| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.4449388209121246, |
| "eval_steps": 500, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.8774386405944824, |
| "epoch": 0.04449388209121246, |
| "grad_norm": 0.12704895436763763, |
| "learning_rate": 0.0002, |
| "loss": 1.2824, |
| "mean_token_accuracy": 0.6995065063238144, |
| "num_tokens": 327680.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.990321047604084, |
| "epoch": 0.08898776418242492, |
| "grad_norm": 0.06498447060585022, |
| "learning_rate": 0.0002, |
| "loss": 0.9436, |
| "mean_token_accuracy": 0.7361853331327438, |
| "num_tokens": 655360.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.8516890451312065, |
| "epoch": 0.13348164627363737, |
| "grad_norm": 0.08354150503873825, |
| "learning_rate": 0.0002, |
| "loss": 0.8984, |
| "mean_token_accuracy": 0.7455797985196113, |
| "num_tokens": 983040.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8647283613681793, |
| "epoch": 0.17797552836484984, |
| "grad_norm": 0.06878677755594254, |
| "learning_rate": 0.0002, |
| "loss": 0.8596, |
| "mean_token_accuracy": 0.7515453889966011, |
| "num_tokens": 1310620.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.8749317795038223, |
| "epoch": 0.22246941045606228, |
| "grad_norm": 0.12942193448543549, |
| "learning_rate": 0.0002, |
| "loss": 0.8933, |
| "mean_token_accuracy": 0.7529007002711297, |
| "num_tokens": 1623688.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8274541616439819, |
| "epoch": 0.26696329254727474, |
| "grad_norm": 0.0724002867937088, |
| "learning_rate": 0.0002, |
| "loss": 0.8346, |
| "mean_token_accuracy": 0.7569744408130645, |
| "num_tokens": 1951368.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8037244468927384, |
| "epoch": 0.3114571746384872, |
| "grad_norm": 0.06898455321788788, |
| "learning_rate": 0.0002, |
| "loss": 0.7998, |
| "mean_token_accuracy": 0.765008519589901, |
| "num_tokens": 2279048.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.7850529387593269, |
| "epoch": 0.3559510567296997, |
| "grad_norm": 0.06743492931127548, |
| "learning_rate": 0.0002, |
| "loss": 0.7893, |
| "mean_token_accuracy": 0.7665843173861504, |
| "num_tokens": 2606728.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.7824164897203445, |
| "epoch": 0.40044493882091214, |
| "grad_norm": 0.0739995464682579, |
| "learning_rate": 0.0002, |
| "loss": 0.7853, |
| "mean_token_accuracy": 0.7667212918400764, |
| "num_tokens": 2934287.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8248968973755837, |
| "epoch": 0.44493882091212456, |
| "grad_norm": 0.07532794028520584, |
| "learning_rate": 0.0002, |
| "loss": 0.8328, |
| "mean_token_accuracy": 0.7640984863042831, |
| "num_tokens": 3253000.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.7598924160003662, |
| "epoch": 0.489432703003337, |
| "grad_norm": 0.07987581938505173, |
| "learning_rate": 0.0002, |
| "loss": 0.7672, |
| "mean_token_accuracy": 0.7705939218401909, |
| "num_tokens": 3580680.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.7683228522539138, |
| "epoch": 0.5339265850945495, |
| "grad_norm": 0.0910281166434288, |
| "learning_rate": 0.0002, |
| "loss": 0.7702, |
| "mean_token_accuracy": 0.7697535201907157, |
| "num_tokens": 3908360.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7364083752036095, |
| "epoch": 0.578420467185762, |
| "grad_norm": 0.0787215381860733, |
| "learning_rate": 0.0002, |
| "loss": 0.7448, |
| "mean_token_accuracy": 0.775777031481266, |
| "num_tokens": 4236040.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7472490265965461, |
| "epoch": 0.6229143492769744, |
| "grad_norm": 0.08141525834798813, |
| "learning_rate": 0.0002, |
| "loss": 0.7503, |
| "mean_token_accuracy": 0.7719789668917656, |
| "num_tokens": 4563587.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.7923195600509644, |
| "epoch": 0.6674082313681868, |
| "grad_norm": 0.1027020588517189, |
| "learning_rate": 0.0002, |
| "loss": 0.7961, |
| "mean_token_accuracy": 0.7704089671373368, |
| "num_tokens": 4884278.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7354328155517578, |
| "epoch": 0.7119021134593994, |
| "grad_norm": 0.08376702666282654, |
| "learning_rate": 0.0002, |
| "loss": 0.7421, |
| "mean_token_accuracy": 0.7752418130636215, |
| "num_tokens": 5211958.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.7320004045963288, |
| "epoch": 0.7563959955506118, |
| "grad_norm": 0.07954639941453934, |
| "learning_rate": 0.0002, |
| "loss": 0.7347, |
| "mean_token_accuracy": 0.7771021112799644, |
| "num_tokens": 5539638.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.726919624209404, |
| "epoch": 0.8008898776418243, |
| "grad_norm": 0.08581581711769104, |
| "learning_rate": 0.0002, |
| "loss": 0.7283, |
| "mean_token_accuracy": 0.7800618261098862, |
| "num_tokens": 5867318.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7299349635839463, |
| "epoch": 0.8453837597330367, |
| "grad_norm": 0.08491721004247665, |
| "learning_rate": 0.0002, |
| "loss": 0.7291, |
| "mean_token_accuracy": 0.778104068338871, |
| "num_tokens": 6194899.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7759820625185967, |
| "epoch": 0.8898776418242491, |
| "grad_norm": 0.08337808400392532, |
| "learning_rate": 0.0002, |
| "loss": 0.7852, |
| "mean_token_accuracy": 0.772493363916874, |
| "num_tokens": 6517162.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7255487650632858, |
| "epoch": 0.9343715239154616, |
| "grad_norm": 0.08438117802143097, |
| "learning_rate": 0.0002, |
| "loss": 0.7265, |
| "mean_token_accuracy": 0.7796106413006783, |
| "num_tokens": 6844842.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.719422459602356, |
| "epoch": 0.978865406006674, |
| "grad_norm": 0.08568471670150757, |
| "learning_rate": 0.0002, |
| "loss": 0.7188, |
| "mean_token_accuracy": 0.7807851985096932, |
| "num_tokens": 7172484.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7186581477140769, |
| "epoch": 1.0222469410456063, |
| "grad_norm": 0.08594680577516556, |
| "learning_rate": 0.0002, |
| "loss": 0.7087, |
| "mean_token_accuracy": 0.7901185231331067, |
| "num_tokens": 7484161.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.6627673536539078, |
| "epoch": 1.0667408231368187, |
| "grad_norm": 0.0888429582118988, |
| "learning_rate": 0.0002, |
| "loss": 0.6608, |
| "mean_token_accuracy": 0.796707707643509, |
| "num_tokens": 7811841.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.6725413352251053, |
| "epoch": 1.1112347052280311, |
| "grad_norm": 0.09360355883836746, |
| "learning_rate": 0.0002, |
| "loss": 0.6687, |
| "mean_token_accuracy": 0.7943419247865677, |
| "num_tokens": 8139521.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.6608762927353382, |
| "epoch": 1.1557285873192435, |
| "grad_norm": 0.09271729737520218, |
| "learning_rate": 0.0002, |
| "loss": 0.6528, |
| "mean_token_accuracy": 0.7991992741823196, |
| "num_tokens": 8467165.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.6472577676177025, |
| "epoch": 1.200222469410456, |
| "grad_norm": 0.0905335545539856, |
| "learning_rate": 0.0002, |
| "loss": 0.6395, |
| "mean_token_accuracy": 0.8023400634527207, |
| "num_tokens": 8794479.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7025618955492974, |
| "epoch": 1.2447163515016686, |
| "grad_norm": 0.09981676936149597, |
| "learning_rate": 0.0002, |
| "loss": 0.6969, |
| "mean_token_accuracy": 0.7927162751555443, |
| "num_tokens": 9115502.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.6550201892852783, |
| "epoch": 1.289210233592881, |
| "grad_norm": 0.09637434035539627, |
| "learning_rate": 0.0002, |
| "loss": 0.6472, |
| "mean_token_accuracy": 0.7994350507855416, |
| "num_tokens": 9443182.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.6535603314638138, |
| "epoch": 1.3337041156840934, |
| "grad_norm": 0.09058881551027298, |
| "learning_rate": 0.0002, |
| "loss": 0.6505, |
| "mean_token_accuracy": 0.8004654511809349, |
| "num_tokens": 9770862.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.6676291272044181, |
| "epoch": 1.3781979977753058, |
| "grad_norm": 0.10104376822710037, |
| "learning_rate": 0.0002, |
| "loss": 0.6605, |
| "mean_token_accuracy": 0.7970087364315986, |
| "num_tokens": 10098521.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.6632239744067192, |
| "epoch": 1.4226918798665182, |
| "grad_norm": 0.09252401441335678, |
| "learning_rate": 0.0002, |
| "loss": 0.6567, |
| "mean_token_accuracy": 0.7999079167842865, |
| "num_tokens": 10425894.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.6941266790032387, |
| "epoch": 1.4671857619577309, |
| "grad_norm": 0.10002182424068451, |
| "learning_rate": 0.0002, |
| "loss": 0.6904, |
| "mean_token_accuracy": 0.7945976346731186, |
| "num_tokens": 10746167.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.6502977207303047, |
| "epoch": 1.5116796440489433, |
| "grad_norm": 0.10251280665397644, |
| "learning_rate": 0.0002, |
| "loss": 0.6447, |
| "mean_token_accuracy": 0.8009829133749008, |
| "num_tokens": 11073847.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.6449682429432869, |
| "epoch": 1.5561735261401557, |
| "grad_norm": 0.10051094740629196, |
| "learning_rate": 0.0002, |
| "loss": 0.6297, |
| "mean_token_accuracy": 0.8034362852573395, |
| "num_tokens": 11401527.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.647579725086689, |
| "epoch": 1.6006674082313683, |
| "grad_norm": 0.09883299469947815, |
| "learning_rate": 0.0002, |
| "loss": 0.6439, |
| "mean_token_accuracy": 0.8008567914366722, |
| "num_tokens": 11729207.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.6341586455702781, |
| "epoch": 1.6451612903225805, |
| "grad_norm": 0.0934380292892456, |
| "learning_rate": 0.0002, |
| "loss": 0.6273, |
| "mean_token_accuracy": 0.805011647939682, |
| "num_tokens": 12056607.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.7022602126002312, |
| "epoch": 1.6896551724137931, |
| "grad_norm": 0.10245587676763535, |
| "learning_rate": 0.0002, |
| "loss": 0.6997, |
| "mean_token_accuracy": 0.7951286539435387, |
| "num_tokens": 12375719.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.6635140493512154, |
| "epoch": 1.7341490545050056, |
| "grad_norm": 0.10037633031606674, |
| "learning_rate": 0.0002, |
| "loss": 0.6532, |
| "mean_token_accuracy": 0.7992691040039063, |
| "num_tokens": 12703399.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.6396854467689991, |
| "epoch": 1.778642936596218, |
| "grad_norm": 0.09362801909446716, |
| "learning_rate": 0.0002, |
| "loss": 0.6319, |
| "mean_token_accuracy": 0.8054976046085358, |
| "num_tokens": 13031079.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.645181667804718, |
| "epoch": 1.8231368186874306, |
| "grad_norm": 0.09476775676012039, |
| "learning_rate": 0.0002, |
| "loss": 0.6366, |
| "mean_token_accuracy": 0.8037867248058319, |
| "num_tokens": 13358733.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.652097150683403, |
| "epoch": 1.8676307007786428, |
| "grad_norm": 0.09551919251680374, |
| "learning_rate": 0.0002, |
| "loss": 0.6399, |
| "mean_token_accuracy": 0.8037777736783027, |
| "num_tokens": 13685998.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.6837869822978974, |
| "epoch": 1.9121245828698554, |
| "grad_norm": 0.10013949871063232, |
| "learning_rate": 0.0002, |
| "loss": 0.6841, |
| "mean_token_accuracy": 0.7974902361631393, |
| "num_tokens": 14005579.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.628454002737999, |
| "epoch": 1.9566184649610678, |
| "grad_norm": 0.13781693577766418, |
| "learning_rate": 0.0002, |
| "loss": 0.6213, |
| "mean_token_accuracy": 0.8075341194868088, |
| "num_tokens": 14333259.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.6827295254438351, |
| "epoch": 2.0, |
| "grad_norm": 0.15781015157699585, |
| "learning_rate": 0.0002, |
| "loss": 0.68, |
| "mean_token_accuracy": 0.7996979187696408, |
| "num_tokens": 14640642.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.5626548573374748, |
| "epoch": 2.0444938820912126, |
| "grad_norm": 0.10359702259302139, |
| "learning_rate": 0.0002, |
| "loss": 0.5339, |
| "mean_token_accuracy": 0.8331230849027633, |
| "num_tokens": 14968322.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.5715467736124993, |
| "epoch": 2.088987764182425, |
| "grad_norm": 0.10643124580383301, |
| "learning_rate": 0.0002, |
| "loss": 0.5351, |
| "mean_token_accuracy": 0.8342131584882736, |
| "num_tokens": 15296002.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.5666955970227718, |
| "epoch": 2.1334816462736375, |
| "grad_norm": 0.10716599971055984, |
| "learning_rate": 0.0002, |
| "loss": 0.5381, |
| "mean_token_accuracy": 0.8319605439901352, |
| "num_tokens": 15623682.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.5521194256842137, |
| "epoch": 2.1779755283648496, |
| "grad_norm": 0.11130277067422867, |
| "learning_rate": 0.0002, |
| "loss": 0.5209, |
| "mean_token_accuracy": 0.8368723064661026, |
| "num_tokens": 15951273.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.6372111544013024, |
| "epoch": 2.2224694104560623, |
| "grad_norm": 0.10940540581941605, |
| "learning_rate": 0.0002, |
| "loss": 0.6048, |
| "mean_token_accuracy": 0.8223052099347115, |
| "num_tokens": 16269048.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.5748868621885777, |
| "epoch": 2.266963292547275, |
| "grad_norm": 0.1133793443441391, |
| "learning_rate": 0.0002, |
| "loss": 0.5473, |
| "mean_token_accuracy": 0.8281539842486382, |
| "num_tokens": 16596728.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.5760611467063427, |
| "epoch": 2.311457174638487, |
| "grad_norm": 0.11276053637266159, |
| "learning_rate": 0.0002, |
| "loss": 0.5512, |
| "mean_token_accuracy": 0.8286992073059082, |
| "num_tokens": 16924408.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.5595883250236511, |
| "epoch": 2.3559510567296997, |
| "grad_norm": 0.10785238444805145, |
| "learning_rate": 0.0002, |
| "loss": 0.534, |
| "mean_token_accuracy": 0.8329534187912941, |
| "num_tokens": 17252088.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.5623729139566421, |
| "epoch": 2.400444938820912, |
| "grad_norm": 0.11915665119886398, |
| "learning_rate": 0.0002, |
| "loss": 0.5318, |
| "mean_token_accuracy": 0.833781997859478, |
| "num_tokens": 17579671.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.6471138700842858, |
| "epoch": 2.4449388209121246, |
| "grad_norm": 0.09796369075775146, |
| "learning_rate": 0.0002, |
| "loss": 0.62, |
| "mean_token_accuracy": 0.8189012482762337, |
| "num_tokens": 17899138.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.5706302203238011, |
| "epoch": 2.489432703003337, |
| "grad_norm": 0.11501935124397278, |
| "learning_rate": 0.0002, |
| "loss": 0.5443, |
| "mean_token_accuracy": 0.8297176495194435, |
| "num_tokens": 18226818.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.5607033468782902, |
| "epoch": 2.5339265850945494, |
| "grad_norm": 0.10209450125694275, |
| "learning_rate": 0.0002, |
| "loss": 0.537, |
| "mean_token_accuracy": 0.8327666461467743, |
| "num_tokens": 18554498.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.5697788536548615, |
| "epoch": 2.578420467185762, |
| "grad_norm": 0.11915192008018494, |
| "learning_rate": 0.0002, |
| "loss": 0.5417, |
| "mean_token_accuracy": 0.8292823225259781, |
| "num_tokens": 18882178.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.5598745912313461, |
| "epoch": 2.622914349276974, |
| "grad_norm": 0.12072969228029251, |
| "learning_rate": 0.0002, |
| "loss": 0.5318, |
| "mean_token_accuracy": 0.8316263392567634, |
| "num_tokens": 19209735.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6062877185642719, |
| "epoch": 2.667408231368187, |
| "grad_norm": 0.11766766756772995, |
| "learning_rate": 0.0002, |
| "loss": 0.5785, |
| "mean_token_accuracy": 0.8279747024178505, |
| "num_tokens": 19531022.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.5801788948476314, |
| "epoch": 2.7119021134593995, |
| "grad_norm": 0.11248841881752014, |
| "learning_rate": 0.0002, |
| "loss": 0.5555, |
| "mean_token_accuracy": 0.8277870997786522, |
| "num_tokens": 19858702.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.5594978570938111, |
| "epoch": 2.7563959955506117, |
| "grad_norm": 0.11471142619848251, |
| "learning_rate": 0.0002, |
| "loss": 0.5343, |
| "mean_token_accuracy": 0.8331717640161515, |
| "num_tokens": 20186382.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.5783086307346821, |
| "epoch": 2.8008898776418243, |
| "grad_norm": 0.12316859513521194, |
| "learning_rate": 0.0002, |
| "loss": 0.5514, |
| "mean_token_accuracy": 0.8270430579781533, |
| "num_tokens": 20514062.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.5552236251533031, |
| "epoch": 2.8453837597330365, |
| "grad_norm": 0.1180945634841919, |
| "learning_rate": 0.0002, |
| "loss": 0.5254, |
| "mean_token_accuracy": 0.833421728014946, |
| "num_tokens": 20841607.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6508079275488854, |
| "epoch": 2.889877641824249, |
| "grad_norm": 0.11594584584236145, |
| "learning_rate": 0.0002, |
| "loss": 0.6277, |
| "mean_token_accuracy": 0.8178864538669586, |
| "num_tokens": 21156705.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.5584865964949131, |
| "epoch": 2.9343715239154617, |
| "grad_norm": 0.10876645147800446, |
| "learning_rate": 0.0002, |
| "loss": 0.5341, |
| "mean_token_accuracy": 0.8327999398112297, |
| "num_tokens": 21484385.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.5605839185416699, |
| "epoch": 2.978865406006674, |
| "grad_norm": 0.12009165436029434, |
| "learning_rate": 0.0002, |
| "loss": 0.5317, |
| "mean_token_accuracy": 0.8333747163414955, |
| "num_tokens": 21812022.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.5539867511162391, |
| "epoch": 3.0222469410456063, |
| "grad_norm": 0.12213966995477676, |
| "learning_rate": 0.0002, |
| "loss": 0.4962, |
| "mean_token_accuracy": 0.8499505198918856, |
| "num_tokens": 22124803.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.48398632630705835, |
| "epoch": 3.0667408231368185, |
| "grad_norm": 0.1170133501291275, |
| "learning_rate": 0.0002, |
| "loss": 0.4345, |
| "mean_token_accuracy": 0.8670241564512253, |
| "num_tokens": 22452483.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.4720599815249443, |
| "epoch": 3.111234705228031, |
| "grad_norm": 0.12524375319480896, |
| "learning_rate": 0.0002, |
| "loss": 0.4173, |
| "mean_token_accuracy": 0.8702847585082054, |
| "num_tokens": 22780163.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.4982055649161339, |
| "epoch": 3.1557285873192438, |
| "grad_norm": 0.11435994505882263, |
| "learning_rate": 0.0002, |
| "loss": 0.4423, |
| "mean_token_accuracy": 0.864942966401577, |
| "num_tokens": 23107835.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.48039583414793013, |
| "epoch": 3.200222469410456, |
| "grad_norm": 0.12139066308736801, |
| "learning_rate": 0.0002, |
| "loss": 0.4226, |
| "mean_token_accuracy": 0.8702625542879104, |
| "num_tokens": 23435179.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.5418343998491764, |
| "epoch": 3.2447163515016686, |
| "grad_norm": 0.13233190774917603, |
| "learning_rate": 0.0002, |
| "loss": 0.4924, |
| "mean_token_accuracy": 0.8546271666884422, |
| "num_tokens": 23753029.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.4849280208349228, |
| "epoch": 3.289210233592881, |
| "grad_norm": 0.12592358887195587, |
| "learning_rate": 0.0002, |
| "loss": 0.4278, |
| "mean_token_accuracy": 0.8678036436438561, |
| "num_tokens": 24080709.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.5037459127604962, |
| "epoch": 3.3337041156840934, |
| "grad_norm": 0.12919606268405914, |
| "learning_rate": 0.0002, |
| "loss": 0.4524, |
| "mean_token_accuracy": 0.8612239480018615, |
| "num_tokens": 24408389.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.4822153463959694, |
| "epoch": 3.378197997775306, |
| "grad_norm": 0.12449780106544495, |
| "learning_rate": 0.0002, |
| "loss": 0.4291, |
| "mean_token_accuracy": 0.8676740288734436, |
| "num_tokens": 24736036.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.4992881424725056, |
| "epoch": 3.4226918798665182, |
| "grad_norm": 0.1114158034324646, |
| "learning_rate": 0.0002, |
| "loss": 0.4433, |
| "mean_token_accuracy": 0.8635369181632996, |
| "num_tokens": 25063393.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.5634101495146752, |
| "epoch": 3.467185761957731, |
| "grad_norm": 0.13151301443576813, |
| "learning_rate": 0.0002, |
| "loss": 0.5292, |
| "mean_token_accuracy": 0.848271718621254, |
| "num_tokens": 25383789.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.5000352062284946, |
| "epoch": 3.511679644048943, |
| "grad_norm": 0.1317589432001114, |
| "learning_rate": 0.0002, |
| "loss": 0.4488, |
| "mean_token_accuracy": 0.8623447984457016, |
| "num_tokens": 25711469.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.5025811523199082, |
| "epoch": 3.5561735261401557, |
| "grad_norm": 0.12309940159320831, |
| "learning_rate": 0.0002, |
| "loss": 0.4515, |
| "mean_token_accuracy": 0.8600758820772171, |
| "num_tokens": 26039149.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.4819103963673115, |
| "epoch": 3.6006674082313683, |
| "grad_norm": 0.12250859290361404, |
| "learning_rate": 0.0002, |
| "loss": 0.4342, |
| "mean_token_accuracy": 0.8647636532783508, |
| "num_tokens": 26366815.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.4777693450450897, |
| "epoch": 3.6451612903225805, |
| "grad_norm": 0.11709456145763397, |
| "learning_rate": 0.0002, |
| "loss": 0.4249, |
| "mean_token_accuracy": 0.8680423229932785, |
| "num_tokens": 26694191.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.5633887559175491, |
| "epoch": 3.689655172413793, |
| "grad_norm": 0.12784965336322784, |
| "learning_rate": 0.0002, |
| "loss": 0.5239, |
| "mean_token_accuracy": 0.8493879333138465, |
| "num_tokens": 27013314.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.48369882851839063, |
| "epoch": 3.7341490545050053, |
| "grad_norm": 0.12846286594867706, |
| "learning_rate": 0.0002, |
| "loss": 0.4369, |
| "mean_token_accuracy": 0.8636931553483009, |
| "num_tokens": 27340994.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.5082808144390583, |
| "epoch": 3.778642936596218, |
| "grad_norm": 0.12857837975025177, |
| "learning_rate": 0.0002, |
| "loss": 0.4617, |
| "mean_token_accuracy": 0.8581709936261177, |
| "num_tokens": 27668674.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.47558383345603944, |
| "epoch": 3.8231368186874306, |
| "grad_norm": 0.12611748278141022, |
| "learning_rate": 0.0002, |
| "loss": 0.4235, |
| "mean_token_accuracy": 0.866823847591877, |
| "num_tokens": 27996329.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.4854692369699478, |
| "epoch": 3.867630700778643, |
| "grad_norm": 0.12649761140346527, |
| "learning_rate": 0.0002, |
| "loss": 0.44, |
| "mean_token_accuracy": 0.863289874792099, |
| "num_tokens": 28323622.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.5147525526583194, |
| "epoch": 3.9121245828698554, |
| "grad_norm": 0.1261458545923233, |
| "learning_rate": 0.0002, |
| "loss": 0.4702, |
| "mean_token_accuracy": 0.8576680824160576, |
| "num_tokens": 28644498.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.5058113448321819, |
| "epoch": 3.9566184649610676, |
| "grad_norm": 0.12469454854726791, |
| "learning_rate": 0.0002, |
| "loss": 0.4509, |
| "mean_token_accuracy": 0.8585198208689689, |
| "num_tokens": 28972178.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.5309454324917916, |
| "epoch": 4.0, |
| "grad_norm": 0.14466610550880432, |
| "learning_rate": 0.0002, |
| "loss": 0.4892, |
| "mean_token_accuracy": 0.8545252420963385, |
| "num_tokens": 29281284.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.4318658284842968, |
| "epoch": 4.044493882091213, |
| "grad_norm": 0.1319647878408432, |
| "learning_rate": 0.0002, |
| "loss": 0.361, |
| "mean_token_accuracy": 0.8950821414589882, |
| "num_tokens": 29608964.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.4118408665060997, |
| "epoch": 4.088987764182425, |
| "grad_norm": 0.12470532953739166, |
| "learning_rate": 0.0002, |
| "loss": 0.333, |
| "mean_token_accuracy": 0.901517990231514, |
| "num_tokens": 29936644.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.4389042042195797, |
| "epoch": 4.133481646273637, |
| "grad_norm": 0.1262277513742447, |
| "learning_rate": 0.0002, |
| "loss": 0.3689, |
| "mean_token_accuracy": 0.8918602123856545, |
| "num_tokens": 30264324.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.40702272206544876, |
| "epoch": 4.17797552836485, |
| "grad_norm": 0.1268932968378067, |
| "learning_rate": 0.0002, |
| "loss": 0.3338, |
| "mean_token_accuracy": 0.9011573985219001, |
| "num_tokens": 30591882.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.4825476065278053, |
| "epoch": 4.222469410456062, |
| "grad_norm": 0.11590472608804703, |
| "learning_rate": 0.0002, |
| "loss": 0.4132, |
| "mean_token_accuracy": 0.8850692108273506, |
| "num_tokens": 30911133.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.4279182717204094, |
| "epoch": 4.266963292547275, |
| "grad_norm": 0.1330350935459137, |
| "learning_rate": 0.0002, |
| "loss": 0.3625, |
| "mean_token_accuracy": 0.893429034948349, |
| "num_tokens": 31238813.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.41735711842775347, |
| "epoch": 4.3114571746384875, |
| "grad_norm": 0.13147839903831482, |
| "learning_rate": 0.0002, |
| "loss": 0.3465, |
| "mean_token_accuracy": 0.896250243484974, |
| "num_tokens": 31566493.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.41771056577563287, |
| "epoch": 4.355951056729699, |
| "grad_norm": 0.1316802203655243, |
| "learning_rate": 0.0002, |
| "loss": 0.3503, |
| "mean_token_accuracy": 0.8955738350749016, |
| "num_tokens": 31894173.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.41876513287425043, |
| "epoch": 4.400444938820912, |
| "grad_norm": 0.13833066821098328, |
| "learning_rate": 0.0002, |
| "loss": 0.3486, |
| "mean_token_accuracy": 0.8961227789521218, |
| "num_tokens": 32221752.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.514089448004961, |
| "epoch": 4.4449388209121246, |
| "grad_norm": 0.12354996800422668, |
| "learning_rate": 0.0002, |
| "loss": 0.4487, |
| "mean_token_accuracy": 0.8752555340528488, |
| "num_tokens": 32538458.0, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 2250, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.35167252010879e+17, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
