{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.09475322451908141,
      "epoch": 0.008003201280512205,
      "grad_norm": 424.0,
      "learning_rate": 1.4516129032258066e-06,
      "loss": 7.240174865722656,
      "mean_token_accuracy": 0.5289772760123015,
      "num_tokens": 42414.0,
      "step": 10
    },
    {
      "entropy": 0.10299791819415986,
      "epoch": 0.01600640256102441,
      "grad_norm": 232.0,
      "learning_rate": 3.0645161290322584e-06,
      "loss": 5.961273574829102,
      "mean_token_accuracy": 0.6124999947845936,
      "num_tokens": 83232.0,
      "step": 20
    },
    {
      "entropy": 0.08830865409690887,
      "epoch": 0.024009603841536616,
      "grad_norm": 300.0,
      "learning_rate": 4.67741935483871e-06,
      "loss": 3.4538761138916017,
      "mean_token_accuracy": 0.6931818276643753,
      "num_tokens": 123546.0,
      "step": 30
    },
    {
      "entropy": 0.08543417068431154,
      "epoch": 0.03201280512204882,
      "grad_norm": 19.75,
      "learning_rate": 6.290322580645162e-06,
      "loss": 0.5229705810546875,
      "mean_token_accuracy": 0.8744318261742592,
      "num_tokens": 165215.0,
      "step": 40
    },
    {
      "entropy": 0.06759846829809249,
      "epoch": 0.040016006402561026,
      "grad_norm": 28.25,
      "learning_rate": 7.903225806451613e-06,
      "loss": 0.09152845144271851,
      "mean_token_accuracy": 0.9596590884029865,
      "num_tokens": 206670.0,
      "step": 50
    },
    {
      "entropy": 0.06551291118375957,
      "epoch": 0.04801920768307323,
      "grad_norm": 3.109375,
      "learning_rate": 9.516129032258065e-06,
      "loss": 0.06548311114311219,
      "mean_token_accuracy": 0.9602272696793079,
      "num_tokens": 248738.0,
      "step": 60
    },
    {
      "entropy": 0.05871303144376725,
      "epoch": 0.056022408963585436,
      "grad_norm": 4.90625,
      "learning_rate": 1.1129032258064516e-05,
      "loss": 0.05862548351287842,
      "mean_token_accuracy": 0.9624999977648259,
      "num_tokens": 289455.0,
      "step": 70
    },
    {
      "entropy": 0.05400295334402472,
      "epoch": 0.06402561024409764,
      "grad_norm": 2.6875,
      "learning_rate": 1.274193548387097e-05,
      "loss": 0.05906534790992737,
      "mean_token_accuracy": 0.9653409078717232,
      "num_tokens": 328191.0,
      "step": 80
    },
    {
      "entropy": 0.05965230115689337,
      "epoch": 0.07202881152460984,
      "grad_norm": 2.03125,
      "learning_rate": 1.4354838709677421e-05,
      "loss": 0.06336120367050171,
      "mean_token_accuracy": 0.9613636344671249,
      "num_tokens": 370714.0,
      "step": 90
    },
    {
      "entropy": 0.0571730526862666,
      "epoch": 0.08003201280512205,
      "grad_norm": 4.1875,
      "learning_rate": 1.596774193548387e-05,
      "loss": 0.06329689621925354,
      "mean_token_accuracy": 0.9653409034013748,
      "num_tokens": 411158.0,
      "step": 100
    },
    {
      "entropy": 0.05321204150095582,
      "epoch": 0.08803521408563425,
      "grad_norm": 3.40625,
      "learning_rate": 1.7580645161290325e-05,
      "loss": 0.05498598217964172,
      "mean_token_accuracy": 0.9681818149983883,
      "num_tokens": 451650.0,
      "step": 110
    },
    {
      "entropy": 0.049898089445196095,
      "epoch": 0.09603841536614646,
      "grad_norm": 2.359375,
      "learning_rate": 1.9193548387096777e-05,
      "loss": 0.05433200597763062,
      "mean_token_accuracy": 0.9721590876579285,
      "num_tokens": 491529.0,
      "step": 120
    },
    {
      "entropy": 0.05273421558085829,
      "epoch": 0.10404161664665866,
      "grad_norm": 4.21875,
      "learning_rate": 1.9911190053285968e-05,
      "loss": 0.052664899826049806,
      "mean_token_accuracy": 0.9704545423388481,
      "num_tokens": 534477.0,
      "step": 130
    },
    {
      "entropy": 0.048857488168869166,
      "epoch": 0.11204481792717087,
      "grad_norm": 3.125,
      "learning_rate": 1.9733570159857907e-05,
      "loss": 0.05310600996017456,
      "mean_token_accuracy": 0.9715909071266651,
      "num_tokens": 573912.0,
      "step": 140
    },
    {
      "entropy": 0.054796313343103975,
      "epoch": 0.12004801920768307,
      "grad_norm": 3.109375,
      "learning_rate": 1.9555950266429843e-05,
      "loss": 0.05551998615264893,
      "mean_token_accuracy": 0.9681818142533303,
      "num_tokens": 613967.0,
      "step": 150
    },
    {
      "entropy": 0.050939775854931216,
      "epoch": 0.12805122048819528,
      "grad_norm": 4.90625,
      "learning_rate": 1.9378330373001776e-05,
      "loss": 0.05050772428512573,
      "mean_token_accuracy": 0.969318176060915,
      "num_tokens": 654625.0,
      "step": 160
    },
    {
      "entropy": 0.047000630758702755,
      "epoch": 0.1360544217687075,
      "grad_norm": 2.484375,
      "learning_rate": 1.9200710479573715e-05,
      "loss": 0.05440411567687988,
      "mean_token_accuracy": 0.9687499947845936,
      "num_tokens": 695387.0,
      "step": 170
    },
    {
      "entropy": 0.05339475516229868,
      "epoch": 0.14405762304921968,
      "grad_norm": 2.21875,
      "learning_rate": 1.902309058614565e-05,
      "loss": 0.05330321192741394,
      "mean_token_accuracy": 0.9715909034013748,
      "num_tokens": 736229.0,
      "step": 180
    },
    {
      "entropy": 0.05177106715273112,
      "epoch": 0.1520608243297319,
      "grad_norm": 2.390625,
      "learning_rate": 1.8845470692717587e-05,
      "loss": 0.053306227922439574,
      "mean_token_accuracy": 0.9749999947845935,
      "num_tokens": 777298.0,
      "step": 190
    },
    {
      "entropy": 0.04644646589877084,
      "epoch": 0.1600640256102441,
      "grad_norm": 3.21875,
      "learning_rate": 1.8667850799289523e-05,
      "loss": 0.046160414814949036,
      "mean_token_accuracy": 0.9767045393586159,
      "num_tokens": 816992.0,
      "step": 200
    },
    {
      "entropy": 0.03921595757419709,
      "epoch": 0.16806722689075632,
      "grad_norm": 2.953125,
      "learning_rate": 1.8490230905861456e-05,
      "loss": 0.046266698837280275,
      "mean_token_accuracy": 0.9744318142533303,
      "num_tokens": 856414.0,
      "step": 210
    },
    {
      "entropy": 0.04731454871071037,
      "epoch": 0.1760704281712685,
      "grad_norm": 3.703125,
      "learning_rate": 1.8312611012433395e-05,
      "loss": 0.049581846594810484,
      "mean_token_accuracy": 0.9710227221250534,
      "num_tokens": 897293.0,
      "step": 220
    },
    {
      "entropy": 0.050612482032738625,
      "epoch": 0.1840736294517807,
      "grad_norm": 3.046875,
      "learning_rate": 1.813499111900533e-05,
      "loss": 0.056270813941955565,
      "mean_token_accuracy": 0.9653409041464329,
      "num_tokens": 937566.0,
      "step": 230
    },
    {
      "entropy": 0.053894570120610295,
      "epoch": 0.19207683073229292,
      "grad_norm": 1.75,
      "learning_rate": 1.7957371225577267e-05,
      "loss": 0.05222816467285156,
      "mean_token_accuracy": 0.9715909041464329,
      "num_tokens": 979785.0,
      "step": 240
    },
    {
      "entropy": 0.04845360988983884,
      "epoch": 0.2000800320128051,
      "grad_norm": 1.6171875,
      "learning_rate": 1.7779751332149203e-05,
      "loss": 0.053466469049453735,
      "mean_token_accuracy": 0.9670454487204552,
      "num_tokens": 1019640.0,
      "step": 250
    },
    {
      "entropy": 0.049828451115172354,
      "epoch": 0.20808323329331732,
      "grad_norm": 1.625,
      "learning_rate": 1.760213143872114e-05,
      "loss": 0.05171402096748352,
      "mean_token_accuracy": 0.9693181782960891,
      "num_tokens": 1060823.0,
      "step": 260
    },
    {
      "entropy": 0.04814320435980335,
      "epoch": 0.21608643457382953,
      "grad_norm": 3.203125,
      "learning_rate": 1.7424511545293075e-05,
      "loss": 0.048413312435150145,
      "mean_token_accuracy": 0.9801136322319508,
      "num_tokens": 1102242.0,
      "step": 270
    },
    {
      "entropy": 0.045647610700689255,
      "epoch": 0.22408963585434175,
      "grad_norm": 2.484375,
      "learning_rate": 1.724689165186501e-05,
      "loss": 0.047308555245399474,
      "mean_token_accuracy": 0.9738636329770088,
      "num_tokens": 1141927.0,
      "step": 280
    },
    {
      "entropy": 0.04609768155496567,
      "epoch": 0.23209283713485393,
      "grad_norm": 2.40625,
      "learning_rate": 1.7069271758436947e-05,
      "loss": 0.05692839026451111,
      "mean_token_accuracy": 0.9681818142533303,
      "num_tokens": 1181811.0,
      "step": 290
    },
    {
      "entropy": 0.05474342008819803,
      "epoch": 0.24009603841536614,
      "grad_norm": 2.015625,
      "learning_rate": 1.6891651865008883e-05,
      "loss": 0.054971718788146974,
      "mean_token_accuracy": 0.9653409063816071,
      "num_tokens": 1223060.0,
      "step": 300
    },
    {
      "entropy": 0.05363150021876208,
      "epoch": 0.24809923969587835,
      "grad_norm": 1.6171875,
      "learning_rate": 1.671403197158082e-05,
      "loss": 0.05203201770782471,
      "mean_token_accuracy": 0.9693181775510311,
      "num_tokens": 1264243.0,
      "step": 310
    },
    {
      "entropy": 0.04393154126009904,
      "epoch": 0.25610244097639057,
      "grad_norm": 2.640625,
      "learning_rate": 1.6536412078152755e-05,
      "loss": 0.05061659216880798,
      "mean_token_accuracy": 0.9704545415937901,
      "num_tokens": 1304761.0,
      "step": 320
    },
    {
      "entropy": 0.0477942802477628,
      "epoch": 0.26410564225690275,
      "grad_norm": 2.75,
      "learning_rate": 1.635879218472469e-05,
      "loss": 0.04757603406906128,
      "mean_token_accuracy": 0.9738636314868927,
      "num_tokens": 1344735.0,
      "step": 330
    },
    {
      "entropy": 0.04573209672234953,
      "epoch": 0.272108843537415,
      "grad_norm": 2.453125,
      "learning_rate": 1.6181172291296626e-05,
      "loss": 0.04777190387248993,
      "mean_token_accuracy": 0.9744318120181561,
      "num_tokens": 1385052.0,
      "step": 340
    },
    {
      "entropy": 0.04807677125791088,
      "epoch": 0.2801120448179272,
      "grad_norm": 2.109375,
      "learning_rate": 1.6003552397868562e-05,
      "loss": 0.049048814177513125,
      "mean_token_accuracy": 0.9693181782960891,
      "num_tokens": 1425851.0,
      "step": 350
    },
    {
      "entropy": 0.04814378765877336,
      "epoch": 0.28811524609843936,
      "grad_norm": 2.109375,
      "learning_rate": 1.58259325044405e-05,
      "loss": 0.04974732995033264,
      "mean_token_accuracy": 0.9698863595724105,
      "num_tokens": 1466568.0,
      "step": 360
    },
    {
      "entropy": 0.049266617721877994,
      "epoch": 0.2961184473789516,
      "grad_norm": 2.296875,
      "learning_rate": 1.5648312611012434e-05,
      "loss": 0.052187031507492064,
      "mean_token_accuracy": 0.9732954487204551,
      "num_tokens": 1507415.0,
      "step": 370
    },
    {
      "entropy": 0.04980740246828645,
      "epoch": 0.3041216486594638,
      "grad_norm": 1.953125,
      "learning_rate": 1.547069271758437e-05,
      "loss": 0.04835581183433533,
      "mean_token_accuracy": 0.9727272689342499,
      "num_tokens": 1548371.0,
      "step": 380
    },
    {
      "entropy": 0.03997177838464268,
      "epoch": 0.31212484993997597,
      "grad_norm": 4.0,
      "learning_rate": 1.5293072824156306e-05,
      "loss": 0.04371632635593414,
      "mean_token_accuracy": 0.9801136322319508,
      "num_tokens": 1590062.0,
      "step": 390
    },
    {
      "entropy": 0.03653254228320293,
      "epoch": 0.3201280512204882,
      "grad_norm": 2.859375,
      "learning_rate": 1.5115452930728242e-05,
      "loss": 0.05101522207260132,
      "mean_token_accuracy": 0.9715909041464329,
      "num_tokens": 1632136.0,
      "step": 400
    },
    {
      "entropy": 0.0425920093410241,
      "epoch": 0.3281312525010004,
      "grad_norm": 2.015625,
      "learning_rate": 1.4937833037300178e-05,
      "loss": 0.04813604354858399,
      "mean_token_accuracy": 0.9744318127632141,
      "num_tokens": 1671462.0,
      "step": 410
    },
    {
      "entropy": 0.03970824745483696,
      "epoch": 0.33613445378151263,
      "grad_norm": 2.90625,
      "learning_rate": 1.4760213143872116e-05,
      "loss": 0.03939797580242157,
      "mean_token_accuracy": 0.9812499947845936,
      "num_tokens": 1712639.0,
      "step": 420
    },
    {
      "entropy": 0.042815997470461296,
      "epoch": 0.3441376550620248,
      "grad_norm": 1.453125,
      "learning_rate": 1.458259325044405e-05,
      "loss": 0.04754864275455475,
      "mean_token_accuracy": 0.9744318135082721,
      "num_tokens": 1753415.0,
      "step": 430
    },
    {
      "entropy": 0.0437370027997531,
      "epoch": 0.352140856342537,
      "grad_norm": 2.515625,
      "learning_rate": 1.4404973357015988e-05,
      "loss": 0.04742822349071503,
      "mean_token_accuracy": 0.9732954494655133,
      "num_tokens": 1792637.0,
      "step": 440
    },
    {
      "entropy": 0.04030659813288366,
      "epoch": 0.36014405762304924,
      "grad_norm": 3.796875,
      "learning_rate": 1.4227353463587924e-05,
      "loss": 0.04023580253124237,
      "mean_token_accuracy": 0.9801136322319508,
      "num_tokens": 1830990.0,
      "step": 450
    },
    {
      "entropy": 0.03681652669911273,
      "epoch": 0.3681472589035614,
      "grad_norm": 2.828125,
      "learning_rate": 1.4049733570159858e-05,
      "loss": 0.04969809353351593,
      "mean_token_accuracy": 0.9721590869128705,
      "num_tokens": 1870848.0,
      "step": 460
    },
    {
      "entropy": 0.04123187513323501,
      "epoch": 0.3761504601840736,
      "grad_norm": 2.21875,
      "learning_rate": 1.3872113676731796e-05,
      "loss": 0.043952861428260805,
      "mean_token_accuracy": 0.9761363573372364,
      "num_tokens": 1911638.0,
      "step": 470
    },
    {
      "entropy": 0.0450622973847203,
      "epoch": 0.38415366146458585,
      "grad_norm": 2.328125,
      "learning_rate": 1.369449378330373e-05,
      "loss": 0.045897161960601805,
      "mean_token_accuracy": 0.9738636322319507,
      "num_tokens": 1950406.0,
      "step": 480
    },
    {
      "entropy": 0.04427487466891762,
      "epoch": 0.39215686274509803,
      "grad_norm": 2.25,
      "learning_rate": 1.3516873889875667e-05,
      "loss": 0.04462568461894989,
      "mean_token_accuracy": 0.9732954502105713,
      "num_tokens": 1990786.0,
      "step": 490
    },
    {
      "entropy": 0.04660661657107994,
      "epoch": 0.4001600640256102,
      "grad_norm": 2.859375,
      "learning_rate": 1.3339253996447603e-05,
      "loss": 0.04628114402294159,
      "mean_token_accuracy": 0.9738636322319507,
      "num_tokens": 2030170.0,
      "step": 500
    },
    {
      "entropy": 0.04326281558896881,
      "epoch": 0.40816326530612246,
      "grad_norm": 2.171875,
      "learning_rate": 1.3161634103019538e-05,
      "loss": 0.05016361474990845,
      "mean_token_accuracy": 0.9738636314868927,
      "num_tokens": 2070211.0,
      "step": 510
    },
    {
      "entropy": 0.04552530206856318,
      "epoch": 0.41616646658663464,
      "grad_norm": 4.09375,
      "learning_rate": 1.2984014209591475e-05,
      "loss": 0.048873734474182126,
      "mean_token_accuracy": 0.9744318135082721,
      "num_tokens": 2110242.0,
      "step": 520
    },
    {
      "entropy": 0.048664613376604396,
      "epoch": 0.4241696678671469,
      "grad_norm": 5.96875,
      "learning_rate": 1.2806394316163411e-05,
      "loss": 0.051538127660751346,
      "mean_token_accuracy": 0.9738636314868927,
      "num_tokens": 2149695.0,
      "step": 530
    },
    {
      "entropy": 0.0454276799748186,
      "epoch": 0.43217286914765907,
      "grad_norm": 2.390625,
      "learning_rate": 1.2628774422735349e-05,
      "loss": 0.04872589409351349,
      "mean_token_accuracy": 0.9721590876579285,
      "num_tokens": 2191269.0,
      "step": 540
    },
    {
      "entropy": 0.04385605392162688,
      "epoch": 0.44017607042817125,
      "grad_norm": 3.546875,
      "learning_rate": 1.2451154529307283e-05,
      "loss": 0.049637827277183535,
      "mean_token_accuracy": 0.9755681768059731,
      "num_tokens": 2233167.0,
      "step": 550
    },
    {
      "entropy": 0.047120464290492235,
      "epoch": 0.4481792717086835,
      "grad_norm": 2.0625,
      "learning_rate": 1.227353463587922e-05,
      "loss": 0.047938373684883115,
      "mean_token_accuracy": 0.9715909048914909,
      "num_tokens": 2273683.0,
      "step": 560
    },
    {
      "entropy": 0.04529402129119262,
      "epoch": 0.4561824729891957,
      "grad_norm": 1.8671875,
      "learning_rate": 1.2095914742451155e-05,
      "loss": 0.0421001136302948,
      "mean_token_accuracy": 0.9789772681891918,
      "num_tokens": 2312821.0,
      "step": 570
    },
    {
      "entropy": 0.040198929510370365,
      "epoch": 0.46418567426970786,
      "grad_norm": 4.125,
      "learning_rate": 1.1918294849023091e-05,
      "loss": 0.045373284816741945,
      "mean_token_accuracy": 0.9772727213799953,
      "num_tokens": 2354833.0,
      "step": 580
    },
    {
      "entropy": 0.04357797284028493,
      "epoch": 0.4721888755502201,
      "grad_norm": 1.1875,
      "learning_rate": 1.1740674955595029e-05,
      "loss": 0.05674033761024475,
      "mean_token_accuracy": 0.9693181782960891,
      "num_tokens": 2395646.0,
      "step": 590
    },
    {
      "entropy": 0.04933104421943426,
      "epoch": 0.4801920768307323,
      "grad_norm": 0.9375,
      "learning_rate": 1.1563055062166963e-05,
      "loss": 0.052300095558166504,
      "mean_token_accuracy": 0.9698863595724105,
      "num_tokens": 2434038.0,
      "step": 600
    },
    {
      "entropy": 0.04834912738297135,
      "epoch": 0.4881952781112445,
      "grad_norm": 4.40625,
      "learning_rate": 1.1385435168738899e-05,
      "loss": 0.04378367662429809,
      "mean_token_accuracy": 0.977840905636549,
      "num_tokens": 2473548.0,
      "step": 610
    },
    {
      "entropy": 0.03797003252257127,
      "epoch": 0.4961984793917567,
      "grad_norm": 3.703125,
      "learning_rate": 1.1207815275310837e-05,
      "loss": 0.04411846697330475,
      "mean_token_accuracy": 0.9761363610625267,
      "num_tokens": 2513657.0,
      "step": 620
    },
    {
      "entropy": 0.037305073122115576,
      "epoch": 0.5042016806722689,
      "grad_norm": 2.265625,
      "learning_rate": 1.1030195381882771e-05,
      "loss": 0.0530417799949646,
      "mean_token_accuracy": 0.9772727228701115,
      "num_tokens": 2553049.0,
      "step": 630
    },
    {
      "entropy": 0.043729909125249836,
      "epoch": 0.5122048819527811,
      "grad_norm": 2.375,
      "learning_rate": 1.0852575488454709e-05,
      "loss": 0.05261260867118835,
      "mean_token_accuracy": 0.9698863580822945,
      "num_tokens": 2593026.0,
      "step": 640
    },
    {
      "entropy": 0.04822997532319277,
      "epoch": 0.5202080832332934,
      "grad_norm": 1.3828125,
      "learning_rate": 1.0674955595026645e-05,
      "loss": 0.05205248594284058,
      "mean_token_accuracy": 0.9727272666990757,
      "num_tokens": 2633344.0,
      "step": 650
    },
    {
      "entropy": 0.04994500142056495,
      "epoch": 0.5282112845138055,
      "grad_norm": 2.484375,
      "learning_rate": 1.0497335701598579e-05,
      "loss": 0.05130913853645325,
      "mean_token_accuracy": 0.9693181797862053,
      "num_tokens": 2672721.0,
      "step": 660
    },
    {
      "entropy": 0.04809544003510382,
      "epoch": 0.5362144857943177,
      "grad_norm": 1.7734375,
      "learning_rate": 1.0319715808170516e-05,
      "loss": 0.043288156390190125,
      "mean_token_accuracy": 0.9772727236151695,
      "num_tokens": 2714199.0,
      "step": 670
    },
    {
      "entropy": 0.041754112322814764,
      "epoch": 0.54421768707483,
      "grad_norm": 2.125,
      "learning_rate": 1.014209591474245e-05,
      "loss": 0.04941850304603577,
      "mean_token_accuracy": 0.9710227191448212,
      "num_tokens": 2757389.0,
      "step": 680
    },
    {
      "entropy": 0.04491965179331601,
      "epoch": 0.5522208883553421,
      "grad_norm": 2.421875,
      "learning_rate": 9.964476021314388e-06,
      "loss": 0.048098990321159364,
      "mean_token_accuracy": 0.975568176060915,
      "num_tokens": 2797266.0,
      "step": 690
    },
    {
      "entropy": 0.04177754216361791,
      "epoch": 0.5602240896358543,
      "grad_norm": 1.8828125,
      "learning_rate": 9.786856127886324e-06,
      "loss": 0.03804603815078735,
      "mean_token_accuracy": 0.9818181775510311,
      "num_tokens": 2837178.0,
      "step": 700
    },
    {
      "entropy": 0.04215870927291689,
      "epoch": 0.5682272909163666,
      "grad_norm": 3.109375,
      "learning_rate": 9.60923623445826e-06,
      "loss": 0.040096724033355714,
      "mean_token_accuracy": 0.9806818142533302,
      "num_tokens": 2876752.0,
      "step": 710
    },
    {
      "entropy": 0.040034703072160484,
      "epoch": 0.5762304921968787,
      "grad_norm": 2.5625,
      "learning_rate": 9.431616341030196e-06,
      "loss": 0.049453753232955935,
      "mean_token_accuracy": 0.97443181052804,
      "num_tokens": 2918121.0,
      "step": 720
    },
    {
      "entropy": 0.04085657097748481,
      "epoch": 0.584233693477391,
      "grad_norm": 3.109375,
      "learning_rate": 9.253996447602132e-06,
      "loss": 0.05398496389389038,
      "mean_token_accuracy": 0.9744318142533303,
      "num_tokens": 2957034.0,
      "step": 730
    },
    {
      "entropy": 0.045270191514282486,
      "epoch": 0.5922368947579032,
      "grad_norm": 2.203125,
      "learning_rate": 9.076376554174068e-06,
      "loss": 0.04788501560688019,
      "mean_token_accuracy": 0.9710227236151695,
      "num_tokens": 2996795.0,
      "step": 740
    },
    {
      "entropy": 0.046037832100410014,
      "epoch": 0.6002400960384153,
      "grad_norm": 2.171875,
      "learning_rate": 8.898756660746004e-06,
      "loss": 0.045071455836296084,
      "mean_token_accuracy": 0.9772727221250535,
      "num_tokens": 3036825.0,
      "step": 750
    },
    {
      "entropy": 0.04801415469264612,
      "epoch": 0.6082432973189276,
      "grad_norm": 2.34375,
      "learning_rate": 8.72113676731794e-06,
      "loss": 0.0479773223400116,
      "mean_token_accuracy": 0.9710227243602276,
      "num_tokens": 3076791.0,
      "step": 760
    },
    {
      "entropy": 0.0462384240119718,
      "epoch": 0.6162464985994398,
      "grad_norm": 2.453125,
      "learning_rate": 8.543516873889876e-06,
      "loss": 0.04638043344020844,
      "mean_token_accuracy": 0.9761363558471203,
      "num_tokens": 3120135.0,
      "step": 770
    },
    {
      "entropy": 0.045585069648222995,
      "epoch": 0.6242496998799519,
      "grad_norm": 2.15625,
      "learning_rate": 8.365896980461814e-06,
      "loss": 0.048996028304100034,
      "mean_token_accuracy": 0.9721590884029865,
      "num_tokens": 3161235.0,
      "step": 780
    },
    {
      "entropy": 0.04545295296120457,
      "epoch": 0.6322529011604642,
      "grad_norm": 3.234375,
      "learning_rate": 8.188277087033748e-06,
      "loss": 0.04392434060573578,
      "mean_token_accuracy": 0.9744318127632141,
      "num_tokens": 3200661.0,
      "step": 790
    },
    {
      "entropy": 0.0439292140188627,
      "epoch": 0.6402561024409764,
      "grad_norm": 5.03125,
      "learning_rate": 8.010657193605684e-06,
      "loss": 0.04489242136478424,
      "mean_token_accuracy": 0.9744318149983883,
      "num_tokens": 3239228.0,
      "step": 800
    },
    {
      "entropy": 0.04578416767762974,
      "epoch": 0.6482593037214885,
      "grad_norm": 3.515625,
      "learning_rate": 7.83303730017762e-06,
      "loss": 0.050331324338912964,
      "mean_token_accuracy": 0.9755681790411472,
      "num_tokens": 3278112.0,
      "step": 810
    },
    {
      "entropy": 0.04590382124297321,
      "epoch": 0.6562625050020008,
      "grad_norm": 1.4453125,
      "learning_rate": 7.655417406749557e-06,
      "loss": 0.048855268955230714,
      "mean_token_accuracy": 0.9761363595724106,
      "num_tokens": 3319328.0,
      "step": 820
    },
    {
      "entropy": 0.04355756618315354,
      "epoch": 0.664265706282513,
      "grad_norm": 2.75,
      "learning_rate": 7.4777975133214935e-06,
      "loss": 0.04029326438903809,
      "mean_token_accuracy": 0.9789772689342499,
      "num_tokens": 3359897.0,
      "step": 830
    },
    {
      "entropy": 0.04515375569462776,
      "epoch": 0.6722689075630253,
      "grad_norm": 2.75,
      "learning_rate": 7.3001776198934286e-06,
      "loss": 0.045958060026168826,
      "mean_token_accuracy": 0.9744318142533303,
      "num_tokens": 3400717.0,
      "step": 840
    },
    {
      "entropy": 0.041079553513554856,
      "epoch": 0.6802721088435374,
      "grad_norm": 2.03125,
      "learning_rate": 7.1225577264653645e-06,
      "loss": 0.04416506886482239,
      "mean_token_accuracy": 0.974431811273098,
      "num_tokens": 3439852.0,
      "step": 850
    },
    {
      "entropy": 0.043181278416886924,
      "epoch": 0.6882753101240496,
      "grad_norm": 3.03125,
      "learning_rate": 6.9449378330373005e-06,
      "loss": 0.04391254186630249,
      "mean_token_accuracy": 0.9789772689342499,
      "num_tokens": 3479654.0,
      "step": 860
    },
    {
      "entropy": 0.04434247007593513,
      "epoch": 0.6962785114045619,
      "grad_norm": 1.8046875,
      "learning_rate": 6.767317939609237e-06,
      "loss": 0.04593832492828369,
      "mean_token_accuracy": 0.9744318164885044,
      "num_tokens": 3520077.0,
      "step": 870
    },
    {
      "entropy": 0.04393934286781587,
      "epoch": 0.704281712685074,
      "grad_norm": 1.6640625,
      "learning_rate": 6.589698046181173e-06,
      "loss": 0.046162942051887514,
      "mean_token_accuracy": 0.977840905636549,
      "num_tokens": 3560952.0,
      "step": 880
    },
    {
      "entropy": 0.04364555573556572,
      "epoch": 0.7122849139655862,
      "grad_norm": 1.640625,
      "learning_rate": 6.412078152753108e-06,
      "loss": 0.0497223436832428,
      "mean_token_accuracy": 0.9744318142533303,
      "num_tokens": 3600585.0,
      "step": 890
    },
    {
      "entropy": 0.04233085287705762,
      "epoch": 0.7202881152460985,
      "grad_norm": 1.7578125,
      "learning_rate": 6.234458259325044e-06,
      "loss": 0.03990237712860108,
      "mean_token_accuracy": 0.9801136337220668,
      "num_tokens": 3643430.0,
      "step": 900
    },
    {
      "entropy": 0.04045454500592314,
      "epoch": 0.7282913165266106,
      "grad_norm": 3.21875,
      "learning_rate": 6.056838365896981e-06,
      "loss": 0.04266665279865265,
      "mean_token_accuracy": 0.9784090861678123,
      "num_tokens": 3684717.0,
      "step": 910
    },
    {
      "entropy": 0.041357538173906506,
      "epoch": 0.7362945178071229,
      "grad_norm": 2.390625,
      "learning_rate": 5.879218472468917e-06,
      "loss": 0.04428337216377258,
      "mean_token_accuracy": 0.9789772689342499,
      "num_tokens": 3724155.0,
      "step": 920
    },
    {
      "entropy": 0.04306626642246556,
      "epoch": 0.7442977190876351,
      "grad_norm": 1.75,
      "learning_rate": 5.701598579040853e-06,
      "loss": 0.048436781764030455,
      "mean_token_accuracy": 0.9749999940395355,
      "num_tokens": 3765538.0,
      "step": 930
    },
    {
      "entropy": 0.043767763464711604,
      "epoch": 0.7523009203681472,
      "grad_norm": 1.5234375,
      "learning_rate": 5.523978685612789e-06,
      "loss": 0.05207340121269226,
      "mean_token_accuracy": 0.9749999955296517,
      "num_tokens": 3805016.0,
      "step": 940
    },
    {
      "entropy": 0.04864149287022883,
      "epoch": 0.7603041216486595,
      "grad_norm": 3.34375,
      "learning_rate": 5.346358792184725e-06,
      "loss": 0.05412702560424805,
      "mean_token_accuracy": 0.9710227236151695,
      "num_tokens": 3844988.0,
      "step": 950
    },
    {
      "entropy": 0.045417454978451136,
      "epoch": 0.7683073229291717,
      "grad_norm": 2.109375,
      "learning_rate": 5.168738898756661e-06,
      "loss": 0.039266571402549744,
      "mean_token_accuracy": 0.9818181782960892,
      "num_tokens": 3884166.0,
      "step": 960
    },
    {
      "entropy": 0.047192305256612596,
      "epoch": 0.7763105242096838,
      "grad_norm": 4.125,
      "learning_rate": 4.991119005328598e-06,
      "loss": 0.0472522109746933,
      "mean_token_accuracy": 0.9778409041464329,
      "num_tokens": 3922599.0,
      "step": 970
    },
    {
      "entropy": 0.044715805305168034,
      "epoch": 0.7843137254901961,
      "grad_norm": 2.03125,
      "learning_rate": 4.813499111900533e-06,
      "loss": 0.043560588359832765,
      "mean_token_accuracy": 0.9727272681891919,
      "num_tokens": 3961577.0,
      "step": 980
    },
    {
      "entropy": 0.04465018446790055,
      "epoch": 0.7923169267707083,
      "grad_norm": 1.6171875,
      "learning_rate": 4.63587921847247e-06,
      "loss": 0.04594630300998688,
      "mean_token_accuracy": 0.9732954502105713,
      "num_tokens": 4001019.0,
      "step": 990
    },
    {
      "entropy": 0.04393655447638593,
      "epoch": 0.8003201280512204,
      "grad_norm": 1.7890625,
      "learning_rate": 4.458259325044405e-06,
      "loss": 0.04262282252311707,
      "mean_token_accuracy": 0.9778409048914909,
      "num_tokens": 4041721.0,
      "step": 1000
    },
    {
      "entropy": 0.042122538023249945,
      "epoch": 0.8083233293317327,
      "grad_norm": 1.8515625,
      "learning_rate": 4.2806394316163415e-06,
      "loss": 0.04339278638362885,
      "mean_token_accuracy": 0.9784090869128704,
      "num_tokens": 4082654.0,
      "step": 1010
    },
    {
      "entropy": 0.04523669999325648,
      "epoch": 0.8163265306122449,
      "grad_norm": 2.0625,
      "learning_rate": 4.1030195381882775e-06,
      "loss": 0.048701035976409915,
      "mean_token_accuracy": 0.9704545423388481,
      "num_tokens": 4123109.0,
      "step": 1020
    },
    {
      "entropy": 0.04366708064917475,
      "epoch": 0.8243297318927572,
      "grad_norm": 3.1875,
      "learning_rate": 3.9253996447602135e-06,
      "loss": 0.043869239091873166,
      "mean_token_accuracy": 0.9749999977648258,
      "num_tokens": 4162927.0,
      "step": 1030
    },
    {
      "entropy": 0.043506618012906985,
      "epoch": 0.8323329331732693,
      "grad_norm": 1.5859375,
      "learning_rate": 3.7477797513321494e-06,
      "loss": 0.04709920883178711,
      "mean_token_accuracy": 0.9772727221250535,
      "num_tokens": 4202732.0,
      "step": 1040
    },
    {
      "entropy": 0.04336636826628819,
      "epoch": 0.8403361344537815,
      "grad_norm": 2.609375,
      "learning_rate": 3.570159857904086e-06,
      "loss": 0.04854160845279694,
      "mean_token_accuracy": 0.9744318127632141,
      "num_tokens": 4242673.0,
      "step": 1050
    },
    {
      "entropy": 0.0459871404862497,
      "epoch": 0.8483393357342938,
      "grad_norm": 2.359375,
      "learning_rate": 3.3925399644760213e-06,
      "loss": 0.0448982447385788,
      "mean_token_accuracy": 0.9772727236151695,
      "num_tokens": 4283487.0,
      "step": 1060
    },
    {
      "entropy": 0.04385776598937809,
      "epoch": 0.8563425370148059,
      "grad_norm": 1.140625,
      "learning_rate": 3.2149200710479577e-06,
      "loss": 0.043155476450920105,
      "mean_token_accuracy": 0.97670454159379,
      "num_tokens": 4322572.0,
      "step": 1070
    },
    {
      "entropy": 0.04259421322203707,
      "epoch": 0.8643457382953181,
      "grad_norm": 1.4765625,
      "learning_rate": 3.0373001776198937e-06,
      "loss": 0.03880903720855713,
      "mean_token_accuracy": 0.9846590884029865,
      "num_tokens": 4363658.0,
      "step": 1080
    },
    {
      "entropy": 0.04217607219179627,
      "epoch": 0.8723489395758304,
      "grad_norm": 1.625,
      "learning_rate": 2.8596802841918296e-06,
      "loss": 0.04083235561847687,
      "mean_token_accuracy": 0.9755681775510311,
      "num_tokens": 4405021.0,
      "step": 1090
    },
    {
      "entropy": 0.04688679115497507,
      "epoch": 0.8803521408563425,
      "grad_norm": 3.734375,
      "learning_rate": 2.682060390763766e-06,
      "loss": 0.05114847421646118,
      "mean_token_accuracy": 0.9687499970197677,
      "num_tokens": 4445257.0,
      "step": 1100
    },
    {
      "entropy": 0.04287364967749454,
      "epoch": 0.8883553421368547,
      "grad_norm": 1.9765625,
      "learning_rate": 2.5044404973357016e-06,
      "loss": 0.04361231029033661,
      "mean_token_accuracy": 0.9784090854227543,
      "num_tokens": 4485832.0,
      "step": 1110
    },
    {
      "entropy": 0.04545789617113769,
      "epoch": 0.896358543417367,
      "grad_norm": 1.8984375,
      "learning_rate": 2.326820603907638e-06,
      "loss": 0.04764930307865143,
      "mean_token_accuracy": 0.9704545415937901,
      "num_tokens": 4526200.0,
      "step": 1120
    },
    {
      "entropy": 0.04260573259089142,
      "epoch": 0.9043617446978791,
      "grad_norm": 1.375,
      "learning_rate": 2.149200710479574e-06,
      "loss": 0.04154365062713623,
      "mean_token_accuracy": 0.9772727213799953,
      "num_tokens": 4566364.0,
      "step": 1130
    },
    {
      "entropy": 0.04537807885790244,
      "epoch": 0.9123649459783914,
      "grad_norm": 1.546875,
      "learning_rate": 1.97158081705151e-06,
      "loss": 0.04880580902099609,
      "mean_token_accuracy": 0.9698863588273525,
      "num_tokens": 4607256.0,
      "step": 1140
    },
    {
      "entropy": 0.043002552184043454,
      "epoch": 0.9203681472589036,
      "grad_norm": 1.15625,
      "learning_rate": 1.793960923623446e-06,
      "loss": 0.04373497366905212,
      "mean_token_accuracy": 0.9749999925494194,
      "num_tokens": 4647531.0,
      "step": 1150
    },
    {
      "entropy": 0.042623581358930095,
      "epoch": 0.9283713485394157,
      "grad_norm": 3.578125,
      "learning_rate": 1.616341030195382e-06,
      "loss": 0.04597070813179016,
      "mean_token_accuracy": 0.9755681745707989,
      "num_tokens": 4688875.0,
      "step": 1160
    },
    {
      "entropy": 0.04425350625788269,
      "epoch": 0.936374549819928,
      "grad_norm": 4.90625,
      "learning_rate": 1.4387211367673181e-06,
      "loss": 0.049137341976165774,
      "mean_token_accuracy": 0.9755681782960892,
      "num_tokens": 4730185.0,
      "step": 1170
    },
    {
      "entropy": 0.04330989782029064,
      "epoch": 0.9443777511004402,
      "grad_norm": 1.9765625,
      "learning_rate": 1.261101243339254e-06,
      "loss": 0.042343682050704955,
      "mean_token_accuracy": 0.975568176060915,
      "num_tokens": 4767986.0,
      "step": 1180
    },
    {
      "entropy": 0.04626658609486185,
      "epoch": 0.9523809523809523,
      "grad_norm": 1.9375,
      "learning_rate": 1.08348134991119e-06,
      "loss": 0.05028271675109863,
      "mean_token_accuracy": 0.9744318135082721,
      "num_tokens": 4807784.0,
      "step": 1190
    },
    {
      "entropy": 0.040159306283749176,
      "epoch": 0.9603841536614646,
      "grad_norm": 5.78125,
      "learning_rate": 9.058614564831261e-07,
      "loss": 0.04466935396194458,
      "mean_token_accuracy": 0.9801136322319508,
      "num_tokens": 4851222.0,
      "step": 1200
    },
    {
      "entropy": 0.040738246936234644,
      "epoch": 0.9683873549419768,
      "grad_norm": 2.046875,
      "learning_rate": 7.282415630550623e-07,
      "loss": 0.04467435479164124,
      "mean_token_accuracy": 0.9778409041464329,
      "num_tokens": 4889974.0,
      "step": 1210
    },
    {
      "entropy": 0.044979390571825206,
      "epoch": 0.976390556222489,
      "grad_norm": 2.921875,
      "learning_rate": 5.506216696269983e-07,
      "loss": 0.04703192114830017,
      "mean_token_accuracy": 0.9727272681891919,
      "num_tokens": 4931481.0,
      "step": 1220
    },
    {
      "entropy": 0.0433165812632069,
      "epoch": 0.9843937575030012,
      "grad_norm": 2.09375,
      "learning_rate": 3.7300177619893427e-07,
      "loss": 0.051091307401657106,
      "mean_token_accuracy": 0.9704545430839062,
      "num_tokens": 4972119.0,
      "step": 1230
    },
    {
      "entropy": 0.04314052956688101,
      "epoch": 0.9923969587835134,
      "grad_norm": 2.09375,
      "learning_rate": 1.9538188277087038e-07,
      "loss": 0.04473470449447632,
      "mean_token_accuracy": 0.9715909063816071,
      "num_tokens": 5013093.0,
      "step": 1240
    },
    {
      "entropy": 0.04248538885976335,
      "epoch": 1.0,
      "grad_norm": 3.421875,
      "learning_rate": 1.7761989342806397e-08,
      "loss": 0.04711778163909912,
      "mean_token_accuracy": 0.9718899483743467,
      "num_tokens": 5051976.0,
      "step": 1250
    }
  ],
  "logging_steps": 10,
  "max_steps": 1250,
  "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": 5.070355748382106e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}