{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 804,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1365940570831299,
      "epoch": 0.0037313432835820895,
      "grad_norm": 1.602013349533081,
      "learning_rate": 0.0002,
      "loss": 2.4923055171966553,
      "mean_token_accuracy": 0.534668356180191,
      "num_tokens": 16376.0,
      "step": 1
    },
    {
      "entropy": 1.237722486257553,
      "epoch": 0.007462686567164179,
      "grad_norm": 1.462984323501587,
      "learning_rate": 0.0002,
      "loss": 2.143401622772217,
      "mean_token_accuracy": 0.5692279934883118,
      "num_tokens": 32699.0,
      "step": 2
    },
    {
      "entropy": 1.4051735997200012,
      "epoch": 0.011194029850746268,
      "grad_norm": 1.1621999740600586,
      "learning_rate": 0.0002,
      "loss": 1.7229645252227783,
      "mean_token_accuracy": 0.5906880646944046,
      "num_tokens": 49192.0,
      "step": 3
    },
    {
      "entropy": 1.366858810186386,
      "epoch": 0.014925373134328358,
      "grad_norm": 0.9362263083457947,
      "learning_rate": 0.0002,
      "loss": 1.3983738422393799,
      "mean_token_accuracy": 0.6390129625797272,
      "num_tokens": 65489.0,
      "step": 4
    },
    {
      "entropy": 1.3347687423229218,
      "epoch": 0.018656716417910446,
      "grad_norm": 1.0186233520507812,
      "learning_rate": 0.0002,
      "loss": 1.2787163257598877,
      "mean_token_accuracy": 0.646392360329628,
      "num_tokens": 82113.0,
      "step": 5
    },
    {
      "entropy": 1.216700166463852,
      "epoch": 0.022388059701492536,
      "grad_norm": 0.5289033651351929,
      "learning_rate": 0.0002,
      "loss": 1.1512999534606934,
      "mean_token_accuracy": 0.6695324331521988,
      "num_tokens": 98392.0,
      "step": 6
    },
    {
      "entropy": 1.1541274189949036,
      "epoch": 0.026119402985074626,
      "grad_norm": 0.3854961097240448,
      "learning_rate": 0.0002,
      "loss": 1.0803110599517822,
      "mean_token_accuracy": 0.6676419377326965,
      "num_tokens": 114971.0,
      "step": 7
    },
    {
      "entropy": 1.06028351187706,
      "epoch": 0.029850746268656716,
      "grad_norm": 0.4441620111465454,
      "learning_rate": 0.0002,
      "loss": 1.0047900676727295,
      "mean_token_accuracy": 0.6877810806035995,
      "num_tokens": 131252.0,
      "step": 8
    },
    {
      "entropy": 0.9763151109218597,
      "epoch": 0.033582089552238806,
      "grad_norm": 0.524559736251831,
      "learning_rate": 0.0002,
      "loss": 0.9419859647750854,
      "mean_token_accuracy": 0.6997241675853729,
      "num_tokens": 147572.0,
      "step": 9
    },
    {
      "entropy": 0.9578081965446472,
      "epoch": 0.03731343283582089,
      "grad_norm": 0.4574742913246155,
      "learning_rate": 0.0002,
      "loss": 0.8855704069137573,
      "mean_token_accuracy": 0.7059380114078522,
      "num_tokens": 163875.0,
      "step": 10
    },
    {
      "entropy": 0.9111354798078537,
      "epoch": 0.041044776119402986,
      "grad_norm": 0.4545396864414215,
      "learning_rate": 0.0002,
      "loss": 0.8169403672218323,
      "mean_token_accuracy": 0.7208303064107895,
      "num_tokens": 179993.0,
      "step": 11
    },
    {
      "entropy": 0.8632445931434631,
      "epoch": 0.04477611940298507,
      "grad_norm": 0.4060600697994232,
      "learning_rate": 0.0002,
      "loss": 0.7969988584518433,
      "mean_token_accuracy": 0.7173275947570801,
      "num_tokens": 196505.0,
      "step": 12
    },
    {
      "entropy": 0.7818893641233444,
      "epoch": 0.048507462686567165,
      "grad_norm": 7.450481414794922,
      "learning_rate": 0.0002,
      "loss": 0.7918776273727417,
      "mean_token_accuracy": 0.7221231460571289,
      "num_tokens": 213075.0,
      "step": 13
    },
    {
      "entropy": 0.7156526148319244,
      "epoch": 0.05223880597014925,
      "grad_norm": 0.5087403059005737,
      "learning_rate": 0.0002,
      "loss": 0.7229958176612854,
      "mean_token_accuracy": 0.7386146485805511,
      "num_tokens": 229493.0,
      "step": 14
    },
    {
      "entropy": 0.7127820700407028,
      "epoch": 0.055970149253731345,
      "grad_norm": 0.3936116397380829,
      "learning_rate": 0.0002,
      "loss": 0.7001363635063171,
      "mean_token_accuracy": 0.7414001673460007,
      "num_tokens": 245887.0,
      "step": 15
    },
    {
      "entropy": 0.6817566901445389,
      "epoch": 0.05970149253731343,
      "grad_norm": 0.37246227264404297,
      "learning_rate": 0.0002,
      "loss": 0.6792046427726746,
      "mean_token_accuracy": 0.743701845407486,
      "num_tokens": 262208.0,
      "step": 16
    },
    {
      "entropy": 0.6676452159881592,
      "epoch": 0.06343283582089553,
      "grad_norm": 0.4071228504180908,
      "learning_rate": 0.0002,
      "loss": 0.6639598608016968,
      "mean_token_accuracy": 0.745721697807312,
      "num_tokens": 278765.0,
      "step": 17
    },
    {
      "entropy": 0.6656136214733124,
      "epoch": 0.06716417910447761,
      "grad_norm": 0.3188609778881073,
      "learning_rate": 0.0002,
      "loss": 0.6388291120529175,
      "mean_token_accuracy": 0.7558866590261459,
      "num_tokens": 294783.0,
      "step": 18
    },
    {
      "entropy": 0.6872565746307373,
      "epoch": 0.0708955223880597,
      "grad_norm": 10.33668327331543,
      "learning_rate": 0.0002,
      "loss": 0.6665768027305603,
      "mean_token_accuracy": 0.7409394383430481,
      "num_tokens": 311094.0,
      "step": 19
    },
    {
      "entropy": 0.6658274680376053,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.34565994143486023,
      "learning_rate": 0.0002,
      "loss": 0.6393768191337585,
      "mean_token_accuracy": 0.7514362335205078,
      "num_tokens": 327592.0,
      "step": 20
    },
    {
      "entropy": 0.6434597074985504,
      "epoch": 0.07835820895522388,
      "grad_norm": 0.3687341809272766,
      "learning_rate": 0.0002,
      "loss": 0.6277059316635132,
      "mean_token_accuracy": 0.757160946726799,
      "num_tokens": 343858.0,
      "step": 21
    },
    {
      "entropy": 0.6272915452718735,
      "epoch": 0.08208955223880597,
      "grad_norm": 0.27142569422721863,
      "learning_rate": 0.0002,
      "loss": 0.6190376281738281,
      "mean_token_accuracy": 0.7603178918361664,
      "num_tokens": 360211.0,
      "step": 22
    },
    {
      "entropy": 0.5943922847509384,
      "epoch": 0.08582089552238806,
      "grad_norm": 0.2911827564239502,
      "learning_rate": 0.0002,
      "loss": 0.5969835519790649,
      "mean_token_accuracy": 0.7661998867988586,
      "num_tokens": 376407.0,
      "step": 23
    },
    {
      "entropy": 0.6011459678411484,
      "epoch": 0.08955223880597014,
      "grad_norm": 0.2360728681087494,
      "learning_rate": 0.0002,
      "loss": 0.5967482328414917,
      "mean_token_accuracy": 0.7657501399517059,
      "num_tokens": 392950.0,
      "step": 24
    },
    {
      "entropy": 0.60666723549366,
      "epoch": 0.09328358208955224,
      "grad_norm": 0.26025551557540894,
      "learning_rate": 0.0002,
      "loss": 0.6145684123039246,
      "mean_token_accuracy": 0.7592208236455917,
      "num_tokens": 409024.0,
      "step": 25
    },
    {
      "entropy": 0.6086552143096924,
      "epoch": 0.09701492537313433,
      "grad_norm": 0.24351224303245544,
      "learning_rate": 0.0002,
      "loss": 0.6115347146987915,
      "mean_token_accuracy": 0.7585291415452957,
      "num_tokens": 425644.0,
      "step": 26
    },
    {
      "entropy": 0.5964881032705307,
      "epoch": 0.10074626865671642,
      "grad_norm": 0.2827071249485016,
      "learning_rate": 0.0002,
      "loss": 0.5943769812583923,
      "mean_token_accuracy": 0.7679151445627213,
      "num_tokens": 442271.0,
      "step": 27
    },
    {
      "entropy": 0.598346620798111,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.22368261218070984,
      "learning_rate": 0.0002,
      "loss": 0.5878981947898865,
      "mean_token_accuracy": 0.7693336755037308,
      "num_tokens": 458551.0,
      "step": 28
    },
    {
      "entropy": 0.6149618774652481,
      "epoch": 0.10820895522388059,
      "grad_norm": 0.2550245225429535,
      "learning_rate": 0.0002,
      "loss": 0.594581127166748,
      "mean_token_accuracy": 0.765242800116539,
      "num_tokens": 475089.0,
      "step": 29
    },
    {
      "entropy": 0.5989812016487122,
      "epoch": 0.11194029850746269,
      "grad_norm": 0.2275591641664505,
      "learning_rate": 0.0002,
      "loss": 0.5958423614501953,
      "mean_token_accuracy": 0.7631504237651825,
      "num_tokens": 491505.0,
      "step": 30
    },
    {
      "entropy": 0.57822684943676,
      "epoch": 0.11567164179104478,
      "grad_norm": 0.24225932359695435,
      "learning_rate": 0.0002,
      "loss": 0.5797860026359558,
      "mean_token_accuracy": 0.7708296924829483,
      "num_tokens": 508228.0,
      "step": 31
    },
    {
      "entropy": 0.5828859359025955,
      "epoch": 0.11940298507462686,
      "grad_norm": 0.26869437098503113,
      "learning_rate": 0.0002,
      "loss": 0.5919716954231262,
      "mean_token_accuracy": 0.7665105015039444,
      "num_tokens": 524850.0,
      "step": 32
    },
    {
      "entropy": 0.5556454360485077,
      "epoch": 0.12313432835820895,
      "grad_norm": 0.19256635010242462,
      "learning_rate": 0.0002,
      "loss": 0.5622851252555847,
      "mean_token_accuracy": 0.7769578099250793,
      "num_tokens": 541178.0,
      "step": 33
    },
    {
      "entropy": 0.5680376440286636,
      "epoch": 0.12686567164179105,
      "grad_norm": 0.26926833391189575,
      "learning_rate": 0.0002,
      "loss": 0.5755432844161987,
      "mean_token_accuracy": 0.7709646821022034,
      "num_tokens": 557762.0,
      "step": 34
    },
    {
      "entropy": 0.5800794959068298,
      "epoch": 0.13059701492537312,
      "grad_norm": 0.178944930434227,
      "learning_rate": 0.0002,
      "loss": 0.5851141214370728,
      "mean_token_accuracy": 0.7640193998813629,
      "num_tokens": 574323.0,
      "step": 35
    },
    {
      "entropy": 0.5877208411693573,
      "epoch": 0.13432835820895522,
      "grad_norm": 0.24447374045848846,
      "learning_rate": 0.0002,
      "loss": 0.5876041650772095,
      "mean_token_accuracy": 0.7654121220111847,
      "num_tokens": 590537.0,
      "step": 36
    },
    {
      "entropy": 0.5542350262403488,
      "epoch": 0.13805970149253732,
      "grad_norm": 0.1687035709619522,
      "learning_rate": 0.0002,
      "loss": 0.555046021938324,
      "mean_token_accuracy": 0.7743215262889862,
      "num_tokens": 606826.0,
      "step": 37
    },
    {
      "entropy": 0.5715240687131882,
      "epoch": 0.1417910447761194,
      "grad_norm": 0.2247871309518814,
      "learning_rate": 0.0002,
      "loss": 0.5669185519218445,
      "mean_token_accuracy": 0.771597757935524,
      "num_tokens": 623434.0,
      "step": 38
    },
    {
      "entropy": 0.5593602806329727,
      "epoch": 0.1455223880597015,
      "grad_norm": 0.194087415933609,
      "learning_rate": 0.0002,
      "loss": 0.5648448467254639,
      "mean_token_accuracy": 0.7727399319410324,
      "num_tokens": 639797.0,
      "step": 39
    },
    {
      "entropy": 0.546249121427536,
      "epoch": 0.14925373134328357,
      "grad_norm": 0.19142727553844452,
      "learning_rate": 0.0002,
      "loss": 0.5554565191268921,
      "mean_token_accuracy": 0.7755379229784012,
      "num_tokens": 656207.0,
      "step": 40
    },
    {
      "entropy": 0.5677359402179718,
      "epoch": 0.15298507462686567,
      "grad_norm": 0.18955466151237488,
      "learning_rate": 0.0002,
      "loss": 0.5726029872894287,
      "mean_token_accuracy": 0.7708010077476501,
      "num_tokens": 672607.0,
      "step": 41
    },
    {
      "entropy": 0.5781030654907227,
      "epoch": 0.15671641791044777,
      "grad_norm": 0.21186234056949615,
      "learning_rate": 0.0002,
      "loss": 0.5711483955383301,
      "mean_token_accuracy": 0.76866614818573,
      "num_tokens": 688771.0,
      "step": 42
    },
    {
      "entropy": 0.5682426989078522,
      "epoch": 0.16044776119402984,
      "grad_norm": 0.18833644688129425,
      "learning_rate": 0.0002,
      "loss": 0.5552631616592407,
      "mean_token_accuracy": 0.7775518447160721,
      "num_tokens": 705172.0,
      "step": 43
    },
    {
      "entropy": 0.568386510014534,
      "epoch": 0.16417910447761194,
      "grad_norm": 0.20943182706832886,
      "learning_rate": 0.0002,
      "loss": 0.5554555058479309,
      "mean_token_accuracy": 0.7801986634731293,
      "num_tokens": 721417.0,
      "step": 44
    },
    {
      "entropy": 0.5791953951120377,
      "epoch": 0.16791044776119404,
      "grad_norm": 0.15064512193202972,
      "learning_rate": 0.0002,
      "loss": 0.5714805126190186,
      "mean_token_accuracy": 0.771676167845726,
      "num_tokens": 737881.0,
      "step": 45
    },
    {
      "entropy": 0.569206565618515,
      "epoch": 0.17164179104477612,
      "grad_norm": 0.199662446975708,
      "learning_rate": 0.0002,
      "loss": 0.583835780620575,
      "mean_token_accuracy": 0.7657788544893265,
      "num_tokens": 753856.0,
      "step": 46
    },
    {
      "entropy": 0.5452835559844971,
      "epoch": 0.17537313432835822,
      "grad_norm": 0.1934865415096283,
      "learning_rate": 0.0002,
      "loss": 0.5569155216217041,
      "mean_token_accuracy": 0.7749973982572556,
      "num_tokens": 770030.0,
      "step": 47
    },
    {
      "entropy": 0.5560597032308578,
      "epoch": 0.1791044776119403,
      "grad_norm": 0.18504005670547485,
      "learning_rate": 0.0002,
      "loss": 0.5633100867271423,
      "mean_token_accuracy": 0.7732223272323608,
      "num_tokens": 786200.0,
      "step": 48
    },
    {
      "entropy": 0.5514497309923172,
      "epoch": 0.1828358208955224,
      "grad_norm": 0.14916570484638214,
      "learning_rate": 0.0002,
      "loss": 0.5459753274917603,
      "mean_token_accuracy": 0.78175088763237,
      "num_tokens": 802595.0,
      "step": 49
    },
    {
      "entropy": 0.5639061480760574,
      "epoch": 0.1865671641791045,
      "grad_norm": 0.16385433077812195,
      "learning_rate": 0.0002,
      "loss": 0.5620872974395752,
      "mean_token_accuracy": 0.7769079953432083,
      "num_tokens": 819107.0,
      "step": 50
    },
    {
      "entropy": 0.5763161331415176,
      "epoch": 0.19029850746268656,
      "grad_norm": 0.16482000052928925,
      "learning_rate": 0.0002,
      "loss": 0.573357343673706,
      "mean_token_accuracy": 0.769774317741394,
      "num_tokens": 835540.0,
      "step": 51
    },
    {
      "entropy": 0.5532338619232178,
      "epoch": 0.19402985074626866,
      "grad_norm": 0.1456056386232376,
      "learning_rate": 0.0002,
      "loss": 0.5551443099975586,
      "mean_token_accuracy": 0.7754847854375839,
      "num_tokens": 852010.0,
      "step": 52
    },
    {
      "entropy": 0.5501885116100311,
      "epoch": 0.19776119402985073,
      "grad_norm": 0.1471378654241562,
      "learning_rate": 0.0002,
      "loss": 0.547960638999939,
      "mean_token_accuracy": 0.7818176299333572,
      "num_tokens": 868397.0,
      "step": 53
    },
    {
      "entropy": 0.5568799078464508,
      "epoch": 0.20149253731343283,
      "grad_norm": 0.14303237199783325,
      "learning_rate": 0.0002,
      "loss": 0.5620520710945129,
      "mean_token_accuracy": 0.7757244110107422,
      "num_tokens": 884557.0,
      "step": 54
    },
    {
      "entropy": 0.5843401253223419,
      "epoch": 0.20522388059701493,
      "grad_norm": 0.15758250653743744,
      "learning_rate": 0.0002,
      "loss": 0.5874170064926147,
      "mean_token_accuracy": 0.767481729388237,
      "num_tokens": 901005.0,
      "step": 55
    },
    {
      "entropy": 0.5804745703935623,
      "epoch": 0.208955223880597,
      "grad_norm": 0.16883815824985504,
      "learning_rate": 0.0002,
      "loss": 0.5828210711479187,
      "mean_token_accuracy": 0.7666165083646774,
      "num_tokens": 917284.0,
      "step": 56
    },
    {
      "entropy": 0.5736707895994186,
      "epoch": 0.2126865671641791,
      "grad_norm": 0.1380566656589508,
      "learning_rate": 0.0002,
      "loss": 0.5679526925086975,
      "mean_token_accuracy": 0.7719616591930389,
      "num_tokens": 933769.0,
      "step": 57
    },
    {
      "entropy": 0.5437489748001099,
      "epoch": 0.21641791044776118,
      "grad_norm": 0.19315236806869507,
      "learning_rate": 0.0002,
      "loss": 0.5477432012557983,
      "mean_token_accuracy": 0.7795370072126389,
      "num_tokens": 950113.0,
      "step": 58
    },
    {
      "entropy": 0.5593954026699066,
      "epoch": 0.22014925373134328,
      "grad_norm": 0.11987848579883575,
      "learning_rate": 0.0002,
      "loss": 0.5614272356033325,
      "mean_token_accuracy": 0.7741457670927048,
      "num_tokens": 966763.0,
      "step": 59
    },
    {
      "entropy": 0.5602267235517502,
      "epoch": 0.22388059701492538,
      "grad_norm": 0.13175706565380096,
      "learning_rate": 0.0002,
      "loss": 0.5578678846359253,
      "mean_token_accuracy": 0.7752770483493805,
      "num_tokens": 983226.0,
      "step": 60
    },
    {
      "entropy": 0.5553758591413498,
      "epoch": 0.22761194029850745,
      "grad_norm": 0.16444535553455353,
      "learning_rate": 0.0002,
      "loss": 0.5669548511505127,
      "mean_token_accuracy": 0.7726316303014755,
      "num_tokens": 999642.0,
      "step": 61
    },
    {
      "entropy": 0.571015253663063,
      "epoch": 0.23134328358208955,
      "grad_norm": 0.14549528062343597,
      "learning_rate": 0.0002,
      "loss": 0.5684346556663513,
      "mean_token_accuracy": 0.7699839025735855,
      "num_tokens": 1015964.0,
      "step": 62
    },
    {
      "entropy": 0.5603262484073639,
      "epoch": 0.23507462686567165,
      "grad_norm": 0.1592172235250473,
      "learning_rate": 0.0002,
      "loss": 0.558625340461731,
      "mean_token_accuracy": 0.7741718143224716,
      "num_tokens": 1032148.0,
      "step": 63
    },
    {
      "entropy": 0.5674083977937698,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.16129815578460693,
      "learning_rate": 0.0002,
      "loss": 0.5629262328147888,
      "mean_token_accuracy": 0.7710614651441574,
      "num_tokens": 1048670.0,
      "step": 64
    },
    {
      "entropy": 0.5649230629205704,
      "epoch": 0.24253731343283583,
      "grad_norm": 0.1565246731042862,
      "learning_rate": 0.0002,
      "loss": 0.5614427328109741,
      "mean_token_accuracy": 0.7731173038482666,
      "num_tokens": 1065136.0,
      "step": 65
    },
    {
      "entropy": 0.5438065677881241,
      "epoch": 0.2462686567164179,
      "grad_norm": 0.14227062463760376,
      "learning_rate": 0.0002,
      "loss": 0.5480823516845703,
      "mean_token_accuracy": 0.7801248729228973,
      "num_tokens": 1081654.0,
      "step": 66
    },
    {
      "entropy": 0.5650558322668076,
      "epoch": 0.25,
      "grad_norm": 0.14410872757434845,
      "learning_rate": 0.0002,
      "loss": 0.5681664943695068,
      "mean_token_accuracy": 0.7684987038373947,
      "num_tokens": 1097916.0,
      "step": 67
    },
    {
      "entropy": 0.5579129904508591,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.15842370688915253,
      "learning_rate": 0.0002,
      "loss": 0.5658090114593506,
      "mean_token_accuracy": 0.7706008553504944,
      "num_tokens": 1114061.0,
      "step": 68
    },
    {
      "entropy": 0.556731291115284,
      "epoch": 0.2574626865671642,
      "grad_norm": 0.14555150270462036,
      "learning_rate": 0.0002,
      "loss": 0.5542295575141907,
      "mean_token_accuracy": 0.7769580781459808,
      "num_tokens": 1130798.0,
      "step": 69
    },
    {
      "entropy": 0.5532342046499252,
      "epoch": 0.26119402985074625,
      "grad_norm": 0.14868402481079102,
      "learning_rate": 0.0002,
      "loss": 0.5550898313522339,
      "mean_token_accuracy": 0.7759661972522736,
      "num_tokens": 1147327.0,
      "step": 70
    },
    {
      "entropy": 0.5650774538516998,
      "epoch": 0.26492537313432835,
      "grad_norm": 0.11625094711780548,
      "learning_rate": 0.0002,
      "loss": 0.5690693259239197,
      "mean_token_accuracy": 0.7702993303537369,
      "num_tokens": 1163575.0,
      "step": 71
    },
    {
      "entropy": 0.5545080304145813,
      "epoch": 0.26865671641791045,
      "grad_norm": 0.14182746410369873,
      "learning_rate": 0.0002,
      "loss": 0.5567182302474976,
      "mean_token_accuracy": 0.7749036997556686,
      "num_tokens": 1180058.0,
      "step": 72
    },
    {
      "entropy": 0.5616336166858673,
      "epoch": 0.27238805970149255,
      "grad_norm": 0.15870311856269836,
      "learning_rate": 0.0002,
      "loss": 0.5611894130706787,
      "mean_token_accuracy": 0.7715721428394318,
      "num_tokens": 1196628.0,
      "step": 73
    },
    {
      "entropy": 0.5505207479000092,
      "epoch": 0.27611940298507465,
      "grad_norm": 0.1406475454568863,
      "learning_rate": 0.0002,
      "loss": 0.5505172610282898,
      "mean_token_accuracy": 0.7769232988357544,
      "num_tokens": 1213406.0,
      "step": 74
    },
    {
      "entropy": 0.5437770038843155,
      "epoch": 0.2798507462686567,
      "grad_norm": 0.16017386317253113,
      "learning_rate": 0.0002,
      "loss": 0.5520621538162231,
      "mean_token_accuracy": 0.7741838246583939,
      "num_tokens": 1230009.0,
      "step": 75
    },
    {
      "entropy": 0.5606179535388947,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.14126917719841003,
      "learning_rate": 0.0002,
      "loss": 0.5628935098648071,
      "mean_token_accuracy": 0.7716085761785507,
      "num_tokens": 1246199.0,
      "step": 76
    },
    {
      "entropy": 0.5349563509225845,
      "epoch": 0.2873134328358209,
      "grad_norm": 0.14994531869888306,
      "learning_rate": 0.0002,
      "loss": 0.5402539968490601,
      "mean_token_accuracy": 0.7839800268411636,
      "num_tokens": 1262462.0,
      "step": 77
    },
    {
      "entropy": 0.5473214983940125,
      "epoch": 0.291044776119403,
      "grad_norm": 0.15640538930892944,
      "learning_rate": 0.0002,
      "loss": 0.5467700362205505,
      "mean_token_accuracy": 0.7780201882123947,
      "num_tokens": 1278827.0,
      "step": 78
    },
    {
      "entropy": 0.5550399571657181,
      "epoch": 0.2947761194029851,
      "grad_norm": 0.128059521317482,
      "learning_rate": 0.0002,
      "loss": 0.5564315319061279,
      "mean_token_accuracy": 0.7799375802278519,
      "num_tokens": 1295190.0,
      "step": 79
    },
    {
      "entropy": 0.5476572811603546,
      "epoch": 0.29850746268656714,
      "grad_norm": 0.13417840003967285,
      "learning_rate": 0.0002,
      "loss": 0.5486627817153931,
      "mean_token_accuracy": 0.7764216214418411,
      "num_tokens": 1311469.0,
      "step": 80
    },
    {
      "entropy": 0.574133113026619,
      "epoch": 0.30223880597014924,
      "grad_norm": 0.13622140884399414,
      "learning_rate": 0.0002,
      "loss": 0.5702267289161682,
      "mean_token_accuracy": 0.7687713354825974,
      "num_tokens": 1327870.0,
      "step": 81
    },
    {
      "entropy": 0.5228832364082336,
      "epoch": 0.30597014925373134,
      "grad_norm": 0.14002114534378052,
      "learning_rate": 0.0002,
      "loss": 0.523066520690918,
      "mean_token_accuracy": 0.7886366546154022,
      "num_tokens": 1343957.0,
      "step": 82
    },
    {
      "entropy": 0.5370830297470093,
      "epoch": 0.30970149253731344,
      "grad_norm": 0.1302386373281479,
      "learning_rate": 0.0002,
      "loss": 0.5404574275016785,
      "mean_token_accuracy": 0.7827296704053879,
      "num_tokens": 1359901.0,
      "step": 83
    },
    {
      "entropy": 0.5551084727048874,
      "epoch": 0.31343283582089554,
      "grad_norm": 0.13283085823059082,
      "learning_rate": 0.0002,
      "loss": 0.5553267598152161,
      "mean_token_accuracy": 0.7749131768941879,
      "num_tokens": 1376475.0,
      "step": 84
    },
    {
      "entropy": 0.5495391339063644,
      "epoch": 0.31716417910447764,
      "grad_norm": 0.1591142863035202,
      "learning_rate": 0.0002,
      "loss": 0.5484066009521484,
      "mean_token_accuracy": 0.7790594100952148,
      "num_tokens": 1392875.0,
      "step": 85
    },
    {
      "entropy": 0.5406418591737747,
      "epoch": 0.3208955223880597,
      "grad_norm": 0.13724540174007416,
      "learning_rate": 0.0002,
      "loss": 0.5506547093391418,
      "mean_token_accuracy": 0.7753994464874268,
      "num_tokens": 1408992.0,
      "step": 86
    },
    {
      "entropy": 0.5459884703159332,
      "epoch": 0.3246268656716418,
      "grad_norm": 0.13824748992919922,
      "learning_rate": 0.0002,
      "loss": 0.5567780137062073,
      "mean_token_accuracy": 0.7764346897602081,
      "num_tokens": 1425097.0,
      "step": 87
    },
    {
      "entropy": 0.5511012375354767,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.12534891068935394,
      "learning_rate": 0.0002,
      "loss": 0.5489964485168457,
      "mean_token_accuracy": 0.775433361530304,
      "num_tokens": 1441310.0,
      "step": 88
    },
    {
      "entropy": 0.5810290426015854,
      "epoch": 0.332089552238806,
      "grad_norm": 0.1321086585521698,
      "learning_rate": 0.0002,
      "loss": 0.5752503871917725,
      "mean_token_accuracy": 0.7668759524822235,
      "num_tokens": 1457821.0,
      "step": 89
    },
    {
      "entropy": 0.5557464361190796,
      "epoch": 0.3358208955223881,
      "grad_norm": 0.13489902019500732,
      "learning_rate": 0.0002,
      "loss": 0.5530065894126892,
      "mean_token_accuracy": 0.7779003828763962,
      "num_tokens": 1474315.0,
      "step": 90
    },
    {
      "entropy": 0.5691788643598557,
      "epoch": 0.33955223880597013,
      "grad_norm": 0.1257353574037552,
      "learning_rate": 0.0002,
      "loss": 0.5645563006401062,
      "mean_token_accuracy": 0.7690286040306091,
      "num_tokens": 1490831.0,
      "step": 91
    },
    {
      "entropy": 0.5488206893205643,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.1616927832365036,
      "learning_rate": 0.0002,
      "loss": 0.5563665628433228,
      "mean_token_accuracy": 0.7770231366157532,
      "num_tokens": 1507083.0,
      "step": 92
    },
    {
      "entropy": 0.5520597100257874,
      "epoch": 0.34701492537313433,
      "grad_norm": 0.12840059399604797,
      "learning_rate": 0.0002,
      "loss": 0.5511025786399841,
      "mean_token_accuracy": 0.7762677520513535,
      "num_tokens": 1523293.0,
      "step": 93
    },
    {
      "entropy": 0.5596688687801361,
      "epoch": 0.35074626865671643,
      "grad_norm": 0.12729911506175995,
      "learning_rate": 0.0002,
      "loss": 0.5565716028213501,
      "mean_token_accuracy": 0.7760972827672958,
      "num_tokens": 1539685.0,
      "step": 94
    },
    {
      "entropy": 0.5472247302532196,
      "epoch": 0.35447761194029853,
      "grad_norm": 0.13684000074863434,
      "learning_rate": 0.0002,
      "loss": 0.5559186339378357,
      "mean_token_accuracy": 0.7763453274965286,
      "num_tokens": 1556176.0,
      "step": 95
    },
    {
      "entropy": 0.5479343831539154,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.16693443059921265,
      "learning_rate": 0.0002,
      "loss": 0.5495245456695557,
      "mean_token_accuracy": 0.7785529047250748,
      "num_tokens": 1572474.0,
      "step": 96
    },
    {
      "entropy": 0.5312614589929581,
      "epoch": 0.3619402985074627,
      "grad_norm": 0.15308798849582672,
      "learning_rate": 0.0002,
      "loss": 0.5377439260482788,
      "mean_token_accuracy": 0.7839500159025192,
      "num_tokens": 1588586.0,
      "step": 97
    },
    {
      "entropy": 0.5481617450714111,
      "epoch": 0.3656716417910448,
      "grad_norm": 0.1866571456193924,
      "learning_rate": 0.0002,
      "loss": 0.5532342195510864,
      "mean_token_accuracy": 0.7758291810750961,
      "num_tokens": 1604895.0,
      "step": 98
    },
    {
      "entropy": 0.5557571947574615,
      "epoch": 0.3694029850746269,
      "grad_norm": 0.12315784394741058,
      "learning_rate": 0.0002,
      "loss": 0.5560829639434814,
      "mean_token_accuracy": 0.7777059823274612,
      "num_tokens": 1621183.0,
      "step": 99
    },
    {
      "entropy": 0.5654265284538269,
      "epoch": 0.373134328358209,
      "grad_norm": 0.16337135434150696,
      "learning_rate": 0.0002,
      "loss": 0.558884859085083,
      "mean_token_accuracy": 0.7751797139644623,
      "num_tokens": 1637656.0,
      "step": 100
    },
    {
      "entropy": 0.5454747378826141,
      "epoch": 0.376865671641791,
      "grad_norm": 0.1486552357673645,
      "learning_rate": 0.0002,
      "loss": 0.5444817543029785,
      "mean_token_accuracy": 0.7836427241563797,
      "num_tokens": 1653856.0,
      "step": 101
    },
    {
      "entropy": 0.5437199473381042,
      "epoch": 0.3805970149253731,
      "grad_norm": 0.14405834674835205,
      "learning_rate": 0.0002,
      "loss": 0.5505189299583435,
      "mean_token_accuracy": 0.7780970484018326,
      "num_tokens": 1670078.0,
      "step": 102
    },
    {
      "entropy": 0.5362537056207657,
      "epoch": 0.3843283582089552,
      "grad_norm": 0.16666138172149658,
      "learning_rate": 0.0002,
      "loss": 0.5472300052642822,
      "mean_token_accuracy": 0.7768100798130035,
      "num_tokens": 1686425.0,
      "step": 103
    },
    {
      "entropy": 0.5530032366514206,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.13793827593326569,
      "learning_rate": 0.0002,
      "loss": 0.5608469247817993,
      "mean_token_accuracy": 0.7765886187553406,
      "num_tokens": 1702784.0,
      "step": 104
    },
    {
      "entropy": 0.5522157400846481,
      "epoch": 0.3917910447761194,
      "grad_norm": 0.1373043954372406,
      "learning_rate": 0.0002,
      "loss": 0.5503177046775818,
      "mean_token_accuracy": 0.7784782648086548,
      "num_tokens": 1718926.0,
      "step": 105
    },
    {
      "entropy": 0.5485086441040039,
      "epoch": 0.39552238805970147,
      "grad_norm": 0.14003051817417145,
      "learning_rate": 0.0002,
      "loss": 0.5415142774581909,
      "mean_token_accuracy": 0.7796619236469269,
      "num_tokens": 1735215.0,
      "step": 106
    },
    {
      "entropy": 0.5622719824314117,
      "epoch": 0.39925373134328357,
      "grad_norm": 0.13935396075248718,
      "learning_rate": 0.0002,
      "loss": 0.553518533706665,
      "mean_token_accuracy": 0.7777030766010284,
      "num_tokens": 1751843.0,
      "step": 107
    },
    {
      "entropy": 0.5561539381742477,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.1298242211341858,
      "learning_rate": 0.0002,
      "loss": 0.5502764582633972,
      "mean_token_accuracy": 0.7798024117946625,
      "num_tokens": 1768182.0,
      "step": 108
    },
    {
      "entropy": 0.5464132577180862,
      "epoch": 0.40671641791044777,
      "grad_norm": 0.1633938103914261,
      "learning_rate": 0.0002,
      "loss": 0.5578470230102539,
      "mean_token_accuracy": 0.778321161866188,
      "num_tokens": 1784569.0,
      "step": 109
    },
    {
      "entropy": 0.5351661443710327,
      "epoch": 0.41044776119402987,
      "grad_norm": 0.1518043428659439,
      "learning_rate": 0.0002,
      "loss": 0.546484112739563,
      "mean_token_accuracy": 0.7804382294416428,
      "num_tokens": 1801161.0,
      "step": 110
    },
    {
      "entropy": 0.5500597059726715,
      "epoch": 0.4141791044776119,
      "grad_norm": 0.12739060819149017,
      "learning_rate": 0.0002,
      "loss": 0.5549509525299072,
      "mean_token_accuracy": 0.7765515446662903,
      "num_tokens": 1817578.0,
      "step": 111
    },
    {
      "entropy": 0.54776930809021,
      "epoch": 0.417910447761194,
      "grad_norm": 0.12031728774309158,
      "learning_rate": 0.0002,
      "loss": 0.5454729795455933,
      "mean_token_accuracy": 0.7791832387447357,
      "num_tokens": 1834035.0,
      "step": 112
    },
    {
      "entropy": 0.5651962906122208,
      "epoch": 0.4216417910447761,
      "grad_norm": 0.14656467735767365,
      "learning_rate": 0.0002,
      "loss": 0.5598954558372498,
      "mean_token_accuracy": 0.7755678594112396,
      "num_tokens": 1850514.0,
      "step": 113
    },
    {
      "entropy": 0.5535902380943298,
      "epoch": 0.4253731343283582,
      "grad_norm": 0.12962885200977325,
      "learning_rate": 0.0002,
      "loss": 0.5499719381332397,
      "mean_token_accuracy": 0.7795900106430054,
      "num_tokens": 1866717.0,
      "step": 114
    },
    {
      "entropy": 0.552685096859932,
      "epoch": 0.4291044776119403,
      "grad_norm": 0.12470420449972153,
      "learning_rate": 0.0002,
      "loss": 0.5580195188522339,
      "mean_token_accuracy": 0.7737019062042236,
      "num_tokens": 1882995.0,
      "step": 115
    },
    {
      "entropy": 0.5454623401165009,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.1673581898212433,
      "learning_rate": 0.0002,
      "loss": 0.5584871172904968,
      "mean_token_accuracy": 0.7741757184267044,
      "num_tokens": 1899039.0,
      "step": 116
    },
    {
      "entropy": 0.5481576770544052,
      "epoch": 0.43656716417910446,
      "grad_norm": 0.12181615084409714,
      "learning_rate": 0.0002,
      "loss": 0.5469480752944946,
      "mean_token_accuracy": 0.7805405110120773,
      "num_tokens": 1915338.0,
      "step": 117
    },
    {
      "entropy": 0.5433652251958847,
      "epoch": 0.44029850746268656,
      "grad_norm": 0.1260775923728943,
      "learning_rate": 0.0002,
      "loss": 0.5456545352935791,
      "mean_token_accuracy": 0.7788949012756348,
      "num_tokens": 1931657.0,
      "step": 118
    },
    {
      "entropy": 0.5526310950517654,
      "epoch": 0.44402985074626866,
      "grad_norm": 0.13798080384731293,
      "learning_rate": 0.0002,
      "loss": 0.5505106449127197,
      "mean_token_accuracy": 0.7771039605140686,
      "num_tokens": 1948105.0,
      "step": 119
    },
    {
      "entropy": 0.5459735095500946,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.12050134688615799,
      "learning_rate": 0.0002,
      "loss": 0.5456549525260925,
      "mean_token_accuracy": 0.7821092307567596,
      "num_tokens": 1964472.0,
      "step": 120
    },
    {
      "entropy": 0.552826315164566,
      "epoch": 0.45149253731343286,
      "grad_norm": 0.13367635011672974,
      "learning_rate": 0.0002,
      "loss": 0.561984121799469,
      "mean_token_accuracy": 0.7741702795028687,
      "num_tokens": 1980870.0,
      "step": 121
    },
    {
      "entropy": 0.5679444372653961,
      "epoch": 0.4552238805970149,
      "grad_norm": 0.13436280190944672,
      "learning_rate": 0.0002,
      "loss": 0.571191668510437,
      "mean_token_accuracy": 0.7690508514642715,
      "num_tokens": 1997312.0,
      "step": 122
    },
    {
      "entropy": 0.5477713048458099,
      "epoch": 0.458955223880597,
      "grad_norm": 0.13245144486427307,
      "learning_rate": 0.0002,
      "loss": 0.5457987189292908,
      "mean_token_accuracy": 0.7796986401081085,
      "num_tokens": 2013538.0,
      "step": 123
    },
    {
      "entropy": 0.5546025335788727,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.1400025635957718,
      "learning_rate": 0.0002,
      "loss": 0.549485981464386,
      "mean_token_accuracy": 0.7795898914337158,
      "num_tokens": 2029913.0,
      "step": 124
    },
    {
      "entropy": 0.551468089222908,
      "epoch": 0.4664179104477612,
      "grad_norm": 0.14955715835094452,
      "learning_rate": 0.0002,
      "loss": 0.5536458492279053,
      "mean_token_accuracy": 0.7741625159978867,
      "num_tokens": 2046110.0,
      "step": 125
    },
    {
      "entropy": 0.5553492605686188,
      "epoch": 0.4701492537313433,
      "grad_norm": 0.15557385981082916,
      "learning_rate": 0.0002,
      "loss": 0.564170241355896,
      "mean_token_accuracy": 0.7710831165313721,
      "num_tokens": 2062481.0,
      "step": 126
    },
    {
      "entropy": 0.5552105009555817,
      "epoch": 0.47388059701492535,
      "grad_norm": 0.13015641272068024,
      "learning_rate": 0.0002,
      "loss": 0.5556001663208008,
      "mean_token_accuracy": 0.7744580060243607,
      "num_tokens": 2078979.0,
      "step": 127
    },
    {
      "entropy": 0.5516892224550247,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.13372646272182465,
      "learning_rate": 0.0002,
      "loss": 0.5488641858100891,
      "mean_token_accuracy": 0.7791122794151306,
      "num_tokens": 2095278.0,
      "step": 128
    },
    {
      "entropy": 0.5511497408151627,
      "epoch": 0.48134328358208955,
      "grad_norm": 0.13201062381267548,
      "learning_rate": 0.0002,
      "loss": 0.5487321615219116,
      "mean_token_accuracy": 0.777876615524292,
      "num_tokens": 2111580.0,
      "step": 129
    },
    {
      "entropy": 0.5581526607275009,
      "epoch": 0.48507462686567165,
      "grad_norm": 0.13924257457256317,
      "learning_rate": 0.0002,
      "loss": 0.5638423562049866,
      "mean_token_accuracy": 0.7699771374464035,
      "num_tokens": 2127900.0,
      "step": 130
    },
    {
      "entropy": 0.542765274643898,
      "epoch": 0.48880597014925375,
      "grad_norm": 0.12968650460243225,
      "learning_rate": 0.0002,
      "loss": 0.5457286834716797,
      "mean_token_accuracy": 0.7792125195264816,
      "num_tokens": 2144176.0,
      "step": 131
    },
    {
      "entropy": 0.5610632598400116,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.12429090589284897,
      "learning_rate": 0.0002,
      "loss": 0.5593065023422241,
      "mean_token_accuracy": 0.7749785333871841,
      "num_tokens": 2160383.0,
      "step": 132
    },
    {
      "entropy": 0.5488081872463226,
      "epoch": 0.4962686567164179,
      "grad_norm": 0.13297075033187866,
      "learning_rate": 0.0002,
      "loss": 0.5489633679389954,
      "mean_token_accuracy": 0.7763438075780869,
      "num_tokens": 2177038.0,
      "step": 133
    },
    {
      "entropy": 0.5711653679609299,
      "epoch": 0.5,
      "grad_norm": 0.15128269791603088,
      "learning_rate": 0.0002,
      "loss": 0.5680216550827026,
      "mean_token_accuracy": 0.7711322754621506,
      "num_tokens": 2193404.0,
      "step": 134
    },
    {
      "entropy": 0.5383014529943466,
      "epoch": 0.503731343283582,
      "grad_norm": 0.12188736349344254,
      "learning_rate": 0.0002,
      "loss": 0.5379375219345093,
      "mean_token_accuracy": 0.7828169614076614,
      "num_tokens": 2209492.0,
      "step": 135
    },
    {
      "entropy": 0.5449882298707962,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.1486186385154724,
      "learning_rate": 0.0002,
      "loss": 0.5506329536437988,
      "mean_token_accuracy": 0.7776465862989426,
      "num_tokens": 2225797.0,
      "step": 136
    },
    {
      "entropy": 0.5486281961202621,
      "epoch": 0.5111940298507462,
      "grad_norm": 0.1376088708639145,
      "learning_rate": 0.0002,
      "loss": 0.5550296306610107,
      "mean_token_accuracy": 0.7759877741336823,
      "num_tokens": 2242003.0,
      "step": 137
    },
    {
      "entropy": 0.5254313051700592,
      "epoch": 0.5149253731343284,
      "grad_norm": 0.11921165883541107,
      "learning_rate": 0.0002,
      "loss": 0.5298187732696533,
      "mean_token_accuracy": 0.7868426591157913,
      "num_tokens": 2258630.0,
      "step": 138
    },
    {
      "entropy": 0.5666594952344894,
      "epoch": 0.5186567164179104,
      "grad_norm": 0.13457800447940826,
      "learning_rate": 0.0002,
      "loss": 0.5722187757492065,
      "mean_token_accuracy": 0.7680158317089081,
      "num_tokens": 2274959.0,
      "step": 139
    },
    {
      "entropy": 0.5818119496107101,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.12537291646003723,
      "learning_rate": 0.0002,
      "loss": 0.5758311748504639,
      "mean_token_accuracy": 0.7662911713123322,
      "num_tokens": 2291409.0,
      "step": 140
    },
    {
      "entropy": 0.5623303800821304,
      "epoch": 0.5261194029850746,
      "grad_norm": 0.10797441005706787,
      "learning_rate": 0.0002,
      "loss": 0.5589613318443298,
      "mean_token_accuracy": 0.7719256728887558,
      "num_tokens": 2307880.0,
      "step": 141
    },
    {
      "entropy": 0.5529690682888031,
      "epoch": 0.5298507462686567,
      "grad_norm": 0.1298912912607193,
      "learning_rate": 0.0002,
      "loss": 0.5505614280700684,
      "mean_token_accuracy": 0.7753282338380814,
      "num_tokens": 2324143.0,
      "step": 142
    },
    {
      "entropy": 0.5444576740264893,
      "epoch": 0.5335820895522388,
      "grad_norm": 0.13483697175979614,
      "learning_rate": 0.0002,
      "loss": 0.5511869788169861,
      "mean_token_accuracy": 0.7745974808931351,
      "num_tokens": 2340373.0,
      "step": 143
    },
    {
      "entropy": 0.5505483150482178,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.13404449820518494,
      "learning_rate": 0.0002,
      "loss": 0.5640406012535095,
      "mean_token_accuracy": 0.770884320139885,
      "num_tokens": 2356651.0,
      "step": 144
    },
    {
      "entropy": 0.5237122476100922,
      "epoch": 0.5410447761194029,
      "grad_norm": 0.1442001610994339,
      "learning_rate": 0.0002,
      "loss": 0.5356744527816772,
      "mean_token_accuracy": 0.7825886309146881,
      "num_tokens": 2372785.0,
      "step": 145
    },
    {
      "entropy": 0.5595418810844421,
      "epoch": 0.5447761194029851,
      "grad_norm": 0.13395285606384277,
      "learning_rate": 0.0002,
      "loss": 0.5614413619041443,
      "mean_token_accuracy": 0.7732272893190384,
      "num_tokens": 2388966.0,
      "step": 146
    },
    {
      "entropy": 0.5672930181026459,
      "epoch": 0.5485074626865671,
      "grad_norm": 0.12952914834022522,
      "learning_rate": 0.0002,
      "loss": 0.5526295304298401,
      "mean_token_accuracy": 0.7793028354644775,
      "num_tokens": 2405504.0,
      "step": 147
    },
    {
      "entropy": 0.5611097514629364,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.12538692355155945,
      "learning_rate": 0.0002,
      "loss": 0.5492501258850098,
      "mean_token_accuracy": 0.7777923196554184,
      "num_tokens": 2421848.0,
      "step": 148
    },
    {
      "entropy": 0.5485778898000717,
      "epoch": 0.5559701492537313,
      "grad_norm": 0.11349925398826599,
      "learning_rate": 0.0002,
      "loss": 0.5477732419967651,
      "mean_token_accuracy": 0.7781182080507278,
      "num_tokens": 2438298.0,
      "step": 149
    },
    {
      "entropy": 0.540116623044014,
      "epoch": 0.5597014925373134,
      "grad_norm": 0.1568114310503006,
      "learning_rate": 0.0002,
      "loss": 0.5526154637336731,
      "mean_token_accuracy": 0.7777610123157501,
      "num_tokens": 2454556.0,
      "step": 150
    },
    {
      "entropy": 0.5628165453672409,
      "epoch": 0.5634328358208955,
      "grad_norm": 0.1343757063150406,
      "learning_rate": 0.0002,
      "loss": 0.5676834583282471,
      "mean_token_accuracy": 0.7719947844743729,
      "num_tokens": 2471114.0,
      "step": 151
    },
    {
      "entropy": 0.5437122881412506,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.12374638766050339,
      "learning_rate": 0.0002,
      "loss": 0.5384945273399353,
      "mean_token_accuracy": 0.7826568484306335,
      "num_tokens": 2487597.0,
      "step": 152
    },
    {
      "entropy": 0.5438667386770248,
      "epoch": 0.5708955223880597,
      "grad_norm": 0.11150524020195007,
      "learning_rate": 0.0002,
      "loss": 0.5401822328567505,
      "mean_token_accuracy": 0.7825683057308197,
      "num_tokens": 2503935.0,
      "step": 153
    },
    {
      "entropy": 0.5377064347267151,
      "epoch": 0.5746268656716418,
      "grad_norm": 0.15077364444732666,
      "learning_rate": 0.0002,
      "loss": 0.5432950258255005,
      "mean_token_accuracy": 0.7775296866893768,
      "num_tokens": 2520134.0,
      "step": 154
    },
    {
      "entropy": 0.5510450154542923,
      "epoch": 0.5783582089552238,
      "grad_norm": 0.13458377122879028,
      "learning_rate": 0.0002,
      "loss": 0.5545709133148193,
      "mean_token_accuracy": 0.7761582881212234,
      "num_tokens": 2536630.0,
      "step": 155
    },
    {
      "entropy": 0.5244418382644653,
      "epoch": 0.582089552238806,
      "grad_norm": 0.1248108372092247,
      "learning_rate": 0.0002,
      "loss": 0.5255698561668396,
      "mean_token_accuracy": 0.7885231226682663,
      "num_tokens": 2553051.0,
      "step": 156
    },
    {
      "entropy": 0.542046919465065,
      "epoch": 0.585820895522388,
      "grad_norm": 0.1421440839767456,
      "learning_rate": 0.0002,
      "loss": 0.5432825684547424,
      "mean_token_accuracy": 0.7801544964313507,
      "num_tokens": 2569390.0,
      "step": 157
    },
    {
      "entropy": 0.5532049536705017,
      "epoch": 0.5895522388059702,
      "grad_norm": 0.1295875906944275,
      "learning_rate": 0.0002,
      "loss": 0.5504398941993713,
      "mean_token_accuracy": 0.7763772010803223,
      "num_tokens": 2585713.0,
      "step": 158
    },
    {
      "entropy": 0.5496713221073151,
      "epoch": 0.5932835820895522,
      "grad_norm": 0.1107042133808136,
      "learning_rate": 0.0002,
      "loss": 0.5504273772239685,
      "mean_token_accuracy": 0.7768300622701645,
      "num_tokens": 2602386.0,
      "step": 159
    },
    {
      "entropy": 0.5469995141029358,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.13561351597309113,
      "learning_rate": 0.0002,
      "loss": 0.5535402297973633,
      "mean_token_accuracy": 0.7760646194219589,
      "num_tokens": 2618862.0,
      "step": 160
    },
    {
      "entropy": 0.5322236269712448,
      "epoch": 0.6007462686567164,
      "grad_norm": 0.12066421657800674,
      "learning_rate": 0.0002,
      "loss": 0.530032753944397,
      "mean_token_accuracy": 0.7871640175580978,
      "num_tokens": 2635191.0,
      "step": 161
    },
    {
      "entropy": 0.5541785657405853,
      "epoch": 0.6044776119402985,
      "grad_norm": 0.11959829926490784,
      "learning_rate": 0.0002,
      "loss": 0.5545668601989746,
      "mean_token_accuracy": 0.7756617814302444,
      "num_tokens": 2651549.0,
      "step": 162
    },
    {
      "entropy": 0.5544012784957886,
      "epoch": 0.6082089552238806,
      "grad_norm": 0.12402931600809097,
      "learning_rate": 0.0002,
      "loss": 0.5454179048538208,
      "mean_token_accuracy": 0.7791251689195633,
      "num_tokens": 2667916.0,
      "step": 163
    },
    {
      "entropy": 0.5556594282388687,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.11536160111427307,
      "learning_rate": 0.0002,
      "loss": 0.5553296804428101,
      "mean_token_accuracy": 0.7747591286897659,
      "num_tokens": 2684398.0,
      "step": 164
    },
    {
      "entropy": 0.5535193085670471,
      "epoch": 0.6156716417910447,
      "grad_norm": 0.12544436752796173,
      "learning_rate": 0.0002,
      "loss": 0.5558847188949585,
      "mean_token_accuracy": 0.7752297967672348,
      "num_tokens": 2700926.0,
      "step": 165
    },
    {
      "entropy": 0.5272173285484314,
      "epoch": 0.6194029850746269,
      "grad_norm": 0.14834193885326385,
      "learning_rate": 0.0002,
      "loss": 0.541037917137146,
      "mean_token_accuracy": 0.7798037081956863,
      "num_tokens": 2717243.0,
      "step": 166
    },
    {
      "entropy": 0.526879757642746,
      "epoch": 0.6231343283582089,
      "grad_norm": 0.14949624240398407,
      "learning_rate": 0.0002,
      "loss": 0.5444039702415466,
      "mean_token_accuracy": 0.7814223617315292,
      "num_tokens": 2733292.0,
      "step": 167
    },
    {
      "entropy": 0.5524511188268661,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.12368033081293106,
      "learning_rate": 0.0002,
      "loss": 0.5564784407615662,
      "mean_token_accuracy": 0.775758758187294,
      "num_tokens": 2749879.0,
      "step": 168
    },
    {
      "entropy": 0.5634000301361084,
      "epoch": 0.6305970149253731,
      "grad_norm": 0.11312047392129898,
      "learning_rate": 0.0002,
      "loss": 0.5528282523155212,
      "mean_token_accuracy": 0.7778469622135162,
      "num_tokens": 2766302.0,
      "step": 169
    },
    {
      "entropy": 0.5637902468442917,
      "epoch": 0.6343283582089553,
      "grad_norm": 0.11576911062002182,
      "learning_rate": 0.0002,
      "loss": 0.5519281029701233,
      "mean_token_accuracy": 0.7786077857017517,
      "num_tokens": 2782590.0,
      "step": 170
    },
    {
      "entropy": 0.5396268516778946,
      "epoch": 0.6380597014925373,
      "grad_norm": 0.15072329342365265,
      "learning_rate": 0.0002,
      "loss": 0.5385048985481262,
      "mean_token_accuracy": 0.7841583043336868,
      "num_tokens": 2799148.0,
      "step": 171
    },
    {
      "entropy": 0.5509935170412064,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.12073057144880295,
      "learning_rate": 0.0002,
      "loss": 0.551585853099823,
      "mean_token_accuracy": 0.7757870107889175,
      "num_tokens": 2815282.0,
      "step": 172
    },
    {
      "entropy": 0.5379815995693207,
      "epoch": 0.6455223880597015,
      "grad_norm": 0.14362798631191254,
      "learning_rate": 0.0002,
      "loss": 0.5470185279846191,
      "mean_token_accuracy": 0.7795873135328293,
      "num_tokens": 2831543.0,
      "step": 173
    },
    {
      "entropy": 0.5499899685382843,
      "epoch": 0.6492537313432836,
      "grad_norm": 0.14761203527450562,
      "learning_rate": 0.0002,
      "loss": 0.547420859336853,
      "mean_token_accuracy": 0.7754547894001007,
      "num_tokens": 2847913.0,
      "step": 174
    },
    {
      "entropy": 0.5515707582235336,
      "epoch": 0.6529850746268657,
      "grad_norm": 0.11489994823932648,
      "learning_rate": 0.0002,
      "loss": 0.5548341870307922,
      "mean_token_accuracy": 0.7761587500572205,
      "num_tokens": 2864397.0,
      "step": 175
    },
    {
      "entropy": 0.5610131025314331,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.14579513669013977,
      "learning_rate": 0.0002,
      "loss": 0.5653943419456482,
      "mean_token_accuracy": 0.76800337433815,
      "num_tokens": 2880757.0,
      "step": 176
    },
    {
      "entropy": 0.5525858253240585,
      "epoch": 0.6604477611940298,
      "grad_norm": 0.12836626172065735,
      "learning_rate": 0.0002,
      "loss": 0.5537892580032349,
      "mean_token_accuracy": 0.7772515565156937,
      "num_tokens": 2897086.0,
      "step": 177
    },
    {
      "entropy": 0.545110210776329,
      "epoch": 0.664179104477612,
      "grad_norm": 0.12226562201976776,
      "learning_rate": 0.0002,
      "loss": 0.5358925461769104,
      "mean_token_accuracy": 0.7812047451734543,
      "num_tokens": 2913360.0,
      "step": 178
    },
    {
      "entropy": 0.5489350408315659,
      "epoch": 0.667910447761194,
      "grad_norm": 0.1556328982114792,
      "learning_rate": 0.0002,
      "loss": 0.554627537727356,
      "mean_token_accuracy": 0.7744484096765518,
      "num_tokens": 2929537.0,
      "step": 179
    },
    {
      "entropy": 0.5485879927873611,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.12704777717590332,
      "learning_rate": 0.0002,
      "loss": 0.5564995408058167,
      "mean_token_accuracy": 0.7742021679878235,
      "num_tokens": 2945902.0,
      "step": 180
    },
    {
      "entropy": 0.5480811595916748,
      "epoch": 0.6753731343283582,
      "grad_norm": 0.15276439487934113,
      "learning_rate": 0.0002,
      "loss": 0.5493203997612,
      "mean_token_accuracy": 0.7759238183498383,
      "num_tokens": 2962308.0,
      "step": 181
    },
    {
      "entropy": 0.5753533393144608,
      "epoch": 0.6791044776119403,
      "grad_norm": 0.14567701518535614,
      "learning_rate": 0.0002,
      "loss": 0.5708569884300232,
      "mean_token_accuracy": 0.7680999636650085,
      "num_tokens": 2978992.0,
      "step": 182
    },
    {
      "entropy": 0.5246053040027618,
      "epoch": 0.6828358208955224,
      "grad_norm": 0.12182111293077469,
      "learning_rate": 0.0002,
      "loss": 0.5235775709152222,
      "mean_token_accuracy": 0.7877005636692047,
      "num_tokens": 2995144.0,
      "step": 183
    },
    {
      "entropy": 0.5230054184794426,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.1372102051973343,
      "learning_rate": 0.0002,
      "loss": 0.5314317941665649,
      "mean_token_accuracy": 0.7832537144422531,
      "num_tokens": 3011644.0,
      "step": 184
    },
    {
      "entropy": 0.5497380346059799,
      "epoch": 0.6902985074626866,
      "grad_norm": 0.13534119725227356,
      "learning_rate": 0.0002,
      "loss": 0.5529069304466248,
      "mean_token_accuracy": 0.7763594686985016,
      "num_tokens": 3027874.0,
      "step": 185
    },
    {
      "entropy": 0.5427591353654861,
      "epoch": 0.6940298507462687,
      "grad_norm": 0.11877557635307312,
      "learning_rate": 0.0002,
      "loss": 0.5460038185119629,
      "mean_token_accuracy": 0.779443547129631,
      "num_tokens": 3044306.0,
      "step": 186
    },
    {
      "entropy": 0.5567215830087662,
      "epoch": 0.6977611940298507,
      "grad_norm": 0.11480985581874847,
      "learning_rate": 0.0002,
      "loss": 0.5519757866859436,
      "mean_token_accuracy": 0.7766963839530945,
      "num_tokens": 3060866.0,
      "step": 187
    },
    {
      "entropy": 0.5475030839443207,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.1427774876356125,
      "learning_rate": 0.0002,
      "loss": 0.5460057258605957,
      "mean_token_accuracy": 0.778233140707016,
      "num_tokens": 3077245.0,
      "step": 188
    },
    {
      "entropy": 0.5509032011032104,
      "epoch": 0.7052238805970149,
      "grad_norm": 0.11019925028085709,
      "learning_rate": 0.0002,
      "loss": 0.5527553558349609,
      "mean_token_accuracy": 0.7769258767366409,
      "num_tokens": 3093658.0,
      "step": 189
    },
    {
      "entropy": 0.5607019364833832,
      "epoch": 0.7089552238805971,
      "grad_norm": 0.13768573105335236,
      "learning_rate": 0.0002,
      "loss": 0.5634750723838806,
      "mean_token_accuracy": 0.7734896093606949,
      "num_tokens": 3110224.0,
      "step": 190
    },
    {
      "entropy": 0.5281394273042679,
      "epoch": 0.7126865671641791,
      "grad_norm": 0.14549201726913452,
      "learning_rate": 0.0002,
      "loss": 0.5353987812995911,
      "mean_token_accuracy": 0.7837551981210709,
      "num_tokens": 3126561.0,
      "step": 191
    },
    {
      "entropy": 0.5594705939292908,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.10756427049636841,
      "learning_rate": 0.0002,
      "loss": 0.5538036823272705,
      "mean_token_accuracy": 0.7771794945001602,
      "num_tokens": 3143007.0,
      "step": 192
    },
    {
      "entropy": 0.5404242128133774,
      "epoch": 0.7201492537313433,
      "grad_norm": 0.14313453435897827,
      "learning_rate": 0.0002,
      "loss": 0.5462848544120789,
      "mean_token_accuracy": 0.7799954414367676,
      "num_tokens": 3159374.0,
      "step": 193
    },
    {
      "entropy": 0.5449017584323883,
      "epoch": 0.7238805970149254,
      "grad_norm": 0.13515527546405792,
      "learning_rate": 0.0002,
      "loss": 0.5421125888824463,
      "mean_token_accuracy": 0.7806603759527206,
      "num_tokens": 3175734.0,
      "step": 194
    },
    {
      "entropy": 0.5471158772706985,
      "epoch": 0.7276119402985075,
      "grad_norm": 0.1266445517539978,
      "learning_rate": 0.0002,
      "loss": 0.5424020290374756,
      "mean_token_accuracy": 0.7797265499830246,
      "num_tokens": 3192162.0,
      "step": 195
    },
    {
      "entropy": 0.5424392372369766,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.1297360211610794,
      "learning_rate": 0.0002,
      "loss": 0.5411167144775391,
      "mean_token_accuracy": 0.7819352447986603,
      "num_tokens": 3208373.0,
      "step": 196
    },
    {
      "entropy": 0.5653031170368195,
      "epoch": 0.7350746268656716,
      "grad_norm": 0.12269576638936996,
      "learning_rate": 0.0002,
      "loss": 0.5693800449371338,
      "mean_token_accuracy": 0.7666560560464859,
      "num_tokens": 3224534.0,
      "step": 197
    },
    {
      "entropy": 0.5433857291936874,
      "epoch": 0.7388059701492538,
      "grad_norm": 0.1304396092891693,
      "learning_rate": 0.0002,
      "loss": 0.5530256032943726,
      "mean_token_accuracy": 0.7773127257823944,
      "num_tokens": 3240748.0,
      "step": 198
    },
    {
      "entropy": 0.5586421191692352,
      "epoch": 0.7425373134328358,
      "grad_norm": 0.13297304511070251,
      "learning_rate": 0.0002,
      "loss": 0.5637133121490479,
      "mean_token_accuracy": 0.7729853093624115,
      "num_tokens": 3257207.0,
      "step": 199
    },
    {
      "entropy": 0.5477910488843918,
      "epoch": 0.746268656716418,
      "grad_norm": 0.10735364258289337,
      "learning_rate": 0.0002,
      "loss": 0.5446949005126953,
      "mean_token_accuracy": 0.7795043736696243,
      "num_tokens": 3273784.0,
      "step": 200
    },
    {
      "entropy": 0.5457836240530014,
      "epoch": 0.75,
      "grad_norm": 0.11384668946266174,
      "learning_rate": 0.0002,
      "loss": 0.5393834114074707,
      "mean_token_accuracy": 0.7845726311206818,
      "num_tokens": 3289944.0,
      "step": 201
    },
    {
      "entropy": 0.5433100759983063,
      "epoch": 0.753731343283582,
      "grad_norm": 0.11221212893724442,
      "learning_rate": 0.0002,
      "loss": 0.5432642698287964,
      "mean_token_accuracy": 0.7775585800409317,
      "num_tokens": 3306164.0,
      "step": 202
    },
    {
      "entropy": 0.5426963418722153,
      "epoch": 0.7574626865671642,
      "grad_norm": 0.10500309616327286,
      "learning_rate": 0.0002,
      "loss": 0.5477461814880371,
      "mean_token_accuracy": 0.7770811319351196,
      "num_tokens": 3322713.0,
      "step": 203
    },
    {
      "entropy": 0.545770674943924,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.11508174985647202,
      "learning_rate": 0.0002,
      "loss": 0.5490862727165222,
      "mean_token_accuracy": 0.7765889465808868,
      "num_tokens": 3338964.0,
      "step": 204
    },
    {
      "entropy": 0.5439680963754654,
      "epoch": 0.7649253731343284,
      "grad_norm": 0.10792317986488342,
      "learning_rate": 0.0002,
      "loss": 0.5460648536682129,
      "mean_token_accuracy": 0.7781796604394913,
      "num_tokens": 3355498.0,
      "step": 205
    },
    {
      "entropy": 0.5327712148427963,
      "epoch": 0.7686567164179104,
      "grad_norm": 0.11194958537817001,
      "learning_rate": 0.0002,
      "loss": 0.5305078625679016,
      "mean_token_accuracy": 0.7843922525644302,
      "num_tokens": 3371658.0,
      "step": 206
    },
    {
      "entropy": 0.5490477979183197,
      "epoch": 0.7723880597014925,
      "grad_norm": 0.11794038116931915,
      "learning_rate": 0.0002,
      "loss": 0.5417205095291138,
      "mean_token_accuracy": 0.78175488114357,
      "num_tokens": 3388034.0,
      "step": 207
    },
    {
      "entropy": 0.5323569253087044,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.11149560660123825,
      "learning_rate": 0.0002,
      "loss": 0.5369465351104736,
      "mean_token_accuracy": 0.7810990363359451,
      "num_tokens": 3404204.0,
      "step": 208
    },
    {
      "entropy": 0.5370678752660751,
      "epoch": 0.7798507462686567,
      "grad_norm": 0.12095826119184494,
      "learning_rate": 0.0002,
      "loss": 0.5443766713142395,
      "mean_token_accuracy": 0.7776306122541428,
      "num_tokens": 3420204.0,
      "step": 209
    },
    {
      "entropy": 0.5488609373569489,
      "epoch": 0.7835820895522388,
      "grad_norm": 0.11064192652702332,
      "learning_rate": 0.0002,
      "loss": 0.5494049787521362,
      "mean_token_accuracy": 0.7764303684234619,
      "num_tokens": 3436433.0,
      "step": 210
    },
    {
      "entropy": 0.5590399503707886,
      "epoch": 0.7873134328358209,
      "grad_norm": 0.12268902361392975,
      "learning_rate": 0.0002,
      "loss": 0.5598021745681763,
      "mean_token_accuracy": 0.7729885578155518,
      "num_tokens": 3452904.0,
      "step": 211
    },
    {
      "entropy": 0.5357785671949387,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.11145606637001038,
      "learning_rate": 0.0002,
      "loss": 0.5335577726364136,
      "mean_token_accuracy": 0.7851356714963913,
      "num_tokens": 3469338.0,
      "step": 212
    },
    {
      "entropy": 0.548455223441124,
      "epoch": 0.7947761194029851,
      "grad_norm": 0.12931586802005768,
      "learning_rate": 0.0002,
      "loss": 0.5509188175201416,
      "mean_token_accuracy": 0.7770029008388519,
      "num_tokens": 3486178.0,
      "step": 213
    },
    {
      "entropy": 0.5470339804887772,
      "epoch": 0.7985074626865671,
      "grad_norm": 0.11273862421512604,
      "learning_rate": 0.0002,
      "loss": 0.5529839992523193,
      "mean_token_accuracy": 0.7753448337316513,
      "num_tokens": 3502510.0,
      "step": 214
    },
    {
      "entropy": 0.5560101717710495,
      "epoch": 0.8022388059701493,
      "grad_norm": 0.11948554962873459,
      "learning_rate": 0.0002,
      "loss": 0.5573582649230957,
      "mean_token_accuracy": 0.7743163257837296,
      "num_tokens": 3518803.0,
      "step": 215
    },
    {
      "entropy": 0.5522631704807281,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.1119312047958374,
      "learning_rate": 0.0002,
      "loss": 0.5562689900398254,
      "mean_token_accuracy": 0.7734942436218262,
      "num_tokens": 3535242.0,
      "step": 216
    },
    {
      "entropy": 0.5601872354745865,
      "epoch": 0.8097014925373134,
      "grad_norm": 0.12234390527009964,
      "learning_rate": 0.0002,
      "loss": 0.555570125579834,
      "mean_token_accuracy": 0.775284618139267,
      "num_tokens": 3551570.0,
      "step": 217
    },
    {
      "entropy": 0.5386732369661331,
      "epoch": 0.8134328358208955,
      "grad_norm": 0.11854761093854904,
      "learning_rate": 0.0002,
      "loss": 0.5382176637649536,
      "mean_token_accuracy": 0.7823415547609329,
      "num_tokens": 3568283.0,
      "step": 218
    },
    {
      "entropy": 0.5387750864028931,
      "epoch": 0.8171641791044776,
      "grad_norm": 0.13731595873832703,
      "learning_rate": 0.0002,
      "loss": 0.5445034503936768,
      "mean_token_accuracy": 0.7794878035783768,
      "num_tokens": 3584639.0,
      "step": 219
    },
    {
      "entropy": 0.5346844494342804,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.11946218460798264,
      "learning_rate": 0.0002,
      "loss": 0.537401556968689,
      "mean_token_accuracy": 0.7808428555727005,
      "num_tokens": 3601189.0,
      "step": 220
    },
    {
      "entropy": 0.5355700850486755,
      "epoch": 0.8246268656716418,
      "grad_norm": 0.12383900582790375,
      "learning_rate": 0.0002,
      "loss": 0.5412791967391968,
      "mean_token_accuracy": 0.7821899056434631,
      "num_tokens": 3617791.0,
      "step": 221
    },
    {
      "entropy": 0.5349564850330353,
      "epoch": 0.8283582089552238,
      "grad_norm": 0.14098820090293884,
      "learning_rate": 0.0002,
      "loss": 0.5327227115631104,
      "mean_token_accuracy": 0.7841495871543884,
      "num_tokens": 3634239.0,
      "step": 222
    },
    {
      "entropy": 0.5680065155029297,
      "epoch": 0.832089552238806,
      "grad_norm": 0.1184559017419815,
      "learning_rate": 0.0002,
      "loss": 0.5632155537605286,
      "mean_token_accuracy": 0.7737922817468643,
      "num_tokens": 3650815.0,
      "step": 223
    },
    {
      "entropy": 0.5402304530143738,
      "epoch": 0.835820895522388,
      "grad_norm": 0.11277454346418381,
      "learning_rate": 0.0002,
      "loss": 0.5410030484199524,
      "mean_token_accuracy": 0.7816927134990692,
      "num_tokens": 3667148.0,
      "step": 224
    },
    {
      "entropy": 0.5298157036304474,
      "epoch": 0.8395522388059702,
      "grad_norm": 0.13301390409469604,
      "learning_rate": 0.0002,
      "loss": 0.5344346165657043,
      "mean_token_accuracy": 0.7849860638380051,
      "num_tokens": 3683497.0,
      "step": 225
    },
    {
      "entropy": 0.5437726527452469,
      "epoch": 0.8432835820895522,
      "grad_norm": 0.13887442648410797,
      "learning_rate": 0.0002,
      "loss": 0.5504146218299866,
      "mean_token_accuracy": 0.7761549800634384,
      "num_tokens": 3699913.0,
      "step": 226
    },
    {
      "entropy": 0.5522396713495255,
      "epoch": 0.8470149253731343,
      "grad_norm": 0.12692731618881226,
      "learning_rate": 0.0002,
      "loss": 0.5581880807876587,
      "mean_token_accuracy": 0.7727026343345642,
      "num_tokens": 3716310.0,
      "step": 227
    },
    {
      "entropy": 0.5726576149463654,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.131325826048851,
      "learning_rate": 0.0002,
      "loss": 0.5614522099494934,
      "mean_token_accuracy": 0.7715796381235123,
      "num_tokens": 3733010.0,
      "step": 228
    },
    {
      "entropy": 0.5361438393592834,
      "epoch": 0.8544776119402985,
      "grad_norm": 0.13466528058052063,
      "learning_rate": 0.0002,
      "loss": 0.5357048511505127,
      "mean_token_accuracy": 0.7809497117996216,
      "num_tokens": 3749018.0,
      "step": 229
    },
    {
      "entropy": 0.5571031421422958,
      "epoch": 0.8582089552238806,
      "grad_norm": 0.11813043057918549,
      "learning_rate": 0.0002,
      "loss": 0.5556023120880127,
      "mean_token_accuracy": 0.7761641889810562,
      "num_tokens": 3765465.0,
      "step": 230
    },
    {
      "entropy": 0.5436211079359055,
      "epoch": 0.8619402985074627,
      "grad_norm": 0.11379984021186829,
      "learning_rate": 0.0002,
      "loss": 0.54551762342453,
      "mean_token_accuracy": 0.7811069935560226,
      "num_tokens": 3781921.0,
      "step": 231
    },
    {
      "entropy": 0.5305088758468628,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.13800835609436035,
      "learning_rate": 0.0002,
      "loss": 0.5387157797813416,
      "mean_token_accuracy": 0.7823031395673752,
      "num_tokens": 3798205.0,
      "step": 232
    },
    {
      "entropy": 0.5435737818479538,
      "epoch": 0.8694029850746269,
      "grad_norm": 0.1275321990251541,
      "learning_rate": 0.0002,
      "loss": 0.5543756484985352,
      "mean_token_accuracy": 0.7763945609331131,
      "num_tokens": 3814523.0,
      "step": 233
    },
    {
      "entropy": 0.5561396330595016,
      "epoch": 0.8731343283582089,
      "grad_norm": 0.12397392094135284,
      "learning_rate": 0.0002,
      "loss": 0.5569422841072083,
      "mean_token_accuracy": 0.7748852670192719,
      "num_tokens": 3830746.0,
      "step": 234
    },
    {
      "entropy": 0.5487532168626785,
      "epoch": 0.8768656716417911,
      "grad_norm": 0.11915898323059082,
      "learning_rate": 0.0002,
      "loss": 0.5395780801773071,
      "mean_token_accuracy": 0.7798616886138916,
      "num_tokens": 3847199.0,
      "step": 235
    },
    {
      "entropy": 0.5489061176776886,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.12740027904510498,
      "learning_rate": 0.0002,
      "loss": 0.5479505062103271,
      "mean_token_accuracy": 0.7766968011856079,
      "num_tokens": 3863562.0,
      "step": 236
    },
    {
      "entropy": 0.547185480594635,
      "epoch": 0.8843283582089553,
      "grad_norm": 0.1286180168390274,
      "learning_rate": 0.0002,
      "loss": 0.5490999221801758,
      "mean_token_accuracy": 0.7770249098539352,
      "num_tokens": 3879961.0,
      "step": 237
    },
    {
      "entropy": 0.5323613584041595,
      "epoch": 0.8880597014925373,
      "grad_norm": 0.14494632184505463,
      "learning_rate": 0.0002,
      "loss": 0.5378727316856384,
      "mean_token_accuracy": 0.7837895452976227,
      "num_tokens": 3896177.0,
      "step": 238
    },
    {
      "entropy": 0.5406072735786438,
      "epoch": 0.8917910447761194,
      "grad_norm": 0.12126541882753372,
      "learning_rate": 0.0002,
      "loss": 0.5427301526069641,
      "mean_token_accuracy": 0.7774636447429657,
      "num_tokens": 3912336.0,
      "step": 239
    },
    {
      "entropy": 0.5496147274971008,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.11748510599136353,
      "learning_rate": 0.0002,
      "loss": 0.5488207340240479,
      "mean_token_accuracy": 0.7754782885313034,
      "num_tokens": 3928813.0,
      "step": 240
    },
    {
      "entropy": 0.5378351211547852,
      "epoch": 0.8992537313432836,
      "grad_norm": 0.1299649029970169,
      "learning_rate": 0.0002,
      "loss": 0.53224778175354,
      "mean_token_accuracy": 0.7841470837593079,
      "num_tokens": 3945121.0,
      "step": 241
    },
    {
      "entropy": 0.5416342914104462,
      "epoch": 0.9029850746268657,
      "grad_norm": 0.11161520332098007,
      "learning_rate": 0.0002,
      "loss": 0.543321430683136,
      "mean_token_accuracy": 0.7811318039894104,
      "num_tokens": 3961420.0,
      "step": 242
    },
    {
      "entropy": 0.5356542766094208,
      "epoch": 0.9067164179104478,
      "grad_norm": 0.13325005769729614,
      "learning_rate": 0.0002,
      "loss": 0.5471915602684021,
      "mean_token_accuracy": 0.7788051515817642,
      "num_tokens": 3978040.0,
      "step": 243
    },
    {
      "entropy": 0.5307867750525475,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.13638627529144287,
      "learning_rate": 0.0002,
      "loss": 0.5353068709373474,
      "mean_token_accuracy": 0.7818553894758224,
      "num_tokens": 3994228.0,
      "step": 244
    },
    {
      "entropy": 0.5468405783176422,
      "epoch": 0.914179104477612,
      "grad_norm": 0.10671847313642502,
      "learning_rate": 0.0002,
      "loss": 0.5478777289390564,
      "mean_token_accuracy": 0.7797961980104446,
      "num_tokens": 4010633.0,
      "step": 245
    },
    {
      "entropy": 0.5650929361581802,
      "epoch": 0.917910447761194,
      "grad_norm": 0.12254519760608673,
      "learning_rate": 0.0002,
      "loss": 0.5589828491210938,
      "mean_token_accuracy": 0.772595688700676,
      "num_tokens": 4026826.0,
      "step": 246
    },
    {
      "entropy": 0.5452629327774048,
      "epoch": 0.9216417910447762,
      "grad_norm": 0.1311108022928238,
      "learning_rate": 0.0002,
      "loss": 0.5453138947486877,
      "mean_token_accuracy": 0.7767230123281479,
      "num_tokens": 4042992.0,
      "step": 247
    },
    {
      "entropy": 0.5520199686288834,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.11586081981658936,
      "learning_rate": 0.0002,
      "loss": 0.5567255020141602,
      "mean_token_accuracy": 0.7729258686304092,
      "num_tokens": 4059303.0,
      "step": 248
    },
    {
      "entropy": 0.53436079621315,
      "epoch": 0.9291044776119403,
      "grad_norm": 0.1261804699897766,
      "learning_rate": 0.0002,
      "loss": 0.5392404794692993,
      "mean_token_accuracy": 0.7807567715644836,
      "num_tokens": 4075705.0,
      "step": 249
    },
    {
      "entropy": 0.547790601849556,
      "epoch": 0.9328358208955224,
      "grad_norm": 0.132975235581398,
      "learning_rate": 0.0002,
      "loss": 0.5582680702209473,
      "mean_token_accuracy": 0.7740280032157898,
      "num_tokens": 4092054.0,
      "step": 250
    },
    {
      "entropy": 0.5463861227035522,
      "epoch": 0.9365671641791045,
      "grad_norm": 0.11676760762929916,
      "learning_rate": 0.0002,
      "loss": 0.5412884950637817,
      "mean_token_accuracy": 0.7824486494064331,
      "num_tokens": 4108403.0,
      "step": 251
    },
    {
      "entropy": 0.548604428768158,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.10695616155862808,
      "learning_rate": 0.0002,
      "loss": 0.5415095686912537,
      "mean_token_accuracy": 0.7786484360694885,
      "num_tokens": 4124703.0,
      "step": 252
    },
    {
      "entropy": 0.5456768721342087,
      "epoch": 0.9440298507462687,
      "grad_norm": 0.11566945910453796,
      "learning_rate": 0.0002,
      "loss": 0.5451362133026123,
      "mean_token_accuracy": 0.7792640030384064,
      "num_tokens": 4141047.0,
      "step": 253
    },
    {
      "entropy": 0.5508268177509308,
      "epoch": 0.9477611940298507,
      "grad_norm": 0.11922704428434372,
      "learning_rate": 0.0002,
      "loss": 0.556510865688324,
      "mean_token_accuracy": 0.7729993313550949,
      "num_tokens": 4157239.0,
      "step": 254
    },
    {
      "entropy": 0.5267818570137024,
      "epoch": 0.9514925373134329,
      "grad_norm": 0.1438627541065216,
      "learning_rate": 0.0002,
      "loss": 0.53831547498703,
      "mean_token_accuracy": 0.779523491859436,
      "num_tokens": 4173559.0,
      "step": 255
    },
    {
      "entropy": 0.5438347607851028,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.11041932553052902,
      "learning_rate": 0.0002,
      "loss": 0.5431445837020874,
      "mean_token_accuracy": 0.7779940068721771,
      "num_tokens": 4189973.0,
      "step": 256
    },
    {
      "entropy": 0.5553971230983734,
      "epoch": 0.9589552238805971,
      "grad_norm": 0.12537860870361328,
      "learning_rate": 0.0002,
      "loss": 0.5558521747589111,
      "mean_token_accuracy": 0.7732774913311005,
      "num_tokens": 4206357.0,
      "step": 257
    },
    {
      "entropy": 0.5424372553825378,
      "epoch": 0.9626865671641791,
      "grad_norm": 0.24047061800956726,
      "learning_rate": 0.0002,
      "loss": 0.5399760603904724,
      "mean_token_accuracy": 0.7818148881196976,
      "num_tokens": 4222536.0,
      "step": 258
    },
    {
      "entropy": 0.5593102723360062,
      "epoch": 0.9664179104477612,
      "grad_norm": 0.11326711624860764,
      "learning_rate": 0.0002,
      "loss": 0.5496351718902588,
      "mean_token_accuracy": 0.7785725742578506,
      "num_tokens": 4238919.0,
      "step": 259
    },
    {
      "entropy": 0.5395167469978333,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.18676528334617615,
      "learning_rate": 0.0002,
      "loss": 0.5366888046264648,
      "mean_token_accuracy": 0.7813075333833694,
      "num_tokens": 4254998.0,
      "step": 260
    },
    {
      "entropy": 0.5502018928527832,
      "epoch": 0.9738805970149254,
      "grad_norm": 0.15462352335453033,
      "learning_rate": 0.0002,
      "loss": 0.5523666739463806,
      "mean_token_accuracy": 0.7749178558588028,
      "num_tokens": 4271391.0,
      "step": 261
    },
    {
      "entropy": 0.5499600321054459,
      "epoch": 0.9776119402985075,
      "grad_norm": 0.1702757626771927,
      "learning_rate": 0.0002,
      "loss": 0.5635721683502197,
      "mean_token_accuracy": 0.7707696408033371,
      "num_tokens": 4287573.0,
      "step": 262
    },
    {
      "entropy": 0.5500475317239761,
      "epoch": 0.9813432835820896,
      "grad_norm": 0.1249007061123848,
      "learning_rate": 0.0002,
      "loss": 0.55130934715271,
      "mean_token_accuracy": 0.7757371217012405,
      "num_tokens": 4304144.0,
      "step": 263
    },
    {
      "entropy": 0.5734081566333771,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.1454128623008728,
      "learning_rate": 0.0002,
      "loss": 0.5684080719947815,
      "mean_token_accuracy": 0.7702711224555969,
      "num_tokens": 4320751.0,
      "step": 264
    },
    {
      "entropy": 0.5342459976673126,
      "epoch": 0.9888059701492538,
      "grad_norm": 0.13839378952980042,
      "learning_rate": 0.0002,
      "loss": 0.5326778888702393,
      "mean_token_accuracy": 0.7842968255281448,
      "num_tokens": 4336967.0,
      "step": 265
    },
    {
      "entropy": 0.5551866143941879,
      "epoch": 0.9925373134328358,
      "grad_norm": 0.123750239610672,
      "learning_rate": 0.0002,
      "loss": 0.5526628494262695,
      "mean_token_accuracy": 0.7780842483043671,
      "num_tokens": 4353571.0,
      "step": 266
    },
    {
      "entropy": 0.5383746922016144,
      "epoch": 0.996268656716418,
      "grad_norm": 0.11935991048812866,
      "learning_rate": 0.0002,
      "loss": 0.5406463742256165,
      "mean_token_accuracy": 0.779077485203743,
      "num_tokens": 4369799.0,
      "step": 267
    },
    {
      "entropy": 0.5555987656116486,
      "epoch": 1.0,
      "grad_norm": 0.18210704624652863,
      "learning_rate": 0.0002,
      "loss": 0.560416579246521,
      "mean_token_accuracy": 0.7698495537042618,
      "num_tokens": 4386293.0,
      "step": 268
    },
    {
      "entropy": 0.5331346839666367,
      "epoch": 1.0037313432835822,
      "grad_norm": 0.15174758434295654,
      "learning_rate": 0.0002,
      "loss": 0.5441968441009521,
      "mean_token_accuracy": 0.7786108106374741,
      "num_tokens": 4402537.0,
      "step": 269
    },
    {
      "entropy": 0.5266798287630081,
      "epoch": 1.007462686567164,
      "grad_norm": 0.15728607773780823,
      "learning_rate": 0.0002,
      "loss": 0.5264585018157959,
      "mean_token_accuracy": 0.7850178331136703,
      "num_tokens": 4418595.0,
      "step": 270
    },
    {
      "entropy": 0.5346386432647705,
      "epoch": 1.0111940298507462,
      "grad_norm": 0.13970735669136047,
      "learning_rate": 0.0002,
      "loss": 0.532387912273407,
      "mean_token_accuracy": 0.7832404673099518,
      "num_tokens": 4434778.0,
      "step": 271
    },
    {
      "entropy": 0.5166874974966049,
      "epoch": 1.0149253731343284,
      "grad_norm": 0.15346941351890564,
      "learning_rate": 0.0002,
      "loss": 0.5192524194717407,
      "mean_token_accuracy": 0.789738655090332,
      "num_tokens": 4450843.0,
      "step": 272
    },
    {
      "entropy": 0.5390005707740784,
      "epoch": 1.0186567164179103,
      "grad_norm": 0.13393612205982208,
      "learning_rate": 0.0002,
      "loss": 0.5412783622741699,
      "mean_token_accuracy": 0.7782677412033081,
      "num_tokens": 4467483.0,
      "step": 273
    },
    {
      "entropy": 0.5343061685562134,
      "epoch": 1.0223880597014925,
      "grad_norm": 0.12204071134328842,
      "learning_rate": 0.0002,
      "loss": 0.5340715646743774,
      "mean_token_accuracy": 0.7810396254062653,
      "num_tokens": 4483686.0,
      "step": 274
    },
    {
      "entropy": 0.5360396206378937,
      "epoch": 1.0261194029850746,
      "grad_norm": 0.14761637151241302,
      "learning_rate": 0.0002,
      "loss": 0.5309356451034546,
      "mean_token_accuracy": 0.7837508767843246,
      "num_tokens": 4499797.0,
      "step": 275
    },
    {
      "entropy": 0.5468345582485199,
      "epoch": 1.0298507462686568,
      "grad_norm": 0.1375226527452469,
      "learning_rate": 0.0002,
      "loss": 0.542205274105072,
      "mean_token_accuracy": 0.7789264917373657,
      "num_tokens": 4516222.0,
      "step": 276
    },
    {
      "entropy": 0.5189393311738968,
      "epoch": 1.0335820895522387,
      "grad_norm": 0.12962016463279724,
      "learning_rate": 0.0002,
      "loss": 0.5189461708068848,
      "mean_token_accuracy": 0.7874213606119156,
      "num_tokens": 4532506.0,
      "step": 277
    },
    {
      "entropy": 0.5298452377319336,
      "epoch": 1.037313432835821,
      "grad_norm": 0.17477856576442719,
      "learning_rate": 0.0002,
      "loss": 0.5365669131278992,
      "mean_token_accuracy": 0.781813234090805,
      "num_tokens": 4548850.0,
      "step": 278
    },
    {
      "entropy": 0.538953885436058,
      "epoch": 1.041044776119403,
      "grad_norm": 0.14079952239990234,
      "learning_rate": 0.0002,
      "loss": 0.5398151278495789,
      "mean_token_accuracy": 0.78059022128582,
      "num_tokens": 4565273.0,
      "step": 279
    },
    {
      "entropy": 0.548811137676239,
      "epoch": 1.044776119402985,
      "grad_norm": 0.20743925869464874,
      "learning_rate": 0.0002,
      "loss": 0.5445910692214966,
      "mean_token_accuracy": 0.7765936106443405,
      "num_tokens": 4581785.0,
      "step": 280
    },
    {
      "entropy": 0.5578295588493347,
      "epoch": 1.0485074626865671,
      "grad_norm": 0.1094159483909607,
      "learning_rate": 0.0002,
      "loss": 0.5556608438491821,
      "mean_token_accuracy": 0.7754440009593964,
      "num_tokens": 4598178.0,
      "step": 281
    },
    {
      "entropy": 0.5403339266777039,
      "epoch": 1.0522388059701493,
      "grad_norm": 0.18311786651611328,
      "learning_rate": 0.0002,
      "loss": 0.5385863184928894,
      "mean_token_accuracy": 0.7840645611286163,
      "num_tokens": 4614421.0,
      "step": 282
    },
    {
      "entropy": 0.5173541605472565,
      "epoch": 1.0559701492537314,
      "grad_norm": 0.15705779194831848,
      "learning_rate": 0.0002,
      "loss": 0.5244402289390564,
      "mean_token_accuracy": 0.7891088575124741,
      "num_tokens": 4630913.0,
      "step": 283
    },
    {
      "entropy": 0.5370561629533768,
      "epoch": 1.0597014925373134,
      "grad_norm": 0.12554816901683807,
      "learning_rate": 0.0002,
      "loss": 0.538566529750824,
      "mean_token_accuracy": 0.7825571298599243,
      "num_tokens": 4647514.0,
      "step": 284
    },
    {
      "entropy": 0.5364163964986801,
      "epoch": 1.0634328358208955,
      "grad_norm": 0.15889275074005127,
      "learning_rate": 0.0002,
      "loss": 0.5405207872390747,
      "mean_token_accuracy": 0.7817329466342926,
      "num_tokens": 4664129.0,
      "step": 285
    },
    {
      "entropy": 0.541067436337471,
      "epoch": 1.0671641791044777,
      "grad_norm": 0.1326969712972641,
      "learning_rate": 0.0002,
      "loss": 0.5383604168891907,
      "mean_token_accuracy": 0.7826975882053375,
      "num_tokens": 4680616.0,
      "step": 286
    },
    {
      "entropy": 0.5240883231163025,
      "epoch": 1.0708955223880596,
      "grad_norm": 0.14208173751831055,
      "learning_rate": 0.0002,
      "loss": 0.5197935700416565,
      "mean_token_accuracy": 0.7880465388298035,
      "num_tokens": 4696909.0,
      "step": 287
    },
    {
      "entropy": 0.5224429070949554,
      "epoch": 1.0746268656716418,
      "grad_norm": 0.13456787168979645,
      "learning_rate": 0.0002,
      "loss": 0.5270411372184753,
      "mean_token_accuracy": 0.7891874015331268,
      "num_tokens": 4713178.0,
      "step": 288
    },
    {
      "entropy": 0.5469739735126495,
      "epoch": 1.078358208955224,
      "grad_norm": 0.13231691718101501,
      "learning_rate": 0.0002,
      "loss": 0.5516743659973145,
      "mean_token_accuracy": 0.7771703153848648,
      "num_tokens": 4729684.0,
      "step": 289
    },
    {
      "entropy": 0.5239145085215569,
      "epoch": 1.0820895522388059,
      "grad_norm": 0.17734842002391815,
      "learning_rate": 0.0002,
      "loss": 0.5298153162002563,
      "mean_token_accuracy": 0.7842872738838196,
      "num_tokens": 4746046.0,
      "step": 290
    },
    {
      "entropy": 0.5460294783115387,
      "epoch": 1.085820895522388,
      "grad_norm": 0.1188027411699295,
      "learning_rate": 0.0002,
      "loss": 0.547230064868927,
      "mean_token_accuracy": 0.7760306596755981,
      "num_tokens": 4762499.0,
      "step": 291
    },
    {
      "entropy": 0.5370855629444122,
      "epoch": 1.0895522388059702,
      "grad_norm": 0.16826674342155457,
      "learning_rate": 0.0002,
      "loss": 0.5313410758972168,
      "mean_token_accuracy": 0.7849053144454956,
      "num_tokens": 4778994.0,
      "step": 292
    },
    {
      "entropy": 0.5488353371620178,
      "epoch": 1.0932835820895523,
      "grad_norm": 0.14899083971977234,
      "learning_rate": 0.0002,
      "loss": 0.5461880564689636,
      "mean_token_accuracy": 0.7785388231277466,
      "num_tokens": 4795307.0,
      "step": 293
    },
    {
      "entropy": 0.5511489063501358,
      "epoch": 1.0970149253731343,
      "grad_norm": 0.13867032527923584,
      "learning_rate": 0.0002,
      "loss": 0.5523304343223572,
      "mean_token_accuracy": 0.774916797876358,
      "num_tokens": 4811717.0,
      "step": 294
    },
    {
      "entropy": 0.5315069109201431,
      "epoch": 1.1007462686567164,
      "grad_norm": 0.1389411985874176,
      "learning_rate": 0.0002,
      "loss": 0.5338950753211975,
      "mean_token_accuracy": 0.7827065885066986,
      "num_tokens": 4828027.0,
      "step": 295
    },
    {
      "entropy": 0.5153618454933167,
      "epoch": 1.1044776119402986,
      "grad_norm": 0.12945152819156647,
      "learning_rate": 0.0002,
      "loss": 0.523125171661377,
      "mean_token_accuracy": 0.7871790528297424,
      "num_tokens": 4844166.0,
      "step": 296
    },
    {
      "entropy": 0.5159175395965576,
      "epoch": 1.1082089552238805,
      "grad_norm": 0.14918646216392517,
      "learning_rate": 0.0002,
      "loss": 0.5206066370010376,
      "mean_token_accuracy": 0.7866144627332687,
      "num_tokens": 4860365.0,
      "step": 297
    },
    {
      "entropy": 0.5476884096860886,
      "epoch": 1.1119402985074627,
      "grad_norm": 0.14630824327468872,
      "learning_rate": 0.0002,
      "loss": 0.5424180626869202,
      "mean_token_accuracy": 0.7797351181507111,
      "num_tokens": 4876899.0,
      "step": 298
    },
    {
      "entropy": 0.5478092730045319,
      "epoch": 1.1156716417910448,
      "grad_norm": 0.12391816824674606,
      "learning_rate": 0.0002,
      "loss": 0.5417909026145935,
      "mean_token_accuracy": 0.7809217274188995,
      "num_tokens": 4893174.0,
      "step": 299
    },
    {
      "entropy": 0.5405594408512115,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.13673117756843567,
      "learning_rate": 0.0002,
      "loss": 0.5294885039329529,
      "mean_token_accuracy": 0.785841554403305,
      "num_tokens": 4909274.0,
      "step": 300
    },
    {
      "entropy": 0.5359626412391663,
      "epoch": 1.123134328358209,
      "grad_norm": 0.12731675803661346,
      "learning_rate": 0.0002,
      "loss": 0.5376664400100708,
      "mean_token_accuracy": 0.7818749994039536,
      "num_tokens": 4925404.0,
      "step": 301
    },
    {
      "entropy": 0.5178922563791275,
      "epoch": 1.126865671641791,
      "grad_norm": 0.12766136229038239,
      "learning_rate": 0.0002,
      "loss": 0.5272905230522156,
      "mean_token_accuracy": 0.7878039330244064,
      "num_tokens": 4941912.0,
      "step": 302
    },
    {
      "entropy": 0.5387950092554092,
      "epoch": 1.1305970149253732,
      "grad_norm": 0.16448946297168732,
      "learning_rate": 0.0002,
      "loss": 0.554564356803894,
      "mean_token_accuracy": 0.7765425443649292,
      "num_tokens": 4958272.0,
      "step": 303
    },
    {
      "entropy": 0.5365511924028397,
      "epoch": 1.1343283582089552,
      "grad_norm": 0.12706293165683746,
      "learning_rate": 0.0002,
      "loss": 0.5372544527053833,
      "mean_token_accuracy": 0.7818904668092728,
      "num_tokens": 4974546.0,
      "step": 304
    },
    {
      "entropy": 0.5357211828231812,
      "epoch": 1.1380597014925373,
      "grad_norm": 0.12718632817268372,
      "learning_rate": 0.0002,
      "loss": 0.5251268148422241,
      "mean_token_accuracy": 0.7874895185232162,
      "num_tokens": 4991175.0,
      "step": 305
    },
    {
      "entropy": 0.5316093266010284,
      "epoch": 1.1417910447761195,
      "grad_norm": 0.13668428361415863,
      "learning_rate": 0.0002,
      "loss": 0.5296725034713745,
      "mean_token_accuracy": 0.7854830622673035,
      "num_tokens": 5007650.0,
      "step": 306
    },
    {
      "entropy": 0.5251090079545975,
      "epoch": 1.1455223880597014,
      "grad_norm": 0.14311696588993073,
      "learning_rate": 0.0002,
      "loss": 0.5251290798187256,
      "mean_token_accuracy": 0.7889134287834167,
      "num_tokens": 5023882.0,
      "step": 307
    },
    {
      "entropy": 0.5323427468538284,
      "epoch": 1.1492537313432836,
      "grad_norm": 0.17292356491088867,
      "learning_rate": 0.0002,
      "loss": 0.5480212569236755,
      "mean_token_accuracy": 0.7765524536371231,
      "num_tokens": 5040164.0,
      "step": 308
    },
    {
      "entropy": 0.5279139578342438,
      "epoch": 1.1529850746268657,
      "grad_norm": 0.14443358778953552,
      "learning_rate": 0.0002,
      "loss": 0.5364061594009399,
      "mean_token_accuracy": 0.7827098667621613,
      "num_tokens": 5056521.0,
      "step": 309
    },
    {
      "entropy": 0.5495095700025558,
      "epoch": 1.1567164179104479,
      "grad_norm": 0.15309947729110718,
      "learning_rate": 0.0002,
      "loss": 0.5432314276695251,
      "mean_token_accuracy": 0.7825538218021393,
      "num_tokens": 5073289.0,
      "step": 310
    },
    {
      "entropy": 0.5516466945409775,
      "epoch": 1.1604477611940298,
      "grad_norm": 0.13423986732959747,
      "learning_rate": 0.0002,
      "loss": 0.5399754047393799,
      "mean_token_accuracy": 0.7783583402633667,
      "num_tokens": 5089961.0,
      "step": 311
    },
    {
      "entropy": 0.5431086272001266,
      "epoch": 1.164179104477612,
      "grad_norm": 0.13997507095336914,
      "learning_rate": 0.0002,
      "loss": 0.5346441268920898,
      "mean_token_accuracy": 0.7826652079820633,
      "num_tokens": 5106473.0,
      "step": 312
    },
    {
      "entropy": 0.551514595746994,
      "epoch": 1.1679104477611941,
      "grad_norm": 0.1294819861650467,
      "learning_rate": 0.0002,
      "loss": 0.5547239184379578,
      "mean_token_accuracy": 0.7737241387367249,
      "num_tokens": 5122719.0,
      "step": 313
    },
    {
      "entropy": 0.5247722268104553,
      "epoch": 1.171641791044776,
      "grad_norm": 0.16079172492027283,
      "learning_rate": 0.0002,
      "loss": 0.5377180576324463,
      "mean_token_accuracy": 0.7820589989423752,
      "num_tokens": 5138963.0,
      "step": 314
    },
    {
      "entropy": 0.5316114723682404,
      "epoch": 1.1753731343283582,
      "grad_norm": 0.14574149250984192,
      "learning_rate": 0.0002,
      "loss": 0.5395583510398865,
      "mean_token_accuracy": 0.7803787589073181,
      "num_tokens": 5155393.0,
      "step": 315
    },
    {
      "entropy": 0.5247842520475388,
      "epoch": 1.1791044776119404,
      "grad_norm": 0.1209016665816307,
      "learning_rate": 0.0002,
      "loss": 0.5220508575439453,
      "mean_token_accuracy": 0.7911726981401443,
      "num_tokens": 5171517.0,
      "step": 316
    },
    {
      "entropy": 0.5263349562883377,
      "epoch": 1.1828358208955223,
      "grad_norm": 0.12773162126541138,
      "learning_rate": 0.0002,
      "loss": 0.5241119265556335,
      "mean_token_accuracy": 0.7861284911632538,
      "num_tokens": 5187720.0,
      "step": 317
    },
    {
      "entropy": 0.5590190440416336,
      "epoch": 1.1865671641791045,
      "grad_norm": 0.13424637913703918,
      "learning_rate": 0.0002,
      "loss": 0.557574450969696,
      "mean_token_accuracy": 0.7739435881376266,
      "num_tokens": 5203974.0,
      "step": 318
    },
    {
      "entropy": 0.5342936366796494,
      "epoch": 1.1902985074626866,
      "grad_norm": 0.11586414277553558,
      "learning_rate": 0.0002,
      "loss": 0.5337927937507629,
      "mean_token_accuracy": 0.7846957296133041,
      "num_tokens": 5220438.0,
      "step": 319
    },
    {
      "entropy": 0.5330416560173035,
      "epoch": 1.1940298507462686,
      "grad_norm": 0.14838160574436188,
      "learning_rate": 0.0002,
      "loss": 0.5382588505744934,
      "mean_token_accuracy": 0.7813979238271713,
      "num_tokens": 5236826.0,
      "step": 320
    },
    {
      "entropy": 0.5252465605735779,
      "epoch": 1.1977611940298507,
      "grad_norm": 0.13898268342018127,
      "learning_rate": 0.0002,
      "loss": 0.5281700491905212,
      "mean_token_accuracy": 0.7860727906227112,
      "num_tokens": 5253233.0,
      "step": 321
    },
    {
      "entropy": 0.5264831036329269,
      "epoch": 1.2014925373134329,
      "grad_norm": 0.12452477961778641,
      "learning_rate": 0.0002,
      "loss": 0.5302609205245972,
      "mean_token_accuracy": 0.7824349999427795,
      "num_tokens": 5269677.0,
      "step": 322
    },
    {
      "entropy": 0.5295979827642441,
      "epoch": 1.205223880597015,
      "grad_norm": 0.12388317286968231,
      "learning_rate": 0.0002,
      "loss": 0.5288482904434204,
      "mean_token_accuracy": 0.7865610122680664,
      "num_tokens": 5286069.0,
      "step": 323
    },
    {
      "entropy": 0.5366425067186356,
      "epoch": 1.208955223880597,
      "grad_norm": 0.1489880532026291,
      "learning_rate": 0.0002,
      "loss": 0.5361911058425903,
      "mean_token_accuracy": 0.7808954417705536,
      "num_tokens": 5302186.0,
      "step": 324
    },
    {
      "entropy": 0.5372333526611328,
      "epoch": 1.212686567164179,
      "grad_norm": 0.12909604609012604,
      "learning_rate": 0.0002,
      "loss": 0.5376520156860352,
      "mean_token_accuracy": 0.7795601338148117,
      "num_tokens": 5318344.0,
      "step": 325
    },
    {
      "entropy": 0.5372686982154846,
      "epoch": 1.2164179104477613,
      "grad_norm": 0.12557928264141083,
      "learning_rate": 0.0002,
      "loss": 0.5373642444610596,
      "mean_token_accuracy": 0.7830566316843033,
      "num_tokens": 5334684.0,
      "step": 326
    },
    {
      "entropy": 0.5335738807916641,
      "epoch": 1.2201492537313432,
      "grad_norm": 0.14373338222503662,
      "learning_rate": 0.0002,
      "loss": 0.5395293235778809,
      "mean_token_accuracy": 0.777466356754303,
      "num_tokens": 5350835.0,
      "step": 327
    },
    {
      "entropy": 0.5396379232406616,
      "epoch": 1.2238805970149254,
      "grad_norm": 0.12644916772842407,
      "learning_rate": 0.0002,
      "loss": 0.5411747097969055,
      "mean_token_accuracy": 0.7809303551912308,
      "num_tokens": 5367227.0,
      "step": 328
    },
    {
      "entropy": 0.5528681576251984,
      "epoch": 1.2276119402985075,
      "grad_norm": 0.1431560516357422,
      "learning_rate": 0.0002,
      "loss": 0.5505959391593933,
      "mean_token_accuracy": 0.7723089307546616,
      "num_tokens": 5383738.0,
      "step": 329
    },
    {
      "entropy": 0.5375695377588272,
      "epoch": 1.2313432835820897,
      "grad_norm": 0.11693650484085083,
      "learning_rate": 0.0002,
      "loss": 0.5364817976951599,
      "mean_token_accuracy": 0.7808503806591034,
      "num_tokens": 5400332.0,
      "step": 330
    },
    {
      "entropy": 0.5434885919094086,
      "epoch": 1.2350746268656716,
      "grad_norm": 0.1216924786567688,
      "learning_rate": 0.0002,
      "loss": 0.5423562526702881,
      "mean_token_accuracy": 0.7797342985868454,
      "num_tokens": 5416943.0,
      "step": 331
    },
    {
      "entropy": 0.5413278341293335,
      "epoch": 1.2388059701492538,
      "grad_norm": 0.1281692087650299,
      "learning_rate": 0.0002,
      "loss": 0.5406221151351929,
      "mean_token_accuracy": 0.7824849933385849,
      "num_tokens": 5433247.0,
      "step": 332
    },
    {
      "entropy": 0.5266992747783661,
      "epoch": 1.242537313432836,
      "grad_norm": 0.12139108777046204,
      "learning_rate": 0.0002,
      "loss": 0.523157000541687,
      "mean_token_accuracy": 0.7888277471065521,
      "num_tokens": 5449732.0,
      "step": 333
    },
    {
      "entropy": 0.5189371854066849,
      "epoch": 1.2462686567164178,
      "grad_norm": 0.1311338096857071,
      "learning_rate": 0.0002,
      "loss": 0.5270359516143799,
      "mean_token_accuracy": 0.787872239947319,
      "num_tokens": 5466164.0,
      "step": 334
    },
    {
      "entropy": 0.5501465052366257,
      "epoch": 1.25,
      "grad_norm": 0.13798867166042328,
      "learning_rate": 0.0002,
      "loss": 0.5585375428199768,
      "mean_token_accuracy": 0.7729622572660446,
      "num_tokens": 5482553.0,
      "step": 335
    },
    {
      "entropy": 0.5365464985370636,
      "epoch": 1.2537313432835822,
      "grad_norm": 0.1312791109085083,
      "learning_rate": 0.0002,
      "loss": 0.5411218404769897,
      "mean_token_accuracy": 0.7805088460445404,
      "num_tokens": 5499006.0,
      "step": 336
    },
    {
      "entropy": 0.5384563654661179,
      "epoch": 1.2574626865671643,
      "grad_norm": 0.12015857547521591,
      "learning_rate": 0.0002,
      "loss": 0.5358060002326965,
      "mean_token_accuracy": 0.7817949503660202,
      "num_tokens": 5515412.0,
      "step": 337
    },
    {
      "entropy": 0.5438619628548622,
      "epoch": 1.2611940298507462,
      "grad_norm": 0.1391054093837738,
      "learning_rate": 0.0002,
      "loss": 0.5440458059310913,
      "mean_token_accuracy": 0.7779539227485657,
      "num_tokens": 5531574.0,
      "step": 338
    },
    {
      "entropy": 0.5251474976539612,
      "epoch": 1.2649253731343284,
      "grad_norm": 0.1520051807165146,
      "learning_rate": 0.0002,
      "loss": 0.528119683265686,
      "mean_token_accuracy": 0.7888564765453339,
      "num_tokens": 5548034.0,
      "step": 339
    },
    {
      "entropy": 0.5147749483585358,
      "epoch": 1.2686567164179103,
      "grad_norm": 0.1558913290500641,
      "learning_rate": 0.0002,
      "loss": 0.5209435224533081,
      "mean_token_accuracy": 0.7878405749797821,
      "num_tokens": 5563996.0,
      "step": 340
    },
    {
      "entropy": 0.5468348115682602,
      "epoch": 1.2723880597014925,
      "grad_norm": 0.14310075342655182,
      "learning_rate": 0.0002,
      "loss": 0.5435556769371033,
      "mean_token_accuracy": 0.7792878299951553,
      "num_tokens": 5580555.0,
      "step": 341
    },
    {
      "entropy": 0.5487774908542633,
      "epoch": 1.2761194029850746,
      "grad_norm": 0.1363477259874344,
      "learning_rate": 0.0002,
      "loss": 0.5494071841239929,
      "mean_token_accuracy": 0.7765398621559143,
      "num_tokens": 5597021.0,
      "step": 342
    },
    {
      "entropy": 0.5432843714952469,
      "epoch": 1.2798507462686568,
      "grad_norm": 0.16833975911140442,
      "learning_rate": 0.0002,
      "loss": 0.5400795340538025,
      "mean_token_accuracy": 0.7790917009115219,
      "num_tokens": 5613414.0,
      "step": 343
    },
    {
      "entropy": 0.5422557294368744,
      "epoch": 1.2835820895522387,
      "grad_norm": 0.13373275101184845,
      "learning_rate": 0.0002,
      "loss": 0.54288649559021,
      "mean_token_accuracy": 0.7796483337879181,
      "num_tokens": 5629714.0,
      "step": 344
    },
    {
      "entropy": 0.5340670198202133,
      "epoch": 1.287313432835821,
      "grad_norm": 0.13807663321495056,
      "learning_rate": 0.0002,
      "loss": 0.5384235978126526,
      "mean_token_accuracy": 0.7823121398687363,
      "num_tokens": 5645981.0,
      "step": 345
    },
    {
      "entropy": 0.5413920432329178,
      "epoch": 1.291044776119403,
      "grad_norm": 0.13238251209259033,
      "learning_rate": 0.0002,
      "loss": 0.5383737087249756,
      "mean_token_accuracy": 0.77981998026371,
      "num_tokens": 5662395.0,
      "step": 346
    },
    {
      "entropy": 0.5349558889865875,
      "epoch": 1.294776119402985,
      "grad_norm": 0.12336546927690506,
      "learning_rate": 0.0002,
      "loss": 0.5374408960342407,
      "mean_token_accuracy": 0.7779416292905807,
      "num_tokens": 5678902.0,
      "step": 347
    },
    {
      "entropy": 0.5325140357017517,
      "epoch": 1.2985074626865671,
      "grad_norm": 0.12998366355895996,
      "learning_rate": 0.0002,
      "loss": 0.5309338569641113,
      "mean_token_accuracy": 0.7861874550580978,
      "num_tokens": 5695621.0,
      "step": 348
    },
    {
      "entropy": 0.5402383208274841,
      "epoch": 1.3022388059701493,
      "grad_norm": 0.15694497525691986,
      "learning_rate": 0.0002,
      "loss": 0.5482033491134644,
      "mean_token_accuracy": 0.7775879204273224,
      "num_tokens": 5711695.0,
      "step": 349
    },
    {
      "entropy": 0.5530188679695129,
      "epoch": 1.3059701492537314,
      "grad_norm": 0.16104641556739807,
      "learning_rate": 0.0002,
      "loss": 0.5578451752662659,
      "mean_token_accuracy": 0.7752558439970016,
      "num_tokens": 5727911.0,
      "step": 350
    },
    {
      "entropy": 0.5479607433080673,
      "epoch": 1.3097014925373134,
      "grad_norm": 0.14000706374645233,
      "learning_rate": 0.0002,
      "loss": 0.544680118560791,
      "mean_token_accuracy": 0.7792343497276306,
      "num_tokens": 5744386.0,
      "step": 351
    },
    {
      "entropy": 0.5423225462436676,
      "epoch": 1.3134328358208955,
      "grad_norm": 0.17503292858600616,
      "learning_rate": 0.0002,
      "loss": 0.5420334935188293,
      "mean_token_accuracy": 0.7802097499370575,
      "num_tokens": 5760730.0,
      "step": 352
    },
    {
      "entropy": 0.5337104350328445,
      "epoch": 1.3171641791044777,
      "grad_norm": 0.12291057407855988,
      "learning_rate": 0.0002,
      "loss": 0.5293710231781006,
      "mean_token_accuracy": 0.7864390015602112,
      "num_tokens": 5777001.0,
      "step": 353
    },
    {
      "entropy": 0.5444219410419464,
      "epoch": 1.3208955223880596,
      "grad_norm": 0.13007858395576477,
      "learning_rate": 0.0002,
      "loss": 0.5497652292251587,
      "mean_token_accuracy": 0.7775230556726456,
      "num_tokens": 5793427.0,
      "step": 354
    },
    {
      "entropy": 0.5331632196903229,
      "epoch": 1.3246268656716418,
      "grad_norm": 0.1571977436542511,
      "learning_rate": 0.0002,
      "loss": 0.5505660772323608,
      "mean_token_accuracy": 0.7779581993818283,
      "num_tokens": 5809922.0,
      "step": 355
    },
    {
      "entropy": 0.5501002669334412,
      "epoch": 1.328358208955224,
      "grad_norm": 0.1255483329296112,
      "learning_rate": 0.0002,
      "loss": 0.5463261604309082,
      "mean_token_accuracy": 0.7779203057289124,
      "num_tokens": 5826456.0,
      "step": 356
    },
    {
      "entropy": 0.5357140451669693,
      "epoch": 1.332089552238806,
      "grad_norm": 0.1408921778202057,
      "learning_rate": 0.0002,
      "loss": 0.5318068861961365,
      "mean_token_accuracy": 0.7848965376615524,
      "num_tokens": 5842766.0,
      "step": 357
    },
    {
      "entropy": 0.5357788354158401,
      "epoch": 1.335820895522388,
      "grad_norm": 0.15629494190216064,
      "learning_rate": 0.0002,
      "loss": 0.5369441509246826,
      "mean_token_accuracy": 0.7827416211366653,
      "num_tokens": 5859138.0,
      "step": 358
    },
    {
      "entropy": 0.5363368988037109,
      "epoch": 1.3395522388059702,
      "grad_norm": 0.11929517239332199,
      "learning_rate": 0.0002,
      "loss": 0.5317768454551697,
      "mean_token_accuracy": 0.783902958035469,
      "num_tokens": 5875566.0,
      "step": 359
    },
    {
      "entropy": 0.5138964056968689,
      "epoch": 1.3432835820895521,
      "grad_norm": 0.13416901230812073,
      "learning_rate": 0.0002,
      "loss": 0.5184313058853149,
      "mean_token_accuracy": 0.7893688827753067,
      "num_tokens": 5891795.0,
      "step": 360
    },
    {
      "entropy": 0.5150919556617737,
      "epoch": 1.3470149253731343,
      "grad_norm": 0.13322052359580994,
      "learning_rate": 0.0002,
      "loss": 0.5180276036262512,
      "mean_token_accuracy": 0.7895152717828751,
      "num_tokens": 5908118.0,
      "step": 361
    },
    {
      "entropy": 0.5254094079136848,
      "epoch": 1.3507462686567164,
      "grad_norm": 0.13677389919757843,
      "learning_rate": 0.0002,
      "loss": 0.5256995558738708,
      "mean_token_accuracy": 0.7870867997407913,
      "num_tokens": 5924616.0,
      "step": 362
    },
    {
      "entropy": 0.545592337846756,
      "epoch": 1.3544776119402986,
      "grad_norm": 0.12930770218372345,
      "learning_rate": 0.0002,
      "loss": 0.547904908657074,
      "mean_token_accuracy": 0.7770174890756607,
      "num_tokens": 5941186.0,
      "step": 363
    },
    {
      "entropy": 0.5370796471834183,
      "epoch": 1.3582089552238805,
      "grad_norm": 0.13048067688941956,
      "learning_rate": 0.0002,
      "loss": 0.5408164262771606,
      "mean_token_accuracy": 0.7810717821121216,
      "num_tokens": 5957574.0,
      "step": 364
    },
    {
      "entropy": 0.5403587371110916,
      "epoch": 1.3619402985074627,
      "grad_norm": 0.15434801578521729,
      "learning_rate": 0.0002,
      "loss": 0.5395029783248901,
      "mean_token_accuracy": 0.7786827087402344,
      "num_tokens": 5973998.0,
      "step": 365
    },
    {
      "entropy": 0.5292467176914215,
      "epoch": 1.3656716417910448,
      "grad_norm": 0.12458949536085129,
      "learning_rate": 0.0002,
      "loss": 0.5264879465103149,
      "mean_token_accuracy": 0.7861236184835434,
      "num_tokens": 5990382.0,
      "step": 366
    },
    {
      "entropy": 0.5326224118471146,
      "epoch": 1.3694029850746268,
      "grad_norm": 0.12966883182525635,
      "learning_rate": 0.0002,
      "loss": 0.5328729152679443,
      "mean_token_accuracy": 0.7840806394815445,
      "num_tokens": 6006691.0,
      "step": 367
    },
    {
      "entropy": 0.5448299199342728,
      "epoch": 1.373134328358209,
      "grad_norm": 0.15954849123954773,
      "learning_rate": 0.0002,
      "loss": 0.5463345050811768,
      "mean_token_accuracy": 0.7785107493400574,
      "num_tokens": 6022875.0,
      "step": 368
    },
    {
      "entropy": 0.5402169525623322,
      "epoch": 1.376865671641791,
      "grad_norm": 0.1721678376197815,
      "learning_rate": 0.0002,
      "loss": 0.5449873208999634,
      "mean_token_accuracy": 0.7793633341789246,
      "num_tokens": 6039437.0,
      "step": 369
    },
    {
      "entropy": 0.522490456700325,
      "epoch": 1.3805970149253732,
      "grad_norm": 0.13083986937999725,
      "learning_rate": 0.0002,
      "loss": 0.5237096548080444,
      "mean_token_accuracy": 0.7891271412372589,
      "num_tokens": 6056060.0,
      "step": 370
    },
    {
      "entropy": 0.5373298078775406,
      "epoch": 1.3843283582089552,
      "grad_norm": 0.14084212481975555,
      "learning_rate": 0.0002,
      "loss": 0.5413792133331299,
      "mean_token_accuracy": 0.7821934223175049,
      "num_tokens": 6072567.0,
      "step": 371
    },
    {
      "entropy": 0.5313082635402679,
      "epoch": 1.3880597014925373,
      "grad_norm": 0.1413281410932541,
      "learning_rate": 0.0002,
      "loss": 0.5352596640586853,
      "mean_token_accuracy": 0.782850906252861,
      "num_tokens": 6088818.0,
      "step": 372
    },
    {
      "entropy": 0.529704749584198,
      "epoch": 1.3917910447761195,
      "grad_norm": 0.128413125872612,
      "learning_rate": 0.0002,
      "loss": 0.5355388522148132,
      "mean_token_accuracy": 0.7817994952201843,
      "num_tokens": 6105190.0,
      "step": 373
    },
    {
      "entropy": 0.547185018658638,
      "epoch": 1.3955223880597014,
      "grad_norm": 0.13032199442386627,
      "learning_rate": 0.0002,
      "loss": 0.5498857498168945,
      "mean_token_accuracy": 0.7774333655834198,
      "num_tokens": 6121814.0,
      "step": 374
    },
    {
      "entropy": 0.5206846371293068,
      "epoch": 1.3992537313432836,
      "grad_norm": 0.1367998868227005,
      "learning_rate": 0.0002,
      "loss": 0.525123119354248,
      "mean_token_accuracy": 0.7872356921434402,
      "num_tokens": 6137850.0,
      "step": 375
    },
    {
      "entropy": 0.5373269021511078,
      "epoch": 1.4029850746268657,
      "grad_norm": 0.1361006647348404,
      "learning_rate": 0.0002,
      "loss": 0.5285162925720215,
      "mean_token_accuracy": 0.7845685184001923,
      "num_tokens": 6154202.0,
      "step": 376
    },
    {
      "entropy": 0.5583214312791824,
      "epoch": 1.4067164179104479,
      "grad_norm": 0.12571895122528076,
      "learning_rate": 0.0002,
      "loss": 0.5490396022796631,
      "mean_token_accuracy": 0.7769980877637863,
      "num_tokens": 6170623.0,
      "step": 377
    },
    {
      "entropy": 0.5345316231250763,
      "epoch": 1.4104477611940298,
      "grad_norm": 0.12002874165773392,
      "learning_rate": 0.0002,
      "loss": 0.5335662961006165,
      "mean_token_accuracy": 0.7847907692193985,
      "num_tokens": 6186995.0,
      "step": 378
    },
    {
      "entropy": 0.5344050079584122,
      "epoch": 1.414179104477612,
      "grad_norm": 0.13620412349700928,
      "learning_rate": 0.0002,
      "loss": 0.5380545854568481,
      "mean_token_accuracy": 0.7809189409017563,
      "num_tokens": 6203244.0,
      "step": 379
    },
    {
      "entropy": 0.5563318729400635,
      "epoch": 1.417910447761194,
      "grad_norm": 0.1582707017660141,
      "learning_rate": 0.0002,
      "loss": 0.5650156140327454,
      "mean_token_accuracy": 0.7710053324699402,
      "num_tokens": 6219704.0,
      "step": 380
    },
    {
      "entropy": 0.5314245671033859,
      "epoch": 1.421641791044776,
      "grad_norm": 0.14954440295696259,
      "learning_rate": 0.0002,
      "loss": 0.5360095500946045,
      "mean_token_accuracy": 0.7819344699382782,
      "num_tokens": 6236108.0,
      "step": 381
    },
    {
      "entropy": 0.5438588559627533,
      "epoch": 1.4253731343283582,
      "grad_norm": 0.13026201725006104,
      "learning_rate": 0.0002,
      "loss": 0.546026885509491,
      "mean_token_accuracy": 0.7794162184000015,
      "num_tokens": 6252551.0,
      "step": 382
    },
    {
      "entropy": 0.5416135936975479,
      "epoch": 1.4291044776119404,
      "grad_norm": 0.12766756117343903,
      "learning_rate": 0.0002,
      "loss": 0.5409781336784363,
      "mean_token_accuracy": 0.7814617455005646,
      "num_tokens": 6268947.0,
      "step": 383
    },
    {
      "entropy": 0.5397155731916428,
      "epoch": 1.4328358208955223,
      "grad_norm": 0.12332645803689957,
      "learning_rate": 0.0002,
      "loss": 0.5322134494781494,
      "mean_token_accuracy": 0.7832454741001129,
      "num_tokens": 6285443.0,
      "step": 384
    },
    {
      "entropy": 0.5389638394117355,
      "epoch": 1.4365671641791045,
      "grad_norm": 0.1402750462293625,
      "learning_rate": 0.0002,
      "loss": 0.537129282951355,
      "mean_token_accuracy": 0.7819686532020569,
      "num_tokens": 6301736.0,
      "step": 385
    },
    {
      "entropy": 0.5305354446172714,
      "epoch": 1.4402985074626866,
      "grad_norm": 0.1382511556148529,
      "learning_rate": 0.0002,
      "loss": 0.5361950993537903,
      "mean_token_accuracy": 0.7800089418888092,
      "num_tokens": 6318226.0,
      "step": 386
    },
    {
      "entropy": 0.5448728799819946,
      "epoch": 1.4440298507462686,
      "grad_norm": 0.12688331305980682,
      "learning_rate": 0.0002,
      "loss": 0.5425639152526855,
      "mean_token_accuracy": 0.7793668359518051,
      "num_tokens": 6334901.0,
      "step": 387
    },
    {
      "entropy": 0.5163320973515511,
      "epoch": 1.4477611940298507,
      "grad_norm": 0.1287093609571457,
      "learning_rate": 0.0002,
      "loss": 0.5205877423286438,
      "mean_token_accuracy": 0.7889581769704819,
      "num_tokens": 6351325.0,
      "step": 388
    },
    {
      "entropy": 0.534174844622612,
      "epoch": 1.4514925373134329,
      "grad_norm": 0.13482384383678436,
      "learning_rate": 0.0002,
      "loss": 0.5346795916557312,
      "mean_token_accuracy": 0.785330280661583,
      "num_tokens": 6367848.0,
      "step": 389
    },
    {
      "entropy": 0.5353098511695862,
      "epoch": 1.455223880597015,
      "grad_norm": 0.12721478939056396,
      "learning_rate": 0.0002,
      "loss": 0.5334895849227905,
      "mean_token_accuracy": 0.7840058505535126,
      "num_tokens": 6384128.0,
      "step": 390
    },
    {
      "entropy": 0.5127712190151215,
      "epoch": 1.458955223880597,
      "grad_norm": 0.3956281244754791,
      "learning_rate": 0.0002,
      "loss": 0.5116028785705566,
      "mean_token_accuracy": 0.7937542349100113,
      "num_tokens": 6400299.0,
      "step": 391
    },
    {
      "entropy": 0.5303019732236862,
      "epoch": 1.462686567164179,
      "grad_norm": 0.26528817415237427,
      "learning_rate": 0.0002,
      "loss": 0.5407310724258423,
      "mean_token_accuracy": 0.7815377861261368,
      "num_tokens": 6416534.0,
      "step": 392
    },
    {
      "entropy": 0.5287035554647446,
      "epoch": 1.4664179104477613,
      "grad_norm": 0.25263166427612305,
      "learning_rate": 0.0002,
      "loss": 0.528439462184906,
      "mean_token_accuracy": 0.7890845090150833,
      "num_tokens": 6432857.0,
      "step": 393
    },
    {
      "entropy": 0.5452031493186951,
      "epoch": 1.4701492537313432,
      "grad_norm": 0.14761021733283997,
      "learning_rate": 0.0002,
      "loss": 0.54726243019104,
      "mean_token_accuracy": 0.7756960988044739,
      "num_tokens": 6449409.0,
      "step": 394
    },
    {
      "entropy": 0.5531849265098572,
      "epoch": 1.4738805970149254,
      "grad_norm": 0.17053969204425812,
      "learning_rate": 0.0002,
      "loss": 0.5543383359909058,
      "mean_token_accuracy": 0.7772884219884872,
      "num_tokens": 6465774.0,
      "step": 395
    },
    {
      "entropy": 0.5687634348869324,
      "epoch": 1.4776119402985075,
      "grad_norm": 0.13562458753585815,
      "learning_rate": 0.0002,
      "loss": 0.5645351409912109,
      "mean_token_accuracy": 0.7718057334423065,
      "num_tokens": 6482042.0,
      "step": 396
    },
    {
      "entropy": 0.544958308339119,
      "epoch": 1.4813432835820897,
      "grad_norm": 0.15165205299854279,
      "learning_rate": 0.0002,
      "loss": 0.538078248500824,
      "mean_token_accuracy": 0.780204176902771,
      "num_tokens": 6498518.0,
      "step": 397
    },
    {
      "entropy": 0.5311676114797592,
      "epoch": 1.4850746268656716,
      "grad_norm": 0.1277652531862259,
      "learning_rate": 0.0002,
      "loss": 0.5337294340133667,
      "mean_token_accuracy": 0.7831934988498688,
      "num_tokens": 6514860.0,
      "step": 398
    },
    {
      "entropy": 0.5275157988071442,
      "epoch": 1.4888059701492538,
      "grad_norm": 0.16418659687042236,
      "learning_rate": 0.0002,
      "loss": 0.5396382212638855,
      "mean_token_accuracy": 0.7796144187450409,
      "num_tokens": 6531282.0,
      "step": 399
    },
    {
      "entropy": 0.5259099006652832,
      "epoch": 1.4925373134328357,
      "grad_norm": 0.126469686627388,
      "learning_rate": 0.0002,
      "loss": 0.5243666768074036,
      "mean_token_accuracy": 0.7873355597257614,
      "num_tokens": 6547659.0,
      "step": 400
    },
    {
      "entropy": 0.5152204781770706,
      "epoch": 1.4962686567164178,
      "grad_norm": 0.14800144731998444,
      "learning_rate": 0.0002,
      "loss": 0.5207810401916504,
      "mean_token_accuracy": 0.7866075932979584,
      "num_tokens": 6563877.0,
      "step": 401
    },
    {
      "entropy": 0.5387643277645111,
      "epoch": 1.5,
      "grad_norm": 0.13282911479473114,
      "learning_rate": 0.0002,
      "loss": 0.5396767258644104,
      "mean_token_accuracy": 0.7800559848546982,
      "num_tokens": 6580214.0,
      "step": 402
    },
    {
      "entropy": 0.5597681552171707,
      "epoch": 1.5037313432835822,
      "grad_norm": 0.13526253402233124,
      "learning_rate": 0.0002,
      "loss": 0.564717173576355,
      "mean_token_accuracy": 0.7698872089385986,
      "num_tokens": 6596727.0,
      "step": 403
    },
    {
      "entropy": 0.5411169677972794,
      "epoch": 1.5074626865671643,
      "grad_norm": 0.1372886449098587,
      "learning_rate": 0.0002,
      "loss": 0.544418215751648,
      "mean_token_accuracy": 0.7771344631910324,
      "num_tokens": 6613231.0,
      "step": 404
    },
    {
      "entropy": 0.5623854994773865,
      "epoch": 1.5111940298507462,
      "grad_norm": 0.14793866872787476,
      "learning_rate": 0.0002,
      "loss": 0.5613381862640381,
      "mean_token_accuracy": 0.7715491652488708,
      "num_tokens": 6629470.0,
      "step": 405
    },
    {
      "entropy": 0.5456371009349823,
      "epoch": 1.5149253731343284,
      "grad_norm": 0.12701347470283508,
      "learning_rate": 0.0002,
      "loss": 0.541689395904541,
      "mean_token_accuracy": 0.7824305295944214,
      "num_tokens": 6646065.0,
      "step": 406
    },
    {
      "entropy": 0.552760511636734,
      "epoch": 1.5186567164179103,
      "grad_norm": 0.13117454946041107,
      "learning_rate": 0.0002,
      "loss": 0.5483608245849609,
      "mean_token_accuracy": 0.7776399701833725,
      "num_tokens": 6662532.0,
      "step": 407
    },
    {
      "entropy": 0.53936767578125,
      "epoch": 1.5223880597014925,
      "grad_norm": 0.14745967090129852,
      "learning_rate": 0.0002,
      "loss": 0.5434844493865967,
      "mean_token_accuracy": 0.7814289182424545,
      "num_tokens": 6678864.0,
      "step": 408
    },
    {
      "entropy": 0.5357921570539474,
      "epoch": 1.5261194029850746,
      "grad_norm": 0.13177861273288727,
      "learning_rate": 0.0002,
      "loss": 0.5398023128509521,
      "mean_token_accuracy": 0.7828291207551956,
      "num_tokens": 6695275.0,
      "step": 409
    },
    {
      "entropy": 0.5142264068126678,
      "epoch": 1.5298507462686568,
      "grad_norm": 0.12137385457754135,
      "learning_rate": 0.0002,
      "loss": 0.5177730917930603,
      "mean_token_accuracy": 0.7905157059431076,
      "num_tokens": 6711799.0,
      "step": 410
    },
    {
      "entropy": 0.5344009697437286,
      "epoch": 1.533582089552239,
      "grad_norm": 0.1399386078119278,
      "learning_rate": 0.0002,
      "loss": 0.5395348072052002,
      "mean_token_accuracy": 0.780754029750824,
      "num_tokens": 6727958.0,
      "step": 411
    },
    {
      "entropy": 0.5387486070394516,
      "epoch": 1.537313432835821,
      "grad_norm": 0.14169862866401672,
      "learning_rate": 0.0002,
      "loss": 0.5429282188415527,
      "mean_token_accuracy": 0.7802885323762894,
      "num_tokens": 6744330.0,
      "step": 412
    },
    {
      "entropy": 0.5318516343832016,
      "epoch": 1.5410447761194028,
      "grad_norm": 0.1238308697938919,
      "learning_rate": 0.0002,
      "loss": 0.532063901424408,
      "mean_token_accuracy": 0.785664901137352,
      "num_tokens": 6760822.0,
      "step": 413
    },
    {
      "entropy": 0.5161900222301483,
      "epoch": 1.544776119402985,
      "grad_norm": 0.13053998351097107,
      "learning_rate": 0.0002,
      "loss": 0.5176095962524414,
      "mean_token_accuracy": 0.7909317165613174,
      "num_tokens": 6777117.0,
      "step": 414
    },
    {
      "entropy": 0.5330372601747513,
      "epoch": 1.5485074626865671,
      "grad_norm": 0.14414377510547638,
      "learning_rate": 0.0002,
      "loss": 0.5398930907249451,
      "mean_token_accuracy": 0.7799312919378281,
      "num_tokens": 6793354.0,
      "step": 415
    },
    {
      "entropy": 0.5430023074150085,
      "epoch": 1.5522388059701493,
      "grad_norm": 0.1265353262424469,
      "learning_rate": 0.0002,
      "loss": 0.5396794080734253,
      "mean_token_accuracy": 0.7781756520271301,
      "num_tokens": 6809614.0,
      "step": 416
    },
    {
      "entropy": 0.5544451475143433,
      "epoch": 1.5559701492537314,
      "grad_norm": 0.12820257246494293,
      "learning_rate": 0.0002,
      "loss": 0.5425227880477905,
      "mean_token_accuracy": 0.7799335867166519,
      "num_tokens": 6826205.0,
      "step": 417
    },
    {
      "entropy": 0.5510530769824982,
      "epoch": 1.5597014925373134,
      "grad_norm": 0.15976029634475708,
      "learning_rate": 0.0002,
      "loss": 0.5494173765182495,
      "mean_token_accuracy": 0.7780761867761612,
      "num_tokens": 6842446.0,
      "step": 418
    },
    {
      "entropy": 0.5290757119655609,
      "epoch": 1.5634328358208955,
      "grad_norm": 0.1367224007844925,
      "learning_rate": 0.0002,
      "loss": 0.5261332392692566,
      "mean_token_accuracy": 0.7868656665086746,
      "num_tokens": 6858543.0,
      "step": 419
    },
    {
      "entropy": 0.5366243869066238,
      "epoch": 1.5671641791044775,
      "grad_norm": 0.15515874326229095,
      "learning_rate": 0.0002,
      "loss": 0.5382797718048096,
      "mean_token_accuracy": 0.7808849364519119,
      "num_tokens": 6874880.0,
      "step": 420
    },
    {
      "entropy": 0.5246019065380096,
      "epoch": 1.5708955223880596,
      "grad_norm": 0.1444631963968277,
      "learning_rate": 0.0002,
      "loss": 0.5281886458396912,
      "mean_token_accuracy": 0.7854573130607605,
      "num_tokens": 6891435.0,
      "step": 421
    },
    {
      "entropy": 0.5308691114187241,
      "epoch": 1.5746268656716418,
      "grad_norm": 0.14639756083488464,
      "learning_rate": 0.0002,
      "loss": 0.5425790548324585,
      "mean_token_accuracy": 0.7789381593465805,
      "num_tokens": 6907821.0,
      "step": 422
    },
    {
      "entropy": 0.5465221256017685,
      "epoch": 1.578358208955224,
      "grad_norm": 0.1638735681772232,
      "learning_rate": 0.0002,
      "loss": 0.5480068325996399,
      "mean_token_accuracy": 0.7775472104549408,
      "num_tokens": 6924228.0,
      "step": 423
    },
    {
      "entropy": 0.5357428267598152,
      "epoch": 1.582089552238806,
      "grad_norm": 0.15004044771194458,
      "learning_rate": 0.0002,
      "loss": 0.534127950668335,
      "mean_token_accuracy": 0.7842502892017365,
      "num_tokens": 6940461.0,
      "step": 424
    },
    {
      "entropy": 0.5376645922660828,
      "epoch": 1.585820895522388,
      "grad_norm": 0.14717955887317657,
      "learning_rate": 0.0002,
      "loss": 0.5365069508552551,
      "mean_token_accuracy": 0.7809850126504898,
      "num_tokens": 6956984.0,
      "step": 425
    },
    {
      "entropy": 0.5195955783128738,
      "epoch": 1.5895522388059702,
      "grad_norm": 0.15700839459896088,
      "learning_rate": 0.0002,
      "loss": 0.5263607501983643,
      "mean_token_accuracy": 0.786608874797821,
      "num_tokens": 6973184.0,
      "step": 426
    },
    {
      "entropy": 0.5346983373165131,
      "epoch": 1.5932835820895521,
      "grad_norm": 0.1265949308872223,
      "learning_rate": 0.0002,
      "loss": 0.5349383354187012,
      "mean_token_accuracy": 0.7829756736755371,
      "num_tokens": 6989588.0,
      "step": 427
    },
    {
      "entropy": 0.528701514005661,
      "epoch": 1.5970149253731343,
      "grad_norm": 0.13123738765716553,
      "learning_rate": 0.0002,
      "loss": 0.5311565399169922,
      "mean_token_accuracy": 0.7848205864429474,
      "num_tokens": 7006032.0,
      "step": 428
    },
    {
      "entropy": 0.5235633179545403,
      "epoch": 1.6007462686567164,
      "grad_norm": 0.1442072093486786,
      "learning_rate": 0.0002,
      "loss": 0.524992048740387,
      "mean_token_accuracy": 0.7862335294485092,
      "num_tokens": 7022556.0,
      "step": 429
    },
    {
      "entropy": 0.5378939062356949,
      "epoch": 1.6044776119402986,
      "grad_norm": 0.14503757655620575,
      "learning_rate": 0.0002,
      "loss": 0.5370591282844543,
      "mean_token_accuracy": 0.7793628126382828,
      "num_tokens": 7038977.0,
      "step": 430
    },
    {
      "entropy": 0.542696550488472,
      "epoch": 1.6082089552238807,
      "grad_norm": 0.14489823579788208,
      "learning_rate": 0.0002,
      "loss": 0.5515229105949402,
      "mean_token_accuracy": 0.7762403339147568,
      "num_tokens": 7055532.0,
      "step": 431
    },
    {
      "entropy": 0.5282143354415894,
      "epoch": 1.6119402985074627,
      "grad_norm": 0.1357962042093277,
      "learning_rate": 0.0002,
      "loss": 0.5251887440681458,
      "mean_token_accuracy": 0.7881295680999756,
      "num_tokens": 7071746.0,
      "step": 432
    },
    {
      "entropy": 0.5541251450777054,
      "epoch": 1.6156716417910446,
      "grad_norm": 0.1309272199869156,
      "learning_rate": 0.0002,
      "loss": 0.5470398664474487,
      "mean_token_accuracy": 0.7779666483402252,
      "num_tokens": 7088233.0,
      "step": 433
    },
    {
      "entropy": 0.5291491746902466,
      "epoch": 1.6194029850746268,
      "grad_norm": 0.13591884076595306,
      "learning_rate": 0.0002,
      "loss": 0.5252074599266052,
      "mean_token_accuracy": 0.7864278256893158,
      "num_tokens": 7104939.0,
      "step": 434
    },
    {
      "entropy": 0.5383701026439667,
      "epoch": 1.623134328358209,
      "grad_norm": 0.15707901120185852,
      "learning_rate": 0.0002,
      "loss": 0.5444713830947876,
      "mean_token_accuracy": 0.7803786247968674,
      "num_tokens": 7121416.0,
      "step": 435
    },
    {
      "entropy": 0.5325479060411453,
      "epoch": 1.626865671641791,
      "grad_norm": 0.18670128285884857,
      "learning_rate": 0.0002,
      "loss": 0.5340004563331604,
      "mean_token_accuracy": 0.7836445271968842,
      "num_tokens": 7137612.0,
      "step": 436
    },
    {
      "entropy": 0.5397835820913315,
      "epoch": 1.6305970149253732,
      "grad_norm": 0.1274837851524353,
      "learning_rate": 0.0002,
      "loss": 0.5392928123474121,
      "mean_token_accuracy": 0.779869869351387,
      "num_tokens": 7153967.0,
      "step": 437
    },
    {
      "entropy": 0.5358160883188248,
      "epoch": 1.6343283582089554,
      "grad_norm": 0.12929777801036835,
      "learning_rate": 0.0002,
      "loss": 0.5390000343322754,
      "mean_token_accuracy": 0.7791162878274918,
      "num_tokens": 7170204.0,
      "step": 438
    },
    {
      "entropy": 0.5452472269535065,
      "epoch": 1.6380597014925373,
      "grad_norm": 0.15463511645793915,
      "learning_rate": 0.0002,
      "loss": 0.546872079372406,
      "mean_token_accuracy": 0.7800528854131699,
      "num_tokens": 7186820.0,
      "step": 439
    },
    {
      "entropy": 0.53532275557518,
      "epoch": 1.6417910447761193,
      "grad_norm": 0.13320156931877136,
      "learning_rate": 0.0002,
      "loss": 0.5344700217247009,
      "mean_token_accuracy": 0.784910649061203,
      "num_tokens": 7203069.0,
      "step": 440
    },
    {
      "entropy": 0.5384882837533951,
      "epoch": 1.6455223880597014,
      "grad_norm": 0.1380772888660431,
      "learning_rate": 0.0002,
      "loss": 0.5379555225372314,
      "mean_token_accuracy": 0.7825993895530701,
      "num_tokens": 7219386.0,
      "step": 441
    },
    {
      "entropy": 0.5263376832008362,
      "epoch": 1.6492537313432836,
      "grad_norm": 0.14290866255760193,
      "learning_rate": 0.0002,
      "loss": 0.5337180495262146,
      "mean_token_accuracy": 0.7823992669582367,
      "num_tokens": 7235440.0,
      "step": 442
    },
    {
      "entropy": 0.5457373559474945,
      "epoch": 1.6529850746268657,
      "grad_norm": 0.14475581049919128,
      "learning_rate": 0.0002,
      "loss": 0.5475543737411499,
      "mean_token_accuracy": 0.7789654433727264,
      "num_tokens": 7251825.0,
      "step": 443
    },
    {
      "entropy": 0.5170267447829247,
      "epoch": 1.6567164179104479,
      "grad_norm": 0.1291104257106781,
      "learning_rate": 0.0002,
      "loss": 0.5238728523254395,
      "mean_token_accuracy": 0.7878426611423492,
      "num_tokens": 7268060.0,
      "step": 444
    },
    {
      "entropy": 0.5119059011340141,
      "epoch": 1.6604477611940298,
      "grad_norm": 0.1312272697687149,
      "learning_rate": 0.0002,
      "loss": 0.5107318162918091,
      "mean_token_accuracy": 0.7920443564653397,
      "num_tokens": 7284406.0,
      "step": 445
    },
    {
      "entropy": 0.5405990779399872,
      "epoch": 1.664179104477612,
      "grad_norm": 0.13291458785533905,
      "learning_rate": 0.0002,
      "loss": 0.5375615358352661,
      "mean_token_accuracy": 0.7826282531023026,
      "num_tokens": 7300830.0,
      "step": 446
    },
    {
      "entropy": 0.5290491580963135,
      "epoch": 1.667910447761194,
      "grad_norm": 0.16867397725582123,
      "learning_rate": 0.0002,
      "loss": 0.5367984175682068,
      "mean_token_accuracy": 0.7823507636785507,
      "num_tokens": 7317376.0,
      "step": 447
    },
    {
      "entropy": 0.5333274900913239,
      "epoch": 1.671641791044776,
      "grad_norm": 0.13572576642036438,
      "learning_rate": 0.0002,
      "loss": 0.5391582250595093,
      "mean_token_accuracy": 0.7835893034934998,
      "num_tokens": 7333710.0,
      "step": 448
    },
    {
      "entropy": 0.5388626307249069,
      "epoch": 1.6753731343283582,
      "grad_norm": 0.13041196763515472,
      "learning_rate": 0.0002,
      "loss": 0.5398558378219604,
      "mean_token_accuracy": 0.7816222161054611,
      "num_tokens": 7350126.0,
      "step": 449
    },
    {
      "entropy": 0.5627462565898895,
      "epoch": 1.6791044776119404,
      "grad_norm": 0.1380966603755951,
      "learning_rate": 0.0002,
      "loss": 0.5585891008377075,
      "mean_token_accuracy": 0.774171307682991,
      "num_tokens": 7366445.0,
      "step": 450
    },
    {
      "entropy": 0.5573850572109222,
      "epoch": 1.6828358208955225,
      "grad_norm": 0.13787388801574707,
      "learning_rate": 0.0002,
      "loss": 0.5412402153015137,
      "mean_token_accuracy": 0.7797098010778427,
      "num_tokens": 7382939.0,
      "step": 451
    },
    {
      "entropy": 0.5532975196838379,
      "epoch": 1.6865671641791045,
      "grad_norm": 0.13220225274562836,
      "learning_rate": 0.0002,
      "loss": 0.5484021902084351,
      "mean_token_accuracy": 0.7746714651584625,
      "num_tokens": 7399139.0,
      "step": 452
    },
    {
      "entropy": 0.5151724368333817,
      "epoch": 1.6902985074626866,
      "grad_norm": 0.21221774816513062,
      "learning_rate": 0.0002,
      "loss": 0.5242787599563599,
      "mean_token_accuracy": 0.7887167632579803,
      "num_tokens": 7415333.0,
      "step": 453
    },
    {
      "entropy": 0.5301668345928192,
      "epoch": 1.6940298507462686,
      "grad_norm": 0.1518767923116684,
      "learning_rate": 0.0002,
      "loss": 0.5330111980438232,
      "mean_token_accuracy": 0.7836803197860718,
      "num_tokens": 7431990.0,
      "step": 454
    },
    {
      "entropy": 0.5392823070287704,
      "epoch": 1.6977611940298507,
      "grad_norm": 0.14736776053905487,
      "learning_rate": 0.0002,
      "loss": 0.5444834232330322,
      "mean_token_accuracy": 0.7808229774236679,
      "num_tokens": 7448245.0,
      "step": 455
    },
    {
      "entropy": 0.5256123840808868,
      "epoch": 1.7014925373134329,
      "grad_norm": 0.14706024527549744,
      "learning_rate": 0.0002,
      "loss": 0.5332489013671875,
      "mean_token_accuracy": 0.7833222001791,
      "num_tokens": 7464255.0,
      "step": 456
    },
    {
      "entropy": 0.5445366203784943,
      "epoch": 1.705223880597015,
      "grad_norm": 0.16646987199783325,
      "learning_rate": 0.0002,
      "loss": 0.5444868206977844,
      "mean_token_accuracy": 0.7794564366340637,
      "num_tokens": 7480416.0,
      "step": 457
    },
    {
      "entropy": 0.5389459282159805,
      "epoch": 1.7089552238805972,
      "grad_norm": 0.12615232169628143,
      "learning_rate": 0.0002,
      "loss": 0.5393764972686768,
      "mean_token_accuracy": 0.78050696849823,
      "num_tokens": 7496810.0,
      "step": 458
    },
    {
      "entropy": 0.5414003133773804,
      "epoch": 1.712686567164179,
      "grad_norm": 0.18230244517326355,
      "learning_rate": 0.0002,
      "loss": 0.547127902507782,
      "mean_token_accuracy": 0.7776285856962204,
      "num_tokens": 7512967.0,
      "step": 459
    },
    {
      "entropy": 0.5501416921615601,
      "epoch": 1.716417910447761,
      "grad_norm": 0.138383686542511,
      "learning_rate": 0.0002,
      "loss": 0.5501201152801514,
      "mean_token_accuracy": 0.7753444314002991,
      "num_tokens": 7529496.0,
      "step": 460
    },
    {
      "entropy": 0.5400792211294174,
      "epoch": 1.7201492537313432,
      "grad_norm": 0.15383076667785645,
      "learning_rate": 0.0002,
      "loss": 0.5444374084472656,
      "mean_token_accuracy": 0.7796078622341156,
      "num_tokens": 7545816.0,
      "step": 461
    },
    {
      "entropy": 0.5446860194206238,
      "epoch": 1.7238805970149254,
      "grad_norm": 0.1254233568906784,
      "learning_rate": 0.0002,
      "loss": 0.5380433797836304,
      "mean_token_accuracy": 0.7813080102205276,
      "num_tokens": 7562338.0,
      "step": 462
    },
    {
      "entropy": 0.5387992411851883,
      "epoch": 1.7276119402985075,
      "grad_norm": 0.1430063545703888,
      "learning_rate": 0.0002,
      "loss": 0.5298895835876465,
      "mean_token_accuracy": 0.7857054173946381,
      "num_tokens": 7578886.0,
      "step": 463
    },
    {
      "entropy": 0.5432053953409195,
      "epoch": 1.7313432835820897,
      "grad_norm": 0.15069834887981415,
      "learning_rate": 0.0002,
      "loss": 0.5394545197486877,
      "mean_token_accuracy": 0.7813818156719208,
      "num_tokens": 7594923.0,
      "step": 464
    },
    {
      "entropy": 0.5428010821342468,
      "epoch": 1.7350746268656716,
      "grad_norm": 0.12793928384780884,
      "learning_rate": 0.0002,
      "loss": 0.5419149398803711,
      "mean_token_accuracy": 0.779750257730484,
      "num_tokens": 7611394.0,
      "step": 465
    },
    {
      "entropy": 0.5389568954706192,
      "epoch": 1.7388059701492538,
      "grad_norm": 0.17946098744869232,
      "learning_rate": 0.0002,
      "loss": 0.5485541820526123,
      "mean_token_accuracy": 0.7761444449424744,
      "num_tokens": 7627602.0,
      "step": 466
    },
    {
      "entropy": 0.5574225187301636,
      "epoch": 1.7425373134328357,
      "grad_norm": 0.26036718487739563,
      "learning_rate": 0.0002,
      "loss": 0.5592665076255798,
      "mean_token_accuracy": 0.7752985954284668,
      "num_tokens": 7644127.0,
      "step": 467
    },
    {
      "entropy": 0.5414848476648331,
      "epoch": 1.7462686567164178,
      "grad_norm": 0.14623212814331055,
      "learning_rate": 0.0002,
      "loss": 0.5441226363182068,
      "mean_token_accuracy": 0.7790254950523376,
      "num_tokens": 7660629.0,
      "step": 468
    },
    {
      "entropy": 0.5366654098033905,
      "epoch": 1.75,
      "grad_norm": 0.14204439520835876,
      "learning_rate": 0.0002,
      "loss": 0.5322362184524536,
      "mean_token_accuracy": 0.7856068909168243,
      "num_tokens": 7676801.0,
      "step": 469
    },
    {
      "entropy": 0.5576929599046707,
      "epoch": 1.7537313432835822,
      "grad_norm": 0.3054654896259308,
      "learning_rate": 0.0002,
      "loss": 0.5612924695014954,
      "mean_token_accuracy": 0.7732748240232468,
      "num_tokens": 7693088.0,
      "step": 470
    },
    {
      "entropy": 0.5312875211238861,
      "epoch": 1.7574626865671643,
      "grad_norm": 0.14751650393009186,
      "learning_rate": 0.0002,
      "loss": 0.5314509868621826,
      "mean_token_accuracy": 0.7852085381746292,
      "num_tokens": 7709211.0,
      "step": 471
    },
    {
      "entropy": 0.5124304220080376,
      "epoch": 1.7611940298507462,
      "grad_norm": 0.16575190424919128,
      "learning_rate": 0.0002,
      "loss": 0.5192998051643372,
      "mean_token_accuracy": 0.7895690947771072,
      "num_tokens": 7725574.0,
      "step": 472
    },
    {
      "entropy": 0.5452256798744202,
      "epoch": 1.7649253731343284,
      "grad_norm": 0.20871040225028992,
      "learning_rate": 0.0002,
      "loss": 0.5548557639122009,
      "mean_token_accuracy": 0.7744195759296417,
      "num_tokens": 7741923.0,
      "step": 473
    },
    {
      "entropy": 0.5397016704082489,
      "epoch": 1.7686567164179103,
      "grad_norm": 0.20561866462230682,
      "learning_rate": 0.0002,
      "loss": 0.5485490560531616,
      "mean_token_accuracy": 0.7755938917398453,
      "num_tokens": 7758130.0,
      "step": 474
    },
    {
      "entropy": 0.5324810743331909,
      "epoch": 1.7723880597014925,
      "grad_norm": 0.14155474305152893,
      "learning_rate": 0.0002,
      "loss": 0.53411865234375,
      "mean_token_accuracy": 0.7829086035490036,
      "num_tokens": 7774449.0,
      "step": 475
    },
    {
      "entropy": 0.5462884604930878,
      "epoch": 1.7761194029850746,
      "grad_norm": 0.13009391725063324,
      "learning_rate": 0.0002,
      "loss": 0.5341886281967163,
      "mean_token_accuracy": 0.7831237614154816,
      "num_tokens": 7790638.0,
      "step": 476
    },
    {
      "entropy": 0.5368072241544724,
      "epoch": 1.7798507462686568,
      "grad_norm": 0.17024141550064087,
      "learning_rate": 0.0002,
      "loss": 0.5320987701416016,
      "mean_token_accuracy": 0.7863357067108154,
      "num_tokens": 7806970.0,
      "step": 477
    },
    {
      "entropy": 0.5397102534770966,
      "epoch": 1.783582089552239,
      "grad_norm": 0.14083008468151093,
      "learning_rate": 0.0002,
      "loss": 0.5472857356071472,
      "mean_token_accuracy": 0.780105248093605,
      "num_tokens": 7823336.0,
      "step": 478
    },
    {
      "entropy": 0.539318397641182,
      "epoch": 1.787313432835821,
      "grad_norm": 0.1556725651025772,
      "learning_rate": 0.0002,
      "loss": 0.5453104972839355,
      "mean_token_accuracy": 0.7782669812440872,
      "num_tokens": 7839709.0,
      "step": 479
    },
    {
      "entropy": 0.520709365606308,
      "epoch": 1.7910447761194028,
      "grad_norm": 0.13168865442276,
      "learning_rate": 0.0002,
      "loss": 0.5259016752243042,
      "mean_token_accuracy": 0.7881457805633545,
      "num_tokens": 7856282.0,
      "step": 480
    },
    {
      "entropy": 0.545460969209671,
      "epoch": 1.794776119402985,
      "grad_norm": 0.14062927663326263,
      "learning_rate": 0.0002,
      "loss": 0.5431666374206543,
      "mean_token_accuracy": 0.7789504528045654,
      "num_tokens": 7872796.0,
      "step": 481
    },
    {
      "entropy": 0.5574689209461212,
      "epoch": 1.7985074626865671,
      "grad_norm": 0.14690035581588745,
      "learning_rate": 0.0002,
      "loss": 0.5516860485076904,
      "mean_token_accuracy": 0.7780975550413132,
      "num_tokens": 7889346.0,
      "step": 482
    },
    {
      "entropy": 0.5629963129758835,
      "epoch": 1.8022388059701493,
      "grad_norm": 0.1399964839220047,
      "learning_rate": 0.0002,
      "loss": 0.5579118132591248,
      "mean_token_accuracy": 0.7713749259710312,
      "num_tokens": 7905664.0,
      "step": 483
    },
    {
      "entropy": 0.5299117416143417,
      "epoch": 1.8059701492537314,
      "grad_norm": 0.14648693799972534,
      "learning_rate": 0.0002,
      "loss": 0.5380065441131592,
      "mean_token_accuracy": 0.7824157625436783,
      "num_tokens": 7922022.0,
      "step": 484
    },
    {
      "entropy": 0.535229966044426,
      "epoch": 1.8097014925373134,
      "grad_norm": 0.14146099984645844,
      "learning_rate": 0.0002,
      "loss": 0.5405075550079346,
      "mean_token_accuracy": 0.7798210084438324,
      "num_tokens": 7938249.0,
      "step": 485
    },
    {
      "entropy": 0.5422546565532684,
      "epoch": 1.8134328358208955,
      "grad_norm": 0.1448005735874176,
      "learning_rate": 0.0002,
      "loss": 0.5470380187034607,
      "mean_token_accuracy": 0.7798198163509369,
      "num_tokens": 7954768.0,
      "step": 486
    },
    {
      "entropy": 0.5476532280445099,
      "epoch": 1.8171641791044775,
      "grad_norm": 0.12321429699659348,
      "learning_rate": 0.0002,
      "loss": 0.5507278442382812,
      "mean_token_accuracy": 0.7762424349784851,
      "num_tokens": 7971197.0,
      "step": 487
    },
    {
      "entropy": 0.5327108651399612,
      "epoch": 1.8208955223880596,
      "grad_norm": 0.12477955222129822,
      "learning_rate": 0.0002,
      "loss": 0.5317014455795288,
      "mean_token_accuracy": 0.784480556845665,
      "num_tokens": 7987444.0,
      "step": 488
    },
    {
      "entropy": 0.5364231020212173,
      "epoch": 1.8246268656716418,
      "grad_norm": 0.13437515497207642,
      "learning_rate": 0.0002,
      "loss": 0.5306515097618103,
      "mean_token_accuracy": 0.786288782954216,
      "num_tokens": 8003853.0,
      "step": 489
    },
    {
      "entropy": 0.5527881532907486,
      "epoch": 1.828358208955224,
      "grad_norm": 0.13964197039604187,
      "learning_rate": 0.0002,
      "loss": 0.5511640906333923,
      "mean_token_accuracy": 0.7761706560850143,
      "num_tokens": 8020229.0,
      "step": 490
    },
    {
      "entropy": 0.5417826473712921,
      "epoch": 1.832089552238806,
      "grad_norm": 0.13031242787837982,
      "learning_rate": 0.0002,
      "loss": 0.5431976318359375,
      "mean_token_accuracy": 0.7820132970809937,
      "num_tokens": 8036611.0,
      "step": 491
    },
    {
      "entropy": 0.5438458323478699,
      "epoch": 1.835820895522388,
      "grad_norm": 0.14999718964099884,
      "learning_rate": 0.0002,
      "loss": 0.5554797649383545,
      "mean_token_accuracy": 0.7736136466264725,
      "num_tokens": 8053063.0,
      "step": 492
    },
    {
      "entropy": 0.5380668789148331,
      "epoch": 1.8395522388059702,
      "grad_norm": 0.15141451358795166,
      "learning_rate": 0.0002,
      "loss": 0.5367497205734253,
      "mean_token_accuracy": 0.7832521051168442,
      "num_tokens": 8069525.0,
      "step": 493
    },
    {
      "entropy": 0.548671081662178,
      "epoch": 1.8432835820895521,
      "grad_norm": 0.1565960794687271,
      "learning_rate": 0.0002,
      "loss": 0.5547431111335754,
      "mean_token_accuracy": 0.7750902473926544,
      "num_tokens": 8085710.0,
      "step": 494
    },
    {
      "entropy": 0.5554158538579941,
      "epoch": 1.8470149253731343,
      "grad_norm": 0.1457189917564392,
      "learning_rate": 0.0002,
      "loss": 0.5487250685691833,
      "mean_token_accuracy": 0.7737509161233902,
      "num_tokens": 8102122.0,
      "step": 495
    },
    {
      "entropy": 0.5348504483699799,
      "epoch": 1.8507462686567164,
      "grad_norm": 0.12991224229335785,
      "learning_rate": 0.0002,
      "loss": 0.5250347256660461,
      "mean_token_accuracy": 0.7887249141931534,
      "num_tokens": 8118639.0,
      "step": 496
    },
    {
      "entropy": 0.546437993645668,
      "epoch": 1.8544776119402986,
      "grad_norm": 0.1408451795578003,
      "learning_rate": 0.0002,
      "loss": 0.5394275188446045,
      "mean_token_accuracy": 0.7830011993646622,
      "num_tokens": 8134829.0,
      "step": 497
    },
    {
      "entropy": 0.5353506356477737,
      "epoch": 1.8582089552238807,
      "grad_norm": 0.14167442917823792,
      "learning_rate": 0.0002,
      "loss": 0.5389787554740906,
      "mean_token_accuracy": 0.7805581986904144,
      "num_tokens": 8151204.0,
      "step": 498
    },
    {
      "entropy": 0.5407468229532242,
      "epoch": 1.8619402985074627,
      "grad_norm": 0.13700367510318756,
      "learning_rate": 0.0002,
      "loss": 0.5465103983879089,
      "mean_token_accuracy": 0.7782802432775497,
      "num_tokens": 8167312.0,
      "step": 499
    },
    {
      "entropy": 0.5279424637556076,
      "epoch": 1.8656716417910446,
      "grad_norm": 0.13276584446430206,
      "learning_rate": 0.0002,
      "loss": 0.534116804599762,
      "mean_token_accuracy": 0.7825368642807007,
      "num_tokens": 8183754.0,
      "step": 500
    },
    {
      "entropy": 0.5125058591365814,
      "epoch": 1.8694029850746268,
      "grad_norm": 0.13462357223033905,
      "learning_rate": 0.0002,
      "loss": 0.5232669711112976,
      "mean_token_accuracy": 0.7883776277303696,
      "num_tokens": 8200117.0,
      "step": 501
    },
    {
      "entropy": 0.5494843274354935,
      "epoch": 1.873134328358209,
      "grad_norm": 0.12468113750219345,
      "learning_rate": 0.0002,
      "loss": 0.5489933490753174,
      "mean_token_accuracy": 0.7755606025457382,
      "num_tokens": 8216477.0,
      "step": 502
    },
    {
      "entropy": 0.5506572872400284,
      "epoch": 1.876865671641791,
      "grad_norm": 0.13064922392368317,
      "learning_rate": 0.0002,
      "loss": 0.5484140515327454,
      "mean_token_accuracy": 0.7770435959100723,
      "num_tokens": 8232845.0,
      "step": 503
    },
    {
      "entropy": 0.5596420019865036,
      "epoch": 1.8805970149253732,
      "grad_norm": 0.13143454492092133,
      "learning_rate": 0.0002,
      "loss": 0.563188910484314,
      "mean_token_accuracy": 0.7689026594161987,
      "num_tokens": 8249050.0,
      "step": 504
    },
    {
      "entropy": 0.5608761608600616,
      "epoch": 1.8843283582089554,
      "grad_norm": 0.14789830148220062,
      "learning_rate": 0.0002,
      "loss": 0.5517562031745911,
      "mean_token_accuracy": 0.778393492102623,
      "num_tokens": 8265557.0,
      "step": 505
    },
    {
      "entropy": 0.542554572224617,
      "epoch": 1.8880597014925373,
      "grad_norm": 0.12269016355276108,
      "learning_rate": 0.0002,
      "loss": 0.5380637645721436,
      "mean_token_accuracy": 0.779810756444931,
      "num_tokens": 8281884.0,
      "step": 506
    },
    {
      "entropy": 0.5385424941778183,
      "epoch": 1.8917910447761193,
      "grad_norm": 0.13651514053344727,
      "learning_rate": 0.0002,
      "loss": 0.5387696027755737,
      "mean_token_accuracy": 0.7821653783321381,
      "num_tokens": 8298240.0,
      "step": 507
    },
    {
      "entropy": 0.5442920178174973,
      "epoch": 1.8955223880597014,
      "grad_norm": 0.13530083000659943,
      "learning_rate": 0.0002,
      "loss": 0.5518702268600464,
      "mean_token_accuracy": 0.7749216854572296,
      "num_tokens": 8314544.0,
      "step": 508
    },
    {
      "entropy": 0.5436656922101974,
      "epoch": 1.8992537313432836,
      "grad_norm": 0.1347070187330246,
      "learning_rate": 0.0002,
      "loss": 0.5466428399085999,
      "mean_token_accuracy": 0.775749608874321,
      "num_tokens": 8330811.0,
      "step": 509
    },
    {
      "entropy": 0.550523191690445,
      "epoch": 1.9029850746268657,
      "grad_norm": 0.14112962782382965,
      "learning_rate": 0.0002,
      "loss": 0.5441328883171082,
      "mean_token_accuracy": 0.7798765450716019,
      "num_tokens": 8347372.0,
      "step": 510
    },
    {
      "entropy": 0.5401795655488968,
      "epoch": 1.9067164179104479,
      "grad_norm": 0.12165411561727524,
      "learning_rate": 0.0002,
      "loss": 0.5412328243255615,
      "mean_token_accuracy": 0.7819360196590424,
      "num_tokens": 8363698.0,
      "step": 511
    },
    {
      "entropy": 0.5304841473698616,
      "epoch": 1.9104477611940298,
      "grad_norm": 0.14741350710391998,
      "learning_rate": 0.0002,
      "loss": 0.5355175733566284,
      "mean_token_accuracy": 0.7825563848018646,
      "num_tokens": 8380004.0,
      "step": 512
    },
    {
      "entropy": 0.5147908627986908,
      "epoch": 1.914179104477612,
      "grad_norm": 0.15777826309204102,
      "learning_rate": 0.0002,
      "loss": 0.5264052152633667,
      "mean_token_accuracy": 0.7863651365041733,
      "num_tokens": 8396196.0,
      "step": 513
    },
    {
      "entropy": 0.515343189239502,
      "epoch": 1.917910447761194,
      "grad_norm": 0.1895928829908371,
      "learning_rate": 0.0002,
      "loss": 0.5264019966125488,
      "mean_token_accuracy": 0.7855760753154755,
      "num_tokens": 8412493.0,
      "step": 514
    },
    {
      "entropy": 0.559312716126442,
      "epoch": 1.921641791044776,
      "grad_norm": 0.13089579343795776,
      "learning_rate": 0.0002,
      "loss": 0.5520235896110535,
      "mean_token_accuracy": 0.7749026119709015,
      "num_tokens": 8428665.0,
      "step": 515
    },
    {
      "entropy": 0.5633579939603806,
      "epoch": 1.9253731343283582,
      "grad_norm": 0.16620893776416779,
      "learning_rate": 0.0002,
      "loss": 0.5516705513000488,
      "mean_token_accuracy": 0.7784135341644287,
      "num_tokens": 8445148.0,
      "step": 516
    },
    {
      "entropy": 0.5537055283784866,
      "epoch": 1.9291044776119404,
      "grad_norm": 0.1369117945432663,
      "learning_rate": 0.0002,
      "loss": 0.5524970293045044,
      "mean_token_accuracy": 0.7750401049852371,
      "num_tokens": 8461516.0,
      "step": 517
    },
    {
      "entropy": 0.5486184060573578,
      "epoch": 1.9328358208955225,
      "grad_norm": 0.1374250203371048,
      "learning_rate": 0.0002,
      "loss": 0.5492095351219177,
      "mean_token_accuracy": 0.7750917971134186,
      "num_tokens": 8478027.0,
      "step": 518
    },
    {
      "entropy": 0.5284916907548904,
      "epoch": 1.9365671641791045,
      "grad_norm": 0.12916702032089233,
      "learning_rate": 0.0002,
      "loss": 0.5282076597213745,
      "mean_token_accuracy": 0.7868338376283646,
      "num_tokens": 8494299.0,
      "step": 519
    },
    {
      "entropy": 0.537935420870781,
      "epoch": 1.9402985074626866,
      "grad_norm": 0.12493748217821121,
      "learning_rate": 0.0002,
      "loss": 0.5427759885787964,
      "mean_token_accuracy": 0.7799347043037415,
      "num_tokens": 8510691.0,
      "step": 520
    },
    {
      "entropy": 0.5350519120693207,
      "epoch": 1.9440298507462686,
      "grad_norm": 0.13286834955215454,
      "learning_rate": 0.0002,
      "loss": 0.5356943607330322,
      "mean_token_accuracy": 0.7813293039798737,
      "num_tokens": 8527053.0,
      "step": 521
    },
    {
      "entropy": 0.5486050397157669,
      "epoch": 1.9477611940298507,
      "grad_norm": 0.1205226331949234,
      "learning_rate": 0.0002,
      "loss": 0.5451630353927612,
      "mean_token_accuracy": 0.7807651460170746,
      "num_tokens": 8543637.0,
      "step": 522
    },
    {
      "entropy": 0.5378240644931793,
      "epoch": 1.9514925373134329,
      "grad_norm": 0.12455619126558304,
      "learning_rate": 0.0002,
      "loss": 0.5361875891685486,
      "mean_token_accuracy": 0.7842144221067429,
      "num_tokens": 8560228.0,
      "step": 523
    },
    {
      "entropy": 0.5332168340682983,
      "epoch": 1.955223880597015,
      "grad_norm": 0.13837487995624542,
      "learning_rate": 0.0002,
      "loss": 0.5303863883018494,
      "mean_token_accuracy": 0.7861669808626175,
      "num_tokens": 8576561.0,
      "step": 524
    },
    {
      "entropy": 0.5415726602077484,
      "epoch": 1.9589552238805972,
      "grad_norm": 0.16169637441635132,
      "learning_rate": 0.0002,
      "loss": 0.5505262613296509,
      "mean_token_accuracy": 0.7772628217935562,
      "num_tokens": 8592912.0,
      "step": 525
    },
    {
      "entropy": 0.5102374628186226,
      "epoch": 1.962686567164179,
      "grad_norm": 0.1450612097978592,
      "learning_rate": 0.0002,
      "loss": 0.5185758471488953,
      "mean_token_accuracy": 0.7882694453001022,
      "num_tokens": 8608991.0,
      "step": 526
    },
    {
      "entropy": 0.531574010848999,
      "epoch": 1.966417910447761,
      "grad_norm": 0.15838107466697693,
      "learning_rate": 0.0002,
      "loss": 0.5359771847724915,
      "mean_token_accuracy": 0.7811269164085388,
      "num_tokens": 8625515.0,
      "step": 527
    },
    {
      "entropy": 0.5479525923728943,
      "epoch": 1.9701492537313432,
      "grad_norm": 0.13464049994945526,
      "learning_rate": 0.0002,
      "loss": 0.5461106300354004,
      "mean_token_accuracy": 0.7799096554517746,
      "num_tokens": 8642125.0,
      "step": 528
    },
    {
      "entropy": 0.5310268998146057,
      "epoch": 1.9738805970149254,
      "grad_norm": 0.13331666588783264,
      "learning_rate": 0.0002,
      "loss": 0.5311203002929688,
      "mean_token_accuracy": 0.7841774821281433,
      "num_tokens": 8658254.0,
      "step": 529
    },
    {
      "entropy": 0.5335047841072083,
      "epoch": 1.9776119402985075,
      "grad_norm": 0.15320999920368195,
      "learning_rate": 0.0002,
      "loss": 0.5368505716323853,
      "mean_token_accuracy": 0.7829203605651855,
      "num_tokens": 8674386.0,
      "step": 530
    },
    {
      "entropy": 0.5317138433456421,
      "epoch": 1.9813432835820897,
      "grad_norm": 0.15745186805725098,
      "learning_rate": 0.0002,
      "loss": 0.5369722247123718,
      "mean_token_accuracy": 0.7837116867303848,
      "num_tokens": 8690815.0,
      "step": 531
    },
    {
      "entropy": 0.5469349026679993,
      "epoch": 1.9850746268656716,
      "grad_norm": 0.193673238158226,
      "learning_rate": 0.0002,
      "loss": 0.5555710196495056,
      "mean_token_accuracy": 0.7731745839118958,
      "num_tokens": 8707211.0,
      "step": 532
    },
    {
      "entropy": 0.5364779233932495,
      "epoch": 1.9888059701492538,
      "grad_norm": 0.1405423879623413,
      "learning_rate": 0.0002,
      "loss": 0.5334070324897766,
      "mean_token_accuracy": 0.785079225897789,
      "num_tokens": 8723715.0,
      "step": 533
    },
    {
      "entropy": 0.5547333508729935,
      "epoch": 1.9925373134328357,
      "grad_norm": 0.16125163435935974,
      "learning_rate": 0.0002,
      "loss": 0.5444945096969604,
      "mean_token_accuracy": 0.7796527445316315,
      "num_tokens": 8739984.0,
      "step": 534
    },
    {
      "entropy": 0.5208631753921509,
      "epoch": 1.9962686567164178,
      "grad_norm": 0.1466934233903885,
      "learning_rate": 0.0002,
      "loss": 0.5199640989303589,
      "mean_token_accuracy": 0.7862742394208908,
      "num_tokens": 8756190.0,
      "step": 535
    },
    {
      "entropy": 0.5480705946683884,
      "epoch": 2.0,
      "grad_norm": 0.13491855561733246,
      "learning_rate": 0.0002,
      "loss": 0.5436497926712036,
      "mean_token_accuracy": 0.7807738333940506,
      "num_tokens": 8772698.0,
      "step": 536
    },
    {
      "entropy": 0.5415555536746979,
      "epoch": 2.003731343283582,
      "grad_norm": 0.16747047007083893,
      "learning_rate": 0.0002,
      "loss": 0.5319045782089233,
      "mean_token_accuracy": 0.7846594750881195,
      "num_tokens": 8788976.0,
      "step": 537
    },
    {
      "entropy": 0.48524464666843414,
      "epoch": 2.0074626865671643,
      "grad_norm": 0.14733654260635376,
      "learning_rate": 0.0002,
      "loss": 0.4873185455799103,
      "mean_token_accuracy": 0.8014557808637619,
      "num_tokens": 8805065.0,
      "step": 538
    },
    {
      "entropy": 0.5235195159912109,
      "epoch": 2.0111940298507465,
      "grad_norm": 0.2031126618385315,
      "learning_rate": 0.0002,
      "loss": 0.5338194370269775,
      "mean_token_accuracy": 0.7841212600469589,
      "num_tokens": 8821553.0,
      "step": 539
    },
    {
      "entropy": 0.5199090093374252,
      "epoch": 2.014925373134328,
      "grad_norm": 0.16197463870048523,
      "learning_rate": 0.0002,
      "loss": 0.5291742086410522,
      "mean_token_accuracy": 0.7838669121265411,
      "num_tokens": 8837975.0,
      "step": 540
    },
    {
      "entropy": 0.5170688033103943,
      "epoch": 2.0186567164179103,
      "grad_norm": 0.18860118091106415,
      "learning_rate": 0.0002,
      "loss": 0.521155059337616,
      "mean_token_accuracy": 0.7889470607042313,
      "num_tokens": 8854383.0,
      "step": 541
    },
    {
      "entropy": 0.5346346199512482,
      "epoch": 2.0223880597014925,
      "grad_norm": 0.16565288603305817,
      "learning_rate": 0.0002,
      "loss": 0.5241870880126953,
      "mean_token_accuracy": 0.7872694879770279,
      "num_tokens": 8870807.0,
      "step": 542
    },
    {
      "entropy": 0.5326488316059113,
      "epoch": 2.0261194029850746,
      "grad_norm": 0.1526976078748703,
      "learning_rate": 0.0002,
      "loss": 0.5272649526596069,
      "mean_token_accuracy": 0.7865242511034012,
      "num_tokens": 8887089.0,
      "step": 543
    },
    {
      "entropy": 0.524365171790123,
      "epoch": 2.029850746268657,
      "grad_norm": 0.17769154906272888,
      "learning_rate": 0.0002,
      "loss": 0.5222117304801941,
      "mean_token_accuracy": 0.7874642908573151,
      "num_tokens": 8903525.0,
      "step": 544
    },
    {
      "entropy": 0.49923795461654663,
      "epoch": 2.033582089552239,
      "grad_norm": 0.14854110777378082,
      "learning_rate": 0.0002,
      "loss": 0.4976331293582916,
      "mean_token_accuracy": 0.7968524843454361,
      "num_tokens": 8919992.0,
      "step": 545
    },
    {
      "entropy": 0.5152603462338448,
      "epoch": 2.0373134328358207,
      "grad_norm": 0.17432616651058197,
      "learning_rate": 0.0002,
      "loss": 0.5216287970542908,
      "mean_token_accuracy": 0.7912430018186569,
      "num_tokens": 8936625.0,
      "step": 546
    },
    {
      "entropy": 0.5111927315592766,
      "epoch": 2.041044776119403,
      "grad_norm": 0.1844320446252823,
      "learning_rate": 0.0002,
      "loss": 0.5220290422439575,
      "mean_token_accuracy": 0.7908796668052673,
      "num_tokens": 8952910.0,
      "step": 547
    },
    {
      "entropy": 0.5109107345342636,
      "epoch": 2.044776119402985,
      "grad_norm": 0.1629939079284668,
      "learning_rate": 0.0002,
      "loss": 0.5073159337043762,
      "mean_token_accuracy": 0.7964547127485275,
      "num_tokens": 8969270.0,
      "step": 548
    },
    {
      "entropy": 0.5256383270025253,
      "epoch": 2.048507462686567,
      "grad_norm": 0.156361922621727,
      "learning_rate": 0.0002,
      "loss": 0.518945038318634,
      "mean_token_accuracy": 0.7902349829673767,
      "num_tokens": 8985362.0,
      "step": 549
    },
    {
      "entropy": 0.5305684953927994,
      "epoch": 2.0522388059701493,
      "grad_norm": 0.1715325117111206,
      "learning_rate": 0.0002,
      "loss": 0.5272260904312134,
      "mean_token_accuracy": 0.784850001335144,
      "num_tokens": 9001550.0,
      "step": 550
    },
    {
      "entropy": 0.5125356316566467,
      "epoch": 2.0559701492537314,
      "grad_norm": 0.17070426046848297,
      "learning_rate": 0.0002,
      "loss": 0.5080939531326294,
      "mean_token_accuracy": 0.794335201382637,
      "num_tokens": 9017869.0,
      "step": 551
    },
    {
      "entropy": 0.5276556760072708,
      "epoch": 2.0597014925373136,
      "grad_norm": 0.18260441720485687,
      "learning_rate": 0.0002,
      "loss": 0.5346400141716003,
      "mean_token_accuracy": 0.7878844290971756,
      "num_tokens": 9034237.0,
      "step": 552
    },
    {
      "entropy": 0.5142727345228195,
      "epoch": 2.0634328358208953,
      "grad_norm": 0.19874036312103271,
      "learning_rate": 0.0002,
      "loss": 0.5225756764411926,
      "mean_token_accuracy": 0.7883211523294449,
      "num_tokens": 9050631.0,
      "step": 553
    },
    {
      "entropy": 0.520856112241745,
      "epoch": 2.0671641791044775,
      "grad_norm": 0.16145063936710358,
      "learning_rate": 0.0002,
      "loss": 0.5232750177383423,
      "mean_token_accuracy": 0.7890350520610809,
      "num_tokens": 9067250.0,
      "step": 554
    },
    {
      "entropy": 0.5234951823949814,
      "epoch": 2.0708955223880596,
      "grad_norm": 0.14688369631767273,
      "learning_rate": 0.0002,
      "loss": 0.5160607099533081,
      "mean_token_accuracy": 0.7932988256216049,
      "num_tokens": 9083463.0,
      "step": 555
    },
    {
      "entropy": 0.5156868994235992,
      "epoch": 2.074626865671642,
      "grad_norm": 0.17150931060314178,
      "learning_rate": 0.0002,
      "loss": 0.5148377418518066,
      "mean_token_accuracy": 0.7930486649274826,
      "num_tokens": 9099856.0,
      "step": 556
    },
    {
      "entropy": 0.5226755812764168,
      "epoch": 2.078358208955224,
      "grad_norm": 0.14815834164619446,
      "learning_rate": 0.0002,
      "loss": 0.5174975395202637,
      "mean_token_accuracy": 0.7925054877996445,
      "num_tokens": 9116005.0,
      "step": 557
    },
    {
      "entropy": 0.5249863862991333,
      "epoch": 2.082089552238806,
      "grad_norm": 0.15850502252578735,
      "learning_rate": 0.0002,
      "loss": 0.5295873284339905,
      "mean_token_accuracy": 0.7859255969524384,
      "num_tokens": 9132278.0,
      "step": 558
    },
    {
      "entropy": 0.518112450838089,
      "epoch": 2.0858208955223883,
      "grad_norm": 0.17276053130626678,
      "learning_rate": 0.0002,
      "loss": 0.5247409343719482,
      "mean_token_accuracy": 0.787254199385643,
      "num_tokens": 9148718.0,
      "step": 559
    },
    {
      "entropy": 0.5190763473510742,
      "epoch": 2.08955223880597,
      "grad_norm": 0.1838274598121643,
      "learning_rate": 0.0002,
      "loss": 0.5258316397666931,
      "mean_token_accuracy": 0.7868564873933792,
      "num_tokens": 9165132.0,
      "step": 560
    },
    {
      "entropy": 0.5210602208971977,
      "epoch": 2.093283582089552,
      "grad_norm": 0.19322465360164642,
      "learning_rate": 0.0002,
      "loss": 0.5251522660255432,
      "mean_token_accuracy": 0.7876996248960495,
      "num_tokens": 9181610.0,
      "step": 561
    },
    {
      "entropy": 0.5228234678506851,
      "epoch": 2.0970149253731343,
      "grad_norm": 0.16540859639644623,
      "learning_rate": 0.0002,
      "loss": 0.5194794535636902,
      "mean_token_accuracy": 0.7875367105007172,
      "num_tokens": 9198070.0,
      "step": 562
    },
    {
      "entropy": 0.5316733568906784,
      "epoch": 2.1007462686567164,
      "grad_norm": 0.17291492223739624,
      "learning_rate": 0.0002,
      "loss": 0.5229303240776062,
      "mean_token_accuracy": 0.7877513617277145,
      "num_tokens": 9214519.0,
      "step": 563
    },
    {
      "entropy": 0.5199796482920647,
      "epoch": 2.1044776119402986,
      "grad_norm": 0.16843076050281525,
      "learning_rate": 0.0002,
      "loss": 0.5121482610702515,
      "mean_token_accuracy": 0.7908738553524017,
      "num_tokens": 9230646.0,
      "step": 564
    },
    {
      "entropy": 0.5120833367109299,
      "epoch": 2.1082089552238807,
      "grad_norm": 0.19006116688251495,
      "learning_rate": 0.0002,
      "loss": 0.5176136493682861,
      "mean_token_accuracy": 0.7903040647506714,
      "num_tokens": 9246948.0,
      "step": 565
    },
    {
      "entropy": 0.5191686302423477,
      "epoch": 2.111940298507463,
      "grad_norm": 0.17130376398563385,
      "learning_rate": 0.0002,
      "loss": 0.5252852439880371,
      "mean_token_accuracy": 0.787716269493103,
      "num_tokens": 9263461.0,
      "step": 566
    },
    {
      "entropy": 0.5416426509618759,
      "epoch": 2.1156716417910446,
      "grad_norm": 0.1967155933380127,
      "learning_rate": 0.0002,
      "loss": 0.5384277701377869,
      "mean_token_accuracy": 0.7818170785903931,
      "num_tokens": 9279972.0,
      "step": 567
    },
    {
      "entropy": 0.5269709751009941,
      "epoch": 2.1194029850746268,
      "grad_norm": 0.20237617194652557,
      "learning_rate": 0.0002,
      "loss": 0.5330915451049805,
      "mean_token_accuracy": 0.7869119197130203,
      "num_tokens": 9296077.0,
      "step": 568
    },
    {
      "entropy": 0.513814777135849,
      "epoch": 2.123134328358209,
      "grad_norm": 0.15175029635429382,
      "learning_rate": 0.0002,
      "loss": 0.5088481307029724,
      "mean_token_accuracy": 0.7935866266489029,
      "num_tokens": 9312515.0,
      "step": 569
    },
    {
      "entropy": 0.5114565044641495,
      "epoch": 2.126865671641791,
      "grad_norm": 0.18045492470264435,
      "learning_rate": 0.0002,
      "loss": 0.5154776573181152,
      "mean_token_accuracy": 0.7927515804767609,
      "num_tokens": 9328705.0,
      "step": 570
    },
    {
      "entropy": 0.5144636854529381,
      "epoch": 2.1305970149253732,
      "grad_norm": 0.17092739045619965,
      "learning_rate": 0.0002,
      "loss": 0.5138760209083557,
      "mean_token_accuracy": 0.7910611927509308,
      "num_tokens": 9344984.0,
      "step": 571
    },
    {
      "entropy": 0.5236328467726707,
      "epoch": 2.1343283582089554,
      "grad_norm": 0.173468679189682,
      "learning_rate": 0.0002,
      "loss": 0.5324080586433411,
      "mean_token_accuracy": 0.7874365448951721,
      "num_tokens": 9361268.0,
      "step": 572
    },
    {
      "entropy": 0.5344418287277222,
      "epoch": 2.138059701492537,
      "grad_norm": 0.18645711243152618,
      "learning_rate": 0.0002,
      "loss": 0.5357177257537842,
      "mean_token_accuracy": 0.7840316742658615,
      "num_tokens": 9377667.0,
      "step": 573
    },
    {
      "entropy": 0.5042809993028641,
      "epoch": 2.1417910447761193,
      "grad_norm": 0.16691593825817108,
      "learning_rate": 0.0002,
      "loss": 0.5108679533004761,
      "mean_token_accuracy": 0.7975978702306747,
      "num_tokens": 9394086.0,
      "step": 574
    },
    {
      "entropy": 0.5369294136762619,
      "epoch": 2.1455223880597014,
      "grad_norm": 0.15230222046375275,
      "learning_rate": 0.0002,
      "loss": 0.5277273058891296,
      "mean_token_accuracy": 0.7882997840642929,
      "num_tokens": 9410646.0,
      "step": 575
    },
    {
      "entropy": 0.5273502916097641,
      "epoch": 2.1492537313432836,
      "grad_norm": 0.1738549768924713,
      "learning_rate": 0.0002,
      "loss": 0.5199886560440063,
      "mean_token_accuracy": 0.7918457239866257,
      "num_tokens": 9427220.0,
      "step": 576
    },
    {
      "entropy": 0.5034138485789299,
      "epoch": 2.1529850746268657,
      "grad_norm": 0.16484306752681732,
      "learning_rate": 0.0002,
      "loss": 0.5062033534049988,
      "mean_token_accuracy": 0.7932880967855453,
      "num_tokens": 9443779.0,
      "step": 577
    },
    {
      "entropy": 0.4976464584469795,
      "epoch": 2.156716417910448,
      "grad_norm": 0.16964028775691986,
      "learning_rate": 0.0002,
      "loss": 0.5047153234481812,
      "mean_token_accuracy": 0.7968799024820328,
      "num_tokens": 9460169.0,
      "step": 578
    },
    {
      "entropy": 0.525898888707161,
      "epoch": 2.16044776119403,
      "grad_norm": 0.1925138384103775,
      "learning_rate": 0.0002,
      "loss": 0.5280981063842773,
      "mean_token_accuracy": 0.7853529304265976,
      "num_tokens": 9476731.0,
      "step": 579
    },
    {
      "entropy": 0.5017904862761497,
      "epoch": 2.1641791044776117,
      "grad_norm": 0.15841910243034363,
      "learning_rate": 0.0002,
      "loss": 0.5021662712097168,
      "mean_token_accuracy": 0.7965309321880341,
      "num_tokens": 9493013.0,
      "step": 580
    },
    {
      "entropy": 0.5236378312110901,
      "epoch": 2.167910447761194,
      "grad_norm": 0.1764245629310608,
      "learning_rate": 0.0002,
      "loss": 0.5250015258789062,
      "mean_token_accuracy": 0.7886385321617126,
      "num_tokens": 9509467.0,
      "step": 581
    },
    {
      "entropy": 0.5085260570049286,
      "epoch": 2.171641791044776,
      "grad_norm": 0.16477051377296448,
      "learning_rate": 0.0002,
      "loss": 0.5121427774429321,
      "mean_token_accuracy": 0.7920020669698715,
      "num_tokens": 9525521.0,
      "step": 582
    },
    {
      "entropy": 0.5277596637606621,
      "epoch": 2.175373134328358,
      "grad_norm": 0.18558384478092194,
      "learning_rate": 0.0002,
      "loss": 0.5275318026542664,
      "mean_token_accuracy": 0.7874853909015656,
      "num_tokens": 9541855.0,
      "step": 583
    },
    {
      "entropy": 0.51266248524189,
      "epoch": 2.1791044776119404,
      "grad_norm": 0.1635850965976715,
      "learning_rate": 0.0002,
      "loss": 0.5173766613006592,
      "mean_token_accuracy": 0.7898177653551102,
      "num_tokens": 9558180.0,
      "step": 584
    },
    {
      "entropy": 0.5215219110250473,
      "epoch": 2.1828358208955225,
      "grad_norm": 0.16504792869091034,
      "learning_rate": 0.0002,
      "loss": 0.5152029395103455,
      "mean_token_accuracy": 0.7906715422868729,
      "num_tokens": 9574414.0,
      "step": 585
    },
    {
      "entropy": 0.5250913053750992,
      "epoch": 2.1865671641791047,
      "grad_norm": 0.1531083583831787,
      "learning_rate": 0.0002,
      "loss": 0.517106294631958,
      "mean_token_accuracy": 0.7902461886405945,
      "num_tokens": 9590957.0,
      "step": 586
    },
    {
      "entropy": 0.5339035540819168,
      "epoch": 2.1902985074626864,
      "grad_norm": 0.1628551036119461,
      "learning_rate": 0.0002,
      "loss": 0.5370537638664246,
      "mean_token_accuracy": 0.7826921045780182,
      "num_tokens": 9607334.0,
      "step": 587
    },
    {
      "entropy": 0.5297398567199707,
      "epoch": 2.1940298507462686,
      "grad_norm": 0.17225806415081024,
      "learning_rate": 0.0002,
      "loss": 0.5343117117881775,
      "mean_token_accuracy": 0.7868678122758865,
      "num_tokens": 9623620.0,
      "step": 588
    },
    {
      "entropy": 0.5233481600880623,
      "epoch": 2.1977611940298507,
      "grad_norm": 0.16789408028125763,
      "learning_rate": 0.0002,
      "loss": 0.5252033472061157,
      "mean_token_accuracy": 0.7889025807380676,
      "num_tokens": 9639976.0,
      "step": 589
    },
    {
      "entropy": 0.5225154161453247,
      "epoch": 2.201492537313433,
      "grad_norm": 0.15730316936969757,
      "learning_rate": 0.0002,
      "loss": 0.5257684588432312,
      "mean_token_accuracy": 0.7879121899604797,
      "num_tokens": 9656150.0,
      "step": 590
    },
    {
      "entropy": 0.5133989825844765,
      "epoch": 2.205223880597015,
      "grad_norm": 0.2035355418920517,
      "learning_rate": 0.0002,
      "loss": 0.5088110566139221,
      "mean_token_accuracy": 0.795003280043602,
      "num_tokens": 9672593.0,
      "step": 591
    },
    {
      "entropy": 0.545501321554184,
      "epoch": 2.208955223880597,
      "grad_norm": 0.21898981928825378,
      "learning_rate": 0.0002,
      "loss": 0.5424832105636597,
      "mean_token_accuracy": 0.7800484746694565,
      "num_tokens": 9689022.0,
      "step": 592
    },
    {
      "entropy": 0.5240471512079239,
      "epoch": 2.2126865671641793,
      "grad_norm": 0.15258178114891052,
      "learning_rate": 0.0002,
      "loss": 0.530670166015625,
      "mean_token_accuracy": 0.7840432077646255,
      "num_tokens": 9705598.0,
      "step": 593
    },
    {
      "entropy": 0.5081499069929123,
      "epoch": 2.216417910447761,
      "grad_norm": 0.20431077480316162,
      "learning_rate": 0.0002,
      "loss": 0.5145601630210876,
      "mean_token_accuracy": 0.7950203865766525,
      "num_tokens": 9721825.0,
      "step": 594
    },
    {
      "entropy": 0.5313006192445755,
      "epoch": 2.220149253731343,
      "grad_norm": 0.1535511016845703,
      "learning_rate": 0.0002,
      "loss": 0.5350795984268188,
      "mean_token_accuracy": 0.7831084877252579,
      "num_tokens": 9738558.0,
      "step": 595
    },
    {
      "entropy": 0.5182467773556709,
      "epoch": 2.2238805970149254,
      "grad_norm": 0.16815528273582458,
      "learning_rate": 0.0002,
      "loss": 0.5121225118637085,
      "mean_token_accuracy": 0.7931758910417557,
      "num_tokens": 9754808.0,
      "step": 596
    },
    {
      "entropy": 0.5362052470445633,
      "epoch": 2.2276119402985075,
      "grad_norm": 0.17921388149261475,
      "learning_rate": 0.0002,
      "loss": 0.5349149107933044,
      "mean_token_accuracy": 0.7824008017778397,
      "num_tokens": 9770959.0,
      "step": 597
    },
    {
      "entropy": 0.535765752196312,
      "epoch": 2.2313432835820897,
      "grad_norm": 0.18234150111675262,
      "learning_rate": 0.0002,
      "loss": 0.5311421751976013,
      "mean_token_accuracy": 0.7860585302114487,
      "num_tokens": 9787452.0,
      "step": 598
    },
    {
      "entropy": 0.5220657736063004,
      "epoch": 2.235074626865672,
      "grad_norm": 0.17075759172439575,
      "learning_rate": 0.0002,
      "loss": 0.5269136428833008,
      "mean_token_accuracy": 0.7844899594783783,
      "num_tokens": 9803883.0,
      "step": 599
    },
    {
      "entropy": 0.5282145738601685,
      "epoch": 2.2388059701492535,
      "grad_norm": 0.15000498294830322,
      "learning_rate": 0.0002,
      "loss": 0.5264820456504822,
      "mean_token_accuracy": 0.7889179140329361,
      "num_tokens": 9820304.0,
      "step": 600
    },
    {
      "entropy": 0.5101103484630585,
      "epoch": 2.2425373134328357,
      "grad_norm": 0.21740344166755676,
      "learning_rate": 0.0002,
      "loss": 0.5152019262313843,
      "mean_token_accuracy": 0.7896611094474792,
      "num_tokens": 9836738.0,
      "step": 601
    },
    {
      "entropy": 0.5106130391359329,
      "epoch": 2.246268656716418,
      "grad_norm": 0.17091670632362366,
      "learning_rate": 0.0002,
      "loss": 0.5152319073677063,
      "mean_token_accuracy": 0.7931098341941833,
      "num_tokens": 9853169.0,
      "step": 602
    },
    {
      "entropy": 0.4893152490258217,
      "epoch": 2.25,
      "grad_norm": 0.22172100841999054,
      "learning_rate": 0.0002,
      "loss": 0.49760663509368896,
      "mean_token_accuracy": 0.7992293536663055,
      "num_tokens": 9869296.0,
      "step": 603
    },
    {
      "entropy": 0.5243295952677727,
      "epoch": 2.253731343283582,
      "grad_norm": 0.18062254786491394,
      "learning_rate": 0.0002,
      "loss": 0.5286306142807007,
      "mean_token_accuracy": 0.7865812629461288,
      "num_tokens": 9885786.0,
      "step": 604
    },
    {
      "entropy": 0.5476033389568329,
      "epoch": 2.2574626865671643,
      "grad_norm": 0.17773279547691345,
      "learning_rate": 0.0002,
      "loss": 0.5444879531860352,
      "mean_token_accuracy": 0.780517116189003,
      "num_tokens": 9902385.0,
      "step": 605
    },
    {
      "entropy": 0.528127908706665,
      "epoch": 2.2611940298507465,
      "grad_norm": 0.1832549273967743,
      "learning_rate": 0.0002,
      "loss": 0.5193147659301758,
      "mean_token_accuracy": 0.7898916155099869,
      "num_tokens": 9918852.0,
      "step": 606
    },
    {
      "entropy": 0.5321992486715317,
      "epoch": 2.264925373134328,
      "grad_norm": 0.17879848182201385,
      "learning_rate": 0.0002,
      "loss": 0.5303541421890259,
      "mean_token_accuracy": 0.7855431735515594,
      "num_tokens": 9935218.0,
      "step": 607
    },
    {
      "entropy": 0.5326562970876694,
      "epoch": 2.2686567164179103,
      "grad_norm": 0.18682479858398438,
      "learning_rate": 0.0002,
      "loss": 0.5352398753166199,
      "mean_token_accuracy": 0.7856907397508621,
      "num_tokens": 9951691.0,
      "step": 608
    },
    {
      "entropy": 0.5085463523864746,
      "epoch": 2.2723880597014925,
      "grad_norm": 0.1803983896970749,
      "learning_rate": 0.0002,
      "loss": 0.5086592435836792,
      "mean_token_accuracy": 0.7945016622543335,
      "num_tokens": 9967916.0,
      "step": 609
    },
    {
      "entropy": 0.5094747915863991,
      "epoch": 2.2761194029850746,
      "grad_norm": 0.179136261343956,
      "learning_rate": 0.0002,
      "loss": 0.5116256475448608,
      "mean_token_accuracy": 0.7906947284936905,
      "num_tokens": 9984543.0,
      "step": 610
    },
    {
      "entropy": 0.5107395797967911,
      "epoch": 2.279850746268657,
      "grad_norm": 0.1843739002943039,
      "learning_rate": 0.0002,
      "loss": 0.5126347541809082,
      "mean_token_accuracy": 0.7935709655284882,
      "num_tokens": 10000841.0,
      "step": 611
    },
    {
      "entropy": 0.510215513408184,
      "epoch": 2.283582089552239,
      "grad_norm": 0.1749034970998764,
      "learning_rate": 0.0002,
      "loss": 0.5147651433944702,
      "mean_token_accuracy": 0.7939617931842804,
      "num_tokens": 10017104.0,
      "step": 612
    },
    {
      "entropy": 0.5255024135112762,
      "epoch": 2.2873134328358207,
      "grad_norm": 0.17081239819526672,
      "learning_rate": 0.0002,
      "loss": 0.5292178988456726,
      "mean_token_accuracy": 0.7868891507387161,
      "num_tokens": 10033661.0,
      "step": 613
    },
    {
      "entropy": 0.5174866542220116,
      "epoch": 2.291044776119403,
      "grad_norm": 0.17191831767559052,
      "learning_rate": 0.0002,
      "loss": 0.5210599899291992,
      "mean_token_accuracy": 0.7896179109811783,
      "num_tokens": 10050314.0,
      "step": 614
    },
    {
      "entropy": 0.5216618627309799,
      "epoch": 2.294776119402985,
      "grad_norm": 0.15834467113018036,
      "learning_rate": 0.0002,
      "loss": 0.5203002095222473,
      "mean_token_accuracy": 0.7914630323648453,
      "num_tokens": 10066724.0,
      "step": 615
    },
    {
      "entropy": 0.5273145735263824,
      "epoch": 2.298507462686567,
      "grad_norm": 0.2328452616930008,
      "learning_rate": 0.0002,
      "loss": 0.5253832340240479,
      "mean_token_accuracy": 0.7888303846120834,
      "num_tokens": 10083031.0,
      "step": 616
    },
    {
      "entropy": 0.5163110569119453,
      "epoch": 2.3022388059701493,
      "grad_norm": 0.1779068261384964,
      "learning_rate": 0.0002,
      "loss": 0.5165371298789978,
      "mean_token_accuracy": 0.7911377400159836,
      "num_tokens": 10099378.0,
      "step": 617
    },
    {
      "entropy": 0.5036482810974121,
      "epoch": 2.3059701492537314,
      "grad_norm": 0.19564983248710632,
      "learning_rate": 0.0002,
      "loss": 0.5062058568000793,
      "mean_token_accuracy": 0.7941041439771652,
      "num_tokens": 10115305.0,
      "step": 618
    },
    {
      "entropy": 0.5294737368822098,
      "epoch": 2.3097014925373136,
      "grad_norm": 0.20425698161125183,
      "learning_rate": 0.0002,
      "loss": 0.5274835228919983,
      "mean_token_accuracy": 0.7902465760707855,
      "num_tokens": 10131724.0,
      "step": 619
    },
    {
      "entropy": 0.538879469037056,
      "epoch": 2.3134328358208958,
      "grad_norm": 0.17032799124717712,
      "learning_rate": 0.0002,
      "loss": 0.5420734882354736,
      "mean_token_accuracy": 0.7821353524923325,
      "num_tokens": 10148541.0,
      "step": 620
    },
    {
      "entropy": 0.5180075243115425,
      "epoch": 2.3171641791044775,
      "grad_norm": 0.1744605004787445,
      "learning_rate": 0.0002,
      "loss": 0.5202793478965759,
      "mean_token_accuracy": 0.7892477065324783,
      "num_tokens": 10164856.0,
      "step": 621
    },
    {
      "entropy": 0.5143277049064636,
      "epoch": 2.3208955223880596,
      "grad_norm": 0.16975471377372742,
      "learning_rate": 0.0002,
      "loss": 0.5149828791618347,
      "mean_token_accuracy": 0.7906354814767838,
      "num_tokens": 10181035.0,
      "step": 622
    },
    {
      "entropy": 0.5316905975341797,
      "epoch": 2.324626865671642,
      "grad_norm": 0.17982955276966095,
      "learning_rate": 0.0002,
      "loss": 0.5393372774124146,
      "mean_token_accuracy": 0.7804277241230011,
      "num_tokens": 10197720.0,
      "step": 623
    },
    {
      "entropy": 0.5204255282878876,
      "epoch": 2.328358208955224,
      "grad_norm": 0.17796067893505096,
      "learning_rate": 0.0002,
      "loss": 0.5173559188842773,
      "mean_token_accuracy": 0.7897128164768219,
      "num_tokens": 10214065.0,
      "step": 624
    },
    {
      "entropy": 0.5189772769808769,
      "epoch": 2.332089552238806,
      "grad_norm": 0.1623089611530304,
      "learning_rate": 0.0002,
      "loss": 0.5185124278068542,
      "mean_token_accuracy": 0.7876813858747482,
      "num_tokens": 10230429.0,
      "step": 625
    },
    {
      "entropy": 0.5281235128641129,
      "epoch": 2.3358208955223883,
      "grad_norm": 0.1610279232263565,
      "learning_rate": 0.0002,
      "loss": 0.5221237540245056,
      "mean_token_accuracy": 0.7886802852153778,
      "num_tokens": 10246874.0,
      "step": 626
    },
    {
      "entropy": 0.5154699981212616,
      "epoch": 2.33955223880597,
      "grad_norm": 0.1539958268404007,
      "learning_rate": 0.0002,
      "loss": 0.511885404586792,
      "mean_token_accuracy": 0.793968677520752,
      "num_tokens": 10263249.0,
      "step": 627
    },
    {
      "entropy": 0.5257471352815628,
      "epoch": 2.343283582089552,
      "grad_norm": 0.17384476959705353,
      "learning_rate": 0.0002,
      "loss": 0.5300131440162659,
      "mean_token_accuracy": 0.7854046821594238,
      "num_tokens": 10279865.0,
      "step": 628
    },
    {
      "entropy": 0.5260587707161903,
      "epoch": 2.3470149253731343,
      "grad_norm": 0.16657280921936035,
      "learning_rate": 0.0002,
      "loss": 0.530137836933136,
      "mean_token_accuracy": 0.7845129370689392,
      "num_tokens": 10296201.0,
      "step": 629
    },
    {
      "entropy": 0.4986217990517616,
      "epoch": 2.3507462686567164,
      "grad_norm": 0.15695442259311676,
      "learning_rate": 0.0002,
      "loss": 0.5049688220024109,
      "mean_token_accuracy": 0.7965218275785446,
      "num_tokens": 10312457.0,
      "step": 630
    },
    {
      "entropy": 0.527582585811615,
      "epoch": 2.3544776119402986,
      "grad_norm": 0.1535857915878296,
      "learning_rate": 0.0002,
      "loss": 0.5306364297866821,
      "mean_token_accuracy": 0.7822943180799484,
      "num_tokens": 10328828.0,
      "step": 631
    },
    {
      "entropy": 0.5110795050859451,
      "epoch": 2.3582089552238807,
      "grad_norm": 0.16837763786315918,
      "learning_rate": 0.0002,
      "loss": 0.5116544365882874,
      "mean_token_accuracy": 0.792694553732872,
      "num_tokens": 10345109.0,
      "step": 632
    },
    {
      "entropy": 0.5123824924230576,
      "epoch": 2.361940298507463,
      "grad_norm": 0.16515249013900757,
      "learning_rate": 0.0002,
      "loss": 0.5103446841239929,
      "mean_token_accuracy": 0.7953357100486755,
      "num_tokens": 10361054.0,
      "step": 633
    },
    {
      "entropy": 0.5145570486783981,
      "epoch": 2.3656716417910446,
      "grad_norm": 0.16593779623508453,
      "learning_rate": 0.0002,
      "loss": 0.5132509469985962,
      "mean_token_accuracy": 0.7917768955230713,
      "num_tokens": 10377534.0,
      "step": 634
    },
    {
      "entropy": 0.5097534209489822,
      "epoch": 2.3694029850746268,
      "grad_norm": 0.17235712707042694,
      "learning_rate": 0.0002,
      "loss": 0.5129758715629578,
      "mean_token_accuracy": 0.7928901016712189,
      "num_tokens": 10394069.0,
      "step": 635
    },
    {
      "entropy": 0.5342534631490707,
      "epoch": 2.373134328358209,
      "grad_norm": 0.16720189154148102,
      "learning_rate": 0.0002,
      "loss": 0.5387343764305115,
      "mean_token_accuracy": 0.7836835235357285,
      "num_tokens": 10410499.0,
      "step": 636
    },
    {
      "entropy": 0.5487189590930939,
      "epoch": 2.376865671641791,
      "grad_norm": 0.16905991733074188,
      "learning_rate": 0.0002,
      "loss": 0.5453655123710632,
      "mean_token_accuracy": 0.7782279253005981,
      "num_tokens": 10426781.0,
      "step": 637
    },
    {
      "entropy": 0.5357492566108704,
      "epoch": 2.3805970149253732,
      "grad_norm": 0.16518373787403107,
      "learning_rate": 0.0002,
      "loss": 0.5330023169517517,
      "mean_token_accuracy": 0.7854349464178085,
      "num_tokens": 10443016.0,
      "step": 638
    },
    {
      "entropy": 0.5225549787282944,
      "epoch": 2.3843283582089554,
      "grad_norm": 0.16202834248542786,
      "learning_rate": 0.0002,
      "loss": 0.5225259065628052,
      "mean_token_accuracy": 0.7870278358459473,
      "num_tokens": 10459201.0,
      "step": 639
    },
    {
      "entropy": 0.5189392864704132,
      "epoch": 2.388059701492537,
      "grad_norm": 0.17072422802448273,
      "learning_rate": 0.0002,
      "loss": 0.5200708508491516,
      "mean_token_accuracy": 0.791137307882309,
      "num_tokens": 10475466.0,
      "step": 640
    },
    {
      "entropy": 0.5257381796836853,
      "epoch": 2.3917910447761193,
      "grad_norm": 0.15416349470615387,
      "learning_rate": 0.0002,
      "loss": 0.5287957787513733,
      "mean_token_accuracy": 0.7854514420032501,
      "num_tokens": 10491844.0,
      "step": 641
    },
    {
      "entropy": 0.5229465961456299,
      "epoch": 2.3955223880597014,
      "grad_norm": 0.1715656965970993,
      "learning_rate": 0.0002,
      "loss": 0.5251911282539368,
      "mean_token_accuracy": 0.787058562040329,
      "num_tokens": 10508043.0,
      "step": 642
    },
    {
      "entropy": 0.5345896184444427,
      "epoch": 2.3992537313432836,
      "grad_norm": 0.14537544548511505,
      "learning_rate": 0.0002,
      "loss": 0.5331792831420898,
      "mean_token_accuracy": 0.7870064675807953,
      "num_tokens": 10524515.0,
      "step": 643
    },
    {
      "entropy": 0.513592854142189,
      "epoch": 2.4029850746268657,
      "grad_norm": 0.1578165739774704,
      "learning_rate": 0.0002,
      "loss": 0.5107499361038208,
      "mean_token_accuracy": 0.7935400903224945,
      "num_tokens": 10541212.0,
      "step": 644
    },
    {
      "entropy": 0.5219082683324814,
      "epoch": 2.406716417910448,
      "grad_norm": 0.15485122799873352,
      "learning_rate": 0.0002,
      "loss": 0.5251319408416748,
      "mean_token_accuracy": 0.7842298001050949,
      "num_tokens": 10557429.0,
      "step": 645
    },
    {
      "entropy": 0.532458558678627,
      "epoch": 2.41044776119403,
      "grad_norm": 0.15033091604709625,
      "learning_rate": 0.0002,
      "loss": 0.5344883799552917,
      "mean_token_accuracy": 0.7839881181716919,
      "num_tokens": 10573640.0,
      "step": 646
    },
    {
      "entropy": 0.5289199203252792,
      "epoch": 2.4141791044776117,
      "grad_norm": 0.18606211245059967,
      "learning_rate": 0.0002,
      "loss": 0.5404102206230164,
      "mean_token_accuracy": 0.780749723315239,
      "num_tokens": 10589928.0,
      "step": 647
    },
    {
      "entropy": 0.5123477131128311,
      "epoch": 2.417910447761194,
      "grad_norm": 0.17192338407039642,
      "learning_rate": 0.0002,
      "loss": 0.5178117156028748,
      "mean_token_accuracy": 0.7884255647659302,
      "num_tokens": 10606192.0,
      "step": 648
    },
    {
      "entropy": 0.5328754633665085,
      "epoch": 2.421641791044776,
      "grad_norm": 0.171258345246315,
      "learning_rate": 0.0002,
      "loss": 0.5254883766174316,
      "mean_token_accuracy": 0.7896005064249039,
      "num_tokens": 10622566.0,
      "step": 649
    },
    {
      "entropy": 0.5208312794566154,
      "epoch": 2.425373134328358,
      "grad_norm": 0.14705301821231842,
      "learning_rate": 0.0002,
      "loss": 0.5148872137069702,
      "mean_token_accuracy": 0.7895035594701767,
      "num_tokens": 10638936.0,
      "step": 650
    },
    {
      "entropy": 0.5213367491960526,
      "epoch": 2.4291044776119404,
      "grad_norm": 0.18489304184913635,
      "learning_rate": 0.0002,
      "loss": 0.5248196721076965,
      "mean_token_accuracy": 0.7895573377609253,
      "num_tokens": 10655160.0,
      "step": 651
    },
    {
      "entropy": 0.5255916714668274,
      "epoch": 2.4328358208955225,
      "grad_norm": 0.17082132399082184,
      "learning_rate": 0.0002,
      "loss": 0.5310555100440979,
      "mean_token_accuracy": 0.7847766876220703,
      "num_tokens": 10671609.0,
      "step": 652
    },
    {
      "entropy": 0.550238773226738,
      "epoch": 2.4365671641791042,
      "grad_norm": 0.20729617774486542,
      "learning_rate": 0.0002,
      "loss": 0.5431200861930847,
      "mean_token_accuracy": 0.7801288962364197,
      "num_tokens": 10688113.0,
      "step": 653
    },
    {
      "entropy": 0.5350706577301025,
      "epoch": 2.4402985074626864,
      "grad_norm": 0.1664358526468277,
      "learning_rate": 0.0002,
      "loss": 0.5351425409317017,
      "mean_token_accuracy": 0.7840354889631271,
      "num_tokens": 10704503.0,
      "step": 654
    },
    {
      "entropy": 0.5174449980258942,
      "epoch": 2.4440298507462686,
      "grad_norm": 0.1792401820421219,
      "learning_rate": 0.0002,
      "loss": 0.5174031257629395,
      "mean_token_accuracy": 0.7921167761087418,
      "num_tokens": 10720444.0,
      "step": 655
    },
    {
      "entropy": 0.5383309870958328,
      "epoch": 2.4477611940298507,
      "grad_norm": 0.18368947505950928,
      "learning_rate": 0.0002,
      "loss": 0.5425350069999695,
      "mean_token_accuracy": 0.7808329910039902,
      "num_tokens": 10736781.0,
      "step": 656
    },
    {
      "entropy": 0.5155188888311386,
      "epoch": 2.451492537313433,
      "grad_norm": 0.1622094362974167,
      "learning_rate": 0.0002,
      "loss": 0.5155419111251831,
      "mean_token_accuracy": 0.7907011955976486,
      "num_tokens": 10753167.0,
      "step": 657
    },
    {
      "entropy": 0.5223165601491928,
      "epoch": 2.455223880597015,
      "grad_norm": 0.1702336072921753,
      "learning_rate": 0.0002,
      "loss": 0.5252817869186401,
      "mean_token_accuracy": 0.7866344302892685,
      "num_tokens": 10769443.0,
      "step": 658
    },
    {
      "entropy": 0.5043924450874329,
      "epoch": 2.458955223880597,
      "grad_norm": 0.15965932607650757,
      "learning_rate": 0.0002,
      "loss": 0.5058254599571228,
      "mean_token_accuracy": 0.7947247475385666,
      "num_tokens": 10785622.0,
      "step": 659
    },
    {
      "entropy": 0.5168613493442535,
      "epoch": 2.4626865671641793,
      "grad_norm": 0.16473491489887238,
      "learning_rate": 0.0002,
      "loss": 0.5218056440353394,
      "mean_token_accuracy": 0.7899257838726044,
      "num_tokens": 10802055.0,
      "step": 660
    },
    {
      "entropy": 0.519745409488678,
      "epoch": 2.466417910447761,
      "grad_norm": 0.16005174815654755,
      "learning_rate": 0.0002,
      "loss": 0.5178749561309814,
      "mean_token_accuracy": 0.7917293906211853,
      "num_tokens": 10818314.0,
      "step": 661
    },
    {
      "entropy": 0.5039018765091896,
      "epoch": 2.470149253731343,
      "grad_norm": 0.1764734983444214,
      "learning_rate": 0.0002,
      "loss": 0.5094691514968872,
      "mean_token_accuracy": 0.7939349561929703,
      "num_tokens": 10834559.0,
      "step": 662
    },
    {
      "entropy": 0.5233125612139702,
      "epoch": 2.4738805970149254,
      "grad_norm": 0.1721738576889038,
      "learning_rate": 0.0002,
      "loss": 0.5221680998802185,
      "mean_token_accuracy": 0.7892500162124634,
      "num_tokens": 10851044.0,
      "step": 663
    },
    {
      "entropy": 0.5263130366802216,
      "epoch": 2.4776119402985075,
      "grad_norm": 0.18548409640789032,
      "learning_rate": 0.0002,
      "loss": 0.5293120741844177,
      "mean_token_accuracy": 0.784709244966507,
      "num_tokens": 10867340.0,
      "step": 664
    },
    {
      "entropy": 0.5338780581951141,
      "epoch": 2.4813432835820897,
      "grad_norm": 0.18075405061244965,
      "learning_rate": 0.0002,
      "loss": 0.527806282043457,
      "mean_token_accuracy": 0.7845254093408585,
      "num_tokens": 10883511.0,
      "step": 665
    },
    {
      "entropy": 0.5315570682287216,
      "epoch": 2.485074626865672,
      "grad_norm": 0.1618206650018692,
      "learning_rate": 0.0002,
      "loss": 0.5266122221946716,
      "mean_token_accuracy": 0.7876931577920914,
      "num_tokens": 10900116.0,
      "step": 666
    },
    {
      "entropy": 0.5129313841462135,
      "epoch": 2.4888059701492535,
      "grad_norm": 0.17066992819309235,
      "learning_rate": 0.0002,
      "loss": 0.5172396898269653,
      "mean_token_accuracy": 0.7914125770330429,
      "num_tokens": 10916502.0,
      "step": 667
    },
    {
      "entropy": 0.5208132937550545,
      "epoch": 2.4925373134328357,
      "grad_norm": 0.16389311850070953,
      "learning_rate": 0.0002,
      "loss": 0.5225825309753418,
      "mean_token_accuracy": 0.7878787815570831,
      "num_tokens": 10932922.0,
      "step": 668
    },
    {
      "entropy": 0.5138225853443146,
      "epoch": 2.496268656716418,
      "grad_norm": 0.19532781839370728,
      "learning_rate": 0.0002,
      "loss": 0.5174206495285034,
      "mean_token_accuracy": 0.7901126593351364,
      "num_tokens": 10949268.0,
      "step": 669
    },
    {
      "entropy": 0.5288973152637482,
      "epoch": 2.5,
      "grad_norm": 0.18305917084217072,
      "learning_rate": 0.0002,
      "loss": 0.5321277976036072,
      "mean_token_accuracy": 0.7827803492546082,
      "num_tokens": 10965727.0,
      "step": 670
    },
    {
      "entropy": 0.5190397500991821,
      "epoch": 2.503731343283582,
      "grad_norm": 0.17453818023204803,
      "learning_rate": 0.0002,
      "loss": 0.5205980539321899,
      "mean_token_accuracy": 0.79007987678051,
      "num_tokens": 10982192.0,
      "step": 671
    },
    {
      "entropy": 0.5292644500732422,
      "epoch": 2.5074626865671643,
      "grad_norm": 0.20140023529529572,
      "learning_rate": 0.0002,
      "loss": 0.5297321677207947,
      "mean_token_accuracy": 0.7868858128786087,
      "num_tokens": 10998542.0,
      "step": 672
    },
    {
      "entropy": 0.5147685110569,
      "epoch": 2.5111940298507465,
      "grad_norm": 0.14919306337833405,
      "learning_rate": 0.0002,
      "loss": 0.5168865919113159,
      "mean_token_accuracy": 0.7919680327177048,
      "num_tokens": 11014794.0,
      "step": 673
    },
    {
      "entropy": 0.5530222207307816,
      "epoch": 2.5149253731343286,
      "grad_norm": 0.15733702480793,
      "learning_rate": 0.0002,
      "loss": 0.5499193072319031,
      "mean_token_accuracy": 0.7764365673065186,
      "num_tokens": 11031367.0,
      "step": 674
    },
    {
      "entropy": 0.5164763256907463,
      "epoch": 2.5186567164179103,
      "grad_norm": 0.1580049991607666,
      "learning_rate": 0.0002,
      "loss": 0.5119961500167847,
      "mean_token_accuracy": 0.7906799763441086,
      "num_tokens": 11047708.0,
      "step": 675
    },
    {
      "entropy": 0.5175507366657257,
      "epoch": 2.5223880597014925,
      "grad_norm": 0.1583462953567505,
      "learning_rate": 0.0002,
      "loss": 0.5124553442001343,
      "mean_token_accuracy": 0.7930836081504822,
      "num_tokens": 11064166.0,
      "step": 676
    },
    {
      "entropy": 0.5099252760410309,
      "epoch": 2.5261194029850746,
      "grad_norm": 0.20494359731674194,
      "learning_rate": 0.0002,
      "loss": 0.5234544277191162,
      "mean_token_accuracy": 0.7916188836097717,
      "num_tokens": 11080383.0,
      "step": 677
    },
    {
      "entropy": 0.5219163745641708,
      "epoch": 2.529850746268657,
      "grad_norm": 0.23150977492332458,
      "learning_rate": 0.0002,
      "loss": 0.5278502106666565,
      "mean_token_accuracy": 0.7889153361320496,
      "num_tokens": 11096929.0,
      "step": 678
    },
    {
      "entropy": 0.5226232409477234,
      "epoch": 2.533582089552239,
      "grad_norm": 0.1982942670583725,
      "learning_rate": 0.0002,
      "loss": 0.520408570766449,
      "mean_token_accuracy": 0.7901227027177811,
      "num_tokens": 11113257.0,
      "step": 679
    },
    {
      "entropy": 0.5349150896072388,
      "epoch": 2.5373134328358207,
      "grad_norm": 0.1624055802822113,
      "learning_rate": 0.0002,
      "loss": 0.5286049842834473,
      "mean_token_accuracy": 0.7859060317277908,
      "num_tokens": 11129395.0,
      "step": 680
    },
    {
      "entropy": 0.528727650642395,
      "epoch": 2.541044776119403,
      "grad_norm": 0.17963045835494995,
      "learning_rate": 0.0002,
      "loss": 0.5162175893783569,
      "mean_token_accuracy": 0.789388120174408,
      "num_tokens": 11145773.0,
      "step": 681
    },
    {
      "entropy": 0.5230469256639481,
      "epoch": 2.544776119402985,
      "grad_norm": 0.17094683647155762,
      "learning_rate": 0.0002,
      "loss": 0.519881546497345,
      "mean_token_accuracy": 0.7906030863523483,
      "num_tokens": 11162276.0,
      "step": 682
    },
    {
      "entropy": 0.5128089189529419,
      "epoch": 2.548507462686567,
      "grad_norm": 0.2320418804883957,
      "learning_rate": 0.0002,
      "loss": 0.5217410922050476,
      "mean_token_accuracy": 0.7926544845104218,
      "num_tokens": 11178886.0,
      "step": 683
    },
    {
      "entropy": 0.5072636231780052,
      "epoch": 2.5522388059701493,
      "grad_norm": 0.20943979918956757,
      "learning_rate": 0.0002,
      "loss": 0.5183896422386169,
      "mean_token_accuracy": 0.7905592769384384,
      "num_tokens": 11195352.0,
      "step": 684
    },
    {
      "entropy": 0.5097889825701714,
      "epoch": 2.5559701492537314,
      "grad_norm": 0.23722805082798004,
      "learning_rate": 0.0002,
      "loss": 0.525998592376709,
      "mean_token_accuracy": 0.7886322885751724,
      "num_tokens": 11211521.0,
      "step": 685
    },
    {
      "entropy": 0.539807990193367,
      "epoch": 2.5597014925373136,
      "grad_norm": 0.17591507732868195,
      "learning_rate": 0.0002,
      "loss": 0.5344754457473755,
      "mean_token_accuracy": 0.7813717275857925,
      "num_tokens": 11227860.0,
      "step": 686
    },
    {
      "entropy": 0.5342103391885757,
      "epoch": 2.5634328358208958,
      "grad_norm": 0.22380377352237701,
      "learning_rate": 0.0002,
      "loss": 0.5241360664367676,
      "mean_token_accuracy": 0.7901870459318161,
      "num_tokens": 11244158.0,
      "step": 687
    },
    {
      "entropy": 0.5307934135198593,
      "epoch": 2.5671641791044775,
      "grad_norm": 0.1869865357875824,
      "learning_rate": 0.0002,
      "loss": 0.5165374279022217,
      "mean_token_accuracy": 0.7908752858638763,
      "num_tokens": 11260377.0,
      "step": 688
    },
    {
      "entropy": 0.5231481194496155,
      "epoch": 2.5708955223880596,
      "grad_norm": 0.18755362927913666,
      "learning_rate": 0.0002,
      "loss": 0.5221779346466064,
      "mean_token_accuracy": 0.7877349406480789,
      "num_tokens": 11276910.0,
      "step": 689
    },
    {
      "entropy": 0.5327332317829132,
      "epoch": 2.574626865671642,
      "grad_norm": 0.17062468826770782,
      "learning_rate": 0.0002,
      "loss": 0.5431654453277588,
      "mean_token_accuracy": 0.7820072174072266,
      "num_tokens": 11293421.0,
      "step": 690
    },
    {
      "entropy": 0.5024554505944252,
      "epoch": 2.578358208955224,
      "grad_norm": 0.2268306016921997,
      "learning_rate": 0.0002,
      "loss": 0.5210269689559937,
      "mean_token_accuracy": 0.7899546772241592,
      "num_tokens": 11309810.0,
      "step": 691
    },
    {
      "entropy": 0.5187607258558273,
      "epoch": 2.582089552238806,
      "grad_norm": 0.16152653098106384,
      "learning_rate": 0.0002,
      "loss": 0.5280716419219971,
      "mean_token_accuracy": 0.785857617855072,
      "num_tokens": 11326292.0,
      "step": 692
    },
    {
      "entropy": 0.5381567031145096,
      "epoch": 2.585820895522388,
      "grad_norm": 0.21659867465496063,
      "learning_rate": 0.0002,
      "loss": 0.5352873802185059,
      "mean_token_accuracy": 0.7846846282482147,
      "num_tokens": 11342795.0,
      "step": 693
    },
    {
      "entropy": 0.5329597592353821,
      "epoch": 2.58955223880597,
      "grad_norm": 0.16229940950870514,
      "learning_rate": 0.0002,
      "loss": 0.5267969965934753,
      "mean_token_accuracy": 0.7852964401245117,
      "num_tokens": 11359010.0,
      "step": 694
    },
    {
      "entropy": 0.5313838422298431,
      "epoch": 2.593283582089552,
      "grad_norm": 0.15890458226203918,
      "learning_rate": 0.0002,
      "loss": 0.523610532283783,
      "mean_token_accuracy": 0.7865879833698273,
      "num_tokens": 11375332.0,
      "step": 695
    },
    {
      "entropy": 0.5463909059762955,
      "epoch": 2.5970149253731343,
      "grad_norm": 0.18537728488445282,
      "learning_rate": 0.0002,
      "loss": 0.5428441166877747,
      "mean_token_accuracy": 0.7778408825397491,
      "num_tokens": 11391757.0,
      "step": 696
    },
    {
      "entropy": 0.5092201456427574,
      "epoch": 2.6007462686567164,
      "grad_norm": 0.15471166372299194,
      "learning_rate": 0.0002,
      "loss": 0.5099161267280579,
      "mean_token_accuracy": 0.7928211987018585,
      "num_tokens": 11408002.0,
      "step": 697
    },
    {
      "entropy": 0.5340137034654617,
      "epoch": 2.6044776119402986,
      "grad_norm": 0.21168433129787445,
      "learning_rate": 0.0002,
      "loss": 0.5471012592315674,
      "mean_token_accuracy": 0.7793057262897491,
      "num_tokens": 11424244.0,
      "step": 698
    },
    {
      "entropy": 0.5221166610717773,
      "epoch": 2.6082089552238807,
      "grad_norm": 0.17584562301635742,
      "learning_rate": 0.0002,
      "loss": 0.5273703336715698,
      "mean_token_accuracy": 0.7858802080154419,
      "num_tokens": 11440796.0,
      "step": 699
    },
    {
      "entropy": 0.5195273905992508,
      "epoch": 2.611940298507463,
      "grad_norm": 0.1487855613231659,
      "learning_rate": 0.0002,
      "loss": 0.5190231800079346,
      "mean_token_accuracy": 0.7890651226043701,
      "num_tokens": 11457341.0,
      "step": 700
    },
    {
      "entropy": 0.5417843014001846,
      "epoch": 2.6156716417910446,
      "grad_norm": 0.17057086527347565,
      "learning_rate": 0.0002,
      "loss": 0.5328060388565063,
      "mean_token_accuracy": 0.7850722223520279,
      "num_tokens": 11473884.0,
      "step": 701
    },
    {
      "entropy": 0.5344310849905014,
      "epoch": 2.6194029850746268,
      "grad_norm": 0.14398950338363647,
      "learning_rate": 0.0002,
      "loss": 0.523583173751831,
      "mean_token_accuracy": 0.7903382778167725,
      "num_tokens": 11490186.0,
      "step": 702
    },
    {
      "entropy": 0.5337594449520111,
      "epoch": 2.623134328358209,
      "grad_norm": 0.16912032663822174,
      "learning_rate": 0.0002,
      "loss": 0.531908392906189,
      "mean_token_accuracy": 0.7852459698915482,
      "num_tokens": 11506306.0,
      "step": 703
    },
    {
      "entropy": 0.5263167321681976,
      "epoch": 2.626865671641791,
      "grad_norm": 0.1840488761663437,
      "learning_rate": 0.0002,
      "loss": 0.5250154733657837,
      "mean_token_accuracy": 0.7851651459932327,
      "num_tokens": 11522847.0,
      "step": 704
    },
    {
      "entropy": 0.528788149356842,
      "epoch": 2.6305970149253732,
      "grad_norm": 0.16106311976909637,
      "learning_rate": 0.0002,
      "loss": 0.5311395525932312,
      "mean_token_accuracy": 0.7847187519073486,
      "num_tokens": 11539030.0,
      "step": 705
    },
    {
      "entropy": 0.5256313234567642,
      "epoch": 2.6343283582089554,
      "grad_norm": 0.18702206015586853,
      "learning_rate": 0.0002,
      "loss": 0.5318302512168884,
      "mean_token_accuracy": 0.7854705154895782,
      "num_tokens": 11555666.0,
      "step": 706
    },
    {
      "entropy": 0.5093299522995949,
      "epoch": 2.638059701492537,
      "grad_norm": 0.18550799787044525,
      "learning_rate": 0.0002,
      "loss": 0.5174624919891357,
      "mean_token_accuracy": 0.7912192046642303,
      "num_tokens": 11571933.0,
      "step": 707
    },
    {
      "entropy": 0.5121369138360023,
      "epoch": 2.6417910447761193,
      "grad_norm": 0.2079463005065918,
      "learning_rate": 0.0002,
      "loss": 0.5148282051086426,
      "mean_token_accuracy": 0.7929982990026474,
      "num_tokens": 11588305.0,
      "step": 708
    },
    {
      "entropy": 0.522641733288765,
      "epoch": 2.6455223880597014,
      "grad_norm": 0.17375299334526062,
      "learning_rate": 0.0002,
      "loss": 0.5264861583709717,
      "mean_token_accuracy": 0.7887682616710663,
      "num_tokens": 11604656.0,
      "step": 709
    },
    {
      "entropy": 0.5334865152835846,
      "epoch": 2.6492537313432836,
      "grad_norm": 0.19092123210430145,
      "learning_rate": 0.0002,
      "loss": 0.5361115336418152,
      "mean_token_accuracy": 0.7828067094087601,
      "num_tokens": 11621215.0,
      "step": 710
    },
    {
      "entropy": 0.5242611020803452,
      "epoch": 2.6529850746268657,
      "grad_norm": 0.17106467485427856,
      "learning_rate": 0.0002,
      "loss": 0.5184009075164795,
      "mean_token_accuracy": 0.7890250086784363,
      "num_tokens": 11637461.0,
      "step": 711
    },
    {
      "entropy": 0.5254231840372086,
      "epoch": 2.656716417910448,
      "grad_norm": 0.18828749656677246,
      "learning_rate": 0.0002,
      "loss": 0.5220206379890442,
      "mean_token_accuracy": 0.7883093357086182,
      "num_tokens": 11653872.0,
      "step": 712
    },
    {
      "entropy": 0.5288625285029411,
      "epoch": 2.66044776119403,
      "grad_norm": 0.19284789264202118,
      "learning_rate": 0.0002,
      "loss": 0.5286439061164856,
      "mean_token_accuracy": 0.7847935855388641,
      "num_tokens": 11670124.0,
      "step": 713
    },
    {
      "entropy": 0.516718402504921,
      "epoch": 2.664179104477612,
      "grad_norm": 0.15631504356861115,
      "learning_rate": 0.0002,
      "loss": 0.5174189805984497,
      "mean_token_accuracy": 0.7906842827796936,
      "num_tokens": 11686470.0,
      "step": 714
    },
    {
      "entropy": 0.5242748856544495,
      "epoch": 2.667910447761194,
      "grad_norm": 0.18477752804756165,
      "learning_rate": 0.0002,
      "loss": 0.5323954224586487,
      "mean_token_accuracy": 0.7831654697656631,
      "num_tokens": 11702937.0,
      "step": 715
    },
    {
      "entropy": 0.5290126353502274,
      "epoch": 2.671641791044776,
      "grad_norm": 0.18125174939632416,
      "learning_rate": 0.0002,
      "loss": 0.5322421193122864,
      "mean_token_accuracy": 0.7816368043422699,
      "num_tokens": 11719240.0,
      "step": 716
    },
    {
      "entropy": 0.5167457982897758,
      "epoch": 2.675373134328358,
      "grad_norm": 0.17377378046512604,
      "learning_rate": 0.0002,
      "loss": 0.51507169008255,
      "mean_token_accuracy": 0.7903617173433304,
      "num_tokens": 11735409.0,
      "step": 717
    },
    {
      "entropy": 0.5312926769256592,
      "epoch": 2.6791044776119404,
      "grad_norm": 0.15681293606758118,
      "learning_rate": 0.0002,
      "loss": 0.5237995982170105,
      "mean_token_accuracy": 0.7873582690954208,
      "num_tokens": 11751881.0,
      "step": 718
    },
    {
      "entropy": 0.5318104922771454,
      "epoch": 2.6828358208955225,
      "grad_norm": 0.16546019911766052,
      "learning_rate": 0.0002,
      "loss": 0.5251667499542236,
      "mean_token_accuracy": 0.7856694608926773,
      "num_tokens": 11768225.0,
      "step": 719
    },
    {
      "entropy": 0.5237040966749191,
      "epoch": 2.6865671641791042,
      "grad_norm": 0.16682998836040497,
      "learning_rate": 0.0002,
      "loss": 0.5231617093086243,
      "mean_token_accuracy": 0.7875684052705765,
      "num_tokens": 11784620.0,
      "step": 720
    },
    {
      "entropy": 0.5202368199825287,
      "epoch": 2.6902985074626864,
      "grad_norm": 0.17968019843101501,
      "learning_rate": 0.0002,
      "loss": 0.5251814723014832,
      "mean_token_accuracy": 0.78860142827034,
      "num_tokens": 11801068.0,
      "step": 721
    },
    {
      "entropy": 0.5152329802513123,
      "epoch": 2.6940298507462686,
      "grad_norm": 0.18279340863227844,
      "learning_rate": 0.0002,
      "loss": 0.5201072096824646,
      "mean_token_accuracy": 0.788761630654335,
      "num_tokens": 11817789.0,
      "step": 722
    },
    {
      "entropy": 0.5422751754522324,
      "epoch": 2.6977611940298507,
      "grad_norm": 0.17404738068580627,
      "learning_rate": 0.0002,
      "loss": 0.5462705492973328,
      "mean_token_accuracy": 0.7783142030239105,
      "num_tokens": 11834379.0,
      "step": 723
    },
    {
      "entropy": 0.5129429250955582,
      "epoch": 2.701492537313433,
      "grad_norm": 0.18814703822135925,
      "learning_rate": 0.0002,
      "loss": 0.5165575742721558,
      "mean_token_accuracy": 0.7904448211193085,
      "num_tokens": 11850770.0,
      "step": 724
    },
    {
      "entropy": 0.5254894196987152,
      "epoch": 2.705223880597015,
      "grad_norm": 0.15707333385944366,
      "learning_rate": 0.0002,
      "loss": 0.5243353247642517,
      "mean_token_accuracy": 0.7852540165185928,
      "num_tokens": 11866937.0,
      "step": 725
    },
    {
      "entropy": 0.5248469561338425,
      "epoch": 2.708955223880597,
      "grad_norm": 0.16662286221981049,
      "learning_rate": 0.0002,
      "loss": 0.525855302810669,
      "mean_token_accuracy": 0.7861558347940445,
      "num_tokens": 11883190.0,
      "step": 726
    },
    {
      "entropy": 0.5052276030182838,
      "epoch": 2.7126865671641793,
      "grad_norm": 0.17599958181381226,
      "learning_rate": 0.0002,
      "loss": 0.5062177181243896,
      "mean_token_accuracy": 0.7966409027576447,
      "num_tokens": 11899680.0,
      "step": 727
    },
    {
      "entropy": 0.5214319974184036,
      "epoch": 2.716417910447761,
      "grad_norm": 0.16147606074810028,
      "learning_rate": 0.0002,
      "loss": 0.5187070965766907,
      "mean_token_accuracy": 0.7905149608850479,
      "num_tokens": 11916105.0,
      "step": 728
    },
    {
      "entropy": 0.5107331871986389,
      "epoch": 2.720149253731343,
      "grad_norm": 0.1838311105966568,
      "learning_rate": 0.0002,
      "loss": 0.5223504900932312,
      "mean_token_accuracy": 0.7879439145326614,
      "num_tokens": 11932339.0,
      "step": 729
    },
    {
      "entropy": 0.5096628740429878,
      "epoch": 2.7238805970149254,
      "grad_norm": 0.19706125557422638,
      "learning_rate": 0.0002,
      "loss": 0.5217396020889282,
      "mean_token_accuracy": 0.7888819575309753,
      "num_tokens": 11948596.0,
      "step": 730
    },
    {
      "entropy": 0.5494230389595032,
      "epoch": 2.7276119402985075,
      "grad_norm": 0.1703970581293106,
      "learning_rate": 0.0002,
      "loss": 0.547927975654602,
      "mean_token_accuracy": 0.777910441160202,
      "num_tokens": 11964932.0,
      "step": 731
    },
    {
      "entropy": 0.5448969304561615,
      "epoch": 2.7313432835820897,
      "grad_norm": 0.18176127970218658,
      "learning_rate": 0.0002,
      "loss": 0.5407285094261169,
      "mean_token_accuracy": 0.7788923978805542,
      "num_tokens": 11981433.0,
      "step": 732
    },
    {
      "entropy": 0.5229097306728363,
      "epoch": 2.7350746268656714,
      "grad_norm": 0.15667755901813507,
      "learning_rate": 0.0002,
      "loss": 0.5159817337989807,
      "mean_token_accuracy": 0.7891620248556137,
      "num_tokens": 11997639.0,
      "step": 733
    },
    {
      "entropy": 0.5191904157400131,
      "epoch": 2.7388059701492535,
      "grad_norm": 0.1721811443567276,
      "learning_rate": 0.0002,
      "loss": 0.5218988656997681,
      "mean_token_accuracy": 0.7904060035943985,
      "num_tokens": 12014130.0,
      "step": 734
    },
    {
      "entropy": 0.5449526757001877,
      "epoch": 2.7425373134328357,
      "grad_norm": 0.17720188200473785,
      "learning_rate": 0.0002,
      "loss": 0.5445029139518738,
      "mean_token_accuracy": 0.7790267616510391,
      "num_tokens": 12030672.0,
      "step": 735
    },
    {
      "entropy": 0.5268261879682541,
      "epoch": 2.746268656716418,
      "grad_norm": 0.18524481356143951,
      "learning_rate": 0.0002,
      "loss": 0.5338873267173767,
      "mean_token_accuracy": 0.7801726907491684,
      "num_tokens": 12047144.0,
      "step": 736
    },
    {
      "entropy": 0.5180369466543198,
      "epoch": 2.75,
      "grad_norm": 0.1833166629076004,
      "learning_rate": 0.0002,
      "loss": 0.5154885649681091,
      "mean_token_accuracy": 0.7930953353643417,
      "num_tokens": 12063456.0,
      "step": 737
    },
    {
      "entropy": 0.520864948630333,
      "epoch": 2.753731343283582,
      "grad_norm": 0.17072772979736328,
      "learning_rate": 0.0002,
      "loss": 0.519443154335022,
      "mean_token_accuracy": 0.7902696430683136,
      "num_tokens": 12079749.0,
      "step": 738
    },
    {
      "entropy": 0.5259118229150772,
      "epoch": 2.7574626865671643,
      "grad_norm": 0.15088406205177307,
      "learning_rate": 0.0002,
      "loss": 0.5221176147460938,
      "mean_token_accuracy": 0.7875189185142517,
      "num_tokens": 12095996.0,
      "step": 739
    },
    {
      "entropy": 0.5308933854103088,
      "epoch": 2.7611940298507465,
      "grad_norm": 0.1881408989429474,
      "learning_rate": 0.0002,
      "loss": 0.5294531583786011,
      "mean_token_accuracy": 0.7857530564069748,
      "num_tokens": 12112351.0,
      "step": 740
    },
    {
      "entropy": 0.5012388527393341,
      "epoch": 2.7649253731343286,
      "grad_norm": 0.1454417109489441,
      "learning_rate": 0.0002,
      "loss": 0.4996565282344818,
      "mean_token_accuracy": 0.798053503036499,
      "num_tokens": 12128539.0,
      "step": 741
    },
    {
      "entropy": 0.5164308771491051,
      "epoch": 2.7686567164179103,
      "grad_norm": 0.2045368105173111,
      "learning_rate": 0.0002,
      "loss": 0.5229124426841736,
      "mean_token_accuracy": 0.7889576256275177,
      "num_tokens": 12145010.0,
      "step": 742
    },
    {
      "entropy": 0.5285548865795135,
      "epoch": 2.7723880597014925,
      "grad_norm": 0.1736793965101242,
      "learning_rate": 0.0002,
      "loss": 0.5349996089935303,
      "mean_token_accuracy": 0.7806331962347031,
      "num_tokens": 12161368.0,
      "step": 743
    },
    {
      "entropy": 0.527101144194603,
      "epoch": 2.7761194029850746,
      "grad_norm": 0.16130943596363068,
      "learning_rate": 0.0002,
      "loss": 0.5260635614395142,
      "mean_token_accuracy": 0.7845276594161987,
      "num_tokens": 12177579.0,
      "step": 744
    },
    {
      "entropy": 0.5165161341428757,
      "epoch": 2.779850746268657,
      "grad_norm": 0.15534961223602295,
      "learning_rate": 0.0002,
      "loss": 0.5140593647956848,
      "mean_token_accuracy": 0.7893622666597366,
      "num_tokens": 12193776.0,
      "step": 745
    },
    {
      "entropy": 0.5335810482501984,
      "epoch": 2.783582089552239,
      "grad_norm": 0.16441547870635986,
      "learning_rate": 0.0002,
      "loss": 0.5323783159255981,
      "mean_token_accuracy": 0.7829417288303375,
      "num_tokens": 12210120.0,
      "step": 746
    },
    {
      "entropy": 0.5203997641801834,
      "epoch": 2.7873134328358207,
      "grad_norm": 0.17424358427524567,
      "learning_rate": 0.0002,
      "loss": 0.523624062538147,
      "mean_token_accuracy": 0.7887652665376663,
      "num_tokens": 12226468.0,
      "step": 747
    },
    {
      "entropy": 0.5201438963413239,
      "epoch": 2.791044776119403,
      "grad_norm": 0.16570104658603668,
      "learning_rate": 0.0002,
      "loss": 0.5204302072525024,
      "mean_token_accuracy": 0.7892085611820221,
      "num_tokens": 12243033.0,
      "step": 748
    },
    {
      "entropy": 0.5477077513933182,
      "epoch": 2.794776119402985,
      "grad_norm": 0.17271164059638977,
      "learning_rate": 0.0002,
      "loss": 0.5525671243667603,
      "mean_token_accuracy": 0.7785156220197678,
      "num_tokens": 12259386.0,
      "step": 749
    },
    {
      "entropy": 0.5182037800550461,
      "epoch": 2.798507462686567,
      "grad_norm": 0.17035937309265137,
      "learning_rate": 0.0002,
      "loss": 0.5240148901939392,
      "mean_token_accuracy": 0.7873071283102036,
      "num_tokens": 12275845.0,
      "step": 750
    },
    {
      "entropy": 0.5358012318611145,
      "epoch": 2.8022388059701493,
      "grad_norm": 0.19006162881851196,
      "learning_rate": 0.0002,
      "loss": 0.5386052131652832,
      "mean_token_accuracy": 0.7809128612279892,
      "num_tokens": 12292079.0,
      "step": 751
    },
    {
      "entropy": 0.5293133556842804,
      "epoch": 2.8059701492537314,
      "grad_norm": 0.1824568659067154,
      "learning_rate": 0.0002,
      "loss": 0.5247379541397095,
      "mean_token_accuracy": 0.787973627448082,
      "num_tokens": 12308540.0,
      "step": 752
    },
    {
      "entropy": 0.534746989607811,
      "epoch": 2.8097014925373136,
      "grad_norm": 0.16650667786598206,
      "learning_rate": 0.0002,
      "loss": 0.5251801013946533,
      "mean_token_accuracy": 0.7861672788858414,
      "num_tokens": 12325263.0,
      "step": 753
    },
    {
      "entropy": 0.5277372971177101,
      "epoch": 2.8134328358208958,
      "grad_norm": 0.15979520976543427,
      "learning_rate": 0.0002,
      "loss": 0.5277374386787415,
      "mean_token_accuracy": 0.7879827320575714,
      "num_tokens": 12341509.0,
      "step": 754
    },
    {
      "entropy": 0.5297356992959976,
      "epoch": 2.8171641791044775,
      "grad_norm": 0.1765962392091751,
      "learning_rate": 0.0002,
      "loss": 0.5333434343338013,
      "mean_token_accuracy": 0.7844502031803131,
      "num_tokens": 12358075.0,
      "step": 755
    },
    {
      "entropy": 0.5149746164679527,
      "epoch": 2.8208955223880596,
      "grad_norm": 0.16600866615772247,
      "learning_rate": 0.0002,
      "loss": 0.5245839357376099,
      "mean_token_accuracy": 0.7882574498653412,
      "num_tokens": 12374444.0,
      "step": 756
    },
    {
      "entropy": 0.5162809640169144,
      "epoch": 2.824626865671642,
      "grad_norm": 0.19397495687007904,
      "learning_rate": 0.0002,
      "loss": 0.5219771265983582,
      "mean_token_accuracy": 0.7887652814388275,
      "num_tokens": 12390953.0,
      "step": 757
    },
    {
      "entropy": 0.5194946527481079,
      "epoch": 2.828358208955224,
      "grad_norm": 0.16699662804603577,
      "learning_rate": 0.0002,
      "loss": 0.519737720489502,
      "mean_token_accuracy": 0.7899023741483688,
      "num_tokens": 12407222.0,
      "step": 758
    },
    {
      "entropy": 0.5087470188736916,
      "epoch": 2.832089552238806,
      "grad_norm": 0.18057791888713837,
      "learning_rate": 0.0002,
      "loss": 0.5064352750778198,
      "mean_token_accuracy": 0.7973962873220444,
      "num_tokens": 12423588.0,
      "step": 759
    },
    {
      "entropy": 0.5269856452941895,
      "epoch": 2.835820895522388,
      "grad_norm": 0.15729346871376038,
      "learning_rate": 0.0002,
      "loss": 0.5306705236434937,
      "mean_token_accuracy": 0.7866159975528717,
      "num_tokens": 12439879.0,
      "step": 760
    },
    {
      "entropy": 0.5460618138313293,
      "epoch": 2.83955223880597,
      "grad_norm": 0.18975763022899628,
      "learning_rate": 0.0002,
      "loss": 0.5474591255187988,
      "mean_token_accuracy": 0.781822144985199,
      "num_tokens": 12456072.0,
      "step": 761
    },
    {
      "entropy": 0.5230747014284134,
      "epoch": 2.843283582089552,
      "grad_norm": 0.15568320453166962,
      "learning_rate": 0.0002,
      "loss": 0.5270999670028687,
      "mean_token_accuracy": 0.7864617109298706,
      "num_tokens": 12472226.0,
      "step": 762
    },
    {
      "entropy": 0.5220855101943016,
      "epoch": 2.8470149253731343,
      "grad_norm": 0.15668445825576782,
      "learning_rate": 0.0002,
      "loss": 0.5187541246414185,
      "mean_token_accuracy": 0.790753573179245,
      "num_tokens": 12488527.0,
      "step": 763
    },
    {
      "entropy": 0.5183272808790207,
      "epoch": 2.8507462686567164,
      "grad_norm": 0.14737525582313538,
      "learning_rate": 0.0002,
      "loss": 0.5153676867485046,
      "mean_token_accuracy": 0.792951300740242,
      "num_tokens": 12504908.0,
      "step": 764
    },
    {
      "entropy": 0.5257730633020401,
      "epoch": 2.8544776119402986,
      "grad_norm": 0.18290433287620544,
      "learning_rate": 0.0002,
      "loss": 0.5267256498336792,
      "mean_token_accuracy": 0.7887216210365295,
      "num_tokens": 12521322.0,
      "step": 765
    },
    {
      "entropy": 0.5082066059112549,
      "epoch": 2.8582089552238807,
      "grad_norm": 0.16560977697372437,
      "learning_rate": 0.0002,
      "loss": 0.5087635517120361,
      "mean_token_accuracy": 0.7943163812160492,
      "num_tokens": 12537501.0,
      "step": 766
    },
    {
      "entropy": 0.5251262933015823,
      "epoch": 2.861940298507463,
      "grad_norm": 0.18704728782176971,
      "learning_rate": 0.0002,
      "loss": 0.5243902802467346,
      "mean_token_accuracy": 0.7889377325773239,
      "num_tokens": 12553757.0,
      "step": 767
    },
    {
      "entropy": 0.5206309631466866,
      "epoch": 2.8656716417910446,
      "grad_norm": 0.2072477638721466,
      "learning_rate": 0.0002,
      "loss": 0.5288045406341553,
      "mean_token_accuracy": 0.7842439860105515,
      "num_tokens": 12569984.0,
      "step": 768
    },
    {
      "entropy": 0.5140768364071846,
      "epoch": 2.8694029850746268,
      "grad_norm": 0.18866559863090515,
      "learning_rate": 0.0002,
      "loss": 0.518936276435852,
      "mean_token_accuracy": 0.7901607006788254,
      "num_tokens": 12586343.0,
      "step": 769
    },
    {
      "entropy": 0.5277619361877441,
      "epoch": 2.873134328358209,
      "grad_norm": 0.17521493136882782,
      "learning_rate": 0.0002,
      "loss": 0.5250133872032166,
      "mean_token_accuracy": 0.7883250266313553,
      "num_tokens": 12602788.0,
      "step": 770
    },
    {
      "entropy": 0.5361880958080292,
      "epoch": 2.876865671641791,
      "grad_norm": 0.17607153952121735,
      "learning_rate": 0.0002,
      "loss": 0.5320221185684204,
      "mean_token_accuracy": 0.7843968868255615,
      "num_tokens": 12619338.0,
      "step": 771
    },
    {
      "entropy": 0.5387411415576935,
      "epoch": 2.8805970149253732,
      "grad_norm": 0.16333559155464172,
      "learning_rate": 0.0002,
      "loss": 0.5377700328826904,
      "mean_token_accuracy": 0.7794961333274841,
      "num_tokens": 12635781.0,
      "step": 772
    },
    {
      "entropy": 0.52068030834198,
      "epoch": 2.8843283582089554,
      "grad_norm": 0.2111354023218155,
      "learning_rate": 0.0002,
      "loss": 0.5309345722198486,
      "mean_token_accuracy": 0.7851005345582962,
      "num_tokens": 12652120.0,
      "step": 773
    },
    {
      "entropy": 0.5023355633020401,
      "epoch": 2.888059701492537,
      "grad_norm": 0.16245239973068237,
      "learning_rate": 0.0002,
      "loss": 0.5070939064025879,
      "mean_token_accuracy": 0.7958957403898239,
      "num_tokens": 12668512.0,
      "step": 774
    },
    {
      "entropy": 0.5165709853172302,
      "epoch": 2.8917910447761193,
      "grad_norm": 0.18636101484298706,
      "learning_rate": 0.0002,
      "loss": 0.5146787166595459,
      "mean_token_accuracy": 0.7920748293399811,
      "num_tokens": 12684901.0,
      "step": 775
    },
    {
      "entropy": 0.5335523337125778,
      "epoch": 2.8955223880597014,
      "grad_norm": 0.17786805331707,
      "learning_rate": 0.0002,
      "loss": 0.5301126837730408,
      "mean_token_accuracy": 0.7868293076753616,
      "num_tokens": 12701220.0,
      "step": 776
    },
    {
      "entropy": 0.5350312739610672,
      "epoch": 2.8992537313432836,
      "grad_norm": 0.17426325380802155,
      "learning_rate": 0.0002,
      "loss": 0.53025883436203,
      "mean_token_accuracy": 0.786578357219696,
      "num_tokens": 12717843.0,
      "step": 777
    },
    {
      "entropy": 0.5273053050041199,
      "epoch": 2.9029850746268657,
      "grad_norm": 0.1676986813545227,
      "learning_rate": 0.0002,
      "loss": 0.5316031575202942,
      "mean_token_accuracy": 0.7859497368335724,
      "num_tokens": 12734270.0,
      "step": 778
    },
    {
      "entropy": 0.5331600606441498,
      "epoch": 2.906716417910448,
      "grad_norm": 0.22916914522647858,
      "learning_rate": 0.0002,
      "loss": 0.539795458316803,
      "mean_token_accuracy": 0.7830531746149063,
      "num_tokens": 12750856.0,
      "step": 779
    },
    {
      "entropy": 0.5255026668310165,
      "epoch": 2.91044776119403,
      "grad_norm": 0.201639786362648,
      "learning_rate": 0.0002,
      "loss": 0.5250778198242188,
      "mean_token_accuracy": 0.788204088807106,
      "num_tokens": 12767459.0,
      "step": 780
    },
    {
      "entropy": 0.520592212677002,
      "epoch": 2.914179104477612,
      "grad_norm": 0.18787626922130585,
      "learning_rate": 0.0002,
      "loss": 0.5184063911437988,
      "mean_token_accuracy": 0.7918865531682968,
      "num_tokens": 12783645.0,
      "step": 781
    },
    {
      "entropy": 0.5173375234007835,
      "epoch": 2.917910447761194,
      "grad_norm": 0.21068905293941498,
      "learning_rate": 0.0002,
      "loss": 0.5156338214874268,
      "mean_token_accuracy": 0.7925535291433334,
      "num_tokens": 12799874.0,
      "step": 782
    },
    {
      "entropy": 0.5314591825008392,
      "epoch": 2.921641791044776,
      "grad_norm": 0.18807783722877502,
      "learning_rate": 0.0002,
      "loss": 0.5354687571525574,
      "mean_token_accuracy": 0.782687246799469,
      "num_tokens": 12816237.0,
      "step": 783
    },
    {
      "entropy": 0.5250366032123566,
      "epoch": 2.925373134328358,
      "grad_norm": 0.19177882373332977,
      "learning_rate": 0.0002,
      "loss": 0.5243748426437378,
      "mean_token_accuracy": 0.788783997297287,
      "num_tokens": 12832733.0,
      "step": 784
    },
    {
      "entropy": 0.5160113275051117,
      "epoch": 2.9291044776119404,
      "grad_norm": 0.20948225259780884,
      "learning_rate": 0.0002,
      "loss": 0.5233412981033325,
      "mean_token_accuracy": 0.7861292213201523,
      "num_tokens": 12849053.0,
      "step": 785
    },
    {
      "entropy": 0.5039379075169563,
      "epoch": 2.9328358208955225,
      "grad_norm": 0.19059154391288757,
      "learning_rate": 0.0002,
      "loss": 0.5079476833343506,
      "mean_token_accuracy": 0.7951397895812988,
      "num_tokens": 12865327.0,
      "step": 786
    },
    {
      "entropy": 0.5442841053009033,
      "epoch": 2.9365671641791042,
      "grad_norm": 0.19541315734386444,
      "learning_rate": 0.0002,
      "loss": 0.5416577458381653,
      "mean_token_accuracy": 0.7812930792570114,
      "num_tokens": 12881791.0,
      "step": 787
    },
    {
      "entropy": 0.5215146541595459,
      "epoch": 2.9402985074626864,
      "grad_norm": 0.17175233364105225,
      "learning_rate": 0.0002,
      "loss": 0.5193598866462708,
      "mean_token_accuracy": 0.787645697593689,
      "num_tokens": 12898001.0,
      "step": 788
    },
    {
      "entropy": 0.5333852842450142,
      "epoch": 2.9440298507462686,
      "grad_norm": 0.1919661909341812,
      "learning_rate": 0.0002,
      "loss": 0.5312644839286804,
      "mean_token_accuracy": 0.7849176079034805,
      "num_tokens": 12914404.0,
      "step": 789
    },
    {
      "entropy": 0.5274566262960434,
      "epoch": 2.9477611940298507,
      "grad_norm": 0.16644608974456787,
      "learning_rate": 0.0002,
      "loss": 0.5266693830490112,
      "mean_token_accuracy": 0.7857722342014313,
      "num_tokens": 12930837.0,
      "step": 790
    },
    {
      "entropy": 0.504246324300766,
      "epoch": 2.951492537313433,
      "grad_norm": 0.22507771849632263,
      "learning_rate": 0.0002,
      "loss": 0.5166245698928833,
      "mean_token_accuracy": 0.7927839010953903,
      "num_tokens": 12947200.0,
      "step": 791
    },
    {
      "entropy": 0.530088871717453,
      "epoch": 2.955223880597015,
      "grad_norm": 0.17820289731025696,
      "learning_rate": 0.0002,
      "loss": 0.5355557203292847,
      "mean_token_accuracy": 0.7830621153116226,
      "num_tokens": 12963577.0,
      "step": 792
    },
    {
      "entropy": 0.5236834436655045,
      "epoch": 2.958955223880597,
      "grad_norm": 0.2109825313091278,
      "learning_rate": 0.0002,
      "loss": 0.5322213172912598,
      "mean_token_accuracy": 0.7841589003801346,
      "num_tokens": 12979785.0,
      "step": 793
    },
    {
      "entropy": 0.5233126282691956,
      "epoch": 2.9626865671641793,
      "grad_norm": 0.18967954814434052,
      "learning_rate": 0.0002,
      "loss": 0.5212454199790955,
      "mean_token_accuracy": 0.7889588326215744,
      "num_tokens": 12995708.0,
      "step": 794
    },
    {
      "entropy": 0.5299918279051781,
      "epoch": 2.966417910447761,
      "grad_norm": 0.16920922696590424,
      "learning_rate": 0.0002,
      "loss": 0.5193845629692078,
      "mean_token_accuracy": 0.7896459102630615,
      "num_tokens": 13012134.0,
      "step": 795
    },
    {
      "entropy": 0.5286006182432175,
      "epoch": 2.970149253731343,
      "grad_norm": 0.20278210937976837,
      "learning_rate": 0.0002,
      "loss": 0.5250504016876221,
      "mean_token_accuracy": 0.7887490838766098,
      "num_tokens": 13028379.0,
      "step": 796
    },
    {
      "entropy": 0.5279938280582428,
      "epoch": 2.9738805970149254,
      "grad_norm": 0.18265850841999054,
      "learning_rate": 0.0002,
      "loss": 0.5340323448181152,
      "mean_token_accuracy": 0.7845529317855835,
      "num_tokens": 13044924.0,
      "step": 797
    },
    {
      "entropy": 0.5393821448087692,
      "epoch": 2.9776119402985075,
      "grad_norm": 0.22119778394699097,
      "learning_rate": 0.0002,
      "loss": 0.5526689291000366,
      "mean_token_accuracy": 0.7805535644292831,
      "num_tokens": 13061419.0,
      "step": 798
    },
    {
      "entropy": 0.5472811162471771,
      "epoch": 2.9813432835820897,
      "grad_norm": 0.168561190366745,
      "learning_rate": 0.0002,
      "loss": 0.5470550060272217,
      "mean_token_accuracy": 0.7784483730792999,
      "num_tokens": 13077766.0,
      "step": 799
    },
    {
      "entropy": 0.5338646024465561,
      "epoch": 2.9850746268656714,
      "grad_norm": 0.23440560698509216,
      "learning_rate": 0.0002,
      "loss": 0.5280591249465942,
      "mean_token_accuracy": 0.7905636876821518,
      "num_tokens": 13094062.0,
      "step": 800
    },
    {
      "entropy": 0.5247876346111298,
      "epoch": 2.9888059701492535,
      "grad_norm": 0.15499182045459747,
      "learning_rate": 0.0002,
      "loss": 0.5195421576499939,
      "mean_token_accuracy": 0.7871391475200653,
      "num_tokens": 13110406.0,
      "step": 801
    },
    {
      "entropy": 0.5327739417552948,
      "epoch": 2.9925373134328357,
      "grad_norm": 0.1991773396730423,
      "learning_rate": 0.0002,
      "loss": 0.5344197750091553,
      "mean_token_accuracy": 0.7860752046108246,
      "num_tokens": 13126593.0,
      "step": 802
    },
    {
      "entropy": 0.5207744166254997,
      "epoch": 2.996268656716418,
      "grad_norm": 0.16794730722904205,
      "learning_rate": 0.0002,
      "loss": 0.5182280540466309,
      "mean_token_accuracy": 0.7904401868581772,
      "num_tokens": 13142897.0,
      "step": 803
    },
    {
      "entropy": 0.517033725976944,
      "epoch": 3.0,
      "grad_norm": 0.16952210664749146,
      "learning_rate": 0.0002,
      "loss": 0.5222631096839905,
      "mean_token_accuracy": 0.7886133790016174,
      "num_tokens": 13159031.0,
      "step": 804
    }
  ],
  "logging_steps": 1,
  "max_steps": 804,
  "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": 1.2260244758336635e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}