{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 142,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.4938980638980865,
      "epoch": 0.014084507042253521,
      "grad_norm": 89.5,
      "learning_rate": 5e-05,
      "loss": 4.519379615783691,
      "mean_token_accuracy": 0.3811451345682144,
      "num_tokens": 2953.0,
      "step": 1
    },
    {
      "entropy": 3.6362282037734985,
      "epoch": 0.028169014084507043,
      "grad_norm": 29.125,
      "learning_rate": 5e-05,
      "loss": 3.7604172229766846,
      "mean_token_accuracy": 0.39103876799345016,
      "num_tokens": 5196.0,
      "step": 2
    },
    {
      "entropy": 3.796808362007141,
      "epoch": 0.04225352112676056,
      "grad_norm": 18.75,
      "learning_rate": 5e-05,
      "loss": 3.3663859367370605,
      "mean_token_accuracy": 0.42154909670352936,
      "num_tokens": 7537.0,
      "step": 3
    },
    {
      "entropy": 3.071028172969818,
      "epoch": 0.056338028169014086,
      "grad_norm": 14.5625,
      "learning_rate": 5e-05,
      "loss": 3.1916425228118896,
      "mean_token_accuracy": 0.4548730403184891,
      "num_tokens": 9584.0,
      "step": 4
    },
    {
      "entropy": 2.7734431624412537,
      "epoch": 0.07042253521126761,
      "grad_norm": 15.0,
      "learning_rate": 5e-05,
      "loss": 2.661928653717041,
      "mean_token_accuracy": 0.5473615825176239,
      "num_tokens": 11344.0,
      "step": 5
    },
    {
      "entropy": 2.4827571511268616,
      "epoch": 0.08450704225352113,
      "grad_norm": 13.6875,
      "learning_rate": 5e-05,
      "loss": 2.691732168197632,
      "mean_token_accuracy": 0.5095682442188263,
      "num_tokens": 13289.0,
      "step": 6
    },
    {
      "entropy": 2.6064285039901733,
      "epoch": 0.09859154929577464,
      "grad_norm": 12.625,
      "learning_rate": 5e-05,
      "loss": 2.7455577850341797,
      "mean_token_accuracy": 0.5099757686257362,
      "num_tokens": 15664.0,
      "step": 7
    },
    {
      "entropy": 2.4781495332717896,
      "epoch": 0.11267605633802817,
      "grad_norm": 12.75,
      "learning_rate": 5e-05,
      "loss": 2.5589702129364014,
      "mean_token_accuracy": 0.5618840456008911,
      "num_tokens": 17331.0,
      "step": 8
    },
    {
      "entropy": 2.512305438518524,
      "epoch": 0.1267605633802817,
      "grad_norm": 10.1875,
      "learning_rate": 5e-05,
      "loss": 2.5893146991729736,
      "mean_token_accuracy": 0.5188388600945473,
      "num_tokens": 19724.0,
      "step": 9
    },
    {
      "entropy": 2.187403589487076,
      "epoch": 0.14084507042253522,
      "grad_norm": 11.8125,
      "learning_rate": 5e-05,
      "loss": 2.4727907180786133,
      "mean_token_accuracy": 0.6085146963596344,
      "num_tokens": 21334.0,
      "step": 10
    },
    {
      "entropy": 2.8956857919692993,
      "epoch": 0.15492957746478872,
      "grad_norm": 10.1875,
      "learning_rate": 5e-05,
      "loss": 2.8553805351257324,
      "mean_token_accuracy": 0.4754949361085892,
      "num_tokens": 23770.0,
      "step": 11
    },
    {
      "entropy": 2.4213987588882446,
      "epoch": 0.16901408450704225,
      "grad_norm": 9.75,
      "learning_rate": 5e-05,
      "loss": 2.4139676094055176,
      "mean_token_accuracy": 0.5475910231471062,
      "num_tokens": 26233.0,
      "step": 12
    },
    {
      "entropy": 2.668797016143799,
      "epoch": 0.18309859154929578,
      "grad_norm": 12.25,
      "learning_rate": 5e-05,
      "loss": 2.769174337387085,
      "mean_token_accuracy": 0.5351328551769257,
      "num_tokens": 27692.0,
      "step": 13
    },
    {
      "entropy": 2.544685661792755,
      "epoch": 0.19718309859154928,
      "grad_norm": 9.5625,
      "learning_rate": 5e-05,
      "loss": 2.4870781898498535,
      "mean_token_accuracy": 0.514883853495121,
      "num_tokens": 30217.0,
      "step": 14
    },
    {
      "entropy": 2.5978153944015503,
      "epoch": 0.2112676056338028,
      "grad_norm": 10.875,
      "learning_rate": 5e-05,
      "loss": 2.639450788497925,
      "mean_token_accuracy": 0.5468774288892746,
      "num_tokens": 31969.0,
      "step": 15
    },
    {
      "entropy": 2.452792704105377,
      "epoch": 0.22535211267605634,
      "grad_norm": 11.5,
      "learning_rate": 5e-05,
      "loss": 2.5285415649414062,
      "mean_token_accuracy": 0.5533104464411736,
      "num_tokens": 34160.0,
      "step": 16
    },
    {
      "entropy": 2.234907388687134,
      "epoch": 0.23943661971830985,
      "grad_norm": 12.0625,
      "learning_rate": 5e-05,
      "loss": 2.2553577423095703,
      "mean_token_accuracy": 0.5958019495010376,
      "num_tokens": 35899.0,
      "step": 17
    },
    {
      "entropy": 2.2781192660331726,
      "epoch": 0.2535211267605634,
      "grad_norm": 10.8125,
      "learning_rate": 5e-05,
      "loss": 2.4401230812072754,
      "mean_token_accuracy": 0.5632342845201492,
      "num_tokens": 37693.0,
      "step": 18
    },
    {
      "entropy": 2.292062520980835,
      "epoch": 0.2676056338028169,
      "grad_norm": 10.5,
      "learning_rate": 5e-05,
      "loss": 2.53348445892334,
      "mean_token_accuracy": 0.5441952645778656,
      "num_tokens": 39793.0,
      "step": 19
    },
    {
      "entropy": 2.5250023007392883,
      "epoch": 0.28169014084507044,
      "grad_norm": 13.5,
      "learning_rate": 5e-05,
      "loss": 2.7173304557800293,
      "mean_token_accuracy": 0.4843936040997505,
      "num_tokens": 42035.0,
      "step": 20
    },
    {
      "entropy": 2.3936269879341125,
      "epoch": 0.29577464788732394,
      "grad_norm": 10.5625,
      "learning_rate": 5e-05,
      "loss": 2.397752046585083,
      "mean_token_accuracy": 0.5632070526480675,
      "num_tokens": 43606.0,
      "step": 21
    },
    {
      "entropy": 2.7215799689292908,
      "epoch": 0.30985915492957744,
      "grad_norm": 9.375,
      "learning_rate": 5e-05,
      "loss": 2.7266643047332764,
      "mean_token_accuracy": 0.4814189076423645,
      "num_tokens": 46160.0,
      "step": 22
    },
    {
      "entropy": 2.4695653319358826,
      "epoch": 0.323943661971831,
      "grad_norm": 8.4375,
      "learning_rate": 5e-05,
      "loss": 2.3790149688720703,
      "mean_token_accuracy": 0.5413414835929871,
      "num_tokens": 48749.0,
      "step": 23
    },
    {
      "entropy": 2.3166028261184692,
      "epoch": 0.3380281690140845,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 2.274803638458252,
      "mean_token_accuracy": 0.5993590205907822,
      "num_tokens": 50770.0,
      "step": 24
    },
    {
      "entropy": 2.285377472639084,
      "epoch": 0.352112676056338,
      "grad_norm": 8.75,
      "learning_rate": 5e-05,
      "loss": 2.2308878898620605,
      "mean_token_accuracy": 0.5828141123056412,
      "num_tokens": 53004.0,
      "step": 25
    },
    {
      "entropy": 2.1157060265541077,
      "epoch": 0.36619718309859156,
      "grad_norm": 8.375,
      "learning_rate": 5e-05,
      "loss": 2.1033847332000732,
      "mean_token_accuracy": 0.5867631435394287,
      "num_tokens": 55108.0,
      "step": 26
    },
    {
      "entropy": 2.5209829807281494,
      "epoch": 0.38028169014084506,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 2.3607499599456787,
      "mean_token_accuracy": 0.5426046550273895,
      "num_tokens": 57080.0,
      "step": 27
    },
    {
      "entropy": 1.974183738231659,
      "epoch": 0.39436619718309857,
      "grad_norm": 7.4375,
      "learning_rate": 5e-05,
      "loss": 1.9418365955352783,
      "mean_token_accuracy": 0.6255736351013184,
      "num_tokens": 59237.0,
      "step": 28
    },
    {
      "entropy": 2.252374231815338,
      "epoch": 0.4084507042253521,
      "grad_norm": 9.875,
      "learning_rate": 5e-05,
      "loss": 2.228358507156372,
      "mean_token_accuracy": 0.5917280316352844,
      "num_tokens": 60850.0,
      "step": 29
    },
    {
      "entropy": 2.327530801296234,
      "epoch": 0.4225352112676056,
      "grad_norm": 9.5625,
      "learning_rate": 5e-05,
      "loss": 2.3329389095306396,
      "mean_token_accuracy": 0.5718153715133667,
      "num_tokens": 62734.0,
      "step": 30
    },
    {
      "entropy": 2.2620502710342407,
      "epoch": 0.43661971830985913,
      "grad_norm": 9.375,
      "learning_rate": 5e-05,
      "loss": 2.1953892707824707,
      "mean_token_accuracy": 0.5675001591444016,
      "num_tokens": 64637.0,
      "step": 31
    },
    {
      "entropy": 2.445267140865326,
      "epoch": 0.4507042253521127,
      "grad_norm": 9.0625,
      "learning_rate": 5e-05,
      "loss": 2.501152276992798,
      "mean_token_accuracy": 0.5355894565582275,
      "num_tokens": 66880.0,
      "step": 32
    },
    {
      "entropy": 2.3873718976974487,
      "epoch": 0.4647887323943662,
      "grad_norm": 9.4375,
      "learning_rate": 5e-05,
      "loss": 2.3468880653381348,
      "mean_token_accuracy": 0.5372899621725082,
      "num_tokens": 68623.0,
      "step": 33
    },
    {
      "entropy": 2.0545941591262817,
      "epoch": 0.4788732394366197,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 2.1947593688964844,
      "mean_token_accuracy": 0.6112199202179909,
      "num_tokens": 70769.0,
      "step": 34
    },
    {
      "entropy": 2.350032150745392,
      "epoch": 0.49295774647887325,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 2.418266773223877,
      "mean_token_accuracy": 0.5104246437549591,
      "num_tokens": 73653.0,
      "step": 35
    },
    {
      "entropy": 2.237062931060791,
      "epoch": 0.5070422535211268,
      "grad_norm": 10.75,
      "learning_rate": 5e-05,
      "loss": 2.4755356311798096,
      "mean_token_accuracy": 0.5747656747698784,
      "num_tokens": 75524.0,
      "step": 36
    },
    {
      "entropy": 2.0994813442230225,
      "epoch": 0.5211267605633803,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 2.044315814971924,
      "mean_token_accuracy": 0.6056438237428665,
      "num_tokens": 77246.0,
      "step": 37
    },
    {
      "entropy": 2.5556156635284424,
      "epoch": 0.5352112676056338,
      "grad_norm": 8.25,
      "learning_rate": 5e-05,
      "loss": 2.446904182434082,
      "mean_token_accuracy": 0.5177429020404816,
      "num_tokens": 79970.0,
      "step": 38
    },
    {
      "entropy": 2.337540030479431,
      "epoch": 0.5492957746478874,
      "grad_norm": 8.5625,
      "learning_rate": 5e-05,
      "loss": 2.216331720352173,
      "mean_token_accuracy": 0.58082115650177,
      "num_tokens": 81876.0,
      "step": 39
    },
    {
      "entropy": 2.043498247861862,
      "epoch": 0.5633802816901409,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 1.9643354415893555,
      "mean_token_accuracy": 0.607293888926506,
      "num_tokens": 84298.0,
      "step": 40
    },
    {
      "entropy": 2.4852016270160675,
      "epoch": 0.5774647887323944,
      "grad_norm": 8.75,
      "learning_rate": 5e-05,
      "loss": 2.571437358856201,
      "mean_token_accuracy": 0.5233406946063042,
      "num_tokens": 86291.0,
      "step": 41
    },
    {
      "entropy": 2.5456899404525757,
      "epoch": 0.5915492957746479,
      "grad_norm": 9.125,
      "learning_rate": 5e-05,
      "loss": 2.4434070587158203,
      "mean_token_accuracy": 0.5472028106451035,
      "num_tokens": 88194.0,
      "step": 42
    },
    {
      "entropy": 2.2191686630249023,
      "epoch": 0.6056338028169014,
      "grad_norm": 9.0625,
      "learning_rate": 5e-05,
      "loss": 2.225029468536377,
      "mean_token_accuracy": 0.6006355881690979,
      "num_tokens": 89922.0,
      "step": 43
    },
    {
      "entropy": 2.2295450568199158,
      "epoch": 0.6197183098591549,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 2.4000232219696045,
      "mean_token_accuracy": 0.580116793513298,
      "num_tokens": 92171.0,
      "step": 44
    },
    {
      "entropy": 2.2142042219638824,
      "epoch": 0.6338028169014085,
      "grad_norm": 8.3125,
      "learning_rate": 5e-05,
      "loss": 2.2619423866271973,
      "mean_token_accuracy": 0.5818447023630142,
      "num_tokens": 94362.0,
      "step": 45
    },
    {
      "entropy": 2.3695861101150513,
      "epoch": 0.647887323943662,
      "grad_norm": 7.21875,
      "learning_rate": 5e-05,
      "loss": 2.4540324211120605,
      "mean_token_accuracy": 0.5298221409320831,
      "num_tokens": 96559.0,
      "step": 46
    },
    {
      "entropy": 1.9680090248584747,
      "epoch": 0.6619718309859155,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 1.9608612060546875,
      "mean_token_accuracy": 0.6151116490364075,
      "num_tokens": 99018.0,
      "step": 47
    },
    {
      "entropy": 2.449136793613434,
      "epoch": 0.676056338028169,
      "grad_norm": 9.375,
      "learning_rate": 5e-05,
      "loss": 2.364119529724121,
      "mean_token_accuracy": 0.5657317340373993,
      "num_tokens": 100885.0,
      "step": 48
    },
    {
      "entropy": 2.109535813331604,
      "epoch": 0.6901408450704225,
      "grad_norm": 6.875,
      "learning_rate": 5e-05,
      "loss": 2.1192779541015625,
      "mean_token_accuracy": 0.5970411896705627,
      "num_tokens": 103357.0,
      "step": 49
    },
    {
      "entropy": 2.3080650568008423,
      "epoch": 0.704225352112676,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 2.351897716522217,
      "mean_token_accuracy": 0.5307685062289238,
      "num_tokens": 105870.0,
      "step": 50
    },
    {
      "entropy": 2.1463415920734406,
      "epoch": 0.7183098591549296,
      "grad_norm": 7.5,
      "learning_rate": 5e-05,
      "loss": 2.2693376541137695,
      "mean_token_accuracy": 0.5934874191880226,
      "num_tokens": 107995.0,
      "step": 51
    },
    {
      "entropy": 2.1181687712669373,
      "epoch": 0.7323943661971831,
      "grad_norm": 8.4375,
      "learning_rate": 5e-05,
      "loss": 2.1629841327667236,
      "mean_token_accuracy": 0.6063478291034698,
      "num_tokens": 110358.0,
      "step": 52
    },
    {
      "entropy": 2.0169546008110046,
      "epoch": 0.7464788732394366,
      "grad_norm": 7.78125,
      "learning_rate": 5e-05,
      "loss": 1.817868709564209,
      "mean_token_accuracy": 0.6273332387208939,
      "num_tokens": 112267.0,
      "step": 53
    },
    {
      "entropy": 2.189443975687027,
      "epoch": 0.7605633802816901,
      "grad_norm": 9.8125,
      "learning_rate": 5e-05,
      "loss": 2.3292086124420166,
      "mean_token_accuracy": 0.6041209101676941,
      "num_tokens": 113899.0,
      "step": 54
    },
    {
      "entropy": 2.1946000456809998,
      "epoch": 0.7746478873239436,
      "grad_norm": 9.3125,
      "learning_rate": 5e-05,
      "loss": 2.2774949073791504,
      "mean_token_accuracy": 0.5620942413806915,
      "num_tokens": 116015.0,
      "step": 55
    },
    {
      "entropy": 2.1172711849212646,
      "epoch": 0.7887323943661971,
      "grad_norm": 7.90625,
      "learning_rate": 5e-05,
      "loss": 2.1236391067504883,
      "mean_token_accuracy": 0.5911316424608231,
      "num_tokens": 118044.0,
      "step": 56
    },
    {
      "entropy": 2.144031971693039,
      "epoch": 0.8028169014084507,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 2.187140703201294,
      "mean_token_accuracy": 0.5624879449605942,
      "num_tokens": 120180.0,
      "step": 57
    },
    {
      "entropy": 2.181178390979767,
      "epoch": 0.8169014084507042,
      "grad_norm": 10.3125,
      "learning_rate": 5e-05,
      "loss": 2.113279104232788,
      "mean_token_accuracy": 0.6004882454872131,
      "num_tokens": 121676.0,
      "step": 58
    },
    {
      "entropy": 2.314671218395233,
      "epoch": 0.8309859154929577,
      "grad_norm": 9.5625,
      "learning_rate": 5e-05,
      "loss": 2.331024169921875,
      "mean_token_accuracy": 0.5824068263173103,
      "num_tokens": 123161.0,
      "step": 59
    },
    {
      "entropy": 2.0048855245113373,
      "epoch": 0.8450704225352113,
      "grad_norm": 9.375,
      "learning_rate": 5e-05,
      "loss": 2.070265054702759,
      "mean_token_accuracy": 0.6141902804374695,
      "num_tokens": 125016.0,
      "step": 60
    },
    {
      "entropy": 1.966452568769455,
      "epoch": 0.8591549295774648,
      "grad_norm": 7.96875,
      "learning_rate": 5e-05,
      "loss": 1.9488521814346313,
      "mean_token_accuracy": 0.5854089111089706,
      "num_tokens": 127349.0,
      "step": 61
    },
    {
      "entropy": 2.1519170999526978,
      "epoch": 0.8732394366197183,
      "grad_norm": 7.5625,
      "learning_rate": 5e-05,
      "loss": 2.1743807792663574,
      "mean_token_accuracy": 0.5636171847581863,
      "num_tokens": 129742.0,
      "step": 62
    },
    {
      "entropy": 2.7546687722206116,
      "epoch": 0.8873239436619719,
      "grad_norm": 7.96875,
      "learning_rate": 5e-05,
      "loss": 2.7343056201934814,
      "mean_token_accuracy": 0.496803916990757,
      "num_tokens": 131743.0,
      "step": 63
    },
    {
      "entropy": 2.421773910522461,
      "epoch": 0.9014084507042254,
      "grad_norm": 7.96875,
      "learning_rate": 5e-05,
      "loss": 2.6149866580963135,
      "mean_token_accuracy": 0.5249136164784431,
      "num_tokens": 134002.0,
      "step": 64
    },
    {
      "entropy": 2.0106727480888367,
      "epoch": 0.9154929577464789,
      "grad_norm": 8.4375,
      "learning_rate": 5e-05,
      "loss": 1.940914273262024,
      "mean_token_accuracy": 0.5855439230799675,
      "num_tokens": 136120.0,
      "step": 65
    },
    {
      "entropy": 2.1143648624420166,
      "epoch": 0.9295774647887324,
      "grad_norm": 8.375,
      "learning_rate": 5e-05,
      "loss": 2.2469482421875,
      "mean_token_accuracy": 0.6145937740802765,
      "num_tokens": 137997.0,
      "step": 66
    },
    {
      "entropy": 2.1894670128822327,
      "epoch": 0.9436619718309859,
      "grad_norm": 8.25,
      "learning_rate": 5e-05,
      "loss": 2.20756459236145,
      "mean_token_accuracy": 0.5872255340218544,
      "num_tokens": 139904.0,
      "step": 67
    },
    {
      "entropy": 1.75501748919487,
      "epoch": 0.9577464788732394,
      "grad_norm": 9.3125,
      "learning_rate": 5e-05,
      "loss": 1.9068934917449951,
      "mean_token_accuracy": 0.6642919778823853,
      "num_tokens": 141166.0,
      "step": 68
    },
    {
      "entropy": 2.204477220773697,
      "epoch": 0.971830985915493,
      "grad_norm": 7.875,
      "learning_rate": 5e-05,
      "loss": 2.286216974258423,
      "mean_token_accuracy": 0.5354761853814125,
      "num_tokens": 143738.0,
      "step": 69
    },
    {
      "entropy": 2.1219914853572845,
      "epoch": 0.9859154929577465,
      "grad_norm": 8.625,
      "learning_rate": 5e-05,
      "loss": 2.274585723876953,
      "mean_token_accuracy": 0.609759122133255,
      "num_tokens": 145439.0,
      "step": 70
    },
    {
      "entropy": 1.9260511100292206,
      "epoch": 1.0,
      "grad_norm": 8.375,
      "learning_rate": 5e-05,
      "loss": 1.8763841390609741,
      "mean_token_accuracy": 0.5979989916086197,
      "num_tokens": 147650.0,
      "step": 71
    },
    {
      "epoch": 1.0,
      "eval_entropy": 2.1684498711371085,
      "eval_loss": 2.284580707550049,
      "eval_mean_token_accuracy": 0.5757196218195096,
      "eval_num_tokens": 147650.0,
      "eval_runtime": 18.8201,
      "eval_samples_per_second": 15.09,
      "eval_steps_per_second": 3.773,
      "step": 71
    },
    {
      "entropy": 2.24437940120697,
      "epoch": 1.0140845070422535,
      "grad_norm": 8.125,
      "learning_rate": 5e-05,
      "loss": 2.148402214050293,
      "mean_token_accuracy": 0.5866604596376419,
      "num_tokens": 149372.0,
      "step": 72
    },
    {
      "entropy": 1.630838692188263,
      "epoch": 1.028169014084507,
      "grad_norm": 7.0,
      "learning_rate": 5e-05,
      "loss": 1.526665449142456,
      "mean_token_accuracy": 0.6816482096910477,
      "num_tokens": 151274.0,
      "step": 73
    },
    {
      "entropy": 2.193234443664551,
      "epoch": 1.0422535211267605,
      "grad_norm": 7.90625,
      "learning_rate": 5e-05,
      "loss": 2.176100969314575,
      "mean_token_accuracy": 0.6048732623457909,
      "num_tokens": 153103.0,
      "step": 74
    },
    {
      "entropy": 1.7758266925811768,
      "epoch": 1.056338028169014,
      "grad_norm": 7.53125,
      "learning_rate": 5e-05,
      "loss": 1.7472716569900513,
      "mean_token_accuracy": 0.6403823792934418,
      "num_tokens": 155291.0,
      "step": 75
    },
    {
      "entropy": 1.8887377083301544,
      "epoch": 1.0704225352112675,
      "grad_norm": 8.125,
      "learning_rate": 5e-05,
      "loss": 1.872934341430664,
      "mean_token_accuracy": 0.6085800677537918,
      "num_tokens": 157847.0,
      "step": 76
    },
    {
      "entropy": 1.6055135428905487,
      "epoch": 1.084507042253521,
      "grad_norm": 7.53125,
      "learning_rate": 5e-05,
      "loss": 1.6859486103057861,
      "mean_token_accuracy": 0.6478956639766693,
      "num_tokens": 160538.0,
      "step": 77
    },
    {
      "entropy": 1.9197329580783844,
      "epoch": 1.0985915492957747,
      "grad_norm": 7.375,
      "learning_rate": 5e-05,
      "loss": 1.929700493812561,
      "mean_token_accuracy": 0.6054274141788483,
      "num_tokens": 162709.0,
      "step": 78
    },
    {
      "entropy": 1.6682755053043365,
      "epoch": 1.1126760563380282,
      "grad_norm": 8.125,
      "learning_rate": 5e-05,
      "loss": 1.899672269821167,
      "mean_token_accuracy": 0.6613650918006897,
      "num_tokens": 164478.0,
      "step": 79
    },
    {
      "entropy": 1.8098062872886658,
      "epoch": 1.1267605633802817,
      "grad_norm": 9.625,
      "learning_rate": 5e-05,
      "loss": 1.7989305257797241,
      "mean_token_accuracy": 0.6530113369226456,
      "num_tokens": 165767.0,
      "step": 80
    },
    {
      "entropy": 2.3263643383979797,
      "epoch": 1.1408450704225352,
      "grad_norm": 8.1875,
      "learning_rate": 5e-05,
      "loss": 2.32468843460083,
      "mean_token_accuracy": 0.5657684504985809,
      "num_tokens": 167374.0,
      "step": 81
    },
    {
      "entropy": 1.6652490496635437,
      "epoch": 1.1549295774647887,
      "grad_norm": 8.625,
      "learning_rate": 5e-05,
      "loss": 1.821378469467163,
      "mean_token_accuracy": 0.6479186862707138,
      "num_tokens": 169216.0,
      "step": 82
    },
    {
      "entropy": 2.157005727291107,
      "epoch": 1.1690140845070423,
      "grad_norm": 8.1875,
      "learning_rate": 5e-05,
      "loss": 2.1137959957122803,
      "mean_token_accuracy": 0.5775181204080582,
      "num_tokens": 171194.0,
      "step": 83
    },
    {
      "entropy": 1.6896095722913742,
      "epoch": 1.1830985915492958,
      "grad_norm": 8.5625,
      "learning_rate": 5e-05,
      "loss": 1.785524606704712,
      "mean_token_accuracy": 0.6791908890008926,
      "num_tokens": 173018.0,
      "step": 84
    },
    {
      "entropy": 2.241276800632477,
      "epoch": 1.1971830985915493,
      "grad_norm": 8.9375,
      "learning_rate": 5e-05,
      "loss": 2.1982414722442627,
      "mean_token_accuracy": 0.5830243527889252,
      "num_tokens": 174822.0,
      "step": 85
    },
    {
      "entropy": 1.8407301902770996,
      "epoch": 1.2112676056338028,
      "grad_norm": 7.65625,
      "learning_rate": 5e-05,
      "loss": 1.752529501914978,
      "mean_token_accuracy": 0.6172220706939697,
      "num_tokens": 176932.0,
      "step": 86
    },
    {
      "entropy": 1.707479178905487,
      "epoch": 1.2253521126760563,
      "grad_norm": 7.96875,
      "learning_rate": 5e-05,
      "loss": 1.6565136909484863,
      "mean_token_accuracy": 0.6619749516248703,
      "num_tokens": 178804.0,
      "step": 87
    },
    {
      "entropy": 1.6418649852275848,
      "epoch": 1.2394366197183098,
      "grad_norm": 7.15625,
      "learning_rate": 5e-05,
      "loss": 1.5298699140548706,
      "mean_token_accuracy": 0.6645439267158508,
      "num_tokens": 181353.0,
      "step": 88
    },
    {
      "entropy": 1.6070191860198975,
      "epoch": 1.2535211267605635,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 1.7814466953277588,
      "mean_token_accuracy": 0.6640422344207764,
      "num_tokens": 183278.0,
      "step": 89
    },
    {
      "entropy": 1.8768837749958038,
      "epoch": 1.267605633802817,
      "grad_norm": 9.0625,
      "learning_rate": 5e-05,
      "loss": 1.9979424476623535,
      "mean_token_accuracy": 0.6454368084669113,
      "num_tokens": 185041.0,
      "step": 90
    },
    {
      "entropy": 1.809422641992569,
      "epoch": 1.2816901408450705,
      "grad_norm": 7.78125,
      "learning_rate": 5e-05,
      "loss": 1.8236162662506104,
      "mean_token_accuracy": 0.6187984794378281,
      "num_tokens": 187646.0,
      "step": 91
    },
    {
      "entropy": 1.8032868206501007,
      "epoch": 1.295774647887324,
      "grad_norm": 7.4375,
      "learning_rate": 5e-05,
      "loss": 1.8008391857147217,
      "mean_token_accuracy": 0.6168265342712402,
      "num_tokens": 189952.0,
      "step": 92
    },
    {
      "entropy": 2.107921302318573,
      "epoch": 1.3098591549295775,
      "grad_norm": 7.71875,
      "learning_rate": 5e-05,
      "loss": 2.077434539794922,
      "mean_token_accuracy": 0.5662726014852524,
      "num_tokens": 192592.0,
      "step": 93
    },
    {
      "entropy": 1.6842139065265656,
      "epoch": 1.323943661971831,
      "grad_norm": 7.5625,
      "learning_rate": 5e-05,
      "loss": 1.6454025506973267,
      "mean_token_accuracy": 0.6700995117425919,
      "num_tokens": 194549.0,
      "step": 94
    },
    {
      "entropy": 1.859749436378479,
      "epoch": 1.3380281690140845,
      "grad_norm": 9.0625,
      "learning_rate": 5e-05,
      "loss": 1.9091846942901611,
      "mean_token_accuracy": 0.6098746657371521,
      "num_tokens": 196448.0,
      "step": 95
    },
    {
      "entropy": 2.018050581216812,
      "epoch": 1.352112676056338,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 2.1190457344055176,
      "mean_token_accuracy": 0.5959457382559776,
      "num_tokens": 198615.0,
      "step": 96
    },
    {
      "entropy": 1.9890158474445343,
      "epoch": 1.3661971830985915,
      "grad_norm": 8.125,
      "learning_rate": 5e-05,
      "loss": 1.960221529006958,
      "mean_token_accuracy": 0.5875463634729385,
      "num_tokens": 201013.0,
      "step": 97
    },
    {
      "entropy": 1.8716930747032166,
      "epoch": 1.380281690140845,
      "grad_norm": 8.5,
      "learning_rate": 5e-05,
      "loss": 2.0004959106445312,
      "mean_token_accuracy": 0.6175930351018906,
      "num_tokens": 202969.0,
      "step": 98
    },
    {
      "entropy": 1.7610785365104675,
      "epoch": 1.3943661971830985,
      "grad_norm": 8.1875,
      "learning_rate": 5e-05,
      "loss": 1.6850508451461792,
      "mean_token_accuracy": 0.6504905521869659,
      "num_tokens": 204906.0,
      "step": 99
    },
    {
      "entropy": 1.9090750515460968,
      "epoch": 1.408450704225352,
      "grad_norm": 8.875,
      "learning_rate": 5e-05,
      "loss": 1.938533067703247,
      "mean_token_accuracy": 0.621729701757431,
      "num_tokens": 206869.0,
      "step": 100
    },
    {
      "entropy": 1.8917019963264465,
      "epoch": 1.4225352112676055,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 2.0368175506591797,
      "mean_token_accuracy": 0.6267035603523254,
      "num_tokens": 208432.0,
      "step": 101
    },
    {
      "entropy": 1.7962620854377747,
      "epoch": 1.436619718309859,
      "grad_norm": 8.375,
      "learning_rate": 5e-05,
      "loss": 1.8455748558044434,
      "mean_token_accuracy": 0.608839362859726,
      "num_tokens": 210949.0,
      "step": 102
    },
    {
      "entropy": 1.724296659231186,
      "epoch": 1.4507042253521127,
      "grad_norm": 10.25,
      "learning_rate": 5e-05,
      "loss": 1.7760952711105347,
      "mean_token_accuracy": 0.6573535352945328,
      "num_tokens": 212595.0,
      "step": 103
    },
    {
      "entropy": 1.8014512360095978,
      "epoch": 1.4647887323943662,
      "grad_norm": 7.6875,
      "learning_rate": 5e-05,
      "loss": 1.8614972829818726,
      "mean_token_accuracy": 0.5946371257305145,
      "num_tokens": 215384.0,
      "step": 104
    },
    {
      "entropy": 1.8427681922912598,
      "epoch": 1.4788732394366197,
      "grad_norm": 8.75,
      "learning_rate": 5e-05,
      "loss": 1.880817174911499,
      "mean_token_accuracy": 0.6408448368310928,
      "num_tokens": 217561.0,
      "step": 105
    },
    {
      "entropy": 1.6576282978057861,
      "epoch": 1.4929577464788732,
      "grad_norm": 9.9375,
      "learning_rate": 5e-05,
      "loss": 1.6814919710159302,
      "mean_token_accuracy": 0.6402300000190735,
      "num_tokens": 219087.0,
      "step": 106
    },
    {
      "entropy": 1.5478721261024475,
      "epoch": 1.5070422535211268,
      "grad_norm": 7.59375,
      "learning_rate": 5e-05,
      "loss": 1.5617148876190186,
      "mean_token_accuracy": 0.6625793874263763,
      "num_tokens": 221887.0,
      "step": 107
    },
    {
      "entropy": 1.923494279384613,
      "epoch": 1.5211267605633803,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 2.0690817832946777,
      "mean_token_accuracy": 0.5978109538555145,
      "num_tokens": 224182.0,
      "step": 108
    },
    {
      "entropy": 1.8152073919773102,
      "epoch": 1.5352112676056338,
      "grad_norm": 7.71875,
      "learning_rate": 5e-05,
      "loss": 1.7398602962493896,
      "mean_token_accuracy": 0.6207661181688309,
      "num_tokens": 226591.0,
      "step": 109
    },
    {
      "entropy": 2.1172706484794617,
      "epoch": 1.5492957746478875,
      "grad_norm": 9.9375,
      "learning_rate": 5e-05,
      "loss": 2.237187385559082,
      "mean_token_accuracy": 0.5684898793697357,
      "num_tokens": 228529.0,
      "step": 110
    },
    {
      "entropy": 1.686636283993721,
      "epoch": 1.563380281690141,
      "grad_norm": 9.125,
      "learning_rate": 5e-05,
      "loss": 1.8167091608047485,
      "mean_token_accuracy": 0.6692909449338913,
      "num_tokens": 229973.0,
      "step": 111
    },
    {
      "entropy": 2.285969853401184,
      "epoch": 1.5774647887323945,
      "grad_norm": 8.6875,
      "learning_rate": 5e-05,
      "loss": 2.308091878890991,
      "mean_token_accuracy": 0.5341236144304276,
      "num_tokens": 232232.0,
      "step": 112
    },
    {
      "entropy": 1.7772781550884247,
      "epoch": 1.591549295774648,
      "grad_norm": 7.09375,
      "learning_rate": 5e-05,
      "loss": 1.7048449516296387,
      "mean_token_accuracy": 0.6309333443641663,
      "num_tokens": 234998.0,
      "step": 113
    },
    {
      "entropy": 1.8632723987102509,
      "epoch": 1.6056338028169015,
      "grad_norm": 6.96875,
      "learning_rate": 5e-05,
      "loss": 1.8152518272399902,
      "mean_token_accuracy": 0.5958341658115387,
      "num_tokens": 237842.0,
      "step": 114
    },
    {
      "entropy": 1.7296560108661652,
      "epoch": 1.619718309859155,
      "grad_norm": 9.125,
      "learning_rate": 5e-05,
      "loss": 1.7719354629516602,
      "mean_token_accuracy": 0.6910708993673325,
      "num_tokens": 239068.0,
      "step": 115
    },
    {
      "entropy": 2.026154547929764,
      "epoch": 1.6338028169014085,
      "grad_norm": 7.71875,
      "learning_rate": 5e-05,
      "loss": 2.067565441131592,
      "mean_token_accuracy": 0.5664031803607941,
      "num_tokens": 241497.0,
      "step": 116
    },
    {
      "entropy": 1.6592170596122742,
      "epoch": 1.647887323943662,
      "grad_norm": 7.65625,
      "learning_rate": 5e-05,
      "loss": 1.7524383068084717,
      "mean_token_accuracy": 0.6437603831291199,
      "num_tokens": 243721.0,
      "step": 117
    },
    {
      "entropy": 1.8260776698589325,
      "epoch": 1.6619718309859155,
      "grad_norm": 8.0,
      "learning_rate": 5e-05,
      "loss": 1.7933809757232666,
      "mean_token_accuracy": 0.6192324906587601,
      "num_tokens": 246193.0,
      "step": 118
    },
    {
      "entropy": 1.6278434097766876,
      "epoch": 1.676056338028169,
      "grad_norm": 7.875,
      "learning_rate": 5e-05,
      "loss": 1.6023352146148682,
      "mean_token_accuracy": 0.6632742285728455,
      "num_tokens": 248065.0,
      "step": 119
    },
    {
      "entropy": 1.5735027194023132,
      "epoch": 1.6901408450704225,
      "grad_norm": 7.3125,
      "learning_rate": 5e-05,
      "loss": 1.5997726917266846,
      "mean_token_accuracy": 0.6820999830961227,
      "num_tokens": 250126.0,
      "step": 120
    },
    {
      "entropy": 1.7936385571956635,
      "epoch": 1.704225352112676,
      "grad_norm": 7.1875,
      "learning_rate": 5e-05,
      "loss": 1.662062644958496,
      "mean_token_accuracy": 0.6343215107917786,
      "num_tokens": 252242.0,
      "step": 121
    },
    {
      "entropy": 2.0255846083164215,
      "epoch": 1.7183098591549295,
      "grad_norm": 7.53125,
      "learning_rate": 5e-05,
      "loss": 1.9866116046905518,
      "mean_token_accuracy": 0.5996693223714828,
      "num_tokens": 254704.0,
      "step": 122
    },
    {
      "entropy": 1.8428587317466736,
      "epoch": 1.732394366197183,
      "grad_norm": 7.0625,
      "learning_rate": 5e-05,
      "loss": 1.786312460899353,
      "mean_token_accuracy": 0.6079171597957611,
      "num_tokens": 257154.0,
      "step": 123
    },
    {
      "entropy": 1.8701068460941315,
      "epoch": 1.7464788732394365,
      "grad_norm": 6.96875,
      "learning_rate": 5e-05,
      "loss": 1.814254641532898,
      "mean_token_accuracy": 0.6146602630615234,
      "num_tokens": 260050.0,
      "step": 124
    },
    {
      "entropy": 1.7365414798259735,
      "epoch": 1.76056338028169,
      "grad_norm": 6.8125,
      "learning_rate": 5e-05,
      "loss": 1.6392512321472168,
      "mean_token_accuracy": 0.631788820028305,
      "num_tokens": 262731.0,
      "step": 125
    },
    {
      "entropy": 1.5750393569469452,
      "epoch": 1.7746478873239435,
      "grad_norm": 8.875,
      "learning_rate": 5e-05,
      "loss": 1.652256727218628,
      "mean_token_accuracy": 0.6727721691131592,
      "num_tokens": 264264.0,
      "step": 126
    },
    {
      "entropy": 1.6853447258472443,
      "epoch": 1.788732394366197,
      "grad_norm": 9.125,
      "learning_rate": 5e-05,
      "loss": 1.8110965490341187,
      "mean_token_accuracy": 0.6612876951694489,
      "num_tokens": 265839.0,
      "step": 127
    },
    {
      "entropy": 1.9790962934494019,
      "epoch": 1.8028169014084507,
      "grad_norm": 8.6875,
      "learning_rate": 5e-05,
      "loss": 1.9432560205459595,
      "mean_token_accuracy": 0.5927930474281311,
      "num_tokens": 268531.0,
      "step": 128
    },
    {
      "entropy": 1.6438206434249878,
      "epoch": 1.8169014084507042,
      "grad_norm": 9.75,
      "learning_rate": 5e-05,
      "loss": 2.005178689956665,
      "mean_token_accuracy": 0.6584273278713226,
      "num_tokens": 270295.0,
      "step": 129
    },
    {
      "entropy": 1.8002346009016037,
      "epoch": 1.8309859154929577,
      "grad_norm": 8.6875,
      "learning_rate": 5e-05,
      "loss": 1.9875435829162598,
      "mean_token_accuracy": 0.6358410120010376,
      "num_tokens": 272470.0,
      "step": 130
    },
    {
      "entropy": 1.811629831790924,
      "epoch": 1.8450704225352113,
      "grad_norm": 9.25,
      "learning_rate": 5e-05,
      "loss": 1.896381139755249,
      "mean_token_accuracy": 0.6606249660253525,
      "num_tokens": 274166.0,
      "step": 131
    },
    {
      "entropy": 1.7871823608875275,
      "epoch": 1.8591549295774648,
      "grad_norm": 8.625,
      "learning_rate": 5e-05,
      "loss": 1.7262006998062134,
      "mean_token_accuracy": 0.6428729891777039,
      "num_tokens": 276354.0,
      "step": 132
    },
    {
      "entropy": 1.5866018235683441,
      "epoch": 1.8732394366197183,
      "grad_norm": 9.875,
      "learning_rate": 5e-05,
      "loss": 1.54076087474823,
      "mean_token_accuracy": 0.6837680488824844,
      "num_tokens": 278044.0,
      "step": 133
    },
    {
      "entropy": 1.9443781971931458,
      "epoch": 1.887323943661972,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 1.935890555381775,
      "mean_token_accuracy": 0.5984784960746765,
      "num_tokens": 279727.0,
      "step": 134
    },
    {
      "entropy": 1.6744503676891327,
      "epoch": 1.9014084507042255,
      "grad_norm": 8.375,
      "learning_rate": 5e-05,
      "loss": 1.667353868484497,
      "mean_token_accuracy": 0.6531110107898712,
      "num_tokens": 281727.0,
      "step": 135
    },
    {
      "entropy": 1.5089002698659897,
      "epoch": 1.915492957746479,
      "grad_norm": 9.5625,
      "learning_rate": 5e-05,
      "loss": 1.7763428688049316,
      "mean_token_accuracy": 0.6703489422798157,
      "num_tokens": 283499.0,
      "step": 136
    },
    {
      "entropy": 1.7822388112545013,
      "epoch": 1.9295774647887325,
      "grad_norm": 8.1875,
      "learning_rate": 5e-05,
      "loss": 1.7652621269226074,
      "mean_token_accuracy": 0.6092153340578079,
      "num_tokens": 286098.0,
      "step": 137
    },
    {
      "entropy": 1.633983552455902,
      "epoch": 1.943661971830986,
      "grad_norm": 8.625,
      "learning_rate": 5e-05,
      "loss": 1.7642450332641602,
      "mean_token_accuracy": 0.6580131351947784,
      "num_tokens": 288020.0,
      "step": 138
    },
    {
      "entropy": 1.9797956943511963,
      "epoch": 1.9577464788732395,
      "grad_norm": 8.0625,
      "learning_rate": 5e-05,
      "loss": 1.8309262990951538,
      "mean_token_accuracy": 0.6048336029052734,
      "num_tokens": 290401.0,
      "step": 139
    },
    {
      "entropy": 1.5087556838989258,
      "epoch": 1.971830985915493,
      "grad_norm": 9.875,
      "learning_rate": 5e-05,
      "loss": 1.6536986827850342,
      "mean_token_accuracy": 0.7231832593679428,
      "num_tokens": 291584.0,
      "step": 140
    },
    {
      "entropy": 1.9238645434379578,
      "epoch": 1.9859154929577465,
      "grad_norm": 8.4375,
      "learning_rate": 5e-05,
      "loss": 1.7226345539093018,
      "mean_token_accuracy": 0.6254735887050629,
      "num_tokens": 293607.0,
      "step": 141
    },
    {
      "entropy": 1.587816596031189,
      "epoch": 2.0,
      "grad_norm": 8.8125,
      "learning_rate": 5e-05,
      "loss": 1.5887134075164795,
      "mean_token_accuracy": 0.6791131645441055,
      "num_tokens": 295300.0,
      "step": 142
    },
    {
      "epoch": 2.0,
      "eval_entropy": 1.972671693479511,
      "eval_loss": 2.2974655628204346,
      "eval_mean_token_accuracy": 0.5787080614499642,
      "eval_num_tokens": 295300.0,
      "eval_runtime": 18.8723,
      "eval_samples_per_second": 15.049,
      "eval_steps_per_second": 3.762,
      "step": 142
    }
  ],
  "logging_steps": 1,
  "max_steps": 284,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 303200469906432.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}