{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.4112416096031666,
      "epoch": 0.009411764705882352,
      "grad_norm": 1.8359375,
      "learning_rate": 3.551401869158878e-05,
      "loss": 3.197305107116699,
      "mean_token_accuracy": 0.5108491905033589,
      "num_tokens": 54223.0,
      "step": 20
    },
    {
      "entropy": 1.4180317347869278,
      "epoch": 0.018823529411764704,
      "grad_norm": 1.1484375,
      "learning_rate": 7.289719626168224e-05,
      "loss": 1.3110086441040039,
      "mean_token_accuracy": 0.7237571679055691,
      "num_tokens": 107652.0,
      "step": 40
    },
    {
      "entropy": 0.3327511426061392,
      "epoch": 0.02823529411764706,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0001102803738317757,
      "loss": 0.30270068645477294,
      "mean_token_accuracy": 0.9089575450867414,
      "num_tokens": 161942.0,
      "step": 60
    },
    {
      "entropy": 0.15523856291547417,
      "epoch": 0.03764705882352941,
      "grad_norm": 0.419921875,
      "learning_rate": 0.00014766355140186915,
      "loss": 0.15582343339920043,
      "mean_token_accuracy": 0.9389361537992954,
      "num_tokens": 216289.0,
      "step": 80
    },
    {
      "entropy": 0.13305185977369546,
      "epoch": 0.047058823529411764,
      "grad_norm": 0.287109375,
      "learning_rate": 0.00018504672897196262,
      "loss": 0.13600366115570067,
      "mean_token_accuracy": 0.9443846270442009,
      "num_tokens": 269496.0,
      "step": 100
    },
    {
      "entropy": 0.13116289507597684,
      "epoch": 0.05647058823529412,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00019881070366699704,
      "loss": 0.13201957941055298,
      "mean_token_accuracy": 0.9454528078436851,
      "num_tokens": 324096.0,
      "step": 120
    },
    {
      "entropy": 0.1289165782276541,
      "epoch": 0.06588235294117648,
      "grad_norm": 0.22265625,
      "learning_rate": 0.00019682854311199209,
      "loss": 0.1311253786087036,
      "mean_token_accuracy": 0.9442193854600192,
      "num_tokens": 378175.0,
      "step": 140
    },
    {
      "entropy": 0.126811383664608,
      "epoch": 0.07529411764705882,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.00019484638255698713,
      "loss": 0.1268758773803711,
      "mean_token_accuracy": 0.9458435796201229,
      "num_tokens": 431977.0,
      "step": 160
    },
    {
      "entropy": 0.12122174641117453,
      "epoch": 0.08470588235294117,
      "grad_norm": 0.490234375,
      "learning_rate": 0.00019286422200198217,
      "loss": 0.12409029006958008,
      "mean_token_accuracy": 0.947612290084362,
      "num_tokens": 485792.0,
      "step": 180
    },
    {
      "entropy": 0.1259708887897432,
      "epoch": 0.09411764705882353,
      "grad_norm": 0.1796875,
      "learning_rate": 0.00019088206144697722,
      "loss": 0.12703568935394288,
      "mean_token_accuracy": 0.9452463738620281,
      "num_tokens": 539402.0,
      "step": 200
    },
    {
      "entropy": 0.12701054341159762,
      "epoch": 0.10352941176470588,
      "grad_norm": 0.1875,
      "learning_rate": 0.00018889990089197224,
      "loss": 0.12911847829818726,
      "mean_token_accuracy": 0.9460463833063841,
      "num_tokens": 592534.0,
      "step": 220
    },
    {
      "entropy": 0.13070587934926153,
      "epoch": 0.11294117647058824,
      "grad_norm": 0.142578125,
      "learning_rate": 0.0001869177403369673,
      "loss": 0.12991806268692016,
      "mean_token_accuracy": 0.9448980402201415,
      "num_tokens": 646540.0,
      "step": 240
    },
    {
      "entropy": 0.12293657497502863,
      "epoch": 0.1223529411764706,
      "grad_norm": 0.125,
      "learning_rate": 0.00018493557978196235,
      "loss": 0.12530051469802855,
      "mean_token_accuracy": 0.9467378713190555,
      "num_tokens": 699482.0,
      "step": 260
    },
    {
      "entropy": 0.12243955316953362,
      "epoch": 0.13176470588235295,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.0001829534192269574,
      "loss": 0.12428406476974488,
      "mean_token_accuracy": 0.9461488857865333,
      "num_tokens": 753538.0,
      "step": 280
    },
    {
      "entropy": 0.12547048437409103,
      "epoch": 0.1411764705882353,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.00018097125867195244,
      "loss": 0.1255722761154175,
      "mean_token_accuracy": 0.9452667482197284,
      "num_tokens": 807637.0,
      "step": 300
    },
    {
      "entropy": 0.12290935553610324,
      "epoch": 0.15058823529411763,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.00017898909811694749,
      "loss": 0.12427603006362915,
      "mean_token_accuracy": 0.9455865450203419,
      "num_tokens": 861500.0,
      "step": 320
    },
    {
      "entropy": 0.12255495986901224,
      "epoch": 0.16,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.00017700693756194253,
      "loss": 0.12343236207962036,
      "mean_token_accuracy": 0.9462181635200977,
      "num_tokens": 915685.0,
      "step": 340
    },
    {
      "entropy": 0.12301232716999948,
      "epoch": 0.16941176470588235,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.00017502477700693757,
      "loss": 0.12390052080154419,
      "mean_token_accuracy": 0.9456650875508785,
      "num_tokens": 969319.0,
      "step": 360
    },
    {
      "entropy": 0.11796908043324947,
      "epoch": 0.17882352941176471,
      "grad_norm": 0.111328125,
      "learning_rate": 0.0001730426164519326,
      "loss": 0.11939283609390258,
      "mean_token_accuracy": 0.948033981397748,
      "num_tokens": 1022179.0,
      "step": 380
    },
    {
      "entropy": 0.1230991912074387,
      "epoch": 0.18823529411764706,
      "grad_norm": 0.095703125,
      "learning_rate": 0.00017106045589692766,
      "loss": 0.12383708953857422,
      "mean_token_accuracy": 0.9462584905326367,
      "num_tokens": 1075540.0,
      "step": 400
    },
    {
      "entropy": 0.11976800877600909,
      "epoch": 0.1976470588235294,
      "grad_norm": 0.09228515625,
      "learning_rate": 0.0001690782953419227,
      "loss": 0.12105886936187744,
      "mean_token_accuracy": 0.947526141256094,
      "num_tokens": 1128777.0,
      "step": 420
    },
    {
      "entropy": 0.120988504961133,
      "epoch": 0.20705882352941177,
      "grad_norm": 0.08203125,
      "learning_rate": 0.00016709613478691775,
      "loss": 0.12126818895339966,
      "mean_token_accuracy": 0.9466797344386577,
      "num_tokens": 1183006.0,
      "step": 440
    },
    {
      "entropy": 0.11960811992175877,
      "epoch": 0.2164705882352941,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.0001651139742319128,
      "loss": 0.12003393173217773,
      "mean_token_accuracy": 0.9471524652093649,
      "num_tokens": 1237005.0,
      "step": 460
    },
    {
      "entropy": 0.12131676562130451,
      "epoch": 0.22588235294117648,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.00016313181367690784,
      "loss": 0.12350617647171021,
      "mean_token_accuracy": 0.9453111827373505,
      "num_tokens": 1290676.0,
      "step": 480
    },
    {
      "entropy": 0.1230114926584065,
      "epoch": 0.23529411764705882,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.00016114965312190289,
      "loss": 0.12403985261917114,
      "mean_token_accuracy": 0.945762087777257,
      "num_tokens": 1344203.0,
      "step": 500
    },
    {
      "entropy": 0.11878803265281021,
      "epoch": 0.2447058823529412,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.00015916749256689793,
      "loss": 0.12019540071487426,
      "mean_token_accuracy": 0.9473085217177868,
      "num_tokens": 1397896.0,
      "step": 520
    },
    {
      "entropy": 0.1214360436424613,
      "epoch": 0.2541176470588235,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.00015718533201189295,
      "loss": 0.12231371402740479,
      "mean_token_accuracy": 0.9460072103887797,
      "num_tokens": 1451050.0,
      "step": 540
    },
    {
      "entropy": 0.1186625265982002,
      "epoch": 0.2635294117647059,
      "grad_norm": 0.1083984375,
      "learning_rate": 0.00015520317145688802,
      "loss": 0.11994267702102661,
      "mean_token_accuracy": 0.9478032253682613,
      "num_tokens": 1504087.0,
      "step": 560
    },
    {
      "entropy": 0.12019197628833353,
      "epoch": 0.27294117647058824,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.00015322101090188306,
      "loss": 0.12003889083862304,
      "mean_token_accuracy": 0.9468427047133445,
      "num_tokens": 1557793.0,
      "step": 580
    },
    {
      "entropy": 0.12021019305102527,
      "epoch": 0.2823529411764706,
      "grad_norm": 0.14453125,
      "learning_rate": 0.0001512388503468781,
      "loss": 0.12203142642974854,
      "mean_token_accuracy": 0.9466403890401125,
      "num_tokens": 1611761.0,
      "step": 600
    },
    {
      "entropy": 0.1219343897420913,
      "epoch": 0.2917647058823529,
      "grad_norm": 0.095703125,
      "learning_rate": 0.00014925668979187315,
      "loss": 0.12166310548782348,
      "mean_token_accuracy": 0.9456497117877006,
      "num_tokens": 1665978.0,
      "step": 620
    },
    {
      "entropy": 0.11833240487612784,
      "epoch": 0.30117647058823527,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0001472745292368682,
      "loss": 0.11984043121337891,
      "mean_token_accuracy": 0.9470193911343813,
      "num_tokens": 1719634.0,
      "step": 640
    },
    {
      "entropy": 0.1223891715053469,
      "epoch": 0.31058823529411766,
      "grad_norm": 0.0859375,
      "learning_rate": 0.00014529236868186324,
      "loss": 0.12163187265396118,
      "mean_token_accuracy": 0.945722034946084,
      "num_tokens": 1773965.0,
      "step": 660
    },
    {
      "entropy": 0.11958962972275913,
      "epoch": 0.32,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.00014331020812685829,
      "loss": 0.12138279676437377,
      "mean_token_accuracy": 0.9464583415538073,
      "num_tokens": 1828369.0,
      "step": 680
    },
    {
      "entropy": 0.11919600740075112,
      "epoch": 0.32941176470588235,
      "grad_norm": 0.1005859375,
      "learning_rate": 0.00014132804757185333,
      "loss": 0.12019137144088746,
      "mean_token_accuracy": 0.9472872346639634,
      "num_tokens": 1881385.0,
      "step": 700
    },
    {
      "entropy": 0.11798580284230412,
      "epoch": 0.3388235294117647,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.00013934588701684837,
      "loss": 0.11908473968505859,
      "mean_token_accuracy": 0.9474243365228177,
      "num_tokens": 1935579.0,
      "step": 720
    },
    {
      "entropy": 0.11969086015596986,
      "epoch": 0.34823529411764703,
      "grad_norm": 0.0830078125,
      "learning_rate": 0.00013736372646184342,
      "loss": 0.12085671424865722,
      "mean_token_accuracy": 0.945514016225934,
      "num_tokens": 1989952.0,
      "step": 740
    },
    {
      "entropy": 0.1185019179712981,
      "epoch": 0.35764705882352943,
      "grad_norm": 0.08544921875,
      "learning_rate": 0.00013538156590683846,
      "loss": 0.11856237649917603,
      "mean_token_accuracy": 0.946198745444417,
      "num_tokens": 2043741.0,
      "step": 760
    },
    {
      "entropy": 0.11969180204905569,
      "epoch": 0.36705882352941177,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.0001333994053518335,
      "loss": 0.12065744400024414,
      "mean_token_accuracy": 0.9467049967497587,
      "num_tokens": 2096959.0,
      "step": 780
    },
    {
      "entropy": 0.12028948199003935,
      "epoch": 0.3764705882352941,
      "grad_norm": 0.09375,
      "learning_rate": 0.00013141724479682855,
      "loss": 0.11995272636413574,
      "mean_token_accuracy": 0.9472199380397797,
      "num_tokens": 2150277.0,
      "step": 800
    },
    {
      "entropy": 0.12013958040624857,
      "epoch": 0.38588235294117645,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.0001294350842418236,
      "loss": 0.12228761911392212,
      "mean_token_accuracy": 0.9452053606510162,
      "num_tokens": 2205079.0,
      "step": 820
    },
    {
      "entropy": 0.12084516463801265,
      "epoch": 0.3952941176470588,
      "grad_norm": 0.08984375,
      "learning_rate": 0.00012745292368681864,
      "loss": 0.12086043357849122,
      "mean_token_accuracy": 0.9460023086518049,
      "num_tokens": 2259132.0,
      "step": 840
    },
    {
      "entropy": 0.1185136326123029,
      "epoch": 0.4047058823529412,
      "grad_norm": 0.0927734375,
      "learning_rate": 0.00012547076313181369,
      "loss": 0.11896246671676636,
      "mean_token_accuracy": 0.947355293482542,
      "num_tokens": 2312926.0,
      "step": 860
    },
    {
      "entropy": 0.12052565137855709,
      "epoch": 0.41411764705882353,
      "grad_norm": 0.07275390625,
      "learning_rate": 0.00012348860257680873,
      "loss": 0.12106109857559204,
      "mean_token_accuracy": 0.9457735672593117,
      "num_tokens": 2367484.0,
      "step": 880
    },
    {
      "entropy": 0.11853996482677757,
      "epoch": 0.4235294117647059,
      "grad_norm": 0.08251953125,
      "learning_rate": 0.00012150644202180377,
      "loss": 0.11953499317169189,
      "mean_token_accuracy": 0.9477898344397545,
      "num_tokens": 2421267.0,
      "step": 900
    },
    {
      "entropy": 0.11955512529239058,
      "epoch": 0.4329411764705882,
      "grad_norm": 0.07373046875,
      "learning_rate": 0.00011952428146679882,
      "loss": 0.12004319429397584,
      "mean_token_accuracy": 0.9473437976092101,
      "num_tokens": 2474674.0,
      "step": 920
    },
    {
      "entropy": 0.12128720586188138,
      "epoch": 0.4423529411764706,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.00011754212091179386,
      "loss": 0.12165406942367554,
      "mean_token_accuracy": 0.946265934035182,
      "num_tokens": 2527789.0,
      "step": 940
    },
    {
      "entropy": 0.1206457834225148,
      "epoch": 0.45176470588235296,
      "grad_norm": 0.07666015625,
      "learning_rate": 0.00011555996035678891,
      "loss": 0.12144509553909302,
      "mean_token_accuracy": 0.9454990841448307,
      "num_tokens": 2581246.0,
      "step": 960
    },
    {
      "entropy": 0.1177017642185092,
      "epoch": 0.4611764705882353,
      "grad_norm": 0.06884765625,
      "learning_rate": 0.00011357779980178395,
      "loss": 0.11779636144638062,
      "mean_token_accuracy": 0.9478017807006835,
      "num_tokens": 2634788.0,
      "step": 980
    },
    {
      "entropy": 0.11786414836533368,
      "epoch": 0.47058823529411764,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.00011159563924677901,
      "loss": 0.11904054880142212,
      "mean_token_accuracy": 0.9468365252017975,
      "num_tokens": 2688397.0,
      "step": 1000
    },
    {
      "entropy": 0.11901227324269711,
      "epoch": 0.48,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.00010961347869177403,
      "loss": 0.11933200359344483,
      "mean_token_accuracy": 0.9470609456300736,
      "num_tokens": 2742281.0,
      "step": 1020
    },
    {
      "entropy": 0.11828972357325256,
      "epoch": 0.4894117647058824,
      "grad_norm": 0.09326171875,
      "learning_rate": 0.00010763131813676907,
      "loss": 0.1188921332359314,
      "mean_token_accuracy": 0.9475983269512653,
      "num_tokens": 2795688.0,
      "step": 1040
    },
    {
      "entropy": 0.12043144796043634,
      "epoch": 0.4988235294117647,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.00010564915758176413,
      "loss": 0.12094131708145142,
      "mean_token_accuracy": 0.9459604099392891,
      "num_tokens": 2849705.0,
      "step": 1060
    },
    {
      "entropy": 0.11978372959420085,
      "epoch": 0.508235294117647,
      "grad_norm": 0.0634765625,
      "learning_rate": 0.00010366699702675917,
      "loss": 0.1203683614730835,
      "mean_token_accuracy": 0.9462303299456835,
      "num_tokens": 2903205.0,
      "step": 1080
    },
    {
      "entropy": 0.11803631293587387,
      "epoch": 0.5176470588235295,
      "grad_norm": 0.072265625,
      "learning_rate": 0.00010168483647175422,
      "loss": 0.1185994267463684,
      "mean_token_accuracy": 0.94720942042768,
      "num_tokens": 2956612.0,
      "step": 1100
    },
    {
      "entropy": 0.11592688583768904,
      "epoch": 0.5270588235294118,
      "grad_norm": 0.07958984375,
      "learning_rate": 9.970267591674926e-05,
      "loss": 0.11661049127578735,
      "mean_token_accuracy": 0.9493322856724262,
      "num_tokens": 3010272.0,
      "step": 1120
    },
    {
      "entropy": 0.12087701242417097,
      "epoch": 0.5364705882352941,
      "grad_norm": 0.078125,
      "learning_rate": 9.772051536174431e-05,
      "loss": 0.12155802249908447,
      "mean_token_accuracy": 0.9461342889815569,
      "num_tokens": 3064001.0,
      "step": 1140
    },
    {
      "entropy": 0.11912537859752774,
      "epoch": 0.5458823529411765,
      "grad_norm": 0.0791015625,
      "learning_rate": 9.573835480673935e-05,
      "loss": 0.11921051740646363,
      "mean_token_accuracy": 0.9465634476393461,
      "num_tokens": 3117457.0,
      "step": 1160
    },
    {
      "entropy": 0.12001162189990282,
      "epoch": 0.5552941176470588,
      "grad_norm": 0.06884765625,
      "learning_rate": 9.37561942517344e-05,
      "loss": 0.12046235799789429,
      "mean_token_accuracy": 0.9466627705842257,
      "num_tokens": 3171279.0,
      "step": 1180
    },
    {
      "entropy": 0.11993242320604622,
      "epoch": 0.5647058823529412,
      "grad_norm": 0.06494140625,
      "learning_rate": 9.177403369672944e-05,
      "loss": 0.11984696388244628,
      "mean_token_accuracy": 0.9461807683110237,
      "num_tokens": 3224002.0,
      "step": 1200
    },
    {
      "entropy": 0.1174244137480855,
      "epoch": 0.5741176470588235,
      "grad_norm": 0.076171875,
      "learning_rate": 8.979187314172448e-05,
      "loss": 0.11823408603668213,
      "mean_token_accuracy": 0.9473186526447535,
      "num_tokens": 3278045.0,
      "step": 1220
    },
    {
      "entropy": 0.11788262957707048,
      "epoch": 0.5835294117647059,
      "grad_norm": 0.08447265625,
      "learning_rate": 8.780971258671953e-05,
      "loss": 0.1186334252357483,
      "mean_token_accuracy": 0.9472660254687071,
      "num_tokens": 3331971.0,
      "step": 1240
    },
    {
      "entropy": 0.11809731125831605,
      "epoch": 0.5929411764705882,
      "grad_norm": 0.07470703125,
      "learning_rate": 8.582755203171457e-05,
      "loss": 0.1175646185874939,
      "mean_token_accuracy": 0.9479816593229771,
      "num_tokens": 3386043.0,
      "step": 1260
    },
    {
      "entropy": 0.11705515258945524,
      "epoch": 0.6023529411764705,
      "grad_norm": 0.0732421875,
      "learning_rate": 8.384539147670962e-05,
      "loss": 0.11811616420745849,
      "mean_token_accuracy": 0.9478495504707098,
      "num_tokens": 3440265.0,
      "step": 1280
    },
    {
      "entropy": 0.11638234835118055,
      "epoch": 0.611764705882353,
      "grad_norm": 0.06298828125,
      "learning_rate": 8.186323092170466e-05,
      "loss": 0.11749498844146729,
      "mean_token_accuracy": 0.9476320147514343,
      "num_tokens": 3495078.0,
      "step": 1300
    },
    {
      "entropy": 0.11721732169389724,
      "epoch": 0.6211764705882353,
      "grad_norm": 0.0703125,
      "learning_rate": 7.988107036669971e-05,
      "loss": 0.11693710088729858,
      "mean_token_accuracy": 0.9483740273863077,
      "num_tokens": 3548416.0,
      "step": 1320
    },
    {
      "entropy": 0.1194566513877362,
      "epoch": 0.6305882352941177,
      "grad_norm": 0.06591796875,
      "learning_rate": 7.789890981169475e-05,
      "loss": 0.11988570690155029,
      "mean_token_accuracy": 0.9467007365077734,
      "num_tokens": 3602131.0,
      "step": 1340
    },
    {
      "entropy": 0.11895696413703263,
      "epoch": 0.64,
      "grad_norm": 0.08056640625,
      "learning_rate": 7.59167492566898e-05,
      "loss": 0.11897348165512085,
      "mean_token_accuracy": 0.9464051406830549,
      "num_tokens": 3656510.0,
      "step": 1360
    },
    {
      "entropy": 0.11814323519356548,
      "epoch": 0.6494117647058824,
      "grad_norm": 0.0751953125,
      "learning_rate": 7.393458870168484e-05,
      "loss": 0.11766921281814575,
      "mean_token_accuracy": 0.9478136103600263,
      "num_tokens": 3709279.0,
      "step": 1380
    },
    {
      "entropy": 0.11508021377958358,
      "epoch": 0.6588235294117647,
      "grad_norm": 0.07275390625,
      "learning_rate": 7.195242814667988e-05,
      "loss": 0.11595516204833985,
      "mean_token_accuracy": 0.948334090039134,
      "num_tokens": 3763113.0,
      "step": 1400
    },
    {
      "entropy": 0.11870876159518957,
      "epoch": 0.668235294117647,
      "grad_norm": 0.08642578125,
      "learning_rate": 6.997026759167493e-05,
      "loss": 0.11904337406158447,
      "mean_token_accuracy": 0.946418273076415,
      "num_tokens": 3816580.0,
      "step": 1420
    },
    {
      "entropy": 0.11650171880610287,
      "epoch": 0.6776470588235294,
      "grad_norm": 0.0771484375,
      "learning_rate": 6.798810703666997e-05,
      "loss": 0.11831188201904297,
      "mean_token_accuracy": 0.9470474667847156,
      "num_tokens": 3870864.0,
      "step": 1440
    },
    {
      "entropy": 0.1193418983835727,
      "epoch": 0.6870588235294117,
      "grad_norm": 0.07568359375,
      "learning_rate": 6.600594648166502e-05,
      "loss": 0.11884019374847413,
      "mean_token_accuracy": 0.9471879187971354,
      "num_tokens": 3923803.0,
      "step": 1460
    },
    {
      "entropy": 0.11842391341924667,
      "epoch": 0.6964705882352941,
      "grad_norm": 0.06591796875,
      "learning_rate": 6.402378592666006e-05,
      "loss": 0.1178665280342102,
      "mean_token_accuracy": 0.9475426636636257,
      "num_tokens": 3977822.0,
      "step": 1480
    },
    {
      "entropy": 0.11864821463823319,
      "epoch": 0.7058823529411765,
      "grad_norm": 0.06787109375,
      "learning_rate": 6.204162537165511e-05,
      "loss": 0.11885442733764648,
      "mean_token_accuracy": 0.9469472885131835,
      "num_tokens": 4032147.0,
      "step": 1500
    },
    {
      "entropy": 0.1202593996655196,
      "epoch": 0.7152941176470589,
      "grad_norm": 0.06982421875,
      "learning_rate": 6.0059464816650144e-05,
      "loss": 0.1202502727508545,
      "mean_token_accuracy": 0.9464103363454341,
      "num_tokens": 4085894.0,
      "step": 1520
    },
    {
      "entropy": 0.11869277339428663,
      "epoch": 0.7247058823529412,
      "grad_norm": 0.064453125,
      "learning_rate": 5.8077304261645196e-05,
      "loss": 0.11944924592971802,
      "mean_token_accuracy": 0.9473904862999916,
      "num_tokens": 4139875.0,
      "step": 1540
    },
    {
      "entropy": 0.11899546729400753,
      "epoch": 0.7341176470588235,
      "grad_norm": 0.06298828125,
      "learning_rate": 5.609514370664024e-05,
      "loss": 0.1177599310874939,
      "mean_token_accuracy": 0.9476923041045666,
      "num_tokens": 4194092.0,
      "step": 1560
    },
    {
      "entropy": 0.11850538682192564,
      "epoch": 0.7435294117647059,
      "grad_norm": 0.06689453125,
      "learning_rate": 5.411298315163529e-05,
      "loss": 0.11867293119430541,
      "mean_token_accuracy": 0.9467376712709665,
      "num_tokens": 4247890.0,
      "step": 1580
    },
    {
      "entropy": 0.11745293159037828,
      "epoch": 0.7529411764705882,
      "grad_norm": 0.08349609375,
      "learning_rate": 5.213082259663032e-05,
      "loss": 0.11821787357330323,
      "mean_token_accuracy": 0.9470843933522701,
      "num_tokens": 4301620.0,
      "step": 1600
    },
    {
      "entropy": 0.11770048486068845,
      "epoch": 0.7623529411764706,
      "grad_norm": 0.07080078125,
      "learning_rate": 5.014866204162537e-05,
      "loss": 0.11723434925079346,
      "mean_token_accuracy": 0.9475229471921921,
      "num_tokens": 4355527.0,
      "step": 1620
    },
    {
      "entropy": 0.11760439816862345,
      "epoch": 0.7717647058823529,
      "grad_norm": 0.0703125,
      "learning_rate": 4.816650148662042e-05,
      "loss": 0.11808444261550903,
      "mean_token_accuracy": 0.9468282666057348,
      "num_tokens": 4409174.0,
      "step": 1640
    },
    {
      "entropy": 0.11781213534995913,
      "epoch": 0.7811764705882352,
      "grad_norm": 0.06884765625,
      "learning_rate": 4.618434093161546e-05,
      "loss": 0.1175337553024292,
      "mean_token_accuracy": 0.9476906027644872,
      "num_tokens": 4464273.0,
      "step": 1660
    },
    {
      "entropy": 0.11711168768815697,
      "epoch": 0.7905882352941176,
      "grad_norm": 0.064453125,
      "learning_rate": 4.420218037661051e-05,
      "loss": 0.11734318733215332,
      "mean_token_accuracy": 0.9468550808727741,
      "num_tokens": 4519274.0,
      "step": 1680
    },
    {
      "entropy": 0.11835422322619707,
      "epoch": 0.8,
      "grad_norm": 0.07177734375,
      "learning_rate": 4.222001982160555e-05,
      "loss": 0.11830483675003052,
      "mean_token_accuracy": 0.9464673962444067,
      "num_tokens": 4572849.0,
      "step": 1700
    },
    {
      "entropy": 0.11916838847100734,
      "epoch": 0.8094117647058824,
      "grad_norm": 0.07177734375,
      "learning_rate": 4.0237859266600596e-05,
      "loss": 0.11912477016448975,
      "mean_token_accuracy": 0.9463759887963533,
      "num_tokens": 4628106.0,
      "step": 1720
    },
    {
      "entropy": 0.11923501612618566,
      "epoch": 0.8188235294117647,
      "grad_norm": 0.068359375,
      "learning_rate": 3.825569871159564e-05,
      "loss": 0.11939848661422729,
      "mean_token_accuracy": 0.9465419482439756,
      "num_tokens": 4681923.0,
      "step": 1740
    },
    {
      "entropy": 0.11746299676597119,
      "epoch": 0.8282352941176471,
      "grad_norm": 0.0634765625,
      "learning_rate": 3.6273538156590684e-05,
      "loss": 0.11812403202056884,
      "mean_token_accuracy": 0.946595299616456,
      "num_tokens": 4735264.0,
      "step": 1760
    },
    {
      "entropy": 0.1170548350084573,
      "epoch": 0.8376470588235294,
      "grad_norm": 0.08056640625,
      "learning_rate": 3.429137760158573e-05,
      "loss": 0.11677987575531006,
      "mean_token_accuracy": 0.9473019450902939,
      "num_tokens": 4788562.0,
      "step": 1780
    },
    {
      "entropy": 0.11945610237307847,
      "epoch": 0.8470588235294118,
      "grad_norm": 0.0732421875,
      "learning_rate": 3.230921704658077e-05,
      "loss": 0.11916428804397583,
      "mean_token_accuracy": 0.9478193432092666,
      "num_tokens": 4842266.0,
      "step": 1800
    },
    {
      "entropy": 0.11968207377940417,
      "epoch": 0.8564705882352941,
      "grad_norm": 0.07177734375,
      "learning_rate": 3.0327056491575818e-05,
      "loss": 0.11990871429443359,
      "mean_token_accuracy": 0.9458895120769739,
      "num_tokens": 4896850.0,
      "step": 1820
    },
    {
      "entropy": 0.1200746615882963,
      "epoch": 0.8658823529411764,
      "grad_norm": 0.07275390625,
      "learning_rate": 2.8344895936570865e-05,
      "loss": 0.11900430917739868,
      "mean_token_accuracy": 0.9468101300299168,
      "num_tokens": 4950134.0,
      "step": 1840
    },
    {
      "entropy": 0.11479769693687558,
      "epoch": 0.8752941176470588,
      "grad_norm": 0.07373046875,
      "learning_rate": 2.6362735381565906e-05,
      "loss": 0.11587966680526733,
      "mean_token_accuracy": 0.9476078327745199,
      "num_tokens": 5003404.0,
      "step": 1860
    },
    {
      "entropy": 0.11819958202540874,
      "epoch": 0.8847058823529412,
      "grad_norm": 0.07666015625,
      "learning_rate": 2.4380574826560954e-05,
      "loss": 0.11736975908279419,
      "mean_token_accuracy": 0.947489620372653,
      "num_tokens": 5057306.0,
      "step": 1880
    },
    {
      "entropy": 0.11933490885421634,
      "epoch": 0.8941176470588236,
      "grad_norm": 0.083984375,
      "learning_rate": 2.2398414271556e-05,
      "loss": 0.11951359510421752,
      "mean_token_accuracy": 0.9462826151400805,
      "num_tokens": 5111179.0,
      "step": 1900
    },
    {
      "entropy": 0.11859381324611604,
      "epoch": 0.9035294117647059,
      "grad_norm": 0.078125,
      "learning_rate": 2.0416253716551043e-05,
      "loss": 0.11757175922393799,
      "mean_token_accuracy": 0.9475957881659269,
      "num_tokens": 5165007.0,
      "step": 1920
    },
    {
      "entropy": 0.11773485708981753,
      "epoch": 0.9129411764705883,
      "grad_norm": 0.08203125,
      "learning_rate": 1.8434093161546088e-05,
      "loss": 0.11632401943206787,
      "mean_token_accuracy": 0.9477220837026834,
      "num_tokens": 5219022.0,
      "step": 1940
    },
    {
      "entropy": 0.11942147044464946,
      "epoch": 0.9223529411764706,
      "grad_norm": 0.0791015625,
      "learning_rate": 1.6451932606541132e-05,
      "loss": 0.1185724139213562,
      "mean_token_accuracy": 0.9460108429193497,
      "num_tokens": 5272951.0,
      "step": 1960
    },
    {
      "entropy": 0.11498618689365685,
      "epoch": 0.9317647058823529,
      "grad_norm": 0.072265625,
      "learning_rate": 1.4469772051536176e-05,
      "loss": 0.1158057689666748,
      "mean_token_accuracy": 0.9480354256927968,
      "num_tokens": 5326883.0,
      "step": 1980
    },
    {
      "entropy": 0.1181730064097792,
      "epoch": 0.9411764705882353,
      "grad_norm": 0.07958984375,
      "learning_rate": 1.2487611496531219e-05,
      "loss": 0.11795371770858765,
      "mean_token_accuracy": 0.946659629791975,
      "num_tokens": 5380844.0,
      "step": 2000
    },
    {
      "entropy": 0.11797150396741926,
      "epoch": 0.9505882352941176,
      "grad_norm": 0.061767578125,
      "learning_rate": 1.0505450941526264e-05,
      "loss": 0.11806412935256957,
      "mean_token_accuracy": 0.9476543284952641,
      "num_tokens": 5434293.0,
      "step": 2020
    },
    {
      "entropy": 0.12068004687316716,
      "epoch": 0.96,
      "grad_norm": 0.0751953125,
      "learning_rate": 8.523290386521308e-06,
      "loss": 0.12049158811569213,
      "mean_token_accuracy": 0.9461670182645321,
      "num_tokens": 5488955.0,
      "step": 2040
    },
    {
      "entropy": 0.12010791432112455,
      "epoch": 0.9694117647058823,
      "grad_norm": 0.06787109375,
      "learning_rate": 6.541129831516353e-06,
      "loss": 0.11887935400009156,
      "mean_token_accuracy": 0.948026841506362,
      "num_tokens": 5543086.0,
      "step": 2060
    },
    {
      "entropy": 0.11950830989517272,
      "epoch": 0.9788235294117648,
      "grad_norm": 0.06396484375,
      "learning_rate": 4.558969276511398e-06,
      "loss": 0.11832374334335327,
      "mean_token_accuracy": 0.9463718678802252,
      "num_tokens": 5597309.0,
      "step": 2080
    },
    {
      "entropy": 0.11795398364774883,
      "epoch": 0.9882352941176471,
      "grad_norm": 0.07080078125,
      "learning_rate": 2.576808721506442e-06,
      "loss": 0.11762050390243531,
      "mean_token_accuracy": 0.9462573926895856,
      "num_tokens": 5650961.0,
      "step": 2100
    },
    {
      "entropy": 0.11789034181274474,
      "epoch": 0.9976470588235294,
      "grad_norm": 0.0673828125,
      "learning_rate": 5.946481665014866e-07,
      "loss": 0.11846446990966797,
      "mean_token_accuracy": 0.9470111202448607,
      "num_tokens": 5704037.0,
      "step": 2120
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.11721415494382381,
      "eval_loss": 0.11758215725421906,
      "eval_mean_token_accuracy": 0.9473637981414795,
      "eval_num_tokens": 5717542.0,
      "eval_runtime": 192.9855,
      "eval_samples_per_second": 5.182,
      "eval_steps_per_second": 2.591,
      "step": 2125
    }
  ],
  "logging_steps": 20,
  "max_steps": 2125,
  "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": 1.3495406394916992e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}