{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0858324715615306,
  "eval_steps": 500,
  "global_step": 2100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.0375,
      "epoch": 0.005170630816959669,
      "grad_norm": 0.27744260430336,
      "learning_rate": 1.9968976215098243e-05,
      "loss": 1.3067,
      "mean_token_accuracy": 0.6855354130268096,
      "num_tokens": 3537600.0,
      "step": 10
    },
    {
      "entropy": 1.88203125,
      "epoch": 0.010341261633919338,
      "grad_norm": 0.17058567702770233,
      "learning_rate": 1.993450534298518e-05,
      "loss": 1.1895,
      "mean_token_accuracy": 0.708834320306778,
      "num_tokens": 7032962.0,
      "step": 20
    },
    {
      "entropy": 1.7921875,
      "epoch": 0.015511892450879007,
      "grad_norm": 0.13976408541202545,
      "learning_rate": 1.9900034470872113e-05,
      "loss": 1.1912,
      "mean_token_accuracy": 0.7101442337036132,
      "num_tokens": 10614996.0,
      "step": 30
    },
    {
      "entropy": 1.7640625,
      "epoch": 0.020682523267838676,
      "grad_norm": 0.13212522864341736,
      "learning_rate": 1.986556359875905e-05,
      "loss": 1.1405,
      "mean_token_accuracy": 0.7191447138786315,
      "num_tokens": 14141355.0,
      "step": 40
    },
    {
      "entropy": 1.70859375,
      "epoch": 0.025853154084798345,
      "grad_norm": 0.1327238529920578,
      "learning_rate": 1.9831092726645986e-05,
      "loss": 1.1139,
      "mean_token_accuracy": 0.7228168249130249,
      "num_tokens": 17721204.0,
      "step": 50
    },
    {
      "entropy": 1.68359375,
      "epoch": 0.031023784901758014,
      "grad_norm": 0.13344530761241913,
      "learning_rate": 1.979662185453292e-05,
      "loss": 1.0762,
      "mean_token_accuracy": 0.729270726442337,
      "num_tokens": 21211732.0,
      "step": 60
    },
    {
      "entropy": 1.6890625,
      "epoch": 0.03619441571871768,
      "grad_norm": 0.13124032318592072,
      "learning_rate": 1.9762150982419856e-05,
      "loss": 1.0805,
      "mean_token_accuracy": 0.7282510459423065,
      "num_tokens": 24751832.0,
      "step": 70
    },
    {
      "entropy": 1.678125,
      "epoch": 0.04136504653567735,
      "grad_norm": 0.1330331563949585,
      "learning_rate": 1.9727680110306793e-05,
      "loss": 1.0846,
      "mean_token_accuracy": 0.727464348077774,
      "num_tokens": 28301590.0,
      "step": 80
    },
    {
      "entropy": 1.66640625,
      "epoch": 0.04653567735263702,
      "grad_norm": 0.1394193321466446,
      "learning_rate": 1.969320923819373e-05,
      "loss": 1.0566,
      "mean_token_accuracy": 0.7313333511352539,
      "num_tokens": 31802125.0,
      "step": 90
    },
    {
      "entropy": 1.65,
      "epoch": 0.05170630816959669,
      "grad_norm": 0.14272728562355042,
      "learning_rate": 1.9658738366080663e-05,
      "loss": 1.0603,
      "mean_token_accuracy": 0.7316244184970856,
      "num_tokens": 35299495.0,
      "step": 100
    },
    {
      "entropy": 1.640625,
      "epoch": 0.05687693898655636,
      "grad_norm": 0.13103780150413513,
      "learning_rate": 1.96242674939676e-05,
      "loss": 1.0617,
      "mean_token_accuracy": 0.7324262440204621,
      "num_tokens": 38770579.0,
      "step": 110
    },
    {
      "entropy": 1.6359375,
      "epoch": 0.06204756980351603,
      "grad_norm": 0.131998211145401,
      "learning_rate": 1.9589796621854536e-05,
      "loss": 1.0699,
      "mean_token_accuracy": 0.7302633821964264,
      "num_tokens": 42276781.0,
      "step": 120
    },
    {
      "entropy": 1.6296875,
      "epoch": 0.0672182006204757,
      "grad_norm": 0.13271059095859528,
      "learning_rate": 1.955532574974147e-05,
      "loss": 1.0701,
      "mean_token_accuracy": 0.7319574892520905,
      "num_tokens": 45791721.0,
      "step": 130
    },
    {
      "entropy": 1.6125,
      "epoch": 0.07238883143743537,
      "grad_norm": 0.1325104981660843,
      "learning_rate": 1.9520854877628406e-05,
      "loss": 1.06,
      "mean_token_accuracy": 0.7329302072525025,
      "num_tokens": 49358721.0,
      "step": 140
    },
    {
      "entropy": 1.615625,
      "epoch": 0.07755946225439504,
      "grad_norm": 0.12658704817295074,
      "learning_rate": 1.9486384005515343e-05,
      "loss": 1.0477,
      "mean_token_accuracy": 0.7350383698940277,
      "num_tokens": 52890301.0,
      "step": 150
    },
    {
      "entropy": 1.63515625,
      "epoch": 0.0827300930713547,
      "grad_norm": 0.13451404869556427,
      "learning_rate": 1.9451913133402276e-05,
      "loss": 1.0558,
      "mean_token_accuracy": 0.733129221200943,
      "num_tokens": 56419066.0,
      "step": 160
    },
    {
      "entropy": 1.62421875,
      "epoch": 0.08790072388831438,
      "grad_norm": 0.1334684193134308,
      "learning_rate": 1.9417442261289213e-05,
      "loss": 1.028,
      "mean_token_accuracy": 0.737556916475296,
      "num_tokens": 59961473.0,
      "step": 170
    },
    {
      "entropy": 1.6125,
      "epoch": 0.09307135470527404,
      "grad_norm": 0.12568821012973785,
      "learning_rate": 1.938297138917615e-05,
      "loss": 1.0458,
      "mean_token_accuracy": 0.7361509501934052,
      "num_tokens": 63434899.0,
      "step": 180
    },
    {
      "entropy": 1.58125,
      "epoch": 0.09824198552223372,
      "grad_norm": 0.12893521785736084,
      "learning_rate": 1.9348500517063083e-05,
      "loss": 1.0201,
      "mean_token_accuracy": 0.7413658320903778,
      "num_tokens": 66996997.0,
      "step": 190
    },
    {
      "entropy": 1.56484375,
      "epoch": 0.10341261633919338,
      "grad_norm": 0.12755005061626434,
      "learning_rate": 1.931402964495002e-05,
      "loss": 1.0229,
      "mean_token_accuracy": 0.7409334897994995,
      "num_tokens": 70549858.0,
      "step": 200
    },
    {
      "entropy": 1.5453125,
      "epoch": 0.10858324715615306,
      "grad_norm": 0.12291249632835388,
      "learning_rate": 1.9279558772836953e-05,
      "loss": 1.0298,
      "mean_token_accuracy": 0.738553661108017,
      "num_tokens": 74086821.0,
      "step": 210
    },
    {
      "entropy": 1.57890625,
      "epoch": 0.11375387797311272,
      "grad_norm": 0.13515888154506683,
      "learning_rate": 1.924508790072389e-05,
      "loss": 1.0365,
      "mean_token_accuracy": 0.7381217360496521,
      "num_tokens": 77581264.0,
      "step": 220
    },
    {
      "entropy": 1.5890625,
      "epoch": 0.1189245087900724,
      "grad_norm": 0.13524512946605682,
      "learning_rate": 1.9210617028610827e-05,
      "loss": 1.0402,
      "mean_token_accuracy": 0.7369829535484314,
      "num_tokens": 81045144.0,
      "step": 230
    },
    {
      "entropy": 1.52734375,
      "epoch": 0.12409513960703206,
      "grad_norm": 0.1315074861049652,
      "learning_rate": 1.917614615649776e-05,
      "loss": 1.0071,
      "mean_token_accuracy": 0.7420039892196655,
      "num_tokens": 84564212.0,
      "step": 240
    },
    {
      "entropy": 1.55703125,
      "epoch": 0.12926577042399173,
      "grad_norm": 0.12327884882688522,
      "learning_rate": 1.9141675284384697e-05,
      "loss": 1.014,
      "mean_token_accuracy": 0.7410085022449493,
      "num_tokens": 88081496.0,
      "step": 250
    },
    {
      "entropy": 1.57890625,
      "epoch": 0.1344364012409514,
      "grad_norm": 0.12867313623428345,
      "learning_rate": 1.910720441227163e-05,
      "loss": 1.0069,
      "mean_token_accuracy": 0.7418998777866364,
      "num_tokens": 91615725.0,
      "step": 260
    },
    {
      "entropy": 1.5484375,
      "epoch": 0.13960703205791106,
      "grad_norm": 0.12194376438856125,
      "learning_rate": 1.9072733540158567e-05,
      "loss": 1.0288,
      "mean_token_accuracy": 0.7407114923000335,
      "num_tokens": 95146899.0,
      "step": 270
    },
    {
      "entropy": 1.57421875,
      "epoch": 0.14477766287487073,
      "grad_norm": 0.1243792474269867,
      "learning_rate": 1.9038262668045503e-05,
      "loss": 1.0273,
      "mean_token_accuracy": 0.7392907023429871,
      "num_tokens": 98742455.0,
      "step": 280
    },
    {
      "entropy": 1.56015625,
      "epoch": 0.1499482936918304,
      "grad_norm": 0.1307244598865509,
      "learning_rate": 1.9003791795932437e-05,
      "loss": 0.9956,
      "mean_token_accuracy": 0.7460803270339966,
      "num_tokens": 102251107.0,
      "step": 290
    },
    {
      "entropy": 1.5421875,
      "epoch": 0.15511892450879008,
      "grad_norm": 0.12532740831375122,
      "learning_rate": 1.8969320923819374e-05,
      "loss": 0.9876,
      "mean_token_accuracy": 0.7475418984889984,
      "num_tokens": 105784628.0,
      "step": 300
    },
    {
      "entropy": 1.5484375,
      "epoch": 0.16028955532574973,
      "grad_norm": 0.13335883617401123,
      "learning_rate": 1.893485005170631e-05,
      "loss": 1.0098,
      "mean_token_accuracy": 0.7438597559928894,
      "num_tokens": 109309371.0,
      "step": 310
    },
    {
      "entropy": 1.54140625,
      "epoch": 0.1654601861427094,
      "grad_norm": 0.12339036166667938,
      "learning_rate": 1.8900379179593244e-05,
      "loss": 1.0017,
      "mean_token_accuracy": 0.7453927099704742,
      "num_tokens": 112813280.0,
      "step": 320
    },
    {
      "entropy": 1.5625,
      "epoch": 0.17063081695966908,
      "grad_norm": 0.1306329220533371,
      "learning_rate": 1.886590830748018e-05,
      "loss": 1.0239,
      "mean_token_accuracy": 0.7398741483688355,
      "num_tokens": 116277173.0,
      "step": 330
    },
    {
      "entropy": 1.55234375,
      "epoch": 0.17580144777662876,
      "grad_norm": 0.1199406087398529,
      "learning_rate": 1.8831437435367117e-05,
      "loss": 1.0089,
      "mean_token_accuracy": 0.7426419973373413,
      "num_tokens": 119775750.0,
      "step": 340
    },
    {
      "entropy": 1.5203125,
      "epoch": 0.1809720785935884,
      "grad_norm": 0.1323672980070114,
      "learning_rate": 1.879696656325405e-05,
      "loss": 0.9968,
      "mean_token_accuracy": 0.7448601007461548,
      "num_tokens": 123248571.0,
      "step": 350
    },
    {
      "entropy": 1.53671875,
      "epoch": 0.18614270941054809,
      "grad_norm": 0.12378353625535965,
      "learning_rate": 1.8762495691140987e-05,
      "loss": 1.0024,
      "mean_token_accuracy": 0.7444860756397247,
      "num_tokens": 126768515.0,
      "step": 360
    },
    {
      "entropy": 1.5515625,
      "epoch": 0.19131334022750776,
      "grad_norm": 0.12648890912532806,
      "learning_rate": 1.8728024819027924e-05,
      "loss": 0.9973,
      "mean_token_accuracy": 0.7450312077999115,
      "num_tokens": 130266395.0,
      "step": 370
    },
    {
      "entropy": 1.5265625,
      "epoch": 0.19648397104446744,
      "grad_norm": 0.12423662841320038,
      "learning_rate": 1.8693553946914857e-05,
      "loss": 0.9861,
      "mean_token_accuracy": 0.7472135782241821,
      "num_tokens": 133821405.0,
      "step": 380
    },
    {
      "entropy": 1.5265625,
      "epoch": 0.20165460186142709,
      "grad_norm": 0.12470203638076782,
      "learning_rate": 1.8659083074801794e-05,
      "loss": 1.0048,
      "mean_token_accuracy": 0.7438580453395843,
      "num_tokens": 137356477.0,
      "step": 390
    },
    {
      "entropy": 1.52734375,
      "epoch": 0.20682523267838676,
      "grad_norm": 0.11875167489051819,
      "learning_rate": 1.862461220268873e-05,
      "loss": 0.9833,
      "mean_token_accuracy": 0.7478334903717041,
      "num_tokens": 140888954.0,
      "step": 400
    },
    {
      "entropy": 1.54609375,
      "epoch": 0.21199586349534644,
      "grad_norm": 0.11470374464988708,
      "learning_rate": 1.8590141330575667e-05,
      "loss": 0.985,
      "mean_token_accuracy": 0.7477921664714813,
      "num_tokens": 144421867.0,
      "step": 410
    },
    {
      "entropy": 1.53515625,
      "epoch": 0.2171664943123061,
      "grad_norm": 0.1266632378101349,
      "learning_rate": 1.85556704584626e-05,
      "loss": 0.9762,
      "mean_token_accuracy": 0.7494876086711884,
      "num_tokens": 147936416.0,
      "step": 420
    },
    {
      "entropy": 1.5125,
      "epoch": 0.22233712512926576,
      "grad_norm": 0.13191717863082886,
      "learning_rate": 1.8521199586349537e-05,
      "loss": 0.9747,
      "mean_token_accuracy": 0.750033575296402,
      "num_tokens": 151400555.0,
      "step": 430
    },
    {
      "entropy": 1.51640625,
      "epoch": 0.22750775594622544,
      "grad_norm": 0.12589062750339508,
      "learning_rate": 1.8486728714236474e-05,
      "loss": 0.9784,
      "mean_token_accuracy": 0.7486616790294647,
      "num_tokens": 154906166.0,
      "step": 440
    },
    {
      "entropy": 1.5203125,
      "epoch": 0.2326783867631851,
      "grad_norm": 0.13004685938358307,
      "learning_rate": 1.8452257842123407e-05,
      "loss": 0.9776,
      "mean_token_accuracy": 0.7488125145435334,
      "num_tokens": 158369458.0,
      "step": 450
    },
    {
      "entropy": 1.503125,
      "epoch": 0.2378490175801448,
      "grad_norm": 0.12742707133293152,
      "learning_rate": 1.8417786970010344e-05,
      "loss": 0.9906,
      "mean_token_accuracy": 0.746845805644989,
      "num_tokens": 161916438.0,
      "step": 460
    },
    {
      "entropy": 1.51640625,
      "epoch": 0.24301964839710444,
      "grad_norm": 0.13093426823616028,
      "learning_rate": 1.838331609789728e-05,
      "loss": 0.995,
      "mean_token_accuracy": 0.7453965723514557,
      "num_tokens": 165415312.0,
      "step": 470
    },
    {
      "entropy": 1.56015625,
      "epoch": 0.2481902792140641,
      "grad_norm": 0.13128837943077087,
      "learning_rate": 1.8348845225784214e-05,
      "loss": 1.0114,
      "mean_token_accuracy": 0.7426946878433227,
      "num_tokens": 168898815.0,
      "step": 480
    },
    {
      "entropy": 1.55703125,
      "epoch": 0.2533609100310238,
      "grad_norm": 0.13076794147491455,
      "learning_rate": 1.831437435367115e-05,
      "loss": 0.9759,
      "mean_token_accuracy": 0.7508574306964875,
      "num_tokens": 172415633.0,
      "step": 490
    },
    {
      "entropy": 1.521875,
      "epoch": 0.25853154084798347,
      "grad_norm": 0.12472254037857056,
      "learning_rate": 1.8279903481558084e-05,
      "loss": 0.9674,
      "mean_token_accuracy": 0.7516848564147949,
      "num_tokens": 175959198.0,
      "step": 500
    },
    {
      "entropy": 1.53515625,
      "epoch": 0.26370217166494314,
      "grad_norm": 0.1270027905702591,
      "learning_rate": 1.824543260944502e-05,
      "loss": 0.9842,
      "mean_token_accuracy": 0.7485461413860321,
      "num_tokens": 179473499.0,
      "step": 510
    },
    {
      "entropy": 1.540625,
      "epoch": 0.2688728024819028,
      "grad_norm": 0.1286272406578064,
      "learning_rate": 1.8210961737331958e-05,
      "loss": 0.9872,
      "mean_token_accuracy": 0.7479561626911163,
      "num_tokens": 182948456.0,
      "step": 520
    },
    {
      "entropy": 1.51015625,
      "epoch": 0.27404343329886244,
      "grad_norm": 0.12400833517313004,
      "learning_rate": 1.817649086521889e-05,
      "loss": 0.9621,
      "mean_token_accuracy": 0.7519203543663024,
      "num_tokens": 186385965.0,
      "step": 530
    },
    {
      "entropy": 1.51328125,
      "epoch": 0.2792140641158221,
      "grad_norm": 0.1264849752187729,
      "learning_rate": 1.8142019993105828e-05,
      "loss": 0.9668,
      "mean_token_accuracy": 0.7518666923046112,
      "num_tokens": 189926499.0,
      "step": 540
    },
    {
      "entropy": 1.50859375,
      "epoch": 0.2843846949327818,
      "grad_norm": 0.12239896506071091,
      "learning_rate": 1.810754912099276e-05,
      "loss": 0.9713,
      "mean_token_accuracy": 0.7509606957435608,
      "num_tokens": 193432394.0,
      "step": 550
    },
    {
      "entropy": 1.5328125,
      "epoch": 0.28955532574974147,
      "grad_norm": 0.12010076642036438,
      "learning_rate": 1.8073078248879698e-05,
      "loss": 0.9739,
      "mean_token_accuracy": 0.7502519965171814,
      "num_tokens": 196907834.0,
      "step": 560
    },
    {
      "entropy": 1.52265625,
      "epoch": 0.29472595656670114,
      "grad_norm": 0.12827058136463165,
      "learning_rate": 1.8038607376766634e-05,
      "loss": 0.9887,
      "mean_token_accuracy": 0.7459660708904267,
      "num_tokens": 200401413.0,
      "step": 570
    },
    {
      "entropy": 1.5109375,
      "epoch": 0.2998965873836608,
      "grad_norm": 0.1177738755941391,
      "learning_rate": 1.8004136504653568e-05,
      "loss": 0.9623,
      "mean_token_accuracy": 0.7526758849620819,
      "num_tokens": 203921751.0,
      "step": 580
    },
    {
      "entropy": 1.5265625,
      "epoch": 0.3050672182006205,
      "grad_norm": 0.1272975504398346,
      "learning_rate": 1.7969665632540504e-05,
      "loss": 0.98,
      "mean_token_accuracy": 0.749792069196701,
      "num_tokens": 207392847.0,
      "step": 590
    },
    {
      "entropy": 1.4953125,
      "epoch": 0.31023784901758017,
      "grad_norm": 0.12402703613042831,
      "learning_rate": 1.793519476042744e-05,
      "loss": 0.9855,
      "mean_token_accuracy": 0.7477928400039673,
      "num_tokens": 210946738.0,
      "step": 600
    },
    {
      "entropy": 1.50078125,
      "epoch": 0.3154084798345398,
      "grad_norm": 0.1203068271279335,
      "learning_rate": 1.7900723888314374e-05,
      "loss": 0.9592,
      "mean_token_accuracy": 0.7534137964248657,
      "num_tokens": 214442947.0,
      "step": 610
    },
    {
      "entropy": 1.54296875,
      "epoch": 0.32057911065149947,
      "grad_norm": 0.12435421347618103,
      "learning_rate": 1.786625301620131e-05,
      "loss": 0.969,
      "mean_token_accuracy": 0.7514902472496032,
      "num_tokens": 217974343.0,
      "step": 620
    },
    {
      "entropy": 1.5265625,
      "epoch": 0.32574974146845914,
      "grad_norm": 0.12747792899608612,
      "learning_rate": 1.7831782144088248e-05,
      "loss": 0.975,
      "mean_token_accuracy": 0.7483082413673401,
      "num_tokens": 221488604.0,
      "step": 630
    },
    {
      "entropy": 1.52734375,
      "epoch": 0.3309203722854188,
      "grad_norm": 0.12488026171922684,
      "learning_rate": 1.779731127197518e-05,
      "loss": 0.9989,
      "mean_token_accuracy": 0.745906400680542,
      "num_tokens": 225024161.0,
      "step": 640
    },
    {
      "entropy": 1.484375,
      "epoch": 0.3360910031023785,
      "grad_norm": 0.11778416484594345,
      "learning_rate": 1.7762840399862118e-05,
      "loss": 0.9765,
      "mean_token_accuracy": 0.7498968541622162,
      "num_tokens": 228517539.0,
      "step": 650
    },
    {
      "entropy": 1.4875,
      "epoch": 0.34126163391933817,
      "grad_norm": 0.11822066456079483,
      "learning_rate": 1.7728369527749055e-05,
      "loss": 0.9769,
      "mean_token_accuracy": 0.7499968707561493,
      "num_tokens": 232050772.0,
      "step": 660
    },
    {
      "entropy": 1.48046875,
      "epoch": 0.34643226473629785,
      "grad_norm": 0.11642944067716599,
      "learning_rate": 1.7693898655635988e-05,
      "loss": 0.9656,
      "mean_token_accuracy": 0.7531766474246979,
      "num_tokens": 235589506.0,
      "step": 670
    },
    {
      "entropy": 1.4984375,
      "epoch": 0.3516028955532575,
      "grad_norm": 0.11931689083576202,
      "learning_rate": 1.7659427783522925e-05,
      "loss": 0.9602,
      "mean_token_accuracy": 0.753033047914505,
      "num_tokens": 239103526.0,
      "step": 680
    },
    {
      "entropy": 1.484375,
      "epoch": 0.35677352637021714,
      "grad_norm": 0.10961424559354782,
      "learning_rate": 1.762495691140986e-05,
      "loss": 0.9283,
      "mean_token_accuracy": 0.7590166687965393,
      "num_tokens": 242628802.0,
      "step": 690
    },
    {
      "entropy": 1.48359375,
      "epoch": 0.3619441571871768,
      "grad_norm": 0.11637769639492035,
      "learning_rate": 1.7590486039296795e-05,
      "loss": 0.9534,
      "mean_token_accuracy": 0.7538342356681824,
      "num_tokens": 246124006.0,
      "step": 700
    },
    {
      "entropy": 1.49921875,
      "epoch": 0.3671147880041365,
      "grad_norm": 0.13185293972492218,
      "learning_rate": 1.755601516718373e-05,
      "loss": 0.9766,
      "mean_token_accuracy": 0.7504629611968994,
      "num_tokens": 249627086.0,
      "step": 710
    },
    {
      "entropy": 1.5203125,
      "epoch": 0.37228541882109617,
      "grad_norm": 0.11748187988996506,
      "learning_rate": 1.7521544295070668e-05,
      "loss": 0.9696,
      "mean_token_accuracy": 0.7497382581233978,
      "num_tokens": 253142510.0,
      "step": 720
    },
    {
      "entropy": 1.50703125,
      "epoch": 0.37745604963805585,
      "grad_norm": 0.12026237696409225,
      "learning_rate": 1.7487073422957605e-05,
      "loss": 0.9348,
      "mean_token_accuracy": 0.7580827891826629,
      "num_tokens": 256734152.0,
      "step": 730
    },
    {
      "entropy": 1.53671875,
      "epoch": 0.3826266804550155,
      "grad_norm": 0.1143966093659401,
      "learning_rate": 1.7452602550844538e-05,
      "loss": 0.9895,
      "mean_token_accuracy": 0.746571522951126,
      "num_tokens": 260325040.0,
      "step": 740
    },
    {
      "entropy": 1.52421875,
      "epoch": 0.3877973112719752,
      "grad_norm": 0.12038221210241318,
      "learning_rate": 1.7418131678731475e-05,
      "loss": 0.9492,
      "mean_token_accuracy": 0.7550131738185882,
      "num_tokens": 263838715.0,
      "step": 750
    },
    {
      "entropy": 1.5015625,
      "epoch": 0.3929679420889349,
      "grad_norm": 0.12316713482141495,
      "learning_rate": 1.7383660806618408e-05,
      "loss": 0.9403,
      "mean_token_accuracy": 0.7563300907611847,
      "num_tokens": 267341909.0,
      "step": 760
    },
    {
      "entropy": 1.50703125,
      "epoch": 0.3981385729058945,
      "grad_norm": 0.11857284605503082,
      "learning_rate": 1.7349189934505345e-05,
      "loss": 0.9357,
      "mean_token_accuracy": 0.7585722267627716,
      "num_tokens": 270844304.0,
      "step": 770
    },
    {
      "entropy": 1.50390625,
      "epoch": 0.40330920372285417,
      "grad_norm": 0.12760131061077118,
      "learning_rate": 1.731471906239228e-05,
      "loss": 0.958,
      "mean_token_accuracy": 0.7544421136379242,
      "num_tokens": 274347980.0,
      "step": 780
    },
    {
      "entropy": 1.4859375,
      "epoch": 0.40847983453981385,
      "grad_norm": 0.11467541754245758,
      "learning_rate": 1.7280248190279215e-05,
      "loss": 0.9432,
      "mean_token_accuracy": 0.756697702407837,
      "num_tokens": 277878589.0,
      "step": 790
    },
    {
      "entropy": 1.453125,
      "epoch": 0.4136504653567735,
      "grad_norm": 0.1142238974571228,
      "learning_rate": 1.724577731816615e-05,
      "loss": 0.9264,
      "mean_token_accuracy": 0.7594365060329438,
      "num_tokens": 281393449.0,
      "step": 800
    },
    {
      "entropy": 1.465625,
      "epoch": 0.4188210961737332,
      "grad_norm": 0.1210210919380188,
      "learning_rate": 1.7211306446053085e-05,
      "loss": 0.9461,
      "mean_token_accuracy": 0.7550443470478058,
      "num_tokens": 284863610.0,
      "step": 810
    },
    {
      "entropy": 1.4859375,
      "epoch": 0.4239917269906929,
      "grad_norm": 0.1252208799123764,
      "learning_rate": 1.7176835573940022e-05,
      "loss": 0.9506,
      "mean_token_accuracy": 0.7554301738739013,
      "num_tokens": 288435640.0,
      "step": 820
    },
    {
      "entropy": 1.46640625,
      "epoch": 0.42916235780765255,
      "grad_norm": 0.12282290309667587,
      "learning_rate": 1.714236470182696e-05,
      "loss": 0.9561,
      "mean_token_accuracy": 0.7522307515144349,
      "num_tokens": 291972853.0,
      "step": 830
    },
    {
      "entropy": 1.48359375,
      "epoch": 0.4343329886246122,
      "grad_norm": 0.13035908341407776,
      "learning_rate": 1.7107893829713892e-05,
      "loss": 0.9588,
      "mean_token_accuracy": 0.7543351411819458,
      "num_tokens": 295448635.0,
      "step": 840
    },
    {
      "entropy": 1.4921875,
      "epoch": 0.43950361944157185,
      "grad_norm": 0.11919616162776947,
      "learning_rate": 1.707342295760083e-05,
      "loss": 0.9802,
      "mean_token_accuracy": 0.748005723953247,
      "num_tokens": 299005991.0,
      "step": 850
    },
    {
      "entropy": 1.4859375,
      "epoch": 0.4446742502585315,
      "grad_norm": 0.12567895650863647,
      "learning_rate": 1.7038952085487762e-05,
      "loss": 0.9692,
      "mean_token_accuracy": 0.7503853142261505,
      "num_tokens": 302531494.0,
      "step": 860
    },
    {
      "entropy": 1.4609375,
      "epoch": 0.4498448810754912,
      "grad_norm": 0.11992225050926208,
      "learning_rate": 1.70044812133747e-05,
      "loss": 0.9247,
      "mean_token_accuracy": 0.760009765625,
      "num_tokens": 306035085.0,
      "step": 870
    },
    {
      "entropy": 1.49609375,
      "epoch": 0.4550155118924509,
      "grad_norm": 0.12116365134716034,
      "learning_rate": 1.6970010341261635e-05,
      "loss": 0.975,
      "mean_token_accuracy": 0.7497194230556488,
      "num_tokens": 309470942.0,
      "step": 880
    },
    {
      "entropy": 1.475,
      "epoch": 0.46018614270941055,
      "grad_norm": 0.11720309406518936,
      "learning_rate": 1.6935539469148572e-05,
      "loss": 0.952,
      "mean_token_accuracy": 0.7535983860492707,
      "num_tokens": 312985893.0,
      "step": 890
    },
    {
      "entropy": 1.49609375,
      "epoch": 0.4653567735263702,
      "grad_norm": 0.11972068250179291,
      "learning_rate": 1.6901068597035505e-05,
      "loss": 0.9631,
      "mean_token_accuracy": 0.7522001743316651,
      "num_tokens": 316493658.0,
      "step": 900
    },
    {
      "entropy": 1.48515625,
      "epoch": 0.4705274043433299,
      "grad_norm": 0.11726847290992737,
      "learning_rate": 1.6866597724922442e-05,
      "loss": 0.9487,
      "mean_token_accuracy": 0.7549257040023803,
      "num_tokens": 319971268.0,
      "step": 910
    },
    {
      "entropy": 1.50390625,
      "epoch": 0.4756980351602896,
      "grad_norm": 0.11658336967229843,
      "learning_rate": 1.683212685280938e-05,
      "loss": 0.9779,
      "mean_token_accuracy": 0.7492849051952362,
      "num_tokens": 323543858.0,
      "step": 920
    },
    {
      "entropy": 1.471875,
      "epoch": 0.4808686659772492,
      "grad_norm": 0.11780999600887299,
      "learning_rate": 1.6797655980696312e-05,
      "loss": 0.9249,
      "mean_token_accuracy": 0.759690934419632,
      "num_tokens": 327128684.0,
      "step": 930
    },
    {
      "entropy": 1.5015625,
      "epoch": 0.4860392967942089,
      "grad_norm": 0.11748922616243362,
      "learning_rate": 1.676318510858325e-05,
      "loss": 0.9644,
      "mean_token_accuracy": 0.752453887462616,
      "num_tokens": 330698096.0,
      "step": 940
    },
    {
      "entropy": 1.5,
      "epoch": 0.49120992761116855,
      "grad_norm": 0.12470361590385437,
      "learning_rate": 1.6728714236470185e-05,
      "loss": 0.9606,
      "mean_token_accuracy": 0.7517075955867767,
      "num_tokens": 334239261.0,
      "step": 950
    },
    {
      "entropy": 1.50078125,
      "epoch": 0.4963805584281282,
      "grad_norm": 0.12562286853790283,
      "learning_rate": 1.669424336435712e-05,
      "loss": 0.9476,
      "mean_token_accuracy": 0.75567626953125,
      "num_tokens": 337713689.0,
      "step": 960
    },
    {
      "entropy": 1.49921875,
      "epoch": 0.5015511892450879,
      "grad_norm": 0.1192832961678505,
      "learning_rate": 1.6659772492244055e-05,
      "loss": 0.9485,
      "mean_token_accuracy": 0.7538661956787109,
      "num_tokens": 341230527.0,
      "step": 970
    },
    {
      "entropy": 1.50234375,
      "epoch": 0.5067218200620476,
      "grad_norm": 0.11119144409894943,
      "learning_rate": 1.6625301620130992e-05,
      "loss": 0.9516,
      "mean_token_accuracy": 0.7550728797912598,
      "num_tokens": 344782636.0,
      "step": 980
    },
    {
      "entropy": 1.5046875,
      "epoch": 0.5118924508790073,
      "grad_norm": 0.12529641389846802,
      "learning_rate": 1.6590830748017926e-05,
      "loss": 0.9704,
      "mean_token_accuracy": 0.750161212682724,
      "num_tokens": 348288352.0,
      "step": 990
    },
    {
      "entropy": 1.54609375,
      "epoch": 0.5170630816959669,
      "grad_norm": 0.12611797451972961,
      "learning_rate": 1.6556359875904862e-05,
      "loss": 0.9823,
      "mean_token_accuracy": 0.7478528261184693,
      "num_tokens": 351754878.0,
      "step": 1000
    },
    {
      "entropy": 1.50390625,
      "epoch": 0.5222337125129266,
      "grad_norm": 0.12376543134450912,
      "learning_rate": 1.65218890037918e-05,
      "loss": 0.9457,
      "mean_token_accuracy": 0.7560109198093414,
      "num_tokens": 355224219.0,
      "step": 1010
    },
    {
      "entropy": 1.503125,
      "epoch": 0.5274043433298863,
      "grad_norm": 0.12620951235294342,
      "learning_rate": 1.6487418131678732e-05,
      "loss": 0.9276,
      "mean_token_accuracy": 0.7603226721286773,
      "num_tokens": 358763362.0,
      "step": 1020
    },
    {
      "entropy": 1.521875,
      "epoch": 0.532574974146846,
      "grad_norm": 0.11898606270551682,
      "learning_rate": 1.645294725956567e-05,
      "loss": 0.9356,
      "mean_token_accuracy": 0.7571122646331787,
      "num_tokens": 362325652.0,
      "step": 1030
    },
    {
      "entropy": 1.5140625,
      "epoch": 0.5377456049638056,
      "grad_norm": 0.12898164987564087,
      "learning_rate": 1.6418476387452606e-05,
      "loss": 0.926,
      "mean_token_accuracy": 0.7587803602218628,
      "num_tokens": 365822057.0,
      "step": 1040
    },
    {
      "entropy": 1.5,
      "epoch": 0.5429162357807652,
      "grad_norm": 0.11916136741638184,
      "learning_rate": 1.638400551533954e-05,
      "loss": 0.9359,
      "mean_token_accuracy": 0.7570879578590393,
      "num_tokens": 369329930.0,
      "step": 1050
    },
    {
      "entropy": 1.48203125,
      "epoch": 0.5480868665977249,
      "grad_norm": 0.12124485522508621,
      "learning_rate": 1.6349534643226476e-05,
      "loss": 0.9515,
      "mean_token_accuracy": 0.7533541798591614,
      "num_tokens": 372877333.0,
      "step": 1060
    },
    {
      "entropy": 1.4828125,
      "epoch": 0.5532574974146846,
      "grad_norm": 0.11895754188299179,
      "learning_rate": 1.631506377111341e-05,
      "loss": 0.9389,
      "mean_token_accuracy": 0.7570245862007141,
      "num_tokens": 376405650.0,
      "step": 1070
    },
    {
      "entropy": 1.47109375,
      "epoch": 0.5584281282316442,
      "grad_norm": 0.12033624947071075,
      "learning_rate": 1.6280592899000346e-05,
      "loss": 0.9536,
      "mean_token_accuracy": 0.7542732715606689,
      "num_tokens": 379955803.0,
      "step": 1080
    },
    {
      "entropy": 1.50859375,
      "epoch": 0.5635987590486039,
      "grad_norm": 0.1252926141023636,
      "learning_rate": 1.6246122026887283e-05,
      "loss": 0.9807,
      "mean_token_accuracy": 0.7493639528751374,
      "num_tokens": 383508886.0,
      "step": 1090
    },
    {
      "entropy": 1.45859375,
      "epoch": 0.5687693898655636,
      "grad_norm": 0.11495286226272583,
      "learning_rate": 1.6211651154774216e-05,
      "loss": 0.9302,
      "mean_token_accuracy": 0.758233267068863,
      "num_tokens": 387045540.0,
      "step": 1100
    },
    {
      "entropy": 1.47890625,
      "epoch": 0.5739400206825233,
      "grad_norm": 0.1155555322766304,
      "learning_rate": 1.6177180282661153e-05,
      "loss": 0.9276,
      "mean_token_accuracy": 0.759257459640503,
      "num_tokens": 390553827.0,
      "step": 1110
    },
    {
      "entropy": 1.48125,
      "epoch": 0.5791106514994829,
      "grad_norm": 0.12705214321613312,
      "learning_rate": 1.6142709410548086e-05,
      "loss": 0.9367,
      "mean_token_accuracy": 0.7574102461338044,
      "num_tokens": 394086014.0,
      "step": 1120
    },
    {
      "entropy": 1.475,
      "epoch": 0.5842812823164426,
      "grad_norm": 0.11858130246400833,
      "learning_rate": 1.6108238538435023e-05,
      "loss": 0.9397,
      "mean_token_accuracy": 0.7564952373504639,
      "num_tokens": 397627950.0,
      "step": 1130
    },
    {
      "entropy": 1.46796875,
      "epoch": 0.5894519131334023,
      "grad_norm": 0.12174364179372787,
      "learning_rate": 1.607376766632196e-05,
      "loss": 0.9317,
      "mean_token_accuracy": 0.758020156621933,
      "num_tokens": 401141051.0,
      "step": 1140
    },
    {
      "entropy": 1.48671875,
      "epoch": 0.594622543950362,
      "grad_norm": 0.12224062532186508,
      "learning_rate": 1.6039296794208893e-05,
      "loss": 0.9298,
      "mean_token_accuracy": 0.7598855435848236,
      "num_tokens": 404694343.0,
      "step": 1150
    },
    {
      "entropy": 1.48671875,
      "epoch": 0.5997931747673216,
      "grad_norm": 0.12275724858045578,
      "learning_rate": 1.600482592209583e-05,
      "loss": 0.9564,
      "mean_token_accuracy": 0.7541142761707306,
      "num_tokens": 408162048.0,
      "step": 1160
    },
    {
      "entropy": 1.48671875,
      "epoch": 0.6049638055842813,
      "grad_norm": 0.12119049578905106,
      "learning_rate": 1.5970355049982766e-05,
      "loss": 0.9241,
      "mean_token_accuracy": 0.7597784161567688,
      "num_tokens": 411698032.0,
      "step": 1170
    },
    {
      "entropy": 1.46875,
      "epoch": 0.610134436401241,
      "grad_norm": 0.11573296785354614,
      "learning_rate": 1.59358841778697e-05,
      "loss": 0.9028,
      "mean_token_accuracy": 0.7644691288471221,
      "num_tokens": 415198276.0,
      "step": 1180
    },
    {
      "entropy": 1.48984375,
      "epoch": 0.6153050672182007,
      "grad_norm": 0.12336082011461258,
      "learning_rate": 1.5901413305756636e-05,
      "loss": 0.9059,
      "mean_token_accuracy": 0.763912558555603,
      "num_tokens": 418689534.0,
      "step": 1190
    },
    {
      "entropy": 1.46953125,
      "epoch": 0.6204756980351603,
      "grad_norm": 0.1194983720779419,
      "learning_rate": 1.5866942433643573e-05,
      "loss": 0.9387,
      "mean_token_accuracy": 0.7568131923675537,
      "num_tokens": 422206745.0,
      "step": 1200
    },
    {
      "entropy": 1.47109375,
      "epoch": 0.6256463288521199,
      "grad_norm": 0.12008671462535858,
      "learning_rate": 1.583247156153051e-05,
      "loss": 0.9287,
      "mean_token_accuracy": 0.759381753206253,
      "num_tokens": 425735490.0,
      "step": 1210
    },
    {
      "entropy": 1.46484375,
      "epoch": 0.6308169596690796,
      "grad_norm": 0.12218579649925232,
      "learning_rate": 1.5798000689417443e-05,
      "loss": 0.9326,
      "mean_token_accuracy": 0.7594467222690582,
      "num_tokens": 429247925.0,
      "step": 1220
    },
    {
      "entropy": 1.4625,
      "epoch": 0.6359875904860393,
      "grad_norm": 0.12128184735774994,
      "learning_rate": 1.576352981730438e-05,
      "loss": 0.9143,
      "mean_token_accuracy": 0.7615148365497589,
      "num_tokens": 432773801.0,
      "step": 1230
    },
    {
      "entropy": 1.4796875,
      "epoch": 0.6411582213029989,
      "grad_norm": 0.12187935411930084,
      "learning_rate": 1.5729058945191316e-05,
      "loss": 0.92,
      "mean_token_accuracy": 0.7607288718223572,
      "num_tokens": 436248156.0,
      "step": 1240
    },
    {
      "entropy": 1.4765625,
      "epoch": 0.6463288521199586,
      "grad_norm": 0.12327364832162857,
      "learning_rate": 1.569458807307825e-05,
      "loss": 0.9249,
      "mean_token_accuracy": 0.7601008296012879,
      "num_tokens": 439808916.0,
      "step": 1250
    },
    {
      "entropy": 1.47734375,
      "epoch": 0.6514994829369183,
      "grad_norm": 0.12044602632522583,
      "learning_rate": 1.5660117200965186e-05,
      "loss": 0.9456,
      "mean_token_accuracy": 0.7569905340671539,
      "num_tokens": 443343747.0,
      "step": 1260
    },
    {
      "entropy": 1.47734375,
      "epoch": 0.656670113753878,
      "grad_norm": 0.12415958940982819,
      "learning_rate": 1.5625646328852123e-05,
      "loss": 0.935,
      "mean_token_accuracy": 0.7574470579624176,
      "num_tokens": 446845796.0,
      "step": 1270
    },
    {
      "entropy": 1.46796875,
      "epoch": 0.6618407445708376,
      "grad_norm": 0.12678299844264984,
      "learning_rate": 1.5591175456739056e-05,
      "loss": 0.9206,
      "mean_token_accuracy": 0.7590867578983307,
      "num_tokens": 450337419.0,
      "step": 1280
    },
    {
      "entropy": 1.48984375,
      "epoch": 0.6670113753877973,
      "grad_norm": 0.12798433005809784,
      "learning_rate": 1.5556704584625993e-05,
      "loss": 0.9467,
      "mean_token_accuracy": 0.7556301057338715,
      "num_tokens": 453821582.0,
      "step": 1290
    },
    {
      "entropy": 1.4796875,
      "epoch": 0.672182006204757,
      "grad_norm": 0.12155309319496155,
      "learning_rate": 1.552223371251293e-05,
      "loss": 0.9339,
      "mean_token_accuracy": 0.7585957705974579,
      "num_tokens": 457299519.0,
      "step": 1300
    },
    {
      "entropy": 1.47265625,
      "epoch": 0.6773526370217167,
      "grad_norm": 0.12246780097484589,
      "learning_rate": 1.5487762840399863e-05,
      "loss": 0.9403,
      "mean_token_accuracy": 0.7562773644924163,
      "num_tokens": 460871201.0,
      "step": 1310
    },
    {
      "entropy": 1.4984375,
      "epoch": 0.6825232678386763,
      "grad_norm": 0.11269519478082657,
      "learning_rate": 1.54532919682868e-05,
      "loss": 0.9263,
      "mean_token_accuracy": 0.7602653086185456,
      "num_tokens": 464382014.0,
      "step": 1320
    },
    {
      "entropy": 1.47890625,
      "epoch": 0.687693898655636,
      "grad_norm": 0.1227000281214714,
      "learning_rate": 1.5418821096173737e-05,
      "loss": 0.9458,
      "mean_token_accuracy": 0.7562500655651092,
      "num_tokens": 467894476.0,
      "step": 1330
    },
    {
      "entropy": 1.48046875,
      "epoch": 0.6928645294725957,
      "grad_norm": 0.12165822833776474,
      "learning_rate": 1.538435022406067e-05,
      "loss": 0.9325,
      "mean_token_accuracy": 0.756926417350769,
      "num_tokens": 471392902.0,
      "step": 1340
    },
    {
      "entropy": 1.4703125,
      "epoch": 0.6980351602895554,
      "grad_norm": 0.12303000688552856,
      "learning_rate": 1.5349879351947607e-05,
      "loss": 0.9064,
      "mean_token_accuracy": 0.7626752257347107,
      "num_tokens": 474902078.0,
      "step": 1350
    },
    {
      "entropy": 1.46328125,
      "epoch": 0.703205791106515,
      "grad_norm": 0.11540233343839645,
      "learning_rate": 1.531540847983454e-05,
      "loss": 0.9407,
      "mean_token_accuracy": 0.7561467230319977,
      "num_tokens": 478433388.0,
      "step": 1360
    },
    {
      "entropy": 1.49140625,
      "epoch": 0.7083764219234746,
      "grad_norm": 0.12478071451187134,
      "learning_rate": 1.5280937607721477e-05,
      "loss": 0.9359,
      "mean_token_accuracy": 0.7584864318370819,
      "num_tokens": 481921966.0,
      "step": 1370
    },
    {
      "entropy": 1.49375,
      "epoch": 0.7135470527404343,
      "grad_norm": 0.1309724599123001,
      "learning_rate": 1.5246466735608412e-05,
      "loss": 0.9271,
      "mean_token_accuracy": 0.7596810460090637,
      "num_tokens": 485379998.0,
      "step": 1380
    },
    {
      "entropy": 1.46015625,
      "epoch": 0.718717683557394,
      "grad_norm": 0.12762272357940674,
      "learning_rate": 1.5211995863495348e-05,
      "loss": 0.9154,
      "mean_token_accuracy": 0.7618087232112885,
      "num_tokens": 488917377.0,
      "step": 1390
    },
    {
      "entropy": 1.45703125,
      "epoch": 0.7238883143743536,
      "grad_norm": 0.12085416913032532,
      "learning_rate": 1.5177524991382283e-05,
      "loss": 0.9189,
      "mean_token_accuracy": 0.7611732959747315,
      "num_tokens": 492457961.0,
      "step": 1400
    },
    {
      "entropy": 1.4421875,
      "epoch": 0.7290589451913133,
      "grad_norm": 0.12487323582172394,
      "learning_rate": 1.5143054119269218e-05,
      "loss": 0.9286,
      "mean_token_accuracy": 0.7590953767299652,
      "num_tokens": 495919216.0,
      "step": 1410
    },
    {
      "entropy": 1.4453125,
      "epoch": 0.734229576008273,
      "grad_norm": 0.1264040321111679,
      "learning_rate": 1.5108583247156155e-05,
      "loss": 0.9369,
      "mean_token_accuracy": 0.7580833673477173,
      "num_tokens": 499447846.0,
      "step": 1420
    },
    {
      "entropy": 1.45859375,
      "epoch": 0.7394002068252327,
      "grad_norm": 0.11700266599655151,
      "learning_rate": 1.507411237504309e-05,
      "loss": 0.9444,
      "mean_token_accuracy": 0.7564094007015228,
      "num_tokens": 502931170.0,
      "step": 1430
    },
    {
      "entropy": 1.4625,
      "epoch": 0.7445708376421923,
      "grad_norm": 0.12091559171676636,
      "learning_rate": 1.5039641502930025e-05,
      "loss": 0.9403,
      "mean_token_accuracy": 0.756866055727005,
      "num_tokens": 506455188.0,
      "step": 1440
    },
    {
      "entropy": 1.43515625,
      "epoch": 0.749741468459152,
      "grad_norm": 0.11989731341600418,
      "learning_rate": 1.500517063081696e-05,
      "loss": 0.9139,
      "mean_token_accuracy": 0.7620751082897186,
      "num_tokens": 509960186.0,
      "step": 1450
    },
    {
      "entropy": 1.475,
      "epoch": 0.7549120992761117,
      "grad_norm": 0.12092899531126022,
      "learning_rate": 1.4970699758703897e-05,
      "loss": 0.9271,
      "mean_token_accuracy": 0.7599273502826691,
      "num_tokens": 513463939.0,
      "step": 1460
    },
    {
      "entropy": 1.478125,
      "epoch": 0.7600827300930714,
      "grad_norm": 0.11744925379753113,
      "learning_rate": 1.4936228886590832e-05,
      "loss": 0.9298,
      "mean_token_accuracy": 0.7581395506858826,
      "num_tokens": 516984058.0,
      "step": 1470
    },
    {
      "entropy": 1.465625,
      "epoch": 0.765253360910031,
      "grad_norm": 0.11444874107837677,
      "learning_rate": 1.4901758014477767e-05,
      "loss": 0.9228,
      "mean_token_accuracy": 0.760771507024765,
      "num_tokens": 520529841.0,
      "step": 1480
    },
    {
      "entropy": 1.4828125,
      "epoch": 0.7704239917269907,
      "grad_norm": 0.12582576274871826,
      "learning_rate": 1.4867287142364704e-05,
      "loss": 0.9261,
      "mean_token_accuracy": 0.7593208611011505,
      "num_tokens": 524080208.0,
      "step": 1490
    },
    {
      "entropy": 1.4671875,
      "epoch": 0.7755946225439504,
      "grad_norm": 0.11210944503545761,
      "learning_rate": 1.4832816270251637e-05,
      "loss": 0.9086,
      "mean_token_accuracy": 0.7635318100452423,
      "num_tokens": 527634182.0,
      "step": 1500
    },
    {
      "entropy": 1.44375,
      "epoch": 0.7807652533609101,
      "grad_norm": 0.12150448560714722,
      "learning_rate": 1.4798345398138574e-05,
      "loss": 0.9147,
      "mean_token_accuracy": 0.7632191896438598,
      "num_tokens": 531156991.0,
      "step": 1510
    },
    {
      "entropy": 1.453125,
      "epoch": 0.7859358841778697,
      "grad_norm": 0.11527998745441437,
      "learning_rate": 1.476387452602551e-05,
      "loss": 0.9302,
      "mean_token_accuracy": 0.7586703896522522,
      "num_tokens": 534685727.0,
      "step": 1520
    },
    {
      "entropy": 1.45859375,
      "epoch": 0.7911065149948294,
      "grad_norm": 0.11719609051942825,
      "learning_rate": 1.4729403653912444e-05,
      "loss": 0.9273,
      "mean_token_accuracy": 0.7598319470882415,
      "num_tokens": 538205698.0,
      "step": 1530
    },
    {
      "entropy": 1.44765625,
      "epoch": 0.796277145811789,
      "grad_norm": 0.11949972808361053,
      "learning_rate": 1.469493278179938e-05,
      "loss": 0.892,
      "mean_token_accuracy": 0.7663702487945556,
      "num_tokens": 541713751.0,
      "step": 1540
    },
    {
      "entropy": 1.47578125,
      "epoch": 0.8014477766287487,
      "grad_norm": 0.1147402822971344,
      "learning_rate": 1.4660461909686317e-05,
      "loss": 0.9391,
      "mean_token_accuracy": 0.75692800283432,
      "num_tokens": 545255223.0,
      "step": 1550
    },
    {
      "entropy": 1.484375,
      "epoch": 0.8066184074457083,
      "grad_norm": 0.11995094269514084,
      "learning_rate": 1.4625991037573252e-05,
      "loss": 0.9238,
      "mean_token_accuracy": 0.7611361384391785,
      "num_tokens": 548770048.0,
      "step": 1560
    },
    {
      "entropy": 1.4578125,
      "epoch": 0.811789038262668,
      "grad_norm": 0.1158953309059143,
      "learning_rate": 1.4591520165460187e-05,
      "loss": 0.9055,
      "mean_token_accuracy": 0.7651141226291657,
      "num_tokens": 552280466.0,
      "step": 1570
    },
    {
      "entropy": 1.49375,
      "epoch": 0.8169596690796277,
      "grad_norm": 0.12239167839288712,
      "learning_rate": 1.4557049293347122e-05,
      "loss": 0.9674,
      "mean_token_accuracy": 0.7507982015609741,
      "num_tokens": 555747931.0,
      "step": 1580
    },
    {
      "entropy": 1.484375,
      "epoch": 0.8221302998965874,
      "grad_norm": 0.11154595762491226,
      "learning_rate": 1.4522578421234059e-05,
      "loss": 0.9261,
      "mean_token_accuracy": 0.7608684718608856,
      "num_tokens": 559290660.0,
      "step": 1590
    },
    {
      "entropy": 1.484375,
      "epoch": 0.827300930713547,
      "grad_norm": 0.1206263080239296,
      "learning_rate": 1.4488107549120994e-05,
      "loss": 0.9171,
      "mean_token_accuracy": 0.7622957825660706,
      "num_tokens": 562788799.0,
      "step": 1600
    },
    {
      "entropy": 1.4640625,
      "epoch": 0.8324715615305067,
      "grad_norm": 0.11691080778837204,
      "learning_rate": 1.4453636677007929e-05,
      "loss": 0.9425,
      "mean_token_accuracy": 0.7569314420223237,
      "num_tokens": 566371887.0,
      "step": 1610
    },
    {
      "entropy": 1.44921875,
      "epoch": 0.8376421923474664,
      "grad_norm": 0.11534351855516434,
      "learning_rate": 1.4419165804894866e-05,
      "loss": 0.9099,
      "mean_token_accuracy": 0.7634938955307007,
      "num_tokens": 569917270.0,
      "step": 1620
    },
    {
      "entropy": 1.46328125,
      "epoch": 0.8428128231644261,
      "grad_norm": 0.11727341264486313,
      "learning_rate": 1.4384694932781799e-05,
      "loss": 0.9037,
      "mean_token_accuracy": 0.7641095459461212,
      "num_tokens": 573393579.0,
      "step": 1630
    },
    {
      "entropy": 1.49375,
      "epoch": 0.8479834539813857,
      "grad_norm": 0.12199830263853073,
      "learning_rate": 1.4350224060668736e-05,
      "loss": 0.9202,
      "mean_token_accuracy": 0.7616257786750793,
      "num_tokens": 576885405.0,
      "step": 1640
    },
    {
      "entropy": 1.49296875,
      "epoch": 0.8531540847983454,
      "grad_norm": 0.11643374711275101,
      "learning_rate": 1.4315753188555672e-05,
      "loss": 0.8988,
      "mean_token_accuracy": 0.7642439842224121,
      "num_tokens": 580339669.0,
      "step": 1650
    },
    {
      "entropy": 1.475,
      "epoch": 0.8583247156153051,
      "grad_norm": 0.1187024861574173,
      "learning_rate": 1.4281282316442606e-05,
      "loss": 0.9011,
      "mean_token_accuracy": 0.764795571565628,
      "num_tokens": 583862692.0,
      "step": 1660
    },
    {
      "entropy": 1.4828125,
      "epoch": 0.8634953464322648,
      "grad_norm": 0.11858861893415451,
      "learning_rate": 1.4246811444329543e-05,
      "loss": 0.9113,
      "mean_token_accuracy": 0.7625495612621307,
      "num_tokens": 587415782.0,
      "step": 1670
    },
    {
      "entropy": 1.4921875,
      "epoch": 0.8686659772492245,
      "grad_norm": 0.12404865026473999,
      "learning_rate": 1.421234057221648e-05,
      "loss": 0.9305,
      "mean_token_accuracy": 0.7590306222438812,
      "num_tokens": 590914299.0,
      "step": 1680
    },
    {
      "entropy": 1.46328125,
      "epoch": 0.8738366080661841,
      "grad_norm": 0.11309592425823212,
      "learning_rate": 1.4177869700103413e-05,
      "loss": 0.9215,
      "mean_token_accuracy": 0.7607256054878235,
      "num_tokens": 594392721.0,
      "step": 1690
    },
    {
      "entropy": 1.47734375,
      "epoch": 0.8790072388831437,
      "grad_norm": 0.11948959529399872,
      "learning_rate": 1.414339882799035e-05,
      "loss": 0.9219,
      "mean_token_accuracy": 0.7603579103946686,
      "num_tokens": 597925926.0,
      "step": 1700
    },
    {
      "entropy": 1.45078125,
      "epoch": 0.8841778697001034,
      "grad_norm": 0.12206869572401047,
      "learning_rate": 1.4108927955877284e-05,
      "loss": 0.9058,
      "mean_token_accuracy": 0.7641239047050477,
      "num_tokens": 601418970.0,
      "step": 1710
    },
    {
      "entropy": 1.4953125,
      "epoch": 0.889348500517063,
      "grad_norm": 0.11994688957929611,
      "learning_rate": 1.4074457083764221e-05,
      "loss": 0.9439,
      "mean_token_accuracy": 0.7552627563476563,
      "num_tokens": 604907270.0,
      "step": 1720
    },
    {
      "entropy": 1.45078125,
      "epoch": 0.8945191313340227,
      "grad_norm": 0.1137346401810646,
      "learning_rate": 1.4039986211651156e-05,
      "loss": 0.9044,
      "mean_token_accuracy": 0.7650101542472839,
      "num_tokens": 608417023.0,
      "step": 1730
    },
    {
      "entropy": 1.47734375,
      "epoch": 0.8996897621509824,
      "grad_norm": 0.1210690587759018,
      "learning_rate": 1.4005515339538091e-05,
      "loss": 0.9318,
      "mean_token_accuracy": 0.7580539882183075,
      "num_tokens": 611892606.0,
      "step": 1740
    },
    {
      "entropy": 1.4609375,
      "epoch": 0.9048603929679421,
      "grad_norm": 0.1245742216706276,
      "learning_rate": 1.3971044467425028e-05,
      "loss": 0.9071,
      "mean_token_accuracy": 0.7634753048419952,
      "num_tokens": 615479462.0,
      "step": 1750
    },
    {
      "entropy": 1.46953125,
      "epoch": 0.9100310237849017,
      "grad_norm": 0.1237822026014328,
      "learning_rate": 1.3936573595311961e-05,
      "loss": 0.9,
      "mean_token_accuracy": 0.7645568609237671,
      "num_tokens": 618981303.0,
      "step": 1760
    },
    {
      "entropy": 1.478125,
      "epoch": 0.9152016546018614,
      "grad_norm": 0.11876232922077179,
      "learning_rate": 1.3902102723198898e-05,
      "loss": 0.9173,
      "mean_token_accuracy": 0.762673944234848,
      "num_tokens": 622483548.0,
      "step": 1770
    },
    {
      "entropy": 1.4609375,
      "epoch": 0.9203722854188211,
      "grad_norm": 0.11670742928981781,
      "learning_rate": 1.3867631851085835e-05,
      "loss": 0.9214,
      "mean_token_accuracy": 0.7605323374271393,
      "num_tokens": 626030553.0,
      "step": 1780
    },
    {
      "entropy": 1.4578125,
      "epoch": 0.9255429162357808,
      "grad_norm": 0.11695141345262527,
      "learning_rate": 1.3833160978972768e-05,
      "loss": 0.9207,
      "mean_token_accuracy": 0.7607087075710297,
      "num_tokens": 629557901.0,
      "step": 1790
    },
    {
      "entropy": 1.45234375,
      "epoch": 0.9307135470527405,
      "grad_norm": 0.1188148483633995,
      "learning_rate": 1.3798690106859705e-05,
      "loss": 0.9227,
      "mean_token_accuracy": 0.7603236854076385,
      "num_tokens": 633058595.0,
      "step": 1800
    },
    {
      "entropy": 1.43671875,
      "epoch": 0.9358841778697001,
      "grad_norm": 0.11865752935409546,
      "learning_rate": 1.3764219234746641e-05,
      "loss": 0.8993,
      "mean_token_accuracy": 0.7662648856639862,
      "num_tokens": 636564235.0,
      "step": 1810
    },
    {
      "entropy": 1.4421875,
      "epoch": 0.9410548086866598,
      "grad_norm": 0.1295095682144165,
      "learning_rate": 1.3729748362633575e-05,
      "loss": 0.9281,
      "mean_token_accuracy": 0.7599412083625794,
      "num_tokens": 640078587.0,
      "step": 1820
    },
    {
      "entropy": 1.42265625,
      "epoch": 0.9462254395036195,
      "grad_norm": 0.11797858774662018,
      "learning_rate": 1.3695277490520511e-05,
      "loss": 0.9165,
      "mean_token_accuracy": 0.761432534456253,
      "num_tokens": 643624605.0,
      "step": 1830
    },
    {
      "entropy": 1.42578125,
      "epoch": 0.9513960703205792,
      "grad_norm": 0.12107110023498535,
      "learning_rate": 1.3660806618407448e-05,
      "loss": 0.9199,
      "mean_token_accuracy": 0.7606053829193116,
      "num_tokens": 647142310.0,
      "step": 1840
    },
    {
      "entropy": 1.42890625,
      "epoch": 0.9565667011375388,
      "grad_norm": 0.10939899832010269,
      "learning_rate": 1.3626335746294381e-05,
      "loss": 0.9083,
      "mean_token_accuracy": 0.7634673655033112,
      "num_tokens": 650597087.0,
      "step": 1850
    },
    {
      "entropy": 1.4453125,
      "epoch": 0.9617373319544984,
      "grad_norm": 0.12392441928386688,
      "learning_rate": 1.3591864874181318e-05,
      "loss": 0.9025,
      "mean_token_accuracy": 0.7637257993221283,
      "num_tokens": 654112458.0,
      "step": 1860
    },
    {
      "entropy": 1.47109375,
      "epoch": 0.9669079627714581,
      "grad_norm": 0.11592892557382584,
      "learning_rate": 1.3557394002068253e-05,
      "loss": 0.9189,
      "mean_token_accuracy": 0.7612792730331421,
      "num_tokens": 657622848.0,
      "step": 1870
    },
    {
      "entropy": 1.45,
      "epoch": 0.9720785935884177,
      "grad_norm": 0.11586037278175354,
      "learning_rate": 1.352292312995519e-05,
      "loss": 0.9115,
      "mean_token_accuracy": 0.7627564072608948,
      "num_tokens": 661133982.0,
      "step": 1880
    },
    {
      "entropy": 1.4546875,
      "epoch": 0.9772492244053774,
      "grad_norm": 0.12101200968027115,
      "learning_rate": 1.3488452257842123e-05,
      "loss": 0.9049,
      "mean_token_accuracy": 0.7641494333744049,
      "num_tokens": 664651190.0,
      "step": 1890
    },
    {
      "entropy": 1.4359375,
      "epoch": 0.9824198552223371,
      "grad_norm": 0.1195937767624855,
      "learning_rate": 1.345398138572906e-05,
      "loss": 0.9232,
      "mean_token_accuracy": 0.7607501983642578,
      "num_tokens": 668185502.0,
      "step": 1900
    },
    {
      "entropy": 1.43515625,
      "epoch": 0.9875904860392968,
      "grad_norm": 0.13262051343917847,
      "learning_rate": 1.3419510513615997e-05,
      "loss": 0.911,
      "mean_token_accuracy": 0.7632193565368652,
      "num_tokens": 671716553.0,
      "step": 1910
    },
    {
      "entropy": 1.43203125,
      "epoch": 0.9927611168562565,
      "grad_norm": 0.13163335621356964,
      "learning_rate": 1.338503964150293e-05,
      "loss": 0.9053,
      "mean_token_accuracy": 0.7645232558250428,
      "num_tokens": 675258147.0,
      "step": 1920
    },
    {
      "entropy": 1.45625,
      "epoch": 0.9979317476732161,
      "grad_norm": 0.12155997008085251,
      "learning_rate": 1.3350568769389867e-05,
      "loss": 0.9054,
      "mean_token_accuracy": 0.7639196395874024,
      "num_tokens": 678787268.0,
      "step": 1930
    },
    {
      "entropy": 1.41796875,
      "epoch": 1.0031023784901758,
      "grad_norm": 0.12953729927539825,
      "learning_rate": 1.3316097897276803e-05,
      "loss": 0.9077,
      "mean_token_accuracy": 0.7628555178642273,
      "num_tokens": 682352660.0,
      "step": 1940
    },
    {
      "entropy": 1.35234375,
      "epoch": 1.0082730093071355,
      "grad_norm": 0.13617122173309326,
      "learning_rate": 1.3281627025163737e-05,
      "loss": 0.8488,
      "mean_token_accuracy": 0.7747729301452637,
      "num_tokens": 685885551.0,
      "step": 1950
    },
    {
      "entropy": 1.38359375,
      "epoch": 1.0134436401240952,
      "grad_norm": 0.12055247277021408,
      "learning_rate": 1.3247156153050673e-05,
      "loss": 0.8776,
      "mean_token_accuracy": 0.770104318857193,
      "num_tokens": 689438133.0,
      "step": 1960
    },
    {
      "entropy": 1.371875,
      "epoch": 1.0186142709410548,
      "grad_norm": 0.12040404230356216,
      "learning_rate": 1.321268528093761e-05,
      "loss": 0.8446,
      "mean_token_accuracy": 0.7767514526844025,
      "num_tokens": 692951376.0,
      "step": 1970
    },
    {
      "entropy": 1.37578125,
      "epoch": 1.0237849017580145,
      "grad_norm": 0.11971385031938553,
      "learning_rate": 1.3178214408824543e-05,
      "loss": 0.8555,
      "mean_token_accuracy": 0.7742941915988922,
      "num_tokens": 696477318.0,
      "step": 1980
    },
    {
      "entropy": 1.38828125,
      "epoch": 1.0289555325749742,
      "grad_norm": 0.11191146075725555,
      "learning_rate": 1.314374353671148e-05,
      "loss": 0.8586,
      "mean_token_accuracy": 0.7740735948085785,
      "num_tokens": 699984789.0,
      "step": 1990
    },
    {
      "entropy": 1.37265625,
      "epoch": 1.0341261633919339,
      "grad_norm": 0.12285640090703964,
      "learning_rate": 1.3109272664598415e-05,
      "loss": 0.8513,
      "mean_token_accuracy": 0.7760735332965851,
      "num_tokens": 703494729.0,
      "step": 2000
    },
    {
      "entropy": 1.3921875,
      "epoch": 1.0392967942088935,
      "grad_norm": 0.1279694139957428,
      "learning_rate": 1.307480179248535e-05,
      "loss": 0.857,
      "mean_token_accuracy": 0.7742870032787323,
      "num_tokens": 707006914.0,
      "step": 2010
    },
    {
      "entropy": 1.38828125,
      "epoch": 1.0444674250258532,
      "grad_norm": 0.1300376057624817,
      "learning_rate": 1.3040330920372287e-05,
      "loss": 0.8541,
      "mean_token_accuracy": 0.7744471371173859,
      "num_tokens": 710573396.0,
      "step": 2020
    },
    {
      "entropy": 1.375,
      "epoch": 1.049638055842813,
      "grad_norm": 0.12188553810119629,
      "learning_rate": 1.3005860048259222e-05,
      "loss": 0.8461,
      "mean_token_accuracy": 0.7757511854171752,
      "num_tokens": 714162377.0,
      "step": 2030
    },
    {
      "entropy": 1.36875,
      "epoch": 1.0548086866597726,
      "grad_norm": 0.12129724025726318,
      "learning_rate": 1.2971389176146159e-05,
      "loss": 0.8318,
      "mean_token_accuracy": 0.7799286842346191,
      "num_tokens": 717683573.0,
      "step": 2040
    },
    {
      "entropy": 1.37265625,
      "epoch": 1.0599793174767322,
      "grad_norm": 0.11378098279237747,
      "learning_rate": 1.2936918304033092e-05,
      "loss": 0.8475,
      "mean_token_accuracy": 0.7767247796058655,
      "num_tokens": 721227222.0,
      "step": 2050
    },
    {
      "entropy": 1.38359375,
      "epoch": 1.065149948293692,
      "grad_norm": 0.1249590516090393,
      "learning_rate": 1.2902447431920029e-05,
      "loss": 0.8305,
      "mean_token_accuracy": 0.7789794862270355,
      "num_tokens": 724706080.0,
      "step": 2060
    },
    {
      "entropy": 1.3953125,
      "epoch": 1.0703205791106516,
      "grad_norm": 0.12428700923919678,
      "learning_rate": 1.2867976559806965e-05,
      "loss": 0.8666,
      "mean_token_accuracy": 0.7721951365470886,
      "num_tokens": 728199061.0,
      "step": 2070
    },
    {
      "entropy": 1.37265625,
      "epoch": 1.0754912099276113,
      "grad_norm": 0.12617260217666626,
      "learning_rate": 1.2833505687693899e-05,
      "loss": 0.8381,
      "mean_token_accuracy": 0.7785492360591888,
      "num_tokens": 731689061.0,
      "step": 2080
    },
    {
      "entropy": 1.384375,
      "epoch": 1.080661840744571,
      "grad_norm": 0.11602634191513062,
      "learning_rate": 1.2799034815580835e-05,
      "loss": 0.8618,
      "mean_token_accuracy": 0.7727059543132782,
      "num_tokens": 735212770.0,
      "step": 2090
    },
    {
      "entropy": 1.36796875,
      "epoch": 1.0858324715615306,
      "grad_norm": 0.11466547101736069,
      "learning_rate": 1.2764563943467772e-05,
      "loss": 0.8158,
      "mean_token_accuracy": 0.7820233702659607,
      "num_tokens": 738666342.0,
      "step": 2100
    }
  ],
  "logging_steps": 10,
  "max_steps": 5802,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 2100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8591552888100094e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}