{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.45226666666666665,
  "eval_steps": 500,
  "global_step": 2120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.2413338124752045,
      "epoch": 0.00021333333333333333,
      "grad_norm": 0.5237457752227783,
      "learning_rate": 0.0,
      "loss": 2.287360191345215,
      "mean_token_accuracy": 0.6451007425785065,
      "num_tokens": 4191.0,
      "step": 1
    },
    {
      "entropy": 1.2207386708921857,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.4350375831127167,
      "learning_rate": 3e-06,
      "loss": 2.19429079691569,
      "mean_token_accuracy": 0.65218452612559,
      "num_tokens": 39906.0,
      "step": 10
    },
    {
      "entropy": 1.1284118384122848,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.49733078479766846,
      "learning_rate": 6.333333333333334e-06,
      "loss": 1.9721708297729492,
      "mean_token_accuracy": 0.6748957321047783,
      "num_tokens": 90428.0,
      "step": 20
    },
    {
      "entropy": 1.3042965233325958,
      "epoch": 0.0064,
      "grad_norm": 0.39715778827667236,
      "learning_rate": 9.666666666666667e-06,
      "loss": 2.214934539794922,
      "mean_token_accuracy": 0.6405263364315033,
      "num_tokens": 140516.0,
      "step": 30
    },
    {
      "entropy": 1.206481871008873,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.5763714909553528,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 2.0720544815063477,
      "mean_token_accuracy": 0.6752387754619121,
      "num_tokens": 182338.0,
      "step": 40
    },
    {
      "entropy": 1.236506675183773,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.5518174171447754,
      "learning_rate": 1.6333333333333335e-05,
      "loss": 1.9000749588012695,
      "mean_token_accuracy": 0.6715238064527511,
      "num_tokens": 235345.0,
      "step": 50
    },
    {
      "entropy": 1.2768938690423965,
      "epoch": 0.0128,
      "grad_norm": 0.6082292199134827,
      "learning_rate": 1.9666666666666666e-05,
      "loss": 1.8587135314941405,
      "mean_token_accuracy": 0.6761457294225692,
      "num_tokens": 279475.0,
      "step": 60
    },
    {
      "entropy": 1.4092280209064483,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.6658427119255066,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 1.7947965621948243,
      "mean_token_accuracy": 0.6667077802121639,
      "num_tokens": 330352.0,
      "step": 70
    },
    {
      "entropy": 1.4339179992675781,
      "epoch": 0.017066666666666667,
      "grad_norm": 1.2848966121673584,
      "learning_rate": 2.633333333333333e-05,
      "loss": 1.6212703704833984,
      "mean_token_accuracy": 0.6736222848296165,
      "num_tokens": 373039.0,
      "step": 80
    },
    {
      "entropy": 1.516643624007702,
      "epoch": 0.0192,
      "grad_norm": 0.5840373039245605,
      "learning_rate": 2.9666666666666672e-05,
      "loss": 1.6039567947387696,
      "mean_token_accuracy": 0.6713629268109799,
      "num_tokens": 422371.0,
      "step": 90
    },
    {
      "entropy": 1.467514592409134,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.5295536518096924,
      "learning_rate": 3.3e-05,
      "loss": 1.576413631439209,
      "mean_token_accuracy": 0.6885988213121891,
      "num_tokens": 462284.0,
      "step": 100
    },
    {
      "entropy": 1.3056970939040184,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.4493468701839447,
      "learning_rate": 3.633333333333333e-05,
      "loss": 1.2876591682434082,
      "mean_token_accuracy": 0.7173117578029633,
      "num_tokens": 508883.0,
      "step": 110
    },
    {
      "entropy": 1.2709181517362595,
      "epoch": 0.0256,
      "grad_norm": 0.5542399883270264,
      "learning_rate": 3.966666666666667e-05,
      "loss": 1.249708652496338,
      "mean_token_accuracy": 0.727140337228775,
      "num_tokens": 555373.0,
      "step": 120
    },
    {
      "entropy": 1.3233605474233627,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.4590514004230499,
      "learning_rate": 4.3e-05,
      "loss": 1.3480740547180177,
      "mean_token_accuracy": 0.7144808873534203,
      "num_tokens": 599533.0,
      "step": 130
    },
    {
      "entropy": 1.241466723382473,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.2866012752056122,
      "learning_rate": 4.633333333333333e-05,
      "loss": 1.275578498840332,
      "mean_token_accuracy": 0.7148015096783638,
      "num_tokens": 657261.0,
      "step": 140
    },
    {
      "entropy": 1.103185237944126,
      "epoch": 0.032,
      "grad_norm": 0.25304746627807617,
      "learning_rate": 4.966666666666667e-05,
      "loss": 1.1829004287719727,
      "mean_token_accuracy": 0.7423724889755249,
      "num_tokens": 704749.0,
      "step": 150
    },
    {
      "entropy": 1.085295742750168,
      "epoch": 0.034133333333333335,
      "grad_norm": 2.8501718044281006,
      "learning_rate": 5.300000000000001e-05,
      "loss": 1.1587767601013184,
      "mean_token_accuracy": 0.7412141926586628,
      "num_tokens": 751635.0,
      "step": 160
    },
    {
      "entropy": 1.0946420103311538,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.2150825560092926,
      "learning_rate": 5.633333333333334e-05,
      "loss": 1.1224396705627442,
      "mean_token_accuracy": 0.7382922798395157,
      "num_tokens": 794488.0,
      "step": 170
    },
    {
      "entropy": 1.0789968609809875,
      "epoch": 0.0384,
      "grad_norm": 0.2265262007713318,
      "learning_rate": 5.966666666666667e-05,
      "loss": 1.1242941856384276,
      "mean_token_accuracy": 0.7424697011709214,
      "num_tokens": 838647.0,
      "step": 180
    },
    {
      "entropy": 1.0386934965848922,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.2709059417247772,
      "learning_rate": 6.3e-05,
      "loss": 1.1370309829711913,
      "mean_token_accuracy": 0.7532300829887391,
      "num_tokens": 884497.0,
      "step": 190
    },
    {
      "entropy": 1.1033389106392861,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.27512305974960327,
      "learning_rate": 6.633333333333334e-05,
      "loss": 1.2109394073486328,
      "mean_token_accuracy": 0.7378732696175575,
      "num_tokens": 931398.0,
      "step": 200
    },
    {
      "entropy": 1.108424139022827,
      "epoch": 0.0448,
      "grad_norm": 0.23238568007946014,
      "learning_rate": 6.966666666666668e-05,
      "loss": 1.1784509658813476,
      "mean_token_accuracy": 0.7302148967981339,
      "num_tokens": 983649.0,
      "step": 210
    },
    {
      "entropy": 0.9056355074048043,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.3005298674106598,
      "learning_rate": 7.3e-05,
      "loss": 0.977406120300293,
      "mean_token_accuracy": 0.7774429574608803,
      "num_tokens": 1024372.0,
      "step": 220
    },
    {
      "entropy": 1.019825778901577,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.2753085494041443,
      "learning_rate": 7.633333333333334e-05,
      "loss": 1.1463540077209473,
      "mean_token_accuracy": 0.7483755856752395,
      "num_tokens": 1073667.0,
      "step": 230
    },
    {
      "entropy": 1.042508740723133,
      "epoch": 0.0512,
      "grad_norm": 0.32724323868751526,
      "learning_rate": 7.966666666666666e-05,
      "loss": 1.098323440551758,
      "mean_token_accuracy": 0.7490729346871376,
      "num_tokens": 1117523.0,
      "step": 240
    },
    {
      "entropy": 1.036650500446558,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.28012844920158386,
      "learning_rate": 8.3e-05,
      "loss": 1.078709030151367,
      "mean_token_accuracy": 0.751419472694397,
      "num_tokens": 1160152.0,
      "step": 250
    },
    {
      "entropy": 1.0605042964220046,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.28786003589630127,
      "learning_rate": 8.633333333333334e-05,
      "loss": 1.1537845611572266,
      "mean_token_accuracy": 0.7475032344460487,
      "num_tokens": 1203153.0,
      "step": 260
    },
    {
      "entropy": 1.0000992961227895,
      "epoch": 0.0576,
      "grad_norm": 0.2648380696773529,
      "learning_rate": 8.966666666666666e-05,
      "loss": 1.0818438529968262,
      "mean_token_accuracy": 0.7550511255860328,
      "num_tokens": 1246554.0,
      "step": 270
    },
    {
      "entropy": 0.9443735256791115,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.29017189145088196,
      "learning_rate": 9.300000000000001e-05,
      "loss": 1.0578758239746093,
      "mean_token_accuracy": 0.7694767877459526,
      "num_tokens": 1286049.0,
      "step": 280
    },
    {
      "entropy": 0.9980478152632714,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.3078839182853699,
      "learning_rate": 9.633333333333335e-05,
      "loss": 1.1183401107788087,
      "mean_token_accuracy": 0.7594211131334305,
      "num_tokens": 1331161.0,
      "step": 290
    },
    {
      "entropy": 0.9343128114938736,
      "epoch": 0.064,
      "grad_norm": 0.253248393535614,
      "learning_rate": 9.966666666666667e-05,
      "loss": 0.9862067222595214,
      "mean_token_accuracy": 0.7730094477534294,
      "num_tokens": 1375777.0,
      "step": 300
    },
    {
      "entropy": 1.0266294315457345,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.23917347192764282,
      "learning_rate": 9.999975737505649e-05,
      "loss": 1.1334312438964844,
      "mean_token_accuracy": 0.7540638357400894,
      "num_tokens": 1421027.0,
      "step": 310
    },
    {
      "entropy": 1.0486552365124227,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.2840607762336731,
      "learning_rate": 9.999891867457112e-05,
      "loss": 1.1424532890319825,
      "mean_token_accuracy": 0.7420963421463966,
      "num_tokens": 1472539.0,
      "step": 320
    },
    {
      "entropy": 0.9518789499998093,
      "epoch": 0.0704,
      "grad_norm": 0.25352534651756287,
      "learning_rate": 9.999748091322068e-05,
      "loss": 0.9646738052368165,
      "mean_token_accuracy": 0.7610545977950096,
      "num_tokens": 1518725.0,
      "step": 330
    },
    {
      "entropy": 1.014043178409338,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.22224737703800201,
      "learning_rate": 9.999544410823167e-05,
      "loss": 1.0504605293273925,
      "mean_token_accuracy": 0.7481713563203811,
      "num_tokens": 1570475.0,
      "step": 340
    },
    {
      "entropy": 0.9679042734205723,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.2880384027957916,
      "learning_rate": 9.999280828400803e-05,
      "loss": 1.0482870101928712,
      "mean_token_accuracy": 0.7645679444074631,
      "num_tokens": 1617068.0,
      "step": 350
    },
    {
      "entropy": 1.077889482676983,
      "epoch": 0.0768,
      "grad_norm": 0.24659322202205658,
      "learning_rate": 9.998957347213085e-05,
      "loss": 1.1968301773071288,
      "mean_token_accuracy": 0.7367698416113854,
      "num_tokens": 1661987.0,
      "step": 360
    },
    {
      "entropy": 1.0857299536466598,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.31553569436073303,
      "learning_rate": 9.99857397113579e-05,
      "loss": 1.1405964851379395,
      "mean_token_accuracy": 0.7424520552158356,
      "num_tokens": 1712075.0,
      "step": 370
    },
    {
      "entropy": 1.0094588652253151,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.1927226334810257,
      "learning_rate": 9.998130704762335e-05,
      "loss": 1.062878704071045,
      "mean_token_accuracy": 0.7529133662581444,
      "num_tokens": 1762001.0,
      "step": 380
    },
    {
      "entropy": 0.95768673568964,
      "epoch": 0.0832,
      "grad_norm": 0.2921101748943329,
      "learning_rate": 9.997627553403699e-05,
      "loss": 1.0986035346984864,
      "mean_token_accuracy": 0.76050655990839,
      "num_tokens": 1805002.0,
      "step": 390
    },
    {
      "entropy": 0.8763241834938527,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.24879010021686554,
      "learning_rate": 9.997064523088384e-05,
      "loss": 0.9313676834106446,
      "mean_token_accuracy": 0.7820899412035942,
      "num_tokens": 1844640.0,
      "step": 400
    },
    {
      "entropy": 1.0211177349090577,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.27772271633148193,
      "learning_rate": 9.996441620562322e-05,
      "loss": 1.1202519416809082,
      "mean_token_accuracy": 0.7511946842074394,
      "num_tokens": 1891181.0,
      "step": 410
    },
    {
      "entropy": 0.9595077157020568,
      "epoch": 0.0896,
      "grad_norm": 0.31688833236694336,
      "learning_rate": 9.995758853288805e-05,
      "loss": 1.0745075225830079,
      "mean_token_accuracy": 0.7604640245437622,
      "num_tokens": 1934494.0,
      "step": 420
    },
    {
      "entropy": 0.963299511373043,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.29464003443717957,
      "learning_rate": 9.995016229448395e-05,
      "loss": 1.0001104354858399,
      "mean_token_accuracy": 0.763674932718277,
      "num_tokens": 1976770.0,
      "step": 430
    },
    {
      "entropy": 1.1388054117560387,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.2581053078174591,
      "learning_rate": 9.994213757938819e-05,
      "loss": 1.2570265769958495,
      "mean_token_accuracy": 0.726725485175848,
      "num_tokens": 2028837.0,
      "step": 440
    },
    {
      "entropy": 0.9712555348873139,
      "epoch": 0.096,
      "grad_norm": 0.3099062740802765,
      "learning_rate": 9.993351448374873e-05,
      "loss": 1.072590446472168,
      "mean_token_accuracy": 0.7607480764389039,
      "num_tokens": 2072441.0,
      "step": 450
    },
    {
      "entropy": 1.0221860893070698,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.25882700085639954,
      "learning_rate": 9.992429311088296e-05,
      "loss": 1.0992252349853515,
      "mean_token_accuracy": 0.7492104887962341,
      "num_tokens": 2119587.0,
      "step": 460
    },
    {
      "entropy": 1.082937440276146,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.23417538404464722,
      "learning_rate": 9.991447357127657e-05,
      "loss": 1.2704851150512695,
      "mean_token_accuracy": 0.7427436165511608,
      "num_tokens": 2171365.0,
      "step": 470
    },
    {
      "entropy": 0.8950243927538395,
      "epoch": 0.1024,
      "grad_norm": 0.27864041924476624,
      "learning_rate": 9.990405598258212e-05,
      "loss": 0.965212631225586,
      "mean_token_accuracy": 0.773382380604744,
      "num_tokens": 2216186.0,
      "step": 480
    },
    {
      "entropy": 0.9513905093073844,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.27456656098365784,
      "learning_rate": 9.989304046961772e-05,
      "loss": 1.0514169692993165,
      "mean_token_accuracy": 0.7602177545428276,
      "num_tokens": 2263172.0,
      "step": 490
    },
    {
      "entropy": 1.1295288607478142,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.2654035985469818,
      "learning_rate": 9.988142716436546e-05,
      "loss": 1.2499847412109375,
      "mean_token_accuracy": 0.7319100961089134,
      "num_tokens": 2309133.0,
      "step": 500
    },
    {
      "entropy": 0.9629665195941925,
      "epoch": 0.1088,
      "grad_norm": 0.2432626485824585,
      "learning_rate": 9.986921620596989e-05,
      "loss": 1.038647174835205,
      "mean_token_accuracy": 0.7657018698751926,
      "num_tokens": 2352188.0,
      "step": 510
    },
    {
      "entropy": 1.0380054742097855,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.22924789786338806,
      "learning_rate": 9.985640774073634e-05,
      "loss": 1.0875710487365722,
      "mean_token_accuracy": 0.7410315036773681,
      "num_tokens": 2400184.0,
      "step": 520
    },
    {
      "entropy": 1.0756568349897861,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.322125107049942,
      "learning_rate": 9.984300192212912e-05,
      "loss": 1.15003662109375,
      "mean_token_accuracy": 0.7418296962976456,
      "num_tokens": 2445609.0,
      "step": 530
    },
    {
      "entropy": 1.037429604679346,
      "epoch": 0.1152,
      "grad_norm": 0.29860854148864746,
      "learning_rate": 9.982899891076973e-05,
      "loss": 1.1256014823913574,
      "mean_token_accuracy": 0.7413557574152947,
      "num_tokens": 2490595.0,
      "step": 540
    },
    {
      "entropy": 0.863083366304636,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.2839767634868622,
      "learning_rate": 9.98143988744349e-05,
      "loss": 0.9366037368774414,
      "mean_token_accuracy": 0.786571592092514,
      "num_tokens": 2537415.0,
      "step": 550
    },
    {
      "entropy": 0.9333060696721077,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.23799379169940948,
      "learning_rate": 9.979920198805464e-05,
      "loss": 1.0223896980285645,
      "mean_token_accuracy": 0.7608667835593224,
      "num_tokens": 2582645.0,
      "step": 560
    },
    {
      "entropy": 1.0654601491987705,
      "epoch": 0.1216,
      "grad_norm": 0.26716434955596924,
      "learning_rate": 9.97834084337101e-05,
      "loss": 1.1589471817016601,
      "mean_token_accuracy": 0.7418766617774963,
      "num_tokens": 2633308.0,
      "step": 570
    },
    {
      "entropy": 1.0902433142066001,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.20525577664375305,
      "learning_rate": 9.976701840063136e-05,
      "loss": 1.1464842796325683,
      "mean_token_accuracy": 0.735442753136158,
      "num_tokens": 2686009.0,
      "step": 580
    },
    {
      "entropy": 0.8992204323410988,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.2524716556072235,
      "learning_rate": 9.975003208519522e-05,
      "loss": 0.9971331596374512,
      "mean_token_accuracy": 0.7749405071139336,
      "num_tokens": 2731814.0,
      "step": 590
    },
    {
      "entropy": 1.0139609590172767,
      "epoch": 0.128,
      "grad_norm": 0.26556891202926636,
      "learning_rate": 9.973244969092282e-05,
      "loss": 1.1074792861938476,
      "mean_token_accuracy": 0.7530097424983978,
      "num_tokens": 2780314.0,
      "step": 600
    },
    {
      "entropy": 0.9936468213796615,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.1984289288520813,
      "learning_rate": 9.971427142847718e-05,
      "loss": 1.0992106437683105,
      "mean_token_accuracy": 0.7568746477365493,
      "num_tokens": 2827832.0,
      "step": 610
    },
    {
      "entropy": 0.9435897715389728,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.25219106674194336,
      "learning_rate": 9.969549751566075e-05,
      "loss": 1.016776180267334,
      "mean_token_accuracy": 0.76652851998806,
      "num_tokens": 2873516.0,
      "step": 620
    },
    {
      "entropy": 0.9641087010502816,
      "epoch": 0.1344,
      "grad_norm": 0.2497103363275528,
      "learning_rate": 9.967612817741272e-05,
      "loss": 1.0406078338623046,
      "mean_token_accuracy": 0.7586069479584694,
      "num_tokens": 2913600.0,
      "step": 630
    },
    {
      "entropy": 1.0591406509280206,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.23842781782150269,
      "learning_rate": 9.965616364580636e-05,
      "loss": 1.1297724723815918,
      "mean_token_accuracy": 0.7443821474909782,
      "num_tokens": 2962337.0,
      "step": 640
    },
    {
      "entropy": 1.0139970764517785,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.2730376124382019,
      "learning_rate": 9.963560416004623e-05,
      "loss": 1.112107276916504,
      "mean_token_accuracy": 0.7496687114238739,
      "num_tokens": 3006113.0,
      "step": 650
    },
    {
      "entropy": 0.9443116314709187,
      "epoch": 0.1408,
      "grad_norm": 0.26775607466697693,
      "learning_rate": 9.961444996646532e-05,
      "loss": 1.020461654663086,
      "mean_token_accuracy": 0.7648672193288804,
      "num_tokens": 3052837.0,
      "step": 660
    },
    {
      "entropy": 1.113273823261261,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.2903483510017395,
      "learning_rate": 9.95927013185221e-05,
      "loss": 1.2222958564758302,
      "mean_token_accuracy": 0.7364178076386452,
      "num_tokens": 3101650.0,
      "step": 670
    },
    {
      "entropy": 0.8284098848700523,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.3110564053058624,
      "learning_rate": 9.957035847679749e-05,
      "loss": 0.905357551574707,
      "mean_token_accuracy": 0.7894617035984993,
      "num_tokens": 3143357.0,
      "step": 680
    },
    {
      "entropy": 0.9675152562558651,
      "epoch": 0.1472,
      "grad_norm": 0.27567023038864136,
      "learning_rate": 9.954742170899172e-05,
      "loss": 1.0641048431396485,
      "mean_token_accuracy": 0.7591987878084183,
      "num_tokens": 3193515.0,
      "step": 690
    },
    {
      "entropy": 0.9688289143145085,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.28721460700035095,
      "learning_rate": 9.952389128992113e-05,
      "loss": 1.0814785957336426,
      "mean_token_accuracy": 0.759615159034729,
      "num_tokens": 3242439.0,
      "step": 700
    },
    {
      "entropy": 0.9880564108490943,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.3781116306781769,
      "learning_rate": 9.949976750151489e-05,
      "loss": 1.100698184967041,
      "mean_token_accuracy": 0.7498437210917472,
      "num_tokens": 3287959.0,
      "step": 710
    },
    {
      "entropy": 1.027633222937584,
      "epoch": 0.1536,
      "grad_norm": 0.2338317185640335,
      "learning_rate": 9.947505063281157e-05,
      "loss": 1.0913351058959961,
      "mean_token_accuracy": 0.750363714993,
      "num_tokens": 3335092.0,
      "step": 720
    },
    {
      "entropy": 1.1284249052405357,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.2954626679420471,
      "learning_rate": 9.944974097995581e-05,
      "loss": 1.2494465827941894,
      "mean_token_accuracy": 0.7387717284262181,
      "num_tokens": 3380879.0,
      "step": 730
    },
    {
      "entropy": 0.9154682122170925,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.2581311762332916,
      "learning_rate": 9.942383884619455e-05,
      "loss": 1.0275691032409668,
      "mean_token_accuracy": 0.7657118752598763,
      "num_tokens": 3430079.0,
      "step": 740
    },
    {
      "entropy": 1.0482445612549782,
      "epoch": 0.16,
      "grad_norm": 0.2496950626373291,
      "learning_rate": 9.939734454187365e-05,
      "loss": 1.1575148582458497,
      "mean_token_accuracy": 0.7434881895780563,
      "num_tokens": 3482687.0,
      "step": 750
    },
    {
      "entropy": 1.008715095371008,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.24042253196239471,
      "learning_rate": 9.937025838443397e-05,
      "loss": 1.0921018600463868,
      "mean_token_accuracy": 0.7544152162969112,
      "num_tokens": 3528140.0,
      "step": 760
    },
    {
      "entropy": 0.9826410934329033,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.20811501145362854,
      "learning_rate": 9.934258069840765e-05,
      "loss": 1.0858916282653808,
      "mean_token_accuracy": 0.7535254985094071,
      "num_tokens": 3578130.0,
      "step": 770
    },
    {
      "entropy": 0.9679107010364533,
      "epoch": 0.1664,
      "grad_norm": 0.3156932294368744,
      "learning_rate": 9.931431181541426e-05,
      "loss": 1.0209306716918944,
      "mean_token_accuracy": 0.7623399093747139,
      "num_tokens": 3625602.0,
      "step": 780
    },
    {
      "entropy": 0.8914617538452149,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.2739625573158264,
      "learning_rate": 9.928545207415675e-05,
      "loss": 1.014828872680664,
      "mean_token_accuracy": 0.7721783280372619,
      "num_tokens": 3671375.0,
      "step": 790
    },
    {
      "entropy": 0.9649161577224732,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.23826555907726288,
      "learning_rate": 9.92560018204174e-05,
      "loss": 1.0278871536254883,
      "mean_token_accuracy": 0.7589392751455307,
      "num_tokens": 3717893.0,
      "step": 800
    },
    {
      "entropy": 0.9766084000468254,
      "epoch": 0.1728,
      "grad_norm": 0.33309969305992126,
      "learning_rate": 9.92259614070538e-05,
      "loss": 1.0395455360412598,
      "mean_token_accuracy": 0.7545588746666908,
      "num_tokens": 3764739.0,
      "step": 810
    },
    {
      "entropy": 1.0871834971010685,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.22862163186073303,
      "learning_rate": 9.919533119399438e-05,
      "loss": 1.1718174934387207,
      "mean_token_accuracy": 0.7301450505852699,
      "num_tokens": 3810938.0,
      "step": 820
    },
    {
      "entropy": 1.0070038817822933,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.2899021506309509,
      "learning_rate": 9.916411154823433e-05,
      "loss": 1.146003818511963,
      "mean_token_accuracy": 0.7569455504417419,
      "num_tokens": 3857657.0,
      "step": 830
    },
    {
      "entropy": 1.116474264860153,
      "epoch": 0.1792,
      "grad_norm": 0.2815554738044739,
      "learning_rate": 9.913230284383112e-05,
      "loss": 1.1874059677124023,
      "mean_token_accuracy": 0.7290896072983741,
      "num_tokens": 3909965.0,
      "step": 840
    },
    {
      "entropy": 0.9500585079193116,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.3055127263069153,
      "learning_rate": 9.90999054619e-05,
      "loss": 1.0213364601135253,
      "mean_token_accuracy": 0.7678412273526192,
      "num_tokens": 3948812.0,
      "step": 850
    },
    {
      "entropy": 0.8651109091937542,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.2479069083929062,
      "learning_rate": 9.906691979060943e-05,
      "loss": 0.9216291427612304,
      "mean_token_accuracy": 0.7761695921421051,
      "num_tokens": 3991940.0,
      "step": 860
    },
    {
      "entropy": 1.1529859654605388,
      "epoch": 0.1856,
      "grad_norm": 0.26249563694000244,
      "learning_rate": 9.903334622517643e-05,
      "loss": 1.2724492073059082,
      "mean_token_accuracy": 0.7244490720331669,
      "num_tokens": 4050343.0,
      "step": 870
    },
    {
      "entropy": 1.0595085561275481,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.294251948595047,
      "learning_rate": 9.89991851678619e-05,
      "loss": 1.1476259231567383,
      "mean_token_accuracy": 0.7438870698213578,
      "num_tokens": 4099181.0,
      "step": 880
    },
    {
      "entropy": 1.0237427443265914,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.251798152923584,
      "learning_rate": 9.896443702796573e-05,
      "loss": 1.1016413688659668,
      "mean_token_accuracy": 0.7469503089785576,
      "num_tokens": 4143951.0,
      "step": 890
    },
    {
      "entropy": 0.9793856598436832,
      "epoch": 0.192,
      "grad_norm": 0.2858197093009949,
      "learning_rate": 9.892910222182196e-05,
      "loss": 1.0976881980895996,
      "mean_token_accuracy": 0.7614006102085114,
      "num_tokens": 4194087.0,
      "step": 900
    },
    {
      "entropy": 1.0538529880344867,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.23789124190807343,
      "learning_rate": 9.889318117279373e-05,
      "loss": 1.196424674987793,
      "mean_token_accuracy": 0.7455965608358384,
      "num_tokens": 4243300.0,
      "step": 910
    },
    {
      "entropy": 0.9650195389986038,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.3590649366378784,
      "learning_rate": 9.885667431126824e-05,
      "loss": 1.0054343223571778,
      "mean_token_accuracy": 0.7639905765652657,
      "num_tokens": 4291788.0,
      "step": 920
    },
    {
      "entropy": 0.8939604975283146,
      "epoch": 0.1984,
      "grad_norm": 0.2963460087776184,
      "learning_rate": 9.881958207465158e-05,
      "loss": 0.9668001174926758,
      "mean_token_accuracy": 0.778343915939331,
      "num_tokens": 4336567.0,
      "step": 930
    },
    {
      "entropy": 0.8603343136608601,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.22635255753993988,
      "learning_rate": 9.878190490736353e-05,
      "loss": 0.9547459602355957,
      "mean_token_accuracy": 0.780703829228878,
      "num_tokens": 4379939.0,
      "step": 940
    },
    {
      "entropy": 1.0914236083626747,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.28582993149757385,
      "learning_rate": 9.874364326083216e-05,
      "loss": 1.1933195114135742,
      "mean_token_accuracy": 0.7399712555110455,
      "num_tokens": 4429828.0,
      "step": 950
    },
    {
      "entropy": 0.9969653740525246,
      "epoch": 0.2048,
      "grad_norm": 0.23508235812187195,
      "learning_rate": 9.87047975934885e-05,
      "loss": 1.080200481414795,
      "mean_token_accuracy": 0.7547310657799244,
      "num_tokens": 4478207.0,
      "step": 960
    },
    {
      "entropy": 0.9463694766163826,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.28800663352012634,
      "learning_rate": 9.866536837076098e-05,
      "loss": 1.0770373344421387,
      "mean_token_accuracy": 0.7622694931924343,
      "num_tokens": 4523626.0,
      "step": 970
    },
    {
      "entropy": 1.017927524447441,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.22783571481704712,
      "learning_rate": 9.862535606506992e-05,
      "loss": 1.1111245155334473,
      "mean_token_accuracy": 0.7514706686139107,
      "num_tokens": 4575636.0,
      "step": 980
    },
    {
      "entropy": 0.9148829184472561,
      "epoch": 0.2112,
      "grad_norm": 0.44951871037483215,
      "learning_rate": 9.85847611558218e-05,
      "loss": 1.0038617134094239,
      "mean_token_accuracy": 0.7708485037088394,
      "num_tokens": 4623275.0,
      "step": 990
    },
    {
      "entropy": 0.9400355100631714,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.2274121642112732,
      "learning_rate": 9.85435841294036e-05,
      "loss": 1.051990795135498,
      "mean_token_accuracy": 0.762999877333641,
      "num_tokens": 4670070.0,
      "step": 1000
    },
    {
      "entropy": 0.9205634713172912,
      "epoch": 0.21546666666666667,
      "grad_norm": 0.2580691874027252,
      "learning_rate": 9.850182547917686e-05,
      "loss": 1.0361328125,
      "mean_token_accuracy": 0.77038114964962,
      "num_tokens": 4713754.0,
      "step": 1010
    },
    {
      "entropy": 0.9357184395194054,
      "epoch": 0.2176,
      "grad_norm": 0.3024005889892578,
      "learning_rate": 9.845948570547187e-05,
      "loss": 1.001820945739746,
      "mean_token_accuracy": 0.7686716303229332,
      "num_tokens": 4754969.0,
      "step": 1020
    },
    {
      "entropy": 0.9788610845804214,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.22762160003185272,
      "learning_rate": 9.841656531558163e-05,
      "loss": 1.0857264518737793,
      "mean_token_accuracy": 0.755905470252037,
      "num_tokens": 4799172.0,
      "step": 1030
    },
    {
      "entropy": 0.8699902944266796,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.3153718113899231,
      "learning_rate": 9.83730648237558e-05,
      "loss": 0.9507910728454589,
      "mean_token_accuracy": 0.7810183942317963,
      "num_tokens": 4841210.0,
      "step": 1040
    },
    {
      "entropy": 1.0130531772971154,
      "epoch": 0.224,
      "grad_norm": 0.254517138004303,
      "learning_rate": 9.832898475119446e-05,
      "loss": 1.0863225936889649,
      "mean_token_accuracy": 0.7473610386252403,
      "num_tokens": 4889983.0,
      "step": 1050
    },
    {
      "entropy": 0.9305486619472504,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.3190745413303375,
      "learning_rate": 9.828432562604197e-05,
      "loss": 1.0010540008544921,
      "mean_token_accuracy": 0.7654982030391693,
      "num_tokens": 4935048.0,
      "step": 1060
    },
    {
      "entropy": 0.8807580441236496,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.25836268067359924,
      "learning_rate": 9.823908798338061e-05,
      "loss": 0.9626541137695312,
      "mean_token_accuracy": 0.7766690820455551,
      "num_tokens": 4974286.0,
      "step": 1070
    },
    {
      "entropy": 0.9813514620065689,
      "epoch": 0.2304,
      "grad_norm": 0.23850619792938232,
      "learning_rate": 9.819327236522411e-05,
      "loss": 1.0773870468139648,
      "mean_token_accuracy": 0.7523389101028443,
      "num_tokens": 5029521.0,
      "step": 1080
    },
    {
      "entropy": 0.9586616240441799,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.32606637477874756,
      "learning_rate": 9.814687932051123e-05,
      "loss": 1.0027738571166993,
      "mean_token_accuracy": 0.7671207025647163,
      "num_tokens": 5074269.0,
      "step": 1090
    },
    {
      "entropy": 0.9240864880383015,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.23331928253173828,
      "learning_rate": 9.809990940509911e-05,
      "loss": 1.0010904312133788,
      "mean_token_accuracy": 0.7657159030437469,
      "num_tokens": 5120632.0,
      "step": 1100
    },
    {
      "entropy": 0.9960295349359513,
      "epoch": 0.2368,
      "grad_norm": 0.27676209807395935,
      "learning_rate": 9.805236318175672e-05,
      "loss": 1.0985889434814453,
      "mean_token_accuracy": 0.7565312668681144,
      "num_tokens": 5167690.0,
      "step": 1110
    },
    {
      "entropy": 0.9320930615067482,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.2536001205444336,
      "learning_rate": 9.800424122015802e-05,
      "loss": 1.0100153923034667,
      "mean_token_accuracy": 0.7664023399353027,
      "num_tokens": 5209976.0,
      "step": 1120
    },
    {
      "entropy": 0.960879421979189,
      "epoch": 0.24106666666666668,
      "grad_norm": 0.2650390565395355,
      "learning_rate": 9.79555440968751e-05,
      "loss": 1.0518557548522949,
      "mean_token_accuracy": 0.7670834749937058,
      "num_tokens": 5257491.0,
      "step": 1130
    },
    {
      "entropy": 1.0790555529296397,
      "epoch": 0.2432,
      "grad_norm": 0.2658112645149231,
      "learning_rate": 9.790627239537144e-05,
      "loss": 1.2411640167236329,
      "mean_token_accuracy": 0.7390920028090477,
      "num_tokens": 5302986.0,
      "step": 1140
    },
    {
      "entropy": 0.9476075693964958,
      "epoch": 0.24533333333333332,
      "grad_norm": 0.2797292470932007,
      "learning_rate": 9.785642670599479e-05,
      "loss": 1.0823354721069336,
      "mean_token_accuracy": 0.765032921731472,
      "num_tokens": 5348946.0,
      "step": 1150
    },
    {
      "entropy": 0.8997275173664093,
      "epoch": 0.24746666666666667,
      "grad_norm": 0.21643367409706116,
      "learning_rate": 9.780600762597005e-05,
      "loss": 0.9590452194213868,
      "mean_token_accuracy": 0.7714703544974327,
      "num_tokens": 5393421.0,
      "step": 1160
    },
    {
      "entropy": 1.0325972460210324,
      "epoch": 0.2496,
      "grad_norm": 0.26943519711494446,
      "learning_rate": 9.775501575939227e-05,
      "loss": 1.0748598098754882,
      "mean_token_accuracy": 0.7444046661257744,
      "num_tokens": 5443639.0,
      "step": 1170
    },
    {
      "entropy": 0.9544263079762458,
      "epoch": 0.2517333333333333,
      "grad_norm": 0.28337323665618896,
      "learning_rate": 9.770345171721929e-05,
      "loss": 1.0133567810058595,
      "mean_token_accuracy": 0.7597964540123939,
      "num_tokens": 5494552.0,
      "step": 1180
    },
    {
      "entropy": 0.9211089439690113,
      "epoch": 0.2538666666666667,
      "grad_norm": 0.23862819373607635,
      "learning_rate": 9.765131611726446e-05,
      "loss": 0.9742342948913574,
      "mean_token_accuracy": 0.7758729934692383,
      "num_tokens": 5540743.0,
      "step": 1190
    },
    {
      "entropy": 1.0495109014213084,
      "epoch": 0.256,
      "grad_norm": 0.2335851490497589,
      "learning_rate": 9.759860958418926e-05,
      "loss": 1.1212799072265625,
      "mean_token_accuracy": 0.7419089064002037,
      "num_tokens": 5592981.0,
      "step": 1200
    },
    {
      "entropy": 1.115493653714657,
      "epoch": 0.2581333333333333,
      "grad_norm": 0.2895198464393616,
      "learning_rate": 9.754533274949575e-05,
      "loss": 1.2449783325195312,
      "mean_token_accuracy": 0.7372826255857945,
      "num_tokens": 5637868.0,
      "step": 1210
    },
    {
      "entropy": 0.9640243485569954,
      "epoch": 0.26026666666666665,
      "grad_norm": 0.23221854865550995,
      "learning_rate": 9.749148625151908e-05,
      "loss": 1.0559998512268067,
      "mean_token_accuracy": 0.7624265968799591,
      "num_tokens": 5684479.0,
      "step": 1220
    },
    {
      "entropy": 0.9518351659178734,
      "epoch": 0.2624,
      "grad_norm": 0.2750278413295746,
      "learning_rate": 9.743707073541978e-05,
      "loss": 1.0608203887939454,
      "mean_token_accuracy": 0.7580445930361748,
      "num_tokens": 5726626.0,
      "step": 1230
    },
    {
      "entropy": 1.0462738864123822,
      "epoch": 0.26453333333333334,
      "grad_norm": 0.21800817549228668,
      "learning_rate": 9.738208685317611e-05,
      "loss": 1.172841453552246,
      "mean_token_accuracy": 0.7489309534430504,
      "num_tokens": 5776891.0,
      "step": 1240
    },
    {
      "entropy": 0.9446908816695213,
      "epoch": 0.26666666666666666,
      "grad_norm": 0.2996140718460083,
      "learning_rate": 9.732653526357612e-05,
      "loss": 1.0297443389892578,
      "mean_token_accuracy": 0.7673991709947586,
      "num_tokens": 5817721.0,
      "step": 1250
    },
    {
      "entropy": 1.0725971311330795,
      "epoch": 0.2688,
      "grad_norm": 0.25890064239501953,
      "learning_rate": 9.727041663220989e-05,
      "loss": 1.2022390365600586,
      "mean_token_accuracy": 0.7427652187645435,
      "num_tokens": 5864952.0,
      "step": 1260
    },
    {
      "entropy": 1.0392154708504677,
      "epoch": 0.27093333333333336,
      "grad_norm": 0.2563996911048889,
      "learning_rate": 9.721373163146148e-05,
      "loss": 1.1769997596740722,
      "mean_token_accuracy": 0.7434597261250019,
      "num_tokens": 5910222.0,
      "step": 1270
    },
    {
      "entropy": 0.9681669734418392,
      "epoch": 0.2730666666666667,
      "grad_norm": 0.2539491653442383,
      "learning_rate": 9.715648094050087e-05,
      "loss": 1.0889246940612793,
      "mean_token_accuracy": 0.7573506951332092,
      "num_tokens": 5958487.0,
      "step": 1280
    },
    {
      "entropy": 0.9369139075279236,
      "epoch": 0.2752,
      "grad_norm": 0.29088094830513,
      "learning_rate": 9.709866524527588e-05,
      "loss": 1.0017178535461426,
      "mean_token_accuracy": 0.7683532252907753,
      "num_tokens": 6002949.0,
      "step": 1290
    },
    {
      "entropy": 1.0060899198055266,
      "epoch": 0.2773333333333333,
      "grad_norm": 0.2669098675251007,
      "learning_rate": 9.704028523850392e-05,
      "loss": 1.0900717735290528,
      "mean_token_accuracy": 0.7483311414718627,
      "num_tokens": 6053462.0,
      "step": 1300
    },
    {
      "entropy": 0.9726419121026992,
      "epoch": 0.27946666666666664,
      "grad_norm": 0.2472907453775406,
      "learning_rate": 9.698134161966363e-05,
      "loss": 1.1085708618164063,
      "mean_token_accuracy": 0.7585345059633255,
      "num_tokens": 6103810.0,
      "step": 1310
    },
    {
      "entropy": 0.9410715244710446,
      "epoch": 0.2816,
      "grad_norm": 0.2315952181816101,
      "learning_rate": 9.692183509498659e-05,
      "loss": 1.04432373046875,
      "mean_token_accuracy": 0.7650494173169136,
      "num_tokens": 6149332.0,
      "step": 1320
    },
    {
      "entropy": 0.9582152135670186,
      "epoch": 0.28373333333333334,
      "grad_norm": 0.28324106335639954,
      "learning_rate": 9.686176637744884e-05,
      "loss": 1.0879361152648925,
      "mean_token_accuracy": 0.7590656578540802,
      "num_tokens": 6193221.0,
      "step": 1330
    },
    {
      "entropy": 0.8448746472597122,
      "epoch": 0.28586666666666666,
      "grad_norm": 0.27708905935287476,
      "learning_rate": 9.680113618676229e-05,
      "loss": 0.9311987876892089,
      "mean_token_accuracy": 0.7856365889310837,
      "num_tokens": 6238920.0,
      "step": 1340
    },
    {
      "entropy": 0.9729965463280678,
      "epoch": 0.288,
      "grad_norm": 0.2840976119041443,
      "learning_rate": 9.673994524936615e-05,
      "loss": 1.0606694221496582,
      "mean_token_accuracy": 0.7592033997178078,
      "num_tokens": 6286552.0,
      "step": 1350
    },
    {
      "entropy": 0.9429876990616322,
      "epoch": 0.29013333333333335,
      "grad_norm": 0.2964062988758087,
      "learning_rate": 9.667819429841817e-05,
      "loss": 1.0691003799438477,
      "mean_token_accuracy": 0.7747758001089096,
      "num_tokens": 6331573.0,
      "step": 1360
    },
    {
      "entropy": 1.0872648723423481,
      "epoch": 0.2922666666666667,
      "grad_norm": 0.2675784230232239,
      "learning_rate": 9.661588407378596e-05,
      "loss": 1.2234331130981446,
      "mean_token_accuracy": 0.7345306605100632,
      "num_tokens": 6383986.0,
      "step": 1370
    },
    {
      "entropy": 0.8609121344983578,
      "epoch": 0.2944,
      "grad_norm": 0.22704993188381195,
      "learning_rate": 9.655301532203797e-05,
      "loss": 0.934294605255127,
      "mean_token_accuracy": 0.7782075613737106,
      "num_tokens": 6426867.0,
      "step": 1380
    },
    {
      "entropy": 0.8974170722067356,
      "epoch": 0.2965333333333333,
      "grad_norm": 0.2712918221950531,
      "learning_rate": 9.648958879643467e-05,
      "loss": 0.9794875144958496,
      "mean_token_accuracy": 0.768825002014637,
      "num_tokens": 6470272.0,
      "step": 1390
    },
    {
      "entropy": 0.9115433268249035,
      "epoch": 0.2986666666666667,
      "grad_norm": 0.2932375371456146,
      "learning_rate": 9.642560525691948e-05,
      "loss": 0.9729208946228027,
      "mean_token_accuracy": 0.7670300453901291,
      "num_tokens": 6513165.0,
      "step": 1400
    },
    {
      "entropy": 0.8882284432649612,
      "epoch": 0.3008,
      "grad_norm": 0.28009793162345886,
      "learning_rate": 9.63610654701097e-05,
      "loss": 0.9678432464599609,
      "mean_token_accuracy": 0.7733970731496811,
      "num_tokens": 6556342.0,
      "step": 1410
    },
    {
      "entropy": 0.8777762867510319,
      "epoch": 0.30293333333333333,
      "grad_norm": 0.23763342201709747,
      "learning_rate": 9.629597020928722e-05,
      "loss": 0.9687582969665527,
      "mean_token_accuracy": 0.7792320027947426,
      "num_tokens": 6601597.0,
      "step": 1420
    },
    {
      "entropy": 0.9160067647695541,
      "epoch": 0.30506666666666665,
      "grad_norm": 0.31182846426963806,
      "learning_rate": 9.623032025438939e-05,
      "loss": 0.9827415466308593,
      "mean_token_accuracy": 0.7696192592382431,
      "num_tokens": 6649545.0,
      "step": 1430
    },
    {
      "entropy": 0.9564003840088844,
      "epoch": 0.3072,
      "grad_norm": 0.23129577934741974,
      "learning_rate": 9.61641163919996e-05,
      "loss": 1.0380489349365234,
      "mean_token_accuracy": 0.7621255874633789,
      "num_tokens": 6696196.0,
      "step": 1440
    },
    {
      "entropy": 1.05502300709486,
      "epoch": 0.30933333333333335,
      "grad_norm": 0.27563098073005676,
      "learning_rate": 9.609735941533788e-05,
      "loss": 1.150872802734375,
      "mean_token_accuracy": 0.7396457836031913,
      "num_tokens": 6745953.0,
      "step": 1450
    },
    {
      "entropy": 0.9179727308452129,
      "epoch": 0.31146666666666667,
      "grad_norm": 0.2697086036205292,
      "learning_rate": 9.603005012425135e-05,
      "loss": 1.0098464012145996,
      "mean_token_accuracy": 0.7658590793609619,
      "num_tokens": 6794497.0,
      "step": 1460
    },
    {
      "entropy": 1.1471355110406876,
      "epoch": 0.3136,
      "grad_norm": 0.2717576026916504,
      "learning_rate": 9.596218932520468e-05,
      "loss": 1.2438765525817872,
      "mean_token_accuracy": 0.7290126971900464,
      "num_tokens": 6841976.0,
      "step": 1470
    },
    {
      "entropy": 0.9211908876895905,
      "epoch": 0.3157333333333333,
      "grad_norm": 0.23845043778419495,
      "learning_rate": 9.589377783127047e-05,
      "loss": 1.0137989044189453,
      "mean_token_accuracy": 0.767640671133995,
      "num_tokens": 6882780.0,
      "step": 1480
    },
    {
      "entropy": 0.8538284994661808,
      "epoch": 0.3178666666666667,
      "grad_norm": 0.23966865241527557,
      "learning_rate": 9.58248164621194e-05,
      "loss": 0.908942985534668,
      "mean_token_accuracy": 0.7838981494307518,
      "num_tokens": 6922694.0,
      "step": 1490
    },
    {
      "entropy": 0.9744522646069527,
      "epoch": 0.32,
      "grad_norm": 0.27710849046707153,
      "learning_rate": 9.575530604401051e-05,
      "loss": 1.0989601135253906,
      "mean_token_accuracy": 0.7573227554559707,
      "num_tokens": 6972307.0,
      "step": 1500
    },
    {
      "entropy": 1.0386228650808333,
      "epoch": 0.3221333333333333,
      "grad_norm": 0.2898867428302765,
      "learning_rate": 9.56852474097812e-05,
      "loss": 1.1454790115356446,
      "mean_token_accuracy": 0.7430783316493035,
      "num_tokens": 7020126.0,
      "step": 1510
    },
    {
      "entropy": 0.9898950323462486,
      "epoch": 0.32426666666666665,
      "grad_norm": 0.2526203393936157,
      "learning_rate": 9.561464139883737e-05,
      "loss": 1.1016074180603028,
      "mean_token_accuracy": 0.7571706652641297,
      "num_tokens": 7068804.0,
      "step": 1520
    },
    {
      "entropy": 0.999041261523962,
      "epoch": 0.3264,
      "grad_norm": 0.24795448780059814,
      "learning_rate": 9.554348885714326e-05,
      "loss": 1.0913206100463868,
      "mean_token_accuracy": 0.7498321965336799,
      "num_tokens": 7120053.0,
      "step": 1530
    },
    {
      "entropy": 0.9470011726021766,
      "epoch": 0.32853333333333334,
      "grad_norm": 0.2526148557662964,
      "learning_rate": 9.547179063721139e-05,
      "loss": 1.0380746841430664,
      "mean_token_accuracy": 0.7631630048155784,
      "num_tokens": 7166553.0,
      "step": 1540
    },
    {
      "entropy": 0.9413013480603695,
      "epoch": 0.33066666666666666,
      "grad_norm": 0.2717132866382599,
      "learning_rate": 9.539954759809226e-05,
      "loss": 1.0266976356506348,
      "mean_token_accuracy": 0.7626717418432236,
      "num_tokens": 7211784.0,
      "step": 1550
    },
    {
      "entropy": 1.0694061882793904,
      "epoch": 0.3328,
      "grad_norm": 0.2959578335285187,
      "learning_rate": 9.532676060536419e-05,
      "loss": 1.1947439193725586,
      "mean_token_accuracy": 0.7360827416181565,
      "num_tokens": 7256909.0,
      "step": 1560
    },
    {
      "entropy": 1.100894968211651,
      "epoch": 0.33493333333333336,
      "grad_norm": 0.2543633282184601,
      "learning_rate": 9.525343053112276e-05,
      "loss": 1.2262410163879394,
      "mean_token_accuracy": 0.7347081542015076,
      "num_tokens": 7305883.0,
      "step": 1570
    },
    {
      "entropy": 0.887219874560833,
      "epoch": 0.3370666666666667,
      "grad_norm": 0.2346959412097931,
      "learning_rate": 9.517955825397056e-05,
      "loss": 0.9671891212463379,
      "mean_token_accuracy": 0.7790701374411583,
      "num_tokens": 7352371.0,
      "step": 1580
    },
    {
      "entropy": 0.8671779796481133,
      "epoch": 0.3392,
      "grad_norm": 0.32672831416130066,
      "learning_rate": 9.510514465900653e-05,
      "loss": 0.9251022338867188,
      "mean_token_accuracy": 0.7759940758347511,
      "num_tokens": 7393383.0,
      "step": 1590
    },
    {
      "entropy": 0.9520958945155144,
      "epoch": 0.3413333333333333,
      "grad_norm": 0.2607707679271698,
      "learning_rate": 9.50301906378154e-05,
      "loss": 1.0628732681274413,
      "mean_token_accuracy": 0.7598425537347794,
      "num_tokens": 7439670.0,
      "step": 1600
    },
    {
      "entropy": 1.0912248834967613,
      "epoch": 0.34346666666666664,
      "grad_norm": 0.2782971262931824,
      "learning_rate": 9.495469708845701e-05,
      "loss": 1.1914597511291505,
      "mean_token_accuracy": 0.7388298735022545,
      "num_tokens": 7485267.0,
      "step": 1610
    },
    {
      "entropy": 0.8908423334360123,
      "epoch": 0.3456,
      "grad_norm": 0.3009212613105774,
      "learning_rate": 9.487866491545554e-05,
      "loss": 0.9948520660400391,
      "mean_token_accuracy": 0.7739639699459075,
      "num_tokens": 7527918.0,
      "step": 1620
    },
    {
      "entropy": 1.0133905045688152,
      "epoch": 0.34773333333333334,
      "grad_norm": 0.22292616963386536,
      "learning_rate": 9.480209502978869e-05,
      "loss": 1.0447552680969239,
      "mean_token_accuracy": 0.7502199374139309,
      "num_tokens": 7573700.0,
      "step": 1630
    },
    {
      "entropy": 0.9277701899409294,
      "epoch": 0.34986666666666666,
      "grad_norm": 0.25027042627334595,
      "learning_rate": 9.472498834887671e-05,
      "loss": 0.9662016868591309,
      "mean_token_accuracy": 0.769710011780262,
      "num_tokens": 7619772.0,
      "step": 1640
    },
    {
      "entropy": 0.9953067198395729,
      "epoch": 0.352,
      "grad_norm": 0.2255147099494934,
      "learning_rate": 9.46473457965715e-05,
      "loss": 1.1140392303466797,
      "mean_token_accuracy": 0.7478319302201271,
      "num_tokens": 7666080.0,
      "step": 1650
    },
    {
      "entropy": 0.9726508021354675,
      "epoch": 0.35413333333333336,
      "grad_norm": 0.23199768364429474,
      "learning_rate": 9.456916830314548e-05,
      "loss": 1.043684482574463,
      "mean_token_accuracy": 0.7573757611215115,
      "num_tokens": 7717686.0,
      "step": 1660
    },
    {
      "entropy": 0.8418352358043194,
      "epoch": 0.3562666666666667,
      "grad_norm": 0.2636827826499939,
      "learning_rate": 9.449045680528041e-05,
      "loss": 0.9572833061218262,
      "mean_token_accuracy": 0.7885566264390945,
      "num_tokens": 7760984.0,
      "step": 1670
    },
    {
      "entropy": 0.9119837798178196,
      "epoch": 0.3584,
      "grad_norm": 0.2830665111541748,
      "learning_rate": 9.441121224605629e-05,
      "loss": 0.9864655494689941,
      "mean_token_accuracy": 0.7652302265167237,
      "num_tokens": 7806335.0,
      "step": 1680
    },
    {
      "entropy": 0.833198781311512,
      "epoch": 0.3605333333333333,
      "grad_norm": 0.2658802568912506,
      "learning_rate": 9.43314355749399e-05,
      "loss": 0.8842006683349609,
      "mean_token_accuracy": 0.7828017815947532,
      "num_tokens": 7846783.0,
      "step": 1690
    },
    {
      "entropy": 0.9806767851114273,
      "epoch": 0.3626666666666667,
      "grad_norm": 0.23127029836177826,
      "learning_rate": 9.425112774777354e-05,
      "loss": 1.116124439239502,
      "mean_token_accuracy": 0.7566492781043053,
      "num_tokens": 7890665.0,
      "step": 1700
    },
    {
      "entropy": 0.8920052781701088,
      "epoch": 0.3648,
      "grad_norm": 0.22740185260772705,
      "learning_rate": 9.417028972676359e-05,
      "loss": 1.005050277709961,
      "mean_token_accuracy": 0.7736168324947357,
      "num_tokens": 7933849.0,
      "step": 1710
    },
    {
      "entropy": 1.0065233081579208,
      "epoch": 0.36693333333333333,
      "grad_norm": 0.34241342544555664,
      "learning_rate": 9.408892248046885e-05,
      "loss": 1.1022210121154785,
      "mean_token_accuracy": 0.7556669354438782,
      "num_tokens": 7981235.0,
      "step": 1720
    },
    {
      "entropy": 0.9596796780824661,
      "epoch": 0.36906666666666665,
      "grad_norm": 0.24347490072250366,
      "learning_rate": 9.40070269837891e-05,
      "loss": 1.058722686767578,
      "mean_token_accuracy": 0.7606610782444477,
      "num_tokens": 8028778.0,
      "step": 1730
    },
    {
      "entropy": 0.8785062082111835,
      "epoch": 0.3712,
      "grad_norm": 0.20483236014842987,
      "learning_rate": 9.392460421795328e-05,
      "loss": 0.9666107177734375,
      "mean_token_accuracy": 0.7833232149481774,
      "num_tokens": 8071410.0,
      "step": 1740
    },
    {
      "entropy": 0.8095596194267273,
      "epoch": 0.37333333333333335,
      "grad_norm": 0.36941683292388916,
      "learning_rate": 9.38416551705078e-05,
      "loss": 0.8782508850097657,
      "mean_token_accuracy": 0.7895680025219918,
      "num_tokens": 8113695.0,
      "step": 1750
    },
    {
      "entropy": 0.9545476101338863,
      "epoch": 0.37546666666666667,
      "grad_norm": 0.23256246745586395,
      "learning_rate": 9.375818083530474e-05,
      "loss": 1.0102588653564453,
      "mean_token_accuracy": 0.7568591669201851,
      "num_tokens": 8161702.0,
      "step": 1760
    },
    {
      "entropy": 0.8932637803256511,
      "epoch": 0.3776,
      "grad_norm": 0.2185504138469696,
      "learning_rate": 9.367418221248989e-05,
      "loss": 0.9898091316223144,
      "mean_token_accuracy": 0.7799834325909615,
      "num_tokens": 8205315.0,
      "step": 1770
    },
    {
      "entropy": 0.9569453194737434,
      "epoch": 0.3797333333333333,
      "grad_norm": 0.34778159856796265,
      "learning_rate": 9.358966030849072e-05,
      "loss": 1.0686969757080078,
      "mean_token_accuracy": 0.7559246391057968,
      "num_tokens": 8252919.0,
      "step": 1780
    },
    {
      "entropy": 1.029954968392849,
      "epoch": 0.3818666666666667,
      "grad_norm": 0.26953041553497314,
      "learning_rate": 9.350461613600449e-05,
      "loss": 1.1169189453125,
      "mean_token_accuracy": 0.746114706993103,
      "num_tokens": 8300355.0,
      "step": 1790
    },
    {
      "entropy": 0.943967767059803,
      "epoch": 0.384,
      "grad_norm": 0.2434380203485489,
      "learning_rate": 9.34190507139859e-05,
      "loss": 0.9968074798583985,
      "mean_token_accuracy": 0.7640544638037682,
      "num_tokens": 8346150.0,
      "step": 1800
    },
    {
      "entropy": 0.960440730303526,
      "epoch": 0.38613333333333333,
      "grad_norm": 0.2441101223230362,
      "learning_rate": 9.333296506763505e-05,
      "loss": 1.059675121307373,
      "mean_token_accuracy": 0.7603183135390281,
      "num_tokens": 8393408.0,
      "step": 1810
    },
    {
      "entropy": 0.8498699732124806,
      "epoch": 0.38826666666666665,
      "grad_norm": 0.2330222874879837,
      "learning_rate": 9.324636022838509e-05,
      "loss": 0.9152523040771484,
      "mean_token_accuracy": 0.7780576214194298,
      "num_tokens": 8440947.0,
      "step": 1820
    },
    {
      "entropy": 0.9348598286509514,
      "epoch": 0.3904,
      "grad_norm": 0.25513678789138794,
      "learning_rate": 9.315923723388986e-05,
      "loss": 1.041547966003418,
      "mean_token_accuracy": 0.7705774419009686,
      "num_tokens": 8486476.0,
      "step": 1830
    },
    {
      "entropy": 0.8954004615545272,
      "epoch": 0.39253333333333335,
      "grad_norm": 0.24745526909828186,
      "learning_rate": 9.307159712801147e-05,
      "loss": 0.9804242134094239,
      "mean_token_accuracy": 0.773698341846466,
      "num_tokens": 8533216.0,
      "step": 1840
    },
    {
      "entropy": 0.9542841285467147,
      "epoch": 0.39466666666666667,
      "grad_norm": 0.3648325800895691,
      "learning_rate": 9.298344096080776e-05,
      "loss": 1.0262937545776367,
      "mean_token_accuracy": 0.7662550717592239,
      "num_tokens": 8579411.0,
      "step": 1850
    },
    {
      "entropy": 0.820501434057951,
      "epoch": 0.3968,
      "grad_norm": 0.26821690797805786,
      "learning_rate": 9.289476978851976e-05,
      "loss": 0.8598980903625488,
      "mean_token_accuracy": 0.7888635769486427,
      "num_tokens": 8623046.0,
      "step": 1860
    },
    {
      "entropy": 0.9134793929755688,
      "epoch": 0.3989333333333333,
      "grad_norm": 0.25643301010131836,
      "learning_rate": 9.280558467355907e-05,
      "loss": 1.0032004356384276,
      "mean_token_accuracy": 0.7722208425402641,
      "num_tokens": 8665201.0,
      "step": 1870
    },
    {
      "entropy": 0.9829332500696182,
      "epoch": 0.4010666666666667,
      "grad_norm": 0.27002307772636414,
      "learning_rate": 9.271588668449503e-05,
      "loss": 1.1377047538757323,
      "mean_token_accuracy": 0.7662092931568623,
      "num_tokens": 8707004.0,
      "step": 1880
    },
    {
      "entropy": 1.0146782457828523,
      "epoch": 0.4032,
      "grad_norm": 0.2506401240825653,
      "learning_rate": 9.262567689604195e-05,
      "loss": 1.124861240386963,
      "mean_token_accuracy": 0.7566848322749138,
      "num_tokens": 8753713.0,
      "step": 1890
    },
    {
      "entropy": 1.0487395107746125,
      "epoch": 0.4053333333333333,
      "grad_norm": 0.24727876484394073,
      "learning_rate": 9.25349563890463e-05,
      "loss": 1.254913330078125,
      "mean_token_accuracy": 0.7382492840290069,
      "num_tokens": 8801703.0,
      "step": 1900
    },
    {
      "entropy": 0.8892258107662201,
      "epoch": 0.40746666666666664,
      "grad_norm": 0.2835586667060852,
      "learning_rate": 9.244372625047372e-05,
      "loss": 0.9624475479125977,
      "mean_token_accuracy": 0.7717373371124268,
      "num_tokens": 8848191.0,
      "step": 1910
    },
    {
      "entropy": 0.9389666721224785,
      "epoch": 0.4096,
      "grad_norm": 0.2468794584274292,
      "learning_rate": 9.235198757339594e-05,
      "loss": 1.0801708221435546,
      "mean_token_accuracy": 0.7686146914958953,
      "num_tokens": 8895860.0,
      "step": 1920
    },
    {
      "entropy": 0.9686854526400566,
      "epoch": 0.41173333333333334,
      "grad_norm": 0.24529697000980377,
      "learning_rate": 9.225974145697775e-05,
      "loss": 1.062663745880127,
      "mean_token_accuracy": 0.7554567009210587,
      "num_tokens": 8945814.0,
      "step": 1930
    },
    {
      "entropy": 0.9051942273974418,
      "epoch": 0.41386666666666666,
      "grad_norm": 0.3458064794540405,
      "learning_rate": 9.216698900646383e-05,
      "loss": 0.9697030067443848,
      "mean_token_accuracy": 0.7712534263730049,
      "num_tokens": 8990326.0,
      "step": 1940
    },
    {
      "entropy": 1.0142885446548462,
      "epoch": 0.416,
      "grad_norm": 0.25167402625083923,
      "learning_rate": 9.20737313331655e-05,
      "loss": 1.0986696243286134,
      "mean_token_accuracy": 0.7525161564350128,
      "num_tokens": 9037859.0,
      "step": 1950
    },
    {
      "entropy": 1.0254975706338882,
      "epoch": 0.41813333333333336,
      "grad_norm": 0.2604680061340332,
      "learning_rate": 9.197996955444732e-05,
      "loss": 1.0939658164978028,
      "mean_token_accuracy": 0.7426734983921051,
      "num_tokens": 9088894.0,
      "step": 1960
    },
    {
      "entropy": 0.9542628638446331,
      "epoch": 0.4202666666666667,
      "grad_norm": 0.2736513614654541,
      "learning_rate": 9.188570479371387e-05,
      "loss": 1.14229097366333,
      "mean_token_accuracy": 0.7659218199551105,
      "num_tokens": 9133717.0,
      "step": 1970
    },
    {
      "entropy": 0.9707967802882195,
      "epoch": 0.4224,
      "grad_norm": 0.2167668491601944,
      "learning_rate": 9.179093818039616e-05,
      "loss": 1.077283763885498,
      "mean_token_accuracy": 0.7643323555588722,
      "num_tokens": 9178671.0,
      "step": 1980
    },
    {
      "entropy": 0.9660324349999427,
      "epoch": 0.4245333333333333,
      "grad_norm": 0.35394829511642456,
      "learning_rate": 9.169567084993814e-05,
      "loss": 1.083635139465332,
      "mean_token_accuracy": 0.7617688804864884,
      "num_tokens": 9221289.0,
      "step": 1990
    },
    {
      "entropy": 0.9027157455682755,
      "epoch": 0.4266666666666667,
      "grad_norm": 0.2674417495727539,
      "learning_rate": 9.159990394378303e-05,
      "loss": 0.980889892578125,
      "mean_token_accuracy": 0.7698795303702355,
      "num_tokens": 9268892.0,
      "step": 2000
    },
    {
      "entropy": 0.952321158349514,
      "epoch": 0.4288,
      "grad_norm": 0.2332596629858017,
      "learning_rate": 9.15036386093598e-05,
      "loss": 1.0610048294067382,
      "mean_token_accuracy": 0.7620892718434333,
      "num_tokens": 9317142.0,
      "step": 2010
    },
    {
      "entropy": 0.8780575722455979,
      "epoch": 0.43093333333333333,
      "grad_norm": 0.2631659209728241,
      "learning_rate": 9.140687600006929e-05,
      "loss": 0.983364200592041,
      "mean_token_accuracy": 0.7832586973905563,
      "num_tokens": 9360075.0,
      "step": 2020
    },
    {
      "entropy": 0.9146695531904697,
      "epoch": 0.43306666666666666,
      "grad_norm": 0.33491751551628113,
      "learning_rate": 9.13096172752704e-05,
      "loss": 0.9856008529663086,
      "mean_token_accuracy": 0.7726532012224198,
      "num_tokens": 9396928.0,
      "step": 2030
    },
    {
      "entropy": 1.0050749816000462,
      "epoch": 0.4352,
      "grad_norm": 0.30352818965911865,
      "learning_rate": 9.121186360026625e-05,
      "loss": 1.0613908767700195,
      "mean_token_accuracy": 0.7496299520134926,
      "num_tokens": 9449045.0,
      "step": 2040
    },
    {
      "entropy": 0.919689030200243,
      "epoch": 0.43733333333333335,
      "grad_norm": 0.33956724405288696,
      "learning_rate": 9.111361614629022e-05,
      "loss": 0.9860305786132812,
      "mean_token_accuracy": 0.7708889573812485,
      "num_tokens": 9498718.0,
      "step": 2050
    },
    {
      "entropy": 0.9274381674826145,
      "epoch": 0.43946666666666667,
      "grad_norm": 0.29058581590652466,
      "learning_rate": 9.101487609049181e-05,
      "loss": 0.9976702690124511,
      "mean_token_accuracy": 0.7693670354783535,
      "num_tokens": 9537791.0,
      "step": 2060
    },
    {
      "entropy": 0.917995036393404,
      "epoch": 0.4416,
      "grad_norm": 0.22210952639579773,
      "learning_rate": 9.091564461592274e-05,
      "loss": 1.01414155960083,
      "mean_token_accuracy": 0.7708904504776001,
      "num_tokens": 9581893.0,
      "step": 2070
    },
    {
      "entropy": 0.8260227806866169,
      "epoch": 0.4437333333333333,
      "grad_norm": 0.31099778413772583,
      "learning_rate": 9.081592291152252e-05,
      "loss": 0.9264348983764649,
      "mean_token_accuracy": 0.7857892021536828,
      "num_tokens": 9624465.0,
      "step": 2080
    },
    {
      "entropy": 0.9573663167655468,
      "epoch": 0.4458666666666667,
      "grad_norm": 0.2500694692134857,
      "learning_rate": 9.071571217210443e-05,
      "loss": 1.042325496673584,
      "mean_token_accuracy": 0.7585179045796394,
      "num_tokens": 9670927.0,
      "step": 2090
    },
    {
      "entropy": 0.8527260765433311,
      "epoch": 0.448,
      "grad_norm": 0.22426480054855347,
      "learning_rate": 9.061501359834108e-05,
      "loss": 0.9534576416015625,
      "mean_token_accuracy": 0.7796914517879486,
      "num_tokens": 9717399.0,
      "step": 2100
    },
    {
      "entropy": 1.0308616764843463,
      "epoch": 0.45013333333333333,
      "grad_norm": 0.2896214723587036,
      "learning_rate": 9.051382839675005e-05,
      "loss": 1.1293525695800781,
      "mean_token_accuracy": 0.7496985673904419,
      "num_tokens": 9766859.0,
      "step": 2110
    },
    {
      "entropy": 1.0453794084489345,
      "epoch": 0.45226666666666665,
      "grad_norm": 0.21886104345321655,
      "learning_rate": 9.041215777967945e-05,
      "loss": 1.128882598876953,
      "mean_token_accuracy": 0.7480149149894715,
      "num_tokens": 9821214.0,
      "step": 2120
    }
  ],
  "logging_steps": 10,
  "max_steps": 9376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.659666001301914e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}