{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 234,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.327302247285843,
      "epoch": 0.012903225806451613,
      "grad_norm": 1.0468459129333496,
      "learning_rate": 0.0,
      "loss": 1.9516217708587646,
      "mean_token_accuracy": 0.5490271672606468,
      "num_tokens": 129656.0,
      "step": 1
    },
    {
      "entropy": 1.3267911523580551,
      "epoch": 0.025806451612903226,
      "grad_norm": 1.0235601663589478,
      "learning_rate": 4.166666666666667e-06,
      "loss": 1.9609016180038452,
      "mean_token_accuracy": 0.545955665409565,
      "num_tokens": 257608.0,
      "step": 2
    },
    {
      "entropy": 1.4163906574249268,
      "epoch": 0.03870967741935484,
      "grad_norm": 0.8620004057884216,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.9539512395858765,
      "mean_token_accuracy": 0.5437170788645744,
      "num_tokens": 385654.0,
      "step": 3
    },
    {
      "entropy": 1.5072677731513977,
      "epoch": 0.05161290322580645,
      "grad_norm": 0.6674020290374756,
      "learning_rate": 1.25e-05,
      "loss": 1.929893136024475,
      "mean_token_accuracy": 0.5428568199276924,
      "num_tokens": 513578.0,
      "step": 4
    },
    {
      "entropy": 1.5772784054279327,
      "epoch": 0.06451612903225806,
      "grad_norm": 0.44926103949546814,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.8096842765808105,
      "mean_token_accuracy": 0.5594104528427124,
      "num_tokens": 643891.0,
      "step": 5
    },
    {
      "entropy": 1.691362425684929,
      "epoch": 0.07741935483870968,
      "grad_norm": 0.3624405860900879,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 1.792994499206543,
      "mean_token_accuracy": 0.5583670735359192,
      "num_tokens": 773552.0,
      "step": 6
    },
    {
      "entropy": 1.7625758796930313,
      "epoch": 0.09032258064516129,
      "grad_norm": 0.32741692662239075,
      "learning_rate": 2.5e-05,
      "loss": 1.7421350479125977,
      "mean_token_accuracy": 0.565700501203537,
      "num_tokens": 902676.0,
      "step": 7
    },
    {
      "entropy": 1.816343978047371,
      "epoch": 0.1032258064516129,
      "grad_norm": 0.31979212164878845,
      "learning_rate": 2.916666666666667e-05,
      "loss": 1.7134227752685547,
      "mean_token_accuracy": 0.5680374577641487,
      "num_tokens": 1031062.0,
      "step": 8
    },
    {
      "entropy": 1.884494736790657,
      "epoch": 0.11612903225806452,
      "grad_norm": 0.3261905312538147,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.697797417640686,
      "mean_token_accuracy": 0.5728070959448814,
      "num_tokens": 1159144.0,
      "step": 9
    },
    {
      "entropy": 1.9154246598482132,
      "epoch": 0.12903225806451613,
      "grad_norm": 0.33981192111968994,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.691472053527832,
      "mean_token_accuracy": 0.5723336413502693,
      "num_tokens": 1286047.0,
      "step": 10
    },
    {
      "entropy": 1.8476576060056686,
      "epoch": 0.14193548387096774,
      "grad_norm": 0.3439367115497589,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.6031842231750488,
      "mean_token_accuracy": 0.5876904204487801,
      "num_tokens": 1414956.0,
      "step": 11
    },
    {
      "entropy": 1.7925567030906677,
      "epoch": 0.15483870967741936,
      "grad_norm": 0.3315640985965729,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.5772321224212646,
      "mean_token_accuracy": 0.5901090204715729,
      "num_tokens": 1544083.0,
      "step": 12
    },
    {
      "entropy": 1.6733195632696152,
      "epoch": 0.16774193548387098,
      "grad_norm": 0.2760685086250305,
      "learning_rate": 5e-05,
      "loss": 1.5256061553955078,
      "mean_token_accuracy": 0.5991538763046265,
      "num_tokens": 1673817.0,
      "step": 13
    },
    {
      "entropy": 1.4815043658018112,
      "epoch": 0.18064516129032257,
      "grad_norm": 0.23523160815238953,
      "learning_rate": 4.9997496794168726e-05,
      "loss": 1.4415483474731445,
      "mean_token_accuracy": 0.6162490248680115,
      "num_tokens": 1801770.0,
      "step": 14
    },
    {
      "entropy": 1.4159721434116364,
      "epoch": 0.1935483870967742,
      "grad_norm": 0.28664934635162354,
      "learning_rate": 4.998998767795805e-05,
      "loss": 1.4667837619781494,
      "mean_token_accuracy": 0.6101680621504784,
      "num_tokens": 1931723.0,
      "step": 15
    },
    {
      "entropy": 1.3478048890829086,
      "epoch": 0.2064516129032258,
      "grad_norm": 0.28385764360427856,
      "learning_rate": 4.9977474155117045e-05,
      "loss": 1.4278314113616943,
      "mean_token_accuracy": 0.6164416000247002,
      "num_tokens": 2061345.0,
      "step": 16
    },
    {
      "entropy": 1.3173148781061172,
      "epoch": 0.21935483870967742,
      "grad_norm": 0.24572592973709106,
      "learning_rate": 4.995995873155958e-05,
      "loss": 1.3927912712097168,
      "mean_token_accuracy": 0.6206158325076103,
      "num_tokens": 2190508.0,
      "step": 17
    },
    {
      "entropy": 1.3462176322937012,
      "epoch": 0.23225806451612904,
      "grad_norm": 0.2069179117679596,
      "learning_rate": 4.99374449148625e-05,
      "loss": 1.3792784214019775,
      "mean_token_accuracy": 0.6236012801527977,
      "num_tokens": 2318941.0,
      "step": 18
    },
    {
      "entropy": 1.39307202398777,
      "epoch": 0.24516129032258063,
      "grad_norm": 0.19426818192005157,
      "learning_rate": 4.9909937213563165e-05,
      "loss": 1.37335205078125,
      "mean_token_accuracy": 0.62324009090662,
      "num_tokens": 2448059.0,
      "step": 19
    },
    {
      "entropy": 1.420734629034996,
      "epoch": 0.25806451612903225,
      "grad_norm": 0.19838440418243408,
      "learning_rate": 4.987744113625665e-05,
      "loss": 1.370849370956421,
      "mean_token_accuracy": 0.6241574659943581,
      "num_tokens": 2577375.0,
      "step": 20
    },
    {
      "entropy": 1.371973842382431,
      "epoch": 0.2709677419354839,
      "grad_norm": 0.18288694322109222,
      "learning_rate": 4.9839963190492576e-05,
      "loss": 1.3033034801483154,
      "mean_token_accuracy": 0.6400853618979454,
      "num_tokens": 2707055.0,
      "step": 21
    },
    {
      "entropy": 1.3838643580675125,
      "epoch": 0.2838709677419355,
      "grad_norm": 0.18097883462905884,
      "learning_rate": 4.979751088147192e-05,
      "loss": 1.318864107131958,
      "mean_token_accuracy": 0.6351697221398354,
      "num_tokens": 2836471.0,
      "step": 22
    },
    {
      "entropy": 1.354823961853981,
      "epoch": 0.2967741935483871,
      "grad_norm": 0.16573207080364227,
      "learning_rate": 4.975009271054409e-05,
      "loss": 1.2960338592529297,
      "mean_token_accuracy": 0.6396285593509674,
      "num_tokens": 2967186.0,
      "step": 23
    },
    {
      "entropy": 1.356062889099121,
      "epoch": 0.3096774193548387,
      "grad_norm": 0.15834170579910278,
      "learning_rate": 4.969771817350445e-05,
      "loss": 1.311288595199585,
      "mean_token_accuracy": 0.638341523706913,
      "num_tokens": 3091283.0,
      "step": 24
    },
    {
      "entropy": 1.3014232963323593,
      "epoch": 0.3225806451612903,
      "grad_norm": 0.15633215010166168,
      "learning_rate": 4.9640397758692715e-05,
      "loss": 1.2794947624206543,
      "mean_token_accuracy": 0.6442421302199364,
      "num_tokens": 3221704.0,
      "step": 25
    },
    {
      "entropy": 1.2741473019123077,
      "epoch": 0.33548387096774196,
      "grad_norm": 0.16600695252418518,
      "learning_rate": 4.957814294489261e-05,
      "loss": 1.2685260772705078,
      "mean_token_accuracy": 0.6439810171723366,
      "num_tokens": 3351359.0,
      "step": 26
    },
    {
      "entropy": 1.2793954610824585,
      "epoch": 0.34838709677419355,
      "grad_norm": 0.1699378341436386,
      "learning_rate": 4.9510966199033174e-05,
      "loss": 1.299481749534607,
      "mean_token_accuracy": 0.6377060264348984,
      "num_tokens": 3480058.0,
      "step": 27
    },
    {
      "entropy": 1.2380398958921432,
      "epoch": 0.36129032258064514,
      "grad_norm": 0.1586979180574417,
      "learning_rate": 4.943888097369216e-05,
      "loss": 1.2551636695861816,
      "mean_token_accuracy": 0.6473096534609795,
      "num_tokens": 3610897.0,
      "step": 28
    },
    {
      "entropy": 1.25519198179245,
      "epoch": 0.3741935483870968,
      "grad_norm": 0.1537477672100067,
      "learning_rate": 4.936190170440208e-05,
      "loss": 1.2700082063674927,
      "mean_token_accuracy": 0.6448632478713989,
      "num_tokens": 3740627.0,
      "step": 29
    },
    {
      "entropy": 1.200623482465744,
      "epoch": 0.3870967741935484,
      "grad_norm": 0.1391288787126541,
      "learning_rate": 4.928004380675941e-05,
      "loss": 1.2029218673706055,
      "mean_token_accuracy": 0.6599779576063156,
      "num_tokens": 3869695.0,
      "step": 30
    },
    {
      "entropy": 1.222929760813713,
      "epoch": 0.4,
      "grad_norm": 0.14511308073997498,
      "learning_rate": 4.9193323673337476e-05,
      "loss": 1.2141369581222534,
      "mean_token_accuracy": 0.6561597287654877,
      "num_tokens": 3998221.0,
      "step": 31
    },
    {
      "entropy": 1.2850614041090012,
      "epoch": 0.4129032258064516,
      "grad_norm": 0.14675162732601166,
      "learning_rate": 4.910175867040377e-05,
      "loss": 1.260993242263794,
      "mean_token_accuracy": 0.6444773599505424,
      "num_tokens": 4127108.0,
      "step": 32
    },
    {
      "entropy": 1.2510194778442383,
      "epoch": 0.4258064516129032,
      "grad_norm": 0.14100538194179535,
      "learning_rate": 4.9005367134442235e-05,
      "loss": 1.2186331748962402,
      "mean_token_accuracy": 0.65362448990345,
      "num_tokens": 4256994.0,
      "step": 33
    },
    {
      "entropy": 1.223235473036766,
      "epoch": 0.43870967741935485,
      "grad_norm": 0.13875682651996613,
      "learning_rate": 4.890416836848127e-05,
      "loss": 1.203627586364746,
      "mean_token_accuracy": 0.6568410098552704,
      "num_tokens": 4385244.0,
      "step": 34
    },
    {
      "entropy": 1.1924351304769516,
      "epoch": 0.45161290322580644,
      "grad_norm": 0.13426443934440613,
      "learning_rate": 4.8798182638228166e-05,
      "loss": 1.1726880073547363,
      "mean_token_accuracy": 0.6638099774718285,
      "num_tokens": 4515175.0,
      "step": 35
    },
    {
      "entropy": 1.2145061641931534,
      "epoch": 0.4645161290322581,
      "grad_norm": 0.1369384229183197,
      "learning_rate": 4.868743116801074e-05,
      "loss": 1.2057712078094482,
      "mean_token_accuracy": 0.6585973352193832,
      "num_tokens": 4644464.0,
      "step": 36
    },
    {
      "entropy": 1.1699982434511185,
      "epoch": 0.4774193548387097,
      "grad_norm": 0.13203217089176178,
      "learning_rate": 4.857193613652711e-05,
      "loss": 1.1678709983825684,
      "mean_token_accuracy": 0.6651179119944572,
      "num_tokens": 4774388.0,
      "step": 37
    },
    {
      "entropy": 1.1900109350681305,
      "epoch": 0.49032258064516127,
      "grad_norm": 0.1409081071615219,
      "learning_rate": 4.845172067240415e-05,
      "loss": 1.196943998336792,
      "mean_token_accuracy": 0.6569633036851883,
      "num_tokens": 4903171.0,
      "step": 38
    },
    {
      "entropy": 1.1959572583436966,
      "epoch": 0.5032258064516129,
      "grad_norm": 0.13402387499809265,
      "learning_rate": 4.8326808849565936e-05,
      "loss": 1.204121470451355,
      "mean_token_accuracy": 0.6561899334192276,
      "num_tokens": 5033677.0,
      "step": 39
    },
    {
      "entropy": 1.1748242527246475,
      "epoch": 0.5161290322580645,
      "grad_norm": 0.13230547308921814,
      "learning_rate": 4.819722568241274e-05,
      "loss": 1.1755125522613525,
      "mean_token_accuracy": 0.6607141047716141,
      "num_tokens": 5163487.0,
      "step": 40
    },
    {
      "entropy": 1.1830326318740845,
      "epoch": 0.5290322580645161,
      "grad_norm": 0.13548049330711365,
      "learning_rate": 4.806299712081172e-05,
      "loss": 1.183053970336914,
      "mean_token_accuracy": 0.661345437169075,
      "num_tokens": 5292899.0,
      "step": 41
    },
    {
      "entropy": 1.180369421839714,
      "epoch": 0.5419354838709678,
      "grad_norm": 0.13295236229896545,
      "learning_rate": 4.792415004490034e-05,
      "loss": 1.1716046333312988,
      "mean_token_accuracy": 0.6627360582351685,
      "num_tokens": 5422025.0,
      "step": 42
    },
    {
      "entropy": 1.1974405944347382,
      "epoch": 0.5548387096774193,
      "grad_norm": 0.12952998280525208,
      "learning_rate": 4.77807122597034e-05,
      "loss": 1.1795024871826172,
      "mean_token_accuracy": 0.6590340510010719,
      "num_tokens": 5551149.0,
      "step": 43
    },
    {
      "entropy": 1.1339516341686249,
      "epoch": 0.567741935483871,
      "grad_norm": 0.13098949193954468,
      "learning_rate": 4.7632712489564926e-05,
      "loss": 1.1292296648025513,
      "mean_token_accuracy": 0.6728876158595085,
      "num_tokens": 5680701.0,
      "step": 44
    },
    {
      "entropy": 1.1452270299196243,
      "epoch": 0.5806451612903226,
      "grad_norm": 0.13148927688598633,
      "learning_rate": 4.748018037239592e-05,
      "loss": 1.1412042379379272,
      "mean_token_accuracy": 0.6679868176579475,
      "num_tokens": 5808493.0,
      "step": 45
    },
    {
      "entropy": 1.1565511226654053,
      "epoch": 0.5935483870967742,
      "grad_norm": 0.12716591358184814,
      "learning_rate": 4.732314645373921e-05,
      "loss": 1.147752046585083,
      "mean_token_accuracy": 0.6678735390305519,
      "num_tokens": 5938296.0,
      "step": 46
    },
    {
      "entropy": 1.1531172394752502,
      "epoch": 0.6064516129032258,
      "grad_norm": 0.13128802180290222,
      "learning_rate": 4.7161642180652464e-05,
      "loss": 1.137876033782959,
      "mean_token_accuracy": 0.6694907322525978,
      "num_tokens": 6067439.0,
      "step": 47
    },
    {
      "entropy": 1.1726240366697311,
      "epoch": 0.6193548387096774,
      "grad_norm": 0.1309962421655655,
      "learning_rate": 4.699569989541074e-05,
      "loss": 1.166740894317627,
      "mean_token_accuracy": 0.6635602861642838,
      "num_tokens": 6197510.0,
      "step": 48
    },
    {
      "entropy": 1.150630384683609,
      "epoch": 0.632258064516129,
      "grad_norm": 0.12693582475185394,
      "learning_rate": 4.6825352829029705e-05,
      "loss": 1.1487030982971191,
      "mean_token_accuracy": 0.667711079120636,
      "num_tokens": 6327768.0,
      "step": 49
    },
    {
      "entropy": 1.1061551421880722,
      "epoch": 0.6451612903225806,
      "grad_norm": 0.12342569977045059,
      "learning_rate": 4.665063509461097e-05,
      "loss": 1.107095718383789,
      "mean_token_accuracy": 0.6762828528881073,
      "num_tokens": 6455846.0,
      "step": 50
    },
    {
      "entropy": 1.1445851027965546,
      "epoch": 0.6580645161290323,
      "grad_norm": 0.1373581886291504,
      "learning_rate": 4.647158168051066e-05,
      "loss": 1.1354420185089111,
      "mean_token_accuracy": 0.6716547980904579,
      "num_tokens": 6585457.0,
      "step": 51
    },
    {
      "entropy": 1.1473966836929321,
      "epoch": 0.6709677419354839,
      "grad_norm": 0.1250847578048706,
      "learning_rate": 4.628822844333278e-05,
      "loss": 1.1474499702453613,
      "mean_token_accuracy": 0.6668257936835289,
      "num_tokens": 6716096.0,
      "step": 52
    },
    {
      "entropy": 1.1469789743423462,
      "epoch": 0.6838709677419355,
      "grad_norm": 0.1268235445022583,
      "learning_rate": 4.6100612100748765e-05,
      "loss": 1.14267897605896,
      "mean_token_accuracy": 0.6698051393032074,
      "num_tokens": 6844341.0,
      "step": 53
    },
    {
      "entropy": 1.1110682487487793,
      "epoch": 0.6967741935483871,
      "grad_norm": 0.12988688051700592,
      "learning_rate": 4.59087702241444e-05,
      "loss": 1.1135189533233643,
      "mean_token_accuracy": 0.6766636297106743,
      "num_tokens": 6971794.0,
      "step": 54
    },
    {
      "entropy": 1.1421085596084595,
      "epoch": 0.7096774193548387,
      "grad_norm": 0.1267767995595932,
      "learning_rate": 4.571274123109606e-05,
      "loss": 1.1490681171417236,
      "mean_token_accuracy": 0.6662089303135872,
      "num_tokens": 7100353.0,
      "step": 55
    },
    {
      "entropy": 1.1042888760566711,
      "epoch": 0.7225806451612903,
      "grad_norm": 0.12492457777261734,
      "learning_rate": 4.551256437767719e-05,
      "loss": 1.110184669494629,
      "mean_token_accuracy": 0.6747554019093513,
      "num_tokens": 7230177.0,
      "step": 56
    },
    {
      "entropy": 1.065716564655304,
      "epoch": 0.7354838709677419,
      "grad_norm": 0.12365498393774033,
      "learning_rate": 4.530827975059715e-05,
      "loss": 1.0647603273391724,
      "mean_token_accuracy": 0.6866239011287689,
      "num_tokens": 7359042.0,
      "step": 57
    },
    {
      "entropy": 1.0687381699681282,
      "epoch": 0.7483870967741936,
      "grad_norm": 0.12762895226478577,
      "learning_rate": 4.5099928259173516e-05,
      "loss": 1.0665113925933838,
      "mean_token_accuracy": 0.6885024607181549,
      "num_tokens": 7488137.0,
      "step": 58
    },
    {
      "entropy": 1.1023950427770615,
      "epoch": 0.7612903225806451,
      "grad_norm": 0.12779441475868225,
      "learning_rate": 4.488755162713975e-05,
      "loss": 1.0950183868408203,
      "mean_token_accuracy": 0.6785372719168663,
      "num_tokens": 7618282.0,
      "step": 59
    },
    {
      "entropy": 1.089103415608406,
      "epoch": 0.7741935483870968,
      "grad_norm": 0.12895165383815765,
      "learning_rate": 4.467119238428975e-05,
      "loss": 1.0878229141235352,
      "mean_token_accuracy": 0.6820144504308701,
      "num_tokens": 7748017.0,
      "step": 60
    },
    {
      "entropy": 1.0883491188287735,
      "epoch": 0.7870967741935484,
      "grad_norm": 0.13146434724330902,
      "learning_rate": 4.445089385796099e-05,
      "loss": 1.0844019651412964,
      "mean_token_accuracy": 0.6833582445979118,
      "num_tokens": 7876986.0,
      "step": 61
    },
    {
      "entropy": 1.1199355721473694,
      "epoch": 0.8,
      "grad_norm": 0.13159561157226562,
      "learning_rate": 4.422670016435792e-05,
      "loss": 1.1150472164154053,
      "mean_token_accuracy": 0.6756617501378059,
      "num_tokens": 8006794.0,
      "step": 62
    },
    {
      "entropy": 1.1050203889608383,
      "epoch": 0.8129032258064516,
      "grad_norm": 0.1264924556016922,
      "learning_rate": 4.3998656199717435e-05,
      "loss": 1.1043422222137451,
      "mean_token_accuracy": 0.6767379865050316,
      "num_tokens": 8135408.0,
      "step": 63
    },
    {
      "entropy": 1.1202740967273712,
      "epoch": 0.8258064516129032,
      "grad_norm": 0.12594330310821533,
      "learning_rate": 4.3766807631318106e-05,
      "loss": 1.1134419441223145,
      "mean_token_accuracy": 0.676059253513813,
      "num_tokens": 8265182.0,
      "step": 64
    },
    {
      "entropy": 1.1008555963635445,
      "epoch": 0.8387096774193549,
      "grad_norm": 0.14352354407310486,
      "learning_rate": 4.353120088833501e-05,
      "loss": 1.089049220085144,
      "mean_token_accuracy": 0.679233193397522,
      "num_tokens": 8395710.0,
      "step": 65
    },
    {
      "entropy": 1.09499129652977,
      "epoch": 0.8516129032258064,
      "grad_norm": 0.12506140768527985,
      "learning_rate": 4.329188315254196e-05,
      "loss": 1.0959919691085815,
      "mean_token_accuracy": 0.6772814318537712,
      "num_tokens": 8525026.0,
      "step": 66
    },
    {
      "entropy": 1.0806580409407616,
      "epoch": 0.864516129032258,
      "grad_norm": 0.12470545619726181,
      "learning_rate": 4.3048902348863116e-05,
      "loss": 1.077396273612976,
      "mean_token_accuracy": 0.6855179369449615,
      "num_tokens": 8655176.0,
      "step": 67
    },
    {
      "entropy": 1.0938902795314789,
      "epoch": 0.8774193548387097,
      "grad_norm": 0.1254391372203827,
      "learning_rate": 4.280230713577564e-05,
      "loss": 1.096771240234375,
      "mean_token_accuracy": 0.6798589378595352,
      "num_tokens": 8784950.0,
      "step": 68
    },
    {
      "entropy": 1.0837120488286018,
      "epoch": 0.8903225806451613,
      "grad_norm": 0.12564809620380402,
      "learning_rate": 4.255214689556557e-05,
      "loss": 1.0887655019760132,
      "mean_token_accuracy": 0.6802218705415726,
      "num_tokens": 8915424.0,
      "step": 69
    },
    {
      "entropy": 1.1309964209794998,
      "epoch": 0.9032258064516129,
      "grad_norm": 0.13434115052223206,
      "learning_rate": 4.229847172443866e-05,
      "loss": 1.1279692649841309,
      "mean_token_accuracy": 0.6699688509106636,
      "num_tokens": 9043958.0,
      "step": 70
    },
    {
      "entropy": 1.0794824212789536,
      "epoch": 0.9161290322580645,
      "grad_norm": 0.13104921579360962,
      "learning_rate": 4.204133242248832e-05,
      "loss": 1.0755493640899658,
      "mean_token_accuracy": 0.6832893788814545,
      "num_tokens": 9173238.0,
      "step": 71
    },
    {
      "entropy": 1.0700343400239944,
      "epoch": 0.9290322580645162,
      "grad_norm": 0.12627503275871277,
      "learning_rate": 4.1780780483522575e-05,
      "loss": 1.0692481994628906,
      "mean_token_accuracy": 0.6846828386187553,
      "num_tokens": 9303178.0,
      "step": 72
    },
    {
      "entropy": 1.0971452593803406,
      "epoch": 0.9419354838709677,
      "grad_norm": 0.1322394162416458,
      "learning_rate": 4.151686808475204e-05,
      "loss": 1.0987398624420166,
      "mean_token_accuracy": 0.6774289608001709,
      "num_tokens": 9432086.0,
      "step": 73
    },
    {
      "entropy": 1.1076685041189194,
      "epoch": 0.9548387096774194,
      "grad_norm": 0.13057094812393188,
      "learning_rate": 4.1249648076341165e-05,
      "loss": 1.1064797639846802,
      "mean_token_accuracy": 0.6764660328626633,
      "num_tokens": 9560333.0,
      "step": 74
    },
    {
      "entropy": 1.0812313556671143,
      "epoch": 0.967741935483871,
      "grad_norm": 0.12581738829612732,
      "learning_rate": 4.0979173970824626e-05,
      "loss": 1.077629566192627,
      "mean_token_accuracy": 0.6832556426525116,
      "num_tokens": 9689775.0,
      "step": 75
    },
    {
      "entropy": 1.067845843732357,
      "epoch": 0.9806451612903225,
      "grad_norm": 0.12489581108093262,
      "learning_rate": 4.070549993239106e-05,
      "loss": 1.0688203573226929,
      "mean_token_accuracy": 0.6856153085827827,
      "num_tokens": 9818901.0,
      "step": 76
    },
    {
      "entropy": 1.070682093501091,
      "epoch": 0.9935483870967742,
      "grad_norm": 0.13314804434776306,
      "learning_rate": 4.0428680766036384e-05,
      "loss": 1.0635384321212769,
      "mean_token_accuracy": 0.6840452700853348,
      "num_tokens": 9948460.0,
      "step": 77
    },
    {
      "entropy": 1.0606148540973663,
      "epoch": 1.0,
      "grad_norm": 0.16886763274669647,
      "learning_rate": 4.0148771906588706e-05,
      "loss": 1.0518985986709595,
      "mean_token_accuracy": 0.6878193020820618,
      "num_tokens": 9997067.0,
      "step": 78
    },
    {
      "entropy": 1.079283893108368,
      "epoch": 1.0129032258064516,
      "grad_norm": 0.13475318253040314,
      "learning_rate": 3.986582940760717e-05,
      "loss": 1.0569158792495728,
      "mean_token_accuracy": 0.6868253797292709,
      "num_tokens": 10126462.0,
      "step": 79
    },
    {
      "entropy": 1.0828415900468826,
      "epoch": 1.0258064516129033,
      "grad_norm": 0.13142913579940796,
      "learning_rate": 3.957990993015683e-05,
      "loss": 1.0775682926177979,
      "mean_token_accuracy": 0.682832732796669,
      "num_tokens": 10256580.0,
      "step": 80
    },
    {
      "entropy": 1.0669744163751602,
      "epoch": 1.038709677419355,
      "grad_norm": 0.12954232096672058,
      "learning_rate": 3.929107073146197e-05,
      "loss": 1.0615326166152954,
      "mean_token_accuracy": 0.6860573291778564,
      "num_tokens": 10385670.0,
      "step": 81
    },
    {
      "entropy": 1.036408707499504,
      "epoch": 1.0516129032258064,
      "grad_norm": 0.13460129499435425,
      "learning_rate": 3.899936965343989e-05,
      "loss": 1.0449131727218628,
      "mean_token_accuracy": 0.6920557096600533,
      "num_tokens": 10515880.0,
      "step": 82
    },
    {
      "entropy": 1.057508446276188,
      "epoch": 1.064516129032258,
      "grad_norm": 0.13466627895832062,
      "learning_rate": 3.8704865111117746e-05,
      "loss": 1.0574533939361572,
      "mean_token_accuracy": 0.6864631548523903,
      "num_tokens": 10645214.0,
      "step": 83
    },
    {
      "entropy": 1.0200782120227814,
      "epoch": 1.0774193548387097,
      "grad_norm": 0.1338030993938446,
      "learning_rate": 3.840761608093456e-05,
      "loss": 1.0181647539138794,
      "mean_token_accuracy": 0.6967797502875328,
      "num_tokens": 10775324.0,
      "step": 84
    },
    {
      "entropy": 1.0154706835746765,
      "epoch": 1.0903225806451613,
      "grad_norm": 0.13213345408439636,
      "learning_rate": 3.8107682088930794e-05,
      "loss": 1.0140069723129272,
      "mean_token_accuracy": 0.6974232569336891,
      "num_tokens": 10906115.0,
      "step": 85
    },
    {
      "entropy": 1.0599654540419579,
      "epoch": 1.103225806451613,
      "grad_norm": 0.13774846494197845,
      "learning_rate": 3.7805123198827857e-05,
      "loss": 1.0535001754760742,
      "mean_token_accuracy": 0.6869458928704262,
      "num_tokens": 11036082.0,
      "step": 86
    },
    {
      "entropy": 1.063461497426033,
      "epoch": 1.1161290322580646,
      "grad_norm": 0.13691790401935577,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.062204360961914,
      "mean_token_accuracy": 0.6850434467196465,
      "num_tokens": 11166308.0,
      "step": 87
    },
    {
      "entropy": 1.053346373140812,
      "epoch": 1.129032258064516,
      "grad_norm": 0.1327267438173294,
      "learning_rate": 3.719237359534087e-05,
      "loss": 1.041867733001709,
      "mean_token_accuracy": 0.6904401481151581,
      "num_tokens": 11296038.0,
      "step": 88
    },
    {
      "entropy": 1.0241582617163658,
      "epoch": 1.1419354838709677,
      "grad_norm": 0.13621629774570465,
      "learning_rate": 3.688230558902725e-05,
      "loss": 1.0226869583129883,
      "mean_token_accuracy": 0.6944883912801743,
      "num_tokens": 11424449.0,
      "step": 89
    },
    {
      "entropy": 1.0044510662555695,
      "epoch": 1.1548387096774193,
      "grad_norm": 0.13091513514518738,
      "learning_rate": 3.656985807418248e-05,
      "loss": 1.0043504238128662,
      "mean_token_accuracy": 0.7012801319360733,
      "num_tokens": 11553352.0,
      "step": 90
    },
    {
      "entropy": 1.0132758170366287,
      "epoch": 1.167741935483871,
      "grad_norm": 0.1394454389810562,
      "learning_rate": 3.6255093620441834e-05,
      "loss": 1.0138417482376099,
      "mean_token_accuracy": 0.6968973129987717,
      "num_tokens": 11683188.0,
      "step": 91
    },
    {
      "entropy": 1.0309394523501396,
      "epoch": 1.1806451612903226,
      "grad_norm": 0.13282909989356995,
      "learning_rate": 3.593807526142261e-05,
      "loss": 1.0259687900543213,
      "mean_token_accuracy": 0.6926954090595245,
      "num_tokens": 11812927.0,
      "step": 92
    },
    {
      "entropy": 1.0551650151610374,
      "epoch": 1.1935483870967742,
      "grad_norm": 0.14189599454402924,
      "learning_rate": 3.56188664821012e-05,
      "loss": 1.0586897134780884,
      "mean_token_accuracy": 0.6854764148592949,
      "num_tokens": 11941868.0,
      "step": 93
    },
    {
      "entropy": 1.0260127633810043,
      "epoch": 1.206451612903226,
      "grad_norm": 0.13907048106193542,
      "learning_rate": 3.529753120609982e-05,
      "loss": 1.030573844909668,
      "mean_token_accuracy": 0.6930540651082993,
      "num_tokens": 12070500.0,
      "step": 94
    },
    {
      "entropy": 1.039493516087532,
      "epoch": 1.2193548387096773,
      "grad_norm": 0.13697883486747742,
      "learning_rate": 3.497413378288541e-05,
      "loss": 1.0335443019866943,
      "mean_token_accuracy": 0.690931960940361,
      "num_tokens": 12199259.0,
      "step": 95
    },
    {
      "entropy": 1.014926478266716,
      "epoch": 1.232258064516129,
      "grad_norm": 0.13661201298236847,
      "learning_rate": 3.464873897488322e-05,
      "loss": 1.005345344543457,
      "mean_token_accuracy": 0.6979024410247803,
      "num_tokens": 12327381.0,
      "step": 96
    },
    {
      "entropy": 1.0336228609085083,
      "epoch": 1.2451612903225806,
      "grad_norm": 0.13816405832767487,
      "learning_rate": 3.432141194450772e-05,
      "loss": 1.0337718725204468,
      "mean_token_accuracy": 0.6914172023534775,
      "num_tokens": 12457229.0,
      "step": 97
    },
    {
      "entropy": 1.030926525592804,
      "epoch": 1.2580645161290323,
      "grad_norm": 0.1320193111896515,
      "learning_rate": 3.39922182411134e-05,
      "loss": 1.0276269912719727,
      "mean_token_accuracy": 0.6948070153594017,
      "num_tokens": 12587371.0,
      "step": 98
    },
    {
      "entropy": 1.0352215021848679,
      "epoch": 1.270967741935484,
      "grad_norm": 0.14135801792144775,
      "learning_rate": 3.3661223787868094e-05,
      "loss": 1.0391088724136353,
      "mean_token_accuracy": 0.6897449269890785,
      "num_tokens": 12714956.0,
      "step": 99
    },
    {
      "entropy": 1.0415681898593903,
      "epoch": 1.2838709677419355,
      "grad_norm": 0.13542622327804565,
      "learning_rate": 3.332849486855144e-05,
      "loss": 1.0490217208862305,
      "mean_token_accuracy": 0.6891317665576935,
      "num_tokens": 12845041.0,
      "step": 100
    },
    {
      "entropy": 1.018684171140194,
      "epoch": 1.2967741935483872,
      "grad_norm": 0.14059163630008698,
      "learning_rate": 3.2994098114281134e-05,
      "loss": 1.0158987045288086,
      "mean_token_accuracy": 0.6964094191789627,
      "num_tokens": 12974429.0,
      "step": 101
    },
    {
      "entropy": 1.0010781586170197,
      "epoch": 1.3096774193548386,
      "grad_norm": 0.14262941479682922,
      "learning_rate": 3.265810049016959e-05,
      "loss": 0.9994345307350159,
      "mean_token_accuracy": 0.7018002942204475,
      "num_tokens": 13103925.0,
      "step": 102
    },
    {
      "entropy": 1.0431874468922615,
      "epoch": 1.3225806451612903,
      "grad_norm": 0.13670195639133453,
      "learning_rate": 3.232056928191376e-05,
      "loss": 1.0475877523422241,
      "mean_token_accuracy": 0.6897815689444542,
      "num_tokens": 13232750.0,
      "step": 103
    },
    {
      "entropy": 1.0185461789369583,
      "epoch": 1.335483870967742,
      "grad_norm": 0.13562367856502533,
      "learning_rate": 3.1981572082320756e-05,
      "loss": 1.014870524406433,
      "mean_token_accuracy": 0.6959144324064255,
      "num_tokens": 13362678.0,
      "step": 104
    },
    {
      "entropy": 1.0220624506473541,
      "epoch": 1.3483870967741935,
      "grad_norm": 0.13380715250968933,
      "learning_rate": 3.164117677777191e-05,
      "loss": 1.020029067993164,
      "mean_token_accuracy": 0.6945825219154358,
      "num_tokens": 13493019.0,
      "step": 105
    },
    {
      "entropy": 1.0095699653029442,
      "epoch": 1.3612903225806452,
      "grad_norm": 0.13702020049095154,
      "learning_rate": 3.1299451534628135e-05,
      "loss": 1.006049394607544,
      "mean_token_accuracy": 0.6977230757474899,
      "num_tokens": 13622495.0,
      "step": 106
    },
    {
      "entropy": 1.0184833258390427,
      "epoch": 1.3741935483870968,
      "grad_norm": 0.14241954684257507,
      "learning_rate": 3.0956464785579124e-05,
      "loss": 0.9997148513793945,
      "mean_token_accuracy": 0.7005062252283096,
      "num_tokens": 13751405.0,
      "step": 107
    },
    {
      "entropy": 1.0199502855539322,
      "epoch": 1.3870967741935485,
      "grad_norm": 0.13611233234405518,
      "learning_rate": 3.061228521593931e-05,
      "loss": 1.0128517150878906,
      "mean_token_accuracy": 0.6958493441343307,
      "num_tokens": 13881358.0,
      "step": 108
    },
    {
      "entropy": 0.9918312728404999,
      "epoch": 1.4,
      "grad_norm": 0.13883398473262787,
      "learning_rate": 3.0266981749893157e-05,
      "loss": 0.9998171925544739,
      "mean_token_accuracy": 0.7012378796935081,
      "num_tokens": 14010084.0,
      "step": 109
    },
    {
      "entropy": 1.008819818496704,
      "epoch": 1.4129032258064516,
      "grad_norm": 0.14105769991874695,
      "learning_rate": 2.9920623536692638e-05,
      "loss": 1.0185983180999756,
      "mean_token_accuracy": 0.6961732059717178,
      "num_tokens": 14139791.0,
      "step": 110
    },
    {
      "entropy": 1.0010607317090034,
      "epoch": 1.4258064516129032,
      "grad_norm": 0.1377599686384201,
      "learning_rate": 2.9573279936809667e-05,
      "loss": 1.0039435625076294,
      "mean_token_accuracy": 0.699637234210968,
      "num_tokens": 14270212.0,
      "step": 111
    },
    {
      "entropy": 1.0302995592355728,
      "epoch": 1.4387096774193548,
      "grad_norm": 0.13768720626831055,
      "learning_rate": 2.9225020508046232e-05,
      "loss": 1.0220448970794678,
      "mean_token_accuracy": 0.6949515119194984,
      "num_tokens": 14400115.0,
      "step": 112
    },
    {
      "entropy": 1.0343796014785767,
      "epoch": 1.4516129032258065,
      "grad_norm": 0.13906188309192657,
      "learning_rate": 2.8875914991604948e-05,
      "loss": 1.0252317190170288,
      "mean_token_accuracy": 0.6952718943357468,
      "num_tokens": 14530182.0,
      "step": 113
    },
    {
      "entropy": 1.0087079033255577,
      "epoch": 1.4645161290322581,
      "grad_norm": 0.14347852766513824,
      "learning_rate": 2.8526033298122985e-05,
      "loss": 1.0005042552947998,
      "mean_token_accuracy": 0.7015687823295593,
      "num_tokens": 14658760.0,
      "step": 114
    },
    {
      "entropy": 1.0079718455672264,
      "epoch": 1.4774193548387098,
      "grad_norm": 0.1353033185005188,
      "learning_rate": 2.8175445493671972e-05,
      "loss": 1.0109742879867554,
      "mean_token_accuracy": 0.6946247220039368,
      "num_tokens": 14787887.0,
      "step": 115
    },
    {
      "entropy": 1.0122466161847115,
      "epoch": 1.4903225806451612,
      "grad_norm": 0.1421288102865219,
      "learning_rate": 2.782422178572682e-05,
      "loss": 1.0127758979797363,
      "mean_token_accuracy": 0.6972710192203522,
      "num_tokens": 14917268.0,
      "step": 116
    },
    {
      "entropy": 1.0002537220716476,
      "epoch": 1.5032258064516129,
      "grad_norm": 0.14423465728759766,
      "learning_rate": 2.7472432509106248e-05,
      "loss": 1.0058646202087402,
      "mean_token_accuracy": 0.6994956210255623,
      "num_tokens": 15044663.0,
      "step": 117
    },
    {
      "entropy": 1.0204123258590698,
      "epoch": 1.5161290322580645,
      "grad_norm": 0.1408473253250122,
      "learning_rate": 2.7120148111887732e-05,
      "loss": 1.0183706283569336,
      "mean_token_accuracy": 0.6959905028343201,
      "num_tokens": 15174619.0,
      "step": 118
    },
    {
      "entropy": 1.0369951203465462,
      "epoch": 1.5290322580645161,
      "grad_norm": 0.1445825695991516,
      "learning_rate": 2.6767439141299865e-05,
      "loss": 1.0403022766113281,
      "mean_token_accuracy": 0.6914126351475716,
      "num_tokens": 15304197.0,
      "step": 119
    },
    {
      "entropy": 0.9871217235922813,
      "epoch": 1.5419354838709678,
      "grad_norm": 0.14201968908309937,
      "learning_rate": 2.6414376229594813e-05,
      "loss": 0.9843333959579468,
      "mean_token_accuracy": 0.7033548727631569,
      "num_tokens": 15432579.0,
      "step": 120
    },
    {
      "entropy": 1.0354598462581635,
      "epoch": 1.5548387096774192,
      "grad_norm": 0.14182494580745697,
      "learning_rate": 2.606103007990371e-05,
      "loss": 1.0324954986572266,
      "mean_token_accuracy": 0.6913894191384315,
      "num_tokens": 15561722.0,
      "step": 121
    },
    {
      "entropy": 1.0236361995339394,
      "epoch": 1.567741935483871,
      "grad_norm": 0.13691937923431396,
      "learning_rate": 2.570747145207796e-05,
      "loss": 1.0316237211227417,
      "mean_token_accuracy": 0.6904159933328629,
      "num_tokens": 15691958.0,
      "step": 122
    },
    {
      "entropy": 1.0215153396129608,
      "epoch": 1.5806451612903225,
      "grad_norm": 0.1487395167350769,
      "learning_rate": 2.5353771148519057e-05,
      "loss": 1.0282378196716309,
      "mean_token_accuracy": 0.6913777440786362,
      "num_tokens": 15821165.0,
      "step": 123
    },
    {
      "entropy": 0.9973925575613976,
      "epoch": 1.5935483870967742,
      "grad_norm": 0.14773723483085632,
      "learning_rate": 2.5e-05,
      "loss": 0.992594838142395,
      "mean_token_accuracy": 0.700667105615139,
      "num_tokens": 15949216.0,
      "step": 124
    },
    {
      "entropy": 1.0059948936104774,
      "epoch": 1.6064516129032258,
      "grad_norm": 0.14315782487392426,
      "learning_rate": 2.4646228851480956e-05,
      "loss": 0.9999693036079407,
      "mean_token_accuracy": 0.7004231512546539,
      "num_tokens": 16078522.0,
      "step": 125
    },
    {
      "entropy": 1.0299797132611275,
      "epoch": 1.6193548387096774,
      "grad_norm": 0.13848090171813965,
      "learning_rate": 2.429252854792205e-05,
      "loss": 1.01283860206604,
      "mean_token_accuracy": 0.6960466951131821,
      "num_tokens": 16208166.0,
      "step": 126
    },
    {
      "entropy": 1.02093555778265,
      "epoch": 1.632258064516129,
      "grad_norm": 0.14542198181152344,
      "learning_rate": 2.39389699200963e-05,
      "loss": 1.0113534927368164,
      "mean_token_accuracy": 0.697401650249958,
      "num_tokens": 16338202.0,
      "step": 127
    },
    {
      "entropy": 0.9917213767766953,
      "epoch": 1.6451612903225805,
      "grad_norm": 0.14419761300086975,
      "learning_rate": 2.358562377040519e-05,
      "loss": 0.9841994047164917,
      "mean_token_accuracy": 0.7038668245077133,
      "num_tokens": 16468262.0,
      "step": 128
    },
    {
      "entropy": 1.0204757675528526,
      "epoch": 1.6580645161290324,
      "grad_norm": 0.14411410689353943,
      "learning_rate": 2.3232560858700137e-05,
      "loss": 1.0233168601989746,
      "mean_token_accuracy": 0.6931110769510269,
      "num_tokens": 16597219.0,
      "step": 129
    },
    {
      "entropy": 1.0202430561184883,
      "epoch": 1.6709677419354838,
      "grad_norm": 0.14136308431625366,
      "learning_rate": 2.287985188811228e-05,
      "loss": 1.0240514278411865,
      "mean_token_accuracy": 0.6934163346886635,
      "num_tokens": 16725933.0,
      "step": 130
    },
    {
      "entropy": 1.0111908614635468,
      "epoch": 1.6838709677419355,
      "grad_norm": 0.14592957496643066,
      "learning_rate": 2.2527567490893758e-05,
      "loss": 1.0206562280654907,
      "mean_token_accuracy": 0.6934522315859795,
      "num_tokens": 16854221.0,
      "step": 131
    },
    {
      "entropy": 1.0256818979978561,
      "epoch": 1.696774193548387,
      "grad_norm": 0.13796362280845642,
      "learning_rate": 2.2175778214273185e-05,
      "loss": 1.0266203880310059,
      "mean_token_accuracy": 0.6940783485770226,
      "num_tokens": 16984319.0,
      "step": 132
    },
    {
      "entropy": 1.0219127759337425,
      "epoch": 1.7096774193548387,
      "grad_norm": 0.14714568853378296,
      "learning_rate": 2.182455450632803e-05,
      "loss": 1.024146556854248,
      "mean_token_accuracy": 0.693178303539753,
      "num_tokens": 17112766.0,
      "step": 133
    },
    {
      "entropy": 1.0336032658815384,
      "epoch": 1.7225806451612904,
      "grad_norm": 0.1392759084701538,
      "learning_rate": 2.1473966701877025e-05,
      "loss": 1.031114101409912,
      "mean_token_accuracy": 0.6899987980723381,
      "num_tokens": 17241888.0,
      "step": 134
    },
    {
      "entropy": 1.0012932121753693,
      "epoch": 1.7354838709677418,
      "grad_norm": 0.13640883564949036,
      "learning_rate": 2.1124085008395054e-05,
      "loss": 0.99483323097229,
      "mean_token_accuracy": 0.699407234787941,
      "num_tokens": 17371518.0,
      "step": 135
    },
    {
      "entropy": 1.023273080587387,
      "epoch": 1.7483870967741937,
      "grad_norm": 0.13954921066761017,
      "learning_rate": 2.0774979491953777e-05,
      "loss": 1.0156586170196533,
      "mean_token_accuracy": 0.6934596672654152,
      "num_tokens": 17501856.0,
      "step": 136
    },
    {
      "entropy": 1.0082550570368767,
      "epoch": 1.761290322580645,
      "grad_norm": 0.14078585803508759,
      "learning_rate": 2.0426720063190335e-05,
      "loss": 1.0049326419830322,
      "mean_token_accuracy": 0.6976227834820747,
      "num_tokens": 17630077.0,
      "step": 137
    },
    {
      "entropy": 1.0113943740725517,
      "epoch": 1.7741935483870968,
      "grad_norm": 0.1388006955385208,
      "learning_rate": 2.0079376463307368e-05,
      "loss": 1.008650302886963,
      "mean_token_accuracy": 0.6971171796321869,
      "num_tokens": 17757628.0,
      "step": 138
    },
    {
      "entropy": 1.0009506717324257,
      "epoch": 1.7870967741935484,
      "grad_norm": 0.1410999745130539,
      "learning_rate": 1.973301825010685e-05,
      "loss": 1.002673864364624,
      "mean_token_accuracy": 0.6974465698003769,
      "num_tokens": 17887287.0,
      "step": 139
    },
    {
      "entropy": 1.0393515154719353,
      "epoch": 1.8,
      "grad_norm": 0.14737561345100403,
      "learning_rate": 1.9387714784060685e-05,
      "loss": 1.0421193838119507,
      "mean_token_accuracy": 0.6883728802204132,
      "num_tokens": 18016015.0,
      "step": 140
    },
    {
      "entropy": 1.01366426050663,
      "epoch": 1.8129032258064517,
      "grad_norm": 0.14167554676532745,
      "learning_rate": 1.904353521442088e-05,
      "loss": 1.0130255222320557,
      "mean_token_accuracy": 0.6935537680983543,
      "num_tokens": 18145940.0,
      "step": 141
    },
    {
      "entropy": 0.9947181940078735,
      "epoch": 1.8258064516129031,
      "grad_norm": 0.14348004758358002,
      "learning_rate": 1.8700548465371874e-05,
      "loss": 0.9804049730300903,
      "mean_token_accuracy": 0.7035569846630096,
      "num_tokens": 18276039.0,
      "step": 142
    },
    {
      "entropy": 1.0215622782707214,
      "epoch": 1.838709677419355,
      "grad_norm": 0.1438874453306198,
      "learning_rate": 1.8358823222228097e-05,
      "loss": 1.0185149908065796,
      "mean_token_accuracy": 0.6950262635946274,
      "num_tokens": 18401790.0,
      "step": 143
    },
    {
      "entropy": 1.003854677081108,
      "epoch": 1.8516129032258064,
      "grad_norm": 0.13913430273532867,
      "learning_rate": 1.801842791767925e-05,
      "loss": 1.0018150806427002,
      "mean_token_accuracy": 0.698628194630146,
      "num_tokens": 18530808.0,
      "step": 144
    },
    {
      "entropy": 0.9781669676303864,
      "epoch": 1.864516129032258,
      "grad_norm": 0.14533355832099915,
      "learning_rate": 1.7679430718086243e-05,
      "loss": 0.9760633707046509,
      "mean_token_accuracy": 0.7056997269392014,
      "num_tokens": 18659766.0,
      "step": 145
    },
    {
      "entropy": 0.9946553856134415,
      "epoch": 1.8774193548387097,
      "grad_norm": 0.13890255987644196,
      "learning_rate": 1.7341899509830416e-05,
      "loss": 1.0002374649047852,
      "mean_token_accuracy": 0.7009364366531372,
      "num_tokens": 18789014.0,
      "step": 146
    },
    {
      "entropy": 1.0132156759500504,
      "epoch": 1.8903225806451613,
      "grad_norm": 0.13903337717056274,
      "learning_rate": 1.700590188571887e-05,
      "loss": 1.013561487197876,
      "mean_token_accuracy": 0.6940024495124817,
      "num_tokens": 18918025.0,
      "step": 147
    },
    {
      "entropy": 0.9755073860287666,
      "epoch": 1.903225806451613,
      "grad_norm": 0.14805617928504944,
      "learning_rate": 1.667150513144856e-05,
      "loss": 0.9804725646972656,
      "mean_token_accuracy": 0.7047407254576683,
      "num_tokens": 19048295.0,
      "step": 148
    },
    {
      "entropy": 1.007135532796383,
      "epoch": 1.9161290322580644,
      "grad_norm": 0.13888275623321533,
      "learning_rate": 1.633877621213192e-05,
      "loss": 1.0063148736953735,
      "mean_token_accuracy": 0.6959036514163017,
      "num_tokens": 19173607.0,
      "step": 149
    },
    {
      "entropy": 0.9916294440627098,
      "epoch": 1.9290322580645163,
      "grad_norm": 0.14132030308246613,
      "learning_rate": 1.6007781758886607e-05,
      "loss": 0.995391309261322,
      "mean_token_accuracy": 0.7005252242088318,
      "num_tokens": 19303083.0,
      "step": 150
    },
    {
      "entropy": 1.035450056195259,
      "epoch": 1.9419354838709677,
      "grad_norm": 0.13970625400543213,
      "learning_rate": 1.567858805549229e-05,
      "loss": 1.0316739082336426,
      "mean_token_accuracy": 0.6913958415389061,
      "num_tokens": 19431668.0,
      "step": 151
    },
    {
      "entropy": 1.0315918624401093,
      "epoch": 1.9548387096774194,
      "grad_norm": 0.1366419643163681,
      "learning_rate": 1.535126102511678e-05,
      "loss": 1.02440345287323,
      "mean_token_accuracy": 0.6905666291713715,
      "num_tokens": 19560373.0,
      "step": 152
    },
    {
      "entropy": 1.0261467769742012,
      "epoch": 1.967741935483871,
      "grad_norm": 0.14188458025455475,
      "learning_rate": 1.5025866217114592e-05,
      "loss": 1.0182201862335205,
      "mean_token_accuracy": 0.6918191909790039,
      "num_tokens": 19688747.0,
      "step": 153
    },
    {
      "entropy": 1.0280335769057274,
      "epoch": 1.9806451612903224,
      "grad_norm": 0.1408587545156479,
      "learning_rate": 1.4702468793900188e-05,
      "loss": 1.0299128293991089,
      "mean_token_accuracy": 0.6898224130272865,
      "num_tokens": 19816935.0,
      "step": 154
    },
    {
      "entropy": 1.0113243460655212,
      "epoch": 1.9935483870967743,
      "grad_norm": 0.13604769110679626,
      "learning_rate": 1.4381133517898804e-05,
      "loss": 1.0048298835754395,
      "mean_token_accuracy": 0.6976648792624474,
      "num_tokens": 19946010.0,
      "step": 155
    },
    {
      "entropy": 1.0134106129407883,
      "epoch": 2.0,
      "grad_norm": 0.1904267519712448,
      "learning_rate": 1.4061924738577403e-05,
      "loss": 0.9954104423522949,
      "mean_token_accuracy": 0.6922345757484436,
      "num_tokens": 19994005.0,
      "step": 156
    },
    {
      "entropy": 0.9955317601561546,
      "epoch": 2.0129032258064514,
      "grad_norm": 0.14227579534053802,
      "learning_rate": 1.3744906379558165e-05,
      "loss": 0.9748439788818359,
      "mean_token_accuracy": 0.7040891200304031,
      "num_tokens": 20123770.0,
      "step": 157
    },
    {
      "entropy": 1.0204933807253838,
      "epoch": 2.0258064516129033,
      "grad_norm": 0.14505904912948608,
      "learning_rate": 1.3430141925817532e-05,
      "loss": 0.9986345171928406,
      "mean_token_accuracy": 0.6985316872596741,
      "num_tokens": 20253618.0,
      "step": 158
    },
    {
      "entropy": 0.9957914128899574,
      "epoch": 2.0387096774193547,
      "grad_norm": 0.1376512050628662,
      "learning_rate": 1.3117694410972748e-05,
      "loss": 0.9839757084846497,
      "mean_token_accuracy": 0.7033893465995789,
      "num_tokens": 20382498.0,
      "step": 159
    },
    {
      "entropy": 1.00312839448452,
      "epoch": 2.0516129032258066,
      "grad_norm": 0.1435370147228241,
      "learning_rate": 1.2807626404659142e-05,
      "loss": 0.988297700881958,
      "mean_token_accuracy": 0.7001642063260078,
      "num_tokens": 20511465.0,
      "step": 160
    },
    {
      "entropy": 0.978405088186264,
      "epoch": 2.064516129032258,
      "grad_norm": 0.13807372748851776,
      "learning_rate": 1.2500000000000006e-05,
      "loss": 0.9760883450508118,
      "mean_token_accuracy": 0.7053746059536934,
      "num_tokens": 20640276.0,
      "step": 161
    },
    {
      "entropy": 0.9854478612542152,
      "epoch": 2.07741935483871,
      "grad_norm": 0.1422271430492401,
      "learning_rate": 1.2194876801172148e-05,
      "loss": 0.985142707824707,
      "mean_token_accuracy": 0.7017852514982224,
      "num_tokens": 20770635.0,
      "step": 162
    },
    {
      "entropy": 0.987832635641098,
      "epoch": 2.0903225806451613,
      "grad_norm": 0.14423753321170807,
      "learning_rate": 1.1892317911069212e-05,
      "loss": 0.9848812818527222,
      "mean_token_accuracy": 0.7014080882072449,
      "num_tokens": 20900634.0,
      "step": 163
    },
    {
      "entropy": 0.9720889776945114,
      "epoch": 2.1032258064516127,
      "grad_norm": 0.14397835731506348,
      "learning_rate": 1.1592383919065442e-05,
      "loss": 0.9813554286956787,
      "mean_token_accuracy": 0.7022952511906624,
      "num_tokens": 21030337.0,
      "step": 164
    },
    {
      "entropy": 0.948253445327282,
      "epoch": 2.1161290322580646,
      "grad_norm": 0.963468074798584,
      "learning_rate": 1.1295134888882258e-05,
      "loss": 0.9537678956985474,
      "mean_token_accuracy": 0.7122210636734962,
      "num_tokens": 21160459.0,
      "step": 165
    },
    {
      "entropy": 0.978998601436615,
      "epoch": 2.129032258064516,
      "grad_norm": 0.14730028808116913,
      "learning_rate": 1.1000630346560117e-05,
      "loss": 0.9974563121795654,
      "mean_token_accuracy": 0.6987469494342804,
      "num_tokens": 21289979.0,
      "step": 166
    },
    {
      "entropy": 0.9566375017166138,
      "epoch": 2.141935483870968,
      "grad_norm": 0.14551083743572235,
      "learning_rate": 1.0708929268538034e-05,
      "loss": 0.9667287468910217,
      "mean_token_accuracy": 0.7059918269515038,
      "num_tokens": 21417697.0,
      "step": 167
    },
    {
      "entropy": 1.0017154440283775,
      "epoch": 2.1548387096774193,
      "grad_norm": 0.1442798674106598,
      "learning_rate": 1.0420090069843167e-05,
      "loss": 1.006643295288086,
      "mean_token_accuracy": 0.6955008804798126,
      "num_tokens": 21547900.0,
      "step": 168
    },
    {
      "entropy": 0.9761978760361671,
      "epoch": 2.167741935483871,
      "grad_norm": 0.13751927018165588,
      "learning_rate": 1.0134170592392836e-05,
      "loss": 0.9718648195266724,
      "mean_token_accuracy": 0.7058127820491791,
      "num_tokens": 21677940.0,
      "step": 169
    },
    {
      "entropy": 0.9624803140759468,
      "epoch": 2.1806451612903226,
      "grad_norm": 0.14026784896850586,
      "learning_rate": 9.851228093411296e-06,
      "loss": 0.953010082244873,
      "mean_token_accuracy": 0.7094407603144646,
      "num_tokens": 21807629.0,
      "step": 170
    },
    {
      "entropy": 1.0061953738331795,
      "epoch": 2.193548387096774,
      "grad_norm": 0.14467491209506989,
      "learning_rate": 9.571319233963627e-06,
      "loss": 0.9934457540512085,
      "mean_token_accuracy": 0.698410265147686,
      "num_tokens": 21937842.0,
      "step": 171
    },
    {
      "entropy": 1.0045757293701172,
      "epoch": 2.206451612903226,
      "grad_norm": 0.1439468115568161,
      "learning_rate": 9.29450006760894e-06,
      "loss": 0.9853917956352234,
      "mean_token_accuracy": 0.7011738419532776,
      "num_tokens": 22067033.0,
      "step": 172
    },
    {
      "entropy": 0.9552254527807236,
      "epoch": 2.2193548387096773,
      "grad_norm": 0.13853858411312103,
      "learning_rate": 9.020826029175384e-06,
      "loss": 0.9448599815368652,
      "mean_token_accuracy": 0.7115063071250916,
      "num_tokens": 22196381.0,
      "step": 173
    },
    {
      "entropy": 0.981830969452858,
      "epoch": 2.232258064516129,
      "grad_norm": 0.13981983065605164,
      "learning_rate": 8.750351923658832e-06,
      "loss": 0.9704583883285522,
      "mean_token_accuracy": 0.7055226489901543,
      "num_tokens": 22325575.0,
      "step": 174
    },
    {
      "entropy": 0.9844187051057816,
      "epoch": 2.2451612903225806,
      "grad_norm": 0.14050233364105225,
      "learning_rate": 8.483131915247968e-06,
      "loss": 0.975999653339386,
      "mean_token_accuracy": 0.7042497023940086,
      "num_tokens": 22454987.0,
      "step": 175
    },
    {
      "entropy": 0.9636645689606667,
      "epoch": 2.258064516129032,
      "grad_norm": 0.13572311401367188,
      "learning_rate": 8.219219516477428e-06,
      "loss": 0.9580243825912476,
      "mean_token_accuracy": 0.7085861191153526,
      "num_tokens": 22584324.0,
      "step": 176
    },
    {
      "entropy": 0.9852519631385803,
      "epoch": 2.270967741935484,
      "grad_norm": 0.14262360334396362,
      "learning_rate": 7.958667577511683e-06,
      "loss": 0.9825783967971802,
      "mean_token_accuracy": 0.7021386325359344,
      "num_tokens": 22713028.0,
      "step": 177
    },
    {
      "entropy": 0.980076439678669,
      "epoch": 2.2838709677419353,
      "grad_norm": 0.14399471879005432,
      "learning_rate": 7.701528275561348e-06,
      "loss": 0.9789690375328064,
      "mean_token_accuracy": 0.7044615298509598,
      "num_tokens": 22841542.0,
      "step": 178
    },
    {
      "entropy": 0.9826332628726959,
      "epoch": 2.296774193548387,
      "grad_norm": 0.14319010078907013,
      "learning_rate": 7.447853104434438e-06,
      "loss": 0.9809717535972595,
      "mean_token_accuracy": 0.7025709077715874,
      "num_tokens": 22969323.0,
      "step": 179
    },
    {
      "entropy": 0.9805971756577492,
      "epoch": 2.3096774193548386,
      "grad_norm": 0.17046643793582916,
      "learning_rate": 7.197692864224365e-06,
      "loss": 0.9834362268447876,
      "mean_token_accuracy": 0.7006140649318695,
      "num_tokens": 23098695.0,
      "step": 180
    },
    {
      "entropy": 0.9558906927704811,
      "epoch": 2.3225806451612905,
      "grad_norm": 0.15952308475971222,
      "learning_rate": 6.951097651136889e-06,
      "loss": 0.9562183618545532,
      "mean_token_accuracy": 0.7097498401999474,
      "num_tokens": 23228396.0,
      "step": 181
    },
    {
      "entropy": 0.9680119752883911,
      "epoch": 2.335483870967742,
      "grad_norm": 0.14320263266563416,
      "learning_rate": 6.708116847458043e-06,
      "loss": 0.9723418354988098,
      "mean_token_accuracy": 0.7073116898536682,
      "num_tokens": 23358823.0,
      "step": 182
    },
    {
      "entropy": 0.9725239872932434,
      "epoch": 2.3483870967741938,
      "grad_norm": 0.1440429538488388,
      "learning_rate": 6.468799111665003e-06,
      "loss": 0.9727824330329895,
      "mean_token_accuracy": 0.7049575448036194,
      "num_tokens": 23488354.0,
      "step": 183
    },
    {
      "entropy": 0.9424054622650146,
      "epoch": 2.361290322580645,
      "grad_norm": 0.13706418871879578,
      "learning_rate": 6.23319236868189e-06,
      "loss": 0.9380854368209839,
      "mean_token_accuracy": 0.714186854660511,
      "num_tokens": 23618360.0,
      "step": 184
    },
    {
      "entropy": 0.9745959714055061,
      "epoch": 2.3741935483870966,
      "grad_norm": 0.1436537653207779,
      "learning_rate": 6.001343800282569e-06,
      "loss": 0.9735644459724426,
      "mean_token_accuracy": 0.704635463654995,
      "num_tokens": 23748436.0,
      "step": 185
    },
    {
      "entropy": 1.0063187405467033,
      "epoch": 2.3870967741935485,
      "grad_norm": 0.14635713398456573,
      "learning_rate": 5.7732998356420796e-06,
      "loss": 1.0041303634643555,
      "mean_token_accuracy": 0.6960858479142189,
      "num_tokens": 23874993.0,
      "step": 186
    },
    {
      "entropy": 0.967486634850502,
      "epoch": 2.4,
      "grad_norm": 0.13961286842823029,
      "learning_rate": 5.549106142039018e-06,
      "loss": 0.9667018055915833,
      "mean_token_accuracy": 0.7070406302809715,
      "num_tokens": 24003568.0,
      "step": 187
    },
    {
      "entropy": 0.9616538360714912,
      "epoch": 2.412903225806452,
      "grad_norm": 0.14085349440574646,
      "learning_rate": 5.328807615710246e-06,
      "loss": 0.9632882475852966,
      "mean_token_accuracy": 0.7090655639767647,
      "num_tokens": 24133856.0,
      "step": 188
    },
    {
      "entropy": 0.976001113653183,
      "epoch": 2.425806451612903,
      "grad_norm": 0.14318986237049103,
      "learning_rate": 5.112448372860257e-06,
      "loss": 0.9712142944335938,
      "mean_token_accuracy": 0.7032363936305046,
      "num_tokens": 24264174.0,
      "step": 189
    },
    {
      "entropy": 0.9615498259663582,
      "epoch": 2.4387096774193546,
      "grad_norm": 0.15484382212162018,
      "learning_rate": 4.900071740826489e-06,
      "loss": 0.9542532563209534,
      "mean_token_accuracy": 0.7100772634148598,
      "num_tokens": 24391008.0,
      "step": 190
    },
    {
      "entropy": 0.9784514307975769,
      "epoch": 2.4516129032258065,
      "grad_norm": 0.1551555097103119,
      "learning_rate": 4.691720249402856e-06,
      "loss": 0.9786233901977539,
      "mean_token_accuracy": 0.704196609556675,
      "num_tokens": 24518432.0,
      "step": 191
    },
    {
      "entropy": 0.9739161878824234,
      "epoch": 2.464516129032258,
      "grad_norm": 0.13926950097084045,
      "learning_rate": 4.487435622322814e-06,
      "loss": 0.9705713391304016,
      "mean_token_accuracy": 0.708015650510788,
      "num_tokens": 24647693.0,
      "step": 192
    },
    {
      "entropy": 0.973455473780632,
      "epoch": 2.47741935483871,
      "grad_norm": 0.13951078057289124,
      "learning_rate": 4.2872587689039484e-06,
      "loss": 0.9669640064239502,
      "mean_token_accuracy": 0.7067858427762985,
      "num_tokens": 24776864.0,
      "step": 193
    },
    {
      "entropy": 0.9738265424966812,
      "epoch": 2.490322580645161,
      "grad_norm": 0.137411966919899,
      "learning_rate": 4.091229775855598e-06,
      "loss": 0.9736723303794861,
      "mean_token_accuracy": 0.7044489830732346,
      "num_tokens": 24905902.0,
      "step": 194
    },
    {
      "entropy": 0.9626506343483925,
      "epoch": 2.5032258064516126,
      "grad_norm": 0.14244906604290009,
      "learning_rate": 3.8993878992512415e-06,
      "loss": 0.9621111154556274,
      "mean_token_accuracy": 0.706295557320118,
      "num_tokens": 25034657.0,
      "step": 195
    },
    {
      "entropy": 0.9638460949063301,
      "epoch": 2.5161290322580645,
      "grad_norm": 0.13963346183300018,
      "learning_rate": 3.711771556667218e-06,
      "loss": 0.9548102021217346,
      "mean_token_accuracy": 0.7104058936238289,
      "num_tokens": 25164329.0,
      "step": 196
    },
    {
      "entropy": 0.9780149757862091,
      "epoch": 2.5290322580645164,
      "grad_norm": 0.14156083762645721,
      "learning_rate": 3.5284183194893488e-06,
      "loss": 0.9741950035095215,
      "mean_token_accuracy": 0.7038740292191505,
      "num_tokens": 25293811.0,
      "step": 197
    },
    {
      "entropy": 1.0000493451952934,
      "epoch": 2.541935483870968,
      "grad_norm": 0.18765178322792053,
      "learning_rate": 3.3493649053890326e-06,
      "loss": 0.9990582466125488,
      "mean_token_accuracy": 0.6991258412599564,
      "num_tokens": 25423153.0,
      "step": 198
    },
    {
      "entropy": 0.9833392351865768,
      "epoch": 2.554838709677419,
      "grad_norm": 0.1423710733652115,
      "learning_rate": 3.1746471709702964e-06,
      "loss": 0.9852808713912964,
      "mean_token_accuracy": 0.7009892612695694,
      "num_tokens": 25551382.0,
      "step": 199
    },
    {
      "entropy": 0.9682489261031151,
      "epoch": 2.567741935483871,
      "grad_norm": 0.13777530193328857,
      "learning_rate": 3.0043001045892695e-06,
      "loss": 0.967422366142273,
      "mean_token_accuracy": 0.7054171785712242,
      "num_tokens": 25681542.0,
      "step": 200
    },
    {
      "entropy": 0.9885143414139748,
      "epoch": 2.5806451612903225,
      "grad_norm": 0.14251641929149628,
      "learning_rate": 2.8383578193475315e-06,
      "loss": 0.9858547449111938,
      "mean_token_accuracy": 0.7019123136997223,
      "num_tokens": 25811593.0,
      "step": 201
    },
    {
      "entropy": 0.9749475717544556,
      "epoch": 2.5935483870967744,
      "grad_norm": 0.13928896188735962,
      "learning_rate": 2.676853546260791e-06,
      "loss": 0.9732533693313599,
      "mean_token_accuracy": 0.7038539499044418,
      "num_tokens": 25940206.0,
      "step": 202
    },
    {
      "entropy": 0.9768851473927498,
      "epoch": 2.606451612903226,
      "grad_norm": 0.1388377547264099,
      "learning_rate": 2.5198196276040782e-06,
      "loss": 0.9758725166320801,
      "mean_token_accuracy": 0.7040918990969658,
      "num_tokens": 26068569.0,
      "step": 203
    },
    {
      "entropy": 0.9973954185843468,
      "epoch": 2.6193548387096772,
      "grad_norm": 0.13934050500392914,
      "learning_rate": 2.3672875104350844e-06,
      "loss": 1.0013916492462158,
      "mean_token_accuracy": 0.6959126144647598,
      "num_tokens": 26198474.0,
      "step": 204
    },
    {
      "entropy": 0.9525961130857468,
      "epoch": 2.632258064516129,
      "grad_norm": 0.14118824899196625,
      "learning_rate": 2.219287740296605e-06,
      "loss": 0.9482128620147705,
      "mean_token_accuracy": 0.7096298635005951,
      "num_tokens": 26326152.0,
      "step": 205
    },
    {
      "entropy": 0.9498050287365913,
      "epoch": 2.6451612903225805,
      "grad_norm": 0.13799995183944702,
      "learning_rate": 2.075849955099668e-06,
      "loss": 0.9489595890045166,
      "mean_token_accuracy": 0.7115756347775459,
      "num_tokens": 26455939.0,
      "step": 206
    },
    {
      "entropy": 0.9663607403635979,
      "epoch": 2.6580645161290324,
      "grad_norm": 0.13604725897312164,
      "learning_rate": 1.937002879188285e-06,
      "loss": 0.9674727916717529,
      "mean_token_accuracy": 0.7058031111955643,
      "num_tokens": 26585781.0,
      "step": 207
    },
    {
      "entropy": 0.9505886733531952,
      "epoch": 2.670967741935484,
      "grad_norm": 0.138391375541687,
      "learning_rate": 1.8027743175872664e-06,
      "loss": 0.9568888545036316,
      "mean_token_accuracy": 0.7089990824460983,
      "num_tokens": 26711858.0,
      "step": 208
    },
    {
      "entropy": 0.9782652631402016,
      "epoch": 2.6838709677419352,
      "grad_norm": 0.13821113109588623,
      "learning_rate": 1.673191150434067e-06,
      "loss": 0.9766222238540649,
      "mean_token_accuracy": 0.702571265399456,
      "num_tokens": 26840657.0,
      "step": 209
    },
    {
      "entropy": 0.9674790501594543,
      "epoch": 2.696774193548387,
      "grad_norm": 0.13552363216876984,
      "learning_rate": 1.5482793275958529e-06,
      "loss": 0.962854266166687,
      "mean_token_accuracy": 0.7062871009111404,
      "num_tokens": 26970561.0,
      "step": 210
    },
    {
      "entropy": 0.9813898727297783,
      "epoch": 2.709677419354839,
      "grad_norm": 0.1387540102005005,
      "learning_rate": 1.428063863472895e-06,
      "loss": 0.9765456914901733,
      "mean_token_accuracy": 0.7029862403869629,
      "num_tokens": 27099808.0,
      "step": 211
    },
    {
      "entropy": 0.9833763092756271,
      "epoch": 2.7225806451612904,
      "grad_norm": 0.13949647545814514,
      "learning_rate": 1.312568831989258e-06,
      "loss": 0.9785590171813965,
      "mean_token_accuracy": 0.7032964751124382,
      "num_tokens": 27227187.0,
      "step": 212
    },
    {
      "entropy": 0.9646457061171532,
      "epoch": 2.735483870967742,
      "grad_norm": 0.1389310359954834,
      "learning_rate": 1.201817361771837e-06,
      "loss": 0.9615578651428223,
      "mean_token_accuracy": 0.709533579647541,
      "num_tokens": 27356425.0,
      "step": 213
    },
    {
      "entropy": 0.9672388881444931,
      "epoch": 2.7483870967741937,
      "grad_norm": 0.13714219629764557,
      "learning_rate": 1.095831631518729e-06,
      "loss": 0.9644992351531982,
      "mean_token_accuracy": 0.7045844122767448,
      "num_tokens": 27485158.0,
      "step": 214
    },
    {
      "entropy": 0.9853720366954803,
      "epoch": 2.761290322580645,
      "grad_norm": 0.13979892432689667,
      "learning_rate": 9.946328655577624e-07,
      "loss": 0.9827160835266113,
      "mean_token_accuracy": 0.7024488374590874,
      "num_tokens": 27613898.0,
      "step": 215
    },
    {
      "entropy": 0.9637850895524025,
      "epoch": 2.774193548387097,
      "grad_norm": 0.1382351517677307,
      "learning_rate": 8.98241329596236e-07,
      "loss": 0.9596185088157654,
      "mean_token_accuracy": 0.7085541039705276,
      "num_tokens": 27742345.0,
      "step": 216
    },
    {
      "entropy": 0.9817634969949722,
      "epoch": 2.7870967741935484,
      "grad_norm": 0.13956692814826965,
      "learning_rate": 8.066763266625282e-07,
      "loss": 0.9818617105484009,
      "mean_token_accuracy": 0.702887199819088,
      "num_tokens": 27871525.0,
      "step": 217
    },
    {
      "entropy": 0.9589646533131599,
      "epoch": 2.8,
      "grad_norm": 0.13627667725086212,
      "learning_rate": 7.199561932405952e-07,
      "loss": 0.9577062129974365,
      "mean_token_accuracy": 0.7097993195056915,
      "num_tokens": 28000658.0,
      "step": 218
    },
    {
      "entropy": 0.9998961240053177,
      "epoch": 2.8129032258064517,
      "grad_norm": 0.1377549171447754,
      "learning_rate": 6.380982955979192e-07,
      "loss": 0.9931961894035339,
      "mean_token_accuracy": 0.6985419914126396,
      "num_tokens": 28129810.0,
      "step": 219
    },
    {
      "entropy": 0.9305282756686211,
      "epoch": 2.825806451612903,
      "grad_norm": 0.13711713254451752,
      "learning_rate": 5.611190263078464e-07,
      "loss": 0.9275537729263306,
      "mean_token_accuracy": 0.7170769050717354,
      "num_tokens": 28259343.0,
      "step": 220
    },
    {
      "entropy": 0.9768733754754066,
      "epoch": 2.838709677419355,
      "grad_norm": 0.13856461644172668,
      "learning_rate": 4.890338009668316e-07,
      "loss": 0.970474123954773,
      "mean_token_accuracy": 0.704489178955555,
      "num_tokens": 28389274.0,
      "step": 221
    },
    {
      "entropy": 0.9825925827026367,
      "epoch": 2.8516129032258064,
      "grad_norm": 0.14358536899089813,
      "learning_rate": 4.2185705510739415e-07,
      "loss": 0.9805284738540649,
      "mean_token_accuracy": 0.7027609497308731,
      "num_tokens": 28518414.0,
      "step": 222
    },
    {
      "entropy": 0.9739194586873055,
      "epoch": 2.864516129032258,
      "grad_norm": 0.13605843484401703,
      "learning_rate": 3.5960224130728857e-07,
      "loss": 0.9668415784835815,
      "mean_token_accuracy": 0.7055631726980209,
      "num_tokens": 28648178.0,
      "step": 223
    },
    {
      "entropy": 0.9678709432482719,
      "epoch": 2.8774193548387097,
      "grad_norm": 0.13870655000209808,
      "learning_rate": 3.0228182649555126e-07,
      "loss": 0.9622299075126648,
      "mean_token_accuracy": 0.707713283598423,
      "num_tokens": 28777130.0,
      "step": 224
    },
    {
      "entropy": 0.9655280485749245,
      "epoch": 2.8903225806451616,
      "grad_norm": 0.1391085535287857,
      "learning_rate": 2.499072894559057e-07,
      "loss": 0.9706152677536011,
      "mean_token_accuracy": 0.707145519554615,
      "num_tokens": 28906037.0,
      "step": 225
    },
    {
      "entropy": 0.9672367423772812,
      "epoch": 2.903225806451613,
      "grad_norm": 0.13682921230793,
      "learning_rate": 2.0248911852807916e-07,
      "loss": 0.9662670493125916,
      "mean_token_accuracy": 0.7067179307341576,
      "num_tokens": 29036308.0,
      "step": 226
    },
    {
      "entropy": 0.9703774452209473,
      "epoch": 2.9161290322580644,
      "grad_norm": 0.14008858799934387,
      "learning_rate": 1.6003680950742728e-07,
      "loss": 0.9692631959915161,
      "mean_token_accuracy": 0.7054286599159241,
      "num_tokens": 29165368.0,
      "step": 227
    },
    {
      "entropy": 0.9861516654491425,
      "epoch": 2.9290322580645163,
      "grad_norm": 0.13893291354179382,
      "learning_rate": 1.2255886374334946e-07,
      "loss": 0.9812330007553101,
      "mean_token_accuracy": 0.7022589892148972,
      "num_tokens": 29295274.0,
      "step": 228
    },
    {
      "entropy": 0.9563653618097305,
      "epoch": 2.9419354838709677,
      "grad_norm": 0.134530171751976,
      "learning_rate": 9.006278643683696e-08,
      "loss": 0.9512709379196167,
      "mean_token_accuracy": 0.7103432267904282,
      "num_tokens": 29424540.0,
      "step": 229
    },
    {
      "entropy": 0.9658341854810715,
      "epoch": 2.9548387096774196,
      "grad_norm": 0.14166159927845,
      "learning_rate": 6.255508513750841e-08,
      "loss": 0.9626063108444214,
      "mean_token_accuracy": 0.7082437798380852,
      "num_tokens": 29553990.0,
      "step": 230
    },
    {
      "entropy": 0.9772498980164528,
      "epoch": 2.967741935483871,
      "grad_norm": 0.13874590396881104,
      "learning_rate": 4.004126844042444e-08,
      "loss": 0.9715877771377563,
      "mean_token_accuracy": 0.7052456364035606,
      "num_tokens": 29683419.0,
      "step": 231
    },
    {
      "entropy": 0.9945057854056358,
      "epoch": 2.9806451612903224,
      "grad_norm": 0.14169548451900482,
      "learning_rate": 2.2525844882964607e-08,
      "loss": 0.98978590965271,
      "mean_token_accuracy": 0.7005648165941238,
      "num_tokens": 29813015.0,
      "step": 232
    },
    {
      "entropy": 0.9796391725540161,
      "epoch": 2.9935483870967743,
      "grad_norm": 0.13883884251117706,
      "learning_rate": 1.0012322041960676e-08,
      "loss": 0.9724296927452087,
      "mean_token_accuracy": 0.7050853669643402,
      "num_tokens": 29942783.0,
      "step": 233
    },
    {
      "entropy": 0.9650707542896271,
      "epoch": 3.0,
      "grad_norm": 0.19312243163585663,
      "learning_rate": 2.503205831277944e-09,
      "loss": 0.9573902487754822,
      "mean_token_accuracy": 0.7067301273345947,
      "num_tokens": 29991313.0,
      "step": 234
    }
  ],
  "logging_steps": 1,
  "max_steps": 234,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.2579498480993567e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}