm": 0.8079819083213806, "learning_rate": 7.02501264260565e-05, "loss": 0.0418, "step": 1460 }, { "epoch": 8.032786885245901, "grad_norm": 0.2653636634349823, "learning_rate": 6.983624721705999e-05, "loss": 0.0389, "step": 1470 }, { "epoch": 8.087431693989071, "grad_norm": 0.48421579599380493, "learning_rate": 6.94207486349601e-05, "loss": 0.0391, "step": 1480 }, { "epoch": 8.142076502732241, "grad_norm": 0.7526752352714539, "learning_rate": 6.900366459984374e-05, "loss": 0.0395, "step": 1490 }, { "epoch": 8.19672131147541, "grad_norm": 0.36561110615730286, "learning_rate": 6.858502916122943e-05, "loss": 0.0336, "step": 1500 }, { "epoch": 8.251366120218579, "grad_norm": 0.4030620753765106, "learning_rate": 6.816487649528783e-05, "loss": 0.0414, "step": 1510 }, { "epoch": 8.306010928961749, "grad_norm": 0.4381183683872223, "learning_rate": 6.774324090205155e-05, "loss": 0.0388, "step": 1520 }, { "epoch": 8.360655737704919, "grad_norm": 0.4365619122982025, "learning_rate": 6.732015680261505e-05, "loss": 0.0408, "step": 1530 }, { "epoch": 8.415300546448087, "grad_norm": 0.6897148489952087, "learning_rate": 6.689565873632458e-05, "loss": 0.0377, "step": 1540 }, { "epoch": 8.469945355191257, "grad_norm": 0.3831244111061096, "learning_rate": 6.646978135795852e-05, "loss": 0.0327, "step": 1550 }, { "epoch": 8.524590163934427, "grad_norm": 0.3431888222694397, "learning_rate": 6.604255943489822e-05, "loss": 0.0368, "step": 1560 }, { "epoch": 8.579234972677595, "grad_norm": 0.6473280191421509, "learning_rate": 6.561402784428974e-05, "loss": 0.0399, "step": 1570 }, { "epoch": 8.633879781420765, "grad_norm": 0.3908807039260864, "learning_rate": 6.518422157019658e-05, "loss": 0.0354, "step": 1580 }, { "epoch": 8.688524590163935, "grad_norm": 0.4484104812145233, "learning_rate": 6.475317570074357e-05, "loss": 0.035, "step": 1590 }, { "epoch": 8.743169398907105, "grad_norm": 0.5294371247291565, "learning_rate": 6.432092542525263e-05, "loss": 0.0373, "step": 1600 }, { "epoch": 8.797814207650273, "grad_norm": 0.35410311818122864, "learning_rate": 6.388750603136973e-05, "loss": 0.0377, "step": 1610 }, { "epoch": 8.852459016393443, "grad_norm": 0.9991357326507568, "learning_rate": 6.345295290218427e-05, "loss": 0.0373, "step": 1620 }, { "epoch": 8.907103825136613, "grad_norm": 0.47563156485557556, "learning_rate": 6.301730151334054e-05, "loss": 0.032, "step": 1630 }, { "epoch": 8.96174863387978, "grad_norm": 0.4704972803592682, "learning_rate": 6.258058743014145e-05, "loss": 0.038, "step": 1640 }, { "epoch": 9.01639344262295, "grad_norm": 0.5292912721633911, "learning_rate": 6.214284630464521e-05, "loss": 0.0362, "step": 1650 }, { "epoch": 9.07103825136612, "grad_norm": 0.6291676759719849, "learning_rate": 6.170411387275472e-05, "loss": 0.0366, "step": 1660 }, { "epoch": 9.12568306010929, "grad_norm": 0.38353613018989563, "learning_rate": 6.126442595130027e-05, "loss": 0.0382, "step": 1670 }, { "epoch": 9.180327868852459, "grad_norm": 0.45063936710357666, "learning_rate": 6.08238184351155e-05, "loss": 0.0335, "step": 1680 }, { "epoch": 9.234972677595628, "grad_norm": 0.3961513042449951, "learning_rate": 6.0382327294107065e-05, "loss": 0.0356, "step": 1690 }, { "epoch": 9.289617486338798, "grad_norm": 0.2651735544204712, "learning_rate": 5.993998857031814e-05, "loss": 0.029, "step": 1700 }, { "epoch": 9.344262295081966, "grad_norm": 1.1500636339187622, "learning_rate": 5.949683837498614e-05, "loss": 0.0337, "step": 1710 }, { "epoch": 9.398907103825136, "grad_norm": 0.4949517548084259, "learning_rate": 5.905291288559458e-05, "loss": 0.0335, "step": 1720 }, { "epoch": 9.453551912568306, "grad_norm": 0.34546446800231934, "learning_rate": 5.860824834291973e-05, "loss": 0.0292, "step": 1730 }, { "epoch": 9.508196721311476, "grad_norm": 0.3333743214607239, "learning_rate": 5.816288104807199e-05, "loss": 0.0394, "step": 1740 }, { "epoch": 9.562841530054644, "grad_norm": 0.3905448913574219, "learning_rate": 5.771684735953242e-05, "loss": 0.0369, "step": 1750 }, { "epoch": 9.617486338797814, "grad_norm": 0.4974684715270996, "learning_rate": 5.7270183690184495e-05, "loss": 0.0352, "step": 1760 }, { "epoch": 9.672131147540984, "grad_norm": 0.4600943624973297, "learning_rate": 5.682292650434147e-05, "loss": 0.0376, "step": 1770 }, { "epoch": 9.726775956284152, "grad_norm": 0.6638236045837402, "learning_rate": 5.6375112314769526e-05, "loss": 0.0349, "step": 1780 }, { "epoch": 9.781420765027322, "grad_norm": 0.8875242471694946, "learning_rate": 5.592677767970704e-05, "loss": 0.038, "step": 1790 }, { "epoch": 9.836065573770492, "grad_norm": 0.983162522315979, "learning_rate": 5.547795919988004e-05, "loss": 0.0345, "step": 1800 }, { "epoch": 9.890710382513662, "grad_norm": 0.26490020751953125, "learning_rate": 5.502869351551414e-05, "loss": 0.0343, "step": 1810 }, { "epoch": 9.94535519125683, "grad_norm": 0.5392178297042847, "learning_rate": 5.457901730334354e-05, "loss": 0.0334, "step": 1820 }, { "epoch": 10.0, "grad_norm": 0.36645573377609253, "learning_rate": 5.4128967273616625e-05, "loss": 0.0345, "step": 1830 }, { "epoch": 10.05464480874317, "grad_norm": 0.6238542199134827, "learning_rate": 5.3678580167099214e-05, "loss": 0.0299, "step": 1840 }, { "epoch": 10.109289617486338, "grad_norm": 0.3955978453159332, "learning_rate": 5.322789275207505e-05, "loss": 0.0321, "step": 1850 }, { "epoch": 10.163934426229508, "grad_norm": 0.43886300921440125, "learning_rate": 5.277694182134422e-05, "loss": 0.0297, "step": 1860 }, { "epoch": 10.218579234972678, "grad_norm": 0.3649499714374542, "learning_rate": 5.232576418921944e-05, "loss": 0.0342, "step": 1870 }, { "epoch": 10.273224043715848, "grad_norm": 0.2916063964366913, "learning_rate": 5.187439668852063e-05, "loss": 0.0272, "step": 1880 }, { "epoch": 10.327868852459016, "grad_norm": 0.5019255876541138, "learning_rate": 5.142287616756809e-05, "loss": 0.0308, "step": 1890 }, { "epoch": 10.382513661202186, "grad_norm": 0.28151676058769226, "learning_rate": 5.09712394871742e-05, "loss": 0.0383, "step": 1900 }, { "epoch": 10.437158469945356, "grad_norm": 0.7377211451530457, "learning_rate": 5.051952351763428e-05, "loss": 0.0322, "step": 1910 }, { "epoch": 10.491803278688524, "grad_norm": 1.962594985961914, "learning_rate": 5.006776513571655e-05, "loss": 0.032, "step": 1920 }, { "epoch": 10.546448087431694, "grad_norm": 0.3466651737689972, "learning_rate": 4.9616001221651693e-05, "loss": 0.0355, "step": 1930 }, { "epoch": 10.601092896174864, "grad_norm": 1.698729157447815, "learning_rate": 4.916426865612201e-05, "loss": 0.0345, "step": 1940 }, { "epoch": 10.655737704918034, "grad_norm": 0.9637352824211121, "learning_rate": 4.8712604317250576e-05, "loss": 0.0313, "step": 1950 }, { "epoch": 10.710382513661202, "grad_norm": 0.6569703817367554, "learning_rate": 4.8261045077590665e-05, "loss": 0.031, "step": 1960 }, { "epoch": 10.765027322404372, "grad_norm": 0.4290389120578766, "learning_rate": 4.780962780111554e-05, "loss": 0.0307, "step": 1970 }, { "epoch": 10.819672131147541, "grad_norm": 0.2776305079460144, "learning_rate": 4.735838934020903e-05, "loss": 0.0338, "step": 1980 }, { "epoch": 10.87431693989071, "grad_norm": 0.426277756690979, "learning_rate": 4.690736653265695e-05, "loss": 0.0271, "step": 1990 }, { "epoch": 10.92896174863388, "grad_norm": 0.890143096446991, "learning_rate": 4.6456596198639866e-05, "loss": 0.0325, "step": 2000 }, { "epoch": 10.98360655737705, "grad_norm": 0.27757057547569275, "learning_rate": 4.600611513772708e-05, "loss": 0.0313, "step": 2010 }, { "epoch": 11.03825136612022, "grad_norm": 0.5478549599647522, "learning_rate": 4.555596012587253e-05, "loss": 0.0349, "step": 2020 }, { "epoch": 11.092896174863387, "grad_norm": 0.5741052031517029, "learning_rate": 4.510616791241243e-05, "loss": 0.0351, "step": 2030 }, { "epoch": 11.147540983606557, "grad_norm": 0.6417993903160095, "learning_rate": 4.46567752170652e-05, "loss": 0.0311, "step": 2040 }, { "epoch": 11.202185792349727, "grad_norm": 0.5397584438323975, "learning_rate": 4.420781872693383e-05, "loss": 0.0355, "step": 2050 }, { "epoch": 11.256830601092895, "grad_norm": 0.19554707407951355, "learning_rate": 4.375933509351071e-05, "loss": 0.0306, "step": 2060 }, { "epoch": 11.311475409836065, "grad_norm": 0.5310783982276917, "learning_rate": 4.331136092968566e-05, "loss": 0.0321, "step": 2070 }, { "epoch": 11.366120218579235, "grad_norm": 0.25652116537094116, "learning_rate": 4.2863932806756845e-05, "loss": 0.0273, "step": 2080 }, { "epoch": 11.420765027322405, "grad_norm": 0.2970091998577118, "learning_rate": 4.241708725144529e-05, "loss": 0.0327, "step": 2090 }, { "epoch": 11.475409836065573, "grad_norm": 0.5687034726142883, "learning_rate": 4.197086074291285e-05, "loss": 0.0312, "step": 2100 }, { "epoch": 11.530054644808743, "grad_norm": 1.2950825691223145, "learning_rate": 4.152528970978432e-05, "loss": 0.0303, "step": 2110 }, { "epoch": 11.584699453551913, "grad_norm": 0.2822929620742798, "learning_rate": 4.108041052717329e-05, "loss": 0.0269, "step": 2120 }, { "epoch": 11.639344262295083, "grad_norm": 0.33775463700294495, "learning_rate": 4.063625951371278e-05, "loss": 0.0325, "step": 2130 }, { "epoch": 11.693989071038251, "grad_norm": 0.40372395515441895, "learning_rate": 4.019287292859016e-05, "loss": 0.0293, "step": 2140 }, { "epoch": 11.748633879781421, "grad_norm": 0.4413204491138458, "learning_rate": 3.975028696858713e-05, "loss": 0.0291, "step": 2150 }, { "epoch": 11.80327868852459, "grad_norm": 0.7298337817192078, "learning_rate": 3.930853776512473e-05, "loss": 0.0253, "step": 2160 }, { "epoch": 11.857923497267759, "grad_norm": 0.37207067012786865, "learning_rate": 3.8867661381313606e-05, "loss": 0.0299, "step": 2170 }, { "epoch": 11.912568306010929, "grad_norm": 0.658452570438385, "learning_rate": 3.8427693809009996e-05, "loss": 0.0294, "step": 2180 }, { "epoch": 11.967213114754099, "grad_norm": 0.632122814655304, "learning_rate": 3.798867096587739e-05, "loss": 0.0331, "step": 2190 }, { "epoch": 12.021857923497267, "grad_norm": 0.4645497798919678, "learning_rate": 3.755062869245441e-05, "loss": 0.0313, "step": 2200 }, { "epoch": 12.076502732240437, "grad_norm": 0.4805966019630432, "learning_rate": 3.711360274922876e-05, "loss": 0.0343, "step": 2210 }, { "epoch": 12.131147540983607, "grad_norm": 0.4689371585845947, "learning_rate": 3.667762881371804e-05, "loss": 0.0324, "step": 2220 }, { "epoch": 12.185792349726777, "grad_norm": 0.5546335577964783, "learning_rate": 3.6242742477556924e-05, "loss": 0.0316, "step": 2230 }, { "epoch": 12.240437158469945, "grad_norm": 0.5335983633995056, "learning_rate": 3.58089792435917e-05, "loss": 0.0322, "step": 2240 }, { "epoch": 12.295081967213115, "grad_norm": 0.9482993483543396, "learning_rate": 3.5376374522981883e-05, "loss": 0.031, "step": 2250 }, { "epoch": 12.349726775956285, "grad_norm": 0.41239190101623535, "learning_rate": 3.494496363230933e-05, "loss": 0.0323, "step": 2260 }, { "epoch": 12.404371584699454, "grad_norm": 0.3172128200531006, "learning_rate": 3.451478179069517e-05, "loss": 0.0247, "step": 2270 }, { "epoch": 12.459016393442623, "grad_norm": 0.5010955333709717, "learning_rate": 3.4085864116924516e-05, "loss": 0.0291, "step": 2280 }, { "epoch": 12.513661202185792, "grad_norm": 0.5820704102516174, "learning_rate": 3.365824562657956e-05, "loss": 0.0347, "step": 2290 }, { "epoch": 12.568306010928962, "grad_norm": 0.2892806828022003, "learning_rate": 3.323196122918091e-05, "loss": 0.0255, "step": 2300 }, { "epoch": 12.62295081967213, "grad_norm": 0.344103068113327, "learning_rate": 3.2807045725337815e-05, "loss": 0.0297, "step": 2310 }, { "epoch": 12.6775956284153, "grad_norm": 0.812961757183075, "learning_rate": 3.2383533803906986e-05, "loss": 0.0296, "step": 2320 }, { "epoch": 12.73224043715847, "grad_norm": 0.8150091767311096, "learning_rate": 3.196146003916084e-05, "loss": 0.029, "step": 2330 }, { "epoch": 12.78688524590164, "grad_norm": 0.38141801953315735, "learning_rate": 3.154085888796486e-05, "loss": 0.0276, "step": 2340 }, { "epoch": 12.841530054644808, "grad_norm": 0.5500876307487488, "learning_rate": 3.11217646869647e-05, "loss": 0.0272, "step": 2350 }, { "epoch": 12.896174863387978, "grad_norm": 0.534770667552948, "learning_rate": 3.0704211649782975e-05, "loss": 0.0262, "step": 2360 }, { "epoch": 12.950819672131148, "grad_norm": 0.8099629282951355, "learning_rate": 3.0288233864226235e-05, "loss": 0.0311, "step": 2370 }, { "epoch": 13.005464480874316, "grad_norm": 0.8436483144760132, "learning_rate": 2.9873865289502112e-05, "loss": 0.031, "step": 2380 }, { "epoch": 13.060109289617486, "grad_norm": 0.4314166307449341, "learning_rate": 2.946113975344694e-05, "loss": 0.0267, "step": 2390 }, { "epoch": 13.114754098360656, "grad_norm": 0.30424031615257263, "learning_rate": 2.9050090949764218e-05, "loss": 0.0289, "step": 2400 }, { "epoch": 13.169398907103826, "grad_norm": 0.3659568727016449, "learning_rate": 2.8640752435273942e-05, "loss": 0.0306, "step": 2410 }, { "epoch": 13.224043715846994, "grad_norm": 0.36199069023132324, "learning_rate": 2.8233157627173062e-05, "loss": 0.0248, "step": 2420 }, { "epoch": 13.278688524590164, "grad_norm": 0.29980045557022095, "learning_rate": 2.7827339800307466e-05, "loss": 0.0321, "step": 2430 }, { "epoch": 13.333333333333334, "grad_norm": 1.3584239482879639, "learning_rate": 2.7423332084455544e-05, "loss": 0.0263, "step": 2440 }, { "epoch": 13.387978142076502, "grad_norm": 0.372349351644516, "learning_rate": 2.7021167461623454e-05, "loss": 0.0259, "step": 2450 }, { "epoch": 13.442622950819672, "grad_norm": 0.32903772592544556, "learning_rate": 2.6620878763352714e-05, "loss": 0.0296, "step": 2460 }, { "epoch": 13.497267759562842, "grad_norm": 0.4593813419342041, "learning_rate": 2.6222498668039798e-05, "loss": 0.0254, "step": 2470 }, { "epoch": 13.551912568306012, "grad_norm": 0.2875046133995056, "learning_rate": 2.5826059698268433e-05, "loss": 0.0242, "step": 2480 }, { "epoch": 13.60655737704918, "grad_norm": 0.7870773673057556, "learning_rate": 2.5431594218154585e-05, "loss": 0.0228, "step": 2490 }, { "epoch": 13.66120218579235, "grad_norm": 0.6613791584968567, "learning_rate": 2.5039134430704292e-05, "loss": 0.0316, "step": 2500 }, { "epoch": 13.71584699453552, "grad_norm": 0.6127548813819885, "learning_rate": 2.4648712375184695e-05, "loss": 0.0249, "step": 2510 }, { "epoch": 13.770491803278688, "grad_norm": 0.5010942220687866, "learning_rate": 2.426035992450848e-05, "loss": 0.0248, "step": 2520 }, { "epoch": 13.825136612021858, "grad_norm": 0.3922024071216583, "learning_rate": 2.3874108782631847e-05, "loss": 0.0271, "step": 2530 }, { "epoch": 13.879781420765028, "grad_norm": 0.43783921003341675, "learning_rate": 2.348999048196634e-05, "loss": 0.0251, "step": 2540 }, { "epoch": 13.934426229508198, "grad_norm": 0.6914840936660767, "learning_rate": 2.310803638080458e-05, "loss": 0.0274, "step": 2550 }, { "epoch": 13.989071038251366, "grad_norm": 0.5838907957077026, "learning_rate": 2.2728277660760273e-05, "loss": 0.0268, "step": 2560 }, { "epoch": 14.043715846994536, "grad_norm": 0.2598058581352234, "learning_rate": 2.2350745324222632e-05, "loss": 0.0252, "step": 2570 }, { "epoch": 14.098360655737705, "grad_norm": 0.26290881633758545, "learning_rate": 2.197547019182545e-05, "loss": 0.0273, "step": 2580 }, { "epoch": 14.153005464480874, "grad_norm": 0.18337023258209229, "learning_rate": 2.1602482899930998e-05, "loss": 0.0238, "step": 2590 }, { "epoch": 14.207650273224044, "grad_norm": 0.7507574558258057, "learning_rate": 2.123181389812897e-05, "loss": 0.0286, "step": 2600 }, { "epoch": 14.262295081967213, "grad_norm": 0.4715905785560608, "learning_rate": 2.086349344675062e-05, "loss": 0.023, "step": 2610 }, { "epoch": 14.316939890710383, "grad_norm": 0.3056642413139343, "learning_rate": 2.0497551614398402e-05, "loss": 0.0276, "step": 2620 }, { "epoch": 14.371584699453551, "grad_norm": 0.39559224247932434, "learning_rate": 2.0134018275491357e-05, "loss": 0.0239, "step": 2630 }, { "epoch": 14.426229508196721, "grad_norm": 0.7475664019584656, "learning_rate": 1.9772923107826113e-05, "loss": 0.0285, "step": 2640 }, { "epoch": 14.480874316939891, "grad_norm": 0.3495936095714569, "learning_rate": 1.941429559015415e-05, "loss": 0.0257, "step": 2650 }, { "epoch": 14.53551912568306, "grad_norm": 0.40958720445632935, "learning_rate": 1.905816499977528e-05, "loss": 0.029, "step": 2660 }, { "epoch": 14.59016393442623, "grad_norm": 0.7965494394302368, "learning_rate": 1.8704560410147415e-05, "loss": 0.0224, "step": 2670 }, { "epoch": 14.6448087431694, "grad_norm": 0.30071818828582764, "learning_rate": 1.8353510688513235e-05, "loss": 0.0246, "step": 2680 }, { "epoch": 14.699453551912569, "grad_norm": 0.36322134733200073, "learning_rate": 1.8005044493543444e-05, "loss": 0.0236, "step": 2690 }, { "epoch": 14.754098360655737, "grad_norm": 0.3105396330356598, "learning_rate": 1.7659190272997185e-05, "loss": 0.0277, "step": 2700 }, { "epoch": 14.808743169398907, "grad_norm": 0.9227954149246216, "learning_rate": 1.7315976261399696e-05, "loss": 0.0243, "step": 2710 }, { "epoch": 14.863387978142077, "grad_norm": 0.254707008600235, "learning_rate": 1.6975430477737308e-05, "loss": 0.0225, "step": 2720 }, { "epoch": 14.918032786885245, "grad_norm": 0.30804020166397095, "learning_rate": 1.6637580723169988e-05, "loss": 0.0279, "step": 2730 }, { "epoch": 14.972677595628415, "grad_norm": 0.27004608511924744, "learning_rate": 1.630245457876181e-05, "loss": 0.0269, "step": 2740 }, { "epoch": 15.027322404371585, "grad_norm": 0.95454341173172, "learning_rate": 1.5970079403229286e-05, "loss": 0.0217, "step": 2750 }, { "epoch": 15.081967213114755, "grad_norm": 0.19432038068771362, "learning_rate": 1.564048233070792e-05, "loss": 0.0229, "step": 2760 }, { "epoch": 15.136612021857923, "grad_norm": 0.27142611145973206, "learning_rate": 1.5313690268537008e-05, "loss": 0.0243, "step": 2770 }, { "epoch": 15.191256830601093, "grad_norm": 0.2761251628398895, "learning_rate": 1.498972989506301e-05, "loss": 0.0282, "step": 2780 }, { "epoch": 15.245901639344263, "grad_norm": 0.25468575954437256, "learning_rate": 1.4668627657461632e-05, "loss": 0.0245, "step": 2790 }, { "epoch": 15.300546448087431, "grad_norm": 0.5863296389579773, "learning_rate": 1.4350409769578705e-05, "loss": 0.0237, "step": 2800 }, { "epoch": 15.3551912568306, "grad_norm": 0.598770022392273, "learning_rate": 1.4035102209790252e-05, "loss": 0.0236, "step": 2810 }, { "epoch": 15.40983606557377, "grad_norm": 1.1667654514312744, "learning_rate": 1.3722730718881633e-05, "loss": 0.0227, "step": 2820 }, { "epoch": 15.46448087431694, "grad_norm": 0.22816218435764313, "learning_rate": 1.3413320797946104e-05, "loss": 0.0201, "step": 2830 }, { "epoch": 15.519125683060109, "grad_norm": 0.1558607667684555, "learning_rate": 1.3106897706303051e-05, "loss": 0.0232, "step": 2840 }, { "epoch": 15.573770491803279, "grad_norm": 0.23202569782733917, "learning_rate": 1.2803486459435887e-05, "loss": 0.0209, "step": 2850 }, { "epoch": 15.628415300546449, "grad_norm": 0.23527105152606964, "learning_rate": 1.2503111826949809e-05, "loss": 0.0202, "step": 2860 }, { "epoch": 15.683060109289617, "grad_norm": 0.5382643938064575, "learning_rate": 1.2205798330549712e-05, "loss": 0.0265, "step": 2870 }, { "epoch": 15.737704918032787, "grad_norm": 0.5895487070083618, "learning_rate": 1.1911570242038356e-05, "loss": 0.0247, "step": 2880 }, { "epoch": 15.792349726775956, "grad_norm": 0.17227207124233246, "learning_rate": 1.1620451581334784e-05, "loss": 0.0247, "step": 2890 }, { "epoch": 15.846994535519126, "grad_norm": 0.30139684677124023, "learning_rate": 1.1332466114513512e-05, "loss": 0.0252, "step": 2900 }, { "epoch": 15.901639344262295, "grad_norm": 0.2650488018989563, "learning_rate": 1.1047637351864254e-05, "loss": 0.0231, "step": 2910 }, { "epoch": 15.956284153005464, "grad_norm": 0.8009262084960938, "learning_rate": 1.0765988545972639e-05, "loss": 0.0234, "step": 2920 }, { "epoch": 16.010928961748633, "grad_norm": 0.2774312198162079, "learning_rate": 1.0487542689821978e-05, "loss": 0.0244, "step": 2930 }, { "epoch": 16.065573770491802, "grad_norm": 0.3976193964481354, "learning_rate": 1.0212322514916139e-05, "loss": 0.0226, "step": 2940 }, { "epoch": 16.120218579234972, "grad_norm": 0.604283332824707, "learning_rate": 9.940350489423811e-06, "loss": 0.0258, "step": 2950 }, { "epoch": 16.174863387978142, "grad_norm": 0.7713361978530884, "learning_rate": 9.671648816344282e-06, "loss": 0.0227, "step": 2960 }, { "epoch": 16.229508196721312, "grad_norm": 0.4896545112133026, "learning_rate": 9.406239431694841e-06, "loss": 0.0209, "step": 2970 }, { "epoch": 16.284153005464482, "grad_norm": 0.3895213305950165, "learning_rate": 9.14414400272003e-06, "loss": 0.0218, "step": 2980 }, { "epoch": 16.338797814207652, "grad_norm": 0.37582123279571533, "learning_rate": 8.885383926122748e-06, "loss": 0.0214, "step": 2990 }, { "epoch": 16.39344262295082, "grad_norm": 0.7232876420021057, "learning_rate": 8.629980326317489e-06, "loss": 0.0214, "step": 3000 }, { "epoch": 16.44808743169399, "grad_norm": 0.3176191747188568, "learning_rate": 8.377954053705805e-06, "loss": 0.0256, "step": 3010 }, { "epoch": 16.502732240437158, "grad_norm": 0.3513713479042053, "learning_rate": 8.129325682974165e-06, "loss": 0.0268, "step": 3020 }, { "epoch": 16.557377049180328, "grad_norm": 0.32964545488357544, "learning_rate": 7.884115511414286e-06, "loss": 0.0219, "step": 3030 }, { "epoch": 16.612021857923498, "grad_norm": 0.2857699394226074, "learning_rate": 7.642343557266134e-06, "loss": 0.024, "step": 3040 }, { "epoch": 16.666666666666668, "grad_norm": 0.5647152662277222, "learning_rate": 7.404029558083653e-06, "loss": 0.0295, "step": 3050 }, { "epoch": 16.721311475409838, "grad_norm": 0.3039594888687134, "learning_rate": 7.169192969123467e-06, "loss": 0.0226, "step": 3060 }, { "epoch": 16.775956284153004, "grad_norm": 0.4706748127937317, "learning_rate": 6.937852961756669e-06, "loss": 0.0248, "step": 3070 }, { "epoch": 16.830601092896174, "grad_norm": 0.35213521122932434, "learning_rate": 6.710028421903625e-06, "loss": 0.0237, "step": 3080 }, { "epoch": 16.885245901639344, "grad_norm": 0.3645097017288208, "learning_rate": 6.4857379484922375e-06, "loss": 0.0209, "step": 3090 }, { "epoch": 16.939890710382514, "grad_norm": 0.5215752720832825, "learning_rate": 6.2649998519396235e-06, "loss": 0.0207, "step": 3100 }, { "epoch": 16.994535519125684, "grad_norm": 0.7735849618911743, "learning_rate": 6.047832152657207e-06, "loss": 0.0221, "step": 3110 }, { "epoch": 17.049180327868854, "grad_norm": 0.3790043294429779, "learning_rate": 5.834252579579685e-06, "loss": 0.0242, "step": 3120 }, { "epoch": 17.103825136612024, "grad_norm": 0.29151690006256104, "learning_rate": 5.624278568717611e-06, "loss": 0.0207, "step": 3130 }, { "epoch": 17.15846994535519, "grad_norm": 0.4057449996471405, "learning_rate": 5.417927261734007e-06, "loss": 0.0221, "step": 3140 }, { "epoch": 17.21311475409836, "grad_norm": 0.3481403887271881, "learning_rate": 5.215215504544995e-06, "loss": 0.0202, "step": 3150 }, { "epoch": 17.26775956284153, "grad_norm": 0.23598790168762207, "learning_rate": 5.016159845944473e-06, "loss": 0.0298, "step": 3160 }, { "epoch": 17.3224043715847, "grad_norm": 0.3719569146633148, "learning_rate": 4.820776536253202e-06, "loss": 0.0195, "step": 3170 }, { "epoch": 17.37704918032787, "grad_norm": 0.3433237373828888, "learning_rate": 4.629081525992118e-06, "loss": 0.0227, "step": 3180 }, { "epoch": 17.43169398907104, "grad_norm": 0.3161409795284271, "learning_rate": 4.441090464580178e-06, "loss": 0.023, "step": 3190 }, { "epoch": 17.48633879781421, "grad_norm": 0.8182094097137451, "learning_rate": 4.256818699056847e-06, "loss": 0.0235, "step": 3200 }, { "epoch": 17.540983606557376, "grad_norm": 0.4030332863330841, "learning_rate": 4.076281272829158e-06, "loss": 0.023, "step": 3210 }, { "epoch": 17.595628415300546, "grad_norm": 0.252172589302063, "learning_rate": 3.89949292444361e-06, "loss": 0.0221, "step": 3220 }, { "epoch": 17.650273224043715, "grad_norm": 1.1839289665222168, "learning_rate": 3.726468086382989e-06, "loss": 0.0222, "step": 3230 }, { "epoch": 17.704918032786885, "grad_norm": 0.3704149127006531, "learning_rate": 3.5572208838881073e-06, "loss": 0.02, "step": 3240 }, { "epoch": 17.759562841530055, "grad_norm": 0.25189173221588135, "learning_rate": 3.391765133804692e-06, "loss": 0.0294, "step": 3250 }, { "epoch": 17.814207650273225, "grad_norm": 0.31096377968788147, "learning_rate": 3.2301143434554147e-06, "loss": 0.0216, "step": 3260 }, { "epoch": 17.868852459016395, "grad_norm": 0.37782198190689087, "learning_rate": 3.072281709537156e-06, "loss": 0.0223, "step": 3270 }, { "epoch": 17.92349726775956, "grad_norm": 0.3171507716178894, "learning_rate": 2.918280117043709e-06, "loss": 0.0253, "step": 3280 }, { "epoch": 17.97814207650273, "grad_norm": 0.46874311566352844, "learning_rate": 2.768122138213858e-06, "loss": 0.0215, "step": 3290 }, { "epoch": 18.0327868852459, "grad_norm": 0.7982349395751953, "learning_rate": 2.621820031505051e-06, "loss": 0.0211, "step": 3300 }, { "epoch": 18.08743169398907, "grad_norm": 0.6364455819129944, "learning_rate": 2.4793857405926044e-06, "loss": 0.0191, "step": 3310 }, { "epoch": 18.14207650273224, "grad_norm": 0.3849509060382843, "learning_rate": 2.3408308933947244e-06, "loss": 0.0212, "step": 3320 }, { "epoch": 18.19672131147541, "grad_norm": 0.6707078814506531, "learning_rate": 2.2061668011231496e-06, "loss": 0.0244, "step": 3330 }, { "epoch": 18.25136612021858, "grad_norm": 0.377101331949234, "learning_rate": 2.0754044573598306e-06, "loss": 0.0197, "step": 3340 }, { "epoch": 18.306010928961747, "grad_norm": 0.531783938407898, "learning_rate": 1.9485545371593807e-06, "loss": 0.0227, "step": 3350 }, { "epoch": 18.360655737704917, "grad_norm": 0.5045121312141418, "learning_rate": 1.8256273961776093e-06, "loss": 0.0202, "step": 3360 }, { "epoch": 18.415300546448087, "grad_norm": 0.6575759053230286, "learning_rate": 1.7066330698261467e-06, "loss": 0.0236, "step": 3370 }, { "epoch": 18.469945355191257, "grad_norm": 0.454512357711792, "learning_rate": 1.591581272453141e-06, "loss": 0.0176, "step": 3380 }, { "epoch": 18.524590163934427, "grad_norm": 0.2828448712825775, "learning_rate": 1.4804813965502374e-06, "loss": 0.0219, "step": 3390 }, { "epoch": 18.579234972677597, "grad_norm": 0.38783037662506104, "learning_rate": 1.3733425119857812e-06, "loss": 0.0281, "step": 3400 }, { "epoch": 18.633879781420767, "grad_norm": 0.42596179246902466, "learning_rate": 1.2701733652643933e-06, "loss": 0.0212, "step": 3410 }, { "epoch": 18.688524590163933, "grad_norm": 0.2547903060913086, "learning_rate": 1.1709823788129292e-06, "loss": 0.0233, "step": 3420 }, { "epoch": 18.743169398907103, "grad_norm": 0.463740736246109, "learning_rate": 1.0757776502929084e-06, "loss": 0.0217, "step": 3430 }, { "epoch": 18.797814207650273, "grad_norm": 0.9804412722587585, "learning_rate": 9.84566951939414e-07, "loss": 0.0242, "step": 3440 }, { "epoch": 18.852459016393443, "grad_norm": 0.23706400394439697, "learning_rate": 8.973577299266178e-07, "loss": 0.0228, "step": 3450 }, { "epoch": 18.907103825136613, "grad_norm": 0.2117275893688202, "learning_rate": 8.141571037598828e-07, "loss": 0.0212, "step": 3460 }, { "epoch": 18.961748633879782, "grad_norm": 0.7138676047325134, "learning_rate": 7.349718656945504e-07, "loss": 0.0217, "step": 3470 }, { "epoch": 19.016393442622952, "grad_norm": 0.38242462277412415, "learning_rate": 6.598084801814563e-07, "loss": 0.0234, "step": 3480 }, { "epoch": 19.07103825136612, "grad_norm": 0.30162400007247925, "learning_rate": 5.88673083339164e-07, "loss": 0.0226, "step": 3490 }, { "epoch": 19.12568306010929, "grad_norm": 0.3553517460823059, "learning_rate": 5.215714824530427e-07, "loss": 0.0234, "step": 3500 }, { "epoch": 19.18032786885246, "grad_norm": 0.2980111539363861, "learning_rate": 4.585091555011856e-07, "loss": 0.0223, "step": 3510 }, { "epoch": 19.23497267759563, "grad_norm": 0.3606838881969452, "learning_rate": 3.994912507072013e-07, "loss": 0.0193, "step": 3520 }, { "epoch": 19.2896174863388, "grad_norm": 0.2700374722480774, "learning_rate": 3.4452258611991083e-07, "loss": 0.0209, "step": 3530 }, { "epoch": 19.34426229508197, "grad_norm": 0.44229865074157715, "learning_rate": 2.9360764922004014e-07, "loss": 0.0194, "step": 3540 }, { "epoch": 19.398907103825138, "grad_norm": 0.3039417266845703, "learning_rate": 2.467505965538519e-07, "loss": 0.0202, "step": 3550 }, { "epoch": 19.453551912568305, "grad_norm": 0.4226384460926056, "learning_rate": 2.0395525339383937e-07, "loss": 0.0221, "step": 3560 }, { "epoch": 19.508196721311474, "grad_norm": 0.7776270508766174, "learning_rate": 1.6522511342643155e-07, "loss": 0.0203, "step": 3570 }, { "epoch": 19.562841530054644, "grad_norm": 0.23865941166877747, "learning_rate": 1.3056333846677705e-07, "loss": 0.0195, "step": 3580 }, { "epoch": 19.617486338797814, "grad_norm": 0.5467799305915833, "learning_rate": 9.997275820062268e-08, "loss": 0.0193, "step": 3590 }, { "epoch": 19.672131147540984, "grad_norm": 0.24738432466983795, "learning_rate": 7.3455869953315e-08, "loss": 0.0209, "step": 3600 }, { "epoch": 19.726775956284154, "grad_norm": 1.1878104209899902, "learning_rate": 5.101483848591881e-08, "loss": 0.0205, "step": 3610 }, { "epoch": 19.781420765027324, "grad_norm": 0.37522733211517334, "learning_rate": 3.2651495818486476e-08, "loss": 0.0213, "step": 3620 }, { "epoch": 19.83606557377049, "grad_norm": 0.3089437782764435, "learning_rate": 1.8367341080510747e-08, "loss": 0.0181, "step": 3630 }, { "epoch": 19.89071038251366, "grad_norm": 0.45112279057502747, "learning_rate": 8.163540388539393e-09, "loss": 0.0207, "step": 3640 }, { "epoch": 19.94535519125683, "grad_norm": 0.6257203817367554, "learning_rate": 2.0409267509569062e-09, "loss": 0.0234, "step": 3650 }, { "epoch": 20.0, "grad_norm": 0.20247182250022888, "learning_rate": 0.0, "loss": 0.0241, "step": 3660 }, { "epoch": 20.0, "step": 3660, "total_flos": 3.885044834331936e+17, "train_loss": 0.05339513023287221, "train_runtime": 3949.2706, "train_samples_per_second": 45.401, "train_steps_per_second": 0.927 } ], "logging_steps": 10, "max_steps": 3660, "num_input_tokens_seen": 0, "num_train_epochs": 20, "save_steps": 10000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.885044834331936e+17, "train_batch_size": 49, "trial_name": null, "trial_params": null }