| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9992069785884219, |
| "eval_steps": 500, |
| "global_step": 630, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.007930214115781126, |
| "grad_norm": 5.831911087036133, |
| "learning_rate": 4.324992481203008e-06, |
| "loss": 0.3658, |
| "mean_token_accuracy": 0.8928569257259369, |
| "num_tokens": 378311.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.01586042823156225, |
| "grad_norm": 2.1580238342285156, |
| "learning_rate": 9.731233082706766e-06, |
| "loss": 0.1998, |
| "mean_token_accuracy": 0.9497495234012604, |
| "num_tokens": 746131.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.023790642347343377, |
| "grad_norm": 1.2876217365264893, |
| "learning_rate": 1.513747368421053e-05, |
| "loss": 0.1267, |
| "mean_token_accuracy": 0.9623284697532654, |
| "num_tokens": 1120541.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.0317208564631245, |
| "grad_norm": 1.1852203607559204, |
| "learning_rate": 2.0543714285714285e-05, |
| "loss": 0.1525, |
| "mean_token_accuracy": 0.9569363176822663, |
| "num_tokens": 1497625.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.03965107057890563, |
| "grad_norm": 1.8743442296981812, |
| "learning_rate": 2.594995488721805e-05, |
| "loss": 0.1745, |
| "mean_token_accuracy": 0.9548388957977295, |
| "num_tokens": 1875615.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.047581284694686754, |
| "grad_norm": 1.139708161354065, |
| "learning_rate": 3.135619548872181e-05, |
| "loss": 0.1247, |
| "mean_token_accuracy": 0.9635096311569213, |
| "num_tokens": 2251890.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.05551149881046788, |
| "grad_norm": 1.131273627281189, |
| "learning_rate": 3.676243609022557e-05, |
| "loss": 0.1306, |
| "mean_token_accuracy": 0.9637615323066712, |
| "num_tokens": 2621447.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.063441712926249, |
| "grad_norm": 1.1811543703079224, |
| "learning_rate": 3.784051927887756e-05, |
| "loss": 0.152, |
| "mean_token_accuracy": 0.9552397429943085, |
| "num_tokens": 2996042.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.07137192704203013, |
| "grad_norm": 0.8909814953804016, |
| "learning_rate": 3.782766416343561e-05, |
| "loss": 0.1375, |
| "mean_token_accuracy": 0.9621576011180878, |
| "num_tokens": 3377601.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.07930214115781126, |
| "grad_norm": 0.9474954009056091, |
| "learning_rate": 3.780493000801809e-05, |
| "loss": 0.1526, |
| "mean_token_accuracy": 0.9590886533260345, |
| "num_tokens": 3751030.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08723235527359238, |
| "grad_norm": 1.1592732667922974, |
| "learning_rate": 3.7772332656427635e-05, |
| "loss": 0.188, |
| "mean_token_accuracy": 0.9521842837333679, |
| "num_tokens": 4131644.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09516256938937351, |
| "grad_norm": 0.8788422346115112, |
| "learning_rate": 3.772989482628866e-05, |
| "loss": 0.1599, |
| "mean_token_accuracy": 0.9550999224185943, |
| "num_tokens": 4512103.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10309278350515463, |
| "grad_norm": 1.4064106941223145, |
| "learning_rate": 3.767764609321513e-05, |
| "loss": 0.1221, |
| "mean_token_accuracy": 0.9651734471321106, |
| "num_tokens": 4893098.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11102299762093576, |
| "grad_norm": 0.9584864377975464, |
| "learning_rate": 3.761562287019881e-05, |
| "loss": 0.1363, |
| "mean_token_accuracy": 0.9598889470100402, |
| "num_tokens": 5271822.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.11895321173671689, |
| "grad_norm": 0.7917718887329102, |
| "learning_rate": 3.754386838223246e-05, |
| "loss": 0.1382, |
| "mean_token_accuracy": 0.9614855170249939, |
| "num_tokens": 5647014.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.126883425852498, |
| "grad_norm": 1.0597527027130127, |
| "learning_rate": 3.746243263618564e-05, |
| "loss": 0.181, |
| "mean_token_accuracy": 0.9508858978748321, |
| "num_tokens": 6012816.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13481363996827914, |
| "grad_norm": 1.0599623918533325, |
| "learning_rate": 3.7371372385954096e-05, |
| "loss": 0.1413, |
| "mean_token_accuracy": 0.9615827381610871, |
| "num_tokens": 6386515.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.14274385408406026, |
| "grad_norm": 0.9258592128753662, |
| "learning_rate": 3.727075109290706e-05, |
| "loss": 0.0975, |
| "mean_token_accuracy": 0.9722805857658386, |
| "num_tokens": 6758951.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.1506740681998414, |
| "grad_norm": 0.909608781337738, |
| "learning_rate": 3.7160638881660015e-05, |
| "loss": 0.1457, |
| "mean_token_accuracy": 0.9616211354732513, |
| "num_tokens": 7130741.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.1586042823156225, |
| "grad_norm": 1.1566433906555176, |
| "learning_rate": 3.7041112491203635e-05, |
| "loss": 0.1571, |
| "mean_token_accuracy": 0.959383350610733, |
| "num_tokens": 7498050.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16653449643140364, |
| "grad_norm": 0.9355993270874023, |
| "learning_rate": 3.691225522142321e-05, |
| "loss": 0.132, |
| "mean_token_accuracy": 0.9633103668689728, |
| "num_tokens": 7870788.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.17446471054718476, |
| "grad_norm": 0.8453308939933777, |
| "learning_rate": 3.677415687504555e-05, |
| "loss": 0.1305, |
| "mean_token_accuracy": 0.9646081447601318, |
| "num_tokens": 8237261.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.1823949246629659, |
| "grad_norm": 0.977247416973114, |
| "learning_rate": 3.6626913695054076e-05, |
| "loss": 0.1653, |
| "mean_token_accuracy": 0.9555340647697449, |
| "num_tokens": 8617139.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19032513877874702, |
| "grad_norm": 1.1029367446899414, |
| "learning_rate": 3.6470628297615407e-05, |
| "loss": 0.1174, |
| "mean_token_accuracy": 0.96415935754776, |
| "num_tokens": 8991532.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.19825535289452814, |
| "grad_norm": 0.6542048454284668, |
| "learning_rate": 3.630540960056459e-05, |
| "loss": 0.1416, |
| "mean_token_accuracy": 0.9587215840816498, |
| "num_tokens": 9370397.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.20618556701030927, |
| "grad_norm": 0.8101022839546204, |
| "learning_rate": 3.613137274749844e-05, |
| "loss": 0.1271, |
| "mean_token_accuracy": 0.9632396280765534, |
| "num_tokens": 9735025.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.2141157811260904, |
| "grad_norm": 0.6271952986717224, |
| "learning_rate": 3.594863902753009e-05, |
| "loss": 0.1172, |
| "mean_token_accuracy": 0.9656077027320862, |
| "num_tokens": 10116982.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.22204599524187152, |
| "grad_norm": 0.7731532454490662, |
| "learning_rate": 3.575733579076067e-05, |
| "loss": 0.1157, |
| "mean_token_accuracy": 0.9643243551254272, |
| "num_tokens": 10484878.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.22997620935765264, |
| "grad_norm": 0.5634732842445374, |
| "learning_rate": 3.55575963595269e-05, |
| "loss": 0.1222, |
| "mean_token_accuracy": 0.964326947927475, |
| "num_tokens": 10859735.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.23790642347343377, |
| "grad_norm": 0.6386313438415527, |
| "learning_rate": 3.534955993548664e-05, |
| "loss": 0.1239, |
| "mean_token_accuracy": 0.9646699249744415, |
| "num_tokens": 11238871.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.2458366375892149, |
| "grad_norm": 1.457926869392395, |
| "learning_rate": 3.513337150260695e-05, |
| "loss": 0.1185, |
| "mean_token_accuracy": 0.9652786076068878, |
| "num_tokens": 11618094.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.253766851704996, |
| "grad_norm": 0.5694575905799866, |
| "learning_rate": 3.4909181726122425e-05, |
| "loss": 0.1229, |
| "mean_token_accuracy": 0.9661637961864471, |
| "num_tokens": 11997741.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2616970658207772, |
| "grad_norm": 0.8648974299430847, |
| "learning_rate": 3.4677146847534146e-05, |
| "loss": 0.1333, |
| "mean_token_accuracy": 0.9635002672672272, |
| "num_tokens": 12366311.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2696272799365583, |
| "grad_norm": 1.0403084754943848, |
| "learning_rate": 3.4437428575722406e-05, |
| "loss": 0.1436, |
| "mean_token_accuracy": 0.9604751408100128, |
| "num_tokens": 12741380.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.2775574940523394, |
| "grad_norm": 0.4657036066055298, |
| "learning_rate": 3.419019397424923e-05, |
| "loss": 0.1005, |
| "mean_token_accuracy": 0.9723070323467254, |
| "num_tokens": 13116264.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2854877081681205, |
| "grad_norm": 0.7393902540206909, |
| "learning_rate": 3.393561534492897e-05, |
| "loss": 0.121, |
| "mean_token_accuracy": 0.9677346348762512, |
| "num_tokens": 13491308.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.2934179222839017, |
| "grad_norm": 0.45850270986557007, |
| "learning_rate": 3.3673870107748426e-05, |
| "loss": 0.1295, |
| "mean_token_accuracy": 0.962225341796875, |
| "num_tokens": 13863363.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.3013481363996828, |
| "grad_norm": 0.9276625514030457, |
| "learning_rate": 3.3405140677219984e-05, |
| "loss": 0.1, |
| "mean_token_accuracy": 0.9689119040966034, |
| "num_tokens": 14237630.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.30927835051546393, |
| "grad_norm": 1.1208475828170776, |
| "learning_rate": 3.312961433525393e-05, |
| "loss": 0.1217, |
| "mean_token_accuracy": 0.966601413488388, |
| "num_tokens": 14619630.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.317208564631245, |
| "grad_norm": 0.734718382358551, |
| "learning_rate": 3.284748310063868e-05, |
| "loss": 0.0896, |
| "mean_token_accuracy": 0.9766981601715088, |
| "num_tokens": 14998117.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3251387787470262, |
| "grad_norm": 0.9824924468994141, |
| "learning_rate": 3.2558943595219744e-05, |
| "loss": 0.1469, |
| "mean_token_accuracy": 0.9596902310848237, |
| "num_tokens": 15370638.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.3330689928628073, |
| "grad_norm": 0.7965070009231567, |
| "learning_rate": 3.226419690687075e-05, |
| "loss": 0.1059, |
| "mean_token_accuracy": 0.9675719797611236, |
| "num_tokens": 15748388.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.34099920697858843, |
| "grad_norm": 0.8247693777084351, |
| "learning_rate": 3.196344844935208e-05, |
| "loss": 0.1152, |
| "mean_token_accuracy": 0.968847393989563, |
| "num_tokens": 16115212.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.34892942109436953, |
| "grad_norm": 0.5400510430335999, |
| "learning_rate": 3.165690781915463e-05, |
| "loss": 0.1086, |
| "mean_token_accuracy": 0.9710552394390106, |
| "num_tokens": 16495021.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.3568596352101507, |
| "grad_norm": 0.5498576760292053, |
| "learning_rate": 3.1344788649428664e-05, |
| "loss": 0.1364, |
| "mean_token_accuracy": 0.9634641230106353, |
| "num_tokens": 16877792.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3647898493259318, |
| "grad_norm": 0.4931318461894989, |
| "learning_rate": 3.1027308461099314e-05, |
| "loss": 0.1195, |
| "mean_token_accuracy": 0.9649541556835175, |
| "num_tokens": 17250761.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.37272006344171293, |
| "grad_norm": 1.09007728099823, |
| "learning_rate": 3.07046885112728e-05, |
| "loss": 0.1317, |
| "mean_token_accuracy": 0.9628723621368408, |
| "num_tokens": 17632124.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.38065027755749403, |
| "grad_norm": 0.6950754523277283, |
| "learning_rate": 3.037715363903869e-05, |
| "loss": 0.1175, |
| "mean_token_accuracy": 0.9647312045097352, |
| "num_tokens": 18004501.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.3885804916732752, |
| "grad_norm": 0.6208556294441223, |
| "learning_rate": 3.004493210877593e-05, |
| "loss": 0.1223, |
| "mean_token_accuracy": 0.9682797908782959, |
| "num_tokens": 18374558.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.3965107057890563, |
| "grad_norm": 0.775266170501709, |
| "learning_rate": 2.9708255451071633e-05, |
| "loss": 0.1335, |
| "mean_token_accuracy": 0.9639385879039765, |
| "num_tokens": 18749304.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40444091990483744, |
| "grad_norm": 0.9818545579910278, |
| "learning_rate": 2.9367358301363695e-05, |
| "loss": 0.179, |
| "mean_token_accuracy": 0.9543587744235993, |
| "num_tokens": 19122884.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.41237113402061853, |
| "grad_norm": 0.6328598856925964, |
| "learning_rate": 2.9022478236419525e-05, |
| "loss": 0.1219, |
| "mean_token_accuracy": 0.9642945945262908, |
| "num_tokens": 19505791.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.4203013481363997, |
| "grad_norm": 0.5261305570602417, |
| "learning_rate": 2.867385560876492e-05, |
| "loss": 0.1135, |
| "mean_token_accuracy": 0.9672774255275727, |
| "num_tokens": 19874827.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.4282315622521808, |
| "grad_norm": 0.8929963111877441, |
| "learning_rate": 2.8321733379178497e-05, |
| "loss": 0.1144, |
| "mean_token_accuracy": 0.9659302532672882, |
| "num_tokens": 20253047.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.43616177636796194, |
| "grad_norm": 0.9355366230010986, |
| "learning_rate": 2.7966356947368384e-05, |
| "loss": 0.1068, |
| "mean_token_accuracy": 0.9688229739665986, |
| "num_tokens": 20634607.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.44409199048374304, |
| "grad_norm": 0.5547499060630798, |
| "learning_rate": 2.7607973980949138e-05, |
| "loss": 0.0928, |
| "mean_token_accuracy": 0.9735658347606659, |
| "num_tokens": 21018415.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.4520222045995242, |
| "grad_norm": 0.748486340045929, |
| "learning_rate": 2.7246834242838194e-05, |
| "loss": 0.144, |
| "mean_token_accuracy": 0.9614684402942657, |
| "num_tokens": 21392870.0, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4599524187153053, |
| "grad_norm": 0.5803206562995911, |
| "learning_rate": 2.6883189417192004e-05, |
| "loss": 0.0958, |
| "mean_token_accuracy": 0.9710721135139465, |
| "num_tokens": 21766369.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.46788263283108644, |
| "grad_norm": 0.7718275785446167, |
| "learning_rate": 2.651729293400326e-05, |
| "loss": 0.1135, |
| "mean_token_accuracy": 0.9700194001197815, |
| "num_tokens": 22147309.0, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.47581284694686754, |
| "grad_norm": 0.5852587819099426, |
| "learning_rate": 2.6149399792481425e-05, |
| "loss": 0.1034, |
| "mean_token_accuracy": 0.9703018367290497, |
| "num_tokens": 22529500.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.4837430610626487, |
| "grad_norm": 0.5625137686729431, |
| "learning_rate": 2.57797663833396e-05, |
| "loss": 0.1013, |
| "mean_token_accuracy": 0.9700919210910797, |
| "num_tokens": 22905496.0, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.4916732751784298, |
| "grad_norm": 0.6591458916664124, |
| "learning_rate": 2.5408650310111714e-05, |
| "loss": 0.1044, |
| "mean_token_accuracy": 0.9704932272434235, |
| "num_tokens": 23282018.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.49960348929421095, |
| "grad_norm": 0.4330291748046875, |
| "learning_rate": 2.503631020962441e-05, |
| "loss": 0.1112, |
| "mean_token_accuracy": 0.9680072784423828, |
| "num_tokens": 23653981.0, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.507533703409992, |
| "grad_norm": 0.455351859331131, |
| "learning_rate": 2.4663005571748806e-05, |
| "loss": 0.0868, |
| "mean_token_accuracy": 0.9742586433887481, |
| "num_tokens": 24038611.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5154639175257731, |
| "grad_norm": 0.6181245446205139, |
| "learning_rate": 2.428899655855785e-05, |
| "loss": 0.0854, |
| "mean_token_accuracy": 0.9756738305091858, |
| "num_tokens": 24419015.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5233941316415543, |
| "grad_norm": 0.5415076613426208, |
| "learning_rate": 2.3914543823015064e-05, |
| "loss": 0.0913, |
| "mean_token_accuracy": 0.972983330488205, |
| "num_tokens": 24793432.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5313243457573354, |
| "grad_norm": 0.4727752208709717, |
| "learning_rate": 2.353990832732133e-05, |
| "loss": 0.103, |
| "mean_token_accuracy": 0.970764434337616, |
| "num_tokens": 25160028.0, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5392545598731165, |
| "grad_norm": 0.8963496088981628, |
| "learning_rate": 2.3165351161046018e-05, |
| "loss": 0.1503, |
| "mean_token_accuracy": 0.9575360476970672, |
| "num_tokens": 25532928.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5471847739888978, |
| "grad_norm": 0.5121340155601501, |
| "learning_rate": 2.279113335916947e-05, |
| "loss": 0.1041, |
| "mean_token_accuracy": 0.9712670624256134, |
| "num_tokens": 25919474.0, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5551149881046789, |
| "grad_norm": 0.681879997253418, |
| "learning_rate": 2.2417515720163462e-05, |
| "loss": 0.1037, |
| "mean_token_accuracy": 0.9695886611938477, |
| "num_tokens": 26290618.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.56304520222046, |
| "grad_norm": 1.0346599817276, |
| "learning_rate": 2.20447586242365e-05, |
| "loss": 0.1211, |
| "mean_token_accuracy": 0.9666536629199982, |
| "num_tokens": 26667987.0, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.570975416336241, |
| "grad_norm": 0.9770829081535339, |
| "learning_rate": 2.1673121851870622e-05, |
| "loss": 0.1685, |
| "mean_token_accuracy": 0.954728877544403, |
| "num_tokens": 27041066.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5789056304520223, |
| "grad_norm": 0.691178023815155, |
| "learning_rate": 2.1302864402776154e-05, |
| "loss": 0.0996, |
| "mean_token_accuracy": 0.9713979303836823, |
| "num_tokens": 27413889.0, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.5868358445678034, |
| "grad_norm": 1.1742725372314453, |
| "learning_rate": 2.0934244315390614e-05, |
| "loss": 0.1119, |
| "mean_token_accuracy": 0.9691739916801453, |
| "num_tokens": 27785470.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.5947660586835845, |
| "grad_norm": 1.2198081016540527, |
| "learning_rate": 2.056751848704752e-05, |
| "loss": 0.1589, |
| "mean_token_accuracy": 0.9572734713554383, |
| "num_tokens": 28159193.0, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6026962727993656, |
| "grad_norm": 0.602756142616272, |
| "learning_rate": 2.0202942494940434e-05, |
| "loss": 0.1037, |
| "mean_token_accuracy": 0.9694911718368531, |
| "num_tokens": 28539690.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6106264869151468, |
| "grad_norm": 1.1811344623565674, |
| "learning_rate": 1.984077041800711e-05, |
| "loss": 0.1116, |
| "mean_token_accuracy": 0.96877019405365, |
| "num_tokens": 28911785.0, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6185567010309279, |
| "grad_norm": 0.7665280103683472, |
| "learning_rate": 1.9481254659857714e-05, |
| "loss": 0.1246, |
| "mean_token_accuracy": 0.9625495433807373, |
| "num_tokens": 29292768.0, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.626486915146709, |
| "grad_norm": 0.7360073924064636, |
| "learning_rate": 1.912464577287066e-05, |
| "loss": 0.1222, |
| "mean_token_accuracy": 0.9631928503513336, |
| "num_tokens": 29663714.0, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.63441712926249, |
| "grad_norm": 0.7629641890525818, |
| "learning_rate": 1.8771192283578552e-05, |
| "loss": 0.1051, |
| "mean_token_accuracy": 0.9696576714515686, |
| "num_tokens": 30043782.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6423473433782713, |
| "grad_norm": 0.5978000164031982, |
| "learning_rate": 1.842114051946607e-05, |
| "loss": 0.1009, |
| "mean_token_accuracy": 0.9704001903533935, |
| "num_tokens": 30417830.0, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6502775574940524, |
| "grad_norm": 0.5232475996017456, |
| "learning_rate": 1.8074734437300252e-05, |
| "loss": 0.0833, |
| "mean_token_accuracy": 0.9746542632579803, |
| "num_tokens": 30802961.0, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6582077716098335, |
| "grad_norm": 0.5504288077354431, |
| "learning_rate": 1.773221545311313e-05, |
| "loss": 0.1309, |
| "mean_token_accuracy": 0.9640195429325104, |
| "num_tokens": 31178942.0, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6661379857256146, |
| "grad_norm": 0.5151984095573425, |
| "learning_rate": 1.7393822273954927e-05, |
| "loss": 0.0968, |
| "mean_token_accuracy": 0.9719594359397888, |
| "num_tokens": 31557788.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6740681998413958, |
| "grad_norm": 0.7249670028686523, |
| "learning_rate": 1.7059790731535253e-05, |
| "loss": 0.1127, |
| "mean_token_accuracy": 0.9684128284454345, |
| "num_tokens": 31931234.0, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.6819984139571769, |
| "grad_norm": 0.9256052374839783, |
| "learning_rate": 1.6730353617868142e-05, |
| "loss": 0.1046, |
| "mean_token_accuracy": 0.9703239917755127, |
| "num_tokens": 32315128.0, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.689928628072958, |
| "grad_norm": 0.9300370812416077, |
| "learning_rate": 1.6405740523035514e-05, |
| "loss": 0.1388, |
| "mean_token_accuracy": 0.960295993089676, |
| "num_tokens": 32685548.0, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.6978588421887391, |
| "grad_norm": 0.48336443305015564, |
| "learning_rate": 1.6086177675182108e-05, |
| "loss": 0.1039, |
| "mean_token_accuracy": 0.9706290900707245, |
| "num_tokens": 33059246.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7057890563045203, |
| "grad_norm": 0.764839768409729, |
| "learning_rate": 1.5771887782853435e-05, |
| "loss": 0.109, |
| "mean_token_accuracy": 0.9691159665584564, |
| "num_tokens": 33430764.0, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7137192704203014, |
| "grad_norm": 0.6993154287338257, |
| "learning_rate": 1.5463089879786523e-05, |
| "loss": 0.0949, |
| "mean_token_accuracy": 0.9726164698600769, |
| "num_tokens": 33808602.0, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7216494845360825, |
| "grad_norm": 0.541619598865509, |
| "learning_rate": 1.5159999172261788e-05, |
| "loss": 0.0955, |
| "mean_token_accuracy": 0.9709022462368011, |
| "num_tokens": 34180985.0, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.7295796986518636, |
| "grad_norm": 0.6246166229248047, |
| "learning_rate": 1.486282688912226e-05, |
| "loss": 0.1021, |
| "mean_token_accuracy": 0.9721058547496796, |
| "num_tokens": 34550143.0, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.7375099127676448, |
| "grad_norm": 0.5300966501235962, |
| "learning_rate": 1.457178013456476e-05, |
| "loss": 0.1075, |
| "mean_token_accuracy": 0.9692781269550323, |
| "num_tokens": 34930154.0, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7454401268834259, |
| "grad_norm": 0.697161078453064, |
| "learning_rate": 1.4287061743805637e-05, |
| "loss": 0.0831, |
| "mean_token_accuracy": 0.9756361067295074, |
| "num_tokens": 35308665.0, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.753370340999207, |
| "grad_norm": 0.585613489151001, |
| "learning_rate": 1.4008870141721574e-05, |
| "loss": 0.0909, |
| "mean_token_accuracy": 0.9729478836059571, |
| "num_tokens": 35681715.0, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7613005551149881, |
| "grad_norm": 0.9559932947158813, |
| "learning_rate": 1.3737399204564117e-05, |
| "loss": 0.1263, |
| "mean_token_accuracy": 0.9682435929775238, |
| "num_tokens": 36064165.0, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7692307692307693, |
| "grad_norm": 0.7164075374603271, |
| "learning_rate": 1.3472838124844096e-05, |
| "loss": 0.1162, |
| "mean_token_accuracy": 0.9657099366188049, |
| "num_tokens": 36439927.0, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7771609833465504, |
| "grad_norm": 0.6004156470298767, |
| "learning_rate": 1.321537127948035e-05, |
| "loss": 0.0952, |
| "mean_token_accuracy": 0.9702759087085724, |
| "num_tokens": 36815220.0, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.7850911974623315, |
| "grad_norm": 0.43146395683288574, |
| "learning_rate": 1.2965178101304418e-05, |
| "loss": 0.0856, |
| "mean_token_accuracy": 0.9753100454807282, |
| "num_tokens": 37184059.0, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.7930214115781126, |
| "grad_norm": 0.6236276626586914, |
| "learning_rate": 1.2722432954010879e-05, |
| "loss": 0.1086, |
| "mean_token_accuracy": 0.9676127433776855, |
| "num_tokens": 37555139.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8009516256938938, |
| "grad_norm": 0.5673333406448364, |
| "learning_rate": 1.2487305010640487e-05, |
| "loss": 0.1346, |
| "mean_token_accuracy": 0.9613774538040161, |
| "num_tokens": 37932593.0, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8088818398096749, |
| "grad_norm": 0.8051638007164001, |
| "learning_rate": 1.2259958135680695e-05, |
| "loss": 0.1084, |
| "mean_token_accuracy": 0.9688047528266907, |
| "num_tokens": 38307934.0, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.816812053925456, |
| "grad_norm": 0.6798306703567505, |
| "learning_rate": 1.2040550770865838e-05, |
| "loss": 0.1069, |
| "mean_token_accuracy": 0.9685763716697693, |
| "num_tokens": 38683643.0, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8247422680412371, |
| "grad_norm": 0.611504077911377, |
| "learning_rate": 1.1829235824756539e-05, |
| "loss": 0.0988, |
| "mean_token_accuracy": 0.9698693156242371, |
| "num_tokens": 39056282.0, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8326724821570183, |
| "grad_norm": 0.7180154919624329, |
| "learning_rate": 1.1626160566175199e-05, |
| "loss": 0.1016, |
| "mean_token_accuracy": 0.9724007725715638, |
| "num_tokens": 39435119.0, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8406026962727994, |
| "grad_norm": 0.5224301815032959, |
| "learning_rate": 1.1431466521571993e-05, |
| "loss": 0.0939, |
| "mean_token_accuracy": 0.9735538899898529, |
| "num_tokens": 39813876.0, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8485329103885805, |
| "grad_norm": 0.5703458189964294, |
| "learning_rate": 1.1245289376392789e-05, |
| "loss": 0.0954, |
| "mean_token_accuracy": 0.9708606123924255, |
| "num_tokens": 40189069.0, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8564631245043616, |
| "grad_norm": 0.8698118925094604, |
| "learning_rate": 1.1067758880517723e-05, |
| "loss": 0.1046, |
| "mean_token_accuracy": 0.9705537259578705, |
| "num_tokens": 40567081.0, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8643933386201428, |
| "grad_norm": 0.773918092250824, |
| "learning_rate": 1.0898998757836425e-05, |
| "loss": 0.0959, |
| "mean_token_accuracy": 0.9754850029945373, |
| "num_tokens": 40945524.0, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8723235527359239, |
| "grad_norm": 0.7313055992126465, |
| "learning_rate": 1.0739126620022823e-05, |
| "loss": 0.0926, |
| "mean_token_accuracy": 0.9718107759952546, |
| "num_tokens": 41330413.0, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.880253766851705, |
| "grad_norm": 0.48010993003845215, |
| "learning_rate": 1.0588253884569649e-05, |
| "loss": 0.0886, |
| "mean_token_accuracy": 0.9730834066867828, |
| "num_tokens": 41716094.0, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.8881839809674861, |
| "grad_norm": 0.9055776596069336, |
| "learning_rate": 1.0446485697139776e-05, |
| "loss": 0.1138, |
| "mean_token_accuracy": 0.9654557406902313, |
| "num_tokens": 42082634.0, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.8961141950832673, |
| "grad_norm": 0.4968336820602417, |
| "learning_rate": 1.03139208582885e-05, |
| "loss": 0.0785, |
| "mean_token_accuracy": 0.9763318955898285, |
| "num_tokens": 42459696.0, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.9040444091990484, |
| "grad_norm": 0.5969836115837097, |
| "learning_rate": 1.0190651754607829e-05, |
| "loss": 0.0809, |
| "mean_token_accuracy": 0.9777667760848999, |
| "num_tokens": 42827539.0, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9119746233148295, |
| "grad_norm": 0.4573867619037628, |
| "learning_rate": 1.0076764294340752e-05, |
| "loss": 0.1015, |
| "mean_token_accuracy": 0.9714258730411529, |
| "num_tokens": 43208449.0, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9199048374306106, |
| "grad_norm": 0.6519758105278015, |
| "learning_rate": 9.972337847510408e-06, |
| "loss": 0.1161, |
| "mean_token_accuracy": 0.9646508097648621, |
| "num_tokens": 43592012.0, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9278350515463918, |
| "grad_norm": 0.6176139712333679, |
| "learning_rate": 9.877445190605806e-06, |
| "loss": 0.0875, |
| "mean_token_accuracy": 0.9729797005653381, |
| "num_tokens": 43976872.0, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9357652656621729, |
| "grad_norm": 0.9633002877235413, |
| "learning_rate": 9.792152455862712e-06, |
| "loss": 0.127, |
| "mean_token_accuracy": 0.9668839693069458, |
| "num_tokens": 44354868.0, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.943695479777954, |
| "grad_norm": 0.6179344654083252, |
| "learning_rate": 9.716519085175029e-06, |
| "loss": 0.1059, |
| "mean_token_accuracy": 0.967372578382492, |
| "num_tokens": 44735550.0, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9516256938937351, |
| "grad_norm": 0.49601414799690247, |
| "learning_rate": 9.650597788668761e-06, |
| "loss": 0.0839, |
| "mean_token_accuracy": 0.9758542478084564, |
| "num_tokens": 45103847.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9595559080095163, |
| "grad_norm": 0.373835027217865, |
| "learning_rate": 9.59443450796747e-06, |
| "loss": 0.1005, |
| "mean_token_accuracy": 0.9721076548099518, |
| "num_tokens": 45476883.0, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.9674861221252974, |
| "grad_norm": 0.7643465995788574, |
| "learning_rate": 9.54806838417483e-06, |
| "loss": 0.07, |
| "mean_token_accuracy": 0.976939582824707, |
| "num_tokens": 45847570.0, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.9754163362410785, |
| "grad_norm": 0.6381562352180481, |
| "learning_rate": 9.511531730596529e-06, |
| "loss": 0.0672, |
| "mean_token_accuracy": 0.9792495489120483, |
| "num_tokens": 46226820.0, |
| "step": 615 |
| }, |
| { |
| "epoch": 0.9833465503568596, |
| "grad_norm": 0.5611590147018433, |
| "learning_rate": 9.48485001022061e-06, |
| "loss": 0.0832, |
| "mean_token_accuracy": 0.9748390138149261, |
| "num_tokens": 46597049.0, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.9912767644726408, |
| "grad_norm": 0.7576779127120972, |
| "learning_rate": 9.468041817971918e-06, |
| "loss": 0.1028, |
| "mean_token_accuracy": 0.9703056037425994, |
| "num_tokens": 46971605.0, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.9992069785884219, |
| "grad_norm": 0.9836002588272095, |
| "learning_rate": 9.461118867752965e-06, |
| "loss": 0.0934, |
| "mean_token_accuracy": 0.9739538550376892, |
| "num_tokens": 47349784.0, |
| "step": 630 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 630, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.135165704756527e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
